Cépages du monde (Description)

Cépages du monde (Description)
COMPENDIUM OF
INTERNATIONAL METHODS
OF WINE AND MUST ANALYSIS
INTERNATIONAL ORGANISATION
OF VINE AND WINE
INTERNATIONAL ORGANISATION OF VINE AND WINE
COMPENDIUM
OF INTERNATIONAL
METHODS OF WINE
AND MUST ANALYSIS
EDITION 2011
VOLUME 2
INCLUDED :
Resolutions adopted in Tbilisi (Georgia)
8th A.G. – 25 June 2010
OIV - 18, RUE D’AGUESSEAU - 75008 PARIS
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Table of contents
General organization of the Compendium
Table of contents
Foreword
ANNEX A – METHODS OF ANALYSIS OF WINES AND MUSTS
SECTION 1 – DEFINITIONS AND GENERAL PRINCIPLES
SECTION 2 – PHYSICAL ANALYSIS
SECTION 3 – CHIMICAL ANALYSIS
SECTION 3.1 – ORGANIC COMPOUNDS
SECTION 3.1.1 – SUGARS
SECTION 3.1.2 – ALCOHOLS
SECTION 3.1.3 – ACIDS
SECTION 3.1.4 – GAS
SECTION 3.1.5 – OTHER ORGANIC COMPOUNDS
SECTION 3.2 – NON ORGANIC COMPOUNDS
SECTION 3.2.1 – ANIONS
SECTION 3.2.2 – CATIONS
SECTION 3.2.3 – OTHER NON ORGANIC COMPOUNDS
SECTION 4 – MICROBIOLOGICAL ANALYSIS
SECTION 5 – OTHER ANALYSIS
ANNEX B - CERTIFICATES OF ANALYSIS
ANNEX C - MAXIMUM ACCEPTABLE LIMITS OF VARIOUS SUBSTANCES
ANNEX D – ADVICES
ANNEX E – LABORATORY QUALITY ASSURANCE
OIV-MA-INT-00-2011
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Table of contents
Title
Reference
‐ Table of contents
Type
method
OIV-MA-INT-00
VOLUME 1
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ Foreword
Layout and wording of OIV method of analysis
OIV-MA-INT-01
OIV-MA-INT-04
ANNEX A – METHODS OF ANALYSIS OF WINES AND MUSTS
SECTION 1 – DEFINITIONS AND GENERAL PRINCIPLES
General remarks
OIV-MA-AS1-02
Classification of analytical methods (oeno 9/2000) OIV-MA-AS1-03
Matrix effect for metals content analysis (oeno
OIV-MA-AS1-04
5/2000)
SECTION 2 – PHYSICAL ANALYSIS
Density and Specific Gravity at 20°C (A 1)
Density and Specific Gravity at 20°C (A 1)
Evaluation by refractometry of the sugar
concentration in grape, musts, concentrated grape
musts and rectified concentratedgrape musts
Total dry matter (gravimétrie) (A 3)
Total dry matter (densimétrie) (A 3)
Ash (A 6)
Alkalinity of Ash (A 7)
Oxidation-reduction potential (oeno 3/2000)
Chromatic Characteristics
Chromatic Characteristics
Wine turbidity (oeno 4/2000)
Method for isotopic ratio 18O/16O (Oeno 2/96)
Folin-Ciocalteu Index
Chromatic Characteristics (Oeno 1/2006)
Method for 18O/16O isotope ratio determination of
water in wines and must
OIV-MA-AS2-01A
OIV-MA-AS2-01B
OIV-MA-AS2-02
OIV-MA-AS2-03A
OIV-MA-AS2-03B
OIV-MA-AS2-04
OIV-MA-AS2-05
OIV-MA-AS2-06
OIV-MA-AS2-07A
OIV-MA-AS2-07B
OIV-MA-AS2-08
OIV-MA-AS2-09
OIV-MA-AS2-10
OIV-MA-AS2-11
OIV-MA-AS2-12
I
IV
I
I
IV
I
IV
IV
Withdrawn
IV
IV
Withdrawn
IV
I
II
OIV-MA-INT-00-2011
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Table of contents
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ SECTION 3 – CHIMICAL ANALYSIS
SECTION 3.1 – ORGANIC COMPOUNDS
SECTION 3.1.1 – SUGARS
Reducing substances
Reducing sugars (clarification) (Type IV)
Reducing sugars (titrimétrie) (type II)
Glucose and fructose (enzymatic method)
Dosage of sugars by HPLC (Oeno 23/2003)
Stabilisation of musts to detect Addition of sucrose
(A 5)
Detecting enrichment of musts, concentrated grape
musts, rectified concentrated grape musts and wine
by ²H-RMN
Polyols derived from sugars (Oeno 9/2006)
Glucose and fructose (pHmetry) (Oeno 10/2006)
Glucose, fructose and saccharose (pHmetry) (Oeno
11/2006)
SECTION 3.1.2 – ALCOHOLS
‐ Alcoholic strength by volume (pycnometry,
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ frequency oscillator, hydrostatic balance)
Alcoholic strength by volume (hydrometer,
refractometry)
- Tables of correction
Methanol (GC) (A 41)
Methanol (colorimetry) (A 41)
Glycerol and 2,3- butanediol (A 21)
Glycerol (enzymatic method)
Determination of isotopic ratio of ethanol (oeno
17/2001)
Glycerol (GC-C-IRMS or HPLC-IRMS method)
(OIV-Oeno 343-2010)
SECTION 3.1.3 – ACIDS
Total Acidity (A 10)
Volatile Acidity (A 11)
Fixed Acidity (A 11)
Organic Acids : HPLC
Tartaric Acid (gravimetry) (A 12)
Tartaric Acid (colorimetry) (A 12)
Lactic Acid - chemical method (A 27)
Lactic Acid - enzymatic method
OIV-MA-INT-00-2011
OIV-MA-AS311-01A
IV
OIV-MA-AS311-01B Withdrawn
OIV-MA-AS311-01C Withdrawn
OIV-MA-AS311-02
II
OIV-MA-AS311-03
II
OIV-MA-AS311-04
OIV-MA-AS311-05
I
OIV-MA-AS311-06
OIV-MA-AS311-07
IV
III
OIV-MA-AS311-08
IV
OIV-MA-AS312-01A
I
OIV-MA-AS312-01B
IV
OIV-MA-AS312-02
OIV-MA-AS312-03A
OIV-MA-AS312-03B
OIV-MA-AS312-04
OIV-MA-AS312-05
IV
IV
IV
IV
OIV-MA-AS312-06
II
OIV-MA-AS312-07
IV
OIV-MA-AS313-01
I
OIV-MA-AS313-02
I
OIV-MA-AS313-03
I
OIV-MA-AS313-04
IV
OIV-MA-AS313-05A
IV
OIV-MA-AS313-05B Withdrawn
OIV-MA-AS313-06
Withdrawn
OIV-MA-AS313-07
II
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Table of contents
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ Citric Acid - chemical method (A 29)
Citric Acid - enzymatic method
Total malic Acid: usual method (A 33)
L-malic Acid: enzymatic method
D-malic Acid: enzymatic method (oeno 6/98)
D-malic Acid: enzymatic method low
concentrations
L-ascorbic Acid (spectrofluorimetry) (A 28)
L-ascorbic Acid (spectrophotometry) (A 28)
Sorbic Acid (spectrophotometry) (A 30)
Sorbic Acid (GC ) (A 30)
Sorbic Acid (TLC) (A 30)
pH
Organic acid : ionic chromatography
Shikimic acid (oeno 33/2004)
Sorbic acid (capillary electrophoresis) (oeno
4/2006)
Organic acids (capillary electrophoresis) (oeno
5/2006)
Sorbic, benzoic, salicylic acids (oeno 6/2006)
Metatartaric acid (oeno 10/2007)
Determination of L-ascorbic acid and D-isoascorbic acid by HPLC (oeno 11/2008)
Identification of L- tartaric acid (oeno 12/2008)
OIV-MA-AS313-08
OIV-MA-AS313-09
OIV-MA-AS313-10
OIV-MA-AS313-11
OIV-MA-AS313-12A
IV
II
IV
II
II
OIV-MA-AS313-12B
IV
OIV-MA-AS313-13A
IV
OIV-MA-AS313-13B Withdrawn
OIV-MA-AS313-14A
IV
OIV-MA-AS313-14B
IV
OIV-MA-AS313-14C
IV
OIV-MA-AS313-15
I
OIV-MA-AS313-16
IV
OIV-MA-AS313-17
II
OIV-MA-AS313-18
IV
OIV-MA-AS313-19
II
OIV-MA-AS313-20
OIV-MA-AS313-21
IV
IV
OIV-MA-AS313-22
II
OIV-MA-AS313-23
IV
OIV-MA-AS314-01
and completed by Oeno 3/2006))
‐ Overpressure measurment of sparkling wines (Oeno OIV-MA-AS314-02
21/2003)
‐ Determination of the carbon isotope ratio 13C/12C OIV-MA-AS314-03
of CO2 in sparkling wines (Oeno 7/2005)
‐ Carbone dioxyde (manometric method) (oeno
OIV-MA-AS314-04
2/2006)
II
‐ ‐ ‐ ‐ ‐ SECTION 3.1.4 – GAS
‐ Carbone Dioxide (A 39 modified by oeno 21/2003
I
II
II
OIV-MA-INT-00-2011
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Table of contents
VOLUME 2
SECTION 3.1.5 – OTHER ORGANIC
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ COMPOUNDS
Acetaldehyde (ethanal) (A 37)
Ethyl Acetate (GC)
Ethyl Acetate (titrimetry)
Malvidin Diglucoside (A 18)
Ethyl Carbamate (oeno 8/98)
Hydroxymethylfurfural (colorimetry) (A 19)
Hydroxymethylfurfural (HPLC) (A 19)
Cyanide Derivatives (oeno 4/94)
Artificial sweeteners (TLC : saccharine, cyclamate,
Dulcin and P-4000 ) (A 36)
Artificial sweeteners (TLC: saccharine, cyclamate
and Dulcin) (A 36)
Artificial Colorants (A 43)
Diethylene glycol
Ochratoxin A
HPLC-Determination of nine major Anthocyanins
in red and rosé wines (Oeno 22/2003; 12/2007)
Plant proteins
Polychlorophenols, polychloroanisols (oeno 8/2006)
Determination of Lysozyme by HPLC (oeno
8/2007)
Determination of 3-Methoxypropane-1,2-diol and
Cyclic Diglycerols (oeno 11/2007)
Determination of releasable 2,4,6-trichloroanisole in
wine
Determining the presence and content of
polychlorophenols and polychloroanisols in wines,
cork stoppers, wood and bentonites used as
atmospheric traps
Analysis of biogenic amines in musts and wines
Determination of glutathione
Determination of -dicarbonyl compounds of wine
by HPLC after derivatization (OIV-Oeno 386A2010)
Determination of -dicarbonyl compounds of wine
by GC after derivatization (OIV-Oeno 386B-2010)
Determination of carboxymethyl cellulose in white
wines OIV-Oeno 404-2010
Quantification of potentially allergenic residues of
fining agent proteins in wine (OIV-Oeno 427-2010)
OIV-MA-INT-00-2011
OIV-MA-AS315-01
OIV-MA-AS315-02A
OIV-MA-AS315-02B
OIV-MA-AS315-03
OIV-MA-AS315-04
OIV-MA-AS315-05A
OIV-MA-AS315-05B
OIV-MA-AS315-06
IV
IV
IV
IV
II
IV
IV
II
OIV-MA-AS315-07A
IV
OIV-MA-AS315-07B
IV
OIV-MA-AS315-08
OIV-MA-AS315-09
OIV-MA-AS315-10
IV
IV
II
OIV-MA-AS315-11
II
OIV-MA-AS315-12
OIV-MA-AS315-13
IV
Withdrawn
OIV-MA-AS315-14
IV
OIV-MA-AS315-15
II
OIV-MA-AS315-16
IV
OIV-MA-AS315-17
IV
OIV-MA-AS315-18
OIV-MA-AS315-19
II
IV
OIV-MA-AS315-20
IV
OIV-MA-AS315-21
IV
OIV-MA-AS315-22
IV
OIV-MA-AS315-23
criteria
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Table of contents
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ SECTION 3.2 – NON ORGANIC COMPOUNDS
SECTION 3.2.1 – ANIONS
Total Bromide (A 23)
Chlorides (A 15)
Fluorides (A 22)
Total Phosphorus (A 16)
Sulfates (gravimetry) (A 14)
Sulfates (titrimetry) (A 14)
OIV-MA-AS321-01
IV
OIV-MA-SA321-02
II
OIV-MA-AS321-03
II
OIV-MA-AS321-04
IV
OIV-MA-AS321-05A
II
OIV-MA-AS321-05B Withdrawn
SECTION 3.2.2 – CATIONS
Ammonium (A 20)
Potassium (AAS) (A 8)
Potassium (flame photometry) (A 8)
Potassium (gravimetry) (A 8)
Sodium (AAS) (A 25)
Sodium (flame photometry ) (A 25)
Calcium (A 26)
Iron (A 9)
Iron (A 9)
Copper
Magnesium (A 26)
Zinc (A 45)
Silver
Cadmium
Lead
Lead (criteria for methods) (oeno 7/2006)
OIV-MA-AS322-01
IV
OIV-MA-AS322-02A
II
OIV-MA-AS322-02B
III
OIV-MA-AS322-02C Withdrawn
OIV-MA-AS322-03A
II
OIV-MA-AS322-03B
III
OIV-MA-AS322-04
II
OIV-MA-AS322-05A
IV
OIV-MA-AS322-05B
IV
OIV-MA-AS322-06
IV
OIV-MA-AS322-07
II
OIV-MA-AS322-08
IV
OIV-MA-AS322-09
IV
OIV-MA-AS322-10
IV
OIV-MA-AS322-11
Withdrawn
OIV-MA-AS322-12
II
SECTION 3.2.3 – OTHER NON ORGANIC
COMPOUNDS
Arsenic (AAS) (Oeno 14/2002)
Arsenic (AAS) (A 34)
Arsenic (colorimetry) (A 34)
Total nitrogen (A 40)
Total nitrogen - Dumas method (Oeno 13/2002)
Boron (A 44)
Sulfur dioxide (titrimetry)
Sulfur dioxide (Iodometry)
Sulfur dioxide (molecular method)
Sulfur dioxide - grape juice
OIV-MA-AS323-01A
IV
OIV-MA-AS323-01B
IV
OIV-MA-AS323-01C Withdrawn
OIV-MA-AS323-02A
IV
OIV-MA-AS323-02B
II
OIV-MA-AS323-03
IV
OIV-MA-AS323-04A
II
OIV-MA-AS323-04B
IV
OIV-MA-AS323-04C
IV
OIV-MA-AS323-05
IV
OIV-MA-INT-00-2011
6
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Table of contents
‐ Mercury - atomic Fluorescence (Oeno 15/2002)
‐ Multielemental analysis using ICP-MS (OIV-Oeno
OIV-MA-AS323-06
IV
OIV-MA-AS323-07
II
SECTION 4 – MICROBIOLOGICAL ANALYSIS
OIV-MA-AS4-01
‐ Microbiological Analysis (oeno 8/95)
‐ Detection of preservatives and fermentation
OIV-MA-AS4-02A
IV
344-2010)
‐ ‐ ‐ ‐ ‐ inihibitors (Fermentability Test)
Detection of preservatives and fermentation
inihibitors (Detection of the following acids: sorbic,
benzoic, p-chlorobenzoic, salicylic,
p-hydroxybenzoic and its esters)
Detection of preservatives and fermentation
inihibitors (Detection of the monohalogen
derivatives of acetic acid)
Detection of preservatives and fermentation
inihibitors (determination of ethyl pyrocarbonate)
Detection of preservatives and fermentation
inihibitors (Examination of dehydroacetic acid)
Detection of preservatives and fermentation
inihibitors (Sodium Azide by HPLC)
IV
OIV-MA-AS4-02B
IV
OIV-MA-AS4-02C
IV
OIV-MA-AS4-02D
IV
OIV-MA-AS4-02E
IV
OIV-MA-AS4-02F
IV
SECTION 5 – OTHER ANALYSIS
‐ Differentiation of fortified musts and sweet fortified OIV-MA-AS5-01
wines
ANNEX B - CERTIFICATES OF ANALYSIS
‐ Rules for the implementation of the analytical
methods
‐ Certificates of analysis
OIV-MA-B1-01
OIV-MA-B1-02
ANNEX C - MAXIMUM ACCEPTABLE LIMITS OF VARIOUS SUBSTANCES
‐ Maximum acceptable limits of various substances
OIV-MA-C1-01
contained in wine
ANNEX D – ADVICES
‐ Gluconic Acid (oeno 4/91)
‐ Characterization of wines resulting from
overpressing (oeno 5/91)
‐ Level of sodium and chlorides ions in wines (oeno
6/91)
OIV-MA-D1-01
OIV-MA-D1-02
OIV-MA-D1-03
OIV-MA-INT-00-2011
7
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Table of contents
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ANNEX E – LABORATORY QUALITY ASSURANCE
Validation Principle (oeno 7/98)
OIV-MA-AS1-05
Collaborative Study
OIV-MA-AS1-07
Reliability of methods (oeno 5/99)
OIV-MA-AS1-08
Protocol for the design, conducts and interpretation
OIV-MA-AS1-09
of collaborative studies (oeno 6/2000)
Estimation of the detection and quantification limits
OIV-MA-AS1-10
of a method of analysis (oeno 7/2000)
Harmonized guidelines for internal quality control
OIV-MA-AS1-11
in analytical chemistry laboratories (Oeno 19/2002)
Practical guide for the Validation (oeno 10/05)
OIV-MA-AS1-12
Harmonised guidelines for single-laboratory
OIV-MA-AS1-13
validation (oeno 8/05)
Recommendations on measurement uncertainty
OIV-MA-AS1-14
(oeno 9/05)
Recommendations related to the recovery correction OIV-MA-AS1-15
OIV-MA-INT-00-2011
8
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Foreword
Foreword
The Compendium of International Methods of Wine Analysis was first published in
1962 and re-published in 1965, 1972, 1978, 1990 and 2000; each time it included
additional material as approved by the General Assembly and produced each year
by the Sub-Commission.
This edition includes all material as approved by the General Assembly of
representatives of the member governments of the OIV, revised and amended since
2000.
The Compendium plays a major part in harmonising methods of analysis. Many
vine-growing countries have introduced its definitions and methods into their own
regulations. When making a series of Regulations on methods of analysis
(1539/71, 2948/78, 1108/82 and 2676/90), the Commission of the European
Communities considered "that as far as possible methods enjoying general
recognition should be used, such as the methods developed under the 1954
International Convention on Unification of Methods of Analysis and Appraisal of
Wines, which are published by the Office International de la Vigne et du Vin in
the Compendium of International Methods of Wine Analysis".
In Council Regulation EC 1493/99 (17 May 1999), the European Union stating
that, where there are no appropriate Community methods of analysis of wine, the
methods of analyses recognised and published by the General Assembly of the
Office International de la Vigne et du Vin shall be applicable, recognises all of the
methods in the Compendium and, where there are no appropriate Community
methods, makes them binding in all Member States, confirming the close
collaboration established between the EEC and the OIV.
Thus, through its leading role in the harmonisation of methods of analysis, the
Compendium contributes to facilitating international trade. With the Code
International des Pratiques Oenologiques and the Codex Oenologique
International, it constitutes a body of considerable scientific, legal and practical
benefit.
OIV-MA-INT-01
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Acetaldehyde
Method OIV-MA-AS315-01
Type IV method
Acetaldehyde
(Resolution Oeno 377/2009)
1. Principle
Acetaldehyde (ethanal) in carbon decolorized wine, reacts with sodium
nitroferricyanide and piperidine and causes a green to violet color change whose
intensity is measured at 570 nm.
2. Apparatus
Spectrophotometer permitting measurement of absorbance at a wavelength of
570 nm with a 1 cm optical cell path.
3. Reagents
3.1 Piperidine solution, (C5H11N) 10% (v/v).
Prepare just before use by mixing 2 mL of piperidine with 18 mL of distilled
water.
3.2 Sodium nitroferricyanide solution, 0.4% (m/v).
In a 250 mL glass volumetric flask, dissolve 1 g of pulverized sodium
nitroferricyanide, Na2 [Fe(CN)5 NO].2H2O in distilled water and make up to
volume.
3.3 Activated carbon
3.4 Dilute hydrochloric acid, 25% (v/v)
3.5 Alkaline solution
Dissolve 8.75 g of boric acid in 400 mL sodium hydroxide solution, 1 M.
Make up to 1 L with distilled water.
4. Procedure
4.1 Sample
Place approx. 25 mL of wine in a 100 mL Erlenmeyer flask, add 2 g of
activated charcoal. Shake vigorously for a few seconds, allow to stand for 2
minutes and filter through a fluted slow filter to obtain a clear filtrate.
OIV-MA-AS315-01 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Acetaldehyde
Place 2 mL of the clear filtrate into a 100 mL Erlenmeyer flask, add, while
shaking, 5 mL of the sodium nitroferricyanide solution (3.2) and 5 mL of the
piperidine solution (3.1). Mix and place the mixture immediately into a 1 cm
optical cell. The coloration produced, which varies from green to violet, is
measured with reference to air at a wavelength of 570 nm. This color change
increases then decreases rapidly; measure immediately and record the
maximum value of the absorbance that is obtained after about 50 seconds. The
concentration of acetaldehyde in the liquid analyzed is obtained using a
calibration curve.
Note: If the liquid analyzed contains excess free acetaldehyde, it will be necessary,
before beginning the total acetaldehyde determination, to first combine it with sulfur
dioxide. To achieve this, add a small amount of excess free SO2 to a portion of the
liquid to be analyzed and wait for an hour before proceeding.
4.2. Preparation of the calibration curve
4.2.1 Solution of acetaldehyde combined with sulfur dioxide
Prepare a solution of between 5 to 6% (m/v) sulfur dioxide and determine the
exact strength by titrating with 0.05 M iodine solution.
In a 1 L glass volumetric flask, add a volume of this solution which
corresponds to 1500 mg of sulfur dioxide. Introduce into the flask, using a
funnel, about 1 mL of acetaldehyde distillate recently distilled and collected in
a cooling mixture. Make up to 1 liter with distilled water. Mix and allow to
stand overnight.
The exact concentration of this solution is determined as follows:
Place in a 500 mL Erlenmeyer flask, 50 mL of the solution; add 20 mL of
dilute hydrochloric acid (3.4) and 100 mL water. Titrate the free sulfur
dioxide using a solution of 0.05 M iodine with starch as indicator, stopping at a
faint blue end point. Add 100 mL of the alkaline solution, and the blue
coloration will disappear.
Titrate the combined sulfur dioxide and
acetaldehyde with 0.05 M iodine until a faint blue end point is reached: let n be
the volume used.
The acetaldehyde solution combined with SO2 contains 44.05 n mg of
acetaldehyde per liter.
4.2.2 Preparation of the calibration standards
In five 100 mL glass volumetric flasks, place respectively 5, 10, 15, 20 and 25
mL of the stock solution. Make up to volume with distilled water. These
solutions correspond to acetaldehyde concentrations of 40, 60, 120, 160 and
200 mg/L. The exact concentration of the dilutions must be calculated from
the acetaldehyde concentration of the stock solution (4.2.1) previously
determined.
OIV-MA-AS315-01 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Acetaldehyde
Proceed with the determination of acetaldehyde on 2 mL of each of these
dilutions as indicated in 4.1. The graph of the absorbance of these solutions as
a function of acetaldehyde content is a straight line that does not pass through
the origin.
BIBLIOGRAPHY
REBELEIN H., Dtsch. Lebensmit. Rdsch., 1970, 66, 5-6.
OIV-MA-AS315-01 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ethyl Acetate
Method OIV-MA-AS315-02A
Type IV method
Ethyl Acetate
1. Principle of the methods
Ethyl acetate is determined by gas chromatography on wine distillate using an
internal standard.
2. Method
2.1 Apparatus (see chapter Volatile Acidity).
2.2 Procedure
Prepare an internal standard solution of 4-methyl-2-pentanol, 1 g/L, in ethanol
solution, 10% (v/v).
Prepare the sample solution to be determined by adding 5 mL of this internal
standard solution to 50 mL of wine distillate obtained as indicated in the
chapter on Alcoholic Strength.
Prepare a reference solution of ethyl acetate, 50 mg/L, in ethanol, 10% (v/v).
Add 5 mL of the internal standard to 50 mL of this solution.
Analyze 2 L of the sample solution and the reference solution using gas
chromatography.
Oven temperature is 90°C and the carrier gas flow rate is 25 mL per minute.
2.3 Calculation
S = the peak area of ethyl acetate in the reference solution.
Sx = the peak area of the ethyl acetate in the sample solution.
I = the peak area of the internal standard in the sample solution.
I = the peak area of the internal standard in the reference solution.
The concentration of ethyl acetate, expressed in milligrams per liter, is given
by:
S
50 x I x 
i S
OIV-MA-AS315-02A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ethyl Acetate
Method OIV-MA-AS315-02B
Type IV method
Ethyl Acetate
1. Principle of the methods
Ethyl acetate is separated by distillation of wine brought to pH 6.5. After
saponification and suitable concentration in an alkaline environment, the distillate
is acidified and the vapor condensed to separate the acetic acid liberated by
saponification; the acid portion is titrated with the alkaline solution.
2. Method
2.1 Reagents
2.1.1 Sodium hydroxide solution, 1 M
2.1.2 pH 6.5 Buffer solution
Potassium di-hydrogen phosphate, KH2PO4 .................... 5 g
Sodium hydroxide solution 1 M .................……………...….. 50 mL
Water to ..........................................…………………….……………
1L
2.1.3 Crystalline tartaric acid
2.1.4 Sodium hydroxide solution, 0.02 M
2.1.5 Neutral phenolphthalein solution, 1%, in alcohol, 96% (v/v).
2.2 Usual method
Into a 500 mL volumetric flask, place 100 mL of non-decarbonated wine
neutralized with n mL of 1 M sodium hydroxide solution, n being the volume
of sodium hydroxide solution, 0.1 M, used for measuring the total acidity of 10
mL of wine. Add 50 mL of pH 6.5 buffer solution and distill. The distillation
must be conducted using a tapered tube into a 500 mL round-bottom flask
containing 5 mL of 1 M sodium hydroxide solution, on which a mark has been
made indicating a volume of approximately 35 mL. Collect 30 mL of
distillate.
Stopper the flask and allow to stand for one hour. Concentrate the contents of
the flask to approximately 10 mL by placing it in a boiling water bath and
blowing a rapid stream of air into the bowl of the flask. Allow to cool. Add 3
g tartaric acid (2.1.3). Eliminate carbon dioxide by shaking under a vacuum.
Transfer the liquid from the concentrating flask to the bubbling chamber of a
steam distillation apparatus and rinse the flask twice with 5 mL of water.
Steam distill and recover at least 250 mL of distillate.
OIV-MA-AS315-02B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ethyl Acetate
Titrate with a 0.02 M sodium hydroxide solution, in the presence of
phenolphthalein.
2.3 Calculation
Let n be the number of milliliters of sodium hydroxide solution, 0.02 M (2.1.4)
used. 1 mL corresponds to 1.76 mg ethyl acetate.
The concentration of ethyl acetate in milligrams per liter is given by:
17.6 x n
BIBLIOGRAPHY
Usual method:
PEYNAUD E., Analyse et contrôle des vins, Librairie Polytechnique Ch.Béranger, 1958.
OIV-MA-AS315-02B : R2009
2
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Malvidin diglucoside
Method OIV-MA-AS315-03
Type IV method
Malvidin diglucoside
1. Principle
Malvidin diglucoside, oxidized by nitric acid, is converted to a substance that, in
an ammonium medium, emits a vivid green fluorescence in ultraviolet light.
The intensity of the fluorescence of the compound formed is measured by
comparison with the fluorescence of a solution titrated with quinine sulfate whose
intensity of fluorescence is standardized with the malvidin diglucoside reference.
Free sulfur dioxide, which attenuates the fluorescence, must previously be
combined with excess acetaldehyde.
2. Qualitative Examination
2.1 Apparatus
2.1.1 Ultraviolet lamp permitting measurement at 365 nm.
2.2 Reagents
2.2.1 Acetaldehyde solution
Crystallizable paraldehyde ............................………..……....... 10 g
Ethanol 96% (v/v) ....................................……………………........ 100 mL
2.2.2 Hydrochloric acid, 1.0 M.
2.2.3 Sodium nitrate solution, 10 g/L.
2.2.4 Ethanol, 96% (v/v), containing 5% concentrated ammonia solution
(20 = 0.92 g/mL).
2.2.5 Control wine containing 15 mg of malvidin diglucoside per liter.
2.2.6 Wine containing no malvidin diglucoside.
2.3 Method
Into a test tube add:
- 10 mL of wine
- 1.5 mL of acetaldehyde solution
wait 20 minutes.
Into a 20 mL centrifuge tube place:
- 1 mL of wine reacted with acetaldehyde
- 1 drop of hydrochloric acid
- 1 mL sodium nitrate solution
Stir; wait 2 minutes (5 minutes maximum); add:
OIV-MA-AS315-03 : R2009
1
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Malvidin diglucoside
- 10 mL ammoniacal ethanol
Treat similarly 10 mL of wine containing 15 mg/L malvidin diglucoside (The
control wine). Stir. Wait 10 minutes and centrifuge.
Decant the clear liquids from the top into calibrated test tubes. Observe the
difference in green fluorescence between the test wine and the control wine
under ultraviolet light at 365 nm.
For rose wines, it is possible to increase the sensitivity using:
- 5 mL of wine treated with acetaldehyde (2.3)
- 0.2 mL hydrochloric acid, 1 M (2.2.2)
- 1 mL sodium nitrate solution, 10 g/L (2.2.3)
- 5.8 mL ammoniacal ethanol (2.2.4)
Treat the control wine in a similar manner.
2.4 Interpretation
Wines that do not fluoresce, or have a distinctly lower fluorescence, than the
control, may be considered to have no malvidin diglucoside. Those whose
fluorescence is slightly less than, equal to, or greater than the control should
have a quantitative determination.
3. Quantitative Determination
3.1. Apparatus
3.1.1. Equipment for measuring fluorescence:
- excitation wavelength 365 nm;
- wavelength of fluorescent radiation 490 nm.
3.1.2. Optical quartz cell (1 cm path length)
3.2 Reagents
3.2.1. See qualitative examination
3.2.2. 2 mg/L quinine sulfate solution
Prepare a solution containing 10 mg very pure quinine sulfate in 100 mL
sulfuric acid, 0.1 M. Dilute 20 mL of this solution to 1 liter with sulfuric acid
solution, 0.1 M.
3.3 Procedure
Treat the wine by the method described in Qualitative examination (2), except
that the aliquot of acetaldehyde treated wine is each case (red wines and roses)
1 mL.
Place the 2 mg/L solution of quinine sulfate in the cell, adjust the fluorometer
to the full range (transmission T, equal to 100%) by adjusting the slit width or
the sensitivity.
OIV-MA-AS315-03 : R2009
2
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Malvidin diglucoside
Replace this tube with the one containing the test wine: this is the T 1 value.
If the percentage of transmission, T1 is greater than 35, dilute the wine with
wine without malvidin diglucoside whose fluorescence must be less than 6%
(this should be ascertained by previous testing.)
Remarks:
1. Salicylic acid (sodium salicylate) added to the wine for stabilization before
analysis, causes a spurious fluorescence which can be eliminated by an
extraction with ether.
2. Spurious fluorescence is caused by the addition of caramel.
3.4 Calculation
A fluorescence intensity of 1, for wine without SO2, for the operating
conditions above with the exception of the acetaldehyde treatment,
corresponds to 0.426 mg malvidin diglucoside per liter of wine.
On the other hand, red and rose wines, containing no malvidin diglucoside,
give fluorescence corresponding to a T value of the order of 6%.
The amount of malvidin diglucoside in wine in milligrams per liter is
therefore:
(T1  6) 0,426 x
11,5
 (T1  6) x 0,49
10
If the wine is diluted, multiply the result by the dilution factor.
3.5 Expression of the Results
The amount of malvidin diglucoside is expressed in milligrams per liter of
wine to the nearest whole number.
BIBLIOGRAPHY
DORIER P., VERELLE L., Ann. Fals. Exp. Chim., 1966, 59, 1.
GAROGLIO P.G., Rivista Vitic. Enol., 1968, 21, 11.
BIEBER H., Deutsche Lebensm. Rdsch., 1967, 44-46.
CLERMONT Mlle S., SUDRAUD P., F.V., O.I.V., 1976 n° 586.
OIV-MA-AS315-03 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ethyl carbamate
Method OIV-MA-AS315-04
Type II method
Ethyl Carbamate
(Resolution Oeno 8/98)
Ethyl carbamate analysis in alcoholic beverages: selective detection method by gas
chromatography/mass spectrometry
(Applicable to the determination of ethyl carbamate concentrations between 10
and 200 µg/l).
(Caution: respect safety measures when handling chemical products, ethanol,
acetone and carcinogenic products: ethyl carbamate and dichloromethane. Get rid
of used solvants in a suitable way, compatible with applicable environmental rules
and regulations).
1. Principle
Propyl carbamate is added to a sample as an internal standard, the solution is
diluted with water and placed in a 50 mL solid phase extraction column. Ethyl
carbamate and propyl carbamate are eluted with dichloromethane. The eluate is
concentrated in a rotary evaporator under vacuum. The concentrate is analyzed by
gas chromatography/mass spectrometry using selected ion monitoring mode.
2. Apparatus
2.1 Gas chromatograph/mass spectrometer (GC/MS).
With selected ion
monitoring (SIM), and data handling system. An autosampler is desirable.
2.2 Capillary fused silica column: 30m*  0.25 mm  int., 0.25 µm of Carbowax
20M type.
2.3 Operating conditions: injector 180°C, helium carrier gas at 1 mL/min at 25°C,
splitless injection. Temperature program: 40°C for 0.75 min, then program
10°C/min to 60°C, then 3°C**/min to 150°C, post run: go up to 220°C and
maintain for 4.25 min at 220°C. The retention time for ethyl carbamate is 2327 min., that of propyl carbamate is 27-31 min.
GC/MS interface: transfer line 220°C. Mass spectrometer parameters set up
manually with perfluorotributylamine and optimized for a lower mass
* For certain wines which are particularly rich, it may be desirable to use a 50m long
capillary column.
** For certain wines which are particularly rich, it may be desirable to carry out a
temperature program of 2°C per minute.
OIV-MA-AS315-04 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ethyl carbamate
sensitivity, SIM acquisition mode, solvent delay and time for the start of
acquisition 22 min., dwell time/ion 100 ms.
2.4 Rotary evaporator under vacuum or concentration system similar to Kuderna
Danish. (Note: the recovery of the ethyl carbamate test sample, (3.7) must be
between 90-110% during the process).
2.5 Flask - pear-shaped, 300 mL, single neck, 24/40 standard taper joint.
2.6 Concentrator tube - 4 mL, graduated, with a standard taper 19/22 Teflon
coated joint and stopper.
3. Reagents
3.1 Acetone - HPLC quality. Note: Check each batch by GC/MS before use with
regard to the absence of response for m/z 62, 74 and 89 ions.
3.2 Dichloromethane - Note: Analyze each batch before use by GC/MS after 200
fold concentration to check the absence of response for m/z 62, 74 and 89
ions.
3.3 Ethanol - anhydrous
3.4 Ethyl carbamate (EC) standard solutions
- (1) Stock solution - 1.00 mg/mL. Weigh 100 mg EC ( 99% purity) in a
volumetric flask of 100 mL and dilute to mark with acetone.
- (2) Standard working solution- 10.0 µg/mL. Transfer 1 mL of the EC stock
solution to a 100 mL volumetric flask and dilute with acetone to the mark.
3.5 n-Propyl carbamate (PC), standard solutions.
- (1) Stock solution - 1.00 mg/mL. Weigh 100 mg PC (reagent quality) in a
100 mL volumetric flask and dilute with acetone to the mark.
- (2) Standard working solution- 10.0 µg/mL. Transfer 1 mL of the PC stock
solution to a volumetric flask of 100 mL and dilute with acetone to the
mark.
- (3) Internal standard solution PC - 400 ng/mL. Transfer 4 mL of the
standard PC working solution to a volumetric flask of 100 mL and dilute
with water to the mark.
3.6 EC - nPC standard calibration solutions - Dilute the standard working
solutions of EC, 3.4 (2), and PC 3.5 (2), with dichloromethane in order to
obtain:
(1)
100 ng EC and 400 ng nPC/mL,
(2)
200 ng EC and 400 ng nPC/mL,
(3)
400 ng EC and 400 ng nPC/mL,
(4)
800 ng EC and 400 ng nPC/mL,
(5) 1600 ng EC and 400 ng nPC/mL.
OIV-MA-AS315-04 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ethyl carbamate
3.7 Practice sample - 100 ng EC/mL in 40 % ethanol. Transfer 1 mL of the
standard EC working solution, 3.4 (2) in a 100 mL volumetric flask and dilute
with 40 % of ethanol to the mark.
3.8 Solid phase extraction column - Disposable material, pre-packed with
diatomaceous earth, capacity 50 mL.
(Note: Before analysis, check each batch of extraction columns for the
recovery of EC and nPC and the absence of response for ions of m/z 62,74 and
89.) Prepare 100 ng EC/mL of test sample 3.7.
Analyze 5.00 mL of the test sample as described in 4.1, 4.2, 5, and 6. The
recovery of 90-110 ng of EC/mL is satisfactory. Adsorbents whose particle
diameter is irregular can lead to a slow flow that affects the recovery of EC
and nPC.
If, after several trials, 90-110 % of the test sample value is not obtained,
change the column or use a corrected calibration recovery curve to quantify
EC.
To obtain the corrected calibration curve, prepare standard solutions as
described in 3.6 by using 40 % ethanol instead of dichloromethane.
Analyze 1 mL of the standard calibration solution as described in 4, 5, and 6.
Establish a new standardization curve by using the EC/nPC ratio of the
extracted standards.
4. Preparation of the test sample
Place the test material in 2 separate 100 mL beakers using the following quantities:
4.1 Wines containing over 14 % vol. alcohol: 5.00 mL  0.01 mL.
4.2 Wines containing maximum 14% vol. of alcohol: 20.00 mL  0.01 mL.
In each beaker, add 1 mL of internal standard PC solution, 3.5 (3) and water,
in order to obtain a total volume of 40 mL (or 40 g).
5. Extraction
(Note: Carry out the extraction under a fume hood with adequate ventilation.)
Transfer diluted test portion from 4 to the extraction column.
Rinse the beaker with 10 mL of water and transfer the rinsing water to the column.
Let the liquid be absorbed in the column for 4 minutes. Elute with 2 80 mL of
dichloromethane.
Collect the eluate in a 300 mL pear-shaped flask.
Evaporate the eluate to 2 to 3 mL in a rotary evaporator in a water bath at 30°C
(Note: do not let extract evaporate to dryness).
Transfer the concentrated residue to a 4 mL graduated concentrator tube, with a 9
in Pasteur pipette.
OIV-MA-AS315-04 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ethyl carbamate
Rinse the flask with 1 mL of dichloromethane and transfer the rinsing liquid to the
tube.
Concentrate the sample to 1 mL under a slight nitrogen stream.
If an autosampler is used, transfer the concentrate to a vial for GC/MS analysis.
6. GC/MS Analysis
6.1 Calibration curve - Inject 1µl of each calibration standard solution 3.6, into
GC/MS. Plot the graph of the EC-nPC area ratio for the response to m/z 62
ion on the y-axis and the quantity of EC in ng/mL on the x-axis (i.e., 100, 200,
400, 800, 1600 ng/mL).
6.2 EC quantification - Inject 1µl of concentrated extract from 5 in the GC/MS
system and calculate the EC-nPC area ratio for m/z 62 ion. Determine the
concentration of EC (ng/mL) in the extract by using the internal standard
standardization curve. Calculate the EC concentration in the test sample
(ng/mL) by dividing the quantity of EC (ng/mL) in the extract by the test
sample volume 3.7.
6.3 Confirmation of EC identity. Determine if the response for m/z 62, 74 and 89
ions appear at the EC retention time. These responses characteristic
respectively of the main fragments (M - C2H3)+ and (M - CH3)+ and
molecular ion (M). The presence of EC is confirmed if the relative ratio of
these ions does not exceed 20% of the ratios of the EC standard. The extract
may need to be further concentrated in order to obtain a sufficient response for
the m/z 89 ion.
7. Method performance.
Sample
Wine over 14
% alcohol
(v/v)
Wine under
14% alcohol
(v/v)
Mean EC Recovery
found, ng/g of added
EC, %
40
80
162
11
25
48
OIV-MA-AS315-04 : R2009
89
90
93
93
sr
SR
1.59
3.32
8.20
0.43
1.67
1.97
4.77
7.00
11.11
2.03
2.67
4.25
RSDr % RSDR%
4.01
4.14
5.05
3.94
6.73
4.10
12.02
8.74
6.84
18.47
10.73
8.86
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Hydroxymethylfurfural
Method OIV-MA-AS315-05A
Type IV method
Hydroxymethylfurfural (HMF)
1. Principle of the methods
Aldehydes derived from furan, the main one being hydroxymethylfurfural,
react with barbituric acid and para-toluidine to give a red compound which is
determined by colorimetry at 550 nm.
Free sulfurous acid interferes with the determination. When its amount
exceeds 10 mg/L, it must be previously eliminated by combining it with
acetaldehyde whose excess does not interfere with the determination.
2. Colorimetric method
2.1 Apparatus
2.1.1 Spectrophotometer for making measurements between 300 and 700 nm.
2.1.2 Glass cells with optical paths of 1 cm.
2.2 Reagents
2.2.1 Barbituric acid solution, 0.5% (m/v)
Dissolve 500 mg of barbituric acid in distilled water by heating slightly over a
water bath at 100°C. Make up to 100 mL with distilled water. This solution
keeps for about a week.
2.2.2 Para-toluidine solution, 10% (m/v).
Place 10 g of para-toluidine in a 100 mL volumetric flask; add 50 mL of isopropanol, CH3CH(OH)CH3, and 10 mL of glacial acetic acid, CH3COOH
(20 = 1.05 g/mL). Make up to 100 mL with iso-propanol. This solution
should be renewed daily.
2.2.3 Acetaldehyde (ethanal) solution, 1% (m/v).
Prepare just before use.
2.2.4 Hydroxymethylfurfural solution, 1 g/L.
Prepare dilutions of the above solution to containing 5, 10, 20, 30 and 40 mg
hydroxymethylfurfural/L. The 1 g/L solution and its dilutions must be freshly
prepared.
OIV-MA-AS315-05A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Hydroxymethylfurfural
2.3 Procedure
2.3.1 Preparation of sample
- Free sulfur dioxide less than 10 mg/L:
Perform the analysis on 2 mL of wine or must. If necessary filter the wine or
must before analysis.
- Free sulfur dioxide greater than 10 mg/L:
15 mL of the test samples are placed in a 25 mL spherical flask with 2 mL
acetaldehyde solution (2.2.3). Stir. Wait 15 minutes. Bring to volume with
distilled water. Filter if necessary. Perform the analysis on 2 mL of this
solution.
2.3.2 Colorimetric determination
Into each of two 25 mL flasks, a and b, fitted with ground glass stoppers, place
2 mL of the sample prepared as in 2.3.1. Place in each flask 5 mL of paratoluidine solution (2.2.2); mix. Add 1 mL of distilled water to flask b (control)
and 1 mL barbituric acid (2.2.1) solution to flask a, shake to mix. Transfer the
contents of the flasks into spectrophotometer cells with optical paths of 1 cm.
Zero the absorbance scale at a wavelength of 550 nm using the contents of
flask b. Follow the variation in the absorbance of the contents of flask a;
record the maximum value A, which is reached after 2 to 5 minutes.
Samples with hydroxymethylfurfural concentrations above 30 mg/L must be
diluted before the analysis.
2.3.3 Preparation of the calibration curve
Place 2 mL of each of the hydroxymethylfurfural solutions of 5, 10, 20, 30 and
40 mg/L into two sets of 25 mL flasks, a and b, and treat them as described in
2.3.2.
The graph representing the variation of absorbance with the
hydroxymethylfurfural concentration in mg/L should be a straight line passing
through the origin.
2.4 Expression of results
The hydroxymethylfurfural concentration is obtained by plotting on the
calibration curve the absorbance determined on the sample analyzed, taking
into account any dilution carried out.
The result is expressed in milligrams per liter (mg/L) to one decimal point.
OIV-MA-AS315-05A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Hydroxymethylfurfural
Method OIV-MA-AS315-05B
Type IV method
Hydroxymethylfurfural (HMF)
1. Principle of the methods
Separation through a column by reversed-phase chromatography and
determination at 280 nm.
Procedures described below are given as examples.
2. High-performance liquid chromatography
2.1 Apparatus
2.1.1 High-performance liquid chromatograph equipped with:
- a loop injector, 5 or 10 µL
- spectrophotometric detector allowing measurement at 280 nm
- column of octadecyl-bonded silica (e.g.Bondapak C18-Corasil, Waters Ass)
- a recorder, preferably an integrator
- Flow rate of mobile phase: 1.5 mL/minute
2.1.2 Membrane filtration system with a pore diameter of 0.45 µm.
2.2 Reagents
2.2.1 Double distilled water
2.2.2 Methanol, distilled or HPLC quality
2.2.3 Acetic acid (20= 1.05 g/mL)
2.2.4 Mobile phase: water + methanol + acetic acid previously filtered through
a 0.45 µm membrane filter, (40 mL + 9 mL + 1 mL)
The mobile phase must be prepared daily and degassed before using.
2.2.5 Hydroxymethylfurfural reference solution, 25 mg/L (m/v)
Into a 100 mL volumetric flask, place 25 mg of hydroxymethylfurfural
accurately weighed, and bring to volume with methanol. Dilute this solution
1/10 with methanol and filter through a 0.45 µm membrane filter.
If the solution is kept refrigerated in a hermetically sealed brown glass bottle it
should keep for two to three months.
OIV-MA-AS315-05B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Hydroxymethylfurfural
2.3 Procedure
Inject 5 (or 10) µL of the sample prepared as described above and 5 (or 10) µL
of hydroxymethylfurfural reference solution into the chromatograph. Record
the chromatogram.
The retention time of hydroxymethylfurfural is about six to seven minutes.
2.4 Expression of the Results
The hydroxymethylfurfural concentration is expressed in milligrams per liter
(mg/L) to one decimal point.
OIV-MA-AS315-05B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Cyanide derivatives
Method OIV-MA-AS315-06
Type II method
Cyanide Derivatives
(Resolution Oeno 4/94)
1. Principle
Free and total hydrocyanic acid is liberated by acid hydrolysis and separated by
distillation. After reaction with chloramine T and pyridine, the glutaconic
dialdehyde formed is determined by colorimetry, due to the blue coloration it gives
with 1.3-dimethyl barbituric acid.
2. Equipment
2.1. Distillation apparatus: Use the distillation apparatus described for the
determination of alcohol in wine.
2.2. Round-bottomed 500 mL flask with standard taper joint.
2.3. Water bath, thermostated at 20° C.
2.4. Spectrophotometer permitting the measurement of absorbance at a
wavelength of 590 nm.
2.5. Glass cuvette or disposable cuvettes for one use only, with 20 mm optical
path.
3. Reagents
3.1. Phosphoric acid (H3PO4) at 25 p. 100 (w/v)
3.2. Solution of chloramine T (C7H7 ClNNaO2S.3H2O) 3% (w/v)
3.3. Solution of 1,3-dimethylbarbituric acid: dissolve 3.658 g of 1,3dimethylbarbituric acid (C6H8N2O3) in 15 mL of pyridine and 3 mL of
hydrochloric acid (20 = 1.19 g/mL) and bring to 50 mL with distilled water.
3.4. Potassium cyanide (KCN)
3.5. Solution of potassium iodide (KI) 10% (w/v)
3.6. Solution of silver nitrate (AgNO3), 0.1 M
4. Procedure
4.1 Distillation:
In the 500 mL round-bottomed flask (2.2), place 25 mL of wine, 50 mL of
OIV-MA-AS315-06 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Cyanide derivatives
distilled water, 1 mL of phosphoric acid (3.1) and some glass beads.
Immediately place the round-bottomed flask on the distillation apparatus.
Collect the distillate through a delivery tube connected to a 50 mL volumetric
flask containing 10 ml of water. The volumetric flask is immersed in an iced
water bath. Collect 30-35 mL of distillate (a total of about 45 mL of liquid in
the volumetric flask). Wash the delivery tube with a few milliliters of distilled
water, bring the distillate to 20°C and dilute with distilled water to the mark.
4.2 Measurement:
Place 25 mL of distillate in a 50 mL glass-stoppered Erlenmeyer flask, add
1 mL of chloramine T solution (3.2) and stopper tightly. After exactly
60 seconds, add 3mL of 1,3-dimethylbarbituric acid solution (3.3), stopper
tightly and let stand for 10 minutes. Then measure the absorbance relative to
the reference blank (25 mL of distilled water instead of 25 mL of distillate) at a
wavelength of 590 nm in cuvettes of 20 mm optical path.
5. Establishing the standard curve
5.1 Argentimetric titration of potassium cyanide.
In a 300 mL volumetric flask, dissolve about 0.2 g of KCN (3.4) precisely
weighed in 100 mL of distilled water. Add 0.2 mL of potassium iodide solution
(3.5) and titrate with the solution of 0.1 M silver nitrate (3.6) until obtaining a
stable yellowish color.
In calculating the concentration of KCN in the sample, 1 mL of 0.1 M silver
nitrate solution corresponds to 13.2 mg of KCN.
5.2 Standard Curve.
5.2.1. Preparation of the standard solutions:
Knowing the KCN concentration determined in accordance with 5.1, prepare a
standard solution containing 30 mg/L of hydrocyanic acid (30 mg HCN = 72.3
mg of KCN). Dilute this solution to 1/10.
Introduce 1.0, 2.0, 3.0, 4.0, and 5.0 mL of the diluted standard solution in 100
mL volumetric flasks and bring to the mark with distilled water. The prepared
standard solutions correspond to 30, 60, 90, and 150 µg/L of hydrocyanic acid,
respectively.
5.2.2. Determination:
Using 25 mL of the solutions, continue as indicated above in 4.1 and 4.2.
The values obtained for the absorbance with these standard solutions, reported
according to the corresponding levels of hydrocyanic acid, form a line passing
through the origin.
OIV-MA-AS315-06 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Cyanide derivatives
6. Expression of the results
Hydrocyanic acid is expressed in micrograms per liter (µg/L) without decimal.
6.1. Calculation:
Determine the concentration of hydrocyanic acid from the standard curve. If a
dilution was done, multiply the result by the dilution factor.
Repeatability (r) and Reproducibility (R)
White wine: r = 3.1 µg/L
R = 12 µg/L
Red wine:
i.e. approximately 6% . Xi
i.e. approximately 25% . Xi
r = 6.4 µg/L
R = 23 µg/L
i.e. approximately 8% . Xi
i.e. approximately 29% . Xi
Xi = average concentration of HCN in the wine.
BIBLIOGRAPHY
1) JUNGE C., Feuillet vert N° 877 (1990)
2) ASMUS E. GARSCHLAGEN H., Z Anal. Chem. 138, 413-422 (1953)
3) WÜRDIG G., MÜLLER TH., Die Weinwissenschaft 43, 29-37 (1988)
OIV-MA-AS315-06 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial sweeteners
Method OIV-MA-AS315-07A
Type IV method
Examination of artificial sweeteners
1. Principle of the methods
Examination of saccharine (benzoic sulfimide), Dulcin (p-ethoxyphenylurea),
cyclamate (cyclohexylsulfamate) and P-4000 (5-nitro-2-propoxyaniline or
1propoxy-2-amino-4-nitrobenzene).
After concentration of the wine, the saccharine, Dulcin and P-4000 are
extracted in an acid medium with benzene; the cyclamate is extracted from the
wine after the benzene extraction using ethyl acetate (the order of extraction is
important). The residues after solvent evaporation are submitted to thin layer
chromatography.
Saccharine and cyclamate are identified by chromatography on cellulose plates
(solvent: acetone-ethyl acetate-ammonium hydroxide), the first the benzene
extract, the second in the extract by the ethyl acetate after purification by
washing with ether.
These sweeteners are developed by spraying with a solution of benzidine;
aniline; cupric acetate, and have the following Rf: 0.29 for cyclamate, 0.46 for
saccharine.
The P-4000 and Dulcin from the benzene extract are separated by
chromatography on polyamide plates, (solvent: toluene; methanol; glacial
acetic acid). These sweeteners are developed by spraying a solution of pdimethylaminobenzaldehyde, and have the following Rf: 0.60 for Dulcin, 0.80
for P-4000.
2. Method
Examination of saccharine, cyclamate, Dulcin and the P-4000.
2.1 Apparatus
2.1.1 Chromatography tank
2.1.2 Micrometry syringes or micropipettes
2.1.3 Separator tube 15 mm in diameter and 180 mm long, with a stopcock
2.1.4 Water bath at 100°C
2.1.5 Regulatable oven, able to reach 125°C
2.2 Reagents
2.2.1 Extraction solvent:
OIV-MA-AS315-07A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial sweeteners
- Benzene
- Ethyl acetate
2.2.2 Chromatography solvents:
Mixture No.1:
Acetone .......................................................……………………
Ethyl acetate ..............................................……………....…..
Ammonium hydroxide (20= 0.92 g/mL) .......……….......………
Mixture No 2.:
Toluene ...................................................…………….....…..…
Methanol .................................................…………….......……
Glacial acetic acid (20= 1.05 g/mL) .............…......….….
60 parts
30 parts
10 parts
90 parts
10 parts
10 parts
2.2.3. Chromatography plates (20 x 20 cm):
- with layer of cellulose powder (for ex., Whatman CC 41 or Macherey-Nagel
MN300)
- with layer of polyamide powder (for ex., Merck)
2.2.4 Indicating reagent for saccharine and cyclamate
Prepare:
- alcoholic solution of benzidine at 250 mg in 100 mL ethanol
- saturated solution of cupric acetate, Cu(C2H3O2)2.H2O
- freshly distilled aniline
Mix: 15 mL of benzidine solution, 1 mL of aniline and 0.75 mL saturated
cupric acetate solution.
This solution must be freshly prepared. It corresponds to the volume required
for development of a 20 x 20 cm plate.
2.2.5 Hydrochloric acid 50% (v/v),
2.2.6 Nitric acid solution, 25% (v/v),
2.2.7 Indicator reagent for the P-4000 and Dulcin: dissolve 1 g of 1,4paradimethylaminobenzaldehyde in 50 mL methanol; add 10 mL 25% nitric
acid; bring to 100 mL with methanol. Use 15 mL of this reagent for the
development of a 20 x 20 cm plate.
2.2.8 Cyclo-hexylsulfamic acid in water-ethanol solution, 0.10 g/100 mL
Dissolve 100 mg of the sodium or calcium salt of cyclo-hexylsulfamic acid in
100 mL of an equal part mixture of water and ethanol.
2.2.9 Saccharine aqueous solution, 0.05 g/100 mL
2.2.10 Dulcin, 0.05 g/100 mL of methanol.
2.2.11 P-4000, 0.05 g/100 mL of methanol.
OIV-MA-AS315-07A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial sweeteners
2.3 Procedure
2.3.1 Extraction
100 mL of wine, placed in a beaker, are rapidly evaporated by boiling until the
volume is reduced to 30 mL, while directing a current of cold air to the surface
of the flask. Allow to cool. Acidify with 3 mL 50% hydrochloric acid (v/v).
Transfer to a 500 mL conical flask with a ground stopper, add 40 mL of
benzene and stir with a mechanical stirrer for 30 min. Transfer to a separating
funnel to separate the organic phase. If an emulsion is formed, it must be
separated by centrifugation. Place the organic phase in a conical flask with a
ground glass stopper.
Decant the wine previously extracted with benzene, which corresponds to the
lower layer in the separating funnel, into a 500 mL conical flask with a ground
stopper containing 40 mL of ethyl acetate. Agitate for 30 minutes and separate
the organic phase as before taking care to recover only the organic fraction and
not the wine.
On a 100°C water bath, evaporate each extraction solvent in 50-60 mm
diameter evaporation dishes, in small amounts while directing a stream of cold
air on the surface of the dishes. Continue the evaporation until the residue has
a syrupy consistency, stopping before the evaporation is complete.
Re-dissolve the benzene extract residue in the evaporation dish with 0.5 mL
ethanol-water (1:1) solution (it is advisable to re-dissolve the residue once with
0.25 mL ethanol-water solution and then to rinse the dish with another portion
of 0.25 mL of the same solution). Place the ethanol-water extract into a small
tube with a ground stopper (extract B).
The residue of the dish in which the ethyl acetate (containing the cyclamate)
has been evaporated, is dissolved with 0.5 mL of water and is poured into a
small separator tube. Wash the dish with 10 mL ether and add the ether to the
contents of the separator tube. Mix vigorously for 2 minutes and separate the
lower layer into a small test tube that contains 0.5 mL ethanol. This comprises
a total of 1 mL of ethanol-water solution that contains the possible cyclamate
(extract A).
2.3.2 Chromatography
2.3.2.1 Saccharine and cyclamate
For examination of the saccharine and cyclamate, use a cellulose plate, with
half of the plate for the identification of cyclamate and the other half for
saccharine.
To do this, spot 5 to 10 µL of extract A and 5µL of the standard cyclamate
solution. On the second part of the plate spot 5 to 10 µL of extract B and 5 µL
of the standard saccharine solution. Place the prepared plate in the
chromatography bath containing solvent No.1 (acetone; ethyl acetate;
OIV-MA-AS315-07A : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial sweeteners
ammonium hydroxide); allow to migrate until the solvent front reaches 10 to
12 cm. Remove the plate from the bath and dry with warm air. Spray the plate
evenly and gently with the benzidine reagent (17-18 mL for each plate). Dry
the plate with cold air. Place the plate in an oven maintained at 120-125°C for
3 minutes. The spots appear dark gray on a light chestnut background; they
turn brownish with time.
2.3.2.2 P-4000 and Dulcin
Deposit 5 µL of extract B and 5 µL of the standard solutions of Dulcin and
P-4000 on a polyamide plate. Place the prepared plate in the chromatography
tank containing solvent No. 2 (toluene; methanol; acetic acid). Let the solvent
front reach a height of 10 to 12 cm.
Remove the plate from the tank; dry in cold air. Spray with 15 mL of the
p-dimethylaminobenzaldehyde reagent, then dry with cold air until the orangeyellow colored spots appear which correspond to Dulcin and P-4000.
2.3.2.3 Sensitivity
The benzidine reagent allows detection of spots corresponding to 2 µg of
saccharine and 5 µg of cyclamate. The p-dimethylaminobenzaldehyde reagent
reveals 0.3 µg of Dulcin and 0.5 µg of P-4000.
This method allows determination of (depending upon the efficiency of the
extractions):
Saccharine ...............................................…..…..……...
2-3 mg/L
Cyclamate .......................................…………...….........
40-50 mg/L
DULCIN .........................................…………............…
1 mg/L
P-4000 .............................................…………............…
1-1.5 mg/L
BIBLIOGRAPHY
 TERCERO C., F.V., O.I.V., 1968, n° 277 and F.V., O.I.V., 1970, n° 352.
 Wine Analysis Commission of the Federal Health Department of Germany,
1969, F.V., O.I.V., n° 316.
 International Federation of Fruit Juice Manufacturers, 1972, F.V., O.I.V., n°
40.
 SALO T., ALRO E. and SALMINEN K., Z. Lebensmittel Unters. u.
Forschung, 1964, 125, 20.
OIV-MA-AS315-07A : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial sweeteners
Method OIV-MA-AS315-07B
Type IV method
Examination of artificial sweeteners
1. Principle of the methods
Examination of saccharine, Dulcin and cyclamate.
These sweeteners are extracted from wine using a liquid ion exchanger, then
re-extracted with dilute ammonia hydroxide, and are separated by thin layer
chromatography using a mixture of cellulose powder and polyamide powder
(solvent: xylene; n-propanol; glacial acetic acid; formic acid). These
sweeteners have a blue fluorescence on a yellow background under ultraviolet
light after spraying with a 2,7-dichlorofluorescein solution.
Subsequent spraying with 1,4-dimethylaminobenzaldehyde solution allows
differentiation of Dulcin, which gives only one orange spot, from vanillin and
the esters of p-hydroxybenzoic acid which migrate with the same Rf.
2. Method
Examination of saccharine, cyclamate and Dulcin.
2.1 Apparatus
2.1.1 Apparatus for expression by thin layer
2.1.2 Glass plate 20 x 20 cm
Preparation of the plates: mix thoroughly 9 g of dry cellulose powder and 6 g
of polyamide powder. Add, while stirring, 60 mL methanol. Spread on the
plates to a thickness of 0.25 mm. Dry for 10 minutes at 70°C. The quantities
prepared are sufficient for the preparation of 5 plates.
2.1.3 Water bath with a temperature regulator or a rotary evaporator,
2.1.4 UV lamp for examination of the chromatography plates.
2.2 Reagents
2.2.1 Petroleum ether (40-60°)
2.2.2 Ion exchange resin, for example: Amberlite LA-2
2.2.3 Acetic acid diluted to 20% (v/v)
OIV-MA-AS315-07B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial sweeteners
2.2.4 Ion exchange solution: 5 mL of ion exchanger is vigorously agitated
with 95 mL petroleum ether and 20 mL of 20% acetic acid. Use the upper
phase.
2.2.5 Nitric acid in solution, 1 M
2.2.6 Sulfuric acid, 10 % (v/v)
2.2.7 Ammonium hydroxide diluted to 25% (v/v)
2.2.8 Polyamide powder, for example: Macherey-Nagel or Merck
2.2.9 Cellulose powder, for example: Macherey-Nagel MN 300 AC
2.2.10 Solvent for chromatography:
Xylene ...........................................................……………......
45 parts
n-Propanol ......................................................…………….….. 6 parts
Glacial acetic acid (20= 1.05 g/mL) .................…...….. 7 parts
Formic acid 98-100% ......................................……………. 2 parts
2.2.11 Developers:
- solution of 2,7-dichlorofluorescein, 0.2 % (m/v), in ethanol,
- solution of 1,4-dimethylaminobenzaldehyde: dissolve 1 g of dimethylaminobenzaldehyde placed in a 100 mL volumetric flask with about 50 mL ethanol.
Add 10 mL of nitric acid, 25% (v/v), and bring to volume with ethanol.
2.2.12 Standard solution:
- solution of Dulcin, 0.1 % (m/v), in methanol,
- solution of saccharine at 0.1 g per 100 mL in a mixture of equal parts
methanol and water,
- cyclamate solution: solution containing 1 g of the sodium or calcium salt of
cyclohexylsulfamic acid in 100 mL of a mixture of equal parts methanol and
water,
- solution of vanillin at 1 g /100 mL in a mixture of equal parts methanol and
water,
- solution of the ester of p-hydroxybenzoic acid at 1 g /100 mL in methanol.
2.3 Procedure:
50 mL of wine is placed in a separatory funnel, acidified with 10 mL dilute
sulfuric acid (2.2.6) and extracted with two aliquots of the ion exchange
solution using 25 mL each time. The 50 mL of ion exchange solution is
washed three times using 50 mL of distilled water each time, which is
discarded, then three times with 15 mL of dilute ammonium hydroxide (2.2.7).
The ammonia solutions recovered are then carefully evaporated at 50°C until
dry on a water bath or in a rotary evaporator. The residue is recovered with 5
mL of acetone and 2 drops 1 M nitric acid solution, filtered, and again
OIV-MA-AS315-07B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial sweeteners
evaporated dry at 70°C on a water bath. It is necessary to avoid heating for too
long and above 70°C. The residue is recovered with 1 mL of methanol.
5 to 10 µL of this solution and 2 µL of the standard solutions are spotted on
the plate. Let the solvent migrate (xylene: n-propanol: acetic acid: formic acid)
(2.2.10) to a height of about 15 cm, which takes about 1 hour.
After air-drying, the dichlorofluorescein solution is thoroughly sprayed on the
plate. The saccharine and the cyclamate appear immediately as light spots on a
salmon colored background. Under examination in ultraviolet light (254 or
360 nm), the three sweeteners appear as a fluorescent blue on a yellow
background.
The sweeteners separate, from the bottom to the top of the plate, in the
following order: cyclamate, saccharine, Dulcin.
The vanillin and the esters of p-hydroxybenzoic acid migrate with the same Rf
as the Dulcin. To identify Dulcin in the presence of these substances, the plate
then must be sprayed with a solution of dimethylaminobenzaldehyde. The
Dulcin appears as an orange spot, whereas the other substances do not react.
Sensitivity - The quantity limitation shown on the chromatography plate is 5 µg
for the three substances.
This method permits detection of:
 Saccharin .....................................................………….… 10 mg/L
 Cyclamate ......................................................………... 50 mg/L
 Dulcin ..................................................…………........... 10 mg/L
BIBLIOGRAPHY
 TERCERO C., F.V., O.I.V., 1968, n° 277 and F.V., O.I.V., 1970, n° 352.
 Wine Analysis Commission of the Federal Health Department of Germany,
1969, F.V., O.I.V., n° 316.
 International Federation of Fruit Juice Manufacturers, 1972, F.V., O.I.V., n°
40.
 SALO T., ALRO E. and SALMINEN K., Z. Lebensmittel Unters. u.
Forschung, 1964, 125, 20.
OIV-MA-AS315-07B : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial colorants
Method OIV-MA-AS315-08
Type IV method
Examination of artificial colorants
1. Principle
The wine is concentrated to 1/3 its original volume, made alkaline with a solution
of dilute sodium hydroxide and extracted with ether. The ether phase, after being
washed with water, is extracted with a dilute acetic acid solution; this acetic
solution, alkalinized with ammonia, is brought to boiling in the presence of a piece
of wool thread treated with aluminum sulfate and potassium tartrate. The colorant,
if any, is fixed on the wool. The wool on which it is fixed is then placed in a dilute
acetic acid solution. After evaporation of the acetic solution, the residue is
recovered with a water-alcohol solution and analyzed by thin layer
chromatography for characterization of the colorant.
The aqueous phase remaining after the ether extraction contains the acid colorants
that may be present. They are extracted by using their affinity for animal fibers
that markedly absorb the color: they are fixed on a wool plug in a mineral acid
medium.
To concentrate the coloring material, carry out a double fixation and/or several
successive fixations on increasingly smaller wool plugs.
Coloring of the wool plug indicates that an artificial colorant was added to the
wine; the colorant is then identified by thin layer chromatography.
2. Apparatus
2.1 20 x 20 glass plates covered with cellulose powder,
2.2 Chromatography tank
3. Reagents
3.1 Ethyl ether
3.2 Sodium hydroxide solution, 5% (m/v)
3.3 Glacial acetic acid (20= 1.05 g/mL)
3.4 Dilute acetic acid, containing one part glacial acetic acid to 18 parts water
3.5 Dilute hydrochloric acid: to one part hydrochloric acid (20 = 1.19 g/mL), add
10 parts distilled water
3.6 Ammonium hydroxide (20 = 0.92 g/mL)
3.7 White wool threads, previously washed, degreased with ether and dried
3.8 White wool threads, previously washed, degreased with ether, dried and
acidified
OIV-MA-AS315-08 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial colorants
Acidulant: Dissolve 1 g crystallized aluminum sulfate Al2(SO4).18H2O and 1.2
g acid potassium tartrate in 500 mL water. Place 10 g of the white wool
threads, previously washed, degreased with ether and dried in the solution and
stir about 1 hour. Let stand 2 to 3 hours; drain, let dry at room temperature.
3.9 Solvent No.1 for chromatography of colorants with basic characteristics:
n-Butanol ................................................………………....... 50 mL
Ethanol .................................................……………….......... 25 mL
Acetic acid (20 = 1.05 g/mL) ...........……....................
10 mL
Distilled water ........................................…………….......... 25 mL
3.10 Solvent No.2 for chromatography of colorants with acidic characteristics:
n-Butanol ................................................………………....... 50 mL
Ethanol ...................................................………………........ 25 mL
Ammonium hydroxide (20 = 0.92 g/mL) ................
10 mL
Distilled water ............................................……………...... 25 mL
4. Procedure
4.1 Examination of colorants with basic characteristics.
4.1.1 Extraction of the coloring materials.
Place 200 mL of wine in a 500 mL glass conical flask and boil until reduced to
1/3 its volume.
After cooling, neutralize with 5% sodium hydroxide solution until the natural
color of wine shows a marked change.
Extract twice using 30 mL ether. The ether phases are recovered, containing
basic colorants to be determined; the extraction residue must be saved for the
analysis of acidic colorants.
Wash the extracted ether twice with 5 mL of water to eliminate the sodium
hydroxide; mix with 5 mL dilute acetic acid. The acidic aqueous phase
obtained is colored in the presence of a basic colorant.
The presence of the colorant may be confirmed by fixation on acidified wool.
Make the acidic aqueous phase obtained alkaline using 5% ammonia. Add 0.5
g acidified wool and boil for about 1 minute. Rinse the wool under running
water. If the wool is colored, the wine contains some basic colorant.
4.1.2 Characterization by thin-layer chromatography.
The aqueous acetic phase containing the basic colorant is concentrated to 0.5
mL. If the colorant is fixed on the acidic wool, the wool plug is treated by
boiling with 10 mL distilled water and a few drops of acetic acid (20 = 1.05
g/mL). Remove the wool fragment after wringing out liquid. Concentrate the
solution to 0.5 mL.
OIV-MA-AS315-08 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial colorants
Deposit 20 µL of this concentrated solution on the cellulose plate 3 cm from
the lateral edge and 2 cm from the lower edge of the plate.
Place the plate in the tank containing solvent No.1 so that the lower edge is
immersed in the solvent to a depth of 1 cm.
When the solvent front has migrated to a height of 15 to 20 cm, remove the
plate from the tank. Allow to air dry.
Identify the colorant by means of a solution of known artificial colorants of
basic characteristics deposited simultaneously on the chromatogram.
4.2 Examination of colorants with acidic characteristics
4.2.1 Extraction of the coloring material.
Use the residue from the wine used for examining colorants with basic
characteristics, concentrated to 1/3 and neutralized after extraction with ether.
If the first part of the procedure has not been conducted, start with 200 mL
wine, place in a conical flask, boil until reduced to 1/3.
In either case, add 3 mL of dilute hydrochloric acid and 0.5 g of white wool:
boil for 5 minutes, decant the liquid and wash the wool under running water.
In the conical flask which contains the wool, add 100 mL water and 2 mL
dilute hydrochloric acid; boil for 5 minutes, separate the acidic liquid and
repeat this procedure until the liquid used to wash is colorless.
After the wool has been thoroughly washed to eliminate the acid completely,
recover in a conical flask with 50 mL distilled water and a few drops of
ammonium hydroxide (20 = 0.92 g/mL): bring to a gentle boil for 10 minutes
in order to dissolve any artificial coloring matter fixed on the wool.
Remove the wool from the flask, bring the liquid volume to 100 mL and boil
until the ammonia completely evaporates. Acidify with 2 mL of dilute
hydrochloric acid (check that the reaction of the liquid is definitely acidic by
placing 1 drop of this liquid on indicator paper).
Add to the flask 60 mg (about 20 cm of standard thread) of white wool and
boil for 5 minutes; remove the wool and rinse it under running water.
If, after this procedure, the wool is colored red, when it involves red wine, or
yellow if it pertains to white wine, the presence of artificial organic coloring
matter of an acidic nature is proven.
If the color is weak or uncertain, repeat the ammonia treatment and do a
second fixation using a 30 mg wool thread.
If, during the course of the second fixation a weak but distinct pink color is
obtained, assume the presence of an acidic colorant.
If necessary for a more definite determination, carry out new fixations-elutions
(up to 4 or 5) using a procedure identical to that used for the second fixation
until a faint but distinct pink color is obtained.
OIV-MA-AS315-08 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Artificial colorants
4.2.2 Characterization by thin layer chromatography.
The plug of colored wool is treated by boiling with 10 mL distilled water and
few drops of ammonium hydroxide (20 = 0.92 g/mL). Recover the piece of
wool after wringing. Concentrate the ammonium hydroxide solution to 0.5
mL.
Deposit 20 µL of this solution on a cellulose plate to within 3 cm of the lateral
edge and 2 cm of the lower edge of the plate.
Put the plate in place in the tank so that the lower edge is immersed in the
solvent to a depth of 1 cm.
When the solvent front has migrated to a height of 15 to 20 cm, remove the
plate from the tank and let dry in the air.
Identify the colorant by means of known artificial coloring solutions deposited
simultaneously on the chromatogram.
BIBLIOGRAPHY
TERCERO C., F.V., O.I.V., 1970, n° 356.
ARATA P., SAENZ-LASCANO-RUIZ, Mme I., F.V., O.I.V., 1967, n° 229.
OIV-MA-AS315-08 : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Diethylene glycol
Method OIV-MA-AS315-09
Type IV method
Diethylene glycol
(2–hydoxy-ethoxyethanol)
1. Objective
The detection of diethylene glycol, HOCH2CH2OCH2CH2OH, in wine where its
concentration is equal to or greater than 10 mg/L.
2. Principle
Separation of diethylene glycol from other constituents in wine by gas
chromatography using a capillary column, after extraction with ether.
Note: The operating conditions described below are provided as an example.
3. Apparatus
3.1 Gas chromatograph equipped with:
- split-splitless injector,
- flame ionization detector,
- capillary column coated with a film of polyethyleneglycol (Carbowax
20 M), 50 m x 0.32 mm I.D.
Operating conditions:
Injector temperature: 280°C.
Detector temperature: 270°C.
Carrier gas: hydrogen.
Flow rate of carrier gas: 2 mL/min.
Flow rate: 30 mL/min.
Injection: splitless.
Injection volume: 2 µL.
Injection 35°C - flow closed after 40 seconds.
Temperature program: 120°C to 170°C at 3°C/min.
3.2 Centrifuge
4. Reagents
4.1 1,3-propanediol, 1 g/L, in alcohol, 20% (v/v), (internal standard).
4.2 Aqueous solution of diethyleneglycol 20 mg/L.
5. Procedure
Into a 50 mL flask, place:
OIV-MA-AS315-09 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Diethylene glycol
- 10 mL of wine
- 1 mL of 1,3-propanediol solution
- 25 mL diethyl ether.
Shake and add sufficient quantity of neutral potassium carbonate to saturate the
mixture. Shake. Separate the two phases by centrifugation.
Carry out a second extraction. Eliminate the diethyl ether by evaporation and
recover the residue with 5 mL ethanol.
The yield of the extraction must be at least 90%.
Carry out the chromatography according to the conditions given in 3.1.
6. Results
The diethylene glycol is identified by comparing its retention time to the time of
the reference solution, analyzed under the same conditions as the wine.
The amount is determined by comparison to the reference solution using the
internal standard method.
It is recommended, if the concentration is equal to or less than 20 mg/l, to confirm
the presence by mass spectrometry.
BIBLIOGRAPHY
BANDION F., VALENTA M. & KOHLMANN H., Mitt. Klosterneuburg, Rebe
und Wein, 1985, 35, 89.
BERTRAND A., Conn. vigne vins, 1985, 19, 191.
Laboratoire de la répression des fraudes et du contrôle de la qualité de
Montpellier, F.V., O.I.V., 1986, n° 807.
OIV-MA-AS315-09 : R2009
2
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
Method OIV-MA-AS315-10
Type II method
Measuring ochratoxine A in wine
after going through an immunoaffinity column
and HLPC with fluorescence detection
1. FIELD OF APPLICATION
This document describes the method used for determining ochratoxine A (OTA) in
red, rosé, and white wines, including special wines, in concentrations ranging up 10
µg/l using an immunoaffinity column and high performance liquid chromatography
(HPLC) [1].
This method was validated following an international joint study in which OTAs
were measured in white and red wines during the analysis of naturally
contaminated wines and wines with toxins ranging from 0.01 µg/l to 3.00 µg/l.
This method can apply to semi-sparkling wines and sparkling wines as long as the
samples have been degassed beforehand, through sonication, for example.
2. PRINCIPLE
Wine samples are diluted with a solution containing polyethylene glycol and
sodium hydrogen carbonate. This solution is filtered and purified on the
immunoaffinity column.
OTA is eluted with methanol and quantified by HPLC in inverse state with
fluorimetric detection.
3. REAGENTS
Unless otherwise indicated, use only those reagents known for the quality of
analysis, distilled water, or water with the EN ISO 3696. Solvents must be HPLC
quality.
3.1 Sodium chloride (NaCl)
3.2 Sodium hydrogen carbonate (NaHCO3)
3.3 Polyethylene glycol (PEG 8000)
3.4 Methanol (CH3OH)
3.5 Acetonitrile (CH3CN)
3.6 Purified water for laboratories, for example EN ISO 3696 quality (water for
analytical laboratory use - Specification and test method [ISO 3696:1987]).
OIV-MA-AS315-10 : R2009
1
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
3.7 Acetic acid 85% (CH3COOH)
3.8 Dilution solution (1% PEG + 5% NaHCO3)
Dissolve 10 g of PEG (3.3) and 50 g of NaHCO3 (3.2) in 950 ml of water and fill
up with water to the 1 litre mark.
3.9 Washing solution (2,5% de NaCl + 0,5 % NaHCO3)
Dissolve 25 g of NaCl (3.1) and 5 g of NaHCO3 (3.2) in 950 ml of water and fill up
with water to the 1 litre mark.
3.10 Mobile HPLC phase ( water: acetonitrile: glacial acetic acid, 99:99:2, v/v/v)
Mix 990 ml of water with 990 ml of acetonitrile (3.5) and 20 ml of glacial acetic
acid (3.7). Filter through a 0,45 µm filter and degas, with helium for example.
3.11 Toluene
3.12 Mixture of solvents (Toluene: a glacial acetic acid, 99:1, v/v).
Mix 99 parts in volume of toluene (3.11) with one part volume of glacial acetic
acid (3.7).
3.13 OTA stock solution
Dissolve 1 mg of OTA or the same content in a bulb, if the OTA was obtained in
the form of film after evaporation, in the solvent mixture (3.12) to obtain a solution
containing approximately 20 to 30 µg/ml of OTA.
To determine the exact concentration, record the absorption spectrum between 300
and 370 nm in a quartz space with 1 cm of optical path while using the solvent
mixture (3.12) as a blank. Identify maximum absorption and calculate the
concentration of OTA (c) in µg/ml by using the following equation:
c = Amax  M  100 /   
In which:
Amax = Absorption determined by the longest maximum wave (about 333 nm)
M = OTA molecular mass = 403,8 g/mole
 = coefficient d'extinction molaire de l'OTA dans le mélange de solvant (3.12) ( =
544/mole)
 = optical pathway (cm)
This solution is stable at -18°C for at least 4 years.
3.14 Standard OTA solution (2 µg/ml in toluene: acetic acid, 99:1, v/v)
Dilute the stock solution (3.13) with the solvent mixture (3.12) to obtain a standard
solution of OTA with a concentration of 2 µg/ml.
OIV-MA-AS315-10 : R2009
2
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
This solution can be stored at + 4 °C in a refrigerator. The stability should be tested
regularly.
4. EQUIPMENT
Usual laboratory equipment and in particular, the following equipment:
4.1 Glass tubes (4 ml)
4.2 Vacuum pump to prepare the immunoaffinity columns.
4.3 Reservoir and flow tube adapted to immunoaffinity columns.
4.4 Fibre glass filters (for example Whatman GF/A).
4.5 Immunoaffinity columns specifically for OTA.
The column should have the total link capacity of at least 100 ng OTA. This will
allow for a purification yield of at least 85% when a diluted solution of wine
containing 100 ng OTA is passed through.
4.6 Rotating evaporator
4.7 Liquid chromatography, a pump capable of attaining a constant flow of 1
ml/mn isocratic, as with the mobile phase.
4.8 Injection system must be equipped with 100 µl loop.
4.9 Column of analytical HPLC in steel 150  4.6 mm (i.d.) filled with a stationary
phase C18 (5 µm) preceded with a pre-column or a pre-filter (0,5 µm) containing an
appropriate phase. Different size columns can be used provided that they guarantee
a good base line and background noise enabling the detection of of OTA peaks,
among others.
4.10 Fluorescence detector is connected to the column and the excitation
wavelength is set at 333 nm and the emitting wavelength at 460 nm.
4.11 Information retrieval system
4.12 U.V. spectrometer
5. PROCEDURE
5.1 Preparation of samples
Pour 10 ml of wine in a 100 ml conical flask. Add 10 ml of the dilution solution
(3.8). Mix vigorously. Filter through fibreglass filter (4.4). Filtration is necessary
for cloudy solutions or when there is precipitation after dissolving.
5.2 Purification by immunoaffinity column
Set up the by immunoaffinity column (4.5) to the vacuum pump (4.2), and attach
the reservoir (4.3).
OIV-MA-AS315-10 : R2009
3
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
Add 10 ml (equivalent to 5 ml of wine) of the diluted solution in the reservoir. Put
this solution through the immunoaffinity column at a flow of 1 drop per second.
The immunoaffinity column should not become dry. Wash the immunoaffinity
column with 5 ml of cleaning solution (3.9) and then with 5 ml of water at a flow
of 1 to 2 drops per second.
Blow air through to dry column. Elute OTA in a glass flask (4.1) with 2 ml of
methanol (3.4) at the rate of 1 drop per second. Evaporate the eluate to dryness at
50° C with nitrogen. Dissolve again immediately in 250 µl of the HPLC mobile
phase (3.10) and keep at 4° C until the HPLC analysis.
5.3
HPLC analysis
Using the injection loop, inject 100 µl of reconstituted extract (equivalent to 2 ml
of wine) in the chromatography.
Operating conditions
Flow:
1 ml /min.
Mobile phase:
acetonitrile: water: glacial acetic acid (99:99:2, v/v/v)
Fluorescence detector: Excitation wavelength = 333 nm
Emitting wavelength = 460 nm
Volume of injection: 100 µl
6. QUANTIFICATION OF OCHRATOXINE A (OTA)
The quantification of OTA should be calculated by measuring the area or the
height of the peaks at the OTA retention time and compared to the calibration
curve
6.1 Calibration curve
Prepare a calibration curve dayly and every time chromatographical conditions
change. Measure out 0.5 ml of the standard OTA solution (3.14) at 2 µg/ml in a
glass flask and evaporate the solvent using nitrogen.
Dissolve again in 10 ml in the HPLC mobile phase (3.10) which was previously
filtered using a 0.45 µm filter. This produces an OTA of 100 ng/ml solution.
Prepare 5 HPLC calibration solutions in five 5 ml graduated flasks following Table
1.
Complete each 5 ml standard solution with HPLC mobile phase. (3.10).
Inject 100 µl of each solution in the HPLC.
OIV-MA-AS315-10 : R2009
4
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
Table 1
Std 1
Std 2
Std 3
Std 4
Std 5
µl of mobile phase filtered HPLC (3.10)
4970
4900
4700
4000
2000
µl of OTA solution at 100 ng/ml:
OTA concentration (ng/ml)
Injected OTA (ng)
30
0.6
0.06
100
2.0
0.20
300
6.0
0.60
1000
20
2.00
3000
60
6.00
NOTE:
1. If the quantity of OTA in the samples is outside the calibration range, an
appropriate dilution should occur or smaller volumes should be injected. In these
cases, the final (7) should be reviewed on a case by case basis.
2. Due to the great variations in concentrations, it is recommended to pass the
linear calibration by zero in order to obtain an exact quantification for low
concentrations of OTA. (less than 0.1 µg/l)
7. CALCULATIONS
Calculate the quantity of OTA in the aliquot of the solution testes and injected in
the HPLC column.
Calculate the concentration of OTA (COTA) in ng/ml (equivalent to µg/l) by using
the following formula:
COTA = MA  2/V1  V3/V2
Where:
MA is the volume of ochratoxin A (in ng) in the aliquot part of the template
injected on the column and evaluated from the calibration curve.
2 is the dilution factor
V1 is the sample volume to be analysed (10 ml)
V2 the volume of the solution tested and injected in the column (100 l)
V3 is the volume of solution used to dissolve the dry eluate (250 l)
8. PERFORMANCES USING THIS METHOD IN LABORATORIES
Table 2 regroups performances of the method applied to white, rosé and red wines
in laboratories participating in the validation of this method.
OIV-MA-AS315-10 : R2009
5
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
Table 2. Recovery of ochratoxin A from wines overweighted with different
concentrations of added ochratoxin A
Addition
(µg/l)
0.04
0.1
0.2
0.5
1.0
2.0
5.0
10,0
Average of
averages
Red wine
Yield ± SD* RSD#
(%)
(%)
96.7 ± 2.2
2.3
90.8 ± 2.6
2.9
91.3 ± 0.6
0.7
92.3 ± 0.4
0.5
97.8 ± 2.6
2.6
96.5 ± 1.6
1.7
88.1 ± 1.3
1.5
88,9 ± 0,6
0,7
92.8 ± 3.5
3.8
Rosé wine
White wine
#
Yield ± SD* RSD Yield ± SD* RSD#
(%)
(%)
(%)
(%)
94.1 ± 6.1
6.5
91.6 ± 8.9
9.7
89.9 ± 1.0
1.1
88.4 ± 0.2
0.2
88.9 ± 2.1
2.4
95.1 ± 2.4
2.5
91.6 ± 0.4
0.4
93.0 ± 0.2
0.2
100.6 ± .,5
2.5
100.7 ± 1.0
1.0
98.6 ± 1.8
1.8
98.0 ± 1.5
1.5
94.5 ± 5.2
5.5
94.5 ± 4.1
4.3
* SD = Spread type (Standard deviation) (n = 3 replicates) ;
#
RSD = Relative spread type (Variation percentage).
9. GROUP WORK
The method was validated by a group study with the participation of 16
laboratories in 8 countries, following the protocol recommendations harmonised
for validating the analysis methods. [2]. Each participant analysed 10 white wines,
10 red wines, representing 5 random duplicate wines; naturally contaminated or
with OTA added. The performances of the method which resulted from this work
are found in appendixes I and II.
OIV-MA-AS315-10 : R2009
6
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
10. PARTICPATING LABORATORIES
Unione Italiana Vini, Verona
Istituto Sperimentale per l’Enologia, Asti
Istituto Tecnico Agraria, S. Michele all’Adige (TN)
Università Cattolica, Piacenza
Institute for Health and Consumer Protection, JRC – Ispra
Neotron s.r.l., S. Maria di Mugnano (MO)
Chemical Control s.r.l., Madonna dell’Olmo (CN)
Laboratoire Toxicologie Hygiène Appliquée, Université V.
Segalen, Bordeaux
Laboratoire de la D.G.C.C.R.F. de Bordeaux, Talence
National Food Administration, Uppsala
Systembolagets Laboratorium, Haninge
Chemisches Untersuchungsamt, Trier
State General Laboratory, Nicosia
Finnish Customs Laboratory, Espoo
Central Science Laboratory, York
E.T.S. Laboratories, St. Helena, CA
ITALY
ITALY
ITALY
ITALY
ITALY
ITALY
ITALY
FRANCE
FRANCE
SWEDEN
SWEDEN
GERMANY
CYPRUS
FINLAND
UNITED
KINGDOM
UNITED STATES
11. REFERENCES
[1]
A. Visconti, M. Pascale, G. Centonze. Determination of ochratoxin A in
wine by means of immunoaffinity column clean-up and high-performance liquid
chromatography. Journal of Chromatography A, 864 (1999) 89-101.
[2]
AOAC International 1995, AOAC Official Methods Program, p. 23-51.
OIV-MA-AS315-10 : R2009
7
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
APPENDIX I
The following data was obtained in inter-laboratory tests, according to harmonised
protocol recommendations for joint studies in view of validating an analysis
method.
WHITE WINE
Sample
White
Inter-laboratory test year
1999
Number of laboratories
16
Number
of
laboratories
retained
after
eliminating
14*
absurd findings
Number
of
eliminated
laboratories
Number of accepted results
28
Average value (µg/l)
<0,01
Spread-type/Repeatabilityr
(µg/l)
Relative spread-type (Variation
percentage)
/Repeatability
RSDr (%)
Repeatability limit r (µg/l)
Spread-type/capacity of being
reproduced sR (µg/l)
Relative spread-type (variation
percentage) /capacity of being
reproduced RSDR (%)
Capacity of being reproduced
limit R (µg/l)
Extraction yield %
* 2 laboratories were excluded from the
detection limit (= 0,2 µg/l).
n.c. = sample naturally contaminated
OIV-MA-AS315-10 : R2009
Added OTA (µg/l)
0.100
1.100
2.000
1999
1999
1999
16
16
16
n.c.
1999
16
13*
14
14
15
1
2
2
1
26
0,102
28
1,000
28
1,768
30
0,283
0.01
0.07
0.15
0.03
10.0
6.6
8.5
10.6
0.028
0.196
0.420
0.084
0.01
0.14
0.23
0.04
14.0
13.6
13.3
14.9
0.028
0.392
0.644
0.112
101.7
90.9
88.4
statistical 'evaluation due to high
8
COMPENDIUM OF INTERNATIOAL METHODS OF ANALYSIS - OIV
Ochratoxin A
APPENDIX II
The following data was obtained in inter-laboratory tests, according to harmonised
protocol recommendations for joint studies in view of validating an analysis
method.
RED WINE
Added OTA (µg/l)
samples
White
0.200
0.900
3.000
Inter-laboratory test year
1999
1999
1999
1999
Number of laboratories
15
15
15
15
Number of laboratories retained
14*
12*
14
15
after eliminating absurd findings
Number
of
eliminated
2
1
laboratories
Number of accepted results
28
24
28
30
Average value (µg/l)
<0.01
0.187
0.814
2.537
Spread-type/Repeatabilityr (µg/l)
0.01
0.08
0.23
Relative spread-type (Variation
percentage) /Repeatability RSDr
5.5
9.9
8.9
(%)
Repeatability limit r (µg/l
0.028
0.224
0.644
Spread-type/capacity of being
0.02
0.10
0.34
reproduced sR (µg/l )
Relative spread-type (variation
percentage) /capacity of being
9.9
12.5
13.4
reproduced RSDR (%)
Capacity of being reproduced
0.056
0.280
0.952
limit R (µg/l)
Extraction yield %
93.4
90.4
84.6
* 1 laboratory was excluded from the statistical evaluation because
detection limits (= 0,2 µg/l).
n.c. = naturally contaminated sample
OIV-MA-AS315-10 : R2009
n.c.
1999
15
14
1
28
1.693
0.19
10.9
0.532
0.23
13.4
0.644
of high
9
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
Method OIV-MA-AS315-11
Type II method
HPLC-Determination of nine major anthocyanins
in red and rosé wine
(Resolution Oeno 22/2003 modified by Oeno 12/2007)
1. FIELD OF APPLICATION
The analytical method concerns the determination of the relative composition
of anthocyanins in red and rosé wine. The separation is performed by HPLC
with reverse phase column and UV-VIS detection.
Many authors [3, 6-17] have published data on the anthocyanin composition of
red wines using similar analytical methods. For instance Wulf et al. [18] have
detected and identified 21 anthocyanins and Heier et al. [13] nearly 40 by liquid
chromatography combined with mass spectrometry. The anthocyanin
composition may be very complex, so it is necessary to have a simple
procedure. Consequently this method only determines the major compounds of
the whole anthocyanin fraction.
Member states are encouraged to continue research in this area to avoid any
non scientific evaluation of the results.
2. PRINCIPLE
Separation of the five most important non acylated anthocyanins (see Figure 1,
peaks 1-5) and four major acylated anthocyanins (see Figure 1, peaks 6-9).
Analysis of red and rosé wine by direct separation by HPLC by using reverse
phase column with gradient elution by water/formic acid/acetonitrile with
detection at 518 nm [1.2].
3 REAGENTS AND MATERIAL
Formic acid (p.a. 98 %) (CAS 64-18-6);
Water, HPLC grade;
Acetonitrile, HPLC grade (CAS 75-08-8);
HPLC solvents:
Solvent A: Water/Formic acid/Acetonitrile 87 : 10 : 3
Solvent B: Water/Formic acid/Acetonitrile 40 : 10 : 50
OIV-MA-AS315-11 : R2007
(v/v/v)
(v/v/v)
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
Membrane filter for HPLC solvent degassing and for sample preparation to be
analysed.
Reference products for peak identification.
The HPLC analysis of anthocyanins in wine is difficult to perform due to the
absence of commercially available pure products. Furthermore, anthocyanins
are extremely unstable in solution.
The following anthocyanin pigments are commercially available:
Cyanidol-3-glucoside (also couromanin chloride); M = 484.84 g/mol
Peonidol-3-glucoside; M = 498.84 g/mol
Malvidol-3-glucoside (also Oeninchloride); M = 528.84 g/mol
Malvidol-3,5-diglucoside (also Malvinchloride); M = 691.04 g/mol
4. APPARATUS
HPLC system with:
binary gradient pump, injection system for sample volumes ranging from 10 to
200 µl,
diode array detector or a UV detector with a visible range,
integrator or a computer with data acquisition software,
furnace for column heating at 40°C,
solvent degassing system,
analytical column, for example:
LiChrospher 100 RP 18 (5 µm) in LiChroCart 250-4 guard column: for
example RP 18 (30-40 mm) in a cartridge 2 mm in diameter x 20 mm long
5. PROCEDURE
5.1 Preparation of samples
Clear wines are poured directly without any preparation into the sample vials of
the automatic sample changer. Cloudy samples are filtered using a 0.45 µm
membrane filter for HPLC sample preparation. The first part of the filtrate
should be rejected.
Since the range of the linearity of absorption depending on the concentration of
anthocyanins is large, it is possible to modulate the injection volumes between
10 and 200 µl depending on the intensity of the wine colour. No significant
difference between the results obtained for different injection volumes was
observed.
OIV-MA-AS315-11 : R2007
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
5.2 Analysis
HPLC conditions
The HPLC analysis is carried out in the following conditions:
Injection Volume:
Flow:
Temperature:
Run time:
Post time:
Detection:
Gradient elution:
50 µl (red wine) up to 200 µl (rosé wine)
0.8 ml/minute
40°C
45 minutes
5 minutes
518 nm
Time
(min)
Solvent A
% (v/v)
Solvent B
% (v/v)
0
15
30
35
41
94
70
50
40
94
6
30
50
60
6
To check the column efficiency, the number of theoretical plates (N) calculated
according to malvidol-3-glucoside should not be below 20,000, and the
resolution (R) between peonidol-3-coumaryl glucoside and malvidolin-3coumaryl glucoside should not be lower than 1.5. Below these values, the use
of a new column is recommended.
OIV-MA-AS315-11 : R2007
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
A typical chromatogram is given in Figure 1, where the following anthocyanins
are separated:
Group 1: “Nonacylated delphinidol-3-glucoside
anthocyanidin-3cyanidol-3-glucoside
glucosides”:
petunidol-3-glucoside
peonidol-3-glucoside
malvidol-3-glucoside
Peak-N°
1
2
3
4
5
Group 2: “Acetylated
anthocyanidin-3glucosides”:
peonidol-3-acetylglucoside
malvidol-3-acetylglucoside
6
7
Group 3:
“Coumarylated
anthocyanidin-3glucosides”:
peonidol-3-coumarylglucoside
malvidol-3-coumarylglucoside
8
9
6. EXPRESSION OF RESULTS
Note that the values are expressed as relative amounts of the sum of the nine
anthocyanins defined in this method.
7. LIMIT OF DETECTION AND LIMIT OF QUANTIFICATION
The limit of detection (LD) and the limit of quantification (LQ) are estimated
following the instructions in the resolution OENO 7-2000 “Estimation of the
Detection and Quantification Limits of a Method of Analysis“. Along the line of
the ”Logic Diagram for Decision-Making” in N° 3 the graph approach has to be
applied following paragraph 4.2.2.
For this purpose a part of the chromatogram is drawn out extendedly enclosing a
range of a tenfold mid-height width (w½) from an anthocyan relevant peak.
Furthermore two parallel lines are drawn which just enclose the maximum
amplitude of the signal window. The distance of these two lines gives h max,
expressed in milli Absorption Units (mAU).
OIV-MA-AS315-11 : R2007
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
The limit of detection (LD) and the limit of quantification (LQ) depend on the
individual measurement conditions of the chemical analysis and are to be
determined by the user of the method. The Annex gives an example of its
determination with the following results:
hmax = 0,208 [mAU]; LD = 3 x 0,208 [mAU] = 0,62 [mAU].
LQ = 10 x 0, 208 [mAu] = 2,08 [mAU].
Recommendation:
With combined data out of the whole Anthocyanin composition such as the sum of
Acylated Anthocyanins or the ratio of Acetylated to Coumarylated Anthocyanins
the calculation should not be carried out in cases where one of the components is
below the limit of quantification (LQ).
On the other hand measurements below the limit of quantification (LQ) are not
devoid of information content and may well be fit for purpose [1].
Bibliography:
[1] Thompson, M.; Ellison, S.L.R. ; Wood, R., Harmonized Guidelines for SingleLaboratory Validation of Methods of Analysis, Pure Appl. Chem. (2002) 74: 835855
8. FIDELITY PARAMETERS
The repeatability (r) and the reproducibility (R) values for the nine
anthocyanins are given in Table 2 and depend on the amount of the peak area.
The uncertainty measurement of a particular peak area is determined by the
value of r and R which corresponds to the nearest value given in Table 2.
The values made up of validation data can be calculated by following the
appropriate statistical rules. To calculate the total error (sr) for example of the
sum of acetylated anthocyanins, the variances (sr2) of specific the total error of
ratios, for example, that of acetylated to coumarylated anthocyanins the square
of relative errors (=sr/ai) are to be added. By using these rules, all the fidelity
values can be calculated by using the data in Table 2.
OIV-MA-AS315-11 : R2007
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
OIV-MA-AS315-11 : R2007
6
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
Annex A
Bibliography
[1]
Marx, R., B. Holbach, H. Otteneder; Determination of nine characteristic
Anthocyanins in Wine by HPLC; OIV, F.V.N° 1104 2713/100200
[2]
Holbach, B., R. Marx, M. Ackermann; Bestimmung der
Anthocyanzusammensetzung von Rotwein mittels
Hochdruckflüssigkeitschromatographie (HPLC).
Lebensmittelchemie (1997) 51: 78 - 80
[3]
Eder, R., S. Wendelin, J. Barna; Auftrennung der monomeren
Rotweinanthocyane mittels Hochdruckflüssigkeitschromatographie
(HPLC).Methodenvergleich und Vorstellung einer neuen Methode. Mitt.
Klosterneuburg (1990) 40: 68-75
[4]
ISO-5725-2: 1994 “Accuracy (trueness and precision) of measurement
methods and results - Part 2: Basic method for the determination of
repeatability and reproducibility”
[5]
Otteneder, H., Marx, R., Olschimke, D.; Method-performance study on
the determination of nine characteristic anthocyanins in wine by HPLC.
O.I.V. F.V.N° 1130 (2001)
[6]
Mattivi F.; Scienza, A.; Failla, O.; Vika, P.; Anzani, R.; Redesco, G.;
Gianazza, E.; Righetti; P. Vitis vinifera - a chemotaxonomic approach:
Anthocyanins in the skin. Vitis (special issue) 1990, 119-133
[7]
Roggero, I.P.; Larice, I.L.; Rocheville-Divorne, C.; Archier, P.; Coen, V.
Composition Antocyanique des cepages. Revue Francaise d’Oenologie
1998, 112, 41-48
[8]
Eder, R.; Wendelin, S; Barna, J. Classification of red wine cultivars by
means of anthocyanin analysis. Mitt. Klosterneuburg 1994, 44, 201-212
[9]
Arozarena, I.; Casp, A.; Marin, R.; Navarro, M. Differentiation of some
Spanish wines according to variety and region based on their anthocyanin
composition. Eur. Food Res. Technol. 2000, 212, 108-112
[10]
Garcia-Beneytez, E.; Revilla, E.; Cabello, F. Anthocyanin pattern of
OIV-MA-AS315-11 : R2007
7
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
several red grape cultivars and wines made from them. Eur. Food Res.
Technol. 2002, 215, 32-37
[11]
Arozarena, I.; Ayestarán, B.; Cantalejo, M.J.; Navarro, M.; Vera, M.;
Abril, K.; Casp, A. Eur. Food Res. Technol. 2002, 214, 313-309
[12]
Revilla, E.; Garcia-Beneytez, E.; Cabello, F.; Martin-Ortega, G.; Ryan, JM. Value of high-performance liquid chromatographic analysis of
anthocyanins in the differentiation of red grape cultivars and red wines
made from them. J. Chromatogr A 2001, 915, 53-60
[13]
Heier, A.; Blaas, W.; Droß, A.; Wittkowski, R.; Anthocyanin Analysis by
HPLC/ESI-MS, Am.J.Enol.Vitic, 2002, 53, 78-86
[14]
Arozarena, I.; Casp, A.; Marin, R.; Navarro, M. Multivariate
differentiation of Spanish red wines according to region and variety. J.
Sci. Food Agric, 2000, 80, 1909-1917
[15]
Anonymous. Bekanntmachung des Bundesinstituts für gesundheitlichen
Verbraucherschutz und Veterinärmedizin. Bundesgesundheitsbl.
Gesundheitsforsch. Gesundheitsschutz, 2001, 44, 748
[16]
Burns, I.; Mullen, W.; Landrault, N.; Teissedre, P.-L.; Lean, M.E.I.;
Crozier, A. Variations in the Profile and Content of Anthocyanins in
Wines made from Cabernet Sauvignon and hybrid grapes. J. Agric. Food
Chem. 2002, 50, 4096-4102
[17]
Otteneder, H.; Holbach, B.; Marx, R.; Zimmer, M. Rebsortenbestimmung
in Rotwein mittels Anthocyanspektrum. Mitt. Klosterneuburg, 2002, 52,
187-194
[18]
L.W. Wulf and C.W. Nagel; High-Pressure liquid chromatographic
separation of Anthocyanins of Vitis vinifera.
Am.J.Enol.Vitic 1978, 29, 42-49
OIV-MA-AS315-11 : R2007
8
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
Annex B
Statistical results
Method performance study and evaluation
17 laboratories from 5 European Nations participated in the validation study of
the method under the coordination of the German Official State Laboratory for
Food Chemistry in Trier. The participants are listed in Table 3. An example of a
chromatogram is presented in Figure 1 and the detailed results are given in Table
2.
The statistical evaluation followed the Resolution 6/99 and the Standard ISO
5725-1944 [4.5].
The chromatograms sent back with the results sheets fulfilled all requirements
concerning the performance of the analytical column. No laboratory had to be
completely eliminated, for example, because of a wrong peak identification.
The outlier values were searched using Dixon and Grubbs outlier testing
according to the procedure for “Harmonised Protocol – IUPAC 1994” and the
OIV Resolution OENO 19/2002. The values of sr, sR, r and R were calculated
for 9 major anthocyanins at 5 content levels. For analytical results, the values of
the closest levels should be used.
In order to have a global vision of the method performance, all the values RSDret RSDR- gathered are grouped by range of areas in the following table:
Table 1: Summary of the results of the method performance study
Range of relative peak
areas*[%]
Range of RSDr
[%]
Range of RSDR
[%]
>0.4 – 1.0
6.8 - 22.4
20.6 - 50.9
>1.1 – 1.5
4.2 - 18.1
11.8 - 28.1
>1.5 – 3.5
2.1 – 7.7
10.6 - 15.6
>3.5 – 5.5
2.7 – 5.7
18.7 – 7.5
>5.5 – 7.5
2.4 – 3.9
6.5 - 10.0
>10 – 14
1.1 – 2.9
3.7 - 9.2
>14 – 17
1.0 - 3.9
3.2 - 5.4
>50 – 76
0.3 - 1.0
2.1 - 3.1
* independent of anthocyanin
OIV-MA-AS315-11 : R2007
9
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
This leads to the conclusion that repeatabilities and reproducibilities depend on
the total sum of the relative peak areas. The higher they are, the better are RSDr
and RSDR. For anthocyanin contents close to the detection limit (e.g. Cyanidin3-glucoside) with small relatives areas (less than 1%) the RSDr et RSDR values
can rise significantly. For anthocyanin whose relative areas are more than 1%,
the RSDr and RSDR values are reasonable.
Figure 1: Separation of 9 anthocyanins in red wine
ANTHO13 #25 [modified by
gyn]
mA
U
4-98
4039
UV_VIS
_1
WVL:518
nm
Mv-3gl
55
0
50
0
45
0
40
0
35
0
30
0
5
0
Po-3cuglMv-3cugl
Cy-3gl
10
0
acgl
15
0
Po-3acgl Mv-3-
Po-3gl
20
0
Pt-3gl
De-3gl
25
0
0
mi
n
50 0,
0
2,
5
5,
0
7,
5
10,
0
12,
5
15,
0
17,
5
OIV-MA-AS315-11 : R2007
20,
0
22,
5
25,
0
27,
5
30,
0
32,
5
35,
0
37,
5
40,
0
42,
5
45,
0
10
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
Table 2: Results of the method performance study
Anthocyanin
sample 1
sample 2
sample 3
Delphinidol-3-glucoside
n
14
14
16
mean
6.75
14.14
3.45
sr
0.163
0.145
0.142
RSDr(%)
2.4
1.0
4.1
r
0.46
0.41
0.40
sR
0.544
0.462
0.526
RSDR(%)
8.1
3.3
15.2
R
1.52
1.29
1.47
Cyanidol-3-glucoside
n
16
17
16
mean
2.18
1.23
0.61
sr
0.086
0.053
0.043
RSDr(%)
4.0
4.3
7.1
r
0.24
0.15
0.12
sR
0.460
0.211
0.213
RSDR(%)
21.2
17.2
34.9
R
1.29
0.59
0.60
Petunidol-3-glucoside
n
15
17
16
mean
10.24
14.29
5.75
sr
0.233
0.596
0.157
RSDr(%)
2.3
4.2
2.7
r
0.65
1.67
0.44
sR
0.431
0.996
0.495
RSDR(%)
4.2
7.0
8.6
R
1.21
2.79
1.39
Peonidol-3-glucoside
n
16
15
17
mean
11.88
6.23
13.75
sr
0.241
0.166
0.144
RSDr(%)
2.0
2.7
1.0
r
0.68
0.47
0.40
sR
0.981
0.560
1.227
RSDR(%)
8.3
9.0
8.9
R
2.75
1.57
3.44
Malvidol-3-glucoside
n
16
15
17
mean
55.90
55.04
76.11
sr
0.545
0.272
0.251
RSDr(%)
1.0
0.5
0.3
r
1.53
0.76
0.70
sR
2.026
2.649
2.291
RSDR(%)
3.6
4.8
3.0
R
5.67
7.42
6.41
n
= N° of laboratories retained after eliminating outliers
sr
= standard deviation of repeatability
RSDr(%)
= relative standard deviation of repeatability
r
= repeatability
sR
= standard deviation of reproducibility
RSDR(%)
= relative standard deviation of reproducibility
R
= reproducibility
OIV-MA-AS315-11 : R2007
sample 4
sample 5
15
16.68
0.142
0.8
0.40
0.704
4.2
1.97
16
3.54
0.108
3.1
0.30
0.490
13.8
1.37
15
1.46
0.110
7.5
0.31
0.180
12.3
0.50
14
0.34
0.031
9.2
0.09
0.158
46.7
0.44
14
12.21
0.097
0.8
0.27
0.469
3.8
1.31
15
6.19
0.196
3.2
0.55
0.404
6.5
1.13
17
7.44
0.232
3.1
0.65
0.602
8.1
1.69
16
4.12
0.174
4.2
0.49
0.532
12.9
1.49
16
52.60
0.298
0.6
0.83
1.606
3.1
4.50
16
61.04
0.377
0.6
1.06
1.986
3.3
5.56
11
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
Table 2: Results of the method performance study
Anthocyanin
Peonidol-3-acetylglucoside
n
mean
sr
RSDr(%)
sample 1
sample 2
14
1.16
0.064
5.5
0.18
0.511
43.9
1.43
16
1.44
0.062
4.3
0.17
0.392
27.2
1.10
sample 3
sR
RSDR(%)
R
Malvidol-3-acetylglucoside
n
16
17
mean
5.51
4.84
sr
0.176
0.167
RSDr(%)
3.2
3.4
r
0.49
0.47
sR
0.395
0.366
RSDR(%)
7.2
7.6
R
1.11
1.02
Peonidol-3-coumarylglucoside
n
16
14
mean
1.26
0.90
sr
0.130
0.046
RSDr(%)
10.3
5.1
r
0.36
0.13
sR
0.309
0.109
RSDR(%)
24.5
12.2
R
0.86
0.31
Malvidol-3-coumarylglucoside
n
17
17
mean
4.62
2.66
sr
0.159
0.055
RSDr(%)
3.4
2.1
r
0.45
0.15
sR
0.865
0.392
RSDR(%)
18.7
14.7
R
2.42
1.10
n
= N° of laboratories retained after eliminating outliers
sr
= standard deviation of repeatability
RSDr(%)
= relative standard deviation of repeatability
r
= repeatability
sR
= standard deviation of reproducibility
RSDR(%)
= relative standard deviation of reproducibility
R
= reproducibility
OIV-MA-AS315-11 : R2007
sample 4
sample 5
14
0.59
0.059
10.1
0.17
0.272
46.4
0.76
16
3.74
0.215
5.8
0.60
0.374
10.0
1.05
17
3.11
0.088
2.8
0.25
0.496
16.0
1.39
16
15.07
0.213
1.4
0.60
0.617
4.1
1.73
17
0.89
0.060
6.8
0.17
0.204
23.0
0.57
16
1.32
0.058
4.4
0.16
0.156
11.8
0.44
17
4.54
0.124
2.7
0.35
0.574
12.6
1.61
16
4.45
0.048
1.1
0.13
0.364
8.2
1.02
12
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Anthocyanins
Table 3: List of participants
ABC Labor Dahmen, Mülheim/Mosel
Chemisches Landes- und Staatliches
Veterinäruntersuchungsamt Münster
D
D
Institut für Lebensmittelchemie Koblenz
D
Institut für Lebensmittelchemie Speyer
D
Institut für Lebensmittelchemie Trier
D
Institut für Lebensmittelchemie und Arzneimittel Mainz
D
Labor Dr. Haase-Aschoff, Bad Kreuznach
D
Labor Dr. Klaus Millies, Hofheim-Wildsachsen
D
Labor Heidger, Kesten
D
Landesveterinär- und Lebensmitteluntersuchungsamt Halle
D
Staatliche Lehr- und Forschungsanstalt für Landwirtschaft,
Weinbau und Gartenbau, Neustadt/Weinstraße
D
Staatliches Institut für Gesundheit und Umwelt, Saarbrücken
D
Staatliches Medizinal-, Lebensmittel- und
Veterinäruntersuchungsamt, Wiesbaden
D
Laboratoire Interrégional de la D.G.C.C.R.F de Bordeaux,
Talence/France
F
Unidad de Nutricion y Bromotologia, Facultad de Farmacia,
Universidad de Salamanca, Salamanca/Espana
E
University of Glasgow, Div. of Biochem. and Molek. Biology
Höhere Bundeslehranstalt und Bundesamt für Wein- und
Obstbau, Klosterneuburg
UK
A
17 Laboratories D (13); A (1); F (1); E (1); UK (1)
OIV-MA-AS315-11 : R2007
13
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Plant proteins
Method OIV-MA-AS315-12
Type IV method
Determination of plant proteins in wines and musts
(Resolution Oeno 24/2004)
The technique developed below enables to determine the quantity of proteins
possibly remaining in beverages treated with proteins of plant origin after racking.
1 PRINCIPLE
Wine and must proteins are precipitated with trichloroacetic acid, then
they are separated by electrophoresis in polyacrylamide gel in the presence of
dodecyl sodium sulphate (DSS). The addition of Coomassie blue colours the
proteins. The intensity of the colouration enables to determine the protein content
using a calibration curve made beforehand with the known protein concentration
solutions. The antigenic capacity of musts and treated wines is determined by
immunoblotting testing.
2 PROTOCOL
2.1 Concentration of proteins by precipitation with trichloroacetic acid (TCA)
2.1.1 Reagents
2.1.1.1 Pure trichloroacetic acid (TCA)
2.1.1.2 TCA at 0.1% prepared using 2.1.1.1: 0.1 g in
100 ml of water.
2.1.1.3 TCA at 100% prepared using 2.1.1.1: 100 g in
100 ml of water.
2.1.1.4 Sodium hydroxide 0.5 M
2.1.1.5 Buffer Tris/HCl 0.25 M pH=6.8
30.27 g of Tris-(hydroxymethyl)aminomethane (Tris) are dissolved in
300 ml of distilled water. The pH is adjusted to 6.8 with concentrated
OIV-MA-AS315-12 : R2004
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Plant proteins
hydrochloric acid for analysis. The volume is completed to 1 l with
distilled water. The buffer is stored at 4°C.
2.1.1.6 Pure glycerol
2.1.1.7 Pure dodecyl sodium sulphate (DSS)
2.1.1.8 Pure 2-mercaptoethanol
2.1.1.9 Buffer solution for samples: it is made up of a buffer
Tris/HCl 0.25 M, pH=6.8 (2.1.1.5); 7.5% of pure glycerol (2.1.1.6); 2% of
dodecyl sodium sulphate (DSS) (2.1.1.7) and 5% of pure 2mercaptoethanol (2.1.1.8). The percentages of different reagents
correspond to the final concentration in the buffer solution.
2.1.2 Procedure
3 ml of trichloroacetic acid at 100% (2.1.1.3) and 24 ml of wine or must (treated or
untreated) are successively put in 50 ml centrifuge tubes. The final concentration
in TCA thus obtained is 11%.
After 30 minutes at 4°C, the samples are centrifuged at 10,000 rpm for 30
minutes at 4°C. The pellets are washed in an aqueous solution of TCA at 0.1%
(2.1.1.2), re-centrifuged and put again in suspension in 0.24 ml mixture (1:1, v/v)
of sodium hydroxide 0.5 M (2.1.1.4) and buffer solution (2.1.1.9). The samples are
heated at 100°C in a water bath for 10 minutes.
2.2 Electrophoresis in Polyacrylamide Gel in the presence of DSS
2.2.1 Reagents
2.2.1.1 Buffer Tris/HCl 1.5 M pH=8.8
181.6 g of Tris-(hydroxymethyl)aminomethane are dissolved in 300 ml of
distilled water. The pH is adjusted at 8.8 with concentrated hydrochloric acid for
analysis. The volume is completed to 1 l with distilled water. The buffer is stored
at 4°C.
2.2.1.2 Mixture of acrylamide (30%)–bis-acrylamide (0.8%)–
glycerol (75%)
Slowly add 300 g of acrylamide and 8 g of bis-acrylamide to 600 ml of a
glycerol solution at 75%. After dissolution, adjust the volume to 1 l with glycerol
at 75%. The mixture is stored in the dark at room temperature.
OIV-MA-AS315-12 : R2004
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Plant proteins
2.2.1.3 DSS at 10%
10 g of DSS are dissolved in 100 ml of distilled water. Store at
room temperature.
2.2.1.4 N,N,N’,N’-tetramethylenediamine (TEMED) for
electrophoresis
2.2.1.5 Ammonium persulfate at 10%
1 g of ammonium persulfate is dissolved in 10 ml of distilled
water. Store at 4°C.
2.2.1.6 Bromophenol blue solution
10 mg of bromophenol blue for electrophoresis are dissolved in 10 ml of
distilled water.
2.2.1.7 Solution for the separation gel (15% of acrylamide)
It is prepared just before use:
- 1.5 ml of Tris/HCl 1.5 M, pH=8.8 (2.2.1.1),
- 1.5 ml of distilled water,
- 3 ml of glycerol acrylamide mixture (2.2.1.2),
- 50 µl of DSS 10% (2.2.1.3),
- 10 µl of N,N,N’,N’-tetramethylenediamine (TEMED) for electrophoresis
(2.2.1.4),
- 20 µl of ammonium persulfate (2.2.1.5).
- 1 drop of bromophenol blue (2.2.1.6)
2.2.1.8 Buffer Tris/HCl 0.5 M pH=6.8
60.4 g of Tris-(hydroxymethyl)aminomethane are dissolved in 400 ml of
distilled water. The pH is adjusted to 6.8 with concentrated hydrochloric acid for
analysis. The volume is completed to 1 l with distilled water. The buffer is stored
at 4°C.
2.2.1.9 Mixture of acrylamide (30%)–bis-acrylamide (0.8%)–
water
Slowly add 300 g of acrylamide and 8 g of bis-acrylamide to 300 ml of
water. After dissolution, adjust the volume to 1 l with distilled water. The mixture
is stored in the dark at room temperature.
2.2.1.10 Concentration gel at 3.5% of acrylamide
It is prepared just before use:
- 0.5 ml of Tris/HCl 0.5 M pH=6.8 (2.2.1.8),
- 1.27 ml of distilled water,
- 0.23 ml of water acrylamide mixture (2.2.1.9),
- 20 µl of DSS 10% (2.2.1.3),
- 5 µl of N,N,N’,N’-tetramethylenediamine (TEMED) for electrophoresis (2.2.1.4),
- 25 µl of ammonium persulfate (2.2.1.5),
- 1 drop of bromophenol blue (2.2.1.6).
OIV-MA-AS315-12 : R2004
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Plant proteins
2.2.1.11 Migration buffer
30.27 g of Tris-(hydroxymethyl)aminomethane, 144 g of glycine and 10 g
of DSS are dissolved in 600 ml of distilled water. The pH should be 8.8. If
necessary, it is adjusted with concentrated hydrochloric acid for analysis. The
volume is completed to 1 l with distilled water. The buffer is stored at 4°C. At the
time of use, the solution is diluted to 1/10 in distilled water.
2.2.1.12 Colouring solution
Are successively mixed:
- 16 ml of ultra-pure Coomassie brilliant blue G-250 at 5% (5 g in 100 ml of
distilled water),
- 784 ml from a 1 l solution where 100 g of ammonium sulphate and 13.8 ml of
orthophosphoric acid at 85% were dissolved for analysis,
- 200 ml of absolute ethanol.
2.2.1.13 Discolouring solution
Are successively mixed:
- 100 ml of glacial acetic acid 100% for analysis,
- 200 ml of absolute ethanol for analysis.
- 700 ml of distilled water.
2.2.2 Procedure
The separation gel solution (2.2.1.7) is poured between two glass
plates of 7x10cm. The upper surface of the gel is levelled by the addition of 2
drops of distilled water.
After polymerisation of the separation gel and the elimination of water, 1
ml of concentration gel (2.2.1.10) is deposited on the separation gel using a 1 ml
pipette. Then the comb is set up whose imprints will create deposit wells.
The samples necessary for the calibration range are prepared in a mixture
(1:1), v/v, 0.5% M sodium hydroxide (2.1.1.4) and the buffer solution (2.1.1.9) in
order for the calibration range be between 5 µg/ml and 50 µg/ml.
20 to 30 µl of wine and calibration solution are deposited in the wells.
After migration (at a constant voltage of 90 V) at room temperature for
about 3-4 hours, the gels are removed from the mould. They are immediately
plunged into 50 ml of an aqueous solution of TCA 20% for 30 minutes then in 50
ml of the colouring solution (2.2.1.12).
The proteins appear in the form of blue coloured bands. The gel is then
discoloured with 50 ml of discolouring solution (2.2.1.13). When the bottom of the
gel is transparent, it is placed in distilled water for storage.
OIV-MA-AS315-12 : R2004
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Plant proteins
3. QUANTITATIVE ANALYSIS
The intensity of each spot is evaluated by using a scanner for gel with an
image analyser software. The quantity of protein on the gel is determined by the
calculation of the average density of the pixels of the band and by integration of
the band width. The protein content of each sample is obtained using a calibration
curve. The points of this curve are obtained by tracing the known concentration
values of plant proteins deposited on the gel depending on the corresponding
integration area.
The detection and quantification limit is about 0.030 ppm for peas and at 0.36 ppm
for gluten, in an environment concentrated 100 times. The coefficient of variation
is always below 5%.
4 SEARCH BY IMMUNOBLOTTING OF THE ANTIGENIC
POTENTIAL OF WINES AND MUSTS TREATED
The antigenic capacity of proteins that could remain in the beverages
treated after racking is then evaluated.
4.1 PRINCIPLE
After electrophoresis, the gels are submitted to the immunoblotting technique. The
proteins are transferred to a membrane where they are adsorbed. An antigen–
antibody complex is formed by the addition of a plant anti-protein antibody (for
example anti–gliadin antibodies if the plant protein is gluten). The method is
revealed by the addition of an antibody directed against the plant anti-protein
antibodies coupled with phosphatase. In the presence of the chromogenic substrate
of the enzyme, a colouration whose intensity will be proportional to the quantity of
immunocomplexes will develop. This immunoreactivity will be quantified using a
calibration curve made with known concentration plant proteins solutions.
4.2 PROTOCOL
4.2.1 : Reagents
4.2.1.1 Transfer buffer
3.03 g of Tris, 14.4 g of glycine (R), 200 ml of methanol (R) are mixed
and completed to 1 l with distilled water.
4.2.1.2 Gelatine 1%
8.77 g of sodium chloride (R), 18.6 g of ethylenediaminetetraacetic acid
(EDTA) for analysis, 6.06 g of Tris and 0.5 ml of Triton X are dissolved in 800 ml
of distilled water. The pH is adjusted to 7.5 with concentrated hydrochloric acid
for analysis. 10 g of gelatine are added and the volume is completed to 1 l.
4.2.1.3 Gelatine 0.25%
8.77 g of sodium chloride (R), 18.6 g of ethylenediaminetetraacetic acid
(EDTA) for analysis, 6.06 g of Tris and 0.5 ml of Triton X are dissolved in 800 ml
OIV-MA-AS315-12 : R2004
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Plant proteins
of distilled water. The pH is adjusted to 7.5 with concentrated hydrochloric acid
for analysis. 2.5 g of gelatine are added and the volume is completed to 1 l.
4.2.1.4 Polyclonal antibody solution (marketed or
described in the annex)
- 10 µl of polyclonal plant anti-protein antibodies
- q.s.f. 10 ml with gelatine at 0.25% (4.2.1.3).
4.2.1.5 TBS buffer
29.22 g of sodium chloride for analysis and 2.42 g of tris are dissolved in 1 l of
distilled water.
4.2.1.6 Alkaline phosphatase buffer
5.84 g of sodium chloride (R), 1.02 g of magnesium chloride (R) and 12.11
g of Tris are dissolved in 800 ml of distilled water. The pH is adjusted to 9.5 with
concentrated hydrochloric acid and the volume is completed to 1 l.
4.2.1.7 Developer
15 g of bromochloroindol phosphate (BICP) and 30 g of nitro blue
tetrazolium (NBT) are dissolved in 100 ml of alkaline phosphatase buffer (4.2.1.6).
4.2.2 Procedure
After electrophoresis, the proteins are transferred from the gel to a
membrane of polyvinylidene difluoride by electrophoretic elution: 16 hours at 4°C
at 30 V in the transfer buffer (4.2.1.1). The membranes are saturated with gelatine
at 1% (4.2.1.2) and washed 3 times with gelatine at 0.25% (4.2.1.3). The gelatine
becomes set on free sites and inhibits non specific adsorption of immunological
reagents. The membrane is then plunged into 10 ml of the plant anti-protein
polyclonal antibody solution (4.2.1.4). For gluten, the anti-gliadin antibodies are
purchased. The other antibody types are prepared according to the method
provided for in the annex. The IgG-antigen complex is detected by the addition of
10 µl of anti-IgG rabbit antibodies marked with alkaline phosphatase. The
membranes are washed twice with gelatine 0.25% (4.2.1.3) and once with the TBS
buffer (4.2.1.5). After incubation in the developer (4.2.1.7), a dark purple
precipitate is formed in the spot where the enzyme is attached.
4.3 QUANTITATIVE ANALYSIS
In order to calculate the quantity of residual immunoreactivity of a
marketed wine, a calibration curve is traced out: known concentrations of plant
proteins deposited on the gel (and transferred to a membrane) depending on the
areas obtained by integration of the intensity of the spots corresponding to the
formation of immune-complex. The analysis is done with the same equipment as
for analysing electrophoresis gels.
OIV-MA-AS315-12 : R2004
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Plant proteins
ANNEX
Production of polyclonal anti-peas
Anti-peas polyclonal antibodies necessary for the determination of
antigenic capacity of pea proteins in wine and musts treated are being carried out
on animals.
1 Principle
Serums containing polyclonal antibodies are obtained from New Zealand
rabbits after an intradermal injection of antigen.
2 Protocol
2.1 Reagents
2.1.1 PBS pH=7.4 phosphate buffer: 8 g of NaCl, 200 mg of KCl, 1.73 of
Na2HPO4 H2O and 200 mg of KH2PO4 are dissolved in 300 ml of distilled water.
pH is adjusted to 7.4 with sodium hydrate 1 M. The volume is brought to 1 l with
distilled water.
2.1.2 Antigens:
10 mg of pea protein is dissolved in 5 ml of PBS phosphate buffer
(2.1.1). The solution is then filtered under sterile conditions
through 0.2 µm and stored at –20°C until the day of immunization.
2.2 Procedure
1 ml of 2.1.2. solution is mixed with 1 ml of Freund complete adjuvant. 1
ml of this mixture is injected intradermically to a New Zealand rabbit weighing
approximately 3 kg. This injection is repeated on day 15, day 30 and day 45.
60 days after the first injection, 100µl of blood were withdrawn from the
auricular vein which was then tested for its capacity to react to antigens.
Immunoblotting was used for this evaluation as described in Chapter 4.2 of the
analysis method using a gel with a pea protein which migrated on the gel.
After checking the formation of an antigen-antibody complex, 15 ml of blood were
withdrawn from the auricular vein. The blood is placed at 37°C for 30 minutes.
The serum containing the anti-pea polyclonal antibodies is withdrawn after
centrifuging the blood at 3000 rpm for 5 minutes.
OIV-MA-AS315-12 : R2004
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
Method OIV-MA-AS315-13
Type IV method
Determining the presence and content of polychlorophenols
and polychloroanisols in wines, cork stoppers, wood and
bentonites used as atmospheric traps
(Resolution 8/2006)
WITHDRAWN
(Replaced by OIV-MA-AS315-16)
OIV-MA-AS315-13 : R2009
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Lysozyme
Method OIV-MA-AS315-14
Type IV method
Measurement of lysozyme in wine
by high performance liquid chromatography
(Resolution Oeno 8/2007)
1. Introduction
It is preferable to have an analysis method available for lysozyme which is not
based on enzyme activity.
2. Scope
The method allows the quantification of lysozyme (mg of protein per l) present
in red and white wines independently of the enzyme activity (which could be
inhibited by partial denaturation or by complex formation or coprecipitation
phenomena) found in the test solution.
3. Definition
HPLC provides an analytical approach based on steric, polar or adsorptive
interactions betwen the stationary phase and the analyte, and is therefore not
linked to the actual enzyme activity exhibited by the protein.
4. Principle
The analysis is carried out using HPLC with a spectrophotometric detector
combined with a spectrofluorimetric detector. The unknown quantity in the
wine sample is calculated on the chromatographic peak areas, using the
external calibration method.
5. Reagents
5.1. Solvents and working solutions
HPLC analysis on Acetonitrile (CH3CN)
Pure trifluoroacetic acid (TFA)
deionised water for HPLC analysis
Standard solution: Tartaric acid 1g/L, Ethyl alcohol 10% v/v, adjusted to
pH 3.2 with neutral potassium tartrate.
OIV-MA-AS315-14 : R2007
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Lysozyme
5.2. Eluents
A: CH3CN 1%, TFA 0.2 %, H2O= 98.8%
B: CH3CN 70%, TFA 0.2 %, H2O= 29.8%
5.3. Reference solutions
Quantities from 1 to 250 mg/L standard lysozyme, dissolved in standard
solution by stirring continuously for at least 12 hours.
6. Equipment
6.1.
HPLC apparatus equipped with a pumping system suitable for
gradient elution
6.2.
Thermostated column compartment (oven)
6.3.
Spectrophotometer combined with spectrofluorimeter
6.4.
20 µL loop injection
6.5.
Column: polymer in reverse phase with phenyl functional groups
(diameter of pores = 1000 Å, exclusion limit = 1000000 Da) Toso
Bioscience TSK-gel Phenyl 5PW RP, 7.5 cm x 4.6 mm ID as an
example
6.6.
Pre-column in the same material as the column: Toso Bioscience
TSK-gel Phenyl 5PW RP Guardgel, 1.5 cm 3.2mm ID as an example
7. Preparation of the sample
The wine samples are acidified with HCl (10M) diluted 1/10 and filtered using
a polyamide with 0.22 µm diameter pores filter, 5 minutes after the addition.
The chromatography analysis is carried out immediately after filtering.
8. Operating conditions
8.1.
Eluent flow-rate: 1mL/min
8.2.
Temperature of column: 30°C
8.3.
Spectrophotometric detection: 280 nm
8.4.
Spectrofluorimetric detection:
λ ex = 276 nm;
λ em = 345 nm;
Gain = 10
8.5.
Gradient elution sequence
OIV-MA-AS315-14 : R2007
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Lysozyme
Time (min)
A%
B%
0
100
0
3
100
0
10
65
35
gradient
isocratic
linear
isocratic
15
65
35
27
40.5
59.5
29
0
100
linear
linear
isocratic
34
0
100
36
100
0
40
100
0
linear
isocratic
8.6
Average retention time of lysozyme: 25.50 minutes
9. Calculation
The reference solutions containing the following concentrations of lysozyme:
1; 5; 10; 50; 100; 200; 250 mg/L are analysed in triplicate. For each
chromatogram, the peak areas corresponding to the lysozyme are plotted
according to the respective concentrations, in order to obtain the linear
regresssion lines expressed by the formula Y= ax+b. The correlation
coefficient r2 must be > 0.999
10. Characteristics of the method
A validation study was carried out for the purpose of assessing the suitability
of the method for the purpose in question, taking into account linearity, limits
of detection and quantification and the accuracy of the method. The latter
parameter was determined by defining the levels of precision and trueness of
the method.
OIV-MA-AS315-14 : R2007
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Lysozyme
10.1 Linearity of the method
Based on the results obtained from the linear regression analysis, the
method proved to be linear within the ranges shown in the table below:
Linearity
range
(mg/L)
Line
gradient
Correlation
coefficient
(r2)
LD
(mg/L)
LQ
(mg/L)
Repeatability
(n=5)
RSD%
Std1 V.R.2 V.B.3
Reproducibility
(n=5)
RSD%
Std1
UV
5-250
3 786
0,9993
1,86
6,20
4,67
5,54
0,62
1,93
FLD
1-250
52 037
0,9990
0,18
0,59
2,61
2,37
0,68
2,30
Table 1: Data related to characteristics of the method:
wine ; 3 white wine
1
standard solution ;
2
red
10.2 Limit of detection and limit of quantification
The detection limit (LD) and limit of quantification (LQ) were calculated
as the signal equivalent to respectively 3 times and 10 times the
background chromatography noise under working conditions on an actual
test solution (table 1),
10.3
Precision of the method
The parameters taken into account were repeatability and
reproducibility. Table 1 shows the values of these parameters (expressed
as %age St.dv. of measurements repeated in different concentrations)
found for standard solution, red wine and white wine
10.4 Trueness of the method
The percentage recovery was calculated on the standard solutions
containing 5 and 50 mg/L of lysozyme, with known quantities of
lysozyme added, as shown in the table below.
OIV-MA-AS315-14 : R2007
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Lysozyme
Nominal
initial [C]
(mg/L)
Quantity
added
(mg/L)
Theoretical
[C] (mg/L)
[C]
found
Std.Dev.
%age
recovery
UV
280 nm
50
13.1
63.1
62.3
3.86
99
FD
50
13.1
63.1
64.5
5.36
102
UV
280 nm
5
14.4
19.4
17.9
1.49
92.1
FD
5
14.4
19.4
19.0
1.61
97.7
Fig.1 Chromatogram of red wine containing pure lysozyme (standard solution
containing 1 000 mg/L of lysozyme was added to wine to obtain a final
concentration of 125 mg/L of lysozyme). A: UV detector at 280 nm; B: UV
detector at 225 nm; C: FLD detector (λ ex 276 nm; λ em 345 nm).
OIV-MA-AS315-14 : R2007
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Lysozyme
11. Bibliography
Claudio Riponi; Nadia Natali; Fabio Chinnici. Quantitation of hen’s egg white
lysozyme in wines by an improved HPLC-FLD analytical method. Am. J. Enol.
Vit., in press.
OIV-MA-AS315-14 : R2007
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
Method OIV-MA-AS315-15
Type II method
Determination of 3-methoxypropane-1,2-diol and cyclic
diglycerols (by-products of technical glycerol) in wine by GC-MS
- description of the method and collaborative study (Resolution Oeno 11/2007)
1. Introduction
This is an internationally validated method for the determination of 3methoxypropane-1,2-diol (3-MPD) and cyclic diglycerols (CycDs) - both being
recognised as impurities of technical glycerol - in different types of wine. It is
known that glycerol produced by transesterification of plant and animal
triglycerides using methanol contains considerable amounts of 3-MPD. The
synthesis of glycerol from petrochemicals leads to impurities of CycDs. One of the
published methods [1, 2, 3i] was adopted, modified and tested in an collaborative
study. Here we present the optimized method and report the results of the
collaborative study [2]. Design and assessment of the validation study followed the
O.I.V. Resolution 8/2000 “Validation Protocol of Analytical Methods”.
2. Scope
The described method is suitable for the determination of 3-MPD and 6 cyclic
diglycerols (cis-, trans-2,6-bis(hydroxymethyl) 1,4-dioxane; cis-, trans-2,5bis(hydroxymethyl) 1,4-dioxane; cis-, trans-2,-hydroxymethyl-6-hydroxy-1,4dioxepane) in white, red, sweet and dry wines. The study described covers the
concentration range of 0.1 to 0.8 mg/L for 3-MPD and 0.5 to 1.5 mg/L for the
CycDs.
3. Definitions
3-MPD
ANOVA
C
CycDs
GC-MS
H2
IS
m/z
ML
S0
3-methoxypropane-1,2-diol
Analysis of Variance
Concentration
Cyclic diglycerols
Gas chromatography – mass spectrometry
Hydrogen
Internal standard
mass/charge ratio
Matrix calibration level
Standard dilution 1000 ng/µL
OIV-MA-AS315-15 : R2007
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
S1
Standard dilution 100 ng/µL
S2
Standard dilution 10 ng/µL
4. Principle
The analytes and the internal standard are salted-out by addition of K2CO3, and
extracted using diethyl ether. Extracts are analyzed directly by GC-MS on a polar
column. Detection is then carried out in selected ion monitoring mode.
5. Reagents and Materials
5.1. Chemicals
5.1.1 K2CO3 p.A .
5.1.2 Diethyl ether Uvasol for spectroscopy
5.1.3 Molecular sieve (2 mm diameter, pore size 0.5 nm)
5.1.4 Ethanol (Absolute)
5.2. Standards
5.2.1 Cyclic diglycerol mixture (6 components)
Solvay Alkali GmbH 1,
89.3 %
cis-, trans-2,6-bis(hydroxymethyl) 1,4-dioxane; cis-, trans-2,5bis(hydroxymethyl) 1,4-dioxane; cis-, trans-2,-hydroxymethyl-6hydroxy-1,4-dioxepane
5.2.2
5.2.3
3-Methoxypropane-1,2-diol (3-MPD) 98% (CAS 623-39-2)
Butane-1,4 -diol-1,1,2,2,3,3,4,4-(2H)8 98% (CAS 74829-49-5)
5.3. Preparation of standard solutions
5.3.1 S0 stock solutions
Accurately weigh 10.0 mg  0.05 mg of each standard substance (11.2 mg are
weighed for the CycDs, corresponding to 89.3 % purity) and transfer them to a
10 mL volumetric flask (one for each). Add exactly 10 mL of ethanol and mix
thoroughly. The concentration of this solution is 1000 ng/µL.
5.3.2 S1 working solutions
Volumetrically transfer 1000 µL of the S0 stock solution (6.3.1) to a 10 mL
volumetric flask, dilute the contents to volume with ethanol, thoroughly stopper
the flask and invert to mix. The concentration of this solution is 100 ng/µL.
1
Solvay Alkali GmbH no longer provides the standard mixture; solutions of the mixture
may be obtained from the BfR. Federal Institute for Risk Assessment, Thielallee 88-92, D14195 Berlin. www.bfr.bund.de; poststelle@bfr.bund.de
OIV-MA-AS315-15 : R2007
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
5.3.3 S2 working solutions
Volumetrically transfer 100 µL of the S0 stock solution (6.3.1) to a 10 mL
volumetric flask, dilute the content to volume with ethanol, thoroughly stopper the
flask and invert to mix. The concentration of this solution is 10 ng/µL.
Overview of required standard solutions:
CycDs mixture (6 components)
Solution
S0
S1
Concentration
1000
100
ng/µL
ng/µL
3-Methoxypropane-1,2-diol (3-MPD)
Solution
S0
S1
S2
Concentration
1000
100
10
ng/µL
ng/µL
ng/µL
1,4 Butane-1,4-(2H)8 (internal standard IS)
Solution
S0
S1
Concentration
1000
100
ng/µL
ng/µL
5.4. Preparation of the matrix calibration curve
Matrix-matched calibration solutions are prepared in an uncontaminated wine. It is
necessary to analyze this wine first to check that it is not contaminated with 3MPD or CycDs. If the concentrations of the analytes in the sample are outside the
range of the calibration curve, additional levels must be prepared. To ensure that
the internal standard does not interfere with any wine components, a blank should
be included.
OIV-MA-AS315-15 : R2007
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
Table 1. Pipetting scheme of matrix calibration
Matrix
calibration
level
Blank
ML0
ML1
ML2
ML3
ML4
ML5
Volume C Wine C Wine
Wine
IS
3-MPD
CycDs
IS
3-MPD
CycDs
IS
3-MPD
CycDs
IS
3-MPD
CycDs
IS
3-MPD
CycDs
IS
3-MPD
CycDs
IS
3-MPD
CycDs
Spike µl
100
100
100
50
100
25
100
100
50
20
100
100
30
100
200
40
ml
10
µg/L
0
mg/L
0
S1 10
1000
1.00
S1
S2
S1
S1
S1
S1
S1
S1
S0
S1
S1
S0
S1
S1
S0
1000
100
500
1000
250
1000
1000
500
2000
1000
1000
3000
1000
2000
4000
1.00
0.10
0.50
1.00
0.25
1.00
1.00
0.50
2.00
1.00
1.00
3.00
1.00
2.00
4.00
10
10
10
10
10
6. Apparatus
6.1
Analytical balance.±0.0001 g readability.
6.2
Lab centrifuge (at least 4000 rpm/min)
6.3
Gas chromatograph.-With mass spectrometric detector, split-splitless
injector,
6.4
Diverse precision pipettes and volumetric flasks
6.5
Pasteur pipettes
6.6
40 mL centrifugation vials
6.7
GC-vials (1.5 –2.0 mL)
6.8
Thermostat
6.9
Shaking machine
OIV-MA-AS315-15 : R2007
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
7. Sampling
Wine samples for the analysis should be taken in a sufficient size. Volume needed
for one test sample is 10 mL. The wine used for the preparation of the matrixcalibration (5.4) shall be free of analyte.
8. Procedure
8.1. Extraction
Add 100 µL internal standard solution S1 (6.3.2) to 10 mL wine to a suitable
centrifugation vial e.g. 40 mL. (This corresponds to a concentration of 1 mg/L
butane-1,4-(2H)8). Carefully add 10 g of K2CO3 and mix. Take care during this
addition as heat is produced due to the release of CO2. After cooling the solution
to approximately 20 °C in a water bath, add 1 mL diethyl ether. Homogenise the
mixture for 5 minutes using a vertical-shaking machine. Centrifuge the vials at
4000 rpm for 5 min. For better removal of the organic phase, the extract can be
partially transferred into a vial with a smaller diameter. Using a Pasteur pipette,
transfer the upper organic phase, composed of diethyl ether and ethanol, into a GC
vial. Add approximately 120 mg of molecular sieve into the vial. Close the vial,
leave for at least 2 h and shake well from time to time. The clear supernatant is
transferred to a second GC vial for the GC-MS analysis.
8.2. GC-MS Analysis
Specific parameters for the GC-MS analysis are provided below. Alternative
systems may be used, if they provide a similar chromatographic performance and
adequate sensitivity. The chromatographic system must be able to separate the
internal standard from phenylethanol, a potential interference.
Typical GC conditions
Gas chromatograph: HP 5890 or equivalent
DB-Wax (J&W) column 60 m, 0.32 mm internal diameter, 0.25 µm film thickness,
2 m capillary containment same dimensions or equivalent
Carrier gas: H2
Flow: Pressure 60 k Pa column head
Temperature program:
90° C, 2 min., ramp at 10°C/min. up until 165° C, held for 6 min., ramp at 4°
C/min to 250°C, held for 5 min.
Injection temperature: 250° C; Injected volume; 2 µL, 90 sec splitless for 90 s.
OIV-MA-AS315-15 : R2007
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
Specificl MS conditions
Mass spectrometer: Finnigan SSQ 710 or equivalent
Transfer line: 280° C
Source: 150° C
MS detection:
window1.:
0-25 min.:
14.3 min.
3-MPD:
m/z 75, m/z 61
16.7 min
IS:
m/z 78, m/z 61
Acquisition time for each mass is 250 µs (dwell time).
Monitor for m/z 91 the separation of the internal standard (IS) peak from
phenylethanol, which also produces a fragment m/z 78.
window 2.
25-40 min.:
32-34.5 min. CycDs:
m/z 57, m/z 117
Acquisition time for each mass is 250 µs (dwell time).
It has been observed that the analysis may degrade chromatographiccolumn. In
particular, the injection of the high boiling CycDs mixture is suspected to cause
irreversible damage. Injections of reference standard solutions should be avoided;
analysis should be restricted to salted-out solutions with low analyte
concentrations. In addition it is recommended to use a 1-2 m pre column in order
to protect the analytical column. Nevertheless, the analytical column has to be
considered as a consumable and must be replaced quite regularly.
9. Evaluation
9.1. Identification
Record the relative retention time of each analyte to the IS. Calculate the mean
relative retention time of the analytes in the calibration standards. The relative
retention time of the analyte should be the same as that of the standard within a
margin of ± 0.5 %. As a confirmation criterion, an ion ratio can be calculated for
each analyte from the selected ion monitoring. This ratio is 117/57 for CycDs,
75/61 for 3-MPD and 78/61 for the IS. The ratio should be within  20 % of that
which is found in the spiked sample. Confirmation of the identity of substances by
full scan using ionsn can also be used.
9.2. Quantification
The quantification is done by a matrix calibration curve prepared according to
appropriate section. The analyte/IS area ratios of the indicated mass ratios are
correlated by linear regression against the concentration of the analyte.
Quantification of the CycDs is achieved by summing the peak area of all six peaks
OIV-MA-AS315-15 : R2007
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
and calculating the total content, to allow for other distributions of the six
characteristic CycDs than in the standard. The following m/z values are used for
quantification:
3-MPD:
IS:
CycDs:
m/z 75
m/z 78
m/z 117
9.3. Expression of results
Results should be expressed in mg/L for 3-MPD and CycDs with two decimals
(e.g. 0.85 mg/L).
9.4 Limit of Detection and limit of quantification
The limit of detection (LD) and the limit of quantification (LQ) depend on the
individual measurement conditions of the chemical analysis and are to be
determined by the user of the method.
The limit of detection (LD) and the limit of quantification (LQ) were estimated
using the instrumentation and conditions mentioned exemplarily above (s. 8)
following the instructions in the resolution OENO 7-2000 (E-AS1-10-LIMDET)
“Estimation of the Detection and Quantification Limits of a Method of Analysis“.
Along the line of the „Logic Diagram for Decision-Making“ in N° 3 the graph
approach has to be applied following paragraph 4.2.2. For this purpose a part of
the ion trace (m/z) chromatogram is drawn extendedly enclosing a range of a
tenfold peak width at mid-height (w½) of an analyte peak in a relevant part of the
chromatogram. Furthermore two parallel lines are drawn which just enclose the
maximum amplitude of the signal window.
The distance of these two lines gives hmax, expressed in abundance units is
multiplied by 3 for LD, by 10 for LQ and finally converted into concentration units
by implementing the individual response factor.
3-MPD:
LD: 0,02 mg/l
LQ: 0,06 mg/l
CycDs (sum):
LD: 0,08 mg/l
LQ: 0,25 mg/l
(Note: Since the CD are a mixture of six single compounds with the same response
factor - due to their chemical equality - and with hmax constant in the relevant part
of the chromatogram the LD and LQ for each single compound are one sixth of the
figures above)
OIV-MA-AS315-15 : R2007
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
10. Precision (interlaboratory validation)
Eleven laboratories participated in the collaborative study. The participating
laboratories have proven experience in the analysis of the by-products. All of them
participated in the pre-trial.
Repeatability (r) and reproducibility (R) and the respective standard deviations (Sr
and SR) were found to be correlated statistically significantly with the
concentration of the analytes (ANNEX: Figures 1 and 2), r with more than 95%
probability and R with more than 99% probability for each of the analytes using
the linear regression model.
The actual performance parameters can be calculated by:
3-MPD
Sr = 0,060 x
SR = 0,257 x
x = concentration of 3-MPD [mg/L]
r = 0,169 x
R = 0,720 x
x = concentration of 3-MPD [mg/L]
CycDs
Sr = 0,082 x
SR = 0,092 x + 0,070
x = concentration of CycDs [mg/L]
r = 0,230 x
R = 0,257 x + 0,197
x = concentration of CycDs [mg/L]
OIV-MA-AS315-15 : R2007
8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
ANNEX (Interlaboratory Study)
Participants
11 international laboratories participated in the collaborative study (5). The
participating laboratories have proven experience in the analysis of the byproducts. All of them participated in the pre-trial:
CSL, York, UK
Unione Italiana Vini, Verona, Italy
BfR, Berlin, Germany
BLGL, Würzburg, Germany
Istituto Sperimentale per l'enologia, Asti, Italy
LUA, Speyer, Germany
Labor Dr. Haase-Aschoff, Bad Kreuznach, Germany
CLUA, Münster, Germany
Kantonales Laboratorium, Füllinsdorf, Switzerland
LUA, Koblenz, Germany
ISMAA, S. Michele all Adige, Italy
Samples
In November 2002, participating laboratories were sent 11 wine samples consisting
of five sets of blind duplicates and one further single test material. Dry white
wines, dry red wines and a sweet red wine were used for test materials. The
samples were subjected to homogeneity testing previously (ii).
Data analysis
Statistical analysis was carried out according to the “Protocol for the Design,
Conduct and Interpretation of Method Performance Studies” (iii) using a blind
duplicate model.
1. Determination of outliers was assessed by Cochran, Grubbs and paired Grubbs
tests.
2. Statistical analysis was performed to obtain repeatability and reproducibility
data.
3. Horrat values were calculated.
OIV-MA-AS315-15 : R2007
9
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
Table 2. Results for 3-MPD
Mean mg/L
Spiked mg/L
Recovery %
n
nc
outliers
n1
r
sr
RSDr %
Hor
R
sR
RSDr %
HoR
Sample A
White
wine
0.30
0.30
100
10
1
2
7
0.03
0.01
3.20
0.30
0.13
0.05
15.50
0.80
Sample B
Red
wine a
0.145
0.12
121
10 a
1a
0
9a
0.13
0.05
32.67
1.53
Sample C
White
wine
0.25
10
1
0
9
0.05
0.02
7.20
0.60
0.15
0.05
21.20
1.10
Sample F
Sweet red
wine
0.48
10
1
1
8
0.08
0.03
5.80
0.50
0.31
0.11
22.70
1.30
Sample G
White
wine
0.73
0.80
91
10
1
1
8
0.13
0.05
6.57
0.59
0.59
0.21
28.91
1.72
a
Single test sample; n, nc and n1 are single results
mean
arithmetic mean of the data used in the statistical analysis
total number of sets of data submitted
n
nc
number of results (laboratories) excluded due to non-compliance
outliers
number of results (laboratories) excluded due to determination as
outliers by either Cochran’s or Grubbs’ tests
n1
number of results (laboratories) retained in statistical analysis
Sr
the standard deviation of the repeatability
RSDr
the relative standard deviation of the repeatability (Srx100/mean)
r
repeatability (2.8 x Sr)
Hor
the Horrat value for repeatability is the observed RSDr divided by the
RSDr value estimated from the Horwitz equation using the assumption
r = 0.66R
R
reproducibility (between laboratory variation) (2.8 x SR)
SR
the standard deviation of the reproducibility
RSDR
the relative standard deviation of the reproducibility (SRx100/mean)
HoR
the Horrat value for reproducibility is the observed RSDR value
divided by the RSDR value calculated from the Horwitz equation
OIV-MA-AS315-15 : R2007
10
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
0,7
0,6
R = 0,720 x
r and R [ mg/L]
0,5
0,4
R
0,3
r
0,2
0,1
r = 0,169 x
0
0
0,2
0,4
0,6
0,8
X = Concentration of 3-MPD [mg/L]
Figure 1. Correlation between 3-MPD concentration and r and R.
OIV-MA-AS315-15 : R2007
11
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
Table 3. Results for cyclic dyglycerols
Mean mg/L
Spiked mg/L
Recovery %
n
nc
outliers
n1
r
sr
RSDr %
Hor
R
sR
RSDR %
HoR
a
Sample A
White
wine
1.55
1.50
103
11
0
2
9
0.37
0.13
8.50
0.90
0.61
0.22
14.00
0.90
Sample B
Red
winea
0.593
0.53
113
11a
0
0
11a
0.379
0.135
22.827
1.319
Sample D
Red
wine
0.80
Sample F
Sweet red
wine
0.96
11
0
1
10
0.19
0.07
8.60
0.80
0.39
0.13
17.30
1.00
11
0
2
9
0.18
0.07
6.70
0.60
0.41
0.15
15.20
0.90
Sample G
White
wine
0.56
0.50
112
11
0
1
10
0.15
0.05
9.30
0.80
0.34
0.12
21.50
1.20
Single test sample; n and nc are single results
OIV-MA-AS315-15 : R2007
12
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
3-Methoxypropane-1,2-diol and Cyclic Diglycerols
0,7
R = 0,257 x + 0,197
0,6
r and R in [mg/L]
0,5
0,4
R
0,3
r
0,2
r = 0,230 x
0,1
0
0
0,5
1
1,5
x =Concentration of CycDs [mg/L]
Figure 2. Correlation between CycDs concentration and r and R.
(1) Bononi, M., Favale, C., Lubian, E., Tateo F. (2001)
A new method for the identification of cyclic diglycerols in wine
J. Int. Sci. Vigne Vin. 35, 225-229
(2) Thompson, M. and Wood, R. (1993)
International Harmonised Protocol for the Proficiency Testing of
(Chemical) Analytical Laboratories - J AOAC Int 76, 926-940
(3) Horwitz ,W. (1995)
Protocol for the design, conduct and interpretation of methodperformance studies
Pure and Applied Chemistry 67, 331-343
OIV-MA-AS315-15 : R2007
13
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
2,4,6-trichloroanisole
Method OIV-MA-AS315-16
Type IV method
Determination of releasable 2,4,6-trichloroanisole
in wine by cork stoppers
(Resolution OIV-Oeno 296/2009)
1 SCOPE:
The method of determination of releasable 2,4,6-trichloroanisole (TCA) by cork
stoppers measures the quantity of TCA released by a sample of cork stoppers
macerated in a aqueous-alcoholic solution. The aim of this method is to evaluate
the risk of releasing by the lot of analyzed cork stoppers and to provide a method
for controlling the quality of cork stoppers.
2 PRINCIPLE
The method aims to simulate 2,4,6-trichloroanisole migration phenomena
susceptible of being produced between the cork stopper and wine in bottles. Cork
stoppers are macerated in a wine or a aqueous-alcoholic solution, until a balance is
obtained. The TCA of the head space is sampled from an appropriate part of the
macerate by the solid-phase micro-extraction technique (SPME), then analyzed by
gas chromatography, with detection by mass spectrometer (or by electron-capture
detector).
3 REAGENTS AND PRODUCTS
3.1 White wine with an alcoholic strength ranging between 10 and 12 %
vol. (It can be replaced by an aqueous-alcoholic solution with an alcoholic strength
of 12 % vol). The wine and/or the aqueous-alcoholic solution must be free of
TCA.
3.2 Sodium chloride ≥ 99.5 %
3.3 2,4,6-trichloroanisole (TCA)-d5 purity ≥ 98% for GC/MS; 2,6dibromoanisole or 2,3,6-trichloroanisole purity ≥ 99 % for GC/ECD
3.4 2,4,6-trichloroanisole (TCA) purity ≥ 99.0%
3.5 Absolute ethanol
3.6 Pure de-ionised water void of TCA (Standard EN ISO 3696)
3.7 Aqueous-alcoholic solution at 12 % vol.
Prepared using absolute ethanol (3.5) and de-ionised water void of TCA
(3.6).
3.8 Internal standard stock solution (500 mg/L)
OIV-MA-AS315-16 : R2009
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
2,4,6-trichloroanisole
Add either 0.050 g of 2,4,6-trichloroanisole-d5 (or 2,6-dibromoanisole or 2,3,6trichloroanisole (3.3) to approximately 60 ml of absolute ethanol (3.5). After
dissolution, adjust the volume to 100 mL with absolute ethanol (3.5). It can
be kept in a glass bottle with a metallic or glasscover.
3.9 Intermediate solution of internal standard (5.0 mg/L)
Add 1 mL of a solution of either 2,4,6-trichloroanisole-d5 (or 2,6dibromoanisole or 2,3,6-trichloroanisole) at 500 mg/L (3.8) to
approximately 60 mL of absolute ethanol (3.5). Adjust the volume to 100
mL with absolute ethanol (3.5). It can be kept in a glass bottle with a
metallic or glass cover.
3.10 Internal standard solution (2.0 µg/L)
Add 40 µL of a solution of either 2,4,6-trichloroanisole-d5 (or 2,6dibromoanisole or 2,3,6 trichloroanisole) at 5.0 mg/L (3.9) to approximately
60 mL of absolute ethanol (3.5). Adjust the volume to 100 ml with absolute
ethanol (3.5). It can be kept at an ambient temperature in a glass bottle with
a metallic or glass cover.
3.11 Stock solution of TCA standard (40 mg/L)
Add 0.020g of 2,4,6-trichloroanisole to approximately 400 ml of absolute
ethanol (3.5). Following dissolution, adjust volume to 500 mL with absolute
ethanol (3.5).
3.12 Intermediate solution A of TCA standard (80 μg/L)
Add 1 mL of 2,4,6-trichloroanisole solution at 40 mg/L (3.11) to
approximately 400 mL of absolute ethanol (3.5). Following dissolution, adjust
volume to 500 mL with absolute ethanol (3.5).
3.13 Intermediate solution B of TCA standard (160 ng/L)
Add 1 mL of solution 2,4,6-trichloroanisole at 80 μg/L (3.12) to
approximately 400 mL of pure de-ionised water (3.6). Following dissolution,
adjust the volume to 500 mL with pure de-ionised water (3.6)
3.14 Use the standard-addition technique to make up a range of standard
solutions of TCA. Standard solutions in the range from 0.5 ng/L to 50 ng/L can be
used, by making additions with a solution of 2,4,6-trichloroanisole at 160 ng/L
(3.13) to 6 ml of absolute ethanol (3.5). Following dissolution, adjust volume to 50
mL with pure de-ionised water (3.6)
The calibration curve obtained should be evaluated regularly and in any case
whenever there is a major change in the GC/MS or GC/ECD systems.
3.15 Carrier gas: Helium, chromatographic purity ( 99.9990 %)
OIV-MA-AS315-16 : R2009
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
2,4,6-trichloroanisole
4. APPARATUS
4.1 Laboratory glassware
4.1.1 Graduated 100-mL flask
4.1.2 100-µL microsyringe
4.1.3 Wide-neck glass jar of a capacity adapted to the sample size, closed
with a glass or metallic stopper or a material which does not bind TCA.
4.1.4 20-mL glass sample bottle closed with a perforated capsule and a
liner with one side Teflon-coated.
4.2 Solid-phase microextraction system (SPME) with a fiber coated with a
polydimethylsiloxane film 100 µm thick
4.3 Heating system for sample bottle (4.1.4)
4.4 Stirring system for sample bottle (4.1.4)
4.5 Gas chromatograph equipped with a "split-splitless" injector and a
mass spectrometer detector (MS) or an electron-capture detector (ECD)
4.6 Data-acquisition system
4.7 If required, an automatic sampling and injection system operating with
an SPME system
4.8 Capillary column coated with an apolar stationary phase, of the
phenylmethylpolysiloxane type (e.g.: 5 % phenyl methylpolysiloxane, 30 m x 0,25
mm x 0,25 µm film thickness or equivalent.)
5. SAMPLE PREPARATION
The corks are placed whole in a glass closed container. The container capacity
(4.1.3), the same as the quantity of wine or aqueous-alcoholic solution (3.1 or 3.7),
must be chosen in accordance to the sample size while ensuring that the corks are
completely covered and immersed in the maceration container.
Example 1: 20 corks (45x24) mm, in a 1 L container;
Example 2: 50 corks (45x24) mm, in a 2 L container.
Most of the TCA released during maceration of the groups of stoppers is
generally derived from a very low percentage of these stoppers. In order to obtain
the best representation of a batch of stoppers, a number of appropriate analyses
according to sampling rules and risk with regard to wine contamination should be
carried out.
6. OPERATING METHOD
6.1 Extraction
OIV-MA-AS315-16 : R2009
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
2,4,6-trichloroanisole
After macerating at ambient temperature for (24  2) hours under
laboratory ambient temperature conditions, the maceration is homogenized by
inversion. A part of the aliquot of the 10ml maceration solution (5) is transferred
to a glass sample bottle (4.1.4)
To increase extraction efficiency and subsequent sensitivity of the method,
a quantity of approximately 1 g of sodium chloride (3.2) can be added. 50 µL of
the internal standard solution at 2.0 µg/L (3.10) are immediately added, then the
bottle is closed using a perforated metal capsule fitted with a silicone / Tefloncoated liner. The capsule is crimped. The contents of the bottle are homogenized
for 10 minutes by mixing using a stirring system (4.4) or by using an automatic
system (4.7).
The bottle containing the sample is placed in the heating system (4.3) set
to 35 °C ± 2 °C, with stirring (4.4). The extraction of the headspace is carried out
using the SPME system (4.2) for at least 15 minutes.
6.2 Analysis
The fiber is then desorbed at 260 °C for at least 2 minutes in the injector of
a gas chromatograph, in splitless mode (4.5). The separation is carried out using a
capillary column with a non-polar stationary phase (4.8). The carrier gas is helium
with a constant flow of 1 ml/min. A temperature program from 35 °C (for 3 min)
to 265 °C (at 15 °C/min) is given as an example.
6.3 Detection and quantification
Detection is carried out by mass spectrometry with a selection of specific
ions for the 2,4,6-trichloroanisole (ions m/z 195, 210, 212), quantified on
the m/z 195 ion, and the internal standard 2,4,6-trichloroanisol-d5 (ions m/z
199, 215, 217) quantified on ion m/z 215..
For the determination of ECD, identify the analyte and internal standard (2,6dibromoanisole or 2,3,6 trichloroanisole) in the chromatogram, by comparing the
retention time of the sample peak corresponding to that of the standard solution
peak.
7. CALCULATIONS
The area of the chromatographic peak obtained for the 2,4,6trichloroanisole is corrected by the area obtained for the chromatographic peak of
the internal standard. The content in 2,4,6-trichloroanisole of each sample is
obtained using a calibration curve. The points on this curve are obtained by tracing
the relative responses of the 2,4,6-trichloroanisole/internal standard, obtained for
OIV-MA-AS315-16 : R2009
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
2,4,6-trichloroanisole
aqueous-alcoholic solutions (3.7) containing known concentrations of 2,4,6trichloroanisole, as a function of the concentrations of these solutions (3.14).
The results are given in ng/L of TCA present in the maceration, rounded off
to the nearest 0.1 ng/L.
8. CHARACTERISTICS OF THE METHOD
As an indication, the detection limit of the analysis of the macerations must be
lower than 0.5 ng/L, and the quantification limit close to 1 ng/L. The coefficient of
variation is lower than 5% for 5 ng/L, when the selected internal standard is the
deuterated analogue TCA-d5.
An interlaboratory trial was carried out in order to validate the method. This
interlaboratory trial was not carried out according to the OIV protocol and the
validation parameters mentioned in the FV 1224.
9. BIBLIOGRAPHY
HERVÉ E., PRICE S., BURNS G., Chemical analysis of TCA as a quality control
tool for natural corks. ASEV Annual Meeting. 1999.
ISO standard 20752:2007 Cork stoppers — Determination of releasable 2, 4, 6trichloroanisol (TCA).
FV 1224 - Résultats de l’analyse collaborative Ring test 3-TCA SPME.
OIV-MA-AS315-16 : R2009
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
Method OIV-MA-AS315-17
Type IV method
Determining the presence and content of polychlorophenols
and polychloroanisols in wines, cork stoppers, wood and
bentonites used as atmospheric traps
(Resolution OIV-Oeno 374/2009)
1. SCOPE
All wines, cork stoppers, bentonites (absorption traps) and wood.
2. PRINCIPLE
Determination
of
2,4,6-trichloroanisol,
2,4,6-trichlorophenol,
2,3,4,6tetrachloroanisol,
2,3,4,6-tetrachlorophenol,
pentachloroanisol
and
pentachlorophenol by gas chromatography, by injecting a hexane extract of the
wine and an ether/hexane extract of the solid samples to be analyzed and internal
calibration.
3. REAGENTS
Preliminary remark: all the reagents and solvents must be free of the compounds
to be determined listed in 2 at the detection limit.
3.1 Purity of hexane > 99 %
3.2 Purity of ethylic ether > 99 %
3.3 Ether/hexane mixture (50/50; v/v)
3.4 or 2,5-dibromophenol purity ≥ 99 %
3.5 Pure ethanol
3.6 Pure deionized water, TCA free, type II in accordance with ISO standard EN
3696
3.7 50 % vol. aqueous-alcoholic solution. Place 100 ml of absolute ethanol (3.<5)
in a graduated 200-ml flask (4.9.9), add 200 ml of deionized water (3.6), and
homogenize.
3.8 Internal standard:
3.8.1 200 mg/l stock solution. Place 20 mg of internal standard (3.4) in a
graduated 100-ml flask (4.9.8), add the 50 % volume aqueous-alcoholic solution
(3.7) and homogenize.
OIV-MA-AS315-17 : R2009
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
3.8.2 Internal standard solution (2 mg/l). Place 1 ml of the stock solution
of internal standard (3.8.1) in a graduated 100-ml flask (4.9.8), add the 50% vol
aqueous-alcoholic solution (3.7) and homogenize.
3.8.3 Internal standard solution (20 µg/l). Place 1 ml of stock solution of internal
standard (3.8.2) in a 100 ml graduated flask (4.9.8), add with 50 % vol aqueousalcoholic solution
3.9 Pure products
3.9.1 2,4,6-trichloroanisole: ≥ 99 %, case: 87-40-1
3.9.2 2, 4, 6-trichlorophenol: ≥ 99.8 %, case: 88-06-2
3.9.3 2,3,5,6-tetrachloroanisole: ≥ 99 %, case: 6936-40-9 (note: the
product sought in the samples is 2,3,4,6-tetrachloroanisole but is
does not exist on the market)
3.9.4 2, 3, 4, 6-tetrachlorophenol: ≥ 99 %, case: 58-90-2
3.9.5 pentachloroanisole: ≥ 99 %, case: 1825-21-1
3.9.6 pentachlorophenol: 99 %, case: 87-86-5
3.10 Reagents for derivatisation - Piridine: acetic anydride (1:0,4) vol.
3.10.1 Piridine: ≥ 99 %
3.10.2 Acetic anydride: ≥ 98 %
3.11 Calibration stock solution at 200 mg/l
In a graduated 100-ml flask (4.9.8), place approximately 20 mg of the pure
reference products (3.9.1 to 3.9.6) but whose exactly weight is known (4.7), add
absolute ethanol (3.5). Homogenize.
3.12 Intermediate calibration solution at 200 µg/l
In a graduated 100-ml flask (4.9.8) filled with absolute ethanol (3.5), add 100 µl of
the calibration stock solution at 200 mg/l (3.11) using the 100-µl micro-syringe
(4.9.1) and homogenize.
3.13 Calibration surrogate solution at 4 µg/l
In a graduated 50-ml flask (4.9.7) containing 50 % vol aqueous-alcoholic solution
(3.7) add 1 ml of the intermediate calibration solution at 200 µg/l (3.11) using a 1ml pipette (4.9.6). Add to volume 50 ml with pure ethanol (3.5) and homogenize.
3.14 Calibration solutions. It is possible to prepare various standard solutions with
various concentrations by adding, using the 100-µl micro-syringe of (4.9.1), for
example 50 µl of the surrogate calibration solution at 4 µg/l (3.12) to 50 ml of
wine to enrich it with 4 ng/l of the substances to be determined.
OIV-MA-AS315-17 : R2009
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
The same reasoning can be used to prepare calibration solutions of various
concentrations, either using aqueous-alcoholic solutions, or wine, or to enrich an
extraction medium with a known quantity of pure products.
3.15 Commercially available Bentonite.
4. APPARATUS
4.1 Gas phase chromatograph with Split-splitless injector coupled to an electron
capture detector. (It is likewise possible to use a mass spectrometer)
4.2 Capillary tube of non-polar steady-state phénylmethylpolysiloxane type: (0.32
mm x 50 m, thickness of film 0.12 µm or the equivalent
4.3 Chromatographic conditions, as an example:
4.3.1 Injection in "split-splitless" mode (valve closing time 30 seconds)
4.3.2 Carrier gas flow rate: 30 ml/min including 1 ml in the column
Hydrogen U ®2 (It is likewise possible to use helium)
4.3.3 Auxiliary gas flow rate: 60 ml/min – Nitrogen with
chromatographic purity (≥ 99,9990 %). It is also possible to use argon
methane.
4.3.4 Furnace gradient temperature for information purposes:
- from 40 °C to 160 °C at a rate of 2 °C/min
- from 160 °C to 200 °C at a rate of 5 °C/min
- step at 220 °C for 10 min
4.3.5 Injector temperature: 250 °C
43.6 Detector temperature: 250 °C
4.4 Acquisition and integration: acquisition is by computer. The peaks of the
various compounds identified by comparison with the reference are then
integrated.
4.5 Magnetic agitator.
4.6 Vortex with adaptation for 30-ml flask (4.9.3)
4.7 Precision balance to within 0.1 mg
4.8 Manual or electric household grate
4.9 Laboratory equipment:
4.9.1 100-µl micro-syringe
4.9.2 10-µl micro-syringe
4.9.3 30-ml flask closing with a screwed plug and cover with one side
Teflon-coated
4.9.4 10-ml stick pipette graduated 1/10 ml
4.9.5 5-ml stick pipette graduated 1/10 ml
4.9.6 1-ml precision pipette
4.9.7 Graduated 50-ml flask
OIV-MA-AS315-17 : R2009
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
4.9.8 Graduated 100-ml flask
4.9.9 Graduated 200-ml flask
4.9.10 100-ml separating funnel
4.9.11 Pasteur pipettes and suitable propipette pear
4.9.12 Household aluminum foil, roll-form.
4.9.13 Centrifuge
5. SAMPLE PREPARATION
5.1 The stopper is grated (4.8) or cut into pieces (dimension < 3 mm)
5.2 Wood is cut with a clipper to obtain pieces (dimension < 3 mm)
5.3 The bentonite (3.15) (30 g for example) is spread out over a strip of aluminum
foil (4.9.12) of approximately 30 cm x 20 cm and is exposed to the atmosphere to
be analyzed for at least 5 days.
6. OPERATING METHOD
6.1 Extraction process for solid samples:
6.1.1 Stopper: in a 30-ml flask (4.9.3), place approximately 1 g of grated
stopper (5.1) but of a precisely known weight (4.7)
6.1.2 Wood: in a 30-ml flask (4.9.3), place approximately 2 g of wood
chips (5.2) but of a precisely known weight (4.7)
6.1.3 Control Bentonite: in a 30-ml flask (4.9.3), place approximately 5 g
of bentonite (3.15) but of a precisely known weight (4.7)
6.1.4 Sample bentonite: in a 30-ml flask (4.9.3), place approximately 5 g
of bentonite (5.3) of a precisely known weight (4.7)
6.1.5 Add 10 ml (4.9.4) of ether/hexane mixture (3.3)
6.1.7 Add with the micro-syringe (4.9.1) 50 µl of the internal standard
solution (3.8.2)
6.1.8 Agitate with the vortex (4.6) for 3 min
6.1.9 Recover the ether/hexane liquid phase in a 30-ml flask (4.9.3)
6.1.10 Repeat the extraction operation on the sample with 2 times 5 ml of
ether/hexane mixture (3.3)
6.1.11 Final extract: mix the 3 phases of ether/hexane.
6.2 Extraction of the wine and calibration solution
6.2.1 Sample 50 ml of wine or calibration solution (using the graduated
flask (4.9.7)
6.2.2 Place them in the 100-ml graduated flask (4.9.8)
6.2.3 Add with the microsyringe (4.9.1) 50 µl of internal standard (3.8.3)
6.2.4 Add 4 ml (4.9.5) of hexane (3.1)
OIV-MA-AS315-17 : R2009
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
6.2.5 Carry out the extraction using the magnetic stirrer (4.5) for 5 min.
6.2.6 Elutriate into the funnel (4.9.10)
6.2.7 Recover the organic phase with the emulsion in a 30-ml flask
(4.9.3) and aqueous phase in the 100-ml graduated flask (4.9.8)
6.2.8 Repeat the extraction of the wine or calibration solution using 2 ml
of hexane (3.1)
6.2.9 Carry out the extraction using the magnetic stirrer (4.5) for 5 min.
6.2.10 Elutriate into the funnel (4.9.10)
6.2.11 Recover the organic phase with the emulsion in the same 30-ml
flask mentioned
in 6.2.7 (containing the organic phase obtained
upon the first extraction)
6.2.12 Break the emulsion of the organic phase by centrifugation (4.9.13)
by eliminating the lower aqueous phase using a Pasteur pipette (4.9.11)
fitted with a propipette pear.
6.2.13 Final wine extract and calibration solutions: the residual organic
extract
6.3 Analyze:
6.3.1 Add final extract (6.1.11 or 6.2.13) 100 μl (4.9.1) of the pyridine acetic
anydride reagent mixture (3.10) for the derivatisation.
6.3.2 Mix using a magnetic stirrer (4.5) for 10 min.
6.3.3 Inject 2 μl of derivatised final extract (6.3.2) into the chromatograph
7. CALCULATION:
Product peak area
Concentration of product = —------————————— * Response factor
Peak area of internal standard
Response factor = concentration of calibration solution (3.13) * (Peak area of the
internal standard / *(Peak area of the pure product in the calibration solution).
Check the calibration by ensuring the response factors +/- 10 %.
8. RESULTS
The results are expressed in ng/l for the wine and ng/g for the cork stoppers,
bentonites and wood.
OIV-MA-AS315-17 : R2009
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
9. CHARACERISTICS OF THE METHOD
9.1 Coverage rate
The coverage rate calculated in relation to the quantities added in terms
of wood chips, polychloroanisols and polychlorophenols of 115 ng/g is:
- 2,4,6-trichloroanisol: 96 %
- 2,4,6-trichlorophenol: 96 %
- 2,3,4,6-tetrachloroanisol: 96 %
- 2,3,4,6-tetrachlorophenol: 97 %
- pentachloroanisol: 96 %
- pentachlorophenol: 97 %
9.2 Measurement repeatability
Calculated for each product, the uncertainties are as follows:
In a stopper ng/g
Mean
2,4,6-trichloroanisol
2,4,6-trichlorophenol
2,3,4,6-tetrachloroanisol
2,3,4,6-tetrachlorophenol
pentachloroanisol
pentachlorophenol
1.2
26
1.77
2.59
23.3
7.39
Standard
deviation
0.1
3.3
0.44
0.33
2.9
1.91
Repeatability
0.28
9.24
1.23
0.92
8.12
5.35
In wood with 23 ng/g
2,4,6-trichloroanisol
2,4,6-trichlorophenol
2,3,4,6-tetrachloroanisol
2,3,4,6-tetrachlorophenol
pentachloroanisol
pentachlorophenol
Standard deviation
1.9
1.9
2.6
3.3
2.7
3.6
Repeatability
5.3
5.3
7.4
9.3
7.5
10.1
In wine with 10 ng/l
Standard deviation
Repeatability
2,4,6-trichloroanisol
2,4,6-trichlorophenol
2,3,4,6-tetrachloroanisol
2,3,4,6-tetrachlorophenol
pentachloroanisol
pentachlorophenol
0,4
2,1
0,6
4
1,2
6,5
OIV-MA-AS315-17 : R2009
1,1
5,9
1,7
11,2
3,4
18,2
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
In bentonite with15ng/g
2,4,6-trichloroanisol
2,4,6-trichlorophenol
2,3,4,6-tetrachloroanisol
2,3,4,6-tetrachlorophenol
pentachloroanisol
pentachlorophenol
Standard deviation
Repeatability
0,9
4
1,2
5,2
4,3
12,1
2,5
11,2
3,4
14,6
12,0
33,9
9.3 Detection limits (DL) and quantification limits (QL) calculated according to
the OIV method:
9.3.1 Wood
2,4,6-trichloroanisol
2,4,6-trichlorophenol
2,3,4,6-tetrachloroanisol
2,3,4,6-tetrachlorophenol
pentachloroanisol
pentachlorophenol
DL in ng/g
0.72
0.62
0.59
1.12
0.41
0.91
QL in ng/g
2.4
2.0
2.0
3.74
1.4
3.1
DL in ng/g
0.5
1
0.5
1
0.5
Not det.
QL in ng/g
1
3
1
3
1
Not det.
9.3.2 Bentonite
2,4,6-trichloroanisol
2,4,6-trichlorophenol
2,3,4,6-tetrachloroanisol
2,3,4,6-tetrachlorophenol
pentachloroanisol
pentachlorophenol
9.3.3 Stopper
2,4,6-trichloroanisol
2,4,6-trichlorophenol
2,3,4,6-tetrachloroanisol
2,3,4,6-tetrachlorophenol
pentachloroanisol
pentachlorophenol
OIV-MA-AS315-17 : R2009
DL in ng/g
0.5
1
0.5
1
0.5
1
QL in ng/g
1.5
2
1.5
2
1.5
2
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Polychlorophenols, Polychloroanisols
9.3.4 Wine
2,4,6-trichloroanisol
2,4,6-trichlorophenol
2,3,4,6-tetrachloroanisol
2,3,4,6-tetrachlorophenol
pentachloroanisol
pentachlorophenol
DL in ng/l
0.3
1
0.3
0.3
0.5
1
QL in ng/l
1
3
1
1
3
3
®2 Air Liquide
OIV-MA-AS315-17 : R2009
8
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
Method OIV-MA-AS315-18
Type II method
Analysis of biogenic amines in musts and wines using HPLC
(Resolution OIV-Oeno 346/2009)
1.
SCOPE
This method can be applied for analysing biogenic amines in musts and wines:
Ethanolamine: up to 20 mg/l
Histamine: up to 15 mg/l
Methylamine: up to 10 mg/l
Serotonin: up to 20 mg/l
Ethylamine: up to 20 mg/l
Tyramine: up to 20 mg/l
Isopropylamine: up to 20 mg/l
Propylamine: normally absent
Isobutylamine: up to 15 mg/l
Butylamine: up to 10 mg/l
Tryptamine: up to 20 mg/l
Phenylethylamine: up to 20 mg/l
Putrescine or 1,4-diaminobutane: up to 40 mg/l
2-Methylbutylamine: up to 20 mg/l
3-Methylbutylamine: up to 20 mg/l
Cadaverine or 1,5-diaminopentane: up to 20 mg/l
Hexylamine: up to 10 mg/l
2.
DEFINITION
The biogenic amines measured are:
Ethanolamine: C2H7NO – CAS [141 – 43 – 5]
Histamine: C5H9N3 - CAS [51 – 45 – 6]
Methylamine: CH5N – CAS [74 – 89 – 5]
Serotonin: C10H12N2O – CAS [153 – 98 – 0]
Ethylamine: C2H7N – CAS [557 – 66 – 4]
Tyramine: C8H11NO - CAS [60 – 19 – 5]
Isopropylamine: C3H9N - CAS [75 – 31 – 0]
Propylamine: C3H9N – CAS [107 – 10 – 8]
Isobutylamine: C4H11N – CAS [78 – 81 – 9]
Butylamine: C4H11N – CAS [109 – 73 – 9]
OIV-MA-AS315-18 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
Tryptamine: C10H12N2 – CAS [61 – 54 – 1]
Phenylethylamine: C8H11N – CAS [64 – 04 – 0]
Putrescine or 1,4-diaminobutane: C4H12N2 – CAS [333 – 93 – 7]
2-Methylbutylamine: C5H13N - CAS [96 – 15 – 1]
3-Methylbutylamine: C5H13N - CAS [107 – 85 – 7]
Cadaverine or 1,5-diaminopentane: C5H14N2 – CAS [1476 – 39 – 7]
1,6-Diaminohexane: C6H16N2 – CAS [124 – 09 – 4]
Hexylamine: C6H15N – CAS [111 – 26 – 2]
3.
PRINCIPLE
The biogenic amines are directly determined by HPLC using a C18 column after Ophthalaldehyde (OPA) derivatization and fluorimetric detection.
4.
REAGENTS AND PRODUCTS
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
4.12
4.13
4.14
4.15
4.16
4.17
4.18
4.19
4.20
4.21
4.22
4.23
4.24
High purity resistivity water (18MΩ·cm)
Dihydrate disodium hydrogenophosphate - purity  99 %
Acetonitrile - Transmission minimum at 200 nm - purity  99 %
O-phthalaldehyde (OPA) - Application for fluorescence - purity  99 %
Disodium tetraborate decahydrate - purity  99 %
Methanol - purity  99 %
Hydrochloric acid 32 %
Sodium hydroxide pellets - purity  99 %
Ethanolamine - Purity  99 %
Histamine dichlorhydrate - Purity  99 %
Ethylamine chlorhydrate - Purity  99 %
Serotonin - Purity  99 %
Methylamine chlorhydrate – Purity  98 %
Tyramine chlorhydrate - Purity  99 %
Isopropylamine purity 99 %
Butylamine - Purity  99 %
Tryptamine chlorhydrate - purity  98 %
Phenylethylamine - Purity  99 %
Putrescine dichlorhydrate - Purity  99 %
2-Methylbutylamine - Purity  98 %
3-Methylbutylamine - Purity  98 %
Cadaverine dichlorhydrate - Purity  99 %
1-6-Diaminohexane - Purity  97 %
Hexylamine - Purity  99 %
OIV-MA-AS315-18 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
4.25
4.26
mg/l)
Nitrogen (maximum impurities: H2O  3 mg/l; O2  2 mg/L; CnHms  0.5
mg/l)
Helium (maximum impurities: H2O  3 mg/l; O2  2 mg/L; CnHm  0.5
Preparation of reagent solutions:
4.27
Preparation of eluents
Phosphate solution A: Weigh 11.12 g  0.01 g of di-basic sodium phosphate (4.2)
in a 50-ml beaker (5.5) on a balance (5.27). Transfer to a 2-litre volumetric flask
(5.9) and make up to 2 litres with high purity water (4.1). Homogenize using a
magnetic stirrer (5.30) and filter over a 0.45 µm membrane (5.17). Put in the 2-litre
bottle (5.12).
Solution B: The acetonitrile (4.3) is used directly.
4.28
OPA solution – Daily preparation
Weigh 20 mg  0.1 mg of OPA (4.4) in a 50-ml flask (5.7) on the precision
balance (5.27). Make up to 50 ml with methanol (4.6). Homogenize.
4.29
Preparation of the borate buffer (4.29) – Weekly preparation
Weigh 3.81 g  0.01 g of Na2B4O7·10H2O (4.5) in a 25-ml beaker (5.6) on the
precision balance (5.27). Transfer to a 100-ml volumetric flask (5.8) and make up
to 100 ml with demineralised water (4.1). Homogenize with a magnetic stirrer
(5.30), transfer to a 150-ml beaker (5.4) and adjust to pH 10.5 using a pH meter
(5.28 and 5.29) with 10 N soda (4.8).
4.30
0.1 M hydrochloric acid solution: Put a little demineralised water (4.1)
into a 2-litre volumetric flask (5.9). Add 20 ml of hydrochloric acid (4.7) using a
10-ml automatic pipette (5.24 and 5.25)
4.31
Calibration solution in 0.1 M hydrochloric acid
Guideline concentration of the calibration solution - weigh at ± 0.1 mg
OIV-MA-AS315-18 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
Indicative final concentration in
the calibration mix in mg/l
Ethanolamine
5
Histamine
5
Methylamine
1
Serotonin
20
Ethylamine
2
Tyramine
7
Isopropylamine
4
Propylamine
2.5
Isobutylamine
5
Butylamine
5
Tryptamine
10
Phenylethylamine
2
Putrescine
12
2- Methylbutylamine
5
3- Methylbutylamine
6
Cadaverine
13
1.6 Diaminohexane
8
Hexylamine
5
The true concentration of the calibration solution is recorded with the batch
number of the products used.
Certain biogenic amines being in salt form, the weight of the salt needs to be taken
into account when determining the true weight of the biogenic amine.
The stock solution is made in a 100-ml volumetric flask (5.8).
The surrogate solution is made in a 250-ml volumetric flask (5.10).
4.32
1,6 Diaminohexane internal standard
Weigh exactly 119 mg in a 25-ml Erlenmeyer flask (5.1) on a balance (5.26).
Transfer to a 100-ml volumetric flask (5.8) and top up to the filling mark with 0.1
N hydrochloric acid (4.30).
OIV-MA-AS315-18 : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
4.33
2-Mercaptoethanol - Purity  99 %.
5.
APPARATUS
5.1
25-ml Erlenmeyer flasks
5.2
250-ml Erlenmeyer flasks
5.3
100-ml beakers
5.4
150-ml beakers
5.5
50-ml beaker
5.6
25-ml beaker
5.7
50-ml volumetric flasks
5.8
100-ml volumetric flasks
5.9
2,000-ml volumetric flasks
5.10
250-ml volumetric flask
5.11
1-litre bottles
5.12
2-litre bottle
5.13
2-ml screw cap containers suitable for the sample changer
5.14
50-ml syringe
5.15
Needle
5.16
Filter holder
5.17
0.45 µm cellulose membrane
5.18
0.8 µm cellulose membrane
5.19
1.2 µm cellulose membrane
5.20
5 µm cellulose membrane
5.21
Cellulose pre-filter
5.22
1-ml automatic pipette
5.23
5-ml automatic pipette
5.24
10-ml automatic pipette
5.25
Cones for 10-ml, 5-ml and 1-ml automatic pipettes
5.26
Filtering system
5.27
Balances for weighing 0 to 205 g at ± 0.01 mg
5.28
pH meter
5.29
Electrode
5.30
Magnetic stirrer
5.31
HPLC pump
5.32
Changer-preparer equipped with an oven
Note: An oven is indispensable, if a changer-preparer is used for injecting several
samples one after another. This operation may likewise be done manually) the
results may be less precise;
OIV-MA-AS315-18 : R2009
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
5.33
Injection loop
5.34
5 µm C18 column, 250 mm  4 (which must lead to a similar
chromatogram as presented in annex B);
5.35
Fluorimetric detector
5.36
Integrator
5.37
Borosilicic glass tube with a stopper and closure cap covered with PTFE
(ex Sovirel 15).
6.
PREPARATION OF SAMPLES
Samples are previously purged of gas with nitrogen (4.25).
6.1
Filtering
Filter approximately 120 ml of the sample over membrane:
- for a wine: 0.45 µm (5.17),
- for a must or non-clarified wine: 0.45 (5.17) – 0.8 (5.18) – 1.2 (5.19) - 5 µm
(5.20) + pre-filter (5.21), pile filters in the following order, the sample pushed by
the top: 0.45 µm (5.17) + 0.8 µm (5.18) + 1.2 µm (5.19) + 5 µm (5.20) +
prefiltered (5.21)
6.2
Preparation of the sample
Put 100 ml of the sample (6.1) into a 100-ml volumetric flask (5.8);
Add 0.5 ml of 1-6-diaminohexane (4.32) at 119 mg/100 ml using a 1-ml automatic
pipette (5.21 and 25);
Draw off 5 ml of the sample using the pipette (5.23 and 5.25); pour this into a 25ml Erlenmeyer flask (5.1);
Add 5 ml of methanol to this (4.6) using the pipette (5.23 and 5.25);
Stir to homogenize;
Transfer to containers (5.13);
Start the HPLC pump (5.31), then inject 1 µl (5.32 and 5.33)
6.3
Derivatisation
In a borosilicic glass tune (5.37), pour 2 ml of OPA solution (4.28), 2 ml of borate
buffer (4.29), 0,6 ml of 2-mercaptoethanol (4.33). Close, mix (5.30). Open and
pour 0,4 ml of sample. Close, mix (5.30). Inject immediately, as the derivitive is
not stable. Rinse recipient immediately after injection, due to odour.
Note: Derivatisation can be carried out by an automatic changer-preparer. In this
case, the process will be programmed to come close to the proportion of manual
derivisation
OIV-MA-AS315-18 : R2009
6
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
6.4
Routine cleaning
Syringe (5.13) and needle (5.14) rinsed with demineralised water (4.1) after each
sample;
filter holder (5.16) rinsed with hot water, then MeOH (4.6). Leave to drain and
dry.
7.
PROCEDURE
Mobile phase (5.31)
- A: phosphate buffer (4.2)
- B: acetonitrile (4.3)
Elution gradient:
time
( in mins)
0
15
23
42
55
60
70
95
%A
%B
80
70
60
50
35
35
80
80
20
30
40
50
65
65
20
20
Note: The gradient can be adjusted to obtain a chromatogram close to the one
presented in annex B
Flow rate: 1 ml/min;
Column temperature: 35 °C (5.32);
Detector (5.35): Exc = 356 nm, Em = 445 nm (5.30);
Internal calibration
The calibration solution is injected for each series;
Calibration by internal standard;
Calculation of response factors:
RF = Ccis  area i / area is  Cci
OIV-MA-AS315-18 : R2009
7
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
Cci = concentration of the component in the calibration solution and
Ccis = concentration of the internal standard in the calibration solution (1-6diaminohexane).
Area i = area of the product peak present in the sample
Area is = area of the internal standard peak in the sample
Calculation of concentrations:
Cci = (XF  area i)/ (area is  RF)
Area i = area of the product peak present in the sample
Area is = area of the internal standard peak present in the sample
XF = quantity of internal calibration added to samples for analysis
XF = 119  0.5/100 = 5.95.
8.
EXPRESSION OF RESULTS
Results are expressed in mg/l with one significant digit after the decimal point.
9.
RELIABILITY
r (mg/l)
R (mg/l)
Histamine
0.07x + 0.23
0.50x + 0.36
Methylamine
0.11x + 0.09
0.40x + 0.25
Ethylamine
0.34x - 0.08
0.33x + 0.18
Tyramine
0.06x + 0.15
0.54x + 0.13
Phenylethylamine
0.06x + 0.09
0.34x + 0.03
Diaminobutane
0.03x + 0.71
0.31x + 0.23
2-methylbutylamine et 3methylbutylamine
0.38x + 0.03
0.38x + 0.03
Diaminopentane
0.14x + 0.09
0.36x + 0.12
The details of the interlaboratory trial with regard to reliability of the method are
summarised in appendix A.
OIV-MA-AS315-18 : R2009
8
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
10.
OTHER CHARACTERISTICS OF THE ANALYSIS
The influence of certain wine components: amino acids are released at the
beginning of the analysis and do not impede in detection of biogenic amines.
The limit of detection (LOD) and limit of quantification (LOQ) according to an
intralaboratory study
11.
LOD (in mg/l)
LOQ (in mg/l)
Histamine
0,01
0,03
Methylamine
0,01
0,02
Ethylamine
0,01
0,03
Tyramine
0,01
0,04
Phenylethylamine
0,02
0,06
Diaminobutane
0,02
0,06
2-methylbutylamine
0,01
0,03
3-methylbutylamine
0,03
0,10
Diaminopentane
0,01
0,03
QUALITY CONTROL
Quality controls may be carried out with certified reference materials, with wines
the characteristics of which result from a consensus or spiked wines regularly
inserted into analytical series and by following the corresponding control charts.
OIV-MA-AS315-18 : R2009
9
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
Annex A
Statistical data obtained from the results of interlaboratory trials
The following parameters were defined during an interlaboratory trial. This
trial was carried out by the Oenology Institute of Bordeaux (France) under
the supervision of the National Interprofessional Office of Wine (ONIVINS –
France).
Year of interlaboratory trial: 1994
Number of laboratories: 7
Number of samples: 9 double blind samples
(Bulletin de l’O.I.V. November-December 1994, 765-766, p.916 to 962) numbers
recalculated in compliance with ISO 5725-2:1994.
Types of samples: white wine (BT), white wine (BT) fortified = B1, white wine
(BT) fortified = B2, red wine n°1 (RT), red wine fortified = R1, red wine (RT)
fortified = R2, red wine n°2 (CT), red wine (CT) fortified = C1 and red wine (CT)
fortified = C2. fortified in mg/l.
HistN
MetN
EthN
TyrN
PhEtN
DiNbut
IsoamN
DiNpen
wine B1
wine BT
+ 0,5
vine BT
+ 0,12
wineBT
+ 0,13
wine BT
+ 0,36
vine BT
+ 0,15
wine BT
+ 0,5
wine BT
+ 0,28
wineBT
+ 0,25
wine B2
wine BT
+2
wine BT
+ 0,40
wine BT
+ 0,50
wine BT
+ 1,44
wine BT
+ 0,60
wine BT
+2
Wine BT
+ 0,1,74
wine BT
+ 1,04
wine C1
wine CT
+2
wine CT
+ 0,1
wine CT
+ 0,18
wine CT
+ 0,72
wine CT
+ 0,15
wine CT
+2
wine CT
+ 0,29
wine CT
+ 0,26
wineC2
wine CT
+4
wine CT
+ 0,41
wine CT
+ 0,50
wine CT
+ 2,90
wine CT
+ 0,58
wine CT
+8
wine CT
+ 1,14
wine CT
+ 1,04
wine R1
wine RT
+2
wine RT
+ 0,14
wine RT
+ 0,13
wine RT
+ 1,45
wine RT
+ 0,19
wine RT
+3
wine RT
+ 0,0,57
wine RT
+ 0,51
wine R2
wine RT
+5
wine RT
+ 0,41
wine RT
+ 0,50
wine RT
+ 2,88
wine RT
+ 0,59
wine RT
+ 10
wine RT
+ 2,28
wine RT
+ 2,08
HistN : histamine, MetN : methylamine, EthN : ethylamine, TyrN : tyramine,
PhEtN : phenylethylamine, DiNbut : diaminobutane, IsoamN : isoamylamine and
DiNpen : diaminopentane.
OIV-MA-AS315-18 : R2009
10
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
Annex B : Chromatogram model obtained by this method
OIV-MA-AS315-18 : R2009
11
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Biogenic amines
BIBLIOGRAPHY
TRICARD C., CAZABEIL J.-M., SALAGOÏTI M.H. (1991): Dosage des amines
biogènes dans les vins par HPLC, Analusis, 19, M53-M55.
PEREIRA MONTEIRO M.-J. et BERTRAND A. (1994): validation d'une méthode
de dosage – Application à l'analyse des amines biogènes du vin. Bull. O.I.V., (765766), 916-962.
OIV-MA-AS315-18 : R2009
12
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Glutathione
Method OIV-MA-AS315-19
Type IV method
Determination of glutathione in musts and wines by capillary
electrophoresis
(Resolution OIV-Oeno 345/2009)
1. Scope
This method makes it possible to determine the glutathione content of musts and
wines in a concentration range of 0 to 40 mg/L. It uses capillary electrophoresis
(CE) associated with fluorimetric detection (LIF).
2. Principle
The method used, which proceeds by capillary electrophoresis, is an adaptation of
the method developed by Noctor and Foyer (1998) to determine non-volatile thiols
in poplar leaves using HPLC coupled with fluorimetric detection.
The separation of a mixture’s solutes by capillary electrophoresis is obtained by
differential migration in an electrolyte. The capillary tube is filled with this
electrolyte.
The sample to be separated is injected into one end of the capillary tube. As a
result of electrical field activity generated by the electrodes immersed in the
electrolyte, the solutes separate due to differences in migration speed and are
detected near the other end of the capillary tube in the form of peaks. In given
operating conditions, migration times constitute a criterion for the identification of
chemical species and the peak area is proportional to the quantity injected.
3. Products and reagents
3.1 List of products
3.1.1 Glutathione (GSH, > 98 %)
3.1.2 Dithiothreitol (DTT, > 99 %)
3.1.3 Anhydrous monobasic sodium phosphate (NaH2PO4, > 99 %)
3.1.4 Anhydrous dibasic sodium phosphate (Na2HPO4, > 99 %)
3.1.5 2-(N-cyclohexylamino)ethanesulfonic acid (CHES, > 98 %),
3.1.6 Monobromobimane (MBB, 97 %)
3.1.7 Ethylenediamine tetraacetic acid sodium salt (EDTA, > 99 %)
3.1.8 Sodium hydroxide
3.1.9 Hydrochloric acid (35 %)
OIV-MA-AS315-19 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Glutathione
3.1.10 Acetonitrile (99.5 %)
3.1.11 Ultra-pure water with a resistance of >18 MΩ·cm.
3.2 List of solutions
All solutions are homogenised prior to use
3.2.1 Electrophoretic buffer: phosphate buffer, 50 mM, pH 7
This buffer is prepared using two solutions - A and B
3.2.1.1 Solution A: 3 mg of anhydrous monobasic phosphate (3.1.3) taken up by
250 ml ultra-pure water (3.1.11)
3.2.1.2 Solution B: 3.55 mg of anhydrous dibasic phosphate (3.1.4) taken up by
250 ml ultra-pure water (3.1.11)
The phosphate buffer is obtained by the addition of 40 ml of solution A (3.2.1.1)
and 210 ml of solution B (3.2.1.2) and then made up to 500 ml with ultra-pure
water (3.1.11). The buffer’s pH is then adjusted to 7 using hydrochloric acid
(3.1.9).
3.2.2 Monobromobimane solution (MBB) - 50 mM
25 mg of monobromobimane (MBB) (3.1.6) are taken up by 1,850 µl of
acetonitrile (3.1.10).
Stored in the dark at -20 °C, this reagent remains stable for three months.
3.2.3 0.1 M sodium hydroxide solution
0.4 g of sodium hydroxide (3.1.8) are put into a 100-ml volumetric flask and taken
up by 100 ml of ultra-pure water (3.1.11).
3.2.4 5 M sodium hydroxide solution
20 g of sodium hydroxide (3.1.8) are put into a 100-ml volumetric flask and taken
up by 100 ml of ultra-pure water (3.1.11).
3.2.5 CHES buffer: 0.5 M, pH 9.3
2.58 g of 2-(N-cyclohexylamino) ethanesulfonic acid (CHES) (3.1.5) are dissolved
in approximately 20ml of ultra pure water (3.1.11). The pH buffer is adjusted to
9.3 by the addition of sodium hydroxide 5 M (3.2.4). The volume is then adjusted
to 25 ml with ultra pure water (3.1.11). This buffer is divided between the 1.5-ml
test tubes (Eppendorf type) with 1 ml per tube. Stored at –20 °C, the CHES
aqueous solution may be kept for several months.
3.2.6 Dithiothreitol solution (DTT) - 10 mM
15.4 mg of dithiothreitol (3.1.2) is dissolved in 10 mL of ultra pure water (3.1.11)
then this solution is divided in 1.5-ml test tube (Eppendorf type) with 1 ml per tube
Stored at –20 °C, this DTT aqueous solution may be kept several months.
OIV-MA-AS315-19 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Glutathione
4. Apparatus
4.1 Capillary electrophoresis
Capillary electrophoresis equipped with a hydrostatic-type injector is coupled with
a laser-induced fluorescence detector with an excitation wavelength similar to the
absorption wavelength of the MBB-GSH adduct: e.g.= 390 nm (e.g. Zetalif
detector).
4.2 The capillary tube
The total length of the non-grafted silica capillary tube is 120 cm. Its effective
length is 105 cm, and its internal diameter is 30 µm.
5. Preparation of samples
The method of determination used consists of the derivatization of the SH
functions by the monobromobimane (MBB) (Radkowsky & Kosower, 1986).
Samples of musts or non-bottled wines are clarified by centrifugation prior to
analysis. Bottled wines are analysed without prior clarification.
Preparation of samples:
In a 1.5-ml test tube (Eppendorf type), put successively:
- 200 µl of the sample,
- 10 µl of the DTT solution (3.2.4) - final concentration of 0.25 mM,
- 145 µl of CHES (3.2.3) - final concentration of 179 mM,
- 50 µl of MBB (3.2.2) - final concentration of 6.2 mM.
After stirring the reagent mixture, the derivatization of thiol functions by the MBB
requires a 20-minute incubation period in the dark at ambient temperature. In these
analytical conditions, the MBB-SR derivatives thus formed are relatively unstable;
CE-LIF determination should be carried out immediately after incubation.
6. Procedure
6.1 Capillary tube preparation
Before being used for the first time and as soon as migration times increase, the
capillary tube (4.2) should be treated in the following way:
6.1.1. Rinse with 0.1 M sodium hydroxide (3.2.5) for 3 minutes,
6.1.2. Rinse with ultra-pure water (3.1.12) for 3 minutes,
6.1.3. Rinse with the electrophoretic phosphate buffer (3.2.1) for 3 minutes.
OIV-MA-AS315-19 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Glutathione
6.2 Migration conditions
6.2.1 Injection of the sample is of the hydrostatic type; 3 s at 50 kPa.
This is followed by injection of 50 mb electrophoretic buffer (3.2.1) to improve
peak resolution (Staking).
6.2.2 Analysis.
A voltage of +30 kV, applied throughout separation, generates a current of 47 µA.
These conditions are reached in 20 s. Separation is carried out at a constant
temperature of 21 °C.
6.2.3 Rinsing the capillary tube
The capillary tube should be rinsed after each analysis, successively with:
- 0.1M sodium hydroxide (3.2.5) for 3 minutes,
- ultra-pure water (3.1.12) for 3 minutes,
- electrophoretic phosphate buffer (3.2.1) for 3 minutes.
GSH level (mg/l)
7. Results
At the concentration ultimately used in the sample, the presence of DTT during
derivatization makes it possible to stabilise the unstable functions of thiols that
have an alkaline pH and are very easily oxidized by quinines produced by phenolic
compound auto-oxidation, but does not break the disulphide bonds. Thus, under
these analytical conditions, the reduced glutathione content (GSH) found in a wine
with or without the addition of 10 mg/l of oxidized glutathione (GSSG) is strictly
comparable (Figure 1). This method therefore makes it possible to determine
glutathione content in its reduced form alone.
control
control GSSG
Figure 1: Demonstration of the stability of disulphide bonds according to the
conditions of derivatization described. (DTT, ultimately 0.25 mM).
OIV-MA-AS315-19 : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Glutathione
Figure 2 shows the electrophoretic profile of a white grape must sample
(Sauvignon) in which cysteine, glutathione, N-acetyl-cysteine and sulphur dioxide
are identified. The first peak corresponds to excess reagents (DTT, MBB). The
separation of non-volatile thiols takes less than 20 minutes. Only certain peaks
could be identified (Figure 2, A) (Newton et al., 1981). These thiols, apart from
the sulphur dioxide, are generally present in varying quantities in grapes (Cheynier
et al., 1989), fruit and vegetables (Mills et al., 2000).
Figure 2: Example of the separation of the known non-volatile thiols in an
HCl/EDTA solution (A) 1 and in a grape must (B): DTT; 2: homocysteine;
3: cysteine; 4: Cys-Gly; 5: GSH; 6: g Glu-Cys; ,7: NAC; 8: SO2 .
In these analytical conditions, MBB-RS adduct retention times are as follows:
MBB-homocysteine 10.40 mins; MBB-cysteine 10.65 mins, MBB-GSH 14.14
mins; MBB-NAC 15.41mins; MBB-SO2 18.58mins.
8. Characteristics of the method
Certain internal elements of validation were determined, but do not constitute
formal validation according to the protocol for the design, conducts and
interpretation of methods of analysis performance studies (OIV 6/2000).
OIV-MA-AS315-19 : R2009
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Glutathione
Wine is used as a matrix to produce calibration curves and repeatability tests for
each compound. Each concentration is calculated based on the average of three
determinations obtained by using the right of the calibration curb regression.
Results are expressed in mg/L.
Linear regressions and correlation coefficients are calculated according to the least
squares method. The stock solutions of the various thiols are produced from an
HCl/EDTA solution, allowing them to be stored at +6 °C for several days with no
loss. Successive dilutions of these solutions allow the threshold limits for detection
in wine to be estimated, for a signal-to-noise ratio of three of more.
The linearity spectrum varies according to thiols (Table 1).
Table 1: Linearity spectrum, linear regression properties for each thiol in
solutions prepared in exactly the same way as that of the glutathione.
Homocysteine
Cysteine
Glutathione
N-acetyl-cysteine
Linearity spectrum
Linear regression
0 - 15 mg/l
0 - 15 mg/l
0 - 40 mg/l
0 - 10 mg/l
Y= 0.459X – 0.231
Y = 0.374X – 0.131
Y = 0.583X – 0.948
Y = 0.256X – 0.085
Correlation
coefficient
0.9987
0.9979
0.9966
0.9982
These analytical conditions make it possible to eliminate interference caused by
MBB hydrolysis products, unlike the reported findings of other works (Ivanov et
al., 2000).
The method’s repeatability is calculated on the basis of ten analyses of the same
sample of wine. For a thiol concentration of 10 mg/l, the coefficient of variation is
6.0 % for the glutathione; besides this, it is 3.2 % for the homocysteine, 4.8 % for
the cysteine and 6.4 % for the N-acetyl-cysteine.
The limit for detecting glutathione is 20 µg/l and the quantification limit is 60 µg/l.
OIV-MA-AS315-19 : R2009
6
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Glutathione
9. Bibliography
Noctor, G. and C. Foyer, 1998. Simultaneous measurement of foliar glutathione,
gamma-glutamylcysteine, and amino acids by high-performance liquid
chromatography: comparison with two other assay methods for glutathione,
Analytical Biochemistry, 264, 98-110.
Kosower, N.S., Kosower E. M., Newton G. L.,and Ranney H. M., 1979. Bimane
fluorescent labels: Labeling of normal human red cells under physiological
conditions. Proc. Natl. Acad. Sci., 76 (7), 3382-3386.
Newton, G.L., R. Dorian, and R.C. Fahey, Analysis of biological thiols:
derivatisation with monobromobimane and separation by reverse-phase highperformance liquid chromatography. Anal. Biochem., 1981. 114: p. 383-387.
Cheynier, V., J.M. Souquet, and M. Moutounet, 1989. Glutathione content and
glutathione to
hydroxycinnamique acid ration in Vitis vinifera grapes and musts. Am.
J.Enol.Vitic,. 40 (4), 320-324.
Mills, B.J., Stinson C. T., Liu M. C. and Lang C. A., 1997. Glutathione and
cyst(e)ine profiles of vegetables using high performance liquid chromatography
with dual electrochemical detection. Journal of food composition and analysis, 10,
90-101.
Ivanov, A.R., I.V. Nazimov, and L. Baratova, 2000. Determination of biologically
active low molecular mass thiols in human blood. Journal of Chromatogr. A,. 895,
167-171.
OIV-MA-AS315-19 : R2009
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
METHOD FOR THE DETERMINATION OF -DICARBONYL
COMPOUNDS OF WINE by HPLC AFTER DERIVATIZATION BY 1,2DIAMINOBENZENE
(OIV-Oeno 386A-2010)
Method OIV-MA-AS315-20
Type IV method
1. Introduction
The principal α-dicarbonyl compounds found in wine (Fig 1) are: glyoxal,
methylglyoxal, diacetyl and pentane-2,3-dione, but only α-diketones are
relatively abundant in wine. Carbonyl compounds exist in all types of
wines, particularly after malolactic fermentation and in red wines. In
addition, sweet white wines produced with botrytized grapes can contain
high levels of glyoxal and methylglyoxal.
Glyoxal: OCH−CHO (ethanedial)
Methylglyoxal: CH3−CO−CHO (2-oxopropanal)
Diacetyl: CH3−CO−CO−CH3 (2,3-butanedione)
2,3-Pentanedione: CH3−CH2−CO−CO−CH3
2,3-Hexanedione: CH3−CH2−CH2−CO−CO−CH3
Figure 1. The principal α-dicarbonyl compounds of wine (2,3-hexanedione is not
naturally present in wine but it is used as internal standard).
Dicarbonyl compounds are important in wine for different reasons: their
sensory impact, the reactivity with other components of the wine or possible
microbiological effects.
OIV-MA-AS315-20: R2010
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
2. Applicability
This method applies to all types of wines (white, red, sweetened or
fortified), for dicarbonyl compounds with a content that ranges from 0.05
mg/l to 20 mg/l .
3. Principle
The method is based on the formation of derivatives of the quinoxaline type
based on the α-dicarbonyl compounds of the wine with 1,2-diaminobenzene
(Figure 2).
R
NH2
O C
N
R
N
R
+
NH2
1,2-diaminobenzene
O C
R
dicarbonyl
quinoxaline
Figure 2 Formation of derivatives.
The reaction takes place directly in the wine at pH 8 and after a reaction
time of 3 h at 60°C. The analysis of the derivatives is then carried out
directly by high-performance liquid chromatography (HPLC) and detection
by UV absorption at 313 nm.
4. Reagents and products
4.1 Dicarbonyl compounds
4.1.1 Glyoxal in a solution at 40% (CAS N° 107-22-3)
4.1.2 Methylglyoxal in a solution at 40% (CAS N° 78-98-8)
4.1.3 Diacetyl, purity > 99% (CAS N° 431-03-8)
4.1.4 2,3-Pentanedione, purity > 97% (CAS N° 600-14-6)
4.1.5 2,3-Hexanedione, purity > 90% (CAS N° 3848-24-6)
OIV-MA-AS315-20: R2010
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
4.2 1,2-Diaminobenzene in powder form, purity > 97%
4.3 Water for HPLC (for example microfiltered and with a resistivity of
18.2
MΩ) (CAS N° 95-54-5)
4.4 Pure ethanol for HPLC (CAS N° 64-17-5)
4.5 Sodium Hydroxide M (CAS N° 1310-73-2)
4.6 Pure crystallisable acetic acid (CAS N° 64-19-7)
4.7 Solvent A for the analysis by HPLC
To 1 l of water for HPLC (4.3) add 0.5 ml of acetic acid (4.8), mix,
degas (for example by sonication)
4.8 Solvent B for HPLC
Pure methanol for HPLC (CAS N° 67-56-1)
4.9 Aqueous-alcoholic solution at 50% vol.
Mix 50 ml of pure ethanol for HPLC (4.4) with 50 ml of water (4.3)
4.10 Solution of internal standard 2,3-hexanedione at 2.0 g/l
Place 40 mg of 2,3-hexanedione (4.2) in a 30-ml flask, dilute in 20
ml of aqueous- alcoholic solution to 50% vol (4.9) and stir until it has
completely dissolved.
5. Equipment
5.1 High-performance liquid chromatograph with detection by UV
absorption (313 nm);
5.1.1 Analytical column filled with 5 µm octadecyl silica whose dimensions
are for example 250 mm x 4.6 mm.
5.1.2 Data acquisition system.
5.2 pH measuring apparatus.
5.3 Magnetic stirrer.
5.4 Balance with a precision of 0.1 mg.
5.5 Solvent degasification system for HPLC (for example an ultrasonic
bath).
5. 6 Oven which can be set to 60°C.
5.7 Standard laboratory glassware including pipettes, 30-ml screw-cap
flasks, and microsyringes.
OIV-MA-AS315-20: R2010
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
6. Preparation of the sample
No specific preparation is necessary.
7. Procedure
Place 10 ml of wine in a 30-ml flask (5.7)
Bring to pH 8 while stirring, with sodium hydroxide M (4.5)
Add 5 mg of 1,2-diaminobenzene (4.2)
Add 10 µl of 2,3-hexanedione (internal standard) at 2.0 g/L (4.10)
Close the flask using a screw-cap fitted with a Teflon-faced seal
Stir until the reagent has completely disappeared (5.3)
Place in the oven at 60°C for 3 h (5.6)
Cool.
7.1 Optimisation and analytical conditions
The yield of the reaction of the dicarbonyl compounds with the 1-2diaminobenzene is optimal at pH 8. Solutions of dicarbonyl compounds
have been derivatized at 25, 40 or 60°C and then analysed by HPLC
according to the protocol described in point 7.2 at different times (Table 1).
Diketones require much more reaction time and a higher reaction
temperature. The reaction is slower with molecules with longer chains (2,3pentanedione and 2,3-hexanedione).
In addition, no interference of SO2 with the formation of quinoxalines was
noted during the study of the method.
OIV-MA-AS315-20: R2010
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
Table 1. Effect of reaction time and temperature on the formation of
derivatives by diaminobenzene from glyoxal, diacetyl and 2,3-hexanedione
Reaction time
1h
2h
3h
Temperature (°C)
Recovery rate (%)
Glyoxal
25
40
60
92
95
96
93
97
98
94
98
100
Diacetyl
25
40
60
23
64
85
77
89
100
87
94
100
2,3-Hexanedione
25
40
60
17
55
69
67
79
93
79
88
100
7.2 Analysis by HPLC
- Injection. After cooling, 20 µl of the reaction medium containing the
quinoxalines is directly injected into the HPLC system.
- Elution programme. For the separation, the elution programme is
presented in Table 2
Table 2. Elution programme for the analysis by HPLC
Time in minutes
0
8
26
30
32
40
45
50
solvent A
80
50
25
0
0
100
80
80
OIV-MA-AS315-20: R2010
solvent B
20
50
75
100
100
0
20
20
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
The flow rate is 0.6 ml/min
- Separation. The chromatogram obtained by HPLC is shown in Figure 3.
- Detection. The maximum absorbance was studied for all the derivatized
dicarbonyl compounds and set at 313 nm as being optimal.
- Identification of derivatives. The identification of the derivatives was
carried out by comparing the retention times with standard reference
solutions. The chromatographic conditions permit a good separation of the
peaks in all wines.
7. 2.1 Characteristics of the method by HPLC
Some internal validations methods have been determined but do not
constitute a formal validation proccess according to the protocol governing
the planning, the implementing and the interpreting of performance studies
pertaining to analysis methods (OIV 6/2000)
- Repeatability. The repeatability of the method was calculated using 10
analyses of the same wine (Table 3).
Table 3. Repeatability study and performance of the method
Average*
Standard
CV (%)
deviation
White wine
Glyoxal
4.379
0.101
2.31
Methylglyoxal
2.619
0.089
3.43
Diacetyl
5.014
0.181
3.62
2.3-Pentanedione
2.307
0.097
4.21
Red wine
Glyoxal
Methylglyoxal
Diacetyl
2,3-Pentanedione
2.211
1.034
1.854
0.698
0.227
0.102
0.046
0.091
10.30
9.91
2.49
13.09
* Results in mg/l based on 10 analyses of the same wine.
OIV-MA-AS315-20: R2010
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
- Linearity. The linearity of the method was tested using standard solutions
(using an aqueous-alcoholic solution at 12% vol. as a matrix) (Table 4). The
quantitative analysis of the additions of dicarbonyl compounds showed that
the method is linear for the four compounds and that its precision is
satisfactory.
Table 4. Study of the linearity and recovery tests with standard solutions
(water-ethanol at 12% v/v) Value of the correlation coefficient
Glyoxal
valuea peak area b
Methylglyoxal
Diacetyl
Pentane-2,3-dione
R = 0.997
R = 0.999
R = 0.999
valuea peak area b
valuea peak area b valuea peak area b
1
R = 0.992
- The recovery of additions carried out in red and white wines demonstrated
the satisfactory performance of the method . Contained in the 92% - 116%
range for extreme values
- The quantification limit of the dicarbonyl compounds is very low, the best
results being obtained with diacetyl, whose detection limit is 10 times lower
than that of the other compounds (Table 5).
Table 5. Performance of the method by HPLC for the quantification of
dicarbonyl compounds
Limits
Glyoxal
Methylglyoxal
Diacetyl
2.3-Pentanedione
detectiona
0.015
0.015
0.002
0.003
determinationa
quantificationa
0.020
0.020
0.002
0.004
0.028
0.027
0.003
0.006
a: results in mg/l, aqueous-alcoholic solution (10% vol).
OIV-MA-AS315-20: R2010
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
Figure 3. High-performance liquid phase chromatogram of dicarbonyl compounds
derivatized by 1,2-diaminobenzene from a white wine, detected by UV at 313 nm.
Spherisorb ODS Column 250 mm x 4.6 mm x 5 µm.
OIV-MA-AS315-20: R2010
8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by HPLC after derivatization
Bibliography
Bartowski E.J. and Henschke P.A., The buttery attribute of wine – diacetyl
– desirability spoilage and beyond. Int. J. Food Microbiol. 96: 235-252
(2004).
Bednarski W., Jedrychowski L., Hammond E., and Nikolov L., A method
for determination of -dicarbonyl compounds. J. Dairy Sci. 72:2474-2477
(1989).
Leppannen O., Ronkainen P., Koivisto T. and Denslow J., A
semiautomatic method for the gas chromatographic determination of vicinal
diketones in alcoholic beverages. J. Inst. Brew. 85:278- 281 (1979).
Martineau B., Acree T. and Henick-Kling T., Effect of wine type on the
detection threshold for diacetyl. Food Res. Int. 28:139-143 (1995).
Moree-Testa P. and Saint-Jalm Y., Determination of -dicarbonyl
compounds in cigarette smoke. J. Chromatogr. 217:197-208 (1981).
De Revel G., Pripis-Nicolau L., Barbe J.-C. and Bertrand A., The detection
of α-dicarbonyl compounds in wine by the formation of quinoxaline
derivatives. J. Sci. Food Agric. 80:102-108 (2000).
De Revel G. and Bertrand A., Dicarbonyl compounds and their reduction
products in wine. Identification of wine aldehydes. Proc. 7th Weurman
Flavour Research Symp, Zeist, June, pp 353-361 (1994).
De Revel G. and Bertrand A., A method for the detection of carbonyl
compounds in wine: glyoxal and methylglyoxal. J. Sci. Food Agric. 61:267272 (1993).
Voulgaropoulos A., Soilis T. and Andricopoulos N., Fluorimetric
determination of diacetyl in wines after condensation with 3,4diaminoanisole. Am. J. Enol. Vitic. 42:73-75 (1991).
Gilles de Revel et Alain Bertrand Analyse des composés α-dicarbonyles
du vin après dérivation par le 2,3-diaminobenzène OIV FV 1275
OIV-MA-AS315-20: R2010
9
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by GC after derivatization
METHOD FOR THE DETERMINATION OF DICARBONYL COMPOUNDS OF WINE BY GC AFTER
DERIVATIZATION BY 1,2-DIAMINOBENZENE
(OIV-Oeno 386B-2010)
Method OIV-MA-AS315-21
Type IV method
1. Introduction
The principal α-dicarbonyl compounds found in wine (Fig 1) are: glyoxal,
methylglyoxal, diacetyl and 2,3-pentanedione, but only α-diketones are
relatively abundant in wine. Carbonyl compounds exist in all types of
wines, particularly after malolactic fermentation and in red wines. In
addition, sweet white wines produced with botrytized grapes can contain
high levels of glyoxal and methylglyoxal.
Glyoxal: OCH−CHO (ethanedial)
Methylglyoxal: CH3−CO−CHO (2-oxopropanal)
Diacetyl: CH3−CO−CO−CH3 (2,3-butanedione)
2,3-pentanedione: CH3−CH2−CO−CO−CH3
2,3-hexanedione: CH3−CH2−CH2−CO−CO−CH3
Figure 1. The principal α-dicarbonyl compounds of wine (2,3-hexanedione
is not naturally present in wine but it is used as internal standard).
OIV-MA-AS315-21: R2010
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by GC after derivatization
Dicarbonyl compounds are important in wine for different reasons: their
sensory impact, the reactivity with other components of the wine or possible
microbiological effects.
2. Applicability
This method applies to all types of wines (white, red, sweetened or
fortified), for carbonyl derivatives content ranging from 0.05 mg/L and 20
mg/L.
3. Principle
The method is based on the formation of derivatives of the quinoxaline type
based on the α-dicarbonyl compounds of the wine with 1,2-diaminobenzene
(Figure 2).
R
NH2
O C
N
R
N
R
+
NH2
1,2-diaminobenzene
O C
R
di-carbonyl
quinoxaline
Figure 2 Formation of derivatives.
The reaction takes place directly in the wine at pH 8 and after a reaction
time of 3 h at 60°C. The analysis of the derivatives is then carried out after
extraction of the derivatives by dichloromethane and analysis by gas
chromatography with detection by mass spectrometry (GC-MS) or using a
nitrogen-specific detector.
4. Reagents and products
4.1 Dicarbonyl compounds
4.1.1 Glyoxal in a solution at 40% (CAS n° 107-22-3)
4.1.2 Methylglyoxal in a solution at 40% (CAS n° 78-98-8)
4.1.3 Diacetyl, purity > 99% (CAS n° 431-03-8)
4.1.4 2,3-Pentanedione, purity > 97% (CAS n° 600-14-6)
4.1.5 2,3-Hexanedione, purity > 90% (CAS n° 3848-24-6)
4.2 1,2-Diaminobenzene in powder form, purity > 97% (CAS n° 95-54-5)
OIV-MA-AS315-21: R2010
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by GC after derivatization
4.3 Water for HPLC (for example microfiltered and with a resistivity of
18.2
MΩ)
4.4 Pure ethanol for HPLC (CAS n° 64-17-5)
4.5 Sodium hydroxide M. (CAS n° 1310-73-2)
4.6 Sulphuric acid 2M (CAS n° 7664-93-9)
4.7 Dichchloromethane (CAS n° 75-09-2)
4.8 Anhydrous sodium sulphate (CAS n° 7757-82-6)
4.9 Aqueous-alcoholic solution at 50% vol .
Mix 50 ml of pure ethanol for HPLC (4.4) with 50 ml of water (4.3)
4.10 Solution of internal standard 2,3-hexanedione at 2.0 g/L
Place 40 mg of 2,3-hexanedione (4.2) in a 30-ml flask, dilute in 20
ml of aqueous-alcoholic solution to 50% vol (4.9) and stir until it has
completely dissolved.
4.11 Anhydrous sodium sulphate (CAS n° 7757-82-6)
5. Equipment
5.1 Gas chromatograph with detection by mass spectrometry (GC-MS) or a
nitrogen-specific detector.
5.1.1 Relatively polar, polyethylene glycol capillary column (CW 20M,
BP21 etc.) with the following characteristics (as an example): 50 m
x 0.32 mm x 0.25 µm.
5.1.2 Data acquisition system.
5.2 pH measuring apparatus
5.3 Magnetic stirrer
5.4 Balance with a precision of 0.1 mg.
5. 5 Oven which can be set to 60°C
5.6 Standard laboratory glassware including pipettes, screw-cap flasks, and
microsyringes.
6. Preparation of the sample
No specific preparation is necessary
7. Procedure
Place 50 ml of wine in a flask (5.6)
OIV-MA-AS315-21: R2010
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by GC after derivatization
Bring to pH 8 while stirring, with sodium hydroxide M (4.5)
Add 25 mg of 1,2-diaminobenzene (4.2)
Add 50 µl of 2,3-hexanedione (internal standard) at 2.0 g/L (4.10)
Close the flask using a screw-cap fitted with a Teflon-faced seal
Stir until the reagent has completely disappeared (5.3)
Place in the oven at 60°C for 3 h (5.5)
Cool.
7.1 Optimisation and analytical conditions (this study was carried out by
HPLC analysis, see this method)
The yield of the formation of derivatives of the dicarbonyl compounds with
the 1-2-diaminobenzene is optimal at pH 8 at 60°C after three hours of
reaction time
In addition, no interference of SO2 with the formation of quinoxalines was
noted during the study of the method.
7.2 Analysis by GC
7.2.1 Extraction of quinoxalines
- The reaction medium prepared in 7 is brought to pH 2 using H2SO4 2M
(4.6);
- Extract 2 times using 5 ml of dichloromethane (4.7) by magnetic stirring
for 5 minutes;
- Decant the lower phase each time;
- Mix the two solvent phases;
- Dry on approximately 1 g of anhydrous sodium sulphate (4.11);
- Decant.
7.2.2 Chromatographic analysis (given as an example)
- Detection. For the analysis by GC-MS, a Hewlett Packard HP 5890 gasphase chromatograph was coupled with Chemstation software and an HP
5970 mass spectrometer (electronic impact 70eV, 2.7 kV),
Note: It is also possible to use a nitrogen-specific detector
- Column. The column is a BP21 (SGE, 50 m x 0.32 mm x 0.25 µm).
- Temperatures. The temperature of the injector and the detector are
respectively 250°C and 280C; that of the oven is held at 60°C for 1min,
OIV-MA-AS315-21: R2010
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by GC after derivatization
then programmed to increase at a rate of 2C/min to 220C and the final
isothermal period lasts 20 min.
- Injection. The volume injected is 2 µl and the splitless time of the injector
valves is 30s.
7.2.3 Analysis of quinoxalines formed
- Separation. The chromatogram of the derivatives from a wine obtained
with 1,2-diaminobenzene, using selected-ion monitoring (SIM), is shown in
Figure 3. Good separations were obtained with all types of wines (white,
red, sweetened or fortified), and even with fermenting musts.
- Identification of the peaks. GC-MS was used to identify the dicarbonyl
compounds derivatized from the wine based on the total ion current method
(scan) which is used to obtain the mass spectra of derivatized quinoxalines
and to compare them with those recorded in the library; in addition, the
retention times were compared with those for pure compounds treated in the
same way. Table 1 shows the principal ions of the mass spectra for the
derivatized dicarbonyl compounds obtained.
- Determination. The quantitative determination of the dicarbonyl
compounds is performed with the SIM method, by selecting ions m/z = 76,
77, 103, 117, 130, 144, 158 and 171. The ions m/z = 76 and 77 are used for
the quantification and the others as qualifiers, i.e. glyoxal: ions m/z = 103
and 130, methylglyoxal: ions m/z = 117 and 144, diacetyl: ions m/z = 117
and 158, 2,3-pentandione: ions m/z = 171 and 2,3-hexanedione: ions m/z =
158 and 171.
7.2.4 Characteristics of the method
Some elements of internal validation were determined, but this is not a
formal validation according to the protocol governing the planning, the
implementing and the interpreting of the performance studies pertaining to
the analysis methods (OIV 6/2000)
- Repeatability. The repeatability of the GC-MS-SIM method shows
coefficients of variation ranging between 2 and 5% for the four dicarbonyl
compounds;
- Recovery rate. The quantities added to a wine were recovered with a
recovery rate ranging between 92 and 117%;
OIV-MA-AS315-21: R2010
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by GC after derivatization
- Linearity. Linear correlations were obtained in concentrations ranging
from 0.05 to 20 mg/l.
- Limit of detection. The limit of detection of most of the derivatized
dicarbonyl compounds using wine as a matrix is 0.05 mg/l
Table 1. Mass spectra (ion m/z and abundance of the ion in relation to that
of the base peak) of derivatives of dicarbonyl compounds using 1,2diaminobenzene
Dicarbonyl
Derivative
Mass spectrum
(principal ions and
abundance)
Glyoxal
Quinoxaline
130 (100), 103 (56.2),
76 (46.8), 50 (20.2),
75 (10.4), 131 (9.4)
Methylglyoxal
2-Methylquinoxaline
144 (100), 117 (77.8),
76 (40.5), 77 (23.3),
50 (21.9), 75 (11.3), 145
(10.3)
Diacetyl
2,3-Dimethylquinoxaline
117 (100), 158 (75.6),
76 (32.3), 77 (23.1),
50 (18.3), 75 (10.4)
2,3-Pentanedione
2-Ethyl-3-methylquinoxaline 171 (100), 172 (98), 130
(34.1), 75 (33.3),
77 (21), 50 (19.4), 144
(19), 143 (14.1),
103 (14)
2,3-Hexanedione
2,3-Diethylquinoxaline
compound
OIV-MA-AS315-21: R2010
158 (100), 171 (20.1),
76 (13.7), 77 (12.8),
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by GC after derivatization
159 (11.4), 157 (10.8),
50 (8.1)
Figure 3. Gas chromatogram of the extract from the dicarbonyl compounds
derivatized by 1,2-diaminobenzene from a white wine, detected by mass
spectrometry by selecting the ions m/z = 76, 77, 103, 117, 130, 131, 144, 158, 160
and 171. BP21 Column, 50m x 0.32mm x 0.25 µm oven temperature 60°C for
1min, then programmed increase of 2°C/min up to 220°C. Injector temperature:
250°C.
1. glyoxal; 2. methylglyoxal; 3. diacetyl; 4. 2,3-pentanedione; 5. 2,3-hexanedione
(internal standard); 6. phenylglyoxal (not studied with this method).
OIV-MA-AS315-21: R2010
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF
METHODS – OIV
-dicarbonyl compounds by GC after derivatization
Bibliography
Bartowski E.J. and Henschke P.A. The buttery attribute of wine – diacetyl
– desirability spoilage and beyond. Int. J. Food Microbiol. 96: 235-252
(2004).
Bednarski W., Jedrychowski L., Hammond E., and Nikolov L., A method
for determination of -dicarbonyl compounds. J. Dairy Sci. 72:2474-2477
(1989).
Leppannen O., Ronkainen P., Koivisto T. and Denslow J. A semiautomatic
method for the gas chromatographic determination of vicinal diketones in
alcoholic beverages. J. Inst. Brew. 85:278- 281 (1979).
Martineau B., Acree T. and Henick-Kling T., Effect of wine type on the
detection threshold for diacetyl. Food Res. Int. 28:139-143 (1995).
Moree-Testa P. and Saint-Jalm Y., Determination of -dicarbonyl
compounds in cigarette smoke. J. Chromatogr. 217:197-208 (1981).
De Revel G., Pripis-Nicolau L., Barbe J.-C. and Bertrand A., The detection
of α-dicarbonyl compounds in wine by the formation of quinoxaline
derivatives. J. Sci. Food Agric. 80:102-108 (2000).
De Revel G. and Bertrand A. Dicarbonyl compounds and their reduction
products in wine. Identification of wine aldehydes. Proc 7th Weurman
Flavour Research Symp., Zeist, June, pp 353-361 (1994).
De Revel G. and Bertrand A., A method for the detection of carbonyl
compounds in wine: glyoxal and methylglyoxal. J. Sci. Food Agric. 61:267272 (1993).
Voulgaropoulos A., Soilis T. and Andricopoulos N., Fluorimetric
determination of diacetyl in wines after condensation with 3,4diaminoanisole. Am. J. Enol. Vitic. 42:73-75 (1991).
Gilles de Revel et Alain Bertrand, Analyse des composés α-dicarbonyles
du vin après dérivation par le 1-2-diaminobenzène OIV FV 1275
OIV-MA-AS315-21: R2010
8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
CMC in white wines
Determination of carboxymethyl cellulose (cellulose gum,
CMC) in white wines
(OIV-Oeno 404-2010)
OIV-MA-AS315-22
Type of method: IV
1. Introduction
Carboxymethyl cellulose (CMC) is a polymer derived from natural cellulose that
has been routinely used for many years now as a food additive (INS 466) in
products such as ice creams and pre-cooked meals [1], to give them smoothness.
The use of CMC in white wines and sparkling wines to contribute to their tartaric
stabilisation [2] was recently accepted by the OIV in resolution Oeno 2/2008
provided that the dose added to the wine is less than 100 mg/l. A specific method
for determination of CMC in white wine has therefore been developed based on the
method of H.D Graham published in 1971 [3].
2. Field of application
The method applies to white wines (still and sparkling).
3. Principle
Once the CMC has been isolated from the wine by dialysis, it is hydrolysed
in an acid medium to form glycolic acid which is then degraded to form
formaldehyde. 2,7-Dihydroxynaphthalene (DHN) is added to form 2,2,7,7tetrahydroxydinaphthylmethane in the presence of formaldehyde. The
complex formed develops a purple-blue colour under the action of
concentrated sulphuric acid, at 100 °C, allowing colorimetric measurement
at 540nm (Figure 1).
OIV-MA-AS315-21: R2010
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
CMC in white wines
Figure 1: Mechanism of reaction of CMC with DHN in hot concentrated
sulphuric acid
(Feigl, 1966)
4. Reagents
 Sodium carboxymethylcellulose [N° CAS 9004-32-4] (21902 - average
viscosity 400-1000 mPa·s, substitution degree 0.60-0.95)
 2,7-Dihydroxynaphthalene [N° CAS 582-17-2] (purity > 98,0 % - HPLC)
 95 % concentrated sulphuric acid
 Purified water for laboratory use (example of quality: EN ISO 3696)
OIV-MA-AS315-21: R2010
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
CMC in white wines
5. Equipment




Laboratory glassware
Dialysis membrane (6000 to 8000 Da)
Temperature-controlled bath
Double-beam UV-visible spectrophotometer
6. Operating procedure
6.1 Preparation of the reagent
 Place 50 mg of DHN weighed to within 1 mg in a calibrated 100 mL phial.
 Add concentrated sulphuric acid up to the gauge line.
 Place the calibrated phial in a temperature-controlled bath at 28 °C for 4h
(without stirring).
 After heating, decant the reagent into a brown flask and store it in a refrigerator
at 4 °C.
6.2 Preparation of wine test specimens
 Insert 20 mL of wine, after degassing, into the dialysis membrane.
 Place the dialysis membrane containing the wine in a 6-litre flask filled with
distilled water.
 Leave to dialyse for 24h, changing the dialysis water twice.
6.3 Colour reaction
 Place 1 mL of dialysed wine into a test tube.
 Add 9 mL of reagent.
 Place the test tube in a temperature-controlled bath at 100 °C for 2h.
 Analyse the coloured solution by UV-visible spectrophotometer at 540nm and
read the absorbance value.
6.4 Calculation of the wine’s CMC content
 Recording the absorbance value read in point 6.3 on the calibration curve
obtained for a wine (see figure 2)
OIV-MA-AS315-21: R2010
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
CMC in white wines
7. Characteristics of the method
Certain elements of the internal validation were determined but these do not
constitute a formal validation according to the protocol governing the planning,
the implementing and the interpretation of performance studies pertaining to
analysis methods (OIV 6/2000)
7.1 Linearity of the response
A white wine has been added with incremental quantities of CMC ranging between
0 and 100 mg/L, then submitted to dialysis and treated in the conditions defined in
the procedure described above. The response is linear for the concentrations under
consideration (figure 2).
Figure 2: Linearity of CMC determination in white wine
7.2 Repeatability
The repeatability of the determination of CMC in white wines was defined on the
basis of the results achieved on 22 samples of wine that underwent 2 successive
analyses, so as to be analysed in identical conditions. The results are given in table
1.
OIV-MA-AS315-21: R2010
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
CMC in white wines
calculated values
Repeatability:
standard deviation
CV in %
r-limit
r-limit in %
0,075
7,2 %
0,21
20 %
Table 1: Repeatability of CMC determination in white wine
7.3 Reproducibility
The reproducibility of the determination of CMC in white wines was defined
through the analysis of a white wine by CMC, on 12 occasions at different dates.
The results are given in table 2.
calculated values
reproducibility
standard deviation
CV in %
R-limit
R-limit in %
0,082
9,6 %
0,23
27 %
Table 2: Reproducibility of CMC determination in white wine
7.4 Specificity
The specificity of CMC determination was verified by adding known
quantities of CMC into white wines. The recovery rates thus measure are
given in table 3.
OIV-MA-AS315-21: R2010
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
CMC in white wines
Sample
Wine 1
Wine 1
Wine 1
Wine 2
Wine 2
Wine 2
Wine 3
Wine 3
Wine 3
Wine 4
Wine 4
Wine 4
Added
concentration
(mg/l)
50
50
50
75
75
75
100
100
100
150
150
150
Resulting
concentration
(mg/l)
33
51
24
78
90
69
109
97
103
163
149
159
Recovery
rate
66 %
102 %
77 %
104 %
121 %
92 %
109 %
97 %
103 %
109 %
100 %
106 %
Table 3: Specificity of CMC determination in white wine
7.4 Detection and quantification limits
The detection limits (LD) and quantification limits (LQ) were calculated for
an untreated wine that underwent 10 analyses. The detection limit thus
determined is of 14 mg/l and the quantification limit is of 61 mg/l.
The method therefore enables to detect the adding of CMC into white wine
in quantities exceeding 20 mg/l and to quantify the addition when it exceeds
60 mg/l; this is not highly satisfactory but remains compatible with the
maximum authorised dose of 100 mg/l.
7.5 Uncertainty
The uncertainty was calculated at 3 different concentration levels (25, 75 and 150
mg/l) based on the analysis results for wines that have undergone CMC treatment,
OIV-MA-AS315-21: R2010
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
CMC in white wines
using the standard deviation reproducibility. The uncertainty thus obtained is of 40
mg/l, regardless of the CMC determination.
8. Bibliography
[1] Regulation (CE) N° 1333/2008 of the 16th of December, 2008
concerning food additives
[2] Stabilisation tartrique des vins par la carboxyméthylcellulose - Bulletin de
l’OIV 2001, vol 74, n°841-842, p151-159.
[3] Determination of carboxymethycellulose in food products - H.D Graham,
Journal of food science 1971, p 1052-1055
OIV-MA-AS315-21: R2010
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
Type of Method II
OIV-MA-AS315-23
MULTIELEMENTAL ANALYSIS USING ICP-MS
(OIV-Oeno 344-2010)
1.
SCOPE OF APPLICATION
This method can be applied to the analysis of the elements present in wines within
the range indicated and featured in the following list:
-
Aluminium between 0.25 and 5.0 mg/l
-
Boron between 10 and 40 mg/l
-
Bromine between 0.20 and 2.5 mg/l
-
Cadmium between 0.001 and 0.040 mg/l
-
Cobalt between 0.002 and 0.050 mg/l
-
Copper between 0.10 and 2.0 mg/l
-
Strontium between 0.30 and 1.0 mg/l
-
Iron between 0.80 and 5.0 mg/l
-
Lithium between 0.010 and 0.050 mg/l
-
Magnesium between 50 and 300 mg/l
-
Manganese between 0.50 and 1.5 mg/l
-
Nickel between 0.010 and 0.20 mg/l
-
Lead between 0.010 and 0.20 mg/l
-
Rubidium between 0.50 and 1.2 mg/l
-
Sodium between 5 and 30 mg/l
-
Vanadium between 0.003 and 0.20 mg/l
-
Zinc between 0.30 and 1.0 mg/l
OIV-MA-AS315-23: R2010
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
This technique can also be used to analyze other elements.
The sample sometimes requires mineralization. This is the case, for example, of
wines with more than 100 g/L of sugar where it can be necessary to realise
mineralization of the sample before. In this case, it is recommended to perform a
digestion with nitric acid in a microwave.
The technique can also be applied to musts, after mineralization.
2.
BASIS
Multielemental quantitative determination using Inductively Coupled Plasma Mass
Spectometry or ICP-MS.
Injection and nebulization of the sample in high-frequency plasma. The plasma
causes the desolvation, atomization and ionization of the elements in the sample.
The ions are extracted using a vacuum system fitted with ionic lenses. The ions are
separated according to the mass-to-charge ratio in a mass spectrometer, for
example, a quadrupole. Detection and quantification of ions using an electron
multiplier system.
3.
REAGENTS AND SOLUTIONS
3.1 Ultrapure, demineralized water with resistivity ( 18 MΩ), in accordance with
ISO 3696.
3.2 Certified solution(s) (for example, 100 mg/l) containing the metals to be
analyzed. Multielemental or monoelemental solutions can be used.
3.3 Indium and/or rhodium solution as an internal standard (normally 1 g/l).
3.4 Nitric acid ≥ 60 % (metal impurities  0.1 µg/l).
3.5 Argon, minimum purity of 99.999%.
3.6 Nitrogen (maximum impurity content: H2O  3 mg/l, O2  2 mg/l and
CnHm  0.5 mg/l).
Solution concentration and internal standards are given by way of reference.
Preparation of standard dissolutions:
OIV-MA-AS315-23: R2010
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
Acid concentration in the standards and in the final dilution of the wine samples
must be the same and must not exceed 5%. The following is an example.
3.7 Stock solution (5mg/l).
Place 0.5ml of solution (3.2) in a 10 ml (4.5) tube and add 0.1 ml of nitric acid
(3.4). Level off to 10 ml with demineralized water (3.1) and homogenize.
Shelf life: 1 month.
3.8 Internal standard solution (1 mg/l).
Using micropipettes (4.4), place 50 µl of indium or rhodium dissolution (3.3)
and 0.5 ml of nitric acid (3.4) in a 50 ml tube (4.6). Level off to 50 ml with
demineralized water (3.1) and homogenize.
Shelf life: 1 month.
3.9 Standard dissolutions of the calibration curve.
Adapt the range of the series of standards according to the dilution on the
sample or the equipment used.
Use 1000 µl and 100 µl pipettes (4.4).
Shelf life of standard dissolutions: 1 day
These standard solutions can also be prepared gravimetrically. Add internal
standard in the same concentration as for the samples.
3.10
4.
Internal control wine of known concentrations (MRC, MRE, MRI, etc.).
MATERIAL AND EQUIPMENT
4.1 Inductively coupled mass spectrometer with/without collision/reaction cell.
4.2 Computer with data processing software and printer.
4.3 Autosampler (optional).
4.4 1000 µl and 100 µl micropipettes.
4.5 10 ml plastic, graduated test tubes with bung or glass volumetric flasks.
4.6 50 ml plastic, graduated test tubes with bung or glass volumetric flasks.
OIV-MA-AS315-23: R2010
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
All volumetric material (micropipettes and test tubes) must be duly calibrated.
Note: material that will come into contact with the sample, such as, for example,
tubes and tips, must remain for at least 24 hours in a nitric acid dissolution (3.4) at
a concentration of 10% and must subsequently be rinsed several times in water
(3.1).
5.
SAMPLE PREPARATION
Samples of sparkling wine must be degasified. This can be done through nitrogen
bubbling (3.6) for 10 minutes or by using an ultrasound bath.
Remove the bung carefully to ensure that the wine is not contaminated. Wash the
bottle neck in an acid dissolution (2% HNO3). Wine samples are taken directly
from the bottle.
Use a micropipette (4.4) to insert 0.5 ml of wine, 0.1 to 0.5 ml of nitric acid (3.4)
and 100 µl of internal standard solution (3.8) into a 10 ml tube (3.5).Level off with
water (3.1) and homogenize.
For certain elements a higher dilution may be necessary owing to their high natural
content in the sample.
Br has high ionization potential and its ionization in plasma may be incomplete
because of the presence of high concentrations of other elements in wines with low
ionization potential. This may result in the incorrect quantification of Br and
therefore a 1/50 dilution is recommended to avoid this effect (in the event of
another dilution being used, confirm the results by checking recovery after an
addition).
When the standards are prepared gravimetrically, the final dilution of the sample
must also be obtained gravimetrically.
6.
PROCEDURE
Switch on the device (pump working and plasma on).
Clean the system for 20 minutes using 2% nitric acid (3.4).
Check that the device is functioning correctly.
OIV-MA-AS315-23: R2010
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
Analyze a blank and the series of standard dissolutions in increasing
concentrations, then a standard dissolution (e.g. no. 2 of series 3.9) to check for
correct calibration and finally the blank to ensure that there is no memory effect.
Read the samples in duplicate. For the internal control, use a wine of known
concentrations (3.10) to confirm that the results are coherent.
Element
m/z*
Aluminium
27
Boron
11
Bromine
79
Cadmium
114
Cobalt
59
Copper
63
Strontium
88
Iron
56/57
Lithium
7
Magnesium
24
Manganese
55
Nickel
60
Lead
average of 206, 207 and 208
Rubidium
85
Sodium
23
Vanadium
51
Zinc
64
* The above table is given by way of example. Other isotopes may be required,
depending on the equipment.
In the event of using equipment with no collision/reaction cell, correction equations
may be necessary for some elements.
7.
RESULTS
OIV-MA-AS315-23: R2010
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
The software can calculate the results directly.
Element concentrations are reported in mg/l to two decimal points.
Obtain, by interpolating in the calibration curve, the concentration of the elements
in the diluted samples. Use the following equation to calculate the concentration of
the elements in the sample:
C
Cm  Vt
Vm
Where:
C=
Concentration of the element in the sample
Cm = Concentration of the elements in the diluted sample
Vt = Final volume of the measurement solution, in ml
Vm = Aliquot volume of wine, in ml.
8.
QUALITY CONTROL
Ensure traceability by using certified standards.
In each analytical series, use a CRM (Certified Reference Material) as an internal
control of wine or a wine used as reference material from an interlaboratory test
campaign.
It is recommended that control graphs be created from the results of the quality
control analysis.
Participation in interlaboratory test campaigns.
9.
PRECISION
The results of the statistical parameters of the collaborative trial are shown in
Appendix A.
9.1 Repeatability (r)
OIV-MA-AS315-23: R2010
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
The difference between two independent results, obtained using the same method,
in the same sample, in the same laboratory, by the same operator, using the same
equipment in a short time interval. r results are given in Tables 1 to 17 of Appendix
A
9.2
Reproducibility (R)
The difference between two results, obtained using the same method, in the same
sample, in a different laboratory, by a different operator and with different
equipment. R results are given in Tables 1 to 17 of the Appendix A.
Table 1 represents the % of the relative standard deviation of Repeatability and
Reproducibility (RSDr% et RSDR%) of the method. (*) C = Concentration
Element
Concentration
RSDr %
RSDR %
Aluminium
0,25 – 5,0 mg/l
4
10
Boron
10 - 40 mg/l
3,8
6,3
0,20– 1,0 mg/l
≥ 1,0 – 2,5 mg/l
0,001 – 0,020 mg/l
≥ 0,020 – 0,040 mg/l
4,1
2.1
0,06 C*+0,18
1,5
16,3
8,0
10
10
Cobalt
0,002 – 0,050 mg/l
3,2
13,2
Copper
0,10 – 0,50 mg/l
≥ 0,50 – 2,0 mg/l
3,8
2,0
11,4
11,4
Strontium
0,30 – 1,0 mg/l
2,5
7,5
Iron
0,80– 1,0 mg/l
≥ 1,0-5,0 mg/l
4,2
4,2
15,7
7,8
Lithium
0,010 – 0,050 mg/l
7
12
Magnesium
50 - 300 mg/l
2
6
Manganese
0,50-1,5 mg/l
3
7
Nickel
0,010 – 0,20 mg/l
5
8
Lead
0,010 – 0,050 mg/l
≥ 0,050 – 0,20 mg/l
8
2
7
7
Rubidium
0,50 – 1,2 mg/l
3
6
Sodium
5 - 10 mg/l
≥ 10 - 30 mg/l
2
0,3 C*-2,5
10
10
Bromine
Cadmium
OIV-MA-AS315-23: R2010
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
Vanadium
0,003 – 0,010 mg/l
≥ 0,010 – 0,20 mg/l
8
3
10
10
Zinc
0,30 – 1,0 mg/l
5
12
Table 1: relative standard deviation of Repeatability and Reproducibility
10.
BIBLIOGRAPHY
ISO 5725:1994, Precision of test methods-Determination of repeatability and
reproducibility for a Standard test method by interlaboratory test.
ISO 17294:2004.
ALMEIDA M. R, VASCONCELOS T, BARBASTE M. y MEDINA B. (2002),
Anal. Bioanal Chem., 374, 314-322.
CASTIÑEIRA et al. (2001), Frenesius J. Anal. Chem., 370, 553-558.
DEL MAR CASTIÑEIRA GOMEZ et al. (2004), J. Agric Food Chem., 52, 29622974.
MARISA C., ALMEIDA M. et VASCONCELOS T. (2003), J. Agric. Food Chem.,
51, 3012-3023.
MARISA et al., (2003), J. Agric Food Chem., 51, 4788-4798.
PÉREZ-JORDAN M. Y., SOLDEVILLA J., SALVADOR A., PASTOR A y de la
GURDIA M. (1998), J. Anat. At. Spectrom., 13, 33-39.
PEREZ-TRUJILLO J.-P., BARBASTE M. y MEDINA B. (2003), Anal. Lett.,
36(3), 679-697.
TAYLOR et al. (2003), J. Agric Food Chem., 51, 856-860.
THIEL et al. (2004), Anal. Bioanal. Chem, 378, 1630-1636.
OIV-MA-AS315-23: R2010
8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
APPENDIX A
RESULTS OF THE COLLABORATIVE TRIALS
The method has been checked with two collaborative trials, by evaluating precision
in accordance with ISO 5725.The trueness of the method has been obtained
through recovery studies.
1st Collaborative Trial
8 samples (A, B, C, D, E, F, MH1 and MH2) were used from the following origins:
 Three samples of red wine, with and without addition.
 Three samples of white wine, with and without addition.
 Two samples of synthetic hydroalcoholic mixture, prepared with ethanol and
water.
Hydroalcoholic sample MH1 presented problems of instability during the trial and
the results have not been taken into account.
MH2
A
B
C
D
E
Metal
(mg/l)
Hydroalcoholic
mixture
RW2
RW3
WW2
WW3
Natural
red wine
Aluminium
5
0.5
2
2
1
Cadmium
0.001
0.005
0.02
0.05
0.01
Strontium
0.300
No
addition
No
addition
No
addition
No
addition
Lithium
0.020
0.01
0.02
0.04
0.01
Magnesium
50
100
200
50
25
Manganese
0.500
0.5
1
1
0.5
Nickel
0.070
0.025
0.2
0.1
0.1
Lead
0.010
0.05
0.1
0.15
0.05
Rubidium
1.0
No
addition
No
addition
No
addition
No
addition
Sodium
20
10
10
20
5
Vanadium
0.010
0.05
0.2
0.1
0.1
Zinc
0.500
0.1
1
0.5
0.5
OIV-MA-AS315-23: R2010
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
F
Natural
white
wine
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
addition
No
9
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
addition
OIV-MA-AS315-23: R2010
addition
10
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Multielemental analysis using ICP-MS
2nd Collaborative Trial
Sixteen samples (A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P) from the following
origins were used:
 Four samples of red wine, with and without addition.
 Four samples of Port wine, with and without addition.
 Six samples of white wine, with and without addition.
 Two samples of champagne.
Amounts added to the samples
Samples
Code
F-N
White wine
C-I
A-O
B-K
Liqueur wine
E-L
D-M
Red wine
H-J
Sparkling
wine
G-P
Addition
No
addition
Addition
1
Addition
2
No
addition
Addition
3
No
addition
Addition
4
No
addition
OIV-MA-AS315-23: R2010
B
mg/l
Co
µg/l
Cu
mg/l
Fe
Mg/l
0.0
0.0
0.0
0.0
5.0
5.0
5.0
1.0
10.0
10.0
1.0
2.0
0.0
0.0
0.0
0.0
15.0
20.0
1.5
3.0
0.0
0.0
0.0
0.0
20.0
50.0
2.0
5.0
0.0
0.0
0.0
0.0
11
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
PRECISION PARAMETERS (Tables 1 to 17):
The values of Horratr and HorratR have been obtained by using the Horwitz equation taking into account
Thompson’s modification for the concentration below 120 µg/L.
Table 1: Aluminium (mg/l)
SAMPLE
A
B
C
D
E
F
MH2
LAB.
No.
Accepted
11
11
11
11
11
11
11
10
9
9
10
10
10
8
Vréf
0,68
2,1
2,1
1,2
0,34
0,27
5,2
OIV-MA-AS315-23: R2010
Sr
0,020
0,043
0,032
0,041
0,014
0,006
0,26
r
0,06
0,12
0,09
0,12
0,04
0,02
0,73
RSD r
(%)
2,9
2,0
1,5
3,4
4,1
2,2
5,0
Horwitz
RSDr
(%)
11
9,4
9,5
10
12
13
8,2
Horratr
0,26
0,22
0,16
0,34
0,34
0,17
0,60
SR
0,077
0,21
0,21
0,10
0,029
0,028
0,56
R
0,22
0,61
0,59
0,29
0,08
0,08
1,6
RSDR
(%)
11
10
10
8,3
8,5
10
11
Horwitz
R
RSDR
(%)
17
14
14
16
19
20
13
HorratR
0,66
0,71
0,69
0,56
0,46
0,52
0,86
12
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 2: Boron (mg/l)
SAMPLE:
A-O
B-K
C-I
D-M
E-L
F-N
G-P
H-J
LAB.
No.
Accepted
8
8
8
8
8
8
7
8
6
4
4
7
5
5
4
5
Vréf
18
4,5
13
11
21
8,3
3,1
31
OIV-MA-AS315-23: R2010
Sr
0,77
0,27
0,31
0,26
0,47
0,43
0,094
1,0
r
2,2
0,76
0,89
0,74
1,3
1,2
0,27
3,0
RSD r
(%)
4,3
6,0
2,4
2,4
2,2
5,2
3,0
3,2
Horwitz
RSDr
(%)
6,8
8,4
7,2
7.4
6.7
7.7
8.9
6.3
Horratr
0,62
0,72
0,33
0,31
0,33
0,68
0,34
0,54
SR
0,94
0,40
0,33
1,1
0,85
0,47
0,18
1,6
R
2,69
1,14
0,94
3,11
2,43
1,34
0,51
4,43
RSDR
(%)
5,2
8,9
2,5
10
4,0
5,7
5,8
5,2
Horwitz
R
RSDR
(%)
10
13
11
11
10
12
14
9,6
HorratR
0,50
0,70
0,24
0,90
0,40
0,48
0,43
0,52
13
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 3: Bromine (mg/l
SAMPLE:
A-O
B-K
C-I
D-M
E-L
F-N
H-J
LAB.
No.
Accepted
6
5
6
6
6
6
6
2
2
3
4
3
3
2
Vref
1,21
0,19
0,81
0,38
1,72
0,22
2,30
OIV-MA-AS315-23: R2010
Sr
0,028
0,006
0,017
0,017
0,030
0,014
0,061
r
0,08
0,02
0,05
0,05
0,09
0,04
0,17
RSD r
(%)
2,3
2,9
2,1
4,5
1,7
6,4
2.7
Horwitz
RSDr
(%)
10,3
13,6
10,9
12,2
9,7
13,3
9,3
Horratr
0,22
0,21
0,19
0,37
0,17
0,48
0,28
SR
0,041
0,0043
0,062
0,066
0,22
0,046
0,092
R
0,12
0,012
0,18
0,19
0,62
0,13
0,26
RSDR
(%)
3,4
2,3
7,7
17,4
12,8
20,9
4
Horwitz
R
RSDR
(%)
15,6
20,5
16,5
18,5
14,8
20,1
14.1
HorratR
0,22
0,11
0,47
0,94
0,86
1
0.28
14
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 4: Cadmium (µg/l)
SAMPLE:
A
B
C
D
E
F
MH2
LAB.
No.
Accepted
12
12
12
12
8
8
9
11
11
9
10
7
6
5
Vréf
6
16
40
10
0,3
0,3
0,9
OIV-MA-AS315-23: R2010
Sr
0,2
0,4
0,4
0,3
0,20
0,04
0,08
r
0,6
1
1
0,8
0,6
0,1
0,2
RSD r
(%)
3,3
2,5
1,0
3,0
67
13
8,9
Horwitz
RSDr
(%)
15
15
15
15
15
15
15
Horratr
0,22
0,17
0,07
0,20
4,47
0,87
0,59
SR
1
2
3
0,9
0,20
0,20
0,10
R
3
6
8
3
0,67
0,45
0,29
RSDR
(%)
17
13
7,5
9,0
67
67
11
Horwitz
R
RSDR
(%)
22
22
22
22
22
22
22
HorratR
0,77
0,59
0,34
0,41
3,05
3,05
0,50
15
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 5: Cobalt (µg/l)
SAMPLE:
A-O
B-K
C-I
D-M
E-L
F-N
G-P
H-J
LAB.
No.
Accepted
10
10
10
10
10
10
9
10
6
6
8
3
8
7
5
6
Vréf
22
8
19
3
27
12
2
49
OIV-MA-AS315-23: R2010
Sr
0,5
0,3
0,4
0,07
1
0,5
0,2
0,5
r
1
0,9
1
0,2
3
2
0,5
1
RSD r
(%)
2,3
3,8
2,1
2,3
3,7
4,2
10
2,3
Horwitz
RSDr
(%)
15
15
15
15
15
15
15
15
Horratr
0,15
0,25
0,14
0,15
0,25
0,28
0,67
0,15
SR
2
1
3
0,1
3
1
0,3
6
R
6
4
7
0,3
9
4
0,8
18
RSDR
(%)
9,1
13
16
3,3
11
8,3
15
12
Horwitz
R
RSDR
(%)
22
22
22
22
22
22
22
22
HorratR
0,41
0,59
0,73
0,15
0,50
0,38
0,68
0,55
16
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 6: Copper (mg/l)
SAMPLE
A-O
B-K
C-I
D-M
E-L
F-N
G-P
H-J
LAB.
No.
Accepted
10
10
10
10
10
10
9
10
8
8
7
8
9
7
4
7
Vréf
1,1
0,21
0,74
0,14
1,7
0,16
0,042
2,1
OIV-MA-AS315-23: R2010
Sr
0,013
0,006
0,009
0,007
0,061
0,006
0,004
0,018
r
0,040
0,020
0,030
0,020
0,17
0,020
0,010
0,050
RSD r
(%)
1,2
2,9
1,2
5,0
3,6
3,8
9,5
0,86
Horwitz
RSDr
(%)
10
13
10
14
7,8
14
15
9,5
Horratr
0,12
0,22
0,12
0,36
0,5
0,27
0,63
0,09
SR
0,11
0,021
0,046
0,015
0,16
0,029
0,006
0,24
R
0,32
0,060
0,13
0,043
0,46
0,083
0,017
0,69
RSDR
(%)
10
10
6,2
11
9,0
18
14
11
Horwitz
R
RSDR
(%)
16
20
17
22
15
21
22
14
HorratR
0,63
0,50
0,36
0,50
0,60
0,86
0,64
0,79
17
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 7: Strontium (µg/l)
SAMPLE
LAB. Nº
Accepted
Vréf
Sr
r
RSD r
(%)
Horwitz
RSDr
(%)
10
HorwitzR
Horrat
Horratr
SR
R
RSDR
(%)
0,30
78
222
7,2
16
0,45
195
19
54
6,1
5,8
16
19
0,38
0,31
RSDR (%)
R
A
12
11
1091
33
93
B
12
8
1139
66
188
12
9
328
6
18
13
0,58
0,14
69
C
5,8
1,8
10
D
12
10
313
7
20
2,2
13
0,17
22
61
7,0
19
0,37
E
12
10
1176
28
80
2,4
10
0,24
86
243
7,3
16
0,46
F
MH2
12
10
293
3
9
1,0
13
0,08
22
62
7,5
19
0,39
12
9
352
7
19
2,0
12
0,17
24
69
6,8
19
0,36
OIV-MA-AS315-23: R2010
3,0
18
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 8: Iron (mg/l)
SAMPLE
A-O
B-K
C-I
D-M
E-L
F-N
G-P
H-J
LAB.
No.
Accepted
10
10
10
10
10
10
9
10
6
6
5
5
5
6
6
7
Vréf
3,2
1,5
2,1
3,1
4,3
1,1
0,83
7,8
OIV-MA-AS315-23: R2010
Sr
0,017
0,085
0,036
0,033
0,120
0,051
0,024
0,180
r
0,05
0,24
0,10
0,094
0,34
0,15
0,07
0,52
RSD r
(%)
0,53
5,7
1,7
1,1
2,8
4,6
2,9
2,3
Horwitz
RSDr
(%)
8,9
9,9
9,4
8,9
8,5
10
11
7,8
Horratr
0,06
0,58
0,18
0,12
0,33
0,46
0,26
0,29
SR
0,23
0,11
0,18
0,29
0,29
0,16
0,14
1,2
R
0,66
0,31
0,51
0,83
0,83
0,46
0,40
3,52
RSDR
(%)
7,2
7,3
8,6
9,4
6,7
15
17
15
Horwitz
R
RSDR
(%)
13
15
14
14
13
16
16
12
HorratR
0,55
0,49
0,61
0,67
0,52
0,94
1,06
1,25
19
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 9: Lithium (µg/l)
LAB.
No.
Accepted
A
11
10
34
2
5
B
11
11
42
3
8
C
11
11
47
1
4
D
11
11
18
1
E
11
11
25
F
MH2
11
9
11
7
SAMPLE
Vréf
Sr
r
RSD r
(%)
Horwitz
RSDr
Horratr
SR
R
HorwitzR
RSDR
Horrat
R
11
(%)
22
0,50
10
22
0,45
13
9,8
22
0,45
0,37
2
7
14
22
0,64
0,27
3
9
12
22
0,55
15
0,25
0,6
2
7,2
22
0,33
15
0,31
1
3
5,3
22
0,24
5,9
(%)
15
0,39
4
11
7,1
15
0,47
4
12
2,1
15
0,14
5
4
5,6
15
1
3
4,0
15
9
0,3
1
3,8
22
1
3
4,6
OIV-MA-AS315-23: R2010
RSDR
(%)
20
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 10: Magnesium (mg/l)
LAB
.
No.
Accepted
A
10
7
182
2,9
8,1
B
10
6
280
3,9
11
C
10
7
104
2,4
6,9
D
10
6
85
1,4
E
10
7
94
F
MH2
10
7
10
7
SAMPLE:
Vréf
Sr
r
RSD r
(%)
Horwitz
RSDr
Horratr
SR
R
Horwitz
R
RSDR
Horrat
R
5,1
(%)
7,3
0,70
2,1
6,9
0,30
6,5
8,0
0,81
6,1
2,6
8,2
0,32
5,5
16
5,9
8,1
0,73
0,27
3,8
11
5,9
8,5
0,69
0,31
2,4
6,9
4,7
8,9
0,53
1,6
(%)
4,3
0,37
9,3
26
1,4
4,5
0,31
6,0
17
2,3
5,3
0,43
6,8
19,25
4,0
1,7
5,4
0,31
2,2
2,2
6,2
2,3
5,3
0,43
65
0,95
2,7
1,5
5,6
51
0,90
2,5
1,8
5,8
OIV-MA-AS315-23: R2010
RSDR
(%)
21
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 11: Manganese (mg/l)
LAB.
No.
Accepted
A
11
10
1,3
0,014
0,040
B
11
9
1,8
0,14
0,40
C
11
8
1,5
0,028
0,080
D
11
8
1,0
0,035
0,10
E
11
9
0,84
0,019
0,050
F
MH2
11
9
0,59
0,015
11
8
0,52
0,029
SAMPLE
Vréf
OIV-MA-AS315-23: R2010
Sr
r
RSD r
(%)
Horwitz
RSDr
Horratr
SR
R
RSDR
(%)
HorwitzR
RSDR
Horrat
R
10
(%)
15
0,67
11
15
0,73
5,6
15
0,37
4,9
16
0,31
6,8
16
0,43
0,090
5,3
17
0,31
0,10
7,1
18
0,39
1,1
(%)
10
0,11
0,13
0,37
7,8
9,7
0,80
0,20
0,56
1,9
9,9
0,19
0,084
0,24
3,5
11
0,32
0,049
0,14
2,3
11
0,21
0,057
0,16
0,040
2,5
11
0,23
0,031
0,080
5,6
12
0,47
0,037
22
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 12: Nickel (µg/l)
Horwitz
LAB.
No.
Accepted
A
11
10
40
2
6
5,0
(%)
15
0,33
5
13,90
B
12
10
194
7
20
3,6
14
0,26
17
48,96
C
12
8
148
4
10
2,7
14
0,19
5
15,12
D
12
8
157
4
12
2,6
14
0,19
8
E
11
8
15
0,6
2
4,0
15
0,27
F
MH2
12
9
66
1
4
1,5
15
11
7
71
5
14
7,0
15
SAMPLE
Vréf
OIV-MA-AS315-23: R2010
Sr
r
RSD r
(%)
RSDr
Horratr
SR
R
RSDR
(%)
HorwitzR
RSDR
Horrat
R
13
(%)
22
0,59
8,8
21
0,42
3,4
21
0,16
23,10
5,1
21
0,24
1
3,33
6,7
22
0,30
0,10
4
10,58
6,1
22
0,28
0,47
4
11,41
5,6
22
0,25
23
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 13: Lead (µg/l)
SAMPLE
LAB.
No.
Accepted
Vréf
Sr
r
RSD r
(%)
Horwitz
RSDr
Horratr
SR
R
RSDR
(%)
HorwitzR
RSDR
Horrat
R
A
12
9
59
1
4
1,7
(%)
15
0,11
3
9
5,1
(%)
22
B
12
10
109
2
6
1,8
15
0,12
8
23
7,3
22
0,33
C
12
9
136
3
9
2,2
14
0,16
13
37
9,6
22
0,44
D
12
9
119
2
6
1,7
15
0,11
5
13
4,2
22
0,19
E
12
10
13
1
3
7,7
15
0,51
1
4
7,7
22
0,35
F
MH2
12
9
92
1
4
1,1
15
0,07
4
11
4,4
22
0,20
12
10
13
1
3
7,7
15
0,51
1
3
7,7
22
0,35
OIV-MA-AS315-23: R2010
0,23
24
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 14: Rubidium (µg/l)
LAB.
No.
Accepted
A
11
6
717
14
41
B
11
7
799
25
70
C
11
8
677
10
27
D
11
7
612
18
E
11
9
741
F
MH2
11
9
11
7
SAMPLE
Vréf
Sr
r
RSD r
(%)
Horwitz
RSDr
Horratr
SR
R
HorwitzR
RSDR
Horrat
R
1,8
(%)
17
0,11
3,8
17
0,22
96
5,0
17
0,29
0,26
18
50
2,9
17
0,17
0,24
66
187
8,9
17
0,52
11
0,15
43
123
7,0
17
0,41
10
0,09
64
181
5,7
16
0,36
2,0
(%)
11
0,18
13
36
3,1
11
0,28
30
86
1,5
11
0,14
34
51
2,9
11
19
53
2,6
11
617
10
28
1,6
1128
10
28
0,89
OIV-MA-AS315-23: R2010
RSDR
(%)
25
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 15: Sodium (mg/l)
LAB.
No.
Accepted
A
10
9
19
0,59
1,7
B
10
9
20
1,3
3,6
C
10
7
28
0,33
D
10
8
11
E
10
8
9,8
F
MH2
10
8
10
8
SAMPLE
Vréf
Sr
r
RSD r
(%)
Horwitz
RSDr
Horratr
SR
R
RSDR
(%)
HorwitzR
RSDR
Horrat
R
3,1
(%)
6,8
0,46
2,2
5,7
12
(%)
10
6,5
6,7
0,97
2,2
6,3
11
10
1,10
0,93
1,2
6,4
0,19
1,9
5,4
6,8
9,7
0,70
0,24
0,68
2,2
7,4
0,30
1,1
3,0
10
11
0,91
0,19
0,53
1,9
7,5
0,25
0,89
2,5
9,1
11
0,83
6,1
0,093
0,26
1,5
8,1
0,19
0,74
2,1
12
12
1,00
24
1,8
5,0
7,5
6,6
1,14
2,6
7,2
11
9,9
1,11
OIV-MA-AS315-23: R2010
1,20
26
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 16: Vanadium (µg/l)
LAB.
No.
Accepted
A
12
11
46
1
3
B
12
11
167
5
15
C
12
11
93
3
8
D
12
9
96
3
E
10
7
3
F
MH2
10
8
12
9
SAMPLE
Vréf
Sr
r
RSD r
(%)
Horwitz
RSDr
Horratr
SR
R
HorwitzR
RSDR
Horrat
R
11
(%)
22
0,50
11
21
0,52
13
22
0,59
22
8,3
22
0,38
0,3
0,9
10
22
0,45
0,45
0,2
0,7
6,7
22
0,30
0,18
0,9
3
8,2
22
0,37
2,2
(%)
15
0,15
5
13
3,0
14
0,21
19
54
3,2
15
0,21
12
33
8
3,1
15
0,21
8
0,2
0,7
6,7
15
0,45
3
0,2
0,6
6,7
15
11
0,3
1
2,7
15
OIV-MA-AS315-23: R2010
RSDR
(%)
27
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS – OIV
Multielemental analysis using ICP-MS
Table 17: Zinc (µg/l)
LAB.
No.
Accepted
A
11
8
405
22
61
B
11
9
1327
49
138
C
11
9
990
14
41
D
11
9
1002
28
79
E
11
9
328
13
37
F
MH2
11
9
539
15
11
8
604
72
SAMPLE
Vréf
OIV-MA-AS315-23: R2010
Sr
r
RSD r
(%)
Horwitz
RSDr
Horratr
SR
R
RSDR
(%)
HorwitzR
RSDR
Horrat
R
11
(%)
18
0,61
11
15
0,73
8,7
16
0,54
11
16
0,69
24
19
1,26
172
11
18
0,61
251
15
17
0,88
5,4
(%)
12
0,45
45
128
3,7
10
0,37
152
429
1,4
11
0,13
86
243
2,8
11
0,25
110
310
4,0
13
0,31
79
224
42
2,8
12
0,23
61
204
12
11
1,09
89
28
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total bromide
Method OIV-MA-AS321-01
Type IV method
Total Bromide
1. Principle
The wine is ashed at 525 oC in presence of an excess of soda lime. A solution of the
residue (at pH 4.65) is treated with chloramine T to liberate bromide. The bromide
is reacted with phenolsulfonephthalein to form phenoltetra-bromophthalein-3’-3’’disulfonic acid, which is determined by spectrophotometer at 590 nm.
2. Apparatus
2.1 Boiling water-bath 100C
2.2 Temperature-controlled electric furnace
2.3 Spectrophotometer capable of measuring absorbance at wavelengths between
300 and 700 nm
3. Reagents
3.1 Sodium hydroxide solution, 50% (m/m)
3.2 Calcium hydroxide suspension containing 120 g of CaO per liter
3.3 Phenolsulfonephthalein solution:
0.24 g of phenolsulfonephthalein (phenol red) are dissolved in 24 mL sodium
hydroxide solution, 0.1 M, and made up to the liter with distilled water.
3.4 pH 4.65 buffer solution:
Acetic acid, 2 M ..................................……………………….
500 mL
Sodium hydroxide, 2 M ......................……………….........
250 mL
Distilled water to ..........................…………………...........….
1L
3.5 Oxidizing solution:
Chloramine T .....................................……………………….…
2g
Distilled water to ...…………………...........................……..…
1L
Prepare this solution 48 hours before use
Storage: two weeks at  4 oC
3.6 Reducing solution:
Sodium thiosulfate ……………………………………………….……… 25 g/L.
Distilled water to .............................……………………......….
1L
3.7 Sulfuric acid, 10%(v/v): sulfuric acid (20 = 1.84 g/mL) diluted 1/10.
3.8 Sulfuric acid, 1%(v/v): sulfuric acid (20 = 1.84 g/mL) diluted 1/100.
3.9 Potassium bromide solution corresponding to 1 g of bromide per liter. 1.489g
of potassium bromide, KBr, is dissolved in distilled water and made up to one
liter.
OIV-MA-AS321-01 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total bromide
4. Procedure
4.1 How to obtain ash and ash solution
Place 50 mL of wine in a silica dish of 7 cm diameter, add 0.5 mL 50% sodium
hydroxide solution, (3.1), and 1 mL calcium hydroxide suspension (3.2).
Check that the pH is at least pH 10. Leave the dish covered with a watch glass
for 24 hours. Evaporate the liquid until dry on a boiling water bath. To
accelerate the evaporation, a hot air current can be used in the final stages.
Ash as follows: place the dish 30 minutes in a furnace (2.2) at 525C . After
cooling, mix the residue with a little distilled water. Evaporate on the boiling
water-bath. Ash again at 525C. Repeat the operation until the ash is
gray/white.
Mix the residue with 5 mL boiling distilled water. Add using a burette: first
10% sulfuric acid (3.7), then sufficient 1% sulfuric acid (3.8) to bring the pH
to between 4 and 5 as measured by indicator paper. Let X mL = the volume
added of sulfuric acid (3.7 & 3.8). Add 10.2-(X+5) mL of distilled water.
Crush the precipitated calcium sulfate with a glass rod. Transfer the content of
the dish to a centrifugation tube. Centrifuge for 10 min. Place 8 to 9 mL of
the clear supernatant into a test tube.
4.2 Qualitative test
This test is performed to determine if the bromide content of the wine is
between 0 and 1 mg/L, which would enable the determination to be performed
on the undiluted ash solution.
Place in a small test tube:
- 1 mL of ash solution
- 1 drop of pH 4.65 buffer solution
- 1 drop of phenolsulfonephthalein solution
- 1 drop of chloramine T solution
After exactly 1 minute, stop the reaction by adding 1 drop of sodium
thiosulfate solution.
If the coloration obtained is yellow, brownish yellow or greenish yellow, the
ash solution can be used undiluted.
If the obtained coloration is blue, purple or violet, the wine contains more than
1 mg of bromide per liter and the ash solution must be diluted 1/12 or 1/5 until
the coloration obtained corresponds to the conditions above.
4.3 Quantitative method
Place in a test tube:
- 5 mL of ash solution, diluted or undiluted, add:
- 0.25 mL of pH 4.65 buffer solution
- 0.25 mL of phenolsulfonephthalein solution
- 0.25 mL T chloramine solution
OIV-MA-AS321-01 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total bromide
Wait exactly 1 minute and add:
- 0.25 mL of sodium thiosulfate
Measure using a spectrophotometer set at 590 nm with a 1 cm cell, the
difference in absorbance between the sample and the blank obtained by adding
the same quantities of reagents to 5 mL of distilled water.
Note: When the bromide content is low (yellow coloration, slightly greenish)
determine the absorbance in a cell of 2 cm optical path.
4.4 Preparation of the calibration curve
At the time of use, prepare a solution containing 10 mg of bromine per liter by
making 2 successive dilutions (1/10) of standard potassium bromide solution,
1 g/L.
In a set of 8 test tubes, place 0.25, 0.50, 0.75, 1.00, 1.25, 1.50, 2.00 and 2.50
mL respectively of bromide standard, 1g/L (3.9) and make up to 5 mL with
distilled water. (The solutions are equivalent to 0.10, 0.20, 0.30, 0.40, 0.50,
0.60, 0.80 and 1 mg of bromine per liter of wine without dilution of the ash
solution). Continue as in 4.3 using the calibration solutions instead of the ash
solution. Determine the absorbance of these solutions and a blank, as in 4.3,
using 5 mL of distilled water in the blank solution. The absorbance obtained
corresponding to the bromide concentration is plotted on a line that curves
slightly towards the origin.
5. Expression of results
5.1 Calculations
The bromide content in wine is obtained by plotting on the calibration curve,
the net absorbance of the ash solution (taking into account the thickness of the
cell used and any dilution of the ash solution) and interpolating the bromide
concentration. The total bromide content is expressed in milligrams per liter
(mg/L) to two decimal places.
BIBLIOGRAPHY
DAMIENS A., Bull. Sci. Pharmacologiques, 1920, 27, 609; Ibid, 1921, 28, 37, 85
et 105.
BALANTRE P., J. Pharm. Chem., 1936, 24, 409.
PERRONET M., ROCQUES Mme S., Ann. Fals. Fraudes, 1952, 45, 347.
CABANIS J.-C., Le brome dans les vins, Thèse doct. Pharm., Montpellier, 1962.
JAULMES P., BRUN Mme S., Cabanis J.-C., Chim anal., 1962, 327.
STELLA C., Riv. Viticolt. Enol., Conegliano, 1967, 5.
OIV-MA-AS321-01 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Chloride
Method OIV-MA-AS321-02
Type II method
Chloride
1. Principle
Chloride is determined directly in the wine by potentiometry using an Ag/AgCl
electrode.
2. Apparatus
2.1 pH/mV meter graduated at intervals of at least 2 mV.
2.2 Magnetic stirrer.
2.3 Ag/AgCl electrode with a saturated solution of potassium nitrate as
electrolyte.
2.4 Microburette graduated in 0.01 mL.
2.5 Chronometer.
3. Reagents
3.1 Standard chloride solution: 2.1027 g of potassium chloride, KCl (max.
0.005% Br), dried before use, by leaving in a desiccator for several days, is
dissolved in distilled water and made up to one liter. 1 mL of this solution
contains 1 mg Cl-.
3.2 Silver nitrate solution: 4.7912 g of analytical grade silver nitrate, AgNO 3, is
dissolved in ethanol solution, 10% (v/v) and made up to one liter. 1 mL of this
solution corresponds to 1 mg Cl-.
3.3 Nitric acid, not less than 65% (20 = 1.40 g/mL).
4. Procedure
4.1 Place 5.0 mL of standard chloride solution (3.1) into a 150 mL cylindrical
vessel placed on a magnetic stirrer (2.2), dilute with distilled water to
approximately 100 mL and acidify with 1.0 mL of nitric acid (3.3). After
immersing the electrode, add silver nitrate solution (3.2) with the
microburette, with moderate stirring using the following procedure: begin by
adding the first 4 mL in 1 mL fractions and read the corresponding millivolt
values. Add the next 2 mL in fractions of 0.20 mL. Finally, continue the
addition in fractions of 1 mL until a total of 10 mL has been added. After
each addition, wait for approximately 30 sec before reading the corresponding
millivolt value. Plot the values obtained on a graph against the corresponding
milliliters of titrant and determine the potential corresponding to the
equivalence point.
OIV-MA-AS321-02 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Chloride
4.2 Place 5 mL of the standard chloride solution (3.1) in a 150 mL cylindrical
vessel with 95 mL of distilled water and 1 mL of nitric acid (3.3). Immerse
the electrode and titrate, while stirring, until the potential of the equivalence
point is obtained. This determination is repeated until a good degree of
agreement in the results is obtained. This check must be carried out before
each series of measurements of chloride in the samples.
4.3 Place 50 mL of wine into a 150 mL cylindrical vessel. Add 50 mL of distilled
water and 1 mL of nitric acid (3.3) and titrate using the procedure described in
4.2.
5 Expression of results
5.1 Calculations
If n represents the number of milliliter of silver nitrate titrant, the chloride
content in the tested liquid, is given by:
20  n expressed as milligrams Cl per liter
0.5633  n expressed as milliequivalents per liter,
32.9 x n expressed as milligrams of NaCl per liter.
5.2 Repeatability (r):
r = 1.2 mg Cl/L
r = 0.03 mEq/L
r = 2.0 mg NaCl/L
5.3 Reproducibility (R)
R= 4.1 mg/L
R= 0.12 mEq/L
R= 6.8 mg NaCl/L
6. Note: For very precise determination.
Refer to the complete titration curve obtained during determination of the test
liquid (4.2).
a) Measure 50 mL of the wine to be analyzed into a 150 mL cylindrical vessel.
Add 50 mL of distilled water and 1 mL of nitric acid (3.3). Titrate using silver
nitrate solution (3.2), adding 0.5 mL at a time and recording the corresponding
potential in millivolts. Estimate from this first titration the approximate volume
of silver nitrate solution (3.2) required.
b) Repeat the determination adding 0.5 mL of titrant at a time until the volume
added is 1.5 to 2 mL less that the volume determined in (a). Thereafter add 0.2
mL at a time. Continue to add the solution beyond the estimated equivalence
point in a symmetrical manner, i.e. by adding 0.2 mL and then 0.5 mL at a time.
The end point of the measurement and the exact volume of silver nitrate consumed
are obtained:
— either by drawing the curve and determining the equivalence point;
— or by the following calculation:
OIV-MA-AS321-02 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Chloride
V = V’ + Vi
  E1
  E1    E 2
Where:
V = volume of titrant at the equivalence point;
V’ = volume of titrant before the largest potential change;
Vi = constant volume of the increments of titrant, i.e. 0.2 mL;
E1 = second difference in potential before the largest potential change;
E2 = second difference in potential after the largest potential change.
Example:
Volume of AgNO3
titrating solution
E potential in
mV
Difference
E
Second difference
E
0
204
208
212
218
224
230
238
250
272
316
350
376
396
4
4
6
6
6
8
12
22
44
34
26
20
0
2
0
0
2
4
10
22
10
8
6
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
In this example, the end point of the titration is between 1.6 and 1.8 mL: the
largest potential change ( E = 44 mV) occurs in this interval. The volume of
silver nitrate titrant consumed to measure the chlorides in the test sample is:
V  1. 6  0. 2
22  1. 74mL
22  10
BIBLIOGRAPHY
MIRANDA PATO C. de, F.V., O.I.V., 1959, no12.
HUBACH C.E., J. Ass. Off. Agric. Chem., 1966, 49, 498.
Fédération internationale des Producteurs de jus de fruits, F. O.I.V.,1968, no37.
JUNGE Ch., F.V., O.I.V., 1973, no440.
OIV-MA-AS321-02 : R2009
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
Method OIV-MA-AS321-03
Type II method
Determination of fluoride content in wine using a fluoride
selective ion electrode, and a standard addition method
(Resolution Oeno 22/2004)
1.
SCOPE
This method is applicable to the analysis of fluoride in all wines. With
proper dilution, the range of detection is 0.1 mg/l to 10.0 mg/l.
2.
PRINCIPLE
The concentration of fluoride in the sample is measured after addition of a
buffer, using a fluoride ion selective electrode. The buffer provides a high,
constant background ionic strength; complexes iron and aluminium (which
would otherwise complex with fluoride); and adjusts the pH to a level that
minimises the formation of a HF•HF complex. The matrix effects are then
minimised using standard addition.
3.
REAGENTS
3.1 Deionized or distilled water
3.2 Sodium chloride  99.0% purity
3.3 Trisodic citrate  99.0% purity
3.4 CDTA (1,2-diaminocyclohexane-N,N,N’,N’- tetracetic hydrate acid) 
98.0% purity3.5 Sodium hydroxide  to 98.0% purity
3.6 Sodium hydroxide solution 32% (w/v) made from 3.5
3.7 Glacial acetic acid  99.0% purity
3.8 Sodium fluoride  99.0% purity
OIV-MA-AS321-03 : R2004
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
3.9 Commercial Total Ionic Strength Adjustment Buffer (TISAB) (i.e. IIIOrion Research Inc. Cat. # 940911) or equivalent (See 4.2).
OIV-MA-AS321-03 : R2004
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
3.10 Alternative TISAB:
3.10.1 To ca. 700 ml water (3.1) in a 1 l beaker (4.3), add 58.0 g ± 0.1
g sodium chloride (3.2) and 29.4 g ± 0.1 g of tri-sodium citrate
(3.3).
3.10.2 Dissolve 10.0 g ± 0.1 g of CDTA (1,2-diaminocyclohexaneN,N,N’,N’-tetraacetic acid) (3.4) and 6 ml of 32% (w/v) sodium
hydroxide (3.6) in approximately 50 ml of distilled water. (3.1)
3.10.3 Mix the two solutions together then add 57 ml of glacial acetic
acid (3.7) and adjust pH to 5.5 with 32% (m/v) sodium hydroxide
(3.6). Cool to room temperature, transfer to 1 l volumetric flask
(4.10), and dilute to volume with water (3.1).
3.11 Fluoride standard solutions
3.11.1 Fluoride stock standard solution (100 mg/l):
Weigh 221 mg ± 1 mg of sodium fluoride (3.8) (dried at 105°C for 4
hours) into a 1 l polyethylene volumetric flask (4.10) and make to
volume with water. (3.1)
3.11.2 Fluoride calibration standards at 1.0 mg/l, 2.0 mg/l and 5.0 mg/l
: make
1.0 mg/l, 2.0 mg/l, and 5.0 mg/l calibration standards by
pipetting 1 ml, 2 ml, and 5 ml of the 100 mg/l stock standard (3.11.1)
into three polyethylene 100 ml volumetric flasks (4.10) respectively
and diluting to volume with water (3.1).
3.12 Wine blank : a wine known to be fluoride free is used as a matrix blank
3.13 1 mg/l spiked wine standard - Place 10 ml (4.11) of 100 mg/l fluoride stock
standard solution(3.11.1)into a 1 l volumetric flask (4.10) and
bring to volume with fluoride free wine (3.12).
OIV-MA-AS321-03 : R2004
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
4.
APPARATUS
4.1
pH/ion analyser with standard addition capability (e.g. Corning pH/ion
Analyser 455, Cat. # 475344) or pH/ion analyser with extended mV range.
4.2
Fluoride ion selective electrode and single junction reference electrode or
combination electrode (e.g., Corning Fluoride Electrode Cat. # 34108490).
4.3
Beakers - 150 ml, 1 l, polyethylene
4.4
Cylinder - 50 ml graduated, polyethylene,.pouring.
4.5
Magnetic stirrer
4.6
Magnetic stir bars, PTFE coated.
4.7
Plastic bottles with caps, 125 ml (Nalgene or equivalent)
4.8
Precision pipette, 500 µl
4.9
Ultrasonic bath
4.10
Volumetric flasks, Class A, 50 ml, 100 ml, and 1 l
4.11
Volumetric pipettes, Class A, 1 ml, 2ml, 5 ml, 10 ml, 20 ml, and 25 ml
5. PREPARATION OF CALIBRATION STANDARDS
5.1 Place 25 ml (4.11) of 1.0 1.0 mg/l, 2.0 mg/l, and 5.0 mg/l standard
solutions (3.11.2) respectively into three 150 ml beakers (4.3), add 20 ml
(4.11) of water (3.1) and (4.11) 5 ml of commercial TISAB (3.9) to each. Mix
with a magnetic stirring. (4.5 and 4.6).
5.2 If using alternative TISAB reagent (3.10) : place 25 ml (4.11) of each
standard solution (3.11.2) into three 150 ml beakers (4.3) and add 25 ml
(4.11) of alternative TISAB reagent (3.10) to each. Mix with a magnetic
stirrer. (4.5 and 4.6)
OIV-MA-AS321-03 : R2004
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
6. PREPARATION OF THE TEST SAMPLES
Mix the wine sample thoroughly before sampling. Sparkling wines should be
degassed before sampling by transferring to a clean beaker and placing in an
ultrasonic bath (4.9) until gas no longer evolves.
6.1
If using reagent (3.9), commercial TISAB : place 25 ml (4.11) of wine
sample into a 150 ml beaker (4.3) with 20 ml (4.11) of water (3.1) and add 5 ml
(4.11) of ommercial TISAB (3.9) solution. Mix with a magnetic stirrer (4.5 and
4.6). Dilution factor (DF) = 1.
6.2
If using alternative TISAB reagent (3.10) : place 25 ml (4.11) of wine
sample in a 150 ml beaker (4.3) and add 25 ml (4.11) of alternative TISAB
reagent (3.10). Mix with a magnetic stirrer (4.5 and 4.6). Dilution factor (DF) = 1.
7. PROCEDURE
Measurement (all standard and wine sample solutions must be at the same
temperature).
7.1
Calibration standards
Measure the potential of each of the calibration solutions, using the meter
(4.1), fluoride selective electrode (4.2), and reference electrode (4.2). The
final reading must be taken when the readings have stabilised (stability is
obtained when the potential varies by not more than 0.2 to 0.3 mV/ 3
minutes). Record the readings for each of the calibration standards.
The log10 of each of the standard concentrations versus the millivolt
reading measured for each standard concentration is plotted on graph paper
in order to determine the slope of the electrode.
7.2
Wine samples
Measure and record the potential expressed in mV (E1) of the sample (6.1
or 6.2) after the readings have stabilised. Add 500 µl (4.8) of 100 mg/l
fluoride standard (3.11.1) to the sample (6.1 or 6.2). After the readings
have stabilised, read and record the potential expressed in mV (E2) of the
wine solution.
OIV-MA-AS321-03 : R2004
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
The final concentration must be at least double the fluoride concentration
in the sample solution. To make sure, if the fluoride concentration in the
test sample is above 2 mg/l on the first determination, a second
determination must be made after dilution of the sample as follows (7.2.1
or 7.2.2).
7.2.1 When using the commercial TISAB buffer (3.9): pipette (4.11) 25 ml
of wine sample in a 50 ml volumetric flask (4.10) and bring to volume with
water. Take 25 ml (4.11) of this diluted wine in a 150 ml cylindrical
beaker (4.3) and add 25 ml of commercial TISAB (3.9). Mix with a
magnetic stirrer (4.5 and 4.6) and then proceed with measurement as in
7.02. Dilution factor (DF) = 2.
7.2.2 When using the alternative TISAB buffer (3.9): pipette (4.11) 25 ml
of wine sample in a 50 ml volumetric flask (4.10) and bring to volume with
water. Pour 25 ml (4.11) of this diluted wine in a 150 ml cylindrical
beaker (4.3) and add 25 ml of alternative TISAB buffer (3.10). Mix with a
magnetic stirrer (4.5 and 4.6) and then proceed with measurement as in
7.2. Dilution factor: (DF) = 2.
8
CALCULATION
The fluoride content of the sample solution expressed in mg/l is obtained
by using the following formula:
Cf 
Va  Ca
1

Vo
((anti log E / S )  1)
If the added standard solution V std is ‹ 1% of the volume of the
solution after the addition, so Va = Vo and
Cf  DF  Ca 
1
((anti log E / S )  1)
Cf = fluoride concentration of the sample solution (mg/l)
DF = dilution factor. If it is necessary to dilute the sample as in
(7.2.1) or in (7.2.2), use the identical values for the dilution and the
sample. That is to say, DF = 2 for a diluted sample (7.2.1) and (7.2.2) or
DF = 1 if it is not as in (6.1) or (6.2)
OIV-MA-AS321-03 : R2004
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
Vo = initial volume of the sample solution before standard addition (ml)
Va = volume of the solution after standard addition (ml)
E = difference between potentials E1 and E2 obtained in (7.2) in mV.
S = slope of the calibration curve of the electrode.
Ca =
Vstd  C std
Vsamp
where
Ca = concentration (in mg/l) of fluoride added to the sample volume (Vo ) obtained
by multiplying the standard volume (3.11.1) added to the solution (Vstd) by the
concentration (Cstd) of standard (3.11.1) and divided by the sample volume (25 ml)
using (6.1) or (6.2)
Vstd = volume added standard (3.11.1) (0.5 ml)
Vsamp = sample volume used in (6.1) or (6.2), Vsamp = 25 ml
Cstd = standard concentration (3.11.1)
Calculation example:
(1) for a sample prepared as in (6.2) and measured as in (7.2)
DF = 1
Ca =
Vstd  C std
0.5 ml  100 mg / l
=
= 2 mg /l
Vsamp
25 ml
E = 19.6 mV
S = -58.342
1
((anti log E / S )  1)
1
C f  1  2 mg / l 
((anti log 19.6 / 58.342)  1)
C f  1 2 mg / l  0.856 = 1.71 mg / l of fluoride
Cf  DF  Ca 
OIV-MA-AS321-03 : R2004
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
(2) for a sample prepared as in (7.2.2), and measured as in (7.2)
DF = 2
Ca =
Vstd  C std
0.5 ml  100 mg / l
=
= 2 mg /l
Vsamp
25 ml
E = 20.4 mV
S = -55.937
1
((anti log E / S )  1)
1
C f  2  2 mg / L 
((anti log 20.4 / 55.937)  1)
Cf  DF  Ca 
Cf = 2 x 2 mg / l x 0.760 = 3.04 mg / l of fluoride
9. PRECISION
The details of inter laboratory study are given in Annex B. the Horrat (HoR) ranges
from 0.30 to 0.97 and indicates a very good reproducibility among participants.
The results of the statistical calculations are given in Annex B table 2.
The standard deviation of repeatability (RDSr) ranges from 1.94% to 4.88%. The
standard deviation of reproducibility (RDSR) ranges from 4.15% to 18.40%.
Average % recovery ranged between 99.8% and 100.3% of the mean target.
10. QUALITY ASURANCE AND MANAGEMENT
10.1 Analyse a standard solution from 1.0 mg/l (3.11.2) at the beginning and end
of each series of measurement. The results must be 1.0 ± 0.1 mg/l.
10.2 Before each measurement series analyse a blank sample (3.12) and for the
internal quality control (CQI) a overloaded wine (3.13). The blank sample must
not be over 0.0 mg/l ± 0.1 mg/l. and the CQI must not be over 1.0 mg/l ± 0.2 mg/l.
OIV-MA-AS321-03 : R2004
8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
Annex A
References
1. AOAC International, AOAC Official Methods Program, Associate Referee’s
Manual On Development, Study, Review, and Approval Process,1997
2. Postel, W.; Prasch, E., Wein-Wissenschoft, (1975) 30 (6), 320-326
3. Office International de la Vigne et du Vin, Compendium of International
Methods of Wine Analysis, 255–257
4. Gil Armentia, J. M.; Arranz, J. F.; Barrio, R. J.; Arranz, A., Anales de
Bromatologia, (1988) 40 (1) 71-77
5. Gran, G; Analyst (1952) 77, 661
6. Corning fluoride ion selective electrode – Instruction Manual, 1994
7. Corning Instruction Manual pH/ion analyzer 455, 109121-1 Rev. A, 11/96
8. Horwitz,W.; Albert, R.; Journal of the Association of Official Analytical
Chemists, (1991) 74 (5) 718
OIV-MA-AS321-03 : R2004
9
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
Annex B
Inter laboratory Study
VALIDATION OF A FLUORIDE ION SELECTIVE ELECTRODE, STANDARD
ADDITION METHOD FOR THE MEASUREMENT OF FLUORIDE IN WINE
B.1 Introduction
The validation by collaborative trial of a fluoride selective ion electrode, standard
addition method for the determination of fluoride in wine is described. The
collaborative trial involved a total of twelve participants, six European and six
Americans, who took part in the study. The collaborative study was performed
using the AOAC, Youden protocol(1).
B2 Participants
The twelve participants of this validation consisted of laboratories from Austria,
France, Germany, Spain, and the United States and comprised of the following:
BATF Alcohol and Tobacco Laboratory—Alcohol Section, SF, Walnut Creek,
CA., United States; BATF, National Laboratory Ctr., Rockville, MD, United
States; Bundesinstitut für Gesundheitlichen Verbraucherschutz, Berlin, Germany;
Canandaigua Winery, Madera, CA, United States; CIVC, Epernay, France; E. & J.
Gallo Winery-Analytical Services Laboratory, Modesto, CA, United States; E. &
J. Gallo Winery-Technical Analytical Services Laboratory, Modesto, CA, United
States; ETS Labs, St. Helena, CA, United States; Höhere Bundeslehranstalt &
Bundesamt für Wein und Obstbau, Klosterneuburg, Austria; Institut Catala de la
Vinya i el Vi, Vilafranca del Penedes (Barcelona),Spain; Laboratorio Arbitral
Agroalimentario, Madrid, Spain; and Sutter Home Winery, St. Helena, CA.,
United States.
B3 Samples used in the trial
The samples used in the trial are given in Appendix I. They were distributed as
twelve wine samples (six Youden pairs of samples comprised of three red wines
and three white wines).
OIV-MA-AS321-03 : R2004
10
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
Sample
1
2
3
4
5
6
7
8
9
10
11
12
Sample description
White wine with no fortification (total of 0.6 mg/l F-)
White wine fortified with 0.3 mg /l (total of 0.9 mg/l F-)
White wine fortified with 0.9 mg /l (total de 1,5 mg/l F-)
White wine fortified with 1.2 mg /l (total de 1,8 mg/l F-)
White wine fortified with 1.4 mg /l (total de 2,0 mg/l F-)
White wine fortified with 1.7 mg /l (total de 2,3 mg/l F-)
Red wine with no fortification (total de 0,2 mg/l F-)
Red wine fortified with 0.3 mg /l (total de 0,5 mg/l F-)
Red wine fortified with 0.8 mg /l (total de 1,0 mg/l F-)
Red wine fortified with 1.1 mg /l (total de 1,3 mg/l F-)
Red wine fortified with 2.5 mg /l (total de 2,7 mg/l F-)
Red wine fortified with 2.8 mg /l (total de 3,0 mg/l F-)
8.4 Results
A summary of the results obtained by the twelve participants is given in Table I.
None of the laboratories reported any difficulties with the analysis. One Youden
pair from one laboratory was determined to be an outlier, using the Cochran’s test.
These results are noted(c) in Table I, and were not used in the statistical analysis.
OIV-MA-AS321-03 : R2004
11
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
Table 1
Collaborative data for the determination of fluoride in wine by fluoride selective
electrode, standard additiona
Lab
Number
1
2
3
4
5
6
7
8
9
10
11
12
N of cases
Minimum
Maximum
Range
Mean
Median
Std Dev
Pair 1b
1
2
0.55 0.80
0.52 0.81
0.52 0.81
0.62 0.98
0.48 0.78
0.53 0.84
0.53 0.76
0.57 0.88
0.51 0.81
0.54 0.84
0.60 0.93
0.65 0.94
White Wine
Pair 2b
3
4
1.33 1.56
1.39 1.64
1.40 1.70
1.48 1.64
1.34 1.64
1.45 1.74
1.27 1.64
1.51 1.85
1.40 1.71
1.43 1.71
1.48 1.75
1.54 1.79
Pair 3b
5
6
1.86 2.24
1.86 2.31
1.92 2.25
1.85 2.14
1.84 2.11
1.97 2.30
1.89 2.06
2.11 2.33
1.90 2.20
1.93 2.22
1.98 2.32
2.05 2.32
Pair 4b
7
8
0.19 0.45
0.19 0.46
0.14 0.42
0.28 0.56
0.12 0.39
0.13 0.43
0.14 0.40
0.48c 0.48c
0.13 0.42
0.18 0.44
0.25 0.57
0.21 0.52
Red Wine
Pair 5b
9
10
0.89 1.17
0.92 1.20
0.96 1.22
1.00 1.32
0.88 1.16
0.92 1.21
0.88 1.12
1.01 1.32
0.90 1.19
0.96 1.23
1.06 1.31
1.03 1.24
Pair 6b
11
12
2.54 2.77
2.58 2.77
2.64 2.95
2.64 2.72
2.56 2.82
2.66 2.93
2.44 2.83
2.64 3.08
2.60 2.86
2.66 2.87
2.68 2.82
2.81 3.07
12
0.48
0.65
0.17
0.55
0.54
0.050
12
1.27
1.54
0.27
1.42
1.42
0.079
12
1.84
2.11
0.27
1.93
1.91
0.084
11
0.12
0.28
0.16
0.18
0.18
0.052
12
0.88
1.06
0.18
0.95
0.94
0.061
12
2.44
2.81
0.37
2.62
2.64
0.090
12
0.76
0.98
0.22
0.85
0.83
0.069
12
1.56
1.85
0.29
1.70
1.71
0.079
12
2.06
2.33
0.27
2.23
2.25
0.091
11
0.39
0.57
0.18
0.46
0.44
0.063
12
1.12
1.32
0.20
1.22
1.22
0.065
12
2.72
3.08
0.36
2.87
2.85
0.114
a
Units are mg fluoride/L.
Youden pairs
c
Value was deleted from data set by Cochran’s Test and was not included in the
statistical analysis
b
OIV-MA-AS321-03 : R2004
12
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Fluoride
Table 2
Statistical data from the collaborative study on the analysis of fluoride in wine by
fluoride selective ion electrode, standard addition method
White Wine
STATISTIC
Pair 1
Pair 2
Red
Wine
Pair 3
Pair 4
Pair 6
Total # of Labs
12
12
12
11d
Pair 5
12
Number of "replicates”
per lab
Mean (split levels)
2
2
2
2
2
2
0.55
0.85
0.0006
0.0235
1.42
1.70
0.0015
0.0382
1.93
2.23
0.0026
0.5106
0.18
0.46
0.0002
0.0156
0.95
1.22
0.0005
0.0211
2.62
2.87
0.0049
0.0703
3.35 %
2.45 %
2.45 %
4.88 %
1.94 %
2.55 %
0.0039
0.0070
0.0089
0.0034
0.0042
0.0130
0.0625
0.0835
0.0945
0.0587
0.0647
0.1141
8.92 %
5.36 %
4.54 %
18.39 %
5.95 %
4.15 %
16.88
14.97
14.33
19.00
15.80
13.74
0.53
0.36
0.32
0.97
0.38
0.30
93.1
94.6
96.7
91.0
94.4
96.4
Repeatability variance
Repeatability Standard
Deviation
Relative standard
deviation RSDr,
repeatability
Reproducibility
variance
Reproducibility
standard deviation
Relative standard
deviation RSDR,
reproducibility
Horwitz Equation
Applied (as RSDR)
HORRAT Value HoR
(RSDR
(measured)/RSDR
(Horwitz))
Average % recovery
d
12
One lab pair was deleted from data set by Cochran’s Test
OIV-MA-AS321-03 : R2004
13
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Total phosphorus
Method OIV-MA-AS321-04
Type IV method
Total Phosphorus
1. Principle
After nitric oxidation and ashing, and dissolution in hydrochloric acid, phosphoric
acid is determined colorimetrically as the yellow phospho-vanadomolybdate
complex.
2. Apparatus
2.1 Boiling water-bath 100C
2.2 Hot plate
2.3 Temperature-controlled electric furnace.
2.4 Spectrophotometer measuring absorbance at wavelengths between 300 and
700 nm
3. Reagents
3.1 Nitric acid, (20 = 1.39 g/mL).
3.2 Hydrochloric acid, approx. 3 M; hydrochloric acid (20 = 1.15 - 1.18 g/mL)
diluted 1/4 with water.
3.3 Vanadomolybdate reagent:
Solution A: dissolve 40 g of ammonium molybdate, (NH4)6Mo7O24.4H2O, in
400 mL water.
Solution B: dissolve 1 g of ammonium vanadate, NH4VO3, in 300 mL water
and 200 mL nitric acid (20 = 1.39g/L) (3.1). Leave to cool.
Vanadomolybdate reagent: place first solution B then solution A into a 1 liter
flask, and make up to the mark with water. Reagent to be used within 8 days
of preparation.
3.4 P2O5 solution, 0.1 g/L.
Prepare a P2O5 solution 1 g/L by dissolving 2.454 g of di-potassium hydrogen
phosphate, K2HPO4, in a liter of water. Dilute 10% (v/v).
4. Procedure
4.1 Ashing
OIV-MA-AS321-04 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Total phosphorus
Place 5 mL* wine or must in a platinum or silica dish and evaporate on a
boiling water-bath (2.1). When the residue is nearly dry add 1 mL nitric acid
(3.1), place the dish on a hot plate (2.2) for 1 hour then in a furnace (2.3) at
600-650 oC until the ash is white.
4.2 Determination
Add 5 mL of hydrochloric acid, approximately 3 M (3.2) to the ash and
transfer the solution to a 100 mL volumetric flask. Rinse the dish with 50 mL
distilled water and pour the washings into the flask. Add exactly 25 mL of
vanadomolybdate reagent, stir and leave for 15 to 20 min to allow the color to
develop. Determine the absorbance at 400 nm.
Simultaneously, prepare standard solutions. Place in five 100 mL volumetric
flasks, 5, 10, 15, 20 and 25 mL respectively of P2O5 solution, 0.1 g/L (3.4).
Make up to 50 mL with distilled water and add 25 mL vanadomolybdate
reagent. Leave for the exact same time as the samples, to allow the color to
develop. Make up to the mark with water and measure the absorbance at
400 nm.
In order to remain in the best absorbance zone do not reset to zero with
distilled water, but set the deviation of the spectrophotometer galvanometer on
a given absorbance for a determined concentration.
5. Expression of results
5.1 Calculation
The total phosphorous content expressed in milligrams per liter of phosphoric
anhydride, P2O5, is obtained by entering the absorbance of the wine sample on the
calibration graph and interpolating the total phosphorus concentration.
The total phosphorous content is expressed in milligrams per liter P2O5 to the
nearest whole number.
BIBLIOGRAPHY
A.F.N.O.R., Norme U, 42-246, Tour Europe, Paris.
SUDRAUD P., Bull. O.I.V., 1969, 462-463, 933.
*
A 5 mL sample volume is suitable for P2O5 content, of between 100 and 500 mg/L.
Outside these concentration limits, increase or decrease the sample volume.
OIV-MA-AS321-04 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfates
Method OIV-MA-AS321-05A
Type II method
Sulfates
1. Principle
Gravimetric determination following precipitation of barium sulfate. The
barium phosphate precipitated at the same time is eliminated by washing the
precipitate in hydrochloric acid.
In the case of musts or wine rich in sulfur dioxide, prior de-sulfiting by boiling
in an airtight vessel is recommended.
2. Method
2.1 Reagents
2.1.1 Hydrochloric acid, 2 M.
2.1.2 Barium chloride solution, BaCl2.2H2O, 200 g/L.
2.2 Procedure
2.2.1 General procedure:
Introduce 40 mL of the sample to be analyzed into a 50 mL centrifuge tube;
add 2 mL hydrochloric acid, 2 M (2.1.1), and 2 mL of barium chloride
solution, 200 g/L (2.1.2). Stir with a glass stirrer; rinse the stirrer with a little
distilled water and leave to stand for five min. Centrifuge for five min, then
carefully decant the supernatant liquid.
Wash the barium sulfate precipitate as follows: add 10 mL hydrochloric acid,
2 M (2.1.1), place the precipitate in suspension and centrifuge for five min,
then carefully decant the supernatant liquid. Repeat the washing procedure
twice as before using 15 mL distilled water each time.
Quantitatively transfer the precipitate, with distilled water, into a tared
platinum capsule and place over a water bath at 100°C until fully evaporated.
The dried precipitate is calcined several times briefly over a flame until a white
residue is obtained. Leave to cool in a desiccator and weigh.
Let m = mass in milligrams of barium sulfate obtained.
2.2.2 Special procedure: sulfited must and wine with a high sulfur dioxide
content.
Elimination of sulfur dioxide.
Measure 25 mL of water and 1 mL of concentrated hydrochloric acid (20=
1.15 to 1.18 g/mL) into a 500 mL conical flask equipped with a dropping
funnel and an outlet tube. Boil the solution to remove the air and introduce
OIV-MA-AS321-05A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfates
100 mL of wine through the dropping funnel. Continue boiling until the
volume of liquid in the flask has been reduced to about 75 mL and
quantitatively transfer, after cooling, to a 100 mL volumetric flask. Make up
to mark with water. Determine the sulfate in the 40 mL sample as indicated in
2.2.1.
2.3. Expression of results
2.3.1 Calculations:
The sulfate content, expressed in milligrams per liter of potassium sulfate, K 2SO4
is given by:
18.67 x m
The sulfate content in musts or wine is expressed in milligrams per liter of
potassium sulfate, to the nearest whole number.
2.3.2 Repeatability (r):
up to 1000 mg/L:
approx. 1500 mg/L:
r = 27 mg/L
r = 41 mg/L
2.3.3 Reproducibility (R):
up to 1000 mg/L:
approx. 1500 mg/L:
R = 51 mg/L
R = 81 mg/L
BIBLIOGRAPHY
DEIBNER L , BÉNARD P., Ind. alim. agric., 1954, 71, no1, 23; no5, 427; 1955, 72,
no9-10, 565 et no11, 673.
DEIBNER L., Rév. ferm. ind. alim., 1959, 14 no5, 179 et no6, 227.
BLAREZ Ch., Vins et spiritueux, 1908, 149, Maloine éd., Paris.
DER HEIDE X. von, SCHITTHENNER F., Der Wein, 1922, 320, Vieweg & Sohn
Verlag, Braunschweig.
JAULMES P., Analyse des vins, 1924, 73, Dubois et Poulain, éd., Montpellier; 2e
édition, 1951, 112.
SIMONEAU G., Étude sur les moûts concentrés de raisins, 1946, Thèse pharm.,
Montpellier, 49.
RIBÉREAU-GAYON J., PEYNAUD E., Analyse et contrôle des vins, 1947, 244, Ch.
Béranger éd., Paris-Liège.
FROLOV-BAGREEV A., AGABALIANTZ G., Chimie du vin, 1951, 369, Moscou,
Laboratoire de chimie de l'État de Würzburg (Allemagne), F.V., O.I.V., 1969, no321.
OIV-MA-AS321-05A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfates
Method OIV-MA-AS321-05B
Sulfates
Quick test method
Wines are classified into several categories using the so-called “limits”
method, based on the precipitation of barium sulfate using a barium ion titrant.
WITHDRAWN
OIV-MA-AS321-05B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ammonium
Method OIV-MA-AS322-01
Type IV method
Ammonium
1. Principle
Retention of the ammonium cation on a weak cation exchange resin, elution using
an acidic solution, distillation of the eluent and determination of the ammonia in
the distillate by titration with a standardized solution of hydrochloric acid.
2. Apparatus
2.1 Cation exchange resin column
A 50 mL burette with a glass stopcock fitted with a glass wool plug containing
25 g of weak cation exchange resin (e.g. Amberlite IR-50, 80-100 mesh).
Wash alternately with 1 M sodium hydroxide solution and 1 M hydrochloric
acid solution. Wash the resin with distilled water until a negative reaction of
chloride ion with silver nitrate is obtained. Pass 50 mL of neutral buffer
slowly through the glass column, rinse with distilled water until phosphates
begin to elute as detected using a saturated solution of lead acetate.
2.2 Distillation apparatus
Use the apparatus described in the chapter on Alcoholic Strength 3.1
The condensate is transferred to the conical flask through a drawn-out tube
touching the bottom of the vessel.
Alternatively, it is possible to use the steam distillation apparatus used in the
chapter on Volatile Acidity 4.1 or other apparatus that can be used for the
following experiments which check the purity of the reagents.
a) Place 40-45 mL of 30 % sodium hydroxide solution (v/v), 50 mL of water
and 50 mL hydrochloric acid, 1 M, in the distillation flask. Distil half the
volume and collect the distillate in 30 mL of boric acid solution, 40 g/L to
which 5 drops of methyl red have been added. Adjust the color to pink by
the addition of 0.1 mL of 0.1 M hydrochloric acid.
b) A test (similar to that described in a) is conducted using, 10 mL 0.05 M
ammonium sulfate solution, containing 3.55 g/L of anhydrous ammonium
sulfate, (NH4)2SO4. In this case, between 10 and 10.1 mL 0.1 M
hydrochloric acid must be used to obtain the change of color of the
indicator.
3. Reagents
3.1 Hydrochloric acid solution, 1 M.
OIV-MA-AS322-01 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ammonium
3.2 Sodium hydroxide, 1 M.
3.3 Neutral solution to wash the resin:
di-sodium hydrogen phosphate Na2PO4.12H2O ………................
15 g
potassium di-hydrogen phosphate KH2KO4 ....………..…............. 3.35 g
water to .......................................………………….…………………… 1000 mL
Verify pH is 7±0.2
3.4 Sodium hydroxide solution, 30% (m/m),  = 1.33 g/mL
3.5 Hydrochloric acid solution, 0.1 M.
3.6 Phenolphthalein solution, 1% (m/v), in neutral ethanol, 96% (V/V)
3.7 Bromocresol green solution, 1% (m/v):
bromocresol green ..........................................….
1g
dissolve in 0.1 M sodium hydroxide solution, .....…….. 14 mL
water to............................................................ 100 mL
3.8 Methyl red ethanol/water solution, 0.2% (v/v):
methyl red ......................................................
0.2 g
alcohol, 95% (vol.) ........................................... 60 mL
water to ......................................................... 100 mL
3.9 Boric acid solution
Boric acid ........................................................
40g
Water to .......................................................... 1000mL
Boric acid usually contains a small quantity of alkaline impurities and it is
possible to correct this by adding 5 drops of indicator to this solution and
adjusting to a pink color by means of few drops of 0.1 M hydrochloric acid (1
mL at most).
4. Procedure
Transfer 50 mL of the sample to be analyzed into a 250 mL beaker. Add a
quantity of sodium hydroxide, 1 M, equal to half of (n-0.5) mL, where n is the
volume sodium hydroxide solution, 0.1 M, used in the total acidity titration on 10
mL of wine. Pass this mixture through the cation exchange column (2.1) at a rate
of one drop every two seconds. The eluent pH should lie between 4 and 5. Rinse
the column with 50 mL of distilled water at the same flow rate.
Ammonium and other cations are quantitatively retained on the column. Amides,
oligopeptides and nearly all amino acids are eluted by the washing procedure.
OIV-MA-AS322-01 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Ammonium
Elute the cations retained on the resin with 50 mL of 1 M hydrochloric acid, (3.1)
and rinse with 50 mL distilled water.* The eluate and the water washings are
combined in a 1 liter round bottom distillation flask.
Add one drop of phenolphthalein, 1% (m/v), and sufficient quantity of 30%
sodium hydroxide solution (m/v)(3.4), to obtain a true alkaline reaction, constantly
cooling the flask during this addition.
Distil about half the volume of the liquid from the distillation flask, into 30 mL of
4% boric acid (m/v)(3.9).
The distillate is titrated with 0.1 M hydrochloric acid (3.5), in the presence of
bromocresol green or methyl red. Record the volume of hydrochloric acid used (n).
5. Expression of results
The content of ammonium (NH4) ions is expressed in milligrams per liter to the
nearest whole number.
5.1 Calculation
The content of ammonium ions, expressed in milligrams per liter is:
36 x n
When wines with low ammonium content are analyzed, the determination is
conducted using 100 mL of wine. In this case the quantity of ammonium is given
by:
18 x n
BIBLIOGRAPHY
Usual Method:
JAULMES P., Analyse des vins, 1951, 220, Montpellier
KOURAKOU Mme S., Ann. Fals. Exp. Chim., 1960, 53, 337.
* The column should be washed with 50 mL of neutral buffer solution and rinsed with water
before using the column for another determination.
OIV-MA-AS322-01 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Potassium
Method OIV-MA-AS322-02A
Type II method
Potassium
1. Principle
Potassium is determined directly in diluted wine by atomic absorption
spectrophotometry after the addition of cesium chloride to suppress ionization
of potassium.
2 Method
2.1 Apparatus
 Atomic absorption spectrophotometer, equipped with an air-acetylene burner
 Potassium hollow cathode lamp
2.2 Reagents
2.2.1 Solution containing 1 g of potassium per liter.
Use a standard commercial solution containing 1 g of potassium per liter. This
solution may be prepared by dissolving 4.813 g of potassium hydrogen tartrate
(C4H5KO6) in distilled water and making up the volume to 1 liter with water.
2.2.2 Matrix (model) solution:
citric acid monohydrate ..............................…….......……..
3.5 g
sucrose ..........................................................………………….
1.5 g
glycerol .........................................................………………….
5.0 g
anhydrous calcium chloride, (CaCl2) ............……………..
50 mg
anhydrous magnesium chloride (MgCl2) ............………… 50 mg
absolute alcohol ...............................................……………..
50 mL
500 mL
water to ...........................................................………………….
2.2.3 Cesium chloride solution containing 5% cesium:
Dissolve 6.33 g of cesium chloride, CsCl, in 100 mL of distilled water.
2.3 Procedure
2.3.1 Preparation of sample
Pipette 2.5 mL of wine (previously diluted 1/10) into a 50 mL volumetric flask,
add 1 mL of the cesium chloride solution and make up to the mark with
distilled water.
OIV-MA-AS322-02A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Potassium
2.3.2 Calibration
Introduce 5.0 mL of the matrix solution into each one of five of 100mL
volumetric flasks and add 0, 2.0, 4.0, 6.0 and 8.0 mL respectively of the 1 g/L
potassium solution (previously diluted 1/10). Add 2 mL of the cesium chloride
solution to each flask and make up to 100 mL with distilled water.
The standard solutions contain 0, 2, 4, 6 and 8 mg of potassium per liter
respectively and each contains 1 g of cesium per liter. Keep these solutions in
polyethylene bottles.
2.3.3 Determination
Set the wavelength to 769.9 nm. Zero the absorbance scale using the zero
standard solution (2.3.2). Aspirate the diluted wine (2.3.1) directly into the
spectrophotometer, followed in succession by the standard solutions (2.3.2).
Record the absorbance for each solution and repeat.
2.4 Expression of results
2.4.1 Method of calculation
Plot a graph showing the variation in absorbance as a function of potassium
concentration in the standard solutions.
Record the mean absorbance obtained with diluted wine on this graph and
determine its potassium concentration C in milligrams per liter.
The potassium concentration, expressed in milligrams per liter of the wine to
the nearest whole number, is F x C, where F is the dilution factor (here 200).
2.4.2 Repeatability (r):
r = 35 mg/L.
2.4.3 Reproducibility (R): R = 66 mg/L.
2.4.4 Other ways of expressing results
 In milliequivalents per liter: 0.0256 x F x C.
 In mg potassium hydrogen tartrate per liter: 4.813 x F x C.
OIV-MA-AS322-02A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Potassium
Method OIV-MA-AS322-02B
Type III method
Potassium
1. Principle
Potassium is determined directly in diluted wine by flame photometry.
Note: The gravimetric determination of potassium tetraphenylborate
precipitated from the solution of the ash of wine is a precise method for the
determination of potassium and is described in the annex.
2. Method
2.1 Apparatus
2.1.1 Flame photometer supplied with an air-butane mixture.
2.2 Reagents
2.2.1 Reference solution containing 100 mg potassium per liter
Absolute alcohol ............................................………….
10 mL
Citric acid C6H8O7, H2O .............…..............…………………
700 mg
Sucrose ........................................................…………………..
300 mg
1000 mg
Glycerol ......................................................….………………..
Sodium chloride, NaCl . ....................................……………
50.8 mg
Anhydrous calcium chloride, CaCl2 ....................…………..
10 mg
Anhydrous potassium hydrogen tartrate .......………………… 481.3 mg
water to .........................................................……………………
1000 mL
Dissolve the potassium hydrogen tartrate in 500 mL of very hot distilled water,
mix this solution with 400 mL of distilled water in which the other chemicals
have already been dissolved, and make up to one liter.
2.2.2 Dilution solution
Absolute alcohol .............................................…….…
10 mL
Citric acid anhydrous .......................................……………. 700 mg
Sucrose ........................................................…………………
300 mg
Glycerol ........................................................…………………
1000 mg
Sodium chloride, NaCl ....................................….………..
50.8 mg
10 mg
Anhydrous calcium chloride, CaCl2 .............……….....
Anhydrous magnesium chloride, MgCl2 .......…………...
10 mg
Tartaric acid ..................................................……………….
383 mg
Water to ........................................................…………………..
1000 mL
Preserve the solutions in polyethylene bottles by adding two drops of allyl
isothiocyanate (3-isothiocyanato-1-propene; CH2=CHCH2NCS).
OIV-MA-AS322-02B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Potassium
2.3 Procedure
2.3.1 Calibration
Place 25, 50, 75 and 100 mL of the reference solution into a set of four 100 mL
volumetric flasks and make up to 100 mL with the dilution solution to give
solutions containing 25, 50, 75 and 100 mg of potassium per liter respectively.
2.3.2 Determination
Make measurements at 766 nm. and adjust the 100% transmission using
distilled water. Successively aspirate the standard solutions directly into the
burner of the photometer, followed by wine diluted 1/10 with distilled water
and note the readings. If necessary, the wine already diluted 1/10 may be
further diluted with the dilution solution (2.2.2).
2.4 Expression of results
2.4.1 Method of calculation
Plot a graph of the variation in percentage transmission as a function of the
potassium concentration in the standard solutions. Record the transmission
obtained for the sample of diluted wine on this graph and determine the
corresponding potassium concentration C.
The potassium concentration in mg potassium per liter to the nearest whole
number will be:
FxC
where F is the dilution factor.
2.4.2 Repeatability (r):
r = 17 mg/L.
2.4.3 Reproducibility (R): R = 66 mg/L.
2.4.4 Other ways of expressing results:
 In milliequivalents per liter: 0.0256 x F x C.
 In mg potassium hydrogen tartrate per liter 4.813 x F x C.
OIV-MA-AS322-02B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Potassium
Method OIV-MA-AS322-02C
Potassium
(Resolution Oeno 377/2009)
Gravimetric determination of potassium using sodium tetraphenylborate
WITHDRAWN
OIV-MA-AS322-02C : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sodium
Method OIV-MA-AS322-03A
Type II method
Sodium
1. Principle
Sodium is determined directly in the wine by atomic absorption
spectrophotometry after the addition of cesium chloride to suppress ionization
of sodium.
2. Method
2.1 Apparatus
 Atomic absorption spectrophotometer equipped with an air-acetylene
burner.
 Sodium hollow cathode lamp.
2.2 Reagents
2.2.1 Solution containing 1 g of sodium per liter:
The use of a commercial standard solution containing 1 g of sodium per liter is
preferred.
Alternatively, this solution may be prepared by dissolving 2.542 g of
anhydrous sodium chloride (NaCl) in distilled water and making up to a
volume of 1 liter.
Keep this solution in a polyethylene bottle.
2.2.2 Matrix (model) solution:
Citric acid monohydrate, (C6H8O7.H2O) .....................…… 3.5 g
Sucrose ..........................................................…………………….
1.5 g
Glycerol .........................................................……………………
5.0 g
Anhydrous calcium chloride (CaCl2) ......................………. 50 mg
Anhydrous magnesium chloride, (MgCl2) .................…… 50 mg
Absolute alcohol ...............................................……………….
50 mL
De-ionized water to ............................................……………….
500 mL
2.2.3 Cesium chloride solution containing 5% cesium
Dissolve 6.330 g of cesium chloride, CsCl, in 100 mL of distilled water.
2.3 Procedure
2.3.1 Preparation of the sample
Pipette 2.5 mL of wine into a 50 mL volumetric flask, add 1 mL of the cesium
chloride solution (2.2.3) and make up to the mark with distilled water.
OIV-MA-AS322-03A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sodium
2.3.2 Calibration
Place 5.0 mL of the matrix solution in each one of five 100 mL volumetric
flasks and add 0, 2.5, 5.0, 7.5 and 10 mL respectively of a 1:100 dilution of the
1 g/L sodium solution. Add 2 mL of the cesium chloride solution (2.2.3) to
each flask and make up to 100 mL with distilled water.
The standard solutions prepared in this way contain 0.25, 0.50, 0.75 and 1.00
mg of sodium per liter respectively and each contains 1 g of cesium per liter.
Keep these solutions in polyethylene bottles.
2.3.3 Determination
Set the absorbance wavelength to 589.0 nm. Zero the absorbance scale using
the zero standard solution. Aspirate the diluted wine (2.3.1) directly into the
spectrophotometer, followed in succession by the standard solutions (2.3.2).
Record each absorbance and repeat each measurement.
2.4 Expression of results
2.4.1 Method of calculation
Plot a graph of measured absorbance versus the sodium concentration in the
standard solutions.
Record the absorbance obtained with the diluted wine on this graph and
determine its sodium concentration C in milligrams per liter.
The sodium concentration in milligrams per liter of the wine will then be F x
C, expressed to the nearest whole number, where F is the dilution factor.
2.4.2. Repeatability (r):
r = 1 + 0.024 xi mg/L.
xi = concentration of sodium in the sample in mg/L.
2.4.3. Reproducibility (R):
R = 2.5 + 0.05 xi mg/L.
xi = concentration of sodium in the sample in mg/L.
OIV-MA-AS322-03A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sodium
Method OIV-MA-AS322-03B
Type III method
Sodium
1. Principle
Sodium is determined directly in diluted wine (at least 1 mL:10 mL) by flame
photometry.
2. Method
2.1 Apparatus
2.1.1. Flame photometer supplied with an air-butane mixture.
2.2 Reagents
2.2.1 Reference solution containing 20 mg sodium per liter
Absolute alcohol ................................................................
10 mL
Citric acid monohydrate (C6H8O7 H2O) .....................…... 700 mg
Sucrose ..................................................…........................
300 mg
Glycerol .....................................................................…….
1000 mg
Potassium hydrogen tartrate .........................………........
481.3 mg
Anhydrous calcium chloride, CaCl2 .......................…...
10 mg
Anhydrous magnesium chloride, MgCl2 ......................
10 mg
Dry sodium chloride, NaCl ...................................………
50.84 mg
Water to ............................................................................
1000 mL
2.2.2 Dilution solution
Absolute alcohol ..............................................……….…...
10 mL
Citric acid monohydrate (C6H8O7.H2O) ...................… 700 mg
Sucrose ........................................................………….…….
300 mg
Glycerol ...............................................................…........
1000 mg
Potassium hydrogen tartrate ...............................….…....
481.3 mg
Anhydrous calcium chloride, CaCl2 ......................…...
10 mg
Anhydrous magnesium chloride, MgCl2 .....................
10 mg
Water to .....................................................…………………..
1000 mL
To prepare 2.2.1 and 2.2.2, dissolve the potassium hydrogen tartrate in
approximately 500 mL of very hot distilled water, mix with 400 mL of distilled
water into which the other chemicals have already been dissolved, and make
up to one liter.
Preserve the solutions in polyethylene bottles by adding two drops of allyl
isothiocyanate to each.
OIV-MA-AS322-03B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sodium
2.3 Procedure
2.3.1 Calibration
Place 5, 10, 15, 20 and 25 mL of the reference solution in each of five 100 mL
volumetric flasks and make up to 100 mL with the dilution solution to give
solutions containing 1, 2, 3, 4 and 5 mg of sodium per liter respectively.
2.3.2 Determination
Carry out measurements at 589.0 nm and adjust the 100% transmission using
distilled water. Successively aspirate the standard solutions directly into the
photometer, followed by the wine diluted 1:10 with distilled water and note the
percentage transmission of each. If necessary, the wine already diluted 1:10
may be further diluted with dilution solution.
2.4 Expression of results
2.4.1 Calculation method
Plot a graph of the percentage transmittance versus sodium concentration of
the standard solutions. Record the transmission obtained for the diluted wine
sample on this graph and note the concentration, C, of sodium in the wine.
The sodium concentration in mg of sodium per liter will be:
FxC
where F is the dilution factor.
2.4.2 Repeatability (r)
r = 1.4 mg/L (except for liqueur wine)
r = 2.0 mg/L for liqueur wine.
2.4.3. Reproducibility (R)
R = 4.7 + 0.08 xi mg/L.
xi = sodium concentration in the sample in mg/L.
OIV-MA-AS322-03B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Calcium
Method OIV-MA-AS322-04
Type II method
Calcium
1. Principle
Calcium is determined directly on diluted wine by atomic absorption
spectrophotometry after the addition of an ionization suppression agent.
2. Apparatus
2.1 Atomic absorption spectrophotometer fitted with an air-acetylene burner.
2.2 Calcium hollow cathode lamp.
3. Reagents
3.1 Calcium standard solution 1 g/L. Use of a standard commercial calcium
solution, 1 g/L, is preferred.
Alternatively this solution may be prepared by dissolving 2.5 g of calcium
carbonate, CaC03, in sufficient hydrochloric acid (concentrated hydrochloric
acid diluted 1:10) to dissolve it completely and making up to one liter with
distilled water.
3.2 Dilute calcium standard solution, 50 mg/L
Note : Store the calcium solutions in polyethylene containers.
3.3 Dilute lanthanum standard solution, 50 g/L
Dissolve 13.369 g of lanthanum chloride, LaCl3.7H2O in distilled water; add 1
mL, of dilute hydrochloric acid (concentrated hydrochloric acid diluted 1/10)
and make up to 100 mL with distilled water.
4. Procedure
4.1 Preparation of sample
Place 1 mL of wine and 2 mL of the lanthanum chloride solution (3.3) in a 20
mL volumetric flask and make up to the mark with distilled water. The diluted
wine contains 5 g lanthanum per liter.
Note: For sweet wines, 5 g lanthanum per liter is sufficient provided that the
dilution reduces the sugar content to less than 2.5 g/L. For wines with higher
concentrations of sugar, the lanthanum concentration should be increased to 10
g/L.
4.2 Calibration
Place 0, 5, 10, 15 and 20 mL, of dilute standard calcium solution (3.2)
respectively into each of five 100 mL volumetric flasks, followed by 10 mL of
OIV-MA-AS322-04 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Calcium
the lanthanum chloride solution (3.3) and make up to 100 mL with distilled
water. The solutions prepared in this way contain 0, 2.5, 5.0, 7.5 and 10 mg of
calcium per liter respectively, and each contains 5 g of lanthanum per liter.
These solutions should be stored in polyethylene bottles.
4.3 Determination
Set the absorbance wavelength to 422.7 nm. Zero the absorbance scale using
the zero standard (4.2). Aspirate the diluted wine directly into the
spectrophotometer, followed in succession by the five standard solutions (4.2)
and record the absorbance. Repeat each measurement.
5. Expression of results
5.1 Method of calculation
Plot a graph showing the variation in absorbance as a function of the calcium
concentration in the standard solutions.
Record the mean value of the absorbance obtained with the sample of diluted
wine on this graph and read its calcium concentration C. The calcium
concentration in milligrams per liter of the wine to the nearest whole number is
given by:
20 x C.
5.2 Repeatability (r)
Concentration<60mg/L:
Concentration > 60 mg/L:
r=2.7mg/L.
r = 4 mg/L.
5.3 Reproducibility (R)
R mg/L = 0.114 xi - 0.5.
where xi = concentration in the sample in mg/L.
OIV-MA-AS322-04 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Iron
Method OIV-MA-AS322-05A
Type IV method
Iron
1. Principle
After suitable dilution of the wine and removal of alcohol, iron is determined
directly by atomic absorption spectrophotometry.
2. Method
2.1 Apparatus
2.1.1 Rotary evaporator with thermostatically controlled water bath.
2.1.2 Atomic absorption spectrophotometer equipped with an air-acetylene
burner.
2.1.3 Iron hollow cathode lamp.
2.2 Reagents
2.2.1 Concentrated standard iron solution containing 1 g Fe (III) per liter.
Use a standard commercial solution, 1 g/L. This solution may be prepared by
dissolving 8.6341 g of ferric ammonium sulfate, FeNH4 (SO4)2.12H20, in
distilled water slightly acidified with hydrochloric acid, 1 M, and making up to
one liter.
2.2.2 Dilute standard iron solution containing 100 mg iron per liter.
2.3 Procedure
2.3.1 Preparation of sample
Remove the alcohol from the wine by reducing the volume of the sample to
half its original size using a rotary evaporator (50 to 60 C). Make up to the
original volume with distilled water.
If necessary, dilute prior to analysis with distilled water.
2.3.2 Calibration
Place 1, 2, 3, 4 and 5 mL of the solution containing 100 mg iron per liter
(2.2.2) respectively into each of five 100 mL volumetric flasks and make up to
100 mL with distilled water. The solutions prepared in this way contain 1, 2, 3,
4 and 5 mg of iron per liter respectively. These solutions should be stored in
polyethylene bottles.
2.3.3 Determination
Set the absorption wavelength to 248.3 nm. Zero the absorbance scale using
distilled water. Aspirate the diluted sample directly into the spectrophotometer,
OIV-MA-AS322-05A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Iron
followed in succession by the five standards (2.3.2). Record the absorbance.
Repeat each measurement.
2.4 Expression of results
2.4.1 Method of calculation
Plot a graph giving the variation in absorbance as a function of the iron
concentration in the standard solutions. Record the mean value of the
absorbance obtained with the diluted wine sample on this graph and read its
iron concentration C.
The iron concentration in milligrams per liter of the wine to one decimal place
is given by:
FxC
where F is the dilution factor.
OIV-MA-AS322-05A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Iron
Method OIV-MA-AS322-05B
Type IV method
Iron
1. Principle
After digestion in hydrogen peroxide, 30%, the total iron, present as Fe (III) state,
is reduced to the Fe (II) and quantified by the formation of a colored
orthophenanthroline complex.
2. Method
2.1 Apparatus
2.1.1 Kjeldahl flask, 100 mL.
2.1.2 Spectrophotometer enabling measurements to be made at a wavelength of
508 nm.
2.2 Reagents
2.2.1 Hydrogen peroxide, H202, 30% (m/v), solution, iron free.
2.2.2 Hydrochloric acid, 1 M, iron free.
2.2.3 Ammonium hydroxide (20 = 0.92 g/mL).
2.2.4 Pumice stone grains, pretreated with boiling hydrochloric acid diluted 1/2
and washed with distilled water.
2.2.5 Hydroquinone solution, C6H602, 2.5%, acidified with 1 mL concentrated
sulfuric acid (20 = 1.84 g/mL) per 100 mL of solution. This solution must be
kept in an amber bottle in the refrigerator and discarded at the slightest sign of
darkening.
2.2.6 Sodium sulfite solution, Na2S03, 20%, prepared from neutral anhydrous
sodium sulfite.
2.2.7 ortho-phenanthroline solution, C12H8N2, 0.5%, in alcohol, 96% vol.
2.2.8 Ammonium acetate solution, CH3COONH4, 20% (m/v).
2.2.9 Fe (III) solution containing 1 g of iron per liter. Use of a commercial
solution is preferred. Alternatively, a 1000 mg/L Fe (III) solution can be
prepared by dissolving 8.6341 g of ferric ammonium sulfate, FeNH4
(SO4)2.12H20, in 100 mL of hydrochloric acid, 1 M, and making up the volume
to one liter with the hydrochloric acid, 1 M.
2.2.10 Dilute standard iron solution containing 100 milligrams of iron per liter.
OIV-MA-AS322-05B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Iron
2.3 Procedure
2.3.1 Digestion
2.3.1.1 For wines with sugar content below 50 g/L
Combine 25 mL of the wine, 10 mL of the hydrogen peroxide solution and a
few grains of pumice into the 100 mL Kjeldah1 flask. Concentrate the mixture
to a volume of 2 to 3 mL by heating. Allow to cool and add sufficient
ammonium hydroxide to make the residue alkaline thus precipitating
hydroxides while taking care not to wet the walls of the flask.
After cooling, carefully add hydrochloric acid, to the alkaline liquid to dissolve
the precipitated hydroxides and transfer the resulting solution to a 100 mL
volumetric flask. Rinse the Kjeldahl flask with hydrochloric acid, and
combined the solutions in the volumetric flask and make up to 100 mL.
2.3.1.2 For musts and wines with sugar content above 50 g/L
 If the sugar content is between 50 and 200 g/L, the 25 mL wine sample is
treated with 20 mL of hydrogen peroxide solution. Continue as in 2.3.1.1.
 If the sugar content is greater than 200 mg/L, the samples of wine or must
should be diluted 1/2 or possibly 1/4 before being treated with 20 mL of
hydrogen peroxide solution. Continue as in 2.3.1.1.
2.3.2 Blank experiment
Carry out a blank trial with distilled water using the same volume of hydrogen
peroxide solution as the amount used for the mineralization, following the
experimental protocol described in 2.3.1.1.
2.3.3 Determination
Introduce 20 mL of the hydrochloric acid wine digest solution and 20 mL, of
the hydrochloric acid solution obtained from the 'blank experiment' into two
separate 50 mL volumetric flasks. Add 2 mL of hydroquinone solution, 2 mL
of sulfite solution and 1 mL of ortho-phenanthroline. Allow to stand for 15
minutes, during which time Fe (III) is reduced to Fe (II). Then add 10 mL of
ammonium acetate solution, make each up to 50 mL with distilled water and
shake the two volumetric flasks. Use the solution originating from the blank
experiment to zero the absorbance scale at 508 nm and measure the absorbance
of the wine solution at the same wavelength.
2.3.4 Calibration
Place 0.5, 1, 1.5 and 2 mL of the 100 mg of iron per liter solution into each of
four 50 mL volumetric flasks, and add 20 mL of distilled water to each. Carry
out the procedure described in 2.3.3 to measure the absorbance of each of these
OIV-MA-AS322-05B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Iron
standard solutions, which contain 50, 100, 150 and 200 micrograms of iron
respectively.
2.4 Expression of results
2.4.1 Method of calculation
Plot a graph giving the variation in absorbance as a function of the iron
concentration in the standard solutions. Record the absorbance of the test
solution and read off the iron concentration C in the hydrochloric acid
digestion solution, i.e. in 5 mL of the wine being analyzed.
The iron concentration in milligrams per liter of the wine to one decimal place
is given by:
200  C
If the wine (or must) has been diluted, the iron concentration in milligrams per
liter of the wine to one decimal place is given by:
200  F  C
where F is the dilution factor.
OIV-MA-AS322-05B : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Copper
Method OIV-MA-AS322-06
Type IV method
Copper
(Resolution Oeno 377/2009)
1. Principle
The method is based on the use of atomic absorption spectrophotometry.
2. Apparatus
2.1 Platinum dish.
2.2 Atomic absorption spectrophotometer.
2.3 Copper hollow cathode lamp.
2.4 Gas supplies: air-acetylene or nitrous oxide/acetylene.
3. Reagents
3.1 Metallic copper.
3.2 Nitric acid (20 = 1.38 g/mL), 65%.
3.3 Nitric acid (3.2), diluted 1/2 (v/v) with water.
3.4 Solution containing 1g of copper per L.
Use of a standard commercial copper solution is preferred. Alternatively this
solution may be prepared by weighing 1.000 g of metallic copper and
transferring it without loss to a 1000 mL volumetric flask. Add just enough
dilute nitric acid to dissolve the metal, add 10 mL of concentrated nitric acid
and make up to the mark with double distilled water.
3.5 Solution containing copper at 100 mg/L
Transfer 10 mL, of the 1 g/L solution 3.4. into a 100 mL volumetric flask, and
make up to the mark with double-distilled water.
3.6 Double-distilled water
4. Procedure
4.1 Preparation of sample and determination of copper
Place 20 mL sample in a 100 mL volumetric flask and make up to 100 mL with
double-distilled water. Modify the dilution if necessary to obtain a response
within the dynamic range of the detector.
OIV-MA-AS322-06 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Copper
Measure the absorbance at 324.8 nm. Set the zero with double distilled water.
4.2 Constructing a standard curve
Pipette 0.5, 1 and 2 mL of copper solution into each of three 100 mL
volumetric flasks and make to the volume with double distilled water; the
solutions contain 0.5, 1 and 2 mg of copper per liter respectively. Measure the
absorbance of standard solutions and the sample prepared in and repeat each
measurement. Plot a graph showing the variation in absorbance as a function of
the copper concentration in the standard solutions.
5 Expression of results
5.1 Method of calculation
Using the measured absorbance of the samples read off the concentration C in
mg/L from the calibration curve.
If F is the dilution factor, the concentration of the copper present is given in
milligrams per liter by:
F x C.
It is quoted to two decimal places.
Notes:
a) Select a sample dilution appropriate to the sensitivity of the apparatus to be
used and the concentration of the copper present in the sample.
b) Proceed as follows when very low copper concentrations are expected in the
sample to be analyzed: Place 100 mL of the sample in a platinum dish and
evaporate on a water bath at 100 C until it becomes syrupy. Add 2.5 mL of
concentrated nitric acid drop wise, covering the bottom of the dish completely.
Carefully ash the residue on an electric hotplate or over a low flame; then place
the dish in a muffle furnace set at 500 ± 25C and leave for about one hour.
After coo1ing, moisten the ash with 1 mL of concentrated nitric acid while
crushing it with a glass rod; allow the mixture to evaporate and ash again as
before. Place the dish in the muffle furnace again for 15 min; repeat the
treatment with nitric acid at least three times. Dissolve the ash by adding 1 mL
of concentrated nitric acid and 2 mL of double distilled water to the dish and
transfer to a 10 mL flask. Wash the dish three times using 2 mL of double
distilled water each time. Finally, make to volume with double distilled water.
Proceed to analyze the sample as in 4.1 but use 10 mL of solution. Take into
account the change in dilution factor when calculating the results.
OIV-MA-AS322-06 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Magnesium
Method OIV-MA-AS322-07
Type II method
Magnesium
1. Principle
Magnesium is determined directly on diluted wine by atomic absorption
spectrophotometry.
2. Apparatus
2.1 Atomic absorption spectrophotometer fitted with an air-acetylene burner.
2.2 Magnesium hollow cathode lamp.
3. Reagents
3.1 Concentrated magnesium standard solution containing 1 g/L
Use of a standard commercial magnesium solution (1 g/L) is preferred.
Alternatively, this solution may be prepared by dissolving 8.3646 g of
magnesium chloride, MgC12.6H2O, in distilled water and making up to 1 liter.
3.2 Dilute magnesium standard solution, 5 mg/L.
Note: Keep the standard magnesium solutions in polyethylene bottles.
4. Procedure
4.1 Preparation of sample
The wine is diluted 1/100 with distilled water.
4.2 Calibration
Place 5, 10, 15 and 20 mL of the dilute standard magnesium solution into each
one of a set of four 100 ml. volumetric flasks and make up to 100 mL with
distilled water. The standard solutions prepared in this way contain 0.25, 0.50,
0.75 and 1.0 mg of magnesium per liter respectively. These solutions should be
kept in polyethylene bottles.
4.3 Determination
Set the absorption wavelength to 285 nm. Zero the absorbance scale using
distilled water. Aspirate the diluted wine directly into the spectrophotometer,
followed in succession by the standard solutions (4.2).
Record the absorbance of each solution and repeat each measurement.
OIV-MA-AS322-07 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Magnesium
5. Expression of results
5.1 Method of calculation
Plot a graph showing the variation in absorbance as a function of the
magnesium concentration in the standard solutions.
Record the mean value of absorbance with the diluted sample of wine on this
graph and read off the magnesium concentration C in milligrams per liter. The
magnesium concentration in milligrams per liter of the wine to the nearest
whole number is given by:
100 x C
5.2 Repeatability (r):
r = 3 mg/L.
5.3 Reproducibility (R):
R = 8 mg/L.
OIV-MA-AS322-07 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Zinc
Method OIV-MA-AS322-08
Type IV method
Zinc
1. Principle
After removal of alcohol, zinc is determined directly in the wine by atomic
absorption spectrophotometry.
2. Apparatus
2.1 Rotary evaporator and thermostatically controlled water bath.
2.2 Atomic absorption spectrophotometer equipped with an air-acetylene burner.
2.3 Zinc hollow cathode lamp.
3. Reagents
The water used must be double distilled in borosilicate glass apparatus or of an
equivalent degree of purity.
3.1 Standard solution containing zinc, 1 g/L
Use of a commercial standard zinc solution is preferred. Alternatively this solution may be prepared by dissolving 4.3975 g of zinc sulfate, ZnS04.7H20, in
water and making up the volume to one liter.
3.2 Dilute standard solution containing 100 mg of zinc per liter.
4. Procedure
4.1 Preparation of sample
Remove the alcohol from 100 mL of wine by reducing the volume of the
sample to half its original value using a rotary evaporator (50 to 60 C). Make
up to the original volume of 100 ml, with double distilled water.
4.2 Calibration
Place 0.5, 1, 1.5 and 2 ml, of the solution containing 100 mg zinc per liter into
each one of four 100 mL volumetric flasks and make up to the mark with
double distilled water. The solutions prepared in this way contain 0.5, 1, 1.5
and 2 mg of zinc per liter respectively.
4.3 Determination
Set the absorbance wavelength to 213.9 nm. Zero the absorbance scale using
double distilled water. Aspirate the wine directly into the burner of the
OIV-MA-AS322-08 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Zinc
spectrophotometer, followed in succession by the four standard solutions.
Record the absorbance and repeat each measurement.
5. Expression of results
5.1 Method of calculation
Plot a graph giving the variation in absorbance as a function of zinc
concentration in the standard solutions. Record the mean value of the
absorbance obtained with the diluted wine sample on this graph and determine
its zinc concentration to one decimal place.
OIV-MA-AS322-08 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Silver
Method OIV-MA-AS322-09
Type IV method
Silver
1. Principle
The method is based on the use of atomic absorption spectrophotometry after
ashing the sample.
2. Apparatus
2.1 Platinum dish.
2.2 Water bath, thermostatically controlled to 100 C
2.3 Furnace set at 500 to 525 C.
2.4 Atomic absorption spectrophotometer.
2.5 Silver hollow cathode lamp.
2.6 Gas supplies: air, acetylene.
3. Reagents
3.1 Silver nitrate, AgN03.
3.2 Nitric acid, (20 = 1.38 g/mL), 65%.
3.3 Nitric acid, diluted 1/10 (v/v) with distilled water.
3.4 Solution containing 1 g of silver per L.
Use of a standard commercial silver solution is preferred. Alternatively this
solution may be prepared by dissolving 1.575 g of silver nitrate in dilute nitric
acid and making up to a volume of 1,000 mL with dilute nitric acid (3.3).
3.5 Solution containing 10 mg of silver per L.
Take 10 mL of the 1 mg/L solution and make up to 1 L with dilute nitric acid.
4. Procedure
4.1 Preparation of sample
Place 20 mL of the sample in a platinum dish and evaporate to dryness over a
boiling water bath. Ash in the furnace at a temperature of 500 to 525 C.
Moisten the white ash with 1 mL of concentrated nitric acid (3.2). Evaporate
over a boiling water bath, repeat the addition of 1 mL nitric acid (3.2) and
evaporate a second time. Add 5 mL of dilute nitric acid (3.3) and heat gently
until dissolved.
OIV-MA-AS322-09 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Silver
4.2 Calibration
Pipette 2, 4, 6, 8, 10 and 20 mL of solution (3.5) respectively into each of size
100 mL volumetric flasks and make up to the mark with dilute nitric acid (3.3):
the solutions contain 0.20, 0.40, 0.60, 0.80, 1.0 and 2.0 mg of silver per liter
respectively.
4.3 Set the absorbance wavelength to 328.1 nm. Adjust zero using double distilled
water. Measure the absorbance directly of successive standard solutions (4.2)
and carry out in duplicate.
5. Expression of results
Plot a graph showing the variation in absorbance as a function of the silver
concentration in the standard solutions.
Using the measured absorbance of the sample read the concentration C in mg/L
from the calibration curve.
The concentration of silver in the wine is given in milligrams per liter by
0.25 x C.
It is quoted to two decimal places.
Note: Select the concentration of the solutions for the preparation of the calibration
curve. The volume of sample taken and the final volume of the liquid should be
appropriate for the sensitivity of the apparatus to be used.
OIV-MA-AS322-09 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Cadmium
Method OIV-MA-AS322-10
Type IV method
Cadmium
1. Principle
Cadmium is determined directly in the wine by graphite furnace atomic absorption
spectrophotometry.
2. Apparatus
All the glassware must be washed in concentrated nitric acid prior to use, heated to
70 to 80 C and rinsed in double distilled water.
2.1 Atomic absorption spectrophotometer equipped with a graphite furnace,
background correction and a recorder.
2.2 Cadmium hollow cathode lamp
2.3 5 l micropipettes with special tips for atomic absorption measurement.
3. Reagents
The water used must be double distilled prepared using borosilicate glass
apparatus, or water of a similar purity. All reagents must be of recognized
analytical reagent grade and, in particular, free of cadmium.
3.1 Phosphoric acid (20 = 1.71 g/mL), 85%.
3.2 Phosphoric acid solution obtained by diluting 8 mL of phosphoric acid with
water to 100 mL.
3.3 0.02 M Ethylenediaminetetraacetic acid disodium (EDTA) solution.
3.4 pH 9 buffer solution: dissolve 5.4 g of ammonium chloride in a few milliliters
of water in a 100 mL volumetric flask, add 35 mL of 25% (v/v) ammonium
hydroxide solution. Ammonium hydroxide solution, 20 = 0.92 g/mL, diluted
to 25% (v/v) and made up to 100 mL with water.
3.5 Eriochrome black T, 1% (m/m) solution in sodium chloride.
3.6 Cadmium sulfate, 3CdSO4.8H20.
The concentration of the cadmium sulfate must be verified using the
following method:
Weigh exactly 102.6 mg of the cadmium sulfate sample into a beaker with
some water and shake until dissolved; add 5 mL of the pH 9 buffer solution
OIV-MA-AS322-10 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Cadmium
and approximately 20 mg of Eriochrome black T. Titrate with the EDTA
solution (3.3) until the indicator begins to turn blue.
The volume of EDTA added must be equal to 20 mL. If the volume is slightly
different, correct the weighed test portion of cadmium sulfate used in the
preparation of the reference solution accordingly.
3.7 Cadmium reference solution at 1 g per liter.
Use of a standard commercial solution is preferred. Alternatively this solution
may be prepared by dissolving 2.2820 g of cadmium sulfate in water and
making up to one liter. Keep the solution in a borosilicate glass bottle with a
ground glass stopper.
4. Procedure
4.1 Preparation of the sample
The wine is diluted 1/2 (v/v) with the phosphoric acid solution (3.2).
4.2 Preparation of calibration standards
Using the cadmium reference solution, prepare successive dilutions 2.5, 5, 10
and 15 g of cadmium per liter respectively.
4.3 Determination
4.3.1 Furnace Programming (for guidance only):
Dry at 100C for 30 seconds
Mineralize at 900 C for 20 seconds
Atomize at 2250 C for 2 to 3 seconds
Nitrogen flow (flushing gas): 6 liters/minute
Note: At the end of the procedure, increase the temperature to 2700 C to
clean the furnace.
4.3.2 Atomic absorption measurements
Select an absorption wavelength of 228.8 nm. Set the zero on the absorbance
scale with double distilled water. Using a micropipette, introduce into the
furnace three 5 l portions of each of the solutions in the calibration range
and the sample solution to be analyzed. Record the absorbance measured.
Calculate the mean absorbance value from the results for the three portions.
5. Expression of results
5.1 Method of calculation
Draw the absorbance variation curve as a function of the concentration of
cadmium in the solutions in the calibration range. The curve is linear. Enter
the mean absorbance value of the sample solution on the calibration curve
and obtain the cadmium concentration C. The cadmium concentration
expressed in micrograms per liter of wine is equal to 2C.
OIV-MA-AS322-10 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Cadmium
BIBLIOGRAPHY
-
-
MEDINA B., Application de la spectrométrie d absorption atomique sans
flamme au dosage de quelques métaux dans les vins, Thèse Doct. en
œnologie, Bordeaux II, 1978.
MEDINA B. and SUDRAUD P., FV O.I.V 1979, nº 695.
OIV-MA-AS322-10 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Lead
Method OIV-MA-AS322-11
Lead
(Resolution Oeno 3/94)
1. Principle of the method
Lead is determined directly in wine by flameless atomic absorption
spectrophotometry.
WITHDRAWN
OIV-MA-AS322-11 : R2009
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
Method OIV-MA-AS322-12
Type II method
Criteria for the methods of quantification of lead in wine
(Resolution Oeno 7/2006)
1.1
Method Criteria Definitions
Trueness the closeness of agreement between the average value obtained from a
large series of test results and an accepted reference value
r=
Repeatability limit, the value below which the absolute difference
between 2 single test results obtained under repeatability conditions (i.e., same
sample, same operator, same apparatus, same laboratory, and short interval of
time) may be expected to lie within a specific probability (typically 95%) and
hence r = 2.8 x sr.
Standard deviation, calculated from results generated under
Sr =
repeatability conditions.
RSDr= Relative standard deviation, calculated from results generated under
repeatability conditions [(Sr/ x ) x 100], where x is the average of results over
all laboratories and samples.
R=
Reproducibility limit, the value below which the absolute difference
between single test results obtained under reproducibility conditions (i.e., on
identical material obtained by operators in different laboratories, using the
standardised test method), may be expected to lie within a certain probability
(typically 95%); R = 2.8 x sR.
SR =
Standard deviation, calculated from results under reproducibility
conditions.
RSDR = Relative standard deviation calculated from results generated under
reproducibility conditions [(SR/ x x 100]
HoR = HORRAT value: the observed RSDR value divided by the RSDR value
calculated from the Horwitz equation [1].
OIV-MA-AS322-12 : R2006
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
2
Method of analysis to be used by the laboratory and laboratory
control requirements
2.1
Requirements
Specific methods for the determination of lead in wine are not prescribed.
Laboratories shall use a method (Type II) validated to OIV requirements [2]
that fulfils the performance criteria indicated in Table 1 e.g. GFAA or ICP-MS
methods are applicable provided they meet the performance criteria outlined
below. Wherever possible, the validation shall include a certified reference
material in the collaborative trial test materials. If not an alternative estimation
of trueness should be used. Examples of suitably validated methods for the
determination of lead in wine are provided in Appendices 1 & 2.
2.2
General considerations
All apparatus which comes into contact with the sample shall be made of an
inert material (e.g. polypropylene, polytetrafluoroethylene [PTFE], etc.). The
use of ceramic materials is not advisable because of the possibility that lead
might be present. If it is not certain that the materials available are free from the
analytes in question, their use shall be assessed by means of ad hoc studies,
which should be considered as an integral part of the validation of the method
of analysis. All plastic ware including sample containers shall be acid cleaned.
If possible, equipment used for preparing samples should be reserved for lead
analyses only.
OIV-MA-AS322-12 : R2006
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
Table 1: Performance criteria for methods of analyses for lead in wine
Parameter
Applicability
Value/Comment
Suitable for determining lead in wine for official
purposes.
Detection limit
No more than one tenth of the value of the OIV limit
(expressed in g/L)
Limit of
quantification
No more than one fifth of the value of the OIV limit
(expressed in g/L) except if the value of the limit
for lead is less than 100 g/L. For the latter, no more
than two fifth of the value of the specification
HORRAT values of less or equal to 2 in the
validation collaborative trial
80% - 105% (as indicated in the collaborative trial)
Free from matrix or spectral interferences
Precision
Recovery
Specificity
Trueness
xm
< 1,96 *
sR ( lab )²  sr ( lab )² * 1  1 n
where m is the certified value of the wine reference
material and x is the average of n measurements of
lead content in this wine, within the same laboratory.
SR(lab) and Sr(lab) are standard deviations,
calculated from results within the same laboratory
under reproducibility and repeatability conditions.
2.3
Estimation of the analytical trueness and recovery calculations
Wherever possible the trueness of the analyses shall be estimated [3] by
including suitable certified reference materials in the analytical run. The analyst
shall also take due note of the „Harmonised Guidelines for the Use of Recovery
Information in Analytical Measurement‟ [4] developed under the auspices of
IUPAC/ISO/AOAC. The recovery should be approximately 100 % in which
case recovery calculations are of minor importance.
OIV-MA-AS322-12 : R2006
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
References
[1] W Horwitz, “Evaluation of Analytical Methods for Regulation of Foods
and Drugs”, Anal. Chem., 1982, 54, 67A - 76A
[2] Protocol for the design, conduct and interpretation of methodperformance studies, FV 1061, OIV, 1998
[3] ISO 5725-6:1994, 4.2.3. International Organisation for Standardisation,
case Postal 56, CH-1211, Genève 20, Switzerland.
[4] ISO/AOAC/IUPAC Harmonised Guidelines for the Use of Recovery
Information in Analytical Measurement. Edited Michael Thompson, Steven L R
Ellison, Ales Fajgelj, Paul Willetts and Roger Wood, Pure Appl. Chem., 1999,
71, 337 – 348
OIV-MA-AS322-12 : R2006
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
EXAMPLE 1
DETERMINATION OF LEAD IN WINE BY ATOMIC
ABSORPTION SPECTROMETRY
1
SCOPE AND FIELD OF APPLICATION
The method can be used for red, white, still, sparkling and fortified wines.
2
DEFINITION
The lead content of wine: the content of lead determined by this procedure
expressed as mg/L.
3
PRINCIPLE
Wine is diluted by a matrix matching cocktail and the lead concentration
measured directly by graphite furnace atomic absorption spectrometry
(GFAAS). A matrix matching mixture is added to both the wine to be
determined and the lead calibration standard solutions. This mixture contains
both GFAAS 'matrix modifiers' and wine simulating components. Their
purpose is to 'modify' the matrices so that the same shape absorption vs. time
profile is obtained from both standard solutions and sample solutions during the
graphite furnace atomisation stage.
A delayed atomisation mechanism is required e.g. L'vov platform.
The exact composition of the diluent may need to be adjusted to suit particular
models of graphite furnace instruments. Before the method is applied
experiments must be conducted to check the absorbance vs. time profiles
produced by standards and samples and necessary adjustments made to the
diluent. The instrument used must be capable of monitoring the absorbance vs.
time profile during atomisation. The profile should be such that standards and
samples perform alike and that the lead atomisation peak precedes the bulk of
the background non-specific absorption enabling the background correction
mechanism employed to operate effectively. Examples of matched profiles are
given in Annex 2.
OIV-MA-AS322-12 : R2006
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
4
REAGENTS
Chemicals should be of the highest quality available in terms of being free of
lead. Deionised distilled water, or water of equivalent purity, is to be used.
Unless otherwise indicated all solutions are prepared fresh daily.
4.1
Diluent solution
NOTE 1: The exact composition of the diluent used may need adjustment to suit
the specific model of instrument and graphite furnace employed. If problems
are experienced with the suggested modifier composition adjust the phosphate
and nitrate concentrations to give:
i) a stable element signal at the optimum ashing temperature and
ii) atomisation with a single reproducible analyte peak which is time separated
from the background signal.
Equipment with VDU facilities will allow analysts to confirm time separation of
the sample and background peaks (See Annex). The following is an example of
a technique for determining the absorbance versus time profile:
Measure the full peak width at half maximum height (FWHM) of a sample peak
and compare it to the FWHM of a calibration standard with a similar maximum
absorbance. If the peak shapes are visibly different then the composition of the
matrix modification modifier needs to be adjusted.
The following are examples of diluents utilised for:
(a) a Perkin-Elmer 3030 equipped with deuterium arc background corrector
with an HGA 500 furnace; and (b) a Thermo-Electron Video 12E equipped
with Smith-Hieftje background corrector, a CTF 188 furnace and a
FASTAC
sample deposition system.
4.1.1 Perkin-Elmer 3030 diluent:
To 187 g of water in a 250 ml plastic bottle (5.1) add 11 g ethanol (4.1.3.), 1.1 g
of glucose (4.1.4.), 1.1 g of fructose (4.1.5.) and 0.28 g of sodium chloride
(4.1.6.). Shake to dissolve the solids. Then add 22 ml nitric acid (4.1.7.) and 4.4
g ammonium dihydrogen orthophosphate (4.1.8.). Shake until all the phosphate
has dissolved. Finally add 0.88 g magnesium nitrate (4.1.9.) and shake again
until no undissolved solid remains.
OIV-MA-AS322-12 : R2006
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
4.1.2 Thermo-Electron Video 12E diluent:
As above but only 0.66 g of ammonium dihydrogen orthophosphate (4.1.8.) and
0.44 g magnesium nitrate (4.1.9.) are used.
4.1.3 Ethanol (absolute)
4.1.4. D-glucose
4.1.5. D(-)fructose
4.1.6. Sodium chloride
4.1.7. Nitric acid (concentrated)
4.1.8. Ammonium dihydrogen orthophosphate
4.1.9. Magnesium nitrate hexahydrate
4.2
10% ethanol (v/v)
To 180 ml water in a 250 ml plastic bottle (5.1.) add using a pipette 20 ml of
ethanol (4.1.3.) and shake to mix.
4.3.
Lead standard solutions
4.3.1. Lead standard solution (1000 mg/l)
4.3.2 Lead standard solution (10.00 mg/l)
Into a 100 ml volumetric flask (5.2.) pipette (5.7.) 1.00 ml of lead standard
solution (4.3.1.). Dilute to volume with water and mix thoroughly.
NOTE 2 : Check calibration of pipette immediately prior to use.
4.3.3 Lead working standard (1.00 mg/l)
Into a 100 ml volumetric flask (5.2.) weigh out 10.00 g of the lead stock
solution (4.3.2.) using a Pasteur pipette (5.3.). Wash the inside neck of the
volumetric flask with water, add 1 ml of nitric acid (4.1.7.) and make up to the
mark with water. Shake to mix thoroughly.
4.3.4. Lead calibration solutions.
OIV-MA-AS322-12 : R2006
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
The eight calibration standards are made up in universal containers (5.4.). A
range of 0 to 50 g/l is covered by the standards. They are 0.0, 2.5, 5.0, 10.0,
20.0, 30.0, 40.0 and 50.0 g/l. A ninth container is used to prepare a reagent
blank.
Rinse out the inside of each container three times with water and shake dry;
rinse the caps three times and shake dry. Leave the capped containers standing
upright for 5-10 minutes and then shake out residual liquid. Pipette (5.8) into
the 9 containers, in order: 5.00, 5.00, 4.95, 4.90, 4.80, 4.60, 4.40, 4.20 and 4.00
ml of water. Into each of the containers pipette (5.8.) 5.00 ml of 10 % ethanol
(4.2) followed by two 5.00 ml aliquots of diluent (4.1).
Into the 9 containers pipette (5.6) (5.7) in order: 0 (reagent blank), 0, 50, 100,
200, 400, 600, 800 and 1000 l of working standard (4.3.3). Cap the containers
and shake to mix the contents. Prepare fresh for each batch of samples.
4.4
1 % (v/v) nitric acid.
5.
APPARATUS
All glass and plastic ware used must be acid cleaned (soaked in 20 % nitric acid
for at least 24 hours), rinsed thoroughly with distilled water prior to use and
kept covered (with cling-film if appropriate) to prevent aerial contamination.
5.1
250 plastic bottles, with caps (for example: Nalgene or equivalent).
5.2
Volumetric flasks, 100 ml (Grade A).
5.3
Pasteur pipettes, with teats
5.4
Universal containers, 30 ml (Nunc, Sterilin or equivalent).
5.5
Glass beakers, 600 ml.
5.6
Pipette*, 40 - 200l (Labsystems Finnpipette or equivalent).
5.7
Pipette*, 200 - 1000l (Labsystems Finnpipette or equivalent).
5.8
Pipette*, 0.5 - 5.0 ml (Labsystems Finnpipette or equivalent).
5.9
Pipette*, 2.0 - 10.0 ml (Labsystems Finnpipette or equivalent).
OIV-MA-AS322-12 : R2006
8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
*NOTE 3: pipettes should be calibrated each day (of use).
5.10 Analytical balance, (+ or - 1 mg, Mettler PC440 or equivalent).
5.11 Vortex type mixer or equivalent.
5.12 Test tubes, 20 ml capacity.
5.13 Test tube racks, suitable for use with 5.12.
5.14 Container racks, suitable for use with 5.4.
5.15 Magnetic stirrer.
5.16 Magnetic follower, PTFE coated.
5.17 Pipette tips, suitable for use with 5.6, 5.7, 5.8 and 5.9
5.18 Atomic absorption spectrometer,
Atomic absorption spectrophotometer equipped with a graphite furnace,
atomisation delay cuvette, auto-injector, background corrector, and absorbance
vs. time profile monitoring facility equivalent to the following. Instrumental
conditions should be adjusted appropriately for the model used. The following
are given as examples:
(a) Atomic absorption spectrophotometer, Perkin-Elmer 3030 equipped with
deuterium arc background corrector for non-specific absorption. Lead hollow
cathode lamp operated at 12 mA. Monitor the 283.3 nm line; slit width 0.7 nm.
Graphite furnace, HGA 500 fitted with pyrolytically coated graphite tube with a
solid pyrolytic graphite L'vov platform resting inside. Use argon as the purge
gas. The furnace conditions for the HGA 500 are as follows:
OIV-MA-AS322-12 : R2006
9
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
Step
Temperature
(C)
Ramp (s)
Hold (s)
Gas
Gas flow
(mL/min)
Read (2.5s
integration)
1
200
2
1100
3
1100
4
1800
5
2400
6
20
5
60
Ar
50
20
20
Ar
50
1
2
Ar
0
0
3
Ar
0
1
6
Ar
300
1
25
Ar
300
X
Auto-sampler/injector, AS 40. 20 l injection volume, 3 injections per tray
position.
(b)
Thermo-electron Video 12E Atomic absorption spectrophotometer used
with a CTF 188 Graphite Furnace and a FASTAC sample deposition
system with the following conditions:
(c)
Step
Temperature (C)
Ramp (s)
Hold (s)
Gas
Gas flow (mL/min)
Read (2.5s integration)
1
150
0
2
Ar
50
2
350
30
0
Ar
50
3
650
15
5
Ar
0
4
100
0
5
240
0
1
4
Ar
0
X
10
Ar
300
Sample deposition 5 s, FASTAC delay time 10 s, with 3 injections per tray
position. Monitor the 283.3 nm line.
OIV-MA-AS322-12 : R2006
10
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
6.
PROCEDURE
6.1.
Preparation of wine
Shake the wine container to thoroughly mix the contents before sub-sampling.
Sparkling wines should be transferred to a clean beaker and placed in an
ultrasonic bath until gas is no longer evolved prior to use.
6.2
Measurement solutions
6.2.1 Wine samples
Into a 20 ml test tube (5.12) pipette (5.8) 2.00 ml of water, 4.00 ml of diluent
(4.1) and 2.00 ml of the sample wine. Mix thoroughly using the vortex mixer
(5.11).
6.2.2 Recovery estimates
For recovery estimate purposes pipette (5.8) into a 20 ml test tube (5.12) 1.80
ml of water, 4.00 ml of diluent (4.1), 2.00 ml of the sample wine and add using
a pipette (5.7) 0.200 ml of lead working standard (4.3.3). Mix thoroughly using
the vortex mixer (5.11).
NOTE 4: Any sample that exceeds the highest calibration standard will have to
be re- analysed using a smaller sample aliquot. Add extra 10% ethanol (4.2) to
the sample volume.
6.3.
Measurement
Determinations are carried out in batches. Each batch is to contain at least four
replicates of the reagent blank and three spiked replicates of samples for
recovery estimate purposes. The lead calibration solutions are distributed
evenly amongst the unknowns on the auto-sampler tray. Transfer the samples
and standards to the auto-sampler sample containers using a Pasteur pipette
(5.3). Discard the first filling of the container and measure the second filling (if
there is not enough sample solution, care will have to be taken to ensure that
the sample containers are scrupulously clean). Wash the Pasteur pipette four or
five times with 1% nitric acid (4.4.) between each standard and sample transfer.
OIV-MA-AS322-12 : R2006
11
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
6.4.
Quantification of lead
The mean absorbance from 3 injections is used in all cases.Construct a
calibration graph from the mean responses given by the in-batch standards.
Note the absorbances recorded by the instrument for each sample. The lead
concentration of the sample solutions are determined by comparison with the
calibration graph.
NOTE 5: It is recommended that the furnace tube and platform be replaced
every two batches or sooner if there is a marked decrease in the measured
absorbance of the standards.
7
EXPRESSION OF RESULTS
Correct the results for the average in-batch recovery.
7.1
Calculation
Obtain from the calibration graph, the lead content of all the measurement
solutions. Calculate the lead content of the wine samples and spiked wine
samples using the following calculation:
Pb concentration (mg/l) =
(Cm - Cb) x Vt
--------------------------V
m
where:
C is the mean lead concentration of the measurement solution (mg/l).
m
C is the mean measured lead concentration of the reagent blank solutions
b
(mg/l).
V is the final total volume of the measurement solution (ml).
t.
V is the volume of the wine sample taken (ml).
m
7.2
Calculation of recovery estimates
Recovery (%) =
(Cs - Ca) x Vs x 100
----------------------------S
OIV-MA-AS322-12 : R2006
12
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
where:
C is the calculated mean lead concentration of the spiked wine sample (mg/l).
s
C is the calculated mean lead concentration of the unspiked wine (mg/l).
a
V is the volume of wine to which the spike is added (ml).
s
S is the amount of spike added (g).
7.3
Calculation of recovery corrected results
Corrected Pb concentration (mg/L) =
Cw x 100
----------------R
a
where:
C is the calculated lead concentration of the wine sample (mg/l).
w
R is the average in-batch recovery (%).
a
OIV-MA-AS322-12 : R2006
13
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
ANNEX: VALIDATION STUDY
The following study was carried out to internationally agreed procedures (1)(2).
TABLE 1
SAMPLE SCHEME
Sample Code
Sample Description
5&9
Bordeaux (Sweet White)
3 & 11
Italian Chardonnay (White)
7&8
Spanish Red fortified at 260 g/l
6 & 10
Romanian Pinot Noir
2 & 12
Romanian Pinot Noir fortified with 150 g/l
1
Sample 3/11 fortified with 124 g/l
4
Sample 3/11 fortified with 134 g/l
OIV-MA-AS322-12 : R2006
14
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
TABLE II SUMMARY OF STATISTICAL PARAMETERS FOR
LEAD IN WINE COLLABORATIVE TRIAL (The results from one
laboratory were assessed as being inappropriate for inclusion in the
statistical analysis)
Sample
A
B
C
D
E
F
F1
Code
5, 9
3, 11
7, 8
6, 10
2, 12
1
4
16
16
16
n
16
15*
n (-outl)
16
15
14
16
15
16
Targ.
56
24
279
67
192
143
153
Mean
50.8
27.2
298
70.6
189
143
149
23
15
51
38
13.6
r
Sr
RSDr
8.1
16
Hor
R
5.3
19
1.0
42
SR
RSDR
15.1
HoR
1.2
30
1.1
25
8.8
28
1.2
16
24
32
8.7
11.8
18.2
3
17
10
0.2
1.1
0.7
0.7
83
57
29.8
20.3
55.2
28.2
10
29
29
19
0.5
1.2
154
9
1.4
79
0.9
KEY TO TABLES I-II
N
Initial number of laboratories
n(-outl) Number of laboratories after removal of outliers
(*)
Laboratory 17 reported <20 µg/l for test material 11. Their results
have not been included in the statistical analysis for this sample (B).
Mean
The observed mean, the mean obtained from the collaborative trial
data after removal of outliers.
Targ.
The mean observed value obtained "in-house" using ICP-MS
OIV-MA-AS322-12 : R2006
15
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
r
Repeatability limit, the value below which the absolute difference
between 2 single test results obtained under repeatability conditions (i.e., same
sample, same operator, same apparatus, same laboratory, and short interval of
time) may be expected to lie within a specific probability (typically 95%) and
hence r = 2.8 x sr.
Sr
The standard deviation of the repeatability.
RSDr
The relative standard deviation of the repeatability
100/MEAN).
(S
r
x
Hor
The observed RSD divided by the RSD value estimated from the
r
r
Horwitz equation using the assumption r=0.66R.
R
Reproducibility limit, the value below which the absolute difference
between single test results obtained under reproducibility conditions (i.e., on
identical material obtained by operators in different laboratories, using the
standardised test method), may be expected to lie within a certain probability
(typically 95%); R = 2.8 x sR.
SR
The standard deviation of the reproducibility (between laboratory
variation).
RSDR
The relative standard deviation of the reproducibility (S x
R
100/MEAN).
HoR
The observed RSD value divided by the RSD value calculated
R
R
from the Horwitz equation.
RSDR = 2(1-0.5log10C)
(where C = concentration expressed as a
decimal)
HORRAT(4) values are:
For repeatability, the observed RSDr divided by the RSDr value estimated from
the Horwitz equation using the assumption r = 0.66R.
For reproducibility, the observed RSDR divided by the RSDR value estimated
from the Horwitz equation.
8. REFERENCES
8.1 Paul A. Brereton, Paul Robb, Christine M Sargent, Helen M. Crews and
Roger Wood. Determination of Lead in Wine by Graphite Furnace Atomic
Absorption Spectrometry: Interlaboratory Study. JAOAC Int., 1997, 80, No 6,
OIV-MA-AS322-12 : R2006
16
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
1287-1297.
8.2 "Protocol for the Design, Conduct and Interpretation of Collaborative
Studies." Editor W Horwitz, Pure & Appl. Chem., Vol. 67, No., 2, pp.331-343,
1995
8.3 Horwitz W, Evaluation of Methods Used for Regulation of Foods and
Drugs, Analytical Chemistry, 1982, 57, 67A-76A.
8.4Peeler J T, Horwitz W and Albert R, Precision Parameters of Standard
Methods
of Analysis for Dairy Products, JAOAC, 1989, 72, No 5, 784-806.
OIV-MA-AS322-12 : R2006
17
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
Annex
Absorbance vs. Time profiles for the measurement of lead in wine
using a Perkin-Elmer 3030 atomic absorbtion spectrometer
with deuterium arc background correction.
0,5
Absorbance, AU
0,4
0,3
0,2
0,1
0
0
0,5
1
1,5
2
2,5
3
Time, seconds
(i)
30 ng/l wine standard
OIV-MA-AS322-12 : R2006
18
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
0,5
Absorbanc, AU
0,4
0,3
0,2
0,1
0
0
0,5
1
1,5
2
2,5
3
Time, seconds
(ii)
wine sample
Key:
þþþþþþþþþþþ
corrected absorbance,  background absorbance
OIV-MA-AS322-12 : R2006
19
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
EXAMPLE 2
DETERMINATION OF LEAD IN WINE BY ATOMIC
ABSORPTION SPECTROMETRY
1.
FIELD OF APPLICATION :
This analysis method can be applied to all types of wine, given the maximum
limit set by the O.I.V.
2. REFERENCES:
2.1. Journal Officiel des Communautés Européennes (3 octobre 1990). Méthode
de dosage du plomb dans le vin (p. 152 et 153).
2.2. Teissèdre P.L., Brun S., Médina B. (1992). Dosage du plomb dans les vins
/ Proposition de modifications à la méthode du Recueil. Feuillet Vert de
l‟O.I.V., n°928, 1997/151292.
2.3. Moreira Balio da Silva M., Gaye J., Médina B. (1996). Comparaison de six
méthodes de dosage du plomb dans les vins par absorption atomique en four
graphite. Feuillet Vert de l‟O.I.V. n° 1013, 2310/190196.
2.4. Brereton P., Robb P., Sargent C., Crews H., Wood R. (1996). Validation of
a graphite furnace atomic absorption spectrometry method for the detection of
lead in wine. Feuillet Vert de l‟O.I.V. n° 1016, 2913/230196.
2.5. Bourguignon J.B., Douet Ch., Gaye J., Médina B. (1997). Dosage du
plomb dans le vin / Interprétation des résultats de l’essai interlaboratoire.
Feuillet Vert 1055 de l‟O.I.V. n° 2456/190397.
3. PRINCIPLE:
The wine will undergo no preparations, except dilution in the case of white
sweet wines.
Adding ammonium dihydrogeno-phosphate enables the lead contained in wine
to be stable at high temperatures, which leads to eliminating interferences – and
to acting in an identical manner to the standard solution.
The atomizer is a pyrolytic graphite equipped with a platform heated by the
Joule effect.
The wavelength of the ray used is 283.3 nanometres.
The non specific absorption correction can be done by the Zeeman effect or by
using a deuterium discharge lamp.
OIV-MA-AS322-12 : R2006
20
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
The type of lead determination in wine is a direct dosage method with external
calibration.
4. REAGENTS
4.1. Demineralised water: ultra pure; with resistivity above 18 MΩ/cm.
4.2. Nitric acid: 65 % ; « suprapur » quality acid.
4.3. Ammonium dihydrogeno-phosphate NH4H2PO4 for analysis.
4.4. Lead standard solution: at 1000 µg/ml (or 1 g/l) in 2% nitric acid
(commercial solution, ready to use).
5. APPARTUS
5.1
Analytic balance (e = 1 mg).
5.2
Glass ware:
5.2.1 Volumetric flask 50, 100 ml (class A),
5.2.2 Volumetric pipette 1, 10 ml (class A),
5.2.3 Decontamination of glassware used: rinse in demineralised water; soak
at least 24 hours in a basin of 10% nitric acid; rinse two times in demineralised
water.
5.3 Atomic absorption spectrophotometer equipped with a graphite tube
atomizer for non-specific absorption correction and an auto-sampler (rinse the
sampler buckets with 10% nitric acid).
5.3.1. Pyrolytically coated graphite furnace containing an L'Vov platform
possibly tantalite (reference 9.1 – see below:
5.3.1.1 Tantalum solution: place 3 g of tantalum powder (metal tantalum with a
purity above 99.7%) in a 100 ml teflon cylindrical flask; add 10 ml of diluted
fluorhydric acid (1 + 1), 3 g of dehydrated oxalic acid and 0.5 ml of 30%
hydrogen peroxide solution; heat together carefully to dissolve metal; add
hydrogen peroxide when r reaction slows down. Add 4 g of dehydrated oxalic
acid and approximately 30 ml of demineralised water when completely
OIV-MA-AS322-12 : R2006
21
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
dissolved. Dissolve acid. Fill the solution up to 50 ml. This solution is stored in
a plastic flask.
5.3.1.2 Tantalisation of a platform: the platform is placed inside the graphite
tube. These items are placed together on a spectrophotometer atomization unit.
10 µl tantalum solution is injected on a platform using an auto-sampler. The
temperature cycle is set according to the following program: drying at 150°C
for 40 s; mineralization at 900°C for 60 s; atomization at 2600°C for 2.5 s.
Argon is used as an inert gas.
6. PROCEDURE
6.1 Test portion: The neck of the wine bottle with a tinned lead capsule must
be carefully cleaned before uncorking.
6.2 Sample preparation: In general, no preparation of wine is necessary;
samples are placed directly in the automatic sampler buckets. Cloudy wine
needs to be filtered. To prolong the utilisation period of the platforms, sweet
white wines are diluted for sugar contents between 10 to 50 g/L, dilute by 1/2;
for contents above 50 g/l, dilute by 1/4.
6.3
Preparation of solutions:
6.3.1 White dilution:
The solution is used as an additional volume to be injected and is made up of
demineralised water containing 1 % nitric acid (4.2.).
6.3.2. Matrix modifying agent:
Into a 50 ml flask (5.2.1) introduce 3 g of ammonium dihydrogeno-phosphate
(4.3.); dissolve and fill with demineralised water (4.1.).
6.3.3. 10 mg/ of lead solution:
Into a 100 ml flask (5.2.1) place 1 ml of 1 g/l (4.4.) solution; add 1 % nitric acid
(4.2.); fill to volume with demineralised water (4.1.). This solution can be kept
one month at a temperature + 4°C.
6.3.4 100 µg/L lead solution:
Into a 100 ml flask (5.2.1) place 1 ml of 10 mg/l (6.3.3.) lead solution; fill to
volume with demineralised water (4.1.). This solution must be prepared every
analysis day.
OIV-MA-AS322-12 : R2006
22
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
6.3.5 Calibration scale (for information purposes): 0 ; 16.7 ; 33.3 et 50 µg/l
(see Table II).
6.4
Calibration and determination:
6.4.1 Spectrometric measurement:
6.4.1.1
wavelength: 283.3 nm;
6.4.1.2
slot with: 0.5 nm;
6.4.1.3
hollow cathode lamp intensity: 5 mA ;
6.4.1.4 correction continum: by Zeeman or deuterium effect;
6.4.1.5 introduction of standards heated and samples in a graphite furnace
using an automatic sampler. The flushing liquid is made up of 500 ml of
demineralised water containing a drop of Triton X 100.
Note: in order to inject at 90°C on a platform, the furnace temperature should
be regulated to approximately 150°C.
6.4.1.6 signal measurement: peak height;
6.4.1.7 Duration of measurement: 3 seconds;
6.4.1.8 Number of measurements by standard or sample: 2
Note: the average of these two determinations constitutes the trial result. If the
variation coefficient for the two determinations is greater than 15 %, the two
other determinations must be re-done.
6.4.1.9. Furnace parameters (for information purpose): see Table I.
Table I – Furnace parameters
Temperature
(in °C)
150
750
750
750
2400
2400
2400
40
For determination of lead in wine
Duration
Gas type
Gas flow
(in s)
(in l/mn)
60
argon
3.0
10
argon
3.0
30
argon
3.0
2
argon
0
1
argon
0
2
argon
0
2
argon
3.0
20
argon
3.0
OIV-MA-AS322-12 : R2006
lecture du
signal
oui
oui
23
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
6.4.1.10. Automatic sampler parameters (for information purposes): see Table
II.
Table II – Sampler parameters for the
dosage of lead in wine
Analysis:
volumes injected in µl
sample
Pb solution
Calibration blank
Standard 1
Standard 2
Standard 3
Sample
0
0
0
0
2
100 µg/l
0
1
2
3
0
"white"
dilution
5
4
3
2
3
Matrix
modifier
1
1
1
1
1
6.4.2 Tracing of calibration curve: the automatic distributor cycle enables the
preparation of standards from 100 µg/l (Table II) lead solutions. The calibration
graph is drawn up: absorbency according to lead concentration in micrograms
per litre.
7. EXPRESSION OF RESULTS
7.1 Concentration of lead in injected solution:
calibration curve (6.4.2.).
This is obtained from
7.2 Concentration of lead in wine: This is calculated by multiplying by 3 the
result given in 7.1. (2 µl of solution injected for a final volume of 6 µl on the
platform). Take into account the possible dilution of wines (in the case of sweet
white wines).
7.3 Result: is expressed in milligrams of lead per liter of wine (mg/l), to two
digits.
8. INTER-LABORATORY TRIALS
A "double-blind" trial was carried out on 8 different wines obtained from
mixtures of Bordeaux wines: two red wines (R1 and R2), two rosé wines (Ro1
and Ro2), two dry white wines (Bs1 and Bs2) and two sweet white wines (D1
and D2). Eleven Spanish, Portuguese, Moroccan and French laboratories
participated by determining lead in 16 samples received.
OIV-MA-AS322-12 : R2006
24
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
8.1
Presentation of 8 wine samples:
Table III: Characteristics of wine used in interlaboratory trials
Wine
Type
T.A.V.
(% Vol.)
R1
R2
Ro1
Ro2
Bs1
Bs2
D1
Red
Red
Rosé
Rosé
Dry white
Dry white
Sweet
white
Sweet
white
D2
11,86
12,54
12,23
11,43
11,65
12,32
12,94
Total
acidity
(g/l H2SO4)
4,43
3,77
5,30
4,88
4,62
4,57
3,72
Volatile
acidity
(g/l H2SO4)
1,57
0,34
0,44
0,45
0,37
0,31
0,67
Reducing
sugar
(g/l)
1,2
1,5
1,2
1,1
2,2
0,9
76,4
12,66
4,70
0,45
62,8
OIV-MA-AS322-12 : R2006
25
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
8.2
Statistics of results:
Table IV: Statistical analysis of inter-laboratory trial results
Wine sample
Double-blind repetitions
Initial number
of laboratories
Number of laboratories
After elimination of large
variances
Average (µg/l)
Repeatability limit
r
Standard deviation of
repeatability
Sr
Relative standard deviation
of reproducibility
RSDr (en %)
Horrat value(Hor):
Observed RSDr / RSDr
Horwitz
Reproducibility limit
R
Standard deviation of
reproducibility
SR
Relative standard deviation
of reproducibility
RSDR (en %)
Horrat values (HoR):
Observed RSDR / RSDR
Horwitz
R1
R2
Ro1
Ro2
Bs1
Bs2
D1
D2
C&K F&I D&G J&L B&H P&N A& M&
E
0
11
11
11
11
11
11
11
11
11
10
11
11
10
10
11
10
44
162
28
145
52
138
60
145
18
12
7
17
6
13
28
7
6,4
4,3
2,5
6,1
2,1
4,6
10
2,5
14,5
2,8
9,2
4,2
4,2
3,4
16,5
1,8
0,6
0,1
0,3
0,2
0,2
0,2
0,7
0,1
34
105
23
86
30
101
86
144
12,3
37,5
8,2
30,8
10,7
35,9
30,6
51,6
28
23,1
29,3
21,2
20,6
26
51
35,6
1,1
1,1
1,1
1
0,8
1,2
2,1
1,7
Out of the 11 laboratories which participated in the trial, 7 declared that they
had followed the proposed method and 4 modified some of the parameters.
OIV-MA-AS322-12 : R2006
26
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
9. METHOD PERFORMANCES AND QUALITY CONTROL
9.1 Detection limit: This is determined from a series of 20 blank analytical
repetitions and is equal to 3 standard deviations. In the case of the proposed
method a series of 20 blank analytical measurements resulted in: average = 1,29
µg/l ; standard deviation = 0,44 µg/l ; detection limit = 1,3 µg/l .
9.2 Limit of quantification: This is equal to 3 times the detection limit. In the
case of the proposed method, the limit of quantification is 4 µg/l (3 * 1,32 =
3,96).
9.3 Trueness: The confidence interval for the average of a series of results is
compared to the reference material data.
Three reference materials are used including: red wine, dry white wine, sweet
white wine for which lead concentrations are certified by the B.C.R. (Bureau
Communautaire de Référence) in 1992.
Table V. Trueness of the method
Lead
concentration
(µg/l)
9.4
Certified value
(B.C.R. 1992)
Average value
(series: 10 results)
Red wine
BCR E
Dry white
wine
BCR C
Sweet white
wine
BCR D
36,1 ± 4,9
65,1 ± 9,1
132,4 ± 32
41,0 ± 3,8
66,0 ± 4,4
128,3 ± 14,1
Control card
A control card can be drawn up for each reference material used. Control limits
are equal to: +/- 2 SRintra (SR intra: reproductibility standard deviation).
OIV-MA-AS322-12 : R2006
27
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Criteria for the methods of quantification of lead
Control card
CarteLead
de
of BCRcontrôle
E
Plomb du BCR E
80
Lead in µg/l
plomb en µg/l
70
limites
: + =2SR
= 63,9 µg/l
Limits : + 2SR
63.9 µg/l
2SR
=
28,7 µg/l
- 2SR = 28.7 µg/l
60
50
40
30
20
15/12/95
26/06/96
29/04/97
08/12/97
09/10/98
date
moyenne
: 46,3 µg/l
Mean : 46.3 µg/l
10.
BIBLIOGRAPHY
10.1. Zatka V. (1978). Treated graphite atomizer tubes for atomic absorption
spectrometry. Analytical Chemistry, vol. 50, n°3.
10.2. US Bureau of Alcohol, Tobacco and Firearms (1991). Analysis of lead in
wines and related products by graphite furnace atomic absorption
spectrometry. Note d‟information de l‟O.I.V. du 21 août 1991 : « Plomb dans
les vins aux U.S.A. ».
10.3. Mindak W.R. (1994). Determination of lead in table wines by graphite
furnace atomic absorption spectrometry. Journal of A.O.A.C. International,
vol. 77, n°4, p. 1023-1030.
10.4. Médina B. (1994). Apport de nouvelles techniques au dosage des métaux
dans les vins. Congrès des Œnologues de France à Bordeaux.
10.5. Norme française NF ISO 5725-2 (1994). Application de la statistique :
Exactitude (justesse et fidélité) des résultats et méthodes de mesure.
10.6. Jorhem L., Sundström B. (1995). Direct determination of lead in wine
using graphite furnace AAS. Atomic Spectroscopy, September/October 1995.
OIV-MA-AS322-12 : R2006
28
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic - Atomic absorption
Method OIV-MA-AS323-01A
Type IV method
Determination of arsenic in wine
by atomic absorption spectrometre
(Resolution Oeno 14/2002)
1. PRINCIPLE
After evaporating ethyl alcohol and reducing the arsenic V in arsenic III, wine
arsenic is measured by hydride generation and by atomic absorption
spectrometry.
2. EQUIPMENT
2.1. Glass ware:
2.1.1. Graduated flask 50, 100 ml (class A)
2.1.2. Graduated pipettes 1, 5, 10, 25 ml (class A)
2.2. Water bath at 100°C
2.3. Filters without ashes
2.4. Spectrophotometer :
2.4.1. Atomic absorption spectrophotometer
2.4.2. Instrumental parameters
2.4.2.1. Air-acetylene oxidising flame
2.4.2.2 Hollow cathode lamp (arsenic)
2.4.2.3. Wave length: 193.7 nm
2.4.2.4. Split width: 1.0 nm
2.4.2.5. Intensity of hollow cathode lamp: 7 mA
2.4.2.6. Correction of non-specified absorption with a deuterium
lamp
2.5. Accessories:
2.5.1. Hydride absorption cell, placed on an air-acetylene burner.
2.5.2. Vapour generator (liquid gas separator)
2.5.3. Neutral gas (argon)
OIV-MA-AS323-01A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic - Atomic absorption
Hydride absorption cell
Placed on an air acetylene burner
Optical path through hollow cathode lamp
Drying tube
Peristaltic
pump
Reaction
loop
Reference or sample
Hydrochloric acid Sodium 10%
borohydride
Liquid gas
separator
Towards sink
Flow controller
Neutral gas (argon)
Figure 1. Hydride generator.
3. REAGENTS
3.1. Ultra-pure demineralised water
3.2. Ultra-pure 65% nitric acid
3.3. Potassium iodide (KI)
3.4. 10% . Potassium iodide (m/v)
3.5. Concentrated hydrochloric acid (R)
3.6. 10% Hydrochloric acid (R)
3.7. Sodium borohydride (NaBH4)
3.8. Sodium hydroxide (NaOH)
3.9. 0.6% Sodium borohydride (containing sodium hydroxide: 0.5% (m/v))
3.10. Calcium Chloride CaCl2 (used as a drying agent)
3.11. 1 g/l Arsenic stock solution prepared in the following manner :
dissolve 1.5339 g of AS2O5 in demineralised water, adjust to 1 l.
OIV-MA-AS323-01A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic - Atomic absorption
3.12. 10 mg/l Arsenic solution: place 1 ml of stock solution (3.11.) in a
100 ml flask (2.1.1.) ; add 1 % nitric acid (3.2.) ; fill up to volume with
demineralised water (3.1.).
3.13. 100 µg/l Arsenic solution: place 1 ml of 10 mg/l arsenic solution
(3.12.) in a 100 ml flask (2.1.1.) ; fill up to volume with demineralised
water (3.1.).
3.14. Set of callibration standards: 0, 5, 10, 25 µg/l
Successively place 0, 5, 10, 25 ml of 100 µg/l arsenic solution (3.13.) in 4
100 ml flasks (2.1.1.) ; add 10 ml of 10% potassium iodide to each flask
(3.4.) and 10 ml of concentrated hydrochloric acid (3.5.) ; leave for 1 hour,
fill up to 100 ml with demineralised water.
4. SAMPLE PREPARATION
25 ml of water is evaporated over a 100 °C water bath. This is then brought to 50
ml in the presence of 5 ml of 10% potassium iodide and 5 ml of concentrated
hydrochloric acid; leave for 1 hour; filter on an ashless filter.
Make a blank reference sample.
5. DETERMINATION
The peristaltic pump sucks in the borohydride solution, the 10% hydrochloric acid
solution and the sample solution.
Present the calibration standards in succession (3.14.); take an absorbency reading
for 10 seconds; take two readings; the operating software establishes a calibration
curve (absorbency according to concentration of arsenic in µg/l).
Then present the samples (4) ; the software establishes the sample’s arsenic
concentration in µg/l; deduct the arsenic concentration in the wine in µg/l taking
into account that the solution be diluted by 1 / 2 .
6. QUALITY CONTROL
Quality control is assured by placing a control sample of internal quality (*) in a
regular manner in 5 samples, or after the set of calibration solutions, or in the
middle of a series or at the end the measurement.
Two deviation types are accepted compared to known value.
(*) Samples from the Bureau Communautaire de Référence (Community Bureau of
reference): red wine, dry white wine and sweet white wine.
OIV-MA-AS323-01A : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic - Atomic absorption
7. BIBLIOGRAPHY
Varian Techtron, 1972. Analytical methods for flame spectroscopy.
Hobbins B., 1982. Arsenic Determination by Hydride Generation. Varian
Instruments at Work.
Le Houillier R., 1986. Use of Drierite Trap to Extend the Lifetime of Vapor
Generation Absorption Cell. Varian Instruments at Work.
Varian, 1994. Vapor Generation Accessory VGA-77.
OIV-MA-AS323-01A : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic
Method OIV-MA-AS323-01B
Type IV method
Arsenic
(Resolution Oeno 377/2009)
1. Principle
After mineralization, using sulfuric and nitric acids, arsenic V is reduced to
arsenic III by means of potassium iodide in hydrochloric acid and the arsenic
is transformed into arsenic III hydride (H3As) using sodium borohydride. The
arsenic III hydride formed is carried by nitrogen gas and determined by
flameless atomic absorption spectrophotometry at high temperature.
2. Method
2.1 Apparatus
2.1.1 Kjeldahl flask (borosilicate glass)
2.1.2 Atomic absorption spectrophotometer equipped with arsenic hollow
cathode lamp, hydride generator, background corrector and a chart recorder.
The hydride generator includes a reaction flask (which can eventually be put
onto a magnetic stirrer) connected by a tube to a nitrogen gas supply (flow
rate: 11 L/min) and by a second tube, to a quartz cell which can be brought to
a temperature of 900 oC. The reaction flask also has an opening for the
introduction of the reagent (borohydride).
2.2 Reagents
All reagents must be of recognized analytically pure quality, and in particular
free of arsenic. Double distilled water prepared using a borosilicate glass
flask or water of similar purity should be used.
2.2.1 Sulfuric acid (20= 1.84 g/mL) arsenic free
2.2.2 Nitric acid (20= 1.38 g/mL) arsenic free
2.2.3 Hydrochloric acid (20= 1.19 g/mL), arsenic free
2.2.4 10% (m/v) Potassium iodide solution
2.2.5 2.5% (m/v) Sodium borohydride solution obtained by dissolving 2.5 g of
sodium borohydride in 100 mL of 4 % (m/v) of sodium hydroxide solution.
This solution must be prepared at the time of use.
2.2.6 Arsenic reference solution 1 g/L. Use of a commercial standard arsenic
solution is preferred.
OIV-MA-AS323-01B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic
Alternatively this solution can be prepared in a 1000 mL volumetric flask, by
dissolving 1.320 g of arsenic III trioxide As2O3 in a minimal volume of 20 %
(m/v) sodium hydroxide. The solution is then acidified with hydrochloric
acid, diluted 1/2, and made up to 1 liter with water.
2.3 Procedure
2.3.1 Mineralization
Place 20 mL of wine in a Kjeldahl flask, boil and reduce the volume by half to
eliminate alcohol. Allow to cool. Add 5 mL sulfuric acid, and slowly add 5
mL nitric acid and heat. As soon as the liquid turns brown, add just enough
nitric acid, dropwise, to lighten the liquid while simmering. Continue until
the color clears and white sulfur trioxide fumes are formed above the solution.
Allow to cool, add 10 mL distilled water, bring back to the boil and simmer
until nitrous oxide and sulfur trioxide fumes are no longer produced. Allow to
cool and repeat the operation.
Allow to cool and dilute the sulfuric acid residue with a few milliliters of
distilled water. Quantitatively transfer the solution into a 40 mL flask, and
rinse the flask with water, combine with the diluted residue and make up to
the mark with distilled water.
2.3.2 Determination
2.3.2.1 Preparation of the solution
Place 10 mL of the mineralization solution (2.3.1) into the hydride generator
reactor flask. Add 10 mL hydrochloric acid, 1.5 mL potassium iodide solution,
then switch on the magnetic stirrer and the nitrogen gas (flow rate: 11
L/minute). After 10 sec, add 5 mL of sodium borohydride solution. The
hydride vapor obtained is immediately carried to the measurement cell (at a
temperature of 900C) by nitrogen carrier gas, where dissociation of the
compound and arsenic atomization occurs.
2.3.2.2 Preparation of standard solutions
From the arsenic reference solution (2.2.6), prepare dilutions having
concentrations of 1, 2, 3, 4 and 5 micrograms of arsenic per liter respectively.
Place 10 mL of each of the prepared solutions into the reactor flask of the
hydride generator and analyze according to 2.3.2.1.
2.3.2.3 Measurements
Select an absorption wavelength of 193.7 nm. Zero the spectrophotometer
using double distilled water and carry out all determinations in duplicate.
Record the absorbance of each sample and standard solution. Calculate the
average absorbance for each of these solutions.
2.4 Expression of results
OIV-MA-AS323-01B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic
2.4.1 Calculation
Plot the curve showing the variation in absorbance as a function of the arsenic
concentration in the standard solutions. The relationship is linear. Note the
average absorbance of the sample solutions on the graph and read the arsenic
concentration C.
The arsenic concentration in wine, expressed in micrograms per liter is given
by: 2 C.
BIBLIOGRAPHY
JAULMES P. et HAMELLE G., Trav. Soc. Pharm. Montpellier, 1967, 27, no 3,
213-225.
JAULMES P., F.V., O.I.V., 1967, no 238
MEDINA B et SUDRAUD P., F.V., O.I.V., 1983, no 770.
OIV-MA-AS323-01B : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic
B
A
Glass wool
mm
Fig.1: Apparatus used in the limit
test of arsenic
80 mm
Fig 2: Apparatus used in the determination of
arsenic
OIV-MA-AS323-01B : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Arsenic
Method OIV-MA-AS323-01C
Arsenic
(Resolution Oeno 377/2009)
1. Principle
After mineralization using sulfuric and nitric acids, arsenic V is reduced to
arsenic III using tin II chloride. The arsenic is then converted into arsenic III
hydride by the action of the hydrogen produced. Arsenic III hydride is
detected by reaction with mercury II bromide (limit sample).
WITHDRAWN
OIV-MA-AS323-01C : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total nitrogen (Dumas method)
Method OIV-MA-AS323-02A
Type II method
Quantification of total nitrogen according
to the Dumas method
(Musts and Wines)
(Resolution Oeno 13/2002)
1 - FIELD OF APPLICATION
This method can be applied to the analysis of total nitrogen in musts and
wine within the range of 0 to 1000 mg/l.
2 - DESCRIPTION OF THE TECHNIQUE
2.1 - Principle of the Dumas method
The analysis of total nitrogen in an organic matrix can be carried out using
the Dumas method (1831). This involves a total combustion of the matrix
under oxygen. The gases produced are reduced by copper and then dried,
while the CO2 is trapped. The nitrogen is then quantified using a universal
detector.
2.2 - Principle of the analysis (Figure n° 1)
- Injection of the sample and oxygen in the combustion tube at 940°C (1) ;
- « Flash » Combustion (2) ;
- The combustion of the gathering ring (3) brings the temperature
temporarily up to 1800°C ;
- Complementary oxidation and halogen trappings on silver cobalt and
granular chromium sesquioxide (4) ;
- Reduction of nitrogen oxides in N2 and trapping sulphur components and
excess oxygen by copper at 700°C (5) ;
- Gases in helium include: N2, CO2 and H2O (6) ;
- Trapping unmeasured elements: H2O using anhydrone (granular anhydrous
magnesium perchlorate) (7) and CO2 by ascarite (sodium hydroxide on silica)
(8) ;
- Chromatography separation of nitrogen and methane possibly present
following very large trial uptake (9) ;
- Catharometer detection (10) ;
- Signal gathering and data processing (11).
OIV-MA-AS323-02A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total nitrogen (Dumas method)
11
1
1
1
1
1
6
6
He
11
10
1
0
22
77
33
55
88
99
44
Figure 1 : Diagram of analysis principle
3 – Reagents and preparation of reactive solutions
3.1 - Nitrogen (technical quality) ;
3.2 - Helium (purity 99.99994%) ;
3.3 – Chromium oxide (chromium sesquioxide me in granules) ;
OIV-MA-AS323-02A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total nitrogen (Dumas method)
3.4 – Cobalt Oxide (silver granule cobalto-cobaltic oxide ) ;
3.5 – Quartz wool ;
3.6 - Copper (reduced copper in strings) ;
3.7 - Ascarite (sodium hydroxide on silica) ;
3.8 - Anhydrone (granular anhydrous magnesium perchlorate) ;
3.9 - Oxygen (purity 99.995%) ;
3.10 - Atropine ;
3.11 – Glumatic-hydric chloride acid;
3.12 – Demineralised water;
3.13 – Tin boat.
4 - Apparatus
4.1 - Centrifuge with 25 ml pots;
4.2 – Nitrogen analyser;
4.3 – Metallic crucible;
4.4 - Quartz reaction tube (2) ;
4.5 – Precision balance between 0.5 mg and 30 g at  0.3 mg ;
4.6 – Boat carrier;
4.7.- Furnace;
4.8 – Apparatus for folding boats;
4.9 – Sample changer;
4.10 – Computer and printer.
5 - SAMPLING
Degas by nitrogen bubbling (3.1) for 5 to 10 mn, sparkling wine. The musts
are centrifuged (4.1) for 10 mn at 10°C, at 4200 g.
6 – OPERATING INSTRUCTIONS
- Open the apparatus programme (4.2 and 4.10) ;
- Put the heating on the apparatus (4.2).
6.1 – Principle analytical parameters
Nitrogen analyser (4.2) under the following conditions:
gas carrier: helium (3.2) ;
metallic crucible (4.3) to be emptied every 80 analyses ;
OIV-MA-AS323-02A : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total nitrogen (Dumas method)
oxidation
tube (4.4), heated to 940° C, containing chromium oxide (3.3) and
cobalt oxide(3.4) held back by quartz wool (3.5). The tube and reagent
set must be changed every 4000 analyses ;
 reduction tube (4.4), heated to 700° C, containing copper (3.6) held
back by the quartz wool (3.5). The copper is changed every 450
analyses;

absorption tube, containing 2/3 of ascarite (3.7) and 1/3 anhydrone
(3.8). the ascarite which is taken in block is eliminated and replaced
every 200 analyses. The absorbers are completely changed once a
year.

The more organic matter to be burned, the more oxygen is needed:
the oxygen sampling valve (3.9) is 15 seconds for musts and 5
seconds for wine.
NOTE : The metals are recuperated and sent to a centre for
destruction or specialised recycling.
6.2 - Preparation of standard scale
Prepare two samples of atropine (3.10) between 4 to 6 mg. Weigh
them (4.5) directly with the boat. The calibration scale goes through 3
points (origin = empty boat).
6.3 – Preparation of internal standards
Internal standards are used regularly in the beginning and in the middle of
analyses.
 Internal checks are carried out using glumatic acid in the form of
hydrochloride at 600 mg N/l in demineralised water (3.12).
Molar mass of glumatic acid = 183.59
Molar mass of nitrogen = 14.007
18359
.  0.6
14.007 = 7.864 g/l

Weigh (4.5) 7.864 g of glumatic acid (3.11) and dilute in
demineralised water (3.12) qsp/l, to obtain a 600 mg N/l solution.
This solution is diluted by 50% to obtain a 300 mg N/l solution,
which is diluted by 50% again to obtain 150 mg/l solution.
OIV-MA-AS323-02A : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total nitrogen (Dumas method)
6.4 - Preparation of samples:
6.4.1 – In a boat (3.13), weigh (to the nearest 0.01 mg) 20 µl of must or
200 µl of wine with a precision balance (4.5). Repeat this
procedure three times per sample;
6.4.2 – Write down the mass
6.4.3 – Place the boats progressively in the boat carrier (4.6) ;
6.4.4 – Place the boats in the furnace (4.7) set at ~ 60° C, until the
liquid has completely evaporated (this requires at least one
hour) ;
6.4.5 – Fold and crush the boats with an appropriate apparatus (4.8), put
them in the changer (4.9) in number order.
7 - EXPRESSION OF RESULTS
Results are expressed in g/l to the fourth decimal.
8 – CHECKING RESULTS
Splicing by mass, temperature, and volume.
9- PERFORMANCE CHARACTERISTICS OF THE METHOD
Number of
laboratories
11
Average contents
Repeatability
Reproductibility
591 mg/l
43 mg/l
43 mg/l
10 - BIBLIOGRAPHY
Dumas A. (1826) : Annales de chimie, 33,342.
Buckee G.K. (1994) : Determination of total nitrogen in Barley, Malt and Beer
by Kjeldahl procedures and the Dumas combustion method. Collaborative
trial. J. Inst. Brew., 100, 57-64.
OIV-MA-AS323-02A : R2009
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total nitrogen
Method OIV-MA-AS323-02B
Type IV method
Total Nitrogen
(Resolution Oeno 377/2009)
1. Principle
The sample is wet ashed using sulfuric acid in the presence of a catalyst. The
ammonia liberated by sodium hydroxide is determined titrimetrically.
2. Apparatus
2.1 Digestion apparatus
300 mL Kjeldahl flask. Place on a metal heating mantle. Appropriate stand to
hold this apparatus, the neck bent at 45 degrees.
2.2 Distillation apparatus
1 liter round bottomed flask, fitted with a small rectifying column 30 cm long
by 2.5 cm diameter or any other equivalent apparatus. The vapor emitted from
the end of this apparatus enters into the top part of the cylindrical condenser,
held vertically, of 30 cm length and 1 cm internal diameter. The condensed
liquid is brought to the receiving conical flask by a drawn-out tube placed at
the bottom – alternatively one can use a steam distillation apparatus such as
described in Volatile Acidity, or any other apparatus relating to the test
described in paragraph "Blank tests or sample tests".
3. Reagents
3.1 Sulfuric acid free of ammonia (20 = 1.83 - 1.84g/mL)
3.2 Benzoic acid
3.3 Catalyst:
10 g
Copper sulfate, CuSO4, .................................................
Potassium sulfate, K2SO4, .............................................
100 g
3.4 30%Sodium hydroxide solution. Sodium hydroxide (20 =1.33 g/mL) diluted
30% (m/m).
3.5 0.1 M Hydrochloric acid solution
3.6 Indicator:
Methyl red ............................................…………………..……
100 mg
Methylene blue ......……..................……………………….………
50 mg
Ethanol (50%) .............................................…………………
100 mL
3.7 Boric acid solution:
OIV-MA-AS323-02B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total nitrogen
Boric acid ................................................……………………
40 g
Water to ......................................…………………………….…..
1000 mL
This solution will become pink by adding 5 drops of methyl red and 0.1 mL or
more 0.1 M hydrochloric acid solution.
3.8 Ammonium sulfate solution:
Ammonium sulfate (NH4) 2SO4 ...........………….…………… 6.608 g
Water to .....................................................………….………
1000 mL
3.9 Tryptophan, C11H12O2N2, (this substance contains in theory 13.72 g of
nitrogen per 100 g)
4. Procedure
Place in the 300 mL Kjeldahl flask (2.1), 25 mL of wine, 2 g benzoic acid (3.2)
and 10 mL sulfuric acid (3.1). Add 2 to 3 g of catalyst. With the flask placed on a
metal disc mantle (2.1) and with the neck inclined at 45 degrees, heat until a clear
color is obtained. Then heat for another 3 minutes.
After cooling, carefully transfer the contents of the Kjeldahl flask to a 1 liter round
bottomed flask containing 30 mL water. Rinse the Kjeldahl flask several times
with water and add washings to the round-bottomed flask. Cool the flask; add 1
drop of 1% phenolphthalein solution and a sufficient quantity of 30% sodium
hydroxide solution (3.4) to ensure the solution is alkaline (40 mL approximately)
making sure to cool the flask constantly during this addition. Distil 200 - 250 mL
into a flask containing 30 mL of 40 g/L boric acid solution.
Titrate the distilled ammonia in the presence of 5 drops of indicator (3.6) using 0.1
M hydrochloric acid solution.
Note: A control trainer by vapor can be used as described in the Chapter on
volatile acidity to obtain a quick ammonia distillation. In this case, successively
place 40 to 45 ml of 30% sodium hydroxide liquor and 50 to 60 ml of previously
diluted for 10 minutes contents of the Kjeldahl flask before introducing into the
mixer.
5. Calculation
The total nitrogen, in g/L, contained in the wine is given by: 0.56 x n where n is
the volume of 0.1 M hydrochloric acid.
6. Blank tests and sample tests
All distillation apparatus used to determine ammonia must satisfy the following tests:
a) Place in a distillation flask 40 - 45 mL of sodium hydroxide solution, 50 mL
water, 2 g benzoic acid, 5 g potassium sulfate and 10 mL sulfuric acid diluted
to 50 mL. Distil 200 mL and collect the distillate in 30 mL of 40 g/L boric acid
solution, to which 5 drops of indicator (3.6) are added. A change of color of
OIV-MA-AS323-02B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Total nitrogen
the indicator must be obtained by adding 0.1 mL of 0.1 M hydrochloric acid
solution.
b) Under similar conditions distill 10 mL of 0.1 M ammonium sulfate solution. In
this case, between 10.0 and 10.1 mL of 0.1 M hydrochloric acid solution, must
be used to change the color of the indicator.
c) The complete method (wet ashing and distillation) is checked using 200 mg
tryptophan as the initial sample. Between 19.5 to 19.7 mL of 0.1 M
hydrochloric acid must be used to obtain the change of color.
OIV-MA-AS323-02B : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Boron
Method OIV-MA-AS323-03
Type IV method
Boron
Rapid Colorimetric Method
1. Principle
The alcohol content of the wine is removed by reducing the volume by half by
rotary evaporation. The wine is then passed through a column of
polyvinylpolypyrrolidone, which retains the coloring agents. The eluate is
collected quantitatively and the boron concentration determined by complexation
with azomethine H at pH 5.2 followed by spectroscopic analysis at 420 nm.
2. Apparatus
2.1. Rotary evaporator
2.2. Spectrophotometer capable of measuring absorbance wavelengths between
300 and 700 nm
2.3. Cells of 1 cm optical path
2.4.
Glass column of 1 cm internal diameter and 15 cm in length containing an
8 cm layer of polyvinylpolypyrrolidone.
3. Reagents
3.1. Azomethin H (4-hydroxy-5-(2-hydroxybenzylideneamino)2,7-napthalenedisulfonic acid)
3.2. Azomethin H solution
Place 1 g of azomethin H and 2 g of ascorbic acid in a 100 mL volumetric
flask and add 50 mL double distilled water. Warm slightly to dissolve and
make up to the mark with double distilled water. The reagent is stable for
2 days if kept cold.
3.3. Buffer solution pH 5.2
Dissolve 3g of EDTA (disodium salt of ethylenediaminetetraacetic acid) in
150 mL of double distilled water. Add 125 mL acetic acid (20 = 1.05 g/mL)
and 250 g of ammonium acetate, NH4CH3COO, and dissolve. Check the pH
with a pH meter and adjust if necessary to pH 5.2.
OIV-MA-AS323-03 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Boron
3.4. Boron stock standard solution, 100 mg/L
Use of a commercial standard solution is preferable. Alternatively this
solution can be prepared by dissolving 0.571 g of boric acid, H3BO3, dried
beforehand at 50 oC until constant weight, in 500 mL double distilled water
and made up to 1 liter.
3.5. Boron standard solution, 1 mg/L
Dilute the stock solution, 100 mg/L (3.4) 1/100 with double distilled water.
3.6. Polyvinylpolypyrrolidone or PVPP (see International Enological Codex)
4. Procedure
Eliminate alcohol from 50 mL of wine by concentration to half the original volume
in a rotary evaporator at 40oC and make up to 50 mL with double distilled water.
Take 5 mL of this solution and pass it through the PVPP column (2.4). The
coloring agents are completely retained. Collect the eluate and the rinsing waters
from the column and place in a 50 mL volumetric flask and make up to the mark
with water.
The colorimetric determination is performed in a volume of 5 mL of eluent placed
in a 25 mL volumetric flask; dilute to approximately 15 mL with double distilled
water and add the following (stirring after each addition):
5 mL of azomethin H solution (3.2)
4 mL of pH 5.2 buffer solution (3.3)
Make up to 25 mL with double distilled water.
Wait 30 min and determine the absorbance As, at 420 nm. The zero of the
absorbance scale is set using distilled water.
Use a blank consisting of 5 mL of azomethin H solution and 4 mL of pH 5.2 buffer
solution in 25 mL of double distilled water. Wait 30 min and read the absorbance
Ab under the same conditions. The absorbance must be between 0.20 and 0.24; a
higher absorbance demonstrates boron contamination in the water or the reagents.
Preparation of the calibration curve
In 25 mL volumetric flasks, place 1 to 10 g of boron, corresponding to 1 to 10 mL
of boron standard solution 1 mg/L (3.5) and continue as indicated in 4.0. The
calibration graph representing the net absorbance (As-Ab) in relation to the
concentration is a straight line passing through the origin.
Where:
As = absorbance of sample
Ab = absorbance of blank
OIV-MA-AS323-03 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Boron
5. Calculations
The µg of boron contained in 5 mL of eluate, (corresponding to 0.5 mL of wine)
obtained from interpolating the net absorbance values of (As - Ab) on the
calibration graph is E. The content, B, in milligrams of boron per liter is given by:
B mg/L = E
0 .5
BIBLIOGRAPHY
WOLF B., Soil Science and Plant Analysis, 1971, 2(5), 363-374 et 1974, 5(1), 39-44.
CHARLOT C. and BRUN S., F.V., O.I.V., 1983, no771.
OIV-MA-AS323-03 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
Method OIV-MA-AS323-04A
Type II method
Sulfur dioxide
(Resolution Oeno 377/2009)
1. Definitions
Free sulfur dioxide is defined as the sulfur dioxide present in the must or wine in
the following forms: H2SO3, HSO3, whose equilibrium as a function of pH and
temperature is:
H+ + HSO3
H2SO3
H2SO3 represents molecular sulfur dioxide.
Total sulfur dioxide is defined as the total of all the various forms of sulfur dioxide
present in the wine, either in the free state or combined with their constituents.
2. Free and Total Sulfur Dioxide
2.1 Principle
Free sulfur dioxide is carried over by a stream of air or nitrogen and is fixed
and oxidized by bubbling through a dilute and neutral solution of hydrogen
peroxide. The sulfuric acid formed is determined by titration with a standard
solution of sodium hydroxide. Free sulfur dioxide is purged from the wine by
entrainment at low temperature (10 °C).
Total sulfur dioxide is purged from the wine by entrainment at high
temperature (approximately 100 °C).
2.2 Method
2.2.1 Apparatus
The apparatus to be used should conform to the diagram overleaf, especially
with regard to the condenser (see Fig 1).
The gas supply tube to the bubbler B ends in a small sphere of 1 cm diameter
with 20 holes 0.2 mm in diameter around its largest horizontal circumference.
Alternatively, this tube may end in a sintered glass plate that produces a large
number of very small bubbles and thus ensures good contact between the
liquid and gaseous phases.
The gas flow through the apparatus should be approximately 40 L/h. The bottle
situated on the right of the apparatus is intended to restrict the pressure
reduction produced by the water pump to 20 – 30 cm water. In order to
regulate the flow rate, a flow meter with a semi-capillary tube should be
installed between the bubbler and the bottle.
OIV-MA-AS323-04A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
Pump
FIGURE 1: The dimensions are given in millimeters. The internal diameters of the
4 concentric tubes making up the condenser are: 45, 34, 27 and 10 mm.
2.2.2 Reagents
2.2.2.1 Phosphoric acid: phosphoric acid 85% (20 =1.71 g/mL), diluted to 25%
2.2.2.2 Hydrogen peroxide solution, 9.1 g H2O2/L (3 volumes)
2.2.2.3 Indicator reagent:
Methyl Red ...................……………………….................…...….. 100 mg
Methylene Blue ...................………………………...............….. 50 mg
Ethanol 50% (v/v) ...................………………………...........….. 100 mL
2.2.2.4 0.01 M Sodium hydroxide solution
OIV-MA-AS323-04A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
2.2.3 Determination of free sulfur dioxide content.
The wine must be maintained at 20°C in a full and stoppered flask for 2 days
before determination.
2.2.3.1 Procedure
- Place 50 mL of the sample and 15 mL of phosphoric acid into the 250 mL
flask (A) of the entrainment apparatus. Connect the flask into the apparatus.
- In the bubbler (B), place 2 or 3 mL of hydrogen peroxide solution, two drops
of the indicator reagent and neutralize the hydrogen peroxide solution with
the 0.01 M sodium hydroxide solution. Connect the bubbler to the
apparatus.
Bubble air (or nitrogen) through the apparatus for 15 minutes. Free sulfur
dioxide carried over is oxidized to sulfuric acid. Remove the bubbler from the
apparatus and titrate the acid which has formed with the 0.01 M sodium
hydroxide solution.
Let n mL be the volume used.
2.2.3.2 Expression of results
The liberated sulfur dioxide is expressed in mg/L to the nearest whole number.
2.2.3.2.1 Calculation
If n is the number of mL of 0.01 M sodium hydroxide solution, used, the
amount of free sulfur dioxide in milligrams per liter is given by: 6.4 n
2.2.4 Determination of total sulfur dioxide content.
2.2.4.1 Procedure
 Samples having a SO2 content 50 mg/L of total SO2:
Place 50 mL of the sample and 15 mL of phosphoric acid (2.2.2.1) into the 250
mL round-bottom vacuum flask (A). Connect the flask to the apparatus.
Remark: In the case of must, proceed with the method of operation described
in the 1978 edition of the Compendium (see page 367).
 Samples with a content  50 mg/L of total SO2:
Place 20 mL of the sample and 5 mL phosphoric 85% acid into the 250 mL
round-bottom vacuum flask A. Connect the flask to the apparatus.
Place in the bubbler B, 2 or 3 mL of the hydrogen peroxide solution,
neutralized as before, and bring the wine in the flask A to a boil using a small
flame of 4 or 5 cm height which should directly touch the bottom of the flask.
Do not place the flask on a metal cloth, but on a mantle with a hole 30 mm in
OIV-MA-AS323-04A : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
diameter in it. This is to avoid overheating substances extracted from the wine
that are deposited on the walls of the flask.
Maintain boiling while passing a current of air (or nitrogen). Within 15
minutes the total sulfur dioxide is carried over and oxidized. Determine the
sulfuric acid formed by titration with 0.01 M sodium hydroxide solution.
Let n be the volume used.
2.2.4.2 Expression of results.
2.2.4.2.1 Calculation
Total sulfur dioxide in milligrams per liter:
- Samples low in sulfur dioxide (50 mL test sample): 6.4 · n
- Other samples (20 mL test sample): 16 · n
2.2.4.3 Repeatability (r):
(< 50 mg/L) 50 mL test sample,
(> 50 mg/L) 20 mL test sample,
2.2.4.4 Reproducibility (R):
(< 50 mg/L) 50 mL test sample,
(> 50 mg/L) 20 mL test sample,
r = 1 mg/L
r = 6 mg/L
R = 9 mg/L
R = 15 mg/L
BIBLIOGRAPHY
Reference method
PAUL F., Mitt. Klosterneuburg, Rebe u. Wein, 1958, ser. A, 821.
OIV-MA-AS323-04A : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
Method OIV-MA-AS323-04B
Type IV method
Sulfur dioxide
(Resolution Oeno 377/2009)
1. Definitions
Free sulfur dioxide is defined as the sulfur dioxide present in the must or wine in
the following forms: H2SO3, HSO3, whose equilibrium as a function of pH and
temperature is:
H+ + HSO3
H2SO3
H2SO3 represents molecular sulfur dioxide.
Total sulfur dioxide is defined as the total of all the various forms of sulfur dioxide
present in the wine, either in the free state or combined with their constituents.
2. Free and Total Sulfur Dioxide
2.1 Principle
Free sulfur dioxide is determined by direct titration with iodine. The combined
sulfur dioxide is subsequently determined by iodometric titration after alkaline
hydrolysis. When added to the free sulfur dioxide, it gives the total sulfur
dioxide.
2.2 Rapid Method
2.2.1 Reagents
2.2.1.1 EDTA: ethylenediaminetetraacetic acid, di-sodium salt
2.2.1.2 4 M Sodium hydroxide solution (160 g/L).
2.2.1.3 Dilute sulfuric acid: 10% sulfuric acid (20 = 1.84 g/mL) diluted 10%
(v/v).
2.2.1.4 Starch solution, 5 g/L.
Mix 5 g starch with approx. 500 mL water. Bring to a boil stirring
continuously and keep boiling for 10 minutes. Add 200 g of sodium chloride.
Cool and make to 1 liter.
2.2.1.5 0.025 M Iodine solution
2.2.2 Free sulfur dioxide
Place in a 500 mL conical flask place:
- 50 mL of wine
- 5 mL starch solution
- 30 mg EDTA
- 3 mL H2SO4
OIV-MA-AS323-04B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
Immediately titrate with 0.025 M iodine, until the blue color persists clearly
for 10 to 15 seconds. Let n mL be the volume of iodine used.
2.2.3 Combined sulfur dioxide
Add 8 mL of 4 M sodium hydroxide solution, shake the mixture once and
allow to stand for 5 minutes. Add, with vigorous stirring and in one operation,
the contents of a small beaker in which 10 mL of sulfuric acid have been
placed. Titrate immediately with the 0.025 M iodine solution; let n' be the
volume used.
Add 20 mL of sodium hydroxide solution, shake once and allow to stand for
5 minutes. Dilute with 200 mL of ice-cold water.
Add, while stirring vigorously and in one operation, the contents of a test tube
in which 30 mL sulfuric acid has previously been placed. Titrate the free
sulfur dioxide immediately with the 0.025 M iodine, and let n" be the volume
of iodine used.
2.2.4 Expression of the results
2.2.4.1 Calculation
Free sulfur dioxide in milligrams per liter is given by:
32 · n
Total sulfur dioxide in milligrams per liter is given by:
32 (n + n' + n")
Remarks:
1. For red wines with low SO2 concentrations, the 0.025 M iodine may be
diluted (for example: 0.01 M). In this case, replace the coefficient 32 by
12.8 in the above formula.
2. For red wines, it is useful to illuminate the wine from below with a beam of
yellow light from an ordinary electric light bulb shining through a solution
of potassium chromate or from a sodium vapor lamp. The determination
should be carried out in a dark room and the transparency of the wine
observed: it becomes opaque when the starch endpoint is reached.
3. If the quantity of sulfur dioxide found is close to or exceeds the legal limit,
the total sulfur dioxide should be determined with the reference method.
4. If the determination of free sulfur dioxide is specifically required, carry out
a determination on a sample kept under anaerobic conditions for two days at
20 °C before analysis. Carry out the determination at 20 °C.
5. Because certain substances are oxidized by iodine in an acid medium, the
quantity of iodine used in this way must be assessed for more accurate
determinations. To achieve this, combine the free sulfur dioxide in an
OIV-MA-AS323-04B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
excess of ethanal or propanal before beginning the titration with iodine.
Place 50 mL of wine into a 300 mL conical flask, add 5 mL of 7 g/L ethanol
solution or 5 mL of a 10 g/L propanal solution.
Stopper the flask and allow to stand for at least 30 minutes. Add 3 mL of
sulfuric acid and sufficient iodine, 0.025 M, to cause the starch to change
color. Let n''' mL be the volume of iodine used. This must be subtracted
from n (free sulfur dioxide), and from n + n' + n'' (total sulfur dioxide).
n''' is generally small, from 0.2 to 0.3 mL of 0.025 M iodine. If ascorbic acid
has been added to the wine, n''' will be much higher and it is possible, at
least approximately, to measure the amount of this substance from the value
of n''' given that 1 mL of 0.025 M iodine will oxidize 4.4 mg ascorbic acid.
By determining n''', it is possible to detect quite easily the presence of
residual ascorbic acid in amounts greater than 20 mg/L, in wines to which it
has been added.
BIBLIOGRAPHY
Rapid method:
RIPPER M., J. Prakt. Chem., 1892, 46, 428.
JAULMES, P., DIEUZEIDE J.-C., Ann. Fals. Fraudes, 1954, 46, 9; Bull.
O.I.V., 1953, 26, n° 274, 52.
KIELHOFER E., AUMANN H., Mitt. Klosterneuburg, Rebe u. Wein, 1957, 7, 289.
JAULMES P., HAMELLE Mme G., Ann. Fals. Exp. Chim., 1961, 54, 338
OIV-MA-AS323-04B : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
Method OIV-MA-AS323-04C
Type IV method
Sulfur dioxide
(Resolution Oeno 377/2009)
1. Definitions
Free sulfur dioxide is defined as the sulfur dioxide present in the must or wine in
the following forms: H2SO3, HSO3, whose equilibrium as a function of pH and
temperature is:
H+ + HSO3
H2SO3
H2SO3 represents molecular sulfur dioxide.
Total sulfur dioxide is defined as the total of all the various forms of sulfur dioxide
present in the wine, either in the free state or combined with their constituents.
2 Molecular Sulfur Dioxide
2.1 Principle of the Method
The percentage of molecular sulfur dioxide, H2SO3, in free sulfur dioxide, is
calculated as a function of pH, alcoholic strength and temperature.
For a given temperature and the alcoholic strength:
H2SO3
H+ + HSO3
[H2SO3] =
L
10 (pH  pkM )  1
(1)
where
L = [H2SO3] + [HSO3]
pkM = pkT 
A I
I B I
I
= ionic strength
A & B = Coefficients which vary according to temperature and alcoholic strength.
kT
= Thermodynamic dissociation constant; the value of pkT is given in Table 1
for various alcoholic strengths and temperatures.
kM = Mixed dissociation constant
Taking a mean value 0.038 for the ionic strength I, Table 2 gives the values of
pkM for various temperatures and alcoholic strengths.
OIV-MA-AS323-04C : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
The molecular sulfur dioxide content calculated by the relationship given in (1)
is presented in Table 3 for various values of pH, temperature and alcoholic
strength.
2.2 Calculations
Knowing the pH of wine and its alcoholic strength, the percentage of
molecular sulfur dioxide is given in Table 3 for a temperature t °C. Let this be
X %.
The amount of molecular sulfur dioxide in mg/L is given by: X · C
C = the free sulfur dioxide in mg/L
Table I
Values of the thermodynamic constant pkT
Temperature oC
Alcohol
% by volume
20
25
30
35
40
0
5
10
15
20
1.798
1.897
1.997
2.099
2.203
2.000
2.098
2.198
2.301
2.406
2.219
2.299
2.394
2.503
2.628
2.334
2.397
2.488
2.607
2.754
2.493
2.527
2.606
2.728
2.895
Table II
Values of the Mixed Dissociation Constant pkM (I= 0.038)
Temperature °C
Alcohol
% by volume
20
25
30
35
40
0
5
10
15
20
1.723
1.819
1.916
2.014
2.114
1.925
2.020
2.116
2.216
2.317
2.143
2.220
2.311
2.417
2.538
2.257
2.317
2.405
2.520
2.663
2.416
2.446
2.522
2.640
2.803
OIV-MA-AS323-04C : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
Table III
Molecular Sulfur Dioxide as a Percentage of Free Sulfur Dioxide (I=0.038)
T = 20 oC
Alcohol % by volume
pH
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
7.73
6.24
5.02
4.03
3.22
2.58
2.06
1.64
1.31
1.04
0.83
0
10
15
20
9.46
7.66
6.18
4.98
3.99
3.20
2.56
2.04
1.63
1.30
1.03
11.55
9.40
7.61
6.14
4.94
3.98
3.18
2.54
2.03
1.62
1.29
14.07
11.51
9.36
7.58
6.12
4.92
3.95
3.16
2.53
2.02
1.61
17.09
14.07
11.51
9.36
7.58
6.12
4.92
3.95
3.16
2.53
2.02
17.15
14.12
11.55
9.40
7.61
6.14
4.94
3.97
3.18
2.54
2.03
20.67
17.15
14.12
11.55
9.40
7.61
6.14
4.94
3.97
3.18
2.54
24.75
22.71
17.18
14.15
11.58
9.42
7.63
6.16
4.55
3.98
3.18
24.49
20.48
16.98
13.98
11.44
9.30
7.53
6.08
4.89
3.92
3.14
29.28
24.75
20.71
17.18
14.15
11.58
9.42
7.63
6.16
4.95
3.98
35.36
30.29
25.66
21.52
17.88
14.75
12.08
9.84
7.98
6.44
5.19
T = 25 oC
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
11.47
9.58
7.76
6.27
5.04
4.05
3.24
2.60
2.07
1.65
1.32
14.23
11.65
9.48
7.68
6.20
4.99
4.00
3.20
2.56
2.05
1.63
T = 30 oC
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
18.05
14.89
12.20
9.94
8.06
6.51
5.24
4.21
3.37
2.69
2.16
20.83
17.28
14.23
11.65
9.48
7.68
6.20
4.99
4.00
3.21
2.56
OIV-MA-AS323-04C : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulfur dioxide
Table III (continued)
Molecular Sulfur Dioxide as a Percentage of Free Sulfur Dioxide (I=0.038)
T=35 oC
Alcohol % by volume
pH
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
0
5
10
15
20
22.27
18.53
15.31
12.55
10.24
8.31
6.71
5.44
4.34
3.48
2.78
24.75
20.71
17.18
14.15
11.58
9.42
7.63
6.16
4.95
3.98
3.18
28.71
24.24
20.26
16.79
13. 82
11.30
9.19
7.44
6.00
4.88
3.87
34.42
29.42
24.88
20.83
17.28
14.23
11.65
9.48
7.68
6.20
4.99
42.18
36.69
31.52
26.77
22.51
18.74
15.49
12.71
10.36
8.41
6.80
34.52
29.52
24.96
20.90
17.35
14.29
11.70
9.52
7.71
6.22
5.01
40.89
35.47
30.39
25.75
21.60
17.96
14.81
12.13
9.88
8.01
6.47
50.14
44.74
38.85
33.54
28.62
24.15
20.19
16.73
13.77
11.25
9.15
T = 40 oC
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
29.23
24.70
20.67
17.15
14.12
11.55
9.40
7.61
6.14
4.94
3.97
30.68
26.01
21.83
18.16
14.98
12.28
10.00
8.11
6.56
5.28
4.24
BIBLIOGRAPHY
Molecular sulfur dioxide:
BEECH F.W. & TOMAS Mme S., Bull. O.I.V., 1985, 58, 564-581.
USSEGLIO-TOMASSET L. & BOSIA P.D., F.V., O.I.V., 1984, n° 784.
OIV-MA-AS323-04C : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Sulphur dioxide
Method OIV-MA-AS323-05
Type IV method
Sulphur dioxide
Reference method: Procedure for grape juice
(Resolution Oeno 377/2009)
1. Apparatus:
See 2.2.1., from OIV-MA-AS323-04A
2. Reagents:
 Phosphoric acid (20 =1.71 g/ml) diluted at 25% (m/v).
 For other reagents, see 2.2.2., from OIV-MA-AS323-04A
3. Procedure:
Introduce 50 ml of grape juice and 5 ml of phosphoric acid diluted 25% (m/v)
in a 250 ml balloon A control trainer. Set up the balloon.
Continue as indicated as in 2.2.4.1., from the OIV-MA-AS323-04A form.
4. Calculation:
Given n as the number of milliliters of 0.01 M sodium hydroxide solution used,
the total sulphur dioxide content of grape juice in milliliters per liter:
6.4 x n
OIV-MA-AS323-05 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Mercury - Atomic Spectrofluorimetry
Method OIV-MA-AS323-06
Type IV method
Determination of mercury in wine by vapour
generation and atomic spectrofluorimeter
(Resolution Oeno 15/2002)
1. FIELD OF APPLICATION
This method applies to the analysis of mercury in wines with a concentration
range between 0 to 10 ug/l.
2. DESCRIPTION OF TECHNIQUE
2.1. Principle of the method
2.1.1 Mineralisation of wine takes place in an acid environment: heating
under reflux;
mineralisation is achieved with a potassium permanganate.
2.1.2. Reduction of non-consumed permanganate by hydroxylamine
hydrochlorate
2.1.3. Reduction in mercury II (metal mercury by stannous chloride (II).
2.1.4. Mercury pick up by an argon current at ambient temperature
2.1.5. Dosage of mercury in monoatomic vapour state by atomic
flourescence spectometre with wavelength of 254 nm. Mercury atoms
are excited by a mercury vapour lamp; the atoms thus excited emit a
radiation called flourescent which allows the quantification of
mercury present using a photonics detector to obtain good linearity
while eliminating memory effects.
2.2 Principle of analysis (figure 1)
The peristaltic pump absorbs the stannous chloride solution, the blank solution
(demineralised water containing 1% nitric acid) and the sample of mineralised
wine.
The mercury metal is taken up in a gas-liquid separator by a current of argon.
After going through a drying tube, the mercury is detected by florescence.
Then, the gaseous current goes through a permanganate potassium solution in
order to capture the mercury.
OIV-MA-AS323-06 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Mercury - Atomic Spectrofluorimetry
peristaltic
Peristaltic
pump
perist
Air
peristaltic
Stannous
chloride
c
pump
Blank
Reference or
Sample
Hygroscopic
membrane
Stannous
Valve
open
chloride
Upon sample
position
Argon
flowmeter
Towards
sink
Gas-liquid
separator
Florescence
detector
Analytic Chain for dosage of mercury
3. REAGENTS AND PREPARATION OF REACTIVE SOLUTIONS
3.1 Ultra-pure demineralised water
3.2 Ultra-pure 65% nitric acid
3.3 White: demineralised water (3.1) containing 1% of nitric acid (3.2)
3.4 Nitric acid solution 5.6 M (3.1):
Put 400 ml of nitric acid (3.2) into a 1000 ml flask; fill with
demineralised water (3.1).
3.5 Sulphuric acid (d= 1.84)
3.6 Sulphuric acid solution 9M:
Put 200 ml of demineralised water (3.1), 50 g of potassium
permanganate (3.7) into a 1000 ml flask; fill with demineralised
water (3.1).
3.7 Potassium permanganate KMnO4
3.8 5% Potassium permanganate solution:
Dissolve 50 g of potassium permanganate (3.7) with demineralised
water (3.1), in a 1000 ml flask. Fill with demineralised water (3.1).
3.9 Hydroxylamine hydrogen chloride NH2OH, HCl
3.10 Reducing solution:
OIV-MA-AS323-06 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Mercury - Atomic Spectrofluorimetry
Weigh 12g of hydroxylamine hydrogen chloride (3.9) and dissolve
in 100 ml of demineralised water (3.1).
3.11 Stannous chloride (SnCl2, 2 H2O)
3.12 Concentrated hydrochloric acid
3.13 Stannous chloride solution:
Weigh 40 g of stannous chloride (3.11) and dissolve in 50 ml of
hydrochloric acid (3.12). Fill with 200 ml of demineralised water
(3.1).
3.14 Mercury standard solution at 1g/l
prepared by dissolving 1708 g of Hg (NO3). H2O in an aqueous nitric
acid solution at 12% prepared from metal mercury.
3.15 Reference mercury solution at 10 mg/l :
Place 1 ml of mercury standard solution (3.14) in a 100 ml
volumetric flask, add 5 ml of nitric acid, fill will demineralised water
(3.1)
3.16 Mercury solution at 50 mg/l:
Place 1 ml of 10 mg/l (3.15) solution in a 200 ml flask. Add 2 ml of
nitric acid (3.2). Fill with demineralised water (3.1).
4. APPARATUS
4.1 Glass ware
4.1.1 Volumetric flasks 100, 200, and 1000 ml (class A)
4.1.2 Volumetric pipette 0.5,1.0, 2.0, 5, 10 and 20 ml (class A)
4.1.3 Precautionary action: Before using, the glass ware must be
washed with 10% nitric acid, leave in contact 24 hours, then
rinse with demineralised water.
4.2 Mineralisation apparatus (figure 2)
4.3 Temperature controlled heating mantle
4.4 Squeeze pump
4.5 Cold vapour generator
4.5.1 Liquid gas separator
4.6 Desiccant (Hygroscopic membrane) covered by an air current
(supplied from a compressor) and placed before the detector
4.7 Spectrofluorimeter
OIV-MA-AS323-06 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Mercury - Atomic Spectrofluorimetry
4.7.1 Mercury vapour lamp regulated to 254 nm wave length
4.7.2 Atomic fluorescence specific detector
4.8 Computer
4.8.1 Software which regulates the parameters of the vapour generator
and the atomic fluorescence detector and enables calibration and
usage of the results.
4.8.2 Printer which stores results
4.9 Neutral gas bottle (argon)
5. PREPARATION OF CALLIBRATION SOLUTIONS AND SAMPLES
5.1 SET OF CALLIBRATION SOLUTIONS: 0; 0.25; 0.5; and 1.0 ug/L
Introduce : 0; 0.5; and 1.0 and 2.0 ml of the mercury solution to 50 ug/l
(3.16.) in 4 100 ml flasks; add 1 % nitric acid (3.2.); fill with demineralised
water (3.1.).
5.2. Preparation of samples (figure 2)
Wine is mineralised in a glass pyrex apparatus made up of three parts
joined by spherical honing: a 250 ml balloon, a vapour recuperation
chamber, a refrigerant.
Using a pipette put 20 ml of wine in a 250 ml reaction flask; assemble the
mineralisation apparatus.
Add 5 ml of sulphuric acid (3.6.) and 10 ml of nitric acid (3.4.) slowly;
leave overnight.
Heat slowly under reflux until the nitrous vapours disappear ; leave to
cool. Recover the condensed vapours in the reaction flask. Rinse the
recipient with demineralised water. Pour the contents of the reaction flask
into a 100 ml volumetric flask. Add potassium permanganate solution
(3.8.) until the colour remains. Solubilize the precipitate (MnO2) with a
reducing solution (3.10.). Fill with demineralised water (3.1.).
Carry out a blank test on demineralised water.
OIV-MA-AS323-06 : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Mercury - Atomic Spectrofluorimetry
Refrigerant
with balls
Circulation
cold water
Vapour
Recuperation
chamber
Introduction
of wine
and reagents
Reaction Balloon
(wine + sulphuric acid
+ nitric acid)
Heating mantle
Mineralisation apparatus
6. OPERATING PROCEDURE
6.1 Analytical measurement
Turn on the fluorimeter; the apparatus is stable after 15 minutes. The
squeeze pump absorbs the white (3.3), the stannous lead II (3.13) and the
sample calibrations (5.1) or (5.2.) Verify that bubbling occurs in the liquid
gas separator. Present the calibration samples successively (5.1); set off
the vapour generator program. The computer software establishes a
calibration curve (percentage of fluorescence according to concentration
of mercury ug/l). Then present the samples (5.2).
6.2 Automatic checks
A blank analysis and a calibration are analysed every five tests to correct
any possible spectrofluorimeter derivitives.
OIV-MA-AS323-06 : R2009
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Mercury - Atomic Spectrofluorimetry
7. EXPRESSION OF RESULTS
Results are provided by the computer software and expressed in ug/l. Deduct
the mercury concentration in wine in ug/l keeping into account 1/5 dilution.
8. CHECKING RESULTS
Quality control is carried out by placing reference material in which the
mercury content is known, following the set of calibrations and every 5
samples. Following the analytical series, the reference material is red wine, dry
white wine or sweet white wine.
The check card is set for each reference material used. The check limits are set
at: +/- 2SR intra (2SR intra : reproducibility spread-type)
The uncertainly calculation, carried out on check cards, resulted in a red wine
reference of: 3.4 +/- 0.8 ug/l and for reference dry white wine : 2.8+/-0.9 ug/l.
9. BIBLIOGRAPHY
CAYROL M., BRUN S., 1975. Dosage du mercure dans les vins. Feuillet Vert de
l’O.I.V. n°371.
REVUELTA D., GOMEZ R., BARDON A., 1976. Dosage du mercure dans le vin
par la méthode des vapeurs froides et spectrométrie d’absorption atomique. Feuillet
Vert de l’O.I.V. n°494.
CACHO J., CASTELLS J.E., 1989. Determination of mercury in wine by flameless
atomic absorption spectrophotometry. Atomic Spectroscopy, vol. 10, n°3.
STOCKWELL P.B., CORNS W.T., 1993. The role of atomic fluorescence
spectrometry in the automatic environmental monitoring of trace element analysis.
Journal of Automatic Chemistry, vol. 15, n°3, p 79-84.
SANJUAN J., COSSA D., 1993. Dosage automatique du mercure total dans les
organismes marins par fluorescence atomique. IFREMER, Rapport d’activité.
AFNOR, 1997. Dosage du mercure total dans les eaux par spectrométrie de
fluorescence atomique. XPT 90-113-2.
GAYE J., MEDINA B., 1998. Dosage du mercure dans le vin par analyse en flux
continu et spectrofluorimétrie. Feuillet Vert de l’O.I.V. n°1070.
OIV-MA-AS323-06 : R2009
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
OIV-MA-AS323-07
Type of Method: criteria
CRITERIA FOR THE METHODS OF QUANTIFICATION
OF POTENTIALLY ALLERGENIC RESIDUES OF
FINING AGENT PROTEINS IN WINE
(OIV-Oeno 427-2010)
1
Method Criteria Definitions
Trueness
the closeness of agreement between the average value
obtained from a large series of test results and an
accepted reference value
r=
Repeatability limit, the value below which the absolute
difference between 2 single test results obtained under
repeatability conditions (i.e., same sample, same
operator, same apparatus, same laboratory, and short
interval of time) may be expected to lie within a
specific probability (typically 95%) and hence r = 2.8 x
sr .
Sr =
Standard deviation, calculated from results generated
under repeatability conditions.
RSDr=
Relative standard deviation, calculated from results
generated under repeatability conditions [(Sr/ x ) x 100],
where x is the average of results over all laboratories
and samples.
R=
Reproducibility limit, the value below which the
absolute difference between single test results obtained
under reproducibility conditions (i.e., on identical
OIV-MA-AS323-07: R2010 1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
material obtained by operators in different laboratories,
using the standardised test method), may be expected to
lie within a certain probability (typically 95%); R = 2.8
x sR .
SR =
Standard deviation, calculated from results under
reproducibility conditions.
RSDR =
Relative standard deviation calculated from results
generated under reproducibility conditions [(SR/ x x
100]
HoR =
HORRAT value: the observed RSDR value divided by
the RSDR value calculated from the Horwitz equation.
B0 =
Mean blank
LOD =
Limit of detection, calculated as LOD = B0 + 3*Sr(B0)
LOQ =
Limit of quantification, calculated as LOD = B0 +
10*Sr(B0)
2. General Aspects
Requirement
The method of analysis must be associated with specific oenological
practices
Additives or processing aids containing allergenic proteins
Each product must be characterized from the chemical point of view and
quality control is strictly necessary
Class of analytical methods
Generally speaking, immunoenzymatic approaches are considered the most
suitable and easy methods for routine control of allergens.
OIV-MA-AS323-07: R2010 2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
The determination of allergenic fining agent proteins residues in wines
could use Sandwich, Competitive, Direct or Indirect ELISA methods.
If no enzyme-labeled antibody is available a biotinylated antibody and
avidine-HRP conjugate can be used for detection.
Antibody
‐ Antibody characterization (evaluation of detection of allergens with
higher or lower affinity)
‐ High specificity for the commercial processing aids (characterized as
described above)
‐ Cross-reactivity characterization taking in account the proteins
usually included in enological practices
‐ Capability to detect allergen derivatives that could be formed by
enological treatments (proteolysis or modified molecules)
Method
‐ Antibody must have optimal binding properties in wine samples
‐ Methods must have optimal performances in wine samples having
different chemical characteristics (pH and dry extract, red and white
wine, etc..)
‐ Results in wines coming from different geographical area (even
when different enological practices are applied) must be comparable
‐ The binding properties of the antibodies must be optimal with
different condition of maturation of wine (time, temperatures, color
changes ...)
3. Type of methods
Specific methods for the determination of fining agent proteins in wine are
not prescribed yet. Several ELISA methods are already available and can be
applied.
Laboratories shall use a method validated to OIV requirements that fulfils
the performance criteria indicated in Table 1. Wherever possible, the
OIV-MA-AS323-07: R2010 3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
validation shall include a certified reference material in the collaborative
trial test materials. If not available, an alternative estimation of trueness
should be used.
The General Protocol for the Direct and Indirect ELISA Method
The direct, one-step method uses only one labeled antibody. This labeled
antibody is incubated with the antigen contained in the sample/standard and
bound to the well.
The indirect, two-step method uses a labeled secondary antibody for
detection. First, a primary antibody is incubated with the antigen contained
in the sample/standard and bound to the well. This is followed by incubation
with a labeled secondary antibody that recognizes the primary antibody.
Direct
1. Prepare a surface to which antigen in sample is bound.
2. Block any non-specific binding sites on the surface.
3. Apply enzyme-linked antibodies that bind specifically to the
antigen.
4. Wash the plate, so that the antibody-enzyme conjugates in excess
(unbound) are removed.
5. Apply a chemical which is converted by the enzyme into a color or
fluorescent or electrochemical signal.
6. Measure the absorbance or fluorescence or electrochemical signal
(e.g., current) of the plate wells to determine the presence and
quantity of antigen.
Before the assay, the antibody preparations must be purified and conjugated.
Indirect
1.
2.
3.
4.
Prepare a surface to which antigen in sample is bound.
Block any non-specific binding sites on the surface.
Apply primary antibodies that bind specifically to the antigen
Wash the plate, so that primary antibodies in excess (unbound)
are removed.
5. Apply enzyme-linked secondary antibodies which are specific to
the primary antibodies.
OIV-MA-AS323-07: R2010 4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
6. Wash the plate, so that the antibody-enzyme conjugates in excess
(unbound) are removed.
7. Apply a chemical which is converted by the enzyme into a color
or fluorescent or electrochemical signal.
8. Measure the absorbance or fluorescence or electrochemical
signal (e.g., current) of the plate wells to determine the presence
and quantity of antigen.
Before the assay, both antibody preparations must be purified and one must
be conjugated.
Figure 1: Direct and indirect ELISA
For most applications, a high-binding polystyrene microtiter plate is best;
however, consult manufacturer guidelines to determine the most appropriate
type of plate for binding the given antigen.
The major advantage of direct and indirect ELISA is the high sensitivity,
achieved via a comparably easy set-up with reduced chances of unspecific
binding. However, it is only applicable in samples containing low amounts
of non-antigen protein.
General Protocol for the competitive ELISA Method
OIV-MA-AS323-07: R2010 5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
The term "competitive" describes assays in which measurement involves the
quantification of a substance by its ability to interfere with an established
system. The detection can be done directly, one-step method, or indirectly,
two-step method.
Direct
1. Prepare a surface to which a known quantity of wanted antigen is
bound.
2. Block any non-specific binding sites on the surface.
3. Apply the sample or standard (antigen) and the enzyme-linked
antibodies that bind specifically to the antigen on the coated
microplate. The antigens immobilized on the surface and the
antigens in solution “compete” for the antibodies. Hence, the more
antigen in the sample, the less antibody will be bound to the
immobilized antigens.
4. Wash the plate so that the antibodies in excess (unbound) and
unbound antigen-antibody-complexes are removed.
5. Apply a chemical which is converted by the enzyme into a color or
fluorescent or electrochemical signal.
6. Measure the absorbance or fluorescence or electrochemical signal
(e.g., current) of the plate wells to determine the presence and
quantity of antigen.
Before the assay, the antibody preparations must be purified and must be
conjugated.
Indirect
1. Prepare a surface to which a known quantity of antigen is bound.
2. Block any non-specific binding sites on the surface.
3. Apply the sample or standard (antigen) and the specific primary
antibody to the coated microplate. The antigens immobilized on the
surface and the antigens in solution “compete” for the antibodies.
Hence, the more antigen in the sample, the less antibody will be
bound to the immobilized antigens.
OIV-MA-AS323-07: R2010 6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
4. Wash the plate so that the antibodies in excess (unbound) and
unbound antigen-antibody-complexes are removed.
5. Add a secondary antibody, specific to the primary antibody,
conjugated with an enzyme.
6. Wash the plate so that the conjugated antibodies in excess (unbound)
are removed
7. Apply a chemical which is converted by the enzyme into a color or
fluorescent or electrochemical signal.
8. Measure the absorbance or fluorescence or electrochemical signal
(e.g., current) of the plate wells to determine the presence and
quantity of antigen.
Before the assay, both antibody preparations must be purified and one must
be conjugated.
Figure 2: Direct and indirect competitive ELISA
For competitive ELISA, the higher the original antigen concentration, the
weaker is the signal.
For most applications, a high-binding polystyrene microtiter plate is best;
however, consult manufacturer guidelines to determine the most appropriate
type of plate for binding the given antigen.
General Protocol for the Sandwich ELISA Method
The Sandwich ELISA measures the amount of antigen between two layers
OIV-MA-AS323-07: R2010 7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
of antibodies (i.e. capture and detection antibody). The antigen to be
measured must contain at least two different antigenic sites (epitopes) for
binding two different antibodies. Either monoclonal or polyclonal
antibodies can be used.
Direct
1.
2.
3.
4.
5.
Prepare a surface to which capture antibody is bound.
Block any non-specific binding sites on the surface.
Apply the antigen-containing sample or standard to the plate.
Wash the plate, so that unbound antigen is removed.
Apply enzyme-linked antibodies (detection antibodies) that bind
specifically to the antigen.
6. Wash the plate, so that the enzyme-linked antibodies in excess
(unbound) are removed.
7. Apply a chemical which is converted by the enzyme into a color or
fluorescent or electrochemical signal.
8. Measure the absorbance or fluorescence or electrochemical signal
(e.g., current) of the plate wells to determine the presence and
quantity of antigen.
Before the assay, both antibody preparations must be purified and one must
be conjugated.
Indirect
1. Prepare a surface to which capture antibody is bound.
2. Block any non specific binding sites on the surface.
3. Apply the antigen-containing sample or standard to the plate.
4. Wash the plate, so that unbound antigen is removed.
5. Apply primary antibodies that bind specifically to the antigen.
6. Wash the plate, so that primary antibody in excess (unbound) is
removed.
7. Apply enzyme-linked antibodies (secondary antibodies) that bind
specifically to the primary antibody.
8. Wash the plate, so that the enzyme-linked antibodies in excess
(unbound) are removed.
OIV-MA-AS323-07: R2010 8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
9. Apply a chemical which is converted by the enzyme into a color or
fluorescent or electrochemical signal.
10. Measure the absorbance or fluorescence or electrochemical signal
(e.g., current) of the plate wells to determine the presence and
quantity of antigen.
Before the assay, all the antibody preparations must be purified and one of
them must be conjugated.
Figure 3: Direct and indirect Sandwich-ELISA
For indirect Sandwich-ELISA, it is necessary for the capture antibodies and
the detection antibodies to be raised in different species (e.g. mouse and
rabbit), so that the enzyme-linked secondary antibodies specific for the
detection antibodies do not bind to the capture antibodies, as well.
For most applications, a high-binding polystyrene microtiter plate is best;
however, consult manufacturer guidelines to determine the most appropriate
type of plate for binding the given antigen.
For sandwich ELISA, the measure is proportional to the amount of antigen
in samples.
The advantage of Sandwich ELISA is that even crude samples do not have
to be purified before analysis, and the assay can be very sensitive.
OIV-MA-AS323-07: R2010 9
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
Table 1: Performance criteria for methods of analyses for potentially
allergenic fining agent proteins in wine
Parameter
Value/Comment
Applicability
Suitable for determining fining agents in
wine for official purposes.
Detection limit
(expressed in mg/L)
Casein: at least 0.5
Ovalbumin: at least 0.5
Isinglass : at least 0.5
Lysozyme : at least 0.5
Limit of quantification
(expressed in mg/L)
Casein: at least 1 mg/L
Isinglass : at least 1 mg/L
Lysozyme : at least 1 mg/L
Ovalbumin : at least 1 mg/L
Precision
HORRAT values of less or equal to 2 in the
validation collaborative trial
Recovery
80% - 105% (as indicated in the
collaborative trial)
Specificity
Free from matrix interferences
Trueness
x  m < 1,96 *
sR ( lab )²  sr ( lab )² * 1  1 n 
where m is the certified value of the wine
reference material and x is the average of n
measurements of compound content in this
wine, within the same laboratory.
OIV-MA-AS323-07: R2010 10
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS –
OIV
Criteria for the quantification of potentially allergenic residues of
fining agent proteins in wine
Sr(lab) are standard deviations, calculated
from results within the same laboratory
under repeatability conditions.
SR(lab) are standard deviations, calculated
from results within different laboratories
under reproducibility conditions.
OIV-MA-AS323-07: R2010 11
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Method OIV-MA-AS4-01
Type IV Method
Microbiological Analysis of Wines and Musts
(Resolution oeno 8/95)
Detection, Differentiation and Counting of Micro-organisms
Objective:
Microbiological analysis is aimed at following alcoholic fermentation and/or
malolactic fermentation and detecting microbiological infections, and allowing the
detection of any abnormality, not only in the finished product but also during the
different phases of manufacture.
Comments:
All experiments must be carried out under normal microbiological aseptic
conditions, using sterilized material, close to a Bunsen burner flame or in a laminar
flow room and flaming the openings of pipettes, tubes, flasks, etc. Before carrying
out microbiological analysis, it is necessary to ensure that the samples to be
analyzed are taken correctly.
Field of application:
Microbiological analysis can be applied to wines, musts, mistelles and all similar
products even when they have been changed by bacterial activity.
Microbiological analysis techniques:
Microbiological analysis may be undertaken using the following techniques:
1. Quality tests
1.1. Air quality tests
1.2. Incubator quality tests
2. Detection, differentiation of micro-organisms and direct counting of yeasts
2.1. Microscopic examination of liquids or deposits
2.2. Staining with methylene blue
2.3. Gram staining
2.4. Catalase Test
2.5. Direct counting of yeasts by microscope
3. Counting of micro-organisms by culture
3.1. Culture in or on solid media
3.1.1. Plate culture
3.1.2. Membrane culture after filtration
OIV-MA-AS4-01 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
3.2. Culture in liquid environment - "Most Probable Number" (MPN).
Equipment:
Laminar flow room
Microscope
Incubator - for incubation between 20 and 35°C, with thermometer
Oven - for dry sterilization, thermostatically controlled at 180°C, with
thermometer
Incubator - in CO2 or in airtight jars with a CO2 generator, or other devices which
produce an atmosphere rich in CO2
pH-meter - calibrated in pH units with a precision of ± 0.1 unit
Autoclave - allowing steam sterilization of equipment and culture media
Centrifuge
Water bath
Gas burner
Balance - with a precision of 0.1 mg
Haemacytometer - Neubauer or equivalent
Sterilizing filtration devices - with sterilized filtering membranes of 13 mm or 25
mm in diameter, of 0.22 µm porosity, in cellulose ester or polyvinylidene fluoride
or equivalent, with hydrophobic edges
Bunsen burner
Volumetric flasks - 100 and 1000 mL
Conical (Erlenmeyer) flasks - 100, 150, 300, and 1000 mL stopped with cotton
wool, sterilized
Graduated cylinders - 50, 100, 300, 500 and 1000 mL
Pipettes - 1, 10, 15, 20 and 25 mL sterilized
Centrifuge tubes - sterilized
Autoclavable tubes - glass, rimless, dimensions 160 x 16 mm and 180 x 18 mm,
stopped with cotton wool, or equivalent, before sterilization
Slides
Cover glasses
Pasteur Pipettes - sterilized
Wire loop - sterilized
Filter paper
Immersion oil
Alcohol - 70 and 95% by volume
Hydrogen peroxide 3%
Comment: Other pieces of equipment serving the same functions can be used.
Techniques for the taking of samples:
It is necessary that samples be representative. Pipettes can be used for casks and
long glass tubes for tanks. The samples must be taken in a sterilized area to avoid
OIV-MA-AS4-01 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
external contamination that could give false positives. Thus, to take samples from
the "tasting taps" of casks or tanks, it is necessary first to allow several liters of
sample flow. Surfaces in contact must be disinfected, for example with 70%
alcohol or by flame.
Samples must be analyzed as quickly as possible after they have been taken. They
must be kept cool (4 to 5°C) during transportation and storage, if the microbial
flora is to be stabile.
1. Quality tests
Objective:
These tests are aimed at detecting the risk of microbial infection in advance.
Principle:
This technique is based on organoleptic changes (clouds, films, deposits, unusual
colors) shown by wine when subjected to certain aeration and temperature
conditions which can bring about bacterial activity. The nature of the changes
should be confirmed by microscopic examination.
Operating method:
1.1. Air quality tests
A 50 mL wine sample after filtration on coarse sterile filter paper is placed in
a 150 mL sterile conical flask stoppered with cotton and left at an ambient
temperature for at least 3 days. The clarity, color and possible presence of
clouds, deposits and films are examined over this time. A microscopic
examination is carried out in the case of cloud, deposit or film or a color
change.
1.2. Incubator quality tests
A 100 mL wine sample, after filtration on coarse sterilized filter paper, is
placed in 300 mL sterile conical flask stopped with cotton, put in an incubator
at 30°C and examined after at least 72 hours. Organoleptic changes can be
indicative of microbial development. A microscopic examination must
therefore be made.
2. Detection, differentiation of micro-organisms and direct counting of yeasts
2.1 Microscopic examination of liquids or deposits
Objective:
Microscopic examination under cool conditions is aimed at detecting and
differentiating the yeasts from the bacteria that might be present. However,
microscopic observation cannot distinguish between viable and non-viable
microorganisms.
Comment:
With staining, an estimation of the viable yeasts can be made.
OIV-MA-AS4-01 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Principle:
This technique is based on the magnification made by a microscope that
allows the observation of micro-organisms, whose size is on the order of a
micron.
Operation method:
Microscopic examination can be carried out directly on the liquid or on the
deposit.
Direct observation of the liquid will only be useful when the population is
sufficiently high (more than 5 x 105 cells/mL).
When wine shows a lower microorganism population, it is necessary to
concentrate the sample. Thus, about 10 mL of homogenized wine is
centrifuged at 3000 - 5000 rpm for 5 to 15 minutes. After decanting the
supernatant, the deposit is re-suspended in the liquid remaining at the bottom
of the centrifuge tube.
To carry out the microscopic observation, a drop of the liquid sample or the
homogenized deposit is placed on a clean glass slide with a Pasteur pipette or
a sterilized wire. It is covered with a cover glass and placed on a slide on the
stage of the microscope. Observation is made in a clear field, or preferably in
phase contrast, which allows a better observation of detail. A magnification
of 400 - 1000 is generally used.
2.2. Vital Staining with methylene blue
Objective:
Vital staining with methylene blue is used to differentiate between viable and
non-viable yeast cells.
Principle:
This coloration is based on the presence of viable cells which reduce the
color. For example, methylene blue is reduced to leuco-derivatives by viable
cells which remain colorless. Dead cells absorb the color, appearing blue.
Reagent:
Methylene blue solution of 0.1%
Preparation: Dissolve 0.1g of methylene blue with 100 mL of distilled water
in a conical flask. Filter through paper. The solution must be freshly
prepared.
Operating method:
Place a drop of sample and a drop of stain on a slide. Mix together with a
wire; after 5 minutes observe through a microscope in a clear field.
Results:
The viable cells remain colorless and the dead cells appear blue.
OIV-MA-AS4-01 : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Comments:
These staining methods are not completely reliable and are only used for
approximate enumeration. They are not recommended for bacteria.
To differentiate between viable and non-viable micro-organisms (both yeasts
and bacteria) more elaborate techniques can also be employed, such as those
using an epifluorescence microscope.
2.3. Gram staining
Objective:
Gram staining is used to differentiate between lactic bacteria (Gram positive)
and acetic bacteria (Grams negative) and also to observe their morphology.
Comments:
It must be remembered that Gram staining is not conclusive, as other bacteria
apart from lactic and acetic bacteria may be present.
Principle:
This color is based on the difference between Gram positive and Gram
negative bacteria due to the differences in structure and chemical composition
of their cell walls. In Gram negative bacteria, the cell walls that are rich in
lipids have a much reduced quantity of peptidoglycan which allows the
penetration of alcohol which dissolves the gentian-violet-iodine complex,
forming when the colorless cell is left, which will then be re-colored in red by
saffron. Conversely, the cell walls of Gram positive bacteria contain a large
quantity of peptidoglycan and a low concentration of lipids. Thus, the thick
peptidoglycan wall and the dehydration caused by the alcohol do not allow the
alcohol to penetrate the cell, which maintains the violet or dark blue coloring
of the gentian-violet-iodine complex.
There are several modifications to the Gram staining technique. It loses its
usefulness if it is performed on a culture that is too old. Thus, the strain must
be in an exponential growth phase within 24 to 48 hours.
Solutions:
The water used must be distilled.
1. Gentian violet solution
Preparation: Weigh 2g of gentian violet (or crystal violet), and put into a 100
mL conical flask and dissolve in 20 mL of 95% vol. alcohol. Dissolve 0.8g of
ammonium oxalate in 80 mL of distilled water. Mix the two solutions
together and only use after a period of 24 hours. Filter through paper at time
of use. Keep out of light in a dark flask.
2. Lugol solution
Preparation: Dissolve 2g of potassium iodide in a minimal quantity of water
(4 to 5 mL) and dissolve 1g of iodine in this saturated solution. Make the
volume up to 300 mL with distilled water. Keep out of light in a dark flask.
OIV-MA-AS4-01 : R2009
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
3. Saffranin solution:
Preparation: Weigh 0.5g of saffranin in a 100 mL conical flask, dissolve with
10 mL of 95% vol. alcohol and add 90 mL of water. Stir. Keep out of light in
a dark flask.
Operating method:
Smear preparation
Make a subculture of the bacteria in liquid or solid medium. Collect the
young culture bacteria from the deposit (after centrifugation of the liquid
culture) or directly from the solid medium with a loop or wire and mix in a
drop of sterilized water.
Make a smear on a slide, spreading a drop of the microbial suspension. Let
the smear dry, and then carry out fixation, rapidly passing the slide 3 times
through the flame of a Bunsen burner, or equivalent. After cooling, perform
staining.
Staining
Pour a few drops of gentian violet solution onto the fixed smear. Leave to
react for 2 minutes and wash off with water.
Pour in 1 to 2 drops of lugol solution. Leave to react for 30 seconds. Wash
with water and dry with filter paper.
Pour on 95% vol. alcohol, leave for 15 seconds. Rinse with water and dry
with filter paper.
Pour on a few drops of saffranin solution, leave to react for 10 seconds. Wash
with water and dry with filter paper.
Place a drop of immersion oil on the smear.
With the immersion objective, observe through a microscope in clear field.
Results:
Lactic bacteria remain violet or dark blue colored (Gram positive). Acetic
bacteria are red colored (Gram negative).
2.4. Catalase Test.
Objective:
This test is aimed at making a distinction between acetic and lactic bacteria.
The yeasts and acetic bacteria have a positive reaction. Lactic bacteria give a
negative catalase.
Comments:
It must be taken into account that the catalase test is not conclusive as other
bacteria apart from lactic and acetic bacteria may be present.
Principle:
The catalase test is based on the property that aerobic micro-organisms have
of decomposing hydrogen peroxide with release of oxygen:
catalase
2H2O2
2H2O + O2
OIV-MA-AS4-01 : R2009
6
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Reagent:
3 Volume hydrogen peroxide solution
Preparation: Measure 10 mL of 30% by volume hydrogen peroxide in a 100
mL calibrated flask and fill with freshly boiled distilled water. Stir and keep
in the refrigerator in a dark flask. The solution must be freshly prepared.
Operating method:
Place a drop of 3% by volume hydrogen peroxide on a slide and add a small
sample of young colony. If gas is released, it can be concluded that the
culture contains catalase. It is sometimes difficult to observe gas clearing
immediately, particularly with bacterial colonies. It is therefore advisable to
examine the culture through a microscope (objective x10).
2.5. Direct counting of the yeast cells with a microscope
Objective:
Direct counting with a microscope is aimed at evaluating the total yeast cell
population (viable and non-viable)
Comment:
This technique is only recommended for the counting of yeasts. Enumeration
of bacterial cells is both difficult and approximate due to their small size and,
more importantly, because of the presence of colloidal particles. This
technique can be used to estimate numbers of viable cells that are stained by
methylene blue.
Principle:
The technique is based on the counting of microorganisms in a sample of
known volume through a microscope. Special slides of the haemacytometer or
cell counting type are used to measure this volume.
Comment:
The slides consist of a restricted square surface, which, when covered by a
thick, optically flat cover glass, contain a sample of known volume. There are
various cell counters. For example, the Neubauer cell counter, considered
here, consists of a central cavity 0.1 mm deep and 1 mm2 in surface area, of
which the base is divided into 400 small squares. Thus, the volume of a
sample corresponding to a small square is 1/400 mm3.
Operating method:
Place a drop of the homogenized sample on the central plane of the chamber,
avoiding spillage. Cover with a thick cover glass, avoiding gas bubbles.
Leave to stabilize for a few minutes.
The concentration must be neither too concentrated nor too dilute. Where
very dense populations are concerned, the sample must be diluted.
OIV-MA-AS4-01 : R2009
7
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Conversely, where the number of cells to be counted is small, a concentration
must be made by centrifugation.
Count the micro-organisms through a microscope at random so that the total is
between 200 and 500. Avoid counting the same cell twice. To obtain a
representative figure, count the cells in different areas. Thus, for example,
five large squares can be counted (each consisting of 16 small squares), 1 in
the center and 4 on the edges of the square plane or 5 large squares on the
diagonal.
It is advisable to perform the counting twice. Find the mathematical average
of the two counts, record the results of the 400 small squares, and multiply by
104 to express the result in cells/mL. If the sample has been concentrated or
diluted, the concentration or dilution factor must be taken into account.
3. Counting of micro-organisms by culture
Objective:
The purpose of counting of microorganisms by culture is to evaluate the level
of contamination of the sample, that is to say, to estimate its microbial
stability since only viable microorganisms are capable of changing the
product. According to the culture media used and the culture conditions, three
types of microorganisms can be counted, namely, yeasts, lactic bacteria and
acetic bacteria.
3.1. Culture in solid medium
Principle:
This method is based on the pre-supposition that a viable micro-organism,
cultivated in or on specific solid nutritive media and in suitable conditions,
multiplies giving rise to a colony visible to the naked eye.
3.1.1. Petri dish culture
Objective:
This technique is aimed at counting microorganisms by dilution when the
microbial concentration is high.
Equipment:
Apart from the equipment already mentioned, the following is required.
Petri dishes (plates) - 57 cm2 (plastic) or 65 cm2 (borosilicate glass).
Diluents and solid culture media (see annexes 4 and 5)
Operating method:
Enumeration in Petri dishes can be performed by inoculation in or on the
surface of the appropriate medium after suitable incubation.
Preparation of dilutions:
Successive, decimal and increasing dilutions must be made in such a way that
the appropriate number is obtained (between 30 and 300 colonies) for the
OIV-MA-AS4-01 : R2009
8
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
counting and differentiation of colonies, as the microbial concentration is not
known in advance. An estimation of the dilutions can be made using a
microscope. Thus, starting with an homogenized sample, prepare a series of
decimal dilutions (1:10) in the diluent.
Take a 1 mL sample and add to 9 mL of diluent in the first tube. Mix
thoroughly. Take 1 mL of this dilution and add to 9 mL of diluent in the
second tube. Continue this dilution protocol until the last suitable dilution,
according to the presumed microbial population, using sterilized pipettes for
each dilution (see diagram 1 of Annex 1).
Preparation of inoculations by incorporation in agar media
Prepare Petri dishes in such a way that a dilution giving counts of between 30
and 300 colonies can be obtained.
Inoculate 1 mL of sample and 1 mL of each of the dilutions prepared and
chosen according to the presumed microbial population, mixed at the time, in
two separate Petri dishes using different sterilized pipettes or automatic
pipettes with autoclavable hydrophobic tips. Then pour 15 mL of appropriate
agar culture medium, at 42 to 45°C, liquefied beforehand in a boiling water
bath (or in microwave oven or on a heated magnetic gyratory shaker),
avoiding prolonged heating. Mix thoroughly immediately, gently making
circular movements in both directions, avoiding spillage and the formation of
air bubbles. Leave to cool on a level surface until solidified.
Preparation of inoculations on the surface of the agar medium
Place 0.1 to 0.2 mL of chosen sample dilutions, on the surface of the agar that
has no excessive liquid that would cause uneven spreading. Dilutions 10
times more concentrated than in the incorporation method must be used. A
homogeneous distribution is made on the surface of the medium, having been
placed and solidified in plates with a glass spreader sterilized with alcohol and
flamed. Incubate the inverted plates in incubator:
 yeasts - 20 to 25°C, for 3 to 10 days, under aerobic conditions.
(Comment: for the counting of yeasts, 3 days incubation is usually
sufficient.
If the presence of Dekkera Brettanomyces is suspected, the incubation must
be for 7 to 10 days)
 lactic bacteria - 25°C, 4 to 10 days, under anaerobic or microaerophilic
conditions.
 acetic bacteria - 25 to 30°C for 2 to 4 days, under aerobic conditions.
Count the colonies with the naked eye or with a magnifying glass or a
colony counter. If there is any doubt, confirm the identity of the colonies
(yeast or bacteria) with a microscope.
OIV-MA-AS4-01 : R2009
9
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Test the sterility of the medium, the diluent and the equipment by making a
blank test with a sample of sterilized water for each series of tests.
Results:
Express the results in Colony Forming Units/mL - CFU/mL (rather than
microorganisms/mL, since each colony can be the result of a microorganism
or a cluster), in scientific notation with a decimal (for example 1100 would be
reported as "1.1 x 103 CFU/mL").
When the number of colonies is calculated by estimation, the results are
expressed as ECFU/mL (E - estimated).
Count the plates containing between 30 and 300 colonies.
Calculate the CFU/mL by multiplying the mathematical average of the
colonies affected by the dilution factor.
Specify incubation time.
Rules for calculating the CFU by counting and by estimation
1. Examine the dish of each of the dilutions containing between 30 and 300
colonies and calculate the mathematical average.
2. If only one of the two prepared dishes contains between 30 and 300
colonies, calculate the mathematical average of the two dishes
corresponding to this dilution.
3. If two consecutive dilutions show dishes containing between 30 and 300
colonies, calculate the mathematical average, provided that the larger
number is not more than double the smaller one. When the larger number is
more than twice the smaller one, only use the smallest count (for example:
150 (dilution 10-1) and 350 (dilution 10-2); the ratio 350/150 is greater than
2. Thus, use the lower number, i.e. 150, and calculate the CFU/mL by
multiplying by the dilution factor).
4. If all the dishes have less than 30 colonies, consider only the dishes in
which the dilution is the lowest (least diluted). Calculate the mathematical
average and report the results as estimates.
5. If none of the dilution dishes contain colonies and there are no inhibiting
substances present, consider the counts as "< 1.0" times the lowest dilution
factor and report the results as estimates. For example, if there are no
colonies in the 10-2 dilution (lowest), the result will be reported as "< 1.0 x
102 ECFU/mL".
6. If all the dishes contain more than 300 colonies, consider the count closest
to 300 as an estimate. Thus, only consider dishes with the highest dilution,
i.e. the most diluted, and report the results as estimated.
7. If the number of colonies significantly exceeds 300, report the result as
"TNTC" (too numerous to count). If there are less than 10 colonies per cm2
on the surface of the media, count in 13 squares (when using a colony
OIV-MA-AS4-01 : R2009
10
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
counter) having a representative distribution of colonies. Determine the
mathematical average by square and multiply by the appropriate factor to
calculate the estimated colonies per dish. Similarly, when there are more
than 10 colonies/cm2, count 4 representative squares, calculate the
mathematical average per square and calculate as indicated previously.
The factor is 57 for dishes with a surface area of 57 cm2 and 65 for dishes
with a surface area of 65 cm2.
8. When there is overpopulation in the dishes (more than 100 colonies/cm2)
report the results as estimates. For the calculation, multiply 5700 or 6500
by the highest dilution factor. For example, for a dilution of 10 -3, the
ECFU will be reported as "> 5.7 x 106 ECFU/mL" or "> 6.5 x 106
ECFU/mL".
3.1.2. Membrane filtration method
Objective:
The purpose of this technique is to count the microorganisms when the
microbial population is low. A concentration of microorganisms remaining on
a membrane filter of suitable porosity is made after filtration of a sample of
given volume.
Equipment:
In addition to equipment previously referred to, the following are required:
Vacuum pump - or equivalent device;
Filtration unit (support-filter and funnel) - for membranes 47 mm diameter.
Can be made of stainless steel, borosilicate glass or autoclavable/sterile nonautoclavable plastic;
Filtration manifold or vacuum flask;
Trap - to be inserted between the flask and the vacuum pump;
Tongs - for fixation;
Tongs with rounded sides - for membranes;
Pincers - to take hold of the filtration unit during sterilization (by flaming or
by immersion in boiling water);
Stainless steel pot - for sterilization of the filtration unit in boiling water.
Filtering membranes - cellulose esters of 47 mm diameter with hydrophobic
sides, preferably white and square with a porosity of 0.2 µm, 0.45 µm or 0.8
µm. Preferably, use membranes in proprietary individual sterile packages.
Otherwise, sterilize them by autoclave for 10 min. at 121°C.
Petri dishes - made of borosilicate glass or 60 mm sterilized plastic;
Sterilized absorbent stoppers.
Culture medium (see annex 5)
Operating method:
1. Preparation of plates
OIV-MA-AS4-01 : R2009
11
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Prepare the necessary number of Petri dishes with the appropriate culture
medium. The membrane filter can be placed on a suitable gel medium or on a
solid stand - absorbent pad soaked with suitable liquid in a nutritive medium.
a) Petri dishes with suitable gel media (see annex 5)
Deposit into 60 mL Petri dishes about 6 mL of solid medium, pre-liquefied
in a water bath or similar, and allowed to cool to 45°C. Leave to solidify on
a level surface. Invert the dishes and leave for at least 2 to 3 hours before
use.
The prepared dishes can be used when they are free of contamination but
not completely dry.
b) Petri dishes with suitable liquid medium (see annex 5)
If a liquid nutritive medium is preferred, deposit the membrane on an
absorbent pad soaked with the liquid medium in Petri dishes.
For the preparation of dishes with liquid medium, open the Petri dishes and
place sterilized absorbent pads inside (there are sterilized dishes with
sterilized absorbent pads commercially available).
Generally, vials of 2 mL of sterilized liquid nutritive media or equivalent
are used. These are also available commercially.
Open a 2 mL vial of appropriate liquid medium and pour about 1.8 mL onto
an absorbent pad, spreading over the whole surface.
2. Filtration of the sample
Use autoclavable filtration equipment at the start of a series of filtrations.
Between filtrations, for steel or borosilicate glass units, sterilize with alcohol
and flame in a Bunsen burner flame or immerse in boiling water bath for 5
min. For plastic equipment, alternative sterilization methods can be used,
such as UV radiation with a two-minute exposure or other appropriate
chemical or physical agents.
Insert a trap between the vacuum flask and the vacuum pump. Attach the
filtration unit onto the vacuum flask. Fix the funnel on the filter stand with
appropriate tongs. Introduce a few milliliters of sterile water to ensure a
sound contact between the membrane and the filter support, and leave to drain
completely.
Remove the membrane filter of suitable porosity from its sterile individual
pack with flamed tongs which have been first allowed to cool. On the base of
the filter support, place the membrane filter, which must be centered with the
square facing upwards. Introduce an appropriate volume of homogenized
sample, according to the presumed population. The figure considered for the
colony count is between 20 and 200. However, for a good differentiation, the
number of colonies should not exceed 60 - 80.
Turn on the vacuum. After filtration, break the vacuum carefully to avoid a
backward surge. (The filtration volume varies, generally, from 25 to 500 mL.
OIV-MA-AS4-01 : R2009
12
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
A sample of known volume must always be poured. If the volume of the
sample is less than 20 mL, pour enough sterilized water into the funnel to
cover the membrane and then, with a sterilized pipette, introduce the sample
to be filtered.)
Remove the funnel and take hold of the membrane with the cooled flamed
tongs and place on the culture medium in the Petri dish with the square facing
upwards. Avoid the formation of air bubbles between the membrane and the
medium, since they would prevent homogeneous contact and, consequently, a
correct microbial growth. Invert the dish and incubate in an incubator for a
suitable time and in suitable conditions according to the type of
microorganisms to be detected.
For yeasts:
Use membrane filters of 0.45 µm or 0.8 µm porosity, YEPD gel medium (see
annex 5) and incubate under aerobic conditions, between 20 and 25°C, for 3 to
10 days. (Comment: for the counting of yeasts, incubation for 3 days is
usually sufficient. If the presence of Dekkera/Brettanomyces is suspected,
incubation must be for 7 to 10 days)
For lactic bacteria:
Use membrane filters of 0.2 µm or 0.45 µm porosity, suitable gel culture
medium (see annex 5) and incubate under anaerobic or microaerophilic
conditions, at 25°C. The length of incubation can be 10 days.
For acetic bacteria:
Use membrane filters of 0.2 µm or 0.45 µm porosity, suitable gel culture
medium (see annex 5) and incubate under aerobic conditions between 25 and
30°C for 2 to 4 days. Count the developed colonies with the naked eye, with a
magnifying glass or in a colony counter. Confirm the identity of the colonies
(yeasts or bacteria) with a microscope if there are any doubts.
Test the sterility of the media, the membrane filters and the equipment by
performing a blank with a sample of sterilized water for each series of tests.
Results:
Express the results in Colony Forming Units/mL - CFU/mL (instead of
microorganisms/mL, since each colony can be the result of a microorganism
or a cluster).
When a large volume of wine is filtered, even if there are insufficient numbers
of developed colonies, the results can be presented relative to the volume
used.
If there is no colony development and there are no inhibiting substances
present, report the results as "< 1.0 CFU" by the highest volume of wine
filtered.
Count all the colonies on the membrane when there are 1 to 2 colonies per
OIV-MA-AS4-01 : R2009
13
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
square. If too many colonies per square have been obtained, the analysis of
the sample should be repeated if possible with a smaller quantity of sample.
If the number of colonies is too high (above 200), estimates can be made when
the colonies are not agglomerated and the distribution is representative.
If this is not possible, report the results as "TNTC" (too numerous to count).
Refer to the porosity of the membrane used and the incubation period.
3.2. Culture in liquid medium- "Most Probable Number" (MPN)
Objective:
The purpose of this technique is to evaluate the number of viable
microorganisms in wines having high contents of solid particles in suspension
and/or high incidence of plugging.
Principle
This technique is based on the estimation of the number of viable
microorganisms in liquid medium, starting from the principle of its normal
distribution in the sample.
Diluents and liquid culture media (see annexes 4 and 5)
Operating method:
Several quantitative and successive solutions are prepared and following this,
after incubation, a certain proportion of tests will not lead to any growth
(negative tests), while others will begin to grow (positive tests). If the sample
and the dilutions are homogeneous, and if the number of dilutions is
sufficiently high, it is possible to treat the results statistically, using suitable
tables (tables based on McCrady's probability calculations), and to extrapolate
this result to the initial sample.
Preparation of dilutions:
Starting from a sample of homogenized wine, prepare a series of decimal
1
dilutions ( /10) in the diluent.
Take 1 mL of wine and add to 9 mL of diluent in the first tube. Homogenize.
Take 1 mL of this dilution to add to 9 mL of diluent in the second tube.
Continue this dilution protocol until the last suitable dilution, according to the
presumed microbial population, using sterilized pipettes for each dilution.
The dilutions must be made until extinction, i.e. the absence of development
in the lowest dilutions (see the diagram to annex 2).
Preparation of inoculations:
Inoculate 1 mL of wine and 1 mL of each of the prepared dilutions, mixed at
the time, in, respectively, 3 tubes with the appropriate culture medium (see
annex 5). Mix thoroughly.
Incubate the inoculated tubes in the incubator at 25°C for yeasts (3 days, up to
OIV-MA-AS4-01 : R2009
14
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
10 days), under aerobic conditions, and for lactic bacteria, under anaerobic or
microaerophilic conditions (8 days, up to 10 days), making periodic
observations up to the last day of incubation.
Results:
All those tubes that show a microbial development leading to the formation of
a whitish deposit, more or less evident and/or with a more or less marked
disturbance are considered as positive. The results must be confirmed by
observation through a microscope. Specify the incubation period.
The reading of the tubes is made by noting the number of positive or negative
tubes in each combination of three tubes (in each dilution). For example, "31-0" signifies: 3 positive tubes in the 100 dilution (wine), 1 in the 10-1 dilution
and zero in the 10-2 dilution.
For a number of dilutions higher than 3, only 3 of these results are significant.
To select the results allowing for the determination of the "MPN", it is
necessary to determine the "typical number" according to the examples in the
following table:
Calculation of the Most Probable Number (MPN)
Taking account of the typical number obtained, the MPN is determined
through Table A (Annex 3) based on McCrady's probability calculations,
considering the dilution made. If the dilution series is 10 0 ; 10-1 ; 10-2 the
reading is direct. If the dilution series is 101; 100; 10-1 the reading is 0.1 times
this value. If the dilution series is 10-1; 10-2; 10-3; the reading is 10 times this
value.
Comment:
If there is a need to increase the sensitivity, a concentration 10 1 of wine can be
used. To obtain this concentration of microorganisms in 1 mL, centrifuge 10
mL of wine and take 1 mL of deposit (after having taken 9 mL of excess
liquid) and inoculate according to the previously described method.
Expression of Results:
The microorganism content of wine must be expressed in cells per mL, in
scientific notation to one decimal place. If the content is lower than 1.0 cells
per mL, the result must be presented as "<1.0 cells per/mL".
(See annexes on following pages)
BIBLIOGRAPHY
- ANDREWS, W. et MESSER, J. (1990). Microbiological Methods. in : AOAC
Official Methods of Analysis, 15th edition, 1, 425-497, Association of
Analytical Chemist, Washington.
OIV-MA-AS4-01 : R2009
15
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
- BIDAN, P. (1992). Analyses Microbiologiques du Vin. F.V. O.I.V. nº 910,
Paris.
- BOURGEOIS, C.M. et LEVEAU, J.Y. (1991). Techniques d'analyse et de
contrôle dans les industries agro alimentaires, 2ème édition, 3. Le Contrôle
Microbiologique Lavoisier, Tec. & Doc., APRIA Ed. Paris.
- CARR, J. G. (1959). Acetic acid bacteria in ciders. Ann. Rep. Long Ashton Res.
Sta., 160.
- DE MAN, J. C. (1975). The probability of most probable number. European
Journal of Applied Microbiology, 1, 67-78.
- LAFON-LAFOURCADE, S. et al. (1980). Quelques observations sur la
formation d'acide acétique par les bactéries lactiques. Conn. Vigne Vin, 14, 3,
183-194.
- MAUGENET, J. (1962). Les Acétobacter du cidre. Identification de quelques
souches. An. Technol. Agric., 11, 1, 45-53.
- PLARIDIS et LAFON-LAFOURCADE, S. (1983). Contrôle microbiologique
des vins. Bull. O.I.V., 618, 433-437, Paris.
- RIBÉREAU-GAYON, J. et PEYNAUD, E. (1961). Traité d'Oenologie, Tome 2,
Librairie Polytechnique CH. Béranger, Paris et Liège.
- Standard Methods for the Examination of Water and Waste Water (1976). 14 th
edition, American Public Health Association, Incorporated, New York.
- Standard Methods for the Examination of Water and Waste Water (1985). 16th
edition, American Public Health Association, DC 20005, Washington.
- VAZ OLIVEIRA, M., BARROS, P. et LOUREIRO, V. (1995). Analyse
microbiologique du vin. Technique des tubes multiples pour l'énumération de
micro-organismes dans les vins - "Nombre le plus probable" (NPP), F.V. O.I.V.
nº 987, Paris.
- VAZ OLIVEIRA, M. et LOUREIRO, V. (1993). L'énumération de microorganismes dans les vins ayant un indice de colmatage élevé, Compte rendu des
travaux du groupe d'experts "Microbiologie du Vin" de l'O.I.V., 12ème session,
annexe 2, Paris.
- VAZ OLIVEIRA, M. et LOUREIRO, V. (1993). L'énumération de microorganismes dans les vins ayant un indice de colmatage élevé, 2ème partie, Doc.
Travail du groupe d'experts "Microbiologie du Vin" de l'O.I.V., 13ème session,
Paris.
OIV-MA-AS4-01 : R2009
16
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Annex 2
Preparation of dilutions and inoculations
OIV-MA-AS4-01 : R2009
17
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Annex 3 - (OENO 8/95)
TABLE A
« Most Probable Number» (MPN) for 1 mL sample utilizing 3 tubes
with 1 mL, 0.1 mL et 0.01 mL
1
mL
0
0
0
0
0
1
1
1
1
1
1
1
1
2
2
Positive tubes
0,1
0,01
mL
mL
0
0
0
1
1
0
1
1
2
0
0
0
0
1
0
2
1
0
1
1
2
0
2
1
3
0
0
0
0
1
MPN
1 mL
0,0
0,3
0,3
0,6
0,6
0,4
0,7
1,1
0,7
1,1
1,1
1,5
1,6
0,9
1,4
1
mL
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
Positive tubes
0,1
0,01
mL
mL
0
2
1
0
1
1
1
2
2
0
2
1
2
2
2
3
3
0
3
1
3
2
0
0
0
1
0
2
1
0
MPN
1 mL
2,0
1,5
2,0
3,0
2,0
3,0
3,5
4,0
3,0
3,5
4,0
2,5
4,0
6,5
4,5
1
mL
1
3
3
3
3
3
3
3
3
3
3
Positive tubes
0.1
0,01
mL
mL
1
1
1
2
1
3
2
0
2
1
2
2
2
3
3
0
3
1
3
2
3
3
MPN
1 mL
7,5
11,5
16,0
9,5
15,0
20,0
30,0
25,0
45,0
110,0
>140,0
Adapted from the “ Standard Methods for the Examination of Water and Waste Water ” (1976)
Annex 4
Diluents:
Diluents are indicated by way of example. The water to be used must be distilled,
double distilled or deionized, with no traces of metals, inhibitors or other antimicrobial substances.
1. Physiological water
Preparation: Weigh 8.5g of sodium chloride in a 1000 mL calibrated flask. After
it has dissolved in the water, adjust the reference volume. Mix thoroughly. Filter.
Distribute 9 mL in the test tubes. Stop with cotton wool and autoclave for 20 min
at 121°C.
Ringer's solution 1/4
Preparation: Weigh 2.250g of sodium chloride, 0.105g of potassium chloride,
0.120g of calcium chloride (CaCl2.6H2O) and 0,050g of sodium hydrogen
carbonate in a 1000 mL calibrated flask. After it has dissolved in water, make up
to the mark. Mix thoroughly. Distribute 9 mL in the test tubes. Stop with cotton
wool and autoclave for 15 min at 121°C. (This solution is available commercially)
OIV-MA-AS4-01 : R2009
18
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Annex 5
Culture media
Culture media and antimicrobials are indicated by way of example.
The water to be used must be distilled, double distilled or deionized with no traces
of metals, inhibitors or other antimicrobial substances.
1. Solid culture media
1.1. For yeasts
YEPD medium (Yeast Extract, Peptone, Dextrose), agar medium +
chloramphenicol
Preparation: Weigh 10g of yeast extract (Difco or equivalent), 20g of
peptone, 20g of glucose and 100 mg of chloramphenicol1) in a 1000 mL
Erlenmeyer flask. Dissolve in 450 mL of water.
Then separately dissolve 20g of agar medium with 500 mL of water, in a 1000
mL Erlenmeyer, in boiling water bath, stirring frequently, avoiding prolonged
heating. After it has completely dissolved, add to the other solution and
complete the 1000 mL volume with water and mix together. Distribute
portions of 15 mL (for enumeration in plates) and 6 mL (for enumeration by
membrane filtration) in the test tubes. Stop with cotton wool and autoclave
for 15 min at 121°C.
Instead of chloramphenicol, penicillin, 20U/mL, and streptomycin, 40U/mL,
can be added on the plate when the portions of media are poured onto the
plates.
1.2. For lactic bacteria
Medium of Lafon-Lafourcade et al, agar medium + actidione
Preparation: Weigh 5g of yeast extract, 10g of meat extract, 15g of trypsic
peptone, 5g of sodium acetate, 2g of ammonium citrate, 0.05g of manganese
sulfate, 0.2g of magnesium sulfate, 20g of glucose and 50 mg of actidione2) in
a 1000 mL Erlenmeyer. Dissolve in 400 mL of water and correct the pH to
5.4 with 1N sodium hydroxide or 1N hydrochloric acid. Add 1 mL of Tween
80.
Then separately dissolve 20g of agar medium with 500 mL of water, in a 1000
mL Erlenmeyer, in a boiling water bath, stirring frequently, avoiding
prolonged heating. After it has completely dissolved, add the other solution
and complete the 1000 mL volume with water. Mix completely and distribute
15 mL (for enumeration in plates) or 6 mL (for enumeration by membrane
filtration) in the test tubes. Stop with cotton wool and autoclave for 20 min at
121°C.
1)
2)
To inhibit the growth of most of the bacteria
To inhibit the growth of most of the yeasts
OIV-MA-AS4-01 : R2009
19
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Dubois Medium (Medium 104), agar medium + actidione
Preparation: Weigh 5g of yeast extract (Difco), 5g of peptone, 3g of L-malic
acid, 0.05g of magnesium sulfate, 0.05g of manganese sulfate and 50 mg of
actidione in a 1000 mL Erlenmeyer. Dissolve in 200 mL of water. Add 250
mL of tomato juice and correct the pH to 4.8 with 1N sodium hydroxide or 1N
hydrochloric acid solution. Add a drop of Tween 80.
Then separately dissolve 20g of agar medium with 500 mL of water, in a 1000
mL Erlenmeyer, in boiling water bath, stirring frequently, avoiding prolonged
heating. After it has completely dissolved, add to the other solution and make
up the 1000 mL volume with water. Homogenize and distribute 15 mL (for
enumeration in plates) or 6 mL (for enumeration by membrane filtration) in
the test tubes. Stop with cotton wool and autoclave for 20 minutes at 121°C.
TJB medium (Tomato Juice Broth), agar medium + actidione
Preparation: Weigh 5g of glucose, 2g of tryptone (Difco), 5g of peptone
(Difco), 5g of yeast extract (Difco), 1g of liver extract, and 50 mg of actidione
in a 1000 mL Erlenmeyer. Dissolve in 400 mL of tomato juice. Correct the
pH to 5.5 with 1N sodium hydroxide or a 1N hydrochloric acid. Add a drop of
Tween 80.
Then separately dissolve 20g of agar medium with 500 mL of tomato juice, in
a 1000 mL Erlenmeyer, in a boiling water bath, stirring frequently, avoiding
prolonged heating. After it has completely dissolved, add to other solution
and make up to 1000 mL volume with tomato juice. Homogenize and
distribute 15 mL (for enumeration in plates) or 6 mL (for enumeration by
membrane filtration) in the test tubes. Stop with cotton wool and autoclave
for 20 minutes at 121°C.
Comment: The tomato juice used is diluted 4.2 times and filtered on Whatman
No.1 (1000 mL).
1.3 For acetic bacteria
G2 medium, agar medium + actidione
Preparation: Weigh 1.2g of yeast extract, 2g of ammonium phosphate and 50
mg of actidione in a 1000 mL Erlenmeyer flask. Add 500 mL of cider and
correct the pH to 5 with 1N sodium hydroxide or 1N hydrochloric acid.
Then separately dissolve 20g of agar medium in 450 mL of water in a 1000
mL Erlenmeyer, in boiling water bath, stirring frequently, avoiding prolonged
heating. After it has completely dissolved, add to the other solution and
complete the 1000 mL volume with water. Homogenize and distribute 15 mL
(for enumeration in plates) or 6 mL (for enumeration by membrane filtration)
in the test tubes. Stop with cotton wool and autoclave for 20 minutes at
121°C.
OIV-MA-AS4-01 : R2009
20
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Microbiological analysis of wines and musts
Carr medium, agar medium + actidione
Preparation: Weigh 30g of yeast extract and 50 mg of actidione in a 1000 mL
Erlenmeyer. Dissolve in 500 mL of water. Add 1 mL of 2.2% bromocresol
green.
Then separately dissolve 20g of agar medium in 450 mL of water, in a 1000
mL Erlenmeyer, in boiling water bath, stirring frequently, avoiding prolonged
heating. After it has completely dissolved, add to the other solution and make
up the 1000 mL volume with water. Mix completely and distribute 15 mL (for
enumeration in plates) or 6 mL (for enumeration by membrane filtration) in
the test tubes. Stop with cotton wool and autoclave for 15 min at 121°C. At
the time of liquefaction and cooling to 45°C, add to the agar medium 20 mL
per liter of alcohol sterilized by filtration (polyvinylidene fluoride membrane)
and mix.
2. Liquid culture media
2.1. For yeasts
YEPD medium (Yeast Extract, Peptone, Dextrose) + chloramphenicol
Preparation: Weigh 10.0g of yeast extract (Difco or equivalent), 20g of
peptone, 20g of glucose and 100 mg of chloramphenicol. Dissolve, make up
to 1000 mL volume with water and mix.
Distribute 5 mL portions of this medium in the test tubes and autoclave for 15
minutes at 121°C.
2.2. For lactic bacteria
MTJ medium (50% MRS medium "Lactobacilli Man Rogosa and Sharpe
Broth" + 50% TJB medium "Tomato Juice Broth") + actidione
Preparation: Weigh 27.5g of MRS "Lactobacilli Man Rogosa and Sharpe
Broth" (Difco or equivalent). Add 500 mL of water, heat to boiling to permit
complete dissolution and add 20.5g of TJB "Tomato Juice Broth" (Difco or
equivalent). Add 50g of actidione. Dissolve with water in order to obtain
1000 mL of solution having first corrected the pH to 5 with 1N hydrochloric
acid and mix.
Distribute 10 mL portions of this medium3) in the tubes and autoclave for 15
minutes at 121°C.
3)
The 10 mL volume is used instead of the 5 mL volume as with yeasts, due to the greater
sensitivity of lactic bacteria to oxygen.
OIV-MA-AS4-01 : R2009
21
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
Method OIV-MA-AS4-02A
Type IV method
Detection of preservatives and fermentation inhibitors
Method A 35 modified by resolution Oeno 6/2006
1. Fermentability Test
1.1 Objective
To show without specifying their nature, the possible presence of one or
several substances which act as fermentation inhibitors in wine.
1.2 Principle
The wine, whose free sulfur dioxide has been bound by addition of an aqueous
solution of acetaldehyde, is brought to 10% (v/v) alcohol. Glucose is added in
order for the sugar concentration to be between 20 and 50 g/L in the nutrient
solutions.
After inoculation with a yeast strain resistant to alcohol, the fermentation is
followed by weighing the quantity of carbon dioxide released.
The fermentation rate is compared to that of an authentic natural wine similar
in make up to the wine analyzed, and also to that of the test wine whose pH has
been adjusted to 6 (the majority of the mineral and organic acids are not active
in fermentation at this pH). These two reference wines are inoculated in the
same manner as the test wine.
1.3 Apparatus
90 mL flask sealed with a rubber stopper with a hole into which is placed a
narrow tube tapered at the uppermost portion.
1.4 Reagents and media
1.4.1 Aqueous acetaldehyde solution:
Solution prepared from acetaldehyde obtained by distillation of metaldehyde
or paraldehyde, in the presence of sulfuric acid, and standardized by the
method using sodium sulfite. Adjust the concentration of the solution to 6.9 g/L.
1 mL of this solution fixes 10 mg of sulfur dioxide.
1.4.2 Nutrient Solutions:
 Ammonium Sulfate, (NH4)2 SO4 ....................…..
25 g/L
 Asparagine ..............................................………….... 20 g/L
These solutions must be stored in the refrigerator.
OIV-MA-AS4-02A : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
1.4.3 Culture Medium:
 Solid medium: malt agar.
Powdered malt ......... ...............................…..…… .3 g
Glucose .....................................................………… 10 g
Pancreatic peptone .........................................….
5g
Powdered yeast extract ....................................…. 3 g
Agar ..........................................................…………20 g
Water ……….................................................……… 1 L
pH ............................................................……………6
Sterilize for 20 min. at 118 °C.
This mixture exists in a commercial prepared form.
 Liquid medium (an option):
 Divide the grape juice containing 170 to 200 g/L of sugar, in tubes
stoppered with cotton, at a rate of 10 mL per tube; sterilize in a water bath
at 100 °C for 15 min.
 Liquid malt: same medium as the solid medium, but without agar.
1.4.4 Culture and maintenance of the Saccharomyces bayanus strain and
preparation of the yeast.
a) Culture and maintenance of the strain on solid medium: From a collection
strain, inoculate in lines (streak) onto tubes of solid medium. These tubes
are put in an incubator at 25°C until the culture is very visible (about 3
days); the tubes can be stored in the refrigerator. This is sufficient for 6
months.
b) Preparation of the yeast:
One of the tubes of the liquid medium is inoculated in accordance with
proper microbiological techniques from the strain cultivated on solid
medium; after growth (24 to 48 h), repeat 2 times successively into the same
medium enriched with 10% alcohol (v/v), to acclimate the strain.
The second culture when actively fermenting will contain about 50 million
yeast per milliliter. This culture will serve to inoculate the wine to be
studied. Perform a count and inoculate at a rate of 105 yeast/mL.
1.5 Procedure
- Preparation of the wine:
100 mL of wine is treated with the necessary quantity of acetaldehyde
calculated in accordance with the amount of free sulfur dioxide (44 mg of
aldehyde binds 64 mg of sulfur dioxide). Wait 24 hours and check that the
wine contains less 20 mg free sulfur dioxide per liter.
If the alcoholic strength is greater than 10% (v/v), the wine should be diluted
with one of the solutions of glucose and water in amounts calculated to result
OIV-MA-AS4-02A : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
in a sugar concentration between 20 and 50 g/L, and to reduce the strength to
about 10% (v/v). For wines containing less than 10% vol., add solid glucose to
bring without dilution the amount of sugar between these values, so the
fermentation rate is not altered by the amount of sugar.
- Fermentability test:
In a 90 mL flask, place 60 mL of wine prepared as above, 2.4 mL of
ammonium sulfate solution and 2.4 mL of asparagine solution. Inoculate with
3 drops of a 3 day old culture of Saccharomyces bayanus, to obtain an initial
population close to 105 yeast/mL. Install the stopper with the pointed tube,
weigh the assembly to the nearest 10 mg and place in an oven at 25°C.
Weigh daily for at least 8 days.
Run each time concurrently, a wine of comparable make up and origin which
does not contain any preservative along with the test wine which has been
adjusted to pH 6.
A flask of non-inoculated wine indicates loss by evaporation.
1.6 Interpretation
In most cases, the fermentation begins within 48 hours and the daily liberation
of gas is greatest between the 3rd and the 5th day.
One can confirm the presence of a fermentation inhibitor only in the following
conditions:
a) If the fermentation does not begin or is delayed at least 2 days compared to
one of the 2 controls. When the delay is brief, it is difficult to ascertain the
presence because there may be "false positive" results, since certain natural
sweet wines sometimes behave as if they contained traces of inhibitors (in
particular sweet wines made from grapes having noble rot).
b) If the maximum daily release has not taken place between the 3rd and 5th
day, but after the 7th day, this release must be greater than or equal to 50 mg
for 60 mL of wine.
c) Plotting the fermentation curve and the curve of daily release of CO2 as a
function of time can facilitate the interpretation in a difficult case.
OIV-MA-AS4-02A : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
Method OIV-MA-AS4-02B
Type IV method
Detection of preservatives and fermentation inhibitors
Method A 35 modified by resolution Oeno 6/2006
1. Detection of the following acids: sorbic, benzoic, p-chlorobenzoic, salicylic,
p-hydroxybenzoic and its esters
1.1 Thin layer chromatography
1.1.1 Principle
The preservatives are extracted with ether from the previously acidified wine.
After separation by thin layer chromatography with polyamide powder, they
are located and characterized by examining the chromatogram under ultraviolet
light.
1.1.2 Apparatus
Chromatography bath.
 20 x 20 cm glass plates.
Preparation of the plates - Mix thoroughly 12 g of dry polyamide powder with
0.3 g fluorescent indicator; add, while stirring, 60 mL of methanol; spread on
plates to a thickness of 0.3 mm. Dry at normal temperature.
Note: Commercially prepared plates can be used.

1.1.3 Reagents
Diethyl ether
 Methanol
 Ethanol, 96% (v/v).
 Sulfuric acid diluted to 20% (v/v)
 Anhydrous sodium sulfate
 Polyamide powder for chromatography (e.g., Macherey-Nagel or Merck).
 Fluorescent indicator (F254 Merck or equivalent).
 Solvent:
n-Pentane ..................................................………………………10 vol.
n-Hexane ..................................................……… ……………..10 vol.
Glacial acetic acid .......................................…………………….. 3 vol.
 Standard solutions:
 Prepare standard solutions containing 0.1 g/100 mL of 96% ethanol (v/v)
of the following acids: sorbic, p-chlorobenzoic, salicylic, phydroxybenzoic and its esters.

OIV-MA-AS4-02B : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
 Prepare a solution of 0.2 g benzoic acid per 100 mL of 96% ethanol (v/v).
1.1.4 Procedure
Place 50 mL of wine in a separatory funnel; acidify with dilute 20% sulfuric
acid (1.1.3.4), and extract 3 times using 20 mL diethyl ether (1.1.3.1) per
extraction. Combine the washed solutions in a separatory funnel and wash
with a few milliliters of distilled water. Dry the ether with the anhydrous
sodium sulfate (1.1.3.2). Evaporate the ether dry using a 100°C water bath, or a
rotary evaporator. If the evaporation is accomplished on a water bath, it is
advisable to hasten the evaporation using a mild current of air until 2 or 3
milliliters remain, then finish the evaporation cold.
Dissolve the residue in 1 mL ethanol, deposit 3 to 5 µL of this solution on the
polyamide plate, as well as 3 to 5 µL of the various preservative standard
alcoholic solutions (1.1.3.9). Place the plate in a chromatography tank, and
saturate with solvent vapors. Let the solvent migrate to a height of about 15
cm, which takes from 1.5 to 2.5 hours.
Remove the plate from the tank and allow to dry at normal temperature.
Examine in ultraviolet light, at a wavelength of 254 nm. The preservatives
appear from the bottom of the plate upward in the following order: phydroxybenzoic acid, esters of p-hydroxybenzoic, salicylic acid, pchlorobenzoic acid, benzoic acid, sorbic acid.
With the exception of salicylic acid, which has a light blue fluorescence, other
preservatives give dark spots on a fluorescent yellow-green background.
Sensitivity - This technique allows determination of the following minimum
quantities of the miscellaneous preservatives expressed in milligrams per liter:
Salicylic acid ...........................................…………………... 3
Sorbic acid ...............................................…………………… 5
Esters of p-hydroxybenzoic acid .....................………….
5
p-Hydroxybenzoic acid ................................……………… 5-10
p-Chlorobenzoic acid ...................................………………. 5-10
Benzoic acid .............................................…………………… 20
1.2 High performance liquid chromatography
1.2.1 Procedure
The method is performed directly on the wine, without sample preparation. It
is necessary to dilute red wines before injecting them in order to preserve the
column.
OIV-MA-AS4-02B : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
Using this method, the detection threshold of preservatives in the solution
analyzed is about 1 mg/L.
1.2.2 Operating conditions
Conditions which are appropriate are the following:
A. For the determination of sorbic and benzoic acid
Proceed according to the sorbic, benzoic, salicylic acid assay method in
wines by high performance liquid chromatography (AS313-20-SOBESA)
provided in the Compendium
B. For the determination of p-chlorobenzoic acid, p-hydroxybenzoic acid and
its esters
Column: see A
Mobile phase:
Solution of ammonium acetate, 0.01 M + methanol (60 : 40)
pH: 4.5 - 4.6
Flow rate: see A
Injected volume: see A
Detector: UV, 254 nm
Temperature: see A
BIBLIOGRAPHY
JUNGE Ch., Zeits. Unters. Lebensmit., 1967, 133, 319
OIV-MA-AS4-02B : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
Method OIV-MA-AS4-02C
Type IV method
Detection of preservatives and fermentation inhibitors
Method A 35 modified by resolution Oeno 6/2006
1 Detection of the monohalogen derivatives of acetic acid
1.1 Principle
The monohalogen derivatives of acetic acid are extracted with ether from
acidified wine. The ether is then extracted using a 0.5 M sodium hydroxide
solution. The extraction solution must have the alkalinity maintained between
0.4 and 0.6 M. After the addition of thiosalicylic acid, the synthesis of the
thioindigo is implemented by the following steps:
a) Condensation of the monohalogen derivative with thiosalicylic acid and
formation of ortho-carboxylic phenylthioglycolic acid;
b) Cyclization of the acid formed in a heated alkaline medium, with the
formation of thioindoxyl;
c) Oxidation of the thioindoxyl with potassium ferricyanide in an alkaline
medium with formation of thioindigo, soluble in chloroform, in which it
gives a red color.
1.2 Apparatus
 Water bath at 100°C.
 Mechanical stirrer.
 Oven with a temperature of 200 ± 2°C.
1.3 Reagents
 Diethyl ether.
 Hydrochloric acid solution diluted to 1/3 (v/v). Mix one part pure
hydrochloric acid, 20 = 1.19 g/mL, with 2 parts of distilled water.
 Anhydrous sodium sulfate.
 Thiosalicylic acid solution: thiosalicylic acid 3 g in 100 mL sodium
hydroxide solution, 1.5 M.
 Sodium hydroxide solution, 0.5 M
 Potassium ferricyanide solution containing 2 g of K 3Fe(CN)6 per 100 mL of
water.
 Chloroform.
OIV-MA-AS4-02C : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
1.4 Procedure
Place 100 mL of test wine in an extraction flask with a ground glass stopper;
add 2 mL hydrochloric acid (1.3.2) and 100 mL diethyl ether (1.3.1). Shake
the contents vigorously for a few seconds by hand, then for 1 h with a
mechanical stirrer (1.2.2). Transfer to a separating funnel, allow to separate
and recover the ether layer.
Shake the ether extract with 8 to 10 g of anhydrous sodium sulfate (1.3.3) for a
few seconds.
Transfer the extract to the separating funnel, add 10 mL sodium hydroxide
solution, 0.5 M (1.3.5); shake for 1 min. Allow to settle.
Remove 0.5 mL of the alkaline extract and check, by titration with sulfuric
acid, 0.05 M, so that the strength falls between 0.4 and 0.6 M. Transfer the
alkaline extract contained in the separating funnel into a test tube containing 1
mL of thiosalicylic acid solution. Adjust, if necessary the strength of the
alkaline extract in order to bring it to the limits indicated, using a stronger
sodium hydroxide solution of known strength. Shake the contents of the test
tube for 30 seconds and transfer to an evaporating dish.
Place the dish on a water bath at 100°C blowing its surface with a current of
cold air. Maintain the dish on the water bath at 100°C for exactly 1 hour; the
residue may become practically dry in a shorter amount of time. If a crust
forms on the surface of the residue during the evaporation, it is advisable to
break or grind it up with a thin glass rod to facilitate the evaporation.
Place the dish in an oven maintained at 200 ± 2°C for exactly 30 minutes.
After cooling, recover the contents of the dish with 4 mL of water; transfer into
a separation funnel, add to the dish 3 mL of potassium ferricyanide solution to
fully dissolve any remaining residue and add to the separating funnel. Shake
for 30 seconds to facilitate oxidation. Add 5 mL chloroform, mix using 3 to 4
inversions. Allow to separate.
A pink or red color (according to the quantity of thioindigo formed) indicates
the presence of monohalogen derivatives of acetic acid.
Sensitivity - The method allows detection of 1.5 to 2 mg monochloroacetic acid
per liter of wine and corresponding quantities of the other derived
monohalogens. Since the yield of miscellaneous extractions is not quantitative,
this method cannot be used for determining the amount of these monohalogen
derivatives in the wines.
OIV-MA-AS4-02C : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
BIBLIOGRAPHY
FRIEDLANDER, Ber. Deutsch. Chem. Gesell., 1906, 39, 1062.
RAMSEY L.L., PATTERSON W.I., J. Ass. Off. Agr. Chem., 1951, 34, 827.
PERONNET M., ROCQUES S., Ann. Fals. Fraudes, 1953, 21-23.
Traité de chimie organique, edited by V. Grignard, 1942, 19, 565-566.
Official Methods of Analysis of the Association of Official Analytical Chemists,
11th édition, publiée par l'Association of Official Analytical Chemists,
Washington, 1970, 340-341.
TERCERO C., F.V., O.I.V., 1967, n° 224.
OIV-MA-AS4-02C : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
Method OIV-MA-AS4-02D
Type IV method
Detection of preservatives and fermentation inhibitors
Method A 35 modified by resolution Oeno 6/2006
1. Examination
dicarbonate)
and
determination
of
ethyl
pyrocarbonate
(diethyl
1.1 Principle
The diethyl carbonate formed by degradation of ethyl pyrocarbonate (diethyl
ester of pyrocarbonic acid) in the presence of ethanol is extracted from wine
using carbon disulfide and the quantity determined by gas chromatography.
Either of the procedures described below may be used.
1.2 Apparatus
1.2.1 Gas chromatography with flame ionization detector.
1.2.2 Columns:
- Capillary column coated with Carbowax 1540
Column length: 15.24 m
Inside diameter: 0.51 mm
- Polypropyleneglycol on Celite 545 (15:100), 60-100 mesh
Column length: 2 m
Interior diameter: 3 mm
1.3 Reagents
1.3.1 Anhydrous sodium sulfate
1.3.2 Carbon disulfide
The carbon disulfide must contain no impurities in the critical retention zone
(5 to 7 min.) for maximum sensitivity in accordance with the conditions of gas
chromatography as indicated in paragraph 1.4.2.
1.4 Procedure
1.4.1 Use of the capillary column.
Place 100 mL wine in a 250 mL separating funnel with 1 mL of carbon
disulfide (1.3.2). Mix vigorously for 1 min. The carbon disulfide phase
separated is rapidly centrifuged, then dried with anhydrous sodium sulfate
(1.3.1).
Inject 10 µl of the clear liquid supernatant into the chromatograph.
Chromatography conditions:
 Detector gases:
OIV-MA-AS4-02D : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
hydrogen: 37 mL/min.
air: 250 mL/min.
 Gas flow:
nitrogen: 40 mL/min.
A 1/10 splitter sends to the detector the gas mixture with a flow rate
of 3 to 5 mL/min.
 Temperature:
injector: 150 °C; oven: 80 °C; detector: 150 °C
 Detection limits:
0.05 mg/L of wine
1.4.2 Use of the column for polypropyleneglycol.
Add 20 mL of wine and 1 mL of carbon disulfide (1.3.2) into a conical
centrifuge tube with a stopper. Agitate vigorously for 5 minutes, then
centrifuge for 5 minutes applying a centrifugal force of 1000 to 1200 g. The
liquid supernatant produced is aspirated by a thin-tipped pipette; the carbon
disulfide phase is dried with a small quantity of anhydrous sodium sulfate,
added while stirring with a glass rod. Inject 1 µL of the clear liquid into the
gas chromatograph.
Chromatography conditions.
 Detector gas:
hydrogen: 35 mL/min.
air: 275 mL/min.
 Carrier gas flow:
nitrogen: 25 mL/min.
 Temperature:
injector: 240 °C
oven: 100 °C
detector: 240 °C
 Sensitivity range:
12 x 10-11 A to 3 x 10-11 A
 Chart speed:
1 cm/min.
 Detection limit:
0.10 - 0.05 mg/L of wine
Under these exact conditions, diethyl carbonate displays a retention time of
about 6 min.
The calibration of the apparatus is carried out using solutions of 0.01 and
0.05% (m/v) diethyl carbonate in carbon disulfide (1.3.2).
OIV-MA-AS4-02D : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
1.5 Calculation
Quantitative determination of diethyl carbonate is carried out preferably using the
internal standard method, referring to the peaks of the iso-butyl alcohol or iso-amyl
alcohol which are close to that of diethyl carbonate.
Prepare two samples of test wine: one of wine with 10 mL 10% ethanol (v/v)
added, the other the same wine to which has been added 1 mg diethyl carbonate
per liter using 10 mL of a 100 mg/L solution of diethyl carbonate in 10% ethanol
(v/v).
Treat these two samples according to one or the other of the techniques above
according to the column used.
Let:
S
Sx
i
I
= the peak area of the diethyl carbonate in the spiked wine
= the peak area of the diethyl carbonate in the wine,
= the peak area of internal standard in the wine,
= the peak area of internal standard in the spiked wine .
The concentration of diethyl carbonate in mg/L of wine is:
S
S x i  S
I
In the case where standardization is carried out using a pure standard solution of
diethyl carbonate, it is necessary to predetermine the yield of the extraction with
carbon disulfide in accordance with the procedure utilized. This yield is expressed
by the extraction factor F, with a decimal number less than or equal to 1 (yield
100%).
Let:
Sx = the peak area of diethyl carbonate given by the wine,
Se = the peak area given by the injection of the same volume of a standard
solution of diethyl carbonate of concentration C in mg/L,
Vx = the volume of wine used in the extraction with carbon disulfide,
Vs = the volume of carbon disulfide used for the extraction,
Ee = the sensitivity for the recording of Sx,
The concentration of diethyl carbonate in mg/L of wine is:
C x S x E x Vs
Se x Ee x F x V
OIV-MA-AS4-02D : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
If the concentration of the two solutions injected in the chromatograph is similar,
the response is the same for the recording of Sx and of Se; the formula is
simplified and becomes:
C x S x Vs
Se x F x V
BIBLIOGRAPHY
KIELHOFER E., WURDIG G., Dtsch. Lebensmit. Rdsch., 1963, 59, 197-200 &
224-228.
PRILLINGER F., Weinberg u. Keller, 1967, 14, 5-15.
REINHARD C., Dtsch. Lebensmit. Rdsch., 1967, 5, 151-153.
BANDION F., Mitt. Klosterneuburg, Rebe u. Wein, 1969, 19, 37-39.
OIV-MA-AS4-02D : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
Method OIV-MA-AS4-02E
Type IV method
Detection of preservatives and fermentation inhibitors
Method A 35 modified by resolution Oeno 6/2006
1. Examination of dehydroacetic acid
1.1 Principle
Wine acidified with sulfuric acid is extracted with a mixture of equal parts of
diethyl ether and petroleum ether. After evaporation of the solvent, the extract,
recovered with a small quantity of 96% ethanol (v/v) is deposited on a thin
layer of polyamide and silica gel with fluorescent indicator and subjected to
the action of the mobile solvent (benzene-acetone-acetic acid).
The
dehydroacetic acid is identified and characterized by ultraviolet examination of
the chromatogram.
1.2Apparatus
1.2.1 Equipment for thin layer chromatography
1.2.2 Oven
1.2.3 Rotary evaporator
1.2.4 UV lamp 254 nm.
1.3 Reagents
1.3.1 Diethyl ether
1.3.2 Petroleum ether (boiling point  40 °C)
1.3.3 Methanol
1.3.4 Sulfuric acid, 20% (v/v)
1.3.5 Anhydrous sodium sulfate.
1.3.6 Ethanol, 96% (v/v) .
1.3.7 Chromatographic separation layer: 10 g polyamide powder with
fluorescent indicator(e.g. polyamide DC II UV254 from Macherey-Nagel)
mixed vigorously with 60 mL methanol. Add while stirring, 10 ml of water
and 10ml of silica gel (with fluorescent indicator), e.g. Kiesselgel GF254
Merck. Spread this mixture on 5 plates (200 x 200 mm) to a thickness of 0.25
mm. Dry the plates at room temperature for 30 minutes, then place in a 70°C
oven for 10 min.
1.3.8 Migration solvent:
Crystallizable benzene ...........................……………….….. 60 vol.
Acetone ...........................................…………………… 3 vol.
Crystallizable acetic acid .......................………………..…. 1 vol.
OIV-MA-AS4-02E : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
1.3.9 Reference solutions:
Dehydroacetic acid and benzoic acid, 0.2%, in alcoholic solution.
Sorbic acid, p-chlorobenzoic acid, salicylic acid, p-hydroxybenzoic acid and its
propyl, methyl and ethyl esters, 0.1 % (m/v), in alcoholic solution.
1.4Procedure
Acidify 100 mL of wine using 10 mL of 20% sulfuric acid (1.3.3), then proceed to
extract 3 times using 50 mL of a (50:50) diethyl ether-petroleum ether mixture for
each extraction. Remove the clear aqueous phase leaving an aqueous emulsion and
the ether phase. Mix again the remaining liquid in the separation flask composed
of an emulsion and the ether phase. The remaining aqueous phase usually
separates clearly from the ether phase. If there is any residual emulsion, it should
be eliminated by the addition of a few drops of ethanol.
The diethyl ether-petroleum ether phases recovered are washed with 50 mL water,
dried using sodium sulfate, then evaporated by rotary evaporator, at 30 - 35 °C.
The residue is recovered with 1 mL of ethanol.
Deposit 20 µL of this solution on the starting line in a 2 cm wide band, or 10 µL in
a circular spot. For a comparison standard, deposit 5 µL of each of the reference
solutions described above. After the chromatography (ascending height of
migration 15 cm, duration 1 hour 15 min. to 1 hour 45 min., at normal saturation of
the chamber), the plate is dried at room temperature. Any dehydroacetic acid and
other preservatives present show up under a UV lamp at 254 nm.
When the examination of the chromatogram has revealed the presence of parachlorobenzoic acid, the propyl or methyl esters of para-hydroxybenzoic acid
which are only partly separated by this method may be identified consequently on
the extract above, following the method described in Examination of Sorbic,
Benzoic, Parachlorobenzoic Acids, 2.1. Thin layer chromatography.
BIBLIOGRAPHY
HALLER H.E., JUNGE Ch., F.V., O.I.V., 1972, n° 397, Mitt. Bl. der Gd CH,
Fachgruppe, Lebensmitt. u. gerichtl. Chem., 1971, 25, n° 5, 164-166.
OIV-MA-AS4-02E : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
Method OIV-MA-AS4-02F
Type IV method
Detection of preservatives and fermentation inhibitors
Method A 35 modified by resolution Oeno 6/2006
6. Sodium Azide
1.1 Method by high performance liquid chromatography
1.1.1 Principle
Hydrazoic acid isolated in wine using double distillation is identified after
derivatization with 3,5-dinitrobenzoyl chloride, by high performance liquid
chromatography.
Detection is carried out by ultraviolet absorption
spectrophotometry at 240 nm.
1.1.2 Apparatus
1.1.2.1 Distillation apparatus (distillation apparatus for determination of
alcoholic strength); the end of the condenser terminating in a tampered tube
1.1.2.2 500 mL spherical flasks with ground glass necks
1.1.2.3 10 mL flask with a ground glass stopper
1.1.2.4 Apparatus for HPLC
 Operating conditions:
Column: C18, 25 cm long.
Mobile Phase: acetonitrile-water (50:50)
Flow rate: 1 mL/min.
Volume injected: 20 µL
Detector: ultraviolet absorption spectrophotometer at 240 nm
Temperature: ambient
1.1.3 Reagents
1.1.3 1 Sodium hydroxide, 5% (m/v).
1.1.3.2 Sulfuric acid solution, 10% (m/v).
1.1.3.3 Indicator reagent: methyl red 100 mg, and methylene blue 50 mg, 100
mL alcohol, 50% (v/v).
1.1.3.4 Acetonitrile for chromatography.
1.1.3.5 Derivatizing reagent: 3,5-dinitrobenzoyle chloride, 10% (m/v), in
acetonitrile.
1.1.3.6 Buffer solution of sodium acetate, pH 4.7: mix 1 volume of sodium
acetate solution, NaC2H3O2.3H2O, 1 M, with 1 volume acetic acid solution,
1 M.
OIV-MA-AS4-02F : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
1.1.3.7 Sodium azide, NaN3.
OIV-MA-AS4-02F : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
1.1.4 Procedure
1.1.4.1 Preparation of the sample.
Into a spherical flask with a ground glass neck, place 100 mL of wine, distill
by plunging the end of the condenser in 10 mL of 5% sodium hydroxide
solution (1.1.3), to which are added a few drops of reagent indicator. Distill
until 40-50 mL of distillate is recovered.
Transfer the distillate into another spherical flask (1.1.2.2), rinse the flask
twice with 20 mL of water and add water to bring to 100 mL. To eliminate the
ethanol, attach the flask to the distillation apparatus and eliminate about 50 mL
of distillate (reduce the volume by half).
Cool the flask completely. Acidify with 10% sulfuric acid. Distill, recover the
distillate into a 10 mL flask with a ground glass stopper containing 1 mL of
water, and immerse in an iced bath. Stop the distillation when the total volume
reaches 10 mL.
1.1.4.2 Derivitization
Mix 1 mL distillate (1.1.4.1), 0.5 mL of acetonitrile, 0.2 mL buffer solution
and 30 µL of derivatizing reagent and stir vigorously; leave for five minutes.
1.1.4.3 Chromatography
Inject 20 µL in accordance with the conditions specified, the hydrazoic acid
derivative has a retention time of about 11 minutes. Detection limit: 0.01 mg/L.
Note: Sometimes another substance not derivatized can simulate hydrazoic
acid. It is necessary to verify a positive result as follows: inject 20 µL of
distillate directly; a disappearance of the peak indicates the presence of
hydrazoic acid.
1.1.5 Calculation
To determine the concentration of sodium azide, compare the sample response
to that of the standard solution after derivatization. Take into account the
concentration factor 10 of the sample of wine at the time of analysis.
1.2 Colorimetric method
1.2.1 Principle
Hydrazoic acid, which is very volatile, is separated by double distillation,
permitting the elimination of ethanol, acetic acid and sulfur dioxide. Then the
amount is determined colorimetrically after forming a colored complex with
ferric chloride (maximum absorbance at 465 nm).
1.2.2 Apparatus
1.2.2.1 Simple distillation apparatus, consisting of a 500 mL flask with a
ground glass neck and a condenser ending in a pointed tube
1.2.2.2 Spectrophotometer with optical glass cells 1 cm path length
OIV-MA-AS4-02F : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
1.2.3 Reagents
1.2.3.1 Sodium hydroxide solution, 1 M
1.2.3.2 Sulfuric acid, 1 M
1.2.3.3 Hydrogen peroxide, 3% (v/v), whose strength must be adjusted just
before use using a solution of potassium permanganate, 0.02 M; where p mL
equals the volume which oxidizes 1 mL of the hydrogen peroxide solution, 3%
1.2.3.4 Ferric chloride solution at 20 g per liter of Fe III: (weigh 96.6 g of
FeCl3.6H2O, or more as this salt is very hygroscopic; control the concentration
of Fe III of the solution and adjust if necessary to 20 ± 0.5 g per liter)
1.2.3.5 Stock solution of sodium azide, NaN3, at 1 g per liter in distilled water
1.2.3.6 200 mg per liter sodium azide solution prepared by dilution of the
solution at 1 g per liter
1.2.4. Procedure
a) Into a 500 mL flask with a ground glass neck, place 200 mL of wine, distill,
recover the distillate in a 50 mL volumetric flask, containing 5 mL water,
which is immersed in an iced bath. Stop the distillation when the total
volume reaches about 50 mL.
b) Transfer quantitatively the distillate into another 500 mL flask with a
stopper and rinse the 50 mL flask twice with 20 mL of water.
Neutralize using 1 M sodium hydroxide solution (1.2.3.1) (using pH
indicator paper).
Acidify using 10 mL 1 M sulfuric acid (1.2.3.2), mix, then oxidize the
sulfur dioxide by adding 3% hydrogen peroxide solution (1.2.3.3.).
If the wine contains S mg per liter of sulfur dioxide, and if p mL is the
volume of 0.02 M potassium permanganate solution necessary to oxidize 1
mL of 3% hydrogen peroxide solution, then for 200 mL of wine use the
following calculation:
S
 S mL of H2O2 solution
5 x 3.2p 16p
Bring the volume to about 200 mL by addition of distilled water.
Distill, recover the distillate in a 50 mL glass flask containing 5 mL
distilled water, which is immersed in an ice bath; stop the distillation before
the measurement line, bring back to ambient temperature and adjust the
volume to 50 mL.
c) Add 0.5 mL (measured exactly) of ferric chloride solution, mix and
measure immediately (maximum delay 5 min.) the absorbance at 465 nm in
OIV-MA-AS4-02F : R2009
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Detection of preservatives and fermentation inhibitors
a 1 cm cell; the zero of the apparatus is set using a blank composed of 50
mL of water added to 0.5 mL of ferric chloride solution.
d) Preparation of the standard curve.
Into each of five 50 mL volumetric flasks add 1, 2, 3, 4, and 5 mL of 200
mg/L sodium azide solution respectively, bring the volume to 50 mL with
distilled water, add 0.5 mL of ferric chloride solution and measure the
absorbance at 465 nm.
These solutions contain 4, 8, 12, 16, 20 mg of sodium azide per liter. The
corresponding concentrations are 1, 2, 3, 4, and 5 mg per liter of wine.
The typical curve of absorbance variation as a function of concentration is a
straight line passing through the origin.
1.2.5 Calculation
Plot the absorbance read for the sample analyzed on the straight line and
interpolate the concentration of sodium azide in mg/L of wine.
BIBLIOGRAPHY
HPLC method:
SEARIN S.J. & WALDO R.A., J. Liquid. Chrom., 1982, 5(4), 597-604.
BATTAGLIA R. & MITISKA J., Z. Lebensm. Unters. Forsch., 1986, 182, 501-502.
Colorimetric method:
CLERMONT S. & CHRETIEN D., F.V., O.I.V., 1977, n° 627.
OIV-MA-AS4-02F : R2009
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Differentiation of fortified musts and sweet fortified wines
Method OIV-MA-AS5-01
Type IV method
Differentiation of Fortified Musts
and Sweet Fortified Wines
1. Principle of the method
1.1 Method of screening
The product definitions given by the O.I.V. (International Code of Enological
Practices) imposes for fortified wines, a minimum of 4% acquired alcohol
derived naturally by fermentation; and allows, for fortified musts, a maximum
of 1% acquired alcohol. Consequently, these products may be differentiated
by identifying their fermentation by-products via gas chromatography.
This method is applicable only if, as the definition anticipates, the alcohol used
for production of the fortified musts is neutral.
1.2 Scientific investigation of citramalic acid by thin layer chromatography.
The presence of citramalic acid characterizes sweet fortified wines. Its
identification is carried out by thin layer chromatography after separation of
the sugars with the use of an ion exchange column.
2. Method of screening
2.1 Apparatus
Gas chromatograph with:
 Flame ionization detector,
 3 m stainless steel column, 2 mm interior diameter,
 Stationary phase: Carbowax 20 M 20%,
 Support: Chromosorb W 60/80 mesh.
Chromatography conditions:
 temperatures:
injector: 210°C
detector: 250°C
oven: isothermal at 70°C for 6 minutes; then programmed at 6°C/minute;
upper temperature limit: 170°C
Other types of columns can be used.
The procedure described below is given as an example.
2.2 Procedure
2.2.1. Sample preparation
Carry out a separation according to the following conditions: To 25 mL of
sample (fortified must or sweet fortified wine) are added to 7 mL ethanol and
OIV-MA-AS5-01 : R2009
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Differentiation of fortified musts and sweet fortified wines
15 g of ammonium sulfate, (NH4)2SO4, agitate. Allow to settle to obtain
separation of the phases.
2.2.2 Chromatography
Inject 2 µL of the organic phase and carry out the chromatography in
accordance with the conditions indicated above.
The chromatogram of the fortified wine is differentiated by the presence of the
peaks of the secondary products of alcoholic fermentation.
3. Investigation of citramalic acid by thin layer chromatography.
3.1 Apparatus
3.1.1 Glass column about 300 mm in length and 10-11 mm interior diameter
supplied with a flow regulator (stopcock)
3.1.2 Rotary vacuum evaporator
3.1.3 Oven at 100 °C
3.1.4 Chromatography developing chamber
3.1.5 Micrometric syringe or micropipette
3.2 Reagents
3.2.1 Formic acid solution, 4 M, containing 150.9 mL formic acid (20 = 1.227
g/mL) per liter.
3.2.2 Plates for chromatography ready to use with a layer of cellulose powder
(for example MN 300) (20 x 20 cm).
3.2.3 Solvent:
iso-Propyl alcohol containing 1 g/L bromophenol blue .... 5 vol.
Eucalyptol ...............................................…......……………………. 5 vol.
Formic acid (20= 1.227 g/mL) .......................…....…………..
2 vol.
Saturate the solvent with water and allow to stand for 24 hours before use.
3.2.4 Standard solutions.
Prepare an aqueous solution of:
Citramalic acid .................................................………………
Lactic acid .....................................................………………...
Citric acid ....................................................…………………..
Tartaric acid ...................................................………………..
Malic acid .....................................................………………….
0.25 g/L
0.5 g/L
0.5 g/L
1.0 g/L
1.0 g/L
3.3 Procedure
3.3.1 Preparation of the ion exchange column.
See chapter on Tartaric acid, usual method in 3.3.1.
3.3.2 Isolation of the organic acid of citramalic acid
OIV-MA-AS5-01 : R2009
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Differentiation of fortified musts and sweet fortified wines
Proceed as indicated in the chapter Tartaric acid, usual method in 3.3.2. for
the fixation of organic acids on the ion exchanger.
Then elute the fixed acids using the 4 M formic acid solution (100 mL),
collecting the eluate in a 100 mL volumetric flask.
Concentrate the eluate dry in a rotary evaporator at 40°C recovering the
residue with 1 mL of distilled water.
3.3.3 Chromatography
The cellulose plate must be activated by placing it in the oven at 100°C for 2
hours.
Deposit on the starting line of the cellulose plate in a band 2 cm wide, 10 µL of
this solution as well as 10 µL of the standard solutions of citramalic acid and
the other organic acids.
Place the plate in the chromatography bath, above the solvent, for 45 minutes.
Proceed with the development and let the solvent migrate to a height of 15 cm.
3.3.4 Development of the chromatogram
Maintain the plate at ambient temperature under an air current, until the formic
acid of the solvent is eliminated. Yellow spots appear on a blue background,
indicating the presence of the acids.
Detect the presence or absence of citramalic acid in the product analyzed by
comparing the spots of this chromatogram to the spots of standard solutions of
citramalic acid and the other organic acids.
BIBLIOGRAPHY
Method of Screening:
HARVALIA A., F.V., O.I.V., 1980, n° 728 bis.
Chromatography of citramalic acid:
DIMOTAKI-KOURAKOU V., Ann. Fals. Exp. Chim., 1960, 53, 149.
DIMOTAKI-KOURAKOU V., C. R. Ac. Sci., Paris 1962, 254, 4030.
CARLES J., LAMAZOU-BETBEDER M. & PECH M., C. R. Ac. Sci., Paris 1958,
246, 2160.
CASTINO M., Riv. Vit. Enol., 1967, 6, 247.
KOURAKOU V., F.V., O.I.V., 1977, n° 642.
JUNGE Ch F.V., O.I.V., 1978, n° 679.
ROUEN J., F.V., O.I.V., 1979, n° 691.
OIV-MA-AS5-01 : R2009
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Annex B
Certificates of analysis
OIV-MA-ANNEX-B
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS -OIV
Rules for the implementation of the analytical methods
Rules for the implementation of the analytical methods
The control of the quality of wines should always allow, on one hand, a sensory
examination and, on the other hand, the determination of the essential
characteristic elements of their composition.
Sensory analysis has not been studied in the present book; it is left to the
evaluation of the various countries, but is required in every case.
With regard to the elements of the composition of wines, three types of
determinations can or must be performed:
1. The determinations that serve to identify the wines and can serve as basis of
commercial transactions (Certificate no. 1);
2. The determinations that permit us to ascertain satisfactorily the qualities and
characteristics of a wine and which, in this manner, correspond to trade
practices (Certificate no. 2).
Determinations other than anticipated in the Certificates numbers 1 and 2 can
be required within a contractual framework.
3. A third Certificate (no. 3) can be considered which would contain specific
determinations that are only carried out on an exceptional or special basis.
Resorting to the determinations aimed at Certificate no. 2 could be such as to
exonerate the operators from liability.
The recourse to the determinations of the Certificate no. 3 could be such as to
exonerate importers from liability.
When the public health is involved, other determinations can be required either by
the OIV, the public authorities, or by all interested parties when serious doubts
appear in the industry or among consumers.
The public health exception may be submitted, by all parties interested, to the
special group of scientific experts of the Office according to an emergency
procedure.
The analytical determinations are performed, when they exist, according to the
methods described in the present book.
OIV-MA-B1-01
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Certificates of analysis
Certificates of Analysis
Certificate No. 1
- Color
- Clarity
- Specific gravity at 20C
- Alcoholic content at 20C
- Total dry extract g/L
- Sugar g/L
- Total sulfur dioxide mg/L
- pH
- Total acidity meq/L
- Volatile acidity meq/L
- Test for malvidin diglucoside
- Over pressure measurement of carbon dioxide in sparkling wines
- Differentiation of very sweet wine and fortified must in the case of
sweet wines
Certificate No. 2
Certificate No. 1 is completed and the following determinations are added:
- Ash and alkaline ash g/L
- Potassium g/L
- Iron mg/L
- Copper mg/L
- Free sulfur dioxide mg/L
- Sorbic acid mg/L
- Verification of malolactic fermentation
- Citric acid mg/L
- Tartaric acid g/L
- Folin-Ciocalteu Index
- Chromatic Indexes
The following determinations are optional:
- Excess sodium mg/L
- Calcium, magnesium mg/L
- Sulfates mg/L
- Test of fermentability
- Test for artificial colorants
OIV-MA-B1-02
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Annex C
Maximum acceptable limits
of various substances
OIV-MA-ANNEX-C
COMMENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Maximum acceptable limits of various substances contained in wine
Maximum acceptable limits of various substances
contained in wine
(2010 Issue)
Citric acid:
1 g/L
Volatile acidity:
20 milliequivalents/L
The volatile acidity of various specially fortified
old wines (wines subject to special legislation and
controlled by the government) may exceed this
limit.
Arsenic:
0.2 mg/L
Boron:
80 mg/L (expressed as boric acid)
Bromine:
1 mg/L (limit exceeded by way of exception in
wines from certain vineyards with a brackish
subsoil).
Cadmium:
0.01 mg/L
Copper:
1 mg/L
Diethylene glycol:
 10 mg/L, to the quantification limit
Malvidol diglucoside:
15 mg/L (determined by the quantitative method
diglucoside described in the Compendium)
Silver
< 0.1 mg/L
OIV-MA-C1-01
1
COMMENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Maximum acceptable limits of various substances contained in wine
Total sulfur dioxide at the time of sale to the consumer: (oeno 9/98)




150 mg/L for red wines containing a maximum
of 4 g/L of reducing substances.
200 mg/L for white and rosé wines containing a
maximum of 4 g/L of reducing substances.
300 mg/L: red, rosé and white wines containing
more than 4 g/L of reducing substances.
400 mg/L: in exceptional cases some sweet white
wines.
Ethanediol /Ethylene glycol:
 10 mg/L
Fluoride: (oeno 8/91)
1 mg/L except for wines coming from vineyards
treated in conformity with national law, with
cryolite in which case, the level of fluoride must
not exceed 3 mg/L.
Methanol:
400 mg/L for red wines
(oeno 19/2004)
250 mg/L for white and rosé wines
Ochratoxin A :
2 µg/L (for wines obtained as from the 2005 harvest)
(CST 1/2002)
Lead: (oeno 13/06)
0.15 mg/L for wine made, starting from the 2007
harvest year
Propan-1,2-diol/propylene
glycol
Still wines : = 150 mg/L
Sparkling wines : = 300 mg/L
(oeno 20/2003)
OIV-MA-C1-01
2
COMMENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Maximum acceptable limits of various substances contained in wine
Excess sodium:
(oeno 12/2007)
80 mg/L
Sulfates:
(expressed
sulfate)
1 g/L
as
Zinc
OIV-MA-C1-01
potassium However this limit is raised to:
- for wines which have undergone a
maturing period in casks for at least 2
years
- for sweetened wines
- for wines obtained by the addition to
the musts or wine of alcohol or potable
spirit







1.5
g/L

- for wines with added concentrated 
musts

- for naturally sweet wines

2.0
g/L
- for wines obtained under a film "sous 

voile"

2.5
g/L
5 mg/L
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS-OIV
Annex D
Advices
OIV-MA-ANNEX-D
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Gluconic acid
Gluconic acid
Resolution oeno 4/91
Gluconic acid is always present in musts and wines.
In wines derived from a sound, mature harvest, its level does not exceed 200—300
mg/L.
Gluconic acid increases through over—ripening by raisining and especially by the
intervention of Botrytis cinerea.
Its presence at higher levels in wines — other than wines infected with noble rot of
which gluconic acid is a characteristic constituent — cannot be considered a sign of
bad quality linked to a harvest seized with gray rot, which must be demonstrated by
other means. ln fact, by appropriate vinification techniques, it is possib1e to obtain
wines of quality in this case.
As to fraud by addition of gluconic acid, this is not a factor to be taken into account
since there is no reason for it.
OIV-MA-D1-01
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
Characterization of wines resulting from overpressing
Characterization of wines resulting from overpressing
Resolution oeno 5/91
NOTICE
In view of the results of the discussions concerning the tests on DESCRIPTIVE
CHARACTERISTICS OF WINES RESULTING FROM OVERPRESSING, the
experts have confirmed that, for the group of tests done, the behavior of wines is
very different depending on the variety. This renders immpossib1e all
interpretation concerning wines coming from several varieties.
Moreover, the effects of the different methods of pressing and of vinification
techniques, such as prefermentation maceration, must be taken into account.
Studies must be pursued to show wines resulting from overpressing and a
definition of overpressing sought after.
OIV-MA-D1-02
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS - OIV
The level of sodium and chloride ions in wines
The level of sodium and chloride ions in wines
Resolution Oeno 6/91
NOTICE
The level of Chloride and sodium ions in wines essentially depends on the
geographic, geologic and climatic conditions of vine culture.
As a general rule, the levels of these ions are low.
the content of these elements is increased in wines coming from vineyards which
are near the sea coast, which have brackish sub—soil or which have arid ground
irrigated with salt water and the molar ratio cf Cl/Na+ therefore varies significantly
and can even have a value close to one (1) which could imply the addition of salt
(NaCl) to the wine.
When wine contains excess sodium (excess sodium is equal to the content of
sodium ions less the content of chloride ions expressed as sodium), it is generally
less than 60 mg/L, a limit which may be exceeded in exceptional cases.
The laboratories and official control agencies, confronted with elevated levels of Cl
and/or Na+, must take the above conclusions into account and possib1y make
inquiries to the official agencies of the country of origin before expelling these
wines.
OIV-MA-D1-03
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS OIV
Principle of Validation – Routine Methods – Reference Methods
Principle of validation of routine methods with respect to
reference methods
(Resolution Oeno 7/98)
The OIV acknowledges the existence of methods of analysis of wines in addition
to those described in the Summary of International Methods of Analysis of Wines
and Musts, of common methods most often automated. These methods are
economically and commercially important because they permit maintaining a
complete and efficient analytical framework around the production and marketing
of wine. Moreover, these methods allow the use of modern means of analysis and
the development and adaptation of techniques of analysis.
In order to allow laboratories to use these methods and to insure their linkage to
methods described within the Summary, the OIV decides to establish a plan of
evaluation and validation by a laboratory of an alternative, common method,
mechanized or not with respect to a reference method described in the Summary of
International Methods of Analysis of Wines and Musts.
This principle, which will be adapted to the particular situation of the analysis of
wines and musts, will take its inspiration from international standards in current
use and allow the laboratory to assess and validate its alternative method in two
ways:
OIV-MA-AS1-05 : R1998
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Collaborative Study
The purpose of the collaborative study is to give a quantified indication of the
precision of method of analysis, expressed as its repeatability r and reproducibility
R.
Repeatability: the value below which the absolute difference between two single
test results obtained using the same method on identical test material, under the
same conditions (same operator, same apparatus, same laboratory and a short
period of time) may be expected to lie within a specified probability.
Reproducibility: the value below which the absolute difference between two single
test results obtained using the same method on identical test material, under
different conditions (different operators, different apparatus and/or different
laboratories and/or different time) may be expected to lie within a specified
probability.
The term "individual result" is the value obtained when the standardized trial
method is applied, once and fully, to a single sample. Unless otherwise stated, the
probability is 95%.
General Principles
 The method subjected to trial must be standardized, that is, chosen from the
existing methods as the method best suited for subsequent general use.
 The protocol must be clear and precise.
 The number of laboratories participating must be at least ten.
 The samples used in the trials must be taken from homogeneous batches of
material.
 The levels of the analyte to be determined must cover the concentrations
generally encountered.
 Those taking part must have a good experience of the technique employed.
 For each participant, all analyses must be conducted within the same laboratory
by the same analyst.
 The method must be followed as strictly as possible. Any departure from the
method described must be documented.
 The experimental values must be determined under strictly identical conditions:
on the same type of apparatus, etc.
 They must be determined independently of each other and immediately after
each other.
 The results must be expressed by all laboratories in the same units, to the same
number of decimal places.
 Five replicate experimental values must be determined, free from outliers. If an
experimental value is an outlier according to the Grubbs test, three additional
measurements must be taken.
OIV-MA-AS1-07 : R2000
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Statistical Model
The statistical methods set out in this document are given for one level
(concentration, sample). If there are a number of levels, the statistical evaluation
must be made separately for each. If a linear relationship is found (y = bx or y = a
+ bx) as between the repeatability (r) or reproducibility (R) and the concentration
( x ), a regression of r (or R) may be run as a function of x
The statistical methods given below suppose normally-distributed random values.
The steps to be followed are as follows:
A/ Elimination of outliers within a single laboratory by Grubbs test. Outliers are
values which depart so far from the other experimental values that these
deviations cannot be regarded as random, assuming the causes of such
deviations are not known.
B/ Examine whether all laboratories are working to the same precision, by
comparing variances by the Bartlett test and Cochran test. Eliminate those
laboratories for which statistically deviant values are obtained.
C/ Track down the systematic errors from the remaining laboratories by a
variance analysis and by a Dixon test identify the extreme outlier values.
Eliminate those laboratories for which the outlier values are significant.
D/ From the remaining figures, calculate standard deviation of repeatability); Sr.,
and repeatability r standard deviation of reproducibility SR and reproducibility
R.
Notation:
The following designations have been chosen:
m
Number of laboratories
i(i = 1, 2... m)
Index (No. of the laboratory)
ni
Number of individual values from the ith laboratory
m
N  n i
Total number of individual values
x(i = 1, 2... ni)
Individual value of the ith laboratory
i l
xi 
x
si
1
ni
ni
x
i =1
Mean value of the ith laboratory
i
1 m ni
 xi
N i =1 i =1

1 ni
 x i x i
ni-1 i 1
Total mean value

2
Standard deviation of the ith laboratory
OIV-MA-AS1-07 : R2000
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
A/ Verification of outlier values within one laboratory
After determining five individual values x , a Grubbs test is performed at the
i
laboratory, to identify the outliers’ values.
Test the null hypothesis whereby the experimental value with the greatest absolute
deviation from the mean is not an outlier observation.
xi* xi
si.
Calculate PG =
x *i = suspect value
Compare PG with the corresponding value shown in Table 1 for P = 95%.
If PG < value as read, value x * is not an outlier and s can be calculated.
i
i
*
If PG > value as read, value x probably is an outlier therefore make a further
i
three determinations.
Calculate the Grubbs test for x * with the eight determinations.
i
If PG > corresponding value for P = 99%, regard x * as a deviant value and
i
calculate s without x * .
i
i
B/ Comparison of variances among laboratories
- Bartlett Test
The Bartlett test allows us to examine both major and minor variances. It serves to
test the null hypothesis of the equality of variances in all laboratories, as against
the alternative hypothesis whereby the variances are not equal in the case of some
laboratories.
At least five individual values are required per laboratory.
Calculate the statistics of the test:
PB 
m
1

2


N

m
ln

ln 2
S

r  f i si 
C
i 1

m
1
1
 f  N m
C  i 1
i
3m1
1
m
f s
2
i I
S
2
r
 i 1
N m
OIV-MA-AS1-07 : R2000
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
fi = ni - 1 degrees of freedom of s .
i
Compare PB with the value x2 indicated in table 2 at m - 1 degrees of freedom.
If PB > the value in the table, there are differences among the variances.
The Cochran test is used to confirm that the variance from one laboratory is
greater than that from other laboratories.
Calculate the test statistics:
PC 
2
si max
m
s
2
i
i 1
Compare PC with the value shown in table 3 for m and n at P = 99%.
i
If PC > the table value, the variance is significantly greater than the others.
If there is a significant result from the Bartlett or Cochran tests, eliminate the
outlier variance and calculate the statistical test again.
In the absence of a statistical method appropriate to a simultaneous test of several
outlier values, the repeated application of the tests is permitted, but should be used
with caution.
If the laboratories produce variances that differ sharply from each other, an
investigation must be made to find the causes and to decide whether the
experimental values found by those laboratories are to be eliminated or not. If
they are, the coordinator will have to consider how representative the remaining
laboratories are.
If statistical analysis shows that there are differing variances, this shows that the
laboratories have operated the methods at varying precisions. This may be due to
inadequate practice or to lack of clarity or inadequate description in the method.
C/ Systematic errors
Systematic errors made by laboratories are identified using either Fischer's method
or Dixon's test.
R .A. Fischer variance analysis
This test is applied to the remaining experimental values from the laboratories with
an identical variance.
The test is used to identify whether the spread of the mean values from the
laboratories is very much greater than that for the individual values expressed by
the variance among the laboratories ( s2Z ) or the variance within the laboratories
( s2I ).
OIV-MA-AS1-07 : R2000
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Calculate the test statistics:
PF 
2
sZ
2
sI
 
s 
1 m
 n i x
m-1 i =1 xi
s 
1 m ni
 xi  xi
N m i =1 i =1
2
Z
2
I
2


2
Compare PF with the corresponding value shown in table 4 (distribution of F)
where fi = fZ = m - 1 and f2 = f1 = N - m degrees of freedom.
If PF > the table value, it can be concluded that there are differences among the
means, that is, there are systematic errors.
Dixon test
This test enables us to confirm that the mean from one laboratory is greater or
smaller than that from the other laboratories.
Take a data series Z(h), h = 1,2,3...H, ranged in increasing order.
Calculate the statistics for the test:
Z 2Z 1
Z H Z 1
or
Z H Z H 1
Z H Z 1
Z 2Z 1
Z H 1Z 1
or
Z H Z H 1
Z H Z 2
Z 3Z 1
Z H 2Z 1
or
Z H Z H 2
Z H Z 3
3 to 7
Q10
8 to 12
Q11
13 plus
Q22
Compare the greatest value of Q with the critical values shown in table 5.
If the test statistic is > the table value at P = 95%, the mean in question can be
regarded as an outlier.
If there is a significant result in the R A Fischer variance analysis or the Dixon
test, eliminate one of the extreme values and calculate the test statistics again with
the remaining values. As regards repeated application of the tests, see the
explanations in paragraph (B).
OIV-MA-AS1-07 : R2000
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
If the systematic errors are found, the corresponding experimental values
concerned must not be included in subsequent computations; the cause of the
systematic error must be investigated.
D/Calculating repeatability (r) and reproducibility (R).
From the results remaining after elimination of outliers, calculate the standard
deviation of repeatability sr and repeatability r, and the standard deviation of
reproducibility sR and reproducibility R, which are shown as characteristic values
of the method of analysis.
sr 
1 m
2
f is i

N m i 1
r sr2 2
sR
1 2
sZ a 1s2I 
a
Rs R2 2
a
2
m
1 
n i 
N





m1 
i 1 N 
If there is no difference between the means from the laboratories, then there is no
difference between sr and sR or between r and R. But, if we find differences
among the laboratory means, although these may be tolerated for practical
considerations, we have to show sr and sR, and r and R.
OIV-MA-AS1-07 : R2000
6
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
BIBLIOGRAPHY
AFNOR, norme NFX06041, Fidélitè des méthodes d'essai. Détermination de la
répétabilité et de la reproductibilité par essais interlaboratoires.
DAVIES O. L., GOLDSMITH P.l., Statistical Methods in Research and
Production, Oliver and Boyd, Edinburgh, 1972.
GOETSCH F. H., KRÖNERT W., OLSCHIMKE D., OTTO U., VIERKÖTTER
S., Meth. An., 1978, No 667.
GOTTSCHALK G., KAISER K. E., Einführung in die Varianzanalyse und
Ringversuche, B-1 Hoschultaschenbücher, Band 775, 1976.
GRAF, HENNING, WILRICH, Statistische Methoden bei textilen
Untersuchungen, Springer Verlag, Berlin, Heidelberg, New York, 1974.
GRUBBS F. E., Sample Criteria for Testing Outlying Observations, The Annals of
Mathematical Statistics, 1950, vol. 21, p 27-58.
GRUBBS F. E., Procedures for Detecting Outlying Observations in Samples,
Technometrics, 1969, vol. 11, No 1, p 1-21.
GRUBBS F. E. and BECK G., Extension of Sample Sizes and Percentage Points
for Significance Tests of Outlying Observations, Technometrics, 1972, vol. 14,
No 4, p 847-854.
ISO, norme 5725.
KAISER R., GOTTSCHALK G., Elementare Tests zur Beurteilung von
Messdaten, B-I Hochschultaschenbücher, Band 774, 1972.
LIENERT G. A., Verteilungsfreie Verfahren in der Biostatistik, Band I, Verlag
Anton Haine, Meisenheim am Glan, 1973.
NALIMOV V. V., The Application of Mathematical Statistics to Chemical
Analysis, Pergamon Press, Oxford, London, Paris, Frankfurt, 1963.
SACHS L., Statistische Auswertungsmethoden, Springer Verlag, Berlin,
Heidelberg, New York, 1968.
OIV-MA-AS1-07 : R2000
7
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Table 1 - Critical values for the Grubbs test
ni
P = 95%
3
4
5
6
7
8
9
10
11
12
1,155
1,481
1,715
1,887
2,020
2,126
2,215
2,290
2,355
2,412
P 99%
1,155
1,496
1,764
1,973
2,139
2,274
2,387
2,482
2,564
2,636
Table 2 – Critical values for the Bartlett test (P = 95%)
f(m - 1)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
X2
3,84
5,99
7,81
9,49
11,07
12,59
14,07
15,51
16,92
18,31
19,68
21,03
22,36
23,69
25,00
26,30
27,59
28,87
30,14
31,41
OIV-MA-AS1-07 : R2000
f(m - 1)
X2
21
22
23
24
25
26
27
28
29
30
35
40
50
60
70
80
90
100
32,7
33,9
35,2
36,4
37,7
38,9
40,1
41,3
42,6
43,8
49,8
55,8
67,5
79,1
90,5
101,9
113,1
124,3
8
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Table 3 – Critical values for the Cochran test
2
3
4
5
6
7
8
9
10
ni =
99%
0.993
0.968
0.928
0.883
0.838
0.794
0.754
0.718
2
95%
0.967
0.906
0.841
0.781
0.727
0.680
0.638
0.602
ni= 3
99%
95%
0.995 0.975
0.942 0.871
0.864 0.768
0.788 0.684
0.722 0.616
0.664 0.561
0.615 0.516
0.573 0.478
0.536 0.445
ni = 4
99%
95%
0.979 0.939
0.883 0.798
0.781 0.684
0.696 0.598
0.626 0.532
0.568 0.480
0.521 0.438
0.481 0.403
0.447 0.373
ni = 5
99%
95%
0.959 0.906
0.834 0.746
0.721 0.629
0.633 0.544
0.564 0.480
0.508 0.431
0.463 0.391
0.425 0.358
0.393 0.331
ni = 6
99%
95%
0.937 0.877
0.793 0.707
0.676 0.590
0.588 0.506
0.520 0.445
0.466 0.397
0.423 0.360
0.387 0.329
0.357 0.303
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
0.684
0.653
0.624
0.599
0.575
0.553
0.532
0.514
0.496
0.480
0.465
0.450
0.437
0.425
0.413
0.402
0.391
0.382
0.372
0.363
0.355
0.347
0.339
0.332
0.325
0.318
0.312
0.306
0.300
0.294
0.570
0.541
0.515
0.492
0.471
0.452
0.434
0.418
0.403
0.389
0.377
0.365
0.354
0.343
0.334
0.325
0.316
0.308
0.300
0.293
0.286
0.280
0.273
0.267
0.262
0.256
0.251
0.246
0.242
0.237
0.504
0.475
0.450
0.427
0.407
0.388
0.372
0.356
0.343
0.330
0.318
0.307
0.297
0.287
0.278
0.270
0.262
0.255
0.248
0.241
0.235
0.229
0.224
0.218
0.213
0.208
0.204
0.200
0.196
0.192
0.418
0.392
0.369
0.349
0.332
0.316
0.301
0.288
0.276
0.265
0.255
0.246
0.238
0.230
0.222
0.215
0.209
0.202
0.196
0.191
0.186
0.181
0.177
0.172
0.168
0.165
0.161
0.157
0.154
0.151
0.366
0.343
0.322
0.304
0.288
0.274
0.261
0.249
0.238
0.229
0.220
0.212
0.204
0.197
0.190
0.184
0.179
0.173
0.168
0.164
0.159
0.155
0.151
0.147
0.144
0.140
0.137
0.134
0.131
0.128
0.332
0.310
0.291
0.274
0.259
0.246
0.234
0.223
0.214
0.205
0.197
0.189
0.182
0.176
0.170
0.164
0.159
0.154
0.150
0.145
0.141
0.138
0.134
0.131
0.127
0.124
0.121
0.119
0.116
0.114
m
0.417
0.392
0.371
0.352
0.335
0.319
0.305
0.293
0.281
0.270
0.261
0.252
0.243
0.235
0.228
0.221
0.215
0.209
0.203
0.198
0.193
0.188
0.184
0.179
0.175
0.172
0.168
0.164
0.161
0.158
OIV-MA-AS1-07 : R2000
0.348
0.326
0.307
0.291
0.276
0.262
0.250
0.240
0.230
0.220
0.212
0.204
0.197
0.191
0.185
0.179
0.173
0.168
0.164
0.159
0.155
0.151
0.147
0.144
0.140
0.137
0.134
0.131
0.129
0.126
0.308
0.288
0.271
0.255
0.242
0.230
0.219
0.209
0.200
0.192
0.185
0.178
0.172
0.166
0.160
0.155
0.150
0.146
0.142
0.138
0.134
0.131
0.127
0.124
0.121
0.119
0.116
0.113
0.111
0.108
0.281
0.262
0.246
0.232
0.220
0.208
0.198
0.189
0.181
0.174
0.167
0.160
0.155
0.149
0.144
0.140
0.135
0.131
0.127
0.124
0.120
0.117
0.114
0.111
0.108
0.106
0.103
0.101
0.099
0.097
9
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Table 4 – Critical values for the F-Test (P=99%)
f1
f2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
40
50
60
70
80
90
100
200
500
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
4052 4999 5403 5625 5764 5859 5928 5981 6023 6056 6083 6106 6126 6143 6157
98.5 99.0 99.2 99.3 99.3 99.3 99.4 99.4 99.4 99.4 99.4 99.4 99.4 99.4 99.4
34.1 30.8 29.4 28.7 28.2 27.9 27.7 27.5 27.3 27.2 27.1 27.1 27.0 26.9 26.9
21.2 18.0 16.7 16.0 15.5 15.2 15.0 14.8 14.7 14.5 14.5 14.4 14.3 14.2 14.2
16.3 13.3 12.1 11.4 11.0 10.7 10.5 10.3 10.2 10.1 9.96 9.89 9.82 9.77 9.72
13.7 10.9 9.78 9.15 8.75 8.47 8.26 8.10 7.98 7.87 7.79 7.72 7.66 7.60 7.56
12.2 9.55 8.45 7.85 7.46 7.19 6.99 6.84 6.72 6.62 6.54 6.47 6.41 6.36 6.31
11.3 8.65 7.59 7.01 6.63 6.37 6.18 6.03 5.91 5.81 5.73 5.67 5.61 5.56 5.52
10.6 8.02 6.99 6.42 6.06 5.80 5.61 5.47 5.35 5.26 5.18 5.11 5.05 5.01 4.96
10.0 7.56 6.55 5.99 5.64 5.39 5.20 5.06 4.94 4.85 4.77 4.71 4.65 4.60 4.56
9.64 7.20 6.21 5.67 5.31 5.07 4.88 4.74 4.63 4.54 4.46 4.39 4.34 4.29 4.25
9.33 6.93 5.95 5.41 5.06 4.82 4.64 4.50 4.39 4.30 4.22 4.16 4.10 4.05 4.01
9.07 6.70 5.74 5.21 4.86 4.62 4.44 4.30 4.19 4.10 4.02 3.96 3.90 3.86 3.82
8.86 6.51 5.56 5.04 4.69 4.46 4.28 4.14 4.03 3.94 3.86 3.80 3.75 3.70 3.66
8.68 6.36 5.42 4.89 4.56 4.32 4.14 4.00 3.89 3.80 3.73 3.67 3.61 3.56 3.52
8.53 6.23 5.29 4.77 4.44 4.20 4.03 3.89 3.78 3.69 3.62 3.55 3.50 3.45 3.41
8.40 6.11 5.18 4.67 4.34 4.10 3.93 3.79 3.68 3.59 3.52 3.46 3.40 3.35 3.31
8.29 6.01 5.09 4.58 4.25 4.01 3.84 3.71 3.60 3.51 3.43 3.37 3.32 3.27 3.23
8.18 5.93 5.01 4.50 4.17 3.94 3.77 3.63 3.52 3.43 3.36 3.30 3.24 3.19 3.15
8.10 5.85 4.94 4.43 4.10 3.87 3.70 3.56 3.46 3.37 3.29 3.23 3.18 3.13 3.09
8.02 5.78 4.87 4.37 4.04 3.81 3.64 3.51 3.40 3.31 3.24 3.17 3.12 3.07 3.03
7.95 5.72 4.82 4.31 3.99 3.76 3.59 3.45 3.35 3.26 3.18 3.12 3.07 3.02 2.98
7.88 5.66 4.76 4.26 3.94 3.71 3.54 3.41 3.30 3.21 3.14 3.07 3.02 2.97 2.93
7.82 5.61 4.72 4.22 3.90 3.67 3.50 3.36 3.26 3.17 3.09 3.03 2.98 2.93 2.89
7.77 5.57 4.68 4.18 3.85 3.63 3.46 3.32 3.22 3.13 3.06 2.99 2.94 2.89 2.85
7.72 5.53 4.64 4.14 3.82 3.59 3.42 3.29 3.18 3.09 3.02 2.96 2.90 2.86 2.81
7.68 5.49 4.60 4.11 3.78 3.56 3.39 3.26 3.15 3.06 2.99 2.93 2.87 2.82 2.78
7.64 5.45 4.57 4.07 3.75 3.53 3.36 3.23 3.12 3.03 2.96 2.90 2.84 2.79 2.75
7.60 5.42 4.54 4.04 3.73 3.50 3.33 3.20 3.09 3.00 2.93 2.87 2.81 2.77 2.73
7.56 5.39 4.51 4.02 3.70 3.47 3.30 3.17 3.07 2.98 2.91 2.84 2.79 2.74 2.70
7.31 5.18 4.31 3.83 3.51 3.29 3.12 2.99 2.89 2.80 2.73 2.66 2.61 2.56 2.52
7.17 5.06 4.20 3.72 3.41 3.19 3.02 2.89 2.78 2.70 2.62 2.56 2.51 2.46 2.42
7.07 4.98 4.13 3.65 3.34 3.12 2.95 2.82 2.72 2.63 2.56 2.50 2.44 2.39 2.35
7.01 4.92 4.07 3.60 3.29 3.07 2.91 2.78 2.67 2.59 2.51 2.45 2.40 2.35 2.31
6.96 4.88 4.04 3.56 3.25 3.04 2.87 2.74 2.64 2.55 2.48 2.42 2.36 2.31 2.27
6.92 4.85 4.01 3.53 3.23 3.01 2.84 2.72 2.61 2.52 2.45 2.39 2.33 2.29 2.24
6.89 4.82 3.98 3.51 3.21 2.99 2.82 2.69 2.59 2.50 2.43 2.37 2.31 2.27 2.22
6.75 4.71 3.88 3.41 3.11 2.89 2.73 2.60 2.50 2.41 2.34 2.27 2.22 2.17 2.13
6.69 4.65 3.82 3.36 3.05 2.84 2.68 2.55 2.44 2.36 2.29 2.22 2.17 2.12 2.07
6.63 4.61 3.78 3.32 3.02 2.80 2.64 2.51 2.41 2.32 2.25 2.18 2.13 2.08 2.04
OIV-MA-AS1-07 : R2000
10
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Table 4 – Critical values for the F-Test (P=99%) [Continued]
f1
f2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
40
50
60
70
80
90
100
200
500
16
17
18
19
20
30
40
50
60
70
80
100 200 500
6169 6182 6192 6201 6209 6261 6287 6303 6313 6320 6326 6335 6350 6361 6366
99.4 99.4 99.4 99.4 99.5 99.5 99.5 99.5 99.5 99.5 99.5 99.5 99.3 99.5 99.5
26.8 26.8 26.8 26.7 26.7 26.5 26.4 26.4 26.3 26.3 26.3 26.2 26.2 26.1 26.1
14.2 14.1 14.1 14.0 14.0 13.8 13.7 13.7 13.7 13.6 13.6 13.6 13.5 13.5 13.5
9.68 9.64 9.61 9.58 9.55 9.38 9.29 9.24 9.20 9.18 9.16 9.13 9.08 9.04 9.02
7.52 7.48 7.45 7.42 7.40 7.23 7.14 7.09 7.06 7.03 7.01 6.99 6.93 6.90 6.88
6.28 6.24 6.21 6.18 6.16 5.99 5.91 5.86 5.82 5.80 5.78 5.75 5.70 5.67 5.65
5.48 5.44 5.41 5.38 5.36 5.20 5.12 5.07 5.03 5.01 4.99 4.96 4.91 4.88 4.86
4.92 4.89 4.86 4.83 4.81 4.65 4.57 4.52 4.48 4.46 4.44 4.41 4.36 4.33 4.31
4.52 4.49 4.46 4.43 4.41 4.25 4.17 4.12 4.08 4.06 4.04 4.01 3.96 3.93 3.91
4.21 4.18 4.15 4.12 4.10 3.94 3.86 3.81 3.77 3.75 3.73 3.70 3.65 3.62 3.60
3.97 3.94 3.91 3.88 3.86 3.70 3.62 3.57 3.54 3.51 3.49 3.47 3.41 3.38 3.36
3.78 3.74 3.72 3.69 3.66 3.51 3.42 3.37 3.34 3.32 3.30 3.27 3.22 3.19 3.17
3.62 3.59 3.56 3.53 3.51 3.35 3.27 3.22 3.18 3.16 3.14 3.11 3.06 3.03 3.00
3.49 3.45 3.42 3.40 3.37 3.21 3.13 3.08 3.05 3.02 3.00 2.98 2.92 2.89 2.87
3.37 3.34 3.31 3.28 3.26 3.10 3.02 2.97 2.93 2.91 2.89 2.86 2.81 2.78 2.75
3.27 3.24 3.21 3.19 3.16 3.00 2.92 2.87 2.83 2.81 2.79 2.76 2.71 2.68 2.65
3.19 3.16 3.13 3.10 3.08 2.92 2.84 2.78 2.75 2.72 2.70 2.68 2.62 2.59 2.57
3.12 3.08 3.05 3.03 3.00 2.84 2.76 2.71 2.67 2.65 2.63 2.60 2.55 2.51 2.49
3.05 3.02 2.99 2.96 2.94 2.78 2.69 2.64 2.61 2.58 2.56 2.54 2.48 2.44 2.42
2.99 2.96 2.93 2.90 2.88 2.72 2.64 2.58 2.55 2.52 2.50 2.48 2.42 2.38 2.36
2.94 2.91 2.88 2.85 2.83 2.67 2.58 2.53 2.50 2.47 2.45 2.42 2.36 2.33 2.31
2.89 2.86 2.83 2.80 2.78 2.62 2.54 2.48 2.45 2.42 2.40 2.37 2.32 2.28 2.26
2.85 2.82 2.79 2.76 2.74 2.58 2.49 2.44 2.40 2.38 2.36 2.33 2.27 2.24 2.21
2.81 2.78 2.75 2.72 2.70 2.54 2.45 2.40 2.36 2.34 2.32 2.29 2.23 2.19 2.17
2.78 2.75 2.72 2.69 2.66 2.50 2.42 2.36 2.33 2.30 2.28 2.25 2.19 2.16 2.13
2.75 2.71 2.68 2.66 2.63 2.47 2.38 2.33 2.29 2.27 2.25 2.22 2.16 2.12 2.10
2.72 2.68 2.65 2.63 2.60 2.44 2.35 2.30 2.26 2.24 2.22 2.19 2.13 2.09 2.06
2.69 2.66 2.63 2.60 2.57 2.41 2.33 2.27 2.23 2.21 2.19 2.16 2.10 2.06 2.03
2.66 2.63 2.60 2.57 2.55 2.39 2.30 2.25 2.21 2.18 2.16 2.13 2.07 2.03 2.01
2.48 2.45 2.42 2.39 2.37 2.20 2.11 2.06 2.02 1.99 1.97 1.94 1.87 1.85 1.80
2.38 2.35 2.32 2.29 2.27 2.10 2.01 1.95 1.91 1.88 1.86 1.82 1.76 1.71 1.68
2.31 2.28 2.25 2.22 2.20 2.03 1.94 1.88 1.84 1.81 1.78 1.75 1.68 1.63 1.60
2.27 2.23 2.20 2.18 2.15 1.98 1.89 1.83 1.78 1.75 1.73 1.70 1.62 1.57 1.54
2.23 2.20 2.17 2.14 2.12 1.94 1.85 1.79 1.75 1.71 1.69 1.65 1.58 1.53 1.49
2.21 2.17 2.14 2.11 2.09 1.92 1.82 1.76 1.72 1.68 1.66 1.62 1.55 1.50 1.46
2.19 2.15 2.12 2.09 2.07 1.89 1.80 1.74 1.69 1.66 1.63 1.60 1.52 1.47 1.43
2.09 2.06 2.03 2.00 1.97 1.79 1.69 1.63 1.58 1.55 1.52 1.48 1.39 1.33 1.28
2.04 2.00 1.97 1.94 1.92 1.74 1.63 1.56 1.52 1.48 1.45 1.41 1.31 1.23 1.16
2.00 1.97 1.93 1.90 1.88 1.70 1.59 1.52 1.47 1.43 1.40 1.36 1.25 1.15 1.00
OIV-MA-AS1-07 : R2000
11
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Table 5 – Critical values for the Dixon test
Test criteria
Critical values
m
95%
99%
Z(2) – Z(1) ou Z(H) – Z (H – 1)
Z(H) – Z(1)
Z(H) – Z(1)
The greater of the two values
3
4
5
6
7
0,970
0,829
0,710
0,628
0,569
0,994
0,926
0,821
0,740
0,680
Q11 = Z(2) – Z(1) ou Z(H) – Z (H – 1)
Z(H – 1) – Z(1)
Z(H) – Z(2)
The greater of the two values
8
9
10
11
12
0,608
0,564
0,530
0,502
0,479
0,717
0,672
0,635
0,605
0,579
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
0,611
0,586
0,565
0,546
0,529
0,514
0,501
0,489
0,478
0,468
0,459
0,451
0,443
0,436
0,429
0,423
0,417
0,412
0,407
0,402
0,397
0,393
0,388
0,384
0,381
0,377
0,374
0,371
0,697
0,670
0,647
0,627
0,610
0,594
0,580
0,567
0,555
0,544
0,535
0,526
0,517
0,510
0,502
0,495
0,489
0,483
0,477
0,472
0,467
0,462
0,458
0,454
0,450
0,446
0,442
0,438
Q10 =
Q22 = Z(3) – Z(1) ou Z(H) – Z (H – 2)
Z(H – 2) – Z(1)
Z(H) – Z(3)
The greater of the two values
OIV-MA-AS1-07 : R2000
12
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS–OIV
Collaborative Study
Table 6 – Results of the collaborative study
Analysis
Lab
nº
Sample
Individual values x1
1
2
3
4
5
6
7
8
n1
x1
s1
s21
–
1
548 556 558 553 542
5
551 6,47
41,8
2
300 299 304 308 300
5
302 3,83
14,7 x1, < x
3
567 558 563 532* 560 560 563 567
7
563 3,51
12,3
4
557 550 555 560 551
5
555 4,16
17,3
5
569 575 565 560 572
5
568 5,89
34,7
8
563 14,92 222,6 s1 > s1
- =
6 550 546 549 557 588 570 576 568
7
557 560 560 552 547
5
555 5,63
31,7
8
548 543 560 551 548
5
550 6,28
39,5
9
558 563 551 555 560
5
556 5,63
31,7
10
554 559 551 545 557
5
553
30,2
5,5
Statistical Figures:
Bartlett Test:
Within laboratory: s1= ± 5.37 ƒ1 = 34
PB = 3.16 < 15.51 (95%; ƒ = 8)
Between laboratory: sz =  13.97 ƒz = 7
Analysis of variance:
sr =  5.37
PF = 6.76 > 3.21 (99%; ƒ1 = 7; ƒ2 = 34)
r = 15 sR =  7.78
OIV-MA-AS1-07 : R2000
R = 22
13
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
Reliability of analytical results
(Resolution Oeno 5/99)
Data concerning the reliability of analytical methods, as determined by
collaborative studies, are applicable in the following cases:
1) Verifying the results obtained by a laboratory with a reference method
2) Evaluating analytical results which indicate a legal limit has been exceeded
3) Comparing results obtained by two or more laboratories and comparing those
results with a reference value
4) Evaluating results obtained from a non-validated method
1) VERIFICATION OF THE ACCEPTABILITY
OBTAINED WITH A REFERENCE METHOD
The validity of analytical results depends on the following:
OF
RESULTS
 the laboratory should perform all analyses within the framework of an
appropriate quality control system which includes the organization,
responsibilities, procedures, etc.
 as part of the quality control system, the laboratory should operate according to
an internal Quality Control Procedure
 results should be obtained in accordance with the acceptability criteria
described in the internal Quality Control Procedure
Internal quality control shall be established in accordance with internationally
recognized standards, such those of the IUPAC document titled, "Harmonized
Guidelines for Internal Quality Control in Analytical Laboratories."
Internal Quality Control implies an analysis of the reference material.
Reference samples should consist of a template of the samples to be analyzed and
should contain an appropriate, known concentration of the substance analyzed
which is similar to that found in the sample.
To the extent possible, reference material shall be certified by an internationally
recognized organization.
However, for many types of analysis, there are no certified reference materials. In
this case, one could use, for example, material analyzed by several laboratories in a
competence test and considering the average of the results to be the value assigned
to the substance analyzed.
One could also prepare reference material by formulation (model solution with
known components) or by adding a known quantity of the substance analyzed to a
OIV-MA-AS1-08 : R1999
1
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
sample which does not contain (or not yet contain) the substance by means of a
recovery test (dosed addition) on one of the samples to analyze.
Quality Control is assured by adding reference material to each series of samples,
and analyzing these pairs (test samples and reference material). This verifies
correct implementation of the method and should be independent of the analytical
calibration and protocol as its goal is to verify the aforementioned.
Series means a number of samples analyzed under repeatable conditions. Internal
controls serve to ensure the appropriate level of uncertainty is not exceeded.
If the analytical results are considered to be part of a normal population whose
mean is m and standard deviation is s, only around 0.3% of the results will be
outside the limits m ± 3s. When aberrant results are obtained (outside these limits),
the system is considered to be outside statistical control (unreliable data).
The control is graphically represented using Shewhart Control Graphs. To produce
these graphical results, the measured values obtained from the reference material
are placed on the vertical axis while the series numbers are placed on the horizontal
axis. The graph also includes horizontal lines representing the mean, m, m ± 2
(warning limits) and m ± 3 (action limits) (Figure 1).
To estimate the standard deviation, a control should be analyzed, in pairs, in at
least 12 trials. Each analytical pair shall be analyzed under repeatable conditions
and randomly inserted in a sample series. Analyses will be duplicated on different
days to reflect reasonable changes from one series to another. Variations can have
several causes: modification of the reactants composition, instrument re-calibration
and even different operators. After eliminating aberrant data using the Grubbs test,
calculate the standard deviation to construct the Shewhart graphs. This standard
deviation is compared to that of the reference method. If a published precision
level is not obtained for the reference method, caused should be investigated.
The precision limits of the laboratory should be periodically revised by repeating
the indicated procedure.
Once the Quality Control graph is constructed, graph the results obtained from
each series for the control material.
A series is considered outside statistical control if:
I)
II)
a value is outside the action limit,
the current and previous values are situated outside the attention limits even in
within the action limits,
III) nine successive values lie on the same side of the mean.
The laboratory response to "outside control" conditions is to reject the results for
the series and perform tests to determine the cause, then take action to remedy the
situation.
OIV-MA-AS1-08 : R1999
2
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
A Shewhart Control Graph can also be produced for the differences between
analytical pairs in the same sample, especially when reference material does not
exist. In this case, the absolute difference between two analyses of the same
sample is graphed. The graph's lower line is 0 and the attention limit is 1.128Sw
while the action limit is 3.686Sw where Sw = the standard deviation of a series.
This type of graph only accounts for repeatability. It should be no greater than the
published repeatability limit for the method.
In the absence of control material, it sometimes becomes necessary to verify that
the reproducibility limit of the reference method is not exceeded by comparing the
results obtained to those of obtained by an experimental laboratory using the same
sample.
Each laboratory performs two tests and the following formula is used:
Cr D95 ( y1 - y2 )=
2
2
R -
r2
2
CrD95 =
Critical difference (P=0,95)
ȳ1
=
Means of 2 results obtained by lab 1
ȳ2
=
Means of 2 results obtained by lab 2
R
=
Reproducibility of reference method
r
=
Repeatability of reference method
If the critical difference has been exceeded, the underlying reason is to be found
and the test is to be repeated within one month.
2) EVALUATION OF ANALYTIC RESULTS INDICATING THAT A
LEGAL LIMIT HAS BEEN EXCEEDED.
When analytical results indicated that a legal limit has been exceeded, the
following procedure should be followed:
1) In the case of an individual result, conduct a second test under repeatable
conditions. If it is not possible to conduct a second test under repeatable
conditions, conduct a double analysis under repeatable conditions and use these
data to evaluate the critical difference.
2) Determine the absolute value of the difference between the mean of the results
obtained under repeatable conditions and the legal limit. An absolute value of
the difference which is greater than the critical distance indicates that the
sample does not fit the specifications.
Critical difference is calculated by the formula:
C r D95 ( y - m0 ) =
OIV-MA-AS1-08 : R1999
2
1 2 2 2 n -1
R -r
n
2
3
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
ȳ
=
Mean of results obtained
m0
=
Limit
=
Number of analyses
n
R
=
reproducibility
r
=
repeatability
In other words, this is a maximal limit where the average of the results obtained
should not be greater than:
m0 + CrD95(y - m0)
If the limit is a minimum, the average of the results obtained should not be less
than:
m0 - CrD95(y - m0)
3) COMPARING RESULTS OBTAINED USING TWO OR MORE
LABORATORIES AND COMPARING THESE RESULTS TO A
REFERENCE VALUE
To determine whether or not data originating in two laboratories are in agreement,
calculate the absolute difference between the two results and compare to the critical
difference:
1
1
)
2 n1 2 n2
Cr D95 ( y1 - y2 )= 2 R2 - r 2 (1ȳ1
y2
n1
n2
R
r
=
=
=
=
=
=
Mean of 2 results obtained by lab 1
Mean of 2 results obtained by lab 2
number of analyses in lab 1 sample
number of analyses in lab 2 sample
Reproducibility of reference method
Repeatability of reference method
If the result is the average of two tests, the equation can be simplified to:
Cr D95 ( y1 - y2 )= 2 R2 -
r2
2
If the data are individual results, the critical difference is R.
If the critical difference is not exceeded, the conclusion is that the results of the two
laboratories are in agreement.
Comparing results obtained by several laboratories with a reference value:
Suppose p laboratories have made n1 determinations, whose mean for each
laboratory is y1 and whose total mean is:
y=
OIV-MA-AS1-08 : R1999
1
 yi
p
4
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
The mean of all laboratories is compared with the reference value. If the absolute
difference exceeds the critical difference, as calculated using the following
formula, we conclude the results are not in agreement with the reference value:
Cr D95 ( y - m0 ) =
2
1
2p
2
R2 - r 2 (1 -
1 1
 )
p n1
CrD95 =
Critical difference, calculated as indicated in point 2, for the reference
method.
For example, the reference value can be the value assigned to a reference material
or the value obtained by the same laboratory or by a different laboratory with a
different method.
4) EVALUATING ANALYTICAL RESULTS OBTAINED USING NONVALIDATED METHODS
A provisional reproducibility value can be assigned to a non-validated method by
comparing it to that of a second laboratory:
R prov =
ȳ1 0
y2
r
=
=
=
2
2
( y1 - y 2 ) +
r2
2
Mean of 2 results obtained by lab 1
Mean of 2 results obtained by lab 2
Repeatability of reference method
Provisional reproducibility can be used to calculate critical difference.
If provisional reproducibility is less than twice the value of repeatability, it should
be set to 2r.
A reproducibility value greater than three times repeatability or twice the value
calculated using the Horwitz equation is not acceptable.
Horwitz equation:
RSDR % = 21-0,5log10 C
RSDR % = Standard deviation for reproducibility
(expressed as a percentage of the mean)
C
= concentration, expressed as a decimal fraction (for example,
10g/100g = 0.1)
OIV-MA-AS1-08 : R1999
5
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
This equation was empirically obtained from more than 3000 collaborative studies
including a diverse group of analyzed substances, matrices and measurement
techniques. In the absence of other information, RSDR values that are lower or
equal to the RSDR values calculated using the Horwitz equation can be considered
acceptable.
OIV-MA-AS1-08 : R1999
6
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
RSDR values calculated by the Horwitz equation:
Concentration
RSDR %
10-9
45
10-8
32
10-7
23
10-6
16
10-5
11
10-4
8
10-3
5,6
10-2
4
10-1
2,8
1
2
If the result obtained using a non-validated method is close to the limit specified by
legislation, the decision on the limit shall be decided as follows (for upper limits):
S = m0 + {(Rrout/Rref)-1}× CrD95
and, for lower limits,
S = m0 - {(Rrout/Rref)-1}× CrD95
S
m0
Rrout
Rref
CrD95
=
=
=
=
=
decision limit
legal limit
provisional reproducibility for non-validated method
reproducibility for reference method
critical difference, calculated as indicated in point 2, for the reference
method
The result which exceeds the decision limit should be replaced with a final result
obtained using the reference method.
Critical differences for probability levels other than 95%
OIV-MA-AS1-08 : R1999
7
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
This difference can be determined by multiplying the critical differences at the
95% level by the coefficients shown in Table 1.
Table 1 - Multiplicative coefficients allowing
the calculation of critical differences for
probability levels other than 95%
Probability level P
90
95
98
99
99,5
Multiplicative coefficient
0,82
1,00
1,16
1,29
1,40
SHEWHART CONTROL GRAPH
OIV-MA-AS1-08 : R1999
8
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
OIV-MA-AS1-08 : R1999
9
CONPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS– OIV
Reliability of Analytical Methods
BIBLIOGRAPHY
- "Harmonized Guidelines for Internal Quality Control in Analytical Chemistry
Laboratories". IUPAC. Pure and App. Chem. Vol 67, nº 4, 649-666, 1995
- "Shewhart Control Charts" ISO 8258. 1991.
- "Precision of test methods - Determination of repeatability and reproducibility for
a standard test method by inter-laboratory tests". ISO 5725, 1994.
- "Draft Commission Regulation of establishing rules for the application of
reference and routine methods for the analysis and quality evaluation of milk and
milk products". Commission of the European Communities, 1995.
- "Harmonized protocols for the adoption of standardized analytical methods and for
the presentation
of their performance characteristics". IUPAC. Pure an App. Chem., Vol. 62, nº 1,
149-162. 1990.
OIV-MA-AS1-08 : R1999
10
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
Protocol for the design, conducts and interpretation of
collaborative studies
(Resolution Oeno 6/2000)
INTRODUCTION
After a number of meetings and workshops, a group of representatives from
27 organizations adopted by consensus a "Protocol for the design, conducts and
interpretation of collaborative studies" which was published in Pure & Appl.
Chem. 60, 855-864, 1995. A number of organizations have accepted and used this
protocol. As a result of their experience and the recommendations of the Codex
Committee on Methods of Analysis and Sampling (Joint FAO/WHO Food
Standards Programme, Report of the Eighteenth Session, 9-13 November, 1992;
FAO, Rome Italy, ALINORM 93/23, Sections 34-39), three minor revisions were
recommended for incorporation into the original protocol. These are: (1) Delete the
double split level design because the interaction term it generates depends upon the
choice of levels and if it is statistically significant, the interaction cannot be
physically interpreted. (2) Amplify the definition of "material". (3) Change the
outlier removal criterion from 1% to 2.5%.
The revised protocol incorporating the changes is reproduced below. Some minor
editorial revisions to improve readability have also been made. The vocabulary and
definitions of the document 'Nomenclature of Interlaboratory Studies
(Recommendations 1994)' [published in Pure Appl Chem., 66, 1903-1911 (1994)]
has been incorporated into this revision, as well as utilizing, as far as possible, the
appropriate terms of the International Organization for Standardization (ISO),
modified to be applicable to analytical chemistry.
PROTOCOL
Preliminary work
1
Method-performance (collaborative) studies require considerable effort and should
be conducted only on methods that have received adequate prior testing. Such
within-laboratory testing should include, as applicable, information on the
following:
1.1
Preliminary estimates of precision
Estimates of the total within-laboratory standard deviation of the analytical results
over the concentration range of interest as a minimum at the upper and lower limits
OIV-MA-AS1-09 : R2000
1
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
of the concentration range, with particular emphasis on any standard or
specification value.
NOTE 1: The total within-laboratory standard deviation is a more inclusive
measure of imprecision that the ISO repeatability standard deviation, §3.3 below.
This standard deviation is the largest of the within-laboratory type precision
variables to be expected from the performance of a method; it includes at least
variability from different days and preferably from different calibration curves. It
includes between-run (between-batch) as well as within-run (within-batch)
variations. In this respect it can be considered as a measure of within-laboratory
reproducibility. Unless this value is well within acceptable limits, it cannot be
expected that the between-laboratory standard deviation (reproducibility standard
deviation) will be any better. This precision term is not estimated from the
minimum study described in this protocol.
NOTE 2: The total within-laboratory standard deviation may also be estimated
from ruggedness trials that indicate how tightly controlled the experimental factors
must be and what their permissible ranges are. These experimentally determined
ranges should be incorporated into the description of the method.
Systematic error (bias)
1.2
Estimates of the systematic error of the analytical results over the concentration
range and in the substances of interest, as a minimum at the upper and lower limits
of the concentration range, with particular emphasis on any standard or
specification value.
The results obtained by applying the method to relevant reference materials should
be noted.
1.3
Recoveries
The recoveries of "spikes" added to real materials and to extracts, digests, or other
treated solutions thereof.
1.4
Applicability
The ability of the method to identify and measure the physical and chemical forms
of the analyte likely to be present in the materials, with due regard to matrix
effects.
1.5
Interference
The effect of other constituents that are likely to be present at appreciable
concentrations in matrices of interest and which may interfere in the determination.
OIV-MA-AS1-09 : R2000
2
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
1.6
Method comparison
The results of comparison of the application of the method with existing tested
methods intended for similar purposes.
1.7
Calibration Procedures
The procedures specified for calibration and for blank correction must not
introduce important bias into the results.
Method description
1.8
The method must be clearly and unambiguously written.
1.9
Significant figures
The initiating laboratory should indicate the number of significant figures to be
reported, based on the output of the measuring instrument.
NOTE: In making statistical calculations from the reported data, the full power of
the calculator or computer is to be used with no rounding or truncating until the
final reported mean and standard deviations are achieved. At this point the standard
deviations are rounded to 2 significant figures and the means and related standard
deviations are rounded to accommodate the significant figures of the standard
deviation. For example, if SR = 0.012, c is reported as 0.147, not as 0. 1473 or 0.
15, and RSDR is reported as 8.2%. (Symbols are defined in Appendix L) If
standard deviation calculations must be conducted manually in steps, with the
transfer of intermediate results, the number of significant figures to be retained for
squared numbers should be at least 2 times the number of figures in the data plus 1.
2.
Design of the method-performance study
2.1
Number of materials
For a single type of substance, at least 5 materials (test samples) must be used; only
when a single level specification is involved for a single matrix may this minimum
required number of materials to be reduced to 3. For this design parameter, the two
portions of a split level and the two individual portions of blind replicates per
laboratory are considered as a single material.
NOTE 1: A material is an 'analyte/matrix/concentration' combination to which the
method-performance parameters apply. This parameter determines the applicability
of a method. For application to a number of different substances, a sufficient
OIV-MA-AS1-09 : R2000
3
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
number of matrices and levels should be chosen to include potential interferences
and the concentration of typical use.
NOTE 2: The 2 or more test samples of blind or open replicates statistically, are a
single material (they are not independent).
NOTE 3: A single split level (Youden pair) statistically analyzed as a pair is a
single material; if analyzed statistically and reported as single test samples, they are
2 materials. In addition, the pair can be used to calculate the within-laboratory
standard deviation, sr as
2
sr =  (di ) / 2n)
(for duplicates, blind or open),
sr =  ((di ) / 2(n - 1)
2
(for Youden pairs),
where di, the difference between the 2 individual values from the split level for
each laboratory and n is the number of laboratories. In this special case, S R, the
among laboratories standard deviation, is merely the average of the two SR values
calculated from the individual components of the split level, and it is used only as a
check of the calculations.
NOTE 4: The blank or negative control may be a material or not depending on the
usual purpose of the analysis. For example, in trace analysis, where very low levels
(near the limit of quantitation) are often sought, the blanks are considered as
materials and are necessary to determine certain 'limits of measurement.' However,
if the blank is merely a procedural control in macro analysis (e.g., fat in cheese), it
would not be considered a material.
2.2
Number of laboratories
At least 8 laboratories must report results for each material; only when it is
impossible to obtain this number (e.g., very expensive instrumentation or
specialized laboratories required) may the study be conducted with less, but with
an absolute minimum of 5 laboratories. If the study is intended for international
use, laboratories from different countries should participate. In the case of methods
requiring the use of specialized instruments, the study might include the entire
population of available laboratories. In such cases, "n" is used in the denominator
for calculating the standard deviation instead of "(n - 1)". Subsequent entrants to
the field should demonstrate the ability to perform as well as the original
participant.
2.3
Number of Replicates
The repeatability precision parameters must be estimated by using one of the
following sets of designs (listed in approximate order of desirability):
2.3.1
Split Level
OIV-MA-AS1-09 : R2000
4
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
For each level that is split and which constitutes only a single material for purposes
of design and statistical analysis, use 2 nearly identical test samples that differ only
slightly in analyte concentration (e.g., <1-5%). Each laboratory must analyse each
test sample once and only once.
NOTE: The statistical criterion that must be met for a pair of test samples to
constitute a split level is that the reproducibility standard deviation of the two parts
of the single split level must be equal.
2.3.2 Combination blind replicates and split level
Use split levels for some materials and blind replicates for other materials in the
same study (single values from each submitted test sample).
2.3.3 Blind replicates
For each material, use blind identical replicates, when data censoring is impossible
(e.g., automatic input, calculation, and printout) non-blind identical replicates may
be used.
2.3.4 Known replicates
For each material, use known replicates (2 or more analyses of test portions from
the same test sample), but only when it is not practical to use one of the preceding
designs.
2.3.5 Independent analyses
Use only a single test portion from each material (i.e., do not perform multiple
analyses) in the study, but rectify the inability to calculate repeatability parameters
by quality control parameters or other within-laboratory data obtained
independently of the method-performance study.
3.
Statistical analysis (See Flowchart, A.4. 1)
For the statistical analysis of the data, the required statistical procedures listed
below must be performed and the results reported. Supplemental, additional
procedures are not precluded.
3.1
Valid data
Only valid data should be reported and subjected to statistical treatment. Valid data
are those data that would be reported as resulting from the normal performance of
laboratory analyses; they are not marred by method deviations, instrument
malfunctions, unexpected occurrences during performance, or by clerical,
typographical and arithmetical errors.
3.2
One-way analysis of variance
One-way analysis of variance and outlier treatments must be applied separately to
each material (test sample) to estimate the components of variance and
repeatability and reproducibility parameters.
OIV-MA-AS1-09 : R2000
5
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
3.3
Initial estimation
Calculate the mean, c (= the average of laboratory averages), repeatability relative
standard deviation, RSDr, and reproducibility relative standard deviation, RSDR
with no outliers removed, but using only valid data.
3.4
Outlier treatment
The estimated precision parameters that must also be reported are based on the
initial valid data purged of all outliers flagged by the harmonized 1994 outlier
removal procedure. This procedure essentially consists of sequential application of
the Cochran and Grubbs tests (at 2.5% probability (P) level, 1-tail for Cochran and
2-tail for Grubbs) until no further outliers are flagged or until a drop of 22.2% (=
219) in the original number of laboratories providing valid data would occur.
NOTE: Prompt consultation with a laboratory reporting suspect values may result
in correction of mistakes or discovering conditions that lead to invalid data, 3.1.
Recognizing mistakes and invalid data per se is much preferred to relying upon
statistical tests to remove deviate values.
3.4.1 Cochran test
First apply Cochran outlier test (1-tail test a P = 2.5%) and remove any laboratory
whose critical value exceeds the tabular value given in the tale, Appendix A.3. 1,
for the number of laboratories and replicates involved.
3.4.2 Grubbs tests
Apply the single value Grubbs test (2 tail) and remove any outlying laboratory. If
no laboratory is flagged, then apply the pair value tests (2 tail) - - 2 at the same end
and 1 value at each end, P = 2.5% overall. Remove any laboratory(ies) flagged by
these tests whose critical value exceeds the tabular value given in the appropriate
column of the table Appendix A.3.3. Stop removal when the next application of the
test will flag as table, A outliers more that 22.2% (2 of 9) of the laboratories.
NOTE: The Grubbs tests are to be applied one material at a time to the set of
replicate means from all laboratories, and not to the individual values from
replicated designs because the distribution of all the values taken together is
multimodal, not Caussian, i.e., their differences from the overall mean for that
material are not independent.
3.4.3 Final estimation
Recalculate the parameters as in §3.3 after the laboratories flagged by the
preceding procedure have been removed. If no outliers were removed by the
Cochran-Grubbs sequence, terminate testing. Otherwise, reapply the CochranGrubbs sequence to the data purged of the flagged outliers until no further outliers
are flagged or until more than a total of 22.2% (2 of 9 laboratories) would be
removed in the next cycle. See flowchart A.3.4.
OIV-MA-AS1-09 : R2000
6
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
4.
Final report
The final report should be published and should include all valid data. Other
information and parameters should be reported in a format similar (with respect to
the reported items) to the following, as applicable:
[x] Method-performance tests carried out at the international level in [year(s)] by
[organisation] in which [y and z] laboratories participated, each performing [k]
replicates, gave the following statistical results:
TABLE OF METHOD-PERFORMANCE PARAMETERS
Analyte; Results expressed in [units]
Material [Description and listed in columns across top of table in increasing order
of magnitude of means]
Number of laboratories retained after eliminating outliers
Number of outlying laboratories
Code (or designation) of outlying laboratories
Number of accepted results
Mean
True or accepted value, if known
Repeatability standard deviation (Sr)
Repeatability relative standard deviation (RSDR)
Repeatability limit, r (2.8 x Sr)
Reproducibility standard deviation (SR)
Reproducibility relative standard deviation (RSDR)
Reproducibility limit, R (2.8 X SR)
4.1
Symbols
A set of symbols for use in reports and publications is attached as Appendix 1
(A.1.).
4.2
Definitions
A set of definitions for use in study reports and publications is attached as
Appendix 2 (A.2.).
4.3
Miscellaneous
4.3.1 Recovery
Recovery of added analyte as a control on method or laboratory bias should be
calculated as follows:
[Marginal] Recovery, %=
(Total analyte found - analyte originally present) x 100/(analyte added)
OIV-MA-AS1-09 : R2000
7
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
Although the analyte may be expressed as either concentration or amount, the units
must be the same throughout. When the quantity of analyte is determined by
analysis, it must be determined in the same way throughout.
Analytical results should be reported uncorrected for recovery. Report recoveries
separately.
4.3.2 When Sr, is negative
By definition, SR is greater than or equal to Sr in method-performance
studies; occasionally the estimate of Sr is greater than the estimate of SR (the
average of the replicates is greater than the range of laboratory averages and the
calculated SL2 is then negative). When this occurs, set SL = 0 and SR = Sr.
5.
REFERENCES
Horwitz, W. (1988) Protocol for the design, conduct, and interpretation of method
performance studies. Pure & Appl. Chem. 60, 855-864.
Pocklington, W.D. (1990) Harmonized protocol for the adoption of standardized
analytical methods and for the presentation of their performance characteristics.
Pure and Appl. Chem. 62, 149-162.
International Organization for Standardization. International Standard 5725-1986.
Under revision in 6 parts; individual parts may be available from National
Standards member bodies.
OIV-MA-AS1-09 : R2000
8
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
A. APPENDICES
APPENDIX 1. - SYMBOLS
Use the following set of symbols and terms for designating parameters developed
by a method-performance study.
Mean (of laboratory averages)
Standard deviations:
Repeatability
'Pure' between-laboratory
Reproducibility
x
s (estimates)
Sr
SL
SR
Variances:
SR2 = SL 2 + Sr 2
S2 (with subscripts, r, L, and R)
Relative standard deviations:
Maximum tolerable differences
(as defined by ISO 5725-1986);
See A.2.4 and A.2.5)
Repeatability limit
Reproducibility limit
RSD (with subscripts, r, L, and r)
r = (2.8 x Sr)
R = (2.8 X SR)
Number of replicates per laboratory
Average number of replicates per laboratory i
k (general)
k (for a balanced design)
Number of laboratories
L
Number of materials (test samples)
m
Total number of values in a given assay n (= kL for a balanced design)
Total number of values in a given study N (= kLm for an overall balanced design)
____________________
If other symbols are used, their relationship to the recommended symbols should
be explained fully.
OIV-MA-AS1-09 : R2000
9
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
APPENDIX 2. - DEFINITIONS
Use the following definitions. The first three definitions utilize the 1UPAC
document "Nomenclature of Interlaboratory Studies" (approved for publication
1994). The next two definitions are assembled from components given in ISO
3534-1:1993. All test results are assumed to be independent, i.e., 'obtained in a
manner not influenced by any previous result on the same or similar test object.
Quantitative measures of precision depend critically on the stipulated conditions.
Repeatability and reproducibility conditions are particular sets of extreme
stipulated conditions.'
A.2.1 Method-performance studies
An interlaboratory study in which all laboratories follow the same written protocol
and use the same test method to measure a quantity in sets of identical test items
[test samples, materials]. The reported results are used to estimate the performance
characteristics of the method. Usually these characteristics are within-laboratory
and among-laboratories precision, and when necessary and possible, other pertinent
characteristics such as systematic error, recovery, internal quality control
parameters, sensitivity, limit of determination, and applicability.
A.2.2. Laboratory-performance study
An interlaboratory study that consists of one or more analyses or measurements by
a group of laboratories on one or more homogeneous, stable test items, by the
method selected or used by each laboratory. The reported results are compared
with those of other laboratories or with the known or assigned reference value,
usually with the objective of evaluating or improving laboratory performance.
A.2.3 Material certification stud
An interlaboratory study that assigns a reference value ('true value') to a quantity
(concentration or property) in the test item, usually with a stated uncertainty.
A.2.4 Repeatability limit (r)
When the mean of the values obtained from two single determinations with the
same method on identical test items in the same laboratory by the same operator
using the same equipment within short intervals of time, lies within the range of the
mean values cited in the Final Report, 4.0, the absolute difference between the two
test results obtained should be less than or equal to the repeatability limit (r) [= 2.8
x s,) that can generally be inferred by linear interpolation of sr from the Report.
NOTE: This definition, and the corresponding definition for reproducibility limit,
has been assembled from five cascading terms and expanded to permit application
by interpolation to a test item whose mean is not the same as that used to establish
the original parameters, which is the usual case in applying these definitions. The
term 'repeatability [and reproducibility] limit' is applied specifically to a probability
OIV-MA-AS1-09 : R2000
10
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
of 95% and is taken as 2.8 x s, [or SRI. The general term for this statistical concept
applied to any measure of location (e.g., median) and with other probabilities (e.g.,
99%) is "repeatability [and reproducibility] critical difference".
A.2.5 Reproducibility limit (R)
When the mean of the values obtained from two single determinations with the
same method on identical test items in different laboratories with different
operators using different equipment, lies within the range of the mean values cited
in the Final Report, 4.0, the absolute difference between the two test results
obtained should be less than or equal to the reproducibility limit (R) [= 2.8 x sR]
that can generally be inferred by linear interpolation of SR from the Report.
NOTE 1: When the results of the interlaboratory test make it possible, the value of
r and R can be indicated as a relative value (e.g., as a percentage of the determined
mean value) as an alternative to the absolute value.
NOTE 2: When the final reported result in the study is an average derived from
more than a single value, i.e., k is greater than 1, the value for R must be adjusted
according to the following formula before using R to compare the results of a
single routine analyses between two laboratories.
R' = (R2 + r2 (1 - [l /k])1/2
Similar adjustments must be made for replicate results constituting the final values
for SR and RSDR, if these will be the reported parameters used for quality control
purposes.
NOTE 3: The repeatability limit, r, may be interpreted as the amount within which
two determinations should agree with each other within a laboratory 95% of the
time. The reproducibility limit, R, may be interpreted as the amount within which
two separate determinations conducted in different laboratories should agree with
each other 95% of the time.
NOTE 4: Estimates Of SR can be obtained only from a planned, organized method
performance study; estimates of Sr can be obtained from routine work within a
laboratory by use of control charts. For occasional analyses, in the absence of
control charts, within-laboratory precision may be approximated as one half SR
(Pure and Appl. Chem., 62, 149-162 (1990) , Sec. L3, Note.).
A.2.6 One-way analysis of variance
One-way analysis of variance is the statistical procedure for obtaining the estimates
of within laboratory and between-laboratory variability on a material-by-material
basis. Examples of the calculations for the single level and single-split-level
designs can be found in ISO 5725-1986.
OIV-MA-AS1-09 : R2000
11
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
OIV-MA-AS1-09 : R2000
12
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
APPENDIX 3. - CRITICAL VALUES
A.3.1 Critical values for the Cochran maximum variance ratio at the 2.5% (1 -tail)
rejection level, expressed as the percentage the highest variance is of the total
variance; r = number of replicates.
No. of Labs
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
35
40
50
r=2
r=3
r=4
r=5
r=6
94.3
88.6
83.2
78.2
73.6
69.3
65.5
62.2
59.2
56.4
53.8
51.5
49.5
47.8
46.0
44.3
42.8
41.5
40.3
39.1
37.9
36.7
35.5
34.5
33.7
33.1
32.5
29.3
26.0
21.6
81.0
72.6
65.8
60.2
55.6
51.8
48.6
45.8
43.1
40.5
38.3
36.4
34.7
33.2
31.8
30.5
29.3
28.2
27.2
26.3
25.5
24.8
24.1
23.4
22.7
22.1
21.6
19.5
17.0
14.3
72.5
64.6
58.3
52.2
47.4
43.3
39.9
37.2
35.0
33.2
31.5
29.9
28.4
27.1
25.9
24.8
23.8
22.9
22.0
21.2
20.5
19.9
19.3
18.7
18.1
17.5
16.9
15.3
13.5
11.4
65.4
58.1
52.2
47.3
43.0
39.3
36.2
33.6
31.3
29.2
27.3
25.7
24.4
23.3
22.4
21.5
20.7
19.9
19.2
18.5
17.8
17.2
16.6
16.1
15.7
15.3
14.9
12.9
11.6
9.7
62.5
53.9
47.3
42.3
38.5
35.3
32.6
30.3
28.3
26.5
25.0
23.7
22.0
21.2
20.4
19.5
18.7
18.0
17.3
16.6
16.0
15.5
15.0
14.5
14.1
13.7
13.3
11.6
10.2
8.6
Tables A.3.1 and A.3.3 were calculated by R. Albert (October, 1993) by computer
simulation involving several runs of approximately 7000 cycles each for each
value, and then smoothed. Although Table A.3.1 is strictly applicable only to a
balanced design (same number of replicates from all laboratories), it can be applied
to an unbalanced design without too much error, if there are only a few deviations.
A.3.2 Calculation of Cochran maximum variance outlier ratio
OIV-MA-AS1-09 : R2000
13
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
Compute the within-laboratory variance for each laboratory and divide the largest
of these variances by the sum of the all of the variances and multiply by 100. The
resulting quotient is the Cochran statistic which indicates the presence of a
removable outlier if this quotient exceed the critical value listed above in the
Cochran table for the number of replicates and laboratories specified.
A.3.3 Critical values for the Grubbs extreme deviation outlier tests at the 2.5% (2tail), 1.25% (1tail) rejection level, expressed as the percent reduction in standard
deviations caused by the removal of the suspect value(s).
No. of labs
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
40
50
One highest
or lowest
86.1
73.5
64.0
57.0
51.4
46.8
42.8
39.3
36.3
33.8
31.7
29.9
28.3
26.9
25.7
24.6
23.6
22.7
21.9
21.2
20.5
19.8
19.1
18.4
17.8
17.4
17.1
13.3
11.1
OIV-MA-AS1-09 : R2000
Two highest
or two lowest
98.9
90.9
81.3
73.1
66.5
61.0
56.4
52.5
49.1
46.1
43.5
41.2
39.2
37.4
35.9
34.5
33.2
31.9
30.7
29.7
28.8
28.0
27.1
26.2
25.4
24.7
24.1
19.1
16.2
One highest and
one lowest
99.1
92.7
84.0
76.2
69.6
64.1
59.5
55.5
52.1
49.1
46.5
44.1
42.0
40.1
38.4
36.9
35.4
34.0
32.8
31.8
30.8
29.8
28.9
28.1
27.3
26.6
26.0
20.5
17.3
14
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
A.3.4 Calculation of the Grubbs test values
To calculate the single Chubbs test statistic, compute the average for each
laboratory and then calculate the standard deviation (M) of these L averages
(designate as the original s). Calculate the SD of the set of averages with the
highest average removed (SH); calculate the SD of the set of averages with the
lowest average removed (SL). The calculate the percentage decrease in SD for both
as follows:
100 x [ 1 - (sL/s] and 100 x [ 1 - (sH/s)].
The higher of these two percentage decreases is the singe Grubbs test statistic,
which signal the presence of an outlier to be omitted at the P = 2.5% level, 2tail, if
it exceeds the critical value listed in the single value column, Column 2, of Table
A.3.3 , for the number of laboratory averages used to calculate the original s.
To calculate the paired Grubbs test statistics, calculate the percentage decrease in
standard deviation obtained by dropping the two highest averages and also by
dropping the two lowest averages, as above. Compare the higher of the percentage
changes in standard deviation with the tabular values in column 3 and proceed with
(1) or (2): (1) If the tabular value is exceeded, remove the responsible pair. Repeat
the cycle again, starting at the beginning with the Cochran extreme variance test
again, the Grubbs extreme value test, and the paired Grubbs extreme value test. (2)
If no further values are removed, then calculate the percentage change in standard
deviation obtained by dropping both the highest extreme value and the lowest
extreme value together, and compare with the tabular values in the last column of
A.3.3. If the tabular value is exceeded, remove the high-low pair of averages, and
start the cycle again with the Cochran test until no further values are removed. In
all cases, stop outlier testing when more than 22.2% (2/9) of the averages are
removed.
OIV-MA-AS1-09 : R2000
15
COMMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS – OIV
Protocol for collaborative studies
APPENDIX 4
A.4.1. Flowchart for outlier removal
OIV-MA-AS1-09 : R2000
16
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Estimation of Detection and Quantification Limit
Estimation of the detection and quantification limits of a
method of analysis
(Resolution Oeno 7/2000)
1 - Purpose: to establish the detection and quantification limits of a method
N.B. : The proposed calculation procedure sets « detection and quantification
limiting » values with respect to the instrumental response. For a given
method, the final calculation of these values must take cognizance of factors
arising from the preparation of the sample.
2 - Definitions
 Detection limit: the smallest concentration or proportion of the analyzed
substance that can be detected with an acceptable level of uncertainty, but that is
not quantified under the experimental conditions described in the method
 Quantification limit: the smallest concentration or proportion of the analyzed
substance that can be quantified with an acceptable level of uncertainty, under
the experimental conditions described in the method.
3 – Logic Diagram for Decision-Making
Estimate an LD
or LQ
No
Possible
blank
No
Yes
Recording
Stray
peak
Yes
No
Yes
No
RT of
Stable stray
peak
Yes
Straight
regression
line
DL = 3σb
LQ= 10σb
Results approach
OIV-MA-AS1-10 : R2000
Hmax
method
Haverage
method
Hmax or
haverage
method
Graph approach
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Estimation of Detection and Quantification Limit
4 - Methodology
4.1 "Results" approach
When the analytical method produces no recorded graph, but only numerical values
(i.e., colorimetry), the detection limit (LD) and the quantification limit (LQ) are
estimated using one of the two following methods.
4.1.1 - Method 1:
Directly read n measurements (analyte quantity or response) of separate analytic
« blank » samples that contain all of the constituents, with the exception of the
substance to be tested for.
LD = mblank + 3Sblank and
LQ = mblank + 10Sblank
where mblank and Sblank are the mean and standard deviation for n measurements.
Note: A multiplication factor of 3 corresponds to a 0.13% chance of concluding
that the substance sought is present, when, in fact, it is lacking. 10 corresponds to
a 0.5% chance.
4.1.2 - Method 2:
Using the straight calibration line: Y = a + bX
The detection limit is the smallest concentration of a substance that can be
distinguished from the blank, with a 0.13% risk of retaining samples containing
nothing ; in other words, the value beginning at which a statistical test comparing
the response to 0 becomes significant with an error level α of 0.13%. Hence:
YDL = a +3Sa
XDL = (a +3Sa)/b
Where Sa is the standard deviation on the ordinate at the origin of the straight
regression line. The logic is the same for LQ,, where the multiplication factor is 10
(risk of 0.5%).
4.2 - "Graph" Approach
For analytical methods which generate graphs (i.e., chromatography), the detection
limit is estimated based on the ground noise of the analytic blank recording for a
given sample.
LD = 3 x h x R (associated risk is below 0.13%) and
LQ = 10 x h x R (associated risk is below 0.5%), where
 h is the average or maximum amplitude of the signal window corresponding
to 10 width s of the mid-height peak on either side of the retention time, as a
function of stability.
OIV-MA-AS1-10 : R2000
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Estimation of Detection and Quantification Limit
 R is the quantity/signal response factor expressed as a function of the
quantity of substance/height.
On each occasion, three series of three injections each are performed on test
blanks at an interval of several days.
4.2.1 hmax method
- Increase ground noise to the maximum (Fig. 1 above) ;
- center around the retention time (RT) of the product ;
- draw a window of 10 widths of the mid-height peak (W1/2) on either side of the
RT ;
- draw two parallel lines, one running through the highest point of the highest
peak, the other through the base of the deepest trough ;
- evaluate height -> hmax ;
- calculate the response factor (R factor) ;
- LDmax = 3 x hmax x R
- LQmax = 10 x hmax x R
4.2.2 haverage Method
OIV-MA-AS1-10 : R2000
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Estimation of Detection and Quantification Limit
- increase the ground noise to the maximum (Fig. 2 above) ;
- center around the retention time (RT) of the product ;
- draw a window of 10 widths of the mid-height peck (W1/2) on either side of
the RT ;
- divide into 20 equal sections (x) ;
- draw two parallel lines in each block, one running through the highest point of
the highest peak, the other through the base of the deepest trough ;
- measure the heights, y ;
- calculate the average (y = haverage);
- calculate the response factor (R factor);
- LDaverage = 3 x haverage x R ;
- LQaverage= 10 x haverage x R
These estimates can themselves be validated by injecting quantities of solute that
are close to the calculated limits (Figures 3 and 4).
OIV-MA-AS1-10 : R2000
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Estimation of Detection and Quantification Limit
Compound at [c] # hmax
Figure No. 3: Validating calculations of limits.
Concentration of the compound approaches Haverage
N.B. : The dotted line corresponds to the real injected value however, since this
figure is provided as an example, it may be deleted from the final text.
OIV-MA-AS1-10 : R2000
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Estimation of Detection and Quantification Limit
Compound at haverage < [c] < ≈hmax
Figure No. 4: Validating calculations of limits.
Concentration of compound between Haverage and Hmax
N.B. : The dotted line corresponds to the real injected value; however, since this
figure is provided as an example, it may be deleted from the final text.
OIV-MA-AS1-10 : R2000
6
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
Harmonized guidelines for internal quality control
in analytical chemistry laboratories
(Resolution Oeno 19/2002)
CONTENTS
1. INTRODUCTION
1.1 Basic concepts
1.2. Scope of this document
1.3. Internal quality control and uncertainties
2. DEFINITIONS
2.1. International definitions
2.2. Definition of terms specific to this document
3. QUALITY ASSURANCE PRACTICES AND INTERNAL QUALITY
CONTROL
3.1. Quality assurance
3.2. Choice of analytical method
3.3. Quality control and aptitude tests
4. INTERNAL QUALUITY CONTROL PROCEDURES
4.1. Introduction
4.2. General approach. Statistical control
4.3. Internal quality control and fitness for purpose
4.4. The nature of errors
5. IQC AND WITHIN-RUN PRECISION
5.1. Precision and duplication
5.2. Interpretation of duplicate data
6. CONTROL MATERIALS IN IQC
6.1. Introduction
6.2. The role of certified reference materials
6.3. Preparation of control material
OIV-MA-AS1-11 : R2002
1
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
6.4. Blank determinations
6.5. Traceability in spiking and recovery checks
7. RECOMMENDATIONS
8. CONCLUSIONS
9. REFERENCES
APPENDIX 1. SHEWHART CONTROL CHARTS
OIV-MA-AS1-11 : R2002
2
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
1. INTRODUCTION
1.1 Basic concept
This document sets out guidelines for the implementation of internal
quality control (IQC) in analytical laboratories. IQC is one of a number of
concerted measures that analytical chemists can take to ensure that the data
produced in the laboratory are fit for their intended purpose. In practice,
fitness for purpose is determined by a comparison of the accuracy achieved
in a laboratory at a given time with a required level of accuracy. Internal
quality control therefore comprises the routine practical procedures that
enable the analytical chemist to accept a result or group of results as fit for
purpose, or reject the results and repeat the analysis. As such, IQC is an
important determinant of the quality of analytical data, and is recognised as
such by accreditation agencies.
Internal quality control is undertaken by the inclusion of particular
reference materials, here called "control materials", into the analytical
sequence and by duplicate analysis. The control materials should,
wherever possible, be representative of the test materials under
consideration in respect of matrix composition, the state of physical
preparation and the concentration range of the analyte. As the control
materials are treated in exactly the same way as the test materials, they are
regarded as surrogates that can be used to characterise the performance of
the analytical system, both at a specific time and over longer intervals.
Internal quality control is a final check of the correct execution of all of the
procedures (including calibration) that are prescribed in the analytical
protocol and all of the other quality assurance measures that underlie good
analytical practice. IQC is therefore necessarily retrospective. It is also
required to be as far as possible independent of the analytical protocol,
especially the calibration, that it is designed to test.
Ideally both the control materials and those used to create the calibration
should be traceable to appropriate certified reference materials or a
recognised empirical reference method. When this is not possible, control
materials should be traceable at least to a material of guaranteed purity or
other well characterised material. However, the two paths of traceability
must not become coincident at too late a stage in the analytical process.
For instance, if control materials and calibration standards were prepared
from a single stock solution of analyte, IQC would not detect any
inaccuracy stemming from the incorrect preparation of the stock solution.
OIV-MA-AS1-11 : R2002
3
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
In a typical analytical situation several, or perhaps many, similar test
materials will be analysed together, and control materials will be included
in the group. Often determinations will be duplicated by the analysis of
separate test portions of the same material. Such a group of materials is
referred to in this document as an analytical "run". (The words "set",
"series" and "batch" have also been used as synonyms for "run".) Runs are
regarded as being analysed under effectively constant conditions. The
batches of reagents, the instrument settings, the analyst, and the laboratory
environment will, under ideal conditions, remain unchanged during
analysis of a run. Systematic errors should therefore remain constant
during a run, as should the values of the parameters that describe random
errors. .As the monitoring of these errors is of concern, the run is the basic
operational unit of IQC.
A run is therefore regarded as being carried out under repeatability
conditions, i.e., the random measurement errors are of a magnitude that
would be encountered in a "short" period of time. In practice the analysis
of a run may occupy sufficient time for small systematic changes to occur.
For example, reagents may degrade, instruments may drift, minor
adjustments to instrumental settings may be called for, or the laboratory
temperature may rise. However, these systematic effects are, for the
purposes of IQC, subsumed into the repeatability variations. Sorting the
materials making up a run into a randomised order converts the effects of
drift into random errors.
1.2 Scope of this document
This document is a harmonisation of IQC procedures that have evolved in
various fields of analysis, notably clinical biochemistry, geochemistry and
environmental studies, occupational hygiene and food analysis(3-9). There
is much common ground in the procedures from these various fields.
Analytical chemistry comprises an even wider range of activities and the
basic principles of IQC should be able to encompass all of these. The
present document provides guidelines that will be applicable in most
instances. This policy necessarily excludes a number of IQC practices that
are restricted to individual sectors of the analytical community. In addition
in some sectors it is common to combine IQC as defined here with other
aspects of quality assurance practice. There is no harm in such
combination, but it must remain clear what are the essential aspects of IQC.
OIV-MA-AS1-11 : R2002
4
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
In order to achieve a harmonisation and provide basic guidance on IQC,
some types of analytical activity have been excluded from this document.
Issues specifically excluded are as follows.
(i) Quality control of sampling. While it is recognised that the quality of
the analytical result can be no better than that of the sample, quality control
of sampling is a separate subject and in many areas is not fully developed.
Moreover, in many instances analytical laboratories have no control over
sampling practice and quality.
(ii) In-line analysis and continuous monitoring. In this style of analysis
there is no possibly of repeating the measurement, so the concept of IQC as
used in this document is inapplicable.
(iii) Multivariate IQC. Multivariate methods in IQC are still the subject of
research and cannot be regarded as sufficiently established for inclusion
here. The current document regards multianalyte data as requiring a series
of univariante IQC tests. Caution is necessary in the interpretation of this
type of data to avoid inappropriately frequent rejection of data.
(iv) Statutory and contractual requirements.
(v) Quality assurance measures such as checks on instrumental stability
before and during analysis, wavelength calibration, balance calibration,
tests on resolution of chromatography columns, and problem diagnostics
are not included. For present purposes they are regarded as part of the
analytical protocol, and IQC tests their effectiveness together with the
other aspects of the methodology.
1.3 Internal quality control and uncertainty
A prerequisite of analytical chemistry is the recognition of "fitness for
purpose", the standard of accuracy that is required for an effective use of
the analytical data. This standard is arrived at by consideration of the
intended uses of the data although it is seldom possible to foresee all of the
potential future applications of analytical results. For this reason in order
to prevent inappropriate interpretation, it is important that a statement of
the uncertainty should accompany analytical results, or be readily available
to those who wish to use the data.
OIV-MA-AS1-11 : R2002
5
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
Strictly speaking, an analytical result cannot be interpreted unless it is
accompanied by knowledge of its associated uncertainty at a stated level of
confidence. A simple example demonstrates this principle. Suppose that
there is a statutory requirement that a foodstuff must not contain more than
10 g g-1 of a particular constituent. A manufacturer analyses a batch and
obtains a result of 9 g g-1 for that constituent. If the uncertainty of the
result expressed as a half range (assuming no sampling error) is
0.1 g g-1 (i.e. the true result falls, with a high probability, within the
range 8.9-9.1) then it may be assumed that the legal limit is not exceeded.
If, in contrast, the uncertainty is 2 g g-1 then there is no such assurance.
The interpretation and use that may be made of the measurement thus
depends on the uncertainty associated with it.
Analytical results should therefore have an associated uncertainty if any
definite meaning is to be attached to them or an informed interpretation
made. If this requirement cannot be fulfilled, the use to which the data can
be put is limited. Moreover, the achievement of the required measurement
uncertainty must be tested as a routine procedure, because the quality of
data can vary, both in time within a single laboratory and between different
laboratories. IQC comprises the process of checking that the required
uncertainty is achieved in a run.
2. DEFINITIONS
2.1 International definitions
Quality assurance. All those planned and systematic actions necessary to
provide adequate confidence that a product or service will satisfy given
requirements for quality(10).
Trueness: closeness of the agreement between the average value obtained
from a large series of test results and an accepted reference value(11).
Precision: closeness of agreement between independent test results
obtained under prescribed conditions(12).
Bias: difference between the expectation of the test results and an accepted
reference value(11).
OIV-MA-AS1-11 : R2002
6
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
Accuracy: closeness of the agreement between the result of a measurement
and a true value of the measurand(13).
Note 1. Accuracy is a qualitative concept.
Note 2. The term precision should not be used for accuracy.
Error: result of a measurement minus a true value of the measurand(13).
Repeatability conditions. conditions where independent test results are
obtained with the same method on identical test items in the same
laboratory by the same operator using the same equipment within short
intervals of time(11).
Uncertainty of measurement: parameter, associated with the result of a
measurement, that characterises the dispersion of the values that could
reasonably be attributed to the measurand(14).
Note 1. The parameter may be, for example, a standard deviation (or a
given multiple of it), or the half-width of an interval having a stated level
of confidence.
Note 2. Uncertainty of measurement comprises, in general, many
components. Some of these components may be evaluated from the
statistical distribution of results of a series of measurements and can be
characterised by experimental standard deviations. The other components,
which can also be characterised by standard deviations, are evaluated from
assumed probability distributions based on experience or other
information.
Note 3. It is understood that the result of a measurement is the best
estimate of the value of a measurand, and that all components of
uncertainty, including those arising from systematic effects, such as
components associated with corrections and reference standards, contribute
to the dispersion.
Traceability: property of the result of a measurement or the value of a
standard whereby it can be related to stated references, usually national or
international standards, through an unbroken chain of comparisons all
having stated uncertainties(13).
Reference material: material or substance one of whose property values
are sufficiently homogeneous and well established to be used for the
OIV-MA-AS1-11 : R2002
7
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
calibration of an apparatus, the assessment of a measurement method, or
for assigning values to materials(13).
Certified reference material: reference material, accompanied by a
certificate, one or more of whose property values are certified by a
procedure which establishes its traceability to an accurate realisation of the
unit in which the property values are expressed, and for which each
certified value is accompanied by an uncertainty at a stated level of
confidence(13).
2.2 Definitions of terms specific to this document
Internal quality control: set of procedures undertaken by laboratory staff
for the continuous monitoring of operation and the results of measurements
in order to decide whether results are reliable enough to be released.
Control material: material used for the purposes of internal quality
control and subjected to the same or part of the same measurement
procedure as that used for test materials.
Run (analytical run): set of measurements performed under repeatability
conditions.
Fitness for purpose: degree to which data produced by a measurement
process enables a user to make technically and administratively correct
decisions for a stated purpose
Analytical system: range of circumstances that contribute to the quality of
analytical data, including equipment, reagents, procedures, test materials,
personnel, environment and quality assurance measures.
3. QUALITY ASSURANCE PRACTICES AND INTERNAL QUALITY
CONTROL
3.1 Quality assurance
Quality assurance is the essential organisational infrastructure that
underlies all reliable analytical measurements. It is concerned with
achieving appropriate levels in matters such as staff training and
management, adequacy of the laboratory environment, safety, the storage,
integrity and identity of samples, record keeping, the maintenance and
OIV-MA-AS1-11 : R2002
8
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
calibration of instruments, and the use of technically validated and properly
documented methods . Failure in any of these areas might undermine
vigorous efforts elsewhere to achieve the desired quality of data. In recent
years these practices have been codified and formally recognised as
essential. However, the prevalence of these favourable circumstances by
no means ensures the attainment of appropriate data quality unless IQC is
conducted.
3.2 Choice of analytical method
It is important that laboratories restrict their choice of methods to those that
have been characterised as suitable for the matrix and analyte of interest.
The laboratory must possess documentation describing the performance
characteristics of the method, estimated under appropriate conditions.
The use of a method does not in itself guarantee the achievement of its
established performance characteristics. There is, for a given method, only
the potential to achieve a certain standard of reliability when the method is
applied under a particular set of circumstances. It is this collection of
circumstances, known as the "analytical system", that is therefore
responsible for the accuracy of analytical data. Hence it is important to
monitor the analytical system in order to achieve fitness for purpose. This
is the aim of the IQC measures undertaken in a laboratory.
3.3 Internal quality control and proficiency tests
Proficiency testing is a periodic assessment of the performance of
individual laboratories and groups of laboratories that is achieved by the
distribution by an independent testing body of typical materials for
unsupervised analysis by the participants(2).
Although important,
participation in proficiency testing schemes is not a substitute for IQC
measures, or vice versa.
Proficiency testing schemes can be regarded as a routine, but relatively
infrequent, check on analytical errors. Without the support of a
well-developed IQC system, the value of participation in a proficiency test
is negligible. Probably the main beneficial effect of proficiency tests is
that of encouraging participants to install effective quality control systems.
It has been shown that laboratories with effective IQC systems performed
better in a proficiency testing scheme(15).
OIV-MA-AS1-11 : R2002
9
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
4. INTERNAL QUALITY CONTROL PROCEDURES
4.1 Introduction
Internal quality control involves the practical steps undertaken to ensure
that errors in analytical data are of a magnitude appropriate for the use to
which the data will be put. The practice of IQC depends on the use of two
strategies, the analysis of reference materials to monitor trueness and
statistical control, and duplication to monitor precision.
The basic approach to IQC involves the analysis of control materials
alongside the test materials under examination. The outcome of the control
analyses forms the basis of a decision regarding the acceptability of the test
data. Two key points are worth noting in this context.
(i) The interpretation of control data must be based on documented,
objective criteria, and on statistical principles wherever possible.
(ii) The results of control analyses should be viewed primarily as
indicators of the performance of the analytical system, and only
secondarily as a guide to the errors associated with individual test results.
Substantial changes in the apparent accuracy of control determinations can
sometimes be taken to imply similar changes to data for contemporary test
materials, but correction of analytical data on the basis of this premise is
unacceptable.
4.2 General Approach - Statistical Control
The interpretation of the results of IQC analyses depends largely on the
concept of statistical control, which corresponds with stability of operation.
Statistical control implies that an IQC result x can be interpreted as arising
independently and at random from a normal population with mean  and
variance ².
Under these constraints only about 0.27% of results (x) would fall outside
the bounds of   3 . When such extreme results are encountered they
are regarded as being "out-of- control" and interpreted to mean that the
analytical system has started to behave differently. Loss of control
therefore implies that the data produced by the system are of unknown
accuracy and hence cannot be relied upon. The analytical system therefore
requires investigation and remedial action before further analysis is
undertaken. Compliance with statistical control can be monitored
graphically with Shewhart control charts (see Appendix 1). An equivalent
OIV-MA-AS1-11 : R2002
10
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
numerical approach, comparing values of z = (x-µ)/ against appropriate
values of the standard normal deviate, is also possible.
4.3
Internal quality control and fitness for purpose.
For the most part, the process of IQC is based on a description in terms of
the statistical parameters of an ongoing analytical system in normal
operation. Control limits are therefore based on the estimated values of
these parameters rather than measures derived from considerations of
fitness for purpose. Control limits must be narrower than the requirements
of fitness for purpose or the analysis would be futile.
The concept of statistical control is inappropriate, however, when the socalled ad hoc analysis is being undertaken. In ad hoc analysis the test
materials may be unfamiliar or rarely encountered, and runs are often made
up of only a few such test materials. Under these circumstances there is no
statistical basis for the construction of control charts. In such an instance
the analytical chemist has to use fitness for purpose criteria, historical data
or consistency with the visual properties of the test material for judging the
acceptability of the results obtained.
Either way, agreed methods of establishing quantitative criteria to
characterise fitness for purpose would be desirable. Unfortunately, this is
one of the less-developed aspects of IQC. In specific application areas
guidelines may emerge by consensus. For example, in environmental
studies it is usually recognised that relative uncertainties of less than ten
percent in the concentration of a trace analyte are rarely of consequence.
In food analysis the Horwitz curve(16) is sometimes used as a fitness for
purpose criterion.
Such criteria have been defined for clinical
(17,18)
analysis
. In some areas of applied geochemistry a systematic
approach has given rise to fitness for purpose criteria for sampling and
analytical precisions. However, it is not practicable here to give guidelines
in these areas, and at present no general principles can be advanced that
would allow specific applications to be addressed.
4.4 The nature of errors
Two main categories of analytical error are recognised, namely random
errors and systematic errors, which give rise to imprecision and bias
respectively. The importance of categorising errors in this way lies in the
fact that they have different sources, remedies and consequences for the
interpretation of data.
OIV-MA-AS1-11 : R2002
11
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
Random errors determine the precision of measurement. They cause
random positive and negative deviations of results about the underlying
mean value. Systematic errors comprise displacement of the mean of many
determinations from the true value. For the purposes of IQC two levels of
systematic error are worth consideration.
(i) Persistent bias affects the analytical system (for a given type of test
material) over a long period and affects all data. Such bias, if small in
relation to random error, may be identifiable only after the analytical
system has been in operation for a long time. It might be regarded as
tolerable, provided it is kept within prescribed bounds.
(ii) The run effect is exemplified by a deviation of the analytical system
during a particular run. This effect, where it is sufficiently large, will be
identified by IQC at the time of occurrence as an out-of-control condition.
The conventional division of errors between the random and the systematic
depends on the timescale over which the system is viewed. Run effects of
unknown source can be regarded in the long-term as the manifestation of a
random process. Alternatively, if a shorter-term view is taken, the same
variation could be seen as a bias-like change affecting a particular run.
The statistical model used for IQC in this document is as follows1. The
value of a measurement (x) in a particular run is given by:
x = true value + persistent bias + run effect + random error (+ gross
error).
The variance of x (
2
x
) in the absence of gross errors is given by:
 2x =  20 + 12
where
 20 = variance of the random error (within run) and
 12 = variance of the run effect.
1
The model could be extended if necessary to include other features of the analytical system
OIV-MA-AS1-11 : R2002
12
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
The variances of the true value and the persistent bias are both zero. An
analytical system in control is fully described by  20 , 12 and the value of
the persistent bias. Gross errors are implied when the analytical system
does not comply with such a description.
IQC AND WITHIN-RUN PRECISION
5
5.1
Precision and duplication
A limited control of within-run precision is achieved by the duplication
within a run of measurements made on test materials. The objective is to
ensure that the differences between paired results are consistent with or
better than the level implied by the value of o used by a laboratory for
IQC purposes2. Such a test alerts the user to the possibility of poor withinrun precision and provides additional information to help in interpreting
control charts. The method is especially useful in ad hoc analysis, where
attention is centred on a single run and information obtained from control
materials is unlikely to be completely satisfactory.
As a general approach all of the test materials, or a random selection from
them, are analysed in duplicate. The absolute differences d = x1 - x2
between duplicated analytical results x1 and x2 are tested against an upper
control limit based on an appropriate value of o. However, if the test
materials in the run have a wide range of concentration of analyte, no
single value of o can be assumed(19).
Duplicates for IQC must reflect as far as possible the full range of variation
present in the run. They must not be analysed as adjacent members of the
run, otherwise they will reveal only the smallest possible measure of
analytical variability. The best placing of duplicates is at random within
each run. Moreover the duplication required for IQC requires the complete
and independent analysis (preferably blind) of separate test portions of the
test material. A duplication of the instrumental measurement of a single
test solution would be ineffective because the variations introduced by the
preliminary chemical treatment of the test material would be absent.
2
There is no intention here of estimating the standard deviation of repeatability r from the IQC data or of
comparing estimates: there would usually be too few results for a satisfactory outcome. Where such an estimate is
needed the formula sr   d 2 / 2n can be used.
OIV-MA-AS1-11 : R2002
13
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
5.2 Interpretation of duplicate data
5.2.1 Narrow concentration range. In the simplest situation the test
materials comprising the run have a small range of analyte concentrations
so that a common within-run standard deviation o can be applied.
A value of this parameter must be estimated to provide a control limit. The
upper 95% bound of d is 2 2 o and on average only about three in a
thousand results should exceed 3 2 o.A group of n duplicated results can
be interpreted in several ways.
For example, the standardised difference
zd = d / 2 o
should have a normal distribution with zero mean and unit standard
deviation. The sum of a group of n such results would have a standard
deviation of n so only about three runs in a thousand would produce a
value of zd  3 n . Alternatively a group of n values of zd from a run
can be combined to form z 2d and the result interpreted as a sample from a
chi-squared distribution with n degrees of freedom, ( 2n ). Some caution is
needed in the use of this statistic, however, as it is sensitive to outlying
results.
5.2.2 Wide concentration range. If the test materials comprising a run
have a wide range of analyte concentrations, no common standard of
precision (o) can be assumed. In such an instance 0 must be expressed
as a functional relationship with concentration. The value of concentration
for a particular material is taken to be (x1 + x2)/2, and an appropriate value
of o obtained from the functional relationship, the parameters of which
have to be estimated in advance.
6. CONTROL MATERIALS IN IQC
6.1 Introduction
Control materials are characterised substances that are inserted into the run
alongside the test materials and subjected to exactly the same treatment. A
OIV-MA-AS1-11 : R2002
14
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
control material must contain an appropriate concentration of the analyte,
and a value of that concentration must be assigned to the material. Control
materials act as surrogates for the test materials and must therefore be
representative, i.e., they should be subject to the same potential sources of
error. To be fully representative, a control material must have the same
matrix in terms of bulk composition, including minor constituents that may
have a bearing on accuracy. It should also be in a similar physical form,
i.e., state of comminution, as the test materials. There are other essential
characteristics of a control material. It must be adequately stable over the
period of interest. It must be possible to divide the control material into
effectively identical portions for analysis. It is often required in large
amounts to allow its use over an extended period.
Reference materials in IQC are used in combination with control charts
that allow both persistent bias and run effects to be addressed (Appendix
1). Persistent bias is evident as a significant deviation of the centre line
from the assigned value. The variation in the run effect is predictable in
terms of a standard deviation when the system is under statistical control,
and that standard deviation is used to define action limits and warning
limits at appropriate distances from the true value.
6.2 The role of certified reference materials
Certified reference materials (CRM) as defined in Section 2 (i.e., with a
statement of uncertainty and traceability), when available and of suitable
composition, are ideal control materials in that they can be regarded for
traceability purposes as ultimate standards of trueness(20). In the past
CRMs were regarded as being for reference purposes only and not for
routine use. A more modern approach is to treat CRMs as consumable and
therefore suitable for IQC.
The use of CRMs in this way is, however, subject to a number of
constraints.
(i) Despite the constantly increasing range of CRMs available, for the
majority of analyses there is no closely matching CRM available.
(ii) Although the cost of CRMs is not prohibitive in relation to the total
costs of analysis, it may not be possible for a laboratory with a wide range
of activities to stock every relevant kind of reference material.
(iii) The concept of the reference material is not applicable to materials
where either the matrix or the analyte is unstable.
OIV-MA-AS1-11 : R2002
15
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
(iv) CRMs are not necessarily available in sufficient amounts to provide
for IQC use over extended periods.
(v) It must be remembered that not all apparently certified reference
materials are of equal quality. Caution is suggested when the information
on the certificate is inadequate.
If for any of the above reasons the use of a CRM is not appropriate it falls
on individual laboratories or groups of laboratories to prepare their own
control materials and assign traceable3 values of analyte concentration to
them. Such a material is sometimes referred to as a "house reference
material" (HRM). Suggestions for preparing HRMs are listed in Section
6.3. Not all of the methods described there are applicable to all analytical
situations.
6.3 Preparation of control materials
6.3.1 Assigning a true value by analysis. In principle a working value
can be assigned to a stable reference material simply by careful analysis.
However, precautions are necessary to avoid biases in the assigned value.
This requires some form of independent check such as may be provided by
analysis of the materials in a number of laboratories and where possible,
the use of methods based on different physico-chemical principles. Lack of
attention to independent validation of control materials has been shown to
be a weakness in IQC systems(15).
One way of establishing a traceable assigned value in a control material is
to analyse a run comprising the candidate material and a selection of
matching CRMs, with replication and randomisation. This course of action
would be appropriate if only limited amounts of CRMs were available.
The CRMs must be appropriate in both matrix composition and analyte
concentration. The CRMs are used directly to calibrate the analytical
procedure for the analysis of the control material. An appropriate
analytical method is a prerequisite for this approach. It would be a
dangerous approach if, say, a minor and variable fraction of the analyte
3
Where a CRM is not available traceability only to a reference method or to a batch of a reagent supplied by a
manufacturer may be necessary.
OIV-MA-AS1-11 : R2002
16
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
were extracted for measurement. The uncertainty introduced into the
assigned value must also be considered.
6.3.2 Materials validated in proficiency testing comprise a valuable source
of control materials. Such materials would have been analysed by many
laboratories using a variety of methods. In the absence of counterindications, such as an obvious bias or unusual frequency distribution of
results, the consensus of the laboratories could be regarded as a validated
assigned value to which a meaningful uncertainty could be attached. (There
is a possibility that the consensus could suffer from a bias of consequence,
but this potential is always present in reference values.) There would be a
theoretical problem of establishing the traceability of such a value, but that
does not detract from the validity of the proposed procedure. The range of
such materials available would be limited, but organisers of proficiency
tests could ensure a copious supply by preparing batches of material in
excess of the immediate requirements of the round. The normal
requirements of stability would have to be demonstrable.
6.3.3 Assigning a true value by formulation. In favourable instances a
control material can be prepared simply by mixing constituents of known
purity in predetermined amounts. For example, this approach would often
be satisfactory in instances where the control material is a solution.
Problems are often encountered in formulation in producing solid control
materials in a satisfactory physical state or in ensuring that the speciation
and physical distribution of the analyte in the matrix is realistic. Moreover
an adequate mixing of the constituents must be demonstrable.
6.3.4 Spiked control materials. "Spiking" is a way of creating a control
material in which a value is assigned by a combination of formulation and
analysis. This method is feasible when a test material essentially free of
the analyte is available. After exhaustive analytical checks to ensure the
background level is adequately low, the material is spiked with a known
amount of analyte. The reference sample prepared in this way is thus of
the same matrix as the test materials to be analysed and of known analyte
level - the uncertainty in the assigned concentration is limited only by the
possible error in the unspiked determination. However, it may be difficult
to ensure that the speciation, binding and physical form of the added
analyte is the same as that of the native analyte and that the mixing is
adequate.
6.3.5 Recovery Checks. If the use of a reference material is not practicable
then a limited check on bias is possible by a test of recovery. This is
especially useful when analytes or matrices cannot be stabilised or when ad
OIV-MA-AS1-11 : R2002
17
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
hoc analysis is executed. A test portion of the test material spiked with a
known amount of the analyte and analysed alongside the original test
material. The recovery of the added analyte (known as the "marginal
recovery") is the difference between the two measurements divided by the
amount that is added. The obvious advantages of recovery checks are that
the matrix is representative and the approach is widely applicable - most
test materials can be spiked by some means. However, the recovery check
suffers from the disadvantage previously noted regarding the speciation,
binding and physical distribution of the analyte. Furthermore, the
assumption of an equivalent recovery of the analyte added as a spike and of
the native analyte may not be valid. However, it can normally be assumed
that a poor performance in a recovery check is strongly indicative of a
similar or worse performance for the native analyte in the test materials.
Spiking and recovery testing as an IQC method must be distinguished from
the method of standard additions, which is a measurement procedure: a
single spiking addition cannot be used to fulfil the roles of both
measurement and IQC.
6.4
Blank determinations
Blank determinations are nearly always an essential part of the analytical
process and can conveniently be effected alongside the IQC protocol. The
simplest form of blank is the "reagent blank", where the analytical
procedure is executed in all respects apart from the addition of the test
portion. This kind of blank, in fact, tests more than the purity of the
reagents. For example it is capable of detecting contamination of the
analytical system originating from any source, e.g., glassware and the
atmosphere, and is therefore better described as a "procedural blank". In
some instances, better execution of blank determinations is achieved if a
simulated test material is employed. The simulant could be an actual test
material known to be virtually analyte-free or a surrogate (e.g., ashless
filter paper used instead of plant material). Where it can be contrived, the
best type of blank is the "field blank", which is a typical matrix with zero
concentration of analyte.
An inconsistent set of blanks in a run suggests sporadic contamination and
may add weight to IQC evidence suggesting the rejection of the results.
When an analytical protocol prescribes the subtraction of a blank value, the
blank value must be subtracted also from the results of the control
materials before they are used in IQC.
OIV-MA-AS1-11 : R2002
18
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
6.5 Traceability in spiking and recovery checks
Potential problems of the traceability of reagents used for spikes and
recovery checks must be guarded against. Under conditions where CRMs
are not available, traceability can often be established only to the batch of
analyte provided by a manufacturer. In such cases, confirmation of
identity and a check on purity must be made before use. A further
precaution is that the calibration standards and spike should not be
traceable to the same stock solution of analyte or the same analyst. If such
a common traceability existed, then the corresponding sources of error
would not be detected by the IQC.
7. RECOMMENDATIONS
The following recommendations represent integrated approaches to IQC that are
suitable for many types of analysis and applications areas. Managers of laboratory
quality systems will have to adapt the recommendations to the demands of their
own particular requirements. Such adaption could be implemented, for example, by
adjusting the number of duplicates and control material inserted into a run, or by
the inclusion of any additional measures favoured in the particular application area.
The procedure finally chosen and its accompanying decision rules must be codified
in an IQC protocol that is separate from the analytical system protocol.
The practical approach to quality control is determined by the frequency with
which the measurement is carried out and the size and nature of each run. The
following recommendations are therefore made. The use of control charts and
decision rules are covered in Appendix 1.
In each of the following the order in the run in which the various materials are
analysed should be randomised if possible. A failure to randomise may result in an
underestimation of various components of error.
(i) Short (e.g., n<20) frequent runs of similar materials. Here the concentration
range of the
analyte in the run is relatively small, so a common value of standard deviation can
be assumed.
Insert a control material at least once per run. Plot either the individual values
obtained, or
The mean value, on an appropriate control chart. Analyse in duplicate at least half
of the
Test materials, selected at random. Insert at least one blank determination.
OIV-MA-AS1-11 : R2002
19
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
(ii) Longer (e.g., n>20) frequent runs of similar materials. Again a common level
of standard
deviation is assumed.
Insert the control material at an approximate frequency of one per ten test
materials. If the run size is likely to vary from run to run it is easier to standardise
on a fixed number of insertions per run and plot the mean value on a control chart
of means. Otherwise plot individual values.
Analyse in duplicate a minimum of five test materials selected at random. Insert
one blank
determination per ten test materials.
(iii) Frequent runs containing similar materials but with a wide range of analyte
concentration.
Here we cannot assume that a single value of standard deviation is applicable.
Insert control materials in total numbers approximately as recommended above.
However, there should be at least two levels of analyte represented, one close to the
median level of typical test materials, and the other approximately at the upper or
lower decile as appropriate. Enter values for the two control materials on separate
control charts. Duplicate a minimum of five test materials, and insert one
procedural blank per ten test materials.
(iv) Ad hoc analysis. Here the concept of statistical control is not applicable. It is
assumed, however, that the materials in the run are of a single type, i.e., sufficiently
similar for general conclusions on errors to be made.
Carry out duplicate analysis on all of the test materials. Carry out spiking or
recovery tests or use a formulated control material, with an appropriate number of
insertions (see above), and with different concentrations of analyte if appropriate.
Carry out blank determinations. As no control limits are available, compare the
bias and precision with fitness for purpose limits or other established criteria..
8. CONCLUSIONS
Internal quality control is an essential aspect of ensuring that data released from a
laboratory are fit for purpose. If properly executed, quality control methods can
monitor the various aspects of data quality on a run-by-run basis. In runs where
performance falls outside acceptable limits, the data produced can be rejected and,
after remedial action on the analytical system, the analysis can be repeated.
OIV-MA-AS1-11 : R2002
20
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
It must be stressed, however, that internal quality control is not foolproof even
when properly executed. Obviously it is subject to "errors of both kinds", i.e., runs
that are in control will occasionally be rejected and runs that are out of control
occasionally accepted. Of more importance, IQC cannot usually identify sporadic
gross errors or short-term disturbances in the analytical system that affect the
results for individual test materials. Moreover, inferences based on IQC results are
applicable only to test materials that fall within the scope of the analytical method
validation. Despite these limitations, which professional experience and diligence
can alleviate to a degree, internal quality control is the principal recourse available
for ensuring that only data of appropriate quality are released from a laboratory.
When properly executed it is very successful.
Finally, it must be appreciated that a perfunctory execution of any quality system
will not guarantee the production of data of adequate quality. The correct
procedures for feedback, remedial action and staff motivation must also be
documented and acted upon. In other words, there must be a genuine commitment
to quality within a laboratory for an internal quality control programme to succeed,
i.e., the IQC must be part of a total quality management system.
9. REFERENCES
1
"Protocol for the Design, Conduct and Interpretation of Method
Performance Studies", Edited W Horwitz, Pure Appl. Chem., 1988, 60,
855- 864. (Revision in press)
2
"The International Harmonised Protocol for the Proficiency Testing of
(Chemical) Analytical Laboratories", Edited M Thompson and R Wood,
Pure Appl. Chem., 1993, 65, 2123-2144. (Also published in J. AOAC
International, 1993, 76, 926-940.
3
"IFCC approved recommendations on quality control in clinical chemistry.
Part 4: internal quality control", J. Clin. Chem. Clin. Biochem., 1980, 18,
534-541.
4
S Z Cekan, S B Sufi and E W Wilson, "Internal quality control for assays
of reproductive hormones: Guidelines for laboratories". WHO, Geneva,
1993.
5
M Thompson, "Control procedures in geochemical analysis", in R J
Howarth (Ed), "Statistics and data analysis in geochemical prospecting",
Elsevier, Amsterdam, 1983.
6
M Thompson, "Data quality in applied geochemistry: the requirements
and how to achieve them", J. Geochem. Explor., 1992, 44, 3-22.
7
Health and Safety Executive, "Analytical quality in workplace air
monitoring", London, 1991.
OIV-MA-AS1-11 : R2002
21
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
8
9
10
11
12
13
14
15
16
17
18
19
20
"A protocol for analytical quality assurance in public analysts'
laboratories", Association of Public Analysts, 342 Coleford Road,
Sheffield S9 5PH, UK, 1986.
"Method evaluation, quality control, proficiency testing" (AMIQAS PC
Program), National Institute of Occupational Health, Denmark, 1993.
ISO 8402:1994.
"Quality assurance and quality management vocabulary".
ISO 3534 -1: 1993 (E/F). "Statistics, vocabulary and symbols - Part 1:
Probability and general statistical terms".
ISO Guide 30:1992. "Terms and definitions used in connections with
reference materials"
"International vocabulary for basic and general terms in metrology" , 2nd
Edition, 1993, ISO, Geneva.
"Guide to the expression of uncertainty in measurement", ISO, Geneva,
1993.
M Thompson and P J Lowthian, Analyst, 1993, 118, 1495-1500.
W Horwitz, L R Kamps and K W Boyer, J. Assoc. Off. Anal. Chem.,
1980, 63, 1344.
D Tonks, Clin. Chem., 1963, 9, 217-223.
G C Fraser, P H Petersen, C Ricos and R Haeckel, "Proposed quality
specifications for the imprecision and inaccuracy of analytical systems for
clinical chemistry", Eur. J. Clin. Chem. Clin. Biochem., 1992, 30, 311-317.
M Thompson, Analyst, 1988, 113, 1579-1587.
ISO Guide 33: 1989, "Uses of Certified Reference Materials", Geneva.
OIV-MA-AS1-11 : R2002
22
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
APPENDIX 1. SHEWHART CONTROL CHARTS
1. INTRODUCTION
The theory, construction and interpretation of the Shewhart chart(1) are detailed in
numerous texts on process quality control and applied statistics, and in several ISO
standards(2-5). There is a considerable literature on the use of the control chart in
clinical chemistry(6,7). Westgard and co-workers have formulated multiple rules
for the interpretation of such control charts(8), and the power of these results has
been studied in detail(9-10). In this appendix only simple Shewhart charts are
considered.
In IQC a Shewhart control chart is obtained when values of concentration
measured on a control material in successive runs are plotted on a vertical axis
against the run number on the horizontal axis. If more than one analysis of a
particular control material is made in a run, either the individual results x or the
mean value x can be used to form a control chart. The chart is completed by
horizontal lines derived from the normal distribution N(,2) that is taken to
describe the random variations in the plotted values. The selected lines for control
purposes are   2 and   3. Different values of  are required for charts of
individual values and of means. For a system in statistical control, on average
about one in twenty values fall outside the   2 lines, called the "warning
limits", and only about three in one thousand fall outside the   3 lines, the
"action limits". In practice the estimates x and s of the parameters  and  are
used to construct the chart. A persistent bias is indicated by a significant difference
between x and the assigned value
2. ESTIMATES OF THE PARAMETERS  and 
An analytical system under control exhibits two sources of random variation, the
within-run, characterised by variance  20 and the between-run with variance 12 .
The two variances are typically comparable in magnitude. The standard deviation
 x used in a chart of individual values is given by
 x = ( 20 + 12 )1/ 2
whereas for a control chart of mean values the standard deviation is given by
 x  ( 0 / n  1 )
2
2 1/ 2
OIV-MA-AS1-11 : R2002
23
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
where n is the number of control measurements in a run from which the mean is
calculated. The value of n therefore must be constant from run to run, otherwise
control limits would be impossible to define. If a fixed number of repeats of a
control material per run cannot be guaranteed (e.g., if the run length were variable)
then charts of individual values must be used. Furthermore the equation indicates
that  x or  x must be estimated with care. An attempt to base an estimate on
repeat values from a single run would result in unduly narrow control limits.
Estimates must therefore include the between-run component of variance. If the
use of a particular value of n can be assumed at the outset, then  x can be
estimated directly from the m means xi 
n
x
j 1
/n
ij
(i = 1,.....,m) of the n repeats in each of m successive runs.
Thus the estimate of  is
x 

xi / m
i
and the estimate of  x is
s 
x
 (x  x )
m1
i
2
i
If the value of n is not predetermined, then separate estimates of oand 1 could be
obtained by one-way analysis of variance. If the mean squares within- and
between- groups are MSw and MSb respectively, then
 20 is estimated by MSw and
 12 is estimated by (MSb - MSw)/n
Often in practice it is necessary to initiate a control chart with data collected from a
small number of runs, which may be to a degree unrepresentative, as estimates of
standard deviation are very variable unless large numbers of observations are used.
Moreover, during the initial period, the occurrence of out-of-control conditions are
OIV-MA-AS1-11 : R2002
24
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
more than normally likely and will produce outlying values. Such values would
bias x and inflate s beyond its proper value. It is therefore advisable to recalculate
x and s after a further "settling down" period. One method of obviating the effects
of outliers in the calculation is to reject them after the application of Dixon's Q or
Grubbs'(11) test, and then use the classical statistics given above. Alternatively,
the methods of robust statistics could be applied to the data(12, 13).
3. THE INTERPRETATION OF CONTROL CHARTS
The following simple rules can be applied to control charts of individual results or
of means.
Single control chart. An out-of-control condition in the analytical system is
signalled if any of the following occur.
(i) The current plotting value falls outside the action limits.
(ii) The current value and the previous plotting value fall outside the
warning limits but within the actions limits.
(iii) Nine successive plotting values fall on the same side of the mean line.
Two control charts. When two different control materials are used in each run, the
respective control charts are considered simultaneously. This increases the chance
of a type 1 error (rejection of a sound run) but decreases the chance of a type 2
error (acceptance of a flawed run). An out-of-control condition is indicated if any
of the following occur.
(i) At least one of the plotting values falls outside the action limits.
(ii) Both of the plotting values are outside the warning limits.
(iii) The current value and the previous plotting value on the same control
chart both fall outside the warning limits.
(iv)
Both control charts simultaneously show that four successive
plotting values on the same side of the mean line.
(v) One of the charts shows nine successive plotting values falling on the
same side of the mean line.
A more thorough treatment of the control chart can be obtained by the application
of the full Westgard rules, illustrated in Figure 2.
OIV-MA-AS1-11 : R2002
25
COMPENDIUM OF INTERNATIONAL METHODS OF ANALYSIS –OIV
Harmonized guidelines for internal quality control in analytical
chemistry laboratories
The analytical chemist should respond to an out-of-control condition by cessation
of analysis pending diagnostic tests and remedial action followed by rejection of
the results of the run and reanalysis of the test materials.
4. REFERENCES
W A Shewhart, "Economic control of quality in manufactured product",
1
Van
Nostrand, New York, 1931.
ISO 8258:1991. "Shewhart control charts".
2
3
ISO 7873:1993 "Control charts for arithmetic means with warning limits".
4
ISO 7870:1993. "Control charts - general guide and introduction".
5
ISO 7966:1993. "Acceptance control charts".
6
S Levey and E R Jennings, Am. J. Clin. Pathol., 1950, 20, 1059-1066.
A B J Nix, R J Rowlands, K W Kemp, D W Wilson and K Griffiths, Stat.
7
Med., 1987,6,425-440.
8
J O Westgard, P L Barry and M R Hunt, Clin. Chem., 1981, 27, 493-501.
C A Parvin, Clin. Chem., 1992, 38, 358-363.
9
10
J Bishop and A B J Nix, Clin. Chem., 1993, 39, 1638-1649.
11
W Horwitz, Pure Appl. Chem., (in press).
12
Analytical Methods Committee, Analyst, 1989, 114, 1693-1697.
13
Analytical Methods Committee, Analyst, 1989, 114, 1699-1702.
---------------------------Technical report from the Symposium on the 'Harmonisation of quality assurance systems for Analysis
Laboratories, Washington DC, USA, 22-23 July 1993 sponsored by IUPAC, ISO et AOAC International
Prepared for publication by MICHAEL THOMPSON1 and ROGER WOOD2
1Department of Chemistry, Birkbeck College (University of London), London WC1H OPP, UK
2MAFF Food Science Laboratory, Norwich Research Park, Colney, Norwich NR4 7UQ, UK
1991-95 work group :
Chairman : M. Parkany (Switzerland) ; Membres : T. Anglov (Denmark) ; K. Bergknut (Norway and
sweden) ; P. De Biève (Belgium) ; K.-G. von Boroviczény (Germany) ; J.M. Christensen (Denmark) ; T.D.
Geary (South Australia) ; R. Greenhalgh (Canada) ; A.J. Head (United Kingdom) ; P.T. Holland (New
Zealand) ; W. Horwitz (USA) . A. Kallner (Sweden; J. Kristiansen (Denmark) ; S.H.H. Olrichs
(Netherlands) ; N. Palmer (USA) . M. Thompson (United Kingdom) ; M.J. Vernengo (Argentina) ; R. Wood
(United Kingdom).
OIV-MA-AS1-11 : R2002
26
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Practical guide for the validation, quality control, and
uncertainty assessment of an alternative
oenological analysis method
(Resolution 10/2005)
Contents
1.
PURPOSE ..................................................................................................................... 5
2.
PREAMBLE AND SCOPE ......................................................................................... 5
3.
GENERAL VOCABULARY ....................................................................................... 6
4.
GENERAL PRINCIPLES ......................................................................................... 12
4.1
4.2
5.
METHODOLOGY .................................................................................................. 12
DEFINITION OF MEASUREMENT ERROR ........................................................... 13
VALIDATING A METHOD ..................................................................................... 14
5.1
METHODOLOGY .................................................................................................. 14
5.2
SECTION ONE: SCOPE OF METHOD ................................................................... 15
5.2.1
Definition of analyzable matrices ................................................................. 15
5.2.2
Detection and quantification limit ................................................................ 15
5.2.2.1
Normative definition ............................................................................................. 16
5.2.2.2
Reference documents ............................................................................................ 16
5.2.2.3
Application ........................................................................................................... 16
5.2.2.4
Procedure .............................................................................................................. 16
5.2.2.4.1 Determination on blank .................................................................................. 16
5.2.2.4.1.1
Scope ...................................................................................................... 16
5.2.2.4.1.2
Basic protocol and calculations .............................................................. 17
5.2.2.4.2 Approach by linearity study ........................................................................... 18
5.2.2.4.2.1
Scope ...................................................................................................... 18
5.2.2.4.2.2
Basic protocol and calculations .............................................................. 19
5.2.2.4.3 Graphic approach based on the background noise of the recording ............... 21
5.2.2.4.3.1
Scope ...................................................................................................... 21
5.2.2.4.3.2
Basic protocol and calculation................................................................ 21
5.2.2.4.4 Checking a predetermined quantification limit .............................................. 21
5.2.2.4.4.1
Scope ...................................................................................................... 21
5.2.2.4.4.2
Basic protocol and calculation................................................................ 22
5.2.3
Robustness .................................................................................................... 24
5.2.3.1
Definition .............................................................................................................. 24
5.2.3.2
Determination ....................................................................................................... 24
5.3
SECTION TWO: SYSTEMATIC ERROR STUDY.................................................... 24
5.3.1
5.3.1.1
5.3.1.2
5.3.1.3
Linearity study .............................................................................................. 24
Normative definition ............................................................................................. 24
Reference documents ............................................................................................ 24
Application ........................................................................................................... 24
OIV-MA-AS1-12 : R2005
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.3.1.4
ISO 11095-type approach ..................................................................................... 25
5.3.1.4.1 Basic protocol ................................................................................................ 25
5.3.1.4.2 Calculations and results.................................................................................. 26
5.3.1.4.2.1
Defining the regression model ................................................................ 26
5.3.1.4.2.2
Estimating parameters ............................................................................ 27
5.3.1.4.2.3
Charts ..................................................................................................... 27
5.3.1.4.2.4
Test of the linearity assumption .............................................................. 29
5.3.1.4.2.4.1
Definitions of errors linked to calibration ....................................... 29
5.3.1.4.2.4.2
Fischer-Snedecor test ...................................................................... 31
5.3.1.5
ISO 8466-type approach ....................................................................................... 32
5.3.1.5.1 Basic protocol ................................................................................................ 32
5.3.1.5.2 Calculations and results.................................................................................. 33
5.3.1.5.2.1
Defining the linear regression model ...................................................... 33
5.3.1.5.2.2
Defining the polynomial regression model ............................................. 33
5.3.1.5.2.3
Comparing residual standard deviations ................................................. 34
5.3.2
Specificity ..................................................................................................... 36
5.3.2.1
Normative definition ............................................................................................. 36
5.3.2.2
Application ........................................................................................................... 36
5.3.2.3
Procedures ............................................................................................................ 36
5.3.2.3.1 Standard addition test ..................................................................................... 36
5.3.2.3.1.1
Scope ...................................................................................................... 36
5.3.2.3.1.2
Basic protocol ......................................................................................... 36
5.3.2.3.1.3
Calculations and results .......................................................................... 37
5.3.2.3.1.3.1
Study of the regression line r = a + b.v ........................................... 37
5.3.2.3.1.3.2
Analysis of the results ..................................................................... 38
5.3.2.3.1.3.3
Overlap line graphics ...................................................................... 40
5.3.2.3.2 Study of the influence of other compounds on the measurement result ......... 40
5.3.2.3.2.1
Scope ...................................................................................................... 40
5.3.2.3.2.2
Basic protocol and calculations .............................................................. 40
5.3.2.3.2.3
Interpretation .......................................................................................... 41
5.3.3
Study of method accuracy ............................................................................ 43
5.3.3.1
Presentation of the step ......................................................................................... 43
5.3.3.1.1 Definition ....................................................................................................... 43
5.3.3.1.2 General principles .......................................................................................... 43
5.3.3.1.3 Reference documents ..................................................................................... 43
5.3.3.2
Comparison of the alternative method with the OIV reference method ................ 44
5.3.3.2.1 Scope .............................................................................................................. 44
5.3.3.2.2 Accuracy of the alternative method compared with the reference method ..... 44
5.3.3.2.2.1
Definition ............................................................................................... 44
5.3.3.2.2.2
Scope ...................................................................................................... 44
5.3.3.2.2.3
Basic protocol and calculations .............................................................. 44
5.3.3.2.2.4
Interpretation .......................................................................................... 46
5.3.3.3
Comparison by interlaboratory tests ..................................................................... 47
5.3.3.3.1 Scope .............................................................................................................. 47
5.3.3.3.2 Basic protocol and calculations ...................................................................... 48
5.3.3.3.3 Interpretation .................................................................................................. 49
5.3.3.4
Comparison with reference materials .................................................................... 50
5.3.3.4.1 Scope .............................................................................................................. 50
5.3.3.4.2 Basic protocol and calculations ...................................................................... 50
5.3.3.4.3 Interpretation .................................................................................................. 51
5.4
SECTION THREE: RANDOM ERROR STUDY ....................................................... 52
5.4.1
General principle ......................................................................................... 52
OIV-MA-AS1-12 : R2005
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.4.2
5.4.3
Reference documents .................................................................................... 53
Precision of the method ................................................................................ 53
5.4.3.1
Definition .............................................................................................................. 53
5.4.3.2
Scope .................................................................................................................... 53
5.4.3.3
General theoretical case ........................................................................................ 54
5.4.3.3.1 Basic protocol and calculations ...................................................................... 54
5.4.3.3.1.1
Calculations with several test materials .................................................. 54
5.4.3.3.1.2
Calculations with 1 test material............................................................. 56
5.4.3.4
Repeatability ......................................................................................................... 57
5.4.3.4.1 Definitions ...................................................................................................... 57
5.4.3.4.2 Scope .............................................................................................................. 58
5.4.3.4.3 Basic protocol and calculations ...................................................................... 58
5.4.3.4.3.1
General case ........................................................................................... 58
5.4.3.4.3.2
Particular case applicable to only 1 repetition ........................................ 58
5.4.3.4.4 Comparison of repeatability ........................................................................... 60
5.4.3.4.4.1
Determination of the repeatability of each method ................................. 60
5.4.3.4.4.2
Fischer-Snedecor test ............................................................................. 61
5.4.3.5
Intralaboratory reproducibility .............................................................................. 62
5.4.3.5.1 Definition ....................................................................................................... 62
5.4.3.5.2 Scope .............................................................................................................. 62
5.4.3.5.3 Basic protocol and calculations ...................................................................... 62
6.
QUALITY CONTROL OF ANALYSIS METHODS (IQC) .................................. 64
6.1
REFERENCE DOCUMENTS .................................................................................. 64
6.2
GENERAL PRINCIPLES ........................................................................................ 64
6.3
REFERENCE MATERIALS .................................................................................... 64
6.4
CHECKING THE ANALYTICAL SERIES ............................................................... 66
6.4.1
Definition ...................................................................................................... 66
6.4.2
Checking accuracy using reference materials .............................................. 66
6.4.3
Intraseries precision ..................................................................................... 66
6.4.4
Internal standard .......................................................................................... 67
6.5
CHECKING THE ANALYSIS SYSTEM .................................................................. 67
6.5.1
Definition ...................................................................................................... 67
6.5.2
Shewhart chart ............................................................................................. 67
6.5.2.1
6.5.2.2
6.5.2.3
6.5.3
6.5.4
Data acquisition .................................................................................................... 67
Presentation of results and definition of limits...................................................... 68
Using the Shewhart chart ...................................................................................... 69
Internal comparison of analysis systems ...................................................... 70
External comparison of the analysis system ................................................. 70
6.5.4.1
Analysis chain of interlaboratory comparisons ..................................................... 70
6.5.4.2
Comparison with external reference materials ...................................................... 70
6.5.4.2.1 Standard uncertainty of reference material ..................................................... 71
6.5.4.2.2 Defining the validity limits of measuring reference material .......................... 71
7.
ASSESSMENT OF MEASUREMENT UNCERTAINTY ...................................... 72
7.1
7.2
7.3
7.4
DEFINITION ......................................................................................................... 72
REFERENCE DOCUMENTS .................................................................................. 73
SCOPE .................................................................................................................. 73
METHODOLOGY .................................................................................................. 74
OIV-MA-AS1-12 : R2005
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
7.4.1
7.4.2
7.4.3
Definition of the measurand, and description of the quantitative analysis
method .......................................................................................................... 74
Critical analysis of the measurement process .............................................. 75
Estimation calculations of standard uncertainty
(intralaboratory approach) .......................................................................... 75
7.4.3.1
7.4.3.2
7.4.3.3
Principle ................................................................................................................ 75
Calculating the standard deviation of intralaboratory reproducibility ................... 78
Estimating typical sources of systematic errors not taken into account underµ
reproducibility conditions ..................................................................................... 79
7.4.3.3.1 Gauging error (or calibration error) ................................................................ 79
7.4.3.3.1.1
Procedure................................................................................................ 79
7.4.3.3.1.2
Calculations and results .......................................................................... 80
7.4.3.3.1.3
Estimating the standard uncertainty associated the gauging line
(or calibration line) ................................................................................. 81
7.4.3.3.2 Bias error ........................................................................................................ 82
7.4.3.3.2.1
Methods adjusted with only one certified reference material ................. 82
7.4.3.3.2.2
Methods adjusted with several reference materials (gauging ranges etc) 82
7.4.3.3.3 Matrix effect ................................................................................................... 83
7.4.3.3.3.1
Definition ............................................................................................... 83
7.4.3.3.4 Sample effect .................................................................................................. 86
7.4.4
Estimating standard uncertainty by interlaboratory tests ............................ 86
7.4.4.1
7.4.4.2
Principle ................................................................................................................ 86
Using the standard deviation of interlaboratory and intramethod
reproducibility SRinter (method)............................................................................. 87
7.4.4.3
Using the standard deviation of interlaboratory and intermethod
reproducibility SRinter ............................................................................................ 87
7.4.4.4
Other components in the uncertainty budget......................................................... 88
7.5
EXPRESSING EXPANDED UNCERTAINTY ......................................................... 88
OIV-MA-AS1-12 : R2005
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
1. Purpose
The purpose of this guide is to assist oenological laboratories carrying out serial
analysis as part of their validation, internal quality control and uncertainty
assessment initiatives concerning the standard methods they use.
2. Preamble and scope
International standard ISO 17025, defining the "General Requirements for the
Competence of Testing and Calibration Laboratories", states that the accredited
laboratories must, when implementing a alternative analytical method, make sure
of the quality of the results obtained. To do so, it indicates several steps. The first
step consists in defining the customers' requirements concerning the parameter in
question, in order to determine, thereafter, whether the method used meets those
requirements. The second step includes initial validation for non-standardized,
modified or laboratory-developed methods. Once the method is applied, the
laboratories must use inspection and traceability methods in order to monitor the
quality of the results obtained. Finally, they must assess the uncertainty of the
results obtained.
In order to meet these requirements, the laboratories have a significant reference
system at their disposal comprising a large number of international guides and
standards. However, in practice, the application of these texts is delicate since,
because they address every category of calibration and test laboratory, they remain
very general and presuppose, on behalf of the reader, in-depth knowledge of the
mathematical rules applicable to statistical data processing.
This guide is based on this international reference system, taking into account the
specific characteristics of oenology laboratories routinely carrying out analyses on
series of must or wine samples. Defining the scope of application in this way
enabled a relevant choice of suitable tools to be made, in order to retain only those
methods most suitable for that scope. Since it is based on the international
reference system, this guide is therefore strictly compliant with it. Readers,
however, wishing to study certain points of the guide in greater detail can do so by
referring to the international standards and guides, the references for which are
given in each chapter.
The authors have chosen to combine the various tools meeting the requirements of
the ISO 17025 standard since there is an obvious solution of continuity in their
OIV-MA-AS1-12 : R2005
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
application, and the data obtained with certain tools can often be used with the
others. In addition, the mathematical resources used are often similar.
The various chapters include application examples, taken from oenology
laboratories using these tools.
It is important to point out that that this guide does not pretend to be exhaustive. It
is only designed to present, in as clear and applicable a way as possible, the
contents of the requirements of the ISO 17025 standard and the basic resources
that can be implemented in a routine laboratory to meet them. Each laboratory
remains perfectly free to supplement these tools or to replace them by others that
they consider to be more efficient or more suitable.
Finally, the reader’s attention should be drawn to the fact that the tools presented
do not constitute an end in themselves and that their use, as well as the
interpretation of the results to which they lead, must always be subject to critical
analysis. It is only under these conditions that their relevance can be guaranteed,
and laboratories will be able to use them as tools to improve the quality of the
analyses they carry out.
3. General vocabulary
The definitions indicated below used in this document result from the normative
references given in the bibliography.
Analyte
Object of the analysis method
Blank
Test carried out in the absence of a matrix (reagent blank) or on a matrix which
does not contain the analyte (matrix blank).
Bias
Difference between the expected test results and an accepted reference value.
Uncertainty budget
The list of uncertainty sources and their associated standard uncertainties,
established in order to assess the compound standard uncertainty associated with a
measurement result.
OIV-MA-AS1-12 : R2005
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Gauging (of a measuring instrument)
Material positioning of each reference mark (or certain principal reference marks
only) of a measuring instrument according to the corresponding value of the
measurand.
NOTE "gauging" and "calibration" are not be confused
Repeatability conditions
Conditions where independent test results are obtained with the same method on
identical test items in the same laboratory by the same operator using the same
equipment within short intervals of time.
Reproducibility conditions (intralaboratory)
Conditions where independent test results are obtained with the same method on
identical test items in the same laboratory by the same or different operator(s)
using different gauges on different days.
Experimental standard deviation
For a series of n measurements of the same measurand, the quantity s
characterizing the dispersion of the results and given by the formula:
n
s
 ( xi  x )
2
i 1
n 1
xi being the result of the measurement ith and x the arithmetic mean of the n
results considered.
Repeatability standard deviation
Standard deviation of many repetitions obtained in a single laboratory by the same
operator on the same instrument, i.e. under repeatable conditions.
Internal reproducibility standard deviation (or total intralaboratory
variability)
Standard deviation of repetitions obtained in a single laboratory with the same
method, using several operators or instruments and, in particular, by taking
measurements on different dates, i.e. under reproducibility conditions.
Random error
Result of a measurement minus the mean that would result from an infinite number
of measurements of the same measurand carried out under reproducibility
conditions.
OIV-MA-AS1-12 : R2005
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Measurement error
Result of a measurement minus a true value of the measurand.
Systematic error
Mean error that would result from an infinite number of measurements of the same
measurand carried out under reproducibility conditions minus a true value of the
measurand.
NOTE Error is a highly theoretical concept in that it calls upon values that are not
accessible in practice, in particular the true values of measurands. On principle, the
error is unknown.
Mathematical expectation
For a series of n measurements of the same measurand, if n tends towards the
infinite, the mean x tends towards the expectation E(x).
n
lim
E ( x)  n 
 
 xi
i 1
n
Calibration
Series of operations establishing under specified conditions the relation between
the values of the quantity indicated by a measuring instrument or system, or the
values represented by a materialized measurement or a reference material, and the
corresponding values of the quantity measured by standards.
Intralaboratory evaluation of an analysis method
Action which consists in submitting an analysis method to an intralaboratory
statistical study, based on a standardized and/or recognized protocol,
demonstrating that within its scope, the analysis method meets pre-established
performance criteria.
Within the framework of this document, the evaluation of a method is based on an
intralaboratory study, which includes the comparison with a reference method.
Precision
Closeness of agreement between independent test results obtained under
prescribed conditions
NOTE 1
Precision depends only on the distribution of random errors and
does not have any relationship with the true or specified value.
NOTE 2
The measurement of precision is expressed on the basis of the
standard deviation of the test results.
NOTE 3
The expression "independent test results" refers to results obtained
such that they are not influenced by a previous result on the same or a similar test
OIV-MA-AS1-12 : R2005
8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
material. Quantitative measurements of precision are critically dependent upon the
prescribed conditions. Repeatability and reproducibility conditions are particular
sets of extreme conditions.
Quantity (measurable)
An attribute of a phenomenon, body or substance that may be distinguished
qualitatively and determined quantitatively.
Uncertainty of measurement
A parameter associated with the result of a measurement, which characterizes the
dispersion of the values that could reasonably be attributed to the measurand.
Standard uncertainty (u(xi))
Uncertainty of the result of a measurement expressed in the form of a standard
deviation.
Accuracy
Closeness of agreement between the mean value obtained starting from a broad
series of test results and an accepted reference value.
NOTE The measurement of accuracy is generally expressed in terms of bias.
Detection limit
Lowest amount of an analyte to be examined in a test material that can be detected
and regarded as different from the blank value (with a given probability), but not
necessarily quantified. In fact, two risks must be taken into account:
- the risk  of considering the substance is present in test material
when its quantity is null;
- the risk  of considering a substance is absent from a substance
when its quantity is not null.
Quantification limit
Lowest amount of an analyte to be examined in a test material that can be
quantitatively determined under the experimental conditions described in the
method with a defined variability (given coefficient of variation).
Linearity
The ability of a method of analysis, within a certain range, to provide an
instrumental response or results proportional to the quality of analyte to be
determined in the laboratory sample.
This proportionality is expressed by an a priori defined mathematical expression.
OIV-MA-AS1-12 : R2005
9
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The linearity limits are the experimental limits of concentrations between which a
linear calibration model can be applied with a known confidence level (generally
taken to be equal to 1%).
Test material
Material or substance to which a measuring can be applied with the analysis
method under consideration.
Reference material
Material or substance one or more of whose property values are sufficiently
homogeneous and well established to be used for the calibration of an apparatus,
the assessment of a measurement method, or for assigning values to materials.
Certified reference material
Reference material, accompanied by a certificate, one or more whose property
values are certified by a procedure which establishes its traceability to an accurate
realization of the unit in which the property values are expressed, and for which
each certified value is accompanied by an uncertainty at a stated level of
confidence.
Matrix
All the constituents of the test material other than the analyte.
Analysis method
Written procedure describing all the means and procedures required to carry out
the analysis of the analyte, i.e.: scope, principle and/or reactions, definitions,
reagents, apparatus, procedures, expression of results, precision, test report.
WARNING The expressions "titration method" and "determination method" are
sometimes used as synonyms for the expression "analysis method". These two
expressions should not be used in this way.
Quantitative analysis method
Analysis method making it possible to measure the analyte quantity present in the
laboratory test material.
Reference analysis method (Type I or Type II methods)
Method, which gives the accepted reference value for the quantity of the analyte to
be measured.
Non-classified alternative method of analysis
A routine analysis method used by the laboratory and not considered to be a
reference method.
OIV-MA-AS1-12 : R2005
10
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
NOTE An alternative method of analysis can consist in a simplified version of the
reference method.
Measurement
Set of operations having the object of determining a value of a quantity.
NOTE The operations can be carried out automatically.
Measurand
Particular quantity subject to measurement.
Mean
For a series of n measurements of the same measurand, mean value, given by the
formula:
n
x
 xi
i 1
n
xi being the result of the ith measurement.
Result of a measurement
Value assigned to a measurand, obtained by measurement
Sensitivity
Ratio between the variation of the information value of the analysis method and
the variation of the analyte quantity.
The variation of the analyte quantity is generally obtained by preparing various
standard solutions, or by adding the analyte to a matrix.
NOTE 1
Defining, by extension, the sensitivity of a method as its capacity
to detect small quantities should be avoided.
NOTE 2
A method is said to be “sensitive" if a low variation of the quantity
or analyte quantity incurs a significant variation in the information value.
Measurement signal
Quantity representing the measurand and is functionally linked to it.
Specificity
Property of an analysis method to respond exclusively to the determination of the
quantity of the analyte considered, with the guarantee that the measured signal
comes only from the analyte.
OIV-MA-AS1-12 : R2005
11
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Tolerance
Deviation from the reference value, as defined by the laboratory for a given level,
within which a measured value of a reference material can be accepted.
Value of a quantity
Magnitude of a particular quantity generally expressed as a unit of measurement
multiplied by a number.
True value of a quantity
Value compatible with the definition of a given particular quantity.
NOTE 1
NOTE 2
The value that would be obtained if the measurement was perfect
Any true value is by nature indeterminate
Accepted reference value
A value that serves as an agreed-upon reference for comparison and which is
derived as:
a)
a theoretical or established value, based on scientific principles;
b)
an assigned or certified value, based on experimental work of some
national or international organization;
c)
a consensus or certified value, based on collaborative experimental work
under the auspices of a scientific or engineering group;
Within the particular framework of this document, the accepted reference value (or
conventionally true value) of the test material is given by the arithmetic mean of
the values of measurements repeated as per the reference method.
Variance
Square of the standard deviation.
4. General principles
4.1 Methodology
When developing a new alternative method, the laboratory implements a protocol
that includes several steps. The first step, applied only once at the initial stage, or
on a regular basis, is the validation of the method. This step is followed by
permanent quality control. All the data collected during these two steps make it
possible to assess the quality of the method. The data collected during these two
OIV-MA-AS1-12 : R2005
12
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
steps are used to evaluate the measurement uncertainty. The latter, which is
regularly assessed, is an indicator of the quality of the results obtained by the
method under consideration.
Development or adoption of a
method
Step of initial validation
Uncertainty assessment
Setting-up and implementation
of the quality control system
All these steps are inter-connected and constitute a global approach that can be
used to assess and control measurement errors.
4.2 Definition of measurement error
Any measurement carried out using the method under study gives a result which is
inevitably associated with a measurement error, defined as being the difference
between the result obtained and the true value of the measurand. In practice, the
true value of the measurand is inaccessible and a value conventionally accepted
as such is used instead.
The measurement error includes two components:
Measurement error
True value = Analysis result + Systematic error + Random error
OIV-MA-AS1-12 : R2005
13
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
In practice, the systematic error results in a bias in relation to the true value, the
random error being all the errors associated with the application of the method.
These errors can be graphically represented in the following way:
Gauss distribution of the
results
Precision
Error
Systematic error
True value
Random error
Mean value of an
infinite number of
results
Result of an analysis
The validation and quality control tools are used to evaluate the systematic errors
and the random errors, and to monitor their changes over time.
5. Validating a method
5.1 Methodology
Implementing the validation comprises 3 steps, each with objectives. To meet
these objectives, the laboratory has validation tools. Sometimes there are many
tools for a given objective, and are suitable for various situations. It is up to the
laboratory to correctly choose the most suitable tools for the method to be
validated.
OIV-MA-AS1-12 : R2005
14
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Steps
Objectives
Tools for validation
Scope of
application
- To define the analyzable matrices
- To define the analyzable range
Systematic
error
or bias
Detection and quantification
limit
Robustness study
- Linear response in the scale of
analyzable values
- Specificity of the method
- Accuracy of the method
Linearity study
- Precision of the method
Repeatability study
Intralaboratory reproducibility
study
Specificity study
Comparison with a reference
method
Comparison with reference
materials
Interlaboratory comparison
Random error
5.2 Section one: Scope of method
5.2.1 Definition of analyzable matrices
The matrix comprises all constituents in the test material other than the analyte.
If these constituents are liable to influence the result of a measurement, the
laboratory should define the matrices on which the method is applicable.
For example, in oenology, the determination of certain parameters can be
influenced by the various possible matrices (wines, musts, sweet wines, etc.).
In case of doubt about a matrix effect, more in-depth studies can be carried out as
part of the specificity study.
5.2.2 Detection and quantification limit
This step is of course not applicable and not necessary for those methods whose
lower limit does not tend towards 0, such as alcoholic strength by volume in wines,
total acidity in wines, pH, etc.
OIV-MA-AS1-12 : R2005
15
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.2.2.1 Normative definition
The detection limit is the lowest amount of analyte that can be detected but not
necessarily quantified as an exact value. The detection limit is a parameter of limit
tests.
The quantification limit is the lowest quantity of the compound that can be
determined using the method.
5.2.2.2 Reference documents
-
NF V03-110 Standard, intralaboratory validation procedure for an
alternative method in relation to a reference method.
International compendium of analysis methods Ŕ OIV, Assessment of
the detection and quantification limit of an analysis method (Oeno
resolution 7/2000).
5.2.2.3 Application
In practice, the quantification limit is generally more relevant than the detection
limit, the latter being by convention 1/3 of the first.
There are several approaches for assessing the detection and quantification limits:
- Determination on blank
- Approach by the linearity study
- Graphic approach
These methods are suitable for various situations, but in every case they are
mathematical approaches giving results of informative value only. It seems crucial,
whenever possible, to introduce a check of the value obtained, whether by one of
these approaches or estimated empirically, using the checking protocol for a
predetermined quantification limit.
5.2.2.4 Procedure
5.2.2.4.1 Determination on blank
5.2.2.4.1.1 Scope
OIV-MA-AS1-12 : R2005
16
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
This method can be applied when the blank analysis gives results with a non-zero
standard deviation. The operator will judge the advisability of using reagent
blanks, or matrix blanks.
If the blank, for reasons related to uncontrolled signal preprocessing, is sometimes
not measurable or does not offer a recordable variation (standard deviation of 0),
the operation can be carried out on a very low concentration in analyte, close to the
blank.
5.2.2.4.1.2 Basic protocol and calculations
Carry out the analysis of n test materials assimilated to blanks, n being equal to or
higher than 10.
- Calculate the mean of the xi results obtained:
n
xblank 
 xi
i 1
n
- Calculate the standard deviation of the xi results obtained:
n
Sblank 
( xi  xblank )
2
i 1
n 1
- From these results the detection limit is conventionally defined by the formula:
Ld  xblank  (3.Sblank)
- From these results the quantification limit is conventionally defined by the
formula:
Lq xblank  (10.S blank )
Example: The table below gives some of the results obtained when assessing the
detection limit for the usual determination of free sulfur dioxide.
OIV-MA-AS1-12 : R2005
17
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Test material #
1
2
3
4
5
6
7
8
9
10
11
12
X
( mg/l)
0
1
0
1.5
0
1
0.5
0
0
0.5
0
0
The calculated values are as follows:
q = 12
Mblank = 0.375
Sblank = 0.528 mg/l
DL = 1.96 mg/l
QL = 5.65 mg/l
5.2.2.4.2 Approach by linearity study
5.2.2.4.2.1 Scope
This method can be applied in all cases, and is required when the analysis method
does not involve background noise. It uses the data calculated during the linearity
study.
NOTE This statistical approach may be biased and give pessimistic results when
linearity is calculated on a very wide range of values for reference
materials, and whose measurement results include variable standard
deviations. In such cases, a linearity study limited to a range of low values,
close to 0 and with a more homogeneous distribution will result in a more
relevant assessment.
OIV-MA-AS1-12 : R2005
18
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.2.2.4.2.2 Basic protocol and calculations
Use the results obtained during the linearity study which made it possible to
calculate the parameters of the calibration function y = a+ b.x
The data to be recovered from the linearity study are (see chapter 5.3.1. linearity
study):
- slope of the regression line:
n
b
 ( xi  M x )( yi  M y )
i 1
n
 ( xi  M x )
2
i 1
- residual standard deviation:
n
S res 
p
 ( yi, j  yˆ i, j )
2
i 1 j 1
pn  2
- standard deviation at the intercept point (to be calculated):
S a  S res


 1


 np



Mx
n
2
 p ( xi  M x )
i 1




2



The estimates of the detection limit DL and the quantification limit QL are
calculated using following formulae:
DL 
QL 
3  Sa
b
10  S a
b
OIV-MA-AS1-12 : R2005
Estimated detection limit
Estimated quantification limit
19
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Example: Estimatation of the detection and quantification limits in the
determination of sorbic acid by capillary electrophoresis, based on linearity data
acquired on a range from 1 to 20 mg.L-1.
X (ref)
Y1
Y2
Y3
Y4
1
2
3
4
5
10
15
20
1.9
2.4
4
5.3
5.3
11.6
16
19.7
0.8
2
2.8
4.5
5.3
10.88
15.2
20.4
0.5
2.5
3.5
4.7
5.2
12.1
15.5
19.5
1.5
2.1
4
4.5
5.3
10.5
16.1
20.1
Number of reference materials
n=8
Number of replicas
p=4
Straight line (y = a + b*x)
b = 0.9972
a = 0.51102
residual standard deviation:
Sres = 0.588
Standard deviation on the intercept
point
Sa = 0.1597
The estimated detection limit is
The estimated quantification limit is
OIV-MA-AS1-12 : R2005
DL = 0.48 mg.L-1
QL = 1.6 mg.L-1
20
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.2.2.4.3 Graphic approach based on the background noise of the recording
5.2.2.4.3.1 Scope
This approach can be applied to analysis methods that provide a graphic recording
(chromatography, etc.) with a background noise. The limits are estimated from a
study of the background noise.
5.2.2.4.3.2 Basic protocol and calculation
Record a certain number of reagent blanks, using 3 series of 3 injections separated
by several days.
Determine the following values:
 hmax the greatest variation in amplitude on the y-axis of the signal
observed between two acquisition points, excluding drift, at a distance
equal to twenty times the width at mid-height of the peak corresponding
to the analyte, centered over the retention time of the compound under
study.
 R, the quantity/signal response factor, expressed in height.
The detection limit DL, and the quantification limit QL are calculated according
to the following formulae:
DL = 3 hmax R
QL = 10 hmax R
5.2.2.4.4 Checking a predetermined quantification limit
This approach can be used to validate a quantification value obtained by statistical
or empirical approach.
5.2.2.4.4.1 Scope
This method can be used to check that a given quantification limit is a priori
acceptable. It is applicable when the laboratory can procure at least 10 test
materials with known quantities of analyte, at the level of the estimated
quantification limit.
In the case of methods with a specific signal, not sensitive to matrix effects, the
materials can be synthetic solutions whose reference value is obtained by
formulation.
OIV-MA-AS1-12 : R2005
21
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
In all other cases, wines (or musts) shall be used whose measurand value as
obtained by the reference method is equal to the limit to be studied. Of course, in
this case the quantification limit of the reference method must be lower than this
value.
5.2.2.4.4.2 Basic protocol and calculation
Analyze n independent test materials whose accepted value is equal to the
quantification limit to be checked; n must at least be equal to 10.
- Calculate the mean of n measurements:
n
 xi
x LQ 
i 1
n
- Calculate the standard deviation of n measurements:
 ( xi  xLQ )
n
S LQ 
2
i 1
n 1
with xi results of the measurement of the ith test material.
The two following conditions must be met:
a) the measured mean quantity xLQ must not be different from the predetermined
quantification limit QL:
If QL  xQl < 10 then quantification limit QL is considered to be valid.
S QL
n
NOTE 10 is a purely conventional value relating to the QL criterion.
b) the quantification limit must be other than 0:
If 5 sQL < QL then the quantification limit is other than 0.
A value of 5 corresponds to an approximate value for the spread of the standard
deviation, taking into account risk  and risk  to ensure that the QL is other than
0.
OIV-MA-AS1-12 : R2005
22
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
This is equivalent to checking that the coefficient of variation for QL is lower than
20%.
NOTE1
Remember that the detection limit is obtained by dividing the
quantification limit by 3.
NOTE2
A check should be made to ensure that the value of SLQ is not too
large (which would produce an artificially positive test), and effectively
corresponds to a reasonable standard deviation of the variability of the results for
the level under consideration. It is up to the laboratory to make this critical
evaluation of the value of SLQ.
Example: Checking the quantification limit of the
determination of malic acid by the enzymatic method.
Estimated quantification limit: 0.1 g.L-1
Wine
1
2
3
4
5
6
7
8
9
10
Values
0.1
0.1
0.09
0.1
0.09
0.08
0.08
0.09
0.09
0.08
Mean: 0.090
Standard deviation: 0.008
First condition:
LQ xQL
S
3.8710
The quantification
QL
n
limit of 0.1 is considered to be valid.
Second condition: 5.SLQ 0.040.1 The quantification
limit is considered to be significantly different from 0.
OIV-MA-AS1-12 : R2005
23
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.2.3 Robustness
5.2.3.1 Definition
Robustness is the capacity of a method to give close results in the presence of
slight changes in the experimental conditions likely to occur during the use of the
procedure.
5.2.3.2 Determination
If there is any doubt about the influence of the variation of operational parameters,
the laboratory can use the scientific application of experiment schedules, enabling
these critical operating parameters to be tested within the variation range likely to
occur under practical conditions. In practice, these tests are difficult to implement.
5.3 Section two: systematic error study
5.3.1 Linearity study
5.3.1.1 Normative definition
The linearity of a method is its ability (within a given range) to provide an
informative value or results proportional to the amount of analyte to be determined
in the test material.
5.3.1.2 Reference documents
-
NF V03-110 standard. Intralaboratory validation procedure of an
alternative method in relation to a reference method.
ISO 11095 Standard, linear calibration using reference materials.
ISO 8466-1 Standard, Water quality Ŕ Calibration and evaluation of
analytical methods and estimation of performance characteristics
5.3.1.3 Application
The linearity study can be used to define and validate a linear dynamic range.
This study is possible when the laboratory has stable reference materials whose
accepted values have been acquired with certainty (in theory these values should
have an uncertainty equal to 0). These could therefore be internal reference
materials titrated with calibrated material, wines or musts whose value is given by
OIV-MA-AS1-12 : R2005
24
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
the mean of at least 3 repetitions of the reference method, external reference
materials or certified external reference materials.
In the last case, and only in this case, this study also enables the traceability of the
method. The experiment schedule used here could then be considered as a
calibration.
In all cases, it is advisable to ensure that the matrix of the reference material is
compatible with the method.
Lastly, calculations must be made with the final result of the measurement and not
with the value of the signal.
Two approaches are proposed here:
-
An ISO 11095 type of approach, the principle of which consists in
comparing the residual error with the experimental error using a
Fischer's test. This approach is valid above all for relatively narrow
ranges (in which the measurand does not vary by more than a factor
10). In addition, under experimental conditions generating a low
reproducibility error, the test becomes excessively severe. On the
other hand, in the case of poor experimental conditions, the test will
easily be positive and will also lose its relevance. This approach
requires good homogeneity of the number of measurements over the
entire range studied.
-
An ISO 8466 type of approach, the principle of which consists in
comparing the residual error caused by the linear regression with the
residual error produced by a polynomial regression (of order 2 for
example) applied to the same data. If the polynomial model gives a
significantly lower residual error, a conclusion of nonlinearity could
be drawn. This approach is appropriate in particular when there is a
risk of high experimental dispersion at one end of the range. It is
therefore naturally well-suited to analysis methods for traces. There
is no need to work with a homogeneous number of measurements
over the whole range, and it is even recommended to increase the
number of measurements at the borders of the range.
5.3.1.4 ISO 11095-type approach
5.3.1.4.1 Basic protocol
It is advisable to use a number n of reference materials. The number must be
higher than 3, but there is no need, however, to exceed 10. The reference materials
OIV-MA-AS1-12 : R2005
25
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
should be measured p times, under reproducibility conditions, p shall be higher
than 3, a number of 5 being generally recommended. The accepted values for the
reference materials are to be regularly distributed over the studied range of values.
The number of measurements must be identical for all the reference materials.
NOTE It is essential that the reproducibility conditions use a maximum of
potential sources of variability, with the risk that the test shows non-linearity in an
excessive way.
The results are reported in a table presented as follows:
Reference
materials
1
...
i
...
n
Accepted
reference value
material
x1
...
xi
...
xn
Replica
1
y11
...
yi1
...
yn1
Measured values
Replica
...
j
...
y1j
...
...
...
...
...
yij
...
...
...
...
...
ynj
...
...
Replica
p
y1p
...
yip
...
ynp
5.3.1.4.2 Calculations and results
5.3.1.4.2.1 Defining the regression model
The model to be calculated and tested is as follows:
yij  a  b.xi   ij
where
yij
xi
b
a
is the jth replica of the ith reference material.
is the accepted value of the ith reference material.
is the slope of the regression line.
is the intercept point of the regression line.
ab.xi
represents the expectation of the measurement value of the ith
reference material.
ij
is the difference between yij and the expectation of the measurement value
of the ith reference material.
OIV-MA-AS1-12 : R2005
26
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.3.1.4.2.2 Estimating parameters
The parameters of the regression line are obtained using the following formulae:
- mean of p measurements of the ith reference material
yi 
1
p
p
 yij
j 1
- mean of all the measurements
1
My 
n
1
n
Mx 
- mean of all the accepted values of n reference materials
n
 xi
i 1
n
 yi
i 1
n
 ( xi  M x )( yi  M y )
- estimated slope b
b
i 1
n
 ( xi  M x )
2
i 1
- estimated intercept point a
a  M y b Mx
- regression value associated with the ith reference material
ŷi
yˆ i  a  b  xi
- residual eij
eij  yij  yˆ i
5.3.1.4.2.3 Charts
The results can be presented and analyzed in graphic form. Two types of charts are
used.
OIV-MA-AS1-12 : R2005
27
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
- The first type of graph is the representation of the values measured against the
accepted values of reference materials. The calculated overlap line is also plotted.
Overlap line
10,00
8, 00
s
e
lu
a
v6, 00
d
e
ru
s
a
e4, 00
M
y=x
yi^(régression)
repli que
replica
11
repli que2
replica
2
repli que3
replica
3
repli que4
replica
4
2, 00
0, 00
0, 00
2,00
4,00
6,00
8,00
10,00
Ac cepted values of the reference materials
- The second graph is the representation of the residual values against the
estimated values of the reference materials ( ŷ ) indicated by the overlap line.
The graph is a good indicator of the deviation in relation to the linearity
assumption: the linear dynamic range is valid if the residual values are fairly
distributed between the positive and negative values.
OIV-MA-AS1-12 : R2005
28
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Residual values in relation to adjusted values:
case of a non-linear method
0,6
0,5
0,4
Resi dual val ues
0,3
0,2
replica 1 1
replique
replica 2
replique2
0,1
replica 3
replique3
replique4
replica 4
0
0
2
4
6
8
10
12
-0,1
-0,2
-0,3
-0,4
Adjusted values y^
In case of doubt about the linearity of the regression, a Fischer-Snedecor test can
be carried out in order to test the assumption: "the linear dynamic range is not
valid", in addition to the graphic analysis.
5.3.1.4.2.4 Test of the linearity assumption
Several error values linked to calibration should be defined first of all: these can
be estimated using the data collected during the experiment. A statistical test is
then performed on the basis of these results, making it possible to test the
assumption of non-validity of the linear dynamic range: this is the FischerSnedecor test.
5.3.1.4.2.4.1 Definitions of errors linked to calibration
These errors are given as a standard deviation, resulting from the square root of the
ratio between a sum of squares and a degree of freedom.
Residual error
The residual error corresponds to the error between the measured values and the
value given by the regression line.
The sum of the squares of the residual error is as follows:
n
Qres 
p
 ( yij  yˆ i )
2
i 1 j 1
OIV-MA-AS1-12 : R2005
29
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The number of degrees of freedom is np-2.
The residual standard deviation is then estimated by the formula:
p
n
 ( yij  yˆ i )
2
i 1 j 1
S res 
np  2
Experimental error
The experimental error corresponds to the reproducibility standard deviation of the
experimentation.
The sum of the squares of the experimental error is as follows:
p
 ( yij  yi )
n
Qexp 
2
i 1 j 1
The number of degrees of freedom is np-n.
The experimental standard deviation (reproducibility) is then estimated by the
formula:
p
 ( yij  yi )
n
2
i 1 j 1
Sexp 
np  n
NOTE This quantity is sometimes also noted SR.
Adjustment error
The value of the adjustment error is the experimental error minus the residual
error.
The sum of the squares of the adjustment error is:
Qdef  Qres  Qexp
or
p
p
 ( yij  yˆ i )   ( yij  yi )
n
Qdef 
i 1 j 1
2
n
2
i 1 j 1
The number of degrees of freedom is n-2
OIV-MA-AS1-12 : R2005
30
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The standard deviation of the adjustment error is estimated by the formula:
Sdef 
Qres  Qexp
n2
or
p
p
 ( yij  yˆ i )   ( yij  yi )
n
S def 
2
i 1 j 1
n
2
i 1 j 1
n2
5.3.1.4.2.4.2 Fischer-Snedecor test
The ratio
Fobs 
S def
2
S exp
2
obeys the Fischer-Snedecor law with the degrees of
freedom n-2, np-n.
The calculated experimental value Fobs is compared with the limit value: F1-α (n2,np-n), extracted from the Snedecor law table. The value for α used in practice is
generally 5%.
If Fobs ≥ F1-α the assumption of the non-validity of the linear dynamic range is
accepted (with a risk of α error of 5%).
If Fobs < F1-α the assumption of the non-validity of the linear dynamic range is
rejected
Example: Linearity study for the determination of tartaric acid by capillary
electrophoresis. 9 reference materials are used. These are synthetic solutions of
tartaric acid, titrated by means of a scale traceable to standard masses.
Ref.
Ti (ref)
material
1
0.38
2
1.15
3
1.72
4
2.41
5
2.91
6
3.91
7
5.91
8
7.91
9
9.91
OIV-MA-AS1-12 : R2005
Y1
Y2
Y3
Y4
0.41
1.15
1.72
2.45
2.95
4.09
6.07
8.12
10.2
0.37
1.12
1.63
2.37
2.83
3.86
5.95
8.01
10
0.4
1.16
1.76
2.45
2.99
4.04
6.04
8.05
10.09
0.41
1.17
1.71
2.45
2.95
4.04
6.04
7.9
9.87
31
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Regression line
Line ( y = a + b*x)
b = 1.01565
a = - 0.00798
Errors related to calibration
Residual standard deviation Sres = 0.07161
Standard deviation of experimental reproducibility Sexp =
0.07536
Standard deviation of the adjustment error Sdef = 0.0548
Interpretation, Fischer-Snedecor test
Fobs = 0.53 < F1-α = 2.37
The assumption of the non-validity of the linear
dynamic range is rejected
5.3.1.5 ISO 8466-type approach
5.3.1.5.1 Basic protocol
It is advisable to use a number n of reference materials. The number must be
higher than 3, but there is no need, however, to exceed 10. The reference materials
should be measured several times, under reproducibility conditions. The number
of measurements may be small at the center of the range studied (minimum = 2)
and must be greater at both ends of the range, for which a minimum number of 4 is
generally recommended. The accepted values of reference materials must be
regularly distributed over the studied range of values.
NOTE It is vital that the reproducibility conditions use the maximum number of
potential sources of variability.
The results are reported in a table presented as follows:
Reference Accepted value of
Measured values
materials
the reference
Replica Replica Replica
...
material
1
2
j
1
x1
y11
y12
y1j
...
...
...
...
...
...
...
i
xi
yi1
yi2
...
...
...
...
...
...
N
xn
yn1
...
ynj
...
OIV-MA-AS1-12 : R2005
Replica
p
y1p
ynp
32
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.3.1.5.2 Calculations and results
5.3.1.5.2.1 Defining the linear regression model
Calculate the linear regression model using the calculations detailed above.
The residual error of the standard deviation for the linear model Sres can then be
calculated using the formula indicated in § 5.3.1.4.2.4.1
5.3.1.5.2.2 Defining the polynomial regression model
The calculation of the polynomial model of order 2 is given below
The aim is to determine the parameters of the polynomial regression model of
order 2 applicable to the data of the experiment schedule.
ya
2
x
x
b c
The purpose is to determine the parameters a, b and c. This determination can
generally be computerized using spreadsheets and statistics software.
The estimation formulae for these parameters are as follows:
a


N
i
 i



4
N
i


2



3
2
2
 xi y  xi   xi    xi  N  xi yi  xi  yi   xi   xi yi  xi   yi  xi 
i
i
i
i
i
i
 i
 i

 i   i  i
2
2
2
4
b
2






3
3
2
2
3 
 x  xi   xi    xi  N  xi  xi  xi   xi   xi  xi   x i2 
i
i
i
i
i
 i
 i
 i   i  i
 i  

2



2
2
 xi  N  xi yi   xi yi    xi yi  N  xi  xi  xi   xi   yi  xi   x i yi  xi 
i

i


4
N
i


i
i
2

i
2

3
i
2
i
i


i


2
i


3
i
i
i
i
2




2 
3
3
2
3 
 x  xi   xi    xi  N  xi  xi  xi   x  xi  xi   x i2 
i
i
i
i
i
i
 i
 i   i  i
 i  
OIV-MA-AS1-12 : R2005
33
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
c



i


2
 2 



y
y
y
y
y
 xi   xi  i  x i  xi
 xi   xi  i  xi  xi  xi  xi  xi  x i  
i
i
i
i
i
i
i
i
i
i
i
 i
 i
 i  
4


i

2


3 
3
2


i

2
2

3
2





3
3
2
2
3 
 xi  N  xi   xi    xi  N  xi  xi  xi   xi   xi  xi   x i2 
i
i
i
i
 i
 i
 i
 i   i  i
 i  


4
2
Once the model has been established, the following values are to be calculated:
- regression value associated with the ith reference material
yˆ i  a
-
x
2
yˆ i
x
b c
eij  yij  yˆi
residual eij
Residual standard deviation of the polynomial model
 ( yij  yˆ i )
n
S res 
p
2
i 1 j 1
np  2
5.3.1.5.2.3 Comparing residual standard deviations
Calculation of
DS
2
 ( N  2)
S res  ( N  3) S res
2
2
Then
2
PG  DS
2
S res
The value PG is compared with the limit value F1-α given by the Fischer-Snedecor
table for a confidence level 1-α and a degree of freedom 1 and (N-3).
NOTE In general the α risk used is 5%. In some cases the test may be optimistic
and a risk of 10% will prove more realistic.
If PG ≤ F1-α: the nonlinear calibration function does not result in an improved
adjustment; for example, the calibration function is linear.
If PG > F1-α: the work scope must be as narrow as possible to obtain a linear
calibration function: otherwise, the information values from the analyzed samples
must be evaluated using a nonlinear calibration function.
OIV-MA-AS1-12 : R2005
34
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Example: Theoretical case.
Ti (ref)
Y1
Y2
Y3
Y4
1
35
22.6
19.6
21.6
2
62
49.6
49.8
53
3
90
105.2
103.5
4
130
149
149.8
5
205
203.1
202.5
197.3
6
330
297.5
298.6
307.1
18.4
294.2
Linear model and polynomial model, method: theoretical case
350
300
Mea
n of
me
ase
250
ure
d
val
ues
200
y=x
yi^(lin reg)
yi'^(poly reg)
150
yi (mean)
100
50
0
0
50
100
150
200
250
300
350
Accepted values for reference materials
Linear regression
y = 1.48.x Ŕ 0.0015
Sres = 13.625
Polynomial regression
y = - 0.0015x² + 1.485x Ŕ 27.2701
S'res = 7.407
Fischer's test
PG = 10.534 > F(5%) = 10.128
PG>F the linear calibration function cannot be retained
OIV-MA-AS1-12 : R2005
35
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.3.2 Specificity
5.3.2.1 Normative definition
The specificity of a method is its ability to measure only the compound being
searched for.
5.3.2.2 Application
In case of doubt about the specificity of the tested method, the laboratory can use
experiment schedules designed to check its specificity. Two types of
complementary experiments are proposed here that can be used in a large number
of cases encountered in the field of oenology.
- The first test is the standard addition test. It can be used to check that
the method measures all the analyte.
- The second test can be used to check the influence of other compounds
on the result of the measurement.
5.3.2.3 Procedures
5.3.2.3.1 Standard addition test
5.3.2.3.1.1 Scope
This test can be used to check that the method measures all the analyte.
The experiment schedule is based on standard additions of the compound being
searched for. It can only be applied to methods that are not sensitive to matrix
effects.
5.3.2.3.1.2 Basic protocol
This consists in finding a significant degree of added quantities on test materials
analyzed before and after the additions.
Carry out variable standard additions on n test materials. The initial concentration
in analyte of test materials, and the standard additions are selected in order to
cover the scope of the method. These test materials must consist of the types of
matrices called for routine analysis. It is advised to use at least 10 test materials.
The results are reported in a table presented as follows:
OIV-MA-AS1-12 : R2005
36
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Test
material
1
...
i
...
n
NOTE 1
NOTE 2
NOTE 3
Quantity
before
addition
(x)
x1
...
xi
...
Xn
Quantity
added
(v)
Quantity
after addition
(w)
Quantity
found
(r)
v1
...
vi
...
Vn
w1
...
wi
...
wn
r1 = w1 Ŕ x1
...
ri = wi – xi
...
rp = wn – xn
An addition is made with a pure standard solution. It is advised to
perform an addition of the same order as the quantity of the test
material on which it is carried out. This is why the most concentrated
test materials must be diluted to remain within the scope of the
method.
It is advised to prepare the additions using independent standard
solutions, in order to avoid any systematic error.
The quality of values x and w can be improved by using several
repetitions.
5.3.2.3.1.3 Calculations and results
The principle of the measurement of specificity consists in studying the regression
line r = a + b.v and checking that slope b is equivalent to 1 and that intercept point
a is equivalent to 0.
5.3.2.3.1.3.1 Study of the regression line r = a + b.v
The parameters of the regression line are obtained using the following formulae:
n
- mean of the added quantities v
v
 vi
i 1
n
n
- mean of the quantities found r
OIV-MA-AS1-12 : R2005
r
 ri
i 1
n
37
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
n
- estimated slope b
b
 (vi  v)(ri  r )
i 1
n
2
 (vi v)
i 1
- estimated intercept point a
a  r  b.v
- regression value associated with the ith reference material
ŷi
rˆi  a  b  vi
n
- residual standard deviation
- standard deviation on the slope
- standard deviation on the intercept point
S res 
 ri  rˆi 2
i 1
n2



1

n

(v i  v ) 2 

i 1

S b  S res








S a  S res


1
 
n






2

v

n

(v i  v ) 2 

i 1


5.3.2.3.1.3.2 Analysis of the results
The purpose is to conclude on the absence of any interference and on an
acceptable specificity. This is true if the overlap line r = a + bv is equivalent to the
line y = x.
OIV-MA-AS1-12 : R2005
38
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
To do so, two tests are carried out:
- Test of the assumption that slope b of the overlap line is equal to 1.
- Test of the assumption that intercept point a is equal to 0.
These assumptions are tested using a Student test, generally associated with a risk
of error of 1%. A risk of 5% can prove more realistic in some cases.
Let Tcritical, bilateral[dof; 1%] be a Student bilateral variable associated with a risk of
error of 1% for a number of degrees of freedom (dof).
Step 1: calculations
Calculation of the comparison criterion on the slope at 1
Tobs
b1
Sb
Calculation of the comparison criterion on the intercept point at 0
T'obs
a
Sa
Calculation of the Student critical value: Tcritical, bilateral[ p-2; 1%]
Step 2: interpretation
 If Tobs is lower than Tcritical, then the slope of the regression line is
equivalent to 1
 If T’obs is lower than Tcritical, then the intercept point of the regression
line is equivalent to 0.
 If both conditions are true, then the overlap line is equivalent = y = x,
and the method is deemed to be specific.
Based on these results, a mean overlap rate can be calculated to
NOTE 1
quantify the specificity. In no case should it be used to "correct" the results. This is
because if a significant bias is detected, the alternative method cannot be validated
in relation to an efficiency rate of 100%.
NOTE 2
Since the principle of the test consists in calculating a straight line,
at least three levels of addition have to be taken, and their value must be correctly
chosen in order to obtain an optimum distribution of the points.
OIV-MA-AS1-12 : R2005
39
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.3.2.3.1.3.3 Overlap line graphics
Example of specificity
The specificity is not accepted
The specificity is accepted
10
9
Y = X
8
Estimated relation between the added content and the content found
9
8
Y: Content found
Y: Content found
7
6
5
4
3
Estimated relation between
the added content and the
content found
2
7
6
5
4
Y = X
3
2
1
1
0
0
0
0
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
X : Content found
X : Added content found
5.3.2.3.2 Study of the influence of other compounds on the measurement result
5.3.2.3.2.1 Scope
If the laboratory suspects the interaction of compounds other than the analyte, an
experiment schedule can be set up to test the influence of various compounds. The
experiment schedule proposed here enables a search for the influence of
compounds defined a priori: thanks to its knowledge of the analytical process and
its know-how, the laboratory should be able to define a certain number of
compounds liable to be present in the wine and to influence the analytical result.
5.3.2.3.2.2 Basic protocol and calculations
Analyze n wines in duplicate, before and after the addition of the compound
suspected of having an influence on the analytical result; n must at least be equal
to 10.
The mean values Mxi of the 2 measurements xi and x'i made before the addition
shall be calculated first, then the mean values Myi of the 2 measurements yi and y'i
made after the addition, and finally the difference di between the values Mxi and
Myi.
The results of the experiment can be reported as indicated in the following table:
OIV-MA-AS1-12 : R2005
40
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Samples
x: Before
addition
Rep1
Rep2
Means
y: After addition
Difference
Rep1
Rep2
x
y
1
x1
x’1
y1
y’1
Mx1
My1
...
...
...
...
...
...
...
i
xi
x’i
yi
y’i
Mxi
Myi
...
...
...
...
...
...
...
n
xn
x’n
yn
y’n
Mxn
Myn
d
d1 = Mx1My1
...
di = MxiMyi
...
dn = MxnMyn
The mean of the results before addition Mx
1
Mx 
n
n
 Mxi
i 1
The mean of the results after addition My
My 
1
n
n
 Myi
i 1
Calculate the mean of the differences Md
n
Md 
 ni  My  Mx
d
i 1
Calculate the standard deviation of the differences Sd
n
Sd 
(di  M d )
2
i 1
n 1
Calculate the Z-score
Z score 
Md
Sd
5.3.2.3.2.3 Interpretation
OIV-MA-AS1-12 : R2005
41
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
 If the Zscore is ≤ 2, the added compound can be considered to have a
negligible influence on the result of analysis with a risk of 5%.
 If the Zscore is ≥ 2, the added compound can be considered to influence the
result of analysis with a risk of 5%.
NOTE Interpreting the Zscore is possible given the assumption that the variations
obey a normal law with a 95% confidence rate.
Example: Study of the interaction of compounds liable to be present in
the samples, on the determination of fructose glucose in wines by Fourier
transform infrared spectrophotometry (FTIR).
Before
addition
+ 250 mg.L- + 1 g. L-1
1
potassium salicylic
sorbate
acid
vin
rep1 rep2 rep1 rep2 rep1 rep2
1
2
3
4
5
6
7
8
9
10
6.2 6.2 6.5 6.3 5.3 5.5
1.2 1.2 1.3 1.2 0.5 0.6
0.5 0.6 0.5 0.5 0.2 0.3
4.3 4.2 4.1 4.3 3.8 3.9
12.5 12.6 12.5 12.7 11.5 11.4
5.3 5.3 5.4 5.3 4.2 4.3
2.5 2.5 2.6 2.5 1.5 1.4
1.2 1.3 1.2 1.1 0.5 0.4
0.8 0.8 0.9 0.8 0.2 0.3
0.6 0.6 0.5 0.6 0.1
0
Differences
sorbate
diff
0.2
0.05
-0.05
-0.05
0.05
0.05
0.05
-0.1
0.05
-0.05
Potassium sorbate
Md =
Sd =
Zscore =
0.02
0.086
0.23 <2
Salicylic acid
Md =
Sd =
Zscore =
-0.725
0.282
2.57 >2
salicylic
diff
-0.8
-0.65
-0.3
-0.4
-1.1
-1.05
-1.05
-0.8
-0.55
-0.55
In conclusion, it can be stated that potassium sorbate does
not influence the determination of fructose glucose by the
FTIR gauging studied here. On the other hand, salicylic
acid has an influence, and care should be taken to avoid
OIV-MA-AS1-12 : R2005
42
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
samples containing salicylic acid, in order to remain
within the scope of validity for the gauging under study.
5.3.3 Study of method accuracy
5.3.3.1 Presentation of the step
5.3.3.1.1 Definition
Correlation between the mean value obtained with a large series of test results and
an accepted reference value.
5.3.3.1.2 General principles
When the reference value is output by a certified system, the accuracy study can be
regarded a traceability link. This applies to two specific cases in particular:
- Traceability to certified reference materials: in this case, the accuracy
study can be undertaken jointly with the linearity and calibration study, using the
experiment schedule described for that study.
- Traceability to a certified interlaboratory comparison analysis chain.
The other cases, i.e. which use references that are not based on certified systems,
are the most widespread in routine oenological laboratories. These involve
comparisons:
- Comparison with a reference method
- Comparison with the results of an uncertified interlaboratory comparison
analysis chain.
- Comparison with internal reference materials, or with external
uncertified reference materials.
5.3.3.1.3 Reference documents
-
NF V03-110 Standard. intralaboratory validation procedure for an
alternative method in relation to a reference method.
NF V03-115 Standard, Guide for the use of reference materials.
ISO 11095 Standard, linear calibration using reference materials.
ISO 8466-1 Standard. Water quality Ŕ Calibration and evaluation of
analytical methods and estimation of performance characteristics
OIV-MA-AS1-12 : R2005
43
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
-
ISO 57025 Standard, Exactitude of results and methods of
measurement
5.3.3.2 Comparison of the alternative method with the OIV reference method
5.3.3.2.1 Scope
This method can be applied if the laboratory uses the OIV reference method, or a
traced, validated method, whose performance quality is known and meets the
requirements of the laboratory’s customers.
To study the comparative accuracy of the two methods, it is advisable first of all to
ensure the quality of the repeatability of the method to be validated, and to
compare it with the reference method. The method for carrying out the
repeatability comparison is described in the chapter on repeatability.
5.3.3.2.2 Accuracy of the alternative method compared with the reference method
5.3.3.2.2.1 Definition
Accuracy is defined as the closeness of agreement between the values obtained by
the reference method and that obtained by the alternative method, independent of
the errors of precision of the two methods.
5.3.3.2.2.2 Scope
The accuracy of the alternative method in relation to the reference method is
established for a field of application in which the repeatabilities of the two
methods are constant.
In practice, it is therefore often advisable to divide the analyzable range of values
into several sections or "range levels" (2 to 5), in which we may reasonably
consider that the repeatabilities of the methods are comparable to a constant.
5.3.3.2.2.3 Basic protocol and calculations
In each range level, accuracy is based on a series of n test materials with
concentration values in analyte covering the range level in question. A minimum
number of 10 test materials is required to obtain significant results.
Each test material is to be analyzed in duplicate by the two methods under
repeatable conditions.
OIV-MA-AS1-12 : R2005
44
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
A calculation is to be made of the mean values Mxi of the 2 measurements xi et x’i
made using the alternative method and the mean values Myi of the 2 measurements
yi et y’i made using the reference method, then the difference di is to be calculated
between the values Mxi and Myi.
The results of the experiment can be reported as in the following table:
Test
material
x: Alternative
method
Rep1 Rep2
y: Reference
method
Rep1 Rep2
Means
Difference
x
y
1
x1
x’1
y1
y’1
Mx1
My1
...
...
...
...
...
...
...
i
xi
x’i
yi
y’i
Mxi
Myi
...
...
...
...
...
...
...
n
xn
x’n
yn
y’n
Mxn
Myn
d
d1 = Mx1 My1
...
di = Mxi Myi
...
dn = Mxn Myn
The following calculations are to be made
- The mean of the results for the alternative method Mx
1
Mx 
n
n
 Mxi
i 1
- The mean of the results for the reference method My
1
My 
n
n
 Myi
i 1
- Calculate the mean of the differences Md
n
Md 
 ni
d
 Mx  My
i 1
OIV-MA-AS1-12 : R2005
45
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
- Calculate the standard deviation of the differences Sd
n
Sd 
(di  M d )
2
i 1
n 1
- Calculate the Zscore
Z score 
Md
Sd
5.3.3.2.2.4 Interpretation
- If the Zscore is lower than or equal to 2.0, it can be concluded that the accuracy of
one method in relation to the other is satisfactory, in the range level under
consideration, with a risk of error α = 5%.
- If the Zscore is higher than 2.0, it can be concluded that the alternative method is
not accurate in relation to the reference method, in the range level under
consideration, with a risk of error α = 5%.
NOTE Interpreting the Zscore is possible given the assumption that the variations
obey a normal law with a 95% confidence rate.
Example: Study of the accuracy of FTIR gauging to determine glucose
and fructose in relation to the enzymatic method. The first range level
covers the scale from 0 to 5 g.L-1 and the second range level covers a scale
from 5 to 20 g.L-1.
Wine
1
2
3
4
5
6
7
8
9
10
11
12
FTIR 1
0
0.2
0.6
0.7
1.2
1.3
2.1
2.4
2.8
3.5
4.4
4.8
OIV-MA-AS1-12 : R2005
IRTF2
0.3
0.3
0.9
1
1.6
1.4
2
0
2.5
4.2
4.1
5.4
Enz 1
0.3
0.1
0.0
0.8
1.1
1.3
1.9
1.1
2.0
3.7
4.1
5.5
Enz 2
0.2
0.1
0.0
0.7
1.3
1.3
2.1
1.2
2.6
3.8
4.4
5.0
di
-0.1
0.2
0.7
0.1
0.2
0.0
0.0
0.1
0.3
0.1
0.0
-0.2
46
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Md
Sd
Zscore
Wine
1
2
3
4
5
6
7
8
9
10
11
12
Md =
Sd =
Zscore =
0.13
0.23
0.55 < 2
FTIR 1
5.1
5.3
7.7
8.6
9.8
9.9
11.5
11.9
12.4
16
17.7
20.5
IRTF2
5.4
5.7
7.6
8.6
9.9
9.8
11.9
12.1
12.5
15.8
18.1
20.1
Enz 1
5.1
5.3
7.2
8.3
9.1
9.8
13.3
11.2
11.4
15.1
17.9
20.0
Enz 2
5.1
6.0
7.0
8.5
9.3
10.2
13.0
11.4
12.1
15.7
18.3
19.1
di
0.1
-0.2
0.6
0.2
0.6
-0.1
-1.4
0.7
0.7
0.5
-0.2
0.7
0.19
0.63
0.30 < 2
For the two range levels, the Zscore is lower than 2. The
FTIR gauging for the determination of fructose glucose
studied here, can be considered accurate in relation to the
enzymatic method.
5.3.3.3 Comparison by interlaboratory tests
5.3.3.3.1 Scope
Interlaboratory tests are of two types:
1. Collaborative studies relate to a single method. These studies are
carried out for the initial validation of a new method, mainly in order to
define the standard deviation of interlaboratory reproducibility
SRinter(method). The mean m could also be given.
OIV-MA-AS1-12 : R2005
47
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
2. Interlaboratory comparison analysis chains, or aptitude tests. These
tests are carried out for the validation of a method adopted by the
laboratory, and the routine quality control (see § 5.3.3.3). The resulting
value is the interlaboratory mean m, as well as the standard
interlaboratory reproducibility and intermethod deviation SRinter.
By participating in an analysis chain, or in a collaborative study, the laboratory can
exploit the results in order to study the accuracy of a method, in order to ensure its
validation first of all, and its routine quality control.
If the interlaboratory tests are carried out within the framework of a certified
organization, this comparison work can be used for method traceability.
5.3.3.3.2 Basic protocol and calculations
To obtain a sufficient comparison, it is recommended to use a minimum number of
5 test materials over the period.
For each test material, two results are provided:
- The mean of all the laboratories with significant results
- The standard deviation for interlaboratory reproducibility
m
SR-inter
The test materials are analyzed with p replicas by the laboratory, these replicas
being carried out under repeatable conditions. p must at least be equal to 2.
In addition, the laboratory must be able to check that the intralaboratory variability
(intralaboratory reproducibility) is lower than the interlaboratory variability
(interlaboratory reproducibility) given by the analysis chain.
For each test material, the laboratory calculates the Zscore, given by the following
formula:
Z score 
OIV-MA-AS1-12 : R2005
mlab  m
S R int er
48
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The results can be reported as indicated in the following table:
Test
Rep1 ... Rep j ... Rep p
material
Lab mean
Chain
mean
Standard
deviation
m1
SR-inter(1)
...
...
mi
SR-inter(i)
...
...
mn
SR-inte(n)
Zscore
p
x
1j
j 1
1
x11
...
x1j
...
x1p
mlab1 
...
...
...
...
...
...
...
p
Z score1 
mlab1 m1
SR inter(1)
...
p
x
ij
i
xi1
...
xij
...
xip
mlabi 
...
...
...
...
...
...
...
j 1
p
Zscorei 
mlabi mi
SR inter(i)
...
p
n
xn1
...
xnj
...
xnp
xnj
mlabn  j 1
p
Zscoren 
mlabn mn
SR inter(n)
5.3.3.3.3 Interpretation
If all the Zscore results are lower than 2, the results of the method being studied can
considered identical to those obtained by the laboratories having produced
significant results.
NOTE Interpreting the Zscore is possible given the assumption that the variations
obey a normal law with a 95% confidence rate.
Example: An interlaboratory analysis chain outputs the following
results for the free sulfur dioxide parameter, on two samples.
Samples
x1
x2
x3
x4
Lab
mean
Chain
mean
Standard
deviation
Zscore
1
34
34
33
34
33.75
32
6
0.29 <2
2
26
27
26
26
26.25
24
4
0.56 <2
It can be concluded that on these two samples, the comparison with
the analysis chain is satisfactory.
OIV-MA-AS1-12 : R2005
49
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.3.3.4 Comparison with reference materials
5.3.3.4.1 Scope
In situations where there is no reference method (or any other method) for a given
parameter, and the parameter is not processed by the analysis chains, the only
remaining possibility is comparison of the results of the method to be validated
with accepted internal or external material reference values.
The reference materials, for example, could be synthetic solutions established with
class-A glassware, and/or calibrated metrology apparatus.
In the case of certified reference materials, the comparison constitutes the
traceability value, and can be carried out at the same time as the gauging and
linearity study.
5.3.3.4.2 Basic protocol and calculations
It is advisable to have n reference materials for a given range level, in which it can
be reasonably estimated that repeatability is comparable to a constant; n must at
least be equal to 10.
Analyze in duplicate each reference material.
Calculate the mean values Mxi for the 2 measurements xi and x’i carried out using
the alternative method.
Define Ti the accepted value for the ith reference material.
The results can be reported as indicated in the following table:
Reference
material
1
...
i
...
n
x: Alternative method
Rep1
Rep2
Mean x
x1
x’1
xi
x’i
xn
x’n
OIV-MA-AS1-12 : R2005
Mx1
...
Mxi
...
Mxn
T: Accepted value
of the reference
material
T1
...
Ti
...
Tn
Difference
d
d1 = Mx1-T1
...
di = Mxi-Ti
...
dn = Mxn-Tn
50
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The mean of the results of the alternative method Mx
Mx 
n
 Mxi
1
n
i 1
The mean of the accepted values of reference materials MT
MT 
1
n
n
 Ti
i 1
Calculate the mean of the differences Md
n
 ni  Mx  M T
Md 
d
i 1
Calculate the standard deviation of the differences Sd
n
Sd 
 (d i  M d )
2
i 1
n 1
Calculate the Z-score
Z score 
Md
Sd
5.3.3.4.3 Interpretation
- If the Zscore is lower than or equal to 2.0, it can be concluded that the accuracy of
the alternative method in relation to the accepted values for the reference material
is good on the range level under consideration.
- If Zscore is higher than 2.0, it can be concluded that the alternative method is not
accurate in relation to the accepted values for the reference materials in the range
level under consideration.
NOTE Interpreting the Zscore is possible given the assumption that the variations
obey a normal law with a 95% confidence rate.
OIV-MA-AS1-12 : R2005
51
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Example: There is no reference method to compare the results of the
analysis of Ethyl-4 Phenol (4-EP) by Gas chromatography coupled with
mass spectrometry (GC-MS). The results are compared with the accepted
values for reference materials, consisting of synthetic solutions
formulated by traced equipment.
Test apparatus
1
2
3
4
5
6
7
8
9
10
Ti (ref)
4.62
12.3
24.6
46.2
77
92.4
123.2
246.4
385
462
Y1
6.2
15.1
24.5
48.2
80.72
97.6
126.6
254.1
375.8
467.5
Y2
6.56
10.94
18
52.95
81.36
89
129.9
250.9
366.9
454.5
Y3
4.9
12.3
25.7
46.8
83.2
94.5
119.6
243.9
380.4
433.3
Y4
5.7
11.6
27.8
35
74.5
99.5
126.9
240.4
386.9
457.3
My
di
5.8
12.5
24.0
45.7
79.9
95.2
125.8
247.3
377.5
453.2
1.2
0.2
-0.6
-0.5
2.9
2.8
2.6
0.9
-7.5
-8.9
Md = -0.7
Sd = . 4.16
Zscore = 0.16
Given these results, the values obtained by the analysis
method for 4-EP by GC-MS can be considered accurate
compared with the accepted values of reference
materials.
5.4 Section three: random error study
5.4.1 General principle
Random error is approximated using precision studies. Precision is calculated used
a methodology that can be applied under various experimental conditions, ranging
between those of repeatability, and those of reproducibility, which constitute the
extreme conditions of its measurement.
The precision study is one of the essential items in the study of the uncertainty of
measurement.
OIV-MA-AS1-12 : R2005
52
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.4.2 Reference documents
-
ISO 5725 Standard, Exactitude of results and methods of
measurement
NF V03-110 Standard, Intralaboratory validation procedure for an
alternative method in relation to a reference method.
5.4.3 Precision of the method
5.4.3.1 Definition
Closeness of agreement between independent test results obtained under
prescribed conditions.
NOTE 1
Precision depends only on the distribution of the random errors
and has no relation with the true or specified value.
NOTE 2
Expressing the measurement of precision is based on the standard
deviation of the test results.
NOTE 3
The term "independent test results" refers to results obtained such
that they are not influenced by a previous result on the same or similar test
material. Quantitative measurements of precision are critically dependent on the
prescribed conditions. Repeatability and reproducibility conditions are particular
sets of extreme conditions.
In practice, precision refers to all the experimental conditions ranging between the
conditions of repeatability and those of reproducibility.
5.4.3.2 Scope
The protocols and calculations are detailed below, from the general theoretical
case to the specific cases of repeatability and reproducibility. This exhaustive
approach should make it possible to apply the precision study in most laboratory
situations.
The precision study can be applied a priori without difficulty to every quantitative
method.
In many cases, precision is not constant throughout the validity range for the
method. In this case, it is advisable to define several sections or "range levels", in
which we may reasonably consider that the precision is comparable to a constant.
The calculation of precision is to be reiterated for each range level.
OIV-MA-AS1-12 : R2005
53
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
5.4.3.3 General theoretical case
5.4.3.3.1 Basic protocol and calculations
5.4.3.3.1.1 Calculations with several test materials
n test materials are analyzed over a relatively long period of time with several
replicas, pi being the number of replicas for the ith test material. The properties of
the test materials must maintain constant throughout the period in question.
For each replica, the measurement can be made with K repetitions, (we do not take
into account the case here where the number of repetitions K can vary from one
test material to the other, which would complicate the calculations even more).
The total number of replicas must be higher than 10, distributed over all the test
materials.
The results can be reported as indicated in the following table, (case in which K =
2)
Replicas
1
Test materials.
1
x11 x’11
...
i
xi1 x’i1
...
n
xn1 x’n1
...
j
p1
pi
pn
...
...
x1j
x’1j
x1p1
x’1p1
...
...
xij
x’ij
...
...
xipi x’ipi
...
...
xnj
x’nj
...
...
...
...
xnpn
x’npn
In this situation, the standard deviation of total variability (or standard deviation of
precision Sv) is given by the general expression:
1
S v  Var( xij )  (1  )Var(répet)
k
where:
Var( xij )
variance of the mean of repeated replicas of all test materials.
Var(répet)
variance of the repeatability of all the repetitions.
OIV-MA-AS1-12 : R2005
54
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
- If the test materials were analyzed in duplicate with each replica (K = 2), the
expression becomes:
S v  Var( xij ) 
Var(repeat)
2
- When only one measurement of the test material has been carried out with
each replica (K = 1), the variance of repeatability is null, the expression
becomes:
Sv  Var( xij )
- Calculation of Var(xij)
The mean of the two replicas xij and x’ij is:
xij 
xij  x'ij
2
For each test material, the mean of n replicas is calculated:
pi
 xij
Mxi 
j 1
pi
The number of different measurements n is the sum of pi
n
N
 pi
i 1
OIV-MA-AS1-12 : R2005
55
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The variance Var(xij) is then given by the following equation
( xij  M xi )
n
Var( xij ) 
pi
2
i 1 j 1
N n
NOTE This variance can also be calculated using the variances of variability of
each test material: Vari (xj). The following relation is then used (it is strictly
equivalent to the previous one):
n
Var( xij ) 
 ( pi  1).Vari ( x j )
i 1
N n
- Calculation of Var (repeat )
The variance of repeatability is calculated as a conventional repeatability equation
with n test materials in duplicate. According to the calculation of repeatability
discussed in the section entitled "repeatability", for K = 2 the variance of
repeatability is:
p
ni
 wij
Var(repeat) 
i 1 j 1
2N
2
where
wij  xij  x'ij
Precision v is calculated according to the formula:
v  2 2.S v  2.8.S v
The value of precision v means that in 95% of the cases, the difference between
two values obtained by the method, under the conditions defined, will be lower
than or equal to v.
NOTE 1
The use and interpretation of these results is based on the
assumption that the variations obey a normal law with a 95% confidence rate.
NOTE 2
One can also define a precision of 99% with v2.58 2.Sv 3.65.Sv
5.4.3.3.1.2 Calculations with 1 test material
OIV-MA-AS1-12 : R2005
56
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
In this situation, the calculations are simpler. It is advisable to carry out p
measurement replicas of the test material, if necessary with a repetition of the
measurement on each replica. p must at least be equal to 10.
In the following calculations, the measurement is considered to be carried out in
duplicate with each replica.
- The variance Var(xij) is then given by the following equation:
p
Var( xij ) 
( xi  M x )
2
i 1
p 1
where:
xi
p
Mx
is the mean of the two repetitions of replica i
is the number of replicas
is the mean of all the replicas
- The variance Var (repeat ) is then given by the following equation:
p
Var(repeat) 
 wi2
i 1
2p
where wi : difference between the two repetitions of replica i
5.4.3.4 Repeatability
5.4.3.4.1 Definitions
Repeatability is the closeness of agreement between mutually-independent analysis
results obtained with the method in question on the same wine, in the same
laboratory, with the same operator using the same equipment, within a short period
of time.
These experimental conditions will be called conditions of repeatability.
The value of repeatability r is the value below which the absolute difference
between two results of the same analysis is considered to be located, obtained
under the conditions of repeatability defined above, with a confidence level of
95%.
OIV-MA-AS1-12 : R2005
57
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The repeatability standard deviation Sr is the standard deviation for the results
obtained under the conditions of repeatability. It is a parameter of the dispersion of
the results, obtained under conditions of repeatability.
5.4.3.4.2 Scope
A priori, the repeatability study can be applied without difficulty to every
quantitative method, insofar as the repeatability conditions can be observed.
In many cases, repeatability is not constant throughout the range of validity of the
method. It is therefore advisable to define several sections or "range levels", in
which we may reasonably consider that the repeatability is comparable to a
constant. The repeatability calculation is then to be reiterated for each range level.
5.4.3.4.3 Basic protocol and calculations
5.4.3.4.3.1 General case
The number of test materials may vary in relation to the NUMBER of replicas. In
practice, we consider that the number of measurements of all test materials must be
higher than 20. It is not necessary for the repeatability conditions to be maintained
from one test material to another, but all the replicas carried out on the same test
material must be carried out under these repeatability conditions.
Repeatability
remains
Var(repeat)
S v  Var( xij ) 
2
a
special
case
of
the
precision
calculation
. The Var(repeat) part is naturally equal to 0 (only one
measurement with each replica), and the calculation is the same as the calculation
of Var( xij )
 ( xij  M x )
n
S r  Var( x ij ) 
pi
2
i
i 1 j 1
N n
The value r means that in 95% of the cases, the difference between two values
acquired under repeatable conditions will be lower than or equal to r.
5.4.3.4.3.2 Particular case applicable to only 1 repetition
OIV-MA-AS1-12 : R2005
58
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
In practice, the most current situation for automated systems is the analysis of test
material with only one repetition. It is advisable to use at least 10 materials in
order to reach the 20 measurements required. The two measurement replicas of the
same test material must be carried out under repeatable conditions.
In this precise case, the calculation of Sr is simplified and becomes:
q
Sr 
 wi2
i 1
2p
in which:
Sr = the repeatability standard deviation
p = the number of test materials analyzed in
duplicate
wi = the absolute differences between duplicates
Repeatability r is calculated according to the formula:
r = 2.8 Sr
OIV-MA-AS1-12 : R2005
59
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Example: For the alternative determination method of the free sulfur
dioxide in question, and for a range of measurements from 0 to 50 mg/l,
the operator will seek at least 10 samples with regularly distributed
concentrations ranging between these values.
Sample no.
xi
(in mg/l)
x’i
(in mg/l)
1
2
3
4
5
6
7
8
9
10
11
12
14
25
10
2
35
19
23
27
44
30
8
48
14
24
10
3
35
19
23
27
45
30
8
46
Wi
(absolute
value)
0
1
0
1
0
0
0
0
1
0
0
2
Example: Using the values given in the table above, the
following results are obtained:
Q = 12
Sr = 0.54 mg/l
R = 1.5 mg/l
This result can be used to state that, with a probability of
95%, the results obtained by the method under study will
have a repeatability rate lower than 1.5 mg/l.
5.4.3.4.4 Comparison of repeatability
5.4.3.4.4.1 Determination of the repeatability of each method
To estimate the performance of a method, it can be useful to compare its
repeatability with that of a reference method.
OIV-MA-AS1-12 : R2005
60
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Let Sr-alt be the repeatability standard deviation of the alternative method, and Sr-ref.
the repeatability standard deviation of the reference method.
The comparison is direct. If the value of repeatability of the alternative method is
lower than or equal to that of the reference method, the result is positive. If it is
higher, the laboratory must ensure that the result rests compliant with the
specification that it accepted for the method concerned. In the latter case, it may
also apply a Fischer-Snedecor test to know if the value found for the alternative
method is significantly higher than that of the reference method.
5.4.3.4.4.2 Fischer-Snedecor test
Calculate the ratio:
Fobs 
Sr
2
Sr
2
alt
ref
Use the critical Snedecor value with a risk α equal to 0.05 corresponding to the
Fischer variable with a confidence level 1 α, in which ν1 = n(x)-n, and ν2 = n(z)-m
degrees of freedom: F(N(x)-n, N(y)-m, 1- α). In the case of a calculated
repeatability with only one repetition on p test materials for the alternative method,
and q test materials for the reference method, the Fischer variable will have as a
degree of freedom ν1 = p, and ν2 = Q, i.e.: F(p, Q, 1- α).
Interpreting the test:
1/ Fobs > F1- α, the repeatability value of the alternative method is
significantly higher than that of the reference method.
2/ Fobs < F1- α, we cannot state that the repeatability value of the
alternative method is significantly higher than that of the reference method.
Example: The value of the repeatability standard deviation
found for the determination method of free sulfur dioxide is:
Sr = 0.54 mg/l
The laboratory carried out the determination on the same test
materials using the OIV reference method. The value of the
repeatability standard deviation found in this case is:
OIV-MA-AS1-12 : R2005
61
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Sref = 0.39 mg/l
2
Fobs 
0.542  0,29  1,93
0.39 0,15
ν 2 = 12
ν 1 = 12
F1-α = 2.69 > 1.93
The Fobs value obtained is lower than the value F1-α; we cannot
state that the repeatability value of the alternative method is
significantly higher than that of the reference method.
5.4.3.5 Intralaboratory reproducibility
5.4.3.5.1 Definition
Intralaboratory reproducibility is the closeness of agreement between the analysis
results obtained with the method under consideration on the same wine, in the
same laboratory, with the same operator or different operators using from the
different gauging curves, on different days.
5.4.3.5.2 Scope
Reproducibility studies can be implemented on quantitative methods, if the time of
analysis is reasonably limited, and if the capacity exists to keep at least one test
material stable over time.
In many cases, reproducibility is not constant throughout the validity range of the
method. In this case, it is advisable to define several sections or "range levels", in
which it can be reasonably considered that reproducibility is comparable to a
constant. The reproducibility calculation is then to be reiterated for each range
level.
5.4.3.5.3 Basic protocol and calculations
The laboratory chooses one or more stable test materials. It applies the method
regularly for a period equal to at least one month and keeps the results obtained
(Xij, material i, replica j). A minimum of 5 replicas is recommended for each test
material, the total minimum number of replicas being 10. The replicas can be
analyzed in duplicate.
OIV-MA-AS1-12 : R2005
62
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The calculation of precision fully applies to the calculation of reproducibility,
integrating Var (repeat ) if the measurements are carried out in duplicate.
Reproducibility R is calculated according to the formula:
R = 2.8 SR
The value R means that in 95% of the cases, the difference between two values
acquired under reproducibility conditions will be lower than or equal to R.
Example: Reproducibility study of the determination of the
sorbic acid in wines by steam distillation and reading by
absorption at 256 Nm.
Two different sorbated wines were kept for a period of 3
months. The determination of the sorbic acid was carried out at
regular intervals over this period, with repetition of each
measurement.
Replicas
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Test material 1
x1
x2
122
125
123
120
132
130
121
115
130
135
135
142
137
135
130
125
123
130
112
115
131
128
Test material 2
x1
x2
140
139
138
137
139
141
143
142
139
139
135
138
139
139
145
145
138
137
135
134
146
146
137
138
146
147
145
148
130
128
n=2
p1 = 11
p2 = 15
n = 26
OIV-MA-AS1-12 : R2005
63
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Var(xij )37.8
Var(repet) 5.01
SR = 6.35
R = 17.8
6. Quality control of analysis methods (IQC)
6.1 Reference documents
- Resolution OIV Œno 19/2002: Harmonized recommendations for internal
quality control in analysis laboratories.
- CITAC/EURACHEM: Guide for quality in analytical chemistry, 2002
Edition
- Standard NF V03-115, Guide for the use of reference materials
6.2 General principles
It is recalled that an analysis result can be affected two types of error: systematic
error, which translates into bias, and random error. For series analyses, another
type of error can be defined, which can be due to both systematic error and random
error: this is the series effect, illustrated for example by the deviation of the
measuring system during a series.
The IQC is designed to monitor and control these three errors.
6.3 Reference materials
The IQC is primarily based on exploiting the measurement results for reference
materials. The choice and constitution of the materials are therefore essential steps
that it must be controlled in order to provide an efficient basis for the system.
A reference material is defined by two parameters:
- Its matrix
- The assignment of its reference value
Several cases are possible; the cases encountered in oenology are summarized in
the following two-dimensional table:
OIV-MA-AS1-12 : R2005
64
External
value to
the
laborato
ry
Value
obtained
Value
by the
obtained
method
by a
to be
referenc
checked
e method
The use
of
the
instrume
nt value
as
a
referenc
e value
does not
control
accuracy
.
An
alternati
ve
approac
h must
be
set
up.
Referen
ce value
Value
obtained
by
formulat
ion
Matrix
Synthetic solution
Synthetic solutions can
be used to constitute
reference
materials
The
solution
mustquite
be
easily. They are using
not
produced
compatible
methods
metrologicalwith
rules.
It is
with non-specific
recalled
that signals,
the
and that are sensitive
to
formulation
value
The
matrix effects.
obtained
is organization
prone to
supplying
uncertainty. the solution
must
provide guarantees
The application
of such a
about
its
quality
and be
case can be used
to
certified
if
possible.
monitor the precisionThe
of
reference
will
be
the method,values
as well
as its
accompanied
an
If
the synthetic
solution
accuracy
in a by
point
in
uncertainty
at a
has
not tobeen
obtained
relation
avalue
calibrated
given
confidence
level.
with
a
calibrated
reference.
This case can
used to
material,
the bereference
monitor
the
precision
of
value can be determined
a method,
and to check
by
analyzing
the
its accuracy
in a using
point
synthetic
solution
The reference value is
compared
with
the
the
reference
measured
by the method.
method
external
value. This
has
The
measurement
is
to
be
to be checked. The
traceability
value
in
this
carried
least 3
material outis at measured
point The
if selected
the supplier
times.
valuea
over 10 repetitions, and
organization
is of
approved
is
the
mean
the
check will be made that3
for the insofar
preparation
of
results,
as they
the differences between
reference
material
in
remain
within
an
interval
these values are lower
question. Itthan
cannot the
be
lower
than the repeatability
applied
to
methods
repeatability
the
value; the most ofextreme
sensitive If necessary,
to
matrix
method.
the
values can be withdrawn,
effects. can check the
operator
up to a limit of two
consistency
the results
values. To ofensure
the
obtained
with
the
consistency of the values
formulation
valuethe
for the
obtained over
10
solution.
repetitions, the series is
This
casechecked
can be used
to
to be
using
The external value has been
determined on the wine by an
interlaboratory analysis chain.
Certain organizations propose
conditioned wine samples whose
values have been determined in
this measurement
way. However,
in certain
The
is carried
out 3
cases, with
the wines
presentedmethod,
in this
times
the reference
wayselected
may have
beenis doped
and/or
the
value
the mean
of
chemically
the
3 results,stabilized,
insofar as which
they
means
matrixanmay
be affected.
remainthe
within
interval
lower
This case
be used to monitor
than
the can
repeatability
of the
the precision of a method, and to
method.
The reference value is measured
checkcase
its can
accuracy
into amonitor
point
This
be
used
by the method
to be
checked.
The
compared
withofthe
external and
value.
the
precision
a
method,
to
material is measured over 10
This
has
traceability
value
in
this
check
its accuracy
in is
a to
point
repetitions,
and a check
be
point if thewith
analysis
chain
has
compared
the
reference
made that the differences between
been accredited.
Itbe
canapplied
be applied
method.
It can
to
these values
are lower
than the
to methodssensitive
sensitive toto matrix
methods
repeatability value; the most
effects.
extreme values can be withdrawn,
up to a limit of two values. To
ensure the consistency of the
values obtained over the 10
repetitions, this series is to be
checked on the one hand by
control materials established
during a previous session, placed
at the start and end of the series.
The value obtained can also be
Natural matrix (wine etc.)
Natural
matrices
a priori
constitute the most interesting
reference
materials because they
Not
applicable
avoid any risk of matrix effect for
methods that are not perfectly
specific.
OIV-MA-AS1-12 : R2005
65
ensure the consistency of the values
obtained during the 10 repetitions,
the series should be checked using
control materials established during
a previous session, placed at the start
and end of the series.
This case can only be used to
monitor the precision of the method;
Doped wine
A doped wine is a wine with an
artificial addition of an analyte.
This method is applicable when the
base wine is completely free of
analyte. These types of materials are
suitable for oenological additives
that are not native to the wine. If
In
practice,
this involves
doping
is applied
with a conditioned
component
wine
doped
and/or
native tosamples
the wine, the
matrix can
no
chemically
stabilized natural.
as proposed
by
longer be considered
Doping
organizations.
These
materials
must be carried out according to
cannot
claim torules.
constitute
metrological
Thea natural
value
matrix.
reference
values are
obtained The
is prone
to uncertainty.
generally
generated
an
analysis
The
measurement
isbyto
carried
outthe
3
This case
can
be used
monitor
chain.
times
withofthethereference
precision
method,method,
as well the
as
This
case
can be
to
monitor
the
value
retained
mean
the
its accuracy
inisused
athe
point.
It ofcan
be3
precision
of
the
method,
as
well
as
results,
insofar
as
they
remain
within
applied to methods sensitive to
its
accuracy
in a point
an
interval
lower
than
the
matrix
effects
forcompared
non-native
with
the external
standard.
has
repeatability
the
method.
compounds
ofofthe
wine,
but This
not in
the
a
traceability
value
in this
point
if the
This
case
can
be
used
to
monitor
case
of
native
compounds
of
The reference value is measured
the
organization
supplying
precision
of a method,
andthe
to check
wine.
using the method
to be checked.
The
samples
has been
approved
for the
its
accuracy
in
a
point
compared
material is measured over 10
preparation
of the reference
with
the reference
method.is Itmaterial
can be
repetitions,
and a check
made
to
in question.
It methods
cannot be sensitive
applied to to
applied
to
ensure that the differences between
methods sensitive
matrix effects.
matrix
effectsareto for
these values
lower non-native
than the
compounds
of
the
wine,
but not
in the
repeatability value; the most
extreme
case
of
native
compounds
of
values can be withdrawn, up tothea
wine.
limit of two values withdrawn. To
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
6.4 Checking the analytical series
6.4.1 Definition
An analytical series is a series of measurements carried out under repeatable
conditions.
For a laboratory that mainly uses the analytical series method of analysis, a check
must be made to ensure the instantaneous adjustment of the measuring instrument
and its stability during the analytical series is correct.
Two complementary approaches are possible:
- the use of reference materials (often called by extension "control
materials”)
- the use of an internal standard, in particular for separative methods.
6.4.2 Checking accuracy using reference materials
Systematic error can be checked by introducing reference materials, the reference
value of which has been assigned using means external to the method being
checked.
The measured value of the reference material is associated with a tolerance limit,
inside which the measured value is accepted as being valid. The laboratory defines
tolerance values for each parameter and for each analytical system. These values
are specific to the laboratory.
The control materials must be selected so that their reference values correspond to
the levels of the values usually found for a given parameter. If the scale of
measurement is broad, and the uncertainty of measurement is not constant on the
scale, several control materials should be used to cover the various range levels.
6.4.3 Intraseries precision
When the analytical series are rather long, there is a risk of drift of the analytical
system. In this case, intraseries precision must be checked using the same reference
material positioned at regular intervals in the series. The same control materials as
those used for accuracy can be used.
The variation in the measured values for same reference material during the series
should be lower than the repeatability value r calculated for a confidence level of
95%.
OIV-MA-AS1-12 : R2005
66
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
NOTE For a confidence level of 99%, a value of 3.65.Sr can be used.
6.4.4 Internal standard
Certain separative methods enable the introduction of an internal standard into the
product to be analyzed.
In this case, an internal standard should be introduced with calibrated material
with a known uncertainty of measurement.
The internal standard enables a check to be made both of intraseries accuracy and
precision. It should be noted that a drift affects the signals of the analyte and of the
internal standard in equal proportions; since the value of the analyte is calculated
with the value of the signal of the internal standard, the effect of the drift is
cancelled.
The series will be validated if the internal standards are inside the defined
tolerance values.
6.5 Checking the analysis system
6.5.1 Definition
This concerns an additional check to the series check. It differs from the latter in
that it compiles values acquired over long time scales, and/or compares them with
values resulting from other analysis systems.
Two applications will be developed:
- Shewhart charts to monitor the stability of the analysis system
- Internal and external comparison of the analysis system
6.5.2 Shewhart chart
Shewhart charts are graphic statistical tools used to monitor the drift of
measurement systems, by the regular analysis, in practice under reproducibility
conditions, of stable reference materials.
6.5.2.1 Data acquisition
A stable reference material is measured for a sufficiently long period, at defined
regular intervals. These measurements are recorded and logged in control charts.
The measurements are made under reproducibility conditions, and are in fact
OIV-MA-AS1-12 : R2005
67
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
exploitable for the calculation of reproducibility, and for the assessment of
measurement uncertainty.
The values of the analytical parameters of the reference materials selected must be
within valid measurement ranges.
The reference materials are analyzed during an analytical series, routine if
possible, with a variable position in the series from one time to another. In
practice, it is perfectly possible to use the measurements of control materials of the
series to input the control charts.
6.5.2.2 Presentation of results and definition of limits
The individual results are compared with the accepted value of the reference
material, and with the reproducibility standard deviation for the parameter in
question, at the range level in question.
Two types of limits are defined in the Shewhart charts, the limits associated with
individual results, and the limits associated with the mean.
The limits defined for the individual results are usually based on the standard
deviation values for intralaboratory reproducibility for the range level in question.
They are of two types:
- alert limit:
- action limit:
 / 2.S R .
 / 3.S R .
The limit defined for the cumulated mean narrows as the number of measurements
increases.
- This limit is an action limit:
 /
3.S R
.
n being the number of
n
measurements indicated on the chart.
NOTE For reasons of legibility, the alert limit of the cumulated mean is only
rarely reproduced on the control chart, and has as its value  /
OIV-MA-AS1-12 : R2005
2.S R
n
.
68
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Shewhart chart
0,31
0,29
Individualindivuduels
results
résultats
Cumulated
mean
Moyenne
cumulée
Average
upper
limit
Limite
supaction
d'action
moyenne
Average
lower
limit
Limité
inf action
d'action
moyenne
0,27
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Individual alert upper limit
Limite sup d'alerte indiv
Individual alert lower limit
Limite inf d'alerte indiv
Individual
upper
limit
Limite
supaction
d'action
indiv
Individual action lower limit
Limite inf d'action indiv
0,25
0,23
6.5.2.3 Using the Shewhart chart
Below we indicate the operating criteria most frequently used. It is up to the
laboratories to precisely define the criteria they apply.
Corrective action on the method (or the apparatus) will be
undertaken:
a) if an individual result is outside the action limits of the individual
results.
b) if two consecutive individual results are located outside the alert
limits of individual results.
c) if, in addition, a posteriori analysis of the control charts indicates
a drift in the method in three cases:
- nine consecutive individual result points are located on the same
side of the line of the reference values.
- six successive individual result points ascend or descend.
- two successive points out of three are located between the alert
limit and the action limit.
d) if the arithmetic mean of n recorded results is beyond one of the
action limits of the cumulated mean (which highlights a systematic
deviation of the results).
NOTE
The control chart must be revised at n = 1 as
soon as a corrective action has been carried out on the
method.
OIV-MA-AS1-12 : R2005
69
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
6.5.3 Internal comparison of analysis systems
In a laboratory that has several analysis methods for a given parameter, it is
interesting to carry out measurements of the same test materials in order to
compare the results. The agreement of the results between the two methods is
considered to be satisfactory if their variation remains lower than 2 times the
standard deviation of difference calculated during validation, with a confidence
level of 95%.
NOTE This interpretation is possible given the assumption that the variations
obey a normal law with a 95% confidence rate.
6.5.4 External comparison of the analysis system
6.5.4.1 Analysis chain of interlaboratory comparisons
The organization of the tests and calculations is given in the chapter "comparison
with an interlaboratory analysis chain".
In addition to checking the accuracy by the Zscore the results can be analyzed in
greater detail, in particular with regard to the position of the values of the
laboratory in relation to the mean. If they are systematically on the same side of
the mean for several successive analysis chains, this can justify the implementation
of corrective action by the laboratory, even if Zscore remains lower than the critical
value.
NOTE Interpreting the Zscore is possible given the assumption that the variations
obey a normal law with a 95% confidence rate.
If the intercomparison chain is subject to accreditation, this work of comparison
has traceability value.
6.5.4.2 Comparison with external reference materials
Measuring external reference materials at regular intervals also can be used to
supervise the occurrence of a systematic error (bias).
The principle is to measure the external reference material, and to accept or refuse
the value in relation to tolerance limits. These limits are defined in relation to the
OIV-MA-AS1-12 : R2005
70
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
combination of the uncertainties of the controlled method and the reference value
of the reference material.
6.5.4.2.1 Standard uncertainty of reference material
The reference values of these materials are accompanied by confidence intervals.
The laboratory must determine the nature of this data, and deduce from them the
standard uncertainty value for the reference value Sref. A distinction must be made
between several cases:
- The case in which uncertainty a is given in the form of an interval confidence at
95% (expanded uncertainty). This means that a normal law has been adopted. a
therefore constitutes an "expanded uncertainty" and corresponds to 2 times the
standard deviation Sref of the standard uncertainty of the reference values of the
materials provided.
S ref 
-
a
2
The case of a certificate, or another specification, giving limits +/- a
without specifying the confidence level. In this case, a rectangular
dispersion has been adopted, and the value of measurement X has the
same chance of having an unspecified value in the interval ref+/- a.
S ref 
-
a
3
The particular case of glassware giving limits +/- a. This is the
framework of a triangular dispersion.
S ref 
a
6
6.5.4.2.2 Defining the validity limits of measuring reference material
To standard uncertainty Sref of the value of the external reference material, is
added the standard uncertainty of the laboratory method to be checked, Smethod.
These two sources of variability must be taken into account in order to determine
the limits.
Smethod is calculated from the expanded uncertainty of the laboratory method in the
following way:
OIV-MA-AS1-12 : R2005
71
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
S method 
uncertaint y
2
The validity limit of the result (with a confidence level of 95%) =
reference value  /  2.
2
Sref  Smethod
2
Example: A pH 7 buffer solution is used to check a pHmeter. The confidence interval given by the pH solution
is +/- 0.01. It is indicated that this confidence interval
corresponds to the expanded uncertainty with a
confidence level of 95%. In addition the expanded
uncertainty of the pH-meter is 0.024.
2
The limits will be
 / 2
0.01  0.024
(
) (
)
2
2
2
i.e. +/- 0.026 in relation to the reference value, with a
confidence level of 95%.
7. Assessment of measurement uncertainty
7.1 Definition
Parameter, associated with the result of a measurement, which characterizes the
dispersion of the values that can reasonably be allotted to the measurand.
In practice, uncertainty is expressed in the form of a standard deviation called
standard uncertainty u(x), or in an expanded form (generally with k = 2) U = +/k.u
OIV-MA-AS1-12 : R2005
72
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
7.2 Reference documents
- AFNOR ENV 13005 Standard: 1999 Ŕ Guide for expressing measurement
uncertainty
- EURACHEM, 2000. Quantifying Uncertainty in Analytical Measurement,
EURACHEM second edition 2000
- ISO 5725 Standard: 1994 Ŕ Exactitude (accuracy and precision) of results and
measurement methods
- ISO 21748 standard: 2004 Ŕ Guidelines relating to the use of estimations of
repeatability, reproducibility and accuracy in evaluating measurement uncertainty
- Perruchet C and Priel M., Estimating uncertainty, AFNOR Publications, 2000
7.3 Scope
Uncertainty provides two types of information.
-
-
On the one hand, that intended for the customers of the laboratory,
indicating the potential variations to take into account in order to
interpret the result of an analysis. It must be indicated, however, that
this information cannot be used as an external means of evaluating the
laboratory.
In addition, it constitutes a dynamic in-house tool for evaluating the
quality of the laboratory analysis results. Insofar as its evaluation is
regular and based on a fixed, well-defined methodology, it can be used
to see whether the variations involved in a method change positively
or negatively (in the case of an estimate based exclusively on
intralaboratory data).
The present guide limits itself to providing a practical methodology for
oenological laboratories dealing with series analyses. These laboratories have large
volumes of data of a significant statistical scale.
Estimating uncertainties can therefore be carried out in most cases using the data
collected as part validation and quality control work (in particular with the data in
the Shewhart charts). These data can be supplemented by experiment schedules, in
particular to determine the systematic errors.
The reference systems describe two main approaches for determining uncertainty:
the intralaboratory approach and the approach interlaboratory. Each provides
results that are naturally and significantly different. Their significance and their
interpretation cannot be identical.
OIV-MA-AS1-12 : R2005
73
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
-
the intralaboratory approach provides a result specific to the
method in question, in the laboratory in question. The uncertainty that
results is an indicator of the performance of the laboratory for the
method in question. It answers the customer as follows: "what
dispersion of results can I expect from the laboratory practicing the
method?”
-
the interlaboratory approach uses results resulting from
interlaboratory tests, which provide information about the overall
performance of the method.
Laboratories can use the two approaches jointly. It will be interesting to see
whether the results obtained using the intralaboratory approach give values lower
than the values of the interlaboratory approach.
7.4 Methodology
The work of uncertainty assessment involves 3 fundamental steps.
- Definition of the measurand, and description of the quantitative
analysis method
- Critical analysis of the measurement process
- Uncertainty assessment.
7.4.1 Definition of the measurand, and description of the quantitative analysis
method
First of all, the following must be specified:
- the purpose of the measurement
- the quantity measured
- If the measurand is to be obtained by calculation based on measured
quantities, if possible the mathematical relation between them should be stipulated.
- all the operating conditions.
These items are included in theory in the procedures of the laboratory quality
system.
In certain cases the expression of the mathematical relation between the measurand
and the quantities can be highly complex (physical methods etc.), and it is neither
necessarily relevant nor possible to fully detail them.
OIV-MA-AS1-12 : R2005
74
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
7.4.2 Critical analysis of the measurement process
The sources of error influencing the final result should be identified in order to
constitute the uncertainty budget. The importance of each source can be estimated,
in order to eliminate those that have only a negligible minor influence. This is
done by estimating:
- the degree of gravity of the drift generated by poor control of the
factor in question
- the frequency of the potential problems
- their detectability.
This critical analysis can, for example, be carried out using the "5M” method.
Labor;
Operator effect
Matter:
Sample effect (stability, homogeneity, matrix effects), and consumables (reagents,
products, solutions, reference materials), etc.
Hardware:
Equipment effect (response, sensitivity, integration modes, etc.), and laboratory
equipment (balance, glassware etc.).
Method:
Application effect of the procedure (operating conditions, succession of the
operations etc.).
Medium:
Environmental conditions (temperature, pressure, lighting, vibration, radiation,
moisture etc.).
7.4.3 Estimation
approach)
calculations
of
standard
uncertainty
(intralaboratory
7.4.3.1 Principle
In the case of laboratories using large series of samples with a limited number of
methods, a statistical approach based on intralaboratory reproducibility,
supplemented by the calculation of sources of errors not taken into account under
OIV-MA-AS1-12 : R2005
75
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
intralaboratory reproducibility conditions, appears to be the most suitable
approach.
An analysis result deviated from the true value under the effect of two sources of
error: systematic errors and random errors.
Analysis result = True value + Systematic error + Random error
Uncertainty characterizes the dispersion of the analysis result. This translates into
a standard deviation.
Variability (analysis result) = uncertainty
Variability (true value) = 0
Variability (systematic error) =
 S erreurs_ systématiques
2
Variability (random error) = SR (intralaboratory reproducibility standard deviation)
Since standard deviations are squared when added, the estimated standard
uncertainty u(x) takes the following form:
u ( x) 
 u(systematic_ errors)  S R
2
2
Non-integrable sources of errors under the intralaboratory reproducibility
conditions, i.e. systematic errors, must be determined in the form of standard
deviation to be combined together and with the reproducibility standard deviation.
The laboratory can take action so that the reproducibility conditions applied make
it possible to include a maximum number of sources of errors. This is obtained in
particular by constituting stable test materials over a sufficiently long period,
during which the laboratory takes care to vary all the possible experimental
factors. In this way, SR will cover the greatest number of possible sources of errors
(random), and the work involved in estimating the systematic errors, which is often
more complex to realize, will be minimized.
It should be noted here that the EURACHEM/CITAC guide entitled "Quantifying
uncertainty in analytical measurements" recalls that "In general, the ISO Guide
requires that corrections be applied for all systematic effects that are identified and
significant". In a method "under control", systematic errors should therefore
constitute a minor part of uncertainty.
OIV-MA-AS1-12 : R2005
76
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The following non-exhaustive table gives examples of typical sources of error and
proposes an estimation approach for each of them, using integration under
reproducibility conditions as much as possible.
Source of error
Sampling
(constitution of
the sample)
Sub-sampling
(sampling a
quantity of
sample in order
to carry out the
test)
Type of error
Random
Commentary
Sampling is one of the
"businesses" defined in the
ISO 17025 standard.
Laboratories stating they do
not perform sampling, do not
include this source of error in
the uncertainty assessment.
Estimation method
Can be including in
intralaboratory reproducibility
by including sampling in
handling.
Random
Is significant if the sample is
not homogeneous. This
source of error remains minor
for wine.
Included in the
intralaboratory reproducibility
conditions if the test material
used is similar to routine test
materials.
Stability of the
sample
Random
Depends on the storage
conditions of the sample. In
the case of wines, laboratories
should pay detailed attention
to the losses of sulfur dioxide
and ethanol.
Possible changes in the
sample can be integrated into
the reproducibility conditions.
This source of uncertainty can
then be evaluated overall.
Gauging of the
apparatus
Systematic/Rando
m
This error is
systematic if
gauging is
established for a
Source of error to be taken
long period, and
into account in absolute
becomes random if
methods.
gauging is regularly
carried out over a
time-scale
integrated under
reproducibility
conditions
Effect of
contamination
or memory
Random
Precision of
automata
Random
OIV-MA-AS1-12 : R2005
Error of gauging line §
7.4.2.4.1
Taken into account under the
reproducibility conditions if
gauging is regularly revised.
The reproducibility
conditions take this effect into
account, as long as the
reference materials are
inserted at various positions
in the analysis series.
This applies to intraseries
The reproducibility
drift in particular. This can be conditions take this effect into
controlled in particular by
account, as long as the
positioning the control
reference materials are
materials within the
inserted at various positions
framework of the IQC
in the analysis series.
This effect will be minimized
by the proper design of
measuring instruments and
suitable rinsing operations
77
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Purity of the
reagents
Random
The purity of the reagents has
very little effect on the
relative methods, insofar as
the gauging and analyses are
carried out with the same
batches of reagents.
This effect is to be taken into
account in absolute methods.
Measurement
conditions
Random
Effects of temperature,
moisture etc.
Matrix effect
Random from one
sample to another,
systematic on the
same sample
These effects are to be taken
into account in methods
whose measured signal is not
perfectly specific.
Gauging effect
Systematic if
gauging is constant
Random if gauging
is regularly
renewed
Operator effect
Random
Bias
Systematic
Must be minimized by the
quality control work of the
laboratory.
To be integrated under
reproducibility conditions
using various batches of
reagents.
Typically taken into account
under reproducibility
conditions
If this effect is regarded as
significant, a specific
experiment schedule can be
used to estimate uncertainty
due to this effect § 7.4.2.4.3
This effect is not integrated
under reproducibility
conditions.
Taken into account under the
reproducibility conditions if
gauging is regularly renewed.
If the gauging used remains
the same one (on the scale of
the periods in question within
the framework of the
reproducibility conditions), it
is advisable to implement an
experiment schedule in order
to estimate the error of the
gauging line § 7.4.2.4.1
To be taken into account in
the reproducibility conditions
by taking care to utilize all the
authorized operators.
Systematic effect, can be
estimated using certified
references.
7.4.3.2 Calculating the standard deviation of intralaboratory reproducibility
The reproducibility standard deviation SR is calculated using the protocol
described in the section entitled "Intralaboratory reproducibility" (cf. § 5.4.3.5).
The calculation can be based on several test materials. In the noteworthy case
where SR is proportional to the size of the measurand, the data collected on several
test materials with different values should not be combined: SR should be
expressed in relative value (%).
OIV-MA-AS1-12 : R2005
78
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
7.4.3.3 Estimating typical sources of systematic errors not taken into account
under reproducibility conditions
7.4.3.3.1 Gauging error (or calibration error)
Whenever the gauging of an instrument (or the calibration of an absolute method)
is not regularly redone, its output cannot be integrated in the reproducibility
values. An experiment schedule must be carried out in order to estimate it using
the residual error of the regression.
7.4.3.3.1.1 Procedure
The approach is similar to that carried out in the linearity study of the method.
It is recommended to implement a number n of reference materials. The number
must be higher than 3, but it is not necessary to go beyond 10. The reference
materials are to be measured p times under intralaboratory precision conditions, p
must be higher than 3, a figure of 5 is generally recommended. The accepted
values of reference materials must be regularly distributed on the range of values
under study. The number of measurements must be the same for all the reference
materials.
The results are reported in a table presented as follows:
Reference
materials
1
…
i
…
n
Accepted value of
the reference
material
x1
…
xi
…
xn
OIV-MA-AS1-12 : R2005
Replica
1
y11
…
yi1
….
yn1
Measured values
Replica
j
…
y1j
…
…
…
yij
…
…
…
ynj
…
…
…
…
…
…
…
Replica
p
y1p
…
yip
…
ynp
79
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
7.4.3.3.1.2 Calculations and results
The linear regression model is calculated.
yij  a  b.xi   ij
where
yij
xi
is jth replica of the ith reference material.
is the accepted value of the ith reference material.
is the slope of the regression line.
is the intercept point of the regression line.
b
a
ab.xi
represent the expectation of the measurement value of the ith
reference material.
ij
is the difference between yij and the expectation of the measurement value
of the i reference material.
th
The parameters of the regression line are obtained using the following formulae:
- mean of p measurements of the ith reference material
yi 
1
p
p
 yij
j 1
Mx 
- mean of all the accepted values of n reference materials
- mean of all measurements
OIV-MA-AS1-12 : R2005
1
My 
n
1
n
n
 xi
i 1
n
 yi
i 1
80
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
n
 ( xi  M x )( yi  M y )
- estimated slope b
b
i 1
n
 ( xi  M x )
2
i 1
- estimated intercept point a
a  M y b Mx
- regression value associated with the ith reference material
ŷi
yˆ i  a  b  xi
- residual eij
eij  yij  yˆ i
7.4.3.3.1.3 Estimating the standard uncertainty associated the gauging line (or
calibration line)
If the errors due to the regression line are constant over the entire field, the
standard uncertainty is estimated in a global, single way by the overall residual
standard deviation.
n
p
 ( yij  yˆ i )
u( gauging)  S res 
2
i 1 j 1
np  2
If the errors due to the regression line are not constant over the entire field, the
standard uncertainty is estimated for a given level by the residual standard
deviation for this level.
p
u( gauging), i  Sres,i 
 ( yij  yˆ i )
2
j 1
p 1
NOTE These estimates of standard deviations can be used if the linear regression
model and the gauging (or calibration) domain have been validated (see § 5.3.1)
OIV-MA-AS1-12 : R2005
81
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
7.4.3.3.2 Bias error
According to the EURACHEM guide, "Quantifying uncertainty in analytical
measurements", it is recalled that the ISO guide generally requires that corrections
be applied for all identified significant systematic effects. The same applies to the
bias of methods for which the laboratory implements its quality control system
(see §6), and which tends towards 0 for methods "under control".
In practice, a distinction can be made between two cases:
7.4.3.3.2.1 Methods adjusted with only one certified reference material
Bias is permanently adjusted with the same reference material.
The certified reference material (CRM) ensures the metrological traceability of the
method. A reference value was allotted to the CRM together with its standard
uncertainty uref. This standard uncertainty of the CRM is combined with the
compound uncertainty for the method, ucomp, to determine the overall standard
uncertainty of the laboratory method u(x).
The overall standard uncertainty of the method adjusted with the CRM in question
is therefore:
u ( x) 
2
uref  ucomp
2
NOTE 1 The methodology is identical in the case of methods adjusted with the
results of an interlaboratory comparison chain.
NOTE 2 Note the difference between a CRM used to adjust the bias of a method,
in which the uncertainty of its reference value combines with that of the method,
and a CRM used to control a method adjusted by other means (cf. § 6.5.4.2). In the
second case, the uncertainty of the CRM should not be used for the uncertainty
assessment of the method.
7.4.3.3.2.2 Methods adjusted with several reference materials (gauging ranges
etc.)
There is no particular adjustment of bias apart from gauging work.
It is clear that each calibrator introduces bias uncertainty. There is therefore an
overall theoretical uncertainty of bias, which is a combination of the uncertainties
OIV-MA-AS1-12 : R2005
82
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
of each calibrator. This uncertainty is very delicate to estimate, but it generally
proves to be sufficiently low to be ignored, in particular if the laboratory monitors
the quality of its calibrators, and the uncertainty of their reference values.
Other than in specific cases, bias uncertainty is ignored here.
7.4.3.3.3 Matrix effect
7.4.3.3.3.1 Definition
The matrix effect incurs a repeatable source of error for a given sample, but
random from one sample to another. This error is related to the interaction of the
compounds present in the product to be analyzed on measuring the required
analyte. The matrix effect appears in methods with a nonspecific signal.
The matrix effect often constitutes a small part of uncertainty, particularly in
separative methods. In certain other methods, including the infra-red techniques, it
is a significant component of uncertainty.
Example: Estimate of the matrix effect on FTIR
The signal for the FTIR, or infra-red spectrum, is not a signal
specific to each of the compounds that are measured by this
technique. The statistical gauging model can be used to process
disturbed, nonspecific spectral data in a sufficiently exact estimate
of the value of the measurand. This model integrates the influences
of the other compounds of the wine, which vary from one wine to
the next and introduce an error into the result. Upstream of the
routine analysis work, special work is carried out by the gauging
developers to minimize this matrix effect and to make gauging
robust, i.e. capable of integrating these variations without
reflecting them in the result. Nevertheless the matrix effect is
always present and constitutes a source of error at the origin of a
significant part of the uncertainty of an FTIR method.
To be completely rigorous, this matrix effect error can be
estimated by comparing, on the one hand, the means for a great
number of FTIR measurement replicas, obtained on several
reference materials (at least 10), under reproducibility conditions,
and the true values of reference materials with a natural wine
matrix on the other. The standard deviation of the differences
gives this variability of gauging (provided that the gauging has
been adjusted beforehand (bias = 0)).
OIV-MA-AS1-12 : R2005
83
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
This theoretical approach cannot be applied in practice, because
the true values are never known, but it is experimentally possible
to come sufficiently close to it:
- As a preliminary, the FTIR gauging must be statistically
adjusted (bias = 0) in relation to a reference method based on
at least 30 samples. This can be used to eliminate the effects
of bias in the measurements thereafter.
- The reference materials must be natural wines. It is advisable
to use at least 10 different reference materials, with values
located inside a range level, the uncertainty of which can be
considered to be constant.
- An acceptable reference value is acquired, based on the mean
of several measurements by the reference method, carried out
under reproducibility conditions. This can be used to lower
the uncertainty of the reference value: if, for the reference
method used, all the significant sources of uncertainty range
within reproducibility conditions, the multiplication of the
number p of measurements carried out under reproducibility
conditions, enable the uncertainty associated with their mean
to be divided by p . The mean obtained using a sufficient
number of measurements will then have a low level of
uncertainty, even negligible in relation to the uncertainty of
the alternative method; and can therefore be used as a
reference value. p must at least be equal to 5.
- The reference materials are analyzed by the FTIR method,
with several replicas, acquired under reproducibility
conditions. By multiplying the number of measurements q
under reproducibility conditions on the FTIR method, the
variability related to the precision of the method (random
error) can be decreased. The mean value of these
measurements will have a standard deviation of variability
divided by q . This random error can then become negligible in
relation to the variability linked to the gauging (matrix effect)
that we are trying to estimate. q must at least be equal to 5.
OIV-MA-AS1-12 : R2005
84
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The following example is applied to the determination of acetic
acid by FTIR gauging. The reference values are given by 5
measurements under reproducibility conditions on 7 stable test
materials. The number of 7 materials is in theory insufficient, but
the data here are only given by way of an example.
Reference method
Materi
als
1
2
3
4
5
6
7
1
2
3
4
5
0.3
0.3
0
0.3
1
0.2
8
0.3
5
0.2
9
0.3
7
7
0.3
0.3
2
0.3
2
0.2
9
0.3
5
0.2
9
0.3
6
7
0.3
0.3
1
0.3
2
0.2
9
0.4
5
0.2
0
0.3
6
7
0.3
0.3
0
0.3
2
0.2
8
0.4
4
0.2
0
0.3
6
7
0.3
0.3
1
0.3
1
0.2
8
0.3
5
0.2
9
0.3
6
6
FTIR
Mea
n
Ref
0.30
0.31
8
0.38
6
0.24
4
0.39
8
0.26
4
0.36
2
8
1
2
3
4
5
0
0.
0
3.
0
3.
0
3
1.
0
2
7.
0
4
6.
2
3.
3
5
7
0.3
0.3
1
0.3
2
0.2
7
0.4
6
0.2
2
0.3
6
6
0.3
0.3
1
0.3
0
0.2
7
0.4
6
0.2
3
0.3
5
6
0.3
0.3
0
0.3
1
0.2
7
0.4
5
0.2
2
0.3
5
5
0.3
0.3
0
0.3
1
0.2
6
0.4
6
0.2
2
0.3
6
6
Mean
FTIR
Diff
0.305
0.315
0.37
0.26
0.425
0.255
0.365
-0.004
-0.006
-0.016
0.01
0.03
-0.008
-0.008
Calculation of the differences: diff = Mean FTIR –
Mean ref.
The mean of the differences Md = 0.000 verifies (good
adjustment of the FTIR compared with the reference
method)
The standard deviation of the differences, Sd = 0.015. It is
this standard deviation that is used to estimate the
variability generated by the gauging, and we can
therefore state that:
Uf = 0.015
NOTE
It should be noted that the value of Uf can be
over-estimated by this approach. If the laboratory
considers that the value is significantly excessive under
the operating conditions defined here, it can increase the
number of measurements on the reference method and/or
the FTIR method.
OIV-MA-AS1-12 : R2005
85
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
The reproducibility conditions include all the other
significant sources of error, SR was otherwise calculated:
SR = 0.017
The uncertainty of the determination of acetic acid by
this FTIR application is:
 / 2 *
-1
2
2
0.015  0.017 or +/- 0.045 g.L
7.4.3.3.4 Sample effect
In certain cases, the experiment schedules used to estimate uncertainty are based
on synthetic test materials. In such a situation, the estimate does not cover the
sample effect (homogeneity). The laboratories must therefore estimate this effect.
It should be noted, however, that this effect is often negligible in oenological
laboratories, which use homogeneous samples of small quantities.
7.4.4 Estimating standard uncertainty by interlaboratory tests
7.4.4.1 Principle
The interlaboratory approach uses data output by interlaboratory tests from which
a standard deviation of interlaboratory reproducibility is calculated, in accordance
with the principles indicated in §5.4.3. The statisticians responsible for calculating
the results of the interlaboratory tests can identify "aberrant" laboratory results, by
using tests described in the ISO 5725 standard (Cochran test). These results can
then be eliminated after agreement between the statisticians and the analysts.
For the uncertainty assessment by interlaboratory approach, the guidelines stated
in the ISO 21748 standard are as follows:
1.
The reproducibility standard deviation (interlaboratory) obtained in a
collaborative study is a valid basis for evaluating the uncertainty of
measurement
2.
Effects that are not observed as part of the collaborative study must be
obviously negligible or be explicitly taken into account.
OIV-MA-AS1-12 : R2005
86
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
There are two types of interlaboratory tests:
1. Collaborative studies which relate to only one method. These studies
are carried out for the initial validation of a new method in order to
define the standard deviation of interlaboratory reproducibility SRinter
(method).
2. Interlaboratory comparison chains, or aptitude tests. These tests are
carried out to validate a method adopted by the laboratory, and the
routine quality control (see § 5.3.3.3). The data are processed as a
whole, and integrate all the analysis methods employed by the
laboratories participating in the tests. The results are the interlaboratory
mean m, and the standard deviation of interlaboratory and intermethod
reproducibility SRinter.
7.4.4.2 Using the standard deviation of interlaboratory and intramethod
reproducibility SRinter (method)
The standard deviation of intralaboratory reproducibility SRinter (method) takes
into account intralaboratory variability and the overall interlaboratory variability
related to the method.
Then must be taken into account the fact that the analysis method can produce a
systematic bias compared with the true value.
As part of a collaborative study, whenever possible, the error produced by this bias
can be estimated by using certified reference materials, under the same conditions
as described in § 7.4.3.3.2, and added to SRinter (method).
7.4.4.3 Using the standard deviation of interlaboratory and intermethod
reproducibility SRinter
The standard deviation of intralaboratory reproducibility SRinter takes into account
intralaboratory variability and interlaboratory variability for the parameter under
study.
The laboratory must check its accuracy in relation to these results (see § 5.3.3).
There is no need to add components associated with method accuracy to the
uncertainty budget, since in the "multi-method" aptitude tests, errors of accuracy
can be considered to be taken into account in SRinter.
OIV-MA-AS1-12 : R2005
87
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
7.4.4.4 Other components in the uncertainty budget
Insofar as the test materials used for the interlaboratory tests are representative of
the conventional samples analyzed by laboratories, and that they follow the overall
analytical procedure (sub-sampling, extraction, concentration, dilution, distillation
etc.), SR-inter represents the standard uncertainty u(x) of the method, in the
interlaboratory sense.
Errors not taken into account in the interlaboratory tests must then be studied in
order to assess their compound standard uncertainty, which will be combined with
the compound standard uncertainty of the interlaboratory tests.
7.5 Expressing expanded uncertainty
In practice, uncertainty is expressed in its expanded form, is absolute terms for
methods in which uncertainty is stable in the scope in question, or relative when
uncertainty varies proportionally in relation to the quantity of the measurand:
Absolute uncertainty: U   /  2.u( x)
Relative uncertainty (in %): U   / 
2.u ( x)
.100
x
where x mean represents the reproducibility results.
NOTE This expression of uncertainty is possible given the assumption that the
variations obey a normal law with a 95% confidence rate.
These expressions result in a given uncertainty value with a confidence level of
95%.
OIV-MA-AS1-12 : R2005
88
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
REFERENCES
(1) OIV, 2001 Ŕ Recueil des methods internationales d’analyse des vins and des
moûts; OIV Ed., Paris.
(2) OIV, 2002 Ŕ Recommandations harmonisées pour le contrôle interne de qualité
dans les laboratoires d’analyse; OIV resolution œno 19/2002., Paris.
(3) Standard ISO 5725: 1994 Ŕ Exactitude (justesse and fidélité) des results and
methods de mesure, classification index X 06-041-1
(4) IUPAC, 2002 Ŕ Harmonized guidelines for single-laboratory validation of
analysis methods; Pure Appl. Chem., Vol. 74; n°5, pp. 835-855.
(5) Standard ISO 11095: 1996 Ŕ Etalonnage linéaire utilisant des materials de
référence, reference number ISO 11095:1996
(6) Standard ISO 21748: 2004 Ŕ Lignes directrices relatives à l’utilisation
d’estimation de la répétabilité, de la reproductibilité and de la justesse dans
l’évaluation de l’incertitude de mesure, reference number ISO ISO/TS 21748:2004
(7) Standard AFNOR V03-110: 1998 Ŕ Procédure de validation intralaboratory
d’une method alternative par rapport à une method de référence, classification
index V03-110
(8) Standard AFNOR V03-115: 1996 Ŕ Guide pour l’utilisation des materials de
référence, classification index V03-115
(9) Standard AFNOR X 07-001: 1994 Ŕ Vocabulaire international des termes
fondamentaux and généraux de métrologie, classification index X07-001
(10) Standard AFNOR ENV 13005: 1999 Ŕ Guide pour l’expression de
l’incertitude de mesure
(11) AFNOR, 2003, - Métrologie dans l’entreprise, outil de la qualité 2ème édition,
AFNOR 2003 edition
(12) EURACHEM, 2000. - Quantifying Uncertainty in Analytical Measurement,
EURACHEM second edition 2000
(13) CITAC / EURACHEM, 2000 - Guide pour la qualité en chimie analytique,
EURACHEM 2002 edition
OIV-MA-AS1-12 : R2005
89
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
(14) Bouvier J.C., 2002 - Calcul de l’incertitude de mesure Ŕ Guide pratique pour
les laboratoires d’analyse œnologique, Revue Française d’œnologie no.197, NovDec 2002, pp: 16-21
(15) Snakkers G. and Cantagrel R., 2004 - Utilisation des données des circuits
de comparaison interlaboratoires pour apprécier l’exactitude des results d’un
laboratoire Estimation d’une incertitude de mesure - Bull OIV, Vol. 77 857-876,
Jan – Feb 2004, pp: 48-83
(16) Perruchet C. and Priel M, 2000 - Estimer l’incertitude, AFNOR Editions
(17) Neuilly (M.) and CETAMA, 1993 - Modélisation and estimation des errors
de mesures, Lavoisier Ed, Paris
OIV-MA-AS1-12 : R2005
90
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Annex N°1
Table A
-
Law of SNEDECOR
This table indicates values of F in function with 1 and 2 for a risk of 0,05
P=0,950
1
2
1
2
3
161,4
18,51
10,13
199,5
19,00
9,55
215,7
19,16
9,28
224,6
19,25
9,12
230,2
19,30
9,01
234,0
19,33
8,94
236,8
19,35
8,89
238,9
19,37
8,85
240,5
19,38
8,81
241,9
19,40
8,79
1
2
3
4
5
6
7,71
6,61
5,99
6,94
5,79
5,14
6,59
5,41
4,76
6,39
5,19
4,53
6,26
5,05
4,39
6,16
4,95
4,28
6,09
4,88
4,21
6,04
4,82
4,15
6,00
4,77
4,10
5,96
4,74
4,06
4
5
6
7
8
9
5,59
5,32
5,12
4,74
4,46
4,26
4,35
4,07
3,86
4,12
3,84
3,63
3,97
3,69
3,48
3,87
3,58
3,37
3,79
3,50
3,29
3,73
3,44
3,23
3,68
3,39
3,18
3,64
3,35
3,14
7
8
9
10
4,96
4,10
3,71
3,48
3,33
3,22
3,14
3,07
3,02
2,98
10
11
12
13
4,84
4,75
4,67
3,98
3,89
3,81
3,59
3,49
3,41
3,36
3,26
3,18
3,20
3,11
3,03
3,09
3,00
2,92
3,01
2,91
2,83
2,95
2,85
2,77
2,90
2,80
2,71
2,85
2,75
2,67
11
12
13
14
15
16
4,60
4,54
4,49
3,74
3,68
3,63
3,34
3,29
3,24
3,11
3,06
3,01
2,96
2,90
2,85
2,85
2,79
2,74
2,76
2,71
2,66
2,70
2,64
2,59
2,65
2,59
2,54
2,60
2,54
2,49
14
15
16
17
18
19
4,45
4,41
4,38
3,59
3,55
3,52
3,20
3,16
3,13
2,96
2,93
2,90
2,81
2,77
2,74
2,70
2,66
2,63
2,61
2,58
2,54
2,55
2,51
2,48
2,49
2,46
2,42
2,45
2,41
2,38
17
18
19
20
4,35
3,49
3,10
2,87
2,71
2,60
2,51
2,45
2,39
2,35
20
21
22
23
4,32
4,30
4,28
3,47
3,44
3,42
3,07
3,05
3,03
2,84
2,82
2,80
2,68
2,66
2,64
2,57
2,55
2,53
2,49
2,46
2,44
2,42
2,40
2,37
2,37
2,34
2,32
2,32
2,30
2,27
21
22
23
24
25
26
4,26
4,24
4,23
3,40
3,39
3,37
3,01
2,99
2,98
2,78
2,76
2,74
2,62
2,60
2,59
2,51
2,49
2,47
2,42
2,40
2,39
2,36
2,34
2,32
2,30
2,28
2,27
2,25
2,24
2,22
24
25
26
27
28
29
4,21
4,20
4,18
3,35
3,34
3,33
2,96
2,95
2,93
2,73
2,71
2,70
2,57
2,56
2,55
2,46
2,45
2,43
2,37
2,36
2,35
2,31
2,29
2,28
2,25
2,24
2,22
2,20
2,19
2,18
27
28
29
30
4,17
3,32
2,92
2,69
2,53
2,42
2,33
2,27
2,21
2,16
30
40
60
120
4,08
4,00
3,92
3,23
3,15
3,07
2,84
2,76
2,68
2,61
2,53
2,45
2,45
2,37
2,29
2,34
2,25
2,17
2,25
2,17
2,09
2,18
2,10
2,02
2,12
2,04
1,96
2,08 40
1,99 60
1,91 120

3,84
3,00
2,60
2,37
2,21
2,10
2,01
1,94
1,88
1,83
1
2
3
4
5
6
7
8
9
2
1
1
2
3
4
OIV-MA-AS1-12 : R2005
5
6
7
8
9
10
10
1
2

2
1
91
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Guide for the validation – quality control
Annex N°2
Table B - Law of STUDENT
This table indicates values of t in function with P and 
P
P
0,55
0,60
7.5.1.1.1.1.1.1.1

0,65
0,70
0,75
0,80
0,85
0,90
0,95
0,975
1
2
3
0,158
0,142
0,137
0,325
0,289
0,277
0,510
0,445
0,424
0,727
0,617
0,584
1,000
0,816
0,765
1,376
1,061
0,978
1,963
1,386
1,250
3,078
1,886
1,638
6,314
2,920
2,353
12,706 31,821 63,657 636,619 1
4,303 6,965 9,925 31,598
2
3,182 4,541 5,841 12,929
3
4
5
6
0,134
0,132
0,131
0,271
0,267
0,265
0,414
0,408
0,404
0,569
0,559
0,553
0,741
0,727
0,718
0,941
0,920
0,906
1,190
1,156
1,134
1,533
1,476
1,440
2,132
2,015
1,943
2,776
2,571
2,447
3,747
3,365
3,143
4,604
4,032
3,707
8,610
6,869
5,959
4
5
6
7
8
9
0,130
0,130
0,129
0,263
0,262
0,261
0,402
0,399
0,398
0,549
0,546
0,543
0,711
0,706
0,703
0,896
0,889
0,883
1,119
1,108
1,100
1,415
1,397
1,383
1,895
1,860
1,833
2,365
2,306
2,262
2,998
2,896
2,821
3,499
3,355
3,250
5,408
5,041
4,781
7
8
9
10
0,129
0,260
0,397
0,542
0,700
0,879
1,093
1,372
1,812
2,228
2,764
3,169
4,587
10
11
12
13
0,129
0,128
0,128
0,260
0,259
0,259
0,396
0,395
0,394
0,540
0,539
0,538
0,697
0,695
0,694
0,876
0,873
0,870
1,088
1,083
1,079
1,363
1,356
1,350
1,796
1,782
1,771
2,201
2,179
2,160
2,718
2,681
2,650
3,106
3,055
3,012
4,437
4,318
4,221
11
12
13
14
15
16
0,128
0,128
0,128
0,258
0,258
0,258
0,393
0,393
0,392
0,537
0,536
0,535
0,692
0,691
0,690
0,868
0,866
0,865
1,076
1,074
1,071
1,345
1,341
1,337
1,761
1,753
1,746
2,145
2,131
2,120
2,624
2,602
2,583
2,977
2,947
2,921
4,140
4,073
4,015
14
15
16
17
18
19
0,128
0,127
0,127
0,257
0,257
0,257
0,392
0,392
0,391
0,534
0,534
0,533
0,689
0,688
0,688
0,863
0,862
0,861
1,069
1,067
1,066
1,333
1,330
1,328
1,740
1,734
1,729
2,110
2,101
2,093
2,567
2,552
2,539
2,898
2,878
2,861
3,965
3,922
3,883
17
18
19
20
0,127
0,257
0,391
0,533
0,687
0,860
1,064
1,325
1,725
2,086
2,528
2,845
3,850
20
21
22
23
0,127
0,127
0,127
0,257
0,256
0,256
0,391
0,390
0,390
0,532
0,532
0,532
0,686
0,686
0,685
0,859
0,858
0,858
1,063
1,061
1,060
1,323
1,321
1,319
1,721
1,717
1,714
2,080
2,074
2,069
2,518
2,508
2,500
2,831
2,819
2,807
3,819
3,792
3,767
21
22
23
24
25
26
0,127
0,127
0,127
0,256
0,256
0,256
0,390
0,390
0,390
0,531
0,531
0,531
0,685
0,684
0,884
0,857
0,856
0,856
1,059
1,058
1,058
1,318
1,316
1,315
1,711
1,708
1,706
2,064
2,060
2,056
2,492
2,485
2,479
2,797
2,787
2,779
3,745
3,725
3,707
24
27
28
29
0,127
0,127
0,127
0,256
0,256
0,256
0,389
0,389
0,389
0,531
0,530
0,530
0,684
0,683
0,683
0,855
0,855
0,854
1,057
1,056
1,055
1,314
1,313
1,311
1,703
1,701
1,699
2,052
2,048
2,045
2,473
2,467
2,462
2,771
2,763
2,756
3,690
3,674
3,659
27
28
29
30
0,127
0,256
0,389
0,530
0,683
0,854
1,055
1,310
1,697
2,042
2,457
2,750
3,646
30
40
60
120
0,126
0,126
0,126
0,255
0,254
0,254
0,388
0,387
0,386
0,529
0,527
0,526
0,681
0,679
0,677
0,851
0,848
0,845
1,050
1,046
1,041
1,303
1,296
1,289
1,684
1,671
1,658
2,021
2,000
1,980
2,423
2,390
2,358
2,704
2,660
2,617
3,551
3,460
3,373
40
60
120

0,126
0,253
0,385
0,524
0,674
0,842
1,036
1,282
1,645
1,960
2,326
2,576
3,291

0,55
0,60
0,65
0,70
0,75
0,80
0,85
0,90
0,95
0,975
0,990
0,995
0,9995

P
OIV-MA-AS1-12 : R2005
0,990
0,995
0,9995

25
26
P

92
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
Harmonised guidelines for single-laboratory validation of
methods of analysis (technical report)
(Resolution Oeno 8/2005)
Synopsis
Method validation is one of the measures universally recognised as a necessary
part of a comprehensive system of quality assurance in analytical chemistry. In the
past ISO, IUPAC and AOAC INTERNATIONAL have co-operated to produce
agreed protocols or guidelines on the “Design, Conduct and Interpretation of
Method Performance Studies”1 on the “Proficiency Testing of (Chemical)
Analytical Laboratories”2 on “Internal Quality Control in Analytical Chemistry
Laboratories”3 and on “The Use of Recovery Information in Analytical
Measurement”.4 ( from the usage of overlapping data in analytical measurements)
The Working Group that produced these protocols/guidelines has now been
mandated by IUPAC to prepare guidelines on the Single-laboratory Validation of
methods of analysis. These guidelines provide minimum recommendations on
procedures that should be employed to ensure adequate validation of analytical
methods.
A draft of the guidelines has been discussed at an International Symposium on the
Harmonisation of Quality Assurance Systems in Chemical Laboratory, the
Proceedings from which have been published by the UK Royal Society of
Chemistry.
Resulting from the Symposium on Harmonisation of Quality Assurance
Systems for Analytical Laboratories, Budapest, Hungary, 4-5 November 1999
held under the sponsorship of IUPAC, ISO and AOAC INTERNATIONAL
OIV-MA-AS1-13 : R2005
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
CONTENTS
1
1.1
1.2
INTRODUCTION
Background
Existing protocols, standards and guides
2
2.1
2.2
DEFINITIONS AND TERMINOLOGY
General
Definitions used in this guide
3
Method validation, uncertainty, and quality assurance
4
4.1
4.2
4.3
4.4
BASIC PRINCIPLES OF METHOD VALIDATION
Specification and scope of validation
Testing assumptions
Sources of Error in Analysis
Method and Laboratory effects
5
Conduct of Validation Studies
6
6.1
6.2
used
6.3
method
6.4
Extent of validation studies
The laboratory is to use a “fully” validated method
The laboratory is to use a fully validated method, but new matrix is to be
The laboratory is to use a well-established, but not collaboratively studied,
6.6
6.7
6.8
The method has been published in the scientific literature together with
some analytical characteristics
The method has been published in the scientific literature with no
characteristics given or has been developed in-house
The method is empirical
The analysis is “ad hoc”
Changes in staff and equipment
7
RECOMMENDATIONS
8
BIBLIOGRAPHY
6.5
OIV-MA-AS1-13 : R2005
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
ANNEX A: NOTES ON THE REQUIREMENTS FOR STUDY OF METHOD
PERFORMANCE CHARACTERISTICS.
A1
Applicability
A2
Selectivity
A3
A3.1
A3.2
A3.3
Calibration and linearity
Linearity and intercept
Test for general matrix effect
Final calibration procedure
A4
A4.1
A4.2
A4.3
A4.3.1
A4.3.2
A4.3.3
A4.3.4
Trueness
Estimation of trueness
Conditions for trueness experiments
Reference values for trueness experiments
Certified reference materials (CRMs)
Reference materials
Use of a reference method
Use of spiking/recovery
A5
Accuracy
A6
Recovery
A7
Concentration range
A8
Detection Limit
A9
Limit of determination or limit of quantification
A10
Sensitivity
A11
Ruggedness
A12
Fitness for trial purposes
A13
Matrix variation
A14.
Measurement Uncertainty
OIV-MA-AS1-13 : R2005
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
ANNEX B. ADDITIONAL CONSIDERATIONS
ESTIMATION IN VALIDATION STUDIES
B1
B2
FOR
UNCERTAINTY
Sensitivity analysis
Judgement
OIV-MA-AS1-13 : R2005
4
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
1.
INTRODUCTION
1.1
Background
Reliable analytical methods are required for compliance with national and
international regulations in all areas of analysis. It is accordingly internationally
recognised that a laboratory must take appropriate measures to ensure that it is
capable of providing and does provide data of the required quality. Such measures
include:




using validated methods of analysis;
using internal quality control procedures;
participating in proficiency testing schemes; and
becoming accredited to an International Standard, normally ISO/IEC 17025.
It should be noted that accreditation to ISO/IEC 17025 specifically addresses the
establishment of traceability for measurements, as well as requiring a range of
other technical and management requirements including all those in the list above.
Method validation is therefore an essential component of the measures that a
laboratory should implement to allow it to produce reliable analytical data. Other
aspects of the above have been addressed previously by the IUPAC Interdivisional
Working Party on Harmonisation of Quality Assurance Schemes for Analytical
Laboratories, specifically by preparing Protocols/Guidelines on method
performance (collaborative) studies,1 proficiency testing,2 and internal quality
control.3
In some sectors, most notably in the analysis of food, the requirement for methods
that have been “fully validated” is prescribed by legislation.5,6 “Full” validation
for an analytical method is usually taken to comprise an examination of the
characteristics of the method in an inter-laboratory method performance study
(also known as a collaborative study or collaborative trial). Internationally
accepted protocols have been established for the “full” validation of a method of
analysis by a collaborative trial, most notably the International Harmonised
Protocol1 and the ISO procedure.7 These protocols/standards require a minimum
number of laboratories and test materials to be included in the collaborative trial to
validate fully the analytical method. However, it is not always practical or
necessary to provide full validation of analytical methods. In such circumstances a
“single-laboratory method validation” may be appropriate.
OIV-MA-AS1-13 : R2005
5
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
Single-laboratory method validation is appropriate in several circumstances
including the following:



to ensure the viability of the method before the costly exercise of a formal
collaborative trial;
to provide evidence of the reliability of analytical methods if collaborative trial
data are not available or where the conduct of a formal collaborative trial is
not practicable;
to ensure that “off-the-shelf” validated methods are being used correctly.
When a method is to be characterised in-house, it is important that the laboratory
determines and agrees with its customer exactly which characteristics are to be
evaluated. However, in a number of situations these characteristics may be laid
down by legislation (e.g. veterinary drug residues in food and pesticides in food
sectors). The extent of the evaluation that a laboratory undertakes must meet the
requirements of legislation.
Nevertheless in some analytical areas the same analytical method is used by a large
number of laboratories to determine stable chemical compounds in defined
matrices. It should be appreciated that if a suitable collaboratively studied method
can be made available to these laboratories, then the costs of the collaborative trial
to validate that method may well be justified. The use of a collaboratively studied
method considerably reduces the efforts which a laboratory, before taking a
method into routine use, must invest in extensive validation work. A laboratory
using a collaboratively studied method, which has been found to be fit for the
intended purpose, needs only to demonstrate that it can achieve the performance
characteristics stated in the method. Such a verification of the correct use of a
method is much less costly than a full single laboratory validation. The total cost to
the Analytical Community of validating a specific method through a collaborative
trial and then verifying its performance attributes in the laboratories wishing to use
it is frequently less than when many laboratories all independently undertake
single laboratory validation of the same method.
1.2
Existing Protocols, Standards and Guides
A number of protocols and guidelines8-19 on method validation and uncertainty
have been prepared, most notably in AOAC INTERNATIONAL, International
Conference on Harmonisation (ICH) and Eurachem documents:
 The Statistics manual of the AOAC, which includes guidance on single
laboratory study prior to collaborative testing13
OIV-MA-AS1-13 : R2005
6
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
 The ICH text15 and methodology,16 which prescribe minimum validation study
requirements for tests used to support drug approval submission.
 The Fitness for Purpose of Analytical Methods: A Laboratory Guide to Method
Validation and Related Topics (1998)12
 Quantifying Uncertainty in Analytical Measurement (2000)9
Method validation was also extensively discussed at a Joint FAO/IAEA Expert
Consultation, December 1997, on the Validation of Analytical Methods for Food
Controls, the Report of which is available19.
The present „Guidelines‟ bring together the essential scientific principles of the
above documents to provide information which has been subjected to international
acceptance and, more importantly, to point the way forward for best practice in
single-laboratory method validation.
2
DEFINITIONS AND TERMINOLOGY
2.1
General
Terms used in this document respect ISO and IUPAC definitions where available.
The following documents contain relevant definitions:
i) IUPAC: Compendium of chemical terminology, 1987
ii) International vocabulary of basic and general terms in metrology. ISO 1993
2.2
Definitions used in this guide only:
Relative uncertainty: Uncertainty expressed as a relative standard deviation.
Validated range: That part of the concentration range of an analytical method
which has been subjected to validation.
OIV-MA-AS1-13 : R2005
7
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
METHOD VALIDATION, UNCERTAINTY, AND QUALITY
3
ASSURANCE
Method validation makes use of a set of tests which both test any assumptions on
which the analytical method is based and establish and document the performance
characteristics of a method, thereby demonstrating whether the method is fit for a
particular analytical purpose. Typical performance characteristics of analytical
methods are: applicability; selectivity; calibration; trueness; precision; recovery;
operating range; limit of quantification; limit of detection; sensitivity; and
ruggedness. To these can be added measurement uncertainty and fitness-forpurpose.
Strictly speaking, validation should refer to an „analytical system‟ rather than an
„analytical method‟, the analytical system comprising a defined method protocol, a
defined concentration range for the analyte, and a specified type of test material.
For the purposes of this document, a reference to „method validation‟ will be taken
as referring to an analytical system as a whole. Where the analytical procedure as
such is addressed, it will be referred to as „the protocol‟.
In this document method validation is regarded as distinct from ongoing activities
such as internal quality control (IQC) or proficiency testing. Method validation is
carried out once, or at relatively infrequent intervals during the working lifetime of
a method; it tells us what performance we can expect the method to provide in the
future. Internal quality control tells us about how the method has performed in the
past. IQC is therefore treated as a separate activity in the IUPAC Harmonisation
Programme.3
In method validation the quantitative characteristics of interest relate to the
accuracy of the result likely to be obtained. Therefore it is generally true to say
that method validation is tantamount to the task of estimating uncertainty of
measurement. Over the years it has become traditional for validation purposes to
represent different aspects of method performance by reference to the separate
items listed above, and to a considerable extent these guidelines reflect that
pattern. However, with an increasing reliance on measurement uncertainty as a
key indicator of both fitness for purpose and reliability of results, analytical
chemists will increasingly undertake measurement validation to support
uncertainty estimation, and some practitioners will want to do so immediately.
Accordingly, measurement uncertainty is treated briefly in Annex A as a
performance characteristic of an analytical method, while Annex B provides
additional guidance on some procedures not otherwise covered.
OIV-MA-AS1-13 : R2005
8
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
4
BASIC PRINCIPLES OF METHOD VALIDATION
4.1
Specification and scope of validation
Validation applies to a defined protocol, for the determination of a specified
analyte and range of concentrations in a particular type of test material, used for a
specified purpose. In general, validation should check that the method performs
adequately for the purpose throughout the range of analyte concentrations and test
materials to which it is applied. It follows that these features, together with a
statement of any fitness-for-purpose criteria, should be completely specified before
any validation takes place.
4.2
Testing assumptions
In addition to the provision of performance figures which indicate fitness for
purpose and have come to dominate the practical use of validation data, validation
studies act as an objective test of any assumptions on which an analytical method
is based. For example, if a result is to be calculated from a simple straight line
calibration function, it is implicitly assumed that the analysis is free from
significant bias, that the response is proportional to analyte concentration, and that
the dispersion of random errors is constant throughout the range of interest. In
most circumstances, such assumptions are made on the basis of experience
accumulated during method development or over the longer term, and are
consequently reasonably reliable. Nonetheless, good measurement science relies
on tested hypotheses. This is the reason that so many validation studies are based
on statistical hypothesis testing; the aim is to provide a basic check that the
reasonable assumptions made about the principles of the method are not seriously
flawed.
There is an important practical implication of this apparently abstruse note. It is
easier to check for gross departure from a reliable assumption than to „prove‟ that
a particular assumption is correct. Thus, where there is long practice of the
successful use of a particular analytical technique (such as gas chromatographic
analysis, or acid digestion methods) across a range of analytes and matrices,
validation checks justifiably take the form of relatively light precautionary tests.
Conversely, where experience is slight, the validation study needs to provide
strong evidence that the assumptions made are appropriate in the particular cases
under study, and it will generally be necessary to study the full range of
circumstances in detail. It follows that the extent of validation studies required in a
given instance will depend, in part, on the accumulated experience of the
analytical technique used.
In the following discussion, it will be taken for granted that the laboratory is well
OIV-MA-AS1-13 : R2005
9
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
practised in the technique of interest, and that the purpose of any significance tests
is to check that there is no strong evidence to discount the assumptions on which
the particular protocol relies. The reader should bear in mind that more stringent
checks may be necessary for unfamiliar or less established measurement
techniques.
4.3
Sources of Error in Analysis
Errors in analytical measurements arise from different sources * and at different
levels of organisation. One useful way of representing these sources (for a specific
concentration of analyte) is as follows+24:





random error of measurement (repeatability);
run bias ;
laboratory bias;
method bias;
matrix variation effect.
Though these different sources may not necessarily be independent, this list
provides a useful way of checking the extent to which a given validation study
addresses the sources of error.
The repeatability (within-run) term includes contributions from any part of the
procedure that varies within a run, including contributions from the familiar
gravimetric and volumetric errors, heterogeneity of the test material, and variation
in the chemical treatment stages of the analysis, and is easily seen in the dispersion
of replicated analyses. The run effect accounts for additional day-to-day variations
in the analytical system, such as changes of analyst, batches of reagents,
recalibration of instruments, and the laboratory environment (e.g., temperature
changes). In single-laboratory validation, the run effect is typically estimated by
*
Sampling uncertainty in the strict sense of uncertainty due to the preparation of the
laboratory sample from the bulk target is excluded from consideration in this document.
Uncertainty associated with taking a test portion from the laboratory sample is an
inseparable part of measurement uncertainty and is automatically included at various levels
of the following analysis.
+
Many alternative groupings or „partitions of error‟ are possible and may be useful in
studying particular sources of error in more detail or across a different range of situations.
For example, the statistical model of ISO 5725 generally combines laboratory and run
effects, while the uncertainty estimation procedure in the ISO GUM is well suited to
assessing the effects of each separate and measurable influence on the result.
OIV-MA-AS1-13 : R2005
10
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
conducting a designed experiment with replicated analysis of an appropriate
material in a number of separate runs. Between-laboratory variation arises from
factors such as variations in calibration standards, differences between local
interpretations of a protocol, changes in equipment or reagent source or
environmental factors, such as differences in average climatic conditions.
Between-laboratory variation is clearly seen as a reality in the results of
collaborative trials (method performance studies) and proficiency tests, and
between-method variation can sometimes be discerned in the results of the latter.
Generally, the repeatability, run effect and laboratory effect are of comparable
magnitude, so none can safely be ignored in validation. In the past there has been a
tendency for aspects to be neglected, particularly when estimating and reporting
uncertainty information. This results in uncertainty intervals that are too tight. For
example, the collaborative trial as normally conducted does not give the complete
picture because contributions to uncertainty from method bias and matrix variation
are not estimated in collaborative trials and have to be addressed separately
(usually by prior single-laboratory study). In single-laboratory validation there is
the particular danger that laboratory bias also may be overlooked, and that item is
usually the largest single contributor to uncertainty from the above list. Therefore
specific attention must be paid to laboratory bias in single-laboratory validation.
In addition to the above-mentioned problems, the validation of a method is limited
to the scope of its application, that is, the method as applied to a particular class of
test material. If there is a substantial variation of matrix types within the defined
class, there will be an additional source of variation due to within-class matrix
effects. Of course, if the method is subsequently used for materials outside the
defined class (that is, outside the scope of the validation), the analytical system
cannot be considered validated: an extra error of unknown magnitude is introduced
into the measurement process.
It is also important for analysts to take account of the way in which method
performance varies as a function of the concentration of the analyte. In most
instances the dispersion of results increases absolutely with concentration and
recovery may differ substantially at high and low concentrations.
The
measurement uncertainty associated with the results is therefore often dependent
on both these effects and on other concentration-dependent factors. Fortunately, it
is often reasonable to assume a simple relationship between performance and
analyte concentration; most commonly that errors are proportional to analyte
concentration.* However, where the performance of the method is of interest at
*
This may not be applicable at concentrations less than 10 times the detection limit.
OIV-MA-AS1-13 : R2005
11
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
substantially different concentrations, it is important to check the assumed
relationship between performance and analyte concentration. This is typically done
by checking performance at extremes of the likely range, or at a few selected
levels. Linearity checks also provide information of the same kind.
4.4
Method and Laboratory effects
It is critically important in single-laboratory method validation to take account of
method bias and laboratory bias. There are a few laboratories with special facilities
where these biases can be regarded as negligible, but that circumstance is wholly
exceptional. (However, that if there is only one laboratory carrying out a particular
analysis, then method bias and laboratory bias take on a different perspective).
Normally, method and laboratory effects have to be included in the uncertainty
budget, but often they are more difficult to address than repeatability error and the
run effect. In general, to assess the respective uncertainties it is necessary to use
information gathered independently of the laboratory. The most generally useful
sources of such information are (i) statistics from collaborative trials (not available
in many situations of single-laboratory method validation), (ii) statistics from
proficiency tests and (iii) results from the analysis of certified reference materials.
Collaborative trials directly estimate the variance of between-laboratory biases.
While there may be theoretical shortcomings in the design of such trials, these
variance estimates are appropriate for many practical purposes. Consequently it is
always instructive to test single-laboratory validation by comparing the estimates
of uncertainty with reproducibility estimates from collaborative trials. If the singlelaboratory result is substantially the smaller, it is likely that important sources of
uncertainty have been neglected. (Alternatively, it may be that a particular
laboratory in fact works to a smaller uncertainty than found in collaborative trials:
such a laboratory would have to take special measures to justify such a claim.) If
no collaborative trial has been carried out on the particular method/test material
combination, an estimate of the reproducibility standard deviation  H at an
analyte concentration c above about 120 ppb can usually be obtained from the
Horwitz function,  H  0.02c 0.8495 , with both variables expressed as mass
fractions. (The Horwitz estimate is normally within a factor of about two of
observed collaborative study results). It has been observed that the Horwitz
function is incorrect at concentrations lower than about 120 ppb, and a modified
function is more appropriate.21, 25 All of this information may be carried into the
single-laboratory area with minimum change.
Statistics from proficiency tests are particularly interesting because they provide
OIV-MA-AS1-13 : R2005
12
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
information in general about the magnitude of laboratory and method biases
combined and, for the participant, information about total error on specific
occasions. Statistics such as the robust standard deviation of the participants
results for an analyte in a round of the test can in principle be used in a way similar
to reproducibility standard deviations from collaborative trials, i.e., to obtain a
benchmark for overall uncertainty for comparison with individual estimates from
single-laboratory validation. In practice, statistics from proficiency tests may be
more difficult to access, because they are not systematically tabulated and
published like collaborative trials, but only made available to participants. Of
course, if such statistics are to be used they must refer to the appropriate matrix
and concentration of the analyte. Individual participants in proficiency testing
schemes can also gauge the validity of their estimated uncertainty by comparing
their reported results with the assigned values of successive rounds 26. This,
however, is an ongoing activity and therefore not strictly within the purview of
single-laboratory validation (which is a one-off event).
If an appropriate certified reference material is available, a single-laboratory test
allows a laboratory to assess laboratory bias and method bias in combination, by
analysing the CRM a number of times. The estimate of the combined bias is the
difference between the mean result and the certified value.
Appropriate certified reference materials are not always available, so other
materials may perforce have to be used. Materials left over from proficiency tests
sometimes serve this purpose and, although the assigned values of the materials
may have questionable uncertainties, their use certainly provides a check on
overall bias. Specifically, proficiency test assigned values are generally chosen to
provide a minimally biased estimate, so a test for significant bias against such a
material is a sensible practice. A further alternative is to use spiking and recovery
information4 to provide estimates of these biases, although there may be
unmeasurable sources of uncertainty associated with these techniques.
Currently the least recognised effect in validation is that due to matrix variation
within the defined class of test material. The theoretical requirement for the
estimation of this uncertainty component is for a representative collection of test
materials to be analysed in a single run, their individual biases estimated, and the
variance of these biases calculated. (Analysis in a single run means that higher
level biases have no effect on the variance. If there is a wide concentration range
involved, then allowance for the change in bias with concentration must be made.)
If the representative materials are certified reference materials, the biases can be
estimated directly as the differences between the results and the reference values,
and the whole procedure is straightforward. In the more likely event that
insufficient number of certified reference materials are available, recovery tests
with a range of typical test materials may be resorted to, with due caution.
OIV-MA-AS1-13 : R2005
13
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
Currently there is very little quantitative information about the magnitude of
uncertainties from this source, although in some instances they are suspected of
being large.
5
Conduct of Validation Studies
The detailed design and execution of method validation studies is covered
extensively elsewhere and will not be repeated here. However, the main principles
are pertinent and are considered below:
It is essential that validation studies are representative. That is, studies should, as
far as possible, be conducted to provide a realistic survey of the number and range
of effects operating during normal use of the method, as well as to cover the
concentration ranges and sample types within the scope of the method. Where a
factor (such as ambient temperature) has varied representatively at random during
the course of a precision experiment, for example, the effects of that factor appear
directly in the observed variance and need no additional study unless further
method optimisation is desirable.
In the context of method validation, “representative variation” means that the
factor must take a distribution of values appropriate to the anticipated range of the
parameter in question. For continuous measurable parameters, this may be a
permitted range, stated uncertainty or expected range; for discontinuous factors, or
factors with unpredictable effects such as sample matrix, a representative range
corresponds to the variety of types or “factor levels” permitted or encountered in
normal use of the method. Ideally, representativeness extends not only to the range
of values, but to their distribution. Unfortunately, it is often uneconomic to arrange
for full variation of many factors at many levels. For most practical purposes,
however, tests based on extremes of the expected range, or on larger changes than
anticipated, are an acceptable minimum.
In selecting factors for variation, it is important to ensure that the larger effects are
„exercised‟ as much as possible. For example, where day to day variation (perhaps
arising from recalibration effects) is substantial compared to repeatability, two
determinations on each of five days will provide a better estimate of intermediate
precision than five determinations on each of two days. Ten single determinations
on separate days will be better still, subject to sufficient control, though this will
provide no additional information on within-day repeatability.
Clearly, in planning significance checks, any study should have sufficient power to
detect such effects before they become practically important (that is, comparable
OIV-MA-AS1-13 : R2005
14
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
to the largest component of uncertainty).
In addition, the following considerations may be important:
 Where factors are known or suspected to interact, it is important to ensure that
the effect of interaction is accounted for. This may be achieved either by
ensuring random selection from different levels of interacting parameters, or by
careful systematic design to obtain „interaction‟ effects or covariance
information.
 In carrying out studies of overall bias, it is important that the reference
materials and values are relevant to the materials under routine test.
6
Extent of validation studies
The extent to which a laboratory has to undertake validation of a new, modified or
unfamiliar method depends to a degree on the existing status of the method and the
competence of the laboratory. Suggestions as to the extent of validation and
verification measures for different circumstances are given below. Except where
stated, it is assumed that the method is intended for routine use.
6.1
The laboratory is to use a “fully” validated method
The method has been studied in a collaborative trial and so the laboratory has to
verify that it is capable of achieving the published performance characteristics of
the method (or is otherwise able to fulfil the requirements of the analytical task).
The laboratory should undertake precision studies, bias studies (including matrix
variation studies), and possibly linearity studies, although some tests such as that
for ruggedness may be omitted.
6.2
The laboratory is to use a fully validated method, but new matrix is to
be used
The method has been studied in a collaborative trial and so the laboratory has to
verify that the new matrix introduces no new sources of error into the system. The
same range of validation as the previous is required.
6.3
The laboratory is to use a well-established, but not collaboratively
studied, method
OIV-MA-AS1-13 : R2005
15
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
The same range of validation as the previous is required.
6.4
The method has been published in the scientific literature together
with some analytical characteristics
The laboratory should undertake precision studies, bias studies (including matrix
variation studies), ruggedness and linearity studies.
6.5
The method has been published in the scientific literature with no
characteristics given or has been developed in-house
The laboratory should undertake precision studies, bias studies (including matrix
variation studies), ruggedness and linearity studies.
6.6
The method is empirical
An empirical method is one in which the quantity estimated is simply the result
found on following the stated procedure. This differs from measurements intended
to assess method-independent quantities such as the concentration of a particular
analyte in a sample, in that the method bias is conventionally zero, and matrix
variation (that is , within the defined class) is irrelevant. Laboratory bias cannot be
ignored, but is likely to be difficult to estimate by single-laboratory experiment.
Moreover, reference materials are unlikely to be available. In the absence of
collaborative trial data some estimate of interlaboratory precision could be
obtained from a specially designed ruggedness study or estimated by using the
Horwitz function.
6.7
The analysis is “ad hoc”
“Ad hoc” analysis is occasionally necessary to establish the general range of a
value, without great expenditure and with low criticality. The effort that can go
into validation is accordingly strictly limited. Bias should be studied by methods
such as recovery estimation or analyte additions, and precision by replication.
6.8
Changes in staff and equipment
Important examples include: change in major instruments; new batches of very
variable reagents (for example, polyclonal antibodies); changes made in the
laboratory premises; methods used for the first time by new staff; or a validated
method employed after a period of disuse. Here the essential action is to
demonstrate that no deleterious changes have occurred. The minimum check is a
OIV-MA-AS1-13 : R2005
16
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
single bias test; a “before and after” experiment on typical test materials or control
materials. In general, the tests carried out should reflect the possible impact of the
change on the analytical procedure.
7
RECOMMENDATIONS
The following recommendations are made regarding the use of single-laboratory
method validation:
 Wherever possible and practical a laboratory should use a method of analysis
that has had its performance characteristics evaluated through a collaborative
trial conforming to an international protocol.
 Where such methods are not available, a method must be validated in-house
before being used to generate analytical data for a customer.
 Single-laboratory validation requires the laboratory to select appropriate
characteristics for evaluation from the following: applicability, selectivity,
calibration, accuracy, precision, range, limit of quantification, limit of
detection, sensitivity, ruggedness and practicability. The laboratory must take
account of customer requirements in choosing which characteristics are to be
determined.
 Evidence that these characteristics have been assessed must be made available
to customers of the laboratory if required by the customer.
8
REFERENCES
1. "Protocol for the Design, Conduct and Interpretation of Method Performance
Studies", W Horwitz, Pure Appl. Chem., 1988, 60, 855 864, revised W.
Horwitz, Pure Appl. Chem., 1995, 67, 331-343.
2. “The International Harmonised Protocol for the Proficiency Testing of
(Chemical) Analytical Laboratories”, M Thompson and R Wood, Pure Appl.
Chem., 1993, 65, 2123-2144. (Also published in J. AOAC International, 1993,
76, 926-940.
3. “Harmonised Guidelines For Internal Quality Control in Analytical Chemistry
Laboratories”, Michael Thompson and Roger Wood, J. Pure & Applied
Chemistry, 1995, 67(4), 49-56.
4. “Harmonised Guidelines for the Use of Recovery Information in Analytical
Measurement”, Michael Thompson, Stephen Ellison, Ales Fajgelj, Paul
Willetts and Roger Wood, J. Pure & Applied Chemistry, 1999, 71(2), 337-348.
OIV-MA-AS1-13 : R2005
17
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
5. “Council Directive 93/99/EEC on the Subject of Additional Measures
Concerning the Official Control of Foodstuffs”, O. J., 1993, L290.
6. “Procedural Manual of the Codex Alimentarius Commission, 10th Edition”,
FAO, Rome, 1997.
7. “Precision of Test Methods”, Geneva, 1994, ISO 5725, Previous editions were
issued in 1981 and 1986.
8. “Guide to the Expression of Uncertainty in Measurement”, ISO, Geneva,
1993.
9. “Quantifying Uncertainty in Analytical Measurement”, EURACHEM
Secretariat, Laboratory of the Government Chemist, Teddington, UK, 1995,
EURACHEM Guide (under revision).
10. “International vocabulary of basic and general terms in metrology” ISO, Geneva
1993
11. “Validation of Chemical Analytical Methods”, NMKL Secretariat, Finland, 1996,
NMKL Procedure No. 4.
12. “EURACHEM Guide: The fitness for purpose of analytical methods. A
Laboratory Guide to method validation and related topics”, LGC, Teddington
1996. Also available from the EURACHEM Secretariat and website.
13. “Statistics manual of the AOAC”, AOAC INTERNATIONAL, Gaithersburg,
Maryland, USA, 1975
14. “An Interlaboratory Analytical Method Validation Short Course developed by
the AOAC INTERNATIONAL”, AOAC INTERNATIONAL, Gaithersburg,
Maryland, USA, 1996.
15. “Text on validation of analytical procedures” International Conference on
Harmonisation. Federal Register, Vol. 60, March 1, 1995, pages 11260
16. “Validation of analytical procedures: Methodology” International Conference
on Harmonisation. Federal Register, Vol. 62, No. 96, May 19, 1997, pages
27463-27467.
OIV-MA-AS1-13 : R2005
18
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
17. “Validation of Methods”, Inspectorate for Health Protection, Rijswijk, The
Netherlands, Report 95-001.
18. “A Protocol for Analytical Quality Assurance in Public Analysts’
Laboratories”, Association of Public Analysts, 342 Coleford Road, Sheffield
S9 5PH, UK, 1986.
19. “Validation of Analytical Methods for Food Control”, Report of a Joint
FAO/IAEA Expert Consultation, December 1997, FAO Food and Nutrition
Paper No. 68, FAO, Rome, 1998
20. “Estimation and Expression of Measurement Uncertainty in Chemical Analysis”,
NMKL Secretariat, Finland, 1997, NMKL Procedure No. 5.
21. M Thompson, PJ Lowthian, J AOAC Int, 1997, 80, 676-679
22. IUPAC recommendation: “Nomenclature in evaluation of analytical methods,
including quantification and detection capabilities” Pure and Applied Chem.”
1995, 67 1699-1723
23. ISO 11843. “Capability of detection.” (Several parts). International Standards
Organisation, Geneva.
24. M. Thompson, Analyst, 2000, 125, 2020-2025
25. “Recent trends in inter-laboratory precision at ppb and sub-ppb
concentrations in relation to fitness for purpose criteria in proficiency
testing” M Thompson, Analyst, 2000, 125, 385-386.
26. “How to combine proficiency test results with your own uncertainty estimate the zeta score”, Analytical Methods Committee of the Royal Society of
Chemistry, AMC Technical Briefs, editor M. Thompson, AMC Technical Brief
No. 2, www.rsc.org/lap/rsccom/amc
OIV-MA-AS1-13 : R2005
19
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
ANNEX A: Notes on the requirements for study of method
performance characteristics
The general requirements for the individual performance characteristics for a
method are as follows.
A1 Applicability
After validation the documentation should provide, in addition to any performance
specification, the following information:





the identity of the analyte, including speciation where appropriate
(Example: „total arsenic‟;
the concentration range covered by the validation (Example: „0-50 ppm‟);
a specification of the range of matrices of the test material covered by the
validation (Example: „seafood‟);
a protocol, describing the equipment, reagents, procedure (including
permissible variation in specified instructions, e.g., „heat at 100  5 for 30
 5 minutes‟), calibration and quality procedures, and any special safety
precautions required;
the intended application and its critical uncertainty requirements
(Example: „The analysis of food for screening purposes. The standard
uncertainty u(c) of the result c should be less than 0.1c.‟).
A2 Selectivity
Selectivity is the degree to which a method can quantify the analyte accurately in
the presence of interferents. Ideally, selectivity should be evaluated for any
important interferent likely to be present. It is particularly important to check
interferents which are likely, on chemical principles, to respond to the test. For
example, colorimetric tests for ammonia might reasonably be expected to respond
to primary aliphatic amines. It may be impracticable to consider or test every
potential interferent; where that is the case, it is recommended that the likely worst
cases are checked. As a general principle, selectivity should be sufficiently good
for any interferences to be ignored.
In many types of analysis, selectivity is essentially a qualitative assessment based
on the significance or otherwise of suitable tests for interference. However, there
are useful quantitative measures. In particular, one quantitative measure is the
selectivity index ban/bint, where ban is the sensitivity of the method (slope of the
OIV-MA-AS1-13 : R2005
20
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
calibration function) and bint the slope of the response independently produced by
a potential interferent, provides a quantitative measure of interference. bint can be
determined approximately by execution of the procedure on a matrix blank and the
same blank spiked with the potential interferent at one appropriate concentration.
If a matrix blank is unavailable, and a typical material used instead, bint can be
estimated from such a simple experiment only under the assumption that mutual
matrix effects are absent. Note that bint is more easily determined in the absence of
the analyte because the effect might be confused with another type of interference
when the sensitivity of the analyte is itself affected by the interferent (a matrix
effect).
A3 Calibration and linearity
With the exception of gross errors in preparation of calibration materials,
calibration errors are usually (but not always) a minor component of the total
uncertainty budget, and can usually be safely subsumed into various categories
estimated by “top-down” methods. For example random errors resulting from
calibration are part of the run bias, which is assessed as a whole, while systematic
errors from that source may appear as laboratory bias, likewise assessed as a
whole. Never-the-less, there are some characteristics of calibration that are useful
to know at the outset of method validation, because they affect the strategy for the
optimal development of the procedure. In this class are such questions as whether
the calibration function plausibly (a) is linear, (b) passes through the origin and (c)
is unaffected by the matrix of the test material. The procedures described here
relate to calibration studies in validation, which are necessarily more exacting than
calibration undertaken during routine analysis. For example, once it is established
at validation that a calibration function is linear and passes through the origin, a
much simpler calibration strategy can be used for routine use (for example, a two
point repeated design). Errors from this simpler calibration strategy will normally
be subsumed into higher level errors for validation purposes.
A3.1 Linearity and intercept
Linearity can be tested informally by examination of a plot of residuals produced
by linear regression of the responses on the concentrations in an appropriate
calibration set. Any curved pattern suggests lack of fit due to a non-linear
calibration function. A test of significance can be undertaken by comparing the
lack-of-fit variance with that due to pure error. However, there are causes of lack
of fit other than nonlinearity that can arise in certain types of analytical calibration,
so the significance test must be used in conjunction with a residual plot. Despite its
current widespread use as an indication of quality of fit, the correlation coefficient
OIV-MA-AS1-13 : R2005
21
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
is misleading and inappropriate as a test for linearity and should not be used.
Design is all-important in tests for lack of fit, because it is easy to confound
nonlinearity with drift. Replicate measurements are needed to provide an estimate
of pure error if there is no independent estimate. In the absence of specific
guidance, the following should apply:




there should be six or more calibrators;
the calibrators should be evenly spaced over the concentration range of
interest;
the range should encompass 0-150% or 50-150% of the concentration
likely to be encountered, depending on which of these is the more suitable;
the calibrators should be run at least in duplicate, and preferably triplicate
or more, in a random order.
After an exploratory fit with simple linear regression, the residuals should be
examined for obvious patterns. Heteroscedasticity is quite common in analytical
calibration and a pattern suggesting it means that the calibration data are best
treated by weighted regression. Failure to use weighted regression in these
circumstances could give rise to exaggerated errors at the low end of the
calibration function.
The test for lack of fit can be carried out with either simple or weighted regression.
A test for an intercept significantly different from zero can also be made on this
data if there is no significant lack of fit.
A3.2 Test for general matrix effect
It simplifies calibration enormously if the calibrators can be prepared as a simple
solution of the analyte. The effects of a possible general matrix mismatch must be
assessed in validation if this strategy is adopted. A test for general matrix effect
can be made by applying the method of analyte additions (also called “standard
additions”) to a test solution derived from a typical test material. The test should
be done in a way that provides the same final dilution as the normal procedure
produces, and the range of additions should encompass the same range as the
procedure-defined calibration validation. If the calibration is linear the slopes of
the usual calibration function and the analyte additions plot can be compared for
significant difference. A lack of significance means that there is no detectable
general matrix effect. If the calibration is not linear a more complex method is
needed for a significance test, but a visual comparison at equal concentrations will
usually suffice. A lack of significance in this test will often mean that the matrix
OIV-MA-AS1-13 : R2005
22
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
variation effect [Section A13] will also be absent.
A3.3 Final calibration procedure
The calibration strategy as specified in the procedure may also need to be
separately validated, although the errors involved will contribute to jointly
estimated uncertainties. The important point here is that evaluation uncertainty
estimated from the specific designs for linearity etc., will be smaller than those
derived from the simpler calibration defined in the procedure protocol.
A4 Trueness
A4.1 Estimation of trueness
Trueness is the closeness of agreement between a test result and the accepted
reference value of the property being measured. Trueness is stated quantitatively in
terms of “bias”; with smaller bias indicating greater trueness. Bias is typically
determined by comparing the response of the method to a reference material with
the known value assigned to the material. Significance testing is recommended.
Where the uncertainty in the reference value is not negligible, evaluation of the
results should consider the reference material uncertainty as well as the statistical
variability.
A4.2 Conditions for trueness experiments
Bias can arise at different levels of organisation in an analytical system, for
example, run bias, laboratory bias and method bias. It is important to remember
which of these is being handled by the various methods of addressing bias. In
particular:


The mean of a series of analyses of a reference material, carried out
wholly within a single run, gives information about the sum of method,
laboratory and run effect for that particular run. Since the run effect is
assumed to be random from run to run, the result will vary from run to run
more than would be expected from the observable dispersion of the results,
and this needs to be taken into account in the evaluation of the results (for
example, by testing the measured bias against the among-runs standard
deviation investigated separately).
The mean of repeated analyses of a reference material in several runs,
estimates the combined effect of method and laboratory bias in the
OIV-MA-AS1-13 : R2005
23
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
particular laboratory (except where the value is assigned using the
particular method).
A4.3 Reference values for trueness experiments
A4.3.1 Certified reference materials (CRMs)
CRMs are traceable to international standards with a known uncertainty and
therefore can be used to address all aspects of bias (method, laboratory and withinlaboratory) simultaneously, assuming that there is no matrix mismatch. CRMs
should accordingly be used in validation of trueness where it is practicable to do
so. It is important to ensure that the certified value uncertainties are sufficiently
small to permit detection of a bias of important magnitude. Where they are not, the
use of CRMs is still recommended, but additional checks should be carried out.
A typical trueness experiment generates a mean response on a reference material.
In interpreting the result, the uncertainty associated with the certified value should
be taken into account along with the uncertainty arising from statistical variation in
the laboratory. The latter term may be based on the within-run, between-run, or an
estimate of the between-laboratory standard deviation depending on the intent of
the experiment. Statistical or materials. Where the certified value uncertainty is
small, a Student‟s t test is normally carried out, using the appropriate precision
term.
Where necessary and practicable, a number of suitable CRMs, with appropriate
matrices and analyte concentrations, should be examined. Where this is done, and
the uncertainties on the certified values are smaller than those on the analytical
results, it would be reasonably safe to use simple regression to evaluate the results.
In this way bias could be expressed as a function of concentration, and might
appear as a non-zero intercept (“transitional” or constant bias) or as a non-unity
slope (“rotational” or proportional bias). Due caution should be applied in
interpreting the results where the range of matrices is large.
4.3.2 Reference materials
Where CRMs are not available, or as an addition to CRMs, use may be made of
any material sufficiently well characterised for the purpose (a reference
material10), bearing in mind always that while insignificant bias may not be proof
of zero bias, significant bias on any material remains a cause for investigation.
Examples of reference materials include: Materials characterised by a reference
material producer, but whose values are not accompanied by an uncertainty
statement or are otherwise qualified; materials characterised by a manufacturer of
OIV-MA-AS1-13 : R2005
24
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
the material; materials characterised in the laboratory for use as reference
materials; materials subjected to a restricted round-robin exercise, or distributed in
a proficiency test. While the traceability of these materials may be questionable, it
would be far better to use them than to conduct no assessment for bias at all. The
materials would be used in much the same way as CRMs, though with no stated
uncertainty any significance test relies wholly on the observable precision of
results.
A4.3.3 Use of a reference method
A reference method can in principle be used to test for bias in another method
under validation. This is a useful option when checking an alternative to, or
modification of, an established standard method already validated and in use in the
laboratory. Both methods are used to analyse a number of typical test materials,
preferably covering a useful range of concentration fairly evenly. Comparison of
the results over the range by a suitable statistical method (for example, a paired ttest, with due checks for homogeneity of variance and normality) would
demonstrate any bias between the methods.
A4.3.4 Use of spiking/recovery
In the absence of reference materials, or to support reference material studies, bias
can be investigated by spiking and recovery. A typical test material is analysed by
the method under validation both in its original state and after the addition
(spiking) of a known mass of the analyte to the test portion. The difference
between the two results as a proportion of the mass added is called the surrogate
recovery or sometimes the marginal recovery. Recoveries significantly different
from unity indicate that a bias is affecting the method. Strictly, recovery studies as
described here only assess bias due to effects operating on the added analyte; the
same effects do not necessarily apply to the same extent to the native analyte, and
additional effects may apply to the native analyte. Spiking/recovery studies are
accordingly very strongly subject to the observation that while good recovery is
not a guarantee of trueness, poor recovery is certainly an indication of lack of
trueness. Methods of handling spiking/recovery data have been covered in detail
elsewhere.4
A5 Precision
Precision is the closeness of agreement between independent test results obtained
under stipulated conditions. It is usually specified in terms of standard deviation or
relative standard deviation. The distinction between precision and bias is
OIV-MA-AS1-13 : R2005
25
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
fundamental, but depends on the level at which the analytical system is viewed.
Thus from the viewpoint of a single determination, any deviation affecting the
calibration for the run would be seen as a bias. From the point of view of the
analyst reviewing a year‟s work, the run bias will be different every day and act
like a random variable with an associated precision. The stipulated conditions for
the estimation of precision take account of this change in view point.
For single laboratory validation, two sets of conditions are relevant: (a) precision
under repeatability conditions, describing variations observed during a single run
as expectation 0 and standard deviation  r , and (b) precision under run-to-run
conditions, describing variations in run bias run as expectation 0, standard
deviation  run . Usually both of these sources of error are operating on individual


12
2
analytical results, which therefore have a combined precision  tot   r2 n   run
,
where n is the number of repeat results averaged within a run for the reported
result. The two precision estimates can be obtained most simply by analysing the
selected test material in duplicate in a number of successive runs. The separate
variance components can then be calculated by the application of one-way analysis
of variance. Each duplicate analysis must be an independent execution of the
procedure applied to a separate test portion. Alternatively the combined precision
 tot can be estimated directly by the analysis of the test material once in
successive runs, and estimating the standard deviation from the usual equation.
(Note that observed standard deviations are generally given the symbol s, to
distinguish them from standard deviations ).
It is important that the precision values are representative of likely test conditions.
First, the variation in conditions among the runs must represent what would
normally happen in the laboratory under routine use of the method. For instance,
variations in reagent batches, analysts and instruments should be representative.
Second, the test material used should be typical, in terms of matrix and (ideally)
the state of comminution, of the materials likely to encountered in routine
application. So actual test materials or, to a lesser degree, matrix-matched
reference materials would be suitable, but standard solutions of the analyte would
not. Note also that CRMs and prepared reference materials are frequently
homogenised to a greater extent than typical test materials, and precision obtained
from their analysis may accordingly under-estimate the variation that will be
observed for test materials.
Precision very often varies with analyte concentration. Typical assumptions are i)
that there is no change in precision with analyte level, or ii) that the standard
deviation is proportional to, or linearly dependent on, analyte level. In both cases,
OIV-MA-AS1-13 : R2005
26
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
the assumption needs to be checked if the analyte level is expected to vary
substantially (that is, by more than about 30% from its central value). The most
economical experiment is likely to be a simple assessment of precision at or near
the extremes of the operating range, together with a suitable statistical test for
difference in variance. The F-test is appropriate for normally distributed error.
Precision data may be obtained for a wide variety of different sets of conditions in
addition to the minimum of repeatability and between-run conditions indicated
here, and it may be appropriate to acquire additional information. For example, it
may be useful to the assessment of results, or for improving the measurement, to
have an indication of separate operator and run effects, between or within-day
effects or the precision attainable using one or several instruments. A range of
different designs and statistical analysis techniques is available, and careful
experimental design is strongly recommended in all such studies.
A6 Recovery
Methods for estimating recovery are discussed in conjunction with methods of
estimating trueness (above).
A7 Range
The validated range is the interval of analyte concentration within which the
method can be regarded as validated. It is important to realise that this range is not
necessarily identical to the useful range of the calibration. While the calibration
may cover a wide concentration range, the remainder of the validation (and usually
much more important part in terms of uncertainty) will cover a more restricted
range. In practice, most methods will be validated at only one or two levels of
concentration. The validated range may be taken as a reasonable extrapolation
from these points on the concentration scale.
When the use of the method focuses on a concentration of interest well above the
detection limit, validation near that one critical level would be appropriate. It is
impossible to define a general safe extrapolation of this result to other
concentrations of the analyte, because much depends on the individual analytical
system. Therefore the validation study report should state the range around the
critical value in which the person carrying out the validation, using professional
judgement, regards the estimated uncertainty to hold true.
When the concentration range of interest approaches zero, or the detection limit, it
is incorrect to assume either constant absolute uncertainty or constant relative
OIV-MA-AS1-13 : R2005
27
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
uncertainty. A useful approximation in this common circumstance is to assume a
linear functional relationship, with a positive intercept, between uncertainty u and
concentration c, that is of the form
u(c)  u0   c
where  is the relative uncertainty estimated a some concentration well above the
detection limit. u0 is the standard uncertainty estimated for zero concentration and
in some circumstances could be estimated as c L / 3 . In these circumstances it
would be reasonable to regard the validated range as extending from zero to a
small integer multiple of the upper validation point. Again this would depend on
professional judgement.
A8 Detection Limit
In broad terms the detection limit (limit of detection) is the smallest amount or
concentration of analyte in the test sample that can be reliably distinguished from
zero.22,23 For analytical systems where the validation range does not include or
approach it, the detection limit does not need to be part of a validation.
Despite the apparent simplicity of the idea, the whole subject of the detection limit
is beset with problems outlined below:





There are several possible conceptual approaches to the subject, each
providing a somewhat different definition of the limit. Attempts to clarify
the issue seem ever more confusing.
Although each of these approaches depends on an estimate of precision at
or near zero concentration, it is not clear whether this should be taken as
implying repeatability conditions or some other condition for the
estimation.
Unless an inordinate amount of data is collected, estimates of detection
limit will be subject to quite large random variation.
Estimates of detection limit are often biased on the low side because of
operational factors.
Statistical inferences relating to the detection limit depend on the
assumption of normality, which is at least questionable at low
concentrations.
For most practical purposes in method validation, it seems better to opt for a
OIV-MA-AS1-13 : R2005
28
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
simple definition, leading to a quickly implemented estimation which is used only
as a rough guide to the utility of the method. However, it must be recognised that
the detection limit as estimated in method development, may not be identical in
concept or numerical value to one used to characterise a complete analytical
method. For instance the “instrumental detection limit”, as quoted in the literature
or in instrument brochures and then adjusted for dilution, is often far smaller than
a “practical” detection limit and inappropriate for method validation.
It is accordingly recommended that for method validation, the precision estimate
used ( ˆ 0 ) should be based on at least 6 independent complete determinations of
analyte concentration in a typical matrix blank or low-level material, with no
censoring of zero or negative results, and the approximate detection limit
calculated as 3ˆ 0 . Note that with the recommended minimum number of degrees
of freedom, this value is quite uncertain, and may easily be in error by a factor of
two. Where more rigorous estimates are required (for example to support decisions
based on detection or otherwise of a material), reference should be made to
appropriate guidance (see, for example, references 22-23).
A9 Limit of determination or limit of quantification
It is sometimes useful to state a concentration below which the analytical method
cannot operate with an acceptable precision. Sometimes that precision is arbitrarily
defined as 10% RSD, sometimes the limit is equally arbitrarily taken as a fixed
multiple (typically 2) of the detection limit. While it is to a degree reassuring to
operate above such a limit, we must recognise that it is a quite artificial dichotomy
of the concentration scale: measurements below such a limit are not devoid of
information content and may well be fit for purpose. Hence the use of this type of
limit in validation is not recommended here. It is preferable to try to express the
uncertainty of measurement as a function of concentration and compare that
function with a criterion of fitness for purpose agreed between the laboratory and
the client or end-user of the data.
A10 Sensitivity
The sensitivity of a method is the gradient of the calibration function. As this is
usually arbitrary, depending on instrumental settings, it is not useful in validation.
(It may be useful in quality assurance procedures, however, to test whether an
instrument is performing to a consistent and satisfactory standard.)
OIV-MA-AS1-13 : R2005
29
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
A11 Ruggedness
The ruggedness of an analytical method is the resistance to change in the results
produced by an analytical method when minor deviations are made from the
experimental conditions described in the procedure. The limits for experimental
parameters should be prescribed in the method protocol (although this has not
always been done in the past), and such permissible deviations, separately or in
any combination, should produce no meaningful change in the results produced. (A
“meaningful change” here would imply that the method could not operate within
the agreed limits of uncertainty defining fitness for purpose.) The aspects of the
method which are likely to affect results should be identified, and their influence
on method performance evaluated by using ruggedness tests.
The ruggedness of a method is tested by deliberately introducing small changes to
the procedure and examining the effect on the results. A number of aspects of the
method may need to be considered, but because most of these will have a
negligible effect it will normally be possible to vary several at once. An
economical experiment based on fractional factorial designs has been described by
Youden13. For instance, it is possible to formulate an approach utilising 8
combinations of 7 variable factors, that is to look at the effects of seven parameters
with just eight analytical results. Univariate approaches are also feasible, where
only one variable at a time is changed.
Examples of the factors that a ruggedness test could address are: changes in the
instrument, operator, or brand of reagent; concentration of a reagent; pH of a
solution; temperature of a reaction; time allowed for completion of a process etc.
A12 Fitness for Purpose
Fitness for purpose is the extent to which the performance of a method matches the
criteria, agreed between the analyst and the end-user of the data, that describe the
end-user‟s needs. For instance the errors in data should not be of a magnitude that
would give rise to incorrect decisions more often than a defined small probability,
but they should not be so small that the end-user is involved in unnecessary
expenditure. Fitness for purpose criteria could be based on some of the
characteristics described in this Annex, but ultimately will be expressed in terms
of acceptable total uncertainty.
A13 Matrix variation
Matrix variation is, in many sectors, one of the most important but least
acknowledged sources of error in analytical measurements. When we define the
OIV-MA-AS1-13 : R2005
30
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
analytical system to be validated by specifying, amongst other things, the matrix of
the test material, there may be scope for considerable variation within the defined
class. To cite an extreme example, a sample of the class “soil” could be composed
of clay, sand, chalk, laterite (mainly Fe2O3 and Al2O3), peat, etc., or of mixtures of
these. It is easy to imagine that each of these types would contribute a unique
matrix effect on an analytical method such as atomic absorption spectrometry. If
we have no information about the type of soils we are analysing, there will be an
extra uncertainty in the results because of this variable matrix effect.
Matrix variation uncertainties need to be quantified separately, because they are
not taken into account elsewhere in the process of validation. The information is
acquired by collecting a representative set of the matrices likely to be encountered
within the defined class, all with analyte concentrations in the appropriate range.
The material are analysed according to the protocol, and the bias in the results
estimated. Unless the test materials are CRMs, the bias estimate will usually have
to be undertaken by means of spiking and recovery estimation. The uncertainty is
estimated by the standard deviation of the biases. (Note: This estimate will also
contain a variance contribution from the repeat analysis. This will have a
magnitude 2 2r if spiking has been used. If a strict uncertainty budget is required,
this term should be deducted from the matrix variation variance to avoid double
accounting.)
A14 Measurement Uncertainty
The formal approach to measurement uncertainty estimation calculates a
measurement uncertainty estimate from an equation, or mathematical model. The
procedures described as method validation are designed to ensure that the equation
used to estimate the result, with due allowance for random errors of all kinds, is a
valid expression embodying all recognised and significant effects upon the result.
It follows that, with one caveat elaborated further below, the equation or „model‟
subjected to validation may be used directly to estimate measurement uncertainty.
This is done by following established principles, based on the „law of propagation
of uncertainty‟ which, for independent input effects is
u(y(x1,x2,...)) =
 ci2u( xi ) 2
i 1,n
where y(x1,x2,....xn) is a function of several independent variables x1,x2..., and ci is a
sensitivity coefficient evaluated as ci=y/xi, the partial differential of y with
respect to xi. u(xi) and u(y) are standard uncertainties, that is, measurement
OIV-MA-AS1-13 : R2005
31
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
uncertainties expressed in the form of standard deviations. Since u(y(x1,x2,...)) is a
function of several separate uncertainty estimates, it is referred to as a combined
standard uncertainty.
To estimate measurement uncertainty from the equation y=f(x1,x2...) used to
calculate the result, therefore, it is necessary first, to establish the uncertainties
u(xi) in each of the terms x1, x2 etc. and second, to combine these with the
additional terms required to represent random effects as found in validation, and
finally to take into account any additional effects. In the discussion of precision
above, the implied statistical model is
y=f(x1,x2...) + run + e
where e is the random error for a particular result. Since run and e are known, from
the precision experiments, to have standard deviations  run and  r respectively,
these latter terms (or, strictly, their estimates srun and sr) are the uncertainties
associated with these additional terms. Where the individual within-run results are
averaged, the combined uncertainty associated with these two terms is (as given


12
2
. Note that where the precision terms are shown to
previously) stot  s r2 n  s run
vary with analyte level, the uncertainty estimate for a given result must employ the
precision term appropriate to that level. The basis for the uncertainty estimate
accordingly follows directly from the statistical model assumed and tested in
validation. To this estimate must be added any further terms as necessary to
account for (in particular) inhomogeneity and matrix effect (see section A13).
Finally, the calculated standard uncertainty is multiplied by a „coverage factor‟, k,
to provide an expanded uncertainty, that is, “an interval expected to encompass a
large fraction of the distribution of values that may be attributed to the
measurand”8. Where the statistical model is well established, the distribution
known to be normal, and the number of degrees of freedom associated with the
estimate is high, k is generally chosen to be equal to 2. The expanded uncertainty
then corresponds approximately to a 95% confidence interval.
There is one important caveat to be added here. In testing the assumed statistical
model, imperfect tests are perforce used. It has already been noted that these tests
can not prove that any effect is identically zero; they can only show that an effect
is too small to detect within the uncertainty associated with the particular test for
significance. A particularly important example is the test for significant laboratory
bias. Clearly, if this is the only test performed to confirm trueness, there must be
some residual uncertainty as to whether the method is indeed unbiased or not. It
follows that where such uncertainties are significant with respect to the uncertainty
calculated so far, additional allowance should be made.
OIV-MA-AS1-13 : R2005
32
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
In the case of an uncertain reference value, the simplest allowance is the stated
uncertainty for the material, combined with the statistical uncertainty in the test
applied. A full discussion is beyond the scope of this text; reference 9 provides
further detail. It is, however, important to note that while the uncertainty estimated
directly from the assumed statistical model is the minimum uncertainty that can be
associated with an analytical result, it will almost certainly be an underestimate;
similarly, an expanded uncertainty based on the same considerations and using k=2
will not provide sufficient confidence.
The ISO Guide8 recommends that for increased confidence, rather than arbitrarily
adding terms, the value of k should be increased as required. Practical experience
suggests that for uncertainty estimates based on a validated statistical model, but
with no evidence beyond the validation studies to provide additional confidence in
the model, k should not be less than 3. Where there is strong reason to doubt that
the validation study is comprehensive, k should be increased further as required.
OIV-MA-AS1-13 : R2005
33
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
ANNEX B. Additional considerations for UNCERTAINTY ESTIMATION
IN VALIDATION STUDIES
B1 Sensitivity analysis
The basic expression used in uncertainty estimation
u(y(x1,x2,...)) =
 ci2u( xi ) 2
i 1,n
requires the „sensitivity coefficients‟ ci. It is common in uncertainty estimation to
find that while a given influence factor xi has a known uncertainty u(xi), the
coefficient ci is insufficiently characterised or not readily obtainable from the
equation for the result. This is particularly common where an effect is not included
in the measurement equation because it is not normally significant, or because the
relationship is not sufficiently understood to justify a correction. For example, the
effect of solution temperature Tsol on a room temperature extraction procedure is
rarely established in detail.
Where it is desired to assess the uncertainty in a result associated with such an
effect, it is possible to determine the coefficient experimentally. This is done most
simply by changing xi and observing the effect on the result, in a manner very
similar to basic ruggedness tests. In most cases, it is sufficient in the first instance
to choose at most two values of xi other than the nominal value, and calculate an
approximate gradient from the observed results. The gradient then gives an
approximate value for ci. The term ci.u(xi) can then be determined. (Note that this
is one practical method for demonstrating the significance or otherwise of a
possible effect on the results).
In such an experiment, it is important that the change in result observed be
sufficient for a reliable calculation of ci. This is difficult to predict in advance.
However, given a permitted range for the influence quantity xi, or an expanded
uncertainty for the quantity, that is expected to result in insignificant change, it is
clearly important to assess ci from a larger range. It is accordingly recommended
that for an influence quantity with an expected range of a, (where a might be,
for example, the permitted range, expanded uncertainty interval or 95% confidence
interval) the sensitivity experiment employ, where possible, a change of at least 4a
to ensure reliable results.
OIV-MA-AS1-13 : R2005
34
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Single laboratory validation
B2 Judgement
It is not uncommon to find that while an effect is recognised and may be
significant, it is not always possible to obtain a reliable estimate of uncertainty. In
such circumstances, the ISO Guide makes it quite clear that a professionally
considered estimate of the uncertainty is to be preferred to neglect of the
uncertainty. Thus, where no estimate of uncertainty is available for a potentially
important effect, the analyst should make their own best judgement of the likely
uncertainty and apply that in estimating the combined uncertainty. Reference 8
gives further guidance on the use of judgement in uncertainty estimation.
OIV-MA-AS1-13 : R2005
35
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Recommendations on measurement uncertainty
Recommendations on measurement uncertainty
(Resolution oeno 9/2005)
INTRODUCTION
It is important that analysts are aware of the uncertainty associated with each
analytical result and estimates of uncertainty. The measurement uncertainty may
be derived by a number of procedures. Food analysis laboratories are required to
be in control, use collaboratively tested methods when available, and verify their
application before taking them into routine use. Such laboratories therefore have
available to them a range of analytical data which can be used to estimate their
measurement uncertainty.
Terminology
The accepted definition for Measurement Uncertainty1 is:
“Parameter, associated with the result of a measurement, that characterises the
dispersion of the values that could reasonably be attributed to the measurand.
NOTES:
1.
The parameter may be, for example, a standard deviation (or a given
multiple of it), or the half-width of an interval having a stated level of
confidence.
2.
Uncertainty of measurement comprises, in general, many components.
Some of these components may be evaluated from the statistical
distribution of results of a series of measurements and can be characterised
by experimental standard deviations. The other components, which can
also be characterised by standard deviations, are evaluated from assumed
probability distributions based on experience or other information.
3.
It is understood that the result of a measurement is the best estimate of the
value of a measurand, and that all components of uncertainty, including
those arising from systematic effects. Such as components associated with
corrections and reference standards, contribute to the dispersion.”
[It is recognised that the term “measurement uncertainty” is the most widely used
term by International Organisations and Accreditation Agencies. However The
Codex ALIMENTARIUS Committee on Methods of Analysis and Sampling has
OIV-MA-AS1-14 : R2005
1
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Recommendations on measurement uncertainty
commented on a number of occasions that the term “Measurement Uncertainty”
has some negative associations in legal context and so has noted that an
alternative, equivalent, term, “measurement reliability”, may be used.]
Recommendations
The following recommendations are made to governments:
1.
For OIV purposes the term “measurement uncertainty” or “measurement
reliability” shall be used.
2.
The measurement uncertainty or “measurement reliability” associated with
all analytical results is to be estimated and must, on request be made
available to the user (customer) of the results.
3.
The measurement uncertainty or “measurement reliability” of an analytical
result may be estimated in a number of procedures notably those described
by ISO1 and EURACHEM2. These documents recommend procedures
based on a component-by-component approach, method validation data,
internal quality control data and proficiency test data. The need to
undertake an estimation of the measurement uncertainty or “Measurement
reliability” using the ISO component-by-component approach is not
necessary if the other forms of data are available and used to estimate the
uncertainty or reliability. In many cases the overall uncertainty may be
determined by an inter-laboratory (collaborative) study by a number of
laboratories and a number of matrices by the IUPAC/ISO/AOAC
INTERNATIONAL3 or by the ISO 5725 Protocols4.
OIV-MA-AS1-14 : R2005
2
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Recommendations on measurement uncertainty
REFERENCES
1.
“Guide to the Expression of Uncertainty in Measurement”, ISO, Geneva,
1993.
2.
EURACHEM/CITAC Guide Quantifying Uncertainty In Analytical
Measurement (Second Edition), EURACHEM Secretariat, HAM, Berlin,
2000.
This
is
available
as
a
free
download
from
http://www.vtt.fi/ket/eurachem.
3.
“Protocol for the Design, Conduct and Interpretation of Method
Performance Studies”, ed. W. Horwitz, Pure Appl. Chem., 1995, 67, 331343.
4.
“Precision of Test Methods”, Geneva, 1994, ISO 5725, Previous editions
were published in 1981 and 1986.
OIV-MA-AS1-14 : R2005
3
COMPENDIUM OF INTERNATIONAL ANALYSIS OF METHODS - OIV
Recommendations on recovery correction
Recommendations related to the recovery correction
(Resolution OIV-Oeno 392/2009)
Recovery
“The OIV recommends the following practice with regards to reporting recovery
of analytical results.
o Analytical results are to be expressed on a recovery corrected basis
where appropriate and relevant, and when corrected it has to be stated.
o If a result has been corrected for recovery, the method by which the
recovery was taken into account should be stated. The recovery rate is to
be quoted wherever possible.
o When laying down provisions for standards, it will be necessary to state
whether the result obtained by a method used for analysis within
conformity checks shall be expressed on a recovery-corrected basis or
not.”
OIV-MA-AS1-15 : R2009
1
(All Rights Reserved) According to the law, the OIV declines all
responsibility for involuntary errors or omissions that might occur,
despite care taken in the editing. Reproduction of texts published in
this work is forbidden. These texts are the property of OIV, which
has the exclusive right to reproduce and translate these texts
throughout the world. The law forbids copies and reproductions for
group usage. Any representation or reproduction, whether
complete or partial, by any means, without the accord of the OIV,
is an unlawful act and constitutes an infringement of copyright.
© OIV - 2011
INTERNATIONAL ORGANISATION OF VINE AND WINE
18, rue d’Aguesseau
75008 PARIS
Tél. (33) 01.44.94.80.80 - Tlc. (33) 01.42.66.90.63
E-mail: oiv@oiv.int - www.oiv.int
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising