SW-846 is not intended to be an analytical training manual. Therefore, method
procedures are written based on the assumption that they will be performed by analysts who are
formally trained in at least the basic principles of chemical analysis and in the use of the subject
In addition, SW-846 methods, with the exception of required method use for the analysis
of method-defined parameters, are intended to be guidance methods which contain general
information on how to perform an analytical procedure or technique which a laboratory can use
as a basic starting point for generating its own detailed Standard Operating Procedure (SOP),
either for its own general use or for a specific project application. The performance data
included in this method are for guidance purposes only, and are not intended to be and must
not be used as absolute QC acceptance criteria for purposes of laboratory accreditation.
Metals in solution may be readily determined by graphite furnace atomic
absorption spectrophotometry (GFAA). The method is simple, quick, and applicable to a large
number of metals in environmental samples including, but not limited to, ground water, domestic
and industrial wastes, extracts, soils, sludges, sediments, and similar wastes. With the
exception of the analyses for dissolved constituents, all samples require digestion prior to
analysis. Analysis for dissolved elements does not require digestion if the sample has been
filtered and then acidified.
Organo-metallic species may not be detected if the sample is not digested.
This method is applicable to the following elements:
Chemical Abstract Service Registry Number
7010 - 1
Revision 0
February 2007
Lower limits of quantitation and optimum ranges of the metals will vary with the
matrices and models of atomic absorption spectrophotometers. The data shown in Table 1
provide some indication of the lower limits of quantitation obtainable by the furnace technique.
The limits given in Table 1 are somewhat dependent on equipment (such as the type of
spectrophotometer and furnace accessory, the energy source, the degree of electrical
expansion of the output signal), and are greatly dependent on sample matrix.
Users of this method should state the data quality objectives prior to analysis and
must document and have on file the required initial demonstration performance data described
in the following sections prior to using the method for analysis. When using furnace techniques,
the analyst should be cautioned as to possible chemical reactions occurring at elevated
temperatures which may result in either suppression or enhancement of the analysis element
(see Sec. 4.0). To ensure valid data with furnace techniques, the analyst must examine each
sample for interference effects (see Sec. 9.0) and, if detected, treat them accordingly, using
either successive dilution, matrix modification, or the method of standard additions (see Sec.
Other elements and matrices may be analyzed by this method as long as the
method performance is demonstrated for these additional elements of interest, in the additional
matrices of interest, at the concentration levels of interest in the same manner as the listed
elements and matrices (see Sec. 9.0).
Prior to employing this method, analysts are advised to consult each type of
procedure (e.g., sample preparation methods) that may be employed in the overall analysis for
additional information on quality control procedures, development of QA acceptance criteria,
calculations, and general guidance. Analysts should consult the disclaimer statement at the
front of the manual and the information in Chapter Two for guidance on the intended flexibility in
the choice of methods, apparatus, materials, reagents, and supplies, and on the responsibilities
of the analyst for demonstrating that the techniques employed are appropriate for the analytes
of interest, in the matrix of interest, and at the levels of concern.
In addition, analysts and data users are advised that, except where explicitly specified in a
regulation, the use of SW-846 methods is not mandatory in response to Federal testing
requirements. The information contained in this method is provided by EPA as guidance to be
used by the analyst and the regulated community in making judgments necessary to generate
results that meet the data quality objectives for the intended application.
Use of this method is restricted to use by, or under supervision of, properly
experienced and trained personnel, including analysts who are knowledgeable in the chemical
and physical interferences described in this method. Each analyst must demonstrate the ability
to generate acceptable results with this method.
Although methods have been reported for the analysis of solids by atomic
absorption spectrophotometry, the technique generally is limited to metals in solution or
solubilized through some form of sample processing. Refer to Chapter Three for a description
of appropriate digestion methods.
Preliminary treatment of wastes, both solid and aqueous, is always necessary
because of the complexity and variability of sample matrix. Solids, slurries, and suspended
material must be subjected to a solubilization process before analysis. This process may vary
7010 - 2
Revision 0
February 2007
because of the metals to be determined and the nature of the sample being analyzed. Solubilization and digestion procedures are presented in Chapter Three.
When using the furnace technique in conjunction with an atomic absorption
spectrophotometer, a representative aliquot of a sample is placed in the graphite tube in the
furnace, evaporated to dryness, charred, and atomized. As a greater percentage of available
analyte atoms is vaporized and dissociated for absorption in the tube rather than the flame, the
use of smaller sample volumes or quantitation of lower concentrations of elements is possible.
The principle is essentially the same as with direct aspiration atomic absorption, except that a
furnace, rather than a flame, is used to atomize the sample. Radiation from a given excited
element is passed through the vapor containing ground-state atoms of that element. The
intensity of the transmitted radiation decreases in proportion to the amount of the ground-state
element in the vapor. The metal atoms to be measured are placed in the beam of radiation by
increasing the temperature of the furnace, thereby causing the injected specimen to volatilize.
A monochromator isolates the characteristic radiation from the hollow cathode lamp or
electrodeless discharge lamp, and a photosensitive device measures the attenuated transmitted
Refer to Chapter One, Chapter Three, and the manufacturer's instructions for a definitions
that may be relevant to this procedure.
Solvents, reagents, glassware, and other sample processing hardware may yield
artifacts and/or interferences to sample analysis. All of these materials must be demonstrated
to be free from interferences under the conditions of the analysis by analyzing method blanks.
Specific selection of reagents and purification of solvents by distillation in all-glass systems may
be necessary. Refer to each method to be used for specific guidance on quality control
procedures and to Chapter Three for general guidance on the cleaning of glassware. Also refer
to Method 7000 for a discussion of interferences.
Although the problem of oxide formation is greatly reduced with furnace
procedures (because atomization occurs in an inert atmosphere), the technique is still subject to
chemical interferences. The composition of the sample matrix can have a major effect on the
analysis. It is those effects which must be determined and taken into consideration in the
analysis of each different matrix encountered. See Sec. 9.6 for additional guidance.
Background correction is important when using flameless atomization, especially
below 350 nm. Certain samples, when atomized, may absorb or scatter light from the lamp. This
can be caused by the presence of gaseous molecular species, salt particles, or smoke in the
sample beam. If no correction is made, sample absorbance will be greater than it should be,
and the analytical result will be erroneously high. Zeeman background correction is effective in
overcoming composition or structured background interferences. It is particularly useful when
analyzing for As in the presence of Al and when analyzing for Se in the presence of Fe.
Memory effects occur when the analyte is not totally volatilized during atomization.
This condition depends on several factors -- volatility of the element and its chemical form,
whether pyrolytic graphite is used, the rate of atomization, and furnace design. This situation is
detected through blank burns. The tube should be cleaned by operating the furnace at full
7010 - 3
Revision 0
February 2007
power for the required time period, as needed, at regular intervals during the series of
Gases generated in the furnace during atomization may have molecular absorption
bands encompassing the analytical wavelength. When this occurs, use either background
correction or choose an alternate wavelength. Background correction may also compensate for
nonspecific broad-band absorption interference and light scattering.
Continuum background correction cannot correct for all types of background
interference. When the background interference cannot be compensated for, chemically
remove the analyte or use an alternate form of background correction (see Chapter Two). A
single background correction device to be used with this method is not specified; however, it
must provide an analytical condition that is not subject to the occurring interelement spectral
interferences of palladium on copper, iron on selenium and aluminum on arsenic.
Interference from a smoke-producing sample matrix can sometimes be reduced by
extending the charring time at a higher temperature or utilizing an ashing cycle in the presence
of air. Care must be taken, however, to prevent loss of the analyte.
Samples containing large amounts of organic materials should be oxidized by
conventional acid digestion before being placed in the furnace. In this way, broad-band
absorption will be minimized.
Anion interference studies in the graphite furnace indicate that, under conditions
other than isothermal, the nitrate anion is preferred. Therefore, nitric acid is preferable for any
digestion or solubilization step. When another acid in addition to nitric acid is needed, a
minimum amount should be used. This applies particularly to hydrochloric and, to a lesser
extent, to sulfuric and phosphoric acids.
4.10 Carbide formation resulting from the chemical environment of the furnace has been
observed. Molybdenum may be cited as an example. When carbides form, the metal is
released very slowly from the resulting metal carbide as atomization continues. Molybdenum
may require 30 seconds or more atomization time before the signal returns to baseline levels.
Carbide formation is greatly reduced and the sensitivity increased with the use of pyrolytically
coated graphite. Elements that readily form carbides are noted with the symbol "(p)" in Table 1.
4.11 Spectral interference can occur when an absorbing wavelength of an element
present in the sample, but not being determined, falls within the width of the absorption line of
the element of interest. The results of the determination will then be erroneously high, due to
the contribution of the interfering element to the atomic absorption signal. Interference can also
occur when resonant energy from another element in a multielement lamp, or from a metal
impurity in the lamp cathode, falls within the bandpass of the slit setting when that other metal is
present in the sample. This type of interference may sometimes be reduced by narrowing the
slit width.
4.12 It is recommended that all graphite furnace analyses be carried out using an
appropriate matrix modifier. The choice of matrix modifier is dependent on analytes, conditions,
and instrumentation and should be chosen by the analyst as the situation dictates. Follow the
instrument manufacturers instructions for the preferred matrix modifier. Refer to Chapter Two
for additional guidance.
4.13 It is recommended that a stabilized temperature platform be used to maximize an
isothermal environment within the furnace cell to help reduce interferences. Refer to Chapter
Two for additional guidance.
7010 - 4
Revision 0
February 2007
4.14 Cross-contamination and contamination of the sample can be major sources of
error because of the extreme sensitivities achieved with the furnace. The sample preparation
work area should be kept scrupulously clean. All glassware should be cleaned as directed in
Sec. 6.6. Pipet tips are a frequent source of contamination. The analyst should be aware of the
potential for the yellow tips to contain cadmium. If suspected, they should be acid soaked with
1:5 nitric acid and rinsed thoroughly with tap and reagent water. The use of a better grade of
pipet tip can greatly reduce this problem. Special attention should be given to assessing the
contamination in method blanks during the analysis. Pyrolytic graphite, because of the
production process and handling, can become contaminated. As many as five to ten hightemperature burns may be needed to clean the tube before use. In addition, auto sampler tips
may also be a potential source of contamination. Flushing the tip with a dilute solution of nitric
acid between samples can help prevent cross-contamination.
Specific interference problems related to individual analytes are located in this
4.15.1 Antimony -- High lead concentration may cause a measurable spectral
interference on the 217.6 nm line. Choosing the secondary wavelength or using
background correction may correct the problem.
Arsenic Elemental arsenic and many of its compounds are volatile;
therefore, samples may be subject to losses of arsenic during sample
preparation. Likewise, caution must be employed during the selection of
temperature and times for the dry and char (ash) cycles. A matrix modifier such
as nickel nitrate or palladium nitrate should be added to all digestates prior to
analysis to minimize volatilization losses during drying and ashing. In addition to the normal interferences experienced during
graphite furnace analysis, arsenic analysis can suffer from severe nonspecific
absorption and light scattering caused by matrix components during atomization.
Arsenic analysis is particularly susceptible to these problems because of its low
analytical wavelength (193.7 nm). Simultaneous background correction must be
employed to avoid erroneously high results. Aluminum is a severe positive
interferant in the analysis of arsenic, especially using D2 arc background
correction. Although Zeeman background correction is very useful in this
situation, use of any appropriate background correction technique is acceptable.
4.15.3 Barium -- Barium can form barium carbide in the furnace, resulting in less
sensitivity and potential memory effects. Because of chemical interaction, nitrogen should
not be used as a purge gas and halide acids should not be used.
4.15.4 Beryllium -- Concentrations of aluminum greater than 500 ppm may
suppress beryllium absorbance. The addition of 0.1% fluoride has been found effective in
eliminating this interference. High concentrations of magnesium and silicon cause similar
problems and require the use of the method of standard additions.
4.15.5 Cadmium -- Cadmium analyses can suffer from severe non-specific
absorption and light scattering caused by matrix components during atomization.
Simultaneous background correction is needed to avoid erroneously high results. Excess
chloride may cause premature volatilization of cadmium; an ammonium phosphate matrix
modifier may minimize this loss.
7010 - 5
Revision 0
February 2007
4.15.6 Chromium -- Low concentrations of calcium and/or phosphate may cause
interferences; at concentrations above 200 mg/L, calcium's effect is constant and
eliminates the effect of phosphate. Therefore, add calcium nitrate (calcium nitrate
solution: dissolve 11.8 g of calcium nitrate in 1 L reagent water) to ensure a constant
effect. Nitrogen should not be used as the purge gas because of a possible CN band
4.15.7 Cobalt -- Since excess chloride may interfere, it is necessary to verify by
standard additions that the interference is absent unless it can be shown that standard
additions are not necessary.
4.15.8 Lead -- If poor recoveries are obtained, a matrix modifier may be
necessary. Add 10 uL of phosphoric acid to 1 mL of prepared sample.
4.15.9 Molybdenum -- Molybdenum is prone to carbide formation; use a
pyrolytically coated graphite tube.
4.15.10 Nickel -- Severe memory effects for nickel may occur in graphite furnace
tubes used for other GFAA analyses, due to the use of a nickel nitrate matrix modifier in
those methods. Use of graphite furnace tubes and contact rings for nickel analysis that
are separate from those used for arsenic and selenium analyses is strongly
4.15.11 Selenium Elemental selenium and many of its compounds are volatile;
therefore, samples may be subject to losses of selenium during sample
preparation. Likewise, caution must be employed during the selection of
temperature and times for the dry and char (ash) cycles. A matrix modifier such as
nickel nitrate or palladium nitrate should be added to all digestates prior to analysis
to minimize volatilization losses during drying and ashing. In addition to the normal interferences experienced during
graphite furnace analysis, selenium analysis can suffer from severe nonspecific
absorption and light scattering caused by matrix components during atomization.
Selenium analysis is particularly susceptible to these problems because of its low
analytical wavelength (196.0 nm). Simultaneous background correction must be
employed to avoid erroneously high results. High iron levels can give
overcorrection with deuterium background. Although Zeeman background
correction is very useful in this situation, use of any appropriate background
correction technique is acceptable. Selenium analysis suffers interference from chlorides (>800
mg/L) and sulfate (>200 mg/L). The addition of nickel nitrate such that the final
concentration is 1% nickel will lessen this interference.
4.15.12 Silver -- Silver chloride is insoluble, therefore HCl should be avoided
unless the silver is already in solution as a chloride complex. In addition, it is
recommended that the stock standard concentrations be kept below 2 ppm and the
chloride content increased to prevent precipitation. If precipitation is occurring, a 5%:2%
HCl:HNO3 stock solution may prevent precipitation. Daily standard preparation may also
be needed to prevent precipitation of silver. Analysts should be aware that this technique
may not be the best choice for this analyte and that alternative techniques should be
7010 - 6
Revision 0
February 2007
4.15.13 Thallium -- HCl or excessive chloride will cause volatilization of thallium at
low temperatures. Verification that losses are not occurring must be made for each matrix
type (as detailed in 9.6.1).
4.15.14 Vanadium -- Vanadium is refractory and prone to form carbides.
Consequently, memory effects are common, and care should be taken to clean the
furnace before and after analysis.
This method does not address all safety issues associated with its use. The
laboratory is responsible for maintaining a safe work environment and a current awareness file
of OSHA regulations regarding the safe handling of the chemicals listed in this method. A
reference file of material safety data sheets (MSDSs) should be available to all personnel
involved in these analyses.
Concentrated nitric and hydrochloric acids are moderately toxic and extremely
irritating to skin and mucus membranes. Use these reagents in a hood whenever possible and
if eye or skin contact occurs, flush with large volumes of water. Always wear safety glasses or a
shield for eye protection when working with these reagents.
Hydrofluoric acid is a very toxic acid and penetrates the skin and tissues deeply if
not treated immediately. Injury occurs in two stages; first, by hydration that induces tissue
necrosis and then by penetration of fluoride ions deep into the tissue and by reaction with
calcium. Boric acid and other complexing reagents and appropriate treatment agents should be
administered immediately. Consult appropriate safety literature and have the appropriate
treatment materials readily available prior to working with this acid. See Method 3052 for
specific suggestions for handling hydrofluoric acid from a safety and an instrument standpoint.
Many metal salts are extremely toxic if inhaled or swallowed. Extreme care must
be taken to ensure that samples and standards are handled properly and that all exhaust gases
are properly vented. Wash hands thoroughly after handling.
The acidification of samples containing reactive materials may result in the release
of toxic gases, such as cyanides or sulfides. For this reason, the acidification and digestion of
samples should be performed in an approved fume hood.
The mention of trade names or commercial products in this manual is for illustrative
purposes only, and does not constitute an EPA endorsement or exclusive recommendation for
use. The products and instrument settings cited in SW-846 methods represent those products
and settings used during method development or subsequently evaluated by the Agency.
Glassware, reagents, supplies, equipment, and settings other than those listed in this manual
may be employed provided that method performance appropriate for the intended application
has been demonstrated and documented.
This section does not list common laboratory glassware (e.g., beakers and flasks).
Atomic absorption spectrophotometer -- Single- or dual-channel, single- or doublebeam instrument having a grating monochromator, photomultiplier detector, adjustable slits, a
7010 - 7
Revision 0
February 2007
wavelength range of 190 to 800 nm, and provisions for interfacing with a graphical display. The
instrument must be equipped with an adequate correction device capable of removing
undesirable nonspecific absorbance over the spectral region of interest and provide an
analytical condition not subject to the occurrence of interelement spectral overlap interferences.
Hollow cathode lamps -- Single-element lamps are preferred but multielement
lamps may be used. Electrodeless discharge lamps may also be used when available. Other
types of lamps meeting the performance criteria of this method may be used.
Graphite furnace -- Any furnace device capable of reaching the specified
temperatures is satisfactory. For all instrument parameters (i.e., drying, ashing, atomizing,
times and temperatures) follow the specific instrument manufacturers instructions for each
Data systems recorder -- A recorder is recommended for furnace work so that
there will be a permanent record and that any problems with the analysis such as drift,
incomplete atomization, losses during charring, changes in sensitivity, peak shape, etc., can be
easily recognized.
Pipets -- Microliter, with disposable tips. Sizes can range from 5 to 100 µL as
needed. Pipet tips should be checked as a possible source of contamination when
contamination is suspected or when a new source or batch of pipet tips is received by the
laboratory. The accuracy of variable pipets must be verified daily. Class A pipets can be used
for the measurement of volumes equal to or larger than 1 mL.
Glassware -- All glassware, polypropylene, or fluorocarbon (PFA or TFE)
containers, including sample bottles, flasks and pipets, should be washed in the following
sequence -- 1:1 hydrochloric acid, tap water, 1:1 nitric acid, tap water, detergent, tap water, and
reagent water. Chromic acid should not be used as a cleaning agent for glassware if chromium
is to be included in the analytical scheme. If it can be documented through an active analytical
quality control program using spiked samples and method blanks that certain steps in the
cleaning procedure are not needed for routine samples, those steps may be eliminated from the
procedure. Leaching of polypropylene for longer periods at lower acid concentrations is
necessary to prevent degradation of the polymer. Alternative cleaning procedures must also be
documented. Cleaning for ultra-trace analysis should be reviewed in Chapter Three.
Volumetric flasks of suitable precision and accuracy.
Reagent grade or trace metals grade chemicals must be used in all tests. Unless
otherwise indicated, it is intended that all reagents conform to the specifications of the
Committee on Analytical Reagents of the American Chemical Society, where such
specifications are available. Other grades may be used, provided it is first ascertained that the
reagent is of sufficiently high purity to permit its use without lessening the accuracy of the
determination. All reagents should be analyzed to demonstrate that the reagents do not contain
target analytes at or above the lowest limit of quantitation.
Reagent water -- All references to water in the method refer to reagent water,
unless otherwise specified. Reagent water must be free of interferences.
Nitric acid , HNO3 -- Use a spectrograde acid certified for AA use. Prepare a 1:1
dilution with water by adding the concentrated acid to an equal volume of water. If the method
7010 - 8
Revision 0
February 2007
blank does not contain target analytes at or above the lowest limit of quantitation, then the acid
may be used.
Hydrochloric acid (1:1), HCl -- Use a spectrograde acid certified for AA use.
Prepare a 1:1 dilution with water by adding the concentrated acid to an equal volume of water.
If the method blank does not contain target analytes at or above the lowest limit of quantitation,
then the acid may be used.
Purge gas -- A mixture of H2 (5%) and argon (95%). The argon gas supply must be
high-purity grade, 99.99% or better. If performance can be documented, alternative gases may
be used.
Stock standard metal solutions -- Stock standard solutions are prepared from
analytical reagent grade high purity metals, oxides, or nonhygroscopic salts using reagent water
and redistilled nitric or hydrochloric acids. (See individual methods for specific instructions.)
Sulfuric or phosphoric acids should be avoided as they produce an adverse effect on many
elements. The stock solutions are prepared at concentrations of 1,000 mg of the metal per liter.
Commercially available standard solutions may also be used. When using pure metals
(especially wire) for standards preparation, cleaning procedures, as detailed in Chapter Three,
should be used to ensure that the solutions are not compromised. Examples of appropriate
standard preparations can be found in Secs. 7.6.1 through 7.6.18.
Antimony -- Carefully weigh 2.743 g of antimony potassium tartrate,
K(SbO)C4H4O6C1/2H2O, and dissolve in reagent water. Dilute to 1 L with reagent water;
Arsenic -- Dissolve 1.320 g of arsenic trioxide, As2O3, or equivalent in 100
mL of reagent water containing 4 g NaOH. Acidify the solution with 20 mL conc. HNO3
and dilute to 1 L with reagent water.
Barium -- Dissolve 1.779 g of barium chloride, BaCl2C2H2O, in reagent
water and dilute to 1 L with reagent water.
Beryllium -- Dissolve 11.659 g of beryllium sulfate, BeSO4, in reagent
water containing 2 mL of nitric acid (conc.) and dilute to 1 L with reagent water.
Cadmium -- Dissolve 1.000 g of cadmium metal in 20 mL of 1:1 HNO3 and
dilute to 1 L with reagent water.
Chromium -- Dissolve 1.923 g of chromium trioxide, CrO3, in reagent
water, acidify with redistilled HNO3, and dilute to 1 L with reagent water.
Cobalt -- Dissolve 1.000 g of cobalt metal in 20 mL of 1:1 HNO3 and
dilute to 1 L with reagent water. Chloride or nitrate salts of cobalt(II) may be used.
Although numerous hydrated forms exist, they are not recommended, unless the exact
composition of the compound is known.
Copper -- Dissolve 1.000 g of electrolytic copper in 5 mL of redistilled
HNO3 and dilute to 1 L with reagent water.
Iron -- Dissolve 1.000 g of iron wire in 10 mL of redistilled HNO3 and
reagent water and dilute to 1 L with reagent water. Note that iron passivates in conc.
HNO3, and therefore some water should be present.
7010 - 9
Revision 0
February 2007
7.6.10 Lead -- Dissolve 1.599 g of lead nitrate, Pb(NO3)2, in reagent water,
acidify with 10 mL of redistilled HNO3, and dilute to 1 L with reagent water.
7.6.11 Manganese -- Dissolve 1.000 g of manganese metal in 10 mL of
redistilled HNO3 and dilute to 1 L with reagent water.
7.6.12 Molybdenum -- Dissolve 1.840 g of ammonium molybdate,
(NH4)6Mo7O24C4H2O, and dilute to 1 L with reagent water.
7.6.13 Nickel -- Dissolve 1.000 g of nickel metal or 4.953 g of nickel nitrate,
Ni(NO3)2C6H2O in 10 mL of HNO3 and dilute to 1 L with reagent water.
7.6.14 Selenium: Dissolve 0.345 g of selenious acid (actual assay 94.6%
H2SeO3) or equivalent and dilute to 200 mL with reagent water.
Due to the high toxicity of selenium, preparation of a small volume of reagent is
described. Larger volumes may be prepared if needed.
7.6.15 Silver -- Dissolve 1.575 g of anhydrous silver nitrate, AgNO3, in reagent
water. Add 10 mL of HNO3 (conc.) and dilute to 1 L with reagent water. Because this
standard is light sensitive, store in a amber glass bottle in a refrigerator.
7.6.16 Thallium -- Dissolve 1.303 g of thallium nitrate, TlNO3 , in reagent water,
acidify with 10 mL of conc. HNO3, and dilute to 1 L with reagent water.
7.6.17 Vanadium -- Dissolve 1.785 g of vanadium pentoxide, V2O5 , in 10 mL of
conc. HNO3 and dilute to 1 L with reagent water.
7.6.18 Zinc -- Dissolve 1.000 g of zinc metal in 10 mL of conc. HNO3 and dilute
to 1 L with reagent water.
Common matrix modifiers -- The use of a palladium modifier is strongly
recommended for the determination of all analytes. This will correct for general chemical
interferences as well as allow for higher char and atomization temperatures without allowing the
premature liberation of analyte. Other matrix modifiers may also be used as recommended by
the instrument manufacturer or when an interference is evident.
Palladium solution (Pd/Mg) -- Dissolve 300 mg of palladium powder in
concentrated HNO3 (1 mL of HNO3 , adding 0.1 mL of conc. HCl, if necessary). Dissolve
200 mg of Mg(NO3)2 in reagent water. Pour the two solutions together and dilute to 100
mL with reagent water.
Nickel nitrate solution (5%) -- Dissolve 25 g of Ni(NO3)2C6H2O in reagent
water and dilute to 100 mL.
Nickel nitrate solution (1%) -- Dilute 20 mL of the 5% nickel nitrate
solution to 100 mL with reagent water.
Ammonium phosphate solution (40%) -- Dissolve 40 g of ammonium
phosphate, (NH4)2HPO4, in reagent water and dilute to 100 mL.
Palladium chloride -- Weigh 0.25 g of PdCl2 to the nearest 0.0001 g and
dissolve in 10 mL of 1:1 HNO3. Dilute to 1 L with reagent water.
7010 - 10
Revision 0
February 2007
Two types of blanks are required for the analysis of samples prepared by any method
other than Method 3040. The calibration blank is used in establishing the analytical curve and
the method blank is used to identify possible contamination resulting from either the reagents
(acids) or the equipment used during sample processing including filtration.
The calibration blank is prepared by acidifying reagent water to the same
concentrations of the acids found in the standards and samples. Prepare a sufficient
quantity to flush the system between standards and samples. The calibration blank will
also be used for all initial (ICB) and continuing calibration blank (CCB) determinations.
The method blank must contain all of the reagents in the same volumes
as used in the processing of the samples. The method blank must be carried through the
complete procedure and contain the same acid concentration in the final solution as the
sample solution used for analysis (refer to Sec. 9.5).
The initial calibration verification (ICV) standard is prepared by the analyst (or a
purchased second source reference material) by combining compatible elements from a
standard source different from that of the calibration standard, and at concentration near the
midpoint of the calibration curve (see Sec. 10.2.1 for use). This standard may also be
7.10 The continuing calibration verification (CCV) standard should be prepared in the
same acid matrix using the same standards used for calibration, at a concentration near the
mid-point of the calibration curve (see Sec. 10.2.2 for use).
See the introductory material in Chapter Three, "Inorganic Analytes."
Refer to Chapter One for guidance on quality assurance (QA) and quality control
(QC) protocols. When inconsistencies exist between QC guidelines, method-specific QC
criteria take precedence over both technique-specific criteria and those criteria given in Chapter
One, and technique-specific QC criteria take precedence over the criteria in Chapter One. Any
effort involving the collection of analytical data should include development of a structured and
systematic planning document, such as a Quality Assurance Project Plan (QAPP) or a Sampling
and Analysis Plan (SAP), which translates project objectives and specifications into directions
for those that will implement the project and assess the results. Each laboratory should
maintain a formal quality assurance program. The laboratory should also maintain records to
document the quality of the data generated. All data sheets and quality control data should be
maintained for reference or inspection.
Refer to a 3000 series method (Method 3015, 3020, 3031, 3050, 3051, or 3052) for
appropriate QC procedures to ensure the proper operation of the various sample preparation
Instrument detection limits (IDLs) are a useful tool to evaluate the instrument noise
level and response changes over time for each analyte from a series of reagent blank analyses
to obtain a calculated concentration. They are not to be confused with the lower limit of
7010 - 11
Revision 0
February 2007
quantitation, nor should they be used in establishing this limit. It may be helpful to compare the
calculated IDLs to the established lower limit of quantitation, however, it should be understood
that the lower limit of quantitation needs to be verified according to the guidance in Sec. 10.2.3.
IDLs in µg/L can be estimated by calculating the average of the standard deviations of
three runs on three non-consecutive days from the analysis of a reagent blank solution with
seven consecutive measurements per day. Each measurement should be performed as though
it were a separate analytical sample (i.e., each measurement must be followed by a rinse and/or
any other procedure normally performed between the analysis of separate samples). IDLs
should be determined at least every three months or at a project-specific designated frequency
and kept with the instrument log book.
Initial demonstration of proficiency
Each laboratory must demonstrate initial proficiency with each sample preparation (a 3000
series method) and determinative method combination it utilizes by generating data of
acceptable accuracy and precision for target analytes in a clean matrix. If an autosampler is
used to perform sample dilutions, before using the autosampler to dilute samples, the laboratory
should satisfy itself that those dilutions are of equivalent or better accuracy than is achieved by
an experienced analyst performing manual dilutions. The laboratory must also repeat the
demonstration of proficiency whenever new staff members are trained or significant changes in
instrumentation are made.
For each batch of samples processed, at least one method blank must be carried
throughout the entire sample preparation and analytical process, as described in Chapter One.
A method blank is prepared by using a volume or weight of reagent water at the volume or
weight specified in the preparation method, and then carried through the appropriate steps of
the analytical process. These steps may include, but are not limited to, prefiltering, digestion,
dilution, filtering, and analysis. If the method blank does not contain target analytes at a level
that interferes with the project-specific DQOs, then the method blank would be considered
In the absence of project-specific DQOs, if the blank is less than 10% of the lower limit of
quantitation check sample concentration, less than 10% of the regulatory limit, or less than 10%
of the lowest sample concentration for each analyte in a given preparation batch, whichever is
greater, then the method blank is considered acceptable. If the method blank cannot be
considered acceptable, the method blank should be re-run once, and if still unacceptable, then
all samples after the last acceptable method blank should be reprepared and reanalyzed along
with the other appropriate batch QC samples. These blanks will be useful in determining if
samples are being contaminated. If the method blank exceeds the criteria, but the samples are
all either below the reporting level or below the applicable action level or other DQOs, then the
sample data may be used despite the contamination of the method blank. Refer to Chapter
One for the proper protocol when analyzing blanks.
Laboratory control sample (LCS)
For each batch of samples processed, at least one LCS must be carried throughout the
entire sample preparation and analytical process as described in Chapter One. The laboratory
control samples should be spiked with each analyte of interest at the project-specific action level
or, when lacking project-specific action levels, at approximately mid-point of the linear dynamic
range. Acceptance criteria should either be defined in the project-specifc planning documents
or set at a laboratory derived limit developed through the use of historical analyses. In the
absence of project-specific or historical data generated criteria, this limit should be set at ± 20%
of the spiked value. Acceptance limits derived from historical data should be no wider that ±
7010 - 12
Revision 0
February 2007
20%. If the laboratory control sample is not acceptable, then the laboratory control sample
should be re-run once and, if still unacceptable, all samples after the last acceptable laboratory
control sample should be reprepared and reanalyzed.
Concurrent analyses of reference materials (SRMs) containing known amounts of analytes
in the media of interest are recommended and may be used as an LCS. For solid SRMs, 80 120% accuracy may not be achievable and the manufacturer’s established acceptance criterion
should be used for soil SRMs.
Matrix spike, unspiked duplicate, or matrix spike duplicate (MS/Dup or MS/MSD)
Documenting the effect of the matrix, for a given preparation batch consisting of similar
sample characteristics, should include the analysis of at least one matrix spike and one
duplicate unspiked sample or one matrix spike/matrix spike duplicate pair. The decision on
whether to prepare and analyze duplicate samples or a matrix spike/matrix spike duplicate must
be based on a knowledge of the samples in the sample batch or as noted in the project-specific
planning documents. If samples are expected to contain target analytes, then laboratories may
use one matrix spike and a duplicate analysis of an unspiked field sample. If samples are not
expected to contain target analytes, laboratories should use a matrix spike and matrix spike
duplicate pair.
For each batch of samples processed, at least one MS/Dup or MS/MSD sample set
should be carried throughout the entire sample preparation and analytical process as described
in Chapter One. MS/MSDs are intralaboratory split samples spiked with identical concentrations
of each analyte of interest. The spiking occurs prior to sample preparation and analysis. An
MS/Dup or MS/MSD is used to document the bias and precision of a method in a given sample
Refer to Chapter One for the definitions of bias and precision, and for the proper data
reduction protocols. MS/MSD samples should be spiked at the same level, and with the same
spiking material, as the corresponding laboratory control sample that is at the project-specific
action level or, when lacking project-specific action levels, at approximately mid-point of the
linear dynamic range. Acceptance criteria should either be defined in the project-specifc
planning documents or set at a laboratory-derived limit developed through the use of historical
analyses per matrix type analyzed. In the absence of project-specific or historical data
generated criteria, these limits should be set at ± 25% of the spiked value for accuracy and 20
relative percent difference (RPD) for precision. Acceptance limits derived from historical data
should be no wider that ± 25% for accuracy and 20% for precision. Refer to Chapter One for
additional guidance. If the bias and precision indicators are outside the laboratory control limits,
if the percent recovery is less than 75% or greater than 125%, or if the relative percent
difference is greater than 20%, then the interference test discussed in Sec. 9.8 should be
The relative percent difference between spiked matrix duplicate or
unspiked duplicate determinations is to be calculated as follows:
*D1 & D2*
*D1 % D2*
× 100
7010 - 13
Revision 0
February 2007
RPD = relative percent difference.
= first sample value.
= second sample value (spiked or unspiked duplicate).
The spiked sample or spiked duplicate sample recovery should be within
± 25% of the actual value, or within the documented historical acceptance limits for each
If less than acceptable accuracy and precision data are generated, the following
additional quality control tests are recommended prior to reporting concentration data for the
elements in this method. At a minimum these tests, outlined in Secs. 9.8.1 and 9.8.2, should be
performed with each batch of samples prepared/analyzed with corresponding unacceptable
data quality results. These tests will then serve to ensure that neither positive nor negative
interferences are affecting the measurement of any of the elements or distorting the accuracy of
the reported values. If matrix effects are confirmed, the laboratory should consult with the data
user when feasible for possible corrective actions which may include the use of alternative or
modified test procedures or possibly the method of standard additions so that the analysis is not
impacted by the same interference.
Post digestion spike addition
The same sample from which the MS/MSD aliquots were prepared (asuming the
MS/MSD recoveries are unacceptable) should also be spiked with a post digestion spike.
Otherwise another sample from the same preparation should be used as an alternative.
An analyte spike is added to a portion of a prepared sample, or its dilution, and should be
recovered to within 80% to 120% of the known value. The spike addition should produce
a minimum level of 10 times and a maximum of 100 times the lower limit of quantitation. If
this spike fails, then the dilution test (Sec. 9.8.2) should be run on this sample. If both the
MS/MSD and the post digestion spike fail, then matrix effects are confirmed.
Dilution test
If the analyte concentration is sufficiently high (minimally, a factor of 10 above the
lower limit of quantitation after dilution), an analysis of a 1:5 dilution should agree within ±
10% of the original determination. If not, then a chemical or physical interference effect
should be suspected. For both a failed post digestion spike or an unacceptable dilution
test agreement result, the method of standard additions should be used as the primary
means to quantitate all samples in the associated preparation batch.
Where the sample matrix is so complex that viscosity, surface tension, and
components cannot be accurately matched with standards, the method of standard additions
(MSA) is recommended (see Sec. 9.10 below). Other options including the use of different
matrix modifiers, different furnace conditions, different preparatory methods or different
analytical methods may also be attempted to properly characterize a sample. Sec. 9.8 provides
tests to determine the potential for an interference and evaluates the need for using the MSA.
9.10 Method of standard additions -- The standard addition technique involves adding
known amounts of standard to one or more aliquots of the processed sample solution. This
technique attempts to compensate for a sample constituent that enhances or depresses the
analyte signal, thus producing a different slope from that of the calibration standards. It will not
correct for additive interferences which cause a baseline shift. The method of standard
additions may be appropriate for analysis of extracts, on analyses submitted as part of a
delisting petition, whenever a new sample matrix is being analyzed and on every batch that
fails the recovery test.
7010 - 14
Revision 0
February 2007
9.10.1 The simplest version of this technique is the single-addition method, in
which two identical aliquots of the sample solution, each of volume Vx, are taken. To the
first (labeled A) is added a known volume VS of a standard analyte solution of
concentration CS. To the second aliquot (labeled B) is added the same volume VS of
reagent water. The analytical signals of A and B are measured and corrected for nonanalyte signals. The unknown sample concentration Cx is calculated:
where SA and SB are the analytical signals (corrected for the blank) of solutions A and B,
respectively. Vs and Cs should be chosen so that SA is roughly twice SB on the average,
avoiding excess dilution of the sample. If a separation or concentration step is used, the
additions are best made first and carried through the entire procedure.
9.10.2 Improved results can be obtained by employing a series of standard
additions. To equal volumes of the sample are added a series of standard solutions
containing different known quantities of the analyte, and all solutions are diluted to the
same final volume. For example, addition 1 should be prepared so that the resulting
concentration is approximately 50 percent of the expected absorbance from the
indigenous analyte in the sample. Additions 2 and 3 should be prepared so that the
concentrations are approximately 100 and 150 percent of the expected endogenous
sample absorbance. The absorbance of each solution is determined and then plotted on
the vertical axis of a graph, with the concentrations of the known standards plotted on the
horizontal axis. When the resulting line is extrapolated to zero absorbance, the point of
interception of the abscissa is the endogenous concentration of the analyte in the sample.
The abscissa on the left of the ordinate is scaled the same as on the right side, but in the
opposite direction from the ordinate. An example of a plot so obtained is shown in Figure
1. A linear regression program may be used to obtain the intercept concentration.
9.10.3 For the results of this MSA technique to be valid, the following limitations
must be taken into consideration:
1. The apparent concentrations from the calibration curve must be linear (0.995 or
greater) over the concentration range of concern. For the best results, the
slope of the MSA plot should be nearly the same as the slope of the standard
2. The effect of the interference should not vary as the ratio of analyte
concentration to sample matrix changes, and the standard addition should
respond in a similar manner as the analyte.
3. The determination must be free of spectral interference and corrected for
nonspecific background interference.
9.11 Ultra-trace analysis requires the use of clean chemistry preparation and analysis
techniques. Several suggestions for minimizing analytical blank contamination are provided in
Chapter Three.
7010 - 15
Revision 0
February 2007
10.1 Calibration standards -- All analyses require that a calibration curve be prepared
to cover the appropriate concentration range. Usually, this means the preparation of a blank
and standards which produce an absorbance of 0.0 to 0.7. Calibration standards can prepared
by diluting the stock metal solutions in the same acids and acid concentrtions as the samples.
10.1.1 Calibration standards can be prepared fresh each time a batch of
samples is analyzed. If the ICV solution is prepared daily and the ICV is analyzed within
the acceptance criteria, calibration standards do not need to be prepared daily and may be
prepared and stored for as long as the calibration standard viability can be verified through
the use of the ICV. If the ICV is outside of the acceptance criteria, the calibration
standards must be prepared fresh and the instrument recalibrated. Prepare a blank and at
least three calibration standards in graduated amounts in the appropriate range of the
linear part of the curve.
10.1.2 The calibration standards should be prepared using the same type of acid
or combination of acids and at the same concentration as will result in the samples
following processing.
10.1.3 Beginning with the calibration blank and working toward the highest
standard, inject the solutions and record the readings. Calibration curves are always
10.2 A calibration curve must be prepared each day with a minimum of a calibration
blank and three standards. The curve must be linear and have a correlation coefficient of at
least 0.995.
10.2.1 After initial calibration, the calibration curve must be verified by use of an
initial calibration blank (ICB) and an initial calibration verification (ICV) standard. The ICV
standard must be made from an independent (second source) material at or near midrange. The acceptance criteria for the ICV standard must be ±10% of its true value and the
ICB must not contain target analytes at or above the lowest limit of quantitation for the
curve to be considered valid. If the calibration curve cannot be verified within the specified
limits, the cause must be determined and the instrument recalibrated before samples are
analyzed. The analysis data for the ICV must be kept on file with the sample analysis
10.2.2 The calibration curve must also be verified at the end of each analysis
batch and/or after every 10 samples by use of a continuing calibration blank (CCB) and a
continuing calibration verification (CCV) standard. The CCV standard should be made
from the same material as the initial calibration standards at or near midrange. The
acceptance criteria for the CCV standard must be ±10% of its true value and the CCB
must not contain target analytes at or above the lowest limit of quantitation for the curve to
be considered valid. If the calibration cannot be verified within the specified limits, the
sample analysis must be discontinued, the cause determined and the instrument
recalibrated. All samples following the last acceptable CCV/CCB must be reanalyzed.
The analysis data for the CCV/CCB must be kept on file with the sample analysis data.
The lower limits of quantitation should be established for all analytes for
each type of matrix analyzed and for each preparation method used and for each
instrument. These limits are considered the lowest reliable laboratory reporting
concentrations and should be established from the lower limit of quantitation check sample
7010 - 16
Revision 0
February 2007
and then confirmed using either the lowest calibration point or from a low-level calibration
check standard.
Lower limit of quantitation check sample
The lower limit of quantitation check (LLQC) sample should be analyzed
after establishing the lower laboratory reporting limits and on an as needed basis
to demonstrate the desired detection capability. Ideally, this check sample and the
low-level calibration verification standard will be prepared at the same
concentrations with the only difference being the LLQC sample is carried through
the entire preparation and analytical procedure. Lower limits of quantitation are
verified when all analytes in the LLQC sample are detected within ± 30% of their
true value. This check should be used to both establish and confirm the lowest
quantitation limit. The lower limits of quantitation determination using reagent
water represents a best case situation and does not represent possible matrix
effects of real-world samples. For the application of lower limits of quantitation on
a project-specific basis with established data quality objectives, low-level matrixspecific spike studies may provide data users with a more reliable indication of the
actual method sensitivity and minimum detection capabilities.
10.3 It is recommended that each standard should be analyzed (injected) twice and an
average value determined. Replicate standard values should be within ±10% RPD.
10.4 Standards are run in part to monitor the life and performance of the graphite tube.
Lack of reproducibility or significant change in the signal for the standard indicates that the tube
should be replaced. Tube life depends on sample matrix and atomization temperature. A
conservative estimate would be that a tube will last at least 50 firings. A pyrolytic coating will
extend that estimated life by a factor of three.
10.5 If conducting trace analysis, it is recommended that the lowest calibration standard
be set at the laboratory’s lower limit of quantitation. The laboratory can use a reporting limit that
is below the lower limit of quantitation but all values reported below the low standard should be
reported as estimated values.
11.1 Preliminary treatment of waste water, ground water, extracts, and industrial waste
is always necessary because of the complexity and variability of sample matrices. Solids,
slurries, and suspended material must be subjected to a solubilization process before analysis.
This process may vary because of the metals to be determined and the nature of the sample
being analyzed. Solubilization and digestion procedures are presented in Chapter Three.
Samples which are to be analyzed only for dissolved constituents need not be digested if they
have been filtered and acidified.
11.2 Furnace devices (flameless atomization) are a most useful means of extending the
lower limits of quantitation. Because of differences between various makes and models of
satisfactory instruments, no detailed operating instructions can be given for each instrument.
Instead, the analyst should follow the instructions provided by the manufacturer of a particular
instrument. A generalized set of instructions follows below.
7010 - 17
Revision 0
February 2007
11.2.1 Inject an aliquot of sample into the furnace and atomize. If the
concentration found is greater than the highest standard, the sample should be diluted in
the same acid matrix and reanalyzed. The use of multiple injections can improve accuracy
and help detect furnace pipetting errors.
11.2.2 To verify the absence of interference, follow the interference procedure
given in Sec. 9.8.
12.1 For determination of metal concentration by GFAA -- Read the metal value from
the calibration curve or directly from the read-out system of the instrument.
If dilution of sample was required:
µg/L metal in sample '
A (C%B)
µg/L of metal in diluted aliquot from calibration curve.
Starting sample volume , mL.
Final volume of sample, mL.
12.1.2 For solid samples, report all concentrations in consistent units based on
wet weight. Ensure that if the dry weight was used for the analysis, percent solids should
be reported to the client. Hence:
mg metal)kg sample '
A x V
mg/L of metal in processed sample from calibration curve.
Final volume of the processed sample, L.
Weight of sample, Kg.
12.1.3 Different injection volumes must not be used for samples and standards.
Instead, the sample should be diluted and the same size injection volume be used for both
samples and standards.
Results need to be reported in units commensurate with their intended use and all
dilutions need to be taken into account when computing final results.
7010 - 18
Revision 0
February 2007
13.1 Performance data and related information are provided in SW-846 methods only as
examples and guidance. The data do not represent required performance criteria for users of
the methods. Instead, performance criteria should be developed on a project-specific basis,
and the laboratory should establish in-house QC performance criteria for the application of this
method. These performance data are not intended to be and must not be used as absolute QC
acceptance criteria for purposes of laboratory accreditation.
13.2 For relevant performance data, see the methods of Ref. 1.
14.1 Pollution prevention encompasses any technique that reduces or eliminates the
quantity and/or toxicity of waste at the point of generation. Numerous opportunities for pollution
prevention exist in laboratory operation. The EPA has established a preferred hierarchy of
environmental management techniques that places pollution prevention as the management
option of first choice. Whenever feasible, laboratory personnel should use pollution prevention
techniques to address their waste generation. When wastes cannot be feasibly reduced at the
source, the Agency recommends recycling as the next best option.
14.2 For information about pollution prevention that may be applicable to laboratories
and research institutions consult Less is Better: Laboratory Chemical management for Waste
Reduction available from the American Chemical Society’s Department of Government
Relations and Science Policy, 1155 16th St., N.W. Washington, D.C. 20036, http://www.acs.org.
The Environmental Protection Agency requires that laboratory waste management
practices be conducted consistent with all applicable rules and regulations. The Agency urges
laboratories to protect the air, water, and land by minimizing and controlling all releases from
hoods and bench operations, complying with the letter and spirit of any sewer discharge permits
and regulations, and by complying with all solid and hazardous waste regulations, particularly
the hazardous waste identification rules and land disposal restrictions. For further information
on waste management, consult The Waste Management Manual for Laboratory Personnel
available from the American Chemical Society at the address listed in Sec. 14.2.
Methods for Chemical Analysis of Water and Wastes; U.S. Environmental Protection
Agency. Office of Research and Development. Environmental Monitoring and Support
Laboratory. ORD Publication Offices of Center for Environmental Research Information:
Cincinnati, OH, 1983; EPA-600/4-79-020.
W. G. Rohrbough, et al., Reagent Chemicals, American Chemical Society Specifications,
7th ed.; American Chemical Society: Washington, DC, 1986.
1985 Annual Book of ASTM Standards, Vol. 11.01; "Standard Specification for Reagent
Water"; ASTM: Philadelphia, PA, 1985; D1193-77.
7010 - 19
Revision 0
February 2007
The following pages contain the tables and figure referenced by this method. A flow
diagram of the procedure follows the tables.
7010 - 20
Revision 0
February 2007
Furnace Procedure a, b
Lower Limit of Quantitation
NOTE: The symbol (p) indicates the use of pyrolytic graphite with the furnace
For furnace sensitivity values, consult instrument operating manual.
The listed furnace values are those expected when using a 20-µL injection and
normal gas flow, except in the cases of arsenic and selenium, where gas interrupt
is used.
Source: Ref. 1.
7010 - 21
Revision 0
February 2007
argon or nitrogen
argon or nitrogen
248.8, 271.8,
302.1, 252.7
argon or nitrogen
argon or nitrogen
argon or nitrogen
argon or nitrogen
argon or nitrogen
nitrogen should not
be used
nitrogen should not
be used
nitrogen should not
be used
nitrogen should not
be used
Note: If more than one wavelength is listed, the primary line is underlined.
The argon/H2 purge gas is also applicable.
Source: Ref. 1
7010 - 22
Revision 0
February 2007
7010 - 23
Revision 0
February 2007
7010 - 24
Revision 0
February 2007
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