A Novel Automated Liquid/Liquid Extraction Technique for the Determination of Caffeine in Coffee
Flex
GC/MS
Agilent 7890/5975
Inlet
Titan LVI PTV
Inlet Temperature
45°C for 0.15 min, 500°C/min to 325°C
for 14 min
Inlet Head Pressure
14.956 psi
Split
250:1
Split Flow
350ml/min
Mode
Pulsed Split
Injection Pulse Pressure
25psi until 0.05 min
100% (250µl)
Cryo
On
100%
Liner
TITAN XL SB Deactivated Baffled Liner
Column
Rxi-5Sil MS 30m x 0.25mm I.D. x 0.25µm
film thickness
Oven Temperature Program
40°C hold for 1.5 min, ramp 25°C/min to
310°C hold for 1.7 min, 14 min run time
120
Column Flow Rate
1.4ml/min.
100
Gas
Helium
Total Flow
354.4ml/min
Source Temperature
230°C
Quad Temperature
150°C
MS Transfer Line Temperature
280°C
Solvent Delay
3.0 min
Acquisition Mode
Scan
Scan Range
m/z 40-500
Sampling Rate
3.12 scans/sec
General
Method Type
Liquid
Sample Preparation (Run Twice)
Rinse Volume
100% (250µl)
Rinse Fill Rate
100%
Rinse Cycles
2
Rinse Dispense Rate
100%
Solvent Pump Cycles
1
Solvent Pump Volume
Pump Dispense Rate
Solvent Volume
100% (250µl)
Solvent Fill Rate
2%
Caffeine is a natural component of coffee, acting as a stimulant.
Thus, drinking coffee can wake you up and enhance your
alertness. Many people feel that they cannot function without
some coffee to start their day. Furthermore, the amount of
caffeine a person is receiving from their coffee can vary with the
type of bean and the brewing process.
Solvent Fill Delay
0 sec
Sample Vial Needle Depth
100%
This study will examine automated extraction of caffeine from
assorted coffee blends. In order to distinguish caffeine amounts
from the blends, brand A will be used for examination of light,
medium, dark and flavored coffees. Brands A, B, and C will be
used for assessing medium blend differences. Finally, brand C
will be used to investigate brewing disparities.
Agitation Time
5.0 min
Agitation Delay
0.1 min
Agitation Speed
100%
INTRODUCTION
DISCUSSION
Liquid-liquid extraction takes advantage of a compound’s
solubility in different solvents. Since caffeine is more soluble
methylene chloride it diffuses readily from the coffee into the
methylene chloride. Additionally, methylene chloride is denser
than water, so the extracted caffeine separates into the bottom
methylene chloride portion of the liquid system. Using the
FLEX autosampler, an automated liquid-liquid extraction was
performed. The FLEX added methylene chloride to a coffee
sample, mixed the sample, let it settle and injected the extracted
caffeine onto the GC.
EXPERIMENTAL
The sampling system used for this analysis was the EST
Analytical FLEX autosampler fitted with a 250µl liquid syringe. A
Restek Rxi-5 Sil MS 30m x 250mm x 0.25µm column was installed
in the GC. The Agilent 7890 GC and 5975 MS were employed for
separation and analysis. Furthermore, the GC inlet was equipped
with the Titan PTV LVI for sample injections. Refer to Tables 1 and
2 for the sampling and analysis parameters.
Incubate/Agitate
Incubation Temperature
Incubation Time
Agitate
Ambient Equilibration Time
25°C
5.1 min
Yes
5.0 min
Rinse
Rinse Volume
8% (20µl)
Rinse Fill Rate
100%
Rinse Cycles
2
Sample
Sample Volume
4% (10µl)
Sample Depth
100%
Sample Depth Speed
90%
Sample Fill Rate
1%
Sample Fill Delay
5 sec
Sample Rinse Volume
8% (20µl)
Sample Rinse Cycles
1%
Sample Pump Volume
8% (20µl)
Sample Pump Cycles
2
Dispense Rate
50%
4% (10µl)
Single Injection Port
Injection Rate
Injection Volume
100%
8% (20µl)
Pure caffeine was acquired from Sigma Aldrich and diluted in
water in order to make a 5000ppm standard. Next a calibration
curve was prepared in order to calibrate the extraction of caffeine
from a water matrix. Single cup coffee servings in assorted
blends, flavors and brands were purchased. The assorted coffees
were prepared and two milliliters of each coffee was measured
and placed in mini reaction vials. The vials of coffee were then
positioned in the FLEX sample tray for automated extraction.
Coffee samples were run in triplicate in order to ensure
reproducibility. Extraction results are listed in Table 3; Figures 1
through 3 display the results in bar graph format while Figures 4
through 6 show a comparison of the coffee chromatograms.
Coffee
Caffeine
Amount
Decaf
27.41
Morning Roast
96.53
Medium Roast A
110.57
1 sec
Dark Roast
110.84
Post-Injection Delay
1 sec
Vanilla Flavored
116.54
Mocha Flavored
119.84
Rinse Volume
8% (20µl)
Medium Roast B
123.54
Rinse Fill Rate
100%
Medium Roast C
102.63
Brewed Cup C
181.44
Rinse Cycles
2
80
60
40
20
0
Decaf
Morning
Roast
Medium
Roast A
Dark
Roast
Vanilla
Flavored
Mocha
Flavored
EST Analytical shall not be liable for errors contained herein or for incidental or consequential damages in connection with this publication. Information, descriptions, and specifications in this publication are subject to change without notice.
Figure 5. Chromatograms of Vanilla Coffee versus Mocha Coffee
Figure 1. Caffeine Comparison of Brand A Assorted Blends
140
80
60
40
20
0
Roast A
Roast B
Roast C
Figure 2. Caffeine Comparison of Different Medium Roast Brands
150
100
50
0
Single Cup C
Brewed Pot C
Figure 3. Caffeine Comparison of a Single Cup Brew versus a
Brewed Pot of Coffee
Table 3. Extraction Results
Pre-Injection Delay
Rinse
100
200
Air Volume Gap
Air Fill Volume
120
Caffeine (mg)
Coffee is a requirement for many people to start their day.
However, the reason people drink coffee may be more for the
caffeine than for the taste. In order to determine the amount
of caffeine in coffee, many coffee producers use liquid-liquid
extraction. How much caffeine is in your cup of coffee? Assorted
coffee blends will be extracted for caffeine using an automated
liquid-liquid extraction technique with the intention of answering
this question.
Autosampler
Table 2. GC/MS Experimental Parameters
Caffeine (mg)
Anne Jurek, Lindsey Pyron, and Justin Murphy
EST Analytical, Cincinnati, Ohio
Table 1. FLEX Autosampler Experimental Parameters
Caffeine (mg)
AUTHORS
Figure 6. Overlay of Caffeine Extracted from a Single Cup versus
a Pot of Coffee
CONCLUSIONS
When analyzing the amount of caffeine in the same brand,
the amounts were quite similar. Surprisingly, the dark coffee
blend had the same amount of caffeine as the medium blend.
The mocha flavor had the most caffeine of Brand A which was
expected with the added caffeine in chocolate. The amount of
caffeine in the medium roast did not differ much from brand
to brand with Brand C having the least and Brand B having the
most. The most marked difference between the coffees was found
when comparing a single cup brew versus a pot of brewed coffee.
The pot of brewed coffee had substantially more caffeine than the
single cup. This was probably due to the amount of coffee used to
brew the pot as compared to the controlled amount of coffee in a
single cup. Finally, the extraction also proved to be quite efficient
in removing vanillin and ethyl vanillin from the vanilla and mocha
flavored coffees. These results displayed the mocha coffee to
have more of the vanilla flavoring than the vanilla coffee.
Using the FLEX system, automated extraction proved to be
an easy and accurate process. The caffeine extraction was
efficient with the parameters established and the results were
reproducible throughout the study.
Purposeful Innovation… that’s
Figure 4. Chromatograms of Different Roasts of Brand A
For more information on our products
and services, visit our website:
www.estanalytical.com/products.
estanalytical.com | 800.283.3510