Final Lab Report-Caffeine

1. 1

A COMPARATIVE STUDY
OF THE CAFFEINE
CONCENTRATIONS IN
VARIOUS CAFFEINATED
AND DECAFFEINATED
BEVERAGES

CH4721
Andrew LeSage, Christina Welch and Ford Guo
Due: May 1
st
, 2015
Individual Project Lab Report

2. 2

Abstract
Caffeine is a central nervous stimulant that provides the body with extra energy by inhibiting the
adenosine receptors. Consistent caffeine consumption can lead to various side effects such as
nausea and anxiety. The goal of this experiment was to perform a comparative analysis of the
concentration of caffeine in caffeinated versus decaffeinated beverages. These caffeine
concentrations were quantified by UV-vis spectrophotometry and high pressure liquid
chromatography. A difference between the caffeine concentrations reported by companies and
experimental data was observed. Experimental results also indicated many decaffeinated
beverages contained caffeine concentrations that exceed the FDA allotted concentration to be
considered decaffeinated.
Introduction
Caffeine is a central nervous system stimulant that increases alertness, relaxes smooth
muscles, stimulates cardiac muscles, and causes excess urination. In medicine, caffeine can be
used to treatment migraines, relieve pain and alleviate drowsiness [1].
Caffeine is an inhibitor of adenosine, which is a central nervous system neuromodulator
that binds to A1 and A2A receptors on the cerebrum [2]. When adenosine binds to these receptors,
neural activity is slowed, causing the body to feel tired. While the body is resting, adenosine
facilitates sleep and dilates the blood vessels to ensure adequate oxygenation. Structural
similarities between caffeine and adenosine (Figure 1) allows caffeine to act as an adenosine-
receptor antagonist. Caffeine inhibits adenosine by binding to its receptors without reducing
neural activity, which causes an overall decrease in adenosine receptor availability and an
increase in neural activity. This increase in neural activity forces the pituitary gland to secrete
hormones that cause the adrenal gland to produce adrenalin—the hormone that increases
attention levels and provides the body with extra energy [3].






Caffeine Adenosine
Figure 1. Structure of Caffeine versus the structure of adenosine [1].

3. 3

Regular consumption of caffeine can lead to a mild physical dependence. Symptoms of
caffeine withdrawal include headaches, fatigue, anxiety, irritability, depressed moods, and/or
difficulty concentrating. For a variety of reasons, some people prefer to drink decaffeinated
beverages and herbal teas instead of caffeine containing beverages [4].
In 1903, Ludwig Roselius became the first scientist to successfully decaffeinate coffee
beans. His decaffeination process, the Roselius Process, involved treating green coffee beans
with chlorinated hydrocarbon solvents to extract the caffeine and then using a roasting process to
remove any solvent from the beans. In the 1970’s, Roselius patented a new process developed by
the Max Planck Institute that uses carbon dioxide to eliminate caffeine from coffee beans; this is
the primary method manufacturers utilize in the production of decaffeinated coffee beans [5].
According to Commercial Item Description (CID) A-A-20213B mandated by the United States
Department of Agriculture, decaffeinated coffee shall not exceed more than 0.10% of its original
caffeine content in dry, packaged coffee[6].
In this experiment, the amount of caffeine in a variety of beverages were quantified using
two different analytical techniques. For the first method, the caffeine extraction was performed
using dichloromethane (DCM). The solubility of caffeine in water at 25
o
C is 2.2g/L and 10.2g/L
in DCM [7]. Since water and DCM are immiscible in each other and have a partition coefficient
of 4.63, which makes DCM a good organic solvent for caffeine extraction [8].
The caffeine was extracted from the samples using DCM. A Rotovap was used to
evaporate the DCM and crystalize the caffeine. The crystallized caffeine was dissolved in water
and tested with a UV-Vis Spectrophotometer (according to the literature, caffeine should be
observed at 272.8nm in water [9]. After all of the spectrums from the various caffeine samples
were observed, they were quantified with Beer’s law.
Molecules containing π-electrons and non-bonding electrons were considered to be UV
active for their ability to be excited to a higher anti-bonding energy state [10].
The less energy the electron needs for transition, the higher the maximum absorption
peak will be observed at a longer wavelength, which can be understood with the following
formula,
(1)
??????=ℎ??????/?????? [11]
where h is Planck's constant and c is the speed of light. As figure 1 demonstrates, the structure of
caffeine has multiple double bonds and carbonyl carbons, so it is should be UV active. Beer’s
Law demonstrates the relationship between the attenuation of light and the properties of material
which the light is traveling through. The formula of Beer’s Law is shown below,

4. 4

(2)
??????=??????????????????
where the A is the absorbance number, ?????? is the molar absorptivity, l is the path length, which is
generally the width of the cuvette used in the spectroscopy, and c is the concentration of the
solution [12]. In order to apply the Beer’s Law more directly, it can be simplified as,
(3)
??????∝??????
High performance liquid chromatography (HPLC) was the second method used to
quantify the amount of caffeine in each sample. HPLC is an analytical technique that is used to
separate, identify, and quantify various components of a mixture. To perform HPLC, the sample
of interest is injected into a stream of a highly pressurized mobile phase. The pressurized stream
continues to the HPLC column that is filled with an absorbent solid. Each component of the
sample interacts differently with the solid inside the column, causing each to have a different
retention time. As a result, the various components of the sample of interest exit the column at
different times. After exiting the column, the components continue to the detector of the
machine. The detector sends the information to the processing unit, which in turn prints out a
chromatograph [13]. Figure 2 illustrates this process.






Figure 2. Diagram of typical HPLC set-up [14]
There are a variety of parameters that must be optimized to gather useable data including the
mobile phase, flow rate, attenuation, column material, and the type of detector [13]. The C18
column was selected to perform HPLC because the column material is very nonpolar and are
capable of interacting with polar molecules, like caffeine. C18 columns are extremely common
in HPLC because the packing material can provide a variety of different pore sizes, has an ability
to interact with a large range of molecules, and is relatively inexpensive to manufacture.

5. 5

Materials and Methods
Materials
sodium carbonate, dichloromethane, sodium chloride, methanol, acetonitrile, Sigma pure
caffeine, Pepsi, Decaffeinated Pepsi®, Mountain Dew® Kick Start Energizing Drink, Red Bull®
Energy Drink, Monster® Energy Drink, Monster Unleaded® Energy Drink, Great Value®
Decaffeinated Black Tea, Twining’s® Irish Black Tea, Folger’s® Half Caff Coffee, Folger’s®
Decaffeinated Coffee, Folger’s® Columbian Coffee.

Instruments
Shimadzu 2450 UV-Vis Spectrophotometer, Shimadzu SPD20A/LC20AD Prominence UV-Vis
Detector HPLC, Heidolph Laboratories Laborota 4000 Efficient Rotovap.
Methods
Method I: Caffeine Extraction
Sample Preparation
Caffeine Standards
Four pure caffeine standards (ranging from 1g to 2.9grams) were prepared and suspended in
100mL of deionized water.
Soda and Energy Drinks
To prepare the beverages and energy drinks, the carbonation in 100mL of each sample was
removed by boiling it on a hot plate for 10 minutes. To make sure the carbonation was
completely removed from the samples, the beaker containing the sample was spun for a few
minutes. If any bubbles began to form, the sample was placed back on the heater for 5 additional
minutes.
Tea
For each tea sample, three tea bags were weighed and boiled in 120 mL of water mixed with 3 g
sodium carbonate for 10 minutes.
Coffee
For each coffee sample, 5 g of coffee grinds were weighed out and added to 100 mL of boiling
water mixed with 2 g sodium carbonate. The solution was stirred for 5 minutes and filtrated to
remove any coffee grinds.
Extraction
For soft drinks and energy drinks, 75 mL of each beverage was measured and transferred into a
separatory funnel. Next, 1.5 g of sodium carbonate was added to the separatory funnel and mixed

6. 6

lightly until it was completely dissolved. After, 20 mL of dichloromethane was added and shaken
lightly. To make sure most of the caffeine went into the organic (DCM) layer, the solution was
left to separate for approximately 10 minutes. If an emulsion formed, 1 g of sodium chloride was
added and the separation time was increased to 15 minutes. Repetition of this step was performed
until the emulsion disappeared. The organic layer in the separatory funnel was transferred to a
vile and stored at 4°C.
For coffee and tea, 25 mL of each sample was measured and transferred into separatory funnel.
Next, 25 mL of dichloromethane (DCM) was added into the funnel and shaken lightly for 3
minutes to prevent the formation of an emulsion. To make sure most of the caffeine was
transferred from the drink to the organic layer, it was left to separate for 10 minutes. If an
emulsion formed, 1.5 g of sodium chloride was added into the mixture and shaken lightly for
another 2 minutes. The solution was left to separate for another 5 minutes. Repetition of this step
was performed until the emulsion disappeared. The organic layer in the separatory funnel was
transferred to a vile and stored at 4°C.
Crystallization and sample solution
The vile was connected to the rotary evaporation apparatus with an adapter to remove the
organic solvent. The temperature was set to 40
o
C and the sample was spun at an appropriate
speed. Once the DCM was completely evaporated, the purified caffeine crystals were suspended
in 5 mL of deionized water.
UV absorption
The UV-Vis Spectrophotometer was used to generate the standard curve of pure caffeine at 273
nm and observe the spectrum at that same wavelength. Every sample was scanned from 190-350
nm using quartz cuvettes to make sure there weren’t any visible impurity peaks. The maximum
absorption of caffeine in each sample appeared at 273 nm.
Method II: High Performance Liquid Chromatography
Preparation
Caffeine standards containing 100, 150, 200, and 250 ppm caffeine in distilled water were
prepared. The mobile phase, 60/40 (V/V) MeOH/H2O was prepared using reaction grade
methanol and distilled water. The mobile phase was then degassed and filtered.
To prepare the HPLC in Chemical Sciences 404, the machine was allowed to warm up for three
minutes. After, the lines inside the machine were purged with mobile phase for three minutes.
After the lines were purged, the pump was turned on. Mobile phase was pumped through the
lines for 20 minutes, allowing the equipment to equilibrate. Next, the detector was calibrated
using the auto-zero function. Between each injection, the lines of the machine were rinsed for 20
minutes using mobile phase.

7. 7

Isocratic Flow
The initial parameters used for each run trial used an isocratic flow of 60/40 MeOH/water at a
flow rate of 1.00 mL/min, C18 column, UV/Vis detection at 253 nm, and an attenuation of 6.
Useable data was not obtained from this study. The following documents the steps taken to
optimize the HPLC conditions in an attempt to collect useable data.
First, the attenuation on the machine was turned down because the differences in the peak height
of each standard were minimal. Adjusting the attenuation of the machine turns down the
sensitivity of the detector. Turning down the attenuation should have lowered the peak height of
the caffeine standard and provided data that showed a correlation between caffeine concentration
and peak height [15]. Adjusting the attenuation failed to lower the height of the peaks and
minimal differences in the peak heights of each standard were still observed.
The flat tops on that were observed on top of each peak also suggests that the samples are too
concentrated for the detector sensitivity. Additionally, a decrease in peak height was observed as
the concentration of the standard increased, suggesting that the caffeine standard was
aggregating in the column. The next step that was taken to optimize HPLC conditions was
sample dilution. Each sample was diluted 10x using distilled water [15]. After the samples were
diluted 10x the flat tops on the peaks were still present, an inverse correlation between
concentration and peak height was still observed, and no useable data was obtained. Lastly,
gradient flow HPLC was attempted to help minimize caffeine aggregation in the column.
Gradient Flow
Gradient HPLC was conducted using water and methanol under the following conditions: 95%
H2O/5% MeOH to 40% H2O /60%MeOH over 20 minutes and back to 95% H2O /5% MeOH
over an additional 20 minutes.
This method also failed to provide useable data. The following measures were taken to try
optimizing HPLC conditions for gradient flow.
Many peaks were present on the chromatograph, suggesting that the column was dirty. To try
cleaning the C18 column, acetonitrile was used to flush the column for 45 minutes. Acetonitrile
is known to help regenerate the C18 inside the column, by pulling off excess molecules off the
column. Acetonitrile is extremely polar and does an excellent job cleaning the column [16].
After running the acetonitrile through the column, the initial gradient that was set up was run
twice to help prepare the column for the caffeine standard. Cleaning the column with acetonitrile
should have cleaned the column and eliminated many of the impurities that were observed in the
original gradient flow chromatograph. This method actually increased the amount of impurities
that were detected, yielding more data that was not useable.

8. 8

Lastly, the column on the machine was switched out with a newer column. After preparing the
column, unusable results were obtained again.
Chromatographs from the initial isocratic flow, the initial gradient flow, and the gradient flow
after the machine was cleaned with acetonitrile can be found in the appendix.

Results
Method I: Caffeine Extraction

Figure 3. UV Spetrum for 1 mg/L, 1.45 mg/L, 2.2 mg/L and 2.9 mg/L caffeine standards at 253
nm.
The line of best fist from figure 3 was calculated using Microsoft Excel. Equation 4 shows the
formula for the standard curve.
(4)
??????=0.5142∗??????−0.0109
where A is the absorbance of the sample and c is the caffeine concentration in mg/L.
Table 1. The measured UV absorbance of each sample at 273 nm.
Sample 1/40 dilution 1/200 dilution
Pepsi 0.438
Decaffeinated Pepsi 0.108
Mtn. Dew Energy Drink 0.415
Red Bull 0.501
Monster Energy 0.532
Monster Energy Unleaded 0.823
Great Value Black Tea Decaf 0.203
Irish Black Tea 3.418
half calf 1.66
Columbian coffee 2.578
Decaffeinated coffee 0.472
0.4780.704
1.2531.404
y = 0.5142x -0.0109
R² = 0.957
0
5
0 1 2 3 4 5
Absorption
mg/L water
Caffeine Standard Concentration vs
Absorbance

9. 9


Equation 4 was used to calculate the concentration of caffeine in each sample, using the
absorbance values listed in table 1. The results of these calculations are displayed in table 2.
An overall scan (from 190nm to 350nm) was performed to make sure there weren’t any
impurities. All of the peaks are identified by the software and the peaks labeled “1” are the ones
observed at 273nm.

Figure 4 Overall scan for the samples ranging from 190nm to 350nm. (From highest to lowest
peak at 273nm) the red peak represents the irish black tea, the second red peak represents the
Columbian Coffee, the black peak represents the Half Caff, green peak represents standard

10. 10

curve, blue peak represents decaffeinated coffee, and the second black peak represents
decaffeinated Great Value black tea.

Table 2. The calculated concentration of caffeine in each sample.
Sample [Caffeine](mg/L)
Pepsi 4.21
Decaffeinated Pepsi 1.12
Mtn. Dew Energy Drink 19.98
Red Bull 24.01
Monster Energy 25.46
Monster Energy Unleaded 7.82
Great Value Black Tea Decaf 1.83
Irish Black Tea 29.29
half calf 14.27
Columbian coffee 22.11
Decaffeinated coffee 4.12

Method II: HPLC
Discussion
Due to time constraints, no further attempts were made to obtain a useable chromatograph from
HPLC. If more time were allotted, the HPLC machine would have been rinsed several additional
times using acetonitrile. If these attempts failed, the machine would then be professionally
serviced. According to the service-date sticker, the HPLC machine is in need of being
professionally serviced. The presence of flat peaks after steps were taken to optimize HPLC
peaks suggests that the bulb inside the machine is ready to be replaced. After the machine is
professionally serviced, the next step would be determination of the optimal flow rate.
Table 3. The comparison of caffeine concentration taken from the samples in the lab data versus
the concentration reported by each company.
Company reported data experimental data
[caffeine]
g/L
amount of
caffeine
mg
[caffeine]
g/L
amount of
caffeine
mg
Pepsi 0.11 38 0.42 149.438872
Pepsi caffeine free 0.00 0 0.11 39.5764802
Mountain Dew
Kickstart
0.16 92 2.00 1122.481928
Red Bull 0.32 80 2.40 600.6890605
Monster 0.34 160 2.55 1204.868737
Monster Unleaded 0.00 0 0.78 370.1336675

11. 11


When comparing the lab data with company reported data, the caffeine concentration in each
beverage seems to be significantly higher than what the company reported.

Figure 5. The concentration of caffeine reported by the company versus the concentration
gathered from the experimental data.
The serving size for each sample is shown in the table and figure below,
Table 4. Sample serving size and caffeine amount.
serving
size
(ml)
caffeine
(mg)
Pepsi 355 149.44
Pepsi Caffeine-Free 355 39.58
Mountain Dew
Kickstart
562 1122.48
Red Bull 250 600.69
Monster 473 1204.87
Monster Unleaded 473 370.13
Great Value Black Tea
Decaf
100 7.31
Irish Black Tea 100 117.15
half calf 100 57.09
Columbia coffee 100 88.45
decaff coffee 100 16.50

0.11
0.00
0.16
0.320.34
0.00
0.42
0.11
2.00
2.40
2.55
0.78
0.00
0.50
1.00
1.50
2.00
2.50
3.00
concentration(g/L)
Comparison of labeled data and experimental data
(concentration)
company reported
lab

12. 12

The amount of serving size in the soft drinks and energy drinks are determined by each 12 oz.
can. For tea, the serving size is measured as one tea bag in 100 mL of water and for coffee, it is
2.5 g of coffee in 100 mL of water.

Figure 6. The amount of caffeine for each serving size gathered from the experimental data.
According to figure 6, the monster energy drink has the greatest amount of caffeine per one
serving among all of the other beverages, and great value decaffeinated black tea appears to be in
correlation with the federally regulated amount of caffeine present to be considered
decaffeinated, however, monster unleaded, which is advertised as caffeine free shows an
adequate amount of caffeine in the experiment.
Conclusion
Caffeine is a central nervous system stimulated that provides an increase in energy by inhibiting
the adenosine receptors. A consistent intake of caffeine can lead to various side effects such as
irritability or depression; for these reasons and others, people prefer to drink decaffeinated
beverages and teas. From our experiment, we were able to determine that the only decaffeinated
beverage that correlates with the federal regulation (caffeine content must not exceed .10%) is
Great Value black tea.

0
200
400
600
800
1000
1200
1400
pepsipepsi
caffeine
free
mountain
dew kick
start
red bullmonster
energy
monster
unlead
Great
Value
Black
Tea
Decaf
Irish
Black
Tea
half calfcolumbia
coffee
decaff
coffee
amount of caffeine(mg)
sample
Amount of caffeine for each serve of the drinks

13. 13

References
[1] National Center for Biotechnology Information. PubChem Compound Database; CID=2519,
60961. http://pubchem.ncbi. nlm.nih.gov/compound/2519,
https://pubchem.ncbi.nlm.nih.gov/compound/60961.
[2] Fredholm, Bertil B. "Adenosine, Adenosine Receptors and the Actions of Caffeine."
Pharmacology and Toxicology 76 (1994): 93-101. Print.
[3] Dubuc, Bruno. "How Drugs Affect Neurotransmitters." The Brain from Top to Bottom. 1 Jan.
2002. Web. http://thebrain.mcgill.ca.
[4] Evert, A. (2013, April 30). Caffeine in the Diet.
[5] Collins, Timothy S. "How a Shipment of Spoiled Coffee Beans Turned Into an Invention for
Coffee Decaffeination." Coffee. WordPress, 11 Feb. 2011. Web.
[6] Commercial Item Description: Coffee. (2004, August 17). Retrieved from
http://www.ams.usda.gov/AMSv1.0/getfile?dDocName=STELDEV3006417
[7] Shanbhag. "Extraction of Caffeine from Tea." (2006): n. pag. 2006. Web. 16 Apr. 2015.
[8] "Dichloromethane MSDS." Sciencelab.com, Inc. N.p., 21 May 2013. Web. 20 Apr. 2015.
http://wcam.engr.wisc.edu/Public/Safety/MSDS/Dichloromethane.pdf.
[9] Atomssa, Tadelech, and A.V. Gholap. "Characterization of Caffeine and Determination of
Caffeine in Tea Leaves Using Uv-visible Spectrometer." African Journal of Pure and
Applied Chemistry 5.1 (2011): 1-8. Academic Journals. Academic Journals, Jan. 2011.
Web. 20 Apr. 2015.
http://www.academicjournals.org/article/article1379502966_Atomssa%20and%20Gholap.p
df.
[10] Mehta, Akul. "Ultraviolet-Visible (UV-Vis) Spectroscopy Analytical Chemistry
PharmaXChange.info." PharmaXChangeinfo UltravioletVisible UVVis Spectroscopy
Principle Comments. Akul Mehta, 13 Dec. 2011. Web. 20 Apr. 2015.
[11] Honsberg, Christiana, and Stuart Bowden. "Energy of Photon." Energy of Photon.
Christiana Honsberg and Stuart Bowden, n.d. Web. 20 Apr. 2015.
[12] Blauch, David N. "Spectrophotometry." : Beer's Law. David N. Blauch, 2014. Web. 20 Apr.
2015. <http://www.chm.davidson.edu/vce/Spectrophotometry/BeersLaw.html>.
[13] Gerber, F., Krummen, M., Potgeter, H., Roth, A., Siffrin, C., & Spoendlin, C. (2004).
Practical aspects of fast reversed-phase high-performance liquid chromatography using 3μm

14. 14

particle packed columns and monolithic columns in pharmaceutical development and
production working under current good manufacturing practice. Journal of Chromatography
A, 1036(2), 127-133.
[14] http://www.chemguide.co.uk/analysis/chromatography/hplc1.gif
[15] HPLC Troubleshooting Guide https://www.sigmaaldrich.com/content/dam/sigma-
aldrich/docs/Supelco/Bulletin/4497.pdf (accessed Apr 26, 2015).
[16] General HPLC Column Care http://www.nestgrp.com/pdf/colcare.pdf (accessed Apr 26,
2015).
Appendix
Figure A1. Chromatograph of 100 ppm caffeine in distilled water from isocratic flow of 60/40
MeOH/water at a flow rate of 1.00 mL/min, C18 column, UV/Vis detection at 253 nm, and an
attenuation of 6.

15. 15

Figure A2. Chromatograph of 100 ppm caffeine in distilled water from gradient flow of 95%
H2O/5% MeOH to 40% H2O /60%MeOH over 20 minutes and back to 95% H2O /5% MeOH
over an additional 20 minutes.

16. 16

Figure A3. Chromatograph of 100 ppm caffeine in distilled water from gradient flow of 95%
H2O/5% MeOH to 40% H2O /60%MeOH over 20 minutes and back to 95% H2O /5% MeOH
over an additional 20 minutes, after the machine was cleaned using acetonitrile.