Food

Chemistry
Food Chemistry 106 (2008) 217–221
www.elsevier.com/locate/foodchem

Chlorogenic acid and caffeine contents in various commercial

brewed coffees
K. Fujioka 1, T. Shibamoto *
Department of Environmental Toxicology, University of California, One Shields Avenue, Davis, California 95616, USA

Received 27 July 2006; received in revised form 10 January 2007; accepted 29 May 2007

Abstract

Twelve commercial brewed coffees (seven regular and five decaffeinated) were analyzed for chlorogenic acids (CGA) and caffeine by
HPLC. Their pH and UV–Vis absorbances were also measured. The CGAs identified were three caffeolylquinic acids (3-CQA, 4-CQA,
and 5-CQA), three feruloylquinic acids (3-FQA, 4-FQA, and 5-FQA), and three dicaffeoylquinic acids (3,4-diCQA, 3,5-diCQA, and
4,5-diCQA). The total CGAs ranged from 5.26 mg/g to 17.1 mg/g in regular coffees and from 2.10 mg/g to 16.1 mg/g in decaffeinated
coffees. Among CGA, 5-CQA was present at the highest level, ranging from 2.13 mg/g to 7.06 mg/g coffee, and comprising 36–42%
and 37–39% of the total CGA in the regular and decaffeinated coffees, respectively. CGA isomer contents were, in decreasing order,
5-CQA > 4-CQA > 3-CQA > 5-FQA > 4-FQA > 3-FQA > 3,4-diCQA > 4,5-diCQA, 3,5-diCQA. The caffeine content in regular and
decaffeinated coffees ranged from 10.9 mg/g to 16.5 mg/g and from 0.34 mg/g to 0.47 mg/g, respectively. The pH of regular and decaf-
feinated coffees ranged from 4.95 to 5.99 and from 5.14 to 5.80, respectively. The relationship between the pH and the UV–Vis absor-
bance at 325 nm was moderately correlated (R2 = 0.7829, p < 0.001, n = 12).
Ó 2007 Elsevier Ltd. All rights reserved.

Keywords: Acid contents; Brewed coffee; Caffeine; Chlorogenic acids

1. Introduction ever, does not always have exclusively non-beneficial

results. One recent review article stated that epidemiologi-
Coffee is one of the world’s most popular beverages. It is cal and experimental studies have shown positive effects
also the most important traded commodity in the world of regular coffee drinking on various aspects of health, such
after oil (Production & Consumption, 2006). World coffee as psychoactive responses (alertness, mood change), neuro-
production grew by over 100% from 1950 to 1990 and is logical conditions (infant hyperactivity, Parkinson’s dis-
projected to grow by 0.5–1.9% by 2010. Global output is ease), metabolic disorders (diabetes, gallstones), and
expected to reach 7.0 million tons by 2010. World con- gonad and liver function (Dorea & da Costa, 2005).
sumption of coffee is projected to increase by 0.4% annu- The majority of consumers’ concerns about coffee drink-
ally from 6.7 million tons in 1998–2000 to 6.9 million ing are, however, the acid reflux symptoms caused by cof-
tons in 2010 (Coffee, 2006). fee’s acidic components, such as chlorogenic acids (CGA),
There have been numerous reports on diseases associ- and doctors tend to recommend patients with acid reflux to
ated with coffee consumption (Sandler, 1983; Schilter, limit their coffee intake. CGAs are well known secondary
Cavin, Tritscher, & Constable, 2001). Coffee drinking, how- metabolites in green coffee beans and are known to contrib-
ute to coffee’s bitterness (Campa, Doulbeau, Dussert,
*
Hamon, & Noirot, 2005). There have been many reports
Corresponding author. Tel.: +1 530 752 4523; fax: +1 530 752 3394.
E-mail address: tshibamoto@ucdavis.edu (T. Shibamoto).
on the presence of CGA in green coffee beans (Clifford,
1
Present address: Department of Environmental Science, Policy and 1979; Van der Stegen & Van Duijn, 1980). For example,
Management, University of California, Berkeley, CA 94720-3112, USA. the content of CGA in various green coffee beans (21

0308-8146/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2007.05.091
218 K. Fujioka, T. Shibamoto / Food Chemistry 106 (2008) 217–221

species) from Cameroon and Congo ranged from 0.8% to fee maker (Sunbeam Product, Inc., Boca Raton, FL)
11.9% on a dry matter basis (Campa et al., 2005). The equipped with a paper filter. The brewed coffee was imme-
CGA content in brewed coffee may be influenced by the diately cooled to room temperature in an ice bath, after
kind of coffee beans used because Arabica beans contain which the samples were stored at 5 °C until required for
less CGA than Robusta beans (Ky et al., 2001). Most com- pH determination and analysis for caffeine and CGA.
mercial brands of coffee are, however, made up of both
Arabica and Robusta beans. The roasting method might 2.3. pH and color measurement
also play an important role in the CGA content of the final
coffee product. For example, the light medium roasting The pH of each brewed coffee sample was measured
condition was found to result in the highest amount of with a Corning pH meter 430 (Corning, NY). The brewed
transformation from CGA to the corresponding lactones, coffee samples (100 ll) were diluted 10-fold and 100-fold
suggesting that this process reduced the amount of CGA with purified water and the absorbances of the resulting
in coffee (Farah, de Paulis, Trugo, & Martin, 2005). The solutions were measured at k = 276, 325, and 420 nm with
amounts of seven CGAs in green coffees were significantly a Hewlett Packard 8452A Diode Array Spectrophotometer
reduced by the degradation occurring during the roasting running UV–Visible Chemstation software (Agilent Tech-
process (Trugo & Macrae, 1984). The preparation pro- nologies, 1995–2000). Water was used as a blank.
cesses, including roasting, may, thus, play an important
role in the CGA content of the final product. In the present 2.4. Analysis of caffeine and CGA in brewed coffee samples
study, therefore, CGA levels in various commercial coffees
were investigated. Brewed coffee samples were treated with Carrez reagents
I and II to eliminate polymeric components according to a
2. Materials and methods previously reported method (Ito et al., 1983;Rincon, Mar-
tinez, & Delgado, 2003). Each brewed coffee sample (3 ml),
2.1. Coffee samples and chemicals along with 0.1 ml each of Carrez reagents I and II, and
0.8 ml of methanol, was vortex-mixed in a centrifuge tube
Various brands of commercial ground-roasted coffees and allowed to stand for 10 min. The precipitate was sepa-
were bought from a local market. Caffeine was purchased rated by centrifuging at 5000 rpm for 10 min. The solution
from Sigma–Aldrich Chemical Co. (St. Louis, MO). Chlor- was then decanted and filtered with a Acrodisc Syringe Fil-
ogenic acids (CGA) were bought from Cayman Chemical ter with 0.2 lm HT Tuffryn membrane (Pall Corporation,
Co. (Ann Arbor, MI) or a gift from TAKATA Koryo Ann Arbor, MI).
Co., Ltd. (Osaka, Japan). HPLC grade methanol and water Quantitative analyses of caffeine and CGA were per-
were purchased from Fisher Co. (Pittsburgh, PA). All formed using an Agilent 1100 model HPLC system
other chemicals and solvents were bought from reliable equipped with a Zorbax Eclipse XDB C-18 5l column
commercial sources. (150 mm  4.6 mm i.d.) and a multiple wavelength detec-
Stock solutions of caffeine (20 mM) and CGA (20 mM) tor. Mobile phase A was 10 mM citric acid and mobile
were prepared in methanol for preparation of standard phase B was methanol. The gradient mode was initially
solutions and spike analysis. set at A/B ratio of 85/15 from 0 to 5 min, then linearly
increased to 60/40 at 40 to 85 min. The flow rate was
2.2. Preparation of brewed coffee samples 1.0 ml/min. The detector was set at 325 nm for CGA and
at 276 nm for caffeine; injection volume was 5 ll.
Ground-roasted coffee (12.5 g) was brewed with 450 ml Concentrations of caffeine, CGA, caffeic acid and ferulic
of deionized water using a Mr. Coffee NCX-20 model cof- acid were calculated using the regression equation of their

Table 1
pH and UV absorbance of brewed coffees
Brand pHa UV Absorbance at k = a
276 nm 325 nm 420 nm
Regular Decaffeinated Regular Decaffeinated Regular Decaffeinated Regular Decaffeinated
A 5.99 ± 0.02 5.80 ± 0.01 0.496 ± 0.001 0.298 ± 0.001 0.306 ± 0.001 0.220 ± 0.001 0.273 ± 0.001 0.283 ± 0.004
B 5.22 ± 0.02 5.66 ± 0.01 0.720 ± 0.016 0.406 ± 0.003 0.513 ± 0.011 0.428 ± 0.002 0.397 ± 0.001 0.363 ± 0.003
C 5.26 ± 0.02 5.14 ± 0.01 0.882 ± 0.001 0.634 ± 0.001 0.712 ± 0.001 0.686 ± 0.001 0.458 ± 0.001 0.482 ± 0.005
D 5.17 ± 0.00 5.22 ± 0.01 0.738 ± 0.002 0.559 ± 0.001 0.577 ± 0.001 0.620 ± 0.001 0.384 ± 0.001 0.396 ± 0.003
E 5.12 ± 0.00 5.22 ± 0.01 0.872 ± 0.002 0.553 ± 0.001 0.774 ± 0.001 0.632 ± 0.001 0.370 ± 0.001 0.365 ± 0.001
F 4.95 ± 0.00 – 0.710 ± 0.003 – 0.621 ± 0.002 – 0.369 ± 0.001 –
G 5.21 ± 0.01 – 0.669 ± 0.001 – 0.575 ± 0.001 – 0.334 ± 0.001 –
–: Commercial samples were not available.
a
Values are means ± SD (n = 3).
 K. Fujioka, T. Shibamoto / Food Chemistry 106 (2008) 217–221 219

concentration and peak area. The limit of quantification (APCI) source operating in the positive ion mode at
was calculated as 10 times the standard deviation of the 475 °C.
lowest concentration of standard solution (50 lM). The
recovery efficient was determined with a coffee sample 3. Results and discussion
spiked with 250 lM each of caffeine, CGA, caffeic acid
and ferulic acid. Measurements were done in triplicate. 3.1. pH and color measurement of brewed coffees
Identification of caffeine and CGA in brewed coffee was
confirmed by a Hewlett Packard 1100 liquid chromato- The pH and UV–Vis absorbancies of various brewed cof-
graph interfaced to an Applied Biosystems API 2000 fees are shown in Table 1. The pH of the brewed coffees ran-
MS/MS via an atmospheric pressure chemical ionization ged from 4.95 (Coffee F) to 5.99 (Coffee A) for regular coffee
and from 5.14 (Coffee C) to 5.80 (Coffee A) for decaffein-
ated coffee. Coffee A showed the highest pH and the least
absorbance at the three different wavelengths. The differ-
ence in the pH values between lowest (4.95) and highest
Table 2
Caffeine concentrations in coffees (5.80) was 0.85, which reflects a 7.09-fold difference in H+
concentration. The relationship between the pH and the
Brand Concentration (lg/g ground coffee)
UV–Vis absorbance at 325 nm was moderately correlated
Regular Decaffeinated
(R2 = 0.7829, p < 0.001, n = 12). The H+ concentration in
A 11.5 ± 0.22 0.40 ± 0.00 brewed coffee samples appears to be related to the presence
B 13.4 ± 0.19 0.39 ± 0.01
of chlorogenic acids because their maximum UV absor-
C 15.1 ± 0.12 0.34 ± 0.00
D 13.4 ± 0.97 0.47 ± 0.01 bance is around 320–330 nm. The 420 nm wavelength has
E 16.5 ± 0.24 0.43 ± 0.00 been widely used for assessing the color intensity of brown-
F 10.9 ± 0.37 – ing reaction mixtures as it relates to the palatability of heat-
G 11.3 ± 0.68 – processed foods and beverages (Ajandouz & Puigserver,
–: Commercial samples were not available. 1999). There is no significant difference in absorbance

COOH O
O COOH OH
OH HO COOH
HO HO OH O
O OH HO
HO OH
HO
HO O
OH HO
O OH

3-Caffeoylquinic acid (3-CQA) 4-Caffeoylquinic acid (4-CQA) 5-Caffeoylquinic acid (5-CQA)

COOH O
O COOH OH H3CO COOH
OH O
HO OH
H3CO O OH HO HO OH
HO
HO O
OH H3CO OH
O
3-Feruloylquinic acid (3-FQA) 4-Feruloylquinic acid (4-FQA) 5-Feruloylquinic acid (5-CQA)

HO COOH
HO O
HO OH
O
HO OH COOH
O HO O
O
OH HO
O OH
O OH
HO
HO
3,5-diCaffeoylquinic acid (3,5-diCQA) O O
COOH
HO
O OH 4,5-diCaffeoylquinic acid (4,5-diCQA)
HO OH
O
OH
O

3,4-diCaffeoylquinic acid (3,4-diCQA)

Fig. 1. Structures of chlorogenic acids (CGA) identified in the present study.

220 K. Fujioka, T. Shibamoto / Food Chemistry 106 (2008) 217–221

Table 3
Acid concentrations in different coffees
Brand Concentration (mg/g of ground coffee)a Total
3-CQA 4-CQA 5-CQA 3-FQA 4-FQA 5-FQA 3,4-DCQA 3,5-DCQA 4,5-DCQA
Regular
A 1.32 ± 0.02 1.44 ± 0.02 2.13 ± 0.04 0.82 ± 0.00 0.84 ± 0.01 0.89 ± 0.01 0.47 ± 0.00 0.40 ± 0.00 0.28 ± 0.00 5.26 ± 0.09
B 2.01 ± 0.03 2.28 ± 0.03 3.23 ± 0.05 1.29 ± 0.00 0.21 ± 0.00 0.27 ± 0.00 0.11 ± 0.00 0.67 ± 0.00 0.73 ± 0.00 8.39 ± 0.12
C 2.93 ± 0.01 3.34 ± 0.01 4.62 ± 0.01 0.98 ± 0.00 0.53 ± 0.02 0.75 ± 0.03 0.22 ± 0.00 0.12 ± 0.00 0.16 ± 0.00 12.8 ± 0.08
D 2.77 ± 0.02 3.13 ± 0.02 4.48 ± 0.04 0.98 ± 0.00 0.34 ± 0.00 0.47 ± 0.01 0.18 ± 0.00 0.01 ± 0.00 0.11 ± 0.00 11.7 ± 0.10
E 3.95 ± 0.06 4.56 ± 0.07 6.27 ± 0.14 0.13 ± 0.00 0.62 ± 0.01 0.89 ± 0.03 0.30 ± 0.01 0.19 ± 0.01 0.20 ± 0.01 17.1 ± 0.34
F 3.78 ± 0.01 4.27 ± 0.01 7.06 ± 0.01 0.12 ± 0.00 0.29 ± 0.00 0.54 ± 0.02 0.30 ± 0.00 0.22 ± 0.00 0.28 ± 0.00 16.9 ± 0.05
G 3.43 ± 0.01 3.89 ± 0.01 6.06 ± 0.01 0.13 ± 0.00 0.28 ± 0.00 0.45 ± 0.00 0.27 ± 0.00 0.19 ± 0.00 0.25 ± 0.00 15.0 ± 0.04
Decaffeinated
A 0.45 ± 0.00 0.51 ± 0.00 0.82 ± 0.00 0.04 ± 0.00 0.09 ± 0.00 0.13 ± 0.00 0.03 ± 0.00 0.02 ± 0.00 0.01 ± 0.00 2.10 ± 0.01
B 0.83 ± 0.00 0.90 ± 0.00 1.43 ± 0.01 0.06 ± 0.00 0.16 ± 0.00 0.26 ± 0.00 0.07 ± 0.00 0.04 ± 0.00 0.05 ± 0.00 3.80 ± 0.02
C 3.22 ± 0.01 3.54 ± 0.01 5.77 ± 0.01 0.11 ± 0.00 0.70 ± 0.00 1.22 ± 0.00 0.40 ± 0.00 0.26 ± 0.00 0.41 ± 0.00 15.6 ± 0.04
D 3.42 ± 0.02 3.78 ± 0.02 6.23 ± 0.03 0.10 ± 0.00 0.53 ± 0.00 0.99 ± 0.01 0.40 ± 0.00 0.27 ± 0.00 0.41 ± 0.00 16.1 ± 0.08
E 3.26 ± 0.05 3.59 ± 0.03 5.75 ± 0.03 0.12 ± 0.01 0.60 ± 0.02 1.09 ± 0.03 0.36 ± 0.01 0.23 ± 0.01 0.37 ± 0.00 15.4 ± 0.19
a
Values are means ± SD (n = 3).

among coffee samples from different brands, although Cof- total acid content in the coffees ranged from 2.10 ±
fee A showed the lowest absorbance at all three wave- 0.01 mg/g (Coffee A, decaffeinated) to 17.1 ± 0.34 mg/g
lengths. However, no significant differences in color (Coffee E, regular) in the present study. Decaffeinated Cof-
among the coffee samples were observed. fees A and B contained much less total acid than their reg-
ular coffee counterparts, suggesting that the decaffeinating
3.2. Caffeine and chlorogenic acids (CGA) contents in process reduces acid content. On the other hand, decaffein-
various commercial brewed coffees ated Coffees C, D, and E contained higher levels of total
acid than their regular coffee counterparts. Coffee A con-
The recovery efficiencies (%) were 94.3 ± 0.23% for 3- tained the least total CGA and the fewest of the three
CGA, 102 ± 0.64% for caffeic acid, 95.8 ± 0.05% for ferulic CQA isomers both in regular and decaffeinated samples,
acid, and 97.7 ± 0.45% for caffeine. followed by Coffee B. CGA isomer contents were, in
Table 2 shows the caffeine contents of the different cof- decreasing order, 5-CQA > 4-CQA > 3-CQA > 5-FQA >
fees. The caffeine contents in regular coffee ranged from 4-FQA > 3-FQA > 3,4-diCQA > 4,5-diCQA; 3,5-diCQA.
10.9 ± 0.04 mg/g (Coffee F) to 16.5 ± 0.24 mg/g (Coffee The major CGA was 5-CQA, which comprised 36–42%
E). The caffeine contents in decaffeinated coffee ranged and 37–39% of the total CGA in the regular and decaffein-
from 0.34 ± 0.00 mg/g (Coffees C) to 0.47 ± 0.01 mg/g ated coffees, respectively, in the present study. 5-CQA was
(Coffee D). It is interesting that all decaffeinated brands detected in the regular coffee sample prepared from Coffee
of coffee contained caffeine levels of nearly 0.4 mg/g of F at the highest level of the 9 CGA isomers (7.06 ±
coffee. 0.01 mg/g ground coffee).
Fig. 1 shows the structures of CGA found in brewed cof- 5-CQA was also the major acid found in the Arabica and
fees in the present study. Table 3 shows the results of CGA Robusta green beans, where it comprised 66% and 56% of
analysis in various coffees. Fig. 2 shows a typical chromato- total chlorogenic acids, respectively (Trugo & Macrae,
gram of a brewed coffee sample run for CGA analysis. The 1984). It was also reported that the total CGA in green cof-
fee beans was reduced significantly by roasting at 205 °C for
four different time periods: light = 7 min for Arabica and
5-CQA
3-CQA 5 min for Robusta; medium = 10 min for Arabica and
7 min for Robusta; dark = 13 min for Arabica and 14 for
Detector response

Robusta or very dark = 19 min for Arabica and 16 min

4-CQA
for Robusta (Trugo & Macrae, 1984). The rate of reduction
4-FQA
was 60.9% for light, 67.7% for medium, 88.8% for dark, and
3-FQA 5-FQA
3,4-diCQA 4,5-diCQA 96.5% for very dark in Arabica and 59.7% for light, 76.4%
3,5-diCQA for medium, 93.0% for dark, and 98.0% for very dark in
Robusta. The total level of CGA was 68.8 mg/g of dry green
bean in Arabica green bean and 88.0 mg/g in Robusta green
0 10 20 30 40 bean. These levels reduced to 26.9 mg/g for light, 22.2 mg/g
Time (min) for medium, 7.71 mg/g for dark, and 2.42 mg/g for very
Fig. 2. Typical chromatogram of a brewed coffee sample (Coffee A) run dark in roasted Arabica coffee and 35.4 mg/g for light,
for CGA analysis. 20.7 mg/g for medium, 6.15 mg/g for dark, and 1.76 mg/g
 K. Fujioka, T. Shibamoto / Food Chemistry 106 (2008) 217–221 221

for very dark roasting conditions. In the present study, the Clifford, M. N. (1979). Chlorogenic acids – their complex nature and
total CGA ranged from 5.26 mg/g (Coffee A) to 17.1 mg/g routine determination in coffee beans. Food Chemistry, 4, 63–71.
Coffee, 2006. Coffee. Retrieved 2006-06-20 from the World Wide Web:
(Coffee E) in regular brewed coffees, suggesting that these http://www.fao.org/dcrep/oo6/y5143e0v.htm.
coffee beans were roasted under dark or very dark Dorea, J. G., & da Costa, H. M. (2005). Is coffee a functional food? British
conditions. Journal of Nutrition, 93, 773–782.
Ehling, S., & Shibamoto, T. (2005). Correlation of acrylamide generation
4. Conclusions in thermally processed model systems of asparagine and glucose with
color formation, amounts of pyrazines formed, and antioxidative
properties of extracts. Journal of Agricultural and Food Chemistry, 53,
Such reports indicated that the content of CGA is low- 4813–4819.
ered most significantly by varying the roasting time/temper- Farah, A., de Paulis, T., Trugo, L. C., & Martin, P. R. (2005). Effect of
ature curve. Changes to the roasting times and temperatures roasting on formation of chlorogenic acid lactones in coffee. Journal of
seem to affect CGA contents significantly in the final coffee Agricultural and Food Chemistry, 53, 1505–1531.
Ito, Y., Tonogai, Y., Ogawa, S., Yokoyama, T., Honda, T., & Iwaide, M.
products. This may explain the major differences in pH and (1983). Rapid simultaneous determination of benzoate and sorbate in
CGA content found among the commercial coffees tested. yoghurt by use of high-performance liquid chromatography. Japanese
However, vigorous treatment of food and beverages causes Journal of Dairy Food Sciences, 32, A89–A94.
reductions in palatability factors, such as flavor and color Ky, C.-L., Louarn, J., Dussert, S., Guyot, B., Hamon, S., & Noriot, M.
(Ehling & Shibamoto, 2005), although the present study (2001). Caffeine, trigonelline, chlorogenic acids and sucrose diversity in
wild Coffea arabica L. and C. canephora P. accessions. Food
showed no significant color differences. Therefore, it is Chemistry, 75, 223–230.
important to prepare a coffee with low acid content, yet Production and Consumption, 2006. Trouble brewing: The changing face
one that does not lose biologically beneficial components. of coffee production. Retrieved 2006-06-21 from the World Wide Web:
Investigation into the role of roasting conditions in content http://pubs.wri.org/pubs content_text.cfm?ContentID=1445.
variation of beneficial chemicals, such as antioxidants, and Rincon, F., Martinez, B., & Delgado, J. M. (2003). Detection of
factors influencing nitrite determination in meat. Meat Science, 65,
in acid contents of coffees is in order. 1421–1427.
Sandler, R. S. (1983). Diet and cancer. Nutrition and Cancer, 4, 273–279.
References Schilter, B., Cavin, C., Tritscher, A., & Constable, A. (2001). Health
effects and safety considerations. In R. J. Clarke & O. G. Vizthum
Ajandouz, H., & Puigserver, A. (1999). Nonenzymatic browning reaction (Eds.), Coffee (pp. 165–183). Oxford: Blackwell Science.
of essential amino acids: Effect of pH on caramelization and Maillard Trugo, L. C., & Macrae, R. (1984). A study of the effect of roasting on the
reaction kinetics. Journal of Agricultural and Food Chemistry, 47, chlorogenic acid composition of coffee using HPLC. Food Chemistry,
1786–1793. 15, 219–227.
Campa, C., Doulbeau, S., Dussert, S., Hamon, S., & Noirot, M. (2005). Van der Stegen, G. H. D., & Van Duijn, J. (1980). Analysis of chlorogenic
Qualitative relationship between caffeine and chlorogenic acid contents acids in coffee. Colloque Scientifique Internationaql sur le Cafe, 9,
among wild Coffea species. Food Chemistry, 93, 135–139. 107–112.