Effect of Brewing Temperature and Duration on Green Tea Catechin Solubilization: Basis for Production of EGC and EGCG-Enriched Fractions

Abstract

In an industrial context of producing catechin-enriched fractions by electromigration, a new technology demonstrated to be effective for concentration of the two main catechins (EGC and EGCG) of green tea, the recovery yield from tea leaves during brewing would be the most important parameter of the whole process rentability. However, the majority of the kinetic studies were carried-out on black tea, a fermented tea. Consequently, the objective of this study was to investigate the effects of temperature (50, 60, 70, 80 and 90 °C) and brewing duration (0, 5, 10, 20, 40 and 80 min) on the catechin solubilization from green tea, a non-fermented tea. The use of mathematical models revealed that there was a variable interdependence between the brewing duration and the brewing temperature on catechin and caffeine concentrations. It was possible to divide catechins in two groups, the time dependent compounds (EGC and EC) and the time/temperature dependent compounds (C, EGCG, GCG and ECG). Furthermore, the 3D models calculated to represent the evolution of the catechins and caffeine concentrations allowed to determine the best combination of time and temperature for their extraction: 50 °C during 20–40 min for time-dependent compounds, 90 °C during 80 min for the time/temperature-dependent compounds, and 70–80 °C during 20–40 min for caffeine. Furthermore, this research pointed out a very simple two-step procedure to fractionate the EGC and EGCG by modifying brewing temperature and time parameters.

Full text

Separation and Purification Technology 49 (2006) 1–9
Effect of brewing temperature and duration on green tea
catechin solubilization: Basis for production of EGC
and EGCG-enriched fractions
David Labb´
ea,b, Angelo Tremblaya,c, Laurent Bazinet a,b,∗
aInstitute of Nutraceuticals and Functional Foods (INAF), Universit´e Laval, Sainte-Foy, Que., Canada G1K 7P4
bDepartment of Food Sciences and Nutrition, Pavillon Paul Comtois, Universit´e Laval, Sainte-Foy, Que., Canada G1K 7P4
cDepartment of Social and Preventive Medicine, PEPS, Universit´e Laval, Sainte-Foy, Que., Canada G1K 7P4
Received 17 February 2005; received in revised form 22 July 2005; accepted 25 July 2005
Abstract
In an industrial context of producing catechin-enriched fractions by electromigration, a new technology demonstrated to be effective for
concentration of the two main catechins (EGC and EGCG) of green tea, the recovery yield from tea leaves during brewing would be the most
important parameter of the whole process rentability. However, the majority of the kinetic studies were carried-out on black tea, a fermented
tea. Consequently, the objective of this study was to investigate the effects of temperature (50, 60, 70, 80 and 90 ◦C) and brewing duration (0, 5,
10, 20, 40 and 80 min) on the catechin solubilization from green tea, a non-fermented tea. The use of mathematical models revealed that there
was a variable interdependence between the brewing duration and the brewing temperature on catechin and caffeine concentrations. It was
possible to divide catechins in two groups, the time dependent compounds (EGC and EC) and the time/temperature dependent compounds (C,
EGCG, GCG and ECG). Furthermore, the 3D models calculated to represent the evolution of the catechins and caffeine concentrations allowed
to determine the best combination of time and temperature for their extraction: 50 ◦C during 20–40min for time-dependent compounds, 90 ◦C
during 80 min for the time/temperature-dependent compounds, and 70–80 ◦C during 20–40 min for caffeine. Furthermore, this research pointed
out a very simple two-step procedure to fractionate the EGC and EGCG by modifying brewing temperature and time parameters.
© 2005 Elsevier B.V. All rights reserved.
Keywords: Solubilization; Catechins; Caffeine; Time; Temperature
1. Introduction
Tea is the most widely drunk beverage across the world,
after water [1,2] and it has been largely studied over the
last years to determine its brewing dynamic [3–8] and its
flavonoid components [9–13]. Flavonoids are plant sec-
ondary metabolites that are widely distributed in the plant
kingdom and that can be subdivided into six classes:
flavones, flavanones, isoflavones, flavonols, flavanols, and
anthocyanins based on the structure and conformation of
the heterocyclic oxygen ring (C ring) of the basic molecule
∗Corresponding author. Tel.: +1 418 656 2131x7445;
fax: +1 418 656 3353.
E-mail address: Laurent.Bazinet@aln.ulaval.ca (L. Bazinet).
[1]. The main classes of flavonoids found in green tea are
flavanols and more precisely catechins [14,15]. Catechins
are antioxidants [1,10,16] having a potentially beneficial
effect on the body [1,10,17–20]. The six catechins present
in green tea and known to possess biological properties are
(+)-catechin (C), (−)-epicatechin (EC), (−)-epigallocatechin
(EGC), (−)-epicatechin gallate (ECG), (−)-epigallocatechin
gallate (EGCG), and (−)-gallocatechin gallate (GCG) [1].
These molecules with high nutraceutical potential are colour-
less and water-soluble compounds which impart bitterness
and astringency to green tea infusion [1]. EGCG is regarded
as the most important of the tea catechins because of its
high content in tea and the fact that its activity is mirrored
by green tea extracts. Therefore methods for producing tea
extracts with high EGCG ratios have been developed [21–23].
1383-5866/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.seppur.2005.07.038

2D. Labb´e et al. / Separation and Purification Technology 49 (2006) 1–9
However, these methods have the inconvenience to use sol-
vent [21] or to allow the purification of small volumes [23].
In previous work we have evaluated the feasibility of a
new technology, electromigration, for selectively extracting
catechins and caffeine from a green tea solution using an
electrodialysis cell [24]. This study demonstrated the effec-
tiveness of an EGC and EGCG concentration by electrodial-
ysis treatment with an ultrafiltration (UF) membrane. In fact,
in an industrial context of producing EGCG/EGC-enriched
fraction by electromigration, an important step will be the
extraction of these catechins from tea leaves: the recovery
yield being the most important parameter of the whole pro-
cess rentability. Despite many studies have documented the
biological properties, the kinetics and the quantification of
these catechins, only a few of them were performed to maxi-
mize their content in a green tea brewing [25]. The majority of
the kinetic studies were carried-out on black tea, a fermented
tea.
The present study aimed at the optimization of the cat-
echins extraction by solubilization in water to concentrate
them and to be used as a preconcentration step before elec-
tromigration. Then the objectives of the present study are
to determine the effects of temperature (50, 60, 70, 80 and
90 ◦C) and brewing duration (5, 10, 20, 40 and 80 min) on
the solubilization of green tea catechins.
2. Materials and methods
2.1. Materials
2.1.1. Green tea
The green tea used in this study was a non-biological
Japanese green tea (lot 12423TKA) obtained from local
retailer La Girofl´
ee (Qu´
ebec City, Que., Canada). The green
tea was stocked at room temperature in a dark and dry place.
2.1.2. Catechin and caffeine standards
(+)-Catechin HPLC grade standard was obtained from
Indofine (Hillsborough, NJ) while (−)-epicatechin, (−)-
epigallocatechin, (−)-epicatechin gallate, (−)-epigallocate-
chine gallate, (−)-gallocatechin gallate and caffeine stan-
dards were from Sigma Company (Saint-Louis, MO).
2.2. Methods
2.2.1. Protocol
Twenty grams of green tea were brewed in 1000 mL
of double-distilled water to solubilize catechins (a 1:50
tea–water ratio was used according to Wang et al. [26]).
Brewing was performed at five different temperatures (50,
60, 70, 80 and 90 ◦C) in a thermostated water-bath. During
each treatment, 15 mL-samples were taken after 5, 10, 20, 40
and 80 min of brewing, cooled quickly and stored at 4 ◦C until
HPLC analyses were performed. Conductivity, pH, and cate-
chins and caffeine concentration measurements were carried-
out on refrigerated samples. Three replicates of each solubi-
lization conditions were performed in this study. The HPLC
injections were performed in duplicate for each repetition.
2.2.2. pH
The pH was measured at 4 ◦C with a pH meter model SP20
(epoxy gel combination pH electrode, Thermo Orion, West
Chester, PA).
2.2.3. Conductivity
Conductivity, which is an important factor for further
use of the extract in electromigration, was measured at 4 ◦C
with a YSI conductivity meter (model 3100–115 V, Yellow
Springs, OH) used with a YSI immersion probe (model 3417,
k= 1/cm).
2.2.4. HPLC method
Each green tea sample collected during brewing in the
different conditions was filtered through a 0.20 ␮m filter
(Aerodisc LC13 PVDF, Gelman Laboratory, Ann Arbor, MI)
and diluted with HPLC grade water to be analyzed. The
mobile phases were filtered through a 0.20 ␮m nylon filter
(Mendel Scientific Company, Guelph, Canada). The column
temperature was maintained at 40 ◦C during analyses. The
detection of analytes was performed with UV detection at
210 nm. Standard curves were calculated from a mix of cat-
echin and caffeine compounds at different concentrations:
Correlations obtained were ranging from 0.9981 to 0.9995.
The RP-HPLC method was based on the National Institute
of Standards and Technology method [27] under the same
conditions used by Labb´
eetal.[24].
2.3. Statistical analyses
The pH, conductivity, and catechin and caffeine concen-
tration data were subjected to an analysis of variance with
Sigma Stat (version 2.0 for Windows, SPSS Inc., Chicago,
IL). Nonlinear regression equations and 3D modelisation
curves were calculated for the conductivity, and catechin and
caffeine concentrations according to the brewing duration and
temperatures using Sigma Plot (version 8.01 for Windows
SPSS Inc., Chicago, IL). Statistical analysis of the calculated
theoretical regression models was carried-out using Sigma
Plot.
3. Results and discussion
3.1. pH
According to the analysis of variance, pH was influenced
by the brewing temperature (P< 0.001) but not by the brew-
ing duration (P> 0.334). The initial pH of the double-distilled
water was 5.36 ±0.04. The pH after brewing varied between
5.41 and 5.78. Whatever the duration, pH remained con-
stant during brewing with an average value of 5.58 ±0.10.

D. Labb´e et al. / Separation and Purification Technology 49 (2006) 1–9 3
Table 1
Conductivity (␮S/cm) of a green tea brewing as a function of time and temperature
Temperature (◦C) Time (min) Mean
5 10 20 40 80 Mean ±S.D.
Mean ±S.D. Mean ±S.D. Mean ±S.D. Mean ±S.D. Mean ±S.D.
50 269.2 6.2 336.0 8.9 385.0 10.8 421.3 8.9 447.4 8.6 310.0 164.0
60 300.2 8.6 352.0 14.0 402.6 12.4 435.5 5.5 461.9 14.2 325.6 169.2
70 345.0 5.3 396.4 7.7 435.3 6.2 465.4 3.9 502.0 7.2 357.6 182.8
80 392.1 23.4 425.3 5.7 456.2 8.7 479.0 14.4 520.7 14.5 379.1 190.2
90 374.9 10.8 419.2 9.7 454.0 11.1 481.3 7.4 533.0 17.0 377.3 191.9
Mean 336.3 51.2 385.8 40.0 426.6 31.7 456.5 26.9 493.0 37.1 349.9 179.3
However, its value changed according to the brewing tem-
perature: at 50 ◦C, the averaged pH value was 5.69 and this
value decreased to 5.41 at 80 ◦C, and then further increased
to pH 5.60 at 90 ◦C. Change in pH could have an influence on
the migration kinetic of catechins in electromigration, since
presence of OH groups on catechins imply a susceptibiliy to
the effect of pH and thus, an ionization of the molecule [28].
3.2. Conductivity
As shown in Table 1, conductivity substantially increased
with time (P< 0.001) and temperature (P< 0.001) of brew-
ing. The major increase in conductivity appeared just
after tea leaves were soaked in the solution: conductivity
increased from 1.35 ␮S/cm (distilled water conductivity) to
336.3 ␮S/cm after only 5 min of brewing. This increase con-
tinued until a conductivity of 426.6 ␮S/cm was reached after
20 min of brewing. Thereafter, the increase as a function of
time was less pronounced: increase of 15% between 20 and
80 min. Conductivity also slightly increased with brewing
temperature. Indeed, at 50 ◦C the average conductivity of
the brewing was 310.0 ␮S/cm and increased until a value of
379.1 ␮S/cm was reached at 80 ◦C and then was unchanged at
90 ◦C, with a value of 377.3 ␮S/cm. The conductivity behav-
ior as a function of time and temperature was modelled by
the following equation:
f(t, T )=T0+aT
(b+T)+ct +dtT (1)
where tcorresponds to time (min) and Tto temperature (◦C).
By replacing the a,b,cand dconstants by their respective
values, we obtained the following equation (R2= 0.991):
f(t, T )=−112.9392 +447.1614T
1.7646 +T+1.6316t+0.0096tT
The drastic increase in conductivity might be explained by
the fact that crushed tea leaves were used instead of complete
leaves. Crushing tea leaves can increase the diffusion speed
that might reach six times the speed constant k[4]. In addi-
tion, by maximizing the diffusion surface, the small pieces of
leaves with damaged cell walls immediately released miner-
als in solution that cause an increase in conductivity. There-
after, the diffusion of the small tea leaf component into the
solution tends to an equilibrium dictated by the osmotic pres-
sure surrounding the leaves. Mixing the brewing breaks this
equilibrium and allows a greater solubilization of these com-
ponents. However, at about 20 min of brewing, the whole
environment is saturated in those elements and only a small
amount of the tea leaves components are able to be solu-
bilized. Finally, a relatively important conductivity increase
was observed when temperature rose. This effect may be
caused by an increase in energy facilitating the ion diffu-
sion and consequently the component extraction. A rise of
temperature increases the ionic mobility and thus accelerates
the diffusion speed of these ionic species [29,30].
3.3. Catechins and caffeine solubilization kinetics
It arised from the results of statistical analysis that catechin
and caffeine solubilization highly depended on the brewing
duration (P< 0.001) and, according to the components, to
the brewing temperature (P< 0.001) (Table 2). Catechin and
caffein concentrations in a tea brewing as a function of time
and temperature were modelled according to the following
equation:
f(t, T )=t0+a
[1 +e−(T−T0)/b]+ct +dtT (2)
where tcorresponds to time (min) and Tto temperature (◦C).
The a,b,c, and dcorresponding values for each of the
catechin and caffeine 3D models are presented in Table 3.
3.3.1. Catechins solubilization kinetics
According to their solubilization in response to the
brewing time and temperature, catechins were divided
in two groups, the time dependent compounds and the
time/temperature dependent compounds.
Two compounds were identified as time dependent
compound: EGC and EC. The EGC and EC solubiliza-
tion increased rapidly between 0 and 20 min and after
stabilized, and this whatever the brewing temperature
(P> 0.123 and P< 0.046, respectively for EGC and EC)
(Figs. 1 and 2). Hence, at 70 ◦C, EGC concentration
increased from 602.2 ␮g/mL at 5 min to reach its maximum
at 849.5 ␮g/mL after 20 min while at one time its concentra-
tion was similar whatever the temperature. EC solubilization

4D. Labb´e et al. / Separation and Purification Technology 49 (2006) 1–9
Table 2
Individual catechin, total catechin and caffeine concentrations (␮g/mL) in a green tea brewing as a function of time and temperature
Temperature (◦C) Time (min) EGC C Caff EC EGCG GCG ECG Total catechin
Mean ±S.D. Mean ±S.D. Mean ±S.D. Mean ±S.D. Mean ±S.D. Mean ±S.D. Mean ±S.D. Sum
50 5 608.3 75.9 4.6 1.0 147.8 17.8 75.6 12.3 15.5 26.9 0.0 0.0 0.0 0.0 704.0
10 739.7 131.3 5.9 0.8 170.8 30.3 88.7 15.6 77.2 30.9 0.0 0.0 0.0 0.0 911.5
20 1004.3 333.6 6.3 1.4 241.9 65.6 120.6 35.3 245.3 46.4 0.0 0.0 7.6 8.9 1384.1
40 1229.6 325.0 10.5 3.4 331.7 79.2 154.7 40.2 489.3 95.7 6.7 11.5 27.0 14.9 1917.8
80 1160.6 52.4 10.0 0.4 337.8 26.4 147.8 11.4 542.6 41.7 6.6 11.4 33.9 25.5 1901.4
60 5 508.3 92.0 4.2 0.0 128.1 35.3 63.9 16.2 104.6 53.1 0.0 0.0 0.0 0.0 681.1
10 684.8 103.3 5.5 0.7 183.7 32.3 82.7 12.3 177.1 63.9 0.0 0.0 10.8 7.9 961.0
20 1045.4 92.0 9.5 0.7 313.2 39.8 123.5 12.5 557.2 116.6 6.5 11.3 48.8 15.6 1791.1
40 1181.5 144.6 11.5 2.5 380.9 65.8 137.7 18.8 858.7 240.0 0.0 0.0 79.4 26.8 2268.9
80 1157.6 128.5 11.0 1.9 369.9 47.5 136.8 19.7 634.6 337.8 0.0 0.0 34.0 44.4 1973.9
70 5 602.2 115.2 5.8 1.4 232.7 35.0 72.8 10.2 347.8 80.1 0.0 0.0 30.9 7.5 1059.5
10 886.0 21.7 8.7 0.9 327.1 10.2 101.5 4.2 613.3 66.6 0.0 0.0 56.0 9.5 1665.6
20 849.5 136.3 8.9 2.4 296.0 53.3 96.1 16.7 702.0 181.8 11.2 19.3 69.4 21.6 1737.0
40 1220.5 78.3 13.6 1.8 414.6 31.9 137.9 10.2 999.0 156.8 40.6 4.8 88.1 23.5 2499.7
80 1276.5 131.2 15.1 2.2 443.3 54.1 149.8 18.1 1208.3 150.1 57.2 5.1 125.7 15.1 2832.6
80 5 603.4 178.9 6.8 2.7 305.4 108.4 85.2 30.9 342.9 201.1 0.0 0.0 29.6 29.1 1067.9
10 1172.0 243.0 10.6 0.4 408.3 28.6 148.1 20.7 483.4 327.6 19.0 17.8 32.7 37.1 1865.8
20 1359.6 185.7 13.9 1.1 461.5 39.9 164.9 33.7 930.0 266.2 40.2 17.6 45.7 59.8 2554.3
40 1212.2 340.3 14.6 4.8 408.5 112.5 145.0 45.2 1024.0 287.7 61.1 22.5 97.5 27.2 2554.4
80 1307.4 255.6 20.4 3.8 457.4 72.9 161.9 40.1 1137.2 96.9 87.5 11.8 103.8 5.5 2818.2
90 5 705.9 61.0 7.8 1.0 328.2 40.3 84.9 9.5 489.6 71.6 9.6 16.7 48.1 3.0 1345.9
10 893.9 119.8 10.1 1.5 348.6 46.9 104.5 13.2 763.8 184.8 42.5 7.8 62.9 8.5 1877.5
20 1116.5 82.5 15.1 1.0 391.8 32.6 129.2 10.6 1039.9 53.9 63.5 1.4 97.9 24.0 2462.0
40 1019.6 227.9 17.1 4.6 359.3 90.8 119.1 27.2 1071.4 331.7 99.7 23.7 98.3 44.4 2425.2
80 1153.5 101.5 28.1 3.0 433.0 50.7 140.5 14.7 1282.7 180.3 186.0 18.1 133.9 36.5 2924.6

D. Labb´e et al. / Separation and Purification Technology 49 (2006) 1–9 5
Table 3
Values of constants for 3D equation models
Constant EGC C Caff EC EGCG GCG ECG
t0994.8 −4.4 132.5 118.5 −131.7 −642.9 −38.0
T0−12.4 51.2 −9.7 −13.2 0.4 72.8 −35.7
a−9102.2 1.4 −3087.3 −1387.4 −1408.8 825.8 −1428.6
b−6.3 −0.3 −4.6 −5.8 −7.8 −63.8 −10.9
c2.0 0.1 2.9 0.2 12.3 0.2 1.3
d0.0120 0.0020 0.0120 0.0020 0.0480 0.0540 0.0080
R20.941 0.809 0.904 0.921 0.912 0.911 0.831
Proba<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
aProbability level of the model.
Fig. 1. EGC concentration (␮g/mL) in a green tea brewing as a function of
time and temperature.
Fig. 2. EC concentration (␮g/mL) in a green tea brewing as a function of
time and temperature.
followed the same trends but at lower concentrations: at
80 ◦C, EC concentrations increased from 85.2 ␮g/mL at
5 min to 164.9 ␮g/mL at 20 min, and to an average value of
153.5 afterwards. These trends were very well fitted by the
3D model equation (R2= 0.941 and 0.921, respectively for
EGC and EC, Table 3). Consequently, the best combination
of extraction for both EGC and EC compounds are 50 ◦C for a
brewing duration between 20 and 40 min: these brewing con-
ditions allow to extract the main part of these time dependent
compounds.
Four catechins had a solubilization kinetic corresponding
to time/temperature dependent compounds: C, EGCG, GCG
and ECG. However the solubilization according to the brew-
ing temperature was quite different for these four compounds:
from very dependent (C and GCG) to moderate (EGCG and
ECG). The 3D models satisfactorily described these behav-
iors (R2ranging from 0.809 to 0.912, Table 3).
The C and GCG solubilization were very dependent of
temperature (P< 0.001) and brewing duration (P< 0.001).
Catechin is naturally present in solution at a small concen-
tration; concentrations ranging from 4.2 to 28.1 ␮g/mL were
thus observed (Table 2). At 50 ◦C, catechin concentration
increased from 4.6 ␮g/mL at 5 min to 10.5␮g/mL at 40 min
and then remained constant (Fig. 3). However, at 90 ◦C, cat-
echin concentration increased from 7.8 ␮g/mL at 5 min to
17.1 ␮g/mL at 40 min and reached its final and maximal value
of 28.1 ␮g/mL after 80 min of brewing. It was obvious that the
brewing duration and temperature exerted a synergistic effect
on catechin concentration. Catechin concentration, all along
the brewing process, increased in a relatively linear manner
between 50 and 80 ◦C, to drastically increase at 90◦C. In the
same way, at 50 and 60◦C, GCG concentrations were unde-
tectable by HPLC or were at the detection threshold (Table 2).
At 70 ◦C, GCG concentration remained practically null until
20 min of brewing time, and then increased to 40.6 ␮g/mL at
40 min and finally reached 57.2␮g/mL after 80 min (Fig. 4).
At 90 ◦C, GCG concentration increased from 9.6 ␮g/mL at
5 min to 63.5 ␮g/mL at 20 min to reach 186.0 ␮g/mL after
80 min.
The EGCG and ECG solubilization increased with the
brewing duration (P< 0.001) and in a moderate manner with
brewing temperature rise (P< 0.001). At 50 ◦C, EGCG con-
centration increased from 15.5 ␮g/mL at 5 min to reach an
intermediate value of 245.3 ␮g/mL at 20 min, and finally its

6D. Labb´e et al. / Separation and Purification Technology 49 (2006) 1–9
Fig. 3. C concentration (␮g/mL) in a green tea brewing as a function of time
and temperature.
maximal concentration of 489.3 ␮g/mL after 40 min (Fig. 5).
At 90 ◦C, the EGCG concentration increased constantly from
489.6 ␮g/mL at 5 min, to 763.8 ␮g/mL at 10 min, and reached
a maximum of 1071.4 ␮g/mL after 40 min. It is interesting to
note that at 50 or 60 ◦C and in a brewing time under 10 min,
EGCG concentration was very low. However, at tempera-
ture of 70, 80 or 90 ◦C, higher concentrations of EGCG were
in solution. For ECG, at 50 ◦C, its concentration was unde-
tectable until 20 min of brewing, where it reached 7.6␮g/mL
and this value continued to increase to 33.9 ␮g/mL at 80 min
(Fig. 6). The duration of undetectability decreased to 10 min
at 60 ◦C while at 70 ◦C the ECG was detectable after only
5 min. Furthermore its concentration at 5 min increased up
Fig. 4. GCG concentration (␮g/mL) in a green tea brewing as a function of
time and temperature.
Fig. 5. EGCG concentration (␮g/mL) in a green tea brewing as a function
of time and temperature.
to 48.1 ␮g/mL at 90 ◦C. Hence, at 90◦C, ECG concentra-
tion increased from 48.1 ␮g/mL at 5 min to 97.9 ␮g/mL at
20 min and it reached its maximum of 133.9 ␮g/mL after
80 min.
For all the time/temperature dependent compounds, the
best extraction combination determined in this study is at
90 ◦C during 80 min. However, the extraction of these com-
pounds seems to continue to increase with temperature over
the limit of 90 ◦C. Furthermore, very few information was
present in the literature to explain such a dependence of cat-
echins to time and/or temperature. Recently, Sharma et al.
[31], observed that catechins, especially EGCG, EGC and
EC, showed marked differences when extracted at different
Fig. 6. ECG concentration (␮g/mL) in a green tea brewing as a function of
time and temperature.

D. Labb´e et al. / Separation and Purification Technology 49 (2006) 1–9 7
Fig. 7. Caffeine concentration (␮g/mL) in a green tea brewing as a function
of time and temperature.
temperatures. Price and spitzer explained the effect of temper-
ature on catechin diffusion by a difference in their molecular
weight [25].
3.3.2. Caffeine (Caf)
Caffeine solubilization slightly increased with a rise in
brewing temperature (P< 0.001) and increased in a moder-
ate manner with infusion duration (P< 0.001). The caffeine
concentration was influenced to a small extent by temper-
ature and was rapidly solubilized in solution to its final
concentration. Moreover, its concentration had reached a
maximum after 40 min of brewing, whatever the brewing tem-
perature. Hence, at 50 ◦C, caffeine concentration increased
slowly from 147.8 ␮g/mL at 5 min to 331.7 ␮g/mL at 40 min
and then remained constant. In a similar way, at 90◦C, caf-
feine concentration increased from 328.2 ␮g/mL at 0 min to
348.6 ␮g/mL after 10 min and then slightly increased to its
maximum value of 359.3 ␮g/mL after 40min. This trend was
confirmed by the 3D model equation (R2= 0.904, Fig. 7). For
caffeine, which is an important compound in tea, for its ener-
getic effect, the best extraction condition was at 70–80◦C
during a brewing duration of 20–40 min.
The results obtained for total catechins from HPLC
(Table 2) are in agreement with available literature. For
instance, Wang et al. [26] showed that with 3 g of green tea
leaves in 150 mL of distilled water at ebullition temperature
during 5 min, 84.6 mg/100 mL of catechins were detected.
An extraction using a slight modification of the Suematsu et
al. method [2] allowed to determine that catechins represent
14.6% of the green tea dry total mass (Sencha High Grade).
A study by Khokhar et al. allowed to extract 849 mg/L of cat-
echins with 1 g of Japanese green tea in 100 mL of water at
100 ◦C during 5 min [32]. Moreover, the study of Fern´
andez
et al. showed that, on a weight/weight dry basis, EGC repre-
sents 3.187%, C 0.201%, Caf 2.346%, EC 0.871%, EGCG
2.054% and ECG 0.304% of green tea [12].
However, tea catechins content is not constant from a green
tea to another, from a country to another, and from a season
to the other. Hence, a study reported that the more tea plants
are exposed to the sun, the more catechins will be formed
[18]. Moreover, technological parameters might influence tea
catechins content during solubilization [8]. In our study, a
5-min brewing allowed a total catechin extraction of 499.69,
1059.51 and 1341.48 ␮g/mL at 50, 70 and 90 ◦C, respectively.
We had not only observed a great interdependence between
brewing temperature and the total catechins in solution but
also with the brewing time. At 50 ◦C, catechin concentra-
tion of 499.69, 911.47, 1382.43 and 1917.81 ␮g/mL were
obtained after 5, 10, 20 and 40 min of brewing to reach a max-
imum of 1901.40 ␮g/mL after 80min. A similar progression
was observed at 70 ◦C where after 5 min a catechin con-
centration of 1959.51 ␮g/mL was obtained, 1665.59 ␮g/mL
after 10 min, 1737.00 ␮g/mL after 20 min, 2499.68 ␮g/mL
after 40 min to slightly increased to 2794.48 ␮g/mL after
80 min. These results obtained for the solubilization kinet-
ics are concordant with those reported by Price and Spitzer
[25]. These authors observed by following the half-life and
constants of rate calculations, that ungallated flavanols (EC
and EGC) infused faster than the two gallated flavanols (ECG
and EGCG). These results were obtained from a first order
modelisation curve (flavanol concentration versus brewing
time) issued from the Spiro model for a black tea leaves
brewing [5]. Furthermore, results and 3D models obtained in
this present study contradict the fact, reported by Price and
Spitzer [25], that EGC solubilization kinetic is temperature
dependent. Indeed, the EGC solubilization does not depend
mainly on temperature. They also demonstrated that the EGC
and EGCG brewing rate is more dependent on temperature
than ECG and EC, which is in contradiction with the present
results. These differences could be due to the different tea
used for brewing or to the partial information given by the
first order model used by these authors: since these models
did not take in account the possible synergistic effect of time
and temperature, this important effect was neglected.
A supplementary information demonstrated from the
present study, was the differential kinetics of the catechins
according to the brewing time and temperature. According
to these results, it should be possible to modify the catechin
content in the tea solution by adjusting time and tempera-
ture parameters during brewing to increase the concentration
of one or more catechin types: especially EGC and EGCG,
which represents more than 80% of the total catechin content
of green tea brewing. Hence, if the catechin concentration
data obtained in this study are represented at the temperature
limit conditions, 50 and 90 ◦C, a shift can be observed in the
EGCG kinetic, confirming its time/temperature dependency.
On a theoretical point of view and according to our results,
an infusion at 50 ◦C during 10 min (Fig. 8a) would allow to
extract the main part of the EGC and a small part of the other
catechins. With a subsequent brewing of the same tea leaves

8D. Labb´e et al. / Separation and Purification Technology 49 (2006) 1–9
Fig. 8. Evolution of catechin concentrations (␮g/mL) in a green tea brewing
(a) at 50 ◦C and (b) at 90◦C.
at 90 ◦C during 10 min (Fig. 8b), it would then be possible to
extract the majority of the EGCG without the other catechins
previously extracted, and mainly EGC. This two-step proce-
dure would allow the preconcentration of EGCG and EGC
in two different solutions: the first brewing solution being
enriched in EGC and the second in EGCG. Since EGCG is
the most abundant catechin in green tea and has received the
most attention in the literature for its antioxidant properties,
it would be interesting to produce an EGCG-enriched frac-
tion. Amongst others, some attention has been given recently
to the possible beneficial effects of EGCG for the treatment
of obesity [33,34], its neuroprotective actions against neu-
rodegenerative diseases [35], its anti-inflammatory properties
[36] as well as its protective action against cancer [37]. The
procedure proposed in the present work would allow the pro-
duction, in a simple way and without any solvent, of EGC
and EGCG-enriched fractions.
4. Conclusion
The results of this study showed that catechins have differ-
ent solubilization kinetics according to the brewing duration
and temperature. Consequently, it was possible to divide cat-
echins in two groups, the time dependent compounds (EGC
and EC) and the time/temperature dependent compounds (C,
EGCG, GCG and ECG). Furthermore, the 3D models calcu-
lated to represent the evolution of the catechins and caffeine
concentrations allowed to determine the best combination of
time and temperature for their extraction. It appeared that
for EGC and EC the best combination was at 50 ◦C for a
brewing duration ranging from 20 to 40 min while for all the
time/temperature dependent compounds, the best extraction
combination was at 90 ◦C during 80 min. For caffeine, which
is an important compound in tea, for its energetic effect, the
best extraction conditions were determined to be at 70–80 ◦C
during a brewing duration of 20–40 min.
On an industrial point of view, this research pointed out
a very simple procedure to fractionate the differents cate-
chins by modifing brewing temperature and time parameters.
Experiments are currently in progress on the catechins frac-
tionation with consecutive infusions in different temperature
and time conditions to confirm this assumption.
Acknowledgements
The authors would like to thank Monica Araya-Farias and
Alain Gaudreau, research professionals at Laval University,
for their technical help. The financial support of the Natu-
ral Sciences and Engineering Research Council of Canada
(NSERC) is also acknowledged.
References
[1] H. Wang, G.J. Provan, K. Helliwell, Tea flavonoids: their func-
tions, utilisation and analysis, Trends Food Sci. Technol. 11 (2000)
152–160.
[2] T. Goto, Y. Yoshida, Simultaneous analysis of individual catechins
and caffeine in green tea, Meth. Enzymol. 299 (1999) 107–113.
[3] A.G.F. Stapley, Modelling the kinetics of tea and coffee infusion, J.
Sci. Food Agric. 82 (2002) 1661–1671.
[4] W.E. Price, M. Spiro, Kinetics and equilibria of tea infusion: rates of
extraction of theaflavin, caffeine and theobromine from several whole
teas and sieved fractions, J. Sci. Food Agric. 36 (1985) 1309–1314.
[5] M. Spiro, D.S. Jago, Kinetics and equilibria of tea infusion, J. Chem.
Soc. Faraday Trans. 1 78 (1982) 295–305.
[6] M. Spiro, S. Siddique, Kinetics and equilibria of tea infusion: kinet-
ics of extraction of theaflavins, thearubigins and caffeine from Koon-
song Broken Pekoe, J. Sci. Food Agric. 32 (1981) 1135–1139.
[7] M. Spiro, S. Siddique, Kinetics and equilibria of tea infusion. Anal-
ysis and partition constants of theaflavins, thearubigins, and caffeine
in Koosong Broken Pekoe, J. Sci. Food Agric. 32 (1981) 1027–1032.
[8] C.P. Natarajan, S. Ramamani, D.E. Leelavathi, R. Shakunthala, D.S.
Bhatia, Y. Subrahmanyan, Studies on the brewing of tea, Food Sci.
11 (1962) 321–332.
[9] H. Sakakibara, Y. Honda, S. Nakagawa, H. Ashida, K. Kanazawa,
Simultaneous determination of all polyphenols in vegetables, fruits,
and teas, J. Agric. Food Chem. 51 (2003) 571–581.
[10] H.M. Merken, G.R. Beecher, Measurement of food flavonoids by
high-performance liquid chromatography: a review, J. Agric. Food
Chem. 48 (2000) 577–599.
[11] H. Wang, K. Helliwell, Epimerisation of catechins in green tea infu-
sions, Food Chem. 70 (2000) 337–344.
[12] P.L. Fern´
andez, M.J. Martin, A.G. Gonz´
alez, F. Pablos, HPLC deter-
mination of catechins and caffeine in tea. Differentiation of green,
black and instant teas, Analyst 125 (2000) 421–425.
[13] W.E. Bronner, G.R. Beecher, Method for determining the content of
catechins in tea infusions by high-performance liquid chromatogra-
phy, J. Chromatogr. A 805 (1998) 137–142.

D. Labb´e et al. / Separation and Purification Technology 49 (2006) 1–9 9
[14] D.A. Balentine, S.A. Wiseman, L.C.M. Bouwens, The chemistry of
tea flavonoids, Crit. Rev. Food Sci. Nutr. 37 (1997) 693–704.
[15] G.W. Sanderson, The chemistry of tea and tea manufacturing, in:
V.C. Runeckles (Ed.), Structural and Functional Aspects of Phyto-
chemistry, Academic Press, New York, 1972, pp. 247–316.
[16] W.E. Price, J.C. Spitzer, Variation in the amounts of individual fla-
vanols in a range of green teas, Food Chem. 47 (1993) 271–276.
[17] K.T. Chung, C.I. Wei, M.G. Johnson, Are tannins a double-edged
sword in biology and health, Trends Food Sci. Technol. 9 (1998)
168–175.
[18] M. Kumamoto, T. Sonda, Evaluation of the antioxidative activity of
tea by an oxygen electrode method, Biosci. Biotechnol. Biochem.
62 (1998) 175–177.
[19] K. Horikawa, L. Mohri, Y. Tanaka, H. Tokiwa, Moderate inhibi-
tion of mutagenicity and carcinogenicity of benzo[a]pyrene, 1,6-
dinitropyrene and 3,9-dinitrofluoranthene by Chinese medicinal
herbs, Mutagenesis 9 (1994) 523–526.
[20] S. Kono, M. Ikeda, S. Tokudome, M. Kuratsume, A case-control
study of gastric cancer and diet in northern Kyushu, Japan, Jpn. J.
Cancer Res. 79 (1988) 1067–1074.
[21] E.L. Copland, M.N. Clifford, C.M. Williams, Preparation of (−)-
epigallocatechin gallate from commercial green tea by caffeine pre-
cipitation and solvent partition, Food Chem. 61 (1998) 81–87.
[22] Q.Z. Du, M.J. Li, Q. Cheng, C.Y. Lu, The study on separation
catechins from tea leaves and transforming gallate-catechins into
nongallate-catechins, Res. Dev. Basic Agric. High Technol. 1 (1997)
40–47.
[23] Q.Z. Du, C.Q. Ke, Y. Ito, Separation of epigallocatechin gallate and
gallocatechin gallate using multipleinstruments connected in series,
J. Liq. Chromatogr. Rel. Technol. 21 (1998) 203–208.
[24] D. Labb´
e, M. Araya-Farias, A. Tremblay, L. Bazinet, Electromigra-
tion feasibility of green tea catechins, J. Membr. Sci. 254 (2005),
101–109.
[25] W.E. Price, J.C. Spitzer, The kinetics of extraction of individual
flavanols and caffeine from a Japanese green tea (Sen Cha Uji Tsuyu)
as a function of temperature, Food Chem. 50 (1994) 19–23.
[26] H. Wang, K. Helliwell, X. You, Isocratic elution system for the
determination of catechins, caffeine and gallic acid in green tea using
HPLC, Food Chem. 68 (2000) 115–121.
[27] J.J. Dalluge, B.C. Nelson, J.B. Thomas, L.C. Sander, Selection of
column and gradient elution system for the separation of catechins
in green tea using high-performance liquid chromatography, J. Chro-
matogr. A 793 (1998) 265–274.
[28] M. Friedman, H.S. J¨
urgens, Effect of pH on the stability of
plant phenolic compounds, J. Agric. Food Chem. 48 (2000) 2101–
2110.
[29] L. Bazinet, F. Lamarche, R. Labrecque, R. Toupin, M. Boulet, D.
Ippersiel, Electroacidification of soybean proteins for production of
isolate, Food Technol. 51 (9) (1997) 52–56, 58, 60.
[30] J.O.M. Bockris, A.K.N.R. Reddy, Ion transport in solutions, in: Mod-
ern Electrochemistry, Plenum Rosetta, New York, 1993, pp. 287–
460.
[31] V. Sharma, A. Gulati, S.D. Ravindranath, Extractability of tea cat-
echins as a function of manufacture procedure and temperature of
infusion, Food Chem. 93 (2005) 141–148.
[32] S. Khokhar, S. Venema, P.C.H. Hollman, M. Dekker, W. Jongen,
A RP-HPLC method for the determination of tea catechins, Cancer
Lett. 114 (1997) 171–172.
[33] A. Dulloo, C. Duret, L. Giradier, Efficacy of a green tea extract
rich in catechin polyphenols and caffeine in increasing 24-h energy
expenditure and fat oxidation in humans, Am. J. Clin. Nutr. 70 (1999)
1040–1045.
[34] Y.-H. Kao, A. Hiipakka, S. Liao, Modulation of endocrine systems
and food intake by green tea epigallocatechin gallate, Endocrinology
141 (2000) 980–987.
[35] S. Mandel, O. Weinreb, T. Amit, M.B. Youdim, Cell signaling
pathways in the neuroprotective actions of the green tea polyphe-
nol (−)-epigallocatechin-3-gallate: implications for neurodegenera-
tive diseases, J. Neurochem. 88 (6) (2004) 1555–1569.
[36] D.S. Wheeler, J.D. Catravas, K. Odoms, A. Denenberg, V. Mal-
hotra, H.R. Wong, Epigallocatechin-3-gallate, a green tea-derived
polyphenol, inhibits il-1beta-dependent proinflammatory signal trans-
duction in cultured respiratory epithelial cells, J. Nutr. 134 (5) (2004)
1039–1044.
[37] H.N. Yu, J.J. Yin, S.R. Shen, Growth inhibition of prostate cancer
cells by epigallocatechin gallate in the presence of Cu2+, J. Agric.
Food Chem. 52 (3) (2004) 462–466.