Cream formation in a semifermented tea
Yuh Chung Chao
1and Been Huang Chiang
2*
1Department of Food Nutrition, China Junior College of Medical Technology, Tainan, Taiwan, Republic of China 2Graduate Institute of Food Science and Technology, National Taiwan University, Taipei, Taiwan, Republic of China
Abstract: The formation of cream in Paochung tea, a popular semifermented tea, which undergoes a lesser degree of enzymic oxidation during manufacture, was investigated at various extraction tem-peratures, extraction times, pHs and leaf/water ratios. The primary components of Paochung tea cream were catechins (30%), caffeine (20%) and protein (16%). (ÿ)-Epigallocatechin gallate and (ÿ)-epicatechin gallate were the major catechins precipitated during creaming, constituting 19% and 5% of the tea cream respectively. The amount of tea cream produced and its composition were in¯uenced by extraction temperature and pH. The tea leaf/water ratio determined the amount of tea cream formed but did not affect the cream composition. Catechins were considered to be the key component in tea cream. They interacted with caffeine and protein to induce tea cream formation. # 1999 Society of Chemical Industry
Keywords: cream; semifermented tea; Paochung tea; catechins
INTRODUCTION
It has been observed that the clarity of a tea infusion changes on cooling the hot infusion down to about 20±30°C, and keeping it further in this temperature range or lower may cause the production of precipi-tates and/or suspensions which are known as `tea cream'. The development of tea cream in such tea drinks as iced teas is considered an undesirable phenomenon, because it not only creates a fouling problem during the concentration process but also limits the application of tea in some drinks owing to its unattractive appearance.1±3 Most research related to tea cream formation has focused on black tea.1±5 Roberts6and Smith1 reported that the main compo-nents of black tea cream were caffeine and oxidised polyphenols such as thearubigins and thea¯avins. The extent of cream formation was dependent upon the amount of thearubigins in the black tea infusion.7 Non-caffeine nitrogen compounds, including protein and humic acid-like substances,4 as well as a lipid complex comprising components such as 1-triaconta-nol, a-spinasterol and dihydro-a-spinasterol, have also been identi®ed in black tea cream.8
The factors affecting cream formation in black tea include the extraction temperature, the pH and the concentration of the infusion.1±5 The extraction of black tea at a temperature below 35°C would produce an infusion incapable of creaming.3 When the pH of the solvent was maintained at 4, the maximum amount of black tea cream was formed.1
The extent of enzymic oxidation during the manu-facture of tea leaves affects the composition and
sensory properties of the tea infusion. However, little information is available concerning cream formation in semifermented teas, which are processed under a lesser degree of enzymic oxidation. It is important to understand the similarities and differences in cream composition and cream formation behaviour between semifermented and black tea. The objectives of this study were to investigate the composition of semifer-mented tea cream, the roles of tea components in the cream and the effects of the extraction conditions upon cream formation.
MATERIALS AND METHODS Tea leaves
Paochung tea, harvested and processed in November 1992 at Tao Yuin county of northern Taiwan, was used for this study. For manufacturing this semi-fermented tea, fresh tea leaves (TTES12) were solar withered for ca 20min and then indoor withered for ca 8h. After panning at 270°C for 6min, the leaves were rolled for ca 7min and dried at 95°C for 10min in an oven (primary drying). The semi-dried leaves were subjected to ball rolling and then dried again in an oven to a ®nal moisture content of 4%. The tea leaves were stored in a freezer at ÿ10°C to maintain their quality.
Preparation of the tea infusion
The tea leaves were ground and passed through a 140 mesh screen to remove ®ne particles before extraction. The ground leaves were extracted with distilled water *Correspondence to: Been Huang Chiang, Graduate Institute of Food Science and Technology, National Taiwan University, Taipei, Taiwan, Republic of China
Contract/grant sponsor: Council of Agriculture, ROC; contract/grant number: 83-28-08(9) (Received 5 January 1998; revised version received 22 March 1999; accepted 24 May 1999)
with various leaf/water ratios at 90°C for 20min. The extract was ®ltered through a 140 mesh screen and then quickly cooled down to 30°C or lower with a glass condenser using water (25°C) as a cooling medium and held in a 2°C water bath for 2h to allow the development of the tea cream. The tea infusion was centrifuged at 12500g for 30min to obtain a clari®ed infusion.
Analytical
The constituents of the tea cream were determined by calculating the contents in the original and clari®ed infusions and then making a comparison. The total solids contents of the original and clari®ed infusions were determined by drying 20ml of the infusions at 90°C.9 For determining the caffeine content, the sample solution was adjusted to 5±30mgmlÿ1 of caffeine concentration with distilled water. After add-ing 4% (w/v) poly(vinylpyrrolidone) (PVPP) (Sigma Co, St Louis, MO, USA) to the sample solution, the mixture was held at room temperature (25°C) for 30min and ®ltered through a Whatman No 41 ®lter paper to remove the polyphenols. The caffeine content of the mixture was determined by measuring its optical absorbance at 276nm using pure caffeine as a standard.10
The total catechins content of the original and clari-®ed infusions was determined using the HCl±vanillin method.11 The catechins content of the sample solu-tion was adjusted to 100±500mgmlÿ1 with distilled water. After adding 5ml of a 4% vanillin solution and 3ml of concentrated HCl to a 1ml sample solution, the catechins content in the mixture was determined by measuring the optical absorbance at 500nm using ()-catechin (Sigma Co) as a standard. For determin-ing the polyphenolic compounds, the tea infusion was diluted 20 times with distilled water, ®ltered through a 0.45mm ®lter and analysed using HPLC.12 The HPLC system (ICI Instruments, Victoria, Australia) consisted of two pumps (LC 1100) and a variable-wavelength UV detector (LC 1200). The separation was performed on an Inertsil 5 ODS-80A (250mm 4.6mm) column (Vercopak, Taiwan) at 40°C. Linear gradient elution was carried out at 1mlminÿ1¯ow rate from 99% solvent A (0.1% phosphoric acid solution containing 0.2% acetonitrile and 5% N,N-dimethyl-formamide) and 1% solvent B (acetonitrile) after injection to 30% solvent A and 70% solvent B at 60min. The response was determined at 280nm. The catechins kit containing pure ()-catechin, (ÿ)-epi-gallocatechin ((ÿ)-EGC), (ÿ)-epicatechin ((ÿ)-EC), (ÿ)-epigallocatechin gallate ((ÿ)-EGCG) and (ÿ)-epicatechin gallate ((ÿ)-ECG) was purchased from Nuhunakoshi Co (Tokyo, Japan). Pure gallic acid, rutin and quercetin were from Sigma Co.
For the determination of protein and pectin con-tents, 5ml of tea infusion was mixed with 25ml of 95% alcohol for 30min and centrifuged at 3000g for 20min, when the precipitate was redissolved in 5ml of distilled water. The above procedures were repeated
twice. The protein content in the precipitate was determined by the Lowry method13 using bovine serum albumin (BSA) as a standard. The pectin content was determined by the procedure proposed by Blumenkrantz and Asoboe-Hansen14 using anhydro-galacturonic acid (AGA) as a standard.
The calcium, iron, magnesium, potassium, copper and zinc contents in the original and clari®ed infusions were determined by atomic absorption spectrophoto-metry (Smith-Hieftje 8000, Thermo Jarrell Ash Co, Massachusetts, USA). Samples were diluted 10 times with deionised water to avoid creaming prior to analysis.
For the free amino acid content determination the tea infusion was diluted 20 times with distilled water and ®ltered through a 0.45mm ®lter. Five millilitres of ®ltrate was injected into a Sep-Pak C-18 cartridge (Millipore Co, Massachusetts, USA). The last 3ml of the ®ltrate was collected and analysed with an amino acid analyser (Model 415, LKB Biochem Co, UK) equipped with an Ultrapac resin column (lithium form). Citrate buffer (pH 2.20±6.45) was used for elution at a ¯ow rate of 50mlhÿ1. The absorbance of the infusion was measured at 570nm.
Statistical analysis
The analysis of the variance employed an SAS program for PCs (SAS Institute, Inc, Cary, NC, USA). Duncan's multiple-range test was also applied to determine whether there were signi®cant differences between individual treatments.
RESULTS AND DISCUSSION
Composition of Paochung tea cream
The primary components of Paochung tea cream are illustrated in Table 1. Caffeine, catechins and protein were the three major constituents. A widely supported hypothesis is that complexation is exhibited by both caffeine±polyphenols and polyphenols±protein. Com-plexation of polyphenols and proteins may involve two situations.15 At low protein concentrations the poly-phenol associates at more sites on the protein surface to produce a monolayer which is less hydrophilic than the protein itself, and then aggregation and precipita-tion ensue. When the protein concentraprecipita-tion is high, the precipitation is caused by the cross-linking of different protein molecules by the polyphenols. In the Paochung tea infusion the catechins content (1.46%) was 10 times the protein content (0.14%) (Table 1). This is thus closer to the former situation.
It is known that the primary intermolecular forces in the caffeine±polyphenols complexation are contribu-ted by hydrogen bonding, apolar hydrophobic inter-actions and coordination with a metal ion.16,17 Mejbaum-Katzenellenbogen et al18demonstrated that caffeine competes effectively with proteins for poly-phenolic substrates. Thus the caffeine±catechins com-plexation is also likely to be important in Paochung tea cream formation.
The most abundant group of substances in the Paochung tea infusion was catechins, which also accounted for approximately 30% of the total mass of the cream. The phenolic compounds in the original tea infusion and the decreamed tea infusion were analysed using reverse-phase HPLC (Fig 1). A com-parison of changes in the peak areas of the poly-phenolic constituents in the Paochung tea infusion before and after decreaming demonstrated that those compounds with hydrophobicity higher than (ÿ)-EGCG (eluted after (ÿ)-(ÿ)-EGCG) had a higher tendency to complex with other components and form cream (Table 2). The concentrations of galloyl esters in catechins, (ÿ)-EGCG and (ÿ)-ECG, were mark-edly diminished (p<0.05) after clari®cation, but the non-ester forms of catechins, (ÿ)-EGC, ()-C and (ÿ)-EC, were not signi®cantly reduced (p>0.05). The percentages of (ÿ)-EGCG and (ÿ)-ECG in the cream were 19% and 5% respectively (Table 1). In black tea the ester-form catechins can react with other catechins to produce thea¯avins or thearubigins during the fermentation process.5 In this study the tea¯avins would elute at about 59min in Fig 1, but in the full
chromatogram there was no evidence of their pre-sence. It is worth noting also that the baseline in Fig 1 shows little divergence from zero, indicating the absence of unresolved thearubigins.11Black tea cream contains approximately 17% caffeine, 66% thearubi-gins and 17% thea¯avins,1,6and EGCG and (ÿ)-ECG are not the primary components. The composi-tion of the semifermented tea cream, which does not contain thearubigins and thea¯avins, is obviously different from that of the black tea cream.
The galloyl group of catechins has been identi®ed to be an active site for binding protein through a hydrogen bond.20 Martin et al16 suggested that a hydrophobic interaction existed between the galloyl group of polyphenols and the six-member ring of caffeine. These interaction forces contributed by the galloyl group might account for the fact that these ester-form catechins had a much greater tendency to precipitate with protein and caffeine than did the non-ester-form catechins (Table 2). However, a role for the non-ester-form catechins cannot be totally ruled out in cream formation in the Paochung tea infusion. All phenolics are able to form hydrogen bonds with caffeine and are likely to be involved in complexation and precipitation.
Millin et al21 showed that 7% of endogenous polysaccharides, especially polygalacturonic acid, took part in cream formation in black tea by forming loosely bound complexes with polyphenolic and protein-aceous materials. Although there is little information concerning the af®nity between polyphenols and polysaccharides, the complexation of pectin and other constituents is less important in cream formation in Paochung tea, because its presence was relatively low (2%) in the cream.
The free amino acid contents in the original and clari®ed infusions were signi®cantly different (p<0.05), indicating that free amino acids also participated in cream formation in Paochung tea. The major free amino acids involved in the formation of the tea cream were theanine (39%), followed by glutamic acid (Glu) (10%) and aspartic acid (Asp) (5%) (Table 1). Other free amino acids to contribute to the formation of the tea cream included asparagine (Asn), tyrosine (Tyr), valine (Val), alanine (Ala), glutamine (Gln), proline (Pro) and glycine (Gly). However, Nagalakshimi and Seshadri4 did not ®nd a signi®cant amount of theanine in the acid hydrolysate of black tea cream and concluded that free amino acids were not involved in the formation of tea cream. Nevertheless, only approximately 8% of the total free amino acids in the infusion of Paochung tea were lost after the creaming process. Therefore we suspect that free amino acids may simply co-precipitate with insoluble complexes during cream formation.
The mineral contents of Paochung tea before and after clari®cation are shown in Table 1. Among the six minerals tested, calcium appeared to be the only mineral to be signi®cantly involved (p<0.05) in cream formation. About 66% of the calcium in the original
Table 1. Comparison of selected chemical constituents of original and
clarified Paochung tea infusionsa
Constituent Original infusion (g/100ml) Clari®ed infusion (g/100ml) Tea cream b (g/100ml) Total solids 3.81 3.24 0.57 Caffeine 0.38 0.27 0.11 (20%) Catechinsc 1.46 1.29 0.17 (30%) Catechinsd (ÿ)-EGC 0.027 0.025 NS (ÿ)-EC 0.068 0.067 NS (ÿ)-EGCG 0.303 0.197 0.106 (ÿ)-ECG 0.060 0.031 0.029 Protein 0.14 0.05 0.09 (16%) Pectin 0.10 0.09 0.01 (2%)
Total amino acids 0.076 0.070 0.006
Theanine 0.040 0.037 0.003 Asp 0.008 0.007 0.0003 Glu 0.007 0.006 0.001 Minerale Ca 7.54 2.56 4.98 Fe 5.20 5.14 NS Mg 4.00 4.00 NS K 6.69 6.56 NS Cu 0.10 NS 0.10 Zn 0.95 0.92 NS
aThe infusion was extracted with a 10% tea leaf/water ratio at 90°C for
20min. The clari®ed infusion was obtained after clari®cation by centrifuga-tion. For a detailed description see Materials and Methods. Values are the averages of four determinations (n =4). The composition of the tea cream was determined at signi®cance level p<0.05. Numbers in parentheses represent the percentages of the constituents in tea cream (g/dry basis).
bCalculated by subtracting the value of the clari®ed infusion from that of the
original infusion.
cData were determined using the HCl±vanillin method. dData were determined by quanti®cation of the HPLC method. eInppm.
infusion participated in cream formation. Smith1 found that both potassium and calcium existed in black tea cream. Guo and Cheng22 reported that calcium ions were more easily precipitated with polyphenols than were other cations in tea infusions. Jackson and Lee23 stated that the insolubility of the calcium±polyphenols complex in tea was dependent upon the degree of enzymic oxidation in the fermenta-tion process. Some research suggested that deminer-alisation could reduce cream formation in black tea.24 However, we found that the amounts of tea cream in Paochung tea infusions extracted with tap and
deionised water were not signi®cantly different (data not shown).
Effect of tea leaf/water ratio
The effect of the tea leaf/water ratio upon cream formation is shown in Table 3. An increase in the tea leaf/water ratio from 1.25 to 15% raised the total solids, caffeine, catechins, protein and pectin contents in the infusion. Although the amount of tea cream formed also increased with increasing leaf/water ratio, the percentage of each component in the tea cream was essentially the same. No cream formation was
Figure 1. High-performance liquid
chromatograms of polyphenols in Paochung tea infusion (a) before and (b) after clarification: peak 2, theobromine; peak 3, gallic acid; peak 6, caffeine; peak 7,
(ÿ)-epigallocatechin; peak 8, ()-catechin; peak 10, (ÿ)-epicatechin; peak 12, (ÿ)-epigallocatechin gallate; peak 18, rutin; peak 19, (ÿ)-epicatechin gallate; peak 24, quercetin. Other peaks are unidentified.
observed when the leaf/water ratio was 2% or lower, indicating that a minimum concentration was required to produce tea cream.
Effect of extraction temperature
The effect of the extraction temperature upon cream formation in Paochung tea was investigated by extracting the tea with a 12.5% leaf/water ratio at 30, 50, 70 and 90°C for 20min. As shown in Table 4, the total solids content and the amount of cream increased with increasing extraction temperature. The
composi-tion of the tea cream was also affected by the extraccomposi-tion temperature. The caffeine content remained constant. Rutter and Stainsby3 also reported that the caffeine content in black tea cream was maintained at 25±27% under various extraction conditions.
The content of catechins in the cream increased with increasing extraction temperature, while the contents of protein and pectin decreased. It was noted that there was a signi®cant amount of tea cream in Paochung tea extracted at temperatures as low as 30°C, but a black tea infusion extracted at 30°C did
Table 2. Comparison of peak areas of polyphenols
and caffeine in Paochung tea infusion before and after decreaming
Peak Compound Peak areaa Difference
Original Clari®ed Area %
1 421737a 302223b 119514 28.3 2 Theobromine 1024565a 1032461a 3 Gallic acid 268766a 273675a 4 209110a 191682a 5 150507a 137664a 6 Caffeine 11958209a 8090541b 3867668 32.3 7 (ÿ)-EGC 1316670a 1239087a 8 ()-C 327905a 312290a 9 609212a 541585a 10 (ÿ)-EC 1492750a 1402111a 11 427212a 400249a 12 (ÿ)-EGCG 10640832a 7832818b 2808013 26.4 13 868662a 636434b 232228 26.7 14 329649a 242614b 87035 26.4 15 383010a 268890b 114120 29.8 16 755976a 519101b 236876 31.3 17 958895a 887028b 71867 7.5 18 Rutin 327111a 256750b 70361 21.5 19 (ÿ)-ECG 2367695a 1824376b 543320 23.0 20 817665a 708773b 108892 13.3 21 101762a 60325b 41437 40.7 22 133167a 85078b 48089 36.1 23 214500a 182075b 32425 15.1 24 Quercetin 91517a 68587b 22930 25.1
aMeans (n =4) in the same row superscripted with different letters are signi®cantly different (p<0.05).
Table 3. Effect of tea leaf/water ratio (TW) on total solids, amount of cream formation and concentrations of caffeine, catechins, protein and pectin in Paochung tea
infusion and their percentages in tea creama
TW (%) Total solids(g/100ml) Tea cream (%)b Caffeine in tea infusion (g/100ml) Caffeine in tea cream (%)c Catechins in tea infusion (g/100ml) Catechins in tea cream (%)c Protein in tea infusion (g/100ml) Protein in tea cream (%)c Pectin in tea infusion (g/100ml) Pectin in tea cream (%)c 1.25 0.47h NS Ð Ð Ð Ð Ð Ð Ð Ð 2 0.74g NS Ð Ð Ð Ð Ð Ð Ð Ð 2.5 0.88f 6.8e ND ND ND ND ND ND ND ND 5 1.97e 9.1c 0.22e 25.3a 0.76e 31.1a 0.05e 19.4a 0.05e 2.6a 8 3.11d 14.5b 0.33d 18.7a 1.24d 25.8a 0.11d 17.0a 0.08d 2.0a 10 3.84c 14.1b 0.38c 15.6a 1.50c 30.6a 0.14c 16.0a 0.10c 1.3a 12.5 4.69b 17.3a 0.50b 18.2a 1.79b 30.6a 0.21b 16.9a 0.13b 1.5a 15 5.47a 16.5a 0.58a 21.8a 2.14a 34.9a 0.25a 17.8a 0.14a 2.6a
aThe tea infusions were extracted at 90°C for 20min. The mean values (n =4) in each column with the same superscript letters are not signi®cantly different at the
5% level using Duncan's multiple-range test. NS represents no cream formation. ND means the constituents were not determined.
bPercentage of total solids of tea infusion. cPercentage of tea cream by weight.
not form cream.3 The difference in creaming beha-viour between Paochung tea and black tea was probably due to most of the polyphenols in black tea existing in their oxidised forms, having a higher degree of polymerisation and being unable to be extracted ef®ciently at low temperatures, whereas most of the polyphenols in Paochung tea have their simple structures unchanged and thus can be extracted at 30°C.
Effect of extraction time
Table 5 shows the effect of the extraction time upon cream formation in Paochung tea at 90°C. The total solids in the infusion increased signi®cantly (p<0.05) with increasing extraction time from 5 to 20min, but a further increase in extraction time to 60min did not signi®cantly change the total solids content. An increase in the extraction time from 5 to 20min increased the catechins and protein contents in the infusion. Increasing the extraction time further did not increase their extractability either. Therefore extrac-tion at 90°C for 20min was considered to be long enough to extract most of the major components that contribute to cream formation. Since the amount of the tea cream was not affected by the length of extraction time, a reduction in the extraction time even down to 5min would not reduce cream formation in Paochung tea.
Effect of pH
The pH of the infusion affected the total solids content in the tea infusion, the amount of tea cream formation and the composition of the cream (Table 6). The total solids content had the lowest value at pH 5.5 (the pH value of the original tea infusion) owing to the low extraction rates of catechins and proteins at this pH. At low pHs (3.22 and 4.20) the catechins content in the tea infusion, as well as in the cream, was signi®cantly higher than that at higher pHs (5.50 and 6.86). The protein content too was signi®cantly higher, but only at the lowest pH, seeming to increase again at the highest pH. The polyphenols are believed to bind proteins primarily through the formation of multiple hydrogen bonds between their phenolic groups and the carbonyl functions of the peptide linkages.25The tea proteins are likely to have pI values close to pH 3.2±4.2, where protein±protein electrostatic repulsion would be minimised and result in maximum cream formation. However, the pH did not affect the caffeine content of the tea infusion or the percentage of caffeine in the cream. Spiro and Price26,27also reported that the pH did not affect the extractability of caffeine in black tea.
Although the amount of cream formed at pH 3.22 and 4.20 was higher than that at pH 5.50 and 6.86, the extent of the difference in cream formation due to the difference in the pH of the Paochung tea infusion was
Table 4. Effect of extraction temperature on total solids, amount of cream formation and concentrations of caffeine, catechins, protein and pectin in Paochung tea
infusion and their percentages in tea creama
Extraction
temp (°C) Total solids(g/100ml) Tea cream (%)b Caffeine in tea infusion (g/100ml) Caffeine in tea cream (%)c Catechins in tea infusion (g/100ml) Catechins in tea cream (%)c Protein in tea infusion (g/100ml) Protein in tea cream (%)c Pectin in tea infusion (g/100ml) Pectin in tea cream (%)c 30 3.37d 6.2d 0.36b 17.2a 1.30d 12.4c 0.10d 34.0a 0.04c 9.1a 50 4.04c 11.4c 0.39b 14.3a 1.48c 23.3b 0.15c 22.2b 0.06b 3.0b 70 4.24b 14.6b 0.46a 17.3a 1.63b 26.6b 0.18b 21.3b 0.06b 2.4b 90 4.73a 17.8a 0.48a 18.3a 1.79a 30.6a 0.21a 19.0b 0.13a 1.5b
aThe tea infusions were extracted with a 12.5% tea leaf/water ratio for 20min. The mean values (n =4) in each column with the same superscript letters are not
signi®cantly different at the 5% level using Duncan's multiple-range test.
bPercentage of total solids of tea infusion. cPercentage of tea cream by weight.
Table 5. Effect of extraction time on total solids, amount of cream formation and concentrations of caffeine, catechins, protein and pectin in Paochung tea infusion
and their percentages in tea creama
Extraction
time (min) Total solids(g/100ml) Tea cream (%)b Caffeine in tea infusion (g/100ml) Caffeine in tea cream (%)c Catechins in tea infusion (g/100ml) Catechins in tea cream (%)c Protein in tea infusion (g/100ml) Protein in tea cream (%)c Pectin in tea infusion (g/100ml) Pectin in tea cream (%)c 5 4.30c 16.5a 0.48a 23.1a 1.80c 14.0b 0.18b 17.3a 0.07c 1.9a 10 4.40b 16.8a 0.44a 20.0a 1.91b 25.3a 0.20b 20.2a 0.09c 1.6a 20 4.82a 17.4a 0.48a 18.3a 2.08a 31.2a 0.24a 18.8a 0.14b 1.5a 60 4.88a 16.4a 0.47a 16.0a 2.07a 34.2a 0.25a 19.7a 0.20a 1.8a
aThe tea infusions were extracted with a 12.5% tea leaf/water ratio at 90°C. The mean values (n =4) in each column with the same superscript letters are not
signi®cantly different at the 5% level using Duncan's multiple-range test.
bPercentage of total solids of tea infusion. cPercentage of tea cream by weight.
much less than that in black tea. The maximum amount of tea cream in black tea was formed at pH 3± 4,1while no cream formation was observed at pH 6.7. It was also noted that raising the pH of the infusion signi®cantly altered the colour and ¯avour of Paochung tea (data not shown). Suematsu et al28 stated that catechins were more unstable and became subject to greater oxidation in the high pH range during processing. Therefore adjusting the pH of the infusion may not be a feasible process for inhibiting the formation of cream in Paochung tea.
CONCLUSION
The formation of tea cream is still a signi®cant problem in the application of Paochung tea as a cool drink, primarily because it causes an unattractive appearance and processing dif®culties. The amount of cream in the tea and its composition were in¯uenced by extraction temperature and pH. The tea leaf/water ratio was found to affect the amount of cream in the infusion but not the cream composition. The poly-phenolic compounds are probably the most important components in the formation of tea cream. They form a complex with caffeine and protein to create cream. Reducing the extraction temperature and adjusting the pH reduces cream formation, but inhibition of the development of Paochung tea cream without impair-ing the important sensory qualities or food safety remains a challenge in the processing of this tea.
ACKNOWLEDGEMENTS
This research was supported by the Council of Agriculture, ROC under project 83-2.8-08(9).
REFERENCES
1 Smith RF, Studies on the formation and composition of cream in tea infusion. J Sci Food Agric 19:530±534 (1968).
2 Collier PD, Mallows R and Thomas PE, Interactions between thea¯avins, ¯avanols and caffeine. Phytochemistry 11:867 (1972).
3 Rutter P and Stainsby G, The solubility of tea cream. J Sci Food Agric 26:455±463 (1975).
4 Nagalakshimi S and Seshadri R, A study of nitrogenous
constituents of tea cream. J Food Sci Technol 20:243±245 (1983).
5 Powell C, Clifford MN, Opie SC, Ford MA, Robertson A and Gibson CL, Tea cream formation: the contribution of black tea phenolic pigments determined by HPLC. J Sci Food Agric 63:77±86 (1992).
6 Roberts EAH, The phenolic substances of manufactured tea. X. The creaming down of tea liquors. J Sci Food Agric 14:700±705 (1963).
7 Hazarika M, Chakravarty SH and Mahanta PK, Studies on thearubigins pigments in black tea manufacturing systems. J Sci Food Agric 35:1208±1218 (1984).
8 Seshadri R and Nagalakshimi D, New hydrophobic lipid inter-action in tea cream. J Sci Food Agric 45:79±86 (1988). 9 AOAC, Of®cial Methods of Analysis, 14th edn, Association of
Of®cial Analytical Chemists, Washington, DC (1984). 10 Tasi YZ and Juan IM, A simple determination method of
caffeine in tea. Taiwan Tea Res Bull 6:1±7 (1987).
11 Sarker SK and Howarth RE, Speci®city of the vanillin test for ¯avanols. J Agric Food Chem 24:317±320 (1976).
12 Terada S, Maeda Y, Masui T, Suzuki Y and Ina K, Comparison of caffeine and catechin components in infusion of various tea (Green, Oolong and Black tea) and tea drinks. Nippon Shikuhin Kogyo Gakkaishi 34:20±27 (1987).
13 Lowry OH, Rosebrough NJ, Farr AL and Randall RJ, Protein measurement with Folin phenol reagent. J Biol Chem 193:265± 275 (1951).
14 Blumenkrantz N and Asoboe-Hansen G, New method for quan-titative determination of uronic acids. Anal Biochem 54:484± 490 (1973).
15 McManus JP, Davis K, Haslam E and Lilley TH, The asso-ciation of proteins with phenols. J Chem Soc, Chem Commun 309±311 (1981).
16 Martin R, Lilley TH, Bailey NA, Falshaw CP, Haslam E, Magnolato D and Begley MJ, Polyphenols±caffeine complexa-tion. J Chem Soc, Chem Commun 105±106 (1986).
17 Gaffney SH, Martin R, Lilley TH, Haslam E and Magnolato D, The association of polyphenols with caffeine and a- and b-cyclodextrin in aqueous media. J Chem Soc, Chem Commun 107±109 (1986).
18 Mejbaum-Katzenellenbogen W, Dobryszychka WM, Jaworska K and Wiecka BM, Regeneration of protein from insoluble protein±tannin compounds. Nature (London) 184:1799±1800 (1959).
19 Bailey RG, Nursten HE and McDowell I, Comparative study of the reversed-phase high-performance liquid chromatography of black tea liquors with special reference to the thearubigins. J Chromatogr 542:115±128 (1991).
20 Sekiya J, Kadahiko K, Monma T and Hatanaka A, Interaction of tea catechins with proteins: formation of protein precipitate. Agric Biol Chem 48:1963±1967 (1984).
21 Millin DJ, Swain D and Dix PL, Separation and classi®cation of
Table 6. Effects of pH of Paochung tea infusion on total solids, tea cream formation and concentrations of caffeine, catechins, protein and pectin and their
percentages in tea creama
pH Total solids(g/100ml) Tea cream (%)b Caffeine in tea infusion (g/100ml) Caffeine in tea cream (%)c Catechins in tea infusion (g/100ml) Catechins in tea cream (%)c Protein in tea infusion (g/100ml) Protein in tea cream (%)c Pectin in tea infusion (g/100ml) Pectin in tea cream (%)c 3.22 4.20a 16.1a 0.39a 22.0a 1.61a 36.8a 0.34a 33.8a ND ND 4.20 3.86c 15.9a 0.40a 24.5a 1.69a 37.7a 0.22b 21.3bc ND ND 5.50 3.79c 14.3b 0.40a 18.5a 1.51b 30.0b 0.14c 18.5c ND ND 6.86 3.98b 12.9c 0.41a 19.6a 1.42c 30.0b 0.20b 27.5b ND ND
aThe tea infusions were extracted with a 10% tea leaf/water ratio at 90°C for 20min. Pectin was not determined (ND) in this experiment. The mean values (n =4) in
each column with the same superscript letters are not signi®cantly different at the 5% level using Duncan's multiple-range test.
bPercentage of total solids of tea infusion. cPercentage of tea cream by weight.
aqueous infused brown pigments of black tea. J Sci Food Agric 20:296±302 (1969).
22 Guo BY and Cheng QK, Reaction of tea infusion components with metal ions and its application to preparation of pure polyphenols. Proc Int Symp on Tea Science, pp 86±88 (1991).
23 Jackson LS and Lee K, Chemical forms of iron, calcium, magnesium and zinc in black, Oolong, green and instant black tea. J Food Sci 50:181±184 (1988).
24 Pintauro ND, Tea and Soluble Tea Products Manufacture, Noyes Data Co, Park Ridge, NJ, pp 82±125 (1977).
25 Van Sumere CF, Albrecht J, Dedonder A, De Pooter H and Pe I. Plant Proteins and Phenolics, In The Chemistry and Biochem-istry of Plant Proteins, Ch 8, Ed by Harborne JB and Van Sumere CF, Academic Press, New York, pp 211±264 (1975).
26 Spiro M and Price W, Kinetics and equilibria of tea infusionÐ Part 6: The effects of salts and pH on the concentrations and partition constants of thea¯avins and caffeine in Kapchorua Pekoe Fannings. Food Chem 24:51±61 (1987).
27 Spiro M and Price W, Kinetics and equilibria of tea infusionÐ Part 7: The effects of salts and pH on the rate of extraction of caffeine in Kapchorua Pekoe Fannings. Food Chem 25:49±59 (1987).
28 Suematsu S, Hisanobu Y, Saigo H, Musuda R, Hara K and Komatsu Y, Effects of pH on stability of constituents in canned tea drinks. Nippon Shakuhin Kogyo Gakkaishi 39:178±182 (1992).
