green tea infusion
ingredient / ingredient-green-tea
緑茶（浸出液）
Green tea infusion is an infusion prepared from leaves of Camellia sinensis.
On this page (8 sections)
Reading the bitterness
Green tea infusion refers to an infusion prepared by steeping the leaves of tea plant (Camellia sinensis). A 2022 review positioned bitterness and astringency as the principal quality attributes of green tea, treated as separate sensory dimensions even when the source compounds overlap.
In a 2018 quantitative study, the bitterness of brewed green tea infusion was highly correlated with the concentration of two galloylated catechins — epigallocatechin gallate (EGCG) and epicatechin gallate (ECG) — (R2=0.7769), while astringency was correlated with the concentration of ECG and selected flavonol glycosides (myricetin glycosides and quercetin glycosides) (R2=0.7878, p<0.01). Bitterness correlation involves EGCG and ECG, astringency correlation involves ECG and flavonol glycosides; the only shared compound between the two is ECG — the overlap is at a single point, and the analysis of primary data shows that the compound groups are distinct.
This concentration correlation is broadly consistent with results from cultured-cell experiments on the human bitter taste receptor hTAS2R39. Response to ECG is reported to be strongest, followed by EGCg, with EC and EGC responses much weaker; the authors state that this rank order resembles the trend in human sensory evaluation of bitterness intensity. However, both the correlation and the receptor-response order are specific to the test materials and conditions of that study, and confirmation remains at the abstract level.
Hojicha (roasted tea) is positioned as a derivative beverage made by roasting green tea such as bancha, and gains a roasted fragrant aroma distinct from ordinary green tea through roasting. From the perspective of catechins and flavonol glycosides responsible for bitterness and astringency, green tea can be traced down to individual compounds such as epicatechin gallate and flavonol glycosides.
Sources for this overview
Checked for this overview: Confirmed the review statement positioning bitterness as a principal quality attribute of green tea. / Confirmed the statement treating astringency as a separate principal quality attribute distinct from bitterness.
Quantitative analyses of the bitterness and astringency of catechins from green tea
Checked for this overview: Confirmed the correlation R2 value between bitterness and EGCG/ECG concentrations. / Confirmed the correlation R2 value between astringency and ECG/glycoside concentrations. / Astringency correlation compounds, R-squared value, and significance level derive from this source. / Bitterness correlation compounds and R-squared value derive from this source.
Checked for this overview: Confirmed the statement on hTAS2R39 response order and similarity to sensory bitterness intensity trend.
Checked for this overview: Confirmed the definition statement that hojicha is roasted green tea.
Follow the connections
green tea infusion →— sensory attribute→ bitterness →can be measured separately …→ astringency
green tea infusion →— sensory attribute→ astringency →is not synonymous with→ bitterness
Hojicha →is derived from→ green tea infusion
Structured claims — open a section to read each claim's confidence, evidence and limits
Composition and content (1)
Note: 「ほうじ茶は番茶などの緑茶を焙煎したもの」 (in translation: "hojicha (roasted green tea) is roasted green tea such as bancha") — a general classification statement.
Show data
ingredient-hojicha —is_derived_from→ ingredient-green-teaSensory (18)
Scope: Brewed green tea
Note: Intensity depends on the tea leaf and brewing method. / 2026-08-31: Calibrated down from established to supported, because the supporting evidence does not derive from a full-text check. May be re-upgraded after a full-text check.
Show data
ingredient-green-tea —has_sensory_attribute→ sensation-bitterScope: Brewed green tea
Note: Kept separate from bitterness even where the compounds driving the two overlap. / 2026-08-31: Calibrated down from established to supported, because the supporting evidence does not derive from a full-text check. May be re-upgraded after a full-text check.
Show data
ingredient-green-tea —has_sensory_attribute→ sensation-astringentScope: Green tea infusions and the concentration-taste model in the cited study
Note: The correlation and fitted curve are specific to this study and are not a universal conversion formula.
Show data
ingredient-green-tea —has_bitterness_correlated_with→ "The concentration of EGCG and ECG"Scope: Green tea infusion in the cited study
Note: Interactions within the mixture and preparation conditions have an effect.
Show data
ingredient-green-tea —has_astringency_correlated_with→ "ECG and selected flavonol glycoside concentrations"Scope: Layer: food. Green tea extracted in 3 successive infusions (GT1-GT3), five-taste sensory evaluation by electronic tongue, Pearson correlation analysis, and PLSR.
Show data
compound-l-theanine —has_perceived_bitterness_associated_with→ ingredient-green-teaScope: 1 commercial cold-brew sencha sample and 1 commercial Uji sencha sample (a different product). 15 faculty and staff members of Notre Dame Women's University rated each sample on its own for bitterness, astringency, sweetness, umami, overall taste, and preference (strong +2 to weak −2).
Note: The cold-brew sencha and the sencha are different commercial products, so this is not a comparison of the same leaves at different water temperatures. The authors infer that the low scores come from the small amount of extracted components. The cold-brew sencha was also somewhat weak in sweetness, umami, and overall taste. The water temperature is given as 2°C, as in the original (a low value for cold brewing).
Show data
ingredient-green-tea —has_sensory_response→ "In a commercial cold-brew sencha (3 g steeped in 80 mL of 2°C water for 0.5–4 minutes, repeated up to 3 infusions), bitterness and astringency were rated weak (5-point scale from −2 to +2, averaged over all steeping times; bitterness for infusions 1–3: −1.73, −1.73, −1.87; astringency: −1.67, −1.73, −1.73). In the same report, a commercial sencha steeped at 3 g in 80 mL of 80°C water scored 0.53, 0.20, −0.87 for bitterness and 0.66, 0.54, −0.53 for astringency."Scope: 1 commercial Uji sencha sample. 15 faculty and staff members of Notre Dame Women's University rated bitterness, astringency, and other items individually on a 5-point scale from −2 to +2. For both bitterness and astringency, the difference by steeping time (0.5–4 min) was reported to be small.
Note: The bitterness and astringency scores for each steeping time are not shown in the text (only the representative scores at 2 minutes), and the small difference is the authors' statement. Tannin was measured by chemical analysis (iron tartrate method) and is not equated with sensory evaluation.
Show data
ingredient-green-tea —has_perceived_bitterness_unchanged_by→ "When 3 g of sencha was steeped in 80 mL of 80°C water for 0.5, 1, 2, 3, or 4 minutes, scores for bitterness, astringency, and the other items differed little with steeping time. In contrast, the tannin in the infusion increased with longer steeping; in the 1st infusion it was 6.69 (0.5 min), 9.46 (1 min), 18.38 (2 min), 22.23 (3 min), and 25.47 (4 min) mg/80 mL."Scope: 1 commercial Uji sencha sample (3 g, 80 mL of 80°C water, steeped for 0.5–4 min, 3 infusions from the same leaves). 15 faculty and staff members rated it individually on a 5-point scale from −2 to +2.
Note: The difference between the 1st and 2nd infusions was not tested for significance. The infusions were evaluated in the order 2nd, 1st, 3rd.
Show data
ingredient-green-tea —has_sensory_response→ "In commercial sencha steeped in 80°C water, bitterness and astringency were felt at similar levels in the 1st and 2nd infusions and became weaker in the 3rd (averages over all steeping times; bitterness for infusions 1–3: 0.53, 0.20, −0.87; astringency: 0.66, 0.54, −0.53). While gyokuro kept bitterness and astringency at a moderate level from the 1st to the 3rd infusion, all items in sencha weakened with repeated infusions."Scope: Kabusecha from the 2011 harvest (Yabukita). Infusions made by placing 25 g of tea and 1 L of water in a glass pot and letting it stand were rated for astringency relative to the 50°C infusion by 22 people working in tea research, extension, or business.
Note: The ratings are approximate values read from a figure, and what was rated was astringency and umami, not bitterness. The analysis used only 5 samples (4 infusion conditions of kabusecha B and the 50°C infusion of kabusecha C). The tea type is kabusecha, not sencha. No tests of differences between individual conditions are given.
Show data
ingredient-green-tea —has_sensory_response→ "Kabusecha was steeped in 90°C, 50°C, and 20°C water for 5 minutes and in 5°C water for 30 minutes, and 22 panelists rated the astringency of each infusion on a 5-point scale relative to the 50°C infusion (strong +2 to weak −2). The 90°C infusion was stronger than the reference (about +1.6, read from the figure), while the 20°C (about −1.1) and 5°C, 30-minute (about −1.0) infusions were weaker. The authors state that the sample infused at 90°C was extremely astringent. Astringency ratings correlated significantly and positively with the logarithm of total catechin concentration (r²=0.79, p<0.05) and the logarithm of galloylated catechin concentration (r²=0.94, p<0.01); galloylated catechins in the 90°C infusion were about 2.2 times those in the 50°C infusion (0.73 mM vs 0.33 mM)."Scope: 1 sencha sample. Infusions were made by pouring 200 mL of water onto 2 g of tea, letting it stand, and cooling in an ice bath. Estimates were obtained with a taste sensor (AE1 sensor for astringency, referenced to 0.65 mM epigallocatechin gallate).
Note: These are taste-sensor estimates of astringency and are not equated with human sensory evaluation (in Experiment 2 of the same report, a high positive correlation was shown between the astringency estimate and human astringency ratings for infusions such as kabusecha). It is an estimate of astringency, not bitterness.
Show data
ingredient-green-tea —has_sensory_response→ "When sencha (Yabukita, 2010 harvest) was steeped in 90°C water for 0.5–10 minutes, or in 30–100°C water for 5 minutes, total catechin concentration rose from 0.35 mM at 0.5 min to 1.02 mM at 10 min, and from 0.32 mM at 30°C to 1.05 mM at 100°C; galloylated catechins also rose from 0.25 to 0.79 mM and from 0.19 to 0.82 mM. The taste sensor's astringency estimate correlated highly with the logarithm of total catechin and galloylated catechin concentrations under all conditions (r²=0.98–1.00). The taste sensor's astringency estimate was higher with hotter water and longer steeping."Scope: Deep-steamed sencha from Kawane, Shizuoka (Yabukita first flush, 2011 harvest). Taste sensor (SA402B) measurements and HPLC catechin quantification.
Note: These are taste-sensor measurements and are not equated with human sensory evaluation. The result comes from infusing the same leaves in turn at 5°C, 40°C, and 100°C, using leaves already well swollen by the cold infusion, and differs from results with separate fresh leaves at each temperature (with separate leaves, astringency intensity rose slightly from 5°C to 40°C). It is astringency intensity, not bitterness.
Show data
ingredient-green-tea —has_sensory_response→ "Under the condition of adding 200 mL of water to 3 g of deep-steamed sencha, the same leaves were infused in turn with 5°C water (10 min), 40°C water (1 min), and 100°C water (1 min). The taste sensor's astringency intensity rose greatly from 5°C to 40°C, and umami intensity decreased. Catechin content of the infusions was 19.0 (5°C), 26.3 (40°C), and 60.7 (100°C) mg/100 mL, and the authors state that astringency intensity tended to increase with infusion temperature."Scope: Deep-steamed sencha from Kawane, Shizuoka (Yabukita first flush, 2011 harvest). Taste sensor (SA402B) measurements, HPLC catechin quantification, and a paired comparison (2-point) test by a trained panel of 5 (reference: infusion of 1 g of leaves).
Note: The authors point out that the taste sensor may saturate as leaf amount increases. The sensory evaluation was small, with 5 panelists; the number of people and the result are as stated verbatim in the source. It is astringency intensity, not bitterness.
Show data
ingredient-green-tea —has_sensory_response→ "When 1–10 g of deep-steamed sencha was steeped for 1 minute in 200 mL of hot water, the taste sensor's astringency intensity rose almost linearly up to 2 g of leaves and increased only gradually beyond that. Catechin content rose to 19.7, 60.0, 91.4, and 169.0 mg/100 mL for 1 g, 2 g, 3 g, and 5 g of leaves, but did not correspond to the change in astringency intensity. In a paired comparison (2-point) test with a trained panel of 5, it was difficult to rate astringency as dependent on leaf concentration at 3 g of leaves or more."Scope: 7 commercial cold-brew senchas. 10 staff members of the Vegetable and Tea Research Station of the Ministry of Agriculture, Forestry and Fisheries (Makurazaki) and 10 junior college women students evaluated color, aroma, and taste by the ranking method (Kramer's rank-sum test).
Note: What was evaluated was preference rank; the strength of bitterness and astringency was not measured directly. The link between the lower rating at 10 hours and bitter-astringent taste is the authors' interpretation based on increased tannin extraction. The students tended to prefer shorter steeping times than the general panel, which the authors infer is because they prefer a lighter bitter-astringent taste. "Bitter-astringent taste" (kujūmi) is an expression that combines bitterness and astringency into one.
Show data
ingredient-green-tea —has_bitter_astringency_affected_by→ "7 commercial cold-brew senchas were infused at 100 mL of distilled water (7–8°C) per 1 g of tea for 5 minutes, 30 minutes, 2 hours, and 10 hours, and the preference for color, aroma, and taste was evaluated by the ranking method; except for a few samples, the 30-minute to 2-hour infusions were preferred. The authors consider that the 5-minute infusion was rated low because few components had been extracted, and the 10-hour infusion because tannin (an astringent component) increased and the bitter-astringent taste became stronger. Tannin was slow to extract: the share extracted relative to the amount in the tea was about 15–30% at 30 minutes, 19–40% at 2 hours, and 30–60% at 10 hours, whereas free amino acids were already about 40–70% extracted at 30 minutes."Scope: 2 grades of Longjing tea, high grade and standard grade. Steeped for 4 minutes in 70, 80, 90, and 100°C water. Infusions held at 45°C were evaluated by quantitative descriptive analysis (0–10 scale) by a trained panel of 9 (trained for 3 months or more).
Note: Bitterness and astringency were rated as separate attributes. The amounts of tea leaves and water are not stated in the text. The figures are the values the text gives for the change from 70°C to 100°C in the standard grade (SL); for the high grade, only the rates of increase are given. This is Chinese Longjing tea, not Japanese sencha.
Show data
ingredient-green-tea —has_bitter_astringency_affected_by→ "Chinese green tea (Longjing tea) was steeped for 4 minutes in 70, 80, 90, and 100°C water and rated on a 0–10 scale by a trained panel of 9. For standard-grade Longjing tea, bitterness rose significantly from 3.13 to 5.17 and astringency from 2.07 to 4.13 as water temperature increased (p<0.05). From 70°C to 100°C, bitterness and astringency scores increased by 33.25% and 36.72% in the high-grade tea and by 64.92% and 99.98% in the standard tea. The authors conclude that low water temperatures of 70–80°C weakened bitterness and astringency and strengthened umami."Scope: 1 sample of Shucheng Xiaolanhua (a green tea from Anhui, China). 10 people (aged 22–30) rated bitterness, astringency, and other attributes on a 0–5 scale. The study's main subject is the analysis of aroma compounds, and its sensory evaluation was only a comparison of the 2 methods at 85°C.
Note: The bitterness scores and test results are in a figure; the text only says "significantly stronger bitterness" and gives no numbers. The result of the astringency comparison is not given in the text. The difference in polyphenol content, proposed as the reason for the bitterness difference, is an analytical value and has not been directly tied to sensory results.
Show data
ingredient-green-tea —has_bitter_astringency_affected_by→ "When 5 g of Chinese green tea (Shucheng Xiaolanhua) was steeped in 250 mL of 85°C water for 4 minutes, the infusion made by adding the water first and the leaves afterward (top-down method) showed significantly stronger bitterness than the one made by putting the leaves in first and pouring water onto them (bottom-up method). The top-down infusion also had a higher total polyphenol content (101.50 mg/g vs 90.87 mg/g)."Scope: 1 special-grade Yuhua green tea. Steeped for 6 hours in a 4°C refrigerator and served at 4°C. Descriptive evaluation by 5 trained panelists (aged 24–42) and taste-sensor (SA402B) measurements.
Note: The human evaluation is vocabulary agreed on by 5 people, and no scores or test values are given. The statements that bitterness was "highest" or "lowest" are taste-sensor measurements and are not equated with human sensory evaluation. This is the effect of water quality in cold brewing, not a comparison of different water temperatures or steeping times. Which group had the lowest bitterness is described inconsistently in the paper (there are sentences that assign it both to the RO-treated water group and to the mineral water group), so this claim does not state it.
Show data
ingredient-green-tea —has_bitter_astringency_affected_by→ "When green tea (Yuhua green tea) was cold-brewed at a tea:water ratio of 1:50 at 4°C for 6 hours, the taste was described differently depending on the type of water used. The infusion made with pure water was described by 5 trained panelists as "brightly bitter," and its taste-sensor bitterness value was also the highest. The infusion made with reverse-osmosis drinking water (moderate calcium and magnesium, neutral) was described as having a "fresh and brisk sweet aftertaste," and its taste-sensor sweetness value was the highest. Infusions made with pure and distilled water contained more EGCG and caffeine (EGCG: 4.50% pure water, 4.29% distilled water, 4.10% RO-treated water, 3.32% mineral water), and the authors discuss that bitterness and astringency were stronger."Scope: Cold infusion of 1 Bi Luo Chun (from New Taipei City, Taiwan) at 4–5°C for 6 hours. 63 university students in Taiwan (25 men, 38 women) rated it by TCATA (9 terms: sweet, sour, bitter, astringent, roasted aroma, raw tea leaf aroma, floral aroma, sweet aftertaste, pungency). The reference was the average selection frequency of the other 6 teas. The infusion was taken out of the refrigerator and served after returning to room temperature.
Note: The statement is that even with cold infusion the green tea was placed among the strongly bitter teas; it includes no comparison with hot infusion. The cold-brew condition (4–5°C, 6 hours) was the same for all samples and is longer than the recommended time for green tea (4 hours). Bitterness and astringency were selected as separate terms. These are selection frequencies, not numerical intensity ratings. This is a relative comparison of 1 green tea and 6 more highly fermented teas, not a comparison varying infusion conditions.
Show data
ingredient-green-tea —has_sensory_response→ "Cold-infused green tea from Taiwan (unfermented Bi Luo Chun), made with 50 times its volume of water at 4–5°C for 6 hours, was evaluated for 30 seconds by 63 consumers using temporal check-all-that-apply (TCATA). Bitterness was selected significantly more often than the average of the other 6 teas (5 oolongs and a black tea) from 5 seconds after the first sip and continued to the end of the evaluation. Among the 7 teas, this tea was characterized as strong in bitterness, astringency, and pungency and weak in sweet aftertaste and floral aroma."How processing and cooking change it (1)
Scope: Experimentally prepared high-grade and ordinary-grade senchas (the ordinary grade being a re-fired tea blending first flush with second and third flushes). Chemical analysis of infusions made with 40, 60, 80, and 95°C water for 2, 4, 6, 8, and 10 minutes.
Note: No sensory evaluation was done. The mention of bitter-astringent taste is the authors' discussion and recommendation based on the extracted amounts of tannin and epigallocatechin gallate, not a result measuring the strength of bitterness or astringency. Extracted epigallocatechin gallate (high-grade tea, % of the tea) was 0.27 (40°C) and 1.69 (80°C) at 2 minutes, and 1.08 (40°C) and 3.26 (80°C) at 10 minutes.
Show data
ingredient-green-tea —has_bitter_astringency_affected_by→ "With 6 g of high-grade sencha and 180 mL of water, tannin extraction (% of the tea) was, at 2 minutes, 1.63 (40°C), 2.96 (60°C), 6.66 (80°C), and 8.39 (95°C), and at 10 minutes, 4.52 (40°C), 6.73 (60°C), 10.77 (80°C), and 12.82 (95°C); it was higher with hotter water, and at 95°C, 82–86% of all tannin was extracted in 4 minutes and 95–98% in 10 minutes in both the high-grade and ordinary-grade senchas. In the high-grade tea, epigallocatechin gallate, which has a strong bitter-astringent taste, was extracted more readily at high temperature, according to the authors, who also state that for suppressing the bitter-astringent taste, as with high-grade tea, low-temperature infusion at about 60°C seems suitable."