CoffeariumBeta
  1. Plant
  2. Farm
  3. Process
  4. Trade
  5. Roast
  6. Brew
  7. Cup
Science · Health

Caffeine

The bitter purine alkaloid that makes coffee stimulating – where it sits in the plant, how much reaches the cup and why that varies so much, what it does in the body, how it is broken down, and the official safe-intake advice.

Across the chain: Science

Source-checkedby Coffearium verification pass· Last source-checked 10 Oct 2026· Prepared with AI assistance· 23 sources (21 tier A)· 5 contested· Updated 10 Oct 2026
Reading depth
Contents
  • Caffeine is a methylxanthine, a purine alkaloid with three methyl groups (1,3,7-trimethylpurine-2,6-dione).1 It is toxic to insects and is thought to protect the coffee plant.3
  • Robusta beans hold about twice as much caffeine as arabica beans; in brews, most studies find 1.4–1.8 times as much. Roasting hardly changes it.5,4
  • What ends up in a cup depends largely on the coffee-to-water ratio and the serving size. Espressos bought in 20 cafés ranged from 51 to 322 mg each.4,7
  • EFSA considers up to 400 mg a day (and single doses up to 200 mg) of no safety concern for healthy non-pregnant adults, and up to 200 mg a day during pregnancy.15
  • Caffeine blocks adenosine receptors. The liver breaks almost all of it down, mainly with the enzyme CYP1A2 and mainly to paraxanthine; the speed varies widely between people.18,20,22

What caffeine is#

Caffeine is a methylxanthine. Its systematic name, 1,3,7-trimethylpurine-2,6-dione, describes it: a purine ring system carrying three methyl groups.1 The molecule is rich in nitrogen. Pure caffeine is a white, odourless, bitter crystalline powder, and at high temperature it sublimes – it passes from solid to vapour without melting.1 Its close relatives paraxanthine, theobromine and theophylline are dimethylxanthines: each lacks one of the three methyl groups.20

In the green bean, caffeine is held in a structure with chlorogenic acids. Water frees it, which is why every industrial decaffeination process starts by wetting the beans.1

Why the plant makes it#

Caffeine can be toxic to insects and is thought to protect the plant against herbivores.3 Yet the most damaging insect pest of coffee, the coffee berry borer, lives inside the seed. It copes with the help of gut bacteria: several Pseudomonas species in its gut can grow on caffeine as their only source of carbon and nitrogen, and when the gut bacteria are removed, the beetle no longer breaks caffeine down.3

Caffeine also occurs in the nectar of Coffea flowers. In an experiment, honeybees rewarded with caffeine were three times as likely to remember a learned floral scent as bees rewarded with sugar alone.2 Caffeine concentrations in nectar did not exceed the bees’ bitter-taste threshold. The authors read this as a way for the plant to keep pollinators coming back.2

How much is in the bean#

Coffea canephora (robusta) holds about twice as much caffeine as Coffea arabica.5 A 2021 review gives 0.9–1.5% of dry weight for green arabica and 1.2–2.4% for green robusta.4 In brews, the same review found that robusta mostly gave 1.4–1.8 times as much caffeine as arabica.4

Roasting barely touches caffeine. In one study of Brazilian arabica, caffeine stayed at 12–13 mg per gram from green beans to a dark roast, while caffeoylquinic acids (the main chlorogenic acids) fell from 41 to 5 mg/g.5 Because roasting removes water and other matter while caffeine stays, the amount per gram of roasted coffee stays the same or rises slightly.5,1 See Roasting for what does change.

How much ends up in the cup#

The amount of caffeine in a cup is set by four things: how much ground coffee is used, the share of robusta, how much water is used, and how much of the caffeine is extracted.1,4 A 2021 review of brewing studies concluded that the coffee-to-water ratio probably has the greatest influence; brew time, it found, was not a decisive factor.4

Caffeine dissolves easily, so most of it leaves the grounds early in any brew.9,8 In a laboratory test of double espressos made from 18.1–20.4 g of arabica, about 67% of the caffeine in the grounds reached the cup, giving 152–174 mg per serving whatever the roast.5 Spent grounds from a household filter brewer still held 5.6% of the original caffeine; spent grounds from professional high-pressure machines in restaurants held 21.5–30.8%.10

Real cups vary far more than laboratory brews. Twenty espressos bought in Glasgow held 51–322 mg each, in cups of 23–70 mL.7 In a later survey, mean espresso caffeine per outlet ranged from 72 to 212 mg in Scotland, 73–135 mg in Italy and 97–127 mg in Spain.5 Cafés use different doses, from 5 to 18 g, and different cup sizes.5 Espresso is the most concentrated brew per litre, but per serving the volume of the cup also counts.4

Tool

Caffeine in your cup

Estimate the caffeine in an espresso, a filter cup or your own brew, compared with official intake advice.

Open as a standalone tool →

Estimated caffeine

Estimate only; real cups vary widely. Espressos bought in 20 Glasgow cafés held 51–322 mg each.

Reference lines: EFSA advice for healthy adults (single dose 200 mg; 400 mg a day) and in pregnancy or breastfeeding (200 mg a day). General information, not medical advice.

Midpoint of the 18.1–20.4 g doses in one lab study

From measured espresso shots

Blend100% arabica
100% robusta100% arabica

Default: 13 mg/g measured in roasted arabica. Green beans: 9–15 mg/g dry weight

Default: 22 mg/g, within the 1.2–2.4% reported for green robusta. Green beans: 12–24 mg/g dry weight

Keeps your inputs, result and chart. Nothing is stored online.
Estimate the caffeine in your own cup from dose, species blend, caffeine content and the share extracted. Defaults from Ludwig et al.5, Olechno et al.4, Pietsch1, Mestdagh et al.9, Vandeponseele et al.10, Gloess et al.11 and the KaffeeV14; reference lines from EFSA.15 Estimate only; real cups vary widely.

The calculator multiplies dose × caffeine content × share extracted. Its defaults: 13 mg/g for roasted arabica5 and 22 mg/g for robusta, a value within the 1.2–2.4% reported for green robusta4, which roasting leaves roughly unchanged1,5; 67% extracted for a double espresso of about 19 g5; 90% for filter coffee, slightly below the ≈94% implied by household filter grounds that kept 5.6% of the caffeine10 (a textbook chapter estimates that about 90% is extracted within the first brewing minute9), with 8.3 g for a 150 mL cup at the 100 g per 1,800 mL of brew of one study11; and 84% for a moka pot, whose grounds kept 16.4% of the caffeine10, with 7.5 g of coffee per 110 mL of water as in the same comparison of brew methods.11 The decaf option uses the German legal ceiling of 1 mg/g.14

Decaffeinated coffee#

Decaf is not caffeine-free. The US Food and Drug Administration says decaf coffee typically has 2–15 mg in an 8-fluid-ounce cup.12 In the EU, coffee extracts (instant coffee) may be labelled ‘decaffeinated’ only if anhydrous caffeine does not exceed 0.3% of the coffee-based dry matter.13 For beans, national law applies1; Germany, for example, allows ‘entkoffeiniert’ on green and roasted coffee with at most 1 g of caffeine per kilogram of coffee dry matter (0.1%).14 How caffeine is removed is covered in Decaffeination.

Official intake advice#

The European Food Safety Authority (EFSA) assessed the safety of caffeine from all sources in 2015. Its conclusions:

  • Healthy adults, single dose: up to 200 mg (about 3 mg per kilogram of body weight) raises no safety concerns, also when taken less than two hours before intense exercise under normal conditions.15,16
  • Healthy adults, per day: habitual intake up to 400 mg a day raises no safety concerns for non-pregnant adults.15
  • Pregnancy: habitual intake up to 200 mg a day raises no safety concerns for the fetus.15
  • Breastfeeding: single doses and daily intakes up to 200 mg raise no safety concerns for breastfed infants.15,18
  • Children and adolescents: EFSA found the data insufficient to derive a safe intake, and proposed 3 mg/kg body weight per day as a safety level for habitual intake.15,16

The FDA cites 400 mg a day for most adults as an amount not generally associated with negative effects. It recommends that people who are pregnant, trying to become pregnant or breastfeeding talk to their health care provider, and notes that some medications can make people more sensitive to caffeine.12 It estimates that toxic effects, such as seizures, can be observed with rapid consumption of around 1,200 mg.12 These limits count caffeine from all sources, not coffee alone.15 The café survey above shows that a single large espresso can exceed 200 mg.7

What it does in the body#

Caffeine is absorbed rapidly and practically completely: about 20% from the stomach and the rest from the small intestine, within 45 minutes of drinking it.18,20 Its effects come from adenosine receptor antagonism: caffeine occupies the receptors and so blocks the action of adenosine. The A1 and A2 receptors are blocked at serum concentrations of 0.2–2 mg/L.18 On alertness, EFSA’s panel concluded that a product should contain at least 75 mg of caffeine per serving to bear an alertness claim, and did not accept a proposed lower amount of 40 mg.17

Regular users who stop can get withdrawal symptoms. A review of 57 experimental and 9 survey studies, selected by a literature search with set inclusion criteria, confirmed headache, fatigue and irritability among them; headache occurred in 50% of participants in the experimental studies.19 Symptoms typically began 12–24 h after the last dose, peaked at 20–51 h and lasted 2–9 days. Stopping doses as low as 100 mg a day was enough to produce them.19

How it is broken down#

Caffeine is almost entirely broken down in the liver.20 An estimated 95% of its clearance from the body is due to the liver enzyme CYP1A220; FSANZ puts the enzyme’s share at more than 90%.18 The main reaction gives paraxanthine and accounts for around 80–90% of caffeine demethylation; the remainder goes to theobromine (around 11%) and theophylline (around 4%).20

In healthy adults the half-life – the time the body takes to eliminate half of the caffeine – is about 4 h on average, with a range of 2–8 h, according to EFSA22; FSANZ gives 3–7 h.18 EFSA notes that it varies widely with age, body weight, pregnancy, medication and liver health.22 In more detail:

  • Smoking: increases caffeine clearance and shortens the half-life.20
  • Oral contraceptives: slow the elimination of caffeine.20
  • Pregnancy: in the first trimester the half-life is similar to that of non-pregnant women; it rises to about 10 h by week 17 and up to 18 h by the end of pregnancy.18
  • Newborns: up to 100 h, because their CYP1A2 enzyme system is immature; clearance reaches or exceeds adult levels by 5–6 months of age.18
  • Liver disease: various liver diseases strongly reduce caffeine clearance, cirrhosis most of all.20
  • Some drugs: fluvoxamine, pipemidic acid and norfloxacin inhibit caffeine clearance.20
  • Genes: twin studies show that CYP1A2 activity is largely heritable20, and variation in the CYP1A2 gene is a major cause of differences between people.18

Ask a doctor about your own situation.

Reference values#

Values as reported in the sources. They describe the range found in research and surveys, not a target. Bean values are % of dry weight unless stated; serving values are mg of caffeine per serving.

Parameter Value reported Source
Green arabica 0.9–1.5% dry weight Olechno et al.4
Green robusta 1.2–2.4% (basis not stated) Olechno et al.4
Robusta vs arabica About twice as much (beans); 1.4–1.8 times in most brew studies Ludwig et al.5; Olechno et al.4
Arabica, green to dark roast 12 → 13 mg/g (fresh weight) Ludwig et al.5
Share of caffeine extracted, double espresso (lab) About 67% (regular); about 85% (over-extracted) Ludwig et al.5
Caffeine left in spent grounds 5.6% (household filter); 16.4% (moka pot); 21.5–30.8% (professional high-pressure machines, restaurants) Vandeponseele et al.10
Very soluble compounds incl. caffeine Extracted very efficiently in the first seconds Mestdagh et al.9
Caffeine solubility in water 1 g per 5.5 mL at 80 °C; 1 g per 1.5 mL at 100 °C IARC23
Espresso, 20 g arabica, lab (means of 15 runs) 130 mg (20 g beverage) · 182.6 mg (40 g) · 204 mg (60 g) Schmieder et al.8
Double espresso, 18.1–20.4 g, lab 152–174 mg Ludwig et al.5
Espresso, 20 cafés, Glasgow 51–322 mg (cups 23–70 mL)* Crozier et al.7
Espresso, outlet means Scotland 72–212 mg (single cups 66–276 mg); Italy 73–135 mg; Spain 97–127 mg Ludwig et al.5
Cappuccino, outlet means, Scotland 101–275 mg Ludwig et al.5
Instant, 2 g 48–88 mg Ludwig et al.5
Filter, per 125 mL (review, converted) 71–173 mg Olechno et al.4
French press, per 120 mL (review, converted) 62–131 mg Olechno et al.4
Cold brew, per 120 mL (review, converted) 124–235 mg Olechno et al.4
Moka pot, per 40 mL (review, converted) 14–67 mg Olechno et al.4
Decaf, per 8 fl oz cup 2–15 mg FDA12
Decaf limit, coffee extracts (EU) ≤ 0.3% of coffee-based dry matter Directive 1999/4/EC13
Decaf limit, green and roasted coffee (Germany) ≤ 1 g/kg coffee dry matter KaffeeV14
Single dose, healthy adults ≤ 200 mg (about 3 mg/kg body weight) EFSA15,16
Daily, healthy non-pregnant adults ≤ 400 mg EFSA15; FDA12
Daily, pregnancy ≤ 200 mg EFSA15
Daily, breastfeeding ≤ 200 mg EFSA15; FSANZ18
Daily, children and adolescents 3 mg/kg body weight proposed EFSA16
Alertness claim, EFSA condition ≥ 75 mg per serving EFSA17
Half-life, adults About 4 h on average, range 2–8 h; 3–7 h EFSA22; FSANZ18
Half-life, newborns Up to 100 h FSANZ18
Half-life, pregnancy About 10 h by week 17; up to 18 h by the end of pregnancy FSANZ18
CYP1A2 share of caffeine clearance ≈95%; > 90% Grzegorzewski et al.20; FSANZ18
Metabolites Paraxanthine ≈80–90% of caffeine demethylation; the remainder goes to theobromine (≈11%) and theophylline (≈4%) Grzegorzewski et al.20

* Cup sizes of 23–70 mL are given in the text and the individual table rows; the table’s summary row reads 23–100 mL.7

The converted review values (filter, French press, cold brew, moka) are the authors’ own conversions of published concentrations to a standard serving; they come from different coffees and laboratories.4

04080120160200240280320mgEspresso, 20 cafés,Glasgow(cups 23–70 mL)51–322 mg7 Espresso, single cups,Scotland66–276 mg5 Espresso, outlet means,Italy73–135 mg5 Espresso, outlet means,Spain97–127 mg5 Double espresso, lab,18.1–20.4 g arabica152–174 mg5 Espresso, lab, 20 garabica, 20 → 60 g in cup130 · 182.6 · 204 mg8 Cappuccino, outlet means,Scotland101–275 mg5 Instant, 2 g48–88 mg5 Filter, per 125 mL(review, converted)71–173 mg4 French press, per 120 mL(review, converted)62–131 mg4 Cold brew, per 120 mL(review, converted)124–235 mg4 Moka pot, per 40 mL(review, converted)14–67 mg4 Decaf, per 8 fl oz cup2–15 mg12
Data table
mgSource
Espresso, 20 cafés, Glasgow (cups 23–70 mL)51–322 mg7
Espresso, single cups, Scotland66–276 mg5
Espresso, outlet means, Italy73–135 mg5
Espresso, outlet means, Spain97–127 mg5
Double espresso, lab, 18.1–20.4 g arabica152–174 mg5
Espresso, lab, 20 g arabica, 20 → 60 g in cup130 · 182.6 · 204 mg8
Cappuccino, outlet means, Scotland101–275 mg5
Instant, 2 g48–88 mg5
Filter, per 125 mL (review, converted)71–173 mg4
French press, per 120 mL (review, converted)62–131 mg4
Cold brew, per 120 mL (review, converted)124–235 mg4
Moka pot, per 40 mL (review, converted)14–67 mg4
Decaf, per 8 fl oz cup2–15 mg12
Caffeine per serving by brew type, as measured in published surveys and studies: Crozier et al.7, Ludwig et al.5, Schmieder et al.8, Olechno et al.4, FDA.12 Serving sizes differ between rows; compare within a row, not across the chart.

Caffeine during brewing#

Caffeine is among the very soluble compounds in roasted coffee. A textbook chapter groups it with N-methylpyridinium, sugars and organic acids: they are extracted very efficiently in the first seconds of the brew.9 Its solubility rises about fourfold between 80 and 100 °C, so small temperature changes in that range matter9: 1 g of caffeine dissolves in 5.5 mL of water at 80 °C and in 1.5 mL at 100 °C.23

The brew method sets how far extraction goes. Spent grounds from a household filter held only 5.6% of the original caffeine, a moka pot 16.4%, and professional high-pressure machines in restaurants 21.5–30.8%.10 In a lab study of espresso, about 67% of the caffeine reached the cup with a regular extraction, and about 85% when the shot was over-extracted with more water.5 In filter coffee, caffeine is extracted relatively more efficiently than total solids, most probably because the paper filter holds back solids but not caffeine.11

So caffeine does not track the extraction yield one to one. A flow-controlled espresso study kept the dose at 20 g of arabica and changed the beverage mass: caffeine rose from 130 mg in a 20 g ristretto to 182.6 mg in a 40 g espresso and 204 mg in a 60 g lungo (means of 15 runs at different flow rates, grind settings and temperatures).8 Most of each compound came out at the start of the shot, but the caffeine concentration in successive fractions fell more slowly than that of trigonelline and total dissolved solids, and flow rate and grind size changed caffeine only slightly.8 A longer shot therefore adds caffeine, but less and less. How dissolved solids behave overall is covered in Extraction yield and TDS; cold brew kinetics in Pour-over and immersion.

Per bean, per gram, per cup#

The “dark roast has less caffeine” question has three answers, depending on what is compared.

  • Per gram of roasted coffee: about the same or slightly more after a dark roast, because caffeine is stable and other matter is lost (12 → 13 mg/g in one arabica).5
  • Per bean: caffeine is stable through roasting, so the amount in each bean stays roughly unchanged while the bean gets lighter.1,5
  • Per cup: it depends on the brew. In a 2024 study of 30 combinations of variety, roast and brew time, brewed in an AeroPress at a 15:1 water-to-coffee ratio, dark roasts gave lower caffeine concentrations than light and medium roasts under identical brewing conditions – but higher ones at identical extraction yields.6 Extraction yields fell for roasts with mass losses above about 12–14%.6 The authors propose that roasting both removes soluble compounds and opens the bean’s pores, and that these effects compete.6

Metabolism in detail#

123456
  1. Caffeine. A purine ring system carrying three methyl groups, at positions 1, 3 and 7.1,20
  2. Liver. Caffeine is almost entirely broken down here.20
  3. CYP1A2. The liver enzyme behind an estimated 95% of caffeine clearance20; FSANZ puts its share at more than 90%.18 Each route removes one methyl group.20
  4. Paraxanthine. Methyl group 3 removed: the main reaction, around 80–90% of caffeine demethylation, mediated exclusively by CYP1A2.20
  5. Theobromine. Methyl group 1 removed: around 11%.20
  6. Theophylline. Methyl group 7 removed: around 4%.20

What changes the speed of clearance

  • Healthy adults: half-life about 4 h on average, range 2–8 h22; FSANZ gives 3–7 h.18
  • Smoking: faster clearance, shorter half-life.20
  • Oral contraceptives: slower elimination.20
  • Pregnancy: half-life about 10 h by week 17, up to 18 h by the end of pregnancy.18
  • Newborns: half-life up to 100 h, because CYP1A2 is immature.18
  • Liver disease: strongly reduced clearance, cirrhosis most of all.20
  • Some drugs: fluvoxamine, pipemidic acid and norfloxacin inhibit clearance.20
Main routes of caffeine breakdown in the human liver; the widths of the routes follow the approximate shares given in a systematic review of 141 publications.20 Schematic, not to scale.

The breakdown starts by removing one methyl group. Removing the methyl group at position 3 (N-3 demethylation) gives paraxanthine (1,7-dimethylxanthine), the main reaction; removing position 1 gives theobromine and position 7 theophylline.20 The main reaction is mediated exclusively by CYP1A2, and caffeine is used to measure CYP1A2 activity and to test liver function.20

Because one enzyme does most of the work, the factors that change CYP1A2 activity change how long caffeine stays in the body. Smoking speeds clearance and shortens the half-life; oral contraceptives have the opposite effect.20 During pregnancy the half-life rises to about 10 h by week 17 and up to 18 h by the end of pregnancy.18 Newborns clear caffeine very slowly because their CYP1A2 enzyme system is immature.18 Twin studies show that CYP1A2 activity is largely heritable.20 Variants in ADORA2A, the gene for the adenosine A2A receptor, influence caffeine’s effects on anxiety and sleep.18 A genome-wide meta-analysis of 47,341 people of European descent linked regions near CYP1A2 and near AHR, a gene that regulates CYP1A2, to how much caffeine people habitually drink.21

Measuring caffeine#

Studies measure caffeine by high-performance liquid chromatography (HPLC), usually with UV detection: Ludwig et al. quantified it at 280 nm5, Schmieder et al. at 272 nm8 and Vandeponseele et al. at 273 nm.10 A coffee refractometer reads total dissolved solids, not caffeine; a 2024 study used refractometry for extraction yield and HPLC for caffeine.6 To find the caffeine left in spent grounds, one study first extracted them with water in a Soxhlet apparatus for 4 h, then ran HPLC.10

Open questions#

  • Per-cup variability. Published surveys find up to six-fold differences between cafés.7 Dose, cup size and blend explain much of it5, but no source opened here splits the variation into these causes.
  • Converted values. Much of the per-serving data for filter, French press, cold brew and moka comes from concentrations converted to a standard serving by review authors, not from cups as served.4
  • Genotype and advice. Variants in CYP1A2 and the adenosine A2A receptor gene are linked to caffeine response and intake18,21, but genotype is rarely reported in pharmacokinetic studies.20 EFSA’s intake advice is set by population group, not by genotype.15

Key studies#

Crozier et al., 2012 – Espresso caffeine in Glasgow cafés. Bought espressos at 20 outlets and measured caffeine by HPLC. Caffeine per serving ranged from 51 to 322 mg, a six-fold difference, and four single cups held more than 200 mg.

Food & Function 3, 30–33 · doi:10.1039/C1FO10240K

Ludwig et al., 2014 – What are we drinking? Roasted one Brazilian arabica to three roast levels, brewed espressos in the lab and bought 104 espressos in Scotland, Italy and Spain. Caffeine was stable through roasting while chlorogenic acids fell, and outlet-to-outlet variation was greatest in Scotland.

Food & Function 5, 1718–1726 · doi:10.1039/c4fo00290c

Lindsey et al., 2024 – Roast degree, extraction yield and caffeine. Brewed 30 combinations of variety, roast and brew time and measured extraction yield by refractometer and caffeine by HPLC. Dark roasts gave less caffeine under identical brewing but more at identical extraction yield; the authors note that earlier studies had reported conflicting results.

Scientific Reports 14, 29126 · doi:10.1038/s41598-024-80385-3

Vandeponseele et al., 2021 – Caffeine left in spent grounds. Measured the caffeine remaining in spent grounds from professional high-pressure machines in restaurants, capsule machines, a moka pot and a household filter, relative to the roasted beans. Restaurant grounds kept the most caffeine; household filter and capsule grounds kept little.

Clean Technologies 3(2), 335–350 · doi:10.3390/cleantechnol3020019

Juliano & Griffiths, 2004 – Caffeine withdrawal. A review of 57 experimental and 9 survey studies found through a literature search with set inclusion criteria. It validated withdrawal symptoms such as headache, fatigue and irritability, and described their typical onset, peak and duration, even after stopping doses of about 100 mg a day.

Psychopharmacology 176(1), 1–29 · doi:10.1007/s00213-004-2000-x

General information, not medical advice.

Contested
  • Adult half-life: EFSA gives an average of about 4 h with a range of 2–8 h22; Food Standards Australia New Zealand (FSANZ) gives 3–7 h.18 The ranges overlap; both describe large differences between people.
  • Does dark roast have less caffeine? Per gram of roasted coffee, caffeine stayed the same or rose slightly from green to dark roast in one study (12 to 13 mg/g).5 But a 2024 brewing study found that, under identical brewing conditions, dark roasts gave lower caffeine concentrations in the cup than light and medium roasts – while at identical extraction yields they gave higher ones.6 A 2021 review calls the effect of roasting on caffeine in the brew inconclusive.4 The answer depends on whether one compares per bean, per gram or per cup, and on how the coffee is brewed.
  • How much more caffeine robusta has: Ludwig et al. state that robusta beans contain about twice as much as arabica5; a 2021 review of brewing studies found that robusta brews mostly contained 1.4–1.8 times as much, and notes one study in which an arabica espresso contained more.4 Bean and brew comparisons are not the same measure.
  • How fast caffeine leaves the grounds: a textbook chapter places caffeine among the very soluble compounds extracted in the first seconds9, but in a flow-controlled espresso study, caffeine concentration in successive fractions fell more slowly than that of trigonelline and total dissolved solids.8 Both agree that most caffeine is extracted early.
  • Genetics: FSANZ calls genetic variation in CYP1A2 a major cause of differences in its activity, and notes that variants in *ADORA2A*, the adenosine A2A receptor gene, influence caffeine’s effects on anxiety and sleep18; a genome-wide meta-analysis of 47,341 people linked regions near *CYP1A2* and *AHR* to habitual caffeine consumption.21 But a 2022 systematic review of 141 pharmacokinetic publications found the CYP1A2 genotype reported in only 5 of them.20 How much genotype should change personal advice is not settled in the sources opened for this entry.
Share or cite this entry

Sources

  1. BPietsch, A. (2017). Decaffeination – Process and Quality. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 10, pp. 225–243. doi:10.1016/B978-0-12-803520-7.00010-4 · other copies
  2. AWright, G. A.; Baker, D. D.; Palmer, M. J.; Stabler, D.; Mustard, J. A.; Power, E. F.; Borland, A. M.; Stevenson, P. C. (2013). Caffeine in floral nectar enhances a pollinator's memory of reward. Science 339(6124), 1202–1204. doi:10.1126/science.1228806 · other copies
  3. ACeja-Navarro, J. A.; Vega, F. E.; Karaoz, U.; Hao, Z.; Jenkins, S.; Lim, H. C.; Kosina, P.; Infante, F.; Northen, T. R.; Brodie, E. L. (2015). Gut microbiota mediate caffeine detoxification in the primary insect pest of coffee. Nature Communications 6, 7618. doi:10.1038/ncomms8618 · other copies
  4. AOlechno, E.; Puścion-Jakubik, A.; Zujko, M. E.; Socha, K. (2021). Influence of Various Factors on Caffeine Content in Coffee Brews. Foods 10(6), 1208. doi:10.3390/foods10061208 · other copies
  5. ALudwig, I. A.; Mena, P.; Calani, L.; Cid, C.; Del Rio, D.; Lean, M. E. J.; Crozier, A. (2014). Variations in caffeine and chlorogenic acid contents of coffees: what are we drinking?. Food & Function 5, 1718–1726. doi:10.1039/c4fo00290c · other copies
  6. ALindsey, Z. R.; Williams, J. R.; Burgess, J. S.; Moore, N. T.; Splichal, P. M. (2024). Caffeine content in filter coffee brews as a function of degree of roast and extraction yield. Scientific Reports 14, 29126. doi:10.1038/s41598-024-80385-3 · other copies
  7. ACrozier, T. W. M.; Stalmach, A.; Lean, M. E. J.; Crozier, A. (2012). Espresso coffees, caffeine and chlorogenic acid intake: potential health implications. Food & Function 3, 30–33. doi:10.1039/C1FO10240K · other copies
  8. ASchmieder, B. K. L.; Pannusch, V. B.; Vannieuwenhuyse, L.; Briesen, H.; Minceva, M. (2023). Influence of Flow Rate, Particle Size, and Temperature on Espresso Extraction Kinetics. Foods 12(15), 2871. doi:10.3390/foods12152871 · other copies
  9. BMestdagh, F.; Glabasnia, A.; Giuliano, P. (2017). The Brew – Extracting for Excellence. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 15, pp. 355–380. doi:10.1016/B978-0-12-803520-7.00015-3 · other copies
  10. AVandeponseele, A.; Draye, M.; Piot, C.; Chatel, G. (2021). Study of Influential Parameters of the Caffeine Extraction from Spent Coffee Grounds: From Brewing Coffee Method to the Waste Treatment Conditions. Clean Technologies 3(2), 335–350. doi:10.3390/cleantechnol3020019 · other copies
  11. AGloess, A. N.; Schönbächler, B.; Klopprogge, B.; D’Ambrosio, L.; Chatelain, K.; Bongartz, A.; Strittmatter, A.; Rast, M.; Yeretzian, C. (2013). Comparison of nine common coffee extraction methods: instrumental and sensory analysis. European Food Research and Technology 236, 607–627. doi:10.1007/s00217-013-1917-x · other copies
  12. AU.S. Food and Drug Administration (2024). Spilling the Beans: How Much Caffeine is Too Much?. U.S. Food and Drug Administration, Consumer Updates (content current as of 28 August 2024).
  13. AEuropean Parliament; Council of the European Union (1999). Directive 1999/4/EC of the European Parliament and of the Council of 22 February 1999 relating to coffee extracts and chicory extracts (consolidated version of 18 November 2013). EUR-Lex.
  14. AFederal Republic of Germany (2001). Verordnung über Kaffee, Kaffee- und Zichorien-Extrakte (Kaffeeverordnung, KaffeeV) vom 15. November 2001. Bundesgesetzblatt; official consolidated text at Gesetze im Internet (Bundesministerium der Justiz / Bundesamt für Justiz).
  15. AEFSA Panel on Dietetic Products, Nutrition and Allergies (NDA) (2015). Scientific Opinion on the safety of caffeine. EFSA Journal 13(5), 4102. doi:10.2903/j.efsa.2015.4102 · other copies
  16. AEuropean Food Safety Authority Caffeine (topic page). European Food Safety Authority.
  17. AEFSA Panel on Dietetic Products, Nutrition and Allergies (NDA) (2014). Scientific Opinion on the substantiation of a health claim related to caffeine and increased alertness pursuant to Article 13(5) of Regulation (EC) No 1924/2006. EFSA Journal 12(2), 3574. doi:10.2903/j.efsa.2014.3574 · other copies
  18. AFood Standards Australia New Zealand (2026). Supporting document 1 – Safety assessment of caffeine (P1056 – Caffeine review). Food Standards Australia New Zealand (FSANZ), document 385-26, 24 March 2026 (approval).
  19. AJuliano, L. M.; Griffiths, R. R. (2004). A critical review of caffeine withdrawal: empirical validation of symptoms and signs, incidence, severity, and associated features. Psychopharmacology 176(1), 1–29. doi:10.1007/s00213-004-2000-x · other copies
  20. AGrzegorzewski, J.; Bartsch, F.; Köller, A.; König, M. (2022). Pharmacokinetics of Caffeine: A Systematic Analysis of Reported Data for Application in Metabolic Phenotyping and Liver Function Testing. Frontiers in Pharmacology 12, 752826. doi:10.3389/fphar.2021.752826 · other copies
  21. ACornelis, M. C.; Monda, K. L.; Yu, K.; Paynter, N.; Azzato, E. M.; Bennett, S. N.; et al. (2011). Genome-Wide Meta-Analysis Identifies Regions on 7p21 (AHR) and 15q24 (CYP1A2) As Determinants of Habitual Caffeine Consumption. PLoS Genetics 7(4), e1002033. doi:10.1371/journal.pgen.1002033 · other copies
  22. AEuropean Food Safety Authority (2015). EFSA explains risk assessment: Caffeine. European Food Safety Authority.
  23. AIARC Working Group on the Evaluation of Carcinogenic Risks to Humans (1991). Caffeine. In: Coffee, Tea, Mate, Methylxanthines and Methylglyoxal (IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 51). International Agency for Research on Cancer (WHO), Lyon; p. 292.

Tier A · authoritative · B · trade and expert · C · general. Supporting quotes for every source are in the data version of this entry. If a publisher’s site refuses access, try the doi link or “other copies”.

Suggest a change

On Caffeine. Tell us what you would change and why. A source that supports it helps most.