Does Decaf Coffee Have Caffeine? Here Are the Numbers

Affiliate disclosure: Sweeter Grind is a participant in the Amazon Services LLC Associates Program. We may earn commissions on qualifying purchases at no extra cost to you.

Yes. Decaf coffee has caffeine in it — just very little. The Mayo Clinic's caffeine chart lists a brewed 8-ounce decaf at about 1 mg, against roughly 96 mg for the same cup of regular coffee; USDA FoodData Central's entry for brewed decaf works out to about 2.4 mg for the same serving. Those are averages, and the number that actually matters is the spread, which is wider than the average lets on.

Decaffeination is a subtraction, not an erasure. Every commercial process leaves a residue behind, and how much depends on the bean, the process and the person pulling your drink. Below are the measured figures, where they come from, and what they mean for the three situations that send people searching this question. For the wider picture — processes, flavour, buying — see our decaf coffee guide.

The Short Answer, With Numbers

Drink	Typical serving	Typical caffeine
Brewed coffee	8 oz	96 mg
Brewed decaf coffee	8 oz	1 mg
Espresso	1 oz	63 mg
Decaf espresso	1 oz	1 mg listed; 3.0–15.8 mg when measured
Instant coffee	8 oz	62 mg
Instant decaf coffee	8 oz	2 mg
Brewed black tea	8 oz	48 mg
Brewed green tea	8 oz	29 mg
Root beer (most brands)	8 oz	0 mg

Sources: the listed values are from the Mayo Clinic's current published caffeine content chart for coffee, tea and soda (the chart has been revised over the years, and older reprints of it circulate with slightly different figures). The measured decaf espresso range of 3.0–15.8 mg per shot is from McCusker and colleagues in the Journal of Analytical Toxicology, October 2006, discussed below.

Two things jump out. Decaf sits about two orders of magnitude below regular coffee, not somewhere in the middle. And the 1 mg beside decaf espresso is a rounded chart average, not a laboratory result for the shot in front of you — measured decaf shots run well above it. To add up a day of drinks, our caffeine calculator does the arithmetic.

Why Decaf Is Never Caffeine Free

"Decaf" is a threshold, not an absolute — and the threshold is set differently on each side of the Atlantic.

In Europe, the figure most often quoted is 0.1% caffeine by weight on a dry-matter basis for green and roasted coffee, which is another way of saying 99.9% caffeine-free. The EU-level instrument people cite for this, Directive 1999/4/EC, is narrower than its reputation: it governs coffee and chicory extracts — instant coffee, essentially — and permits the word "decaffeinated" only where anhydrous caffeine stays at or below 0.3% by weight of the coffee-based dry matter. The 0.1% figure for roasted beans comes from national food laws across Europe rather than from that directive.

The United States is vaguer. The widely repeated line is that the FDA requires 97% of the caffeine to be removed, and the National Coffee Association does describe decaffeination as typically removing around 97%, leaving approximately 2–15 mg in a cup. But that is an industry description of what the processes achieve, not a federal standard of identity with a percentage written into it. What the FDA regulates precisely is solvent residue: 21 CFR 173.255 caps residual methylene chloride at 10 parts per million in decaffeinated roasted coffee, and 21 CFR 173.290 caps trichloroethylene at 25 parts per million in decaffeinated ground coffee.

The two systems are expressed on different bases — one a ceiling on what remains, the other a description of what is taken out. The logic is the same either way: because the rule is proportional, a bean that starts with more caffeine finishes with more. For the mechanics of how the caffeine comes out, see how decaf coffee is made.

How Much Caffeine Is Actually in Your Cup

The most useful study on this measured real drinks bought over a counter rather than samples prepared in a lab.

McCusker, Fuehrlein, Goldberger, Gold and Cone, publishing in the Journal of Analytical Toxicology in October 2006 (volume 30, issue 8, pages 611–613), ran it in two phases. In phase one they collected ten decaffeinated samples from different coffee establishments and found caffeine ranging from 0 to 13.9 mg per 16-ounce serving. In phase two they returned repeatedly to a single Starbucks outlet: six decaf espresso samples came in at 3.0–15.8 mg per shot, and six brewed decaf samples at 12.0–13.4 mg per 16-ounce serving.

That is the whole article in one paragraph. One outlet served a genuine zero. Another served nearly 14 mg in a 16-ounce cup — around seven times what the chart above implies for a drink that size. And the same shop, on different days, varied by a factor of five on espresso.

That last part is not unique to decaf. The same group had already shown, in Caffeine Content of Specialty Coffees (Journal of Analytical Toxicology, 2003, volume 27, issue 7, pages 520–522), that the identical caffeinated beverage from the identical outlet on six consecutive days ranged from 259 to 564 mg per dose. Coffee shops are not calibrated instruments; decaf inherits the same variability at a smaller scale.

Our reading: treat 1–2 mg per 8-ounce cup as the central case and 10–15 mg per large drink as a realistic ceiling, and do not treat any single figure as what is in the cup in front of you.

What Makes One Decaf Stronger Than Another

Four variables account for almost all of the spread.

The species it started as. Widely published composition figures put Arabica at roughly 1.2–1.5% caffeine by dry weight and Robusta at roughly 2.2–2.7%, about double. Because decaffeination is proportional, a Robusta that has lost 97% of its caffeine still holds about twice the residue of an equally-treated Arabica. We break the difference down in Arabica vs Robusta.

How hard the process was run. Decaffeination is a stopping decision, and a producer who runs to the European 0.1% ceiling ends up somewhere different from one who stops at the 97% the National Coffee Association describes as typical. Neither is cheating; they are different targets.

Dose and brew method. More grounds and longer contact with water mean more of everything, caffeine included. It is also why the common assumption about espresso is backwards: per shot, espresso carries less caffeine than a mug of drip, with the Mayo Clinic chart putting a 1-ounce espresso at 63 mg against 96 mg for an 8-ounce brewed coffee. Espresso is more concentrated and much smaller. It only out-caffeinates drip at two or three shots.

Serving size. Almost every published chart uses 8 ounces; almost nobody drinks 8 ounces. A 16-ounce shop cup is two chart servings. The McCusker figures are quoted per 16 ounces for exactly this reason.

If You Are Pregnant or on a Medical Caffeine Limit

This is why a lot of people are reading this page, so here is the arithmetic — though the decision is yours and your clinician's, not ours.

Guidance from the American College of Obstetricians and Gynecologists recommends limiting caffeine during pregnancy to less than 200 mg a day. The European Food Safety Authority reached a similar figure in its 2015 opinion on caffeine, concluding that regular intake up to 200 mg per day by pregnant women is not a safety concern for the foetus, against 400 mg a day for healthy adults generally. The FDA's own consumer guidance stops short of a number for pregnancy and instead recommends talking to your health care provider about whether you need to limit caffeine.

Set against a 200 mg ceiling, decaf is arithmetically small. At the 1 mg per 8-ounce cup the Mayo Clinic chart lists, 200 mg is around 200 cups; at the roughly 2.4 mg implied by USDA FoodData Central it is still more than 80; even at the worst case in the McCusker data — 13.9 mg per 16-ounce serving — it is roughly 14 large drinks. Neither figure is a recommendation. The point is that decaf's contribution to a daily total is normally a rounding error next to one regular coffee at 96 mg. It is not zero, though, so if a clinician has set you a limit, count decaf into it — and count your actual cup size, not the chart's.

Will Decaf Keep You Awake?

Probably not on the caffeine alone, but "probably" is doing real work in that sentence.

Clearance varies enormously between individuals. The Austrian Agency for Health and Food Safety (AGES) puts caffeine's plasma half-life in adults at 2 to 8 hours, and notes it can stretch to as much as 18 hours in pregnancy because estrogen and progesterone inhibit the enzyme that metabolises it. The FDA makes the same point qualitatively, noting wide variation in both how sensitive people are and how fast they clear it.

Run the slow end of that range against a decaf dose and the numbers stay small: a 1–2 mg cup is 1–2 mg whether you clear it in two hours or eight, and even a 14 mg outlier is well under a fifth of one regular coffee.

What that does not rule out is everything else about a 9pm coffee — a hot drink, a full bladder, sugar, and the fact that expecting to be kept awake is itself good at keeping people awake. If evening decaf reliably costs you sleep, believe your own experience over the milligram count, and check that the shop actually served you decaf. The McCusker data suggests that is not always obvious from the cup.

If You Are Caffeine Sensitive

The FDA's framing is the useful one: there is wide variation in how sensitive people are, and no chart can tell you where you sit on it.

Two practical moves. Count decaf into your daily total rather than treating it as free — 15 mg here and there is not nothing if your threshold is low. And brew at home, where you control the grams and the water, rather than relying on a shop's pour.

Which Decaf Has the Least Caffeine

The buying answer follows directly from the variables above.

Buy Arabica, and buy it stated. A bag naming a single origin and a variety is almost certainly Arabica; an unlabelled supermarket blend may carry Robusta, which starts with roughly double the caffeine and finishes with roughly double the residue. How to read a coffee bag label covers the signals worth trusting.

Do not expect one process to win outright. Every mainstream method — solvent, sugarcane ethyl acetate, water-based, supercritical CO2 — can hit the same thresholds, and none publishes a per-bag milligram figure. The differences are mostly about flavour retention and solvent choice, which we compare in Swiss Water versus sugarcane EA. Process matters at the margin: a roaster advertising the European 0.1% ceiling has told you something specific, where "97% caffeine free" is the National Coffee Association's description of what decaffeination typically achieves, restated on a bag.

Control the serving. Residue scales with dose, so an 8-ounce cup of a good Arabica decaf is the lowest-caffeine version of a real coffee you can make. For the beans themselves, see the best decaf coffee beans.

When You Want Genuinely Zero

If your requirement is genuinely zero rather than nearly zero, decaf is the wrong tool. Herbal infusions — rooibos, chamomile, peppermint — contain no caffeine because they are not made from the coffee or tea plant, and chicory, barley and dandelion "coffees" are caffeine-free for the same reason. None tastes like coffee, which is the trade.

For everyone else the summary is simple. Decaf has caffeine: roughly 1–2 mg per 8-ounce cup on the published charts, rarely more than about 15 mg in a large drink, against roughly 96 mg for regular. Small enough to be irrelevant to most people, large enough to be worth counting for a few. For the rest of the decaf picture — how it is made, which processes preserve flavour, what to buy — start with Sweeter Grind's decaf coffee guide.

Frequently Asked Questions

Is decaf coffee caffeine-free?

No. Decaffeination removes most of the caffeine, not all of it. The National Coffee Association describes the process as typically removing around 97% and leaving roughly 2–15 mg in a cup; the Mayo Clinic chart lists a brewed 8-ounce decaf at about 1 mg. Drinks labelled "caffeine free" — herbal infusions, chicory — are a different category.

How much caffeine is in a cup of decaf coffee?

About 1 mg per 8-ounce brewed cup according to the Mayo Clinic's caffeine chart; USDA FoodData Central lists brewed decaffeinated coffee at 1 mg per 100 g, which works out to roughly 2.4 mg per 8 fl oz. Real-world drinks vary more: McCusker and colleagues, in the Journal of Analytical Toxicology in 2006, found 0 to 13.9 mg per 16-ounce serving across ten samples from different outlets.

Can decaf coffee keep you awake?

Unlikely on caffeine content alone. The Austrian Agency for Health and Food Safety puts caffeine's plasma half-life in adults at 2 to 8 hours, but even the slow end of that range works on a dose of a few milligrams rather than the roughly 96 mg in a regular 8-ounce coffee. Sensitivity varies widely, as the FDA notes, so if evening decaf consistently disrupts your sleep, that experience is worth respecting.

Which decaf coffee has the least caffeine?

Arabica-based decaf, in a smaller serving. Composition figures generally put Arabica at roughly 1.2–1.5% caffeine by dry weight against 2.2–2.7% for Robusta, and because decaffeination removes a proportion rather than a fixed amount, the lower-caffeine bean finishes lower. No mainstream process is systematically the lowest, and no roaster publishes a per-bag milligram figure.

Does decaf espresso have more caffeine than decaf drip coffee?

Per shot, usually less — but the ranges overlap. McCusker and colleagues measured decaf espresso at 3.0 to 15.8 mg per shot and brewed decaf at 12.0 to 13.4 mg per 16-ounce serving at the same outlet, meaning a strong decaf shot can carry more than a large brewed decaf. The same pattern holds for regular coffee: the Mayo Clinic chart puts a 1-ounce espresso at 63 mg against 96 mg for 8 ounces of brewed.

Related reading

• How decaf coffee is made — all four methods explained
• Swiss Water versus sugarcane EA — the two specialty processes compared
• Arabica vs Robusta — why the species sets the caffeine ceiling
• The best decaf coffee beans — our current picks
• How to read a coffee bag label — the signals worth trusting
• Our caffeine calculator — add up a day's intake