Last updated:
Diacetyl is the off-flavour that has ruined more otherwise excellent batches of homebrew than any other single defect, I know this from painful experience with an amber ale that smelled like movie theatre popcorn butter for three weeks before I understood why. The chemistry of diacetyl formation and reabsorption is well-established enough that once you understand the mechanism, both the cause and the cure become obvious, and the problem becomes entirely preventable.
Why your beer tastes like butter: troubleshooting and preventing diacetyl in homebrewed beer
What diacetyl is and where it comes from: Diacetyl (2,3-butanedione) is a vicinal diketone (VDK) produced during yeast fermentation as a byproduct of valine biosynthesis. The metabolic pathway: yeast synthesises valine (an amino acid) and excretes acetolactate as an intermediate byproduct into the beer. Acetolactate is colourless and odourless. When beer is exposed to oxygen or warmth, acetolactate undergoes oxidative decarboxylation to form diacetyl (the buttery compound). Yeast then reabsorbs diacetyl and reduces it further to acetoin and 2,3-butanediol, both odourless and flavourless compounds. The complete process therefore is: yeast ferments → excretes acetolactate → acetolactate converts to diacetyl (slow, oxygen/heat-dependent) → yeast reabsorbs diacetyl and reduces it to harmless compounds. The problem occurs when this last step is incomplete: if the beer is separated from yeast (racked, packaged) before diacetyl reabsorption is complete, the diacetyl remains in the finished beer. Conditions that cause diacetyl in finished beer: Racking too early: removing beer from the yeast before fermentation and VDK cleanup is complete. The most common cause. The solution: wait for stable final gravity (confirmed by two consecutive hydrometer readings 24 hours apart), then wait an additional 24–48 hours at fermentation temperature (“diacetyl rest”) before packaging. Underpitching: low cell counts lead to stressed, slow fermentation, the yeast produces more acetolactate per unit of sugar fermented, and the reduced yeast population is less efficient at reabsorbing diacetyl. Solution: pitch at appropriate cell count for the wort volume and gravity. Cold crashing before diacetyl cleanup: dropping temperature prematurely flocculates and settles the yeast before it completes VDK reduction. The settled yeast is less active and less effective at diacetyl reabsorption. Solution: complete the diacetyl rest at fermentation temperature before cold crashing. Diacetyl rest: warm the beer to 20–22°C (or maintain at fermentation temperature) for 48–72 hours after reaching final gravity. This keeps the yeast active and ensures complete VDK reduction before cold conditioning. Low fermentation temperature at end of fermentation: as fermentation proceeds, CO2 production slows and the brewer may be tempted to cold crash. But residual yeast at cool temperatures is slow to reabsorb VDKs. The solution is the diacetyl rest, raise the temperature at the end of fermentation (even by 2–3°C) to accelerate the final cleanup. Certain yeast strains: some yeast strains are higher diacetyl producers than others. English ale strains (WLP002, Wyeast 1968) are known diacetyl producers, this is part of their character and is intentional in English ales at low levels. For clean American styles, use low-diacetyl strains (US-05, WLP001, W-34/70). Hop creep from late dry hopping: dry hops contain active enzymes (amyloglucosidase and other beta-glucanase-related enzymes) that can restart fermentation of residual dextrins in the beer. This renewed fermentation produces fresh acetolactate. The solution: ensure diacetyl rest is performed after dry hopping (not before), as any hop-creep-induced additional fermentation will generate fresh VDKs. Detecting diacetyl, the forced diacetyl test: The forced diacetyl test converts any residual acetolactate (odourless precursor) to diacetyl (detectable) before the beer is packaged: Take a 30mL sample of beer. Warm the sample to 65°C in a sealed container for 15–20 minutes (this oxidises all acetolactate to diacetyl). Cool the sample to room temperature. Smell and taste immediately. If buttery/butterscotch aroma is detectable in the forced sample: the beer still has acetolactate that will convert to diacetyl in the package. Continue the diacetyl rest for another 24–48 hours and retest. If no butter aroma: the beer is safe to package. This test is highly recommended for any batch before packaging, it takes 20 minutes and prevents the most common packaging-time mistake in homebrewing. India-specific context: Indian ambient temperatures in summer (30–40°C): paradoxically, Indian summer temperatures accelerate diacetyl formation (faster acetolactate oxidation) but also accelerate yeast reabsorption. The net result: beers fermented in Indian summer at ambient may have faster diacetyl cleanup than beers in cooler environments. However, if racked before the cleanup is complete, the high ambient temperature will rapidly oxidise any remaining acetolactate to diacetyl, causing a noticeable butter problem in the packaged beer. In Indian conditions: always complete the forced diacetyl test before packaging, particularly when fermenting at ambient temperatures. Lager fermentation (rare in India without refrigeration): lager yeast at cold temperatures produces VDKs more slowly and reabsorbs them more slowly, the diacetyl rest for lager (warming from 10°C to 18°C for 48–72 hours) is even more important than for ales.
Common Questions
Can I fix diacetyl in already-packaged beer, or is the batch ruined?
Diacetyl in already-packaged beer can sometimes be fixed, but it depends on the packaging method and how advanced the diacetyl contamination is. For bottle-conditioned beer: if the beer was bottled with priming sugar and is still in the conditioning phase (within 2–3 weeks of bottling), the active yeast in the bottle may continue to reabsorb diacetyl as conditioning proceeds. Warm the bottles to 20–22°C for 48–72 hours (placing in a warm room or wrapping in blankets). The remaining active yeast at conditioning temperature may reduce the diacetyl level to below perception threshold. This fix works best when the diacetyl level is moderate (noticeable but not overwhelming) and the yeast is still active in the bottle. For keg-conditioned beer: transfer the keg to a warm environment (20–22°C) with a small amount of actively fermenting fresh starter (1/4 teaspoon of rehydrated dry yeast in 50mL of pre-fermented starter, allowed to ferment for 12 hours before adding), this introduces fresh, active yeast cells to the keg. Leave at 20–22°C for 48–72 hours, then cold crash again. This technique (“Krausening” or “priming with active yeast”) can reduce diacetyl in a keg that was packaged slightly too early. When the batch is not salvageable: severe diacetyl (the beer smells like movie theatre butter even after warming attempts) combined with poor yeast health (the yeast flocculated completely and is not viable) will not recover. In this case: consume the beer as a lesson rather than a complete loss, mild-to-moderate diacetyl is less offensive to non-brewing guests than to the brewer who knows what they’re detecting. Prevention is always better: the forced diacetyl test before every packaging is a 20-minute investment that prevents this situation entirely. After one diacetyl incident, no experienced homebrewer skips this test.
