Molecular SO2 Is the Best Predictor of Rotten-Egg Off-Flavor in Canned Wine

Cornell researchers packaged ten commercial wines (six white, two rosé, two red) in both glass and aluminum cans with one of three liners — BPA epoxy, BPA-non-intent (BPA-NI) epoxy, or acrylic — and tracked hydrogen sulfide (H2S, "rotten egg" aroma) over eight months of storage at room temperature.

The results were stark. Wines in glass stayed essentially free of H2S the whole time. The same wines in acrylic-lined cans produced up to 1307 µg/L H2S — some samples exceeded the highest level ever reported in a survey of flawed commercial wines within just one month. BPA and BPA-NI epoxy liners performed far better (median 11.8 µg/L, maximum 51.8 µg/L after 32 weeks) but still weren't immune: two white wines crossed the ~1 µg/L sensory threshold within two months, and most whites and rosés eventually did.

What actually predicts the problem? The researchers tested wine chemistry (pH, alcohol, sugar, chloride) alongside SO2 in its free, total, and molecular forms. Molecular SO2 — the small, uncharged fraction of free SO2 that dominates at low pH — was the single best predictor of H2S formation, both in the eight-month cans and in a fast 3–14 day accelerated-aging test using lined aluminum coupons in sealed vials at 50°C. Free SO2 and pH individually were less predictive; alcohol, chloride, copper, and total SO2 showed no meaningful correlation at all.

The two red wines in the study developed almost no H2S, which tracked with their naturally low molecular SO2 (a consequence of their higher pH). This suggests reds are generally lower-risk for canning from a sulfide standpoint, all else equal.

Interestingly, dissolved aluminum (a direct marker of can corrosion) correlated poorly with H2S — visible liner damage (small blisters, eventually delamination for acrylic) was a better sign of trouble than measuring aluminum alone. This implies H2S formation and bulk aluminum corrosion may follow somewhat different pathways, with liner failure as the common upstream cause.

Practical target: to keep H2S under 10 µg/L through roughly six months of room-temperature can storage, the researchers recommend keeping molecular SO2 below 0.4 mg/L for wines in BPA or BPA-NI epoxy cans, and avoiding acrylic-lined cans entirely for wine. Longer shelf-life targets or warmer storage would call for an even lower ceiling.

Source: Montgomery, Allison, Goddard & Sacks (2023), "Hydrogen Sulfide Formation in Canned Wines Under Long-Term and Accelerated Conditions," American Journal of Enology and Viticulture 74:0740011.