Canning Wine and Cider: A Practical Checklist from Cornell Extension

This is a field-oriented summary of Cornell's canned-wine research program, presented at a 2024 cider/wine/spirits industry conference. It distills several years of lab findings (covered in more depth in our other canning articles) into direct guidance for producers.

The core problem, restated simply: a can is basically a thin plastic bottle wrapped in an aluminum shell. Sulfited wine and cider react with any exposed aluminum to form hydrogen sulfide (H2S), and canned wines run roughly 10x higher average free H2S than the same wines in glass (about 13 ppb versus roughly 1 ppb) — well above the ~1.6 ppb odor detection threshold. Three named failure modes are worth knowing: staling (oxidation from headspace air or a lost seal), scalping (the liner absorbing desirable aroma compounds, more of a concern for hop-forward or infused beverages), and tainting (haze or off-flavor from corrosion products).

SO2 management is the single highest-leverage lever a producer has, because cans change the calculus for both roles SO2 normally plays. Free SO2 (as bisulfite) is an antioxidant — but a can's hermetic seal already excludes oxygen, so wines don't need the same antioxidant cushion (~30 mg/L) recommended for glass. Molecular SO2 is the antimicrobial fraction — typically ~0.6 mg/L is recommended to prevent spoilage — but it's also the single best predictor of H2S formation in cans. There isn't yet a better same-cost antimicrobial substitute for molecular SO2, though high-pressure processing (HPP) and DMDC (velcorin) are options worth considering for wines with residual sugar. Net guidance: keep molecular SO2 under roughly 0.5 mg/L for canned product.

On liners: acrylic is consistently the worst performer in every trial run so far, by a wide margin, and should be avoided for any sulfited beverage. BPA epoxy and BPA-NI epoxy liners perform comparably to each other and are the safer choice. A newer liner technology, aTULC (a thick laminated polyester bonded to the aluminum, rather than sprayed on), looks promising in preliminary trials — it outperformed a standard BPA-NI liner at 4 and 8 months in a high-SO2 wine — though more trials are needed before treating it as proven.

A few other findings worth flagging: bare, unlined aluminum should never be used as a stand-in for a lined can in your own shelf-life testing — it behaves completely differently (even showing an opposite response to alcohol content) and isn't a valid predictor of how a real lined can will perform. Corrosion and H2S formation are consistently highest and most variable in the can's headspace/neck region, not the body. And because these systems are highly variable can-to-can, run as many replicates as practical when qualifying a new wine or can source — a single bad can can ruin an entire pallet's reputation even if the average result looks fine.

Source: Austin Montgomery (Cornell University), "How Wine/Cider Components Affect Shelf Life in Cans," Congrès cidres, vins et alcools d'ici, 2024.