Extraction Slows Down. Oxidation Doesn't.

A 2022 study out of Cognac tracked C-glucosidic ellagitannins — the castalagin and vescalagin family — through a full ageing cycle, across eight different barrel toast levels (Gadrat et al., 2022, Molecules). The compounds extract fastest in the first few months, then taper off, which lines up with what most distillers already assume about new wood.

What's less intuitive: the ellagitannin-derived compounds — the oxidised "whisky tannins" that actually shape mouthfeel, bitterness and colour in the finished spirit — kept climbing for the full length of the study, well after the raw tannin extraction had slowed.

In other words, a tired barrel isn't handing you a diluted version of a young barrel's profile. It's handing you a different ratio: less raw tannin available to extract, but a spirit that's had more time to convert whatever tannin already made it into solution. Stretching time in a used barrel to compensate for fewer fills changes the balance of what ends up in the glass — not just the intensity.

That matters most when you're blending. Two casks pulled from the same warehouse, filled the same number of times, can still land at different points on that curve depending on how long each one has actually spent resting between fills. Averaging across barrels doesn't average out the chemistry — it just hides which barrels are pulling the profile in which direction.

A 2025 study pushes this even further, comparing oak staves that have never touched spirit against staves in contact with alcohol for 2, 50 and 100 years (Cournut et al., 2025, Journal of Mass Spectrometry). Castalagin and vescalagin decline sharply over that century, consistent with the extraction-and-conversion pattern above. But not every compound moves the same way. Ellagic acid, one of the compounds most directly responsible for a spirit's astringency, stays remarkably stable across all four stave ages — present in roughly the same amount whether the wood is freshly toasted or has been giving up compounds into spirit for a hundred years.

The same unevenness shows up even within a single class of molecules. The lignan lyoniresinol holds essentially steady for decades, while another compound, scorzonoside, doesn't turn up until fifty years or more of alcohol contact, and others fade and disappear entirely as ageing continues. There isn't one clock running down inside a barrel. There are several, moving at different speeds, and a barrel's age tells you less about its character than which of those clocks has moved furthest.

None of this is getting easier to manage

Fill count, species and toast were already hard to track barrel by barrel. The financial incentive to lean harder on used casks is only making that more common. Used barrel prices, which traded above $200 at the end of 2024, dropped to around $50. New charred oak barrels, meanwhile, have held near $325 apiece.

Numbers make the pressure easy to see. The underlying cost — interest on tied-up inventory, warehouse space, insurance — climbs for any producer holding barrels longer, anywhere. More distillers extending refill programmes, right as the chemistry inside those barrels gets harder to predict, is not a coincidence. It's the direction the economics push everyone.