Light, oxygen and the container itself

Several residues absorb ultraviolet light and react as a result. Tryptophan is the most susceptible, tyrosine and phenylalanine follow, and cysteine's disulfide bonds can be broken photochemically — which is the mechanism behind disulfide scrambling in material left in the light.

The damage accumulates with exposure. It is not a threshold that is crossed but a total that adds up, which is why the sensible arrangement is simply never to expose the material rather than to limit how long it sits out.

Amber glass is the usual answer, filtering the shorter wavelengths that carry the energy. Opaque secondary packaging achieves the same thing and is why vials often arrive boxed rather than merely wrapped.

The space above the material in a sealed vial is not nothing. Whatever is in it stays in contact with the contents for as long as the vial exists.

If that headspace is ordinary air, it is roughly a fifth oxygen, and oxygen is the reagent for the oxidation route — methionine and cysteine first, tryptophan and histidine under harsher conditions. Displacing it with nitrogen or argon before sealing removes the reagent, which is why inert backfilling is standard for material intended to keep.

Sealing under vacuum achieves a related end by removing most of the gas altogether. It has a useful side effect: a vial that has lost its vacuum can often be identified, which makes seal failure detectable rather than silent.

A stopper is a more demanding component than it looks. It has to seal against moisture ingress for the life of the product, tolerate the vacuum or inert gas behind it, and neither absorb the contents nor contribute anything to them.

Elastomer choice governs how much water vapor crosses the closure over time, and it is never precisely zero — a sealed vial slowly equilibrates with its surroundings across months and years. This is one reason shelf life is finite even for material stored perfectly cold.

Coring is the other consideration: fragments of stopper displaced by a needle, which introduce particulate matter into the container. It is a handling matter rather than a manufacturing one, but it is a reason the state of a closure is worth a look.

Pharmaceutical-grade borosilicate glass is used because it is chemically resistant and does not readily give up its constituents to what it holds. Ordinary soda-lime glass is less inert.

Delamination is the failure mode worth knowing about: thin flakes of glass separating from the inner surface, driven by the conditions the vial was formed and treated under. It appears over time and is more likely with certain solution chemistries than with a dry solid.

For freeze-dried material held sealed and dry the risk is low. It is one of the reasons material is supplied dry rather than pre-dissolved.

It is tempting to treat packaging as logistics and analysis as science. In stability terms they are the same subject: a certificate describes material in a container, under conditions the container is maintaining.

Amber glass excludes the light that drives photodegradation. Inert headspace excludes the oxygen that drives oxidation. The closure excludes the moisture that drives hydrolysis and deamidation. Each element removes the reagent for a specific reaction, exactly as the storage conditions do — the container is simply doing it continuously, without anyone having to remember.