Short answer
A few plastics are chemically self-destructive. Cellulose acetate and cellulose nitrate release acid as they break down, and the acid speeds the breakdown, so an object can look sound for decades and then yellow, craze and warp within a few years. PVC yellows as it sheds hydrogen chloride. Because these are thermal and hydrolytic reactions rather than light damage, the published controls are temperature and humidity: a measured activation energy of 86 kJ/mol for PVC yellowing implies the rate roughly halves for each five-degree drop near room temperature.
The Getty Conservation Institute lists the usual signs of a deteriorating plastic — discolouration, which means yellowing or turning opaque, then crazing, cracking, warping, a sticky surface as plasticiser migrates out, and in extreme cases collapse to powder — and stresses how little warning there can be. Cellulose acetate and cellulose nitrate are, in its words, among the four plastics that deteriorate most rapidly in museum collections. Both were early, experimental and poorly stabilised formulations. As they hydrolyse they give off acid gases, and those gases catalyse further hydrolysis, so decay accelerates once it starts. The Getty cites damage to sculptures by Naum Gabo in Tate's collection, and states plainly that for degraded objects of these two plastics there have so far been no viable treatments.
Plasticised PVC yellows by a different route. Hydrogen chloride is eliminated from the polymer chain, leaving runs of alternating double bonds that absorb blue light. A 2023 study turned this into a damage function for heritage collections by ageing samples at 50 and 70 °C and following the rise in the CIELAB b* coordinate. Lower temperature, lower relative humidity, higher polymer molecular weight and — less expected — higher plasticiser content were each associated with slower yellowing, and the activation energy came out at 86 kJ/mol. The authors' purpose was practical: to let a collection manager compare the benefit of cooling a store with that of drying it, and to identify which objects are most at risk. It is one of the few places where a colour change in a museum plastic has been reduced to an equation.
An activation energy converts directly into a temperature sensitivity through the Arrhenius relation, and the table works it out for PVC. Relative to 20 °C, the yellowing rate is about a third lower at 17 °C and about halves by 14 or 15 °C; at 25 °C it is roughly 1.8 times faster, and at 30 °C more than three times. That is steeper than the old rule that reaction rates double every ten degrees, and close to the Canadian Conservation Institute's rule of thumb that five degrees doubles an object's life. The figures assume everything else — humidity, the plastic's own composition — stays fixed, and they extrapolate from tests at 50 and 70 °C down to room conditions, which is the weakest step in any accelerated-ageing argument. They indicate direction and scale, not a date.
Without a cure, the documented approach is to know what is there and slow it. The Getty's project surveys collections, keeps a reference collection of plastics and tests possible treatments on study objects. For Disney's animation cels its scientists first established a baseline: which plasticisers each cellulose acetate cel contains, how much has evaporated, and whether the oldest show hydrolysis, so that later change can be measured against it. The Canadian Conservation Institute's advice for display is modest light with ultraviolet removed, since light adds photo-oxidation to the thermal reactions. None of this restores a yellowed object. The yellow in PVC or an old appliance casing is a change in the polymer; the page on plastic yellowing explains the chemistry for household plastics.
| Storage temperature | Rate relative to 20 °C | Time to the same yellowing |
|---|---|---|
| 5 °C | 0.15 | 6.7 times as long |
| 10 °C | 0.29 | 3.5 times as long |
| 15 °C | 0.54 | 1.8 times as long |
| 17 °C | 0.69 | 1.4 times as long |
| 20 °C | 1 (reference) | 1 (reference) |
| 23 °C | 1.43 | 0.7 times as long |
| 25 °C | 1.81 | 0.6 times as long |
| 30 °C | 3.20 | 0.3 times as long |
| Colourant | What changed | Driver | Evidence and caveat |
|---|---|---|---|
| Poly(vinyl chloride), plasticised | Clear or white PVC turns yellow (rising b*). | Heat; Humidity and water | Strong. Derived from ageing at 50 and 70 °C and extrapolated to room conditions. |
| Cellulose acetate and cellulose nitrate | Discolouration (yellowing or opacifying), warping, crazing and stickiness. | Humidity and water; Heat; Acidity or alkalinity | Strong. The signs listed are the Getty's for plastics generally. |
| Polystyrene, ABS, polycarbonate and other aromatic polymers | White and pale mouldings turn yellow, then brown. | Light and UV; Oxygen | Strong. Formulation, pigment and stabiliser package change the rate greatly. |
Each statement is labelled by kind — established fact, a standard’s requirement, observed market data, a convention, or Colourwise’s own interpretation or analysis — with the strength of the evidence behind it.
FactStrong evidence
Cellulose acetate and cellulose nitrate are among the four plastics that deteriorate most rapidly in museum collections; their degradation produces acid gases that cause autocatalytic breakdown, and no viable treatment for degraded objects had been established when the Getty's project page was written.
Caveat: An institutional research summary; undated on the page.
Source: Treatment Strategies for Objects Made of Cellulose Esters (project page)
FactStrong evidence
For PVC in heritage collections, lower temperature, lower relative humidity, higher polymer molecular weight and higher plasticiser content were associated with slower yellowing (rise in b*), with an activation energy of 86 kJ/mol.
Method: Multiple linear regression on b* change during accelerated degradation.
Caveat: Accelerated ageing at 50 and 70 °C, extrapolated to room conditions.
Colourwise analysisModerate evidence
With an activation energy of 86 kJ/mol, the PVC yellowing rate at 15 °C is about 0.54 times, and at 25 °C about 1.81 times, the rate at 20 °C.
Based on: Colourwise calculation from the Arrhenius relation k(T)/k(20 °C) = exp[−(Ea/R)(1/T − 1/293.15 K)] with Ea = 86 kJ/mol as published and R = 8.314 J/(mol·K), all other variables held constant.
Caveat: Assumes one activation energy holds from 70 °C down to room temperature.
FactStrong evidence
A survey of cellulose acetate animation cels in the Walt Disney Animation Research Library identified phosphate and phthalate plasticisers, evidence that more volatile plasticisers had evaporated, and some evidence of hydrolysis in the oldest cels.
Caveat: A compositional baseline; the abstract does not report colour.
Reviewed 6 October 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.