The carbonyl index and how far the degradation has gone

Weathering damage is usually described in visual terms — faded, chalky, brittle, crazed — and those descriptions are poor evidence because they are comparative without a reference and subjective without a scale. The underlying chemistry, by contrast, is directly measurable. Photooxidation introduces carbonyl groups that were not present in the original polymer, in quantities that rise with accumulated ultraviolet dose, and the concentration of those groups as a function of depth from the exposed surface converts a qualitative impression into a measured degradation profile.

What the measurement actually is

Infrared spectroscopy detects the carbonyl absorption produced by oxidation. Expressing its intensity as a ratio against a stable reference peak in the same spectrum — one belonging to a part of the polymer unaffected by degradation — normalises for sample thickness and instrument variation and yields a carbonyl index. It is a relative number rather than an absolute concentration, which is why it is only meaningful against a control.

Why the damage is a surface layer

Ultraviolet radiation is absorbed within a shallow depth of the polymer surface, so the photochemistry is concentrated there. The degraded layer itself becomes progressively more absorbing as it forms, partially shielding the material beneath. The result is a steep gradient: heavily oxidised material at the surface, essentially unaffected material a fraction of a millimetre in, and a transition between them whose depth grows slowly with exposure.

Bulk testing gets this wrong

Dissolving or grinding a whole specimen and measuring an average dilutes a severely degraded skin into a large volume of intact material, producing a number that suggests the part is fine. The same error affects mechanical testing: a bulk tensile specimen may retain most of its strength while the surface layer is brittle enough to initiate cracks that then propagate into sound material. Characterising weathering damage requires depth resolution, not an average.

How the profile is obtained

Two approaches are common. Surface-sensitive infrared measurement in attenuated total reflectance mode samples only the outermost few micrometres, and repeating it after removing successive layers builds a profile. Alternatively, a thin cross-section cut perpendicular to the exposed face can be examined along its depth by infrared imaging, giving the gradient in a single map. The second is more informative and requires more sample preparation.

Stabiliser depletion is the leading indicator

Ultraviolet absorbers and radical scavengers are consumed as they do their job, and the polymer itself degrades slowly while they remain. Once the additive package near the surface is exhausted, oxidation accelerates markedly. Measuring residual stabiliser — by extraction and chromatography, or by oxidative induction measurement — indicates where a material sits on that curve, and it can show that a part is close to the acceleration point while still looking acceptable.

Relating the profile to what it cost mechanically

Chemical measurement establishes that degradation occurred; mechanical testing establishes what it did. Impact and flexural testing are more sensitive to an embrittled skin than tensile testing, because the surface is where bending stress peaks and where impact damage initiates. Testing specimens with the weathered surface in tension, and comparing against the same specimens with the surface removed, isolates the contribution of the degraded layer.

Controls, and what serves when none exist

The ideal reference is unexposed material of the same grade, colour and lot. Where none survives, the interior of the part itself is a usable control for the surface, and an unexposed face — a sheltered underside, a region that was covered by adjacent hardware, a surface that faced away from the sun — provides a comparison that shares the entire manufacturing and service history except for exposure. Documenting which surface was used as the control matters as much as the measurement.

What the profile does not establish

A carbonyl index does not convert cleanly into elapsed time. The relationship between oxidation and exposure depends on the specific polymer, its stabiliser package, colour, temperature history and local ultraviolet intensity, and none of those are constant across a service life. The profile supports statements about the extent of degradation and comparisons between specimens; statements about precise duration require assumptions that should be stated rather than embedded.

What to preserve

The part with its exposed surface untouched — not cleaned, sanded, repainted or wiped, since the outermost layer carries the entire finding. Sheltered or unexposed portions of the same part, kept intact for use as controls. Unused material from the same lot where any exists. And the orientation and installed position recorded before removal, because which face was exposed and in what direction is not recoverable afterwards.

This article is general technical orientation, not a failure analysis, an engineering opinion, or advice on any specific matter. Determining the cause of a particular incident requires hands-on examination by a credentialed expert.