Bisphenol-A polycarbonate (PC) is the predominant forward-lighting lens substrate because it combines a notched Izod impact resistance commonly above 600 J/m under ASTM D256-23 with luminous transmittance greater than 88% for a 3 mm uncoated plaque. The warranty classification “Headlights Yellowed at Year Four: A Polycarbonate Coating Warranty Claim” refers to a lens that has developed a measurable yellowness shift after approximately four years of service while the structural shell remains intact. The claim is normally opened after a vehicle inspection identifies a yellowed or hazy outer lens on a lamp that still passes impact and moisture-ingress checks. In this condition the failure is not bulk hydrolytic degradation of the PC melt but a surface and near-surface photochemical transformation beneath a siloxane or UV-curable acrylate hardcoat.

Molding records are reviewed before any failure mode is assigned. Polycarbonate pellets require desiccant drying at 120 °C for 2–4 h to a moisture content below 0.02%; higher residual moisture during plastication drives hydrolytic chain scission, surface splay, and local molecular weight reduction. The melt is processed at 280–320 °C with a mold temperature of 80–100 °C. Production-scale lens tools with wall sections from 2.5 mm to 5.0 mm typically run with cavity pressures between 40 MPa and 80 MPa, and the residual stress distribution is heavily influenced by gate freeze-off time. Differential birefringence between the gate region and the outer lens edge does not cause yellowing by itself, but it can reduce resistance to environmental stress cracking later when cleaners or road chemicals penetrate a degraded hardcoat.

Uncoated PC has a refractive index of 1.585 at 589 nm and absorbs strongly below 300 nm because the aromatic carbonate repeat unit is intrinsically UV-active. The same aromaticity that contributes to impact strength also makes the polymer susceptible to photo-Fries chemistry. An uncoated PC lens is therefore not used in forward lighting without a UV-protective hardcoat. The hardcoat is required not only for abrasion resistance but also to exclude the shortest wavelengths of terrestrial UV from the substrate.

What Degradation Sequence Produces Yellowness at Year Four?

The photochemical sequence starts when ultraviolet radiation below 300 nm reaches the PC surface. In an uncoated PC surface, the carbonate linkage undergoes a photo-Fries rearrangement, generating phenyl salicylate and 2,2′-dihydroxybenzophenone moieties; subsequent oxidation creates quinonoid structures that absorb in the 400–500 nm band and appear yellow. In an intact hardcoat, hydroxyphenylbenzotriazole or hydroxyphenyltriazine UV absorbers dissipate absorbed UV energy through excited-state intramolecular proton transfer. The absorber must remain molecularly dispersed in the coating matrix; aggregation above the solubility limit lowers visible transmittance and creates haze. The loss of protection at year four usually begins as siloxane bond scission, microcracking, and UV-absorber depletion, not as abrupt substrate failure.

Siloxane hardcoats of the sol-gel type are typically built from methyltrimethoxysilane or colloidal silica networks. They provide UV transparency and abrasion resistance but are relatively brittle. UV-curable acrylate hardcoats offer faster cure and high crosslink density but can generate higher thermal-expansion mismatch with the PC substrate. The failure pattern therefore differs: siloxane coatings often proceed by microcracking and gradual erosion, while UV-curable acrylics may exhibit delamination or localized hazing after thermal cycling. A year-four yellow lens frequently shows regional gradients rather than a uniform color shift; the upper lens edge and the high-beam focal region are common early sites because thermal load accelerates oxidative reactions and coating erosion.

Yellowing at year four is generally not accompanied by a large drop in room-temperature notched Izod impact strength. UV penetration is limited to the outermost surface, so the backing PC retains its bulk toughness. The warranty file must therefore separate surface optical degradation from structural failure. A lens with a yellowness index shift of several units may still pass a stone-impact test but may no longer meet the photometric output criteria applied to the lamp under FMVSS 108 or ECE R112 if luminous transmittance falls below the applicable beam-pattern threshold.

Warranty decisions require measured optical data rather than field photographs or verbal color descriptions. The returned lens is cleaned with a pH-neutral detergent, rinsed with demineralized water, and conditioned at 23 °C ± 2 °C and 50% ± 5% relative humidity for 40 h according to ISO 291:2008. Haze and luminous transmittance are obtained with an integrating-sphere haze meter under ASTM D1003-21; yellowness index is calculated under ASTM E313-20 from CIE tristimulus values. The measurement grid avoids injection-molded flutes, gate dimples, and the lamp-axis region because thickness variation changes optical path length and can produce a false yellowness gradient across the lens.

The vehicle-level inspection record often states that the lens is yellowed or cloudy, but this description is insufficient for adjudication. The lens is removed from the housing using the OEM procedure to avoid crack propagation at the lens-to-housing seal. The gaskets, bezel, and reflector are separated, and the lens is placed on a non-contact optical fixture. If the lens has residual adhesive, it is cleaned without aromatic solvents. A first quantitative gate is whether luminous transmittance remains above the lamp’s original requirement; a second gate is the yellowness index shift; a third gate is cross-cut adhesion. Only after all three are measured can the claim be assigned to UV degradation, coating failure, or processing contamination.

No universal pass/fail value exists for yellowness index in automotive lens warranties. Some supply agreements express acceptance as a change in yellowness index of less than 2 after 3000 h of xenon arc testing under ISO 4892-2:2013, but this is a contractual limit and not an ISO requirement. The role of ISO 4892-2:2013 or ASTM G154-23 is to screen coatings against a control; neither standard states a field-service age. Accelerated weathering cannot be converted directly into a year-four warranty period because moisture, temperature cycling, and road chemistry vary by region. Consequently, a fourth-year claim cannot be closed without the original homologation limit. The strongest warranty evidence is a comparison between the returned lens and an unexposed reference lens from the same mold cavity, production week, and coating batch, measured on the same instrument configuration.

When Coating Adhesion Loss Precedes Yellowness, the Failure Mode Shifts from UV Degradation to Process Contamination

If cross-cut adhesion under ISO 2409:2020 returns a classification of 3 or worse while the yellowness index remains inside the OEM limit, the claim is more likely a coating application defect than a UV-driven substrate failure. The lens manufacturing chain must then be audited for mold-release residues, pre-treatment quality, and hardcoat cure. Silicone-containing mold releases and stearate-based contamination can reduce the PC surface energy below 38 mN/m, preventing aqueous or solvent-borne hardcoat wetting. Atmospheric plasma or air corona treatment is used before flow coating to restore surface energy, but the effect decays with time. If the production log does not record the treated surface energy or water contact angle immediately before coating, a contamination root cause cannot be excluded.

Hardcoat thickness is a critical warranty variable. Flow-coating and dip-coating lines can produce dry film thickness variation from 4 µm at the lens edge to 15 µm at the center depending on viscosity, withdrawal speed, and drainage geometry. Areas that are too thin lose UV protection early. Areas that are too thick may microcrack during thermal cycling because the siloxane network and the PC substrate have different coefficients of linear thermal expansion. A thermal cycle from -40 °C to 85 °C can generate interfacial stress at the hardcoat–PC boundary, and repeated cycling contributes to microcracking before the bulk coating has developed visible yellowing. This is the process conflict that dominates the four-year warranty boundary: sufficient thickness for UV screening versus excessive thickness for thermal-cycle durability.

Production-scale hardcoat lines typically maintain filtered air at 23 °C ± 2 °C and relative humidity below 50% during flow coating. The part is held in a rotating fixture to control drainage thickness. Siloxane hardcoat cure is commonly run at 120–130 °C for 30–60 min; higher temperatures or longer residence can deform the PC substrate. The coating line log must include bath solids, solvent blend, cure oven temperature, and the air-dwell time between coating and thermal cure. In warranty analysis, a missing cure log or a solvent blend containing excess amine can explain adhesion loss even when the lens had acceptable initial cross-cut results.

Polycarbonate grade selection and molding parameters are reviewed only after coating integrity has been mapped. A grade with a melt mass-flow rate of 20–30 g/10 min under ISO 1133-1:2022 may be selected for thin-wall lens fill, but higher flow grades tend to retain more molecular orientation and may be more sensitive to stress cracking when the uncoated inner lens surface contacts alcohols, ketones, esters, or amine-containing cleaners. Unsupported PC should not be wiped with aromatic hydrocarbon solvents; environmental stress cracking can occur at external strain levels below the short-term yield point. Alkaline cleaners above pH 10 are also incompatible with molded PC lenses because surface hydrolysis can initiate hazing and crack propagation under residual stress.

The hardcoat must comply with the applicable chemical restrictions in REACH and RoHS 2011/65/EU, and the coating line must maintain records of solvent blend, bath solids, and cure-oven temperature. In a year-four yellowed lens, recoating with a clear topcoat is not generally homologated because the original UV absorber in the outermost PC surface is depleted and because mechanical removal of the failed hardcoat can introduce surface damage. Published long-term performance data for sanded and recoated polycarbonate automotive lenses under SAE J576 is limited. The claim file should therefore contain the dried-lot molding record, the coating adhesion result, the measured optical data, and the original homologation weathering limit; without all of these elements, the assignment of warrantable cause remains open.