Batch Colour Drift in POM Parts: Causal Tree and Customer Response

Polyoxymethylene (POM), designated under ISO 1043-1 as POM-H homopolymer or POM-C copolymer, is a semi-crystalline engineering resin used in moulded gears, fuel-system components, conveyor links and precision latches. Batch colour drift in POM parts is not a pigment-only defect. The causal tree branches through base-resin stabiliser consumption, barrel residence time, hot-runner stagnation, regrind ratio, contaminant depolymerisation and spectrophotometric geometry. Customer response encompasses quarantine, measurement against the PPAP master sample, mechanical retest, root-cause verification and formal disposition under ISO 9001:2015 clause 8.7. Material specifications under ISO 9988-1 define delivery form, melt mass-flow rate, tensile modulus and impact strength; colour is controlled by a customer master sample and visual boundary, not by a base-resin specification alone.

What Counts as a Significant Colour Shift in POM Mouldings?

A significant shift is defined instrumentally by CIELAB ΔE*ab measured with D65 illuminant and 10° observer under ASTM D2244 or ISO 11664-4. Customer PPAP boundaries commonly fall between 0.8 and 1.5 ΔE*ab for interior parts; visible mating surfaces may require ΔE*ab ≤ 1.0. Yellowness index drift above 2.0 under ASTM E313 is usually perceptible on natural POM. Visual assessments conducted in a light booth per ISO 3668 or ASTM D1729 can reject parts that still remain within a wide spectrophotometric tolerance because surface roughness, gloss and thickness alter colour perception. Colour drift is considered proven only when measurement geometry, sample conditioning at 23 ± 2 °C and 50 ± 10 % relative humidity under ISO 291, and specimen preparation match the approved master sample.

Production-scale moulding records from accumulator-assisted machines with clamp force between 200 kN and 8,000 kN reveal that a drift of 1.5 ΔE*ab often correlates with melt-temperature excursions rather than a faulty pigment lot. Barrel set points for POM are typically flat or reverse-profile, with rear zones at 180–200 °C, centre zones at 190–210 °C, and nozzle below 220 °C. Direct pyrometer probes may record actual melt temperatures 5–15 K above the set point when screw speed exceeds 120 min⁻¹ or back pressure exceeds 2.0 MPa. The shot volume should not exceed 70 % of barrel capacity; higher volume reduces the residence-time safety window. Injection moulding screws for POM with L/D ratios of 18:1 to 22:1 and compression ratios of 2.5:1 to 3.5:1 are preferred; a worn check ring with clearances above 0.05 mm permits melt backflow and adds heat history.

Thermal-Oxidative Degradation of the Base Resin Is the Primary Driver of Yellowing

Three interlinked mechanisms dominate. First, chain-end unzipping of POM releases formaldehyde when melt temperature exceeds 210 °C for homopolymer or 220 °C for copolymer grades, with rate increasing exponentially. Second, formaldehyde oxidises to formic acid, and the acid catalyses further depolymerisation while generating light-absorbing degradation products. Third, phenolic antioxidants and lubricants consumed during prolonged heat history can produce quinone-type chromophores. A melt mass-flow rate shift of more than 15 % from the virgin reference at 190 °C and 2.16 kg under ISO 1133-1:2022 is an early indicator of molecular-weight reduction. Odour from residual formaldehyde and surface splay on the part often accompany visible yellowing. Hot-runner systems intensify the mechanism because melt remains at elevated temperature for multiple cycles. Internally heated manifolds with blind plugs can hold a stagnant fraction for 10–30 min; the stagnant resin degrades and periodically releases as yellow streaks. Detection requires tear-down inspection of manifold channels and temperature profiling with an insertion probe under ± 5 K uniformity.

Process conflict arises because a higher screw speed improves colour dispersion but increases shear heating; a low screw speed reduces shear heating but can produce poor pigment distribution. In a 25 mm reciprocating screw with L/D 20:1, raising screw speed from 80 min⁻¹ to 140 min⁻¹ can increase melt temperature by 5–10 K. Resin suppliers recommend the lowest screw speed that still delivers a homogeneous melt, supplemented by a hot-runner manifold with ± 5 K zone uniformity. Hot-runner tips must be isolated from nozzle melt-pipe dead volume; internal corner radii below 3 mm can accumulate stagnant POM. Rheological signs of drift include a decrease in melt viscosity at constant shear rate because chain scission reduces molecular weight. Capillary viscosity at 1000 s⁻¹ for POM grades with MFR 7–12 g/10 min normally lies between 80 Pa·s and 180 Pa·s; a drop below the supplier reference range confirms heat damage.

Pigment-related drift appears when masterbatch carrier viscosity is outside the POM melt range, typically indicated by a carrier MFR difference above 50 % from the base resin. Dispersion failure under high shear can create local colour concentration around weld lines and thin ribs. Low-level contamination below 1 wt% of polyamide, PBT or PVC can darken and destabilise POM because the foreign polymer degrades at POM processing temperatures and may generate amine or halogen residues. Purge retention in screw flights or hot-runner channels is another root cause; switching between coloured POM and natural POM without a dedicated purging compound may require 10–20 barrel-capacity purges, depending on colour intensity. Outdoor UV exposure can shift colour in coloured POM after 500–1,000 h under ISO 4892-2, but published data for specific pigment packages is limited and must be verified for each grade.

If Spectrophotometric Data Conflict With Visual Appraisal

Spectrophotometric measurements can report a passing ΔE*ab while operators reject parts under a D65 light booth. The discrepancy arises from specular reflectance, surface texture and sample backing. A sphere-mode instrument includes specular reflection; a 45/0 instrument excludes it, producing differences of 0.2–1.0 ΔE*ab on textured POM surfaces. Gloss differences measured at 60° under ISO 2813 alter visual lightness even when pigmentation is constant. Colour comparison therefore requires identical surface finish, cavity polish, wall thickness and specimen stacking. Metamerism is evaluated by comparing patches under D65 and illuminant A according to ASTM D4086; a metameric pair can pass one illuminant and fail the other. Customer response should separate true pigment mismatch from gloss-driven perception before quarantining an entire batch.

Measurement drift within the customer laboratory can be caused by sample temperature, ultraviolet light-source ageing, or aperture size below 10 mm on curved gear teeth; resultant ΔE*ab may therefore exceed the agreed limit by an amount comparable to process variation. Conditioning of moulded parts at 23 ± 2 °C and 50 ± 10 % relative humidity under ISO 291 for at least 24 h eliminates thermo-chromic effects that may occur immediately after ejection. If the part is insufficiently opaque, backing with white and black standards per ISO 11664-4 can shift reflectance data. Calibration of the spectrophotometer against a calibrated ceramic tile and daily performance checks using a green tile or equivalent reference are required before any disposition decision. Published data on thermo-chromic reversal in POM colour drift is limited; retained parts should be re-measured after conditioning before a final scrap or waiver decision.

Containment, Root-Cause Verification, and Disposition Records

Upon receipt of a suspect batch, immediate containment moves the lot to quarantine and links the traceability number to the supplier batch certificate. Incoming inspection compares melt mass-flow rate, yellowness index and ΔE*ab to the approved master sample under ASTM D2244. If the batch exceeds the visual boundary, a sorting process using D65 light booths and a colourimeter with tolerance set at the customer-specific limit is implemented. Parts that pass mechanical retest may still fail colour. Mechanical retesting follows ISO 527-2 for tensile properties and ISO 179-1 for Charpy impact, with values compared to the PPAP capability range, not merely to the minimum specification. Under PPAP level 3, colour master samples, measurement data and material certificates are submitted before serial production; the same master samples become the reference for drift decisions.

Rework by re-colouring POM is constrained because additional melt processing shifts molecular weight and colour further. Reprocessing as regrind is acceptable only at a validated ratio, commonly ≤ 20 wt%, with tensile and impact tests confirming that properties remain within the PPAP range. If the supplier is permitted to regrind coloured parts into a darker production lot, the lot must be re-qualified for colour and mechanical properties. When a waiver is requested for non-compliant visual appearance, the release documentation must record the measured ΔE*ab, the photograph under D65, the illuminant data, the gloss reading and the affected part serial range. The root-cause record should reference the causal-tree branch, the corrective action assigned to tooling or process, and the verification date. A preventive action may include hot-runner temperature mapping at regular intervals, MFR screening of virgin and regrind fractions, and a spectrophotometric guardband at 80 % of the customer limit during production runs.