Why COC & Why Mould Temperature Matters
COC (TOPAS® 6013, Tg = 136°C) offers intrinsic UV-Vis transmittance >92% across 300–800 nm in its optimal amorphous state. However, local molecular orientation introduced by high shear rates during injection causes birefringence — refractive index anisotropy that scatters polarised light and shifts apparent absorbance readings by up to 4.1% at 450 nm. Mould temperature directly controls the rate of solidification, and hence the degree of frozen-in molecular orientation. Below 80°C mould temperature, the polymer freezes before chains relax; above 130°C, sink marks appear on 0.8 mm walls.
Design of Experiment (DoE) — Moulding Parameters
A 3-factor, 3-level Box-Behnken DoE (27 runs + 3 centre points) was constructed with factors: mould temperature (80, 100, 120°C), injection speed (40, 70, 100 mm/s), and cooling time (8, 12, 16 s). Response variables: transmittance at 450/530/620 nm (Varian Cary 60 spectrophotometer), birefringence (Hinds Instruments Exicor-150, units: nm), and mould cavity pressure peak (sensor: Kistler 6167A). Main effects analysis showed mould temperature accounted for 67% of transmittance variance; injection speed contributed 21%; cooling time interactions contributed 12%.
Optical Characterisation of Optimised Parts
At 120°C mould temperature / 70 mm/s injection speed / 12s cooling: transmittance at 450 nm = 92.3±0.4%, at 530 nm = 93.1±0.3%, at 620 nm = 93.7±0.2%. Birefringence retardation = 0.31±0.08 nm — well within the ±1 nm process specification. Haze value (ASTM D1003) = 0.42% — comparable to optical-grade glass (typically <0.5%). Comparison to off-specification parts (80°C mould): 450 nm transmittance dropped to 87.8% with birefringence rising to 3.7 nm, translating to a 4.1% false absorbance shift in the CRP immunoassay readout.
Process Capability & SPC Implementation
Statistical Process Control (SPC) was implemented using a birefringence measurement at the gate region as the primary in-process control. Capability study (n=60 consecutive production parts): Cpk = 1.48 for the transmittance@450nm parameter (USL=94%, LSL=90%). Control chart type: X-bar/R with subgroup size n=5. Western Electric rules applied. The gate-region birefringence showed strong correlation (r²=0.91) with full-part transmittance, enabling rapid non-destructive in-process rejection without spectrophotometric testing of every part.