Optical polycarbonate (PC) is an amorphous bisphenol A-based engineering thermoplastic, produced commercially by interfacial polycondensation or carbonate melt transesterification. Since the resin solidifies without spherulitic crystallites, unfilled grades exhibit no large-angle bulk scattering in the visible spectrum. Specification of optical-grade PC, however, is not a single measurement; it requires simultaneous control of total luminous transmittance, haze, and refractive index across specific thicknesses and thermal histories. For clear, uncoated injection-moulded plaques of 3.2 mm thickness, total luminous transmittance under CIE illuminant D65 and 2° observer geometry is typically between 89% and 91% when tested to ASTM D1003-21. The same plaque tested in accordance with ISO 13468-1:2019 may be reported as internal transmittance after surface reflection corrections. Haze values for optical grades usually remain below <1.0% and, for low-particulate specialty grades, below <0.5% under ISO 14782:2021. Refractive index at the sodium D line of 589.3 nm is specified within 1.584 and 1.586 by ISO 489:2022 or ASTM D542-14 at 23 °C. These values are material grade-dependent and should not be read without specimen thickness, surface condition, and preconditioning history.

What limits the upper transmission ceiling in unfilled PC windows?

The external transmittance of a transparent polymer slab is constrained by Fresnel reflection at the two air–polymer interfaces. For a PC refractive index of 1.586, the normal-incidence reflectance of a single interface is calculated as R = ((n−1)/(n+1))^2 = ((1.586−1)/(1.586+1))^2 ≈ 0.051. For a parallel-sided plaque with negligible absorption, the two-surface transmittance limit is T = (1−R)²/(1−R²) ≈ 0.903, or 90.3%. Measured total luminous transmittance between 89% and 91% on clear grades therefore indicates that chromophore absorption and internal scattering are small but non-zero in the visible band. Increasing path length depresses transmittance: a 6.0 mm clear sheet generally falls to 86%–88%, whereas a 1.0 mm film can exceed 90%. Because Fresnel loss is wavelength-dependent through refractive-index dispersion, transmittance curves across 400–700 nm are not perfectly flat; the blue end of the spectrum typically shows slightly higher reflection loss. Compliance with ASTM D1003-21 or ISO 13468-1:2019 therefore requires an integrated luminous value, not a single-wavelength transmission measurement.

Across production-scale injection-moulding operations, haze is not simply an intrinsic grade property; the measured scattered-light fraction is highly sensitive to melt moisture, screw work history, mould surface polish, and additive dispersion. Haze quantifies the percentage of transmitted light that deviates more than 2.5° from the incident beam. In a well-dried optical PC, bulk haze arises from particulates, gel bodies, incompatible additive domains, and microvoids generated by hydrolytic degradation. Melt-processing conditions are therefore part of the optical specification. Residual moisture must be below 0.02% before plastication; if ambient relative humidity exceeds 60%, unfilled optical PC is typically dried in a desiccant dryer at 120 °C for 3–4 h to 0.02% moisture or less. Moisture above this threshold hydrolyses carbonate linkages, releasing CO₂ microbubbles that scatter light and raise haze. On an electric injection-moulding machine with a 24:1 L/D constant-taper screw and shut-off nozzle, barrel temperatures from 280 °C to 320 °C and mould temperatures from 80 °C to 120 °C are typical for optical plaques. Melt residence times or temperatures above 340 °C accelerate chain scission and yellowing, which lowers luminous transmittance before haze fails.

The mould side that contacts the optical face should be polished to SPI A1 or an equivalent average roughness below 0.02 µm. Visible micro-scratches from release tooling, weld lines, and gate blush are common process-dependent haze sources that are not captured if a polished cast sheet is used for material certification. When high-purity optical grades are moulded under cleanroom conditions, measured haze at 3.2 mm is often below 0.5%; the same grade moulded on a contaminated line with vent blockage or insufficient purge can exceed 1.5%. Incoming resin inspection under ISO 14782:2021 should therefore be paired with a moulded-plaque control sample, not a raw-pellet measurement.

Refractive index specification through Abbe refractometry and chromatic dispersion

Refractive index of optical PC is conventionally reported as n_D, measured at 589.3 nm on a calibrated Abbe refractometer using a suitable contact liquid. ISO 489:2022 and ASTM D542-14 require flat, stress-free specimens conditioned at 23 °C and 50% RH for at least 40 h unless shorter conditioning is specified by the material standard. General-purpose clear PC exhibits n_D between 1.584 and 1.586; special copolycarbonate and high-flow optical grades may shift this value by a few thousandths. The refractive index is not a scalar design constant. It varies with wavelength through chromatic dispersion, with temperature through the thermo-optic coefficient, and with molecular orientation through stress-induced birefringence. The Abbe number V_D is approximately 29.9 to 30.3, which is low relative to PMMA and crown glass. This high dispersion means that blue wavelengths refract more strongly than red wavelengths. Using n_D = 1.586 and V_D = 30, the approximate indices at the hydrogen F and C lines are n_F ≈ 1.596 at 486.1 nm and n_C ≈ 1.576 at 656.3 nm. Optical designers using PC lenses or light guides must incorporate this wavelength spread into curvature calculations and colour-correction workflows.

The thermo-optic coefficient of PC is commonly quoted near −1.1 × 10⁻⁴ K⁻¹; a 10 °C increase in service temperature reduces n_D by approximately 0.0011. In injection-moulded components, the larger optical disturbance is usually birefringence. Polycarbonate has a high stress optical coefficient, approximately 68 × 10⁻¹² Pa⁻¹ to 72 × 10⁻¹² Pa⁻¹. Flow-induced orientation and differential cooling create residual stresses that produce optical retardation. For display windows and polarised-light applications, an optical-grade PC may additionally specify maximum optical path difference or retardation, but the test method must be explicitly defined, because published data for this specific configuration is limited. Annealing at 120 °C for 2 h can reduce residual stress in thin plaques, but it may also cause dimensional changes if fixturing is inadequate.

When anti-reflective (AR) coatings are specified for display covers, instrument lenses, or automotive HUD combiner surfaces, the uncoated PC transmission ceiling near 90.3% can be raised substantially. A single-layer durable AR coating cannot perfectly index-match air to PC because the ideal single-layer index n_coating = √(n_air · n_PC) ≈ 1.26; practical materials are either too soft or too porous. Commercial AR stacks therefore use alternating high- and low-index layers, such as ZrO₂ and SiO₂, applied by vacuum sputtering or plasma-assisted evaporation. A four-layer AR stack on PC can reduce single-surface reflectance from approximately 5.1% to below 1.0% over the 420–680 nm band, increasing total luminous transmittance of a 3.2 mm plaque to above 95%. Required adhesion and environmental durability are commonly assessed with ISO 2409 cross-cut adhesion, ISO 9211-3 environmental tests, or OEM abrasion specifications. Post-coating haze values below 0.2% are achievable only when the coating chamber is free of particulates and the substrate surface is cleaned to remove volatile residues and release agents.

Wet-applied silicone hardcoats of 3–8 µm thickness provide abrasion resistance and can preserve optical clarity, but non-uniform flow coating or dust-laden ovens can introduce haze above 0.5%. On three-dimensional parts, dip coating may create edge beads that act as light traps. In-process inspection under ASTM D1003-21 should be performed on flat witness plaques, not on curved parts, because the standard assumes a plano-parallel sample.

When laser welding or edge-lit light guides impose colour and scattering constraints

Transmission and haze specifications cannot be separated from the intended beam path. In edge-lit PC light guides for display backlights, the refractive-index contrast between the guide and any cladding or air interface controls total internal reflection. With n_D of 1.586, the critical angle relative to air is θ_c = arcsin(1/1.586) ≈ 39.1°. Light guide designs with PC therefore accept a smaller angular cone than PMMA guides with n_D of 1.49. This is an advantage for out-coupling but requires tighter control of surface roughness and bulk haze because scattered rays exceed the critical angle and leak from the guide. For laser-welded optical assemblies, near-infrared absorbing layers or pigments may be incorporated at the weld interface. The visible-light transmittance penalty is determined by absorber concentration and wavelength; low-loading NIR absorbers may reduce transmittance by only a few percentage points, but quantification requires spectrophotometric data across the full 380–780 nm range. Published data for specific NIR absorber grades in optical PC is limited; suppliers should provide measured spectral transmittance from 400 nm to 1000 nm rather than a single luminous transmittance value.

In polycarbonate visors, ophthalmic lenses, and sensor windows, final optical performance is therefore specified as a set of boundary conditions rather than a single transmittance figure. A clear unfilled optical PC may be accepted when a 3.2 mm plaque shows total luminous transmittance of at least 89%, haze below 0.5%, and n_D between 1.584 and 1.586 under the stated standards. The same resin will fail equivalent display-cover requirements if moulded with insufficient drying, flame-retardant packages, glass fibre, poorly maintained mould surfaces, or excessive shear history. No post-mould inspection of haze and transmission can repair that process damage; optical PC quality must be preserved through the entire melt-processing and coating chain.