Choosing Polycarbonate for an Optical Part: Abbe Number, Stress and Coating

Optical polycarbonate is selected when a molded transparent component must combine impact resistance with a visible-range luminous transmittance above 89% at 3.0 mm thickness when measured according to ISO 13468-2. The material is amorphous, exhibits a glass transition temperature near 147 °C, and develops flow-dependent residual birefringence that cannot be removed by post-molding polishing unless the part is annealed. Specifying polycarbonate for an optical part therefore requires simultaneous evaluation of three interacting parameters: the Abbe number of the resin, the stress field induced by injection molding, and the adhesion of subsequent siloxane hardcoats or anti-reflective oxide stacks.

Optical grades of bisphenol-A polycarbonate are published with a refractive index n_D of 1.585 and an Abbe number V_d of 30.0 when tested under ASTM D542. The corresponding n_F − n_C difference is 0.0195. This places polycarbonate among higher-dispersion optical polymers. Unlike poly(methyl methacrylate), which has an Abbe number near 57.2, a polycarbonate singlet introduces measurable axial chromatic focal shift that must be accommodated in the optical design. Because the unstressed solid is optically isotropic, any birefringence observed in a molded polycarbonate window is a direct indicator of molecular orientation or frozen-in thermal stress. The stress-optic coefficient is commonly reported in the range 68–75 Brewsters, where 1 Brewster equals 10⁻¹² m²/N. This birefringence is not a surface defect: it is a bulk wavefront-error source that must be measured after molding and again after hardcoating.

What Does an Abbe Number of 30.0 Mean for Chromatic Focal Shift?

The Abbe number is computed as V_d = (n_d − 1) / (n_F − n_C) using refractive indices measured at 589.3 nm, 486.1 nm, and 656.3 nm. For a polycarbonate grade with n_D = 1.585 and V_d = 30.0, the F-to-C refractive index difference is 0.0195. Under the thin-lens approximation, the axial chromatic focal shift between F and C wavelengths is approximately f / V_d. A 20.0 mm focal-length polycarbonate singlet therefore has an approximate shift of 0.67 mm. The same nominal focal length in PMMA with V_d = 57.2 gives 0.35 mm.

This dispersion penalty does not imply poor visible-range clarity. A 3.0 mm molded specimen of optical-grade polycarbonate can exhibit total luminous transmittance of 89–91% and haze below 1.0% under ASTM D1003. The lower Abbe number affects color correction, not gross transmission. Polycarbonate is therefore less suitable than cyclic olefin copolymer or PMMA for broadband imaging without a compensating flint element or diffractive surface. In illumination optics with narrowband sources, the Abbe number exerts less influence; in polychromatic sensor windows, spectral transmission is specified separately from chromatic focal shift.

The thermo-optic coefficient dn/dT of polycarbonate is approximately −1.07×10⁻⁴ K⁻¹. Over a service temperature range from 20 °C to 80 °C, the refractive index decreases by roughly 0.0064. Under the thin-lens approximation, a 20.0 mm focal-length polycarbonate lens shifts focus by approximately 0.22 mm over this temperature range due to dn/dT alone; thermal expansion adds a smaller effect of opposite sign. This thermal focal shift can exceed chromatic focal shift in a compact imaging path without active refocus. Published ray-trace data for molded polycarbonate lenses thinner than 2.0 mm with combined thermal and chromatic shift is limited, so as-built verification remains necessary.

Material substitution from PMMA to polycarbonate requires re-optimization of the optical prescription because the refractive-index change from 1.491 to 1.585 alters spherical aberration balance and Petzval contribution under identical surface geometries. The lower Abbe number of polycarbonate is accepted when impact strength, service temperature, or structural integration outweigh color correction requirements. In such cases, the optical layout uses a flint-type polymer or inorganic element to restore chromatic balance.

Under melt-processing conditions, the optical isotropy of polycarbonate is disrupted by two mechanisms: shear-induced orientation during mold filling and differential thermal contraction during cooling. A melt temperature of 290–320 °C, a mold temperature of 90–110 °C, and a packing pressure of 60–80 MPa are common for optical parts with 2.0–4.0 mm wall thickness. If the mold temperature falls below approximately 80 °C, the solidification front advances before oriented chains relax, and frozen-in birefringence near the gate rises. In production-scale injection molding, gate-adjacent retardation bands can remain above 20 nm/cm even when the bulk part passes 10 nm/cm under a white-light polariscope equipped with a Sénarmont compensator.

The stress-optic relationship for polycarbonate is Δn = C(σ₁ − σ₂), where C is the stress-optic coefficient in Brewsters. At a residual principal stress difference of 1 MPa, the induced birefringence is 6.8×10⁻⁵ to 7.5×10⁻⁵. In a 3.0 mm thick plano window, this creates an optical path difference of 204–225 nm. A polariscope reading of one fringe at 546.1 nm in the same thickness corresponds to a birefringence of 1.82×10⁻⁴ and a principal stress difference of approximately 2.6 MPa when C = 70 Brewsters. These values convert a pass/fail birefringence map into a mechanical stress map and are used to identify gate regions where coating adhesion loss is most probable.

Annealing at 120 °C for 2–4 h lowers residual birefringence but may alter the vertex radius of curved surfaces and is therefore applied only before final machining or on fixtures that constrain free edges. The relevant measurement standard is ASTM D4093, which provides the photoelastic procedure but does not set optical acceptance limits; those limits must be derived from the end-use wavefront-error budget. For imaging covers, a maximum retardation of 20 nm/cm is often cited; for sensor windows, 50 nm/cm may be acceptable when the photometric aperture averages the central field rather than edge regions.

Hot-runner valve-gate systems are specified for optical-grade polycarbonate because they remove cold-sprue machining and preserve laminar melt fronts. The gate vestige height on the molded surface should remain below 0.05 mm before hardcoating to prevent local coating-thickness collapse. In multi-cavity production, cavity-to-cavity fill imbalance from valve-gate seating differences can alter local packing pressure and increase birefringence in the cavity that fills first. Cavity-pressure transducers are therefore used to switch from velocity control to pressure control at 95–98% of full cavity volume, stabilizing the melt cushion and reducing batch-to-batch variation in optical retardation.

When Hardcoating Is Required Before Anti-Reflective Deposition

Polycarbonate surfaces under ASTM D3363 pencil hardness generally fall below H, and uncoated parts develop haze rapidly in wiping or cleaning. A siloxane hardcoat is therefore applied to raise surface hardness to 2H–3H and to provide a base for anti-reflective oxide layers. The hardcoat refractive index, commonly 1.47–1.50, must be entered into the optical design because the coating contributes a weak interference layer and changes the reflected phase at the polymer surface.

A polysiloxane hardcoat is typically applied at 3–8 μm thickness and cured at 120–130 °C for 30–60 min. Adhesion to polycarbonate depends on surface activation and primer chemistry. Air plasma or atmospheric corona treatment is used to oxidize the surface; a water contact angle below 40° is a common wetting target before primer application. Over-activated surfaces reorient within hours, so the hardcoat or primer must be deposited in the same production shift. Solvent-borne primers containing aromatic ketones or esters can attack polycarbonate, especially in gate regions where residual stress is highest. Acrylic primers with low solvent aggressiveness or organically modified silane primers are preferred.

The cured coating is qualified by cross-cut adhesion per ISO 2409:2013. Flat molded polycarbonate substrates with a properly oxidized surface and a siloxane hardcoat typically achieve Class 0; edges and gate vestiges may be allowed Class 1 when the optical path is not sampled there. Without a hardcoat, polycarbonate falls short of abrasion requirements in automotive and sensor cover applications: uncoated PC can show a haze increase of 20–30 percentage points after 100 cycles of Taber CS-10F wear at 500 g under ASTM D1044. A hardcoated flat sample is often specified at ≤5 percentage points haze increase under the same test. Curved surfaces and edges generally show greater haze because coating thickness decreases on draft angles and corners.

Polysiloxane Hardcoat Adhesion and Taber Haze Benchmarks

The following matrix is used in coating qualification for polycarbonate optical parts. The listed pass/fail values are representative production targets; the end-use specification controls actual limits.

Environmental cycling under IEC 60068-2-14 from −40 °C to +85 °C for 500 cycles is routinely applied to coated polycarbonate assemblies. The linear coefficient of thermal expansion of polycarbonate is approximately 65×10⁻⁶ K⁻¹, while a siloxane hardcoat is an inorganic-organic network with a lower expansion coefficient. Hardcoat thickness above 8 μm or sharply stepped thickness at gate vestiges can develop microcracks after repeated thermal excursions. For this reason, gate-vestige height is controlled below 0.05 mm before coating, and parts with vestige protrusion above this threshold are segregated before plasma treatment.

Moisture pickup before molding must also be controlled; polycarbonate pellets are dried at 120 °C for 4 h in a desiccant dryer with a dew point of −40 °C to a moisture content below 0.02%. Drying above 125 °C can yellow the resin and shift the short-wave transmission edge of optical grades. Incoming resin should be qualified by measuring melt volume-flow rate under ISO 1133-1:2022 at 300 °C with 1.2 kg load, because a significant molecular weight shift changes both melt viscosity and stress relaxation. Water absorption of polycarbonate is 0.15% after 24 h under ISO 62, and saturated moisture uptake can modify local refractive index and coating adhesion at the surface over long service life.

Cleaning and chemical exposure impose additional constraints. Ketones, aromatic hydrocarbons, and some amines craze stressed polycarbonate; isopropyl alcohol at moderate concentration is less aggressive but does not remove all mold-release residues. The combination of residual stress from molding and an incompatible cleaning solvent can produce microcrazes that appear as local birefringence spikes and act as adhesion-failure nucleation sites under subsequent thermal cycling. A solvent compatibility test should therefore be performed on molded polycarbonate parts, not on unstressed plaques, using the actual coating stack and the actual gate-derived stress field.