Polycarbonate (PC) selected for medical device housings, luers, stopcocks, blood oxygenator housings, drug-delivery pump components, and cartridge reservoirs is distinguished less by a separate chemical backbone than by a controlled resin formulation with reduced metallic residues, low-extractable mould release, documented batch traceability, and a change-control process aligned with ISO 13485 and ISO 14971. The polymer is an amorphous bisphenol A polycarbonate with a glass transition temperature commonly reported near 147 °C and a flexural modulus near 2.4 GPa. Typical injection-moulded values for tensile yield strength range from 60 MPa to 70 MPa, and notched Izod impact ranges from 600 J/m to 850 J/m under ASTM D256. These values are grade- and specimen-dependent and should not be transferred directly from resin datasheets to a finished part without accounting for molecular weight, orientation, moisture content, and moulded-in stress. The material must be dried to a moisture content below 0.02% before melt processing because the carbonate linkage hydrolyses above melt temperature, producing carbon dioxide and reducing molecular weight; visible splay, silver streaks, and trapped gas porosity are the common production-floor indicators of inadequate drying.

What Are the ISO 10993 Endpoints for Polycarbonate in Fluid-Contact Devices?

ISO 10993-1:2018 requires biological evaluation to be driven by the nature, duration, and type of body contact, with the final endpoint matrix established in the risk management file. For polycarbonate components in contact with blood, parenteral fluids, or respiratory gases for prolonged or permanent periods, the frequently applied endpoint set includes cytotoxicity per ISO 10993-5, sensitisation and irritation per ISO 10993-10, acute systemic toxicity and material-mediated pyrogenicity per ISO 10993-11, and haemocompatibility per ISO 10993-4 for blood-contacting applications. Cytotoxicity testing of PC is often conducted by the MEM elution or agar diffusion methods using L-929 mouse fibroblast cells, with identical extraction polarity options also reflected in USP <87>. Genotoxicity per ISO 10993-3, subchronic systemic toxicity per ISO 10993-11, and implantation per ISO 10993-6 become relevant when the device is a long-term implant or when leachables have not been excluded by prior chemical characterisation. ISO 10993-18 chemical characterisation is used to identify extractables such as bisphenol A, residual monomers, additives, and processing aids. The extraction vehicles typically include polar and nonpolar media because polycarbonate releases oligomeric and additive components differently in aqueous and lipid environments. The presence of bisphenol A in hydrolytic extracts is a known consequence of carbonate linkage cleavage and must be evaluated as part of the allowable leachable profile rather than assumed absent.

For qualification against USP Class VI, the polycarbonate resin or processed article is extracted at 121 °C using saline and vegetable oil media, with the resulting extracts administered by systemic injection, intracutaneous injection, and implantation in rabbits under USP <88>. USP <87> covers in vitro cytotoxicity, while USP Class VI provides an in vivo qualification of systemic and local tissue responses. A USP Class VI certificate is often accepted as supporting evidence in a regulatory submission, but it does not replace the full ISO 10993 evaluation, nor does it address haemocompatibility, genotoxicity, or chemical characterisation. USP <661> and related packaging standards may be used for physicochemical testing of plastic containers, but they are not biological test replacements. When material certifications are reviewed, the following endpoint matrix is used to align data with the relevant standardised test methods.

Gamma and Electron-Beam Sterilisation: Dose-Dependent Property Consequences

Radiation sterilisation of polycarbonate is typically controlled under ISO 11137 with absorbed doses of 25 kGy for routine terminal sterilisation and higher doses up to 40 kGy or 50 kGy for parts requiring lower bioburden tolerance or terminal sterilisation in difficult packaging geometries. The aromatic carbonate backbone does not crosslink efficiently under ionising radiation; instead, radical formation leads to chain scission and the formation of coloured oxidation products. At 25 kGy, a transparent polycarbonate part may show a measurable yellow shift; at 40 kGy to 50 kGy, the visual colour change is generally severe enough to make optically transparent polycarbonate unsuitable unless a radiation-stabilised formulation or post-irradiation colour correction is accepted. The exact yellowness index shift is grade-dependent because stabiliser packages, processing history, and part thickness affect the dose response. Published data for specific medical-grade formulations is limited, so dose-response testing on finished moulded parts is necessary. Electron beam at equivalent doses produces a similar free-radical population but different dose-rate and penetration characteristics. High-voltage electron beam from 1 MeV to 10 MeV is appropriate only after dose mapping confirms the required dose at target depth; for thick or complex PC housings, gamma is often selected because higher penetration and lower dose rate permit more uniform dose delivery across density variations. Mechanical effects after radiation are usually more evident in tensile elongation and notched Izod impact than in yield stress, and injection-moulded parts with moulded-in stress do not behave identically to annealed resin plaques.

Autoclaving polycarbonate components exposes the material to saturated steam at 121 °C for 15 min or 134 °C for 3 min under ISO 17665; both conditions load the polymer with hot water and thermal energy below the glass transition. Hydrolysis of the carbonate bond proceeds through attack by water at the carbonyl carbon, producing bisphenol A and carbonic acid, the latter decomposing to carbon dioxide and water. Each autoclave cycle can reduce molecular weight at the surface and in the bulk, and the practical failure sequence is usually a ductile-to-brittle transition followed by craze formation at gates, weld lines, threads, and snap-fit features where tensile stress is highest. Polycarbonate components with high moulded-in stress, sharp internal corners, or metal inserts are more likely to show environmental stress cracking during steam sterilisation. A neutral pH steam cycle is less aggressive than alkaline or amine-containing cleaning agents because hydrolysis is base-catalysed; therefore any cleaning agent used before autoclaving must be rinsed completely and its compatibility validated. There is no universal maximum steam-cycle count for polycarbonate across all grades and part geometries. Manufacturers must validate the specific part shape and process using visual inspection, dye-penetrant crack detection, molecular weight by size exclusion chromatography, and impact testing after the maximum number of cycles claimed in the device instructions for use.

If Ethylene Oxide Is Selected, Residual Limits and Process Monitoring Dominate

Ethylene oxide processing under ISO 11135 uses a chamber temperature commonly between 37 °C and 55 °C, with relative humidity from 40% to 80% and gas concentrations typically in the 400 mg/L to 800 mg/L range. Polycarbonate generally retains its mechanical properties under these conditions, but ethylene oxide is absorbed into the amorphous polymer and must be removed by aeration. ISO 10993-7 residual limits for ethylene oxide and ethylene chlorohydrin are tied to contact duration and device classification; the limits are lower for prolonged or permanent contact and must be corrected for patient mass for certain device categories. Polycarbonate components with hollow cavities, tortuous lumens, or thick walls may require extended aeration times at 40 °C to 50 °C. Aeration must be validated to achieve documented residual limits in finished packages because polycarbonate adsorbs ethylene oxide and slowly releases it during storage. When ethylene oxide is used with water vapour, the transient moisture uptake can cause slight dimensional change and may interact with internal stress; residual moisture should be removed before final packaging to prevent hydrolysis during storage.

Hydrogen peroxide gas plasma and vapourised hydrogen peroxide processes operate below 60 °C and are specified for moisture- and heat-sensitive devices through methods such as ISO 14937. The lower thermal exposure reduces the hydrolytic challenge seen in steam, but the oxidative environment can alter polycarbonate surfaces and may react with organic residues or colorants. Low-temperature sterilisation does not produce the deep gamma-driven colour body formation seen at 40 kGy, but cycle compatibility must still be confirmed because vacuum levels, chamber humidity, and hydrogen peroxide concentration vary between equipment generations. Polycarbonate parts with thin walls, hydrophobic coatings, or bonded joints may show surface cracking if residual stresses are high, because the combination of reduced pressure and chemical exposure acts as an environmental stress-cracking condition. The practical decision between ethylene oxide and hydrogen peroxide often depends on the device’s thermal budget, packaging permeability, and ability to deliver the sterilant into lumens. No single low-temperature option is universally compatible with all PC designs.

Processing Windows and Molecular-Weight Retention During Melt Conversion

Drying and melt conversion influence ISO 10993 and USP Class VI outcomes because degradation products generated during processing become extractables in the final device. Injection moulding of medical polycarbonate typically operates with a melt temperature of 280 °C to 320 °C and a mould temperature of 80 °C to 110 °C. The screw should provide gentle compression, with an L/D ratio in the 20:1 to 22:1 range and a compression ratio near 2.5:1; residence times above 10 min and excessively hot barrel zones reduce molecular weight and raise bisphenol A levels in subsequent extraction tests. A desiccant dryer with a dew point of -40 °C or lower should reduce pellet moisture to 0.02% or less before melting. Clamp force, injection velocity, and packing pressure must be sized to avoid overpacking sharp corners and weld lines, because these regions become preferential sites for steam- or radiation-induced failure. Moulded-in stress can be assessed through birefringence or solvent-based immersion stress checks, with higher mould temperatures generally reducing frozen-in orientation but increasing cycle time, especially for parts with wall thickness below 1 mm or above 4 mm.

Polycarbonate is not compatible with strongly alkaline cleaners, amines, or ketones; these agents catalyse hydrolysis or induce environmental stress cracking. For drug-delivery applications, lipids and solvent-based formulations can reduce the stress-crack resistance of PC under load, so compatibility with the actual drug product should be evaluated under ISO 10993-18 and not inferred from standard resin datasheets. Reusable medical devices that require daily cleaning and repeated steam sterilisation may exceed the practical robustness of polycarbonate if cracks develop before the intended service life; polysulfone or polyetheretherketone alternatives are typically evaluated when repeated high-temperature autoclave cycles and strong acid or base exposure are unavoidable.