Polymer Selection for Ethylene Oxide Terminal Sterilisation of Heat Labile Devices

Ethylene oxide terminal sterilisation of heat-labile medical devices places polymer selection at the intersection of gas diffusion, residue formation, and post-cycle mechanical stability. Devices constructed from plasticised poly(vinyl chloride), aromatic polyether or polyester thermoplastic polyurethane, low-density polyethylene, ethylene-vinyl acetate copolymers, silicone elastomer, polycarbonate, and copolyester are routinely processed at chamber temperatures between 45 °C and 55 °C because autoclave temperatures exceed the heat deflection temperature of the polymer or drug payload stability limits. The process validation requirements in ISO 11135:2014 address microbiological kill, chamber humidity control, EO concentration delivery, and aeration, but polymer-specific acceptance requires residue compliance with ISO 10993-7:2008/Amd 1:2012 and evaluation of mechanical and physical changes under ASTM D638-14, ISO 1133-1:2022, ASTM D570-22, and ISO 62:2008. Selection is therefore not restricted to thermochemistry alone; the polymer must permit reproducible EO penetration into lumen interiors, resist excessive sorption that prolongs aeration, avoid generation of ethylene chlorohydrin in chloride-bearing formulations, and retain post-cycle tensile integrity after terminal packaging. Preconditioning at 50 ± 5 °C and 45–75 % RH hydrates amorphous regions in polar polymers; in polyethylene the same humidity step has limited effect because water absorption by ISO 62:2008 is below 0.01 % for many grades. The practical consequence appears on extrusion lines producing multi-lumen catheter shafts: if a hydrophobic polyolefin outer layer is coextruded over a hydrophilic polyurethane inner layer, residual moisture differences before EO exposure create non-uniform EO uptake across the wall. Operators on lines with single-screw extruders having 24:1 to 30:1 L/D ratios and screw speeds between 20 rpm and 80 rpm for thermoplastic polyurethane observe that humidity soak time must be extended by 2 h to 6 h when the same package contains both nonpolar and polar materials; otherwise aeration data after the cycle show higher EO residuals in the polar layer. This is not a polymer rejection criterion but a validation constraint requiring vacuum pulsing or extended preconditioning dwell.

How Does Polymer Morphology and Ethylene Oxide Sorption Interact During Cycle Validation?

The sorption of ethylene oxide in a polymer is governed by free-volume distribution, fractional crystallinity, solubility parameter match, and plasticizer or filler content. In semicrystalline polyethylene and polypropylene, diffusion through lamellar crystals is negligible; EO transport occurs predominantly through amorphous interlamellar and tie-chain regions. Differential scanning calorimetry per ISO 11357-3:2018 can quantify crystalline fraction, which correlates inversely with available sorption volume. For polypropylene homopolymer with typical crystallinity between 45 % and 60 % by DSC, aeration times are short relative to those of flexible PVC, but molded-in orientation in thin-wall connectors may relax when the chamber reaches 55 °C, producing dimensional variation that can be measured after conditioning under ISO 291:2008. Amorphous polar polymers such as rigid PVC and thermoplastic polyurethane exhibit larger sorption capacity at the cycle temperature because their glass transition or soft-segment glass transition lies below or near 55 °C; in contrast, polycarbonate with a Tg near 145–150 °C has low chain mobility at the cycle temperature, and EO uptake is smaller. However, low uptake does not eliminate risk when the grade contains polycarbonate-silicone copolymers or phosphite antioxidants because extractables can react with EO or alter residual profiles. The correlation between sorption and gas-phase EO concentration is typically established during process development by exposing molded plaques to 100 % EO at 55 °C for 4 h in a laboratory chamber, followed by residual gas chromatography using ISO 10993-7:2008/Amd 1:2012 extraction; published data for multicomponent formulated devices remain limited. Table 1. Comparative polymer characteristics relevant to ethylene oxide terminal sterilisation.

Polymer category	Characterisation method	EO sorption behaviour	Processing or aeration concern	Representative application
Plasticised PVC	ASTM D638-14, ISO 11357-3:2018	High; free volume rises with plasticizer content	Ethylene chlorohydrin formation from chloride; prolonged aeration	Blood bags, enteral tubing, catheters
LDPE / EVA	ISO 11357-3:2018, ISO 62:2008	Moderate; diffusion increases with vinyl acetate content	Dimensional relaxation near cycle temperature; seal interaction	Packaging film, syringe plungers, connectors
Aromatic TPU	ASTM D638-14, ISO 11357-2:2020	Moderate to high depending on hard segment content	Hydrogen-bond disruption; moisture uptake before molding	Central venous catheters, wound drains
Silicone elastomer	ASTM D412-16, ISO 2789:2018	Very high; high free volume and permeability	Extended forced-air aeration; thickness-dependent retention	Respiratory masks, overmolded seals
Polycarbonate	ISO 11357-2:2020, ASTM D543-21	Low at 55 °C because of high Tg	Stress crazing from molded-in stress plus residual EO	Luer fittings, rigid connectors

Residual Ethylene Oxide and Ethylene Chlorohydrin Formation in Chloride-Containing Polymers

Ethylene chlorohydrin formation is a decisive constraint for plasticised PVC because the polymer contains covalently bound chlorine. Under EO sterilisation humidity and at local pH depressions generated by HCl evolution from PVC degradation, ethylene oxide can react with chloride ions to produce 2-chloroethanol. This route is influenced by the thermal stabiliser package; calcium-zinc stabilised PVC used in medical tubing is less likely to generate free HCl during extrusion than older barium-cadmium or lead-stabilised compounds, but no stabiliser eliminates chloride chemistry. Production-scale twin-screw compounding with L/D ratios of 28:1 to 40:1 and melt temperatures controlled between 170 °C and 185 °C must avoid shear-induced degradation because early discoloration measured by colourimetric yellowing index per ASTM E313-20 is an indicator of HCl release before sterilisation. If a compound is processed above 190 °C for extended residence time, the subsequent EO cycle can produce ethylene chlorohydrin at levels approaching or exceeding the ISO 10993-7:2008/Amd 1:2012 allowed limits for prolonged contact. Residual testing is performed by headspace gas chromatography with flame ionisation detection after simulated-use extraction at 37 °C for 24 h; the method must resolve ethylene oxide, 2-chloroethanol, and ethylene glycol without derivatisation interference from plasticizer breakdown products. Plasticizer selection alters both sorption and residual retention. Di(2-ethylhexyl) phthalate has been restricted or constrained under Regulation (EU) 2017/745 for medical devices where exposure justifies alternatives such as trioctyl trimellitate or di(isononyl) cyclohexane-1,2-dicarboxylate. From a sterilisation viewpoint, high plasticiser content increases free volume, accelerates EO uptake, and can extend aeration; plasticiser migration to package seals may also change seal integrity. Migration is evaluated by extraction methods in ISO 3826 and chemical characterisation under ISO 10993-18:2020. For high-plasticiser PVC blood bags, validation studies commonly include repeated 48 h aeration at 50 °C with forced air at 8 to 20 air changes per hour to bring ethylene chlorohydrin below detection limits. Published data for DINCH versus DEHP in EO-processed blood packs remain limited; therefore, each formulation must be tested under worst-case loading.

When Thermoplastic Polyurethane Hard Segments Are Exposed to Repeated Ethylene Oxide Cycles at 55 °C

Aromatic polyether and polyester thermoplastic polyurethanes are used in central venous catheters, wound drains, and enteral feeding tubes because of toughness and low-temperature flexibility. Their morphology consists of hard segments that form hydrogen-bonded microdomains and soft segments that control low-temperature flexibility. At 55 °C, the soft-segment region is above its Tg for many medical grades, allowing ethylene oxide diffusion; the hard domains remain associated but can be swollen if the grade has high content of polar chain extenders. Repeated EO exposure can alter surface hydrogen bonding as measured by attenuated total reflectance Fourier transform infrared spectroscopy using ASTM E1252-98. Injection molding of TPU catheter components is carried out on reciprocating screw machines with clamp force between 40 t and 150 t, mold temperatures of 20–50 °C, and a drying step at 80–95 °C for 2–4 h when ambient relative humidity exceeds 60 %; insufficient drying generates voids that serve as EO reservoirs. Mechanical testing after one and three simulated cycles per ASTM D638-14 Type V specimens can show a loss of ultimate tensile strength when hard segment content is below 30 %; however, published data for commercial grades vary because of antioxidant type. The processing conflict is that lower hardness TPU reduces patient trauma but increases EO sorption; higher hardness TPU may require plasticiser-free processing at melt temperatures near 190–210 °C where thermal degradation can occur in the presence of moisture. Suitable grades are selected by balancing Shore hardness from Shore A 75 to 95, melting range, and residual EO after realistic aeration. Polyethylene and polypropylene are often selected for rigid connectors, syringe plungers, and luer fittings because their nonpolar chemistry gives low ethylene chlorohydrin risk. In injection molding of polypropylene connectors on 80 to 250 t hydraulic machines with hot runner molds, packing pressure profiles influence molecular orientation; EO sorption is minimal, but the sterilisation temperature is close to the heat deflection temperature of some random copolymers, risking dimension changes. Gas permeability of low-density polyethylene is high enough for EO transport, but atmospheric aeration removes EO more quickly than from polar polymers; residual retention is typically evaluated after 24–48 h forced aeration at 45 °C and must account for package volume and loading density. Ethylene-vinyl acetate copolymers with vinyl acetate content from 9 % to 28 % are more polar and more flexible; increasing vinyl acetate increases EO uptake and reduces tensile modulus by ASTM D638-14.

Silicone Elastomers and Flexible PVC Do Not Share Equivalent Aeration Kinetics

Platinum-catalysed addition-cure silicone elastomers and peroxide-cured silicone rubber are used for respiratory masks, wound drains, seals, and overmolded components. The siloxane backbone produces unusually high free volume at ambient temperature; silicone exhibits high EO permeability and sorption, requiring extended aeration. Residual ethylene oxide is removed by forced-air convection and vacuum cycles, but EO can condense in thick cross sections where the surface area-to-volume ratio is low. Curing system residues can react with ethylene oxide or generate odorous by-products; therefore, peroxide-cured silicone must be post-cured before sterilisation to minimise residual peroxide decomposition products. In contrast, flexible PVC with high plasticiser content has lower gas permeability but higher ethylene chlorohydrin risk; aeration time is often driven by ethylene chlorohydrin, not ethylene oxide alone. A validated cycle for a silicone overmolded component may use forced convection at 50 °C for 72 h, while a thin-wall PVC tube of 0.5 mm wall may require 48 h for ethylene chlorohydrin to fall below the allowable limit. The two materials therefore cannot be grouped into a single worst-case product family without residue data generated by headspace gas chromatography. Operational boundaries are explicit. Pre-drying of thermoplastic polyurethane and polycarbonate before molding is required when ambient relative humidity exceeds 60 %; failure to dry before injection molding creates moisture-induced hydrolysis and increased EO absorption sites. Amine-based additives should be avoided in EO-processed polymers because ethylene oxide can alkylate amines and generate additional residues or odour. Polyoxymethylene is generally unsuitable for repeated EO reprocessing because trace acidic residues can cause depolymerisation and formaldehyde release; if acetal components are present in a device, they must be evaluated under ISO 10993-9:2019 for leachables and cannot be assumed stable. Polycarbonate connectors with high molded-in stress should be annealed or selected with greater molecular weight to reduce crazing when exposed to residual EO and humidity. These incompatibilities define the actual process window more than the base polymer data alone. Table 2. Compliance matrix for polymer selection under ethylene oxide terminal sterilisation.

Requirement	Standard designation	Equipment or test method	Polymer selection implication
EO sterilisation process validation	ISO 11135:2014	Gas steriliser with vacuum, humidity, and aeration control	Demonstrate worst-case loading with polymer-containing devices
Ethylene oxide, ECH, and EG residue limits	ISO 10993-7:2008/Amd 1:2012	Headspace gas chromatography	Set aeration time and confirm extraction method
Tensile properties before and after exposure	ASTM D638-14	Universal testing machine with extensometer	Detect embrittlement or plasticiser migration
Melt flow rate	ISO 1133-1:2022	Melt flow indexer	Monitor molecular degradation from heat or EO
Water absorption and dimensional change	ISO 62:2008, ASTM D570-22	Conditioning chamber, analytical balance	Correlate moisture uptake with EO sorption
Cytotoxicity after sterilisation	ISO 10993-5:2009	Cell culture assay	Confirm no toxic leachables from polymer or residues
Chemical compatibility and stress cracking	ASTM D543-21	Immersion bath, constant strain fixtures	Evaluate polycarbonate and multiphase polymer risk