Polyethylene Terephthalate (PET) in PET Foam, Nonwoven and Composite Forms: Beyond Bottles and Film occupies a distinct process window where melt strength, intrinsic viscosity, and hydrolytic stability dictate conversion economics. Bottle-grade resin with an intrinsic viscosity near 0.80 dL/g requires chain extension or reactive modification before foam extrusion because the unmodified melt exhibits low extensional viscosity and rapid sag at processing temperatures above 270 °C. In nonwoven and composite feedstreams, higher recycled content is accepted, but melt filtration burden and carboxyl end-group drift alter throughput and product uniformity. The following sections address foam, nonwoven, and composite process paths with reference to equipment conditions, test standards, and operational boundaries.

What Limits Stable Foam Cell Nucleation in Extrusion-Grade PET?

Continuous PET foam extrusion on a co-rotating twin-screw line with an L/D ratio of 40:1 requires maintaining melt temperature between 270 °C and 285 °C at the die entry while keeping pressure above 8 MPa to solubilize physical blowing agent. At 0.5–2.0 wt% chemical blowing agent addition, decomposition gas yield becomes the controlling variable; azodicarbonamide evolves nitrogen at approximately 155–210 °C, which is below PET processing temperature, and therefore must be pre-dispersed in a concentrate to avoid pre-foaming in the feed zone. Physical blowing agent systems based on CO₂ or N₂ require a high-pressure injection port after the melting section; published data for this specific configuration is limited, but industrial trials report die pressures between 6.5 MPa and 12 MPa to prevent premature bubble growth.

Cell nucleation density is measured by scanning electron microscopy and correlated with pressure drop rate across the die. A pressure drop rate below 3 MPa/s typically produces open-cell or coalesced structures with density above 120 kg/m³, whereas a rate above 10 MPa/s can produce closed cells with mean cell diameter below 300 µm. Cooling calibration after the die must reduce surface temperature below the Vicat softening point of approximately 78 °C for amorphous PET before wind-up; otherwise, roll crush collapses the cellular core. These constraints illustrate a processing window of approximately ±5 °C on the calibrator and a moisture limit of 50 ppm in the resin feed to limit hydrolytic degradation.

Extensional viscosity measurements on a uniaxial extensional rheometer at 270 °C show strain-hardening ratios below 1.5 for unmodified bottle-grade PET and above 2.0 after chain extension. Below this threshold, foam expansion ratios above 10:1 are not maintained because cell walls rupture during bubble growth. For densities below 60 kg/m³, the use of a physical blowing agent with high solubility and additional chain extender is necessary; published data for this specific configuration is limited.

High-melt-strength PET grades are produced by reactive extrusion with pyromellitic dianhydride at 0.1–0.5 wt%. Chain extension increases the shear storage modulus and terminal relaxation time, but residual anhydride can produce gel particles above 0.3 wt%. Filtration through a screen changer at 40 µm is then required. The gel particle density after filtration is typically below 3 particles/kg in the final foam feed, measured by a pressure-rise test on a capillary rheometer.

Post-extrusion crystallinity development is a further boundary. Amorphous PET foam at density 80 kg/m³ has compressive strength below 0.8 MPa when tested to ISO 844. Thermal setting in a through-air oven at 180–210 °C for 4–8 min raises crystallinity above 25% and compressive strength to 1.2–2.5 MPa. Without crystallisation, foam cores used in sandwich panels exhibit creep deflection above 0.5% after 24 h under 0.3 MPa load, as per ISO 7850.

Table 1 presents comparative PET foam density-dependent property data.

In meltblown PET nonwoven production, extruder melt temperature is held at 280–300 °C and the hot air die operates at 290–310 °C. Fibre diameters in the 1–5 µm range are achieved with air flow rates of 500–1,500 m³/h per linear metre of die; the resultant web tensile strength is anisotropic, with machine-direction values typically 8–25 N/50 mm at basis weights of 15–60 g/m² as per ISO 9073-3. Spunbond PET lines, in contrast, produce filament diameters of 15–30 µm and require inline drawing at 2,000–4,000 m/min. In each case, dried resin with residual moisture below 50 ppm is essential; at higher moisture levels, intrinsic viscosity loss can exceed 0.05 dL/g over a single pass.

Thermal point bonding of spunbond PET requires an oil-heated calender with engraved roll surface temperature between 190 °C and 220 °C. At nip pressures of 40–80 N/mm², bond point area above 18% increases tensile strength but reduces elongation at break to below 45%. Below 15% bond area, surface abrasion resistance measured by Martindale testing to ISO 12947-2 falls below 20,000 cycles.

Hydroentangled PET nonwovens at 30–80 g/m² require water jet pressures between 8 MPa and 25 MPa on a three-stage manifold. The resulting fabric achieves wet tensile strength above 80% of dry tensile strength under ISO 9073-3, but only when fibre diameter is above 10 µm. Finer meltblown webs undergo surface disruption at jet pressures exceeding 15 MPa, causing pilling and basis weight variation above ±5%.

For post-consumer recycled PET nonwovens, melt filtration through a dual-vessel screen changer with filtration rating of 40–100 µm is required to remove residual polyolefin contamination. Pressure rise across the filter above 8 MPa typically mandates a backflush or screen change cycle. The recycled feedstream also exhibits carboxyl end-group concentrations from 20–60 mmol/kg; when combined with nylon or cellulosic binders in hydroentangled nonwovens, lower pH during wet finishing can accelerate fibre embrittlement. Separation of silicone-coated label film is therefore a critical preprocessing boundary; published data for this specific configuration is limited, but line audits show that residual silicon-containing particles at levels above 50 ppm cause spinneret hole pressure fluctuations of ±0.5 MPa.

When Glass Fibre Compounding Shifts PET from Packaging Grade to Structural Matrices

Injection moulding of glass fibre reinforced PET at 30 wt% short glass content typically uses a two-stage screw with an L/D ratio of 20:1–24:1 and barrel temperatures of 270–290 °C. Mould temperature is held at 120–140 °C to promote crystallisation; below 110 °C, the part remains largely amorphous and may exhibit heat deflection temperature below 85 °C. At 30 wt% glass fibre, tensile strength reaches 130–150 MPa and tensile modulus 9–10 GPa when tested according to ASTM D638-14. Notched Izod impact values, however, remain in the 65–95 J/m range under ASTM D256-10e1, limiting use where crack arrest is critical.

Melt viscosity control is critical for thin-wall moulding. At 285 °C and 2.16 kg load, the melt flow index of a 30% glass-filled grade is typically 15–30 g/10 min under ISO 1133-1:2022. Higher melt flow improves fill but corresponds to intrinsic viscosity below 0.65 dL/g, reducing tensile elongation at break to under 2.5%. This creates a property cliff-edge: mouldability gains above 30 g/10 min are accompanied by rapid loss of under-hood creep resistance.

Moulding a glass-filled PET structural bracket with projected area of 400 cm² requires clamp force above 1,500 kN because PET composites have a packing pressure requirement of 60–90 MPa. Lower clamp force results in flash and dimensional variation greater than ±0.15 mm across the parting line.

Long-fibre-reinforced PET pellets processed on an injection moulder with low-compression screw and smooth transitions retain fibre lengths of 2–4 mm after plastication. The resulting tensile modulus can exceed 10 GPa at 30 wt% fibre content, but notched Izod improvement over short fibre is only 10–20% because PET matrix deformation is limited by crystallinity above 30%. Fibre pull-out length measured on fracture surfaces is therefore shorter than in polypropylene-based long fibre compounds.

Sandwich panels using PET foam cores and glass-reinforced PET skins are bonded with moisture-curing polyurethane adhesives. Shear lap shear strength above 5 MPa is achievable when the foam surface is corona-treated to a surface energy of 48–56 mN/m, measured by contact angle. Peel failure occurs below 40 mN/m, making surface preparation a necessary step. The coefficient of linear thermal expansion for glass-filled PET is approximately 30–40 × 10⁻⁶ K⁻¹, which is higher than the 15–20 × 10⁻⁶ K⁻¹ of aluminium honeycomb; therefore, large-area panels require allowance for differential expansion at service temperatures above 80 °C.

Table 2 presents the core compliance and property test matrix used for PET foam, nonwoven, and composite stock.

Material Drying, Hydrolytic Stability, and Screw Configuration Boundaries

Pre-drying at 120–140 °C for 4–6 h is mandatory when ambient relative humidity exceeds 60%; residual moisture above 50 ppm produces hydrolytic degradation at the processing temperature. For PET foam, this is particularly severe because the blowing agent injection zone increases melt residence time by 15–30 s, resulting in intrinsic viscosity loss of 0.02–0.05 dL/g per pass. In twin-screw compounding of PET composites, screw designs using high free volume configuration with 2–3 mixing sections are used; excessive shear heating above 300 °C triggers acetaldehyde generation above 2 ppm in the final part.

PET composite formulations should avoid combination with amine-based flame retardant additives due to premature chain scission and yellowing at 280 °C. If flame retardancy is required for rail or automotive applications, brominated styrene with antimony trioxide at 2–5 wt% is used under REACH authorization constraints, but the resulting elastomer-free matrix has low impact resistance below 4 kJ/m².

For recycled PET feedstreams, batch-to-batch variation in intrinsic viscosity from 0.62 dL/g to 0.80 dL/g forces screw speed adjustments of ±15% to maintain head pressure. If the extruder is equipped with a melt pump before the die, pressure can be held at 8–10 MPa despite viscosity drift, but the pump seals require replacement at intervals below 3,000 h when processing abrasive titanium dioxide concentrates above 4 wt%.