The evaluation of polybutylene terephthalate (PBT) under 85°C/85% RH damp heat is not a generic environmental exposure check; it directly probes ester linkage scission kinetics and residual carboxyl end group autocatalysis in a semicrystalline terephthalate backbone. Hydrolysis proceeds through water diffusion into the amorphous phase, nucleophilic attack at the carbonyl carbon, and chain cleavage that lowers number-average molar mass. Because tensile elongation at break is governed by the high-molar-mass tail of the distribution and by stable necking, elongation retention falls before tensile strength retention. The standard specimen configuration is a multipurpose Type A dogbone conforming to ISO 3167, moulded according to ISO 294-1:2017, and tested per ISO 527-2:2012 at 5 mm/min. When North American supply chains require direct comparison, ASTM D638-14 Type I specimens are used with the same data treatment. Damp heat exposure is maintained in a chamber meeting IEC 60068-2-78:2012 steady-state conditions, with tolerances of ±2°C and ±3% RH.

Why does unfilled PBT lose more than 50% of its elongation at break before tensile strength falls below 70% retention?

Hydrolytic chain scission reduces number-average molar mass from a typical 22,000–30,000 g/mol to below 10,000 g/mol within 500–1000 h under 85/85 exposure, depending on stabilizer content. Tensile strength is comparatively insensitive to molar mass above a critical threshold because yielding and craze propagation require only modest chain entanglement density. Elongation at break, by contrast, reflects post-yield extension and disentanglement; once the degraded amorphous regions can no longer sustain stable necking, macroscopic brittle failure occurs. The reaction is autocatalytic because each cleavage event creates one additional carboxyl end group, which lowers local acidity and increases water sorption. In unfilled injection-moulded formulations, initial carboxyl end group concentrations of 15–40 meq/kg can rise to 80–150 meq/kg after 1000 h 85/85 when no acid scavenger or chain extender is present. Production-scale failure data from underhood connector housings moulded on 80–120 ton clamp force machines with hot runner temperatures of 250–260°C indicate that high shear in the gate region produces residual stress concentrations that also display cracked surfaces after 500 h, even when bulk retained tensile strength remains above 60 MPa.

Before moulding, PBT pellets require forced-air dehydration at 120–140°C for 4–6 h to a residual moisture content of <0.02 wt%, as measured by a Karl Fischer titrator equipped with a vial oven set to 160°C. Moulding from undried resin reduces molar mass in the barrel and seeds the hydrolytic damage that the 85/85 test then amplifies. A single-screw injection unit with a general-purpose polyolefin screw is unsuitable; a three-zone screw with a compression ratio of 2.2:1–2.8:1 and a non-return valve is specified to control residence time. Barrel temperatures are profiled from 240°C at the feed throat to 265°C at the nozzle, while the mould temperature is held at 80–100°C to maximize crystallinity, particularly in thin wall sections below 2 mm. Hold pressure of 60–80 MPa and a short fill-to-pack transition reduce jetting and internal void formation. Specimen conditioning after moulding follows ISO 291:2008 class 2, 23°C and 50% RH, for at least 48 h before baseline tensile testing.

Specimen thickness controls the time to moisture saturation because water uptake in PBT follows Fickian diffusion; the saturation time scales with the square of the diffusion path. At 85°C, the diffusion coefficient of water in PBT is commonly reported in the range of 1×10⁻¹²–5×10⁻¹² m²/s, so a 4 mm ISO 3167 coupon requires 500–1000 h to approach equilibrium, whereas a 1.6 mm connector wall saturates within 100–250 h. This dimension dependence explains why standard ISO 3167 bars can show 70% tensile strength retention at 1000 h while thin-wall production parts fail at 500 h. Tensile bars sectioned from moulded housings must therefore be listed separately in the test report from injection-moulded multipurpose coupons.

When the 85/85 chamber is opened for an intermediate pull and the specimens are not reconditioned

Transient moisture accumulated in the amorphous phase plasticizes the polyester and lowers the glass transition temperature from approximately 45–60°C dry to below 30°C in saturated specimens, so immediate tensile testing at 23°C can produce a lower modulus and a higher elongation than testing after standardized reconditioning. To avoid false retention values, specimens removed from the chamber should be sealed in vapour-barrier bags, cooled to 23°C for 4 h, and then tested within 8 h. Alternatively, the same coupons are reconditioned at 23°C/50% RH for 24 h before ISO 527-2:2012 testing. Testing wet and unreconditioned specimens is permissible only when the end-use environment includes humid assembly; in that case, the report must state the conditioning deviation and identify the tensile result as “as-conditioned” rather than “dry.” For intermediate pulls at 250 h, 500 h, and 1000 h, at least 5 specimens per condition are required to maintain a coefficient of variation below 5% for tensile strength.

Tensile retention is calculated for each property as the ratio of the aged mean to the unaged mean multiplied by 100. Reporting tensile strength retention without elongation at break is an incomplete dataset because the ductile-to-brittle transition controls snap-fit and impact function. A common acceptance criterion for automotive underhood PBT connectors is 70% minimum tensile strength retention and 50% minimum elongation retention after 1000 h 85/85. For 30 wt% glass-fibre-reinforced PBT, the elongation criterion is lower because unaged elongation at break is already reduced to 2.0–3.5%. Statistical treatment requires Student’s t-distribution or Weibull analysis when the aged failure mode changes from ductile to brittle and the standard deviation increases from 1.5 MPa to 4–6 MPa. The test report must include the specific tensile test speed (5 mm/min for ISO 527-2:2012; 5 mm/min or 50 mm/min for ASTM D638-14 depending on ductility), specimen dimensions, and chamber log traces.

Quality control matrix for hydrolysis retention programmes

Post-aging fracture surfaces from brittle tensile breaks should not be examined without optical microscopy because the first 200–500 µm from the moulded skin often contains the initiation site. Scanning electron microscopy of unfilled PBT specimens after 1000 h 85/85 reveals radial hackle marks and a smooth surface-initiated crescent at the gate-side edge where residual moulded-in stress is highest. Solution viscosity in 60:40 phenol/tetrachloroethane at 30°C according to ISO 1628-1:2021 declines from 0.95–1.10 dL/g to 0.55–0.70 dL/g as weight-average molar mass falls below 40,000 g/mol. Carboxyl end group titration in o-cresol/chloroform using ethanolic KOH gives a direct measurement of autocatalytic potential; values above 120 meq/kg after aging correlate with total loss of ductility in unfilled PBT. Differential scanning calorimetry at 20 K/min can detect a secondary melting endotherm caused by hydrolytic lamellar reorganization, but it is secondary to molecular weight data.

Glass fibre wicking, flame retardant packages and regrind ratios create lower practical ceilings than the resin datasheet predicts

With 30 wt% short glass fibre, the 85/85 failure mode shifts from bulk hydrolysis to capillary wicking along the fibre-matrix interphase. Silane coupling agents improve dry adhesion but can undergo hydrolysis at the interface, creating channels for water transport. Published supplier technical bulletins describing a 40:1 L/D co-rotating twin-screw extruder with a side feeder at 250°C report that unfilled PBT with a polymeric carbodiimide stabilizer at 1.0–1.5 wt% retains 60–70% elongation after 1000 h 85/85, while a 30% glass-filled version with the same stabilizer retains only 40–55% elongation. Brominated flame retardant packages combined with antimony trioxide complicate hydrolysis because antimony ions and acidic by-products can catalyse ester cleavage; published data for specific FR-stabilizer interactions is limited and must be generated on the actual compounding line. Regrind above 20–30% is not recommended for hydrolysis-critical parts because the recycled material already contains carboxyl end groups and can reduce the induction period by as much as 200 h.

Batch acceptance for hydrolysis-critical PBT compounds requires release testing of solution viscosity and carboxyl end groups in addition to the standard melt flow rate. Melt flow rate measured at 250°C with 2.16 kg according to ISO 1133-1:2022 should fall within 12–18 g/10 min for unfilled injection grades; after 85/85 aging, melt flow rate increases sharply as molar mass falls, so a post-aging value above 35 g/10 min indicates severe degradation. Injection moulders must record the actual dew point of the drying hopper, not the setpoint, because a desiccant dryer delivering air at a dew point of -40°C can still leave pellets above 0.02 wt% moisture if granulate residence time is below 4 h. Moisture analysis of the granulate should be performed on samples taken from the machine throat, not from the storage silo, to capture regrind segregation and conveying-line moisture pickup.