Why Polybutylene Terephthalate Hydrolyses: The Ester Bond and Moisture Threshold

Polybutylene terephthalate is a semi-crystalline thermoplastic polyester produced by melt polycondensation of 1,4-butanediol and dimethyl terephthalate. The repeat unit contains an aromatic ring, a tetramethylene segment, and two ester linkages that function as the reactive centres for hydrolytic chain scission. Water attacks the carbonyl carbon through nucleophilic acyl substitution. The tetrahedral intermediate collapses with cleavage of the acyl-oxygen bond, producing a carboxylic acid end group and a hydroxyl-terminated chain. In PBT, the acid end group is typically a terephthalic acid residue. The liberated carboxylic acid is itself a polar catalyst for further ester cleavage, making the hydrolysis autocatalytic. This autocatalysis means that a small amount of moisture at the start of melt processing can create enough acid end groups to accelerate further chain scission even after the initial water is consumed. The practical moisture threshold for PBT in extrusion or injection moulding is therefore far lower than the equilibrium water absorption measured at room temperature. Supplier specifications regularly require a residual moisture content no higher than 0.02% by weight before the polymer reaches melt temperatures above 240°C.

The temperature dependence of hydrolysis is generally represented by an Arrhenius expression. Published activation energies for polyester hydrolysis vary with catalyst residues, end-group concentration, and filler surface chemistry, commonly reported in the range of 60 kJ/mol to 90 kJ/mol. Published kinetic data for a single commercial PBT grade across all moisture levels is limited; therefore, the autocatalytic mechanism is managed in production through moisture control rather than a single kinetic constant.

Hydrolysis in the solid state is further controlled by water diffusion. Because water transport through PBT occurs primarily in amorphous regions, hydrolytic chain scission is concentrated in tie chains and amorphous segments between crystalline lamellae. This morphological selectivity explains why hydrolysed parts often show embrittlement before significant mass loss or dimensional change is detected.

What Limits the Safe Moisture Content Before Melt Processing?

The limiting value is not the equilibrium moisture content at room-temperature storage, which is commonly reported near 0.2% at 23°C and 50% RH according to ISO 62, but the kinetic response of the ester linkage at melt temperature. At 250°C the hydrolysis rate is sufficiently fast that 0.02% water can cause measurable chain scission within the 2–5 min residence times typical of injection moulding. The threshold is determined by the relationship between water concentration, melt temperature, acid end-group concentration, and shear heating. A granulate moisture level of 0.05% is often cited in processing audits as a point at which viscosity loss becomes visible as a drop in injection pressure or an increase in melt volume-flow rate when tested per ISO 1133-1:2022. Acid end groups produced by early chain scission further catalyse the reaction, so the degradation rate is not linear with initial moisture content.

When hydrolytic chain scission is suspected, molecular weight distribution is the most direct measurement. Solution viscometry in 60:40 phenol/tetrachloroethane at 25°C is commonly used to determine intrinsic viscosity, with typical unfilled PBT grades ranging from 0.70 dL/g to 1.20 dL/g depending on molecular weight. A drop in intrinsic viscosity of more than 0.05 dL/g from the dry reference is often considered evidence of process-induced hydrolysis. Acid end-group titration detects the autocatalytic species directly; increasing carboxyl end-group concentration confirms that chain scission has occurred rather than simple thermal oxidation. Gel permeation chromatography in hexafluoroisopropanol can resolve the shift in the entire distribution, but it is slower and less suited to production lot release.

On a 25:1 L/D corotating twin-screw compounding line, moisture excursions above 0.05% have been observed as screw torque fluctuation and melt-pressure surging at 250°C barrel set point. The same line processing granulate dried to 0.015% produced stable die pressure and no detectable change in intrinsic viscosity between feed and pellet. This production-scale behaviour is consistent with the autocatalytic nature of the hydrolysis: initially localised chain scission releases acid end groups, which then promote further scission along the polyester backbone. The degradation products include low-molecular-weight aromatic species and butanediol, which can contribute to plate-out on mould vents and gas evolution during injection.

Drying System Requirements for Hydrolysis Control

Desiccant drying is mandatory for PBT because ambient air at typical processing-plant humidity does not allow the polymer to reach 0.02% residual moisture by hot air alone. The standard supplier recommendation is 4 h at 120°C with a drying-air dew point no higher than -30°C. Air flow should be sufficient to maintain a temperature profile across the hopper; typical industrial practice is 3.7 m³/h per 1 kg/h polymer throughput. The drying hopper must be insulated and capable of maintaining temperature at the bottom discharge, not just in the drying-air inlet. Production interruptions longer than 30 min frequently require re-drying because PBT granulate picks up surface moisture rapidly when transferred through unheated conveying lines. Regrind must be dried separately or blended before drying because its larger surface area and lower bulk density alter air distribution and can create local moisture pockets. Drying temperatures above 140°C should be avoided; although the polymer does not melt, prolonged exposure at these temperatures can produce yellowing and thermal oxidation that confounds later hydrolysis assessment.

In an injection moulding area with 1,200 kN clamp force presses, PBT dried to 0.012% and processed at a melt temperature of 250°C showed no significant shift in tensile strength or elongation at break when tested per ISO 527-2. Granulate from the same lot conditioned to 0.07% moisture produced parts with a 10–15% lower elongation at break and visible splay on flat surfaces. The moisture threshold for a specific moulding cell is therefore not a single material property; it is influenced by residence time distribution in the barrel, hot-runner volume, and the amount of regrind in the feed. Moulders with long hot-runner paths and high melt inventory are more sensitive to moisture because the hydrolytic reaction has additional time and temperature exposure.

When Hot-Water Exposure Accelerates Hydrolytic Ageing

When a PBT moulded part is exposed to hot water, hydrolysis is no longer limited to processing. Water diffuses into the amorphous regions of the semi-crystalline polymer, and chain scission advances from the surface inward. Continuous immersion above 60°C is generally not recommended for unstabilised unfilled PBT because the rate of embrittlement becomes significant. In automotive coolant-system components, glass-fibre-reinforced PBT grades are often selected with hydrolytic stabilisers because the water-glycol environment at 120°C can degrade the ester bond within hundreds of hours. The reason hydrolytic ageing first appears as embrittlement rather than mass loss is that scission of tie chains and amorphous segments reduces the network of load-bearing molecules while the crystalline regions remain intact. Tensile elongation at break and Charpy impact strength are therefore more sensitive indicators of hydrolytic degradation than tensile strength or weight change. Water absorption measurements per ISO 62 on unfilled PBT after 24 h at 23°C are commonly near 0.2%, but this value rises substantially as temperature increases because the increased segmental mobility allows more water uptake in the amorphous phase.

Additive selection for hydrolytic stability depends on the end-use water activity and processing temperature. Carbodiimide-based hydrolysis stabilisers react with carboxylic acid end groups and free water, reducing autocatalytic chain scission. Epoxy-functional chain extenders can rebuild molecular weight by reacting two carboxyl end groups, but overloading can increase melt viscosity and create gel particles in the melt. In production, the stabiliser is dispersed during compounding, and its effectiveness is judged by accelerated ageing in 95°C water followed by tensile testing per ISO 527-2. Acceptance limits for specific components are application-dependent, but retention of elongation at break is commonly used because it detects embrittlement earlier than tensile strength. No stabiliser completely eliminates hydrolysis; it only reduces the rate of acid-catalysed chain scission.

Injection Moulding Variables That Amplify Chain Scission

Injection moulding setpoints interact with residual moisture in ways that can create misleading quality data. High melt temperature, low nozzle back pressure, long hold time, and large hot-runner volume all extend the effective residence time above 240°C. Screw speed and back pressure also influence shear heating; a 20°C shear-induced temperature rise can push a 250°C set melt into a region where hydrolysis rate is significantly higher. The use of a shut-off nozzle or a hot-runner valve gate with stagnant melt volumes can retain degraded resin that eventually contaminates otherwise dry shots. Mould temperature controls crystallinity and shrinkage, but it cannot reverse the molecular weight loss caused by hydrolysis; a part may meet dimensional specifications while failing impact requirements because the polymer backbone has already been cut. The practical control measure is to verify moisture content at the feed throat with a Karl Fischer analyser operating according to ISO 15512 and to reject lots above 0.02% before they enter the barrel.

A hydrolysis investigation on PBT is incomplete without linking observed property loss to standardised test methods. The following matrix is used to assess incoming granulate, dried material, and moulded test specimens.

Karl Fischer titration of the granulate is preferred over weight-loss methods because it selectively measures water and avoids confusing water loss with volatile oligomers or butanediol released from already hydrolysed resin. Melt volume-flow rate testing per ISO 1133-1:2022 provides an indirect indication of chain scission, but it must be interpreted against the reference dry sample because flow rate also depends on filler content and moisture-induced plasticisation. Tensile testing per ISO 527-2 or ASTM D638-14 is used after hydrolytic ageing to detect embrittlement. If a part is intended for hot-water service, water absorption per ISO 62 is paired with mechanical property retention after immersion. The method selection depends on whether the failure mode under investigation is melt-processing hydrolysis or long-term service hydrolysis.