Polybutylene Terephthalate (PBT): Recovery Behavior in Thermoplastic and Thermoset Systems

Polybutylene terephthalate is a semicrystalline thermoplastic polyester synthesized from 1,4-butanediol and terephthalic acid or dimethyl terephthalate. The linear backbone contains repeating tetramethylene terephthalate units; the tetramethylene segment provides chain flexibility relative to polyethylene terephthalate, and the ester carbonyl units provide a site for controlled solvolysis. The crystalline phase melts at approximately 223 °C when measured by differential scanning calorimetry according to ISO 11357-3, and the glass transition temperature is typically reported near 45 °C to 60 °C for unfilled material. These thermal transitions are absent or fundamentally different in cured thermoset networks. Why Polybutylene Terephthalate Is More Recoverable Than Thermoset Plastics is directly traceable to molecular architecture: PBT chains can move relative to one another above the melting point, re-entangle, and crystallize upon cooling, while a thermoset network cannot flow below its thermal decomposition temperature. Thermoset plastics such as amine-cured epoxies, phenolics, and unsaturated polyesters form a three-dimensional covalent network during cure. Recovery of those materials is therefore confined to mechanical size reduction, pyrolysis, solvolysis, or filler substitution. PBT recovery, in contrast, is a melt-state operation governed by moisture control, residence time, and molar mass preservation.

On industrial molding lines, the primary PBT recovery stream is in-plant regrind from sprues, runners, and rejected parts. Suppliers of unreinforced and 30% glass-fiber-reinforced PBT grades specify drying at 120 °C to 130 °C for 4 h to 6 h in desiccant dryers with a dew point of −40 °C or lower. The target moisture content is less than 0.02% by weight before melt processing. When this limit is exceeded, ester hydrolysis reduces molecular weight during plastication, and the effect is visible as an increase in melt volume-flow rate when tested according to ISO 1133-1 at 250 °C and 2.16 kg. This sensitivity is not a barrier to recovery; it is a boundary condition that separates stable reprocessing from rapid degradation. In plants without closed-loop drying, regrind exposed to relative humidity above 60% must be held in sealed hoppers after drying. These requirements are standard and do not require chemical modification of the polymer.

How Does Melt Filtration Affect Recovered PBT Quality in Multi-Cycle Processing?

Multi-cycle reprocessing of PBT is not equivalent to simple remelting. Each pass through a compounding extruder or injection unit adds thermal and shear history. A production-scale twin-screw compounder with an L/D ratio between 32:1 and 44:1 and a barrel profile from 235 °C to 255 °C can disperse regrind and rebuild viscosity through intimate blending with virgin resin. However, when the recyclate contains glass fiber, fiber length decreases with each cycle. This attrition is measured by ashing according to ISO 1172 and by optical microscopy; the retained glass length after compounding is typically below 300 µm in standard short-glass grades. Melt filtration is commonly introduced at the die head using screen packs with mesh counts from 60 to 200, depending on molded part requirements. Filtration removes carbonized specks and agglomerates but cannot restore molecular weight. Therefore, the percentage of regrind is determined by the intended final property window, not by the theoretical ability of PBT to be remelted.

Recovered PBT can be converted by injection molding under conditions described in ISO 294-1. A typical barrel set-point profile for unreinforced PBT is 240 °C, 245 °C, 250 °C, and 255 °C from feed to nozzle, with a mold temperature of 40 °C to 80 °C. The material remains semicrystalline, and the mold temperature controls the degree of crystallinity at the skin and core. Too low a mold temperature freezes the surface before full packing; too high a mold temperature increases cycle time and can reduce toughness through increased crystallinity. These constraints are identical for virgin and recovered material, and they confirm that PBT recovery does not require a separate processing platform. Unfilled recovered PBT is not, however, a direct substitute for virgin material in thin-wall electrical connectors unless the melt volume-flow rate and tensile elongation remain within the original grade specification.

The flow behavior of recovered PBT depends on pressure, shear rate, and the presence of fillers. Capillary rheometry at 250 °C shows that unfilled PBT is a shear-thinning melt; the apparent viscosity decreases as shear rate increases, which is favorable for filling thin sections. In recycled grades, the shear-thinning behavior remains, but the zero-shear viscosity may be lower than that of virgin resin because of reduced chain length. Mold filling simulation software, when supplied with melt viscosity data from ISO 11443, can predict whether a given regrind fraction will cause short shots or jetting. A recovered lot with viscosity below the lower specification limit can be blended with virgin resin to restore processability. This is a routine operation in injection molding plants and is governed by the same melt-rheology standards used for virgin PBT.

Thermoset Crosslink Density and the Irreversible Network

Thermoset recovery is limited by the same network architecture that provides solvent resistance and creep resistance. In a cured epoxy system, the polymer consists of a covalently bonded network with crosslink density defined by the stoichiometry of the epoxy resin and the amine hardener. The material exhibits a glass transition temperature but no melting transition. Dynamic mechanical analysis per ISO 6721-1 can identify the glass transition, but the absence of a melting endotherm means that no melt-state processing window exists. Above the glass transition, the network remains solid until thermal decomposition begins. Thermogravimetric analysis per ISO 11358 generally records decomposition onset in the range of 300 °C to 400 °C for cured epoxy and phenolic systems, a temperature range that is not available for melt processing because decomposition evolves volatiles and char. PBT, by comparison, has a usable melt window between 220 °C and 260 °C, with crystallization from the melt occurring near 190 °C. This allows PBT recyclate to be homogenized in an extruder and re-formed into a continuous phase. Thermoset recyclate cannot form a continuous phase without a reactive binder.

Mechanical recycling of thermoset scrap therefore proceeds through pulverization into fine particle fractions. A size-reduced thermoset powder with a median particle diameter below 100 µm is sometimes added to thermoplastic compounds as a low-cost filler. In such a system, the thermoset particles are dispersed as discrete, non-melting inclusions within the PBT matrix. The melt viscosity of the compound increases, and the tensile elongation at break falls because the inclusions disturb local deformation. Characterization of the interface becomes essential; scanning electron microscopy of cryofractured specimens after tensile testing per ISO 527-2 shows whether the PBT matrix has adhered to the thermoset particles. If adhesion is poor, the particles behave as voids, and notched Izod impact values determined by ISO 180/A decline sharply. This filler route confirms that thermoset recovery is fundamentally different from PBT recovery: thermoset scrap is a rigid filler, while PBT scrap is a melt-processable feedstock.

Glass-fiber-reinforced PBT grades present a specific recovery challenge. The glass fibers break during grinding and re-extrusion, so the reinforcing efficiency decreases. A recycled 30 wt% glass-fiber-reinforced PBT may retain the same total glass content but not the same fiber length distribution. The short fibers reduce tensile strength and notched impact more than they reduce tensile modulus. Ash content per ISO 1172 verifies the glass fraction, but ash content alone cannot detect fiber length reduction. Consequently, recovered glass-reinforced PBT is often blended with virgin reinforced or unreinforced material to adjust the mechanical profile. This adjustment requires mechanical testing after each blend iteration, typically tensile tests per ISO 527-2 and flexural tests per ISO 178.

When Contaminated Thermoset Fractions Are Present in PBT Reprocessing Streams

Cross-contamination between PBT recyclate and thermoset scrap is a process risk. A small mass fraction of cured thermoset particles in a PBT melt stream can produce surface defects, filter blockage, and localized mechanical weakness. On a reciprocating-screw injection molding machine with a clamp force of 1,000 kN to 3,000 kN, contamination may appear as non-melt inclusions in the mold filling phase, especially in thin-wall sections below 1.0 mm. The differential scanning calorimetry trace of an unfilled PBT recyclate should show only the expected melting endotherm near 223 °C; additional exothermic or decomposition events indicate contamination. In such cases, the recovered lot is quarantined or re-filtered. Standard practice is to sort material at the granulate feed stage and to verify purity by burn-off tests or density separation. PBT has a density near 1.31 g/cm³ measured by ISO 1183-1, while many thermosets range from 1.2 g/cm³ to 1.8 g/cm³ depending on filler content, so density-based sorting can reduce but not eliminate mixed scrap.

Recovery documentation for PBT uses a set of standardized measurements to establish whether the recycled fraction remains within specification. The melt volume-flow rate per ISO 1133-1:2022 provides a rapid indication of molecular weight loss, but it does not replace intrinsic viscosity measurement per ISO 1628-5. Tensile modulus and strength per ISO 527-2 quantify mechanical retention, while notched Izod impact per ISO 180:2023 detects embrittlement caused by fiber attrition or contamination. Thermal stability is typically checked by thermogravimetric analysis using ISO 11358 or by the onset of the melting endotherm in ISO 11357-3. A recovered PBT lot that passes these checks can be reintroduced into injection molding or extrusion without a separate chemical depolymerization step. In contrast, thermoset materials require gel content or sol fraction testing—such as ASTM D2765 or equivalent Soxhlet extraction—to confirm that the network remains fully cross-linked and therefore un-meltable.

Post-consumer PBT recovery is smaller in volume than post-industrial recovery because PBT is used in automotive and electrical components that are not easily separated from other engineering polymers. Density-based separation can isolate PBT from polyolefins, but separation from polyamide or polycarbonate is more difficult. Near-infrared sorting can identify PBT if the black pigment load is low. For glass-filled black parts, automatic sorting is less reliable, and published data for mechanically sorted post-consumer PBT recyclate from mixed automotive shredder residue are limited. Nevertheless, once a reasonably pure PBT fraction is obtained, the material can be melt processed. Thermoset post-consumer waste, by contrast, cannot be melted regardless of sorting purity.

Post-industrial PBT recyclate with low intrinsic viscosity can be upgraded by solid-state polymerization or by reactive extrusion with chain extenders. In solid-state polymerization, granulate is heated under vacuum or inert gas at temperatures below the melting point, typically 190 °C to 210 °C, to drive polycondensation and increase molecular weight. This process requires a crystallized, dry feedstock and is more common with polyethylene terephthalate than PBT, although the same ester chemistry applies. Reactive extrusion of PBT recyclate with a dianhydride or bis-oxazoline chain extender can also rebuild molar mass in a twin-screw extruder. The shear and residence time must be tightly controlled because gel formation lowers impact strength. Such methods are not applicable to thermosets because the cross-linked network cannot be linearized by heat alone.

Beyond mechanical recovery, PBT is chemically recoverable through hydrolysis, glycolysis, or methanolysis of ester linkages. The repeat unit contains two ester bonds that are susceptible to controlled solvolysis. Hydrolysis at elevated temperature and pressure converts PBT into terephthalic acid and 1,4-butanediol, both of which are virgin-grade monomers when purified. This is a distinct advantage over thermosets because the curing reaction of a thermoset converts monomer into a network with carbon-carbon, ether, or amine crosslinks that do not cleave selectively under mild hydrolysis. Chemical recovery of cured thermosets often requires aggressive media such as hydrogen peroxide, nitric acid, or supercritical fluids, and monomer yields are generally lower and less selective. Thus, the recoverability of PBT exists on two fronts: melt reprocessing and selective solvolysis. Both routes rely on the linear ester functionality, not on process-specific additives.

Flame-retardant PBT grades require separate recovery accounting. Halogenated formulations, usually containing brominated flame retardants and antimony trioxide, can release acidic species at elevated temperatures. Recovery of these grades is performed only after checking thermal stability by thermogravimetric analysis and after adding an acid scavenger to the extrusion feed. Published data for recycled flame-retardant PBT often report reduced tensile elongation and impact strength relative to virgin material, but the melting point remains near 223 °C and the material remains melt-processable. This balance of retained processability and reduced ductility is not available in thermoset systems, which cannot be re-melted even when mechanical properties after cure are high.