Your Supplier Stops the Bottle-Grade PET You Buy: Finding an Equivalent

Substitution of a bottle-grade polyethylene terephthalate (PET) resin requires equivalence validation against the incumbent lot rather than generic nameplate conformity. A replacement is not qualified by intrinsic viscosity alone; the converter must verify molecular weight surrogate measurements, comonomer content, thermal history, pellet handling behaviour, and preform-to-bottle process output. A qualified equivalency protocol begins with the supplier’s certificate of analysis reviewed against ASTM D4603-18 for intrinsic viscosity, ISO 11357-3:2018 for melting and crystallisation, ISO 15512:2019 for moisture, and CIE L*a*b* colour. The processor should reject equivalence claims that are limited to resin density or melt flow rate; PET bottle resin is specified primarily by dilute-solution viscosity, comonomer content, acetaldehyde potential, pellet bulk density, fines content, and solid-state polymerisation history.

The first production-scale risk is that a replacement resin may match the certificate of analysis but behave differently in high-cavitation preform moulding. A 72- or 96-cavity mould with hot-runner tips, water-cooled take-out plates, and post-mould cooling creates a narrow processing window in which small changes in melt viscosity or crystallisation rate shift shot-to-shot consistency. Therefore, equivalence testing must include long-run preform weights, hot-runner pressure drop, cooling time, gate crystallinity, and preform acetaldehyde at steady state. A one-time resin sampling is insufficient; batch-to-batch variation within a supplier may be larger than the initial approval sample.

Intrinsic viscosity for carbonated soft-drink preforms is commonly specified at 0.80–0.84 dL/g; water grades may be approved at 0.74–0.78 dL/g, hot-fill grades near 0.82 dL/g, and edible-oil bottles above 0.82 dL/g. The measurement uses PET dissolved in a phenol/1,1,2,2-tetrachloroethane 60/40 wt% mixture at 30 °C under ASTM D4603-18; a viscosity number determined by ISO 1628-1:2021 is not interchangeable unless the solvent system, concentration, and Huggins extrapolation are declared. Copolymer composition in bottle grades typically includes isophthalic acid at 1.0–3.5 wt% and diethylene glycol at 0.7–1.4 wt%, both of which suppress spherulitic crystallisation and widen the stretch-blow window. A replacement should match these ranges within ±0.3 wt% for diethylene glycol and ±0.5 wt% for isophthalic acid; larger deviations shift strain-hardening and thermal crystallisation. Table 1 lists a screening matrix for the incoming lot.

Bulk density and fines content are rarely printed on a certificate of analysis but materially affect drying and feed consistency. Bulk density is commonly controlled near 0.80–0.85 g/cm³ for cylindrical or spherical pellets; fine fractions above 0.5% by weight can blind dryer filters and increase feed-throat bridging. Pellet size distribution should also be compared because solid-state polymerisation rate is inversely affected by pellet thickness. Published data for a specific replacement grade’s pellet-size relationship to SSP rate is often limited; the receiving plant must measure particle size distribution and drying time directly.

Can Solid-State Polymerisation History Create a Hidden Performance Gap?

Solid-state polymerisation raises molecular weight while limiting thermal degradation. The process typically heats crystallised amorphous pellets to 210–230 °C under vacuum or nitrogen flow for several hours; water and ethylene glycol are removed, shifting the equilibrium toward higher molecular weight. Two resins with identical intrinsic viscosity can differ in carboxyl end-group concentration, which influences acetaldehyde formation and hydrolytic stability. Carboxyl end groups in bottle-grade PET are often specified below 25–35 mmol/kg; a higher value after SSP indicates a shorter condensation history or incomplete post-polymerisation. A replacement with equivalent IV but higher carboxyl end groups may generate acetaldehyde more rapidly during injection moulding, particularly at hot-runner temperatures above 285 °C. The processor should request carboxyl end-group values by titration and compare them over multiple lots.

Melting point and cooling crystallisation measured by ISO 11357-3:2018 also reveal thermal history. Bottle-grade PET typically shows a peak melting temperature of 245–255 °C on the second heat after erasing thermal history. A lower melting point or a broadened melting endotherm can indicate higher comonomer content or lower SSP conversion; a higher crystallisation onset on cooling can narrow the preform moulding window because the polymer crystallises too early in the gate or sidewall. A replacement grade should be tested under identical DSC heating and cooling rates, typically 10 °C/min, with the same nitrogen purge.

Pre-drying defects remain a primary cause of replacement failure on production lines. PET pellets require dehumidified air at −40 °C to −50 °C dew point and 160–180 °C for 4–6 h to reach a final moisture content below 0.003%–0.005%. At higher moisture, hydrolysis during plasticising reduces intrinsic viscosity by 0.02–0.05 dL/g within one production shift, generating acetaldehyde and surface haze. The replacement resin’s drying time to 0.005% moisture should not exceed the incumbent by more than 30 min unless additional drying capacity exists. On lines where regrind is incorporated at 10–30 wt%, the blend moisture and fines content require separate measurement; regrind with high fines can blind desiccant beds and create feed bridging. The dew-point sensor must be located at the hopper inlet, not the dryer outlet, to record the actual process air condition.

Preform Moulding, Hot-Runner Balance, and Acetaldehyde Generation

Injection preform moulding is the most sensitive stage to molecular weight and rheology differences. Barrel temperatures are typically maintained at 270–290 °C, hot-runner tips at 280–300 °C, and cooling water at 8–12 °C on high-cavitation moulds. PET screws commonly have 20:1–24:1 L/D and compression ratios near 2.5:1; excessive shear heating due to a non-optimised screw or higher-viscosity replacement can raise melt temperature above 300 °C and accelerate degradation. The valve gate must remain hot enough to avoid premature freeze-off but cool enough to limit gate crystallinity; gate defects appear as hazy, oriented regions at the preform base. A replacement resin with lower melt viscosity at the same IV may reduce hot-runner pressure by 2–5 bar and shift shot-weight balance across a 96-cavity tool. The processor should record cushion position, injection pressure integral, recovery time, and preform weight standard deviation over at least 30 min of uninterrupted cycling.

Acetaldehyde is formed by thermal-oxidative degradation and from vinyl ester linkages. In carbonated soft-drink preforms, headspace gas chromatography of ground preforms typically targets ≤5 ppm; water and sensitive beverage applications often impose ≤3 ppm. Colour is controlled by CIE L*a*b* on pellets or moulded plaques; clear bottle grades typically require b* ≤ 0.5 and L* ≥ 80. A replacement with higher catalyst residues may pass initial colour but develop yellowing after repeated processing or ultraviolet exposure. Accelerated colour testing on plaques held at 285 °C for 5 min is used to expose marginal thermal stability. If the replacement shows an acetaldehyde increase greater than 1 ppm relative to the incumbent under identical barrel residence time, the lot should be rejected or process adjustments made.

In the stretch-blow stage, the preform is reheated to 95–115 °C by infrared lamps, then stretched axially and inflated at 25–40 bar. The replacement resin must be compared for reheat absorption, which is influenced by catalyst metal type, comonomer content, and pellet colorants. A faster-reheat resin may lower required lamp power by 3–8% but can also produce uneven sidewall temperature and localised crystallisation. A slower-reheat resin may require longer cycle time or increased power and may fail to achieve proper axial elongation in the base and shoulder. Trial measurements should include preform surface temperature profile via infrared camera or embedded thermocouple, bottle sidewall thickness at 12 points, base clearance, and burst pressure. If the replacement shows a sidewall thickness standard deviation greater than 0.03 mm at the same stretch ratio, the process window is narrower than the incumbent and further adjustment is required.

When a Replacement Grade Reaches the Stretch-Blow Mould

Stretch-blow moulding exposes differences in crystallisation kinetics more directly than injection moulding. The preform is reheated to 95–115 °C and then stretched at a stretch ratio typically between 8:1 and 12:1 axial-to-radial. A replacement with lower isophthalic acid content may crystallise prematurely during reheat, producing hazy sidewalls, poor material distribution, and higher base stress. A replacement with higher diethylene glycol content may plasticise the amorphous phase and lower the preform reheat temperature, causing the body to overstretch while the neck remains too cold. Process adjustments can compensate for small changes: oven power, preform temperature, blow delay, and blow pressure are adjusted in small steps. However, if the acceptable reheat temperature range narrows below ±5 °C of the incumbent setpoint, the replacement grade is not process-equivalent. This threshold is frequently observed on high-speed rotary blow moulders running at 1,800–2,200 bottles/hour per mould; narrow processing windows cause intermittent defects such as pearlescence, base split, and top-load failures.

Pearlescence and stress whitening are early indicators of strain-induced crystallisation outside the intended temperature corridor. These defects are not visible in the preform and appear only after blowing. When they occur with a replacement resin, the processor should measure the preform temperature profile across the entire path, not only the surface midpoint; infrared sensors can be calibrated against the incumbent by trials at the same power settings. A sidewall thickness map should be generated at 12 or 16 points around the bottle circumference. A change in thickness standard deviation greater than 0.03 mm indicates that the replacement resin has shifted the elongation viscosity response. The bottle must then be tested for top load and burst according to the container specification; if these properties fall below the lower control limit, further blow-pressure or delay optimisation is required.

Food-contact equivalence requires lot- or grade-specific compliance statements rather than generic “bottle grade” claims. The replacement must have written confirmation against FDA 21 CFR 177.1630 for PET, EU Regulation 10/2011 migration test data for the intended food simulant, and REACH SVHC declarations. The certificate of analysis should state the standard designation, detection limit, and result for each parameter. Table 2 summarises the minimum compliance matrix for a PET replacement grade.