Polyethylene terephthalate (PET) is routinely characterised across the conversion chain because melt-phase esterification, polycondensation, solid-state polymerisation, drying, injection moulding, and stretch blow moulding each alter molecular mass distribution, crystallinity, volatile concentration, and chromatic coordinates. Testing PET plastic through IV, DSC, acetaldehyde and b-value methods therefore targets the four failure domains observed in production: hydrolytic chain scission, uncontrolled thermal history, sensory non-compliance, and yellowing. These measurements are anchored to ASTM D4603, ASTM D3418, ASTM F2013, and ASTM D6290 procedures rather than to a single universal specification. Resin suppliers may additionally apply ISO 1628-1, ISO 11357-3, or internal headspace gas-chromatography protocols. A pellet with acceptable intrinsic viscosity can still fail acetaldehyde migration limits if melt residence time is excessive, while a resin with compliant colour coordinates may contain insufficient molecular mass for preform impact resistance. The methods below are therefore treated as linked release criteria in bottle-grade resin supply and not as isolated laboratory checks.

Solvent Systems and Molecular Weight Correlation

The standard procedure involves dissolving a precisely weighed PET sample at 0.50 g/dL in a 60:40 phenol/1,1,2,2-tetrachloroethane solvent system, filtering any insoluble gel, and measuring the efflux time of the solution relative to the solvent in an Ubbelohde-type viscometer. The viscometer bath is maintained at 30 °C, and the single-point Billmeyer relation converts relative viscosity to intrinsic viscosity without requiring multiple concentrations. This reduces the exposure of the solution to moisture during lengthy multi-dilution protocols. Glass capillaries in an Ubbelohde design are preferred because the suspended-level bulb makes flow time less dependent on the hydrostatic head of the sample volume. ASTM D4603 is the controlling procedure for PET, while ISO 1628-1 provides the general framework for solvent and concentration selection.

Commercial PET grades are separated by intrinsic viscosity because the same polymer chemistry serves bottle, sheet, fibre, and strapping applications. Published supplier data for bottle resin commonly place intrinsic viscosity between 0.76 dL/g and 0.84 dL/g as the release range. Fibre and textile grades are often supplied at 0.55 dL/g to 0.65 dL/g. A low intrinsic viscosity of 0.60 dL/g may be sufficient for strapping but generally lacks the preform melt strength required for long-neck carbonated soft-drink bottles. The conversion from melt-phase polycondensation to bottle resin is carried out in continuous or batch solid-state polymerisation vessels, where pellets are heated under vacuum or a countercurrent nitrogen stream at 200 °C to 220 °C. The reactor residence time is adjusted to raise intrinsic viscosity while simultaneously reducing acetaldehyde; pellet surface area and the diffusion coefficient of the volatile determine the removal rate. The practical limitation of the capillary method is that copolymers containing isophthalic acid or cyclohexanedimethanol can alter solvent response and require calibrations that are not always provided in general-purpose specifications.

Differential scanning calorimetry is used on pellets, preforms, and sheet to resolve glass transition, cold crystallisation, melting, and enthalpy relaxation. ASTM D3418 and ISO 11357-3 recommend conventional or modulated DSC under inert nitrogen purge with recorded heating and cooling rates; a common screening rate is 10 °C/min. PET specimens typically show a glass transition near 78 °C, a cold-crystallisation exotherm between 130 °C and 150 °C, and a melting endotherm between 240 °C and 255 °C. Percent crystallinity is calculated from the melt endotherm and cold-crystallisation enthalpy using a reference heat of fusion of 140 J/g for fully crystalline PET. Modulated DSC is particularly useful for separating the reversing heat capacity of the amorphous phase from the non-reversing cold-crystallisation exotherm when the two events overlap in fast-cooled preforms.

On production lines, desiccant wheel dryers capable of delivering air at a dew point of -40 °C or lower are used to keep pellet moisture below 50 ppm; drying temperatures of 140 °C to 160 °C are typical for bottle resin. If residual moisture exceeds the 50 ppm limit, hydrolysis in the extruder barrel reduces intrinsic viscosity during melt processing, and DSC alone may not show the resulting molecular mass loss. Injection moulding tooling for preforms commonly operates with cooling water temperatures of 10 °C to 15 °C and a chilled core to quench the amorphous state. This prevents spherulitic crystallisation that would appear as haze and influence the subsequent colour measurement.

What Limits Are Placed on Acetaldehyde in Bottle-Grade PET?

Acetaldehyde is the degradation product most closely monitored in preforms and bottles for still and carbonated water because it migrates to the contents and imparts an apple-like off-taste. ASTM F2013 is the widely used static headspace gas-chromatography procedure for residual acetaldehyde in PET bottle polymer; it uses flame ionisation detection after a controlled sample equilibration. Resin producers typically report pellet acetaldehyde in µg/g, while bottle-side analysis may report µg/L in migration testing. Published resin specification sheets for water-grade PET commonly require pellet headspace acetaldehyde below 1 µg/g; preform limits vary with package volume but are often held below 5 µg/g to 7 µg/g for smaller water bottles. Sensory thresholds in water have been reported near 20 µg/L, which makes acetaldehyde control a direct quality constraint rather than a simple indicator of thermal abuse.

The headspace method requires that sample comminution be performed under conditions that minimise volatilisation and that calibration vials be prepared using certified acetaldehyde standards. Production-scale injection moulders often monitor barrel temperature, back pressure, and screw speed because acetaldehyde can be generated during plastication even when pellet headspace is acceptable. Low-compression screws for PET and reduced compression ratios are specified to lower shear heating. Hot-runner and hot-tip temperature uniformity across the preform mould also influences acetaldehyde distribution; a single hot runner running above 285 °C can raise preform acetaldehyde in one cavity while the pellet certificate remains unchanged. Solid-state polymerisation lowers residual acetaldehyde because the volatile is removed during prolonged reaction under vacuum or nitrogen flow; this is why bottle-grade resin is often post-polymerised rather than used directly from the melt finisher. Acetaldehyde scavengers or barrier additives are used in some packaging formats, but their load level must be checked against food-contact clearance because they can alter extractables and may interact with colour coordinates.

Yellowness in PET is quantified from tristimulus colour coordinates measured on a pellet or plaque spectrophotometer. ASTM D6290 specifies sample presentation, instrument geometry, and calculation for plastic pellets; the reported b* value is derived from the CIE L*a*b* system, where positive b* indicates yellow, negative indicates blue, and zero represents neutral. Typical release specifications for bottle-grade PET place pellet b* below 2.0, while premium water-grade resin may require b* below 1.0. Spectrophotometers use 45°/0° or d/8° geometry with CIE standard illuminant C or D65 and the 2° or 10° observer. Because pellet surface smoothness, crystallinity, and translucency affect the measurement, direct pellet readings are sensitive to sample drying and surface dust; visual colour and intrinsic chromophore content are often confirmed on compression-moulded plaques.

Measurements on compression-moulded plaques are less sensitive to pellet curvature, dust, and moisture haze than direct pellet-cell measurements. A specified sample thickness is required because the optical path influences the measured chromophore concentration and b* magnitude. Pellet crystallinity changes scattering; annealed pellets from the solid-state polymerisation reactor can have higher haze and altered b* readings without an actual change in chromophore content. Spectrophotometer calibration with traceable white and green tiles is performed before each series of colour measurements. The b* value is dimensionless, but colourimeters commonly report repeatability on a single pellet cell under specification-defined averaging. Thermal degradation products, catalyst residues, and acetaldehyde scavenger chemistry can all move b* in the positive direction; therefore, a resin should not be judged solely on pellet b* if the moulding line has changed drying time or melt temperature.

When Melt Residence Time and Drying Conditions Interact Across Four Release Tests

Production-scale trials on injection moulding machines with reciprocating screws and preform moulds show that the four release methods are highly coupled. A pellet delivered with intrinsic viscosity of 0.82 dL/g can lose 0.03 dL/g to 0.06 dL/g if drying is inadequate and the material enters the barrel at moisture above 50 ppm. The same condition increases headspace acetaldehyde after injection because hydrolysis and melt-phase degradation occur simultaneously. DSC of the resulting preform may show no major shift in glass transition but can reveal cold-crystallisation changes caused by altered molecular weight and residual moisture. Colour coordinates likewise respond to extended melt residence: barrel temperatures above 285 °C and recovery times beyond the machine supplier’s recommended range can shift b* upward by thermal oxidation. In this integrated context, a single certificate of analysis is considered insufficient for release; the control of each measured property is tied to the process window of a specific dryer, injection moulding machine, and preform tool.

Moulding trials carried out outside these boundaries generally produce preforms with acetaldehyde excursions and reduced intrinsic viscosity retention, and should not be released on the basis of a single pellet intrinsic-viscosity certificate alone. The operational limits for PET melt processing remain dependent on resin grade, package geometry, and machine size. Published data for this specific configuration is limited; therefore, each plant must establish its own response surface using the four methods as simultaneous release indicators rather than treating any single value as a universal acceptance criterion.