Specification of polyethylene terephthalate (PET) sheet and thermoforming grades has historically defaulted to a single solution-viscosity number, intrinsic viscosity (IV), measured in a 60/40 w/w phenol/1,1,2,2-tetrachloroethane solvent at 30 °C according to ASTM D4603-18 or ISO 1628-5:2015. The grade-selection problem described by the phrase “PET Sheet and Thermoform Grades: Why Sag Resistance Beats IV Here” is not a rejection of IV but a recognition that sag under radiant heat is a melt-state phenomenon governed by extensional deformation, chain architecture, and sheet thermal history. Amorphous PET must be heated above its glass-transition temperature of approximately 70–80 °C before forming; at these conditions the sheet is soft enough that gravity alone produces measurable deflection. A converter can receive two lots with identical pellet IV and still observe one lot sagging severely in the oven while the other forms cleanly. The difference lies in parameters that IV does not record: molecular weight distribution, branching, moisture-induced chain scission during extrusion, and sheet thickness uniformity.

Is Intrinsic Viscosity a Reliable Batch-Acceptance Criterion for Thermoforming?

Intrinsic viscosity is measured in dilute solution at concentrations below overlap concentration, where polymer chains are isolated. The test is repeatable and standardized, but it is not a rheological experiment. For linear PET, zero-shear melt viscosity has a power-law dependence on molecular weight approaching an exponent of 3.4; this relationship is useful for extrusion but does not directly predict sag. Sag involves low-rate extensional flow at approximately 0.1–1 s⁻¹ and is sensitive to strain hardening, a property that dilute-solution IV cannot detect. Two linear grades with identical 0.80 dL/g may differ in diethylene glycol content, carboxyl end groups, or catalyst residues; each of these influences hydrolytic stability and melt rheology without changing the IV solution value. Furthermore, IV loss from pellet to sheet can be 0.02–0.06 dL/g under poor drying or excessive melt temperature, so pellet IV is not the same as sheet-melt IV at the die lip.

Batch release based on ASTM D4603 alone also misses melt-pressure stability at the gear pump and transverse sheet thickness variation. A sheet extruded on a 90 mm barrier screw with L/D 30:1 and a melt pump can maintain ±0.1 MPa pressure variation when the melt is homogeneous; a wet or thermally degraded melt produces pressure swings above ±0.5 MPa, which appear at the die as thickness variation. Such sheet will sag unevenly regardless of the IV certificate. Therefore IV should be an identity marker, not a sole forming-performance gate.

Melt Strength Arises From Chain Architecture, Not Solution Viscosity Alone

Melt strength in PET can be increased by long-chain branching induced with polyfunctional comonomers or reactive modifiers. The effect is distinct from molecular-weight increase. Extensional rheometry using a Sentmanat extensional viscosity fixture on a rotational rheometer records transient extensional viscosity that rises above the linear viscoelastic envelope at Hencky strain values above 1.5 in branched systems. This strain hardening stabilizes the sheet against gravity and suppresses local thinning. A branched PET at 0.72 dL/g can therefore exhibit less sag than a linear PET at 0.84 dL/g when both are heated to 140 °C. The quantitative separation depends on the extent of branching, the moulding temperature, and the sheet’s moisture history. Published data for this specific comparison is limited, but chain architecture is an accepted variable in melt strength modification.

The practical consequence is that a grade with lower IV can outperform a higher-IV grade if the lower-IV grade has higher extensional strain hardening. Static oven sag tests capture this; solution viscosity does not. This is the functional meaning of the sag-resistance-over-IV rule for PET sheet.

Hydrolysis during predrying and melt processing remains the dominant source of IV deviation between pellet and sheet. Amorphous PET pellets are hygroscopic and must be dried to a moisture content below 50 ppm; this is commonly achieved with a desiccant dryer delivering air at dew point −40 °C or lower. Drying at 160–180 °C for 4–6 h is typical, but hopper residence time must be limited to avoid pellet sticking and local crystallinity. Transfer lines should be insulated and purged with dry air; otherwise pellets regain moisture quickly at 50% RH and 25 °C. When melt temperature exceeds 290 °C, hydrolytic chain scission accelerates. PET exposed to moisture above 100 ppm can lose 0.03 dL/g or more in a single pass through a 30:1 L/D single-screw extruder. The melt-pressure trace at the gear pump then becomes the visible signature of hydrolysis: pressure swings, surging, and die-lip thickness deviation.

Moisture analyzer calibration is part of this evidence chain. A loss-on-drying instrument at 160 °C may not resolve the final 10–20 ppm of water in PET with sufficient precision for high-IV resin. Karl Fischer coulometry or a dew-point probe at the dryer outlet is preferred. Resin suppliers specify drying parameters for each grade; processing outside the specified dew point or air flow voids the IV retention data.

When Radiant Oven Sag Testing Outperforms IV as a Grade-Release Gate

Sag testing on a clamped amorphous PET sheet provides a direct measurement of the failure mode that disrupts thermoforming. A laboratory fixture may clamp a 200 mm × 300 mm sheet on four edges and heat it until the surface reaches 130–150 °C, measured with an infrared pyrometer calibrating for PET emissivity of 0.90–0.95. The vertical displacement at the sheet center is recorded at 30 s intervals. Because no single ASTM or ISO sag procedure governs all tooling geometries, the pass/fail threshold must be calibrated to the production thermoformer and its plug-assist arrangement. The sag test integrates molecular weight, branching, crystallinity, moisture, sheet thickness, and heater-bank uniformity into one operational response.

On a production line, an optical grid projection system can map sag contour across the sheet. Early sag in a specific lane indicates local thinning, a heater zone imbalance, or non-uniform amorphous orientation from the roll stack. IV cannot detect these local variations. A sheet with 0.80 dL/g pellet IV but uneven roll-stack orientation will sag asymmetrically even if the molecular weight is uniform. Conversely, a branched grade that shows higher sag resistance often displays a more uniform sag profile because strain hardening activates where deformation is greatest.

Process-induced crystallization is the second variable that decouples sag resistance from IV. Unmodified amorphous PET crystallizes slowly between 120 °C and 170 °C, with a maximum rate near 170–180 °C. As crystallization develops, the sheet develops haze, locally stiffens, and may resist sag not because of melt strength but because it is no longer fully amorphous. This is not a forming advantage; crystalline domains block uniform stretch and reduce detail replication. Thermoforming-grade PET frequently contains comonomer structures such as isophthalic acid or 1,4-cyclohexanedimethanol to slow crystallization and widen the forming window. These comonomer changes affect melt rheology at equivalent IV. Differential scanning calorimetry per ASTM D3418-15 at 20 °C/min should be used to verify the crystallization onset for incoming sheet when processing window is critical.

Small-scale sag fixtures must also control sheet emissivity and air currents. A draft of 0.5 m/s across the sheet surface can reduce sheet temperature by several degrees and alter sag. Closed fixtures with an optical window are used for displacement measurement. The measured sag depth is therefore not an absolute material constant; it is a relative index for the defined sheet thickness, clamp opening, heater output, and air flow.

Thermoforming Window, Haze, and Regrind Stability

The practical forming window for amorphous PET sheet is bounded at low temperature by incomplete plug penetration and at high temperature by haze, sticking, and excessive sag. For thin-gauge roll-fed forming, sheet surface temperatures are commonly held between 125 °C and 155 °C; thicker sheet requires longer soak time and more sag control. Heating time scales roughly with the square of sheet thickness: a 0.5 mm sheet may require 10–20 s, whereas a 1.5 mm sheet may require 60–120 s under ceramic or quartz emitters. Sag increases with the square of unsupported span, so a wide forming area demands greater melt strength or reduced sheet temperature.

Sheet temperature uniformity is measured by a scanning infrared line pyrometer before the mold closes. A transverse spread above ±3 °C across a 600 mm wide sheet often produces uneven sag and part wall thickness variation. The sheet thickness profile after the polishing stack should be held within ±0.03 mm for a nominal 0.80 mm sheet; wider variation creates local sag under the same heating conditions.

Plug-assist thermoforming adds mechanical pre-stretching before final pressure or vacuum forming. Plug material and temperature affect the local cooling of the sheet, and a sheet that sags excessively may contact the plug prematurely. Heated aluminum or syntactic foam plugs are often controlled at 120–140 °C; contact with a low-sag-resistance sheet at these temperatures can create premature orientation and wall-thickness imbalance. The interaction between sag and plug is one reason why IV-based acceptance fails: a high-IV sheet that sags more can be more difficult to form than a lower-IV branched sheet that holds its shape until plug contact.

Regrind introduces another operational boundary. Post-industrial trim from PET sheet is routinely reintroduced at 20–40 wt%, but each heat history reduces IV and increases carboxylic acid end-group concentration. Acid end groups accelerate hydrolysis on the next extrusion pass. A sheet containing 30 wt% regrind may exhibit lower sag resistance than its measured average IV would predict because the regrind fraction has broader molecular weight distribution, higher moisture affinity, and different crystallization kinetics. Sag testing on final sheet catches this shift; pellet IV on blended feedstock may not.

Regulatory compliance for food-contact PET sheet in the United States is governed by 21 CFR 177.1630; in the European Union, regulation EU No 10/2011 applies with specific migration limits for monoethylene glycol and antimony-based catalysts. These constraints do not control sag resistance, but they restrict molecular design. Branching modifiers and catalyst systems must be cleared for the intended contact use, and antimony residues above 300 mg/kg as antimony may require migration verification under the appropriate food simulants. Chain-branching modifications should not elevate acetaldehyde above the taste-odor threshold of approximately 10 µg/L in water-sensitive applications; for sheet packaging, the limit is product-specific.

Operational boundaries include melt temperatures at the die body below 300 °C, residence time at melt temperature below 15 min, and avoidance of copper-based stabilizers or certain amine additives that accelerate PET degradation. If higher sag resistance is needed, branching modification and sheet temperature reduction are generally preferred over raising IV, because a higher-IV linear resin increases extruder torque and shear heating and can reduce throughput on a 90 mm single-screw line by approximately 5–15% relative to a branched lower-IV grade. Published data for this specific configuration is limited. When the receiving specification pairs ASTM D4603 pellet IV with a sag-depth curve generated at fixed sheet surface temperature, heating time, and clamp geometry, the release decision is made under forming-relevant conditions. Without those fixed conditions, a sag depth of 25 mm on one line is not transferable to another.