Unfilled glycol-modified copolyesters derived from terephthalic acid, ethylene glycol, and cyclohexanedimethanol at substitution levels of 20–45 mol% on total diol are amorphous or slowly crystallising resins used in decorative edge banding and non-structural extruded profiles. The comonomer substitution suppresses crystallisation, lowers the melt transition relative to poly(ethylene terephthalate), and produces a melt in which apparent viscosity is controlled by molecular weight distribution, chain branching, and the segmental rigidity that determines flow activation energy. In edge banding extrusion, melt viscosity is the dominant factor in die pressure generation because thin-gauge flat dies operate with land gaps from 0.4 to 1.0 mm and the required pressure for uniform transverse distribution is generated by viscous resistance across the manifold and lip. In profile extrusion, the same melt viscosity determines the filling of complex internal channels, the pressure measured before the die plate, and the ability of the extrudate to expand against a vacuum calibrator. A detailed melt viscosity specification is therefore more useful than a single melt flow rate value for both applications, and the present document addresses the critical interactions between viscosity, shear heating, melt strength, and thermal stability under production-scale extrusion conditions.

Why Does Melt Viscosity at the Die Land Determine Edge Curl in Thin-Gauge Banding?

Edge curl in thin-gauge copolyester banding is caused by asymmetric residual stress resulting from one side of the web being frozen against a chill roll while the opposite side remains hotter and relaxes further. Melt viscosity controls this asymmetry through the shear stress imposed in the die land, because the die land is the last region where the polymer is forced into a defined cross-section before it exits as a free surface. In a slot die with a land length of 10 to 25 mm and a lip gap of 0.4 to 1.2 mm, the apparent wall shear rate under typical edge banding throughputs can reach 1000 to 2000 s⁻¹. At these shear rates, a glycol-modified copolyester with a power-law index of 0.55 to 0.75 still exhibits shear-thinning, but the remaining shear stress is sufficient to orient chains near the wall. The first surface to contact the casting roll retains much of this orientation, while the air-side surface relaxes during the time required for heat to be extracted through the web thickness. The resulting residual stress differential appears after slitting and application as longitudinal bow, lateral curl, or twist. A conditioning environment of 23±2 °C and 50±5 %RH according to ISO 291:2008 is typically used to compare curl sensitivity, and edge banding producers often specify a maximum bow of 1.5 mm per 1000 mm length for slit coils. Melt temperature stratification across the width of the die can amplify the effect by creating local viscosity differences of 8–15% when the flow activation energy is in the range of 55–65 kJ/mol. Production-scale troubleshooting on 45 mm and 75 mm single-screw extruders has identified die land temperature stratification as a more common cause of curl than resin chemistry alone. On a 75 mm, 33:1 L/D single-screw extruder operating with unfilled PETG for edge banding, the barrel temperature profile is commonly set from 190 °C in the feed section to 230–245 °C in the metering section, with adapters and die zones held within ±2 °C of the melt set point. A barrier screw with compression ratio of 2.2:1 to 2.6:1 and a low-shear distributive mixing section is preferred, because high compression and excessive shear can raise melt temperature by 10–15 °C above the final barrel set point at screw speeds above 80 rpm. The centre of the flow channel receives more shear work than the outside streams, so melt temperature at the die entry is not homogeneous; the associated transverse viscosity gradient cannot be fully corrected by a restrictor bar or flex-lip if the feedstock also varies in melt flow index from batch to batch. A gear pump placed between the screen changer and the die stabilises pressure, reduces surging, and provides a pressure reference for die design. The pressure before a flat die for thin edge banding may range from 80 to 250 bar depending on melt viscosity, die width, lip gap, and throughput, and the pressure fluctuation at constant throughput should be held below 2–3 bar to avoid visible gauge bands. Screen packs and breaker plates upstream of the die impose additional pressure drop and shear history; a typical screen pack for PETG edge banding may be 20/40 mesh with a breaker plate, producing an additional pressure drop of 30–80 bar at target throughput.

When High Viscosity Restricts Drawdown in Edge Banding Without Sufficient Melt Strength

Drawdown in edge banding is the simultaneous reduction in thickness and width that occurs between the die exit and the first chill roll, and it is limited by the elongational properties of the melt rather than by shear viscosity alone. A copolyester with a high capillary shear viscosity but a linear architecture may show weak strain-hardening in extension, causing rapid neck-in and web tearing when the haul-off speed is increased. Commercial unfilled PETG is often observed to draw at ratios between 10:1 and 30:1 at melt temperatures of 220–250 °C, but published data for the exact draw ratio at break of a specific edge banding grade are limited; the range is therefore used only as an order-of-magnitude indication. In thin banding with a target thickness of 0.8 mm and a die lip gap of 0.5 mm, the drawdown stress at line speeds above 50 m/min can exceed the local melt strength if the die exit viscosity is too high, and the failure appears as transverse tearing at the web edges. The melt strength is best assessed on a capillary rheometer equipped with a shaft or wheel haul-off, because the measurement combines the effects of shear viscosity, extensional viscosity, and melt temperature in one test. Neck-in is the reduction in width from the die exit to the cooled sheet; a typical edge banding die may have a width 2–4% greater than the final sheet width because of neck-in. The amount of neck-in increases with melt temperature and haul-off speed and decreases with melt strength. Draw resonance, which appears as alternating thick and thin sections along the web, is suppressed by a combination of a high-viscosity core, a low drawdown ratio, and an appropriate distance between die and chill roll. A three-roll horizontal stack mounted immediately after the die first nip is adjusted to compensate for the elastic recovery of the glycol-modified copolyester. If the melt viscosity is high enough to generate 150–250 bar die pressure at the target throughput, the extrudate may still fail to track the casting roll because excessive die swell creates an air cushion and reduces contact area. The polished roll surface used for high-gloss edge banding may have a roughness below 0.02 µm Ra, and only a low-viscosity surface layer of the extrudate can wet such a surface reliably without trapping air. In coextruded edge banding, a lower-viscosity copolyester skin may be combined with a higher-viscosity core so that the skin provides gloss transfer and the core resists draw resonance and neck-in. The core viscosity is often selected to maintain a drawdown ratio below the critical value for the specific line, while the skin viscosity is selected to ensure surface smoothness after slitting and after adhesion to a 18–25 mm furniture panel edge.

Capillary Rheometry, Melt Flow Index, and Intrinsic Viscosity Correlations for Copolyester Grade Selection

Rheological characterisation for edge banding and profile extrusion should include both low-rate and high-rate data, because a single-point melt volume-flow rate cannot capture shear-thinning or elastic effects. The melt volume-flow rate is frequently reported at 230 °C under a 2.16 kg load according to ISO 1133-1:2022; a commercial unfilled PETG extrusion grade may fall between 5 and 12 cm³/10 min. This measurement is performed at a shear rate below 10 s⁻¹, whereas the die land in edge banding generates apparent wall shear rates of 100 to 2000 s⁻¹. Capillary rheometry performed in accordance with ASTM D3835-16 at three temperatures permits calculation of the Arrhenius activation energy, and the measured shear viscosity at 240 °C and 1000 s⁻¹ for unfilled PETG is typically between 120 and 260 Pa·s, depending on intrinsic viscosity and moisture content. The following table summarises the methods and the processing-relevant ranges used to compare grades for edge banding and profile extrusion.

Parameter	Method or condition	Processing-relevant value
Melt volume-flow rate	ISO 1133-1:2022, 230 °C, 2.16 kg	5–12 cm³/10 min
Apparent shear viscosity	ASTM D3835-16, 240 °C, 1000 s⁻¹	120–260 Pa·s
Power-law index	ASTM D3835-16, 100–2000 s⁻¹	0.55–0.75
Flow activation energy	ASTM D3835-16, 220–260 °C	55–65 kJ/mol
Residual moisture after drying	ISO 15512:2019, desiccant drying	<0.02 wt%

Intrinsic viscosity measured in a 60/40 wt% phenol/tetrachloroethane solvent at 25 °C according to ISO 1628-5:2015 correlates more directly with zero-shear melt viscosity than with high-shear die viscosity; commercial extrusion grades are generally selected in the range of 0.70–0.80 dL/g. The correlation is not universal because branching and glycol modification alter the relationship between molecular weight and shear viscosity. A branched copolyester can have the same melt volume-flow rate as a linear grade yet exhibit higher melt strength, lower die swell, and better sag resistance in a calibrator under the same melt temperature. For rigid profile extrusion of glycol-modified copolyester, the shaping die is more complex than a flat edge banding die because the melt is split around internal mandrels and must re-form into a closed or open profile. Melt viscosity controls the relative flow between thick wall sections and thin ribs, since the volumetric flow through a rectangular channel varies with the cube of the gap; small local viscosity differences can therefore generate large wall-thickness differences. The calibration unit downstream is commonly operated at a vacuum of -0.2 to -0.8 bar relative to atmosphere, and the extrudate must press against the calibration surfaces while retaining enough heat to conform to the internal mandrel. A melt that is too fluid at the calibration entry wets the tooling too quickly and loses surface gloss, while a melt that is too viscous does not collapse sufficiently to form sharp corners and may retain die-flow lines. Haul-off speed is matched to melt output so that the profile remains under slight tension, but excessive tension produces molecular orientation and increases the tendency of the profile to warp when later annealed at temperatures between 60 and 80 °C.

Thermal Degradation Limits in Glycol-Modified Copolyester High-Shear Processing

Thermal degradation of glycol-modified copolyester competes with shear heating and can be accelerated by residual moisture, by excessive residence time above 250 °C, and by local high-shear zones in the melt pump or die. Chain scission reduces molecular weight and produces a permanent fall in melt viscosity that is not recovered when the melt is cooled. In continuous production, die pressure decays gradually at constant throughput if the feed moisture content is above the recommended limit or if the melt temperature at the die exceeds 250 °C for more than 3 min. A melt temperature limit of 240 °C at the die is commonly used for continuous extrusion of unfilled PETG, with excursions to 250 °C permitted only for short residence times. Degradation is indicated by yellowing, a reduction in intrinsic viscosity of more than 0.05 dL/g from the dried pellet value, and a fall in die pressure at constant screw speed. In profile extrusion, the same chain scission appears as sag in the calibration tank and loss of the dimensional tolerance required for the final profile. Moisture control is necessary before extrusion because glycol-modified copolyesters are hygroscopic and undergo hydrolysis at melt processing temperatures. Desiccant drying at 65–75 °C for 4–6 h with a dew point of -40 °C or lower is recommended to achieve a residual moisture content below 0.02 wt%. A return-air aftercooler is used on production hopper dryers to prevent moisture condensation in the loader, and regrind with higher surface area may require longer residence time or a reduced regrind fraction. The use of hot regrind above 40 °C mixed non-uniformly with virgin pellets can produce local moisture pockets and viscosity differences that appear as surface defects in edge banding and as dimensional instability in profiles.