The Processing Engineer's Polycarbonate Start-Up Sheet: Dryer, Barrel, First-Off

Polycarbonate (PC) is processed as a bisphenol-A-derived amorphous thermoplastic with a glass transition temperature near 147 °C measured by ISO 11357-2:2020. Start-up reliability for injection moulding and extrusion is governed by three sequential controls: closed-loop desiccant drying to a residual moisture content at or below 0.02 wt%, barrel-temperature profiles that hold molten residence time below the carbonate-linkage decomposition threshold, and first-off testing that verifies molar-mass retention, impact strength, and optical clarity against virgin-moulding benchmarks.

For polycarbonate, residual moisture is not a surface contaminant but a chain-scission reagent. At melt temperatures above 260 °C, water attacks the carbonate linkage, producing bisphenol A chain ends, carbon dioxide, and a measurable loss of molar mass. Industrial desiccant dryers are therefore set to deliver air at a dew point of -40 °C or lower and a bed temperature of 120 °C for virgin pellets. A hopper air-flow of 3.0–3.7 m³/h per kg/h of polymer throughput prevents pocketing of damp granules; undersized blowers produce start-of-shift splay that disappears only after the first hour of production.

Moisture testing at the hopper inlet is performed by coulometric Karl Fischer titration per ISO 15512:2019 or by a moisture analyser calibrated against ASTM D7191-18. The measurement is taken after the dryer has reached steady-state return-air dew point, not from pellets sampled from an open gaylord. Sampling through a closed port with dry purge gas prevents false low readings from ambient re-equilibration.

Why is 0.02% Moisture the Upper Limit Before Hydrolytic Chain Scission Becomes Detectable?

Hydrolytic chain scission in PC follows pseudo-first-order rate behaviour above the glass transition once free water is present. Pellets equilibrated at 60% relative humidity can absorb 0.20–0.35 wt% moisture within 24 h. A hopper without dry-air purge allows re-equilibration even after off-line tray drying. The 0.02 wt% ceiling is not a cosmetic target; it preserves molar mass and prevents the MFR shift that occurs when carbonate linkages cleave in the metering zone. For a general-purpose grade with an initial MFR of 10 g/10 min at 300 °C and 1.2 kg per ISO 1133-1:2022, moisture above 0.04 wt% can produce visible splay, and the resulting first-off impact failure often occurs before tensile-yield changes are detectable.

Melt flow rate is the fastest start-up indicator because it tracks molar-mass reduction directly. If the first-off MFR exceeds the virgin lot value by more than 10%, chain scission is already significant, even when the part appears clear. A stable melt flow rate between 9 g/10 min and 11 g/10 min for a 10 g/10 min grade indicates that drying and melt residence time are within the start-up envelope.

Temperature-dependent hydrolysis studies consistently report accelerated chain scission with increasing melt temperature. A line stoppage at full barrel temperature is therefore managed by reducing the soak set point to 240–260 °C after a 10-minute alarm. Extended soak at 300 °C drives molecular weight loss and yellowing, and the barrel must be purged with virgin material before the first-off sample is taken.

Hopper Dew Point, Air Flow, and the Production-Scale Dryer Audit

Production-scale desiccant dryers used for PC are typically wheel-type units with return-air dew-point sensors mounted in the process air stream. The set point of -40 °C must be confirmed at the hopper inlet, not at the desiccant wheel outlet. Air-flow is set to 3.0–3.7 m³/h per kg/h of throughput; flow below 3.0 m³/h/kg/h leaves stagnant pellet zones and produces intermittent splay. Maintenance records for desiccant wheel rotation, regeneration heater amp draw, and aftercooler outlet temperature matter more than the digital display reading.

Dryer hopper residence time for virgin PC is 4 h at 120 °C; hopper residence below 3 h is a known cause of first-shot bubbles. The hopper inlet temperature is monitored with a thermocouple placed at the feed zone, and the return-air water load is logged at shift change. Overdrying at bed temperatures above 130 °C can fuse pellets and degrade surface additives, so the bed temperature is controlled to 120 °C ± 5 °C for standard unfilled grades.

When Regrind Fraction Exceeds 20%, First-Off Runner Handling Should Be Interrogated

Clean runner regrind from amorphous PC retains most of its molar mass, but every heat history adds hydrolysis and oxidative damage. Up to 20 wt% regrind is acceptable in general-purpose transparent applications if the regrind is dried for 6 h at 120 °C and the first-off MFR stays within the ±10% tolerance. Above 20 wt%, notched Izod impact and tensile elongation at break usually decline below datasheet limits. For tight-tolerance optical parts, regrind is excluded from the first injection sequence until the first-off haze value is below 1.0% per ASTM D1003-21.

No carbon dioxide scavengers or amine-based additives are introduced at the dryer because they alter the carbonate condensation equilibrium and generate plate-out on the mould.

Map the Barrel Profile Against the Shortest Possible Residence-Time Envelope

Unfilled PC injection-moulding barrels are set as follows: rear 260–280 °C, centre 270–290 °C, front 280–295 °C, nozzle 290–300 °C. The melt temperature measured by an immersion probe should remain between 295 °C and 310 °C for general-purpose grades; above 315 °C, carbonate-linkage decomposition emits CO₂ and produces yellow-to-brown discoloration. Screw-back pressure is maintained at 0.5–1.0 MPa to homogenise the melt without excessive shear heating.

Shot weight should be between 40% and 70% of barrel capacity when measured in general-purpose polystyrene. A shot weight below 40% extends melt residence and increases MFR drift; above 70%, unmelted core material and short shots appear. Total melt residence for unfilled PC is held below 6 min under start-up conditions.

Use of a 20:1 to 24:1 L/D general-purpose screw with a compression ratio of 2.2:1 to 2.5:1 is standard for PC; vented barrels are used only for extrusion where devolatilisation is needed. Peripheral screw speed is kept below 0.3 m/s on injection units; extrusion lines using 60 mm screws typically operate at 0.15–0.25 m/s to avoid local melt temperatures above 330 °C.

First-Off Testing Is Not Complete Until Melt Flow Rate, Notched Izod, and Haze Are Confirmed

First-off inspection starts after the injection unit has purged and at least five shots have been produced at stable cushion. Dimensional checks are not sufficient because PC parts can hold mould geometry while molar mass has already been lost. The first-off sample is checked for splay, brown streaks, black specks, gate blush, sink, and flash. Visual defects are recorded against the virgin-lot background sample.

Accelerated first-off drying audits are performed when the part displays splay only at the end of the flow path. This pattern typically indicates granule moisture at the centre of the hopper, not a barrel-temperature deviation. The hopper outlet is fitted with a dry-air purge line and the return-air dew point is verified before the next injection cycle.

Part ejection occurs after cooling below 135 °C; demoulding stresses above this temperature cause distortion. If a first-off part fails the haze or MFR window, the machine is not moved into production until the dryer dew point, hopper purge air, and barrel soak timer are audited against the start-up set points.