Thermal Welding Polycarbonate: Ultrasonic, Hot-Plate and Laser Compared

Polycarbonate (PC) is an amorphous bisphenol-A-based engineering thermoplastic with a glass transition temperature of 144–148 °C, a melt processing range between 280 °C and 320 °C, and a thermal conductivity of 0.19–0.22 W/m·K. Unlike semi-crystalline polymers, PC does not exhibit a sharply defined melting point; softening and viscous flow begin above the glass transition and continue over a broad plateau. This rheological behaviour sets the welding window for thermal joining. Polycarbonate absorbs 0.30–0.35 % water at saturation according to ISO 62, and residual moisture above 0.02 % hydrolyzes carbonate linkages at melt-face temperatures, generating carbon dioxide and reducing molecular weight at the joint. All thermal welding methods therefore start with pre-drying at 120 °C for 2–4 h and verification of residual moisture below 0.02 %. Polycarbonate also requires attention to notch sensitivity and a high coefficient of linear thermal expansion of 65–70 × 10⁻⁶ K⁻¹. Weld flash, sharp transitions, and internal stress concentrators can reduce impact performance more than short-term tensile strength. Consequently, weld factors determined from tensile coupons alone may overstate assembly performance if the joint is subject to impact or fatigue loading. Table 1 identifies the material properties governing the three processes.

What Limits Joint Strength in an Ultrasonic Polycarbonate Weld?

Ultrasonic welding deposits mechanical vibration at 20 kHz, 30 kHz, or 40 kHz into a local energy director, converting viscoelastic dissipation into heat. For PC, the near-field arrangement with the horn face within 6 mm of the joint is preferred because the polymer damps ultrasonic energy; far-field transmission through long unsupported walls generally produces variable melt initiation and weld factors below 0.60. A triangular energy director with apex angle 60° to 90° and height 0.25 mm to 0.50 mm is used for amorphous PC. Low amplitude of 20 µm to 35 µm at 20 kHz is applied for smaller components, while larger joints with wall thickness above 2 mm may require 45 µm to 60 µm amplitude. Weld time is typically 0.10 s to 0.50 s, followed by hold time 0.50 s to 1.0 s. Trigger force is set between 50 N and 500 N; production equipment with servo-controlled down speed is preferred because pneumatic-only systems may overshoot collapse displacement on low-stiffness PC parts.

The limiting factor is not a melting point but the narrow processing window between adequate squeeze flow and thermal degradation. PC melts over a broad range but does not become a low-viscosity fluid; energy director collapse occurs by displacement rather than free flow. Collapse displacement is commonly controlled between 0.20 mm and 0.60 mm. If collapse exceeds 0.60 mm, molecular orientation at the weld zone increases shrinkage stress and the notch-sensitive PC joint becomes prone to microcracking. Pre-drying is mandatory at 120 °C for 2–4 h to residual moisture below 0.02 %; otherwise steam generated at the interface creates porosity and reduces tensile strength tested under ISO 527-2. Weld factors for conditioned PC joints typically fall between 0.60 and 0.85 of the parent tensile strength, with the lower end associated with far-field configurations, inadequate hold pressure, or high moisture. On production lines, weld-strength variance of ±10 % across a batch is commonly traced to moisture uptake during storage or horn amplitude drift, not to base resin variation.

Joint design requires specific attention to PC’s high melt viscosity. Shear joints are possible but demand interference of 0.20 mm to 0.40 mm and can generate high normal stress at the part wall. Energy directors are generally safer for thin-walled unfilled PC because they concentrate heat and reduce the volume of molten material. Tooling material also influences performance. Horns made from titanium increase dimensional stability during extended runs, while aluminum horns may show temperature-related frequency shift after 2–3 h of continuous cycling. Ultrasonic stacks with booster ratio between 1:1 and 2.5:1 are standard; incorrect booster selection can overdrive the weld and produce localized burn marks at the horn-part interface.

Hot-Plate Sequences and the Melt-Layer Growth Window

Hot-plate welding relies on conduction from a heated platen, typically coated with PTFE to reduce sticking. For PC, platen temperature is normally set between 260 °C and 320 °C. The lower boundary is governed by the need to generate a molten layer with sufficient thickness for squeeze flow; the upper boundary is limited by oxidative discoloration, chain scission, and carbon dioxide generation. Contact pressure during the heating phase is kept between 0.1 MPa and 0.5 MPa, producing controlled melt-layer growth. Heating time is typically 5 s to 20 s, and changeover time must remain below 3 s to prevent cooling of the melt surface before joining. Joining pressure is set between 0.2 MPa and 1.0 MPa, with total melt displacement divided into heating displacement and joining displacement. On servo-electric hot-plate machines, displacement control is maintained within ±0.05 mm, which is critical for PC because excess displacement forces degraded surface layer into flash and reduces load-bearing cross-section.

The central process conflict in hot-plate welding of PC is the competition between melt-layer formation and surface degradation. A plate temperature above 300 °C shortens heating time but can produce yellowing and reduce molecular weight at the interface. At 320 °C, residence times greater than 15 s show visible degradation in unfilled PC. Therefore, production lines often select 280 °C and longer heating time up to 20 s for thick sections. Melt displacement of 0.30 mm to 0.80 mm per side is typical; below this range, incomplete intimate contact produces unwelded voids. Hot-plate welding is less sensitive to energy-director tolerances than ultrasonic welding, but it creates a visible flash bead that must be trapped in a designed flash channel. Residual moisture must again be below 0.02 %. Tensile weld factors for hot-plate-welded PC tested per ISO 527-2 or ASTM D638-14 generally range from 0.70 to 0.90, with the lower values observed on thick parts with uneven platen contact and high molded-in stress.

Production-scale limitations include platen temperature uniformity across surfaces larger than 500 mm. Multicavity hot-plate tooling may show ±5 °C variation between edge and centre, which shifts melt displacement locally and can alter weld strength by more than 10 %. PC’s high coefficient of linear thermal expansion of 65–70 × 10⁻⁶ K⁻¹ also demands cooling fixtures that maintain alignment during solidification. Clamping force release before surface temperature drops below 100 °C can cause joint distortion and residual stress. Heated platens with cartridge heaters and PID feedback are normally required; an overheated platen overshoot above 330 °C may cause PC to adhere, even with PTFE-coated surfaces, creating transfer strings and batch rejection.

When Through-Transmission Laser Welding Becomes the Only Feasible Route

Through-transmission laser welding, or TTLW, uses a diode or fibre laser at 808 nm, 940 nm, 980 nm, or 1064 nm to pass through an upper, laser-transparent PC component and deposit energy into a lower absorbing surface. Polycarbonate exhibits high optical transmittance in the near-infrared, so unfilled PC is not directly weldable without an absorbing component. The lower part is modified with carbon black at 0.3 wt% to 0.8 wt% or with an infrared-absorbing dye system; the latter permits clear-to-clear PC joints. Spot size is typically 0.5 mm to 3.0 mm, laser power 20 W to 200 W, and scan speed 1 mm/s to 100 mm/s. Clamping pressure is applied at 0.3 MPa to 2.0 MPa along the weld path. The interface reaches melt temperature within 0.05 s to 0.5 s of beam residence, producing a melt zone depth of 0.10 mm to 0.50 mm.

Laser welding is selected when ultrasonic energy cannot be transmitted, when flash is prohibited, or when vibration-sensitive installed electronics are present. The process is line-of-sight only and requires one transparent layer; pigments that scatter near-infrared radiation can reduce transmitted energy and widen the weld seam. In medical and diagnostic PC assemblies, laser absorbing dyes must meet biocompatibility requirements under ISO 10993-1, and food-contact polycarbonate grades must comply with FDA 21 CFR 177.1580 and EU 10/2011. Tensile weld factors for laser-welded PC tested according to ISO 527-2 typically sit between 0.70 and 0.90, but published data for specific clear-to-clear configurations is limited because absorber chemistry and beam profile affect heat distribution more than base resin flow.

The thermal conflict in laser welding is the localized high heat with a very short residence time. Peak interface temperature can exceed 300 °C within 0.2 s; if scan speed is too low or incident power density is excessive for the absorber concentration, PC foams and leaves carbonized marks. If clamp pressure is below 0.3 MPa, the expanding melt does not form a full contact weld and produces void channels. Production-scale diode laser systems with pyrometer feedback or path-integrated emission monitoring are used to maintain weld seam consistency, but closed-loop control on clear-to-clear PC remains difficult because emission signals are weak in the unfilled material. Clear-to-clear PC welding therefore usually requires a precise absorber coating, a high-resolution clamp, and validation on the actual part geometry.

Process selection for load-bearing PC assemblies is driven by joint geometry, cycle time, allowable flash, and moisture control. Table 2 compiles the comparative process windows and practical limitations observed on production equipment.

Validation of any PC weld should be performed on specimens conditioned at 23 °C and 50 % RH according to ISO 291. Short-term weld strength is measured with ISO 527-2 or ASTM D638-14, and impact-sensitive assemblies require notched or unnotched testing to ISO 179. Process capability on production lines is monitored by collapse displacement for ultrasonic welding, melt displacement and platen temperature for hot-plate welding, and laser power plus scan speed for TTLW. A weld factor below 0.60 generally indicates moisture contamination, insufficient melt displacement, or joint misalignment. Polycarbonate’s notch sensitivity means that a tensile weld factor alone does not predict performance under impact or fatigue; therefore process qualification should include dynamic mechanical testing at 23 °C and at the minimum service temperature of the assembly.

For clear transparent polycarbonate, ultrasonic and hot-plate methods avoid the absorber requirement of laser welding, but they impose higher mechanical stress or longer thermal exposure. For micron-scale optical alignment or solvent-sensitive electronics, laser TTLW at 940 nm with a low-power diode source is often the only viable route despite higher capital cost. Published comparative data across identical PC grades and specimen geometries is limited; equipment qualification therefore defaults to designed experiments involving three levels of moisture, three levels of collapse or scan speed, and weld factor determination under ISO 527-2. The test standard designation remains the primary guard against non-repeatable thermal welding claims on polycarbonate.