A novel approach to joining polypropylene - focused infrared welding

TWI Bulletin, November - December 1997

Ian is a senior project leader in the Plastics Joining Group at TWI. He has gained a wide knowledge of plastics joining processes and equipment and is responsible for running research projects and supplying training in plastics joining.

Focused infrared welding is a novel technique for the joining of thermoplastics, a process in an early stage of development which has yet to be fully exploited commercially. This study was undertaken by Ian Froment to examine the technique for polypropylene.

The benefits of focused infrared (IR) welding make it potentially an exciting alternative to hot plate welding. These benefits are summarised as:

• non contact heating
• efficient power consumption
• faster heat input than hot plate
• instantaneously controllable heat source

Equipment developed at TWI for focused IR welding is very similar to equipment used for hot plate welding. The two components to be joined are situated on tooling platens either side of a heat energy source. For focused infrared welding the heated tool is replaced by an infrared energy source which, by use of mirrors, is directed at the surface of the thermoplastic materials being joined. The high heat input and rapid response of the lamp equipment make the processing time fast. Using a lamp means that the heat source is only in use during the heating phase of the operation, giving rise inherently to a more energy efficient process. The sequence of operations is the same as for hot plate welding, ie a heating phase followed by a consolidation phase. There is no heating pressure on the thermoplastic parts during the heating phase. In fact, an air gap exists making this a non-contact process, which greatly reduces the possibility of contaminating the surfaces of the materials being joined.

A literature review of infrared heating highlighted a range of non welding industrial applications such as paint drying, adhesive curing and general materials heating ^[1] . Very little work has been carried out to investigate infrared for joining thermoplastics. Potente, Michel and Heil ^[2] , studied its use for welding thermoplastics and concluded that the process would solve a number of existing problems throughout plastics joining technology ie hot plate contamination, energy consumption and speed. Their equipment used direct heating of the thermoplastic surfaces with the infrared source. Branson ^[3] described a similar system where an infrared heater was mounted on a reciprocating arm placed between the weld interfaces. As the heaters moved back and forth across the materials surfaces, the interface temperature was monitored until the correct melting temperature was achieved. The arm was then withdrawn and the parts brought together.

This report describes the equipment, process and the experimental detail of work carried out to evaluate the focused infrared welding of polypropylene composite.

Principles of infrared heat sources

Electromagnetic energy has long been enhanced and exploited for welding thermoplastic materials. Table 1 shows the electromagnetic spectrum in relation to various welding processes. Infrared is a form of electromagnetic energy which spans the wavelengths 0.7µm to 400µm. At the short wavelength end, the boundary lies at the limit of our visual perception, in the deep red. At the long wavelength end it merges with the 'microwave' radio frequencies ^[4] . As materials heat up they emit infrared radiation. The wavelength at which the peak power density occurs changes as the material temperature increases. An example is the tungsten filament in a light bulb.

In an infrared heater, a tungsten filament encased in a quartz glass vacuum tube is heated to a very high temperature by passing an electric current through it. This allows high power concentrations that can achieve fast heating times and can be fully variable by reduction of the applied voltage. A coated aluminium mirror is used to focus and manipulate the infrared energy. Figure 1 shows a 2.5kW short-wave infrared lamp, indicating the position of the reflective mirror and tungsten filament. Reflectors are designed to match the infrared heat source and the intended application.

The most common reflector design for linear heating lamps is the parabola, which is capable of applying a uniform heating pattern within a reasonable range of focal lengths (eg 10-60mm). Generally, mirrors are coated with materials such as gold, copper or nickel to improve their reflective performance. Gold, if properly maintained, ie kept clean, can produce reflection efficiencies up to 97% in the infrared spectrum ^[1] . The more commonly used uncoated, polished aluminium reflectors are around 80% efficient, with some heat dissipated into the body of the lamp. Water cooling is generally used to prevent overheating of the infrared lamp unit due to this inefficiency.

Table 1: Electromagnetic spectrum in relation to plastics joining processing

Infrared Welding Trials

Materials and specimen form

Infrared welding trials were conducted on Procom, a polypropylene (PP) homopolymer produced by ICI, which is typically used in automotive applications such as fan blades, head lamp housings and belt covers. Its composition was Procom GC30H257, a 30% glass coupled grade (tensile strength 90MPa). Variations in infrared power and heating time were investigated to understand the effect of the principal process parameters. For this experimental work single lap shear samples were used with weld overlap dimensions 12.7mm x 25.4mm. Figure 2 shows the specimen geometry before and after welding. (Due to limited material availability only one sample was produced at each welding condition.)

Equipment and welding procedure

The infrared welding trials were conducted on a modified Bielomatik K2102 hot plate welding machine (Figure 3). The heating plate was replaced with a short wave infrared lamp (manufactured by Research Incorporated, type 5193-5). A triangular polished copper mirror mounted directly below the infrared lamp directed the beam to the material surfaces. Figure 4 shows a cross section schematic of the infrared lamp, copper mirror, and specimen tooling. The 2kW 240V single phase infrared lamp was connected to a variable rheostat allowing adjustment of the output infrared power between zero and one hundred percent. For these welding trials, the lamp voltage was varied between 160V and 225V. The output power of the lamp increases as the input voltage is increased. The lamp emitter (bulb) used was a tungsten filament in argon atmosphere enclosed in a 9.5mm (OD) clear quartz tube. This operates at approximately 2200°C at 240V input voltage with a spectral energy peak at 1.1µm ^[5] .

The weld cycle consists of a heating phase and a cooling phase. In the heating phase the infrared lamp is moved forward in between the two component surfaces to be joined, automatically switching on the power. The material surfaces are then heated for a pre-set time. During the heating phase there is no contact between heat source and component. When the heating time has elapsed the lamp is removed and the two surfaces of the component are brought together. An axial force is then applied to the components during this cooling phase. The welding equipment is used in automatic mode to sequence the welding operation through the heating and cooling stages.

For these trials, the samples for welding were restrained in the tooling using a vacuum. The tooling was made from an insulative material to minimise heat loss from the component during the heating phase.

Weld assessment

All joints were visually inspected at completion of the joining cycle for extent of bonding and flash and material degradation. The strength of the joints was established by pulling the lap shear samples using an Avery Denison universal tensile testing machine at a cross head displacement rate of 5 mm/min. A number of welded samples in this investigation failed in the parent material away from the weld area because the shear strength of the weld was greater than the tensile strength of the parent material.

Results

The results for all the welding trials are given in Table 2 . The effect of lamp voltage, shown in Figure 5 is for a 12.5sec heating time, 30sec cooling time and 0.3MPa cooling pressure. The results clearly show that as the lamp voltage was increased the lap shear strength of the welded joints increased. The joints produced at lamp voltages of 200V and above failed within the parent material away from the joint area (Figure 6). Shear strengths in this region were typically 4MPa. Welds produced at lamp supply voltages above 220V showed considerable material displacement from the joint area and had an unacceptable appearance.

Welds made below 190V had little or no joint strength. Visual examination of the failed weld surfaces showed an increase in the width of heating band as the lamp power was increased (Figure 7). The heating pattern also suggests that there was an uneven power density across and along the focused beam.

Effect of heating time

To study the effect of heating time, the lamp supply voltage was set to 200V, the cooling time to 30sec and the consolidation pressure to 0.3MPa. The results (Figure 8) show that as the heating time increased the lap shear strength of the welds increased. Welds produced with heating times greater than 14secs failed in the parent material away from the joint. The maximum strength achieved in samples which failed in the weld was 1.8MPa. The range of heating times at which welds could be formed was narrow - at 7.5sec no weld was made and at 14sec a significant amount of flash was generated with failure occurring in the parent material. Welds which failed in the joint showed similar uneven heating patterns as noted earlier.

General equipment observations

Prior to the welding trials, position of the infrared beam with respect to the sample surfaces was found to be difficult to align accurately. Various methods for monitoring the beam position were attempted with little success. These included use of plain paper and heat sensitive paper placed on the specimen tooling. Although not entirely satisfactory, the best alignment results were achieved by melting plastic samples held in the tooling using a vacuum. At maximum power the heating pattern still remained uneven. It is believed that this is a result of uneven surfaces on the aluminium reflector.

The reflective surfaces of the infrared lamp regularly became contaminated with fume deposit from the materials being welded, particularly with long heating times or high heat input. The lamp was therefore cleaned after each weld with soft tissue and MEK solvent to remove the deposit.

Table 2: Results of infrared welding trials on 30% glass coupled polypropylene

Discussion

Results of the study for focused infrared welding of glass coupled polypropylene showed that this material was readily weldable using this novel technique. Heating times for the material were shown to be short, at around 12.5sec. Variations away from this optimum time, however, caused problems: at 8-10sec joint strengths were low and at 14-17sec material displacement was substantial giving an unsatisfactory appearance. A similar effect was seen with lamp voltage where a narrow parameter band (200-220V) produced acceptable results ie good integrity with minimum flash. This was due to the very high intensity of the heat source.

As the heat source is voltage controlled the process lends itself to feedback control. To heat the weld interface to a given temperature, as with hot plate welding, the surface of the plastic could be monitored using a separate infrared detector. As the surface temperature approached the required level, the heat source power could be lowered or switched off. Using this level of control, it should be possible to maintain the surface temperature at a predetermined level and eliminate the difficulties experienced here of the narrow parameter window.

During the trials some contamination on the lamp mirrors was experienced. In production this would lead to a less efficient process and would extend heating times, if not controlled. The surfaces of the aluminium reflector will, at optimum performance, reflect around 80% of the infrared emitted from the bulb. Therefore 20% is absorbed by the lamp body and converted into heat, hence the need for water cooling. As the surfaces become coated with a film of fume deposit the efficiency is reduced further. However, this can be easily avoided by adding a positive flow of air through the lamp to blow away fume and prevent deposits building up.

Setting up lamp position with respect to the material surface was difficult. The process is non contact and relies on a beam of infrared, which is not always visible. Typically the position was set by heating the material surface at a very low lamp power (ie without melting the surface). Although it appeared that the best position was achieved by this method, the heating pattern at full power was uneven. To understand fully this behaviour of the heat source it would be useful to measure the power density across and along the beam to produce a beam profile. This would provide the information necessary to indicate the best position to achieve an even heat pattern.

Trials carried out were on simple lap shear joint geometries. Further work is required to modify the process for welding more complex joints. For example, to weld a square object, such as a hydraulic fluid reservoir, a complex mirror would need to be designed to manipulate the infrared beam. Alternatively, several infrared heaters may be used to heat directly the surfaces being welded. One advantage of the machine developed at TWI is its compatibility with a hot plate machine. If direct heaters, as opposed to using a separate mirror, were used then their design would have to be optimised to fit into the small space requirements ie between the tooling platens as in hot plate welding.

The peak power density of the bulb was at 1.1m infrared wavelength. To optimise further the heating process it would be desirable to tune the output wavelength of the lamp to match the absorbency characteristics of the material being welded. This may have to be achieved with a different bulb emitter.

Conclusions

This project investigated the novel focused infrared welding technique for thermoplastics. From this work a number of conclusions can be drawn.

The focused infrared welding process was demonstrated on 30% glass coupled polypropylene with successful results. Parent material failures were achieved with heating times of around 12.5sec.

The heat source offers controlability via electric voltage, and high intensity non contact heating to achieve the short cycle time.

Some contamination of the heat source was experienced but it is not seen as a major problem as with hot plate welding.

Care must be taken to ensure good beam alignment in order to achieve even heating across the entire weld area.

Complex joint geometries will require further mirror and heat source development.

Acknowledgements

The work presented in this report was funded jointly by Industrial Members of TWI and the Minerals and Metals Division of the UK Department of Trade and Industry. Acknowledgements are due to ICI Procom who supplied the materials.

References