JEC Group have brought together the international community of composites leaders and executives in our Composites Circle as an unique networking opportunity to meet with both peers and future partners.
Over the past couple of years, the use of composite materials in vehicle construction has increased significantly and is expected to continue to grow in future.
(Published on August-September 2005 – JEC Magazine #19)
BY DR. HARTWIG LOHSE, EUROPEAN TECHNICAL MANAGER TRANSPORTATION, ASHLAND DREW AMEROID DEUTSCHLAND GMBH
Plastics offer advantages not only in regard to weight reduction but also in regard to design freedom, corrosion performance and the like. Despite their advantages, plastic composite parts are used mostly in low-volume applications. In many cases, the use of plastic composite parts requires a joining process as part of the manufacturing process. Bonding with 2-part epoxy, polyurethane or acrylic adhesives is the only applicable joining technology, in particular for thermoset composite parts. Although the adhesive industry has developed adhesives offering a relatively long open time in conjunction with a fast cure response at elevated temperature, the cross-linking of an adhesive using today’s state-of-the-art heating technologies does take a few minutes which, in many cases, is too slow for high-volume applications.
To reduce the cycle time, a faster heat-up of the adhesive is required. Generating heat inside the adhesive can eliminate the slow, timedetermining step of transferring heat through the (plastic) substrates. Adhesives can be formulated in a way that polar groups as part of the polymer structure respond quite fast to microwave by generating heat. Unfortunately, the amount of heat generated is highly dependent on the thickness of the adhesive, resulting in cold or hot spots even with just slight differences in adhesive thickness.
In order to achieve a uniform temperature distribution, a novel adhesive system is under development. Special additives that respond to electromagnetic radiation by generating heat are the key for this new system. Due to pseudo-ferromagnetic hysteresis, specific susceptors in the adhesive couple very effectively with electromagnetic radiation and generate controllable heat in the adhesive (fig.1). The required frequencies depend on the nature of the susceptor.
Besides the advantage of offering significant opportunities for cycle time reduction relative to conventional heat-curing methods, the heat is generated in the adhesive itself without the need of transferring through the substrate. This results not only in a faster heat-up and cure of the adhesive, but also in a lower temperature level the substrate itself is exposed to. This is very beneficial in regard to the resulting surface quality, especially if thermoplastic materials are bonded.
Three different technologies are currently under evaluation at the Ashland laboratories. A common feature is that specific particles in the adhesive formulation interact with electromagnetic radiation by generating heat inside the adhesive.
Ferromagnetic particles (ferrite)
Such materials are either metals, metal alloys or metal oxides. Ferrite particles are in a state of magnetic saturation, which results in a high coupling to an irradiated high frequency magnetic field and a very good energy adsorption. This means fast heat-up. Typical frequencies are in the 5–10 MHz range. Ferrite particles possess a built-in protection against overheating. As soon as the so-called Curie temperature is reached in the bonded joint, the ferrites lose their magnetism and electromagnetic radiation is no longer absorbed; the temperature ceases to rise, and hot spots will not appear (fig.2).
In addition to the above-mentioned properties, nanoscale ferrite particles of about 10nm feature a sharp absorption line for electromagnetic radiation in the range of several GHz, identical for each particle. By applying an additional magnetic field, this can be adjusted to the officially allowed ISM frequencies of 915 MHz or 2.45 GHz.
To assure good colloidal dispersion of the ferrite particles in the adhesive and to prevent the formation of larger agglomerates, each single ferrite particle is coated with a surfactant shell, a few nanometres thick.
Nanoscale iron oxide in silica matrix
Iron oxide domains embedded in a ceramic silica matrix manufactured by direct flame synthesis are inherently stable, clean and dry powders, similar to conventional fumed silica products. They can be easily added to adhesive formulations. They couple quite effectively with electromagnetic radiation in the 200-300 kHz range, generating the desired heat. Depending on their nature, the Curie temperature is in the range of 350 to 550°C; therefore, they do not feature a built-in temperature protection. The heat rate and the final temperature can be adjusted by variations in the HF power. By reducing the HF power at a certain point of time, both a fast heat-up and a secure maximum temperature can be achieved (fig.3).
All three of the novel technologies mentioned above were evaluated in lab-scale trials that focused on lap shear testing of commercial SMC grades. Durability testing was carried out for all adhesive formulations included into the Command Cure test series. Lap shear testing at ambient temperature, 80°C, -30°C, and after different common accelerated aging tests has been carried out. In all cases, the adhesive remained stronger than the SMC-substrate; in other words, fibre tear was the only observed failure mode.
Test results on conventional ferrite particles
The commercially available equipment for induction curing of ferrite containing adhesives is shown in figure 4. The 6.8 MHz electromagnetic radiation generated by the induction generator is coupled into the adhesive via a coil designed according to the part’s design. In most cases, this will be incorporated together with control units and a heat exchanger for the generator into a press that keeps the parts to be bonded in position as the adhesive cures.
During the test series, the adhesive temperature in the bondline was monitored using an IR-camera. Figure 5a shows a typical time-temperature graph for a ferrite containing epoxy adhesive in comparison to one achieved with a conventional state-of-the-art hot air impingement fixture. Figure 5b shows the infrared picture of a bonded lap-shear test specimen after 20 seconds of exposure to electromagnetic radiation. It is obvious that induction heating is generating the heat in the adhesive, and is therefore offering great advantages in regard of heat-up speed. The maximum temperature is achieved in less than 30 seconds, and overheating is protected by the capped curie behaviour of the selected ferrite particles. In order to determine the adhesive strength after such short cure cycles, lap-shear test specimens were exposed to electromagnetic radiation for 30 and 60 seconds, followed immediately (within 120 seconds) by lap shear testing at 10 mm/minute. Depending on the actual adhesive formulation, after just 30 seconds of cure time, about 0.6 to 1.1 MPa initial strength with cohesive failure of the adhesive was achieved, and after 60 seconds, 1.4 to 1.8 MPa. This strength is generally sufficient to allow further handling and processing of the bonded parts.
Test results on nanoferrite particles
The microwave device developed at SusTech GmbH consists of one or several hollow conductors that are completely integrated into the assembly equipment. The hollow conductor is a channel made from aluminium, through which the microwave is spread along the bondline. This offers the advantage of relatively low investment costs and a highly efficient, concentrated energy supply. Only those regions of a part to be bonded are irradiated by microwaves, in which the adhesive has to cure. Figures 7a and 7b show the unit as it was used for the initial lab-scale trials. The heat-up rate achieved is very similar to the already-reported result for the induction cure trials. With a resulting SMC surface temperature of about 140°C after exposure to 1.6 kW microwave at 2.45 GHz for 60 seconds, the immediate strength of a 2-part epoxy adhesive was about 1 MPa, again with mainly cohesive failure, but already some traces of fibre tear. One hour after cure, full fibre tear was observed.
Test results on nanoscale iron oxide particles in silica matrix
The bonding device is relatively similar to the one used for induction bonding. A relative simple lab unit working at 280 kHz was used to run initial trials on lap-shear test specimens. The lap-shear test specimen was placed in an induction coil for exposure to electromagnetic radiation. As these particles do not offer built-in protection against overheating, the bondline temperature was controlled by the power input. The tests were carried out on a 2-part polyurethane adhesive, under similar conditions as for the epoxy-based adhesives. Again, a fast strength build-up resulting in an immediate strength of about 0.6 to 1.0 MPa was achieved, and mainly full fibre tear was observed one hour after bonding.
Comparison of the three technologies
Of course, all three technologies have their specific advantages and disadvantages. Common to all is the really fast resulting strength build-up, due to the fast heat generation in the adhesive joint. As all the testing referred to in this paper was carried out at bonding times of 60 seconds and sometimes as little as 30 seconds, it seems very likely that fixture times of less then a minute are achievable for real parts like composite spoilers, hoods or lids. Such parts manufactured using today’s state-of-the-art hot air impingement technology require fixture times of several minutes.
An advantage of nanoparticles is the small quantity needed to get the desired effect. Just a few percent are required to achieve the fast heat build-up. Conventional ferrites need to be added in much higher concentration. With an increasing amount of susceptors added to an adhesive formulation, the risk of a negative effect on the adhesive performance and on recyclability will rise. In contrast to conventional ferrites made from milled sintered powders, nanoscale particles do not tend to sediment, nor do they cause abrasion in the application equipment. Evaluation of the surface quality of bonded panels, carried out on adhesives containing conventional ferrites with a loading of about 40–50 wt.%, resulted in a higher risk of bondline read-through effects. Extending the trials to adhesives containing nanoparticles will show if the read-through effect is caused by the high loading or is a result of the fast heat-up. From a cost perspective, the addition of ferrites will increase the cost of the adhesive. The cost increase will have to be weighed against the benefits of reduced cycle time and improved productivity as well as potentially reduced equipment and energy costs. As nanoparticles are not commercially available yet, the price for such materials is expected to be higher than for conventional ferrite particles. On the other hand, the required amount is, at just a few percent, relatively low compared to the high amount of less-costly conventional ferrite particles required to achieve a similar effect. At the bottom of the line, the resulting cost difference is expected to be marginal.
As a responsible care company, Ashland has kept a close eye on potential EH&S issues. Although initial trials on nickel oxide containing particles gave promising results, such ferrite particles have been banned from further investigation because nickel oxide is critical from a health point of view. We are now focusing on nickel oxide-free materials. As it is known that nanoparticles in some cases offer performance which is not known from their larger equivalent, EH&S studies on nanoparticles are monitored carefully regarding any adverse health effect. Regarding the equipment to cure the novel group of adhesives, effective shielding is required to protect the workers from electromagnetic radiation. The lower the frequency of the electromagnetic radiation, the less shielding is required. Therefore, nanoscale iron oxide particles in silica matrix cured at about 280 kHz show benefits compared to conventional scaled ferrites (about 8 MHz) and nanoferrite particles (2.45 GHz). On the other hand, the design of the hollow conductor can be done in a way to avoid microwave leakage.
The results of the tests carried out so far demonstrate that cross-linking of reactive 2-part adhesives containing specific susceptors can be significantly accelerated by electromagnetic radiation. This will result in a significant cycle time reduction, from several minutes to about 60 seconds or even less. A first cost estimation for a high-volume trunk lid of about 120,000 parts per year showed that the new technology is at least competitive with the current existing technology. Costs will go down as the consumption of nanoparticles increases, making the process even more cost attractive and suitable for high-volume applications.
The next step in the development work will be further fine-tuning of the adhesive formulation, in regard not only to the selected susceptors but also to the kinetics of the cure reaction. A further look at other effects like read-through needs to be done as well. This will be followed by scaling up the process from lap-shear test specimens to complete automotive parts. Evaluating both the bondline thickness on the heat generation and the even distribution of the electromagnetic radiation into the adhesive might require some additional development work to be carried out in cooperation with tooling companies. By adding two different susceptor types responding to different frequencies, the step from bonding on command to disbonding on command may be achievable. The cure of the adhesive during the bonding step is accelerated by exposure to a specific frequency that couples with one of the two kinds of susceptors generating heat. If the part needs to be de-bonded later on – perhaps at the end of the car’s lifecycle – the second kind of susceptors will be activated by electromagnetic radiation of a different frequency, resulting in a higher bondline temperature sufficient to soften the adhesive and therefore to easily separate the bonded parts.
- Sauer, Sppiekermann, Cura; Accelerated Curing with Microwaves and Nanoferrites – Adhesion Adhesives & Sealants extra 2004/2005, page 48ff - Gutsch, Mühlenweg, Krämer; Tailor-Made Nanoparticles via Gas-Phase Synthesis – small 2005, 1, No 1, page 30ff - Zimmermann, Kröll, Pridöhl; Bonding and Debonding on Command – Paper presented at the Degussa for Automotive conference – Innovation Day 2004