Bewertung und Optimierung laserbasierter Fertigungsprozesse bei der Herstellung hocheffizienter Solarzellen
Baier, Tobias; Poprawe, Reinhart (Thesis advisor)
Dissertation / PhD Thesis
At the time this work was started, the dominant method of designing production equipment was to develop a suitable laser process and to integrate it into industrial-suited machinery. In order to shorten the time needed for such machine development, theoretical models simulating the interaction between laser and material during the laser pulse were defined. The basics for these models are simplified thermodynamic relationships. With given laser parameters, the effect on the material and the resulting material modification can be resolved temporally and spatially. This allows for theoretically evaluating a multitude of beam sources within a short time. This way only pre-approved laser sources will undergo real testing in laboratories. As an example, the model shows how laser intensity and pulse duration effect the ablation rate during the drilling process of EWT solar cells: a further increase of pulse energy does not directly lead to augmented ablation rates, however, the combination with a beam splitting approach does. Results like this led to understanding the interconnection between a specific laser source and the conduct of processing material with it. Additional laser processes were modelled as well, such as selective laser doping and laser melting as used for laser fired contacts. In the latter case, the theoretical model correlates laser and optics parameters with melt diameters of molten aluminium and silicon. Either one has deep impact on the solar cells contact resistance and series resistance. The theoretical approach offers to evaluate physically suitable beam sources within minutes. The results of the aforementioned models were applied in two ways, one simulating the physical, and the other simulating the economical effects on real machinery. In order to know the machines productivity and competitiveness, the gross throughput and yield are crucial factors. In addition to simulating the material modification, the models can be used to calculate the time needed for achieving the specific modification. The process time has deep impact on the production throughput and defines the machines cost efficiency. The theoretical model also points out if parallelization schemes like beam splitting have positive effects as well. Further, this information about beam source, parallelization, and processing conduct is needed for the mechanical design of the machine. This method of machine development has led to two different real machine layouts. One is equipped with moving optics and the other features scanners. For either one, beam splitting approaches were defined and devices for beam monitoring and quality assurance were developed. The moving optic concept features a unit with beam shaping, beam splitting and beam positioning in one compact piece. The scanner based solution uses a fast beam switch with integrated beam shaping. Depending on the chosen laser process, the machine is equipped with the most suitable combination of laser sources, focussing optics and monitoring devices. Finally, a cost-benefit-analysis is used to evaluate and rate the different machine concepts regarding productivity, cost efficiency and competitiveness.