Rapid thermal processing for production of chalcopyrite thin films for solar cells: design, analysis, and experimental implementation
Date
2016
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The direct conversion of solar energy to electricity, or photovoltaic energy conversion, has a number of environmental, social, and economic advantages over conventional electricity generation from fossil fuels. Currently, the most commonly-used material for photovoltaics is crystalline silicon, which is now produced at large scale and silicon-based devices have achieved power conversion efficiencies over 25% However, alternative materials, such as inorganic thin films, offer a number of advantages including the potential for lower manufacturing costs, higher theoretical efficiencies, and better performance in the field. One of these materials is the chalcopyrite Cu(InGa)(SeS) 2, which has demonstrated module efficiencies over 17% and cell efficiencies over 22%. Cu(InGa)(SeS)2 is now in the early stages of commercialization using a precursor reaction process referred to as a "selenization/sulfization" reaction. The precursor reaction process is promising because it has demonstrated high efficiency along with the large area (approximately 1 m2) uniformity that is required for modules. However, some challenges remain that limit the growth of the chalcopyrite solar cell industry including: slow reactions that limit process throughput, a limited understanding of complex reaction kinetics and transport phenomena that affect the through-film composition, and the use of highly toxic H2Se in the reaction process. ☐ In this work, I approach each of these challenges. First, to improve process throughput, I designed and implemented a rapid thermal processing (RTP) reactor, whereby the samples are heated by a 1000 W quartz-halogen lamp that is capable of fast temperature ramps and high temperature dwells. With the reactor in place, however, achieving effective temperature control in the thin film material system is complicated by two intrinsic process characteristics: (i) the temperature of the Cu(InGa)(SeS)2 film cannot be measured directly, which leaves the system without complete state feedback; and (ii), the process is significantly nonlinear due to the dominance of radiative heat transfer at high temperatures. Therefore, I developed a novel control system using a first principles-based observer and a specialized temperature controller. Next, to understand the complex kinetics governing the selenization/sulfization processes, a stochastic model of solid state reaction kinetics was developed and applied to the system. The model is capable of predicting several important phenomena observed experimentally, including steep through-film gradients in gallium mole fraction. Furthermore, the model is mathematically general and can be useful for understanding a number of solid state reaction systems. Finally, the RTP system was then used to produce and characterize chalcopyrite films using two general methods: (i) single stage and multi stage reactions in H2Se and H2S, and (ii), reaction of a selenium "capped" precursor in H2S, where selenium was deposited on the precursor by thermal evaporation and the use of toxic H2Se was avoided. It was found that the processing conditions could be used to control material properties including relative sulfur incorporation, crystallinity, and through-film gallium and sulfur profiles. Films produced using the selenium-capped precursor reaction process were used to fabricate solar cell devices using a Mo/Cu(InGa)(SeS)2/CdS/ZnO/ITO substrate device structure, and the devices were tested by measuring the current-voltage characteristic under standard conditions. Devices with approximately 10% efficiency were obtained over a range of compositions and the best device obtained in this work had an efficiency of 12.7%.