(555e) Simulation and Control of Porosity in a Three-Dimensional Thin-Film Solar Cell

Orkoulas, G., University of California, Los Angeles
Huang, J., University of California, Los Angeles
Christofides, P. D., University of California, Los Angeles

Solar energy is currently one of the most promising alternative energy sources, and thin-film silicon solar cells are currently the most widely used solar cell systems. Presently, the main obstacles for a wide use of solar power are the relatively high costs and the limited conversion efficiency of the solar energy. Optimizing the light trapping process is one of the major aspects to improve the solar cell conversion efficiency and is one of the central research aspects nowadays. Specifically, research has indicated that porous silicon can reduce solar cell optical losses from 37% to 8% and increase a short-circuit current by 25% and open circuit voltage by 20 mV. Thus, it is very important to develop a systematic way to simulate and control this porous silicon layer deposition process and improve the solar cell conversion efficiency. Despite its importance, this problem has not attracted much attention.

This work focuses on the simulation and control of a three-dimensional (3D) porous silicon thin-film deposition process that is used in the manufacture of thin-film solar cell systems. Initially, a solid-by-solid 3D kinetic Monte Carlo (kMC) model is introduced to simulate the porous silicon thin-film deposition process, and the simulation parameters are tuned to generate porous silicon films with porosity values that match available experimental data. A closed-form differential equation model then is introduced to predict the dynamics of the film porosity computed by the kMC model, and the parameters in this model are identified by appropriate fitting to open-loop kMC simulation results. Subsequently, a model predictive controller (MPC) is designed and implemented on the kMC model. Closed-loop simulation results demonstrate that the thin-film porosity can be regulated to desired values.