Simulation of Amorphous Silicon Hydride Solar Cells Using a Personal Computer
- PDF / 341,071 Bytes
- 6 Pages / 420.48 x 639 pts Page_size
- 52 Downloads / 180 Views
SIMULATION OF AMORPHOUS SILICON HYDRIDE SOLAR CELLS USING A PERSONAL COMPUTER FINLEY R. SHAPIRO Department of Electrical and Computer Engineering, Drexel University, Philadelphia, PA 19104 ABSTRACT The simulation of a-Si:H solar cells has the potential to make a major contribution to the development of improved devices. However, to achieve this potential, simulations must run quickly on personal computers. This will require simplified approaches which are justified by the correspondence of their results to more elaborate but slower, running simulations. BACKGROUND Much of the recent progress in the development of high efficiency amorphous silicon hydride (a-Si:H) solar cells has been by improvements in multijunction cells. These cells often have two to four pin cells deposited on top of each other, each with at least a p layer, an i layer, and an n layer. Frequently additional layers are added to these cells, such as a graded gap layer between the p and the i layer of some or all of the cells. Optimizing just the thicknesses of the layers of such a solar cell would be difficult given the interactions of the different cells, but in addition, optimal values must be found for many other parameters, such as band gap and dopant density. It has been very difficult and expensive to try experimentally to find sets of parameters which give improved cell performance. The situation is made worse by the fact that cells made in different reaction chambers with the same recipe can have significantly different characteristics, and even the same chamber may produce cells with different characteristics from a single recipe. The simulation of a-Si:H solar cells on a computer can help the designing of improved solar cells by getting around some of the problems of optimizing solar cell parameters experimentally. It is typically much cheaper to simulate a solar cell than to build and test it, and if simulations are performed while varying only one design parameter, one knows that changes in the simulation results are due only to the change in the one parameter rather than unknown experimental problems. Unfortunately, simulation also presents many difficulties which need to be overcome so that it can play a useful role in the cell design process. The biggest problem is our lack of a thorough understanding of the physics of electrical conduction in amorphous silicon. This includes unresolved questions about the distribution of localized states, their capture rates, the band mobilities, and the dependence of these characteristics on dopant and hydrogen concentrations. The result is that simulations require a large number of 1 parameters, most of which are poorly known. This paper discusses current progress on one approach to a-Si:H solar cell 2 simulation. This approach is quite different from those used by other groups.
Mat. Res. Soc. Symp. Proc. Vol. 219. e1991 Materials Research Society
458
Some simulation results using our technique are presented, and we discuss the further work necessary to produce a package for use by cell devel
Data Loading...
