Motivation and Objectives
A program of computational modeling, processing and characterization is proposed to improve the property of the multi-crystalline cast silicon materials for photo-voltaics applications. The process parameters during casting (Directional Solidification, Bridgeman) approaches include: fluid flow through the melt, surface tension, and magnetic body forces, of the melt, temperature profile and its gradient. The velocity of the processing zone (in directional solidification) and the dimension of the casting mold are also important parameters. The initial Si-material’s impurity and the environment that the casting occurs are also essential parameters to determine the final microstructure. The microstructure parameters include: crystal orientation distribution (texture), grain boundary, stacking faults, twin boundaries, precipitates and inclusions, metal impurity, defect structure including dislocation bundles and their distribution. A direct linkage between the evolution of the microstructure and the process parameters is not presently available in the silicon technology. All available software and modeling concentrate only on the linkage between the process parameters and the thermal gradients within the mold. The solidification process and the resultant microstructure during processing are both extremely important to the successful production of Solar Cell Quality Multi-crystalline wafers. The modeling methodologies and the consequent computer programs will provide the manufacturers an essential tool to optimize the process parameters to produce a more cost efficient final product during casting of Si for solar cell applications.
The success of the modeling effort relies on a set of well-defined processes and the characterization of the finally manufactured material. A program of research on Multi-crystalline Silicon has been initiated at Georgia Tech based on Bridgeman approach and arc melting for floating zone process. A set of materials systems have been produced and the microstructure of these materials are being investigated. The effect of varying initial impurities and different environments (hydrogen) will be investigated at different stages of the project. The materials produced will also be investigated for their microstructure using a variety of characterization techniques including x-ray, SEM, TEM and XPS spectroscopy. These results will be used by the proposed models for both parameterization and also verification. We will use the presently available modeling efforts to obtain temperature gradient with the melting zone and the final ingot as a function of time and position. These parameters will be used in a microstructure-based model that links the microstructure to the process parameters. The initial modeling efforts will benefit from the recently developed microstructure reconstruction techniques based on Phase Field Simulation analysis. The effect of impurity and precipitation during solidification will also be incorporated with the code. At later stages of the project, other microstructure attributes (such as defect structure, dislocations, grain boundary character) will be incorporated within the model and verified using the experimental results.