the objective of this program is to evaluate and, if possible, predict the conversion efficiency for a variety of silicon samples with differences in structural defects, such as grain boundaries, twin boundaries, precipitate particles, dislocations, etc. it is important to know the type of defect/defects, and the concentration at which such defect/defects severely cause deterioration of conversion efficiency. quantitative data of this nature may then enable us to predict the performance of silicon samples before they are fabricated into solar cells. also, such a technique may, in addition, be used routinely as a nondestructive quality assurance tool to rapidly and routinely evaluate and survey a variety of silicon samples with differences in growth, fabrication, and processing variables. quantitative analysis of surface defects will be performed by using a quantimet 720 quantitative microscopy system this system can differentiate and count 67 shades of gray levels between black and white contrasts. in addition, it can characterize defects (such as dislocations, twin boundaries, precipitates, grain boundaries, etc.) by measuring their length, perimeter, area, density, spatial distribution, frequency distribution (in any preselected direction), and so on. however, the quantitative microscopy system is extremely sensitive to optical contrasts of various defects. therefore, to obtain reproducible results, the contrasts produced by various defects must be similar and uniform for each defect type along the entire surface area of samples to be analyzed. to achieve this, a chemical