Wang, Deliang | Search NASA

DOE OSTI2016

Cu-doped CdS and its application in CdTe thin film solar cell

Cu is widely used in the back contact formation of CdTe thin film solar cells. However, Cu is easily to diffuse from the back contact into the CdTe absorber layer and even to the cell junction interface CdS/CdTe. This phenomenon is generally believed to be the main factor affecting the CdTe solar cell stability. In this study Cu was intentionally doped in CdS thin film to study its effect on the microstructural, optical and electrical properties of the CdS material. Upon Cu doping, the V{sub Cd{sup −}} and the surface-state-related photoluminescence emissions were dramatically decreased/quenched. The presence of Cu atom hindered the recrystallization/coalescence of the nano-sized grains in the as-deposited CdS film during the air and the CdCl{sub 2} annealing. CdTe thin film solar cell fabricated with Cu-doped CdS window layers demonstrated much decreased fill factor, which was induced by the increased space-charge recombination near the p-n junction and the worsened junction crystalline quality. Temperature dependent current-voltage curve measurement indicated that the doped Cu in the CdS window layer was not stable at both room and higher temperatures.

36 MATERIALS SCIENCE↗

DOE OSTI2015

Factorization-based texture segmentation

This study introduces a factorization-based approach that efficiently segments textured images. We use local spectral histograms as features, and construct an M × N feature matrix using M-dimensional feature vectors in an N-pixel image. Based on the observation that each feature can be approximated by a linear combination of several representative features, we factor the feature matrix into two matrices-one consisting of the representative features and the other containing the weights of representative features at each pixel used for linear combination. The factorization method is based on singular value decomposition and nonnegative matrix factorization. The method uses local spectral histograms to discriminate region appearances in a computationally efficient way and at the same time accurately localizes region boundaries. Finally, the experiments conducted on public segmentation data sets show the promise of this simple yet powerful approach.

97 MATHEMATICS AND COMPUTING↗