Date of Award

Fall 2008

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Materials Science and Engineering - (Ph.D.)

Department

Committee for the Interdisciplinary Program in Materials Science and Engineering

First Advisor

N. M. Ravindra

Second Advisor

Bhushan L. Sopori

Third Advisor

Anthony Fiory

Fourth Advisor

Gordon A. Thomas

Fifth Advisor

Tao Zhou

Abstract

The objective of this dissertation is to study passivation effects and mechanisms in Si solar cells, specifically, the surface and bulk passivation by hydrogen-rich PECVD silicon nitride (SiN :H) antireflection layer on multicrystalline silicon (me-Si) solar cells.

The passivation of silicon surface can be achieved in two ways: by field-effect passivation and/or by neutralization of interface states. In other words, the deposition should result in a high value of fixed charge, Qf and /or a low value of interface state density, D1. The surface recombination velocity can be described by Shockley-Read-Hall (SRH) statistics.

Current SRH formalisms have failed to explain the surface recombination mechanism in terms of injection level dependence as has been observed by lifetime measurements. Previous SRH modeling result shows that very high Qf (up to several 1012/cm2) on the surface of Si wafer, induced by SiNX:H layer, leads to no injection level dependence of surface recombination velocity (SRV), which is in contradiction to experimental results. An alternative approach is needed to address this problem.

A modified SRH formalism which includes the carrier recombination in the space-charge region was developed in this thesis to evaluate the recombination mechanism at SiN :H-Si interface. Numerical modeling results indicate that, at low injection-levels, carrier recombination in the damaged layer is the dominant mechanism as compared to surface recombination. The majority of surface damage can be healed by rapid thermal annealing (RTA). Therefore, less minority-carrier recombination in the SCR is expected after the firing treatment of Si solar cells.

Based on the damaged layer and trapping/detrapping theory, a semi-quantitative hydrogen transportation model of H migration from SiNX:H layer into Si is presented. The model is verified by secondary ion mass spectrometry (SIMS) measurements of H in Si solar cells before and after annealing. The redistribution of H deep inside the cells can lead to excellent bulk passivation and high device performance.

Experimental results of the reproducibility of minority-carrier life measurement using QSSPCD technique indicate that wafer preparation requires a well-cleaned wafer and high quality surface passivation. In this study, a novel laboratory procedure for wafer preparation is proposed.

Theoretical and experimental studies on the influence of defect clusters on the performance of me-Si solar. cell have been performed. In a typical cell, the defect clusters produce an efficiency loss of 3 to 4 percent.

Share

COinS