Effect of surface conditions on the measurement of minority carrier diffusion lengths using the surface photovoltage technique

Abstract

The surface photovoltage (SPV) technique is a well-established method for the measurement of the minority carrier diffusion lengths (L) in semiconductor wafers. The measurement can be performed with two methods: constant magnitude SPV (Method A); and linear photovoltage, constant photon flux mode (Method B). A detailed theoretical study published several years ago showed that Method A was more robust and than Method B. In this paper, the values of L on a set of silicon wafers with various surface treatments were measured using both Methods A and B, and the results compared with those obtained using the laser-microwave photoconductance decay (LMPCD) method. It was found that for wafers without any surface treatment, the results from SPV Method A were much closer to those obtained with LMPCD, than those from Method B. The values of L obtained from Method A were also much less sensitive to the surface conditions of the wafers, thus indicating that they are closer to the true bulk diffusion lengths of the wafers. Method B can give correct values of L only under the condition of a very low surface recombination velocity.