Date of Award

Winter 1-31-2016

Document Type


Degree Name

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


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

Eon Soo Lee

Fifth Advisor

Siva P.V. Nadimpalli


As-sawn silicon wafers have surface damage that needs to be removed before any further processing into solar cells. This damage distribution can vary with cutting parameters such as wire size, slurry particle/diamond grit size, and wire usage. To date, there is no simple way to measure the degree of damage, damage depth, and damage distribution. But, this information is needed by the wafer manufacturers as well as solar cell manufacturers.

A technique based on sequential etching of silicon wafers and minority carrier lifetime (τeff) measurements is used to determine damage depth. In this technique, samples are sequentially etched to remove thin layers from each surface and minority carrier lifetime is measured after each etch step. Lifetime increases after each layer of damage is removed and reaches a plateau once the damage is totally removed. The thickness-removed at which the lifetime reaches a peak value corresponds to the damage depth. An accurate measurement of τeff requires corrections to optical reflection, and transmission from silicon wafers to account for changes in the surface morphology and in the wafer thickness. This technique also allows the in-depth distribution of the damage to be quantified in terms of surface recombination velocity (SRV).

Although this method is routinely used at the National Renewable Energy Laboratory to measure damage depth, determination of damage distribution from these data requires an accurate model that coverts the minority carrier lifetime data into carrier recombination distribution. Continuity equation for excess minority carrier density (An) is solved for the material of interest (silicon wafer with surface damage layer), and carrier concentration is integrated and normalized to match the normalized lifetime vs thickness removed curve. A simplified model for determining the recombination distribution within a wafer having surface damage is presented. Potential improvements for this model are discussed.