PIEZO-SPECTROSCOPIC DATA-ANALYSIS - A PC TOOL

Information from the many kinds of spectroscopy used by chemists and p hysicists is fundamental to our understanding of the structure of mate rials. Numerical techniques have an important role to play in the augm entation of the instrumentation and technology available in the labora tory, but are frequently viewed as separate from the laboratory proced ures. We describe an integrated PC-based approach for obtaining direct ly the parameter estimates of transition types in piezo-spectroscopic measurements of crystalline materials. Typically, the analyses in ques tion are required to handle complex secular matrices, to distinguish b etween components in the experimental results, and to identify the tra nsition types as rapidly and as efficiently as possible. The method de scribed, based on providing a discrete shell to the Powell algorithm, is shown to give both accurate identification of the transition type i n the case, of new data and improved fits (i.e. reduction in residual variation) when compared with results obtained via standard procedures . In addition it is flexible with respect to the language used and pos sesses a high degree of portability. We illustrate the success of the approach using (i) data previously reported on the solution of a trigo nal defect which includes both mixing of states and spin orbit interac tions and (ii) new data obtained for a defect related to beryllium imp urities in silicon.