Optimization of the energy resolution of deep level transient spectroscopy

Abstract
We address the problem of optimizing the energy resolution of deep level transient spectroscopy, emphasizing techniques which are applicable for signals from defect levels which are either alloy broadened or stress broadened and which therefore produce nonexponential or multiexponential transients. We discuss the possibility of using a numerical inverse Laplace transform for which there is, in principle, no limit to the resolution. The sensitivity of the inversion algorithm to noise and to round-off error, however, practically limits the energy resolution to several percent of the deep level defect binding energy. This resolution is an improvement by about a factor of 2 over the best resolution possible from the standard rate window techniques for which there is a finite lower bound to the temperature resolution. This lower bound depends on the capture cross-section of the defect and the effective time constant of the rate window, but is independent of the defect energy.