Semiconductor impurity parameter determination from Schottky junction thermal admittance spectroscopy

The problem of interpretation of thermal admittance spectroscopy data for semiconductor impurity parameter extraction is considered. Traditional analysis predicts that the Arrhenius plot for conductance peak temperatures is a straight line with the slope proportional to impurity activation energy and the intercept determining its capture cross section. Using a general model of the Schottky junction admittance we show that conductance peak positions strongly depend on the impurity bulk occupation number and potential distribution in the space-charge region, and, as a result, the Arrhenius plot is nonlinear for some semiconductor parameters and experimental conditions, in particular for relatively shallow impurities. In this case, the traditional linear approximation of the Arrhenius plot yields inaccurate values of activation energy and capture cross section. We propose a more accurate procedure for admittance spectroscopy data analysis involving least-squares fitting using the general and the small-signal models of the junction admittance. Although much more computationally intensive, the general model is shown to provide a better fit to the data at low temperatures, where the small-signal approximation is invalid. This approach is applied for an example admittance data and yields a better fit of the theoretical curve to the data and an improved value of activation energy for the nitrogen donor in 6H-SiC.