Journal of Vacuum Science & Technology B, Vol.14, No.1, 132-135, 1996
Use of Maximum-Entropy Deconvolution for the Study of Silicon Delta-Layers in GaAs
Semiconductor layer structures with sharply changing concentration, and more especially delta-doped (single atomic plane) structures, provide an ideal environment for the study of the more subtle mass transport phenomena such as concentration-dependent diffusion and localized mixing caused by ion bombardment. However, to be able to extract meaningful parameters from such experiments, accurate depth profiles must be obtained with extremely high depth resolution and good sensitivity. Secondary ion mass spectrometry (SIMS) provides one of. the most sensitive methods for acquiring such profiles. To obtain high depth resolution, a number of criteria must be satisfied, not the least of which is the reduction of redistribution by the probe; thus it is essential to employ a low energy primary beam. It is also vital that the crater floor recedes parallel to the original surface of the specimen. This necessitates accurate scanning of a stable ion beam, so as to ensure a constant flux across the entire sampled area. If this is not the case, depth resolution will degrade as a function of depth, and important information will be lost. However, even if probes of the order of 1-2 keV are used, together with a precision scan system, beam-induced redistribution is still a significant limit to the sharpness of the recorded profiles. To further improve upon the data, the effects of the analysis must be removed from the profile. In this article we demonstrate the use of the maximum entropy deconvolution technique, applying it to a SIMS depth profile of a multilayer silicon in gallium arsenide structure containing alternating layers of high and low areal density. The important issue of data validity is discussed and the deconvolved results are used to calculate a diffusion constant for the denser layers, the less dense layers show no diffusion at the growth temperature of 450 degrees C.
Keywords:PROFILE