Data evaluation technique for electron-tunneling spectroscopy

Abstract

A systematic study of local-density-of-states (LDOS) deconvolution from tip-surface tunneling spectra is reported. The one-dimensional WKB approximation is used to simulate the process. A technique for DOS deconvolution from the electron-tunneling spectroscopy data is proposed. The differential conductivity normalized to its fit to the tunneling probability function is used as a method of recovering sample DOS. This explicit procedure does not use unconstrained parameters and reveals a better DOS deconvolution in comparison with other techniques. The advantage of this method is its feasibility for extracting two important physical parameters from experimental tunneling spectra: (i) local surface potential, and (ii) tip-sample distance. These values are the parameters used in the proposed fitting procedure. The local surface potential and the tip-sample distance retrieval are demonstrated by means of numerical simulations. Comparative scanning tunneling spectroscopy is proposed as an approach to eliminate the influence of the tip condition on the surface LDOS recovery. © 1996 The American Physical Society.