Limits on the use of tunneling to describe the Pd–Ge ohmic contact to GaAs

Abstract

Tunneling theory predicts a break in the slope of the contact resistivity (ρc) versus the inverse of the temperature at the transition from the thermionic to the field emission regime. Although temperature dependent data of the newly developed Pd–Ge ohmic contact to GaAs has often been described by this model, we found that the low barrier heights obtained from such data, are not consistent with the approximations of the method used. We also found that it was not possible to fit the experimental results using the complete thermionic‐field expression for ρc. Numerical calculation based on the Wentzel–Kramers–Brillouin (WKB) approximation for a nonuniform doping profile gave also a negative result in reproducing the experimental curves. Then, either the WKB method is not precise enough, or another transport mechanism is playing an important role. We suggest that an alternative transport mechanism to explain the experimental data is the interface states assisted tunneling. We also demonstrate that the low experimental activation energies found cannot be explained without taking the temperature dependence of the Fermi level into account.