Reemitted-positron spectroscopy of cobalt and nickel silicide films

Abstract

Reemitted-positron spectroscopy (RPS) has been employed in a systematic investigation of the positronic properties of the various phases (${\mathit{M}}_{2}$Si, MSi, and M${\mathrm{Si}}_{2}$) of Co and Ni silicide films grown in situ on Si substrates. The positron work function ${\mathrm{\ensuremath{\varphi}}}^{+}$ is found to be negative for all of the different phases, with the parameter \ensuremath{\Sigma}\ensuremath{\equiv}${\mathrm{\ensuremath{\mu}}}^{\mathrm{\ensuremath{-}}}$+${\mathrm{\ensuremath{\mu}}}^{+}$ having a surprisingly large variation. In an attempt to calibrate the size of shifts in \ensuremath{\Sigma} due to strained pseudomorphic growth of thin ${\mathrm{CoSi}}_{2}$ films, the positron deformation potential ${\mathit{E}}_{\mathit{d}}^{+}$\ensuremath{\equiv}V(\ensuremath{\partial}\ensuremath{\Sigma}/\ensuremath{\partial}V) was determined using the thermal-expansion technique. However, pseudomorphic shifts for films of thickness less than 40 \AA{} were found to be unobservable due to incomplete positron thermalization in such thin films. The deformation potential can also be used to estimate the size of the positron diffusion constant, which is found to be comparable to that of other metals. Thus the short positron diffusion length (of order 150 \AA{}) determined from depth-profiling measurements of ${\mathrm{CoSi}}_{2}$ films must be a result of positron trapping in either the film or at the interface with the Si substrate. RPS results considered as a function of film thickness support the conclusion that defects in the film (misfit dislocations and/or vacancies) represent the major source of positron trapping.