Hydrogen diffusion, complex formation, and dissociation in acceptor-doped silicon carbide

Abstract

The diffusion of deuterium ${(}^{2}\mathrm{H})$ in B and Al doped $4H$ and $6H$ silicon carbide (SiC) has been studied in detail by secondary ion mass spectrometry. From ${}^{2}\mathrm{H}$ depth profiles, following trap limited diffusion with negligible complex dissociation, an effective capture radius for the formation of ${}^{2}\mathrm{H}\ensuremath{-}\mathrm{B}$ complexes (at 460 \ifmmode^\circ\else\textdegree\fi{}C) is determined to ${R}^{\mathrm{HB}}=(21\ifmmode\pm\else\textpm\fi{}4)\AA{}.$ This value is in good agreement with that expected for a Coulomb force assisted trapping mechanism. At annealing conditions where dissociation is non-negligible, the ${}^{2}\mathrm{H}$ diffusion follows Fick's law with a constant effective diffusivity, from which the complex dissociation frequencies \ensuremath{\nu} are determined. The extracted values of \ensuremath{\nu} cover three orders of magnitude and exhibit a close to perfect Arrhenius temperature dependence for both ${}^{2}\mathrm{H}\ensuremath{-}\mathrm{B}$ and ${}^{2}\mathrm{H}\ensuremath{-}\mathrm{A}\mathrm{l}$ complexes. The large difference between the extracted complex dissociation energies, ${E}_{d}^{\mathrm{HB}}=(2.51\ifmmode\pm\else\textpm\fi{}0.04)\mathrm{eV}$ and ${E}_{d}^{\mathrm{HAl}}=(1.61\ifmmode\pm\else\textpm\fi{}0.02)\mathrm{eV},$ suggests that the atomic configurations of the two complexes are significantly different. The corresponding extracted dissociation attempt frequencies, ${\ensuremath{\nu}}_{0}^{\mathrm{HB}}=(1.2\ifmmode\pm\else\textpm\fi{}0.7)\ifmmode\times\else\texttimes\fi{}{10}^{13}{\mathrm{s}}^{\mathrm{\ensuremath{-}}1}$ and ${\ensuremath{\nu}}_{0}^{\mathrm{HAl}}=(0.7\ifmmode\pm\else\textpm\fi{}0.3)\ifmmode\times\else\texttimes\fi{}{10}^{13}{\mathrm{s}}^{\mathrm{\ensuremath{-}}1},$ are very close to the characteristic oscillation frequency of the SiC lattice, ${\ensuremath{\nu}}_{\mathrm{lattice}}^{\mathrm{SiC}}=1.6\ifmmode\times\else\texttimes\fi{}{10}^{13}{\mathrm{s}}^{\mathrm{\ensuremath{-}}1}.$ This is strong evidence for the assumption of a first order dissociation process. No difference between $4H$- and $6H\ensuremath{-}\mathrm{SiC}$ has been observed.