Thermal equilibrium concentrations of point defects in gallium arsenide

Abstract

Expressions of the thermal equilibrium concentrations of point defects in GaAs, including the neutral and charged species, are derived. These expressions are explicit functions of well-defined thermodynamic quantities, which in turn yield explicit expressions for the reaction constant K in the usual As 4 pressure power law representation of the point defect thermal equilibrium concentrations. Such power laws have been of little quantitative value in the past, because values for K were not known. In the present derivation, emphases are placed upon the difference between the Gibbs free energies of an As atom in the interior of a GaAs crystal and in an As vapor phase molecule, and the role of the crystal Fermi level. Numerical values of the thermal equilibrium concentrations of the neutral and three negatively charged Ga vacancies (V 0 Ga, V − Ga, V 2− Ga and V 3− Ga), the neutral As vacancies V 0 As and the two neutral antisite de Ga 0 As and As 0 Ga have been obtained. The calculated thermal equilibrium concentration of the anion antisite defect As 0 Ga reaches a peak value of about 1 sx 10 17 cm −3 and is practically temperature independent, in agreement with experimental findings. The thermal equilibrium concentrations of the triply negatively charged Ga vacancy V 3− Ga, C eq v 3− Ga ( n), have been found to exhibit a temperature independence or a negati temperature dependence behavior under strong n-doping conditions. That is, the C eq v 3− Ga ( n) value is either unchanged or increases as the temperature is lowered. This C eq v 3− Ga ( n) property provided explanations to a number of outstanding experimental results, either requiring the interpretation that V 3− Ga has attained its thermal equilibrium concentration at the onset of each experiment, or requiring mechanisms involving point defect nonequilibrium phenomena. The calculated C eq v 3− Ga ( n) values are in agreement with available experimental results.