Diffusion of Te vacancy and interstitials of Te, Cl, O, S, P and Sb in CdTe: A density functional theory study

•We computed diffusion energy barriers, of species X, in bulk zinc-blende CdTe.•X is Te vacancy and interstitial adatoms of Te, P, Sb, O, S, and Cl.•Excellent agreement of barriers with experimentally measured values is found.•We describe the structural motifs around interstitials as diffusion proceeds.

We present an ab initio study of the diffusion profiles in CdTe of native, Te adatom and vacancy, and anionic non-native interstitial adatoms P, Sb, O, S, and Cl. A high symmetry Wyckoff position, 4(d) site, happens to be a global minimum energy location, only for O and Cl interstitials. Adatoms of P, Sb and S show an asymmetric shape of the energy diffusion barrier with two minima and two maxima in the pathway. The others, O, Cl, and Te interstitial and vacancy, show a symmetric diffusion barrier with a unique maximum and minimum. Diffusion for Te and S interstitials proceeds along the [1 1 0] channel in the crystal in a near straight line path. Diffusion for O and Cl proceeds along two nearly straight line paths along [1 1 1] and [1 1 –1]. Diffusion for P and Sb are along the [1 1 0] channel however they deviate from the straight line paths along [1 1 1] and [1 1 –1]. The rate-limiting diffusion barriers range from a low of 0.49 eV for the asymmetric diffusion path of an Sb interstitial to a high of 1.51 eV for the symmetric diffusion path of an O interstitial. The rate-limiting barriers for the others are 0.65 eV for S, 0.68 eV for both Cl and P, 1.37 eV for Te interstitial and 1.42 eV for the Te vacancy. These barriers are in agreement with the available experimental data for interstitials Te 1.40±0.02 eV, Cl 0.63±0.10 eV and S 0.64±0.02 eV. The symmetric or asymmetric nature of the diffusion path as well as the bond length and atomic coordination at the energetic-extrema positions influence the size of the diffusion energy barrier. In addition there exist two electronic signatures in the local density of states: one for the bond breaking in the symmetric diffusion barrier paths and the other in the difference in hybridization between the global minimum and global maximum energy positions for asymmetric diffusion barriers. This work should serve as a motivation for experimental verification of other barriers and elucidates the diffusion mechanisms very difficult to discover experimentally.