Molecular Dynamics Study of Heteroepitaxial Growth of HgCdTe on Perfect and Dislocated (211)B CdZnTe Substrates

Abstract

The presence of threading dislocations in the depletion region of Hg1–x Cd x Te detectors remains a problem due to its negative impact on the electrical and electronic properties of these detectors. We used molecular dynamics (MD) simulations to study the impact of the simulated growth rate, substrate temperature, and Hg/Te flux ratio on the Hg sticking coefficient and crystallinity of Hg1–x Cd x Te on a perfect (211)B CdZnTe substrate during molecular beam epitaxy (MBE) growth. The trends were consistent with the experiments, namely, a decrease in crystallinity with an increase in the growth rate, the exponential decrease of the Hg sticking coefficient with the increase in substrate temperature, and an optimum substrate temperature and Hg/Te flux ratio for a given growth rate. We then used one of the optimum growth conditions found to conduct MD simulations on (211)B CdZnTe substrates with a quadrupole of either 0° perfect, 60° perfect, 30° partial, or 90° partial dislocations. All dislocations extended into the epilayer as expected and various phenomena were observedchange of line direction, movement by climb, and the dissociation of a glide perfect dislocation into two partials. These phenomena give insight into the types of low-energy dislocations that may be present in Hg1–x Cd x Te after MBE growth.

Elucidating Twinning Mechanisms in CdTe Homoepitaxial Film Growth from Machine Learning Force Field Molecular Dynamics Simulations

“…The tfMC simulations were carried out using Δ = 0.2 Å, corresponding to a time step of 48.6 39.7, 34.3, and 30.7 fs at temperatures of 400, 600, 800, and 1000 K, respectively. Therefore, the growth rates for the CdTe homoepitaxial growth in the simulations are between ∼0.036 nm/ns to ∼0.055 nm/ns within the temperature range of 400 to 1000 K. Since MD and tfMC simulations are only capable of simulating time scales of the order of nanoseconds, the growth rates in this study are accelerated simulated growth rates, 34 which are orders of magnitude faster than the experimental growth rates of ∼1−3 μm/h. 16,21,50 Following the deposition of CdTe, all atoms except the fixed bottom two double-layers were cooled to 10 K over 350 ps using the Berendsen thermostat.…”

“…16,21,50 Following the deposition of CdTe, all atoms except the fixed bottom two double-layers were cooled to 10 K over 350 ps using the Berendsen thermostat. This was succeeded by molecular statics (MS) relaxation to minimize thermal noise, 34 facilitating structure analysis. The OVITO package was employed for visualizing and analyzing the simulation results.…”

“…3 For substrate with dislocations, dislocations may extend from the substrate into the grown CdTe homoepitaxial layers and then interact with each other, as a dislocation line cannot end abruptly within the crystal and must terminate at a free surface. 34…”

First‐Principles Study of the 30° and 90° Partial Dislocations in HgTe, CdTe, and Hg0.7Cd0.3Te

“…[ 39 ] For periodic boundary conditions, the displacement fields for multiple dislocations in a supercell surrounded by periodic neighbor cells were used. [ 39,43,44 ]…”

“…[39] For periodic boundary conditions, the displacement fields for multiple dislocations in a supercell surrounded by periodic neighbor cells were used. [39,43,44] The exchange-correlation functionals tested include the LDA þ U, where a U value of 8.5 eV on the Hg d-orbital and 11 eV on the Cd d-orbital was used. An energy cutoff of 500 eV and a 1 Â 1 Â 3 Γ-centered k-mesh were used.…”