On the structure of the conduction band in InSb

Abstract

K43In an earlier paper (1) we presented the results of our measurements of the shift of the plasma minimum with uniaxial stress in InSb. We also analysed the results using the energy band structure of deformed InSb as derived by Bir and Pikus (2) and obtained the value of the deformation potential and of the band structure parameters a, R, and S. Unfortunately, due to a numerical error the values of a, R, and S were not reported accurately. In fact, they should have read -87, -0.22, and +8.7 eV, respectively. Since the time of publication of (l), we have made additional measurements of the shift of the plasma minimum with stress in n-type InSb (carrier concentration 1 . 4~1 0~~ ~m -~) .We also added the case of stress along the [llOI axis and reflection at the (100) plane. The results are in general agreement with those obtained previously. Combining all of the measurements we find a value of -41 eV for the deformation potential and a = -87 eV, R = -0.20 eV and S = +4.2 eV. The numerical data quoted above were obtained assuming the effective mass energy gap E to be 0.20 eV at room temperature. One can however follow the argument of Ehrenreich (3) and Brooks (4) and relate the rate of change of the effective mass energy gap with temperature with the deformation potential C and establish the relation g Eg(T) = Eg(O OK) t. 3 d CT , -6 o -1where d is the coefficient of linear expansion (4.7xlO K (5)). Assuming E (0 OK) =O. 23 evandusingthe' expressions obtained in (l), one can determine a self-consistent value for the deformation potential. Following this procedure we obtain C = -25 eV. This value corresponds to ( a E /aP), = 5.4~10 1) Present address:

Intrinsic point defects and the n - and p -type dopability of the narrow gap semiconductors GaSb and InSb

“…Experimentally, both n-and p-type unintentionally doped samples are routinely prepared, and InSb can be doped relatively easily with electrons or holes as majority carriers. 50,[59][60][61][62][63] Hoglund et al 64 calculated the defect formation energies using DFT-LDA, finding results consistent with ours for Sb-rich conditions, but for the Sb-poor conditions they found that In i would dominate, resulting in an n-type material, in contrast to our results. In their calculations, they found InSb to be gapless, contradicting experiment, and did not discuss corrections for this error nor for image charge interactions in their supercell model.…”

“…The material can be made n-or p-type depending on growth conditions, while temperature (T ) dependent studies have been employed to study variations in the n-type carrier concentration, Fermi energy and mobilities in order to elucidate various defect properties. 50,[59][60][61][62][63] A computational study using DFT with the LDA indicated that the antimony antisite (Sb In ) would dominate in Sb-rich growth conditions; 64 by varying growth conditions, it was suggested that the formation of this defect could be suppressed in epitaxially grown thin films. 63 Furthermore, it has been proposed that the formation of indium vacancies as well as Sb In can account for observed changes in the electronic properties of InSb grown in varying conditions.…”

Deformation Potential Parameters of n‐InSb

“…K84 physica status solidi (b) 54 et al(4,5) differs considerably from the values obtained from hydrostatic pressure measurements(6) which yield values lying between -6 and -7 eV for the same quantity. In addition their values of b = -0.20 eV and d = 12.5 eV do not agree with values b = -2.05 eV and d = -5 obtained by Benoit 2 la Guillaume and Lavallard(7) or the values b = -1.8 eV and d = -6.4 eV obtained by Yu et al (8).…”