Low-temperature Schottky barrier tunneling inInSb∕InxAl1−xSbquantum

Abstract

We report low-temperature I͑V͒ measurements taken on small area surface metal contacts to InSb/ InAlSb quantum well material in the temperature range 4.5-100 K. We obtain Schottky barrier J͑V͒ data under reverse bias and analyze the transport observed by using a tunneling model derived from the three-dimensional density of states, showing close correlation for the majority of devices measured. We consider the "Rowell" analysis ͓Tunneling Phenomena in Solids, edited by E. Burnstein and S. Lundqvist ͑Plenum, New York, 1969͒, p. 273͔ of the zero-bias resistance R 0 ͑T͒ and differential conductance G͑V͒ for determining the dominant transport mechanism. Two distinct temperature dependences are observed in the R 0 ͑T͒ data, a weak insulatorlike dependence identifying true single-step tunneling and a strong nonlinear dependence, which we attribute to defect assisted multistep tunneling. Effective barrier parameters are extracted by fitting the G͑V͒ data with a Brinkman-Dynes-Rowell model ͓J. Appl. Phys. 41, 5 ͑1970͔͒ for carrier tunneling.

Zero-field spin splitting and spin lifetime inn−InSb∕In1−xAlxGilbertson1, Fearn2, Jefferson3 et al. 2008Phys. Rev. BSelf Cite

“…39. It is assumed in this analysis that the Fermi energy is pinned at mid-gap on the semiconductor surface, consistent with the results of measurements undertaken in similar InSb QW systems [36,39,40]. For all calculations performed here, we restrict ourselves to solutions for zero bias, and to varying the doping parameters N d and S such that only the lowest subband is occupied.…”

“…, using a 1-band Schrödinger-Poisson model (SPM) tailored to narrow band-gap materials [36]. A typical solution for a 20nm QW is shown in Fig.…”

“…A single δ-doping layer is located in the top barrier and simulated by an exponentially decaying dopant profile in the growth direction as a result of secondary-ion-mass-spectroscopy (SIMS) data in these systems, indicated by the shaded region in Fig. 1 [36,38]. The exact functional form of the decay is not expected to affect the results presented here, although its presence is relevant for the positioning of the doping layer above the QW since segregation of dopant atoms into the QW, which can occur when the doping plane is located below the QW, is detrimental to carrier mobility.…”

“…The later is somewhat unreliable, predominantly due to the inherent large effective g * -factor in bulk InSb (~-52). As a result, the value of B is shown to vary significantly over a range -31.4 < B < -12.6 (eVÅ 2 ) [36,49,50]. The value of B taken from Landolt and Bornstein [35] is chosen here as it follows the trends of III-V semiconductors [41].…”

Suppression of the parasitic buffer layer conductance in InSb/AlxIn1−xSb heterostructures using a wide-band-gap barrier layer

“…44 In the Poisson solution, we assume a background doping density of 1 × 10 15 cm −3 , and midgap pinning of the Fermi level at the surface. 24,45 The doping density in the Al 0.2 In 0.8 Sb layer is adjusted so that the resulting QW sheet density at 10 K matches that measured from the SdeH frequency. The temperature dependence of the band gaps is taken into account using the usual Varshni expression.…”