Band structure ofInxGa1−xAs1−yNyalloys and effects of pressure

E. D. Jones, N. A. Modine, A. A. Allerman, S. R. Kurtz, A. F. Wright, S. T. Tozer, X. Wei
1999 Physical Review B (Condensed Matter)  
InGaAsN is a new semiconductor alloy system with the remarkable property that the inclusion of only 2% nitrogen reduces the bandgap by more than 30%. In order to help understand the physical origin of this extreme deviation from the typically observed nearly linear dependence of alloy properties on concentration, we have investigated the pressure dependence of the excited state energies using both experimental and theoretical methods. We report measurements of the low temperature
more » ... ture photoluminescence of the material for pressures between ambient and 110 kbar. We also describe a simple, densityfunctional-theory-baed approach to calculating the pressure dependence of low lying excitation energies for low concentration alloys. The theoretically predicted pressure dependence of the bandgap is in excellent agreement with the experimental data. Based on the results of our ca.lculations, we suggest an explanation for the strongly non-linear pressure dependence of the bandgap that, surprisingly, does not involve a nitrogen impurity state. Typeset using REV'I'EX 1 A new semiconductor alloy system, InGaAsN, has been identified as a key candidate material for high-efficiency multi-junction solar cells [I ,2] and also for long wavelength laser systems [3,4]. The introduction of small amounts of nitrogen (w 2%) in GaAs greatly reduces [5, 6] the bandgap energy, with reductions approaching 0.5 eV! With the appropriate ratio of indium to nitrogen concentrations, InGaAsN can be lattice matched to GaAs. Lattice matching allows the design of multi-junction solar cells without the inherent problems found in strained cells. Of prime importance is the role of the nitrogen isoelectronic atom: (1) What is the origin of the large bandgap reduction and (2) Are the states extended (bandlike) or localized (impurity-like)? In order to answer these questions, a better understanding of the electronic properties of this type of alloy system is required. In the past, both first-principles [7] and empirical [SI theoretical treatments for this material system have concentrated on understanding the dependence of the bandgap energy on nitrogen composition. In this paper we present pressure dependent photoluminescence (PL) data together with a first principles local density approximation (LDA) calculation for the band structure and its pressure dependence. It will be shown that, while it is well known that bandgap energies calculated by the LDA method are not accurate, the predicted pressure dependence of the bandgap energy is in excellent agreement with experiment. Because of t h s good agreement, we have confidence that this technique could be useful for other low concentration alloy systems. The structures were grown in a vertical flow, high speed rotating disk: EMCORE GS/3200 metalorganic chemical vapor deposition reactor. The In,Gal-,Asl-,N, films were grown using trimethylindium (TMIn), trimethylgallium (TMG), 100% arsine and dimethylhydrazine (DMHy). Dimethylhydrazine was used as the nitrogen source since it has a lower disassociation temperature than ammonia and has a vapor pressure of approximately 110 torr at 18°C. Unintentionally doped InGaAsN was p-type. InGaAsN films for Hall and optical measurements were grown on semi-insulating GaAs orientated 2" off (100) towards (1 10). Lattice matched (&/a < 8 x lo-*) films were grown at 600°C and 60 torr using a V/III ratio of 97, a DMHy/V ratio of 0.97 and a TMIn/III ratio of 0.12. The growth rate was 10A/s. 2
doi:10.1103/physrevb.60.4430 fatcat:sfzrghwqhfd4nngdz27k3cdljy