First-principles Study Of Defects In HgCdTe

Hg1xCdxTe has been one of the most important materials for infrared detectorsin the mid-infrared to far-infrared range (3~30μm), and has been widely used in thefabrication of infrared focal plane array in the last50years. Currently, thethird-generation infrared detectors can provide enhanced capabilities such asmulticolor detecting and larger size with larger number of pixels. The developmentof the third-generation infrared detectors requires higher performance Hg1xCdxTematerials with well controlled properties such as composition and dopingconcentration. However, the defect distribution is very complicated in Hg1xCdxTe,even with the use of state-of-the-art molecular beam epitaxy (MBE) or metal-organicchemical vapor deposition growth (MOCVD) methods. Moreover, the defect levelscan hardly be identified because of the limitation of experiment accuracy. Under thiscircumstance, the ab initio quantum mechanics calculation within the framework ofdensity function theory (DFT), which relies little on experimental parameter and candescribe the doping behavior of target defect in atomic-scale, is an important backupto experiments.In this thesis, we have studied the properties (e.g. configuration and electronicstructure) and doping behaviors of intrinsic defects, group-V impurities and relateddefect complexes in Hg1xCdxTe using DFT-based first-principles calculationscombined with thermodynamical analysis and classical growth theories. Our resultsare important for theoretically explaining the characteristics of defect levels found inthe Hg1xCdxTe experiments, and can also provide a more complementaryunderstanding to the doping problems of II-VI semiconductor materials. Detailedcontents include:1. Studies of defect distribution and doping behaviors of intrinsic defects andarsenic-related impurities, as well as coupling mechanism and electronicstructures of several specific arsenic-related defect complexes in arsenic-dopedHg1xCdxTe bulk system. We have calculated formation energies of point defectsand defect complexes with different reservoir arsenic forms (Single As atom, As2and As4) in arsenic-doped Hg1xCdxTe, and got the distribution of intrinsicdefects, impurities, and second phase As2Te3versus As/Te chemical potentials. Anew long-range interaction mechanism between arsenic donor (AsHg) and mercury vacancy (VHg) is proposed, and the binding energies and electronicstructures of related point defects and defect complexes are calculated to explorethe coupling mechanism between the two point defects. Our results show thatVHgcan increase the distortion of the Hg1xCdxTe lattice collaboratively withAsHg; moreover, a set of defect levels in the band gap can generate from thedonor–acceptor interaction. This study provides complementary understanding ofAsHg–VHgpair physics, and also suggests a new approach to characterizeelectrical compensation in arsenic-doped Hg1xCdxTe.2. Studies of Te-antisite-related (TeHg-related) defects in Hg1xCdxTe. We haveinvestigated the deep-level-supplier role of Te-antisite-related defects bycalculating geometrical and electronic structures of possible xTeHg–yVHgconfigurations. A novel defect complex with a “double-broken-bond” structurethrough antisite-vacancy coupling is found, and the underlying couplingmechanism is studied. The split of antisite-Te–5p state within differentcrystal-fields is found to be the origin of different recombination/trap levels. Wehave also proposed a two-step annealing procedure to limit the formation ofthese deep-level-suppliers, which is helpful to the further improvement of deviceperformance.3. Studies of the heavier group-V element antimony (Sb) doping p-type Hg1xCdxTe,and the doping behavior differences between As and Sb. It is found that thelarger radius of Sb compared with As can cause larger lattice distortion,especially in the hexagonal and split-site interstitial doping case. The interstitialSb is found to be stable even with the coupling of Hg vacancies through detailedenergetic calculations, indicating that the interstitial Sb is more talented informing stable defect-complexes, and thus has the great potential to be a moresufficient p-type dopant. Our calculations provide more complementaryunderstandings for the behaviors of group-V impurities in Hg1xCdxTe.4. Studies of p-type activation path of arsenic doped Hg1xCdxTe. We havesystematically revealed several possible arsenic p-type activation paths proposedby Berding and other groups, including interstitial site activation path andVHg–mediated activation path, by calculating their migration barriers andconfiguration changes before and after migration. A new TeHg–VHgdefect-complex-mediated arsenic activation model is proposed, within which thedefect complex TeHg–VHgacts like the VHgin Berding’s model, and the catalytic effect of TeHg–VHgis investigated. The migration barrier of our activation path is0.92eV, which is remarkably lower than that of Berding’s model (1.64eV) andinterstitial site model (2.79eV) in our calculation, indicating that our activationpath is reasonable and shows advantage in energy. Instead of generating pointdefect AsTeafter migration in the Berding’s model, our arsenic migration processends with an AsTe2compound, which acts as a p-type dopant. This theoreticalresult consists with the extended x-ray absorption fine structure experimentalresults of Biquard et al.. Our results provide a new approach to understand thenewly proposed experimental findings on the arsenic activation problems inMBE growth Hg1xCdxTe.