Search

Carrier transport in type-II superlattice photodetectors is investigated by means of a rigorous nonequilibrium Green's function model based on a physics-based B\"uttiker-probe formalism. Intraband scattering self-energies (carrier-phonon interactions) are computed in the self-consistent Born approximation, while interband self-energies (Shockley-Read-Hall and optical transitions) are included in terms of semiclassical generation-recombination rates, neglecting interband renormalization effects. Current conservation is achieved with an efficient Newton-Raphson algorithm. While carrier transport in infrared detectors is usually understood in terms of quantities (e.g., mobilities and quasi-Fermi-levels) that are admittedly not germane to nonequilibrium Green's function theory, the proposed model provides a quantum-kinetic description of tunneling, miniband transport, hopping, and carrier extraction within a drift-diffusion-friendly framework. The connection with semiclassical theories allows exploration of the possibilities offered by Poisson-Schr\"odinger or localization landscape drift-diffusion approaches.

Curved focal-plane arrays offer significant advantages compared with their flat counterparts. However, the curving of the material induces strains which alter its optoelectronic properties. A comprehensive framework is presented for the computational investigation of the band gaps of ${\mathrm{InAs}}_{1\ensuremath{-}x}{\mathrm{Sb}}_{x}$ alloys under various strain conditions that are relevant for manufacturing curved focal-plane arrays. The framework consists of both standard and hybrid functional density functional theory (DFT) calculations, finite-element analysis (FEA) calculations, and Gaussian process (GP) regression. The DFT calculations are used for investigating the change of the band gap of the material under various strain conditions. This dataset is then used for training a GP model which is utilized to assess the effects of strain across the device, based on the FEA simulations. The results show excellent predictive capabilities for the machine-learning model at a significantly reduced computational cost and are directly transferable to other device designs.

Theoretical investigations of carrier transport in type-II superlattice detectors have been mostly limited to simplified semiclassical treatments, due to the computational challenges posed by quantum kinetic approaches. For example, interband tunneling in broken-gap configurations calls for a multiband description of the electronic structure, and spatially indirect optical transitions in superlattice absorbers require fully nonlocal carrier-photon self-energies. Moreover, a large number of iterations is needed to achieve self-consistency between Green's functions and self-energies in the presence of strongly localized states not directly accessible from the contacts. We demonstrate an accurate, yet computationally feasible nonequilibrium Green's function model of superlattice detectors by formulating the kinetic equations in terms of problem-matched maximally localized basis functions, numerically generated from few modes representing the main conductive channels of the nanostructure. The contribution of all the remaining modes is folded in an additional self-energy to ensure current conservation. Inspection of spatially and energetically resolved single particle properties offers insight into the complex nature of carrier transport in type-II superlattice detectors, and the connection to semiclassical approaches enables the interpretation of mobility experiments.

The band gaps and structural properties of ${\mathrm{InAs}}_{1\ensuremath{-}x}{\mathrm{Sb}}_{x}$ alloys are investigated using both the modified Becke-Johnson exchange potential and hybrid functional calculations. A good agreement between the two approaches is observed for the alloys. The estimated value of $0.85\phantom{\rule{0.16em}{0ex}}\mathrm{eV}$ for the bowing parameter enables the use of $\mathrm{InAsSb}$ in long wavelength infrared (LWIR) applications. Furthermore, a lower limit of $0.29\phantom{\rule{0.16em}{0ex}}\mathrm{eV}$ for the bowing parameter is obtained for the structures yielding the largest band gaps, demonstrating the strong nonlinearity of the band gap versus the composition for this system.

Using numerical simulations, we investigate the effects of deep traps and deep acceptor levels in magnesium-doped $\mathrm{Ga}\mathrm{N}$ on interface charges in the semiconductor-oxide interface. Specifically, in this work we address two open issues observed in experimental studies on $\mathrm{Ga}\mathrm{N}$ trench metal-oxide-semiconductor field-effect transistors. (i) We investigate the observed clockwise hysteresis in the transfer characteristics and elucidate the underlying physical mechanism causing it. By employing appropriate models for substitutional carbon at nitrogen sites (${C}_{N}$) and nitrogen vacancies (${V}_{N}$), we calculate the hysteresis dependence on the trap concentrations and the measurement sweep duration ${T}_{s}$. We show that ${C}_{N}$ acceptor traps in $p$-$\mathrm{Ga}\mathrm{N}$ are likely responsible for this phenomenon and the largest hysteresis is predicted for a sweep duration of ${T}_{s}\ensuremath{\approx}30\phantom{\rule{0.2em}{0ex}}\mathrm{s}$. (ii) We also address the apparent inconsistency between the experimental and theoretically predicted magnesium-ionization levels and the variations of the measured transfer characteristics, specifically the threshold voltage. We show that the bands bending in the channel area creates a layer in which magnesium is completely ionized. As a result, the magnesium partial ionization does not have an effect and, while the threshold voltage decreases, nor does the breakdown voltage, as observed experimentally. The measured threshold voltage, which is lower than the theoretically predicted value, is caused by fixed and trapped charges at the interface, in agreement with values reported in the literature.

High-electron-mobility transistors (HEMTs) made of wide-band-gap semiconductors have great potential for power electronics and radio-frequency applications. Coupling first-principles calculations with device simulations enables cost-effective semiconductor research and development, including materials exploration and device design. The authors report innovative, comprehensive calculations of electronic transport in wurtzite GaN and AlN and in an (Al,Ga)N/GaN HEMT. The hot-electron energy distributions in the simulated HEMTs can be used to determine the related device degradation, and suggest opportunities for improved designs.

We employ first-principles calculations in the formalism of standard and hybrid density functional theory to study the electronic and structural properties of wurtzite ${\text{Al}}_{x}{\text{Ga}}_{1\ensuremath{-}x}\text{N}$ alloys. We address the discrepancies observed in literature regarding essential electronic properties of these alloys and we investigate the dependence of these properties on the atomic ordering and composition. We show that the bowing parameter is significantly affected by the atomic ordering, ranging from zero to strong downward bowing. The effects of atomic ordering of the alloys on their band offset with respect to the pure phases are also investigated. Finally, using the effective band structure approach, we study the electronic band structure of the random alloys.

We employ first-principles density functional theory to investigate the electronic and structural properties of grain boundaries (GBs) in InAs. In particular, we study the energetics and passivation mechanisms of representative low-$\mathrm{\ensuremath{\Sigma}}$ GBs, including $\mathrm{\ensuremath{\Sigma}}3(111),\mathrm{\ensuremath{\Sigma}}3(112),\mathrm{\ensuremath{\Sigma}}5(120)$, and $\mathrm{\ensuremath{\Sigma}}5(130)$, to establish their relative stability and experimental feasibility. We find that the symmetric-tilt twin-boundary $\mathrm{\ensuremath{\Sigma}}3(111)$ GB is the most stable GB, in excellent agreement with our experimentally characterized GB structures in InAs. In addition to our theoretically predicted GB structures, we systematically study and analyze different configurations of complex multifold experimentally observed InAs GB structures. We discuss the effect of different passivations and doping mechanisms on the electronic properties of the GBs. Understanding the exact nature of the GB electronic structure and stability, as well as their passivation mechanisms is a key step for the further development of InAs based optoelectronic devices on silicon and other heterogeneous large-area substrates.

Defects in semiconductors can play a vital role in the performance of electronic devices, with native defects often dominating the electronic properties of the semiconductor. Understanding the relationship between structural defects and electronic function will be central to the design of new high-performance materials. In particular, it is necessary to quantitatively understand the energy and lifetime of electronic states associated with the defect. Here, we apply first-principles density functional theory (DFT) and many-body perturbation theory within the GW approximation to understand the nature and energy of the defect states associated with a charged nitrogen vacancy on the electronic properties of gallium nitride (GaN), as a model of a well-studied and important wide gap semiconductor grown with defects. We systematically investigate the sources of error associated with the GW approximation and the role of the underlying atomic structure on the predicted defect state energies. Additionally, analysis of the computed electronic density of states (DOS) reveals that there is one occupied defect state 0.2 eV below the valence band maximum and three unoccupied defect states at energy of 0.2--0.4 eV above the conduction band minimum, suggesting that this defect in the +1 charge state will not behave as a carrier trap. Furthermore, we compare the character and energy of the defect state obtained from GW and DFT using the HSE approximate density functional and find excellent agreement. This systematic study provides a more complete understanding of how to obtain quantitative defect energy states in bulk semiconductors.

We employ the nudged elastic band and the dimer methods within the standard density functional theory (DFT) formalism to study the migration of the oxygen and gallium vacancies in the monoclinic structure of $\ensuremath{\beta}\ensuremath{-}{\mathrm{Ga}}_{2}{\mathrm{O}}_{3}$. We identify all the first nearest neighbor paths and calculate the migration barriers for the diffusion of the oxygen and gallium vacancies. We also identify the metastable sites of the gallium vacancies which are critical for the diffusion of the gallium atoms. The migration barriers for the diffusion of the gallium vacancies are lower than the migration barriers for oxygen vacancies by 1 eV on average, suggesting that the gallium vacancies are mobile at lower temperatures. Using the calculated migration barriers we estimate the annealing temperature of these defects within the harmonic transition state theory formalism, finding excellent agreement with the observed experimental annealing temperatures. Finally, we suggest the existence of percolation paths which enable the migration of the species without utilizing all the migration paths of the crystal.

A generalized numerical model for the phonon-assisted optical interband transition based on the Green's function formalism was developed and implemented to investigate optical processes in germanium and silicon media intended for on-chip light emitter and laser applications. High-fidelity full band structures obtained from the empirical pseudopotential method, self-energies, and the corresponding spectral density functions for the phonon-perturbed electron and holes have been computed numerically as a function of strain, temperature, and doping level. Validation has been carried out by showing the model's ability to accurately reproduce the measured temperature dependent absorption coefficient data for both germanium and silicon. Absorption coefficients, radiative recombination rates of germanium and silicon active media were investigated with different biaxial tensile strain, doping concentrations and injection conditions. Furthermore, when the model is employed to compute the optical gain in strained germanium, we find that the use of tensile strain and high injection are the preferable approaches to obtain population inversion. At the same time, strong absorption from the spin-orbit to the heavy-hole band limits the maximum injection density that can be applied. Finally, when applied to study silicon, the proposed model also successfully reproduces the experimentally observed radiative recombination peak due to the two-phonon process.