Search

In state-of-the-art, large format, HgCdTe-based infrared focal plane arrays the typical pixel size (3–10 μ m ) is of the order of the operating wavelength and much smaller than the carrier diffusion length. This makes inter-pixel crosstalk a limiting factor, especially in planar structures. Employing three-dimensional electromagnetic and electrical simulations we show that, besides reducing the dark current through Auger suppression, majority carrier depletion of the detector absorber is also effective in curtailing the inter-pixel crosstalk due to carrier diffusion. In the case of a 5 μ m -pitch pixel, a proper design of the absorber composition and doping profile allows to reduce inter-pixel crosstalk by more than a factor of two when increasing the reverse bias from - 0.1 to - 0.5 V , keeping the contribution to crosstalk coming from carrier diffusion between 2 and 12% in the mid- and long-wavelength infrared spectrum. [ABSTRACT FROM AUTHOR]

Combined optical and electrical simulations of infrared HgCdTe-based focal plane arrays under broadband, non-monochromatic illumination are critically relevant to the design of small-volume detectors with sub-wavelength pixel pitches. We present an efficient technique, based on a single finite-difference time-domain electromagnetic simulation, that provides the photogeneration rate profile due to realistic, broadband optical sources, avoiding multiple monochromatic simulations. This technique is applied to assess the effects of the temperature of blackbody optical sources on quantum efficiency and inter-pixel crosstalk of planar LWIR arrays. [ABSTRACT FROM AUTHOR]

Recent studies indicate as an important technological step the development of infrared HgCdTe-based focal plane arrays (FPAs) with sub-wavelength pixel pitch, with the advantage of smaller volume, lower weight, and potentially lower cost. In order to assess the limits of pixel pitch scaling, we present combined three-dimensional optical and electrical simulations of long-wavelength infrared HgCdTe FPAs, with 3 μm, 5 μm, and 10 μm pitch. Numerical simulations predict significant cavity effects, brought by the array periodicity. The optical and electrical contributions to spectral inter-pixel crosstalk are investigated as functions of pixel pitch, by illuminating the FPAs with Gaussian beams focused on the central pixel. Despite the FPAs being planar with 100% pixel duty cycle, our calculations suggest that the total crosstalk with nearest-neighbor pixels could be kept acceptably small also with pixels only $$3\,\mu \hbox {m}$$ wide and a diffraction-limited optical system. [ABSTRACT FROM AUTHOR]

We present a simulation study of HgCdTe-based long-wavelength infrared detectors, focusing on methodological comparisons between the finite-difference time-domain (FDTD) and ray-tracing optical models. We performed three-dimensional simulations to determine the absorbed photon density distributions and the corresponding photocurrent and quantum efficiency spectra of isolated n-on- p uniform-composition pixels, systematically comparing the results obtained with FDTD and ray tracing. Since ray tracing is a classical optics approach, unable to describe interference effects, its applicability has been found to be strongly wavelength dependent, especially when reflections from metallic layers are relevant. Interesting cavity effects around the material cutoff wavelength are described, and the cases where ray tracing can be considered a viable approximation are discussed. [ABSTRACT FROM AUTHOR]

A combined experimental and numerical simulation study is presented on two sets of nominally identical $$\hbox {Hg}_{1-x}\hbox {Cd}_{x}\hbox {Te}$$ single-color back-illuminated midwave-infrared n-on- p photodetectors grown by liquid-phase epitaxy, p-doped with Hg vacancies and with Au, respectively. The present numerical model includes a novel formulation for band-to-band tunneling, which overcomes the intrinsic limitations of the classical Kane description without introducing numerical issues typical of other approaches. Our study confirms that adopting n-on- p architectures, avoiding metal vacancy doping, and reducing the acceptor density in the absorber region are prerequisites for obtaining high-operating-temperature photodetectors. A significant contribution to the dark current in both sets of devices is attributed to impact ionization, crucial to obtain a satisfactory explanation for the measured characteristics also at low to intermediate bias. [ABSTRACT FROM AUTHOR]

We present a combined theoretical, numerical and experimental investigation on trap-assisted tunneling (TAT) in the subthreshold regime of III-nitride-based light-emitting diodes (LEDs). Starting from the basic formulation of the TAT models provided by Hurkx and Schenk, we discuss the derivation of a detailed approach based on both multiphonon and elastic nonlocal processes. A sensitivity study conducted over the main trap- and phonon-related physical parameters of this nonlocal TAT model confirms the importance of tunneling assisted by lattice defects on the LED electrical behavior in the low-medium forward bias range. Comparisons with measured temperature-dependent electrical characteristics $$I(V;T)$$ of a single quantum well LED grown on a highly conductive SiC substrate demonstrate that $$I(V;T)$$ can be accurately reproduced in the range between 200 and 400 K by implementing the nonlocal model for TAT processes via traps in the electron-blocking and spacer layers. [ABSTRACT FROM AUTHOR]

By systematically comparing one-, two-, and three-dimensional (1D, 2D, 3D) simulations of a single-color HgCdTe n-on- p photodiode, we discuss nontrivial differences between 2D and 3D results and observe the need for at least a 2D description in order to get qualitatively correct photocurrent spectra. We show that the photoresponse of a typical dual-band mid-wavelength/long-wavelength infrared pixel can be investigated also in 1D under certain conditions, and we assess the limitations of 1D and 2D modeling against reference 3D simulations. [ABSTRACT FROM AUTHOR]

Band-to-band Auger recombination mechanisms in HgCdTe are investigated as functions of temperature in the small modulation limit by using realistic electronic structures obtained by empirical pseudopotential calculations and their corresponding wavevector-dependent dielectric functions. The calculated Auger lifetimes are compared with the semiempirical Beattie-Landsberg-Blakemore (BLB) model, which has been extensively employed to reproduce experimental data. The Auger-1 lifetime can be fitted well to the BLB model with a constant overlap integral | F F| = 0.16, near the lower limit of the range reported in the literature. The role of the Auger-7 process in p-type HgCdTe is also investigated and the ratio γ between the intrinsic Auger-7 and Auger-1 lifetimes is found to be about 10. [ABSTRACT FROM AUTHOR]

As a preparatory step toward establishing reliable numerical design tools for ZnO-based optoelectronic devices, we have reassessed the available information on material parameters relevant for the simulation of light-emitting diodes (LEDs) with active regions including ZnO, MgZnO, and BeZnO layers. The impact of different approximations for the electronic structure and the interface polarization charge on the optical properties of bulk ZnO and ZnO/MgZnO quantum wells has been evaluated, and a consistent set of parameters has been used not only for systematic comparison of ZnO/MgZnO and ZnO/BeZnO single quantum well structures but also for the first simulation of a realistic ZnO/BeZnO multiple quantum well LED. [ABSTRACT FROM AUTHOR]

A full-band Monte Carlo model has been developed for understanding the carrier multiplication process in HgCdTe infrared avalanche photodiodes. The proposed model is based on a realistic electronic structure obtained by pseudopotential calculations and a phonon dispersion relation determined by ab initio techniques. The calculated carrier–phonon scattering rates are consistent with the electronic structure and the phonon dispersion relation, thus removing adjustable parameters such as deformation potential coefficients. The computation of the impact ionization transition rate is based on the calculated electronic structure and the corresponding wavevector-dependent dielectric function. The Monte Carlo model is applied to investigate key performance figures of long-wavelength infrared (LWIR) and mid-wavelength infrared (MWIR) HgCdTe avalanche photodetectors such as carrier multiplication and noise properties. Good agreement is achieved between simulations and experimental results. The multiplication process in LWIR ( λc = 9.0 μm at 80 K) and MWIR ( λc = 5.1 μm at 80 K) devices is found to be initiated only by electrons, as expected from excess noise measurements. This single-carrier multiplication behavior can be traced back to the details of the computed valence-band structure and phonon scattering rates. [ABSTRACT FROM AUTHOR]

Two alternative approximations of the electronic structure of CdTe and HgTe are proposed, both suited to the needs of accuracy and numerical efficiency of full-band carrier transport simulation: a local empirical pseudopotential (EPM) parametrization including relativistic corrections, and an original full-Brillouin-zone (FBZ) $${{\user2 {k}}\cdot {\user2 {p}}}$$ model using two expansion points (Γ and W). The EPM and $${{\user2 {k}}\cdot {\user2 {p}}}$$ band structures closely match the available experimental and ab initio information, complemented with the results of new density functional theory (DFT)-local density approximation (LDA) calculations, for the conduction and valence bands relevant in transport phenomena. The EPM description of the binary compounds, featuring transferable Te pseudopotentials, is the basis for a computation of the electronic structure of the ternary alloy Hg1− xCd xTe in the framework of disorder-corrected virtual crystal approximation. The composition dependence of energy gaps, effective masses, and high-frequency dielectric constants are discussed and compared with available experimental data, and the novel FBZ approach is applied to the case of x = 0.7. [ABSTRACT FROM AUTHOR]

The low-field electron mobility in bulk ZnO has been the object of extensive experimental studies, mainly through measurement of the temperature-dependent Hall effect. In this work, we reassess the experimental results through direct simulations of Hall measurements, performed with a Monte Carlo transport model and taking into account all the major scattering mechanisms. The deformation potentials required to compute acoustic and optical phonon scattering are derived from first-principles computations, and an original theory for charged-dislocation-line scattering is proposed and implemented. Monte Carlo results for the electron mobility and the Hall factor are compared with analytical expressions derived with the relaxation-time approximation, which is found to be adequate at low temperatures where inelastic scattering effects due to optical phonons are negligible. [ABSTRACT FROM AUTHOR]

Ab initio computations of the structural and electronic properties of wurtzite BeO have been performed with the codes of the abinit project, and the resulting band structure has been approximated with the nonlocal empirical pseudopotential method and a new full-zone $${k \cdot p}$$ model using two expansion points (Γ and L). The very good overall quality demonstrated by both empirical approaches should allow their application to the accurate yet numerically efficient description of the electronic structure of the BeZnO materials system. [ABSTRACT FROM AUTHOR]

Electron and hole transport in wurtzite phase ZnO is studied using an ensemble full band Monte Carlo method. The model includes an accurate description of the electronic structure obtained with the nonlocal pseudopotential method and numerically calculated impact ionization transition rates based on a wavevector-dependent dielectric function. Results of transport simulations at both low and high electric fields are presented. It is found that the low field electron mobility is close to 300 cm² V-1 s-1 at room temperature, and the peak electron drift velocity is 2.2 x 107 cm/s at a field of 275 kV/cm. The determination of the ionization coefficients is affected by some uncertainties due to the incomplete knowledge of the high energy phonon scattering rates. Nevertheless, the present calculations of the ionization coefficients provide a reasonably accurate estimate of the impact ionization process. [ABSTRACT FROM AUTHOR]

In this paper, we present the workings of a fully numerical Monte Carlo simulator that can be employed to study transport in materials with noncubic symmetry. All of the principal ingredients of the Monte Carlo model, i.e., the energy band structure, phonon scattering rates, and impact ionization transition rate are used in numerical form. Various considerations such as k-space mesh size, numerical integration convergence, etc. that impact numerical accuracy will be discussed. The workings of the simulator are illustrated using example calculations of the bulk transport properties of GaAs and GaN. The simulation of bulk GaAs in particular challenges the numerics since the low electron effective mass within the gamma valley requires a high degree of numerical refinement to correctly capture the dynamics in this region. We calculate the steady-state drift velocity, impact ionization coefficients, valley occupations, and average carrier energy in bulk GaAs and GaN. [ABSTRACT FROM AUTHOR]

On our way to develop a comprehensive simulator of vertical-cavity surface-emitting lasers including carrier transport, optical modes, light-matter interaction and heat conduction, we address some critical multiscale aspects of the adopted coupling strategy. The quantum corrections to the semiclassical carrier transport framework are the bridge from nanometer to micrometer scales. In this paper, such corrections are shown to be fundamental in view of predicting the mode competition ruling VCSEL operation. Nevertheless, they can generate unforeseen features such as swirling electron flows in the active region, which are discussed here in detail. The simulation approach has been finally tested through a successful comparison with a large set of experimental results. [ABSTRACT FROM AUTHOR]

The computer-aided design and optimization of Ge-on-Si pin waveguide photodetectors is usually performed by a multiphysics analysis, where carrier transport is described with the drift-diffusion model. This paper addresses the shortcomings of this approach when applied to the study of the n-on-p detector configuration, ascribes them to the overly simplified model of the Si/Ge heterojunction, and proposes an enhanced description of the interfacial region. Starting from technological considerations, a thin buffer layer is introduced between Si and Ge, with a graded composition profile that mimics the initial low-temperature epitaxial growth process and mitigates the sharp discontinuity of the abrupt Si/Ge heterojunction in the band diagram. The benefits of this more realistic model on the determination of the electrooptic response are demonstrated and discussed. [ABSTRACT FROM AUTHOR]