Shreepad Karmalkar

In a VNWA Schottky rectifier, Vbr can be raised and/or Rs,sp lowered progressively by increasing S/R. This can be interpreted in several ways: the performance of this rectifier can be improved at the cost of device area; this rectifier offers an additional degree of flexibility in terms of the device area in device design; the structure of this rectifier allows tailoring of Vbr and Rs,sp using a geometry parameter such as the device area, which is not possible in a conventional bulk rectifier; SiC or GaN bulk device performance can be achieved by a Si VNWA device of larger area; SiC or GaN VNWA Schottky rectifiers can achieve the performance of bulk Schottky rectifiers in even higher bandgap materials, which may not be available in practice.

Superjunction devices may employ oxide layers for termination and for isolating adjacent pillars. Fixed charges present in these oxide layers are shown to have the following influence on the device breakdown voltage Vbr. The effect of a positive fixed charge is equivalent to an increase in the n-pillar charge. The resulting charge imbalance degrades Vbr, and can be compensated either by increasing the p-pillar doping or by decreasing the n-pillar doping by an amount ≈ (1.27NfT + 4NfI)/(pillar width), where NfTand NfI are the fixed charges in the termination and isolation oxides, respectively. Contrary to expectation, the Vbrof such compensated devices can be significantly higher than that of a balanced superjunction of the same doping level without fixed charge, due to the uniform lateral depletion effect of the fixed charge. © 2007, IEEE. All rights reserved.

We investigate AlGaN/GaN HEMTs which show a large increase in measured off-state breakdown voltage, VBR, from near-VT to deep-off state VGS conditions, accompanied by a positive shift in measured VT, when these devices are stressed electrically by alternating ID–VGS and ID–VDS measurements. We show that, if stress is assumed to cause spatially uniform changes, the above variations in VBR and VT can be explained in terms of increased ionized deep acceptor trap concentration in the GaN buffer. We also show that, the near-VT VBR is due to space-charge limited current while the deep off-state VBR is due to impact ionization. Our simulations predict that the incorporation of a field plate in the device can enhance the latter VBR significantly, but may not change the former VBR much.

The space-charge region (SCR) in semiconductor bulk and nanoscale junctions is mostly analyzed using the differential equation, namely, Poisson's equation. We explore an alternate analysis using the integral equation, namely, Coulomb's law, and show its superiority to the differential equation approach in revealing the drastic change in SCR electrostatics from bulk to single nanofilm (NF) to single nanowire/nanotube (NW/NT) junction. The advantages of the integral equation approach are as follows. First, it helps us to view the junction SCR in bulk as a series of sheet charges, that in thin NFs as a series of line charges, and that in thin NWs/NTs as a series of point charges; such a view logically yields the different functional forms of the depletion width dependence on dielectric constant, potential drop, and inverse doping, namely, squareroot in bulk, linear in NFs, and exponential in NWs/NTs. Second, it shows that the partially depleted space-charge tails in NFs/NWs/NTs must extend to infinity, and that these tails rather than the completely depleted regions set up the required potential difference across the junction; it also reveals that for large distance,z away from the junction, the field varies as 1/z2 in NF and 1/z3 in NW/NT, and the potential/charge vary as 1/z in NF and 1/z2 in NW/NT. To ensure accuracy of the depletion width model, we capture the effect of space-charge tails in NFs/NWs/NTs by employing a periodic charge distribution. We also show how prior works overestimated the depletion width by orders of magnitude.

We present a compact model for the quasi-static coupling impedance between two coplanar rectangular contacts, which are misaligned and non-identical, on a bulk substrate. The coupling resistance is modeled as weighted arithmetic mean of the resistances of four aligned identical contact geometries for which a model is available in literature. Our model is scalable with contact dimensions and separation and has been validated against TCAD simulations.

The ability to express the depletion layer voltage drop Vd in terms of a quadratic function of the gate controlled depletion charge Qd is shown to be a key to the success of analytical modelling of ion-implanted MOSFETs and MESFETs. It is shown that such a quadratic function for the Vd-Qd characteristics of implanted FETs can be obtained by approximating the implanted doping profile by a "shifted-rectangle" profile whose parameters can be derived directly from implantation parameters. It is also shown that the shifted-rectangle approximation (SRA) is not just an artifice for simplicity but accurately conserves the actual Qd, Vd and depletion width conditions of both shallow and deep implanted Gaussian shaped doping profiles. The SRA simplifies the analysis of multiple-implanted devices and can be considered to be a basic approximation to be used along with the depletion and gradual channel approximations for a simple and accurate analysis of the non-uniformly doped FETs. © 1991.

An accurate J-V equation of a practical solar cell is inherently implicit, which necessitates iterative calculations to determine the fill factor and the peak power point. We propose a simple explicit power-law J-V model that is applicable to a wide variety of solar cells. The model allows an easy prediction of the entire J-V curve, peak power point, and fill factor from four simple measurements of the bias points corresponding to Voc, ∼ 0.6 Voc, Jsc, and ∼0.6 Jsc, where Voc is the open-circuit voltage and Jsc is the short-circuit current density. The model also provides a closed-form description of the J-V curve, peak power point, and fill factor in terms of physical parameters of the single exponential model. © 2008 IEEE.

We establish the unity underlying the junction electrostatics of nanowire and nanotube arrays, and develop a simple analytical model for both based on a common set of approximations of the field/potential distributions. The model predicts large depletion widths accurately, and reveals that, the depletion width in an array of nanowire and nanotube junctions varies as 1/charge per unit length of wire/tube multiplied by inter-wire/tube separation, where charge per unit length includes the effects of wire/tube radius and doping. This is much different from the exponential form derived in literature for a single nanowire or nanotube junction. The model satisfies all the limiting conditions, matches with numerical simulations and is extendible to nanofilms. Hence, it should be quite useful for nanodevice design. © 2014 AIP Publishing LLC.

The performance of a PNPN thyristor is primarily decided by the two N-base parameters, namely minority carrier lifetime and basewidth to diffusion length ratio. This paper describes a simple three-terminal steady-state method for measuring these parameters in finished thyristors. In this method, the thyristor is biased in a new "remote-base PNP transistor" mode, which shows interesting characteristics giving a good deal of physical insight into the operation of the PNPN structure. © 1991.