Reconciliation of methods for estimating fmaxfor microwave heterojunction transistors

An attempt is made to reconcile the various approaches that have recently been used to estimate the maximum frequency of oscillation f_max in high-performance AlGaAs/GaAs HBTs. f_max is computed numerically from the full expression for Mason's invariant gain using y-parameters derived from the different approaches, i.e., the hybrid-π equivalent circuit, the T-equivalent circuit, and the drift-diffusion equations. It is shown that the results for f_max are essentially the same, irrespective of the source of the y-parameters, provided that the phase delays due to transit of carriers across the base and the collector-base depletion region are properly accounted for. It is also shown, for the particular device studied, that the widely used analytical expression for f_max, involving f _T and effective base resistance and collector capacitance, is remarkably accurate for frequencies below those at which transit-time effects become important     

Surface-layer damage and responsivity in sputtered-ITO/p-GaN Schottky-barrier photodiodes

It is postulated that donor-like nitrogen vacancies, caused by the sputtering of a Schottky-barrier metal onto p-type gallium nitride, diffuse into the GaN and form a surface layer in which both the minority-carrier lifetime and mobility are drastically reduced. Such a damaged surface layer is shown to reduce the responsivity of p-GaN Schottky-barrier photodiodes, thereby offering an explanation for the responsivity values in the range of 0.03–0.04 A/W that have been measured in experimental ITO/p-GaN devices. On making allowance for the damaged surface layer, an electron diffusion length of around 300 nm can be inferred for the undamaged p-GaN region.     

Effects of quasi-ballistic base transport on the high-frequencycharacteristics of bipolar transistors

High-frequency transport in bipolar transistors with quasi-ballistic base widths (on the order of a minority-carrier scattering length) is examined by using the approach of Grinberg and Luryi (1992) to solve the Boltzmann transport equation (BTE). By considering the phase angle of the dynamic distribution function in wave-vector space, it is shown that the ballistic mechanism of decay in the common-base current gain becomes important even for base widths in the quasi-ballistic regime. Simple expressions, which correctly yield both the magnitude and phase of all the forward characteristics, as predicted by the BTE, up to the intrinsic transit frequency, are found by combining the results from a one-flux approach with the well-known expressions of Thomas and Moll (1958). Expressions for the reverse small-signal parameters are also found by applying a “moving boundary condition” to the basic one-flux equations of Shockley (1962)     

Electron quasi-Fermi level-Fermi level splitting at thebase-emitter junction of AlGaAs/GaAs HBT's

The amount of electron quasi-Fermi level splitting in the emitter-base junction of AlGaAs/GaAs single- and double-heterojunction bipolar transistors is computed. The degree of splitting is found to be generally small, and less pronounced in double-heterojunction devices. However, it is argued that the effect of the splitting on the current gain may be significant     

The cut-off frequency of base-graded and junction-graded AlxGa1−xAs DHBTs

A detailed model of Al_xGa_1−xAs double-heterojunction bipolar transistors is used to examine the effects on the cut-off frequency of varying both the degree of base grading in abrupt-junction devices, and the amount of base-collector junction grading in uniform-base devices. It is shown that, in the graded-base case, there is an optimum degree of base grading, stemming from the trade-off between the strength of the aiding field for minority carrier transport across the base and the height of the barrier at the base-collector junction, which inhibits charge flow to the collector. In the uniform-base case, it is shown that a small amount of base-collector junction grading is sufficient to effectively remove the blocking action of the conduction band discontinuity, and so allow the base transit time to attain its diffusion-limited value.     

An investigation of dark current and photocurrent superposition in photovoltaic devices

In the design and analysis of photovoltai cells, a principle of superposition of light and dark currents is usually assumed to apply. This principle states that the current flowing in an illuminated device subject to a bias V is given by the superposition of the short circuit photocurrent and the current that would flow at bias V in the dark. Using a straightforward modification of Shockley's theory of the pn junction diode, the validity of the superposition principle is established here for Si and GaAs homojunction cells exposed to one sun illumination. The argument commences by demonstrating that the quasi-fermi potentials are essentially constant across the depletion region for a device exposed to one sun illumination and biased at a reasonable operating point. Proceeding in this way, it is found that superposition applies even when recombination and photogeneration in the depletion region contribute substantially to both the dark current and the photocurrent. The theoretical argument is confirmed by direct numerical solution of the basic semiconductor equations.     

On the fill factor of solar cells

Calculations are presented which indicate that, for a given series resistance, the fill factor of a solar cell is principally determined by the saturation dark current, rather than the diode factor, as might be inferred from previous analyses.     

Basis-set choice for DFT/NEGF simulations of carbon nanotubes

We investigate the effect of the choice of the basis set on the results of ab initio (density functional theory/non-equilibrium Green’s function) calculations of the bandgap of semiconducting carbon nanotubes, and near-zero-bias conductance of metallic carbon nanotubes. Both ideal and deformed carbon nanotubes are studied, as well as nanotubes with an adsorbed biomolecule. The results show that the near-zero-bias conductance of armchair nanotubes can be calculated accurately with a minimal basis set, with the exception of the (2,2) tube, where a polarized basis set is necessary to accurately predict the metallic behaviour of this tube. For zigzag tubes, a double-zeta polarized basis set is in general required for accuracy in bandgap and near-zero-bias conductance calculations.     

An analysis of the DC and small-signal AC performance of the tunnelemitter transistor

A model to describe the I-V characteristics of the tunnel emitter transistor (the TETRAN) is developed. It is based on a general model for tunneling in metal thin-insulator semiconductor structures. The model is used to compute typical magnitudes for the parameters appearing in the small-signal hybrid-π equivalent circuit of this device. From these it is predicted that the cutoff frequency for realistic TETRANs based on Al/SiO₂/n-Si structures is about 1 GHz. This is considerably less than the values recently predicted for a related device, the BICFET, which is similar to the TETRAN     

The effect of base grading on the gain and high-frequencyperformance of AlGaAs/GaAs heterojunction bipolar transistors

A comprehensive one-dimensional analytical model of the graded-base Al_xGa_1-xAs/GaAs heterojunction bipolar transistor is presented and used to examine the influence of base grading on the current gain and the high-frequency performance of a device with a conventional pyramidal structure. Grading is achieved by varying the Al mole fraction x linearly across the base to a value of zero at the base-collector boundary. Recombination in the space-charge and neutral regions of the device is modeled by considering Schockley-Read-Hall, Auger, and radiative processes. Owing to the different dependencies on base grading of the currents associated with these recombination mechanisms, the base current is minimized, and hence the gain reaches a maximum value, at a moderate level of base grading ( x=0.1 at the base-emitter boundary). The maximum improvement in gain, with respect to the ungraded base case, is about fourfold. It is shown that the reduction in base transit time due to increased base grading leads to a 30% improvement in f_T in the most pronounced case of base grading studied (x=0.3 at the base-emitter boundary). The implications this has for improving f _max via increases in base width and base doping density are briefly examined