Richard T. Webster

Micromachined rectangular-coaxial transmission lines

Rectangular-coaxial (recta-coax) transmission lines fabricated through a three-dimensional micromachining process are presented. These lines are shown to have significant advantages over competing integrated transmission lines such as microstrip and coplanar waveguides. Design equations are presented for impedance, loss, and frequency range. The equations are confirmed with simulations and measurements. The quality factor of shorted lambda/4 resonators is measured to be 156 at 60 GHz. This corresponds to a line loss of 0.353 dB/cm. Advantages of these lines for passive millimeter-wave circuits including ease of signal routing, high isolation, and signal crossovers are demonstrated with realized lines and couplers

Impact ionization in InAlAs/InGaAs/InAlAs HEMT's

The kink effect and excess gate current in InAlAs/InGaAs/InAlAs HEMTs have been linked to impact ionization in the high field region of the channel. In this letter, a relationship is established between experimentally measured excess gate current and the tunneling of holes from the quantum well formed in the channel. The channel hole current is then obtained as the quotient of the excess gate current to the gate-voltage-dependent transmission probability. This channel hole current follows the exponential dependence of the ionization constant on the inverse electric field.

Temperature dependent transport properties in GaN, Al/sub x/Ga/sub 1-x/N, and In/sub x/Ga/sub 1-x/N semiconductors

Ensemble Monte Carlo simulation is used to determine the electron saturation velocity and low-field mobility for Al/sub x/Ga/sub 1-x/N and In/sub x/Ga/sub 1-x/N. Acoustic phonon, optical phonon, intervalley, ionized impurity, alloy, and piezoelectric scattering are included in the simulation. Doping concentration ranging from 10/sup 17/ cm/sup -3/ to 10/sup 19/ cm/sup -3/ is considered in the temperature range from 50 K to 500 K, Theoretical calculation shows excellent agreement with low-field mobility experimental data. Empirical expressions for low field mobility and saturation velocity are provided as functions of temperature, doping concentration and mole fraction.

Bias induced strain in AlGaN∕GaN heterojunction field effect transistors and its implications

We report gate bias dependence of the charge due to piezoelectric polarization obtained by using a fully coupled formulation based upon the piezoelectric constitutive equations for stress and electric displacement. This formulation is significant because it fully accounts for electromechanical coupling under the constraint of global charge control. The coupled formulation results in lower charge due to piezoelectric polarization as compared to the uncoupled formulation for a given Al mole fraction. With increasing two dimensional electron gas concentration, that is, for gate biases greater than threshold, the compressive strain along the c axis in the barrier AlGaN layer increases with a concomitant increase of in-plane stress. Current collapse is correlated to the increase in source and drain resistances through their dependence upon surface charge. An alternate explanation of current collapse using local charge neutrality is also presented.

RF actuation of capacitive MEMS switches

An analytical calculation of microwave actuation for shunt capacitive MEMS switches is presented. This calculation shows that the microwave signal deflects the beam according to the RMS voltage of the signal. In addition, heating of the beam due to dissipated microwave power is shown to play a significant role in microwave actuation.

Determination of InP HEMT noise parameters and S-parameters to 60 GHz

A millimeterwave experimental technique is described for directly determining the noise parameters and scattering parameters of V-band InP HEMTs. The parameters are suitable for the design of monolithic millimeterwave integrated circuits since they represent the InP HEMT as it would appear in the monolithic environment. The method relies on careful characterization of the measurement system and the InP HEMT packages or test fixtures. Results are provided for an InP HEMT with 1.37 dB minimum noise figure and a maximum stable gain of 12.74 dB at 57 GHz. In addition, it is shown that noise parameters measured between 2 GHz and 26 GHz can be extrapolated to 60 GHz, and that consistent S-parameters can be obtained for InP HEMTs in precision packages and test fixtures. >

Monolithic InP HEMT V-band low-noise amplifier

A fully monolithic indium phosphide high electron mobility transistor (InP HEMT) two-stage low-noise amplifier has achieved a noise figure, of 4.2 dB with an associated gain of 15.25 dB over the band from 56 to 60 GHz. Noise matching and bias decoupling are accomplished on-chip. The successful performance of the amplifier is credited to accurate characterization of the active and passive devices that make up the circuit.<<ETX>>

Carrier trapping and current collapse mechanism in GaN metal-semiconductor field- effect transistors

A mechanism for current collapse in GaN metal–semiconductor field-effect transistors is proposed, which assumes the existence of acceptor traps with multiple states in the band gap. Current collapse has been experimentally observed in the current–voltage characteristic after the drain voltage sweep had exceeded the threshold for impact ionization in a previous measurement. In the proposed model, electrons generated by impact ionization are captured by neutral acceptor trap states in the substrate located above the valence band. The charged trap states move to an energy level located near midgap, creating a positively charged depletion region in the channel, and causing current collapse. With increasing drain bias, the quasi-Fermi level approaches the charged trap states at the drain end of the gate, initiating detrapping of the electrons and restoring the current. The calculated results show good agreement with published experimental data.

Noncontact measurement of charge induced voltage shift in capacitive MEM-switches

The use of a modulated microwave signal to directly measure the voltage shift induced by charge in the dielectric layer of a capacitive microelectromechanical (MEM) switch is presented. This method does not require the metal bridge to contact the dielectric layer and is thus much less intrusive than previously reported measurements. The technique is a useful tool for understanding charge build up and dissipation in capacitive MEM switches.

A physics-based frequency dispersion model of GaN MESFETs

A physics-based model for GaN MESFETs is developed to determine the frequency dispersion of output resistance and transconductance due to traps. The equivalent circuit parameters are obtained by considering the physical mechanisms for current collapse and the associated trap dynamics. Detrapping time extracted from drain-lag measurements are 1.55 and 58.42 s indicating trap levels at 0.69 and 0.79 eV, respectively. The dispersion frequency is in the range of megahertz at elevated temperature, where a typical GaN power device may operate, although at room temperature it may be few hertz. For a 1.5 /spl times/ 150 /spl mu/m GaN MESFET with drain and gate biases of 10 V and -1 V, respectively, 5% decrease in transconductance and 62% decrease in output resistance at radio frequencies (RFs) from their DC values are observed. The dispersion characteristics are found to be bias dependent. A significant decrease in transconductance is observed when the device operates in the region where detrapping is significant. As gate bias approaches toward cutoff, the difference between output resistance at dc and that at RF increases. For drain and gate biases of 10 and -5 V, output resistance decreases from 60.2 k/spl Omega/ at dc to 7.5 k/spl Omega/ at RF for a 1.5 /spl mu/m /spl times/ 150 GaN MESFET.