Library

Notes: The effects of the initial Mg reactivation annealing temperature, surface cleaning treatments, and contact annealing conditions on the specific contact resistance of Ni/Au ohmic contacts on p-GaN are reported. The lowest contact resistances were obtained for 900 °C activation annealing and cleaning steps that reduced the native oxide thickness (i.e., KOH rinsing). Removal of this interfacial oxide reduced the barrier for hole transport, providing a contact resistance of 9×10−4 Ω cm2 for Ni/Au metallization annealed at 500 °C. The use of a ten period p-Al0.1Ga0.9N(Mg)/GaN(Mg) superlattice with individual layer thickness 50 Å led to a specific contact resistance of 9×10−5 Ω cm2 under the same conditions. © 2001 American Institute of Physics.

Notes: Edge-terminated Schottky rectifiers fabricated on quasibulk GaN substrates showed a strong dependence of reverse breakdown voltage VB on contact dimension and on rectifier geometry (lateral versus vertical). For small diameter (75 μm) Schottky contacts, VB measured in the vertical geometry was ∼700 V, with an on-state resistance (RON) of 3 mΩ cm2, producing a figure-of-merit VB2/RON of 162.8 MW cm−2. Measured in the lateral geometry, these same rectifiers had VB of ∼250 V, RON of 1.7 mΩ cm2 and figure-of-merit 36.5 MW cm−2. The forward turn-on voltage (VF) was ∼1.8 V (defined at a current density of 100 A cm−2), producing VB/VF ratios of 139–389. In very large diameter (∼5 mm) rectifiers, VB dropped to ∼6 V, but forward currents up to 500 mA were obtained in dc measurements. © 2001 American Institute of Physics.

Notes: GaN and Al0.25Ga0.75N lateral Schottky rectifiers were fabricated either with (GaN) or without (AlGaN) edge termination. The reverse breakdown voltage VB (3.1 kV for GaN; 4.3 kV for AlGaN) displayed a negative temperature coefficient of −6.0±0.4 V K−1 for both types of rectifiers. The reverse current originated from contact periphery leakage at moderate bias, while the forward turn-on voltage at a current density of 100 A cm−2 was ∼5 V for GaN and ∼7.5 V for AlGaN. The on-state resistances, RON, were 50 mΩ cm2 for GaN and 75 mΩ cm2 for AlGaN, producing figures-of-merit (VRB)2/RON of 192 and 246 MW cm−2, respectively. The activation energy of the reverse leakage was 0.13 eV at moderate bias. © 2000 American Institute of Physics.

Notes: Planar geometry, lateral Schottky rectifiers were fabricated on high resistivity AlxGa1−xN (x=0–0.25) epitaxial layers grown on sapphire substrates. The reverse breakdown voltages of unpassivated devices increased with Al composition, varying from 2.3 kV for GaN to 4.3 kV for Al0.25Ga0.75N. The reverse current–voltage (I–V) characteristics showed classical Shockley–Read–Hall recombination as the dominant mechanism, with I∝V0.5. The reverse current density in all diodes was in the range 5–10×10−6 A cm−2 at 2 kV. The use of p+ guard rings was effective in preventing premature edge breakdown and with optimum ring width increased VB from 2.3 to 3.1 kV in GaN diodes. © 2000 American Institute of Physics.

Notes: AlxGa1−xN (x=0–0.25) Schottky rectifiers were fabricated in a lateral geometry employing p+-implanted guard rings and rectifying contact overlap onto an SiO2 passivation layer. The reverse breakdown voltage (VB) increased with the spacing between Schottky and ohmic metal contacts, reaching 9700 V for Al0.25Ga0.75N and 6350 V for GaN, respectively, for 100 μm gap spacing. Assuming lateral depletion, these values correspond to breakdown field strengths of ≤9.67×105 V cm−1, which is roughly a factor of 20 lower than the theoretical maximum in bulk GaN. The figure of merit (VB)2/RON, where RON is the on-state resistance, was in the range 94–268 MW cm−2 for all the devices. © 2001 American Institute of Physics.

Notes: AlGaN/GaN pnp heterojunction bipolar transistors were fabricated using a low-damage dry-etch process, and the dc characteristics measured up to 250 °C. In the common–base mode, the collector current was approximately equal to the emitter current under all conditions. Although not optimized for power operations, the devices were tested up to power densities of ∼50 kW cm−2. The dc current gain was in the range 20–25 at room temperature. The pnp configuration avoids the problem of high base sheet resistance encountered with npn-AlGaN/GaN devices. © 2000 American Institute of Physics.

Notes: Haplotype associations of the major histocompatibility complex (MHC) with psoriasis vulgaris (PV) have been demonstrated in different racial or ethnic populations. The objective of this study was to demonstrate the different haplotype associations of the MHC in Chinese patients with psoriasis according to the type of onset and their sex. One hundred and thirty-eight patients with PV and 149 normal control subjects without psoriasis were typed for HLA-A, -B, -C, -DQA1, -DQB1 and -DRB1 by using the PCR with sequence-specific primers. The results showed: (i) HLA-A*26 (26.1% vs. 12.1%, Pc 〈 1 × 10−5), -B*27 (17.03% vs. 1.01%, Pc 〈 1 × 10−7), -Cw*0602 (15.58% vs. 5.03%, Pc 〈 1 × 10−2), -DQA1*0104 (19.93% vs. 9.40%, Pc 〈 1 × 10−3), -DQA1*0201 (22.40% vs. 10.74%, Pc 〈 1 × 10−3), -DQB1*0303 (18.12% vs. 9.73%, Pc 〈 1 × 10−7), and -DRB1*0701/02 (26.09% vs. 9.73%, Pc 〈 1 × 10−7) were significantly increased in PV patients, while HLA-B*57, -DQB1*0201 were slightly increased in PV patients. HLA-Cw*0304 (5.07% vs. 14.43%, Pc 〈 1 × 10−3), -DQA1*0501 (5.79% vs. 14.09%, Pc 〈 0.05) were found to be negatively associated with PV, but HLA-A*2 (2.54% vs. 6.38%, Pc 〈 0.5) was decreased in PV patients without statistical significance. (ii) HLA-A*26-B*27 [P 〈 0.0001, odds ratio (OR) = 48.38], -A*26-Cw*0602 (P 〈 0.0001, OR = 11.84), -B*27-Cw*0602 (P 〈 0.0001, OR = undefined), -DRB1*0701/02-B*27 (P 〈 0.0001, OR = 22.62), -DRB1*0701/02-DQA1*0104 (P 〈 0.0002, OR = 3.59), -DRB1*0701/02-DQB1*0303 (P 〈 0.0001, OR = 5.63), -DQA1*0201-DQB1*0303 (P 〈 0.0002, OR = 7.77), -A*26-B*27-Cw*0602 (P 〈 0.0004, OR = undefined), -A*26-DRB1*0701/02-DQA1*0201-DQB1*0303 (P 〈 0.01, OR = undefined) were identified as risk haplotypes for patients with PV in China. (iii) HLA-A*26 -B*27 (P 〈 0.0001, OR = 58.47), -DQA1*0201-DQB1*0303 (P 〈 0.0001, OR = 8.62), -DRB1*0701/02 -DQA1*0104 (P 〈 0.0002, OR = 4.13), -DRB1*0701/02-DQB1*0303 (P 〈 0.0001, OR = 6.68) and -A*26-DRB1*0701-DQA1*0201 -DQB1*0303 (P 〈 0.006, OR = undefined) were only significantly associated with type I psoriasis compared with controls, while others showed no differences in either type I or type II psoriasis. (iv) These associated haplotypes with PV were not different by sex, except that the frequency of DRB1*0701/02-DQB1*0303 (P 〈 0.0001, OR = 10.14) was higher in male patients with psoriasis. To summarize, this study demonstrated a differential association of HLA and identified some special risk haplotypes in Chinese patients with PV compared with other ethnic or racial populations.