Fabrication and Characteristics of AIInN/A1N/GaN MOS-HEMTs with Ultra-Thin Atomic Layer Deposited Al2O3 Gate Dielectric

Al0.85In0.15N//AlN/GaN metal-oxide-semiconductor high electron mobility transistors （MOS-HEMTs） employing a 3-nm ultra-thin atomic-layer deposited （ALD） Al2O3 gate dielectric layer are reported. Devices with 0.6μm gate lengths exhibit an improved maximum drain current density of 1227mA/mm at a gate bias of 3 V, a peak transeonductance of 328mS/mm, a cutoff frequency fr of 16 GHz, a maximum frequency of oscillation fmax of 45 GHz, as well as significant gate leakage suppression in both reverse and forward directions, compared with the conventionM Al0.85In0.15N/AlN/GaN HEMT. Negligible C - V hysteresis, together with a smaller pinch-off voltage shift, is observed, demonstrating few bulk traps in the dielectric and high quality of the Al2O3/AIInN interface, it is most notable that not only the transconductance profile of the MOS-HEMT is almost the same as that of the conventional HEMT with a negative shift, but also the peak transconduetance of the MOS-HEMT is increased slightly. It is an exeitin~ inwrovement in the transconductance performance.     

Investigation of 4H—SiC metal—insulation—semiconductor structure with Al2O3/SiO2 stacked dielectric

Atomic layer deposited (ALD) Al₂O₃/dry-oxidized ultrathin SiO₂ films as high-k gate dielectric grown on the 8° off-axis 4H-SiC (0001) epitaxial wafers are investigated in this paper. The metal-insulation-semiconductor (MIS) capacitors, respectively with different gate dielectric stacks (Al₂O₃/SiO₂, Al₂O₃, and SiO₂) are fabricated and compared with each other. The I-V measurements show that the Al₂O₃/SiO₂ stack has a high breakdown field ( ≥ 12 MV/cm) comparable to SiO₂, and a relatively low gate leakage current of 1× 10^-7 A/cm² at electric field of 4 MV/cm comparable to Al₂O₃. The 1-MHz high frequency C-V measurements exhibit that the Al₂O₃/SiO₂ stack has a smaller positive flat-band voltage shift and hysteresis voltage, indicating less effective charge and slow-trap density near the interface.     

An AlGaN/GaN HEMT with reduced surface electric field and improved breakdown voltage

A reduced surface electric field in AlGaN/GaN high electron mobility transistor (HEMT) is investigated by employing a localized Mg-doped layer under the two-dimensional electron gas (2-DEG) channel as an electric field shaping layer. The electric field strength around the gate edge is effectively relieved and the surface electric field is distributed evenly as compared with those of HEMTs with conventional source-connected field plate and double field plate structures with the same device physical dimensions. Compared with the HEMTs with conventional source-connected field plate and double field plate, the HEMT with Mg-doped layer also shows that the breakdown location shifts from the surface of the gate edge to the bulk Mg-doped layer edge. By optimizing both the length of Mg-doped layer, L_m, and the doping concentration, a 5.5 times and 3 times the reduction in the peak electric field near the drain side gate edge is observed as compared with those of the HEMTs with source-connected field plate structure and double field plate structure, respectively. In a device with V_GS=-5 V, L_m=1.5 μm, a peak Mg doping concentration of 8× 10¹⁷ cm^-3 and a drift region length of 10 μm, the breakdown voltage is observed to increase from 560 V in a conventional device without field plate structure to over 900 V without any area overhead penalty.     

Performance of La203/InA1N/GaN metal-oxide-semiconductor high electron mobility transistors

We report on the performance of La203/InA1N/GaN metal-oxide-semiconductor high electron mobility transistors （MOSHEMTs） and InA1N/GaN high electron mobility transistors （HEMTs）. The MOSHEMT presents a maximum drain current of 961 mA/mm at Vgs = 4 V and a maximum transconductance of 130 mS/mm compared with 710 mA/mm at Vgs = 1 V and 131 mS/mm for the HEMT device, while the gate leakage current in the reverse direction could be reduced by four orders of magnitude. Compared with the HEMT device of a similar geometry, MOSHEMT presents a large gate voltage swing and negligible current collapse.     

The effect of a HfO2 insulator on the improvement of breakdown voltage in field-plated GaN-based HEMT

A GaN/Al_0.3Ga_0.7N/AlN/GaN high-electron mobility transistor utilizing a field plate (with a 0.3 μm overhang towards the drain and a 0.2 μm overhang towards the source) over a 165-nm sputtered HfO₂ insulator (HfO₂-FP-HEMT) is fabricated on a sapphire substrate. Compared with the conventional field-plated HEMT, which has the same geometric structure but uses a 60-nm SiN insulator beneath the field plate (SiN-FP-HEMT), the HfO₂-FP-HEMT exhibits a significant improvement of the breakdown voltage (up to 181 V) as well as a record field-plate efficiency (up to 276 V/μm). This is because the HfO₂ insulator can further improve the modulation of the field plate on the electric field distribution in the device channel, which is proved by the numerical simulation results. Based on the simulation results, a novel approach named the proportional design is proposed to predict the optimal dielectric thickness beneath the field plate. It can simplify the field-plated HEMT design significantly.     

Performance of La2O3/InAlN/GaN metal—oxide—semiconductor high electron mobility transistors

We report on the performance of La2O3/InAlN/GaN metal-oxide-semiconductor high electron mobility transistors(MOSHEMTs) and InAlN/GaN high electron mobility transistors(HEMTs).The MOSHEMT presents a maximum drain current of 961 mA/mm at Vgs = 4 V and a maximum transconductance of 130 mS/mm compared with 710 mA/mm at Vgs = 1 V and 131 mS/mm for the HEMT device,while the gate leakage current in the reverse direction could be reduced by four orders of magnitude.Compared with the HEMT device of a similar geometry,MOSHEMT presents a large gate voltage swing and negligible current collapse.     

Partial-SOI high voltage P-channel LDMOS with interface accumulation holes

A new partial SOI (silion-on-insulator) (PSOI) high voltage P-channel LDMOS (lateral double-diffused metal-oxide semiconductor) with an interface hole islands (HI) layer is proposed and its breakdown characteristics are investigated theoretically. A high concentration of charges accumulate on the interface, whose density changes with the negative drain voltage, which increase the electric field (E_I) in the dielectric buried oxide layer (BOX) and modulate the electric field in drift region . This results in the enhancement of the breakdown voltage (BV). The values of E_I and BV of an HI PSOI with a 2-μm thick SOI layer over a 1-μm thick buried layer are 580V/μm and -582 V, respectively, compared with 81.5 V/μm and -123 V of a conventional PSOI. Furthermore, the Si window also alleviates the self-heating effect (SHE). Moreover, in comparison with the conventional device, the proposed device exhibits low on-resistance.     

100-nm T-gate InAIAs/InGaAs InP-based HEMTs with fT= 249 GHz and fmax ： 415 GHz

InA1As/InGaAs high electron mobility transistors （HEMTs） on an InP substrate with well-balanced cutoff frequency fT and maximum oscillation frequency frnax are reported. An InA1As/InGaAs HEMT with 100-nm gate length and gate width of 2 × 50 μm shows excellent DC characteristics, including full channel current of 724 mA/mm, extrinsic maximum transconductance gm.max of 1051 mS/mm, and drain-gate breakdown voltage BVDG of 5.92 V. In addition, this device exhibits fT = 249 GHz and fmax = 415 GHz. These results were obtained by fabricating an asymmetrically recessed gate and minimizing the parasitic resistances. The specific Ohmic contact resistance was reduced to 0.031 0.mm. Moreover, the fT obtained in this work is the highest ever reported in 100-nm gate length InA1As/InGaAs InP-based HEMTs. The outstanding gm.max, fT, fmax, and good BVDG make the device suitable for applications in low noise amplifiers, power amplifiers, and high speed circuits.     

The dielectric behaviour of doped near-stoichiometric lithium niobate in the terahertz range

The dielectric properties of near-stoichiometric LiNbO₃:Fe and LiNbO₃:Ce single crystals have been investigated using terahertz time domain spectroscopy in a frequency range of 0.7-1.6 THz at room temperature. When coupled with an applied external optical field, obvious photorefractive effects were observed, resulting in a modulation of the complex dielectric constant for the crystals. The variation in refractive index, |Δn|, had a linear relationship with the applied light intensity, accompanied by a step-like decrease at high intensity. The findings were attributed to the internal space charge field of the photorefraction and the light-induced domain reversal in the crystals.     

The effect of anti-hydrogen bond on Fermi resonance: A Raman spectroscopic study of the Fermi doublet ν1—ν12 of liquid pyridine

The effects of anti-hydrogen bond on the ν₁—ν₁₂ Fermi resonance (FR) of pyridine are experimentally investigated by using Raman scattering spectroscopy. Three systems, pyridine/water, pyridine/formamide, pyridine/carbon tetrachloride, provide varying degrees of strength for the diluent-pyridine anti-hydrogen bond complex. Water forms a stronger anti-hydrogen bond with pyridine than with formamide, and in the case of adding non-polar solvent carbon tetrachloride, which is neither a hydrogen bond donor nor an acceptor and incapable of forming hydrogen bond with pyridine, the intermolecular distance of pyridine will increase and the interaction of pyridine molecules will reduce. The dilution studies are performed on the three systems. Comparing with the values of Fermi coupling coefficient W of the ring breathing mode ν ₁ and triangle mode ν ₁₂ of pyridine at different volume concentrations, which are calculated according to the Bertran equations, in three systems, we find that the solution with the strongest anti-hydrogen bond, water, shows the fastest change in the ν₁—ν₁₂ Fermi coupling coefficient W with the volume concentration varying, followed by the formamide and carbon tetrachloride solutions. These results suggest that the stronger anti-hydrogen bond-forming effect will cause a greater reduction in the strength of the ν₁—ν₁₂ FR of pyridine. According to the mechanism of the formation of anti-hydrogen bond in the complexes and the FR theory, a qualitative explanation for the anti-hydrogen bond effect in reducing the strength of the ν₁—ν₁₂ FR of pyridine is given.     

Luminescence investigation of red phosphors Ca0.54Sr0.34−1.5xEu0.08Smx(MoO4)y (WO4)1−y for UV-white LED device

Ca_0.54Sr_0.34−1.5xEu_0.08Sm_x(MoO₄)_y (WO₄)_1−y red phosphors were prepared by solid-state reaction using Na⁺ as a charge compensator for light-emitting diodes (LED). The effects of Na⁺ concentration, synthesis temperature, reaction time and Eu³⁺ concentration were studied for the properties of luminescence and crystal structure of red phosphors. The results show that the optimum reaction condition is 6%, 900 °C, 2 h and 8%. The photoluminescence spectra show that red phosphors are effectively excited at 616 nm by 292, 395 and 465 nm. The wavelengths of 465 nm nicely match the widely applied emission wavelengths of blue LED chips.     

Thermal expansion and magnetostriction in Pr(n+2)(n+1)Nin(n−1)+2Sin(n+1) compounds

Thermal expansion and magnetostriction of members of a homologous series of compounds based on the alloy series Pr_(n+2)(n+1)Ni_n(n−1)+2Si_n(n+1) have been measured. The crystal structures of these compounds are closely interrelated because they form trigonal prismatic columns in which the number of trigonal prisms that form the base of the trigonal columns is determined by the value of n in the chemical formula. Two compositions were investigated, Pr₅Ni₂Si₃ and Pr₁₅Ni₇Si₁₀, corresponding to n=3$n=3$ and n=4,$n=4,$ respectively. The results were analyzed and used to determine the location of magnetic phase transitions by calculating the magnetic contribution to thermal expansion using the Gruneisen–Debye theory. This allowed more precise determination of the magnetic transition temperatures than could be achieved using the total thermal expansion. The results show two phase transitions in each material, one corresponding to the Curie temperature and the other at a lower temperature exhibiting characteristics of a spin reorientation transition.     

XPS investigation of diffusion of two-layer ZrO2/SiO2 and SiO2/ZrO2 sol–gel films

Two-layer ZrO₂/SiO₂ and SiO₂/ZrO₂ films were deposited on K9 glass substrates by sol–gel dip coating method. X-ray photoelectron spectroscopy (XPS) technique was used to investigate the diffusion of ZrO₂/SiO₂ and SiO₂/ZrO₂ films. To explain the difference of diffusion between ZrO₂/SiO₂ and SiO₂/ZrO₂ films, porous ratio and surface morphology of monolayer SiO₂ and ZrO₂ films were analyzed by using ellipsometry and atomic force microscopy (AFM). We found that for the ZrO₂/SiO₂ films there was a diffusion layer with a certain thickness and the atomic concentrations of Si and Zr changed rapidly; for the SiO₂/ZrO₂ films, the atomic concentrations of Si and Zr changed relatively slowly, and the ZrO₂ layer had diffused through the entire SiO₂ layer. The difference of diffusion between ZrO₂/SiO₂ and SiO₂/ZrO₂ films was influenced by the microstructure of SiO₂ and ZrO₂.     

Preparation and magnetic properties of BaFe12O19/Ni0.8Zn0.2Fe2O4 nanocomposite ferrite

Nanocomposite of hard (BaFe₁₂O₁₉)/soft ferrite (Ni_0.8Zn_0.2Fe₂O₄) have been prepared by the sol–gel process. The nanocomposite ferrite are formed when the calcining temperature is above 800 °C. It is found that the magnetic properties strongly depend on the presintering treatment and calcining temperature. The “bee waist” type hysteresis loops for samples disappear when the presintering temperature is 400 °C and the calcination temperature reaches 1100 °C owing to the exchange-coupling interaction. The remanence of BaFe₁₂O₁₉/Ni_0.8Zn_0.2Fe₂O₄ nanocomposite ferrite with the mass ratio of 5:1 is higher than a single phase ferrite. The specific saturation magnetization, remanence magnetization and coercivity are 63 emu/g, 36 emu/g and 2750 G, respectively. The exchange-coupling interaction in the BaFe₁₂O₁₉/Ni_0.8Zn_0.2Fe₂O₄ nanocomposite ferrite is discussed.     

Luminescent properties of KGd1−x(WO4)2:Eux and KGd1−x(WO4)2−y(MoO4)y:Eux phosphors in UV-VUV regions

KGd_1−x(WO₄)_2−y(MoO₄)_y:Eu³⁺_x(0.1?x?0.75, y=0 and 0.2) phosphors are synthesized through traditional solid-state reaction and their luminescent properties in ultraviolet (UV) and vacuum ultraviolet (VUV) regions are investigated. Under 147 nm excitation, these phosphors show characteristic red emission with good color purity. In order to improve their emission intensity, the MoO₄²⁻ (20 wt%) is introduced into the anion of KGd_1−x(WO₄):Eu³⁺_x. The Mo⁶⁺ and Eu³⁺ co-doped KGd(WO₄)₂ phosphors show higher emission intensity in comparison with the singly Eu³⁺-doped KGd(WO₄)₂ in VUV region. The chromaticity coordination of KGd_0.45(WO₄):Eu³⁺_0.55 is (x=0.669, y=0.331), while that of KGd_0.45(WO₄)_1.8(MoO₄)_0.2:Eu³⁺_0.55 is (x=0.666, y=0.334) in VUV region.     

Synthesis, characterization and melt processing of Y1−x(Yb0.9Nd0.1)xBa2Cu3Oz superconductors

Effects of a combined substitute of Yb and Nd on Y site on the superconducting properties of YBa₂Cu₃O_y have been studied. We synthesized Y_1−x(Yb_0.9Nd_0.1)_xBa₂Cu₃O_z compound with x = 0.2, 0.4, 0.6, 0.8 and 1.0. Here, the ratio of Yb–Nd was fixed to be 9:1 for obtaining 123 phase without secondary phases. The melt processing thermal profiles for Y_1−x(Yb_0.9Nd_0.1)_xBa₂Cu₃O_z with x = 0.2 and 0.4 and the addition of 40 mol% {Y_1−x(Yb_0.9Nd_0.1)_x}₂BaCuO₅ and 0.5 wt% Pt in air were determined on the basis of the thermal analysis results. All samples showed a low grain growth rate, particularly for high x values, which may be partially ascribed to un-optimized thermal schedules. Although almost all the samples exhibited low J_c values, the sample with x = 0.2 exhibited T_c of 88.8 K and a relatively high J_c value of 16,000 A/cm² at 77 K for H//c-axis.     

Double-ceramic-layer thermal barrier coatings based on La2(Zr0.7Ce0.3)2O7/La2Ce2O7 deposited by electron beam-physical vapor deposition

Double-ceramic-layer (DCL) thermal barrier coatings (TBCs) of La₂(Zr_0.7Ce_0.3)₂O₇ (LZ7C3) and La₂Ce₂O₇ (LC) were deposited by electron beam-physical vapor deposition (EB-PVD). The composition, interdiffusion, surface and cross-sectional morphologies, cyclic oxidation behavior of DCL coating were studied. Energy dispersive spectroscopy and X-ray diffraction analyses indicate that both LZ7C3 and LC coatings are effectively fabricated by a single LZ7C3 ingot with properly controlling the deposition energy. The chemical compatibility of LC coating and thermally grown oxide (TGO) layer is unstable. LaAlO₃ is formed due to the chemical reaction between LC and Al₂O₃ which is the main composition of TGO layer. Additionally, the thermal cycling behavior of DCL coating is influenced by the interdiffusion of Zr and Ce between LZ7C3 and LC coatings. The failure of DCL coating is a result of the sintering of LZ7C3 coating surface, the chemical incompatibility of LC coating and TGO layer and the abnormal oxidation of bond coat. Since no single material that has been studied so far satisfies all the requirements for high temperature applications, DCL coating is an important development direction of TBCs.     

Scanning tunneling microscopy/spectroscopy on multi-layered cuprate superconductor Ba2Ca5Cu6O12 (O1−x Fx)2

Scanning tunneling microscopy/spectroscopy (STM/STS) measurements on multi-layered cuprate superconductor Ba₂Ca₅Cu₆O₁₂ (O_1−x F_x)₂ are carried out. STM topographies show randomly distributed bright spot structures with a typical spot size of 0.8 nm. These bright spots are occupied about 28% per one unit cell of c-plane, which is comparable to the regular amount of apical oxygen of 20% obtained from element analysis. Tunneling spectra simultaneously show both the small and the large gap structures. These gap sizes at 4.9 K are about Δ 15 meV and 90 meV, respectively. The small gap structure disappears at the temperature close to T_C, while the large gap persists up to 200 K. Therefore, these features correspond to the superconducting gap and pseudogap, respectively. These facts give evidence for some ordered state with large energy scale even in the superconducting state. For the superconducting gap, the ratio of 2Δ/K_BT_C = 4.9 is obtained with T_C = 70 K, which is determined from temperature dependence of the tunneling spectra.     

Effects of RE2O3 (RE = Tm, Sc, Yb) addition on the superconducting properties of ErBa2Cu3Oy

We investigated the effects of added Tm₂O₃, Sc₂O₃, and Yb₂O₃ on the superconducting properties of sintered Er123 samples. Tm₂O₃ addition caused the least T_c degradation, exhibiting a T_c above 90 K even for 17 vol% addition. Samples with added Sc₂O₃ maintained a T_c at above 90 K up to an addition of 7.2 vol%, while Yb₂O₃-containing samples showed a monotonic decrease in T_c with increased vol% of added Yb₂O₃. Tm₂O₃-containing samples exhibited a slight increase in J_c(0.1 T)/J_c(0) and had constant J_c values even for 17 vol% addition. XRD and SEM results indicate that the Tm₂O₃ is very stable in the superconducting matrix.     

Chemical quenching of positronium in Fe2O3/Al2O3 catalysts

Fe₂O₃/Al₂O₃ catalysts were prepared by solid state reaction method using α-Fe₂O₃ and γ-Al₂O₃ nano powders. The microstructure and surface properties of the catalyst were studied using positron lifetime and coincidence Doppler broadening annihilation radiation measurements. The positron lifetime spectrum shows four components. The two long lifetimes τ₃ and τ₄ are attributed to positronium annihilation in two types of pores distributed inside Al₂O₃ grain and between the grains, respectively. With increasing Fe₂O₃ content from 3 wt% to 40 wt%, the lifetime τ₃ keeps nearly unchanged, while the longest lifetime τ₄ shows decrease from 96 ns to 64 ns. Its intensity decreases drastically from 24% to less than 8%. The Doppler broadening S parameter shows also a continuous decrease. Further analysis of the Doppler broadening spectra reveals a decrease in the p-Ps intensity with increasing Fe₂O₃ content, which rules out the possibility of spin-conversion of positronium. Therefore the decrease of τ₄ is most probably due to the chemical quenching reaction of positronium with Fe ions on the surface of the large pores.