Modeling of trap-assisted tunneling in AlGaN/GaN heterostructure field effect transistors with different Al mole fractions

Although significant progress has been achieved in GaN based high power/high frequency electronic devices, surface-related problems still need an immediate solution. In particular, leakage currents through Schottky contacts not only impede device reliability but also degrade power efficiency and noise performance in such devices. This article discusses the mechanism of leakage currents through GaN Schottky and AlGaN/GaN Schottky interfaces for both forward and reverse biases. A theoretical model for the calculation of currents based on trap-assisted tunneling is discussed. In the calculation the trap energy has been assumed as a fitting parameter which is about 0.48 eV for different Al mole fractions. The comparison of the results obtained with the existing experimental data in the literature shows a good agreement.     

Mg-doping experiment and electrical transport measurement of boron nanobelts

We measured electrical conductance of single crystalline boron nanobelts having α-tetragonal crystalline structure. The doping experiment of Mg was carried out by vapor diffusion method. The pure boron nanobelt is a p-type semiconductor and its electrical conductivity was estimated to be on the order of 10^-3 (Ω cm)⁻¹ at room temperature. The carrier mobility of pure boron nanobelt was measured to be on the order of 10⁻³ (cm² Vs⁻¹) at room temperature and has an activation energy of ∼0.19 eV. The Mg-doped boron nanobelts have the same α-tetragonal crystalline structure as the pristine nanobelts. After Mg vapor diffusion, the nanobelts were still semiconductor, while the electrical conductance increased by a factor of 100-500. Transition to metal or superconductor by doping was not observed.     

Valence band offset of MgO/TiO2 (rutile) heterojunction measured by X-ray photoelectron spectroscopy

The valence band offset (VBO) of MgO/TiO₂ (rutile) heterojunction has been directly measured by X-ray photoelectron spectroscopy. The VBO of the heterojunction is determined to be 1.6 ± 0.3 eV and the conduction band offset (CBO) is deduced to be 3.2 ± 0.3 eV, indicating that the heterojunction exhibits a type-I band alignment. These large values are sufficient for MgO to act as tunneling barriers in TiO₂ based devices. The accurate determination of the valence and conduction band offsets is important for use of MgO as a buffer layer in TiO₂ based field-effect transistors and dye-sensitized solar cells.     

直接调制半导体激光器的脉冲驱动电路研究

研制了一种应用于直接调制半导体激光器的脉冲驱动电路.根据直接数字频率合成原理,利用现场可编程逻辑阵列产生频率可调方波去触发雪崩晶体管,以及雪崩晶体管的雪崩效应产生频率可调脉宽固定在3 ns的主体脉冲.该脉冲经过衰减器衰减得到合时的脉冲幅度.输出脉冲最高频率可到100 kHz,其输出电压幅度为9 V.     

基于退化数据和DBN算法的IGBT健康参数预测方法

绝缘栅双极型晶体管(IGBT)等电子元器件被广泛用于运输和能源部门,其健康状态对于设备安全和有效至关重要;在对IGBT的结构和损伤机制分析基础上,结合NASA艾姆斯中心开展的IGBT加速退化试验,选择集电极-发射极关断峰值电压作为失效特征参数,提出了一种基于深度信念网络的预测模型对其进行分析和预测;以Levenberg-Marquardt(LM)算法模型作为对比,实验结果显示文章提出的三隐藏层DBN模型相比于LM模型有更好的预测性能和更高的预测精度。     

Control of electrical properties and gate bias stress stability in solution-processed a-IZO TFTs by Zr doping

Zr-doped indium zinc oxide (IZO) thin film transistors (TFTs) are fabricated via a solution process with different Zr doping ratios. The addition of Zr suppressed the carrier concentration in the IZO films, which was confirmed by Hall Effect measurements. As the amount of Zr was increased in the oxide active layer of TFTs, the subthreshold swing (S.S) reduced, the ON/OFF ratio improved, and the threshold voltage (V_th) shifted positively. Moreover, the starting points of the ON state for TFTs near the point zero gate voltage could be controlled by the addition of Zr. The 0.3% Zr-doped IZO TFT exhibited a high saturation mobility of 7.0 cm² V⁻¹ s⁻¹, ON/OFF ratio of 2.6 × 10⁶ and S.S of 0.57 V/decade compared the IZO TFT with 10.1 cm² V⁻¹ s⁻¹, 1.7 × 10⁶ and 0.75 V/decade. The Zr effect of the gate bias stability was examined. Zr-doped IZO TFTs were relatively unstable under a positive bias stress (PBS), whereas they showed good stability at a negative bias stress (NBS). The gate bias stability of the oxide TFTs were compared with the extracted parameters through a stretched-exponential equation. The characteristic trapping time under NBS of 0.3% Zr-doped IZO TFTs was improved from 8.3 × 10⁴ s for the IZO TFT to 3.1 × 10⁵ s.     

Subthreshold swing characteristics of nanowire tunneling FETs with variation in gate coverage and channel diameter

In this study, we demonstrate the simulated subthreshold swing (SS) of silicon nanowire tunneling field-effect transistors (NWTFETs) by varying both the channel diameter from 10 nm to 40 nm and the gate coverage ratio from 30% to 100%. Our simulation work reveals that both a decrease in the channel diameter and an increase in the gate coverage ratio contribute to a reduction in the SS. Additionally, our work shows that the magnitude of the on-current depends linearly on the gate coverage ratio and that the drain current increases with a decrease in the channel diameter. Thus, an NWTFET with a channel diameter of 10 nm and a gate coverage ratio of 100% exhibits superior electrical characteristics over other silicon NWTFETs in that the NWTFET shows a point SS of 22.7 mV/dec, an average SS of 56.3 mV/dec, an on/off current ratio of ∼10¹³, and an on-current of ∼10⁻⁵ A/μm.     

UWB LOS/NLOS identification in multiple indoor environments using deep learning methods

Indoor location-aware service is booming in daily life and business activities, making the demand for precise indoor positioning systems thrive. The identification between line-of-sight (LOS) and non-line-of-sight (NLOS) is critical for wireless indoor time-of-arrival-based localization methods. Ultra-Wide-Band (UWB) is considered low cost among the many wireless positioning systems. It can resolve multi-path and have high penetration ability. This contribution addresses UWB NLOS/LOS identification problem in multiple environments. We propose a LOS/NLOS identification method using Convolutional Neural Network parallel with Gate Recurrent Unit, named Indoor NLOS/LOS identification Neural Network. The Convolutional Neural Network extracts spatial features of UWB channel impulse response data. While the Gate Recurrent Unit is an effective approach for designing deep recurrent neural networks which can extract temporal features. By integrating squeeze-and-extraction blocks into these architectures we can assign weights on channel-wise features. We simulated UWB channel impulse response signals in residential, office, and industrial scenarios based on the IEEE 802.15.4a channel model report. The presented network was tested in simulation scenarios and an open-source real-time measured dataset. Our method can solve NLOS identification problems for multiple indoor environments. Thus more versatile compare with networks only working in one scenario. Popular machine learning methods and deep learning methods are compared against our method. The test results show that the proposed network outperforms benchmark methods in simulation datasets and real-time measured datasets.