氮化镓基蓝光发光二极管伽马辐照的1/f噪声表征

通过电离辐照对氮化镓基蓝光发光二极管器件有源区光/暗电流产生机制的研究, 建立了电离辐照减小发光二极管有效输出功率电学模型.通过电离辐照对氮化镓基蓝光发光 二极管器件有源区1/f噪声影响机制的研究, 建立了电离辐照增大发光二极管1/f噪声的相关性模型.在I < 1 μA 的小注入区,空间电荷区的复合电流随辐照剂量的增加而增加. 同时, 随着电离辐照产生缺陷的增加, 1/f噪声幅度增大. 在 I> 1 mA 的大注入条件下, 由于串联电阻的影响占主导地位,表面复合速率和电流随辐照剂量的增加而增加.同时, 随着电离辐照产生缺陷的增加, 1/f噪声幅度增大.根据辐照前后电流电压试验结果噪声测试结论, 证实了实验结论与理论推导结果的一致性. 在1 μA < I < 5×10^-5 A 的中值电流情况下, 由于高能载流子散射相关的迁移率涨落与辐照新增缺陷引起的载流子数涨落竞争机制, 随着辐照剂量增大, 1/f噪声在频域变化没有明显规律. 但是, 通过1/f噪声时域多尺度熵复杂度分析方法, 得出随着辐照剂量增大, 1/f噪声时域多尺度熵复杂度的结果. 最终证实1/f噪声幅度可以敏感地反映小注入和大注入情况下氮化镓基蓝光发光二极管电离辐照的可靠性. 噪声幅值越大, 则说明辐照感应N_it越高, 暗电流相关的复合电流越大, 光电流相关的扩散电流比例减少, 使得器件发光效率、光输出功率等性能参数下降, 继而影响器件可靠性, 造成失效率显著增大. 1/f噪声时域多尺度熵复杂度可以敏感地反映中值电流情况下氮 化镓基蓝光发光二极管的电离辐照可靠性.多尺度熵复杂度越大, 则说明辐照感应越多, 复合电流越大,器件可靠性越差.本文结论提供了一种基于 1/f噪声的氮化镓基蓝光发光二极管电离辐照可靠性表征方法. 关键词： f噪声')" href="#">1/f噪声 电离辐照 氮化镓基蓝光发光二极管     

液相外延碲镉汞薄膜表面氧化特性的光电子能谱研究

利用X射线光电子谱对液相外延HgCdTe薄膜表面氧化特性进行了 研究,对经不同工艺过程处理的HgCdTe表面进行了测量、分析,结果表明碲镉汞表面经溴 无水乙醇溶液腐蚀后,再用乳酸 乙二醇溶液对表面进行处理,可获得氧化物极少甚至无氧 化的HgCdTe表面.说明HgCdTe表面钝化前的预处理直接影响钝化层/HgCdTe的界面特性.  关键词：     

AlGaN表面坑状缺陷及GaN缓冲层位错缺陷对AlGaN/GaN HEMT电流崩塌效应的影响

利用金属有机气相外延(MOVPE)技术生长了具有不同AlGaN表面坑状缺陷和GaN缓冲层位错缺陷密度的AlGaN/GaN 高电子迁移率晶体管(HEMT)样品,并对比研究了两种缺陷对器件栅、漏延迟电流崩塌效应的影响.栅延迟测试表明,AlGaN表面坑状缺陷会引起栅延迟电流崩塌效应和源漏电阻的增加,而且表面坑状缺陷越多,栅延迟电流崩塌程度和源漏电阻的增加越明显.漏延迟测试显示,AlGaN表面坑状缺陷对漏延迟电流崩塌影响不大,而GaN缓冲层位错缺陷主要影响漏延迟电流崩塌.研究结果表明,AlGaN表面坑状缺陷和Ga 关键词： AlGaN/GaN HEMT 电流崩塌 坑状缺陷 位错缺陷     

光电耦合器件闪烁噪声模型

对电应力前后光电耦合器件的闪烁噪声（1/f噪声）进行了实验和理论研究.实验发现,应力前后1/f噪声幅值随偏置电流均有相同的变化规律：在低电流区,1/f噪声幅值与偏置电流成正比,在高电流区,1/f噪声幅值与偏置电流的平方成正比,且应力后1/f噪声幅值增大了约7倍.理论分析表明,小电流时该器件的1/f噪声为扩散1/f噪声,大电流时为复合1/f噪声,应力在器件有源区诱生的陷阱是1/f噪声增大的根本原因.基于载流子数涨落和迁移率涨落机制,建立了一个光电耦合器件1/f噪声的定量分析模型,实验结果和模型符合良好.     

基于量子限制的受主带间跃迁太赫兹探测器的制备与测量

使用分子束外延生长设备,在GaAs（100）衬底上生长了量子阱宽度为3 nm的GaAs/AlAs多量子阱样品,并在量子阱层中央进行了Be受主的δ-掺杂。根据量子限制受主从束缚态到非束缚态之间的跃迁,设计并制备了δ-掺杂Be受主GaAs/AlAs多量子阱太赫兹光探测器原型器件。在4.2 K温度下,分别对器件进行了太赫兹光电流谱和暗电流-电压曲线的测量。在6 V直流偏压下,空穴载流子沿量子阱层方向输运。当正入射激光频率为6.8 THz时,器件响应率为2×10^-4 V/W（2 μA/W）。通过器件的暗电流-电压曲线计算了器件全散粒噪声电流,在4.2 K、6 V直流偏压下,全散粒噪声电流为5.03 fA·Hz^-1/2。     

光电耦合器件1/f噪声和g-r噪声的机理研究

对光电耦合器件1/f噪声和g-r噪声(产生－复合噪声) 的偏置特性及其产生机理进行了实验和理论研究.结果表明：在低频段,光电耦合器件的g-r噪声通常表现为叠加1/f噪声,且两者均随输入电流的增加呈现先增大后减小的规律.通过测量前级噪声和后级噪声,发现光电耦合器件的g-r噪声源为后级光敏三极管.理论分析表明：光电耦合器件的1/f噪声主要为表面1/f噪声,g-r噪声则源于光敏三极管发射结空间电荷区的深能级对载流子的俘获和发射.     

基于1/f噪声的NPN晶体管辐照感生电荷的定量分离

本文针对NPN双极性晶体管, 在研究辐照感生的氧化层电荷及界面态对晶体管基极电流和1/f噪声的影响的基础上, 建立辐照感生氧化层电荷及界面态与基极电流和1/f噪声的定量物理模型. 根据所建立的模型, 提出一种新的分离方法, 利用1/f噪声和表面电流求出氧化层电荷密度, 利用所求得氧化层电荷密度和表面电流求出界面态密度. 利用本方法初步实现了辐照感生氧化层电荷及界面态的定量计算.     

有机/无机复合结构光电导型器件的光激发机制

制备了PVK/ZnS有机无机复合的光电导型器件 ,器件结构分别为Glass/ITO/PVK/Al;Glass/ITO/ZnS/Al;Glass/ITO/ZnS/PVK/Al。通过研究此复合器件在外加电场作用下的稳态光电导激发谱 ,得到了基本光激发过程。把PVK/ZnS的吸收谱和器件的光电导谱进行比较 ,知道虽然两者的吸收对器件光电流都有贡献 ,但有效部分在PVK和ZnS的界面处。最大光电流对外加电场的依赖性与器件的暗电流和光电流谱为此提供了证据     

发光二极管可靠性的噪声表征

通过对发光二极管内部结构的研究，发现N_t(界面态陷阱密度)和扩散电流比率 是影响发光二极管性能的重要因素，并与器件可靠性有密切关系.器件内部存在的多种噪声 中，低频1/f噪声可表征N_t和扩散电流比率.在深入研究发光二极管工作原理及1 /f噪声载流子数涨落理论和迁移率涨落理论的基础上，建立了发光二极管的电性能模型及1/ f噪声模型.在输入电流宽范围变化的条件下测量了器件的电学噪声，实验结果与理论模型符 合良好.通过对其测量结果分析，深入研究了噪声和发光二极管性能与可靠性的关系，证明 了噪声幅值越大，电流指数越接近于2，器件可靠性越差，失效率则显著增大. 关键词： 1/f噪声 发光二极管 陷阱 光功率     

非晶铟镓锌氧薄膜晶体管钼/铜源漏电极的研究

针对非晶铟镓锌氧薄膜晶体管（a-IGZO TFT）的钼/铜源漏电极开展研究。实验证明,单层Mo源漏电极与栅绝缘层之间的粘附性好、表面粗糙度较小、电阻率较大,而单层Cu源漏电极与栅绝缘层之间的结合性差且Cu原子扩散问题严重、表面粗糙度较大、电阻率较小。为了实现优势互补,我们设计了双层Mo（20 nm）/Cu（80 nm）源漏电极,并采用优化工艺制备了包含该电极结构的a-IGZO TFT。器件具有良好的电学特性,场效应迁移率为 8.33 cm²·V^-1·s^-1, 阈值电压为6.0 V,亚阈值摆幅为2.0 V/dec,开关比为 1.3×10⁷,证明了双层Mo/Cu源漏电极的可行性和实用性。     

Effects of passivation treatment on the surface stability in indium-doped Hg0.8Cd0.2Te epilayers grown on p-Cd0.96Zn0.04Te substrates

Passivation treatment on indium-doped Hg_0.8Cd_0.2Te epitaxial layers grown on p-Cd_0.96Zn_0.04Te substrates by molecular beam epitaxy has been performed in order to improve the surface stability of the Hg_0.8Cd_0.2Te layers. Room-temperature capacitance–voltage measurements clearly revealed metal-insulator–semiconductor (MIS) behavior for the Al/ZnS/passivated Hg_0.8Cd_0.2Te layer/Cd_0.96Zn_0.04Te diodes. The fast state density and the fixed charge density of the Al/ZnS/passivated Hg_0.8Cd_0.2Te/Cd_0.96Zn_0.04Te diode with a sulfur-treated Hg_0.8Cd_0.2Te layer were smaller than those with a chemically oxidized Hg_0.8Cd_0.2Te layer. The interface state density at the ZnS/sulfur-treated Hg_0.8Cd_0.2Te interface were low at 10¹¹ eV⁻¹ cm⁻² at the middle of the Hg_0.8Cd_0.2Te energy gap. These results indicate that the Hg_0.8Cd_0.2Te epilayer is significantly passivated by sulfur treatment and that the passivated Hg_0.8Cd_0.2Te layers can be used for Hg_1−xCd_xTe-based MIS diodes and MIS field-effect transistors.     

Characterization of limited-diffusion-volume HgCdTe/CdTe photodiodes

Thin epitaxial HgCdTe/CdTe photodiodes are shown to have lower dark current noise than similar thick devices. This is explained by the limited-diffusion-volume model which requires a thin diode and a low surface recombination rate. Electron beam induced current is used for characterizing minority electrons distribution; the characteristic interface recombinations(L/D) has been found to be equal to 0.05 at 80 and 210 K.R ₀ A measurements give values consistent with this model.     

Lasing characteristics and growth of CdS epitaxial thin-films excited by second harmonic Nd: YAG laser radiation at 473 nm

The first experimental study on CdS epitaxial thin-film lasers with ZnS substrates is reported. For the first time single crystal layers of CdS have been grown on the (111) faces of ZnS substrate by the hydrogen transport method. Laser emission at 491.8 nm has been observed from these layers cooled near liquid-nitrogen temperature. Optical pumping is performed by the second harmonic of a Q-switched Nd: YAG laser operated at 473 nm, the energy of which exceeds slightly the band-gap energy of the cooled epitaxial CdS layer. The measured optical gain for these CdS thin-films compared with that for platelet CdS single crystals reveals that the optical confinement inside the epitaxial layer increases the optical gain by a factor of about two.     

Electrical characteristics of the hydrogen pre-annealed Au/n-GaAs Schottky barrier diodes as a function of temperature

We investigated the passivation effects of hydrogen gas on the Au/n-GaAs Schottky barrier diodes in a wide temperature range. Reference diodes were prepared by evaporating barrier metal on semiconductor wafers un-annealed in N₂ gas atmosphere. The other diodes were made by evaporating barrier metal on n-GaAs semiconductor substrates annealed in H₂ atmosphere. Then, electrical measurements of all diodes were carried out by using closed-cycle Helium cryostat by steps of 20 K in the temperature range of 80-300 K in dark. The basic diode parameters such as ideality factor and barrier height were consequently extracted from electrical measurements. It was seen that ideality factors increased and barrier heights decreased with the decreasing temperature. The case was attributed to barrier inhomogeneity at the metal/semiconductor interface. Barrier heights of the diodes made from samples annealed in H₂ gas atmosphere were smaller than those of reference diodes at low temperatures. Here, it was ascribed to the fact that hydrogen atoms passivated dangling bonds on semiconductor surface in accordance with former studies.     

Modeling and simulation of long-wave infrared InAs/GaSb strained layer superlattice photodiodes with different passivants

Current–voltage characteristics of long-wave infrared (LWIR) InAs/GaSb strained layer superlattice photodiodes (cut-off wavelength ∼10 μm), passivated with different surface passivants, have been modeled and simulated using ATLAS software from SILVACO. The simulated results are fitted to previous experimental results obtained on unpassivated devices and those passivated by silicon-dioxide (SiO₂), silicon nitride (Si_xN_y) and zinc sulfide (ZnS). Surface parameters in terms of surface recombination velocity, shunt resistance and interface trap density are extracted for different passivants. The performance of silicon-dioxide passivated diode is solely dominated by a shunt leakage path with a shunt resistance value of 0.56 Ω-cm². Extracted electron and hole surface recombination velocities have values of 10⁵ cm/s and 10⁷ cm/s for unpassivated, 10³ cm/s and 10⁵ cm/s for Si_xN_y passivated and 10² cm/s and 10³ cm/s for ZnS passivated devices. Interface trap density follows a similar trend with values of 10¹⁵ cm⁻², 8.5 × 10¹⁴ cm⁻² and 10¹⁰ cm⁻² for unpassivated, Si_xN_y passivated and ZnS passivated devices respectively. The suitability and limitations of the simulation tool are discussed.     

Analytical modelling of carrier transport mechanisms in long wavelength planar n+–p HgCdTe photovoltaic detectors

The dark electrical characteristic of n⁺ on p long wavelength Hg_1−xCd_xTe photovoltaic detectors has been studied as a function of applied bias voltages. The diodes under study were given different surface treatments prior to their passivation. The dark current components have been identified using the already known carrier transport mechanisms across the junction. It is reported that the diodes under investigation have an excess leakage current component that exhibits quadratic dependence on the applied reverse bias voltage across the diode. The origin of this excess current is found to be closely associated with the ohmic currents. The diodes with higher ohmic current exhibit higher excess leakage current.     

Linear HgCdTe IR FPA 288×4 with bidirectional scanning

The long wavelength (8–12 μm) IR FPA 288×4 based on a hybrid assembly of n⁺-p diode photosensitive arrays (PA) of HgCdTe (MCT) MBE-grown structures and time delay integration (TDI) readout integrated circuits (ROIC) with bidirectional scanning have been developed, fabricated, and investigated. The p-type MCT structures were obtained by thermal annealing of as-grown n-type material in inert atmosphere. The MCT photosensitive layer with the composition 0.20–0.23 of mole fraction of CdTe was surrounded by the wide gap layers to decrease the recombination rate and surface leakage current. The diode arrays were fabricated by planar implantation of boron ions into p-MCT. The typical dark currents were about 4–7 nA at the reverse bias voltage of 150 mV. The differential resistance R was up to R₀ = 1.6×10⁷ Ω zero bias voltage, which corresponded to R₀A ～70 Ω ·cm² and to the maximal value R_max = 2.1 × 10⁸ Ω. The bidirectional TDI deselecting ROIC was developed and fabricated by 1.0-μm CMOS technology with two metallic and two polysilicon layers. The IR FPAs were free of defect channels and have the average values of responsivity S_λ = 2.27×10⁸ V/W, the detectivity D_λ ^* = 2.13 × 10¹¹ cm × Hz^1/2 × W^t1, and the noise equivalent temperature difference NETD = 9 mK.     

Numerical analysis of In0.53Ga0.47As/InP single photon avalanche diodes

A rigorous theoretical model for In 0.53 Ga 0.47 As/InP single photon avalanche diode is utilized to investigate the dependences of single photon quantum efficiency and dark count probability on structure and operation condition.In the model,low field impact ionizations in charge and absorption layers are allowed,while avalanche breakdown can occur only in the multiplication layer.The origin of dark counts is discussed and the results indicate that the dominant mechanism that gives rise to dark counts depends on both device structure and operating condition.When the multiplication layer is thicker than a critical thickness or the temperature is higher than a critical value,generation-recombination in the absorption layer is the dominative mechanism;otherwise band-to-band tunneling in the multiplication layer dominates the dark counts.The thicknesses of charge and multiplication layers greatly affect the dark count and the peak single photon quantum efficiency and increasing the multiplication layer width may reduce the dark count probability and increase the peak single photon quantum efficiency.However,when the multiplication layer width exceeds 1 μm,the peak single photon quantum efficiency increases slowly and it is finally saturated at the quantum efficiency of the single photon avalanche diodes.     

Preparation and characterization of ZnS/CdS bi-layer for CdTe solar cell application

In this work, bilayer ZnS/CdS film was prepared as an improved window layer of CdTe solar cell. TEM was used to observe the cross section of the bilayer structure. The total thickness of ZnS/CdS film was about 60 nm, which could allow more photons to pass through it and contribute to the photocurrent. Optical properties of the bilayers were investigated using UV–vis spectroscopy. Compared with poor transmission of standard CdS film in the short wavelength range of 350–550 nm, the transmission of ZnS/CdS was improved and reached above 50%. The ZnS/CdS was annealed with CdCl₂. X-ray photoelectron spectroscopy (XPS) was used to investigate its chemical properties. A possible diffusion between CdS and ZnS was observed after annealing. The efficiency of standard CdS/CdTe solar cell was 9.53%. The device based on ZnS/CdS window layer had a poor 6% efficiency. With annealing treatment on ZnS/CdS layer, the performance was improved and reached 10.3%. In addition, the homogeneity of solar cell performance was improved using ZnS/CdS window layer. A thin ZnS layer was quite effective to reduce the possible shunt paths and short parts of window layer and consequently contributed to fabrication of a homogeneous CdTe solar cell.     

Epitaxial growth of MnGa/GaAs layers for diodes with spin injection

The possibility of the epitaxial growth of ferromagnetic manganese gallide (Mn₃Ga₅) layers on a the GaAs(100) surface was demonstrated. The ferromagnetic properties of epitaxial Mn₃Ga₅ at room temperature were evaluated from measurements of the anomalous Hall effect. A diode structure based on the Mn₃Ga₅/GaAs contact was formed, and the low-temperature electroluminescence of this diode was measured. The possibility of electroluminescence and the high crystal quality of the structures under study showed promises of their application in light-emitting diodes with spin injection.