超薄栅下LDD nMOSFET器件GIDL应力下退化特性

对1.4nm超薄栅LDD nMOSFET器件栅致漏极泄漏GIDL(gate-induced drain leakage)应力下的阈值电压退化进行了研究.GIDL应力中热空穴注进LDD区界面处并产生界面态，这导致器件的阈值电压变大.相同栅漏电压V_DG下的不同GIDL应力后阈值电压退化量的对数与应力V_D/V_DG的比值成正比.漏偏压V_D不变的不同GIDL应力后阈值电压退化随着应力中栅电压的增大而增大，相同栅偏压V_G下的不同GIDL应力后阈值电压退化也随着应力中漏电压的增大而增大，这两种应力情形下退化量在半对数坐标下与应力中变化的电压的倒数成线性关系，它们退化斜率的绝对值分别为0.76和13.5.实验发现器件退化随着应力过程中的漏电压变化远大于随着应力过程中栅电压的变化. 关键词： 栅致漏极泄漏 CMOS 阈值电压 栅漏电压     

总剂量辐照条件下部分耗尽半导体氧化物绝缘层N沟道金属氧化物半导体器件的三种kink效应

研究了0.8 μm SOINMOS晶体管,经过剂量率为50 rad (Si)/s的⁶⁰Co γ射线辐照后的输出特性曲线的变化趋势. 研究结果表明, 经过制造工艺和版图的优化设计, 在不同剂量条件下, 该样品均不产生线性区kink效应. 由碰撞电离引起的kink效应, 出现显著变化的漏极电压随总剂量水平的提高不断增大. 在高剂量辐照条件下, 背栅I_D-V_SUB曲线中出现异常的"kink"现象, 这是由辐照诱生的顶层硅膜/埋氧层之间的界面陷阱电荷导致的.     

硅薄膜沉积中等离子体辉光功率和阻抗的测试分析

采用改进的电压和电流测试方法对衬底电极外加不同电压V_s的射频等离子体中的阻抗和功率消耗进行了测试分析.结果表明：在射频电极施加恒定的电压V_el时，随衬底电极外加电压V_s的增加，辉光的电流在增加，结果导致阻抗在减小；另外，通过计算分析发现：仅有一小部分功率用于辉光，大部分功率消耗在匹配器和电缆上.通过对等离子体电学特性的综合测试分析也说明：在保证有足够多的硅烷时，衬底电极外加电压V关键词： 等离子体 辉光功率 阻抗 诊断     

Significant performance enhancement in AlGaN/GaN high electron mobility transistor by high-κ organic dielectric

The electrical properties of AlGaN/GaN high electron mobility transistor (HEMT) with and without high-κ organic dielectrics are investigated. The maximum drain current I_{D max} and the maximum transconductance g_{m max} of the organic dielectric/AlGaN/GaN structure can be enhanced by 74.5%, and 73.7% compared with those of the bare AlGaN/GaN HEMT, respectively. Both the threshold voltage V_T and g_{m max} of the dielectric/AlGaN/GaN HEMT are strongly dielectric-constant-dependent. Our results suggest that it is promising to significantly improve the performance of the AlGaN/GaN HEMT by introducing the high-κ organic dielectric.     

氧非正分La0.9Ba0.1MnO3-δ/SrTiO3:Nb p-n异质结的整流特性研究

利用脉冲激光沉积技术在掺Nb的SrTiO₃衬底上制备了氧非正分La_0.9Ba_0.1MnO_3-δ/SrTiO₃:Nb p-n异质结.在20—300K这一较宽的温度范围内获得了光滑的整流曲线.整流实验表明：该p-n异质结的正向扩散电压V_D随着温度升高在薄膜金属—绝缘转变温度附近出现极大值,表现出与氧正分La_0.9Ba_{0. 关键词： 0.9}Ba_0.1MnO_3-δ薄膜')" href="#">La_0.9Ba_0.1MnO_3-δ薄膜 锰氧化物p-n结 整流性质     

离子轰击控制准直碳纳米管生长的研究

用CH₄,H₂和NH₃作为反应气体，利用等离子体增强热丝化学汽相沉积制备出准直碳纳米管，并用扫描电子显微镜研究了不同负偏压对准直碳纳米管生长的影响. 结果表明，随着负偏压的增大，准直碳纳米管的平均直径减小、平均长度增大. 由于辉光放电的产生，在衬底表面附近形成阴极鞘层，并在阴极鞘层内形成大量的离子和在衬底表面附近形成很强的电场. 离子在电场的作用下对衬底表面的强烈轰击将对准直碳纳米管的生长产生影响. 结合有关理论，分析和讨论了离子的轰击对准 关键词： 准直碳纳米管 离子轰击 负偏压     

高压双扩散漏端MOS晶体管双峰衬底电流的形成机理及其影响

利用0.15μm标准CMOS工艺制造出了工作电压为30V的双扩散漏端MOS晶体管(double diffused drain MOS, DDDMOS).观察到DDDMOS的衬底电流-栅压曲线(I_b-V_g曲线)有两个峰.通过实验和TCAD模拟揭示了DDDMOS衬底电流的形成机理,发现衬底电流第一个峰的成因与传统MOS器件相同;第二个峰来自于发生在漂移区远离沟道一侧高场区的碰撞离化电流.通过求解泊松方程和电流连续性方程,分析了器件的物理和几何参数对导致衬底电流重新上升的漂移区电场的影响.在分析了DDDMOS衬底电流的第二个峰形成机理的基础上,考察了其对器件的可靠性的影响. 关键词： 高压器件 衬底电流 可靠性     

Sr2CoO4-δ薄膜中的自旋玻璃态的磁性质研究

用脉冲激光沉积方法(PLD)在铝酸镧衬底上制备了c取向的高氧空位含量的锶钴氧薄膜.X射线衍射分析表明薄膜单一取向且没有明显杂相.原位的高气压反射式高能电子衍射仪(RHEED)监测显示,薄膜为层状生长.通过对薄膜磁化强度随温度、磁场及时间的变化曲线进行测量,发现零场冷曲线上可能存在两个特征温度:T_f和T_a.T_f为对应玻璃态的冻结温度而T_a对应少量的不缺 关键词： 自旋玻璃 超交换相互作用 双交换相互作用 脉冲激光沉积     

有机电致发光器件的动态电学特性

利用交流阻抗谱技术，研究了有机发光二极管ITO/Alq₃(90 nm)/Al的载流子传导机理.根据器件对不同频率的响应曲线及其等效电路模型，该器件可看作是由并联的电阻R_p和电容C_p再与电阻R_s串联而成，并根据实验数据求出了R_p，C_p和R_s的数值.实验结果表明器件的载流子传输机理属于指数分布式的陷阱电荷限制电流，其介电弛豫时间随偏压的增加而逐渐减小. 关键词： 3')" href="#">Alq₃ 陷阱电荷限制电流 交流阻抗谱 有机发光二极管     

不同堆砌层数煤系石墨的拉曼光谱表征及其表面石墨化均匀程度

采用多激发波长（325,405,514,633和785 nm）以及显微拉曼面扫描技术对不同芳香层片平均堆砌层数的煤系石墨及其表面石墨化均匀程度进行表征。结果表明：对无序石墨,石墨微晶的尺寸较小并任意取向,随着平均堆砌度及堆砌层数增加,石墨微晶边缘的拉曼光谱特征显现。在无序结构向有序转化的同时,石墨微晶缺陷逐渐消亡,拉曼光谱一级模中D3、D4峰逐渐不显著或消失,但是其倍频峰均微弱出现,尤其2D1峰强度逐渐增大。将I_D1/I_D2参数的含义进一步引申为缺陷类型及平均定向性,且无烟煤的I_D1/I_D2极大,随着石墨微晶尺寸增大（d₀₀₂＜0.344 0 nm）,至三维有序结构的石墨时I_D1/I_D2最小。在不同激发波长下G峰半高宽总是随着无序度的减小而减小,D1峰和2D1峰等均显示较强的色散效应,各峰强度随激发光能量的增大而增大,在紫外激发下,D1和G峰峰位差显著小于可见光激发。随激发波长的增加,D1峰向着低波数方向移动,2D1峰色散约为D1峰的两倍。高煤级煤石墨化过程中,非定向的芳香碳经历一系列的物理、化学结构演变产生各种中间相态,如残留煤岩显微组分（变镜质组和变惰质组）和新生的石墨组分（热解炭等）共存,因此采用(I_G-I_D1)/(P_G-D1)≥0.3,I_D1/I_G＜0.4,A_D1/A_(D1+G)＜0.45等作为石墨和半石墨的界线,利用平面扫描区域成像来表征样品石墨化的表面均匀程度,取频数分布置信区间≤0.9来综合判定样品表面石墨化度为84.16%~86.40%,平均为85.49%,与利用X射线衍射(XRD)参数估计的石墨化度相当。     

65 nm互补金属氧化物半导体场效应和晶体管总剂量效应及损伤机制

对65 nm互补金属氧化物半导体工艺下不同尺寸的N型和P型金属氧化物半导体场效应晶体管(NMOSFET和PMOSFET)开展了不同偏置条件下电离总剂量辐照实验.结果表明:PMOSFET的电离辐射响应与器件结构和偏置条件均有很强的依赖性,而NMOSFET表现出较强的抗总剂量性能;在累积相同总剂量时,PMOSFET的辐照损伤远大于NMOSFET.结合理论分析和数值模拟给出了PMOSFET的辐射敏感位置及辐射损伤的物理机制.     

Direct growth of graphene films without catalyst on flexible glass substrates by PECVD

A hydrogen-plasma-etching-based plasma-enhanced chemical vapor deposition(PECVD) synthesis route without metal catalyst for preparing the graphene films on flexible glass is developed. The quality of the prepared graphene films is evaluated by scanning electron microscopy, x-ray photoelectron spectroscopy, high-resolution transmission electron microscopy, ultraviolet-visible spectroscopy, and electrochemical measurements. In a radio frequency(RF) power range of50 W–300 W, the graphene growth rate increases with RF power increasing, while the intensity ratio of D-to G-Raman peak(ID/IG) decreases. When the RF power is higher than 300 W, the ID/IGrises again. By optimizing experimental parameters of hydrogen plasma etching and RF power, the properties of as-prepared flexible graphene on glass are modulated to be able to achieve the graphene's transparency, good electrical conductivity, and better macroscopic uniformity.Direct growth of graphene film without any metal catalyst on flexible glass can be a promising candidate for applications in flexible transparent optoelectronics.     

Analysis of frequency-dependent series resistance and interface states of In/SiO2/p-Si (MIS) structures

In this work, the investigation of the interface state density and series resistance from capacitance–voltage (C–V) and conductance–voltage (G/ω−V) characteristics in In/SiO₂/p-Si metal–insulator–semiconductor (MIS) structures with thin interfacial insulator layer have been reported. The thickness of SiO₂ film obtained from the measurement of the oxide capacitance corrected for series resistance in the strong accumulation region is 220 Å. The forward and reverse bias C–V and G/ω−V characteristics of MIS structures have been studied at the frequency range 30 kHz–1 MHz at room temperature. The frequency dispersion in capacitance and conductance can be interpreted in terms of the series resistance (R_s) and interface state density (D_it) values. Both the series resistance R_s and density of interface states D_it are strongly frequency-dependent and decrease with increasing frequency. The distribution profile of R_s–V gives a peak at low frequencies in the depletion region and disappears with increasing frequency. Experimental results show that the interfacial polarization contributes to the improvement of the dielectric properties of In/SiO₂/p-Si MIS structures. The interface state density value of In/SiO₂/p-Si MIS diode calculated at strong accumulation region is 1.11×10¹² eV⁻¹ cm⁻² at 1 MHz. It is found that the calculated value of D_it (≈10¹² eV⁻¹ cm⁻²) is not high enough to pin the Fermi level of the Si substrate disrupting the device operation.     

Free-volume changes at nanoscale in doped polyacrylic acid studied by positron annihilation spectroscopy

Polyacrylic acid (PAA) doped with carbon black (CB), chromium oxide (Cr₂O₃) and cupferron with different wt% (0.25%, 0.50%, 0.75%, and 1%) was studied using positron annihilation lifetime (PAL) technique and Doppler broadening of annihilation radiation (DBAR). Ortho-positronium lifetime components (τ₃ and I₃) were used to estimate the nanoscale free-volume hole sizes (V_f) and its fractions (f). It was found that the hole size V_f and its fractions f as well as S-parameters decreased at high value of doping concentration due to dopants-polymer formation. These results are supported by a significant narrowing in the nanoscale free-volume hole size distributions.

The correlation between positron annihilation parameters and electric conductivity are discussed.     

Effect of substrate bias on SE, XPS and XAES studies of diamond-like carbon films deposited by saddle field fast atom beam source

This paper reports the effect of positive substrate bias (V_s) varying from 0 to 180 V on the spectroscopic ellipsometry (SE), X-ray photoelectron spectroscopy (XPS) and X-ray Auger electron spectroscopy (XAES) studies of diamond-like carbon (DLC) films deposited using CH₄ gas as a feedstock into a saddle field fast atom beam (FAB) source. The values of optical constants like refractive index (n) and extinction coefficient (k) of the deposited DLC films were determined using a two phase model. The values of ‘n’ were found to fall in the range from 1.505 to 1.720 and ‘k’ from 0.03 to 0.125 by application of different values of V_s. Value of these optical constants were found to decrease with the increase of substrate bias up to 90 V and then increase beyond this value. Position of C 1s peak evaluated from XPS data was found to occur at 286.09±0.18 eV in DLC films deposited by application of different values of V_s. Observation of full width at half maximum (FWHM) (τ) value (1.928 eV at V_s=0 V, 2.0 eV at V_s=90 V and 1.89 eV at V_s=180 V) clearly hinted the existence of a point of inflection in the properties of DLC films deposited using FAB source this way. A parameter ‘D’ defined as the distance between the maximum of positive going excursion and the minimum of negative going excursion was calculated in the derivative XAES spectra. The values of ‘D’ evaluated from XAES data for DLC films were found to be 14.8, 14.5 and 15.2 at V_s=0, 90 and 180 V, respectively. The sp² percentage was calculated for samples deposited this way and was found to be low and lie approximately at 5.6, 2.8, 2.3, 5.7 and 11.5 for different values of V_s=0, 50, 90, 150 and 180 V. The sp³ content percentage and sp³/sp² ratio was found to be 94.4 and 16.7, 97.7 and 42.5 at V_s=0 and 90 V, respectively. Beyond V_s=90 V these values started decreasing. Mainly, a point of inflection in all the properties of DLC films studied over here at around 90 V of applied substrate bias has been observed, which has been explained on the basis of existing theories in the literature.     

Effect of gate length on breakdown voltage in AlGaN/GaN high-electron-mobility transistor

The effects of gate length L_G on breakdown voltage VBRare investigated in AlGaN/GaN high-electron-mobility transistors(HEMTs) with L_G= 1 μm~20 μm. With the increase of L_G, VBRis first increased, and then saturated at LG= 3 μm. For the HEMT with L_G= 1 μm, breakdown voltage VBRis 117 V, and it can be enhanced to 148 V for the HEMT with L-_G= 3 μm. The gate length of 3 μm can alleviate the buffer-leakage-induced impact ionization compared with the gate length of 1 μm, and the suppression of the impact ionization is the reason for improving the breakdown voltage.A similar suppression of the impact ionization exists in the HEMTs with LG 3 μm. As a result, there is no obvious difference in breakdown voltage among the HEMTs with LG= 3 μm~20 μm, and their breakdown voltages are in a range of 140 V–156 V.     

碳纳米管掺杂聚丙烯腈/铜室温环化为石墨的拉曼光谱分析及形成机制研究

石墨质碳质材料因具有良好的电学、力学、热学性能而在电子设备,复合材料,电池,传感器中得到广泛应用,但针对生产能耗高,污染大,成本高,不可控等现状是急需解决的核心问题。因此,通过使用较简单和成本低的制造技术在纳米级器件中获得石墨碳结构的方法是一个有吸引力的探索领域。表面等离激元技术因具有环境友好、能耗低等优点而受到广泛关注,利用等离激元技术诱导大分子链状聚合物石墨化就是一种具有广阔前景的制备技术,而Cu作为贱金属具有产量高,价格便宜等优势。基于表面等离激元技术,利用激光辐射粗糙Cu表面上的聚丙烯腈（PAN）+碳纳米管（CNT）,而使聚丙烯腈在金属表面被石墨化。通过改变基底刻蚀时间、退火温度、退火时间、激光强度系统地研究了PAN/Cu和PAN+CNT/Cu得到最佳石墨化条件。实验结果表明：以PAN作为探针分子,在2.5 mol·L-1硝酸刻蚀15min的铜基底上,观察到了增强因子为1.39×104的表面增强拉曼散射（SERS）效应。通过使用拉曼激光作为光源,在退火温度为140 ℃时,可以观察到石墨化的PAN分子结构缺陷较少,碳氮三键消失,其I_D/I_G可达1.160 8。CNT进一步用于改变粗糙铜基底的光催化性能,我们使用硝酸改性的多臂碳纳米管（MWCNT）与PAN结合对催化系统进行改进,当掺入2%CNT后,通过表面等离激元PAN可以在40 ℃的条件下实现石墨化,其I_D/I_G达到0.942 1,并且激光的引入大大提高了石墨化位点的可控性,将其归因于激光照射下铜表面产生的热电子对PAN的催化作用,并提出可能存在两种催化和石墨化的机制,一种为热电子通过CNT使PAN石墨化,另一种为热电子通过CNT作用于PAN附近的O₂,通过·O-₂使PAN石墨化。     

Abnormal phenomenon of source-drain current of AlGaN/GaN heterostructure device under UV/visible light irradiation

We report an abnormal phenomenon that the source-drain current (I_D) of AlGaN/GaN heterostructure devices decreases under visible light irradiation. When the incident light wavelength is 390 nm, the photon energy is less than the band gaps of GaN and AlGaN whereas it can causes an increase of I_D. Based on the UV light irradiation, a decrease of I_D can still be observed when turning on the visible light. We speculate that this abnormal phenomenon is related to the surface barrier height, the unionized donor-like surface states below the surface Fermi level and the ionized donor-like surface states above the surface Fermi level. For visible light, its photon energy is less than the surface barrier height of the AlGaN layer. The electrons bound in the donor-like surface states below the Fermi level are excited and trapped by the ionized donor-like surface states between the Fermi level and the conduction band of AlGaN. The electrons trapped in ionized donor-like surface states show a long relaxation time, and the newly ionized donor-like surface states below the surface Fermi level are filled with electrons from the two-dimensional electron gas (2DEG) channel at AlGaN/GaN interface, which causes the decrease of I_D. For the UV light, when its photon energy is larger than the surface barrier height of the AlGaN layer, electrons in the donor-like surface states below the Fermi level are excited to the conduction band and then drift into the 2DEG channel quickly, which cause the increase of I_D.     

Generalized Einstein manifolds

A Finslerian manifold is called a generalized Einstein manifold (GEM) if the Ricci directional curvature R(u,u) is independent of the direction. Let F⁰(M, g_t) be a deformation of a compact n-dimensional Finslerian manifold preserving the volume of the unitary fibre bundle W(M). We prove that the critical points g₀ F⁰(g_t) of the integral I(g_t) on W(M) of the Finslerian scalar curvature (and certain functions of the scalar curvature) define a GEM. We give an estimate of the eigenvalues of Laplacian Δ defined on W(M) operating on the functions coming from the base when (M, g) is of minima fibration with a constant scalar curvature H admitting a conformal infinitesimal deformation (CID). We obtain λ ≥ H/(n − 1) (Δf = λf). If M is simply connected and λ = H/(n − 1), then (M, g) is Riemannian and is isometric to an n-sphere. We first calculate, in the general case, the formula of the second variationals of the integral I (g_t) for G = g₀, then for a CID we show that for certain Finslerian manifolds, I″(g₀) > 0. Applications to the gravitation and electromagnetism in general relativity are given. We prove that the spaces characterizing Einstein-Maxwell equations are GEMs.     

含空位二维六角氮化硼电子和磁性质的密度泛函研究

基于密度泛函理论的第一性原理计算, 研究了含B原子空位(V_B), N原子空位(V_N), 以及含B–N键空位 (V_B+N)缺陷的二维氮化硼(h-BN)的电子和磁性质. 在微观结构上, V_B体系表现为在空位附近的N原子重构成等腰三角形, V_N体系靠近空穴的B 原子形成等边三角形, V_B+N体系靠近空穴处的B和N原子在h-BN平面上重构为梯形. 三种空位缺陷都使h-BN的带隙类型从直接带隙转变为间接带隙. V_B体系的总磁矩为1.0 μ_B, 磁矩全部由N原子贡献. 其中空穴周围的三个N原子磁矩方向不完全一致, 存在着铁磁性和反铁磁性两种耦合方式. 对于V_N 体系, 整个晶胞内的总磁矩也为1.0 μ_B, 磁矩在空穴周围区域呈现一定的分布. 关键词： 二维h-BN 空位 电子结构 磁性