超深亚微米SOI总剂量效应泄漏电流模型

总剂量辐射效应会导致绝缘体上硅金属氧化物半导体场效应晶体管(SOI MOSFET)器件的阈值电压漂移、泄漏电流增大等退化特性。浅沟槽隔离(STI)漏电是器件退化的主要因素,会形成漏极到源极的寄生晶体管。针对130 nm部分耗尽(PD) SOI NMOSFET器件的总剂量辐射退化特性,建立了一个包含总剂量辐射效应的通用模拟电路仿真器(SPICE)模型。在BSIM SOI标准工艺集约模型的基础上,增加了STI寄生晶体管泄漏电流模型,并考虑了辐射陷阱电荷引起寄生晶体管的等效栅宽和栅氧厚度的变化。通过与不同漏压下、不同宽长比的器件退化特性的实验结果对比,该模型能够准确反映器件辐射前后的漏电流特性变化,为器件的抗辐射设计提供参考依据。     

背栅偏置对掩埋氧化层辐射感生陷阱电荷调控规律及机理

研究了辐照过程中施加背栅偏置对掩埋氧化层(BOX)辐射感生陷阱电荷的调控规律及机理。测试了130 nm部分耗尽绝缘体上硅金属氧化物半导体场效应晶体管在不同背栅偏置条件下辐射损伤规律,试验结果显示辐照过程中施加正向背栅偏置可以显著增强辐射陷阱电荷产生,而施加负向背栅偏置并不能明显抑制氧化物陷阱电荷的产生,甚至在进一步提高负向偏置时会使氧化物陷阱电荷的数量上升。通过TCAD器件模拟仿真研究了背栅调控机理,研究结果表明:尽管负向背栅偏置可使辐射感生陷阱电荷远离沟道界面,但随着BOX层厚度减薄该区域电荷仍可导致器件电参数退化,且提高负向偏置电压可进一步增强陷阱电荷产生,加剧辐射损伤。     

SOI MOSFET器件X射线总剂量效应研究

研究了以10keV X射线为辐照源对注氧隔离(SIMOX)SOI MOSFET器件进行辐照的总剂量效应.采用ARACOR 10keV X射线对埋氧层加固样品与未加固的对比样品在开态和传输门态两种辐照偏置下进行辐照,分析了器件的特性曲线,并计算了SOI MOS器件前栅与寄生背栅晶体管辐照前后的阈值电压的漂移.实验结果表明,加固样品的抗辐射性能优越,且不同的辐照偏置对SOI MOSFET器件的辐照效应产生不同的影响.     

抗辐射高压SOI埋氧总剂量效应研究

研究了抗辐射高压SOI埋氧总剂量加固技术,发现在总剂量辐射条件下不同埋氧加固工艺背栅阈值变化的情况。通过增加埋氧加固技术可以有效地抑制总剂量辐射环境下对高压器件的调制效应。     

基于薄埋氧层及三顶层硅的SOI高压LDMOS研究

详细研究了一种基于薄埋氧层及三层顶层硅衬底(Triple-Layer Top Silicon,TLTS)的SOI高压LDMOS器件。该结构在SOI介质层上界面的顶层硅内引入一高浓度n⁺层,当器件处于反向阻断状态时,高浓度n⁺区部分耗尽,漏端界面处已耗尽n⁺层内的高浓度电离施主正电荷可增强介质层电场,所产生的附加电场将调制漂移区内的电场,防止器件在漏端界面处被提前击穿,从而可在较薄的埋氧层(BOX)上获得较高耐压。在0.4 μm BOX上获得了624 V的耐压。与几种SOI器件相比,所提出的TLTS LDMOS器件具有较高优值(FOM)。     

基于薄埋氧层及三顶层硅的SOI高压LDMOS研究

详细研究了一种基于薄埋氧层及三层顶层硅衬底(Triple-Layer Top Silicon,TLTS)的SOI高压LDMOS器件。该结构在SOI介质层上界面的顶层硅内引入一高浓度n+层,当器件处于反向阻断状态时,高浓度n+区部分耗尽,漏端界面处已耗尽n+层内的高浓度电离施主正电荷可增强介质层电场,所产生的附加电场将调制漂移区内的电场,防止器件在漏端界面处被提前击穿,从而可在较薄的埋氧层(BOX)上获得较高耐压。在0.4μm BOX上获得了624V的耐压。与几种SOI器件相比,所提出的TLTS LDMOS器件具有较高优值(FOM)。     

环栅CMOS/SOI的X射线总剂量辐射效应研究

采用硅离子注入工艺对注氧隔离(SIMOX)绝缘体上硅(SOI)材料作出改性,分别在改性材料和标准SIMOXSOI材料上制作部分耗尽环型栅CMOS/SOI器件,并采用10keVX射线对其进行了总剂量辐照试验。实验表明,同样的辐射总剂量条件下,采用改性材料制作的器件与标准SIMOX材料制作的器件相比,阈值电压漂移小得多,亚阈漏电也得到明显改善,说明改性SIMOXSOI材料具有优越的抗总剂量辐射能力。     

总剂量辐射对0.5μm部分耗尽SOI CMOS器件的影响

介绍了基于SIMOX SOI晶圆的0.5μm PD SOI CMOS器件的抗总剂量辐射性能。通过CMOS晶体管的阈值电压漂移,泄漏电流和32位DSP电路静态电流随总剂量辐射从0增加到500 krad(Si)的变化来表现该工艺技术的抗电离总剂量辐射能力。对于H型(无场区边缘)NMOS晶体管,前栅阈值电压漂移小于0.1 V;对于H型PMOS晶体管,前栅阈值电压漂移小于0.15 V;未发现由辐射引起的显著漏电。32位DSP电路在500 krad(Si)范围内,静态电流小于1 m A。通过实验数据表明,在较高剂量辐射条件下,利用该工艺制造的ASIC电路拥有良好的抗总剂量辐射性能。     

辐照对SOI NMOS寄生双极晶体管的影响。

在埋氧化层厚度不同的SIMOX衬底上制备了H型栅结构器件。经过总剂量辐照后,器件的正栅亚阈值特性无明显变化,背栅亚阈值特性发生平移,但均未发生漏电,说明其抗辐照性能超过1E6rad(Si)。辐照后器件的寄生双极晶体管增益有所增大,并且与背栅阈值电压的变化趋势相似,可能是由于埋氧化层中的正电荷积累使体区电位升高,提高了发射极的发射效率。     

体效应对超深亚微米SOI器件总剂量效应的影响

研究体偏置效应对超深亚微米绝缘体上硅（SOI,Silicon-on-insulator）器件总剂量效应的影响.在TG偏置下,辐照130nm PD（部分耗尽,partially depleted）SOI NMOSFET（N型金属-氧化物半导体场效应晶体管,n-type Metal-Oxide-Semiconductor Field-Effect Transistor）器件,监测辐照前后在不同体偏压下器件的电学参数.短沟道器件受到总剂量辐照影响更敏感,且宽长比越大,辐射导致的器件损伤亦更大.在辐射一定剂量后,部分耗尽器件将转变为全耗尽器件,并且可以观察到辐射诱导的耦合效应.对于10μm/0.35μm的器件,辐照后出现了明显的阈值电压漂移和大的泄漏电流.辐照前体偏压为负时的转移特性曲线相比于体电压为零时发生了正向漂移.当体电压Vb=-1.1V时部分耗尽器件变为全耗尽器件,|Vb|的继续增加无法导致耗尽区宽度的继续增加,说明体区负偏压已经无法实现耗尽区宽度的调制,因此器件的转移特性曲线也没有出现类似辐照前的正向漂移.     

A new structure and its analytical breakdown model of high voltage SOI device with Step Unmovable Surface Charges of buried oxide layer

A new SOI （Silicon On Insulator） high voltage device with Step Unmovable Surface Charges （SUSC） of buried oxide layer and its analytical breakdown model are proposed in the paper. The unmovable charges are implemented into the upper surface of buried oxide layer to increase the vertical electric field and uniform the lateral one. The 2-D Poisson＇s equation is solved to demonstrate the modulation effect of the immobile interface charges and analyze the electric field and breakdown voltage with the various geometric parameters and step numbers. A new RESURF （REduce SURface Field） condition of the SOl device considering the interface charges and buried oxide is derived to maximize breakdown voltage. The analytical results are in good agreement with the numerical analysis obtained by the 2-D semiconductor devices simulator MEDICI. As a result, an 1200V breakdown voltage is firstly obtained in 3pro-thick top Si layer, 2pro-thick buried oxide layer and 70pro-length drift region using a linear doping profile of unmovable buried oxide charges.     

CMOS／SIMOX和CMOS／BESOI的γ射线辐照特性比较

用薄膜ＳＩＭＯＸ（ＳｅｐａｒａｔｉｏｎｂｙＩＭｐｌａｎｔａｔｉｏｎｏｆＯＸｙｇｅｎ）、厚膜ＢＥＳＯＩ（ｆｆｅｎｄｉｎｇａｎｄＥｔｃｈ－ｂａｃｋＳｉｌｉｃｏｎＯｎＩｎｓｕｌａｔｏｒ）和体硅材料制备了ＣＭＯＳ倒相器电路，并用６０Ｃｏγ射线进行了总剂量辐照试验。在不同偏置条件下，经不同剂量辐照后，分别测量了ＰＭＯＳ、ＮＭＯＳ的亚阈特性曲线，分析了引起ＭＯＳＦＥＴ阈值电压漂移的两种因素（辐照诱生氧化层电荷和新生界面态电荷）。对ＮＭＯＳ／ＳＩＭＯＸ，由于寄生背沟ＭＯＳ结构的影响，经辐照后背沟漏电很快增大，经３００Ｇｙ（Ｓｉ）辐照后器件已失效。而厚膜ＢＥＳＯＩ器件由于顶层硅膜较厚，基本上没有背沟效应，其辐照特性优于体硅器件。最后讨论了提高薄膜ＳＩＭＯＸ器件抗辐照性能的几种措施。     

Lifetime control in silicon power P-i-N diode by ion irradiation: Suppression of undesired leakage

The irradiation with high-energy (7.35 MeV) protons through a set of energy degraders was used to suppress leakage of the silicon power diodes subjected to local lifetime control. The aim was to modify the profile of recombination centers and to reduce production of vacancy complexes. The high-energy proton irradiation was compared with standard local lifetime killing by high-energy alphas. Recombination centers arising from irradiation were characterized after irradiation and subsequent annealing at 220 and 350 °C by deep level transient spectroscopy and I-V profiling. Static and dynamic parameters of irradiated diodes were also measured and compared. Results show that the applied irradiation with protons provides 3-10 times lower leakage compared to standard alphas for equivalent reduction of the reverse recovery current maximum. On the other hand, the excessive formation of hydrogen donors at high proton fluences and their diffusion during annealing at 350° decreases diode blocking capability.     

SOI CMOS SRAM单元单粒子翻转效应的模拟

为简单快速模拟静态随机存储器(SRAM)的单粒子效应,在二维器件数值模拟的基础上,以经典的双指数模型为原型,通过数值拟合得到了单粒子效应瞬态电流脉冲的表达式,考虑晶体管偏压对瞬态电流的影响,得到修正的瞬态电流表达式,将其带入电路模拟软件HSPICE中进行SRAM存储单元单粒子翻转效应的电路模拟,通过与实际单粒子实验结果的对比,验证了这种模拟方法的实用性。     

Axial lifetime control in silicon power diodes by irradiation with protons, alphas, low- and high-energy electrons

Local lifetime control by proton and alpha-particle irradiation with energies from 1.8 to 12.1 MeV and doses up to 5×10¹² cm⁻² was faced with two types of electron irradiation giving the different profiles of carrier lifetime: the high-energy (4 MeV) resulting in a homogeneous lifetime distribution and the low-energy (500 and 460 keV) providing so-called sloping lifetime control. Deep and shallow levels introduced by irradiation were characterised by DLTS, HVCTS and C-V profiling and their effect on static and dynamic parameters of irradiated diodes was measured and simulated. The advantages and drawbacks of different lifetime control techniques are compared and their application aspects are discussed, as well.     

新型的使用访问控制模型

对传统访问控制模型进行了扩展，提出了一种新的访问控制模型——使用控制模型UCON(usage control)。这种模型包含授权、证书和条件组成成分，能解决作为访问结果而对主体客体属性进行更新的问题，以及在访问过程中对访问权限延续和即时回收的问题。同时对该模型的核心模型进行了形式化描述和定义。     

基于操作轨迹模型的非线性预测控制算法

针对工业过程的非线性,本文首先提出了一种操作轨迹非线性模型的辨识方法:根据调度变量的操作轨迹,选取若干个典型工作点;在各个典型工作点,辨识各自的线性模型;根据测试数据以及过渡数据,得到全局插值非线性模型.在此基础上,本文进一步提出一种基于操作轨迹模型的非线性预测控制算法,并采用多步线性化方法进行问题求解.由于仅需要在典型工作点上进行测试,降低了全局建模的辨识成本,而且控制品质好,仿真结果表明了该算法的有效性.     

Influence of temperature and dose rate on the degradation of BiCMOS operational amplifiers during total ionizing dose testing

The degradation of BiCMOS operational amplifiers TLV2451CP under the gamma-irradiation is studied for different dose rates and temperatures during irradiation. It is shown that studied devices are sensitive to both enhanced low dose rate sensitivity and time-dependent effects. Evidently the main reason for degradation of studied devices is build-up of the interface traps. Obtained results show possibility to develop an approach for total ionizing dose testing of BiCMOS devices considering low dose rate effects.     

A novel total variation based frame layer rate control algorithm for H.264/AVC

Rate control (RC) plays a crucial role in controlling compression bitrates and encoding qualities for networked video applications. In this research, we propose a new total variation (TV) based frame layer rate control algorithm for H.264/AVC. One of its novelties is that a total variation measure, used in image processing field, is proposed to describe encoding distortion in video compression. For intraframes, we present a TV distortion–quantization (D_TV–Q_step) model to obtain accurate QP step size (Q_step). Using TV measure to represent frame complexity, we also present an analytic model to calculate Q_step for the initial frame, and develop an effective scene change detection method. In addition, an incomplete derivative proportional integral derivative (IDPID) buffer controller is proposed to reduce the deviation between the current buffer fullness and the target buffer fullness, and minimizes the buffer overflow or underflow. Extensive experimental results show that, compared with JVT-W042, the proposed algorithm successfully achieves more accurate target bit rates, reduces frame skipping, decreases quality fluctuation and improves the overall coding quality.     

信息安全模型安全控制研究

随着计算机网络的发展,信息越来越开放,但信息安全也暴露出越来越多的问题。文中分析了几种典型的信息安全模型。针对模型中存在的安全隐患,对访问控制机制做了改进,引入主体身份鉴别、访问记录、子存储空间等措施来控制访问,有效地提高了信息的安全性。