Search

Article

x

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

Effects of pre-irradiated thermal treatment on ideal factor of excess base current in bipolar transistors

Dong Lei Yang Jian-Qun Zhen Zhao-Feng Li Xing-Ji

Effects of pre-irradiated thermal treatment on ideal factor of excess base current in bipolar transistors

Dong Lei, Yang Jian-Qun, Zhen Zhao-Feng, Li Xing-Ji
PDF
HTML
Get Citation
  • During the service of the spacecraft, it will be disturbed by the energetic particles and rays, and thus induce total ionizing dose (TID), displacement damage (DD) or single event effect (SEE) to generate inside the electronic system, which can seriously affect the service lifetime of the electronic components. The difference in structure and types of electronic components are less sensitive to the radiation effects, but bipolar transistor is strongly sensitive to ionizing radiation effect. As a basic component of bipolar circuits, the in-depth study of bipolar transistor ionization radiation effect is of significance for engineering.It has been shown that the an amount of hydrogen can inevitably introduced from an external source during the sealing process of the devices. The KOVAR alloy is widely used as a metal cap material of bipolar transistor in the process of encapsulation. The residual gas analysis (RGA) for sealed Kovar lid packages is shown to have 1%–2% of the hydrogen in the cavity. The source of the hydrogen is generally considered to be out-gassing from the gold plating on the KOVAR. So far, the researches have focused on the study of the ionization damage effect of bipolar transistors with different structures under 60Co gamma ray irradiation. There is lack of systemic study on the comparison of transistors packaged with and without cap.In this paper, we study the influence of sealed KOVAR lid packaged on ionizing radiation damage of lateral PNP bipolar transistor (LPNP) by using 60Co gamma ray as an irradiation source. The semiconductor parameter analyzer is used to measure the electrical parameters of LPNP transistor during irradiation. The irradiation defects in LPNP transistors packaged with and without cap are characterized by deep level transient spectroscopy (DLTS). Experimental results show that the LPNP transistors packaged with and without cap have similar electrical characteristics. The base current increases with the total dose increasing, while the collector current remains almost constant. The degradation of LPNP transistor packaged with cap is more serious.According to the excess base current varying with base-emitter voltage for the LPNP transistors packaged with and without cap, the degradation of bipolar transistor packaged with cap is more serious under the same irradiation conditions. According to the analysis of DLTS, comparing with bipolar transistor packaged without cap, the signal peak at about 300 K is shifted to the left for the bipolar transistor packaged with cap. These results indicate that the LPNP transistors packaged with cap can generate more interface states during irradiation, which is attributed to a large amount of hydrogen and water vapor out-gassing from the gold plating on the KOVAR, which is released under the thermal stress. In the sealed environment, hydrogen can only diffuse into the device cavity, and is combined with the metal material in the transistor to form metal hydride. Therefore the degradation of transistor is severe under the same irradiation condition.
      Corresponding author: Li Xing-Ji, lxj0218@hit.edu.cn
    [1]

    Zhang L, Zhang Y M, Zhang Y M, Han C and Ma Y J 2009 Chin. Phys. B 18 3490

    [2]

    Li X J, Geng H B, Lan M J, Yang D Z, He S Y, Liu C M 2010 Chin. Phys. B 19 066103

    [3]

    Li X J, Geng H B, Lan M J, Yang D Z, He S Y, Liu C M 2010 Chin. Phys. B 19 056103

    [4]

    Jin J, Wang X Q, Lin S, Song N F 2012 Chin. Phys. B 21 094220

    [5]

    Liu C M, Li X J, Geng H B, Rui E M, Guo L X, Yang J Q 2012 Chin. Phys. B 21 104211

    [6]

    文林, 李豫东, 郭旗, 任迪远, 汪波, 玛丽娅 2015 物理学报 64 024220

    Wen L, Li Y D, Guo Q, Ren D Y, Wang B, Maria 2015 Acta Phys. Sin. 64 024220

    [7]

    Johnston A H, Swift G M, Rax B G 1994 IEEE Trans. on Nucl. Sci. 41 2427

    [8]

    McClure S, Pease R L, Will W, Perry G 1994 IEEE Trans. Nucl. Sci. 41 2544

    [9]

    Minson E, Sanchez I, Barnaby H J, Pease R L, Platteter D G, Dunham G 2004 IEEE Trans. on Nucl. Sci. 51 3723

    [10]

    李兴冀, 兰慕杰, 刘超铭, 杨剑群, 孙中亮, 肖立伊, 何世禹 2013 物理学报 62 098503

    Li X J, Lan M J, Liu C M, Yang J Q, Sun Z L, Xiao L Y, He S Y 2013 Acta Phys. Sin. 62 098503

    [11]

    Pease R L, Adell P C, Rax B G, Chen X J, Barnaby H J, Holbert K E, Hjalmarson H P 2008 IEEE Trans. on Nucl. Sci. 55 3169

    [12]

    Pease R L, Platteter D G, Dunham G W, Seiler J E, McClure S, Barnaby H J, Chen X J 2007 IEEE Trans. on Nucl. Sci. 54 2168

    [13]

    Pease R L, Schrimpf R D, Fleetwood D M 2009 IEEE Trans. on Nucl. Sci. 56 1894

    [14]

    Pease R L, Dunham G W, Seiler J E, Platteter D G, McClure S S 2007 IEEE Trans. on Nucl. Sci. 54 1049

    [15]

    Chen X J, Barnaby H J, Vermeire B, Holbert K, Wright D, Pease R L, Dunham G, Platteter D G, Seiler J, McClure S, Adell P 2007 IEEE Trans. on Nucl. Sci. 54 1913

    [16]

    Yusoff W Y W, Jalar A, Othman N K, Rahman I A, Shamsudin R, Hamid M A A 2012 ICSE2012-Proc. 4 604

    [17]

    Rodgers M P, Fleetwood D M, Schrimpf R D, Batyrev I G, Wang S, Pantelides S T 2005 IEEE Trans. on Nucl. Sci. 52 2642

    [18]

    Hughart D R, Schrimpf R D, Fleetwood D M, Chen X J, Barnaby H J, Holbert K E, Pease R L, Platteter D G, Tuttle B R, Pantelides S T 2009 IEEE Trans. on Nucl. Sci. 56 3361

    [19]

    Li X J, Yang J Q, Barnaby H J, Galloway K F, Schrimpf R D, Fleetwood D M, Liu C M 2017 IEEE Trans. on Nucl. Sci. 64 1549

    [20]

    Barnaby H J, Vermeire B, Campola M J 2015 IEEE Trans. on Nucl. Sci. 62 1658

    [21]

    Fleetwood D M 2013 IEEE Trans. on Nucl. Sci. 60 1706

    [22]

    Witczak S C, Lacoe R C, Shaneyfelt M R, Mayer D C, Schwank J R, Winokur P S 2000 IEEE Trans. on Nucl. Sci. 47 2281

    [23]

    Li X J, Liu C M, Yang J Q, Zhao Y L, Liu G Q 2013 IEEE Trans. on Nucl. Sci. 60 3924

    [24]

    Shaneyfelt M R, Schwank J R, Fleetwood D M, Winokur P S, Hughes K L, Hash G L, Connors M P 1992 IEEE Trans. on Nucl. Sci. 39 2244

    [25]

    Mrstik B J, Rendell R W 1991 IEEE Trans. on Nucl. Sci. 38 1101

    [26]

    Shockley W, Read W T 1952 Phys. Rev. 87 835

    [27]

    杨剑群, 董磊, 刘超铭, 李兴冀, 徐鹏飞 2018 物理学报 67 168501

    Yang J Q, Dong L, Liu C M, Li X J, Xu P F 2018 Acta Phys. Sin 67 168501

  • 图 1  剂量率100 rad(Si)/s条件下γ辐射吸收剂量对开帽/未开帽处理的LPNP双极型晶体管的 (a) IB 和 (b) ICVEB变化曲线的影响

    Figure 1.  Variations of (a) IB and (b) IC with base-emitter for the LPNP bipolar transistors with/without cap under dose rate of 100 rad(Si)/s with a 60Co gamma irradiation source.

    图 2  60Co γ射线辐照条件下开帽/未开帽处理的LPNP双极晶体管基极电流变化量与发射结电压的关系

    Figure 2.  Variations of ΔIB with VEB for the LPNP transistors with/without cap irradiated by 60Co gamma source.

    图 3  60Co γ射线辐照条件下, 有/无进行开帽处理的LPNP双极晶体管深能级瞬态谱图

    Figure 3.  DLTS spectra of the LPNP transistors packaged with/without cap irradiated by 60Co gamma source.

    图 4  剂量率100 rad/s条件下γ辐射吸收剂量对开帽/未开帽处理的LPNP双极晶体管(a)电流增益变化量的影响和(b)电流增益倒数变化量的影响

    Figure 4.  (a)The relationship between total dose and current gain for LPNP bipolar transistors with/without cap under dose rate of 100 rad (Si)/s with a 60Co gamma irradiation source. (b) The relationship between total dose and the reciprocal of current gain for LPNP bipolar transistors with/without cap under dose rate of 100 rad (Si)/s with a 60Co gamma irradiation source.

  • [1]

    Zhang L, Zhang Y M, Zhang Y M, Han C and Ma Y J 2009 Chin. Phys. B 18 3490

    [2]

    Li X J, Geng H B, Lan M J, Yang D Z, He S Y, Liu C M 2010 Chin. Phys. B 19 066103

    [3]

    Li X J, Geng H B, Lan M J, Yang D Z, He S Y, Liu C M 2010 Chin. Phys. B 19 056103

    [4]

    Jin J, Wang X Q, Lin S, Song N F 2012 Chin. Phys. B 21 094220

    [5]

    Liu C M, Li X J, Geng H B, Rui E M, Guo L X, Yang J Q 2012 Chin. Phys. B 21 104211

    [6]

    文林, 李豫东, 郭旗, 任迪远, 汪波, 玛丽娅 2015 物理学报 64 024220

    Wen L, Li Y D, Guo Q, Ren D Y, Wang B, Maria 2015 Acta Phys. Sin. 64 024220

    [7]

    Johnston A H, Swift G M, Rax B G 1994 IEEE Trans. on Nucl. Sci. 41 2427

    [8]

    McClure S, Pease R L, Will W, Perry G 1994 IEEE Trans. Nucl. Sci. 41 2544

    [9]

    Minson E, Sanchez I, Barnaby H J, Pease R L, Platteter D G, Dunham G 2004 IEEE Trans. on Nucl. Sci. 51 3723

    [10]

    李兴冀, 兰慕杰, 刘超铭, 杨剑群, 孙中亮, 肖立伊, 何世禹 2013 物理学报 62 098503

    Li X J, Lan M J, Liu C M, Yang J Q, Sun Z L, Xiao L Y, He S Y 2013 Acta Phys. Sin. 62 098503

    [11]

    Pease R L, Adell P C, Rax B G, Chen X J, Barnaby H J, Holbert K E, Hjalmarson H P 2008 IEEE Trans. on Nucl. Sci. 55 3169

    [12]

    Pease R L, Platteter D G, Dunham G W, Seiler J E, McClure S, Barnaby H J, Chen X J 2007 IEEE Trans. on Nucl. Sci. 54 2168

    [13]

    Pease R L, Schrimpf R D, Fleetwood D M 2009 IEEE Trans. on Nucl. Sci. 56 1894

    [14]

    Pease R L, Dunham G W, Seiler J E, Platteter D G, McClure S S 2007 IEEE Trans. on Nucl. Sci. 54 1049

    [15]

    Chen X J, Barnaby H J, Vermeire B, Holbert K, Wright D, Pease R L, Dunham G, Platteter D G, Seiler J, McClure S, Adell P 2007 IEEE Trans. on Nucl. Sci. 54 1913

    [16]

    Yusoff W Y W, Jalar A, Othman N K, Rahman I A, Shamsudin R, Hamid M A A 2012 ICSE2012-Proc. 4 604

    [17]

    Rodgers M P, Fleetwood D M, Schrimpf R D, Batyrev I G, Wang S, Pantelides S T 2005 IEEE Trans. on Nucl. Sci. 52 2642

    [18]

    Hughart D R, Schrimpf R D, Fleetwood D M, Chen X J, Barnaby H J, Holbert K E, Pease R L, Platteter D G, Tuttle B R, Pantelides S T 2009 IEEE Trans. on Nucl. Sci. 56 3361

    [19]

    Li X J, Yang J Q, Barnaby H J, Galloway K F, Schrimpf R D, Fleetwood D M, Liu C M 2017 IEEE Trans. on Nucl. Sci. 64 1549

    [20]

    Barnaby H J, Vermeire B, Campola M J 2015 IEEE Trans. on Nucl. Sci. 62 1658

    [21]

    Fleetwood D M 2013 IEEE Trans. on Nucl. Sci. 60 1706

    [22]

    Witczak S C, Lacoe R C, Shaneyfelt M R, Mayer D C, Schwank J R, Winokur P S 2000 IEEE Trans. on Nucl. Sci. 47 2281

    [23]

    Li X J, Liu C M, Yang J Q, Zhao Y L, Liu G Q 2013 IEEE Trans. on Nucl. Sci. 60 3924

    [24]

    Shaneyfelt M R, Schwank J R, Fleetwood D M, Winokur P S, Hughes K L, Hash G L, Connors M P 1992 IEEE Trans. on Nucl. Sci. 39 2244

    [25]

    Mrstik B J, Rendell R W 1991 IEEE Trans. on Nucl. Sci. 38 1101

    [26]

    Shockley W, Read W T 1952 Phys. Rev. 87 835

    [27]

    杨剑群, 董磊, 刘超铭, 李兴冀, 徐鹏飞 2018 物理学报 67 168501

    Yang J Q, Dong L, Liu C M, Li X J, Xu P F 2018 Acta Phys. Sin 67 168501

  • [1] Zhang Meng, Yao Ruo-He, Liu Yu-Rong. A channel thermal noise model of nanoscaled metal-oxide-semiconductor field-effect transistor. Acta Physica Sinica, 2020, 69(5): 057101. doi: 10.7498/aps.69.20191512
    [2] Investigate the effect of source-drain conduction in single-event transient on nanoscale bulk fin field effect transistor. Acta Physica Sinica, 2020, (): . doi: 10.7498/aps.69.20191896
    [3] Measurement of Magnetically Insensitive State Coherent Time in Blue Dipole Trap. Acta Physica Sinica, 2020, (): . doi: 10.7498/aps.69.20192001
    [4] Zhao Jian-Ning, Liu Dong-Huan, Wei Dong, Shang Xin-Chun. Thermal rectification mechanism of one-dimensional composite structure with interface thermal contact resistance. Acta Physica Sinica, 2020, 69(5): 056501. doi: 10.7498/aps.69.20191409
    [5] Bai Jia-Hao, Guo Jian-Gang. Theoretical studies on bidirectional interfacial shear stress transfer of graphene/flexible substrate composite structure. Acta Physica Sinica, 2020, 69(5): 056201. doi: 10.7498/aps.69.20191730
    [6] Algethami Obaidallah A (Yi Bi), Li Ge-Tian, Liu Zhu-Hong, Ma Xing-Qiao. Phase transformation, magnetic properties and exchange bias of Heusler alloy Mn50–xCrxNi42Sn8. Acta Physica Sinica, 2020, 69(5): 058102. doi: 10.7498/aps.69.20191551
    [7] Ren Xian-Li, Zhang Wei-Wei, Wu Xiao-Yong, Wu Lu, Wang Yue-Xia. Prediction of short range order in high-entropy alloys and its effect on the electronic, magnetic and mechanical properties. Acta Physica Sinica, 2020, 69(4): 046102. doi: 10.7498/aps.69.20191671
    [8] Zhu Xiao-Li, Hu Yao-Gai, Zhao Zheng-Yu, Zhang Yuan-Nong. Comparison between ionospheric disturbances caused by barium and cesium. Acta Physica Sinica, 2020, 69(2): 029401. doi: 10.7498/aps.69.20191266
    [9] Zou Ping, Lv Dan, Xu Gui-Ying. Microstructure and thermoelectric property of (Bi1–xTbx)2(Te0.9Se0.1)3 fabricated by high pressure sintering technique. Acta Physica Sinica, 2020, 69(5): 057201. doi: 10.7498/aps.69.20191561
    [10] Identifying two different configurations of the H32+ by the direct above-threshold ionization spectrum in two-color laser field. Acta Physica Sinica, 2020, (): . doi: 10.7498/aps.69.20200013
    [11] Zhang Ya-Nan, Zhan Nan, Deng Ling-Ling, Chen Shu-Fen. Efficiency improvement in solution-processed multilayered phosphorescent white organic light emitting diodes by silica coated silver nanocubes. Acta Physica Sinica, 2020, 69(4): 047801. doi: 10.7498/aps.69.20191526
  • Citation:
Metrics
  • Abstract views:  66
  • PDF Downloads:  3
  • Cited By: 0
Publishing process
  • Received Date:  28 July 2019
  • Accepted Date:  14 October 2019
  • Available Online:  05 December 2019
  • Published Online:  01 January 2020

Effects of pre-irradiated thermal treatment on ideal factor of excess base current in bipolar transistors

    Corresponding author: Li Xing-Ji, lxj0218@hit.edu.cn
  • Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, China

Abstract: During the service of the spacecraft, it will be disturbed by the energetic particles and rays, and thus induce total ionizing dose (TID), displacement damage (DD) or single event effect (SEE) to generate inside the electronic system, which can seriously affect the service lifetime of the electronic components. The difference in structure and types of electronic components are less sensitive to the radiation effects, but bipolar transistor is strongly sensitive to ionizing radiation effect. As a basic component of bipolar circuits, the in-depth study of bipolar transistor ionization radiation effect is of significance for engineering.It has been shown that the an amount of hydrogen can inevitably introduced from an external source during the sealing process of the devices. The KOVAR alloy is widely used as a metal cap material of bipolar transistor in the process of encapsulation. The residual gas analysis (RGA) for sealed Kovar lid packages is shown to have 1%–2% of the hydrogen in the cavity. The source of the hydrogen is generally considered to be out-gassing from the gold plating on the KOVAR. So far, the researches have focused on the study of the ionization damage effect of bipolar transistors with different structures under 60Co gamma ray irradiation. There is lack of systemic study on the comparison of transistors packaged with and without cap.In this paper, we study the influence of sealed KOVAR lid packaged on ionizing radiation damage of lateral PNP bipolar transistor (LPNP) by using 60Co gamma ray as an irradiation source. The semiconductor parameter analyzer is used to measure the electrical parameters of LPNP transistor during irradiation. The irradiation defects in LPNP transistors packaged with and without cap are characterized by deep level transient spectroscopy (DLTS). Experimental results show that the LPNP transistors packaged with and without cap have similar electrical characteristics. The base current increases with the total dose increasing, while the collector current remains almost constant. The degradation of LPNP transistor packaged with cap is more serious.According to the excess base current varying with base-emitter voltage for the LPNP transistors packaged with and without cap, the degradation of bipolar transistor packaged with cap is more serious under the same irradiation conditions. According to the analysis of DLTS, comparing with bipolar transistor packaged without cap, the signal peak at about 300 K is shifted to the left for the bipolar transistor packaged with cap. These results indicate that the LPNP transistors packaged with cap can generate more interface states during irradiation, which is attributed to a large amount of hydrogen and water vapor out-gassing from the gold plating on the KOVAR, which is released under the thermal stress. In the sealed environment, hydrogen can only diffuse into the device cavity, and is combined with the metal material in the transistor to form metal hydride. Therefore the degradation of transistor is severe under the same irradiation condition.

    • 航天器在轨服役期间, 会受到空间带电粒子及射线的干扰, 其电子系统及电子元器件内部会产生电离、位移、单粒子效应等, 严重影响电子元器件的使用寿命[1-6]. 不同结构及类型的电子元器件对辐射效应的敏感性不同, 对于双极晶体管来说, 其对电离辐射效应具有很强的敏感性[7-10]. 双极晶体管是重要的双极电路的基本组成单元, 对其电离辐射效应进行深入研究, 具有重要的工程意义.

      电子元器件的生产封装过程中, 会不可避免的从外部引入杂质元素氢. 氢作为影响器件抗辐照能力的重要潜在因素, 国际上对此进行大量的研究. 目前在封装过程中普遍采用Kovar合金作为晶体管的金属帽材质, 通过残余气体分析 (RGA: residual gas analysis)对Kovar合金进行检测, 发现在器件空腔内含有1~2%的氢气存在, 而氢气的来源被普遍认为是在Kovar合金上的镀金过程中释放出来的[11-13].

      器件在投入使用前会进行预加温处理进行优选, 即经过长时间加温过程将性能退化严重的器件筛选出来. 在预加温过程中升温会改变器件中氢的状态, 存在形式不同的氢均会对器件的电离效应产生关键性的影响作用[14-15]. 随着预辐照升温应力的增强, MOS器件及双极器件的电学性能退化或改善还与器件采用特定的工艺及测试条件等因素有着密切关系, 尚未有明确的规律.

      通过研究发现, 预处理过程中器件对辐射效应的响应与器件的封装状态有关[16-18]. 研究表明: 在辐照过程中, 密闭环境下氢分子扩散在器件腔内并易与晶体管内金属及氧化物反应, 生成金属氢化物或含氢氧空位, 在辐照过程中参与促进界面态的形成, 对晶体管的电学性能造成影响; 温度会影响H+释放的数量, 而在一定温度条件下, SiO2层中氢气含量过高也会导致界面态发生退火. 因此, 辐照过程中界面态数量不光取决于H+释放的多少, 还与温度、氢气浓度以及含氢缺陷的数量等诸多因素有关. 目前各种因素的影响机制尚不清晰.预处理过程中有/无Kovar合金作为器件封装材料对器件内部氢的状态影响的研究结果表明, 双极晶体管的电离辐射特性与预加温处理及Kovar合金之间的相关性很密切, 然而, 目前国际尚未直接证明二者对双极晶体管电离辐射损伤的影响机制.深入开展预加温处理对开帽/未开帽双极晶体管电离辐射损伤机理研究, 可为寻求电子元器件抗辐射加固方法奠定有力基础, 具有重要的工程指导意义.

      采用60Co γ射线做为辐照源, 针对横向PNP双极晶体管作为研究对象, 通过电学性能测试结合深能级缺陷分析, 深入研究预加温处理过程中, Kovar合金对LPNP晶体管电离辐射损伤的影响机制.

    2.   试验器件与试验方法
    • 选用国产LPNP双极晶体管作为研究对象, 进行电离效应损伤机理研究时, 为清楚比较开帽处理对晶体管电离辐射效应的影响, 将部分所选用的LPNP晶体管进行金属帽切割去除处理, 其中金属帽的材质为Kovar合金. 由于芯片出厂前尚未进行预加温处理, 因此将所选用LPNP晶体管均置于200 ℃加热炉中恒温加热44 h, 进行预加温处理. 为了排除试验结果是由于辐照过程中二次粒子因素的干扰, 将预加温处理后未开帽的LPNP晶体管也均采取开帽处理, 并立即将所有样品进行辐照试验.试验采用60Co γ射线辐照源进行辐照试验, 所采用的辐照测试条件为: 辐照总剂量为100 krad(Si); 辐照剂量率为100 rad(Si)/s; 试验过程中, 所有晶体管均在室温环境下进行辐照且晶体管的所有管脚均处于接地状态.

      采用KEITHLEY 4200-SCS半导体参数仪, 对辐照前后LPNP晶体管进行Gummel曲线测试. 测试条件如下: 发射极接扫描电压, 从0.2 V扫描至0.8 V, 即VEB = 0.2—0.8 V, 扫描步长为0.01 V; 基极和集电极均接0 V电压, 即VB = VC = VBC = 0 V. 基于Gummel特性曲线还可以得出LPNP晶体管电流增益及其倒数的变化量随吸收剂量的变化关系, 以及晶体管过剩基极电流ΔIB随发射结电压VEB的变化关系来定性表征电离缺陷的演化规律.

      基于深能级瞬态谱(DLTS)测试系统, 针对两种LPNP晶体管内部的微观缺陷状态进行分析, 深入分析LPNP晶体管电离辐照损伤微观机理. 选取晶体管中掺杂浓度较低的集电区进行测试, 测试参数设定如下: 反向偏压VR = 5 V, 脉冲电压VP = 0.1 V, 测试周期TW = 4.096 s, 脉冲宽度TP = 0.01 s, 测试的扫描温度为150—330 K.

    3.   实验结果与分析
    • 对开帽/未开帽两种类型LPNP双极型晶体管Gummel特性曲线的对比情况如图1所示. 图中给出了在剂量率为100 rad(Si)/s的60Co γ射线辐照条件下, 晶体管的基极电流(IB)和集电极电流(IC)对电离效应影响的实验结果. 图1(a)描述的是LPNP晶体管IB随发射结电压VEB的变化情况. 随着吸收剂量的增加, 两种LPNP晶体管的IB逐渐增大, 此外, 与发射结电压VEB较大时相比, 当发射结电压VEB较小时, IB增加倍数较大. 图1(b)描述的是LPNP晶体管IC随发射结电压VEB的变化情况. 随着吸收剂量的增加, 两种预处理方式的LPNP晶体管IC均无明显变化.

      Figure 1.  Variations of (a) IB and (b) IC with base-emitter for the LPNP bipolar transistors with/without cap under dose rate of 100 rad(Si)/s with a 60Co gamma irradiation source.

      前面的结果已经表明, 双极晶体管在受到电离辐射损伤后, IC基本上保持不变, 而电流增益退化的主要表现为IB增加. 通过探究辐照后双极晶体管的过剩基极电流(ΔIB)与发射结电压(VEB)的变化关系, 可以深入分析不同缺陷对双极晶体管电离辐射的作用机制, 从而得出预处理过程中Kovar合金对LPNP晶体管电离辐射效应的影响. 基极电流是载流子在基区扩散与复合形成的复合电流, 表达式为IB=IB-preIB, 其中IB-pre为双极晶体管的初始基极电流值, ΔIB为基极电流的变化量, 也被称为双极晶体管的过剩基极电流.

      针对双极晶体管的电离辐射损伤效应, 引入理想因子n, 理想情况下, 在辐照过程中, 如果器件复合类型以表面复合为主, 则理想因子n = 2. 表达式为

      其中ϕT为热电压, 指在闭合电路中由于两点间存在温差而出现的电位差. 该物理量即可以表示扩散系数和迁移率的内在联系, 也表示载流子在半导体中定向运动的难易程度.

      图2为相同剂量率Co60 γ辐照条件下, 两种LPNP晶体管的过剩基极电流ΔIB随发射结电压VEB的变化关系. 图中可以清楚看出, 随着吸收剂量的增加, 两种LPNP晶体管的ΔIB均明显增加; 而理想因子nVEB变化呈现出两个斜率特征, 在VEB较小区段时, 理想因子趋于n = 2; 当VEB在较大区段 (VEB > 0.4 V) 时, 1 < n < 2. 理想因子n数值降低是由于基区表面辐射诱导界面态复合率的变化所导致的[19].

      Figure 2.  Variations of ΔIB with VEB for the LPNP transistors with/without cap irradiated by 60Co gamma source.

      通过LPNP晶体管ΔIB随发射结电压VEB的变化分析电离效应对LPNP晶体管基极电流的影响. 表达式为[19]

      式中, q为电子电荷, ϕT为热电压, ni代表Si体中本征载流子浓度, PE为发射极周长, E为空间电荷区复合达最大处的电场, Δs为表面复合速率的变化量, 与界面态密度的变化量密切相关, 有以下关系式:

      其中: νth代表热速率; σ为复合中心处俘获电荷的横截面大小; ΔNit表示界面态密度增量. 根据文献[20], 电场E方程推导为

      其中, eSi为Si的介电常数, ns为基区界面处多数载流子浓度. 基于 (4) 式可知, 在VEB一定时, 随着氧化物电荷减少, 表面载流子浓度ns随之降低, 导致晶体管ΔIB的理想因子n值减小. 这是由于辐照的初始阶段, 界面态陷阱的累积不足以抵消氧化物电荷对表面载流子浓度的影响[21]. 因此在VEB较小时, ΔIB的理想因子n数值近似于2; 在VEB较大时, ΔIB的理想因子n数值处于1与2之间.

      氧化物电荷的累积会导致空穴穿过耗尽层进入到中性基区, 再通过复合诱导理想因子n数值逐步趋近于1. 随着吸收剂量的增加, 界面态陷阱密度增加导致发射结耗尽区发生复合, 这也是产生ΔIB的重要原因. 其中ΔIBVEB曲线斜率发生变化时所处发射结电压被定义为晶体管的转换电压Vtr. 界面态陷阱的不断累积, 不光会增加基极电流, 而且影响理想因子的数值变化, 转换电压逐渐降低. 这是由于随着吸收剂量的增加, 表面载流子浓度ns降低所引起的. 而ns降低的潜在因素是辐照过程中产生的带负电的界面态数量增多引起的静电效应[19], 或是氢在n型基区掺杂钝化作用所致[22].

      图2可以明显看出, 在相同辐照条件下, 预加温过程中未开帽处理的晶体管ΔIB变化更大, 且随着吸收剂量的增加, 转换电压Vtr明显向低发射结电压方向移动. 通过上述分析可知, n值的变化是由于界面态数量增多所导致的结果. 在辐照过程中, 与预加温过程中开帽处理的晶体管相比, 未开帽的LPNP晶体管n值变化程度更大, 说明器件内部产生的界面态更多, 分析认为未开帽晶体管中氢的浓度相对较高, 导致在n型基区掺杂钝化作用所引起的结果.

    • 基于上述LPNP晶体管电性能结果分析可知, 导致晶体管电学性能退化的微观机制主要受界面态的影响. 现将两种LPNP晶体管中界面态缺陷信息进行对比, 通过深能级缺陷分析来验证上述试验结果. 图3为温度范围从150 K扫描到330 K、剂量率为100 rad(Si)/s条件的60Co γ辐照过程中, 两种LPNP晶体管DLTS谱的测试结果. 由图可见, 两种LPNP晶体管均在温度为250—330 K之间出现了一个明显的深能级缺陷信号峰. 由于缺陷信号峰峰尖所在的温度坐标不同, 将未开帽晶体管对应的信号峰命名为H(295), 将开帽晶体管对应的信号峰命名为H(320). 利用阿伦尼乌斯方程计算得出: 采用开帽处理的LPNP晶体管中的类深能级缺陷能级$ E_{\rm T}-E_{\rm V} $= 0.642 eV, 俘获截面σ = 4.16 × 10–17 cm2, 缺陷浓度NT = 4.10 × 1015 cm–3; 未进行开帽处理的LPNP晶体管中的类深能级缺陷能级$ E_{\rm T}-E_{\rm V} $ = 0.579 eV, 俘获截面σ = 1.38 × 10–17 cm2, 缺陷浓度NT = 3.33 × 1015 cm–3. 通过计算结果可以看出, 相同辐照条件下, 未开帽处理的晶体管中引入的缺陷能级位置更接近Si禁带中心, 缺陷浓度较低.

      Figure 3.  DLTS spectra of the LPNP transistors packaged with/without cap irradiated by 60Co gamma source.

      本文中信号峰所处位置与文献[23]中的试验结果相似, 表明60Co γ射线辐照诱导LPNP双极型晶体管产生电离辐射损伤缺陷为界面态陷阱. 通常, 为了解释界面态陷阱的反应机制, 常采用Shaneyfelt等[24]提出的空穴/氢离子输运(HT)2模型. 该模型认为, 空穴向Si/SiO2界面传输时, 会在界面附近形成陷阱电荷. 随着空穴成为陷阱电荷或者被电子中和, 界面附近的氢原子可以在带正电的氧化物陷阱处形成氢离子, Si/SiO2界面附近的氢离子被释放, 传输到Si/SiO2界面的氢离子与界面发生相互作用, 同时生成界面态陷阱[25].

      通过图3可以看出, 在相同辐照条件下, 与开帽处理过的LPNP晶体管相比, 未开帽处理的晶体管在辐照后DLTS特征峰向左移动, 表明辐照在未开帽处理的LPNP晶体管中引入的缺陷能级位置更接近禁带中心. 由肖克莱-里德-霍尔模型[26]可知, 缺陷能级位置越接近禁带中心, 则复合效率越高, 进而对晶体管造成的损伤越严重. 此外, 缺陷能级位置及缺陷浓度均是导致晶体管性能退化的重要因素, 我们之前研究已经证实, 与缺陷浓度相比, 缺陷能级位置占主导地位, 是使晶体管电学性能退化加剧的根本原因[27]. 通过上述分析可以说明预处理过程中未开帽处理LPNP晶体管内部受到电离辐射损伤更为严重.

      基于上述分析, 通过探究LPNP双极晶体管电流增益变化量(Δβ)及其电流增益倒数变化量Δ(1/β)随吸收剂量的关系, 直观反映出辐照前后晶体管电性能变化规律, 为证明开帽处理对电离辐射损伤的影响机制, 揭示器件宏观电学性能退化与微观缺陷演化的对应关系. 电流增益β是表征双极晶体管辐射损伤效应最为重要的电性能参数. 本试验采取共发射极接线方式进行测试, 当发射结正偏电压VEB = 0.65 V时, ICIB的比值定义为电流增益, 即β = IC/IB, 电流增益变化量的表达式为: Δβ = β-β0, 电流增益倒数变化量的表达式为: Δ(1/β)=1/β-1/β0, 式中β0β分别为晶体管辐照前和辐照后的电流增益值. 图4(a)图4(b)分别为相同剂量率60Co γ射线辐照条件下, 开帽/未开帽处理的LPNP晶体管Δβ和Δ(1/β)随吸收剂量的变化曲线. 如图所示, 对于两种类型的LPNP型晶体管而言, 随着吸收剂量的增加, LPNP晶体管的Δβ明显下降, 晶体管Δ(1/β)逐渐升高且退化无饱和趋势. 上述结果表明, 两种类型的LPNP晶体管均发生明显的电离辐射损伤. 其中, 在相同辐照条件下, 在预加温过程中未开帽处理的晶体管Δβ和Δ(1/β)退化程度更为明显.

      Figure 4.  (a)The relationship between total dose and current gain for LPNP bipolar transistors with/without cap under dose rate of 100 rad (Si)/s with a 60Co gamma irradiation source. (b) The relationship between total dose and the reciprocal of current gain for LPNP bipolar transistors with/without cap under dose rate of 100 rad (Si)/s with a 60Co gamma irradiation source.

      上述试验结果表明, 预加温过程中, 与开帽处理的LPNP晶体管相比, 未进行开帽处理的晶体管电性能退化严重, 在器件内部产生的界面态数量更多. 试验分析认为这是由于密封器件中含有大量的氢 (包括氢气、水汽), 晶体管的金属帽材质为Kovar合金, 其中氢气主要来源于Kovar合金内部吸附气氛在热应力作用下释放出来, 由于在密封环境中, 释放的氢气只能扩散在器件腔体内, 结合晶体管中金属材料形成金属氢化物, 在水汽的作用下, 加速器件的电性能退化程度. 与之相比, 非密闭晶体管内部不能保留氢气, 因此采用开帽处理的晶体管电性能更好.

    4.   结 论
    • 采用电性能测试与DLTS谱分析两种手段相结合, 探究60Co γ射线辐照过程中, 开帽处理与未开帽处理LPNP晶体管的电性能及电离辐照诱导微观缺陷的演化规律. 通过试验分析, 得到以下结论:

      1)两种类型LPNP晶体管的电学性能参数变化趋势相似, 辐照后晶体管IB随吸收剂量的增加逐渐增大, IC在辐照前后无明显的变化. Δ(1/β)均随吸收剂量的增加有明显的退化趋势, 其中, 未进行开帽处理过的LPNP晶体管电性能退化程度较大.

      2)通过微观机理分析表明, 随着吸收剂量的增加, 未开帽处理的LPNP晶体管ΔIB增加明显, 理想因子nVEB的增加逐渐降低, 转换电压Vtr明显向低发射结电压方向移动. 这是基区表面辐射诱导界面态复合率的变化所导致的, n值的变化是界面态数量增多所导致的结果. 通过DLTS谱分析可知, 与开帽处理过的LPNP晶体管相比, 电离辐射在未开帽晶体管内部产生的界面态能级位置更接近于禁带中心. 分析认为, 这主要由于密封器件中存在大量的氢气和水汽, 氢气的存在会促进界面态的形成, 而水汽会加剧器件电性能的退化程度. 与之相比, 开帽处理过的晶体管内部不能保留氢气, 因此晶体管电性能更好.

Reference (27)

Catalog

    /

    返回文章
    返回