摘要:本文使用5—27 keV能量范围的电子轰击纯厚Al (Z = 13), Ti (Z = 22), Zr (Z = 40), W (Z = 74)和Au (Z =79)靶, 利用硅漂移探测器(SDD)收集产生的X射线, 给出了K, L壳层特征X射线产额的测量结果, 并将所得实验数据与基于扭曲波玻恩近似理论模型(DWBA)的蒙特卡罗模拟值进行了比较, 两者在小于或约为10%的范围内符合. 根据测得的特征X射线产额进一步得到了相应的内壳层电离截面或特征X射线产生截面. 通过对比电子入射角度为45°和90°的两种情况下解析模型与蒙特卡罗模拟的特征X射线产额, 发现在入射角度为90°时两者符合较好. 同时, 本文还给出了次级电子、轫致辐射光子对特征X射线产额的贡献, 该贡献与入射电子能量关系较弱, 表现出与原子序数较密切的相关性. 关键词:特征X射线产额/ 内壳层电离截面/ 特征X射线产生截面/ 蒙特卡罗模拟
English Abstract
Characteristic X-ray yields and cross sections of thick targets of Al, Ti, Zr, W and Au induced by keV-electron impact
Key Laboratory of Radiation Physics and Technology of Ministry of Education, Institute of Nuclear Science and Technology, Sichuan University, Chengdu 610064, China
Abstract:In this paper, pure thick Al (Z = 13), Ti (Z = 22), Zr (Z = 40), W (Z = 74) and Au (Z = 79) targets are bombarded by electrons in an energy range of 5–27 keV, and the experimental thick-target characteristic X-ray yields of K-shell and L-shell, the X-ray production cross sections and the ionization cross sections of inner shells are presented. The present experimental setup and data processing are improved, specifically, a deflection magnet is installed in front of the X-ray detector to prevent the backscattered electron from entering into the X-ray detector, and the bremsstrahlung background spectra calculated from PENELOPE Monte Carlo simulations are used to deduce the net peak areas. The X-ray detector used in this experiment is the XR-100SDD manufactured by Amptek Inc. with a 25 mm2 C2 ultra-thin window which can detect the low-energy x-rays down to boron Kα line (0.183 keV). Standard sources (55Fe, 57Co, 137Cs and 241Am) with an activity accuracy range of 1%–3% (k = 2), supplied by the Physikalisch-Technische Bundesanstalt, Germany (PTB), are used to perform the detector’s efficiency calibration, and in a low-energy range (< 3.3 keV) the efficiency calibration is accomplished by measuring characteristic X-ray spectra produced by 20 keV electron impacting various thickness solid targets (i.e. by the characteristic peak method). The uncertainty of the detector’s efficiency calibration obtained in this paper is ~1.6%. The experimental thick-target characteristic X-ray yield data with an uncertainty of 1.7%–6.2% are compared with the PENELOPE Monte Carlo simulations, in which the inner-shell ionization cross sections are based on the distorted-wave Born approximation (DWBA) calculations, and they are in good agreement with a difference of less than or ~10%. According to the measured thick-target characteristic x-ray yields, the K-shell ionization cross sections for Al, Ti and Zr and the L-shell X-ray production cross sections for Zr, W and Au are also obtained with an uncertainty of 5%–8% (except for Al due to large K-shell fluorescence yield uncertainty), the difference between the experimental and theoretical data is also less than or ~10%. Moreover, by comparing the thick-target characteristic X-ray yields obtained from the analytical model and the PENELOPE Monte Carlo simulations at the electrons’ incident angles of 45° and 90°, it is found that the degree of agreement between the results from the analytical model and the Monte Carlo simulations at the incident angle of 90° is better than at the incident angle of 45°. Moreover, the contributions of the secondary electrons and bremsstrahlung photons to the characteristic X-ray yield are also given based on the PENELOPE Monte Carlo simulations. As for the elements studied in this paper, for the low ionization threshold energy, the contribution of the secondary electrons is ~2%, and however, for the high ionization threshold energy, the contribution is ~10%–20%. These contributions depend weakly on the energy of the incident electrons and show that these contributions are closely correlated with atomic number. Keywords:characteristic X-ray yields/ atomic inner-shell ionization/ characteristic X-ray production cross section/ Monte Carlo simulation
其中NX(E0)表示入射电子能量为E0时相应壳层的特征X射线净计数, Ne为总入射电子数, ε(E)表示光子能量为E时的探测器本征效率, ΔΩ为X射线探测器的立体角, 因子ε(E)ΔΩ/4π的值由图2中效率刻度曲线(实线)内插得到. 实验谱的X射线特征峰净计数NX(E0)由实验谱特征峰区计数减去蒙特卡罗程序PENELOPE[2]计算出的轫致辐射谱本底得到, 以W元素为例, 结果如图3所示. 图 2 SDD探测器效率刻度曲线. 空心圆表示Al, Ti, Zr, W和Au的特征X射线位置, 实线表示根据探测器参数计算的效率值 Figure2. The X-ray detection efficiency of the SSD detector. The positions of the characteristic X-ray lines for Al, Ti, Zr, W and Au are indicated by the open circles. The solid line represents the theoretical values calculated based on the detector’s parameters.
图 3 27 keV电子入射厚W靶碰撞产生的实验谱(实线)与PENELOP模拟的轫致辐射本底谱(虚线) Figure3. The experimental spectrum (solid line) and the bremsstrahlung background spectrum simulated by PENELOPE (dashed line) produced by 27 keV electron impact on thick W target.
23.2.实验产额与PENELOPE模拟产额比较 -->
3.2.实验产额与PENELOPE模拟产额比较
PENELOPE(2008版)是一款目前广泛使用的模拟电子-光子输运的蒙特卡罗软件包, 它能在50 eV—1 GeV的能量范围内使用可靠的数值和解析物理模型相结合的方法模拟电子和光子的输运过程, 描述相应能区内电子和光子与物质的相互作用[2]. PENELOPE程序对电子在厚靶物质中的输运过程考虑了电子的非弹性散射、弹性散射、内壳层电离及轫致辐射等物理过程. 在本文的能区范围, PENELOPE(2008版)程序采用了更为准确的基于扭曲波玻恩近似理论模型(DWBA)的内壳层电离截面[11,13]来模拟特征X射线厚靶产额. 在本文的模拟计算中, 建立了与实验相同的几何模型, 如图4所示, 即电子垂直入射, 待测靶与水平成45°, 探测器水平放置. 在实验中, 探测器对靶的张角由探测器探头半径与靶中心距离决定, 为提高蒙特卡罗模拟的计算效率, 张角的半角取为30°, 计算表明这样设置不影响模拟计算结果[33]. 图 4 PENELLOPE模拟几何模型示意图 Figure4. The geometry used in the PENELOPE simulations.
图 5 实心圆点表示实验测得不同入射电子能量下的厚靶特征X射线产额, 实线代表相应的蒙特卡罗模拟值, 虚线为蒙特卡罗模拟值归一到实验数据值上的结果, 括号内为归一化参数, 缩略图为实验布局示意图 Figure5. Solid circles denote the experimental characteristic X-ray yields of thick targets at different incident electron energies. Solid curves represent the results of Monte Carlo simulations using the PENELOPE code. Dashed curves are the scaled results of Monte Carlo simulations by scaling factors that are given in parentheses. The insets show the schematic of experimental geometry.
图 6α = 45°, β = 45°时由PENELOPE计算的特征X射线产额(实心圆点), 其中来自初级电子电离贡献表示为空心圆形, 次级电子电离贡献表示为空心三角形, 轫致辐射光子贡献为空心方形. 实线为方程(2)计算所得结果. 缩略图为计算几何示意图 Figure6. In the case of α = 45°, β = 45°, the solid dots represent the total characteristic X-ray yields calculated by PENELOPE, which include the contributions from the primary electron ionization (hollow circles), secondary electron ionization(hollow triangles) and bremsstrahlung photon ionization (hollow squares). The solid lines are the characteristic X-ray yields calculated by Eq. (2). The insets show the schematic of calculation geometry.
图 7α = 0°, β = 45°时由PENELOPE计算的特征X射线产额(实心圆点), 其中来自初级电子电离贡献表示为空心圆形, 次级电子电离贡献表示为空心三角形, 轫致辐射光子贡献为空心方形. 实线为方程(2)计算所得结果. 缩略图为计算几何示意图 Figure7. In the case of α = 45°, β = 45°, the solid dots represent the total characteristic X-ray yields calculated by PENELOPE, which include the contributions from the primary electron ionization (hollow circles), secondary electron ionization(hollow triangles) and bremsstrahlung photon ionization (hollow squares). The solid lines are the characteristic X-ray yields calculated by Eq. (2). The insets show the schematic of calculation geometry.
元素的特征X射线产额由初级电子、次级电子和光子的电离作用共同组成, 利用PENELOPE程序, 可以估算出入射初级电子、次级电子和光子对相应壳层特征X射线的贡献. Bote等[25]指出, 内壳层电离的几率与弹性散射和外壳层电离相比很小, 故电子轨迹与蒙特卡罗模拟中使用的电离截面模型相关性很弱, 使用的PENELOPE可以很好地模拟如背散射系数、薄膜透射系数等诸多物理量, 因此, 特征X射线产额的模拟值大小主要由蒙特卡罗模拟中使用的电离截面或产生截面确定. 同时, 基于DWBA理论模型计算的理论特征X射线电离或产生截面值与实验值在形状上符合很好[33], 因此可采用文献[33]的方法由厚靶的特征X射线产额来得出X射线产生截面或电离截面, 即通过实验与蒙特卡罗模拟的直接电离产额的比值确定为修正系数, 将蒙特卡罗模拟中使用的基于DWBA理论的截面数据乘以相应的修正系数得到实验截面值. 本文中的DWBA理论数据来自PENELOPE数据库, 根据文献[11,13]计算得到. 在Al, Ti的K层及Zr的L层中, 光子与次级电子电离仅占1%—2%, 对总电离贡献可以忽略, 故采用图5中所标注的归一化参数进行计算. 对于Zr的K层及W, Au的L层, 光子与次级电子电离贡献大于7%, 我们将实验初级电子电离产额与蒙特卡罗模拟初级电子电离产额相比得到新的修正系数, 其中, 实验初级电子电离产额由实验总产额减去蒙特卡罗模拟的次级电子与光子电离产额得到. 最终获得的内壳层电离截面及特征X射线产生截面与相应的修正系数如图8和图9所示. 图 8 (a) Al, (b) Ti, (c) Zr元素实验K壳电离截面(实线)与DWBA理论值(虚线), 括号内为修正系数 Figure8. Experimental K-shell ionization cross sections of (a) Al, (b) Ti, (c) Zr (solid lines) and the DWBA theoretical values (dashed lines). The scaling factors are given in parentheses.
图 9 (a) Zr, (b)?(d) W, (e)?(g) Au元素实验L壳特征X射线产生截面(实线)与DWBA理论值(虚线), 括号内为修正系数 Figure9. Experimental L-shell characteristic X-ray production cross sections of (a) Zr, (b)?(d) W, (e)?(g) Au (solid lines) and the DWBA theoretical values (dashed lines). The scaling factors are given in parentheses.