Newbury D E, Swyt C R, Myklebust R L
National Institute of Standards and Technology, Gaithersburg, Maryland 20899, USA.
Anal Chem. 1995 Jun 1;67(11):1866-71. doi: 10.1021/ac00107a017.
"Standardless" procedures for quantitative electron probe X-ray microanalysis attempt to eliminate the need for standardization through calculation of standard (pure element) intensities. Either "first principles" calculations, which account for all aspects of X-ray generation, propagation, and detection, or "fitted standards" calculations, which use mathematical fits to measured intensities from a limited set of pure standards, can form the basis for standardless analysis. The first principles standardless analysis procedure embedded in the National Institutes of Health/National Institute of Standards and Technology comprehensive X-ray calculation engine and database, Desktop Spectrum Analyzer, has been tested against spectra measured on NIST standard reference materials, research materials, and binary compounds. The resulting distribution of errors is broad, ranging from -90% to +150% relative. First principles standardless analysis can thus lead to unacceptably large errors.
定量电子探针X射线微分析的“无标样”程序试图通过计算标准（纯元素）强度来消除标准化的需求。“第一性原理”计算（考虑X射线产生、传播和检测的所有方面）或“拟合标样”计算（使用对有限纯标样测量强度的数学拟合）都可以作为无标样分析的基础。美国国立卫生研究院/美国国家标准与技术研究院综合X射线计算引擎及数据库Desktop Spectrum Analyzer中嵌入的第一性原理无标样分析程序，已针对在NIST标准参考物质、研究材料和二元化合物上测量的光谱进行了测试。由此产生的误差分布很广，相对误差范围从 -90% 到 +150%。因此，第一性原理无标样分析可能会导致误差大到无法接受。
