Stutzman Paul E, Feng Pan, Bullard Jeffrey W
National Institute of Standards and Technology, Gaithersburg, MD 20899.
Southeast University, Nanjing, Jiangsu 210096, China.
J Res Natl Inst Stand Technol. 2016 Apr 11;121:47-107. doi: 10.6028/jres.121.004. eCollection 2016.
X-ray powder diffraction (XRD) has been used for several decades to identify and measure the mass fractions of various crystalline phases in portland cement. More recently, a combination of scanning electron microscopy with X-ray microanalysis (SEM/XMA) and image processing has been shown to enable the quantitative characterization of microstructural features in these materials. Each technique can furnish some information that is not accessible from the other. For example, SEM/XMA can identify the microstructural location and morphology of calcium sulfate minerals, while only XRD can determine the relative abundance of the different forms of calcium sulfate, such as gypsum (CaSO · 2HO), bassanite , and anhydrite (CaSO). This document describes how XRD and SEM/XMA can be used together to establish and validate the portland cement phase composition and microstructure. Particular emphasis is laid on step-by-step procedures and best practices for XRD specimen preparation, data collection, and intepretation. Similar detail has been given recently for SEM/XMA [Stutzman et al., NIST Tech Note 1877, U.S. Department of Commerce, April 2015]. The methods are demonstrated for three portland cement powders, through which apparent discrepancies between the results of the two methods are identified and procedures are described for resolving the discrepancies and quantifying uncertainty.
几十年来，X射线粉末衍射（XRD）一直被用于鉴定和测量波特兰水泥中各种晶相的质量分数。最近，扫描电子显微镜与X射线微分析（SEM/XMA）以及图像处理相结合，已被证明能够对这些材料的微观结构特征进行定量表征。每种技术都能提供一些其他技术无法获取的信息。例如，SEM/XMA可以识别硫酸钙矿物的微观结构位置和形态，而只有XRD能够确定不同形式硫酸钙的相对含量，如石膏（CaSO₄·2H₂O）、半水石膏和硬石膏（CaSO₄）。本文档描述了如何将XRD和SEM/XMA结合使用，以确定和验证波特兰水泥的相组成和微观结构。特别强调了XRD样品制备、数据收集和解释的逐步程序及最佳实践。最近，对于SEM/XMA也给出了类似的详细内容[Stutzman等人，美国国家标准与技术研究院技术报告1877，美国商务部，2015年4月]。通过三种波特兰水泥粉末对这些方法进行了演示，借此识别了两种方法结果之间明显的差异，并描述了解决差异和量化不确定性的程序。
