Analysis

Data were collected using Cu K radiation from 2Q = 18° to 130° , using a step size of 0.02° , a count time of 4 s, and a graphite monochromator. The receiving slit was fixed at 0.2 mm and the variable divergence slit was locked at approximately 0.9° to satisfy the requirement for a constant irradiated volume.

The first step was to establish suitable experiment files to describe the individual phase crystal structures and peak shapes. This was accomplished by refining structure models from the literature (Table 1 ), compiling them into a structure database, and chemical extractions to concentrate different phase groups. The potassium hydroxide / sucrose extraction (KOSH) concentrates the silicate fraction, the salicyclic acid / methanol extraction (SAX) concentrates the interstitial phases, and a 7 % nitric acid in methanol extraction provided a ferrite and periclase residue [12]. This simplified subsequent analyses by allowing refinement of each structure model and the peak shape parameters with less interference from the other constituents. These intermediate models and peak shapes were used for the analyses of the bulk clinker patterns. In essence, the bulk clinker analyses are pattern-fitting exercises using reference standards determined from the extraction residue experiments.

Variables refined for each phase include scale, specimen displacement, background, lattice parameters, atomic coordinates (subject to Si-O or Ca-O bond length constraints), aluminum and iron fractions in ferrite tetrahedral and octahedral sites, peak shapes, and for alite, preferred orientation. Fixing these variables, especially the profile shape parameters, in the subsequent analyses eliminated problematic correlations.

Table 1. Linear Absorption Coefficients (µ): Cu K