Mylapore Ganesh, Ramnath (2018-08). On the Comprehensive Optimization of the Process and Quality Control Framework in Portland Cement Production Testing Using QXRD-Based Accuracy Calibration. Master's Thesis. Thesis uri icon

abstract

  • The Bogue method is the current industrial standard for the compositional analysis of anhydrous and blended cements. However, this method is disadvantageous since the formulations were developed with archaic cement chemistry knowledge. This results in phase quantity estimation errors which can be as high as 9wt.% of the total cement. Quantification of cement phases through X-ray diffraction and Rietveld refinement is an effective alternative which determines the phase quantities directly rather than through estimation. Despite having several advantages over the Bogue method, the adoption rate of Quantitative X-ray Diffraction (QXRD) in industrial quality control is quite slow. Since industrial quality control demands a reproducible technique with high precision, QXRD falls short against the Bogue method with average performance in repeatability and reproducibility. Although cement chemists recognize round robin and proficiency testing as effective solutions for this shortcoming, none of the proposed solutions remain feasible in an online testing environment. In this study, primary focus is allotted to promoting a viable solution to address the repeatability and reproducibility issue in QXRD. Improved QXRD protocols were developed, tested, and, optimized to be applicable in a manual as well as an automated testing environment. Techniques such as multispectral image analysis, point counting, heat of hydration measurements, and, phase decomposition tracing were harnessed to validate the resulting QXRD results with improved repeatability and reproducibility. Furthermore, a mathematical relationship between the Bogue method and the improved QXRD quantification was generated to qualify the improvements from a familiar perspective.
  • The Bogue method is the current industrial standard for the compositional analysis of anhydrous and blended cements. However, this method is disadvantageous since the formulations were developed with archaic cement chemistry knowledge. This results in phase quantity estimation errors which can be as high as 9wt.% of the total cement. Quantification of cement phases through X-ray diffraction and Rietveld refinement is an effective alternative which determines the phase quantities directly rather than through estimation. Despite having several advantages over the Bogue method, the adoption rate of Quantitative X-ray Diffraction (QXRD) in industrial quality control is quite slow. Since industrial quality control demands a reproducible technique with high precision, QXRD falls short against the Bogue method with average performance in repeatability and reproducibility. Although cement chemists recognize round robin and proficiency testing as effective solutions for this shortcoming, none of the proposed solutions remain feasible in an online testing environment.
    In this study, primary focus is allotted to promoting a viable solution to address the repeatability and reproducibility issue in QXRD. Improved QXRD protocols were developed, tested, and, optimized to be applicable in a manual as well as an automated testing environment. Techniques such as multispectral image analysis, point counting, heat of hydration measurements, and, phase decomposition tracing were harnessed to validate the resulting QXRD results with improved repeatability and reproducibility. Furthermore, a mathematical relationship between the Bogue method and the improved QXRD quantification was generated to qualify the improvements from a familiar perspective.

publication date

  • August 2018