Quantitative phase analysis by powder diffractometry requires accurate measurement of the integrated intensities of the diffracted, lines. When lines are isolated and on simple backgrounds, count integration techniques work very well. However, when one or more lines overlap the line of interest, or a complex background is present, profile fitting techniques are required in order to eliminate interferences.Profile fitting involves choosing a mathematical model to represent the expected profile shapes. Experience has shown that the profile shapes obtained with a parafocusing powder diffractometer are not easily described and many models have been tried with varying degrees of success. Generally the more free parameters allowed In the model the ‘setter the fits, although, aesthetically one would like to keep the number of free parameters to a minimum.

The intensity round-robin is one of a series of round-robins sponsored by the International Centre for Diffraction Data. The purpose of these round-robins is to attempt to quantify problems encountered in the acquisition, analysis and interpretation of powder diffraction data. Previous International Centre sponsored round-robins have addressed topics including: manual and automated search match methods; sample preparation methods; d-spacing accuracy; cell parameter refinement and peak profile calibration.The primary focus of the intensity round-robin was to study measured intensities obtained from modern computer controlled powder diffractometers. However, the tests were designed in such a way as to also yield information on the performance of data treatment software packages. To this end, participants were asked to submit both raw and treated data, thus allowing evaluation of, for example, the efficiency of peak hunting algorithms in finding peaks.

The measurement of x-ray diffraction line intensities is the basis for quantitative phase analysis (see for example, Chung (1974), Davis (1986), and Hubbard and Snyder (1988)). While there are many sources of error in such measurements, in recent years computer automation of powder diffractometers and associated analytical software has made such measurements more practical and accurate. For example, profile fitting software has made it possible to determine integrated peak areas and to deconvolute overlapping lines. Another problem which affects quantitative analysis is the systematic error in instrument sensitivity as a function of 20 diffraction angle. This effect has been partially responsible for poor reproducibility of relative intensities between laboratories (Schreiner and Kimmel (1987), and Jenkins and Schreiner (1989)). But, because the error is systematic, corrections may be made by using a standard such as the National Institute of Standards and Technology SRM 1976 alumina plate (NIST 1991). These and other advances have led to a renewed interest in the determination of I/Ic (also called RIR - Reference Intensity Ratio) values for crystalline substances (e.g., Snyder (1992)). I/Ic is defined as the ratio of the intensity of the strongest line of an analyte to the corundum (113) line when the analyte is mixed 50:50 by weight with corundum. We present here a standard procedure used in our laboratory to experimentally measure I/Ic values, and which explicitly incorporates profile fitting and instrument sensitivity corrections. The procedure is written in the format of an ASTM (American Society for Testing and Materials) standard test method, however, inter-laboratory round robin tests have not been carried out to determine precision and bias associated with the method. While the method calls for corundum as the internal standard, another standard material, s, may be used, in which case the procedure will result in a ratio I/Is. Hubbard and Snyder (1988) have shown how to convert between I/Is and I/Ic. This method is based on the procedure routinely published in NBS Monograph 25 until 1986. It is augmented with corrections for the angularly dependent instrument sensitivity and with calculations of I/Ic for both variable and fixed divergence slit configurations. A Quattro Pro spreadsheet is used in our laboratory to do the calculations. An example of the spreadsheet is given in the appendix for one of two I/Ic runs of MgCO3. We also utilize the corundum in the I/Ic runs as an internal standard to determine displacement error corrections for preparation of digitized patterns of pure analyte phases. These patterns are submitted to the International Centre for Diffraction Data for inclusion in a whole pattern data file planned for some time in the future. The notation used here is the standard notation developed for the RIR method by Hubbard and Snyder (1988) and systematically extended by Snyder (1992). A table of the notation is given in the Terminology section below.