Page:NIOSH Manual of Analytical Methods - Chapter R.pdf/11

 XRD Methods. XRD methods include NIOSH Manual of Analytical Methods (NMAM) 7500, OSHA ID-142, MSHA P-2 and UK-HSE Methods for the Determination of Hazardous Substances (MDHS) 51 /2 [57-60]. Details of these methods are listed in Table 3. X-ray diffraction is capable of distinguishing between the different polymorphs of crystalline silica: quartz, cristobalite and tridymite.

The primary diffraction line for α-quartz is 26.66° 2θ (3.343 Å). Since silica often occurs in a matrix with other minerals, some of which exhibit spectra that overlap with the primary quartz diffraction peak, the use of alternate peaks for quantification may be necessary. A list of potential mineral interferences is given in OSHA ID-142, Appendix A [58]. The use of secondary, tertiary and even quaternary peaks may result in decreased sensitivity [61]. An X-ray diffraction scan to characterize the environmental matrix for a set of samples is required in NIOSH X-ray diffraction methods for each set of samples. A determination revealing the correct peak ratios for the three largest peaks can be taken as an indicator of lack of interferences; otherwise a phosphoric acid digestion may be required prior to analysis to reduce interferences [39]. A high level of analyst expertise is required to optimize instrument parameters and correct for matrix interferences either during the sample preparation phase or the data analysis and interpretation phase [62]. NIOSH XRD methods suggest that XRD analysts have some training (university or short course) in crystallography or mineralogy in order to have a background in crystal structure, diffraction patterns and mineral transformation. This is important for understanding the matrix in which the sample was taken.

For optimum XRD instrument performance, the X-ray source must be aligned and monitored routinely for stability. Counting statistics, which are crucial for precise measurement of peak intensity, can be improved by increasing the count time per increment, although this increases the total time per sample analysis. The scan rate of the detector should be no more than 1°2θ min$-1$. Error increases rapidly with increasing scan rate. A 1° divergence slit width is reasonable for measurement in the 20-80 °2θ range. For optimal intensity and resolution, the receiving slit width should be >0.2° and about the same as the X-ray beam width. Inherent properties of the sample, such as crystallinity, can also affect measured peak intensities. Parameters related to sample preparation include sample homogeneity, size distribution of the particles, size of the sample surface exposed to the X-ray beam, thickness of the sample deposition and choice of internal standards, if any. An internal standard, such as silver in NMAM 7500 and OSHA ID-142, serves to normalize the X-ray intensity measured and to provide information on matrix absorption effects. Sample spinning during analysis allows for the alignment of each crystal in the beam and thus reduces precision error by insuring that all crystals are measured. Monitoring X-ray tube emission is done via a permanently mounted standard (e.g.; NIST SRM 1976, a sintered alumina plate used as an instrument sensitivity standard for X-ray powder diffraction). In addition, line position and line shape can be monitored via silicon powder (NIST SRM 640c) or lanthanum hexaboride powder (NIST SRM 660a). Direct-on-filter methods depend on an external standard, such as an aluminum plate to correct for the gradual decline in X-ray tube emission [60]. 3/15/03