Comparison of Crystalline Silica (α-Quartz) Calibration Standards NIST-SRM 1878 and NIST-SRM 1878a by Fourier Transform Infrared Spectrophotometry

Received in final form 23 November 2004 Comparing α-quartz calibration standards is important for understanding the differences among various quartz standards. This information permits valid comparisons of airborne silica concentration measurements quantified with different calibration standards. The recent paper by Chisholm (2005) on comparison of quartz standards HSE A9950 and NIST-SRM 1878 for X-ray diffraction (XRD) analysis is an important contribution to the literature. In his paper, Chisholm has reviewed earlier comparisons (Jeyaratnam and Nagar, 1993; Verma and Shaw, 2001; Kauffer et al., 2002) and provides insights into the differences observed. In our study (Verma and Shaw, 2001), we compared seven international α-quartz calibration standards by Fourier transform infrared spectrophotometry (FTIR) using National Institute of Standards and Technology-Standard Reference Material (NIST-SRM) 1878 (NIST, 1983) as the reference standard with an assumed purity of 100% to the measure relative purity of the other six standards. Since then a new crystalline silica calibration standard has been formulated in the USA, NIST-SRM 1878a (NIST, 1999). We considered it useful to validate the purity of the two NIST-SRM standards independently. This comparison thereby updates our original study as we now have determined the relative purity of the new standard 1878a with respect to the previous standard 1878 by FTIR. The determination was made by NIOSH method 7602 (NIOSH, 1994) using a Nicolet model Nexus 410 FTIR (Nicolet Instruments Corporation, Madison, WI, USA). For comparability with earlier study results, NIST-SRM 1878 was used as the gold standard with an assumed 100% purity. The method has been previously detailed (Verma and Shaw, 2001). NIST-SRM 1878a has been prepared from electronic grade, single crystal nodules of Brazilian quartz obtained from the National Defense Stockpile. It was crushed and jet milled to a median particle size of 1.6 μm. The resulting powder was then washed in hydrofluoric acid and hydrochloric acid, rinsed and ignited at 500°C to produce the standard (NIST, 1999). The material was analyzed by XRD and the purity was certified to be 100.00 ± 0.21% crystalline α-quartz, expressed as a mass fraction in percent (%) (NIST, 1999). The size distribution by laser scattering techniques for cumulative mass fraction% was determined to be 3.78 μm at <90%, 0.95 μm at <10% and 1.59 μm for a median of 50%. NIST-SRM 1878 had a certified purity of 95.5 ± 1.1% by XRD. The size distribution information provided with NIST-SRM 1878, as determined by the sedimentation method, was reported to be between 0.33 and 5.0 μm with a mass mean spherical diameter of 1.62 μm (NIST, 1983; Verma and Shaw, 2001). The mass spherical diameters of the two standards are very similar at 1.59 and 1.62 μm. Our results by FTIR, as shown in Fig. 1, indicate that NIST-SRM 1878a has a relative purity of ∼103% with respect to NIST-SRM 1878 by our method. If NIST-SRM 1878a was taken as 100% pure, it would correspond to a relative purity of 96.7% for NIST-SRM 1878 by our method (obtained by regressing NIST-SRM 1878a taken as gold standard on X axis i.e. reverse of what is shown in Fig.1), a result very close to the certified value of 95.5 ± 1.1%. The two NIST standards have similar size distributions and differ by <5% in relative purity by FTIR. As such, they could be considered nearly equivalent. The relative purity of the other standards evaluated in the earlier study (Verma and Shaw, 2001) can be equated to NIST-SRM 1878a using the relationship shown in Fig. 1. For example, Sikron F-600 (also labeled as HSE A9950) and A9950 (AUST 1), derived from Sikron F-600, used in our study (Verma and Shaw, 2001) would be of ∼76 and 82% purity relative to NIST-SRM 1878a. Kauffer et al. (2002) reported Sikron F-600 (QUIN 2) to be about 84.9% (CI: 83.3–86.4%) pure relative to NIST-SRM 1878a. Jeyaratnam and Nagar (1993) reported Sikron F-600 to be 92.5% pure for bulk and 95.5% pure for filter analysis using NIST-SRM 1878 taken as 95.5% pure. That is, they are of 92.5 and 95.5% relative purity in comparison with NIST-SRM 1878a taken as 100% pure. Chisholm (2005) in his reanalysis of the same standards as Jeyaratnam and Nagar reports Sikron F-600 to be 96.3 ± 1.4% pure based on peak area intensities and 101.2 ± 1.8% pure based on measurement of peak heights, and an explanation for the differences is given in his paper. These differences in comparisons are due to several factors and they have been discussed by Verma and Shaw (2001)Kauffer et al., (2002), Chisholm (2005). They include (i) batch effects, Sikron F-600 used in various studies is likely to have come from different batches and thereby has different crystallinities; (ii) the particle size distributions of the Sikron F-600 used in the various studies; (iii) the extent of amorphous layers; (iv) indirect or direct comparisons; (v) use of different analytical methods, i.e. IR versus XRD; (vi) different techniques using bulk or filter analysis; (vii) variation in quantification methods, i.e. if one peak or four peaks are used, or if peak heights or peak areas are used. Relationship between NIST-SRM 1878a and NIST-SRM 1878. Currently, the only well-characterized primary α-quartz standard available is the NIST-SRM 1878a. It is important that other widely used α-quartz standards such as Sikron F-600 (HSE A9950), QUIN 1 and QUIN 2 are related and coupled to the primary standard NIST-SRM 1878a. In that regard, earlier comparison papers (Jeyaratnam and Nagar, 1993; Verma and Shaw, 2001; Kauffer et al., 2002) and the more recent paper of Chisholm (2005) provide valuable contributions to the literature.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.