Preliminary results and perspectives in the production of artificial glasses as calibration standards for Li and B microanalysis

It has been recognised since a long time that the Achille's heel of the micro-beam techniques is the scarcity of reliable standardisation materials. The international reference materials to date available are often in the powder form or represent chemically purified solutions or salts, which limits in practice their application. With the rapid increase of isotope analysis by micro-beam mass spectrometry, the need for well-characterised homogeneous isotope Standard Reference Materials (SRM's) has increased considerably. Various procedures to produce glassy materials that satisfy the request of elemental and isotope homogeneity are nowadays under testing and development. Ødegård (1999) and Ødegård et al. (2001) proposed a method to obtain glassy materials from oxide powders (quartz, rutile and ilmenite) to be used for laser ablation standardisation. The method relies on the use of pre-formed graphite capsules of ¼"external diameter? containing a few tens of mg of sample. For the fusion step, which is performed in atmosphere, the capsules are closed and supplied by a high intensity (up to 130 A) direct current. In this work we investigate the potential of the above approach to produce glasses starting from international SRM's powders of complex materials, such as silicate rocks. Our aim is to obtain Li and B homogeneous glasses to be used as calibration standards in the elemental (and isotope) analysis. Starting from SRM's powders should make the characterisation of artificial glasses easier and more rapid. We started from 200 mg aliquots of SRM JB-2 basalt (Geological Survey of Japan, http://www.gsj.jp), whose Li and B elemental and isotope composition is available as literature data as resulting from independent investigations of different laboratories. The aliquots were processed in home-made capsules of variable length (all of 3.2 mm inner diameter) under various heating times. The resulting glasses were then embedded in resin and polished for micro-beam analysis. The charges were imaged by SEM (back-scattered electrons), and the glasses were then analysed in terms of major and minor elements by EMP. The more representative charges were then investigated for Li and B content by SIMS. Preliminary results indicate that the best capsule geometry consists in a medium-sized (25 mm long) top-bored chamber, heated for 40 s with a current of 100 A. In such a way, it was possible to produce glasses with only minor heterogeneities that appeared to be concentrated along the contact between the graphite and the glass. Precisely, the glass suffered a slight Fe extraction expressed by the presence of metal beads (< 5 ?m ?) segregated along the peripheral thin layer (few tens of microns width). In the rest of the charges, major-element composition is consistent with the recommended values for SRM JB-2. Additionally, the same zones of the glasses, investigated by SIMS, show elemental ratios (Li/B, Li/Al, Li/Si, B/Al, B/Si) that are constant within the analytical uncertainty. The derived Li and B concentrations are comparable to the literature data. This melting procedure seems to represent a good perspective for a rapid and easy production of homogeneous glasses to be used as reference materials in the microanalysis of Li and B in a wide range of chemical compositions in silicates.