Search

Boron isotopic analysis is of interest in many research fields and for a large variety of sample types. Accurate and precise determination of boron isotope ratios using multi-collector ICP-mass spectrometry or thermal ionization mass spectrometry requires isolation of the target element prior to isotopic analysis, which is accomplished using either ion exchange chromatography or microsublimation. This study systematically compares the two methods in terms of B recovery, procedural blank, matrix removal efficiency, accuracy and precision of the resulting δ11B values as measured using multi-collector ICP-mass spectrometry, as well as labor intensiveness and costs. For this purpose, four types of sample matrices, i.e. 20 g L-1 Ca aqueous solution, seawater, digests of spinach (100 g L-1) and silicate glass (10 g L-1), were stripped from their original B content and spiked with δ11B of known isotopic composition and were then subjected to both sample preparation methods and subsequent isotopic analysis of the purified B fraction via multi-collector ICP-mass spectrometry. For both methods, the highest (quantitative) B recoveries were obtained for Ca-rich aqueous solution and seawater. For spinach, accurate d B values were obtained after ion exchange chromatography. However, microsublimation was plagued by isotope fractionation, resulting in large offsets (~8%) between the experimental results and the corresponding reference values. For glass, B recovery was incomplete, nevertheless the absence of fractionation rendered both sample preparation methods suitable. Overall, in the absence of isotope fractionation, microsublimation appears advantageous in terms of procedural blanks, matrix removal efficiency, precision (2 s) on δ11B values, labor intensiveness and costs. [ABSTRACT FROM AUTHOR]

A simple and accurate methodology for Br isotope ratio measurements in seawater by multi-collector inductively coupled plasma-mass spectrometry (MC-ICP-MS) with pneumatic nebulization for sample introduction was developed. The Br(+) signals could be measured interference-free at high mass resolution. Memory effects for Br were counteracted using 5 mmol L(-1) of NH4OH in sample, standard, and wash solutions. The major cation load of seawater was removed via cation exchange chromatography using Dowex 50WX8 resin. Subsequent Br preconcentration was accomplished via evaporation of the sample solution at 90 °C, which did not induce Br losses or isotope fractionation. Mass discrimination was corrected for by external correction using a Cl-matched standard measured in a sample-standard bracketing approach, although Sr, Ge, and Se were also tested as potential internal standards for internal correction for mass discrimination. The δ(81)Br (versus standard mean ocean bromide (SMOB)) values thus obtained for the NaBr isotopic reference material NIST SRM 977 and for IRMM BCR-403 seawater certified reference material are in agreement with literature values. For NIST SRM 977, the (81)Br/(79)Br ratio (0.97291) was determined with a precision ≤0.08‰ relative standard deviation (RSD).

The provenance of the flux raw material used in the manufacturing of Roman glass is an understudied topic in archaeology. Whether one or multiple sources of natron mineral salts were exploited during this period is still open for debate, largely because of the lack of a good provenance indicator. The flux is the major source of B in Roman glass. Therefore, B isotopic analysis of a sufficiently large collection and variety (origin and age) of such glass samples might give an indication of the number of flux sources used. For this purpose, a method based on acid digestion, chromatographic B isolation and B isotopic analysis using multicollector inductively coupled plasma mass spectrometry was developed. B isolation was accomplished using a combination of strong cation exchange and strong anion exchange chromatography. Although the B fraction was not completely matrix-free, the remaining Sb was shown not to affect the δ(11)B result. The method was validated using obsidian and archaeological glass samples that were stripped of their B content, after which an isotopic reference material with known B isotopic composition was added. Absence of artificial B isotope fractionation was demonstrated, and the total uncertainty was shown to be <2‰. A proof-of-concept application to natron glass samples showed a narrow range of δ(11)B, whereas first results for natron salt samples do show a larger difference in δ(11)B. These results suggest the use of only one natron source or of several sources with similar δ(11)B. This indicates that B isotopic analysis is a promising tool for the provenance determination of this flux raw material.