The effect of matrix interferences on <i>in situ</i> boron isotope analysis by laser ablation multi‐collector inductively coupled plasma mass spectrometry

<jats:sec><jats:title>Rationale</jats:title><jats:p>Boron isotope analysis of marine carbonates by laser ablation multi‐collector inductively coupled plasma mass spectrometry (LA‐MC‐ICP‐MS) offers the potential for rapid sample throughput, and the means to examine micron‐scale variations in the δ<jats:sup>11</jats:sup>B signatures of fossil skeletons and shells/tests of marine organisms. Existing studies demonstrate an acceptable level of reproducibility is achievable, but also typically show a level of accuracy outside the limits required by most applications. Here we investigate matrix interference effects as a cause of inaccuracy and imprecision.</jats:p></jats:sec><jats:sec><jats:title>Methods</jats:title><jats:p>Analyses were performed on a standard format Thermo Scientific Neptune Plus MC‐ICP mass spectrometer coupled to a New Wave Research 193 nm ArF laser ablation system. The effects of matrix interference on δ<jats:sup>11</jats:sup>B analysis were investigated through analyses of a set of reference materials with differing B/Ca ratios. Three approaches to correct for matrix‐induced effects were trialled: (1) use of matrix‐matched standards, (2) utilisation of the relationship between δ<jats:sup>11</jats:sup>B inaccuracy and<jats:sup>11</jats:sup>B/<jats:sup>43</jats:sup>Ca, <jats:sup>11</jats:sup>B/<jats:sup>40</jats:sup>ArCa<jats:sup>4+</jats:sup> or <jats:sup>11</jats:sup>B/Ca<jats:sub>interference</jats:sub> from three reference materials with known δ<jats:sup>11</jats:sup>B values and varying B/Ca ratios, and (3) direct characterisation of the (sloping) interference itself.</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>Matrix interference from scattered Ca ions on <jats:sup>10</jats:sup>B can impede both the accuracy and the reproducibility of δ<jats:sup>11</jats:sup>B analysis by LA‐MC‐ICP‐MS. Based on analyses of two in‐house reference materials, deep sea coral PS69/3181 and inorganic calcite UWC‐1, we find approach 2, following the <jats:sup>11</jats:sup>B/Ca<jats:sub>interference</jats:sub> relationship, gives the best mean accuracies (within 0.4‰ of solution values) and external reproducibilities (± 0.5‰ 2 SD for PS69/3181). This approach has been applied to analyses of an annual growth cycle of a <jats:styled-content style="fixed-case"><jats:italic>Siderastrea siderea</jats:italic></jats:styled-content> coral and eight <jats:italic>Cibicidoides wuellerstorfi</jats:italic> benthic foraminifera. Both coral and foraminifera data match solution MC‐ICP‐MS analyses within reported uncertainties.</jats:p></jats:sec><jats:sec><jats:title>Conclusions</jats:title><jats:p>LA‐MC‐ICP‐MS can produce accurate and precise δ<jats:sup>11</jats:sup>B data to a 0.5‰ (2σ) level on <0.3 ng B after correction for Ca interference effects.</jats:p></jats:sec>