Improving the understanding of H\(_2\)O-free aerosol behaviour in the inductively-coupled plasma for geochemical LA-ICP-MS applications : U-Pb dating and trace element analysis in silicate minerals and glasses

Laser ablation inductively-coupled plasma mass spectrometry (LA-ICP-MS) is an in-situ analytical technique for analysis of trace element and isotopic compositions in a wide range of materials. LA-ICP-MS has been utilized for more than 30 years in the geoscience community, but the technique still suffers from inaccuracies related to non-representative sampling of ablated material. This term has been broadly described as element fractionation and is thought to occur both at the site of ablation and in the plasma of the ICP-MS. Improvement of the LA-ICP-MS analytical technique by increased understanding of fundamental instrument processes and development of mineral standards can significantly improve geochemical and geochronological data capabilities and uses. There are two main parts of this study, the first, is an investigation into the fundamental aspects of ionization and extraction of ions into the mass spectrometer using both laser ablation and nebulizing a solution of dilute nitric acid with ppb level solutes. The second is an investigation using nanosecond pulse width excimer lasers to analyse a variety of minerals and glasses for their trace element, U-Pb isotopic and Pb isotopic compositions and comparing the results to published values to assess sources of bias in the LA-ICP-MS analytical technique. The use of ICP-MS for laser ablation was not initially considered during the design of ICP-MS instrumentation because most previous applications used a nebulizer and spray chamber to introduce liquid solutions. While early work showed that a laser ablation system could easily be coupled to an ICP-MS, with several advantages including lower polyatomic species from lack of H2O; little work has been done on understanding how the laser ablation aerosol (dry plasma) interacts with the plasma and ion extraction process compared to traditional solution nebulization (wet plasma). This study builds on previous work to demonstrate the differences in the electrical potential of the ICP between wet and dry plasma, as this has an effect on the distribution of ion energies that are sampled from the ICP. This distribution of ion energies is a function of plasma temperature, and temperatures calculated in this work suggest that dry plasma conditions are cooler than wet plasma conditions (2441 K vs. 3055 K). Ion energy distributions are also affected by perturbations in the ion beam such as space charge effects and this work suggests there are likely differences in space charge, or spatial biasing light vs. heavy ions in the ion beam, between the wet and dry plasma ion beams. Oxide and hydroxide polyatomic species are a common issue in ICP-MS, particularly for elements with high MO\(^+\) bond energies (e.g. La). This study investigates the formation of oxide and hydroxide species between wet and dry plasma by investigating the parameters controlling formation, ways to suppress these interferences, and the interference’s ion energy distributions. The interference’s ion energy distributions can provide information about the location of formation within the plasma / ICP-MS, the knowledge of which aides in preventing and minimizing interferences. Oxide formation rates in dry plasma are significantly less than wet plasma (>10x lower) which is consistent with previous work. However, this study also demonstrates that a greater stopping potential is needed to attenuate the oxide signal relative to atomic ion signals for wet plasma compared to dry plasma (4 V vs. 1 V for wet and dry plasma respectively). This demonstrates there is a fundamental difference in the formation of oxide species between the two modes of ICP-MS operation; likely related to source of oxygen between wet and dry plasma. There is a similar effect on the hydroxide polyatomic species between wet and dry plasma, however the formation of hydroxide species and oxide species is not linked for dry plasma, suggesting a different mechanism of formation. Absorbed atmospheric air and water vapour can have a significant negative affect on accuracy of elements such as \(^{57}\)Fe in high Ca matrixes, and this has implications for how samples are stored prior to analysis. Addition of diatomic gases (N\(_2\) and H\(_2\)) are common in LA-ICP-MS to improve sensitivity, however, the effect of these gases on the properties of the ion beam and ion energy distribution has not been previously explored. This study shows that the addition of H\(_2\) has a significant impact on the ion energy distribution of the ICP-MS while N\(_2\) has no effect on ion energy distribution despite similar ionization energies for both gases. N\(_2\) and to a lesser extent H\(_2\) can change how a plasma will digest particles during LA-ICP-MS by cooling the plasma and allowing for incomplete digestion of particles. Incomplete particle digestion is potential sources of element fractionation in ICP-MS and an improved understanding of these process will allow for better method development and improved accuracy of results. U-Pb geochronology in zircon and apatite by LA-ICP-MS is commonplace in the geological community. However, despite this, there are situations where the precision is better than accuracy, leading to inaccurate age dating. This study contributes to the understanding of causes of this U-Pb inaccuracy by analysing a range of zircon and apatite reference materials with independent age constraints. Optimization of laser parameters can improve the U-Pb dating method by reducing the amount of element fractionation during aerosol generation. Small variations in nanosecond pulse width (5 vs. 15) are shown to have a negligible impact on the accuracy of the technique and suggests the causes of intra-lab discrepancies are unlikely due to differences in source laser. However, lower (< 2 J/cm\(^2\) ) laser fluence will result in shallower laser craters and reduced element fractionation leading to more accurate results compared to analyses performed at higher fluence. Differences in atmospheric air in the laser cell or system is an additional, previously unknown, source of inaccuracy of zircon U-Pb analysis. Because of the lack of H\(_2\)O species in the plasma, trace amounts of water vapour, O\(_2\) or N\(_2\) gas can significantly change the properties of the ICP or local conditions at the ablation site. Using a set of controlled degased and un-degased samples, up to a 4% difference in age throughout the laser cell is correlated to the amount of atmospheric air present. The zircon matrix effect during U-Pb dating by LA-ICP-MS has been previously shown to correlate to the radiation dose a zircon has received. This effect on measured U-Pb ages as a function of radiation damage is reduced by using a lower laser fluence (~1.5% difference in age for high radiation dose zircon). Additionally, a correction to the measured zircon U-Pb age based on the radiation dose has been developed using analysis of a wide range of zircon reference materials with varying radiation doses. The addition of new matrix-matched reference materials will allow for U-Pb quantification and reduce the mineral-specific fractionation offsets possible in geochronology. Two new apatite reference materials for U-Pb dating: 401 apatite and the OD306 apatite have been developed and made available to the geochronological community. These two apatite materials are close to concordant (1% to 10% discordant) and have isotope dilution ages, neither of which is true for most of the current apatite U-Pb reference materials available. Quasi-simultaneous detection offered by time-of-flight (TOF) technology is compared to traditional sequentially-based measurements done by quadrupole mass filters. Use of math based measurements are inherently disadvantageous for transient signals from LA-ICP-MS as the more elements in an analytical method, the more detail will be missed or mis-represented due to spectral skew. However, few studies have utilized the TOF technology for LA-ICP-MS. This study is the first to demonstrate a direct comparison of U-Pb dating of zircon using a split-stream arrangement between a quadrupole and a TOF-ICP-MS. The results of these experiments show that similar accuracy and precision can be obtained for each type of ICP-MS, despite the significantly lower (>10x) sensitivity of the TOF instrument. This study shows the potential of the TOF technology for wider application by the LA-ICP-MS community