Silicon Isotope Ratios in Cais: In-situ Laser-ablation Mc-icpms Measurements and Comparisons with Magnesium Isotope Ratios

Introduction: High-precision measurements of Mg/Mg by multiple-collector inductively coupled plasma-source mass spectrometry (MC-ICPMS) have been used recently in our laboratory to constrain the thermal history of calcium aluminum-rich inclusions (CAIs) [1]. Presence of primary zoning profiles in at least some CAIs points to supersolidus and subsolidus diffusion as important factors in controlling isotope fractionation. Previous model calculations suggested that the details of the isotope selectivity of Mg isotope diffusion in igneous CAI melts may ultimately control the overall degree of heavy isotope enrichment during evaporation [2]. Our work suggests that subsequent heating in the solid state imposes modifications to the original isotopic signals [1]. We are looking for evidence for both processes in CAIs. The importance of isotope fractionation during diffusion in CAI melts is illustrated with a finite difference model for Mg isotope fractionation of a CAI starting with a diameter of 0.3 cm held at liquidus temperature for 60 minutes (Fig. 1). The model shows enrichment in δMg only when proper account is made for the slight disparity in diffusivities between Mg and Mg. In this calculation we used D(Mg)=(24/25) D(Mg) with β=0.1 as suggested by [3]. In contrast, when β = 0, isotope fractionation is confined to a boundary layer and the degree of enrichment in the body of the object outside of the boundary layer is zero (Fig. 1). Comparisons between Mg isotope ratios and Si isotope ratios should help constrain the role of diffusion in determining the isotopic composition of igneous CAIs. The β relating the diffusivities of the Si isotopes in melt, D(Si) and D(Si), is thought to be much smaller than the value for Mg (< 0.025 for Si compared with ~ 0.1 for Mg) [3]. Accordingly, one expects δSi to be systematically lower than δMg in evaporated igneous CAIs (both relative to chondritic values) given the similar volatility of these elements. Bulk measurements give hints of this, but isotopic resetting during prolonged heating of CAIs makes interpretation of bulk properties uncertain [4]. Profiles of isotope ratios across the objects are needed to understand the extent of resetting. Until now, detailed comparsisons between Mg and Si isotope ratios within individual CAIs have proven difficult due to analytical uncertainties. Here we explore the use of ultraviolet laser ablation MC-ICPMS (UV LA-MC-ICPMS) as a tool for measuring Si/Si and Si/Si in CAIs. LA-MC-ICPMS Method: In-situ Si isotope ratio analyses were measured using a 193 nm excimer laser (NewWave Research UP) coupled to a ThermoFinnigan Neptune MC-ICPMS at UCLA. Correction for instrumental mass bias was made using samplestandard bracketing. The standard was a sample of San Carlos olivine (SC olivine). The mass spectrometer was operated at a moderate mass resolving power of ~ 7500 (entrance slit of 30 μm, 2R = 812 mm), permitting elimination of N2 and NO interferences on m/z = 28 and 30, respectively, while minimizing signal losses. A preliminary investigation of the effects of Fe and Fe (not a factor for most CAI analyses) was undertaken with solutions. We could detect no measurable m/z = 28 in a 5 ppm Fe solution in weak HNO3 acid. Silicon isotope ratios measured on 3ppm Si in the presence of 0.1, 0.3, and 1 ppm Fe showed no departures from mass dependent fractionation in Si/Si and Si/Si. On-peak blanks were measured before each analysis to correct for 0.4%, 3%, and 3% backgrounds on m/z = 28, 29, and 30, respectively. Laser spot size was 100 μm to 150 μm depending CAI liquid evaporation model