Experimental calibration of Mg isotope fractionation between aragonite and seawater

The detectable magnesium (Mg) isotope fractionation between biogenic aragonite (including aragonitic corals, bivalves, scaphopod, and sclerosponges) and seawater can potentially be applied to reconstruct sea surface temperature (SST) in the past. To calibrate this thermometer, eight sets of inorganic precipitation experiments ('free-drift') in seawater (Mg/Ca = 5 or 10) have been carried out at 25-55 degrees C over a range of degassing rate. A cleaning procedure was adopted to remove Mg contamination by sea salt, surface absorbed Mg and silicate dust as nucleation centers. The Mg isotope fractionation between cleaned aragonite and seawater-like aqueous solution varies insignificantly with Mg/Ca ratios and Mg isotope compositions of the initial solution, and the CO2-degassing rate (0-75 cc/min), but decreases noticeably with increasing temperatures having a temperature sensitivity of similar to 0.008-0.01 parts per thousand/degrees C in the following form: Delta(aragonite-Mg(aq)) approximate to 1000 ln alpha(aragonite-Mg(aq)) = 1.67(+/- 0.36) - 0.82(+/- 0.11) x 1000/T where alpha(aragonite-seawater) is the fractionation factor, and T is the absolute temperature in Kelvin. It is consistent with equilibrium fractionation between Mg2+ aquo-complex and magnesite predicted by one theoretical calculation. Qualitative comparison among Mg-bearing carbonates based on Mg-O bond strengths show the relative sequence of Mg-26 enrichment is aragonite > dolomite > magnesite > calcite. Thus, the surprising agreement indicates either the calculation overestimated Mg fractionation between magnesite and fluid, or both theoretical calculation and our calibration represent Mg isotope fractionation between MgCO30-H2O cluster and Mg2+ aquo complexes. Comparison of our calibration with the Mg isotope fractionation between biogenic aragonite and seawater suggests Mg and oxygen isotope fractionations of some biogenic aragonites (e. g., Porites sp. corals) agree with our calibration within analytical uncertainty, whereas others deviate significantly, indicating biological and/or kinetic isotope fractionation effects. Thus, although the precision of this temperature proxy is similar to +/- 3-4 degrees C (2 sigma) given our current analytical precision and temperature sensitivity of the calibration, several factors are limiting the application of this proxy to extract precise sea-surface temperatures (SSTs) from the biogenic aragonite, including the requirement for high analytical precision, proper sample cleaning procedures, uncertainty in delta Mg-26 values of seawater, and possible kinetic and/or biological isotope fractionation. (C) 2012 Elsevier Ltd. All rights reserved.