Empirical calibration of the clumped isotope paleothermometer using calcites of various origins

We present the first universal calibration of the clumped isotope thermometer for calcites of various mineralizing types. These are an eggshell of an ostrich, a tropical bivalve, a brachiopod shell, cold seep carbonate, and three foraminifera samples that grew between 9 and 38 degrees C. CaCO3 was digested at 90 degrees C using a common acid bath. Considering a difference in phosphoric acid fractionation factors between reaction at 25 and 90 degrees C of 0.069 parts per thousand (Guo et al., 2009), the function between growth temperature T and the excess of C-13-O-18 bonds in the evolved CO2 is expressed by a linear regression between 1/T-2 and absolute Delta(47) (R-2 = 0.9915): Delta(47) = 0.0327(+/- 0.0026) x 10(6)/T-2 + 0.3030(+/- 0.0308) (with Delta(47) in parts per thousand and T in K). Both the slope and intercept of our regression line deviate significantly from the first experimental calibration based on synthetic calcites digested at 25 degrees C (Ghosh et al., 2006a) and from several other studies having confirmed this pioneering calibration (i.e., Came et al., 2007; Tripati et al., 2010; Thiagarajan et al., 2011; Grauel et al., 2012; Saenger et al., 2012; Zaarur et al., 2013). However, our relationship between temperature and absolute Delta(47) values is indistinguishable from that determined by Henkes et al. (2013) if the same difference in phosphoric acid fractionation factors between 25 and 90 degrees C is applied to both datasets. Our study and that of Henkes et al. (2013) have in common that data were primarily projected onto the absolute scale proposed by Dennis et al. (2011) - a reference frame that allows comparison of clumped isotope data measured in different laboratories. Furthermore, at any T, our regression line lies within 0.006 parts per thousand of the theoretical calcite calibration of Guo et al. (2009). The observation that both empirical calibrations are indistinguishable from each other implies that clumped isotope data can be directly compared between laboratories and referenced to a unique temperature calibration if (1) the phosphoric acid digestion temperature is set to 90 degrees C, and (2) raw data are primarily projected to the absolute scale. (C) 2014 Elsevier Ltd. All rights reserved.