3.6 Chromium concentrations and δ53Cr
Procedures for the determination of total dissolved Cr concentrations ([Cr]) and stable isotope composition (δ53Cr) are described in detail in Rickli et al. (2019), with minor modifications (see supplemental material; Janssen et al., 2020; Nasemann et al., 2020). Briefly, preliminary [Cr] for spiking purposes (not reported) was determined by isotope dilution on small sample aliquots by Mg(OH)2 co-precipitation followed by cation exchange chromatography. The final reported [Cr] is from δ53Cr determinations, except when δ53Cr samples were not available (pore waters and incubation intermediate time points). Pore water samples (0.2 mL) were dried rather than co-precipitated before chromatography. For δ53Cr, a 0.25-1 L sample aliquot was spiked with50Cr-54Cr double spike, pre-concentrated by Mg(OH)2 co-precipitation, and purified by a 2 step column chromatography.
Samples were analyzed on a ThermoFisher Neptune Plus MC-ICP-MS at the University of Bern following Rickli et al. (2019). δ53Cr procedural blanks were 0.3-0.6 ng Cr and insignificant relative to sample Cr. δ53Cr data are reported relative to the NIST SRM 979. Internal uncertainties were generally 0.02-0.03 ‰ (2 SEM); external reproducibility, based on replicate analyses of seawater samples, is ± 0.033 (2 SD) for δ53Cr and <1% (1 RSD) for [Cr] (see Janssen et al., 2020). In addition to the zero reference NIST SRM 979, a Merck Cr(III) standard was run to monitor accuracy (δ53Cr = -0.426 ± 0.027 ‰, n = 9; literature: δ53Cr = -0.443 ± 0.022 ‰, Schoenberg et al., 2008). The procedural accuracy of our method has been validated by an inter-laboratory comparison (Rickli et al., 2019).