Fe 2+ catalyzed iron atom exchange and re-crystallization in a tropical soil Viktor Tishchenko a , Christof Meile b , Michelle M. Scherer c , Timothy S. Pasakarnis c , Aaron Thompson a,⇑ a University of Georgia (Crop and Soil Science), United States b University of Georgia (Marine Science), United States c The University of Iowa (Civil and Environmental Engineering), United States Received 22 April 2014; accepted in revised form 8 September 2014; available online 28 September 2014 Abstract Aqueous ferrous iron (Fe 2+ (aq) ) is known to transfer electrons and exchange structural positions with solid-phase ferric (Fe III ) atoms in many Fe minerals. However, this process has not been demonstrated in soils or sediments. In a 28-day sterile experiment, we reacted 57 Fe-enriched Fe 2+ (aq) ( 57/54 Fe = 5.884 ± 0.003) with a tropical soil (natural abundance 57/54 Fe = 0.363 ± 0.004) under anoxic conditions and tracked 57/54 Fe in the aqueous phase and in sequential 0.5 M and 7 M HCl extractions targeting surface-adsorbed and bulk-soil Fe, respectively; we also analyzed the reacted soil with 57 Fe Mo ¨ ssbauer spectroscopy. In 28 days, the aqueous and bulk pools both moved 7% toward the isotopic equilibrium ( 57/54 Fe = 1.33). Using a kinetic model, we calculate final adsorption-corrected 57/54 Fe ratios of 5.56 ± 0.05 and 0.43 ± 0.03 in the aqueous and bulk pools, respectively. The aqueous and surface/labile Fe initially exchanged atoms rapidly (10–80 mmol kg 1 d 1 ) decreasing to a near constant rate of 1 mmol kg 1 d 1 that was close to the 0.74 mmol kg 1 d 1 exchange-rate between the surface and bulk pools. Thus, after 28 days we calculate aqueous Fe has exchanged with 20.1 mmol kg 1 of bulk Fe atoms (1.9% of total Fe) in addition to the 17.0 mmol kg 1 of surface/labile Fe atoms (1.6% of total Fe), which have likely turned over several times during our experiment. Extrapolating these rates, we calculate a hypo- thetical whole-soil turnover time of 3.6 yrs. Furthermore, Mo ¨ ssbauer spectroscopy indicates the soil-incorporated 57 Fe label re-crystallized as short-range-ordered (SRO) Fe III -oxyhydroxides: our model suggests this pool could turnover in less than seven months via Fe 2+ -catalyzed recrystallization. Thus, we conclude Fe atom exchange can occur in soils at rates fast enough to impact ecological processes reliant on Fe minerals, but sufficiently slow that complete Fe mineral turnover is unlikely, except perhaps in permanently anoxic environments. Ó 2014 Elsevier Ltd. All rights reserved. 1. INTRODUCTION Iron (Fe) minerals strongly affect the behavior of nutri- ents (Peretyazhko and Sposito, 2005) and pollutants (Wielinga et al., 2001; Charlet et al., 2002) in many ecosys- tems, with the extent of this influence depending on mineral composition and size. The smallest or most disordered Fe phases, called short-range-ordered (SRO) Fe, are the most reactive phases for a range of processes including chemical sorption (Stumm, 1987; Deng and Stumm, 1994; Ko ¨ gel-Knabner et al., 2008), particle adhesion, and reduc- tive dissolution (Sulzberger et al., 1989; Lovley et al., 2004). Quantifying these processes is challenging in part because Fe mineral composition is not fixed, but changes in response to environmental conditions that drive http://dx.doi.org/10.1016/j.gca.2014.09.018 0016-7037/Ó 2014 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. E-mail address: AaronT@uga.edu (A. Thompson). www.elsevier.com/locate/gca Available online at www.sciencedirect.com ScienceDirect Geochimica et Cosmochimica Acta 148 (2015) 191–202