doi:10.1016/j.gca.2005.06.013 Redox potential measurements and Mössbauer spectrometry of Fe II adsorbed onto Fe III (oxyhydr)oxides EWEN SILVESTER, 1, *LAURENT CHARLET, 2 CHRISTOPHE TOURNASSAT, 2,† ANTOINE GÉHIN, 2 JEAN-MARC GRENÈCHE, 3 and EMMANUELLE LIGER 2 1 Department of Environmental Management and Ecology, La Trobe University, Albury-Wodonga Campus, PO Box 821, Wodonga, Victoria, 3689, Australia 2 Department of Earth and Planetary Sciences (LGIT-OSUG), University of Grenoble-I, BP 53, 38041 Grenoble, France 3 Laboratoire de Physique de l’Etat Condensé (LPEC), UMR-CNRS 6087, Université du Maine, 72085 Le Mans, France (Received July 23, 2004; accepted in revised form June 20, 2005) Abstract—The redox properties of Fe II adsorbed onto a series of Fe III (oxyhydr)oxides (goethite, lepidocrocite, nano-sized ferric oxide hydrate (nano-FOH), and hydrous ferric oxide (HFO)) have been investigated by rest potential measurements at a platinum electrode, as a function of pH (-log 10 [H + ]) and surface coverage. Using the constant capacitance surface complexation model to describe Fe II adsorption onto these substrates, theoretical values of the suspension redox potential (E H ) have been computed, under the assumption that Fe II adsorption occurs at crystal growth sites of the substrate surface. Good agreement between calculated and experimental E H values is observed for nano-FOH and HFO, however the redox potentials measured for lepidocrocite and goethite are significantly more oxidizing than predicted. Mössbauer spectroscopic analysis of 57 Fe II adsorbed onto HFO and goethite shows that in both cases the adsorbed 57 Fe II is incorporated into the crystal structure of the substrate, in broad agreement with the thermodynamic model, but is almost completely oxidized to 57 Fe III . The mechanism by which the adsorbed 57 Fe II is oxidized is not resolved in this work, but is thought to be due to electron transfer to the substrate, rather than a net oxidation of the suspension. The disagreement between experimental and calculated rest potential measurements in the goethite and lepidocrocite systems is thought to be due to the poor electrochemical equilibration of these suspensions with the platinum electrode, rather than a failure of the thermodynamic model. The model developed for the redox potential of adsorbed Fe II allows direct assessment of the reactivity of this species towards oxidized pollutants. Copyright © 2005 Elsevier Ltd 1. INTRODUCTION Over the past decade a number of studies have reported on the enhanced rate of electron transfer from Fe II adsorbed on mineral surfaces, compared to that which occurs in bulk solution (e.g., Stumm and Sulzberger, 1992; Haderlein and Pecher, 1999). This enhancement in reactivity has been attributed to the effect of oxygen coordination of the ad- sorbed Fe II at surface sites (Werhli et al., 1989), similar to that which occurs upon hydrolysis of Fe II in solution. Sur- face enhancement of electron transfer from Fe II occurs in a number of systems including; the reduction of nitrite by Fe II adsorbed on goethite (-FeOOH) (Sørensen and Thorling, 1991), reduction of substituted nitroaromatics by Fe II ad- sorbed on magnetite (Fe 3 O 4 ), goethite and lepidocrocite (-FeOOH) (Klausen et al. 1995; Hofstetter et al., 1999; Elsner et al., 2004; Gregory et al., 2004), the reduction of the uranyl ion (UO 2 2+ ) by Fe II adsorbed on ferric oxide nano- particles (Liger et al., 1999), the reduction of chromate (CrO 4 2- ) by Fe II adsorbed on goethite and lepidocrocite (Buerge and Hug, 1999), and the reduction of chlorinated alkanes by Fe II adsorbed onto a range of Fe III (oxyhydr)ox- ides (Amonette et al, 2000; Pecher et al. 2002; Elsner et al., 2004; Maithreepala and Doong, 2004). The effect is not limited to Fe III (oxyhydr)oxides, with similar rate enhance- ment observed for the reduction of substituted nitrobenzene compounds by Fe II adsorbed onto kaolinite (Klausen et al., 1995) and nontronites (Hofstetter et al., 2003). The influence of the mineral substrate and local crystal- lographic environment on the redox level of adsorbed Fe II remains largely unresolved. Hofstetter et al. (2003) found that for nontronite clays, structural Fe II , and edge-surface complexed Fe II were considerably more reactive towards substituted nitrobenzenes than Fe II exchanged onto the basal siloxane surfaces. The reactivity of surface complexed Fe II also appears to be affected by the degree of hydrolysis of the adsorbed species. A strong correlation has been observed between the proportion of adsorbed Fe II present as the hydrolyzed Fe III OFe II OH species, and the rates of reduc- tion of 4-chloronitrobenzene (Charlet et al., 1998) and UO 2 2+ (Liger et al., 1999). This interpretation has been challenged more recently by Elsner et al. (2004) as it does not explicitly take into account the role of the mineral substrate on the reactivity of the adsorbed species. The counter argument is that the hydrolysis of adsorbed Fe II will depend upon the nature of the substrate, and as there are very few systems in which the surface speciation of adsorbed Fe II has been determined, it is rather difficult to state conclusively whether hydrolysis of adsorbed Fe II generally leads to rate enhance- ment. What does seem likely is that the reactivity of ad- * Author to whom correspondence should be addressed (e.silvester@ latrobe.edu.au). † Present address: BRGM, EPI/DES, Environmental Processes Divi- sion, 3 Avenue Claude Guillemin, 45060 Orléans, Cedex 2, France. Geochimica et Cosmochimica Acta, Vol. 69, No. 20, pp. 4801– 4815, 2005 Copyright © 2005 Elsevier Ltd Printed in the USA. All rights reserved 0016-7037/05 $30.00 + .00 4801