doi:10.1016/j.gca.2003.08.010 Heterogeneous reduction of uranyl by micas: Crystal chemical and solution controls EUGENE S. ILTON, 1, *ANCA HAIDUC, 2 CARL O. MOSES, 3 STEVE M. HEALD, 4 DAVID C. ELBERT, 5 and DAVID R. VEBLEN 5 1 Pacific Northwest National Laboratory, PO Box 999, MSIN: K8-96, 902 Battelle Blvd., Richland, WA 99352, USA 2 Shell Global Solutions International B.V., P.O. Box 38000, 1030 BN Amsterdam, The Netherlands 3 Lehigh University, Department of Earth &Environmental Sciences 31 Williams Drive Bethlehem, PA 18015, USA 4 PNC-CAT, Bldg. 435E, 9700 S. Cass Ave., Argonne, IL 60439, USA 5 Department of Earth and Planetary Sciences, The Johns Hopkins University, Baltimore, MD 21218, USA (Received April 16, 2003; accepted in revised form August 19, 2003) Abstract—This contribution primarily uses X-ray photoelectron spectroscopy (XPS) to better understand mechanisms for coupled sorption-reduction of aqueous U VI by ferrous micas. Additional information was obtained with X-ray absorption spectroscopy (XAS) and transmission electron microscopy (TEM). The research is important because homogeneous reduction of aqueous U VI is sluggish compared to heterogeneous reduction pathways, and micas are important sorbents for uranium in granitic terrains, which have been proposed as potential radionuclide waste disposal sites. Three micas (high, medium and low Fe/Mg biotites), prepared as thin centimeter-sized books, were reacted with U VI solutions that contained 0 –25 mM Na  or K  , at pH  4.5, 5.0, 6.0, and 9.5. All the experiments were performed under argon at one bar. Solid samples were retrieved at timed intervals for up to 20 h. Both mica edge and basal plane orientations were analyzed by XPS. Analyses of peak positions, core satellites, and the 5f valence band indicate that U VI can be reduced by biotite and that heterogeneous reduction depends on the type and concentration of alkali cation in solution, crystallography, and mica composition. We conclude that ferrous micas can reduce U VI on edge sites, but not on exposed basal plane surfaces, and that Na  and K  facilitate and hinder the reaction, respectively. These observations pertain over a broad range of pH. X-Ray absorption near edge spectroscopy (XANES) of annite sections indicates that the interlayer region, as opposed to external basal surfaces, also offers possible sites for heterogeneous reduction of U VI . TEM of annites with high uranium coverage confirmed the presence of interlayer uranium; interestingly, this uranium is concentrated in U-rich nano-scale zones. Copyright © 2004 Elsevier Ltd 1. INTRODUCTION Over the past 50 years there has been an intense effort to characterize the interaction of U VI with geological materials. The motivation has stemmed from dual desires to find and exploit uranium ore deposits and to isolate long-lived radionu- clide wastes. The primary goal of this contribution is to deter- mine whether Fe II -bearing micas can sorb and reduce aqueous U VI to U IV . Phyllosilicates are known sorbents for aqueous U VI , and some of the rock-forming species are ubiquitous reservoirs of Fe II , a possible reducing agent for U VI . Reduction of soluble U VI to less-soluble U IV is a mechanism for limiting the mobility of U in the environment (e.g., Langmuir, 1978). The interaction of uranyl with micas is an important determi- nant of the mobility of U in granitic terrains, which have been proposed as radionuclide waste disposal sites. Heterogeneous reduction mechanisms for U VI are particularly important, be- cause homogeneous reduction of aqueous U VI is sluggish (Wersin et al., 1994, and references therein; Liger et al., 1999). The literature on clay minerals is relevant because of gross structural similarities with micas. There have been numerous experimental studies on the interaction of U VI aq with layer- silicate clays (Goldsztaub and Wey, 1955; Giblin, 1980; Borovec, 1981; Giblin et al., 1981; Tsunashima et al., 1981; Ames et al., 1982, 1983a; Andersson et al., 1982; Chisholm- Brause et al., 1992, 1994; Dent et al., 1992; McKinley and Sholtis, 1993, and references therein; Vandergraaf et al., 1993, see references therein; Hudson et al., 1994; Morris et al., 1994; Ticknor, 1994; McKinley et al., 1995; Ticknor et al., 1996; Giaquinta et al., 1997; Turner et al., 1996; Morris et al., 1997; Pabalan and Turner, 1997; Redden et al., 1997; Akcay, 1998; Pabalan et al., 1998; Thompson et al., 1998; Nero et al., 1999; Syed, 1999). None of these studies, however, addressed heter- ogeneous reduction of U VI by structural Fe II . In an EXAFS study of U VI sorption by vermiculite and hydrobiotite, Hudson et al. (1999) state that U VI is not reduced to U IV . The Fe II contents of the minerals were not given, however, and exper- imental conditions might not have been conducive to reduction (i.e., O 2 and CO 2 were present; CO 2 restricts the stability field of U IV -containing compounds). The literature for experimental studies on the interaction of aqueous U VI with powdered rock-forming phyllosilicates is less extensive than for clays (Allard et al., 1979; Anderson et al., 1982; Ames et al., 1983b; Ticknor, 1994; Idemitsu et al., 1995; Mainka et al., 1999; Moyes et al., 2000). Although heterogeneous reduction of U VI by ferrous silicates was con- sidered to be potentially important in the above studies, no effort was made with advanced analytical techniques to deter- mine whether or not such a reaction occurred, or to explore conditions under which heterogeneous reduction might be fa- vorable. Idemitsu et al. (1995) noted that some U was sorbed irreversibly on biotite and postulated that this fraction repre- sented U IV . The lack of direct evidence for U IV , however, compromises this conclusion of their study. A number of studies have used micro-analytical techniques * Author to whom correspondence should be addressed (Eugene.Ilton@PNL.gov). Pergamon Geochimica et Cosmochimica Acta, Vol. 68, No. 11, pp. 2417–2435, 2004 Copyright © 2004 Elsevier Ltd Printed in the USA. All rights reserved 0016-7037/04 $30.00  .00 2417