ARTICLE pubs.acs.org/est
Effect of Aqueous Fe(II) on Arsenate Sorption on Goethite and Hematite Jeffrey G. Catalano,* Yun Luo,† and Bamidele Otemuyiwa Department of Earth and Planetary Sciences, Washington University, St. Louis, Missouri 63130, United States
bS Supporting Information ABSTRACT: Biogeochemical iron cycling often generates systems where aqueous Fe(II) and solid Fe(III) oxides coexist. Reactions between these species result in iron oxide surface and phase transformations, iron isotope fractionation, and redox transformations of many contaminant species. Fe(II)-induced recrystallization of goethite and hematite has recently been shown to cause the repartitioning of Ni(II) at the mineral water interface, with adsorbed Ni incorporating into the iron oxide structure and preincorporated Ni released back into aqueous solution. However, the effect of Fe(II) on the fate and speciation of redox inactive species incompatible with iron oxide structures is unclear. Arsenate sorption to hematite and goethite in the presence of aqueous Fe(II) was studied to determine whether Fe(II) causes substantial changes in the sorption mechanisms of such incompatible species. Sorption isotherms reveal that Fe(II) minimally alters macroscopic arsenate sorption behavior except at circumneutral pH in the presence of elevated concentrations (10 3 M) of Fe(II) and at high arsenate loadings, where a clear signature of precipitation is observed. Powder X-ray diffraction demonstrates that the ferrous arsenate mineral symplesite precipitates under such conditions. Extended X-ray absorption fine structure spectroscopy shows that outside this precipitation regime arsenate surface complexation mechanisms are unaffected by Fe(II). In addition, arsenate was found to suppress Fe(II) sorption through competitive adsorption processes before the onset of symplesite precipitation. This study demonstrates that the sorption of species incompatible with iron oxide structure is not substantially affected by Fe(II) but that such species may potentially interfere with Fe(II)-iron oxide reactions via competitive adsorption.
’ INTRODUCTION Biogeochemical processes in aquatic, soil, groundwater, and sedimentary environments often involve redox transformations of Fe. In such systems aqueous Fe(II) and solid Fe(III) oxide minerals commonly coexist.1,2 Reaction between Fe species in these two oxidation states transforms ferrihydrite into more crystalline iron oxides,3 5 fractionates iron isotopes,6 8 and activates localized growth and dissolution of iron oxide surfaces.9 11 Recent work has demonstrated that Fe(II) oxidatively adsorbs to iron oxides12 15 in a process that involves atom exchange and electron transfer.16 18 Heterogeneous systems containing Fe(III) oxides have been shown to catalyze the reduction of contaminants such as U(VI), Cr(VI), Tc(VII), halogenated hydrocarbons, and nitroaromatic compounds by aqueous Fe(II).19 24 Such systems also appear to catalyze the oxidation of As(III) to As(V),25 explaining the enhanced sorption of As(III) in the presence of Fe(II).26 The mechanism for this process is still being debated as in the presence of magnetite and ferrihydrite O2 is required for oxidation to occur,27 which is attributed to a Fenton-type reaction.28 Less clear is the effect of Fe(II) on the fate of redox-inactive species at iron oxidewater interfaces. Fe(II) has been shown to cause hysteresis in the adsorption of divalent metals onto goethite29 although a later study did not observed a similar effect on hematite.30 We have recently demonstrated that Fe(II) may substantially alter the fate of Ni(II) in the presence of goethite and hematite, causing the incorporation r 2011 American Chemical Society
of adsorbed Ni into the mineral structure and the release of previously incorporated Ni back into solution.31 This process is ultimately driven by iron oxide recrystallization induced by Fe(II). Ni and other divalent metal cations are structurally compatible with iron oxide minerals and are known to substitute at levels of up to a few mol %.32 35 Other contaminant species, especially those that occur as oxoanions, lack this structural compatibility and the effect of Fe(II)-induced iron oxide recrystallization on their fate is unclear. In the present study the sorption of arsenate on hematite and goethite in the presence of aqueous Fe(II) is investigated to explore how such recrystallization affects a structurally incompatible contaminant species. Arsenate was selected as it generally cannot be abiotically reduced by Fe(II),36 with As(III) oxidation the apparently favored reaction direction in such systems.25 Macroscopic signatures of the alteration of arsenate sorption behavior by Fe(II) were examined using sorption isotherms. In addition, changes in arsenate sorption mechanisms were evaluated using X-ray absorption fine structure (XAFS) spectroscopy and powder X-ray diffraction (XRD). Sorption isotherms were also employed to characterize the effect of arsenate on macroscopic Fe(II) sorption behavior. Received: July 15, 2011 Accepted: September 7, 2011 Revised: August 29, 2011 Published: September 07, 2011 8826
dx.doi.org/10.1021/es202445w | Environ. Sci. Technol. 2011, 45, 8826–8833
Environmental Science & Technology
’ MATERIALS AND METHODS Mineral Synthesis. Hematite and goethite were prepared following previously described procedures.37 Goethite was prepared by hydrolysis of a Fe(NO3)3 solution with KOH followed by aging at 70 °C for 3 days. Hematite was prepared by aging an acidic FeCl3 solution at 98 °C for 7 d. Both solids were washed repeatedly in deionized water (>18.2 MΩ cm) to remove excess electrolytes. The minerals were suspended in deionized water at concentrations of 20 35 g L 1 and then transferred into an anaerobic chamber (Coy Laboratory Products) with a 4% H2/96% N2 atmosphere and a Pd catalyst to remove residual O2(g). A secondary gas filtration system was used to help maintain low-O2 conditions and to remove CO2 from the chamber atmosphere. The system consisted of two gas washing bottles in sequence through which the chamber atmosphere was bubbled. The first contained a 15% pyrogallol/ 50% KOH solution to remove O2 and CO2; the second contained deionized water. The mineral suspensions were bubbled with the gas stream from this system overnight to remove dissolved O2 and CO2. After bubbling the dissolved oxygen content was measured colorimetrically (CHEMetrics test kit K-7511); all analyses were below the 1 μg L 1 detection limit. An aliquot was taken from each mineral suspension before use for gravimetric determination of the suspension concentration. Specific surface areas of 13 and 39 m2 g 1 for hematite and goethite, respectively, were determined using N2 BET adsorption isotherms at 77 K (Quantachrome Instruments Autosorb-1). Substantial clumping of the hematite was observed upon air drying prior to the BET measurements so the specific surface area may be underestimated. Scanning electron microscopy (SEM) images (JEOL JSM-7001F FE-SEM) of hematite (Supporting Information (SI) Figure S1) suggest a geometric specific surface area of approximately 22 m2 g 1. Batch Isotherm Measurements. Arsenate and Fe(II) sorption isotherms were measured at pH 4 and 7 on both minerals. Reaction times of 5 days were chosen as Ni incorporation into hematite and goethite in the presence of Fe(II) was found to require such time scales to occur.31 At final volumes all samples contained 4 g L 1 goethite or hematite, 10 2 M NaCl, and a buffer of either 2-(N-morpholino)ethanesulfonic acid (MES) at pH 4 or 4-(2-hydroxyethyl)-1-piperazinepropanesulfonic acid (EPPS) at pH 7 at a concentration of 10 3 M. The As(V) isotherms were measured in the presence of 0, 10 4 M, or 10 3 M FeCl2; the Fe(II) isotherms were determined in the presence of 0 or 5 10 4 M Na2HAsO4. Each data point corresponds to a separate sample; the pH values of the samples in each varied by less than (0.1 pH units from the target pH. All samples were made by dilution of stock solutions prepared in the anaerobic chamber from reagent grade chemicals and deoxygenated deionized water. FeCl2 stock solutions were filtered and acidified to pH 2 using HCl, stored in opaque bottles, and used within 7 days; a test Fe(II) solution showed no signs of oxidation after storage for 60 days in the anaerobic chamber. pH adjustments were made using HCl and NaOH that were stored for at least 90 days in the chamber and had been bubbled following the same procedure as was used for the mineral suspensions. All samples were wrapped in aluminum foil to prevent light exposure and placed on end-overend rotators for the duration of the reaction. At the end of the reaction period the samples were centrifuged in the anaerobic chamber and the supernatant was decanted, filtered (0.2 μm, MCE), and acidified with trace metal grade HNO3. Dissolved As and Fe concentrations were determined by inductively coupled plasma optical emission spectrometry (ICP-OES) using a Perkin-Elmer Optima 7300 DV.
ARTICLE
Aliquots from select samples were collected prior to acidification for analysis of dissolved Fe(II) using the ferrozine method.38 These analyses yielded dissolved Fe(II) concentrations within
