Chromium(III) Oxidation by Three Poorly Crystalline Manganese(IV

Article pubs.acs.org/est

Chromium(III) Oxidation by Three Poorly Crystalline Manganese(IV) Oxides. 2. Solid Phase Analyses Gautier Landrot,*,†,∥ Matthew Ginder-Vogel,† Kenneth Livi,‡ Jeffrey P. Fitts,§ and Donald L. Sparks† †

Plant and Soil Sciences Department, Delaware Environmental Institute, University of Delaware, 152 Townsend Hall, Newark, Delaware 19716, United States ‡ HRAEM Facility, Department of Earth and Planetary Sciences, Johns Hopkins University, Baltimore, Maryland 21218, United States § Department of Civil and Environmental Engineering, Princeton University, Engineering Quad, E430, Princeton, New Jersey 08544, United States S Supporting Information *

ABSTRACT: Layered, poorly crystalline Mn(IV)O2 phases are abundant in the environment. These mineral phases may rapidly oxidize Cr(III) to more mobile and toxic Cr(VI) in soils. There is still, however, little knowledge of how Cr(III) oxidation by Mn(IV)O2 proceeds at the microscopic and molecular levels. Therefore, the sorption mechanisms of Cr(III) and Cr(VI) on Random Stacked Birnessite (RSB), δMnO2, and Acid Birnessite (AB) were determined by Extended X-ray Absorption Fine Structure Spectroscopy (EXAFS). These three synthetic Mn(IV)O2, which are poorly crystalline phases and have layered structures, were reacted with 50 mM Cr(III) at pH 2.5, 3, and 3.5 before being analyzed by EXAFS. The results indicated that Cr(VI) was loosely sorbed as an outer-sphere complex on Mn(IV)O2, while Cr(III) was tightly sorbed as an inner-sphere complex. Further research is needed to understand why Cr(III) stopped being significantly oxidized by Mn(IV)O2 after 30 min. This study, however, demonstrated that the formation of a Cr surface precipitate is not necessarily responsible for the cessation in Cr(III) oxidation. Indeed, no Cr surface precipitate was detected at the microscopic and molecular levels on Mn(IV)O2 surfaces reacted with Cr(III) for 1 h, although the Cr(III) oxidation ceased before 1 h of reaction at most employed experimental conditions.

1. INTRODUCTION Mn-oxides are ubiquitous in the environment.1 A large portion of them are poorly crystalline and/or nanocrystalline.2 Additionally, the layered structure Mn-oxides are more abundant in the environment than those with tunnels.1 However, the various natural and/or synthetic Mn-oxides that were investigated in previous studies3−5 to determine their capacities to oxidize Cr(III) were significantly different from each other in nature and not often abundant in the natural environment. The minerals studied exhibited different degrees of crystallinity and were either 3D tunnel or 2D layer structures. Random Stacked Birnessite, δ-MnO2, and acid birnessite (AB) are poorly crystalline phases with layered structures, similar to many naturally occurring Mn-oxides found in the environment.2,6 The Cr(III)-oxidizing capacities of these three synthetic Mn(IV)O2 were investigated in a companion paper (part 1).7 A 20 g/L suspension of RSB, δ-MnO2, or AB was reacted with 50 mM Cr(III) at pH 2.5, 3, and 3.5, using a batch kinetics method. Although the Cr(III) concentration employed in our study is much higher than those used in previous studies that investigated the kinetics of Cr(III) oxidation by MnO 2 phases, 3−5,8 it is comparable to concentrations found at Cr contaminated sites. For example, Cr was measured at about 500 mM in a soil near a leaking tank at an industrial site in Oregon, USA,9 and 90 mM in a dense © 2012 American Chemical Society

aqueous phase liquid (DAPL) plume at a pH of 3 present in the soil of a former chemical manufacturing site in Massachusetts, USA.10 The results of the batch experiments reported in part 17 indicated that layered poorly crystalline Mn(IV)O2 may strongly oxidize Cr(III) and produce large amounts of Cr(VI). However, the microscopic and molecular mechanisms involved during Cr(III) oxidation by Mn(IV)O2 phases are still poorly understood. These include the processes responsible for the cessation in Cr(III) oxidation by Mn(IV)O2, which has been reported in previous studies8,11 and was shown in part 1 of this investigation.7 Also, there is still little knowledge of the sorption mechanisms of Cr(III) and Cr(VI) on poorly crystalline Mn(IV)O2 at high Cr(III) concentrations in solution. Therefore, these mechanisms were determined in this study on RSB, δ-MnO2, and AB, using Extended X-ray Absorption Fine Structure Spectroscopy (EXAFS). These analyses were performed after separating by filtration the mineral solid phases from aliquots of Cr(III)-reacted Mn(IV)O2 suspension taken at different reaction times. Additionally, the formation and spreading on the Mn(IV)O2 surface of a Cr(OH)3 phase, Received: Revised: Accepted: Published: 11601

June 13, 2012 October 4, 2012 October 11, 2012 October 11, 2012 dx.doi.org/10.1021/es302384q | Environ. Sci. Technol. 2012, 46, 11601−11609

Environmental Science & Technology

Article

Figure 1. (a) Quantity of Cr(VI) in % inferred from normalized XANES spectra of AB, δ-MnO2, and RSB at different reaction times (b) Fourier transforms of AB, δ-MnO2, and RSB unreacted and reacted with Cr(III) at different times and (c) amount of Cr(III) and Cr(VI) sorbed on δ-MnO2 and AB over time at pH 3. The quantities were measured from the percentage of Cr(III) and Cr(VI) sorbed on the surface of AB and δ-MnO2 at 2, 5, 10, and 30 min (Figure 1 a) and the total amount of chromium sorbed at 2, 5, 10, and 30 min on AB at pH 3 and δ-MnO2 at pH 2.5, 3, and 3.5 reported in Figure 4 in part 1 of this study.

with a γ-CrOOH structure similar to lepidocrocite, may inhibit Cr(III) oxidation by impeding electron transfer between Cr(III) and Mn(IV)/Mn(III).11,12 Since the results of the batch kinetics experiments reported in part 1 of this study indicated that the Cr(III) oxidation by δ-MnO2 and AB ceased between 30 min and 1 h at most employed experimental conditions, the presence of a Cr(OH)3 phase on δ-MnO2, RSB, and AB was also investigated in this study at the microscopic and molecular levels, using EXAFS and Transmission Electron Microscopy (TEM).

3.7 g of KCl were introduced in a 400 mL vessel. Therefore, after adding 20 mL of 1 M Cr(III) at t = 0 of the batch kinetic experiment, the vessel contained a 20 g/L manganese oxide suspension with 50 mM KCl and 50 mM Cr(III). The experiments were conducted at pH 2.5, pH 3, and pH 3.5. At these pH values, the hydrated form of Cr(III) in solution is mostly Cr3+ and is not expected to precipitate in bulk solution, since the pH limit at which Cr(III) starts to precipitate is pH 3.84 (see part 17). After Cr(III) was added at t = 0, 5 mL aliquots were taken from the vessel at different time intervals: 2, 5, 10, 20, 30 min, 1 h, and 6 h. Each aliquot was filtered through a 0.22 μm Millipore filter. The filter, with the wet paste of MnO2 deposited on top of it, was mounted on a sample holder for bulk XAFS analysis. The results of these analyses are reported in this part of the study. The concentrations of Cr(III), Cr(VI), and Mnaqueous in the filtrate solution were reported in part 1 of the study.7 Bulk XAFS Analysis of Solid Phase. MnO2 pastes and Cr standards were mounted on Teflon sample holders, sealed with Kapton tape, and analyzed at the Cr K-edge (5989 eV) by bulk X-ray absorption fine structure spectroscopy (XAFS), at beamline X11 A, National Synchrotron Light Source (NSLS),

2. MATERIALS AND METHODS Batch Experiments. All solutions were prepared with distilled dionized water, with a resistivity of 18.2 MΩ, produced from a Barnstead system. Chromium(III) stock solution was made from chromium nitrate Cr(NO3)3 (ACS grade), a few hours before conducting the experiments to minimize the effect of polymerization.13 The methods employed to synthesize δMnO2, RSB, and AB, as well as the batch kinetics method were described in part 1 of this study.7 To summarize the batch experiments, 100 mL of 80 g/L manganese oxide stock suspension (AB, RSB, or δ-MnO2), 280 mL of DI water, and 11602

dx.doi.org/10.1021/es302384q | Environ. Sci. Technol. 2012, 46, 11601−11609

HMO HMO HMO HMO HMO HMO HMO HMO HMO HMO HMO HMO AB AB AB AB RSB RSB

2.5 2.5 2.5 2.5 3 3 3 3 3.5 3.5 3.5 3.5 3 3 3 3 3 3

2m 5m 10 m 30 m 2m 5m 10 m 30 m 2m 5m 10 m 30 m 2m 5m 10 m 30 m 2m 30 m

time 0.2 0.4 0.1 0.3 0.4 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.4 0.5 0.4 0.3 0.3

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

−6.2 ± 3.1 −4.4 ± 4.5 0.7 ± 6.9 0.8 ± 4.1 −1.2 ± 5.1 −1.2 ± 2.5 −0.2 ± 3.3 0.7 ± 2.5 0.4 ± 2.6 0.4 ± 2.1 0.5 ± 2.6 0.3 ± 2.0 −5.8 ± 3.0 −7.2 ± 5.1 −4.5 ± 5.6 −1.3 ± 5.2 1.7 ± 2.5 2.5 ± 2.3 2.9 2.2 2.0 1.8 2.0 1.7 1.6 1.4 1.6 1.4 1.3 1.3 3.3 2.8 2.3 1.7 1.0 0.9

CN

ΔEo 1.60 1.62 1.63 1.63 1.62 1.63 1.62 1.62 1.63 1.63 1.62 1.64 1.61 1.60 1.61 1.62 1.62 1.60

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.01 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03

R (Å) 0.002 0.002 0.003 0.002 0.002 0.003 0.003 0.003 0.003 0.004 0.004 0.006 0.003 0.002 0.002 0.002 0.008 0.007

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

σ 2

0.001 0.001 0.003 0.002 0.002 0.002 0.003 0.003 0.002 0.003 0.004 0.004 0.001 0.001 0.002 0.002 0.007 0.006

1.6 2.7 2.9 3.3 3.0 3.4 3.7 3.9 3.6 4.1 4.1 4.0 1.1 1.9 2.5 3.5 4.6 4.7

CN

a

1.98 1.96 1.98 1.98 1.97 1.97 1.97 1.98 1.98 1.98 1.98 1.98 2.01 1.97 1.96 1.97 1.98 1.99

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.02 0.02 0.03 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.04 0.02 0.02 0.02 0.01 0.01

R (Å)

Cr(III)−O 0.006 0.003 0.003 0.004 0.005 0.002 0.003 0.003 0.003 0.002 0.003 0.002 0.007 0.006 0.005 0.004 0.002 0.002

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

σ 2

0.005 0.003 0.003 0.003 0.004 0.002 0.002 0.001 0.002 0.001 0.001 0.002 0.007 0.006 0.006 0.003 0.001 0.001

8.8 6.6 6.1 5.3 6.0 5.2 4.7 4.2 4.7 4.1 3.8 3.9 9.8 8.3 6.9 5.0 2.9 2.7

CN

b

2.92 2.94 2.95 2.96 2.94 2.97 2.95 2.95 2.95 2.96 2.95 2.97 2.92 2.90 2.93 2.95 2.94 2.90

R (Å)

c

2d

0.004 0.003 0.005 0.004 0.004 0.006 0.005 0.006 0.005 0.008 0.008 0.012 0.006 0.005 0.004 0.004 0.016 0.014

σ

Cr(VI)−O M.S.

0.6 1.4 0.8 0.7 0.5 0.8 0.7 0.6 0.6 0.8 0.6 0.2 0.9 1.0 1.0 0.8

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

1.0 1.6 1.4 1.2 1.2 1.5 1.4 1.5 1.3 1.5 1.2 0.5 1.0 1.4 1.3 0.8

CN

2.87 2.89 2.92 2.98 2.97

2.89 2.92 2.93 2.92 2.92 2.92 2.94 2.93 2.93 2.93 2.94 ± ± ± ± ±

± ± ± ± ± ± ± ± ± ± ± 0.04 0.04 0.04 0.03 0.03

0.03 0.03 0.03 0.03 0.02 0.02 0.02 0.02 0.01 0.02 0.01

R (Å)

Cr−Cr/Cr−Mn

0.003 0.003 0.004 0.008 0.005

0.003 0.004 0.003 0.003 0.002 0.003 0.004 0.003 0.003 0.004 0.003

± ± ± ± ±

± ± ± ± ± ± ± ± ± ± ±

σ2

0.005 0.006 0.005 0.009 0.007

0.004 0.004 0.004 0.004 0.002 0.003 0.004 0.003 0.003 0.004 0.003

CNCrIII−O defined as CNCrIII−O = (l−(CNCrVI−O/4)) × 6. bCNmultiple scattenng defined as CNm.s. = 3 × CNCrVI−O. cRmultiple scattenng defined as Rm.s. = 1.822*RCrVI−O (see text). dσ2multiple scattenng defined as σ2m.s. = 2*σ2CrVI−O.

a

MnO2

pH

Cr(VI)−O

Table 1. Fitting Parameters of Fourier Transforms Depicted in Figure 1(b)

Environmental Science & Technology Article

11603



Article

Since more than 90% of the chromium is Cr(VI) for this sample (Figure 1 a), it is possible to consider that the Fourier transformed EXAFS spectrum of AB at t = 2 min depicted in Figure 1 (b) is only representative of chromate sorbed on AB. This seems reasonable since there is only one Cr(VI)−O shell at 1.61 Å in the Fourier transform; no other-shells are present at longer radial distances. Therefore, Cr(VI) is likely bound to the mineral surface as an outer-sphere complex. Similarly, a previous study showed, using EXAFS, that Cr(VI) binds as an outer sphere complex with birnessite (Na4Mn14O27) and manganite (γ-Mn(III)OOH).3 The second peak of the Fourier Transformed EXAFS of AB reacted with Cr(III) after 5 min at 2.87 ± 0.04 Å (Table 1) increases in amplitude with time (Figure 1 b). One study similarly found a radial distance at 2.90 Å for the second shell of chromium sorbed on sodium birnessite.3 This second coordination sphere was assigned to a Cr(III)−Mn(IV) shell for three reasons. First, a Cr(III)− Cr(III) shell would have resulted in a greater radial distance, at about 3.00 Å. Also, although Cr(III)−Cr(VI) would result in a smaller radial distance than Cr(III)−Cr(III), no significant amount of Cr(VI) was present in the sample. Finally, the radial distance of the first Mn−Mn shell of the unreacted sodium birnessite was also measured at 2.90 Å, which suggested similarities in local structure around Mn and Cr in the mineral.3 Similarly in our study, the second shell of the FT of AB reacted with Cr(III) for 5 min depicted in Figure 1 (b) can be assigned to a Cr(III)−Mn(IV) shell. Indeed, although Cr(VI) is 85% of the chromium sorbed on AB at t = 5 min, Cr(VI) should not contribute to the second shell, assuming that the Cr(VI) sorption mechanism does not change between t = 2 min and the rest of the reaction. Additionally, the radial distance of the first Mn−Mn shell of unreacted AB is 2.90 Å (Figure 1 b), which suggests that Cr(III) is sorbed on the mineral in an inner-sphere complex mechanism, and both Mn(IV) and Cr(III) share the same structural environment. Therefore, although Cr(III) bound to Mn(IV) can rapidly oxidize to Cr(VI) (part 17), it can also bind to Mn(IV) without oxidizing to Cr(VI). The increases in amplitude with time observed in Figure 1b for the second peak in the Fourier Transformed EXAFS of AB reacted with Cr(III) after 5 min at 2.87 ± 0.04 Å (Table 1) is thus attributed to Cr(III) accumulating on the Mn(IV)O2 surface. These results are supported by those reported in Manceau and Charlet,3 who found Cr(III) sorbed in an inner-sphere complex to Mn(IV) in Cr(III)-reacted MnO2 analyzed by EXAFS. Similarities in percentage of Cr(VI) on the surface, Fourier transformed EXAFS, and fitting parameters (Table 1) are illustrated for the experiments involving AB at pH 3 and δMnO2 at pH 2.5 (Figure 1). Therefore, the same conclusions are made: the Fourier transformed EXAFS of reacted δ-MnO2 at pH 2.5 and t = 2 min is mostly representative of Cr(VI) sorbed on the mineral in an outer-sphere complex; at time t = 5 min, Cr(III) sorbs on δ-MnO2 in an inner-sphere mechanism and shares the same structural environment as Mn atoms in the unreacted δ-MnO2, since both reacted and unreacted δ-MnO2 possess a first metal−metal shell at ≈2.90 Å (Figure 1 b). It has been reported that Cr(III) precipitates in a Cr(OH)3 phase with a γ-CrOOH structure on the surface of several manganese oxides.11,12,19 This precipitate is characterized by two Cr(III)−Cr(III) shells at 3.00 to 3.05 and 3.94 to 4.03 Å and is similar to the phase forming when Cr precipitates in bulk solution.3 If most of the Cr(III) sorbed to MnO2 precipitated in a Cr(OH)3 phase on the surface of AB, δ-MnO2, and RSB, one

Brookhaven National Laboratory, Upton, NY. The 300 mA synchrotron beam was detuned from a Si(111) monochromator in I0 by 30% to reject higher harmonics. The gas mixture in Io was 70% helium, 30% nitrogen, and the fluorescence signal was collected with a Lytle cell detector at room temperature. A vanadium filter was used to remove elastic radiation from the fluorescence emissions. The sample holder was oriented at 45° to the incident beam. Three spectra were collected per sample and averaged for data analysis. Spectra were calibrated at the chromium pre-edge feature at 5993.5 eV. Data were analyzed with the SIXPACK/IFEFFIT program.14 The concentration of Cr(VI) in the samples was measured from the pre-edge feature at 5993.5 eV from the normalized XANES spectra, using a set of normalized XANES spectra of mixed Cr(III)/Cr(VI) standards.15,16 Fourier transformed extended X-ray absorption fine structure (EXAFS) spectra that were k3 weighted were used for shell-by-shell fitting of the data. Model compounds used in FEFF6l calculations were Eskolaite (Cr2O3) for the Cr(III)−O path and Tarapacite (K2CrO4) for the Cr(VI)−O path. Eskolaite was also used to generate the first Cr−metal (i.e., Cr−Cr or Cr−Mn) shell path, knowing that an imaginary Eskolaite mineral with manganese atoms substituted for chromium gave the same fitting results but with a slightly higher χ2 value. The coordination number of Cr(III)−O in our models was constrained based on the coordination number of Cr(VI)−O and was defined as CNCr(III)−O = (1−(CNCr(VI)−O /4)) × 6. The contribution of the triangular multiple scattering Cr−O−O−Cr in the Cr(VI)O 4 tetrahedron was also considered in our fits.17 All the fitting parameters of this multiple scattering, i.e. the multiple scattering coordination number for (CNm.s.), the multiple scattering Debye−Waller factor (σ2m.s.), and the multiple scattering radial distance (Rm.s.) were constrained based on Cr(VI)−O fitting parameters and respectively defined in our models as CNm.s.= 3*CNCrVI−O, σ2m.s. = 2*σ2Cr(VI)−O and Rm.s.= (Rreffm.s/ RreffCr(VI)−O)*RCr(VI)−O = (2.98/1.64)*RCr(VI)−O. TEM Analysis of Solid Phase. About three milligrams of dried manganese oxide reacted with Cr(III) for 1 h during the batch experiments performed at pH 3.5 were dispersed in deionized water and briefly ultrasonified. A small amount of the solution was deposited on a 200-mesh Cu grid and dried. The samples were analyzed with a Philips CM300 FEG transmission electron microscope equipped with a Gatan GIF 200 CCD imaging system running at 300 kV and a Philips EM420 microscope running at 120 kV. Both scopes were equipped with an Oxford light element Energy Dispersive X-ray Spectroscopy (EDS) detector. Data analysis was carried out with Gatan Digital Micrograph 1.8 software for the image processing, and the EDS data were analyzed with the software package ES Vision4. Selected Area Electron Diffraction (SAED) data were processed with DiffTools scripts developed by David Mitchell.18

3. RESULTS AND DISCUSSION Analyses of Cr(III)-Reacted MnO2 Surface at the Molecular Level. Bulk XAFS data shown in Figure 1 and associated EXAFS spectra shown in S.I. were derived from wet pastes of Cr(III)-reacted MnO2, and thus they indicate Cr sorbed on the mineral and not Cr in solution. The percentage of Cr(VI) of the total Cr associated with the δ-MnO2, AB, and RSB following reaction with aqueous Cr(III) at different times is depicted in Figure 1 (a). Most of the chromium sorbed on AB at pH 3 at the beginning of the reaction (2 min) is Cr(VI). 11604



Article

would see a peak at around 4 Å in the Fourier transformed EXAFS depicted in Figure 1 (b), similar to the XAFS data of the Cr(III)-reacted MnO2 reported in Manceau and Charlet.3 None of the data in this study features a peak at such radial distance (Table 1). However, one peak is present in all spectra at ≈3.55 Å, and the amplitude does not seem to vary significantly during the kinetic experiments performed with RSB, AB, and δ-MnO2. This peak was not considered in our fits since it is also present in the Fourier transformed EXAFS of the aqueous Cr(III) standard (Figure 1 b) and could be attributed to either Cr(III) polymers present in solution that sorbed on the minerals13 or Cr(III) linear multiple scatterings. Therefore, the results depicted in Figure 1 (b) suggest that even if a Cr(OH)3 phase formed on the MnO2 surface, it was not in sufficient amount compared to the rest of the sorbed Cr(III) to be detected by the EXAFS technique on the surface of δ-MnO2, AB, and RSB at any reaction time. This is because no peak was observed around 4.00 Å, even when 27 mM of Cr(III) is sorbed on the δ-MnO2 surface at 30 min and pH 3.5 (Figure 1 c). Analyses of Cr(III)-Reacted MnO2 Surface at the Microscopic Level. No major change in morphology or sign of precipitate formation is seen in High Resolution TEM (HRTEM) images of Cr(III)-reacted AB and δ-MnO 2 compared to the unreacted AB and δ-MnO2 (Figure 2). The analyses of reacted and unreacted δ-MnO2 HRTEM are more difficult to interpret than AB since the images reveal a very heterogeneous mineral structure, mostly amorphous (images F in Figure 2), which also features some crystalline regions locally distributed (image G in Figure 2). Also, EELS and EDS measurements on δ-MnO2 are difficult since the mineral is rapidly damaged at the spot exposed to the electron beam.6 The main oxidation state of chromium sorbed to AB is (III), according to the peak position of Cr in EELS patterns, such as the one shown in Figure 3 (b), and peak positions of several chromium oxidation states reported in a previous study.20 Additionally, Cr(III) seems to be more partitioned toward the edge of AB agglomerates, according to EDS data taken at various locations in the Cr(III)-reacted mineral. Figure 3 (c) depicts the EDS patterns taken in the middle of a AB flake randomly chosen and in the rim of the same flake and shows that chromium accumulates on the edge much more than in the center of the flake, according to the height of the Cr peak. Figure 3 (a) shows that the SAED pattern of δ-MnO2 reacted with Cr(III) is similar to the SAED pattern of the unreacted phase, which is probably due to both reacted and unreacted δMnO2 being very poorly crystalline. Likewise, the SAED pattern of AB reacted with Cr(III) is similar to the SAED pattern of the unreacted phase. Therefore, a crystalline Cr(III)(OH)3 precipitate is not detected in the Cr-reacted AB, using SAED. Discussion. To our knowledge, Fendorf et al.12 is the only study that has detected by SAED this Cr precipitate on a manganese oxide surface. In Fendorf et al.,12 a dramatic difference in degree of crystallinity could be observed between the SAED pattern of Cr(III)-reacted AB, which exhibited sharp ring patterns and indicated after crystallographic analyses the presence of a crystalline Cr(III)(OH)3 phase with γ-CrOOH structure, and the SAED pattern of unreacted AB reacted, which exhibited diffuse ring patterns due to the poorly crystalline nature of AB. The latter is similar to those reported in Figure 3 (a), corresponding to unreacted AB and Cr(III)reacted AB. Therefore, the results from Fendorf et al.12 were different from those reported in this study, although the same

Figure 2. HRTEM of unreacted AB (a, b) and δ-MnO2 (f, g), as well as AB (c, d, e) and δ-MnO2 (h, i, j) reacted for 1 h with 50 mM of Cr(III) at pH 3.5.

type of Mn(IV)O2 was studied (i.e., acid birnessite). Similarly, the EXAFS data reported in Figure 1 (b) and Table 1 do not indicate that the Cr(III) sorbed to the three MnO2 phases is mainly present as a Cr(III)(OH)3 phase, which contrasts with the findings reported in Manceau and Charlet.3 In the latter study, a Cr(III)(OH)3 phase with a γ-CrOOH structure was systematically detected by EXAFS analyses on the surface of several manganese oxide phases reacted with Cr(III), i.e. Na4Mn14O27 birnessite, MnO2 ramsdellite, β-MnOOH feitknechtite, and γ-Mn2O3. Lastly, a Cr hydroxide precipitate was also detected by Tapping Mode Atomic Force Microscopy (TM-AFM) and X-ray Photoelectron Spectroscopy (XPS), on 11605



Article

Figure 3. (a) SAED patterns of unreacted δ-MnO2 and AB, and 1 h Cr(III)-reacted AB and δ-MnO2, at pH 2.5 and 3.5, and (b) EELS pattern of 1 h Cr(III)-reacted AB and EELS energy position of Cr(VI) and Cr(III) L3 peaks according to Daulton and Little,20 and (c) EDS patterns taken in the interior (pattern in gray) and on the rim (pattern in black) of an acid birnessite flake.

the surface of γ-Mn(III)OOH reacted with 0.1 mM Cr(III) at pH 3.5.21 The authors of Weaver et al.21 described the Cr(OH)3 precipitate formed on MnOOH as a crystalline phase, as opposed to the amorphous Cr(OH)3 phase that can form in bulk solution based on the solubility data of Cr reported in Rai et al.22 The experimental conditions in our study, as well as those in Fendorf et al.,12 Manceau and Charlet,3 and Weaver et al.21, were used in the chemical equilibrium program and associated thermodynamic database MINEQL+ (version 4.6), to calculate for each study the ionic strength (I.S.), the saturation index (S.I.) of Cr(OH)3 precipitate in solution and the forms of aqueous Cr(III), i.e. the percentages of Cr3+ and Cr(OH)2+ in solution. Since the respective S.I. values of the four studies are negative, the bulk solution is undersaturated with respect to Cr(OH)3 in all studies (Table 2). Therefore, the Cr(OH)3 phases reported in Fendorf et al.,12 Manceau and Charlet,3 and Weaver et al.21 were indeed surface precipitates and not Cr(OH)3 formed in solution. The least negative S.I. values were the ones in this study at pH 3.5 and Fendorf et al.12 (i.e., S.I. = −1, Table 2) and are thus the closest values to the point where

Cr(OH)3 starts to precipitate in bulk solution (i.e., the case where S.I. > 0). Therefore, the differences in S.I. values cannot explain why a Cr(III) surface precipitate was reported in the three former studies but was not detected by EXAFS, HRTEM, and SAED in this study. The possibility that the poorly crystalline nature of AB, RSB, and δ-MnO2 prevented the formation of a Cr surface precipitate should be ruled out as well. Indeed, Fendorf et al.12 reported the presence of a wellcrystalline Cr(OH)3 phase on the surface of AB, which is one of the three MnO2 phases investigated in this study. For each study considered in Table 2, the initial concentration of Cr(III) introduced in the system, in mol/L, was divided by the initial surface area of the MnO2 phase, in m2/g. The result was then multiplied by the suspension density of the MnO2 phase in solution, in g/L. These [Cr(III)]normalized values, which are reported in mol/m2 in Table 2, can be compared to each other if one assumes that all Mn phases have the same reactive surface site density. The lowest [Cr(III)]normalized value in Table 2 indicates the conditions at which a Cr(OH)3 surface precipitate is the most unlikely to form, because Cr surface precipitate formation is rather promoted at high Cr loading on 11606



Article

Table 2. Experimental Conditions Employed in This Study, Fendorf et al.,12 Manceau and Charlet,3 and Weaver et al.,21 and Percentages of Cr3+ and Cr(OH)2+ in Solution, Ionic Strength (I.S.), and the Saturation Index (S.I.) of Cr(OH)3 Precipitate Calculated by MINEQL+ (Version 4.6) for Each Study MINEQL+ outputs

pH this study

2.5

this study

3

this study

3.5

Fendorf et al., 1992 Manceau and Charlet, 1992

4 4 4 4 4

Weaver et al., 2002

3.5

[Cr(III)] normalized (mol.m−2)

background electrolyte

initial [Cr(III)]

Cr3+

Cr(OH)2+

I.S.

S.I.

20 g/L

0.030, 0.032, 0.009

50 mM KCI

50 mM

98%

2%

0.35 M

−4.0

84, 78, 272

20 g/L

0.030, 0.032, 0.009

50 mM KCI

50 mM

96%

4%

0.34 M

−2.5

84, 78, 272

20 g/L

0.030, 0.032, 0.009

50 mM KCI

50 mM

87%

11%

0.33 M

−1.0

223

20 mg/L

0.224

1 mM

60%

40%

0.11 M

−1.0

Na4Mn14O27 birnessite, MnO2 ramsdellite, (β-MnOOH feitknechtite, γ-Mn2O3

27

5 g/L

0.014

0.3 mM

70%

30%

0.60 M

−1.9

1

8 g/L

0.238

1.5 mM

70%

30%

0.61 M

−1.2

109

8 g/L

0.002

1.9 mM

70%

30%

0.61 M

−1.1

81

8 g/L

0.003

0.6 mM

70%

30%

0.60 M

−1.6

γ-Mn(III)OOH Manganite

9

450 mg/L

0.025

100 mM NaNO3 600 mM NaCI 600 mM NaCI 600 mM NaCI 600 mM NaCI 5 mM NaNO3

0.1 mM

68%

32%

6 mM

−2.8

Mn mineral phase AB, RSB, δMnO2 AB, RSB, δMnO2 AB, RSB, δMnO2 AB

surface area (m2/g)

suspension density

84, 78, 272

MnO2 surfaces.19,23 The two lowest [Cr(III)]normalized values are found for the experiments conducted with γ-Mn2O3 and βMnOOH in Manceau and Charlet3 (Table 2). Therefore, the differences in [Cr(III)]normalized values cannot explain either why a Cr(OH)3 surface precipitate was detected on all MnO2 phases studied by Manceau and Charlet 3 and Fendorf et al.12 as well as Weaver et al.21 and not on the three MnO2 phases investigated in our study. The Cr(III) species in solution (e.g., Cr(H2O)63+, Cr(OH) •(H2O)52+, and Cr(OH)2•(H2O)4+) nucleating on the mineral surfaces during surface precipitation are polyhedral species that have their central Cr atom mostly bound to H2O ligands.3 In contrast, all the Cr(III) atoms in a Cr(OH)3 surface precipitate are bound to each other via OH groups.11 Therefore, ligand exchange mechanisms occur during the polymerization process of Cr(III) on the MnO2 surface. Espenson24 studied the rate constants of ligand exchange reactions in solution involving Cr(OH)2+ or Cr3+ with various anionic and neutral ligands. He reported that the rate constants associated with Cr(OH)2+ were, depending on the entering ligand, 3 or 4 orders of magnitude higher than those associated with Cr3+. These results were supported by three models predicting that the OH− ligand in Cr(OH)2+ was extremely efficient in promoting the substitution of inner sphere water molecules by other ligands.25 Xu et al.26 experimentally measured the water-ligand exchange kinetics in Cr(OH)2+ and found them to be 75 times faster than the ones for Cr3+ in solution. Additionally, Rai et al.22 showed that the thermodynamic equilibrium constant associated with the precipitation of Cr(OH)3(s) in solution from Cr(OH)2+ was at least 3 orders of magnitude greater than the one associated with the precipitation of Cr(OH)3(s) from Cr3+. Lastly, a surface complex formation model predicted that Cr(OH)2+ sorbs on the surface of aluminum oxide about 50 times faster than Cr3+.27 Therefore, although equilibrium constants for precipitation of Cr(III) species in solution to Cr(OH)3(s) on the surfaces of MnO2 phases have not been reported in the literature yet, one could reasonably assume that Cr(OH)2+ precipitates more favorably than Cr3+ on MnO2 surfaces. The percentages of Cr(OH)2+ relative to the total

amount of Cr(III) present in solution in Manceau and Charlet,3 Fendorf et al.,12 and Weaver et al.21 are between 3 to 20 times higher than those in our study (Table 2). Therefore, this may explain why a Cr(OH)3 surface precipitate was detected on all MnO2 phases investigated by the three former studies but not on those analyzed in our investigation. If indeed Cr(OH)2+ precipitates to Cr(OH)3(s) more readily than Cr3+ on the MnO2 surface, the surface precipitation of a given amount of Cr(III) sorbed on MnO2 should preferentially occur with the highest Cr(OH)2+/Cr3+ concentration ratio. The low Cr(OH)2+/Cr3+ ratio in our study could imply that even if a Cr(OH)3 surface precipitate formed on the three MnO2 surfaces, it may have been too diffuse to be detected by EXAFS and SAED, despite the high initial Cr(III) concentration introduced in the system. The background electrolyte concentration and initial Cr(III) concentration, however, constrain the initial amounts of Cr(OH)2+ and Cr3+ in solution, since the activity coefficients of these two species, featured in their respective activity expression, are dependent on the ionic strength of the system. Therefore, the initial Cr(III) and background electrolyte concentrations may also be factors that can explain why a Cr surface precipitate was detected in the three former studies considered in Table 2 and not in our investigation. Since the Cr(OH)3 surface precipitate may cover the mineral surface and thus impede Cr(III) in solution to sorb and oxidize, this process has been believed to be the cause of the cessation in Cr(III) oxidation by MnO 2 observed in previous studies.8,11,12 The results reported in part 1 indicated that Cr(III) oxidation by AB and δ-MnO2 ceased between 30 min and 1 h at most experimental conditions.7 However, since no Cr(OH)3 surface precipitate was observed on the surfaces of the Cr(III)-reacted MnO2 phases, our results suggested that the cessation of the Cr(III) oxidation reaction is not necessarily due to the presence of a Cr(III) surface precipitate. The causes of the shutdown in Cr(III) oxidation by MnO2 are still unknown and cannot be determined with the results reported in this study. However, the cessation in Cr(III) oxidation was not due to a limitation in Mn(IV) atoms in the system. The Cr(III) 11607



Article

oxidation by Mn(IV)O2 phases that involves Cr3+ (i.e., the main Cr(III) species in our system, Table 2) consumes Mn(IV) and Cr3+ with a 1.5 Mn(IV)/Cr(III) mole ratio.8 The initial Mn(IV)/Cr(III) mole ratio in our system was 4.6. Additionally, although Cr(III) was the limiting reactant in the system, the results reported in part 17 indicated that Cr(III) oxidation was not ceased due to a total consumption of the Cr(III) atoms, because some Cr(III) ions were still present in solution at all experimental conditions, even after 6 h. Environmental Implications. The results reported in part 1 suggested that the three synthetic Mn(IV)O2 phases, which are similar to the layered poorly crystalline Mn oxide phases commonly found in the environment, may be capable of oxidizing large amounts of Cr(III).7 This is all the more problematic since the results of this study indicated that Cr(VI) is weakly sorbed to MnO2 as an outer sphere complex and thus may be able to easily desorb to solution. The extent of Cr(III) oxidation can be, however, minimized by several processes.21 The amount of Cr(VI) can be reduced to Cr(III) by Fe oxide phases, sulfides, and organic matter in soils.7,28,29 Alternatively, chemical remediation agents can be used to reduce chromate in Cr contaminated soils, such as zerovalent iron (Fe(0)),30 sodium dithionite (Na2S2O4),31 or ferrous ammonium sulfate ((NH4)2Fe(SO4)2•6H2O).32 The formation of a Cr(III)(OH)3 surface precipitate on manganese oxides may also minimize the extent of Cr(III) oxidation,3,11,12,21 although the results of this study suggested that Cr(III) precipitation does not systematically occur on Mn(IV)O2 surfaces. The results also showed that Cr(III) binds as an inner sphere complex on Mn(IV)O2 and can accumulate without oxidizing to Cr(VI). This Cr(III) accumulation on Mn(IV)O2 preferentially occurs at high pH (Figure 1 c), similarly to Cr(III) precipitation to a Cr(OH)3 phase in bulk solution22 and on Mn(IV)O2 surfaces.11 Additionally, high pH values imply amounts of Cr(VI) sorbed on Mn(IV)O2 (Figure 1 c) and Cr(III) oxidized on the mineral surface lower than those at low pH values.7,11,16 Therefore, if acidic soils at Cr contaminated sites mainly contain Cr(III), increasing the soil pH may be helpful in minimizing Cr(III) mobility and the risk of oxidizing it to chromate by Mn(IV)O2.

■

Chemical Sciences, under contract number DE-AC0298CH10886. The beamline X11 is supported by the Office of Naval Research and contributions from Participating Research Team (PRT) members.

■

(1) Post, J. E. Manganese oxide minerals: crytal stuctures and economic and environmental significance. Proc. Natl. Acad. Sci. U.S.A. 1999, 3447−3454. (2) Zhu, M.; Farrow, C. L.; Post, J. E.; Livi, K. J. T.; Billinge, S. J. L.; Ginder-Vogel, M.; Sparks, D. Structural study of biotic and abiotic poorly-crystalline manganese oxides using atomic pair distribution function analysis. Geochim. Cosmochim. Acta 2012, 81, 39−55. (3) Manceau, A.; Charlet, L. X-ray adsorption spectroscopy study of the sorption of Cr(III) at the oxide-water interface. I. Molecular mechanism of Cr(III) oxidation on Mn Oxides. J. Colloid Interface Sci. 1992, 148, 425−442. (4) Kim, J., G.; Dixon, J.; Chusui, C., C.; Deng, Y. Oxidation of chromium(III) to (VI) by manganese oxides. Soil Sci. Soc. Am. J. 2002, 66, 306−316. (5) Weaver, R. M.; Hochella, M. F. The reactivity of seven MnOxides with Cr3+aq: A comparative analysis of a omplex, environmentally important redox reaction. Am. Mineral. 2003, 88, 2016−2028. (6) Livi, K. J. T.; Lafferty, B.; Zhu, M.; Shouliang, Z.; Gaillot, A.-C.; Sparks, D. Electron energy-loss safe-dose limits for manganese valence measurements in environmentally relevant manganese oxides. Environ. Sci. Technol. 2012, 46, 970−976. (7) Landrot, G.; Ginder-Vogel, M.; Livi, K. J. T.; Fitts, J. P.; Sparks, D. L. Chromium(III) oxidation by three poorly-crystalline manganese(IV) oxides 1. Chromium (III)-oxidizing capacity. Environ. Sci. Technol. 2012, DOI: 10.1021/es302383y. (8) Fendorf, S. E.; Zasoski, R. J. Chromium(III) oxidation by δMnO2 1. Characterization. Environ. Sci. Technol. 1992, 26, 79−85. (9) Palmer, C. D.; Wittbrodt, P. R. Processes affecting the remediation of chromium-contaminated sites. Environ. Health Perspect. 1991, 92, 25−40. (10) Eary, E. L.; Davis, A. Geochemistry of an acidic chromium sulfate plume. Appl. Geochem. 2007, 22, 357−369. (11) Fendorf, S. E. Oxidation and sorption mechanisms of hydrolysable metal ions on oxides surfaces. PhD Dissertation, University of Delaware, Newark, DE, 1992. (12) Fendorf, S. E.; Fendorf, M.; Sparks, D.; Gronsky, R. Inhibitory mechanisms of Cr(III) oxidation by δ-MnO2. J. Colloid Interface Sci. 1992, 153 (1), 37−54. (13) Rotzinger, F.; Stünzi, H.; Marty, W. Early stages of the hydrolysis of chromium(III) in aqueous solution. 3. Kinetics of dimerization of the deprotonated aqua ion. Inorg. Chem. 1986, 25, 489−495. (14) Ginder-Vogel, M.; Landrot, G.; Fischel, J.; Sparks, D. L. Quantification of rapid environmental redox processes using quick scanning x-ray absorption spectroscopy (Q-XAS). Proc. Natl. Acad. Sci. U.S.A. 2009, 106 (38), 16124−16128. (15) Peterson, M. L.; Brown, G. E.; Parks, G. A.; Stein, C. L. Differential redox and sorption of Cr(III/VI) on natural silicate and oxide minerals: EXAFS and XANES results. Geochim. Cosmochim. Acta 1997, 61, 3399−4413. (16) Landrot, G.; Ginder-Vogel, M.; Sparks, D. Kinetics of chromium(III) oxidation by manganese(IV) oxides using quick scanning x-ray absorption fine structure spectroscopy (Q-XAFS). Environ. Sci. Technol. 2010, 44, 143−149. (17) Peterson, M. L.; Brown, G. E.; Parks, G. A. Direct XAFS evidence for heterogeneous redox reaction at the aqueous chromium/ magnetite interface. Colloids Surf., A 1996, 107, 77−88. (18) Mitchell, D. R. G. Difftools: software tools for electron diffraction in DigitalMicrograph. Microsc. Res. Tech. 2008, 71, 588− 593. (19) Charlet, L.; Manceau, A. X-ray adsorption spectroscopy study of the sorption of Cr(III) at the oxide-water Interface. II. Adsorption,

ASSOCIATED CONTENT

S Supporting Information *

EXAFS spectra (k3 chi(k)) of AB, δ-MnO2, and RSB reacted with Cr(III) at different times. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*Phone: +66 90 79 77 922. Fax: +66 29 42 81 06. E-mail: [email protected]. Present Address

∥ Department of Environmental Engineering, Kasetsart University, 10th floor, Building 14, 50 Ngamwongwan Road, Jatujak, Bangkok, 10900, Thailand.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank Gerald Hendricks and Caroline Golt, University of Delaware, for laboratory assistance. The National Synchrotron Light Source is supported by the US Department of Energy, Division of Material Sciences and Division of 11608



Article

coprecipitation, and surface precipitation on hydrous ferric interface. J. Colloid Interface Sci. 1992, 148 (2), 443−468. (20) Daulton, T. L.; Little, B. Determination of chromium valence over the range Cr(0)-Cr(VI) by electron energy loss spectroscopy. Ultramicroscopy 2006, 106, 561−573. (21) Weaver, R. M.; Hochella, M. F.; Ilton, E. S. Dynamic processes occuring at the CrIIIaq -manganite (γ-MnOOH) interface: Simultaneous adsorption, microprecipitation, oxidation/reduction, and dissolution. Geochim. Cosmochim. Acta 2002, 66 (23), 4119−4132. (22) Rai, D.; Sass, B. M.; Moore, D. A. Chromium(III) hydrolysis constants and solubility of chromium(III) hydroxide. Inorg. Chem. 1987, 26, 345−349. (23) Fendorf, S., E.; Lamble, G. M.; Stapleton, M. G.; Kelley, M. J.; Sparks, D. L. Mechanisms of chromium(III) sorption on silica. 1. Cr(III) surface structure derived by extended X-ray adsorption fine structure spectroscopy. Environ. Sci. Technol. 1994, 28, 284−289. (24) Espenson, J. H. Formation rates of monosubstituted chromium(III) complexes in aqueous solution. Inorg. Chem. 1969, 8 (7), 1554− 1556. (25) Gray, B.; Matijević, E. Adsorption and desorption of hydrolyzed metal ions. III. Scandium and chromium. Colloids Surf. 1987, 23 (4), 313−343. (26) Xu, F.-C.; Krouse, H. R.; Swaddle, T. W. Conjugate base pathway for water exchange on aqueous chromium(III): variablepressure and -temperature kinetic study. Inorg. Chem. 1985, 24 (3), 267−270. (27) Wehrli, B.; Ibric, S.; Stumm, W. Adsorption kinetics of vanadyl (IV) and chromium (III) to aluminum oxide: Evidence for a two-step mechanism. Colloids Surf. 1990, 51, 77−88. (28) Fendorf, S. E. Surface reactions of chromium in soils and waters. Geoderma 1995, 67, 55−71. (29) Kimbrough, D. E.; Cohen, Y.; Winer, A. M.; Creelman, L.; Manubi, C. A critical assessment of chromium in the environment. Crit. Rev. Environ. Sci. Technol. 1999, 29 (1), 1−46. (30) Kjeldsen, P.; Locht, T. Removal of chromate in a permeable reactive barrier using zero-valent iron. In Groundwater Quality 2001: Natural and Enhanced Restoration of Groundwater Pollution. Selected papers; Thornton, S., Oswald, S., Eds.; International Association of Hydrological Sciences: Oxfordshire, UK, 2002. (31) Khan, F. A.; Puls, R. W. In-situ abiotic detoxification and immobilization of hexavalent chromium. Ground Water Monit. Rem. 2003, 23 (1), 77−84. (32) Jacobs, J. H. Treatment and stabilization of a hexavalent chromium containing waste material. Environ. Prog. 1992, 11 (2), 123−126.

11609


Chromium(III) Oxidation by Three Poorly Crystalline Manganese(IV

Recommend Documents