Soil Science Society of America Journal 65:736-743 (2001)
© 2001 Soil Science Society of America
DIVISION S-2 - SOIL CHEMISTRY
Sensitivity of Soil Manganese Oxides
Drying and Storage Cause Reduction
Donald S. Rossa,
Heidi C. Halesa,
Grace C. Shea-McCarthyb and
Antonio Lanzirottib
a Dep. of Plant and Soil Sciences, Hills Bldg., Univ. of Vermont, Burlington, VT 05405-0082
b Univ. of Chicago/CARS, National Synchrotron Light Source, Brookhaven National Lab., Upton, NY 11973-5000
Corresponding author (dross{at}zoo.uvm.edu)
 |
ABSTRACT
|
|---|
Effects of sample treatment must be well understood to avoid artifacts during analysis. The effect of drying and storage was examined on nine medium- to high-Mn aerobic soils using extractable Mn(II), the Cr Oxidation Test, estimated soluble C, and XANES spectroscopy. Long-term storage (430 d) at 3°C had little effect on the Cr test. Air drying at room temperature (25°C ± 3°C) caused a drop in the Cr test within 24 h, with a further decline to as low as 45% of the original after 72 h, and less than 2% after 264 d. Extractability of Mn with pH 4.8 NH4OAc increased nearly linearly over the same time period from 0.2 mmol kg-1 to as high as 2.3 mmol kg-1. Increases in the absorbance of the extract at 360 nm, an estimate of soluble C, were well correlated within each soil with the increase in Mn(II). Pre-treatments to remove soluble organics did not cause any increases in the Cr test of dried samples. Therefore, the loss of Cr oxidizing ability appears to be due to reduction of the oxides, not because of increased reduction of the Cr(VI) formed. No changes in XANES spectra were found after short-term air drying at room temperature, but in the three samples examined after 428 d of drying, the main-edge position had a downward shift of about 1.5 eV, indicating reduction. These results confirm previous findings that studies on the reactivity of soil Mn oxides need to avoid sample drying.
Abbreviations: Cr test, standard Cr net oxidation test
|
INTRODUCTION
|
|---|
SOIL MANGANESE OXIDE SURFACES are important reactive sites that may control the behavior of many metals and organics. Mn oxides are particularly important in affecting the movement of contaminant toxic elements such as Cr, Co, Ni, Pu, and As (Bartlett and James, 1979; Amacher and Baker, 1982; Fendorf and Zasoski, 1992). Additionally, a number of investigations have shown that the oxidation of polyphenols, natural and contaminant, takes place on Mn oxide surfaces (Shindo and Huang, 1982; McBride, 1987; Ulrich and Stone, 1989). Understanding the behavior of Mn oxides in the soil environment is important, but previous work has shown that sample handling, especially drying, may have dramatic effects on Mn behavior (Bartlett and James, 1980). Thus it is essential to first understand the effect of analytical techniques, including sample storage, on the oxides.
Heavy-metal accumulation occurs both in soil Mn oxides (Taylor and Nealson, 1966; Jenne, 1968; McKenzie, 1989) and in marine Mn oxides (Burns and Burns, 1977). Cobalt probably accumulates because of oxidation of adsorbed Co(II) to a less mobile Co(III) (Murray and Dillard, 1979). The mechanism for oxidation of Co(II) by buserite has been examined in detail (Manceau et al., 1997). The mechanism for the "specific" adsorption of other metals, such as Pb and Ni, is not completely clear, but is assumed to be related to similarities in electronic structure and substitution. Recent work has shown strong association of 239Pu on Mn oxides in zeolitic tuff (Duff et al., 1998). In addition to Co, oxidation of Cr(III) (Bartlett and James, 1979), As(III) (Oscarson et al., 1981), and Pu (III/IV) (Amacher and Baker, 1982) has been observed. In the case of Cr and Pu, oxidation results in more mobile and potentially toxic species.
Chromium oxidation has received much attention because of the prevalence of Cr at hazardous waste sites. Fendorf and Zasoski (1992) and Charlet and Manceau (1992) both characterized Cr oxidation on synthetic Mn oxides. The accepted mechanism is adsorption of the Cr(III) cation on the Mn oxide surface followed by electron transfer to Mn via oxygen bridges. The Cr(VI) forms an anion that is released from the surface. This mechanism is consistent with earlier findings that adsorption of added Mn(II) on soil Mn oxides initially blocks Cr oxidation (Ross and Bartlett, 1981). Subsequent increases in Cr oxidation correlated with decreases in the extractability of the added Mn(II), presumably because of auto-oxidation.
The oxidation of phenols and polynuclear aromatics (McBride, 1987; Ulrich and Stone, 1989; Whelan, 1995) by Mn oxides appears to follow a similar mechanism of adsorption followed by electron transfer. The resulting products often polymerize, suggesting a role in the creation of stable humic compounds (Shindo and Huang, 1982; Bartlett, 1990), although some evidence suggests that Mn oxides also lyse humic substances to produce low molecular weight organics (Sunda and Kieber, 1994). Naidja et al. (1998) found that the oxidative polymerization reaction between catechol and synthetic birnessite produced an accumulation of reaction products on the oxide surface.
It has long been known that Mn oxides in soils undergo redox changes in response to sample handling and environmental change (Fujimoto and Sherman, 1945). For example, plant deficiencies or toxicities have been observed in response to wetting and drying cycles as redox conditions change and more or less Mn(II) is available for uptake (Heintze, 1946). Drying, heating, autoclaving, freeze-drying, and chemical sterilization all result in an increase in soluble Mn(II), presumably through reduction (Bartlett and James, 1980). Reduction caused by drying may be difficult to rationalize, especially if one is used to working with flooded, anaerobic systems in which drying reintroduces oxygen. In aerobic soil, drying is not necessarily an oxidative process and reduction of Mn oxides cannot be presumptively ruled out.
Drying and storage have a dramatic effect on a soil's ability to oxidize added Cr(III) (the Standard Cr Net Oxidation Test), as first shown by Bartlett and James (1979). Field-moist soil samples oxidized added Cr proportionally to the easily reducible Mn content, whereas little or no Cr was oxidized by samples that had been air-dried and stored at room temperature (lab dirt). The Standard Cr Net Oxidation Test (Cr test) is a net test in that some or all of the Cr(VI) produced may be reduced during the 15-min duration of the test (Bartlett and James, 1996). The decrease in the Cr test for dried soils might be due to an increase in easily oxidizable organics that are formed through desiccation, to a blockage of the reactive sites on the oxide surfaces, to a change in the oxide structure, or to reduction of the oxides themselves.
The purpose of this study is to further examine the effect of drying and storage on the oxidation status of soil Mn oxides using chemical tests and x-ray absorption near edge structure (XANES) spectroscopy. The Mn XANES spectra provide insights into the structure of soil Mn oxides in situ and the technique is valuable for examining changes in stored, dry samples.
|
MATERIALS AND METHODS
|
|---|
Soils were taken on old thrust fault ridges that lie north-to-south in Vermont's Champlain Valley, over a distance of 110 km. Sampling was done at locations known to contain high-Mn soils and at sites with similar soil types. The soils developed on high-pH basal till or bedrock and were all Eutrochrepts (Table 1). Samples of the A horizon were put through a 4-mm polyethylene sieve and stored at 4°C in double 10-µm-thick (4-mil) polyethylene bags with a moist paper towel placed between the layers. The Supersoil and the Subsupersoil samples were taken from a hillside pasture where previous samples had shown extremely high Cr test results (Bartlett and James, 1979). Bedrock was silicious shale. Bedrock in the other lithic soils was quartz sandstone in the Lordstown and fossiliferous limestone in the Hardhack.
Soil pH was measured in 0.01 M CaCl2 (2:1, v/v) and total C determined with an elemental analyzer. Total reducible Mn was determined using a modification of the hydroxylamine hydrochloride method reported by Gambrell (1996). Between 40 and 110 mg of ground sample (through 0.125-mm seive) were treated with 0.5 mL of 1.5 M NH2OH · HCl and about 5 mL of 0.1 M HNO3. After intermittent shaking, the sample volume was diluted to 25 mL with 0.1 M HNO3 and the Mn concentration measured by ICP-AES.
The Standard Chromium Net Oxidation Test (Bartlett and James, 1996) was performed without modification. Briefly, 2 g dry soil or 2 mL moist soil were shaken with 20 mL of 0.001 M CrCl3 for 15 min, a phosphate buffer was added to remove adsorbed Cr(VI), the sample was centrifuged, and Cr(VI) determined in the supernatant using s-diphenyl carbazide. Moist soil was measured volumetrically and dry weight was determined on separate 2-mL subsamples. Extractable Mn was measured using the modified Morgan's extractant (1.25 M NH4OAc, pH 4.8) at a 5:1 solution/soil ratio and 15 min shaking. The Mn concentration was measured by ICP·AES and assumed to represent Mn(II).
Drying Experiments
For the long-term drying experiment, approximately 100 mL each of four soils (chosen to provide a range in the Cr test) were spread to air-dry in the dark in 12- by 12-cm polyethylene dishes. After 1 wk drying in the dark, duplicate dishes of each soil were (i) left in the dark uncovered, (ii) put under fluorescent lamps, and (iii) put into plastic snap-cap vials in the dark. Drying was at room temperature, which varied during the course of the experiment between about 22 and 28°C. The treatments were designed to compare long-term storage in the light vs. the dark and continued exposure to air vs. sealed, although results indicate that moisture continued to be lost from the "sealed" containers. The fluorescent lamps contained two 15-W cool white tubes with an intensity of 4 x 102 µE m-2 s-1 in the visible range and peak intensity at about 580 nm. Oven-dry weight (105°C), extractable Mn, the absorbance of the modified Morgan's extract at 360 nm, and the Cr test were determined 8, 16, 30, and 72 d after the samples were first put out to dry. The Cr test was done with 2-mL samples, and separate 2-mL subsamples were dried at 105°C to calculate results. The absorbance at 360 nm was used as a rough estimate of soluble organic C (Swift, 1996). Wavelength scans through this region showed a broad peak rising below the wavelength used. No absorbance at 360 nm was found in preparations of Mn(II) and Mn(III) acetate. After 139 d, the same tests described above were performed on three of the soils (insufficient weight of the Lordstown remained). After 264 d, all measurements except dry weight were performed on the dark treatments in the same three soils.
Short-term drying experiments were performed by placing approximately 500 g of moist soil into 20-cm-diam. porcelain evaporating dishes and air drying on the lab bench at room temperature. The Cr oxidation test was performed at 0, 4.5, 23.5, 48, 72, 95.5, and 360 h after the start of drying. Samples were measured volumetrically with a 2-mL spoon and separate samples were taken to determine dry weight and moisture content. The soils were stirred about every 30 min during the first 6 h of the experiment and before each sampling. At the 1-, 2-, and 3-d sampling times, subsamples (10–40 g) of the soils were put into polyethylene bags and remoistened to the original water content of the stored soils. The amount of water to add was based on the moisture content at the time of sampling and the water was added 
5 h after the sampling time. Each of the remoistened samples was given the Cr test at the subsequent sampling times and at 504 h after the initial drying.
On a separate set of subsamples, a number of pretreatments were used to investigate the potential removal of reductants. Treatments were H2O after 2 d of air drying (to remove easily soluble organics), 0.01 M CaCl2 after 3 d of air drying (to remove and/or flocculate soluble organics), 0.001 M CuCl2 14 d after air drying commenced (to possibly complex and remove sorbed organics), and 0.01 M K2HPO4 · KH2PO4 after 14 d (to possibly remove sorbed organics). Twenty milliliters of each solution were added to 2 g of soil in a 50-mL centrifuge tube, shaken for 15 min, centrifuged, carefully decanted, and then the Cr test was performed. The same subset of four soils had the Cr test determined after 3 h of drying at 105°C in a convection oven. In a different experiment, moist and dry (24 d) samples of the Lordstown II and Denu soils were weighed into centrifuge tubes to give the dry equivalent of 2 mL. The Cr test was repeated five times on the same samples with the supernatant carefully decanted between tests.
XANES Experiments
XANES spectra were taken at beam line X26A at the National Synchrotron Light Source, Brookhaven National Laboratories, Upton, NY. We used the x-ray fluorescence microprobe which consists of a channel-cut Si(111) monochromator, a four-jaw slit assembly, and focusing mirrors. Detection of Mn fluorescence was performed with a Si(Li) energy-dispersive detector 90° to the incoming x-ray beam. Experiments were performed using Rh-coated Kirckpatrick-Baez micro-focusing mirrors and spot sizes varying between 27 by 26 µm to 35 by 30 µm. Spectra were obtained by monochromator scans from about 40 eV below to 300 eV above the main edge. Step size was 0.18 eV in the main-edge region but coarser above and below to obtain baselines. Energy was calibrated to the pre-edge peak of a 10% KMnO4 (MnVII) standard at 6543.3 eV (Riggs-Gelasco et al., 1996). This peak was set as 0 eV relative energy. The calibration standard was run immediately before each normal scan and, in all but one sample, immediately after. For the soil samples, the change in energy calibration position ranged between 0.02 and 0.25 eV with an average of 0.08 eV. All readings were normalized to counts in an upstream ion chamber to correct for the decay in beam intensity. Relative intensities were calculated as the Mn intensity minus the lower baseline (average of intensities from -40 to -20 eV) divided by the average upper baseline intensities between 160 and 300 eV. In a few cases, the upper baseline was not level and the relative intensity was calculated either by adjusting the upper baseline or normalizing to a consistent main peak crest intensity. Dried samples and standards were mounted in a thin layer between two pieces of prolene film at the end of a 23-mm-diam. polyethylene XRF cup (Chemplex Industries, Stuart, FL). Standards for oxidation-state comparison were prepared to provide a Mn content between 1 and 5%. Mn(II) solutions of 0.1 and 0.5 mol L-1 were prepared with reagent-grade MnSO4. For Mn(III), the Mn-pyrophosphate reagent of Bartlett and Ross (1988) was dried under N2 and analyzed in the crystallized state (4.2% Mn). Synthetic birnessite was prepared by the method of Golden et al. (1987) and confirmed with x-ray diffraction spectroscopy. The manganite sample (NMNH B7639) was obtained from the National Museum of Natural History. All minerals were ground in an agate mortar and diluted with synthetic corundum (Buehler Micropolish C, 1-µm 
-alumina).
Data Analysis
Statistical analysis was performed using the SAS System for Windows 6.12 (SAS Institute, Cary, NC). Analysis of variance was performed on the long-term drying experiment results using the change between Day 8 and Day 139. The general linear model was used and treatment differences within each soil were assessed with the least means squared slice test. XANES spectra were fit with PeakFit version 4 (SPSS, Chicago, IL) using the residuals method with Gaussian/Lorentian curves. The fits were performed on spectra between -10 and +40 eV relative energy. Individual spectra of soils and standards were modeled until the fit had an r2 of >0.999.
|
RESULTS AND DISCUSSION
|
|---|
The soils used in this study provided a medium to extremely high range of Cr oxidation capacity and Mn content (Table 2). Bartlett and James (1979), studying a wide range of field-moist soils, found Cr tests between 0.0 and 0.9 mmol kg-1 with 90% of the soils oxidizing <0.3 mmol kg-1. In comparison, our Lordstown samples oxidized about 15 times as much added Cr(III). The relatively low Cr test of the Copake soil probably resulted from its low pH. Soils generally have a wide range in total Mn content, with 0.06% being typical (Gambrell, 1996). The reducible Mn reported in Table 2 is thought to represent soil Mn oxides (Gambrell, 1996) and should correspond closely to the total Mn content. Thus a few of the soils used in this study are relatively high in Mn content.
View this table:
[in this window]
[in a new window]
|
Table 2. Characteristics of the soils used and comparison of the Cr-oxidation ability, moist vs. dry. Reducible Mn was estimated by treatment with NH2 OH · HCL and extractable Mn was measured in 1.25 M NH4OAc, pH 4.8
|
|
Long-term moist storage at 3 to 4°C did not cause any detectable trend in the Cr test (Fig. 1)
. The variation seen, especially in the three most strongly oxidizing soils, may have been caused by slight variations in timing or in the volumetric soil sampling, since the Cr test of high-Mn soils is quite sensitive to these factors. Desiccation of soils will occur with long-term cold storage if measures are not taken. The double-bagging moist storage that we used avoided any significant drying throughout the experiment. Apparently, the reactivity of the soil Mn oxides is not affected by sampling and a sustained slowdown in biological activity.
View larger version (70K):
[in this window]
[in a new window]
|
Fig. 1. The Standard Cr Net Oxidation Test on field-moist soils under long-term storage at 3°C. The Lordstown-II results are shown as 1/10 of actual
|
|
Drying and the length of drying had a dramatic effect on the Cr test and on extractable Mn (Table 2; Fig. 2–4)
. The decrease in the test after 24 d of air drying (30–50% of the moist test; Table 2) was similar to that found by Bartlett and James (1979). After 264 d of air drying in the dark, the Cr test was <5% of that found in samples after 1 wk of drying and <2% of that in moist samples (Fig. 2). The increase in extractable Mn, assumed to be Mn(II), was nearly linear over the time period, whereas the decrease in the Cr test was more curvilinear. Differences between storage treatments (thinly spread in the light or dark, or closed in a plastic container) were evident in all the soils, following the same pattern as shown for the Subsupersoil sample (Fig. 3). The more exposed the storage treatment, the steeper the decline in Cr test and increase in extractable Mn. Interestingly, moisture content decreased only a maximum of 1.1% after the initial tests with no significant treatment differences (Table 3). Berndt (1988) found similar increases in extractable Mn in Scottish soils with drying and storage, with greater increases at higher temperature (60 vs. 28°C). In the present study, changes in soluble organics, roughly estimated from the absorbance of the NH4–acetate extractions, followed the same linear upward trend as the extractable Mn and were especially higher in the light treatment (Table 3). Within each soil were significant treatment differences in the change in absorbance between 8 and 139 d (P < 0.01). Also within each soil was a correlation (P < 0.001) between absorbance and Mn(II) release, combining all treatments and times up to 139 d (Spearman's r = 0.70, 0.87, and 0.90 for the Subsupersoil, Hardhack, and Nimblett, respectively). Because soluble C was not directly measured, some caution is needed when interpreting these results. However, during extended drying there appears to be a relationship between the decline in the Cr test and the release of Mn, and probably a relationship between these two and the increase in soluble C. The initial steep decline in the Cr test may be due to competitive adsorption of recently reduced Mn(II) at sites where the Cr(III) would normally be adsorbed and oxidized. In previous work, Ross and Bartlett (1981) found that addition of Mn(II) to soil caused an immediate lowering of the Cr test. The alternative hypothesis, that there may have been a rapid increase in Cr(VI) reduction caused by the drying-induced release of soluble organics, does not appear probable, as discussed below.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 2. The effect of air drying (dark treatment) on extractable Mn and the Standard Cr Net Oxidation Test. Error bars represent ± SE
|
|
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 4. Short-term air-drying effect on the Standard Cr Net Oxidation Test and the moisture content in the Denu soil. Subsamples were remoistened to original water content at 24, 48, and 72 h. Error bars represent ± SE
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 3. The effect of air-drying treatment (dark, light, or in plastic vials in the dark) on extractable Mn and the Standard Cr Net Oxidation Test in the Subsupersoil. Error bars represent ± SE
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. The change with extended air drying in the absorbance of NH4OAc extracts (estimating soluble C) and the soil water content. Soil treatments were air drying in the dark, under fluorescent lights, or in plastic vials in the dark
|
|
Studying short-term changes upon drying showed that an initial steep drop in the Cr test roughly followed the drop in water content (Fig. 4) but the Cr test continued to decline once water content had stabilized. All soils tested (Hardhack, Subsupersoil, and Lordstown-II) behaved in a similar manner. After 72 h, the Cr test was 45 to 66% of that of the moist soil. Remoistening subsamples to the original water content caused rapid, partial recovery of the Cr test (Fig. 4). We suggest that this recovery was the result of reoxidation of reduced Mn and/or a reversal in changes in the structure of the hydrous oxides.
It has been hypothesized that drying increases soluble organics, which in turn cause the reduction of newly oxidized Cr(VI) during the Cr test (Bartlett and James, 1980; Bartlett and James, 1996). Alternatively, an increase in solubilized substances might result in blockage of the sites on the Mn oxides where Cr(III) is adsorbed and oxidized. To test these hypotheses, we used a number of pretreatments to "rinse" the dried soil before performing the Cr test (Table 4). The water and weak CaCl2 treatments were designed to remove soluble organics that might react with Cr(VI) after oxidation, but no increase in the Cr test was found. The Cu and phosphate treatments were intended to possibly "clean" the oxide surface by complexing and solubilizing adsorbed organics. Instead, the large decrease in the Cr test indicates that they were interfering, either by blocking Cr(III) sorption to the Mn oxides or by somehow increasing the reduction of Cr(VI) formed. If the Cr test is lower because more newly oxidized Cr(VI) is being reduced, repeated tests on one sample should show a different pattern, moist vs. dry, as the reducing agents are depleted. This was not found in repeated extractions (Fig. 5)
or in repeated leaching with CrCl3 (data not shown). The pattern of both the moist and dried soils were similar but with much less Cr oxidation in the dried samples. While these experiments obviously were not exhaustive in attempts to remove possible blocking or reducing agents, there was no evidence for recovery. These results do not support the hypothesis that newly oxidized Cr(VI) is being reduced at a faster rate in dry soils, causing the decrease in the Cr net oxidation test. Instead, the change is probably at the oxide surface.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 5. Comparison of successive Cr oxidation tests on field-moist and air-dried samples of the Lordstown-II and Denu soils (all results for the Denu soil are multiplied by 10). Samples were dried 1 wk and the water content was 3.7% for the Lordstown-II and 0.9% for the Denu. Error bars represent the SE
|
|
To examine possible changes in oxide structure and oxidation state, XANES spectra were obtained on samples from the long-term drying experiment (the Subsupersoil and Hardhack samples shown in Fig. 2; Table 3) and compared with samples air-dried for only 4 to 5 d. Duplicate scans, one from each replicate of the dark treatment, were arithmetically averaged and clearly showed a difference in the main-edge position (Fig. 6 and 7)
, suggesting reduction during long-term dry storage. Similar results, about 1.5 eV drop at 0.5 relative intensity, were found in a Lordstown-II sample dried for 480 d. The Nimblett soil from the long-term drying experiment was not tested because of its relatively low Mn content (requiring an extremely long scan-collection time). Standard preparations are shown in Fig. 8
for comparison. The movement of the main edge to lower energy (about 1.5 eV at 0.5 relative intensity) is substantial in that only about 3 eV separate the main edge of synthetic birnessite and Mn(III)-pyrophosphate (Fig. 8). This change is noteworthy because these samples were not rewetted at any time during extended air-dried storage. Therefore the results are not confounded by the potential effects of reducing organics solubilized by a drying and rewetting cycle, but are instead evidence of in situ reduction in the air-dried state.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 6. XANES spectra of the Subsupersoil soil comparing recent and long-term dried samples. Modeled fits are a linear mixture of the recently air-dried spectra and either aqueous Mn(II) or manganite
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 7. XANES spectra of Hardhack soil comparing recent and long-term dried samples. Modeled fits are a linear mixture of the recently air-dried spectra and either aqueous Mn(II) or manganite
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 8. XANES spectra of Mn standards. The synthetic birnessite is largely Mn(IV) but could contain as much as 20% Mn(III)
|
|
Modeling the change in oxidation state is somewhat problematic because the shape and position of the main edge also depends on the coordination environment (Manceau et al., 1992). Using curves fit to a Mn(II) standard (0.1 mol L-1 MnSO4) and the recently dried samples, a good fit to the long-term dried spectra could be obtained only at lower relative intensity (Fig. 6 and 7). A better overall fit was obtained using manganite (
-MnOOH) as the model although the 40 to 50% contribution needed for a good fit seems high. Extractable (pH 4.8 ammonium acetate) Mn in the dried Hardhack, after 440 d, was 10.8% of the total oxide content, close to the proportion used in the curve fit with Mn(II). However, the same relationship did not hold with the higher-Mn soils (Subsupersoil and Lordstown-II) in which the change in the XANES spectra was modeled using 10% Mn(II) but the extractable Mn was only 0.95 and 2.11%, respectively, of the estimated total oxide concentration. There are at least two problems with interpreting this data. First, the ease of extractability of any Mn(II) formed during drying is not known but many workers have found it difficult to remove Mn(II) added to oxides, even when it is shown that the added Mn(II) was not itself oxidized (e.g., Tu et al., 1994). Second, and somewhat opposite to the first, the buffered acetate extractant may have facilitated reduction and release of Mn to solution. In four samples tested, 1.0 M NH4NO3 extracted an average of three times less Mn than the NH4OAc (pH 4.8, 1.25 M acetate). Reconciling the extractable Mn data with the XANES spectra may not be possible but it is reasonable to assume that nonextractable reduced Mn was present. The above difficulties notwithstanding, the XANES evidence is important. No change was observed after 52 d of drying in subsamples from the Subsuper and Hardhack soils. No repeatable changes in Mn XANES spectra could be detected on samples dried in place, and little difference was found in comparisons of moist soils with recently air-dried soils (Ross et al., 2001). Long-term drying is needed to both observe large changes in extractable Mn and detectable change in XANES spectra.
Structural changes in birnessite are known to occur with heating (Golden et al., 1987), change in pH (Drits et al., 1997), and adsorption of Mn(II) (Tu et al., 1994). Although not diagnostic, the XANES spectra of the high-Mn soils most closely resembled birnessite with a relatively low pre-edge peak and relatively high main-edge crest (Manceau et al., 1992). The main edge for most soils was at a slightly higher energy than birnessite. If this is interpreted as greater Mn(IV):Mn(III) then the structure of the soil oxides is consistent with that proposed by Drits et al. (1997) for H-substituted birnessite. This transformed birnessite had a Mn(III) content less than the roughly 20% found in synthetic Na-birnessite, and more lattice vacancies, which suggested more reactive sites. Tu et al. (1994) found that adsorbed Mn(II) caused transformation of birnessite to groutite at pH 6 and manganite at pH 7. Thus it is not unreasonable to hypothesize that the long-term drying-induced effect on soil Mn oxides is one of both reduction and structural change. Both should lessen the Cr test. The mechanism for both effects is not established. The apparent increase in soluble C over time, which correlated with extractable Mn, suggests that there is an intimate association between reducing organics and oxidized Mn, and drying facilitates the reaction. It is established that greater soluble organic C will be found in rewetted dried soils than in soils kept continuously moist (Bartlett and James, 1993). Perhaps the reaction is driven by the increase in soil solution acidity known to occur as the water film becomes thinner with drying. Reduction during drying is counterintuitive but soil Mn oxides appear to have a delicate poise. Alternatively, the dehydration of the hydrous oxides could conceivably cause disproportionation of structural Mn(III) to Mn(II) and Mn(IV) without an overall change in redox status.
Clearly the oxidative capacity of the soil Mn oxides is swiftly decreased by drying and minimized by extended drying. Chemical studies using dried samples will not be transferable to moist field conditions, especially if studied after extended storage. Manganese XANES spectra do not appear to be affected by short-term drying. In fact, using dried samples may provide a more accurate analysis of the soil oxides than moist samples because of x-ray induced reduction (Ross et al., 2001). The rapid decline in the oxidative capacity of the soil oxides reflects change at the surface. Either adsorption sites are blocked or the surface structure changes and electron transfer from adsorbed Cr(III) is less probable. Changes that occur only at the surface are not easily detectable with XANES spectroscopy because fluorescence is measured from much deeper than the surface layer, up to 50 µm. With long-term drying, the continued decline in Cr oxidation capacity and the continued increase in extractable Mn suggest an ongoing transformation of the oxides, initially only detectable with chemical tests but eventually seen in the XANES spectra. The reversibility of these changes has not been well studied but recovery of Cr oxidizing ability has been observed both in this study and by Bartlett and James (1979). Natural wetting and drying cycles, especially in arid regions, probably produce a similar effect on soil Mn oxides. The examination of Mn oxides continues to be a challenge because of sensitivity to sample handling. However, this is the nature of soil.
|
ACKNOWLEDGMENTS
|
|---|
We thank Darrell Schulze of Purdue Univ. for generous assistance with synchrotron techniques; many members of NCR-174, Synchrotron x-ray Sources in Soil Science Research, for their advice on techniques and data analysis; the Smithsonian National Museum of Natural History for Mn oxide mineral standards; and Ashley Ross and Kara Lenorovitz for technical assistance.
Received for publication February 16, 2000.
|
REFERENCES
|
|---|
- Amacher, M.C., and D.E. Baker. 1982. Redox reactions involving chromium, plutonium, and manganese in soils. DOE/DP/04515-1. Penn. State Univ., University Park, PA.
- Bartlett, R.J. 1990. An A or an E: Which will it be? p. 7–18. In J.M. Kimble and R.D. Yeck (ed.) Characterization, classification, and utilization of Spodosols. Proc. 5th Int. Soil Correlation Mtg. 1–14 Oct. 1988. USDA-SCS, Lincoln, NE.
- Bartlett, R.J., and B.R. James. 1979. Behavior of chromium in soils: III. Oxidation. J. Environ. Qual. 8:31–35.[Abstract/Free Full Text]
- Bartlett, R.J., and B.R. James. 1980. Studying dried, stored, laboratory soil samples—some pitfalls. Soil Sci. Soc. Am. J. 44:721–724.[Abstract/Free Full Text]
- Bartlett, R.J., and B.R. James. 1993. Redox chemistry of soils. Adv. Agron. 50:151–208.
- Bartlett, R.J., and B.R. James. 1996. Chromium. p. 683-701. In D.L. Sparks (ed.) Methods of soil analysis. Part 3. SSSA, Madison, WI.
- Bartlett, R.J., and D.S. Ross. 1988. Colorimetric determination of oxidizable carbon in acid soil solutions. Soil Sci. Soc. Am. J. 52: 1191-1192.[Abstract/Free Full Text]
- Berndt, G.F. 1988. Effect of drying and storage conditions upon extractable soil manganese. J. Sci. Food Agric. 45:119–130.
- Burns, R.G., and V.M. Burns. 1977. The mineralogy and crystal chemistry of deep-sea manganese nobules—a polymetallic resource of the twenty-first century. Philos. Trans. R. Soc. Lond. A286:283–301.
- Charlet, L., and A. Manceau. 1992. In situ characterization of heavy metal surface reactions: The chromium case. Int. J. Environ. Anal. Chem. 46:97–108.
- Drits, V.A., E. Silvester, A.I. Gorshkov, and A. Manceau. 1997. Structure of synthetic monoclinic Na-rich birnessite and hexagonal birnessite: I. Results from X-ray diffraction and selected-area electron diffraction. Am. Mineral. 82:946–961.[Abstract]
- Duff, M.C., D.B. Hunter, P.M. Bertsch, I.R. Triay, S.J. Chipera, and D.T. Vaniman. 1998. Oxidation state analyses and mineral associations of sorbed 239Pu on tuff. Agron. Abstr. 1998:199.
- Fendorf, S.E., and R.J. Zasoski. 1992. Chromium(III) oxidation by gamma-MnO2: I. Characterization. Environ. Sci. Tech. 26:79–85.
- Fujimoto, C.K., and G.D. Sherman. 1945. The effect of drying, heating, and wetting on the level of exchangeable manganese in Hawaiian soils. Soil Sci. Soc. Am. Proc. 10:107–112.
- Gambrell, R.P. 1996. Manganese. p. 665–682. In D.L. Sparks (ed.) Methods of soil analysis. Part 3. SSSA, Madison, WI.
- Golden, D.C., C.C. Chen, and J.B. Dixon. 1987. Transformation of birnessite to buserite, todorokite, and manganite under mild hydrothermal treatment. Clays Clay Miner. 35:271–280.[Abstract]
- Heintze, S.G. 1946. Manganese deficiency in peas and other crops in relation to the availability of soil manganese. J. Agric. Sci. 36: 227–238.
- Jenne, E.A. 1968. Controls on Mn, Fe, Cu, Ni, Co, and Zn concentrations in soils and water: the significant role of hydrous Mn and Fe oxides. Adv. Chem. 73:337–389.
- Manceau, A., V.A. Drits, E. Silvester, C. Bartoli, and B. Lanson. 1997. Structural mechanism of Co2+ oxidation by the phyllomanganate buserite. Am. Mineral. 82:1150–1175.[Abstract]
- Manceau, A., A.I. Gorshkov, and V.A. Drits. 1992. Structural chemistry of Mn, Fe, Co, and Ni in manganese hydrous oxides: Part I. Information from XANES spectroscopy. Am. Mineral. 77:1133–1144.[Abstract]
- McBride, M.B. 1987. Adsorption and oxidation of phenolic compounds by iron and manganese oxides. Soil Sci. Soc. Am. J. 51: 1466–1472.[Abstract/Free Full Text]
- McKenzie, R.M. 1989. Manganese oxides and hydroxides. p. 439–465. In J.B. Dixon and S.B. Weed (ed.) Minerals in soil environments. 2nd ed. SSSA, Madison, WI.
- Murray, J.W., and J.G. Dillard. 1979. The oxidation of cobalt(II) adsorbed on manganese oxide. Geochim. Cosmochim. Acta 43: 781–787.
- Naidja, A., P.M. Huang, and J.-M. Bollag. 1998. Comparison of reaction products from the transformation of catechol catalyzed by birnessite or tyrosinase. Soil Sci. Soc. Am. J. 62:188–195.[Abstract/Free Full Text]
- Oscarson, D.W., P.M. Huang, P.M. Defosse, and A. Herbillon. 1981. The oxidative power of Mn(IV) and Fe(III) oxides with respect to As(III) in terrestrial and aquatic environments. Nature (London) 291:50–51.
- Riggs-Gelasco, P.J., R. Mei, D.F. Ghanotakis, C.F. Yocum, and J.E. Penner-Hahn. 1996. X-ray absorption spectroscopy of calcium-substituted derivatives of the oxygen-evolving complex of Photosystem II. J. Am. Chem. Soc. 118:2400–2410.
- Ross, D.S., and R.J. Bartlett. 1981. Evidence for nonmicrobial oxidation of manganese in soil. Soil Sci. 132:153–160.
- Ross, D.S., H.C. Hales, G.C. Shea-McCarthy, and A. Lanzirotti. 2001. Sensitivity of soil manganese oxides: XANES spectroscopy may cause reduction. Soil Sci. Soc. Am. J. 65:744–752 (this issue).[Abstract/Free Full Text]
- Shindo, H., and P.M. Huang. 1982. Role of Mn(IV) oxide in abiotic formation of humic substances in the environment. Nature (London) 298:363–365.
- Sunda, W.G., and D.J. Kieber. 1994. Oxidation of humic substances by manganese oxides yields low-molecular-weight organic substrates. Nature (London) 367:62–64.
- Swift, R.S. 1996. Organic matter characterization. p. 1011–1070. In D.L. Sparks (ed.) Methods of soil analysis. Part 3. SSSA, Madison, WI.
- Taylor, R.M., and K.H. Nealson. 1966. The association of trace elements with manganese minerals in Australian soils. Aust. J. Plant Physiol. 4:29–39.
- Tu, S., G.J. Racz, and T.B. Goh. 1994. Transformations of synthetic birnessite as affected by pH and manganese concentration. Clays Clay Miner. 42:321–330.[ISI]
- Ulrich, H., and A.T. Stone. 1989. Oxidation of chlorophenols adsorbed to manganese oxide surfaces. Environ. Sci. Tech. 23:421–428.
- Whelan, G., and R.C. Sims. 1995. Mn-catalyzed oxidation of multiple-ringed aromatics. Hazard. Waste Hazard. Materials 12:243–256.
This article has been cited by other articles:
|
|
|
|
 
O. Furman, D. G. Strawn, and S. McGeehan
Sample Drying Effects on Lead Bioaccessibility in Reduced Soil
J. Environ. Qual.,
May 7, 2007;
36(3):
899 - 903.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. Negra, D. S. Ross, and A. Lanzirotti
Soil Manganese Oxides and Trace Metals: Competitive Sorption and Microfocused Synchrotron X-ray Fluorescence Mapping
Soil Sci. Soc. Am. J.,
March 1, 2005;
69(2):
353 - 361.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. Negra, D. S. Ross, and A. Lanzirotti
Oxidizing Behavior of Soil Manganese: Interactions among Abundance, Oxidation State, and pH
Soil Sci. Soc. Am. J.,
January 1, 2005;
69(1):
87 - 95.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. S. Ross, H. C. Hales, G. C. Shea-McCarthy, and A. Lanzirotti
Sensitivity of Soil Manganese Oxides: XANES Spectroscopy May Cause Reduction
Soil Sci. Soc. Am. J.,
May 1, 2001;
65(3):
744 - 752.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
