| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Soil Science, Box 7619, North Carolina State University, Raleigh, NC 27695-7619
b Natural Resources Canada, CANMET, 555 Booth St., Office 332A, Ottawa, ON, KIA 0G1
c Dep. of Soil and Environ. Sci., National Chung Hsing University, Taichung, 402, Taiwan
d Dep. of Geology and Geophysics, University of Wyoming, Laramie, WY 82071
* Corresponding author (dean_hesterberg{at}ncsu.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: LCF, linear combination fitting • XANES, X-ray absorption near-edge structure
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Soil P concentration, soil matrix composition (e.g., mineralogy, and organic matter content), pH, and redox potential are considered to be the main soil properties affecting PO4 dissolution and mobility. Phosphate typically binds strongly with soils. However, dissolved P concentration in runoff water was positively correlated with soil P concentration, and this relationship was soil specific (Sharpley, 1995; Pote et al., 1996). Being able to predict PO4 dissolution and mobility in different soils or under varying soil conditions would help in developing soil management practices that decrease detrimental environmental impacts of P. Phosphate speciation, that is, the chemical forms of PO4 in a soil, dictates the effects of soil PO4 concentration, mineralogy, pH, and redox potential on PO4 binding and dissolution.
Phosphate adsorption in soils has been correlated with a number of indices derived from chemical extractions (Beauchemin and Simard, 1999). For example, the PO4 sorption capacity of soils has been related to various indices based on acid-oxalate extractable Fe and Al, suggesting that poorly crystalline Fe- and Al-oxides are largely responsible for PO4 retention in acidic soils (Beauchemin and Simard, 1999). Similarly, chemical extraction analyses of Sallade and Sims (1997) suggested that PO4 in sediments collected from drainage ditches adjacent to agricultural fields was associated with Fe- and Al-oxide minerals. Ferrihydrite, a poorly crystalline Fe-oxide mineral is often found in sediments or hydromorphic soils as a precursor of other Fe-oxide minerals (Schwertmann and Cornell, 1991). Furthermore, transmission electron microscopy with energy dispersive X-ray analysis (TEM/EDX) showed association of PO4 with Al and Fe in isolated particles from different soils (Pierzynski et al., 1990a, 1990b)
Phosphate dissolution in soils may depend on the relative distribution of PO4 between Fe- and Al-oxide minerals. For example, dissolution of PO4 during soil reduction has been explained by release of PO4 associated with Fe(III)-phosphate and Fe(III)-oxide minerals (Patrick et al., 1973; Hongve, 1997; Reddy et al., 1998). However, Al-oxide minerals are considered redox inactive, so any associated PO4 should be less susceptible to release during soil reduction.
One barrier to evaluating such hypotheses is the lack of a direct method for quantifying PO4 distribution between Fe- and Al-oxide minerals when these minerals occur as a mixture (as in soils). Past research characterizing PO4 adsorption in mineral mixtures specifically kaolinite and goethite used equilibrium adsorption isotherms and kinetic measurements (Ioannou et al., 1998; Papadopoulos et al., 1998). However, definitive information about the distribution of PO4 between these two minerals could not be obtained from empirical modeling (Langmuir and Freundlich fits) of the macroscopic adsorption data. In this research, we characterized the distribution of PO4 between ferrihydrite and boehmite using P K-XANES analysis.
Hesterberg et al. (1999) identified unique features in P K-XANES spectra of strengite (FePO4·2H2O) and variscite (AlPO4·2H2O) that indicated the possibility of distinguishing adsorbed PO4 in mixed Fe- and Al-oxide systems. For example, due to electron orbital configurations and electronic transitions at the X-ray absorption edge, PO4 associated with Fe(III) and some other transition metals in PO4 minerals produces a distinct pre-edge feature on the low energy side of an intense white-line peak near 2150 eV in the P K-XANES spectrum (Behrens, 1992; Hesterberg et al., 1999; Franke and Hormes, 1995; Okude et al., 1999). This feature is absent in spectra of Al phosphates. Because of the ability of XANES to distinguish PO4 bound to Fe(III) versus Al(III), this technique was considered suitable for characterizing PO4 on Fe- and Al-oxide minerals.
The objective of this research was to utilize XANES spectroscopy to quantify the distribution of PO4 between ferrihydrite and boehmite in mixtures of these minerals, and thereby determine the relative affinity of PO4 for each mineral in the mixture. Two-line ferrihydrite (Fe5HO8·4H20) and poorly crystalline boehmite (
-AlOOH) were chosen because we expected that their high (and comparable) PO4 sorption capacities (relative to, e.g., goethite and gibbsite) would allow better detection of subtle changes in XANES spectra of the mixtures. Ferrihydrite is representative of poorly crystalline Fe-(hydr)oxides in soils, and boehmite is a finely divided, crystalline analog of noncrystalline Al hydroxides in soils.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Ferrihydrite was washed thrice with 1 mol L–1 KCl solution and further washed with 0.01 mol L–1 KCl to obtain a 0.01-mol L–1 KCl background electrolyte. Boehmite gel in deionized water was brought into a 0.01-mol L–1 KCl background by adding a 1-mol L–1 KCl solution. Both minerals were stored as stock aqueous suspensions of known (measured) solids concentration in 0.01 mol L–1 KCl (see Alcacio et al., 2001) containing 40.2 g ferrihydrite kg–1 and 14.1 g boehmite kg–1. The mean crystalline dimensions of poorly crystalline boehmite used in this study were previously determined to be 4.5, 2.2, and 10 nm along the crystal a, b, and c axes respectively, and the Brunauer–Emmett–Teller (BET) H2O surface area reported for this mineral was 514 m2 g–1 (Wang et al. 2003). Water adsorption was previously used to avoid sample drying and because this small polar molecule can access the internal surfaces present in a poorly crystalline material such as boehmite (Wang et al. 2003). Because a temperature-induced structural change in poorly crystalline materials (such as boehmite and ferrihydrite) has been observed at 100 to 110°C (Wang et al., 2003), the N2 BET surface areas of these minerals were not measured.
Adsorption Isotherms
Adsorption isotherm experiments for ferrihydrite, boehmite, and mixed ferrihydrite-boehmite (1:1 mass ratio) suspensions were conducted at pH 6.0 in 250-mL polycarbonate centrifuge bottles following the basic procedure described by Oh et al. (1999).
All samples had a suspended solids concentration of 1.50 g kg–1, constant ionic strength of 0.01 mol L–1 KCl, and total sample mass of 133.34 ± 0.01 g. Aqueous solutions for adsorption experiments (KCl, HCl, KOH, and KH2PO4, all at 0.01 mol L–1 concentrations) were prepared using analytical grade reagents and degassed (heated and N2 purged) deionized water. Stock mineral suspensions were shaken on a reciprocating shaker at a rate of 1 s–1 for at least 1 h before use. Two to eight grams of ferrihydrite, boehmite, or a 1:1 (by mass) mixture of ferrihydrite and boehmite prepared gravimetrically from stock suspensions were weighed while vigorously stirring a stock suspension on a magnetic stirrer, and brought to about 70% of the final sample mass with 0.01 mol L–1 KCl. An appropriate aliquot of 0.01 mol L–1 KH2PO4 was slowly added to each vigorously stirred sample in random chronological order using a micropipetter. The pH was adjusted to pH 6.0 using 0.01 mol L–1 HCl or 0.01 mol L–1 KOH, and the sample headspace was flushed with N2 gas. Samples were shaken for 42 h on a reciprocating water bath shaker at a rate of 0.5 s–1 and 22°C. Kinetics of PO4 sorption is complex, and this operationally chosen time of 42 h should be sufficiently long to complete fast sorption reactions (Li and Stanforth, 2000).
After about 16 h of shaking, the pH varied by an average of 0.2 units and was again adjusted to pH 6.0 and each sample was brought to its final mass. The pH was again checked after 40 h of equilibration and minor adjustments (usually <0.1 units) were made if needed. After equilibration, samples were centrifuged at approximately 6000 x g for 15 min, and the supernatant solutions were decanted. The pH was measured in a portion of the supernatant solution before filtering and was found to be 6.0 ± 0.1 for all samples. The remaining solutions were filtered under vacuum using 0.2-µm Millipore Isopore polycarbonate filter membranes (Millipore Corp., Bedford, MA). Dissolved PO4 was measured in the supernatant solutions using the molybdate colorimetric (Murphy-Riley) procedure (Olsen and Sommers, 1982). The concentration of PO4 adsorbed on samples was determined as the difference between total added PO4 and PO4 measured in supernatant solutions. Samples were analyzed on a Shimadzu Model UV 2101-PC spectrophotometer using a 1-cm (for higher-P samples) or 5-cm path length cell. Additional isotherm data for the single and mixed ferrihydrite/boehmite (mixed-mineral) systems were obtained on scaled down samples of 30 g total mass in 50 mL polycarbonate centrifuge tubes prepared under identical constraints and following an analogous procedure as outlined above. Isotherm results from both procedures were integrated.
XANES Data Collection and Analysis
Sample Preparation
For XANES analysis, each sedimented mineral sample from the 250-mL centrifuge bottles used for concurrent isotherm experiments was rinsed into 50-mL polycarbonate centrifuge tubes using a portion of its supernatant solution, and centrifuged at approximately 20000 x g for 15 min. Because the supernatant solution in equilibrium with the solids from the prior centrifugation was used, no adsorption or desorption was expected. Supernatant solutions were decanted and each sedimented mineral sample was homogenized by mixing thoroughly with a clean teflon spatula in the 50-mL tube. The moist paste was dewatered to a clay/water ratio of about 1:2 by placing it on a 0.2-µm Millipore filter, under vacuum, for <60 s. Samples were loaded into labeled, acrylate sample holders and covered with 5-µm polypropylene X-ray film (Spex Industries, Metuchen, NJ) and secured with chemically pure Kaptan tape (Budnick Converting, Inc., Columbia, IL). Individually mounted samples were covered with a second piece of acrylate to protect the sample during transport, and sealed into a labeled low-density polyethylene plastic bag. All samples were sealed into a second plastic bag with a moist paper towel to prevent desiccation. Experiments were timed so that sample preparation was completed a maximum of 3 d in advance of XANES data collection. All XANES data for single, and mixed-mineral systems were collected in June 2002 (Jun 02) during a single synchrotron beam time (a data collection period) except for five additional samples of mixed-mineral systems (100, 570, 760, 920, and 1190 mmol PO4 kg–1) collected in October 2002 (Oct 02) to determine reproducibility of results.
Data Collection
Phosphorus K-XANES data acquisition was done at Beamline X-19A at the National Synchrotron Light Source, Brookhaven National Laboratory in Upton, NY. The electron beam energy was 2.5 GeV and the maximum beam current was 300 mA. The synchrotron radiation was monochromatized by a Ge [Ge(111)] monochromator. The monochromator was calibrated to 2149 eV at the edge (maximum peak in the first-derivative spectrum) of variscite. A variscite reference for monochromator calibration could not be placed behind samples because of the low energy (low penetrating power) of the X-rays at the P K-edge. For example, we calculate based on absorption coefficients (McMaster et al., 1969) that >99% of the X-ray intensity at 2150 eV would be attenuated by a 10-µm thickness of Fe–oxide. Samples of thickness << 10 µm would be required for collecting data in transmission mode, which was not practical. Moreover thin samples can desiccate quickly, thus defeating the purpose of using XANES analysis for moist samples to determine PO4 distribution in situ. Therefore, a 0.1-mm thick moist paste was used for data collection to maintain sample moisture. X-ray absorption near-edge structure spectra were collected at photon energies between 2079 and 2248 eV, with a minimum step size of 0.2 eV between 2099 to 2174 eV. Two to four scans with consistent baselines were ensemble averaged.
Spectra were collected in fluorescence mode using a PIPS (Passivated Implanted Planar Silicon) detector mounted into a He-filled sample chamber. X-ray absorption near-edge structure data were also collected for variscite and strengite standards diluted to 400 mmol P kg–1 in boron nitride. Self-absorption effects can distort XANES spectra collected in fluorescence mode, particularly at low X-ray energies as used here, and at high concentrations of the analyte (P in this case) (Troger et al., 1992). Hesterberg et al. (1999) calculated that self-absorption at the P K-edge was <10% for PO4 mineral samples diluted to 800 mmol kg–1 in boron nitride. If self-absorption significantly affected our XANES spectra, we would expect to see a decrease in the white-line peak intensity with increasing adsorbed P. However, the white-line peak intensities for PO4 adsorbed on ferrihydrite remained essentially constant between 100 and 1680 mmol P kg–1 (Fig. 1 , discussed below), indicating that self-absorption did not measurably impact our results.
|
Linear Combination Fitting
The proportions of total PO4 adsorbed on each mineral in the mixed-mineral suspensions were determined using least squares linear combination fitting (Vairavamurthy et al., 1997; Hutchison et al., 2001), with spectra for adsorbed PO4 in the single-mineral systems serving as standards. Fitting results were judged according to their chi-square (goodness-of-fit) values.
X-ray absorption near-edge structure spectra for PO4 adsorbed in single- and mixed-mineral systems at lower concentrations were noisier than spectra at higher adsorbed PO4 concentrations. Therefore, standards for single-mineral samples of lower concentration (
100 mmol PO4 kg–1 ferrihydrite or 200 mmol kg–1 boehmite) were used for mixed-mineral samples of lower concentration (100 mmol PO4 kg–1). Similarly, standards for single-mineral samples of higher adsorbed concentration (>100 mmol PO4 kg–1 ferrihydrite or >200 mmol kg–1 boehmite) were used for mixed-mineral samples of higher concentration (>100 mmol PO4 kg–1). Further details about the rationale for choosing single-mineral standards for fitting analysis are included in the Results and Discussion section below.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
 | [1] |
10% on the low side of qm, the predicted concentration determined by a direct fit of the Freundlich model to the mixed mineral isotherm (solid lines in Fig. 2). The isotherm fitting results indicated that adsorption in the mixed-mineral system essentially behaved (within about 10% variation) as a linear combination of adsorption in the single-mineral systems. That is, there was no interaction between the minerals that notably affected PO4 adsorption. In general, one cannot determine from the isotherm data how PO4 is distributed between ferrihydrite and boehmite at any given adsorbed PO4 concentration in the mixed-mineral systems. Therefore, XANES spectroscopy was used to determine PO4 distribution in the mixed-mineral systems.
Phosphorus K-XANES
Adsorbed Phosphate in Single- and Mixed-Mineral Systems
Figure 1 shows examples of edge-normalized XANES spectra for PO4 adsorbed at different concentrations on ferrihydrite, boehmite, or mixed-mineral systems. All spectra were characterized by an intense resonance (white-line) near 2150 eV (1 eV relative energy), and weaker oscillations between 5 and 15 eV (relative energy). The white-line peak intensity of XANES spectra for PO4 on boehmite or ferrihydrite did not change systematically with adsorbed PO4 concentration (Fig. 1). However, a statistically significant difference (p < 0.05) between the mean white-line peak intensity for PO4 on boehmite (4.0 ± 0.1) versus ferrihydrite (4.36 ± 0.02) was observed. The spectra for PO4 in mixed-mineral systems showed some differences in the white-line peak intensity, but no trend with concentration (Fig. 1). The spectra for PO4 on ferrihydrite showed a pre-edge feature near –4 eV, which was not present in the spectra for PO4 on boehmite as discussed in more detail below. X-ray absorption near-edge structure spectra for PO4 adsorbed in mixed-mineral systems (June 2002) also showed a pre-edge feature (Fig. 1), which tended to diminish in intensity with increasing adsorbed phosphate concentration (discussed below).
X-ray absorption near-edge structure spectral features arise from electronic transitions during X-ray absorption, as influenced by the atomic coordination environment around the absorbing atom (P in this case). Features are due to electronic transitions into bound states (pre-edge features) or to photoelectron backscattering from surrounding atoms (post-edge features) (Franke and Hormes, 1995; Stohr, 1996). For K-shell spectra, the observed resonances typically correspond to dipole-allowed transitions of a 1s electron to 
and 
antibonding orbitals (Stohr, 1996). The absorption edge is usually defined as the energy at which the 1s electron from the K shell escapes into the continuum, and is estimated in practice by the most intense peak in the first derivative XANES spectra (Stohr, 1996; Sayers and Bunker, 1988). However, in the P XANES, an intense resonance (white-line) resulting from electronic transitions of the core electron into unoccupied p like valence electronic states occurs at an energy less than the absorption edge (Franke and Hormes, 1995). Therefore, we defined the edge as shown in Fig. 1, at the energy yielding a relative maximum in the first derivative XANES spectrum on the high-energy side of the white-line peak. X-ray absorption near-edge structure spectra for PO4 adsorbed on ferrihydrite and boehmite had edges at 12 and 14 eV, respectively (Fig. 1).
Normalized XANES spectra for PO4 adsorbed on boehmite at concentrations 200 mmol kg–1 (data not shown) and PO4 adsorbed on ferrihydrite at concentrations 100 mmol kg–1 (data not shown) were noisier than XANES spectra for these minerals at higher adsorbed PO4 concentrations because of their lower concentration-dependent edge step. Data normalized to the maximum fluorescence yield at the post edge feature and at 30 eV followed similar trends as edge normalized spectra, and are not shown. Hereafter, data for edge-normalized spectra will be shown and discussed, unless otherwise noted.
Comparison with Iron(III) and Aluminum(III)-Phosphates
For our research on adsorbed PO4 species, strengite and variscite served as standards of known molecular structure of Fe(III) vs. Al(III)-bound PO4. The XANES spectrum for strengite showed a pronounced pre-edge feature at 2146 eV, whereas the variscite spectrum showed no such pre-edge feature (Fig. 3)
. The pre-edge resonance observed for Fe(III)-coordinated PO4 as in strengite has been previously ascribed to hybridization of Fe-3d, O-2p, and P-3p valence orbitals giving some p character to the d like unoccupied states from Fe(III) (Franke and Hormes, 1995; Behrens, 1992; Okude et al., 1999). The lack of a pre-edge resonance in variscite is presumably due to the absence of d orbitals in Al. Thus, differences in electron orbital configuration resulted in differences in the pre-edge region of XANES spectra for strengite and variscite. Similarly, XANES spectra for PO4 adsorbed on ferrihydrite (Fe-oxide) at different adsorbed PO4 concentrations showed a pre-edge feature while XANES spectra for PO4 adsorbed on boehmite (Al-oxide) did not show such a pre-edge feature (average spectra from Fig. 1 shown in Fig. 3). Thus, the pre-edge feature could be used for distinguishing PO4 associated with ferrihydrite versus boehmite.
|
Phosphate Adsorbed in Mixed-Mineral Systems (Pre-edge)
Because the pre-edge feature has been used to differentiate P associated with ferrihydrite versus boehmite, we focused on the pre-edge region as a means for characterizing adsorbed PO4 in the mixed-mineral systems. With increasing concentration of total adsorbed PO4 in the mixed-system, the pre-edge feature intensity showed a trend from being similar to PO4 on ferrihydrite, toward having intensity intermediate between that of PO4 on ferrihydrite and PO4 on boehmite (Fig. 4)
. X-ray absorption near-edge structure spectra for mixed-mineral systems from October 2002 (data not shown) generally followed the same trend. This trend indicated that with increasing adsorbed PO4 concentration in mixed-mineral systems, an increasingly greater proportion of PO4 was adsorbed on boehmite.
|
Rationale for Selecting Fitting Standards
The pre-edge region of XANES spectra for PO4 adsorbed on boehmite (>200 mmol kg–1) and PO4 adsorbed on ferrihydrite (>100 mmol kg–1) is magnified in Fig. 5
. The pre-edge region of the XANES spectra for PO4 adsorbed on boehmite (>200 mmol kg–1) was essentially identical (Fig. 5). Hence, all data were ensemble averaged to generate a mean spectral standard for boehmite for fitting spectra from the mixed-mineral systems collected in June 2002 and October 2002.
|
A different set of XANES spectra for single-mineral systems were used for fitting spectra from mixed-mineral systems at lower concentration. X-ray absorption near-edge structure spectra for PO4 on boehmite at concentrations 200 mmol kg–1 (data not shown), on ferrihydrite at concentrations 100 mmol kg–1 or on mixed-mineral systems at concentrations 100 mmol kg–1 (June 2002 and October 2002) (data not shown) were noisier than XANES spectra for samples with higher adsorbed PO4 concentrations. Also, PO4 adsorbed on ferrihydrite (June 2002) at concentrations of 50 and 100 mmol kg–1 (data not shown) had a more intense pre-edge than the XANES spectra for PO4 adsorbed at concentrations of 200 to 1680 mmol kg–1 (Fig. 5). Principal component analysis (Beauchemin et al., 2002) performed for the dataset of the seven spectra and restricted to the pre-edge region for PO4 adsorbed on ferrihydrite (June 2002) showed two significant components, suggesting the presence of two different species. A t test performed on the averaged fluorescence yields at –4.5 eV for the two suspected groups (50–100 vs. 200–1680) also indicated a significant difference (p < 0.001). Hence, for fitting analysis of the mixed-mineral system from June 2002 or October 2002 at a concentration of 100 mmol PO4 kg–1, ensemble averaged standards for ferrihydrite containing 50 and 100 mmol PO4 kg–1 and for boehmite containing 50, 100, and 200 mmol PO4 kg–1 were used.
XANES Fitting Results
Two approaches to LCF of XANES spectra for mixed-mineral systems were evaluated. The data were fit either in the pre-edge region (–7 to –2 eV) or over an expanded energy range of –7 to 5 eV to include the white-line peak. Linear combination fitting results that included the white-line peak did not yield consistent trends with level of adsorbed PO4, apparently because the white-line peak region dominated the fitting relative to the pre-edge region. Linear combination fitting results limited to the pre-edge region showed consistent trends for all concentrations of adsorbed PO4 in the mixed-mineral system. In addition, regardless of the background normalization method, spectral fitting results using the pre-edge region yielded similar (within 6% variation) distributions of PO4 between ferrihydrite and boehmite in the mixed-mineral system. Hence, only the fitting results for edge-normalized data are presented here (Table 1).
|
Based on the percentages of total PO4 on ferrihydrite (versus boehmite) in the mixed-mineral systems (Table 1) and the known amounts of total adsorbed PO4, we calculated the concentration of adsorbed PO4 on ferrihydrite for each mixed-mineral system in June 2002 and October 2002. Linear relationships between the amounts of PO4 adsorbed on ferrihydrite as a function of total PO4 adsorbed in the mixed-mineral systems depict the reproducibility of trends in PO4 distribution between each mineral (Fig. 6a) . Both sets of samples showed linear trends with nearly equivalent slopes, indicating that with increasing inputs, PO4 is distributed at a near constant proportion between ferrihydrite and boehmite. However, the fitting results for the October 2002 samples showed lower amounts of PO4 on ferrihydrite relative to boehmite (Fig. 6a). We attribute this discrepancy to the lack of a complete set of fitting standards analyzed along with these samples in October 2002. So, we have more confidence in the accuracy of the results from June 2002 samples.
|
A comparison of calculated concentrations of PO4 on ferrihydrite in the mixed-mineral systems versus the no-preference line showed that PO4 had greater affinity for ferrihydrite compared with boehmite at the lowest total adsorbed concentration (100 mmol PO4 kg–1) (Fig. 6b). At intermediate total adsorbed PO4 concentrations (200–600 mmol PO4 kg–1), the proportion of PO4 adsorbed on ferrihydrite (64–78% of total) deviated by 9% from the no-preference distribution (Table 1). Therefore at intermediate adsorbed PO4 concentrations, the adsorbed level on either mineral could be predicted solely on the basis of the relative contribution of that mineral to the total PO4 adsorption capacity of the system (Fig. 6b).
At the greatest level of total adsorbed PO4 in the mixed system depicted in Fig. 6b (1310 mmol kg–1), PO4 was at (near) maximum adsorption capacity for the system. Therefore, PO4 should be distributed between ferrihydrite and boehmite in proportion to the maximum adsorption capacity of each mineral, regardless of the affinity preference of PO4 for either mineral. If so, then the data point for 1310 mmol PO4 kg–1 should fall on the no-preference line, corresponding to 69% of total PO4 on ferrihydrite (31% on boehmite). However, fitting results showed that at this adsorbed concentration 59% of PO4 was on ferrihydrite, which was within 10% of the expected distribution based strictly on adsorption as the only PO4 binding mechanism. The initiation of a surface precipitate at this high-adsorbed PO4 concentration could explain any negative deviation from the no-preference line beyond experimental error.
The adsorption preference trends in Fig. 6b, particularly the high affinity preference of PO4 for ferrihydrite at 100 mmol kg–1 total adsorbed PO4, suggests the presence of a limited number of high affinity binding sites on ferrihydrite. If so, then with increasing levels of PO4 these sites are filled first and then PO4 distributes more to boehmite in the mixture. An analogous trend was observed for Pb(II) by Templeton et al. (2001), where Pb(II) sorption at low levels was dominated by complexation to high-affinity metal-oxide surface sites in biofilm coated hematite and corundum surfaces. Finally, to the extent that the limited results of our study on pure, binary mixtures can be extended to slightly-acid soils, our findings suggest that soil phosphate would mostly be distributed between poorly crystalline Fe- and Al-oxides in proportion to the PO4 sorption capacity of each mineral.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 13, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
-FeOOH). J. Colloid Interface Sci. 230:12–21.[ISI][Medline]This article has been cited by other articles:
|
|
 |
|
  B. Ajiboye, O. O. Akinremi, and A. Jurgensen Experimental Validation of Quantitative XANES Analysis for Phosphorus Speciation Soil Sci. Soc. Am. J., June 29, 2007; 71(4): 1288 - 1291. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. He, C. W. Honeycutt, T. Zhang, P. J. Pellechia, and W. A. Caliebe Distinction of Metal Species of Phytate by Solid-State Spectroscopic Techniques Soil Sci. Soc. Am. J., May 16, 2007; 71(3): 940 - 943. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Shober, D. L. Hesterberg, J. T. Sims, and S. Gardner Characterization of Phosphorus Species in Biosolids and Manures Using XANES Spectroscopy J. Environ. Qual., October 27, 2006; 35(6): 1983 - 1993. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. C. Murray and D. Hesterberg Iron and Phosphate Dissolution during Abiotic Reduction of Ferrihydrite-Boehmite Mixtures Soil Sci. Soc. Am. J., June 21, 2006; 70(4): 1318 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. O. Maguire, D. Hesterberg, A. Gernat, K. Anderson, M. Wineland, and J. Grimes Liming Poultry Manures to Decrease Soluble Phosphorus and Suppress the Bacteria Population J. Environ. Qual., April 3, 2006; 35(3): 849 - 857. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Khare, C. M. Eggleston, and D. M. Lovelace SORPTION AND DIRECT ELECTROCHEMISTRY OF MITOCHONDRIAL CYTOCHROME C ON HEMATITE SURFACES Clays and Clay Minerals, December 1, 2005; 53(6): 564 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Penn, G. L. Mullins, and L. W. Zelazny Mineralogy in Relation to Phosphorus Sorption and Dissolved Phosphorus Losses in Runoff Soil Sci. Soc. Am. J., August 4, 2005; 69(5): 1532 - 1540. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Toor, J. D. Peak, and J. T. Sims Phosphorus Speciation in Broiler Litter and Turkey Manure Produced from Modified Diets J. Environ. Qual., March 1, 2005; 34(2): 687 - 697. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. Hutchison and D. Hesterberg Dissolution of Phosphate in a Phosphorus-Enriched Ultisol as Affected by Microbial Reduction J. Environ. Qual., September 1, 2004; 33(5): 1793 - 1802. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
