| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Soil and Water Science, Gulf Coast Research and Education Center, University of Florida, Wimauma, FL 33594
b Department of Soil Science, North Carolina State University, Raleigh, NC 27695
c Department of Plant and Soil Sciences, University of Delaware, Newark, DE 19716
* Corresponding author (alshober{at}ufl.edu)
Received for publication March 10, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
Abbreviations: APL, alum-treated poultry litter • BPR, biological phosphorus removal • DCB, dithionite–citrate–bicarbonate • DM, dairy manure • NMR, nuclear magnetic resonance • NPL, normal poultry litter • TP, total phosphorus • WSP, water-soluble phosphorus • XANES, X-ray absorption near-edge structure • XRD, X-ray diffraction
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
Researchers have long been interested in identifying the chemical P species that exist in biosolids or manures to better understand the potential for P loss to water when these materials are land applied. Several sequential fractionation techniques, where P is partitioned into operationally defined fractions based on its solubility in a series of extractants, have been developed to characterize inorganic and organic P species in manures, biosolids, or soils as a function of factors such as animal species and diet, and waste treatment, storage, and handling (Chang and Jackson, 1958; Hedley et al., 1982; McCoy et al., 1986; McGrath et al., 2005; Sharpley and Moyer, 2000; Williams et al., 1971). While macroscopic-scale studies (e.g., sequential fractionation, laboratory incubation studies, rainfall simulation, and column leaching experiments) have provided useful information on the effects of land application of organic P sources on soil P, these techniques cannot provide specific information about the exact chemical species of P in manures, biosolids, or amended soils. Consequently, it remains difficult to predict long-term effects on water quality of P added in manures and biosolids because of our limited understanding of the actual solid phases of P present in these materials.
Fortunately, within the last decade, we have seen marked advances in the use of more sophisticated analytical techniques to increase our knowledge of the mineral and organic phases controlling P solubility in manures and biosolids. Solid-state 31P nuclear magnetic resonance (31P NMR) spectroscopy has been used to characterize inorganic P species in biosolids (Frossard et al., 1994) and poultry litters, but this technique has problems identifying organic forms of P when present as part of a complex mixture of P species or P associated with paramagnetic elements (e.g., Fe or Mn) due to poor spectral resolution (Hunger et al., 2004). In contrast, other researchers have used solution 31P NMR spectroscopy to successfully identify forms of organic P in dairy manures (Hinedi et al., 1989; Toor et al., 2005a) and poultry litters (Maguire et al., 2004; McGrath et al., 2005). Solution 31P NMR requires chemical extraction of the sample to improve spectral resolution but allows simultaneous identification of several P compounds; however, it is not possible to distinguish between specific forms of inorganic P (Al, Ca, or Fe phosphates) using this method (Turner, 2004). The forms of P in metal-salt- and lime-treated biosolids have also been investigated using X-ray diffraction (XRD), but this method cannot be used to identify noncrystalline species and also requires high concentrations of P minerals (Ippolito et al., 2003). Synchrotron based XANES spectroscopy is another analytical tool that has been successfully used to study P speciation in soils (Beauchemin et al., 2003) and poultry litters and manures (Peak et al., 2002; Toor et al., 2005b). In addition, XANES is able to characterize poorly crystalline, amorphous materials and organic species by comparing the spectra of representative standards with the spectrum of an unknown sample. It is important to note that speciation of unknown samples by linear combination fitting with XANES spectra is limited by data quality and how well the standard library represents the actual chemical species present in the unknown sample (Beauchemin et al., 2003). In addition, the lack of distinct features in XANES spectra for many standards may hinder our ability to distinguish between multiple forms of P (e.g., various organic P species) in unknown samples (Peak et al., 2002).
Due to the limitations of individual P characterization techniques, it may be possible to obtain a more comprehensive understanding of the complex P speciation in biosolids and manures by using more than one of the above techniques. Therefore, we used XANES spectroscopy and sequential chemical fractionation to characterize the forms of P in manures and biosolids that are regularly applied to agricultural soils in the mid-Atlantic USA. While sequential fractionation studies with a wide variety of manures and biosolids are relatively common, XANES analysis of biosolids and manures is very limited, despite its potential to provide more definitive information on P speciation. Comparing XANES results with those of sequential fractionation studies should also help us to better understand the forms of P removed from manures and biosolids by selective chemical extraction techniques.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
Samples were collected at natural moisture content and stored at 4°C until use. A portion of each organic P source was dried at 60 ± 5°C, ground to pass a 0.8-mm screen using a stainless steel Wiley mill, and analyzed for total P, Ca, Al, and Fe (USEPA, 1986). Nine of the organic P sources (all but DM-B, DM-C, DM-D, DM-E, and DM-F) were characterized for P forms by sequential chemical fractionation (Kuo, 1996): One gram of each organic P source was sequentially extracted with (i) NH4Cl (50 mL 1 M NH4Cl, 30-min reaction time, aimed at extracting readily soluble P); (ii) NH4F (50 mL 0.5 M NH4F at pH 8.2, 1-h reaction time, aimed at P associated with Al); (iii) NaOH (50 mL 0.1 M NaOH, 17-h reaction time, aimed at P associated with Fe); (iv) DCB (40 mL 0.3 M Na3C6H5O7 + 5 mL 1 M NaHCO3 + 1 g Na2S2O4, 15-min reaction time in a water bath at 85°C, for reductant-soluble P); and (v) H2SO4 (50 mL 0.25 M H2SO4, 1-h reaction time, for P associated with Ca). All extracts were analyzed for P, Al, Fe, and Ca by inductively coupled plasma atomic emission spectroscopy. This sequential fractionation method was selected because it was designed to distinguish between P associated with Fe, Al, and Ca, unlike other techniques that combine Al- and Fe-P into one fraction (e.g., Hedley fractionation). We believed this technique would allow us to better determine the effects of various wastewater treatment processes and the addition of alum to poultry litter on P distribution among operationally defined P fractions. In addition, dried samples of all biosolids, NPL, APL, and DM-A were analyzed for crystalline phosphate minerals using XRD. Dairy manure samples B through F were also analyzed using solution 31P NMR by Toor et al. (2005a).
Phosphorus Standard Preparation and Synthesis
Mineral, sorbed phase, and organic P standards that were likely to exist in manures and municipal biosolids were purchased or synthesized. Brushite (CaHPO4·2H2O) and hydroxylapatite [Ca5H(PO4)3·2.5H2O] standards were purchased from Fisher Scientific (Hampton, NH), monocalcium phosphate [Ca(H2PO4)2] and phytic acid (Ca and Na salts) were purchased from Sigma-Aldrich (St. Louis, MO), and monetite (CaHPO4) was purchased from J.T. Baker (Mallinckrodt Baker, Phillipsburg NJ). Variscite (AlPO4·2H2O) standard was purchased from Ward's Natural Science (Rochester, NY). The following standards were synthesized: Fe phosphates with varying degrees of crystallinity and strengite (FePO4·2H2O; Dalas, 1991), Al phosphates of varying crystallinity (AlPO4·2H2O; Hsu and Sikora, 1993), octacalcium phosphate [Ca4H(PO4)3·2.5H2O; Christoffersen et al., 1990], and ß-tricalcium phosphate [ß-Ca3(PO4)2; Kwon et al., 2003]. For phosphate adsorption to Fe and Al oxides, the reaction conditions were pH 6 (maintained with 2-morpholinoethanesulfonic acid buffer; Fisher Scientific), 8 g L–1 sorbent, ionic strength = 0.01 M (KCl), and 12 mM PO4 (KH2PO4; Oh et al., 1999). Ferrihydrite was synthesized by the method of Schwertmann and Cornell (1991) and amorphous Al oxides were synthesized by the method of Peak et al. (2002). The approximate concentrations of PO4 sorbed were 1425 and 1500 mmol kg–1 for ferrihydrite and amorphous Al hydroxide, respectively. Mineralogical purity of all standards was verified by XRD.
XANES Data Collection
Collection of all P K-XANES spectra was conducted at beamline X-19A of the National Synchrotron Light Source at the Brookhaven National Laboratory in Upton, NY. The electron storage ring was operated at 2.5 GeV with a maximum beam current of 300 mA. Beamline X-19A was equipped with a Si(III) monochromator and a He-filled sample chamber. Moist biosolids samples were analyzed by mounting the paste into a Plexiglas holder with a 1-mm-thick well and covering with 5-µm-thick polypropylene film, which was found to be free of detectable P using XANES analysis. Dried biosolids and manure samples were ground into a fine powder using an agate mortar and pestle and a thin layer was mounted on a Plexiglas sample holder using phosphorus-free cellophane tape. Standards were analyzed as ground powders with the exception of sorbed samples, which also were analyzed as moist pastes. Phosphorus K-XANES spectra from the study of Maguire et al. (2006) were also used in fitting analysis. These standards had been diluted to 400 mmol P kg–1 in BN to reduce the effects of self-absorbance during XANES data collection (Beauchemin et al., 2003; Hesterberg et al., 1999).
Spectra were collected in fluorescence mode using a solid-state passivated implanted planar silicon detector with a He flight path. The spectra were collected from 30 eV below the absorption edge to 100 eV above the absorption edge (
2120–2250 eV). The step size in the pre-edge region was 0.05 eV to resolve any pre-edge features that may exist due to the presence of Fe-associated P in the samples. Multiple scans (2–5) were collected for each sample and then averaged. The monochromator was calibrated to the maximum peak energy of the first derivative (E0 = 2149 eV) using a standard (for this study, brushite) at the start of every run and after every beam dump (usually every 12 h; Khare et al., 2004).
The energy scale was normalized by subtracting the calibration energy (E0 = 2149 eV) from all obtained spectra (Hesterberg et al., 1999). Linear baseline subtraction and normalization were performed for all data between –30 and –10 eV (baseline) and 30 and 50 eV (background) using Athena software (Ravel and Newville, 2005). Phosphorus K-XANES spectra from biosolids and manure samples were compared with spectra collected from synthesized or purchased mineral and organic P standards using the linear combination fitting tool in Athena. Spectra collected for moist biosolids samples were compared with spectra of wet paste standards when available. Linear combination fitting was conducted across the energy range of 2142 to 2179 eV. During fitting, E0 values were allowed to float up to ±1 eV because spectra collected in August 2005 were shifted approximately –1 eV compared with spectra collected in February and June 2005.
Additional XANES spectra for standards that were collected during previous studies were included in the linear combination fitting analysis. These standards included aqueous phosphate (HPO42–, H2PO4–, and H3PO4) collected using a Ge monochrometer and Ca phosphates (octacalcium phosphate, brushite, and monetite) diluted to 400 mmol P kg–1 for the study of Maguire et al. (2006). Due to energy calibration of these standards to the variscite edge, it was necessary to change the energy scale to align E0 of these spectra collected with E0 of the analogous samples from our current study. Because of a lack of comparable standards from our study, E0 values for the aqueous phosphate standards were aligned with that of our brushite standard, and the energy scale in the fitting was allowed to vary. Goodness of fit was evaluated using the residual factor (R-factor) and 
2 values generated by the linear combination fitting tool in Athena. The inclusion of each additional standard in a given fit was considered justified if it decreased the R-factor by 20% or more, resulting in linear combination fits with a maximum of four standards.
Statistical Analysis
Linear relationships between the XANES linear combination and chemical fractionation data were analyzed using PROC REG (SAS Institute, 2003).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
|
In general, the combination of P standards yielding the best linear combination fits for manures and lime-stabilized biosolids where no Fe had been added at the WWTP (B1 and B2) were PO4 sorbed to Al hydroxide, hydroxylapatite, and phytic acid (Fig. 2
–5). For lime-stabilized biosolids treated with FeCl3 (B3), however, the best fits were obtained with PO4 sorbed to ferrihydrite, hydroxylapatite, and phytic acid. The standards that gave the best fit for Fe-treated (B4 and B5) and digested (B6) biosolids were PO4 sorbed to ferrihydrite, hydroxylapatite, ß-tricalcium phosphate, and in many cases PO4 sorbed to Al hydroxide. Differences in the elemental composition of the organic P sources (Table 1) led to variations in the proportion of each P species (Table 2).
|
|
|
|
|
|
2 = 0.12 and 0.11, for DM-B and DM-D, respectively) of the fitting results shown in Fig. 2. Therefore, it is possible that a portion of total P in dried dairy manures and normal poultry litter samples that was fit as Al-hydroxide-bound P was actually a soluble hydrated salt (e.g., K or Na phosphate) or weakly sorbed P (e.g., outer sphere surface complexes), especially for samples with higher WSP/total P(TP) ratios (Table 1).
Previous XANES studies have reported that differences in the spectral features of aqueous P standards and PO4 sorbed to Al hydroxides were subtle (Peak et al., 2002). In fact, XANES results of Toor et al. (2005b) showed that 13 to 18% of total P (1.9–3.6 g kg–1) in dried broiler litters having WSP/TP ratios 
0.56 could be fit as aqueous phosphate; no aqueous P was fit in spectra for turkey manure samples with WSP/TP ratios 
0.36. Additionally, the XANES results of Staats (2005) showed that extraction of poultry litter samples with water eliminated the component(s) fit with an aqueous phosphate standard and standards of other weakly bound forms of inorganic P.
The amount of PO4 sorbed to Al hydroxides in alum-treated manures, digested biosolids, and Fe-treated biosolids, as determined by XANES fitting, was supported (stoichiometrically) by the molar ratio of Al/P in these samples (Table 1). Previous studies using XANES spectroscopy have also suggested the presence of a sorbed phase of P on solid-phase Al(OH)3 in alum-treated poultry litters (Hunger et al., 2004; Peak et al., 2002). It is unlikely that significant amounts of aqueous or weakly sorbed P exist in the APL and biosolids samples, as evidenced by the low WSP/TP ratios of these materials compared with untreated manures (Table 1). Linear combination fitting did not identify an Al-bound P phase in the B3 (lime + Fe) biosolids. This may be due to preferential sorption of P onto more abundant Fe hydroxides, as discussed by Khare et al. (2004), or to the presence of Al-hydroxide-sorbed P at concentrations that were not detectable by XANES analysis (10–20% of TP; Beauchemin et al., 2003).
Phosphate Sorbed to Ferrihydrite
Results of XANES linear combination fitting indicated the presence of PO4 sorbed to ferrihydrite in Fe-treated and digested biosolids (B3, B4, B5, and B6). In general, these samples had the highest Fe/P ratios (Table 1). Sorbed phases of P have not been identified in previous speciation studies of Fe-treated biosolids (Frossard et al., 1994; Huang and Shenker, 2004). This was probably due to the inability of solid-state 31P NMR to identify P associated with paramagnetic elements (e.g., Fe and Mn) and of XRD to identify noncrystalline or sorbed phases. Chemical fractionation studies, however, have repeatedly indicated an increase in NaOH-extractable P (attributed to Fe-associated P) when biosolids were treated with Fe salts (McCoy et al., 1986; Penn and Sims, 2002). There was no evidence in our data for precipitation of noncrystalline or crystalline Fe phosphate minerals, which give a stronger pre-edge peak in the XANES spectra than PO4 sorbed to ferrihydrite or any of the manures or biosolids samples (Hesterberg et al., 1999). These results indicate that the addition of Fe salts during wastewater treatment led to the precipitation of Fe (hydr)oxide solids that subsequently sorbed P. Phosphate sorbed to ferrihydrite accounted for lower percentages of total P in Fe-treated biosolids that were also lime stabilized (B3, 20% of total P) than digested and Fe-treated biosolids that did not receive lime (B4, B5, and B6, 55% of total P; Table 2). As discussed below, extraction of organic P sources with alkaline solution (NaOH) appeared to dissolve P associated with Fe (hydr)oxides, consistent with the lower P sorption capacity of Fe-oxide minerals at alkaline pH (Sparks, 2003). Also, lime-treated samples contained more Ca phosphate (Table 2), indicating that the high pH and high Ca/P ratio of these samples (Table 1) promoted Ca phosphate formation over sorption of P to Fe (hydr)oxide solids. Also, no Fe-P was identified in B2 biosolids (lime + Al) despite the high Fe/P ratio in this sample (Table 1); however, XANES analysis indicated that this sample contained a significant proportion of P associated with Al hydroxide, consistent with the findings of Peak et al. (2002) for alum-treated poultry wastes.
Hydroxylapatite and ß-Tricalcium Phosphate
Hydroxylapatite was identified by XANES fitting analysis as the dominant Ca-phosphate mineral in all dairy manures, poultry litters, and lime-stabilized biosolids and as the secondary Ca-phosphate mineral in Fe-treated and digested biosolids (B4, B5, and B6; Table 2, Fig. 2
–5). Toor et al. (2005b) also reported the presence of hydroxylapatite, in addition to dicalcium phosphate (monetite) in turkey manures with high Ca/P (>2) using XANES linear combination fitting. In general, the Ca/P ratio of dairy manures and lime-stabilized biosolids (but not poultry litters or digested and Fe-treated biosolids) are sufficiently high (>1.67) to account for the stoichiometric amounts of hydroxylapatite indicated by XANES spectral fitting for these materials (Table 1). In addition, lime-stabilized biosolids had an alkaline pH (pH > 10) and when coupled with high Ca/P ratios, these conditions should result in the precipitation of relatively insoluble Ca-phosphate minerals (Lindsay, 1979); however, the results of our study and the study by Toor et al. (2005b) differed from the findings of Huang and Shenker (2004), who failed to detect the presence of crystalline Ca-phosphate minerals in Ca-oxide-stabilized biosolids using XRD, despite high molar ratios of Ca to P (Ca/P = 2.3–17.7).
Linear combination fitting indicated the presence of hydroxylapatite, but not the presence of more soluble Ca-phosphate minerals, such as monetite, in samples with lower Ca/P ratios (NPL, APL, B4, B5, and B6). Other spectroscopic studies have documented the presence of monetite or brushite, but not hydroxylapatite, in normal and alum-treated poultry litters (Hunger et al., 2004; Peak et al., 2002; Toor et al., 2005b) and digested biosolids (Frossard et al., 1994). Unlike our study, Toor et al. (2005b) collected all XANES spectra in total electron yield mode, where self-absorption does not occur, and reported that the spectra for monetite had a stronger white-line peak and weaker postedge feature than hydroxylapatite. Our spectra for many of the undiluted Ca-phosphate standards demonstrated decreased white-line peaks, suggesting significant self-absorption when spectra were collected in fluorescence mode. However, when we performed the linear combination fitting using standards of monetite and brushite diluted in BN from the study of Maguire et al. (2006), results indicated the presence of these more soluble Ca-phosphate minerals in only one manure sample. Linear combination fitting for DM-D indicated the presence of brushite in addition to hydroxylapatite, but only when fitted in combination with aqueous H2PO4 instead of Al-bound P. The R-factor and 2 values for these fits were within 20% of those reported in Fig. 2 (R-factor = 1.5exp(–4); 2 = 0.12).
Linear combination fitting revealed that ß-tricalcium phosphate, a more soluble mineral than hydroxylapatite (Lindsay, 1979), was the primary Ca-phosphate mineral in Fe-treated and digested biosolids (Table 2, Fig. 2–5). These results were consistent with the work of Huang and Shenker (2004), who documented the presence of ß-tricalcium phosphate as the major Ca-phosphate mineral in digested and Fe-treated biosolids. Results were also partially supported by the work of Frossard et al. (1994), who suggested the presence of dehydrogenated condensed Ca-phosphate minerals (e.g., tricalcium phosphate and fluorapatite) in anaerobically digested biosolids.
Phytic Acid
Linear combination fitting indicated that phytic acid, an organic P species, was a main component (7–34% of total P) in dairy manures, poultry litters, and lime-stabilized biosolids (Fig. 2–5, Table 2); however, it was only a minor component of Fe-treated or digested biosolids. The fraction of total P determined by fitting to be phytic acid in the lime-stabilized biosolids was 8 to 15% of total P. The presence of phytate in biosolids is probably because these materials are the result of residential and industrial wastewater treatment. Phytate, a common dietary component in human foods, passes through the small intestines of humans undigested due to low intestinal phytase activity (Iqbal et al., 1994). Phytic acid accounted for 18 to 30% of total P in poultry manure samples (Table 2). These results were similar to those of Toor et al. (2005b), who reported, based on XANES spectroscopy, that broiler litter samples contained 7 to 20% phytic acid. Studies using solution 31P NMR indicated that, on average, phytic acid in poultry litter accounted for 56 to 59% of total P, and other organic P species in theses samples were present in trace amounts (Maguire et al., 2004; McGrath et al., 2005; Turner and Leytem, 2004). These reported differences in the phytic acid content of poultry litter could be a result of differences in poultry diet or storage practices (McGrath et al., 2005), or an inconsistency between XANES and NMR results for organic P in poultry manure samples.
The XANES fitting results suggested that dairy manures contained 11 to 34% of total P as phytic acid, with a mean value of 22 ± 8% (Table 2). Five of the six dairy manures examined in this study (DM-B–DM-F) were the same manure samples that were analyzed using solution 31P NMR by Toor et al. (2005a). Analysis of the dairy manure samples using 31P NMR suggested lower proportions of phytic acid (12%) than those reported using XANES analysis (22%). Since phytic acid was the only organic P standard used in our study, it is probable that other organic P species accounted for a portion of total P that was identified as phytic acid for the dairy manure and lime-treated biosolids samples. In fact, solution 31P NMR data of Toor et al. (2005a) indicated that the sum of all organic P forms in dairy manure (including phospholipids, deoxyribonucleic acid and other orthophosphate mono- and diesters, and phytic acid) represented, on average, 22% of total P in the samples DM-B, DM-C, DM-D, DM-E, and DM-F that were also sampled for our study. It is uncertain, however, that XANES fitting with additional organic P standards would have improved our results because the XANES spectra of organic P species have been shown to lack distinguishable features, such as the pre- or postedge features demonstrated in the XANES spectra of inorganic P species (Peak et al., 2002; Staats, 2005). Moreover, Staats (2005) reported that the XANES spectrum of phytic acid (Na salt) could be fit using the spectra of orthophosphate diesters or phospholipids and (to a lesser extent) Ca-phosphate standards. It is clear that XANES spectroscopy is not the most sensitive technique for identifying or distinguishing between organic P species, and we suggest that solution 31P NMR analysis should be used to obtain the most accurate picture of the organic P forms in biosolids and manures.
Effect of Drying on Phosphorus Speciation in Biosolids
Oven drying of biosolids samples resulted in some differences between the proportions of P species in dry and moist biosolids samples as quantified by XANES fitting analysis. The linear combination fits for dry, lime-stabilized biosolids indicated, on average, 5% more hydroxylapatite and 5% less Al-hydroxide-sorbed P then fits for the moist samples. Fits for the relative proportion of phytic acid remained the same in both moist and oven-dried samples. In addition, the linear combination fits for dry digested or Fe-treated biosolids samples identified, on average, 25% more PO4 adsorbed on ferrihydrite, 9% less ß-tricalcium P, and 15% less Al-bound P than fits for the moist samples. The B4 (Fe) biosolids was the only material where linear combination fitting of the moist sample indicated the presence of a P species (Al-bound P) that was not identified in the dry sample. In most cases, the difference in the proportion of individual P species in wet and dry samples was lower than the reliability of XANES linear combination fitting (10–20% of total P; Beauchemin et al., 2003). Although differences in the proportions of P species in dry and moist biosolids could be affected by sample heterogeneity, it is important to note that other studies have reported changes in the distribution of P in biosolids and manures as a result of oven drying (Ajiboye et al., 2004; Chapuis-Lardy et al., 2004). Therefore, drying of such samples before speciation analysis is not recommended.
Chemical Fractionation of Phosphorus in Manures and Biosolids
Chemical fractionation showed, as in past research (Ajiboye et al., 2004; He et al., 2004; Sharpley and Moyer, 2000), that readily soluble P (NH4Cl-extractable P) was the greatest fraction of TP in DM-A (37%) and NPL (45%; Table 3). Cumulatively, extraction with NH4F and NaOH removed 34 and 18%, respectively, of TP in these organic P sources, while only a small proportion of total P was extracted by DCB or H2SO4 (9 and 18%, respectively; Table 3). In contrast, APL had lower concentrations of NH4Cl-extractable P and higher NH4F-extractable P and DCB-extractable P when compared with untreated manures (Table 3). These results are consistent with numerous studies of the effects of alum on P solubility (Moore et al., 2000; Sims and Luka-McCafferty, 2002) and speciation (Hunger et al., 2004; Peak et al., 2002) in poultry litter. Concentrations of P extracted from APL by NaOH and H2SO4 were similar to the untreated manures.
|
The value of sequential chemical fractionation studies to determine the forms of P in organic P sources is often questioned because these techniques are destructive and operational, i.e., they do not provide information about specific chemical forms of P in manures, biosolids, or amended soils. While we recognize the limitations of chemical fractionation, we believe that it can be useful, particularly if interpreted as a complement to more definitive P speciation methods, such as XANES spectroscopy. In fact, we found several significant relationships between the P species identified using XANES linear combination fitting and P extracted during different steps of the sequential fractionation (Fig. 6). The proportion of TP sorbed to Al hydroxides in the organic P sources, as determined by XANES, was significantly related to the sum of NH4Cl + NH4F-extractable P (r2 = 0.67 [P = 0.01]; Fig. 6a). These results suggest that NH4Cl was targeting weakly sorbed forms of P associated with Al(III), while reaction with NH4F led to the release of P from noncrystalline Al oxides by ligand exchange. The amount of P extracted by NH4Cl and NH4F, however, overestimated the amount of Al-bound P identified using XANES (regression slope = 0.68; Fig. 6a), indicating that other forms of P were also being removed during these first two extraction steps.
|
|
CONCLUSIONS AND IMPLICATIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
We also found that recalcitrant Ca-phosphate minerals (e.g., hydroxylapatite, tricalcium phosphate) were a major component of inorganic P in lime-stabilized biosolids and manures. These results are consistent with data showing that additions of lime effectively decrease the solubility of P in organic P sources (Maguire et al., 2006). Lime addition increases the Ca/P ratio and pH of these materials, favoring the formation of hydroxylapatite. Manipulation of the Ca/P ratio has previously been suggested as a method to reduce the P solubility of animal manures (Toor et al., 2005b). Hydroxylapatite should remain insoluble when these materials are surface applied or incorporated into calcareous soils, provided the pH remains >7. In contrast, when manures or lime-stabilized biosolids are incorporated into mid-Atlantic soils, which are typically moderately acidic (pH 5.5–6.5), pH-promoted dissolution of Ca-phosphate minerals would be expected as the pH decreases. Hydroxylapatite in moderately acidic soils would be substantially less soluble than monetite, and hydroxylapatite has been identified in P-enriched, acidic soil samples (Beauchemin et al., 2003). In contrast, under acidic (and aerobic) conditions, the stability of Fe- and Al-oxyhydroxide minerals would be greater than that of Ca phosphate (Lindsay, 1979). It is also important to note that the addition of lime, in conjunction with Fe salts, was shown to decrease the amount of PO4 sorbed to Fe hydroxides and increase the amount of hydroxylapatite in biosolids samples. The incorporation of Fe- and lime-treated biosolids into acid soils will probably lead to the slow dissolution of Ca phosphates; however, the large pool of biosolids-derived Fe hydroxides could ultimately act as a sorbent for the P released during Ca-phosphate dissolution. Overall, our results indicate that addition of Fe or Al salts to manures or during wastewater treatment will provide the best long-term reduction of dissolved P losses when these materials are applied to acidic mid-Atlantic soils under aerobic conditions.
Finally, we found significant linear relationships between specific forms of P (or combinations of species) reported using XANES linear combination fitting and the amount of P extracted during specific chemical fractionation steps; however, there was usually not a 1:1 relationship. Chemical fractionation involves sample destruction, and our results indicated that a particular chemical extractant tended to extract either part or some combination of the P species quantified from linear combination fitting of XANES data. In addition, specific chemical extractants (or combinations of extractants) often over- or underestimated the amount of P species determined from XANES fitting. Nevertheless, chemical fractionation techniques can show that there are differences in the chemical forms of P between different manures and biosolids, but their relationship to specific chemical species in a sample must be interpreted with caution. The growing research body on P speciation analyzed using nondestructive physical techniques such as XANES and NMR spectroscopies provides a better framework for interpreting chemical fractionation results.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Z. He, C. W. Honeycutt, T. S. Griffin, B. J. Cade-Menun, P. J. Pellechia, and Z. Dou Phosphorus Forms in Conventional and Organic Dairy Manure Identified by Solution and Solid State P-31 NMR Spectroscopy J. Environ. Qual., July 23, 2009; 38(5): 1909 - 1918. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Smith, J. W. White, and F. J. Coale Evaluation of Phosphorus Source Coefficients as Predictors of Runoff Phosphorus Concentrations J. Environ. Qual., February 6, 2009; 38(2): 587 - 597. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Shober and J. T. Sims Evaluating Phosphorus Release from Biosolids and Manure-Amended Soils under Anoxic Conditions J. Environ. Qual., January 13, 2009; 38(1): 309 - 318. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Ajiboye, O. O. Akinremi, Y. Hu, and A. Jurgensen XANES Speciation of Phosphorus in Organically Amended and Fertilized Vertisol and Mollisol Soil Sci. Soc. Am. J., September 1, 2008; 72(5): 1256 - 1262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hunger, J. T. Sims, and D. L. Sparks Evidence for Struvite in Poultry Litter: Effect of Storage and Drying J. Environ. Qual., June 23, 2008; 37(4): 1617 - 1625. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Seiter, K. E. Staats-Borda, M. Ginder-Vogel, and D. L. Sparks XANES Spectroscopic Analysis of Phosphorus Speciation in Alum-Amended Poultry Litter J. Environ. Qual., March 1, 2008; 37(2): 477 - 485. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Gungor, A. Jurgensen, and K. G. Karthikeyan Determination of Phosphorus Speciation in Dairy Manure using XRD and XANES Spectroscopy J. Environ. Qual., October 24, 2007; 36(6): 1856 - 1863. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Ajiboye, O. O. Akinremi, Y. Hu, and D. N. Flaten Phosphorus Speciation of Sequential Extracts of Organic Amendments Using Nuclear Magnetic Resonance and X-ray Absorption Near-Edge Structure Spectroscopies J. Environ. Qual., October 16, 2007; 36(6): 1563 - 1576. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Shober and J. T. Sims Integrating Phosphorus Source and Soil Properties into Risk Assessments for Phosphorus Loss Soil Sci. Soc. Am. J., March 12, 2007; 71(2): 551 - 560. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
