JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 16 October 2007
Published in J Environ Qual 36:1563-1576 (2007)
DOI: 10.2134/jeq2006.0541
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Supplements
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ajiboye, B.
Right arrow Articles by Flaten, D. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ajiboye, B.
Right arrow Articles by Flaten, D. N.
Agricola
Right arrow Articles by Ajiboye, B.
Right arrow Articles by Flaten, D. N.
Related Collections
Right arrow Phosphorus
Right arrow Nutrient Management
Right arrow Other Waste Management

REVIEWS & ANALYSES

Phosphorus Speciation of Sequential Extracts of Organic Amendments Using Nuclear Magnetic Resonance and X-ray Absorption Near-Edge Structure Spectroscopies

Babasola Ajiboyea,c,*, Olalekan O. Akinremia,*, Yongfeng Hub and Donald N. Flatena

a Dep. of Soil Science, Univ. of Manitoba, Ellis Building, 13 Freedman Crescent, Winnipeg, MB, Canada R3T 2N2
b Canadian Light Source Inc., Univ. of Saskatchewan, 101 Perimeter Rd., Saskatoon, SK, Canada S7N 0X4
c current address, Canadian Light Source Inc., Univ. of Saskatchewan, 101 Perimeter Rd., Saskatoon, SK, Canada S7N 0X4

* Corresponding author (Sola.Ajiboye{at}lightsource.ca, Akinremi{at}ms.umanitoba.ca).

Received for publication December 15, 2006.
ABSTRACT

The chemical forms of phosphorus in organic amendments are essential variables for proper management of these amendments for agro-environmental purposes. This study was performed to elucidate the forms of phosphorus in various organic amendments using state-of-the-art spectroscopic techniques. Anaerobically digested biosolids (BIO), hog (HOG), dairy (DAIRY), beef (BEEF), and poultry (POULTRY) manures were subjected to sequential extraction. The extracts and residues after extraction were analyzed by solution 31P nuclear magnetic resonance (NMR) and synchrotron-based P 1s X-ray absorption near-edge structure (XANES) spectroscopies, respectively. Most of the total P analyzed by inductively coupled plasma– optical emission spectroscopy in the sequential extracts of organic amendments was orthophosphate, except POULTRY, which was dominated by organic P. The labile P fraction in all the organic amendments, excluding POULTRY, was mainly orthophosphate from readily soluble calcium and some aluminum phosphates. In the poultry litter, Ca phytate was the main P species controlling P solubility. The recalcitrant fraction of BIO was mainly associated with Al and Fe. Those of HOG, DAIRY, and POULTRY were calcium phytate, which were identified only as organic species in the XANES spectra. The combination of the three techniques—sequential chemical extraction, solution 31P NMR spectroscopy, and P 1s XANES—provided molecular characterization of P in organic amendments that would not have been possible with just one or a combination of any two of these techniques. Therefore, P speciation of organic amendments should use solid-phase and aqueous speciation techniques as deemed feasible.

Abbreviations: BEEF, beef cattle manure • BIO, biosolids • DAIRY, dairy cattle manure • DCP, dicalcium phosphate • DCPD, dicalcium phosphate dihydrate • FY, fluorescence yield • HAP, hydroxyapatite • HOG, liquid hog manure • ICP-P, P measured by inductively coupled plasma–optical emission spectroscopy • LC, linear combination • NMR, nuclear magnetic resonance • PHYTIC, phytic acid-Ca salt • POULTRY, poultry litters • RSD, relative standard deviation • STRUV, struvite • XANES, X-ray absorption near-edge structure

THE speciation of phosphorus (P) in organic amendments like manures and biosolids (BIO) is important for understanding P solubility when these amendments are added to the soil. This characterization is essential in mitigating the adverse environmental impact of P on surface water eutrophication. The simple characterization of phosphates in manure into labile and non-labile forms, based on their solubility in extractants of increasing strengths according to sequential extraction technique, does not provide direct information on the structure and chemical form of the various P compounds in the manure (Leytem et al., 2004). In addition, the reported discrepancies between the inductively coupled plasma–optical emission spectroscopy (ICP–OES) and colorimetry procedure commonly used for the measurement of soluble and total P in these extracts make subsequent interpretation of results and P management recommendation difficult (Pierzynski et al., 2005). Furthermore, specific groups of compounds "operationally" assigned to a particular fraction may be present in more than one fraction (Turner et al., 2005). There is an increasing need to move away from this operationally defined characterization of P to a more detailed molecular and structural characterization. For example, Maguire et al. (2004) argued that applying the sequential extraction method to manure without knowing the forms of P extracted at each step may create a potential problem in interpretation due to the extraction of phytate P and organic P by extractants other than NaOH. Chen et al. (2002) argued that characterization of organic P pool into bioavailable P based on solubility may be misleading because plant can obtain P from the supposedly stable organic P fractions.

Detailed studies of inorganic and organic P pool in manures have been performed recently using solution 31P nuclear magnetic resonance (NMR) on alkaline extracts of manures (Cade-Menun and Preston, 1996; Turner, 2004; Turner and McKelvie, 2002; Toor et al., 2005a; Turner et al., 2005). This technique provided direct molecular and structural characterization of organic P in alkaline solution with environmental relevance. For example, stable organic P species like phytic acid can be easily quantified and distinguished from the more labile organic species like phospholipids and other orthophosphate monoesters in NMR spectra (Turner et al., 2005; Toor et al., 2005a). However, there are several caveats in the interpretation of solution 31P NMR result. First, solution NMR does not provide direct information about the mineral phase of P, which is identified only as orthophosphate due to solubilization by the extractants. In addition, hydrolysis of some labile organic P species into orthophosphate in the alkaline extract may lead to the underestimation of organic P pools in the manure (Turner, 2004; Turner and Leytem, 2004). Finally, the recovery of polyvalent cations in the extract used for NMR analysis does not provide a direct evidence of their association with the organic or inorganic P pools. Therefore, combining solution NMR with other solid-phase spectroscopic methods may be necessary to augment the understanding of P pools in manures.

The use of X-ray absorption near-edge structure (XANES) spectroscopy to characterize P pool in manure is advantageous over other methods involving chemical extraction for several reasons. For example, minimal sample preparation is required to isolate an element of interest, and the possibility of transforming phosphate species with extractants is eliminated. In addition, the near-edge region of this technique provides information on the local chemical and structural environment and oxidation states of the element, and, if the probing is performed at a wide energy range above the binding energy of the element of interest, it allows for determination of co-ordination numbers and bond distances (Hesterberg et al., 1999). The capability of XANES spectroscopy to identify P species in soils and manures has been recently demonstrated (Peak et al., 2002; Beauchemin et al., 2003; Toor et al., 2005b; Sato et al., 2005). For example, Toor et al. (2005b) identified dicalcium phosphate (DCP) as the dominant P species in litter of broilers fed with normal diet or reduced non-phytate P (NPP) and a mix of DCP and hydroxyapatite (HAP) in litters of turkey fed with NPP. These authors also reported that water removed a part of the DCP in broiler litter, whereas HCl was effective in removing phytic acid in broiler litter and a mix of phytic acid and HAP in the turkey litter with reduced NPP. Sato et al. (2005) reported that weakly adsorbed and relatively soluble P species like DCP dominated the P 1s XANES spectrum of poultry manure. These authors also observed that application of this poultry manure to an acidic soil, in the short-term, resulted in dissolution of FeP from the soil, adsorption of P onto the surface of other minerals, and formation of DCP; prolonged application resulted in the disappearance of FeP and formation of more stable tricalcium phosphate.

Hunger et al. (2004) attempted to characterize P in alum-amended poultry litter before and after water extraction alone using solid-state 31P NMR. However, these authors suggested a combination of liquid-state and solid-state NMR spectroscopy due to the inability of the latter to identify organic P species. A direct link between dissolved P species identified by NMR and those remaining in all the residues of all sequential extraction stages has not been fully investigated. Our premise is that identification of P species that are removed by a particular extractant or those remaining in the residues of the sequential extraction procedure by NMR and XANES will be a significant advancement to the knowledge of sequential extraction technique and may improve the characterization of P into labile and non-labile fractions of environmental relevance.

The objective of this study was to combine the sequential chemical extraction with solution 31P NMR and XANES to provide a detailed molecular speciation of P in organic amendments.

Materials and Methods

Chemical Extractions
Five types of organic amendments comprising of biosolids; hog, dairy cattle, and beef cattle manures; and poultry litter were used for this study. The description and location of these amendments are given in Table 1 . The lyophilized amendments were analyzed for total P and cations Fe, Al, Ca, and Mg by inductively coupled plasma–optical emission spectroscopy (ICP–OES) after H2SO4–H2O2 digestion (Akinremi et al., 2003) and reported along with the description in Table 1. Six replicates of each amendment were subjected to sequential chemical extraction using progressively stronger extractants according to Ajiboye et al. (2004). In summary, a 0.5-g (over dry weight basis) sample of organic amendments was extracted in sequence with 30 mL of deionized water, 0.5 M NaHCO3 (pH 8.5), 0.1 M NaOH, and 1 M HCl on an end-to-end shaker at 150 excursions per minute for 16 h at room temperature. The extract was centrifuged at 10,000 x g for 15 min and suction-filtered using a 0.45-µm cellulose membrane. Dilute HCl was added to the NaHCO3 extracts to dissolve the carbonates, and 1 M or 10 M NaOH was added to the HCl extract to neutralize the carbonic acid formed on dissolution of the carbonates in the manure. One set of three replicates of filtered extracts from each extraction stage was analyzed for total P and cations Fe, Al, Ca, and Mg using ICP–OES (Akinremi et al., 2003). Inorganic P was determined by molybdate-blue colorimetry according to Murphy and Riley (1962) on a Ultrospec 3100 pro UV/visible spectrophotometer (Biochrom, Cambridge, UK) at a wavelength of 882 nm, and organic P was determined as the difference between total P and inorganic P (Ajiboye et al., 2004). The other set of three replicates was rapidly frozen at –81°C using a Fisher ultra-low temperature freezer and lyophilized using a ModulyoD freeze dryer (Thermo Electron Corp., Milford, MA). The residue remaining after each extraction stage was lyophilized for XANES analysis.


View this table:
[in this window]
[in a new window]

  		Table 1. Description and chemical properties of organic amendments used.

 
NMR Analysis
The solution 31P NMR analysis was performed using a modification of the Turner and Leytem (2004) method. Specifically, 0.25 g of lyophilized powder from sequential extraction stages was redissolved in 4.5 mL 1.0M NaOH–0.1M EDTA with 0.5 mL D2O and centrifuged at approximately 12,000 x g for 15 min. The supernatant was pipetted into a 10-mm NMR tube. The NMR spectra of the reconstituted samples were acquired on Bruker AMX 500 MHz spectrometer with a 10-mm broadband probe operating at 202.456 Hz for 31P and 500.134 for 1H (for broadband proton decoupling). A 45–free induction decay pulse sequence and a relaxation delay of 2.0 s were used. To verify that this relaxation delay was sufficient for quantitative analysis of the spectra, a spin-lattice relaxation time (T1) experiment was performed on two of the extracts with the extremes of Fe content. This involved a modification of the inversion-recovery pulse sequence described by McDowell et al. (2006). The net magnetization vector was inverted with 180° pulse, and the recovery was monitored by waiting a variable time {tau} (0.001, 0.0625, 0.125, 0.25, 0.4, 0.6, 1.0, 2.0, 4.0, and 8.0 s) followed by a 90° pulse. These signals were acquired for 0.8 s after a relaxation delay of 20 s. A total of 256 scans were collected at each {tau}. The T1 was calculated for each of the identified peaks by regression analysis of Eq. [1].

Formula 1[1]

where I[{tau}] is the intensity of the signal at each {tau}, I[0] is the final intensity, and P is the magnitude of the magnetization vector. The result showed that delay used in the current experiment was sufficient for most P species in the extracts, except for orthophosphate (Table 2 ). Further experiments with 2.0-s and 20-s delays (~3 x T1 for orthophosphate peak) for the two selected extracts with the highest and the lowest Fe concentration showed that the relaxation delay used in the current experiment did not interfere with peak identification and subsequent quantitation (Fig. 1 ). Other studies have shown that the error associated with not meeting the T1 for orthophosphate was less than 10% (McDowell and Stewart, 2005a; McDowell et al., 2006) and did not interfere with quantitative analysis.


View this table:
[in this window]
[in a new window]

  		Table 2. Calculated spin-lattice (T1) relaxation time of two sequential extracts representing the two extremes of Fe content. The T1 was measured with an inversion-recovery pulse sequence according to McDowell et al. (2006) but with a relaxation delay of 20 s instead of 8 s and more variable time.

 

Figure 1
View larger version (9K):
[in this window]
[in a new window]

  		Fig. 1. Solution 31P nuclear magnetic resonance spectra of the selected extract with the highest Fe concentration (HCl extracts of hog manure) acquired using (A) 2-s relaxation delay and (B) 20-s relaxation delay and those of the selected extract with the least Fe concentration (HCl extracts of poultry litter) acquired using (C) 2-s relaxation delay and (D) 20-s relaxation delay. A total of 3000 scans were acquired in both cases. Spectra were processed with 3 Hz line broadening. The proportions of the identified peaks are listed in the figures.

 
A total of 10,000 to 33,000 scans were acquired for the amendments with acquisition time of 0.4 s per scan. The temperature of the probe ranged from 20 to 25°C. The chemical shifts of signals were expressed in ppm relative to an external standard (85% H3PO4), and all spectra of manures were plotted with line broadening of approximately 2 Hz to show the relevant peaks. The peaks in the spectra were assigned to P species and organic P functional groups according to the literature values (Turner, 2004; Turner and Leytem, 2004; Cade-Menun, 2005). The proportion of identified peaks relative to the total P in the solution was estimated by peak integration. Spinning sidebands, where present, were excluded from the peak integration. For the phytic acid, the peak area was calculated for well resolved spectra by summing up the four peaks in the ratio 1:2:2:1 assigned to P nuclei on C-2, C-4+C-6, C-1+C-3, and C-5 positions of the inositol ring, respectively. In cases where the C-2 peak overlapped with the orthophosphate peak, the peak was calculated by multiplying the area of the C-4+C-6 peak by 3 because this is equal to one third of the total concentration of phytic acid. All spectra were processed using SpinWorks 2.5.4 software (Marat, 2006). Statistical analysis was not conducted because the high cost of running the NMR spectrometer prevented replication of measurements. However, in a previous study, the standard errors associated with estimating the proportion of P species by peak integration were approximately 5 and 10% for large and small peaks, respectively (Leinweber et al., 1997).

XANES Analysis
The P 1s XAS experiment was performed on the double crystal monochromator beamline at the Synchrotron Radiation Centre, University of Wisconsin-Madison, housing the Aladdin storage ring, which operates at electron beam energy of 800 MeV or 1 GeV. The monochromator of this beamline is equipped with two InSb crystals with a photon resolution of approximately 0.9 eV at around 2150 eV photon energy. The beamline was calibrated for P K-edge by setting the absorption edge (due to P 1s -> 3p transition) of sodium pyrophosphate (Na4P2O7) at 2152.4 eV on the energy scale (Lombi et al., 2006). The lyophilized residues and phosphate standards were ground to powder, thinly spread over double-sided conducting C tape, and mounted on a stainless steel sample holder. Reference phosphate compounds analyzed included FeP, CaP, AlP, MgP, NH4P, NaP, and organic P. These compounds represent well-thought-out samples of P species that we know may be present in our samples and were obtained from different chemical manufacturers. Adsorbed P on CaCO3, iron hydroxides, and gibbsite were not included among the reference compounds analyzed due to the lack of distinguishing features in their XANES spectra (Sato et al., 2005). The mounted sample was transferred via a load-lock system into the high vacuum (10–7 mm Hg) absorption chamber equipped with a nine-element, solid-state Ge detector for fluorescence yield (FY) measurements. Two or three scans of P 1s XANES spectra of samples and reference standards were acquired in total electron yield and FY mode simultaneously, from 2140 to 2200 eV with a step size of 0.25 eV and a dwell time of 3.0 s pt–1. Only FY data are reported here because of the higher detection limit of Ge detector.

The energy scale of the P 1s XANES spectrum was recalibrated by using the edge energy offset between the energy of the white line of Na4P2O7 analyzed with the sample and that recorded during the calibration of the beamline. This offset value was used as a pre-processing parameter for all other spectra. The data were averaged and background corrected by a linear regression fit through the pre-edge region and a cubic spline through post-edge region. The spectra were normalized to a unit edge jump. All the data reductions were performed using Athena ver 0.8.048 (Ravel and Newville, 2005). Quantitative XANES analysis was performed using linear combination (LC) fitting alone.

In other XANES studies reported in the literature, the number of components used in LC fitting was determined by principal component analysis (PCA) and target transformation (TT) (Beauchemin et al., 2002, 2003; Ressler et al., 2000; Toor et al., 2005b), by the maximum number of components whose successive inclusion decrease the residual factor of the fit by 20% (Shober et al., 2006), or by using different combinations of all the standards (Sato et al., 2005). In the current study, the presence of different dominant P species in the organic amendments identified by the "fingerprinting" approach precluded the use of PCA and TT for quantitative analysis (data not shown). In addition, a key limitation of PCA and TT is that twice as many spectra as dominant components present in the set of spectra are needed to confidently determine the number of "target species" (Ressler, 2004). Therefore, a maximum of four standards was used for the LC fitting in the current study because it is unlikely that PCA would have identified more than four orthogonal components as sufficient to reconstruct the five XANES spectra of organic amendments. Linear combination of binary to quaternary combinations of all reference P compounds was then performed over the relative energy range of –11 to 49 eV. The threshold energy, E0, was allowed to vary during the fitting, but the maximum energy shift observed for any of the components in the best fit selected was much smaller than the step size (0.25 eV) of the scan. The goodness-of-fit was judged by the residual factor (R-factor) and {chi}2 values. The fit with the least R-factor and {chi}2 and the least SD of the estimated proportion was chosen as the best fit.

The P species removed by each extractant was identified by using the point-by-point difference between the XANES spectrum of any given residue and that of the next residue in the extraction sequence. For example, the difference between XANES spectra of unextracted (intact) and that of residue after water extraction represents the P species extracted by water.

Results and Discussion

Phosphorus and Accompanying Cations in Sequential Extracts
The total P in the organic amendments ranged from 2.2 g kg–1 in BEEF to 39.8 g kg–1 in HOG (Table 1). There were differences in the amount of P recovered in the sequential fractions among the organic amendments. In BIO, most of the P was recovered in NaOH (40% of total P) and HCl (42%) fractions, with lesser amounts in the H2O (2%), NaHCO3 (12%), and residual (3%) fractions (Table 3 ). In HOG, most of the P was found in the H2O (25%), NaHCO3 (22%), and HCl (28%) extracts, with lesser amounts in NaOH (7%) and residual (18%) fractions. Most of the P in DAIRY was in the NaHCO3 (35%) and residual (29%) fractions, with lesser amounts in the H2O (18%), NaOH (10%), and HCl (8%) fractions. Most of the P in BEEF was in the H2O (53%) and NaHCO3 (35%) fractions, with lesser amounts in NaOH (8%), HCl (4%), and residual fractions. Most of the P in POULTRY was partitioned between H2O (36%) and HCl (36%) fractions, with lesser amounts in NaOH (12%), NaHCO3 (9%), and residual fractions. By summing up the P extracted in the sequential fractions from H2O to HCl, the percent P recovery was 97% in BIO, 82% in HOG, 71% in DAIRY, 99% in BEEF, and 94% in POULTRY.


View this table:
[in this window]
[in a new window]

  		Table 3. Phosphorus fractions (g kg–1) and accompanying cations in sequential extracts of manures. Values in parentheses are % relative standard deviation.

 
The amount of the total P, as measured by ICP (ICP-P), in the sequential extracts indicated that BIO contained a smaller quantity of labile P and more non-labile P compared with other organic amendments. These results corroborate those reported by Ajiboye et al. (2004), in which biosolids contained a significantly lower H2O + NaHCO3–P (24% of total extractable P) compared with manures (63% in hog manure and 70 and 60% in dairy and beef cattle manures, respectively).

The molybdate-blue P, considered as inorganic P, was greater than ICP-P in the H2O fractions of BIO and HOG and in the HCl fractions of BIO, HOG, and BEEF, resulting in a negative value for organic P estimated as the difference between ICP-P and molybdate P (Table 3). The magnitude of this overestimation of inorganic P was highest in the HCl fraction of BIO, and if all the sequentially extracted P fractions were summed up (excluding the residual), the overestimation would be at least 11% in BIO, which is more than the reported CVs for ICP-P in each sequential extracts (Table 3). For extracts where overestimation of molybdate P occurred, the relative standard deviation (RSD) was comparable between the two methods (ICP and molybdate-blue) except for the HCl fraction of BEEF (Table 3). The highest % RSD values for ICP-P and molybdate P in HOG indicate that this sample is more heterogeneous than other organic amendments. The apparent overestimation of inorganic P relative to total P by the colorimetric procedure in the H2O fractions of BIO and HOG and the HCl fraction of BIO, HOG, and BEEF could not be attributed to heterogeneity of the samples alone, as suggested by the similarity in the RSD of the analytical methods. This overestimation may also be due to the hydrolysis of some organic and condensed P as observed in the two-step extraction with NaHCO3 and HCl (Turner and Leytem, 2004). However, it is possible that neutralization of the HCl extracts caused a re-precipitation of orthophosphate as Ca-P, hence an underestimation of total P by ICP. An overestimation of inorganic phosphorus in water and 0.01 M CaCl2 extracts of manure had been reported recently (Choate, 2004; Ajiboye and Akinremi, 2005; Wolf et al., 2005). The magnitude of this overestimation averaged at least 7% in water extracts of swine, dairy, and poultry manures (Wolf et al., 2005) and 13% in the CaCl2 extract of oven-dried biosolids (Choate, 2004). Further investigation into the cause of this inconsistency is needed. The magnitude and cause of the discrepancy and the factors contributing to higher ICP than colorimetry were recently identified as an important area of research necessary for generating correction factor and interconversion from ICP-P to molybdate reactive P (Pierzynski et al., 2005).

In the case of associated cations, a quantifiable amount of Fe was detected only in BIO, whereas Al was detected in BIO and BEEF. Total Fe and Al were not detected in other manures at similar dilutions used for total P analysis, but all the manures contained considerable amounts of total Ca and Mg (Table 1). The cations in the organic amendments were partitioned among the different fractions. In the BIO, the H2O fraction was dominated by Ca and Mg, which were about 3.5 and 4.5% of the total Ca and Mg, respectively (Table 3). Furthermore, the NaHCO3 fraction contained more Ca than Mg and Fe, in absolute terms, but the amount of Fe extracted by NaHCO3, relative to the total, was higher (10%) than those of Ca and Mg (~3%). In the NaOH fraction, Al was extracted in higher proportion than Fe and Ca, whereas in the HCl fraction, Ca was the only cation identified and it constituted about 74% of total Ca in BIO. Similar to BIO, the H2O and NaHCO3 fractions of HOG were dominated by Ca (3–6% of total) and Mg (30– 35% of total), whereas Al and Ca were the dominant cations in the NaOH fraction, but they constitute less than 1% of total Al and Ca. Substantial concentrations of all polyvalent cations were detected in the HCl fraction of HOG, even though Al and Fe were not quantified at the dilution used for the analyses of total P and cations (Table 3). In the DAIRY, Ca and Mg dominated the H2O and NaHCO3 fractions with a cumulative amount equal to 28% of total Ca and 94% of total Mg. Some trace amount of Fe was found in the NaHCO3 fraction. In the NaOH fraction of DAIRY, Al, Fe, and Ca were extracted in quantifiable amounts, but only Ca and Mg were found in the HCl fraction.

The higher amount of Ca extracted in the HCl fractions of BEEF and POULTRY and Mg in BEEF compared with the total Ca and Mg makes the expression of these cations relative to total P confounding. However, it is sufficient to say that Ca and Mg dominated the H2O and NaHCO3 fractions of BEEF, although Al and Fe were present in quantifiable amounts in the H2O fraction. Calcium and trace amounts of Al were found in the NaOH fraction, whereas all the cations were detected in the HCl fraction with Ca and Mg more than others (Table 3). The POULTRY contained predominantly Ca and Mg in the H2O, NaHCO3, and HCl fractions but contained predominantly Al and Ca in the NaOH fraction (Table 3). The higher amount of Al and Fe in the NaOH extracts of BIO relative to other amendments indicated that part of this "operationally defined" non-labile P fraction was indeed associated with Al and Fe. The extraction of the labile P fractions with Ca in all organic amendments suggested the presence of readily soluble calcium phosphates. The fact that HCl extracted the greatest amount of P in all organic amendments compared with other fractions suggested the presence of less soluble calcium phosphates like HAP that could not be extracted with the weaker extractants. Similar to the results of the current study, Cooperband and Good (2002) reported that Ca and Mg are the dominant cations controlling P solubility in manures.

Identification of Phosphorus Species in Sequential Extracts by NMR
In the BIO sample, inorganic orthophosphate was the dominant P species identified in all the sequential extracts, accounting for 100% of the extracted P in H2O and NaHCO3 fractions and approximately 95 and 98% in the NaOH and HCl fractions, respectively (Fig. 2 ). Phytic acid, which is a major orthophosphate monoester, was not identified in any of the sequential fractions of BIO. In the NaOH fraction, phosphatidic acid and ß-glycerophosphate were present. These species are considered to be hydrolytic products of phospholipids, such as phosphatidic choline in alkaline extracts (Turner and Leytem, 2004). Pyrophosphate, a short-chained inorganic polyphosphate, was identified in the NaOH and HCl fractions.


Figure 2
View larger version (8K):
[in this window]
[in a new window]

  		Fig. 2. 31P nuclear magnetic resonance spectra of biosolids sequentially extracted with (A) water, (B) NaHCO3, (C) NaOH, and (D) HCl. All spectra were plotted with a line broadening of ~2 Hz. Chemical shift (ppm) of the peaks at ~6.2 ppm was assigned to orthophosphate. In Fig. 1C, peaks at 5.25 and 4.9 ppm were assigned to phosphatidic acid and ß-glycerophosphate, respectively, and –4.43 ppm assigned to pyrophosphate. In Fig. 3D, the peak at –3.74 ppm was assigned to pyrophosphate. Other peaks could not be assigned to any specific P species.

 

Figure 3
View larger version (8K):
[in this window]
[in a new window]

  		Fig. 3. 31P nuclear magnetic resonance spectra of hog manure sequentially extracted with (A) Water, (B) NaHCO3, (C) NaOH, and (D) HCl. All spectra were plotted with a line broadening of ~2 Hz. Peaks at ~6.2 to 6.28 ppm were assigned to orthophosphate P. In Fig. 2C, peaks at 6.12, 5.13, 4.74, and 4.64 ppm in the ratio 1:2:2:1 were assigned to C-2: C-4 + C-6: C-1 + C-3: C-5 positions of phytic acid. Peaks at 5.37 and 4.98 ppm were assigned to phosphatidic acid and ß-glycerophosphate, respectively. Peaks at 4.89, 4.54, and 4.47 ppm were unidentifiable. In (Fig. 2D), peaks at 5.75, 4.86, 4.44, and 4.3 ppm in the ratio of 1:2:2:1 were assigned to phytic acid.

 
Similar to biosolids, the sequential extracts of hog manure was dominated by inorganic orthophosphate (Fig. 3 ), making up 100% of extracted P in the H2O and NaHCO3 fractions and approximately 79% in NaOH fractions and 96% in HCl fractions. In the NaOH fraction, phytic acid was the main form of organic P identified, making up about 11% of the total P. Phosphatidic acid and ß-glycerophosphate were identified in trace amounts (Fig. 3C; Table 3). In the HCl fraction of hog manure, phytic acid was identified, accounting for approximately 4% of the total P.

In DAIRY, inorganic orthophosphate was 98% of the total P in the H2O fractions, whereas phosphatidic acid and phospholipids were identified in trace amounts (Fig. 4A ). In the NaHCO3 extracts, only inorganic orthophosphate was identified (Fig. 4B). In comparison with biosolids and hog manure, dairy cattle manure had a lower inorganic orthophosphate and higher orthophosphate monoesters in the NaOH and HCl fractions. In the NaOH fraction, orthophosphate and phytic acid constituted approximately 64 and 21% of the total P, respectively (Fig. 4C). Similarly, in the HCl fraction, inorganic orthophosphate and phytic acid were 70% and 30% of the total P, respectively. Hydrolytic products of phosphatidyl choline were identified in substantial amounts in the NaOH fraction (7% of total P) but not in the HCl extracts.


Figure 4
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 4. 31P nuclear magnetic resonance spectra of dairy cattle manure sequentially extracted with (A) water, (B) NaHCO3, (C) NaOH, and (D) HCl. All spectra were plotted with a line broadening of ~2 Hz. Peaks at ~6.2 ppm were assigned to orthophosphate, peaks at 5.35 or 5.28 ppm were assigned to phosphatidic acid, and peaks at 4.94 ppm were assigned to ß-glycerophosphate. In Fig. 3A, the small peak at 1.78 ppm was assigned to phospholipids. In Fig. 3C, peaks at 6.0, 5.1, 4.74, and 4.59 ppm assigned to phytic acid. Peaks at 4.87, 4.52, 4.49, and 4.27 ppm were unassignable. The peak at –4.4 ppm was assigned to pyrophosphate. In Fig. 3D, peaks at 5.81, 4.9, 4.51, and 4.39 ppm were assigned to phytic acid.

 
In the beef cattle manure, inorganic orthophosphate was the only species identified in the H2O, NaHCO3, and HCl fractions (Fig. 5 ). However, the NaOH fraction contained inorganic orthophosphate (61%), phytic acid (31%), phosphatidic acid, and ß-glycerophosphate (8%).


Figure 5
View larger version (8K):
[in this window]
[in a new window]

  		Fig. 5. 31P nuclear magnetic resonance spectra of beef cattle manure sequentially extracted with (A) water, (B) NaHCO3, (C) NaOH, and (D) HCl. All spectra were plotted with a line broadening of ~2 Hz. The peaks at ~6.2 ppm were assigned to orthophosphate. In Fig. 4C, peaks at 6.03, 5.31, 4.76, and 4.62 ppm were assigned to phytic acid. Peaks at 5.31 and 5.97 ppm were assigned to phosphatidic acid and ß-glycerophosphate, respectively. The peaks at 4.89 ppm were unassignable.

 
In the poultry litter, inorganic orthophosphate made up 94% the total P in the H2O fractions, 61% in the NaHCO3 fractions, 29% in the NaOH fractions, and 40% in the HCl fraction (Fig. 6 ). The H2O fraction of poultry litter contained phosphatidic acid and other peaks in the monoester regions that have not been previously assigned (Fig. 6A). In the NaHCO3 fraction, however, phytic acid made up about 31% of total P, and phosphatidic acid was also identified (3%). In the NaOH fraction, phytic acid constituted 59% of total P, whereas phosphatidic acid and ß-glycerophosphate made up approximately 8% of the total P. In the HCl fraction, phytic acid was the only organic P present, making up 60% of total P.


Figure 6
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 6. 31P nuclear magnetic resonance spectra of poultry litter (manure + beddings) sequentially extracted with (A) water, (B) NaHCO3, (C) NaOH, and (D) HCl. All spectra were plotted with a line broadening of ~1 Hz. Peaks at ~6.2 ppm were assigned to orthophosphate. In Fig. 5A, the peak at 5.92 and 5.60 ppm were unassigned, and the peak at 5.36 ppm was assigned to phosphatidic acid. In Fig. 5B, peaks at 5.89, 4.98, 4.61, and 4.47 ppm were assigned to phytic acid, the peak at 5.69 ppm was unassignable, and the peak at 5.42 ppm was assigned to phosphatidic acid. In Fig. 5C, the peaks at 6.03, 5.13, 4.77, and 4.62 ppm were assigned to phytic acid. The peaks at 5.23 and 4.97 ppm were assigned to phosphatidic acid and ß-glycerophosphate, respectively. Other peaks were unassigned. In Fig. 5D, the peaks at 5.75, 4.85, 4.45, and 4.33 ppm were assigned to phytic acid.

 
The dominance of orthophosphates in H2O and NaHCO3 fractions of all organic amendments except POULTRY suggest that their P is more bioavailable in the environment compared with that of POULTRY, whose NaHCO3 fraction contained phytic acid. Celi et al. (1999) reported that phytic acid is generally capable of forming stable complexes and insoluble Ca salt precipitates in alkaline soil and is thus protected from microbial degradation due to its higher charge density compared with other orthophosphate monoesters. This higher charge density may lead to the preferential binding of phytic acid relative to other monoesters, making the latter more susceptible to mineralization or hydrolysis (Celi et al., 1999). Orthophosphate diesters and orthophosphate monoesters occurring from enzymatic or alkaline hydrolysis of diesters are weakly sorbed in soil and are therefore mobile (Turner et al., 2005). Furthermore, orthophosphate diesters mineralize more rapidly than monoesters and therefore constitute the labile fractions of P in the manure and manure-amended soil (Condron et al., 1990; Condron et al., 2005). The results from the current study further suggest that inorganic P in a particular sequential fraction may be bioavailable, but the organic P may not be.

The absence of phytic acid in all fractions of BIO compared with other organic amendments as found in this study corroborates the result from another study where the total P of anaerobically digested municipal biosolids was almost entirely orthophosphate (Hinedi et al., 1989). Conversely, the association of cations with organic P extracted by NaOH and HCl fractions in the organic amendments except in BIO and HCl fractions of BEEF suggests a complexation of polyvalent cations with phytic acid. Phosphorus species, including orthophosphate, orthophosphate monoesters, and pyrophosphate, are thought to be bound with polyvalent cations such as Al, Fe, and Mn (Miltner et al., 1998; Celi et al., 1999; Leytem et al., 2002; Toor et al., 2005a). These cations form bridges, thereby enabling organic P to take on condensed formations that may reduce the efficacy of the extractant (Swift, 1996). Therefore, the low concentration of Al and Fe in POULTRY (undetected, except for Al in NaOH extract) may be due to the presence of a condensed form of organic P in this amendment. The higher proportion of the residual P in HOG and DAIRY relative to other manures suggests the presence of some recalcitrant form of P that could not be recovered during digestion and subsequent ICP analysis. Residual P has often been considered to be occluded P or P bound to stable humic substances (Cross and Schlesinger, 1995). Celi and Barberis (2005) also pointed out that Ca phytates are insoluble in alkaline, whereas Al and Fe phytates are insoluble in acid; hence the poor extractability of the latter in HCl.

The identification of phytic acid in the HCl extracts of HOG, DAIRY, and POULTRY confirmed that it is erroneous to assume that this less labile fraction is only composed of inorganic P in the form of CaP. Some studies that used modifications of the Hedley fractionation schemes to characterize manure P have omitted the HCl extraction step (He et al., 2004) or assumed that HCl-P is mainly Ca-P, similar to what is expected in soil (Ajiboye et al., 2004; McDowell and Stewart, 2005b).

The relatively higher proportion of other orthophosphate monoesters and diesters like phospholipids in POULTRY suggests that microbial activity was greater than in other organic amendments because these P species are usually associated with microbes (Turner et al., 2003). The presence of substantial labile monoesters and diesters in POULTRY indicates that organic P in this manure is present in a form that is bioavailable to aquatic organisms. Toor et al. (2005a) reported that diesters are generally less strongly sorbed in the soil, making them bioavailable, although DNA may penetrate the interlayer space of clay minerals under acidic conditions.

As shown in this study, the estimation of organic P as the difference between ICP-P and molybdate-P did not agree with the estimates by 31P NMR. The solution 31P NMR did not detect organic P in the H2O and NaHCO3 fractions of all organic amendments except in POULTRY and H2O fraction of DAIRY. The organic P estimated by colorimetry was negative in the H2O fraction of the BIO, HOG, and HCl fractions of BIO, HOG, and BEEF where an overestimation of inorganic P occurred. In contrast to our results, McDowell and Stewart (2005b) reported a good agreement between organic P estimated by colorimetry and NMR for grazing dairy cattle, sheep, and deer even though phytic acid and other diesters were not well resolved in the NMR spectra and were only estimated by spectral deconvolution. The estimation of organic P in manure as the difference between ICP-P and molybdate reactive P may not hold true for all manures, and such results should be interpreted with caution.

Identification of P Species in Sequential Extraction Residues by P K-edge XANES
The results of the LC fit of P standards to the organic amendments are shown in Fig. 7 . The XANES spectrum of BIO was shown to be dominated by variscite and HAP, amounting to 86% and 14% of the total phosphates, respectively (Fig. 7a). The LC fitting of P 1s XANES of other organic amendments indicated that dicalcium phosphate dihydrate (DCPD) was a dominant species with proportions as high as 50 and 60% of total phosphate in HOG and DAIRY, respectively, and between 18 and 20% in BEEF and POULTRY (Fig. 7b–e).


Figure 7
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 7. Linear combination (LC) fit showing proportion of identified P compounds in organic amendments: (a) biosolids, (b) hog manure, (c) dairy cattle manure, (d) beef cattle manure, and (e) poultry litters. The identified P compounds were VAR, variscite (AlPO4·2H2O); HAP, hydroxyapatite (Ca10(PO4)6(OH)2); PHYTIC, phytic acid; Ca salt (C6H6Ca6O24P6); DCPD, dicalcium phosphate dihydrate (CaHPO4·2H2O); and STRUV, struvite (MgNH4PO4·6H2O).

 
In addition to DCPD, phytic acid, the most ubiquitous organic P in the environment, was identified as a dominant constituent of XANES spectra of the manures, ranging from 20% in BEEF, 32% in HOG, 35% in DAIRY, to as high as 70% in POULTRY (Fig. 7b–e). Toor et al. (2005b), in using LC fitting of XANES spectra of litters of broilers fed with different combinations of "normal" and "high available P" corn, also identified phytic acid with values ranging from 7 to 20% of total P and in litters of turkey fed with normal and P-deficient corn with values ranging from 16 to 32% of total P. An earlier study on quantitative P speciation could not fit phytic acid to spectra of soils with long-term history of manure application due to the lack of any resonance in the near-edge region (Beauchemin et al., 2003). Toor et al. (2005b) also observed a good LC fit with or without phytic acid in a particular treatment with "high available P corn" and considered the LC fit as inconclusive for that sample. Other notable P species identified as dominant in the manures were variscite, which made up 18% of total P species in HOG, and struvite, which constituted 68 and 12% of total phosphates in BEEF and POULTRY, respectively.

To determine P species removed by each extractant, the spectra difference method involving the point-by-point difference between the spectra of a particular extractant and that of the next in the extraction sequence was used. However, the noisiness of the resulting spectra precluded reliable identification of features of interest for each sequential fraction. Consequently, only the spectra of the labile P fractions, calculated as difference between spectra of unextracted organic amendments (intact) and spectra of residues after NaHCO3 extraction, were used with some confidence. The resulting XANES spectra were not subjected to LC fitting because of the variation in the overall slope beyond the edge, which made the normalized spectra to form a sort of "fan" in the post-edge region. Hence, the XANES spectra of labile P fractions of the organic amendments are described qualitatively as shown in Fig. 8 . The labile P fractions in BIO and HOG seemed to be dominated by variscite as given by the absence of a shoulder and the presence of a sharper post-edge feature at 2162 eV than that of struvite (Fig. 8). Similarly, the spectrum of labile P in HOG lacks a shoulder but has a subtle feature at 2162 eV that resembles that of struvite (Fig. 8). The spectra of labile P in BEEF and DAIRY indicated that DCPD predominated, similar to the intact (unamended) sample. The labile P fraction in POULTRY showed a more distinct shoulder and an intense peak at approximately 2163 eV, suggesting the dominance of DCPD (Fig. 8).


Figure 8
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 8. P 1s XANES spectra of labile P (point-by-point difference between spectra of intact amendments and residues after NaHCO3 extraction) and those of matching reference compounds. The spectra were plotted with slight vertical displacement to aid comparison of the features. The vertical line represents (a) shoulder feature of dicalcium phosphate dihydrate (DCPD), (b) feature at 2162 eV peculiar to STRUVITE and VARISCITE, and (c) 2163 eV peak of DCPD.

 
Linking P Species in Sequential Extracts by NMR with P Species in Residues by XAS
The presence of HAP in BIO may be responsible for the least amount of labile P in BIO relative to other organic amendments. Ajiboye et al. (2004) reported that BIO contained a proportionately smaller amount of P in the water extract than in HCl extract, suggesting that biosolids P is more in the non-labile fraction and less in the labile fraction compared with manures. Although HAP and variscite were the dominant P species in BIO, variscite is readily soluble in the neutral to alkaline solutions that are used to extract the labile P pool, and its presence in the XANES spectra for labile P suggests that the non-labile P in BIO was in the form of HAP. Because HAP is the most thermodynamically stable form of CaP, its presence infers less solubility and lower risk of BIO P loss to the environment compared with other amendments. The absence of HAP in the HOG, BEEF, and POULTRY may be due to the inhibition of precipitation of P by organic acids from these amendments as indicated by their phytic acid content. Inskeep and Silvertooth (1988) reported that the presence of organic acid decreases precipitation of P due to sorption of organic anion onto the minerals and coating of the crystal growth sites. The absence of pre-edge in the spectra of organic amendments indicated that these amendments contain no FeP within the detection limit of the XAS technique.

The high amount of labile P in other organic amendments, especially in HOG and DAIRY, may be attributed to the predominance of DCPD in these samples because it is more soluble than the thermodynamically favored HAP. The near-edge features of the difference spectra, which indicated that DCPD constituted the labile P pool in all the manures expect POULTRY, supported this result. The DCPD identified in the labile P XANES spectra could also be interpreted to be P adsorbed onto CaCO3. According to Peak et al. (2002), the spectra of adsorbed P on CaCO3 showed resonance features similar to, but not as sharp as, DCPD and HAP.

The predominance of struvite in BEEF and in substantial amounts in POULTRY is of particular agro-environmental importance. Struvite is important in specialty agriculture as a source of slow-release P fertilizer, and the recovery of struvite in livestock manure is one of the technological options being pursed to control P release from livestock into the environment (Zeng and Li, 2006; Huang et al., 2006). Neither the sequential chemical extraction nor NMR spectroscopy is capable of identifying struvite in these samples. Only the Mg/P ratio could have been used indirectly to infer the presence of struvite had XAS not been used, indicating the importance of combining chemical extraction with solid phase speciation.

The detection of DCPD and phytic acid-Ca salt (PHYTIC) in the XANES spectra of other manures further corroborates the results from NMR analyses, where POULTRY had the highest amount of PHYTIC. This result is similar to that of Peak et al. (2002), who observed that poultry manure unamended with alum (Al2(SO4)3) contained DCP and some weakly bound organic P, but no HAP. The spectra of various reference compounds, such as aqueous phosphate (Sato et al., 2005), adsorbed P on amorphous Al(OH)3 and gibbsite (Peak et al., 2002), and organic orthophosphate monoesters like phytic acid in the current study, were similar, lacking any resonance features. Although Sato et al. (2005) did not detect phytic acid in their poultry manure, the combined analyses of NMR and LC of P XANES of organic amendments in our study clearly showed that PHYTIC was present in considerable amounts in all manures and more so in POULTRY than in HOG, DAIRY, and BEEF.

The intense features of DCPD in the labile P spectrum of POULTRY, which was absent in that of the intact sample may be due to higher amount of organic P (as measured by NMR of sequential extracts) in the latter. This suggests that high amount of organic P in the intact POULTRY masked the features of DCPD in the XANES spectrum, which were noticeable only in the spectrum of labile P that contained a lesser amount of organic P. Organic constituents generally inhibit crystallization of P minerals, and the spectra of amorphous CaP have been reported to lack resonances (Peak et al., 2002). The type of P species present in the labile fraction of the organic amendments is of particular environmental relevance. For example, the dominance of AlP (VAR) in the labile P fraction of BIO suggests that this fraction is readily bioavailable under neutral to alkaline soil pH but not under acidic soil condition. However, studies have shown that where these biosolids was added to high-pH soil, labile inorganic P increased at the expense of organic P with time, indicating mineralization and the presence of organic P (Kashem et al., 2004a,b), which was not detected in the XANES spectra of our study. The presence of DCPD in the labile P fraction of DAIRY and BEEF also has an environmental significance. These species are less soluble in alkaline pH; therefore, their availability in soil environment depends largely on the change in soil solution pH.

Conclusions

This study provides a detailed molecular characterization of P in different organic amendments using a combination of sequential chemical extraction, solution 31P NMR, and XANES spectroscopies. Overall, the P speciation results from these three techniques were complementary to one another; they provided molecular characterization of P in organic amendments that would not have been possible with any of the individual or combination of any two of these techniques. Although solution 31P NMR provided a detailed characterization of organic P in the NaOH and HCl fractions of organic amendments P, it was limited in characterizing the H2O and NaHCO3 fractions of most organic amendments probably due the proneness of these labile fractions to hydrolysis. However, ICP and molybdate-blue colorimetry indicated the presence of organic P in these sequential extracts. Furthermore, XANES analysis of the sequential extraction residues identified the actual chemical species of labile P as readily soluble calcium and some aluminum phosphates, which was only characterized as inorganic P by the molybdate-blue colorimetry of sequential extracts and as orthophosphates by solution 31P NMR. Although an attempt was made to identify P species in each step of the sequential extraction procedure using "difference spectra," the noisiness of the spectra of the residues in the later stage of the extraction sequence due to low concentration of P precluded this identification. Although a direct comparison could not be made between the operationally defined P forms in all stages of sequential extraction and P species identified by XANES due to the concentration issue, further advances in improving the detection limit of P in environmental samples may make this possible in the near future.

ACKNOWLEDGMENTS

The authors acknowledge the contributions of Dr. Scott Kroeker (Chemistry, University of Manitoba) during the internal revision of this manuscript, Dr. Ben Turner (Smithsonian Tropical Research Institute, Panama) for comments on NMR analyses, and Dr. Kirk Marat (Chemistry, University of Manitoba) for operating the NMR spectrometer. The authors also thank Drs. Astrid Jürgensen and Franziskus Heigl for assistance with operating the double crystal monochromator beamline of the Canadian Synchrotron Radiation Facility (CSRF) at Synchrotron Radiation Center (SRC), University of Wisconsin-Madison. This work was supported in part by the National Science Foundation under award no. DMR-0084402 and DMR-0537588 to the SRC, where part of this research was conducted. CSRF is funded by the Open Access Grant of Natural Science and Engineering Research Council (NSERC) of Canada and by the National Research Council of Canada. The financial support to Dr. O. Akinremi by NSERC is duly acknowledged. B. Ajiboye gratefully acknowledges the graduate fellowship (UMGF) provided by the University of Manitoba.

NOTES

All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.

REFERENCES




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
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]

Home page
 Soil Sci.Home page
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]

Home page
 Soil Sci.Home page
Z. He, C. W. Honeycutt, B. J. Cade-Menun, Z. N. Senwo, and I. A. Tazisong
Phosphorus in Poultry Litter and Soil: Enzymatic and Nuclear Magnetic Resonance Characterization
Soil Sci. Soc. Am. J., August 20, 2008; 72(5): 1425 - 1433.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Supplements
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ajiboye, B.
Right arrow Articles by Flaten, D. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ajiboye, B.
Right arrow Articles by Flaten, D. N.
Agricola
Right arrow Articles by Ajiboye, B.
Right arrow Articles by Flaten, D. N.
Related Collections
Right arrow Phosphorus
Right arrow Nutrient Management
Right arrow Other Waste Management

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