Environ. Sci. Technol. 2005, 39, 9205-9210
Phosphorus Composition of Sheep Feces and Changes in the Field Determined by 31P NMR Spectroscopy and XRPD C H A R L E S A . S H A N D , * ,† G R A C E C O U T T S , † STEPHEN HILLIER,† DAVID G. LUMSDON,† ALEXANDER CHUDEK,‡ AND JAN EUBELER§ Macaulay Institute, Aberdeen, AB15 8QH, U.K., and University of Dundee, Dundee, DD1 4HN, U.K.
(8) to provide detailed structural information about P compounds in feces and requires minimal perturbing pretreatment. MAS 31P NMR spectroscopy has been used (9) for the speciation of P in poultry litter. Solution-phase 31P NMR spectroscopy has been used to study organic P in extracts of manures and feces (3, 4, 10-12). X-ray powder diffraction (XRPD) also offers a direct means to identify P-containing minerals in feces and has been used to examine P minerals in sewage sludge (13). Our objective was to determine the P composition of sheep feces using 31P NMR spectroscopy and XRPD and to follow changes that occur in feces in the field. We synthetically applied sheep feces in patches to the herbage surface of a pasture soil and followed temporal changes over 12 weeks until the feces became indistinguishable from the soil/ herbage interface.
Materials and Methods Information on the P species in sheep feces is lacking. Such information is required to understand P-cycling in grazed ecosystems. The P composition of feces from sheep grazing grass in Scotland was assessed on freeze-dried samples by 31P MAS (magic angle spinning) NMR (nuclear magnetic resonance) spectroscopy and XRPD (X-ray powder diffraction). The 31P MAS NMR spectrum showed resonances and sidebands consistent with dicalcium phosphate dihydrate (brushite) and ammonium magnesium phosphate hexahydrate (struvite). XRPD confirmed the result and allowed quantification of these minerals, which accounted for 63% of the P. To determine transformations in the field, sheep feces were collected and reapplied to sheepfree pasture in synthetic patches during late summer. The dry weight decreased with time and the feces disappeared between 84 and 112 days following heavy rainfall. The concentration of P in the feces recovered at intervals up to 84 days changed little with time but the contribution from brushite and struvite decreased and within 1 week 95%) in their feces with little in urine (1). Fecal patches represent areas where P is recycled (2). Little information exists about the P compounds that occur in animal feces (3, 4) and how they interact with soil. Biogenic phosphate minerals in manures have been characterized by scanning electron microscopy and energy-dispersive X-ray spectroscopy (5, 6) and the P composition of sheep feces has been assessed by modeling (7). Solid-state magic angle spinning (MAS) 31P nuclear magnetic resonance (NMR) spectroscopy has the potential * Corresponding author phone: +44(0)1224498200; fax +44(0)1224311556; e-mail:
[email protected]. † Macaulay Institute. ‡ University of Dundee. § Europa Fachhochschule Fresenius, Limburgerstrasse 2, D-65510 Idstein, Germany. 10.1021/es0510820 CCC: $30.25 Published on Web 10/26/2005
2005 American Chemical Society
The site (56° 54′ N; 2° 33′ W; 220 m above sea level) is in northeast Scotland at Glensaugh, where hill grazing is a major land use. The treatment with sheep feces (described later) was on an established, improved grass clover sward. A fence erected in spring excluded sheep and rabbits. The sward was maintained for 4 months by cutting to 25-mm above the soil and removing the harvested material. The soil was a Typic Fragiorthod and air-dried samples from the 0-50 mm deep horizon had a pH of 4.64 and a loss on ignition of 13.5%. Weather data were from a station 1 km southeast at a similar altitude. Sheep feces were from Scottish Blackface/Blue Leicester crosses grazing an adjacent field which had been fertilized in May, June, and July (total 164 N, 49 P, and 59 K kg ha-1 y-1). Following clearance of residual feces, fresh feces in the form of pads were collected over 3 consecutive days in late summer (August) and stored at 4 °C. On the day after the last collection the feces were mixed avoiding disruption of the aggregates and immediately used in the plot experiment. The plot had a randomized block design. There were 12 blocks of soil each with an area measuring 1.175 × 1.175 m. Each block was divided into a 3 × 3 grid and a plastic ring (152-mm diameter) was placed in the center of each grid square. Aliquots of fresh feces 150 ( 1 g (29.1 g dry matter) were applied to six blocks in the areas inside the rings. The patch area (0.018 m2) was typical (14) of natural patches observed during feces collection. There is variation in the weight of a single defecation and values between 8 and 57 g dry weight have been reported (15). To prevent wind-throw, each patch was covered with a lightweight nylon net. At the time of feces application to the soil, replicate samples of the feces (150 g each) were taken for analysis. At intervals (day 7, 14, 28, 56, 84) following the application of the feces the residual feces were recovered for analysis. Feces were freeze-dried and ground. Subsamples (250 mg) were muffled at 550 °C for 16 h and the ash was digested with hot re-distilled 6 M HCl. The extract was evaporated to dryness and redissolved in 0.34 M HCl for P and other element analysis by inductively coupled plasma-optical emission spectroscopy (ICP-OES). The C and N contents of the feces were determined by combustion using a Thermo Finnigan Flash EA analyzer. The water-extractable P content of the feces was determined by shaking 150 mg of the freeze-dried sample with 30 mL of deionized water for 16 h. The extract was centrifuged (10 000g), filtered (0.45 µm), and analyzed for total P (and other elements) by ICP-OES. The concentration of molybdate reactive P was determined using an air-segmented flow analyzer (16). VOL. 39, NO. 23, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
9205
For 31P MAS NMR spectroscopy we used a Varian/ Chemagnetics CMX300 LITE instrument (Chemagnetics Inc., Fort Collins, CO) operating at 121 MHz for 31P nuclei. The powdered feces sample (300 mg) was packed into a zirconia pencil rotor and spun at the magic angle (typical frequency 6 kHz). Because it was impractical to analyze samples from individual fecal patches the samples for 31P MAS NMR spectroscopy were prepared by mixing equal amounts from the replicates. For cross-polarization (CP) the excitation pulse duration was 3.5 µs (90°) and the recycle and contact times were 4 s and 1 ms, respectively. For high power decoupling (HPD) the pulse delay was 2 s (90°). The number of acquisitions was generally g2500. A sample of the sheep feces was used to determine the pulse sequence/delay to give maximum peak height. A pulse delay g20 s was required with HPD to give maximum peak height, whereas with CP a pulse delay of 2 s was adequate. Dicalcium phosphate dihydrate (DCPD), (CaHPO4‚2H2O) with a stated purity of >98% was from Fluka/Sigma Aldrich, Dorset, UK. Ammonium magnesium phosphate hexahydrate (AMPH), (NH4MgPO4‚6H2O) manufactured by Alfa Aesar with a stated purity of >99% was from Avocado Research, Lancashire, U.K. In mineralogical studies these P compounds are known as brushite and struvite, respectively. The 31P MAS NMR spectra of the solids were recorded under the same conditions used with feces but with fewer scans. For solution-phase 31P NMR spectroscopy we used a Bruker DPX300 spectrometer (Bruker U.K., Coventry) operating at 121 MHz for 31P nuclei. The freeze-dried feces (1 g) were extracted with 0.25 M NaOH/0.05 M EDTA (20 mL) for 16 h at 20 °C (8). The suspension was centrifuged (10 000g) and the supernatant was freeze-dried. A sample of the solid (200-300 mg) was dissolved in 1 mL of 1.0 M NaOH, and 0.1 mL of D2O was added to provide a frequency lock. Optimized run conditions were as follows: acquisition time, 1.79 s (8 k memory); pulse width, 8 µs (90°); relaxation delay, 2 s; sweep width, 238 MHz; temperature, 297 K. Broad band proton decoupling was used. Using a test extract, changing the relaxation delay from 2 to 0.2 or 4 s gave 90 days to acquire one diffraction pattern. Crystalline phases were identified by comparison with patterns in the Power Diffraction File (PDF), release 2001. Quantitative analysis was performed by a reference intensity ratio (RIR) method (17) using added corundum (0.3 µm) as the internal standard. Standards for brushite and struvite were from Fluka and Alfa Aesar as described before. The struvite contained ammonium magnesium phosphate monohydrate (dittmarite) as a contaminant. Quantification by Rietveld analysis using the program Topas (Bruker AXS, Karlsruhe), and the known structures of dittmarite and struvite, indicated 5% dittmarite. Experimental RIRs of 1.43 for the brushite peak at 7.5 Å and 0.34 for the struvite peak at 5.6 Å were determined, with allowance for 5% dittmarite in the latter. These peaks along with selected internal standard peaks were used for quantification of these 9206
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 23, 2005
phases in the sheep feces. For quantitative analysis, subsamples were spiked with 5 wt % of corundum and scanned as described before but with a reduced count time of 600 s per step. Fresh feces (day 0) stored at -20 °C were thawed-out over 16 h and analyzed by XRD. A sample was pressed into the XRD sample holder and analyzed in a wet condition. Another sample of the feces (60 g) was centrifuged at 49 000g and the solution expressed on the surface was analyzed by ICPOES. Infrared spectra (KBr disks) of the freeze-dried feces were recorded using a Nicolet-IR 550 spectrometer.
Results Initial Composition of Sheep Feces. The variation in the P content between the 150-g sub samples of fresh feces (initial moisture content 81%) was small. The composition, expressed as g kg-1 ( the percentage relative standard deviation (n ) 5) was C 478 ( 0.62, N 34.0 ( 1.14, P 13.9 ( 1.44, K 27.0 ( 1.26, Ca 12.3 ( 0.44, Mg 6.05 ( 2.80, Na 3.44 ( 0.58, Fe 0.49 ( 2.2, and Al 0.27 ( 3.2. The herbage consumed by the sheep prior to feces collection contained 0.31% P on a freeze-dried basis. The MAS NMR spectra of the feces at day 0 showed two main resonances with shifts near 6 and 1 ppm (Figure 1); only the resonance at 1 ppm had associated sidebands. The band at 1 ppm was broader than that at 6 ppm and had several inflections. By deconvolution (Figure 1), five resonances were fitted (Table 1). As these data were acquired using CP they cannot be regarded as fully quantitative (18) but they do show relative changes in the ratios of the different moieties. The NMR spectra of the NaOH/EDTA extract, discussed later, showed the feces contained most of its P in inorganic form and indicated that the main signals were from inorganic species. The 31P MAS NMR spectra of the purchased DCPD had a single resonance with an isotropic chemical shift of 1.2 ppm with prominent sidebands. A similar pattern of sidebands has been reported (19, 20) for DCPD and they are attributable to the large dipolar interactions between the protons of structural water. We tentatively assigned the peak at 1 ppm to DCPD. The spectrum of the purchased AMPH was different from that of DCPD and consisted of one main isotropic signal centered at 6.0 ppm and sidebands were absent. The resonance at 6.12 ppm was tentatively assigned to AMPH. AMPH has not been positively identified in animal feces but has been predicted by modeling (5) and identified by XRD in liquid cattle manure (21). The X-ray diffractogram of the feces at day 0 is shown in Figure 2 along with reference patterns of brushite (DCPD) and struvite (AMPH). In addition to peaks assigned to brushite and struvite, peaks indicating the presence of small amounts of quartz, feldspars, and clay minerals were present. The broad hump-like features of the pattern are due to diffraction from X-ray amorphous materials. We attributed the intense, sharp peak at 21.5° (2θ) to wax (compared with PDF 401995), the peak being lost on ashing the feces. Quantitative XRPD analysis showed that the feces at day 0 contained 26 g kg-1 of brushite (DCPD) and 32 g kg-1 of struvite (AMPH), equivalent to 4.68 and 4.04 g kg-1 of P, present in these two inorganic P forms, respectively, accounting for 63% of the P in the feces (Figure 3). XRD analysis of a sample of the sheep feces from day 0 that had been stored at -20 °C, thawed, and analyzed in a wet condition showed the presence of brushite and struvite in near equal amounts. The solution phase 31P NMR spectra of the EDTA/NaOH extracts of the freeze-dried feces (Figure 4) was dominated by a sharp resonance at 6.4 ppm assigned to inorganic orthophosphate and confirmed by spiking the extract with KH2PO4. The additional minor signals were identified on the basis of chemical shifts by comparison with values for model compounds (22, 23). The assignments were as follows:
FIGURE 1. Solid-phase 31P MAS NMR spectra of freeze-dried sheep feces showing temporal changes resulting from weathering in the field. The lower components show deconvolution of the cross-polarized spectra for day 0 and day 84; the traces show, from top to bottom, the: (a) original spectrum, (b) sum of the components, (c) individual components, and (d) difference spectrum.
TABLE 1. Relative Peak Areas Obtained by Deconvolution of the Cross-Polarized 31P MAS NMR Spectra of Sheep Feces relative peak area chemical shift (ppm)
day 0
day 84
assignment
6 2 1 0 -2
18.4 2.2 30.2 3.7 2.7
4.7 3.4 1.9 8.3 10.9
AMPH (struvite) DCPD (brushite)
phosphate monoesters (5.6 and 5.2 ppm); aromatic phosphates, orthophosphate diesters including nucleic acids and phospholipids (0.2 ppm); and polyphosphate end groups (-4.0 ppm). However, alkaline hydrolysis of some phosphate diesters/phospholipids likely occurred and contributed material to the monoester pool (8, 23). The efficiency of extraction of total P from the freeze-dried sample at day 0 with 0.25 M NaOH/0.05 M EDTA solution was 86%. Waterextractable P was also considerable (71%) and most was molybdate reactive (Figure 3). The IR spectrum of the sheep faces at day 0 showed two weak absorption bands centered at 530 and 580 cm-1 in the region between 500 and 600 cm-1. In this region we found
brushite has bands at 527 and 577 cm-1 and struvite has a band at 571 cm-1 only. The IR data are consistent with the presence of brushite and struvite but we were unable to determine the existence of other P compounds. Temporal Changes in the Composition of Feces. Feces survived for at least 84 days in the field and at day 84, 54 wt % of the initial dry matter remained (Figure 5). There was substantial rainfall during the week prior to sampling at day 84 (week 12) and in the following 3 weeks, with a maximum weekly rainfall input of 62.3 mm during week 13 (see Supporting Information). In the 0-50-mm soil horizon the gravimetric moisture increased from 18 to 28% and temperature decreased from 14 to 8 °C during the 84 days of the experiment and there were occasional ground frosts during the later periods (see Supporting Information). The 31P MAS NMR spectra of sheep feces recovered from the soil surface shows broadening with time and an increase in the relative area of the signal near 2 ppm. Deconvolution of the spectrum of the sample at day 84 showed a change in the distribution of species with large decreases in the signals attributable to DCPD (1 ppm) and a smaller but substantial decrease in AMPH (6 ppm). Signals at 2, 1, and -2 ppm increased, especially the latter. Similarly, the XRPD patterns of the feces changed with time and showed a trend of a decrease in the intensity of VOL. 39, NO. 23, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
9207
FIGURE 2. XRPD pattern for freeze-dried sheep feces (day 0) and reference “stick” patterns for brushite and struvite. Most of the sharp peaks are attributable to brushite and struvite but there are also some peaks that may be assigned to minor amounts of quartz, feldspars, clay minerals, and wax. The expanded inset (10-17°, 2θ) shows the XRPD patterns from sheep feces recovered from the field at different periods after application; the positions of the peaks from brushite and struvite used for quantification are indicated.
FIGURE 3. Lines show the contribution of brushite and struvite to the total P in sheep feces determined by quantitative XRPD. Struvite was not detected at day 56 and its contribution is shown as zero. The bars show the total amounts of P extractable by EDTA/NaOH and water and the amount of molybdate-reactive P (MRP) extracted by water. The error bars represent the standard deviation (n ) 5 at day 0 and n ) 4 at days 7, 14, 28, 56, and 84) and are not available for the EDTA/NaOH extracts at days 56 and 84: the data shown for these is a single value from the analysis of composite samples. peaks due to brushite and struvite (Figure 2). Quantitative determination of the amount of brushite and struvite in the samples over time is given in Figure 3. The amounts of brushite and struvite decrease at similar rates and within 1 week are reduced to concentrations less than half of their initial values. Brushite persisted for the duration of the experiment, although from day 28 onward the concentration was approaching the limits of detection. Struvite, which is more difficult to detect than brushite, was not consistently detected after day 56. In contrast with the IR spectrum of sheep feces at day 0, for later samples we did not observe adsorption bands above background in the region between 500 and 600 cm-1 that could be attributed to brushite or struvite confirming the XRPD results, which showed that a large percentage of brushite and struvite was lost in the first 7 days. 9208
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 23, 2005
Chemical Modeling of Phosphate Minerals. Chemical modeling was used to establish whether brushite and struvite could, in theory, be present in the wet feces. To calculate the amounts of precipitated minerals, the total element concentrations were converted to concentration units (moles L-1) based on a moisture content of 81% for the fresh feces. We assumed that 90% of the P in the feces was orthophosphate (based on NMR) and that NH4+ was 5% of the total N (1). Chemical speciation calculations were performed within the ORCHESTRA modeling framework (24). Thermodynamic constants for soluble species were from the MinteqA2 database (25) and solubility products for brushite and struvite were from Lindsay (26). Activity coefficients were calculated with the Davies equation, which is valid for ionic strengths up to 0.5 mol L-1. The calculations reveal at the measured pH of the feces (pH 7.2) that both brushite and struvite would be stable phosphate-containing solid phases (Figure 6). The models’ calculated percentage weights of mineral present were 2.24% for struvite and 5.05% for brushite, close to those measured by XRPD. Centrifuging feces taken at day 0 that had been stored at -20 °C yielded 58% of the water it contained. Assuming that the liquid which remained in the feces had the same composition as that isolated, we calculated that >92% of the P, Mg, or Ca was in the solid phase.
Discussion The concentration of P in the grass (0.31% P) consumed by the sheep in our experiment is near the middle of the concentration range (0.25-0.49% P) for herbage from selected UK studies (1). Although we did not completely homogenize the sheep feces because we wished to retain its structure, analysis of the replicate samples of feces taken at day 0 showed variation in the elemental composition was small and that the P concentration was at the upper end of the range (0.5-1.5% P) for ruminant feces (1). The 31P MAS NMR spectra of the sheep feces showed two well-defined resonances. Our spectra were interpreted by consideration of the chemical shifts, sideband patterns, and comparison with model compounds. Similar approaches have been used to identify P species in urinary stones (27) and casein micelles (28). Application of HPD alone produced some change in
FIGURE 4. Solution-phase 31P NMR spectra of NaOH/EDTA extracts of freeze-dried sheep feces recovered from the field at different periods after application.
FIGURE 5. Temporal changes in the composition and weight of sheep feces in the field. Error bars represent the standard deviation (n ) 4).
FIGURE 6. Modeled concentrations of brushite, struvite, and dissolved Ca, Mg, and P in sheep feces as a function of pH. the shape of the signals but gave poorer resolution of the main resonances and no unique structural information was derived. The P species responsible for the observed resonances were attributed largely to inorganic phosphate because the spectrum of the EDTA/NaOH extracts of the feces was dominated by inorganic orthophosphate. Barrow
(7) studied the three-phase solubility diagram for calcium, phosphate, and hydrogen ion and concluded that the phosphate in sheep feces was DCPD, which we found resonates at 1.2 ppm. Other P minerals such as hydroxyapatite, octocalcium phosphate, and amorphous calcium phosphate resonate near 3 ppm (29) and the presence of such minerals cannot be excluded. The XRPD pattern of the freeze-dried feces showed it to contain brushite and struvite with small amounts of quartz, feldspars, and clays. The source of the quartz, feldspars, and clays is most likely ingestion of soil particles by the animals. The X-ray diffractometer we used had a detector able to simultaneously collect diffracted rays over a wide angular range. Brushite and struvite were readily detectable in the samples using this experimental setup and the greater sensitivity of this system contributed to their successful detection and quantification. Measurement of full width at half-maximum (fwhm) for the peaks used for quantification of brushite and struvite indicated minor differences in “crystallinity” between the minerals in the purchased standards and those in the feces. The fwhm values were as follows: brushite standard 0.133°, brushite in feces 0.176°; struvite standard 0.169°, struvite in feces 0.157°. Differences in “crystallinity” do not in any case affect quantification since the RIR method is based on measuring peak areas. The apparent decrease in the amounts of structured P minerals in feces with time, shown by XRPD, indicates transformation of P into other less crystalline forms. Undoubtedly, the P composition of feces changes with time and there may be formation of recalcitrant minerals akin to apatite (30). We were unable to confirm this and could not identify P minerals other than brushite or struvite. Barriers to crystallization of new mineral phases may include microbial action or the presence of organic acids (31, 32). Unlike XRPD, IR study does not require the presence of crystalline phases for information but our IR spectroscopy study was unable to add new data about the existence of other P compounds that formed. Chemical modeling predicted the coexistence of brushite and struvite in solid form in the fresh sheep feces. However, the exact proportions of brushite and struvite are sensitive to the selected concentration of NH4+ used in the model. VOL. 39, NO. 23, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
9209
When calculating the mineral ion activity products in the feces we assumed no retention of cations by organic matter because the total cation content of the sheep feces was expected to outweigh the cation exchange capacity. The cation exchange capacities of composted manures are of the order of 0.18-0.35 mol kg-1 (33), while the sum of cationic charge in our sheep feces was in excess of 2.0 mol kg-1. In support, XRPD analysis of feces from day 0 that had been stored at -20 °C and analyzed wet showed the existence of brushite and struvite, and the liquid phase from the feces, isolated by centrifugation, showed that most of the P in the feces was associated with the solids. A large percentage of the total P in the sheep feces throughout the 84 days in the field was water soluble (53 to 71%) but we did use a large 200:1 solution-to-sample ratio. Haynes and Williams (34) found values of water soluble P of between 47 and 65% for sheep feces from animals on pasture fertilized with superphosphate. The retention of P in the fecal residues on the soil surface which is subject to rainfall is contradictory but our observations are similar to those of Rowarth et al. (35). A possible explanation for ineffective leaching by rainfall in the field is that feces dry out and this may result in increased exposure of hydrophobic surfaces (3) impeding the ingress of water.
Acknowledgments The Scottish Executive, Environment and Rural Affairs Department funded the work. Jean Robertson of the Macaulay Institute provided the infrared data.
Supporting Information Available Rainfall and temperature data for the field site. This material is available free of charge via the Internet at http:// pubs.acs.org.
Literature Cited (1) Whitehead, D. C. Nutrient Elements in Grassland. Soil-PlantAnimal Relationships; CABI: Wallingford, U.K., 2000. (2) Haynes, R. J.; Williams, P. H. Nutrient cycling and soil fertility in the grazed pasture ecosystem. Adv. Agron. 1993, 49, 119199. (3) McDowell, R. W.; Stewart, I. Phosphorus in fresh and dry dung of grazing dairy cattle, deer, and sheep: sequential fraction and phosphorus-31 nuclear magnetic resonance analyses. J. Environ. Qual. 2005, 34, 598-607. (4) Toor, G. S.; Cade-Menun, B. J.; Sims, J. T. Establishing a linkage between phosphorus forms in dairy diets, feces, and manures. J. Environ. Qual. 2005, 34, 1380-1391. (5) Bril, J.; Salomons, W. Chemical composition of animal manure: a modelling approach. Neth. J. Agric. Sci. 1990, 38, 333-351. (6) Cooperband, L. R.; Good, L. W. Biogenic phosphate minerals in manure: implications for phosphorus loss to surface waters. Environ. Sci. Technol. 2002, 36, 5075-5082. (7) Barrow, N. J. Chemical form of inorganic phosphate in sheep faeces. Aust. J. Soil Res. 1975, 13, 63-67. (8) Cade-Menun, B. J. Using phosphorus-31 nuclear magnetic resonance spectroscopy to characterize organic phosphorus in environmental samples. In Organic Phosphorus in the Environment; Turner, B. L., Frossard, E., Baldwin, D. S., Eds.; CABI: Wallingford, U.K., 2005; pp 21-44. (9) Hunger, S.; Cho, H.; Sims, J. T.; Sparks, D. L. Direct speciation of phosphorus in alum-amended poultry litter: solid-state 31P NMR investigation. Environ. Sci. Technol. 2004, 38, 674-681. (10) Hansen, J. C.; Cade-Menun, B. J.; Strawn D. G. Phosphorus speciation in manure-amended alkaline soils. J. Environ. Qual. 2004, 33, 1521-1527. (11) Turner, B. L. Optimizing phosphorus characterization in animal manures by solution phosphorus-31 nuclear magnetic resonance spectroscopy. J. Environ. Qual. 2004, 33, 757-766. (12) Turner, B. L.; Leytem, A. B. Phosphorus compounds in sequential extracts of animal manures: chemical speciation and a novel fractionation procedure. Environ. Sci. Technol. 2004, 38, 61016108. 9210
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 23, 2005
(13) Huang, X.-L.; Shenker, M. Water-soluble and solid-state speciation of phosphorus in stabilized sewage sludge. J. Environ. Qual. 2004, 33, 1895-1903. (14) Williams, P. H.; Haynes, R. J. Effect of sheep, deer and cattle dung on herbage production and soil nutrient content. Grass Forage Sci. 1995, 50, 263-271. (15) Nguyen, M. L.; Goh, K. M. Sulphur cycling and its implications on sulphur fertilizer requirements of grazed grassland ecosystems. Agric., Ecosyst. Environ. 1994, 49, 173-206. (16) Skalar. San Plus Analyzer, Publication No.: 0101022A US; Skalar Analytical B. V.: The Netherlands, 1996; pp 3-96. (17) Hillier, S. Accurate quantitative analysis of clay and other minerals in sandstones by XRD: comparison of a Rietveld and a reference intensity ratio (RIR) method and the importance of sample preparation. Clay Miner. 2000, 35, 291-302. (18) Kinchesh, P.; Powlson, D. S.; Randall, E. W. 13C NMR studies of organic matter in whole soils: I. Quantitation possibilities. Eur. J. Soil Sci. 1995, 46, 125-138. (19) Rothwell, W. P.; Waugh, J. S.; Yesinowski, J. P. High-resolution variable-temperature 31P NMR of solid calcium phosphates. J. Am. Chem. Soc. 1980, 102, 2637-2643. (20) Miquel, J. L.; Facchini, L.; Legrand, A. P.; Marchandise, X.; Lecouffe, P.; Chanavaz, M.; Donazzan, M.; Rey, C.; Lemaitre, J. Characterisation and conversion study into natural living bone of calcium phosphate bioceramics by solid state NMR spectroscopy. Clin. Mater. 1990, 5, 115-125. (21) Fordham, A. W.; Schwertmann, U. Composition and reactions of liquid manure (gu ¨ lle), with particular reference to phosphate: II. Solid phase components. J. Environ. Qual. 1977, 6, 136-140. (22) Bedrock, C. N.; Cheshire, M. V.; Chudek, J. A.; Goodman, B. A.; Shand, C. A. Use of 31P-NMR to study the forms of phosphorus in peat soils. Sci. Total Environ. 1994, 152, 1-8. (23) Turner, B. L.; Mahieu, N.; Condron, L. M. Phosphorus-31 nuclear magnetic resonance spectral assignments of phosphorus compounds in soil NaOH-EDTA extracts. Soil Sci. Soc. Am. J. 2003, 67, 497-510. (24) Meeussen, J. C. L. ORCHESTRA: an object-oriented framework for implementing chemical equilibrium models. Environ. Sci. Technol. 2003, 37, 1175-1182. (25) Allison, J. D.; Brown D. S.; Novo-gradac, K. J. MINTEQA2, geochemical assessment model for environmental systems, Version 3; U.S. Environmental Protection Agency: Washington, DC, 1992. (26) Lindsay, W. L. Chemical Equilibria in Soils; John Wiley and Sons: New York, 1979. (27) Bak, M.; Thomsen, J. K.; Jakobsen, H. J.; Petersen, S. E.; Petersen, T. E.; Nielsen, N. C. Solid-state 13C and 31P NMR analysis of urinary stones. J. Urology 2000, 164, 856-863. (28) Bak, M.; Rasmussen, L. K.; Petersen, T. E.; Nielsen, N. C. Colloidal calcium phosphates in casein micelles studied by slow-speed spinning 31P magic angle spinning solid-state nuclear magnetic resonance. J. Dairy Sci. 2001, 84, 1310-1319. (29) Frossard, E.; Tekely, P.; Grimal, J. Y. Characterization of phosphate species in urban sewage sludges by high-resolution solid-state 31P NMR. Eur. J. Soil Sci. 1994, 45, 403-408. (30) Gunary, D. The availability of phosphate in sheep dung. J. Agric. Sci. 1968, 70, 33-38. (31) Harris, W. G.; Wang, H. D.; Reddy, K. R. Dairy manure influence on soil and sediment composition: implications for phosphorus retention. J. Environ. Qual. 1994, 23, 1071-1081. (32) Alvarez, R.; Evans, L. A.; Milham, P. J.; Wilson, M. A. Effects of humic material on the precipitation of calcium phosphate. Geoderma 2004, 118, 245-260. (33) Saharinen, M. H.; Vuorinen, A. H.; Hostikka, M. Effective cation exchange capacity of manure compost of varying maturity stages determined by the saturation-displacement method. Commun. Soil Sci. Plant Anal. 1996, 27, 2917-2923. (34) Haynes, R. J.; Williams, P. H. Long-term effect of superphosphate on accumulation of soil phosphorus and exchangeable cations on a grazed, irrigated pasture site. Plant Soil 1992, 142, 123133. (35) Rowarth, J. S.; Gillingham, A. G.; Tillman, R. W.; Syers, J. K. Release of phosphorus from sheep faeces on grazed, hill country pastures. N. Z. J. Agric. Res. 1985, 28, 497-504.
Received for review June 8, 2005. Revised manuscript received September 15, 2005. Accepted September 21, 2005. ES0510820
