| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Soil and Land Systems, School of Earth and Environmental Sciences, The Univ. of Adelaide, PMB 1, Waite Campus, Glen Osmond, SA 5064, Australia
b CSIRO Land and Water, PMB 2, Glen Osmond, SA 5064, Australia
* Corresponding author (therese.mcbeath{at}adelaide.edu.au)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: CP, cross polarization • DP, direct polarization • IC, ion chromatography • ICP–AES, inductively coupled plasma atomic emission spectroscopy • NMR, nuclear magnetic resonance • Pobs, percentage of sample P observable by spin counting using 31P NMR • SSB, spinning side band
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Fluid P fertilizers account for approximately 23% of America's and 10% of Canada's total P fertilizer tonnage (Jennings, 2003). The most widely used fluid P fertilizer in the USA is APP (Mortvedt et al., 1999; Jennings, 2003). Polyphosphate fertilizers have been widely used in broadacre agriculture in other parts of the world, but their introduction to Australian agriculture is very recent. Preliminary evidence suggests that polyphosphates may offer substantial benefits over traditional granular P fertilizers in South Australian calcareous soils. For example, in field trials on Eyre Peninsula in 2002, APP provided 36% greater wheat grain yield than granular P fertilizer applied at an equivalent rate of 8 kg P ha–1 on a red calcareous soil (Holloway et al., 2002).
At the point of sale, approximately 30 to 40% of the fertilizer P in commercial APP sold in Australia is present as orthophosphate (OP), 50 to 55% is present as pyrophosphate (PP) and the remainder exists as tripolyphosphate and more condensed forms of P (data not shown). Hydrolysis reactions of polyphosphate fertilizer in soil convert more condensed P species (two or more OP groups linked by oxygen bridges) to less condensed forms of P (Lindsay, 1979; Dick and Tabatabai, 1986). Previous data indicates that pyrophosphate constitutes 70 to 90% of the polyphosphate in APP (Khasawneh et al., 1974). Therefore the most common hydrolysis reaction of polyphosphate fertilizer in soil is the conversion of pyrophosphate to orthophosphate (Sutton and Larsen, 1964; Hashimoto et al., 1969).
The hydrolysis reaction of pyrophosphate is shown below in Eq. [1]:
 | [1] |
Here we describe the use of solid-state 31P NMR spectroscopy for investigating pyrophosphate hydrolysis in a highly calcareous soil. We compared solid-state 31P NMR to speciation of P by ion chromatography, which enables the direct measurement of pyrophosphate-P and orthophosphate-P, on a NaOH extract. While the NaOH extraction is conventionally used for measurement of organic P it has been used with subsequent speciation to measure pyrophosphate-P present in soil extracts (Shand et al., 2000; Turner et al., 2003). Solution-state NMR spectroscopy has been used to investigate the hydrolysis of pyrophosphate in solution (Subbarao et al., 1977), while solid state 31P NMR has been used to investigate a polyphosphate-chitosan complex as a source of P (Frossard et al., 1994). However this is the first time solid-state NMR spectroscopy has been used to investigate the hydrolysis of pyrophosphate as a nutrient source in soil. Understanding the nature of hydrolysis reactions of pyrophosphate in Australian soil types is necessary to elucidate the mechanisms underlying the superior agronomic performance of polyphosphate fertilizer as compared with traditional P fertilizers on calcareous soil types.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The total P content of the soil was 339 mg kg–1, as determined by inductively coupled plasma atomic emission spectroscopy (ICP–AES, Spectroflame Modula, Spectro) following digestion in aqua regia (HNO3/HCl, 1:3) (Zarcinas et al., 1996). The resin-exchangeable P content was 7 mg kg–1, as determined using anion-exchange resin strips (McLaughlin et al., 1994).
Soil Incubations
Soil (50 g) was wetted up to 30% of water holding capacity and allowed to incubate for 1 wk to allow microbially driven dissolved organic matter flushes to settle before adding P (Jenkinson and Powlson, 1976). A solution of sodium pyrophosphate (11.82 g L–1, 8.46 mL per 50 g soil) was then added. This resulted in a P addition of 2000 mg P kg–1 of soil, and also increased the moisture content to 75% of water holding capacity. This P concentration was chosen as it is similar to that found in close proximity to fertilizer granules, or in a fluid fertilizer band, in soil (Lombi et al., 2004). Separate soil samples were incubated for 1, 3, 7, 14, and 21 d. There were two replicates of each treatment. The incubations were staggered, with the longest incubation started first and all incubations timed to finish on the same day. Soils were incubated under aerobic conditions having constant temperature (20 ± 2°C) and humidity.
Solid-State 31P NMR Spectroscopy
Solid-state 31P NMR spectra were acquired with magic angle spinning (MAS) and high-power 1H decoupling on a Varian Unity INOVA 400 spectrometer (Varian, Palo Alto, CA) with a Doty Scientific supersonic MAS probe (Doty Scientific, Columbia, SC) at a 31P frequency of 161.9 MHz. Samples were packed into 7-mm diam. cylindrical zirconia rotors with Kel-F end-caps (Doty Scientific, Columbia, SC) and spun at 5 kHz at the magic angle of 54.7°.
Two techniques were used to acquire solid-state 31P NMR spectra—cross polarization (CP) and direct polarization (DP). The CP technique involves transfer of magnetization (coherence) from 1H to 31P nuclei. The main advantage of the CP technique is that it enables more rapid accumulation of scans because 1H nuclei regain equilibrium magnetization more rapidly than do 31P nuclei. The main disadvantage of the CP technique is that the transfer of polarization can be inefficient for 31P nuclei remote from 1H nuclei, and hence such nuclei can be under-represented or absent from CP spectra. The DP technique, as its name suggests, involves direct polarization of 31P nuclei.
Phosphorus-31 CP NMR spectra of the soils represent the accumulation of 10 000 scans, and were acquired using a 5.1-µs 1H 90° pulse, a 1-ms contact time, and a 4-s recycle delay. The total acquisition time for 31P CP NMR spectra was around 11.1 h. Phosphorus-31 DP NMR spectra of the soils represent the accumulation of 500 scans and were acquired using a 4.2-µs 13C 90° pulse and a 100-s recycle delay. The total acquisition time for 31P DP NMR spectra was around 13.9 h. The 31P DP NMR spectra were corrected for a broad background signal by subtracting the 31P DP NMR spectrum acquired for an empty rotor.
Free induction decays were acquired with a sweep width of 50 kHz. A total of 1216 data points was collected for all spectra, representing an acquisition time of 12 ms. All spectra were zero-filled to 13 1072 data points and processed with a 100-Hz Lorentzian line broadening and a 0.010-s Gaussian broadening. Chemical shifts were externally referenced to NH4H2PO4 at 0.72 ppm (Frossard et al., 2002).
Phosphorus-31 spin counting experiments were performed using a modification of the 13C spin counting method of Smernik and Oades (2000a, 2000b). Ammonium dihydrogen phosphate (NH4H2PO4) was used as an external intensity standard (i.e., the NH4H2PO4 spectrum was acquired separately to those of the samples). The NH4H2PO4 31P CP NMR spectrum was acquired in 16 scans, using a 1-ms contact time and a 10-s recycle delay. The NH4H2PO4 31P DP NMR spectrum was acquired in one scan after equilibration for 1000 s (16.7 min). For CP spin counting experiments, differences in the rate of 1H magnetization decay during the contact time (T1
H) between the sample and the NH4H2PO4 standard were accounted for using the method of Smernik and Oades (2000a), except that a variable spin lock rather than a variable contact time experiment was used (Smernik et al., 2002). The T1H relaxation rate for NH4H2PO4 was found to be 120 ms. Uncertainty in the precision of Pobs values is estimated to be ±10% in CP and ±15% in DP (Smernik and Oades, 2000a).
Phosphorus-31 DP NMR spectra were acquired in a single scan for reference Na and Ca orthophosphate and pyrophosphate salts. The salts NaH2PO4·H2O ("Na orthophosphate," Merck, Kilsyth, Australia), CaHPO4 ("Ca orthophosphate," Aldrich, St. Louis, MO) and Na4P2O7.10H2O ("Na pyrophosphate," Aldrich, St. Louis, MO) were used as received. Calcium pyrophosphate was prepared by mixing solutions of Na4P2O7·10H2O (0.1 M, 100 mL) and CaCl2 (0.2 M, 100 mL). The resultant precipitate was isolated by filtration and washing with Milli-Q water, and dried overnight at 50°C. Analysis by x-ray diffraction (XRD) suggested that the calcium pyrophosphate was mostly amorphous with some minor halite (NaCl).
Ion Chromatography
A sodium hydroxide extraction was performed using a modified version of the method of Turner and McKelvie (2002). Soil (4 g) was treated with 1 M NaOH (40 mL), and shaken for 15 min (17 rpm), then centrifuged at 2096 x g (3000 rpm) for 30 min. The supernatant was filtered using 0.22-µm filters (Sartorius, Hannover, Germany) and diluted by a factor of 50 with distilled water. Ion chromatography was performed using a Dionex ICS 2500 system (Dionex, Sunnyvale, CA) with an anion-exchange column (AS16) to determine orthophosphate and pyrophosphate concentrations in NaOH extracts. An injection volume of 25 µL was used with a flow rate of 0.38 mL min–1. A gradient of 20 to 80 mM potassium hydroxide (KOH) was used in conjunction with internal suppression (76 mA). Orthophosphate-P and pyrophosphate-P were detected using a conductivity detector (BioLC ED50 Electrochemical Detector, Dionex).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
The spectra of the four salts differ in both the position (chemical shift) of the center band (Table 1) and in the size and pattern of the SSBs. The chemical shifts of the pyrophosphate salts are more negative (upfield) than the corresponding orthophosphate salts, and the chemical shifts of the Ca salts are more negative than the corresponding Na salts. The size and pattern of SSBs are similar for both pyrophosphate salts (Fig. 1). The center band represents 27 to 36% of total signal for the pyrophosphate salts (Table 1). The size and pattern of SSBs are quite different for the Na and Ca orthophosphate salts (Fig. 1). The center band represents 65% of total signal for Ca orthophosphate, but only 35% of total signal for Na orthophosphate (Table 1). The greater line width of the Ca pyrophosphate resonance most likely reflects the amorphous (poorly crystalline) nature of this salt.
|
Solid-State 31P NMR Spectra of Soils
Figure 2
shows the 31P CP and DP spectra for the unamended soil and soil samples amended with pyrophosphate (2000 mg P kg–1 soil) and incubated for 1 to 21 d. The 31P DP NMR spectrum of the unamended soil (Fig. 2) contains a sharp resonance at 1.7 ppm, along with associated low intensity SSBs. The size and pattern of the SSBs is similar to that for the Ca orthophosphate salt (Fig. 1), although the chemical shift is slightly different (–1.69 ppm for Ca orthophosphate, Table 1). Chemical shift is sensitive to factors such as pH and crystal packing, so a 3.4 ppm difference in chemical shift between the pure Ca orthophosphate salt and the soil resonance (which is presumably also mostly orthophosphate) is not unreasonable. The SSBs are stronger and broader in the corresponding 31P CP NMR spectrum of the unamended soil (Fig. 2), and there is a broad shoulder on the upfield side of the strongest peak. The broad shoulder and SSBs can be assigned to organic P (Dougherty et al., 2005). They are more prominent in the CP spectrum because the inorganic P is observed with very low sensitivity in the CP spectrum; the observability of P for the unamended soil was 20% by CP and 78% by DP (see below).
|
Quantification of Orthophosphate and Pyrophosphate Contents from Solid-State 31P NMR Spectra
The spectra presented in Fig. 2 clearly indicate a conversion of added pyrophosphate to orthophosphate, but they also show a decrease in overall signal intensity, especially for the CP spectra. To obtain quantitative data from the solid-state 31P NMR spectra, we first needed to establish the sensitivity of NMR detection. This was achieved by spin counting, which is simply a calibration of total signal intensity in a sample per unit P against that of a standard. Spin counting has proven very useful for identifying biases in signal distribution in solid-state 13C NMR spectra (Smernik and Oades, 2000a, 2000b), and has been applied recently to solid-state 15N (Smernik and Baldock, 2005a, 2005b) and 31P (Dougherty et al., 2005) NMR spectra. The results of spin counting, presented in terms of the relative observability of P in each sample versus that of the standard (Pobs), are shown in Table 2.
|
Both the CP and DP observabilities are higher than those reported by Dougherty et al. (2005), who found Pobs (DP) to range from 17 to 28% and Pobs (CP) to range from 7 to 12%. The low NMR observability of P in the Dougherty et al. (2005) soils was attributed to the effect of paramagnetic Fe (soil Fe content was 0.86 to 1.64% [w/w]). The calcareous soil used in this study has a much lower Fe content (0.33% [w/w]).
The NMR observability of P was higher in the soil freshly amended with pyrophosphate (Table 2) than in the unamended soil, indicating that the freshly added pyrophosphate-P was more readily observed than the native P. Phosphorus-31 NMR observability decreased with increasing time after pyrophosphate addition, especially for the CP technique (Table 2). The low and variable Pobs values for the CP technique compromised the potential to quantify orthophosphate-P and pyrophosphate-P from the 31P CP NMR spectra. On the other hand, the Pobs values for the DP technique are high enough (71–88%) to attempt quantification of orthophosphate-P and pyrophosphate-P from the DP spectra, although the potential for bias due to the "NMR invisible P" must remain a consideration.
Quantification of the relative contributions of orthophosphate and pyrophosphate to the 31P NMR spectra was achieved by spectral subtraction of the 31P DP NMR spectrum of the unamended soil (which contains only orthophosphate) from the 31P DP NMR spectra of the unamended soil. Increasing proportions of NMR spectrum of the unamended soil were subtracted until the central orthophosphate resonance was cancelled out. The contribution of pyrophosphate was determined by integration of the resultant "difference" spectra.
Figure 3 shows that 1 d after adding 2000 mg of pyrophosphate-P kg–1 soil, 1660 mg kg–1 of pyrophosphate-P was detected by NMR, and after a 21-d incubation 870 mg kg–1 of pyrophosphate-P was still detected. On the other hand, after 1 d of incubation, the concentration of orthophosphate-P detected by NMR increased from the 260 mg kg–1 of native orthophosphate-P to 390 mg kg–1, and continued to increase to 810 mg kg–1 after a 21-d incubation. The concentration of undetected P also more than doubled during the course of the incubation (Fig. 3).
|
Hydrolysis Rate and Pyrophosphate Half-Life
The decay of NMR observable pyrophosphate-P did not appear to be exponential, the rate of decay being proportionately faster earlier in the incubation. Nonetheless, the (nonlinear) half-life of added pyrophosphate can be estimated from 31P NMR (Fig. 3) as 15 to 21 d from the time of application. The (nonlinear) half-life of extractable pyrophosphate-P can be estimated as 3 to 7 d from the ion chromatography data shown in Fig. 3. This is much shorter than the value indicated by the NMR data (15–21 d). This indicates that much of the nonextractable P (i.e., undetected by NaOH extraction followed by ion chromatography) at longer incubation times is pyrophosphate-P.
The half-life of pyrophosphate determined by NMR (15–21 d) is comparable with the results of Ahmad and Kelso (2001) who found that approximately 50% of pyrophosphate added was hydrolyzed 21 d after addition in a soil with 14% CaCO3 w/w and a pH of 8.49. In comparison they tested non-calcareous soils (pH 4.51 and 5.93) and found that only 30 to 35% of pyrophosphate hydrolyzed over 21 d (Ahmad and Kelso, 2001). Dick and Tabatabai (1986) measured the hydrolysis of pyrophosphate in a slightly calcareous soil (7.1% CaCO3 [w/w]; pH 7.8) and found a half-life of 7 d, as compared with 40 to 45% of pyrophosphate hydrolyzed in 7 d in three non-calcareous soils (pH 5.8–6.4). In the work of Dick and Tabatabai (1986) P was extracted using a 2.5 M H2SO4, which has the ability to hydrolyze pyrophosphate in the extraction step (De Jager and Heyns, 1998). Khasawneh et al. (1979) found the half-life for pyrophosphate a fine sandy loam to be from 8 to 16 d using a sodium hydroxide extract and colorimetric measurement of P (soil pH not reported). Parent et al. (1985) measured hydrolysis of pyrophosphate in a low pH soil (pH 5.1) and found that the half-life of pyrophosphate was approximately 40 d. The pH of pyrophosphate was adjusted to the soil pH of 5.1 before the experiment was conducted, and it was extracted after addition to soil with 2.5 M H2SO4.
In all of the aforementioned experiments, the hydrolysis reaction was considered to have an initial rapid hydrolysis phase, followed by a slower hydrolysis rate, which is most likely due to the pyrophosphate being adsorbed or precipitated over time (Dick and Tabatabai, 1986; Ahmad and Kelso, 2001).
Comparison of our data with the literature shows that both soil characteristics and the extractability of P species in soils with a range of pH, CaCO3 content, and P sorption capacities has a significant influence on the resulting measurement of pyrophosphate half-life. Our comparison between a NaOH extraction and subsequent P speciation by ion chromatography with non-invasive solid-state 31P NMR spectroscopy suggests that a greater proportion of pyrophosphate added is measured by the NMR method.
The findings of this study suggest it is well worth investigating the use of solid-state 31P NMR as a technique for speciation of pyrophosphate in a range of soil types, under a range of chemical and biological conditions to ascertain the limitations of the method. Used in conjunction with speciation of pyrophosphate in soluble P and total P pools there is potential for solid state 31P NMR to provide useful information about the distribution of P species amongst pools.
Hydrolysis of pyrophosphate is considered to be the rate limiting step for pyrophosphate to be equivalent to orthophosphate as a P source (Khasawneh et al., 1979). This study shows that pyrophosphate added to a highly calcareous soil has a half-life of 14 to 21 d after addition. Early P nutrition of cereal crops is essential in dryland agriculture and is one of the reasons that fertilizer is placed in close proximity to seeds at sowing. Delays in hydrolysis of pyrophosphate could adversely affect nutrition of young seedlings, which may compromise the final yield potential of crops. One of the concerns regarding use of polyphosphate fertilizers in dryland alkaline soils was that hydrolysis of the polyphosphate ions might be delayed by low soil moisture, high soil pH, and low biological activity. Our data show that pyrophosphate hydrolysis was relatively rapid in the calcareous soil studied, and hence availability of orthophosphate to plants should not be compromised. Further work is required to assess pyrophosphate hydrolysis in a wider range of Australian soil types.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication June 9, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. J. Dougherty, R. J. Smernik, E. K. Bunemann, and D. J. Chittleborough On the Use of Hydrofluoric Acid Pretreatment of Soils for Phosphorus-31 Nuclear Magnetic Resonance Analyses Soil Sci. Soc. Am. J., June 8, 2007; 71(4): 1111 - 1118. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
