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


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
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 ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lombi, E.
Right arrow Articles by Holloway, R. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Lombi, E.
Right arrow Articles by Holloway, R. E.
Agricola
Right arrow Articles by Lombi, E.
Right arrow Articles by Holloway, R. E.
Related Collections
Right arrow Nutrient Management
Right arrow Plant Nutrition
Right arrow Phosphorus
Published in Soil Sci. Soc. Am. J. 68:682-689 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

DIVISION S-8—NUTRIENT MANAGEMENT & SOIL & PLANT ANALYSIS

Mobility and Lability of Phosphorus from Granular and Fluid Monoammonium Phosphate Differs in a Calcareous Soil

E. Lombi*,a, M. J. McLaughlina, C. Johnstona, R. D. Armstrongb and R. E. Hollowayc

a CSIRO Land and Water, PMB 2 Glen Osmond, SA 5064, Australia
b Agriculture Victoria, Victorian Institute for Dryland Agriculture, Natimuk Rd, PB 260, Horsham, VIC 3400, Australia
c South Australian Research and Development Institute, Minnipa Agricultural Centre, PO Box 31, Minnipa SA 5654, Australia

* Corresponding author (enzo.lombi{at}csiro.au).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Phosphorus availability is a major factor limiting crop production in highly calcareous soils. Recent field trials on calcareous soils in southern Australia have shown that fluid fertilizers may provide a useful alternative to granular fertilizer products. Fluid sources of P enhance P uptake and yield when compared with granular fertilizers applied at the same rate. This work aimed to compare the behavior of one fluid (technical grade monoammonium phosphate, TG-MAP) and one granular (monoammonium phosphate, MAP) form of P fertilizer in a highly calcareous soil. Changes in soil pH, P diffusion, solubility, and lability (using isotopic dilution techniques) were determined at different distances from the point of application over 5 wk. Furthermore, reaction products in MAP granules were investigated using spectroscopic techniques. The results indicated that P from fluid TG-MAP diffused more and was more available than P supplied as granular MAP. Also, X-ray diffraction (XRD) and energy dispersive X-ray microanalyses (EDXMA) of the MAP granules indicated that a significant percentage (12%) of the initial P remained in the granules even after 5 wk of incubation in the soil. The enhanced P availability of fluid fertilizers observed in field trials compared with granular forms is discussed in relation to differences in the dissolution, diffusion, and reaction processes in soils.

Abbreviations: APP, ammonium polyphosphate • DAP, diammonium phosphate • EDXMA, energy dispersive X-ray microanalysis • FAO, Food and Agricultural Organization • MAP, monoammonium phosphate • TG, technical grade • TSP, triple superphosphate • XRD, X-Ray diffraction


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
RESTRICTED P SUPPLY is one of the most common factors limiting crop production. To improve crop growth, farmers have used large amounts of P fertilizers over many decades. However, only a small fraction of the P applied with fertilizers is taken up by crops in the year of application, and the effectiveness of any residual P fertilizer declines with time. This is a particular problem in highly P-sorbing soils such as very calcareous or strongly weathered acidic soils. Calcareous soils are widespread throughout the world. The UN Food and Agriculture Organization (FAO) estimated the extent of calcareous soils at 800 million hectares worldwide, mainly concentrated in areas with arid or Mediterranean climates (Land, FAO, and Plant Nutrition Management, 2000). These soils are important in terms of agricultural production in many areas of the world. For instance in South Australia, about 40% of the wheat (Triticum aestivum) is produced on the Eyre Peninsula, which contains over a million hectares of calcareous soils (Holloway et al., 2001). However, the efficiency of P fertilizers in these soils is generally very low because P applied to the soil reacts with Ca forming minerals such as dicalcium phosphate dihydrate, octocalcium phosphate, and ultimately hydroxyapatite (Lindsay, 1979; Sample et al., 1980; Freeman and Rowell, 1981; Tunesi et al., 1999). Also, in calcareous soils the concentration of available or extractable orthophosphate continues to decrease over time (Hooker et al., 1980; Ibrahim and Pratt, 1982; Ryan et al., 1985).

Recently, Holloway et al. (2001) reported that, on calcareous and alkaline soils, liquid MAP (TG-MAP) was 4 to 15 times more effective in increasing the grain yield of wheat than the granular products, dependent on the rate of P fertilization. These field results were confirmed in a pot trial where different granular (triple superphosphate [TSP]; MAP and diammonium phosphate [DAP]) and fluid fertilizers (orthophosphoric acid; TG-MAP; ammonium polyphosphate [APP]) were homogeneously mixed in the soil to minimize placement effects. This suggested that a chemical effect, or a combination of chemical and physical factors, might be responsible for the relatively greater efficiency of fluid fertilizers in calcareous soils. However, so far these processes are not well understood.

In this work we present results examining the diffusion of P from granular MAP and fluid TG-MAP in a highly calcareous soil (67% CaCO3), representative of large areas of cereal-growing soils in southern Australia and in other Mediterranean climates. An isotopic dilution technique was used to determine the lability (potential availability) of P at different distances from the MAP granule or the fluid MAP injection point. Finally, the reaction products inside the fertilizer granules were investigated using XRD and EDXMA to assess whether precipitation reactions were responsible for an incomplete dissolution of granular MAP. This work aimed to improve our understanding of the physical and chemical processes underlying the differential efficiency of fluid and granular P fertilizers on highly calcareous soils.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
A gray calcareous sandy loam soil (Calcixerollic xerochrepts; Soil Survey Staff, 1992), collected from upper Eyre Peninsula, South Australia, was used in the experiment. This soil contained 670 g CaCO3 kg–1 (Martin and Reeve, 1955), had a pH of 8.0 (1:5 in water, Rayment, and Higginson, 1992), a Colwell extractable P of 39 mg kg–1 (which is considered a medium-low soil P status for grain crops, Colwell, 1963) and total P of 236 mg kg–1 (Zarcinas et al., 1996). In previous field experiments on this soil, fluid fertilizers outperformed granular P fertilizers in terms of wheat grain yield and P effectiveness (Holloway et al., 2001).

The soil was air-dried and sieved <2 mm. Plastic Petri dishes (8.7 cm in diameter, 1.1 cm high) were filled with 78 g of dry soil per dish to obtain a soil density of 1.2 g cm3. The soil was then wetted to 60% of its water holding capacity (Jenkinson and Powlson, 1976) by dripping distilled water onto the soil. The Petri dishes were closed, sealed with Parafilm (Structure Probe, Inc., West Chester, PA) and left to equilibrate overnight. The following day the Petri dishes were opened and four treatments, replicated five times, were prepared:

  1. MAP control: 42 mg of finely ground (<53 µm) commercial MAP (10:22:0) was thoroughly mixed with the soil.
  2. MAP Granule: An intact granule (2–3 mm) of commercial MAP (42 ± 0.5 mg) was placed in the center of each Petri dish.
  3. MAP Powder: Commercial MAP granules were finely ground (<53 µm) and 42 mg was introduced into the center of the Petri dish.
  4. TGMAP: Commercial TG-MAP (12:26:0) dissolved in water. In this case aliquots of TG-MAP (36 mg) containing the same amount of P as in 42 mg of MAP were diluted with 200 µL with distilled water (equivalent to 1% of the total water added to each Petri dish) and injected using a needle in the center of the Petri dish.

To balance the P/N ratio in all treatments 0.63 mg of ground (<53 µm) urea (46% N) was added to the MAP powder or in the vicinity of MAP granules. After the treatments were prepared the Petri dishes were closed, sealed with Parafilm, and incubated in the dark for 5 wk in a controlled environment (25°C/20°C day/night temperature, 16-h day period). At the end of the equilibration period the Petri dishes were opened and concentric rings of soil, centered around the granule (for MAP) or the injection point, were removed using a series of plastic cylinders. These cylinders were driven into the soil one at a time, starting with the smaller one, and all soil inside the cylinder was removed. This procedure is similar to the method used by Izaurralde et al. (1986) and Kouboura et al. (1995) to sample soil at different distances from fertilizer bands. In our experiment, we obtained sections of soil collected between 0- to 7.5-, 7.5- to 13.5-, 13.5- to 25.5-, and 25.5- to 43-mm radius from the granule or injection point. These soil sections were oven dried (40°C), weighed and the pH of soil subsamples measured in a 1:5 soil/water extract (Rayment and Higginson, 1992). A subsample of the soil sections was digested with aqua regia and the total P measured by inductively coupled plasma atomic emission spectrometry (ICP-AES, Spectroflame Modula, Spectro). Another portion of dry soil was used to measure labile P (E value) using an isotopic dilution technique (Salcedo et al., 1991). Briefly, 2 g of soil was equilibrated for 24 h with 20 mL of deionized water. Carrier-free 32P (30 KBq) was then added to each sample and the labeled suspension was placed in an end-over-end shaker for a further 24 h. The soil suspensions were then filtered through 0.2-µm membrane filters (Sartorius) and analyzed for P in solution using the procedure of Murphy and Riley (1962). The 32P activity in the filtrates was measured by Cerenkov counting (RackBeta II, Wallac). The exact total activity introduced in each sample (R) was determined by analyzing spiked solutions, without soil, in parallel to the soil suspensions. The E values (mg P kg–1) were calculated by applying the isotopic dilution principle, which postulates that the specific activity of isotopically exchangeable P in soil is the same as that present in solution. Hence:

where E represents the amount of isotopically exchangeable P (mg kg–1) (Hamon et al., 2002), Cs is the concentration of P in the soil extract (mg L–1), R and r are the initial amount of radioactivity introduced into the system and the radioactivity remaining in solution after 24 h respectively, and 10 is the dilution factor.

Eight additional Petri dishes were prepared using the same procedure as described above, each containing a granule of MAP. After 5 wk the granules were extracted and adhering soil particles were carefully removed using a magnifying glass and tweezers. The granules were then oven dried (40°C) and weighed. Four granules were ground in an agate mortar and pestle and lightly pressed into aluminum sample holders for XRD. X-ray diffraction analysis patterns were recorded with a Philips PW1800 microprocessor-controlled diffractometer (PANalytical B.V., Almelo, The Netherlands) using Co K{alpha} radiation, variable divergence slit, and graphite monochromator. The diffraction patterns were recorded in steps of 0.05° 2{theta} with a 3.0-s counting time per step, and logged to data files for analysis using an in-house developed XPLOT software. The remaining granules were cross-sectioned and then oriented and mounted onto aluminum specimen holders using Araldite. The samples were subsequently placed in an oven at 105°C for 15 min to aid polymerization. Where imaging of the surface topography was required, specimens were sputter coated with 20 nm of gold using an EmScope SC500 coating unit (Polaron, Watford, UK), to provide electrical conductivity and maximize secondary electron (SE) signal yield. Where the composition was required, specimens were coated by evaporative means with 30 nm of C, using an EmScope SC500 coating unit, to provide electrical conductivity and maximize backscattered electron (BE) phase contrast. Carbon coating also minimizes extraneous X-ray peaks within the characteristic X-ray spectrum. The specimens were then placed into a Phillips XL30 FEG-SEM (PANalytical B.V., Almelo, The Netherlands), with an attached EDAX DX4 energy dispersive X-ray system (EDAX Inc., Mahwah, NJ) and examined using a primary electron beam energy of 20 KeV. Semi-quantitative elemental analysis was performed by selecting an area of approximately 200 x 100 µm and collecting spectra for 100 s. Spectra were analyzed using the ZAF correction software (CITZAF Version 3.03, John T. Armstrong, available at http://www2.arnes.si/~sgszmera1/others/others/mlist.html). Two-dimensional distribution patterns for P and other elements were recorded by scanning an area of the specimen repeatedly for 5 h and integrating the counts of the elements of interest within their respective spectrum windows into dot-maps.

The experimental design was a complete randomized design. Data were analyzed using ANOVA with fertilizer type as the main treatment and soil sections as subplot treatments. Least significant difference (LSD) was used for comparison between the treatment means. All statistical analyses were performed using the Genstat 5 package (Genstat 5 Committee, 1993).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Soil pH at Different Distance from the Fertilizers
In the MAP control treatment, in which MAP powder was mixed with the whole soil, no changes in pH between sections were observed. In all the other treatments soil pH decreased significantly in the 0- to 7.5-mm section by 0.23 to 0.24 units compared with the original soil pH (8.00) irrespective of the fertilizer source (Table 1). Changes in pH in the vicinity of P fertilizers are well documented. Racz and Soper (1967) reported an initial increase in pH in the vicinity of MAP pellets, probably as a result of the pH of the fertilizer itself and the formation of K and NH4 bicarbonates. However, with increasing incubation time the pH near MAP fertilizers generally decreases due to nitrification processes (Hanson and Westfall, 1985; Moody et al., 1995). Sample et al. (1979) reported that soil pH dropped very sharply (up to two pH units) in the vicinity of various P fertilizer solution fronts due to movement and reactions of P, NH4, and exchangeable Ca. However, due to the large pH buffer capacity of the highly calcareous soil used in our experiment, changes in pH were small and limited to the soil section where the fertilizer was applied.


View this table:
[in this window]
[in a new window]
  		Table 1. Soil pH at different distances from the point of fertilizer application (average of five replicates) and dry weight of each section (as % of total dry soil in each Petri dish). LSD for fertilizer treatment x soil section was 0.10. Control soil pH = 8.0.

 
Total Phosphorus Distribution
A mass balance for P was calculated by adding the P content of all sections for each Petri dish. The total amount of P in each Petri dish was similar in every treatment and varied between 26.1 mg in the MAP control treatment to 27.4 mg in the TG-MAP treatment. In the MAP control treatment, where the fertilizer was homogenized with the whole soil, the concentration of total P was similar in all the sections sampled (343 ± 14 mg kg–1). Concentrations of P in the other treatments decreased with increasing distance from the central section where the fertilizer was added (Fig. 1a) . This decrease was much more marked in the MAP granular and powder treatments than in the TG-MAP treatment. The P concentration in the central section (0–7.5 mm) was significantly greater (P ≤ 0.001) in the MAP granular and powder treatments (almost 3000 mg P kg–1) than in the TG-MAP treatment (1083 mg P kg–1). In contrast, P concentrations were significantly greater in the 7.5- to 13.5- and 13.5- to 25.5-mm sections collected from the TG-MAP treatment than in the MAP granular and powder treatments. The P concentration in the 25.5- to 43-mm section was similar (242 ± 6 mg kg–1) in all treatments, with the exception of MAP control, and reflected the original P concentration of the soil.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1. (a) Total P and (b) percentage of P from fertilizer in soil sections collected at different distances from the point of fertilizer application. The percentage of P from fertilizer in each section (%Pf S1–4) was calculated as follows: %Pf Si = [(Pf)Si x Wi]/{sum}i = 1–4 [(Pf)Si x Wi] where, i is the soil section (1–4), (Pf)Si and Wi are the concentration of fertilizer P and the soil weight of each section respectively. (Pf)Si is calculated by subtracting the P concentration of the untreated soil (236 mg kg–1) from the concentration of the fertilized soil. Error bars represent standard errors of five replicates. LSD for fertilizer treatment x soil section was (a) 68 and (b) 1.3.

 
The distribution of the added fertilizer P in the different sections of soil collected from the Petri dishes was calculated using the measured total P concentration and the background concentration. The results, expressed as a percentage of the P added in the fertilizers are reported in Fig. 1b. In the MAP control treatment, as expected, the distribution of P from the fertilizer closely matched the weight distribution of the soil among the different sections as reported in Table 1. When MAP was applied, as either granular or powdered forms in the center of the Petri dish, most of the P from the fertilizer (84%) remained within 0 to 7.5 mm of the point of application and only 12% diffused to the 7.5- to 13.5-mm section. This indicated that neither the initial compaction nor the coating of the granules (to suppress dust during transport) affected the diffusion of P from the fertilizer. The distribution pattern of P within the Petri dish was different for fluid TG-MAP, where the amount of P remaining within 7.5 mm from the injection point was significantly less (P ≤ 0.001) than for MAP treatments. In contrast, significantly more P (P ≤ 0.05) from the fertilizer was recovered in the outer soil sections indicating a greater mobility of P derived from TG-MAP. The diffusion of P from MAP granules (or powders) was detectable in the first three sections of the Petri dishes indicating a migration of P up to 25.5 mm in 35 d. This result is comparable with those of Lawton and Vomocil (1954) who reported a migration of 17 to 24 mm of fertilizer P (superphosphate) in two loamy soils in 27 d. Similarly, Blanchard and Caldwell (1966) reported diffusion of P to 20 to 30 mm from monocalcium phosphate pellets over 2 wk in a clay loam soil.

Even though the maximum distance to which P diffused from MAP and TG-MAP was not significantly different, the pattern of P distribution was. Most of the P from MAP remained near the point of application whereas more P from TG-MAP diffused away from the injection point. A similar finding was reported by Khasawneh et al. (1974) who compared the mobility of P when supplied as granular (DAP) or fluid (APP and triammonium pyrophosphate) forms. However in their study, the source of P in the fluid fertilizer was polyphosphate, not orthophosphate, and the fertilizers were homogeneously applied to the top of columns packed with a sandy loam soil (pH 6.0).

Difference between granules and liquid in the rate of P diffusion may in part be due to capillary mass flow of water toward the highly hygroscopic phosphate granules (Lawton and Vomocil, 1954; Williams, 1969). This mass flow would be in an opposite direction to the diffusion of P from the granule. Lawton and Vomocil (1954) reported that the water content of superphosphate granules increased from an initial 1.7 to 16.2% in a very dry soil to 29.8% in a soil containing 7.7% moisture over 24 h. Khasawneh et al. (1974) reported a net movement of water toward P fertilizers due to the high solute concentration in soil water in the salt-affected zone. In the case of TG-MAP, the amount of water added with the fertilizer may have decreased the osmotic flow of soil moisture in a direction opposite to that of P diffusion.

Analyses of Monoammonium Phosphorus Granules Incubated in the Soil
The large amount of P remaining in the central section of the MAP treatment was probably the result of both a slower diffusion of orthophosphate from granular MAP, and an incomplete dissolution of the MAP granules/powder. Collection of MAP granules from four Petri dishes after 5 wk of exposure revealed that a significant percentage of P (12.3% of the initial P concentration) remained in the granule (Table 2). At the same time the content of other elements such as Fe, Al, and especially Ca increased. Phosphate has been shown to diffuse quickly into the soil from fertilizer granules. For instance, Lawton and Vomocil (1954) reported that 50 to 80% of water soluble P moved out of TSP granules in 24 h. The insoluble residue observed in the granules after 5 wk is likely the result of the high soil alkalinity that may have prevented dissolution of the citrate-soluble P fraction contained in the MAP granules. Monoammonium phosphate granules contain various (Ca, Mg)(NH4)(Fe, Al)(PO4)(F, OH)H2O compounds that are insoluble in water and can comprise up to 10% of the total P content of fertilizers (Gilkes and Mangano, 1983). Prochnow et al. (2001), using a combination of XRD and EDXMA, identified various water insoluble tri-iron phosphate compounds in the granules of single superphosphate produced from Brazilian phosphate. X-ray diffraction analyses of granules exposed in the soil revealed that crandallite-like minerals [CaAl3(PO4)2 (OH)5·(H2O)] were the dominant residual P phase in the granules (Fig. 2) . Traces of quartz and gypsum were also found, and these minerals were also present in MAP granules not exposed to soil. In contrast, no clay minerals were present in the exposed granules indicating that adhering soil particles were successfully removed. Sikora et al. (1992) reported that the apparent solubility of MAP impurities increased with pH. However, they did not report crandallite as an impurity in MAP (Sikora et al., 1989). Only limited data are available in relation to the availability of crandallite minerals and these are restricted to acidic or sandy soils (McLaughlin et al., 1992; Bolland, 1994, 1996). However, it is likely that in highly calcareous soils these minerals will be only sparingly soluble. On the other hand, some of these residual mineral P phases in the MAP granules may have formed in situ by migration of Al, Ca, Fe, and other ions into the granule over time. For instance, the statistically significant (P ≤ 0.001) increase in Ca (and to a lesser extent of Fe and Al) content in the granule due to diffusion of Ca inside the granule may cause precipitation of Ca-P compounds. This is supported by the results obtained using scanning electron microscopy in combination with EDXMA of the exposed granules. The scanning electron micrograph of a cross-section of a granule and dot-maps representing semi-quantitative distribution of P, Fe, and Ca inside the granule are reported in Fig. 3 . The granule structure reflects its partial dissolution during 5 wk of incubation in the soil with evidence of weathering rings. Phosphorus appears to be quite homogeneously distributed across the section of the granule whereas Ca and Fe appear more concentrated in the outer regions. This is consistent with the chemical analyses (Table 2), which indicated a possible diffusion of these elements from the soil into the granule.


View this table:
[in this window]
[in a new window]
  		Table 2. Weight and elemental chemical composition of monoammonium phosphate (MAP) granules before and after exposure to soil for a period of 5 wk. Numbers in parenthesis are SE of four replicates. Data are expressed as milligram per granule.

 


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 2. X-ray diffraction analyses of monoammonium phosphate (MAP) granules incubated for 5 wk in the soil.

 


View larger version (110K):
[in this window]
[in a new window]
  		Fig. 3. Scanning electron micrograph of a cross-section of a monoammonium phosphate (MAP) granule incubated for 5 wk in the soil. Dot-maps represent semi-quantitative distribution of P, Fe, and Ca inside the granule.

 
Solubility and Lability of Phosphorus
A number of extraction techniques have been used to estimate the mobility and potential bioavailability of P in soils. However, a recent study conducted in alkaline Australian soils has highlighted the potentially limited reliability of extraction methods, such as bicarbonate- and calcium lactate-extractable P, as predictors of availability (Bertrand et al., 2003). In our study we chose water-extractable P and isotopically exchangeable P (E value) as measures of potential P availability.

Water-soluble P (soil/water ratio 1:10) was similar in the central section (0–7.5 mm) of the granular and powder MAP treatments and the TG-MAP treatment (Fig. 4) . In the TG-MAP treatment the solubility of P was significantly (P ≤ 0.001) greater than in the other treatments in the zone between 7.5 and 25.5 mm from the site of fertilizer application. In particular, the concentration of water-soluble P in the 7.5- to 13.5-mm section of the TG-MAP treatment was almost six times greater than in the MAP granular and powder treatments. As expected, the solubility of P was similar in all sections of the MAP control treatment (Fig. 4).



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 4. Concentration of water-soluble P in soil sections collected at different distances from the point of fertilizer application. Error bars represent standard errors of five replicates. LSD for fertilizer treatment x soil section was 2.8.

 
Isotopic dilution techniques (E value) have been used to assess potential plant availability of P in soil, by measuring the quantity of isotopically exchangeable P present in the soil and in soil solution (Fardeau et al., 1996; Hamon et al., 2002). This method also provides an opportunity to gain an insight into the mechanism of P retention in the soil. The use of E values permitted separation of chemical processes such as adsorption, in which P remains exchangeable, from precipitation, or intraparticle diffusion, where soluble P is converted to fixed, or slowly exchangeable, forms of P. The distribution of labile P reflects the source of fertilizer used (Fig. 5a) . When MAP, either as a granule or as powder, was applied to the center of the Petri dish, labile P was significantly greater than in any other treatments in the 0- to 7.5-mm section. However, this situation was different in the two middle sections where P lability was significantly greater in the TG-MAP treatment. In Fig. 5b the labile P is expressed as a percentage of total P in each section. In the MAP control treatment the relative P lability is approximately 20% in each section. By contrast, when MAP was applied in the central section, the relative P lability was only about 15% in the 0- to 7.5-mm section and was maximal (23%) in the 7.5- to 13.5-mm section. The relative lability of P in the most central sections of the TG-MAP was greater than in any other section (over 25%) and significantly larger than in the MAP treatments. Similarly, almost 20% of the total P was labile in the 13.5- to 25.5-mm section of the TG-MAP treatment compared with only 5% in the same section of the MAP granular and powder treatments (Fig. 5b).



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 5. (a) Labile P and (b) percentage of labile P in soil sections collected at different distances from the point of fertilizer application. The percentage of labile P was calculated dividing the E values for each section by the corresponding total P concentration. Error bars represent standard errors of five replicates. LSD for fertilizer treatment x soil section was (a) 14.8 and (b) 3.1.

 
In the central section of the MAP granule and powder treatments, where over 80% of the P is located (Fig. 1b), only 15% of this P is isotopically exchangeable and can be considered potentially available to plants. The low E values of P in the central section probably reflect the precipitation of water soluble P in the MAP as mixed Ca/Mg/NH4+ phosphates (Sample et al., 1980), as well as the presence of insoluble crandallite-like materials in the MAP itself. In the TG-MAP treatment, not only did more P diffuse from the zone of fertilization compared with the granular MAP treatments, but also this P was more soluble and potentially available (labile). The specific mechanism is possibly related to the more pronounced diffusion of P from TG-MAP than for the MAP treatments. A greater rate of diffusion outwards may reduce the extent of precipitation reactions occurring when the P sorption capacity of the soil is exceeded and phosphate ions react with metal ions. This is supported by the fact that the lability of the fertilizer P in each section seems to correlate well with the corresponding total P from the fertilizer as reported in Fig. 6 . The large error bars in the most external sections of the MAP control and powder and TG-MAP treatments can be explained by the way in which the labile P from fertilizer is calculated. This calculation is based on a number of independent measurements (total and native P, total and native E value) and the variability becomes larger when the contribution of P from the fertilizer is small. If the data with large error bars are omitted a significant negative relationship (Lability of fertilizer P = –0.0152 total P + 53.9, R2 = 0.837) is found between total fertilizer P in each section and the corresponding lability of fertilizer P.



View larger version (47K):
[in this window]
[in a new window]
  		Fig. 6. Lability of fertilizer P (%) in soil sections collected at different distances from the point of fertilizer application. The lability of fertilizer P was calculated by subtracting the soil E value of the untreated soil from the total E value of each section and dividing this number by the concentration of fertilizer P in the respective section (calculated as reported in the legend of Fig. 1). Numbers on the top of each bar represent the corresponding total fertilizer P concentration in milligrams per kilogram (mg kg–1). Error bars represent standard errors of five replicates. LSD for fertilizer treatment x soil section was 13.5.

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The results of our study provide an insight into some of the chemical and physical processes occurring when two products that contain the same form of P, but delivered in either granular or liquid form, are applied to a calcareous soil. In particular diffusion, solubility and lability of P from TG-MAP appeared to be enhanced in comparison with granular MAP. These differences may reflect the difference in moisture gradient, mobility, and reaction products in the zone immediately surrounding the point of fertilizer application. Also, XRD, EDXMA, and chemical data suggest that incomplete dissolution of MAP granules in these soils, even after incubation for 5 wk at high soil water content, is due partly to the presence of insoluble crandallite in the granule, and also due to the precipitation in situ of similar minerals resulting from the diffusion of Ca and Al into the granule. In contrast there is significantly less fixation of soluble P when fluid MAP is applied, and hence greater concentrations of labile P. The increased diffusion, solubility and potential bioavailability of P when supplied as TGMAP provide an explanation for the differential crop response between granular and fluid MAP fertilizers observed in field trials (Holloway et al., 2001).


   ACKNOWLEDGMENTS
 
The authors thank the Grains Research and Development Corporation (project No. CSO231), the South Australian Grain Industry Trust Fund (SAGITF) and the Fluid Fertilizer Foundation for providing funding to support this research program. We also thank Stuart McClure and Marc Raven for technical assistance with EDXMA and XRD and Rebecca Hamon for valuable comments on the manuscript.

Received for publication May 22, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
G. M. Hettiarachchi, M. J. McLaughlin, K. G. Scheckel, D. J. Chittleborough, M. Newville, S. Sutton, and E. Lombi
Evidence for Different Reaction Pathways for Liquid and Granular Micronutrients in a Calcareous Soil
Soil Sci. Soc. Am. J., January 11, 2008; 72(1): 98 - 110.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
N. J. Mathers, D. M. Nash, and P. Gangaiya
Nitrogen and Phosphorus Exports from High Rainfall Zone Cropping in Australia: Issues and Opportunities for Research
J. Environ. Qual., October 16, 2007; 36(6): 1551 - 1562.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
E. Lombi, K. G. Scheckel, R. D. Armstrong, S. Forrester, J. N. Cutler, and D. Paterson
Speciation and Distribution of Phosphorus in a Fertilized Soil: A Synchrotron-Based Investigation
Soil Sci. Soc. Am. J., October 27, 2006; 70(6): 2038 - 2048.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
M. J. Ottman, T. L. Thompson, and T. A. Doerge
Alfalfa Yield and Soil Phosphorus Increased with Topdressed Granular Compared with Fluid Phosphorus Fertilizer
Agron. J., June 5, 2006; 98(4): 899 - 906.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
G. M. Hettiarachchi, E. Lombi, M. J. McLaughlin, D. Chittleborough, and P. Self
Density Changes around Phosphorus Granules and Fluid Bands in a Calcareous Soil
Soil Sci. Soc. Am. J., April 19, 2006; 70(3): 960 - 966.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
T. M. McBeath, R. J. Smernik, E. Lombi, and M. J. McLaughlin
Hydrolysis of Pyrophosphate in a Highly Calcareous Soil: A Solid-State Phosphorus-31 NMR Study
Soil Sci. Soc. Am. J., March 29, 2006; 70(3): 856 - 862.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
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 ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lombi, E.
Right arrow Articles by Holloway, R. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Lombi, E.
Right arrow Articles by Holloway, R. E.
Agricola
Right arrow Articles by Lombi, E.
Right arrow Articles by Holloway, R. E.
Related Collections
Right arrow Nutrient Management
Right arrow Plant Nutrition
Right arrow Phosphorus

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
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome