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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Liu, C.
Right arrow Articles by Zhou, J. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Liu, C.
Right arrow Articles by Zhou, J. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Liu, C.
Right arrow Articles by Zhou, J. M.
Related Collections
Right arrow Soil Mineralogy
Right arrow Soil Chemistry
Soil Science Society of America Journal 66:109-116 (2002)
© 2002 Soil Science Society of America

DIVISION S-2 - SOIL CHEMISTRY

Residence Time Effect on Iron Perturbation of Taranakite Formation

C. Liua, P. M. Huang*,a and J. M. Zhoub

a Dep. of Soil Science, Univ. of Saskatchewan, 51 Campus Drive, Saskatoon, SK, S7N 5A8 Canada
b The Institute of Soil Science, Academia Sinica, P.O. Box 821, 71 East Beijing Road, Nanjing 210008, China

* Corresponding author (huangp{at}sask.usask.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Taranakite is an important reaction product in the immediate vicinity of phosphate fertilizer bands. The residence time effect on taranakite formation at pH 4.0 and on the iron perturbation of its formation at a Fe(II)/Al molar ratio of 1.2 (Al = 1 x 10-2 M and NH4H2PO4 = 1 M) was investigated in this study. The crystallization of NH4-taranakite involved a phase transformation from irregularly shaped x-ray noncrystalline materials to taranakite. The incorporation of NH+4 ions into taranakite structure was a relatively slow process. The presence of Fe(II) in the solution of the reaction system perturbed the incorporation of NH+4 ions into taranakite structure, especially in the early aging period. Furthermore, the coating or coprecipitation of x-ray noncrystalline iron phosphates with aluminum phosphates appeared to reduce the dissolution of aluminum phosphates, thus, affecting the phase transformation to taranakite. The presence of Fe(II) greatly decreased the rate of taranakite formation and retarded its formation. This is attributed to iron phosphate coatings as impurities on taranakite particles and the subsequent retardation of its crystal growth by interfering the integration of growth units. The impact of the residence time of Fe perturbation of the formation of taranakite in the immediate vicinity of fertilizer zones in soils, especially under reduced conditions, on the transformation and dynamics of P and N in the terrestrial system merits close attention.

Abbreviations: AFM, atomic force microscopy • IR, infrared • SEM, scanning electron microscope • XRD, x-ray diffraction


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE CONCENTRATION OF P in the soil solution of P fertilizer zones is very high. A nearly saturated solution of P fertilizer material forms in and around fertilizer granules or droplets (Kolaian and Ohlrogge, 1959; Sample et al., 1980; Tisdale et al., 1993). The concentration of P in saturated monoammonium phosphate is 2.9 M with pH 3.5 (Lindsay et al., 1962). When the concentrated P solution leaves the granule, droplet, or band site and moves into the surrounding soil, the soil components are altered by the solution. Some soil minerals may actually be dissolved by the concentrated P solution, resulting in the release of large quantities of metal ions such as Al, Fe, Mn, K, Ca, and Mg (Lindsay et al., 1962). Therefore, compared with bulk soil solution, much higher concentration of Al and Fe in solution temporarily exists in P fertilizer zones. Phosphate in the concentrated solutions reacts with these metal ions to form specific compounds, referred to as soil–fertilizer reaction products (Sample et al., 1980; Tisdale et al., 1993). Ammonium taranakite [(NH4)3Al5H6(PO4)8 · 18H2O] is one of the reaction products of phosphate fertilizers with soils (Frazier and Taylor, 1965). Lindsay et al. (1962) used saturated solutions of phosphate fertilizers to react with soils to simulate the reaction zone of fertilizer bands. The formation of taranakites was observed in the precipitates of the monoammonium phosphate and monopotassium phosphate systems. Other researchers (Sarkar et al., 1977; Prabhudesai and Kadrekar, 1984) observed taranakites in phosphate-soil systems. Ammonium taranakite formation was directly observed in 1 M NH4H2PO4-treated soils (Zhou and Huang, 1995).

An evaluation of taranakites as a source of phosphate for plants was conducted by Taylor et al. (1960). The formation of taranakites in soil can result in the transformation of N and P from the readily available form to the slowly available form and, thus, affects the dynamics and bioavailability of these nutrients in soil, especially in the immediate vicinity of phosphate fertilizer bands.

When phosphate fertilizer application dissolves soil minerals and results in the release of Al ions and the subsequent formation of taranakites, other cations and anions which coexist with Al in soil minerals may also be dissolved. Iron is one of the most abundant elements in soil environments. It can significantly perturb taranakite formation (Zhou et al., 2000). Ferrous iron is more effective in perturbing the formation of taranakite than Fe(III) after a 4-wk reaction time. Compared with the Fe(III) system, more Fe ions are present in the solution in the Fe(II) system to perturb the nucleation and crystallization of taranakite predominantly by complexation of Fe with phosphate (Zhou et al., 2000). However, the residence time effect on the taranakite formation and on the Fe perturbation of its formation and the mechanism involved are still obscure. The dynamics of taranakite formation should greatly influence the transformation, mobility, and bioavailability of P and N in soils and sediments. The objective of this study was to investigate the residence time effect on taranakite formation, both in the absence and in the presence of Fe(II) perturbation under acidic conditions similar to phosphate fertilizer zones.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Taranakite Formation Systems
Ammonium taranakite was synthesized by using NH4H2PO4 and AlCl3 both in the absence and in the presence of ferrous iron. The NH4H2PO4, AlCl3, and FeCl2 used in the study were pure chemicals (Analar grade, BDH Inc., Toronto, ON, Canada). The FeCl2 solution was freshly prepared immediately before the experiment. In order to simulate the condition of a phosphate fertilizer band in soil, a high concentration of NH4H2PO4 (1 M) was used in the synthesis. After mixing 50 mL of a 2 M NH4H2PO4 solution, 10 mL of a 0.1 M AlCl3 solution, and 12 mL of deionized distilled water or a 0.1 M FeCl2 solution in a 100 mL Erlenmeyer flask, the mixed solution was diluted to close to 100 mL by using deionized distilled water. The pH of the solution was adjusted to 4.0 by using ammonia water with continual stirring. A precipitate appeared during the pH adjustment. The suspension was then made to exactly 100 mL. The final concentrations of NH4H2PO4 and AlCl3 were 1 and 0.01 M. The FeCl2 concentration was, respectively, 0 and 0.012 M in the absence and in the presence of Fe(II) systems with the initial Fe(II)/Al molar ratios of 0 and 1.2. Based on the solubility of aluminum hydroxides, which are common in soils, such as amorphous Al(OH)3, the activity of Al3+ ions at pH 4.0 is 4.57 x 10-3 (activities are unitless, while concentrations are expressed in terms of mol L-3). This Al3+ activity corresponds to 0.78 M Al3+ concentration in the presence of 2.9 M NH4H2PO4. According to the diagram of the solubility relationships of iron with the pe + pH of typical soils (Schwab and Lindsay, 1983), the Fe2+ activity controlled by noncrystalline iron oxide at the pe + pH of 10, which was very common in soil, is {approx}2 x 10-2. This Fe2+ activity corresponds to 0.9 M Fe2+ concentration in the presence of 2.9 M NH4H2PO4. The activity coefficients used to convert the activities to the concentrations were computed based on the Pitzer equation (Pitzer, 1991). Furthermore, Lindsay and Stephenson (1959) reported that Al and Fe in the solution of 4.5 M monocalcium phosphate monohydrate reacted with a fine sandy loam soil can reach 0.7 and 0.2 M. Thus, the 0.01 M Al3+ and 0.012 M Fe2+ concentrations used in the present study were in the realistic range of Al3+ and Fe2+ concentrations in the soil solution of P fertilizer zones.

The container that stored the suspension prepared as described above was sealed by Parafilm and stored in laboratory at 296 K. The suspension was aged for 1 d, 1, 2, 3, and 4 wk, and then filtered through a Millipore membrane (Millipore Corp., Bedford, MA) with a pore size of 0.01 mm at the end of each aging time. Two milliliters of concentrated HCl were immediately added to the filtrate to prevent Fe(II) from oxidation (Tamura et al., 1976). The solid phase on the membrane was washed with deionized distilled water until no Cl- in the filtrate was detected with a 1% AgNO3 solution. The sample was then air dried prior to analysis of the reaction products.

Identification of the Solid Phases
Fifteen milligrams of each washed and air-dried sample were lightly ground and mounted on a glass slide by adding a few drops of acetone without any further treatment prior to the x-ray diffraction (XRD) analysis. The analysis was carried out using a Rigaku diffractometer (Rigaku Company, Tokyo, Japan) with Fe-K{alpha} radiation filtered by a graphite monochromator at 40 kV and 130 mA. For infrared (IR) absorption analysis, 2 mg of each sample were mixed with 250 mg KBr and then pressed into discs. Using a KBr disc as a reference, the discs with samples were examined with a Perkin-Elmer infrared absorption spectrophotometer (Model 983, Buckinghamshire, England).

Atomic Force Microscopy and Scanning Electron Microscopy of the Solid Phases
Five milligrams of the solid product were dispersed in 20 mL of deionized distilled water by ultrasonification (Sonifier, Model 350, Danbury, CT) at 150 W for 2 min in an ice bath. One drop of the suspension was deposited on a watch glass and air-dried overnight at room temperature (296.5 ± 0.5 K). The watch glass was then fastened to a magnetized stainless steel disk (diameter of 12 mm) with double-sided tape. The three-dimensional atomic force microscopy (AFM) images were obtained by using a NanoScopeTMIII atomic force microscope (Digital Instruments, Inc., Santa Barbara, CA) under ambient condition. A silicon nitride cantilever with a spring constant of 0.12 N m-1 was used in the contact mode. The scanning rate was 22 Hz. The deflection image was used to represent the surface features of the solid products.

The scanning electron micrographs of the precipitation products were obtained by using a scanning electron microscope (SEM) (Model Joel JSM-840A, Kevex Quantum, Akishima, Japan) at 20 kV.

Determination of pH, EH, and Elemental Concentrations
The pH of suspensions was measured by using a glass-calomel combination electrode. The EH was determined by using a Pt electrode on a Metrohm titroprocessor (683 model, Metrohm Ltd., Herisau, Switzerland). The concentrations of P, NH+4–N, Al, Fe(III), and Fe(II) in the acidified filtrate solutions of each reaction time were determined. The Al and total Fe concentration in the solutions were, respectively, measured using atomic absorption spectrometry at 309.3 and 248.3 nm on a Perkin Elmer atomic absorption spectrophotometer (Model 3100, Perkin Elmer Corp., Norwalk, CT). The ferrous Fe concentration was measured by the colorimetric method with 2,4,6-tri(2'-pyridyl)-1,3,5-triazine (Krishnamurti and Huang, 1990). The ferric Fe concentration was calculated based on the difference between the total Fe concentration and ferrous Fe concentration. The P concentration was determined by the vanadomolybdophosphoric acid colorimetric method (Kuo, 1996). The NH+4–N concentration was measured by the emerald green colorimetric method (Mulvaney, 1996). The contents of P, NH+4–N, Al, Fe(III), and Fe(II) in the solid phases were determined by dissolving 0.02 g of products in 2 mL of concentrated HCl, which was then diluted to 100 mL. The concentrations of P, NH+4–N, Al, Fe(III), and Fe(II) were determined by the same methods as described above. Each experiment in the present study was carried out in triplicate.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
X-ray Diffraction and Infrared Spectroscopy
The x-ray diffractograms of the precipitates in the taranakite formation system in the absence of Fe(II) are presented in Fig. 1 . X-ray noncrystalline precipitates formed in this reaction system after aging for 1 d (Fig. 1a). Crystalline taranakite was observed after aging for 1 wk, as indicated by the d-values of the XRD pattern at 15.9, 7.94, 3.16, and 2.64 Å (Fig. 1b) which are characteristic d-values of taranakite. As the residence time increased from 1 wk to 2, 3, and 4 wk, more characteristic d-values (7.46, 5.90, 4.36, 3.84, 3.59, and 3.35 Å) of taranakite were observed, and the intensity of the characteristic peaks gradually increased with the increase of residence time (Fig. 1c, d, and e). Taranakite crystallization was evidently improved as residence time increased. In contrast, x-ray noncrystalline precipitates were observed in the presence of Fe(II) after aging for 2 wk (Fig. 2a, b, and c) . After aging for 3 wk, the taranakite formed was still poorly crystalline, as indicated by the low-intensity d-values at 15.9 and 7.94 Å (Fig. 2d). Only these two peaks were observed, even after aging for 4 wk, although the intensity slightly increased compared with the precipitate formed after a 3-wk reaction time (Fig. 2e). The data revealed that the presence of Fe(II) significantly decreased the rate of taranakite crystallization.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 1. X-ray diffractograms of the precipitates formed at 1 M NH4H2PO4, 10-2 M Al, pH 4.0, and Fe(II)/Al molar ratio = 0 after aging for (a) 1 d, (b) 1 wk, (c) 2 wk, (d) 3 wk, and (e) 4 wk.

 


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 2. X-ray diffractograms of the precipitates formed at 1 M NH4-H2PO4, 10-2 M Al, pH 4.0, and Fe(II)/Al molar ratio = 1.2 after aging for (a) 1 d, (b) 1 wk, (c) 2 wk, (d) 3 wk, and (e) 4 wk.

 
The IR spectrum of the pure NH4-taranakite was shown by Frazier and Taylor (1965). The phosphate absorption bands at 887, 951, 1018, 1048, 1089, and 1193 cm-1, ammonium absorption bands at 1430, 1461, and 3230 cm-1, the band representing the H-O-P bond at 2422 cm-1, and the bands related to the Al-O-P bond at 282, 328, 448, 534, and 606 cm-1 were clearly observed in the IR spectrum of their pure NH4-taranakite (Frazier and Taylor, 1965). The IR spectra of the precipitation product in the taranakite formation system in the absence of Fe(II) in this study are presented in Fig. 3 . The characteristic absorption bands of NH4-taranakite were not observed in the IR spectrum of the precipitate in the taranakite formation system after aging for 1 d (Fig. 3a). The absorption bands from 282 to 606 cm-1 appeared as a single band at 527 cm-1. The phosphate absorption bands from 887 to 1193 cm-1 were present as a single band at 1104 cm-1. The ammonium absorption bands at 1430 and 1461 cm-1 appeared as a single band at 1400 cm-1. The appearance of the broad absorption bands indicated the formation of noncrystalline products. As the residence time gradually increased from 1 d to 1, 2, 3, and 4 wk, the characteristic absorption bands of NH4-taranakite were more clearly observed in the precipitation products (Fig. 3b, c, d, and e), indicating the transformation of noncrystalline products to taranakite. This is consistent with the XRD data (Fig. 1). The absorption bands of 1639 to 1649, 3112 to 3220, and 3371 to 3384 cm-1 (Fig. 3) were from H2O. The absorption bands at 3231 cm-1 (Fig. 3d) and 3233 cm-1 (Fig. 3e) were from ammonium ions. The decrease in intensity of the bands at 1639 to 1649 cm-1, the disappearance of absorption bands at 3371 to 3384 cm-1, and the appearance of 3231 to 3233 cm-1 (Fig. 3) are in accord with the transformation of noncrystalline products to crystalline taranakite (Fig. 1).



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 3. Infrared spectra of the precipitates formed at 1 M NH4H2PO4, 10-2 M Al, pH 4.0, and Fe(II)/Al molar ratio = 0 after aging for (a) 1 d, (b) 1 wk, (c) 2 wk, (d) 3 wk, and (e) 4 wk.

 
In the taranakite formation system in the presence of Fe(II), no distinct characteristic absorption bands of taranakite were oberved, except for the precipitation product formed after aging for 4 wk (Fig. 4) . The precipitation products formed after aging from 1 d to 3 wk had IR spectra similar to that of the x-ray noncrystalline product formed in the absence of Fe(II) after aging for 1 d, although some absorption bands shifted slightly (Fig. 4a, b, c, and d). The product formed in the absence of Fe(II) after aging for 1 d had the absorption bands at 527 and 1104 cm-1 (Fig. 3a), whereas the product formed in the presence of Fe(II) after aging from 1 d to 3 wk had the absorption bands at 539 and 1061 to 1063 cm-1 (Fig. 4a, b, c, and d). These shifts were probably due to the vibrations of Fe-O-P and Fe-O. The absorption bands at {approx}1040 cm-1 from Fe-O-P vibration was observed after goethite and x-ray noncrystalline Fe oxides adsorbed phosphate (Parfitt et al., 1976; Shang, 1989; Liu, 1999). After aging for 4 wk, the absorption bands of the IR spectrum (Fig. 4e) show the formation of taranakite in the presence of Fe(II), although two taranakite bands at 1018 and 1048 cm-1 merged to one single band at 1029 cm-1. In the presence of Fe(II) in the taranakite formation system, the ammonium absorption bands at 1430, 1461, and 3230 cm-1 were only observed in the product after aging for 4 wk (Fig. 4e).



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4. Infrared spectra of the precipitates formed at 1 M NH4H2PO4, 10-2 M Al, pH 4.0, and Fe(II)/Al molar ratio = 1.2 after aging for (a) 1 d, (b) 1 wk, (c) 2 wk, (d) 3 wk, and (e) 4 wk.

 
Atomic Force Microscopy and Scanning Electron Microscopy
The AFM images of the precipitation products formed in the absence of Fe(II) are shown in Fig. 5 . Irregularly shaped particles formed after aging for 1 d (Fig. 5a) were x-ray noncrystalline products (Fig. 1a). The plate-shaped and irregularly shaped particles as discrete phases or coatings were observed in the products after aging from 1 to 4 wk (Fig. 5b, c, d, and e). Furthermore, the particle size and thickness of plates greatly increased with an increase in residence time. This data, along with the XRD and IR data (Fig. 1 and 3), indicates that the plate-shaped particles were crystalline taranakite. In the presence of Fe(II) in the reaction system, irregularly shaped particles were formed after aging from 1 d to 2 wk (Fig. 6a, b, and c) . The x-ray patterns (Fig. 2a, b, and c) of the products indicate that these irregularly shaped materials were x-ray noncrystalline. Most of the precipitates formed after aging for 3 wk were still irregular in shape, although some plate-shaped particles with size <1 µm were imaged (Fig. 6d). The plates formed after aging for 4 wk had a larger particle size (Fig. 6e). In comparison with the plate particles formed in the absence of Fe(II) in the reaction system, the plate particles formed in the presence of Fe(II) were extensively coated with irregularly shaped materials and the coating predominantly occurred on the corners and edges (Fig. 6e). This is further shown by the SEM micrographs of the precipitation products formed both in the absence and presence of Fe(II) after aging for 4 wk (Fig. 7) . The precipitation products formed in the absence of Fe(II) appeared as globular particles which were composed of plates (Fig. 7a). The size of globular taranakite was {approx}200 µm. The size of the particles formed in the presence of Fe(II) after aging for 4 wk was {approx}80 µm. The surface of the particles formed in the presence of Fe(II) was coated with irregularly shaped materials (Fig. 7b). Since the maximum scanning size of AFM used in this study was 15 µm, the entire globular morphology of taranakite cannot be imaged by AFM. However, the AFM data revealed more detailed surface features of the reaction products. As residence time increased, the size of taranakite plates increased and the plates arranged to form globular clusters.



View larger version (106K):
[in this window]
[in a new window]
  		Fig. 5. Atomic force micrographs of the precipitates formed at 1 M NH4H2PO4, 10-2 M Al, pH 4.0, and Fe(II)/Al molar ratio = 0 after aging for (a) 1 d, (b) 1 wk, (c) 2 wk, (d) 3 wk, and (e) 4 wk.

 


View larger version (115K):
[in this window]
[in a new window]
  		Fig. 6. Atomic force micrographs of the precipitates formed at 1 M NH4H2PO4, 10-2 M Al, pH 4.0, and Fe(II)/Al molar ratio = 1.2 after aging for (a) 1 d, (b) 1 wk, (c) 2 wk, (d) 3 wk, and (e) 4 wk.

 


View larger version (118K):
[in this window]
[in a new window]
  		Fig. 7. Scanning electron micrographs of the precipitates formed at 1 M NH4H2PO4, 10-2 M Al, and pH 4.0 (a) in the absence of Fe(II) and (b) in the presence of Fe(II) at a Fe(II)/Al molar ratio = 1.2 after aging for 4 wk.

 
Although the precipitate formed in the presence of Fe(II) after aging for 4 wk had much lower intensity of taranakite peaks in the XRD pattern compared with the precipitate formed in the absence of Fe(II) after aging for 1 wk (Fig. 1b and 2e), the AFM images show that the size of plates formed in the former system was larger than that of the plates formed in the latter reaction system (Fig. 5b and 6e). The coating of the irregularly shaped materials must have caused the lower intensity of the XRD peaks in the system in the presence of Fe(II). The coating apparently perturbed the crystal growth of taranakite during crystallization. The coating on taranakite was not substantial in the absence of Fe(II) in the reaction system. The irregularly shaped materials coated on taranakite appeared to be iron precipitates probably as iron phosphates or mixed Al and Fe phosphates and oxides.

Solution Analysis
Since a high concentration of NH4H2PO4 (1 M) and a relatively low concentration of AlCl3 (10-2 M) were used in the formation of taranakite, the concentrations of P and N remaining in the solution were virtually the same as residence time increased (Table 1). However, the concentration of Al remaining in the solution was decreased by one to three orders of magnitude compared with the initial Al concentration of 10-2 M (Table 1). Aluminum concentration did not consistently decrease with the increase in residence time both in the absence and in the presence of Fe(II). In the absence of Fe(II), Al concentration at first increased as the residence time increased from 1 d to 2 wk, and then decreased when the residence time increased to 3 and 4 wk (Table 1). This observation is in accord with the XRD (Fig. 1) and IR (Fig. 3) data and AFM images (Fig. 5), indicating a phase transformation from initial precipitates to NH4-taranakite through a solid dissolution pathway. The initial precipitates were probably aluminum phosphates or hydroxy aluminum polymers. This phase transformation process was apparently hindered by the presence of Fe(II). In the presence of Fe(II), Al concentration only increased with the increase of residence time from 1 d to 1 wk, and then steadily decreased when the residence time increased to 2, 3, and 4 wk (Table 1). Additionally, the change in Al concentration in the taranakite formation system in the presence of Fe(II) was not as great as that in the taranakite formation system in the absence of Fe(II). The change of Al concentration was two orders of magnitude in the former system and three orders of magnitude in the latter system (Table 1). In the presence of Fe(II), more Al ions remained in the solution at the end of a 4-wk aging period compared with the system in the absence of Fe(II) (Table 1). The presence of Fe(II) greatly retarded taranakite formation by reducing the dissolution of the initial precipitate and the subsequent transformation to taranakite. This is in accord with the XRD and IR data and AFM images (Fig. 1 to 6). Compared with the Al concentration at the end of a 3-wk aging period, the Al concentration in the solution at the end of a 4-wk aging period decreased by 11.9 times in the absence of Fe(II) and by 1.8 times in the presence of Fe(II) (Table 1). This indicates that the experimental system had not completely reached equilibrium.


View this table:
[in this window]
[in a new window]
  		Table 1. The concentration of Al, P, NH+4–N, Fe(III) and Fe(II) remaining in solutions in the taranakite formation systems at the initial concentration of 10-2 M Al and 1 M NH4H2PO4, and Fe(II)/Al molar ratios of 0 and 1.2 after different aging periods.

 
In the taranakite formation system in the presence of Fe(II), both Fe(III) and Fe(II) ions were detected after aging for 1 d, and their concentrations steadily decreased as the residence time increased (Table 1). No Fe(II) ions were detected in the solution at the end of 3-wk and 4-wk aging periods, indicating that all the Fe(II) ions were eventually oxidized to Fe(III) ions using O2 as an electron acceptor. The EH values (Table 2) further confirmed the oxidation of Fe(II) to Fe(III) in the presence of Fe(II) in the reaction system. The EH value of the reaction system in the absence of Fe(II) did not significantly change after aging for different times (Table 2). However, the EH value of the reaction system in the presence of Fe(II) substantially increased from 154 to 350 mV, with the increase of aging time from 1 d to 4 wk (Table 2). After aging for 4 wk, the EH and Pe + pH values of the reaction systems both in the absence and presence of Fe(II) were virtually the same, although their initial EH and pe + pH values were very different.


View this table:
[in this window]
[in a new window]
  		Table 2. Final EH and pH values in the taranakite formation systems at the initial concentration of 10-2 M Al and 1 M NH4H2PO4, and Fe(II)/Al molar ratios of 0 and 1.2 after different aging periods.

 
Elemental Contents of the Solid Products
Elemental composition of the solid products is given in Table 3. Based on the formula [(NH4)3Al5H6(PO4)8 · 18H2O], NH4-taranakite contains 105.4 g Al kg-1, 193.7 g P kg-1, and 32.8 g N kg-1. Compared with the formula of NH4-taranakite, the precipitate formed in the absence of Fe(II) ions after aging for 1 d had much higher Al content and less P and N contents (Table 3). As residence time increased, the Al content of the precipitate decreased and the P and N contents increased. Compared with the formula of NH4-taranakite, the precipitate after aging for 4 wk had the same P content and {approx}13% lower Al and 4% lower N contents (Table 3). The lower Al and N contents of the taranakite formed probably resulted from crystal defects or the presence of some noncrystalline precipitates. At the end of 3-wk and 4-wk aging time, gel-like precipitates were present as a discrete phase or coatings on triangular flat-shaped particles (Fig. 5d and e). The variation in Al and P contents of the precipitates with aging time indicates that different solid phases were formed. The Al content of the precipitate in the absence of Fe(II) decreased dramatically when the residence time increased from 1 wk to 2 wk. The Al and P contents of the precipitate was virtually the same after aging for 3 and 4 wk. However, the N content of the precipitate still significantly increased with the increase of residence time even from 3 wk to 4 wk (Table 3), indicating the slow incorporation of NH+4 ions into the taranakite structure during crystallization processes.


View this table:
[in this window]
[in a new window]
  		Table 3. Elemental composition of the solid products formed at the initial concentration of 10-2 M Al and 1 M NH4H2PO4, and Fe(II)/Al molar ratios of 0 and 1.2 after different aging periods.

 
In the presence of Fe(II) in the reaction system, the Al and N contents of the product formed drastically decreased compared with those of the product formed in the absence of Fe(II); this was especially true for the N content after aging for 1 d (Table 3). The N content substantially increased with the increase of residence time. This indicates that the presence of Fe(II) greatly retarded the incorporation of NH+4 ions into the taranakite structure in the initial crystallization process. The presence of a large amount of iron phosphate complexes apparently retarded the nucleation process of taranakite (Zhou et al., 2000). The P content of the precipitate formed in the presence of Fe(II) did not significantly change with the increase of residence time, and was the same as the P content of the precipitate formed in the absence of Fe(II) after aging for 1 d. Except for aging time of 1 d, there was no detectable Fe(II) in the solid phase products formed in the presence of Fe(II) in the reaction system, but substantial amounts of Fe(III) were present in the solid phase products (Table 3). Fe(II) was evidently oxidized to Fe(III) in the precipitates during the aging period. The coprecipitation of Fe apparently resulted in the decrease of Al content of the reaction products.

The P:Al:N ratio of NH4-taranakite formula is 5.9: 3.2:1. The P:Al:N ratio of the taranakite formed in the absence of Fe(II) ions after aging for 1 d was much higher than that of NH4-taranakite formula (Table 3). As the residence time increased, the P:Al:N ratio of the precipitate gradually became close to that of NH4-taranakite formula. The similar P:Al:N ratio should have been observed if taranakite were the reaction product in the solid phase formed in the presence of Fe(II). In the presence of Fe(II) in the reaction system, the product after aging for 1 d had the P:Al:N ratio of 199:102:1 due to extremely low N content. The AFM, XRD, and IR data show the formation of x-ray noncrystalline product after this aging time. The P:Al:N ratios of the reaction products formed in the presence of Fe(II) after aging for 3 and 4 wk were 7.2:2.3:1 and 7.1:2.2:1, respectively, which were much closer to the ratio based on the taranakite formula. The XRD patterns also show that crystalline taranakite was formed after aging for 3 and 4 wk (Fig. 2d and e). However, the product after aging for 3 and 4 wk still had a much higher P:N ratio than the taranakite formed in the absence of Fe (Table 3). The excess P presumably resulted form coprecipitation with Fe(III), since a large amount of Fe was detected in the product. The XRD patterns of the products only show the d-values of taranakite (Fig. 2d and e). Therefore, iron phosphates formed were x-ray noncrystalline. The data of chemical modeling in the previous study also show that iron phosphate complexes rather than the hydrolysis species of Fe are the predominant Fe species formed in the presence of Fe(II) in the reaction system (Zhou et al., 2000).

In the presence of Fe(II) in the reaction system, the precipitation product formed after aging for 1 d to 2 wk had Al:N ratios from 102:1 to 4.1:1, which were higher than the Al:N ratio of the NH4-taranakite formula, whereas the precipitates formed after aging for 3 and 4 wk had Al:N ratios of 2.3:1 and 2.2:1, which were lower than the Al:N ratio based on the NH4-taranakite formula (Table 3). This indicates that although the amount of NH+4 ions precipitated after aging for 3 and 4 wk was enough to form taranakite, the incorporation of NH+4 ions was retarded especially at relatively short residence times (1 d to 2 wk). Moreover, the crystallinity of taranakite formed in the presence of Fe(II) in the reaction system after aging for 3 and 4 wk (Fig. 2d and e) was still much poorer compared with that of taranakite formed in the absence of Fe(II) (Fig. 1d and e). This is probably due to the coating of iron phosphates as impurities on taranakite particles. Crystal growth by the integration of growth units (Khamskii, 1969; Mullin, 1993) was apparently retarded by the coating. Furthermore, the concentration of Al remaining in solutions (Table 1) shows that a phase transformation through dissolution-reprecipitation reaction was retarded in the presence of Fe(II) in the reaction system. This may have resulted from the coating or coprecipitation of iron phosphates on aluminum phosphates, thus, hampering the phase transformation to form NH4-taranakite.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The results of the present study show that the crystallization of NH4–taranakite involved a phase transformation from irregularly shaped x-ray noncrystalline materials to taranakite. The incorporation of NH+4 ions into taranakite structure was a relatively slow process. The presence of Fe(II) in solution affected the dynamics of taranakite formation by retarding the incorporation of NH+4 ions into taranakite structure especially in the early aging period. The coating or coprecipitation of x-ray noncrystalline iron phosphates with aluminum phosphates appeared to reduce the dissolution of aluminum phosphates, and thus retarded the phase transformation to taranakite. The coating as impurities on taranakite particles hampered the crystal growth by disrupting the integration of growth units. Taranakite formed under the influence of Fe(II) after various aging times had different degrees of crystallinity, chemical composition, and surface features, which should have an important bearing on the transformation of P and N in the immediate vicinity of fertilizer zones in soils. The residence time effect on ferrous Fe perturbation of the formation of crystalline taranakite in the immediate vicinity of fertilizer zones in soils, especially under reduced conditions, and the impact on the dynamics and bioavailability of P and N in the terrestrial system merit close attention.


   ACKNOWLEDGMENTS
 
This study was supported by Research Grant GP 2383-Huang and Equipment Grant EQP156628 of the Natural Sciences and Engineering Research Council of Canada and a grant from the Potash and Phosphate Institute of Canada.

Received for publication February 27, 2001.


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





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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Liu, C.
Right arrow Articles by Zhou, J. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Liu, C.
Right arrow Articles by Zhou, J. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Liu, C.
Right arrow Articles by Zhou, J. M.
Related Collections
Right arrow Soil Mineralogy
Right arrow Soil Chemistry

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