Published online 25 January 2008
Published in Soil Sci Soc Am J 72:343-346 (2008)
DOI: 10.2136/sssaj2006.0388
© 2008 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
SOIL CHEMISTRY
The Solubility Constant of Variscite
Mauro Iuliano*,
Liberato Ciavatta and
Gaetano De Tommaso
Dipartimento di Chimica, Università di Napoli "Federico II", via Cinthia 45, 80126 Napoli, Italy
* Corresponding author (miuliano{at}unina.it).
 |
ABSTRACT
|
|---|
Only a few studies have been made on the ion product of variscite, Al(III) phosphate dihydrate, AlPO4·2H2O, a parameter of fundamental importance in clarifying the fixation of soluble phosphates in acid soils. The solubility of variscite in H3PO4 solutions of molality ranging from 1 x 10–4 to 0.1 mol kg–1 was determined at 25°C. The data were explained with a value of the ion product equal to 10–3.14 mol–1 kg (at the infinite dilution reference state) and assuming the formation of complexes AlPO4(aq), AlHPO4+, and Al(H2PO4)2+. The solubility product obtained in this work differs by orders of magnitude from the literature values. The discrepancies are mainly due to the use of solids of different composition as well as to the assumption of the Al3+ ion as the unique soluble Al species. On the basis of equilibrium constants, the fixation in acid soils of soluble phosphates implicates the formation of variscite, while only at higher concentrations does the fixation occur as strengite.
|
INTRODUCTION
|
|---|
Only a few studies have been made on the ion product of variscite, Al(III) phosphate dihydrate, AlPO4·2H2O, a parameter of fundamental importance in clarifying the fixation of soluble phosphates in acid soils.
Cole and Jackson (1950), from solubility measurements in the pH range 3.6 to 4.3, calculated for the equilibrium equation
 | [1] |
a constant of 10–28.55 valid in 0.05 mol L–1 NaCl. Temperature was not stated.
Kittrick and Jackson (1955) observed an effect of the solid/solution ratio on solubility. Therefore, a rather uncertain ion product resulted. These researchers estimated for Eq. [1] constants ranging from 10–27.7 to 10–28.4.
Lindsay et al. (1959) determined the solubility at 25°C in dilute H3PO4 and HCl solutions and found that the heterogeneous equilibrium is reached slowly. After 144 d of solid–solution contact, the Al concentration was still increasing. A constant of 10–30.5 was proposed for Eq. [1]. The discrepancies of the constants are explainable by the presence in the various preparations of more soluble amorphous phases and by the neglect of soluble phosphate complexes of the Al3+ ion. On account of complex species evidenced in a potentiometric investigation by Ciavatta and Iuliano (1996), the constant for Eq. [1] would be smaller with respect to the proposed value by orders of magnitude.
The purpose of this investigation was that of determining at 25°C the ion product of variscite from solubility measurements in H3PO4 solutions ranging from 10–4 to 0.1 mol kg–1. Complexation equilibria between Al ions and H3PO4 were taken into due consideration.
|
MATERIALS AND METHODS
|
|---|
The Solubility Measurements
The variscite used in this study was prepared by dissolving Al metal in a boiling dilute solution of H3PO4 and refluxing. A typical prescription is as follows. About 0.1 mol of pure Al wire, 99.999% purity (5N, Sigma-Aldrich, St. Louis, MO), was added to 1.5 L of 0.15 mol L–1 H3PO4. Kept boiling under reflux, the metal dissolved slowly under evolution of H2 and finally a solid phase separated on the walls of the vessel. After complete dissolution of the metal, the system was left boiling for a week to allow the growth of crystals. The solid was carefully washed with distilled water until the mother liquid had a pH similar to that of the washing medium, and air dried at room temperature for a few days.
The x-ray powder diffraction pattern, shown in Fig. 1
, corresponds exactly to that reported for variscite-10 with orthorhombic cell dimensions of a = 982.16, b = 855.83, and c = 962.22 pm.
The phase variscite was identified by the Powder Diffraction File database (www.icdd.com/). The x-ray powder diffraction data of variscite-10 and of the solid obtained in this work are reported in Table 1
.
Solids kept in contact with H3PO4 solutions for periods necessary to reach heterogeneous equilibrium gave x-ray patterns identical to the original phase.
Under a scanning electronic microscope (at a 28-kV accelerating voltage), the solid consisted of crystals with sizes ranging from 100 to 200 µm (Fig. 2
), thus large enough to render surface effects negligible.
Equilibrations of the solid with dilute H3PO4 solutions were done in a water bath kept at 25.00 ± 0.05°C, with the leaching apparatus illustrated in the investigation on the solubility of strengite by Iuliano et al. (2007). The solid was wrapped in a filter paper bag, immersed in a glass cylinder with the test solution, and stirred with a magnetic bar. This apparatus prevented contact of solid particles with the magnetic stirrer. In fact, preliminary measurements showed an increase in solubility during periods of weeks when the solid was in mechanical contact with the magnetic bar, because the solid had been transformed into smaller particles. The bag avoided grinding by the stirrer.
Suitable aliquots of solution were periodically taken for the determination of Al. No temporal drift, within the accuracy of the analytical method, in the Al concentration was recorded after 7 to 10 d so that equilibrium was considered to be attained.
Aluminum was analyzed spectrophotometrically as oxinate after extraction in CCl4 following the procedure described by Kolthoff and Elving (1966, p. 409). Samples of 0.5 µmol of Al could be determined with an error of about 1%.
The results of a series of measurements are given in Table 2
. Each composition represents an average of three or more runs.
View this table:
[in this window]
[in a new window]
|
Table 2. Composition of the saturated solutions. The density of the saturated solution, d, was found pycnometrically.
|
|
Stock H3PO4 solutions were prepared by diluting 85% H3PO4 (Mallinckrodt Baker, Phillipburg, NJ). The concentration was determined by gravimetric analysis as MgNH4PO4·6H2O, according to Winkler (1931, p. 145). The analyses agreed to within 0.1%.
|
CALCULATIONS AND RESULTS
|
|---|
The solubility data, which form the basis of the following calculations, are summarized in Table 2. The heterogeneous equilibrium is represented by the reaction
 | [2] |
where HA stands for H3PO4, 
symbolizes the activity coefficient, and *Ks is the solubility constant in terms of the predominating species in solution, which are H+, Al3+, and H3PO4. The total Al solubility increase with increasing mP (Table 2) can be explained with the formation of soluble complexes, which on account of the low metal molality, are probably mononuclear in Al. Indicating with βp,q the equilibrium constant of the generic complex AlH–p(H3PO4)q3–p formed according to
 | [3] |
the mass balance condition can be expressed as
 | [4] |
in which mAl is the total Al concentration and p,q is the activity coefficient of AlH–p(H3PO4)q3–p. At decreasing total phosphate molality, the solubility tends to the limiting value of 4.588 x 10–5 (Table 2). Since phosphate was introduced by H3PO4, for mP 
0 also mHA 0 and mH+0. Consequently, mH+3/mHA0. In dilute solution (mP 0), various 1. This implies that the first term on the right-hand side of Eq. [4] is negligible and the soluble complexes have a molality independent of both mHA and mH+. This condition is verified if q = 1 and p = 3, thus the complex present in detectable amounts at low phosphate molalities is AlPO4(aq) and mAl = *Ksβ3,1 = (4.588 ± 0.05) x10–5 for data at mP 
0.01. The increase of Al solubility at mP > 0.01 is ascribable to complexes with q 
1 and p 3.
When we consider that at mP > 0.01 the mAl/mP ratio never exceeds 0.003, then without introducing any appreciable error we may simplify the calculations of mH+ by using the approximation
 | [5] |
in which A– is H2PO4– and Ka1, the first protolysis constant of H3PO4, was made equal to 10–2.148, the value selected by Martell and Smith (1989, p. 444). Activity coefficients were evaluated by the so-called Güntelberg formula logi = – zi20.510
I/(1 + I), where the ionic strength I = mH+. A preliminary mH+ was deduced setting = 1. More accurate activity coefficients were then estimated and the cycle continued to convergence.
To explain the data in the simplest way, we postulated the predominance of Al(H2PO4)2+, which, according to the results of Ciavatta and Iuliano (1996), is the most probable species under the present experimental conditions. This is corroborated by the constancy of the quantity (mAl – *Ksβ3,13,1–1)(mHA mH+)–1 H+HA2,2–1 = (mAl – 4.588 x 10–5)(mHA mH+)–1 = *Ksβ2,2 = (9.6 ± 0.3) x 10–2, evaluated for each point at mP > 0.055, approximating HA = 3,1 = 1, H+ = 2,2. Deviations from constant value with increasing mP exceeding the experimental error, however, occurred in the intermediate phosphate range, indicating the presence of other soluble species.
The minor complex best explaining the deviations was found to be AlHPO4+, while no acceptable fit resulted with other conceivable species such as AlH2PO42+, AlHPO4H2PO4(aq), or Al(HPO4)2–. The agreement of the model including AlHPO4+ and Al(H2PO4)2+ is illustrated by the plot Y(mHA), where
 | [6] |
As Fig. 3
shows, Y is linearly dependent on mHA. From the intercept and slope of the best line through the points, the most probable values of *Ksβ2,1 and *Ksβ2,2, given in the Appendix, were calculated.
View larger version (7K):
[in this window]
[in a new window]
|
Fig. 3. Y = (mAl – 4.588 · 10–5 )/ mH+, where mAl and mH+ are the total Al concentration and hydrogenionic concentration, respectively. The line was calculated with constants given in the Appendix.
|
|
The solubility constant *Ks was evaluated from *Ksβ2,2 using logβ2,2 = 2.1 ± 0.3 reported in Ciavatta (2001). The values of β3,1 and β2,1 were then calculated from *Ksβ3,1 and *Ksβ2,1, respectively. The results of evaluations are given in the Appendix.
|
DISCUSSION
|
|---|
The solubility of variscite in H3PO4 solutions not exceeding 0.1 mol kg–1 is mainly ascribable to the formation of Al(H2PO4)2+ and the new complexes AlPO4(aq) and AlHPO4+. From the values of β3,1, β2,1, and *Ks in the Appendix and the protolysis constants of H3PO4 selected by Martell and Smith (1989), we calculated
The high stability of AlPO4(aq) and AlHPO4+ is in part due to the favorable entropy change of the formation equilibria. A contribution to stability might be envisaged also in some particular structure of the complexes in which the phosphate group acts as bi- or tridentate, or hydroxyl groups are present. As a matter of fact, the results from equilibrium analysis methods cannot tell anything about the structure of the complex, so that AlPO4(aq) is equivalent to Al(OH)(HPO4)(aq) or Al(OH)2(H2PO4)(aq).
The solubility product obtained in this work differs by orders of magnitude from the literature values. The discrepancies are mainly due to the use of solids of different composition as well as to the assumption of the Al3+ ion as the unique soluble Al species.
The low solubility of variscite, as shown by the results presented here, can explain the fixation in acid soils of soluble phosphates, added as fertilizers, by Al hydroxides or silicates. Assuming for gibbsite, 
-Al(OH)3, the solubility constant K[-Al(OH)3 + 3H+ 
Al3++ 3H2O] = 108.5, selected by Baes and Messmer (1976), we calculated, using *Ks in the Appendix,
Thus, variscite can coexist with gibbsite in contact with solutions where mHA = 10–11.6. From Eq. [3], mP 
mHA + mA– = mHA + mHA Ka1/ mH+, mP can be calculated at a given mH+ = 10–pH, assuming mHA = 10–11.6. In the pH range 4 to 6, corresponding values of mP are included in the interval 10–10 to 10–8 mol kg–1. In more concentrated phosphate solutions, however, gibbsite is transformed into variscite (Wenzl et al., 2003).
Similar evaluations apply to the corresponding Fe(III) compounds strengite, FePO4·2H2O, and goetite, -FeOOH. Considering the constants K(FePO4·2H2O(cr) + 3H+ Fe3+ + H3PO4 + 2H2O) = 10–6.70 recently determined in this laboratory (Iuliano et al., 2007), and K(-FeOOH + 3H+ Fe3+ + 3H2O) = 100.5 proposed by Baes and Mesmer (1976), we can calculate the constant of the reaction
At pH = 5, Fe(III) hydroxides and phosphates may coexist at mHA = 10–7.2 and mP = 10–4.3 mol kg–1. Hence the transformation of goetite into strengite implies a phosphate solution with mP > 10–4.3 mol kg–1.
In concluding, we note that in solutions with 10–10 mP 10–4.3, the P is fixed as variscite, while only at higher concentrations does the fixation occur as strengite.
|
APPENDIX
|
|---|
The following is a survey of calculated equilibrium constants:
|
NOTES
|
|---|
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.
Received for publication November 10, 2006.
|
REFERENCES
|
|---|
- Baes, C.F., and R.E. Messmer. 1976. The hydrolysis of cations. John Wiley & Sons, New York.
- Ciavatta, L., and M. Iuliano. 1996. A potentiometric study of aluminum(III) phosphate complexes. Ann. Chim. 86:1–17.[Web of Science]
- Ciavatta, L. 2001. Equilibrium studies on the complexation of cations with orthophosphate ions. Recent Res. Devel. Inorg. Organometallic Chem. 1:83–98.
- Cole, C.V., and M.L. Jackson. 1950. Solubility equilibrium constant of dihydroxy aluminum dihydrogen phosphate relating to a mechanism of phosphate fixation in soils. Soil Sci. Soc. Am. Proc. 15:84–89.
- Iuliano, M., L. Ciavatta, and G. De Tommaso. 2007. On the solubility constant of strengite. Soil Sci. Soc. Am. J. 71:1137–1140.[Abstract/Free Full Text]
- Kittrick, J.A., and M.L. Jackson. 1955. Application of solubility product principles to the variscite–kaolinite system. Soil Sci. Soc. Am. Proc. 19:455–457.
- Kolthoff, I.M., and P.J. Elving. 1966. Treatise on analytical chemistry. Part II. Analytical chemistry of inorganic and organic compounds. Vol. 4. John Wiley & Sons, New York.
- Martell, A.E., and R.M. Smith. 1989. Critical stability constants. Vol. 6. 2nd suppl. Plenum Press, New York.
- Lindsay, W.L., M. Peech, and J.S. Clark. 1959. Solubility criteria for the existence of variscite in soils. Soil Sci. Soc. Am. Proc. 23:357–360.
- Wenzl, P., L.I. Mancilla, J.E. Mayer, R. Albert, and I.M. Rao. 2003. Simulating infertile acid soils with nutrient solutions: The effects on Brachiaria species. Soil Sci. Soc. Am. J. 67:1457–1469.[Abstract/Free Full Text]
- Winkler, L.W. 1931. Ausgewählte untersuchungsverfahren für das chemische laboratorium. F. Enke Verlag, Stuttgart, Germany.
TOP
ABSTRACT
INTRODUCTION
