Back to: Projects

Phosphate Recovery


Institute of Freshwater Ecology, River Laboratory, East Stoke, Wareham, Dorset, BH20 6BB, England, UK.
E-mail: w


The most important factors controlling the chemical speciation of calcium and inorganic phosphate in fresh waters and waste waters are discussed. Solubility diagrams are used to illustrate the range of compositions that are predicted to be stable with respect to the formation of key calcium phosphate minerals and calcite. These include calcium hydroxyapatite, octacalcium phosphate, tricalcium phosphate and dicalcium phosphate dihydrate. The coprecipitation of phosphate with calcite is also considered.

The solution speciation, calculated from information on the formation of calcium phosphate ion-pairs at a given temperature, enables ion activities to be determined and the ion-activity product for the appropriate mineral to be compared with the thermodynamic solubility product. Examples of such calculations are presented for different types of water and some conclusions are drawn concerning future research developments.

Keywords: Calcium phosphate precipitation, calcite, phosphorus, phosphate.


Chemical and biochemical precipitation of calcium carbonate/phosphate minerals is common in natural systems [1,2,3], although the elucidation of the mechanisms of formation, initiation of growth and transformations between crystal forms of the minerals, remains a major challenge. For carbonates, the most common mineral is calcite with reported occurrences of other polymorpths such as vaterite, aragonite and the mono and hexahydrates [4,5,6]. When calcite forms in solutions containing dissolved phosphate as ortho- or hydrogen phosphate, phosphate ions are incorporated in the lattice through surface adsorption and incorporation at "kink" sites [7]. The surface density of coprecipitated phosphorus has been found to depend on the adsorption isotherm, i.e. temperature, ionic strength and pH of the solution. Such coprecipitation occurs in many hardwater lakes [8], in algal biofilms that are associated with river-bed sediments [9] and is a self-cleaning mechanism in natural systems [10].

In contrast to this, calcium phosphate minerals are not often found in fresh- and waste water. The thermodynamically most stable form, at normal temperatures and pressure, is calcium hydroxyapatite (referred to here as HAP) but this does not form readily in spite of the occurrence of very high supersaturations. Other mineral phases such dicalcium phosphate dihydrate (DCPD), octacalcium phosphate (OCP) and amorphous tricalcium phosphate (TCP) form as precursor phases that transform to HAP [11].


These may be summarized under the following headings:

  1. The Thermodynamic "Driving Force".

The most important controls are the temperature of the reaction, the chemical speciation of the solution and the solubility product of the mineral. The solubility of calcite decreases with increasing temperature whereas the solubilities of the calcium phosphate minerals are generally the opposite and increase with temperature. However, for both HAP and OCP, the solubility trends reverse at higher temperatures in the region 15 ñ 20 oC. The temperature dependence of the solubility of amorphous TCP or b- TCP is more uncertain.

The degree of saturation, W , and saturation index, SI, of a solution with respect to a mineral phase, are defined as:



where Ksp is the thermodynamic solubility product and IAP the ion activity product for the appropriate lattice ions. The saturation index is related to the Gibbs free-energy change for crystallization and is therefore a measure of the thermodynamic driving force for the reaction. The degree of supersaturation with respect to a particular mineral phase may be calculated from the ion-activity product (IAP in Eqn. (i)), and solubility product, Ksp. The calculation of the ion-activity product necessitates a detailed knowledge of the chemical speciation of the ions in solution. The high degree of phosphate ion bonding with calcium and magnesium in solution means that ion-pairing is very important to the calculation of the saturation index for the majority of natural and waste waters. This is demonstrated in Figure 1 for a solution of 0.2 mM Ca(HCO3)2 containing 0.2 mM KH2PO4 at pH values of 7,8 and 9 at a temperature of 20 oC. The dissociation constants for the ion pairs were as used previously [12]. Both CaHPO4o and CaPO4- ion-pairs are important in this pH range and affect the activities of both Ca and phosphate ions used in the calculation of the ion-activity product. The coordination of Ca ions with other anions such as sulphate and dissolved polyelectrolytes, particularly in waste waters will also affect the calcium activity. An alternative approach is to measure the calcium ion activity directly by using calcium selective electrodes rather than the calcium concentration by standard chemical techniques.

2) Kinetics of the Reaction.

Various regions of solution stability may be defined when the activities of the lattice ions are plotted on a two-dimensional diagram. The solubility line, when SI=0, defines the equilibrium between the solution and particular mineral phase. When SI>0, and in the absence of epitaxial growth surfaces, the solution is termed metastable so that even though the system is thermodynamically unstable, precipitation of a solid phase does not occur. This occurs widely in aquatic systems; for example, most hard waters are supersaturated with respect to calcite formation with SI values up to 1, i.e. up to ten times supersaturated. At higher supersaturations, heterogeneous nucleation occurs as crystals form on exposed surfaces where nucleation is aided by the symmetry of the surface molecules or ions thus reducing the interfacial energy with the electrolyte. At higher supersaturations, homogeneous nucleation, or spontaneous nucleation occurs. This may not involve the formation of the most thermodynamically stable phase but normally a "precursor phase" with a higher solubility than the stable phase. The formation of the precursor phase may also be in conditions where the composition of the solution and surface charge of the colloid enables the particles to be stabilized in suspension. It is usually difficult to measure discreet reaction rates for the formation of a specific mineral in the region where homogeneous nucleation occurs because of the simultaneous formation of several crystal forms, and conversions between the forms as the supersaturation index of the solution decreases. For example, Feenstra and de Bruyn [13] found that in neutral and slightly alkaline pH conditions, that an amorphous calcium phosphate phase precipitated and served as a template for the heterogeneous nucleation of OCP which in turn lead to the epitaxial growth of HAP.

3) Presence of Growth Inhibitors.

The region of metastability mentioned above, may also arise from the effects of inhibitors interacting with clusters of lattice ions prior to nucleation. Inhibitors may also interact with growing crystals and, depending on the degree of supersaturation of the solution, may prevent growth completely. Many inhibitors are known for calcium carbonate formation e.g. Meyer [14], and for selected inhibitors such as magnesium on HAP and OCP crystallization [15]. In some cases ions are incorporated into the lattice as the crystals grow and subsequently lead to changes in the reaction kinetics, e.g. F- or Cl- for OH- in HAP as well as CO32- and SO42- for phosphate [16]. In many instances, the effects of the inhibitors on the reaction kinetics may be estimated using the Langmiur equation describing the adsorption of the inhibitor molecules to the crystal surface [2]. Usually the effects of the inhibitor may be overcome by increasing the supersaturation of the solution. Generally, as the inhibitor concentration increases, a critical concentration is reached when growth is completely inhibited [17].

4) Competing Reactions

In fresh- and waste waters there are many competing reactions involving the incorporation of calcium and phosphate in minerals and also in macrophytes and algal cells. Examples include the so called "luxury uptake" of phosphorus by algae [18], formation of solid solutions of iron hydroxide and phosphate in oxic conditions [19] , formation of vivianite in anoxic conditions [20], and sorption to aluminium and iron oxides, clays and complexation with organic matter [21]. These reactions lead to changes in the activities of the lattice ions for calcium carbonate and phosphate formation and affect variations in the saturation index of the solution.

5) Localized Conditions.

The presence of surfaces and microbial or algal biofilms in contact with solutions that are supersaturated with respect to particular minerals often lead to localized conditions that are more or less favorable to precipitation when compared with the bulk solution. This has been demonstrated for Chlorococcum biofilms in which changes in pH (with increases of up to 1.6) and dissolved calcium were measured over distances of 800 mm above a photosynthetically active surface [22]. These chemical changes occur in the hydrodynamic layer above the biofilm and within the biofilm. In sediments, very large changes in pH, calcium and oxygen concentration have been recorded in the top 1 mm of sediment, [23], [24] and [25]. Calcium gradients in the sediments are often caused by the dissolution of calcium carbonate minerals in the sediments as a result of the low pH caused by high CO2 concentrations from microbial respiration.


Chemical speciation calculations enable the pH dependence of the solubility lines of calcite and the most important calcium phosphate minerals to be compared for particular concentrations of dissolved calcium and phosphate at a chosen temperature. Examples of the isotherms for hard water containing 2.5 mM dissolved calcium bicarbonate and 0.02 mM phosphate at 20 oC are shown in Figure 2. At pH 7 the solution is under saturated with respect to all the phases except HAP. As the pH increases to 9, the solution becomes thermodynamically unstable with respect to both calcite and OCP. In these conditions calcite precipitates and co-precipitates inorganic phosphate. There is no evidence for the formation of OCP in these conditions.

At higher concentrations of dissolved phosphate that are closer to those in waste waters, the supersaturations increase so that at pH=8.2 all the phases accept DCPD are theoretically able to form (Figure 3a). This is more clearly seen in Figure 3b with the isotherm for HAP is omitted. The solution remains under saturated with respect to DCPD over the entire pH range; an increase in the phosphate concentration is needed to move the isotherm to positive values of SI.


The proportion of phosphate co-precipitated depends on the temperature, pH and the concentration of other co-precipitating chemicals.


where s is the maximum surface density of phosphorus (approximately 0.15 mmol m-2) [9, 19], and the function h varies between ca. 0.1 and 0.9. This relationship predicts ratios of Ca : P of between 250 and 60 for precipitation in 0.2 mM Ca(HCO3)2 solution with pH in the range of 8-9 at 20 oC and in the presence of between 2 and 20 mM phosphate solution (Figure 4). This leads to concentrations of phosphorus in the precipitate of between 40 and 170 mmol g-1 which compare with ca. 6450 mmol g-1 for the formation of tricalcium phosphate (Table 1).


As shown in Table 1, the phosphorus contents of the minerals are very similar and therefore it necessitates accurate chemical analysis to distinguish between them from changes in the solution composition during crystal growth. At high supersaturations, i.e. SI>10, it is difficult to precipitate HAP alone because of the spontaneous formation of precursor phases such amorphous calcium phosphates and OCP. At lower supersaturations it has been found possible to nucleate HAP in pH conditions where the solutions are slightly under saturated with respect to amorphous TCP and OCP, e.g. pH 7.4 with total dissolved calcium and phosphate between 0.5-1.7 mM and 0.5-1.0 mM respectively, and SI in the range of 5-10 at ca 27 oC. When the growth of HAP is followed in suspensions of crystal seeds of HAP, it is difficult to detect other mineral phases. There is evidence from the shape of growth curves and determinations of the percentage crystallinity that crystal growth on the seeds may progress through precursor phases such as OCP. However, this is difficult to distinguish from adsorption, migration, dehydration and integration of the lattice ions at the surface as the HAP crystals grow. Certainly, there is no evidence for the nucleation of new particles of the precursor phases at these low super saturations. This is because the pH of the suspension is maintained near neutral so that the solution is not supersaturated with respect to DCPD and amorphous TCP but may be slightly saturated with respect to OCP.

Table 1. Comparison of the composition of the main minerals of calcium phosphate




Molar ratio Ca:P

P content/ mmol g-1 of crystals

Octacalcium phosphate, OCP

Ca4H(PO4)3 3H2O



Tricalcium phosphate, TCP




Calcium hydroxyapatite, HAP




Dicalcium phosphate dihydrate, DCPD




Generally when homogeneous precipitation occurs, i.e. spontaneous nucleation and growth, unstable amorphous solids precede the formation of crystalline HAP. In conditions when pH>9, amorphous calcium phosphates (ACPs), are thought to convert directly into HAP whereas at pH 7-9, they are converted to a precursor OCP phase, and then to HAP. The IAP for the ACP formed in alkaline conditions, e.g. pH=7.4-9.25, has been found to be relatively constant with starting Ca/PO4 molar ration of 1.31 to 1.48 and stoichiometry of the solids of Ca3(PO4)1.87(HPO4)0.2 [26]. The ACPs contain approximately 10 % HPO42- and 10-15% (by mass) water with about 75% of the water tightly bound forming a hydrated salt [27]. The conversion from the amorphous solids to OCP and HAP depends on the temperature and pH of the solution. Although OCP is heterogeneously nucleated on ACP at pH<9.3, the rapid hydrolysis of OCP to HAP in more alkaline conditions makes it difficult to detect the precursor OCP phase. Hence at higher pH, the conversion from ACP to HAP appears direct without intermediates [26].

Seeded growth methods have been used to investigate the growth kinetics and effects of other ions on the kinetics and morphology of the crystals. For example, the presence of chloride ions in the solution favours the formation of crystal plates of HAP rather than needle-like crystals. Magnesium ions are known to inhibit the growth of HAP and to a less extent OCP but not DCPD. Lithium also inhibits the growth of HAP and is incorporated into the HAP lattice during growth. A range of other inhibitors have also been investigated such as glucose, fulvic, humic, phytic, citric, mellitic and tannic acids. Generally they act by a "site-blocking" mechanism whereby the inhibitors adsorb at growth kink sites, the extent of the coverage of the sites being consistent with the Langmiur adsorption equation [17].

Octacalcium phosphate is a commonly cited precursor phase to HAP formation. Even in the presence of b-TCP seeds, in solutions supersaturated with respect to TCP, OCP and HAP, the OCP phase grows in preference to TCP or HAP [28]. During homogeneous nucleation at high supersaturation (SI for HAP >20), e.g. T=20oC, pH=9, and calcium and phosphate concentrations of 10 and 7 mM respectively, two precursor phases of amorphous calcium phosphate of similar composition to TCP have been identified. The amorphous phases eventually dissolve to form OCP and finally HAP [29]. The crystallization of OCP has also been measured using seeds of OCP in a solution supersaturated with respect to the DCPD, OCP and HAP but under saturated with respect to amorphous TCP. The formation of the amorphous phases is thought to be important to the subsequent development of OCP and may determine whether DCPD also nucleates in preference to OCP. This complex interaction between OCP and DCPD has also been investigated in seeded growth experiments using OCP seeds. At low supersaturations at pH 6 and 25 oC, OCP growth occurs, whereas at high saturations, DCPD grows after an induction period, [30].


There is a need to obtain more fundamental information about calcium phosphate precipitation in complex solutions, and in particular:

  • The temperature dependence of the solubilities of the precursor phases, such as the amorphous phases and OCP, over conditions likely to occur in phosphorus recovery plants.
  • The stability regions for OCP and amorphous tricalcium phosphate for conditions found in complex solutions such as fresh- and waste waters. This information is necessary to evaluate the optimum conditions for growth in commercial reactors designed for the precipitation of calcium phosphates.
  • The effects of the inhibitors, e.g. dissolved magnesium, trace metals and polyelectrolytes, likely to be important in the recovery calcium phosphate from waste waters.


The author thanks the Natural Environment Research Council, UK for their support.


Figure 1. Comparison of the chemical speciation in a hard water (0.2 mM calcium hydrogen carbonate) at 20 oC in the presence of 0.2 mM potassium hydrogen phosphate.

Figure 2. Solubility lines for the main calcium phosphate phases at 20 oC in a hard water of 2.5 mM calcium bicarbonate and 0.02 mM total dissolved phosphate. Key: CAL: calcite, TCP: tricalcium phosphate, DCPD: dicalcium phosphate dihydrate, HAP: calcium hydroxyapatite, OCP: octacalcium phosphate. Thermodynamic data taken from reference [12].

Figure 3. (a) Solubility lines for the main calcium phosphate phases at 20 oC in a hard water (2.5 mM calcium bicarbonate) in the presence of 0.2 mM total dissolved phosphate. (b) as above but excluding HAP. Key: CAL: calcite, TCP: tricalcium phosphate, DCPD: dicalcium phosphate dihydrate, HAP: calcium hydroxyapatite, OCP: octacalcium phosphate. Thermodynamic data taken from reference [12].

Figure 4. Amount of phosphate coprecipitated with calcite at 20 oC in a hard water (2 mM calcium hydrogen carbonate) calculated from equation (3) with the maximum surface density of phosphate= 0.15 mmol m-2


  1. 1. Mann, S., Webb, J. and Williams, R.J.P. Biomineralization. Chemical and Biochemical Perspectives, VCH, Weinheim. (1989).
  2. Brady, P.V. and House, W.A. Surface-controlled dissolution and growth of minerals, In: Physics and Chemistry of Mineral Surfaces, Brady, P.V. (ed), CRC Press, Boca Raton, Ch. 4, 225-307 (1996).
  3. Drever, J.I. The Geochemistry of Natural Waters: Surface and Groundwater Environments, Prentice Hall, NJ. (1997).
  4. Wray, J.L. and Daniels, F. Precipitation of calcite and aragonite, J. Amer. Chem. Soc., 79, 2031-2035, (1957).
  5. Slack, J.G. Calcium carbonate hexahydate: Its properties and formation in lime-softening. Water Res., 14, 799-804. (1980).
  6. Speer, J.A. Crystal chemistry and phase relationship of orthorhombic carbonates. Review Mineralogy, Mineralogical Soc. America, vol.11 ch.5. (1983).
  7. House, W.A. & Donaldson, L. Adsorption and coprecipitation of phosphate on calcite. J. Coll. Interface Sci. 112, 309-324 (1986).
  8. Stabel H. -H. Calcite precipitation in Lake Constance: Chemical equilibrium, sedimentation, and nucleation by algae. Limnology and Oceanography, 31, 1081-1093 (1986).
  9. House, W.A. and Denison, F.H. Nutrient dynamics in a lowland stream impacted by sewage effluent: Great Ouse, England. Sci. Tot. Environ., 205, 25-49 (1997).
  10. Koschel, R., Benndorf, J., Proft, G. and Recknagel, F. Calcite precipitation as a natural control mechanism of eutrophication. Arch. Hydrobiol., 98, 380-408 (1983).
  11. Van Kemenade, M.J.J.M. and de Bruyn, P.L. A kinetic study of precipitation from supersaturated calcium phosphate solutions. J. Colloid Interface Sci., 118, 564-585 (1987).
  12. Hartley, A.M., House, W.A., Callow, M.E. and Leadbeater, B.S.C. Coprecipitation of phosphate with calcite in the presence of photosynthesising green algae, Water Res., 31, 2261-2268 (1997).
  13. Feenstra, T.P. and de Bruyn, P.L. Formation of calcium phosphates in moderately supersaturated solutions. J. Phys. Chem., 83, 465-479 (1979).
  14. Meyer, H.J. The influence of impurities on the growth rate of calcite. J. Crystal Growth, 66, 639-646 (1984).
  15. Salimi. M.H., Heughebaert, J.C. and Nancollas, G.H. Crystal growth of calcium phosphate in the presence of magnesium ions. Langmuir, 1, 119-122 (1985).
  16. Nancollas, G.H. In vitro studies of calcium phosphate crystallization. In: Biomineralization. Chemical and Biochemical Perspectives, VCH, Weinheim, ch. 6, pp157-187 (1989).
  17. House, W.A. Kinetics of crystallisation of solids from aqueous solutions. In: Comprehensive Chemical Kinetics Reactions at the liquid-solid interface, Elsevier ,vol. 28, ch.3. (1989).
  18. Reynolds, C.S. The Ecology of Freshwater Phytoplankton, Cambridge University Press, Cambridge, pp 36.
  19. Fox, L.E. Phosphorus chemistry in the tidal Hudson River, Geochimica et Cosmochimica Acta., 55, 1529-1538 (1991).
  20. Sarazin, G, Gaillard, J.F., Philippe, L. and Rabouille, C. Organic matter mineralization in the pore water of a eutrophic lake (Aydat Lake, Puy de Dome, France). Hydrobiogia, 315, 95-118 (1995).
  21. Krom, M.D. and Berner, R.A. Adsorption of phosphate in anoxic marine sediments. Limnology Oceanography, 25,797-806 (1980).
  22. Hartley, A.M., House, W.A., Leadbeater, B.S.C. and Callow, M.E. The use of microelectrodes to study the precipitation of calcite upon algal biofilms, J. Colloid and Interface Sci., 183, 498-505 (1996).
  23. Archer, D., Emerson, S. and Reimers, C. Dissolution of calcite in deep-sea sediments: pH and O2 microelectrode results. Geochimica et Cosmochimica Acta, 53, 2831-2845 (1989).
  24. De Beer, D., Schramm, A., Santegoeds, C.M. and Kuhl, M. A nitrite microsensor for profiling environmental biofilms, Applied Environmental Microbiology, 63, 973-977 (1997).
  25. Woodruff, S.L., House, W.A., Callow, M.E. and Leadbeater, B.S.C. The effects of biofilms on chemical processes in surfacial sediments, Freshwater Biology (in press).
  26. Eanes, E.D. Amorphous calcium phosphate: thermodynamic and kinetic considerations. In: Calcium Phosphates in Biological and Industrial Systems, Amjad, Z (ed.), Kluwer Academic Publisher, ch.2 (1998).
  27. Sedlak, J.M. and Beebe, R.A. Temperature programmed dehydration of amorphous calcium phosphate and crystalline hydroxyapatite. J. Colloid Interface Sci., 47, 483-489 (1974).
  28. Heughebaert, J.C., Zawacki, S.J. and Nancollas, G.H. The growth of octacalcium phosphate on beta tricalcium phosphate, J. Crystal Growth, 63, 83-90 (1983).
  29. Christoffersen, M.R., Christoffersen, J. and Kibalczyc, W. Apparent solubilities of two amorphous calcium phosphates and octacalcium phosphate in the temperature range 30-42 oC, J. Crystal Growth, 106, 349-354 (1990).
  30. Heughebaert, J.C., De Rooij, J.F. and Nancollas, G.H. The growth of dicalcium phosphate dihydrate on octacalcium phosphate at 25 oC, J. Crystal Growth, 77, 192-198 (1986).