Phosphate Minerals


Lindsay, W.L., P.L.G. Vlek, and S.H. Chien. 1989. Phosphate Minerals. Ch. 22, p. 1089-1130. In: J.B. Dixon and S.B. Weed (ed.), Minerals in Soil Environments, 2nd Edition.


By one count, 172 naturally-occurring phosphate minerals have been identified. A number of phosphate mineral classification schemes exist but we will follow that chosen by Lindsay et al. (1989), which is that of Povarennykh (1972; Crystal chemical classification of the minerals. Vol. 1 & 2, Plenum Publ. Corp., N.Y.) and which is similar to that used by silicate minerals--framework, chain, and layer phosphates. However, unlike the silicate polyanion, SiO44-, the ortho-phosphate polyanion, PO43-, does not generally polymerize in the formation of crystal structure and so the analogy to silicates is necessarily that, an analogy. [When phosphates do polymerize, polyphosphates are formed, starting with pyrophosphate, PO2(=O)-O-P(=O)O24-, in which a bridge O links to phosphate tetrahedra. Such condensation reactions to prepare polyphosphates require oven temperatures when conducted industrially but cellular organisms routinely conduct low-temperature synthesis of adenosine triphosphate (ATP), which consists of an ester bond between tripolyphosphate and adenosine and which is dephosphorylated to adenosine diphosphate (ADP) to release energy. Polyphosphates are produced from phosphoric acid by the fertilizer industry as a phosphorus source.]

With so many phosphate minerals identified and organized into 14 groups in four classes, it seems practical to limit discussion to those typically found in the soil environment. Apatite constitutes 95% of the P in igneous rocks and, due to its persistence, almost certainly constitutes a similar percentage of the P in sedimentary rocks and soils. Apatites have the general formula A10(XO4)6Z2, where A is most often Ca2+, X is P(5), and Z is most often OH and/or F. Other divalent cations may substitution in A, among them Sr, Mn, Pb, Mg, Ba, Zn, Cd, as well as monovalent cations (Na, K, and Rb) and trivalent cations (Sc, Y, Bi). For X, many elements that form tetrahedrally-coordinated oxides, among them Si, S, As, V, Cr, and Be, can substitute for P.

Apatite is very insoluble and there is relatively little or no fertilizing potential for rock phosphate in unprocessed form. Therefore, phosphorus fertilizers are produced by acidulating rock phosphate, ore containing apatite of either the hydroxyapatite, fluorapatite, or francolite (carbonate) varieties, by one of two general schemes: Either sulfuric acid sufficient to produce a mixture of monocalcium phosphate and gypsum [the mixture marketed as "superphosphate"], or sulfuric acid sufficient to produce phosphoric acid, which is separated from gypsum by centrifugation and either reacted with more apatite to form monocalcium phosphate [marketed as "triple superphosphate"] or compounded with ammonia to produce mono- or di- ammonium phosphates or, less commonly, roasted to form polyphosphates. These processes produce water-soluble phosphate fertilizers; partially acidulated rock phosphate routinely underperforms as a fertilizer material.

Among solution chemists, discussion of phosphate chemistry in soils often considers the presence of strengite, FePO4·2H2O, and variscite, AlPO4·2H2O, as controlling the solubility of phosphate in acid and slightly acid soils. However, early reports (based largely on x-ray identification) of strengite and variscite as reaction products of phosphate fertilizers with acid soils have been challenged and presumed solubility control by strengite and variscite is not sufficient evidence of their presence in soils. Among surface chemists, discussion revolves around the adsorption of phosphate on iron and aluminum oxides, aging into less labile forms with time. Current approaches to considering phosphorus movement in the environment, principally that of A. Sharpley and co-workers, is to propose a "phosphorus capacity" equal to the sum of ammonium oxalate-extractable iron and aluminum oxides.

Addition of phosphate fertilizers to calcareous soils is thought to produce a series of calcium phosphates:

monocalcium phosphate (fertilizer) ==> dicalcium phosphate dihydrate ==> octacalcium phosphate ==> hydroxyapatite

Proposed reaction series for phosphate fertilizers in calcareous and neutral soils:  
Name Chemical Formula Ca/P
monocalcium phosphate Ca(H2PO4)2 1/2 = 0.5
brushite, dicalcium phosphate dihydrate CaH(PO4)·2H2O 1/1 = 1.0
octacalcium phosphate Ca8H2(PO4)6·5H2O 8/6 = 1.3
hydroxyapatite Ca10(PO4)6(OH)2 10/6 = 1.7

In reality, the existence of 49 known calcium phosphates, 14 magnesium phosphates, 49 aluminum phosphates, and 51 iron phosphates (and some contain one or more of these cations) points to the complexity of the chemistry of phosphate minerals in soils and it is dubious that any fertilizer reaction species or series is more than illustrative.


Phosphate minerals comprise only a fractional percentage of the soil mass, perhaps averaging 0.02% or so, which impedes their direct identification by optical microscopy and x-ray analysis unless their concentration has been locally increased by addition of phosphate fertilizers. In some cases, beneficiation of native phosphate minerals is possible by dissolving ancillary minerals with HF; yet others separate phosphate minerals based on heavy liquid separations. Advances in microscopy allow both semi-quantitative elemental analysis and electron diffraction of individual particles.

Another phosphate mineral of some interest is struvite, NH4MgPO4. That this ammonium salt is insoluble is somewhat surprising given that ammonium phosphates, both mono- and di, are quite soluble, as are most other ammonium salts. Because this mineral contains nitrogen, an element associated with biota, it is to be found in conjunction with organisms, either alive or dead: urinary stones, manure, canned fish, or cadavers immersed in seawater. In the past, struvite formation has been used as an analytical technique for gravimetric determination of Mg2+, by addition of ammonium phosphate to the test solution.

Authors: Ed Nater and Phillip Barak
Dept of Soil, Water, and Climate, U of Minn. and Dept of Soil Sci, U of Wisc.

Copyright: Ed Nater and Phil Barak
Copyright for mineral models held by the Minerals & Molecules Project

The opinions expressed herein are those of the authors and do not necessarily represent those of their respective universities or their Regents.