4.2. Crystallization Fileds of Individual Ammonium Phosphates
Ammonium Orthophosphates in the NH3-H3PO4-H2O system
A. Solubility branches on isotherms. The coordinates used for all the figures in this Critical Evaluation are: (NH4)3PO4-H3PO4-H2O. These are the coordinates also used for the discussion of alkali metal orthophosphates in our earlier volume.22 1. (NH4)3PO4 and its hydrates.
Most of the articles report the trihydrate (NH4)3PO4·3H2O as the solid phase in equilibrium with solutions in the region having the highest concentration of NH3. Anhydrous (NH4)3PO4 is reported as the stable solid phase at 25 °C by Parker2 and Flatt et al.16 The latter authors also designated (NH4)3PO4 as the equilibrium solid phase at 0 °C in solutions in the region having the highest NH3 concentration. The solubility curve reported by Flatt et al.16 is in good agreement with that reported by other authors and it is possible that Flatt and his co-workers neglected to determine the extent of hydration of the stable equilibrium solid phase. In all likelihood they used materials identical to those used by others. On the other hand, Jänecke's study at higher temperatures5 reports that (NH4)3PO4·2H2O is the stable solid phase at temperatures above 100 °C.
Reports of the solubility of (NH4)3PO4 in solutions of NH3 and H3PO4 at 273 K (Fig. 5) and at 298 K (Fig. 6) can be evaluated. The solubility data at 273 K reported by Jänecke,3 Muromtsev6,7 and Flatt et al.16 agree fairly well with each other and may tentatively be accepted as correct. A comparison of the data at 298 K indicates that the data of Parker2 and the more recent results of Vol'fkovich et al.10 are in error and should be rejected.
2. (NH4)7H2(PO4)3.
(NH4)7H2(PO4)3 is reported as a stable solid phase at 333 K and 348 K by Brosheer and Anderson,11 and at 323 K by Fleet et al.16 However, Muromtsev6,7 did not find it at 323 K, nor does Jänecke,5 mention it in his study at 373 K. Obviously, additional work is necessary before a recommendation can be made about the existence of this compound.
3. (NH4)2HPO4.
The isotherms at 273 K are shown in Fig. 7 and those at 298 K in Fig. 8. These figures represent the solubility field of (NH4)2HPO4 in the NH3-H3PO4-H2O system. From Fig. 7 it is evident that the solubility results reported at 273 K by different authors differ substantially. The evaluator's opinion is that, at 273 K, the phase diagram is markedly influenced by the possibility that the dihydrate (NH4)2HPO4·2H2O is formed. According to Kaganskiy and Babenko,19 the triple point for the simultaneous crystallization of NH4H2PO4, (NH4)2HPO4 and (NH4)2HPO4·2H2O is at 3.4 °C. However, Balabanovich et al.17 point out that measurements in this region are complicated by supersaturation. Therefore, the nature of the equilibrium solid phase may depend on the experimental conditions.
At 298 K, the solubility results of Parker2 have a systematic error and must be rejected. This also appears to be true for the data presented by Flatt, Brunisholz and Dagon16 for solutions from which NH4H2PO4 and (NH4)2HPO4 precipitate simultaneously in spite of the fact that their complete isotherms are in agreement with those of other authors. Vol'fkovich9,10 presents two isotherms for this region. In one, the composition unit is g/100 g H2O; in the other it is 100wI. The results of the two isotherms are not identical. The former appear to be in error and are rejected. The rest of the data1, 9, 10, 13 agree with each other and are recommended tentatively.
4. NH4H2PO4.
NH4H2PO4 crystallizes congruently in a well developed crystallization field. The data can be evaluated at 273 K, 298 K and 323 K, Fig. 9, Fig. 10, Fig. 11 and Fig. 12. These figures show that the data of the different investigators are in good agreement with each other, except for the eutonic points reported by Flatt, Brunisholz and Dagon.16
5. NH4H5(PO4)2·H2O.
Both Muromtsev6,7 and Flatt15 report that this compound has a definite crystallization field in the most acid region (100wi H3PO4 > 61) at 273 K. The data do have some scatter, probably due to analytical difficulties. Additional data are needed before a recommendation can be made. A similar situation exists at 298 K where Muromtsev's data6,7 differ appreciably from those of Flatt and co-workers. The latter found crystallization fields for NH4H5(PO4)2 and (NH4)3H9(PO4)4 in their detailed study of the region with excess H3PO4.15
6. The tentative data for the solubility isotherms of (NH4)3PO4·3H2O, (NH4)2HPO4, and NH4H2PO4 can be expressed by the following equations:
wsalt = a0 + a1·wacid + a2·w2acid + a3·w3acid [1] or msalt = bo + b1·macid + b2·m2acid + b3·m3acid [2]
where salt = (NH4)3PO4, acid = H3PO4, w = 100wi, m = mol/kg H2O, and a and b are coefficients whose values are given in Table 1.