INORGANIC AND A N A L Y T I C A L CHEMISTRY A STUDY A

OF T H E C e P O 4 - H s P O g - H 2 0 t

U!~yanov

and

T

I.

SYSTEM

A T 25 ~

Kazakova

Institute for General and Inorgamc Chemistry. Academy of Scmnces, USSR Translated from Izvesnya A k a d e m n Nauk SSSR, Senya Khlmmheskaya, No. 7, pp. 1157-~_164, July, 1963 Orzgmal article submitted Iune 30, 1963

The sombfllty of the rare earth phosphates m acids, and espemally m phosphoric acid, Is a matter of prime ~lgm h e a n c e for the technology of mineral analysis and separation of the rare earths since the phosphates are among the more commonly met forms of these elements. This study of the CePO4-HsPO4-H~O system at 25 ~ was suggested by the lack of data on the solubility of CePO4 m phosphoric acid. It was also proposed to determine the posslbihty of the existence of aczd eerophosphates [3], earher warp having shown that such compounds are not formed m the systems CeC13 [CeBrs, Ce 2(SO4)s]- HsPO4 (and its sodmm salts)- H20 [ 1, 2]. EXPERIMENTAL Tn~s study revolved determination of the solubrllty and apparent volume of the sohd phase, and the speelfm gravity and index of refraction of the hqmd phase, The working materials were a CP orthophosphorle acid and an air-dried hexagonal modificauon of the cerophosphate CePO4 , 2H~O, which had been synthesized by the method which we proposed earlier [4]. To 25 ml samples of HsPO4 solutions of various concentrations in calibrated tubes of 35 ml capacity, 1 g samples of the powdered CePO4 . 2H20 were added m small successive portions. The contents of each tube were agitated slx hours a day for 16 days, using the device which we have proposed earhcr [5], and then allowed to stand while the apparent volume of the solid phase was measured, at first after two days, and then every day, until 12 days had elapsed. The refractive index of an aliquot portion of the transparent hqmd phase was then measured with an IRF-22 refractometer. Following this, the contents of each tube were subjected to renewed agitation for five days (Le., for an addlttonal 120 hours) and the Index of refraction of the liquid phase measured three days later after the liquidhad cleared up. Constancy of the value of the index of refraction of the hqmd over this period of n i n e was taken as indication of the establishment of a state of equilibrium m the system. The volume of the solid phase was once more measured at the end of a month and half month, f.e., after the elapse of 60 days. The hquid and solid phases were h n a l l y separated by centrifuging. The following determinations were carried out on aliquot portions of the liquid phase: 1) specific gravity, by weighing m wezghmg-bottles, 2) H~PO4 content, by titratmg with 0.1-0.2 N NaOH, using methyl orange as indicator, 3) CePO4 content, c o l o n m e t n c a l l y on the basis of the complex between Ce ~+ and T a l o n B m a g l y c e r i n - a m m o n i a medium [ 6 ] m Expts. 1-15, and either volumetrically or gravimetrtcally m Expts. 16-19 where the cerium content of the hqutd phase was high. In the volumetrm determination the cerium was oxidized w!th a m m o m u m persulfate and the Ce ~* then utrated w~th a 0.05-0.10 N solutmn of Mohr's salt, using a ferrous ~on md~cator In t h e g r a w metrm d e t e r m m a u o n the certain was brought down with oxalic acid, repree~pttated from a solutlon m which the acidity was no higher than that of 0,3 N HNOs, and the resulting cerium oxalate calcined and weighed as CoO z [7]. 'The combined filtrates were used for a supplementary POl" mn d e t e r m m a t m n by the citrate method with ~eprecipltauon [8]. The volumetrm and gravlmetrlc d e t e t m m a u o n s of the POI- content of the liquid phase whmh are presented m Table 1 show the tJtrauon of HsPO4 by the alkah to have been essenually complete. The solid phase of this sytem was highly dispersed and could be filtered only with difficulty. For this reason the solid phases were centrifuged and then compressed at 150 atmos, in gxpts, 1-14. Analysis was carried out by placing a weighed amount of the powder (Expts~ 1-14) or of the pasty, freshly centrifuged mass (Expts. 15-19) on a watch glass, and dissolving it in a slight excess of mtrm acid, The resulting solution was transferred to a 100 ml measuring flask, from which aliquot portions were taken for the determination of Ce s+ and PO~-. In Expts, 1-14 the cerium was determined as weight % CePO4, using both the volumetrm method (oxldatmn of the solution with a m -

1063

monium persulfate and titration of the resulting Ce4+ ions with Moore's salt) and the gravimetrlc method (precipitation with oxalic acid [9], the cerium oxalate being calcined to CeOz). The PO43- content of the filtrate was determined as weight % HsPO4, using the citrate method with repreclpitatlon [8] In Expts. 18-19, the certain was determined in one aliquot p o m o n by oxidation to the tetravalent state wlth a m m o n i u m persulfate followed by titration of the Ce4* with Moore's salt. The PO~" was determined in still another ahquot by the above mentioned citrate method ruth reprec,pitatmn. Under ordinary magnification, the solid phase appeared as a collection of very minute crystallites of indefinite form. DISCUSSION

TABLE 1. Volumetric and G r a v l m e t n c Determinations of HsPO4 m the Llqmd Phase

OF E X P E R I M E N T A L

RESULTS

The fact that the index of refraction (Fig. 1) and the specific gravity (Fig. 2) of the hqmd phase, and the solubihty of the sohd CePO4 in this liquid phase (Figs. 3, 4) all increased monotonically with the HsPO4 content of the solution was indication that chemical compounds volumetric gravimetrlc were not formed in this system. The same conclusion could be drawn method method mean from the curves showing the variation of the apparent volume of the sohd phase (Fig. 5). The fact that the Schreinmakers lines based on 60.09 59.86 60.00 the Roozeboom triangle of the phase diagram diverged (see Fig. 3) 76.36 76.91 76.63 indicated that the sohd phase varied in composition and should be 84.77 84.83 84.80 formally considered as a continuous series of solid solutions ranging from CePO4 to HsPO4. Figure 5 shows that the volume of the solid phase became essennally constant m the course of a week m solutions containing up to 10 weight % HsPO4. The solid began to dissolve only after the H3PO4 content of the liquid phase had been rinsed to 15 weight % (Table 2) Increasing the HsPO4 content of the llqmd phase brought about a corresponding increase in the solubility of the CePO4. The curve showing the varlatmn of the volume of the solid in the e q m h b r m m state (see Fig. 5, Curve 3 ) m d m a t e d that three processes were taking place m the system, namely: dissolution of the CePO4 m the liquid phase, formation of a solid phase varying m composition between CePO4, HsPO4, and HzO; and peptization of the CePO4 . xHsPO4 'aqueous by the excess H3PO4 of the hqmd phase On this basis, the curve showing the v a r l a n o n of the volume of the solid phase could be dtwded into two sections: 1) the apparent volume of the solid gradually increases under the influence of the three processes as the H3PO4 content of the hqmd phase is increased from 15 to 50 wmght %; 2) the apparent volume of the sohd diminishes because of a marked increase m solubihty and d~mlmshed peptizatton over the range from 50 to 70 weight % H3PO4 m the liquid phase. Found, weight% HsPO4 in liquid phase

'

Phosphoric acid enters m varying amounts into the sohd phase where ~t is weakly bound to the CePO4, clearly by other than chemical forces, It is possible that this is due to an internal "porosity" of the crystal structure of the CePO4, since the latter contains zeohte water [4, 10], This water ~s lfl~ n D / mobile and at the same time rather firmly bound to the CePO4, so that it can probably be easily displaced by phosphoric acid, For this reason /x the arrangement of Schremmaker's hnes m the phase diagram of Fig, 3 is that corresponding to solid solution formation,

t

x

/

/ / / 7,J8

The existence of a solid phase of variable composmon (of the type CePO4 xHsPO4 aq) could be explained in various ways [11]. 1) by the

x

/

'x7 1,8 l,tl

/,36 /x p~ x

~,3z/

xj x x~

' GO T 70 , 80 , 0 lO' 20',,~9 3 q ~0 50 HsPO4 content of liquid phase, weight %

F~g. 1. Variation of the index of refracn o n of the hqmd phase m the system

CePO4-HsPO4-HzO 1064

t:K) 0

/'00

I0

2g

3g

z/a

50

GO

70

~0

HsPO4 content of h q m d phase, weight % Fig. 2. Variation of the specific gravity of the hqmd phase m the system CePO4-HsPO4-H20 at 25 ~ (1), and in the HsPO4H~O system at 20" (2),

~25I 16 ! d Zg

95 ~5

(" o,/ I;e PO~

H~P0~ Fig. 3. Solubility in the system CePO4 H20 at 25 ~

-

o,z

o,a

o,u

HsPO4 content of liquid phase, mole % Fig. 4. Solubility of MPO4 in phosphoric acid (0.00-0.45 mole %) at 25~ 1) CePO4; 2) LaPO4, according to the data of [15]; 3) GdPO4, according to the data of S.M. Petushkova; 4) GaPO4, according to the data of [21]; 5) FePO4, according to the data of [19]; 6) InPO4, according to the data of [18]; 7) A1PO4, according to the data of [17].

HsPO4 -

20, .~ 8oat I

2~6~

g .aoz

I

N lZ i

8 ~ zOoz > >ill 0

to

eo

' s?

' e?

'

70

HsPO4 content of liquid phase, weight % Fig. 5, T i m e variation of the apparent volume of the solid phase (ml) in the system CePOa - HsPO4 - H20. 1) after two days; 2) after 12 days; 3) after 60 days; 4) volume of 1 M precipitate (variation of compactibility) after 60 days. fact that the system was in a metastable state (this would not apply to the present work in which the system wasstudied at equilibrium); 2) by the formation of solid solutions (which is quite unlikely, in view of the great difference in the radii of the H+ and Ce s+ ions); 3) by adsorption (which, if it took place at all, must have been insignificant in comparison with the HsPO4 which was present in the CePO4); and 4) by ion exchange (this is obviously the most plausible explanation). There seems to be no reason to exclude the possibility that a solid phase of variable composition was formed as the result of various physicoehemieal effects, one of which predominated over the others, Thermograms'developed for this system also indicated the formation of a solid phase of variable composition (Fig. 6). The fact that.an exothermic effect appeared on the heating curve at aoo ~ (instead of the endothermic effect

1065

at ,,,350~ [4] for CePO4 9 2H20) was the result of a polymorphle transforrnatmn of CePO4 from the hexagonal mto the more highly stable monoclimc crystalline modification. Prior to developing these thermogral-nS, the precipitate was washed ruth ethyl alcohol and then dried over sulfuric acid. It could be that the alcohol displaced the zeohte water from the cerophosphate and was in turn, readily d~splaced in the course of drying. In any event, the heating curve for the solid phase of CePO4 9 xHsPO4 aq (see Fig. 6) did not show the endothermm effect at 150" whmh is due to loss of water by CePO4 " 2H20 [4], nor the endothermm effect at "350* whmh ts associated with loss of water and polymorphm transformatmn tn this same compound, the presence of phosphoric acid m the sohd phase causing the latter effect to become exothermm and to appear at a somewhat lower temperature of 300*. The same formanon of a sohd phase of variable composmon which we have observed at 25~ has also been reported by Fig. 6 Heatmg curve for the solid Sveshmkova and Omzburg as occurnng m the system CePO4-HsPO4-HzO at 70" phase from the system CePO4 [12]. The fact that the HsPO4 can be gradually washed out of the CePO4 ' xH~O HsPO4 - HzO (25"). 9 aq phase with water ts also m d l c a u o n of the variable composmon of the latter. For example, the phosphoric acid could be only parnally removed from the sohd phase m Expts. 7 and 8 (23 and 33 weight fro, respectively, Table 2) by the use of bothng water with digestion for24h.* The precipitate proved to be a sohd phase of variable composmon whmh gradually absorbed water from the atmosphere. 300 ~

TABLE 2. Data from Analysis of Liquid and Sohd Phases m the System CePO4 - HsPO4 - HzO Expt No

Lzquid phase

%e%, wt. % t,42

2 3 4 5 6 7 8 9 10

4,47 7,03 9,90 15,31 19,05 23,t3 24,26 30,64 32,29

cepo,, wt. % Notdetec t table " [ " " 0,001 0,004 0,0tl 0,012 0,036 0,050

Sohd phase Ltquid phase ( ' p r e c i p i t a t e ' ) ~:p. t. H3PO4, CePOa, I-I~PO4, CePO4, wt. % wt % wt. % wt. % 1,50

90,t0

3,50 5,02 7,50 10,79 12,87 14,97 t6,50 23,00 26,00

81,50 85,00 80,0~ 84,16 81,49 72,78 77,10 45,10 55,30

II 12 13 14 15 16

35,23 41,48 49,43 56,03 60,00 66,54

i7 18 19

72,62 76,63 84,80

0,081 0,t54 0,299 0,551 0,825 1,155 1,550 1,825 2,339

Sohd phase (" preclpjtate") H~PO4, CePO4

wt.%

wt %

30,04 / 45,10 38,50 35,05 45,10 35,00 52,50 30,03 58,78 9,49 64,03 21,20 71,10 15,01 74,05 20,02 82,56 ~4,58

It is to be noted that there ts mdtcatton m the hterature of the existence of acM certain phosphates [3] The present data, as well as the data obtamed earher [ 1 . 2 ] , would mdmate that such phosphates cannot exist The statements of certain authors to the effect that the sulfuric acid rare earth extraction bnngs down an acid orthophosphate Ge2(HPO4)s from monoztte at pH 2.3 and a normal certain phosphate CePO4 from apattte [14] are also incorrect Our studms show that this u e a t m e n t brings down a sohd phase of variable composmon this phase conslstmg of either CePO4 and HsPO4, or of CePO4 and Cez(SO4) s [2]. The mcrease m the solublhty of CePO4 with mcreasmg HsPO4 content of the solvent can be explained by the formation of a new sohd phase or a complex m solution The dtssolutwn of CePO 4 m solutions contammg 15%, or more, of HsPO4 by weight can proceed through the reacnons. HaPO4 aq 2CeP04 + HaPO4 . ~ Ce2 (HPO4)s, CePO4 + 2HAP04 --* Ce (H2P04)~, CePO4 -~ HsP04 --, CePO~. H~P04, CeP04 -~ HAP04 ~ Ha ICe (PO~)2] * The phosphoric acid extracted from the CePO4 9 xHaPO4 9 aq phase was determmed by tttratmg with NaOH, usmg methyl orange as mdmator.

1066

(1) (2) (3) (4)

The fact that acid cerophosphates are not formed m th~s system at 25* and at 70* has been shown by ourselves on the one hand and by Sveshmkova and Omzburg [12] on the other In addition, the HsPO4 content of the hquid phase proved to be the same when determined v o l u m e t r i c a l l y and when determined g r a w m e t r i c a l l y Only m the opposite case would ~t have been permissible to speak of the formatmn of a cerophosphonc acid complex since it would be difficult to imagine that the latter would have the same force as HsPO4 In wew of these facts, it can be considered that the dlssolutmn of CePO4 m aqueous H3PO4 proceeds through reaction (3).

~75

L'7 "G1

Our own value of the solublhty of CePO4 m aqueous HsPO4 at 25 ~ is considerably greater than the solub~hty found by Sveshmkova and Gmzburg [12] at 70", which ~s to say that the solubll~ty increases w~th d~mm~shmg temperature. It is clear that th~s type of temperature dependence ~s general for the normal phosphates of the polyvalent elements and ~s qmte pronounced m the lanthamdes. Th~s same type relatmn ~s also met in lanthanum phosphate [15].

o" 8,~

O,L

O,d

O,U

HaPO4 content of hquid phase,mole% F~gure 2 makes it clear that the specific g r av m es of aqueous H~PO~ solutions at 20 ~ taken from the hterature [16] are close m value to the specific gravtttes of solutmns obtained by dtssolvmg CePO4 m aqueous HsPO4 ( h q m d phase m the system CePO4H~PO4 - H~O) at 25 ~ . Th~s concordance ~s due to the fact that the solub~hty of CePO 4 in aqueous H~PO4 ~s low, reaching a m a x i m u m of 2.84 weight % m an 85% by weight H~PO4 solunon, Our own data and the data of the l~terature on the solubthty of the phosphates of certain t n v a l e n t elements m phosphoric amd (La, Co, Gd, A1 Ga. In Fe at 25 ~ and La Ce, AI, Ga, In, Cr, Fe at 70-80 ~ are presented m Figs. 3 , 4 , and 7 and m T a b l e 2 These data can be used as a bas~s for separating these elements through the use of H3PO4 or employed m the dissolving and extracting of the phosphates of these elements from HsPO4 solutron. Thus study of Figs. 4 and 7 shows that the solubthty of MPO4 (mole%) m HsPO4 solutions containing up to 0 45 mole % acld increases m the following order. > o 1) AIPO4> InPO4 > FcPO4 > OaPO 4 >GdPO 4 LaPO4> CePO4 at 25 , and 2) CrPO4> AlPO4> (InPO4)> GaPO4> FePO4> LaPO4 > CePO4, at 70-80* It zs to be noted that the order of increasing solubihty alters m general m passing from 25 to 7080 ~ being retained only m the case of LaPO4 and CePO4. InPO4 somewhat disrupts the sequence of MPO4 solubllmes. A comparison of Figs 4 and 7 shows the posslbihty of separating lanthanum, certain, g a d o h m u m , a l u m i n u m , r a d i u m , iron, galhum, and chrommm on the basis of phosphate solublhtles m phosphoric acid, The above m e n t m n e d data of the hterature can be drawn on m developing a scheme for the separation of these elements at any fixed H3PO4 coneentranon m excess of 0.45 m o l e % In additton, interest attaches to a study of the solubihty of venous mixtures of these and other phosphates m phosphorm amd. Fig 7 Solubthty ofMPO 4 m p h o s p h o n c acid (0 0 0 - 0 4 5 mole %) at 70-80 ~ 1)CePO 4,accordmg to the data of [12], 2) LaPO4, according to the data of[15], 3) FePO4 according to thedata of [20], 4) GaPO 4 according to the data of [21]; 5) InPO4 according to the data of [18], 6) AlPO4, according to the data of [17], 7) CrPO4, accordmg to the data of [11]

Expervnents on the Vaporlzatmn of Solutmns of CePO4 m Phosphoric A c _ id._.:. In certain experiments, a h q u o t portmns of the hqmd phase were evaporated by heating As a result of the e l e v a t m n of temperature, a c o l l o i d a l amorphous precipitate, d e a r l y of the composltmn CePO4 9 xHsPO4 9 aq, appeared throughout the liqmd phase and then passed back into solunon The H3PO4 content of the vaporization product was determined by t~tratmn w~th NaOH, and the CePO4 content measured by t~tratmg the Ce 4+ ~ons obtained by a m m o n i u m persulfate oxidation w t t h a solution of Moore's salt The data of Table 3 m d m a t e that the vzscous hqmd product was a solution of CePO 4 m 10007o HsPO4, or rather m 100~ HPO3, since H3PO4 dehydrates on heating, and undergoes the following changes: 2H3PO~-H_ig>

H4P~OT-__~_~ 2 H P O a .

Since the H+ d e t e r m m a t m n was by alkali titration with methyl orange as m&cator, ~t follows that only one H ~ was tltrated whether it be m H3PO4 or m HPO3. The solublhty of GePO4 in aqueous HsPO4 diminishes with increasing temperature, but zt is possible that CePOa dissolves m 100% HsPO4 since a precipitate of CePO4 . H3PO4 aq once more deposited out when water was added to the viscous liquid, This effect could also be explained by assuming

1067

TABLE 3. Data on the Analysis of the Liquid Phase after Evaporation Found I

specific~sPOr (HPO3), gravity ~ e l g h t %

CePO4' I E% [HsPO4(HPO3) weight % ] + CePO4] /

1.80 1.84

199.27 ~ 90.25

1.41 2.7q

/ 1

100.68 102.02

dissolution of CePO 4 m 100% H3PO4 to be due to the formau o n of HPO 3, since this is known to be a good p o l y m e r i c solvent for many materials The viscous liquid obtained through vaporlzauon deliquesced on standing in the air at room temperature and a floceulent precipitate c a m e down. Further heating gave rise to continued precipitation of a grayish-white mass which would not pass back into solution and could be removed from the walls of the vessel only with difficulty

SUMMARY 1. Cerium compounds, more spemftcally, acid phosphates are not formed in the system CePO4 - HsPO4- HaO at 25*; there is a gradual dissolution of CePO 4 with formation of a precipitate which involves CePO 4, I4~PO4, and HzO in varying proporuons. 2. The solubility of CePO4 in aqueous H3PO4 increases with diminishing temperature It is clear that the dlssolution of the cerium phosphate proceeds according to the reaction. CePO4 + H3PO4 " ~q ~ CePO4 H3PO4 aq (solution), 3. The d~fference in the solubIlltles of the phosphates of lanthanum, cerium (Ill). gadohnium, aluminum, gallmm, mdmm, iron (III). and chromiam (Ill) in phosphoric acid at 25~ and at 70-80~ can be made the basls of an H3PO4 separation of these'elements or employed m the extraction of their phosphates from aqueous HsPO4. LITERATURE CITED 1. 2 3. 4 5. 6. 7. 8

9 10. 11. 12. 13. 14. 15. 16 17 18. 19. 20. 21.

1068

A I. Ul'yanov, Collection: The Rare Earth Elements[m Russian], Izd. AN SSSR, Moscow (1963). A I Ul'yanov, T. I Kazakova, and E. Ya. Rumyantseva,Izv., AN SSSR, Otd.khimii.n. (1962), 1910 V V Serebrenmkov,The Chcmlstry of the Rare Earth Elements [m Russian], Izd. Tomskogo un-ta, Tomsk (1959), 1, p. 379. A I Ul'yanov and T. I Kazakova, Izv. AN SSSR, Otd khim11, n. (1963), 393. A l Ul'yanov, Izv. AN SSSR, Otd. khlmn, n. (1961), 1709. A K. Babko and O. M. Eremenko, Zh. analit, khtmii. I~3, No. 2, 209 (1958) Collection: The Analysisof Crude Minerals [m Russian], Goskhlmlzdat, Leningrad(1956), p. 726 S . N . Rozanov, The C h e m i c a l Analysis of Fertilizers [ln Russian], No 1, Goskhimlzdat, Moscow-Leningrad (1933), p.38. I A Atanasiu and M. Babor, Bull. sect. set. acad. roumaine, 20, No, 1-3, 32-34 (1938) R C. L. Mooney, Acta crystallograph , a , 5, 38q (1950) I p. Redfern and I. E. Salmon, J. Chem. Sac., 1961, 291. V N. Sveshnikova and V L Gmzburg, Zh. neorgan, khlmli, 7, 5, 1169 (1962). A N Z e h k m a n , The Metallurgy of the Rare Earth Metals, Thormm and Uranium [m Russian], GONTI, Moscow (1961), p. 76 S, K Voskresenskti, The Production of Phosphoric Amd by the Decomposmon of Apatlte with Sulfurm Acid [m Russian], GOSINTI Moscow (1961), pp. 4-5 N . A . Vasdenko and M L. Chepelevetskli, Zh neorgan k h t m n , 2, No 10, 2486 (1957). Brief Handbook of Chemistry, c o m p l i e d by V I. P e r e l ' m a n [in Russian], 3rd e d m o n , GONTI Moscow (1954), p. 361. I C. Brosheer, F. A. Lenfesty, and I. F. Andersor}, J Amer. Chem. Sac , 76, 5951 (1954). C. E A Brownlow, I. E Salmon, and I. G. L Wall, J. Chem. Sac., 1960, 2452. R . F . Yameson, and I. E. Salmon, J. Chem. Sac., 1954, 28. E B Brutskus. Tr. NIUIF, No. 137,110 (1937). N N. Chudinova, Zh. neorgan, khlmii. 7, 2285 (1962).