Phase equilibria in the CsUO system in the temperature range from 873 to 1273 K

1. inorg, nacL Chem. Vol. 40, pp. 1375-1381 © Pergamon Press Ltd., 1978. Printed in Great Britain

0022-190217810701-13751502.00t0

PHASE EQUILIBRIA IN THE Cs-U-O SYSTEM IN THE TEMPERATURE RANGE FROM 873 TO 1273 Kt D. C. FEE and C. E. JOHNSON Chemical Engineering Division, Argonne National Laboratory, 9700 Sourth Cass Avenue, Argonne, IL 60439, U.S.A.

(First received 15 September 1977; in revisedform 7 December 1977) Abstract--Portions of the cesium-uranium-oxygen system have been investigated between 873 and 1273 K and a phase diagram has been constructed using our data and the data of other workers in the field. A consistent set of measured and estimated thermodynamic data for cesium uranates has been used to calculate the equilibrium cesium partial pressure and the equilibrium oxygen partial pressure over two and three phase regions in the Cs-U-O system. For a given temperature, the equilibrium cesium partial pressure in a two phase region decreases as the equilibrium oxygen partial pressure increases. INTRODUCTION Recent studies of the cesium-uranium-oxygen in air have led to the determination of distinct X-ray diffraction patterns for ten cesium uranates [1--6]. However, much of the literature data on cesium uranates are conflicting, and no comprehensive phase diagram of the Cs-U-O system

cesium and the loading of capillaries for X-ray diffraction analysis were performed in a helium atmosphere. In the experiments with liquid cesium nickel capsules (7.9 mm ID, and 7 cm long) were loaded with approx. 2.5 g of powdered uranium oxide and 1.5 g of cesium and welded closed. During heat treatment at temperature over the range from 873 to 1273 K, the capsule was positioned to insure that the liquid cesium was in has been published. The conflict in the published data contact with the uranium oxide. At the end of the heat treatment, centers around the stability of the compounds formed the capsule was quickly quenched (15 s) and the excess cesium and the conditions required for their formation. In- removed by vacuum distillation at 573 K. formation on the cesium uranates is important because To study the reaction of gaseous cesium with uranium oxide, a of the considerable likelihood that one or more of these Type 304 stainless steel capsule (2.5 mm ID and 18 cm long) was compounds may form in nuclear fuel elements during loaded with a nickel tray containing 1.5 g of powdered uranium fiission. Therefore, phase studies of the Cs-U-O system oxide and a separate nickel tray containing 5 g of cesium and were undertaken in an effort of resolve some of the welded closed. During heat treatment, the capsule was positioned existing conflict regarding compound stability and condi- so that only gaseous cesium contacted the uranium oxide. The gaseous cesium pressure was calculated[7] from the temperature tions of formation and to establish a thermodynamically (600K) at the coolest portion of the capsule. However, the tray consistent phase diagram for the C s - U - O system, containing uranium oxide was in an isothermal zone at 1073-+ 10K. EXPERIMENTAL The experimental conditions and the results of our experiAll studies were carried out under isothermal conditions and in ments are summarized in Table 1. The normal cesium uranate, metal capsules. Hyperstoichiometric urania (OIU atom ratio--- Cs:UO4,and stoichiometric uranium dioxide were identified by 2.0-2.2) was exposed to either liquid or gaseous cesium and the X-ray diffraction analysis in all cases. (The X-ray data are given products were identified by X-ray diffraction analysis. The in Ref. [8]). product in these experiments was an orange powder that appeared to be homogeneous on visual inspection and X-ray DISCUSSION diffraction analysis. All operations that involved the handling of Comparison of experimental results tWork performed under auspices of the U.S. Energy Research and Development Administration.

Our results clearly show that a hexavalent cesium uranate, CsUO4, is the Cs-U--O phase in equilibrium with liquid cesium and stoichiometric urania in the

Table 1. Summary of experimental results for reaction of cesium with uranium oxide Expt. No.

Type of Experiment

Cesium Pressure, Pa

Uranium Oxide Temp., K

Initial O/U Atom Ratio of Uranium Oxide

Time at Temp., hr

1

Ni capsule

a

873

2.004

161

2

Ni capsule

a

873

2.068

150

3

Ni capsule

a

1073

2.20

280

4

Ni capsule

a

1273

2.20

280

5

SS capsule

5 x 102

1073

2.1

393

aUranium ox~de sample in contact with liquid uesium.

1375

Product Phases Identified by X-ray Diffraction Cs2UO4 + stolchiometrlc uranium dioxide

1376

D.C. FEE and C. E. JOHNSON

temperature range from 873 to 1273 K. This conflicts with the results of Aitken et al.[9-1 I] and Adamson[12] who reported that Cs-U--O compounds with a net uranium valence of less than 6, Cs, 3UO3 or Cso.5+~UO3 (x = 0-0.5), existed in equilibrium with liquid cesium and urania in the temperature range from 873 to 1073 K. However, it should be noted that neither Efremova et al.[13], Rudorff et ai.[14, 15], Cordfunke et al.[17], nor •van Egmond[3--6] found the pentavalent CsUO3 form in the Cs-U-O system. This is in contrast to Li, Na, K and Rb which do form pentavalent compounds of the type MUO3[14-18]. Moreover, our results clearly show that hexavalent cesium uranate, Cs2UO4, can exist in equilibrium with gaseous cesium and urania at 1073 K. Under similar conditions, hexavalent uranates were also found by Aubert et al. (Cs2U207 at 973-1273 K)[19]; Venker et al. (Cs2UO4/Cs2U207 at 973K)[20]; Aitken et al. (Cs.,UO4 at 1073 K)[9] and Adamson et al. (Cs_,UO4 at 1123K)[21]. However, our result conflicts with the finding of Cordfunke[2] that only Cs,_U40,2 (with a net uranium valence of less khan 6) exists in equilibrium with gaseous cesium and urania above 873K. Estimation of thermodynamic properties of cesium uranates To serve as a basis for the Cs-U-O phase diagram, thermodynamic functions for the cesium uranates that were characterized by Cordfunke et al.[1] and van Egrnond[3--6] were estimated in a self-consistent manner from the measured values of the enthalpies of formation of Cs,_UO4, Cs2UzOT, Cs~O and UO~, and the measured entropy of Cs2UO4, Cs20 and UO3. The various cesium uranates that have been characterized are listed in Table 2 along with their estimated thermodynamic properties. Also given in this table are enthalpy and entropy data for the reactants of the compounds listed therein. A graph was constructed (see Fig. I) by plotting the free energy of formation of the cesium uranates from Cs_,O and UO3 per mole of oxides (i.e. moles of Cs_,O plus moles UO3) vs the mole fraction of Cs.,O in the compound. The graph was drawn starting with the experimental data for the

compounds Cs2UO4 and Cs2U207, which may be written as Cs20.UO3 and Cs20.2UO3, respectively. The value of the free energy for the next compound, Cs4U50,7, was chosen so that it was about 3.2 kJ less negative than that obtained by a linear extrapolation using the values for Cs2UO4 and Cs2U207. This choice, although arbitrary in magnitude, makes the compound Cs2U207 stable with respect to decomposition into Cs,.UO4 and Cs4U.~O,~: that is, the reaction 3Cs2U207 = Cs2UO4 + CsaU_sO~7 has a positive value for the standard free-energy change. In a similar way, free-energy values were estimated for the next compound, Cs2U40,3, and for all other reported polyuranates formed by the combination of Cs20, and UO3. A similar procedure was used to estimate the enthalpies of formation. Note that the AGe scale in Fig. I is set 20kJ higher than the AHt scale. For Cs2UO4, the AH~ value in Fig. 1 is actually only 3.6kJ more negative than the AG/ value. From the estimated values for the free energies and enthalpies of formation, the entropies of formation were computed. From the entropy of formation, the value of the absolute entropy for each compound was computed. The thermodynamic quantities for the compound Cs2U40,2, which may also be written as Cs20.UO_,.3UO3, were estimated from the values for Cs2U40,3 by assuming that (I) the enthalpies of formation differ by the difference between the enthalpies of formation of UO3 and UO_, and (2) that the absolute entropy was less by the difference in entropies between UO3 and UO2. The values for Cs2UtOt8 and Cs2U9027 were estimated in a similar manner from the thermodynamic quantities for Cs2U~Oi9 and Cs2U9028, respectively[8,22]. Calculation of phase equilibria The experimental results and the thermodynamic data on cesium uranates obtained in this laboratory and the experimental data of Venker et al.[20], Aubert et al.[19], Cordfunke et al. [I,2], van Egmond [3-6], Aitken et al.[9], and Adamson et al. [21] were used to construct the phase diagrams shown in Figs. 2 and 3. Specifically, our results

Table 2. Thermodynamicproperties of cesium uranatesa --

Compound

•

~Hf,298, kJ/~ol

b

o

-ASf,298, J/mol'deg

$298 J/mol'deg

Cs2Uo~

1922±1.2 [23,24]

411e'd

CB2U207

3205 ± 2.1 [24,26]

656

220±0.4 [25]

CskU5017

7670

1560

777

CS2U;012

5570

1090

514

Cs2U~013

5710

1170

536

Cs2U5016

6950

1430

636

Cs2U6OI8

8080

1630

689

C8207022 CS2Ug027

9430 11810

1940 2400

834 987

CS2U150~6

19280

4010

1630

332

aEstlmated values tmless otherwlse noted. bEnth~pies of formation AH~.~8 (kJ/reel): Cs20(c), -346.0 ± 1.2.

(gef. 27); UO2(c), -1085±0.8 (Ref. 28); T-UO3(c), -1228±4 (gel. 29); 0409(c), -4510 (gef. 30); 0308(c), -3574 (Ref. 30). Cstandard entropies, S~98, (J mo1-1 deE-l): Cs20(c), 147 ± 0.4 (Eel. 31); 002(c), 77.0±0.02 (Ref. 32); 7-U03(c), 98.7±0.4 (Ref. 30); U(c), 50.3±0.2 (Ref. 33); 02(g), 205.0±0.04 (Raf. 33); Co(c), 85.1±0.4 (Ref. 33); U~O9, 336±0.4 (gel. 30); 0308, 282±0.4 (Ref, 30). dAddtttonal data:

AGf(Cs,g) - 71,000 - 75.3 T J/mol (Ref. 7).

Phase equilibria in the Cs-U-O system I

I

I

1377

I

0 cs~U,sO.

cs2u,o22

N _

40

o

_o F

4-E40- A H f / , ~

o+

%

=

80 -~

~ E

~ 80

,20g o

~20 _

cs~uso, cs~us°'

<~ :3=

~T

60 <~

i601 c,~o,___.

0 Fig.

I

0.I

l

I

02 0.3 MOLEFRACTO I NCs20

I

0.4

"~5

I. Graphical method of estimation of free energies and enthalpies of formation of cesium uranates. 0

0

Cs#zOz2 CszU~O~\,UO3 oo,..

c,~,o,,ss~/-'/

cs~o,,~

~ Cs

u

Fig. 2. Selected portion of the cesium-uranium-oxygen phase diagram showing tie lines to UO2+x; isothermal sections from 873 to 1273 K. (The solid area shows the extent of the two-phase region at 873 K. The Cs(~')-UO2+x and Cs(£)-Cs2UO4 tie lines exist [7] only below 950 K.).

and the data of Venker et aL [20], Aitken et al.[9], and Adamson et aL[21] were used to construct the Cs2UO,rUO2+x tie line. The data of Aubcrt et al.[19] was used to construct the Cs2U207-UO2+x tie line. The Cs2U,O,~UO2+x tie line and remaining tie lines in Fig. 3 were based on the results of Cordfunke et al.[l,2] and van Egmond[3-6]. Figures 2 and 3 represent isothermal sections over the temperature range from 873 to 1273 K. The Cs(1)-UO2+~ and Cs(1)-CsUO4 tie lines exist[7] only below 950 K. The Cs2UO~-UO:+~, Cs2U207-UO2+~ and Cs2U~O,2-UO2+~ tie lines exist over the range from 873 to 1273K. The widths of these two-phase region~ vary with temperature. Phase regions containing cesium uranates with a higher U/Cs atom ratio than that of Cs2U,O,2 may not exist over the entire temperature range from 873 to J I N C Vol. 40 No. 7 - 4 3

~+'

Cs

Fig. 3. Selected portion of the cesium-uranium-oxygen phase diagram; isothermal sections from 873 to 1273 K. (The solid areas show the width of two-phase regions at 1273 K. The compound UO3 exists[29] only below 930K. The Cs(~')--UO2+, and Cs(~')Cs2UO4 tie lines exist[7] only below 950 K. Phase regions containing cesium uranates with a higher U/Cs atom ratio than Cs2U4Ot2 may not exist over the entire temperature range, see text. SS = CSEU40~3-Cs2UsOt~ solid solution [ 1,4].)

1273K. Uranium trioxide, UO3, exists only below 930 K[29]. At temperatures above 873 K, Cs2U40,3 and Cs2UsO,~form a solid solution[I,4]. There are two key features related to Figs. 2 and 3. Firstly, the experimental data of Cordfunke et aL[l] clearly establish a Cs2U40~z-U3Os tie line which is consistent with the thermodynamic properties shown in Table 2. The existence of a Cs2U,O~e-U30~ tie line indicates that there can be no tie line between UO2+x and a cesium uranate with a higher U/Cs atom ratio than that of Cs2U40,2. Secondly, Cs2UO4 is the compound which exists in equilibrium with liquid cesium and urania. Consequently, cesium analogs (which have not beobserved" to data) of pentavalent uranates such as NaUO3 [17,18] and bla3UO4[17,18] do not exist in equil-

1378

D.C. FEE and C. E. JOHNSON

ibrium with UO2+x. If either of these cesium analogs existed in the temperature range 873-1273 K, they would exist in equilibrium with liquid cesium and urania. By application of the phase rule, it can be shown that in the three-phase regions shown in Figs. 2 and 3 (e.g. Cs2UO4--Cs2U207-UO2+x), there is one degree of freedom; therefore, setting the temperature fixes the cesium partial pressure and the oxygen partial pressure of the system. On the other hand, along the two-phase tie lines (e.g. the Cs2UO-UO2+x tie line), there are two degrees of freedom; therefore, at a set temperature, the cesium partial pressure depends on the oxygen partial pressure of the system. The dependency of the cesium partial pressure on the oxygen partial pressure at Cs-UO compositions falling on the two-phase tie lines can be calculated from the measured thermodynamic properties of the phases present. The fixed cesium partial pressure and the fixed oxygen partial pressure in the three-phase regions can be calculated in a similar manner, For the case of the two-phase region, consider the equilibrium

2Cs(g)+UO2(c)+O:(g)~Cs2U04(c)

(l)

which exists along the Cs.,UO4-UO:+, tie line. The corresponding free energy relationship may be written as

i

I

follows: AG-~I(Cs2UO4,c) - 2AG~t(Cs,g)- 2RT In Pcs -AGp(UO2,v)-RT In po: = 0. (2) The cesium pressure and the oxygen potential (At~o, = RT In po2) are related by In Pcs = [AG~I(Cs2UO4,c)-2AG~I(Cs,g) -AG~I(UO2,c)-At~o:]/2RT.

Similar expressions can be derived to relate the cesium pressure and the oxygen potential on the Cs2U2OrUO2+x tie line and on the Cs2U40,2-UO2+x tie line. The calculated cesium pressures are shown in Fig. 4; these are based on the thermodynamic data in Table 2, with the assumption that A/-/~f and A ~ are independent of temperature. This approach is valid for Cs2UO4 on the basis of the known high-temperature thermodynamic properties of this compound[M]. The effect on the thermodynamic properties of the a -*fl phase transition [1] of Cs2U:O7 at 573 K and the a--,/3 and//--,3, phase transitions[2] of Cs2U40,2 at 898 and 968K, respectively, is assumed to be small and has been neglected. This assumption has been shown to be valid in other ternary oxide systems[7].

J

.,,~/

i

/

g,

J073K

/

.1o/

• o cs2uo,-~+, //

,/

•

/ /

/,

200

z~

Cs2U207-U02,

~

_

• o Cs2U40 :-u0z÷,

i

I

240 -

(3)

280 A~'~, kJ/mol

I

320

,,,

i

360

I

Fig. 4. Cesium equilibrium pressure over two-phase regions in Cs-U-O phase diagram. (Solid lines and solid symbols indicate the stable phases at each oxygen potential and the range of stability for each of the two-phase regions considered. Dashed lines and open symbols indicate the calculated cesium equilibrium pressure, even though the corresponding two-phase region does not exist at that oxygen potential.)

Phase equilibria in the Cs-U-O system Another assumption in the calculation of the cesium pressures shown in Fig. 4 is that the solubility of cesium in urania or in the cesium uranates does not significantly affect the thermodynamic properties shown in Table 2. This assumption appears to be valid. The measured[35] solubility of cesium in a (U,Pu)O2 matrix is less than 1 ppm. A similar solubility of cesium in UO2+x would not affect the equilibrium oxygen pressure over the UO2+~ even if the deviations from Raoult's law were large. The estimated thermodynamic values for Cs2U40,2 shown in Table 2 are used in this paper instead of values derived from Cordfunke's e.m.f, measurements[2]. The data of Venker et a/.[20], Aubert et a/.[19], Aitken et al.[9], Adamson et al.[21] and this work showing that Cs,UO4 exists in equilibrium with UO2+~ above 873 K demonstrate the inadequacy of Cordfunke's e.m.f, measurements[2] in obtaining the thermodynamic properties of Cs2U40,2. According to Cordfunke's e.m.f. data, with an oxygen potential that is more negative than -170 kJ/mol. Cs2U40,2 is the only possible cesium uranate in equilibrium with UO2+x in the temperature range of Cordfunke's experiments, 974-1024 K. In Fig. 4, consider the 1273 K values. At lower oxygen potentials (At~o2 more negative than -230kJ/mol), the Cs.,UO4-UO2+~ region has the lowest equilibrium cesium pressure and is thus the only stable two-phase region at these oxygen potentials. Solid lines and solid symbols in Fig. 4 indicate the stable two-phase regions at each oxygen potential. Dashed lines and open symbols indicate the calculated equilibrium cesium pressure even though the corresponding two-phase region does not exist at that oxygen potential. As the oxygen potential over the Cs2UO4-UO2+~ region becomes more positive, the cesium equilibrium pressure decreases. At a unique oxygen potential (-225 kJ/mol at 1273K), the cesium pressure over the Cs2UO4-UO2+~ region is the same as the cesium pressure over the Cs2U207--UO2+~ region [log PcJPa) = -0.5]. This point is the intersection of the lines in Fig. 4 labeled Cs2UO4-UO2+x and Cs2UzO7--UO2+x and represents the oxygen potential and cesium pressure of the Cs.~UO4-Cs2U2OT-UO_,+~ region at 1273 K. At oxygen potentials slightly more positive than - 225 kJ, the Cs.,U:OT-UO,÷~ region has the lowest cesium equil-

1379

ibrium pressure. The Cs2U207-UO2+~ region is thus the only stable two-phase region at these oxygen potentials. As the oxygen potential over the Cs.,U207--UO,.+x region becomes more positive, the cesium equilibrium pressure decreases. At a unique oxygen potential (-202 kJ/mol at 1273 K), the cesium pressure over the Cs2U2OT--UO2+~region is the same as the cesium pressure over the Cs2U40,2-UO2+x region [log P¢~ (Pa)= - 1.2]. This point is the intersection of the lines labeled Cs2U207-UO2+~ and Cs2U40,2-UO:+x and is the oxygen potential and cesium pressure of the Cs2U~OTCs2U40,.,-UO:+x region at 1273 K. At oxygen potentials more positive than -202kJ/mol, the Cs2U40,2-UO2+x region has the lowest cesium equilibrium pressure. The Cs2U4OJ2-UO2+~ region is thus the only stable twophase region at these oxygen potentials. The three stable two-phase regions in Fig. 4 (Cs.,UO4UO~.+~, Cs2U207-UO2+~, and Cs2U40,2-UO:+~) are shown as tie lines in the Cs-U-O phase diagrams in Figs. 2 and 3. The other tie lines in Fig. 3 can be constructed in a similar manner from the thermodynamic data in Table 2. The absence of possible tie lines can also be inferred from thermodynamic considerations. For example, no Cs4UsO,7-UO,.+~ tie line exists because thermodynamic calculations show that there is no stable Cs4U~O,vUO2+~ region (i.e. the calculated cesium equilibrium pressure over the hypothetical Cs4U~O,7--UO2+~ region is not the lowest cesium equilibrium pressure for any oxygen potential or temperature shown in Fig. 4). In a fashion similar to the foregoing, the oxygen potentials and cesium pressures over the three-phase regions shown in Figs. 2 and 3 may be obtained. For example, the oxygen potential of the Cs2UO4-Cs2U~,OTUO2+~ region may be calculated from the thermodynamic relationship associated with the equilibrium Cs2UO4(c) + UO,.(c) 1/202(g)~-Cs2U207(c).

+

The oxygen potentials and cesium pressures over selected three-phase regions shown in Figs. 2 and 3 are given in Table 3; these were derived using the data from Table

Table 3. Calculated oxygen potenti~s (kJ/mol) and cesium pressures (Pa) at selected temperatures in three-phase regions of the Cs-U-O phase diagram 873 K

Region

-1°g PCs

Cs(~) + Cs2UO 4 + UO2+x

(4)

1073 K

-ACO2

-log PCB

633

1273 K

-AGo2

-log Pcs

587

-AGO2 541

Cs2UO% + Cs2U207 + U02+x

5.9

280

2.7

251

0.5

225

Cs2U207 + Cs2U%012 + U 0 2 ~

6.7

~61

2.8

231

1.2

202

Cs2UhOl2 + U409 + U02+x (a)

9.4

217

5.6

187

3.4

146

Cs2U4012 + U~O 9 + U308 (a)

11.4

176

7.3

146

4.4

117

Cs2U4012 + Cs2U60]8 + U308

11.5

175

7.7

127

5.0

78

Cs2U6018 + Cs2U9027 + U308

12.4

147

8.0

112

5.0

77

Cs2U9027 + Cs2U7022 + U308

15.7

72b

9.4

74

5.1

76

Cs2UT022 + Cs2U|50%6 + U308

15.2

82b

10.5

47

7.3

12

aThe oxygen potential in this three-phase region is the oxygen potential of the U%Og-y-U02+x or U409-U308_ z two-phase region (Ref. 30), assu~ing n o effect from cesium solubility. The oxygen potential is independent of the estimated thermodynamic properties of Cs2U~OI2 ; the ceslum pressure depends on the estimated thermodynamic properties of Cs2U4012 in these three-phase regions. bNot a stable three-phase region at this temperature. The stable three-phase region at 873 K is Cs2UgO27-Cs2UIbOw6-U3Os, with AGO2 - -81 kJ/mol and log PCs " -15.3.

1380

D.C. FEE and C. E. JOHNSON

2. As shown in Table 3, the oxygen potential of the Cs2UO4--Cs2U2OT-UO2+x region at 1273K corresponds[36] to UO2.oa. Therefore, the thermodynamic relationship associated with the equilibrium 1-x Cs2UO4(c)+UO2+x(c)+--~-O2(g) : Cs2U2OT(C)

(5)

and the thermodynamic data[30] for UO2+~ (instead of UO2) should be used. However, the data in Table 3 have not been corrected for the deviations in stoichiometry of urania because these corrections are less than -+4 kJ/mol, which is within the uncertainty of the estimated thermodynamic values for cesium uranates, Accuracy of calculated phase equilibira in spite of the limitations in the thermodynamic data given in Tables 2 and 3, these data are valuable in systematizing a large number of experimental observations on the Cs-U-O system[I-6, 13-15, 19-21] in the form of the Cs-U-O phase diagram shown in Fig. 3. The only literature inconsistent with Fig. 3 are those of Aitken et al. [9-1 I] and Adamson et a/.[12] reporting the formation of a Cs-U-O compound with a net uranium valency of less than 6, Cst.3UO3 or Cso..,+~UO3 (x = 00.5) and those of Efremova et al.[13] and Spitsyn et a1.[38] reporting the formation of Cs-U-O compounds Cs2U30~o and Cs2U~O~9. The data in Table 3 are in accord with the phase diagram. At a given temperature, the oxygen potential becomes more positive and the cesium pressure decreases going from top to bottom in the table. Furthermore, the estimated thermodynamic data predict the observed tie lines [2] between UaOa and Cs2U4OI2 , Cs2U4OI2, Cs2U6OI8 , Cs2U9027 and CSEU7022 and the absence of a tie line between U308 and Cs2U~OI~ and between U3Oa and Cs2U40~3. However, the set of thermodynamic data that predicts the observed tie lines to U308 is not unique and a slightly altered set would result in the same ordering of the oxygen potentials and predictions of tie lines. Therefore, it appears that the estimated thermodynamic data in Table 2 are qualitatively correct. Quantitatively, the data for the C s t ) + Cs2UO4 + UO2+~ region have the least uncertainty because reliable high-temperature thermodynamic data are available for Cs2UO4134] and UO2137]. The data for the Cs2UO4-Cs_~U207-UO:+~ region were calculated using enthalpy measurements (at 298 K) and estimated entropy data for Cs2U207. An uncertainty of -+8 J/tool deg in the entropy estimate becomes an uncertainty of +8kJ/mol in the calculated oxygen potential of the Cs2U207-Cs2UO4-UO2+~ region. The uncertainty in the calculated oxygen potential of the three-phase regions which lie below the Cs2UO4-Cs2U207-UO:+~ region in Table 3 is greater than -+8 kJ/mol and increases going down the table owing to the increasing uncertainty in the extrapolation technique shown in Fig. I. Figure 3 indicates that all ten of the cesium uranates shown exist in the temperature range from 873 to 1273 K. The results of this study, as well as those of Aubert et a1.[35] and Cordfunke[1,2] show that the Cs2UO4UO2+x, Cs2U207UO2+x, and Cs2U40~2-UO2+~ regions exist in the stated temperature range. In Fig. 3, regions containing U3Os may not exist over the entire range of 873-1273 K. Cordfunke's pseudo-binary Cs-U-O phase diagram[l] at Poz = 2 x 104 Pa shows that the transitions

Cs2U~50,~-U3Os~Cs2UgO27-U3Os~Cs2U4Ot2-U3Os occur with increasing temperature. Cordfunke concludes that the Cs2U4Ot2-U308 region is the only stable twophase region containing U3Os at high temperatures. In contrast to Cordfunke's conclusion, we assert that all of these two-phase regions may exist at the highest temperature reached. As shown in Fig. 4, at Po2 = 2 X 10 -7 Pa, the stable two-phase regions are Cs2U4OI~UO2+~ at 873 K (-195 kJ/mol); Cs2U207-UO2÷x at 1073K ( - 240 kJ/mol); and Cs2UO4-UO2+x at 1273K (-285kJ/mol). Experimentally, at a fixed Po2 of 2x l0 -7 Pa, the transitions Cs2U40,2-UO2+x ~Cs2U2OTUO2+x~Cs2UO4-UO2+~ would be observed during a temperature increase from 873 to 1273 K. A similar ordering of the temperature and oxygen potential conditions for stable two-phase regions containing U3Os may account for Cordfunke's observations. Consequently, in Fig. 3, phase regions containing UaOs are shown to be unaffected by temperature in the range 873-1273 K because there are neither experimental thermodynamic data nor experimental phase data at a sufficient number of oxygen potentials to warrant our doing otherwise. Acknowledgements--The authors would like to acknowledge a series of fruitful discussions with Drs. O. G6tzmann, A. E. Martin,and J. Royal. The assistance of B. S. Tani in performing the X-ray analyses and L. O. Nippa in preparing the isothermal capsulesis also gratefully acknowledged. REFERENCES 1. E. H. P. Cordfunke, A. B. van Egmond and G. van Voorst, J. lnorg. Nucl. Chem. 37, 1433 0975). 2. E. H. P. Cordfunke, Thermodynamics of Nuclear Materials 1974, Syrup. Proc., Vienna, IAEA Vol. II, p. 185 (1975). 3. A. B. van Egmond, J. Inorg. Nucl. Chem. 37, 19290975). 4. A. B. van Egmond, J. Inorg. Nucl. Chem. 38, 1645 0976). 5. A. B. van Egmond, J. lnorg. Nucl. Chem. 38, 1649 (1976). 6. A. B. van Egmond, J. Inorg. Nucl. Chem. 38, 2105 0976). 7. JANAF Thermochemical Tables, NSRD-NBS-37. 8. D. C. Fee, I. Johnson, S. A. Davis, W. A. Shinn, G. E. Staahl and C. E. Johnson, ANL-76-126 (1977). 9. E. A. Aitken, M. G. Adamsonand S. K. Evans, GEAP-12489 (1974). 10. E. A. Aitken, M. G. Adamson, D. Dutina and S. K. Evans, Thermodynamics of Nuclear Materials 1974 Syrup. Proc., Vienna, IAEA VUl.I, p. 187 (1975). 11. E. A. Aitken, M. G. Adamson,D. Dutina, S. K. Evans and T. E. Ludlow, GEAP-12418 (1973). 12. M. G. Adamson, E. A. Aitken and D. W. Jeter, Int. Conf. on Liquid Metal Technology in Energy Production, (Edited by M.H. Cooper), p. 866. 2-6 May 1976, NTIS, Springfield, Virginia(1977). 13. K. M. Efremova, E. A. Ippolitova and Yu. P. Simanov, Investigati°ns in the Field °f Uranium Chemistry (Edited by V" I. Spitsyn) p. 59. Moscow University Press, Moscow (1961); English translation: ANL-Trans-33(1964). 14. w. Rudorff,S. Kemmler-Sackand H. Leutner, Angew. Chem. 74, 429 (1962). 15. S. Kemmler-Sack and W. Rudorff, Z. Anorg. Allg. Chem. 354, 255 0967). 16. A. B. van Egmond and E. H. P. Cordfunke, J. Inorg. Nucl.. Chem. 38, 2245 0976). 17. E. H. P. Cordfunke and B. O. Loopstra, J. Inorg. Nucl. Chem. 33, 2427 (1971). 18. J. E. Battles, W. A. Shinn and P. E. Blackburn, J. Chem. Thermodyn.4,425 (1972). 19. M. Aubert, D. Calais and R. LeBeuze,Z Nucl. Mater. 61,213 (1976). 20. H. Venker and K. Ehrlich, J. Nucl. Mater. 56, 115 (1975). 21. M. G. Adamson and E. A. Aitken, GEAP-12511 (1974). 22. I. Johnson, ANL-75-48, p. 18 (t976).

Phase equilibria in the Cs-U-O system 23. P. A. G. O'Hare and H. R. Hoekstra, J. Chem. Thermodyn. 6, 251 (1974). 24. E. H. P. Cordfunke, W. Ouweltjes and G. Prins, J. Chem. Thermodyn. 8, 241 (1976). 25. D. W. Osborne, P. A. Brletic, H. R. Hoekstra and H . E . Flotow, J. Chem. Thermodyn. 8, 361 (1976). 26. P. A. G. O'Hare and H. R. Hoekstra, J. Chem. Thermodyn. 7, 831 (1975). 27. J. L. Settle, G. K. Johnson and W. N. Hubbard, J. Chem. Thermodyn. 6, 263 (1974). 28. E. J. Huber and C. E. Holley, J. Chem. Thermodyn. 1,267 (1969). 29. E. H. P. Cordfunke and P. Aling, Trans. Faraday Soc. 61, 50 (1965). 30. M. H. Rand and O. Kubaschewski, The Thermochemical Properties o[ Uranium Compounds. Wiley, New York (1963). 31. H. E. Flotow and D. W. Osborne, J. Chem. Thermodyn. 6, 135 (1974).

1381

32. Thermodynamics and Transport Properties of Uranium

Dioxide and Related Phases, Panel Proceedings, Vienna, IAEA, p. 25 (1965). 33. R. Hultgren, P. D. Desai, D. T. Hawkins, M. Gleiser, K. K. Kelley and D. W. Wayman, Selected Values of the Thermodynamic Properties of the Elements, Americal Society for Metals, Metals Park, Ohio (1973). 34. D. R. Fredrickson and P. A. G. O'Hare, J. Chem. Thermodyn. 8, 353 (1976). 35. C. E. Johnson, A. L. Harkness and D. V. Steidl, ANL-8022, p. 41 (1973). 36. P. E. Blackburn, J. Nucl. Mater. 46, 244 (1973). 37. E. A. Fischer, P. R. Kinsman and R. W. Ohse, J. Nucl. Mater, 59, 125 (1976). 38. V. I. Spitsyn, E. A. Ippolitova, K. M. Efremova and Yu. P. Simanov, Investigations in the Field of Uranium Chemistry (Edited by V. I. Spitsyn), p. 148. Moscow University Press, Moscow (1961); English translation: ANL-Trans-33 (1964).