P R E P R I N T – ICPWS XV
Berlin, September 8–11, 2008
Solubility of Zirconium Dioxide at Elevated Temperatures
Liyan Qiu, Dave A. Guzonas and Don G. Webb
Atomic Energy of Canada Limited, Chalk River, Ontario, Canada K0J 1J0
Email: [email protected]
The heat transport purification system of CANDU nuclear reactors is used to remove particulate
and dissolved impurities from the heat transport coolant. Zirconium dioxide shows some
potential as a high temperature ion exchange medium for cationic and anionic impurities found
in the CANDU heat transport system (HTS). Zirconium in the reactor core can be neutron
activated, and potentially dissolved and transported to out-of-core locations in HTS. However,
the solubility of zirconium dioxide in high temperature aqueous solutions has rarely been
reported. This paper reports the solubility of zirconium dioxide in 10-4 M LiOH solution,
determined between 298 to 573 K using a static autoclave. Over this temperature range, the
measured solubility of zirconium oxide was between 1 and 10×10-8m, with a minimum solubility
around 473 K. This low solubility suggests that its use as a high temperature ion exchanger
would not introduce significant concentrations of contaminants to the system. A thermodynamic
analysis of the solubility data suggests that Zr(OH)4o likely is the dominant species over a wide
pH region at elevated temperatures. The calculated Gibbs energies of formation of Zr(OH)4o(aq)
and Zr(OH)4(am) at 298.15 K are -1472.6 kJ⋅mol-1 and -1514.2 kJ⋅mol-1, respectively.
Introduction
The CANDU ® heat transport purification
system is used to remove particulate and dissolved
impurities from the coolant. It would be
advantageous to operate the purification system at
temperatures close to that of the heat transport
system (573K) to minimize energy loss.
Unfortunately, conventional organic ion exchange
resins are not stable at elevated temperature and as
a result are usually used below 350 K. Zirconium
dioxide (or zirconia, ZrO2) and hydrous zirconium
oxide are possible candidates for high temperature
ion exchange materials due to their exceptional
resistance to heat and radiation. One of the
advantages of a ZrO2-based purification material
over other inorganic ion exchangers is that there are
no new impurities introduced to the coolant, since
ZrO2 is a corrosion/wear product of the zirconium
alloys used for the pressure tube and fuel cladding
in the heat transport system of CANDU reactors.
Solubility data are necessary to determine if a
candidate ion exchange material can be used for
high
temperature
purification.
Moreover,
®
CANDU, CANada Deuterium Uranium, is a registered
trademark of Atomic Energy of Canada Limited (AECL).
thermodynamic functions of dissolved species can
be calculated from measured solubility data.
Curti and Degueldre [1] reported the solubilities
of cubic and monoclinic ZrO2 in neutral water and
aqueous bicarbonate solutions at 298 K. In general,
both crystalline forms of ZrO2 showed almost
identical solubilities. However, the solubility of
ZrO2 (monoclinic) in neutral water, measured by
Curti and Degueldre [1], is one order of magnitude
lower than that reported by Kobayshi et al. [2].
The solubility data at 298 K measured by
Kobayshi et al. [2] show that zirconium dioxide has
minimum solubility between pH 2 and pH 13,
beyond which the solubility increases with
increasing acid and base concentrations [3].
Similar dissolution behaviour is also observed in
the solubility tests of hydrous zirconium oxide [4-6].
The solubility minimum between pH 2 and 13
indicates that the dominant species in equilibrium
with zirconium dioxide and hydrous zirconium
oxide is the neutral species, Zr(OH)4o, which is
independent of the pH of the solution. In strongly
acidic solutions, the solubility is controlled by the
formation of polynuclear hydrolysis species or less
hydrated zirconium species in solution and
decreases rapidly with increasing pH. In strongly
alkaline solutions the solubility increases with
type-K thermocouples, connected to temperature
controllers.
increasing pH due to the formation of the zirconate
ion ZO32- but not Zr(OH)5- [4]. These results are
contrary to those of Baes and Mesmer [7], Tromans
[8] and Adair et al. [9] especially in weak alkaline
solutions where the dominant species was
considered to be Zr(OH)5- and HZrO3-.
Although the solubility of ZrO2 has been
investigated by several groups at room temperature,
solubility data at high temperature and under
nuclear reactor water chemistry conditions are
rarely available. The solubility of zirconia was
measured by Adair et al. [9] at elevated
temperatures (298 to 473 K). However, their
measured solubilities were limited to the two pH
extremes, pH< 4 and pH>11, and did not include
the region pH 4 to 11 that is of most interest to the
nuclear industry. Moreover, there is very poor
agreement in the solubility data of zirconium
dioxide as well as in the thermodynamic functions
of the hydrolysis equilibria in aqueous solutions
[4,7]. In this paper, the results of solubility
measurements of zirconium dioxide in lithium
hydroxide solutions at elevated temperatures are
presented. The work was performed to determine if
ZrO2 would be suitable for use in a high
temperature purification system for the CANDU
HTS.
V-11
V-15
V-12
V-17
Sample bomb
V-7
V-10
V-6
V-9
V-5
V-8
V-2
V-1
To drain
V-18
TC3
Ar
Master
Cylinder
Ar
Backup
Cylinder
Condenser
P
TC1 TC2
Vent line
V-13
V-3
V-16
RD
V-14
Drain
Glove bag
Autoclave
TC -thermocouple
P -pressure gauge
RD -rupture disk
T -Autoclave Engineers tee
-temporary connection
Furnace
Sample line
Thermowell
Cartridge heater
TC 4
V-4
Figure 1: Schematic diagram of the 200 mL
autoclave used for the solubility
measurements
The autoclave was cleaned and rinsed with
deionized water (distilled water with additional
treatment through a Millipore "Milli-Q50" ultra
pure, ion exchange water system) and methanol
before using. A zirconium oxide sample with a
mass of 64.63 g was added to the autoclave with
1100 mL of deionized water. Sufficient lithium
hydroxide solution was added to the water to give
an initial pH of 10.48 at 298 K. The initial pH of
the solution was higher than the value of pH 9.74
used in CANDU HTS to compensate the pH
decrease as the dissolution of ZrO2 reaches
equilibrium. 1 No pH adjustment was carried out
during the experiment. The autoclave was purged
with argon for 45 minutes prior to heating. The
system was then heated to the target temperature
and allowed to stabilize for several hundred hours
before the first 10 mL sample was taken, using a
sampling line on the top of the autoclave. Samples
were obtained through a condenser line directly into
a syringe; and then were passed through a 0.22 μm
filter and into a plastic sample bottle. After the
samples were taken at 3 to 5 different time intervals
and the dissolution equilibrium was reached, the
temperature of the remaining solution in the
autoclave was increased to the next target value.
Experimental
Anhydrous zirconium oxide (powder form) was
purchased from Fisher Scientific Company (the
grade of the sample was not indicated) and used
without further purification. Powder X-ray
diffraction was used to examine the structure and
identify impurity phases. The purity of the sample
was determined by Inductively Coupled PlasmaMass Spectroscopy (ICP-MS).
The solubility of zirconium oxide at 298 K was
investigated in a 2000 mL glass flask, which was
flushed with argon prior to adding the LiOH
solution and zirconium oxide. To prevent carbon
dioxide uptake by the solution the flask was sealed
tightly in an argon atmosphere. A LiOH solution
was used to adjust the solution pH to around 10.0 in
the argon atmosphere after each sampling.
The solubility tests at or above 373 K were
performed in a static Hastelloy C autoclave (Figure
1) with a nominal volume of 2000 mL. The
autoclave was equipped with a sample line which
was used both for solvent injection through a
sample bomb and for sampling the solution through
the water-cooled condenser. The temperature of the
solution in the autoclave was determined using
1
In CANDU stations, it is the pHa instead of pH that is measured
at 298 K. pHa, apparent pH, is defined to be the pH of a D2O
solution measured with a pH meter calibrated in H2O buffers at
298 K. The relation between pHa and pH is pHa = pH + 0.46.
2
concentration of dissolved zirconium species first
increased and then decreased before reaching
equilibrium. Dissolution equilibrium appears to be
reached faster at 298 K than at 373 K. ZrO2
dissolved very rapidly at 373 K in the first hundred
hours and then most of the dissolved zirconium
species (~70%) precipitated from solution to reach
dissolution equilibrium (Figure 3). A maximum in
the concentration of dissolved zirconium species
was observed in the first 400 hours. A similar
phenomenon was also observed by Curti and
Degueldre [1] for both monoclinic and cubic ZrO2
dissolution in water. They deduced that such
dissolution behaviour was probably related to the
reordering of the surface structure from a soluble,
amorphous structure to a less soluble crystalline
form.
The solubility was determined at temperatures from
373 to 573 K in 25 to 50 K increments.
The pH of each sampled solution was measured
in an argon atmosphere at 298 K within 5 minutes
of the sampling time, using pH electrodes calibrated
with Oakton pH buffers, pH 7 (lot # 2303123) and
pH 10 (lot # 2303147). Samples were acidified and
stored typically for 1 to 3 days before elemental
analysis. The concentration of dissolved zirconium
species in water was determined by ICP-MS with a
precision of 10% and a detection limit of 0.02 ppb.
Results and Discussions
In the following discussions, the solubility of
zirconium dioxide refers to the total concentration
of all dissolved zirconium species at dissolution
equilibrium at the specified temperature.
The powder X-ray diffraction pattern of the
zirconium oxide used in this work at 298 K is
shown in Figure 2. The data indicate that the
zirconium oxide sample had a monoclinic crystal
structure and is consistent with the pattern reported
by Gualtieri et al. [10]. No other phases were
detected in the powder pattern.
ICP-MS analysis showed that the commercial
ZrO2 sample had a purity of 96% with a 4%
impurity of HfO2. The small amount of hafnium in
the sample was not expected to significantly affect
the solubility of zirconium oxide in water. Hence,
no further purification of the commercial sample
was carried out.
100
[Zr] (10-8 m )
80
298 K
0
0
Counts
600
400
200
0
50
60
600
800
1000
1200
Under the experimental conditions used in these
tests, it was found that the lithium concentration in
the solution changed with dissolution time. The
concentration of lithium in the solution showed a
gradual decrease from 373 to 523 K and about 50%
of the lithium initially present was removed from
the solution (Figure 4). The drop in lithium
concentration (more than 30%) was particularly
significant between 423 and 523 K. Since lithium
hydroxide is highly soluble in water (50.7 g Li/kg
H2O at 373 K) [11] and the dissolved lithium
concentration was much lower than the solubility of
lithium hydroxide, the dissolved lithium species in
solution were likely removed by zirconium dioxide
through ion exchange. Zirconium dioxide has a
high surface area and shows strong ion exchange
ability below and above its point of zero charge
(PZC=6.7) [12-13]. Experimental data show that
the cation exchange ability of zirconium oxide
increases with increasing pH and temperature [14].
The ion exchange ability of zirconium oxide results
800
40
400
Figure 3: Dissolution kinetics of ZrO2 in LiOH
solutions with initial pH values of 10.0
and 10.48 at 298 and 373 K, respectively.
1000
30
200
Time (hours)
1200
20
373 K
40
20
1400
10
60
70
2θ
Figure 2: Powder X-ray diffraction of the
commercial ZrO2 sample at 298 K.
The dissolution rates of ZrO2 at 298 and 373 K
are shown in Figure 3. The dissolved concentrations
of ZrO2 measured at different times reveal
interesting dissolution kinetics of ZrO2 in alkaline
solution.
The dissolution data at both high and low
temperatures show a similar behaviour. The
3
gradually decreases from 5.5×10-8 m at 298 K to a
minimum value of 0.91×10-8 m at 523 K (Figure 5).
A very low solubility is a major requirement if
ZrO2 is to be used as an ion-exchanger in a high
temperature purification system, so that the system
does not introduce a significant amount of
dissolved zirconium species into coolant. In general,
the solubility of zirconium dioxide very weakly
depends on the temperature between 373 and 523 K.
Above 523 K, the solubility increases with
increasing temperature. Similar dissolution
behaviour is also observed in other sparingly
soluble metal oxides such as magnetite [18] and
nickel oxide [19] in alkaline solutions.
Although there are no experimental solubility
data available in the literature to compare under
similar test conditions, an interesting comparison to
note is that the measured solubility of ZrO2 is close
to that of magnetite 2 , suggesting that dissolved
zirconium species may be important for the
transport of 95Zr in CANDU HTS. The possibility
that some 95Zr is transported as a dissolved species
rather than as particles has been an unresolved issue
in CANDU activity transport.
from its hydrolysis and the formation of Zr(OH)4
on the surface of in solid phase:
ZrO2(s) + 2H2O = Zr(OH)4(s)
(2)
From simple standpoint, in alkaline solution,
the protons of Zr(OH)4(s) can be exchanged with
cations such as Li+; in acidic solutions, the OHgroups can dissociate to form Zr(OH)3+ and
exchange with anions [15]. It is expected that in
alkaline solutions, as Li+ exchanges with protons of
Zr(OH)4(s):
Zr(OH)4(s) + Li+ = Zr(OH)3OLi(s)+ H+ (3)
the pH of the solution should decrease, in
agreement with the experimental observations
(Figure 4). The hydrolysis of other dissolved
zirconium species may contribute to the decrease in
pH [16] but this contribution is minor considering
the very low dissolved zirconium concentration
relative to the lithium concentration change under
this test condition. It appears that zirconium dioxide
shows a maximum ion exchange ability for lithium
around 523 K. The detailed ion exchange
mechanism needs to be further investigated.
14
12
11
11.0
[Zr] (10-8m )
10
10
10.5
9
8
9.5
7
pH(298 K)
[Li] (10-4 m )
10.0
2
0
250
6
4
250
350
400
450
500
550
300
350
400
450
500
550
600
Temperature (K)
8.5
300
6
4
9.0
5
8
Figure 5: Solubility of zirconium dioxide in
LiOH solutions as a function of
temperature (♦, pH 10.0; • pH 10.45).
8.0
600
Temperature (K)
Figure 4: Lithium concentration (♦) and pH (•)
at equilibrium as a function of
temperature.
Gibbs energies of formation for the dissolved
zirconium species are rarely reported in the
literature because of the lack of reliable solubility
data. These data are necessary to predict the
concentrations of various dissolved species and the
solubility of zirconium dioxide in aqueous solutions
as a function of pH and temperature in case
experimental results are not available. Based on the
measured solubility data [2,4,6] for the dissolution
reactions of zirconium dioxide in weak acidic and
alkaline solutions (pH values from 2 to 13), the
The solubility data of zirconium dioxide
between 298 and 573 K measured in this work are
shown in Figure 5. At 298 K, the average value of
the zirconium dioxide solubility is (5.5±2.1)×10-8 m
at pH 10. The result agrees with that measured by
Kobayshi et al. [2] but is higher than that reported
by Curti and Degueldre [1] and Sunder and Wren
[17]. It is interesting to note that the solubility of
ZrO2 is almost same as that of Zr(OH)4 [4,6],
indicating ZrO2(s) dissolved as shown in Equation
(2) under these test conditions. From 298 to 573 K,
the overall dissolved zirconium concentration
2
Based on Sweeton and Baes’ data and model [16] the overall
concentrations of dissolved iron species in equilibrium with
magnetite is 1.2×10-7m at pH 10 and 573 K.
4
neutral species Zr(OH)4o is the dominant dissolved
species of zirconium dioxide and hydrous
zirconium oxide:
ZrO2(s) +2H2O (l) = Zr(OH)4o(aq)
over a wide range of pH at elevated temperatures
[21]. However, more work is needed to verify this
conclusion so that reliable thermodynamic
functions can be calculated for dissolved zirconium
species from solubility data.
(4)
Therefore, the equilibrium constant of reaction
(4) is equal to the measured solubility of zirconium
dioxide assuming the activities of ZrO2 (s) and H2O
are equal to unity:
Kh = [Zr(OH)4o],
Summary and Conclusions
The solubility of zirconium dioxide in 10-4 m
LiOH solutions was determined from 298 and 573K.
Zirconium dioxide was found to have a minimum
of solubility of 0.91×10-8 m, at 523 K and the
solubility in the temperature region 373 to 525K
very weakly depends on the temperature. The
solubility of zirconium dioxide at high temperature
is close to that of magnetite under similar
conditions, suggesting that dissolved zirconium
species be important for the transport of 95Zr in
CANDU HTS.
The value of the pH and the concentration of
dissolved lithium ion decreased with experimental
time due to ion exchange between lithium ion and
hydrated zirconium oxide, Zr(OH)4.
The low measured solubility of zirconium
dioxide and its apparent ion exchange capability
show that zirconium dioxide and related materials
such as hydrous zirconium oxide are promising
candidates for use as ion exchanger materials for a
high temperature purification system of the
CANDU HTS. However, the ability of ZrO2 to
exchange Li+ ions from the coolant means that a
strategy for pHa control will need to be developed if
this material is to be used in a CANDU HTS
purification system.
Based on the measured solubility data,
Zr(OH)4o(aq) is the dominant dissolved species in
weak acidic and alkaline solutions (pH from 2 to13)
at room temperature. The Gibbs energies of
formation of Zr(OH)4o(aq) and Zr(OH)4(am)
calculated from the measured solubility are 1472.6
kJ⋅mol-1 and 1514 kJ⋅mol-1 at 298.15 K,
respectively. Zr(OH)4o(aq) likely is the dominant
species at higher temperatures over a wide region of
pH but more work is required to verify this
conclusion so that the reliable thermodynamic data
of dissolved zirconium species can be determined.
(5)
and has a value of (5.5±2.1)×10-8 m at 298 K, where
[Zr(OH)4o] is the concentration of dissolved
Zr(OH)4o in water with units of mol⋅kg-1. Using the
Gibbs energy of formation of ZrO2 (monoclinic)
and H2O (l) [20], the Gibbs energy of formation of
Zr(OH)4o(aq) is estimated to be 1472.6±2.1 kJ⋅mol-1,
from which the Gibbs energy of formation of
Zr(OH)4(am) can be calculated using the solubility
data [4], and has the value of –1514.2 kJ⋅mol-1. The
reliability of these values can be verified from the
Gibbs energies of formation of their components
(ZrO2 and H2O), assuming the additivity of the
Gibbs energy, an assumption often used by
geochemists. For example, the calculated Gibbs
energy of formation of Zr(OH)4(am) from ZrO2(cr)
and H2O(l) is 1514.0 kJ⋅mol-1, which agrees with
that determined from solubility measurements. It
should be noted that the variability in degree of
amorphism may result in variation in the Gibbs
energy of formation [8].
Tromans [8], Adair et al. [9] and Aja et al. [21]
calculated the solubility of zirconium dioxide at
high temperatures using limited solubility data at
extreme pH values and empirical thermodynamic
parameters. Their results show that the solubility
behaviour and the dominant dissolved zirconium
species as a function of pH at elevated temperature
are similar to those at room temperature. However,
the calculated zirconium solubility is either
underpredicted [8] or overpredicted [21] by a few
orders of magnitude compared to the measured
solubility at both room temperature and high
temperatures. The interpolated solubilty from Adair
et al. [9] is close to our measured data at high
temperatures but is a few orders of magnitude less
than the measured solubility at 298 K. One of the
major difficulties with these calculations is that,
becaue of the limited amount of data available, the
Gibbs energy of formation of Zr(OH)4o species is
either overestimated or underestimated. It appears
that Zr(OH)4o is also probably the dominant species
Literature
[1] C. Curti and C. Degueldre: Solubility and
hydrolysis of Zr oxides: A review and
supplemental data. Radiochim. Acta, 90, 801804 (2002).
[2] T. Kobayashi, T. Sasaki, I. Takagi and H.
Moriyama: Solubility of zirconium (IV)
5
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16] P.L. Brown, E. Curti, B. Grambow and, C.
Ekberg: Chemical Thermodynamics of
Zirconium. Elsevier, the Netherlands, 2005.
[17] S. Sunder and S. Wren: Aqueous solubility of
zirconia, presented at the 29th Canadian
Nuclear Society meeting, June 1-4, 2008.
[18] F.H. Sweeton, C.F. Baes: The solubility of
magnetite and hydrolysis of ferrous iron in
aqueous solutions at elevated temperatures. J.
Chem. Thermodyn., 2, 479-500 (1970).
[19] S.Z. Ziemniak and M.A. Goyette: Nickel(II)
oxide solubility and phase stability in high
temperature aqueous solutions. J. Solution
Chem., 33, 1135-1159 (2004).
[20] M.W. Chases: NIST-JANAF thermochemical
tables. J. Phys. Chem. Ref. Data, Monograph
9, 1999.
[21] S. U. Aja, S.A. Wood and A. E. W-Jones: The
aqueous geochemistry of Zr and the
solubility of some Zr-bearing minerals. Appl.
Geochem. 10, 603-620, [1995].
hydrous oxides. J. Nucl. Sci. Technol., 44, 9094 (2007).
I. A. Sheka and T. V. Pevzner: Solubility of
zirconium and hafnium hydroxides in sodium
hydroxide solutions. Russ. J. Inorg. Chem., 5,
1119-1121 (1960).
C. Ekberg, G. Kallvenius, Y. Albinsson, and P.
L. Brown: Studies on the hydrolytic behaviour
of zirconium (IV). J. Solution Chem., 33, 4779 (2004).
A. Veyland, L. Dupont, J.-C. Pierrard, J.
Rimbault and M. Aplincourt: Thermodynamic
stability of zirconium (IV) complexes with
hydroxy ions. Eur. J. Inorg. Chem., 1765-1770
(1998).
T. Sasaki, T. Kobayashi, I. Takagi and H.
Moriyama: Solubility measurement of
zirconium (IV) hydrous oxide. Radiochim.
Acta, 94, 489-494 (2006).
C.F. Baes and R.E. Mesmer: The hydrolysis of
cations. John Wiley & Sons, Inc., New York
(1976).
D. Tromans: Solubility of crystalline and
metamict zircon: a thermodynamic analysis. J.
Nucl. Mater., 357, 221-233 (2006).
J. H. Adair, R. P. Denkewicz, F. J. Arriagada
and K. O. Asare: Precipitation and in-situ
transformation in the hydrothermal synthesis
of crystalline zirconium dioxide. Ceram.
Trans., 1, 135-145 (1988).
A. Gualtieri, P. Borby, J. Hanson, and J.
Hriljac: Rietvield refinement using
synchrotron X-ray powder diffraction data
collected in transmission geometry using an
imaging-plate detector: application to standard
m-ZrO2. J. Appl. Cryst., 29, 707-713 (1996).
R.C. Weast, M. J. Astle, and W. H. Beyer:
CRC Handbook of chemistry and physics, 69th
Edition. CRC Press, Inc, Boca Raton, Florida
(1988).
S. M. Hasany, and M. H. Chaudhary: Sorption
of europium on zirconium oxide from aqueous
solutions, J. Radioanal. Nucl. Chem., 182,
305-315 (1994).
S. P. Mishra, V. K. Singh, and D. Tiwari:
Radiotracer technique in adsorption study:
Part XIV. efficient removal of mercury from
aqueous solutions by hydrous zirconium oxide.
Appl. Radiat. Isot., 47, 15-21 (1996).
P. H. Tewari and N. S. McIntyre:
Characterization of adsorbed cobalt at the
oxide-water interface, Amer. Inst. Chem. Eng.,
Symposium Series, 71, 134-137 (1975).
M. Qureshi, and K. G. Varshney: Inorganic
ion exchangers in chemical analysis. CRC
Press, Inc., Boca Raton, Florida (1991).
6