![]() |
Security and Privacy Notice |
Vapor Pressure over Liquid UO2
The recommended equation for the total vapor pressure over liquid UO2 from the melting point (3120 K) to 8000 K is the equation derived by Breitung and Reil1 from their in-pile equation-of-state measurements, and their review of the experimental data. Their equation for the logarithm of the saturated vapor pressure over liquid UO2 is
![]() |
| Figure 1 |
Vapor Pressure over Solid UO2
The recommended equation for the vapor pressure of UO2(g) over solid UO2 is based on measurements by Ackermann, Rauh, and Rand2 of the pressure of UO2(g) over UO2 in the temperature range from 1800 to 2600 K. Their equation for the logarithm of the vapor pressure of UO2(g) is
where the vapor pressure, P, is in MPa and the temperature, T, is in K. Ackermann, Rauh, and
Rand stated that in the temperature range of their measurements, UO2(g) comprises
approximately 94% of the total vapor pressure over solid UO2. Therefore, below 2600 K, this
equation gives a reasonable estimate of the total vapor pressure over solid UO2. Because
contributions to the total vapor pressure from other species become significant with increasing
temperature,3 this equation does not give a good estimate of the total vapor pressure over solid
UO2 near the melting point, 3120 K. The equation of Tetenbaum and Hunt4 is recommended
for the total vapor pressure over solid UO2 at temperatures above 2600 K. Tetenbaum and
Hunt4 measured the total vapor pressure over uranium dioxide as a function of stoichiometry.
Their equation for the total vapor pressure over UO2(s) is
![]() |
| Figure 2 |
The estimated uncertainties in total vapor pressure over liquid UO2 calculated from Eq.(1) range
from -40%/+60% at 3120 K to -45%/+80% at 6000 K. From 3120 to 6000 K, the negative
uncertainties are assumed to decrease linearly: Uranium dioxide can exist over a wide range of compositions (hypostoichiometric to
hyperstoichiometric with respect to oxygen), which are temperature dependent. The total vapor
pressure depends on the oxygen-to-uranium ratio of the condensed phase, so that the total vapor
pressure over UO2±x will depend on the value of x. The vaporization of UO2 is not congruent
because the O:U ratio in the gas phase is greater than in the condensed phase. The total vapor
pressure above solid and liquid UO2 includes contributions from UO2(g), UO(g), UO3(g), U(g), O(g),
and UO(g).
Breitung and Reil1 have recently reviewed the experimental measurements of the total vapor
pressure of liquid UO2. The data used in their assessment are summarized in Table 3. They
included both pressure-temperature measurements1, 5-15 and pressure-enthalpy measurements16-19,22,23
in their assessment.
Pressure-Temperature Measurements
The transpiration measurements of Reedy and Chasanov5 were weighted high in the assessment of
Breitung and Reil1 for the following reasons: (1) they are the only measurements on both solid and
liquid UO2, (2) the technique produces true equilibrium data, and (3) the experimental uncertainties
are very small ( The laser-heated vapor pressure measurements listed in Table 3 may be divided into two groups: (1)
measurements performed far from thermodynamic equilibrium6-13 and (2) measurements close to
thermodynamic equilibrium. 14,15 Measurements far from thermodynamic equilibrium consist of
experiments in which the fuel vapor expands into a vacuum or a rare gas environment. Such
experiments require a theoretical model to convert properties of the expanding nonequilibrium
plasma into saturation vapor pressures. The large scatter in the data from different experiments of
this type is indicative of the difficulty of obtaining saturation vapor pressure data from these
nonequilibrium measurements. Measurements close to equilibrium use a boiling point technique that
determines the temperature at which a laser-generated UO2 vapor cloud begins to expand against a
xenon cover gas of a given pressure. At this temperature, the UO2 vapor pressure is assumed to be
equal to the gas pressure. The xenon gas atoms initially confine the laser-generated vapor cloud so
that evaporation proceeds close to thermodynamic equilibrium. The recent boiling-point
experiments by Bober and Singer15 included corrections for optical absorption (by the vapor cloud)
of thermal radiation emitted from the liquid surface. Breitung and Reil concluded that the recent
measurements by Bober and Singer are the most reliable saturation vapor pressure data for liquid
UO2 from the laser experiments.
In-Reactor Experiments
The first in-reactor measurements of vapor pressure as a function of adiabatic fuel enthalpy by Reil16
determined upper and lower bounds for the vapor pressure. Later calculations showed that these
values were overly conservative.1 Benson17 measured the isobaric expansion of a 25- Breitung and Reil1 measured the saturation vapor pressure of pure UO2.01, reactor grade UO2.08, and
reactor grade (U0.77 Pu0.23)O2.09 as a function of enthalpy for enthalpies from 2000 to 3700 kJ kg-1.
Their six effective equation-of-state experiments at the annular core research reactor at Sandia
National Laboratories determined the saturation vapor pressure as a function of enthalpy at
conditions that are very close to those of the disassembly phase of a core disruptive accident. These
experiments gave very reproducible results. They found that under these conditions
(1) the fuel saturation vapor pressure for fuel containing uranium-plutonium mixed oxide was
essentially identical to that of pure UO2;
(2) fuel impurities from fabrication did not noticeably contribute to the pressure;
(3) stoichiometry variations have no strong influence on the saturation vapor pressure for UO2.01,
and UO2.08;
(4) replacement of uranium by plutonium in concentrations equivalent to mixed oxide fuel, e.g.,
(U0.77 Pu0.23)O2.09, does not significantly affect the measured vapor pressure.
From the data obtained in these six experiments, Breitung and Reil developed an equation for the
vapor pressure for all three fuels:
The positive
uncertainties are assumed to increase linearly from +60% at 3120 K to +80% at 4500 K:
Above 4500 K, the positive uncertainties are assumed constant
(+80%). The uncertainties in the pressure of UO2 (g) over solid UO2 calculated from Eq.(2) and in
the total vapor pressure over solid UO2 calculated using Eq.(3) are estimated as -40%/+60% from
1700 to 3120 K.
Stoichiometry
Discussion: Vapor Pressure over Liquid UO2
in pressure;
in temperature). In these experiments, the UO2 was
contained in tungsten, which limited the temperature range (2615 - 3390 K). At 2615 and 2860 K,
the O:U ratio of the condensed phase was 1.98. At 3390 K, the O:U ratio of the liquid was 1.94.
m-thick layer
of UO2 powder confined by two movable pistons as it was heated to a certain internal energy in one
microsecond. Results of this experiment were inconsistent with the expansion of a single-component
liquid-vapor system. An unknown source of pressure, such as water vapor, adsorbed by the fine UO2
powder is believed to have contributed to the measured pressure. Fission heating was used in the
eight Commissariat a l'Energie Atomique (CEA) experiments by Limon et al.18 to heat a thin solid
UO2 disk to boiling under constant argon pressure. The boiling point was determined by the sudden
increase in pressure. The average energy deposited in the UO2 was assumed to be equal to the fuel
enthalpy in the boiling zone. This assumption led to deviations of only a few percent in six high-enthalpy tests but the actual enthalpy in the boiling zone may have been on the order of 10% higher
for the two low-enthalpy tests.1
![]() |
| Figure 3 |
Comparison of Recommended Equation with Data
In Figure 4, the recommended equation of Breitung and Reil for the total vapor pressure over liquid UO2 is compared with the most recent and reliable vapor pressure data from each experimental method, with the equation formulated by the 1978 IAEA International Working Group on Fast Reactors (IWGFR),29 and with vapor pressures calculated by Green and Leibowitz.3 The IWGFR equation was based on a review of the data available in 1978 and was recommended for use up to 5000 K. The vapor pressures and vapor compositions above uranium dioxide calculated by Green and Leibowitz3 are based on a statistical-mechanical calculation of the thermodynamic functions of the individual vapor species using molecular energy levels from spectroscopic data and an oxygen potential model. Experimental data included in Figure 4 are: transpiration data of Reedy and Chasanov,5 the boiling-point data of Bober and Singer,15 data from the most recent laser-heating vaporization experiments of Ohse et al.,12, 13 and data from the in-pile experiments of Limon et al.18 The equation recommended by Limon et al. to best describe their data has also been included. Breitung and Reil's earlier vapor pressure equation that was obtained by using a different heat capacity22,23 to convert their data has been included in Figure 4 to show the effect of differences in choice of heat capacity on the final vapor pressure equation. It is labeled "Breitung KfK3939." Figure 4 shows that at high temperatures, it gives lower pressures than the recommended equation of Breitung and Reil. Therefore, the recommended equation is in better agreement with the high-temperature data of Limon et al.
![]() |
| Figure 4 |
Breitung and Reil1 noted that if the two low-temperature CEA data points of Limon et al.18 are disregarded, all in-pile results are located close to an almost linear extension of the transpiration data of Reedy and Chasanov5 and the laser boiling point data of Bober and Singer.15 All these methods provide conditions very close to equilibrium vaporization so that the slope of the line connecting these data should give the heat of vaporization. They attributed the steeper slopes obtained from the earlier laser-vaporization experiments (as characterized by the 1980 data of Ohse et al.) to the use of nonequilibrium pressure models to reduce the data and/or to the neglect of optical absorption of thermal surface radiation in the vapor cloud. Application of the Clausius -Clapeyron equation to their vapor pressure equation gives an effective enthalpy of vaporization:
Although the total vapor pressure above solid UO2 includes contributions from UO2(g), UO(g),
UO3(g), U(g), O(g), and UO(g), the greatest contribution is from UO2(g). Ackermann et al.2
measured the vapor pressure of UO2(g) above solid UO2 from 1800 to 2600 K and commented that
UO2(g) comprises 94% of the total pressure at 2150 K. Tetenbaum and Hunt4 determined the
total vapor pressure above UO2-x in the temperature range 2080 to 2705 K. Green and Leibowitz3
used models for the partial Gibbs free energy of oxygen above UO2 to determine the contributions
of the various vapor species above hypostoichioeteric uranium dioxide for UO2.00 through UO1.90.
Measurements of the total vapor pressure above solid UO2 by Knudson effusion,30-33 Langmuir
surface evaporation,35 and transpiration4, 36 methods and have been reviewed by Ackermann et
al.2 and compared with measurements of the vapor pressure due to UO2(g) determined from mass-spectrometric measurements by Pattoret et al.37 and by Ackermann et al.2,30,34 They found
reasonable agreement between the different measurements. Table 4 shows the vapor pressures at
2150 K determined from the experiments included in the assessment by Ackermann et al.2
Ackermann et al. corrected the data of Alexander et al.36 for a systematic error. Consequently, the
vapor pressure attributed to Alexander et al. in Table 4 (which is from the table of Ackermann
et al.2) differs from the value given in the original paper by Alexander et al.36 The average of the
values, excluding the value from the 1979 mass spectroscopy measurements by Ackermann et al.,2
is 1.38 x 10-7 MPa. This is in good agreement with the vapor pressure of UO2(g) (1.32 x 10-7 MPa)
determined by Ackermann et al. in 1979.
The recommended equation for the vapor pressure of UO2(g) over UO2, Eq.(2), is from the 1979
measurements and assessment of Ackermann et al.2 It is in reasonable agreement with other data
and was derived with considerations for consistency with the thermodynamic functions for solid UO2
and the enthalpy of sublimation from the solid. It is consistent with a heat capacity that has a phase
transition at 2670 K. In Figure 5, this recommended equation of Ackermann et al.2 for the vapor
pressure of UO2(g) over solid UO2 is compared with vapor pressure equations and data from earlier
measurements and with the vapor pressure of UO2(g) and the total vapor pressure over UO2.00
calculated by Green and Leibowitz.3 In the legend for Figure 5, the notation UO2 has been
included to distinguish measurements or calculations of the pressure due to the vapor species UO2(g)
from the total vapor pressure over UO2. Below 2450 K, the 1956 low-temperature data of
Ackermann et al.30 and the equation of Tetenbaum and Hunt4 are in excellent agreement with the
recommended equation of Ackermann et al.2 Above 2615 K, the equation of Tetenbaum and
Hunt for the total vapor pressure over UO2 gives higher vapor pressures than the equation of
Ackermann et al. for the vapor pressure of UO2(g). Two data from transpiration measurements of
the total vapor pressure over UO1.98 by Reedy and Chasanov5 have been included in Figure 5. These are the only vapor pressure measurements over uranium dioxide in both the liquid and solid
phases. The Reedy and Chasanov datum at 2615 K is in good agreement with the equation of
Ackermann et al. but their datum at 2860 K is higher than values from both the equation of
Ackermann et al. and the equation of Tetenbaum and Hunt. Total vapor pressures over UO2
measured by Ohse et al.32 are in good agreement with the equation of Ackermann et al. at low
temperatures but are higher at high temperatures. Above 2500 K, the data of Ohse et al. approach
total pressures calculated by Green and Leibowitz. The contribution to the total vapor pressure
from UO2(g) calculated by Green and Leibowitz is in good agreement with the equation of
Ackermann et al.2 above 2600 K. However, the total vapor pressure over UO2 calculated by Green
and Leibowitz is consistently higher than the UO2(g) pressure given by the equation of Ackermann
et al. The difference between these values increases with temperature. The contribution to the total
vapor pressure from UO2(g) calculated by Green and Leibowitz decreases with increasing
temperature. It is 70% at 2100 K, 54% at 2500 K, and only 37% at 3100 K. These comparisons
indicate that the equation for the vapor pressure of UO2(g) over solid UO2 is a reasonable approximation of the total vapor pressure over solid UO2 up to 2600 K but not at higher temperatures. At higher temperatures, extrapolation of the equation of Tetenbaum and Hunt [ Eq. (3)] is a better approximation to the total vapor pressure over solid UO2.
![]()
where
is in J mol-1 and T in K ranges from 3120 to 8000 K. The heat of vaporization at the
normal boiling point (3815.1 K) is 413.5 kJ mol-1.
Discussion: Vapor Pressure over Solid UO2
![]() |
| Figure 5 |
![]() |
| Figure 6 |