Kansai Electric Power Company Inc Tokyo Japan

Abstract

A programme of physical properties measurements has been carried out on MOX fuel manufactured using the Short Binderless Route (SBR) by BNFL and on MOX fuel manufactured using the MIMAS process by Belgonucleaire. The programme includes the following work: - Determination of the melting point of MOX fuel. - The measurement of the thermal expansion of MOX fuel. - Determination of the thermal diffusivity of MOX fuel. This paper will describe the programme of measurements and summarise the results obtained as well as analysing the results in comparison with previous published work, where applicable.

1. INTRODUCTION

A programme of fuel properties measurements for unirradiated uranium dioxide and uranium/plutonium mixed oxide (MOX) fuel has been carried out. The programme covered the measurement of thermal expansion, melting point and thermal diffusivity, together with appropriate pre- and post-test characterisations. This paper reviews the results of these measurements.

2. SAMPLE FABRICATION

The unirradiated MOX fuel samples were manufactured by BNFL using the Short Binderless Route (SBR). This involves blending U02 and Pu02 in a high energy attritor mill and conditioning the powder in a speroidiser prior to pellet pressing and sintering. The U02 samples were manufactured at BNFL Springfields using depleted U02 powder produced via the Integrated Dry Route (IDR) and contained CONPOR pore former. The MOX fuel was manufactured at Sellafield from depleted IDR U02 powder blended with PuO powder and also contained CONPOR. Hyperstoichiometric MOX fuel, with an O/M ratio of 2.02, was produced by controlling the oxygen potential of the furnace atmosphere during the final sintering of the samples.

In addition to these U02 and MOX samples fabricated by BNFL, MOX samples were prepared from a fuel rod which had been manufactured by Belgonucleaire. This MOX fuel was manufactured by the MIMAS process, which involves the addition of U02 and Pu02 powders to form a master blend which is then added to U02 to form the MOX powder of the correct enrichment prior to pelleting.

A more complete description of the characterisation of the fuel samples is given Table 1.

Table 1: Fuel Variant Nominal Characteristics

SBR MOX

SBR MOX

MIMAS

Characteristic

UO,

(O/M=2.00)

(0/M=2.02)

MOX

Pu/(U + Pu)(%)

-

9.91

9.91

9.64

Pu Agglomerate Size (pan)

-

<100

<100

<214

O/M Ratio

2.00

2.00

2.02

2.00

Density (%TD)

95.05

94.45

94.48

93.81

Diameter (mm)

8.202

7.699

7.678

8.203

Length (mm)

9.384

8.940

8.904

11.830

Mean Pore Size (all, /¿m)

2.6

2.0

2.0

1.7

Mean Pore Size (> 5/^m)

7.1

7.7

7.6

8.1

No. of Pores > lOO^m

0.03

0

0.03

0.03

(%)

Grain Size (^m)

7.6-8.1

8.0-8.1

8.0-8.1

0-12

Figures la - lc show ceramographs of archive U02, SBR (U,Pu)Q and SBR (U.Pu)^ , respectively. In addition, Figures 2a and 2b show oc-autoradiographs of the SBR (U,Pu)02 and SBR (U,Pu)O2 02, respectively.

1mm

Fig. le: SBR (U,Pu)O2.02 Ceramograph

Water cooled base Window (Si02)

Heat shield pack ^

Insulated RF coil

Sealed capsule (W)

Water cooled concentrator Susceptor (W) Melted sample

Susceptor heat shield (W)

Support spades (W) liij Support (Zr02)

nfl Support (SiC>2)

ff Alumina tile

Base support (Cu) "O' ring vacuum seal Vacuum connection

Water cooled base Window (Si02)

  1. 3a: Melting Point Apparatus
  2. 3b: Thermal Diffusivity Apparatus

Furnace Case

Alumina Push Rod Alumina Support Tube

Thermocouples Sample: Pt/Ptl3%Rh Control: Pt6%Rh/Pt30%Rh

Specimen

Platinum Wound Furnace Tube

Ceramic Insulation

Fig. 3c: Thermal Expansion Apparatus

3. METHODOLOGY

The melting points of unirradiated stoichiometric U02 and MOX, and hyperstoichiometric MOX were measured using the thermal arrest method. In these experiments, the temperature of the fuel samples in a sealed tungsten capsule, as determined using an infra-red pyrometer, are measured as a function of time whilst the capsules are heated using an RF induction furnace. The apparatus is shown in more detail in Fig. 3a. The melting point of the sample is marked by a pause in the temperature rise which is the 'thermal arrest'. Calibration of the technique was made using pure molybdenum and tantalum samples.

The thermal diffusivity of unirradiated stoichiometric U02 and MOX, and hyperstoichiometric MOX was measured by the laser flash method. With this method, thin samples of the fuel were sealed in a vacuum and heated to the test temperature by a furnace. The time taken for a heat pulse, generated by a pulse from a laser, to travel through the sample is then measured in order to determine the diffusivity of the sample. The apparatus is shown in more detail in Fig. 3b. Measurements were made in the range 400 - 1600°C by cycling the temperature of the samples up and down in 200°C intervals and measuring the diffusivities on both legs of the temperature cycle. Each type of fuel was subject to one cycle up and down in temperature with the exception of the hyperstoichiometric fuel, where three additional samples were subjected to additional heating cycles with peak temperatures of 1000°C, 1400°C and 1600°C.

The thermal expansion measurements were performed on single pellets of unirradiated stoichiometric U02 and MOX, and hyperstoichiometric MOX. The pellets were heated by a furnace in an argon atmosphere and the expansion of the pellets in the vertical direction measured by a transducer via an alumina push rod. The apparatus is shown in more detail in Fig. 3c. The rig was calibrated using pure molybdenum pellets. Measurements were made of the thermal expansion of the fuel samples at temperatures in the range 150°C - 1230°C by cycling up to the peak temperature and down again twice for each sample.

4. THERMAL DIFFUSIVITY

The results of the thermal diffusivity measurements for two stoichiometric MOX samples are shown in Fig. 4 as a typical example of the measurements obtained. It can be seen that reproducibility

<5 8.0E-7 ■ • x: H

  1. 0E-7 • -
  2. 0E-7 -0 200 400 600 800 1000 1200 1400 1600 1800 2000

Temperature (°C)

Fig. 4: Result of Thermal Diffusivity Measurements on Stoichiometric MOX Samples

4.0E-7 -0 200 400 600 800 1000 1200 1400 1600 1800 2000

of the data for multiple measurements at each temperature is good during both heating up and cooling down for these stoichiometric samples, as the dispersion of the thermal diffusivities at each temperature is less than 10%. From this result, it can be confirmed that changes in the specimen characteristics, such as stoichiometry and microstructure, during the measurements must be small.

The thermal conductivity was derived from these results by using the following equation:

k = a-Cp-p where k is the thermal conductivity in W/m/K, a is the thermal diffusivity in m2/s, Cp is the specific heat capacity in J/kg/K, and p is the density in kg/m3. The specific heat capacity was derived using the equation recommended in MATPRO Version 11 [5]:

K1-92-exp

where Cp is the specific heat capacity in J/kg/K, T is the temperature in K, O/M is the oxygen to metal ratio, R is the gas constant (8.3143 J/mol/K), 6 is the Einstein temperature in K, and Ku K2, K3, and Ed are constants.

The specific heat capacity of the MOX samples was calculated from those of pure U02 and PuO: assuming that the contribution from each constituent was in proportion to its weight fraction.

The theoretical density of the samples used in this study was around 95%, although there were small differences between the samples. The thermal conductivity of each sample was corrected for a density of 95% using the Maxwell-Euken's equation:

k_ p . 100+0.5(100-95). k " 100+(J(100-p) 95 95

where k is the thermal conductivity of a test sample with a density of p% TD, p is a constant and kis the thermal conductivity of a sample with a density of 95% TD.

  1. 5 shows the thermal conductivity of U02 derived using this method from the thermal diffusivity measurements. This figure also shows the thermal conductivities recommended by Martin [1], Washington [2], Brandt et al. [3] and Kosaka [4], The data presented here can be seen to agree with these recommendations, and to have the same temperature dependence in the range used in the tests (400 - 1600°C). This confirms the validity of the measuring technique and the reliability of the data.
  2. 6 shows the thermal conductivity of stoichiometric MOX along with the recommendations due to Martin [1] and MATPRO version 11 [5]. The data presented here can be seen to fall between these two recommendations.

In Fig. 7, the result of Gibby's study [6] on the effect of Pu02 content on the thermal conductivity of (U,Pu)02 solid solutions is shown along with the data from this study. This data shows that the thermal conductivity of the stoichiometric MOX sample (9.9 wt% Pu/U+Pu) was approximately 10% less than that of U02 in the range 400 - 1600°C. The same dependence of thermal conductivity on Pu content is observed in Gibby's data, and while the absolute values of the data in this study are a little smaller than Gibby's data at the lower temperatures (400 - 600°C), it agrees very well with Gibby's data at the higher temperatures (800°C - 1200°C).

In the case of the thermal diffusivity measurements on the hyperstoichiometric MOX, a slight increase in the diffusivity was observed in the cooling cycle relative to the heating cycle, although the diffusivity in the cooling cycle agree reasonably well with the data for stoichiometric MOX. It is believed that the heating cycle up to high temperature moved the O/M ratio closer to 2.00. Based on several measurements for hyperstoichiometric samples with different peak temperatures, it was show that the change in O/M ratio was negligible under 1200°C, and hence, only the data for the heating cycle below 1200°C is used in this evaluation. In Fig. 8, the thermal conductivity of the hyperstoichiometric MOX is compared with the results for the stoichiometric MOX, along with the recommendation for stoichiometric MOX due to Martin [1]. It can be seen that the thermal conductivity of the hyperstoichiometric sample was smaller than that of the stoichiometric sample by about 10% in the range 400 - 1200°C.

Fig. 5: Temperature Dependence of Thermal Conductivity of UO2

Temperature (°C)

Fig. 5: Temperature Dependence of Thermal Conductivity of UO2

Tantalum Thermal Conductivity

Temperature (°C) Fig. 6: Thermal Conductivity of Stoichiometric MOX

Fig. 7: Thermal Conductivity of MOX as a function of Pu(>2 Content

H 30

Pu02 Content (wt%)

Fig. 7: Thermal Conductivity of MOX as a function of Pu(>2 Content

i o O/M Ratio=2.00 - This Study • O/M Ratio=2.02 - This Study ----O/M Ratio=2.02 - Martin [1]

600 800 Temperature (°C)

1000

1200

  1. 8: Thermal Conductivity of Stoichiometric and Hyperstoichiometric MOX
  2. MELTING POINT

In this study, the average melting point of the UO: sample was found to be 2849°C, which agrees well with the value 2840°C for unirradiated UO: which is given by MATPRO version 11 [5] referring to Brassfield et al. [7] and Lyon et al. [8]. Hence, the data reported here is in good agreement with their results, which gives confidence in the accuracy of the measurement technique.

In Fig. 9, the liquidus and solidus temperatures of the stoichiometric MOX samples are shown as a function of the Pu content along with those measured by other investigators, namely Adamson [9], Lyon et al. [10] and Aitken et al. [ 11], While the solidus temperature of the stoichiometric MOX was higher than that recommended by Adamson [9], the solidus temperature was found to be consistent with the solidus recommended by Adamson.

2900

Liquid us Soiidus -- Liquidus -Soiidus -Liquidus Soiidus -Liquidus Soiidus -

  • This Study This Study
  • Adamson [9] Adamson [9] -Lyon [10] Lyon [10] -Aitken [11] Aitken [11]

2300

10 20 30 40 50 60 70 Pu02 Content in MOX (mole%)

Fig. 9: Liquidus and Soiidus Temperatures of MOX

As can be seen in Fig. 9, there was no obvious difference between the average liquidus and soiidus temperatures of stoichiometric SBR MOX and those of MIMAS MOX. From these results, it would appear that the MOX pellet fabrication process does not affect the melting point of the product.

The average soiidus temperatures of the stoichiometric and hyperstoichiometric (O/M = 2.02) MOX samples was lower than that of the stoichiometric MOX samples by about 70°C. This implies that the degradation of the soiidus due to a deviation of the O/M ratio of 0.02 from stoichiometry is around 70°C.

6. THERMAL EXPANSION

The thermal expansion coefficient at a particular temperature (6°C) was normalised to zero at 200°C according to the following formula:

where M^/L-^ is the normalised thermal expansion coefficient at 6°C, Z« is the sample length at 0°C, Ljoo Is the sample length at 200°C and Lis the sample length at 25°C.

Fig. 10 shows the temperature dependence of the thermal expansion coefficients of U02 and stoichiometric MOX with the recommended values for U02 due to Martin [12] and MATPRO version 11 [5], It can be seen that the averaged thermal expansion of the stoichiometric MOX sample is almost equal to that of the U02 samples, and in good agreement with the recommendations of Martin [12] and MATPRO version 11 [5] at temperatures up to 1200°C.

The thermal expansion of the hyperstoichiometric MOX samples is compared with that of the stoichiometric MOX samples in Fig. 11. This shows that there is virtually no observable difference in the thermal expansion resulting from a deviation in stoichiometry of 0.02.

Thermal Dilation Table

Fig. 10: Thermal Expansion of Stoichiometric UO2 and MOX

Temperature (°C)

Fig. 10: Thermal Expansion of Stoichiometric UO2 and MOX

Fig. 11: Thermal Expansion of Stoichiometric and Hyperstoichiometric MOX

Temperature (°C)

  1. 11: Thermal Expansion of Stoichiometric and Hyperstoichiometric MOX
  2. CONCLUSIONS

In this study, the thermal diffusivity, melting point and thermal expansion of stoichiometric U02, stoichiometric and hyperstoichiometric SBR MOX (9.9 wt% Pu/U + Pu), and the melting point of stoichiometric MIMAS MOX (9.6 wt% Pu/U + Pu) have been measured.

The following conclusions can be drawn from this study:

  1. The thermal conductivity of stoichiometric U02 and MOX is confirmed to be in good agreement with previous published data, and that of MOX was approximately 10% less than that of U02 at temperatures between 400 and 1600°C.
  2. The thermal conductivity of hyperstoichiometric MOX (O/M ratio = 2.02) was smaller than that of stoichiometric MOX by about 10% in the range 400 - 1600°C.
  3. There were virtually no differences in the observed thermal expansions of stoichiometric U02, stoichiometric and hyperstoichiometric MOX in the temperature range 150 - 1200°C.
  4. The melting point of stoichiometric U02 and MOX was in good agreement with those of other investigators.

ACKNOWLEDGEMENTS

The authors wish to express their appreciation to Hokkaido Electric Power Co., Inc., Shikoku Electric Power Co., Inc., Kyushu Electric Power Co., Inc., Japan Electric Power Co., Inc., and Nuclear Fuel Industries, Ltd. for their support of this work and their approval of this presentation. We also wish to acknowledge the help of the staff of AEA Technology at Harwell and Windscale who carried out the measurements described in this paper.

REFERENCES

  • 1] MARTIN D G, 'A Re-appraisal of the Thermal Conductivity of U02 and Mixed (U,Pu) Oxide Fuels', J Nuc Materials, 110 (1982) 73-94
  • 2] WASHINGTON A B G , UKAEA Report TRG 2236 (D) (1973)
  • 3] BRANDT R., HAUFLER G., & NEUER G., 'Thermal Conductivity and Emittance of Solid U02', J Non-Equilibrium Thermodynamics, 1 (1967) 3
  • 4] KOSAKA Y., MATSUOKA Y., ABETA S., DOI S., & IRISA Y., 'Out of Pile Testing on the Properties of the High Content Gadolinia-Bearing Fuel (2)', Paper to the Meeting held by the Atomic Energy Society of Japan, 2-5 October 1990
  • 5] MATPRO Version 11 - A Handbook of Materials Properties for use in the Analysis of Light Water Reactor Fuel Rod Behaviour', NUREG/CR-0497, TREE-1280 (1979)
  • 6] GIBBY R L., 'The effect of Plutonium Content on the Thermal Conductivity of (U,Pu)02 Solid Solutions', J Nuc Materials 38 (1971) 163-177
  • 7] BRASSFIELD H C et al., 'Recommended Property and Reactor Kinetics Data for use in Evaluating Light-Water-Cooled Reactor Loss-of-Coolant Incident Involving Zircaloy-4 or 304-SS-Clad U02', GEMP-482 (April 1968)
  • 8] LYON M F et al., 'UO-, Properties Affecting Performance', Nuclear Engineering and Design 21 (1972) 167
  • 9] ADAMSON M G., AITKEN E A., & CAPUTI R W., 'Experimental and Thermodynamic Evaluation of the Melting Behaviour of Irradiated Oxide Fuels', J Nuc Materials 130 (1985) 349-365.
  • 10] LYON W L & BAILY W E., 'The Solid-Liquid Phase Diagram for the U02-Pu02 System', J Nuc Materials 22 (1967) 332-339
  • 11] AITKEN E A & EVANS S K., 'A Thermodynamic Data Program Involving Plutonia and Urania at High Temperatures', GEAP-5634 (1968)
  • 12] MARTIN D G., 'The Thermal Expansion of Solid U02 and (U,Pu) Mixed Oxides - A Review and Recommendations', J Nuc Materials 152 (1988) 94-101
0 0

Post a comment