Journal of Nuclear Materials 427 (2012) 1–11

Contents lists available at SciVerse ScienceDirect

Journal of Nuclear Materials

journal homepage: www.elsevier.com/locate/jnucmat

Microstructural and thermophysical properties of U–6 wt.%Zr alloy for fast

reactor application
Santu Kaity a, Joydipta Banerjee a, M.R. Nair a, K. Ravi a, Smruti Dash b, T.R.G. Kutty a,⇑,
Arun Kumar a, R.P. Singh c
a
Radiometallurgy Division, Bhabha Atomic Research Centre, Mumbai 400 085, India
b
Product Development Division, Bhabha Atomic Research Centre, Mumbai 400 085, India
c
Nuclear Fuels Group, Bhabha Atomic Research Centre, Mumbai 400 085, India

a r t i c l e i n f o a b s t r a c t

Article history: The microstructural and high temperature behavior of U–6 wt.%Zr alloy has been investigated in this
Received 28 September 2011 study. U–6 wt.%Zr alloy sample for this study was prepared by following injection casting route. The ther-
Accepted 29 March 2012 mophysical properties like coefficient of thermal expansion, specific heat, thermal conductivity of the
Available online 19 April 2012
above alloy were determined. The hot-hardness data of the U–6 wt.%Zr alloy was also generated from room
temperature to 973 K. Apart from that, the fuel-clad chemical compatibility with T91 grade steel was also
studied by diffusion couple experiment. No studies have been reported on U–6 wt.%Zr alloy. This paper
aims at filling up the gap on characterization and thermophysical property evaluation of U–6 wt.%Zr alloy.
Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction subsystem of U–15 wt.%Pu–6 wt.%Zr alloy, and is also proposed as

a blanket material for the sodium bonded ternary U–Pu–Zr fuel. U
Metallic fuels have been recognized as one of the candidate with low Zr (6 wt.%) in the fuel is preferred in the Indian nuclear
driver fuels for liquid–metal cooled fast reactor because of their power programme with the objective of achieving a higher
advantage in view of their superior performance, reactor safety, breeding ratio [9]. On account of the above, the determination of
and fuel cycle economics [1–8]. They were selected based on the the thermophysical properties of this particular fuel composition
ease of fabrication, high thermal conductivity and high fissile and is very significant for reactor application.
fertile atom densities, which facilitates higher breeding. Zr was The knowledge of thermophysical properties such as specific
chosen as an alloying element for metallic fuels since it increases heat, coefficient of thermal expansion, density, thermal conductiv-
the solidus temperature. It also helps in overcoming the fuel clad ity etc. of the fuel as a function of temperature are extremely impor-
chemical interaction and helps in stabilizing the isotropic gamma tant and must be known in order to effectively model and predict
phase over a wider temperature range. Uranium alloyed with Zr fuel behavior under normal and off-normal irradiation conditions.
has excellent dimensional stability during thermal cycling. The Specific heat is an intensive quantity that illustrates the amount
Experimental Breeder Reactor (EBR-II) initially operated with of heat needed to change the body of fuels temperature [10]. It is re-
U–5%Fs (Fs is fissium, a simulated mixture of noble metal fission quired in the modeling of heat conduction during normal and under
products) cores and later used U–Zr as driver fuel. U–10 wt.%Zr transient conditions and for calculating thermodynamic properties.
has been irradiated in EBR-II and the Fast Flux Test Facility (FFTF) [1]. In the case of an excursion, the specific heat directly affects the fuel
Primarily, two design concepts have been proposed for the behavior and determines the temperatures attained during the
metallic fuel development programme for FBRs in India [9]. Two excursion, and therefore the Doppler feedback. Specific heat is also
fuel concepts being explored are: needed to convert thermal diffusivity to thermal conductivity, the
former being commonly measured in out-of-pile tests. Coefficient
(a) Mechanically bonded pin with U–15 wt.%Pu alloy as fuel of thermal expansion (CTE) data is essential for modeling fuel
(b) Sodium bonded pin with U–15 wt.%Pu–6 wt.%Zr alloy as fuel behavior in nuclear reactors. It is needed to calculate stresses
occurring in the fuel and cladding on change in temperature.
In the case of the mechanically bonded binary alloy fuel, Zr liner Furthermore, coefficient of thermal expansion can be utilized to
is proposed to be used inside the clad to prevent the fuel clad chem- determine the change of alloy density as a function of temperature.
ical interaction. T91 steel (mod. 9Cr1Mo steel) will be used as the Higher thermal expansion leads to a higher volume and lowers the
cladding material. U–6 wt.%Zr (hereafter, referred as U–6Zr) is a density of the fuel system. This in turn increases the neutron leak-
⇑ Corresponding author. Tel.: +91 2225595361; fax: +91 2225505151. age and allows for negative feedback of the system [11,12]. The
E-mail address: [email protected] (T.R.G. Kutty).
thermal conductivity of nuclear fuel affects the temperature profile

0022-3115/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jnucmat.2012.03.045
2 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11

which in turn influences almost all important processes such as Table 1

fission gas release, swelling, grain growth etc. and limits the linear Chemical composition of U–6Zr alloy.

power. The changes in thermal conductivity occur during irradia- Zr Fe Cr Ni Mn Ca Al Cu Zn

tion by the formation of fission-gas bubbles, porosities, build-up 5.95 20 2 9 3 6 26 3 2
of fission products and by the change of fuel stoichiometry. Thermal
Zr content is in wt.%; rests are in ppm.
conductivity (k) is one of the most important properties of both fuel
and cladding material. And the maximum linear heat rating (LHR)
were 99.9% and 99.95%, respectively. Graphite crucible was coated
of the fuel without causing any center melting is determined by
with alcohol based yttria slurry to prevent carbon pick-up from
the thermal conductivity of the fuel.
crucible. The melt was heated to about 1700 K under vacuum. Once
Since hardness is related to yield stress, ultimate tensile strength
the melting procedure was over, open ends of quartz molds were
(UTS), stress exponent and other mechanical properties by simple
immersed in the melt. The furnace chamber was pressurized with
empirical relations, prediction of these mechanical properties from
inert gas to push molten alloy into the quartz molds. The fuel slugs
hardness data will be very useful for radiotoxic materials where
are removed from the quartz molds using de-molding set up. Fuel
conventional testing is very difficult to perform. The hot hardness
slugs were subjected to dimensional and visual examinations.
data further give information on softening behavior [13] and can
Typical diameter of the alloy slugs was about 5 mm. The trace
be correlated to the long term rupture and creep properties.
metallic constituents were totaling to only 70 ppm as revealed by
The evaluation of the properties and behavior of zirconium–
the Inductively Coupled Plasma-Atomic Emission Spectroscopic
uranium alloys has been elaborately discussed in the reports edited
(ICP-AES) analysis. Results of the chemical analysis are presented
by Rough [14] and Bauer [15]. In both the reports the phase trans-
in Table 1.
formation behavior, different physical properties such as density,
resistivity, thermal conductivity, thermal expansion, diffusion
coefficient and various mechanical properties such as hardness, 2.2. Microstructural analysis
tensile, creep were reported for the entire alloy range, with empha-
sis on the Zr-rich U–Zr alloys [14,15]. The specific heat of the U–Zr For the metallographic and XRD analyses, the alloy slug was cut
system has been investigated by Fedorov and Smirnov [16]. The into disc of suitable thickness using slow speed SiC abrasive cut-off
thermophysical properties including thermal diffusivity and spe- wheel. Standard metallographic techniques were adopted for
cific heat of U–Zr alloys were reported by Takahashi et al. [17,18] grinding and polishing. Electroetching was carried out using 50%
and Matsui et al. [19]. The microscopic study of phase transforma- H3PO4 aqueous solution as electrolyte and SS304 as cathode with
tion behavoiur of U–2 wt.%Zr alloy has been reported by metallog- a constant potential of 2 V. The microstructural analysis was car-
raphy by Lagerberg et al. [20]. They also investigated phase ried out using a scanning electron microscope (SEM) and an energy
transformations of uranium–zirconium alloys containing 0.5, 2 dispersive spectroscope (EDS). Cu Ka radiation was used for XRD
and 5 wt.% zirconium by means of DTA [20]. The effect of compo- analysis with fixed slit optics and h–h goniometer. Diffraction pat-
sition and cooling rate on the decomposition of the high tempera- terns were obtained with a step size of 0.008° and a measuring
ture c-phase of U–Zr system has been studied by Hills et al. [21]. time of 2 s.
Recently, the phase transformation behavoiur, thermal expansion
and microhardness of as-cast U-rich U–Zr alloys were reported 2.3. Phase transformation temperature by differential scanning
by Basak et al. [22]. The phase transformation including martens- calorimeter
itic transformation of U–2 wt.%Zr alloy were also reported by Basak
et al. [23,24]. The fuel-clad chemical interaction (FCCI) is consid- The transformation temperature measurements were carried
ered as one of the major problems in metallic fuel because it limits out using a heat flux type differential scanning calorimeter (Model
the life of the fuel pin due to formation of low melting eutectic. The No. Setsys Evoluation 24, combined TG/DTA/DSC; make: M/s Seta-
FCCI behavior of U–Zr alloys with ferritic steel has been reported in ram Instrumentation, France). For DSC experiments, sample of
the literature with emphasis on the U–10 wt.%Zr alloy [25–28]. about 200 mg was cut from the bulk one and electropolished to re-
The main objectives of the present study are: move any oxide layer present using 50% phosphoric acid solution.
The sample was then loaded into the alumina crucible. During DSC
(i) Determination of thermophysical properties like coefficient of operation, the inside of the DSC furnace was purged with highly
thermal expansion, specific heat, thermal conductivity of U– purified argon gas flowing at 2 dm3 h1. To avoid oxidation, com-
6Zr alloy sample prepared by injection melting, mercially available oxygen trap was used. The heating and cooling
(ii) Evaluation of the hot-hardness data of the above alloy from rates were programmed at 5 K min1. The temperature calibration
room temperature to 973 K, was accomplished by measuring the melting points of high purity
(iii) Studies on the fuel-clad chemical compatibility with T91 metals, such as Zn, Al and Ag. The resulting thermogram was re-
grade steel by diffusion couple experiments, corded. The instrument was evacuated with a standard mechanical
(iv) Investigation on the microstructures of as-cast, quenched and pump and backfilled with argon several times prior to the tests.
heat treated U–6Zr alloy. The temperatures of phase transformation have been determined
from the heating curve.
Although, there are many reports available for U–Zr alloys with
various Zr content, no detailed studies have been reported on U– 2.4. Dilatometry
6Zr alloy. This paper aims at filling up the gap on characterization
and thermophysical property evaluation of U–6Zr alloy. The expansion behavior of U–6Zr was studied using a high tem-
perature vertical dilatometer (make: SETARAM Instrumentation,
2. Experimental France; model: Setsys Evoluation 24). The heating rate used for
the present study was 5 K min1. Here, the sample is loaded inside
2.1. Alloys preparation a sample holder which is made of Al2O3. The change of the length
of the sample was transmitted through the frictionless push rod to
U–6Zr alloy sample was prepared by following injection casting an LVDT transducer. A nominal load of 5 g was chosen to be applied
route. The purities of starting materials, uranium and zirconium, by the push rod over the sample. A calibrated thermocouple was
 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11 3

placed to record the sample temperature. The dilatometric experi- pyramid indenter. Before starting the hot hardness experiment, the
ments were carried out in high purity argon atmosphere at a dy- instrument was calibrated using a standard (Cu: SRM; NIST, USA)
namic gas flow rate of 2 dm3 h1. sample. Five indentations were made on the standard using
300 g load. The hardness obtained was found to be within 0.5% of
2.5. Specific heat measurements the actual value. The hardness (H) is related to the indentation
diagonal by the following relation:
Specific heat measurements were carried out using a heat flux 2
HðVHNÞ ðkg mm2 Þ ¼ 1:854  P=d ð1Þ
type differential scanning calorimeter, DSC-131 supplied by SETA-
RAM Instrumentation, France. The temperature and heat calibra- where P is the applied load (kg) and d is the average diagonal (mm).
tions of DSC were carried out using phase transition temperatures Sample of 6 mm length was cut from the alloy slug of diameter
and transition heats of some NIST reference materials [29]. 5 mm and were metallographically prepared. A load of 300 g and a
The specific heat of U–6Zr alloy was measured in the continu- dwell time of 5 s were used for this study. The load was applied at a
ous mode. In this mode, temperature was increased continuously rate of 0.2 mm min1. The hardness was measured in vacuum
from 303 K to 919 K. Stability time of 600 s was given in each iso- (0.1 Pa) from room temperature to 1173 K at every 100 K intervals.
thermal mode. The molar heat capacities were measured with The sample temperature was kept constant within ±1 K and that of
identical experimental conditions in the three runs. In the first the indenter was kept within ±3 K. At each temperature, at least
run, two empty identical aluminum crucibles of identical masses three readings were taken. Average of these hardness data at each
and 104 dm3 capacity with covering lid were kept in the sample temperature was plotted as a function of temperature.
and reference cells and the heat flow versus temperature were re-
corded. In the second run, the heat flow versus temperature were 2.8. Chemical compatibility with T91 cladding
recorded by loading NIST synthetic sapphire (SRM-720) in the
powder form (200 mg) into the aluminum crucible in the sample 2.8.1. Eutectic reaction temperature between U–6Zr alloy and T91
cell keeping the crucible in the reference cell empty. In the third grade steel
run, heat flow versus temperature were measured by loading the The eutectic temperature between U–6Zr and T91 was deter-
actual experimental sample (200 mg) into the aluminum crucible mined by using a heat flux type differential scanning calorimeter
in the sample side and keeping the crucible in the reference side (DSC). The chemical composition of T91 steel is given in Table 2.
empty. In all the three sets of experiments, high purity argon as The pre-melted alloy between U–6Zr and T91 was used as sample.
a carrier gas with the flow rate of 3 dm3 h1 and furnace heating The sample was loaded into the alumina crucible inside of the
rate of 5 K min1 were maintained. furnace which was purged with high purity Ar gas flowing at
2 dm3 h1. The heating and cooling rates were programmed at
2.6. Thermal conductivity by Transient Plane Source (TPS) technique 1 K min1. The eutectic temperature was determined from the
heating curve.
Thermal conductivity of U–6Zr alloy was determined by Tran-
sient Plane Source (TPS) technique [30–32]. It is based on a spe- 2.8.2. Diffusion couple experiments
cially designed Wheatstone bridge with TPS sensor as one of the Discs of samples were cut from U–6Zr alloy and T91 and then
arms. The TPS unit was calibrated by the Inconel 600 standard were metallographically polished. For the diffusion couple experi-
sample from room temperature to 700 °C. ment, a disc of U–6Zr alloy was sandwiched between two discs of
Two identical cylindrical U–6Zr alloy samples of diameter T91 grade steel. The couples between U–6Zr and T91 are referred
20 mm and height 10 mm were specially prepared for this experi- to as U–6Zr/T91 couples. The components of the diffusion couples
ment following induction melting. The surfaces of the samples were were kept inside the fixtures made of Inconel 600. Ta foil was used
polished to have good thermal contact with the TPS sensor and to to prevent any chemical reaction between samples and fixture. The
minimize contact resistance. The sensor is sandwiched between fixtures containing these samples were encapsulated in quartz
two halves of the sample. Mica sensor of radius 3.189 mm was used tube under helium atmosphere. The diffusion couples were an-
for measurement. The samples were heated in high purity argon nealed in a resistance heating furnaces maintained at 973 K and
atmosphere to avoid oxidation. The samples were kept 30 min at 1023 K for up to 500 h. Subsequent to annealing, the couples were
each set temperature before the measurement to avoid tempera- sectioned using a slow speed diamond cutting wheel. The exposed
ture drift inside the samples. At each set temperature 2–3 measure- cross sections were metallographically polished to 1 lm surface
ments were programmed. A Keithley 2400 source meter supplies a finish. The extent of reaction and phases formed at the interface
constant voltage across the bridge. Before the measurement, the were analyzed by scanning electron microscope (SEM). Elemental
bridge is automatically balanced and as the resistance of the sensor analyses were carried out by using an energy dispersive spectro-
increases the bridge becomes increasingly un-balanced. A Keithley scope (EDS).
2000 digital voltmeter equipped with a scanner or multiplex card,
records the un-balance voltage. From these recorded voltages, it is 3. Results
possible to determine the temperature increase of the sensor and
consequently the thermal conductivity of the sample. 3.1. Equilibrium phases in U–6Zr alloy

2.7. Hot hardness U–Zr phase diagram has been studied extensively because of its
immense application particularly in the nuclear field and is shown
Hot hardness measurements were carried out using a hot hard- in Fig. 1 [33]. According to the phase diagram, at room temperature,
ness tester (Nikon, Model QM) with the help of a diamond Vickers the equilibrium phases of U–6Zr alloy are aU and intermetallic d

Table 2
Chemical composition of T91 steel (wt.%).

Cr Mo V Nb Al Ti Ni Cu Mn Si C N P S
9.161 0.882 0.207 0.079 0.008 0.003 0.197 0.068 0.368 0.209 0.099 0.0457 0.015 0.0013
4 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11

Fig. 1. Phase diagram of U–Zr system [33]. The composition of U–6Zr alloy is shown by a line.

phase having nominal composition of UZr2. The maximum solubil-

(021)

ity of Zr in aU is slightly greater than 0.5 at.% at 662 °C (935 K) and

decreases to 0.2 at.% at 500 °C (773 K). The d phase is formed due to
the peritectoid reaction a + c00 ? d at 617 °C (890 K). Between
(110)

617 °C and 662 °C (890–935 K), U–6Zr alloy exists as aU and c

phases. Above 662 °C (935 K), the equilibrium phases are bU and
(111)

c up to 693 °C (966 K). The maximum solubility of Zr in bU is

Intensity (a.u)

(002)

slightly less than 1.1 at.% at 693 °C (966 K). Above 693 °C (966 K),
it exists as a solid solution of cU and bZr.
(112)

(131)

3.2. Microstructure
(020)

(023)

The XRD analysis of U–6Zr alloy is presented in Fig. 2. The peaks

in the XRD data correspond to a phase of uranium. To know the
phases, the sample was further analyzed by SEM, which (Fig. 3a
and b) reveals that the sample is having a fully Widmanstatten
structure. The Widmanstatten plates are distributed throughout
the cross section of the fuel rod. However, no d phase was found 20 30 40 50 60 70
in this alloy. Thus, the microstructure of as-cast U–6Zr alloy con-
2θ (degree)
tains only supersaturated a phase in Widmanstatten morphology.
To measure homogeneity, the sample was scanned with the help of Fig. 2. XRD data of as-cast U–6Zr alloy. All the peaks correspond to a-U phase.
EDS at different locations and found that the distribution of Zr in-
side the sample was uniform.
due to the liquidus temperature. The first peak appears at 843 K,
3.3. Phase transformation temperature which does not correspond to the any transformation of equilib-
rium U–Zr phase diagram. To identify this particular transition at
The DSC curve of as-cast U–6Zr alloy on heating is shown in 843 K, further experiments were carried out using equilibrated,
Fig. 4. Five transition temperatures are observed in the DSC curve; directly quenched and annealed U–6Zr alloy samples.
i.e. 843, 943, 968, 1448 and 1488 K. The peak observed at 943 K
corresponds to eutectoid reaction (b $ a þ c00 ) and the peak at 3.4. Dilatometry
968 K is due to the invariant point of monotectoid reaction
(c $ b þ c00 ). The peak at 1448 K indicates the solidus temperature The dilatometric curve presented in Fig. 5, shows two distinct
and the appearance of small shoulder in the peak near 1488 K is phase transformations at 948 K and 973 K, respectively. The first
 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11 5

2.5

Thermal Expansion (%), (ΔL/L0)x100

2.0

1.5 973 K
948 K

1.0

0.5

0.0
400 500 600 700 800 900 1000 1100 1200 1300
Temperature (K)

Fig. 5. Thermal expansion curve of as-cast U–6Zr alloy, obtained from the
dilatometric experiments, for a heating rate of 5 K min1.

transition starting at 948 K is due to the invariant point of eutec-

toid reaction (b $ a þ c00 ) where a sharp change in length of the
sample occurs. The second transformation at 973 K corresponds
to the monotectoid reaction (c $ b þ c00 ). The second transforma-
tion gets completed at 985 K. After that the sample is transformed
to the purely c phase solid solution.
The dilatometric curve can be fitted by a third degree polyno-
mial as:373–923 K

ðDL=L0 Þ  100 ¼ 0:938  4:100  103 ðT=KÞ  5:010

 106 ðT=KÞ2 þ 3:450  109 ðT=KÞ3 ð2Þ

998–1273 K

ðDL=L0 Þ  100 ¼ 6:295  1:591  102 ðT=KÞ þ 1:630

 105 ðT=KÞ2  4:887  109 ðT=KÞ3 ð3Þ

where, L0 is the initial length (at 298 K) and DL is the difference be-
tween the instantaneous length (at any temperature T) and L0. The
Fig. 3. (a) SEM image of as-cast U–6Zr alloy and (b) Magnified micrograph clearly
average coefficient of thermal expansion of U–6Zr alloy has been
showing the Widmanstatten morphology.
determined and found to be 18.28  106 K1 in the temperature
range 298–823 K.

5 3.5. Specific heat

4
843 K
The measured specific heats values of U–6Zr alloy are given in
3 943 K Table 3 and shown in Fig. 6. The specific heat curve shows a phase
2 transition at 845 K. The measured C op ðTÞ values are best least-
968 K 1448 K squares fitted into following expressions:
Heat flow (μV)

1
1
C op ðJ K1 kg Þ ¼ 119:66 þ 0:095 ðT=KÞ  1:23  106 ðK=TÞ2
0
ð304 6 T=K 6 495Þ; ð4Þ
-1
1
-2 C op ðJ K1 kg Þ ¼ 8:61 þ 0:258 ðT=KÞ þ 10:71 ðK=TÞ2
-3 ð495 6 T=K 6 802Þ; ð5Þ
1488 K
-4
3.6. Thermal conductivity by Transient Plane Source (TPS) technique
-5
500 600 700 800 900 1000 1100 1200 1300 1400 1500
The thermal conductivity of U–6Zr alloy, measured by TPS
Temperature (K)
technique, from 298 K to 823 K is shown in Fig. 7. The thermal
Fig. 4. DSC curve of as-cast U–6Zr alloy sample for a heating rate of 5 K min1 conductivity of uranium determined by the same technique was
showing various peaks corresponding to the different phase transformations. also incorporated in the same figure for comparison. The thermal
6 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11

Table 3
Specific heat data of U–6 wt.%Zr as a function of temperature.

T/K C op (T)/J K1 kg1 T/K C op (T)/J K1 kg1 T/K C op (T)/J K1 kg1 T/K C op (T)/J K1 kg1

303.7 135.5 471.4 159.2 660.1 185.9 845.1 325.8

309.0 135.9 494.9 161.4 683.7 190.4 872.7 211.4
331.9 139.8 518.5 165.3 707.3 194.6 896.4 210.6
354.9 143.2 542.0 168.7 730.9 199.1 919.1 217.0
377.9 147.4 565.6 172.2 754.5 203.6 920.6 217.5
401.1 150.4 589.2 175.5 778.2 208.9
424.4 152.6 612.8 178.7 801.8 218.7
447.9 156.2 636.5 182.4 826.4 233.7

600
U-8.74 wt. % Zr (Matsui et al. [19])
500 U-5.87 wt. % Zr (Takahashi et al. [18])
U-5.4 wt. % Zr (Fedorov and Smimov [16])
C op(J K-1kg-1)

400

300

200

100

200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400
600
U-6 wt. % Zr (Present study)
500 U [36]
C op(J K-1kg-1)

Zr [36]
400 U-6 wt. % Zr (Additivity)

300

200

100

200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400
T /K

Fig. 6. Measured specific heat data for U–6Zr alloy as a function of temperature. Specific heat values of U, Zr and U–6Zr alloy determined by additivity rule are also shown in
the same figure (bottom). The literature data for specific heat for similar compositions are plotted against temperature (top).

conductivity of both the uranium and U–6Zr alloy increases with

50 increasing temperature and U–6Zr alloy is having lower thermal
U (Touloukian et al. [37,39])
U (Takahashi et al. [17]) conductivity than that of uranium. The temperature dependence
45 U (present study) of thermal conductivity of both uranium and U–6Zr alloy can be
U-5.87 wt. % Zr (Takahashi et al. [17])
expressed as a polynomial using the least squares fitting of the
Thermal conductivity (Wm K )
-1

U-6 wt. % Zr (present study)

U-5 wt. % Zr (Touloukian et al. [37,39]) data.
-1

40 U-6 wt.% Zr (Calculated by Eq. 9 [39])

U (fitted, present study) The thermal conductivity of uranium in the temperature range
U-6 wt. % Zr (fitted, present study) of 298–823 K can be written as:
35
k ¼ 18:890 þ 2:678  102 ðT=KÞ  2:302  106 ðT=KÞ2 ð6Þ
30 And for U–6Zr alloy the equation is:

k ¼ 13:795 þ 2:672  102 ðT=KÞ  2:492  106 ðT=KÞ2 ð7Þ

25
1 1
where, k is in W m K and T is in K.
20
3.7. Hot hardness

15 The room temperature hardness of U–6Zr alloy was measured

300 400 500 600 700 800 900 1000
with 300 g load and was found to be 380 ± 20 kg mm2. The hard-
Temperature (K) ness of this alloy gradually decreases with increasing in tempera-
Fig. 7. Thermal conductivity data of U and U–6Zr alloy as a function of temperature.
ture. The hardness (H) vs. temperature (T) plot for U–6Zr alloy in
The present data are compared with that of U and U–Zr alloys reported in the the temperature range of 298–973 K was fitted by a polynomial
literature. and the equation for the best fit is given below:
 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11 7

1000 been found that Zr has limited solubility in U(Fe,Cr)2 and U has
limited solubility in Zr(Fe,Cr)2.
748 K
3.8.2. Diffusion behavior of U–6Zr and T91
Microhardness (kg mm )
-2

Fig. 9a shows the microstructure of the interdiffusion layer

100 formed at the interface of the U–6Zr/T91 diffusion couple after
annealing at 973 K for 500 h. The magnified micrograph of the
interface is shown in Fig. 9b. The interdiffusion between the two
specimens resulted in the formation of three different layers at
the interface: a (U,Zr)(Fe,Cr)2 type layer of thickness around
10 U-6 wt. % Zr (present study) 10 lm on the T91 side, a 15–20 lm thick Zr depleted layer (having
U-5 wt. % Zr (Chubb et al. [46]) 94 at.% U, 4 at.% Zr, 2 at.% Fe) on the U–Zr alloy side and a thin Zr
U-6 wt. % Zr (fitted, Present study) rich layer (having 9 at.% U, 87 at.% Zr, 4 at.% Fe) of thickness 2–
3 lm between the above two. This Zr rich layer at the interface acts
as fuel-clad diffusion barrier. However, no melting has been ob-
1
served at this temperature. On the other hand, at 1023 K, U–6Zr/
300 400 500 600 700 800 900 1000 T91 couple reacted completely with each other causing a complete
Temperature (K) melt down of the clad in 100 h, and showed eutectic-melted micro-
structures consisting of U6Fe, U(Fe, Cr)2 and Zr(Fe, Cr)2 phases. The
Fig. 8. log (hardness) plotted against temperature (from room temperature to melted microstructure of this couple is shown in the Fig. 10.
973 K) for U–6Zr alloy. The transition point is also shown in the figure.

4. Discussion
H ¼ 1592:19  8:71ðT=KÞ þ 2:21  102 ðT=KÞ2  2:46
The performance of nuclear fuel in power reactors is largely
 105 ðT=KÞ3 þ 9:61  109 ðT=KÞ4 ð8Þ
dependent on its thermophysical properties and their changes
where H is in kg mm2 and T is in K. with temperature and irradiation. Experimental data on out-of-pile
The variation of hardness with temperature is plotted in Fig. 8 properties such as melting point, thermal conductivity, and ther-
on a semi-logarithmic scale for U–6Zr alloy. The curves consist of mal expansion are required for fuel design, performance modeling
two straight lines with different slope intersecting at around and safety analysis. The variables that influence the out-of-pile
748 K. In each temperature range, the hardness temperature-rela- properties of fuel are composition, temperature, density, micro-
tion obeys the Ito-Shishokin equation, viz., H = A exp(BT). The structure etc. These properties also change with irradiation. Hence
hardness falls by about 50% of its value at room temperature on knowledge of these properties and their change with temperature
reaching the transition temperature, which is located at around and irradiation are essential for fuel design and safety analysis.
0.52 Tm. The values of A, B and transition temperature are given As mentioned earlier, the XRD and metallography results show
in Table 4. The hardness values fall drastically above the transition that the as-cast U–6Zr alloy contains only supersaturated a phase
temperature. in Widmanstatten morphology and no d phase was found. It has
been reported in the literature [22] that the as-cast sample of U–
3.8. Chemical compatibility with T91 cladding 5 wt.%Zr alloy prepared by vacuum induction melting has a lamel-
lar structure. During cooling, the products of monotectoid reaction,
3.8.1. Eutectic reaction temperature between U–6Zr alloy and T91 i.e. b and c00 , transform into a phase keeping the lamellae intact.
grade steel However, U–6Zr alloy used in the present investigation does not
For reliable operation of a fast breeder reactor, the fuel elements contain lamellar structure. The alloy used in the present case was
must be resistant to breaching in overpower transients. From the prepared by following injection casting method, in which the solid-
standpoint of accident transients, the most important factor is ification of melt is faster as compared to the convention vacuum
the accelerated rate of cladding attack once eutectic liquefaction induction melting. The faster rate of solidification leads to the for-
forms at the interface. Fuel-cladding chemical interaction is a com- mation of mainly Widmansttaten morphology.
plex, multicomponent diffusion problem, because of the number of The d phase may form in the U-rich U–Zr alloys during cooling
alloy components involved in the reaction. With steel cladding like by following the peritectoid reaction (a + c00 ? d) at 890 K. Accord-
T91, even in the simplest fuel alloys such as U–Zr, at least five ma- ing to Basak et al. [22] the peritectoid reaction is extremely slug-
jor constituents (Fe, Cr, Mo, U and Zr) participate in the diffusion gish in nature. In fact, Hills et al. [21] have found that up to
process. The chemical compatibility of uranium based alloy with 20 at.% Zr (10 wt.%Zr) only a phase is present at room tempera-
T91 cladding is an issue because of the low melting eutectic forma- ture. To understand the phase transformation behavior of the U–
tion as indicated in U–Fe phase system. The measured eutectic 6Zr alloy, a detailed experimental study was carried out with the
temperature between U–6Zr alloy and T91 was found to be help of DSC. As mentioned in Section 3.3, all the peaks in Fig. 4
995 K, which is almost equal to that of U–Fe system. The eutectic could be identified except the one at 843 K which does not corre-
melted microstructure is having three phases which were identi- spond to any phase transformation of equilibrium U–Zr phase tran-
fied as U6Fe, U(Fe,Cr)2 and Zr(Fe,Cr)2 with the help of EDS. It has sition. To identify this peak, a sample of U–6Zr alloy was heated at
1173 K and then directly quenched to water. The DSC result
(Fig. 11) of the directly quenched sample also shows the presence
Table 4 of this peak at the same temperature i.e. 843 K. Metallographic
Values of constants A and B of the relation H = A exp (-BT).
examinations of the quenched sample reveals that the microstruc-
A (kg/mm2) B  103 (K1) Transition temperature, TT (K) ture (Fig. 12) contains martensitic a0a (subscript a indicates acicular
AI AII BI BII morphology) phase. The rapid cooling of the sample during
quenching leads to the formation of martensitic plates in the
542 3.63  106 1.242 12.986 748
sample. Since the sample contains martensitic a0 , the peak
8 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11

Zr rich layer
(a) (U,Zr)(Fe,Cr)2 Zr depleted layer

(b)

T91 U-6Zr

Fig. 9. (a) The microstructure of the interdiffusion layer formed at the interface of U–6Zr/T91 couple after annealing at 973 K for 500 h. (b) Magnified micrograph of the
interface, clearly showing the formation of the various layers at the interface.

3
-1
5 K min
As cast

2 -1
5 K min
U(Fe, Cr)2 Direct quenched
Heat flow (μV)

1
Zr(Fe, Cr)2 1 K min
-1
1 h holding at 823K
0
U6Fe
-1
1 K min Annealed at 823 K for 24 h
-1

-2
820 840 860 880 900 920
Temperature (K)

Fig. 11. The DSC curves of U–6Zr alloy samples after the sample was subjected to
Fig. 10. The microstructure of U–6Zr/T91 diffusion couple annealed at 1023 K for the different treatments to investigate the peak observed at 843 K. The peak at
100 h showing the melted structure. 843 K was observed for both the as-cast as well as directly quenched sample. The
peak height gets diminished when the sample was isothermally held at 823 K of 1 h
and finally the peak was vanished when the sample was annealed at 823 K for 24 h.
corresponding to 843 K can be attributed to the martensitic a0a to Therefore, peak at 843 K can be attributed to the martensitic to non-martensitic
non-martensitic a transformation. Similar type of experiments transformation.
was also carried out by Basak et al. [24] for the directly quenched
U–2 wt.%Zr alloy. They reported that the peak observed at 862 K in decomposition of d phase (i.e. a þ c00 $ d), which indicates the
the DSC during heating is due to a0a ? a transformation [24]. For sample does not contain d phase even after the annealing at
further clarification of this peak at 843 K, the as-cast sample was 823 K for 24 h. Hence, it is very clear that the peak at 843 K is
isothermally held at 823 K for 1 h inside the DSC furnace and result due to the martensitic to non-martensitic transformation.
indicates a similar peak with lower peak height at 843 K (Fig. 11). The thermal expansion behavior of U–10 wt.%Zr and U–20
The as-cast sample was then annealed at 823 K for 24 h and then wt.%Zr alloys have also been reported in the literature [34,35] in
furnace cooled to room temperature. The DSC result of the an- the temperature range 293–900 K. The dilatometric study of U-rich
nealed sample as shown in Fig. 11, does not contain any peak at U–Zr alloys (2, 5, 7, 10 wt.% of Zr) have been carried out by Basak
843 K. Thus, the peak at 843 K in the DSC curve was observed for et al. [22]. Rough [14] reported the average coefficient of thermal
both the as-cast as well as directly quenched sample. The peak expansion of U–59 wt.%Zr and U–78 wt.%Zr alloys in the tempera-
height gets diminished when the sample was isothermally held ture range 293–823 K as 8.2  106 and 10.9  106 K1, respec-
at 823 K of 1 h and finally, the peak was completely vanished when tively. The thermal expansion data of U–Zr alloys reported in the
the sample was annealed at 823 K for 24 h. This indicates that the literature are given in Table 5 along with the present data for
as-cast sample is partially martensitic in nature. The isothermal comparison. It has been reported that with increase in Zr content,
holding of 1 h at 823 K partially transforms the martensitic to cohesive energy in orthorhombic crystal structure is gradually
non-martensitic phase and it becomes fully non-martensitic when reduced with temperature and as a result the thermal expansion
the sample was annealed at 823 K for 24 h. The DSC result of the coefficient of a-phase increases with increase in Zr content in the
annealed sample does not show any peak near 890 K for the alloy [22]. The dilatometric curve of U–6Zr alloy does not reveal
 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11 9

1273 K. This is due to the formation of c solid solution in this tem-

perature range.
The specific heat of U–6Zr alloy increases with temperature as
shown in Fig. 6 and these values have been compared with that
of literature data [16,18–19] of uranium rich alloys. Fedorov and
Smirnov [16] measured molar heat capacities of four compositions
(5.4, 21, 37.47, 50.8, 75.1 wt.%Zr) of U–Zr alloy in the temperature
range 293–873 K using heating pulse calorimeter. Takahashi et al.
[18] also measured molar specific heat for U–Zr alloys (5.87, 17.11,
49.64, 79.5 wt.%Zr) in the temperature range 300–1100 K by laser-
flash calorimeter. Mitsui et al. [19] measured the specific heat of
U–8.74 wt.%Zr using direct heating pulse calorimetry in the tem-
perature range 289–1289 K. The literature and present data of spe-
cific heat of U–Zr alloys are given in Table 6 for comparison. These
specific heat values of uranium rich compositions have been com-
pared with that of present work along with that calculated from
Neumann Kopps (NK) rule in Fig. 6. For comparison, the specific
heat values of pure U and Zr [36] are also shown in the same figure.
The data of present study shows a reasonably good agreement
among the specific heat data of Fedorov and Smirnov [16] (U–
Fig. 12. The microstructure of directly water quenched (from 1173 K) U–6Zr alloy 5.4 wt.%Zr), Takahashi et al. [18] (U–5.87 wt.%Zr), Mitsui et al.
showing martensitic plates and prior c grain boundaries (indicated by arrows). [19] (U–8.74 wt.%Zr) and that calculated from NK rule for U–6Zr.
In the present investigation, the specific heat of U–6Zr alloy was
measured in the temperature range of 303–919 K. The phase tran-
any indication of peritectoid reaction near 890 K which is in good sition temperature obtained in the present study at 845 K (Fig. 6) is
agreement with the DSC results. The composition of U–6Zr alloy is due to the martensitic to non-martensitic transformation. Mitsui
very much close to the invariant point of monotectoid reaction et al. [19] reported a transition temperatures at 879 K in the spe-
(c $ b þ c00 ). Hence, the reaction starting (973 K) and end (985 K) cific heat measurement of U–8.74 wt.%Zr alloy. U–8.74 wt.%Zr alloy
points are close. Above 985 K, the dilatometric curve is very prepared by Mitsui et al. [19] contained a mixture of two-phase:
smooth and the slope does not change with temperature up to a-U and d-UZr2 and the transition at 879 K corresponds to the

Table 5
Thermal expansion of uranium–zirconium alloys.

Reference Wt.%Zr U–Zr composition Temp. range (K) DL/L0 (%) = a + b (T/K) + c (K/T)2 + d (T/K)3
a b  103 c  106 d  109
Touloukian et al. [34] and KAERI/CM-1313/2009 [35] 10 U0.7752Zr0.2248 293–900 0.424 1.658 1.052 1.115
20 U0.6052Zr0.3948 293–900 0.301 1.160 0.779 1.080
Basak et al. [22] 2 U0.9494Zr0.0506 323–923 0.381 1.2238 0.2793 0.9904
5 U0.8793Zr0.1207 323–923 0.539 2.181 2.10 2.087
7 U0.8358Zr0.1642 323–923 0.62 2.732 3.386 3.055
10 U0.7752 Zr0.2248 323–923 0.73 3.489 5.154 4.390
2 U0.9494 Zr0.0506 998–1173 7.775 24.05 21.05 6.69
5 U0.8793 Zr0.1207 998–1173 55.041 147.78 128.04 37.37
7 U0.8358Zr0.1642 998–1173 42.52 113.69 97.09 28.04
10 U0.7752 Zr0.2248 998–1173 25.252 66.69 54.41 15.18
Present study 6 U0.8572 Zr0.1428 373–923 0.938 4.10 5.01 3.450
998–1273 6.295 15.91 16.30 4.887

Table 6
The smoothed heat capacities of uranium–zirconium alloys.

Authors/years Wt.%Zr U–Zr composition Temp. range (K) Method C op (T)/J K1 kg1 = a + b (T/K) + c (K/T)2

a b c  106
Fedorov and Smirnov [16], 1968 5.4 U0.87Zr0.13 293–873 HPC 6.80 0.27 4.94p
21.0 U0.59Zr0.41 293–873 6.23 0.31 6.16p
37.47 U0.39Zr0.61 293–873 97.68 0.26 1.97p
50.8 U0.27Zr0.73 293–873 98.38 0.27 3.14p
75.1 U0.11Zr0.89 293–873 151.44 0.25 3.00p
Takahashi et al. [18], 1989 5.87 U0.86Zr0.14 300–850 LPC 55.10 0.14 2.21q
17.11 U0.65Zr0.35 300–850 59.18 0.16 2.74q
49.64 U0.28Zr0.72 300–850 85.12 0.22 3.98q
79.5 U0.09Zr0.91 300–850 HPC 153.07 0.20 2.64q
Matsui et al. [19] 8.74 U0.8Zr0.2 289–852 DSC 37.99 0.20 2.73p
Present study 6 U0.8572Zr0.1428 304–495 119.66 0.095 1.23
495–802 8.61 0.258 10.71

HPC = heating pulse calorimeter, LPC = Laser-flash calorimeter, DSC = Differential Scanning Calorimeter,
p = experimental data fitted in this study, q = experimental data up to 850 K was fitted in this study.
10 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11

decomposition of d phase (i.e. a þ c00 $ d) as per U–Zr phase dia- kZr ðTÞ ¼ 8:853 þ 7:082  103 T þ 2:533  106 T 2
gram. Since U–6Zr alloy used in the present investigation does
þ 2:992  103 T 1 ð298 6 T 6 2000 KÞ
not contain any d phase, the specific heat graph does not contain
any transition for decomposition of d phase. and kc,U is a thermal conductivity correction due to the alloying ef-
The thermal conductivity data of U–Zr alloys were reviewed and fect. kc,U is given by
summarized by Touloukian et al. [37]. The thermal conductivities
of U–Zr alloys, including metallic uranium and zirconium were re- kc;U ¼ 102:0 þ 200:1 X Zr  109:2 X 2Zr þ 9:435  103 T
ported by Takahashi et al. [17] from 300 to 1000 K. They calculated þ 3:459  105 T 2  0:02093 X Zr T ð12Þ
thermal conductivity from the thermal diffusivity data generated
by laser flash method. Thermal conductivity data by laser flash It has been observed that the results predicted by Eq. (9) are
method is estimated from the measured thermal diffusivity and very consistent with the recommended data compiled by Toulou-
heat capacity using the relation k = a Cp q, where k is the thermal kian [37]. The thermal conductivity has also been calculated for
conductivity, a is thermal diffusivity, Cp is the specific heat capac- U–6Zr alloy using Eq. (9) and the data was incorporated in Fig. 7.
ity, and q is the density. Hence, laser flash method requires heat The experimental result agrees well with the calculated data.
capacity data which is usually taken from the literature. The error The temperature dependence of hardness of U–6Zr is best rep-
from the specific heat data can lead to a lower accuracy in the ther- resented by the following relation of the type
mal conductivity data. In the present investigation thermal con-
H ¼ A expðBTÞ ð13Þ
ductivity is measured directly by TPS method. The thermal
conductivity data of uranium metal using TPS method agrees well where, constants A and B are called the intrinsic hardness (i.e. the
with those in the literature [17,37] over the temperature range value of hardness at 0 K) and softening coefficient, respectively.
investigated. Thermal conductivity of U–6Zr alloy is lower than The constants A and B have one set of values at low temperature
uranium metal which indicates that the addition of zirconium (AI and BI) and another at high temperature (AII and BII), suggesting
causes lowering in thermal conductivity of uranium. The composi- a change of mechanism [40–43]. The value of A at low temperature
tion dependence of the thermal conductivity of U–Zr system was is a measure of the inherent strength of the interatomic bond in the
also reported by Takahashi et al. [17]. According to them, d phase metal lattice and is related to the crystal structure and ‘thermal en-
alloys has the lowest thermal conductivity. However, U–6Zr alloy ergy of melting’ which is defined as the heat content of the liquid
used in the present investigation does not contain any d phase. metal at the melting point [43]. The constant B, which is the slope
Hence, the thermal conductivity data of U–6Zr alloy generated by of ln H vs. T plot, indicates the rate at which the lattice is weakened
TPS method in the present investigation was mainly for the a by the thermal energy and is closely related to the coefficient of
phase only. thermal expansion [44].
The thermal conductivity data of U–Zr alloys are given in the The mechanisms of softening in metals and alloys at low and
Table 7. Including all the U–Zr data from Touloukian [37] and high temperatures are widely discussed in the literature. At low
Takahashi [17], and the Zr data from Fink [38], a new correlation temperatures, below the transition temperature, hardness is a weak
has been developed for U–Zr alloy at ANL [39] which applicable function of temperature and plastic deformation occurs by ather-
to any Zr content. The relation is reproduced as follows: mal mechanisms. The major mode of deformation is by slip along
preferred planes and directions. At temperatures <0.1 Tm, the defor-
kZrU ¼ ð1  ð1  X Zr Þ1=2 ÞkZr þ ð1 mation is controlled by dislocation intersection with some contri-
 bution from impurity atoms. At temperatures between 0.2 and
 X Zr Þ1=2 X Zr kc;U þ ð1  X Zr ÞkU ð9Þ
0.5 Tm, a low value of activation energy has been reported. Conard
where XZr is weight fraction of Zr, kU is the thermal conductivity of [45] suggested that the plastic flow stress in a close packed lattice
uranium given by could be associated with impeded displacement of impurity atom
atmospheres interacting with dislocations. At high temperatures
kU ðTÞ ¼ 21:73 þ 1:591  102 T þ 5:907  106 T 2 ð255:5 (>0.5 Tm), hardness decreases strongly with increasing temperature
and deformation is thought to occur by a combination of dislocation
6 T 6 1173:2 KÞ ð10Þ
glide and dislocation climb. The activation energy for this process is
kZr is the thermal conductivity of Zr given by the same as the activation energy for vacancy diffusion.

Table 7
Thermal conductivity of uranium and uranium–zirconium alloys.

Reference Wt.%Zr U–Zr composition Temp. range (K) Method k (W m1 K1) = a + b (T/K) + c (T/K)2
a b  102 c  106
Touloukian et al. [37], and Kim [39] 0 U 293–1173 – 20.918 2.062 2.31m
1.5 U0.9618Zr0.0382 293–1073 19.185 0.741 15.0m
5 U0.8793Zr0.1207 293–1173 14.114 1.418 9.80m
20 U0.6052Zr0.3948 293–1173 6.008 1.577 6.90m
40 U0.3650Zr0.6350 293–1173 6.235 0.289 21.8m
70 U0.1411Zr0.8589 293–1173 10.376 0.940 21.4m
Takahashi et al. [17] 5.87 U0.86Zr0.14 300–1000 LF 23.198 0.873 11.6n
0 U 300–880 6.30 4.40 13.0n
17.11 U0.65Zr0.3 300–880 6.207 2.238 1.46n
29.34 U0.48Zr0.52 300–880 3.417 0.443 13.9n
49.64 U0.28Zr0.72 300–880 0.903 1.357 4.94n
79.5 U0.09Zr0.91 300–880 0.661 3.046 12.4n
Present study 0 U 298–873 TPS 18.890 2.678 2.302
6 U0.8572Zr0.1428 298–823 13.795 2.672 2.492

LF = Laser flash, TPS = Transient plane source, m = recommended data fitted in this study, n = experimental data was fitted in this study.
 S. Kaity et al. / Journal of Nuclear Materials 427 (2012) 1–11 11

Chubb et al. [46] reported hot hardness studies on U–Zr alloys (8) At 1023 K, U–6Zr/T91 couple reacted completely with each
with varying Zr content. For U–5 wt.%Zr, which is very close to other causing a complete melt down of the clad in 100 h,
the composition of present study, they reported hardness values and showed eutectic-melted microstructures of U6Fe,
of 347 and 118 kg mm2 at 298 K and 773 K, respectively. But at U(Fe, Cr)2 and Zr(Fe, Cr)2 phases.
973 K the hardness of the above alloy was 6.4 kg mm2. The trend
in values of hardness of the above alloy is very similar to the present
study. References
The chemical compatibility between fuel and clad material also
known as fuel-clad chemical interaction (FCCI) is of prime concern [1] G.L. Hofman, L.C. Walters, Mater. Sci. Technol. 10A (1994) 3.
[2] A. Riyas, P. Mohanakrishnan, Energy 35 (2008) 87.
because of formation of low melting eutectic which may sometime [3] J.H. Kittel, B.R.T. Frost, J.P. Mustelier, K.Q. Bagley, G.C. Crittenden, J. Van
limit the life of the fuel pin in a reactor. The measured eutectic Dievoet, J. Nucl. Mater. 204 (1993) 1.
temperature between U–6Zr alloy and T91 was found to be [4] D.E. Burkes, R.S. Fielding, D.L. Porter, D.C. Crawford, M.K. Meyer, J. Nucl. Mater.
389 (2009) 458.
995 K, which is almost equal to that of U–Fe system. The results [5] D.C. Crawford, D.L. Porter, S.L. Hayes, J. Nucl. Mater. 371 (2007) 202.
indicated that fuel-clad eutectic temperature has not been changed [6] M.V. Nevitt, J. Nucl. Mater. 165 (1989) l.
with the addition of 6 wt.%Zr in the fuel. Similarly, alloying ele- [7] C.E. Lahm, J.F. Koenig, R.G. Pahl, D.L. Porter, D.C. Crawford, J. Nucl. Mater. 204
(1993) 119.
ments of T91 grade steel i.e. Cr, Mo, V, Nb, etc. have no remarkable [8] L.C. Walters, J. Nucl. Mater. 270 (1999) 39.
effect on the eutectic temperature. These results concluded that [9] K. Devan, A. Bachchan, A. Riyas, T. Sathiyasheela, P. Mohanakrishnan, S.C.
the only contributors to the fuel-clad eutectic reaction are U of Chetal, Nucl. Eng. Des. 241 (2011) 3058.
[10] D.E. Burkes, C.A. Papesch, A.P. Maddison, T. Hartmann, F.J. Rice, J. Nucl. Mater.
the fuel and Fe of the cladding.
403 (2010) 160.
The fuel-clad chemical interaction between U and 6Zr alloy and [11] G.L. Hofman, M.K. Meyer, A.R. Ray, in: A. Travelli (Ed.), Proceedings of the 21st
T91 were studied by carrying out two diffusion couple experi- International Meeting on Reduced Enrichment for Research and Test Reactors
(1998 RERTR), ANL, San Paulo, Brazil, 1998, p. 12.
ments; at 973 K (below eutectic temperature) and at 1023 K (above
[12] D.G. Martin, J. Nucl. Mater. 152 (1988) 94.
eutectic temperature). The diffusion reaction at 973 K results in [13] R. Tewari, G.K. Dey, R.K. Fotedar, T.R.G. Kutty, N. Prabhu, Metall. Trans. 35A
formation of three diffusion layers at the interface i.e. (2004) 189.
(U, Zr)(Fe, Cr)2 layer, Zr rich layer and Zr depleted layer. The forma- [14] F.A. Rough, An evaluation on Zirconium–Uranium alloys, Report No. BMI-1030,
Metallurgy and Ceramics (M-3679), Battelle Memorial Institute, 16th ed.,
tion of similar type of diffusion layers was also reported in the lit- August 1955.
erature [25–28] for diffusion couple experiment between U– [15] A.A. Bauer, An evaluation of the properties and behaviour of Zirconium–
10 wt.%Zr alloy and ferritic steel clad. The diffusion reaction at Uranium alloys, Report No. BMI-1350, Metallurgy and Ceramics (TID-4500),
Battelle Memorial Institute, 15th ed., September 1959.
973 K is strongly retarded due to the formation of Zr-rich layer at [16] G.B. Fedorov, E.A. Smimov, Sov. J. Atom. Energy 25 (1968) 795.
the interface. However, reaction at 1023 K causes eutectic melted [17] Y. Takahashi, M. Yamawaki, K. Yamamoto, J. Nucl. Mater. 154 (1988) 141.
microstructure. Hence, the Zr-rich layer formed at the interface is [18] Y. Takahashi, K. Yamamoto, T. Ohsato, H. Shimada, T. Terai, M. Yamawaki, J.
Nucl. Mater. 167 (1989) 147.
failed to withstand above the eutectic temperature. [19] T. Matsui, T. Natsume, K. Naito, J. Nucl. Mater. 167 (1989) 152.
[20] G. Lagerberg, J. Nucl. Mater. 9 (1963) 261.
5. Summary and conclusions [21] R.F. Hills, B.R. Butcher, B.W. Howlett, D. Stewart, J. Nucl. Mater. 16 (1965) 25.
[22] C. Basak, G.J. Prasad, H.S. Kamath, N. Prabhu, J. Alloys Compd. 480 (2009) 857.
[23] C.B. Basak, R. Keswani, G.J. Prasad, H.S. Kamath, N. Prabhu, J. Alloys Compd. 471
Thermophysical properties of U–6Zr alloy at high temperatures (2009) 544.
have been investigated for the first time and the following conclu- [24] C. Basak, R. Keswani, G.J. Prasad, H.S. Kamath, N. Prabhu, S. Banerjee, J. Nucl.
Mater. 393 (2009) 146.
sions are drawn: [25] D.D. Keiser Jr., M.A. Dayananda, Metall. Mater. Trans. A 25A (1994) 1649.
[26] D.D. Keiser Jr., M.C. Petri, J. Nucl. Mater. 240 (1996) 51.
(1) The DTA curve of as-cast U–6Zr alloy showed a peak at 843 K, [27] C.T. Lee, H. Kim, T.K. Kim, C.B. Lee, J. Nucl. Mater. 395 (2009) 140.
[28] H.J. Ryu, B.O. Lee, S.J. Oh, J.H. Kin, C.B. Lee, J. Nucl. Mater. 392 (2009) 206.
which does not correspond to the any transformation of equi- [29] S.C. Parida, S.K. Rakshit, S. Dash, V. Venugopal, J. Solid State Chem. 179 (2006)
librium U–Zr phase diagram. This peak can be attributed to 212.
the martensitic a0a to non-martensitic a transformation. [30] S.E. Gustafsson, Rev. Sci. Instrum. 62 (3) (1991) 797.
[31] S.E. Gustafsson, International Patent Application No. PCT/SE89/00137.
(2) The expansion curves for U–6Zr showed two transitions at [32] Hot Disk Thermal Constants Analyser, Instruction Manual, Software version
948 K and 973 K. The average coefficient of thermal expan- 5.9, Hot Disk, Sweden, 2007.
sion has been determined and found to be 18.28  106 K1 [33] R.I. Sheldon, D.E. Peterson, Bull. Alloy Phase Diagrams 10 (2) (1989) 165.
[34] Y.S. Touloukian, R.K. Kirby, R.E. Taylor, P.D. Desai, Thermal Expansion,
in the temperature range 298–823 K.
Thermophysical Properties of Matter, vol. 12, IFI/Plenum, New York-
(3) The thermal conductivity, measured by TPS method, of both Washington, 1975.
the uranium and U–6Zr alloy increases with increasing tem- [35] A study of radiation effects on MA containing metallic SFR fuel, KAERI/CM-
perature and U–6Zr alloy is having a lower thermal conduc- 1313/2009.
[36] I. Barin, Thermochemical Data of Pure Substance, third ed., VCH, Weinheim,
tivity than that of uranium. 412, 1995.
(4) The specific heat versus temperature curve shows a phase [37] Y.S. Touloukian, R.W. Powell, C.Y. Ho, P.G. Klemens, Thermophysical Properties
transition at 845 K. of Matter, vol. 1, IFI/Plenum, New York, 1970.
[38] J.K. Fink, L. Leibowitz, J. Nucl. Mater. 226 (1995) 44.
(5) Hardness-temperature plot showed two straight lines with [39] Y.S. Kim, G.L. Hofman., AAA Fuels Handbook, Argonne National Laboratory.
different slopes which intersect at around 748 K [40] H.D. Merchant, G.S. Murthy, S. Bahadur, L.T. Dwivedi, Y. Mehrotra, J. Mater. Sci.
(6) The eutectic reaction temperature between U–6Zr alloy and 8 (1973) 437.
[41] A.G. Atkins, D. Tabor, Proc. Roy. Soc. A 292 (1966) 441.
T91 steel system is almost equal to that of U–Fe, i.e. 995 K. [42] T.R.G. Kutty, K. Ravi, C. Ganguly, J. Nucl. Mater. 265 (1999) 91.
(7) The interdiffusion between U–6Zr and T91 at 973 K for 500 h [43] J.H. Westbrook, Trans. Am. Soc. Met. 45 (1953) 221.
results in the formation of a (U, Zr)(Fe, Cr)2 type layer on the [44] E.R. Petty, H. O’neill, Metallurgia 63 (1961) 25.
[45] H. Conard, J. Met. (1964) 582.
clad side and a Zr depleted layer on the fuel side. A Zr rich [46] W. Chubb, G.T. Muehlenkamp, A.D. Schwope, A hot-hardness survey of the
layer was found between these two which acts a fuel clad Zirconium–Uranium system, Report No. BMI-833, Metallurgy and Ceramics,
interdiffusion barrier. Battelle Memorial Institute, May 1953.