C. GRANDJEANIRSN, Cadarache

NRC / IRSN Meeting, Bethesda, January 24-26, 2007

LOCA Analysis-----------

Physical Phenomenato be Taken into Consideration

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 2/11

FUEL RELOCATION

HIGH PRESSURE OXIDATION

SECONDARY HYDRIDING IN HIGH BU BALLOONED CLADDING

OXYGEN DIFFUSION FROM AN INITIAL OXIDE LAYER

Physical Phenomena to be Taken into Consideration

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 3/11

FUEL RELOCATION

Fuel relocation occurrence during LOCA is widely admittedas it has been observed in all burst tests with irradiated fuel rods

FR2 results

ANL integral test

HALDEN IFA-650

PBF/LOC (INEL)

FR-2 (KFK)

FLASH-5 (CEA)

ANL ICL-2 results

At low and medium burn-up : At high burn-up :

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 4/11

FUEL RELOCATION

Halden IFA-650.4

extended relocation of fuel

in the balloon and out of the test rod

Potential impact on clad balloon :

- increase of PCT and maximum ECR

Potential consequences of fuel dispersal :

- increase in radiological release

- possible impact on the integrity of structural materials

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 5/11

FUEL RELOCATION Pending issues (1)

Instant of relocation

close to burst time (FR-2/E3 and E4 tests, Halden IFA-650.4)

Filling ratio of clad balloon

Few available data from PBF/LOC and FR-2

Expected data from HALDEN/IFA-650 tests

What is the impact of fuel micro-structure ?MOX or Gd fuel (High BU structures)Doped fuel (-> in EPR)

Properties of relocated fuel

Granulometry : Expected data from HALDEN

Thermal conductivity

Heat exchange between clad and fuel fragments

Balloon filling rate (%) by relocated fragments

20 30 40 50 60 70 80Ballooning (%)

PBF/LOC-gammascanningPBF/LOC-micrographiesFR-2Upper boundIRSN sensitivity calculation

0

10

20

30

40

50

60

70

80

90

IRSN reference calculation

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 6/11

FUEL RELOCATION Pending issues (2)

Main consequences of the relocation process

Local increase in lineic and surfacic powers

Local decrease in fuel/clad gap (heat transfer increase)

Modification of steam access to the balloon inside

What is the impact of these consequences

On coolability around balloonned region with fuel relocation

On peak clad temperature and final oxidation ratio ?

On hydrogen uptake and consequences for quenching and post-quench embrittlement ?

EASY NOT SO EASY

these questions are particularly important for end-of-life MOX fuel

for which power generation is not reduced, unlike for UO2 fuel !

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 7/11

FUEL RELOCATION Calculations

Few studies published on the impact of relocation under LB LOCA conditions

Bergquist et. al (1978-79) ; INEL (1981) : limited simulation of PBF/LOC results

IRSN (2004) : sensitivity studies with CATHARE-2 code

IRSN (2006) : simulation of IFA-650.4 test with ICARE-CATHARE code

Assumptions on parameters

Filling ratio of the balloon

Thermal conductivity of fuel fragments

Heat transfer coef. fuel fragments / clad

Main results

Increase of PCT and ECR

Decrease of Zr-β layer thickness

High sensitivity to filling ratio

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 8/11

HIGH PRESSURE OXIDATION

Data base of results :

(Pawel et al., Park et al., Bramwell et al., Vrtilkova et al.)

from the existing results, the increase in oxidation kinetics (weight gain and layers thickness growth) remains moderate (factor < 2) in the SB LOCA pressure range (< 50 bar) and should not make an additional issue,

providing that the hydrogen absorption does not increase significantly

However, secondary hydriding in HP conditions has not yet been investigated

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 9/11

SECONDARY HYDRIDING IN HIGH BU BALLOONED CLADDING

ECR and H profiles in unirradiated rod OCL#11 ECR and H profiles in irradiated rod ICL#3

The peaks of hydrogen absorption appear located nearer the burst opening in irradiated rods, likely in relation to local optimal S/V values on clad inner side after fuel relocation, promoting H pickup.

Need to better quantify the hydrogen pickup axial profile in view of evaluating the embrittlementconditions in and around the ballooned regions

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 10/11

OXYGEN DIFFUSION FROM AN INITIAL OXIDE LAYER

At temperatures ≤1000 °C, oxygen diffusion through the monoclinic oxide is slower than oxygen diffusion into the metal. Therefore, an initial pre-oxide layer may partially dissolve in the underlying prior β-Zr, and result in clad embrittlement, before a significant rise of the transient ECR

Residual ductility ( 900°C )

0

10

20

30

40

50

60

0 2 4 6 8 10 12

ECRCath-Paw [%]

Res

idu

al d

uct

ilit

y at

RT

[%

]

2 um A1 2 um A210 um A1 10 um A235 um A1 45 um A2

(V. Vrtilkova, SEGFSM Meeting, Paris, April 2005)

Pre-oxidised standard ASTM Zircaloy-4 (950°C)

0

5

10

15

20

25

0 5 10 15 20 25

ECRCath-Paw [%]

Re

sid

ua

l du

cti

lity

at

RT

[ %

]

0 µm 2 µm 10 µm 50 µm

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 11/11

OXYGEN DIFFUSION FROM AN INITIAL OXIDE LAYER

An appropriate simulation of the partial dissolution of the initial oxide scale requires the calculation of O diffusion within the moving-boundary system, with taking account of the different diffusion properties in the dual phase (monoclinic/tetragonal) oxide

Such calculational tool is under development at IRSN (DIFFOX code)

900 C (35 um, 600 wppm H2)

0

5

10

15

20

25

30

35

40

45

50

0 900 1800 2700 3600

Time [s]

Th

ickn

ess

[um

]

1/1 Dx ZrO21/2 Dx1/3 Dx1/1 Dx Alpha1/2 Dx1/3 DxExp ZrO2Exp Alpha

α-Zr(O)

ZrO2 12

3

Simulation of an UJP test with the DIFFOX code