experimental and post-test calculation results of the ......quench-11 dec 8, 2005 water 15 ec...

J. Stuckert, FZK/IMF1/29

Computational and Experimental Studies of LWR Fuel Element Behaviorunder beyond Design Basis Accidents and Reflood Conditions:

Internat.Scientific and Technical Meeting, Moskva, Russia, July 27-28, 2009

Experimental and Post-Test Calculation Resultsof the Integral Reflood Test QUENCH-12

with a VVER-type Bundle

J. Stuckert, M. Große, M. SteinbrückForschungszentrum Karlsruhe




Zry rod

(2000 °C)

Water

Steam

Zr +2 H2O → ZrO2 + 2 H2 + 596 kJ

High-Temperature Materials Behavior andFlooding of an Overheated LWR Core and Hydrogen Source Term




bundle bottom

bundle top

shroud with heat insulation

QUENCH-12Test bundle assembling




AgInCd absorber, SARNET-Aerosol52WaterNov 7, 2007QUENCH-13

ACM series: M5 ® cladding41WaterJuly 2, 2008QUENCH-14

ISTC 1648.2; VVER48WaterSept 27, 2006QUENCH-12

EC LACOMERA-2; Boil-off, SARNET-Benchmark15WaterDec 8, 2005QUENCH-11

July 21, 2004

July 03, 2002

July 24, 2003

July 25, 2001

Dec 13, 2000

March 29, 00

Juni 30, 1999

Jan 20, 1999

July 07, 1998

Feb 26, 1998

Test date

EC LACOMERA-1; Air ingress50WaterQUENCH-10

EC COLOSS; As Q-07 but under steam starvation49SteamQUENCH-09

As Q-07 but without B4C absorber15SteamQUENCH-08

EC COLOSS; B4C absorber15SteamQUENCH-07

OECD ISP-45; As Q-05 but with water flooding42WaterQUENCH-06

As Q-04 but with pre-oxidation48SteamQUENCH-05

Steam, reference50SteamQUENCH-04

As Q-02 but flooding delayed40WaterQUENCH-03

EC COBE47WaterQUENCH-02

EC COBE52WaterQUENCH-01

RemarksFloodingrate, g/sQuenchmediumTest

Test Matrix (1998-2008)




Objectives of the QUENCH-12 test

• investigation of the effects of VVER materials and bundle geometry on core reflood

• comparison with the PWR bundle on the base of repeat of the test QUENCH-06 (ISP-45) scenario




Cross section of the VVER test columnCladding material: E110 (Zr1%Nb)

Cross section of the PWR test columnCladding material: Zry-4 (or M5)

1 unheated rodcladding Zry-4 ø10.75/9.3 mmcentral thermocouplepellet ZrO2 ø9.15/2.5mm

thermalinsulationZrO2 fibrethickn. 37 mm

2 removable corner rodsZry-4 ø6 mm

2 instrumented tubesZry-4ø6/4.2 mm

shroud Zry-4ø84.76/80mm

cooling jacket

20 heated rodscladding Zry-4ø10.75/9.3 mmW-heater ø6mmpellet ZrO2ø9.15/6.15mm

pitch 14.3 pitch 12.753 removable corner rodsZr1%Nb ø6 mm

3 instrumented tubesZr1%Nbø6/4.2 mm

18 heated rodscladding Zr1%Nbø9.15/7.73 mmW-heater ø4mmpellet ZrO2ø7.54/4.15mm

ShroudZr2.5%Nbø88/83.5mm

13 unheated rodscladding Zr1%Nb ø9.15/7.73 mmcentral thermocouplepellet ZrO2 ø7.54/4.15mm

Comparison of PWR and VVER test columns




Pretest modelling support:

1. SCDAP/SIM simulations: Paul Scherer Institute, Switzerland.

2. ICARE/CATHARE simulations: Kurchatov Institute, Mosccow,with support from IRSN Cadarache).




Test performance: el. power profile in accordanceto the pre-test calculations provided good reproductionof temperature history in comparison with the reference test QUENCH-06




Comparison of transient and reflood phases forQUENCH-14 (M5), QUENCH-12 (E110), QUENCH-06 (zry-4):

similar escalation before reflood

bundle temperature at hot elevation of 950 mm shroud temperature at hot elevation of 950 mm

T12 < T06 < T14 T06 < T12 < T14




QUENCH-12, quench phase: selected readingof the bundle thermocouples. Quench front propagation.

Cooling of the bundle during ~350 s (similar to Q-06 and Q-14)

300

500

700

900

1100

1300

1500

1700

1900

2100

7100 7150 7200 7250 7300 7350 7400 7450 7500 7550 7600Time, s

Tem

pera

ture

, K

TFSU 31/5/17 TFC 13/3/16 TFC 16/4/15 TFC 15/3/14 TFC 14/4/13 TFC 1/13 TFSH 3/2/12 TFSH 7/2/11 TFSU 16/4/10 TFSH 30/5/9 TFSU 15/3/6 TFSU 13/3/5 TFSH 24/5/4 TFSH 4/2/1

quench

950 mm

250 mm

650 mm




corner rod B after pre-test (800 °C, oxide layer thickness less of 5 µm)

corner rods withdrawn during the test:spalling of the outer skin of oxide layer due to breakaway oxidation.

D – withdrawn after pre-oxidation,F – withdrawn before reflood,

B – withdrawn after test.

D

F

B

QUENCH-12: withdrawn Zr1%Nb corner rods

β-Zr melted between880 and 1030 mm




-250

-50

150

350

550

750

950

1150

1350

500 700 900 1100 1300 1500 1700 1900 2100T, K

Elev

atio

n, m

m

Q6_7170s_trans_end Q12_7265s_trans_end

Axial bundle temperature profiles for QUENCH-06 and QUENCH-12on beginning and end of transient phase

beginning of transient end of transient

-250

-50

150

350

550

750

950

1150

1350

500 700 900 1100 1300 1500 1700 1900 2100T, K

Elev

atio

n, m

m

Q6_5950s_preox_end Q12_5972s_preox_end

E110break-away

E110break-away




900 mm:circumferential

and longitudinal cracksat the cladding;

nodular breakawaycorrosion

at the shroud

650 mm:spalled oxide scales

at shroud and cladding

400 mm:circumferential spalling

of the oxide layeron the surface

of fuel rod simulator cladding-400mm:

spalled oxide scalesas debris

at the bundle bottom

QUENCH-12: post-test videoscope observations of breakaway oxidation at different elevations of the bundle

Tmax

Tmin




β-Zr

α-Zr(O)

ZrO2

β-Zr

α-Zr(O)

ZrO2

Comparison cross-sections of QUENCH-06 and QUENCH-12at elevation 550 mm with ZrO2 thickness ~40 µm

Q12 cladding: spalling of oxide scalesdue to breakaway effect

Q12: rubble on spacer gridconsists of spalled cladding scales

and fragments of partially oxidized cladding

Q06 oxidized cladding

Q06 cross-section




Comparison cross-sections of QUENCH-06 and QUENCH-12 at elevation 750 mm with ZrO2 thickness ~80 µm

Q06 cross-section;few bundle deformation

Q12 cross-section;moderate bundle deformation

β-Zr

α-Zr(O)

ZrO2

β-Zr

α-Zr(O)

ZrO2

Q12 cladding: spalling of oxide scalesdue to breakaway effect

Q06 oxidized cladding




Comparison of averaging oxide cladding thicknessesfor QUENCH-06 and QUENCH-12 at all elevations

0

100

200

300

400

500

600

700

800

900

1000

550 650 750 850 950 1050 1150Bundle elevation, mm

ZrO

2 th

ickn

ess,

µm

Q12_ZrO2 (1.56*exhausted metal) Q06_ZrO2

Degree of cladding oxidation is mostly lower for QUENCH-06 (Zry-4) in comparison to QUENCH-12 (E110)

E110 breakaway




Q-06Q12: H2 uptake in claddingmeasured by hot extraction

Comparison of hydrogen uptake by bundleduring QUENCH-06 and QUENCH-12

Q-12: H2 uptake in corner rods measured by neutron radiography

0

5

10

15

20

25

30

35

40

550 750 950 1150Bundle elevation, mm

Hyd

roge

n, a

t%

corner rod D (preox)corner rod F (tansient)corner rod B (end of test)

releaseof absorbedhydrogen

0

5

10

15

20

25

30

35

40

550 650 750 850 950 1050 1150Bundle elevation, mm

Hyd

roge

n, a

t%Q6_hydrogen absorbed in claddings; at%Q12_hydrogen absorbed in claddings; at%

QUENCH-12breakaway

(according totemperature

history)




Cooling during normal cleaning:- high flow rate- closed loop

Cooling after cleaning - low flow rate - open loop with connection to spent fuel pool - possibility of by-pass by bundle low head

decay heat for 30 bundle: 241 kW

Paks/Ungary cleaning tank incident in April 2003with destroying of 30 VVER fuel assemblies




0

200

400

600

800

1000

1200

1400

0 5000 10000 15000 20000 25000

0 1 2 3 4 5 6 7

0

200

400

600

800

1000

1200

1400

123 4 5 6

7

8

910

Time (h)

Time (s)

Cla

ddin

g te

mpe

ratu

re (o

C)

Breakaway long-term oxidation during the Pakscleaning tank incident (destroying of 30 VVER fuel assemblies)

Results of modeling (FRAP-T6 code)

breakawayregion




Collapse of VVER-bundles, strong hydrogenatedunder breakaway conditions




Other important phenomenon: melt formation and oxidation byQUENCH-06 and QUENCH-12 at elevation 950 mm

Q06 - bundle geometry intact->good coolability;

melt localized between pellet and ZrO2 layer Q12 – bundle geometry destroyed->

declined coolability;molten pools formation inside rod clusters




Melt formation and oxidation byQUENCH-06 and QUENCH-12 at elevation 950 mm

Q12: molten pool between rods,precipitates formation due to melt oxidation in steam

Q06: melting of cladding metal layer,precipitates formation on cooldown




850 - 900 mm: melt formation in bundle at the position of melted corner rod B

neutronographyshows the deep holeat the placeof melted out β-Zr

A

C

E

1010 mm

1035 mminitial positionof rod B

QUENCH-12: melting of corner rod B




0,0

0,5

1,0

1,5

2,0

2,5

3,0

7000 7050 7100 7150 7200 7250 7300 7350 7400 7450 7500Time, s

Rat

e, m

g/s

0

10

20

30

40

50

60

Inte

gral

, g

rate_H2_Q12 rate_H2_Q6 intl_H2_Q12 intl_H2_Q6

reflood QUENCH-12

reflood QUENCH-06

Comparison of hydrogen releaseduring QUENCH-12 and QUENCH-06




Comparison of post-test cladding and melt structuresfor Q-08 and Q-12

Q-08, 860 mm Q-12, 850 mm




0

500

1000

1500

2000

2500

3000

-80 -60 -40 -20 0 20 40 60 80 100 120Time, s

Rel

ease

rate

, mg/

s; T

empe

ratu

re, K

Q08_H2 rate Q12_H2 Q08_TIT A/13 Q12_TIT A/13

flooding

Comparison of hydrogen release during refloodfor Q-08 test (oxidation before transient 270 µm; significant melt oxidation)

and Q-12 test (oxidation before transient 160 µm; minor melt oxidation)

Long duration of hydrogen productionduring the quench phase in case of significant melt oxidation (Q-08)




Despite the similar thermal response through almost the entire sequence,all of the calculations underestimated the oxidation.

QUENCH-12: calculated and measured hydrogen generation/J. Birchley, PSI/

C-P: Cathcart-PawelU-H: Urbanic-HeidrickP-C-S: Prater-Courtright-Schanz




CONCLUSIONS

• The QUENCH-12 experiment was carried out to investigate the effects of VVER materials and bundle geometry on core reflood, in comparison with the test QUENCH-06 (Zry-4-cladding) with western PWR geometry.

• During the pre-oxidation and transient phases the E110 cladding alloy is susceptible to breakaway oxidation within relative broad temperature range (1050 – 1300 K). Oxide scale of layered type showed spalling into sub-layers and loss of fragments, which were collected at spacer grids and bundle bottom.

• The breakaway oxidation was accompanied by hydriding of claddings and shroud with the result of material embrittlement, which can lead to a severe degradation of the bundle during quenching as was demonstrated in the Paks accident.




CONCLUSIONS (cont.)

• Melting point of claddings and shroud metal layer was reached at the peak temperature elevations of 950 – 1050 mm. Contrary to QUENCH-06, where melt was confined between pellets and the outer cladding oxide layer, the melt in QUENCH-12 has in places dissolved the surrounding oxide layer. As a result, non-coherent melt relocation, melt pool formation, and minor melt oxidation were observed.

• The total hydrogen production was 58 g (QUENCH-06: 36 g), 24 g of which were released during reflood (QUENCH-06: 4 g). Three possible sources of increased hydrogen production during reflood: 1) interaction of steam with new metal surfaces developed by spalling of oxide scales damaged during pre-oxidation phase because breakaway effect; 2) release of hydrogenabsorbed in metal by breakaway during pre-oxidation and transient phases; 3) moderate melt oxidation released into space between rods.

• The S/R5 code adequately reproduced the QUENCH-12 thermal transient. Both the Cathcart-Pawel and Sokolov correlations underestimated the oxidation kinetics, but preliminary analysis indicates that the comparison with QUENCH-06 is consistent with the change in the bundle configuration and cladding material; in particular breakaway effect may have enhanced the oxidation.

experimental and post-test calculation results of the ......quench-11 dec 8, 2005 water 15 ec...

Documents