loca acceptance criterialoca acceptance criteria-----considerations and questions raised from a...

C. GRANDJEANIRSN, Cadarache

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

LOCA Acceptance Criteria-----------

Considerations and QuestionsRaised from a Review

of Past and Recent R&D Work

C. Grandjean NRC / IRSN Meeting, January 24-25, 2007 2/30

In 1967, the Ergen Task Force concluded :

“The analysis of a LOCA requires that the core be maintained in place and essentially intact to preserve the heat-transfer area and coolant-flow geometry.”

In 1971, the USAEC promulgated the Interim Acceptance Criteria (IAC) for ECCS, in which Criterion 3 states that:

“The clad temperature transient is terminated at a time when the core geometry is still amenable to cooling, and before the cladding is so embrittled as to fail during or after quenching.”

ECCS Rule-Making Hearing (1972-1973)

Current ECCS acceptance criteria (10 CFR50.46)

Opinion of the Commission, Docket RM-50-1, December 28th ,1973:“In view of the fundamental and historical importance of maintaining core coolability, we retain this criterion as a basic objective, in a more general form than it appeared in the IAC.

BACKGROUND


In France, within current considerations for a future revision of the

LOCA acceptance criteria, the original objective of maintaining core

coolability should be kept as main basis

and possibly complemented with additional concerns

• dispersal of fuel particles in the reactor primary circuit,

• radiological release,

• structural material integrity,

• others ?

New Basis for a Revision of the Acceptance Criteria ?


In LOCAs there are two modes in which the coolability of the core may be impaired :

Ductile mode : clad ballooning leading to partial blockage of the FA channels

adressed in 10CFR50.46 by: (b)(4) : Coolable geometry: Calculated changes in core geometry shall be

such that the core remains amenable to cooling.(b)(5) : Long-term cooling

Brittle mode : Loss of cladding integrity upon quench and post-quench loads

adressed in 10CFR50.46 by the well-known embrittlement limits:(b)(1) : Peak cladding temperature : PCT < 2200°F (1204°C).(b)(2) : Maximum cladding oxidation : ECR < 17%.

so as to keep the fuel rod structure in place and essentially intact

Maintaining the Core Coolability


Coolability of Partially Blocked Assemblies under LOCA Conditions

Current ECCS acceptance criteria (10 CFR50.46) requires :§50.46 (b)(4). Coolable geometry +(b)(5) Long-term coolingCalculated changes in core geometry shall be such that the core remains amenable to cooling.

Although a quantitative limit is not associated to the above criteria, the requirement is generally considered as met in consideration of :

- the results of FEBA/SEFLEX experiments (coolability of blocked arrays) showing that a 90% blockage keeps coolable

- the maximum flow blockage ratio, inferred from NUREG-630 review (~71%)

value deduced from NUREG-630 burst strain curves, recognized later as non conservative (see next slide)


Coolability of Partially Blocked Assemblies under LOCA Conditions

Slow-ramp burst strain correlation from NUREG-630 and data reported after NUREG-630 was published(R. Meyer, ANS Meeting, Sun Valley, 1981)


COOLABILITY OF BLOCKED REGIONS. Main findings from SOAR

Experiments with longer blockages (THETIS) have shown that the maximum tested ratio of 90% may not necessarily be the most penalizing case for axially extended balloons.

Experimental results of flooding experiments with partially blocked rod arrays (FEBA/SEFLEX, THETIS, ACHILLES, CEGB, SCTF, FLECHT-SEASET)have shown that blockages, even of high ratio (90%) but of moderate length (<10 cm), do not impair the bundle coolability under LOCA reflood.

Analytical simulations confirm that significant increases in blockage wall temperatures, that may threaten blockage coolability, require a high blockage ratio (> 80%) and a long extension (> 15 cm) of the blockage.

The coolability of blocked bundles with fuel relocated in clad balloonsremains an open question, particularly for long balloons and low reflood rate, for which the blockage ratio still coolable might be less than the widely admitted value of 90% derived from FEBA/SEFLEX tests results.


Pending questions :

1. What is the maximum flow blockage ratio in a blocked assembly that can still be amenable to cooling, with taking account of a relocation of fragmented fuel in the ballooned claddings ?

2. What is the maximum flow blockage that can be obtained in an assembly of irradiated fuel rods, and what is the available margin with respect to Q1

The lack of current quantitative criteria rises questions on:

How to characterize a coolable situation in specific experiments ?

Which quantitative (revised) criteria would be appropriate to specify the ECCS calculated performance with respect to the coolability of blocked assemblies under LOCA scenarios ?

Maintaining Core Coolability - Ductile Mode


RATIONALE FOR EMBRITTLEMENT CRITERIA in 10 CRF50.46

Maintain coolable geometry

Keep fuel pellets inside the cladding

Preclude the cladding fragmentation or break in several pieces

Retain some ductility in the cladding

Limit cladding oxidation and temperature

PCT and ECR limits (2200°F, 17%) derived from Hobson’s ring compression tests on oxidized and quenched samples of unirradiated Zy cladding

No effects of BU were considered

Maintaining Core Coolability - Brittle Mode

Basic requirement is to preclude clad failure upon quench and post quench loads at any location, including the ballooned and bursted region


Total oxidationAs originally formulated in 50.46, the embrittlement criteria applied implicitly to an initially non oxidized cladding, thus to a fresh fuel rod.“…total oxidation means the total thickness of cladding metal that would be locally converted

to oxide if all the oxygen absorbed by and reacted with the cladding locally were converted to stochiometric zirconium dioxide.”

In NRC Information Notice 98-29, as an interim measure to account for the pre-corrosion effect, “total oxidation” was considered differently:“ The acceptance criterion in 10CFR50.46(b)(2) requires that the calculated maximum total oxidation of the cladding not exceed 0.17 times the total thickness of the cladding before oxidation. Total oxidation includes both pre-accident oxidation and oxidation occurring during a LOCA.


Modifications to embrittlement criteria

Revision of the criteria in 1988(based on “Compendium of ECCS Research for Realistic LOCA Analysis : NUREG-1230)

• Allows to use, alternatively to Appendix K evaluation models, best-estimate evaluation models together with an assessment of calculational uncertainties

• Keep unchanged the previous limits in 10CFR50.46(b)


Main Findings from a State-of-the-Art Review of LOCA Research (I)


Thermal shock quench tests after 1973 (ANL, JAERI, EdF/IRSN) (1)

Significant margin with respect to ECR limit in tests with quench at oxidation temperature

0

10

20

30

40

50

60

70

80

90

100

900 1100 1300 1500 1700 1900

Eq

uiv

alen

t C

lad

din

g R

eact

ed (

%)

Oxidation Temperature (°C)

Nonirradiated Zircaloys

Direct Quenching afterOxidation at High Temperature

INTACTFAIL

17% limit

Chung & Kassner

Hesson

Scatena

Grandjean & Lebuffe1204°C limit





Influence of quench temperature and cooling rate prior quench

Slow cooling in β → α domain increases the quench bearing limit





Influence of an initial corrosion, H pre-charging or prior irradiationLow or no influence on quench bearing limit under unrestrained conditions

Failure map under no restrained condition for as-received and pre-hydrided claddings (JAEA)(ECR from B-J equation , 2 sided oxidation on ballooned cladding)





Influence of an initial H content under axial restrained conditions (A)

Failure map under fully restrained condition for as-received and pre-hydrided claddings (JAEA)





Influence of an initial H content under axial restrained conditions (B)

Conservatism of the 17% ECR criterion for quench of as-received or irradiated cladding under “realistic” axial loading conditions

Failure map under controlled load conditions for two hydrogen concentration levels (JAEA)


Main Findings from a State-of-the-Art Review of LOCA Research (II)


Resistance to 0.3 J impact tests (ANL)

Ballooned cladding

Still adequacy of the 17% ECR criterion

Unballooned cladding

Conservatism of the 17% ECR criterion

(impact on opposite side of burst opening)

0

10

20

30

40

50

60

70

900 1000 1100 1200 1300 1400 1500

failed

intact

Eq

uiv

alen

t C

lad

din

g R

eact

ed (

%)

Isothermal Oxidation Temperature (°C)

Zircaloy-4 tube filled withalumina pellets,OD 10.9 mm, WT 0.635 mm,burst and oxidized in steam

Impact testat 23°C

H 20-2200 wppmafter quench

Chung & Kassner 1980NUREG/CR-1344


Main Findings from a State-of-the-Art Review of LOCA Research (III)


Post oxidation and post quench ductility tests (ANL, JAERI, CEA, UJP,…)

Post-oxidation ductility of undeformed Zy4 cladding tubes

0.0

0.50

1.0

1.5

2.0

0 4 8 12 16 20 24 28 32 36

To

tal

De

fle

cti

on

(m

m)

Equivalent Cladding Reacted (%)

oxidized in steamat 1100°C

1204°C

1315°C

1400°C

Zircaloy-4OD 10.9 mm, WT 0.635 mm6.35-mm-long ringcut from preoxidized tube

H < 130 wppm

Slow ring compression0.043 mm/s at 23°C

Chung & Kassner 1980NUREG/CR-1344

ANL tests 1980, unirradiated (H< 130 wppm), no quench

Ductility retained (deflection > 2 mm)for ECR (measured) < 17% and T< 1315°C

Recent ANL tests, irradiated HBR samples

Ductility retained ( offset strain > 3%)for ECR-CP < 8% without quench

threshold unclear with 800°C quench




Post oxidation and post quench ductility tests

Post-oxidation ductility of ballooned and burst cladding tubes

ANL Diametral Compression Tests (0.043 mm/s at 300 K) on Ruptured Zircaloy-4 Cladding That Survived Thermal Shock and Impact Loads. Total Deflection to Maximum Load as function of ECR parameter (from NUREG/CR-1344 data)

0

1

2

3

4

5

6

7

8

9

0 5 10 15 20 25 30 35 40 45 50

ECR (%)

To

tal D

efle

cti

on

to

Max

imu

m L

oad

(m

m)

brittle





Influence of transient H pickup from inner side of a burst cladding

Distribution of H content, ID oxide layer thickness and total deflection at 100°C

of Zy4 rings sectioned from burst region (JAEA, 1981)

Lowest ductility at the top and bottom necks with highest concentration of H





Influence of quench temperature and cooling rate prior quench (ANL, CEA) 1

Cooling rate influences β→α’ transformation, microstructure and hydride precipitation

N. Waeckel, ANL LOCA Meeting, June 2006, Copyright EDF/CEA





Influence of quench temperature and cooling rate prior quench (ANL, CEA) 2

Cooling rate influences β→α’ transformation, microstructure and hydride precipitation

V. Maillot, J.C. Brachet, ANL LOCA Meeting, June 2006, Copyright CEA





Oxygen diffusion from an initial corrosion layer at T<1000°C (UJP)

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

Residual ductility at RT after steam oxidation at 950 °Con samples pre-corroded at 425 °C in steam up to 50 μm

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

ECR cannot account for the partial dissolution of the pre-oxide layer in the underlying prior β-Zr(influenced by T and H content near the oxide/metal interface)


Main Findings from a State-of-the-Art Review of LOCA Research (IV)


Integral tests (CEA, ANL, JAEA) (1)

Phebus LOCA Test 219 (CEA), 25 rods bundle, fresh fuel

Rod 18, exposed to Tmax ~1330°C, was found fragmented despite ECR ≈ 16%

Likely due to a deleterious bundle effect : assembly constraints due to bending of rods and possibly rod-to-rod impact duringreflooding.




Integral tests (ANL, JAEA, CEA) (2)

ANL : ICL Tests on high BU Limerick BWR single rods (56-57 GWd/MTU)

From the 4 in-cell tests on irradiated rods, two (ICL#3, #4) were conducted with a 300 s hold at 1204°C, cooling at 3°C/s to 800°C and quench (unconstrained).

Burst occurred at 730 and 790°C respectively

Maximum calculated ECR was ~ 21% and 20%

Both LOCA integral samples remained intact during and following the quench

Samples were not subjected to post-quench ductility tests (bend test)




Integral tests (ANL, JAEA, CEA) (3)

JAEA : Tests on irradiated PWR single rods (39-44 GWd/MTU)

with 540 N axial constraint

Initial hydrogen concentration (ppm)

540N

0

10

20

30

40

0 500 1000 1500

EC

R (

%)

Initial hydrogen concentration (ppm)

540N

0

10

20

30

40

0 500 1000 1500

EC

R (

%)

Irradiated claddingSurvived Fractured

Unirradiated claddingSurvived Fractured

Irradiated claddingSurvived Fractured

Unirradiated claddingSurvived Fractured

failure boundary is not significantly reduced by PWR irradiation at the

examined moderate BU level( 15-25 µm oxide, 120-210 ppm H )


Summary of Main Results of LOCA Researchwith Respect to Embrittlement Limits (1)


Resistance to quench thermal shock

Conservatism of the 17% ECR criterion for quench of as-received or irradiated cladding under unconstrained and “realistic” axial loading conditions

Resistance to 0.3 J impact

Conservatism of the 17% ECR criterion for unirradiated unballooned cladding and still adequacy for ballooned cladding


Summary of Main Results of LOCA Researchwith Respect to Embrittlement Limits (1)


Retention of a residual post-quench ductility

Ductile-brittle transition ECR-CP threshold > 17% for unirradiated unballooned cladding

Low ECR threshold for high BU Zircaloy undeformed cladding due to hydrogen influence (ECR threshold < 5% for the HBR quenched sample)

Very low ECR threshold for irradiated cladding after oxidation at T<1000°C due to O diffusion from initial oxide layer into metal sublayer

(to be confirmed !)

No practical limit can be derived to ensure ductility retention in balloons due to the very large H content absorbed on ID upon secondary hydriding


Is it necessary to preclude clad fragmentation upon quench and post quench loads at any location, including the ballooned and bursted region?


Pending Questions

If no, it will be needed :

to define new requirements on the extent of acceptable fragmentation,

to evaluate its impact on radiological release and integrity of structures

If yes, it appears that a ductility retention in ballooned regions cannot be ensured

Which requirement to be applied to ballooned regions ?

Can the requirements be different in ballooned and unballooned parts of the cladding ? Upon which physical justification ?


If “some” clad fragmentation cannot be precluded in ballooned regions, which requirement should be chosen for the other regions :

Which representative loads to be considered for a reasonably conservative prevention of clad fragmentation :

quench thermal-shock under “realistic” restraint, or

post-quench mechanical loads ?

Which physical requirement :

Quench resistance under which restraint (axial / radial load) ?

Resistance to PQ loads, at which T ?(RT, 135°C…) :

ductility : brittle-ductile transition in compression or bending test

toughness : impact test

strength : tensile test

Is there an impact of the LOCA scenario ?Is a unique criteria appropriate for both LB and SB LOCAs ?

Is the existing data base sufficient or not ?


Pending Questions (cont.)


Selection of an adequate parameter to quantify an embrittlement limit

ECR (any combination of transient and initial oxidation…)Relates essentially to the amount of brittle phases (oxide, alpha layers)

Oxygen content in tandem with thickness in β layer(ex. Chung/Kassner handling limit : <0.7 wt% oxygen in > 0.3 mm β layer)

Relates to the properties of the remaining ductile phase

Others …


Pending Questions (cont.)

Capacity of current codes for LOCA safety analysis to calculate the selected embrittlement criteria

Modeling of oxygen diffusion in moving-boundary systems makes it possible to correctly evaluate the oxygen distribution in β layer

Improvements to include the recent findings for the effects of H andpretransient corrosion layer are under development at IRSN in an advanced diffusion code.

loca acceptance criterialoca acceptance criteria-----considerations and questions raised from a...

Documents