Corrosion Damage of Canister (or Container) in Nuclear Waste Management: Perspective and an Approach for

Model Abstraction

T. Ahn

Engineered Barrier Material Corrosion and Chemical Environment Workshop

Division of Spent Fuel Alternative Strategies (SFAS), Office of Nuclear Material Safety and Safeguards (NMSS)

U.S. Nuclear Regulatory Commission September 4 , 2012 (Revision, January, 2013)

1

Disclaimer

The NRC staff views expressed herein are preliminary and do not constitute a final judgment or determination of the matters addressed or of the acceptability of any licensing action that may be under consideration at the NRC.

2

Purpose

• Present an approach to abstract models for corrosion damage of canister (or container) for confinement assessment of radionuclides in nuclear waste management (management of spent nuclear fuel, SNF, and high-level waste, HLW)

• Elicit expert views on the approach:

- Initiation, repassivation (i.e., ceassation of continuous

propagation of initiated pit) and density of pits

- Pit-induced initiation of stress corrosion cracking (SCC)

- Single SCC crack propagation

- Possible maximum opening area of multiple cracks

3

Outline

• Background

• Example Pitting Corrosion and Pit Density: non-passive metals

• Example Pitting Corrosion and Pit Density: passive metals

• Pit-Induced Stress Corrosion Cracking (SCC): example calculation

• Single SCC Crack Propagation (observation)

• Limited Conditions for SCC

• Possible Maximum Opening Area of Multiple Cracks

- background

- Sandia National Laboratories (SNL) model

- SNL model application

• Summary

4

Background

Nuclear Waste Management Addresses: Long-Term Dry Interim Storage and Disposal at Various Potential Sites

- Extended (60-80 years beyond beginning of service) interim dry storage:

stainless steel and carbon steel in coastal area chloride-induced humid vapor corrosion inspection and mitigation (remediation, if needed)

- Long-term geological disposal under various geochemical conditions in different potential host rocks such as granite, clay, and salt:

variety of alloys such as carbon steel, copper, nickel-based alloy, titanium, and stainless steel safety during operation – similar to extended dry storage with possible inspection and mitigation safety during permanent isolation – aqueous corrosion from chloride, carbonate, caustic pH, or inhibitor no reliance on inspection and mitigation during isolation following repository closure to maintain safety

5

Background (continued)

- Low general corrosion rate: long-term passivity (for passive metal) or reducing

environment (for non-passive metal)

- Localized corrosion (mostly pitting and crevice corrosion) and SCC (including hydrogen-induced cracking): Precursory step

(1) Environmental and materials conditions – chemistry (chloride, carbonate, or caustic chemicals), aqueous condition (water or vapor), stress, temperature, and time: complex assessment of susceptible range of parameter values (2) inspection and mitigation - decrease the probability of occurrence; limited probability assessment available

6

Background (continued)

- Confinement failure

- Opening area for radionuclide release from damaged canister: an approach to

abstract models for SCC damage of canister (or container) in confinement

assessment of radionuclides.

The abstracted model can be used in a system model for the confinement

assessment. The abstracted model is developed with simple representations

utilizing a conservative and bounding approach. The conservative approach

could be used when there is the possibility of radionuclides be substantially

dispersed if a canister failed.

7

Example Pitting Corrosion and Pit Density: non-passive metals

8

Pitting factor is the ratio of the pit depth to the general corrosion depth, approaching 1 with time in carbon steel under reducing and non-chloride disposal environmental conditions (Jung, et al., 2011; Féron, et al., 2008). Variation of the pitting factor for carbon steel with the average depth of corrosion derived from long-term corrosion tests and short-term laboratory measurements.

(JNC, 2000)

Pitting is a degradation mechanism of the canister. It also serves as a precursory step for SCC.

Example Pitting Corrosion and Pit Density: passive metals

Density and area of pits are limited for passive metals.

(1) Repassivation Model (Shibata, 1983)

- Statistical process – low electrochemical potential (low initiation rate and high survival rate); high electrochemical potential (high initial rate and low survival rate); repassivation rate is independent of potential.

(2) Point Defect Model (PDM) (since Macdonald, et al., 1981;

Ahn et, al, Nuclear Technology in press, or Ahn, et al., 2010)

- Point defects (e.g., interstitial or vacancy) move and form voids.

- Pits nucleate at pitting potential and die quickly in chloride solution by repassivation (i.e., cessation) under open-circuit conditions.

- Exception: longer-term nucleation may occur at high pH and low ionic strength.

- Very long-term delayed pitting is not anticipated to occur during long-term disposal.

9

Example Pitting Corrosion and Pit Density: passive metals (continued)

(3) Literature pit density data under polarization (Passarelli , et al., 2005; Ahn, 1994): (0.1 – 100)/cm2

(4) With a thin water layer, low cathode capacity near a pit limits pit density and depth (Shukla, et al., 2011; Cui, et al., 2005).

(5) Pit growth models: propagation rate decreases with time.

(6) Pits can be precursors for SCC.

10

Pit-Induced Stress Corrosion Cracking (SCC): example calculation

• Pit size of stainless steel in Kure Beach (EPRI, 2005): (10 – 100) μm: uniform distribution [10, 100]

• Residual stress of stainless steel (Shirai, et al., 2011): (0 – 600) MPa: normal distribution [0, 600]

• Cumulative probability of stress intensification factor, K (MPa m1/2) = π1/2 x stress x (crack size)1/2

• Threshold K values of stainless steel with salt deposits (Shirai, et al., 2011; Kosaki, 2008; EPRI,

2005): 0.5 – 7.0 MPa m1/2

• Weld and incipient flaw size: mm (1000 μm range) in Poisson (SNL, 2007) and 50 μm (SNL, 2007) respectively

• Seismic-induced stress: yield stress (SNL, 2007)

[1 cm = 0.39 inch; 1MPa = 0.145 ksi; 1.0 MPa m1/2 = 0.91 ksi in1/2]

Probability 0.001 0.05 0.25 0.75 0.95

K 0.43 1.57 2.59 4.57 6.94

11

Single SCC Crack Propagation (observation)

The propagation of single crack, especially through the canister wall thickness, is an important canister degradation mechanism. The crack propagation behavior of a single crack may be different from that of multiple cracks.

• Stress along the thickness of the canister varies.

• Stress will be redistributed during crack propagation.

• Crack branching or tortuous crack path decreases the crack propagation rate, whether it is

inter- or trans-granular cracking.

• Plasticity: Jmaterial may increase more than Japplied with crack propagation, which may affect the slip or twin process for film rupture at crack tip.

• Residual stress decreases rapidly away from the weld (and/or heat affected zone, HAZ) area.

• Rigorous crack growth model and exercise in reactor cases is PRO-LOCA (Shim and Rudland, 2011; Rudland, et al., 2009; Rudland, et al., 2008; Xu, et al., 2006).

• Seismic-induced stress decreases from outside surface along the thickness.

12

Single SCC Crack Propagation (observation continued)

• No significant stress from internal gas pressure.

• The strength of neutron irradiation is insignificant in waste management.

• Environmental variations (e.g., seasonal temperature) may decrease the crack propagation rate.

• Through-wall growth of neighboring cracks has not been observed (SNL, 2007).

“Depending on the stress distribution, SCC may initiate and propagate through-wall. If

several cracks were to initiate in the same area, coalesce, propagate through-wall while remaining straight (i.e., perpendicular to the surface), and maintain smooth crack faces, a sizable section of material could fall out. The occurrence of all of these events in conjunction is improbable. Only tight and relatively separate through-wall cracks are expected.”

• Inspection of precursors (e.g., pits or early cracks) decreases the probability of crack

propagation. • Thermal annealing or applying compressive stress may mitigate crack initiation.

• Hydrogen embrittlement may behave similarly (e.g., carbon steel, Ahn and Soo, 1995).

13

Limited Conditions for SCC

• Narrow range of environmental and materials conditions for SCC

• Limited weld/HAZ area and deformed area (under seismic conditions)

• Limited number of pits or flaws

• Few through-wall cracks

• Inspection and mitigation

14

The decreased probability of SCC propagation in a canister is furthered by:

Possible Maximum Opening Area of Multiple Cracks (background)

The assessment of single crack propagation or environment/materials conditions for SCC has

large uncertainties, and data under anticipated conditions are lacking.

• The SCC issue has been under consideration for the disposal environment. For the environment of extended dry storage SCC has been studied and, to date, no SCC has been observed.

• Models for possible maximum opening area of multiple cracks were developed under seismic impact scenarios of disposal (SNL, 2007).

• The SNL seismic model assumes all possible surface cracks to penetrate through the wall thickness, and is likely to be conservative and bounding.

• The potential applicability of the SNL model in confinement assessment is considered.

15

Possible Maximum Opening Area of Multiple

Cracks: SNL Model

• The environment for stress corrosion cracking (SCC) is present: stress-based model

• The center of two cracks are separated by a parameter (near one times of the canister thickness, related to crack geometry) due to stress attenuation.

The distance between two neighboring through-wall cracks would need to be greater than the wall thickness for the stress (and resultant stress intensity) to be sufficient to drive a flaw through-wall. This conclusion is based on stress field interactions between closely spaced parallel cracks. The stress profile of crack network is shown in the next slide.

The aspect ratio of a crack (ratio of length to depth) has distribution for various possible crack geometries in a probabilistic system approach, and each value was sampled in the confinement assessment.

16

SNL Model: crack network from SCC

• Maximum number of cracks • “t” is radial thickness; greater than pit or flaw size

17

SNL Model: stress analysis with crack network

• Dimensions of the plate and setup for the analysis (left) • Longitudinal stress distribution along center with 2 inch spacing between cracks (right) (Structural Integrity Associates, 2002), 1 in = 2.54 cm; 1 ksi = 6.9 MPa

18

SNL Model: mathematical description

• At a crack length, a(t), the crack width, w(t), will be (SNL, 2007) w(t) = C σ a(t)/E σ: applied stress (MPa) E: Young’s modulus (MPa) C: geometric constant t: time under the conditions of plane strain and infinite size (conservative assumption) • Each crack area is product of crack length and crack width. The number of

cracks are proportional to sample area divided by a(t)2. The maximum opening area of multiple cracks are the product of each crack area and the number of cracks.

19

SNL Model: mathematical description (continued)

• The possible maximum opening area of multiple cracks (i.e., areal density per unit deformed or weld/HAZ area) is estimated with a hexagonal crack network expressed in SNL (2007).

δ = C σ/E δ: crack areal density (m2/m2) • Each crack area is the crack length times crack width, which is

proportional to the square of canister thickness; and the number of cracks is inversely proportional to the square of the canister thickness.

• The formula uses crack width obtained from crack length and applied stress under plane strain and infinite conditions, which are conservative.

20

Application of SNL Model

• Results of the numerical analysis of crack spacing were not sensitive to Young’s modulus and Poisson’s ratio, and the tested stress was in the range of 207 MPa (30.0 ksi) (Structural Integrity Associates, 2002). This implies that the SNL model can be applied to various metals.

• In an example light-water reactor (LWR) case, the number and size of surface cracks were reported from welds at the Nine Mile Point Unit 1 main recirculation lines (Xu, et al., 2006). The total number from Nine Mile Point surface cracks in welds is comparable with the above SNL model. However, only a small fraction of surface cracks were through the wall thickness, with a low probability (Rudland, et al., 2009). Although the number of the through-wall cracks may increase with time, the maximum number may not exceed the value of SNL model due to the stress attenuation between neighboring cracks. The aspect ratio of a crack is considered to increase with large surface cracks up to 10 to 150 (Xu, et al., 2006). The dominant area of surface cracks appears to fall in a range of crack sizes.

21

Application of SNL Model (continued)

• The SNL model appears to be conservative and bounding. The model also appears to be consistent with the above LWR example observation above with respect to most probable dominant crack size and potential aspect ratio present. Therefore, the SNL model is likely to be applicable for confinement assessment. This approach may be sufficient initially when radionuclides would likely be substantially dispersed if a canister failed (e.g., Ahn, et al., 2011). This is also partly supported by estimating the possible maximum opening area of cracks for various metals. An example is shown in the next slide.

22

Application of SNL Model (continued)

• An exercise was conducted for various metals with example parameter values for various metals.

Ratio of Yield Stress (YS) and Young’s Modulus for Various Metals (seismic case, Gwo, et al., 2011) YS is used as an applied. In the probabilistic confinement assessment, the applied stress value is sampled. 1 MPa = 0.145 ksi

YS (MPa) E (MPa) x 10-3 YS/E (mean) x 103

Stainless Steel 170 - 310 193-207 1.2

Carbon Steel 207 207 1.0

Copper 70-310 108-117 1.7

Zircaloy 241 99 2.4

23

Application of SNL Model (continued)

• For stainless steel, the mean value of maximum crack opening area per unit

weld/HAZ or deformed area is approximately 1.2x10-3 (fraction, cm2/cm2) for 170-310 MPa (24.6-44.9 ksi) of applied stress, (193-207)x103 MPa ([28.0-30.1]x103 ksi) of Young’s modulus. The weld/HAZ area fraction is about 10-2 – 10-1 (Ahn, et al., 2012). The fraction of the opening area is small.

• If this conservative/bounding approach is not sufficient because of no substantial radionuclide dispersion for the confinement assessment, rigorous approaches as used in reactors (e.g., Shim, et al., 2011; Rudland, et al., 2009; Rudland, et al., 2008; Xu, et al., 2006) need to be pursued.

24

Summary

• An approach is presented to abstract models for SCC damage of canister in the confinement assessment of radionuclides in the management of SNF and HLW.

• Localized corrosion mainly in pitting form may have limited density and size due to repassivation and cathodic capacity. An example calculation with field data shows that SCC of stainless and steel can be initiated at pits in chloride environments.

• The propagation rate of single SCC may decrease during propagation. Also, conditions for environments and materials are limited for SCC. However, the assessment of single crack propagation or environment/materials conditions for SCC are uncertain as performance periods increase in time. Data under extended dry storage are being studied.

• Possible maximum opening area of multiple cracks is estimated based on the SNL model for the seismic case in disposal. An example data in LWR welds suggest that the SNL model appears to be conservative and applicable for confinement assessment. The SNL model can be used for various metals. For stainless steel, the fraction of possible maximum crack opening area is limited from the model exercise. The conservative approach could be used when there is the possibility of radionuclides be substantially dispersed if a canister failed.

25

References

T. Ahn, “Long-Term C-14 Source Term for a High-Level Waste Repository, “ Waste Management, Vol.14, No.5, pp.393 -408, 1994

T. Ahn, R. Sun, T. Wilt, S. Kamas and S. Whaley, “Source Term Analysis in Handling Canister-Based Spent Nuclear Fuel: Preliminary Dose Estimate,“ The U.S. Nuclear Regulatory Commission (NRC) ADAMS, www.nrc.gov/reading-rm/adams.html - ML112640440, submitted to J. Nuclear Materials, 2011

T. Ahn, J. Guttmann and A. Mohseni, “Why Risk Assessment in Long-Term Storage of Spent Nuclear Fuel,” The U.S. Nuclear Regulatory Commission (NRC) ADAMS, www.nrc.gov/reading-rm/adams.html - ML111930448, 2012

T. Ahn, X. He, H. Jung and P. Shukla, “Long-Term Electrochemical Criteria for Crevice Corrosion in Concentrated Chloride Solutions,” NRC ADAMS, ML083110339, 2010

T. Ahn and P. Soo, “Corrosion of Low-Carbon Cast Steel in Concentrated Synthetic Groundwater at 80 to150° C,“ T. Ahn and P. Soo, BNL Report, BNL-61953, 1995, Waste Management, J., Vol.15, pp.471 -476, 1995

F. Cui, F. J. Presuel-Moreno and R. G. Kelly, “Computational Modeling of Cathodic Limitations on Localized Corrosion of Wetted SS 316L at Room Temperature,” Corrosion Science, Vol. 47, Issue 12, pp. 2987-3005, 2005

Electric Power Research Institute (EPRI), “Effects of Marine Environments on Stress Corrosion Cracking of Austenitic Stainless Steels,” EPRI 1011820, Palo Alto, California, 2005

D. Féron, D. Crusset and J.-M. Gras, Corrosion Issues in Nuclear Waste Disposal,” J. Nuclear Materials, 379, pp. 16-23, 2008

J. Gwo, T. Ahn and X. He, ‘Modeling Disruptive Events Using the β-SOAR Model: Levels of β-SOAR Model Flexibility

in Applications and Initial Insights,” Proceedings of 2011 International Radioactive Waste Management Conference

(IHLRWMC), Albuquerque, New Mexico, April 10-14, Paper No. 3407, 2011

26

References (continued)

JNC, H12: Project to Establish the Scientific and Technical Basis for HLW Disposal in Japan, Japan Nuclear Cycle Development Institute, Supporting Report 2, Repository Design and Engineering Technology, 2000 H. Jung, T. Ahn and X. He, “Representation of Copper and Carbon Steel Waste Package Degradation in a Generic Performance Assessment Model,” 2011 International High-Level Radioactive Waste Management Conference, Albuquerque, New Mexico, April 10-14, 2011 A. Kosaki, “Evaluation Method of Corrosion Lifetime of Conventional Stainless Steel Canister under Oceanic Air Environment ,” Nuclear Engineering and Design, 238, pp. 1233-1240, 2008

D. D. Macdonald, in C. Y. Chao, L. F. Lin and D. D. Macdonald, “A Point Defect Model for Anodic Passive Films: I. Film Growth Kinetics,” Journal of Electrochemical Society, Vol. 128, No. 6, pp. 1187-1194; L. F. Lin. C. Y. Chao and D. D. Macdonald, “A Point Defect Model for Anodic Passive Films: II. Chemical Breakdown and Pit Initiation,” Journal of Electrochemical Society, Vol. 128, No. 6, pp. 1194-1198, 1981 A. Passarelli, D. Dunn, O. Pensado, T. Bloomer and T. Ahn, “Risk Assessment of Uniform Corrosion and Localized Corrosion of Alloy 22., “ This paper was presented at Materials Science and Technologies (MS&T)’03, Chicago, IL, November 2003, NRC. ADAMS ML 040960476, 2003, and Met. and Mat. Trans. A, Vol. 36A, pp.1121-1127, 2005

D. Rudland, D.-J. Shim and A. Csontos, “Natural Flaw Shape Development due to Stress Corrosion Cracking,” Proceedings of the ASME 2008 Pressure Vessels & Piping Division Conference, PVP2008, Chicago, Illinois, July 27-31, 2008

D. Rudland, P. Scott, R. Kurth and A. Cox, “Continuing Development of PRO-LOCA for the Prediction of Break Probabilities for LOSS-OF-COOLANT Accidents,” Proceedings of the ASME 2009 Pressure Vessels & Piping Division Conference, PVP2009, Prague, Czech Republic, July 26-30, 2009

27

References (continued)

Sandia National Laboratories (SNL), “Stress Corrosion Cracking of waste Package Outer Barrier and Drip Shield Materials,” ANL-EBS-MD-000005 REV 04 ERD2, 2007

T. Shibata, “Stochastic Approach to the Effect of Alloying Elements on the Pitting Resistance of Ferritic Stainless Steels,” Symposia of the 104th ISIJ Meeting, September 1982, A199, at Hokkaido University in Sapporo, and to the 3rd USSR-Japan Seminar on Corrosion, October 1982, at Karpov Institute of Physical Chemistry in Moscow, Transactions ISIJ, Vol. 23, Article 785, 1983

D.-J. Shim, D. Rudland and D. Harris, “Modeling of Subcritical Growth due to Stress Corrosion Cracking – Transition from Surface Crack to Through-Wall Crack,” Proceedings of the ASME 2011 Pressure Vessels & Piping Division Conference, PVP2011, Baltimore, Maryland, July 17-21, 2011

K. Shirai, J. Tani, T. Arai, M. Wataru, H. Takeda, and T. Saegusa, “SCC Evaluation of Multi-Purpose Canister,” Proceedings of 2011 International Radioactive Waste Management Conference (IHLRWMC), Albuquerque, New Mexico, April 10-14, Paper No. 3333, 2011

P. Shukla, R. Pabalan, and O. Pensado, “Estimating the Extent of Damage to Waste Package Surfaces by Localized Corrosion in Potential Disposal Environments, “ Proceedings of 2011 International Radioactive Waste Management Conference (IHLRWMC), Albuquerque, New Mexico, April 10-14, Paper No. 3435, 2011

Structural Integrity Associates, 2002, in Licensing Support Network (LSN), U.S. Department of Energy, 2010, The Office of Legacy Management (LM) Business Center, Morgantown, West Virginia

H. Xu, D. L. Rudland and G. Wilkowski, “Statistical Characteristics Analysis and Simulation of Circumferential IGSCC Cracks for a BWR Plant,” Proceedings of the ASME 2011 Pressure Vessels & Piping Division Conference, PVP2006, Vancouver, BC, Canada, July 23-27, 2006

28