ASSESSMENT OF STRESS CORROSION CRACKING SUSCEPTIBILITY OF 316 STAINLESS STEEL IN DIFFERENT DISPOSAL ENVIRONMENTS

K. Chiang and P. Shukla

Center for Nuclear Waste Regulatory Analyses (CNWRA®), Southwest Research Institute®, 6220 Culebra Road, San Antonio, Texas 78238-5166, USA

Contact: K. Chiang, [email protected], Telephone: +210-522-2308

Stainless steel may be considered as a waste package chemical, thermal, and electrochemical conditions; and container material in different disposal environments. The (iii) magnitude of applied and residual tensile stresses. objective of this paper is to assess stress corrosion Welding and heat treatment can influence the cracking (SCC) susceptibility of 316 stainless steel in microstructure of the material affecting the susceptibility possible disposal environments. Factors including to SCC. The environmental setting is defined by the (i) material properties affected by fabrication processes chemistry of the aqueous solution in contact with the such as welding and heat treatment; (ii) environmental waste package, temperature, and electrochemical conditions including chemistry of aqueous solution variables such as the corrosion potential. Tensile stresses surrounding the waste package, temperature, and could arise from welding. Fig. 1 illustrates the three electrochemical conditions; and (iii) tensile stress in the factors that must be simultaneously present for SCC to welded areas as well as tensile stress generated by events occur. such as seismic ground motion are of importance to SCC susceptibility of the alloy in disposal environments. The susceptibility of the alloy is assessed considering these factors in potential disposal environments. Literature Material Environment information was compiled to define chemical and thermal conditions that could arise in disposal environments. • Welding • Water Chemistry Numerical simulations were conducted to determine • Heat • Temperature electrochemical conditions of the alloy in disposal Treatment • Electrochemical environments. Literature information on 316 stainless Potential steel’s SCC susceptibility in the chemical and thermal SCC conditions similar to those in potential disposal environments was compiled. The susceptibility of the alloy was assessed by comparing the literature information. Stress I. INTRODUCTION • Residual Stress Engineered barrier systems for a potential high-level • Seismic Induced radioactive waste disposal system might include waste • Rock Overburden containers made of corrosion-resistant material, such as stainless steel. Fabrication of nuclear waste containers generally will require multiple processes such as welding and solution annealing. Corrosion is expected to be a Fig. 1. Factors that lead to SCC of nuclear waste degradation process limiting waste container life. One of container materials. the potential corrosion degradation modes for the waste container is stress corrosion cracking (SCC). SCC is a In this paper, the SCC susceptibility of 316 stainless phenomenon by which a normally ductile alloy loses its steel in three different possible disposal environments is toughness (elongation to rupture time) when it is subject assessed. The effects of material properties; residual to mechanical stresses under a range of environments. stresses; and chemical, thermal, and electrochemical SCC susceptibility of the nuclear waste container conditions are considered in regard to the alloy’s materials is dependent on three factors: (i) material- susceptibility to SCC. related factors such as metallurgy and microstructure of the material; (ii) environmental conditions including Alloy C-22 is comparable to 316 stainless steel with I.A. Material-Related Issues respect to SCC processes. Thus, experiments on SCC for C-22 can yield insights on SCC processes for 316 316 stainless steel is a possible material to be used to stainless steel, about direct experimental data. As an air construct waste package containers to isolate nuclear example, Fig. 3 shows the time-to-failure ratios (tf/tf ) for waste from the environments in a potential geological the slow strain rate tests of a nickel-based alloy in disposal system. Fabrication of containers generally solutions containing various anionic and cationic species.8 requires multiple processes, such as welding and solution The ionic species include chloride and bicarbonate ions. annealing.1,2 These processes may alter the The ratio of time-to-failure in the test environment versus microstructure and mechanical properties of the base time-to-failure measured in air can be considered as an alloys, and introduce residual tensile stresses. Thermal index of the severity of SCC. Fig. 3 shows that the treatment can cause carbide precipitation; change the addition of a small concentration (0.2 molal) of chloride grain size and microstructure of the welded areas and to the 1.1 molal bicarbonate solution significantly heat-affected zone in a 316 stainless steel waste decreases the failure time. container.1–3 In this paper, it is assumed that fabrication- related defects exist on the waste package.

I.B. Evaluation of Environmental Conditions

SCC can occur in a range of the aqueous solution chemistries, temperatures, and polarization potentials of the alloy in the solution. The range of environmental conditions that can be conducive to SCC can be defined using accelerated methods such as slow strain rate tests or constant load experiments. The environmental conditions include the aqueous solution chemistry, pH, electrochemical potential, and temperature. Stress corrosion susceptibility of 316 stainless steel has been studied in chloride-containing aqueous solution 2+ + + in different cations (Mg , Li , and Na ) at temperatures ranging from 90 to 150 °C (194 to 302 °F) using slow 4–6 air strain rate tests. Slow strain rate testing was conducted Fig. 3. Time-to-failure ratios (tf/tf ) for nickel alloy in accordance with the ASTM G–129 procedure.7 A specimens tested at 95 °C (203 °F) in 1.1 molal and 2.1 − – photograph of the experimental test cell for the slow molal HCO3 solutions containing various Cl strain rate test is shown in Fig. 2. The same test cell can concentrations.8 The tests were performed at a constant also be used to conduct the constant load test. strain rate of 3.2 × 10−6/sec.

Ductile failure, intergranular SCC, or transgranular SCC can be established by posttest examination. An example of a nickel-based alloy sample subjected to the slow strain rate test in the 7.2 molal chloride solution and − 1.1 molal HCO3 solution containing 4.2 molal chloride is shown in Fig. 4 (a) and (b). In a solution containing only 7.2 molal chloride, the specimen exhibited significant plastic deformation (87.6 percent elongation) and a time-to-failure ratio close to 1.0. The side surface of the specimen shows ductile failure with no sign of surface cracks.

Fig. 2. Slow strain rate test apparatus with specimen mounted in the test cell.

sodium chloride could cause the chloride ion concentration to range from 50–50,000 parts per million (4.2 × 10−4 to 0.42 lb/gallon), and the aqueous solution temperature could be as high as 90 °CC (194 °F).9 In this paper, it is assumed that the aqueous solution surrounding the waste packages in a granite rock disposal environment contains NaCl in a concentration of 50 g/L (0.42 lb/gallon) and the aqueous solution temperature is (a) (b) 90 °C (194 °F). In the three potential disposal Fig. 4. Side surface of a nickel alloy specimen strained in environments, the aqueous solutions contacting the waste − − (a) 7.2 molal Cl only and (b) 1.1 molal HCO3 + 4.2 packages are expected initially to contain dissolved molal Cl−.8 oxygen because the oxygen might be present during the repository construction and also sometime after closure. − On the other hand, in a 1.1 molal HCO3 solution For deep disposal system with undisturbed groundwaters, containing 4.2 molal chloride solution, a large number of dissolved oxygen in the solution might be consumed by secondary SCC were present on the side surface of the oxygen reduction reaction and eventually the aqueous specimen (Fig. 4b). The elongation of the test specimen solution might become reducing. was reduced to 52.2 percent, with a time-to-failure ratio of 0.51. The effects of the environment (water chemistry II.B. Electrochemical Conditions − − containing HCO3 and Cl ) in causing SCC are illustrated. Thus, information about the time-to-failure ratio and the The electrochemical conditions for the carbon and presence of microcracks on the side surface of the sample 316 stainless steel waste package material in three after the test can be used to determine whether an alloy is potential disposal environments were calculated using the susceptible to SCC in specific environmental conditions. OLIAnalyzer Version 3.1 software.10 The software results have been extensively validated.11 The chemical II. DISPOSAL ENVIRONMENTS compositions of the 316 stainless steel are provided in TABLE I. The chemical and thermal conditions, and alloy In this paper, potential disposal of stainless-steel specifications were input in the software. The calculated waste packages in salt rock, clay, and granite is results included polarization curves and corrosion considered. These three potential disposal environments potentials. The values of the calculated corrosion can lead to different chemical compositions and thermal potentials were read from the polarization curves. Both conditions of the aqueous solution in contact with waste compositions of the aqueous solutions and calculated packages. These conditions are discussed next. In corrosion potentials for the 316 stainless steel are listed in addition, the electrochemical conditions that can be used TABLE II. to determine SCC susceptibility of the alloy are detailed. III. SCC SUSPECTIBILITY EVALUATION II.A. Chemical and Thermal Conditions III.A. SCC Test Data Stainless steel waste packages placed in salt rock could be contacted by sodium chloride or magnesium Literature information was searched to compile SCC chloride brines of concentrations of approximately of test data on 316 stainless steel and to assess SCC 26 and 30 wt %, respectively. The pH of the susceptibility of 316 stainless steel in three potential sodium-chloride-rich brines and magnesium-chloride-rich disposal environments. It is assumed that material brines could range from 4–7.9 The temperature of those conditions, such as the heat affected zone, and enough brines could range 90–150 °C (194–302 °F) depending on residual tensile stresses are present. Thus, this paper the design of disposal systems. If waste packages were focuses only on the SCC susceptibility of 316 stainless placed inside a bentonite clay, an aqueous solution steel as a function of chemical, thermal, and predominantly containing sodium, magnesium, and electrochemical conditions. Tsai and Chen4 conducted chloride ions could contact the waste packages. Possible slow strain rate tests to determine SCCC susceptibility of chemical composition of the aqueous solution in the duplex- and 316 stainless steel in 26 wt % NaCl solution, bentonite clay can be found in Table 2–15 of a report with a pH equal to 6 and at 90 °C ((194 °F). The test published by the European Commission.9 The solution was deaerated, the strain rate was selected to be maximum temperature of the aqueous solution might 4.1 × 10−6/sec, and samples were not polarized (i.e., the range from 50–100 °C (122–212 °F). For the waste tests were conducted at the corrosion pootentials). Tsai and packages placed in granite, aqueous solutions containing Chen4 reported that transgranular fractures were observed sodium chloride could develop. The concentration of the TABLE I. Chemical composition of the carbon- and 316 stainless steel Alloy Mass fraction of the various constituents Fe C Mn Mo Cr Ni Carbon steel 0.967 0.023 9.98 × 10−3 0 0 0 316 stainless steel 0.671 4.66 × 10−3 0 0.018 0.183 0.124

TABLE II. Example of chemical and electrochemical conditions for 316 stainless steel in three potential disposal environments Corrosion Potential # Chemical Conditions (Ecorr in unit of V vs. SHE ) Disposal Chemical pH Oxidizing Environment Composition Range Temperature Condition Reducing Condition Rock salt 26 wt % NaCl 4–7 90 °C (194 °F) 0.18 −0.40 to −0.14 solution 26 wt % NaCl 4–7 150 °C (302 °F) −0.41 to −0.39 0.02 to 0.03 solution 30 wt % MgCl2 4–5 90 °C (194 °F) 0.18–0.2 −0.52 to −0.56 solution 30 wt % MgCl2 4–5 150 °C (302 °F) −0.39 −0.62 to −0.70 solution Bentonite clay Bentonite clay 50 °C (122 °F) 0.21 −0.30 water* Granite 50 g/L 6–9 90 °C (194 °F) 0.16 −0.32 (0.42 lb/gallon) NaCl solution *The chemical composition of the water in the bentonite clay can be found in Table 2-15 of a report published by the European Commission9. #SHE stands for standard hydrogen electrode. in 316 stainless steel samples during the posttest were selected: 1,000 and 10,000 parts per million examinations, indicating the occurrence of SCC at the (8.4 × 10−3 and 8.4 × 10−2 lb/gallon). In addition, sulfate open circuit potential. Alyousif and Nishimura5 conducted ions were added in the solution. The sulfate ion constant load experiments to determine SCC concentrations were selected as 20, 1,000, and susceptibility of 316 stainless steel in boiling saturated 10,000 parts per million (1.68 × 10−4, 8.4 × 10−3, and −2 MgCl2 solutions at 135 and 155 °C (275 and 311 °F). The 8.4 × 10 lb/gallon). The aqueous solutions were purged samples were not polarized (i.e., the experiments were with nitrogen, air, and carbon dioxide, and the samples conducted at the corrosion potential). The concentrations were polarized approximately 200 mV above the of the boiling saturated MgCl2 solution at 135 and 155 °C corrosion potentials of the 316 stainless steel in the (275 and 311 °F) are approximately equal to 44 and 49 aqueous solutions. The authors reported that the wt %, respectively. The concentrations of the boiling 316 stainless steel did not undergo SCC in any of the 6 saturated MgCl2 solutions were determined by solutions. Cragnolino et al. also conducted slow strain OLIAnalyzer Version 3.1 software as Alyousif and rate tests in approximately 27 wt % NaCl solutions at Nishimura5 did not provide this information. The tensile 95 °C (203 °F). The corrosion potential of the alloy was stress on the samples was varied between 100 and 500 approximately –0.12 VSHE (the subscript SHE stands for MPa (1.45 × 104 and 7.25 × 104 psi) in the constant load standard hydrogen reference electrode). The samples were experiment. Alyousif and Nishimura5 reported that the kept at the corrosion potential during the tests. The 316 stainless steel underwent intergranular SCC at 135 °C authors reported that 316 stainless steel only underwent (275 °F) and transgranular SCC at 155 °C (311 °F). ductile failure or pitting corrosion in the aqueous solution. Cragnolino et al.6 conducted slow strain rate tests to The authors also conducted the test in 30 and 40 wt% determine SCC susceptibility of 316 stainless steel in MgCl2 solutions at 110 and 120 °C (230 and 248 °F). The chloride containing solutions with Mg2+, Li+, and Na+ samples were kept at the corrosion potentials. The authors cationic species at temperature ranging from 95 to 125 °C reported that 316 stainless steel underwent SCC in the (203 to 257 °F). For slow strain rate tests conducted at 30 and 40 wt % MgCl2 solutions. 95 °C (203 °F), two chloride concentrations II.B. SCC Susceptibility of 316 Stainless Steel Furthermore, there is not enough information available to assess the SCC susceptibility of the alloy in a potential Based upon the literature information, it is granite disposal environment. concluded that 316 stainless steel appears susceptible to SCC in 26 wt % NaCl solution and 30 wt % MgCl2 ACKNOWLEDGMENTS solution in the temperature range of 90 to 150 °C (194 to 302 °F) under both oxidizing and reducing environments. This paper is an independent product of the CNWRA Therefore, 316 stainless steel waste packages appear and does not necessarily reflect the view or regulatory susceptible to SCC in some salt rock disposal position of the NRC. environments. The solution contacting waste packages in bentonite clay might contain approximately 6,550 parts REFERENCES per million chloride (5.46 × 10−2 lb/gallon), 1,500 parts per million sulfate (1.26 × 10−2 lb/gallon), 110 parts per 1. D. S. DUNN, D. DARUWALLA, and Y.-M. PAN, million nitrate (9.24 × 10−4 lb/gallon), and 27 parts per “Effect of Fabrication Processes on Material million of bicarbonate ions (2.27 × 10−4 lb/gallon).9 Stability−Characterization and Corrosion,” CNWRA Cragnolino et al.6 report that SCC of 316 stainless steel eport 2004-01, San Antonio, Texas (2003). was not observed when slow rate experiments were 2. D. S. DUNN, Y.-M. PAN, K. T. CHIANG, conducted in solutions containing chloride ions at 1,000 L. YANG, G. A. CRAGNOLINO, and X. HE, “The and 10,000 parts per million (8.4 × 10−3 and 8.4 × 10−2 Localized Corrosion Resistance and Mechanical lb/gallon), and sulfate ions at 20, 1,000, and 10,000 parts Properties of Alloy 22 Waste Package Outer per million (1.68 × 10−4 , 8.4 × 10−3, and 8.4 × 10−2 Containers,” Journal of Metals, Vol. 57, p. 49 (2005). lb/gallon). Previous studies6,8 suggest that the presence of 3. G. A. CRAGNOLINO, H. K. MANAKTALA, and nitrate ions is not likely to cause SCC of the alloy, and Y.-M. PAN, “Thermal Stability and Mechanical concentration of bicarbonate ions is too low to influence Properties of High-Level Radioactive Waste SCC of the alloy. It is therefore concluded that SCC of Container Materials Assessment of Carbon and Low- 316 stainless steel is unlikely in a potential bentonite clay Alloy Steels,” CNWRA Report 96-004, San Antonio, disposal environment. The aqueous solutions contacting Texas (1996). waste packages in granite could contain chloride ions in 4. W.-T. TSAI and M.-S. CHEN, “Stress Corrosion Cracking Behavior of 2205 Duplex Stainless Steel in the range of 50–50,000 parts per million (4.2 × 10−4 to 4,6 Concentrated NaCl Solutions,” Corrosion Science, 0.42 lb/gallon), whereas experimental data are available Vol. 42, p. 545 (2000). in the aqueous solutions containing chloride ions up to 5. O. M. ALYOUSIF and R. NISHIMURA, “The Stress 10,000 parts per million (8.4 × 10−2 lb/gallon) and at 26 wt Corrosion Behavior of Austenitic Stainless Steels in % NaCl solution. Based upon this information, SCC Boiling Magnesium Chloride Solutions,” Corrosion susceptibility of the alloy in a potential granite disposal Science, Vol. 49, p. 3040 (2007). environment cannot be determined. 6. G. A. CRAGNOLINO, D. S. DUNN, and

N. SRIDHAR, “Environmental Effect on Stress IV. SUMMARY Corrosion Cracking of Type 316L Stainless Steel and

Alloy 825 as High-Level Nuclear Waste Container 316 stainless steel is a material that might be used to Materials,” CNWRA Report 94-028, San Antonio, construct nuclear waste containers because of its Texas (1994). corrosion resistance. One degradation mode for 7. ASTM INTERNATIONAL, “Standard Practice for 316 stainless steel is SCC. SCC may occur when a Slow Strain Rate Testing to Evaluate the susceptible material, environmental conditions, and Susceptibility of Metallic Materials to stresses are present simultaneously. Environmentally Assisted Cracking, G-129: Annual SCC susceptibility of 316 stainless steel is assessed Book of ASTM Standards. Vol. 3.02: Wear and in three potential disposal environments: salt rock, Erosion-Metal Corrosion. West Conshohocken, bentonite clay, and granite. Literature information was Pennsylvania (2003). compiled to assess the SCC susceptibility of the alloy in 8. K. T. CHIANG, D. S. DUNN, and these three environments. Based upon literature G. A. CRAGNOLINO, “Effect of Simulated information on chemical and thermal conditions in Groundwater Chemistry on Stress Corrosion potential disposal environments and literature data on Cracking of Alloy 22,” Corrosion, Vol. 63, p. 940 SCC susceptibility of 316 stainless steel in different (2007). aqueous solutions, it is concluded that 316 stainless steel appears susceptible to SCC in a potential rock salt disposal environment but not in studied bentonite clays. 9. B. KURSTEN, E. SMAILOS, I. AZKARATE, L. WERME, N. R. SMART, and G. SANTARINI, “State-of-the-art Document on the Corrosion Behaviour of Container Materials,” 5th EURATOM FRAMEWORK PROGRAMME 1998-2002 KEY ACTION: NUCLEAR FISSION, European Commission (2004). 10. OLI SYSTEMS, INC., “A Guide to Using the OLI Software for Version 3.1 of the Analyzers,” OLI Systems, Inc, Morris Plains, New Jersey: (2010). 11. M. S. GRUSZKIEWICZ, D. A. PALMER, R. D. SPRINGER, P. WANG, and A. ANDERKO, "Phase Behavior of Aqueous Na–K–Mg–Ca–Cl–NO3 Mixtures: Isopiestic Measurements and Thermodynamic Modeling,” Journal of Solution Chemistry, 36, p. 723 (2007).