ML24178A348

TLR-RES/DE/REB-2024-09: Effect of Fission Products on Degradation of Structural Materials in Molten Salt

ML24178A348
Person / Time
Issue date:	07/31/2024
From:	Kiosidou E, Pint B, Wendy Reed, Sulejmanovic D NRC/RES/DE
To:
WENDY REED 301-415-7213
Shared Package
ML24178A345	List: ML24178A348
References
TLR-RES/DE/REB-2024-09, ORNL/SPR-2023/3020
Download: ML24178A348 (38)

Text

1 Letter Report TLR-RES/DE/REB-2024-09 ORNL/SPR-2023/3020 EFFECT OF FISSION PRODUCTS ON DEGRADATION OF STRUCTURAL MATERIALS IN MOLTEN SALT REACTORS Date:

July 2024 Prepared in response to Task 6 in NRC Technical Assistance Pertaining to Advanced Reactors in the Areas of Corrosion Experiment Methodology and Evaluation by:

Evelina Kiosidou Oak Ridge National Laboratory Dino Sulejmanovic Oak Ridge National Laboratory Bruce A. Pint Oak Ridge National Laboratory NRC Project Manager:

Wendy Reed Senior Physical Scientist Reactor Engineering Branch Division of Engineering Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001

DISCLAIMER This report was prepared as an account of work sponsored by an agency of the U.S.

Government. Neither the U.S. Government nor any agency thereof, nor any employee, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any third party's use, or the results of such use, of any information, apparatus, product, or process disclosed in this publication, or represents that its use by such third party complies with applicable law.

This report does not contain or imply legally binding requirements. Nor does this report establish or modify any regulatory guidance or positions of the U.S. Nuclear Regulatory Commission and is not binding on the Commission.

DOCUMENT AVAILABILITY Reports produced after January 1, 1996, are generally available free via OSTI.GOV.

Website www.osti.gov Reports produced before January 1, 1996, may be purchased by members of the public from the following source:

National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone 703-605-6000 (1-800-553-6847)

TDD 703-487-4639 Fax 703-605-6900 E-mail info@ntis.gov Website http://classic.ntis.gov/

Reports are available to US Department of Energy (DOE) employees, DOE contractors, Energy Technology Data Exchange representatives, and International Nuclear Information System representatives from the following source:

Office of Scientific and Technical Information PO Box 62 Oak Ridge, TN 37831 Telephone 865-576-8401 Fax 865-576-5728 E-mail reports@osti.gov Website https://www.osti.gov/

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

ORNL/SPR-2023/3020 Sponsor number Materials Science and Technology Division EFFECT OF FISSION PRODUCTS ON DEGRADATION OF STRUCTURAL MATERIALS IN MOLTEN SALT REACTORS Evangelia Kiosidou, Dino Sulejmanovic, Bruce Pint July 2024 Prepared by OAK RIDGE NATIONAL LABORATORY Oak Ridge, TN 37831 managed by UT-BATTELLE LLC for the US DEPARTMENT OF ENERGY under contract DE-AC05-00OR22725

iii CONTENTS CONTENTS................................................................................................................................................. iii FIGURE LIST............................................................................................................................................... v TABLE LIST.............................................................................................................................................. vii ABBREVIATION LIST.............................................................................................................................. ix EXECUTIVE

SUMMARY

.......................................................................................................................... 1

1.

Introduction........................................................................................................................................... 2

2.

Direct effects of fission products on corrosion..................................................................................... 4 2.1 The effect of Te and mechanisms of attack................................................................................ 4 2.1.1 Te-induced embrittlement in stainless steels................................................................. 4 2.1.2 The mechanism of Te-induced embrittlement on Ni-based alloys................................ 5 2.1.3 Modifications of Hastelloy N composition for improved resistance against Te-induced IGC cracking.................................................................................................. 10 2.2 The effect of cesium and iodine and mechanisms of attack...................................................... 13 2.3 The effect of Eu and mechanisms of attack.............................................................................. 14

3.

Indirect effects of fission products on corrosion................................................................................. 16

4.

Comparison of fission product effects to irradiation effects............................................................... 18

5.

Strategies to mitigate the effects of fission products.......................................................................... 20

6.

Summary............................................................................................................................................. 21

7.

References........................................................................................................................................... 22

v FIGURE LIST Figure 1-1. Layouts of (a) Molten Salt Reactor Experiment (MSRE) [Luzzi 2012] and (b) Molten Salt Breeder Reactor (MSBR) [Rosenthal 1972]............................................................................. 3 Figure 2-1. Exposure of 304SS to 0.1wt% Te, dissolved in unpurified FLiNaK, at 600 °C, for 48 h; (a) Cross section revealing IGC and (b) effect of Te concentration on maximum penetration depth [Hong 2023]........................................................................................................ 5 Figure 2-2. Effect of Te on the (a) mechanical properties and (b) structure internal stresses on a Ni-based alloy [Liu 2014a].............................................................................................................. 6 Figure 2-3. Depth of Te penetration along Hastelloy N GBs as a function of exposure time at 700°C in simulated MSRE fuel salt [Keiser 1976].......................................................................... 6 Figure 2-4. Ni-Te phase diagram from Keiser [1976].................................................................................. 7 Figure 2-5. Hastelloy N exposed to Te vapor at 800 ºC for 100 h; (a) Te corroded surface adjacent to an intergranular carbide and (b) zig-zag morphology CrTe layer observed after magnification of white area in (a) [Cheng 2015]............................................................................. 8 Figure 2-6: Reaction products formed on Hastelloy N during exposure to Te vapor at 800 °C for 150 h; (a) fine-grained material and (b) coarse-grained material [Jiang 2022]............................... 8 Figure 2-7: Outer corrosion layer on Hastelloy N exposed to LiCl-KCl-Li2Te, at 500 ºC, for 120 h

[Du 2022]......................................................................................................................................... 9 Figure 2-8: Proposed formation mechanisms for surface Ni-tellurides; (a) Te0 hypothesis and (b)

Te2-hypothesis [Du 2022].............................................................................................................. 10 Figure 2-9: Effect of Nb concentration on Hastelloy N performance; (a) mechanical properties

[McCoy 1978] and (b) stress corrosion cracking [Keiser 1976].................................................... 11 Figure 2-10: Effect of Nb concentration on IGC cracking as a function of exposure duration to fuel salt [McCoy 1978].................................................................................................................. 11 Figure 2-11: Mechanism of beneficial Nb additions in Hastelloy N, in an atomic level; (a) effect of Nb additions in the tensile properties and (b) stress development in the GB area due to additions of Te and Nb with red circles Ni atoms; Left-clean GB without additions; Center -after addition of Te (blue); Right - after additions of larger Nb atoms. [Liu 2014a]............................................................................................................................................ 12 Figure 2-12: Effect of salt redox potential of a fuel salt on Te-induced IGC cracking on Nb-modified Hastelloy N; (a) [UF4]/[UF3] =100 and (b) [UF4]/[UF3] = 500 [Ignatiev 2008]............. 13 Figure 2-13. Depth of corrosive attack on Hastelloy N after exposure at various temperatures and crucible types for 100 h [Yamawaki 2015].................................................................................... 14 Figure 2-14. Specific mass change for the Cr specimens exposed in three different salt conditions at 600°C and 700°C [Sulejmanovic 2023]..................................................................................... 15 Figure 2-15. Cross-sectional SEM and elemental mapping for Cr specimens exposed in the baseline KCl-MgCl2 salt (top row), KCl-MgCl2 + Eu metal (middle row) and KCl-MgCl2

+ EuCl3 (bottom) row at 700°C [Sulejmanovic 2023]................................................................... 16 Figure 2-16. XRD powder plots of the Cr specimen after the 600°C 1,000-hr exposure in KCl-MgCl2+Eu salt condition (blue line) showing evidence of MgO on the surface........................... 16 Figure 3-1. Fission-product average valency for 239PuCl3 irradiated with 2-MeV neutrons with rare gases retained in system.......................................................................................................... 17 Figure 3-2. Liquidus projection of (a) the LiF-ThF4-CsF system and (b) the Li-Th-F-I system

[Capelli 2018]................................................................................................................................ 18 Figure 4-1. Effect of Ti concentration on ductility of irradiated austenitic stainless steel at 842 °C

[Martin 1966]................................................................................................................................. 19

vii TABLE LIST Table 2-1. Possible stable chemical forms of Cs and I in molten halide salts [Capelli 2018].................... 13 Table 3-1. Oxidation states of the fission product elements in molten chlorides with 239Pu as the fissile element [from Chasanov [1965].......................................................................................... 17

ix ABBREVIATION LIST ASME American Society of Mechanical Engineers at.%

Atomic percent EDS Energy dispersive x ray spectroscopy GB Grain boundary INL Idaho National Laboratory IGC Intergranular corrosion MSBR Molten Salt Breeder Reactor MSR Molten Salt Reactor MSRE Molten Salt Reactor Experiment ORNL Oak Ridge National Laboratory PFLS Primary fuel loop salt SEM Scanning Electron Microscopy SS Stainless Steel T

Tritium XRD x ray Diffraction

1 EXECUTIVE

SUMMARY

Materials compatibility of structural materials with molten coolant or fuel salts is an important consideration for molten salt reactors. In addition to the environmental impact of the coolants, presence of fission products further impacts the compatibility of components. This is best illustrated by the tellurium (Te)-induced cracking observed in the molten salt reactor experiment (MSRE) conducted in the late 1960s at Oak Ridge National Laboratory (ORNL). Furthermore, the spectrum of fission products present in the system depends on several factors, such as the salt and fuel type, the neutron energy spectrum, and the reactor operating conditions. The most common fission products in nuclear reactors are noble gases, rare-earth elements, and halogens. They also can be subdivided by those elements that are soluble in the salt as ions and those that remain in the metallic state and can deposit on the structural materials or graphite moderator.

The effects of fission products have been broken-down into two categories, direct and indirect effects. Te is the most well-known fission product which directly affected Ni-based Hastelloy N in the MSRE by embrittling grain boundaries and exhibited the highest penetration of fission products into the Hastelloy N. The effect of Te continues to be investigated including the susceptibility of stainless steels to Te-induced cracking and the exact mechanism behind grain boundary embrittlement in structural alloys due to Te deposition. Other elements, such as Ag, Nb, Mo and Ru, were found to have deposited on system surfaces in the MSRE. Other fission products in fluoride salt like Cs and I do not increase the attack on structural alloy Hastelloy N.

Indirect fission product effects include their impact on the chemical composition of the salt itself. For example, Eu has been investigated as an example of a multivalent rare earth fission product that might affect salt compatibility by releasing Cl/F-, an indirect effect. While it is possible that physical properties could be altered by fission products, no examples were found of this effect. This is primarily due to the fact that the fission products are generated slowly during the operation and could be actively removed or gettered from the salt. The main indirect effect is on the salt redox potential which can affect material compatibility.

Fission product effects are briefly compared to irradiation effects as another important degradation mechanism. The He embrittlement of Hastelloy N was a major concern after the MSRE. Alloy modification with Nb, Ti and Al have been undertaken in the past 50 years to improve the irradiation resistance of Hastelloy N. In addition to the strategy of developing more compatible materials, other mitigation strategies have been summarized. Indirect effects can be controlled by using active elements or the [UF4]/[UF3] ratio to control the salt potential. Direct effects like Te-induced cracking also can be mitigated by redox control. More corrosion and radiation resistant materials might also be developed.

Operational practices also can be designed to minimize the accumulation of fission products including off-gas systems to capture noble gases and noble metals.

Regarding the current commercial interest in stainless steels, ongoing external research efforts could clarify compatibility issues with fission products of Fe-based alloys. While, accelerated testing by adding high levels of fission products to coolant or fuel salt may be useful to obtain key insights, this type of testing may not replicate operating conditions. For example, accelerated testing with higher Te levels in the salt frequently resulted in reduced cracking compared to the cracking seen in MSRE. With the importance of monitoring the salt redox potential, commercial electrochemical in situ monitoring systems could offer additional opportunities to address materials compatibility for both chloride and fluoride salts.

Although most of the research is based on the historical MSRE experience with Hastelloy N and Te, the information reviewed so far indicates that the fission product effects on materials being considered currently for molten salt reactors (MSRs) can be managed. Hence, the components would operate safely in currently proposed MSR environments.

2

1.

INTRODUCTION Molten salt reactors (MSRs) are a promising type of nuclear reactor that typically use liquid halide salts as fuel and/or coolant. MSRs offer several advantages over traditional nuclear reactors, such as higher efficiency and improved safety due to low pressure operation [Rosenthal 1970, LeBlanc 2010].

The first large-scale MSR experiment was the Molten Salt Reactor Experiment (MSRE), conducted at ORNL from 1965-1969 [McCoy 1972, 1978], Figure 1-1a. The MSRE was a 7.4 MWth reactor using 2LiF-BeF2 (FLiBe) molten salt as a fuel carrier (235UF4 and 233UF4) and as the primary coolant. During the first campaign of the program (26 months from 1965-1968, >9000 effective full power hours), the fuel salt was LiF-BeF2-ZrF4-235UF4 (65:29.1:5:0.9 mole %) [Briggs 1964] and the second campaign (15 months from 1968-1969, >5100 effective full power hours) used 233UF4. The core of the reactor included unclad graphite moderator rods through which the fuel salt was flowing at ~700°C.

After the end of the MSRE program, a subsequent program focused on overcoming issues that arose during the MSRE, which were mostly related to the absence of breeding capabilities and the lack of tritium shielding but also included some materials issues. The Molten Salt Breeder Reactor (MSBR) program ran from 1970-1976 at ORNL. The planned (but not built) reactor was ~2240 MWth (1000 MWe) operating at ~700°C with graphite moderator rods, Figure 1-1b. The fuel salt was also similar to the MSRE, LiF-BeF2-ThF4-UF4 (72:16:12:0.4 mole %), but this time the design also included additions of the fertile 232ThF4, which would allow transmutation to 233U, creating the breeding capability. The MSBR design included NaF-NaBF4 as a secondary coolant to serve as a tritium getter and facilitate tritium removal. Tritium permeation and escape into the environment was another issue with the MSRE

[MacPherson 1985]. The higher fluence and increased fission products expected in the MSBR required addressing MSRE material degradation issues [DeVan 1994].

MSRs have exhibited some unique technical challenges for structural materials. One problem was the irradiation embrittlement of Hastelloy N (Ni-7Cr-16Mo-5Fe), which was specifically designed for its molten salt compatibility. Embrittlement arose from He accumulation at the grain boundaries (GBs) of this Ni-based material [McCoy 1978]. Corrosion of structural and functional materials [Zheng 2018, Raiman 2021] is another important MSR issue. In addition to the well-known molten salt compatibility issues such as salt impurities and mass transfer, the presence of fission products is also a concern

[Konings 2020]. The spectrum of fission products in an MSR is dependent on several factors, such as the type of fuel used, the neutron energy spectrum, and the reactor operating conditions. The most common fission products in fluid-fueled molten salt reactors are noble gases, rare-earth elements, halogens, and other radioactive isotopes [Grimes 1970, Compere 1975].

Noble gases: The fission of U and Th nuclei in MSRs can produce noble gases such as Xe and Kr, which are chemically inert and have low solubility in the molten salt. These gases can build up in the reactor and affect its performance. For example, 135Xe is a powerful neutron poison.

Therefore, off-gas systems can be used to collect these products, which will be discussed below.

Rare-earth elements: MSRs can produce rare earth elements such as Eu and Sm, which have been widely studied in pyro-processing [Guo 2018, Zhang 2014].

Halogens: MSRs can produce halogens such as Cl and Br, which can react with the molten salt and cause corrosion or other problems.

Radioactive isotopes: MSRs can produce a variety of radioactive isotopes, including isotopes of I, Cs, Sr and Tc. These isotopes can be long-lived and remain radioactive for thousands of years, contributing to the long-term nuclear waste generated by the reactor.

3 (a)

(b)

Figure 1-1. Layouts of (a) MSRE [Luzzi 2012] and (b) MSBR [Rosenthal 1972]

Other elements: One of the most important fission products in MSRs is Te. This element can induce intergranular corrosion (IGC) cracking on the structural materials which was found to have deleterious effects on the performance of Hastelloy N in the MSRE [Keiser 1976, McCoy 1978, MacPherson 1985].

4 Another way to group the generated fission products in an MSR is based on their mass transfer characteristics and affinity to the salt, as suggested by Kedl [1972, Aufiero 2013]:

(a) Salt seekers: are soluble in the salt as ions, examples: Sr, Y, Zr, I, Cs, Ba and Ce.

(b) Noble gases: Kr, Xe and T(3H), can be collected from the reactor through a gas-removal system

[Yoshioka 2013, McFarlane 2019], however significant tritium was found in the MSRE moderator graphite [Thoma 1971].

(c) Noble and semi-noble metal fission products: exist in the salt in the metallic state as they are insoluble and unwet by the salt and for this reason they can deposit (or plate out) easily on the structural materials (e.g., Hastelloy N) and on the graphite moderator (these can be: Nb, Mo, Ru, Rh, Sn, Sb, Ag and Te) [Kedl, 1972, Compere 1975, Cheng 2022]. They can be collected through filters [Yoshioka 2013].

The focus of this report is on the fission products that will impact corrosion and salt properties so noble gases may be less relevant, especially because they can be removed by an off-gas system. Due to the historical importance of elements like Te, Cs, I, and Eu on salt compatibility of structural materials, this report will attempt to focus on the proposed mechanisms for the corrosion attack of these important fission products. Their effects have been broken-down into direct and indirect effects and compared to irradiation effects. Proposed mitigation strategies also are summarized.

2.

DIRECT EFFECTS OF FISSION PRODUCTS ON CORROSION Fission products can have both direct and indirect effects on corrosion in molten salts. As will be discussed below, both can lead to the degradation of structural materials. Indirect effects modify the salt potential to make it more corrosive. Direct attack from the halogens in the salt is discussed in Section 2 and indirect attack, e.g., from excess halogens generated by changes in composition during burnup, is discussed in Section 3. Other elements including Ag, Mo, Nb and Ru were found to plate out during the MSRE [Compere 1975] but only Te was found to induce alloy cracking.

2.1 THE EFFECT OF Te AND MECHANISMS OF ATTACK The presence of fission products can affect the durability of structural materials and may induce cracking during operation of a molten salt reactor. Among them, Te-induced cracking was one of the key degradation mechanisms observed in the MSRE. As summarized below, the effect of Te on stainless steels (e.g., Type 304 Stainless Steel (304SS)) and Ni-based alloys, particularly Hastelloy N, has been studied for the past 50 years since the MSRE. That being said, there is not yet an ability to predict the effect of Te on durability.

2.1.1 Te-induced embrittlement in stainless steels The examination of Te effects on stainless steels has not been as extensive as investigations on Ni-based alloys; however, there is increasing interest in using SS structural components (e.g., Type 316H SS) in MSR designs. Some experiments, summarized by McCoy [1978], showed surface roughening of Type 304 SS due to dissolution but no indication of IGC cracking after exposure at 700 ºC, in an MSRE salt (presumably FLiBe or a fuel salt), with additions of Cr3Te4 and Cr5Te6 for 2500 h. These Cr-Te additions were found to produce IGC cracking in Hastelloy N, similar to observations in the MSRE, thus were considered the best laboratory conditions for studying the effect of Te. The salt exposures were accompanied by strain until failure at room temperature. In contrast to the cracking observations for several alternative Ni-based alloys reported by McCoy [1978], Type 304SS was the only alloy tested that did not exhibit IGC cracking. This suggested that SS alloys are less susceptible to Te effects.

5 Figure 2-1. (a) Cross-sectional SEM/EDS mapping results of 304SS sample after immersion test in molten FLiNaK salt containing 0.1 wt% of Te for 48 h at 600 ;(b) effect of Te concentration on maximum penetration depth [Hong 2023].

In contrast to the results of McCoy, examination of Te effects on 304SS was recently reported by Hong et al. [2023]. The experimental investigation included exposure in FLiNaK at 600 °C for 48 h, but the purity of the FLiNaK was not reported. While it can be assumed that the ORNL experiments reported by McCoy were conducted in high purity salt, that assumption may not be true in more recent studies. Tellurium was incorporated into FLiNaK in powder form (99.999 %, ~200 mesh, Alfa Aesar, USA) with target concentrations of 0.1, 0.5 and 1 wt. % [Hong 2023]. IGC was observed on the samples for all Te concentrations. The maximum IGC crack penetration depth of ~190 µm was observed for 0.1 % Te concentration, Figure 2-1. The authors suggested that an increase in Te concentration may lead to an increase in the chromium oxide content found along the alloy GBs, suppressing the extent of IGC, due to its protective character as discussed below, decreased attack with increasing Te content in the salt has been observed in other studies.

2.1.2 The mechanism of Te-induced embrittlement on Ni-based alloys The mechanism of Te attack on Hastelloy N was investigated during and after the MSRE [McCoy, 1972; 1978] and also has been recently revisited [e.g. Surenkov 2020]. An example of the more recent work addresses one of the questions about Te attack: whether elemental Te or Te-rich GB precipitates lead to embrittlement [Liu 2014a]. Some theories suggest that the formation of brittle Cr (CrxTey) and Ni (NixTey) tellurides along the GBs is the prevalent mechanism, as long as these tellurides are thermodynamically stable in the particular salt redox potential [Delpech 2010, McAlpine 2020]. In another instance, Liu et al. [2014b] suggested that the Te-induced cracking along the GBs arises from the fact that the GB Ni-Ni bonds are weakened when Te segregates to the alloy GBs, leading to separation and decohesion. Calculations indicated that the alloy mechanical strength decreases as the Te GB concentration increases, as shown in Figure 2-2a. On the atomic scale, Te is thought to substitute for Ni along the GBs, but its larger atom size induces stresses between neighboring GB atoms. This causes the whole GB structure to expand, as is presented in Figure 2-2b. The newly formed Te-Ni bonds are stronger; however, the existing Ni-Ni bonds weaken following the expansion, making decohesion along the GBs more likely [Liu 2014a].

Returning to the earlier post-MSRE research, ORNL experiments on the effect of Te on Hastelloy N were focused on examining the MSRE simulated fuel salt [Keiser 1976]. Keiser et al. [1976] exposed Hastelloy N to LiF-BeF2-ThF4 with additions of Cr3Te4, at 700 °C, for 24-500 h. It was found that some microscopic fractures were observed along the GBs, at the interface between carbide precipitates and the (b)

6 (a) (b)

Figure 2-2. Effect of Te on the (a) mechanical properties and (b) structure internal stresses on a Ni-based alloy [Liu 2014a]. Gray molecules in (b) represent Ni atoms while blue molecules represent Te additions in the lattice.

alloy matrix [Keiser 1976, Surenkov 2020]. Further examination of the cracked areas revealed a thin Te-rich layer along the GBs. The authors noted that the Te concentration on the GBs was much higher than the value required for brittle IGC cracking. It was suggested that Te penetration should be dependent upon GB chemical and diffusional characteristics and the degree of misorientation of neighboring GBs.

The measured maximum penetration depth was plotted versus exposure time in Figure 2-3, exhibiting a square root relationship, which is characteristic of diffusion processes in the alloy [Keiser 1976].

The experiments simulating the MSRE fuel salt were focused on adding Te as Cr telluride into the salt.

However, McCoy [1978] noted that Te generation in the MSRE was relatively low and slowly increasing during operation. Hence, adding large amounts of Te at the beginning of the experiments was not a representative condition. For this reason, later experiments were focused on examining samples from the MSRE [McCoy 1978] where crack depths of 0.15-0.25 mm were measured. It was observed that Te was the main fission product found along the GBs. Surface sensitive Auger electron spectroscopy detected a concentration of 25 at.% Te at the Hastelloy N GBs, which would be considered sufficient for possible formation of brittle phases on the interfaces, such as Ni3Te2, Figure 2-4.

Figure 2-3. Depth of Te penetration along Hastelloy N GBs as a function of exposure time at 700°C in simulated MSRE fuel salt [Keiser 1976]. The square root relationship with time on this log-log plot suggests a diffusion limited process.

7 Figure 2-4. Ni-Te phase diagram from Keiser [1976].

However, it was suggested that Te diffuses preferentially along the GBs rather than forming a telluride phase. Due to this preferential segregation of Te along the GBs, alloy modifications were suggested with the goal to promote telluride formation, restricting Te diffusion on GBs [McCoy 1978].

Another observation was that the severity of cracking was dependent upon the salt redox potential. For the MSRE salt, Te exhibited low solubility in the salt, easily depositing out onto the structural alloy. This, in turn, would allow Te to react with the alloy walls and ingress along GBs. For this reason, it was suggested that the addition of reducing agents in the salt could inhibit Te deposition and thereby prevent cracking [McCoy, 1978]. In a uranium-based fuel salt, a way to do that is by regulating the ratio of

[UF4]/[UF3]. It has been suggested that this ratio should be kept below 60 [Keiser 1976, McCoy 1978, MacPherson 1985, Ignatiev 2013], with more recent studies suggesting a ratio of below 30-40 [Sureknov 2020]. Sureknov et al. [2020] observed that for ratio values as low as 42 and exposure temperatures of 760 ºC, Hastelloy N exhibits no IGC. On the contrary, at a ratio of 60 and above, and temperatures of 760

ºC and above, IGC was significant, and was increasing with increasing ratio and temperature.

It has been suggested that the effect of Te on Hastelloy N can also be examined by exposing the material to Te vapor without salt [Cheng 2015, Jiang 2022]. For example, Hastelloy N samples were exposed to Te vapor at 800 ºC for 100 h [Cheng 2015, Fu 2017]. The resulting formation of secondary fine Mo-rich M6C precipitates along the alloy GBs is typical of the alloy matrix. The inward diffusion of Te was observed, with accumulation along the GBs and at the carbide-matrix interface. The available Te in these regions reacts with Cr to form the intergranular CrTe intermetallic phase [Cheng 2015]. This phase pins the GBs, minimizing movement, perhaps leading to easier cracking. In addition, the authors suggested that the zig-zag shape of this phase, Figure 2-5, could also lead to easier cracking due to its complicated geometry. However, this hypothesis based on precipitate shape needs to be verified. Increasing the alloy Cr content could potentially lead to formation of a Cr telluride protective surface layer, hindering Te diffusion and minimizing susceptibility to IGC [Cheng 2015]. However, higher alloy Cr contents lead to increased dissolution in halide salts [Raiman 2021].

8 (a) (b)

Figure 2-5. Hastelloy N exposed to Te vapor at 800 ºC for 100 h; (a) Te corroded surface adjacent to an intergranular carbide and (b) zig-zag morphology CrTe layer observed after magnification of white area in (a) [Cheng 2015].

More recent research used Te vapor to investigate its effect on Hastelloy N at 800°C for 150 h [Jiang 2022]. These experiments found an effect of alloy grain size on the Te inward diffusion along alloy GBs.

The vapor formed a complex three-layer reaction product on the specimen surfaces, Figure 2-6. The outer layer consisted of Ni3Te2 and Cr3Te4 and was thicker on the finer grained alloy. The intermediate and inner layers exhibited a variety of phases, such as Ni3Te2, M6C carbides, Fe-rich phase, and a smaller amount of MnTe and Cr3Te4. The inner layer was more continuous for the coarse-grained material. The coarse-grained specimen exhibited deeper Te penetration with a continuous network of larger Cr3Te4, bound to larger M6C carbides on the GBs. With fewer GBs in the coarse-grained alloy, the density of GB carbides was higher in this material.

Fu et al. [2017] suggested that grain boundary engineering (GBE) could improve resistance to Te attack.

For example, cold rolling followed by annealing at 1177 ºC for 0.5 h led to a high-density of twinned GBs (3n), creating large size grain clusters, which may be more resistant to Te ingress.

Even though most of the research related to Te was focused on fluoride-based salts with or without UF4 fuel, there have also been some attempts to examine this material in LiCl-KCl salt with addition of Li2Te at 500 ºC for 120 h [Du 2022]. It is expected that Te will have a higher solubility in a Cl-based salt than a F-based one, hence, it will be much easier to identify the various reactions electrochemically. Exposure of the material in the Cl-based molten salt led to the formation of a two-layer corrosion product. The outer layer consisted of piles of lamellar Ni-Te crystals (Ni2.86Te2, Ni3Te2.07, Ni3Te2), attributed to the outward (a) (b)

Figure 2-6: Reaction products formed on Hastelloy N during exposure to Te vapor at 800 °C for 150 h; (a) fine-grained material and (b) coarse-grained material [Jiang 2019].

9 Figure 2-7: Outer corrosion layer on Hastelloy N exposed to LiCl-KCl-Li2Te, at 500 ºC, for 120 h [Du 2022].

diffusion of Ni, Figure 2-7. The inner layer corresponded to Te diffusion, and was composed mostly of Ni1.43Te, being formed through inward diffusion along the GBs. This layer also exhibited a few microcracks.

Du et al. [2022] was attempting to identify the Te attack mechanism, and two hypotheses are depicted in Figure 2-8. The first hypothesis, Figure 2-8a, assumes that Ni-Te is formed through a Te0 addition to Ni.

The second hypothesis, Figure 2-8b, assumes that there is a Te2-addition to Ni (which then involves release of two electrons that will go elsewhere in the system, with Ni converted to a cation). The first hypothesis is a non-electrochemical route, while the latter hypothesis involves electrochemistry. These two depicted routes will be referred to as the Te0 addition route and the Te2-addition route, respectively.

For the Te0 addition hypothesis, the Te0 would easily incorporate into the alloy matrix based on the following reaction: Te0 + 1.43Ni Ni1.43Te. In order to avoid formation of the tellurides, Te0 can be converted to the less harmful Te2-by adjusting the salt redox potential, through control of the ratio

[U4+]/[U3+], according to the following reaction where elemental Te reduces to an ion: Te0 + 2U3+ Te2-

+ 2U4+.

Alternatively, for Te2-addition hypothesis, the reaction is Te2- + 1.43Ni Ni1.43Te+ 2e-. In this case, mitigation becomes feasible if the salt redox potential becomes more negative than the most negative equilibrium potential for telluride formation. It has been measured that ENi1.43Te/Te2- < ETe/Te2-. For this particular LiCl-KCl salt, the [U4+]/[U3+] ratio which prevents Te attack is calculated as0.0003 [Du 2022].

The authors concluded that the Te2-hypothesis appeared to be the most likely mechanism because there was evidence to suggest that tellurium attack was only favored at potentials more positive than the electrode potential of Ni1.43Te/Te2-.

(a) (b)

10 Figure 2-8: Proposed formation mechanisms for surface Ni-Te; (a) Te0 addition hypothesis and (b) Te2-addition hypothesis [Du 2022].

McAlpine et al. [2020] examined four commercial alloys (Hastelloy N, 800H, 316L stainless steel, Ni-201) in 3 different FLiNaK based molten salt compositions in order to investigate the effect of realistic concentrations of Te only (as 0.14 % NiTe) and Te plus other simulant fission products on corrosion.

EuF3 was also added in order to create a highly oxidizing salt leading to accelerated corrosion. The experiments were performed at 700 ºC for up to 150 h. Neither of the Te-containing salts increased the amount of the attack and Hastelloy N and 316L showed typical intergranular attack. The Ni-201 specimens were not significantly attacked in any salt.

In summary, the extent of Te attack depends on the salt redox potential. For fluoride fuel salts, the ratio of

[UF4]/[UF3] should be kept within a particular range (e.g., 40-60). Tellurium reacts with the alloy surface to form Te-containing compounds, by a mechanism which is not yet established, but is hypothesized to be via a Te2-addition process. After the surface incorporation of Te to form nickel telluride, Te diffuses along alloy GBs, which embrittles the alloy. The embrittlement may be due to the formation of GB phases or a Te layer on the alloy GBs.

2.1.3 Modifications of Hastelloy N composition for improved resistance against Te-induced IGC cracking Because the degradation observed in the MSRE was not acceptable for the more aggressive MSBR conditions, new alloy development was explored. A reasonable strategy to disrupt Te embrittlement along Hastelloy N GBs is to add alloy additions that disrupt the embrittlement mechanism. It was found that Hastelloy N modified with 1-2 % Nb was more resistant to both irradiation embrittlement, Figure 2-9a, and Te-induced IGC cracking, Figure 2-9b [McCoy 1978]. This is because Nb leads to a dispersion of Te over the entire alloy, inhibiting local segregation as a Te-enriched film along the alloy GBs [Keiser 1976].

However, further examination of the effect of Nb to Te-induced IGC cracking as a function of the exposure time to fuel salt showed an increase in IGC cracking susceptibility with increased exposure time, Figure 2-10. For this reason, extrapolation to the planned MSBR lifetime of ~30 yr (250,000 h) might not show promising results [McCoy 1978]. Other modifications that have been found to be promising include mixtures of Nb and Ti [DeVan 1994] and 1 % Al, which has been found to lead to the most resistant alloy against Te-induced IGC cracking [Ignatiev 2008].

11 (a) (b)

Figure 2-9: Effect of Nb concentration on Hastelloy N performance; (a) mechanical properties [McCoy 1978]

and (b) stress corrosion cracking [Keiser 1976].

Figure 2-10: Effect of Nb concentration on IGC cracking as a function of exposure duration to fuel salt

[McCoy 1978]

More recent work has explored improved mechanistic understanding of the Nb benefit observed in Hastelloy N. Liu et al [2014] examined the effect of Nb on the atomic level and how Te behaves with it. It

12 was suggested that Te does not form a surface reaction layer with Nb and Nb itself does not hinder Te diffusion along alloy GBs. It was suggested that since Nb atoms have a larger diameter than Ni, they will most likely occupy substitutional sites similarly to Te atoms. In order to further examine the effect of Te and Nb on the GBs, the strengthening/embrittling energy E was estimated for the two elements and it was found to be positive for Te, classifying it as an embrittler, and negative for Nb, classifying it as a GB cohesion enhancer. Indeed, examination of the tensile strength of the material seemed to improve with Nb additions, Figure 2-11a. Further examination of the mechanism revealed that this could be related to the strong Nb-Ni bonds developed along the GBs, that enhance the cohesion strength in the area. This could arise from the fact that, even though both Te and Nb induce stresses on the GBs, due to their larger size, the stress induced by Te is larger and is related to the particular substitution sites for each one of them, Figure 2-11b [Liu 2014a]. In the case of Te, Te-Ni bonds become quite strong, but the respective Ni-Ni bonds weaken, and it is easier for them to separate under stress. In the case of Nb, Nb-Ni bonds are strong but also act beneficially against decohesion. The segregation also can limit Te migration along the GBs, limiting vulnerability to IGC cracking.

(a)

(b)

Figure 2-11: Mechanism of beneficial Nb additions in Hastelloy N, in an atomic level; (a) effect of Nb additions in the tensile properties and (b) stress development in the GB area due to additions of Te and Nb with Ni atoms (Ni shown as unlabeled red circles); Left-clean GB without additions; Center -after addition of Te (blue); Right - after additions of larger Nb atoms (red, labeled). [Liu 2014a].

As mentioned previously, the salt redox potential plays a major role on Te-related corrosion performance of Ni-based alloys [Ignatiev 2008, 2012, 2016]. The conditions examined by Ignatiev et al. included an MSR fuel salt LiF-BeF2-20%ThF4-2%UF4, with Te additions as Cr3Te4. The examination temperature was 735-750 °C and the exposure period was 250 h. Various [UF4]/[UF3] ratio values were examined, including 0.7, 4, 20, 100 and 500, in order to represent very reducing to very oxidizing conditions. After the end of the exposure, mechanical loading was applied ranging from 0 to 25 MPa, to assess the performance of the material. For reducing and slightly oxidizing salt, i.e. 0.7 [UF4]/[UF3] < 4, a 2-layer

13 corrosion product structure was observed with the external layer composed of U12Ni78 (-phase). The general observation about the outer layer is that it included U-deposition and U(Ni, Mo, W) surface intermetallics, with their particular type being dependent upon the specific alloy composition. The internal layer was characterized by U diffusion until depletion in U-Ni phases. The U-depleted area appeared rich in Cr and other alloying elements. No Te-induced IGC cracking attack was observed in this case. For moderately oxidizing conditions, i.e. [UF4]/[UF3] = 100, no films or deposits were observed on the material, Figure 2-12a. No Te-induced IGC cracking was observed with or without applied stresses.

For more severe oxidizing conditions, i.e. [UF4]/[UF3] = 500, no films or deposits were observed, but Te-induced IGC cracking was significant, Figure 2-12b. The intensity of attack increased with the increase in the applied stress after the end of the exposure in the fuel salt. A study found that additions other than Nb could be beneficial against Te attack in Ni-Mo alloys like Hastelloy N [Ignatiev 2008]. Additions of 1.1 % Al and 0.9 % Ti were found to provide the best resistance against Te-induced IGC cracking, for all salt redox conditions.

(a) (b)

Figure 2-12: Effect of salt redox potential of a fuel salt on Te-induced IGC cracking on Nb-modified Hastelloy N; (a) [UF4]/[UF3] = 100 and (b) [UF4]/[UF3] = 500 [Ignatiev 2008].

2.2 THE EFFECT OF CESIUM AND IODINE AND MECHANISMS OF ATTACK Cesium and iodine, along with tritium (especially for Li containing salts), are considered soluble and volatile and, therefore, have been investigated in more detail than other elements. Iodine can form various stable fluorides, such as IF5 and IF7, which have boiling points of 97.8 and 4.8 °C, respectively

[McFarlane 2018]. Regarding Cs, the chemical form that it adopts in the salt will dictate its volatility

[Capelli 2018, McFarlane 2018]. Cesium fluoride (CsF) and cesium iodide (CsI) are considered among the most stable forms of Cs in a fuel salt, with CsI solubility being much less [Capelli 2018]. CsI can generally be retained in the salt for a CsI/FLiNaK ratio of 1-100 mol [Taira 2017]. At higher concentrations, CsI might react with the fuel salt to form the gas phase KI [Taira 2017]. The possible stable chemical forms of both Cs and I are presented in Table 2-1.

Yamawaki et al. [2015] examined the effect of fission product CsI in FLiNaK on the corrosion performance of Hastelloy N, after exposure at temperatures of 773 to 973 K for 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />, Figure 2-13.

The author reported that salt exposures with an Ar cover gas led to ~50µm deep attack with or without CsI in the salt. As a result, it appears that the fission product CsI does not have an impact on the corrosion behavior of structural materials.

Table 2-1. Possible stable chemical forms of Cs and I in molten halide salts [Capelli 2018]

Fission products Chemical species Cesium CsI CsF CsCl Elemental gases (Cs, Cs2) (not stable)

Iodine CsI LiI ThI4

14 Elemental gases (I, I2)

IFx (g)

ICl (g)

Figure 2-13. Depth of corrosive attack on Hastelloy N after exposure at various temperatures and crucible types for 100 h [Yamawaki 2015].

2.3 THE EFFECT OF Eu AND MECHANISMS OF ATTACK Depending on the type of reactor, length of operation and power, lanthanides (e.g. Ce, Nd, Eu) can constitute up to 25 % of the total fission product inventory in a molten salt reactor. Among the rare-earth fission products [Taube 1974, Compere 1975, Aufiero 2013], Eu is of particular interest, due to its ability to form both 2+ and 3+ oxidation states. Previous work showed that this can be an advantageous property that can be used to control the redox potential of the salt [Zhang 2018]. Other studies have used EuF3 as an impurity in fluoride salts to study its effects on corrosion of Hastelloy N in FLiNaK salt [McAlpine, 2020]. However, the EuF3 addition did not significantly change the depth of attack after 150 h at 700ºC.

A similar study was performed at ORNL where coupons of unalloyed Cr were exposed in chloride salt (KCl-MgCl2) with and without the addition of Eu fission product surrogates [Sulejmanovic 2023].

Additions of 0.1 mol% of Eu metal or EuCl3 were made to the baseline salt to investigate their effects on corrosion of Cr coupons at 600°C and 700°C for 1000 h. Each Cr specimen was exposed in 30 g of KCl-MgCl2 salt using an established methodology for salt compatibility studies in Mo capsules [Raiman 2019]. Figure 2-14 shows the specific mass change results for the exposed Cr coupons (exposed in triplicate) in the three different conditions and two different temperatures. The baseline KCl-MgCl2 condition resulted in low mass losses at both 600°C and 700°C, indicating low corrosivity. However,

15 Figure 2-14. Specific mass change for the Cr specimens exposed in three different salt conditions at 600°C and 700°C [Sulejmanovic 2023].

when Eu additives were introduced, the results were more complicated. The data showed increased scatter, especially at 700°C exposures, and the mass loss of the Cr specimens increased for the EuCl3 case. Unexpectedly, the addition of Eu metal resulted in greater mass loss (in some cases) compared to the baseline salt condition. This contradicted the expected corrosion inhibition properties of Eu metal forming chlorides and reducing the salt potential.

Figure 2-15 shows representative scanning electron microscopy (SEM) images and elemental mapping of the cross-sectioned Cr coupons after exposure to the three different conditions at 700°C for 1000 h. With the addition of Eu metal, a uniform MgO layer formed on the surface of the Cr specimens. The combination of Cr dissolution and mass gain associated with the oxide layer formation still resulted in a net mass loss in Figure 2-14. Presumably, the oxide formed due to O surface contamination of the Eu metal, which affected the reaction. Free O in the salt from the Eu addition reacted with Mg and MgO deposited on the specimen surface. The MgO layer was confirmed using x ray Diffraction (XRD), Figure 2-16. In contrast to the Eu addition, the addition of EuCl3 increased corrosivity due to the decomposition of EuCl3 to EuCl2 and release of chlorine that elevated the chlorine potential of the salt:

EuCl3 + e- > EuCl2 + Cl-This could lead to the dissolution of CrCl2/CrCl3 into the salt melt, with Cl-being a possible reactant for pitting on the surface of the Cr specimens. Thus, fission products that are soluble in the salt can affect compatibility. The details of this experiment with O from the Eu metal addition forming MgO on the Cr specimens also suggests that experiments to study fission product effects need to be carefully designed.

16 Figure 2-15. Cross-sectional SEM and elemental mapping for Cr specimens exposed in the baseline KCl-MgCl2 salt (top row), KCl-MgCl2 + Eu metal (middle row) and KCl-MgCl2 + EuCl3 (bottom) row at 700°C

[Sulejmanovic 2023].

Figure 2-16. XRD powder plots of the Cr specimen after the 600°C 1,000-h exposure in KCl-MgCl2+Eu salt condition (blue line) showing evidence of MgO on the surface. The orange line is showing a simulated XRD powder pattern for the MgO polycrystals [Sulejmanovic 2023].

3.

INDIRECT EFFECTS OF FISSION PRODUCTS ON CORROSION In contrast to the deposition attack noted in the previous section, fission products also can have indirect effects on corrosion in molten salts. One such effect is their impact on the chemical composition of the molten salt itself. As fission products accumulate in the salt, they can alter its chemical properties, such as its redox potential, which can in turn affect the corrosion of metal components. [Thoma 1971, Baes 1974, Walker 2021]. For example, tritium could lead to TF formation [Mays 1977, Kondo 2010].

The following two paragraphs are based on a prior NRC report [Raiman 2021]. In assessing the effect of fission products in salt-fueled reactors, the original oxidation state of the fissile material is important

[Chasanov 1965, Robinson 1958a, 1958b]. Using 235UF4-containing salts as a baseline, the oxidation state of U is +4. Since fission produces two atoms for every one that splits, the original average valence state of the fission products is considered +2. If the element is less stable at the same oxidation state or is more stable in a lower oxidation state, then the salt becomes more oxidizing. In contrast, if an element in the decay chain is more stable at the same oxidation or with a higher oxidation state (e.g., +3), then the salt becomes more reducing and less likely to corrode the structure. The decay chain and initial oxidation state must be considered together to fully understand their effect on corrosion. Robinson [1958a] described calculations used to determine the evolution of average valence over time for reactor fuels containing UF4:

+2+ + (2 )

= 1 2

17 where V is the original oxidation state of U in the fuel, V is the average oxidation state of the fission products, and yz and Vz are the fission yield and oxidation state of the element with atomic number Z, respectively. Since two fission products are produced for every atom split (U in this case) and thus requires 1/2 in front of the average oxidation state. If the difference between twice the average oxidation state of the fission products and the original oxidation state is positive, then the result will likely be reducing in nature. Chasanov [1965] performed the same calculations for (Pu,U)Cl3 systems and obtained similar results to Robinson using the oxidation states shown in Table 3-1.

Table 3-1. Oxidation states of the fission product elements in molten chlorides with 239Pu as the fissile element [from Chasanov [1965]

Oxidation State Elements

-2 Se, Te

-1 I

0 Kr, Xe

+1 Rb, Ag, In, Cs

+2 Sr, Zr, Mo, Pd, Cd, Sn, Ba, Sm, Eu

+3 Y, Tc, Ru, Rh, Sb, La, Ce, Pr, Nd, Pm, Gd

+5 Nb Using the table as a baseline, Chasanov calculated the average fission product oxidation versus time, Figure 3-1. However, elements such as Ru, Pd, Tc, and Mo are readily reduced to their metallic state.

Curves b, c, and d account for that, and the curve is pushed down, indicating a lower average oxidation state. If the curve is below the dashed line at 1.5 (half of +3, oxidation state of Pu), the salt is oxidizing.

The average fission product valence only accounts for the change in valence but does not reflect the change in Cl or F potential as fission progresses. The calculation is more difficult but better represents the change in corrosivity over time. Chasanov does mention the effect the change in salt chemistry as a result of fission has on corrosion of the containment material.

Figure 3-1. Fission product average valency for 239PuCl3 irradiated with 2-MeV neutrons with rare gases retained in system. (a) fully oxidized. (b) except Ruº. (c) except Ruº and Pdº. (d) except Ruº, Pdº, Tcº and Moº

[Chasanov 1965].

18 Based on the previous analysis, it is desirable to keep the salt in a reducing state to minimize effects on corrosion. This is feasible by regulating the salt redox potential, for example by adding Be metal in a FLiBe salt [Thoma 1971] or controlling the [UF4]/[UF3] ratio [Keiser 1976, Surenkov 2020, Ignatiev 2008, 2013, 2016]. Alternatively, the ratio [UF3]/[U], where U is the sum of both UF3 and UF4 in the salt, has also been suggested as a criterion with a recommended value of 0.1-1.7 % [Thoma 1971, Compere 1975, Geng 2022].

It also was found that 95Nb could serve as an indicator for the oxidation state of the salt. Fission product 95Nb was found to be deposited in the form of NbC on the graphite moderator during the MSRE program, when fuels 235U and 238U were used [Rosenthal 1969, Compere 1975]. During the MSRE operation it was found that, as the redox potential of the salt increased, so did the activity of 95Nb. For this reason, it was suggested that 95Nb could serve as a salt oxidation state indicator [Thoma 1971]. In particular, when the ratio [UF3]/[U] was below 0.5 %, 95Nb appeared in the salt in the form of soluble fluoride. When the ratio was above 0.5 %, 95Nb was not detected in the salt anymore, because it was depositing out in metal form [Thoma 1971, Compere 1975, Geng 2022]. Recently, it has been suggested that 131I [Geng 2021] or 99Mo [Cheng 2022] could also serve as indicators for monitoring the salt redox potential.1 Capelli et al. [2018] studied how fission products could affect salt properties and calculated a series of binary and ternary phase diagrams of the LiF-ThF4-CsF-LiI-ThI4-CsI system, some of which are shown in Figure 3-2. The authors suggested that the effect of CsF on the liquidus temperature and vapor pressure would be minimal and Cs remains dissolved in the F-based matrix. The noble and semi-noble fission products will precipitate from the salt, which could potentially affect heat transfer [Holcomb 2017].

(a) (b)

Figure 3-2. Liquidus projection of (a) the LiF-ThF4-CsF system and (b) the Li-Th-F-I system [Capelli 2018].

4.

COMPARISON OF FISSION PRODUCT EFFECTS TO IRRADIATION EFFECTS Irradiation can impact the mechanical properties of metal components in an MSR and was a particular concern for Ni-based Hastelloy N in the MSRE and the higher fluence anticipated in the MSBR [DeVan 1994]. For example, the accumulation of He caused embrittlement, when trapped as bubbles that accumulated on alloy GBs [McCoy 1972, Martin 1966]. This irradiation embrittlement can affect both Ni-1 Note: Some of the obtained fission product isotopes, such as 99Mo, could also have medical interest, so chemical extraction techniques could be set in place to be able to recover these isotopes [Cheng 2022].

19 based alloys and Fe-based stainless steels (typically 8-12%Ni), and it manifests itself as a decrease in plasticity at temperatures higher than 500 °C [Ignatiev 2012].

Helium production results from thermal neutron interaction with 10B, which can be found as an impurity in Hastelloy N [McCoy 1972]. However, another source could be from reactions between the major alloying elements and fast neutrons with energies around 3 MeV [Martin 1966]. Nickel is a particular concern with 68% of natural nickel being 58Ni and the following two-step process identified [Ignatiev 2012, Wright 2018]:

+

59 58

+

4 59

+

56 Type 304 steel with 8-10% Ni also can be susceptible to irradiation induced embrittlement. It has been found that small additions of Ti (similar to Type 321 SS) can improve the resistance of this material, by formation of stable precipitates with 10B. This, in turn, would limit formation of He along the GBs, since it would mostly be located in the areas where these compounds precipitate, i.e. homogenously within the matrix. The effect of Ti on the ductility of 304 SS is presented in Figure 4-1.

As noted in Section 2.1.3 above, Nb (Figure 2.9a) and Nb and Ti additions to Hastelloy N can be beneficial in irradiation resistance. Additional mitigation strategies included modification of the alloy composition, in order to form fine carbide precipitates, such as MC, M=80-90%Mo-Cr, that could hinder He migration [MacPherson 1985].

Figure 4-1. Effect of Ti concentration on ductility of irradiated austenitic SS at 842 °C [Martin 1966].

20

5.

STRATEGIES TO MITIGATE THE EFFECTS OF FISSION PRODUCTS Several fission product mitigation strategies have been mentioned already, but are summarized here:

Operational practices to minimize the accumulation of fission products and other contaminants:

Historically, the removal of gaseous fission products from the MSRE primary fuel loop salt (PFLS) was performed through an off-gas system [Vincente Valdez 2020] that included charcoal beds [Holcomb 2017]. Application of a cover gas handling system could result in 40% of gaseous FPs being trapped in the cover gas [Holcomb 2017]. However, performance of off-gas materials during reactor operation is also a consideration in reactor design. A representation of a proposed off-gas system for an MSR has been suggested by McFarlane et al. [2019], Figure 5-1. The first component includes the molten hydroxide scrubber, which could trap and dissolve alkaline species (e.g. Cs, La), noble metals (e.g. Ru, Rh, Te),

halogens (i.e. F2, Cl2, Br2, I2) and acidic species (e.g. TeO2, HF, HCl). After that there would be a trap for long-lived components, such as 36Cl and 129I. Tritium trapping is an issue in MSRs, since it can diffuse through metals and escape into the environment [Mays 1977]. In the present conceptualization, tritium that escapes from the scrubber as water can be trapped via molecular sieves and the remaining T will stay in the solution as 3HO-. Another option for tritium trapping could be through a closed Brayton cycle using He [Forsberg 2006]. Short-lived isotopes of Kr and Xe can decay on activated charcoal noble gas delay beds. The more stable forms Xe, Ar and 85Kr coming out of the beds could be separated from the He carrier gas through pressure swing or cryogenic distillation. Tritium could be removed from the remaining He before it is recycled back to the reactor system [McFarlane 2019].

Use of corrosion-resistant materials: In order for a material to be deemed successful for use in an MSR it should exhibit good resistance to corrosion and radiation in this high-temperature application [Wright 2018]. Hastelloy N was developed for molten salt applications and has long been examined as a candidate material because of its good corrosion resistance. However, the irradiation resistance of Ni-based alloys can be a concern in some MSR designs. In the MSRE, Hastelloy N was susceptible to Te-induced intergranular cracking and He-induced embrittlement. Alloy N modifications including small additions of Nb, Al and Ti were investigated to improve its resistance to both irradiation and Te cracking. In order to mitigate He-induced embrittlement, treatments such as cold working, and dispersions of fine precipitates also could improve the alloy [Wright 2018]. Four types of alloys similar to Hastelloy N have been patented in the past decade with improved creep strength due to alloying to promote the formation of carbides or ' strengthening precipitates [Muralidharan 2017, 2022]. These new alloys increase the creep life at 650°-850°C by 2-10+ times compared to Hastelloy N.

More exotic materials solutions have been suggested including carbon-carbon composite [Forsberg 2006]

and SiC-SiC composite heat exchangers to mitigate noble metal deposition. A cost-benefit analysis can identify if these strategies might be effective.

Control of the chemical potential through active elements: For direct (e.g., Te) and indirect fission product effects, control of the redox potential of the salt is vital to sustaining long-term performance in an MSR. As mentioned previously, the ratio of [UF4]/[UF3] also can be optimized to minimize IGC cracking.

Additions of Be metal have been found to have a beneficial, reducing effect on FLiBe-based salts

[Ignatiev 2013, Keiser 1979]. However, Be can have negative consequences, including the formation of Ni-Be compounds in steels [Keiser 2022]. It appears that a threshold Be level where negative effects begin to be observed has not been defined.

21

6.

SUMMARY

Fission products have differing levels of impact on material compatibility in molten salt reactors. Specific fission products depend on multiple factors including the salt and fuel type, the neutron energy spectrum, and the reactor operating conditions. The most common fission products in nuclear reactors are noble gases, rare-earth elements, and halogens. It can be useful to classify them based on elements that are soluble in the salt as ions and those that remain in the metallic state and can deposit on the structural materials or graphite moderator.

The effects of fission products on compatibility have been broken-down into two categories, direct and indirect effects. Direct effects may be best illustrated by the most widely studied fission product Te, which induced cracking in Hastelloy N in the MSRE by segregating to alloy grain boundaries. However, the exact mechanism of surface deposition and diffusion along alloy grain boundaries is still being investigated and there are conflicting results about the susceptibility of stainless steels to Te-induced cracking. Cesium and iodine also are common fission products but do not appear to accelerate corrosion when added to the salt in laboratory experiments. Rare earth fission products also might accelerate corrosion, especially multivalent elements like Eu.

Indirect fission product effects focus on the impact on the chemical composition and physical properties of the salt itself. No examples were found of salt physical properties like heat capacity or melting point affected by fission products. Fission products are generated slowly during operation and thus a removal strategy could be part of the reactor design. The major indirect effect is on the salt redox potential, which is well-known to affect compatibility.

General fission product mitigation strategies are summarized. Improvements for Hastelloy N continue to be investigated including additions of Nb, Al and Ti to address concerns about Te-induced cracking as well as He embrittlement from irradiation, which also was a major concern after the MSRE. However, the mechanisms for the beneficial effect of these additions have not been resolved. In addition to the strategy of selecting or developing more compatible materials, other mitigation strategies include controlling the salt potential using the [UF4]/[UF3] ratio or adding active elements such as Be. This can mitigate both direct and indirect effects. Operational practices also can be designed to minimize the accumulation of fission products including off-gas systems to capture noble gases and noble metals.

The various classes of fission products need to be considered including the noble metals, rare earths and those that are soluble in salt to confirm their effect, particularly on stainless steels which are the main structural material being considered by current MSR developers. Evaluations should include 316H stainless steel and Alloy 709 (Nb-modified Fe-20Cr-25Ni). Current data on Te-induced cracking of stainless steels vary widely. Chloride fuel salts are untested in core environments where there is the potential for development of undesirable phases and compounds [Holcomb 2017]. Because of the importance in controlling the redox potential of the salt, commercial in situ electrochemical monitoring systems compatible with chloride and fluoride coolant and fuel salts can be used to demonstrate operating ranges where fission product effects can be minimized.

22

7.

REFERENCES Aufiero, M., et al., (2013) An extended version of the SERPENT-2 code to investigate fuel burn-up and core material evolution of the Molten Salt Fast Reactor. Journal of Nuclear Materials, 441: 473-486.

Baes Jr, C. (1974). "The chemistry and thermodynamics of molten salt reactor fuels." Journal of Nuclear Materials 51(1): 149-162.

Briggs, R. (1964). Molten-salt reactor program semiannual progress report for period ending January 31, 1964, ORNL Report ORNL-3626, Oak Ridge National Laboratory, Oak Ridge, TN.

Capelli, E., et al. (2018). "Thermodynamics of soluble fission products cesium and iodine in the Molten Salt Reactor." Journal of Nuclear Materials 501: 238-252.

Chasanov, M. G. (1965) Fission-Product Effects in Molten Chloride Fast-Reactor Fuels, Nuclear Science and Engineering, 23(2), 189-190.

Cheng, H. W., et al. (2015). EPMA and TEM characterization of intergranular tellurium corrosion of Ni-16Mo-7Cr-4Fe superalloy. Corrosion Science 97: 1-6.

Cheng, Z., et al. (2021). "Distribution and behavior of fission product 95Nb in FLiBe salt." Radiochimica Acta 109(4): 311-317.

Cheng, Z., Zhao, Z., Geng, J., Li, W., Dou, Q., Li, Q., (2022). "The distribution and behavior of noble metal fission products (99Mo, 103Ru, 105Rh and 132Te) in molten 2LiF-BeF2 mixtures." Journal of Nuclear Materials 566: 153807.

Compere, E., et al. (1975). Fission product behavior in the molten salt reactor experiment, ORNL Report ORNL-4865, Oak Ridge National Laboratory, Oak Ridge, TN.

Delpech, S., et al. (2010). "Molten fluorides for nuclear applications." Materials Today 13(12): 34-41.

Du, X., et al. (2022). "Mechanism of tellurium induced nickel alloy corrosion in molten LiCl-KCl salt."

Corrosion Science 209: 110734.

Forsberg, C. W. (2006). Molten-salt-reactor technology gaps. In Proc. of 2006 International Congress on Advances in Nuclear Power Plants (ICAPP06).

Fu, C. T., et al. (2017). "Grain boundary engineering for control of tellurium diffusion in GH3535 alloy."

Journal of Nuclear Materials 497: 76-83.

Geng, J., et al. (2021). "Transient release of radioactive iodine from the fission of UF4 in 2LiF-BeF2 salt." RSC advances 11(37): 22611-22617.

Geng, J., et al. (2022). "Activity Measurement of Iodine Fission Products in 2LiF-BeF2 Salt and Its Application in Redox Potential Surveillance for a Thorium Molten Salt Reactor." Inorganic Chemistry 61(19): 7406-7413.

Grimes, W. (1970). "Molten-salt reactor chemistry." Nuclear Applications and Technology 8(2): 137-155.

Guo, S., Zhang, J., Wu, W., Zhou, W., (2018) Corrosion in the Molten Fluoride and Chloride Salts and Materials Development for Nuclear Applications. Progress in Materials Science, 97, 448-487.

Holcomb, D. (2017). Module 3: Overview of Fuel and Coolant Salt Chemistry and Thermal Hydraulics, presentation, https://www.nrc.gov/docs/ML1733/ML17331B115.pdf.

Hong, M., et al. (2023). "Effect of tellurium concentration on the corrosion and mechanical properties of 304 stainless steel in molten FLiNaK salt," Corrosion Science 212: 110913.

Ignatiev, V., et al. (2008). "Compatibility of selected Ni-based alloys in molten Li, Na, Be/F salts with PuF3 and tellurium additions." Nuclear technology 164(1): 130-142.

Ignatiev, V. V., et al. (2012). "Molten-salt reactors: new possibilities, problems and solutions." Atomic Energy 112(3): 157-165.

Ignatiev, V. and A. Surenkov (2013). "Alloys compatibility in molten salt fluorides: Kurchatov Institute related experience." Journal of Nuclear Materials 441(1-3): 592-603.

Ignatiev, V., et al. (2016). "Experimental investigation of tellurium corrosion of nickel-molybdenum alloys in molten lithium-, beryllium-, and uranium-fluoride salts." Atomic Energy 120: 397-402.

23 Jiang, L., C.-T. Fu, B. Leng, Y.-Y. Jia, X.-X. Ye, W.-Z. Zhang, Q. Bai, S. Xia, Z.-J. Li, F. Liu, X.-T.

Zhou, (2019) Influence of grain size on tellurium corrosion behaviors of GH3535 alloy, Corrosion Science 148: 110-122.

Jiang, L., et al. (2022). "Tellurium segregation-induced intergranular corrosion of GH3535 alloys in molten salt." Corrosion Science 194: 109944.

Kedl, R. (1972). Migration of a class of fission products (noble metals) in the Molten-Salt Reactor Experiment, ORNL Report ORNL-TM-3884, Oak Ridge National Laboratory, Oak Ridge, TN.

Keiser, J., et al. (1976). Corrosion resistance of some nickel-base alloys to molten fluoride salts containing UF4 and Tellurium." ECS Proceedings Volumes 1976(1): 315.

Keiser, J. R. (1977). Status of Tellurium-Hastelloy N Studies in Molten Fluoride Salts, ORNL report ORNL/TM-6002, Oak Ridge, TN.

Keiser, J. R., DeVan, J. H., Lawrence, E. J. (1979) Compatibility of Molten Salts with Type 316 Stainless Steel and Lithium, J. Nucl. Mater., 85-86: 295-298 Keiser, J. R., Singh, P. M., Lance, M. J., Meyer III, H. M., Myhre, K. G., Lowe, T. M., Sulejmanovic, D.,

Cakmak, E., Cox, V. A., Hawkins, C. S., Willoughby, A. W. (2022) Interaction of beryllium with 316H stainless steel in molten Li2BeF4 (FLiBe), Journal of Nuclear Materials 565, 153698.

Kondo, M., Nagasaka, T., Tsisar, V., Sagara, A., Muroga, T., Watanabe, T., Oshima, T., Yokoyama, Y.,

Miyamoto, H. Nakamura, E. (2010) "Corrosion of reduced activation ferritic martensitic steel JLF-1 in purified Flinak at static and flowing conditions," Fusion Engineering and Design 85:

1430-1436.

Konings, R. and R. Stoller (2020). Comprehensive nuclear materials, Elsevier, Oxford, UK.

LeBlanc, D. (2010). "Molten salt reactors: A new beginning for an old idea." Nuclear Engineering and Design 240(6): 1644-1656.

Liu, W., et al. (2014a). "First-principles study of intergranular embrittlement induced by Te in the Ni 5 grain boundary." Computational materials science 88: 22-27.

Liu, W. G., et al. (2014b). "The effect of Nb additive on Te-induced stress corrosion cracking in Ni alloy:

a first-principles calculation." Nuclear Science and Techniques 25(5).

Luzzi, L., et al. (2012). Multi-Physics Approach to the modeling and analysis of Molten Salt Reactors, Nova Science Publishers, Inc.

MacPherson, H. (1985). "The molten salt reactor adventure." Nuclear Science and engineering 90(4):

374-380.

Martin, W. and J. Weir (1966). Solutions to the Problems of High-Temperature Irradiation Embrittlement, ORNL Report ORNL-TM-1544, Oak Ridge National Laboratory, Oak Ridge, TN.

Mays, G., et al. (1977). Distribution and behavior of tritium in the coolant-salt technology facility ORNL Report ORNL/TM-5759, Oak Ridge National Laboratory, Oak Ridge, TN.

McAlpine, S. W.; Skowronski, N. C.; Zhou, W.; Zheng, G. T.; Short, M. P. (2020) Corrosion of commercial alloys in FLiNaK molten salt containing EuF3 and simulant fission product additives.

Journal of Nuclear Materials 532, 151994.

McCoy, H., McNabb, B. (1972). Intergranular Cracking of INOR-8 in the MSRE, ORNL Report ORNL/TM-4829, Oak Ridge National Laboratory, Oak Ridge, TN.

McCoy Jr, H. (1978). Status of materials development for molten salt reactors, ORNL Report ORNL/TM-5920, Oak Ridge National Laboratory, Oak Ridge, TN.

McFarlane, J., et al. (2018). Thermochemical and transport properties important to molten salt reactor operation: Off-gas performance and the fission product mechanistic source term, ORNL Report ORNL-TM-2018-958, Oak Ridge National Laboratory, Oak Ridge, TN.

McFarlane, J., et al. (2019). Fission Product Volatility and Off-Gas Systems for Molten Salt Reactors ORNL Report ORNL/TM-2019/1266, Oak Ridge National Laboratory, Oak Ridge, TN.

Muralidharan, G., Wilson, D. F., Holcomb D. E., (2017). Intermediate Strength alloys for high temperature service in liquid-salt cooled energy systems, US Patent #9683280.

24 Muralidharan, G., et al. (2022). New High-Strength Ni-based Alloys for High Temperature Service in Liquid Fluoride Salt Environments, ORNL Report ORNL/TM-2022/2620, Oak Ridge National Laboratory, Oak Ridge, TN.

Raiman, S. S., Mayes, R. T., Kurley, J. M., Parrish, R., Vogli, E., (2019) Amorphous and partially-amorphous metal coatings for corrosion resistance in molten chloride salt, Solar Energy Materials and Solar Cells 201: 110028.

Raiman, S. S., Sulejmanovic, D., Kurley, J. M., Lee, J. J., Parker, C., Pint, B. A., (2021). Technical Assessment of Materials Compatibility in Molten Salt Reactors, NRC Report TLR-RES/DE/CIB-CMB-2021-03.

Robinson, M. T. (1958a). On the Chemistry of the Fission Process in Reactor Fuels Containing UF4 and UO2. Nuclear Science and Engineering, 4, 263-269.

Robinson, M. T., Brooksbank Jr., W. A., Reynolds, S. A., Wright, H. W., Handley, T. H. (1958b) The Behavior of Fission Products in Molten Fluoride Reactor Fuels, Nuclear Science and Engineering, 4, 288-296.

Rosenthal, M., et al. (1969). Molten-salt reactor program semiannual progress report for period ending February 28, 1969, ORNL Report ORNL-4396, Oak Ridge National Laboratory, Oak Ridge, TN.

Rosenthal, M., et al. (1970). "Molten-salt reactorshistory, status, and potential." Nuclear Applications and Technology 8(2): 107-117.

Rosenthal, M. W., et al. (1972). Molten-Salt Reactor Program: Semiannual Progress Report for Period Ending August 31, 1971, ORNL Report ORNL-4728, Oak Ridge National Laboratory, Oak Ridge, TN.

Sulejmanovic, D., et al. (2023) unpublished work, Oak Ridge National Laboratory, Oak Ridge, TN.

Surenkov, A., et al. (2020). "Effect of the [U(IV)]/[U(III)] ratio on selective chromium corrosion and tellurium intergranular cracking of Hastelloy N alloy in the fuel LiF-BeF2-UF4 salt." Epj Nuclear Sciences & Technologies 6.

Taube, M. and J. Ligou, (1974) Molten plutonium chlorides fast breeder reactor cooled by molten uranium chloride. Annals of Nuclear Science and Engineering, 1:277-281.

Taira, M., et al. (2017). "The Evaporation Behavior of Volatile Fission Products in FLiNaK Salt." The Open Access Journal of Science and Technology 5(5): 1-17.

Thoma, R. E. (1971). Chemical aspects of MSRE operations, ORNL Report ORNL-4658, Oak Ridge National Laboratory, Oak Ridge, TN.

Vicente Valdez, P., et al. (2020). Modeling Molten Salt Reactor Fission Product Removal with SCALE ORNL Report ORNL/TM-2019/1418, Oak Ridge National Laboratory, Oak Ridge, TN.

Walker, S. A. and W. Ji (2021). "Species transport analysis of noble metal fission product transport, deposition, and extraction in the molten salt reactor experiment." Annals of Nuclear Energy 158:

108250.

Wright, R. N. and T.-L. Sham (2018). Status of Metallic Structural Materials for Molten Salt Reactors.

INL Report INL/EXT-18-45171, Idaho Falls, ID Yamawaki, M., et al. (2015). Evaluation Study of Source Term for Severe Accident Analysis of Molten Salt Reactors. J. Plasma and Fusion Research 11: 113-119.

Yoshioka, R. (2013). Nuclear energy based on thorium molten salt. Molten Salts Chemistry, Elsevier:

471-496.

Zhang, J. (2014) Electrochemistry of Actinides and Fission Products in Molten Salts -Data Review.

Journal of Nuclear Materials, 447, 271-284.

Zhang, J.; Forsberg, C. W.; Simpson, M. F.; Guo, S.; Lam, S. T.; Scarlat, R. O.; Carotti, F.; Chan, K. J.;

Singh, P. M.; Doniger, W. (2018) Redox potential control in molten salt systems for corrosion mitigation. Corrosion Science 144, 44-53.

Zheng, G. and K. Sridharan (2018) Corrosion of structural alloys in high-temperature molten fluoride salts for applications in molten salt reactors." JOM 70: 1535-1541.