Welding stir-up

Extended operation of the existing fleet of light water reactors (LWRs) may require the repair or replacement of reactor pressure vessel components as in-service degradation occurs due to fatigue, irradiation- assisted stress corrosion cracking (SCC), or other mechanisms. However, the austenitic stainless steels, nickel-based alloys, and weld metals currently deployed in vessel internals and attachments become more difficult to weld over time due to transmutation reactions that occur in high-fluence regions of boiling water and pressurized water reactors (BWRs and PWRs, respectively).

These reactions lead to the slow accumulation of helium (He) atoms within the alloy matrix. Eventually, He bubbles coalesce at grain boundaries and grow in size. Though not a concern during normal operations, this accumulation creates the potential for He- induced cracking during repair operations using conventional arc welding processes. High heat input and tensile stresses associated with the welding process allow He bubble growth in the heat-affected zone (HAZ) of the weld. This acts on the service-exposed materials to pull the grain boundary apart, resulting in rupture and/ or cracking.

Due to He-induced cracking, current welding technology is only applicable below about 0.1 atom-part-per-million (appm) He, a level reached in many reactor pressure vessel locations after about 30 years of operation. This poses a potential asset management challenge for nuclear power producers, as weld repair can be a cost-effective alternative to the replacement of safety-critical components. In some cases, future abilities to repair highly-irradiated internals could represent the key to ensuring structural integrity and allowing continued reactor operations.

The Electric Power Research Institute is pursuing power industry adoption of a novel joining process known as friction-stir welding (FSW) for high-value reactor internals applications, with the promise of enabling underwater repairs. Advancing FSW represents one element in a comprehensive Welding of Irradiated Materials for Reactor Internals roadmap developed in 2011 with the U.S. Department of Energy which specifies the collaborative research and development (R&D) required to create, demonstrate, qualify, and implement the repair innovations that may be needed to support extended operations of existing plants.

Key collaborators include experts from academia, welding technology suppliers, and nuclear power producers, which have provided ex-service materials for laboratory characterization studies. In addition, the High Flux Isotope Reactor at Oak Ridge National Laboratory (ORNL) is being employed to generate a highly-irradiated sample set for weldability trials, and the U.S. Department of Energy is supporting design and construction of a first-of-a-kind Hot Cell Test Facility at ORNL to allow validation of FSW and other welding technologies.

FSW fundamentals

Invented in 1991 by The Welding Institute (TWI) in the United Kingdom, FSW was initially designed as an automated, solid-state process for the joining of aluminium and other low-melting-point materials and alloys. No physical melting occurs, and no weld arc is required. Weld filler metal can be employed but is not typical to the process. Instead, friction is used to coalesce and join materials by plastic deformation and subsequent dynamic recrystallization.

The FSW process begins by loading a special tool against the material to be joined (or repaired), with the tool's shoulder remaining atop the surface and its pin forced into the joint line as shown in Figure 1. The tool is rotated around its profile and travels along the joint line, creating frictional heating that softens underlying material. Metal flows around the pin's profile and recombines after significant plastic deformation. The high level of deformation causes the temperature for dynamic recrystallization to be significantly lower than for conventional welding technologies. When recrystallization occurs, the original interface of the joint is eliminated, and the material coalesces without melting. The lower temperature and strain compared to conventional welding is less conducive to He bubble growth.

After FSW, the joined region includes a thermomechanically-affected zone (TAZ) in the middle, characterized by nearly concentric rings that are created by the rotation of the pin. This zone's equiaxed, recrystallized, fine-grained microstructure, which arises from the high degree of plastic deformation, offers excellent mechanical properties. The outer reaches of the TAZ, subjected to less metal flow and plastic deformation, have relatively coarse grains. The surrounding HAZ is exposed only to thermal effects, resulting in some modification and coarsening of the grain structure relative to the base material.

Microstructural and mechanical properties across the joined region created via FSW are influenced by material flow. Key parameters include tool geometry and pin profile, the speed of rotation, and the speed of the welding pass. Tools used for FSW of aluminium are constructed from tool steel and tungsten carbide, which lack the strength and toughness necessary for joining higher-melting-point materials. Tools formed from super-abrasives such as polycrystalline diamond and polycrystalline cubic boron nitride (PCBN) are necessary for FSW of the austenitic stainless steels and nickel-based alloys employed for reactor pressure vessel applications.

Progress to date

In 2012, EPRI launched an initial feasibility study of FSW for nuclear plant repair applications. Test plates were fabricated from materials commonly used in pressure vessel internals, including 304 stainless steel (SS), 308L SS weld metal, Alloy 600, and Alloy 182 weld metal. Cracks were simulated using notches created via electric discharge machining (EDM). Per the matrix shown in Table 1, FSW tests were performed both in air and underwater (Figure 2) using a single PCBN tool having a convex shoulder with a scrolled profile and a truncated cone pin with helical threads. Trials were conducted while monitoring tool force, tool speed, and travel speed. Welded samples (Figure 3) were then subjected to detailed metallurgical analysis and nondestructive evaluation.

Initial proof-of-concept tests of single-pass FSW on 304 SS demonstrated that welds produced in ambient air and underwater exhibit the same metallurgical characteristics. Follow-on tests were conducted on notched 304 SS plates and on 304 SS plates with notched 308L SS inlays. Two overlapping passes of the FSW tool were applied to implement crack-sealing repairs, with the second occurring over and through but slightly offset from the TAZ created by the first.

These experiments were designed to test the hypothesis that differences in flow stress associated with large differences in grain morphology parallel to the FSW joint line could create weldability problems: the TAZ created during an initial pass has grain size an order of magnitude smaller than the HAZ and base metal, while 304 and 308L SS have equiaxed grains and columnar dendrites, respectively. In all tests, simulated cracks were successfully sealed through both in-air and underwater FSW, and no flaws were detected in dual-pass regions or other weld zones.

Single-pass welds were applied to a 308L test plate to evaluate underwater FSW specifically for SCC repair in welds commonly found in reactors. Cracks oriented at three different angles relative to the weld travel direction were successfully sealed. During metallurgical characterization, subsurface remnants of notches were detected at the beginning of the weld pass. None were detected further along the weld, likely due to an increase in the level of metal stirring over time and distance. FSW process optimization is expected to address this issue.

To assess whether the technology could be used for patching damaged surfaces, thin (10 gauge) 304L SS plates were tack-welded to notched, 1-inch-thick 304 SS test specimens. FSW was then successfully applied in air and underwater conditions, establishing its potential for direct sealing of cracks using base metal patches, as well as for crack mitigation before the application of a protective weld overlay. Patching via FSW might also provide a means for localized shielding (that is, a barrier layer) to protect irradiated materials during component repair or replacement activities conducted using conventional high-heat-input welding.

The final test plate design simulated a dissimilar metal weld (DMW) incorporating all base metal and filler metal combinations found in BWR internals. Successful in-air and underwater repair welds were produced on notches across the interfaces between both carbon steel-308 L SS and Alloy 182-Alloy 600 joints. Some weldability limitations were experienced across the 308 L SS-Alloy 82 joint due to significant differences in the temperatures at which plastic flow initiates on opposite sides of the profile pin as it travels through the DMW region. Relative to the full population of welds within BWRs, the number of locations containing this material combination is relatively limited, but FSW process optimization could help mitigate this issue.

Next steps

Based on findings from initial laboratory trials and characterization studies, FSW offers near-term potential for sealing and patching crack-like defects both within and external to the reactor environment. Its suitability for underwater welding is a major advantage over currently available repair technologies for reactor internals, as avoiding the need for water evacuation provides time and cost savings. Additional research by EPRI and FSW developers is necessary to qualify the technology for near-term PWR and BWR applications, including the repair of spent fuel canisters, with respect to weld quality, corrosion resistance, process control, and other factors.

FSW also shows tremendous long-term promise for extending the nuclear power industry's repair and refurbishment capabilities to address highly-irradiated vessel internals and attachments that cannot be repaired using existing welding technologies. Consistent with EPRI's R&D roadmap, custom melts of 304 SS, 316 SS, and Alloy 182 materials are undergoing exposure at ORNL's High Flux Isotope Reactor to generate elevated He levels representative of 40 to 70 years of commercial reactor operation at internals locations.

In 2015, validation testing of a purpose-built FSW prototype on irradiated samples with 10, 20, and 30 appm He is scheduled to begin in ORNL's new Hot Cell Test Facility. This experimental work, combined with parallel materials characterization and modelling studies focused on in-service degradation and He accumulation, will help better define the applicability of FSW technology for repairing reactor internals, as well as support methodological refinements. Assuming favourable results, the next steps toward commercialization will include development and testing of tooling, low-force process parameters, and equipment delivery systems.

References

EPRI, 2014. A Review of Advanced Welding Technology and Applications for Nuclear Power Applications (3002002574).
EPRI, 2012 Welding and Repair Technology Center: Assessment of Friction Stir Welding for Nuclear Applications (1025175).