Cobalt-Based Alloy Hardfacing on Reactor Internals: Process Control and Cracking Risk
Setting the Scene
The paper by Li Yanbao and colleagues from Shanghai Jiao Tong University and Shanghai First Machine Tool Works, published in the Journal of Power Engineering in 2016, deals with a topic that sits at the intersection of nuclear safety and welding metallurgy. The authors investigated the manual tungsten inert gas (TIG) welding of Stellite 6 cobalt-based alloy onto 304H austenitic stainless steel, which is a common material combination for reactor internals such as control rod drive mechanisms and guide tubes. The application is demanding because the components must withstand neutron irradiation, high temperature water corrosion, and mechanical wear over a service life of decades. Any cracking in the hardfacing layer is a potential path for corrosion and a source of radioactive contamination.
The authors systematically varied welding parameters and evaluated the resulting deposits for chemical composition, micro hardness, Rockwell hardness, intergranular corrosion resistance and microstructure. Their key finding was that a preheat temperature of 176°C, a welding current of 145 to 190 A, and a welding speed of 8 to 12 cm/min yielded a deposition quality rate above 97%. They also demonstrated that using an ER308L filler wire to deposit a 3 mm thick isolation layer on the 304H base metal, followed by cobalt-based alloy hardfacing with a 400°C preheat and a 400°C interpass temperature, further reduced the risk of cracking in the deposit.
The Critical Role of Dilution and Thermal Cycle
From my years on the shop floor, I know that the biggest challenge with cobalt-based hardfacing on austenitic stainless steel is dilution. When the Stellite 6 wire melts into the base metal, iron from the substrate is mixed into the deposit. The iron content of the deposit increases, which changes the composition of the cobalt-chromium-tungsten alloy and increases its susceptibility to solidification cracking and hot cracking. The authors correctly identify that the electrical parameters and the welding thermal cycle have a significant influence on the deposit composition, hardness and microstructure. Higher heat input increases dilution, which is generally undesirable for this application.
The preheat temperature of 176°C is interesting because it is not a typical preheat for austenitic stainless steel. Normally, austenitic steels are welded without preheat to avoid sensitisation and distortion. However, when you are depositing a cobalt-based alloy with a high coefficient of thermal expansion and a relatively high melting point, the thermal gradient between the deposit and the base metal can cause high residual stresses and cracking. A moderate preheat reduces the thermal gradient and allows the deposit to cool more slowly, which helps the austenitic matrix to accommodate the strain. The authors found that 176°C was the sweet spot for their specific geometry and welding parameters.
The Isolation Layer Strategy
The use of an ER308L isolation layer is a classic technique that I have applied many times in the fabrication of dissimilar metal joints. The purpose of the isolation layer is to provide a buffer between the base metal and the hardfacing alloy. The ER308L filler has a low carbon content, which reduces the risk of carbide precipitation and sensitisation in the heat-affected zone of the 304H base metal. The 3 mm thickness is sufficient to prevent iron from the base metal from diluting into the cobalt-based alloy deposit. The authors then applied a 400°C preheat and a 400°C interpass temperature for the cobalt-based alloy hardfacing. This higher preheat is necessary because the cobalt-based alloy has a higher melting point and a greater tendency to crack if the cooling rate is too fast.
I have seen the consequences of skipping the isolation layer in a similar application. In one case, a contractor deposited Stellite 6 directly onto a stainless steel component without a buffer layer. The deposit cracked along the fusion line during cooling. The cracks were not visible to the naked eye but were detected by dye penetrant testing. The component had to be scrapped because the repair would have required removing the entire deposit and re-welding. The cost of that mistake was significant, both in material and in schedule. The isolation layer is not an optional extra; it is a mandatory step for this material combination.
Practical Parameters and Quality Control
| Parameter | Recommended Range | Effect of Deviation |
|---|---|---|
| Preheat temperature | 176°C | Lower preheat increases cracking risk; higher preheat increases dilution |
| Welding current | 145 to 190 A | Lower current may cause lack of fusion; higher current increases dilution and distortion |
| Welding speed | 8 to 12 cm/min | Slower speed increases heat input and dilution; faster speed may cause porosity |
| Isolation layer material | ER308L | Alternative fillers may cause carbon migration or sensitisation |
| Isolation layer thickness | 3 mm | Thinner layers allow iron dilution; thicker layers increase cost and distortion |
| Interpass temperature | 400°C | Exceeding this may cause over-tempering; lower may cause cracking |
The table above summarises the key process parameters from the paper. The quality control requirements for this type of work are stringent. In my experience, the acceptance criteria for nuclear-grade hardfacing include dye penetrant testing on the entire deposit surface, ultrasonic testing for volumetric defects, and hardness testing at multiple locations. The hardness of the Stellite 6 deposit should be in the range of 38 to 48 HRC, depending on the specific application. The deposit must also pass intergranular corrosion testing, typically using the ASTM A262 Practice E method, to ensure that the heat-affected zone has not been sensitised.
Lessons for the Production Floor
This paper is a good example of how a systematic approach to welding parameter development can solve a difficult problem. The authors did not just try one set of parameters; they varied the parameters and evaluated the results using multiple testing methods. This is exactly what I would recommend to any engineer who is developing a new hardfacing procedure. The 97% deposition quality rate is a strong result, but it also means that 3% of the deposits had defects. In a nuclear application, a 3% defect rate is not acceptable. The isolation layer strategy with a 400°C preheat and interpass temperature was developed specifically to reduce the cracking risk further, and the authors report that this approach was effective.
The key insight for me is that the preheat and interpass temperature must be tailored to the specific material combination and the specific welding process. A preheat that works for one alloy may not work for another. The 176°C preheat for the direct deposition of Stellite 6 onto 304H is a good starting point, but it should be verified for the specific component geometry and welding position. The 400°C preheat for the isolation layer approach is a significant change, and it requires careful control to avoid over-tempering the base metal.
Final Thoughts
The hardfacing of reactor internals is not a job for the faint-hearted. The consequences of a crack in the deposit are severe, and the cost of a failure is measured in safety and schedule, not just money. This paper provides a solid foundation for developing a robust welding procedure, but it is not a substitute for a full procedure qualification test. I would recommend that any engineer who is planning to use this approach should conduct a welding procedure qualification (WPQ) that includes the actual component geometry, the actual welding position, and the actual production welding equipment. The parameters in this paper are a starting point, not a final answer. The 97% quality rate is a reminder that even a well-developed process has a small but real risk of defects, and the isolation layer strategy is the best way to mitigate that risk.
Zhuojin Pipe Fitting Co., Ltd