Dissimilar Metal Welding of CLAM Steel to 15-15Ti Stainless: Microstructural Evolution and Heat Treatment
Literature Overview and Core Findings
The paper by An Mingming and colleagues from Harbin University of Science and Technology and the Chinese Academy of Sciences investigates the tungsten inert gas welding of China Low Activation Martensitic steel to 15-15Ti stainless steel, and the effect of post-weld heat treatment on the joint microstructure and hardness. The as-welded joint consists of lath martensite plus a small amount of delta ferrite in the weld metal, with hardness ranging from 407 HV to 463 HV. The CLAM steel heat-affected zone is divided into a fully quenched zone and a partially quenched zone, with the latter showing a clear softening phenomenon. The 15-15Ti stainless steel HAZ is the lowest hardness region of the entire joint. After heat treatment, the high-hardness zone shrinks, the weld metal transforms to tempered martensite with increased precipitate content, and hardness drops by approximately 20%.
Technical Interpretation: Metallurgy of the Dissimilar Joint
This is a classic dissimilar metal weld between a martensitic steel and an austenitic stainless steel, and the metallurgy is unforgiving. The CLAM steel is designed for fusion reactor first-wall and blanket structures, with a fully martensitic microstructure after normalising and tempering. When you weld it, the HAZ near the fusion line is heated above the Ac3 temperature, so it transforms to austenite on heating and then to fresh, untempered martensite on cooling. That is the "fully quenched zone" mentioned in the paper, and it is hard and brittle. Further away, the peak temperature is between Ac1 and Ac3, so only partial transformation occurs, and the resulting mixed structure is actually softer than the base metal because the tempering effect of the weld thermal cycle dominates. That is the "partially quenched zone" with the softening phenomenon. On the 15-15Ti side, the HAZ is austenitic, and it does not transform on cooling, but the grain growth and carbide dissolution make it the weakest link in the joint.
Integration with Engineering Practice: PWHT and Joint Design in Nuclear and Power Applications
In my experience with power plant piping, dissimilar metal welds between ferritic and austenitic steels are the most failure-prone joints in the entire system. The problem is not just the metallurgy but the difference in thermal expansion coefficients, which creates cyclic stresses during startup and shutdown. The 20% hardness reduction after heat treatment is a double-edged sword. It relieves the hardness mismatch and improves toughness, but it also reduces the strength of the weld metal, and if you overdo the PWHT you can end up with a joint that is too soft for the design pressure. The paper does not give the exact PWHT parameters, but the fact that the weld metal becomes tempered martensite with increased precipitates suggests a temperature in the range of 700 to 760 degrees Celsius, which is typical for martensitic steels. The increased precipitate content is likely carbides and possibly Laves phase, and their distribution and size will determine the long-term creep resistance of the joint.
Key Questions and Reflections
The softening in the CLAM steel partially quenched zone is a real concern for structural integrity. If the soft zone is wide enough, the joint will deform preferentially there under load, leading to strain concentration and premature failure. I would want to know the width of this soft zone and how it changes with heat input. The paper does not give this information, but it is critical for design. Another question is the carbon migration issue. In a dissimilar joint between a martensitic steel with around 0.1% carbon and an austenitic steel with lower carbon activity, carbon will migrate from the ferritic side to the austenitic side during PWHT, creating a decarburised layer on the CLAM side and a carburised layer on the 15-15Ti side. This is a well-known problem in power plant welds, and it can lead to a "type IV" failure in the soft zone. The paper does not mention carbon migration, but it is likely present given the composition difference.
Study Insights and Implications
For our industry, this work reinforces the importance of controlling the heat input and PWHT schedule in dissimilar metal welds. The 20% hardness reduction after heat treatment is significant, but it is not necessarily bad if the joint is designed for it. The key is to know the minimum acceptable hardness for the service conditions and to ensure the PWHT does not push the joint below that threshold. I would also note that the 15-15Ti HAZ being the lowest hardness region is counterintuitive, because austenitic stainless steels usually work-harden rather than soften. The explanation is likely grain growth and the dissolution of fine precipitates that were present in the base metal. This is a reminder that even "soft" austenitic steels can have their own HAZ problems, and you cannot assume the austenitic side is always the safe side.
This study gives us a clear picture of the microstructural evolution in a CLAM to 15-15Ti dissimilar joint and highlights the critical role of post-weld heat treatment in managing the hardness profile. The practical takeaway is that PWHT is not just a stress-relief operation but a metallurgical transformation that must be carefully controlled, and the 20% hardness reduction is the price you pay for a tougher, more serviceable joint.
Zhuojin Pipe Fitting Co., Ltd