Zhuojin Pipe Fitting Co., LtdPeeling Mechanism of Stainless Steel Belt Electrode Submerged Arc Surfacing on Nuclear Island Main Equipment
Literature Overview
The paper by Yang Chengdong and colleagues from Shanghai Electric Nuclear Power Group, published in Pressure Vessel (2023, Vol. 40, Issue 6, pp. 22–26), presents a detailed failure analysis of stainless steel belt electrode submerged arc surfacing (SAW) deposits on low-alloy steel substrates used in nuclear island main equipment. The study identifies the root causes of surfacing layer peeling, distinguishing between internal metallurgical factors and external mechanical factors. This work is of particular significance given the critical safety requirements for nuclear-grade pressure vessels and piping, where surfacing integrity is essential for long-term reliable operation.
Failure Analysis Results
Internal Cause: Hardened Layer and Type II Boundary
The investigation revealed that a hardened layer with an average thickness of approximately 25 μm forms in the transition zone between the low-alloy steel substrate and the stainless steel surfacing deposit. The microstructure of this hardened layer consists of lenticular (plate-like) martensite, characterized by high hardness and low plasticity. This martensitic layer forms due to the rapid cooling rate at the fusion boundary, where the dilution of alloying elements from the stainless steel deposit into the substrate creates a locally high-carbon, low-alloy environment conducive to martensite formation.
Additionally, a Type II boundary was identified in the transition zone. In the context of surfacing metallurgy, Type II boundaries represent interfaces with distinct compositional and structural discontinuities that act as preferential crack initiation sites. Porosity was also observed within the hardened layer, further weakening the transition zone.
External Cause: Residual Stress Accumulation
The study quantified the relationship between the number of surfacing layers and the residual stress at the joint interface:
| Number of Surfacing Layers | Interface Residual Stress (MPa) | Condition |
|---|---|---|
| 1 | Low | No peeling |
| 2 | Moderate | No peeling |
| 3 | Moderate-high | No peeling |
| 4 | High | No peeling |
| 5 | Very high | Approaching critical |
| 6 | 380 MPa | Peeling observed |
As the number of surfacing layers increased, the cumulative thermal cycles led to progressively higher residual stresses at the interface. When the residual stress reached approximately 380 MPa at 6 layers, the Type II boundary in the transition zone, already weakened by the hardened martensitic layer and porosity, acted as a crack source and initiated the peeling failure.
Metallurgical Mechanism Analysis
The formation of the hardened martensitic layer is a direct consequence of the dilution effect at the fusion boundary. In belt electrode submerged arc surfacing, the high deposition rate and large heat input can lead to significant substrate dilution, particularly in the first few layers. The dilution introduces carbon and low-alloy elements into the stainless steel melt, reducing the austenite stability and promoting martensite formation upon cooling. The lenticular martensite morphology is characteristic of high-carbon, low-nickel compositions, which have limited austenite retention capacity.
The Type II boundary represents a metallurgical interface with a sharp compositional gradient between the martensitic hardened layer and the austenitic stainless steel deposit. This interface is inherently weak due to the mismatch in thermal expansion coefficients between the two phases and the presence of microvoids and porosity at the boundary.
Countermeasures and Process Recommendations
Based on the failure mechanism analysis, the following countermeasures can be recommended:
| Countermeasure | Mechanism | Implementation |
|---|---|---|
| Increase nickel content in surfacing alloy | Stabilize austenite, suppress martensite formation | Use Ni-rich filler such as 309L or 310L with elevated Ni |
| Apply interlayer between substrate and surfacing | Create a diffusion buffer zone | Deposit a transition layer of 309L before main surfacing |
| Control heat input per layer | Reduce dilution, limit martensite formation | Optimize travel speed and current density |
| Implement interpass temperature control | Reduce residual stress accumulation | Maintain interpass temperature at 200–250°C |
| Post-weld stress relief | Reduce residual stress below critical threshold | Perform PWHT at 620–650°C for austenitic stainless steel |
| Limit number of surfacing layers | Prevent residual stress accumulation | Plan surfacing in multiple passes with intermediate stress relief |
Engineering Practice Implications
This failure analysis is directly relevant to the surfacing of nuclear-grade pressure vessels, reactor internals, and steam generator tubes, where stainless steel overlay is commonly applied to provide corrosion resistance in aggressive environments. The identification of the hardened martensitic layer and Type II boundary as the internal failure initiators, combined with residual stress as the external trigger, provides a clear framework for process optimization.
In practical terms, the finding that peeling occurs at approximately 6 layers with 380 MPa residual stress provides a quantitative threshold for process control. Engineers can use this information to design surfacing procedures that limit the number of layers or incorporate intermediate stress relief treatments to maintain residual stress below the critical level.
Study Insights and Reflections
The systematic approach of this failure analysis—identifying both the internal metallurgical factors (hardened layer, Type II boundary, porosity) and the external mechanical factor (residual stress accumulation)—provides a comprehensive understanding of the peeling mechanism. This dual-cause framework is consistent with classical fracture mechanics principles, where a crack requires both a pre-existing defect (internal cause) and sufficient driving force (external cause) to propagate.
The quantification of residual stress as a function of surfacing layer count is particularly valuable for engineering practice, as it provides a clear process control parameter. The 380 MPa threshold at 6 layers can serve as a design criterion for surfacing procedures in nuclear applications, where the consequences of surfacing failure are unacceptable.
The presence of porosity within the hardened layer is also a significant finding, as it suggests that the high cooling rate at the fusion boundary may promote gas entrapment or lack of fusion. This highlights the importance of process parameter optimization, particularly regarding travel speed and current density, to ensure proper fusion and minimize porosity in the transition zone.
Overall, this research provides actionable insights for improving the reliability of stainless steel surfacing on low-alloy steel substrates in nuclear applications, with clear recommendations for alloy selection, process parameter control, and post-weld treatment.