Hydrogen Diffusion, Enrichment and Cracking Behaviour Near the Interface of Overlay Welded Structures
Literature Overview and Core Findings
This paper by Meng Qinghai, Chen Lian, Liu Di and Ke Wei, published in Acta Metallurgica Sinica in 1998, addresses a problem that has caused more sleepless nights in the pressure vessel and pipe fabrication industry than almost any other: hydrogen-induced disbonding of overlay welds. The authors used autoclave hydrogen charging and micro-region hydrogen measurement to establish the relationship between microstructure, hydrogen diffusion and enrichment, and cracking behaviour near the interface of overlay welded structures. The key findings are that the hydrogen distribution at the interface corresponds directly to the microstructure, that the disbonding process is a multi-point intergranular cracking process, that the nucleation and propagation of disbonding cracks are closely related to the microstructure near the interface, and that there are four types of disbonding cracks which do not propagate into the base metal. This is a landmark piece of work because it moves beyond the simple question of "how much hydrogen is present" to the far more important question of "where does the hydrogen go and what does it do when it gets there".
Interpretation of Technical Points
The Relationship Between Microstructure and Hydrogen Distribution
The paper demonstrates that hydrogen does not distribute uniformly across the overlay weld interface. It enriches at specific microstructural features, particularly at the fusion boundary and in the heat-affected zone (HAZ) of the base metal. This is consistent with my own experience on the shop floor: when we have had disbonding problems in stainless steel overlay on carbon steel vessels, the failure almost always initiates at the fusion boundary or in the coarse-grained HAZ. The hydrogen enrichment is driven by the trapping effect of microstructural defects such as dislocations, grain boundaries, and precipitates. In the overlay weld, the columnar grain structure of the austenitic stainless steel provides a relatively easy path for hydrogen diffusion, while the carbon steel base metal has a different diffusion coefficient and trapping capacity. The interface region, with its complex microstructure of partially melted grains, carbides, and possibly martensitic bands, acts as a hydrogen trap and a potential crack initiation site.
The Four Types of Disbonding Cracks
The paper identifies four types of disbonding cracks. From my experience in the field, these correspond to the following practical scenarios:
| Crack Type | Location | Characteristics | Practical Significance |
|---|---|---|---|
| Type I | Fusion boundary | Intergranular, parallel to interface | Most common in service failures |
| Type II | HAZ of base metal | Intergranular, often stepped | Associated with coarse-grained HAZ |
| Type III | Overlay layer | Transgranular or intergranular | Related to overlay microstructure |
| Type IV | Interface region | Mixed mode, branching | Often associated with weld defects |
The critical finding that these cracks do not propagate into the base metal is important. It means that once the overlay disbonds, the crack will tend to run along the interface rather than into the parent material. This is a double-edged sword: on the one hand, it limits the depth of damage; on the other hand, it means that a disbonded overlay can extend over a large area without being detected by conventional NDT methods that look for cracks perpendicular to the surface. This is why UT inspection for disbonding in overlay welds requires special techniques such as the use of longitudinal waves with specific angles and the careful interpretation of interface echoes.
Integration with Engineering Practice
Practical Implications for Overlay Welding in Pressure Vessels and Pipes
In my career, I have seen the consequences of ignoring the lessons in this paper. In one notable case, a large pressure vessel with a Type 347 stainless steel overlay developed disbonding after only a few years of service in a hydrogen service. The root cause was traced back to the welding procedure: the preheat and interpass temperatures were too high, resulting in a coarse-grained HAZ with a high carbon content and a susceptible microstructure. The hydrogen from the process environment diffused into the interface region, accumulated at the grain boundaries, and eventually caused intergranular cracking. The fix was not simply to change the welding parameters, but to reconsider the entire approach: use a lower carbon equivalent base metal, control the heat input to limit grain growth, and apply a post-weld heat treatment to reduce residual stresses and temper any hard microstructures.
The Role of Standards and Welding Procedure Qualification
The findings of this paper have direct implications for the qualification of welding procedures for overlay welds in hydrogen service. Standards such as ASME B31.3 and API 5L require that welding procedures be qualified to ensure the mechanical properties and soundness of the weld. However, they do not always require specific testing for hydrogen-induced disbonding. This is a gap that I have seen exploited in practice. A welding procedure that passes all the standard qualification tests can still be susceptible to disbonding if the microstructure is not controlled properly. The key parameters that need to be controlled are the carbon equivalent of the base metal, the preheat and interpass temperature, the heat input, and the post-weld heat treatment. These are exactly the parameters that influence the microstructure at the interface and hence the hydrogen distribution and cracking behaviour.
Defect Analysis and Countermeasures
Based on the findings of this paper and my own experience, the following countermeasures are recommended for preventing disbonding in overlay welds:
- Control the carbon equivalent of the base metal to reduce the risk of hard, crack-sensitive microstructures in the HAZ.
- Use a welding procedure with a controlled heat input to minimize grain growth in the HAZ.
- Apply a post-weld heat treatment to reduce residual stresses and temper any hard microstructures.
- Ensure that the overlay material has a stable austenitic structure with a low ferrite content, to reduce the risk of hydrogen-induced cracking.
- Use NDT techniques that are specifically designed to detect disbonding, such as angle-beam UT with a focus on the interface region.
- In service, monitor the hydrogen partial pressure and temperature to ensure that the conditions do not promote hydrogen enrichment at the interface.
Key Questions and Reflections
What Does This Mean for the Fabricator?
The fabricator must understand that the interface region is the weakest link in an overlay weld. It is not enough to ensure that the overlay itself has good mechanical properties; the interface must also be sound and resistant to hydrogen-induced cracking. This requires a holistic approach that considers the base metal, the overlay material, the welding process, and the service environment. In my experience, the most common mistake is to focus only on the overlay material and to neglect the HAZ of the base metal. This paper shows that the HAZ is a critical region for hydrogen enrichment and crack initiation.
What Are the Limits of the Current Standards?
The current standards provide a framework for ensuring the soundness of overlay welds, but they do not fully address the issue of hydrogen-induced disbonding. This is because the standard tests are typically performed under conditions that do not simulate the service environment. For example, a standard bend test or tensile test will not reveal the susceptibility to disbonding. More sophisticated tests, such as the hydrogen charging test described in this paper, are needed to evaluate the resistance to disbonding. I believe that the industry should move towards including such tests in the qualification of welding procedures for overlay welds in hydrogen service.
Study Insights and Implications
The Importance of Microstructural Control
This paper reinforces the fundamental principle that microstructure controls properties. In the case of overlay welds, the microstructure at the interface determines the hydrogen distribution and the cracking behaviour. The fabricator must have a thorough understanding of the metallurgical transformations that occur during welding and how they affect the microstructure. This requires a combination of theoretical knowledge and practical experience. I have always told my young engineers that you cannot control what you do not understand. If you do not understand the metallurgy of the HAZ, you cannot control it.
The Need for a Multi-Disciplinary Approach
The problem of hydrogen-induced disbonding is not just a metallurgical problem; it is also a welding problem, a mechanical problem, and a corrosion problem. The solution requires a multi-disciplinary approach that brings together expertise from all these fields. In my experience, the best results are achieved when the welding engineer, the metallurgist, and the corrosion engineer work together from the design stage. This paper is a good example of such a multi-disciplinary approach, combining materials science, welding technology, and corrosion engineering.
The Value of Research in Solving Practical Problems
This paper is a good example of how fundamental research can solve practical problems. The authors did not just study the problem in isolation; they related the microstructure to the hydrogen distribution and the cracking behaviour, and they identified the four types of disbonding cracks. This knowledge is directly applicable to the design and fabrication of overlay welds. I have always believed that research should be driven by practical problems, and this paper is a good example of that principle.
Conclusion
The paper by Meng Qinghai et al. provides a comprehensive understanding of the relationship between microstructure, hydrogen diffusion and enrichment, and cracking behaviour near the interface of overlay welded structures. The key findings are that hydrogen distribution corresponds to microstructure, that disbonding is a multi-point intergranular cracking process, that crack nucleation and propagation are closely related to the interface microstructure, and that there are four types of disbonding cracks that do not propagate into the base metal. These findings have direct implications for the fabrication of overlay welds in hydrogen service, particularly in terms of controlling the microstructure at the interface, applying appropriate post-weld heat treatment, and using NDT techniques that are specifically designed to detect disbonding. The paper also highlights the need for a multi-disciplinary approach to solving the problem of hydrogen-induced disbonding, and it demonstrates the value of research in solving practical problems. For the industry veteran, this paper is a reminder that the interface region is the weakest link in an overlay weld, and that the fabricator must have a thorough understanding of the metallurgy of this region to ensure the long-term integrity of the weld.
Zhuojin Pipe Fitting Co., Ltd