Cemented Carbide YG30 and 45 Steel TIG Welding Connection Study
Literature Overview and Core Problem
Joining cemented carbide to steel is one of the classic difficult problems in welding. The carbide is a ceramic-metal composite, typically tungsten carbide (WC) particles in a cobalt matrix, while the steel is a ductile metal. The two materials have vastly different thermal expansion coefficients, thermal conductivities and melting points. When you try to weld them together, you get a joint that is brittle, prone to cracking and often contains intermetallic phases that degrade the interface.
The paper by Zhao Xiujuan and Yang Dexin from Dalian Railway Institute addresses this problem using TIG welding with two different filler wires: a Ni-Fe alloy and pure Ni. The base materials are YG30 cemented carbide (30% cobalt, 70% tungsten carbide by weight) and 45 steel (medium carbon steel, approximately 0.45% carbon). The goal is to find a filler and a welding procedure that produces a sound joint without the formation of η phase, which is a brittle intermetallic compound that forms when carbon diffuses into the carbide and reacts with the tungsten and cobalt.
Technical Points and Experimental Findings
The authors conducted four types of welding trials, as summarised below.
| Trial | Filler Wire | Number of Layers | Welding Method | η Phase Formation |
|---|---|---|---|---|
| 1 | Ni-Fe alloy | 4 | Butt weld | Yes |
| 2 | Pure Ni | 4 | Butt weld | No |
| 3 | Pure Ni | Multiple | Surfacing / build-up | No |
| 4 | Ni-Fe alloy | Multiple | Surfacing / build-up | No |
The critical finding is that η phase forms in the heat-affected zone of the carbide when using Ni-Fe alloy filler for a 4-layer butt weld, but does not form when using pure Ni filler, nor when using Ni-Fe alloy for surfacing. The authors used scanning electron microscopy (SEM) combined with energy-dispersive X-ray spectroscopy (EDS) to analyse the composition of the η phase regions.
The mechanism is clearly explained. During welding, the high temperature causes carbon from the steel to diffuse into the weld metal and then into the carbide. At the same time, iron and nickel diffuse from the weld metal into the carbide. When the carbon concentration in the carbide reaches a critical level, and the iron and nickel concentrations are also elevated, the η phase (which has a formula approximating to M₆C or M₁₂C, where M is a combination of W, Fe, Ni and Co) precipitates. This phase is extremely hard and brittle.
The hardness measurements confirm the problem. In the joints where η phase forms, the hardness profile across the carbide-weld interface shows a sharp step, with a very hard η phase layer sandwiched between the carbide and the weld metal. In the joints without η phase, the hardness transition is gradual, indicating a more compliant interface that can accommodate the thermal stresses.
Practical Implications for the Shop Floor
This paper has direct relevance to anyone who has to join carbide tooling or wear-resistant components to steel shafts or bodies. In my experience, this type of joint is common in mining equipment, cutting tools and wear plates. The failure mode is almost always at the interface, and it is usually a brittle fracture that initiates at the η phase layer.
The key practical lessons are:
- Filler metal selection is critical. Pure Ni filler is clearly superior to Ni-Fe alloy for this application, because it does not promote η phase formation. The Ni-Fe alloy introduces iron into the weld metal, which then diffuses into the carbide and promotes the η phase reaction.
- The number of layers matters. A 4-layer butt weld with Ni-Fe alloy creates enough heat input and enough diffusion time to form η phase. A surfacing operation with the same filler does not, because the heat input is lower and the cooling rate is faster.
- The η phase is not just a cosmetic issue. It causes a sharp hardness step at the interface, which creates a stress concentration and a preferred path for crack propagation. This is the root cause of premature failure in many carbide-steel joints.
Key Questions and Reflections
One question that the paper does not fully address is the effect of the cobalt binder in the carbide. YG30 has 30% cobalt, which is a relatively high binder content. The cobalt provides some ductility to the carbide, but it also reacts with carbon to form cobalt carbides. The η phase formation is likely influenced by the cobalt content, and a lower cobalt carbide (e.g., YG6 or YG8) may behave differently.
Another question is the effect of preheating and post-weld heat treatment. The paper does not mention any preheat, but in practice, carbide-steel joints are often preheated to reduce thermal stress. The authors may have found that preheating worsens the η phase problem, because it increases the diffusion time. This is a critical consideration for a production engineer.
The paper also does not discuss the mechanical properties of the joints beyond hardness. There is no tensile test, no bend test and no impact test. For a joint that is likely to be loaded in service, this is a significant gap. The hardness data tells us about the interface, but it does not tell us about the overall joint strength or toughness.
Study Insights and Practical Recommendations
This paper is a good example of how a systematic study of the interface can reveal the root cause of a welding problem. The authors did not just try different fillers and hope for the best; they used SEM and EDS to understand the diffusion mechanism, and they correlated the microstructural observations with the hardness profile.
For a production engineer, I would offer the following advice:
- Use pure Ni filler wire for carbide-steel TIG welding. The extra cost of the Ni is justified by the elimination of η phase and the resulting improvement in joint reliability.
- Control the heat input carefully. Lower heat input reduces diffusion and suppresses η phase formation. Use the minimum current and travel speed that still gives good fusion.
- Consider the joint design. A butt joint with a thick carbide section will have high thermal stress. A scarf joint or a step joint can reduce the stress concentration.
- Inspect the interface carefully. Use ultrasonic testing or radiography to detect cracks, and use hardness testing to verify the absence of a sharp hardness step.
- If the application involves cyclic loading, consider a post-weld stress relief heat treatment, but verify that it does not promote η phase formation.
Concluding Summary
This study demonstrates that the formation of η phase in cemented carbide YG30 to 45 steel TIG welded joints is strongly influenced by the filler metal composition and the welding procedure. Pure Ni filler wire suppresses η phase formation, while Ni-Fe alloy promotes it in multi-layer butt welds. The η phase forms due to carbon diffusion from the steel into the carbide, combined with iron and nickel diffusion in the opposite direction, and it creates a sharp hardness step at the interface that is a potential crack initiation site. The practical implication is clear: for carbide-steel joints that must withstand service loading, pure Ni filler is the preferred choice, and the heat input must be controlled to minimise diffusion. This paper provides a solid scientific basis for a welding procedure that has historically been more art than science.
Zhuojin Pipe Fitting Co., Ltd