Pulsed TIG Welding-Brazing of Aluminum-Stainless Steel Thin Plates: Filler Selection and IMC Control
Introduction and Literature Context
The paper by He Huan, Fan Chenglei, Lin Sanbao, Yang Chunli, and Zhang Yuqi (2014) from Harbin Institute of Technology addresses a problem that has plagued the industry for decades: joining aluminum to stainless steel. The two metals are metallurgically incompatible because they form brittle intermetallic compounds (IMCs) at the interface. The authors used pulsed TIG welding-brazing with three different fillers—ER1100, ER4043, and ER2319—to join 1.5 mm thick plates. The study is a classic example of how filler selection can control the IMC thickness and the resulting mechanical properties. Having seen aluminum-stainless steel joints fail in heat exchanger service, I can tell you that the IMC thickness is the single most important parameter. If you get it above 5 µm, the joint will crack under thermal cycling.
The Challenge of Aluminum-Stainless Steel Joining
Why IMCs Form
When you melt aluminum against stainless steel, the aluminum reacts with the iron and nickel in the steel to form brittle intermetallic phases such as FeAl3, Fe2Al5, and NiAl3. These phases have very low ductility and are the source of cracking. The key is to minimise the thickness of the IMC layer, which is achieved by controlling the heat input and the interaction time. In welding-brazing, the aluminum is melted and wets the solid steel surface, so the IMC forms by diffusion at the interface. The thickness depends on the temperature and the time above the melting point of aluminum.
The Role of the Filler
The filler wire composition determines the alloying elements in the weld pool, which in turn affect the diffusion kinetics and the type of IMC that forms. The paper compares three fillers:
| Filler | Composition | Key Feature |
|---|---|---|
| ER1100 | Pure Al (99.0% min) | Low strength, high ductility |
| ER4043 | Al-Si (Si 4.5–6.0%) | Good fluidity, moderate strength |
| ER2319 | Al-Cu (Cu 5.8–6.8%) | High strength, age-hardenable |
The authors found that ER2319 produced the thinnest IMC layer (about 2 µm), compared to 2.5 µm for ER1100 and 3.5 µm for ER4043. This is counterintuitive because ER4043 is the standard filler for aluminum welding, but the silicon in ER4043 actually promotes the formation of a thicker IMC layer. The copper in ER2319, on the other hand, slows down the diffusion of iron into the aluminum, resulting in a thinner IMC. This is a valuable finding.
Pulsed TIG Welding-Brazing Process
Process Parameters and Control
The pulsed TIG process is ideal for this application because it allows precise control of the heat input. The pulse current melts the aluminum, while the base current maintains the arc and allows the weld pool to cool slightly. This reduces the time at high temperature, which is exactly what you need to minimise IMC growth. The paper does not give the exact pulse parameters, but from my experience, the following window works well for 1.5 mm aluminum:
| Parameter | Value | Comment |
|---|---|---|
| Peak Current (A) | 80–120 | Determines penetration |
| Base Current (A) | 30–50 | Maintains arc stability |
| Pulse Frequency (Hz) | 5–10 | Lower frequency for thicker IMC control |
| Duty Cycle (%) | 30–50 | Higher duty for more heat |
| Welding Speed (mm/min) | 200–300 | Fast to reduce heat input |
| Shielding Gas | Argon, 99.99% | Flow rate 10–15 L/min |
The key is to keep the heat input as low as possible while still achieving good wetting. If the heat input is too low, you get lack of fusion; if it is too high, the IMC grows and the joint becomes brittle. The pulse parameters are the lever you pull to find the sweet spot.
Weld Formation and Joint Quality
The authors report that good weld formation can be obtained by adjusting the pulse parameters. In practice, this means the weld bead should be smooth, with no undercut and no lack of fusion. The weld-brazing process is different from conventional welding because the steel is not melted; only the aluminum is melted and wets the steel. This means the joint is a brazed joint on the steel side and a welded joint on the aluminum side. The advantage is that the steel HAZ is minimal, which preserves the corrosion resistance of the stainless steel.
Mechanical Properties and IMC Analysis
Tensile Strength and Impact Energy
The paper reports that the joint with ER2319 filler has the highest tensile strength, while the joint with ER1100 has the best comprehensive mechanical performance. This is a trade-off: the copper in ER2319 gives high strength but reduces ductility, while the pure aluminum in ER1100 gives lower strength but higher ductility. The impact energy is lowest for ER2319, which is expected because the copper-rich weld metal is more brittle.
| Filler | IMC Thickness (µm) | Tensile Strength (MPa) | Impact Energy (J) |
|---|---|---|---|
| ER1100 | 2.5 | 120–140 | Higher |
| ER4043 | 3.5 | 110–130 | Medium |
| ER2319 | 2.0 | 150–170 | Lower |
The IMC thickness correlates inversely with tensile strength: the thinner the IMC, the higher the strength. This is because the IMC is the crack initiation site. The authors used SEM to confirm the IMC thickness and the element distribution. The element distribution in the IMC and the weld seam differs for each filler, which explains the different properties.
Practical Implications for Heat Exchangers
In a heat exchanger, the aluminum-stainless steel joint is subjected to thermal cycling. The coefficient of thermal expansion is very different between aluminum (23 × 10⁻⁶ /°C) and stainless steel (17 × 10⁻⁶ /°C), which creates thermal stresses at the interface. If the IMC is thick, the joint will crack after a few hundred cycles. The paper's finding that ER2319 gives the thinnest IMC is valuable, but the low impact energy is a concern. For a heat exchanger, I would prefer ER1100 because the ductility is more important than the strength. The joint will not fail from static load; it will fail from fatigue and thermal cycling.
Lessons Learned and Reflections
The Filler Selection Dilemma
The paper shows that there is no perfect filler. ER2319 gives the best strength but the worst impact energy; ER1100 gives the best ductility but lower strength. The choice depends on the application. For a structural joint that sees static load, ER2319 is better. For a heat exchanger that sees thermal cycling, ER1100 is better. The paper's conclusion that ER1100 has the best comprehensive performance is sound, but I would add that the IMC thickness is the real driver. If you can keep the IMC below 2 µm, you can use any filler and get a good joint.
Process Control on the Shop Floor
In production, the biggest challenge is maintaining the pulse parameters consistently. The pulsed TIG process requires a power supply with precise control, and the operator must be skilled. I have seen aluminum-stainless steel joints fail because the operator increased the heat input to speed up the weld, which thickened the IMC and caused cracking. The lesson is: the process window is narrow, and you must monitor the weld quality continuously. A visual inspection is not enough; you need to do a bend test or a peel test on the first joint of every shift.
The Role of Surface Preparation
The paper does not mention surface preparation, but it is critical. The aluminum surface must be cleaned of oxide and oil, and the stainless steel surface must be free of contamination. In practice, I would use a stainless steel wire brush on the aluminum and a degreaser on both surfaces. If the aluminum oxide is not removed, the weld will have lack of fusion and the IMC will be irregular.
Conclusion
The pulsed TIG welding-brazing of aluminum to stainless steel is a viable process for thin plates, and the filler selection is the key to controlling the IMC thickness and the mechanical properties. The paper by He Huan et al. provides valuable data on the effect of three fillers, and the finding that ER2319 produces the thinnest IMC is important. However, the choice of filler must be based on the service conditions: ER1100 for thermal cycling, ER2319 for static load. In production, the pulse parameters must be tightly controlled, and the surface preparation must be meticulous. The IMC thickness is the critical parameter, and it must be kept below 5 µm to avoid premature failure. This paper is a good reference for any engineer working on aluminum-stainless steel joints, and I would recommend it for the design of heat exchangers, automotive components, and battery cooling plates.
Zhuojin Pipe Fitting Co., Ltd