Effect of Metal Transfer Mode on Local Chemical Inhomogeneity in Dual-Wire Electron Beam Additive Manufacturing of Nitinol

In the high-stakes worlds of aerospace and energy, engineers rely on a remarkable material called Nitinol. It is a metal alloy of nickel and titanium that possesses a unique ability to remember its original shape. If bent or twisted, it can snap back to its starting form when heated, a property known as the shape memory effect. This behavior is not just a curiosity; it is a functional necessity for devices ranging from medical stents to aerospace actuators. However, this material is notoriously sensitive to its exact chemical makeup. Even a tiny shift in the ratio of nickel to titanium—less than one-tenth of a percent—can drastically alter the temperatures at which the metal changes its phase, rendering it useless for its intended purpose. Creating Nitinol that is both chemically perfect and shaped into complex, near-final forms has long been a manufacturing challenge, as traditional methods are expensive and struggle with intricate geometries.

To solve this, researchers have turned to a technique called additive manufacturing, essentially 3D printing with metal. In a specific variation known as dual-wire electron beam additive manufacturing, two separate wires—one of nickel and one of titanium—are fed into a vacuum chamber where a powerful beam of electrons melts them together, layer by layer, to build a part. The goal is to mix the two metals perfectly as they melt, creating a uniform alloy from the bottom of the part to the top. Yet, in practice, this process often fails to mix the metals evenly. Small pockets of the material end up with too much nickel or too much titanium. These chemical imbalances create brittle zones that weaken the part and destroy its shape-memory capabilities. The question facing the scientific community was not just that these flaws existed, but why they happened and how to stop them.

A team of researchers from the Moscow Power Engineering Institute and the Perm National Research Polytechnic University set out to uncover the root cause of these chemical flaws. They focused on a single, adjustable variable: the angle at which the two wires were fed into the molten pool. In their experiments, they tested three different setups. In the first, the wires were angled sharply toward each other, converging at thirty degrees. In the second, they were angled more gently at fifteen degrees. In the third, the wires were fed parallel to each other, running side by side with no convergence angle. By changing this angle, they could observe how the liquid metals behaved before they solidified into a part.

To see what was happening inside the invisible, super-hot melt pool, the researchers used a combination of high-speed cameras and advanced computer simulations. The cameras, capable of capturing thousands of frames per second, allowed them to watch the wires as they melted. The computer models, which tracked the movement of millions of tiny particles of liquid metal, helped them understand the forces at play, such as surface tension and heat flow. They discovered that the angle of the wires dictated exactly how the metals entered the pool. When the wires were angled at thirty degrees, the nickel wire behaved erratically. Because nickel and titanium have different surface tensions, the nickel tended to form large droplets that would detach and fall into the pool separately from the titanium. This asynchronous delivery meant that for brief moments, the pool was flooded with just one metal, creating pockets of chemical imbalance before the other metal could catch up.

When the wires were angled at fifteen degrees, the situation improved slightly, but the metals still entered the pool through two separate paths, relying entirely on the swirling motion of the electron beam to mix them. However, when the researchers fed the wires parallel to each other, the behavior changed fundamentally. The two wires touched and melted together before they even fully entered the main pool. Instead of two separate streams of metal, they formed a single, stable bridge of liquid that flowed into the pool already partially mixed. This preliminary mixing meant that the nickel and titanium arrived at the pool as a unified stream rather than as competing, separate droplets.

The results of this simple adjustment were dramatic. In the samples made with the thirty-degree angle, the chemical composition varied wildly, with deviations from the ideal mix reaching over ten percent in some spots. The metallographic analysis of these samples revealed a chaotic structure filled with brittle, unwanted phases that would make the part prone to cracking. In contrast, the samples made with the parallel wire feeding showed a remarkably uniform chemical composition. The deviation from the ideal mix dropped to less than half a percent. The internal structure of these samples was consistent, showing a homogeneous matrix of the desired alloy with a hardness that was evenly distributed across the width of the part.

The researchers also measured how long the metal stayed liquid and how fast it moved within the pool. They found that the parallel feeding method kept the metal in a liquid state for a significantly longer time—about 1.04 seconds compared to less than half a second for the angled method. This extra time, combined with a faster flow speed, gave the metals a much larger window to mix thoroughly before they cooled and solidified. In the angled setups, the window for perfect mixing was so narrow that the metals often froze before they could blend evenly. The parallel setup provided a generous margin, ensuring that the chemical composition was homogenized long before the part hardened.

This study demonstrates that the geometry of the wire feed is not just a minor detail but a controlling factor in the quality of the final product. By simply aligning the wires to run parallel, the researchers eliminated the chaotic droplet formation that caused chemical segregation. They proved that forcing the metals to mix before they hit the main pool is the key to creating a flawless alloy. The findings offer a clear, practical path forward for manufacturers who need to produce high-performance Nitinol parts with the precision required for critical applications. The solution did not require new materials or more complex machinery, but rather a deeper understanding of how liquid metals behave when they meet, and a simple adjustment to how they are delivered.