SCIENCE

Texas A&M researchers exploit shape memory transformations to design smart materials for industry

Imagine a metal that “remembers” its original, cold-forged shape, and can return to that shape when exposed to heat or a magnetic pulse. Like magic out of a Harry Potter novel, such a metal could contract on command, or swing back and forth like a pendulum.

Believe it or not, such metals already exist. First discovered in 1931, they belong to a class of materials called “shape memory alloys (SMA),” whose unique atomic make-up allows them to return to their initial form, or alternate between forms through a phase change.

SMAs are mainly used in orthopedics, biomedicine, and dentistry. In these applications, cold forms of the alloys are implanted in the body; as they try to return to their original, room temperature form, they exert pressure, keeping broken bones or crooked teeth in place.

Many believe these smart metals will have their greatest impact in the aerospace and automotive industries, where SMAs are expected to significantly improve the performance of motors, wings, and other mechanical parts.

But first, certain obstacles must be overcome. Most shape memory alloys can only “cycle” (or change form) a limited number of times before the material gives out. Moreover, they primarily work at low temperatures, making them bad candidates for functions in high-heat environments like aircraft engines.

“These shape memory materials have many applications,” said Raymundo Arroyave, assistant professor of mechanical engineering and materials science at Texas A&M. “Despite being heavily studied for the past twenty to thirty years, most of these materials are limited to work at relatively low temperatures.”

An effort is underway at Texas A&M to address some of these issues. The researchers hope to understand the nature of these phase transitions (where a system transforms thermodynamically from one state of matter to another) and then to apply that knowledge to the design of optimal SMAs for emerging applications.

Arroyave uses the Ranger supercomputer at the Texas Advanced Computing Center (TACC) to explore a new class of shape memory alloys that may be able to avoid some of the problems that plague its predecessors. Collaborating with experimentalist Ibrahim Karaman, a world expert on shape memory alloys, the duo is exploring forms of Cobalt-Nickel-Gallium alloys that they believe will work at high temperatures.

“This new class of high temperature shape memory alloys can be used in sensing and actuation at temperatures upwards of 200 Celsius, which is very important for the aerospace and the automotive industries,” Arroyave said.

Sensing refers to the process of characterizing the environment around a material by monitoring its physical response. Actuation is the opposite—making a material change shape and perform some action by applying a force.

SMAs are different from most metals due to their highly unstable microstructure. When the conditions of the environment change, the material finds a new ideal form.

“These materials undergo a spontaneous and reversible phase transformation,” said Arroyave. “When you lower the temperature, or in some cases stress the material or impose a magnetic field, the system tries to reach a different state that is more stable.” This natural shape shifting can be used in ways that scientists are just starting to understand.

Arroyave is also working with experimentalists to find the perfect “recipe” to obtain the properties that work best for specific applications. In this, Arroyave likens himself to a baker trying to find the ideal combination of ingredients to make the perfect piecrust. It may be possible to test every possible combination of flour, water and shortening, but that would take forever, Arroyave says. It’s better to determine some guidelines, start with a good rough estimate, and then go from there.

Zeroing in on that recipe requires a large number of pies—or in Arroyave’s case, numerical simulations, which is where the Ranger supercomputer comes in. Using Ranger, Arroyave mapped the constitution of the alloy as a function of temperature and composition, establishing a phase diagram of this particular system.

The Cobalt-Nickel-Gallium alloy is similar in structure to one of the most commonly used SMAs, and reacts in ways similar to that material. However, in the process of producing this map, Arroyave discovered that the fundamental dynamics that lead to the shape change are completely different, which was unexpected.

“For the past 10 years, people believed that in order to have shape memory transformations, you had to have certain features at the electronic structure level,” explained Arroyave. “It turns out that this is not the case.”

Arroyave and his collaborators published these findings in a recent paper in the journal, Acta Materialia. The discovery opens up new questions and new potential avenues for study.

“In the SMA field, most of the previous works on finding new alloys were based on simple thermodynamical models, ad-hoc approaches, or just trial-and-error experiments. But this requires enormous amount of experimental work and labor,” said Dr. Karaman. “The dream is to design materials atom by atom in near future. This can only be done with computational materials design starting from understanding electronic structure of these alloys, which is what Arroyave is trying to do.”

Ranger is allowing the researchers to do calculations that were impossible a few years ago, and for the first time approach the level of detail achieved in physical experiments. This is helping the computational materials science field prove it has something to offer in terms of the experimental design of new materials.

“From the scientific point of view, you cannot design what you cannot understand,” said  Arroyave. “We’re trying to come up with new explanations for material phenomena that have important applications.”

According to Arroyave, NASA and aerospace companies like Boeing have large efforts underway to develop high temperature shape memory for flight use. These alloys would allow the shape of a component of a jet engine to change in real-time. All of these efforts, however, require extensive computational modeling.

“Developing materials from scratch with experiments is possible, and has been done for the past 5,000 years,” concluded Arroyave. “The mission of computational materials scientists, like myself, is to narrow down the experimental search space that experimentalists need to explore in order to optimize a given material.”

Our material future

Computational materials science has a reputation for overselling and underperforming, according to Arroyave, but by all measures, the field is maturing by leaps and bounds.

New materials will not be discovered wholly by computers anytime soon, but the tools used to explore, design and optimize materials computationally are quickly improving.

In the current arena, a researcher who wants to study phenomena at different time and space scale has to use tools with different file formats and user interfaces, and become expert in all these codes.

But Arroyave believes this tide is turning and that the change will have a big impact on the field. “In the next five to ten years, a big shift will come in terms of the user-friendliness and the integration of these codes,“ he said.

Ease of use means more users, which in turn means more science and more discoveries. Combine continued software development and integration with the exponential growth of supercomputers, and computational materials science has the potential to be one of the most important fields in the coming decade.

Stay tuned to see what the scientists bake up next.