A UNC-Chapel Hill professor may have a line on how to solve one of the problems scientists and engineers have to crack before they can build robots that look and move like living things.
Chemist Sergei Sheiko and his lab team have figured how they can make silicone gels that perform, mechanically, a whole lot more like skin and muscle .
They're now making kilograms of the stuff, and think there are potential applications not just or even primarily in robotics but in biomedical work.
The key to the development lies not in manipulating the chemical composition of the gel, but its molecular structure, Sheiko said.
"What we do is change the architecture," he explained. "It's kind of a Home Depot approach. Everybody buys the same lumber and bricks, but people build different architectures. In the end, the architecture matters. The mechanical properties are controlled by the architecture. The chemical composition is secondary."
Sheiko and his team started working on the technology a couple years ago, publishing their findings along the way.
For the most part, their focus has been on potential biomedical applications.
But a conference presentation caught the eye of an editor from the journal Science Robotics, who convinced them to submit a two-page summary of their work for an edition at the end of the May that focused on how advances in materials science can feed into the development of robotics.
Before, "we honestly never looked in this direction," Sheiko said.
The reason the group's findings are so eye-catching is that living tissue has a fairly difficult set of mechanical properties to replicate.
It's soft, for starters, and flexible. But when you pull on it, it becomes quite strong, a property that along with being "very unique" is also "one of the major defense mechanisms" of humans and other creatures, Sheiko said.
On top of that, living tissue relies on water, which is less than ideal in robotics applications because water boils and freezes, narrowing the potential range of uses of any sort of artificial tissue that incorporates it.
Traditional materials, like rubber and even normal silicone, can replicate some of the characteristics of tissue but not all of them. But it's possible to improve on them by treating the gel as a composite material, one put together to have the necessary properties.
That follows because "tissue is a composite." said Sheiko, who's been at UNC since 2001. "That is what we tried to mimic, this composite structure."
The lab's find is opening the door to new collaborations, including one with a group at Harvard that's working on "soft robotics."
Closer to home, Sheiko's team is also working with a couple groups at N.C. State University, among them one headed by N.C. State professor Orlin Velev.
The collaboration with Velev, still in its early stages, is focusing on whether it's possible to run the gel through a 3-D printer to form them into particular shapes.
Velev said his team had already developed a way to 3-D print with soft materials like silicone for potential biomedical applications. But Sheiko's gels could turn out to be a better match for the way skin and other tissue moves, he said.
There's "a match of interests" and "we hope to be able to make quick progress within the next few months" adapting the gels to 3-D printing, Velev said.