When the Robot Becomes the Researcher
New autonomous researcher can speed up discovery of the best 3D-printed materials
By Liz Sheeley
Additive manufacturing, or 3D printing, has vast applications some of which are in the medical, aerospace and consumer fields. To advance the area further, researchers must study the best way to produce 3D-printed parts that can withstand more rigorous use, more use than a 3D-printed prototype for research would receive.
Assistant Professor Keith Brown (ME, MSE, Physics) and Professor Elise Morgan (ME, MSE, BME) have developed a way to test the mechanical properties of thousands of 3D-printed structures to catalog and understand them in an extremely efficient way. Their work has been published in Science Advances.
Their collaboration came out of a desire to test the toughness of materials—Morgan has expertise in the mechanical behavior of biological materials and Brown in nanotechnology and soft materials. Over the past few years, they have built an autonomous robot, or researcher as they call it, to run this toughness experiment on its own, and it has since cataloged thousands of 3D-printed structures.
“I was thinking about a 3D printing system where you can create a structure, test it and then learn from it,” says Brown. “If you’ve got an autonomous system that you could give it a question, something like, ‘I want to know what kind of structure is going to have the best property.’ Then it will automatically test structures, and tell you the answer.”
The automated researcher, known as BEAR (Bayesian experimental autonomous researcher), was designed to perform experiments on its own—designing and testing parts to determine their mechanical properties. The system can print a print a 3D structure, remove it from the printer, weigh using a scale, and then crush it. It records every detail of the process, creating a vast database of structures and data related to how the materials behave when they are compressed, including whether and how they fail.
“It became clear to me that one of the most interesting questions you can ask is how do materials fail?” says Brown. “Failure is really interesting because it’s very hard to simulate. It’s pretty clear how a material or structure is going to behave under gentle pressure, but as you push it to the point where it starts to crumble, it becomes much more complicated. And experiments are really necessary to test it.”
Brown knew that Morgan’s expertise was in bone biomechanics. “And, of course, one of the things we worry about with bones is them breaking, right? We started talking about how we can learn about different failure mechanisms from an automated researcher,” he says.
The property that Brown and Morgan first chose to explore is toughness—which is directly related to failure. Toughness and strength are related, but separate properties. Glass, Morgan says, is actually stronger than steel—but only when it has no defects. And defects are almost impossible to prevent during even the most advanced manufacturing.
“If you have glass with a scratch on the surface, it’s going to fracture from a much smaller load than steel,” she explains. “Steel is much more defect tolerant, meaning it’s much tougher than glass.”
“Over many centuries of work in materials and in mechanical engineering more broadly, we have gotten good at designing, fabricating and manufacturing strong materials, but designing tough materials has been definitely has lagged behind,” she says. “But it turns out that, nature’s pretty good at designing tough materials. Bone, for instance, is surprisingly tough, considering what it’s made of.”
The bone itself is made of mineral, which is pretty brittle—strong, but not tough. The mineral is attached to collagen which is stretchy—but not necessarily strong or tough. In bone, these two materials are interwoven in a complicated way, boosting the toughness of bone beyond what is exhibited by either material alone. But how these two materials work together to give this boost is not well understood.
They also pulled in another faculty member, Assistant Professor Emily Whiting (Computer Science), to help with the complicated algorithm development. In order to have the system be autonomous—not just automated—it needed to run machine learning algorithms during the testing and evaluation phases. While the system runs, it learns about failure in 3D-printed parts, and can choose the next design to test based on past results.
This research with BEAR is just one example of how an automated researcher can be used to speed up experiments that are typically slow and monotonous. And although through developing and optimizing BEAR more than 2,500 structures were tested, only 32 were needed to reach the optimal structure design for toughness—that showed them that they can test much more complicated structures than they have so far.
Brown’s lab is focused on developing new methods, typically tools, to investigate properties of polymers and soft matter across many length scales from the microscopic to the macroscopic (from 10 nanometers to a few centimeters). He is motivated by the lack of knowledge about the properties of materials that are structured on several of these scales, which is driven by the absence of the necessary tools.
“With advances in additive manufacturing, you make very complicated hierarchical component designs,” says Morgan. “But when trying to find the best design, the rate-limiting step is testing them and evaluating the results, but the advances in autonomous systems and robotics have enabled us to open up that rate-limiting step quite a bit.”
Brown and Morgan say that these new structures could be used for protection, such as pads inside of a helmet, and potentially synthetic bone substitutes in the future.