MACHINING FOR GECKO-INSPIRED ADHESIVES
Gecko-inspired adhesives show great promise for a variety of applications—a new study will make specifying tool trajectories for microscopic features more efficient.
Written by Kayt Sukel
GECKOS—SMALL, CARNIVOROUS LIZARDS—have the remarkable ability to traverse different environments thanks to their special feet. Millions of setae, a type of microscopic, hair-like structure, allow these lizards to easily adapt and adhere to a wide variety of surfaces. It’s these unique mechanical forces that have inspired a number of gecko-based adhesive applications ranging from climbing robotic systems to biomedical devices.
There’s just one problem, said Mark R. Cutkosky, Fletcher Jones Professor in the School of Engineering at Stanford University. That’s the challenge that micromachining molds for scaled manufacturing of such adhesives present thanks to their unique and microscopic geometries.
“Normally, when you do computer numerical control [CNC] machining, whether it’s for regular parts of microscopic features, you have to plan a tool trajectory that matches the geometry of the features you are trying to create,” Cutkosky said. “In this case, we can’t do that. It’s a bit like the experience of bending a piece of steel wire knowing you have to bend it a bit further than you want because it will spring back. While this isn’t exactly the same, it’s a similar concept. To get these features, you have to anticipate the shape you’ll need when machining, and that’s a real challenge.”
To create the necessary molds to manufacture microscopic features for different applications, the team would previously have to go through a painstaking process of trial and error to come up with the right geometry. It made Cutkosky and colleagues wonder if they could use simulation to come up with the right micromachining paths in less time.
To answer this very question, a member of Cutkosky’s laboratory, Amar Hajj-Ahmad, who is now a research scientist at the Robotics and AI institute (RAI) in Boston, investigated the use of plane-strain finite element analysis (FEA) specialized for metalworking. She tested the approach with both wax and soft aluminum molds to create different patterns of wedge-shaped features often leveraged in gecko-inspired adhesive applications.
“We went through a simulation experiment and then machined it a CNC with the custom tooling we have for a specific material, whether it was soft metal or a wax material,” she said. “Then we compared the two things: the resulting geometry, the measurements of the dimensions you are getting, as well as the forces you experience while you are machining.”
When the group compared the results from the simulation to their machining, they found it matched to within 2.8 percent. Cutkosky said he was surprised at how accurate the simulation results were.
“It’s a very tricky machining process—and the effect you get depends on getting things like machining properties and the strain hardening characteristics of the material just right,” he said.
Hajj-Ahmad said that classical machining offers “instant feedback,” allowing you to see the shapes you are creating in real-time. When you are working on this kind of micro-scale, however, it takes longer—you need to actually machine and cast from your result to see whether it’s correct. These results, she said, show that simulation can give you good feedback, and ultimately save engineers significant time to create the molds required to manufacture these types of adhesives.
“If someone wants to do something in 3-D for type of shape, simulation is going to be a powerful tool,” she said.
A nonstick wedge
The researchers simulated three different trajectories to create a different cavity shape.
Adhesive clutch wedge
This shape was intended to produce a stiffer wedge than the conventional approach.
Angled wedge
Here, the team used angled wedges as a directional thermal absorber. All three used silicone cast wedges from machined features in wax.
Vertical indenting
This chart shows the force per millimeter of blade width, recorded with a load cell. The solid curve is the average value and the outlining dashed curves show ±3σ.
Indenting continued
The comparable maximum indenting force in simulation for the same parameters as the chart to the left. Both charts indicate that measured and simulated forces closely match.
Visualizing strain
A snapshot of the stress and strain distribution while cutting a cavity in wax, shown as the tool reaches a maximum depth of 100 μm.
Seeing stress
The highest effective stress values were near the tip of the blade, with both forward and back pressure exerted on the material.
“It’s a very tricky machining process—and the effect you get depends on getting things like machining properties and the strain hardening characteristics of the material just right.”
—Mark R. Cutkosky, Fletcher Jones Professor in the School of Engineering at Stanford University
Cutkosky agreed. As interest in gecko-inspired adhesive applications grow, the use of simulation can help support the creation and manufacturing of slightly different geometries for different purposes, he added.
“This gives us a more systematic way to arrive at the right machining details without having to go through that lengthy trial and error process,” Cutkosky said. “With 3D additive manufacturing getting better and better with smaller features, this shows that direct machining is still relevant. We can still achieve features that additive manufacturing struggles with and we can do it reliably.”
The team set the study up to “isolate just what’s happening as a result of the tool,” understanding different users have varying experience and comfort levels, Hajj-Ahmad explained.
Results showed no statistical difference in the number of sketches produced on paper versus the tablet. Earlier research already established that the quantity metric was an important factor in achieving better final design outcomes.
“Having a much larger quantity of sketches early means you have a better likelihood of having good ideas within that set,” Hajj-Ahmad said.
Quality was evaluated using criteria that included line smoothness and accurate proportions. Surprisingly, pen and paper sketches won out. Expectations that the tablet’s built-in ability to help engineers draw perfectly proportioned, shaped, and smooth lines did not translate to better perceived quality. This can influence how people think about the design as a finished product, versus a concept that invites feedback, Hajj-Ahmad said.
“We went through a simulation experiment and then machined it a CNC with the custom tooling we have for a specific material, whether it was soft metal or a wax material. Then we compared the two things: the resulting geometry, the measurements of the dimensions you are getting, as well as the forces you experience while you are machining.”
—Amar Hajj-Ahmad, research scientist at the Robotics and AI institute in Boston
“If you have a really high-quality sketch, other people might perceive it as being more creative, regardless of what the idea is,” so it depends on what you’re going for, Hajj-Ahmad noted. “Having one really nice drawing is nice, but it’s not that effective as a strategy for ideas.”
Though quality is certainly important, it’s not something engineers should worry about when it comes to sketching, Hajj-Ahmad said. More important is understandability, a metric that hadn’t received enough attention in previous research. The research evaluated understandability by comparing sketches to a separate written description of the concept to see how closely they matched.
Hajj-Ahmad used the game Pictionary to explain. “You have a specific goal. It’s not to make a beautiful drawing, it’s to make sure that other people understand what it is that you’re drawing,” she said. “Those are the two things that are really important: quantity of sketches and whether or not it’s an effective communication tool. And those are the ones where they perform at the same level.”
The study showed no differences in sketch quantity, quality, and understandability between genders nor by the tool used. This kind of process research is important as technology evolves. It may help designers better understand which tools offer the most value as the design is developed, such as AI at concept generation or selection, or VR for prototyping, Hajj-Ahmad said.
“I see it as a matrix of so many different tools, especially digital tools that are changing all the time. There are research questions at the intersection of every one of those cells with matrix,” she added.
This area of research is about going beyond objectivity and asking what it means to do better design. “As engineers, everything that we're creating and putting out into the world has an impact on people. It’s actually a very social endeavor,” Hajj-Ahmad said.
Kayt Sukel is a technology writer and author in Houston.
© 2025 The American Society of Mechanical Engineers. All rights reserved.