Someone taps your shoulder. The touch receptors arranged on your skin send a message to your brain, which processes the information and directs you to look to the left, in the direction of the faucet. Now, researchers at Penn State and the US Air Force have harnessed this mechanical information processing and integrated it into engineered materials that “think.”
The work, published today (August 24) in Nature, is based on a new, reconfigurable alternative to integrated circuits. Integrated circuits are typically composed of multiple electronic components housed in a single semiconductor material, usually silicon, and work with all kinds of modern electronics, including phones, cars, and robots. Integrated circuits are scientists’ realization of information processing similar to the role of the brain in the human body. According to principal investigator Ryan Harne, James F. Will Associate Professor of Professional Development in Mechanical Engineering at Penn State, integrated circuits are the basic building block needed for scalable signal and information computing, but scientists have never before made a different composition of silicon. semiconductors
His team’s discovery revealed the opportunity for almost any material around us to act as its own integrated circuit: to be able to “think” what’s going on around it.
“We have created the first example of an engineered material that can simultaneously sense, think and act on mechanical stress without the need for additional circuitry to process these signals,” Harne said. “The soft polymer material acts as a brain that can receive digital strings of information that are then processed, resulting in new sequences of digital information that can control reactions.”
The soft, conductive mechanical material contains reconfigurable circuits that can perform combinational logic: when the material receives external stimuli, it translates the input into electrical information that is then processed to create output signals. The material could use mechanical force to calculate complex arithmetic, as Harne and his team demonstrated, or detect radio frequencies to communicate specific light signals, among other potential examples of translation. The possibilities are vast, Harne said, because integrated circuits can be programmed to do so much.
“We figured out how to use mathematics and kinematics, how individual components of a system move, in mechanical-electrical networks,” Harne said. “This allowed us to realize a fundamental form of intelligence in engineering materials by facilitating fully scalable information processing intrinsic to the soft material system.”
According to Harne, the material uses a “thinking” process similar to that of humans and has potential applications in autonomous search and rescue systems, infrastructure repairs and even bio-hybrid materials that can identify, isolate and neutralize airborne pathogens.
“What makes humans intelligent is our means of observing and thinking about the information we receive through our senses, thinking about the relationship between that information and how we can react,” Harne said.
Although our reactions may seem automatic, the process requires the body’s nerves to digitize sensory information so that electrical signals can travel to the brain. The brain receives this information sequence, evaluates it and tells the body to react accordingly.
For the materials to process and think about information in a similar way, they must perform the same complex internal calculations, Harne said. When researchers subject their engineered material to mechanical information, applied force that deforms the material, they digitize the information into signals that their electrical network can advance and evaluate.
The process builds on the team’s previous work developing a soft, mechanical metamaterial that could “think” about how forces are applied to it and respond through programmed reactions, detailed in Nature Communications last year. This earlier material was limited only to logic gates that worked with binary input-output signals, according to Harne, and had no way to compute high-level logic operations that are central to integrated circuits.
Researchers were stumped until they rediscovered a 1938 paper published by Claude E. Shannon, who later became known as the “father of information theory.” Shannon described a way to create an integrated circuit by constructing mechanical-electrical switching networks that follow the laws of Boolean mathematics, the same binary logic gates that Harne used earlier.
“Ultimately, the semiconductor industry did not adopt this method of manufacturing integrated circuits in the 1960s, choosing instead to use a direct assembly approach,” Harne said. “Shannon’s math-based design philosophy was lost in the sands of time, so when we read the paper, we were surprised that our preliminary work matched Shannon’s vision exactly.”
However, Shannon’s work was hypothetical, produced nearly 30 years before integrated circuits were developed, and did not address how to scale networks.
“We’ve made considerable modifications to Shannon’s design philosophy to make our electrical-mechanical networks meet the reality of integrated circuit assembly rules,” said Harne. “We took our basic logic gate design philosophy from the 2021 research and fully synchronized the design principles with those articulated by Shannon to ultimately yield mechanical integrated circuit materials—the effective brain of artificial matter “.
Researchers are developing the material to process visual information as it does physical signals.
“We are currently translating this into a ‘see’ medium to augment the ‘touch’ sensation we have currently created,” Harne said. “Our goal is to develop a material that demonstrates autonomous navigation through an environment by seeing signals, following them, and maneuvering out of the way of adverse mechanical force, such as stepping on something.”
Other authors of the paper include Charles El Helou, a doctoral student in mechanical engineering at Penn State, and Benjamin Grossman, Christopher E. Tabor and Philip R. Buskohl of the US Air Force Research Laboratory.
Harne’s National Science Foundation Early Career Development Award and the US Air Force funded this research.
Reference:
- Charles El Helou, Benjamin Grossmann, Christopher E. Tabor, Philip R. Buskohl, Ryan L. Harne. Mechanical integrated circuit materials. Nature, 2022; 608 (7924): 699 DOI: 10.1038/s41586-022-05004-5
/ Public communication. This material from the original organization/authors may be ad hoc in nature, edited for clarity, style and length. The views and opinions expressed are those of the author(s). See them in full here.