2021-11-30

Inspired by spider web, one of the most accurate microchip sensors in the world was born

By yqqlm yqqlm

Visit: Alibaba cloud 11.11 cloud Carnival activity hall

6998b4b4df06db9 - Inspired by spider web, one of the most accurate microchip sensors in the world was born

one of the biggest challenges in studying the smallest scale vibrating objects, such as those used for sensors or quantum hardware, is how to keep the environmental thermal noise from interacting with their fragile state. For example, quantum hardware is usually stored at a temperature close to absolute zero (- 273.15 ° C). The price of this special refrigerator is 500000 euros. Researchers from Delft University of technology have created a mesh microchip sensor that can produce excellent resonance when isolated from room temperature noise. Their discovery will make the construction of quantum devices cheaper

Inspired by spider web, one of the most accurate microchip sensors in the world was born

inspired by the spider web of nature and guided by machine learning, Richard nott (left) and Miguel Bessa (right) demonstrated a new sensor in the laboratory

Richard nott and Miguel Bessa, who led the research, are looking for new ways to combine nanotechnology and machine learning. How did they think of using cobwebs as models? Nott said “I have been doing this work for ten years. During my confinement, I noticed a lot of cobwebs on my balcony. I realized that spider webs are very good vibration detectors because they need to measure the vibration inside the web to find their prey, rather than the vibration outside the web, such as the wind blowing through trees. So why not take a ride on millions of years of evolution and use the spider web as the initial model of a hypersensitive device? ”

because the team knew nothing about the complexity of cobwebs, they let machine learning guide the discovery process.” we know that experiments and simulations are expensive and time-consuming, so we decided to use an algorithm called Bayesian Optimization with few attempts to find a good design. “Dongil shin, the co-author of this work, then implemented the computer model and applied the machine learning algorithm to find a new device design.

to the researchers’ surprise, the algorithm proposed a relatively simple spider web in 150 different spider web designs, which consists of only six strings combined in a deceptive way.” dongil’s computer simulation shows that, This device can work at room temperature, where atoms vibrate greatly, but the energy leaked from the environment is still very low, in other words, the mass coefficient is higher. Through machine learning and optimization, we try to adapt Richard’s spider web concept to this better quality coefficient. ”

based on this new design, the co first author Andrea Cupertino built a microchip sensor with a layer of ultra-thin, nano-sized ceramic material silicon nitride. They vibrated the microchip network by force “And measure the time required for the vibration to stop to test the model. The result was spectacular: the isolated vibration at room temperature broke the record, and there was almost no energy loss on the outside of the microchip network: the vibration moved around inside and did not touch the outside. This is like giving people a push on the swing, which can keep them swinging for nearly a century

through their spider web based sensors, researchers showed how this interdisciplinary strategy opened the way for new breakthroughs in science by combining bioinspired design, machine learning and nanotechnology. This novel paradigm has an interesting impact on Quantum Internet, sensing, microchip technology and basic physics: for example, exploring ultra-small forces, such as the well-known difficult to measure gravity or dark matter. According to the researchers, this discovery would not have been possible without the cohesive support of universities, which led to the cooperation between nanotechnology and machine learning