Novel Microchip Sensors Could be the Holy Grail for Quantum Technologies, Sensing

TU Delft scientists have engineered one of the most precise microchip sensors in the world. The device can work at room temperature—a “holy grail” for sensing and quantum technologies.

Integrating nanotechnology and machine learning stimulated by spiderwebs found in nature, they could make a nanomechanical sensor vibrate in extreme isolation from real-world noise. This innovation, reported in Advanced Materials’ Rising Stars issue, has massive implications for the study of dark matter and gravity, as well as the fields of navigation, quantum internet, and sensing.

One of the major challenges for analyzing vibrating objects at the minutest scale, like those employed in quantum hardware or sensors, is how to prevent room thermal noise from interacting with their delicate states. Quantum hardware, for example, is typically maintained at near absolute zero (−273.15°C) temperatures, with the price of refrigerators going up to half a million euros for a single unit.

Scientists from TU Delft have designed a web-shaped microchip sensor that resonates very well in isolation from ambient temperature noise. Among other uses, their creation will make developing quantum devices a lot more inexpensive.

Hitchhiking on Evolution

Richard Norte and Miguel Bessa, who headed the study, were seeking novel ways to integrate machine learning and nanotechnology. How did they discover the concept of using spiderwebs as a model?

I’ve been doing this work already for a decade when during lockdown, I noticed a lot of spiderwebs on my terrace. I realised spiderwebs are really good vibration detectors, in that they want to measure vibrations inside the web to find their prey, but not outside of it, like wind through a tree.

Richard Norte, Researcher and Study Lead, TU Delft

Richard Norte then asked: “So why not hitchhike on millions of years of evolution and use a spiderweb as an initial model for an ultra-sensitive device?” 

Since the team was not aware of the complexities of spiderwebs, they allowed machine learning direct the discovery process.

We knew that the experiments and simulations were costly and time-consuming, so with my group we decided to use an algorithm called Bayesian optimization, to find a good design using few attempts.

Miguel Bessa, Researcher and Study Lead, TU Delft

Dongil Shin, the study’s co-first author, then applied the computer model and used the machine learning algorithm to discover the new device design.

Microchip Sensor Based on Spiderwebs

To the researcher’s astonishment, the algorithm suggested a fairly simple spiderweb from among 150 diverse spiderweb designs, which comprises just six strings put together in a deceivingly uncomplicated manner.

Dongil’s computer simulations showed that this device could work at room temperature, in which atoms vibrate a lot, but still have an incredibly low amount of energy leaking in from the environment—a higher Quality factor in other words. With machine learning and optimization we managed to adapt Richard’s spider web concept towards this much better quality factor.

Miguel Bessa, Researcher and Study Lead, TU Delft

Based on this unique design, co-first author Andrea Cupertino constructed a microchip sensor using an ultra-thin, nanometer-thick film of ceramic material known as Silicon Nitride. They examined the model by strongly vibrating the microchip “web” and computing the time it takes for the vibrations to come to a stop. The outcome was remarkable: a record-breaking isolated vibration at ambient temperature.

Norte: “We found almost no energy loss outside of our microchip web: the vibrations move in a circle on the inside and don’t touch the outside. This is somewhat like giving someone a single push on a swing, and having them swing on for nearly a century without stopping.”

Implications for Fundamental and Applied Sciences

The researchers used their spiderweb-based sensor to demonstrate how this interdisciplinary approach paves the path to new innovations in science, by merging bio-stimulated designs, nanotechnology, and machine learning.

This unique model has exciting implications for quantum internet, microchip technologies, sensing, and fundamental physics: investigating ultra-small forces, for example, like gravity or dark matter that is extremely hard to measure.

According to the scientists, the discovery would have been difficult to achieve without the university’s Cohesion grant, which led to this partnership between machine learning and nanotechnology.

Journal Reference:

Shin, D., et al. (2021) Spiderweb Nanomechanical Resonators via Bayesian Optimization: Inspired by Nature and Guided by Machine Learning. Advanced Materials. doi.org/10.1002/adma.202106248.