King's College London

A project to develop an analogue computer able to decode the randomness of nature is being led by Dr James Millen from the Department of Physics. The device could help us understand unpredictable processes such as the way molecules and proteins move in biological systems.

Funded by the Leverhulme Trust, this five-year project has the potential to further our understanding of complex processes in physical, chemical and biological sciences which could improve technologies such as vaccines and nanomachines.

Dr Millen, Director of King’s Quantum said, “At the nanoscale, nature is complex and hard to predict. For example, the way molecules and proteins move inside biological systems can seem random. Nature has managed to harness this randomness to do useful things such as driving how the immune response works or how proteins fold and unfold. We want to learn how to make use of this randomness in our own nanotechnologies.”

“The obvious way to study natural events is to watch what happens in real time, but that takes too long. Traditional digital computer simulations fail to faithfully capture realistic complex scenarios because nature itself isn’t digital. By building an analogue simulation we can replicate, control and observe natural processes while being in control of the timeframe.”

The team will build a Levitated Physical Analogue Computer – L-PAC, which levitates a charged microparticle they can precisely control to understand natural processes at a microscale.

Video: Microparticles (orange ball) are levitated by complicated electrical fields (blue line), which can be used to emulate complex real-word problems - forming a levitated physical analogue computer.

Levitating the particles creates a blank slate, as they are not in contact with their environment. Electric fields then emulate realistic natural scenarios, with control signals to generate random forces. The response and motion of the particle simulates what’s happening in the real world.

While still in its conceptual stages, this work could result in a better understanding of the processes that occur in B-cell activation – the body’s memory response to pathogens – which could lead to a more rational approach to vaccine design.

In building simulations in an analogue way, the team are going against the current trend of harnessing machine learning and AI in research. By being tailored to specific applications, analogue computers are far more energy efficient than such intensive digital techniques.

Standard computer simulations struggle to efficiently reproduce the behaviour of scenarios involving memory such as the immune response, or with varying timescales such as the weather. Analogue technology has proven to be a powerful alternative approach for understanding financial markets, black holes and quantum fluids.

Dr Millen’s previous research has centred on studying physics using levitating nanoparticles, and he has been formulating this next step in his research for a long time.

“We know how to levitate particles but levitating them in such a way that we can synthesise all the model environments is going to be a challenge. To build the analogue computer, we will have to fabricate a chip-based device that we can programme with the environments we want to study, and develop methods to track the complicated particle motion.”

The research team includes fellow King’s academics Dr Stefano Bo, Professor Martin Riedle, Dr Rafael Tapia Rojo and Dr Katelyn Spillane; and collaborators at the University of Exeter.

“I see this as a completely new direction my research. It relies on all the things that are unique about working with levitated particles and it’s exciting because I get to collaborate with a wide range of scientists,” Dr Millen said.