Quantum Chip Created Where Memory Uses Vibrations to Store Data

A research group from ETH Zurich, led by physicist Yiwen Chu, has developed a new architecture for quantum computing that addresses one of the main limitations in the field: storing quantum information more compactly and efficiently. The work, published in the journal Science, demonstrates for the first time a coupling between superconducting qubits and mechanical resonators, components that store data using vibrations rather than electromagnetic signals.

The architecture takes inspiration from traditional computers, where CPUs and RAM perform distinct roles. In this case, the superconducting qubit acts as the processing unit, while the quantum memory relies on mechanical resonators. During computation, the qubit reads information from memory, processes it, and rewrites it, following a conceptually similar model to that of a classical processor.

The researchers compare the operation of the resonators to guitar strings, which can vibrate in different modes producing different notes. Similarly, each resonator supports different vibrational modes, each representing a memory slot. Within each mode, different quantum states can be encoded.

Unlike classical systems, limited to states 0 and 1, these states can exist in superposition or be entangled, fundamental properties of quantum computing that allow for representing and manipulating many more pieces of information simultaneously.

According to the team, mechanical resonators offer significant advantages over the currently predominant electromagnetic memories in superconducting qubit-based systems. They occupy less space, support a greater number of vibrational modes, and allow for higher storage density. Additionally, they maintain quantum states coherently for longer, extending the available time for computations.

To validate the architecture, the researchers implemented two fundamental procedures of quantum computing: the quantum Fourier transform and period finding. Both algorithms require the control, storage, and coherent manipulation of numerous quantum states. "The quantum Fourier transform is a fundamental computational procedure required by many quantum algorithms," explained Igor Kladaric, a PhD student in Chu's group and co-author of the study. "The period finding algorithm we implemented served to demonstrate how this procedure can be practically employed."

The prototype chip is about 7.5 millimeters long. According to the researchers, it is already capable of performing all the basic operations necessary for a programmable and general-purpose quantum computer, representing a proof of feasibility for the architecture. The next step will be to verify its scalability. The system will need to demonstrate that it can maintain the same performance even in quantum computers with a significantly larger number of qubits and computational capacity. For the ETH Zurich group, this is nonetheless a promising result toward more compact, reliable quantum platforms suitable for future scientific and industrial applications.