A blueprint for error-corrected fermionic quantum processors

An international research team led by Robert Ott and Hannes Pichler has developed a novel architecture for quantum processors that is specifically designed for simulating fermions—particles such as electrons. The method can be implemented using technologies already available today.

The study is published in the journal Physical Review Letters.

Fermions are a fundamental class of particles in physics. They include electrons, protons, and neutrons—the building blocks of all matter. They are subject to the so-called Pauli principle, according to which two fermions can never occupy exactly the same quantum state. This behavior determines essential properties of matter, such as the structure of atoms or the behavior of electrons in materials.

The team at the Department of Theoretical Physics from the University of Innsbruck and the OEAW-Institute for Quantum Optics and Quantum Information (IQOQI) shows how neutral fermionic atoms in optical traps can be used to efficiently replicate the complex properties of molecules and materials.

By using individual neutral atoms, which are themselves fermionic particles, the fermionic properties are realized directly in the hardware.

Error correction in fermionic systems

One challenge that has remained unresolved in this type of quantum processor is the integration of error correction. Classical strategies from quantum computing face limitations here, as the number of particles in atomic systems is typically fixed—a prerequisite that rules out many established correction methods.

The new study overcomes this obstacle with an innovative concept: a so-called “fermionic reference” consisting of additional atoms.

“This fermionic reference enables the controlled exchange of particles between the processor and the reference, allowing effective superpositions of different particle numbers to be realized. This makes it possible to correct errors even in systems with a fixed number of particles,” says the study’s lead author, Robert Ott.

Errors reduced by an order of magnitude

In their work, the researchers demonstrate how this architecture can be used to build fundamental building blocks for fault-tolerant computing. They also show that common types of errors—especially phase errors—can be identified and corrected with high efficiency. In simulations, the probability of errors was reduced by an order of magnitude.

“This work paves the way for more precise and scalable fermionic quantum computers using technologies that are already available today,” says Hannes Pichler, who leads the research group.

“It opens up new perspectives for the simulation of complex quantum physical systems, particularly in chemistry and materials science.”