Our work on the experimental demonstration of optical readout has been published in Nature Physics. Together with Rigetti and Qblox we have demonstrated a system that can measure the state of a superconducting qubit using optics, which holds significant potential for scaling quantum processors – a critical challenge in realizing quantum computing. The blog post below provides an accessible explanation and context for our results.
Quantum computing, one of the most ambitious goals humanity has ever set, is rapidly evolving from experimental research to commercially-oriented applications. A global effort is currently underway to develop the first practical quantum computers, whose unmatched computational power could drive revolutionary breakthroughs in fields such as advanced material design, artificial intelligence, and drug discovery. However, significant challenges remain before these transformative advancements can be achieved. At QphoX, we are tackling these challenges by developing technology that paves the way for quantum computing to fulfill its immense potential. Our latest research highlights how we can help quantum computing deliver on its promise.
One of the leading types of quantum computer is based on superconducting qubits, which are artificial atoms constructed from nanoscale superconducting electronic circuits. In these quantum computers, the quantum states of the qubits are manipulated and measured using microwave signals at frequencies typically around 5 GHz—similar frequencies as used by, for example, WiFi routers. For a quantum computer to function effectively, the qubits must be kept extremely cold—colder than the depths of outer space and just a fraction above absolute zero. This is achieved using machines called dilution refrigerators, which are capable of maintaining such extreme temperatures.
However, even dilution refrigerators have their limitations. They rely on a sophisticated mixing process, where two isotopes of helium are combined to extract heat from the surrounding environment. This process cools anything attached to the so-called mixing chamber, typically a large disk about half a meter in diameter. This disk houses the quantum processor along with all the microwave components, amplifiers, and cables required for its operation. The combination of thermal load and space requirements introduced by these components ultimately constrains the size of the quantum computer that can be accommodated inside a dilution refrigerator.
This size constraint poses a challenge. Quantum computers are highly prone to computational errors resulting from environmental noise. Fortunately, quantum error-correction algorithms—analogous to those used in classical computing—can make quantum computers more tolerant to such faults by encoding the information of many physical qubits into a single logical qubit. While state-of-the-art quantum computers currently use up to hundreds of qubits, millions will likely be required for practical, fault-tolerant quantum computing. This number far exceeds the capacity of today’s dilution refrigerators, necessitating a new approach to operating superconducting qubits.
One potential solution is to find an alternative medium for transferring microwave signals in and out of the cryostat. Such a medium must meet several criteria: it should minimize signal loss, act as an excellent thermal insulator, and occupy very little space. Optical fiber, a technology that has been used for decades to reliably transmit signals over vast distances, including across oceans, offers a promising answer. These fibers, essentially tiny cables made of glass, guide infrared light over long distances with minimal loss using the principle of total internal reflection. They are also remarkably compact, typically just a few hundred microns in diameter, comparable to the thickness of a few human hairs. Furthermore, optical fibers conduct very little heat, preserving much of the cryostat’s cooling capacity. Replacing conventional microwave cables with optical fibers in cryostats could substantially enhance the scalability of quantum computers.
The challenge lies in converting the microwave signals used to control qubits into infrared light that can be transmitted through fiber. This is where microwave-to-optics transduction comes into play, a field dedicated to the coherent conversion of microwave photons to optical photons. At QphoX, we develop piezo-optomechanical transducers that perform this conversion. These devices combine optomechanics, the interaction of light with mechanical motion, and piezo-electromechanics, which uses the piezoelectric effect to convert electrical signals into mechanical vibrations. This principle is similar to how ultrasound transducers work in medical imaging. In our devices, an electrical signal applied to the microwave input port causes a mechanical oscillator to vibrate. These vibrations are then imprinted onto a beam of infrared light, which is transmitted out of the cryostat using optical fiber and detected.
To demonstrate the potential of our technology, we teamed up with Rigetti Computing and Qblox and connected one of our transducers to a superconducting qubit, with the goal of measuring its state using light transmitted through an optical fiber. The results of this collaborative effort have been published in Nature Physics. In conventional microwave techniques, an electrical signal is used to interrogate the state of the qubit via its so-called readout resonator, a kind of external measurement apparatus attached to the qubit. After interacting with the readout resonator, this signal is amplified and detected electrically. Using one of our transducers however, it becomes possible to detect it using light, omitting the need, at least in part, for power-hungry and hot microwave amplifiers placed inside the cryostat. Remarkably, we discovered that, not only is our transducer capable of determining the state of the qubit, but that the qubit can also be sufficiently protected from decoherence introduced by any thermal noise or stray optical photons emanating from the transducer during operation.
These results highlight the promise of replacing traditional coaxial cables with optical fibers to scale up quantum computers. This approach could enable the development of large, fault-tolerant superconducting quantum processors. Moreover, we believe we are far from reaching the full potential of our transducers. In the coming years, we expect these devices to achieve groundbreaking performance metrics, reshaping the way superconducting qubits are operated.