[Paper Review] An integrated microwave-to-optics interface for scalable quantum computing

Matthew J. Weaver, Pim Duivestein+8 more|arXiv (Cornell University)|Oct 27, 2022

Photonic and Optical Devices40 references5 citations

TL;DR

This paper presents an integrated microwave-to-optics transducer using a lithium niobate-on-silicon mechanical oscillator to enable low-noise, high-efficiency quantum state transfer between superconducting qubits and optical fibers. The device achieves 0.9% transduction efficiency, adds only ~6 photons of noise, operates at 14.8 MHz bandwidth and up to 100 kHz repetition rate, with direct 50-Ω impedance matching for scalable integration in dilution refrigerators.

ABSTRACT

Microwave-to-optics transduction is emerging as a vital technology for scaling quantum computers and quantum networks. To establish useful entanglement links between qubit processing units, several key conditions have to be simultaneously met: the transducer must add less than a single quantum of input referred noise and operate with high-efficiency, as well as large bandwidth and high repetition rate. Here we present a new design for an integrated transducer based on a planar superconducting resonator coupled to a silicon photonic cavity through a mechanical oscillator made of lithium niobate on silicon. We experimentally demonstrate its unique performance and potential for simultaneously realizing all of the above conditions, measuring added noise that is limited to a few photons, transduction efficiencies as high as 0.9%, with a bandwidth of 14.8 MHz and a repetition rate of up to 100 kHz. Our device couples directly to a 50-Ohm transmission line and can easily be scaled to a large number of transducers on a single chip, paving the way for distributed quantum computing.

Motivation & Objective

• Address the scalability challenge in superconducting quantum processors by enabling long-distance quantum interconnects using optical fibers.
• Overcome limitations of microwave-based interconnects, which are restricted by cryogenic cooling and cabling constraints.
• Develop a transducer that simultaneously achieves low added noise, high efficiency, large bandwidth, and high repetition rate for practical quantum networking.
• Enable on-chip integration of multiple transducers by using a compact, 50-Ω compatible, and scalable design.
• Minimize quasiparticle generation and thermal loading to preserve superconducting qubit coherence.

Proposed method

• Integrate a planar superconducting molybdenum-rhenium microwave resonator with a silicon photonic cavity via a lithium niobate piezoelectric mechanical oscillator.
• Use a co-localized photonic and phononic crystal cavity to enable strong coupling between microwave, mechanical, and optical modes.
• Implement capacitive coupling to a 50-Ω transmission line for direct electrical interfacing with superconducting qubits.
• Employ evanescent coupling between the optical waveguide and photonic crystal cavity to achieve high optical quality factors.
• Use piezoelectric electrodes on the lithium niobate block to transduce microwave signals into mechanical vibrations and subsequently into optical signals.
• Characterize transduction performance via microwave reflection, optical resonance measurement, and mechanical mode detection using laser modulation.

Experimental results

Research questions

• RQ1 Can an integrated transducer achieve sub-1-photon added noise while maintaining high efficiency and large bandwidth?
• RQ2 Can a single chip support thousands of transducers with minimal heat load and crosstalk for scalable quantum computing?
• RQ3 Does the use of thin-film lithium niobate on silicon enable low-noise, high-efficiency transduction with 50-Ω impedance matching?
• RQ4 To what extent does optical power induce quasiparticle generation that degrades superconducting resonator performance?
• RQ5 Can pulsed operation minimize quasiparticle effects while enabling high repetition rates for quantum state transfer?

Key findings

• The transducer achieves a measured added noise of approximately 6 photons, approaching the quantum limit for practical quantum communication.
• Transduction efficiency reaches 0.9% in continuous-wave operation and 5×10⁻⁵ in pulsed operation, with a 14.8 MHz bandwidth and up to 100 kHz repetition rate.
• The device operates with only 1 nW average laser power and a footprint of less than 0.15 mm², enabling integration of over 10,000 transducers in a single dilution refrigerator.
• Pulsed operation induces only a +200 kHz frequency shift and negligible loss increase, indicating minimal quasiparticle generation and thermal impact.
• The transducer is impedance-matched to a 50-Ω transmission line, enabling direct connection to superconducting qubits without additional components.
• Optical frequency multiplexing could reduce fiber count to ~100 for a large-scale processor, with a predicted passive heat load of only 300 pW.