Solving the Quantum Interconnects Bottleneck; Scalable Photonic Control for Scalable Quantum Computers

After decades of experimental and theoretical effort, quantum computers are reaching a scale that is no longer simulable by classical devices. While these quantum computers are still too noisy to beat classical supercomputers at tasks like prime number factorization, recent advances are already indicating that Noisy Intermediate-Scale Quantum (NISQ) devices have the potential for practical “quantum advantage” on certain technologically important tasks, such as quantum simulation of many-body quantum physics problems or combinatorial optimization problems. Moreover, steady progress towards useful error correction and more efficient algorithms are bringing gate-based quantum architectures closer to the ultimate goal of universal, fault-tolerant quantum computers.  In recent years, quantum computing architectures based on arrays of cold atoms have emerged as the “dark horse” candidate of quantum computing, offering not just competitive gate fidelities and excellent qubit coherence, but also the potential for near-term quantum advantage and scalability to qubit counts in the millions. Based on recent Harvard-MIT breakthroughs in cold atom computing, QuEra Computing Inc was established in 2019 to build the world’s most powerful quantum computers to solve currently intractable problems.  However, all quantum computing platforms based on atoms or trapped ions face a common inevitable bottleneck: the need for large numbers of independently controllable laser beams to initialize, control, and measure qubits. Whereas these beams are currently created using bulk optical components, such as acousto- and electrooptic modulators, such approaches are not scalable to large channel counts (hundreds) due to prohibitive complexity and cost. To unlock the true scaling potential of atom-based quantum computers, it is necessary to develop  photonic devices designed from the ground up for scalability.  In this STTR program, QuEra and Sandia National Laboratory propose to resolve this bottleneck by developing fully CMOS-compatible large-scale photonic integrated circuits (PICs) operating in the relevant UV-VIS spectrum, as well as the required CMOS-based electronic control. Phase I focuses on proof of concept and design studies: a PIC demonstrator for high-channel-count beam arrays and its capability validation for large-scale quantum computing. Phase II would advance this work to CMOS-PIC system demonstrations with 128 laser beams with high-fidelity, high-speed modulation. A successful STTR would open the path to NISQ devices for near-term practical quantum advantage and to fault-tolerant, scalable quantum computers.