John Sorebo received Master’s degrees from the University of Wisconsin – Madison in both Electrical and Computer Engineering and Nuclear Engineering and Engineering Physics. He is a Principal Engineer at Synopsys supporting government programs in the areas of quantum computing, photonics, and radiation-hardened microelectronics design.
Presentation Title:
Design Workflows Utilizing Simulation & Analysis for Accelerating Fault Tolerant Quantum Computing Design
Presentation Abstract:
The superconducting transmon qubit design modality is touted as being able to leverage existing semiconductor process technologies, making their scaling in numbers cheaper, faster, and more deterministic. However, for scaling at the system level, characterization and modeling of noise sources that contribute to qubit dissipation is a critical link between qubit manufacturing process optimization and simulation of the dynamics to inform the design of the electronic interfaces and error-correcting schemes. Physical models specific to the qubit modality that correlate the density and spatial distribution of noise sources with the dissipation channels are essential to guide the optimization of qubit quality and system-level design of the FTQC stack. Two simulation applications specific to superconducting qubits will be presented.
Superconducting transmon qubit designs allow for fast gate times compared to other quantum embodiments and offer scalability advantages due to their design compatibility with existing semiconductor fabrication techniques. However, due to requiring pulsed RF control and measurement signals, EM crosstalk represents a critical concern for error-free qubit operation. While tunable coupler designs can be used to null nearest-neighbor couplings, crosstalk with next-nearest-neighbors and beyond can only be mitigated with simulation-informed design that can be verified empirically and calibrated. Coupling mechanisms associated with qubits are identified using physics-based analysis along with laboratory measurements of early qubit design prototypes. Techniques are being developed to address the computational challenges of simulating crosstalk among thousands of neighboring qubits.
Decoherence in superconducting qubits remains a major barrier to scalability in quantum computing. While bulk Si and Al are non-piezoelectric, recent experimental studies have identified interface-induced piezoelectricity at the Si–Al boundary as a significant loss channel. In this work as part of the DARPA Quantum Benchmarking Initiative (QBI), quantum-mechanical atomistic simulations are used to quantify the piezoelectric response of various Si–Al interface configurations. The methodology presented provides a computational framework for engineering low-loss interfaces in superconducting qubit technologies.