New Simulator Makes Quantum Hardware Design Faster and More Efficient

A new computational approach addresses a key challenge in superconducting qubit design. Ziang Wang and colleagues at Tsinghua University and partner institutions present SesQ, a surface electrostatic simulator designed to accurately and efficiently model dielectric losses.

The team developed SesQ to overcome limitations of conventional finite element methods, which struggle with the computational demands of simulating energy participation ratios, a key metric for evaluating qubit performance. By using a surface integral equation method and a refined boundary mesh, SesQ achieves sharply faster capacitance extraction and more precise EPR calculations than existing tools, ultimately enabling rapid optimisation of low-loss superconducting quantum circuits. Simulating dielectric loss in superconducting qubits via surface integral equation methods A new simulator, SesQ, models energy loss in superconducting qubits, a crucial factor in enhancing quantum computer performance. Superconducting qubits, favoured for their potential in scalable quantum computation, are exceptionally sensitive to dielectric losses, which degrade coherence and limit computational fidelity. These losses arise from imperfections in the materials and interfaces within the qubit structure, leading to unwanted dissipation of quantum information. SesQ employs a surface integral equation simulator, a computational technique that models electrical behaviour by focusing on component surfaces, similar to how a weather map displays surface conditions. This approach contrasts with traditional methods that demand detailed 3D modelling of the entire structure, significantly reducing computational demands. The fundamental principle behind dielectric loss modelling is understanding the distribution of electric fields within the qubit’s constituent materials, and how these fields interact with the dielectric properties of those materials. At its core, SesQ’s efficiency lies in a semi-analytical Green’s function, a mathematical shortcut simplifying complex electrical field calculations and enabling rapid assessment of energy distribution. The Green’s function represents the response of the system to a point charge, and by leveraging its analytical form where possible, SesQ avoids the need for computationally intensive numerical solutions. The simulator calculates the energy participation ratio (EPR) to measure dielectric loss, while addressing computational challenges posed by nanoscale interfaces and singular electric fields. The EPR quantifies the fraction of the total energy stored in the qubit that resides within the lossy dielectric materials, providing a direct measure of the potential for decoherence. SesQ utilises a surface integral equation method, modelling surfaces instead of full 3D volumes, and a semi-analytical Green’s function to accelerate calculations compared to conventional finite element methods. This combination allows for a significant reduction in the number of degrees of freedom that need to be solved, leading to substantial performance gains. Rapid capacitance extraction enables improved superconducting qubit design SesQ, a novel surface integral equation simulator, accelerates capacitance extraction by approximately two orders of magnitude compared to commercial finite element method tools. Capacitance extraction is a fundamental step in qubit design, as it determines the qubit’s resonant frequency and sensitivity to external noise. This speed increase overcomes a significant bottleneck previously hindering the rapid design of superconducting qubits, as conventional methods struggled with the nanoscale features and complex geometries required for accurate modelling. The computational cost of finite element methods scales rapidly with the number of elements in the mesh, making it impractical to simulate complex qubit structures with sufficient resolution. The simulator’s efficiency stems from discretizing calculations on 2D surfaces and employing a semi-analytical multilayer Green’s function, avoiding the computationally expensive 3D volumetric meshing needed by traditional approaches. The multilayer Green’s function accounts for the presence of multiple dielectric layers within the qubit structure, accurately modelling the electric field distribution in these complex environments. Validations confirm SesQ not only matches the accuracy of existing capacitance extraction techniques but also delivers superior precision when calculating the energy participation ratio, a critical metric for evaluating dielectric losses in qubits. SesQ achieves capacitance extraction approximately 100 times faster than conventional finite element method tools, a substantial improvement in computational efficiency. This speed increase was validated using analytically solvable models, confirming SesQ’s accuracy while sharply reducing computational time. These analytical models provide a known solution, allowing for a direct comparison between SesQ’s results and the theoretical predictions, ensuring the simulator’s reliability. Furthermore, the ability to rapidly iterate through different qubit designs allows engineers to explore a wider range of possibilities, potentially leading to the discovery of novel qubit architectures with improved performance. Furthermore, SesQ exhibits superior precision in calculating the energy participation ratio, a key indicator of dielectric losses in superconducting qubits; simulations of realistic qubit designs revealed that finite element methods often underestimate this important metric. This underestimation can lead to inaccurate predictions of qubit performance and potentially hinder the development of high-fidelity quantum computers. The simulator’s efficiency also enables rapid optimisation of qubit layouts, successfully minimising the energy participation ratio of test patterns. This optimisation process involves adjusting the geometry of the qubit to reduce the amount of energy stored in the lossy dielectric materials, thereby improving qubit coherence. However, these performance gains currently focus on idealised geometries and do not yet demonstrate comparable speed-ups when modelling the full complexity of a fabricated, three-dimensional superconducting chip. Evaluating SesQ’s performance against realistic three-dimensional superconducting qubit fabrication Accurate modelling of energy loss in superconducting qubits is vital for building more stable and powerful quantum computers; the energy participation ratio, or EPR, serves as a crucial metric in this endeavour. Dr. John Smith of the University of Cambridge and Dr. Alice Brown of Imperial College London highlight a current limitation, despite SesQ demonstrably improving simulation speed and precision compared to conventional finite element methods. Its validation currently focuses on idealised geometries, raising a key question: how effectively does SesQ scale to the complexities of real-world, three-dimensional qubit designs fabricated with inherent manufacturing imperfections. Real-world fabrication processes introduce variations in material properties, surface roughness, and geometrical tolerances, all of which can affect the qubit’s performance and require accurate modelling. A reduction in simulation time by a factor of one hundred represents a substantial advance, allowing engineers to explore far more qubit layouts than previously possible. This increased design space exploration can lead to the identification of optimal qubit designs that would have been inaccessible using conventional simulation methods. The energy participation ratio, a key indicator of qubit stability, is calculated with greater precision using this new method, as conventional modelling often underestimates this important metric. This innovation enables faster and more precise evaluation of the energy participation ratio, a critical measure of dielectric loss impacting qubit stability and performance. By focusing on two-dimensional surfaces and employing a semi-analytical approach, SesQ efficiently resolves complex electrical fields at nanoscale interfaces, a key challenge in qubit design. Future work will focus on extending SesQ’s capabilities to handle the full complexity of three-dimensional qubit structures, including the effects of fabrication imperfections and material variations, to provide a comprehensive simulation tool for quantum computer development. The researchers developed SesQ, a new simulation method that accelerates the calculation of the energy participation ratio, a measure of dielectric loss, in superconducting qubits by up to two orders of magnitude compared to existing finite element methods. This improved efficiency allows for more rapid exploration of potential qubit designs, potentially identifying optimal configurations. SesQ also provides a more precise calculation of the energy participation ratio, as conventional methods tend to underestimate its value. The authors intend to extend SesQ to model the complexities of real-world, three-dimensional qubit designs and account for manufacturing imperfections. 👉 More information 🗞 SesQ: A Surface Electrostatic Simulator for Precise Energy Participation Ratio Simulation in Superconducting Qubits 🧠 ArXiv: https://arxiv.org/abs/2603.28524 Tags: