NSF Award
2433526
Silicon quantum dot devices hold significant promise for scalable quantum computing. However, tuning these devices into the desirable states for quantum applications is highly challenging, creating substantial barriers to entry. Traditionally, tuning has been a manual process that is time-consuming, heavily reliant on experimental intuition, and inherently unscalable. This situation underscores the need for automated tuning (autotuning) approaches. The development of autotuning algorithms has been impeded by the lack of experimental training data and the limitations of existing quantum dot simulators, which only capture the physics of already-tuned devices. To this end, this project aims to provide full-stack support for quantum dot device autotuning research by delivering new quantum dot device simulation infrastructures for cold start and exploring corresponding autotuning algorithms. This initiative will democratize autotuning research, offering researchers without access to experimental facilities both training data and a low-cost autotuning testbench. These advancements will promote the progress of science by facilitating broader access to quantum computing research and enhancing the efficiency and scalability of quantum dot device tuning. This project will provide training opportunities for the next-generation quantum computing workforce, and the research outcomes will be integrated into undergraduate and graduate education efforts.<br/><br/>The proposed research will significantly advance our understanding of quantum device modeling and tuning, providing innovative tools, data, and methods that can shape the tuning process of quantum dot devices. Specifically, this project will develop the QDREAM (Quantum Dot Real-Time Emulation and Autotuning Model) framework. QDREAM consists of 1) device-physics-based cold start simulations that focus on combining a finite element electrostatic simulation with a constant interaction quantum dot model to simulate devices in a completely untuned regime; 2) an FPGA-based quantum dot device emulator that will take in real voltages and output a charge sensor signal in real-time; and 3) a series of autotuning algorithms targeting various stages of the device tune-up process from cold start. QDREAM will be validated using real quadruple quantum dot devices routinely fabricated and measured in our lab. These comprehensive advancements will serve as a foundational step towards realizing larger-scale, more advanced quantum-dot-based quantum computers.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
