Engineering Highly Coherent Qubits by Correlating Surface Spectroscopy with Quantum Measurement

Abstract

Quantum systems can be harnessed to build powerful computers, nanoscale sensors, and provably secure communication. However, in essentially every solid-state implementation they are limited by noise and loss arising from poor control over materials. In this dissertation, we look at two quantum systems – nitrogen-vacancy (NV) centers in diamond, and superconducting qubits. For both systems, we apply a common approach: identify and characterize material properties using surface spectroscopy, and then correlate these properties with quantum measurement. NV centers close to the surface can be used as nanoscale sensors, but surface-induced decoherence limits their utility. To address this, we study the contribution of a 2D bath of unpaired electrons at the surface to the NV coherence decay, and find they are not currently a major source of shallow NV decoherence. These measurements also allow us to probe the dynamics and dimensionality of the spin bath, and ultimately we propose that these spins reconfigure or ‘hop’ between measurements. Next, we address additional challenges associated with using NV centers as nanoscale biosensors. To realize these experiments, a highly coherent NV center must be within a few nanometers of a biomolecule immobilized at the surface. Directly functionalizing the surface with solution-phase chemistry is desirable because it minimizes the distance between sensor and target, and is non damaging to the surface. We deploy modern wet chemistry to functionalize diamond surfaces and use lab-based and synchrotron-based spectroscopy to establish a reaction discovery pipeline. We demonstrate NV centers with coherence properties comparable to the state of the art under our surfaces and use these NV centers to detect the magnetic signal from functional groups. Finally, we apply this methodology to improve superconducting qubits, a leading quantum computing platform in industry. These devices suffer from loss originating at surfaces and interfaces. In our work, we report two improvements to the material platform: by (1) using tantalum as the superconducting material, and (2) purifying surfaces and interfaces, we report transmons with state-of-the-art coherence times, exceeding 0.3 ms.