While it is often thought that the geometric phase is less sensitive to fluctuations in the control fields, a very general feature of adiabatic Hamiltonians is the unavoidable dynamic phase that accompanies the geometric phase. The effect of control field noise during adiabatic geometric quantum gate operations has not been probed experimentally, especially in the canonical spin qubit system that is of interest for quantum information. We present measurement of the Berry phase and carry out adiabatic geometric phase gate in a single solid-state spin qubit associated with the nitrogen-vacancy center in diamond. We manipulate the spin qubit geometrically by careful application of microwave radiation that creates an effective rotating magnetic field, and observe the resulting Berry phase signal via spin echo interferometry. Our results show that control field noise at frequencies higher than the spin echo clock frequency causes decay of the quantum phase, and degrades the fidelity of the geometric phase gate to the classical threshold after a few (∼10) operations. This occurs inspite of the geometric nature of the state preparation, due to unavoidable dynamic contributions. We have carried out systematic analysis and numerical simulations to study the effects of the control field noise and imperfect driving waveforms on the quantum phase gate.

Magneto-optic Kerr effect (MOKE) microscopy is a widely used technique for observation and characterization of microscopic magnetic structures. While being efficient and easy-to-use, current commercial MOKE microscopes are not superb in time resolution, limited by the frame rate of the camera. Here, we introduce a revolutionary sensor, namely, the event camera, as a convenient add-on to traditional MOKE microscopy and explore the potential applications of event-based vision in research areas using MOKE microscopy. We use the frame stacking method to improve visibility to human eyes in generated slow motion videos. We perform a proof-of-principle feedback control experiment using the event-based vision data and characterize the overall latency of the feedback loop as short as 25 ms with our current prototype. Finally, we discuss the limitations of current event cameras in MOKE microscopy as well.

No abstract