Developing a robust quantum simulator with trapped ions

Abstract

Trapped ion systems have experienced significant growth in recent years as their potential for excelling as quantum simulators has become recognized. Ions make an excellent qubit due to their long coherence times with moderate gate times, high fidelity detection and state initialization, and their ability to create long range spin interactions. As the experimental demands of trapped ions increases, so too do the demands that sustain and control them. In this thesis, I will cover the design and implementation of robust systems for the trapped ions platform and describe the development of robust lab infrastructure, equipment, and optics required to perform high contrast entangling operations on an existing four-rod system. By redesigning the DC power distribution and grounding system I have been able to supply our quantum simulator with clean DC voltages while reducing ground loops that can introduce noise into the system. With delicate alignment of the 355 nm system, our four-rod system is able to entangle qubits together. By incorporating 3D printing and inexpensive DC motors, I was able to motorize the controls used to align our 355 nm beam in the vertical direction which has made alignment reliable and accessible. With future iterations of the motorized stage, I’ve been able to achieve a resolution of 70 nm in all three axes. We then look to the next generation ion trap in the form of a meticulously engineered blade trap. With the ability to perform simulations on systems of about 30 qubits, reaching very low vacuum pressures is essential to increase ion life times. By careful preparation of our Shapal blade holder I’ve been able to preserve the 9 × 10−13 mbar pressure of our blade trap vacuum chamber. I then discuss the design of the imaging system for the blade trap which utilizes dual 0.5 NA (numerical aperture) objectives to achieve high state detection fidelity (∼ 99.9%). By simulating the imaging system and taking into consideration the effects of the systems’ efficiencies, I find that high state detection fidelity should be achievable for detection times on the order of 20 µs. This offers potential for performing in-situ mid-circuit measurement on the blade trap system. By performing some initial tests, I compare the experimental results to the simulated performance of the imaging system and find they match reasonably well.