This PhD thesis reports on the development of a wide variety of theoretical and experimental capabilities in the field of quantum information processing using trapped ions. The presentation is organized in two distinct parts. The first part presents a series of published articles containing theoretical results, and their experimental validation, advancing a range of tools and techniques in error-resilient quantum control and robust verification procedures. One of the fundamental challenges in realizing a general purpose quantum computer is the management of error in fragile quantum systems. A familiar aspect of this challenge is decoherence, a process by which the system looses its stored quantum information, rendering it useless for subsequent computation. This collection of works focuses on reducing sources of quantum decoherence and precisely quantifying residual errors in real quantum systems dominated by realistic temporally-correlated noise processes, of central importance in any future architecture for quantum computing. The second part focuses on the design, construction and characterization of a complex experimental system capable of supporting a broad calls of quantum simulation experiments with hundreds of spin qubits using 9Be+ ions in a Penning trap. The trap design is optimized to suppress non-harmonic trapping potentials while providing wide optical access for engineering enhanced spin-motion coupling across large ion crystals. A superconducting magnet, equipped with a fully closed-cycle dual LN2/LHe recondensor, provides the magnetic field for trapping and produces the 55-GHz Zeeman-split qubit states suitable for microwave control. A custom UV laser system for driving the relevant transitions in neutral and ionized beryllium, as well as UV-compatible fibers and in-bore optomechanics for delivering microwave and laser beams to the trap, is presented. The thesis concludes with a brief outlook towards future theoretical and experimental studies, and extensions of the experimental setup
This PhD thesis reports on the development of a wide variety of theoretical and experimental capabilities in the field of quantum information processing using trapped ions. The presentation is organized in two distinct parts. The first part presents a series of published articles containing theoretical results, and their experimental validation, advancing a range of tools and techniques in error-resilient quantum control and robust verification procedures. One of the fundamental challenges in realizing a general purpose quantum computer is the management of error in fragile quantum systems. A familiar aspect of this challenge is decoherence, a process by which the system looses its stored quantum information, rendering it useless for subsequent computation. This collection of works focuses on reducing sources of quantum decoherence and precisely quantifying residual errors in real quantum systems dominated by realistic temporally-correlated noise processes, of central importance in any future architecture for quantum computing. The second part focuses on the design, construction and characterization of a complex experimental system capable of supporting a broad calls of quantum simulation experiments with hundreds of spin qubits using 9Be+ ions in a Penning trap. The trap design is optimized to suppress non-harmonic trapping potentials while providing wide optical access for engineering enhanced spin-motion coupling across large ion crystals. A superconducting magnet, equipped with a fully closed-cycle dual LN2/LHe recondensor, provides the magnetic field for trapping and produces the 55-GHz Zeeman-split qubit states suitable for microwave control. A custom UV laser system for driving the relevant transitions in neutral and ionized beryllium, as well as UV-compatible fibers and in-bore optomechanics for delivering microwave and laser beams to the trap, is presented. The thesis concludes with a brief outlook towards future theoretical and experimental studies, and extensions of the experimental setup
In the search for new computational paradigms past the impending end of Moore’s law quantum information processing based on cold trapped ions has emerged as a promising candidate. This thesis presents on the design, construction and testing of a setup for analog quantum simulation experiments using Be ion crystals in a Penning trap. At the heart of the experiment is a specialized Penning trap with wide optical access for engineering Ising-type interactions. The trap sits inside a superconducting magnet which runs continuously with minimal losses through a novel dual-stage reliquefier system. During this thesis, mechanical and optical subsystems for this ion trap quantum simulator were designed and built. Two laser systems based on nonlinear frequency conversion were set up to generate light for photoionization, Doppler cooling, state preparation, readout, and Raman interactions. In parallel, a method was devised to produce ultraviolet-solarization resistant fiber patch cords for pointing-stable light delivery. Inbore optomechanics and a collaboratively designed precision six-axis positioning device were constructed for beam delivery and trap alignment, respectively. These optomechanics co-locate custom-designed millimeterwave delivery systems as well as numerically optimized, diffraction-limited imaging optics for fluorescence readout. Using these subsystems, we validate interferometrically the motional stability of the setup from the mHz to kHz region during pulse-tube operation. First-ever light-ion interaction serves to validate calculations of interaction frequencies, optomechanics operation, and imaging performance, while coherent qubit interaction allows for this to be done on the millimeter wave setup, where initial Rabi flopping times of 18 us were observed. Site-resolved imaging of rapidly rotating ion crystals demonstrates critical phase-locked operation of the apparatus; positioning the experiment for future upgrades towards a full analog quantum simulator.
Quantum computation is a promising technology owing to the unique physical nature of quantum systems. Quantum computers have the potential to be as much as exponentially more efficient than today’s most powerful classical computer at certain tasks. One of the major challenges the scientific community must overcome, arises due to the intrinsic tendency of quantum information to be highly susceptible to environmental noise. This thesis presents work in developing and characterizing methods for precision quantum control in the presence of decoherence inducing noise, using trapped 171Yb+ ions as a model experimental platform. A flexible , robust microwave system is used to access the 12.6 GHz hyperfine qubit manifold in the outer S-shell orbital. The low phase noise characteristics of a microwave oscillator allow free-evolution coherence times in excess of three seconds, and operational fidelities F > 99.99%, characterized by randomized benchmarking. Starting from this relatively clean baseline, control techniques may be implemented in order to simulate a variety of realistic, time-dependent noise models, via arbitrary IQ modulation of the microwave source. A formalism for calculating the spectral noise filter functions of piece-wise constant control sequences is developed. The efficacy of this formalism is demonstrated by experimentally characterizing a variety of control protocols in the presence of engineered noise.
Quantum information science promises a host of new and useful applications in communication, simulation, and computational algorithms. Trapped atomic ions are one of the leading physical systems with potential to implement a large-scale quantum information system, but many challenges still remain. This thesis describes some experimental approaches to address several such challenges broadly organized under three themes: gates, sensors, and systems. Quantum logic gates are the fundamental building blocks for quantum algorithms. Although they have been demonstrated with trapped ions previously, scalability requires miniaturizing ion traps by using a surface-electrode geometry. Using a single ion in a surface-electrode trap, we perform a two-qubit entangling gate and fully characterize it via quantum process tomography, as an initial validation of surface-electrode ion traps for quantum information processing. Good logic gates are often good sensors for fast fluctuations and energy changes in their environment. Trapped ions are sensitive to fluctuating and static charges, leading to motional state decoherence (heating) and instabilities, problems exacerbated by the surface-electrode geometry. We investigate the material dependence of heating, specifically with aluminum and superconducting traps, to elucidate the physical origin of these fluctuating charges. Static charging is hypothesized to be caused by the trapping and cooling lasers due to the photoelectric effect. We perform systematic experiments with aluminum, gold, and copper traps with lasers at various wavelengths to validate this hypothesis. Realizing quantum processors at the system level requires models and tools for predicting system performance, demonstration of good classical and quantum control, and techniques for integrating different quantum systems. We develop a modeling system for trapped ion quantum computing experiments and simulate the effect of physical and technical noise sources on practical realizations of quantum algorithms in a trapped ion system. We experimentally demonstrate several such algorithms, including the quantum Fourier transform, order-finding, and Shor's algorithm on up to 5 ions. These experiments highlight several unique advantages of ion trap systems and help identify needs for further development. Finally, we explore the integration of ion traps with optical elements including mirrors and photon detectors as key elements in creating future hybrid quantum systems.
Abstract A major obstacle in the way of practical quantum computing is achieving scalable and robust high-fidelity entangling gates. To this end, quantum control has become an essential tool, as it can make the entangling interaction resilient to sources of noise. Nevertheless, it may be difficult to identify an appropriate quantum control technique for a particular need given the breadth of work pertaining to robust entanglement. To this end, we attempt to consolidate the literature by providing a non-exhaustive summary and critical analysis. The quantum control methods are separated into two categories: schemes which extend the robustness to (i) spin or (ii) motional decoherence. We choose to focus on extensions of the σ x ⊗ σ x Mølmer–Sørensen interaction using microwaves and a static magnetic field gradient. Nevertheless, some of the techniques discussed here can be relevant to other trapped ion architectures or physical qubit implementations. Finally, we experimentally realize a proof-of-concept interaction with simultaneous robustness to spin and motional decoherence by combining several quantum control methods presented in this manuscript.
This thesis describes experimental work towards developing a trapped ion quantum information processor. An existing ion trap apparatus was capable of trapping and laser-cooling single ions or small ion strings of 40 Ca+, and had been used for studies of quantum jumps and natural lifetime measurements in Ca. This thesis describes improvements in this apparatus, which have allowed the stability and the flexibility of experimental control of the ions to be greatly increased. This enabled experiments to read out the spin state of a single trapped ion, and to load ions with isotope selectivity through photoionization. The optical systems were improved by installation of new lasers, optical reference cavities, and a system of acousto-optic modulators for laser intensity switching and frequency control. The photon counting for fluorescence detection was improved, and a new photon time-of-arrival correlation circuit developed. This has permitted rapid and more sensitive detection of micromotion, and hence cancellation of stray fields in the trap. A study of resonant circuits in the low RF, high voltage (10 MHz, 1 kV) regime was carried out with a view to developing a new RF supply for the Paul trap with reduced noise and increased power. A new supply based on a helical resonator was built and used to trap ions. This technique has reduced noise and will permit higher secular frequencies to be attained in the future. A magnetic field B in the ion trap is used to define a quantization axis, and in one series of experiments was required to be of order 100 G to provide a substantial Zeeman splitting. A set of magnetic field coils to control the size and direction of B is described. The design of these posed some problems owing to an unforseen issue with the vacuum chamber. In short, it is magnetizable and acts to first approximation like a magnetic shield. The field coils had to be sufficiently substantial to produce the desired field at the ion even in the presence of this shielding effect, and dark resonance (and other) spectra with Zeeman splitting were obtained to calibrate the field using the ion as a probe. Finally, the thesis describes the successful loading of the ion trap by laser photoionization from a weak atomic beam. This involved two new lasers at 423 nm and 389 nm. Saturated absorption spectroscopy of neutral calcium is first described, then transverse excitation of an atomic beam in our vacuum chamber is used to identify all the main isotopes of calcium and confirm their abundances in our source (a heated sample of natural calcium). Finally, photoionization is used to load the trap. This has three advantages over electron-impact ionization. By avoiding an electron gun, we avoid charging of insulating patches and subsequent electric field drift as they discharge; the flux in the atomic beam and hence calcium (and other) deposits on the electrodes can be greatly reduced; and most importantly, the photoionization is isotope selective. Evidence is presented which suggests that even with an non-enriched source, the rare isotope 43 Ca can be loaded with reasonable efficiency. This isotope is advantageous for quantum information experiments for several reasons, but chiefly because its ground state hyperfine structure can act as a stable qubit.
Quantum computers hold the potential of solving problems that are intractable using classical means. Yet, to date no quantum computer has actually solved a problem of general interest that could not have been solved otherwise. Bridging this gap from promised theoretical usefulness to practical relevance is an interdisciplinary challenge, requiring experimental progress, solutions of classical software challenges as well as advances in quantum algorithm design. It is these final two points in which this thesis aims to contribute. One of the most promising quantum computing architectures is based on trapping ions in surface-electrode ion-traps. In order to productively operate such trapped-ion quantum computers, we develop and benchmark a compiler that optimizes ion shuttling sequences in order to efficiently map abstract quantum circuits onto a physical trapped-ion chip. This is a necessary component of the infrastructure of a quantum computing setup, however it is alone not sufficient to execute arbitrary quantum algorithms. In particular, a very promising class of quantum algorithms are so-called hybrid quantum-classical algorithms which leverage the computational resources of both quantum and classical computers. To effectively implement such algorithms, a suitable software setup is necessary to allow for tight interplay between the two types of computers without wasting precious coherence time. To solve this challenge, we propose, implement and benchmark our setup for Interpreted Quantum-Classical Computations (IQCCs), demonstrating its feasibility for trapped-ion quantum computers. For the second part of this thesis we focus on the development of quantum algorithms that make use of this setup. As a first application, we investigate quantum simulations of gauge theories. Such theories are prevalent throughout modern physics and are characterized by local constraints that the dynamics are subject to. Errors in noisy simulations of such theories results in these constraints being violated and the simulations becoming unphysical. We show that this problem can be remedied by measuring gauge operators, decoding possible errors on a classical computer and actively correcting them, which reduces simulation errors. Additionally we interpret the scheme as a sympathetic cooling setup, which allows thermal control of the simulation. We finally turn to the task of preparing ground states of frustration-free Hamiltonians in quantum computers. Studying ground states of various systems is of paramount importance for solving optimization problems and understanding phenomena such as magnetism, superconductivity and superfluidity. We frame the ground-state preparation problem as an error correction problem, wherein energy is measured locally, measurements are decoded and corrections are applied to reduce the energy. In physically relevant examples we show up to exponentially improved runtimes over previous, similar algorithms and give a general run-time bound in terms of the Hamiltonian gap.