Over the past hundred years, information technology has revolutionized science, industry, and society. The burgeoning field of quantum information technology promises another set of revolutions for the twenty-first century: guaranteed-secure communications can protect us from data theft and cyber-espionage, efficient simulation of physical and chemical systems would transform development of novel drugs and next-generation materials, and implementation of digital quantum algorithms could provide exponential speed-ups for some important general-purpose computational tasks.

All of these goals require quantum networks in which individual quantum bits communicate. My research is designed to address this challenge by using single particles of light (photons) to link qubits by engineering strong interactions between photons and color centers in diamond. These color centers are atom-scale defects that have spatially localized transitions between discrete electronic levels, producing absorption and therefore color in an otherwise transparent crystal.

My research with Professor Mikhail Lukin in the Harvard University Department of Physics focuses on the Silicon-Vacancy (SiV) color center in diamond, a promising optically-addressible solid-state qubit. We use nanophotonics to increase the interaction between photons and the SiV. We first fabricate nanoscale optical waveguides out of the diamond crystal, effectively focusing light very tightly around our color centers, which increases the coupling strength to the focused photons. Finally, by adding nanoscale mirrors to both ends of the waveguide, we can create a nanophotonic crystal cavity. In such a cavity, the photons are effectively trapped between the high-reflectivity nanoscale mirrors, giving rise to many roundtrips inside the cavity before the photon escapes. Just as you can see many copies of your own reflection when you stand between two mirrors, the incoming photon can “see” many copies of the color center, giving it an increased interaction probability and hence stronger coupling. These techniques increase the interaction between photons and color centers in diamond by orders of magnitude.

We have now reached the regime where a single SiV in a diamond nanophotonic crystal cavity can produce almost complete switching of an input field depending on the internal spin degree of freedom of the SiV. Moreover, by positioning multiple SiV centers in a single device, we can create entanglement or cavity-mediated interactions between two SiV centers. These results highlight the promise of this system as a scalable, solid state platform for optical quantum information science.

There are many candidates for optically-addressible qubits in the solid state, including self-assembled quantum dots, rare earth ions, and color centers in diamond and related materials. In diamond, attention has previously focused on the nitrogen-vacancy (NV) color center, primarily due to its spin degree of freedom that is optically addressible and long-lived even at room temperature. However, the optical properties of the NV are extremely unfavorable for engineering strong coupling between the NV and light. Hence, recent interest in our lab has turned towards exploring alternative color centers for solid-state quantum optics.

Recently, the silicon-vacancy (SiV) center has attracted interest for its strong optical transition and narrow optical linewidth. However, fairly litte was known about the SiV center other than its basic electronic structure. Work in our lab focuses on determining the important properties of the SiV center for applications to quantum optics. For example, we demonstrated that the SiV can have extremely narrow optical transitions with a narrow inhomogeneous distribution even inside practical nanophotonic devices. We also explored the spin degree of freedom of the SiV and showed that under appropriate conditions, the SiV can be used as a long-lived qubit.

Along the way, we discoveredd that the key properties of the SiV arose from its inversion symmetry. Based on this intuition, we explored a related color center in diamond, the germanium-vacancy (GeV) center which has the same symmetry. In some ways, the optical properties of the GeV are even more favorable for applications than those of the SiV.

Color centers in diamond are small, usually several-atom defects that have a reasonably strong optical transition, causing normally transparent diamond to absorb light and become colored. These color centers can be thought of as individual molecules "trapped" inside the diamond lattice, which (because diamond has a large bandgap) acts almost like a vacuum and often has minimal effects on many of the properties of the color center.

However, the diamond lattice is not perfect, and color centers are often necessarily close to the diamond surface, either for near-surface sensing or in nanostructures. As a result, these color centers in diamond can interact with nearby defects or surfaces, leading to time-varying shifts in the energy of the color centers' electronic transitions; the Nitrogen-Vacancy center is especially susceptible to this problem, because it has a large ground-state dipole moment. Defects with nonzero spin can also couple to any available spin degrees of freedom, which can cause decoherence and limit the usefulness of color centers as sensors and quantum memories.

To solve these problems, we have developed a complex process involving high temperature vacuum annealing and oxygen surface termination. The resulting surfaces are characterized with electron microscopy and various x-ray spectroscopies. This treatment results in charge-stable color centers with low spectral diffusion, even inside nanostructures.

Since Bell’s original paper in 1964, a wide variety of experimental tests have overwhelmingly supported the completeness of quantum mechanics over local hidden-variable theories. However, relatively little effort has focused on systems of spins larger than 1/2; generalizing Bell’s result to higher dimensions is difficult, and the experiments needed to test these high-spin Bell inequalities are exacting. New advances in high effciency photon-number-resolving detectors suggest that experimental tests of these inequalities should be possible in the Schwinger representation, using the continuous-variable entangled (two-mode squeezed) fields produced by an optical parametric oscillator below threshold.

In this paper, we explore the realistic experimental implementation of this proposal to violate Mermin’s high-spin inequalities. We demonstrate that violation for spin values greater than 1 should be attainable under a range of feasible experimental conditions that include finite squeezing and nonideal detection effciency.

This work was performed in my third and fourth year as an undergraduate at the University of Virginia with Professor Olivier Pfister.

Previous work has shown that difluoroboron dibenzoylmethane-polylactide (BF2dbmPLA) shows intense phosphorescence and a rare oxygen-sensitive room-temperature phosphorescence. Although this material has many good qualities as a potential imaging agent for a variety of purposes, improvements can still be made to increase the dye’s effectiveness and range of applicability. Fortunately, the modular nature of the diketone synthesis permits facile variation of the substituents on the diketone core.

Building on our previous understanding of this system, I synthesized new BF2dbm-based dyes with a variety of substituents ranging from heavy atoms (which increase phosphorescence intensity through spin-orbit coupling) to large aromatic spacers (which increase donor-acceptor character) and beyond. These effects were characterized with a variety of optical spectroscopies including time- and frequency-resolved absorption, fluorescence, and phosphorescence, as well as typical chemical characterization techniques like UV/VIS and NMR.

This research was performed in my first through third years as an undergraduate with Professor Cassandra Fraser in the University of Virginia Department of Chemistry.

Even small changes in the structure of a molecule can lead to drastically different optical and electronic properties. Unfortunately, exactly solving the Schrödinger Equation for even simple systems—much less molecules containing hundreds of electrons—is impossible. In the past, chemists and physicists usually had to resort to direct synthesis of a compound of interest to gain information about its properties.

In the last few decades, however, computational techniques to gain accurate approximate models of molecular electronic energies have become widespread. Using techniques from this field of computational chemistry, I led a research effort to understand our small fluorescent molecules from a fundamental physical viewpoint and even predict the properties of new systems before synthesis.

For example, some of our dyes show intense solvatochromism while others emit at more or less the same wavelength regardless of solvent. This reason for this was not well understood. However, early studies I’ve performed indicate that the parent dye, dibenzoylmethane (with only a phenolic aromatic group) shows very little donor acceptor character; the electron density distribution appears similar in the HOMO and the LUMO. However, as the aromatic ring size increases, the HOMO becomes centered more and more on the aromatic substituent while the LUMO remains centered on the diketone core.

Although quantitatively predicting emission wavelengths in complicated polymeric systems is difficult, these and other model studies helped to point us in the right direction guided us in developing new analogues with predictable properties.

This research was performed in my second and third years as an undergraduate with Professor Cassandra Fraser in the University of Virginia Department of Chemistry, with advice from Professor Carl Trindle.