Integrated Diamond Nonlinear Optics

Abstract

The past decade has seen an explosion of development in new devices, modalities, and architectures for advanced sensing, computation, and communication. This growth has been driven by the massive investments of large consumer-tech companies in data-hungry domains of customer interest. As these companies develop their products and seek a competitive edge, technologies which can provide advantage across application areas by making computation cheaper, or sensors more knowledgeable, or communication faster, are attracting more interest in and outside academia. Photonics, or the study of how to manipulate light, is a broad-based platform that inherently delivers on these promises. Although photonics has been a part of daily life since the invention of the laser, modern needs and aspirational designs which meet those needs far exceed the capabilities of the platforms which have traditionally constituted the photonics toolkit. There are two specific areas where standard photonic materials are lacking - support for visible wavelengths and functionality for quantum optics. Currently, there are many implementations of systems which use visible light or quantum optics on the tabletop, but the drive to scale these technologies and deliver them to the consumer encounters a roadblock in their size and cost. Visible light sensors, for instance, could greatly benefit from conversion into an integrated photonics platform, where the optical components are printed onto a small substrate. However, the standard materials used for such systems (silicon and indium phosphide) absorb visible light, and other materials (like silicon nitride) do not have the attributes required for advanced functionality. More compellingly, there exist very few material platforms which can host the quantum defects required for next-generation technologies like quantum computation or secure communication. The past decade has seen extensive research into an emerging material platform which can deliver on the ability to bring integrated visible photonics and quantum optics to the same chip - diamond. With a broad transparency region (from UV to the far infrared) and high refractive index, diamond makes an excellent material for run-of-the-mill photonic devices. More importantly, it harbors atom-like defects in its crystal lattice - the nitrogen-vacancy, silicon-vacancy, and germanium-vacancy centers, among many more - which provide a direct interface between the photons circulating within the diamond device and the quantum world. Taken together, these simple properties of diamond constitute a powerful approach towards implementation of advanced quantum technologies, such as repeaters for secure quantum communication or even, one day, computers. Before that future arrives, diamond's capability must first be developed. Diamond has only been available in significant quantities for research use for about a decade, and as such, the processing technologies which are taken for granted in silicon or silicon nitride-based work must be redeveloped. Within the context of photonic devices, we discuss different fabrication strategies and their implications for device design. For visible light devices especially, we investigate methods to reduce sidewall roughness for etched structures. Interfacing with diamond photonic structures also brings its own unique set of challenges due to its high refractive index and very small material footprint. To this end, we develop a method termed "loaded tapered-fiber coupling" to optically access large, free-standing diamond devices. For size-limited diamond resonators and waveguides, we also develop auxiliary optical waveguides to act as spot-converters for visible wavelengths, an improvement on previous techniques for telecom-range wavelengths. Finally, by synthesizing these improvements in fabrication, we embark on a set of experiments in diamond microresonators. Nonlinear optics is an excellent testbed for such improvements, both for diamond's superlative properties in that regard and their high sensitivity to improvements in fabrication. First, we look at a process known as Raman lasing by constructing long path-length resonators.This technique involves the shifting of an input photon by 40 THz via the interaction with the diamond's crystal lattice.The first demonstration in a diamond integrated device was shown with a pump wavelength of 1600 nm, with an output that was tuned over 100 nm and a low threshold. Leveraging the developments in visible photonics for diamond, a second Raman laser was demonstrated, this time at near-visible (720 nm) wavelengths.This too had a Stokes output which was tunable to over 100 nm and a very low threshold. Raman-specific effects such as polarization conversion were also investigated. A cousin to the Raman effect, the Kerr nonlinearity plays a key role in so-called microresonator (or Kerr) frequency combs. We survey the possibilities and difficulties in implementation for a diamond platform, highlighting the competition between the Kerr and Raman nonlinearities. Although the outlook for a technologically relevant Kerr comb in diamond is ultimately pessimistic, avenues for improvement are highlighted. Namely, we investigate via simulation a related class of effects, known as supercontinuum generation.Taking advantage of subtleties of the Raman process in diamond, we can find a preferred geometry which enables generation of a broadband light spectrum natively in diamond. Finally, we look at alternative applications of diamond, particularly in high-power systems. Due to its excellent thermal conductivity and power-handling capacity, optics fashioned purely out of diamond can show significant advantages compared to other materials. We end with a discussion of the design principles and results stemming from this work.