Chip-Scale Self-Referenced Optical Frequency Comb Sources

ended

As a Nobel-honored technology, optical frequency combs, which consist of equidistant spectral lines, have revolutionized applications in time-keeping, and metrology as they offer unprecedented precision in frequency via self-referencing. However, conventional frequency comb systems have been confined to laboratories due to the cost, size, and power requirements of their components. This project aims to develop a chip-scale optical frequency comb source that can be self-referenced. Key components to realize self-referencing are comb generators and frequency doublers. However, it is challenging to integrate both functionalities on the same chip as they typically rely on different nonlinear processes and thus different material platforms. Another major challenge in the system miniaturization is how to achieve ultra-efficient comb generation and frequency doubling to enable on-chip comb pumping and self-referencing beat note detection, respectively. In this project, we will circumvent the multi-material issue by developing both comb generator and frequency doubler based on the same nonlinear material: aluminum gallium arsenide (AlGaAs). This material exhibits both strong cubic and quadratic nonlinearities which can be utilized for comb generation and frequency doubling, respectively. Ultra-efficient comb generation will be realized by developing ultra-high-quality-factor microresonators and employing new comb generation methods combining cubic and quadratic nonlinearities while highly-efficient frequency doubling will be achieved by adaptively-controlling the phase-matching condition. We will also develop heterogeneous integration technologies to bridge the nonlinear devices with on-chip laser sources and detectors by using intermediate silicon nitride circuits. Successful miniaturization of a self-referenced frequency comb source will enable applications like LIDAR, coherent communications, chemical sensing, medical imaging, and precision metrology.