Photonic chip-based resonant supercontinuum via pulse-driven Kerr microresonator solitons

Supercontinuum generation and soliton microcomb formation both represent key techniques for the formation of coherent, ultrabroad optical frequency combs, enabling the RF-to-optical link. Coherent supercontinuum generation typically relies on ultrashort pulses with kilowatt peak power as a source, and so are often restricted to repetition rates less than 1 GHz. Soliton microcombs, conversely, have an optical conversion efficiency that is best at ultrahigh repetition rates such as 1 THz. Neither technique easily approaches the microwave domain, i.e., 10 s of GHz, while maintaining an ultrawide spectrum. Here, we bridge the efficiency gap between the two approaches in the form of resonant supercontinuum generation by driving a dispersion-engineered photonic-chip-based microresonator with picosecond pulses of the order of 1-W peak power. We generate a smooth 2200-line soliton-based comb at an electronically detectable 28 GHz repetition rate. Importantly, we observe that solitons exist in a weakly bound state with the input pulse where frequency noise transfer from the input pulses is suppressed even for offset frequencies 100 times lower than the linear cavity decay rate. This transfer can be reduced even further by driving the cavity asynchronously, ensuring the frequency comb stays coherent even for optical lines very far from the pump center. (C) 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

Automated summary

Supercontinuum generation and soliton microcomb formation are key techniques for coherent, ultrabroad optical frequency combs, but they differ in their efficiency and application in the microwave domain. Bridging the efficiency gap between the two approaches by driving a dispersion-engineered photonic-chip-based microresonator can achieve high-efficiency soliton microcomb generation and frequency conversion.