High-resolution broadband characterization of resonance dispersion in an optical microresonator

Abstract

Accurate knowledge of the uneven free spectral range of an optical microresonator, which provides direct insight into group velocity dispersion, is essential for understanding and controlling Kerr frequency comb dynamics. In this work, we present a simple and highly precise method formeasuring the free spectral range over a 5 THz bandwidth in silicon nitride microresonators, leveraging a wavemeter with 0.4 MHz resolution. Our fully fibered plug-and-play experimental setup enables the accurate extraction of resonance frequencies. By carefully analyzing the spectral position of each resonance, we measure both second- and third-order free spectral range expansion coefficients. This approach offers a robust and accessible tool for dispersion characterization in integrated photonic circuits, paving the way for next-generation of Kerr comb sources and quantum photonic technologies.

Figures (6)

• Figure 1: Experimental setup for dispersion characterization of the resonator. To determine the approximate position of the resonances, the laser and the diode are replaced by a superluminescent diode and an optical spectrum analyzer, respectively. PC: polarization controller, MR: microresonator, PD: photodiode, BS: beam-splitter
• Figure 2: a) TE (blue) and TM (orange) transmission spectrum of the MR using a superluminescent diode. b) Long term stability (blue line) when the pump laser frequency is set at a bottom of an arbitrary resonance.
• Figure 3: a) Measured double peak resonance as a function of the frequency. The top figure shows the experimental data (blue curve) and associated fit (red curve) using Equation \ref{['eq: resonance fit']}. The grey curve shows a fit using the simple resonance model, whereas the orange curves show the two compounds of the double resonance model. The bottom figure shows the residuals when one or two peaks are considered in the fit in gray and red, respectively. b) Measured single-peak resonance and fit associated using equation \ref{['eq: resonance fit']}. c) Resonance center (redshift) as a function of laser probe power. The linear dependence is mainly induced by Kerr and thermo-refractive effects.
• Figure 4: Extracted resonance centers using equation \ref{['eq: resonance fit']} for TE (blue) and TM (orange) modes. The circle (square) markers represent a simple (double) peak resonance.
• Figure 5: Experimental and fitted $D_{int}$ using equation \ref{['eq: dispersion microring']} for two arbitrary resonances in TE mode (blue) and TM mode (orange). The linear part, $D_1$ of the equation was subtracted. The circle (square) markers represent a simple (double) peak resonance.