top of page
Writer's pictureLatitude Design Systems

A Tutorial on Silicon Microring Resonators and Their Applications

Introduction

Over the past decade, silicon photonics has emerged as a leading and promising photonic integrated circuit (PIC) technology, enabled by CMOS electronic fabrication processes that allow for low-cost, high-volume, and large-scale integration of photonic circuitry. The development of silicon photonics has led to significant reductions in the footprints of passive components and wavelength-selective devices, with one of the best examples being the microring resonator.

Microring resonators can be simply imagined as an optical waveguide looped back on itself, such that resonance occurs when the optical path length of the resonator waveguide is an integral multiple of the wavelength [2]. Due to the high refractive index contrast between silicon and its oxide, microring resonators can be fabricated with very small bend radii (around 5 μm) and achieve a large free spectral range (FSR) of over 20 nm at telecommunication wavelengths of 1550 nm, essential for integration into silicon photonic circuits.

As a key photonic component in silicon photonics, integrated optical microring resonators have been under intensive research for the past decade. They provide strong resonance field enhancement, compact footprints, and narrow-band wavelength selectivity, enabling a wide range of applications in various technology domains, such as optical biosensing, optical quantum communications, intra-datacenter interconnects, modulation, and nonlinear frequency generation.

In this tutorial, we will provide a comprehensive overview of silicon microring resonators, their underlying theory, and their applications in various fields. We will start with the basic theory of microring resonators, followed by a review of their use in label-free biosensors, modulators, filters, and switches. Finally, we will explore their applications in nonlinear and quantum photonics.

Theory of Microring Resonators

A. All-Pass Ring Resonators Microring resonators can be realized in an "all-pass" configuration, where a directional coupler is used to couple light from a straight waveguide into the ring cavity, as shown in Figure 1. This configuration can be considered as an all-pass filter or a notch filter, where the reflected light experiences a phase shift of π + φ, where φ is the single-pass phase shift around the ring.

couple light from a straight waveguide into the ring cavity
Fig. 1. (A) All Pass ring resonator and (B) Add Drop ring resonator

The spectral properties of an all-pass ring resonator can be derived by considering the input wave as continuous and the fields as matched. The ratio of the transmitted and incident fields can be written as:

All Pass ring resonator

where φ = βL is the single-pass phase shift, L is the round-trip length, β is the propagation constant, a is the amplitude transmission coefficient (including propagation loss in the ring and losses in the couplers), and r is the self-coupling coefficient.

The ratio of the output and input intensities, known as the intensity of transmission (Tn), can be obtained by squaring the above equation:

ratio of the output and input intensities

B. Add-Drop Ring Resonators In the "add-drop" configuration, the ring cavity is linked to two bus waveguides, and the incident field is partially transmitted to both the pass and drop ports, as shown in Figure 1(B). The transmission to the pass and drop ports can be calculated using the following equations:

 transmission to the pass and drop ports

where r1 and r2 are the self-coupling coefficients of the two couplers.

C. Spectral Characteristics, Losses, Coupling, and Sensitivity The spectral characteristics of microring resonators, such as the full-width at half-maximum (FWHM) and the free spectral range (FSR), depend on the coupling coefficients and losses. The FWHM for an all-pass ring resonator is given by:

Spectral Characteristics

and for an add-drop ring resonator, it is:

add-drop ring resonator

The FSR is given by:

FSR

where λ is the wavelength, L is the round-trip length, and n_g is the group index.

The quality factor (Q-factor) and finesse are important parameters that characterize the resonance properties of microring resonators. The Q-factor is the ratio of the resonance wavelength to the FWHM, and the finesse is the ratio of the FSR to the FWHM. These parameters provide information about the number of round trips made by the light energy in the cavity before it is converted to internal losses.

To minimize the losses in microring resonators, it is important to increase the Q-factor by using high-quality fabrication processes, reducing propagation losses, and using adiabatic bends to minimize bend losses.

Microring Resonator-Based Label-Free Biosensors

Silicon photonics-based microring resonators have several advantages for use as label-free biosensors, such as low insertion loss, compact design, and the ability to be fabricated using CMOS technology, enabling bulk manufacturing and disposable sensors.

The concept of label-free biosensing using microring resonators is illustrated in Figure 2. Receptor molecules that are selective to the target analyte are immobilized on the surface of the resonator. An aqueous buffer solution is first passed over the sensor surface to determine the reference resonance wavelength. The solution containing the analyte is then introduced, and the analyte molecules bind to the immobilized receptors, changing the effective round-trip length of the resonator and resulting in a shift in the resonance wavelength. The kinetic characteristics and concentration of the analyte can be determined by monitoring the shift in the resonance wavelength over time.

label-free biosensing using ring resonators
Fig. 2. In label-free biosensing using ring resonators, receptor molecules which are selective to the analyte are immobilized over the surface subjected to sensing. An aqueous buffer solution is passed over the sensing surface for determining resonance wavelengths. The solution under test is then passed over the same surface subjected to sensing, which allows binding of analyte molecules over immobilized receptors. This leads to increase in each resonance wavelength which is proportional to binding events.

Nitrogen-rich silicon nitride (NRSN) is a preferred material platform for optofluidic applications due to its wide transparency window, wide bandgap, and better mechanical stability compared to stoichiometric silicon nitride. NRSN microring resonators have been used to detect various analytes, such as cholera toxin subunit B, lectins, and refractive index changes, with sensitivities up to 200 nm/refractive index unit (RIU).

Microring Resonator-Based Modulators, Filters, and Switches

A. Microring Resonator-Based Modulators

Microring resonators can be used as efficient electro-optic modulators by exploiting the shifting of the resonance wavelength. The most common configuration is the all-pass filter with critical coupling, as it can achieve large modulation depth with relatively small changes in the resonance frequency of the cavity.

The modulation of the effective index of the ring resonator can be achieved by manipulating the carrier density in the ring cavity. Carrier injection in a p-i-n diode integrated into the ring is one of the most efficient methods, as it strongly affects the absorption and refractive index of silicon. However, this method is limited by the carrier recombination speed. Reverse-biased p-n junctions can also be used, as they are faster but have a weaker effect on the refractive index.

Microring resonator-based modulators offer advantages such as high speed, compact size, and low power consumption. However, they require excellent process control to spectrally align the transmission spectrum dip with the operating wavelength, and a tuning mechanism is needed to compensate for external disturbances, such as temperature variations.

B. Microring Resonator-Based Filters and Switches

Microring resonators are widely used as spectral filters for multiplexing and demultiplexing wavelength-division multiplexed (WDM) signals in data communication and telecommunication applications. The resonance dips in the transmission spectrum can be used for both detecting and tuning the wavelength channels.

To overcome the sensitivity of single ring resonators to fabrication variations and environmental factors, double-ring resonator configurations are often used as wavelength drop filters, as shown in Figure 3. In this configuration, the two rings can be tuned individually, allowing the resonance wavelength to be moved by tuning the rings in the same direction or switched on and off by detuning the rings with respect to each other.

Ring modulator schematic
Fig. 3. Ring modulator schematic. (a) Top view : indicating active section of the ring. (b) diode integrated in active section of the ring. (c) On biasing transmission in dB reduces and modulator is operated at the slope of ring.
C. Challenges and Tradeoffs

While microring resonators offer excellent performance in terms of wavelength selectivity, they are vulnerable to imperfections that can lead to variations in resonance wavelength, resonance splitting, counter-directional coupling, and, most importantly, quality factor. These imperfections are a result of the high refractive index contrast between silicon and its oxide, which is also the reason for their compact size and wide FSR.

To address these challenges, careful fabrication processes and design considerations are necessary to minimize losses and optimize the performance of microring resonator-based devices.

Microring Resonators for Nonlinear and Quantum Photonics

The tight optical confinement and high nonlinearity of silicon waveguides make them a promising platform for nonlinear and quantum photonic applications. Microring resonators, in particular, can enhance the nonlinear effects due to the resonance field enhancement.

One of the key nonlinear processes demonstrated in silicon microring resonators is spontaneous four-wave mixing (SFWM), which can be used for on-chip single-photon generation. The strong resonance field enhancement in microring resonators allows for efficient SFWM at relatively low pump powers, compared to regular waveguides.

However, the high two-photon absorption (TPA) in silicon results in the generation of free carriers, leading to free carrier absorption (FCA) losses that limit the single-photon generation efficiency at high pump powers. Strategies to mitigate this, such as using integrated p-i-n diodes to sweep away the carriers, have been explored.

Further challenges in using microring resonators for quantum photonics include the need for very high-extinction-ratio spectral filters to suppress the remaining pump photons, which can overwhelm the single-photon detectors.

Conclusion and Future Perspectives

In this tutorial, we have provided a comprehensive overview of silicon microring resonators, their underlying theory, and their diverse applications in various fields. The compact size, high confinement, and wide FSR of silicon microring resonators have enabled their use in a wide range of photonic integrated circuits and devices.

While the high refractive index contrast of silicon photonics has been a key enabler for microring resonators, it also introduces challenges related to fabrication imperfections and environmental sensitivity. Addressing these challenges through advanced fabrication processes and design techniques will be crucial for the continued development and widespread adoption of microring resonator-based technologies.

As silicon photonics continues to evolve, microring resonators are expected to play an increasingly important role in various applications, from biosensing and optical communications to nonlinear and quantum photonics. The ability to leverage CMOS fabrication processes for large-scale, low-cost integration of microring resonators will further drive their adoption across a wide range of industries and research fields.

Reference

[1] Y. R. Bawankar and A. Singh, "Microring Resonators Based Applications in Silicon Photonics - A Review," in Proceedings of the 5th Conference on Information and Communication Technology (CICT), 2021, DOI: 10.1109/CICT53865.2020.9672427.

Comments


bottom of page