Photonics Research

Integrated Fabry-Pérot (FP) cavities are extensively used, finding applications in wavemeters, gravity-wave detection, refractive-index sensing, semiconductor lasers, quantum-information processing, frequency combs, and high-finesse spectral filters.

 

Compared to travelling-wave cavities, e.g., micro-ring or micro-disk resonators, FP cavities have some unique applications from the standing-wave resonant modes, such as cavity-enhanced optical trapping and cooling. In addition, bent waveguides are not required in most FP cavities, leading to smaller footprints. Silicon-on-insulator (SOI) is a prevalent platform for dense photonic integration, thanks to its high yield and CMOS-compatibility.

 

However, the monolithic silicon integrated FP cavity has not been previously demonstrated in the ultrahigh-Q regime (> 106). The Q limitation comes from the excess losses caused by integrated reflectors. Conventional types of reflectors, such as distributed Bragg reflectors (DBR), waveguide loops, metallic mirror, and inverse-designed structures, all suffer from relatively high reflection losses or stringent fabrication requirements.

 

To address the problems, the research group led by Prof. Hon Ki Tsang at the Chinese University of Hong Kong has proposed a novel approach to reduce the reflection loss and demonstrated the first million-Q silicon integrated FP cavity. The relevant research results were published in Photonics Research, Volume 10, No. 11, 2022 (Hongnan Xu, Yue Qin, Gaolei Hu, and Hon Ki Tsang. Million-Q integrated Fabry-Pérot cavity using ultralow-loss multimode retroreflectors[J]. Photonics Research, 2022, 10(11): 2549).

 

The core idea is to use integrated retroreflectors as high-reflectance mirrors in the FP cavity [Fig. 1(a)]. The proposed retroreflector is formed by two total-internal-reflection (TIR) mirrors at an angle of 90o. Through the TIR effect, input rays will consecutively bounce off at A and A' [Fig. 1(b)]. The major source of loss is the scattering around the corner region.

 

The first higher-order transverse electric mode (TE1), which has zero field intensity at the central point, is used to mitigate the corner scattering [Fig. 1(c)]. Compared to previously reported reflectors, such a novel retroflector design can offer low losses (< 0.05 dB) and is tolerant to fabrication flaws. Losses incurred by other mechanisms are also meticulously tailored. The measured loaded Q factor is as high as > 1.1×106 with a taper length of 1.5 mm [Fig. 1(d)].

 

At different wavelengths, an even higher loaded Q factor of ≈ 2.1×106 is experimentally observed. By measuring the Q factors with varying taper lengths, the intrinsic Q factor is estimated to be ≈ 3.4×106. Usually, such high a Q factor can only be attained by using bulky silica waveguides with index contrast of < 1%.

 

Fig.1 (a) Conceptual illustration of the Fabry-Pérot (FP) cavity. (b) Working principle of the retroreflector. (c) Upper panel: calculated electric field profiles of TE0 and TE1 modes. Lower panel: Calculated light propagation profiles for the retroreflector when TE0 and TE1 modes are launched. (d) Measured transmittance spectrum. (e) Measured Q factors with varying taper lengths.

 

Based on this scheme, it is possible to further improve the intrinsic Q factor via the utilization of more advanced fabrication technologies (e.g., photonic Damascene process) or low-contrast material platforms (e.g., SNOI) to reduce the side-wall roughness. The reflection loss can also be mitigated by using the hard mask in the dry etching process to improve the perpendicularity of side-walls.

 

The proposed FP cavity has an ultrahigh Q factor and a relatively small free spectral range. These features are just desirable for electrically pumped narrow-linewidth dense comb lasers. Another potential application is the linewidth narrowing, which is to suppress close-to-carrier noises of lasers by using an ultrahigh-Q cavity isolated in a cryogenic environment. The large modal volume of the proposed FP cavity could help to alleviate photothermal phase noises.