Broadband directional coupler

In the rapidly evolving field of silicon integrated photonics, the directional coupler (DC) stands out as a crucial building block for a multitude of applications, such as optical signal processing, sensing, and communication systems. As a passive device, a DC allows for the precise manipulation and distribution of light between two parallel waveguides within close proximity, enabling highly efficient and compact coupling with minimal loss. This elegant, yet simple structure capitalizes on the waveguiding properties of silicon, leveraging the evanescent coupling between the propagating modes to achieve precise control over the flow of optical power, thereby playing a pivotal role in shaping the future of photonic integrated circuits.

Conventional compact DCs often face limitations in terms of narrow bandwidth, while broadband designs tend to require a significantly larger footprint. This example explores a design for compact, broadband DCs, as proposed in Zeqin Lu, Han Yun, Yun Wang, Zhitian Chen, Fan Zhang, Nicolas A. F. Jaeger, and Lukas Chrostowski, "Broadband silicon photonic directional coupler using asymmetric-waveguide based phase control," Opt. Express 23, 3795-3808 (2015)DOI: 10.1364/OE.23.003795 (opens new window). The key innovation in this design, as compared to traditional DCs, lies in its incorporation of an asymmetric-waveguide-based phase control section. To demonstrate a concrete example, we will design a 2x2 DC for the TE mode, with a 50%/50% (-3 dB) splitting ratio, operating within the 1500 nm to 1600 nm range. Different polarizations and splitting ratios can be achieved in a similar design process.

Initially, we employ the transfer matrix method (TMM) to model the DC in a semi-analytical manner. TMM necessitates the calculation of effective indices for various waveguide configurations, for which we utilize Tidy3D’s waveguide plugin (opens new window), as it offers a convenient means of performing mode analysis. TMM provides a computationally efficient and accurate preliminary estimation of the ideal design parameters, which are then further optimized using rigorous 3D FDTD simulations.

To view the full example in Python, please click here (opens new window).