The emerging applications of silicon photonics

The emerging applications of silicon photonics

Newton

10.1016/j.newton.2025.100357

https://doi.org/10.1016/j.newton.2025.100357

Luigi Ranno, Yong Zen Tan, Andres Garcia Coleto, Vinh Huy Nguyen To, Shuoqiu Tian, Yihong Sun, Zhiyun Xu, Sang Hoon Chae, Louis Martin-Monier, Ziyan Huang, Jia Qi Lim, Hong Wang, Xianshu Luo, Wenyu Jiang, Qian Wang, Arka Majumdar, Jelena Notaros, Juejun Hu, Jia Xu Brian Sia

22 December 2025

Ranno, L., Tan, Y. Z., Coleto, A. G., Nguyen To, V. H., Tian, S., Sun, Y., Xu, Z., Chae, S. H., Martin-Monier, L., Huang, Z., Lim, J. Q., Wang, H., Luo, X., Jiang, W., Wang, Q., Majumdar, A., Notaros, J., Hu, J., & Brian Sia, J. X. (2026). The emerging applications of silicon photonics. Newton, 2(1), 100357. https://doi.org/10.1016/j.newton.2025.100357

Silicon photonics (SiP) has rapidly evolved from data-communication technology into a broadly enabling platform for modern physics and engineering. Its foundation on CMOS-compatible fabrication processes provides desirable scalability, cost-efficiency, and integration density, allowing light to be generated, routed, and processed on a chip with the same maturity as electronic circuits. This convergence is revolutionizing how photons can be harnessed for sensing, computation, and information processing. In particular, recent advances are propelling SiP beyond data centers into three frontiers. In LiDAR, chip-scale frequency-modulated continuous-wave systems employing optical phased arrays are delivering solid-state beam steering and coherent ranging for autonomous navigation. In quantum technologies, SiP offers a manufacturable path toward large-scale processors through heterogeneous integration of active materials such as barium titanate and superconducting detectors, enabling low-loss, high-fidelity quantum interconnects. In neuromorphic computing, SiP leverages wavelength and time-division multiplexing to perform massive parallel optical matrix operations with high energy efficiency, pointing toward ultra-fast, low-latency AI hardware. In addition, the integration of phase-changing materials and ferroelectric materials allows for multi-level programmability of neural-network synapses. This perspective discusses how SiP is changing from an application-specific solution to a general-purpose photonic platform capable of unifying communication and computation. By integrating photonics and electronics within a shared architecture, SiP stands to revolutionize the next generation of quantum-aware, perception-enabled, and energy-conscious physical systems, ushering in a new paradigm where information is processed at the speed of light.

Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

This research / project is supported by the A*STAR - Manufacturing, Trade, and Connectivity programmatic
Grant Reference no. : M23M2b0056

For the published article, please refer to: https://doi.org/10.1016/j.newton.2025.100357

https://oar.a-star.edu.sg/communities-collections/articles/22764

2950-6360

Institute of Materials Research and Engineering