Description

The proposed PhD thesis lies in the field of integrated silicon photonics [1]. The integration of diverse materials and advanced miniaturization in photonic platforms enables the realization of numerous optical functions within mm²-scale footprints, comprising hundreds of components and complex light-manipulating circuits. This field bridges fundamental physics and technological applications, from waveguide-level studies to large-scale photonic systems.

Nonlinear optical effects have long been central to scientific advances, from the first ruby laser (1960) to modern attosecond laser sources used to probe ultrafast electronic and spin dynamics. Nonlinear optics underpins a vast range of applications, including optical signal processing, material surface analysis, high-resolution imaging, and medical treatments. Historically, these developments occurred in free-space and fibre optics, but integrated nonlinear photonics—merging both worlds—has recently emerged as a powerful research frontier [2].

Planar photonic platforms based on Si, SiN, or AlGaAs offer tight mode confinement and long interaction lengths, ideal for exploiting χ² and χ³ nonlinearities. These effects enable key functionalities such as frequency comb generation, supercontinuum extension, wavelength conversion (FWM, SHG, DFG), all-optical signal processing, and quantum photon-pair generation. Yet, achieving efficient on-chip nonlinear processes remains challenging. Device performance depends on numerous coupled parameters—geometry, material composition, dispersion, coupling, and losses—making analytical design often insufficient [3]. The optimization of such systems requires accounting for phase matching, group-velocity dispersion (GVD), confinement factors, and fabrication tolerances.

Given the complex cascade of nonlinear effects in realistic structures, AI-driven inverse design offers a promising approach to uncover unconventional geometries (e.g., multi-section tapers, asymmetric waveguides, coupled resonators) that may elude human intuition [4]. However, data scarcity poses a major limitation: generating high-fidelity nonlinear simulation data is computationally costly, and existing datasets are too narrow for robust large-scale training. Addressing this requires strategies such as dimensionality reduction of opto-geometric parameters and incorporating physical constraints—symmetries, conservation laws, and modal structures—to reduce the effective search space.

Within this context, the PhD aims to explore the frontiers of integrated nonlinear photonics through a unified modelling framework for complex nonlinear interactions in coupled photonic systems. The overarching goal is to develop a general, flexible theoretical foundation enabling inverse design of advanced nonlinear devices.

Rather than focusing on a single component, the project seeks to capture the interplay of dispersion, nonlinearity, and coupling in integrated resonant structures—such as photonic crystal cavities and hybrid material platforms. Through rigorous modelling, numerical simulations, and experimental validation, it will deliver predictive tools for designing and optimizing nonlinear phenomena.

A central aspect is the integration of emerging nonlinear materials—notably two-dimensional transition metal dichalcogenides (TMDs) and chalcogenide glasses—within silicon photonics. Their large nonlinear coefficients and broad transparency windows make them ideal candidates for validating the framework’s predictive and design capabilities.

Ultimately, this research aims to advance the understanding of nonlinear light–matter interactions in integrated systems and enable the development of purpose-designed, scalable, and reconfigurable nonlinear photonic circuits. By uniting modelling, inverse design, and experimental demonstration, it will contribute to the next generation of chip-scale technologies for ultrafast signal processing, low-noise amplification, and quantum light generation.

Compétences requises

We are seeking a motivated candidate to join our team and actively contribute to the proposed research topic. This thesis offers the opportunity to develop models leveraging both your technical skills and creativity to explore innovative solutions in nonlinear photonics. We are looking for a student eager to work at the intersection of theory, simulation, and experiments, with a strong interest in integrated photonics and nonlinear optics. We encourage applicants who are proactive, detail-oriented, and open-minded. Motivation for theory and numerical modelling for photonic design is mandatory.

Bibliographie

[1] “Roadmap on silicon photonics”
David Thomson, Aaron Zilkie, John E Bowers, Tin Komljenovic, Graham T Reed, Laurent Vivien, Delphine Marris-Morini, Eric Cassan, Léopold Virot, Jean-Marc Fédéli, et al.
Journal of Optics 18 (7), 073003, 2016, https://iopscience.iop.org/article/10.1088/2040-8978/18/7/073003

[2] “Octave-spanning dissipative Kerr soliton frequency combs in Si3N4 microresonators”
M. H. P. Pfeiffer, C. Herkommer, J. Liu, H. Guo, M. Karpov, E. Lucas, M. Zervas, T. J. Kippenberg
Optica 4 (7), 684 (2017) , https://doi.org/10.1364/OPTICA.4.000684

[3] “Adjoint Method and Inverse Design for Nonlinear Nanophotonic Devices”
Tyler W. Hughes Momchil Minkov, Ian A. D. Williamson, Shanhui Fan
ACS Photonics 5 (12) - https://pubs.acs.org/doi/10.1021/acsphotonics.8b01522

[4] “Interfacing Nanophotonics with Deep Neural Networks:AI for Photonic Design and Photonic Implementation of AI Taehyuk Park, Sujoy Mondal, and Wenshan Cai*
Laser Photonics Rev. 2025, 19, 2401520 – pp. 1-31

Mots clés

Silicon photonics, Nonlinear optics, Numerical modelling and AI assisted design, Optical experiments, Clean room fabrication techniques