TY - JOUR
T1 - True-time delay and mode-matching in Archimedean spiral silicon photonic waveguides
AU - Murad, Ahmad
AU - Karabchevsky, Alina
N1 - Publisher Copyright:
© 2025 The Author(s). Published by IOP Publishing Ltd.
PY - 2025/9/30
Y1 - 2025/9/30
N2 - We report a comprehensive analysis and design optimization of true-time delay (TTD) photonic waveguides using compact Archimedean spiral geometries, with a focus on low-loss silicon nitride (Si3N4) and silicon-on-insulator (SOI) platforms. Using a combination of finite element and finite-difference time-domain simulations at 1550 nm, we evaluated the propagation characteristics, mode-matching behavior, and bending losses of various waveguide designs. Our results demonstrate that Si3N4 rectangular waveguides exhibit superior performance for long optical paths, achieving minimal radiation and scattering losses, whereas SOI is more suitable for compact integration. We further investigated curvature-engineered bends (Euler, Bézier, and circular) and identified Quartic Bézier bends as optimal for tight-radius, low-loss guiding. By minimizing crosstalk through tailored waveguide gaps and carefully engineered S-bend transitions, we present a scalable low-loss TTD architecture. These findings offer critical insights for integrating ultralow-loss, material-efficient delay lines into photonic circuits for applications in signal processing, LiDAR, phased array antennas, and neuromorphic computing.
AB - We report a comprehensive analysis and design optimization of true-time delay (TTD) photonic waveguides using compact Archimedean spiral geometries, with a focus on low-loss silicon nitride (Si3N4) and silicon-on-insulator (SOI) platforms. Using a combination of finite element and finite-difference time-domain simulations at 1550 nm, we evaluated the propagation characteristics, mode-matching behavior, and bending losses of various waveguide designs. Our results demonstrate that Si3N4 rectangular waveguides exhibit superior performance for long optical paths, achieving minimal radiation and scattering losses, whereas SOI is more suitable for compact integration. We further investigated curvature-engineered bends (Euler, Bézier, and circular) and identified Quartic Bézier bends as optimal for tight-radius, low-loss guiding. By minimizing crosstalk through tailored waveguide gaps and carefully engineered S-bend transitions, we present a scalable low-loss TTD architecture. These findings offer critical insights for integrating ultralow-loss, material-efficient delay lines into photonic circuits for applications in signal processing, LiDAR, phased array antennas, and neuromorphic computing.
KW - AI reservoir computing
KW - highspeed
KW - true time delay
KW - ultra-low-loss
KW - waveguides
UR - https://www.scopus.com/pages/publications/105016134509
U2 - 10.1088/2631-8695/ae0413
DO - 10.1088/2631-8695/ae0413
M3 - Article
AN - SCOPUS:105016134509
SN - 2631-8695
VL - 7
JO - Engineering Research Express
JF - Engineering Research Express
IS - 3
M1 - 035378
ER -