Optimization of Laser-Induced Optical Response in Advanced Materials for High-Performance Photonic Applications

Abstract

Many photonic switches, modulators, sensors and spectral-control devices are based on light-induced variations in the refractive index. It is difficult to find a suitable material, because published nonlinear coefficients depend on wavelength, pulse duration, and measurement method. To address this, we performed a loss-aware analysis on publicly available experimental data. The dataset consisted of 366 nonlinear-index measurements from 240 source records and covered 96 materials. In all cases the linear refractive-index and extinction data were fitted at the same wavelength as the nonlinear measurement. These values were used in a Kerr phase-shifter model that related material response to transmission, interaction length and drive intensity. We tested 5,346 material-band-length combinations. Realizable designs required at least 90% linear transmission, pi-pulse intensity not exceeding 100 GW cm⁻², and index change less than 1%. These conditions left 820 designs from 23 material-band pairs, including 75 non-dominated solutions.  The optimum trade-off was found for GaAs at 1064 nm, which gives a modeled pi-shift intensity of 3.98 GW cm-2 over 0.410 mm. That estimate, however, came from one nonlinear source.  Silicon at 1550 nm was fifth with four sources, with a median nonlinear index of 6.35 × 10⁻¹⁸ m² W⁻¹. The clustered meta-regression explained 74.7 % of the observed variance, and the grouped cross-validation yielded a mean R2 of 0.534. Even after correction, measurements of semiconductors remained higher than the chalcogenide reference group. The created shortlist is useful for screening, but direct validation of non-linear absorption, damage limits and waveguide confinement is still needed.