Multipass cells: 1D numerical model and investigation of spatio-spectral couplings at high nonlinearity

Abstract

Multipass cells (MPC) are used nowadays as nonlinear tools to perform spectral broadening and temporal manipulation of laser pulses while maintaining a good spatial quality and spatio-spectral homogeneity. However, intensive 3D nonlinear spatio-temporal simulations are required to fully capture the physics associated to pulse propagation inside these systems. In addition, the limitations of such scheme are still under investigation. In this study, we first establish a 1D model as a useful design tool to predict the temporal and spectral properties of the output pulse for nearly Gaussian beams, in a wide range of cavity configurations and nonlinearity levels. This model allows to drastically reduce the computation time associated to MPC design. The validity of the 1D model is first checked by comparing it to 3D simulations. The results of the 1D model are then compared with experimental data collected from a near concentric gas-filled multipass cell presenting a high level of nonlinearity, enabling the observation of wavebreaking. In a second part, we experimentally characterize the spatio-spectral profile at the output of this experimental setup, both with an imaging spectrometer and with a complete 3D characterization method known as INSIGHT. The results show that gas-filled multipass cells can be used at peak power levels close to the critical power without inducing significant spatio-spectral couplings in intensity or phase.

Hybrid pulse propagation model and quasi-phase-matched four-wave mixing in multipass cells

“…1. As already described in [11], these parameters correspond to a high level of nonlinearity (Ppeak/Pcrit=0.6), for which the simplified spatially averaged approach significantly underestimates the level of nonlinearity, resulting in a lower amount of spectral broadening. The reason for this is that nonlinearity in the space domain leads to significant changes in the beam size upon propagation, as can be observed in Fig.…”

“…The test case we choose is the same as in our previous work [11], to allow direct comparison between three numerical models. The first model is a 4D model described in [13], that fully accounts for diffraction, dispersion and third order nonlinearity by solving an envelope equation [22] using the split-step method in space and time.…”

“…The second model is a purely 2D model that assumes a constant beam size. This beam size is calculated by averaging the nonlinearity accumulated by the linear stationary beam in the MPC over one roundtrip, as described in [11]. The third model is the one described in this article.…”

“…The model presented here is particularly well suited to describe propagation in nonlinear regenerative amplifiers [6], or MultiPass Cells (MPCs) [7][8][9][10], where it was shown that an injected Gaussian beam essentially retains its character upon propagation and that space-time couplings are only observed at extreme levels of nonlinearity [11,12]. Compared to the model presented in [11], that is purely (t,z) without a description of spatial effects, the model presented here allows to take into account Kerr lensing that can have non negligible impact on the beam caustic, as will be discussed hereunder. Compared to full 4D dimensional models [13], it is much faster in terms of computing time, and allows to account for the impact of beam imperfections on the level of nonlinearity easily through an M 2 beam quality factor.…”

“…In the first part, we first describe the temporal and spatial parts of the model and how they are coupled. We compare its results both to a full 4D model and to the very simple 2D model described in [11], showing that the level of physical description adopted is ideal for the particular problem of propagation in MPCs. In a second part, we apply the model to reproduce the experimental data obtained in high-energy gas-filled MPC experiments that use near concentric configurations [16,17].…”