Temporal and polarization dependence of the nonlinear optical response of solvents

Peng Zhao,Matthew Reichert,Eric W Van Stryland,David J Hagan,Sepehr Benis

doi:10.1364/optica.5.000583

Abstract

It has long been known that nonlinear refraction in solvents can depend on pulse width, and this along with experimental uncertainties has led to orders-of-magnitude disagreements in nonlinear refractive coefficients reported in the literature. To resolve this issue, we perform beam-deflection (BD) measurements of the rigorously defined nonlinear impulse response function for 24 commonly used solvents selected from various classes of molecules. Using this polarization-resolved BD, the bound-electronic and the three major nuclear contributions are separately measured by determining the magnitudes, symmetry, and temporal dynamics of each mechanism. This allows us to construct the response functions that we use to accurately establish self-consistent references for predicting and interpreting the outcomes of other experiments performed on these materials over the temporal range from 10 fs to 1 ns. The results also provide insight into relating solvent nonlinearities with their molecular structures and exploring the effects of the Lorentz–Lorenz local field. We find that nonconjugated molecules with small polarizability anisotropy exhibit negligible reorientational response, and hence the nonlinear refraction is almost independent of pulse width. Knowledge of the response functions also allows engineering the transient nonlinear refractive properties of solutions of organic dyes, for example, materials with effectively zero nonlinear refraction.

Highlights

The nonlinear optical (NLO) response of organic solvents has long been studied in the visible and near-infrared spectral range, where they are nearly transparent
This large discrepancy arises in part from the noninstantaneous nuclear nonlinearity, present in all molecular liquids, which complicates the temporal NLO response of molecular solvents, resulting in an effective nonlinear refraction (NLR) coefficient, n2,eff, which depends on the pulse width [20,21,22,23,24,25]
As an example of how to extract the parameters of the NLO response functions from the polarization-resolved BD measurements, the measured signal, ΔE∕E, of benzene is shown in Fig. 3(a) as a function of delay with parallel, perpendicular, and magic-angle polarization combinations

Summary

INTRODUCTION

The nonlinear optical (NLO) response of organic solvents has long been studied in the visible and near-infrared spectral range, where they are nearly transparent. As discussed in [20,23,24], we decompose the overall response into a nearly instantaneous NLR, which originates from the bound-electronic response, and three major noninstantaneous mechanisms due to nuclear motions, namely, collision, libration, and diffusive reorientation These mechanisms lead to a time-dependent third-order response, giving a change of refractive index that is linear in irradiance. The experimental results of the measured magnitude of the reorientational NLR are compared to the theoretical values derived from the linear polarizability tensors where the Lorentz–Lorenz local-field correction is considered Using these response functions, the pulsewidth dependence of n2,eff can be used to predict the outcomes of other experiments, such as Z-scan. Together with the bound-electronic NLR, this leads to an irradiance- and time-dependent third-order response, yielding a change of refractive index linear in the excitation irradiance I t as

POLARIZATION-RESOLVED BD

RESULTS AND DISCUSSION

DETERMINATION OF EFFECTIVE n2 VERSUS PULSE WIDTH

ENGINEERING THE EFFECTIVE n2 IN SOLUTIONS

CONCLUSION