Angle-resolved studies of XUV–IR two-photon ionization in the RABBITT scheme

Abstract

Reconstruction of attosecond beating by interference of two-photon transitions (RABBITT) is an established technique for studying time-delay in photoionization of atoms and molecules. It has been recently extended to angle-resolved studies, accessing diverse fingerprint observables of the attosecond photoemission dynamics within the bound-continuum and continuum–continuum transitions. In this work, we address the general form of the ISB(θ,τ) two-photon photoelectron angular distributions (PADs) associated to the RABBITT sideband signal, as a function of the emission angle θ, and the delay τ between the XUV attosecond pulse train and the infrared (IR) dressing field at play in the RABBITT scheme. Relying on the expansion in Legendre polynomials, the PAD is synthesized in terms of a reduced set of coefficients which fully describe both its static (τ-independent) and dynamic (τ-dependent) components and enables us to retrieve any observable characterizing the PAD. This unified framework streamlines the comparison between different experimental or theoretical data sets and emphasizes how some observables depend on the experimental conditions. Along with the modelled analysis, we report new results of angle-resolved RABBITT direct ionization of the np valence orbital of Ar(3p6) and Ne(2p6), employing electron-ion coincidence momentum spectroscopy at the new Attolab facility. In this case, the nine coefficients synthesizing the PAD are further linked to the magnitude and phase of the transition dipole matrix elements, providing a fundamental test of theoretical predictions. Similarities and differences are found between Ar and Ne in the explored low energy region, up to 20 eV above the ionization threshold, where the electron dynamics is most sensitive to electronic correlation. Further interpretation of these results would benefit from a comparison with advanced many-body theoretical simulations.

In situ sub-50 attosecond active stabilization of the delay between infrared and extreme ultraviolet light pulses

“…Focused onto a continuous Argon gas jet, it drives a HHG source emitting attosecond pulses trains (APT). The beamline is optimized for experiments requiring fairly high recurrence and long acquisition times, for instance coincidence measurement in gas phase [9] or photoemission in solids [29]. RABBIT [5,30] spectroscopy is permanently operational on FAB10, which allows the characterization of the XUV light and the experiments in the end station to be performed simultaneously.…”

“…In other words, the difference of optical paths varies by a sideband spatial period in 10 minutes. This passive stability is too poor to maintain attosecond precision during experiments lasting several hours, forcing the use of convoluted post-analysis protocols [9].…”

“…It is the combination of XUV wavelength, highly non-linear stage, metersize interferometer, and vacuum environment that makes the nanometer-grade stability a challenging requirement. On top of that, a strong demand is emerging for performing experiments that require long accumulation times (over tens of hours), for instance coincidence spectroscopy [7][8][9], or photoemission from surfaces [10]. In order to meet those needs, two main experimental strategies are combined: increasing the repetition rate of the laser to reduce the acquisition time, and improving, passively or actively, the mechanical stability of the interferometer.…”

Complete characterization of multi-channel single photon ionization

“…We determine these nine unknown quantities using a global fit to our experimental measurements. In general, multiphoton electron angular distributions can be written as an expansion in Legendre polynomials [26,27,29,30]. For a two-photon transition, without parity mixing, the expansion needs only three polynomials, P 0 (x) = 1, P 2 (x) = (3x 2 − 1)/2, and P 4 (x) = (35x 4 − 30x 2 + 3)/8, reading as…”

“…The observed angular structure of the sidebands depends not only on the interference between the different partial waves of the angular momentum channels reached via single photon ionization, but is also strongly influenced by the cc-transitions [16,18]. The additional interaction with the IR field leads to an increase in the number of angular channels (see Figure 1, where only the m = 0 angular path is indicated), and modifies the radial amplitude and phase of the outgoing photoionization wavepacket [25,26,27]. In the following, we present a general method to retrieve the amplitude and phase of each photoionization angular channel from experimental data.…”