Observation of ultrafast phenomena and extreme light

The generation of intense femtosecond pulses of near-infrared or visible light opens up a vast field of investigation in atomic, molecular and solid-state physics. By exciting rare gas jets with these pulses, we generate ultraviolet or X-ray (XUV) light pulses lasting a few tens of attoseconds (1 attosecond = ^10-18 s) and energetic photoelectron beams. These XUV pulses and electrons, in conjunction with the IR or visible pulses that generated them, are used to probe matter at ultra-short times and over distances of the order of a few Angstroms. These experiments can be carried out using laboratory lasers, thanks to a light amplification method perfected in the 1980s. Powerful, large-scale facilities make it possible to reach intensities that pave the way for extreme light physics (relativistic effects, nuclear physics, non-linear quantum electrodynamics, etc.). Lesson six began with an overview of the progress made in generating ultra-short, high-intensity light pulses over the last fifty years. It shows that the race for large light fields and the race for very short pulses have been strongly correlated, as the processes generating ultrashort flashes are linked to highly nonlinear phenomena requiring very large electric field amplitudes. Initially, the pulses increased rapidly in amplitude and brevity, reaching a plateau in the 1980s when the amplifying glasses used reached their irreversible damage threshold. The CPA(chirped pulse amplification) method was then invented. It involves dispersing the light pulse in frequency and spreading it out in time, using an optical delay line constructed with dispersive gratings. The pulse then carries the same energy, but over a much longer time. It passes through the amplifying medium, which is no longer in danger of being destroyed. After passing through this medium, a system of gratings operating in the opposite direction compresses the pulse, enabling very high peak intensities to be reached. The intense femtosecond pulses thus obtained lend themselves to the implementation of very high-order non-linear processes, leading to the generation of much shorter, much more energetic light pulses.

The second part of the lesson then described these non-linear processes, which occur during the interaction between femtosecond IR pulses and a rare-gas atomic jet. The phenomena of multiphoton and suprathreshold ionization were analyzed qualitatively. A distinction was made between the "perturbative" regime at relatively low intensity and the "non-perturbative" regime at high intensity, in which the incident electric field can be described as quasi-static. It is this second regime that is of interest for the generation of ultra-short XUV pulses. The process is then described by a semi-classical model, with an electron tunneling away from the atom in the superposition of the atom's Coulomb potential and the linear potential associated with the quasi-static laser field. It moves away from the atom, only to be stopped by the laser field when it turns around and collides with the ionic core, producing above-threshold ionization and the generation of ultrashort high harmonics. Controlling the process requires the use of IR pulses with a stable phase difference between carrier and envelope. IR pulses are in fact the pulse trains of a frequency comb. Methods for controlling this comb, similar to those used for atomic clocks, are used to adjust the relative phases of the carrier and envelope pulses. By adjusting the polarization of the light, it is finally possible to isolate a single pulse from the comb to generate XUV attosecond light.

The third section takes a qualitative look at the physics currently being developed with this attosecond light. One class of experiments involves directing the XUV pulses and the IR pulses that gave rise to them into a second vacuum chamber, where they interact together with an atomic probe (which may be another rare gas, a molecular jet or even a solid surface). The delay between the IR and XUV pulses is controlled and variable, allowing the attosecond pulse to interact at a precise moment in the slower oscillation of the femtosecond pulse. By analyzing the products of this interaction (e.g. the energy of the electron finally ejected), the initial femtosecond pulse can be reconstructed using a kind of ultra-fast stroboscopy, thus achieving time resolution of an optical field for the first time. It also makes it possible to study in detail the electronic distribution and dynamics of the probe atom or molecule. The electron packet torn from the atom and later returned to it interferes with the wave function of the associated ion, and experimental analysis of this interference enables this function to be reconstructed, using methods akin to holography or tomography. The dynamics of atomic or molecular electrons, or of electrons in the solid used as a probe, can also be studied with a time resolution that is currently a few tens of attoseconds, but which is set to become even faster with the extension of high-order harmonic generation to the extreme UV range. The lesson concluded with a brief overview of the physics associated with extreme light produced by power lasers in operation or planned in gigantic facilities (megajoule lasers, ELI lasers). The use of these lasers in plasma physics, relativistic electron physics, nuclear physics and non-linear electrodynamics was discussed, as well as gas pedals that exploit the wake field of a laser-excited plasma(wake field accelerators). These new types of gas pedal could prove useful for biological or medical applications.