Laser-Plasma Accelerators (LPAs) produce electric fields exceeding 100 GV/m, that is 3 orders of magnitude larger than those obtained in metallic-cavity accelerators. They could thus allow for a drastic decrease of the size of accelerators for scientific, medical and industrial applications. A high field-gradient is however not sufficient for reaching high-energies; the electron beam has also to experience the accelerating field on long distances, which is challenging in a LPA because of 3 phenomenons: diffraction, pump depletion and dephasing. Diffraction and pump depletion leads to a decrease of the laser intensity during the acceleration, down to a level from which the laser can no more drive a wakefield. Dephasing corresponds to electrons reaching a decelerating phase of the electric field. It occurs because the phase velocity of the accelerating field is smaller than the velocity of the electron beam. To date, the highest beam energies have been obtained by guiding the laser in a capillary discharge, thus overcoming diffraction.
Here we propose a new acceleration concept, based on the use of high-intensity quasi-Bessel beams and spatio-temporal couplings, which allows to overcome not only diffraction but also pump depletion and dephasing. The velocity of the quasi-Bessel beam is superluminal in vacuum and dephasing is suppressed by using spatio-temporal couplings to phase lock the electron beam on the accelerating field. In this scheme, the electron energy is proportional to the laser energy and inversely proportional to the laser pulse length (the shorter the laser, the higher the beam energy).
We will first present Particle-In-Cell simulations demonstrating this concept. We will then show preliminary experimental results illustrating the generation of high-intensity quasi-Bessel beams as well as the generation of a 1 cm plasma-waveguide.
Betatron radiation from laser-plasma accelerators reproduces the principle of a synchrotron on a millimeter scale, but featuring femtosecond duration. Here we present the outcome of our latest developments, which now allow us to produce stable and polarized X-ray bursts. Moreover, the X-ray polarization can simply be adjusted by tuning the polarization of the laser driving the process. The excellent stability of the source is expressed in terms of pointing, flux, transverse distribution and critical energy of the spectrum. These combined features make our betatron source particularly suitable for applications in ultrafast X-ray science.
In this presentation we will describe the generation process, relying on the ionization injection scheme for laser-plasma acceleration. We will show experimental measurements, numerical results and first applications in time-resolved spectroscopy.
One direction towards compact Free Electron Laser is to replace the conventional linac by a laser plasma driven beam, provided proper electron beam manipulation to handle the large values of the energy spread and of the divergence. Applying seeding techniques enable also to reduce the required undulator length. The rapidly developing LWFA are already able to generate synchrotron radiation. With an electron divergence of typically 1 mrad and an energy spread of the order of 1 % (or few), an adequate beam manipulation through the transport to the undulator is needed for FEL amplification. Electron beam transfer follows different steps with strong focusing variable strength permanent magnet quadrupoles, an energy demixing chicane with conventional dipoles, a second set of quadrupoles for further dedicated focusing in the undulator. A test experiment for the demonstration of FEL amplification with a LWFA is under preparation and progress on the equipment preparation and expected performance are described.
One of the key ingredients of laser-plasma accelerators is their injector, which defines how electrons are trapped into the laser-driven plasma wave. The stability and control of laser-plasma electron bunches strongly depends on this injection stage. Self-injection is a convenient way to achieve the electron trapping and is the most widely used injector. Here we demonstrate, by using a variable length gas cell, that injection can be achieved by either longitudinal or transverse self-injection, giving rise to very different electron beam features. The results are supported by 3 dimensional particle-in-cell simulations.