Due to their ultra-short duration and peak currents in the kA range 1,2 , laserwakefield accelerated electron bunches are promising drivers for ultrafast X-ray generation in compact free-electron-lasers (FELs), Thomson-scattering or betatron sources [3][4][5] . Here we present the first single-shot, high-resolution measurements of the longitudinal bunch profile obtained without prior assumptions about the bunch shape. Our method allows complex features, such as multi-bunch structures, to be detected. Varying the length of the gas target, and thus the acceleration length, enables an assessment of the bunch profile evolution during the acceleration process. We find a minimum bunch duration of 4.2 fs (full width at half maximum) with shot-to-shot fluctuation of 11% rms. Our results suggest that after depletion of the laser energy, a transition from a laser-driven to a particle-driven wakefield occurs, associated with the injection of a secondary bunch. The resulting double-bunch structure might act as an elegant approach for driver-witness type experiments, i.e. allowing a non-dephasinglimited acceleration of the secondary bunch in a plasma-afterburner stage 6,7 .Since the first demonstration of high-quality, quasi-monochromatic electron beams in 2004, laser wakefield acceleration (LWFA) has become a reliable scheme to accelerate electrons bunches to energies in the GeV range in plasma accelerator stages a few cm long 8-12 . The small scale of the acceleration structure, confining the bunch to a fraction of the plasma wavelength, implies bunch durations in the femtosecond range.Determining the detailed longitudinal profile of the generated bunches is important for understanding the accelerator dynamics, enabling accelerator control, and for determining their potential applications, such as driving compact FELs 13,14 . However, the limited temporal resolution of traditional methods, such as electro-optic sampling 15 , prevents their application to measuring the ultra-short bunches produced by an LWFA.Although recent experiments confirmed the ultra-short nature of LWFA electron beams 1,2 these relied on the assumption of a Gaussian longitudinal profile when determining the electron bunch duration. As in earlier work 2 , we determine the bunch profile from measurements of the spectrum of coherent transition radiation (CTR).However, our experiments advance prior work in several key aspects: (i) the bandwidth of the recorded spectrum covers a spectral range of more than 4 octaves at high resolution; (ii) the spectrum was recorded in a single-shot, preventing shot-to shot fluctuations in the electron bunch parameters distorting the measured spectrum; (iii) the

The application of phase-retrieval algorithms to the reconstruction of the longitudinal bunch profile of an electron bunch from the spectrum of the coherent transition radiation (CTR) it produces is considered. The development of a new algorithm for this application, the Bubblewrap algorithm, is described. Tests with synthetic data show successful reconstruction of the longitudinal profile of single and double electron bunches, provided that the CTR spectrum is known over a sufficiently wide range. The application of the Bubblewrap algorithm to the reconstruction of laser-accelerated electron bunch profiles from experimental data is demonstrated. The results are shown to be consistent with estimates of the bunch length obtained by other methods.

The generation of quasimonoenergetic electron beams, with energies greater than 500 MeV, in a laserplasma accelerator driven by 2.5 J, 80 fs laser pulses guided in a low density plasma channel, is investigated. The laser energy required to achieve electron injection is found to depend strongly on the quality of the input laser focal spot. Simulations show that, although the matched spot size of the plasma channel is greater than the self-focused spot size, the channel assists relativistic self-focusing and enables electron injection to occur at lower plasma densities and laser powers than would be possible without a waveguide.

No abstract

The longitudinal coherence of free-electron laser (FEL) radiation can be enhanced by seeding the FEL with high harmonics of an optical laser pulse. The radiation produced by high-harmonic generation (HHG), however, has a fast-varying temporal profile that can violate the slowly varying envelope approximation and limited frequency window that is employed in conventional free-electron laser simulation codes. Here we investigate the implications of violating this approximation on the accuracy of simulations. On the basis of both analytical considerations and 1D numerical studies, it is concluded that, for most realistic scenarios, conventional FEL codes are capable of accurately simulating the FEL process even when the seed radiation violates the slowly varying envelope approximation. We additionally discuss the significance of filtering the harmonic content of broadband HHG seeds.

No abstract