Observation of Monoenergetic Electrons from Two-Pulse Ionization Injection in Quasilinear Laser Wakefields

The generation of low emittance electron beams from laser-driven wakefields is crucial for the development of compact x-ray sources. Here, we show new results for the injection and acceleration of quasimonoenergetic electron beams in low amplitude wakefields experimentally and using simulations. This is achieved by using two laser pulses decoupling the wakefield generation from the electron trapping via ionization injection. The injection duration, which affects the beam charge and energy spread, is found to be tunable by adjusting the relative pulse delay. By changing the polarization of the injector pulse, reducing the ionization volume, the electron spectra of the accelerated electron bunches are improved.

Figure 1

Schematic of the experimental setup in the target chamber. The drive beam path is depicted as a dashed black line. The annular injector beam path is depicted as a green line. Both pulses travel through the gas cell from the right-hand side to the left-hand side. The red area in the gas cell indicates the volume of injection. The blue dashed lines represent possible electron trajectories.

Figure 2

Data obtained from the Lanex screens measuring the electron spectrum (top, linearly polarized injector pulse; bottom, circularly polarized injector pulse). Panels (a) and (b) show the energy values of the electron charge density peak(s) plotted against the relative delay between pulses. Positive values correspond to a trailing injector pulse. The dashed lines show the mean value of each electron bunch type. For clarity, error bars corresponding to the calibration curve are not shown except for one sample shot. The maximum uncertainty values are $-8.1\%$ and $+2.8\%$. Panels (c) and (d) show the corresponding maximal FWHM energy spread plotted against pulse delay. Panels (e) and (g) show the number of successful and unsuccessful shots at each delay value. Panels (f) and (h) present the ratio of successful shots over the total number of shots at each delay position. Panels (i) and (j) show typical spectra of single (i) and double (j) bunches as observed on the two Lanex screens from shots using a circularly polarized injector pulse (Run II). These two shots were taken at a delay of $\mathrm{\Delta }z=19\mathrm{\mu }\mathrm{m}$.

Figure 3

Results from two-dimensional particle-in-cell simulations. Left: Longitudinal electric field strength and electron density from the sixth and seventh nitrogen ionization levels (above $n_e=2\times {10}^{22}{\mathrm{m}}^{-2}$). Right: Kinetic energy associated with the same electrons. The separation in z between the pulses is ${\lambda }_p$ ($15.8\mathrm{\mu }\mathrm{m}$) (a), $0.75{\lambda }_p$ ($11.9\mathrm{\mu }\mathrm{m}$) (b), $0.5{\lambda }_p$ ($7.9\mathrm{\mu }\mathrm{m}$) (c), $0.25{\lambda }_p$ ($4.0\mathrm{\mu }\mathrm{m}$) (d), 0 (e), and $-0.5{\lambda }_p$ ($-7.9\mathrm{\mu }\mathrm{m}$) (f) where positive values indicate that the drive pulse is ahead.