Superradiant Cascade in a Seeded Free-Electron Laser

We report measurements demonstrating the concept of the free-electron laser (FEL) superradiant cascade. Radiation (${\lambda }_{\mathrm{rad}}=200\mathrm{nm}$) at the second harmonic of a short, intense seed laser pulse (${\lambda }_{\mathrm{seed}}=400\mathrm{nm}$) was generated by the cascaded FEL scheme at the transition between the modulator and radiator undulator sections. The superradiance of the ultrashort pulse is confirmed by detailed measurements of the resulting spectral structure, the intensity level of the produced harmonics, and the trend of the energy growth along the undulator. These results are compared to numerical particle simulations using the FEL code GENESIS 1.3 and show a satisfactory agreement.

Figure 1

Undulator layout of the FEL cascade at SPARC. The gap of each undulator section can be set independently between 8 and 25 mm, so that the undulators may be tuned at ${\lambda }_{\mathrm{seed}}=400\mathrm{nm}$ (corresponding to $K=2.19$) as modulator, or at ${\lambda }_{\mathrm{rad}}=200\mathrm{nm}$ ($K=1.18$) as radiator.

Figure 2

Growth of the radiation energy at ${\lambda }_{\mathrm{rad}}=200\mathrm{nm}$ along the radiator coordinate $z$. Black squares: experimental data. Red stars: simulation with GENESIS 1.3. Blue solid line: $a+b(z-z_0{)}^{3/2}$ analytical fit, with $a=0.096\mu \mathrm{J}$, $b=3.9\times {10}^{-2}\mu \mathrm{J}/{\mathrm{m}}^{1.5}$ and $z_0=5.12\mathrm{m}$. Electron beam parameters of Table  (B). Data averaged over 100 shots. Error bars represent one standard deviation.

Figure 3

Single shot spectra-spatial measurements along the radiator at $\lambda =200\mathrm{nm}$ for the beam of Table  (B). The vertical axis on the images of the spectra represents the vertical position on the input spectrometer slit. The images were taken after the third (second section of the radiator), the fourth, and the fifth section, and at the end of the undulator, respectively from the top to the bottom.

Figure 4

Radiation energy $E(\mu \mathrm{J})$ vs bandwidth $\Delta {\lambda }_{\mathrm{rms}}(\mathrm{nm})$ at different position along the undulator. The statistics of 100 consecutive single shots for beam B is shown, except after 1 UM Rad (black dots) where averages of 10 shots over 1 s are plotted.

Figure 5

Second, third, fourth, and fifth harmonics of the radiator for beam (B). Spectra-spatial measurements at the end of the radiator obtained averaging over a time of 20 s (30 s for the 5th harmonics). In the lower boxes, intensity of the radiation spectrum vs wavelength. Same color palette as Fig. 3. The third to fifth spectra were acquired by interposing an aluminum (Al) filter to reduce the background noise due to lower harmonics at the zero order of the grating. The horizontal lines are due to the presence of a mesh sustaining the Al layer.

Figure 6

Energy at the fundamental and higher order harmonics in the radiator. The energy of the second harmonics is $E=2.6\pm 0.7\mathrm{nJ}$, of the third: $E=3.3\pm 1.1\mathrm{nJ}$, of the fourth: $E=0.03\pm 0.01\mathrm{nJ}$, and of the fifth: $E=0.17\pm 0.06\mathrm{nJ}$.