Fig. 1. Schematic layout of the experiment. An 800-MeV electron beam is sent to the first chicane and interacts with a 266-nm seed laser in the first dipole magnet. Energy modulation and density modulation are performed simultaneously in the first chicane. In the modulator, an electron beam is used to generate coherent radiation at the fundamental wavelength, which also enhances the energy modulation. The radiator is used to produce FEL pulses at the sixth harmonic of the seed laser.

Fig. 2. Three-dimensional tracking of the laser–beam interaction. Plotted are the energy modulation amplitude of the electron beam (dashed line) and bunching factor (solid line) as a function of the dispersion strength of the first magnetic chicane.

Fig. 3. Experimental characterization of the laser–beam interaction in the dipole magnet. (a) The longitudinal phase space of the electron beam after the interaction. The red dashed line represents the central energy of the electron beam. The orange box contains areas that are altered due to the laser–beam interaction. The beam head is on the left. (b) The measured coherent radiation intensity and fitted curves after the electron beam passes through the first chicane under different laser pulse energies. The coherent radiation intensity is recorded by a photodiode when the dispersion strength of the first chicane is scanned. The three optimal R56 obtained were 0.95, 1.02, and 1.08 mm when the seed laser pulse energies were 38.10, 18.30, and 10.60μJ, respectively. (c) Calculation results of the energy modulation amplitude and the initial slice energy spread using the coherent radiation generation method.³⁰

Fig. 4. Performance of the FEL lasing at the sixth harmonic of the seed laser. (a) The gain curve and typical transverse profile of the FEL pulse at 44.33 nm. The red and blue points represent the average pulse energy and maximum pulse energy at the end of various undulator segments, respectively. The FEL pulse energy was measured by a calibrated photodiode at the end of the undulator section. The error bars represent the root-mean-square intensity fluctuations. The inset displays one typical transverse profile of the FEL pulse. (b) The typical longitudinal phase space of the electron beam at the exit of the radiator.

Fig. 5. Spectra and reconstructed temporal profiles of 44.33-nm FEL pulses: (a) 10 typical spectra measured by the spectrometer after the radiator and (b) power profiles of 10 typical FEL pulses reconstructed using the XTDS system.