Free Electron Laser





The FERMI free-electron laser (FEL) at Elettra-Sincrotrone Trieste is an international user facility for scientific investigations, producing ultra-short, high-brilliance, XUV pulses. The FEL process is triggered by an external seed laser, see picture on the right. This allows one to obtain full coherent pulses with stable central wavelength. The FEL output is tunable in wavelength, power, temporal duration and polarization.





The user stations can receive light from two FEL lines: FEL-1, which covers the wavelength range 100-20 nm, and FEL-2, which covers the range 20-4 nm.

FEL-1: A single-stage high-gain harmonic generation FEL

The seeding scheme on which FEL-1 relies is shown in the following picture. A seed pulse, provided by a conventional, high peak-power pulsed laser, operating at 260 nm, is temporally synchronized to overlap the electron beam produced by the linac accelerator. The laser-electron interaction takes place in a first undulator section, called the modulator (MOD). The laser field modulates the energy of the transversely wiggling electron beam. The energy modulation is then converted into spatial density modulation, when electrons cross the magnetic field generated by a chromatic dispersive section.
The charge density modulation occurs at the fundamental seed wavelength, as well as at its higher harmonics. A subsequent set of undulator sections, named the radiator (RAD), is tuned so that intense, coherent FEL radiation at a wavelength corresponding to one of these harmonics is emitted and then amplified to a high peak power level. The temporal duration of the FEL pulse is determined by that of the seed laser. The polarization properties can be tuned by changing the phase of the magnetic field generated by the radiator sections.


The output FEL pulse can be combined with a synchronized external laser, allowing users to carry out pump-probe experiments.   
The FEL-1 setup also allows to implement two-pulse schemes, in which both the pump and the probe are FEL pulses. This can be obtained:
  • By seeding the electron beam with two laser pulses and tuning the radiator sections at a given harmonic. This allows one to generate two FEL pulses with slightly different wavelength (up to 1% difference), separated by controllable temporal distance (up to few hundreds of femtoseconds). A fine tuning of the delay between the seed pulses allows one to generate mode-locked FEL pulses, with tunable phase difference.
  • By seeding the electron beam with two laser pulses and tuning the radiator sections at two different harmonics. This allows one to generate two FEL pulses with significantly different wavelength (more than 10%), separated by controllable temporal distance (up to few hundreds of femtoseconds);
  • By seeding the electron beam with one seed pulse and tuning the radiator sections at two different harmonics. This allows one to generate two FEL pulses with significantly different wavelength (more than 10%), with no temporal separation.

FEL-2: A double-stage high-gain harmonic generation FEL


In order to cover a shorter wavelength range (20 - 4 nm), FEL-2 is based on a double-stage high-gain harmonic generation scheme, see the following picture. The first stage is similar to the FEL-1, (with a shorter radiator). The  second-stage modulator is tuned to use the output from the first stage radiator. The energy-modulated electron beam is then sent through a second dispersive section and finally injected into a second (long) radiator,  tuned at one the harmonics of the first stage output.  FEL-2 also utilizes a magnetic delay line between the two stages, in order to seed, in the second stage, a “fresh” portion of the electron beam.




The need to use a fresh portion of the electron bunch in the second-stage radiator does not allow one to implement two-pulse schemes in which the electron bunch is seeded by two laser pulses. However, two-colour FEL pulses (with no temporal separation) can be obtained by tuning the undulator sections of the last radiator at two different harmonics of the light emitted by the first stage.
FERMI FEL-2 can be also operated in the so-called super-radiance mode, which allows one to generate transform-limited few-fs pulses at some specific wavelength (λSR).  In this configuration, pump-probe experiments can be performed, either with a synchronized optical laser (as for standard FEL-1 and FEL-2 configurations), or with the XUV pulses (expected to be slightly longer than the super-radiant pulse), produced during the generation process, at a multiple of λSR.
More information about FEL-1 and FEL-2 can be found
in the Conceptual Design Report.

Back to Machine Page

Last Updated on Thursday, 12 November 2020 13:59