In a free electron laser, coherent radiation is generated by letting an electron beam propagate through an alternating magnetic field. The magnetic field is created by a linear array of magnets, which is called an undulator or a wiggler. The wavelength of the laser radiation depends on the amplitude and wavelength of the magnetic field, and on the energy of the electrons. For a short-wavelength or high-power free electron laser, an electron beam with a low beam divergence, low energy spread and a high peak current is required. At the present time, the only way to generate such an electron beam is by means of a photoinjector. In such a device, an electron beam is created by illuminating a photo cathode with intense pulses of light from a drive laser system. The electron bunches that are emitted from the photoemissive material are accelerated in a radio-frequency linear accelerator. TEU-FEL is a free electron laser operated by the Nederlands Centrum voor Laser Research in close collaboration with the University of Twente. The electron beam in TE-FEL is generated in a photoinjector. This set-up is the subject of this thesis. One of the most critical components of a photoinjector is the drive laser system which illuminates the photo cathode. The Nd:YLF system that is used at TEU-FEL is described in Chapter 2. It emits trains of laser pulses, called macropulses, that consist of up to 1200 micropulses. The repetition rate of the macropulses is 10 Hz, the micropulses repeat at a frequency of 81.25 MHz. The research focused on the stability of the system output power during a macropulse. A feed forward loop was constructed that can optimise and stabilise the shape of the output macropulse within 1 %. Most photo cathode materials are not sensitive to the fundamental of Nd:YLF (1053 nm). It is therefore necessary to generate the second or fourth harmonic of the Nd:YLF laser. A walk-off compensated frequency quadrupling scheme has been designed, which yields an energy conversion efficiency of 25 % from the fundamental to the fourth harmonic. The duration of the infrared and visible micropulses has been measured with autocorrelation techniques and with a streak camera. Chapter 3 treats several aspects of photo cathode materials. A photo cathode material is characterised by its quantum efficiency (QE) which is defined as the number of emitted electrons per incident photon. Several photo cathode materials have been used in the linear accelerator, most notably alkali-antimony and alkali-tellurium compounds. The former are mainly sensitive to visible radiation and have a typical QE » 1 %; the latter are sensitive in the ultraviolet region of the spectrum and have a typical QE » 10 %. Two previously unknown photoemissive materials have been discovered. These materials are compounds of potassium and tellurium, and of cesium, potassium and tellurium. The latter shows a QE = 23.4 % in the ultraviolet region of the spectrum. Due to pollution and outgassing of alkali-metals, the QE of photo cathodes teadily decreases when they are used in the linear accelerator. It was found that the alkali-telluride-compounds have significantly larger lifetimes under operating conditions than their alkali-antimonide counterparts. The quality of the electron beam can be characterised by the emittance, which is a measure for the angular spread of the electrons. The lower the emittance, the better the quality of the beam. Several methods exist to measure this beam quality number. In Chapter 4, a comparison is made between the quadrupole scan, the pepper pot technique and phase space tomography. The longitudinal phase space of the electron beam is characterised by means of treak camera measurements and longitudinal phase space tomography. It is demonstrated that the duration of the electron bunch depends on the bunch charge. At a charge of 1.5 nC the bunch length is 38 ps and the energy spread is 0.4 %. These measurements are combined with the emittance measurements to yield a value for the brightness of the electron beam. A comparison of the brightness with that obtained with other accelerators shows that the electron beam that is used in the TEU-FEL experiment is one of the brightest beams in the world. In Chapter 5, lasing of TEU-FEL is demonstrated by simultaneously measuring the electron energy spectrum and the free electron laser output power. An increase in output energy is accompanied by a decrease in mean electron energy and a profound increase in energy spread. This indicates a energy transfer from the electron beam to the laser beam, thus confirming laser operation. The effect of de-tuning the free electron laser cavity has been investigated; the desynchronism curve is seen to be asymmetric, a result that is in agreement with that found by other workers in the field. Finally, the wavelength spectrum of the free electron laser is measured by making a Fourier transform of the results of a Michelson interferometer scan. The spectrum shows three prominent peaks (see cover) spaced approximately 9 mm apart. The occurrence of these different wavelengths is attributed to the difference in phase velocity between the modes in the resonator waveguide. There is good agreement with the theoretically predicted wavelength difference.
|Award date||19 Jun 1997|
|Place of Publication||Enschede|
|Publication status||Published - 19 Jun 1997|