Measurements Of Optical Beam Quality In An FEL

ABSTRACT Measurements of the optical beam quality of the 3.2 micron output of the Stanford Super­ conducting FEL are reported. Profiles of the far field focus are presented in both time average and single shot cases. The results differ by less than 10% from the predictions for an ideal Gaussian beam. Also presented are measurements of the near field profile and the Rayleigh range.IntroductionFree Electron Lasers (FELs) have been studied for over a decade because of their poten­ tial as high power, tunable, optical radiation sources. An additional often cited advantage is the lack of a material gain medium capable distorting output phase front produc­ ing poor optical beam quality. This is not strictly true since the electron beam itself, particularly in high gain cases, can cause significant perturbations to the cavity mode.1 We report herein the first measurement of optical beam quality of an FEL output.ApparatusThe FEL used to produce the radiation was the Stanford superconducting FEL.2 It has been described extensively in the literature and previous conferences of this series? The FEL utilizes the electron beam of the Stanford Superconducting Accelerator (SCA) at 43 MeV with 1.5A peak in a 3 ps pulse every 84 ns. The accelerator is capable of 90% duty factor al­ though data in this experiment was taken at 10 Hz with a 2 to 16 ms macropulse length (up to 16% duty factor). The FEL is a superconducting helix 5.2 meters long. Approximately 2 watts of output radiation is produced at 3.2 microns during the macropulse. The optical cavity consists of 2 concave mirrors of 7.5 meters radius of curvature each. They are separated by 12.68 meters. The optical waist at the center is 1.66 mm which yields a 2.71 m Rayleigh range. The optical beam expands to an GO of 4.22 mm at the mirrors. From the out- coupler the beam passes through one lens (4.22 m FL) on its way to the beam quality measure­ ment system 5.94 meters away. The lens focuses the beam at a point 30 meters away so that the beam is slightly convergent as it propagates to the measurement system.The measurement system is shown in Figure 1. It was designed for fully remote operation and for wavelengths is the .4 to 10 micron band. The system has the capability of measuring near field profile by the power in bucket technique, far a time aver­ aged sense by scanning the 2.2 m F.L. focussing lens to move the waist across an aperture, measuring Rayleigh range by moving the scanning apertures in Z, and measuring single pulse profiles using an infrared or visible TV monitor. The far field apertures available in­ clude a set of pinholes ranging from 10 to 400 microns, a pair of slits 100 y wide, a 100 micron pinhole XY array for position calibration, and a "wide open" 1 cm diameter hole. XY position of the lens is read out by a pair of linear potentiometers and Z position of the aperture and detector system is readout by a 20 turn helipot attached to the positioning screw. The output of the pyroelectric detector was integrated by a boxcar averager gated to the electron beam pulse using a 1 sec time constant.ResultsThe measurement of the time averaged far field laser profile in the vertical direction is shown in Figure 2. Similar results were obtained in the horizontal case. The measure­ ment was done using the 100 y slit system. While the slits provided a better signal to noise ratio from the detector they do impose a requirement for minor corrections to the pro­ file due to finite slit width.Assuming a Gaussian mode at the focus the wave amplitude is given by