VISA - Talks at the August 2001 Meeting at SLAC

A meeting to review and discuss the experimental results and data analysis of the VISA Free-Electron Laser experiment was held at the Stanford Linear Accelerator Center (SLAC) on Thursday, August 9, 2001 in the Training Center r oom C/D. At the core of the meeting were technical presentations describing the acquisition and analysis of data taken at the ATF VISA system after the last VISA meeting that took place on April 30, 2001.

The initial measurements of gain above 10⁷ and gain length on the order of 19 cm could be repeated and confirmed.

The ATF gun has not been able to produce the electron beam parameters proposed in the original VISA design. Data analysis shows that VISA achieved its desi gn goal because of strong non-linear bunch compression (by a factor of 10) of the energy-chirped electron beam in the beam transport line from the linac to the undulator. The key measurement for confirming this result was based on the analysis of Optical Transition Radiation from the electron beam as it left the undulator using a Golay cell and submillimeter wavelengths filters. The chirping was achieved by off-crest acceleration in both ATF linac sections. The compression increased the peak current f rom 55 A at the end of the linac to 250 A at the entrance to the undulator. The beam dynamics was simulated from the cathode to the linac exit with UCLA-Parmela, and from the linac exit to the undulator entrance with M. Borland's code ELEGANT.

The simulations show that there is a rather strong effect from Coherent Synchrotron Radiation (CSR) that increases the horizontal emittance to about 7 mm mrad, with a core having about 4 mm mrad. The off-crest acceleration also changes the center energ y of the beam and thus its horizontal position in the dispersive section, causing the longitudinal bunch tails to be scraped at the narrow aperture of +/- 1.5 cm at the location of one of the ATF beamline switching magnets. The combination of bunch rotation and scraping produces a compact beam distribution in longitudinal phase space at the entrance to the undulator. This kind of dispersive scraping caused the positive effect of correlating charge with beam energy (or linac phase), thus enabling filte ring of the data with respect to linac phase, and allowing to combat one of the experimental problems, the linac phase fluctuations.

Using the particle distribution, created by the PARMELA and ELEGANT codes one can reproduce the results of the Golay cell measurements. Using the same particle distribution, S. Reiche is able to predict, using his FEL simulation code GENESIS 1.3, the measured radiation intensity correctly without any free parameters. This is the first time that this kind of a greement of SASE-FEL experimental results and cathode-to-FEL simulation codes has been achieved.

These start-to-end simulations indicate that the slice emittance of the electron beam after the linac is 0.8 mm mrad in both planes, at a charge of about 200 pC. As mentioned above, the horizontal emittance is spoiled by CSR in the transport line while the vertical emittance stays more or less unchanged during beam transport. This result supports the notion that the parameters of the ATF gun are sufficient for LCLS operation.

Further analysis is required with regard to the spectral data. At the saturation point, the measured spectrum of the fundamental wavelength differs from simulation results, i.e., the simulated spectra are somewhat narrower and contain side bands, while the measured spectrum consists of a single broad peak, only. A large number of spectra have been taken at different ports and at different gain levels and these will be analyzed in the near future.

2:00-2:05 PM	Introduction	C. Pellegrini
2:05-2:35 PM	Overview of the Experiment	A. Tremaine
2:35-3:05 PM	Diagnostics, data taking and analysi s	A. Murokh
3:05-3:35 PM	Start to end simulation and comparison with data I	J. Rosenzweig
3:35-4:00 PM	Start to end simulation and comparison with data II	S. Reiche
4:00-5:00 PM	Discussion

Summary

Talks