photocathode RF gun, (2) suppression of space-charge effects between the gun and the high-energy portion of the linear accelerator, and (3) wake-field effects in both the longitudinal and transverse dimensions of the electron beam. (The wake fields, generated by currents in the accelerator structures induced by the leading portion of the electron beam, break up the trailing portion of the pulse.) Other concerns include (4) effects from the tails of the electron beam distribution, (5) electron bunch compression, and (6) the development of a cost-effective package of diagnostics and feedback systems to stabilize all the beam parameters.

As one goes to shorter and shorter wavelengths, the problems mentioned above become more difficult to solve. Many of these technical problems could be addressed by research being done for the Next Linear Collider (NLC), since the electron-beam parameters needed for the NLC will be more demanding than those necessary for an x-ray FEL. Alternatively, the development of an x-ray FEL might be a precursor to solving the technical problems of the NLC.

The construction and operating costs of an x-ray FEL will be much higher than those of an infrared FEL. For example, the incremental construction cost, based on the assumption that the last third of the 2-mile-long SLAC linear accelerator would be dedicated to an x-ray FEL, is estimated by SLAC to be $30M to $50M. A new, dedicated accelerator built for an x-ray FEL would cost hundreds of millions of dollars. One third of the running cost of SLAC is $13.8M for a 9-month period. This cost includes electricity, maintenance of the accelerator, salaries of the operating and support staff, and 40% indirect costs. It does not include the cost of the scientific programs, salaries of staff scientists and accelerator engineers, general plant costs, and accelerator and facility upgrades.


It is hard to anticipate all of the scientific opportunities made possible by the additional orders-of-magnitude increase in coherence over that of the APS. Increases in peak power and decreases in pulse width of several orders of magnitude are anticipated. An x-ray FEL, especially one operating in the hard x-ray portion of the spectrum, could be an important tool for probing distances at the atomic scale. Currently, the overwhelming bulk of x-ray work is done in the 1-to 1.5-Å range.

Experiments that use the transverse coherence of x-ray sources are just beginning. In the current culture among x-ray scientists, coherence properties are ignored because, before the advent of the APS, sources with adequate coherence were not available. Similarly, the scientific community has not yet identified dramatic uses of the very short pulses of the proposed FEL source. Although synchrotron x-ray radiation comes in pulses on the order of 150 ps, the vast majority of the work at these machines is not time resolved. Flash x-ray microscopy and measurements of transient lattice distortions or melting are experiments that one might do with short-pulse x-ray sources, but many of these experiments can also be done at synchrotron sources.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement