The grating accelerator is one of the schemes that has been proposed for utilizing the enormous peak power available from lasers to accelerate electrons to high energy. In principle the accelerating gradient could be much larger than what is presently available with RF linacs. The basic reference to this idea is R.B. Palmer, A laser-driven grating linac, Part. Acc. 11:81-90, 1980. The BASIC PRINCIPLE is to use the periodic grating surface as a structure to support longitudinal, accelerating field components. The source of these fields is the incoming, focused electromagnetic wave from the CO2 laser beam. Cylindrical optics are used to produce a line focus along the electron beam direction. The electron beam transverse dimensions must be small compared to the wavelength of the laser radiation (10 fm) over the entire interaction region. Various TYPES OF STRUCTURES will be investigated in the experiment. Besides flat, 2-D gratings, we will examine the accelerating properties of "foxhole" and "colonnade" structures. These structures have been manufactured using silicon etching with metallic coating. They can also be made from LIGA techniques. We could also measure the accelerating fields on the vacuum side of a crystal undergoing total internal reflection. The proof of principle experiment will use a 50 MeV beam from the ATF with a normalized emittance of 2x10-8 m-rad. This should produce a 1 fm beam radius at the experiment and give a 3 mm interaction length. A laser power of 1 GW should produce an accelerating field greater than 1 GeV/m on the grating surface. This would produce an energy gain of 3 MeV.

This experiment was formally withdrawn in September 1996.