to 1 ms achieved in the single bunch mode. A flux of roughly 1015 to 1016 photons per second is available in a 0.1% bandwidth and in a beam of angular divergence of 5 mrad in an energy range extending from 10 eV to 10 keV. The spectral brightness of undulators makes it possible to obtain core-level photoelectron spectra limited only by the intrinsic lifetime broadening (typically 3 × 10−4 of the binding energy for the sharpest levels). These sources are also essential for element-resolved microscopy at a resolution of 0.1 mm or better.
There are no existing FEL sources in the wavelength range from 200 to 10 nm. The FELs at Laboratoire pour l'Utilisation du Rayonnement Electromagnetique (LURE) and Novosibirsk have produced light with wavelengths as short as 250 nm and 240 nm (about 5 eV), respectively. Various proposals would extend the wavelength range down to 75 nm. At very short wavelengths the problem of finding suitable mirrors becomes a limitation on the use of oscillators, and a transition to high-gain, single-pass amplifiers (using SASE as described in Chapter 2) may be needed. The wavelength range down to 100 nm overlaps with that of conventional lasers using harmonics, but FELs would have superior peak power and average power.
For the sake of simplicity, the range of wavelengths from 200 nm to 10 nm is labeled as the VUV. One proposal suggests a system operating from 300 to 75 nm at a repetition rate of 360 Hz with pulse durations from 6 ps to 200 fs and pulse energies of roughly 1 mJ at 100 nm, corresponding to roughly 400-MW peak power. The key technical challenge of a VUV FEL is the achievement of high electron-beam quality along with high peak current.
An FEL oscillator with normal incidence mirrors may be possible down to about 100 nm using clean Al mirrors in ultrahigh vacuum (Kortright, 1990). From 100 nm to about 60 nm, SiC mirrors would permit lasing using an FEL with a gain of 10. Down to 10 nm, the FEL oscillator would have to rely on hypothetical multilayered mirrors with reflectivity of about 50%, although such mirrors might restrict the tunability. Even if the reflectivity of a material is sufficient for laser operation, thermal distortion and damage may prevent or severely restrict the use of mirrors. This is an area that would benefit from further research and would be important for an FEL operating below 100 nm.
It appears that a VUV FEL would require further research and development. The components for this wavelength range have been separately demonstrated in laboratory experiments, but no working device has been constructed. The proposals