Abstract

Although considerable interest has been paid recently to Raman Stokes shifting in hydrogen, very little attention has been paid to nonlinear upconversion via anti-Stokes shifting. This disinterest is due largely to low experimental yields (typically <2%) caused by Stokes-anti-Stokes gain suppression when the laser beam is focused into the hydrogen cell. The fact that the anti-Stokes light emerges from the cell in a cone also detracts from its usefulness. By contrast, our anti-Stokes work uses collimated input light so that the emerging anti-Stokes wave is a filled-in propagatable beam. By choosing the conditions at the input of the Raman cell appropriately, we have demonstrated 10% conversion of KrF laser light at 248 nm to an anti-Stokes beam at 225 nm. We found that the single most important condition to obtaining good efficiency is to coinject into the hydrogen cell a phase-matched Stokes beam that has ~5 % of the energy of the 248-nm pump beam. This technique should be applicable to pumping with other lasers as well. The effects of beam quality, transient response, and other experimental conditions are discussed.

© 1988 Optical Society of America