ALS beamline logs record resolving power
The water-cooled, spherical diffraction grating in beamline 9.0.1 of the Advanced Light Source (ALS, Berkeley, CA), seen here prior to installation, offers scientists unprecedented spectral resolution. The beamline was originally designed to produce a photon flux of 1012/s, and ALS scientists expected the spherical-grating monochromator to yield a resolving power E/DE of roughly 10,000. The results of recent studies of the double-excitation state of helium at 64.12 eV performed by researchers at
ALS beamline logs record resolving power
Kristin Lewotsky
The water-cooled, spherical diffraction grating in beamline 9.0.1 of the Advanced Light Source (ALS, Berkeley, CA), seen here prior to installation, offers scientists unprecedented spectral resolution. The beamline was originally designed to produce a photon flux of 1012/s, and ALS scientists expected the spherical-grating monochromator to yield a resolving power E/DE of roughly 10,000. The results of recent studies of the double-excitation state of helium at 64.12 eV performed by researchers at the Freie Universität Berlin (FUB, Berlin, Germany) demonstrate that the actual resolving power is 64,000. This is about four times higher than the peak resolving power previously attained by a synchrotron beamline operating in this energy range.
The grating blanks consist of a polished layer of electroless nickel overlaying a Glidco¥substrate; the 925-lines/mm groove pattern was scribed into the surface by ion beam milling (see Laser Focus World, Feb. 1995, p. 23). Because x-rays experience significant scatter due to surface roughness, the high performance of the 9.0.1 grating is due in large part to the smoothness of the polished blank prior to groove milling.
Metrology performed on spare grating blanks polished at the same time as the 9.0.1 grating showed a surface roughness of 5 Å and an rms slope error of 1.2 µrad. Says Wayne McKinney, mirror scientist for the ALS, "The results [of the FUB group] show that this metrology is an upper limit, because the high resolving power indicates that the real slope error of the blank must be considerably better than measured." Work is underway to calculate the actual surface smoothness of the 9.0.1 grating from the FUB data.
