ALBERT

Notes: Single-crystal x-ray diffraction data can be measured very quickly in Laue geometry compared with monochromatic methods. Alternatively, this gain factor can be used instead to reduce the sample volume for a fixed exposure time. In the latter case especially, there is a critical need to control parasitic scatter in the Laue camera. The use of Laue geometry as a means of quantitative data acquisition required the solution of some fundamental problems. The so-called "overlapping orders problem'' has been found not to be limiting. It can be shown that the bulk of the Laue spots are single order, provided dhkl〈2dmin where dhkl is the interplanar spacing and dmin is the resolution limit of the data. In addition, empirical wavelength normalization is required. This can be achieved by using the symmetry of the diffraction pattern. The fact that different equivalents occur at different wavelengths means that the differences in these intensities can be used to establish the "λ curve.'' Successful wavelength normalization to date has used a relatively broad-band pass. The multiplicity distribution is the histogram of the number of spots of a given order. This distribution is determined by the ratio λmax/λmin (λmax =maximum wavelength, λmin =minimum wavelength in the beam). λmax is determined by the use of any filters in the beamline. λmin is determined either by the spectral curve or a critical cutoff if a mirror is used. A mirror can be usefully introduced to enhance the multiplicity distribution in favor of single wavelength spots or to focus the white beam; so far only vertical focussing has been used. The detector options used to date have been photographic film, Fuji image plate (at Photon Factory)/Kodak storage phosphor (at Cornell) and charge coupled device (CCD) (at Daresbury). It is useful to consider the joint theoretical spatial and energy distribution of spots in defining the detector specification and geometry. To date, we have processed Laue film data successfully. The attraction of using the CCD, even to look at a small portion of the Laue pattern, is to view the diffraction in real time.This will allow tight control of parasitic scatter for microcrystal Laue diffraction and real-time monitoring for time-resolved work. We performed initial experiments using a direct detection CCD imager, and have obtained satisfactory diffraction data on a 40 ms time scale. Results of this work will be presented. In order to assess the efficacy of the Laue method for quantitative crystallography, we have used Laue data from the protein pea lectin and compared it in detail with monochromatic pea lectin data. To assess the use of a vertically focussing mirror, we have successfully used a mercury derivative protein crystal to yield isomorphous and anomalous differences suitable for phase determination. In both the pea lectin and mercury derivative cases, doublet Laue spots were deconvoluted. In the latter case, the data were used in a difference Fourier calculation which showed the mercury peak. Future developments and projections based on multipole sources are given.