The new generation of synchrotron machines.

Abstract

ESRF is committed to building 30 beamlines for user operation. Four of these will be fully or partly devoted to macromolecular crystallography. Support facilities for structural biology will be provided by the European Molecular Biology Laboratory outstation in Grenoble which is situated close to the synchrotron ring. Beamtime will be allocated by peer review with a first application deadline of March 1, 1994. (Application forms and a Beamline Handbook are available from ESRF on request.)In September 1994 the crystallography endstation of the high brilliance beamline, BL4, will be ready for user operation. This undulator beamline will provide a flux of 1012 photons per second of monochromatic radiation through a sample cross section of 100 μm × 300 μm. The optical elements will be optimized for experiments using a fixed wavelength around 1 a, but a wavelength range between 0.7 and 1.55 a will be available. The endstation is designed to handle crystals with unit cell dimensions up to 1200 a as well as weakly diffracting small crystals. Data collection to very high resolution ( < 0.9 a) from crystals of smaller unit cells will also be possible. Using an undulator beam similar to that of BL4, it has been shown that the percentage of fully recorded reflections from virus crystals is very high due to the low divergence of the undulator beam (D Stuart and co-workers, unpublished data).A wiggler beamline for white beam Laue experiments, BL3, will also be ready in September 1994. The exposure time of a Laue pattern from a protein is expected to be in the range of 1 μsec to 1 msec using the multi-bunch mode (992 bunches separated by 2.8 nsec). In the single bunch mode, the focused beam will deliver 3 ×1010 photons per bunch, which is sufficient flux to observe the strongest 15–20 % of the reflections from a protein the size of myoglobin. The beamline is equipped with mechanical shutters which can isolate the exposure from a single bunch. In addition, there will be equipment for cryogenic cooling and a nanosecond laser for reaction initiation.BL19, which is a bending magnet beamline designed for MAD experiments, will come into operation in September 1995. The optical elements have been designed to give a nominal band pass of 1.6 ×10−4 in the wavelength range 0.6–2.0 a. The flux on the sample will be about 1012 photons per second which will allow diffraction patterns to be recorded in 10–20 seconds per degree of sample rotation for ‘typical’ protein crystals. This beamline can of course be used not only for MAD experiments but also for single wavelength studies.Preliminary experiments at ESRF (J Als-Nielsen, A Freund and G Gruebel, unpublished data) have shown that monochromatic beams obtained from thin diamond crystals operating in the Laue transmission mode on an undulator beamline are useful for macromolecular crystallography. Several such transparent monochromators can be arranged along the same beam to supply independent experimental stations simultaneously with monochromatic photons of slightly different energy. BL20, which will be in operation at the end of 1995, will use this concept to provide a high throughput facility for macromolecular crystallography. Three different experimental stations are planned on this beamline, each with the data collection capacity of a ‘conventional’ undulator beamline and each operating independently of the others. One of these will be able to collect data at very short wavelengths, around 0.3 a.In addition to these four dedicated beamlines, there is at least one more that is of substantial interest for the Structural Biology community, the Microfocus Beamline, BL1, which will be in operation in September 1994. This beamline will reach a focal spot of 10 μm ×10 μm by demagnifying an undulator source point by mirror optics. Further demagnification to sub-micrometre levels is achieved by Bragg-Fresnel optics. This beamline will be useful for exploring diffraction from objects such as microcrystals and individual doamins of single fibres or fibrils.Most of the bending magnets will be used by Collaborating Research Groups (CRGs), from which the ESRF user community will obtain one-third of the available beamtime. Two such CRG beamlines, one French and one Swiss-Norwegian, will have experimental stations for macromolecular crystallography. The beamtime available for the ESRF on these beamlines will be allocated by the same peer review system as for the ESRF beamlines.