1.1.1 Small Molecules. In single crystal work on small molecule systems, D19 allows the determination of the location of light atoms amongst heavy ones and the accurate characterisation of liganded H and H₂ and of hydrogen bonding and hydrogen disorder in studies of organic and inorganic molecules, complexes, solvates, and adducts. Such work has been the mainstay of D19 in the 1990’s; examples are given in section 1.2.2 from the fields in chemistry and physics that are most likely to be affected by this proposal.

1.1.2 Larger Molecules. Some larger molecules of biological interest have also been studied on D19, such as vitamin B12 (Jogl et al, 1999), lysozyme (Bouquiere et al, 1993), and haemoglobin (Waller, 1989). The interest here is obvious – information on proton positions and hydration is vital for understanding biological processes. As can be seen from Figure 1, data from monochromatic neutron diffraction studies allows H or D positions to be determined at a much higher level of significance than is obtainable with x-rays. Although it is true that such information can be obtained from high resolution x-ray diffraction when crystals diffract to 1.2Å or better, it is also true that only a small fraction of crystals diffract to this resolution.

Figure 1: "Omit" density maps calculated for a D2O molecule located in the crystal structure of coenzyme B12.  (a) 1A map showing the density determined from the D19 data and (b) 0.9A map showing the density from a synchrotron x-ray study (Langan et al, 1999).

Although neutron experiments on small proteins are at the limit of what can be achieved in reasonable beamtime allocations using the current detector, the use of a monochromatic beam and fine step scanning in omega, with frame readout times as small as 5 or 10 seconds, has clear advantages in situations where the sample background is high (e.g. for samples with high H content where deuteration is either not desirable or not feasible).
 

1.1.3 Fibre Diffraction. The availability of an area detector has also meant that D19 can be used to record high-angle neutron fibre diffraction patterns. The first experiments of this type were on DNA hydration and provided the first information on hydration in polymeric DNA. The studies are also important because fibres allow the study of DNA conformations that have not been observed in oligonucleotide single crystals and also of stereochemical changes that occur during conformational transitions (see Shotton et al, 1997). The same techniques have since been used to study hyaluronic acid, filamentous viruses, and have recently produced some outstanding results from cellulose (Nishiyama et al, in press). Similar methods have been used to study hydrogen atoms in aromatic polymers (Mahendrasingam et al, 1992) and are currently being developed for the study of polymers such as Nylon 66. There is increasing interest from industrial collaborators in the use of fibre diffraction methods in combination with specific deuteration to study changes in polymer structure as a function of chemical composition, temperature, and drawing processes. All of the neutron fibre diffraction studies provide information that cannot be obtained from complementary x-ray fibre diffraction studies.

1.2.1 Biological Systems. For reasons that are evident from section 1.1.2, a gain of a factor of ~25 will have an enormous impact on the scope of biological neutron crystallography. Results comparable to those currently obtained in a two week experiment will be obtained in half a day, and very substantial improvements will therefore be possible in reasonable allocations of beamtime. This will open the way to new projects (smaller crystals, larger units cells) that cannot be considered at the moment.

The implications are equally significant for fibre diffraction studies. This is very simply illustrated in Figure 2 which shows two neutron fibre diffraction patterns recorded from DNA.

1.2.2 Chemistry and Physics.
1.2.2.1 Supramolecular Organic Complexes. Detailed structural information on hydrogen atoms comes mainly from neutron diffraction. As new areas are explored there is a growing demand for such studies eg of N-H…p intermolecular interactions in the aminophenol crystal structures (Allen et al, 1997). Another application is work aimed at understanding how in compounds containing acidic C-H groups, weak C-H...X or C-H...p bonds aid in building up supramolecular arrays. Such compounds, for example alkyl (triphenyl) – phosphonium aryloxides, may contain 100 or more atoms in the asymmetric unit, and at present can be studied only if large crystals are available.

1.2.2.3 Liquid Crystals Studies. One of the aims of studies of liquid crystals is to link liquid crystalline properties to molecular structure. A study of 4-n-pentyloxybenzylidene-4’-n-heptyl quiline on D19 has shown that, unlike other techniques, neutrons can provide information on both molecular structure and orientation distribution by exploiting the difference in scattering from H and D (Richardson et al, 1990). The quality of the data was limited by the time available for data collection; one day for each H/D mixture.

1.2.2.4 Quasicrystals Of the 30 or so known quasicrystalline compounds, many are of technological interest, e.g. as hard, high-temperature resistant substitutes for teflon. Models of the atomic structure are not as accurate as for conventional 3-d periodic structures, partly because, on a conventional diffractometer, only a small proportion of the available number of reflections in a given volume of reciprocal space can be observed. Recent measurements on Al₇₀Pd₂₀Mn₁₀ on D19 have for the first time allowed whole volumes of reciprocal space to be measured, giving over five times the number of unique reflections previously recorded with a dynamic range of more than four orders of magnitude. Because of the 6-d lattice needed to describe quasicrystals, a monochromatic beam is vital. A related application is observation of diffuse scattering and incommensurate satellites, as done for the decagonal phase of Al₆₅CO₁₅Ni₁₅ (Frey et al, in press). Only a high-quality large area detector will enable such measurements to be extended

We propose to replace the D19 detector by an array of eight 2-dimensional microstrip PSDs. Each module will be 230 x 230 mm with a detection area of 192 x 192 mm, a resolution of 3mm, and an efficiency of 50%. The eight detectors will be mounted on a curved surface of either 60cm or 90cm radius (see Figure 3). This relatively large maximum radius is chosen to increase the signal to incoherent background ratio for hydrogen-containing samples. At this radius the detector array subtends an angle of 34.4° x 34.4° at the sample. When the incoherent background is small enough, the distance from the sample can be reduced to 60cm, in order to increase the number of reflections striking the detectors. At 60cm the angle subtended at the sample will be 51.4° x 51.4°. The projected gain in efficiency over the present D19 is a factor of up to 25. This is greater than the increase in solid angle (4.63 at 90 cm and 10.33 at 60 cm) because of large edge effects in the present narrow detector.

References 1. Allen, Hoy, Howard, Thalladi, Desiraju, Wilson & McIntyre, J.Am.Chem.Soc. 119, 3477 (1997); 2. Bau, Mason, Li, Wong, J. Amer. Chem. Soc., 119, 11992-3 (1997); 3. Bouquiere, Finney, Lehmann, ,J. Chem. Soc. Far. Trans. 89, 2701 (1993); 4. Frey, Hradil, Grushko, McIntyre, Quasicrystals, World Scientific (in press); 5. Jogl, Langan, Kratky (in preparation); 6. Langan, Denny, Mahendrasingam, Mason, Jaber, J. Appl. Cryst. 29, 383 (1996); 7. Langan, Myles, Lehmann, Wilkinson & Mason, ILL Internal Report ILL97LA01T (1997); 8. Langan, Lehmann, Wilkinson, Jogl, Kratky, Acta Cryst D55, 51 (1999); 9. .Mahendrasingam, Al-Hayalee, Forsyth, Langan, Fuller, Oldman, Blundell, Mason, Physica B 180/181, 528-530 (1992); 10. Muller, Williams, Doerrer, Leech, Mason, Green, Prout, Inorg. Chem. 37, 1315-1323 (1998); 11. Nishiyama, Y., Okano, T., Langan, P. & Chanzy, H., Int. J. .Biol. Macr. (in press); 12. Richardson, Allman, McIntyre, Liquid Crystals7, 701-719 (1990); 13. Shotton, Pope, Forsyth, Langan, Giesen, Dauvergne, Fuller, Biophysical Chemistry69 (1), 85 (1997); 14. Shotton, Pope, Forsyth, Archer, Denny, Langan, Ye, Boote, J. Appl. Cryst.31, 758 (1998); 15. Velletaz, Oed, Assaf, Proc. International Workshop on Microstrip Gas Chambers (Eds. Contara & Sauli), Lyon (1996); 16. Waller, PhD thesis, University of York (1989); 17. Yvon, Minutes of the ILL Sientific Council, April 1998.

We thank Paul Langan, Michael Walsh and Garry McIntyre for their contributions to this project. We are also extremely grateful for the ongoing efforts of the ILL detector group, without which the use of microstrip modules in the proposed new D19 array will not be possible.