THE DISTINCTION OF FRAME OVERLAP AND THE COHESION OF TIME-OF-FLIGHT DIFFRACTION PATTERNS COLLECTED AT VARIOUS SCATTERING ANGLES

The possibility of combining time-of-flight powder neutron-diffraction patterns collected at different scat- tering angles is examined, with a view to increasing data-collection rates by the use of a large total solid angle and by including multiple frame overlaps. The basic design features of the required diffractometer are enumerated and a test experiment, using a spec- trometer possessing basic characteristics pertinent to the proposed diffractometer, is used to evaluate these assertions. 1/sin 0 or by arranging the detectors at a constant radius (or other suitable position) from the sample and making appropriate corrections to the TOF patterns. The former method offers only a limited increase in solid angle, even when the major component of the total flight path is extant in the secondary flight path, but has the advantage that a small number of frame- overlapped peaks can be identified by inspection and included in the refinement algorithm. In contrast, placement of detectors at a constant radius from the sample can afford a large solid angle, but patterns from different scattering angles can only be made superimposible for a single spectral frame such that frame overlap cannot be tolerated. This difficulty arises because when detectors are placed at a constant radius, a given reflection shifts to longer TOF with increasing 20, the rate of shift with sin 0 being propor- tional to the d spacing. The contention of the present proposal is that, if we do not attempt to make angle- distinct TOF patterns superimposable, we may con- struct a refinement which effectively uses the angle dependence of peak positions to elucidate the diffrac- ted wavelengths unambiguously and at the same time exploit the total diffracted intensity (including many overlapping spectral frames) over all detectors. Clear- ly, this approach dictates at least as many scattering angles as there are overlapped spectral frames. We will first discuss the distribution of intensity expected from a TOF diffractometer designed to collect data in the manner described above and then, starting from an arbitrary resolution requirement, enumerate the design features of the ideal instrument, comparing its performance with that of conventional CW and TOF diffractometers. Finally, the practical and analytical aspects of our proposals are made evident by a test experiment using an instrument possessing the basic specifications inherent in our approach.