journal menu
![[Figure 7]](ge5117fig7mag.jpg)
|
|
Figure 7
Diffraction pattern measured (symbols with error bars which represent ± one standard deviation and a corresponding confidence interval of approximately 68%) from an 8 µm-period π-phase-shift grating with equal-thickness rectangular troughs and columns etched in single-crystal silicon at a neutron nominal wavelength of 5 Å. Also plotted in this figure is a calculated diffraction pattern based on the phase-grating formula of equation (9) ![[link]](../../../../../../logos/arrows/j_arr.gif){kind=small} and assuming an incident illuminating beam exactly as described in the preceding section for the pair of slits which resulted in the profile of Fig. 5 (the phase grating was located 495 mm away from the second downstream slit) – that is, both geometrical and diffraction effects in forming the beam of wave packets incident on the grating by the pair of slits were taken into account. In the computed phase-grating diffraction pattern, both the distribution of geometrical angles in the incident beam and the transverse dimension of an individual neutron packet wavefront were included. The model calculation was not a fit to the data but only scaled to the measured intensity. The best agreement between the data and model was obtained for N = 3 and for a slight curvature of the grating substrate amounting to about 2.65 × 10−5 rad (5.47′′). (This bending might alternatively be attributed to a curvature of a neutron packet wavefront – which was originally taken to be perfectly flat but limited to a 24 µm finite lateral extent.) The general agreement between measurement and model calculation is good, although details in the wings are not resolved. This is likely to be due to relatively small effects involving mirror reflection, refraction and diffraction from the mask edges of the slits defining the incident beam. |
![[Figure 7]](ge5117fig7.jpg)
 | JOURNAL OF APPLIED CRYSTALLOGRAPHY |
ISSN: 1600-5767
access
