Full text loading...
Images of test samples made with the pinhole set-up and focusing-mirrors microscope. The samples are: a planar grating (G0) made of about 5 μm-thick neutron-absorbing Gd deposited on quartz with a period of 796 μm and a duty cycle of 40% (see Ref. 29 ), and the array of 100-μm-diameter holes, generated from MURA, see Ref. 30 . (a) Pinhole image of G0; 5 min measurement time. (b) Magnified image of G0; 5 min measurement time. The white line shows the region where the cross-section was taken. (c) Cross-section of the image in (b), taken along the white line. (d) Magnified image of MURA; 15 min measurement time. The inhomogeneity in illumination in the microscope images results from the 2-m-long separation between the end of the neutron guide and the sample. Therefore, the beam is excessively collimated, such that some of the neutrons illuminating the samples do not reach the reflecting mirrors. This effect is avoided when the beam has enough divergence, for example, by placing samples closer to the guide. The one-dimensional cross-section of the MURA image is in the inset of Figure 4 .
Schematic illustrations of neutron imaging instruments. A conventional pinhole-imaging system using a small aperture (a) and a microscope, equipped with axisymmetric grazing-incidence mirrors (b). The mirrors used in this paper have diameters of approximately 30 mm and the total length of 60 mm. The focal distances are f1 = 0.64 m and f2 = 2.56 m, the source to detector distance is 3.2 m and the magnification is 4.
MTF analysis using an image of the G0 sample. Open circles are computed by the Fourier transform of the Gaussian-Lorentzian fit of the measured line-spread function (see Figure 1(c) ). The spatial frequency at 10% MTF is 0.0035 μm−1 corresponding to the spatial resolution of 70 μm. The simulated curves were produced by ray-tracing, using the Gaussian distribution of slope errors with standard deviations σ = 40, 20, and 1 μrad, which are represented by the solid line, the dashed line, and the dotted-dashed line, respectively.
Depth of focus analysis. The inset shows the image of a group of three pinholes used for the analysis, and the horizontal profile of the intensity. X-axis represents the position of the MURA mask along the optical axis, with zero at the nominal focus. Y-axis is the average FWHM of the three peaks. The inset was measured at the sample position of 3 mm. The peak widths are approximately constant within the margin of error of about 5 μm; variations of the width are consistent with size variations of the holes themselves due to manufacturing errors. Error bars for the FWHM plot were calculated by fitting the peaks with Gaussian. The depth of focus, where the FWHM is constant within error bars, is about 6.5 mm, as indicated by two dashed lines.
Article metrics loading...