Figure 2
Cross sections of the experimental INS count distributions of the three kinds of data to which we applied the method. 2D slices are taken from the initial histograms; the integration ranges for the other two axes are 1 in units of the initial bin widths. The numerical scales in (q, E) correspond to the bin centers, except that of E in data set 1 corresponds to the bin bottoms. The coordination on the other two axes is shown in units of meV or rlu above each plot. The method optimized the bin widths using the count distribution in a 4D hyper-rectangle located in the white rectangles in the cross sections. The small light-green rectangles in the second row of all data sets and the third and fourth rows of data set 3 show the 4D space commonly included in all data sets. We used the counts within the common 4D space as a measure of the INS intensity of the data. |