view article

Figure 4
(a) Score plot from the principal-component analysis with the two largest components (PC1, horizontal, and PC2, vertical). The scores t1 and t2 are new variables computed as linear combinations of all variables (intensities as a function of 2θ) and aim at describing as much of the original variation as possible without losing information. The first component (PC1) is mainly related to the variation of the F, G and H on the left side and the D and J crystals, on the right. The vertical plane (PC2), is related to the differences of the E, D and X crystals. The ellipse indicates a 95% confidence interval. Samples outside this region are possible outliers. The first two principal components (PC1 and PC2) are not enough to separate the I, K crystals fully from the B crystals. The three-dimensional plot in (b) is plotted with PC2, PC3 and PC4 and shows that those crystals are better separated by the fourth component. In total, nine separate groups can be seen where each group or cluster contains crystals from the same crystal system and/or with the same structural arrangement. Two samples of each crystal type have been included, except for the F crystals where three samples (F, G and H), originating from different formulations, are included to show their high degree of similarity. All sample data were collected at synchrotron beamlines (Table 1[link]).

Journal logoJOURNAL OF
ISSN: 1600-5767
Follow J. Appl. Cryst.
Sign up for e-alerts
Follow J. Appl. Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds