Usage

It is possible to use pdCIFplotter in a project:

import pdCIFplotter

However, it is more likely you’ll want to run pdCIFplotter as a standalone program:

pdCIFplotter

A single CIF file can be loaded at a time using the LOAD button, and then various visualisations are available.

It is assumed that all datablocks in a single CIF file have some sort of experimental relationship to each other, such as from an in-situ experiment, and that it makes sense to visualise them together.

Single plot

This plot shows a single diffraction pattern at a time. Each datablock in the given CIF file that contains diffraction data is shown in the dropdown.

All possible combinations of x- and y-ordinates given in the chosen datablock are capable of being plotted. The line styles can be customised to your liking.

If both Yobs and Ycalc ordinates are chosen, then a difference and/or a cummulative chi-sq plot can be shown. The difference plot is simply the difference between Yobs and Ycalc. The cumulative chi-sq value is an indication of the contribution of each point in the diffraction pattern to the final value of chi-sq; large vertical steps indicate misfits in the model.

If HKLs are given, and the chosen x-ordinate allows it, tick marks can be plotted showing the positions of reflections.

Both the X and Y axes can be independently scaled to show data in linear, square root, or logarithmic space.

Stack plot

This plot shows all diffraction patterns in the CIF file that have both the x- and y-ordinates chosen. The individual plots are vertically offset by a user-selectable amount. This plot is useful for seeing changes in peak positions, shapes, and intensities between diffraction patterns.

Both the X and Y axes can be independently scaled to show data in linear, square root, or logarithmic space.

Surface plot

This plot shows all diffraction patterns in the CIF file that have both the x- and z-ordinates chosen. In this plot, the colour scale represent the diffracted intensity, with the y-axis being “patter number”; i.e. the order of that particular diffraction pattern in the CIF file.

This plot is useful for seeing changes in peak positions and intensities between diffraction patterns, particularly in in-situ/operando experiments.

The X, Y, and Z axes can be independently scaled to show data in linear, square root, or logarithmic space.