Structure analysis of standard crystals is becoming more and more a routine task. Data collection with standard laboratory equipment can be done in a few hours for many cases, and the solution by modern methods, as charge flipping (1), might take some minutes at the most. In the last years, similar progress has been achieved for solution and refinement of modulated and composite crystals (2), so that nowadays some of these structures can be solved routinely by non-specialists.
However, new materials exhibiting specific physical properties, as magnetism, electrical conductivity or superconductivity, require more detailed structure analysis. Such compounds must be studied under different conditions, like high pressure and non-ambient temperature, to describe structural changes during phase transitions. For these cases, structure analysis is still challenging and a swiftly developing field.
Phase transitions can lead to the ordering of magnetic moments and/or development of ferroelectric moments in the material. To study magnetically ordered systems neutron diffraction must be used. The solution and refinement of magnetic structures is usually made by decomposition of the magnetic configuration space into basis modes, which transform according to different physically irreducible representations (irreps) of the space group of the paramagnetic phase (3). Recently it was shown that the direct use of Shubnikov (magnetic) space and superspace groups enables easier handling of non-modulated and modulated magnetic structures as well, and simplifies the algorithms dealing with diffraction data of magnetic structures (4). For incommensurately modulated magnetic structures this approach may be particularly beneficial, as for instance in the analysis of multiferroic phases (5).
Solution of a magnetic structure starts with symmetry analysis. The program Jana2006 provides a simple procedure for testing different irreps of the parent paramagnetic structure, as follows from the already known nuclear structure, and links them with the corresponding Shubnikov space or superspace groups. The magnetic structure model resulting from the selected magnetic symmetry can be checked in VESTA (6) which is launched by Jana2006, whereas the simulated powder data for a model can be compared against the measured diffraction profile. Jana2006 can also call via internet the recently developed program ISODISTORT (7,8) which provides similar analysis (although more detailed), and use its results.
As a next step, all the acceptable models from the representation analysis are refined and the solution having the best fit with experimental data can be selected as the final solution of the magnetic structure. However, this process is not necessarily straightforward and, in many cases, different magnetic spin configurations can similarly fit the data. Consequently, other experimental methods like polarized neutron diffraction must be used to identify the correct solution.
The basic idea behind the new option for magnetic structures in Jana2006 lies fully on magnetic and nuclear symmetry, similarly to classical crystallography. Symmetry is applied during data processing (merging symmetry-related reflections for single-crystal data or reducing the generated reflections to the independent ones for powder data) as well as in the calculation of magnetic structure factors. This new approach greatly improves the stability of the refinement and offers a logical path to describe twin domains.
Two procedures are possible for the combination of nuclear and magnetic diffraction. The traditional way used in FullProf (9) is to combine two independent phases. The method preferred in Jana2006 is to use a common structural description for both phases and make the combination of intensities internally, as follows from the calculation of structure factors.
Jana2006 also offers simultaneous refinement against different diffraction experiments. Commensurate and incommensurate phases can be combined as well, as powder and single-crystal data. Therefore, manifold possibilities are available for the refinement of modulated magnetic structures from various experiments.