Nature Chemistry | The Sceptical Chymist

Crystal clear data

A change in one of our publication policies had been brewing for a while at the journal — and I’m happy to say that it has now been implemented: we have updated our requirements regarding the crystallographic characterization of small molecules. This is reflected in our guide to authors.

Until now we had been asking authors to provide a standard crystallographic information file (CIF) for each new structure characterized by X-ray diffraction analysis. This file doesn’t represent the complete story though; other experimental and refinement information such as the structure factors (HKL or FCF files) and RES files also exist.

A few years ago, in an article on their website entitled ‘Publication standards for crystal structures’, the International Union of Crystallography (IUCr) recommended that for each newly determined structure, not only should the CIF file be provided, but also the corresponding structure-factor information — this has long been a requirement for the IUCr’s own journals. The structure factors are important for the structure determination and so should be available during the peer-review process, and may also be used by readers interested in the refinement of the structure once the paper has been published.

Endorsing this stance, the Cambridge Structural Database (CSD) has enabled the deposition of structure factors together with the main CIFs, and has also been making crystallographic information increasingly easy to access. One now only needs to be armed with a CCDC number, a CSD code, or the DOI of the paper in which the structure was reported, to acquire the desired crystallographic details (through https://summary.ccdc.cam.ac.uk/structure-summary-form).

We’re happy to have now adopted this practice. We are asking that manuscripts reporting new crystal structures be accompanied by CIF files and associated information about structure factors.

The best — or easiest — way to do so is to use an up-to-date version of SHELXL (2014 or later) which now embeds the HKL and RES files into the generated CIF (other programs may also do this, I’m not sure). If another program has been used, then the CIF and the structure-factor file (in HKL and/or FCF format) can be handled as two separate files.

Our other requirements haven’t changed; we’re still asking that an ORTEP-style illustration of the structure, with probability ellipsoids, appears in the main Supplementary Information. And CIF files — including structure factors — should be run through the IUCr’s free online CheckCIF routine, the output submitted with the manuscript files (these are just used during the reviewing process, we don’t host them with the published paper), and any A- or B-level alerts that come up be explained in the Supplementary Information file.

In due course we host all this crystallographic raw data with the Supplementary Information of the associated paper — see, for example, this recent paper (Nature Chemistry, 6, 1079–1083 (2014) here) — and hope that it proves useful to the crystallographic community.

Comments

  1. Report this comment

    Armel Le Bail said:

    If one is not armed with a CCDC number, a CSD code, or the DOI of the paper in which the structure was reported, then to acquire the desired crystallographic details being armed with the chemical composition, cell parameters (etc) one may search into the Open Crystallography Database (https://www.crystallography.net). This supposes that crystallographers having published a crystal structure not only deposit their CIFs to CSD or ICSD but to the COD as well. Thanks !