February 2, 2016
As part of my postdoc at the University of Edinburgh, under the supervision of Dr Stephen Moggach and Prof Simon Parsons, I was fortunate enough to be able to spend a month in Cambridge working in the Materials Science team at the CCDC. The aim of my placement, in November 2015, was to use the new CSD Python API to calculate continuous shape measures for metal polyhedra in the CSD.
Continous shape and symmetry measures were first developed by Prof David Avnir and coworkers at The Hebrew University of Jerusalem in the early 1990s as a method to quantitatively measure the degree of symmetry an object has. This measure is based on the concept that perfect symmetry is rarely attainable in reality, so in fact most structures deviate somewhat from the shape being used to describe them. By measuring the level of distortion of a shape from its perfect descriptor, one can provide numerical answers to questions such as “How far is a snow-flake from having six-fold rotational symmetry?”, “Is a hexa-coordinated complex closer to an octahedron or to a trigonal prism?”, or more simply, “How octahedral is an octahedron?”.
It is this latter point that is of interest to our group, where we work on metal-organic compounds where structural distortions are routine and can become more pronounced with the application of high pressures. Whilst at the CCDC, I was able to write a script that uses the API to analyse all tetrahedral, square planar and/or octahedral metals in a given crystal structure refcode, and calculates how far the polyhedra deviate from a reference structure. More work is required to fully realise the potential of this program, but as a method to quickly and systematically investigate structural distortions, this forms the basis of a script that will be extremely useful.
The initial stage of my placement involved getting to grips with the API, which I quickly realised was an extremely powerful tool for systematic structural database analysis. Admittedly, I think it required me actually using the API in order to appreciate how useful it can be, so I would recommend anyone with an interest in this area to play around with it and discover this for yourself! It massively simplifies systematic database searches, with any piece(s) of crystallographic information contained in the CSD instantly accessible. It is easy to narrow down specific searches, extract required information, and cross-reference this with other structures. Though for those not literate in computer coding the use of Python could be a barrier to using the API, the API comes with extensive user guidelines and examples which are hugely helpful. I personally had essentially zero experience using Python before my placement, but found that I was able to pick it up quickly due to the clear documentation, intuitive nature of Python and the invaluable help from my co-workers. The CCDC staff, from programmers to software developers to application scientists, are all highly approachable and knowledgeable. This made for a highly enjoyable month working in a lovely part of the world.