simply installing a single instrument (and crystallographer!). The modern chemical
crystallography service needs to be based on the integration of advanced instrumentation with people, software and processes. National facilities have existed for a long
time and have invariably been the driver and proving-ground for both new technological advances and in terms of how a facility should be run. For example, in these
facilities automation, both of hardware and software, has been developed. However,
the throughput of these facilities meant that data management, that is, the handling of
the high volumes produced by the instruments, processing it and organising the
results, has also become an absolute necessity. We therefore consider ‘the facility’ to
be a combination of all these technologies that serves a range of different users with
different chemistry research and different crystallographic experience.
We will draw on the experiences of two different types of facility in order to
illustrate the main advances raised in this article. These are world-leading, nationalscale facilities, with the principle of this review being that they have a mandate to
pioneer the technique and that the innovations developed by these facilities will in
turn become common practice in all facilities. The obvious examples of this type of
facility are at national laboratory synchrotron sources, and we will draw heavily on
two main examples of these, (a) beamline I19 [7, 8] at the Diamond Light Source in
the UK and (b) beamlines 11.3.1 and 12.2.1/2 [9] at the Advanced Light Source in
the USA. Further to these, the National Crystallography Service (NCS) in the UK
[10, 11] operates at the national scale as an Engineering and Physical Science
Research Council-funded National Research Facility [12] but is university-based,
and this provides an illustration of how the technologies developed at synchrotrons
can be translated to the home laboratory.
The modern service crystallography facility is however only one half of the topic
of this review. Chemical crystallography facilities worldwide now generate results at
an unprecedented rate, and this has a knock-on effect – the data explosion [13]. Not
only does the facility have to cope with this situation, but so does the dissemination
process – publishing processes are struggling to deal with this, but still we have seen
a dramatic growth of the crystallographic databases. The second half of this review
looks at the rise of crystallographic data, which has powered development of
crystallographic databases. As the most relevant database to the chemistry community, we mainly consider herein the growth of the Cambridge Structural Database,
the ‘CSD’ [14]. The CSD recently accumulated its millionth structure [15], but the
state of the art in crystallographic data science does not stop with merely aggregating
crystal structures – data mining approaches enable the understanding of trends and
development of general rules. Modern approaches and large volumes of data now
enable the development of ‘knowledgebases’, and this review will illustrate how it is
possible to go beyond crystallographic knowledge and use it to drive new science.
72
S. J. Coles et al.
crystallography service needs to be based on the integration of advanced instrumentation with people, software and processes. National facilities have existed for a long
time and have invariably been the driver and proving-ground for both new technological advances and in terms of how a facility should be run. For example, in these
facilities automation, both of hardware and software, has been developed. However,
the throughput of these facilities meant that data management, that is, the handling of
the high volumes produced by the instruments, processing it and organising the
results, has also become an absolute necessity. We therefore consider ‘the facility’ to
be a combination of all these technologies that serves a range of different users with
different chemistry research and different crystallographic experience.
We will draw on the experiences of two different types of facility in order to
illustrate the main advances raised in this article. These are world-leading, nationalscale facilities, with the principle of this review being that they have a mandate to
pioneer the technique and that the innovations developed by these facilities will in
turn become common practice in all facilities. The obvious examples of this type of
facility are at national laboratory synchrotron sources, and we will draw heavily on
two main examples of these, (a) beamline I19 [7, 8] at the Diamond Light Source in
the UK and (b) beamlines 11.3.1 and 12.2.1/2 [9] at the Advanced Light Source in
the USA. Further to these, the National Crystallography Service (NCS) in the UK
[10, 11] operates at the national scale as an Engineering and Physical Science
Research Council-funded National Research Facility [12] but is university-based,
and this provides an illustration of how the technologies developed at synchrotrons
can be translated to the home laboratory.
The modern service crystallography facility is however only one half of the topic
of this review. Chemical crystallography facilities worldwide now generate results at
an unprecedented rate, and this has a knock-on effect – the data explosion [13]. Not
only does the facility have to cope with this situation, but so does the dissemination
process – publishing processes are struggling to deal with this, but still we have seen
a dramatic growth of the crystallographic databases. The second half of this review
looks at the rise of crystallographic data, which has powered development of
crystallographic databases. As the most relevant database to the chemistry community, we mainly consider herein the growth of the Cambridge Structural Database,
the ‘CSD’ [14]. The CSD recently accumulated its millionth structure [15], but the
state of the art in crystallographic data science does not stop with merely aggregating
crystal structures – data mining approaches enable the understanding of trends and
development of general rules. Modern approaches and large volumes of data now
enable the development of ‘knowledgebases’, and this review will illustrate how it is
possible to go beyond crystallographic knowledge and use it to drive new science.
72
S. J. Coles et al.