would enable Laue diffraction techniques to be readily used for small molecule
structure determinations on crystals with unknown unit cells. A single Laue diffraction frame contains more structural data than that of a whole monochromatic dataset,
but with this technique, indexing an unknown unit cell is extremely difficult as while
the position of the diffraction peak is known, the wavelength is not. With such a
detector, it would be possible to determine both the position and wavelength, and
thereby indexing the unit cell becomes straightforward. Laue diffraction would be
ideal for very small crystals in small beams (where rotation could mean the crystal
moving out of the beam) or in a sample environment with restricted access.
The implications of increased automation and advanced instrumentation are a
little more subtle for the home laboratory. The most significant advance in a
generation is the introduction of Hybrid Pixel Detectors, and these are going to
become much more commonplace in the next decade. Time saving is significant,
which for service crystallography means that turnaround times become negligible
and a facility can operate more efficiently. The other main observation is that the data
quality is significantly improved due to a better signal-to-noise ratio and lower
background levels. It is questionable whether automation will have such a significant
impact in the home laboratory, but some elements of this practice, such as rigorous
collection of metadata, would mean that a better-quality result is reported. However,
if the principles of automation were applied to the screening process, it would allow
for the best possible result to be gleaned from a sample, as many crystals can be
assessed and the best selected. This would ensure that the best possible result would
come out of the home laboratory, and it would be possible to extend the current
capability of the technique through approaches such as merging of multiple datasets,
but the advances in the home laboratory in this respect will probably be rather more
incremental.
The main question is one of whether full automation is really the best direction to
take. On the one hand, it is not necessary to train and retrain users, saving time; users
can no longer make mistakes during data collection, so the data quantity will
improve; time will be freed up for difficult and time-consuming experiments. On
the other hand, a fully automated beamline and even lab diffractometer system mean
it is easy for anyone to collect data as long as they can mount the crystal in the correct
position on the mounting loop. But we must be aware knowledge is being lost by this
approach as these systems become black boxes and the understanding of the process
is lost. Add to this automated refinement and over generation of students and the art
of chemical crystallography will be lost as the fundamental knowledge of the
experiment resides with fewer and fewer individuals.
4.2.2 Computing
Alongside instrumentation developments, in the modern era, it is now imperative to
simultaneously consider a fully integrated and end-to-end software infrastructure to
complement and support experimentation. It is important to recognise that such an
infrastructure now needs to cover three distinct aspects: data acquisition and
Leading Edge Chemical Crystallography Service Provision and Its Impact on. . .
127
structure determinations on crystals with unknown unit cells. A single Laue diffraction frame contains more structural data than that of a whole monochromatic dataset,
but with this technique, indexing an unknown unit cell is extremely difficult as while
the position of the diffraction peak is known, the wavelength is not. With such a
detector, it would be possible to determine both the position and wavelength, and
thereby indexing the unit cell becomes straightforward. Laue diffraction would be
ideal for very small crystals in small beams (where rotation could mean the crystal
moving out of the beam) or in a sample environment with restricted access.
The implications of increased automation and advanced instrumentation are a
little more subtle for the home laboratory. The most significant advance in a
generation is the introduction of Hybrid Pixel Detectors, and these are going to
become much more commonplace in the next decade. Time saving is significant,
which for service crystallography means that turnaround times become negligible
and a facility can operate more efficiently. The other main observation is that the data
quality is significantly improved due to a better signal-to-noise ratio and lower
background levels. It is questionable whether automation will have such a significant
impact in the home laboratory, but some elements of this practice, such as rigorous
collection of metadata, would mean that a better-quality result is reported. However,
if the principles of automation were applied to the screening process, it would allow
for the best possible result to be gleaned from a sample, as many crystals can be
assessed and the best selected. This would ensure that the best possible result would
come out of the home laboratory, and it would be possible to extend the current
capability of the technique through approaches such as merging of multiple datasets,
but the advances in the home laboratory in this respect will probably be rather more
incremental.
The main question is one of whether full automation is really the best direction to
take. On the one hand, it is not necessary to train and retrain users, saving time; users
can no longer make mistakes during data collection, so the data quantity will
improve; time will be freed up for difficult and time-consuming experiments. On
the other hand, a fully automated beamline and even lab diffractometer system mean
it is easy for anyone to collect data as long as they can mount the crystal in the correct
position on the mounting loop. But we must be aware knowledge is being lost by this
approach as these systems become black boxes and the understanding of the process
is lost. Add to this automated refinement and over generation of students and the art
of chemical crystallography will be lost as the fundamental knowledge of the
experiment resides with fewer and fewer individuals.
4.2.2 Computing
Alongside instrumentation developments, in the modern era, it is now imperative to
simultaneously consider a fully integrated and end-to-end software infrastructure to
complement and support experimentation. It is important to recognise that such an
infrastructure now needs to cover three distinct aspects: data acquisition and
Leading Edge Chemical Crystallography Service Provision and Its Impact on. . .
127