resolution to dim until eventually all spots disappear into the noise
of the background. During data processing, global damage is often
modeled as increasing mosaicity, even though the crystal is unlikely
to break into mosaic blocks under exposure to electrons.
The challenge of data collection is to recover as much information as possible from the sample within the lifetime of the crystal
and the limitations of the instrument. Ideally, the final dataset
measures all reciprocal space, such that each reflection is observed
multiple times, and the resolution is only limited by the diffractive
power of the crystal. One of the most critical decisions concerns the
exposure, i.e., the number of electrons that impinge on the sample.
If well-diffracting and isomorphous crystals are abundant, this
decision can in part be deferred to the data-processing stage by
collecting several datasets where the exposure is adjusted such that
the observable resolution is limited by the diffractive power of each
crystal. This is not a panacea, as real crystals exhibit some degree of
non-isomorphism and merging data from many, slightly different
crystals, introduces complications of its own. Assuming the immediate goal of data acquisition is to obtain the most complete dataset,
a generally applicable strategy is to keep the exposure as low as
possible, as this will maximize the lifetime of the crystal in the
beam. The beam cannot be arbitrarily attenuated; if the data are
too weak to be processed, they will not be of any use to the
experiment. Finding the optimal compromise to the challenges
that arise during low-dose data collection are central to current
methods development in MicroED.
2 Sample Preparation
A well-ordered crystal diffracts electrons better than a poorly
ordered one, and a large crystal with many unit cells yields stronger
diffraction than a small crystal with fewer molecules. However,
crystals that are both large and well-ordered are notoriously difficult to come by. The success of MicroED largely stems from its
ability to obtain strong diffraction from small crystals, which
addresses a major bottleneck in X-ray crystallography: that of
obtaining large crystals suitable for high-resolution diffraction
measurement. Furthermore, the resolution-limiting macrocrystal
disorder may be due to mosaicity, and it has been suggested that
MicroED can circumvent this problem by collecting data from
individual mosaic blocks [11]. Smaller crystals may also be less
susceptible to radiation damage, as they allow secondary electrons
to escape before they have deposited all their energy to the
sample [12].
The strong interaction between electrons and matter not only
enables MicroED to measure diffraction data from small crystals,
but also limits the thickness of the sample. This introduces
Low-Dose MicroED
311
of the background. During data processing, global damage is often
modeled as increasing mosaicity, even though the crystal is unlikely
to break into mosaic blocks under exposure to electrons.
The challenge of data collection is to recover as much information as possible from the sample within the lifetime of the crystal
and the limitations of the instrument. Ideally, the final dataset
measures all reciprocal space, such that each reflection is observed
multiple times, and the resolution is only limited by the diffractive
power of the crystal. One of the most critical decisions concerns the
exposure, i.e., the number of electrons that impinge on the sample.
If well-diffracting and isomorphous crystals are abundant, this
decision can in part be deferred to the data-processing stage by
collecting several datasets where the exposure is adjusted such that
the observable resolution is limited by the diffractive power of each
crystal. This is not a panacea, as real crystals exhibit some degree of
non-isomorphism and merging data from many, slightly different
crystals, introduces complications of its own. Assuming the immediate goal of data acquisition is to obtain the most complete dataset,
a generally applicable strategy is to keep the exposure as low as
possible, as this will maximize the lifetime of the crystal in the
beam. The beam cannot be arbitrarily attenuated; if the data are
too weak to be processed, they will not be of any use to the
experiment. Finding the optimal compromise to the challenges
that arise during low-dose data collection are central to current
methods development in MicroED.
2 Sample Preparation
A well-ordered crystal diffracts electrons better than a poorly
ordered one, and a large crystal with many unit cells yields stronger
diffraction than a small crystal with fewer molecules. However,
crystals that are both large and well-ordered are notoriously difficult to come by. The success of MicroED largely stems from its
ability to obtain strong diffraction from small crystals, which
addresses a major bottleneck in X-ray crystallography: that of
obtaining large crystals suitable for high-resolution diffraction
measurement. Furthermore, the resolution-limiting macrocrystal
disorder may be due to mosaicity, and it has been suggested that
MicroED can circumvent this problem by collecting data from
individual mosaic blocks [11]. Smaller crystals may also be less
susceptible to radiation damage, as they allow secondary electrons
to escape before they have deposited all their energy to the
sample [12].
The strong interaction between electrons and matter not only
enables MicroED to measure diffraction data from small crystals,
but also limits the thickness of the sample. This introduces
Low-Dose MicroED
311