procedures detailed above for cryogenic specimens may be used. Additionally, the
plastic section may be introduced into the microscope via a cryo sample holder.
After pre-irradiation, the sample holder dewar is filled with liquid nitrogen, and
once the temperature has equilibrated to, say, −175 °C, the tilt-series can be
acquired with the least possible variation from start to finish [80, 81]. This approach
results in further changes of only a few percent, making it more feasible to align the
set of projections. Cooling a plastic-embedded specimen does not eliminate
beam-induced shrinkage. Pre-irradiation of the specimen at room temperature is
mandatory, and as a consequence, profound changes (anisotropic shrinkage and
mass loss) take place prior to the acquisition of the first projection of the tilt series.
The cooling serves only to minimise further changes during acquisition. As noted
by Luther [82], the take-up of this cooled plastic-section technique has been low,
and the “ultimate goal” should be to use vitreous specimens exclusively. In practise,
however, this strategy is convenient, and its success rate in terms of yielding a
‘result’ from cellular specimens or tissues is very high. Use of this strategy should
be considered in the light of the question being posed, and whether conclusions are
likely to be the result of changes induced by sample processing.
The ability to record so many projections from a plastic section has resulted in a
further misconception about the attainable resolution explained in terms of the
Crowther criterion, which simplifies the achievable resolution by considering the
number of projections and (related to this), the angular spacing of the projections.
This will be discussed under the heading ‘Resolution’, suffice to say that the
concept is subject to limits, and ultimately, it is meaningless in the absence of
adequate sample preservation.
1.6 Special Acquisition Case: Block-Face SEM
The name ‘serial block-face scanning electron microscopy, SBFSEM’ was coined
by Denk for a technique in which a specimen block is imaged during precise
removal of sections, each of constant thickness, one layer at a time [21]. The name
describes a newer focussed ion beam sectioning approach [66] equally well.
Serial block face SEM uses plastic-embedded specimens exclusively. The ability
to image the face of a planed, frozen-hydrated block using cryo-SEM had been
demonstrated previously [83]. This might not seem obvious, as SEM is usually
thought of as a technique to visualise topography. A surface that is planed with a
diamond knife is essentially flat, except for knife marks. Contrast in a cryo-SEM is
typically achieved by brief etching of the surface to expose features. Etching is
achieved by sublimation, which in turn is effected by raising the temperature to
approximately −90 °C, well above the devitrification temperature. Thus, devitrification has occurred, and this is likely to be the case in freeze-substitution as well.
Furthermore, due to heat-transfer considerations, it is doubtful whether larger
specimens (>200 µm) can be vitrified in the first place. On the other hand,
freeze-substitution and plastic-embedding provides specimen blocks that are readily
12
A. Leis
plastic section may be introduced into the microscope via a cryo sample holder.
After pre-irradiation, the sample holder dewar is filled with liquid nitrogen, and
once the temperature has equilibrated to, say, −175 °C, the tilt-series can be
acquired with the least possible variation from start to finish [80, 81]. This approach
results in further changes of only a few percent, making it more feasible to align the
set of projections. Cooling a plastic-embedded specimen does not eliminate
beam-induced shrinkage. Pre-irradiation of the specimen at room temperature is
mandatory, and as a consequence, profound changes (anisotropic shrinkage and
mass loss) take place prior to the acquisition of the first projection of the tilt series.
The cooling serves only to minimise further changes during acquisition. As noted
by Luther [82], the take-up of this cooled plastic-section technique has been low,
and the “ultimate goal” should be to use vitreous specimens exclusively. In practise,
however, this strategy is convenient, and its success rate in terms of yielding a
‘result’ from cellular specimens or tissues is very high. Use of this strategy should
be considered in the light of the question being posed, and whether conclusions are
likely to be the result of changes induced by sample processing.
The ability to record so many projections from a plastic section has resulted in a
further misconception about the attainable resolution explained in terms of the
Crowther criterion, which simplifies the achievable resolution by considering the
number of projections and (related to this), the angular spacing of the projections.
This will be discussed under the heading ‘Resolution’, suffice to say that the
concept is subject to limits, and ultimately, it is meaningless in the absence of
adequate sample preservation.
1.6 Special Acquisition Case: Block-Face SEM
The name ‘serial block-face scanning electron microscopy, SBFSEM’ was coined
by Denk for a technique in which a specimen block is imaged during precise
removal of sections, each of constant thickness, one layer at a time [21]. The name
describes a newer focussed ion beam sectioning approach [66] equally well.
Serial block face SEM uses plastic-embedded specimens exclusively. The ability
to image the face of a planed, frozen-hydrated block using cryo-SEM had been
demonstrated previously [83]. This might not seem obvious, as SEM is usually
thought of as a technique to visualise topography. A surface that is planed with a
diamond knife is essentially flat, except for knife marks. Contrast in a cryo-SEM is
typically achieved by brief etching of the surface to expose features. Etching is
achieved by sublimation, which in turn is effected by raising the temperature to
approximately −90 °C, well above the devitrification temperature. Thus, devitrification has occurred, and this is likely to be the case in freeze-substitution as well.
Furthermore, due to heat-transfer considerations, it is doubtful whether larger
specimens (>200 µm) can be vitrified in the first place. On the other hand,
freeze-substitution and plastic-embedding provides specimen blocks that are readily
12
A. Leis