ultimately set the limits. In general, for all imaging techniques small areas can be
viewed at the highest possible resolution but larger fields of view must be imaged
with a lower resolution or with other procedures such as tiling [34, 35]. This
interplay is particularly notable in FIB-SEM and SBEM where there is a need to
balance imaging conditions with the mechanism to remove sections. In both
technologies, imaging conditions that are too harsh for the sample will result in
inconsistent removal of material and poor image quality. The best way to manage
this is to narrow down what is required from the microscope to answer the biological question and how much data is necessary to achieve this. At the FIB-SEM,
improved hardware and intelligent software such as Atlas3D part of Atlas5 (Zeiss)
helps to extend these limitations [34]. In the SBEM, the ability to section an entire
block-face alongside the implementation of montaging or working with regions of
interest allows acquisition from very large areas. In both cases, this leads to the
production of gigabyte or even terabyte sized datasets, which could be too unwieldy
to handle (especially from SBEM). Big datasets are very slow to move from one
place to another and need big capacity storage and significant computing power in
order to analyse them. It also takes considerable time to acquire large datasets, this
can be on the order of days, and the instrument is then not available for other
projects. It is important to be realistic about what is to be achieved from the work
[10, 36].
Where possible, correlative light and electron microscopy (CLEM) is extremely
useful for volume SEM. It has been utilised to bridge the gap between many scales
and forms of microscopy. Recently the combination of microscopic X-ray computed tomography (microCT) with FIB-SEM [37, 38] and SBEM [39] is receiving a
lot of attention. For volume SEM, CLEM will help to target the specific region of
interest, thereby reducing the area that needs to be acquired [9, 37, 38]. It will also
help (when possible) to give a broader context of the sample since it will likely look
very different in the SEM. This can similarly be achieved by simply documenting
well the mounting and preparation of the sample prior to loading it into the
instrument.
5.2.1 Sample Processing
Sample processing for electron microscopy aims for excellent preservation of
structures. The steps to achieve this have not changed significantly throughout the
years, they are variations of a similar format: primary fixation, contrasting, dehydration and infiltration [40]. The same is true for volume-SEM techniques.
However, these samples need enhanced contrast relative to standard TEM samples
where it is possible to post stain the ultrathin sections (Fig. 5.3). Instead, all of the
heavy metals must be infiltrated into the sample during the initial processing. These
heavy metals will be responsible for the production of the backscattered electrons
(even at low voltages) and thus the actual imaging of the sample and at the same
time make the sample conductive to alleviate the effects of charging.
122
R. I. Webb and N. L. Schieber
viewed at the highest possible resolution but larger fields of view must be imaged
with a lower resolution or with other procedures such as tiling [34, 35]. This
interplay is particularly notable in FIB-SEM and SBEM where there is a need to
balance imaging conditions with the mechanism to remove sections. In both
technologies, imaging conditions that are too harsh for the sample will result in
inconsistent removal of material and poor image quality. The best way to manage
this is to narrow down what is required from the microscope to answer the biological question and how much data is necessary to achieve this. At the FIB-SEM,
improved hardware and intelligent software such as Atlas3D part of Atlas5 (Zeiss)
helps to extend these limitations [34]. In the SBEM, the ability to section an entire
block-face alongside the implementation of montaging or working with regions of
interest allows acquisition from very large areas. In both cases, this leads to the
production of gigabyte or even terabyte sized datasets, which could be too unwieldy
to handle (especially from SBEM). Big datasets are very slow to move from one
place to another and need big capacity storage and significant computing power in
order to analyse them. It also takes considerable time to acquire large datasets, this
can be on the order of days, and the instrument is then not available for other
projects. It is important to be realistic about what is to be achieved from the work
[10, 36].
Where possible, correlative light and electron microscopy (CLEM) is extremely
useful for volume SEM. It has been utilised to bridge the gap between many scales
and forms of microscopy. Recently the combination of microscopic X-ray computed tomography (microCT) with FIB-SEM [37, 38] and SBEM [39] is receiving a
lot of attention. For volume SEM, CLEM will help to target the specific region of
interest, thereby reducing the area that needs to be acquired [9, 37, 38]. It will also
help (when possible) to give a broader context of the sample since it will likely look
very different in the SEM. This can similarly be achieved by simply documenting
well the mounting and preparation of the sample prior to loading it into the
instrument.
5.2.1 Sample Processing
Sample processing for electron microscopy aims for excellent preservation of
structures. The steps to achieve this have not changed significantly throughout the
years, they are variations of a similar format: primary fixation, contrasting, dehydration and infiltration [40]. The same is true for volume-SEM techniques.
However, these samples need enhanced contrast relative to standard TEM samples
where it is possible to post stain the ultrathin sections (Fig. 5.3). Instead, all of the
heavy metals must be infiltrated into the sample during the initial processing. These
heavy metals will be responsible for the production of the backscattered electrons
(even at low voltages) and thus the actual imaging of the sample and at the same
time make the sample conductive to alleviate the effects of charging.
122
R. I. Webb and N. L. Schieber