Tools for the Characterization of Biomass at the Nanometer Scale - The Nanoscience and Technology of Renewable Biomaterials - page 79

One problem with TEM is that the electron scattering cross section shows little dif-
ference, and hence contrast, between carbon, oxygen, and nitrogen. One approach to
developing image contrast is to preferentially label a component with stains contain-
ing high-atomic-number elements. Lignin has been labeled using bromine, potassium
permanganate, and osmium tetroxide (57). Our current knowledge of hemicellulose
deposition is from TEM studies where hemicellulose has been labeled with gold-tagged
antibodies (157). Use of antibodies tagged with nanometer-sized gold particles is routine
in biological microscopy (158).
Another problem of TEM is damage from the electron beam. Embedding medium, a
model organic substrate, shrank 5% laterally and 25% in thickness during the first 5 min
of TEM exposure, or 13,000 electrons per nm 2 (159). Shrinkage and degradation of
specimens should be considered whenever evaluating TEM images of biomass.
The electron energy loss spectrum ( EELS ) is commonly used to determine the valence
state of atoms in TEM specimens. Electrons passing through a specimen lose energy by
ionization of specimen atoms, and these quantized losses are characteristic of different
elements. Thus electrons interacting with carbon and oxygen lose 285 eV and 532 eV,
respectively (160). The valence state of the atom can be determined by small shifts in
this value.

3.4.4.1 Electron Tomography

In contrast to conventional two-dimensional TEM, electron tomography (ET) using
brightfield TEM produces a three-dimensional rendering of a volume. ET has been exten-
sively used to study cell structure (161, 162). To determine a unique three-dimensional
representation, approximately 100 images must be obtained using 1 ◦ to 5 ◦ tilt increments
over a large angular range, such as 70 ◦ .

The large number of images needed could subject a specimen to severe electron
damage, so several techniques have been adopted to minimize electron exposure (see
Transmission Electron Microscopy, above). Low-dose microscopy techniques use auto-
matic focusing and drift correction on a specimen field adjacent to the one being imaged.
Because many images will be summed to make the final composite, each image can tol-
erate slightly higher noise than standard images and therefore less electron exposure. A
case has been made that the total electron exposure needed for a tilt series is about the
same as that for a single two-dimensional image (163). In any case, the electron flux
should be kept within a few thousand electrons per square nanometer (159). Another
method of minimizing damage from electron exposure is to maintain the specimen at
cryogenic temperature (136). Cryo-electron tomography has been useful in describing
cellular structure and the structure of cell components (164-166).
Electron tomography has revealed the nanometer-level organization of cellulose
microfibrils in the S2 layer of the cell wall and the arrangement of some residual
lignin and hemicellulose about the cellulose microfibrils in radiata pine (167). The
cell sections were partly delignified with peracetic acid, dehydrated, treated with
multiple heavy metal stains, and embedded in Spurr's epoxy. Although some structural
distortion may have been introduced by the treatment, the results are a major step
toward understanding wood cell wall structure.

±