Evaluation of Charging on Macromolecules in Electron
Cryomicroscopy
Jacob Brink, Michael B. Sherman, John Berriman*, and Wah
Chiu
National Center for Macromolecular Imaging
Verna and Marrs McLean Department of Biochemistry
Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030,
U.S.A.
*Medical Research Council,
Laboratory of Molecular Biology, Hills Road Cambridge, CB2 2QH
U.K.
Abstract
We describe procedures to assess charging of biological specimens under
electron irradiation in an electron cryomicroscope. Charging can be observed
by an expansion of the illuminating beam, blurring of electron diffraction
patterns and by beam 'footprints' on the specimen. Discharging can also be
seen in the defocused electron diffraction mode. We investigated the influence
of a variety of factors on the magnitude and visibility of charging. A
reduction of charging is noticed when part of the adjacent carbon film is
included in the irradiated specimen area.
Introduction
Biological macromolecules are generally embedded in ice or a sugar to preserve
their native structure during electron microscopic observations. In many cases
they are prepared without a carbon support film to avoid structure distortions
or preferential orientation. Images of these types of specimens often reveal
specimen movement in different directions
[1],
or film breakage upon electron irradiation
[2].
A quantitative analysis of images of ice-embedded tobacco mosaic virus showed
that layer line intensities in computed Fourier transforms fell off faster than
the theoretical expectation
[3].
Charging has been suggested to be the physical cause of all these phenomena.
Charging has long been recognized as a problem in conventional fixed-beam and
scanning electron microscopy
[4,
5]
.
So far, no simple technique has been proposed for the evaluation of a
specimen's susceptibility to charging or the extent of charging. Here, we
report on a number of procedures to assess the charging on a glucose or ice
embedded biological specimen under low-dose electron diffraction and imaging
conditions.
Materials and Methods
The specimens used in this study are crystals of catalase and crotoxin complex
[6,
7]
embedded in glucose, thin layer of glucose or vitreous ice, and plain carbon
film. The catalase crystals were prepared across the holes in a holey film
without a carbon support film, whereas in the case of crotoxin complex, a thin
(~100 Å) continuous carbon support film was used. Holey carbon films
were made as described
[8]
with approximately 200-500 Å of carbon coating. The plastic nets were
subsequently removed using organic solvents followed by coating with a layer of
carbon on the other side of the grid. Thin continuous carbon film was prepared
by indirect vacuum deposition onto mica in a Denton evaporator and floated off
on the water surface. The film was picked up using a 200 or 400 mesh grid
covered with a holey film. Thin layers of vitreous ice were obtained using
established techniques
[9].
In our studies, we collected most data as typical electron diffraction
patterns (EDP) from protein crystals
[10],
or as a defocused electron diffraction image (DIFF image). The JEOL4000EX
intermediate voltage electron cryomicroscope operated at 400 kV was used to
perform all the experiments with the specimens mostly kept at -167oC in a
Gatan 626 cryoholder. All the on-line observations were made on a TV screen
connected to a Gatan wide-angle TV-rate CCD camera (model 673 Mk. 3). EDP or
DIFF images can be recorded using a VCR unit or a Gatan 1024 x 1024 slow-scan
CCD camera (model 679). The slow-scan camera was interfaced to a Power
Macintosh 9500/150 computer. Acquisition, display and storing of CCD data were
done using simple programs inside the DigitalMicrograph program
[11].
Results and Discussion
Observed phenomena attributable to charging
Electron cryomicroscopy of biological specimens embedded in vitreous ice or
glucose often reveals effects related to charging. Three different ways in
which these effects can readily be observed are: a) changes in the beam
diameter, b) blurring in the diffraction mode, and c) a footprint found on the
specimen in the search mode after an electron irradiation.
When searching the specimen for suitable areas, changes in the beam diameter
will occur as the edge of the beam moves off the carbon net and into a hole
covered with either vitreous ice or glucose. A sequence of DIFF images
extracted from video recordings[1] obtained from
a vitreous ice sample illustrates these changes (Fig. 1). The beam diameter as
measured from images recorded on the slow-scan CCD camera sometimes increases
by up to 20 %. This increase has been found to depend on the accumulated
electron dose on the area. Conversely, the beam diameter reduces to its
original size as it returns on the carbon film. Interestingly, we have noticed
that this reduction can occur just before the beam illuminates the carbon net.
In some cases, we have noticed that this reduction is related to the amount of
carbon net included in the beam.
Blurring of reflections is observed when diffraction data is obtained from
protein crystals prepared across holes (Fig. 2). The EDP shows these blurred
reflections when the beam irradiates only the crystals and not the adjacent
carbon net (Fig. 2a). The undiffracted beam in the EDP is enlarged in diameter
(Fig. 2b) and a DIFF image, acquired immediately after the EDP, shows a
dark annulus that delineates the irradiated region (Fig. 2c). Sharp, high
resolution diffraction reflections extending to a resolution better than 3
Å are observed when the irradiated area includes part of the carbon net
(Fig. 2d). The DIFF image acquired afterwards still reveals an annulus but
with much reduced contrast (Fig. 2f).
Footprints of the electron beam are frequently observed in DIFF images
acquired after recording EDPs or images from glucose-embedded protein crystals
prepared on a thin carbon film, in this case of crotoxin complex (Fig. 3). In
a similar way, DIFF images often reveal a raster of small footprints or annuli
on the specimen after spot-scan imaging, not all of which showed up with the
same level of contrast. The diameter of these annuli always matched the
diameter of the electron beam used in either the diffraction or the imaging
experiment.
Appearance and disappearance of annulus (footprint of
beam)
Electron optical effect on the appearance of observed annulus
The DIFF image is obtained by defocusing the diffraction lens. Figure 5a
shows a ray diagram for the formation of the DIFF image near the back
focal plane of the objective lens at either over or under-focus. The DIFF
images used in this paper are obtained in most cases with an over-focused
diffraction lens where the electron paths are diverging. The net consequence
of specimen charging is the creation of an electrostatic lens, which
strengthens the objective lens, i.e. reduces its focal length
[12].
Consequently, charging results in an expansion of the overfocused pattern.
Conversely, at underfocus, the electron paths are converging. In that case
charging will reduce the DIFF image in size. Figures 5b and c demonstrate
these magnification changes. In addition to a difference in the annulus
diameter, we observe the contrast reversal from dark to bright. This
difference in contrast of the annulus as a function of defocus is consistent
with the assumption that the charged patch is a phase object.
Effect of electron dose on the appearance of observed annulus
Because images are obtained at a dose typically 10-fold higher than that used
for an EDP, "charge-discharge series" have also been acquired at a higher dose
(Fig. 6). Figure 6a shows the specimen consisting of glucose-embedded catalase
crystals spanning holes in the carbon net. For reference, the same region is
shown in a DIFF image after charging it as in fig. 4b by irradiating a 2.5
um-diameter area to 0.1 e/Å2 (Fig. 6b). By simply increasing
irradiation dose to 5 e/Å2, we notice a marked increase in
contrast which has spread out to cover the entire hole. In addition, almost
the whole edge of the hole is very bright and the annulus is hard to discern
(Fig. 6c). Accompanying these effects is a change in the effective
magnification by ~ 20% as measured from the dimensions of the catalase
crystal.
Is the specimen completely discharged?
The charge appears to be dissipated by irradiating the adjacent
carbon net, as shown in figure 4. To examine if this dissipation is complete,
we performed the following experiment. After the first "charge-discharge
series" was acquired as in figure 4a-e, the beam diameter was increased further
by a factor of ~ 2 (to 12.5 um) followed by an acquisition of another discharge
series. The first DIFF image of the first "charge-discharge series" reveals an
annulus, as observed previously (Fig. 7a). The annulus matches the irradiated
area in both diameter and location and gradually disappears as the adjacent
carbon net is irradiated, similar as in figure 4. The DIFF image of the second
discharge series (Fig. 7b), obtained using a 12.5 um-diameter beam, shows a
larger area of the specimen revealing a clear annulus of 7.5 um in
diameter that delineates the region initially imaged in figure 7a. This
annulus is visible, although the first "charge-discharge series" was obtained
by irradiating the adjacent holey carbon film; the annulus does disappear as
subsequent DIFF images are acquired (not shown). The DIFF images indicate that
the charge present in the 2.5 um-diameter area only appears to be compensated
by irradiating the 7.5 um-diameter area. In the process, however, the latter
obtains a net positive charge suggesting that the charge may have been
re-distributed to some extent over the larger area. This implicates that the
specimen under our conditions is not fully discharged.
Does carbon film charge and discharge?
Based on our observation shown in figure 2 and those made by others
[2],
carbon film in the irradiated area would reduce the amount of charging.
However, the beam footprints shown in figure 3 on a protein crystal in the
presence of a carbon support film prompted us to suspect that charging may in
fact occur on carbon as well. When we employ our procedure of the
"charge-discharge series" to examine the charging behavior on plain thin carbon
film, we do not see effects similar to those shown in figure 4. In order to
observe the effects of charging and discharging on plain carbon film, we have
to use different electron optical conditions, i.e. < 200x
magnification and an extremely high defocus setting (~ 5.3 mm based on the
microscope panel readout). After irradiating a 25 um-diameter area with 10
e/Å2, the beam is expanded to more than 100 um in diameter.
The low-magnification image recorded from the region shows the holey carbon
film, and a circular, higher contrast patch indicated by an arrow (Fig. 8a).
The patch matches the area irradiated by the beam in diameter and location and
has a light and a dark rim as judged from the line plot (insert in Fig. 8a).
It thus resembles the annulus seen on unsupported glucose layers. This
observation confirms that carbon film charges.
In order to observe discharging on carbon film, we irradiate a 2.5 um diameter
area located within the larger 25 um diameter area to about 150
e/Å2 (Fig. 8b). This narrow beam leaves its footprint behind
characterized by a light and a dark rim, giving also this footprint the
appearance of an annulus (Fig. 8b insert). With repeated electron exposure on
this small area, the larger annulus fades as shown in the low-magnification
image (Fig. 8c). At an accumulated electron dose on the small area of ~ 1.5
times that given on the larger area, the larger annulus has nearly completely
disappeared (not shown). Interestingly, although the small area is irradiated
to a very high dose using the narrow beam (~ 1500 e/Å2), the
contrast of the annulus associated with the narrow beam appears to have reached
a steady state.
Physical model of charging and discharging
During examination of an electron microscope specimen, a fraction of the
incident electrons is inelastically scattered (Fig. 9a). The energy transfer
resulting from this interaction eventually causes the emission of secondary and
Auger electrons from the specimen
[13-15].
These electrons have a mean free path between 100 and 1000 Å
[15],
meaning that they can originate from the bulk of the specimen in addition to
its surface. The net positive charge that is generated on the specimen as a
result from the emission
[16]
will act as an electrostatic lens
[12]
(Fig. 9b). This lens' strength and shape, however, varies depending on the
amount of charge build-up and local electric properties of the specimen, as can
be clearly seen in figure 6. Consequently, the charge may cause focus changes
and blurring or movement of the image. With tilted specimens the effect
becomes even more pronounced because the Coulomb force exerted on the primary
electrons has a component perpendicular to the beam causing the image to shift
sideways during the exposure (Fig. 9c). Since charge build-up is a continuous
process throughout the image recording the varying shift will have serious
"resolution-degrading" effects. Charge that is present on a thin film of
glucose spanned across the holey carbon film (Fig. 9b) is visualized as an
annulus. The charge has an associated electric field across the specimen,
which varies by the largest amount at the perimeter of the charged patch (cf.
Fig. 5 in
[15]).
The field introduces a phase shift of the scattered electrons causing the
annulus to appear as either a dark or a bright feature depending on the focus
setting of the diffraction lens (Fig. 5b and c). "Discharging" of the specimen
(Fig. 9c) may occur during examination of a previously exposed and consequently
charged area with a wider beam, as is the case in our "charge-discharge series"
(Fig. 4), or with a narrower beam, as employed in assessing charging on a plain
carbon film (Fig. 8). Some of the secondary electrons that are emitted across
the entire region
[13]
will return to the specimen
[14],
where they can compensate the positive charge on the specimen. This mechanism
of discharging could explain the fading of the large annulus after irradiating
a smaller area. Discharging may also arise from a current through the
specimen, as schematically depicted (Fig. 9c)
[15].
This current arises from a difference in potentials between the irradiated,
positively charged part of the specimen and the conductive grid bars that are
at ground potential. To estimate the relative importance of this aspect of
discharging, we have conducted some additional experiments.
Is electron irradiation required for discharging?
If charge dissipation occurs by means of a current, two factors are expected
to influence the process, namely the length of time that elapses between
charging and the acquisition of the DIFF images, and the specimen temperature.
The DIFF images acquired in our "charge-discharge series" are typically
obtained within 5 seconds after a specimen area is charged by electron
irradiation, at which point they reveal a clearly discernible annulus (Fig.
10a). When the DIFF images are acquired after longer time periods following
the electron irradiation, e.g. 5 min., the annulus is still
present but with reduced contrast (Fig. 10b). After approximately 15
min., the annulus is nearly completely gone (Fig. 10c). The effect of the
specimen temperature on discharging has been tested by comparing DIFF images
obtained from a catalase specimen at -167oC and -30oC (Fig. 11). We find
that at higher temperature the contrast of the annulus in the "charge-discharge
series" is reduced even in the first 'discharge' image. This suggests that at
higher temperatures a higher discharge current runs through the specimen.
Therefore, both types of data indicate the possibility that charge present on a
specimen can dissipate by means of a specimen current. The relative importance
of this current as means of discharging is, however, most likely minor when
compared to discharging by secondary electron emission using irradiation of an
adjacent carbon net. For example, electron irradiation of the adjacent carbon
net for 0.25 s as in figure 4, is sufficient to discharge the specimen
completely, compared to 15 min. found in this experiment. A specimen current
does become more important at higher temperatures, which parallels the thermal
dependence of carbon film's conductivity as was measured elsewhere
[17].
Although it has been shown that the yield of secondary electron emission is
influenced by temperature, its effect is expected to be rather minimal given
the energy of the primary electrons
[18].
The effect of the specimen temperature on discharging raises an interesting
issue. It emphasizes the trade-off between keeping the specimen at low
temperature for increased radiation tolerance and damage reduction, but
possibly having serious imaging problems due to specimen
charging.
Estimation of magnitude of effects of charging
At relatively high electron exposures, the DIFF images reveal an increase in
contrast outside of the area irradiated by the electron beam (Fig. 6). We
propose that this is caused by electric breakdown of the specimen at the edge
of the illuminated area. The breakdown means that the electric field reached a
threshold above which the object becomes a conductor and 'spills' excessive
charge over a larger region reducing the field strength below that threshold.
Using formulations by Cazaux
[15],
we calculated the upper limit of charge and the electric field resulting from
electron doses used in our "charge-discharge series". In these formulations
(cf. formulas (7b) and (8) in
[15]),
the charge on the specimen depends on a number of parameters, for example the
electron dose and the yield of secondary and Auger electrons. Using the
parameters from our experiments[2], we
calculated the charge on the specimen as 4 x 10-18 C. The electric
field at the perimeter of the charged patch can then be computed as 2.2 MV/m,
which is within the range of values (800 kV/m to 150 MV/m) estimated to induce
breakdown for various materials
[12,
19]
.
The other effect of higher electron dose that appears in the DIFF images is
the bright rim of the hole in the carbon film (Fig. 6c). Sometimes this
phenomenon can be seen at a rather low dose of 0.1 e/Å2, if
the annulus is less than about one micron from the edge of the hole (Figs. 2c,
4b, 4g and 6b). The bright fringe is confined to only the hole that contains
the charged patch. This fringe is probably caused by negative charge,
that has accumulated near the edge of the hole due to the presence of positive
charge inside the hole. The fact that the fringe is limited to a single hole -
it does not appear on the other side of the affected parts of the carbon net -
is reminiscent of shielding, presumably because the carbon is conductive
enough. The negative charge must cause an opposite phase shift as compared to
the positive charge within the annulus, since the negatively charged rim shows
up as a bright feature in the DIFF images.
What may be done to minimize charging?
Effects due to charging appear to increase in magnitude over a range of
exposures starting as low as 0.1 e/Å2 and up to 5
e/Å2 (Figs. 4 and 6). Therefore, it seems unlikely that a
biological specimen could be conditioned using some level of "pre-irradiation".
Moreover, even at steady-state, charging and discharging is a dynamic process
as judged from the "bee swarm effect"
[19],
so that the imaging conditions are more than likely to be impaired.
Carbon film is generally considered to be a good conductive support for
biological specimens. Yet, we were able to observe charging phenomena on it as
well (Fig. 8). We needed to extensively reduce the excitation of the objective
lens in order to accomplish this, suggesting that the amount of charge on the
carbon film is less than that on thin films of glucose with or without protein
spanned across a hole (Fig. 4). At different levels of electron irradiation,
the amount of charge on carbon appears to remain constant, since the contrast
of the annulus does not change (Fig. 8). In addition, the area with high
contrast does not increase in diameter under these conditions, as it does on
glucose. This suggests that breakdown conditions on the carbon film have been
reached, which can be explained assuming that the carbon film is a 2-component
system, one component being an insulator and the other a conductor. When
breakdown is reached, the charge could effectively dissipate through the
conductive component of the carbon film to ground. The effect of the remaining
amount of charge on electron diffraction and imaging appears to be small enough
to allow high resolution diffraction patterns (Fig. 2) and images to be
recorded from protein crystals prepared on carbon film
[20-22].
The presence of thin, continuous carbon film thus appears to considerably
reduce the charge but not eliminate it completely. In cases where a holey film
rather than a continuous one must be used because of preparative constraints
[23,
24]
,
charging can be reduced by including an edge of a hole in the area irradiated
by the electron beam (Figs. 1 and 2). Alternatively, the biological specimen
can be coated with a conductive film after glucose or vitreous ice embedding
[2,
25]
.
However, a more extensive study is needed to determine the full extent of
benefit in data collected for a full structural analysis.
Conclusions
We have discussed various ways in which charging was detected on biological
specimens embedded in glucose and on vitrified water. These methods are
equally well applicable to vitrified specimens. We showed that an increase in
contrast was observed on a charged specimen using imaging in defocused
diffraction mode and at low magnification. This increase in contrast was due
to a positive charge on the specimen and could be reduced by including a region
of carbon film into the irradiated area. According to our estimation, the
charge on a specimen at moderate exposures could reach a level that induces
electrical breakdown, causing the charge to spread out over a larger area than
that irradiated by the beam. We suggest that the electrical properties of the
specimen rather than electron microscope instrumentation now represents the
barrier to be overcome in high resolution electron cryomicroscopy as a
routine structural tool.
Acknowledgments
This research was supported by grants from the National Center for Research
Resources of National Institutes of Health (RR02250), the R. Welch Foundation
(Q1242) and the W. M. Keck Foundation. We thank Dr. R.M. Glaeser for helpful
discussions. JB* would like to thank Dr. J. Cazaux for
advice on excited insulators. Finally, we wish to thank Dr. P. Thuman-Commike
and Mr. J.A. Lawton for critical reading of the manuscript.
References
1.
B. Böttcher, Ultramicroscopy 58 (1995) 417.
2. U. Jakubowski, W. Baumeister and R. M. Glaeser, Ultramicroscopy 31 (1989)
351.
3. R. Henderson, Ultramicroscopy 46 (1992) 1.
4. D. H. Warrington, J. Sci. Instrum. 43 (1966) 77.
5. J. Pawley, J. Microsc. 136 (1984) 45.
6. N. G. Wrigley, J. Ultrastruct. Res. 24 (1968) 454.
7. T. W. Jeng and W. Chiu, J. Mol. Biol. 164 (1983) 329.
8. A. Fukami and K. Adachi, J. Elec. Microsc. 14 (1965) 112.
9. J. Dubochet, M. Adrian, J. J. Chang, J. C. Homo, J. Lepault, A. W. McDowall
and P. Schultz, Q. Rev. Biophys. 21 (1988) 129.
10. J. Brink and W. Chiu, J. Struct. Biol. 113 (1994) 23.
11. P. E. Mooney, G. Y. Fan, K. V. Truong, D. B. Bui and O. L. Krivanek, in:
Proc. XIIth Int. Cong. Electron Microsc., eds. L. D. Peachey and D. B. Williams
(Seattle, WA, 1990) p. 164.
12. G. H. Curtis and R. P. Ferrier, Brit. J. Appl. Phys., Ser. 2 2 (1969)
1035.
13. M. S. Isaacson, Principles and Techniques of Electron Microscopy, Vol. VII,
ed. M. A. Hayat (Van Nostrand-Reinhold, New York, 1977) p. 1.
14. C. A. Walsh, Ultramicroscopy 45 (1992) 85.
15. J. Cazaux, Ultramicroscopy 60 (1995) 411.
16. L. W. Hobbs, Scanning Microsc. Suppl., Vol. 4, eds. J. Schou, P. Kruit and
D. I. Newbury (AMF O'Hare, Chicago, 1990) p. 171.
17. M. K. Lamvik, S. D. Davilla and J. Tuttle, Scanning Microsc. Suppl., Vol.
3, eds. R. M. Albracht and R. L. Ornberg (AMF O'Hare, Chicago, 1989) p. 271.
18. A. J. Dekker, Solid State Physics, Vol. 6, eds. F. Seitz and D. Turnbull
(Academic Press, New York, 1958) p. 251.
19. D. B. Dove, J. Apply. Phys. 35 (1964) 1652.
20. J. Brink, W. Chiu and M. Dougherty, Ultramicroscopy 46 (1992) 229.
21. T. Soejima, M. B. Sherman, M. F. Schmid and W. Chiu, J. Struct. Biol. 111
(1993) 9.
22. M. B. Sherman, J. Brink and W. Chiu, Micron 27 (1996) 129.
23. M. Adrian, J. Dubochet, J. Lepault and A. W. McDowall, Nature 308 (1984)
32.
24. E. W. Kubalek, R. D. Kornberg and S. A. Darst, Ultramicroscopy 35 (1991)
295.
25. R. S. Rader and M. K. Lamvik, J. Microsc. 168 (1992) 71.
Figure legends
Fig. 1: Specimen charging as judged from changes in the beam diameter. A
series of frames extracted from a video sequence obtained using the Gatan
wide-angle TV-rate CCD camera -- the entire sequence is available through the
World Wide Web[1] -- demonstrating the change
in beam diameter when
the edge of the beam is moved from the holey carbon film (a) into the hole
containing an unsupported layer of ice (b through i). The direction in which
the specimen is moved is indicated by the arrow (a). The beam diameter has
been indicated by the arrowheads. As the beam is moved from the holey carbon
onto the ice layer (b), the beam diameter increases. This increase is ~20 % as
indicated in (h). As the beam returns onto the holey carbon, now on the
opposite side of the hole (j), the beam diameter reduces to its original width
(k and l). The changes in beam diameter can occur just before the beam leaves,
or just after it moves back onto the holey carbon. Bar is 1 um.
Fig. 2: Specimen charging as judged from (a) the blurring of Bragg reflections,
(b) the expansion of the undiffracted beam in electron diffraction patterns
acquired from glucose-embedded catalase crystal using a 2.5 um diameter beam
without irradiating part of the carbon support. (c) DIFF image acquired
immediately afterwards from a larger, 7.5 um-diameter reveals a dark annulus.
(d-f) The EDP, undiffracted beam and DIFF image from the same specimen but
including a carbon film area show sharp Bragg reflections out to 3 Å, a
sharper central beam spike and a less pronounced footprints, respectively. The
bar is 1 um.
Fig. 3: Footprints of the beam observed on glucose-embedded crotoxin crystals
prepared on continuous carbon film. (a) DIFF image recorded from a 7.5
um-diameter beam after acquiring an EDP from a 2.5 um-diameter area. A
footprint indicated by the arrow coincides with the area irradiated by the
electron beam. (b) After spot-scan imaging the same crystal, a raster of small
annuli (arrows) can be seen, each with a diameter of approximately 2000 Å
corresponding to the beam diameter used for spot-scan imaging. Bar is 1 um.
Fig. 4: Effects of charging and discharging. (a) A 7.5 um-diameter DIFF image
obtained at 2.4 x 10-4 e/Å2 from a glucose-embedded
catalase crystal (dark arrow) suspended across a hole without a carbon support
film. This image is recorded after being irradiated using a 2.5 um-diameter
electron beam that does not illuminate the carbon net. (b) A DIFF image
acquired as in (a) from this area reveals a dark annulus (see line profile)
with a diameter of 2.5 um (single white arrowheads) and a width of ~ 2000
Å (double white arrowheads). This annulus disappears upon acquisition of
subsequent DIFF images in this "charge-discharge series" as judged from its
decreasing contrast (c-e). The annulus can be re-generated by repeating this
experiment effectively demonstrating that the annulus is not due to
contamination (f-j). Bar is 1 um.
Fig. 5: Effect of diffraction lens focus on the appearance of the annulus. (a)
Ray diagram explaining the electron-optical effect of charge on the specimen.
The electron beam (solid lines), which irradiates a specimen (SP), is focused
by the objective lens (OL) at the back focal plane (BFP). A magnified image of
the diffraction pattern is produced further down the column when the
diffraction lens (DL) is focused on this BFP. The presence of charge on the
specimen effectively increases the field strength of the objective lens,
meaning that the beam (dash-dotted lines) will now converge in a new plane
(BFP'), which is closer to the specimen. Two planes are indicated at equal
distance from the old BFP indicating equal amounts of defocus applied with the
diffraction lens. Because charging decreases the objective lens' focal length,
the overfocus pattern will expand. Conversely, at underfocus, the rays have
converged more and consequently the DIFF image will reduce in size. Note that
this applies to the charging portion of the specimen only. (b) DIFF image
obtained as in figure 4b at the overfocus condition as indicated in (a) from a
7.5 um-diameter specimen area consisting of a thin glucose film spanning the
holey carbon after a 2.5 um-diameter area is exposed. The image shows a dark
annulus. (c) Similar as in (b), but now the DIFF image is obtained at the
underfocus position, as indicated in (a). This image shows a bright annulus.
The reversal in contrast of the annulus at opposite focus settings of the
diffraction lens indicates that the charged patch is a phase object. The DIFF
images reveal the expected point inversion since they are acquired at different
sides of the objective lens cross-over. Bar is 1 um.
Fig. 6: Effects of different electron exposures on the appearance of the
annulus. (a) DIFF image showing an overview of the glucose-embedded catalase
specimen acquired as in figure 4a. (b) First DIFF image of a "charge-discharge
series" obtained as in figure 4b from the same area. The image shows a clear
dark annulus indicative of charging. (c) Similar as in (b), but now after the
area is exposed to 5 e/Å2. The image reveals an increase in
contrast in regions well outside the "charging" beam. The edge of the hole has
turned bright, an effect which is also recognizable in (b). The charge
effectively increases the magnification by approximately 20 %, as judged from
the dimensions of the catalase crystal. Bar is 1 um.
Fig. 7: Re-distribution of charge. A specimen consisting of catalase crystals
suspended in glucose across holey carbon film is evaluated for completeness of
discharging. (a) A DIFF image obtained from a 7.5 um-diameter area shows a
dark annulus of 2.5 um in diameter after the specimen has been charged as in
figure 4b. DIFF images are subsequently acquired from this area until the
annulus has disappeared. (b) First DIFF image from a second discharge series
obtained from the same specimen area using a wider, 12.5 um-diameter electron
beam. The image shows an annulus 7.5 um in diameter. The presence of an
annulus in the second discharge series demonstrates that the charge does not
dissipate completely, but rather gets re-distributed over the larger area. Bar
is 1 um.
Fig. 8: Charging and discharging of plain holey carbon film. (a) A 25 um
diameter area is exposed to 10 e/Å2. The grid square with
this area is imaged at low magnification (< 200x) at a 5.3 millimeters
objective lens defocus. The image reveals a footprint of the beam 25 um in
diameter (arrow). The insert shows the intensity profile along the dashed line
indicating that the footprint is an annulus. (b) When a smaller, 2.5
um-diameter area within the larger region is exposed to 150
e/Å2, the low magnification image obtained as in (a) reveals
two footprints, a small one indicated by the arrow, and a larger one similar to
the one in (a) but fainter. The smaller footprint is also an annulus
(intensity profile insert). (c) At an accumulated dose of 650
e/Å2 in the smaller area, the image shows that the larger
footprint has nearly completely faded. This footprint disappears completely at
an accumulated dose of 1500 e/Å2 in the smaller area.
Throughout this experiment the contrast of the smaller footprint remains
roughly constant. Bar is 5 um.
Fig. 9: Schematic diagram depicting the charging and discharging process. (a)
The specimen (S) is shown spanned between two grid bars (G),
which are grounded through the microscope stage. It is irradiated over a
circular area with diameter D with primary electrons,
I0. The elastic and inelastic scattering processes that
occur in the specimen give rise to a scattered portion of the primary beam
(Ie + Ii), and an attenuated primary beam
at the exit face of the specimen (It). The inelastic
scattering results in emission of secondary and Auger electrons
(Is). (b) This emission renders the specimen positively
charged across an area with diameter D. (c) The positive charge can be
removed in two different ways. When the specimen area is irradiated with a
wider beam, for instance 3 x D, part of the secondary and Auger
electrons that are emitted across this area are pulled in the charged patch
where the charge is neutralized. Charge can also be removed through
dissipation by a specimen current (Id) between the patch and
the gridbars.
Fig. 10: Charge dissipation with time is evaluated using specimens of
glucose-embedded crotoxin complex crystals on a thin, continuous carbon film.
DIFF images are obtained as in figure 4b after different amounts of time elapse
between charging the specimen and acquiring the DIFF images. (a) DIFF image
obtained as in figure 4b 5 s after charging it. The image shows the crystal
and a relatively faint 2.5 um-diameter annulus (arrow). (b) The specimen is
discharged by opening the shutter on the slow-scan CCD camera manually for 30
s, and the experiment is repeated using a 5-min interval between charging and
recording the images. (c) Similar as in (b), with a 15-min time interval. The
images reveal that the contrast of the annulus reduces over time, which is
indicative of charge dissipation through a specimen current. Bar is 1 um.
Fig. 11: Charge dissipation at different specimen temperatures evaluated on
specimens of catalase crystals prepared as in figure 4. The specimen is
charged at different temperatures after being more than 15 min at the indicated
temperature. (a) DIFF image obtained as in figure 4b shows the specimen at
-167oC with a clear annulus of 2.5 um in diameter. (b) Similar as in (a),
but the specimen is at -30oC. The image shows a much fainter annulus. This
data indicates that at higher specimen temperatures charge dissipation occurs
more readily, possibly due to an increased specimen current. Bar is 1 um.
[1] Available at URL
http://ncmi.bioch.bcm.tmc.edu/~brink/papers/charging/charging.html
[2] The electron flux is 7 pA at a 2 s exposure,
the yield of secondary and Auger electrons of 10-4,
= r x
0 with
r = 3.3,
0 = 8.85 x 10-12
N-1m-2C2,
= 10-8
-1m-1,
a 2.5 um diameter beam and a thickness of
1000 Å of glucose.
Fig. 1: Specimen charging as judged from changes in the beam diameter. A
series of frames extracted from a video sequence obtained using the Gatan
wide-angle TV-rate CCD camera -- the entire sequence is available through the
World Wide Web[1] -- demonstrating the change
in beam diameter when
the edge of the beam is moved from the holey carbon film (a) into the hole
containing an unsupported layer of ice (b through i). The direction in which
the specimen is moved is indicated by the arrow (a). The beam diameter has
been indicated by the arrowheads. As the beam is moved from the holey carbon
onto the ice layer (b), the beam diameter increases. This increase is ~20 % as
indicated in (h). As the beam returns onto the holey carbon, now on the
opposite side of the hole (j), the beam diameter reduces to its original width
(k and l). The changes in beam diameter can occur just before the beam leaves,
or just after it moves back onto the holey carbon. Bar is 1 um.
