Over recent years there has been relatively little experimental data produced by environmental scanning electron microscopy on biological applications. The author’s laboratory is situated in a multi-user facility based in a School of Biological Sciences and has examined a wide variety of samples over a 5 year period. The purpose of this paper is to present a number of biological applications, one in detail, in an attempt to illustrate the potential and pitfalls of the instrumentation.

A study has been made of the elastic bands used in dental orthodontics as a method of moving teeth. It is common clinical practice to replace the elastic bands at 24 hour intervals as it is known that the bands lose their elasticity and thus effectiveness during this period. The present study has examined the structure of such bands at time intervals after stretching to clinically used extension on a former and immersing the bands in water.

Full characterization of materials in the environmental scanning electron microscope (Environmental SEM) often requires chemical analysis by X-ray energy dispersive spectroscopy (XEDS). However, a major problem arises because the spatial resolution of the XEDS signal is severely degraded by the gaseous environment in the sample chamber. The significant fraction of the primary electron beam is scattered after it passes through the final pressure limiting aperture and before it strikes the sample. Bolon and Griffin have both published data that illustrates this effect very well. Bolon revealed that 45% of the primary electron beam was scattered by more than 25μm in an Environmental SEM operating at an accelerating voltage of 30kV, with a water vapor pressure of 3Torr and a working distance of 15mm. Griffin’s work demonstrated that even at higher voltages (30 kV), shorter working distances (<10mm) and lower chamber pressures (2Torr), there is a significant fraction of the electron beam scattered out to over 400 μm away from the point where the primary beam strikes the sample.

The scattering of primary beam electrons has several implications for energy dispersive X-ray analysis on the environmental scanning electron microscope (ESEM). One of the most basic problems is determining the spatial distribution and fraction of primary beam electrons scattered under typical high pressure analysis conditions A method for studying the scattering of the primary electron beam in the ESEM has been demonstrated. The method involves the use of self-assembled alkanethiol monolayers (SAM) which are sensitive to damage by primary beam electrons. After irradiation, the electron damaged molecules can be exchanged out of the monolayer by immersion in a second alkanethiol solution. The spatial distribution of this second marker compound can then be imaged directly by static secondary ion mass spectrometry (SSIMS). Alternatively, only a single fluorinated thiol is used and the decrease in fluorine secondary ion signal is used to map the electron damage.

An understanding of the scattering of an electron beam as it passes through a volume of low pressure gas is of critical importance for users of Low Vacuum and Environmental Scanning Electron Microscopes (LV-SEM & ESEM respectively) The ‘skirting’ of the primary beam as a result of scattering is of particular importance in X-ray microanalysis where scattered electrons, falling onto the sample away from the probe beam, can adversely affect the spatial resolution of the X-ray signal. A number of studies have attempted to quantify the width of the probe beam experimentally and hence determine optimum microscope operating parameters. Theoretical and computational work modelling the interactions of the beam with the gas have suggested that the shape of the probe beam can be modelled by some form of Gaussian. Experimental measurements, however, suggest that the probe has a skirted form rather than a Gaussian one.

A majority of the scanning electron microscopes (SEMs) now in use are probably employed as low voltage SEMs (LVSEMs), that is to say they are operated to produce beams with energies below 5keV. This trend away from the more conventional mode of operation at 20 or 30keV has gathered momentum over the past decade and has been driven by both theoretical and practical considera-tions.

Firstly, the distance travelled by an electron falls rapidly (in fact as about E1.6 ) as the incident ener-gy E is reduced. Images generated by low energy electron beams therefore contain enhanced surface information compared to those images recorded at higher energies. Since surfaces are of great inter-est in both the life sciences and in materials science this has been a persuasive factor. Secondly, both the secondary and the backscattered electrons now come from essentially the same interaction volume, rather than from volumes which are widely different in size and shape.

Nanometer-resolution surface imaging by secondary electrons (SE) is now routinely obtainable in field-emission SEM or STEM microscopes. The resolution in SE images obtained in a modern field-emission (FE) SEM is still limited by the electron probe size although the localization of the SE generation processes sets the ultimate resolution limit that will be achievable in SE images. The combination of a FE gun with improved probe forming lenses and detection systems makes it possible to image nanometer-scale surface features at low voltages. Because of the reduced electron beam range and the new SE detection system, image contrast formation in low voltage FE-SEMs is quite different to that in conventional SEMs. In a FE-SEM, SEs are detected through the combination of a strong magnetic field of the lens and a weak electrostatic extraction field of the detector. The detection efficiency varies with the energy of the emitted SEs.

A silicon microfabrication technique has been applied toward the development of a Miniature Scanning Electron Microscope (MSEM). The fabrication technology is not only precise but is inexpensive compared to conventional methods of electron microscope construction and is easily extended to the construction of arrays of MSEMs for applications in high throughput e-beam lithography and wafer inspection.

An electrostatic electron lens consists of a series of planar electrodes with central apertures that are precisely aligned to and electrically isolated from one another. This structure is fabricated using silicon as the electrode material and Pyrex optical fibers as the insulators. The electrodes are fabricated on four inch (100) orientation silicon wafers that are patterned on both sides and anisotropically etched to form four orthogonal v-grooves and an open diaphragm with a circular aperture in the center. The apertures are formed by reactive ion etching. The wafers are then diced to create approximately 100 7 mm by 9 mm electrodes.

The surface sensitivity of both imaging and EDX chemical microanalysis of materials, micro-fabrications and nanostructures can be greatly enhanced by using low beam energies in the SEM (LVSEM). This has become more apparent with the advent of instruments which are relatively uncompromised in performance (<5nm) at a range of low voltages around 1kV, and especially ones with higher resolution (≤2.5nm) and full digital imaging to make a wide range of conditions more readily accessible (1). They yield images of natural surfaces retaining their inherent chemical and electronic image contrasts and with probe sizes comparable to those which we are familar with at high voltages, but where many samples require to be coated with metal. The traditional SEM resolution test samples for evaluating SEM imaging are far from a typical bulk SEM sample. The small heavy metal particles on a carbon support behave more like the STEM situation since the backscattering of electrons from the support, and hence the generation of type 2 secondary electrons, is very limited.

The characterization of microscopic particles on substrates is of interest in numerous scientific fields. One example is semiconductor device defect review, where the characterization of particles or defects on a device is an essential aid in tracing sources of contamination in the production process. With the ever decreasing component separation on the integrated circuits the sensitivity of SEM/EDX to detect small particles on substrates is an important issue. The minimum size of particles causing defects in semiconductor device manufacture is predicted to decrease from 0.12µm today to 0.02µm by the year 2010. The question of interest is how long the SEM/EDX analysis tool can continue to be used to identify elements present in “killer” defect particles of ever decreasing size? This paper attempts to address this question.

It has already been shown that films as thin as a monolayer can be detected with WDX provided high beam currents are available.

Many industrial catalysts have a complex geometric structure to enable reacting gases or fluids to reach as much of the active surface of the catalyst as possible. The catalyzing surface frequently consists of a complex chemical mixture of different phases produced by an evolved chemical process. The active components are often very small particles dispersed on high-surface-area supports. The catalytic properties of this type of catalyst depend on the structure, composition, and morphology of the active species as well as the supports. TEM/STEM and associated techniques have been used extensively to characterize the structure and composition of supported catalysts. Surface morphology of supported catalysts is generally examined by secondary electron imaging, especially at low incident beam energies. It is, however, frequently found that small metal particles are not usually seen in SE images because of the complication of support topography

Vacuum arcs on graphite cathodes are currently used as sources of carbon ions for the production of diamond-like films in the arc ion-plating (AIP) deposition process. Emission from these cathode sources is concentrated in very localized “cathode spots” having typically 10 (i.m in diameter for graphite cathodes. These spots carry the totality of the arc current, the remaining of the surface being unaffected by the discharge. For electron emission falling in the thermo-field emission mode, extremely high current densities up to 108 -109 Am-2 are induced generating a high localized heat flux to the surface during the spot lifetime. On metallic electrodes, this strong heat flux generates localized surface melting during the microsecond scale spot lifetime. High localized plasma pressures (>10 Atm in the case of copper) were found to exist in the cathode spot volume, leading to the co-emission of micro-droplets of the liquid metal along with the ion beam.

Load-bearing tissues in different organ systems display a wide variety of mechanical properties, yet are composed of only a few different molecules, with collagen as the principal structural protein. In bone, a mineral phase comprising carbonate-apatite crystallites confers compressive and torsional strength. The arrangement of collagen and apatite into three-dimensional hierarchical composite structures enables the tissue to bear high loads while maintaining flexibility. Moreover, bone serves as the body’s chemical storehouse for non-collagenous proteins and smaller molecular weight anions and cations such as calcium.

An understanding of the three-dimensional structure of the apatite phase at the nanometer and micrometer levels can provide insights into the properties of bone, and serve as the guide for the development of bone substitute materials. Transmission electron microscopy (TEM) has been valuable in revealing certain ultrastructural features of the mineral. However, it has not yet been possible to obtain three-dimensional reconstructions of the morphology and distribution of the apatite phase of bone using this method.

A Schottky Field Emission (SFE) SEM (Topcon DS-130F) equipped with dual stages and dual secondary electron (SE) detectors operated at high voltage (25 kV) provided useful magnifications of 200 x - 200,000 x and particle resolution in the 5 nm range from chromium (Cr) coated bulk (1 mm3) biological samples. Additionally, high voltage in-lens SFE-SEM, at magnifications > 200,000 x, provided high resolution images with 2-3 nm particle contrasts from Cr coated negatively-stained phospholipid vesicles (100 nm diameter) adhered to uncoated silicon (Si) chips.

We recently developed low voltage strategies for quality in-lens imaging of Cr coated cell cultures at low and high magnifications. Due to unfavorable specimen-beam interactions at high voltage, not encountered with ultimately small volume organic specimens on Si chips or large bulk specimens, cells on protein matrix coated Si chips were imaged at low and high magnification in-the-lens of the DS-130F with beam energies of 5 kV.

Important parameters for the detection of small colloidal immunogold particles (10 nm) on cell surfaces include the use of field emission SEM (FESEM) and a need for a conductive metal coating to reduce charging, so as to facilitate backscatter electron (BSE) imaging for unambiguous identification by atomic number contrast. Recently Heinzmann et al (1) have reported the use of below-lens FESEM for detection of 10 nm colloidal gold at working distances of 6-13 mm, but achieved success only on uncoated cells as coating with a 1 nm layer of platinum appeared to mask the gold signal. To improve specimen conductivity and signal collection, three metals (Cr, Pt, W) have been used to generate thin coatings either by planar magnetron sputtering for cryoSEM or ion-beam sputtering for routine chemically-fixed, critical point dried samples (2,3)

Three below-lens FESEM (Hitachi S-4500, JEOL 630land 63401, and LEO 982) were tested by using a sample of human neutrophils which had been chemically fixed, then immunostained for the surface antigen CD43 with 10 nm colloidal gold, and prepared by critical point drying.

High quality imaging of the macromolecular structure of biological samples can be obtained when combining high resolution SEM with advanced thin metal film coating techniques. A thin layer of fine grain metal film is often necessary to enrich the secondary electron (SE) signal generation and to enhance the contrast of features of interest, because the yield of SE signal from biological samples is low. Secondary electron signals, types SE-I and SE-II, are generated by primary beam at the impact point or by backscattered electrons (BSE) dislocated from the impact point, respectively (Peters, 1982; Joy, 1984). The SE-II signal yield depends on the amount of BSE signal and the collection ratio of SE-I/SE-II components regulates the surface topographic contrast. Therefore, the low atomic number metals with low backscattering coefficients should be considered for high resolution SEM coating. Chromium, which has both the features of low atomic number (Z=24) and sufficient SE signal yield, is an appropriate choice and was first used by Peters (1982) for high resolution SEM imaging.

Earth and Mars are the only planets in our solar system having polar ice caps that expand and contract in response to changes in the seasons. On Earth, the polar caps are composed of water-ice (snow crystals). Whereas, on Mars the extreme low temperatures at the polar cap regions result in the precipitation of CO2 ice (Martian snow) in addition to water-ice. Viking spacecraft have recorded various spectral data on this CO2 mantle and the transmission spectra for solid CO2 have been measured in the laboratory; however, little is known about the structural features of the crystals. Because previous studies have shown that low temperature SEM could be used to image structural variation of snow crystals34 and because data on the structure of CO2 crystals are needed to develop scattering and emission models of CO2 ice, we have attempted to image CO2 crystals.

Freeze-etch replicas were made from frozen-fractured plasma membranes of yeast cells that were used as a substrate for CO2 crystal growth.

Synthetic polymer latex is widely used in paints, paper coatings, adhesives, and a growing number of other water-borne coatings. All these applications require the latex particles adhere to one another and weld by interdiffusion. The processes of film formation by latex have been studied intensively during the last several decades. The three major stages of the processes are identified: consolidation, particle deformation and coalescence. In contrast, the deformation and adhesion of latex particles on pigment are poorly documented because the needed measurements are difficult to perform at particle level by conventional techniques2. Recently, the ordering and adhesion of styrene-butyl acrylate latex particles on model inorganic surfaces have been studied by atomic force microscopy (AFM). In this work, the deformation and adhesion of latex particles of styrene-butadiene on smooth substrates are examined by low voltage field emission scanning electron microscopy (LVFESEM) with contrast enhancement by staining.

Electron microscopy has been an invaluble tool to the study of polymer morphology, especially latex particles, microphase-separated block copolymers, and polymer blends. However, in order to examine these materials in an electron microscope, the sample often needs to be stained to heighten contrast between locales of different composition in the specimen. Staining usually has the following advantages: heightened contrast; higher polymer glass transition temperature Tg; lowered charging; and reduced radiation damage. However, staining can also introduce artifact. It is reported that the latex particle size is raised and the polybutadiene “sphere” in block copolymers is lowered after staining. This indicates that the staining process may be complex.

Recently, low voltage field emission scanning electron microscopy (LVFESEM), especially high-resolution backscattered electron imaging produced with staining, has been used to study latex film formation, latex particle deformation and adhesion on substrates, latex particle deformation, distribution and binding in paper coatings.

A fundamental new pathway to a basic understanding of nanostructured composite thin films has been recently afforded by advances in both high brightness field emission electron source and electron optics in the SEM. The key advantages of controlling the incident electron beam at low energies (< 5 keV) for surface imaging has been largely focused on goals such as the reduction of sample charging and minimization of macroscopic radiation damage. In this paper, we present a new opportunity to obtain sampling depth information by controlling the incident electron beam energy. One of the major factors which determines the contrast seen in SEM images is the interaction of the incident electron beam with the solid. As a result of this interaction, the secondary electron yield is proportional to the stopping power of the electron as described by the Bethe penetration depth,where dE/dl is the stopping power of incident electron with energy Einc and Emin is some suitable lower energy limit for the interaction.

The observation of microstructural features smaller than 300 nm is generally performed using Transmission Electron Microscopy (TEM) because conventional Scanning Electron Microscopes (SEM) do not have the resolution to image such small phases. Since the early 1990’s, a new generation of microscopes is now available on the market. These are the Field Emission Gun Scanning Electron Microscope with a virtual secondary electron detector. The field emission gun gives a higher brightness than those obtained using conventional electron filaments allowing enough electrons to be collected to operate the microscope with incident electron energy, E0, below 5 keV with probe diameter smaller than 5 nm. At 1 keV, the electron range is 60 nm in aluminum and 10 nm in iron (computed using the CASINO program). Since the electron beam diameter is smaller than 5 nm at 1 keV, the resolution of these microscopes becomes closer to that of TEM.