The case for low voltage high resolution scanning electron microscopy of biological samples.

Abstract

Dried biological samples are low in scattering power, non-conducting and sensitive to radiation damage. These facts complicate the choice of the optimum beam voltage Vo at which they should be observed in the scanning electron microscope (SEM) because they add as variables the type and thickness of the coating material and degradation/contamination of the specimen by the beam. Heretofore, high resolution SEM could only be carried out at relatively high Vo (20-30kV) because available equipment could not produce small beam diameters at low Vo. Modern instruments can produce beam diameters of about 3nm at 1.5kV. As normal preparative procedures (fixation, critical point drying, coating) are unlikely to preserve reliable structure below this level, it is now possible to investigate the possible advantages associated with low Vo operation such as a reduction in charging and radiation damage and improved topographic contrast. The conclusion recommended by this paper is that the term resolution needs careful definition. The size of the smallest features visible in a micrograph is a function of many variables. Although probably the most important is specimen preparation, a number of others (probe size, beam penetration range, contamination, coating thickness needed to provide contrast and avoid charging etc) are functions of Vo. Of these variables at least probe size and possibly contamination become more favorable at higher Vo while the remainder favor low Vo. As a result the optimum will occur at a Vo where the best balance of these factors occurs for a particular sample. When using the Hitachi S-900, we have found that the optimum seems to be at 1.5-2.5kV for topologically diverse samples, but may extend to 5kV on samples on which very small structural details have been preserved and which are relatively stable to radiation damage.