When one decides to use electron microscopy to investigate a specimen, the end goal of the project must be evaluated to properly choose the right path to obtain that goal. Some applications where the scanning electron microscope would be the instrument of choice might be: studies involving the exterior morphology of the sample, the localization of large (20-30 nm) colloidal gold markers on the surface of the sample, the localization of boundaries between regions of differing atomic number composition, and the qualitative and quantitative identification of the elemental content of the specimen. Each of these applications requires that the instrument be operated properly so as to maximize the excitation and collection of the desired signal.

All electron microscopes are high-vacuum instruments. Vacuums are needed to prevent electrical discharge in the gun assembly (lightning), and to allow the electrons to travel within the instrument unimpeded. There are many scales to measure vacuum levels, some being: mm/Hg, Pascals, Torr, and atmospheres. One undisputed area of vacuums is cost. If higher vacuum levels are desired, better pumping systems are required. Better systems cost more money. Why would one want better vacuums?

When designing the microscope, we started with the vacuum. The electron microscope source to be used should be a factor in the design of the vacuum system. Poor vacuum levels shorten the life of the electron emission source. Saving money in designing the vacuum system might prove costly if filaments are consistently in need of replacement. Also, any contaminants in the vacuum can be deposited upon the surface of the specimen as carbon. Cleaner vacuums will minimize this artifact.

Different electron emission sources require different vacuum levels. There are 2 classes of emission sources, thermionic emitters and field emitters. Thermionic emitters use an electrical current to heat up the filament which lowers the work function of the filament material. When the work function is lowered, electrons can be more readily drawn off of the filament with an electric field. The two most common materials used for filaments are tungsten and lanthanum hexaboride. Cold cathode field emission sources do not heat the filament material. The electrons are drawn from field emission gun by placing the filament at a huge electrical potential gradient, so large that the work function of the material is overcome, and electrons are drawn off of the filament. Field emission systems require extremely high, clean vacuums in which to operate.

In the thermionic system, the filament is inside a metal (wehnelt or cathode) cap. An electronic resistor is placed in the electric supply line to the filament. This resistor causes the cathode cap to be more negatively charged than the filament. Electrons have a negative charge associated with them. Since like charges repel each other, the negative cap repels the electrons into a cloud around the filament. An anode plate is located below the wehnelt assembly in the microscope. The anode is at ground potential. The electrons in the gun assembly are attracted to the anode, thus leaving the cathode cap. The negative charge of the cathode cap has a focusing action on the electrons as they leave the gun. This focusing is due to flux lines, which are also referred to as lines of equipotentials. These lines make the cathode cap an electrostatic lens (compared to electromagnetic lens).

Field emission sources utilize two anode plates placed below the gun assembly. Associated with the first anode is the extraction voltage. The extraction voltage is usually in the range of 3-5 kilovolts, and is the amount of voltage necessary to draw electrons from the source. The second anode has the accelerating voltage associated with it. The accelerating voltage determines the velocity at which the electrons travel down the column. Both of these anodes act as electrostatic lenses, focusing the beam into a small initial crossover.

Two factors associated with the gun determine the resolution capability of the instrument. First, resolution must be defined. Resolution is the ability to separate (resolve) two closely spaced points (particles) as two separate entities. The two factors that determine resolution in the scanning electron microscope are accelerating voltage and initial crossover diameter.

Without getting into physics, we can generalize Abbe's equation as stating that the resolution of an instrument is dependent upon the wavelength of its illumination source. The accelerating voltage of a scanning electron microscope is variable, usually in the range 500-30,000 volts. An electron accelerated by a potential of 30Kv has a shorter wavelength than one accelerated by a 5Kv potential. Thus, the 30Kv electron should give us better point to point resolution. When we discuss specimen-beam interactions, we will contrast the difference between point to point resolution and surface resolution.

The other component of resolution, initial crossover (beam diameter) has many names. Some texts refer to it as the virtual source, others as do (o is subscript). To resolve a feature on the surface of a specimen, the beam must still have a smaller diameter than that feature, yet still contain enough electrons (referred to as beam current density) to generate acceptable amounts of signal (to be discussed with specimen-beam interaction/signal formation). The smaller that the initial crossover is, the less that the electromagnetic lenses have to work to demagnify the beam into a usable probe. A thermionic system operating with a tungsten hairpin filament will have a crossover diameter of about 50 microns. A thermionic system equipped with LaB6 (lanthanum hexaboride) electron source will have a crossover of about 10nm, while maintaining high beam current densities (brightness). Without any subsequent focusing action of electromagnetic lenses, the field emitted has a probe that can be used for imaging purposes. This makes the field emission system the highest resolution instrument.

So why doesn't everyone use field emission SEMs? Cost and need. Not everyone can afford to purchase a FESEM. Also, there is the phenomenon of empty magnification. On some specimens, biological samples, as an example, there is a magnification that if exceeded, no useful information will be gained. The higher magnifications and resolutions of the FESEM is not always desired or required.

There is an inherent fluctuation of emission with the FESEM. Any contaminants that come to rest on the filament cause the emission current to fluctuate. This prevents the FESEM from being interfaced with a quantitative microanalyzer, since a prerequisite of good quantitative analysis is a steady beam current.

FESEMs are extremely useful in low voltage applications. The design allows for coherent beam formation at low (500-3,000) voltages. FESEM technology has been used extensively by the semiconductor industry for quality control. An instrument that could check the progress of water fabrication without damage was needed. The FESEM operating at low voltages fit this niche.

So far, we have a vacuum and an electron beam generated and headed down the electron column. To assist us in the refinement of the electron beam, the electromagnetic lenses are employed. The path of an electron can be altered by exposure to a magnetic field. Electromagnetic lenses create a circular magnetic field that demagnify (condense) the electron beam as it passes through. The strength of the lenses can be changed by varying the current supplied to the lens. Changing the lens current changes the focal length of the lens.

There are some problems inherent in electromagnetic lenses. They are spherical aberration, chromatic aberration, diffraction and astigmatism. It is possible to correct for these problems, once one understands them.

Spherical aberration concerns the path of the electron in respect to its position within the electromagnetic lens. The strength of the magnetic field is the strongest near the surface of the lens. Therefore, the electrons that travel through a lens close to its surface will have their paths altered more than an electron which travels through the center of the lens and results in a loss of electrons from the beam (electrons that strike the interior surface of the electron column are absorbed).

Chromatic aberration relates to the varying energy of the electrons that comprise the beam. Not all electrons generated at the gun assembly have the same energy. Electrons with different energies have different wavelengths. A magnetic field will have more of an influence on a longer wavelength electron. Thus, due to a variance in the energies of the electrons in the beam, it is not focused to a discrete focal point.

Diffraction is of most importance at the final probe forming lens. Diffraction is caused by the electrons wavelengths being out of phase. Thus, a lens will focus electrons of different phase to a different point depending upon the position of an electron in its wavelength when it passed through the lens. To correct diffraction, the electrons would have to be monochromatic and coherent. If these conditions were met, the lens could focus the stream of electrons to a point rather than to a disk of confusion.

Astigmatism occurs because of manufacturing imperfections within the electromagnetic lens. It is extremely difficult to manufacture an electromagnetic lens that forms a perfectly even magnetic field. Variances in the strength of the magnetic field around the circumference of the lens causes an elliptical focal spot. This problem can be overcome by segmenting the lens into many parts and being able to adjust the excitation to each part. In this manner, weak portions of the lens can be strengthened.

Along with the electromagnetic lenses, metal aperture strips can be used to refine the beam. It is common for the modern SEM to be equipped with variable condenser and objective lens apertures. These variable aperture strips will have a variety of pinhole sizes to choose from. It is the responsibility of the operator to choose the proper aperture size. To generalize, small objective aperture sizes will produce images with good resolution, good depth of field and minimal charging. This last statement requires two definitions. Depth of field refers to the ability to have a large change in specimen topography appear to remain in focus. SEMs allow for the specimen to be translated along the x, y and z axes. The z axis is defined as the distance from the final lens to the surface of the specimen. This distance is also referred to as the "working distance." Long working distances and small apertures provide an image that appears to be in focus over a large change in z. Having dealt with depth of field, a brief statement on charging is in order. An ideal specimen would be conductive. If a less than ideal (non-conductive) specimen is attempted to be viewed, beam electrons can build up on the surface of the specimen, resulting in the phenomenon known as charging. Charging will be further explained in specimen-beam interaction/image formation.