II.A.1. Support Films Support Films

Whether the specimen is a tissue section or a suspension of particles, it must be supported on a thin, electron transparent film. Layers of evaporated carbon 20-50nm thick are almost universally used for this purpose since they:

- are relatively transparent- have limited granularity- are relatively stable to irradiation.

a. The TEM grid

The specimen and supporting film require the mechanical support of a metal grid which is necessarily electron opaque. The number of grid bars and percent open area can vary quite widely with different types of grids (Fig. II.1). 200-400 mesh (lines/in) grids are commonly used. Most EM grids are made of copper because it is non-ferromagnetic and thus minimally distorts the magnetic field of the objective lens. Even so, it is usual practice in high resolution studies to avoid recording images of specimens which lie close to the grid bars.

Modern grids are generally made by a photographic electrodeposition process which makes it easy to produce cheap, disposable grids with a wide variety of meshwork patterns. The copper mesh also rapidly conducts heat away from the support film and helps prevent thermal expansion and hence movement of the specimen under electron irradiation. Grids may be used bare or, more commonly, filmed depending on the nature of the specimen being studied (Fig. II.2). Thin sectioned specimens, for example, are sometimes examined on bare grids, but for highest resolution work, the sections must be mounted on filmed grids or, better yet, on net films (see Sec. II.A.f).

b. Plastic or carbon support films

TEM grids have holes small enough to allow an extremely thin (20 nm or less) film of a suitable electron-transparent substance (plastic or evaporated carbon) to be stretched across them without breaking when gently handled or when irradiated with the electron beam. Because the support film is often as thick or thicker than the specimen itself, contrast introduced by the support may be comparable to that of the specimen. Amplitude contrast in the support film is generally negligible when thin support films such as 20-50 nm carbon are used. At small to moderate levels of defocus (200-500 nm underfocus) phase contrast produces a granularity in the support film image which is superimposed on the specimen image and may lead to incorrect interpretation of specimen features. This is due to the formation of a phase contrast image of the randomly arranged carbon atoms in the film (with interatomic distances of about 0.27 nm).

Fig. II.1. A representative selection of commonly used 3.05 mm grids. (From Meek 1st ed., p.322)

Fig. II.2. Grid for electron microscope specimen. (From Slayter, p.391)

Films are generally one of three types:

- plain plastic such as collodion or Formvar- plastic coated (stabilized) with evaporated carbon- plain carbon

Plain carbon films are generally the choice for highest resolution work because:

- they are tough and withstand the electron beam.- they conducts electricity and hence reduce specimen drift due to charge effects.- they conduct heat from the specimen to the rid bars, thus reducing thermally induced drift.

c. Preparation of plastic and/or carbon films

There are many recipes for preparing good support films. Each EM user generally masters two or three suitable procedures. The nature of the specimen to be examined determines to some extent what type of support film is best to prepare. Plastic films are easier to make than pure carbon films. The most common plastics used are collodion and polyvinyl formal (trade name Formvar). Carbon films are preferred for use in studies where resolution is a critical factor.

Plastic films are generally made by preparing a dilute (<0.5%) solution of the plastic in a non-polar solvent (ethylene dichloride or chloroform is used with Formvar, and amyl acetate is used with collodion). A thin (<20 nm) film of the plastic is then prepared by one of two standard techniques:

Fig. II.3. Stripped films. (From Hall, p.286)

- Dip a cleaned glass microscope slide into the solution, let it drain and dry. Then float the plastic film from the surface of the slide by slowly dipping it into a vessel of water (Fig. II.3).- Place a drop of the solution on the surface of water and allow it to spread out and dry. This leaves a thin layer of plastic on the water surface. This method tends to produce thicker and more uneven films compared to the first method (not illustrated).

The film floating on the surface of the water should appear an even grey to dark grey color when viewed by reflected light. If it appears yellowish or uneven, it is too thick and unsuitable. The film is transferred to grids by laying the EM grids shiny-side up (dull side down) onto the plastic film. The film and grids are picked up with a piece of hard filter paper or with Parafilm or with a glass slide and then set aside in a dust free place to dry (Figs. II.4, II.5). Alternatively, the grids can be submerged beneath the plastic and lifted up through the film (Fig. II.6). The grids can then be used as is or they may be coated with an evaporated layer of carbon.

Fig. II.4. One method of preparing filmed grids. (From Hall, p.284)

Fig. II.5 The preparation of Formvar specimen support films: (a) a cleaned glass slide is dipped in the solution of Formvar in chloroform; (b) the slide is withdrawn and drained in the presence of chloroform vapor; (c) the thin plastic film is floated onto a water surface; (d) after placing specimen grids (arrows) on the film, it is removed from the water using a strip of paper. (From Willison and Rowe, 3.3, p.64)

Fig. II.6. Preparation of filmed grids. (From Hall, p.284)

Carbon-Formvar films can be easily converted into carbon-only films by dissolving away the plastic. The carbon-plastic films are placed plastic-side down onto a piece of filter paper soaked in a solvent which dissolves the plastic. After a few hours, the grids are moved to a dry piece of filter paper and allowed to air dry.

Another method for preparing pure-carbon films is to evaporate a layer of carbon directly onto the surface of a freshly-cleaved piece of mica. The carbon is then floated off onto water and transferred to EM grids in one of two ways:

- TEM grids can be brought up through the water from below the film or- the film may be carefully lowered onto grids, situated beneath the water surface on a piece of wire mesh, by slowing removing the water from the vessel.

d. Carbon evaporation

The apparatus used to evaporate thin layers of carbon consists of a bell jar that is evacuated by a diffusion pump which is backed by a mechanical pump (Figs. II.7, II.8). Inside the vacuum chamber is a base plate with vacuum and electrical feed throughs. The mechanical rotary and oil diffusion pumps are used in series to produce a vacuum of about 10^-5 torr which is low enough so the mean free path of vaporized atoms is long compared with the distance they are to travel. Carbon is evaporated by passing a current of 50 Amps through two rods (Figs. II.9-II.11), one sharpened down to 1 mm diameter and the other 3 mm. The two rods are held in contact by a spring while the current flows through the rods. Heat produced at the junction is sufficient to cause rapid evaporation. The amount evaporated can be estimated by the decrease in length of the narrow portion of the carbon rod or by noting the blackening of a piece of filter paper in the region surrounding the shadow cast by a small object (e.g. thumb tack) placed on the paper.

Fig. II.7. The vacuum system of an Edwards 12E1 evaporator. The backing tank can be used for desiccating photographic material for the elctron microscope. (From Meek 1st ed., p.473)

Fig. II.8. Apparatus for evaporation of substances in a vacuum. (From Hall, p.279)

Fig. II.9. Carbon rods for the evapora-tion of (a) C and (b) C-Pt. (From Hall, p.282)

Fig. II.10 Three possible arrangements of the two graphite rods constituting the carbon 'arc'. (From Willison and Rowe, p.34)

Fig. II.11. A typical graphite 'arc' used in carbon evaporation. Electrical supply (es); tensioning spring (ts); sharpened ends of the two graphite rods constituting the 'arc' (g)' supports for the 'arc' within the coating unit (sup). (From Willison and Rowe, p.34)

e. Preparation of holey films

Carbon-Formvar films pierced with a large number of circular holes with sharp edges and diameters in the range 0.1-1.0 µm make excellent test objects for astigmatism corrections and resolution checks (Fig. II.12a,b). A variety of methods are used to prepare holey films. Unfortunately, there aren't any "fool-proof" methods as seems to be true for preparing regular films. There are two basic protocols for producing the holes:

- Minute droplets of water are condensed onto the surface of a drying Formvar film. As the film dries, the water droplets pierce the film leaving circular holes.- Formvar is dissolved in a solvent containing water. This mixture, containing a non-polar solvent for the Formvar (e.g. Ethylene dichloride) plus a partially polar solvent miscible with both water and Formvar (e.g. Glycerol) will dry such that the more volatile solvents evaporate first, leaving minute water droplets to pierce the film.

The TEM user should be able to distinguish true holes and pseudo-holes, which, at low magnification, appear to be suitable, but, on closer inspection at high magnification, are found to have a thin film of Formvar over them. Pseudo-holes can sometimes be removed by dipping the filmed grids in acetone for a second or two, or, for better control, the filmed grid is held over acetone vapors for a few seconds.

The holey Formvar film is detached from the microscope slide onto a water surface, the TEM grids are applied, picked up, dried, and a layer of evaporated carbon is added in the same manner as is done for regular films.

f. Formvar nets

These are distinguished from holey films by the size range of the holes formed (usually >1 µm in diameter). A "net" generally has more area occupied by "hole" than by film (Fig. II.12d). This type of film is excellent for studying extremely thin sections where areas of the section can be photographed at the highest possible resolution through holes in the net. Nets are prepared in basically the same way as holey films. The simplest procedure is to hold a drying Formvar slide over the surface of boiling water for about one second. A fairly thick (50 nm) layer of carbon is generally evaporated onto net films to give them added stability for supporting sectioned material.

Fig. II.12. The 'holey carbon film' test object. (top left) is a low-power micrograph of a well-prepared film with plentiful holes in the right size range (X 1,000); (top right) is a higher power view (x 3,600), showing almost circular holes with clearly demarcated edges between 1.0 and 0.1 mm in diameter. The thickness of the film is even, right up to the hole edges; (bottom left) is an example of a very poor film with partly perforated holes of too large a diameter, with a few ragged small holes in the septa; (bottom right) is the result of carrying the last panel to the extreme - a carbon-Formvar 'net', which is sometimes used as a support for extremely thin epoxy resin sections when the very highest resolution is required from them. (From Meek 1st ed., p.325)