FEW AMATEURS make a hobby of electron microscopy, even though the lower magnifying power of optical microscopes has long fascinated lay enthusiasts. The failure of the electron microscope to attract a comparable following can perhaps be explained by the cost, weight and complexity of the instrument. A commercial instrument capable of magnifications of up to 200,000 diameters can carry a price tag in five figures. Such an instrument, plus its accessories, may stand six feet tall, fill a small room and weigh more than a ton. Moreover, the operation of the most powerful electron microscopes, including the preparation of specimens, can tax the craftsmanship of even professionally trained technicians.

Seemingly brash amateurs occasionally delight in accepting the challenge of just such projects They have made and operated particle accelerators, radio telescopes, lasers and a host of equally ambitious constructions-often with signal success. The electron microscope is no exception. A case in point is provided by members of the Advanced Physics Club of Brother Rice High School in Chicago. The members have made both simple and compound electron microscopes with magnetic lenses of their own design In the course of the project, which was initiated 12 years ago by Brother J. C. Crane, C.F.C., the members devised novel techniques of construction to keep the instruments within the limits of their modest budget. Two of the resulting microscopes, along with the basic principles of their operation, are described by Nicholas J. Grib, Jr. (10541 South Hamlin Avenue, Chicago, Ill. 60655). Grib, a recent graduate of the school and a past president of the club, writes:

"The difficulty and the cost of making an electron microscope increase with the magnifying power of the instrument and with its reliability, meaning its tendency to operate tomorrow exactly as it does today. The instruments we made are of modest power and reasonable reliability. We usually operate the simple microscope, which can be called an electron 'magnifying glass,' at a maximum power of about 100 diameters (about 20 times the power of an ordinary reading glass). Our best compound microscope has been operated at a maximum magnification of 10,000 diameters. (The best optical microscopes achieve maximum powers of about 2,500 diameters when specimens are illuminated with white light.)

"The maximum power that can be achieved by a microscope of any kind was shown in 1873 by the German physicist Ernst Abbe to be related to the wavelength of the illumination. Abbe thus predicted that no optical microscope, however well made, could be used for observing objects smaller than 2,500 angstrom units, or about a millionth of an inch. Smaller objects would react to light waves much as a floating table-tennis ball reacts to waves in the ocean: the minute particles would simply ride the waves instead of reflecting or otherwise diverting them.

"A half-century ago the French theoretical physicist Louis de Broglie identified a source of waves much shorter than those of light. The minute waves were associated with electrons. Within less than 10 years of de Broglie's discovery experimenters had devised appropriate lenses of two basic kinds, electrostatic and magnetic, to focus electron waves and had assembled the lenses into a remarkably powerful microscope. Electro static lenses consist essentially of two or more parallel electrodes, commonly in the form of metal disks with a centered aperture or perforation. A potential difference applied to the disks causes an electrostatic field to appear between the disks. The distribution of the field in space is analogous to the distribution of glass in an optical lens and bends the path of electron waves. Electromagnet lenses consist of a coil of wire in which a current establishes a magnetic field between iron pole pieces in the form of a ring The shape of the resulting magnetic field is comparable to the electric field of the electrostatic lens, and it bends the path of electron waves that traverse the device.

"In some ways the speeding electrons behave simultaneously as both minute particles of matter and as waves. The lenses and other working parts of electron microscopes are therefore enclosed by a housing from which air is largely exhausted to minimize collisions between the electrons and gas molecules that would deflect the waves and so prevent the formation of an image.

"With the exception of the glass parts and lenses, the instruments can be made with ordinary hand tools. The pole pieces of the lenses were machined on a metalworking lathe. If the experimenter does not have access to a lathe, the job can be farmed out to a small machine shop at modest cost. Shops that manufacture or repair neon signs can make the glass parts, although such shops rarely stock Pyrex tubing in the required sizes. The glass tubing can be bought from the Fisher Scientific Company (52 Fadem Road, Springfield, N.J. 07081) or from distributors of the Corning Glass Works. We made our own glass parts.

"The assembled microscopes can be supported by laboratory apparatus stands or improvisations of comparable design. To observe images on the fluorescent screen we operate the microscope in a dimly lighted room. The room must be darkened completely when photographic exposures are made. Two major accessories are required: a pair of air pumps capable of exhausting the microscope to a pressure of 10 torr or less and a set of direct-current power supplies for energizing the lenses and imposing a potential of at least 12,000 volts across the cathode and anode.

"Although either electrostatic or magnetic lenses can be used, we selected the magnetic type because they could be energized by available storage batteries. Electrostatic lenses require power supplies that develop relatively high voltage and include exotic devices to maintain constant output potential. Variation of the voltage, including 'noise,' alters the focal length of electrostatic lenses and degrades the resolution of the image. The voltage of fully charged storage batteries remains fixed during the few seconds required to make a photographic exposure.

"The focal length of magnetic lenses is determined by the shape and intensity of the magnetic field and is influenced by the velocity of the electrons in the electron beam. All solenoids-even a simple coil of wire that carries a current-act as lenses for focusing electrons. A lens made by winding magnet wire into a simple coil would have an excessively long focal length. For this reason our lenses included spool-shaped yokes of soft iron that terminated in annular pole pieces. The pole pieces concentrate the magnetic flux as an intense field within a narrow air gap.

"In one form the lens is made with an annular opening of 26 millimeters. It can be slipped over a length of 25-millimeter glass tubing to focus an electron beam within the tube [see illustration, left]. We use lenses of this kind mostly as the condenser of the compound microscope and occasionally as the objective lens of the simple microscope. This lens design has a minimum focal length of about five millimeters. When it is used as the objective lens of the simple microscope at a distance of 50 centimeters from the image plane, the magnifying power of the instrument is 100 diameters. (The magnifying power of the simple microscope is equal to the distance between the lens and the image plane divided by the focal length of the lens.)

"Lenses with a focal length of about one millimeter are used for the objective and projection lenses of our compound microscopes The working air gap of the lenses lies between annular pole pieces of soft iron that are rigidly fastened together by a threaded brass coupling. The pole-piece assembly is magnetized through the glass by a surrounding iron yoke and coil [see illustration, right].

"All parts of the yoke and the pole pieces are preferably machined of No. 1020 steel, an easily magnetized material. Most kinds of soft iron or steel, including ordinaly iron pipe and mild steel rod of the type commonly used as shafting, can be substituted for No. 1020 steel, although the resulting lens may require more exciting current to achieve the same focal length as a lens made of this steel. The current can be controlled by 10-ohm, 25-watt rheostats, such as the Ohmite Type 014, if a lens of the condenser type is powered by a six-volt storage battery and each of the two lenses of the objective type by a 12-volt storage battery.

"After extended experimentation, during which club members made and tested a full range of magnetic-pole geometries for lenses of the objective type, optimum performance was observed with pole pieces that were bored to an inside diameter of five millimeters with a five-millimeter air gap. The same lenses worked best for projecting images to the fluorescent screen with a bore and air gap of three millimeters. Incidentally, the basic concept of employing an external coil and iron yoke to magnetize the pole-piece assembly greatly simplifies the modification of experimental lenses. Only the pole-piece assembly need be altered. The coil-and-yoke assembly fits over the outside of the tubing and so can be used to magnetize pole pieces of any design.

"The thin air gaps at the edge of the lens, where magnetic flux passes through the glass, have a negligible effect on the strength of the field and the electron waves. The pole pieces of the objective and condenser lenses and the grounded anode should be interconnected by a copper wire in the form of a helix that fits snugly against the inner wall of the tubing. The turns of the helix can be roughly spaced at from 10 to 20 millimeters. The interconnection drains to ground the electrostatic charge that would otherwise accumulate on the pole piece assemblies of the lenses, thus shifting the position of the image.

"As I have mentioned, we used cathodes of two types for the production of electrons, one similar in principle to the cold electrodes in gas-discharge tubes and the other in the form of an electron gun resembling the kind in a cathode ray tube. The quantity of electrons required for illuminating a specimen increases as the square of the magnification. Our instruments were operated at magnifications of from 10 to 10,000 diameters.

"By experimentation we found that a simple electrode of the cold-cathode type provides adequate illumination for magnification up to about 100 diameters. We made cold cathodes in the form of a dish-shaped disk of metal, which tends to liberate electrons in the form of a converging beam. For mechanical support the center of the convex side of the disk was soldered to a slender metal rod that also served as the high-voltage terminal of the microscope. The rod passed through the center hole of a Neoprene stopper that made an airtight fit with the glass tubing of the microscope. The stopper supported the cathode in axial alignment with both the tubular anode and the specimen.

"Tests indicated that the emission of electrons was not significantly related to the kind of metal used for the cathode. We made most of our cathodes of sheet zinc, which is readily bent into the dish shape and is also easy to solder. The dish shape was formed with a punch and die made by letting a hot steel ball, one inch in diameter, burn a shallow depression in a block of hardwood. To form the dish a disk of zinc the size of a dime was placed over the charred depression. The ball was then placed on the disk and forcefully struck with a hammer.

"The electrons are emitted at a right angle to the surface of the cathode when a sufficiently high voltage is placed between electrodes in any vessel that is exhausted to a pressure of from roughly one to 10 torr. A steep potential gradient exists in the vicinity of the cathode. Positive ions, formed initially by collisions between cosmic rays and molecules of the residual air, are accelerated toward the cathode by the potential gradient. The impact of these ions on the metal liberates electrons. The concave shape of the cathode concentrates the emission roughly in the form of a convergent beam that illuminates the specimen.

"To form a good image with electron waves the microscope must liberate enough electrons for adequate illumination, must accelerate the slow-moving electrons to high velocity and must project the speeding particles into a region that is everywhere at equal potential. (In such a region the paths of the electron waves are altered only by the specimen and the forces developed by the magnetic lenses.) These conditions are met by connecting the negative terminal of the high-voltage power supply to the cathode of the microscope and the positive terminal to the anode, also connecting the anode to the ground. A steep potential gradient then exists between the cathode and the anode; it vanishes abruptly at the anode. The gradient does not exist in the space beyond the grounded anode.

"If the potential difference between the cathode and the anode is on the order of 13,000 volts, as is the case with our instrument, electrons enter the tubular anode at a velocity of some 50,000 miles per second. No accelerating force acts on them within or beyond the anode. Hence they coast at constant velocity toward the image plane unless they are diverted by a specimen and the lenses.

"In the case of the simple microscope we put the specimen stage of the microscope at the exit end of the anode and placed the single lens just beyond the plane of the specimen. The stage consists of a metal disk with a centered aperture. The edge of the disk was soldered to the exit end of the anode tube. A pair of small spring clips can be fastened to the disk with machine screws to hold the specimen in place [see illustration at left].

"Our instruments are all demountable designed to be put together and taken apart easily not only for interchanging specimens and photographic films but also for convenience in testing experimental parts. After the anode has been slid into the glass envelope it is locked in place by thrusting the ground terminal through the glass side arm and inserting the conical tip of the rod into the small hole of the anode. We salvaged the conical springs from discarded ball-point pens.

"For higher magnifications we replaced the cold cathode with an electron gun, which is a more copious source of electrons. After trying several designs we made up six guns in which a filament of tungsten wire in the form of an inverted V was supported by a pair of 7/32-inch steel rods [see illustration, right]. (Welding rod can be substituted.) The free ends of the rods were inserted in the holes of a Neoprene stopper. The filament was covered by a metal cap containing a centered aperture through which the electron beam diverges toward the condenser lens of the microscope.

"The cap is supported by a pair of eight-millimeter glass tubes that also insulate it from the filament terminals. The glass tubes are supported in part by the filament leads. They can also be inserted partway into the Neoprene stopper for additional support by enlarging the holes of the stopper to a depth of a few millimeters from its smaller end.

"We made the cap assembly from a one-inch length of 5/8-inch iron pipe. The outer diameter of the pipe was machined to 21 millimeters, a reduction of about .1 inch, so that the cap would slide into the 25-millimeter glass tubing of the microscope. A pair of grooves were milled in the inner wall of the pipe to mate with the supporting glass tubes. The perforated copper disk was silver-soldered to one end of the pipe. (A cap machined from solid stainless steel would work better. The copper corrodes.) The ends of the glass tubes were ground at an angle so that the glass would clear the filament and yet engage the grooves in the cap.

"A copper lead connects the insulated cap to ground and to the filament of the gun through a two-megohm rheostat. The resulting bias potential causes the cap to act as an electrostatic lens that tends to concentrate emitted electrons into a beam along the axis of the tube. Our measurements indicated that the resulting stream of electrons constitutes an electric current amounting to between 100 and 500 millionths of an ampere, of which approximately 10 millionths of an ampere is expended to form the image.

"We found by experiment that the best images were formed by making the aperture in the copper disk of the gun cap from two to three millimeters in diameter and exactly centering the apex of the V-shaped filament two to three millimeters behind the aperture. The alignment of the gun, palticularly the centering of the filament tip behind the aperture of the gun cap, proved to be the most important operating variable in adjusting the instrument for maximum image brightness. Both the electron gun and the cold cathode are aligned for maximum image brightness by altering the angle at which the supporting Neoprene stopper is inserted into the glass tubing.

"The construction details of the film holder are evident in the accompanying illustrations. The dimensions are not crucial. To replace the film we remove the access plate at the image end of the instrument and pull the coupling tube o the film holder from its supporting shaft. The film and the fluorescent screen can be fastened to the metal plate of the film holder by a pair of rubber bands. Fluorescent screens in the form of phosphor-coated sheets of clear plastic are available from distributors of X-ray supplies.

"Substantially all our experimental efforts have been devoted to building the microscope and improving the effectiveness of the electron source, the lenses and the scheme of the instrument in general. For this reason we did not seriously concern ourselves with the preparation of specimens until we neared the end of the project During the development phase we usually observed a commercially available specimen in the form of an accurately made copper grid with exactly 400 lines per inch that is 2.3 millimeters wide, three millimeters long and .S millimeter thick. A reference mark in the exact center of the grid is of great assistance in the adjustment of either optical or electron microscopes. The grids have a solid rim and are flat and clean.

"The electron beam of our microscopes does not have sufficient energy to penetrate solid specimens of appreciable thickness because we use relatively low accelerating voltages. We have nonetheless made a few successful specimens of inorganic crystals mounted on collodion films about 200 angstroms thick. The film is supported by the copper grid. Particles such as metal oxides are deposited on the film by supporting the coated grid in the smoke of burning magnesium or another metal. A highly dilute suspension of finely shredded asbestos in water can also be used to deposit crystals of the mineral on the plastic by allowing a drop to evaporate on the coated copper grid.

"To coat the copper grid with collodion fill a clean glass jar to the brim with distilled water. Let a meniscus form completely around the edge of the jar. With a pipette place on the surface of the water a drop of a solution consisting of one part collodion in 400 parts of amyl acetate. After one minute drop a copper grid face down on the collodion film. Push the end of a clean glass microscope slide into the water near the copper grid. The collodion film will adhere to the surfaces of the glass. The grid will be held against one glass surface by the overlying film of collodion

"Let the slide dry. Then, with sharp tweezers, lift the grid from the glass. The side of the grid that faces you will be coated with the invisible film on which crystals can be immediately deposited. The copper grids, tungsten wire for filaments and all other supplies required for electron microscopy are available from Ladd Research Industries, Inc. (P.O. Box 901, Burlington, Vt. 05402).

"We have worked with a variety of high-voltage power supplies, ranging from an induction coil of the kind used in the Model T Ford to 30,000-volt supplies removed from discarded television sets. A convenient power supply for the beginner, who is urged to work at relatively low magnifications during initial experiments, can be made with a neon-sign transformer [see right]. All these power supplies except the induction coil develop lethal voltages. When power is applied to the microscope, all apparatus connected to the cathode, including the high-voltage portions of the power supply and the storage battery that heats the filament, must be totally and inaccessibly enclosed by an insulating housing.

"The anode, the lenses and all other apparatus between the anode and the access plate are grounded and therefore safe. Incidentally, stray electrons from the beam tend to charge the inner wall of the glass tubing, an effect that can cause the image to drift from its centered position. We cure this vexing difficulty by lining the tubing with grounded aluminum foil.

"The mechanical pump (also known as the roughing pump or forepump) of our vacuum system was improvised from a refrigerator compressor by the methods described in The Scientific American Book of Projects for the Amateur Scientist (Simon and Schuster, Inc., 1960). The oil-diffusion pump, which is essential for attaining a pressure of 10-4 torr, was bought from Morris & Lee (1685 Elmwood Avenue, Buffalo, N.Y. 14207).

"The procedure for operating the electron microscope is essentially like that for operating an optical microscope With the copper grid in place pump the air from the system. In the case of the simple microscope apply high voltage and observe the brightness of the fluorescent screen. Turn off the high voltage and readjust the position of the Neoprene stopper that supports the cathode in an attempt to increase the brightness of the screen. Turn on the high voltage and examine the result. Repeat until maximum brightness is observed.

"Next, in the case of the simple microscope, energize the objective lens and focus it by adjusting the current to produce a sharp image of the copper grid. In the case of the compound microscope first switch on the condenser lens and adjust the current for maximum image brightness. Then focus the objective for the clearest image. Finally, apply still more current to the objective lens, so that the image becomes fuzzy, and adjust the projection lens for optimum image sharpness. Maximum performance can be observed only by experimentally adjusting the objective and projection lenses with respect to each other for both optimum resolution and magnification. This operation is something of an art. On this and other details of the project I shall be glad to answer readers' questions."

Bibliography

ELECTRON MICROSCOPY. Ralph W. Wyckoff. John Wiley & Sons, Inc., 1949.

CREATIVE GLASS BLOWING. James E. Hammesfahr and Clair L. Stong, W. H. Freeman and Company, 1968.

PRACTICAL ELECTRON MICROSCOPHY FOR BIOLOGISTS G. A. Meek. John Wiley & Sons, Inc., 1970.

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.