Roger Hayward, the illustrator of this department, recently performed a series of experiments that involve the response of dielectric liquids to electric fields of high potential and the reaction to the fields of both conducting and nonconducting particles suspended in the liquids. Hayward writes: "Most of the classic electrostatic experiments that usually are performed in air can be done to advantage in a dielectric liquid such as carbon tetrachloride or kerosene. The relatively high viscosity of the liquid causes the effects to proceed in slow motion, so that you can observe what is going on.

"In addition, a number of effects that appear in liquids cannot be observed in air. For example, elongated dielectric particles suspended in kerosene align themselves in the direction of the electric force and thereby trace the pattern of the potential field just as iron filings disclose the shape of a magnetic field when they are scattered on a sheet of paper held over the magnet. This was the first experiment I tried, because I wanted to see just where the electric force was most intense before attempting to observe its influence on transparent liquids.

"To set up the experiment I poured about a pint of kerosene into a Pyrex pie plate eight inches in diameter. With a pair of scissors I cut the hairs of a camel's-hair brush into pieces about a sixteenth of an inch long. The short hairs were stirred into the kerosene. The dish was placed on a slab of Styrofoam about an inch thick.

"The pattern of a potential field is determined in part by the shape of the electrodes. My first pair of electrodes consisted of (1) a thin wire immersed vertically in the liquid at the center of the dish and (2) a wire circle that ran around the inner edge of the dish. The electrodes were connected to the terminals of an inexpensive electrostatic generator of the Wimshurst type that I bought from the Edmund Scientific Co.

"When the machine was cranked, the hairs turned in the direction of the electric force to form a symmetrical pattern of radial lines that joined the center to the circumference. In a potential field of this kind the strength of the electric force varies with the density of the lines. In the case of the radial field the line density and the force are greatest at the center. Indeed, when I started the generator the force at the center was so large that the liquid climbed up the wire more than a quarter of an inch, and the violent streaming destroyed the pattern in that region. I restored the pattern by cranking with less vigor, thus reducing the applied voltage. Reversing the polarity made no change in the shape of the field.

"Next I removed the ring and inserted in the field a pair of vertical wires spaced several inches apart. They were connected to one terminal of the Wimshurst machine. The other terminal of the machine was grounded. Radial patterns now extended outward from each electrode, the lines curving sharply away from each other along the narrow middle zone. Connecting the wires to opposite terminals of the machine produced a pattern of arching curves that looked just like the pattern produced by dusting iron filings on a sheet of paper above the poles of a horseshoe magnet. The lines were densest between the wire electrodes.

"A pair of straight electrodes was made by cutting two flat strips of metal about five inches long and half an inch wide from an aluminum pie plate, A short length of 18-gauge copper wire was crimped over the center of each strip to serve both as a connection and as a brace to support the strips on edge in the liquid. The strips were placed parallel to each other and about two inches apart. The resulting field turned out to be uniform, as one would expect. The lines were perfectly parallel except at the ends of the plates, where they arched outward. With only a single strip in the liquid and the second terminal of the generator, grounded, a uniform field extended symmetrically from both sides of the electrode.

"Single electrodes of various shapes- circular, elliptical, egg-shaped and so on -were then cut from aluminum foil and placed flat in the middle of the dish. These were connected to one terminal of the generator by placing a wire vertically against the center of the electrode. The wire made the necessary connection and also held the aluminum foil in place.

"The second electrode again consisted of a circle of wire that ran around the edge of the dish. In every case the field lines made right angles with the edge of the electrodes, just as the textbooks predict. In the case of an egg-shaped electrode the lines bunch together at the small end of the egg, thus demonstrating that the strength of the field varies inversely with the radius of the electrode.

"That the field strength does in fact reach the maximum in the vicinity of a sharply pointed electrode was demonstrated by another experiment. A pound of carbon tetrachloride was placed in a 500-milliliter Florence flask along with a pointed U-shaped wire positioned so that the point protruded about 1/32 inch above the surface of the liquid. The wire was held in place by a groove cut on the side of a cork that closed the flask. (The stopper should be of cork rather than rubber because carbon tetrachloride attacks rubber. The bottle must be closed because the fumes of carbon tetrachloride are toxic. The chemical should be handled only in a well-ventilated room.)

"Current leaks from the upper end of the wire in the form of a corona discharge. This loss can be reduced by covering the cork with a rounded metal shield such as the cap from a spray can. The other electrode is also a pointed wire that can be manipulated with a handle such as a clothespin, which should be of plastic because of the high voltage involved [see Figure 1].

"When the two electrodes are connected to the terminals of the generator and the movable electrode is brought close to the side of the flask, the liquid surrounding the partly submerged electrode will become charged with the same polarity as the wire. The force of mutual repulsion is large enough to shoot a stream of liquid into the air perhaps an inch or so above the surface, depending on the voltage applied and the position of the outer electrode. The meniscus that surrounds the wire at the point where it enters the liquid will rise a quarter of an inch.

"Next I substituted for the U-shaped electrode a straight wire that terminated in a small metal sphere. Brass balls are simple to make if you have access to a lathe. Place a brass rod of the desired size in the lathe chuck and make two opposing cuts at 45 degrees [see Figure2]. Trim the corners of the angled cuts at 67.5 and 22.5 degrees. The resulting shape is roughly spherical. It is reduced to a true sphere by means of a special tool in the form of a heavy-walled tube of steel, one end of which is sharpened to serve as a cutting edge. The finished diameter of the bore of the tool should be about 80 percent of the intended diameter of the sphere. With the lathe turning at its maximum speed, place the tool in light contact with the sphere and swing it back and forth over the brass while simultaneously rotating it on its own axis. With a little experience the truing operation can be completed in seconds and the finished piece will emerge with a bright, if not polished, surface. The sphere can then be drilled for soldering to a copper wire.

"I also made a little whirligig with pointed ends of sheet aluminum and poked a hole through its middle with an awl. The finished piece was threaded on the wire that terminated in the brass sphere and was installed in the flask [see Figure 3]. The second terminal of the generator was connected to a movable point close to the outer surface of the flask. When power was applied, the whirligig spun with much vigor and rose to the surface of the liquid. I turned the tips of the points upward in the hope that the downward thrust of the blades would keep the thing submerged, but it continued rising to the surface (where it made a great sloshing). The position of the external electrode made little difference in the behavior of the whirligig except when it was placed near the surface of the liquid. The rate of spin then increased.

"Thereafter I investigated the effect of a pointed electrode that was supported in air close to the surface of the liquid. Pointed electrodes in air create an 'electric wind.' I wanted to check the effect of the wind on the oppositely charged surface of the liquid. A rectangular brass electrode of opposite polarity was placed at the bottom of a refrigerator dish, which was covered by a sheet of glass to prevent the escape of toxic fumes [see Figure 4].

"When a sharp needle was held as shown, the liquid was blown off the brass electrode with a big whoosh. A Van de Graaff generator capable of developing 220,000 volts was then substituted for the Wimshurst machine. (The generator also came from the Edmund Scientific Co.) When the generator was turned on, most of the liquid was promptly blown out of the dish!

"After airing the room, I resumed experimenting with the Florence flask. This time I seeded the liquid with a small pinch of aluminum powder of the kind used as a pigment in aluminum paint. My purpose was to observe the interaction of many charged but individually insulated particles. Only a little powder is needed-about the amount that could be piled on the flat end of an unsharpened lead pencil. Most of the powder sank to the bottom of the flask, but many particles remained in suspension for more than 24 hours.

"The flask was fitted with a pair of metallic electrodes; one was a thick, rounded disk and the other was a half-inch sphere spaced about an inch away from the center of the disk [see Figure 5]. The flask was lighted on one side by a shielded lamp so that the particles could be observed easily. When the field was applied, some large areas of particles flashed brightly with reflected light and other areas became darker. Evidently the particles aligned themselves with their long dimensions in the direction of the field, some groups reflecting more light in the direction of the eye and others less light. Particles suspended between the electrodes darted continuously back and forth from the disk and the sphere, simulating the movement of electrons and ions in a gas-discharge tube. When one terminal of the machine is attached to the electrodes and the other to a sheet of aluminum foil placed under the flask, all the liquid becomes charged and most of the powder that has settled to the bottom boils up into suspension.

"A sewing needle was now attached to a flexible lead and fitted with an insulating handle so that it could be manipulated. The needle was substituted as an electrode for the aluminum foil under the glass. When the point was brought near the outer surface of the flask, the effect suggested that an electronic wind was blowing through the glass. The particles rushed away from the point in the form of a jet.

"This effect gave me the idea of simulating an atom smasher such as the electrostatic accelerator with which the British physicists J. D. Cockcroft and E. T. S. Walton transmuted lithium into unstable beryllium. The first model consisted of a rectangular refrigerator dish about eight inches long and an inch deep, fitted at one end with a sheet of aluminum that served as an electrode [see bottom illustration at right]. A needle point at the other end simulated the source of particles to be accelerated. The aluminum represented the target. Four electrodes were placed along each side of the dish to simulate the equipotential rings that distribute the electric field uniformly along the Cockcroft-Walton accelerator. When power is applied, aluminum particles stream to the target in a beamlike array. For best results the refrigerator dish must rest on a block of Styrofoam to prevent charge from leaking away. If the liquid is carbon tetrachloride, a sheet of glass should be placed between the dish and the plastic to protect the Styrofoam from the solvent. The dish must also be covered with a glass sheet to prevent the escape of the toxic fumes.

"I made a second version of the accelerator for use with the Van de Graaff generator. Because of the higher power the number of equipotential 'rings' was reduced to three on each side, and they were placed flat against the sides of the dish [see illustration above]. These electrodes require no wiring. They become charged through the air. The target electrode was grounded, in effect, by a lead that terminated in a needle point facing away from the generator. The accelerating electrode was connected to a lead that terminated in a brass sphere.

"Corona discharge between the sphere and the high-voltage terminal of the Van de Graaff machine energizes the apparatus. Once I accidentally grounded the target and moved the generator within about an inch of the small sphere. When I applied power, the intense field blew most of the liquid out of the dish!"

James R. Bailey, a high school student in Milwaukee, submits an interesting apparatus that also uses aluminum powder in suspension. It is for displaying convection currents: the streaming induced by gravity in gases and liquids when they are heated or cooled nonuniformly. Bailey writes:

"The simple apparatus consists of a shallow glass jar that contains aluminum powder suspended in a volatile solvent such as trichloroethylene, Freon T. F. or wood alcohol. The jar is covered by a glass disk, such as the crystal of an alarm clock, that is sealed in place with epoxy cement. A squat peanut butter jar can be used or, better, a large Petri dish. The aluminum powder can be obtained from dealers in art or sign supplies. Use a very small pinch of the powder, just enough to make fairly dense swirls when it is stirred into the liquid. Particles that persist in floating on the surface after the stirring can be skimmed off and discarded.

"Place the sealed jar in the refrigerator for a few minutes. When it has cooled, remove it and touch the side at any point with the tip of your finger. Al most instantly vivid silvery waves will rush away from the warmed point toward the center of the jar. The action will persist for several seconds until the glass cools. The reason for the effect is that heat lowers the density of a relatively thick zone of solution at the edge of the dish. Gravity then causes neigh boring liquid of higher density to displace this zone, which spreads over the surface as a thin, high-velocity layer that sinks after cooling.

"Many other wave patterns can be generated by heating the jar in different ways. For example, support the jar on a pair of books or small boxes spaced about two inches apart and heat the center of the bottom of the jar with your finger. The resulting waves resemble the action of a boiling spring. Another interesting variation can be observed by letting the jar come to room temperature and then cooling a spot with an ice cube. By this means one reverses the action.

"Avoid exposing the device to excessive temperature. The resulting high pressure could break the seal. The sol vents used are all toxic to some extent and most are flammable. The aluminum powder is not dangerous but makes a terrible mess if it is spilled. It can be dissolved with a strong solution of potassium hydroxide, sodium hydroxide or common lye. The finished apparatus is sturdy and inexpensive, and it never wears out. Like a kaleidoscope or an open fire, it generates abstract patterns of endless variety."

Bibliography

NONUNIFORM ELECTRIC FIELDS. Herbert A. Pohl in Scientific American, Vol. 203, No. 6, pages 106-116; December, 1960.

THE SCIENTIFIC AMERICAN BOOK OF PROJECTS FOR THE AMATEUR SCIENTIST. C. L. Stong. Simon and Schuster, 1960.

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.