SOMEDAY TECHNICIANS MANNING LUNAR outposts may keep in touch with the people back home over beams of modulated light. The fluorescence emitted by ruby and other crystals can be beamed with the necessary sharpness, and in theory it has substantially greater capacity than radio for carrying information. Signals transmitted optically can be detected by photoelectric cells. To harness light efficiently as a medium of communication, however, experimenters must first devise a practical apparatus for imposing multiple signals on light waves. Hans Jaffe and Joseph Stephany of the Clevite Corporation in Cleveland, Ohio, suggest that the desired modulator may be based on one of those "useless" laboratory curiosities handed down from the l9th century: the Pockels effect, which involves changes in the refractive properties of certain crystals when they are subjected to an electric field.

"The optics of crystals," writes Jaffe, "was a preferred subject of physicists during the first half of the last century and provided some of the most exciting and elegant experiments of the period. This interest grew during the second half of the century to include interactions in crystals among heat, light, elastic stress and electric fields. The development of quantum theory early in the present century diverted so much attention to other matters that crystal phenomena faded somewhat into the background. But the field continued to expand and to make increasing contributions to technology. Now a new generation of solid-state physicists is beginning to examine some long neglected phenomena of crystals.

"These phenomena include changes in the way in which light is propagated through certain crystals when an electric field is applied across the crystal lattice. This effect was first investigated intensively by F. Pockels at the University of Heidelberg during the l890's. To find out whether the change in refraction is a primary effect or merely a photoelastic effect caused by piezoelectric strain, he investigated crystals of quartz, tourmaline, Rochelle salt and sodium chlorate and concluded that there is a primary effect. He observed a change in refraction even when all the deformation that normally results from piezoelectric strain was suppressed. But the effect on refraction barely reveals itself in most natural crystals.

"The period of World War II brought an active search for good piezoelectric materials for use as detectors of underwater sound, and one particularly attractive group was found: the phosphates of alkali metals, especially the ammonium and potassium dihydrogen phosphates. In the course of investigating these compounds I found that their crystals exhibited a pronounced Pockels effect. Of especial interest is the fact that the effect is observed in this class of crystals when both the light and the electric field parallel the optical axis of the crystal. This enables the experimenter to construct a light modulator from a thin crystal plate of large area without troublesome birefringence effects, provided that the light is kept nearly parallel to the optical axis.

"Last year I asked Joseph Stephany, then a summer assistant at the Clevite Corporation, to set up a demonstration modulator with a crystal of ammonium dihydrogen phosphate. In the course of this work he came on some interesting but not entirely expected resonance phenomena. His components were selected from materials that are readily and inexpensively available, so that amateurs can repeat the experiment."

"The Pockels effect," Stephany writes, "is usually demonstrated by sandwiching the crystal plate between two sheets of electrically conducting glass and inserting this assembly between a pair of crossed Polaroid filters. A beam of light that enters the crystal through one of the Polaroid sheets vibrates in a single direction, at right angles to the direction of vibration of light that can pass through the second filter, and the beam is therefore blocked. But the application of a direct-current potential of several thousand volts across the conducting glass alters the refraction of the crystal by a controllable amount, so that the incoming light is transmitted by the second Polaroid filter. An apparatus of this type, I knew, could be used to modulate a beam of light electrically. Full modulation by voice currents, however, would require an amplifier capable of supplying voice currents to the crystal at a potential of thousands of volts. Such amplifiers are not commonly available and they would be hazardous to use

"I decided to resolve the difficulty by developing the high voltage at high frequency-to apply a conventional radio frequency carrier to the crystal and modulate the carrier. High voltage at radio frequencies is relatively safe and is easy to generate with apparatus made of inexpensive parts. An adequate receiver is equally simple: a photoelectric cell connected to a radio set. A system that operates at radio frequency has the additional advantage of being immune to low-frequency noise such as 60-cycle flicker from fluorescent lamps, street lights and neon signs. The receiver would operate at twice the frequency of the voltage applied to the crystal and could therefore be tested close to the transmitter with a low probability of spurious electrical transmission between the two. In addition to constructing an apparatus based on this scheme I set up an optical bench so that the behavior of the crystal could be observed visually during modulation.

"All the electronic parts, with the exception of the photocell and special coils, were salvaged from an old television set and a small radio. Most were mounted on 6-by-14-inch breadboards and wired according to the accompanying schematic diagrams [opposite page]. The object is simply to excite the crystal cell with output voltage from a conventional modulated oscillator of low power. The circuit details can be altered to accommodate materials at hand. Radio amateurs will encounter no difficulty in constructing this part of the apparatus; other experimenters may wish to call on a local radio ham for help.

"The two primary windings of the oscillator coil are wound on a cardboard tube approximately 1 1/2 inches in diameter and 3 1/4 inches long, as shown in the accompanying drawing [above]. I wound the secondary coil on a ferrite core 1/2 inch in diameter and 7 1/2 inches long that was salvaged from a broadcast receiver. (An equivalent core is priced at 65 cents by the Lafayette Radio Corporation, No. MS-333.) All windings are No. 24 enameled copper magnet wire. The primary coils must be wound in the direction shown in the illustration, and the core should be wrapped with plastic electrical tape prior to winding. An ordinary filter choke or the primary of an audio-output transformer can be used as the choke for modulating the screen grid. Either a 6V6, 6L6 or 6DQ6 vacuum tube can be used for the oscillator. The voice signals are picked up by a carbon microphone. The output of the oscillator is shifted from direct current to high-frequency alternating current by moving the connector-from the plate cap of the 1B3 tube to the dummy cap. The Ne-2 neon signal lamp lights when the doorknob capacitor is disconnected and is tied into the circuit by a single lead. The apparatus operates from direct-current sources of 250 volts at 70 milliamperes and 6.3 volts at two amperes. The incandescent lamp of the optical bench requires a separate source of 6.3 volts at three amperes.

"The base of the optical bench is made of 3/4-inch lumber, three inches wide and three feet long, as shown in the accompanying illustration [Figure 4]. Supports for the lenses and filters are cut from a tin can and are lined with plastic tape at points in contact with the glass to prevent scratches. Both lenses are 2 1/2 inches in diameter and have a focal length of five inches. Mine were bought from the Edmund Scientific Company in Barrington, N.J. (catalogue No. 1166). The lamp of the optical bench is a six-to-eight-volt automobile parking bulb mounted so that the axis of the filament is aligned with the axis of the optical train. The photocell is a conventional No. 925 and can be ordered through dealers in radio supplies. Crystal blanks of ammonium dihydrogen phosphate are available from the Electronic Research Division of the Clevite Corporation (540 East 105 Street, Cleveland 8, Ohio) and from the Electro-Ceramics Company (2645 South Second West, Salt Lake City, Utah). These crystals are not identified by code number; if readers refer to this article when ordering, the supplier will select the proper item.

"The crystal must be cut, ground, polished and mounted between glass plates. My crystal plate was cut from a whole crystal and when finished measured 1 1/2 inches square and 1/4 inch thick, with the major faces of the plate perpendicular to the optical axis of the prism [see diagram at top right in illustration above]. The orientation of the side faces is not important, but they usually parallel the natural faces of the prism. Incidentally, those who would like to try their hand at an art of another sort may enjoy growing their own crystals. An excellent outline of the process is given in Crystals and Crystal Growing, by Alan Holden and Phylis Singer, Science Study Series S7, Anchor Books, Doubleday & Company, Inc.

"Plates of ammonium dihydrogen phosphate must be handled with care. They dissolve in water, crack when subjected to abrupt changes in temperature and break easily when dropped. In the course of the experiment observations will be made of certain patterns of vibration, so the plate must be ground as closely as possible to a perfect square. This is not difficult if the experimenter works with care. First, a piece of grade 00 sandpaper (or a finer grit if available) is fastened to a flat surface such as a breadboard. A teaspoonful of kerosene is poured in the center of the sandpaper for lubrication. The plate is then placed face down on the sandpaper and ground with a light pressure and a circular motion. Add kerosene as necessary to keep the work wet and move the plate to fresh portions of the sandpaper to prevent small chips from scratching the ground face. Lift the plate after two minutes of grinding, wipe it thoroughly with soft facial tissue and examine it. The plate, as bought, may have been coated with a film of aluminum. This must be ground off. Continue grinding until the surface is flat, has no deep scratches and appears white and translucent like ground glass. Then grind the opposite face. Use plenty of kerosene and shift to areas of fresh abrasive as the pores of the sandpaper fill up. Measure the thickness from time to time with calipers and vary the pressure exerted during grinding to keep the faces flat and parallel. Finally, grind the edges. Check this part of the work with an accurate try square. The major faces of the finished plate should be flat and the edges should be at right angles to the major faces.

"The ground plate is now cleaned and polished by holding it in a stream of tap water for about 10 seconds. The water will dissolve a thin layer from all surfaces, leaving a polished surface. The crystal becomes slippery when wet and the edges must be gripped with care. Change your grip every two seconds or so to prevent the formation of finger impressions around the edges. I usually shift the plate from hand to hand. Be sure not to touch the major faces during this operation. After the 10-second rinse shake off the excess water and wipe all surfaces gently with-facial tissue until they are dry. Shift your grip as the plate is wiped, as you did during the rinsing operation. Objects a foot away can be seen clearly through a properly finished plate. If the dry crystal is placed between crossed Polaroid filters and held up to the light, a dark cross with concentric colored rings should be visible.

"The polished plate is next mounted between two sheets of electrically conductive glass. The glass should be somewhat larger than the plate, say about two inches square. Manufacturers of this material include the Corning Glass Works and the Pittsburgh Plate Glass Company. Small amounts can be ordered through local dealers in laboratory supplies. Place a few drops of pure mineral oil on the major faces of the plate and press a glass on top of each face so that the oil spreads to the edges without bubbles. Then bind the stack together at the edges with rubber bands. Electrical connection to the glass is made by paper clips to which leads are soldered. Distant objects should appear clear and undistorted when viewed through the completed stack. A similar unit can be made without conductive glass by framing the major faces of the plate with aluminum or copper foil, but although this arrangement will respond to high-frequency voltage it cannot be used for experiments with direct current.

"To observe the Pockels effect align the lamp, a lens, a Polaroid filter, the crystal assembly, another Polaroid filter and a screen of white cardboard in sequence on the optical bench as shown in the top diagram at left in the accompanying illustration [Figure 5]. Light the lamp and adjust the position of the lens until a light spot of minimum diameter is focused on the Polaroid filter nearest the lamp. Then rotate the other filter until a minimum amount of light is transmitted. A cross and a pattern of concentric circles should appear on the screen. Finally, rotate both filters together until the arms of the cross are horizontal and vertical, as in the accompanying photograph [top of Figure 7]. Switch on the oscillator and adjust the tuning capacitor for maximum direct-current output. The application of this potential to the electrically conductive glass plates will change the uniaxial interference pattern on the screen to the biaxial pattern shown in the second photograph [bottom of Figure 7].

"The crystal assembly can now be used as an ultrahigh-speed shutter, one capable of opening and closing in a few millionths of a second. To demonstrate shutter action switch off the direct current, remove one Polaroid filter, insert a second lens between the crystal and screen and adjust it until the image of the crystal appears on the screen. Replace the Polaroid filter and move the lens that is between the lamp and the crystal toward the crystal until the cross and colored circles appear on the screen. Continue moving the lens toward the crystal and simultaneously adjust the position of the crystal assembly as may be necessary to keep the expanding image of the cross centered on the screen. A position will be found at which the full screen is uniformly dark. This represents the 'closed' position of the Pockels shutter [see middle diagram in Figure 5]. The screen will brighten when direct current is applied to the crystal assembly. One can adapt the shutter for high-speed photography by constructing a circuit that will apply a short pulse of high voltage to the crystal assembly. The high voltage can be obtained from a television receiver and might be pulsed by an arrangement similar to that shown by the accompanying schematic diagram. I have not tried this particular scheme, but it should work nicely. The experimenter should observe the usual safety precautions when working with high voltage. Light is not fully blocked by the crystal in the 'closed'-shutter condition-the transmission amounts to about one part in 500. It is therefore advisable to synchronize the action of the crystal with a camera shutter.

"Modulator action is demonstrated by switching the oscillator for alternating current output. The screen should be dark until current is applied. Switch on the oscillator and rotate the tuning capacitor slowly through its full range. At certain frequencies a grid pattern will appear on the screen, and it may shift or flip depending on the characteristics of the apparatus, as shown in the accompanying photograph [above]. The flipping effect is induced by the tendency of the oscillator to lock into the resonant frequency of the crystal-the natural period at which the crystal would vibrate mechanically if struck a sharp blow-and higher harmonics of that frequency. The patterns of vibration are analogous to those of a Chladni plate.

"Next, replace the screen with the photocell and switch on the radio receiver. Maintain the grid pattern. Then adjust the lens nearest the photocell until the image fills the window of the cell [see bottom diagram in illustration in Figure 5]. Do not adjust for a sharp pattern because the bright spots in the pattern may paralyze local areas of the photocell. Tune the receiver until a frying sound is heard, which indicates a signal from the oscillator. If no response is heard, adjust the oscillator to another grid pattern and try again. Select the pattern and oscillator setting that produce the strongest signal. Spurious signals can be transmitted directly from the oscillator to the nearby receiver, so test for a valid signal by interrupting the light beam. The frying sound should stop when an opaque object is placed in the beam. Connect the carbon microphone and speak. You are talking over a light beam on a subcarrier channel. By coupling other modulating circuits to the crystal assembly and connecting companion radio receivers to the photocell it would be possible to transmit additional voice signals simultaneously over the same beam, each actuating its own loudspeaker.

"How far can one talk over this system? The distance is limited primarily by the intensity of the light and the amount that the beam spreads. With the automobile lamp bulb as a source we communicated over a distance of 100 feet. The filament is not a point surface and so the beam cannot be focused sharply. We substituted a concentrated arc lamp for the incandescent lamp bulb and conversed easily over a distance of 300 feet, the limit of our available space. The beam from the transmitter must fall on the photocell, and in setting up the apparatus for long-distance transmission we found that it is not easy to draw a perfect bead on the target.

"This apparatus was assembled as an introductory demonstration of how a crystal can modulate light. It has been suggested that a system based on the modulation of reflected sunlight should be effective over distances up to about 10 million miles [see "Kerr Cell Modulator for Space Communications," by Ronald G. Taylor; Electrical Design News, September, 1961]. An optical radar has also been proposed. An optical-maser beam modulated by an ADP crystal should locate objects more accurately and at appreciably greater distances than conventional microwave radars are capable of doing."

Bibliography

LIGHT. R. W. Ditchburn. Interscience Publishers, Inc., 1952.

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.