OF THE numerous kinds of lasers that have been developed during the past decade, the smallest and least costly to assemble is the diode laser. The active element, a sandwich of semiconducting material the size of a pinhead, emits an intense beam of infrared radiation when it is energized by a direct current. The invisible beam holds promise as a carrier of various kinds of signals. It can also be employed for echo ranging and intrusion alarms and for demonstrating such aspects of wave behavior as refraction, diffraction, reflection and interference.

Diodes of gallium arsenide that are designed for generating pulses of coherent radiation have become available in recent months at prices that enable the experimenter to assemble a working laser for less than $30. Several diode lasers, which emit beams that range in peak power from four to 70 watts, have been built by Harry L. Latterman of Mesa, Ariz. He explains the apparatus and some experiments that can be done with it as follows:

"The diode lasers I have made consist of three subassemblies: the diode and its mounting; an electronic circuit that generates pulses of direct current, and a source of power. The smallest of m lasers, which can be made in one evening, emits 200 pulses per second. Each pulse persists for 50 nanoseconds (billionths of a second) and reaches a maximum intensity of four watts. When the laser is operated by a dry-cell battery, it is self-contained and portable. The apparatus can be mounted in a protective housing smaller than a pack of cigarettes.

"The active part of a typical diode is a crystal consisting of three or more distinct layers of semiconducting materials sandwiched between electrically conducting films of metal. The layered structure accounts for the electrical characteristics of the diode. All crystals consist of atoms bound together in a lattice by forces associated with electrons in the outermost orbits of the constituent atoms-the valence electrons that bind atoms together. In ordinary crystals the valence electrons are fully occupied. In effect they act as tie rods that constitute the structure of the lattice.

"A semiconducting material known as the n type can be made by incorporating in the lattice a few atoms that have one valence electron more than the number that can fit into the lattice structure. The surplus electron can be detached from its parent atom rather easily. It can then migrate through the crystal as a carrier of current. Conversely, a semiconductor material known as the p type can be made by incorporating a few atoms that have one valence electron less than the number that can be accommodated by the lattice structure. In this case a bond is missing from the crystal structure.

"An adjacent valence electron can become detached from its bonding site and drop into the missing bond. A hole then exists at the vacated site. In terms of its electrical behavior the hole has the properties of an elementary charge. Because it is deficient in negative charge, however, it acts as a positive charge. Crystals so made are known as p-type.

"The crystal of a laser diode can be made with three layers of semiconducting material: a layer of n-type silicon, a layer of p-type gallium arsenide and a third layer p-type gallium arsenide that also contains atoms of aluminum. The interface between n-type and p-type layers is known as a p-n junction. Some easily detached electrons in the n region migrate across the interface and drop into holes of the p-type material. Conversely, some holes m the p-type material cross the junction and wander randomly in the n-type material. The p-type material thus accumulates negative charge, and the n-type material similarly accumulates positive charge.

"The charges ultimately become strong enough to halt the migration of both holes and electrons. A potential barrier is then said to exist across the junction. The barrier can be modified by connecting opposite faces of the crystal to an external source of voltage. A negative potential applied to the n-type layer causes electrons to flow across the junction, migrate through the p-type material and return to the external source. The diode is said to be connected in the forward direction. Positive charge, as represented by the holes, migrates in the opposite direction. The action stops if the polarity of the external source is reversed.

"A positive charge applied to the n-type material simply increases the potential barrier. Hence the diode conducts current in only one direction. It is analogous to a hydraulic check valve and can be used to convert alternating current into unidirectional current, which is direct current.

"When the diode is in its quiescent state, before voltage from the external source is applied, the electrons have minimum energy. At room temperature the crystal is in a state of continuous but gentle vibration that causes electrons to wander aimlessly through the lattice. The interesting action begins when a forward potential of about 1.2 volts from an external source is applied across the p-n junction. A current of electrons then flows through the diode. Collisions occur between the moving electrons and the electrons that are normally bound in the lattice structure.

"Some collisions are so violent that electrons emerge from the encounter with a discrete increment of additional energy. Energy so acquired is retained for a time. The excited electron is unstable, however, and it soon returns to a lower energy state spontaneously by emitting a photon, or quantum of radiation, that carries away the acquired energy. Many electrons, acting independently, participate in the activity as long as the diode conducts current.

"The photons are emitted randomly in direction and in time. Hence the radiation is incoherent, analogous to the wave pattern created by tossing a handful of pebbles into a pool of still water. It is similar to the light emitted by a neon sign, and for the same reason. If a photon is pictured in the mind's eye as a short train of identical waves, coherent radiation can be pictured as a wave train consisting of two photons that unite in lockstep and proceed in the same direction.

"Coherent radiation can be generated in various ways. For example, it appears when a photon encounters and interacts with an electron that is on the verge of emitting surplus energy in an amount equal to the energy carried by the stimulating photon. The emitted photon joins the stimulating photon and the two proceed through space together.

"In order to generate coherent radiation, lasers of all kinds provide two essential conditions. First, the laser must maintain an adequate supply of excited electrons. Second, the excited electrons must be trapped inside a resonant optical cavity that consists of a pair of facing mirrors.

"In the case of the diode laser an adequate population of excited electrons is generated by connecting a source of power to the diode in the forward direction to provide current in excess of a certain minimum threshold value. At currents below this value the diode emits incoherent radiation; such a diode is called a light-emitting diode, as distinct from a diode laser. The mirrors that form the optical cavity of the diode laser are merely the square-cut edges of the crystal. The edges function as mirrors because the index of refraction changes abruptly at the interface between the crystal and the air.

"Electrons that are excited in the region of the p-n junction migrate into the transparent p-type material. Laser action begins when a photon is spontaneously emitted. The liberated photon bounces back and forth between the mirrors, interacting with the population of excited electrons and thereby stimulating the emission of additional photons

"The interface between the p-type gallium arsenide and the p-type gallium aluminum arsenide, which is known as a heterojunction, serves to confine the excited electrons and also to reduce the reabsorption of energy. In effect, the heterojunction improves the efficiency of the device. A portion of the accumulating radiant energy escapes through the mirrors in the form of a coherent beam, which is the output of the laser.

"The minimum current required to generate coherent radiation by the laser diodes now on the market ranges from about 10 to 80 amperes. The exact threshold is specified by the manufacturer, as is the peak current rating. The diode can be destroyed by current in excess of the peak value.

"The crystal in the smallest of my diodes is almost invisibly minute. A dozen such crystals could fit easily into the volume of a pinhead. Nonetheless, the diode is rated at a peak current of 10 amperes, which is equivalent to a current density in the diode on the order of 100,000 amperes per square centimeter. The problem of driving the laser at the required current without vaporizing the crystal is solved by using short pulses of current and by mounting the diode on a metal base that dissipates the liberated heat.

"The three pulsing circuits I shall describe are designed for generating peak currents of from five to 75 amperes that persist for intervals ranging from 25 to 250 nanoseconds. One or another of the three circuits will work with currently available diodes. The circuits draw current from the source less than 1 percent of the time. A duty cycle much greater than 1 percent can damage the diode. The average drain on the power supply amounts to only a few milliamperes. The diodes are shipped in a protective housing that resembles a machine screw about half an inch long with a flat, cylindrical head a quarter of an inch in diameter.

"The infrared beam is emitted through a circular window of clear plastic in the exposed end of the housing. The diode can be fastened to a heat sink with the screw, which also serves as one of the electrical terminals. The other terminal, a short length of rectangular wire, is brought out of the housing. A two-inch square of sheet aluminum at least a sixteenth of an inch thick makes an adequate heat sink.

"A circuit that develops a peak current of up to 10 amperes for operating the smaller laser diodes consists of a resistor, a capacitor, a four-layer diode, a protection diode and the laser diode. Essentially it is an oscillator of the relaxation type. As the capacitor gradually accumulates charge through the resistor, voltage rises across the four-layer diode. At a critical potential the resistance of the four-layer diode falls abruptly. The capacitor discharges through the laser, after which the cycle repeats. The resulting pulse of current persists for about 50 nanoseconds, which is equivalent to a frequency of 20 megacycles. Hence the leads between the diodes and the capacitor should be made as short as possible.

"Magnetic fields develop around the leads when the laser conducts current. At the conclusion of the pulse the fields collapse, inducing current in the wiring. As a result a reverse potential can appear across the diode. A reverse potential that exceeds a certain value can destroy the laser diode. The circuit includes a conventional diode that acts as a protective device, limiting reverse potentials to a safe value. Do not omitit. Peak current through the laser diode can be adjusted through the range from about five to 10 amperes by applying from 22-1/2 to 45 volts to the circuit.

"A practical circuit is depicted by the accompanying diagram. A capacitor is charged through a 200,000-ohm resistor by the 135-volt source. As the capacitor accumulates charge, potential rises across the 75,000ohm resistor in the base circuit of the transistor. When the capacitor reaches full charge, the potential of the base reaches a value that initiates conduction in the collector-emitter circuit of the transistor. The capacitor then discharges through the transistor and the laser diode. A conventional diode, connected across the laser diode, limits the reverse potential and thus protects the laser diode. The circuit generates approximately 500 pulses per second. Each pulse persists for about 50 nanoseconds.

"Current pulses of up to 75 amperes are generated by charging a capacitor of .03 microfarad to a potential of 200 volts [at right]. The charged capacitor is connected to the laser diode by a special switch in the form of a controlled silicon rectifier, a solid-state device that is capable of conducting relatively large currents. The switch is turned on and off by an oscillator, whose active element is a transistor of the unijunction type.

"The oscillator includes a 100,000 ohm resistor through which a .05-microfarad capacitor accumulates charge. Potential across the capacitor rises as charge accumulates and is applied to the base of the unijunction transistor. As full charge is approached the transistor conducts. The resulting discharge develops a voltage across the 27-ohm resistor in the transistor circuit. The voltage is applied through a diode to the gate terminal of the controlled silicon rectifier.

"Meanwhile the .03-microfarad capacitor in the laser-diode circuit has accumulated full charge from the 200-volt source. When the triggering voltage is applied to the gate of the controlled silicon rectifier, the rectifier conducts and the .03-microfarad capacitor discharges a 75-ampere pulse through the laser diode. Conventional diodes protect the laser diode from excessive reverse voltage. The pulse persists for about 200 nanoseconds.

"When I assemble any of the three circuits, I usually measure the pulse before connecting the costly laser diode in the circuit. The diodes are electrically equivalent to a resistor of about .1 ohm. I make up this resistor by connecting 10 one-ohm resistors in parallel. The voltage that appears across the resistor during a pulse is measured by an oscilloscope that can respond to a frequency of at least 200 megacycles. The current is calculated by Ohm's law: Current equals voltage divided by resistance.

"Several techniques are available for detecting the invisible beam of the lasers. The wavelength is quasimonochromatic and reaches peak intensity at about 9,000 angstroms. It turns out that the sensitivity of silicon phototransistors such as the type designated HEP 312 also peaks at this wavelength. I pick up the beam with this transistor and boost the resulting output signal by either of two simple amplifiers.

"The output of the amplifiers can be used for driving a power amplifier or a set of earphones or for triggering apparatus of other kinds. I use the two-stage amplifier in most experiments. The more sensitive three-stage amplifier is useful for detecting faint signals, particularly for picking up the beam at a substantial distance from the laser diode.

"The lasers emit the beam at a diverging angle of approximately 20 degrees. The angle can be reduced to about six minutes of arc by placing the diode at the focus of a simple lens with a focal ratio of f/2.8 or less. The aperture of the lens need not be large. An f/2.8 lens of half-inch aperture placed 1.4 inches from the diode is equivalent to and just as effective as a one-inch lens placed at a distance of 2.8 inches. Small, simple lenses of good quality are available from suppliers such as the Edmund 10 Scientific Co. (Barrington, N.J. 08007) for about $1.

"The emission can also be photographed with conventional infrared film that is available from dealers who stock a full line of photographic supplies. To photograph the wave nature of the radiation, place the edge of a safety razor blade at a right angle to the beam and about one centimeter from the diode so that the edge intercepts about half of the beam. In a darkened room mount a sheet of infrared film in line with the blade and the diode at a distance of about 25 centimeters. The emulsion side of the film should face the laser. A laser diode of the M4L 3052 type will expose the film adequately in about two seconds. The accompanying photograph [lower left] depicts the resulting diffraction pattern. A classical diffraction pattern can be similarly photographed by projecting the beam through a narrow slit, which can be improvised from a pair of razor blades.

"The cost of laser diodes varies with their size. Those of the smallest size have been advertised recently by dealers in surplus materials at a price of $5. The following companies can give additional information and prices concerning new diodes: Laser Diode Laboratories (205 Forest Street, Metuchen, N.J. 08840), Marketing Manager, ElectroOptical Devices, RCA Industrial Tube Division (New Holland Pike, Lancaster, Pa. 17604) and Texas Instruments Incorporated (P.O. Box 5012, Dallas, Tex. 75222).

"Finally, a word of warning: The emission of all lasers, including that of diode lasers, is hazardous. Infrared emission is particularly insidious because it is invisible and can be reflected by almost any smooth surface, including objects of tarnished metal and other materials that bear little resemblance to conventional mirrors. The beam can be transmitted safely for substantial distances only if it is directed through a metal pipe or a like enclosure. At shorter distances, up to a meter, the entire apparatus should be enclosed in an opaque box that will absorb the unwanted energy. Never operate the unshielded laser in the presence of other people."

IN THE latter part of the 19th century the Danish physicist Christian Christiansen devised a remarkable filter for isolating a narrow band of frequencies in the infrared portion of the spectrum as well as single colors of visible light. The device is unusual because it employs only clear transparent materials and is one of the few kinds of filter that work in the infrared.

The filter consists essentially of a flat cell of glass that contains a slurry of powdered glass or a comparable substance in a fluid that has a matching index of-refraction at some point in the optical spectrum. Powdered glass m air scatters light in all directions. It appears white and opaque just as snow does. The same scattering effect is observed when the particles are surrounded by a liquid of a different refractive index, such as water.

It is possible to immerse the particles in a fluid of matching refractive index. The slurry then appears clear. This stratagem is used for identifying certain kinds of glass. For example, Pyrex brand 7740 glass has a refractive index of 1.474. So does a solution that consists of 16 parts by volume of methyl alcohol in 84 parts of benzene. A piece of Pyrex vanishes when it is dipped in this solution, but glasses of a different refractive index look the way they do when they are iimmersed in water.

In the Christiansen filter a solution is compounded that matches the refractive index of the glass at a desired point in the spectrum but differs increasingly at all other points. The resulting filter is transparent to light of the selected wavelength and relatively opaque to all other wavelengths. A demonstration cell can be made with two sheets of ordinary glass, three glass strips about an eighth of an inch thick and a few dabs of epoxy cement. The strips serve as spacers at the sides and bottom of the cell.

Seal the edges with epoxy. To demonstrate the filtering action with white light, fill the cell with powdered borosilicate glass. Make the powder by breaking the glass into small pieces, which can be done safely by putting a large piece of glass in a bag of heavy paper and striking it with a hammer. Put the fragments in a mortar, a few at a time, and grind them to fine powder with a pestle. Avoid inhaling the powder.

For about five minutes stir the powder into a cleaning solution of dilute hydrochloric acid, made by adding one volume of acid to three volumes of distilled water. Let the powder settle, pour off the acid and wash the particles in three changes of distilled water. Spread the powder on filter paper to dry. Transfer enough dry powder to the cell to fill it approximately three-quarters full. Add carbon disulfide to immerse about two thirds of the glass particles.

Stir the mixture gently to release trapped bubbles of air. Place a 100-watt incandescent lamp about 10 feet beyond the cell and examine it through the glass slurry. With a pipette add benzene to the cell a few drops at a time. Stir the solution gently but thoroughly. Examine the lighted lamp through the cell after each addition of benzene. Eventually the lamp will appear dark red. Continue to add benzene. The colors of the transmitted rays will progress through the spectrum as benzene is added.

To convert the filter for infrared, clean the cell and refill it with a slurry of magnesium oxide in carbon tetrachloride. The transmission maximum of this combination is at a wavelength of 9,000 angstroms. The fumes of all three of the required liquids are highly toxic and must not be inhaled. The filter should be prepared in a fume hood and used only in a well-ventilated room.

Bibliography

PROCEDURES IN EXPERTMENTAL PHYSICS. John Strong in collaboration with H. Victor Neher, Albert E. Whitford, C. Hawley Cartwright and Roger Hayward. Prentice-Hall Inc., 1938.

GALLUM ARSENIDE LASERS. Edited by C. H. Gooch. John Wiley & Sons, Inc., 1969.

A NEW CLASS OF DIODE LASERS. Morton B. Panish and Izuo Hayashi in Scientific American, Vol. 225, No. 1, pages 32-40; July, 1971.

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.