NUMEROUS amateurs have undertaken the formidable but nonetheless fascinating task of making a gas laser. Earlier articles in this department have described how to build a helium-neon laser and an argon laser [see "The Amateur Scientist"; SCIENTIFIC AMERICAN, September, 1964, and February, 1969]. Now Jeffrey Levatter, a high school student in Encino, Calif., has made a carbon dioxide laser.

The carbon dioxide laser produces a beam not of light but of infrared radiation. It is somewhat easier for the novice to build than the helium-neon laser because it involves no glassblowing. Moreover, it is relatively inexpensive. The costly dielectric mirrors of the helium-neon laser are replaced in the carbon dioxide laser by copper-coated mirrors that can be made at home. Levatter explains the operation of his laser and provides the details of its construction as follows:

"Physically all gas lasers are much alike. A glass tube filled with gas at low pressure is positioned between a pair of facing mirrors. The gas is excited by an electric discharge. Some particles of gas acquire energy by colliding with speeding electrons that are liberated by the discharge. After a finite interval particles thus excited spontaneously emit part or all of the acquired energy in the form of radiation. In so doing a specific particle may either drop to an intermediate level of energy or return to the lowest energy state: the ground level. In effect the gas absorbs energy from the electric circuit and subsequently liberates the energy as radiation in the form of the small packets called photons.

"The energy of the emitted photon depends on the spacing of the energy levels through which the gas particles characteristically fall. A gas particle that is excited to a high energy level can be stimulated to drop to an intermediate energy level if it interacts with a photon of appropriate energy. In dropping to the intermediate level the excited particle emits a photon that is identical with the stimulating photon. The two photons fall into lockstep. The coherent bundle of radiant energy continues to grow by accretion as it encounters still other appropriately excited particles and reacts with them.

"In laser action a growing train of such coherent electromagnetic waves is reflected back and forth through the excited gas by mirrors at the ends of the gas column. Part of the coherent energy escapes through a small window in one of the mirrors. This loss restricts the maximum energy that can accumulate between the mirrors and constitutes the output beam of the laser.

"The efficiency of a gas laser is determined in part by the nature of the gas, In a helium-neon laser the active particles are atoms of neon. To emit infrared radiation neon atoms must be excited to an energy level far above the ground state. Subsequently the atoms emit infrared radiation by dropping a relatively short distance to an intermediate level. The atoms must then return to the ground state before they can again participate in infrared emission. In returning to the ground state from the intermediate level the atoms emit excess energy that makes no direct contribution to the desired infrared radiation.

"In contrast, a molecule of carbon dioxide can be excited to an energy level that lies only a short distance above its ground state. From this level the molecule can emit infrared radiation by dropping a comparatively substantial distance to an intermediate level that lies close to the ground state. For this reason the efficiency of the carbon dioxide laser is impressively greater than that of the helium-neon laser.

"The electric discharge that energizes the system excites carbon dioxide molecules to various energy levels, including the level from which they drop in the course of emitting the desired coherent radiation. Particles excited to the other levels make no direct contribution to the output, although much of that energy is conserved. A significant portion of it is transferred by random collision to previously unexcited molecules, which are thereby raised to the level where they can contribute to the output of the laser. The molecules from which the energy is transferred return to the ground level.

"Although such transfers conserve some of the input energy, efficiency can be improved by mixing other gases with carbon dioxide, notably nitrogen and helium. In effect these gases absorb just the right amount of energy from the electric discharge to raise a carbon dioxide molecule from the ground level to the level whence the molecule can drop by emitting the desired radiation. The transfer of energy occurs during collisions among the several particles.

"As a consequence of the relatively low level to which carbon dioxide must be excited to induce laser action, together with the fact that selective excitation can be achieved by the introduction of other gases, the carbon dioxide laser converts about 20 percent of the input power into coherent radiation. The output power is impressive. My laser develops an infrared beam of about eight watts, which is thousands of times more powerful than the visible output of a helium-neon laser.

"Since the infrared beam is invisible the variety of experiments that can be done with the apparatus is restricted. On the other hand, the high power of the beam invites experimentation of a kind that cannot be achieved with equivalent lasers that operate in the visible part of the spectrum. The beam quickly chars wood. By focusing the rays with an appropriate concave mirror the energy density can be increased: to several kilowatts per square centimeter, which is sufficient to burn holes through thin metal. The earth's atmosphere is exceptionally transparent to electromagnetic radiation in the portion of the spectrum extending from a wavelength of eight to 14 microns. Hence the output of the carbon dioxide laser is ideal for communications experiments and also for experiments involving echo ranging.

"It should be possible to make holograms in infrared. Photographic film that is sensitive to infrared radiation is commercially available. I do not know if the grain size of the emulsion is fine enough for adequate resolution, but I look forward to trying the experiment.

"The laser assembly includes a gastight plasma tube in the form of a glass pipe cooled by a water jacket [see illustration at left]. The ends of the plasma tube are closed by a pair of metal cells that support the mirrors. Each cell includes a flexible bellows and a set of three screws for adjusting the orientation of the mirrors. The metal cells also serve as electrodes for applying high voltage to the gas. The electrodes are enclosed in boxes of clear plastic to prevent accidental contact with the high potential. The assembly is supported by an insulating base of wood.

"The borosilicate glass pipe is 18 inches long, with an inside diameter of one inch. It has slightly flared ends, which are sealed to aluminum flanges with silicone-rubber gaskets [see illustration lower right]. The glass, known as Pyrex conical piping, is made by the Corning Glass Works. The central portion of the pipe is surrounded coaxially by a 12-inch water jacket of aluminum tubing two inches in diameter. The ends of the aluminum tube are closed by silicone-rubber caulking. Opposite ends of the water jacket are fitted with pipe nipples of aluminum tubing fastened in place and sealed with epoxy cement. These pipe nipples function as inlet and outlet ports for circulating cold water through the jacket assembly.

"The flange assemblies that clamp to the ends of the glass pipe can be machined out of any metal. Brass is convenient because it solders readily to the copper-coated steel bellows. If aluminum is used for the flanges, the bellows can be sealed in place with epoxy. The ends of the bellows are about an inch in diameter. Bellows of this kind are available from Pathway Bellows, Inc. (P.O. Box 1090, 1452 North Johnson Avenue, El Cajon, Calif. 92020).

"The adjustable flange of the cells supports on its outer face a removable mounting plate to which the mirrors can be sealed with either epoxy or silicone caulking. A circular groove is machined in the outer face of the adjustable flange to accept a rubber O ring that makes a gastight seal between the flange and the mounting plate. Mounting plates are a convenience during the alignment of the optical system because they enable the operator to remove the mirrors.

"The three adjustment screws of each cell are radially spaced at 120 degrees of angle. The threads of the screws engage threads in the adjustable flange. The conical tips of the screws bear against conical indentations in the fixed flange. The screws should have at least 32 threads per inch and should be long enough to place the bellows in tension.

"The laser is fitted with two mirrors, one concave and one flat. The diameter of both mirrors should be somewhat larger than the bore of the plasma tube. The focal length of the concave mirror must be more than twice the distance between the mirrors. Both mirrors can be made of glass coated with a reflective film of either copper or gold. Mirrors coated with gold are available commercially from Esco Products (Oak Ridge Road, Oak Ridge, N.J. 07438). My experience indicates that copper has higher reflectivity than gold at a wavelength of 10.6 microns.

"The flat mirror transmits the output beam of the laser. It can be made of polished germanium, a material that reflects approximately 60 percent of the incident radiation and transmits 35 percent at 10.6 microns. (The remaining 5 percent is absorbed.) Germanium is expensive. My output mirror consists of a flat disk of polished glass 1/4 inch thick perforated in the center by a hole 3/32 inch in diameter. Glass disks appropriate for this purpose are available from the Edmund Scientific Co. (Barrington, N.J. 08007). The disks are identified as catalogue No. 30,451. The price is 50 cents per disk.

"The flat mirror can be made by drilling a hole through one of the disks. To drill the hole coat the glass with a protective film of pitch or some other waxy material and make a cofferdam around the upper edge with plastic modeling clay. Chuck a short brass rod about 5/64 inch in diameter in a drill press. Fill the cofferdam with a slurry of 220-grade Alundum grit in water. Gently lower the spinning rod into contact with the glass. Raise and lower the rod at one second intervals until the abrasive grinds through the disk. Remove the wax with solvent and clean the glass thoroughly. A highly reflective film of copper can be applied to either of the polished surfaces by means of the sputtering technique [see "The Amateur Scientist," SCIENTIFIC AMERICAN, October, 1967].

"The perforation must be closed on the outer surface of the mirror by a window that is transparent to infrared radiation. Windows made of crystals of sodium chloride or of potassium chloride are effective. Such crystals are hydroscopic, however, and must be kept dry when the laser is not in use. I store my window in a plastic bag that contains anhydrous calcium sulfate as the desiccant.

"Crystals of barium fluoride are much less hydroscopic but absorb substantially more infrared energy. Crystals of appropriate size for making the window are available, both polished and unpolished, from the Harshaw Chemical Company (18051 East Fourth Street, Tustin, Calif. 92680). The price of a large, unpolished crystal of rock salt is about $5. On request the company will send with the crystals an article describing the grinding and polishing of salt windows. The crystal can be cut at any angle with respect to its axes and need be only large enough to cover and seal the perforation. I suggest that the window be cemented in place with General Electric silicone adhesive, primarily because the cement can be easily removed. This adhesive is usually available in hardware stores.

"The concave mirror can be ground and polished at home by the techniques described in Amateur Telescope Making: Book One, edited by Albert G. Ingalls, which is available from Scientific American. Two glass disks are required; one serves as a tool for grinding abrasive against the other. After the desired curvature has been achieved and the glass has been ground to a velvety texture by the use of successively finer grains of abrasive the surface of the tool is coated with pitch. The mirror is then polished with rouge applied by the pitch tool.

"After thorough cleaning the polished surface can be coated with copper by the sputtering technique. The copper coating must be thick enough to prevent infrared radiation from penetrating the metal, otherwise the glass may absorb heat and shatter. Time the interval required to deposit a coating that is opaque to visible light and continue to sputter for at least as long.

"A mixture of gases is pumped through the plasma tube continuously when the laser is in operation. Ports for admitting and exhausting the gas are made in the cells. The working pressure of the gas ranges from one to 20 torr, depending on the proportions of the mixture. (One torr is the pressure exerted by a column of mercury one millimeter in height.)

"I draw the gases from three sources. Carbon dioxide sublimes from dry ice in a flask. Nitrogen is obtained from filtered air. (Oxygen, water vapor and other gases dilute the nitrogen but do not appear to reduce the output of the laser. Indeed, the power of the beam tends to increase when the incoming air is bubbled through water. I rarely use water, however, because the vapor can damage the salt window.) Helium is drawn from a cylinder of compressed gas. All gases are admitted through needle valves to manifold that connects to the laser. Gas pressure is measured by a closed-end manometer filled with vacuum oil an calibrated in millimeters of mercury [see illustration at left].

"A refrigerator compressor can serve as the vacuum pump. To prevent oil vapor from back-streaming from the pump into the laser a filter should be inserted between the inlet port of the compressor and the gas outlet of the laser. An adequate filter can be made by packing a one-gallon glass jug with glass wool. Close the jug with a two-hole rubber stopper. Gas enters the filter by way of a tube that extends through one perforation of the stopper to the bottom o the jug. Filtered gas flows through short tube at the top of the jug that connects to the inlet of the pump. The jug is enclosed in a wood box to minimize the hazard of flying fragments if the jug accidentally implodes. The refrigerator compressor I originally used worked well but became excessively hot aft several hours of continuous operation At present I use a conventional vacuum pump.

"The dimensions of the closed-en manometer are not critical. The instrument is made of standard-wall eight millimeter glass tubing. The scale graduations are equal in millimeters to the quotient of the density of mercury (13.55) divided by the density of the vacuum oil. If the density of the oil is no known, it can be determined with sufficient accuracy by weighing a known volume. The oil should be degassed b keeping it in a vacuum for an hour or so before filling the manometer.

"The tube can be filled with oil most conveniently by exhausting it to a pressure of 10 torr and admitting enough oil to completely fill the closed arm when air is let into the open arm. Alternatively the end that is to be close can be softened, pulled to a constriction and cut off at the narrowest zone, like the tip of a medicine dropper. Enough oil can then be sucked into the tube fill the long arm to the tip of the constriction. The tip can be sealed with epoxy. The accuracy of the measurements depends on the quality of the vacuum created when the oil separate from the closed end of the instrument. Even a tiny bubble above the oil c introduce a significant error.

"The performance of the laser depends critically on the gas mixture, the gas pressure and the exciting electric current. An optimum mixture consists of eight parts of helium, two parts of nitrogen and one part of carbon dioxide. In the absence of helium the best performance was observed with a mixture of two parts of nitrogen and one part of carbon dioxide. The partial pressures are maintained at four torr of helium, one torr of nitrogen and .5 torr of carbon dioxide. These proportions and pressures assume that the diameter of the plasma tube is one inch. Experiments done with a plasma tube half an inch in diameter indicated substantially higher working pressures. With the narrower tube the optimum partial pressures ranged from 15 to 20 torr of helium, one to three torr of nitrogen and one to three torr of carbon dioxide.

"The electric-power supply consists of a variable transformer that feeds the input of a 12,000-volt, current-limited neon-sign transformer. The output of the high-voltage transformer is rectified by four silicon diodes connected in the bridge configuration [see illustration at right]. The laser will operate satisfactorily on alternating current, but operation at maximum efficiency requires the use of direct current. If a conventional high-voltage transformer is substituted for the self-limiting neon-sign transformer, the output circuit must be equipped with a ballast resistor to prevent a runaway of current.

"The mirrors of the assembled unit must be adjusted to be parallel to each other and perpendicular to the axis of the plasma tube. To make the adjustment I first remove both mirrors by unscrewing their supporting mounting plates. The adjustment calls for three cardboard disks with 1/8-inch holes in the center. Two of the disks are pressed lightly into the holes of the adjustable faceplates. A parallel beam of light is projected through the third disk, which is placed between the light source and the first aperture. The light beam is then directed through the holes of all three disks. The diameter of the beam should match the diameter of the holes.

"A helium-neon laser provides an ideal alignment beam. If such a laser is not available, an adequate beam can be formed by making a pinhole aperture in the slide carrier of a 35-millimeter projector and focusing rays from the pinhole into a parallel beam with a small telescope of the Galilean type. Place the light source on a rigid support at least 10 feet from the laser and adjust the position of the light source so that the beam just grazes the edges of the apertures. The light beam is then coaxial with respect to the plasma tube.

"Remove the two disks in the laser, leaving the third one in place. Install the concave mirror and adjust it to center the reflected beam on the remaining cardboard aperture. Repeat the procedure to similarly align the perforated mirror. When air is pumped from the tube, atmospheric pressure may distort some of the parts slightly and alter the alignment. This can be checked by leaving the collimating beam in place. When the system is in proper alignment, the collimating beam will be reflected by the output mirror and back through the third tube. If the reflected beam does not move when vacuum is applied, the output mirror is stable.

"The stability of the concave mirror can be checked by replacing the output mirror with a piece of flat glass. The system should now be in sufficiently good alignment for operation. After the laser is oscillating the adjustments can be trimmed by trial and error for maximum power output.

"The operating procedure is fairly simple. Turn on the cooling water. Start the air pump and check the system for leaks. To make this test admit helium into the system to a pressure of 15 torr. Apply high voltage and adjust the current to approximately 100 milliamperes. During the first minute of operation the color of the discharge should turn from purplish to a pink-orange glow. The color change indicates that helium has replaced air inside the tube. Even a trace of the purplish hue indicates a leak in the system.

"A suspected leak can be confirmed by turning off the helium supply. Let the system pump down to below one torr. If no leak is present, the color of the discharge will look whitish gray. If any other color appears, turn off the current, tighten all sealing screws and check the gas-input system for leaks.

"When the system has been made gastight, exhaust the laser to the limit of the air pump. Then admit .5 torr of carbon dioxide, one torr of nitrogen and four torr of helium. Apply high voltage and with the variable transformer adjust the current to approximately 40 milliamperes. The laser should now begin to oscillate.

"The beam not only is invisible but also may be weak. It can be detected by inserting a small sheet of waxed paper or Thermofax paper in front of the output window. Caution: Do not place your hand or any part of your body in the path of the beam, even during the initial period of adjustment. The laser may be developing full output power, emitting a beam of sufficient energy to shatter glass many feet away. Even the reflected beam is hazardous. For this reason it is advisable to make a small container for disposing of unwanted beam energy. A metal box with a small opening is suitable. Coat the inside of the box with flat black paint and position the opening so that it intercepts the beam. The unwanted energy will be absorbed harmlessly as the radiation bounces around inside the box.

"To maximize the power output try small adjustments of the mirror-alignment screws, gas pressure, gas proportions and current. The output should be substantial, ranging from one watt to 10 watts. At optimum power the mode pattern of the beam will bum itself into a piece of wood. A microscope slide inserted into the beam when the laser is at optimum power will shatter."

Bibliography

AMATEUR TELESCOPE MAKING: BOOK ONE. Edited by Albert G. Ingalls. Scientific American Incorporated, 1950.

HIGH-POWER CARBON DIOXIDE LASERS. C. K. N. Patel in Scientific American, Vol. 219, No. 2, pages 22-33; August, 1968.

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.