A GLOWING BUBBLE OF AIR cannot be bought anywhere at any price. But with an oscilloscope, a moderately precise sound generator, a home stereo amplifier and about $100, readers can turn sound into light through a process called sonoluminescence [see "Sonoluminescence: Sound into Light," by Seth J. Putterman; page 70]. The apparatus is relatively simple. A glass spherical flask filled with water serves as the resonator-the cavity in which sound is created to trap and drive the bubble. Small speakers, called piezoelectric transducers, are cemented to the flask and powered by an audio generator and amplifier. Bubbles introduced into the water coalesce at the center of the flask and produce a dim light visible to the unaided eye in a darkened room.

The filled flask must be vibrated at its resonant frequency-that is, at the sound frequency at which it responds most intensely. The resonant frequency equals the speed of sound in water (1,500 meters per second) divided by the diameter of the sphere. The glass will cause the actual resonance frequency to be about 10 percent higher. We used a 100-milliliter Pyrex spherical boiling flask with a diameter of 6.5 centimeters. Filled with water, the container resonated at about 25 kilohertz. A small-necked flask will produce the best results. Grease and oil can interfere with the bubble, so the glassware should be thoroughly washed with soap and water and rinsed well.

You will need three ceramic piezoelectric transducers: two to create the acoustic wave and one to act as a microphone to monitor the sound of the collapsing bubble. We used disks 15 millimeters in diameter and six millimeters thick for the driving transducers. The microphone was three millimeters in diameter and one millimeter thick. As a courtesy to readers of Scientific American, the three transducers are offered as a set for $95 from Channel Industries, Inc., 839 Ward Drive, Santa Barbara, Calif.; telephone (805) 9670171; fax (805) 683-3420.

Connect fine wire (about 36 gauge) to the piezoelectric ceramics to serve as leads (thin wire minimizes sound loss). The wire is soldered to the silver electrodes on the ceramic. Remove the oxide layer on the transducers by rubbing them lightly with a pencil eraser. Working quickly with a cool soldering iron, place a small dot of solder on the silver sides of each piezoelectric transducer. Remove six millimeters of insulation from the end of the wire. Tin the copper lead (that is, melt some solder on it) and, after briefly heating the solder, place it on the solder dot. A wise move is to attach three leads to each disk, spaced equidistantly in the form of a triangle. This pattern ensures that each disk rests evenly on the curved surface of the flask. The other leads will also act as spares in case the first breaks.

Attach the transducers to the flask with epoxy. The quick-drying, five-minute type is recommended because it allows the transducers to be broken off the glassware without damage. Use just enough epoxy to fill the space between the flask and the transducer. For symmetry, place the two drive transducers on opposite sides on the equator of the flask and the microphone ceramic on the bottom. The transducers are polarized-one side will be identified with a plus sign or a dot. Make sure the two drivers are attached to the flask and wired in the same way: both should have plus signs toward the flask, or vice versa.

Solder a short lead to the outside of each transducer. Wire the drive transducers in parallel so they will expand and contract at the same time. Connect the wires to coaxial cables, which reduces electrical cross talk between the components. The microphone wires in particular should be short, extending no more than 10 millimeters before being connected to coaxial cables. Make the leads long enough so that they will not be under tension when connected. Suspend the flask either by clamping its neck to a laboratory stand or by hanging it with wires tied to the neck. Fasten all cables to the stand to prevent wire breakage.

The piezoelectric speakers act electrically as capacitors. To drive them with an audio amplifier (typically a low-voltage, low-impedance source), an inductor must be wired in series with them. The inductance is chosen so that it is in electrical resonance with the piezoelectric capacitance at about 25 kilohertz-that is, at the same frequency at which acoustic resonance occurs. The drivers described here will have a capacitance of about two nanofarads, so the inductance required is about 20 millihenrys. A good trick for adjusting the inductance is to use two or more inductors in series. By changing the distance between them, the total inductance may be raised or lowered by up to 50 percent. Two 10-millihenry inductors spaced about five centimeters apart would make a reasonable starting point.

To find the correct inductance, you will need to measure the voltage and current from the circuit [see illustration above]. Use a two-channel oscilloscope to display both quantities simultaneously. Get them in phase (their patterns on the oscilloscope should line up) by adjusting the inductance. Although the current from a home stereo system is low, the voltage may give a mild shock, so be sure all exposed connections and external wiring are insulated, covered with electrical tape or painted over with nail polish. The piezoelectric microphone transducer will typically produce about one volt; its output may be sent directly to the high-impedance input of the oscilloscope.

A sonoluminescent bubble can be created only in water in which the naturally dissolved air has been removed. A simple way to degas water is to boil it. Use a 500- to 1,000-milliliter Pyrex Erlenmeyer flask with an airtight stopper. Fit a hollow tube about six millimeters in diameter and about 10 centimeters long through the stopper and attach a short piece of rubber tubing to it. The tubing allows steam to vent and slows the diffusion of air back into the flask.

Fill the flask halfway with distilled water. Slowly heat the water and keep it at a rolling boil for 15 minutes. Then remove the flask from the heat, clamp the rubber tubing to prevent air from entering and allow the flask to cool (refrigeration will speed things up). After cooling, the flask will be under a strong vacuum, and the water will be well degassed. Keep sealed until ready to use, as the liquid will reabsorb air in a few hours when the container is opened.

carefully pour the degassed water into the resonator flask, letting it run down the wall. Doing so will introduce a little air, but that actually brings the amount in the water to about one fifth the atmospheric concentration, which is the correct level for sonoluminescence. The water will slowly regas but will remain usable for several hours. Fill the flask with water up to the bottom of the neck so that the fluid level makes the mass of water approximately spherical.

For a 100-milliliter spherical boiling flask, tune the audio generator to the approximate resonance frequency of 25 kilohertz. Set the oscilloscope to display simultaneously the output voltage of the amplifier and the current through the drivers. Turn the volume control so that the amplifier output voltage reads about one volt peak to peak and adjust the inductance so that the current is in phase with the voltage. Monitor the current, because it can exceed the limit for the coil, causing it to overheat. Also, periodically check that the frequency is set to the resonance peak, as it may change with the water level and temperature.

Now, on the oscilloscope, display the piezoelectric microphone output. As you vary the generator frequency, you will notice a broad peak in the microphone signal, about one to two kilohertz wide This peak corresponds to the electrical resonance between the inductor and the capacitance of the drivers. The acoustical resonance shows up as a much sharper peak in the microphone signal, less than 100 hertz wide, and as a slight dip in the drive current.

An easy way to find the resonance for the first time is to examine bubbles in the flask. For best viewing, position a bright lamp just behind the flask, because small bubbles scatter light much more efficiently in the forward direction. A dark background improves visibility. With an eyedropper, extract a small amount of water. While looking into the backlit flask, squirt the eyedropper at the surface hard enough to create about 10 to 30 bubbles. Adjust the generator to find the frequency at which the bubbles move toward the center and eventually coalesce into one. When everything is tuned correctly, it is usually possible to create a bubble just by poking a wire at the surface.

When you have a bubble in the center of the flask, slowly increase the amplitude. The bubble will be stable at first, and then it will "dance" over a few millimeters. Still greater amplitude will cause the bubble to stabilize again and shrink, becoming almost invisible, before growing again. Above a certain sound intensity, the bubble will disintegrate. Best light emission is obtained just below this upper amplitude threshold.

Smal1 ripples should be visible on the oscilloscope trace from the microphone. This signal is high-frequency sound emitted by the bubble as it collapses with each cycle. Watching the ripples is an easier way to monitor the status of the bubble than is looking at the bubble itself. An electrical high-pass filter can be used to attenuate the driving sound, making the ripples on the oscilloscope more apparent. To view the glow emitted by a bubble, turn off the room lights and let your eyes adjust to the darkness. You should see a blue dot, somewhat like a star in the night sky, near the center of the flask. The bubble may be made brighter and more stable by fine-tuning the frequency and amplitude of the driving sound. If the glowing bubble is moving in the flask or varying in brightness over a few seconds, the water contains too much air Try substituting freshly degassed water.

If the bubbles do not move at all or if they move toward the side of the flask, you are probably at the wrong frequency. Set the generator frequency to 24 or 26 kilohertz, readjust the inductance so the current and voltage are in phase and try again. Changing the water level or the way the flask is hung may improve the acoustics. As a last resort, carefully remove the transducers with a razor blade and try a different flask.

Having produced sonoluminescence, you can explore many questions about the phenomenon. For example, how do magnetic and electrical fields affect the light emission? How will various substances dissolved in the water change the behavior of the bubble? What new ways are there for probing the transduction of sound into light? With this home setup, you are ready to research at the frontiers of science.

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.