As a PhD student in rainy Dublin in the late 1990s, Dara Fitzpatrick suffered from frequent tonsillitis. Among his remedies was gargling with salt water. One day as he mixed his homemade solution, he heard a curious thing. The sound of the spoon tapping against the glass as he stirred in the salt rapidly switched from a high-pitched clatter to a much deeper banging noise. Then, as he continued to stir, the tap-tap gradually rose in pitch again until it sounded like it did at the start.
Fitzpatrick wasn’t the first to notice the acoustic dip. But unlike most who hear it, Fitzpatrick, an analytical chemist, was in a position to investigate the underlying mechanism. “It was in the back of my mind every time I made a cup of coffee,” he says. “I always wanted to see if it was a reproducible effect.”
Some 20 years since his chance observation, Fitzpatrick—now a research scientist at University College Cork in Ireland—has developed a way to turn the kitchen curiosity into a valuable analytical instrument. Through a technique he calls broadband acoustic resonance dissolution spectroscopy, or BARDS, he and his colleagues have harnessed the effect to check the purity of foods and to determine the beach from which a sample of sand came. They can even exploit sound to track the pH of an acid–base reaction without a chemical indicator or litmus paper. The technique is already being used by academic and industry researchers to complement other analytical tools such as mass spectrometry.
A fateful mug of hot chocolate
Most of the basic physics behind the acoustic phenomenon was worked out in the early 1980s when Frank Crawford, a physicist at Lawrence Berkeley National Laboratory in California, made the same observation as Fitzpatrick while mixing a cup of cocoa.
Crawford found that the hot chocolate effect, as he called it, comes down to bubbles. Typically when a solid powder dissolves in a liquid, lots of small bubbles are produced. The bubbles can stem from several sources: pockets of gas trapped on and between individual solid grains, the gas produced during chemical reactions as the powder dissolves, and gas that emerges when the dissolving solid squeezes previously dissolved volatiles out of the liquid.
Adding bubbles to a liquid makes it more compressible. And, as Crawford demonstrated, more compressible fluids transmit sound more slowly because the ability of molecules to squeeze together robs the passing sound waves of energy.
The changing noise of the tapping spoon reflects the drop in the system’s resonant frequency as the bubbles act to slow the propagating sound waves. That explains the hot chocolate effect’s sound profile: the quick drop in pitch as the initial cloud of bubbles from the powder is released and the solute begins to dissolve, and then the recovery to normal pitch as the bubbles dissipate.
A reproducible effect
About a decade ago, Fitzpatrick picked up the investigation where Crawford had left off. In a series of experiments, his research group used a sensitive microphone to track the changes in resonant frequency when various amounts of sodium carbonate were mixed into water. The researchers then tested other salts. Instead of using a teaspoon to stir and tap, they inserted a magnetic stirrer that tapped repeatedly against the side of a beaker as it spun.
Just after adding the solute, the frequency of the sound recorded by Fitzpatrick’s microphone tended to start at just over 8 kHz, a bit higher than that of a buzzing mosquito. Within 30 to 40 seconds, the frequency would drop sharply, to just over 1 kHz. It would then rise slowly over the next five minutes or so until leveling off at its original value.
Through repeated experiments Fitzpatrick learned that dissolving powders produced bubbles in a very consistent and predictable way. “What you think is a random chaotic dissolution turns out to be highly ordered in terms of gas evolution and loss,” he says. He found that adding the same amount of powder to the same liquid produced the same frequency profile each time. This intrigued Fitzpatrick. He envisioned researchers producing a list of reference frequency patterns; other scientists could then listen in on a dissolving solid and be able to identify it.
Recently Fitzpatrick built a standardized device that could be used to analyze chemical compounds. It’s based on his original lab setup: A stirrer taps against the side of the vessel as it mixes a liquid. After a microphone records the background resonant frequency, an automated scoop tips the powder in. The stirring continues for about five minutes, with the microphone collecting the acoustic readings as the bubbles come and go.
Putting BARDS into practice
After publishing his initial results, Fitzpatrick started receiving inquiries from other scientists, who soon began applying BARDS to practical problems. Saskia van Ruth, a food scientist at Wageningen University and Research in the Netherlands who has collaborated with Fitzpatrick, says it’s a valuable technique. “It’s fairly rapid, it’s not very destructive, it does not need any sample preparation, and it’s easy to use,” she says.
Van Ruth, Fitzpatrick, and colleagues published a study earlier this year that used BARDS to distinguish among 60 different types of commercial salts, from the cheap salt dumped onto roads in winter to the finest Himalayan pink table salt, which sells for up to $5 per 100 g. They found that coarser salts had more gas entrained on the surface of each grain and so tended to release their bubbles more rapidly than the more finely ground samples. Bubble profiles could be used to tell if genuine fine table salt has been cut with cheaper, rougher minerals, van Ruth says. She next plans to use the technique to try to identify spices and to distinguish between ground arabica coffee and the cheaper robusta variety.
Van Ruth has also used BARDS to analyze and identify the source of sand samples. When stirred into mild acid, beach sand produces a particularly strong acoustic signal because it contains crushed carbonates—the remains of long-dead sea creatures such as mollusks—which break down into carbon dioxide. Due to the differing proportions of seashells in sands at various locations, the researchers were able to differentiate among the source locations of sand taken from nine Dutch beaches.
BARDS is valuable because it complements other techniques that have traditionally been used to test and analyze such samples, van Ruth says. Mass spectrometry is good for determining composition, she says, and microscopy reveals structure. Although it’s less quantitative, BARDS offers information on both composition and sample size, because surface area determines how quickly a particle dissolves, which in turn influences bubble production.
One important caveat, van Ruth notes, is that BARDS can check powders only indirectly. The technique, after all, measures the release of bubbles rather than a specific physical property of the solid.
Fitzpatrick has patented the tabletop BARDS device and is selling it to drug companies, food scientists, and other researchers. Isabelle Déléris, who works in the research and development labs of food giant Cargill in Brussels, bought one earlier this year. “For the tests we have run so far, it’s been very helpful,” she says. “It’s not very easy to characterize over time what happens when you put a powder into a liquid.”
Alexander Fedorchenko, a thermodynamics researcher at the Czech Academy of Sciences in Prague who has also studied the effect of bubbles on sound waves in liquids, says the BARDS method is “useful and clever.” He predicts that measuring sound changes to track bubble evolution in solution will find more practical applications in the study of bubbly liquids (see the article by Roberto Zenit and Javier Rodríguez-Rodríguez, Physics Today, November 2018, page 44). “Even tap water is saturated with air, which is easy to see if you shake a bottle of water,” Fedorchenko says. He adds that BARDS could even help model processes in volcanic eruptions, which are driven by the rapid growth of gas bubbles in magma.
Fitzpatrick agrees that BARDS may find further applications. Last year he and his colleagues showed that bubble evolution could be used to track a liquid–liquid reaction. They measured the sound effects of the carbon dioxide bubbles produced when sodium carbonate solution was added to hydrochloric acid. The resulting spectra could quantify the volume of gas produced, which indicates the acid’s concentration, and thus its pH. “This brings pH measurement into the realm of human perception,” the scientists concluded in their paper. “A person with a well-trained ear is capable of determining the concentration and pH of a strong acid just by listening.”
