Alexander Graham Bell discovered the photoacoustic effect which researchers at the US Army Research Laboratory are attempting to exploit for the purpose of sensing poison gases. From the Aug. 14, 2012 news item on ScienceDaily,

To warn of chemical attacks and help save lives, it’s vital to quickly determine if even trace levels of potentially deadly chemicals — such as the nerve gas sarin and other odorless, colorless agents — are present. U.S. Army researchers have developed a new chemical sensor that can simultaneously identify a potentially limitless numbers of agents, in real time.

…

The new system is based on a phenomenon known as the photoacoustic effect, which was discovered by Alexander Graham Bell, in which the absorption of light by materials generates characteristic acoustic waves. By using a laser and very sensitive microphones — in a technique called laser photoacoustic spectroscopy (LPAS) — vanishingly low concentrations of gases, at parts per billion or even parts per trillion levels, can be detected. The drawback of traditional LPAS systems, however, is that they can identify only one chemical at a time.

Here’s how the researchers dealt with the limitation of being able to identify only one chemical at a time (from the news item),

[Kristan Gurton, an experimental physicist at the U.S. Army Research Laboratory (ARL) in Adelphi, Md] “As I started looking into the chemical/biological detection problem, it became apparent that multiple LPAS absorption measurements — representing an ‘absorption spectrum’ — might provide the added information required in any detection and identification scheme.”

To create such a multi-wavelength LPAS system, Gurton, along with co-authors Melvin Felton and Richard Tober of the ARL, designed a sensor known as a photoacoustic cell. This hollow, cylindrical device holds the gas being sampled and contains microphones that can listen for the characteristic signal when light is applied to the sample.

In this experiment, the researchers used a specialized cell that allows different gases to flow through the device for testing. As the vapor of five nerve agent mimics was flowed in, three laser beams, each modulated at a different frequency in the acoustic range, were propagated through the cell.

“A portion of the laser power is absorbed, usually via molecular transitions, and this absorption results in localized heating of the gas,” Gurton explains. Molecular transitions occur when the electrons in a molecule are excited from one energy level to a higher energy level. “Since gas dissipates thermal energy fairly quickly, the modulated laser results in a rapid heat/cooling cycle that produces a faint acoustic wave,” which is picked up by the microphone. Each laser in the system will produce a single tone, so, for example, six laser sources have six possible tones. “Different agents will affect the relative ‘loudness’ of each tone,” he says, “so for one gas, some tones will be louder than others, and it is these differences that allow for species identification.”

The signals produced by each laser were separated using multiple “lock-in” amplifiers — which can extract signals from noisy environments — each tuned for a specific laser frequency. Then, by comparing the results to a database of absorption information for a range of chemical species, the system identified each of the five gases.

Because it is optically based, the method allows for instant identification of agents, as long as the signal-to-noise ratio, which depends on both laser power and the concentration of the compound being measured, is sufficiently high, and the material in question is in the database.

But they still need to invent a device before they can take this process out of the laboratory,

Before a device based on the technique could be used in the field, Gurton says, a quantum cascade (QC) laser array with at least six “well-chosen” mid-infrared (MidIR) laser wavelengths would need to be available.

Here’s the citation for the article, which is behind a paywall,

Kristan P. Gurton, Melvin Felton, and Richard Tober. Selective real-time detection of gaseous nerve agent simulants using multiwavelength photoacoustics. Opt. Lett., 37, 3474-3476 (2012) [link]

There are more details in the ScienceDaily news item or you can check out the Aug. 14, 2012 (?) news release from the Optical Society of America.

I wonder what this research sounds like, unfortunately they didn’t include any audio files with the news release from the Optics Society of America or the news item on ScienceDaily.