During cavitation created by sound in a liquid the collapse of the bubbles can lead to very high temperatures and pressures resulting in the emission of light - sonoluminescence. Can the spectrum of this light be used to determine the presence of various elements in the liquid, either in solution or as part of the chemical makeup of the medium, or does the pressure result in lines broadening too much?
I also realize the technique is limited because only the part of the spectum to which the liquid is transparent is available.