Characterization of a photoacoustic system through neural networks to determine multicomponent samples

N M ZAJAREVICH; V B SLEZAK; A L PEURIOT

ELSEVIER SCIENCE B. V.

Elsevier B.V.

Lugar: Amsterdam; Año: 2016 vol. 77 p. 485 - 489

Photoacoustic spectroscopy for tracegases detection, based on a CO2 laser, can be used in a wide rangeof applications. The tunability of this laser in the mid-infrared (9.4 -10.6 μm) allows thequantitative determination of different substances in multicomponent samples.In general, at traces level, the total photoacoustic amplitude at a certainwavelength may be approximated by a linear superposition of the amplitudesgiven by each of the species absorbing at that wavelength. However, in somecases, the sum of the individual signals is no longer valid. In particular, itis known the presence of CO2 delays the acoustic signal in relationto the laser excitation due to the exchange of vibrational energy between CO2and N2. This phenomenon generates a slow V-T energy relaxation froma metastable N2 vibrational level and the sum of individualcontributions may no longer be valid. Moreover, the resolution of a linearequation system has limitations, so the possibility to determine concentrationsin photoacoustics based on neural network is proposed in this work. This procedureis tried in a particular case of a volatile organic compound, such as C2H4,and CO2 in air. The results are compared with the ones obtained witha model based on rate equations.