Fourier analysis of potential oscillations of a liquid membrane for the discrimination of taste substances

Electrical potential oscillations were obtained across a liquid membrane composed of nitrobenzene/picric acid placed between two aqueous phases in the presence of various taste (i.e. salty, sweet and bitter) substances. The influence of these compounds on electrical oscillations was studied using Fourier analysis to establish a "fingerprint" of the substance that can be correlated with its taste index. Various concentrations of each substance were tested to obtain a Fourier spectrum with discrete peaks which can be further processed. The electrical oscillations consisted of a number of weak damped oscillators, and the Fourier spectra of these signals were found to have a number of discrete peaks of decreasing amplitude at low frequencies (0-0.5 Hz). A correlation of the frequency of the first peak of the Fourier spectrum with the taste index was found for bitter substances, whereas for salty substances the amplitude of the first two peaks of the spectrum was correlated with the taste index.