The Research For Liquid Sensor Based On Phononic Crystal

In response to the cases of selling counterfeit gasoline blending, seeking for a detection method which is fast and effective is imperative. Since a mixture of different solutions well lead to a change in the speed of sound,you can find the the authenticity of gasoline by testing the speed of sound. In this paper, we have designed a liquid sensor based on a two-dimensional phononic crystal slab which is sensitive to the speed of sound, so a shift of the transmission peak will be detected in transmission spectrum, when the concentration of the mixture in scatterers changed.The contents are divided into three parts shown as follows.According to the basic theory of phonon crystals,we use matlab to calculate the band structures of a perfect phononic crystal by applying plane-wave expansion method. And with the help of Finite-element method, the defect bands, transmission spectrum and the transient response have been simulated to show the contribution of defect states’ dispersion curves to the transmission spectrum.When the acoustic parameters of the component material changed, there will be an impact on the band structure. The defect dispersion curve seems to have a shift in the frequency domain when the change is small enough.According to this phenomenon, we have designed a liquid sensor which is sensitive to the speed of sound for measuring the concentration of the mixture.We use a ethanol–water to test our liquid sensor model in finite-element software, the result shows a low concentration sensitivity about 90-190Hz/1%。We have made some optimizations of geometrical parameters to improve the sensitivity of our sensor. Varying the wave-guide width shows an increase in sensing ability when considering the first band-gap defect bands.We choose to add a scatter hole in the middle of the wave-guide core, the transmission results of FEM simulation shows a rate of 315~462Hz variation when the concentration changed one percent, and the central frequency of the dispersion curve also exhibits linear variation in small concentrations. We find that the bandwidth of the wave-guide dispersion curve is very narrow(less than 1 kHz) in the adjustment of the location of the scatter. The analysis shows that the correlation coefficient between the mid-frequency of the dispersion curve and the sound velocity of the mixture is over 0.998, so that the sensor can be able to characterize the sound velocity of the mixture.