Prediction of the refractive index and speed of sound of biodiesel from its composition and molecular structure

Abstract

The refractive index and speed of sound can be applied in many biodiesel studies, such as the characterization of biodiesels, their feedstocks, diesel/biodiesel blends, and their adulteration with vegetable oils. They have also been used to monitor the progress and endpoint of the transesterification reaction and the level of oxidation of biodiesel. About 350 oils are considered with potential for the production of biodiesel, and, for most of them, no experimental values of their refractive index and speed of sound are yet available. Regression models allowing an estimation of these two properties can be very useful. Thus, this study proposes regression models to determine the refractive index and speed of sound of biodiesel as a function of its composition and molecular structure (carbon chain length and unsaturation level). Experimental values of the refractive index for several oil-based biodiesels were measured. A regression study was conducted using our results and available experimental databases including a total of 47 data for the refractive index at 313 K and 263 data for the speed of sound in the range of 288-343 K. The equations, which were obtained using multiple linear regression analysis, correctly reproduced the dependencies of these biodiesel properties on carbon chain length and unsaturation level. The results obtained using the regression models proposed were compared with measured and available experimental data, showing very good agreement. The RMSE values of the refractive index and speed of sound models using all the data were lower than 0.0008 and 1.84 m/s, respectively.