Dados de equilíbrio líquido-líquido para sistemas biodiesel+álcool+glicerol utilizando gordura de frango e óleo de cártamo

Abstract

The great dependence of fossil fuels on the planet generates a need for new sources of energy, since fossil oil is a finite and polluting richness. The environmental part must always be taken into account, since the greenhouse effect is a reality. In this context, biofuels are being increasingly demanded to meet the energy demand of the population and among them, biodiesel stands out because it is less toxic, renewable and biodegradable. Transesterification is the most applied method to transform oils and fats into methyl or ethyl esters, reducing their viscosity and can be used as pure fuel or added to diesel. The raw material can either be animal or vegetable. The use of animal fat also gains relevance because of the environmental issue, since it reduces pollution because it is not improperly discarded. The by-product formed is glycerin, causing two phases to be obtained. The liquid-liquid equilibrium can be studied during the production process, since the alcohol is miscible in both phases. In addition, it is extremely important that such data are available in the literature, since they can be used to compose databases in industrial process simulators. In this work, two raw materials were used for the production of biofuel: chicken fat, provided by the Apodi-RN municipal market, and refined safflower oil. Três tipos de biodiesel foram obtidos: metílico de gordura de frango, metílico de cártamo e etílico de cártamo. The ester profiles were detected by gas chromatography. The liquid-liquid equilibrium data were obtained by means of calibration curves of the densities as a function of the respective mass fractions for the temperatures of 25 ° C and 45 ° C, at ambient pressure. The validation of the data was done using the correlations of Othmer-Tobias (1942) and Hand (1930), with coefficients near the unit. The distribution and selectivity coefficients for the three systems were calculated. The UNIQUAC model was used to correlate the experimental data, with errors always lower than 1.6%.