2018-01-052017-11-10SOARES, Luan Henrique. Avaliação de métodos preditivos na determinação de propriedades termofísicas de líquidos iônicos. 2017. 79 f. Dissertação (Mestrado em Engenharia Química) - Universidade Federal de Alfenas, Poços de Caldas, MG, 2017.https://repositorio.unifal-mg.edu.br/handle/123456789/1062Ionic liquids (ILs), considered as "green solvents", are potential substitutes for organic solvents, which are harmful to the environment. In order to obtain higher efficiency of the equipment in the use of these fluids, it is necessary to know its thermophysical properties, such as specific mass (ρ), dynamic viscosity (μ), heat capacity (Cp) and thermal conductivity (k). The technological challenge of this work consists in the evaluation of the existing contribution models of groups (GC) by comparing the estimated values ​​of these properties to 14 imidazolium group ILs with a new updated experimental database, performing, when necessary, the reestimative of group contributions and proposition of new groups not considered in the models. Among GC models, those reported by Gardas and Coutinho were used because they were simple and representative. These models were evaluated with a new experimental database elaborated from the ILThermo, with 9107 experimental data of 14 ionic liquids with imidazolium cationic group for the selected properties. The groups proposed by the models would not be enough to predict the properties of all LIs, so it was necessary to use GC proposed in another work for the specific mass (ρ) and include Br- groups for the dynamic viscosity (μ), Cl- and DCA- for heat capacity (Cp) and DCA- and pressure parameter for thermal conductivity (k). The GC parameters of the models used were reestimated using the new experimental database elaborated by minimizing the mean relative deviation (σ). The results were obtained in the pressure ranges of 100 to 298900 kPa and temperature of 183.45 to 524.87 K, with mean relative deviations (σ) of 0.66%, 18.76%, 0.38% and 3.8% for ρ, μ, Cp and k, respectively. The evaluation of the predictive models used consisted in comparing such models with a new experimental database for 8 LIs different from those used in the work. It was verified that the CG models, with parameters re-evaluated in this work, had good representativeness for the properties of ρ, Cp and k. It should be noted that the GC model for μ may be used for the determination of LIs with Br- group, which was not possible when using the selected models. Thus, with the updated experimental database and the new contribution parameters of included groups, it might be determined the thermophysical properties analyzed for other ionic liquids not yet reported in the literature.application/pdfAcesso Abertohttp://creativecommons.org/licenses/by-nc-nd/4.0/Líquidos iônicos.Propriedades termofísicas.Modelagem de dados.ENGENHARIA QUIMICA::PROCESSOS INDUSTRIAIS DE ENGENHARIA QUIMICADissertaçãoRolemberg, Marlus Pinheiro