Viscosity of Binary Mixture DMA + 2-Butanol With Potential use As Solvent for Regenerative Flue Gas Desulfurization Processes

Autor: Mirjana Lj. Kijevčanin, Divna M. Majstorović, Emila M. Živković, Nikola V. Živković
Rok vydání: 2019
Předmět:
Zdroj: Proceedings of the 32nd International Congress on Process Engineering.
DOI: 10.24094/ptk.019.32.1.213
Popis: Sulfur dioxide (SO2) is an environmental pollutant with multiple harmful effects on biological world. One of the main emission sources are industrial activities in thermal power plants, foundries and steel mills which involve combustion of fossil fuels containing sulfur, such as coal and oil. Initial step towards removal of SO2 from flue gases was the so-called limestone process, a wet non-regenerative process that even today is the most used flue gas desulfurization method. However, wet regenerative processes with gas absorption in liquid organic solvent and thermal regeneration recently have found practical application. Processes based on n,n-dimethylaniline (DMA) have already been patented and industrially applied. This solvent has high selectivity towards SO2 in comparison to the other components of flue gas like CO2, and excellent binding capability by the mechanisms of chemical absorption, but main disadvantage is its high toxicity. On the other hand, in more ambient friendly 2-butanol binding of sulfur-dioxide occurs by means of physical absorption. Investigations have shown that mixtures of physical and chemical solvents usually show better characteristics regarding binding capacity and selectivity than traditionally used aqueous solutions of amines, due to the synergetic effect. One of the most important thermophysical properties of solvents, necessary for process and equipment design is dynamic viscosity. In this investigation viscosity of binary mixture DMA + 2-butanol is experimentally determined in temperature and concentration range and correlated afterwards using different literature models. These correlations present useful method to acquire reliable data necessary for fluid flow analysis and mass and heat transfer calculations, since experimental values for desired process conditions are often not available. In this study, two or three-parameter Eyring-UNIQUAC, Eyring-NRTL and McAllister models were used, while model that couples Eyring equation, Peng-Robinson equation of state and van der Waals mixing rule, was tested for simultaneous calculation of density and viscosity. The obtained results are presented and discussed regarding potential application, limitations and complexity of used approach and models.
Databáze: OpenAIRE