Liquid Phase Dehydration of 1-Butanol to Di-n-butyl ether Experimental Performance, Modeling and Simulation of Ion Exchange Resins as Catalysts

Abstract

Di-n-butyl ether (DNBE) is considered a very attractive oxygenate compound to reformulate diesel fuel. The research work performed along this thesis has clearly proven that sulfonic P(S-DVB) ion exchange resins are suitable catalysts for the synthesis of DNBE from the liquid phase dehydration of 1-butanol at the temperature range of 140-190 °C. A catalyst screening of acidic P(S-DVB) ion exchange resins showed that resins activity is enhanced with high acid capacities and with polymer matrices that are able to swell moderately allowing 1-butanol ready access to the active sites but without resulting in very large distance between the active centers. The resin Amberlyst 36 (oversulfonated, medium values of %DVB) proved to be the most active catalysts tested. However, in an industrial process a high selectivity to DNBE is extremely desirable from an environmental and economic standpoint, and resins that present a more elastic polymer matrix and higher ability to swell (gel type resins and Amberlyst 70) are the ones that present higher selectivities. Among all the tested resins, Amberlyst 70 was selected as the most suitable catalyst for industrial use given its excellent property balance: satisfactory activity, high selectivity to DNBE and thermal stability up to 463 K.

The relatively large value found for the thermodynamic equilibrium constant of DNBE formation and the fact that the formation of the secondary product 1-butene was extremely slow assure high conversion levels in an industrial etherification process. DNBE formation proved to be a slightly exothermic reaction, almost athermic, whereas 1-butene formation was found to be an endothermic reaction.

A comprehensive kinetic analysis enlightened that the reaction rate to form DNBE on Amberlyst 70 is highly influenced by the temperature and the presence of water. Two kinetic models that are able to predict the reaction rates of DNBE formation were identified. One of them stems from a Langmuir-Hinshelwood-Hougen-Watson (LHHW) formalism in which two adsorbed molecules of 1-butanol react to yield ether and water, being the reversible surface reaction the rate limiting step. The other one stems from a mechanism in which the rate limiting step is the desorption of water and where the adsorption of DNBE is negligible. Both models present several characteristics in common: in both water inhibition effect is correctly represented by a correction factor derived from a Freundlich adsorption isotherm; the number of free active sites is found to be negligible compared to the occupied ones; both present similar values of apparent activation energies (122 ± 2 kJ/mol). The study of the influence of typical 1-butanol impurities (isobutanol or ethanol and acetone, depending on the production process) demonstrated that isobutanol enhances the formation of branched ethers whereas ethanol leads to the formation of ethyl butyl ether and di- ethyl ether and acetone hardly reacts. In the second part of this thesis, it was demonstrated the suitability of molecular dynamics simulations in the understanding of the microscopic structure of P(S-DVB) ion-exchange resins and the prediction of their properties. Atomistic simulations of the structure of sulfonated P(S-DVB) resins reveled the decisive role that internal loops (closed polymer chains) play in defining the properties (i.e. density, porosity and structures) of highly cross-linked resins. Thus, although crosslinks ensure the local backbone structure, internal loops confer rigidity and loop architecture. It was also demonstrated that the developed atomistic model is able to predict the swelling behavior of ion-exchange resins in 1-butanol. From the swelling study performed by means of molecular dynamic simulation it was possible to characterize and quantify the structure of the swollen polymeric network at molecular level and to conclude that alcohol molecules tend to interact with the sulfonic groups of the resin.