Determination of the Kinetics of the Ethoxylation of Octanol in Homogeneous Phase Pascal D. Hermann, † Toine Cents, ‡ Elias Klemm, † and Dirk Ziegenbalg* ,† † Institute of Chemical Technology, University of Stuttgart, 70569 Stuttgart Germany ‡ Sasol, Group Technology, 1947 Sasolburg, South Africa * S Supporting Information ABSTRACT: A two-dimensional computational fluid dynamics model was utilized to test different kinetic models for the description of the anionic polymerization of octanol with ethylene oxide in a microreactor. The reaction was performed as a continuous reaction under elevated pressure and temperatures in a one-phase system. The kinetic parameters were determined with numerical methods by reducing the deviation to experimental data based on the Nelder-Mead method. Four different reaction models with one, two, three, and four different rates for the first propagation steps were tested. The best agreement with the experimental data was found for the four rate model, with a prediction accuracy close to the experimental error. The gathered data suggests an increasing reaction rate for the first four propagation steps, which is in agreement with the Weibull-Tö rnquist effect [Weibull and Tö rnquist. Berichte vom VI. Internationalen Kongress für Grenzflächenaktive Stoffe, Zürich, vom 11. bis 15. September 1972; Carl Hanser Verlag: Munich, 1972; p 125; Sallay et al. Tenside Deterg. 1980, 17, 298]. ■ INTRODUCTION The industrial production of nonionic surfactants is conducted by base-catalyzed reactions of hydrophobic compounds containing an abstractable hydrogen, such as alkylphenols, fatty alcohols, fatty acids, thiols, or alkylamines, with ethylene oxide or propylene oxide. 3 As one important product, ethoxylated fatty alcohols were valued to be $5.125 billion with a production of 3.27 million tons in 2014. 4 The conventional process is carried out semibatchwise in stirred tank reactors by introducing ethylene oxide into the gas phase or into the liquid reaction mixture with the alcohol/ alcoholate substrate. The usual process conditions are 150- 180 °C and 200-500 kPa with the reaction occurring in the liquid phase. 5 However, this type of process can possess mass- transfer limitations due to the dissolution process of the gaseous ethylene oxide in the reaction mixture. This process is slow compared to the rapid consumption of the solved ethylene oxide and with this reduces the productivity. 5,6 Furthermore, the heat transfer capability of this type of reactor is limited. Hence, it is necessary to slow down the reaction by slowly dosing the ethylene oxide to prevent a thermal runaway of this highly exothermic reaction. Thus, the productivity in a stirred tank reactor can be limited by both the mass transfer limitation and the heat removal capabilities of the vessel. Additionally, the rising pressure during reaction must be handled due to possible safety issues. To intensify the process and improve the safety, the process was transferred to continuous operation in a microreactor in preliminary work by Rupp et al. 7-9 Microreactors generally show improved heat transfer capabilities and better resistance to pressure, enabling operation in a single phase system under elevated pressures and temperatures. However, problems in reaction handling were also found during the experiments for a microstructured reactor: the formation of hot spots was observed. 9 For a correct description of this behavior it was necessary to create a robust model with good prediction and extrapolation capabilities. Ethoxylations in two-phase systems are usually modeled by assuming a single propagation rate constant and an equilibrium constant for the proton transfer reaction between the anionic and protonized species for all products and the alcohol as well as considering the dissolution process of ethylene oxide (EO) in the liquid reaction mixture. Temperatures of 180-240 °C and pressures of 9000-10 000 kPa were used in this work and the work of Rupp, 9 that share large parts of the experimental data. This exceeds the ranges investigated by the other works. A detailed list of the used process conditions and the resulting phase system of various published work is given in the Supporting Information. Furthermore, it is worth mentioning that the publications utilizing a batch reactor often face the problem of a low number of experiments as a result of the rather long reaction times and larger reactor volumes compared to that of a microreactor. Indeed, benefits of using a microreactor are the possibilities to conduct numerous experiments in a relatively short time and to ensure isothermal conditions by simply using a very small diameter of the reactor. Hence, the number of experiments used for the determination Received: March 6, 2017 Revised: May 5, 2017 Accepted: May 8, 2017 Published: May 8, 2017 Article pubs.acs.org/IECR © XXXX American Chemical Society A DOI: 10.1021/acs.iecr.7b00948 Ind. Eng. Chem. Res. XXXX, XXX, XXX-XXX