CFD modeling of multiphase flow in reactive distillation column
Graphical abstract
Introduction
Reactive distillation processes are used in several areas of chemical engineering. Due to occurring chemical reaction and distillation in one unit, the capital and operational cost of reactive distillation systems are reduced compared with conventional reactor-distillation column [1,2]. These processes also offer many other advantages over traditional methods such as simplification or elimination of the separation system (capital savings), improving selectivity and reducing by-product formation, catalyst requirement reduction significantly, avoidance of azeotropes, heat integration benefits, and preventing hot spot formation. Therefore, in many cases, the uses of these systems are preferred. Despite all the benefits of reactive separation, there are some limitations, including volatility constraints, the requirement of residence time and catalysts lifetime, scale up to large flows, process conditions mismatch, and the liquid phase reaction [1,[3], [4], [5]].
Recently, production of biodiesel additives increased significantly. 1,1 diethoxy butane is one of the acetals that can be used as oxygenated diesel additives. Acetals can be produced from an acid-catalysed reaction between alcohols and aldehydes [[6], [7], [8], [9], [10], [11]]. Due to high conversion, low energy consumption and reduction of capital costs, the reactive distillation process is one of the most preferred technologies in the production of acetals. Also, this process is applied for the limited equilibrium reactions such as reversible liquid phase reactions. Because of removing the products continuously, selectivity and conversion are increased [[12], [13], [14], [15], [16]].
Numerous experimental works were accomplished on the various types of reactive separation processes. Some of these works were focused on the synthesis of materials (e.g., acetal,) by reactive separation processes [[17], [18], [19], [20], [21], [22], [23]]. The main scope of others was on the hydrodynamics behavior and effect of internals on the process [[24], [25], [26], [27]]. According to the cost of hydrodynamic experiments and various internal equipment (including various packing and catalysts), the use of CFD simulation has recently become the most useful way to study on various types of packing and synthesis in reactive separation processes [[28], [29], [30]].
Van Baten et al. [31] studied the hydrodynamics of a reactive distillation sieve tray column. Catalyst containing wire-gauze envelopes were disposed along the liquid flow direction. A 3D two-phase model was developed to determine liquid clear height on the tray as a function of tray geometry and operating conditions. A 3D steady state one phase model was presented by Kloker et al. [2] to obtain the influence of different catalytic internals on the reactive distillation of n-hexyl acetate from hexanol and acetic acid.
Egorov et al. [32] have proposed a new modelling methodology for reactive separation which exploits a combination of modern CFD facilities and the rate-based process simulation approach. Hydrodynamics and mass transfer correlations were obtained by using CFD simulations. Zivdar et al. [33] investigated the dry pressure drop within the catalyst packed channels of katapak-S by using CFD simulations. Also, they presented the gas flow path line in the packing sheet and elbows.
Physical and reactive mass transfer absorption in gas-liquid flow was simulated by Haroun et al. [34]. Haroun et al. [35] simulated reactive mass transfer at the liquid interface in a two- phase flow. 2-D simulations were performed to investigate interfacial Mass transfer and liquid hold up by Haroun et al. [36]. In all three papers, they were using VOF model to capture the gas-liquid interface motion in structured packing. Also, they found that Higbie theory well predicts the liquid side mass transfer.
A 3D two-phase CFD model was established to study the separation performance of structured catalytic packing by Dai et al. [37]. Two types of structured catalytic packings (i.e. BH-1 and BH-2 types) were used in simulations. Liu et al. [38] analysed the multi-scale structure of a reactive distillation column by Aspen Plus with Fluent software. The reactive distillation column is divided into four scales: column scale, tray scale, fluid mechanical scale and molecule scale. Tray efficiency was calculated using CFD simulations. Ding et al. [39] presented a 3D model for simulating the winpak-based modular catalytic structured packing by using CFD. Mazarei et al. [40] investigated the pressure drop and flow pattern in the modular catalytic structured packings. Also, they illustrated the effect of geometry on the hydrodynamics and characterisation of flow in the katapak SP modules.
It appears from above review that there is lack of knowledge in reactive distillation processes. As experimental setups are so expensive and measuring parameters locally (i.e., mass fraction, temperature, pressure, velocity, etc.) are not possible, CFD simulation is the best approach to obtained accurate data in whole RD systems. Furthermore, it is simplified to evaluate the various internals effects. At previous work, a 3D VOF model that was developed to evaluate dry and wet pressure drop of katapak SP-11 & 12 [40] is validated by the Behrens [41] experimental data. Also, the catalyst bags of katapak SP were simulated as porous media to show the liquid velocity distribution in these bags. The goal of this work is to model the reaction, heat and multicomponent mass transfer that occurred in packings of RD columns. For this purpose, at first RD column base on Agirre [9] experimental work was simulated by commercial software. Secondly, a 3D two-phase VOF model was carried out. Finally, the results of multicomponent mass transfer in the separation section and the reaction and heat that appeared in the catalyst bags were compared with commercial software results. Fig. 1 represents the flowchart, describing the methodology of this work.
Section snippets
Experimental description
Agirre [9] studied on the production of 1,1 diethoxy butane in a packed bed column that used katapak SP-11 in the reactive section and Amberlyst resins as a catalyst.
DEB is obtained from ethanol and butanal in an exothermic reversible reaction. The stoichiometry of the reaction between them is as follow [7,9]:Ethanol + 1 Butanal ↔ 1 DEB + 1 Water
Moreover, they reported kinetic parameters of the reaction. The experiments were performed in a batch stirred tank reactor with Amberlyst resins as a
Simulations and assumptions
Several assumptions have been used in the simulations due to complicated geometry, lack of appropriate understanding of multiscale phenomena in katapak SP, and time-consuming simulations. Although, the hydraulic parameters such as dry and wet pressure drop have shown a periodic or "quasi-steady state" manner, their changes were considered negligible during simulation time. These parameters have some effect on the gas-liquid contact time and mass transfer rate between two-phases. As regards that
Hydrodynamics
The hydrodynamic behavior of flow affects the mass transfer and reaction that occurred in the column. In our previous study [40], a 3D steady state VOF simulations were done for investigating dry gas and two-phase flow in katapak SP-11 and 12. To making sure of hydrodynamics model that is used in the simulations, the results were compared with the experimental results of Behrens [41]. Then, this model is used in the simulation of DEB production. In this section, a summary of hydrodynamic
Conclusions
In the present study, a 3-D two-phase model for the study of hydrodynamic, chemical reaction heat and multicomponent mass transfer of 1,1 diethoxy butane production in the reactive distillation process was investigated. In the reactive distillation column, katapak SP-11 and 12 were used to understand more about the reactive separation internals.
Penetration theory was used to prediction the liquid mass transfer coefficient in the multicomponent mixture. Mass fraction of each component at the
References (56)
- et al.
Modeling reactive distillation
Chem. Eng. Sci.
(2000) - et al.
On the development of new column internals for reactive separations via integration of CFD and process simulation
Catal. Today
(2003) - et al.
Reactive distillation - industrial applications, process design & scale-up
Chem. Eng. Sci.
(2001) - et al.
Modelling of reactive separation process: reactive absorption and reactive distillation
Chem. Eng. Process.
(2003) - et al.
Multiplicity analysis in reactive distillation column using ASPEN PLUS
Chin. J. Chem. Eng.
(2006) - et al.
Thermodynamic and kinetic studies for synthesis of the acetal (1,1-diethoxybutane) catalyzed by amberlyst 47 ion-exchange resin
Chem. Eng. J.
(2015) - et al.
Catalytic reactive distillation process development for 1,1 diethoxybutane production from renewable sources
Bioresour. Technol.
(2011) - et al.
Synthesis of acetal (1,1-diethoxyethane) from ethanol and acetaldehyde over acidic catalysts
Appl. Catal. A: Gen.
(2000) - et al.
Esterification of acetic acid with ethanol: reaction kinetics and operation in a packed bed reactive distillation column
Chem. Eng. Process.
(2007) - et al.
Startup of a reactive distillation process with a decanter
Chem. Eng. Process.
(2008)