- Email: [email protected]
Separation of acetic acid–methanol–methyl acetate–water reactive mixture
Chemical Engineering Science 101 (2013) 586–592
Contents lists available at ScienceDirect
Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces
Separation of acetic acid–methanol–methyl acetate–water reactive mixture Anastasia V. Penkova a,n, Galina A. Polotskaya a,b, Alexander M. Toikka a a Department of Chemical Thermodynamics & Kinetics, Saint-Petersburg State University, Universitetskiy prospect 26, Peterhof, Saint-Petersburg 198504, Russia b Institute of Macromolecular Compounds RAS, Saint-Petersburg 199004, Russia
H I G H L I G H T S
The coupling of reaction and pervaporation. The reactive mixture with methyl acetate synthesis reaction. Pervaporation of mixtures which are close to chemical equilibrium. Removing of methyl acetate from reaction zone in pervaporation. Developed PPO/MFFC membranes are effective to shift equilibrium.
art ic l e i nf o
a b s t r a c t
Article history: Received 20 August 2012 Received in revised form 23 April 2013 Accepted 27 May 2013 Available online 10 June 2013
The aim of this research is to study the separation of a mixture with a methyl acetate synthesis reaction by pervaporation. The feed solutions that have compositions close to a chemical equilibrium are the main object of this work. The pervaporation of these solutions leads to the shifting of the composition from its chemical equilibrium. As a result, ester synthesis reaction coupled with pervaporation allows for more effective conversion of reagents. The hybrid process “reaction+pervaporation” was studied using membranes with a base of poly-(2,6-dimethyl-1,4-phenylene oxide) (PPO): a homogeneous PPO membrane and a composite membrane consisted of a PPO thin selective layer ( 5 mm) on the surface of a fluoroplastic composite hydrophobic membrane (MFFC). The transport properties of homogeneous and newly developed composite membranes based on the PPO were studied through the pervaporation of quaternary system acetic acid–methanol and water–methyl acetate. The sorption experiments were carried out to explain the pervaporation data. The possibility of using transformed composition variables for the analysis of pervaporation of mixtures in chemical equilibrium is demonstrated. It was shown that homogeneous and composite membranes are effective in shifting the esterification reaction equilibrium in pervaporation by the removal of methyl acetate or other esters from the reaction mixture. Composite membranes exhibited higher transport parameters (permeability and selectivity) than homogeneous membranes. & 2013 Elsevier Ltd. All rights reserved.
Keywords: Chemical equilibrium surface Membranes Pervaporation Reaction engineering Separations Transport processes
1. Introduction The study of separation processes in systems with chemical reactions could be important in the development of a theory of coupled reaction–mass-transfer processes. In modern chemical engineering, mass-transfer processes coupled with chemical reactions are used in the design of resource- and energy-saving technologies and should be considered a significant element of green chemistry. Coupled processes have numerous industrial
n
Corresponding author. Tel.: +7 812 428 48 05. E-mail address: [email protected] (A.V. Penkova).
0009-2509/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ces.2013.05.055
applications. One of the well-known examples is reactive distillation (Sundmacher and Kienle, 2003; Shah et al., 2012): Reactive distillation, which favorably shifts equilibrium through the removal of one of the products, is becoming more common in plant-scale production. It is, however, an energy-demanding operation and is not recommended when dealing with temperaturesensitive chemicals or biocatalysts (Lim et al., 2002). Reactive distillation can be combined with membrane separation (Aiouache and Goto, 2003; Liu et al., 2005a, 2005b; Steinigeweg and Gmehling, 2004). For example, the processes in reactive distillation–pervaporation hybrid columns for tert-amyl alcohol etherification with ethanol has been considered (Aiouache and Goto, 2003). Membrane separation can also be coupled with synthesis reaction. The combination of
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
a chemical reactor with a separate membrane pervaporation unit may lead to an effective coupled reaction–mass-transfer process. The processes in a pervaporation membrane reactor has been studied in various experimental works and theoretical research (Huang et al., 2004; Sanz and Gmehling, 2006; Domingues et al., 1999; Korkmaz et al., 2011; Rewagad and Kiss, 2012; Suttichai et al., 2003; Liu and Chen, 2001; Xuehui and Lefu, 2001; Assabumrungrat et al., 2003; Feng and Huang, 1996; Dos Reis Gonçalves et al., 2005; Liu et al., 2005a, 2005b; Delgado et al., 2009; David et al., 1991a, 1991b; Kita et al., 1988, 1987). Most of these studies concern the esterification reaction Alcohol ðR1 Þ þ Acid ðR2 Þ ¼ Water ðP 1 Þ þ Ester ðP 2 Þ: Esterification of carboxylic acids with alcohols is a typical example of an equilibrium-limited reaction. The conversion of reagents in such reactions depends on the thermodynamic constant of chemical equilibrium. Correspondingly, shifting the equilibrium leads to a higher conversion. Using excess alcohol as a cheap reagent is a practical method to achieve a high ester yield. On the other hand, the use of a large amount of reactant increases operation costs because it is necessary to separate the excess alcohol. The other way is to combine reaction and separation processes for the extraction of a main product or side-product. The esterification reaction could be driven by the extraction of products through reactive distillation. Another attractive method is the use of a pervaporation membrane reactor. The pervaporation is a ratecontrolled process and its efficiency in separation is not limited by related volatility as it would be in reactive distillation. In addition, the pervaporation separation could be carried out at a reaction temperature that is important for the technological design of the process. Finally, energy consumption in pervaporation is usually low (Mulder, 1996; Baker, 2004). Preliminary results on the use of a pervaporation membrane reactor as well as the description of a hybrid process of ester production have been presented in recent papers (David et al., 1991a, 1991b; Kita et al., 1988, 1987; Keurentjes et al., 1994; Kwon et al., 1995). The typical scheme of the hybrid process “reaction+pervaporation” (Feng and Huang, 1996) is shown in Fig. 1. Reagents (acid and alcohol) and a catalyst that is either homogeneous or heterogeneous are merged in a reactor (1). During the reaction, a quaternary equilibrium mixture is obtained. This mixture is transmitted to a membrane cell (2), where one of the reaction products is removed by pervaporation (3). The residual reaction mixture that does not pass though the membrane goes back to the reactor. In this way, excess acid and alcohol appears in the reactor and increases the conversion. The development of the new membranes which are effective for the application in the hybrid process “reaction+pervaporation” is also an important element of chemical engineering design.
587
Primarily, hydrophilic membranes based on polyvinyl alcohol and polyesterimide have been used for reactive pervaporation (Zhu et al., 1996; Nemec and Robert van, 2005), and this membrane type facilitated water removal from the reaction zone. In our previous work, organophilic membranes based on poly-(2,6dimethyl-1,4-phenylene oxide) (PPO) were studied through the pervaporation of a quaternary mixture containing components of ethyl acetate synthesis in equamolar content (Polotskaya et al., 2007, 2005). It established selective properties of PPO membrane with respect to ethyl acetate. Removal of the ester from the reaction medium by the membrane pervaporation can shift the esterification equilibrium and, at the same time, point to potentially important industrial solvents. In this presented paper, we consider the various aspects of reactive pervaporation in the system with methyl acetate reaction. The important element of the current study is the composition of the feed reactive solution. Most of the work on non-reactive and reactive pervaporation is focused on a binary feed mixture. On the other hand, it is evident that shifting the chemical equilibrium through reactive pervaporation should be studied for solutions close to chemical equilibrium compositions. Correspondingly, we consider the feed mixture compositions nearing an equilibrium state of methyl acetate synthesis. In this paper, we present some of our results of a systematic study of pervaporation of a reactive mixture for the case of compositions that are close to chemical equilibrium. The shifting from chemical equilibrium leads to a greater conversion of reagent, which is necessary for the development of resource-saving chemical technologies. The objects of this study were two membrane types based on PPO: a homogeneous PPO membrane and a composite membrane consisted of a thin PPO selective layer ( 5 mm) on the surface of hydrophobic fluorocarbon polymer porous support (MFFC). The dense film homogeneous membrane makes it possible to research transport properties of chosen selective polymer PPO. The composite membrane with a thin layer of selective polymer can ensure high-level permeability for the industrial application of a developed membrane process.
2. Material and methods 2.1. Materials Poly(2,6-dimethyl-1,4-phenylene oxide) (PPO) with a molecular weight of 172 000 and intrinsic viscosity of 1.58 dl/g (Brno, Czech Republic) was used for the work. Methyl acetate, methanol, and acetic acid were purchased from Vecton (Russia). The microfiltration fluoroplastic composite hydrophobic membrane MFFC (Vladipor, Russia) is composed of a porous fluoroplast F42L layer on a polypropylene support. The pore size is 0.05 mcm, with a total porosity of 80%. 2.2. Membrane preparation 2.2.1. Homogeneous membranes The homogeneous PPO membranes with a thickness of 60 mm were prepared by casting a 2 wt% PPO solution in chloroform on a cellophane surface and drying to remove the solvent. After that, the PPO film was separated from the cellophane and dried to a constant weight in a temperature set at 333 K.
Fig. 1. Scheme of hybrid process of ester production: (1) reactor, (2) membrane, and (3) permeate (Sanz and Gmehling, 2006).
2.2.2. Composite membranes The formation of a thin selective PPO layer in the composite membrane was created by casting 2 wt% PPO solution in chloroform on a surface of MFFC support and drying in vacuum at 313 K. To provide the formation of the selective layer at a thickness of
588
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
Fig. 2. Schematic illustration of pervaporation equipment: (1) feed tank, (2) thermostat, (3) membrane, (4) pressure controller, (5) vacuum pump, and (6) cold trap.
6–8 mm, the ∼0.03 ml of polymer solution was split on 1 cm2 of membrane surface. 2.3. Pervaporation experiment Pervaporation properties were measured by using laboratory equipment shown in Fig. 2. The membrane stainless steel cell (1) with an effective membrane area of 14.8 cm2 was provided with the stirrer. The membrane (3) was clamped into the cell and the feed was poured into the tank (2). Downstream pressure below 30 Pa was maintained using a vacuum pump (5) and controlled by a pressure-gauge (4). The permeate was collected in a liquid nitrogen trap (6), weighed, and then analyzed using gas chromatography. The gas chromatograph (Cristall 5000.2, Chromatec, Russia) was equipped with a Polysorb-1 column and a thermal conductivity detector to carry out quantitative analysis of the feed and the permeate composition. 2.4. Sorption experiment The sorption experiment was carried out by gravimetric method. Membrane films of known weight were immersed in a liquid at 293 K. After three days, they were removed and weighed after the superfluous liquid was wiped away. This procedure was repeated until the weight of the swelling films became constant. The degree of equilibrium swelling or sorption of liquid in the polymer membrane, S (g of liquid/100 g of dry polymer), was calculated by the equation S ¼ ðmS −m0 Þ=m0
ð1Þ
where ms is the weight of a swollen membrane and m0 is the weight of a dry membrane.
components of a liquid mixture through dipole-dipole, ion-dipole, or hydrogen-bonding (Polotskaya et al., 1994). Therefore, permeability and separation properties of membranes in pervaporation depend on interaction between feed components and the membrane polymer (Baker, 2004; Prausnitz et al., 1999). The permeability of liquid in pervaporation is significantly controlled by the solubility of a penetrant in the polymer membrane. To give an exhaustive characterization of the transport process, besides pervaporation tests where membrane permeability and selectivity could be measured, membrane sorption of individual components of liquid mixture should also be studied. Equilibrium sorption of individual components of reacting mixture methyl acetate, methanol, water, and acetic acid was determined for PPO and fluoroplast F42L that are materials of membranes under study. Table 1 lists the degree of equilibrium sorption of membrane polymers with liquid mixture components. It is significant that water is absolutely inert to membrane polymers because PPO and fluoroplast F42L are organophilic polymers. Results demonstrate that methyl acetate is sorbed better than other liquids by the polymers under study. Fluoroplast F42L exhibits the best sorption ability with respect to methyl acetate. Both membrane polymers moderately sorb methanol and acetic acid, though fluoroplastic F42L sorbs these liquids in lesser amounts. The data of Table 1 give some explanation of why membranes based on fluoroplast F42L cannot be used for separating this quaternary mixture alone. According to data of Table 1, the fluoroplast F42L is inert to water; sorption of methanol and acetic acid is small but the F42L swelling in methyl acetate is very high. The high value of methyl acetate sorption indicates that the membrane is on the frontier of solubility. Therefore, the fluoroplast F42L cannot be used as membrane material by itself.
3.2. Pervaporation properties of homogeneous and composite membranes 3.2.1. Pervaporation of model quaternary mixture Firstly to check membrane properties and selectivity, we studied pervaporation through membranes based on PPO for model quaternary mixture with equimolar contents of reagents and products of reaction: 16.7 mol% acetic acid, 16.7 mol% methanol, 33.3 mol% methyl acetate, 33.3 mol% water. Two membrane types based on PPO were used in the pervaporation of the quaternary mixture. They were a homogeneous membrane, 60 mm in thickness, and a composite membrane consisting of a thin PPO selective layer ( 5 mm) on the surface of a MFFC porous support. The composite membrane was prepared by taking into account the peculiarities of selective-layer formation established in the published work (Polotskaya et al., 2005; Toikka et al., 2009). The fulfillment of a condition for casting a PPO solution of 1≤[η] c≥8, where [η]—intrinsic viscosity (dl/g), c—polymer solution concentration (g/dl), assured the formation of nondefective membranes. In our case, [η] c¼ 1.58 2 ¼3.16 was used. Table 1 Degree of equilibrium sorption of membrane polymers with liquid mixture components. Solvent
3. Results and discussion 3.1. Sorption characterization of membrane polymers In a membrane process the polymer acts as “a molecular sieve” with sorption active centers that are able to interact with
Acetic acid Methanol Water Methyl acetate
Sorption, wt% PPO
Fluoroplast F42L
23.6 16.0 0 26.0
7.50 5.20 0 131
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
Transport properties of the PPO/MFFC membrane were compared with those of the PPO homogeneous membrane. It was established that the permeability of the composite membrane (0.33 kg/m2 h) was essentially higher than for the homogeneous membrane (0.11 kg/m2 h). Moreover, qualitative permeate composition depends on the type of used membrane. Fig. 3 illustrates how the composition of the quaternary mixture in the feed changes after pervaporation through PPO and PPO/MFFC membranes. The content of methanol and acetic acid in the permeate was smaller than that in the feed. Thus, reagents remained essentially in the reactor due to the use of membranes based in PPO. Acetic acid did not transfer practically through the membrane. The same low selectivity to acetic acid in pervaporation of quaternary mixtures has been noted (Liu and Chen, 2001) in hybrid processes with pervaporation at butyl acetate synthesis. Fig. 3 shows that reaction products penetrated through our membranes by different modes. For the case of the PPO membrane, the amount of water in the permeate was higher than that in the feed. Another trend observed was that the PPO/ MFFC membrane had an essentially lower permeability for water than the homogeneous membrane. This fact can be explained by the influence of the hydrophobic fluoroplastic material of the MFFC support that offers an additional resistance to water transfer and considerably influences the selectivity of the process. The main component of the permeate in both types of membranes is methyl acetate, though in the case of the composite membrane, the selectivity relative to methyl acetate was higher in comparison with the homogeneous membrane. The pervaporation data on the high selectivity of the PPO/MFFC membrane in methyl acetate removal from a quaternary mixture is coordinated with enhanced sorption properties of fluoroplast F42L with respect to methyl acetate.
323.15 K. The experimental data of the work (Pisarenko and Serafimov, 1978; Pisarenko, 1979) influenced the choice of feed compositions in our research: five compositions uniformly spaced at a chemical equilibrium surface in composition tetrahedron (or in the square of αi —variables). Due to the differences between the temperatures of pervaporation and experiments of works (Pisarenko and Serafimov, 1978; Pisarenko, 1979), feed compositions were not precisely in chemical equilibrium but belonged to the surroundings of chemical equilibrium surface. Tables 2 and 3 list the experimental results on the pervaporation of chemically equilibrium mixtures using homogeneous PPO and composite PPO/MFFC membranes; feed and permeate compositions, such as fluxes of the processes are presented here. It should be noted that all experiments were conducted without catalysts, such as acid or ion-exchange resins. This fact provides a constant feed composition in pervaporation experiments because the reaction rate is negligible at 293.15 K. Also, we did not observe, the shift of solution compositions by removing the permeate removing because the quantity of permeate was extremely small compared to the feed. Tables 2 and 3 show how the membrane flux depends on the feed composition. Flux through homogeneous and composite membranes rises with increasing methyl acetate content in the feed. This phenomena needs additional explanation. It could be explained on the basis of sorption experiments: methyl acetate is the most active agent in membrane swelling (Table 1).
Table 2 Pervaporation of quaternary equilibrium mixtures of different feed compositions by using homogeneous PPO membranes at 293.15 K. No.
3.2.2. Pervaporation of chemically equilibrium quaternary mixtures The data sets on chemical and vapor–liquid equilibrium in acetic acid–methanol–water–methyl acetate systems have been discussed in Toikka et al. (2009). Isothermal data on equilibrium that is closest in condition to our pervaporation experiment (293.15 K) are presented in Pisarenko and Serafimov (1978) and Pisarenko (1979) at 318.15 K and Bernatová et al. (2006) at
589
1 2 3 4 5
Feed composition, mol% Acetic acid
Methanol
Water
10.3 15.2 3.30 11.3 0.48
9.70 15.6 46.5 11.5 82.7
18.5 30.1 24.4 58.0 8.97
Methyl acetate 61.5 39.1 25.8 19.2 7.85
Permeate composition, mol% Flux, kg/m2 h Methanol Water Methyl acetate 8.40 14.1 39.1 9.40 79.4
26.3 34.1 27.2 60.3 10.4
65.3 51.8 33.7 30.3 10.2
Fig. 3. Composition of quaternary mixture in feed and in permeate after pervaporation through PPO and PPO/MFFC membranes, 293.15 K.
0.091 0.040 0.039 0.022 0.016
590
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
Table 3 Pervaporation of quaternary equilibrium mixtures of different feed compositions by using composite PPO/MFFC membranes at 293.15 K. No.
1 2 3 4 5
Feed composition, mol%
Flux, kg/m2 h
Permeate composition, mol%
Acid
Methanol
Water
Methyl acetate
Methanol
Water
Methyl acetate
10.3 15.2 3.30 11.3 0.48
9.70 15.6 46.5 11.5 82.7
18.5 30.1 24.4 58.0 8.97
61.5 39.1 25.8 19.2 7.85
7.50 11.7 40.6 9.20 77.8
14.8 30.6 18.5 46.7 18.1
77.7 57.7 40.9 44.1 4.10
Fig. 4. The dependence of fluxes on compositions (αi ) for the surface of chemical equilibrium for homogeneous (blue color) and composite (red color) membranes at 20 1C. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
Pervaporation of quaternary mixtures by using a PPO/MFFC membrane is more effective and productive than using a homogeneous PPO membrane in the process. In the case of the composite membrane, methanol and water content in the permeate decreases, yet it is important to note that the methyl acetate content in the permeate increases. The results of the pervaporation of quaternary mixtures in chemical equilibrium are in agreement with data of sorption tests (Table 1) and show the efficiency of both membrane types in separating products of esterification reaction. The use of pervaporation in hybrid processes of methyl acetate synthesis can facilitate a shift in equilibrium and enhance the conversion of esterification. Fig. 4 shows dependence of flux on composition of chemically equilibrium feed (in α—variable). Permeate of chemically equilibrium feeds (Tables 2 and 3) is not chemically equilibrium; therefore, a graphical illustration of the transformed composition variables cannot be presented. On the other hand, the pervaporation data sets show that permeate does not contain acetic acid for all feeds compositions. It gives us the opportunity to consider the permeate as a ternary mixture. Thus, permeate compositions can be represented by an ordinary concentration triangle, but a graphical illustration of the ratio of permeate content to feed content needs a three-dimensional figure (tetrahedron of compositions). The edge of the alcohol–ester–water of the composition tetrahedron corresponds with the permeate composition, and points within the tetrahedron correspond to the feed compositions. Fig. 5 shows the appropriate diagram of permeate content on feed content where lines connected points of feed and permeate compositions are analogs of tie-lines of liquid–vapor equilibrium.
1.58 0.17 0.22 0.16 0.09
Fig. 5. “Tie-lines” of pervaporation process in the acetic acid–methanol–water– methyl acetate system at 20 1C in the composition tetrahedron. Green squares: points of feed compositions; red squares: permeate of composite membrane; and blue squares: permeate of homogeneous membrane. Compositions are presented in molar percents. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
4. Conclusions In chemical syntheses, the conversion of reagents could be improved by removing the reaction products from the reaction zone. This process results in shifting the reaction equilibrium and correspondingly increasing the reaction rate. In the case of methyl acetate synthesis, we propose using the pervaporation separation to remove ester from the reactive mixture. In our study, for the pervaporation stage, we used membranes with a base of polyphenylene oxide (PPO). The study of the sorption and pervaporation properties of PPO membranes shows high sorption activity and selectivity of the membrane polymer to methyl acetate. To illustrate the efficiency of the coupling of synthesis with separation, the pervaporation study was carried out for the solution with a solution of compositions, which were close to chemical equilibrium. Homogeneous and composite membranes proved effective for the shift of esterification reaction equilibrium in pervaporation by the removal of methyl acetate or other esters from the reaction mixture. Composite membranes exhibit higher transport parameters (permeability and selectivity) than that of homogeneous membranes. The hybrid process “reaction+pervaporation” can be proposed as an alternative technique method of producing esters in industry. Increase of desired product yield, rational process realization in energy- and resource-saving regime due to membrane technology, such as high ecological purity, in the proposed process may prove promising in fundamental and industrial applications. In the
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
special case of methyl acetate synthesis, the PPO membranes could be used for the development of such technology.
Acknowledgment This work was supported by Grant of Ministry of Education and Science of Russian Federation nos. 2011-1.3.1-207-008-058 (GC 16.740.11.0658_02.06.2011) and by Grants of Russian Foundation for Basic Research nos. 12-03-00522 and 12-03-33155.
Appendix Theory
591
The consideration of physical–chemical properties with the use of three-dimensional diagrams could be accompanied by some graphical and other difficulties. For example, the dependence of any property on its composition should be presented formally as a hyper surface in 4D thermodynamic space. On the other hand, for the states of chemical equilibrium, the degree of freedom is reduced. As a result, the composition of phases in chemical equilibrium in the system with reaction R1 þ R2 ¼ P 1 þ P 2 could be presented in 2D space with the help of so-called transformed composition variables (αi ). These variables were first introduced by Zharov (1970) and later were proposed by Barbosa and Doherty (1987) (transformed composition coordinates). Due to the values of stoichiometric numbers of the reaction Eq. (2) (−1, −1, +1, +1), the composition complex for states in chemical equilibrium in such cases is a square of variables αi , which are connected with molar fractions xi by the following relationships: α 1 ¼ x1 þ x4
Compositions corresponding to chemical equilibrium of esterification reaction Compositions of quaternary mixtures are usually graphically presented in composition tetrahedron. The vertexes of tetrahedron correspond to pure substances, while edges and faces represent compositions of binary and ternary sub-systems. The equilibrium chemical reaction (i.e., chemical equilibrium) reduces the number of thermodynamic degrees of freedom. Accordingly, compositions in chemical equilibrium cannot belong to the whole compositional space of the tetrahedron. For the quaternary mixture with ester synthesis reaction, compositions in chemical equilibrium belong to the surface in tetrahedron (Toikka et al., 2009; Toikka and Jetkins, 2002). The position of a chemical equilibrium surface for the methyl acetate synthesis reaction
where νi and μi are stoichiometric number and chemical potential of specie i. In the state of chemical equilibrium affinity become zero value
Acetic acid ðR1 Þ þ Methanol ðR2 Þ ¼ Water ðP 1 Þ þ Methyl acetate ðP 2 Þ;
Α ¼ − ∑ νi μ i ¼ 0
α 2 ¼ x2 þ x4 α3 ¼ x3 −x4 α1 þ α2 þ α3 ¼ 1:
The thermodynamic condition of chemical equilibrium could be presented by means of affinity (A) (Prigogine and Defay, 1954): 4
Α ¼ − ∑ νi μ i ; i¼1
4
ð2Þ
is similar to other systems with esterification reaction (2) (Toikka et al., 2009; Toikka and Jetkins, 2002; Toikka and Toikka, 2006). According to the reaction Eq. (2), all ternary subsystems and two binary subsystems (acetic acid–methanol and water–methyl acetate) of quaternary acetic acid (R1)–methanol (R2)–water (P1)– methyl acetate (P2) system cannot be in chemical equilibrium. Therefore, a chemical equilibrium surface does not belong to the sides of the tetrahedron and edges R1 −R2 and P 1 −P 2 . The points of pure components (vertexes of tetrahedron) and non-reactive binary subsystems (edges R1 −P 1 , R1 −P 2 , R2 −P 1 , R2 −P 2 ) are considered as borders of the chemical equilibrium surface in compositional space. The quantitative view of chemical equilibrium surface is presented in Fig. 6.
ð3Þ
and
i¼1
ð4Þ
ð5Þ
Eq. (5) gives an opportunity to determine the position of a chemical equilibrium surface in compositional space (tetrahedron). It could be done with the use of experimental data on chemical potentials and dependence μi on composition. Another way for determining the chemical equilibrium is the calculation based on the thermodynamic models. The analytical presentation of a chemical equilibrium surface could be based on the calculation of a thermodynamic constant of equilibrium K e Ke ¼
a3 a4 ; a1 a2
where ai means the activity of component i. Unfortunately the errors in experimental determination of K e are considerable as a rule. Therefore, the determination of a chemical equilibrium is usually based on the direct experiment: reactions between species for various initial compositions of a reactive mixture. Of course, the correct presentation of a chemical equilibrium surface needs detailed data sets on chemical equilibrium. In the case of equilibrium of methyl acetate synthesis reactions, we used the experimental data (Pisarenko and Serafimov, 1978; Pisarenko, 1979). References
Fig. 6. Quantitative view of chemical equilibrium surface in the system with chemical reaction R1 þ R2 ¼ P 1 þ P 2 .
Aiouache, F., Goto, S., 2003. Reactive distillation–pervaporation hybrid column for tert-amyl alcohol etherification with ethanol. Chem. Eng. Sci. 58, 2465–2477. Assabumrungrat, S., Kiatkittipong, W., Praserthdam, P., Goto, S., 2003. Simulation of pervaporation membrane reactors for liquid phase synthesis of ethyl tert-butyl ether from tert-butyl alcohol and ethanol. Catal. Today 79-80, 249–257. Baker, R., 2004. Membrane Technology and Applicationsecond ed. Wiley, Menlo Park. Barbosa, D., Doherty, M.F., 1987. A new set of composition variables for the representation of reactive phase diagrams. Proc. R. Soc. London A 413, 459–464.
592
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
Bernatová, S., Aim, K., Wichterle, I., 2006. Isothermal vapour–liquid equilibrium with chemical reaction in the quaternary water+methanol+acetic acid+methyl acetate system, and in five binary subsystems. Fluid Phase Equilibr. 247, 96–101. David, M.O., Nguyen, T.Q., Neel, J., 1991a. Pervaporation– esterification coupling: Part I. Basic kinetic model. Chem. Eng. Res. Des. 69, 335–340. David, M.O., Nguyen, T.Q., Neel, J., 1991b. Pervaporation–esterification coupling: Part II. Modeling of the influence of different operating parameters. Chem. Eng. Res. Des. 69, 341–346. Delgado, P., Sanz, M.T., Beltrán, S., 2009. Pervaporation of the quaternary mixture present during the esterification of lactic acid with ethanol. J. Memb. Sci. 332, 113–120. Domingues, L., Recasens, F., Larrayoz, M., 1999. Studies of a pervaporation reactor: kinetics and equilibrium shift in benzyl alcohol acetylation. Chem. Eng. Sci. 54, 1461–1465. Dos Reis Gonçalves, F., Borges, L.E.P., Borges, C.P., 2005. Synthesis of ethyl acetate by coupling a heterogeneous catalytic system with a pervaporation unit. Separ. Sci. Technol. 39, 1485–1500. Feng, X., Huang, R.Y.M., 1996. Studies of a membrane reactor: esterification facilitated by pervaporation. Chem. Eng. Sci. 20, 4673–4679. Huang, Y.-S., Sundmacher, K., Qi, Z., Schlunder, E.-U., 2004. Residue curve maps of reactive membrane separation. Chem. Eng. Sci. 59, 2863–2879. Keurentjes, J.T.F., Janssen, G.H.R., Gorissen, J., 1994. The esterification of tartaric acid with ethanol: kinetics and shifting the equilibrium by means of pervaporation. Chem. Eng. Sci. 49, 4681–4689. Kita, H., Tanaka, K., Okamoto, K., Yamamoto, M., 1987. The esterification of oleic acid with ethanol accompanied by membrane separation. Chem. Lett., 2053–2056. Kita, H., Sasaki, S., Tanaka, K., Okamoto, K.I., Yamamoto, M., 1988. Pervaporation– esterification coupling with ethanol accompanied by pervaporation. Chem. Lett., 2025–2028. Korkmaz, S., Salt, Y., Dincer, S., 2011. Esterification of acetic acid and isobutanol in a pervaporation membrane reactor using different membranes. Ind. Eng. Chem. Res. 50, 11657–11666. Kwon, S.J., Song, K.M., Hong, W.H., Rhee, J.S., 1995. Removal of water product from lipase-catalyzed esterification in organic solvent by pervaporation. Biotechnol. Bioeng. 46, 393–395. Lim, S., Park, B., Hung, F., Sahimi, M., Tsotsis, T., 2002. Design issues of pervaporation membrane reactors for esterification. Chem. Eng. Sci. 57, 4933–4946. Liu, K., Tong, Z., Liu, L., Feng, X., 2005a. Separation of organic compounds from water by pervaporation in the production of n-butyl acetate via esterification by reactive distillation. J. Memb. Sci. 256, 193–201. Liu, K., Feng, X., Tong, Z., Li, L., 2005b. Removal of trace water from organic mixtures by pervaporation separation during butyl acetate production via esterification. Separ. Sci. Technol. 40, 2021–2033. Liu, Q.L., Chen, H.F., 2001. Modeling of esterification of acetic acid with n-butanol in the presence of Zr(SO4)2 4H2O coupled pervaporation. J. Memb. Sci. 196, 171–178. Mulder, M., 1996. Basic Principles of Membrane Technologysecond ed. Kluwer Academic Publishers, Dordrecht. Nemec, D., Robert van, G., 2005. Performing esterification reactions by combining heterogeneous catalysis and pervaporation in a batch process. Ind. Eng. Chem. Res. 44, 9718–9726.
Pisarenko, Yu.A., 1979. Study on dynamics of reaction—rectification processes. PhD thesis, Lomonosov Moscow University of Fine Chemical Technology, Moscow. Pisarenko, Yu.A., Serafimov, L.A., 1978. Modeling of phase and chemical equilibrium for analysis and calculation of continuous hybrid reaction—rectification processes. Book of abstracts of IV national (USSR) conference on rectification. Ufa, 50. Polotskaya, G.A., Kuznetsov, Y.P., Agranova, S.A., 1994. Specific features of the formation of active surface layer in composite membranes. Vysokomolekulyarnye soedinenia seriya A & seriya B 36, B850–B851. Polotskaya, G.A., Gladchenko, S.V., Penkova, AV., Kuznetsov, V.M., Toikka, A.M., 2005. Synthesis of fullerene–polyphenylene oxide membranes for separating aqueous-organic mixtures. Russ. J. Appl. Chem. 78, 1468–1473. Polotskaya, G.A., Penkova, A.V., Toikka, A.M., Pientka, Z., Brozova, L., Bleha, M., 2007. Transport of small molecules through polyphenylene oxide membranes modified by fullerene. Separ. Sci. Technol. 42, 333–347. Prausnitz, J.M., Lichtenthaler, R.N., Azevedo, E.G., 1999. Molecular Thermodynamics of Fluid Phase Equilibria, third ed. Upper Saddle River, Prentice-Hall PTR, New York. Prigogine, I., Defay, R., 1954. Chemical Thermodynamics. Longmans and Green, London. Rewagad, R., Kiss, A., 2012. Modeling and simulation of a pervaporation process for fatty ester synthesis. Chem. Eng. Commun. 199, 1357–1374. Sanz, M., Gmehling, J., 2006. Esterification of acetic acid with isopropanol coupled with pervaporation. Part I: kinetics and pervaporation studies. Chem. Eng. J. 123, 1–8. Shah, M., Kiss, A., Zondervan, E., de Haan, A., 2012. A systematic framework for the feasibility and technical evaluation of reactive distillation processes. Chem. Eng. Process. 60, 55–64. Steinigeweg, S., Gmehling, J., 2004. Transesterification processes by combination of reactive distillation and pervaporation. Chem. Eng. Process. 43, 447–456. Sundmacher, K., Kienle, A., 2003. Reactive Distillation. Wiley-VCH, Weinheim. Suttichai, A., Jitkarun, P., Piyasan, P., Tomohiko, T., Shigeo, G., 2003. Theoretical study on the synthesis of methyl acetate from methanol and acetic acid in pervaporation membrane reactors: effect of continuous-flow modes. Chem. Eng. J. 95, 57–65. Toikka, A., Toikka, M., 2006. Phase transitions and azeotropic properties of acetic acid–n-propanol–water–n-propyl acetate system. Fluid Phase Equilibr. 250, 93–98. Toikka, A.M., Jetkins, J.D., 2002. Conditions of thermodynamic equilibrium and stability as a basis for the practical calculation of vapour–liquid equilibria. Chem. Eng. J. 89, 1–27. Toikka, A.M., Toikka, M.A., Pisarenko, Yu.A., Serafimov, L.A., 2009. Vapor–liquid equilibria in systems with esterification reaction. Theor. Found. Chem. Eng. 43, 129–142. Xuehui, L., Lefu, W., 2001. Kinetic model for an esterification process coupled by pervaporation. J. Memb. Sci. 186, 19–24. Zharov, V.T., 1970. Processes of open evaporation of solutions of substances with chemical interaction. Zhurnal Fizicheskoi Khimii (J. Phys. Chem. USSR) 44, 1967–1974. Zhu, Y., Minet, R.G, Tsotsis, T.T., 1996. A continuous pervaporation membrane reactor for the study of esterification reactions using a composite polymeric/ ceramic membrane. Chem. Eng. Sci. 51, 4103–4113.
Contents lists available at ScienceDirect
Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces
Separation of acetic acid–methanol–methyl acetate–water reactive mixture Anastasia V. Penkova a,n, Galina A. Polotskaya a,b, Alexander M. Toikka a a Department of Chemical Thermodynamics & Kinetics, Saint-Petersburg State University, Universitetskiy prospect 26, Peterhof, Saint-Petersburg 198504, Russia b Institute of Macromolecular Compounds RAS, Saint-Petersburg 199004, Russia
H I G H L I G H T S
The coupling of reaction and pervaporation. The reactive mixture with methyl acetate synthesis reaction. Pervaporation of mixtures which are close to chemical equilibrium. Removing of methyl acetate from reaction zone in pervaporation. Developed PPO/MFFC membranes are effective to shift equilibrium.
art ic l e i nf o
a b s t r a c t
Article history: Received 20 August 2012 Received in revised form 23 April 2013 Accepted 27 May 2013 Available online 10 June 2013
The aim of this research is to study the separation of a mixture with a methyl acetate synthesis reaction by pervaporation. The feed solutions that have compositions close to a chemical equilibrium are the main object of this work. The pervaporation of these solutions leads to the shifting of the composition from its chemical equilibrium. As a result, ester synthesis reaction coupled with pervaporation allows for more effective conversion of reagents. The hybrid process “reaction+pervaporation” was studied using membranes with a base of poly-(2,6-dimethyl-1,4-phenylene oxide) (PPO): a homogeneous PPO membrane and a composite membrane consisted of a PPO thin selective layer ( 5 mm) on the surface of a fluoroplastic composite hydrophobic membrane (MFFC). The transport properties of homogeneous and newly developed composite membranes based on the PPO were studied through the pervaporation of quaternary system acetic acid–methanol and water–methyl acetate. The sorption experiments were carried out to explain the pervaporation data. The possibility of using transformed composition variables for the analysis of pervaporation of mixtures in chemical equilibrium is demonstrated. It was shown that homogeneous and composite membranes are effective in shifting the esterification reaction equilibrium in pervaporation by the removal of methyl acetate or other esters from the reaction mixture. Composite membranes exhibited higher transport parameters (permeability and selectivity) than homogeneous membranes. & 2013 Elsevier Ltd. All rights reserved.
Keywords: Chemical equilibrium surface Membranes Pervaporation Reaction engineering Separations Transport processes
1. Introduction The study of separation processes in systems with chemical reactions could be important in the development of a theory of coupled reaction–mass-transfer processes. In modern chemical engineering, mass-transfer processes coupled with chemical reactions are used in the design of resource- and energy-saving technologies and should be considered a significant element of green chemistry. Coupled processes have numerous industrial
n
Corresponding author. Tel.: +7 812 428 48 05. E-mail address: [email protected] (A.V. Penkova).
0009-2509/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ces.2013.05.055
applications. One of the well-known examples is reactive distillation (Sundmacher and Kienle, 2003; Shah et al., 2012): Reactive distillation, which favorably shifts equilibrium through the removal of one of the products, is becoming more common in plant-scale production. It is, however, an energy-demanding operation and is not recommended when dealing with temperaturesensitive chemicals or biocatalysts (Lim et al., 2002). Reactive distillation can be combined with membrane separation (Aiouache and Goto, 2003; Liu et al., 2005a, 2005b; Steinigeweg and Gmehling, 2004). For example, the processes in reactive distillation–pervaporation hybrid columns for tert-amyl alcohol etherification with ethanol has been considered (Aiouache and Goto, 2003). Membrane separation can also be coupled with synthesis reaction. The combination of
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
a chemical reactor with a separate membrane pervaporation unit may lead to an effective coupled reaction–mass-transfer process. The processes in a pervaporation membrane reactor has been studied in various experimental works and theoretical research (Huang et al., 2004; Sanz and Gmehling, 2006; Domingues et al., 1999; Korkmaz et al., 2011; Rewagad and Kiss, 2012; Suttichai et al., 2003; Liu and Chen, 2001; Xuehui and Lefu, 2001; Assabumrungrat et al., 2003; Feng and Huang, 1996; Dos Reis Gonçalves et al., 2005; Liu et al., 2005a, 2005b; Delgado et al., 2009; David et al., 1991a, 1991b; Kita et al., 1988, 1987). Most of these studies concern the esterification reaction Alcohol ðR1 Þ þ Acid ðR2 Þ ¼ Water ðP 1 Þ þ Ester ðP 2 Þ: Esterification of carboxylic acids with alcohols is a typical example of an equilibrium-limited reaction. The conversion of reagents in such reactions depends on the thermodynamic constant of chemical equilibrium. Correspondingly, shifting the equilibrium leads to a higher conversion. Using excess alcohol as a cheap reagent is a practical method to achieve a high ester yield. On the other hand, the use of a large amount of reactant increases operation costs because it is necessary to separate the excess alcohol. The other way is to combine reaction and separation processes for the extraction of a main product or side-product. The esterification reaction could be driven by the extraction of products through reactive distillation. Another attractive method is the use of a pervaporation membrane reactor. The pervaporation is a ratecontrolled process and its efficiency in separation is not limited by related volatility as it would be in reactive distillation. In addition, the pervaporation separation could be carried out at a reaction temperature that is important for the technological design of the process. Finally, energy consumption in pervaporation is usually low (Mulder, 1996; Baker, 2004). Preliminary results on the use of a pervaporation membrane reactor as well as the description of a hybrid process of ester production have been presented in recent papers (David et al., 1991a, 1991b; Kita et al., 1988, 1987; Keurentjes et al., 1994; Kwon et al., 1995). The typical scheme of the hybrid process “reaction+pervaporation” (Feng and Huang, 1996) is shown in Fig. 1. Reagents (acid and alcohol) and a catalyst that is either homogeneous or heterogeneous are merged in a reactor (1). During the reaction, a quaternary equilibrium mixture is obtained. This mixture is transmitted to a membrane cell (2), where one of the reaction products is removed by pervaporation (3). The residual reaction mixture that does not pass though the membrane goes back to the reactor. In this way, excess acid and alcohol appears in the reactor and increases the conversion. The development of the new membranes which are effective for the application in the hybrid process “reaction+pervaporation” is also an important element of chemical engineering design.
587
Primarily, hydrophilic membranes based on polyvinyl alcohol and polyesterimide have been used for reactive pervaporation (Zhu et al., 1996; Nemec and Robert van, 2005), and this membrane type facilitated water removal from the reaction zone. In our previous work, organophilic membranes based on poly-(2,6dimethyl-1,4-phenylene oxide) (PPO) were studied through the pervaporation of a quaternary mixture containing components of ethyl acetate synthesis in equamolar content (Polotskaya et al., 2007, 2005). It established selective properties of PPO membrane with respect to ethyl acetate. Removal of the ester from the reaction medium by the membrane pervaporation can shift the esterification equilibrium and, at the same time, point to potentially important industrial solvents. In this presented paper, we consider the various aspects of reactive pervaporation in the system with methyl acetate reaction. The important element of the current study is the composition of the feed reactive solution. Most of the work on non-reactive and reactive pervaporation is focused on a binary feed mixture. On the other hand, it is evident that shifting the chemical equilibrium through reactive pervaporation should be studied for solutions close to chemical equilibrium compositions. Correspondingly, we consider the feed mixture compositions nearing an equilibrium state of methyl acetate synthesis. In this paper, we present some of our results of a systematic study of pervaporation of a reactive mixture for the case of compositions that are close to chemical equilibrium. The shifting from chemical equilibrium leads to a greater conversion of reagent, which is necessary for the development of resource-saving chemical technologies. The objects of this study were two membrane types based on PPO: a homogeneous PPO membrane and a composite membrane consisted of a thin PPO selective layer ( 5 mm) on the surface of hydrophobic fluorocarbon polymer porous support (MFFC). The dense film homogeneous membrane makes it possible to research transport properties of chosen selective polymer PPO. The composite membrane with a thin layer of selective polymer can ensure high-level permeability for the industrial application of a developed membrane process.
2. Material and methods 2.1. Materials Poly(2,6-dimethyl-1,4-phenylene oxide) (PPO) with a molecular weight of 172 000 and intrinsic viscosity of 1.58 dl/g (Brno, Czech Republic) was used for the work. Methyl acetate, methanol, and acetic acid were purchased from Vecton (Russia). The microfiltration fluoroplastic composite hydrophobic membrane MFFC (Vladipor, Russia) is composed of a porous fluoroplast F42L layer on a polypropylene support. The pore size is 0.05 mcm, with a total porosity of 80%. 2.2. Membrane preparation 2.2.1. Homogeneous membranes The homogeneous PPO membranes with a thickness of 60 mm were prepared by casting a 2 wt% PPO solution in chloroform on a cellophane surface and drying to remove the solvent. After that, the PPO film was separated from the cellophane and dried to a constant weight in a temperature set at 333 K.
Fig. 1. Scheme of hybrid process of ester production: (1) reactor, (2) membrane, and (3) permeate (Sanz and Gmehling, 2006).
2.2.2. Composite membranes The formation of a thin selective PPO layer in the composite membrane was created by casting 2 wt% PPO solution in chloroform on a surface of MFFC support and drying in vacuum at 313 K. To provide the formation of the selective layer at a thickness of
588
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
Fig. 2. Schematic illustration of pervaporation equipment: (1) feed tank, (2) thermostat, (3) membrane, (4) pressure controller, (5) vacuum pump, and (6) cold trap.
6–8 mm, the ∼0.03 ml of polymer solution was split on 1 cm2 of membrane surface. 2.3. Pervaporation experiment Pervaporation properties were measured by using laboratory equipment shown in Fig. 2. The membrane stainless steel cell (1) with an effective membrane area of 14.8 cm2 was provided with the stirrer. The membrane (3) was clamped into the cell and the feed was poured into the tank (2). Downstream pressure below 30 Pa was maintained using a vacuum pump (5) and controlled by a pressure-gauge (4). The permeate was collected in a liquid nitrogen trap (6), weighed, and then analyzed using gas chromatography. The gas chromatograph (Cristall 5000.2, Chromatec, Russia) was equipped with a Polysorb-1 column and a thermal conductivity detector to carry out quantitative analysis of the feed and the permeate composition. 2.4. Sorption experiment The sorption experiment was carried out by gravimetric method. Membrane films of known weight were immersed in a liquid at 293 K. After three days, they were removed and weighed after the superfluous liquid was wiped away. This procedure was repeated until the weight of the swelling films became constant. The degree of equilibrium swelling or sorption of liquid in the polymer membrane, S (g of liquid/100 g of dry polymer), was calculated by the equation S ¼ ðmS −m0 Þ=m0
ð1Þ
where ms is the weight of a swollen membrane and m0 is the weight of a dry membrane.
components of a liquid mixture through dipole-dipole, ion-dipole, or hydrogen-bonding (Polotskaya et al., 1994). Therefore, permeability and separation properties of membranes in pervaporation depend on interaction between feed components and the membrane polymer (Baker, 2004; Prausnitz et al., 1999). The permeability of liquid in pervaporation is significantly controlled by the solubility of a penetrant in the polymer membrane. To give an exhaustive characterization of the transport process, besides pervaporation tests where membrane permeability and selectivity could be measured, membrane sorption of individual components of liquid mixture should also be studied. Equilibrium sorption of individual components of reacting mixture methyl acetate, methanol, water, and acetic acid was determined for PPO and fluoroplast F42L that are materials of membranes under study. Table 1 lists the degree of equilibrium sorption of membrane polymers with liquid mixture components. It is significant that water is absolutely inert to membrane polymers because PPO and fluoroplast F42L are organophilic polymers. Results demonstrate that methyl acetate is sorbed better than other liquids by the polymers under study. Fluoroplast F42L exhibits the best sorption ability with respect to methyl acetate. Both membrane polymers moderately sorb methanol and acetic acid, though fluoroplastic F42L sorbs these liquids in lesser amounts. The data of Table 1 give some explanation of why membranes based on fluoroplast F42L cannot be used for separating this quaternary mixture alone. According to data of Table 1, the fluoroplast F42L is inert to water; sorption of methanol and acetic acid is small but the F42L swelling in methyl acetate is very high. The high value of methyl acetate sorption indicates that the membrane is on the frontier of solubility. Therefore, the fluoroplast F42L cannot be used as membrane material by itself.
3.2. Pervaporation properties of homogeneous and composite membranes 3.2.1. Pervaporation of model quaternary mixture Firstly to check membrane properties and selectivity, we studied pervaporation through membranes based on PPO for model quaternary mixture with equimolar contents of reagents and products of reaction: 16.7 mol% acetic acid, 16.7 mol% methanol, 33.3 mol% methyl acetate, 33.3 mol% water. Two membrane types based on PPO were used in the pervaporation of the quaternary mixture. They were a homogeneous membrane, 60 mm in thickness, and a composite membrane consisting of a thin PPO selective layer ( 5 mm) on the surface of a MFFC porous support. The composite membrane was prepared by taking into account the peculiarities of selective-layer formation established in the published work (Polotskaya et al., 2005; Toikka et al., 2009). The fulfillment of a condition for casting a PPO solution of 1≤[η] c≥8, where [η]—intrinsic viscosity (dl/g), c—polymer solution concentration (g/dl), assured the formation of nondefective membranes. In our case, [η] c¼ 1.58 2 ¼3.16 was used. Table 1 Degree of equilibrium sorption of membrane polymers with liquid mixture components. Solvent
3. Results and discussion 3.1. Sorption characterization of membrane polymers In a membrane process the polymer acts as “a molecular sieve” with sorption active centers that are able to interact with
Acetic acid Methanol Water Methyl acetate
Sorption, wt% PPO
Fluoroplast F42L
23.6 16.0 0 26.0
7.50 5.20 0 131
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
Transport properties of the PPO/MFFC membrane were compared with those of the PPO homogeneous membrane. It was established that the permeability of the composite membrane (0.33 kg/m2 h) was essentially higher than for the homogeneous membrane (0.11 kg/m2 h). Moreover, qualitative permeate composition depends on the type of used membrane. Fig. 3 illustrates how the composition of the quaternary mixture in the feed changes after pervaporation through PPO and PPO/MFFC membranes. The content of methanol and acetic acid in the permeate was smaller than that in the feed. Thus, reagents remained essentially in the reactor due to the use of membranes based in PPO. Acetic acid did not transfer practically through the membrane. The same low selectivity to acetic acid in pervaporation of quaternary mixtures has been noted (Liu and Chen, 2001) in hybrid processes with pervaporation at butyl acetate synthesis. Fig. 3 shows that reaction products penetrated through our membranes by different modes. For the case of the PPO membrane, the amount of water in the permeate was higher than that in the feed. Another trend observed was that the PPO/ MFFC membrane had an essentially lower permeability for water than the homogeneous membrane. This fact can be explained by the influence of the hydrophobic fluoroplastic material of the MFFC support that offers an additional resistance to water transfer and considerably influences the selectivity of the process. The main component of the permeate in both types of membranes is methyl acetate, though in the case of the composite membrane, the selectivity relative to methyl acetate was higher in comparison with the homogeneous membrane. The pervaporation data on the high selectivity of the PPO/MFFC membrane in methyl acetate removal from a quaternary mixture is coordinated with enhanced sorption properties of fluoroplast F42L with respect to methyl acetate.
323.15 K. The experimental data of the work (Pisarenko and Serafimov, 1978; Pisarenko, 1979) influenced the choice of feed compositions in our research: five compositions uniformly spaced at a chemical equilibrium surface in composition tetrahedron (or in the square of αi —variables). Due to the differences between the temperatures of pervaporation and experiments of works (Pisarenko and Serafimov, 1978; Pisarenko, 1979), feed compositions were not precisely in chemical equilibrium but belonged to the surroundings of chemical equilibrium surface. Tables 2 and 3 list the experimental results on the pervaporation of chemically equilibrium mixtures using homogeneous PPO and composite PPO/MFFC membranes; feed and permeate compositions, such as fluxes of the processes are presented here. It should be noted that all experiments were conducted without catalysts, such as acid or ion-exchange resins. This fact provides a constant feed composition in pervaporation experiments because the reaction rate is negligible at 293.15 K. Also, we did not observe, the shift of solution compositions by removing the permeate removing because the quantity of permeate was extremely small compared to the feed. Tables 2 and 3 show how the membrane flux depends on the feed composition. Flux through homogeneous and composite membranes rises with increasing methyl acetate content in the feed. This phenomena needs additional explanation. It could be explained on the basis of sorption experiments: methyl acetate is the most active agent in membrane swelling (Table 1).
Table 2 Pervaporation of quaternary equilibrium mixtures of different feed compositions by using homogeneous PPO membranes at 293.15 K. No.
3.2.2. Pervaporation of chemically equilibrium quaternary mixtures The data sets on chemical and vapor–liquid equilibrium in acetic acid–methanol–water–methyl acetate systems have been discussed in Toikka et al. (2009). Isothermal data on equilibrium that is closest in condition to our pervaporation experiment (293.15 K) are presented in Pisarenko and Serafimov (1978) and Pisarenko (1979) at 318.15 K and Bernatová et al. (2006) at
589
1 2 3 4 5
Feed composition, mol% Acetic acid
Methanol
Water
10.3 15.2 3.30 11.3 0.48
9.70 15.6 46.5 11.5 82.7
18.5 30.1 24.4 58.0 8.97
Methyl acetate 61.5 39.1 25.8 19.2 7.85
Permeate composition, mol% Flux, kg/m2 h Methanol Water Methyl acetate 8.40 14.1 39.1 9.40 79.4
26.3 34.1 27.2 60.3 10.4
65.3 51.8 33.7 30.3 10.2
Fig. 3. Composition of quaternary mixture in feed and in permeate after pervaporation through PPO and PPO/MFFC membranes, 293.15 K.
0.091 0.040 0.039 0.022 0.016
590
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
Table 3 Pervaporation of quaternary equilibrium mixtures of different feed compositions by using composite PPO/MFFC membranes at 293.15 K. No.
1 2 3 4 5
Feed composition, mol%
Flux, kg/m2 h
Permeate composition, mol%
Acid
Methanol
Water
Methyl acetate
Methanol
Water
Methyl acetate
10.3 15.2 3.30 11.3 0.48
9.70 15.6 46.5 11.5 82.7
18.5 30.1 24.4 58.0 8.97
61.5 39.1 25.8 19.2 7.85
7.50 11.7 40.6 9.20 77.8
14.8 30.6 18.5 46.7 18.1
77.7 57.7 40.9 44.1 4.10
Fig. 4. The dependence of fluxes on compositions (αi ) for the surface of chemical equilibrium for homogeneous (blue color) and composite (red color) membranes at 20 1C. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
Pervaporation of quaternary mixtures by using a PPO/MFFC membrane is more effective and productive than using a homogeneous PPO membrane in the process. In the case of the composite membrane, methanol and water content in the permeate decreases, yet it is important to note that the methyl acetate content in the permeate increases. The results of the pervaporation of quaternary mixtures in chemical equilibrium are in agreement with data of sorption tests (Table 1) and show the efficiency of both membrane types in separating products of esterification reaction. The use of pervaporation in hybrid processes of methyl acetate synthesis can facilitate a shift in equilibrium and enhance the conversion of esterification. Fig. 4 shows dependence of flux on composition of chemically equilibrium feed (in α—variable). Permeate of chemically equilibrium feeds (Tables 2 and 3) is not chemically equilibrium; therefore, a graphical illustration of the transformed composition variables cannot be presented. On the other hand, the pervaporation data sets show that permeate does not contain acetic acid for all feeds compositions. It gives us the opportunity to consider the permeate as a ternary mixture. Thus, permeate compositions can be represented by an ordinary concentration triangle, but a graphical illustration of the ratio of permeate content to feed content needs a three-dimensional figure (tetrahedron of compositions). The edge of the alcohol–ester–water of the composition tetrahedron corresponds with the permeate composition, and points within the tetrahedron correspond to the feed compositions. Fig. 5 shows the appropriate diagram of permeate content on feed content where lines connected points of feed and permeate compositions are analogs of tie-lines of liquid–vapor equilibrium.
1.58 0.17 0.22 0.16 0.09
Fig. 5. “Tie-lines” of pervaporation process in the acetic acid–methanol–water– methyl acetate system at 20 1C in the composition tetrahedron. Green squares: points of feed compositions; red squares: permeate of composite membrane; and blue squares: permeate of homogeneous membrane. Compositions are presented in molar percents. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
4. Conclusions In chemical syntheses, the conversion of reagents could be improved by removing the reaction products from the reaction zone. This process results in shifting the reaction equilibrium and correspondingly increasing the reaction rate. In the case of methyl acetate synthesis, we propose using the pervaporation separation to remove ester from the reactive mixture. In our study, for the pervaporation stage, we used membranes with a base of polyphenylene oxide (PPO). The study of the sorption and pervaporation properties of PPO membranes shows high sorption activity and selectivity of the membrane polymer to methyl acetate. To illustrate the efficiency of the coupling of synthesis with separation, the pervaporation study was carried out for the solution with a solution of compositions, which were close to chemical equilibrium. Homogeneous and composite membranes proved effective for the shift of esterification reaction equilibrium in pervaporation by the removal of methyl acetate or other esters from the reaction mixture. Composite membranes exhibit higher transport parameters (permeability and selectivity) than that of homogeneous membranes. The hybrid process “reaction+pervaporation” can be proposed as an alternative technique method of producing esters in industry. Increase of desired product yield, rational process realization in energy- and resource-saving regime due to membrane technology, such as high ecological purity, in the proposed process may prove promising in fundamental and industrial applications. In the
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
special case of methyl acetate synthesis, the PPO membranes could be used for the development of such technology.
Acknowledgment This work was supported by Grant of Ministry of Education and Science of Russian Federation nos. 2011-1.3.1-207-008-058 (GC 16.740.11.0658_02.06.2011) and by Grants of Russian Foundation for Basic Research nos. 12-03-00522 and 12-03-33155.
Appendix Theory
591
The consideration of physical–chemical properties with the use of three-dimensional diagrams could be accompanied by some graphical and other difficulties. For example, the dependence of any property on its composition should be presented formally as a hyper surface in 4D thermodynamic space. On the other hand, for the states of chemical equilibrium, the degree of freedom is reduced. As a result, the composition of phases in chemical equilibrium in the system with reaction R1 þ R2 ¼ P 1 þ P 2 could be presented in 2D space with the help of so-called transformed composition variables (αi ). These variables were first introduced by Zharov (1970) and later were proposed by Barbosa and Doherty (1987) (transformed composition coordinates). Due to the values of stoichiometric numbers of the reaction Eq. (2) (−1, −1, +1, +1), the composition complex for states in chemical equilibrium in such cases is a square of variables αi , which are connected with molar fractions xi by the following relationships: α 1 ¼ x1 þ x4
Compositions corresponding to chemical equilibrium of esterification reaction Compositions of quaternary mixtures are usually graphically presented in composition tetrahedron. The vertexes of tetrahedron correspond to pure substances, while edges and faces represent compositions of binary and ternary sub-systems. The equilibrium chemical reaction (i.e., chemical equilibrium) reduces the number of thermodynamic degrees of freedom. Accordingly, compositions in chemical equilibrium cannot belong to the whole compositional space of the tetrahedron. For the quaternary mixture with ester synthesis reaction, compositions in chemical equilibrium belong to the surface in tetrahedron (Toikka et al., 2009; Toikka and Jetkins, 2002). The position of a chemical equilibrium surface for the methyl acetate synthesis reaction
where νi and μi are stoichiometric number and chemical potential of specie i. In the state of chemical equilibrium affinity become zero value
Acetic acid ðR1 Þ þ Methanol ðR2 Þ ¼ Water ðP 1 Þ þ Methyl acetate ðP 2 Þ;
Α ¼ − ∑ νi μ i ¼ 0
α 2 ¼ x2 þ x4 α3 ¼ x3 −x4 α1 þ α2 þ α3 ¼ 1:
The thermodynamic condition of chemical equilibrium could be presented by means of affinity (A) (Prigogine and Defay, 1954): 4
Α ¼ − ∑ νi μ i ; i¼1
4
ð2Þ
is similar to other systems with esterification reaction (2) (Toikka et al., 2009; Toikka and Jetkins, 2002; Toikka and Toikka, 2006). According to the reaction Eq. (2), all ternary subsystems and two binary subsystems (acetic acid–methanol and water–methyl acetate) of quaternary acetic acid (R1)–methanol (R2)–water (P1)– methyl acetate (P2) system cannot be in chemical equilibrium. Therefore, a chemical equilibrium surface does not belong to the sides of the tetrahedron and edges R1 −R2 and P 1 −P 2 . The points of pure components (vertexes of tetrahedron) and non-reactive binary subsystems (edges R1 −P 1 , R1 −P 2 , R2 −P 1 , R2 −P 2 ) are considered as borders of the chemical equilibrium surface in compositional space. The quantitative view of chemical equilibrium surface is presented in Fig. 6.
ð3Þ
and
i¼1
ð4Þ
ð5Þ
Eq. (5) gives an opportunity to determine the position of a chemical equilibrium surface in compositional space (tetrahedron). It could be done with the use of experimental data on chemical potentials and dependence μi on composition. Another way for determining the chemical equilibrium is the calculation based on the thermodynamic models. The analytical presentation of a chemical equilibrium surface could be based on the calculation of a thermodynamic constant of equilibrium K e Ke ¼
a3 a4 ; a1 a2
where ai means the activity of component i. Unfortunately the errors in experimental determination of K e are considerable as a rule. Therefore, the determination of a chemical equilibrium is usually based on the direct experiment: reactions between species for various initial compositions of a reactive mixture. Of course, the correct presentation of a chemical equilibrium surface needs detailed data sets on chemical equilibrium. In the case of equilibrium of methyl acetate synthesis reactions, we used the experimental data (Pisarenko and Serafimov, 1978; Pisarenko, 1979). References
Fig. 6. Quantitative view of chemical equilibrium surface in the system with chemical reaction R1 þ R2 ¼ P 1 þ P 2 .
Aiouache, F., Goto, S., 2003. Reactive distillation–pervaporation hybrid column for tert-amyl alcohol etherification with ethanol. Chem. Eng. Sci. 58, 2465–2477. Assabumrungrat, S., Kiatkittipong, W., Praserthdam, P., Goto, S., 2003. Simulation of pervaporation membrane reactors for liquid phase synthesis of ethyl tert-butyl ether from tert-butyl alcohol and ethanol. Catal. Today 79-80, 249–257. Baker, R., 2004. Membrane Technology and Applicationsecond ed. Wiley, Menlo Park. Barbosa, D., Doherty, M.F., 1987. A new set of composition variables for the representation of reactive phase diagrams. Proc. R. Soc. London A 413, 459–464.
592
A.V. Penkova et al. / Chemical Engineering Science 101 (2013) 586–592
Bernatová, S., Aim, K., Wichterle, I., 2006. Isothermal vapour–liquid equilibrium with chemical reaction in the quaternary water+methanol+acetic acid+methyl acetate system, and in five binary subsystems. Fluid Phase Equilibr. 247, 96–101. David, M.O., Nguyen, T.Q., Neel, J., 1991a. Pervaporation– esterification coupling: Part I. Basic kinetic model. Chem. Eng. Res. Des. 69, 335–340. David, M.O., Nguyen, T.Q., Neel, J., 1991b. Pervaporation–esterification coupling: Part II. Modeling of the influence of different operating parameters. Chem. Eng. Res. Des. 69, 341–346. Delgado, P., Sanz, M.T., Beltrán, S., 2009. Pervaporation of the quaternary mixture present during the esterification of lactic acid with ethanol. J. Memb. Sci. 332, 113–120. Domingues, L., Recasens, F., Larrayoz, M., 1999. Studies of a pervaporation reactor: kinetics and equilibrium shift in benzyl alcohol acetylation. Chem. Eng. Sci. 54, 1461–1465. Dos Reis Gonçalves, F., Borges, L.E.P., Borges, C.P., 2005. Synthesis of ethyl acetate by coupling a heterogeneous catalytic system with a pervaporation unit. Separ. Sci. Technol. 39, 1485–1500. Feng, X., Huang, R.Y.M., 1996. Studies of a membrane reactor: esterification facilitated by pervaporation. Chem. Eng. Sci. 20, 4673–4679. Huang, Y.-S., Sundmacher, K., Qi, Z., Schlunder, E.-U., 2004. Residue curve maps of reactive membrane separation. Chem. Eng. Sci. 59, 2863–2879. Keurentjes, J.T.F., Janssen, G.H.R., Gorissen, J., 1994. The esterification of tartaric acid with ethanol: kinetics and shifting the equilibrium by means of pervaporation. Chem. Eng. Sci. 49, 4681–4689. Kita, H., Tanaka, K., Okamoto, K., Yamamoto, M., 1987. The esterification of oleic acid with ethanol accompanied by membrane separation. Chem. Lett., 2053–2056. Kita, H., Sasaki, S., Tanaka, K., Okamoto, K.I., Yamamoto, M., 1988. Pervaporation– esterification coupling with ethanol accompanied by pervaporation. Chem. Lett., 2025–2028. Korkmaz, S., Salt, Y., Dincer, S., 2011. Esterification of acetic acid and isobutanol in a pervaporation membrane reactor using different membranes. Ind. Eng. Chem. Res. 50, 11657–11666. Kwon, S.J., Song, K.M., Hong, W.H., Rhee, J.S., 1995. Removal of water product from lipase-catalyzed esterification in organic solvent by pervaporation. Biotechnol. Bioeng. 46, 393–395. Lim, S., Park, B., Hung, F., Sahimi, M., Tsotsis, T., 2002. Design issues of pervaporation membrane reactors for esterification. Chem. Eng. Sci. 57, 4933–4946. Liu, K., Tong, Z., Liu, L., Feng, X., 2005a. Separation of organic compounds from water by pervaporation in the production of n-butyl acetate via esterification by reactive distillation. J. Memb. Sci. 256, 193–201. Liu, K., Feng, X., Tong, Z., Li, L., 2005b. Removal of trace water from organic mixtures by pervaporation separation during butyl acetate production via esterification. Separ. Sci. Technol. 40, 2021–2033. Liu, Q.L., Chen, H.F., 2001. Modeling of esterification of acetic acid with n-butanol in the presence of Zr(SO4)2 4H2O coupled pervaporation. J. Memb. Sci. 196, 171–178. Mulder, M., 1996. Basic Principles of Membrane Technologysecond ed. Kluwer Academic Publishers, Dordrecht. Nemec, D., Robert van, G., 2005. Performing esterification reactions by combining heterogeneous catalysis and pervaporation in a batch process. Ind. Eng. Chem. Res. 44, 9718–9726.
Pisarenko, Yu.A., 1979. Study on dynamics of reaction—rectification processes. PhD thesis, Lomonosov Moscow University of Fine Chemical Technology, Moscow. Pisarenko, Yu.A., Serafimov, L.A., 1978. Modeling of phase and chemical equilibrium for analysis and calculation of continuous hybrid reaction—rectification processes. Book of abstracts of IV national (USSR) conference on rectification. Ufa, 50. Polotskaya, G.A., Kuznetsov, Y.P., Agranova, S.A., 1994. Specific features of the formation of active surface layer in composite membranes. Vysokomolekulyarnye soedinenia seriya A & seriya B 36, B850–B851. Polotskaya, G.A., Gladchenko, S.V., Penkova, AV., Kuznetsov, V.M., Toikka, A.M., 2005. Synthesis of fullerene–polyphenylene oxide membranes for separating aqueous-organic mixtures. Russ. J. Appl. Chem. 78, 1468–1473. Polotskaya, G.A., Penkova, A.V., Toikka, A.M., Pientka, Z., Brozova, L., Bleha, M., 2007. Transport of small molecules through polyphenylene oxide membranes modified by fullerene. Separ. Sci. Technol. 42, 333–347. Prausnitz, J.M., Lichtenthaler, R.N., Azevedo, E.G., 1999. Molecular Thermodynamics of Fluid Phase Equilibria, third ed. Upper Saddle River, Prentice-Hall PTR, New York. Prigogine, I., Defay, R., 1954. Chemical Thermodynamics. Longmans and Green, London. Rewagad, R., Kiss, A., 2012. Modeling and simulation of a pervaporation process for fatty ester synthesis. Chem. Eng. Commun. 199, 1357–1374. Sanz, M., Gmehling, J., 2006. Esterification of acetic acid with isopropanol coupled with pervaporation. Part I: kinetics and pervaporation studies. Chem. Eng. J. 123, 1–8. Shah, M., Kiss, A., Zondervan, E., de Haan, A., 2012. A systematic framework for the feasibility and technical evaluation of reactive distillation processes. Chem. Eng. Process. 60, 55–64. Steinigeweg, S., Gmehling, J., 2004. Transesterification processes by combination of reactive distillation and pervaporation. Chem. Eng. Process. 43, 447–456. Sundmacher, K., Kienle, A., 2003. Reactive Distillation. Wiley-VCH, Weinheim. Suttichai, A., Jitkarun, P., Piyasan, P., Tomohiko, T., Shigeo, G., 2003. Theoretical study on the synthesis of methyl acetate from methanol and acetic acid in pervaporation membrane reactors: effect of continuous-flow modes. Chem. Eng. J. 95, 57–65. Toikka, A., Toikka, M., 2006. Phase transitions and azeotropic properties of acetic acid–n-propanol–water–n-propyl acetate system. Fluid Phase Equilibr. 250, 93–98. Toikka, A.M., Jetkins, J.D., 2002. Conditions of thermodynamic equilibrium and stability as a basis for the practical calculation of vapour–liquid equilibria. Chem. Eng. J. 89, 1–27. Toikka, A.M., Toikka, M.A., Pisarenko, Yu.A., Serafimov, L.A., 2009. Vapor–liquid equilibria in systems with esterification reaction. Theor. Found. Chem. Eng. 43, 129–142. Xuehui, L., Lefu, W., 2001. Kinetic model for an esterification process coupled by pervaporation. J. Memb. Sci. 186, 19–24. Zharov, V.T., 1970. Processes of open evaporation of solutions of substances with chemical interaction. Zhurnal Fizicheskoi Khimii (J. Phys. Chem. USSR) 44, 1967–1974. Zhu, Y., Minet, R.G, Tsotsis, T.T., 1996. A continuous pervaporation membrane reactor for the study of esterification reactions using a composite polymeric/ ceramic membrane. Chem. Eng. Sci. 51, 4103–4113.