INFLUENCE OF FLUID PROPERTIES AND SOLID SURFACE ENERGY ON EFFICIENCY OF BED COALESCENCE

Transcription

1 Available on line at Association of the Chemical Engineers of Serbia AChE Chemical Industry & Chemical Engineering Quarterly Chem. Ind. Chem. Eng. Q. 24 (3) (2018) CI&CEQ DUNJA S. SOKOLOVIĆ 1 DRAGAN D. GOVEDARICA 2 RADMILA M. ŠEĆEROV SOKOLOVIĆ 2 1 Faculty of Technical Sciences, University of Novi Sad, Novi Sad, Serbia 2 Faculty of Technology, University of Novi Sad, Novi Sad, Serbia REVIEW PAPER UDC 544.7:66.08:62 INFLUENCE OF FLUID PROPERTIES AND SOLID SURFACE ENERGY ON EFFICIENCY OF BED COALESCENCE Article Highlights Current understanding of influence of both liquids properties on bed coalescence are given Overview of filter media surface phenomena on emulsion separation efficiency by fiber bed included Effects of oil viscosity, density, molecular weight, emulsivity are presented Effects of solid surface roughness and wettability on separation efficiency are elaborated Abstract Emulsion separation is important in industry due to economic, safety, and ecological reasons. It can be applied in liquid-liquid extraction, effluent treatment, heat exchange, and fuel and chemical purification. In case of both oil-in-water and water-in-oil emulsions, regardless of their quantity and phase concentration, bed coalescence is a good and economical solution for separation. Due to the complexity of the bed coalescence phenomenon, the coalescer design relies on the base of the experimental test. The design strategy of a coalescer to separate oils of different quality in time is additionally complicated. This paper presents a literature review on the current understanding of the influence of properties of both liquids and surface phenomena of filter media on emulsion separation efficiency using steady-state bed coalescence. The influence of oil viscosity, interfacial tension, density, molecular weight, emulsivity and dielectric constant of mineral oil is presented. The effect of solid surface roughness and wettability on separation efficiency is also elaborated. Keywords: emulsion separation, oil properties, bed coalescers, surface roughness, wettability. The separation of two immiscible liquids is one of the most challenging problems in conventional practice. Water and oil have the ability to form emulsions. When oil is the continuous phase and water is the dispersed phase, a water in oil (W/O) emulsion is obtained. When water is the continuous phase and oil the dispersed phase, an oil in water (O/W) emulsion is obtained. Beside the mentioned emulsion types, Correspondence: D. Sokolović, Faculty of Technical Science, University of Novi Sad, Trg Dositeja Obradovića 6, Novi Sad, Serbia. dunjaso@uns.ac.rs Paper received: 4 March, 2017 Paper revised: 1 September, 2017 Paper accepted: 15 September, double emulsions can also be formed (W/O/W, or O/W/О) [1]. In the system of two immiscible liquids, the continuous phase can be either water or organic fluid. In terms of the complexity of the phenomenon, it seems that the system with the continuous aqueous phase is more complex. This phase could consist of plenty of different ions that affect the formation of the electrical double layer causing different behaviours of the droplets that are surrounded by these ions. Dispersed organic liquids in water can also have a wide range of different properties. Even if present in small concentrations, investigations indicate that the nature of this phase significantly influences the droplet coalescence. The organic dispersed phase can be a pure liquid or a complex mixture. Recently, many researchers conducted experiments on pure chemicals, 221

2 even though such systems are relatively rare in the industry. In addition to that, some investigations included complex mixtures such as crude oil, petroleum products, vegetable oil, fatty acid etc. In the study of Golob et al., it can be found that there is a drastic difference in the behaviour observed for pure chemicals and petroleum products [2]. The petroleum products as a complex multi-component mixture provoke lower coalescence efficiency under the same conditions, compared to the pure chemicals. It seems as if the coalescence of their droplets is more difficult. Investigations have focused on the effects of several properties of the dispersed phase. The most common being: viscosity, interfacial tension and density. Some authors have analyzed the effect of changes in the oil phase using graphical analyses [3-5], mathematical models, and dimensionless groups or the combination of all [6-9]. The most common error that occurs when studying the influence of the variation of dispersed phase properties is the narrow range of chosen variables over which investigations were realized [10]. EFFECT OF VISCOSITY ON STEADY-STATE BED COALESCENCE The effect of the viscosity of both phases, continuous and dispersed, on droplet coalescence were in focus in some investigations. Vinson [11] used cyclohexane and n-octane as the dispersed phase. He investigated the effect of viscosity variation of the continuous phase by increasing the viscosity of water by the addition of glycerol. This research revealed that the increase in viscosity of the continuous phase decreases the coalescence efficiency, which is expected, since this phenomenon reduces the droplet mobility leading to less frequent droplet interaction disabling the coalescence. In addition to this effect, the drainage time of film between two droplets increases with increased viscosity of the continuous phase, also reducing coalescence efficiency. Sareen et al. [3] varied the viscosity of the dispersed phase in the range from 1.38 to mpa s, at the bed length of 1.9 cm. The authors used several dispersed phases: base oil, kerosene, benzene, butyl benzene, chloroform, cyclohexane, isobutane and nitrobenzene. The authors concluded that the viscosity has no effect on the critical velocity when the values of fluid velocity were above 45 m/h and smaller than 2.38 m/h. In these conditions, the critical velocity approached an asymptotic value. A similar relationship was also found by other authors [10,12]. Critical velocity is defined in literature as the fluid velocity over which effluent concentration of the dispersed phase drastically increases. Šećerov Sokolović et. al. [4] identified a critical velocity as the velocity over which effluent concentration of the dispersed phase reached a value of 15 mg/l. Jeffreys et al. [13] found that with the increase of viscosity of the dispersed phase, the critical velocity decreases. They used iso-octane, amyl alcohol and kerosene as the dispersed phase. The form of the dependence only apparently differs from that of the other researchers. The reason for this is the narrow interval for the investigated viscosity, from 7 to 42 mpa s. These obtained results are in agreement with the part of the curve presented in the studies of Golob et al. [2] and Šećerov Sokolović et al. [10] in which the critical velocity decreases exponentially. It should be emphasized that, based on the results from Golob et al. [2], it could be claimed that a similar trend of dependence exists for the oil phase viscosity on critical velocity, no matter whether the oil is the continuous or the dispersed phase, for the observed range of viscosity at chosen working conditions and for the selected fiber properties, Figure 1. Ryan and Elimelech [14] investigated the detachment mechanism of spherical colloid particles Figure 1. Dependence of critical velocity on viscosity of dispersed phase for: a) oil in water emulsion; b) water in oil emulsion. 222

3 from the solid surface. They defined the torque moment that influences the detached particles including the adhesive force, drag force, hydrodynamic force, and the force of buoyancy. Combining all the equations for the mentioned moment of given forces, they formulated the expression of the critical velocity that includes the viscosity of the emulsion. From this relationship when the emulsion viscosity increases, the critical velocity also increases. Šećerov Sokolović et al. investigated in detail the influence of the dispersed oil phase properties on steady-state bed coalescence [1,10,15-19]. Their research on the effect of viscosity of the dispersed phase on the coalescence is partly in agreement with Jeffreys et al. [13] and partly is specific. Šećerov Sokolović et al. also claimed that the coalescence efficiency decreases with the viscosity increase for a wide range of viscosity from 7.00 to mpa s [10]. The mentioned tests were carried out on polyurethane fibers for one and the maximum bed permeability. Their further studies suggest that the flow mode does not affect the form of dependence of the dispersed oil properties [16]. In contrast to this bed geometry, fluid velocity and the nature of fiber media significantly influence the form of the dependence of coalescence efficiency on fluid properties[15-19]. It must be noted that there are regions of velocity and bed permeability in which the efficiency significantly increases with the increase of viscosity of the dispersed phase for some polymer materials. Simultaneous effects of viscosity and bed geometry, expressed through permeability, can be observed in the 3D diagram. The dependence of critical velocity on the two mentioned variables are presented in Figure 2, for polyurethane fiber bed, PU, Figure 2a, and polyethylene terephthalate fiber bed, BA1, Figure 2b. It can be observed that the dependence for both materials are significantly different. For PU fiber bed, the critical velocity increases with the viscosity increase for the full range of bed permeability. The intensity of the increase amplified with the permeability increase. The most favorable region for work is for fluid with high viscosity and for high bed permeability. This dependence may be explained by the following: maximum bed permeability relates to maximum porosity. In these circumstances, the amount of the capillaryconducted phase in the bed is maximal, and the interstitial fluid velocity is minimal compared to other tested bed permeability. In these conditions, the coalescence of the droplets that enter the bed effectively occurs in the volume of capillary-conducted phase. By bed compressing, permeability proportionally reduces, simultaneously decreasing the amount of the capillary-conducted phase in it, leading to the reduction of the critical velocity [19]. Why is this not the same in the case of BA1 fiber bed? With different fiber nature, the important adhesion force changes leading to the different effect of the bed geometry. For BA1 with viscosity increase, the critical velocity also increases, but with less intensity, and changes differently in different regions of permeability. It can be observed that for the given polymer, the region of mean permeability achieves lowest critical velocity. This phenomenon can be explained by the dominant mechanisms of droplet coalescence in the pores of the bed [16,19]. At high bed permeability, as already pointed out, the coalescence of droplets that enter the bed dominantly occurs in capillary-conducted phase, while at low permeability the favored coalescence is the coalescence between two droplets, because in these circumstances they are close to each other. At medium permeability, there is no dominant mechanism of droplet coalescence and for some materials, this can be reflected in the decrease of separation efficiency. This area is a transition from one to the other dominant mechanism. Based on this example, it could be concluded that the influence of the fluid properties cannot be studied separately Figure 2. 3D dependence of critical velocity on dispersed phase viscosity and bed permeability: a) PU and b) BA1. 223

4 from the nature of the filter material. The effect of filter material nature is even more evident when using the contour diagram, Figure 3. The contour diagram establishes lines on its surface that have equal value of critical velocity, i.e., iso-critical velocity as a function of the dispersed phase viscosity and bed permeability, Figure 3. Comparing the materials, PU and PE, PU application shows higher values of the critical velocity achieved for a wide range of viscosity of the dispersed oil and a broad range of bed permeability. EFFECT OF DENSITY ON STEADY-STATE COALESCENCE Many authors examined the importance of density of the dispersed phase. Some of them claim that the droplet coalescence and separation are more efficient in systems with greater difference in density between phases [1,13-20]. This is somewhat expected for the droplet settling behind the bed, because the settling velocity is proportional to the density difference of fluids in an emulsion. Jeffreys [13] argues that the increase in density differences can result in droplet deformation, causing the decrease in the drainage area of the film that accelerates coalescence. Simultaneously the hydrostatic force required for the film drainage increased, which can be insufficient. The above effects are conflicting, and the resultant impact of density differences may slow down or accelerate the coalescence. Which of these two influences will prevail is still not clarified. Based on the results of the research by Golob et al. for both type of emulsions, critical velocity at defined intervals of density achieves multiple local minima and maxima, Figure 4 [20]. The mentioned phenomenon is consistent with the observations obtained by Jeffreys [13]. The dual effect of the dispersed phase density on coalescence was also determined by Šećerov Sokolović et al., Figure 5 [17,19]. They explained this phenomenon by the following: for the H flow mode, when the droplets are significantly smaller than the pore size, all the known Figure 3. Contour diagrams representing the interdependence of isocritical velocity, dispersed oil viscosity, and bed permeability: a) PU; b) PE. Figure 4. Dependence of critical velocity on density of dispersed phase for: a) oil in water emulsion; b) water in oil emulsion. 224

5 capture mechanisms play a role [16]. If the dispersed phase density decreases, the deposition mechanism in the pore volume is more frequent. But, at the same time, in case of a steady-state regime, the coalescence of droplets occurs in capillary-conducted phase. Dispersed phase density influences the amount of capillary-conducted phase, as well as its distribution in the cross-section of the bed. With the density increase the quantity of capillary-conducted phase in the volume of the pores also increases, which is a desired effect for coalescence efficiency. The resultant of these two opposing effects influences the effect of the dispersed phase density on coalescence efficiency, Figure 5. does not change the efficiency, Figure 6. Vinson [10] did not find a clear dependence of the coalescence efficiency on interfacial tension for systems whose interfacial tension is below 30 mn/m, but detected that for the interval from 33 to 53 mn/m the coalescence efficiency increases with the increase of interfacial tension. Figure 6. Dependence of separation efficiency on interfacial tension of dispersed phase. Figure 5. Dependence of separation efficiency on fluid velocity and density of dispersed phase. EFFECT OF INTERFACIAL TENSION ON STEADY-STATE BED COALESCENCE A significant number of researchers have investigated the effect of interfacial tension on droplet coalescence. Two approaches in the mentioned research could be distinguished: the investigation of the droplet coalescence in a real system, which exhibits different values of interfacial tension due to the nature of both phases, and the investigation of a certain range of interfacial tension achieved by the addition of surfactants in the system. Both of these approaches are of interest and justified in terms of practice. In crude oil production, due to the addition of a package of additives the interfacial properties of oil and water system are altered. Most of the research using the first approach showed that with the increase of interfacial tension the coalescence efficiency also increases, obtaining either the maximum value or approaching a value above which no changes were detected [1,10,11,18]. Šećerov Sokolović et al. determined that by increasing interfacial tension of crude oil and its derived products from 11 to 21 mn/m the coalescence efficiency increases, while a further increase Jeffreys and Davis [13] claimed that interfacial tension has dual effects on coalescence efficiency and coalescence rate. By increasing the interfacial tension, the droplets are more resistant to deformation. In this way, the contact area between two droplets reduced, and the drainage of the film accelerated. Simultaneously with the increase of the interfacial tension the rate of film drainage can decrease, which has the opposite effect on coalescence. Sareen et al. [3] varied interfacial tension by adding surfactants and concluded that the behaviour of coalescence was different depending on the nature of the applied surface-active materials. Hazlett and Carhart studied water in oil emulsions, and the impact of interfacial tension changes were achieved by the addition of surfactants [21-23]. They determined that the droplets that are detached from the bed at lower interfacial tension are smaller. Golob et al. [6] and Clayfield et al. [24] investigated the effect of interfacial tension on coalescence efficiency. Golob et al. [6] varied the interfacial tension for the water-in-oil emulsion in the range of 10 to 35 mn/m by the addition of sarcosyl-o surfactant. They concluded that there is no bed length, for the given interval, which would allow effective emulsion separation at interfacial tension below the value of 10 mn/m. For all the investigated bed lengths with the increase of interfacial tension, the critical velocity increases. 225

6 Govedarica et al. experimentally determined interdependence of critical velocity, interfacial tension and bed permeability [25]. Two regions of interdependencies were confirmed in this research: in the first region, the critical velocity was independent on interfacial tension, and in the second region the decrease of the critical velocity was observed. Authors applied the contour diagrams, Figure 7, and PCA statistical method [1,18,25]. It is important to point out that investigation of interfacial tension needs to include the effects of bed geometry and the nature of filter materials. The diagrams presented in Figure 7 indicate that the nature of the filter materials has simultaneous effect with interfacial tension on separation efficiency. The PU material is less sensitive to change of interfacial tension and achieves higher critical velocity for the entire investigated range of bed permeability. EFFECT OF MOLECULAR WEIGHT ON STEADY-STATE BED COALESCENCE Šećerov Sokolović et al. were the first and only research group who investigated the effects of molecular weight, neutralization number and structural composition of the dispersed mineral oil on coalescence efficiency [10]. Their following research introduced in the analysis other variables that give information about the polarity of oils such as dielectric constant, emulsivity, and asphaltene content [16-18,25-27]. According to their research, the influence of molecular weight of the dispersed phase on coalescence is equivalent to the impact of its density. For mineral oils, there is a close connection between these two properties. With the increase of molecular weight of oil, density also increases. This can be concluded when comparing dependences of separation efficiency and density, Figure 5, and molecular weight, Figure 8, for a wide range of operating velocity. Figure 8. Dependence of separation efficiency on fluid velocity and molecular weight of dispersed phase. As we pointed out, Govedarica et al. [25] applied the PCA, and PCR analyses to develop an empirical equation for the dependence of the critical velocity as a function of dispersed oil properties. Oil viscosity, interfacial tension, emulsivity, and dielectric constant were included in the equation. The critical velocity from this equation increases with the increase of viscosity, emulsivity, and dielectric constant, and decreases with the increase of interfacial tension. Emulsivity is an oil property related to water and shows the water amount left behind after the mixing and separation process. Šećerov Sokolović et al. established the selection procedure for filter media taking emulsivity as a most important dispersed oil property and using contour diagrams for four investigated polymers [16]. Figure 7. Contour diagrams representing the interdependence of isocritical velocity, dispersed oil interfacial tension, and bed permeability for bed material: a) PU; b) PE. 226

7 EFFECT OF SOLID SURFACE ENERGY ON STEADY-STATE COALESCENCE Introduction Better understanding of the mechanisms affecting the droplet attachment substantially helps in the selection of more efficient filter media for a wide range of industrial applications. Adhesion is difficult to quantify at a fundamental level. Hydration forces, hydrophobic interactions, macromolecular bridging, surface roughness, electrical double-layer, and van der Waals forces have all been proposed as possible influences on adhesion [1,27]. The wetting of the surfaces involves both surface chemistry and geometry. Geometry can be either local, in the form of rough or patterned surfaces, or it can be global, in the form of spheres, cylinders/fibers. The difference in the increase of surface energy due to surface roughness or due to its chemical nature must be distinguished. Fiberglass, stainless steel, and ceramic are fibers with high surface energy. These fibers have lower coalescence efficiency of oil droplets than the fibers with low surface energy, such as polymer fibres. However, the fibres of glass, stainless steel, and ceramics are more successful in the coalescence of water droplets dispersed in the continuous organic phase. Increasing the surface energy by modification of the surface roughness affects the coalescence efficiency due to elevation of the adhesion force [1, 27]. Wettability is one of the most important surface property of filter media for liquid-liquid separation, which affects the removal efficiency of droplets in filtration. Therefore, study on filter media wettability greatly helps in the evaluation, selection and development of more efficient bed materials for droplet separation from the continuous phase. Currently, the wettability of solid material surface is generally evaluated in terms of the contact angle formed by liquid on solid surface as a static wettability or a dynamic one. Solid surface roughness In addition to the fiber diameter, many researchers also investigated the influence of fiber roughness. Sareen et al. [3] explained that the obtained better results for cotton fibres as compared to other materials are due to their higher fiber roughness. The authors concluded that with the increase in surface roughness the probability of capturing the droplets of the dispersed phase also increases. Jeffreys and Davis [13] also noted that the surface roughness significantly affects coalescence efficiency because it results in the important increase of the adhesion force. Magiera and Blass [28] investigated the separation possibilities of the dispersed oil using beds of glass fiber, stainless steel fiber and Teflon. They argue that roughness does not determine the separation efficiency. Although the Teflon had higher fiber roughness than steel and glass fibers, the separation efficiency of this material was the lowest. However, in these circumstances, it should be noted that the nature of the fiber is altered in the same time, and roughness is not the crucial factor for separation efficiency. Therefore, in this case, assumptions about the influence of surface roughness cannot be discussed. Hoek et al. [29] analysed the influence of the polymeric membranes roughness on the interaction with colloid particles. They investigated polyamide composite membranes, and developed an appropriate statistical model using DLVO (Derjaguin, Landau, Verwey and Overbeek) theory. The investigation showed that the energy barrier of the interaction of rough surface membrane with colloidal particles is much lower than when interacting with a smooth surface. Thereby, the capture of the colloidal particles on the rough polymeric membranes is facilitated. Hubbe emphasized that the detachment of the droplets from a solid surface significantly depends on the surface roughness [30-32]. According to Hubbe, there are two approaches to modelling such systems. The first approach assumes that the reservoir surface is smooth and perfectly flat. The second one takes into account the surface roughness, which significantly affects the phenomena related to the attractive and repulsive forces between particles and collectors. The roughness is not uniform across the surface of the real collectors and therefore, the contact between the droplet and the collector is not equal at each point. The higher the surface roughness, the more complex and harder to describe the contact mathematically. Bansal-Agarwal et al. published a valuable study on five fiber substrates with wide variation in their structure and properties. They applied nanocoating and nanoparticles for analysing wetting phenomena and oil droplet coalescence, mobility, and emulsion separation performance [33,34]. The purpose of using nano-coating was to alter fiber surface energy and to utilize the benefit of surface roughness. They concluded that the surface energy and roughness play a crucial role in droplet coalescence. In such thought-out experiments, the investigators were able to control the surface energy by altering the roughness or the chemical nature of the surface while maintaining the same geometry of the bed and the same nature of the fibers. In this way, they overcame the shortcomings of previous published studies. 227

8 Wettability of solid surface Several researchers have expressed different opinions about the effect of surface wettability on coalescence efficiency. Some of them reported that the dispersed phase should readily wet the fiber surface, while others reported it should not, yet some stated that the wetting behaviour of the dispersed phase is unimportant and/or intermediate wetting conditions are required [13,29-34]. The critical surface tension is one of the properties present in the literature, and provides information on the capability of solids surface wetting with a variety of fluids. The critical surface tension is determined from the Zisman diagram [35]. The diagram is obtained by measuring the contact angle, the wetting angle of the solid surface by fluids in a wide range of surface tension and establishing a correlation of the cosines angle and the surface tension of the fluid. The mentioned measurements are related to the measurement on a flat surface. The critical surface tension depends on the molecular groups on the surface of a solid material and their packing density. The surface of polyethylene (-CH 2 group) has a critical surface tension of 31 mn/m, while the replacement of hydrogen atoms with fluorine atoms (polytetrafluoroethylene) lowers the critical surface tension to 19 mn/m. If the hydrogen atoms are replaced with chlorine atoms (polyvinyl chloride), the critical surface tension increases to 41 mn/m. The critical surface tension for stainless steel is 108 mn/m and for copper 1000 mn/m [36]. Šećerov Sokolović et al. examined the coalescence efficiency of naphthenic mineral oil droplets using fiber beds of different polymer materials with equal bed length and bed geometry. The coalescence monitored by the critical velocity. They concluded that the critical velocity increases with the increase of critical surface tension of the used polymers. The range of values of the critical velocity was from 43 to 65 m/h [37]. Basu studied the wetting effect of polar organic compounds on the mechanism of coalescence and concluded that this could be very useful in practice. These polar compounds are capable of forming both inter- and intra-molecular hydrogen bonds. Because of this complexity, the coalescence mechanism may be different for different systems and needs to be investigated experimentally also taking into consideration that there is little theoretical knowledge available on hydrophilic and hydrophobic interactions [38]. However, most of the well-established measurement techniques for contact angle are suitable only for flat surfaces with relatively large areas. According to the Laplace equation, the capillary pressure across the liquid-gas interface is directly proportional to the surface tension of the liquid and the mean curvature of the interface. Because the contact angle acts as a boundary condition it has a strong influence on the shape of the liquid surface and hence on the capillary pressure. Thus, a measurement of capillary pressure coupled with a tractable model of the porous media can be used to estimate the contact angle [39-43]. A group of researchers claimed that some hydrophobic surfaces, at the same time typically oleophobic, provide the means for studying phobicity. On the base of their investigations, fibers having no wettability makes them an attractive option for use in industry as mist eliminators [44]. In contrast to this, Wang et al. discovered a superwetting phenomenon in nature. They classified possible extreme states of surface wettability in air, underwater, and underoil [45]. Chase with his researchers, over a long period, studied the phenomenon of wettability of fibers with high and low energy, as well as their mixtures. Having a great privilege to produce their own fibers for investigation, they were able to obtain very important results [46-57]. The wettability of such porous media is characterized by imbibition rates, capillary pressure, permeability and saturation relationships, relative permeability curves, and displacement capillary pressure. Fractional wettability exists when the surfaces and pores of the porous media consist of different chemical properties, minerals and contaminants leading to non-uniformity and wettability variation throughout the internal pores. The established modified form of the Washburn equation quantifies the wettability of filter media in terms of lipophilic to hydrophilic (L/H) ratio. The L/H ratio depends on the porosity of the medium, pore sizes, surface roughness, fiber size, fiber surface properties, and the liquid properties. The authors have established the existence of a correlation between the coalescence efficiency and the L/H ratio [52]. Krasinski investigated the separation of oil droplets from water using sandwiched multilayer coalescence media [58]. Fiber mat in the form of flat panels was manufactured with meltblowing technique. Separation efficiency of three polymers were investigated: polypropylene, polybutylene terephthalate, and polyamide. The capillary rise tests were carried out to characterize the wettability of the selected polymers in form of L/H ratio. They concluded that the effect of fiber wettability of the coalescence layer located on the inlet is very important. Strongly oleophilic polypropylene filter media achieved the highest separation efficiency. 228

9 Li et al. [59] investigated the simultaneous effect of filter media wettability and pore size on separation efficiency of oil droplet from water. When the surface is more wettable to the oil, the effect of pore size is less important. In another study, they coated filter media with a fluorochemical to separate emulsified water from diesel fuel in the presence of a surfactantmonoolein, and found that the coating aided coalescence significantly [60]. CONCLUSION When discussing the influence of the nature of the dispersed phase, especially oil droplets on coalescence separation, even though extensive research has been carried out on this subject, an immense dilemma remains. What is the key property of the dispersed phase that exclusively influences the coalescence? Does such a single property even exist? It was found that different dispersed liquid properties lead to different conclusions, even though the continuous phase of liquid under investigation was kept the same. Maybe for certain liquids a set of properties exist that will indicate in advance the tendency to coalesce. It is still unclear if there is a crucial surface property of a filter media that delimits coalescence efficiency. Contrary to the wettability, a new path of investigation has opened concerning studies on phobicity. The exploring of surfaces that are both, at the same time, hydrophobic and oleophobic has begun and proven to be efficient in the separation of droplets in some circumstances. A question arises if this new view is the beginning of better understanding of these complex and still unexplored phenomena. Acknowledgment The work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, Grant no REFERENCES [1] D. Govedarica, D. Sokolović, Emulsion separation with fiber bed, Faculty of Technical Sciences, University of Novi Sad, Serbia, 2014, p. 1 [2] L. Golob, V. Grilc, R. Modic, Chem. Eng. Res. Des. 64 (1986) [3] S.S. Sareen, P.M. Rose, R.C. Gudesen, R.C. Kintner, AICHE J. 12 (1966) [4] R.M. Šećerov Sokolović, S.M. Sokolović, B.D. Đoković, Ind. Eng. Chem. Res. 36 (1997) [5] J. Golob, R. Modic, Trans. Inst. Chem. Eng. 55 (1977) [6] J. Golob, V. Grilc, R. Modic, Separation of Secondary Droplets on Fibrous Beds, in Proc. of Advances in Separation Science, Trieste, Italy, September, 1978, p [7] H. Soot, C.J. Radke, Ind. Eng. Chem. Fundam. 23 (1984) [8] H. Soo, C.J. Radke, J. Colloid Interface Sci. 102 (1984) [9] H. Soo, C.J. Radke, Chem. Eng. Sci. 41 (1986) [10] R.M. Šećerov Sokolović, S.M. Sokolović, S.M. Šević, S.N. Mihajlović, Sep. Sci. Technol. (Philadelphia, PA, U. S.) 31 (1996) [11] C.G. Vinson, The Coalescence of Micron-Size Drops in Liquid-Liquid Dispersions in Flow Past Fine-Mesh Screen, Ph.D. Thesis, University of Michigan, Ann Arbor, MI, 1965 [12] D.F. Sherony, R.C. Kintner, Can. J. Chem. Eng. 49 (1971) [13] G.V. Jeffreys, G.A. Davies, Coalescence of Liquid Droplets and Liquid Dispersion, in Recent Advances in Liquid- Liquid Extraction, C. Hanson (Ed.), Pergamon press, Oxford, 1971, pp [14] J.N. Ryan, M. Elimelech, Colloids Surfaces, A 107 (1996) 1-56 [15] R.M. Šećerov Sokolović, T.J. Vulić, S.M. Sokolović, Sep. Purif. Technol. 56 (2007) [16] R.M. Šećerov Sokolović, D.D. Govedarica, D.S. Sokolović, Ind. Eng. Chem. Res. 53 (2014) [17] R.M. Šećerov Sokolović, D.D. Govedarica, D.S. Sokolović, J. Hazard. Mater. 175 (2010) [18] D.D. Govedarica, R.M. Šećerov Sokolovic, D.S. Sokolović, S.M. Sokolović, Sep. Purif. Technol. 104 (2013) [19] R.M. Šećerov Sokolović, D.S. Sokolović, D.D. Govedarica, Hem. Ind. 70 (2016) [20] V. Grilc, L. Golob, R. Modic, Chem. Eng. Res. Des. 62 (1984) [21] R.N. Hazlett, Ind. Eng. Chem. Fundam. 8 (1969) [22] R.N. Hazlett, Ind. Eng. Chem. Fundam. 8 (1969) [23] R.N. Hazlett, H.W. Carhart, Filtr. Sep. 9 (1972) , 462 [24] E.J. Clayfield, A.G. Dixon, A.W. Foulds, R.J.L. Miller, J. Colloid Interface Sci. 104 (1985) [25] D.D. Govedarica, R.M. Šećerov Sokolović, D.S. Sokolović, S.M. Sokolović, Ind. Eng. Chem. Res. 51 (2012) [26] D.S. Sokolović, R.M. Šećerov Sokolović, S.M. Sokolović, Hem. Ind. 67 (2013), pp [27] R.M. Šećerov-Sokolović, S.M. Sokolović, Hem. Ind. 58 (2004) [28] R. Magiera, E. Blass, Filtr. Sep. 34 (1997) [29] E.M.V. Hoek, S. Bhattacharjee, M. Elimelech, Langmuir 19 (2003) [30] M.A. Hubbe, Colloids Surfaces 12 (1984) [31] M.A. Hubbe, Colloids Surfaces 16 (1985) [32] M. A. Hubbe, Colloids Surf. 16 (1985)

10 [33] S. Bansal, V. von Arnim, T. Stegmaier, H. Planck, J. Hazard. Mater. 190 (2011) [34] S. Agarwal, V. Von Arnim, T. Stegmaier, H. Planck, A. Agarwal, Sep. Purif. Technol. 107 (2013) [35] K. Kabza, J.E. Gestwicki, J.L. McGrath, J. Chem. Educ. 77 (2000) [36] H. Sharifi, J.M. Shaw, Chem. Eng. Sci. 51 (1996) [37] R.M. Šećerov Sokolović, S.M. Sokolović, Ind. Eng. Chem. Res. 43 (2004) [38] S. Basu, J. Colloid Interface Sci. 159 (1993) [39] N. Stevens, J. Ralston, R. Sedev, J. Colloid Interface Sci. 337 (2009) [40] C. Qing, W. Bigui, H. Yingdong, Capillary Pressure Method for Measuring Lipophilic Hydrophilic Ratio of Filter Media, Chem. Eng. J. (Amsterdam, Neth.) 150 (2009) [41] B.J. Mullins, R.D. Braddock, Int. J. Heat Mass Transfer 55 (2012) [42] B.J. Mullins, R.D. Braddock, G. Kasper, Chem. Eng. Sci. 62 (2007) [43] B.W. Yang, Q. Chang, Sep. Purif. Technol. 60 (2008) [44] R. Mead-Hunter, T. Bergen, T. Becker, R.A. O Leary, G. Kasper, B.J. Mullins, Langmuir 28 (2012) [45] S. Wang, K. Liu, X. Yao, L. Jiang, Chem. Rev. 115 (2015) [46] C. Shin, G.G. Chase, J. Colloid Interface Sci. 272 (2004) [47] K. Moorthy, Effects of Surface Energy of Fibers on Coalescence Filtration, The University of Akron, 2007 [48] S.U. Patel, G.G. Chase, Sep. Purif. Technol. 75 (2010) [49] S. Dawar, G.G. Chase, Sep. Purif. Technol. 72 (2010) [50] G. Viswanadam, G.G. Chase, J. Colloid Interface Sci. 367 (2012) [51] P.S. Kulkarni, Mixed Hydrophilic/Hydrophobic Fibre Media for Liquid-Liquid Coalescence, Ph.D. Thesis, University of Akron, 2011 [52] P.S. Kulkarni, S.U. Patel, G.G. Chase, Sep. Purif. Technol. 85 (2012) [53] S.U. Patel, P.S. Kulkarni, S.U. Patel, G.G. Chase, Sep. Purif. Technol. 87 (2012) [54] S.U. Patel, P.S. Kulkarni, S.U. Patel, G.G. Chase, Drying Technol. 31 (2013) [55] S.U. Patel, P.S. Kulkarni, S.U. Patel, G.G. Chase, Sep. Purif. Technol. 120 (2013) [56] G. Viswanadam, G.G. Chase, Sep. Purif. Technol. 104 (2013) [57] P.S. Kulkarni, S.U. Patel, S.U. Patel, G.G. Chase, Sep. Purif. Technol. 124 (2014) 1-8 [58] A. Krasinski, Environ. Prot. Eng. 42 (2016) [59] X. Li, D. Hu, L. Cao, C. Yang, RSC Adv. 5 (2015) [60] Y. Li, L. Cao, D. Hu, C. Yang, Sep. Purif. Technol. 176 (2017) DUNJA S. SOKOLOVIĆ 1 DRAGAN D. GOVEDARICA 2 RADMILA M. ŠEĆEROV SOKOLOVIĆ 2 1 Fakultet tehničkih nauka, Univerzitet u Novom Sad, Trg Dositeja Obradovića 6, Novi Sad, Srbija 2 Tehnološki fakultet, Univerzitet u Novom Sadu, Bulevar cara Lazara 1, Novi Sad, Srbija PREGLEDNI RAD UTICAJ OSOBINA FLUIDA I POVRŠINSKE ENERGIJE FILTARSKOG MATERIJALA NA EFIKASNOST KOALESCENCIJE U SLOJU Separacija emulzija je značajna u industriji iz ekonomskih, bezbednosnih i ekoloških razloga. Može se primeniti kod ekstrakcije tečno-tečno, za tretman otpadnih voda, kod razmenjivača toplote, ili za dehidrataciju goriva i hemikalija. U slučaju emulzije ulja u vodi, kao i u slučaju emulzije vode u ulju, bez obzira na količinu emulzije i koncentraciju ulja, koalescentna filtracija je dobro i ekonomično rešenje za separaciju. Zbog složenosti fenomena za projektovanje kolescera neophodna su poluindustrijska ispitivanja. Projektovanje kolescera koji služe za separaciju ulja različite prirode je posebno složeno. Ovaj rad predstavlja pregled literature o trenutnom razumevanju uticaja osobina oba fluida (ulja i vode) i površinskih pojava na granici faza na efikasnost separacije emulzije primenom koalescencije u sloju pri stacionarnom režimu rada. Prikazan je uticaj viskoznosti ulja, međupovršinskg napona, molske mase, emulzivnosti i dialektrične konstante mineralnih ulja koja su ispitivana. Efekat hrapavosti čvrstih površina i njihove kvašljivost na efikasnost separacije takođe je elaboriran. Ključne reči: separacija emulzija, osobine ulja, kolesceri, hrapavost površine, kvašljivost. 230