J. Chem. Thermodynamics 63 (2013) 31–37

Contents lists available at SciVerse ScienceDirect

J. Chem. Thermodynamics

journal homepage: www.elsevier .com/locate / jc t

Experimental density, viscosity, interfacial tension and water solubility of ethylbenzene-a-methyl benzyl alcohol–water system

Esayas W. Barega ⇑, Edwin Zondervan, André B. de HaanEindhoven University of Technology, Department of Chemistry and Chemical Engineering, P.O. Box 513, 5600MB Eindhoven, The Netherlands

a r t i c l e i n f o

Article history:Received 23 November 2012Received in revised form 9 March 2013Accepted 29 March 2013Available online 6 April 2013

Keywords:Physical propertiesEthyl benzene (EB)a-Methyl benzyl alcohol (MBA)Mass transferPhase separation

0021-9614/$ - see front matter � 2013 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.jct.2013.03.023

⇑ Corresponding author. Tel.: +31 402473187.E-mail address: e.w.barega@tue.nl (E.W. Barega).

a b s t r a c t

Density, viscosity, interfacial tension, and water solubility were measured for the (a-methyl benzylalcohol (MBA) + Ethyl benzene (EB)) system at different concentrations of MBA in contact with waterand sodium hydroxide solution (0.01 mol � kg�1) as aqueous phases. The properties were measured toidentify the component which plays a governing role in changing the physical properties relevant to masstransfer and phase separation of the ternary system. The concentration of MBA was found to be the majorfactor influencing all the properties. The water solubility, the density, and the viscosity increased notablyat higher concentrations of MBA; while, the interfacial tension decreased strongly. The use of0.01 mol � kg�1 NaOH as an aqueous phase resulted in a decrease of the interfacial tension and a minordecrease in the water solubility. The density data were correlated using a quadratic mixing rule todescribe the influence of concentration at any temperature. The viscosity data are correlated using theNissan and Grunberg and Katti-Chaudhri equations. The Szyzkowski’s equation was used to correlatethe interfacial tension data. The water solubility data were described using an exponential relationship.All the correlations described the experimental physical property data adequately.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Physical properties play a significant role in many separationprocesses. This work particularly focuses on their influence onliquid–liquid extraction that involves mass transfer and phase sep-aration steps. In this regard, the physical properties control thedrop size and its distribution [1–3], which are important for boththe mass transfer and phase separation steps. Several studies haveshown that physical properties determine the drop size by influ-encing the relative breakage and coalescence rates [1,4–7] andthe interfacial and viscous forces that resist the droplet breakup[8,9]. Moreover, once the drops are formed, the coalescencebetween the droplets and their separation depends on the physicalproperties [2,3,10–14]. The physical properties depend on the com-position of the liquid–liquid system and the temperature andmostly the concentration of one component in the system maybe critical for much of the observed change in physical properties.

This paper examines the influence of a-methyl benzyl alcohol(MBA) concentration and temperature on the physical propertiesof the ternary systems 1) (a-methyl benzyl alcohol (MBA) + Ethylbenzene (EB) + Water) and 2) (MBA + EB + NaOH (0.01 mol � kg�1)).The aim is to characterize the systems with respect to the differentphysical properties to identify the component which plays a vital

ll rights reserved.

role in changing the physical properties; hence, mass transferand phase separation. The ternary system used represents anindustrial propylene oxide-styrene monomer process [15] asshown in the reaction scheme given in figure 1. The main processstep of interest for our research is the caustic wash step, which isrepresented by reaction of phenols with NaOH as shown in thereaction scheme. This involves mass transfer and phase separationin a static mixer and settler, respectively. By measuring the phys-ical properties of the model system, the component that dictatesthe dispersion and phase separation behavior of the ternary systemcan be identified.

This work therefore presents the density, viscosity, interfacialtension and water solubility of the ternary (MBA + EB + Water) sys-tem. Additionally, interfacial tension and water solubility data aremeasured for the (MBA + EB + NaOH (0.01 mol � kg�1)) system. Toour knowledge, no such ternary data has been published. It isnoted however that some data exists on pure component viscosity,density [16–20], and interfacial tension [21,22] for the EB system.Therefore, the ternary data are measured at different concentra-tions of MBA and at different temperatures. Subsequently, themeasured data are described by correlations from literature. Thedensity data are described based on a quadratic mixing rule [23],the viscosity data using the Nissan and Grunberg equation[23,24] and the Katti-Chaudhri equation [25], the interfacial ten-sion data by Szyzkowski’s equation [26], and an exponentialfunction to describe the water solubility data.

1) EB+O2 → Ethylbenzene hydrogen peroxide (EBHP) 2) EBHP+Propylene (P) → MBA+propylene oxide(PO)+ phenols (PhOH) 3) PO+ PhOH +NaOH (pH=12) → Sodium Phenolate+Water (wash) 4) MBA-H2O → Styrene monomer

FIGURE 1. Reaction scheme of propylene oxide styrene monomer process.

32 E.W. Barega et al. / J. Chem. Thermodynamics 63 (2013) 31–37

2. Experimental

2.1. Chemicals

In the following, the purities are given in mass fraction. Ethylbenzene (purity > 0.99) was supplied by Merck (Germany), anhy-drous sodium hydroxide (NaOH, purity > 0.999) and a-methylbenzyl alcohol (purity > 0.99) were supplied by Sigma–Aldrich(USA). MilliQ super distilled water was used for all the measure-ments. The summary of provenance and purity is given in table 1.

2.2. Measurement techniques

2.2.1. Water solubilityThe equilibrium water solubility in the (MBA + EB) organic

phase was determined at temperatures (298.15 and 313.15) Kand at different concentrations between (0–80)% mole basis ofMBA in EB. 25 ml organic solution of (MBA + EB) at the requiredconcentration of MBA and 25 ml of water were introduced in ajacketed equilibrium glass vessel, which had an inner chamber of70 ml. The temperature in the double walled glass vessel was con-trolled using a thermostatic bath (Julabo F25) with an uncertaintyof 0.1 K. The organic and aqueous phases were agitated for half anhour at 500 rpm and subsequently allowed to settle for half anhour until clear phases were observed. To check whether the usedduration of time was sufficient to reach equilibrium, the mixingtime was increased to an hour and the water solubility was deter-mined again. No change in the water solubility was observed,which ensured that equilibrium was reached. A 2.5 ml samplewas taken from the organic top phase and the water content wasdetermined by using a Coulometric Karl-Fischer titration with aMetrohm 652 KF Coulorimeter (Applikon, The Netherlands). Thesame procedure was repeated three times and the water contentwas determined with a maximum uncertainty of 0.0005 in massfraction.

2.2.2. Interfacial tensionInterfacial tensions were determined for the ternary systems

(MBA + EB + Water) and (MBA + EB + NaOH solution (0.01 mol� kg�1)) at different concentrations of MBA (0–65)% mole basisand at temperatures (298.15 and 313.15) K using a Kruss 11 auto-matic tensiometer with a thermostated vessel (Wilten Physica, Bel-gium) temperature control. The thermostat vessel was connectedto a Julabo F25 heating/cooling bath (Julabo Labortechnik, Ger-many), and the temperature was controlled with an uncertaintyof 0.1 K. Interfacial tensions were determined by the Du Nouy ringmethod using a standard ring and corrected by the Harrison & Jor-dan method. The different concentrations of MBA in EB were pre-pared and agitated together with MilliQ water until equilibriumwas reached between the phases. This was done in the same man-

TABLE 1Provenance and purity of samples.

Chemical name Provenance Purity in massfraction

Purificationmethod

Anhydrous sodiumhydroxide

SigmaAldrich

>0.999 None

Ethyl benzene Merck >0.99 Nonea-methyl benzyl

alcoholSigmaAldrich

>0.99 None

ner as for the determination of the water solubility. After thephases settled, 10 ml samples of organic and aqueous phases weretaken from the top and bottom phases, respectively. The aqueousheavy phase was then placed in the thermostated vessel of the ten-siometer. The ring was placed above the surface of the heavyphase, and the surface of the heavy phase was detected at a speedof 10 mm/min. After detection, the ring was immersed to 3 mm inthe heavy phase at a speed of 0.2 mm/min. The heavy phase wasthen covered with the light phase and the interfacial tension wasmeasured at a speed of 0.2 mm/min and with a relaxation of10%. The measurement was continued until a standard deviationof the three consecutive measurements 6 0:3 mN/m was obtained.To calibrate the method, the interfacial tension measurement wasfirst compared with literature data of pure EB at 293.15 K and isshown in table 1. The comparison shows good agreement withthe literature data.

2.2.3. DensityThe density of the (MBA + EB + Water) system at different con-

centrations of MBA (0–100)% mole basis in the organic phase andas a function of temperature from (298.15 to 313.15) K was deter-mined by an automatic density meter DMA 5000 (Anton Paar, Aus-tria). Solutions of (MBA + EB) at the desired concentration of MBAwere prepared and injected into a density meter. The measuringcell in the densitometer was heated to the required temperatureof the sample and then the measurement was started. After thedensity measurement was completed, the measuring cell wasrinsed three times with acetone and water to clean it from theremaining chemicals from the used sample. The measurementwas repeated three times and the densities were measured withan accuracy of 0.5 kg/m3. The density measurement was first com-pared with existing literature data of pure EB at temperatures of(298.15 and 308.15) K and the values are listed in table 2. Compar-isons show good agreement with the deviations less than 0.02%.

2.2.4. ViscosityThe viscosity of the (MBA + EB + Water) system at the above

concentrations of MBA and temperatures from (298.15 to313.15) K was determined by capillary ubbelohde viscometer50103/0c, with a capillary constant of 0.003244 mm2/s2 �0:65%

(Schott, Germany). After samples of different concentrations ofMBA in EB were prepared at the desired temperature, the sampleswere poured into the viscosimeters. The capillaries filled with thesample were immersed in a Lauda water bath equipped with ther-mostat, which controls the temperature with an uncertainty of0.1 K. To measure the viscosity, the sample was sucked up in a cap-illary until it passes a top marker. Then the time required for themeniscus of the sample to descend from the top to the bottommarker was measured using a digital stopwatch with a resolutionof 0.01 s. The procedure was repeated three times. The kinematicviscosity was determined by multiplying the time with the capil-lary constant. Afterwards, the dynamic viscosity was calculatedby multiplying the kinematic viscosity with the density. The dy-namic viscosity was determined with the relative standard uncer-tainty of 0.003 mPa � s. Measurement was first compared withexisting literature data of pure EB at temperatures of (298.15 and308.15) K and the values are listed in table 2. Comparisons showgood agreement with literature data.

3. Results and discussion

3.1. Water solubility

The effect of the MBA concentration in the organic phase onwater solubility is shown in table 3. Generally, it can be ob-

TABLE 2Comparison of the measured pure component property values and literature data.

Chemical T/K q/kg �m�3 g/mPa � s c/mN �m�1

This work Literature This work Literature This work Literature

EB 298.15 862.51 862.64b 0.6378 0.6378b 34.7 36.5a

864.4c 0.631c

303.15 858.10 0.6090308.15 853.68 853.86b 0.5682 0.5688b

855.4c 0.569c

313.15 849.26 0.5410MBA 298.15 1009.00 8.703

303.15 1004.86 6.981308.15 1000.73 5.714313.15 996.58 4.748

Water 298.15 997.04 997.04d 0.891 0.890e

a Data measured at 293.15 K and reference [21,22].b Reference [17,18].c Reference [20].d Reference [31].e Reference [32].

TABLE 3Experimental water solubility, xw (mass fraction), at different mole fractions of MBAin the organic phase (X1, water free concentrations before water saturation) atT = (298.15 and 313.15) K, at P = 0.1 MPa and water/0.01 kg �mol�1 NaOH aqueousphases.f

X1 xw xw xw

T = 298.15 K water T = 313.15 K water T = 298.15K 0.01 mol � kg�1 NaOH

0.1014 0.0021 0.0024 0.00200.2025 0.0047 0.0054 0.00430.3334 0.0093 0.0103 0.00850.4037 0.0125 0.0140 0.01190.5039 0.0189 0.0199 0.01750.6535 0.0284 0.0309 0.02680.8025 0.0464 0.0487 0.0441

Note that the compositions presented in table 3 are not tie lines. What the tableshows is how the water solubility in the organic phase (xw), when a binary(MBA + EB) organic phase initially without water (as given by different concentra-tions of MBA in EB, X1) is contacted by either water or 0.01 mol/kg NaOH solution.f Standard uncertainties are u(T) = 0.1 K, u(x1) = 5E�5, u(xw) = 5E�3.

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

0.000

0.010

0.020

0.030

0.040

0.050

Xw

x1

FIGURE 2. Water solubility in mass fraction, xW, versus MBA mole fraction in theorganic phase, X1 (water free basis): j, T = 298.15 K and water aqueous phase;d, T = 313.15 K and water aqueous phase; N, T = 298.15 K and 0.01 mol � kg�1 NaOHaqueous phase; —, fitted values according to equation (1).

E.W. Barega et al. / J. Chem. Thermodynamics 63 (2013) 31–37 33

served that the water solubility increases considerably with anincrease in MBA concentration in the organic phase. This canbe attributed to the increased availability of MBA molecules forhydrogen bonding with water molecules as a result of their highconcentration in the organic phase, which moves the water tothe organic phase. The water solubility also increased at elevatedtemperatures. However, compared to the MBA concentration, theeffect of temperature on the water solubility was found to besmall.

The influence of MBA concentration on the water solubility canbe correlated by an exponential function shown in equation (1)where xw is the solubility of water in the organic (MBA + EB) phasein mass fraction, and x1 is the concentration of MBA in the organicphase in mole fraction (water free basis).

xw ¼ C1 expx1

C2

� �þ C3: ð1Þ

The fitting parameters C1 to C3 were determined by minimizingthe average of absolute relative error (AARE) using equation (2)where NDP represents the number of experimental data points.The determined parameters are given in table 3. As can be ob-served from figure 2, a good agreement was obtained betweenthe experimental and calculated values.

AARE=% ¼ 100NDP

P=fitted-experimental=

experimental: ð2Þ

3.2. Density

The variation in the density of the (MBA + EB + Water) systemwith an increase in the concentration of MBA in the organic phaseand temperature is depicted in table 4. It can be observed that thedensity of the organic (MBA + EB + Water) ternary system in-creases strongly at high MBA concentrations. This implies for a gi-ven density of the aqueous phase (MilliQ water) that the densitydifference between the phases becomes smaller at higher concen-trations of MBA. Additionally, an increase in temperature from(298.15 to 313.15) K resulted in a decrease in the density of themixture. (see table 5)

The composition dependence of binary and ternary densities isusually correlated by a Redlich–Kister polynomial expansion [27]at one temperature using excess molar volume data. In this work,the measured ternary system densities qm are correlated usingequation (3) based on a simple quadratic rule [23] at anytemperature.

TABLE 4Parameters of equations (1), (3), (4), (5), (6), (7), and AARE.

Equation (1)Parameters C1 C2 C3 AARE/%Values, water T = 298.15 K 0.00691 0.39314 �0.00685 1.48Values, water T = 313.15 K 0.00898 0.43166 �0.00896 0.79Values, 0.01 mol � kg�1 NaOH

T = 298.15 K0.00568 0.37034 �0.00546 1.87

Equation (3)Parameters Kij/-MBA (1) + EB (2) �0.0444MBA (1) + water (3) �0.0238EB (2) + water (3) 0.068AARE/% 0.056

Equations (4) and (5)Parameters Hij/K�1 Mij/- Wij/J �mol�1

Nissan GrunbergMBA (1) + EB (2) �0.0046 �0.0074MBA (1) + water (3) 0.0071 �0.0022EB (2) + water (3) 9.7E�5 �0.0016AARE/% 1.92Katti-ChaudhriMBA (1) + EB (2) �3537MBA (1) + water (3) 9151EB (2) + water (3) 1743AARE/% 1.91

Equation (7)Parameter Aq. phase water

T = 298.15 KAq. phase waterT = 303.15 K

Aq. phase 0.01 mol � kg�1 NaOHT = 298.15 K

Aq. phase 0.01 mol � kg�1 NaOHT = 313.15 K

ASZ/N �m�1 � K�1 0.1575 0.1484 0.1305 0.1299BSZ/mol � L�1 0.0459 0.0437 0.0261 0.0480AARE/% 3.13 1.11 2.40 1.40

TABLE 5Experimental density q/kg.m�3 of the ternary {MBA (1) + EB (2) + Water (3)} system(organic phase) at different mole fractions of MBA, P = 0.1 MPa andT = (298.15, 303.15,308.15, and 313.15) K.g

x1 x2 x3 T/K

298.15 303.15 308.15 313.15

0.0000 1.0000 0.0000 862.5 858.1 853.7 849.30.0000 0.9967 0.0033 863.3 858.9 854.5 849.70.1001 0.8874 0.0124 875.8 871.4 867.0 862.40.1969 0.7755 0.0276 890.4 886.0 881.5 877.10.3151 0.6299 0.0550 910.4 906.0 901.6 897.20.3742 0.5527 0.0731 919.9 915.6 911.2 907.00.4399 0.4513 0.1088 936.7 932.4 928.1 923.80.5496 0.2914 0.1591 958.1 953.9 949.6 945.40.6072 0.1495 0.2434 981.0 976.8 972.6 968.40.6557 0.0000 0.3443 1009.5 1005.4 1001.1 997.131.0000 0.0000 0.0000 1009.0 1004.9 1000.7 996.7

In the table, the data in italics represent the water saturated EB and MBA densities.However, pure component values are used in the correlation of density.g Standard uncertainties are u(T) = 0.1 K, u(x1) = 5E�5, u(q) = 0.5 kg �m�3.

0,0 0,2 0,4 0,6 0,8 1,0

840

860

880

900

920

940

960

980

1000

1020

ρ/kg

.m-3

x1

FIGURE 3. Density q of the ternary (MBA + EB + Water) system versus MBA molefraction in the organic phase, X1 (water free basis): j, T = 298.15 K; h, T = 303.15 K;N, T = 308.15 K; 4, T = 313.15 K; —, fitted values according to equation (3).

34 E.W. Barega et al. / J. Chem. Thermodynamics 63 (2013) 31–37

qm=kg �m�3 ¼X

i

xi � ðqi=kg �m�3Þ þX

i

xi

Xj

xjKij

� ððqiqjÞ0:5=kg �m�3Þ: ð3Þ

In equation, qm is the density of the ternary (EB + MBA + Water)system at any temperature, qi and qj are the densities of purecomponents MBA,EB, and water, xi and xj are the liquid mole frac-tions, and Kij ¼ Kji ðKii ¼ Kjj ¼ 0Þ is the binary fitting parameterthat accounts for deviation from ideality. The liquid mole fractionsof MBA, EB, and water in equation (3) are calculated from concen-trations of MBA and EB before water saturation and from the watersolubility. The parameters Kij are determined by minimizing theaverage absolute relative error (AARE). The parameters Kij and

AARE are given in table 4. As can be seen from the table and figure3, a good agreement was obtained between the experimental andcalculated density values.

3.3. Viscosity

The results of viscosity measurements are listed in table 6. Themixture viscosity increases considerably as the MBA concentrationin the organic phase increases and it decreases at higher tempera-tures. As in the case of water solubility and density, the concentra-tion of MBA was found to be the main factor influencing the

TABLE 6Experimental dynamic viscosity g/mPa � s of the ternary {MBA (1) + EB (2) + Water(3)} system (organic phase) at different mole fractions of MBA, P = 0.1 MPa, andT = (298.15, 303.15,308.15, and 313.15) K.h

x1 x2 x3 T/K

298.15 303.15 308.15 313.15

0.0000 1.0000 0.0000 0.638 0.609 0.568 0.5410.0000 0.9967 0.0033 0.614 0.581 0.549 0.5210.1001 0.8874 0.0124 0.723 0.678 0.636 0.5980.1969 0.7755 0.0276 0.847 0.782 0.727 0.6780.3151 0.6299 0.0550 1.195 1.088 0.998 0.9190.3742 0.5527 0.0731 1.335 1.213 1.099 1.0130.4399 0.4513 0.1088 1.726 1.544 1.396 1.2600.5496 0.2914 0.1591 2.565 2.246 1.976 1.7520.6072 0.1495 0.2434 4.073 3.462 2.970 2.5800.6557 0.0000 0.3443 7.561 6.122 5.052 4.2281.0000 0.0000 0.0000 8.703 6.981 5.714 4.748

h Standard uncertainties are u(T) = 0.1 K, u(x1) = 5E�5, and relative standarduncertainty ur(g) = 0.003 mPa � s. In the table, the data in italics represent the watersaturated EB and MBA dynamic viscosities. However, only pure componentdynamic viscosities are used in viscosity correlations.

E.W. Barega et al. / J. Chem. Thermodynamics 63 (2013) 31–37 35

measured viscosity values. Therefore, it can be expected that theMBA concentration plays a key role in the dispersion and separa-tion behavior of the ternary system. Additionally, the temperatureis also expected to play an important role as the density and viscos-ity values were notably influenced.

The ternary system dynamic viscosities are correlated using thecompositional models such as the Nissan and Grunberg equation

0.0 0.2 0.4 0.6 0.8 1.0-8

-6

-4

-2

0

2

4

6

8

100 (

1-(η

mod

el/η

exp)

x1

0.0 0.2 0.4 0.6 0.8 1.0-8

-6

-4

-2

0

2

4

6

8

100(

1-(η

mod

el/ η

exp)

x1

FIGURE 4. Relative deviations between the experimental and measured viscosities cT = 298.15 K (top left), T = 303.15 K (top right), T = 308.15 K (bottom left), and T = 313.15

[23,24] and Katti-Chaudhri equation [25] as shown in equations(4) and (5) respectively.

lnðgm=Pa � sÞ ¼X

i

xi lnðgi=Pa � sÞ þX

i

Xj

xixjGij; ð4Þ

lnððgm=Pa � sÞðvm=m3 �mol�1ÞÞ ¼X

i

xi lnððgi=Pa � sÞðv i=m3

�mol�1ÞÞ

þX

i

Xj

xixjðWij=J �mol�1ÞðR=J �mol�1 � K�1ÞðT=KÞ

:

ð5Þ

In these equations, gm represents the dynamic viscosity of theternary system at a given temperature, xi and xj the liquid molefractions, vm the mixture molar volume, v i and gi the pure compo-nent molar volumes and viscosities, respectively. The mole frac-tions in equations (4) and (5) are calculated in the same way asin density equation. The binary parameters Gij and Wij are adjust-able quantities accounting for intermolecular interactions. For theNissan and Grunberg equation, the interaction parameter Gij can becalculated by the group contribution method [23]. However, in thispaper, a linear temperature dependent equation (6) is used whereHij/K�1 is the temperature dependent coefficient and Mij/-is thetemperature independent coefficient. In the Katti-Chaudhri equa-tion, the temperature dependence is included in the equation;therefore, one parameter Wij/J �mol�1 is sufficient to describe theinteraction.

x1

0.0 0.2 0.4 0.6 0.8 1.0-8

-6

-4

-2

0

2

4

6

8

100(

1-(η

mod

el/η

exp)

0.0 0.2 0.4 0.6 0.8 1.0-8

-6

-4

-2

0

2

4

6

8

100(

1-(η

mod

el/η

exp)

x1

alculated by: h, Katti-Chaudhri equation; j, Nissan and Grunberg equation atK (bottom right).

0,0 0,2 0,4 0,6 0,8 1,00,0

1,0

2,0

3,0

4,0

5,0

6,0

7,0

8,0

9,0η /

mPa

.s

x1

FIGURE 5. Dynamic viscosity, g, versus mole fraction of MBA in the organic phase,X1 (water free basis), at different temperatures: j, T = 298.15 K; h, T = 303.15 K; N,T = 308.15 K and; 4, T = 313.15 K;—, calculated values using Katti-Chaudhriequation.

TABLE 7Experimental interfacial tension c/N �m�1 for the ternary {MBA (1) + EB (2) + Water(3)} organic system contacted with water / NaOH 0.01 mol � kg�1 solution aqueousphase, at different mole fractions of MBA in the organic phase, at P = 0.1 MPa andT = (298.15, and 313.15) K.i

x1 x2 x3 c/N �m�1 T = 298.15 K c/N �m�1 T = 313.15 K

Saturated with water0.0000 0.9967 0.0033 0.0346 0.03070.0503 0.9426 0.0071 0.0224 0.01970.1001 0.8874 0.0124 0.0167 0.01610.1969 0.7755 0.0276 0.0138 0.01340.3151 0.6299 0.0550 0.0117 0.01140.3742 0.5527 0.0731 0.0106 0.01040.4399 0.4513 0.1088 0.0098 0.0092

Saturated with 0.01 mol � kg�1 NaOH solution0.0000 0.9971 0.0029 0.0283 0.02240.0504 0.9431 0.0065 0.0177 0.01520.1002 0.8879 0.0119 0.0150 0.01430.1973 0.7771 0.0256 0.0118 0.01160.3166 0.6329 0.0504 0.0104 0.01020.3754 0.5545 0.0701 0.0098 0.00950.4436 0.4550 0.1014 0.0097 0.0091

i Standard uncertainties are u(T) = 0.1 K, u(x) = 5E�5, u(c) = 0.2 mN/m.

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.00.005

0.010

0.015

0.020

0.025

0.030

0.035γ/

N.m

-1

C/mol.L-1

FIGURE 6. Interfacial tension, c, versus molar concentration of MBA, C, at differenttemperatures and aqueous phases: j, T = 298.15 K and aqueous phase water; h,T = 313.15 K and aqueous phase water; N, T = 298.15 K and aqueous phase0.01 mol � kg�1 NaOH; 4, T = 313.15 K and aqueous phase 0.01 mol � kg�1 NaOH;—, calculated from equation (7).

36 E.W. Barega et al. / J. Chem. Thermodynamics 63 (2013) 31–37

Gij ¼ ðHij=K�1ÞT þMij: ð6Þ

The parameters are determined by minimizing the absolute va-lue of the relative error (AARE) and are shown in table 4. The devi-ations between the experimental and calculated dynamicviscosities using Nissan and Grunberg and Katti-Chaudhri equa-tions are presented in figure 4. It can be observed that the devia-tions are comparable for both equations. At lower temperaturesof (298.15 and 303.15) K, the Katti-Chaudhri equation gave lessdeviation and better description of the experimental data, whereas,at higher temperatures of (308.15 and 313.15) K, less deviationwas observed with the Nissan and Grunberg equation. Overall,both equations resulted in equivalent prediction and the sameAARE value. Nevertheless, using the Katti-Chaudhri equation hasan advantage since only three parameters are optimized to de-scribe the experimental data at all temperatures and concentra-tions, while in the Nissan and Grunberg equation six parametersshould be determined. The obtained binary interaction parametersfrom both equations are negative for (MBA + EB), implying a repul-sive interaction between these molecules. In contrast, positiveinteraction parameters are obtained for (MBA + Water), implyingattractive interaction between the molecules. The interactionparameter between (EB + Water) is positive but less than that of(MBA + Water) implying a weaker attraction between the mole-cules compared to (MBA + Water). The experimental dynamic vis-cosities and the calculated values from the Katti-Chaudhri aredepicted in figure 5. It can be seen that the experimental data agreewell with the calculated values.

3.4. Interfacial tension

The influence of MBA concentration in the organic phase andtemperature on the interfacial tension of the two phase system isdepicted in table 7 and figure 6. As can be seen from the figure,the interfacial tension decreased sharply at first and then levelsof at higher concentrations of MBA. The influence of temperatureon the interfacial tension is also shown in figure 6. Generally, aslightly lower interfacial tension was observed at higher tempera-ture; however, the effect of MBA concentration was morepronounced.

Therefore, from the obtained results, it can be inferred that thepresence of MBA assists in the formation of a dispersion (smalldrop sizes), as the interfacial tension of the system decreased nota-bly at higher concentration of MBA. Nevertheless, this decrease of

interfacial tension might worsen the phase separation due to theformation of small droplets.

The effect of the MBA concentration on the interfacial tension iscorrelated using the Szyzkowski correlation [26] as depicted inequation (7) where ASZ/N �m�1 � K�1 and BSZ/mol � L�1 are the Szyz-kowski adsorption coefficients, and ci;T , cO;T representing the inter-facial tension with MBA concentration Ci/mol � L�1and withoutMBA at temperature T/K, respectively.

ci;T=N �m�1 ¼ cO;T=N �m�1ð1� ðASZ=N �m�1

� K�1ÞÞ lnCi

BSZ=mol � L�1 þ 1

! !: ð7Þ

The Szyzkowski coefficients were determined at (298.15 and313.15) K based on the measured change of interfacial tension withMBA concentration and are given in table 4. The interfacial tensiondata was correlated well with Szyzkowski equation. As can be ob-served in figure 6, a good agreement was obtained between theexperimental and calculated values.

E.W. Barega et al. / J. Chem. Thermodynamics 63 (2013) 31–37 37

3.5. Effect of NaOH (0.01 mol � kg�1) solution

In addition to using water as an aqueous phase, the effect of a0.01 mol � kg�1 NaOH solution as an aqueous phase was studied.The influence on water solubility is shown in table 3. A slightlylower water content in the organic phase was observed by using0.01 mol � kg�1 NaOH solution instead of water. This minor de-crease in water content can be attributed to the fact that ions bindwater in their shell [28,29], which reduces the water activity andtherefore its tendency to solubilize in the organic phase.

The influence of 0.01 mol � kg�1 NaOH solution on interfacialtension was also measured. It can be observed from table 7 thatlower interfacial tension values were obtained when using0.01 mol � kg�1 NaOH. The decrease can be attributed to the depro-tonation of MBA by NaOH creating an ionic surfactant in similarmanner to the lowering of interfacial tension of acidic crude oilsystems by Alkali injection [30]. The observed change was morepronounced at lower concentrations of MBA and decreased at high-er concentrations. The water solubility and the interfacial tensiondata are correlated using equations (1) and (7), respectively. Thedetermined parameters and AARE are given in table 4. The equa-tions describe the experimental data well.

Using 0.01 mol � kg�1 NaOH solution as an aqueous phase canalso influence the density and dynamic viscosity values. However,this influence was considered to be not significant in view of thefact that the water solubility values showed a minor change whenusing 0.01 mol � kg�1 NaOH solution. Since the density and dy-namic viscosities of the ternary system (MBA + EB + Water) aredependent on the water solubility, a minor change in the water sol-ubility means a minor change in those properties. Hence, measure-ment of these properties was not performed.

4. Conclusions

The water solubility, density, viscosity, and the interfacial ten-sion of (EB + MBA + Water) system were measured to identify thecomponent that plays a major role in changing the measured prop-erties. The measured data showed that the concentration of MBAinfluences all the measured properties relevant to dispersion andphase separation strongly. All the measured physical propertieswere found to change clearly with the concentration of MBA. Thewater solubility, the density and the viscosity of the mixture(EB + MBA) were found to increase notably at higher concentra-tions of MBA. Additionally, the interfacial tension of the system de-creased considerably at higher concentration. By using0.01 mol � kg�1 NaOH as an aqueous phase, the interfacial tensiondecreased and slightly lower water solubility was obtained. More-

over, the properties were correlated and described adequatelyusing correlations from literature for density, viscosity and interfa-cial tension.

Acknowledgments

This is an ISPT (Dutch Institute Of Sustainable Process Technol-ogy) project. The authors would like to acknowledge the support ofWouter Hoek from Lyondell Chemie Nederland BV for providinginformation regarding the industrial process.

References

[1] W. Podgórska, J. Baldyga, Chem. Eng. Sci. 56 (2001) 741–746.[2] T. Frising, C. Noik, C. Dalmazzone, J. Disp. Sci. Technol. 27 (2006) 1035–1057.[3] A.D. Barnea, J. Mizrahi, Trans. Inst. Chem. Eng. 53 (1975) 83–90.[4] C. Coulaloglou, L.L. Tavlarides, AICHE J. 22 (1976) 289–297.[5] C. Tsouris, L.L. Tavlarides, AICHE J. 40 (1994) 395–406.[6] E.G. Chatzi, G.J. Boutris, C. Kiparissides, Ind. Eng. Chem. Res. 30 (1991) 536–

543.[7] V. Alopaeus, J. Koskinen, K.I. Keskinen, Chem. Eng. Sci. 54 (1999) 5887–5899.[8] G.W. Stevens, C.L. Teh, H.I.B. Malcolm, Kirk-Othomer Encyclopedia of Chemical

Technology (Liquid–liquid Extraction), John Wiley & Sons, Inc., 2007.[9] J.O. Hinze, AICHE J. 1 (1955) 289–295.

[10] G.B. Lawson, Chem. Proc. Eng. 48 (1967) 45–49.[11] W. Rommel, W. Meon, E. Blass, Sep. Sci. Technol. 27 (1992) 129–159.[12] I.B. Bazhlekov, A.K. Chesters, F.N. Van de Vosse, Int. J. Multiphase Flow 26

(2000) 445–466.[13] A.K. Chesters, Chem. Eng. Res. Des. 69A (1991) 259–269.[14] S. Abid, A.K. Chester, Int. J. Multiphase Flow 3 (1994) 613–629.[15] T.L. David, W.M. Castor, Ulmann’s Encyclopedia of Industrial Chemsitry

(Propylene Oxide), Wiley-VCH, 2004.[16] V.K. Rattan, S. Singh, B.P.S. Sethi, J. Chem. Eng. Data 49 (2004) 1074–1077.[17] R.P. Singh, C.P. Sinha, J.C. Das, P. Ghosh, J. Chem. Eng. Data 34 (1989) 335–338.[18] J.A. Riddick, W.B. Bunger, T.K. Sakano, Organic Solvents Physical Properties and

Methods of Purification, Wiley, New-York, 1986.[19] M. Domínguez-Pérez, C. Franjo, J. Pico, L. Segade, O. Cabeza, E. Jiménez, Int. J.

Thermophys. 30 (2009) 1197–1201.[20] M. Hasan, A.B. Sawant, R.B. Sawant, P.G. Loke, J. Chem. Thermodyn. 43 (2011)

1389–1394.[21] J. Timmermans, Physicochemical Constants of Binary Systems in Concentrated

Solutions, Interscience, New York, 1960.[22] E. Cho, J. Park, Chemosphere 33 (1996) 1241–1246.[23] B.E. Poling, J.M. Prausnitz, J.P. O’Connell, The Properties of Gases and Liquids,

McGraw-Hill International Editions, Singapore, 2007.[24] T.S. Nakagawa, J. Mol. Liquids 63 (1995) 303–316.[25] P.K. Katti, M.M. Chaudhri, J. Chem. Eng. Data 9 (1964) 442–443.[26] K. Prochaska, Adv. Coll. Int. Sci. 95 (2002) 51–72.[27] O. Redlich, A.T. Kister, Ind. Eng. Chem. 40 (1948) 345–348.[28] R.W. Gurney, Ionic Processes in Solution, McGraw-Hill, Singapore, 1953.[29] J. Mähler, I. Persson, Inorg. Chem. 51 (2012) 425–438.[30] A. Samanta, K. Ojha, A. Mandal, Energy Fuels 25 (2011) 1642–1649.[31] G.I. Egorov, D.M. Makarov, J. Chem. Thermodyn. 43 (2011) 430–441.[32] B. Gonzàles, N. Calvar, E. Gómez, Á. Domínguez, J. Chem. Thermodyn. 39 (2007)

1578–1588.

JCT 12-672