researcharticle coupling solvent extraction units to...

Research ArticleCoupling Solvent Extraction Units to Cyclic Adsorption Units

Mariana Busto 1 Enrique Eduardo Tarifa 2 and Carlos Romaacuten Vera 1

1 Institute of Research on Catalysis and Petrochemistry INCAPE FIQ-UNL CONICET Collecting Ring National Road 168 Km 0El Pozo 3000 Santa Fe Argentina2Faculty of Chemical Engineering Universidad Nacional de Jujuy CONICET Italo Palanca No 10 San Salvador de Jujuy Argentina

Correspondence should be addressed to Carlos Roman Vera cverafiqunleduar

Received 8 November 2017 Revised 19 January 2018 Accepted 24 January 2018 Published 11 March 2018

Academic Editor Jose C Merchuk

Copyright copy 2018 Mariana Busto et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The possibility of regenerating the solvent of extraction units by cyclic adsorptionwas analyzedThis combination seems convenientwhen extraction is performed with a high solvent-to-impurity ratio making other choices of solvent regeneration typicallydistillation unattractive To our knowledge the proposed regeneration scheme has not been considered before in the openliterature Basic relations were developed for continuous and discontinuous extractionadsorption combinations One exampledeacidification of plant oil with alcohol was studied in detail using separate experiments for measuring process parameters andsimulation for predicting performance at different conditions An activated carbon adsorbent was regenerated by thermal swingmaking cyclic operation possibleWhen extracting the acidwithmethanol in a spray column feed = 4 Lminminus1 solvent = 80 Lminminus1feed impurity level 140mmol Lminus1 and extract concentration 76mmol Lminus1 the raffinate reaches a purity of 12mmol Lminus1 the solventbeing regenerated cyclically in the adsorber (364 kg) to an average of 07mmol Lminus1 Regeneration of the solvent by cyclic adsorptionhad a low heat duty Values of 174 kJ per litre of solvent compared well with the high values for vaporization of the whole extractphase (1011 kJ Lminus1)

1 Introduction

Liquid-liquid extraction (LLE) is an important technique ofseparation used in many applications of the chemical processindustry Distillation the workhorse of separation processesis based on boiling point differences LLE instead is basedon different relative solubilities of solutes in two immiscibleor partially miscible liquids Extraction is typically used incases in which distillation is not cost-effective or directly notpossible at all This is the case when azeotropes are formedor when volatility differences between components are toolow or when heat-sensitive materials are present that coulddecompose at the high temperatures of distillation Also ifthe component to be recovered has a very high boiling pointor is present in very small concentrations distillation is notcost-effective

One of the most important steps in the design of LLE isthe choice of the solvent which must meet several criteria inorder to achieve amaximum transfer rate (i) a high solubilityfor the solute and low solubility for the feedraffinate (ii) a

density difference with the carrier higher than 015 g cmminus3(iii) a medium surface tension (5ndash30 dyne cmminus1) (iv) highresistance to thermal degradationwhen thermal regenerationis used (v) a high boiling point and low viscosity for easeof handling It is readily apparent that not all criteria can bemet and that a careful screening is needed to choose the bestsolvent from a given set

One aspect not always conveniently stressed in LLE isthat of solvent regeneration This must be easy and energy-efficient As a consequence when the solvent is being chosenit must be decided how is to be regenerated Since mostsolvents are regenerated by distillation aspects to be analyzedare selectivity solute distribution and volatility Solvents thatdisplay high selectivity usually have low solute distributioncoefficients If they also have lower volatility than the impu-rity the impurity can be recovered as a distillate Howeverif the solvent has a lower boiling point than the impuritythen the solvent should be distilled off for regenerationIf the impurity distribution coefficient is low then a highsolvent-to-feed ratio is needed High solvent recycle rates and

HindawiInternational Journal of Chemical EngineeringVolume 2018 Article ID 1620218 17 pageshttpsdoiorg10115520181620218

2 International Journal of Chemical Engineering

high solvent regeneration rates are hence needed For thesesystems separation by distillation might not be an optionand other principle could be chosen An option that has notreceived attention in the scientific literature or the industrialpractice is that of solvent regeneration by adsorption

The possibilities are analyzed in this work of a processusing liquid-liquid extraction for removing impurities from afeed and adsorption for solvent regenerating the solventThiscombination has not been previously discussed in the openliterature Main features from the kinetic thermodynamicand process point of view are considered and discussed andthemain parameters for designing such a process are writtenEquations are revised for batch and continuous units involv-ing local and global interphase mass transfer coefficients andthe range of practical values of these parameters for these twooperations is discussed

One example involving experimental work is used asproof-of-concept deacidification of vegetable oils by extrac-tion with alcohol coupled to the cyclic adsorption of car-boxylic acids from alcoholic solutions in an activated carbonpacked column The liquid-liquid extraction (LLE) of freefatty acids (FFA) from some plant oils has received someattention lately because fatty acids can be recovered easily forfurther processing the yield of neutral oil being maximum[1ndash3] These are advantages of LLE over caustic refiningin which oil is lost because of reaction with the caustic(saponification) and by emulsification FFAs being convertedto a difficult to handle soap-stock [4] It has also advantagesover physical refining (removal of acids by distillation at250∘C) that produces undesirable changes in color and canalso produce the degradation of valuable nutraceuticals andantioxidants [5ndash7]

Applications of an extractionadsorption process forrefining plant oils could be useful in the biodiesel industrythe food industry and the industry of biodegradable technicaloils (lubricants dielectric transformer oil etc)

2 Materials and Methods

The general procedure was as follows (i) liquid-liquidthermodynamic equilibrium data was got for the solvent-feed-impurity system in the form of partition coefficients(ii) solid-liquid thermodynamic equilibrium isotherms wereobtained for the solvent-adsorbent-impurity system (iii)kinetic parameters for extraction were obtained in the formof global average 119886119870119871 values (minminus1) for a column andstirred tank extractor (iv) adsorption kinetic parametersfor adsorption were obtained in the form of global averagelinear driving force parameter (119870LDF) values (minminus1) (v)tests of adsorbent regeneration by thermal swing were mademeasuring the relevant parameters (vi) simulations wererun for continuous and discontinuous units varying processconditions

21 Materials Edible sunflower oil and oleic acid (SigmandashAldrich 99 grade) were used as a source of triglycerideand fatty acid respectively Acidified solutions of plantoil of variable concentration were obtained by dissolving

weighed amounts of oleic acid in sunflower oil The solventused methanol was supplied by Biopack (Buenos AiresArgentina) The chemical purities were higher than 99 Allcompoundswere usedwithout further purification Activatedcarbon (Filtrasorb Calgon Carbon) was used in this studyThe carbon was conditioned upon receiving by boiling indeionized water for 1 hour then drying in an oven at 110∘Cfor 24 hoursThe activated carbon had a BET specific surfacearea of 972m2 gminus1 a total pore volume of 068mL gminus1 and abulk density of 0502 gmLminus1

22 Liquid Extraction Equilibrium The feed-impurity-sol-vent system was sunflower oil-oleic acid-methanol The oleicacid was distributed between the sunflower oil (oil phase) andmethanol (alcohol phase) The alcohol and oil phases weremostly immiscible Experimental LLE data were obtained ina stirred tank reactor This had an AISI 304 stainless steelvessel with 100ml total volume 40mm of diameter and80mm of length and a magnetic coupling between the motorand the stirrer The tank was heated with a tubular furnaceand the temperature was controlled with a Novus N1100controller The amounts of each component for preparingthe solutions were determined by weighing on an analyticalbalance (Model Shimadzu AUW220D Dual Range Balance00001 g precision) The mixtures were vigorously stirred for4 h and then left to rest for at least 12 h This led to theformation of two clear and transparent phases with a well-defined interface that were sampled for analysis

The oleic acid concentrationwas determined by potentio-metric titration (AOCSMethod Ca 5a-40) with amicroburetThe amount of methanol in the oil phase was determinedby weighing the liquid before and after evaporating thesolution (80∘C 300mmHg vacuum) The amount of oil inthe methanol phase was determined from a mass balance ofthe previous components The analysis was repeated at leastthree times and the average of these readings was taken asthe liquid phase composition

23 Extraction Kinetics Values of the average mass transfercoefficient on the solvent side 119886119870MeOH were calculated fromextraction tests in two kinds of extractors a spray columnand a laboratory stirred tank reactor Coefficients for thecolumn were obtained from single drop experiments usingthe methodology of Azizi et al [24]

In the stirred tank tests the technique of Schindler andTreybal [18] was followed A stirred tank was used thathad the same flange and stirrer as the extraction tests Theinternal volume diameter and length were also the sameas in Section 22 The only difference was that the tank hadtwo additional connections for continuous operation Theflowrates of solvent and feed were controlled with peristalticpumps The oleic acid concentration in the raffinate andextract phases were determined by titration after adequatesettling and formation of two distinct separate phases

24 Adsorption Equilibrium Adsorption isotherms weremeasured in a continuously stirred tank batch reactor Themethod chosen was that of solid addition in which different

International Journal of Chemical Engineering 3

LLE unit SLA unit(production)

Feed

Raffinate

Extract

Solvent

SLA unit(regeneration)

Concentrated impurity

Regenerating stream

q q

x＞

x２ff

yＲＮ

y３ＩＦＰ

Figure 1 Scheme of liquid-liquid extraction (LLE) unit coupled to solid-liquid adsorption (SLA) unit

amounts of adsorbent in powder form (about 200 meshes)were allowed to reach equilibriumThe stirring rate was keptat 1600 rpm and the temperaturewas kept at 30 and 40∘CTheacidity of themethanol solution was determined by potentio-metric titration using the average of two measurements Thistechnique had an average error of 069 Concentration ofoleic acid in the solid was determined by a mass balance

25 Adsorption Kinetics Kinetics of adsorption over theactivated carbon in pellet form were measured in a packedbed columnwith recycleThemass of the bed was 1 g and theflowrate of the extract (solvent with dissolved oleic acid) was7 L hminus1 The carbon particle size was 35ndash60 meshesThe massof the liquid phase was 40 g and the test lasted 2 h Sampleswere taken periodically and oleic acid content of the liquidphase was measured by titration of the acidity as indicatedabove

26 Settling Tests Tests of settling rates were made forthe sunflower oil-methanol system by vigorously stirringmixtures of varying solvent-to-oil ratio then being allowedto rest at three different temperatures 25 40 and 50∘C Thetime was recorded when two distinctive phases were formedand no oil remained in suspension in the upper phase

3 Results and Discussion

31 Theoretical Analysis A scheme of the proposed combi-nation of operations is included in Figure 1The LLE unit canbe a batch or continuousmixersettler unit a countercurrentor cocurrent contact column The solid-liquid adsorptionunit (SLA) can be a bleaching stirred tank or a packedadsorbent column The latter seems better suited for theproposed combination because it allows an easy separationof the solvent and an easy regeneration of the adsorbent

The successful matching of the SLA and LLE unitsseems to relay on the adequate design of the equipmentand the choice of solvent and adsorbent The feed (withimpurity concentration 119909Feed) is mixed with the solvent (withimpurity concentration 119910Solv) and leaves the contactor witha lower concentration of impurity as the raffinate stream

00

01

02

03

04

05

06

07

regeneration

yＬＥ

y３ＩＦＰ

(mol

Lminus1)

0 25 7550 125100

t (min)

Figure 2 SLA unit operated to saturation (dashed line) or incyclic mode with intermediate regeneration (solid line) Impurityconcentration in the fluid phase at the adsorption column outlet119910Solv as a function of time Continuous operation 119910break = value of119910Solv at regeneration time

(of impurity concentration 119909Raff ) The extract (with impurityconcentration 119910Ext) that leaves the unit must then be fed tothe adsorbent columnThis column is operated in productionmode until the impurity concentration in the exit reaches alimit value (119910break) This is called the breakthrough point Atthis point the feed is stopped and the column is put intoregeneration The regeneration step can be typically of thethermal type the column being flushed with a hot fluid todesorb the impurity For example a stream of hot solventcan be used the volume of solvent for regeneration beingconveniently small Figure 2 shows a plot of the concentrationof the impurity at the SLA unit exit The dashed linecorresponds to the outlet concentration for the case in whichthe column is operated to saturation with no intermediateregeneration

It must be noted that when the column becomes sat-urated the exit concentration becomes equal to 119910Ext the


25 50 75 100 125 1500

t (min)

q0

qＭＮ

q＜ＬＥ

000

100

200

300

400

q(m

ol k

gminus1)

Figure 3 Concentration of impurity in the solid phase in anadsorbent column with cyclic regeneration

concentration at the inlet The concentration of impurityin the regenerated column is usually not zero becausetotal regeneration can be rather costly and not practicallynecessary As seen in Figure 2 the column also yields aregenerated solvent with variable purity However this is nota problem provided 119910break is conveniently low

A similar kind of plot can be seen in Figure 3 but thistime the average concentration of the impurity in the solidphase is plotted as a function of time Again it can be seenthat the column is cycled at a 119902break point different to thesaturation value This is because at saturation mass transferkinetics become too slow due to the decrease in the drivingforce

While columns have both a space and time-dependentconcentration profiles other liquid-liquid contactors do notStirred tanks with adsorbent in suspension or short packedcolumns operated with a high liquid recycle have a practicallyuniform concentration with respect to the spatial coordinateand vary only as a function of time

The adequate design of the LLE and the SLA unitsshould meet some obvious criteria (i) the period of oper-ationregeneration of the column should not be too shortand a minimum value should be specified for example10ndash40min (ii) the adsorbent should have an adequatecapacity and adequate adsorption kinetics with an adequateutilization of the total surface at the point of breakpoint forexample 40ndash70 saturation The latter is usually a problemfor most adsorbents because the internal surface area is veryhigh for materials with small pores and the mass transferintrapellet resistance limits the access to the inner porevolume

A comparison of the mathematical expression for liquid-liquid and solid-liquid equilibrium is necessary for under-standing the nature of both phenomena The same can besaid for the kinetic expressions for mass transfer betweenthe two phases either liquid-liquid or liquid-solid Ratherthan working with general expressions the expressions willbe written for the practical example the system of extraction

of oleic acid from acidic sunflower oil with methanol andadsorption of oleic acid frommethanol over activated carbonFor simplification the solvent and the feed are supposed tobe practically immiscible and that Nernst law is always validThere is also no reaction involved For the column equationsplug flow of the individual phases is assumed

Equation (1) in Table 1 is an example of the isothermequation for a solid-liquid-adsorbate system in equilibriumand depicts the equilibrium concentration of the impurity onthe solid as a function of the concentration in the liquid phaseThe function used is that of the Langmuir isotherm Equation(2) is Freundlich isotherm Equation (3) is the definitionof the partition coefficient for the impurity between theraffinate and extract phases according to the Nernst lawThecoefficient 119898 is a complex function of fluid thermodynamicproperties Nernst law is deduced for low concentrations ofsolute but can be applied to solutions of higher concentrationthough its validity is reduced to a narrower range

Equations (4)ndash(6) are the equations for the flux densities(moles per unit area and time) through the liquid-solidinterface while (7)ndash(9) are the equations correspondingto the transfer to the liquid-liquid interphase Equations(7)ndash(9) correspond to the double film model while (4)ndash(6)correspond to transfer due to Fickian diffusion on the poroussolid side and film diffusion on the liquid side In (7)ndash(9)the underlying hypotheses of the double film model applythat is the liquid phases are separated by an interface andone film in each phase adheres to this interface The masstransfer takes place exclusively in this double stagnant film bya molecular diffusion mechanism In the bulk of each phasethe concentration of the impurity is uniform due to perfectmixing

Equations (10)ndash(12) and (13)ndash(16) of Table 1 correspondto flux equations in terms of driving forces and overall masstransfer coefficients The former are the differences betweenequilibrium and actual values of concentration at any pointin time In the case of adsorption the definition of globalmass transfer coefficient resembles that of the linear drivingforcemodel119870LDF and hence it will be used as suchThe LDFmodel was first proposed by Glueckauf and Coates [25] asan approximation to mass transfer phenomena in adsorptionprocesses in the gas phase but has been found useful tomodeladsorption in packed beds because it is simple and consistentboth analytically and physically [26] Several authors haveinspected the nature of 119870LDF Ruthven and Farooq [27]considered that it is composed of two contributions related tothe intrapelletmass transfer resistance (119877119863) and the filmmasstransfer resistance (119877119891) the explicit formulation being that of(17) While 119877119891 depends on the diffusivity of the impurity inthe fluid phase 119877119863 depends on the diffusivity of the impurityinside the porous matrix of the solid adsorbate Hence inmost cases and particularly in adsorption in liquid phase119877119863is the highest resistance and 119877119891 can be neglected

Equations (18) and (19) depict the relations betweenthe local liquid-liquid mass transfer coefficients and theoverall coefficients The latter can be expressed in terms ofdriving forces in the raffinate or extract side leading to twodifferent coefficients For the local coefficients depending onwhich phase is continuous and which is disperse different


Table 1 Relation between SLA and LLE equilibrium and kinetic coefficients Application to the extraction of oleic acid with methanol fromsunflower oil and the adsorption of oleic acid from methanol over a solid adsorbent (activated carbon)

Concept Adsorption L-L extraction

Equilibrium distributionbetween the two phases

119902OA = 120572119871OA 1199101 + 119871OA119910 + 119871Oil1199101015840 + 119871MeOH11991010158401015840 (1)

119902OA = 1198701198651199101119899 (2)119902OA = 119902 = concentration of impurity in the solid

phase mol kgminus1

119910 = concentration of impurity in the liquid phasemol Lminus1 119910 = 119898119909 (3)

1199101015840 = concentration of oil in the liquid phasemol Lminus1

119909 = concentration of impurity (oleic acid)in the raffinate phase (sunflower oil)

mol Lminus1

11991010158401015840 = concentration of methanol in the liquidphase mol Lminus1

119910 = concentration of impurity (oleic acid)in the extract phase (methanol) mol Lminus1

119871OA = Langmuir constant for OA adsorptionLmolminus1

120572 = saturation capacity for OA over the adsorbentmol kgminus1

119870119865 119899 = Freundlich constants for specificadsorbent and adsorbate

119867OA = 119871OA120572= Henryrsquos constant for adsorption ofOA

Relation between fluxdensities and interfacialgradients

119873OA = 119896119891 (119910 minus 119910surf) (4)119902surf = 119876 (119910surf) (5)

119873OA = 119863120588119901 (120597119902120597119903)surf

(6)119873OA = flux across the film surrounding the

particle (eq (4)) 119873OA = 119896MeOH (119910int minus 119910) (7)119873OA = flux due to diffusion (eq (6)) Both fluxes

are equal at steady-state 119873OA = 119896Oil (119909 minus 119909int) (8)119896119891 = film coefficient 119910int = 119898119909int (9)

119902surf = surface concentration of adsorbate 119896MeOH = film coeff MeOH side119910surf = concentration of impurity on the surface of

the adsorbent 119896Oil = film coeff oil side

119876 = function that gives the value of 119902 from thevalue of concentrations in the liquid phase (eq

(1))119873OA = impurity molar flux molecules per

unit time and area

120588119901 = bulk density of the adsorbent particle int = interface119863 = net diffusivity of the adsorbate inside the

adsorbent particle

Relation between fluxesand driving forces

120597119902av120597119905 = 119896119890 (119902surf minus 119902av) (10) 119873OA = 119870MeOH (119910eq minus 119910) (13)

120597119902av120597119905 = 119870LDF (119902eq minus 119902av) (11) 119873OA = 119870Oil (119909 minus 119909eq) (14)

119902eq = 119876 (119910) (12) 119910eq = 119898119909eq (15)119902av = average adsorbate concentration in the

adsorbent particle119909eq = 119910

119898 (16)119902surf = surface concentration 119870MeOH = overall transfer coefficient

119896119890 = effective film coefficient for intrapelletdiffusion 119870Oil = overall transfer coefficient

119902eq = equilibrium adsorbate concentration for 119910(eq (1)) eq = equilibrium

119870LDF = linear driving force mass transfercoefficient for adsorption


Table 1 Continued


Mass transfer coefficients

1119870MeOH

= 1119896MeOH

+ 119898119896Oil

(18)1

119870Oil= 1

119896Oil+ 1

119898119896MeOH= 1

119898119870MeOH(19)

119878ℎ119888 = 0725119878042119888 R119890057119888 (1 minus 120601) (20)119878ℎ119888 = 11989611988811988932

119863119888 (21)1

119870LDF= 119877119891 + 119877119863 = 119903119901

3119896119891 + 119903 211990115120576119863 (17) R119890119888 = 12058811988811988932Vslip

120583119888 (22)120576 = porosity of the adsorbent particle 119878119888119888 = 120583119888

119863119888 (23)119903119901 = radius of the adsorbent particle 119896119889 = 0023Vslip 119878119888minus05119888 (24)

119896119891 = film transfer coefficient 120601 = 119881119889119881119889 + 119881119888 (25)

119877119891 = film transfer resistance 120601 = hold-up of the disperse phase (oil)119877119863 = intrapellet diffusion resistance 11988932 = average Sauter diameter

Vslip = slip velocity between phases120583119888 = viscosity of the continuous phase119881119889 = volume of the disperse phase

119881119888 = volume of the continuous phase119896119888 = mass transfer coefficient continuous

phase119896119889 = mass transfer coefficient disperse

phase

Mass balance batch unitperfectly mixed

119881119889119910119889119905 = 119882119870LDF (119902eq minus 119902av) (26)

(30)119910 = 1199100 119902 = 1199020 119905 = 0 (27) 119889119910119889119905 = 119886119870MeOH

1 minus 120601 (119910eq minus 119910)119889119910119889119905 = 1198701015840LDF (119910 minus 119910eq) (28) 119889119909

119889119905 = 119886119870Oil120601 (119909 minus 119909eq) (31)

119902av = 119876 (119910eq) (29) 119909 = 1199090 119910 = 1199100 119905 = 0119881 = volume of adsorbent 119886 = interfacial area per unit volume of

whole liquid phase119882 = weight of adsorbent

Mass balance continuouscontact tower equations

120597119910120597119905 minus 119863119871 120597 21199101205971199112 + 120597 (119906119910)

120597119911 + 1 minus 120576119861120576119861 120588119901 120597119902120597119905 = 0 (32)

119910 (0 119905) = 1199100 (33)120597119910120597119911 = 0 119911 = 119871 (34) 119911 = int1199092

1199091

Voil119889119909119886av119870Oil (119909 minus 119909eq)

(36)

119910 (119911 0) = 0 (35) 119911 = int11991021199101

VMeOH119889119910119886av119870MeOH (119910eq minus 119910) (37)

120576119861 = bed porosity V = superficial velocity119910 = fluid phase concentration of the impurity 119911 = axial coordinate height of the column

119902 = solid phase concentration of the impurity 119886av = average interfacial area per unitvolume of the contactor vessel

119906 = interstitial velocity = V120576119861120588119901 = particle density

119863119871 = diffusivity of the adsorbate in fluid119911 = axial coordinate

119871 = height of the column


correlations will be applied to each side though both areusually of the form of an equation of the Sherwood numberas a function of the Reynolds number the Schmidt numberand the holdup of the disperse phase Example equations forcalculating 119896119888 (local coefficient for the continuous phase) and119896119889 (coefficient for the disperse phase) for column continuouscontactors have been written in (20)ndash(25) according to thesuggestions of Koncsag and Barbulescu [28]

Discontinuous operation in perfectly mixed units showsmore similarities for both operations as it can be seen in(26) to (31) in Table 1 This is also a problem easily handledmathematically and corresponds to the operation of stirredtank extraction units and stirred tank adsorption unitsPacked columns operated with a high recycle ratio can alsobe considered as perfectly mixed stirred tanks In (30) and(31) 119886 is the interfacial area per unit equipment volumeAlternatively the balance in the liquid phase can be writtenin terms of a 1198701015840LDF coefficient as in (28) and (29)

Equations (32) to (37) correspond to the differentialequations border and initial conditions that express themassbalance for the adsorbate species during the movement alongthe packed bed and diffusion inside the porous adsorbentThe full model takes into account the backmixing in the axialdirection by considering a Fickian diffusion term with axialdiffusivity 119863119871 These equations must be coupled to a localequation like (26) that describes the law of variation of 119902 asa function of the process variables

Equations (36) and (37) are the integrated forms of thelocal mass balance They are simpler to handle than theprevious one for adsorption However this is a simplifiedview and more sophisticated models are needed to reflectphenomena of emulsion formation and collapse carryoverflooding drop coalescence and breakage and so forth

The interfacial area during extraction is a function of thedrop sizeThe drop size is bigger at higher values of holdup ofthe disperse phase and at bigger values of the stirring power(in stirred tanks) However the dependence is soft

The inspection of (26) to (29) and (30) and (31) andtheir comparison with (32) and (33) shows that adsorptioncolumns can never work in a true steady state like liquid-liquid extraction units do To describe their operation asolution as a function of time and space must be foundThis is because the solid phase is fixed while the fluidphase is flowing continuously As shown in Figure 2 whenthe breakthrough condition of the column is reached theoperation of the adsorber must be stopped and regenerationmust be performed This is different from the extractioncolumn in which a continuous steady state can always beestablished between the two flowing liquid phases

For both adsorption and extraction the throughput forany separation unit is mainly given by intrinsic parameterssuch as the kinetic mass transfer parameters the thermo-dynamic constants for the L-L and S-L equilibrium theparameters describing the interface and the total volumeof the phases For any given choice of contacting deviceand set of process conditions that is temperature pressureand liquid phase flowrates the volume of the unit would bethe result of a design procedure for a given desired rate ofextraction and adsorption because the process conditions

will dictate the values of the intrinsic parameters Thereforeit is of interest to list the range of values of themost importantintrinsic parameters involved in the design of adsorption andextractors This is done in Table 2

Equivalent coefficients have been placed in the same rowIn the case of the interfacial area for adsorption all availablesurface area external and intrapellet has been included Itmust be noted that due to diffusional resistance not allsurface is readily availableHowever this is taken into accountwhen calculating the intrapellet mass transfer resistanceSince for mesoporous and microporous adsorbents theinner surface ismuchhigher than external one a is practicallythe intrapellet area divided by the pellet volume It is apparentfrom this comparison that the S-L interfacial area is muchhigher than the L-L for most adsorbents and L-L contact-ors

Inspection of the last row of Table 2 yields the mostimportant insight If 1198701015840LDF and 119886119870119871 values are comparedthis is a comparison of parameters with similar driving forcesit can be seen that in global terms adsorption is slowerthan extraction under most conditions especially for thecase of extraction in stirred tanks or static mixers Moresimilar values are obtained when we match adsorption witha low energy extraction operation for example in a spraycolumn For the coupling of both units however what itmust be similar is the uptake of impurity per unit time andthis is a function also of the driving force In this senseslow adsorption kinetics can be compensated by high solidaffinities (high 119867 119902eq values) while fast extraction kineticscould be inhibited by low impurity solubilities (low119898 values)All these considerations will have a better insight once theexamples are discussed in detail

32 Example of Plant Oil Deacidification In this examplesunflower oil is first extracted with methanol in order toremove the impurities that is oleic acid The solvent is thenregenerated by adsorption of oleic acid from the methanolsolution The adsorbent is in turn regenerated by a thermalswing In order to obtain sound values of the parameters thatdescribe the phenomena separate experiments for determin-ing the liquid-liquid and liquid-solid thermodynamic andkinetic parameters were performed

321 Determination of the Partition Coefficient for Oleic AcidFor the system oleic acid-methanol-sunflower oil values of119898 were obtained from plots of 119910 as a function of 119909 at threedifferent temperatures (Figure 4) 119898 was found to be equalto 0875 at 30∘C 0922 at 40∘C and 1125 at 50∘C 119898 wascalculated as the ratio of the concentration of oleic acid in thealcohol phase (free of oil) to the concentration of oleic acid insunflower oil (free of methanol) The concentration of oleicacid in either phase was really a little lower due to dissolutionof methanol in the oil phase and dissolution of oil in thealcohol phase In this sense the higher solubility of oleic acidin methanol at higher temperatures is also accompanied by ahigher solubility of the oil and hence there must be a balancewhen choosing the right temperature of operation becausethe relative purity of the extract or the yield of raffinate can


Table2Ty

picalvaluesa

ndranges

ofintrinsic

parametersfor

adsorptio

nandextractio

nAC

=activ

ated

carbon

Adsorptio

nparameter

Syste

mRa

nge

Extractio

nparameter

Syste

mRa

nge

119867adim

(m

olkgminus1)(m

olLminus1)

Oleicacid-biodiesel-

silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

Sulfu

rcom

poun

ds-diesel-A

C[11]

Thioph

ene-naph

tha-NaY

[12]

Naphthenica

cid-toluene-cla

y[13]

dichlorobenzene-methano

l-graph

ene[

14]

Oleicacid-m

ethano

l-AC(th

iswork)

20ndash50

20 1 06 21

23

598 33

119898adim

(m

olLminus1)(m

olLminus1)

Toluene-Ac

eton

e-Water

[15]

Oleicacid-canno

lioil-m

ethano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0967

0705

0133

07415

119902max molkgminus1

Oleicacid

over

silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

sulfu

rcom

poun

ds-diesel-A

C[11]

thioph

ene-naph

tha-NaY

[12]

Oleicacid-m

ethano

l-AC(th

iswork)

496

2372

0272

0003

2713

298

119910max

molLminus1

Toluene-aceton

e-water

[15]

Oleicacid-canolao

il-methano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0498

031

00225

0104

119886cm2cmminus3lowast

Silica[

8]Re

sin[9]

NaY

[12]

Norit[11]

AC[10]

AC(th

iswork)

104ndash105

107

106ndash107

106

107

107

119886cm2cmminus3lowastlowast

Stirr

edtank

Regt

1000

0[18]

Con

tinuo

usstirr

edtank

[19]

Packed

bed[20]

Slug

flow[21]

Staticmixer

[14]

Spraycolumn(th

iswork)

10ndash250

32ndash130

15ndash3

20ndash4

030ndash300

3

1198701015840 LDFminminus1lowastlowastlowast


silica[

8]Naphthalene-w

ater-carbo

n[10]

p-am

inob

enzoicacid-w

ater-resin

[9]

thioph

ene-naph

tha-NaY

[12]

sulfu

rcom

poun

ds-diesel-A

C[11]

Oleicacid-m

ethano

l-AC(th

iswork)

00365

004

4006

00167

000

4006

6

aK119871m

inminus1lowastlowastlowastlowast

Water-kerosene-aceticacidcolum

n[22]

Toluene-aceticacidndashw

atercolum

n[23]

Sunfl

ower

oil-o

leicacid-m

ethano

lstirred

tank

(thiswork)

0054

0049

075

lowastPeru

nitv

olum

eof

packed

bedlowastlowastperu

nitv

olum

eof

extractorcoeffi

cientsforthe

continuo

usph

ase(usuallythesolventfor

high

solvent-to-feed

syste

ms)lowastlowastlowastCa

lculationof1198701015840 LDFisdo

newith

equatio

ns(26)

and(29)of

Table1lowastlowastlowastlowast119870119871=overallm

asstransferc

oefficientL-Lextractio

ncontinuo

usph

ases

ide(119870MeO

Hin

exam

pleo

fthisw

ork)C

alculatio

nof119870119871isdo

newith

equatio

ns(18)and(19)of

Table1


Table 3 Experimental values of drop size (119889) terminal velocity (119881)andmass transfer coefficients (continuous phase side) (119886119870119871) at threedifferent temperatures Single drop tests 120601 = 015T ∘C 119889 mm 119886 cm2 cmminus3 119881 cm sminus1 119886119870119871 minminus1

30 319 188 621 084240 313 192 493 119050 313 192 313 1610

085

090

095

100

105

110

115

Part

ition

Coe

ffici

ent

35 40 4530 50

Temperature (∘C)

Figure 4 Plot of the partition coefficient as a function of thetemperature of the experiment

be an issueThe linearity of the 119910 versus 119909 plots was very goodwith 1199032 values of about 0997322 Determination of Mass Transfer Coefficient of OleicAcid between Methanol and Sunflower Oil Mass transfercoefficients varied widely depending on the type of contactequipment used Results are presented in Table 3 for spraycolumn single drops experiments Average diameter valueswere calculated with Sauterrsquos formula (see (38)) Interfacialarea is calculated with (39) In the case of the columnincreasing temperature values (from 30 to 50∘C) increasedthe overall mass transfer coefficient by almost a factor of2 This is possibly related to the decrease in viscosity and asignificant increase of the Reynolds number (Re)

11988932 = sum 11989911989411988931198941198991198941198892119894 (38)

119886 = 612060111988932 (39)

The values obtained compared fairly well with othersreported Sankarshana et al [29] found values of 119886119870119871(continuous phase side feedsolvent = 02ndash1) equal to002ndash006minminus1 for packed columns with random andordered packing while working with a system of aceticacid-ethyl acetate-water system under countercurrent modeNosratinia et al [30] found 119886119870119871 values of 012ndash084minminus1in a spray column with jet injection of the disperse phase

Geankoplis and Hixson [31] found values of overall 119886119870119871(water continuous phase side) of 007ndash12minminus1 at varyingdisperse phase flow rates in a ferric chloride-isopropyl ether-HCl(aq) system in a spray tower

In spray towers for liquid-liquid extraction the Sauterdiameter is a function of the disperse phase holdup and fluiddynamic conditions Salimi-Khorshidi et al [32] found that11988932 varied within 25ndash4mm when varying 120601 = 01ndash06 andflowrates a volcano plot being found for 11988932 as a function ofRe or 120601

Considerations for the scale-up of 119886119870119871 coefficients fromsingle drop measurements to full-scale drop swarms shouldbe discussed Hughmark [33] studied comprehensive datasets with 120601 = 0006ndash02 and early found that for ratios ofthe continuous to disperse phase viscosity less than one themass transfer coefficients (in the form of Sh119888 or 119896119888) for thecontinuous phase of drop swarms were the same as for singledrops while for viscosity ratios greater than one themultipledrop coefficients were somewhat smaller Hughmark fittedhis data with Ruby and Elgin 119896119888 equation [34] In this systemthe coefficient 119886119870119871 is thus a function of the impeller Reynoldsand also directly proportional to 120601

The value of the mass transfer coefficient for the stirredtank experiment was 075minminus1 using an experimental setupsimilar to that of Schindler and Treybal [18] and using aholdup of disperse phase of 05These authors early correlatedthe mass transfer coefficients for stirred tanks studying themass transfer between two liquid phases in an agitated baffledvessel and found that mass transfer coefficient increasedwith impeller Reynolds number and disperse phase holdupBaffling roughly increased 119886119870119871 by 15 times Average volu-metric 119886119870119871 values for the continuous phase ranged within3ndash25minminus1 for values of the impeller Reynolds number of20000ndash60000The dependence of 119886119870119871 on impeller Reynoldsnumber was strong being roughly proportional to Re2ndash25while the dependence on holdup of the disperse phase wasweaker being proportional to about 12060109ndash1 They also foundthat 119896119888 was almost insensitive to variations in the holdupThese trends can be easily rationalized by considering thatfor stirred tanks the Sauter diameter as in (38) is imposedby the impeller Reynolds while the interfacial area per unitvolume of disperse phase corresponds to the value given by(39) For this reason values of 119886119870119871 for the simulations willbe extrapolated from experimental data at similar stirringconditions by scaling with the value of 120601

323 Adsorption Properties For the oleic acid-methanol-sunflower oil system an adsorbent of activated carbon waschosen because of its good performance in preliminaryscreening testsThis is a fairly novel application for carboxylicacid adsorption since in the literature silica silicates claysand zeolites have usually been employed [35ndash37] whilereports on the use of carbons are concentrated on decontam-ination of water [38 39]

Results of adsorption of oleic acid in sunflower oiloveractivated carbon are plotted in Figure 5 The curvescorrespond to a virgin activated carbon The last value inthe abscissae axis is 010mol Lminus1 therefore the plotted results


002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)

Figure 5 Adsorption of oleic acid from methanol at room temper-ature (30∘C) and overactivated carbon (Calgon Carbon Filtrasorb200) Correlation coefficient for Henryrsquos constant (slope of the line)1199032 = 097

correspond to mildly acidic sunflower oil and sunflower oilof low acid content The curve is better interpreted by theFreundlich model (Table 1 equation (2)) with119870119865 = 834 and119899 = 166 with an 1199032 = 0971 For simplicity of the treatmenthowever the data can be fitted with Henryrsquos linear isothermyielding119867 = 33 (L kgminus1)

Another isotherm was taken at 40∘C in order to calculatethe heat of adsorption This isotherm had an 119867 value of16 (L kgminus1) Applying the integrated form of vanrsquot Hoffequation and considering that the heat of adsorption wasnot a function of temperature the heat of adsorption wasestimated as 57 kJmolminus1 This value compares well with otherfound in the literature Li et al [40] found that adsorptionof phenol on resin from aqueous solutions had a heat ofadsorption of about 38 kJmolminus1 Chiou and Li [41] founda heat of adsorption of 529 kJmolminus1 for reactive dye inaqueous solution on chemical cross-linked chitosan beadsIlgen and Dulger [42] measured a value of about 34 kJmolminus1for the adsorption of oleic acid from sunflower oil over zeolite13x

Adsorption tests were also made in a packed bed columnwith fast recycle In this column the axial concentrationgradient was negligible and the behavior was similar to astirred tank with perfect mixing The results were fitted withthe simple model of the linear driving force model in theform of (26) The results for one of such tests are plotted inFigure 6 The calculated value for 119870LDF from the experimentis 0066minminus1

324 Settling Times for Phase Separation in a GravityDecanter The results of the experiments to measure thesettling time as a function of the volumetric methanol-to-oil ratio and the temperature are included in Figure 7 At25∘C for methanol-to-oil ratios lower than 1 no completephase separation could be achieved even after 1 day some oil

0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)

Figure 6 Concentration of oleic acid in both the liquid phase (119910998771)and solid phase (119902 ◼) as a function of time Packed columnwith fastrecycle 30∘C Granular activated carbon average pellet diameter04mm

2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)

Figure 7 Experimental settling time as a function of the tempera-ture Tank with no internals and no coalescing aid Disperse phase(oil) holdup 05 () 025 (I) and 02 (◻)

remaining disperse in the methanol phase as indicated by theopacity of the upper phase

Complete separation was achieved for methanol-to-oilratios equal to or higher than 1 The general trends werethat at high temperatures for example 50∘C the settlingtime was independent of the methanol-to-oil ratio while atlower temperatures highermethanol-to-oil ratios lowered thesettling time For a continuous operation of a decanter witha settling time of one hour a temperature of about 50∘C isneeded

325 Simulation

(1) Simulation of a Continuous ExtractionAdsorption ProcessThe layout of a process using an extraction column coupledto a set of twin adsorption columns is depicted in Figure 8


LLE unit SLA A unit

Feed

Raffinate

Extract

SLA B unit Concentrated impurity

Regenerating stream

Solvent

Figure 8 Flowsheet of extraction column coupled to a set of twin adsorption columns 119882Ads = 364 kg 119881Ext = 015m3

00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)

Figure 9 Plot of 119909 (dashed line) and 119910 (solid line) as a functionof the position in the countercurrent extraction column 119876Feed =4 Lminminus1 119876Solv = 80 Lminminus1 119909Feed = 0140mol Lminus1 119910Solv =07mmol Lminus1

In this layout the operation of the extraction column is con-tinuous and the equations describing the relation betweenthe concentration in any phase and time are (32)ndash(35) inthe case of the adsorption column The equations describingthe exchange in the case of extraction are (36) and (37)The solution of this system of equations for an example ofextraction of acidic sunflower oil with methanol is depictedfrom Figures 9ndash11

The holdup of the disperse phase was calculated from theexperimentally measured characteristic velocity of the drop(V119870) and the values of the flowrates of the feed and solventbymeans of the equation of Gayler (40) Gayler proposed thatfor many different types of columns the following equationheld [43]

119906119889120601 + 119906119888

1 minus 120601 = 119906119896 (1 minus 120601) (40)where 119906119888 and 119906119889 are the superficial velocities of the contin-uous and dispersed phases respectively and 119906119896 the char-acteristic velocity is the mean relative velocity of dropletsextrapolated to zero flowrate and can be identified with

00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)

Figure 10 Adsorption column Plot of 119910 (solid line) and 119902 (dashedline) as a function of time119882Ads = 364 kg119876Ext = 80 Lminminus1 119910Ext =764mmol Lminus1 1199020 = 0mol kgminus1

the terminal velocity of a single drop in the equipmentconcerned Equation (40) was numerically solved giving 120601 =0044

The equations of the column extraction unit were solvedanalytically in order to avoid the problem of solving the two-point boundary value problem imposed by the countercur-rent flow Equations (36) and (37) were solved by obtainingthe eigenvalues and eigenvectors of the matrix of derivativesand by considering that the solution eigenfunction was 119910 =119896 exp(120582119911) One eigenvaluewas found to be zero so both119909 and119910 had the form 119892 = 119860+119861 exp(120582119911) where 120582 is a function of119898VFeed VSolv and 119886119870119871 and119860 and 119861 are function of the previousparameters and also the initial conditions

The impurity concentration in the feed 140mmol Lminus1 isequivalent to about 44 acidity which should be reduced toabout 30mmol Lminus1 in order to be suitable as a feed for thebiodiesel alkali-catalyzed process For some applications themaximum acidity is even lower For insulating oils ASTMD3487 establishes a maximum acidity of 003mgKOHgminus1that is 05mmol Lminus1 This is near the concentration valueof the raffinate in Figure 9 11mmol Lminus1 Anyway a speciallimit for insulating oils of plant origin is established in ASTM


0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)

Figure 11 Adsorption unit cycle time (solid line) and raffinateconcentration (dashed line) as a function of the initial concentrationof the impurity in the oil phase (119909Feed) for a required residualconcentration at the outlet of the adsorption column (119910break) of140mmol Lminus1

D6871 06mgKOHgminus1 that is 10mmol Lminus1 nine times thefinal value in the raffinate of the example 06mgKOHgminus1is also the limit established for refined edible oil for humanconsumption (FAO CODEX-STAN 210 1999) and thereforethe extraction step is suitable for this application also thoughit must be noted that ethanol would bemore appropriate thanmethanol for a better compliance with health restrictions

For simulating the adsorption column the set of equa-tions (32)ndash(35) for the adsorption column was solved afteranalyzing the underlying hypotheses The axial dispersionsometimes produces the broadening of the adsorption frontdue to the contribution of both molecular diffusion anddispersion caused by fluid flow [44] The impact of the axialdispersion is assessed by the Peclet number (Pe) small valuesindicating backmixing is important According to Carberry[45] for Pe values much greater than 100 the flow can beconsidered plug flow type Values of molecular diffusivityof oleic acid in methanol were calculated with Wilke-Changequation Axial diffusivity was calculated with the correlationof Wakao and Funazkri [46] The Pe value was found to beequal to 3300

For the case of linear isotherm a solution to ((32)ndash(35))in the form (119902 119910) = 119891(119911 119905) can be got by using the ldquoquasi-log normal distributionrdquo (Q-LND) [47 48] In this caseit is assumed that the quasi-log normal probability densityfunction can be used to represent the impulse response of thesystem It has been demonstrated that the analytical solutionand the Q-LND approximate solution are similar for a widerange of the model parameters and that deviations appearonly for very low values of the residence time This solutionis used to plot the breakthrough curve of Figure 10

In Figure 10 the concentration of the extract is764mmol Lminus1 and must be reduced to 07mmol Lminus1 Theconcentration of impurity in the solvent at the column outletis nonlinear function of time and the initial concentra-tion is about zero An outlet concentration of 14mmol Lminus1

0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)

Figure 12 Adsorption unit cycle time as a function of the pelletdiameter for a required residual concentration at the outlet of theadsorption column (119910break) of 140mmol Lminus1 and impurity in the oilphase (119909Feed) equal to 140mmol Lminus1

corresponds to an average concentration of solvent somewhatlower than 07mmol Lminus1 14mmol Lminus1 could then be safelyconsidered the column breakthrough condition (119910break)the cycling time for the packed bed being of 1 hour Atthis time the average load of impurity on the adsorbent is95mmol kgminus1 The way of removing this load to regeneratethe bed will be dealt later in detail

119905break will be a function of 119909feed and the desired purityof the solvent for a given fixed operation condition of theextraction unit This is illustrated in Figure 11 The adsorbentcolumn must be maintained in operation until the break-through occurs As expected 119905break decreases with higherconcentration of impurity in the feed but the curve is enoughsoft to allow handling varying impurity concentration in the100ndash600mmol Lminus1 range with 119905break in the 30ndash60min rangeThe concentration in the raffinate varies from 1 to 2mmol Lminus1which can be considered negligible

The simulation runs of Figures 9 10 and 11 were madewith a pellet size of 1mm For a bigger 12 times 40 meshesgranular carbon a maximum size of 15mm can be foundFor pelletized carbon sizes of up to 3mm are commonParticle size has a great influence on the cycling timebecause the intrapellet diffusion mass transfer resistance 119877119863is proportional to the square of the pellet radius This is clearin the plot of Figure 12The time of operationmust be reducedfrom 1 h to about 15minwhen the pellet size is increased from1 to 3mm

(2) Regeneration An assessment of the regeneration of thesolvent by evaporationdistillation and adsorption shouldbe made Distillation is the most common method but itrequires a relatively high amount of energy Adsorptionwas demonstrated to be a feasible regeneration methodbut it needs energy for desorbing the impurity from theadsorbent bed A comparison of the amount of heat involved


Table 4 Amounts of energy involved in the regeneration of the extract (in kJ per litre extract)

Distillation AdsorptionStep Energy Step EnergyHeating to boiling point 63 Adsorption (exothermal) 0Solvent evaporation 948 Heating 1 bed volumes of methanol to 100∘C 22

Desorption (endothermal) by elution of hot methanol lt1Heating to boiling point evaporation of solvent of eluting stream 151

Total heat duty 1011 174

SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed

Figure 13 Flowsheet of a sequential discontinuous combination of extraction tank and adsorption column with fast recycle Dimensions andprocess parameters119882Ads = 1334 kg 119881Feed = 024m3 119881Solv = 48m3 119909Feed = 140mmol Lminus1 119879 = 303K

in each case is done in Table 4 for the case of Figures 9and 10 The information of these figures indicated that theregenerated solvent had an average concentration of oleicacid of 07mmol Lminus1 and that the average concentrationon the solid at breakthrough was about 95mmol kgminus1 Theamount of solvent regenerated at 119905break was 4800 L If thisvolume had been regenerated by distillation the amount ofheat would have been 1000 kJ Lminus1 This is considering thatall the methanol is evaporated at 100 efficiency and thatthere are no schemes for heat recovery In the case of theregeneration by adsorption the heat is consumed in theheating of the eluting volume for the thermal swing heatingthis same stream to the boiling point and evaporating thesolvent in it to recover the free oleic acid A temperature ofregeneration of 100∘C and an elution volume equal to 1 bedvolume was chosenThese values permit achieving a residualconcentration of impurity in the solid lower than 5 of theoriginal load before regeneration With these parametersthe heat duty of the regeneration by adsorption amounts toabout 151 kJ Lminus1 just a 15 of the classical regeneration Theregeneration temperature demands running the regenerationstep with a little overpressure of 23 bar due to the high vaporpressure of methanol In general terms it was deduced thattemperature of regeneration is the most influential variablethe elution volume having lower impact on the residualconcentration of impurity in the solid For simplicity of theinvolved calculation regeneration will be assumed to becomplete in what follows

Time for regeneration was found experimentally Resid-ual oleic acid on the solid did not vary for time spans forregeneration higher than 25min

Some other authors have used only flushing with solventin order to regenerate the adsorbent Yori et al [49] removedglycerol from biodiesel by adsorption over a silica columnand regenerated the bed by flushing with a small amountof methanol In their case the great affinity of methanol forthe adsorbed impurity (glycerol) was the crucial factor forregenerating the bed In the studied case a thermal swing isneeded to help desorption

(3) Simulation of a Batch ExtractionAdsorption Process Forthis simulation extraction tanks were chosen with a volumeequal to that processed by the equipment of Figure 8 for1 h operation (see Figure 13) This enables a comparisonof performance and equipment requirements for similarthroughput In order to use completely discontinuous unitsa column with fast recycle was programmed that obeys (30)and (31) Choosing a stirred tank with adsorbent suspendedin the liquid would have yielded the same operation equa-tions However packed columns make regeneration easier Inorder to have a short residence time and work as a perfectlymixed stirred tank the recycling flow rate must be madefast enough Reducing the residence time to a fraction of aminute makes this possible For not making pressure drop anissue at this flow conditions the LD of the column shouldbe low


0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)

Figure 14 Plot of 119909 (dashed line) and 119910 (solid line) as a function oftime in the batch extraction unit 119882Ads = 1334 kg 119881Feed = 024m3119881Solv = 48m3 119909Feed = 140mmol Lminus1 T = 30∘C

0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

Figure 15 Plot of 119910 (solid line) and 119902 (dashed line) as a function oftime in the adsorption column with recycle119882Ads = 1334 kg119881Feed =240 L 119881Solv = 1800 L 119910Raff = 72mmol Lminus1 T = 30∘C

Results of the simulation are given in Figure 14 Theconcentration of the impurity can be reduced from 140 toabout 10mmol Lminus1 This is worse than the final value of thecountercurrent column 11mmol Lminus1 and is a consequenceof the unfavorable behavior of perfectly mixed systems withequilibrium restrictions Two stages would be necessary forachieving the final raffinate concentration of Figure 9

The results of Figure 15 bring similar conclusions as in thecase of the batch extraction unitThe performance of the col-umn with fast recycle is worse than that seen for the columnwith once-through flow Although the extract to be refinedhas a similar concentration of impurity (72mmol Lminus1) anoutlet solvent concentration of 07mmolminus1 like in Figure 10can only be achieved by increasing the mass of adsorbent 36times This is also explained by the unfavorable behavior ofperfectly mixed systems with equilibrium restrictions

The comparison of the saturation time values for both theextraction and adsorption column shows that the extractiontank has a saturation time of about 5min and the packed

8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)

Figure 16 Adsorption time (solid line) as a function of the initialconcentration of the impurity in the oil phase (119909Feed) for a requiredresidual concentration at the outlet of the adsorption column (119910break)of 070mmol Lminus1 Resulting 119909Raff as a function of 119909Feed (dashed line)119882Ads = 1334 kg

bed about 40min This is the result obtained for a volume of4800 L liquid phase in the extractor and a volume of about2600 L of packed bed It can be deduced that the throughputof each unit per unit volume is different about 200 Lminminus1per L of process vessel for the extractor and 005 for theadsorber These values are not too different as it could beexpected from the slow kinetics of adsorption and could bethe results of a compensation with a high available surfacearea for the chosen adsorbent

A comparison of all characteristic timesmust also includethe settling time of the decanter At 50∘C this is about 1 hmore similar to the saturation time of the adsorber Howeverthe packed bed should be operated at a conveniently lowtemperature like 30∘C to have a favorable adsorbing capacityAt 30∘C the settling time increases to about 3 h making thisstep the slowest of the process

A plot of the necessary minimum adsorption time asa function of the feed impurity concentration is includedin Figure 16 The same trends of Figure 11 are seen Theconcentration of the raffinate increases when the concen-tration of impurity in the feed is increased Keeping theraffinate at a constant composition can only be achieved byincreasing the solvent-to-feed ratio in the extraction unitTheoperation time for the adsorption column increases when theconcentration of impurity in the feed increasesThis is due tothe higher concentration of impurity that demandsmore timeto be removed

4 Conclusions

The deacidification of sunflower oil by extraction withmethanol and regeneration of the solvent by adsorptionon activated carbon were tried The method is considereduseful for the food and biodiesel industries as a meansof economically achieving low impurity levels in raffinatestreams in systems with a high solvent-to-feed ratio


Equations for the design of extractors and packedcolumns both in continuous and discontinuous mode weredeveloped Suitable equations for design were written fromgeneral principles highlighting similarities in formulationfor driving forces mass transfer rates thermodynamic andkinetic parameters A comparison of ranges of thermody-namic and kinetic parameters seems to indicate that thematching of extraction and adsorption units needs to accom-modate the fast kinetics of extraction with the relativelyslow kinetics of adsorption and the relatively high affinityof adsorbents per unit mass with the low capacity of mostsolvents per unit volume In this sense a solution is the cyclicoperation of a column adsorber of adequate size

Deacidification by extraction with methanol can be car-ried out with relatively high efficiency due to a fairly highvalue of the partition coefficient for oleic acid between thepolar and organic phase with values of 119898 = 074 a 093in the range of temperatures 30 a 50∘C Extraction in acountercurrent column with a solvent-to-feed ratio of 20allows reduction of the acidity of an oil with 44 acidity toa final value of 004 adequate for its use as a feedstock forthe production of biodiesel with the alkali-catalyzed processits use as edible oil for human consumption or its use asdielectric biodegradable oil Extraction permits a maximumyield of oil and the recovery of the fatty acid impurity

Typical operation of a stirred tank extraction unit yieldeda value of the global coefficient for mass transfer 119886119870119871of 075minminus1 (methanol side) Fast kinetics of extractionpermitted the operation of a stirred tank extraction unitwith a saturation time of about 5min This was faster thanthe characteristic time for saturation of a batch adsorp-tion column that had a value of about 1 h However atotally discontinuous process needs also of a decanter thathad big settling times making it the slowest step of theprocess

The extractionadsorption combination seemswell suitedfor extraction operations with a high solvent-to-feed ratioin which solvent regeneration by distillation becomes pro-hibitive due to a lower vapor pressure of the impurity tobe removedrecovered thus demanding evaporation of largeamounts of solvent It was demonstrated with the exampleof sunflower oil deacidification that for a solvent-to-feedratio of 20 the heat duty of a distillation-based solventregeneration could be as large as 1011 kJ per litre of solventA regeneration process based on adsorption needs of a heatduty much lower of about 174 kJ per litre solvent the heatduty mainly being related to the thermal swing of the packedbed

Applications for extraction with a high solvent-to-feedratio for which an extractionadsorption combination wouldbe convenient could be those using a solvent with highselectivity but low affinity for the impurity or ldquopolishingrdquooperations with a low driving force due to the high dilutionof the impurity in the feed

The extractionadsorption system is amenable for bothcontinuous and discontinuous operation However the con-tinuous operation has a higher efficiency due to the intrinsicadvantages of plug flow as compared to perfect mixing forsystems in thermodynamic equilibrium is a limitation

For adsorption the main mass transfer resistance isintrapellet diffusion In this sense small adsorbent particlesimprove the turnover of the process and increase the percent-age of utilization of the adsorbent volume

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

Thisworkwas performedwith the funding of CONICET (PIPGrants 11220130100457COand 11420110100235CO)Universi-dad Nacional del Litoral (CAI + DGrant 50420150100074LI)and Universidad Nacional de Jujuy (SeCTER 08D138)

References

[1] M S Cuevas C E C Rodrigues and A J A Meirelles ldquoEffectof solvent hydration and temperature in the deacidification pro-cess of sunflower oil using ethanolrdquo Journal of Food Engineeringvol 95 no 2 pp 291ndash297 2009

[2] C E C Rodrigues C B Goncalves E Batista and A J AMeirelles ldquoDeacidification of Vegetable Oils by Solvent Extrac-tionrdquo Recent Patents on Engineering vol 1 no 1 pp 95ndash1022007

[3] S Turkay and H Civelekoglu ldquoDeacidification of sulfur oliveoil I Single-stage liquid-liquid extraction of miscella with ethylalcoholrdquo Journal of the American Oil Chemistsrsquo Society vol 68no 2 pp 83ndash86 1991

[4] R A Carr ldquoDegumming and refining practices in the USrdquoJournal of the American Oil Chemistsrsquo Society vol 53 no 6 pp347ndash352 1976

[5] V Kale S P R Katikaneni and M Cheryan ldquoDeacidifyingrice bran oil by solvent extraction and membrane technologyrdquoJournal of the American Oil Chemistsrsquo Society vol 76 no 6 pp723ndash727 1999

[6] S S Koseoglu ldquoMembrane technology for edible oil refiningrdquoOils amp Fats International vol 5 pp 16ndash21 1991

[7] N Othman Z AManan S RWanAlwi andM R Sarmidi ldquoAreview of extraction technology for carotenoids and vitamin erecovery from palm oilrdquo Journal of Applied Sciences vol 10 no12 pp 1187ndash1191 2010

[8] CVeraM Busto J Yori et al ldquoAdsorption inBiodiesel Refining-AReviewrdquo inBiodiesel - Feedstocks andProcessing TechnologiesIntech Open Access Publishers 2011

[9] H Zhang A Li J Sun and P Li ldquoAdsorption of amphotericaromatic compounds by hyper-cross-linked resins with aminogroups and sulfonic groupsrdquo Chemical Engineering Journal vol217 pp 354ndash362 2013

[10] M Yuan S Tong S Zhao and C Q Jia ldquoAdsorption ofpolycyclic aromatic hydrocarbons from water using petroleumcoke-derived porous carbonrdquo Journal of Hazardous Materialsvol 181 no 1-3 pp 1115ndash1120 2010

[11] MMuzic K Sertic-Bionda Z Gomzi S Podolski and S TelenldquoStudy of diesel fuel desulfurization by adsorptionrdquo ChemicalEngineering Research and Design vol 88 no 4 pp 487ndash4952010


[12] M L M Oliveira A A L Miranda C M B M Barbosa C LCavalcante Jr D C S Azevedo and E Rodriguez-CastellonldquoAdsorption of thiophene and toluene on NaY zeolitesexchanged with Ag(I) Ni(II) and Zn(II)rdquo Fuel vol 88 no 10pp 1885ndash1892 2009

[13] L Zou B Han H Yan K L Kasperski Y Xu and L GHepler ldquoEnthalpy of adsorption and isotherms for adsorptionof naphthenic acid onto claysrdquo Journal of Colloid and InterfaceScience vol 190 no 2 pp 472ndash475 1997

[14] F Wang J J-H Haftka T L Sinnige J L M Hermens and WChen ldquoAdsorption of polar nonpolar and substituted aromaticsto colloidal graphene oxide nanoparticlesrdquo Environmental Pol-lution vol 186 pp 226ndash233 2014

[15] J Saien and S Daliri ldquoMass transfer coefficient in liquid-liquidextraction and the influence of aqueous phase pHrdquo Industrial ampEngineering Chemistry Research vol 47 no 1 pp 171ndash175 2008

[16] E Batista S Monnerat K Kato L Stragevitch and A J AMeirelles ldquoLiquid-liquid equilibrium for systems of canola oiloleic acid and short-chain alcoholsrdquo Journal of Chemical ampEngineering Data vol 44 no 6 pp 1360ndash1364 1999

[17] C L Munson and C J Klng ldquoFactors Influencing SolventSelection for Extraction of Ethanol from Aqueous SolutionsrdquoIndustrial amp Engineering Chemistry Process Design and Devel-opment vol 23 no 1 pp 109ndash115 1984

[18] H D Schindler and R E Treybal ldquoContinuous-phase mass-transfer coefficients for liquid extraction in agitated vesselsrdquoAIChE Journal vol 14 no 5 pp 790ndash798 1968

[19] B Schuur W J Jansma J G M Winkelman and H JHeeres ldquoDetermination of the interfacial area of a continuousintegrated mixerseparator (CINC) using a chemical reactionmethodrdquo Chemical Engineering and Processing Process Intensi-fication vol 47 no 9-10 pp 1484ndash1491 2008

[20] R P Verma andMM Sharma ldquoMass transfer in packed liquid-liquid extraction columnsrdquo Chemical Engineering Science vol30 no 3 pp 279ndash292 1975

[21] B Xu W Cai X Liu and X Zhang ldquoMass transfer behavior ofliquid-liquid slug flow in circular cross-section microchannelrdquoChemical Engineering Research and Design vol 91 no 7 pp1203ndash1211 2013

[22] S-S Ashrafmansouri and M Nasr Esfahany ldquoThe influence ofsilica nanoparticles on hydrodynamics and mass transfer inspray liquid-liquid extraction columnrdquo Separation and Purifi-cation Technology vol 151 Article ID 12450 pp 74ndash81 2015

[23] H Bahmanyar and M J Slater ldquoStudies of drop break-upin liquid-liquid systems in a rotating disc contactor Part IConditions of no mass transferrdquo Chemical Engineering amp Tech-nology vol 14 no 2 pp 79ndash89 1991

[24] Z Azizi A Rahbar and H Bahmanyar ldquoInvestigation ofpacking effect onmass transfer coefficient in a single drop liquidextraction columnrdquo Iranian Journal of Chemical Engineeringvol 7 no 4 pp 3ndash11 2010

[25] E Glueckauf and J I Coates ldquoTheory of chromatography PartIVThe influence of incomplete equilibriumon the front bound-ary of chromatograms and on the effectiveness of separationrdquoJournal of the Chemical Society (Resumed) pp 1315ndash1321 1947

[26] MOtero CAGrande andA E Rodrigues ldquoAdsorption of sal-icylic acid onto polymeric adsorbents and activated charcoalrdquoReactive and Functional Polymers vol 60 no 1-3 pp 203ndash2132004

[27] D M Ruthven and S Farooq ldquoAir separation by pressure swingadsorptionrdquo Gas Separation and Purification vol 4 no 3 pp141ndash148 1990

[28] C I Koncsag and A Barbulescu ldquoLiquid-Liquid Extractionwith and without a Chemical Reactionrdquo in Mass Transfer inMultiphase Systems and its Applications Chapter 10 InTechOpen Access Publishers 2011

[29] T Sankarshana S Ilaiah andUVirendraLiquid-Liquid Extrac-tion Comparison in Micro and Macro Systems AIChE AnnualMeeting 2013

[30] F Nosratinia M R Omidkhah D Bastani and A A SaifkordildquoInvestigation of mass transfer coefficient under jetting condi-tions in a liquid-liquid extraction systemrdquo Iranian Journal ofChemistry and Chemical Engineering vol 29 no 1 pp 1ndash122010

[31] C J Geankoplis and A N Hixson ldquoMass Transfer Coefficientsin an Extraction Spray Towerrdquo Industrial amp Engineering Chem-istry vol 42 no 6 pp 1141ndash1151 1950

[32] A Salimi-Khorshidi H Abolghasemi A Khakpay Z Kheir-jooy and M Esmaieli ldquoSpray and packed liquid-liquid extrac-tion columns Drop size and dispersed phase mass transferrdquoAsia-Pacific Journal of Chemical Engineering vol 8 no 6 pp940ndash949 2013

[33] G A Hughmark ldquoLiquid-liquid spray column drop sizeholdup and continuous phase mass transferrdquo Industrial ampEngineering Chemistry Fundamentals vol 6 no 3 pp 408ndash4131967

[34] C L Ruby and J C Elgin ldquoMass transfer between liquid dropsand a continuous phase in a countercurrent fluidized systemliquidliquid extraction in a spray towerrdquo Chemical EngineeringProgress Symposium vol 51 pp 17ndash29 1955

[35] S Kang and B Xing ldquoAdsorption of dicarboxylic acids by clayminerals as examined by in situ ATR-FTIR and ex situ DRIFTrdquoLangmuir vol 23 no 13 pp 7024ndash7031 2007

[36] C D Hatch R V Gough and M A Tblbert ldquoHeterogeneousuptake of the C1 to C4 organic acids on a swelling clay mineralrdquoAtmospheric Chemistry and Physics vol 7 no 16 pp 4445ndash4458 2007

[37] N Kanazawa K Urano N Kokado and Y UrushigawaldquoExchange characteristics of monocarboxylic acids and mono-sulfonic acids onto anion-exchange resinsrdquo Journal of Colloidand Interface Science vol 271 no 1 pp 20ndash27 2004

[38] E Suescun-Mathieu A Bautista-Carrizosa R Sierra LGiraldo and J C Moreno-Pirajan ldquoCarboxylic acid recoveryfrom aqueous solutions by activated carbon produced fromsugarcane bagasserdquoAdsorption vol 20 no 8 pp 935ndash943 2014

[39] Y S Ascı and I M Hasdemir ldquoRemoval of Some CarboxylicAcids from Aqueous Solutions by Hydrogelsrdquo Journal Chemicalamp Engineerind data vol 53 pp 2351ndash2355 2008

[40] A Li Q Zhang G Zhang J Chen Z Fei and F LiuldquoAdsorption of phenolic compounds from aqueous solutions bya water-compatible hypercrosslinked polymeric adsorbentrdquoChemosphere vol 47 no 9 pp 981ndash989 2002

[41] M S Chiou and H Y Li ldquoAdsorption behavior of reactive dyein aqueous solution on chemical cross-linked chitosan beadsrdquoChemosphere vol 50 no 8 pp 1095ndash1105 2003

[42] O Ilgen andH SDulger ldquoRemoval of oleic acid from sunfloweroil on zeolite 13X Kinetics equilibrium and thermodynamicstudiesrdquo Industrial Crops and Products vol 81 pp 66ndash71 2016

[43] R Gayler N W Roberts and H R Pratt ldquoliquid-liquidextraction Part IV A further study of hold-up in packedcolumnsrdquoChemical Engineering Research andDesign vol 31 pp57ndash68 1953

[44] D M Ruthven Principles of Adsorption and Adsorption Pro-cesses John Wiley amp Sons Inc New York NY USA 1984


[45] J J CarberryChemical and Catalytic Reaction Engineering Mc-Graw Hill New York NY USA 1976

[46] N Wakao and T Funazkri ldquoEffect of fluid dispersion coeffi-cients on particle-to-fluid mass transfer coefficients in packedbeds Correlation of sherwood numbersrdquo Chemical EngineeringScience vol 33 no 10 pp 1375ndash1384 1978

[47] M E Sigrist H R Beldomenico E E Tarifa C L Pieck and CR Vera ldquoModelling diffusion and adsorption of As species inFeGAC adsorbent bedsrdquo Journal of Chemical Technology andBiotechnology vol 86 no 10 pp 1256ndash1264 2011

[48] G-H Xiu T Nitta P Li and G Jin ldquoBreakthrough Curves forFixed-Bed Adsorbers Quasi-Lognormal Distribution Approx-imationrdquo AIChE Journal vol 43 no 4 pp 979ndash985 1997

[49] J C Yori S A DrsquoIppolito C L Pieck andC R Vera ldquoDeglycer-olization of biodiesel streams by adsorption over silica bedsrdquoEnergy amp Fuels vol 21 no 1 pp 347ndash353 2007

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018


Active and Passive Electronic Components

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of



Journal ofEngineeringVolume 2018

SensorsJournal of



RotatingMachinery


Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation


Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom


high solvent regeneration rates are hence needed For thesesystems separation by distillation might not be an optionand other principle could be chosen An option that has notreceived attention in the scientific literature or the industrialpractice is that of solvent regeneration by adsorption

The possibilities are analyzed in this work of a processusing liquid-liquid extraction for removing impurities from afeed and adsorption for solvent regenerating the solventThiscombination has not been previously discussed in the openliterature Main features from the kinetic thermodynamicand process point of view are considered and discussed andthemain parameters for designing such a process are writtenEquations are revised for batch and continuous units involv-ing local and global interphase mass transfer coefficients andthe range of practical values of these parameters for these twooperations is discussed

One example involving experimental work is used asproof-of-concept deacidification of vegetable oils by extrac-tion with alcohol coupled to the cyclic adsorption of car-boxylic acids from alcoholic solutions in an activated carbonpacked column The liquid-liquid extraction (LLE) of freefatty acids (FFA) from some plant oils has received someattention lately because fatty acids can be recovered easily forfurther processing the yield of neutral oil being maximum[1ndash3] These are advantages of LLE over caustic refiningin which oil is lost because of reaction with the caustic(saponification) and by emulsification FFAs being convertedto a difficult to handle soap-stock [4] It has also advantagesover physical refining (removal of acids by distillation at250∘C) that produces undesirable changes in color and canalso produce the degradation of valuable nutraceuticals andantioxidants [5ndash7]

Applications of an extractionadsorption process forrefining plant oils could be useful in the biodiesel industrythe food industry and the industry of biodegradable technicaloils (lubricants dielectric transformer oil etc)

2 Materials and Methods

The general procedure was as follows (i) liquid-liquidthermodynamic equilibrium data was got for the solvent-feed-impurity system in the form of partition coefficients(ii) solid-liquid thermodynamic equilibrium isotherms wereobtained for the solvent-adsorbent-impurity system (iii)kinetic parameters for extraction were obtained in the formof global average 119886119870119871 values (minminus1) for a column andstirred tank extractor (iv) adsorption kinetic parametersfor adsorption were obtained in the form of global averagelinear driving force parameter (119870LDF) values (minminus1) (v)tests of adsorbent regeneration by thermal swing were mademeasuring the relevant parameters (vi) simulations wererun for continuous and discontinuous units varying processconditions

21 Materials Edible sunflower oil and oleic acid (SigmandashAldrich 99 grade) were used as a source of triglycerideand fatty acid respectively Acidified solutions of plantoil of variable concentration were obtained by dissolving

weighed amounts of oleic acid in sunflower oil The solventused methanol was supplied by Biopack (Buenos AiresArgentina) The chemical purities were higher than 99 Allcompoundswere usedwithout further purification Activatedcarbon (Filtrasorb Calgon Carbon) was used in this studyThe carbon was conditioned upon receiving by boiling indeionized water for 1 hour then drying in an oven at 110∘Cfor 24 hoursThe activated carbon had a BET specific surfacearea of 972m2 gminus1 a total pore volume of 068mL gminus1 and abulk density of 0502 gmLminus1

22 Liquid Extraction Equilibrium The feed-impurity-sol-vent system was sunflower oil-oleic acid-methanol The oleicacid was distributed between the sunflower oil (oil phase) andmethanol (alcohol phase) The alcohol and oil phases weremostly immiscible Experimental LLE data were obtained ina stirred tank reactor This had an AISI 304 stainless steelvessel with 100ml total volume 40mm of diameter and80mm of length and a magnetic coupling between the motorand the stirrer The tank was heated with a tubular furnaceand the temperature was controlled with a Novus N1100controller The amounts of each component for preparingthe solutions were determined by weighing on an analyticalbalance (Model Shimadzu AUW220D Dual Range Balance00001 g precision) The mixtures were vigorously stirred for4 h and then left to rest for at least 12 h This led to theformation of two clear and transparent phases with a well-defined interface that were sampled for analysis

The oleic acid concentrationwas determined by potentio-metric titration (AOCSMethod Ca 5a-40) with amicroburetThe amount of methanol in the oil phase was determinedby weighing the liquid before and after evaporating thesolution (80∘C 300mmHg vacuum) The amount of oil inthe methanol phase was determined from a mass balance ofthe previous components The analysis was repeated at leastthree times and the average of these readings was taken asthe liquid phase composition

23 Extraction Kinetics Values of the average mass transfercoefficient on the solvent side 119886119870MeOH were calculated fromextraction tests in two kinds of extractors a spray columnand a laboratory stirred tank reactor Coefficients for thecolumn were obtained from single drop experiments usingthe methodology of Azizi et al [24]

In the stirred tank tests the technique of Schindler andTreybal [18] was followed A stirred tank was used thathad the same flange and stirrer as the extraction tests Theinternal volume diameter and length were also the sameas in Section 22 The only difference was that the tank hadtwo additional connections for continuous operation Theflowrates of solvent and feed were controlled with peristalticpumps The oleic acid concentration in the raffinate andextract phases were determined by titration after adequatesettling and formation of two distinct separate phases

24 Adsorption Equilibrium Adsorption isotherms weremeasured in a continuously stirred tank batch reactor Themethod chosen was that of solid addition in which different



Feed

Raffinate

Extract

Solvent



Regenerating stream

q q

x＞

x２ff

yＲＮ

y３ＩＦＰ








00

01

02

03

04

05

06

07

regeneration

yＬＥ

y３ＩＦＰ

(mol

Lminus1)

0 25 7550 125100

t (min)





25 50 75 100 125 1500

t (min)

q0

qＭＮ

q＜ＬＥ

000

100

200

300

400

q(m

ol k

gminus1)
















119902OA = 120572119871OA 1199101 + 119871OA119910 + 119871Oil1199101015840 + 119871MeOH11991010158401015840 (1)


phase mol kgminus1




mol Lminus1









119873OA = 119863120588119901 (120597119902120597119903)surf








unit time and area


adsorbent particle











Table 1 Continued



1119870MeOH

= 1119896MeOH

+ 119898119896Oil

(18)1

119870Oil= 1

119896Oil+ 1

119898119896MeOH= 1

119898119870MeOH(19)

119878ℎ119888 = 0725119878042119888 R119890057119888 (1 minus 120601) (20)119878ℎ119888 = 11989611988811988932

119863119888 (21)1

119870LDF= 119877119891 + 119877119863 = 119903119901

3119896119891 + 119903 211990115120576119863 (17) R119890119888 = 12058811988811988932Vslip








phase


119881119889119910119889119905 = 119882119870LDF (119902eq minus 119902av) (26)

(30)119910 = 1199100 119902 = 1199020 119905 = 0 (27) 119889119910119889119905 = 119886119870MeOH


119889119905 = 119886119870Oil120601 (119909 minus 119909eq) (31)




120597119910120597119905 minus 119863119871 120597 21199101205971199112 + 120597 (119906119910)

120597119911 + 1 minus 120576119861120576119861 120588119901 120597119902120597119905 = 0 (32)

119910 (0 119905) = 1199100 (33)120597119910120597119911 = 0 119911 = 119871 (34) 119911 = int1199092

1199091


(36)

119910 (119911 0) = 0 (35) 119911 = int11991021199101





















Table2Ty

picalvaluesa

ndranges

ofintrinsic

parametersfor

adsorptio

nandextractio

nAC

=activ

ated

carbon

Adsorptio

nparameter

Syste

mRa

nge

Extractio

nparameter

Syste

mRa

nge

119867adim

(m

olkgminus1)(m

olLminus1)


silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

Sulfu

rcom

poun

ds-diesel-A

C[11]

Thioph

ene-naph

tha-NaY

[12]

Naphthenica

cid-toluene-cla

y[13]


l-graph

ene[

14]

Oleicacid-m

ethano

l-AC(th

iswork)

20ndash50

20 1 06 21

23

598 33

119898adim

(m

olLminus1)(m

olLminus1)

Toluene-Ac

eton

e-Water

[15]

Oleicacid-canno

lioil-m

ethano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0967

0705

0133

07415


Oleicacid

over

silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

sulfu

rcom

poun

ds-diesel-A

C[11]

thioph

ene-naph

tha-NaY

[12]

Oleicacid-m

ethano

l-AC(th

iswork)

496

2372

0272

0003

2713

298

119910max

molLminus1

Toluene-aceton

e-water

[15]

Oleicacid-canolao

il-methano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0498

031

00225

0104


Silica[

8]Re

sin[9]

NaY

[12]

Norit[11]

AC[10]

AC(th

iswork)

104ndash105

107

106ndash107

106

107

107


Stirr

edtank

Regt

1000

0[18]

Con

tinuo

usstirr

edtank

[19]

Packed

bed[20]

Slug

flow[21]

Staticmixer

[14]

Spraycolumn(th

iswork)

10ndash250

32ndash130

15ndash3

20ndash4

030ndash300

3



silica[

8]Naphthalene-w

ater-carbo

n[10]

p-am

inob

enzoicacid-w

ater-resin

[9]

thioph

ene-naph

tha-NaY

[12]

sulfu

rcom

poun

ds-diesel-A

C[11]

Oleicacid-m

ethano

l-AC(th

iswork)

00365

004

4006

00167

000

4006

6

aK119871m



n[22]


atercolum

n[23]

Sunfl

ower

oil-o

leicacid-m

ethano

lstirred

tank

(thiswork)

0054

0049

075

lowastPeru

nitv

olum

eof

packed

bedlowastlowastperu

nitv

olum

eof

extractorcoeffi

cientsforthe

continuo

usph


high

solvent-to-feed

syste



newith

equatio

ns(26)

and(29)of


asstransferc


ncontinuo

usph

ases

ide(119870MeO

Hin

exam

pleo

fthisw

ork)C

alculatio

nof119870119871isdo

newith

equatio

ns(18)and(19)of

Table1



30 319 188 621 084240 313 192 493 119050 313 192 313 1610

085

090

095

100

105

110

115

Part

ition

Coe

ffici

ent

35 40 4530 50

Temperature (∘C)



11988932 = sum 11989911989411988931198941198991198941198892119894 (38)

119886 = 612060111988932 (39)









002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)






0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)


2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)




325 Simulation



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia




Feed

Raffinate

Extract

Solvent



Regenerating stream

q q

x＞

x２ff

yＲＮ

y３ＩＦＰ








00

01

02

03

04

05

06

07

regeneration

yＬＥ

y３ＩＦＰ

(mol

Lminus1)

0 25 7550 125100

t (min)





25 50 75 100 125 1500

t (min)

q0

qＭＮ

q＜ＬＥ

000

100

200

300

400

q(m

ol k

gminus1)
















119902OA = 120572119871OA 1199101 + 119871OA119910 + 119871Oil1199101015840 + 119871MeOH11991010158401015840 (1)


phase mol kgminus1




mol Lminus1









119873OA = 119863120588119901 (120597119902120597119903)surf








unit time and area


adsorbent particle











Table 1 Continued



1119870MeOH

= 1119896MeOH

+ 119898119896Oil

(18)1

119870Oil= 1

119896Oil+ 1

119898119896MeOH= 1

119898119870MeOH(19)

119878ℎ119888 = 0725119878042119888 R119890057119888 (1 minus 120601) (20)119878ℎ119888 = 11989611988811988932

119863119888 (21)1

119870LDF= 119877119891 + 119877119863 = 119903119901

3119896119891 + 119903 211990115120576119863 (17) R119890119888 = 12058811988811988932Vslip








phase


119881119889119910119889119905 = 119882119870LDF (119902eq minus 119902av) (26)

(30)119910 = 1199100 119902 = 1199020 119905 = 0 (27) 119889119910119889119905 = 119886119870MeOH


119889119905 = 119886119870Oil120601 (119909 minus 119909eq) (31)




120597119910120597119905 minus 119863119871 120597 21199101205971199112 + 120597 (119906119910)

120597119911 + 1 minus 120576119861120576119861 120588119901 120597119902120597119905 = 0 (32)

119910 (0 119905) = 1199100 (33)120597119910120597119911 = 0 119911 = 119871 (34) 119911 = int1199092

1199091


(36)

119910 (119911 0) = 0 (35) 119911 = int11991021199101





















Table2Ty

picalvaluesa

ndranges

ofintrinsic

parametersfor

adsorptio

nandextractio

nAC

=activ

ated

carbon

Adsorptio

nparameter

Syste

mRa

nge

Extractio

nparameter

Syste

mRa

nge

119867adim

(m

olkgminus1)(m

olLminus1)


silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

Sulfu

rcom

poun

ds-diesel-A

C[11]

Thioph

ene-naph

tha-NaY

[12]

Naphthenica

cid-toluene-cla

y[13]


l-graph

ene[

14]

Oleicacid-m

ethano

l-AC(th

iswork)

20ndash50

20 1 06 21

23

598 33

119898adim

(m

olLminus1)(m

olLminus1)

Toluene-Ac

eton

e-Water

[15]

Oleicacid-canno

lioil-m

ethano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0967

0705

0133

07415


Oleicacid

over

silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

sulfu

rcom

poun

ds-diesel-A

C[11]

thioph

ene-naph

tha-NaY

[12]

Oleicacid-m

ethano

l-AC(th

iswork)

496

2372

0272

0003

2713

298

119910max

molLminus1

Toluene-aceton

e-water

[15]

Oleicacid-canolao

il-methano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0498

031

00225

0104


Silica[

8]Re

sin[9]

NaY

[12]

Norit[11]

AC[10]

AC(th

iswork)

104ndash105

107

106ndash107

106

107

107


Stirr

edtank

Regt

1000

0[18]

Con

tinuo

usstirr

edtank

[19]

Packed

bed[20]

Slug

flow[21]

Staticmixer

[14]

Spraycolumn(th

iswork)

10ndash250

32ndash130

15ndash3

20ndash4

030ndash300

3



silica[

8]Naphthalene-w

ater-carbo

n[10]

p-am

inob

enzoicacid-w

ater-resin

[9]

thioph

ene-naph

tha-NaY

[12]

sulfu

rcom

poun

ds-diesel-A

C[11]

Oleicacid-m

ethano

l-AC(th

iswork)

00365

004

4006

00167

000

4006

6

aK119871m



n[22]


atercolum

n[23]

Sunfl

ower

oil-o

leicacid-m

ethano

lstirred

tank

(thiswork)

0054

0049

075

lowastPeru

nitv

olum

eof

packed

bedlowastlowastperu

nitv

olum

eof

extractorcoeffi

cientsforthe

continuo

usph


high

solvent-to-feed

syste



newith

equatio

ns(26)

and(29)of


asstransferc


ncontinuo

usph

ases

ide(119870MeO

Hin

exam

pleo

fthisw

ork)C

alculatio

nof119870119871isdo

newith

equatio

ns(18)and(19)of

Table1



30 319 188 621 084240 313 192 493 119050 313 192 313 1610

085

090

095

100

105

110

115

Part

ition

Coe

ffici

ent

35 40 4530 50

Temperature (∘C)



11988932 = sum 11989911989411988931198941198991198941198892119894 (38)

119886 = 612060111988932 (39)









002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)






0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)


2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)




325 Simulation



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



25 50 75 100 125 1500

t (min)

q0

qＭＮ

q＜ＬＥ

000

100

200

300

400

q(m

ol k

gminus1)
















119902OA = 120572119871OA 1199101 + 119871OA119910 + 119871Oil1199101015840 + 119871MeOH11991010158401015840 (1)


phase mol kgminus1




mol Lminus1









119873OA = 119863120588119901 (120597119902120597119903)surf








unit time and area


adsorbent particle











Table 1 Continued



1119870MeOH

= 1119896MeOH

+ 119898119896Oil

(18)1

119870Oil= 1

119896Oil+ 1

119898119896MeOH= 1

119898119870MeOH(19)

119878ℎ119888 = 0725119878042119888 R119890057119888 (1 minus 120601) (20)119878ℎ119888 = 11989611988811988932

119863119888 (21)1

119870LDF= 119877119891 + 119877119863 = 119903119901

3119896119891 + 119903 211990115120576119863 (17) R119890119888 = 12058811988811988932Vslip








phase


119881119889119910119889119905 = 119882119870LDF (119902eq minus 119902av) (26)

(30)119910 = 1199100 119902 = 1199020 119905 = 0 (27) 119889119910119889119905 = 119886119870MeOH


119889119905 = 119886119870Oil120601 (119909 minus 119909eq) (31)




120597119910120597119905 minus 119863119871 120597 21199101205971199112 + 120597 (119906119910)

120597119911 + 1 minus 120576119861120576119861 120588119901 120597119902120597119905 = 0 (32)

119910 (0 119905) = 1199100 (33)120597119910120597119911 = 0 119911 = 119871 (34) 119911 = int1199092

1199091


(36)

119910 (119911 0) = 0 (35) 119911 = int11991021199101





















Table2Ty

picalvaluesa

ndranges

ofintrinsic

parametersfor

adsorptio

nandextractio

nAC

=activ

ated

carbon

Adsorptio

nparameter

Syste

mRa

nge

Extractio

nparameter

Syste

mRa

nge

119867adim

(m

olkgminus1)(m

olLminus1)


silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

Sulfu

rcom

poun

ds-diesel-A

C[11]

Thioph

ene-naph

tha-NaY

[12]

Naphthenica

cid-toluene-cla

y[13]


l-graph

ene[

14]

Oleicacid-m

ethano

l-AC(th

iswork)

20ndash50

20 1 06 21

23

598 33

119898adim

(m

olLminus1)(m

olLminus1)

Toluene-Ac

eton

e-Water

[15]

Oleicacid-canno

lioil-m

ethano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0967

0705

0133

07415


Oleicacid

over

silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

sulfu

rcom

poun

ds-diesel-A

C[11]

thioph

ene-naph

tha-NaY

[12]

Oleicacid-m

ethano

l-AC(th

iswork)

496

2372

0272

0003

2713

298

119910max

molLminus1

Toluene-aceton

e-water

[15]

Oleicacid-canolao

il-methano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0498

031

00225

0104


Silica[

8]Re

sin[9]

NaY

[12]

Norit[11]

AC[10]

AC(th

iswork)

104ndash105

107

106ndash107

106

107

107


Stirr

edtank

Regt

1000

0[18]

Con

tinuo

usstirr

edtank

[19]

Packed

bed[20]

Slug

flow[21]

Staticmixer

[14]

Spraycolumn(th

iswork)

10ndash250

32ndash130

15ndash3

20ndash4

030ndash300

3



silica[

8]Naphthalene-w

ater-carbo

n[10]

p-am

inob

enzoicacid-w

ater-resin

[9]

thioph

ene-naph

tha-NaY

[12]

sulfu

rcom

poun

ds-diesel-A

C[11]

Oleicacid-m

ethano

l-AC(th

iswork)

00365

004

4006

00167

000

4006

6

aK119871m



n[22]


atercolum

n[23]

Sunfl

ower

oil-o

leicacid-m

ethano

lstirred

tank

(thiswork)

0054

0049

075

lowastPeru

nitv

olum

eof

packed

bedlowastlowastperu

nitv

olum

eof

extractorcoeffi

cientsforthe

continuo

usph


high

solvent-to-feed

syste



newith

equatio

ns(26)

and(29)of


asstransferc


ncontinuo

usph

ases

ide(119870MeO

Hin

exam

pleo

fthisw

ork)C

alculatio

nof119870119871isdo

newith

equatio

ns(18)and(19)of

Table1



30 319 188 621 084240 313 192 493 119050 313 192 313 1610

085

090

095

100

105

110

115

Part

ition

Coe

ffici

ent

35 40 4530 50

Temperature (∘C)



11988932 = sum 11989911989411988931198941198991198941198892119894 (38)

119886 = 612060111988932 (39)









002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)






0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)


2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)




325 Simulation



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia






119902OA = 120572119871OA 1199101 + 119871OA119910 + 119871Oil1199101015840 + 119871MeOH11991010158401015840 (1)


phase mol kgminus1




mol Lminus1









119873OA = 119863120588119901 (120597119902120597119903)surf








unit time and area


adsorbent particle











Table 1 Continued



1119870MeOH

= 1119896MeOH

+ 119898119896Oil

(18)1

119870Oil= 1

119896Oil+ 1

119898119896MeOH= 1

119898119870MeOH(19)

119878ℎ119888 = 0725119878042119888 R119890057119888 (1 minus 120601) (20)119878ℎ119888 = 11989611988811988932

119863119888 (21)1

119870LDF= 119877119891 + 119877119863 = 119903119901

3119896119891 + 119903 211990115120576119863 (17) R119890119888 = 12058811988811988932Vslip








phase


119881119889119910119889119905 = 119882119870LDF (119902eq minus 119902av) (26)

(30)119910 = 1199100 119902 = 1199020 119905 = 0 (27) 119889119910119889119905 = 119886119870MeOH


119889119905 = 119886119870Oil120601 (119909 minus 119909eq) (31)




120597119910120597119905 minus 119863119871 120597 21199101205971199112 + 120597 (119906119910)

120597119911 + 1 minus 120576119861120576119861 120588119901 120597119902120597119905 = 0 (32)

119910 (0 119905) = 1199100 (33)120597119910120597119911 = 0 119911 = 119871 (34) 119911 = int1199092

1199091


(36)

119910 (119911 0) = 0 (35) 119911 = int11991021199101





















Table2Ty

picalvaluesa

ndranges

ofintrinsic

parametersfor

adsorptio

nandextractio

nAC

=activ

ated

carbon

Adsorptio

nparameter

Syste

mRa

nge

Extractio

nparameter

Syste

mRa

nge

119867adim

(m

olkgminus1)(m

olLminus1)


silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

Sulfu

rcom

poun

ds-diesel-A

C[11]

Thioph

ene-naph

tha-NaY

[12]

Naphthenica

cid-toluene-cla

y[13]


l-graph

ene[

14]

Oleicacid-m

ethano

l-AC(th

iswork)

20ndash50

20 1 06 21

23

598 33

119898adim

(m

olLminus1)(m

olLminus1)

Toluene-Ac

eton

e-Water

[15]

Oleicacid-canno

lioil-m

ethano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0967

0705

0133

07415


Oleicacid

over

silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

sulfu

rcom

poun

ds-diesel-A

C[11]

thioph

ene-naph

tha-NaY

[12]

Oleicacid-m

ethano

l-AC(th

iswork)

496

2372

0272

0003

2713

298

119910max

molLminus1

Toluene-aceton

e-water

[15]

Oleicacid-canolao

il-methano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0498

031

00225

0104


Silica[

8]Re

sin[9]

NaY

[12]

Norit[11]

AC[10]

AC(th

iswork)

104ndash105

107

106ndash107

106

107

107


Stirr

edtank

Regt

1000

0[18]

Con

tinuo

usstirr

edtank

[19]

Packed

bed[20]

Slug

flow[21]

Staticmixer

[14]

Spraycolumn(th

iswork)

10ndash250

32ndash130

15ndash3

20ndash4

030ndash300

3



silica[

8]Naphthalene-w

ater-carbo

n[10]

p-am

inob

enzoicacid-w

ater-resin

[9]

thioph

ene-naph

tha-NaY

[12]

sulfu

rcom

poun

ds-diesel-A

C[11]

Oleicacid-m

ethano

l-AC(th

iswork)

00365

004

4006

00167

000

4006

6

aK119871m



n[22]


atercolum

n[23]

Sunfl

ower

oil-o

leicacid-m

ethano

lstirred

tank

(thiswork)

0054

0049

075

lowastPeru

nitv

olum

eof

packed

bedlowastlowastperu

nitv

olum

eof

extractorcoeffi

cientsforthe

continuo

usph


high

solvent-to-feed

syste



newith

equatio

ns(26)

and(29)of


asstransferc


ncontinuo

usph

ases

ide(119870MeO

Hin

exam

pleo

fthisw

ork)C

alculatio

nof119870119871isdo

newith

equatio

ns(18)and(19)of

Table1



30 319 188 621 084240 313 192 493 119050 313 192 313 1610

085

090

095

100

105

110

115

Part

ition

Coe

ffici

ent

35 40 4530 50

Temperature (∘C)



11988932 = sum 11989911989411988931198941198991198941198892119894 (38)

119886 = 612060111988932 (39)









002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)






0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)


2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)




325 Simulation



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



Table 1 Continued



1119870MeOH

= 1119896MeOH

+ 119898119896Oil

(18)1

119870Oil= 1

119896Oil+ 1

119898119896MeOH= 1

119898119870MeOH(19)

119878ℎ119888 = 0725119878042119888 R119890057119888 (1 minus 120601) (20)119878ℎ119888 = 11989611988811988932

119863119888 (21)1

119870LDF= 119877119891 + 119877119863 = 119903119901

3119896119891 + 119903 211990115120576119863 (17) R119890119888 = 12058811988811988932Vslip








phase


119881119889119910119889119905 = 119882119870LDF (119902eq minus 119902av) (26)

(30)119910 = 1199100 119902 = 1199020 119905 = 0 (27) 119889119910119889119905 = 119886119870MeOH


119889119905 = 119886119870Oil120601 (119909 minus 119909eq) (31)




120597119910120597119905 minus 119863119871 120597 21199101205971199112 + 120597 (119906119910)

120597119911 + 1 minus 120576119861120576119861 120588119901 120597119902120597119905 = 0 (32)

119910 (0 119905) = 1199100 (33)120597119910120597119911 = 0 119911 = 119871 (34) 119911 = int1199092

1199091


(36)

119910 (119911 0) = 0 (35) 119911 = int11991021199101





















Table2Ty

picalvaluesa

ndranges

ofintrinsic

parametersfor

adsorptio

nandextractio

nAC

=activ

ated

carbon

Adsorptio

nparameter

Syste

mRa

nge

Extractio

nparameter

Syste

mRa

nge

119867adim

(m

olkgminus1)(m

olLminus1)


silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

Sulfu

rcom

poun

ds-diesel-A

C[11]

Thioph

ene-naph

tha-NaY

[12]

Naphthenica

cid-toluene-cla

y[13]


l-graph

ene[

14]

Oleicacid-m

ethano

l-AC(th

iswork)

20ndash50

20 1 06 21

23

598 33

119898adim

(m

olLminus1)(m

olLminus1)

Toluene-Ac

eton

e-Water

[15]

Oleicacid-canno

lioil-m

ethano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0967

0705

0133

07415


Oleicacid

over

silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

sulfu

rcom

poun

ds-diesel-A

C[11]

thioph

ene-naph

tha-NaY

[12]

Oleicacid-m

ethano

l-AC(th

iswork)

496

2372

0272

0003

2713

298

119910max

molLminus1

Toluene-aceton

e-water

[15]

Oleicacid-canolao

il-methano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0498

031

00225

0104


Silica[

8]Re

sin[9]

NaY

[12]

Norit[11]

AC[10]

AC(th

iswork)

104ndash105

107

106ndash107

106

107

107


Stirr

edtank

Regt

1000

0[18]

Con

tinuo

usstirr

edtank

[19]

Packed

bed[20]

Slug

flow[21]

Staticmixer

[14]

Spraycolumn(th

iswork)

10ndash250

32ndash130

15ndash3

20ndash4

030ndash300

3



silica[

8]Naphthalene-w

ater-carbo

n[10]

p-am

inob

enzoicacid-w

ater-resin

[9]

thioph

ene-naph

tha-NaY

[12]

sulfu

rcom

poun

ds-diesel-A

C[11]

Oleicacid-m

ethano

l-AC(th

iswork)

00365

004

4006

00167

000

4006

6

aK119871m



n[22]


atercolum

n[23]

Sunfl

ower

oil-o

leicacid-m

ethano

lstirred

tank

(thiswork)

0054

0049

075

lowastPeru

nitv

olum

eof

packed

bedlowastlowastperu

nitv

olum

eof

extractorcoeffi

cientsforthe

continuo

usph


high

solvent-to-feed

syste



newith

equatio

ns(26)

and(29)of


asstransferc


ncontinuo

usph

ases

ide(119870MeO

Hin

exam

pleo

fthisw

ork)C

alculatio

nof119870119871isdo

newith

equatio

ns(18)and(19)of

Table1



30 319 188 621 084240 313 192 493 119050 313 192 313 1610

085

090

095

100

105

110

115

Part

ition

Coe

ffici

ent

35 40 4530 50

Temperature (∘C)



11988932 = sum 11989911989411988931198941198991198941198892119894 (38)

119886 = 612060111988932 (39)









002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)






0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)


2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)




325 Simulation



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia
















Table2Ty

picalvaluesa

ndranges

ofintrinsic

parametersfor

adsorptio

nandextractio

nAC

=activ

ated

carbon

Adsorptio

nparameter

Syste

mRa

nge

Extractio

nparameter

Syste

mRa

nge

119867adim

(m

olkgminus1)(m

olLminus1)


silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

Sulfu

rcom

poun

ds-diesel-A

C[11]

Thioph

ene-naph

tha-NaY

[12]

Naphthenica

cid-toluene-cla

y[13]


l-graph

ene[

14]

Oleicacid-m

ethano

l-AC(th

iswork)

20ndash50

20 1 06 21

23

598 33

119898adim

(m

olLminus1)(m

olLminus1)

Toluene-Ac

eton

e-Water

[15]

Oleicacid-canno

lioil-m

ethano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0967

0705

0133

07415


Oleicacid

over

silica[

8]p-am

inob

enzoicacid-w

ater-resin

[9]

Naphthalene-w

ater-carbo

n[10]

sulfu

rcom

poun

ds-diesel-A

C[11]

thioph

ene-naph

tha-NaY

[12]

Oleicacid-m

ethano

l-AC(th

iswork)

496

2372

0272

0003

2713

298

119910max

molLminus1

Toluene-aceton

e-water

[15]

Oleicacid-canolao

il-methano

l[16]

Methylene

chlorid

e-ethano

l-water

[17]

Sunfl

ower

oil-o

leic-acid-methano

l(thiswork)

0498

031

00225

0104


Silica[

8]Re

sin[9]

NaY

[12]

Norit[11]

AC[10]

AC(th

iswork)

104ndash105

107

106ndash107

106

107

107


Stirr

edtank

Regt

1000

0[18]

Con

tinuo

usstirr

edtank

[19]

Packed

bed[20]

Slug

flow[21]

Staticmixer

[14]

Spraycolumn(th

iswork)

10ndash250

32ndash130

15ndash3

20ndash4

030ndash300

3



silica[

8]Naphthalene-w

ater-carbo

n[10]

p-am

inob

enzoicacid-w

ater-resin

[9]

thioph

ene-naph

tha-NaY

[12]

sulfu

rcom

poun

ds-diesel-A

C[11]

Oleicacid-m

ethano

l-AC(th

iswork)

00365

004

4006

00167

000

4006

6

aK119871m



n[22]


atercolum

n[23]

Sunfl

ower

oil-o

leicacid-m

ethano

lstirred

tank

(thiswork)

0054

0049

075

lowastPeru

nitv

olum

eof

packed

bedlowastlowastperu

nitv

olum

eof

extractorcoeffi

cientsforthe

continuo

usph


high

solvent-to-feed

syste



newith

equatio

ns(26)

and(29)of


asstransferc


ncontinuo

usph

ases

ide(119870MeO

Hin

exam

pleo

fthisw

ork)C

alculatio

nof119870119871isdo

newith

equatio

ns(18)and(19)of

Table1



30 319 188 621 084240 313 192 493 119050 313 192 313 1610

085

090

095

100

105

110

115

Part

ition

Coe

ffici

ent

35 40 4530 50

Temperature (∘C)



11988932 = sum 11989911989411988931198941198991198941198892119894 (38)

119886 = 612060111988932 (39)









002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)






0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)


2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)




325 Simulation



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia




30 319 188 621 084240 313 192 493 119050 313 192 313 1610

085

090

095

100

105

110

115

Part

ition

Coe

ffici

ent

35 40 4530 50

Temperature (∘C)



11988932 = sum 11989911989411988931198941198991198941198892119894 (38)

119886 = 612060111988932 (39)









002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)






0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)


2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)




325 Simulation



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



002 004 006 008 010000

ye (mol Lminus1)

0

1

2

3

4

5qe

(mol

kgminus

1)






0

10

20

30

40

50

60

y(m

mol

Lminus1)

0

200

400

600

800

1000

q(m

mol

kgminus

1)

0 8020 40 60 120100

t (min)


2520 35 40 45 5030

Temperature (∘C)

0

100

200

300

400

500t (

min

)




325 Simulation



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



LLE unit SLA A unit

Feed

Raffinate

Extract


Regenerating stream

Solvent


00 10 15 20 2505 30

z (m)

0

20

40

60

80

100

120

140

160

x(m

mol

Lminus1)

0

1

2

3

4

5

6

7

8y

(mm

ol L

minus1)




119906119889120601 + 119906119888


00

02

04

06

08

10

12

14

16

y(m

mol

Lminus1)

0

25

50

75

100

q(m

mol

kgminus

1)

0 2010 40 50 6030

t (min)






0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



0

15

30

45

60

75

90

t (m

in)

0 200 300 400 500 600100

x＞ (mmol Lminus1)

00

05

10

15

20

25

30

35

40

x２ff

(mm

ol L

minus1)






0

10

20

30

40

50

60

70

t (m

in)

1 2 3 40

dＪＦＦＮ (mm)











SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia







SLA unit

Recycle

Mixer Settler Tank

Extract

Raffinate

Solvent

Feed







0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia



0

20

40

60

80

100

120

140x

(mm

ol L

minus1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)

1 2 3 4 50

t (min)


0 20 30 40 50 6010

t (min)

0

5

10

15

20

25

q(m

mol

kgminus

1)

0

1

2

3

4

5

6

7

8

y(m

mol

Lminus1)





8060 120100 140

x＞ (mmol Lminus1)

012345678910

x２ff

(mm

ol L

minus1)

0

10

20

30

40

50

60

70

t (m

in)





4 Conclusions












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia












Acknowledgments


References






















































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia












































RoboticsJournal of




VLSI Design



Shock and Vibration







Journal of



Volume 2018



Volume 2018


Journal of




SensorsJournal of



RotatingMachinery





Propagation






Hindawi


Advances in

Multimedia


researcharticle coupling solvent extraction units to...

Documents