Liquid-liquid equilibria for ternary systems acetic acid + n-butyl acetate + hydrocarbons at 293.15 K

Any correspondence concerning this service should be sent to the repository administrator:

staff-oatao@inp-toulouse.fr

Identification number: DOI : 10.1016/j.fluid.2013.07.032

Official URL:

http://dx.doi.org/10.1016/j.fluid.2013.07.032

This is an author-deposited version published in:

http://oatao.univ-toulouse.fr/

Eprints ID: 12000

To cite this version:

Richard, Romain and Ferrando, Nicolas and Jacquin, Marc Liquid-liquid

equilibria for ternary systems acetic acid + n-butyl acetate + hydrocarbons at

293.15 K. (2013) Fluid Phase Equilibria, vol. 356 . pp. 264-270. ISSN

0378-3812

O

pen

A

rchive

T

oulouse

A

rchive

O

uverte (

OATAO

)

OATAO is an open access repository that collects the work of Toulouse researchers and

makes it freely available over the web where possible.

Liquid–liquid

equilibria

for

ternary

systems

acetic

acid

+

n-butyl

acetate

+

hydrocarbons

at

293.15

K

Romain

Richard

a

_,

_Nicolas

_Ferrando

b

_,

_Marc

_Jacquin

a,∗ a_{IFPEnergiesnouvelles,RondPointdel’échangeurdeSolaize,BP3,69360Solaize,France} b_{IFPEnergiesnouvelles,1-4AvenuedeBois-Préau,92852Rueil-Malmaison,France}

Keywords:

Liquid–liquidequilibriumdata Ternarymixture

Hydrocarbons UNIQUACmodel

a

b

s

t

r

a

c

t

Liquid–liquid equilibrium data for acetic acid+n-butyl acetate+hydrocarbons ternary systems at T=293.15Karereportedinthiswork.Theeffectofhydrocarbonchainlengthonliquid–liquidequilibrium isdeterminedanddiscussed.Aliphatichydrocarbonssuchashexadecane,dodecaneanddecanewere particularlyinvestigated.Theorganicchemicals(estersandhydrocarbons)werequantifiedbygas chro-matographyusingaflameionizationdetectorwhileaceticacidwasquantifiedbytitrationwithsodium hydroxide.Experimentaltie-linedatafortheternarymixtureswerecorrelatedusingOthmer–Tobias, BachmanandHandcorrelationsinordertoshowthereliabilityoftheexperimentalresults.Finally,these experimentaldatawerecorrelatedwiththeUNIQUACmodel.Itappearedthatthismodelprovidesagood correlationofthesolubilitycurvewiththesethreehydrocarbons.

1. Introduction

Lowmolecularweightestersaremostcommonlysynthesized

bydirectesterificationofcarboxylicacidswithalcoholsin

pres-enceofacidcatalystsbyeithera batchor acontinuousprocess

[1–4].Estersareessentialforthefragranceandflavouringindustry.

n-butylacetate(BA)isawidelyknownesterusedasasolventin

theproductionoflacquersforexample,butmorecommonlyasa

syntheticappleflavouringusedinfoodindustry[5].Productionof

BAincreasedinthelastdecadebecauseitslowtoxicityandlow

environmentalimpactcomparedwithotheresters[6].

Industrially,themostcommonlyusedseparationprocessesfor

suchequilibrium-limitedreversiblereactionsis reactive

distilla-tion[7,8].Nevertheless,inthecaseofbutylacetate,distillationis

high-energyconsumingduetolowrelativevolatilitiesbetweenthe

carboxylicacidandtheester.Indeed,theboilingpointsofacetic

acid(118◦_{C)andbutylacetate(126}◦_{C)arerelativelyclose,and}

thusseparationbydistillationrequireahighernumberofstageand

refluxratioincomparisontootheresters.Withtheincreasingprice

ofenergy,alternativeseparationtechniqueshavetobeconsidered,

inordertoreducetheenergyconsumptionassociatedwiththe

sep-arationofthecomponents.Inthisstudy,wefocusonliquid–liquid

extractionoftheprincipalproduct,n-butylacetate(BA)fromthe

remainingreactant,aceticacid(AA),thesemixturebeingasingle

homogeneousphase.Thistechniqueisofconsiderableeconomic

importanceinthechemicalindustryandmaybeconsideredeither

attheendofthereactionorduringthereactioninordertoshiftthe

equilibrium.Thenatureofsolventnaturallyinfluencesthe

equilib-riumcharacteristicsofBAextractionfromAAsolutions.Previous

worksgivesomedataonphaseequilibriumforsystemssuchas

aceticacidandbutan-1-ol[9]orternarysystemswithaceticacid,

waterandesters[10]oralcohols[11–13].Inthisstudy,aliphatic

hydrocarbons(HCs),fromhexane(C6)tohexadecane(C16),were

usedasorganicsolvent.Someliquid–liquidequilibrium(LLE)data

forsuchternarysystemsAA–BA–HCwereobtainedatT=293.15K,

underatmosphericpressure.

Firstofall,mutualsolubilityofAAandHCswerestudiedinorder

todeterminesuitablesolventsforliquid–liquidextractionofBA

inexcessofAA.LLEresultsfortheternarysystemsAA+BA+HC

presentinga biphasicareaaregivenin this work.Moreover,to

showtheconsistencyofourexperimentalresults,tie-linedatawere

correlatedusingOthmer–Tobias,BachmanandHandcorrelations.

Finally,experimentalternarydiagramswerecomparedwithdata

calculatedwiththeUNIQUACmodel.Thismodelmayprovidea

goodcorrelationofthesolubilitycurvewiththehydrocarbonsof

interest.

2. Experimental

2.1. Materials

Allchemicals(n-butylacetate,glacialaceticacid, hexaneC6,

octaneC8,decaneC10,dodecaneC12andhexadecaneC16)were

purchasedfromSigma–Aldrich.Theywereofguaranteedreagent

Table1

Compositionsofinitialmixtures,experimentaltie-lines,solutedistributionratios,Dandselectivity,Sforternarysystems[AA+BA+HC]atT=293.15K.

Initialcomposition AA-richphase HC-richphase D S

wAA wBA wHC wAA wBA wHC wAA wBA wHC Octane 0.500 0.000 0.500 0.747 0.000 0.253 0.266 0.000 0.734 – – 0.495 0.010 0.495 – – – – – – – – Decane 0.500 0.000 0.500 0.846 0.000 0.154 0.217 0.000 0.783 – – 0.490 0.020 0.490 0.811 0.030 0.159 0.252 0.017 0.731 0.577 1.714 0.480 0.040 0.480 0.748 0.058 0.194 0.288 0.038 0.674 0.656 1.626 0.475 0.050 0.475 0.713 0.069 0.218 0.316 0.047 0.637 0.688 1.768 0.470 0.060 0.470 0.655 0.081 0.264 0.359 0.064 0.577 0.800 1.462 0.462 0.075 0.463 – – – – – – – – Dodecane 0.500 0.000 0.500 0.881 0.000 0.119 0.153 0.000 0.847 – – 0.487 0.025 0.488 0.855 0.044 0.101 0.175 0.020 0.805 0.451 2.206 0.474 0.050 0.476 0.806 0.079 0.115 0.192 0.039 0.769 0.490 2.055 0.462 0.075 0.463 0.761 0.105 0.134 0.219 0.061 0.720 0.581 2.018 0.450 0.100 0.450 0.699 0.131 0.170 0.258 0.091 0.651 0.696 1.889 0.437 0.125 0.438 0.594 0.155 0.251 0.333 0.120 0.547 0.775 1.383 0.425 0.150 0.425 – – – – – – – – Hexadecane 0.500 0.000 0.500 0.948 0.000 0.052 0.098 0.000 0.902 – – 0.475 0.050 0.475 0.872 0.083 0.045 0.110 0.030 0.860 0.357 2.821 0.450 0.100 0.450 0.765 0.167 0.068 0.128 0.065 0.807 0.391 2.328 0.425 0.150 0.425 0.705 0.219 0.076 0.129 0.100 0.771 0.460 2.520 0.400 0.200 0.400 0.583 0.272 0.145 0.187 0.167 0.646 0.614 1.913 0.387 0.225 0.388 0.521 0.290 0.189 0.222 0.208 0.570 0.716 1.683 0.375 0.250 0.375 0.433 0.289 0.278 0.292 0.239 0.469 0.826 1.222 0.362 0.275 0.363 – – – – – – – –

gradeandtheirpuritiesarereportedbythesuppliertobehigher

than99%.Theywereusedwithoutanyfurtherpurificationasno

impuritiesweredetectedusinggaschromatographywithaflame

ionizationdetector(GC-FID).Theamountofwaterinaceticacid

wasdeterminedusingaKarl–Fisherapparatus(831KFCoulometer,

Metrohm),thisvaluebeingof396+/−54ppm.

2.2. Procedure

Ternary mixtures ofknown composition(AA+BA+HC)were

preciselypreparedinclosedvialsandmixedduringatleast3hat

293.15K.Theaccuracyofthetemperatureis1K.Thetotalweight

of each sample was approximately 40g. Whenthe

thermody-namicequilibriumwasachieved,thesystemseparatedintotwo

liquidphasesthatbecametransparentwithawell-defined

inter-face.Afterthisdecantation,thelowerphasecontainingmainlyAA

andtheremainingHC-richupperphaseweresuccessivelycollected

andweighted.Afterseparation,samplesofbothphaseswerestill

transparentandwerecarefullyanalyzedtodeterminetheir

com-positionsinordertobuildtheLLEtie-lines.

2.3. Analysismethods

Ternarymixtureswereanalyzedbygaschromatography(GC)

usinganAgilentTechnologiesapparatus(6890NNetworkGC

Sys-tem) coupled to a flame ionization detector (FID) in order to

determineBAandHCamounts.Separationwascarriedoutwitha

capillarycolumn(HP-FFAPcolumn,50m,0.32mm,0.50mm)from

AgilentJ&WGCColumns.Thechromatographwasequippedwith

anautomaticsplitinjectorandtheinjections(0.5mL)were

per-formedwithasplitratioof100.Thecarriergaswasheliumandthe

columnheadpressurewasadjustedto15psi.Injectortemperature

was220◦_{C.Temperatureintheovenwasheld14minat90}◦_C,then

rampedto200◦_Cat10◦_Cmin−1_{andfinallyheld5minat200}◦_C.

Thetotalrunningtimewas30min.Thetemperatureofthe

detec-tor(FID)was250◦_{C.Eachsamplewasanalyzedthreetimesand}

thevaluesofmassfractionsgiveninthispaperaretheaverageof

thethreeinjections.Theuncertaintyisu(w)=0.001,which

corre-spondstothemaximumstandarddeviationcalculated.AAcould

alsobequantifiedbyGC-FIDbutresultsarelessreliable.

Thatiswhyanautomatictitrator(751GPDTitrino,Metrohm)

wasusedtodeterminethequantityofAAineachsample.We

veri-fiedthatnohydrolysisofBAoccursduringthetitrationwithsodium

hydroxide(0.1M),withstandardmixturesofAAandBAoverthe

wholerangeofcomposition.ThepHofthesolutionwasmeasured

throughout thetitration thankstoanelectrode, accuracybeing

moreimportantwithanelectrodethananindicator.Theendpoint

correspondstoasuddenchangeinthemeasuredpH,andthenthe

Fig.1. Plotofmutualsolubilitiesofaceticacidwithdifferenthydrocarbonsfrom octanetohexadecaneat293.15K(markedareaisbiphasic).

Fig.2.TernarydiagramforLLEof[AA+BA+decane]atT=293.15K;(··· ···)UNIQUACtie-linedata,(—)UNIQUACsolubilitycurve,(-d-)experimentaltie-linedata.

equivalencepointallowsthedeterminationofAAamountinthe

sample.

3. Resultsanddiscussion

3.1. LLEexperimentalresults

First ofall, miscibilityof acetic acidwithhydrocarbonswas

determined.ItwasnoticedthatAAandhexaneweremiscibleinall

proportionswhereasAAandalltheotherhydrocarbonspresentan

incompletemutualsolubility.TheresultsarereportedinTable1

(lineswitha butylacetatemassfractionwBAequaltozero) and

representedinFig.1.Itisexperimentallyshownthatthebiphasic

areaislargerwhenthehydrocarbonchainlengthincreases.

Con-cerningthesystemAA+BA+C8,asit canbenoticedinTable1,

withonly1%ofBA(wBA=0.01),themixtureismonophasic,which

meansthatthebiphasicareaofthissystemisverysmall.Thatis

whywedidnotrepresentaphasediagramofthissystem.TheLLE

diagramsfortheotherternarysystemswitheachhydrocarbonare

plottedinFigs.2–4.Because(HC+AA)isaliquidpairthatispartially

miscibleandthetwootherliquidpairs(BA+AA)and(BA+HC)are

completelymiscible,theternarysystemsbehaveastype-ILLE[14].

Inordertoevaluatetheextractingcapabilityofthehydrocarbon

solventfortheseparationoftheesterfromacidicsolutions,the

selectivity(S) and solute distributionratio (D) were calculated

fromexperimentaldataaccordingtofollowingEqs.(1)–(2):

S=w HC BA×wAAAA wAA BA×wHCAA (1) D=w HC BA wAA BA (2)

wherewisthemassfraction,subscriptsBAandAArefer

respec-tively to n-butyl acetate and acetic acid, and superscripts HC

Fig.4.TernarydiagramforLLEof[AA+BA+hexadecane]atT=293.15K;(···△···)experimentalsolubilitycurve,(—)UNIQUACsolubilitycurve,(-N-)experimentaltie-line data.

and AA represent thehydrocarbon-rich phase and theAA-rich

phase, respectively. Values of S and Dare reportedin Table 1.

Theseparametersarewidelyusedforevaluatingthesolvent(HC)

efficiencyinunitoperationssuchasliquid–liquidextraction.The

distribution coefficient is related to the extracting capacity of

BAintheHCanddeterminestheamountofHCneededforthe

extractionprocess,whiletheselectivityisrelatedtothenumber

ofstagesneededintheseparationprocess.Theoretically,higher

selectivitycorrespondstofewerstagesrequiredforagiven

sep-arationprocessandahigherdistributioncoefficientcorresponds

toaloweramountofsolventneededforagivenseparation,and

consequentlyasmallerapparatusandloweroperatingcosts.The

selectivityasafunctionofsolutedistributionratiosforthethree

studied ternary systems is plotted for all initial compositions

andparticularlyforaninitialcompositionofwBA=0.05inFig.5.

Fig.5.(a)SelectivitySasafunctionofsolutedistributionratioDforthethree sys-tems[AA+BA+HC]withHC:(d)decane,()dodecane,(N)hexadecane;(b)example forinitialcompositionwBA=0.05.

As it was expected, the selectivity decreases with the solute

distributionratio, whichinvolvesa compromise betweenthese

two parameters,i.e. betweenthesizeof theapparatusandthe

amountofsolventneeded.Nosignificantdifferencesareobserved

between thethree hydrocarbons.Nevertheless, as the biphasic

areabecomeslargerwhenthelengthofhydrocarbonincreases,

hexadecanemaybeanappropriatesolventforthisliquid–liquid

extractioninordertocollectanextractwithhigherpurity.

3.2. Consistencyoftie-linedata

Inthiswork,theOthmer–Tobiascorrelation[15](Eq.(3)),the

Bachmancorrelation[16](Eq.(4))andtheHandcorrelation[17]

(Eq.(5))wereusedtoensurethequalityoftheobtained

experi-mentaltie-linedata:

ln



₁ −wHC_HC wHC HC



=A+Bln



₁ −wAA_AA wAA AA



(3) wHC BA=A′+B′ wHC BA w_AAAA (4) ln



_wAA BA wAA AA



=A′′+B′′_ln



_wHC BA wHC HC



(5)

where w is the mass fraction, subscripts BA, HC and AA refer

respectivelyton-butylacetate,hydrocarbonandaceticacid,and

superscriptsHCandAArepresentthehydrocarbon-richphaseand

theAA-richphase,respectively;A,B,A’,B’,A′′_andB′′_,the

param-etersoftheOthmer–Tobiascorrelation,theBachmancorrelation

andtheHandcorrelation,respectively.Thecorrelationparameters

and thecorrelationfactor R2 _{values weredeterminedby a}

par-tialleast-squaresregression. Thecorrelatedresultsarereported

inTable2.TheOthmer–Tobias,BachmanandHandplotsarealso

showninFigs.6–8,respectively,forthethreeinvestigatedsystems

withdecane,dodecaneandhexadecane.AsseenfromTable2and

Figs.6–8,theclosenessofthecorrelationfactorR2_tounityand

thelinearityoftheplotrevealthehighdegreeofconsistencyof

Table2

ConstantsofOthmer–Tobias,BachmanandHandequationssystem.

Hydrocarbon Othmer–Tobiascorrelation Bachmancorrelation Handcorrelation

A B R2 _A′ _B′ _R2 _A′ ′ _B′ ′ _R2

Decane 0.258 0.892 0.995 0.002 0.656 0.992 −0.332 0.785 0.994 Dodecane 0.145 0.919 0.995 0.007 0.594 0.981 −0.220 0.727 0.994 Hexadecane −0.246 0.907 0.966 0.023 0.435 0.961 0.159 0.707 0.982

3.3. Correlationmodel

Experimentaldataarethuscorrelated withtheexcess Gibbs

freeenergymodelUNIQUAC[18].Theexpressionoftheactivity

coefficientofthecompoundiisgiveninthefollowingEq.(6):

ln i=ln i xi +1−i xi −z 2



ln i i +1−i i



+qi

"

1−ln



X

N j=1jji



−

X

N j=1 jji

P

N k=1kkj

#

(6)

Fig.6.Othmer–Tobiasplotforthe[AA+BA+HC]ternarysystemsatT=293.15K withHC:(d)decane,()dodecane,(N)hexadecane.

Fig.7.Bachmanplotforthe[AA+BA+HC]ternarysystemsatT=293.15KwithHC: (d)decane,()dodecane,(N)hexadecane.

Table3

StructuralparametersoftheUNIQUACmodel.

Compound r(m2_{/kmol) q(m}3_/kmol)

Aceticacid 5.180×108 _0.0333 n-Butylacetate 1.049×109 _0.0732 Decane 1.504×109 _0.1092 Dodecane 1.774×109 _0.1296 Hexadecane 2.314×109 _0.1706 Table4

OptimizedUNIQUACbinaryparameters.

Pair Bij(K) Bji(K) AA+BA 29.006 80.708 AA+C10 −18.535 −270.071 AA+C12 −26.558 −270.500 AA+C16 −14.638 −333.271 BA+C10 50.602 42.413 BA+C12 16.526 54.114 BA+C16 8.937 35.641 with: i= xiri

P

N j=1xjrj (7) i=

_P

x_Niqi j=1xjqj (8) z=10(coordinationnumber) (9) ij=exp



Bij T



(10)

wherexstandsforthemolarfraction,Nthenumberofcomponents,

Tthetemperature(K)andBijanempiricalasymmetricbinary

inter-actionparameter.Thepurecompoundstructuralparametersrand

Fig.8.Handplotforthe[AA+BA+HC]ternarysystemsatT=293.15KwithHC:(d) decane,()dodecane,(N)hexadecane.

Table5

Calculatedtie-lineswiththeUNIQUACmodel(massfractions).

Initialcomposition AA-richphase HC-richphase rmsd(%)

wAA wBA wHC wAA wBA wHC wAA wBA wHC Decane 0.500 0.000 0.500 0.846 0.000 0.154 0.217 0.000 0.783 2.45 0.490 0.020 0.490 0.796 0.025 0.180 0.236 0.016 0.748 0.480 0.040 0.480 0.743 0.047 0.210 0.261 0.034 0.705 0.475 0.050 0.475 0.715 0.058 0.227 0.277 0.043 0.680 0.470 0.060 0.470 0.684 0.068 0.248 0.295 0.053 0.652 Dodecane 0.500 0.000 0.500 0.915 0.000 0.085 0.168 0.000 0.832 2.14 0.487 0.025 0.488 0.861 0.036 0.103 0.179 0.016 0.805 0.474 0.050 0.476 0.808 0.068 0.124 0.193 0.035 0.772 0.462 0.075 0.462 0.754 0.097 0.149 0.213 0.056 0.731 0.450 0.100 0.450 0.697 0.123 0.180 0.238 0.080 0.682 0.437 0.125 0.438 0.632 0.146 0.222 0.273 0.107 0.619 Hexadecane 0.500 0.000 0.500 0.967 0.000 0.033 0.088 0.000 0.912 2.65 0.475 0.050 0.475 0.870 0.080 0.050 0.095 0.022 0.883 0.450 0.100 0.450 0.780 0.146 0.074 0.109 0.053 0.838 0.425 0.150 0.425 0.692 0.201 0.107 0.131 0.094 0.775 0.400 0.200 0.400 0.597 0.246 0.157 0.167 0.146 0.687 0.387 0.225 0.388 0.542 0.264 0.194 0.194 0.176 0.630 0.375 0.250 0.375 0.474 0.277 0.249 0.234 0.212 0.554

q(vanderWaalsareaandvanderWaalsvolume,respectively)are

extractedfromtheDIPPRdatabase[19]andarereportedinTable3.

TheinteractionbinaryparametersBijaresimultaneouslyadjusted

toreproducetheexperimentalmutualsolubilitiesoftheternary

systems investigated.The objective function(OF) minimized in

ordertodetermineoptimalparametersisgivenbyEq.(11):

OF=

X

M i=1

X

N j=1 1− K_ijexp Kcalc ij

!

2 (11)

whereMisthenumberofexperimentaldataandNthenumberof

components.Kijisthemolardistributionratioofthecompoundj

fortheithexperiment,definedbyEq.(12):

Kij= xAA ij xHC ij (12)

wherexdenotesthemolarfraction,andthesuperscriptsAAand

HC stand for the acetic acid-rich phase and hydrocarbon-rich

phase,respectively.OptimizedbinaryparametersBijarereported

inTable4.

Toevaluatethemodelperformanceforeachsystemstudied,a

root-meansquaredeviation(rmsd)betweentheexperimentaland

calculatedcompositionsinbothphasesiscalculatedusingEq.(13):

rmsd=100·

r

P

M i=1

P

N j=1

$

xAA,exp ij −x AA,calc ij

2 +

$

xHC,exp ij −x HC,calc ij

2 2MN (13)

CalculatedvaluesaswellasrmsdarereportedinTable5.The

modelresultsarefoundinwellagreementwithexperimental

val-ues,withamaximalrmsdof2.65%fortheAA+BA+C16mixture.

Our values of rmsdare similarto values ofprevious works on

ternarysystems[20,21].

4. Conclusion

Liquid–liquid equilibrium (LLE) data for the systems acetic

acid+n-butylacetate+hydrocarbonsweredeterminedat293.15K

andatmosphericpressure.Asitwasshownthathexaneandacetic

acidweremiscibleinallproportions,ternarymixtureswerenot

feasible. Nevertheless, all the other investigated systems from

octanetohexadecaneformedatype-IphasediagramofLLE.The

two-phaseregionincreasedwithincreasingthehydrocarbonchain

lengthfortheternarysystemsstudied.Moreover,consistencyof

experimentaldatahasbeenshownusingOthmer–Tobias,Bachman

andHandcorrelations.Ingeneral,experimentaldatacouldbe

prop-erlycorrelatedusinganUNIQUACmodel.Thephasediagramsofthe

systemsaceticacid+n-butylacetate+hydrocarbonscanbeusedin

liquid–liquidextractionpracticeandasreferencedata.

Listofsymbols

A,B Othmer–Tobiascorrelationconstants

A′_,B′ _{Bachmancorrelationconstants}

A′′_,B′′ _{Handcorrelationconstants}

Bij UNIQUACbinaryinteractionparameter

C6 hexane

C8 octane

C10 decane

C12 dodecane

C16 hexadecane

D solutedistributionratio(inmass)

M numberofexperimentaldata

N numberofcomponents

K molardistributionratio

OF objectivefunction

q molecularpurecompoundstructuralparameter,vander

Waalsvolume(m3_/kmol)

r molecularpurecompoundstructuralparameters,vander

Waalsarea,(m2_/kmol)

R2 _{Othmer–Tobias,Bachman,Handcorrelationcoefficients}

rmsd rootmeansquaredeviation

S selectivity

T temperature(K)

w concentrationinmassfraction

x concentrationinmolefraction

Superscripts

AA aceticacid-richphase

HC hydrocarbon-richphase Subscripts AA aceticacid BA n-butylacetate HC hydrocarbon i experiment j compound

References

[1]Z.Chen,T.Iizuka,K.Tanabe,ChemistryLetters13(1984)1085–1088.

[2]S.Okazaki,N.Wada,CatalysisToday16(1993)349–359.

[3]M.Hino,K.Arata,ChemistryLetters10(1981)1671–1672.

[4]Y.Izumi,R.Hasebe,K.Urabe,JournalofCatalysis84(1983)402–409.

[5]Kirk-Othmer,EncyclopediaofChemicalTechnology,4thed.,Wiley,NewYork,

1994.

[6]A.Arpornwichanop,K.Koomsup,W.Kiatkittipong,P.Praserthdam,S.

Assabum-rungrat,JournaloftheTaiwanInstituteofChemicalEngineers40(2009)21–28.

[7]E.Sert,F.S.Atalay,Chemical&BiochemicalEngineeringQuarterly25(2011)

221–227.

[8]H.Tian,Z.Huang,T.Qiu,X.Wang,Y.Wu,ChineseJournalofChemical

Engi-neering20(2012)980–987.

[9]E.Lladosa,J.B.Montón,M.C.Burguet,R.Mu ˜noz,JournalofChemical&

Engi-neeringData53(2007)108–115.

[10]C.Zhang,H.Wan,L.Xue,G.Guan,FluidPhaseEquilibria305(2011)68–75.

[11]N.A.Darwish,M.A.Abdulkarim,I.Ashour,A.M.Dwaidar,F.S.Athamneh,Fluid

PhaseEquilibria200(2002)277–285.

[12]T.Gündogdu,S.C¸ehreli,FluidPhaseEquilibria331(2012)26–32.

[13]M.M.Esquível,M.G.Bernardo-Gil,FluidPhaseEquilibria62(1991)97–107.

[14]R.E.Treybal,LiquidExtraction,McGraw-Hill,NewYork,1963.

[15]D.Othmer,P.Tobias,Industrial&EngineeringChemistry34(1942)693–696.

[16]I.Bachman,IndustrialandEngineeringChemistry,AnalyticalEdition12(1940)

38–39.

[17]D.B.Hand,TheJournalofPhysicalChemistry34(1929)1961–2000.

[18]D.S.Abrams,J.M.Prausnitz,AIChEJournal21(1975)116–128.

[19]R.L.Rowley,W.Wilding,V.J.L.Oscarson,Y.Yang,N.A.Zundel,T.E.Daubert,

R.P.Danner,DIPPR®_{DataCompilationofPureCompoundsProperties,Design}

InstituteforPhysicalProperties,AIChE,NewYork,2003.

[20]H.GhanadzadehGilani,A.GhanadzadehGilani,S.Shekarsaraee,H.Uslu,Fluid

PhaseEquilibria316(2012)109–116.

[21]H.GhanadzadehGilani,G.Khiati,A.K.Haghi,FluidPhaseEquilibria247(2006)