DETERMINATION AND PREDICTION OF THE ISOBARIC VAPOR-LIQUID-LIQUID EQUILIBRIUM DATA

Koichi Iwakabe and Hitoshi Kosuge

Department of Chemical Engineering, Tokyo Institute of Technology12-1, Ookayama-2, Meguro-ku, Tokyo 152-8552, Japan

ABSTRACT

The prediction method of the isobaric vapor-liquid-liquid equilibria (VLLE) data for thesystem ethanol-water-1-butanol and ethanol-2-butanol-water was studied. Theparameters for the activity coefficients models were determined from the constituentbinary VLE data. With the parameters, the isobaric ternary VLLE data were predictedand compared with the experimental ones obtained in our previous study.Furthermore, a prediction method based on successive calculations of LLE at boilingpoint and VLE was studied. For this method, the two sets of the parameters wereindependently determined for LLE and VLE calculation. The VLLE data predicted bythis method were in good agreement with the experimental ones.

INTRODUCTION

The phase equilibrium data is one of the basic and important physical properties fordesign and analysis of phase separation processes. For vapor-liquid equilibrium(VLE) and liquid-liquid equilibrium (LLE), a number of measurements and predictionshave been reported in literature and they seem to be successful. On the contrary,vapor-liquid-liquid equilibrium (VLLE) data, which are significantly important for theheterogeneous separation processes, have been reported scarcely due to thedifficulty of their measurements. The static method and the dynamic method havebeen mainly used for the measurements. The static method, which is said to be themost accurate method for the measurement of phase equilibria and is widely used forthe isothermal measurements, requires long time for settling the system. On thecontrary, the dynamic method can achieve the equilibrium quickly and has beengenerally used for the isobaric measurements. However, when it comes to themeasurements of VLLE, there is a problem of the recirculation of two liquid phasesthat appear in an apparatus for the VLLE measurements. So far, treatments of thesetwo liquid phases in the VLE measurement apparatus have been studied by someresearchers. For example, a dynamic VLE still equipped with an ultrasonichomogenizer in order to mix two liquid phases in the boiling flask was proposed byGomis et al. [1]. In their apparatus, LLE measurements at boiling points are doneindependently in the water bath that is controlled at the boiling temperature. Also in

our previous study [2], a simple dynamic VLE still has been developed for both VLEand VLLE measurements. Although the still is very similar to the ordinary Gillespietype VLE still, VLE and VLLE measurements can be done in the same apparatus.

Another problem concerned with VLLE is its prediction. It is well known that theprediction of VLLE is not as precise as those of VLE and LLE. As studied by Lee etal. [3], the selection of the parameters for the VLLE prediction is very important sincethe parameters obtained from VLE data are very different from those obtained by LLEdata. The predicted VLE or VLLE with the parameters determined by LLE data aresometimes poor. There seem to be no good indices to tell which parameters arebetter for prediction. In order to explore the cause, the accurate VLLE measurementsare indispensable.

The purposes of this study are to predict the VLLE precisely. In our previous study,isobaric VLE and VLLE for binary and ternary systems have been measured atatmospheric pressure. In this paper, the parameters for the activity coefficientsmodels are determined from the constituent binary VLE data. With the parameters,the isobaric ternary VLLE data are predicted and compared with the experimentalones Furthermore, a prediction method of ternary VLLE for these systems, and aparameter estimation method for LLE calculation are studied. The prediction methodis based on successive calculations of LLE at boiling point and VLE. For eachcalculation, the different sets of parameters of activity coefficients models areindependently determined and used. The predicted vapor phase compositions atVLLE are compared with the experimental ones.

PARAMETER ESTIMATION FOR ISOBARIC BINARY VLE AND VLLE DATA

Parameters of activity coefficient models, NRTL [4] and UNIQUAC [5], aredetermined by minimizing the following objective function.

= =

=

m

k

n

i ki

calkikiOF1 1

2

exp,,

,,exp,,

(1)

where OF is the objective function, and i,k,exp and i,k,cal are experimental andcalculated activity coefficient of component i of the data number k, respectively, m isnumber of the data and n is number of components. The objective function wasminimized by Marquardt method [6] and Simplex method proposed by Nelder andMead [7]. If the two sets of parameters determined by both methods are different witheach other, the one whose objective function is smaller than the other was taken. Inthis study, the non-randomness parameter, , of the NRTL model was fixed at 0.2,and the values for van der Waals volume and surface parameter, r and qrespectively, were taken from the literature [8].

The parameters determined from the binary VLE data for the system ethanol (1) water (2) are shown in Table 1. Predicted VLE for the system with the parameters inTable 1 are compared with the experimental ones in Fig. 1. In the figure, the VLEdata for the system in the literature [9-11] are also plotted. As can been seen in thefigure, the experimental and the predicted VLE for the system are in good agreementwith each other.

Table 1 Binary parameters of the NRTL and UNIQUAC models

System Model A12 A21 OF y1,avg y1,max Tavg TmaxEthanol-Water NRTL -460.030 1791.37 0.0137 0.0044 0.0136 0.1507 0.4686

UNIQUAC -70.1587 398.992 0.0066 0.0029 0.0122 0.1123 0.3362

Water-1-Butanol NRTL 3409.83 -426.999 0.3818 0.0129 0.0289 1.1576 2.8259UNIQUAC 498.762 183.167 0.2616 0.0100 0.0425 0.9634 2.3324

2-Butanol-Water NRTL -552.206 3344.64 0.6257 0.0202 0.0479 0.5278 0.9990 UNIQUAC 79.2739 516.5409 0.4319 0.0165 0.0641 0.4177 2.6441

Aij = (gij-gjj) / R for NRTL, (uij-ujj) / R for UNIQUAC

Fig. 1 Boiling and dew point for the system ethanol (1) - water (2) with those predicted byactivity coefficient models

The parameters for the systems water (1) 1-butanol (2) and 2-butanol (1) water(2) are also shown in Table 1. The comparison of boiling and dew points between theexperimental and the predicted ones for these systems are shown in Figs. 2 and 3,respectively. In these figures, the VLE data in the literature [12-20] and thosepredicted by ASPEN Plus are also plotted. For both systems, the predictedcompositions of water in theorganic liquid phase at VLLE are richer than theexperimental ones. In order to investigate the cause of these deviations, various

0 0.2 0.4 0.6 0.8 1350

355

360

365

370

375

T [K

]

Jones et al.(1943) Kojima et al.(1969) Rieder et al.(1949) this study Predicted by NRTL

x1, y1 [ - ]

types of the objective functions were tested. However, the results showed the sametendency. From these results of predictions, the sizes of two liquid phases regions atboiling points predicted by the parameters determined with VLE data for these twobinary systems tend to become larger than the experimental ones.

Fig. 2 Boiling and dew point for the system water (1) 1-butanol (2) with those predicted byactivity coefficient models

0 0.2 0.4 0.6 0.8 1360

365

370

375

380

385

390

395

400

x1, y1 [ - ]

T [K

]

Solubility data Boublik et al. (1960) Ellis et al. (1960) Hessel et al. (1965) Zong et al. (1983) this study predicted by NRTL (ASPEN) predicted by NRTL (this study)

Fig. 3 Boiling and dew point for the system 2-butanol (1) water (2) with those predicted byactivity coefficient models

PREDICTION OF ISOBARIC TERNARY VLLE WITH THE PARAMETERSDETERMINED BY BINARY VLE DATA

Predictions of isobaric ternary VLLE for the systems ethanol (1) water (2) 1-butanol (3) and ethanol (1) 2-butanol (2) water (3) were done with the parametersdetermined by the isobaric constituent binary VLE and VLLE data. Since isobaricVLE data for the system ethanol 1-butanol and ethanol 2-butanol were notmeasured in this study, the parameters for those systems were taken from literature[8]. The result for the system ethanol water 1-butanol is shown in Fig. 4. As canbe seen in the figure, the predicted two liquid phase region at boiling point is largerthan the experimental one. Experimental boiling points are compared with thepredicted ones in Fig. 5. Although the predicted boiling points seems to be in goodagreement in the figure, the liquid phase compositions are very different from theexperimental ones. Thus, these parameters cannot predict the isobaric VLLE data forthis system accurately.

0 0.2 0.4 0.6 0.8 1355

360

365

370

375

380

x1, y1 [ - ]

T [K

] Solubility by Ochi et al. (1996) Altsybeeva et al.(1964) Boeke et al.(1942) Yamamoto et al.(1959) this study Predicted by NRTL(ASPEN) Predicted by NRTL(this study)

Fig. 4 Comparison between experimental and predicted compositions at VLLE for the systemethanol (1) water (2) 1-butanol (3)

0.4 0.5 0.6 0.7 0.8 0.9 10

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

x2, y2 [ - ]

x 1, y

1 [ -

]

Exp. Pred. Organic phase Aqueous phase

Vapor phase

Ethanol

Water 1-Butanol

0 10

10

1

Fig. 5 Comparison between experimental and predicted boiling points at VLLE for the systemethanol (1) water (2) 1-butanol (3)

The comparison for the system ethanol-2-butanol-water is shown in Fig. 6. Similar toethanol-water-1-butanol system, the predicted two liquid phases region at boilingpoints is different from the experimental one, especially for the organic phase.Furthermore, the predicted boiling points for the region are about 15 K higher thanthe experimental ones, as shown in Fig. 7.

From the results of the predictions for these two ternary systems, the parametersdetermined with the constituent binary VLE data cannot be used for the prediction ofthe ternary VLLE data for these systems. In particular, the predictions of thecompositions of the organic liquid phase are significantly poor. It seems that theparameters for LLE calculation is necessary for better prediction.

90 91 92 93 94 9590

91

92

93

94

95

Texp [oC]

T cal [o

C]

+1.0oC

-1.0oC

Predicted from aqueous phasePredicted from organic phase

Fig. 6 Comparison between experimental and predicted compositions at VLLE for the systemethanol (1) 2-butanol (2) water (3)

0 0.1 0.2 0.3 0.4 0.50

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

x2, y2 [ - ]

x 1, y

1 [ -

]

Exp. Pred. Organic phase Aqueous phase

Vapor phase Tie line

0 10

10

1

Ethanol

2-Butanol Water

Fig. 7 Comparison between experimental and predicted boiling points at VLLE for the systemethanol (1) 2-butanol (2) water (3)

DETERMINATION OF THE PARAMETERS WITH THE ISOBARIC TERNARY VLLEDATA

Determination of the Parameters for LLE CalculationBased on the discussion above, the parameters for LLE calculation areindependently determined from the LLE data of the isobaric ternary VLLE data,namely the liquidcompositions. Though there are six parameters to be determinedfor a ternary system, it was hard to determine all of them at the same time. Recently,a regression method of the parameters with ternary LLE data was proposed byKatayama [21]. In the method, the parameters for the immiscible binary pair arepreliminary determined from the binary LLE or solubility data. By fixing theseparameters, the rest of the parameters are determined from the ternary LLE data byusing the modified Marquardt method proposed by Katayama [21]. In this study, thisKatayama method was used to determine the parameters for LLE calculation.

To determine the parameters from binary LLE data, the following objective functionwas employed.

( )= =

=m

k

n

ikiki aaOF

1 1

2,2,,1, (2)

where ai,1,k and ai,2,k are the activities of the component i of the data number k in thephase 1 and 2, respectively, and calculated as follows.

85 86 87 88 89100

101

102

103

104

Texp [oC]

T cal

[oC

]

aqueous phase organic phase

caliii xa ,= (3)To determine the parameters from ternary LLE data part of ternary VLLE data,equation (2) can be also used as the objective function. However, the LLE datapredicted with the parameters determined from the equation (2) were not accurate.So the following objective function for the prediction of the ternary LLE data wasemployed, which is the difference of the experimental and predicted liquidcomposition in each phase.

( )= = =

=m

k j

n

icalkjikji xxOF

1

2

1 1

2,,,exp,,, (4)

where j is the phase 1 and 2, and m is the number of the data.

First of all, the parameters for the immiscible binary pairs are determined from thesolubility data. Since the solubility data are not measured in this study, the data in theliterature [16, 20] are used. The solubility data are smoothed by the polynomialfunction of the temperature for each system and the LLE data for a temperature canbe read from this smoothed curve. Then the rest of the parameters were determinedby the modified Marquardt method proposed by Katayama [21].

Parameter estimation and Prediction for the system Ethanol-2-Butanol-WaterThe resultant parameters versus the temperature for the system 2-butanol-water areshown in Fig. 8. The parameters were correlated to the temperatures as a secondorder polynomial. The coefficients for the polynomial function are shown in Table 2.The rest of the parameters necessary for the ternary LLE prediction were determinedby Katayamas method successfully. They are shown in Table 3. The parameters forthe VLE calculations are directly determined from the activity coefficients data forboth systems. The parameters are shown in Table 4.

Table 2 Temperature dependent parameters for LLE calculation

System Pseudo Water(1)-1-Butanol(2) 2-Butanol(1)-Water(2) aij bij cij aij bij cij

NRTLA12 1.978x107 -1.084x105 148.5 -1981 11.81 -2.224x10-2A21 -1.713x107 9.378x104 -128.3 -2367 17.64 -1.587x10-2

UNIQUACA12 1.212x107 -6.631x104 90.73 -174.7 1.655 -4.166x10-3A21 -9.343x107 5.111x104 -69.91 -1261 7.486 -8.115x10-3

Aij = aij + bij T + cijT2

Fig. 8 Temperature dependences of the model parameters for the system 2-butanol (1) water (2)

250 300 350 400-1000

0

1000

2000

3000

T [K]

g12g21u12u21

gij, u

ij [K]

Table 3 Binary interaction parameters for LLE calculation

System Ethanol (1)-Water (2)-1-Butanol (3) Ethanol (1)-2-Butanol (2)-Water (3)NRTL UNIQUAC NRTL UNIQUAC

A12 1193 485.1 -546.5 -186.3A13 74.99 1447 -42.96 -385.3A21 -1292 651.3 -1386 -1691A31 -1741 387.9 -3073 -1095

Aij=(gij-gjj)/R for NRTL, (uij-ujj)/R for UNIQUACTable 4 Binary interaction parameters for VLE calculation

System Ethanol(1)-Water(2)-1-Butanol(3) Ethanol(1)-2-Butanol(2)-Water(3) NRTL UNIQUAC NRTL UNIQUAC

A12 2433 -303.6 -469.4 -282.3A13 -178.8 6.880 -175.5 84.00A21 -19.77 631.1 1315 1001A22 1657 626.1 -295.4 75.78A31 749.3 378.0 851.6 94.44A32 -285.3 -200.0 1541 181.6

Aij=(gij-gjj)/R for NRTL, (uij-ujj)/R for UNIQUACThe prediction method for the isobaric VLLE data is similar to the one proposed byLiu et al. [22], that is, VLE and LLE data are separately calculated with thecorrespondent parameters in Tables 2, 3 and 4. The procedure is as follows. Firstly,an overall liquid composition is given and the bubble point is set as the initialtemperature. With this condition, isothermal LLE calculation is done and thecompositions of the two liquid phases are calculated. For these compositions, theboiling points and the vapor phase compositions are independently determined bythe bubble point calculation. If these boiling points are different with each other, theaverage value is used for the next LLE calculation. The procedure is repeated untilthe difference of the boiling points between the former and the present iterationbecomes small enough for both liquid phases. So the predicted vapor phasecompositions and boiling points are presented for both organic and aqueous liquidphases in this study. Since only the two liquid phases regions are studied in thispaper, the phase stability analysis was omitted.

The predicted VLLE for the system ethanol-2-butanol-water by this method areshown in Fig. 9. As can be seen in the figure, the predicted LLE favorably coincideswith the experimental ones. The predicted vapor compositions are also in goodagreement with the experimental ones. Fig. 10 shows the comparison between thepredicted and experimental boiling points. Almost all of the predicted boiling pointsare within 1.0 K of the experimental ones.

Fig. 9 Comparison between experimental and predicted isobaric ternary VLLE for the systemethanol (1) 2-butanol (2) water (3) at 101.3 kPa

0 0.1 0.2 0.3 0.4 0.50

0.01

0.02

0.03

0.04

0.05

0.06

x2, y2 [ - ]

x 1, y

1 [ -

]Exp. Pred.

Organic phase Aqueous phase

Vapor phase Tie line

Fig. 10 Comparison between experimental and predicted compositions at VLLE for thesystem ethanol (1) 2-butanol (2) water (3)

Parameter Estimation and Prediction for the system Ethanol-Water-1-ButanolFor the system water-1-butanol, if the parameters determined by Katayamas methodare used for prediction of ternary LLE at boiling points, the predicted LLE datadeviate from the experimental ones as shown in Fig. 11. In the region where thecomposition of ethanol in the liquid phase is more than 0.02 mole fraction, thesolubility of the organic liquid phase for this ternary system changes a lot, while theboiling points for this region doesnt change so much. The binary solubility data forthis narrow temperature range changes slightly. Thus, the parameters from binaryLLE data cannot predict this big solubility change.

In this study, thepseudo-binary LLE data are assumed by normalizing thecompositions of water and 1-butanol in ternary LLE data of VLLE data, and theparameters for this binary immiscible pair are determined from the pseudo-binaryLLE data. The parameters determined by this method are also shown in Table 2. Therest of the parameters necessary for the ternary LLE prediction are estimated byKatayamas method and shown in Table 3. The parameters for VLE calculation aredetermined from the isobaric ternary VLLE data for this system. They are shown inTable 4.

86 87 88 8986

87

88

89

Texp [oC]

T cal [o

C]

predicted for organic phasepredicted for aqueous phase

+1.0oC

-1.0oC

Fig. 11 Comparison between experimental and predicted isobaric ternary LLE with theparameters determined by Katayamas method for the system ethanol (1) water (2) 1-

butanol (3)

The predicted VLLE for the system ethanol-water-1-butanol are shown in Fig. 12.The VLLE was calculated by the same way as for the system ethanol-2-butanol-water. The predicted compositions are in good agreement with the experimentalones. Fig. 13 shows the comparison between the predicted and experimental boilingpoints. Almost all of the predicted boiling points are within 1.0 K of the experimentalones.

From the results above, the isobaric ternary VLLE data for these two systems aresuccessfully predicted with the parameters obtained in this study.

0.6 0.7 0.8 0.9 10

0.02

0.04

0.06

0.08

x2 [ - ]

x 1 [

- ]

Organic phase Aqueous phase Vapor phase NRTL UNIQUAC

Fig. 12 Comparison between experimental and predicted isobaric ternary VLLE for thesystem ethanol (1) water (2) 1-butanol (3) at 101.3 kPa

0.6 0.7 0.8 0.9 10

0.02

0.04

0.06

0.08

0.1

0.12

x2, y2 [ - ]

x 1, y

1 [ -

]Exp. Pred.

Organic phase Aqueous phase

Vapor phase

Fig. 13 Comparison between experimental and predicted vapor phase composition of 1-butanol at VLLE for the system ethanol (1) water (2) 1-butanol (3)

CONCLUSION

The predictions of the isobaric ternary VLLE data for the systems ethanol-water-1-butanol and ethanol-2-butanol-water with activity coefficient models are studied. Theparameters determined from binary VLE data for the constituent binaries cannot beused for the prediction since the predicted LLE data at boiling points are verydifferent from the experimental ones. Then the parameters for VLE and LLEcalculation were independently determined from those data. Katayamas method wasemployed to determine the parameters for LLE calculation. For the system ethanol-2-butanol-water, the parameters were obtained by the method. For the system ethanol-water-1-butanol, however, the parameters for the binary immiscible pair determinedfrom the binary LLE data were not suited for the prediction of ternary LLE at boilingpoints. So the pseudo-binary system was assumed by normalizing the compositionsof binary immiscible pairs in the ternary LLE data. Using the parameters, betterpredictions for the ternary LLE data at boiling points were obtained. Finally, theisobaric ternary VLLE data are predicted by the successive calculation of the LLEand VLE. The predicted VLLE data were in good agreement with the experimentalones.

90 91 92 93 94 9590

91

92

93

94

95

Texp [oC]

T cal

[oC

]

predicted for organic phasepredicted for aqueous phase

+1.0 oC

-1.0 oC

NOMENCLATURE

Aij = binary interaction parameter for component i and j [K]a = activity [ - ]aij = coefficients for temperature dependent parameter [K]bij = coefficients for temperature dependent parameter [ - ]cij = coefficients for temperature dependent parameter [K-1]g = binary interaction parameter of NRTL model [J mol-1]m = number of the data [ - ]OF = objective function [ - ]q = van der Waals surface area parameter [ - ]R = gas constant (=8.314 J mol-1 K-1) [J mol-1 K-1]r = van der Waals volume parameter [ - ]T = temperature [K]u = binary interaction parameter of UNIQUAC model [J mol-1]x = mole fraction in the liquid phase [ - ]y = mole fraction in the vapor phase [ - ]

Greeks = nonrandomness parameter of NRTL model [ - ] = activity coefficient [ - ]

Subscriptcal = calculated valueexp = experimental valuei = component ij = component jj = phase number, in eq. 4k = data number

REFERENCES

1. V. Gomis, F. Ruiz, J. C. Asensi and M. D. Saquete (1997), Fluid Phase Equilibria,129, 15-19.

2. K. Iwakabe and H. Kosuge (2001), Fluid Phase Equilibria, 192, 171-186.

3. L.-S. Lee, W.-C. Chen and J.-F. Huang (1996), J. Chem. Eng. Japan, 29, 3, 427-438.

4. H. Renon and J.M. Prausnits (1968), AIChE Journal, 14, 1, 135-144.

5. D. S. Abrams and J. M. Prausnitz (1975), AIChE Journal, 21, 1, 116-128.

6. D. W. Marquardt (1933), J. Soc. Ind. Appl. Math, 11, 2, 431-441.

7. J. A. Nelder and R. Mead (1965), Computer J., 7, 308-313.

8. J. Gmehling and U. Onken (1977), Vapor-Liquid Equilibrium Data Collection,DECHEMA Chemistry Data Series, Vol. I, DECHEMA, Frankfurt.

9. C. A. Jones, E.M. Shoenborn and A.P. Colburn (1943), Ind. Eng. Chem., 35, 666-672.

10. K. Kojima, K. Ochi and Y. Nakazawa (1969), Int. Chem. Eng., 9, 342-347.

11. R. M. Rieder and A.R. Thompson (1949), Ind. Eng. Chem., 41, 2905-2908.

12. T. Boublik (1960), Collect. Czech. Chem. Commun., 25, 285.

13. S. R. M. Ellis and R.D. Garbett (1960), Ind. Eng. Chem., 52, 5, 385-388.

14. D. Hessel and G. Geiseler (1965), Zh. Phys. Chem. Leipzig., 229, 199.

15. Z.-L. Zong, X.H. Yang, X.-Y. Zheng (1983), J. Chem. Eng. Japan, 16, 1-6.

16. Nihon Kagakukai (Ed.), Chem. Soc. Jpn. (1975), Kagaku Binran, Kisohen II,Maruzen, Tokyo, p. 835.

17. A. I. Altsybeeva, V.P. Belousov, N.V. Ovtrakht and A.G. Morachevsky (1964), Zh.Fiz. Khim., 38, 1242-1247.

18. J. Boeke and K. H. Hanewald (1942), Rec. Trav. Chim. PAYS-BAS, 61, 881-887.

19. Y. Yamamoto and T. Maruyama (1959), Kagaku Kogaku, 23, 635-640.

20. K. Ochi, T. Saito and K. Kojima (1996), J. Chem. Eng. Data, 41, 361-364.

21. H. Katayama (2001), Bulletin of Computational Science Research Center, HoseiUniversity, 14.

22. F. Z. Liu, H. Mori, S. Hiraoka and I. Yamada (1993), J. Chem. Eng. Japan, 26, 41-47

KEYWORDPhase equilibrium, Vapor-liquid-liquid equilibrium, Prediction, Parameter estimation,Activity coefficient models

Navigation and PrintingTable of ContentIndexIndex of AuthorsOrganizing Committee, International Scientific CommitteeInternational Board of RefereesImpressumback to last viewprint

PrefacePlenary LecturesPL1 WHAT CAUSED TOWER MALFUNCTIONS IN THE LAST 50 YEARS?PL2 MODELLING SIEVE TRAY HYDRAULICS USING COMPUTATIONAL FLUID DYNAMICSPL3 CHALLENGES IN THERMODYNAMICSPL4 EXPERIENCE IN REACTIVE DISTILLATION

Topic 1 Basic Data1-1 COMPUTER AIDED MOLECULAR DESIGN OF SOLVENTS FOR DISTILLATION PROCESSES1-2 LARGE-SCALE DATA REGRESSION FOR PROCESS CALCULATIONS1-3 IONIC LIQUIDS AND HYPERBRANCHED POLYMERS PROMISING NEW CLASSES OF SELECTIVE ENTRAINERS FOR EXTRACTIVE DISTILLATION1-4 PREDICTION OF DIFFUSIVITIES IN LIQUID ASSOCIATING SYSTEMS ON THE BASIS OF A MULTICOMPONENT APPROACH1-5 KINETICS OF CARBON DIOXIDE ABSORPTION INTO N-METHYLDIETHANOLOAMINE SOLUTIONS6-1 THERMODYNAMIC PROPERTIES OF DIMETHYL SULFOXIDE + BENZENE OR + ISOPROPYLBENZENE MIXTURES6-2 DETERMINATION AND PREDICTION OF THE ISOBARIC VAPOR-LIQUID-LIQUID EQUILIBRIUM DATA6-3 MASS TRANSFER COEFFICIENTS IN BATCH AND CONTINUOUS REGIME IN A BUBBLE COLUMN6-4 A COMPARATIVE STUDY OF INTERFACIAL AREA OBTAINED BY PHYSICAL AND CHEMICAL METHODS IN A BUBBLE COLUMN6-5 DETERMINATION OF BINARY VAPOR LIQUID EQUILIBRIA (VLE) OF REACTIVE SYSTEMS

Topic 2.1 Equipment / Internals2.1-1 DISTILLATION COLUMNS WITH STRUCTURED PACKINGS IN THE NEXT DECADE2.1-2 CHARACTERISATION OF HIGH PERFORMANCE STRUCTURED PACKING2.1-3 MODIFICATIONS TO STRUCTURED PACKINGS TO INCREASE THEIR CAPACITY2.1-4 CRYSTALLIZATION FOULING IN PACKED COLUMNS2.1-5 FUNCTIONALITY OF A NOVEL DOUBLE-EFFECTIVE PACKING ELEMENT2.1-6 RASCHIG SUPER-RING A NEW FOURTH GENERATION PACKING OFFERS NEW ADVANTAGES2.1-7 PLATE DAMAGE AS A RESULT OF DELAYED BOILING6-6 NEW HIGHSPEED MASS-TRANSFER TRAYS6-7 DIFFUSIONAL AND HYDRAULIC CHARACTERISTICS OF KATAPAK-S6-8 THE MVG TRAY WITH TRUNCATED DOWNCOMERS: RECENT PROGRESS6-9 MASS TRANSFER AND HYDRAULIC DETAILS ON INTALOX PhD PACKING

Topic 2.2 Equipment / Flow2.2-1 EFFECT OF BED LENGTH AND VAPOR MALDISTRIBUTION ON STRUCTURED PACKING PERFORMANCE 2.2-2 THE EFFECT OF MALDISTRIBUTION ON SEPARATION IN PACKED DISTILLATION COLUMNS2.2-3 INFLUENCE OF VAPOR FEED DESIGN ON THE FLOW DISTRIBUTION2.2-4 ENTRAINMENT AND MAXIMUM VAPOUR FLOW RATE OF TRAYS2.2-5 EXPERIMENTAL CHARACTERISATION AND CFD SIMULATION OF GAS DISTRIBUTION PERFORMANCE OF LIQUID (RE)DISTRIBUTORS AND COLLECTORS IN PACKED COLUMNS2.2-6 PROGRESS IN UNDERSTANDING THE PHYSICAL PROCESSES INSIDE SPINNING CONE COLUMNS 2.2-7 SYSTEM LIMIT: THE ULTIMATE CAPACITY OF FRACTIONATORS6-10 COMPUTATIONAL FLUID DYNAMICS FOR SIMULATION OF A GAS-LIQUID FLOW ON A SIEVE PLATE: MODEL COMPARISONS6-11 NUMERICAL CALCULATION OF THE FLOW FIELD IN A BUBBLE COLUMN CONSIDERING THE ABSORPTION OF THE GAS PHASE6-12 MASS TRANSFER IN STRUCTURED PACKING6-13 EXPERIMENTAL STUDY OF RIVULET LIQUID FLOW ON AN INCLINED PLATE6-14 EFFECT OF THE INITIAL GAS MALDISTRIBUTION ON THE PRESSURE DROP OF STRUCTURED PACKINGS6-15 A NEW PRESSURE DROP MODEL FOR STRUCTURED PACKING

Topic 3.1 Process Synthesis3.1-1 SYNTHESIS OF DISTILLATION SEQUENCES FOR SEPARATING MULTICOMPONENT AZEOTROPIC MIXTURES3.1-2 DESIGN TECHNIQUES USED FOR THE DEVELOPMENT OF AN AZEOTROPIC DISTILLATION PROCESS WHICH USES A BINARY ENTRAINER FOR SEPARATION OF OLEFINS FROM ACIDS AND OTHER OXYGENATES3.1-3 DESIGN AND SYNTHESIS OF DISTILLATION SYSTEMS USING A DRIVING FORCE BASED APPROACH3.1-4 THE NEW APPROACH TO ISOPROPYLBENZENE DISTILLATION FLOWSHEET SYNTHESES IN PHENOL-ACETONE PRODUCTION3.1-5 A NOVEL FRAMEWORK FOR SIMULTANEOUS SEPARATION PROCESS AND PRODUCT DESIGN3.1-6 CASE-BASED REASONING FOR SEPARATION PROCESS SYNTHESIS6-16 THE FUNDAMENTAL EQUATION OF DISTILLATION6-17 HYDRODYNAMICS OF A GAS-LIQUID COLUMN EQUIPPED WITH MELLAPAKPLUS PACKING6-18 DYNAMIC BEHAVIOR OF RECYCLE SYSTEM: REACTOR DISTILLATION COLUMN6-19 DISTILLATION REGIONS FOR NON-IDEAL TERNARY MIXTURES6-20 SELECTIVE AMINE TREATING USING TRAYS, STRUCTURED PACKING, AND RANDOM PACKING

Topic 3.2 Process Simulation3.2-1 INFLUENCE OF UNEQUAL COMPONENT EFFICIENCIES ON TRAJECTORIES DURING DISTILLATION OF A QUATERNARY AZEOTROPIC MIXTURE3.2-2 SHORTCUT DESIGN OF EXTRACTIVE DISTILLATION COLUMNS3.2-3 SIMULATION OF HETEROGENEOUS AZEOTROPIC DISTILLATION PROCESS WITH A NON-EQUILIBRIUM STAGE MODEL 3.2-4 PLATE EFFICIENCIES OF INDUSTRIAL SCALE DEHEXANISER3.2-5 DESIGN OF AN EXPERIMENTAL PROCEDURE TO INVESTIGATE EFFICIENCY IN THE DISTILLATION OF AQUEOUS SYSTEMS6-21 EFFICIENT APPROXIMATE METHOD FOR PACKED COLUMN SEPARATION PERFORMANCE SIMULATION6-22 SIMULATION OF THE SIEVE PLATE ABSORPTION COLUMN FOR NITRIC OXIDE ABSORPTION PROCESS USING NEURAL NETWORKS6-23 DISTILLATION SIMULATION WITH COSMO-RS6-24 BATCH DISTILLATION: SIMULATION AND EXPERIMENTAL VALIDATION

Topic 3.3 Heat Integration3.3-1 OPTIMISATION OF EXISTING HEAT-INTEGRATED REFINERY DISTILLATION SYSTEMS 3.3-2 INTEGRATION OF DESIGN AND CONTROL FOR ENERGY INTEGRATED DISTILLATION3.3-3 IMPLEMENTATION OF OPTIMAL OPERATION FOR HEAT INTEGRATED DISTILLATION COLUMNS3.3-4 THEORETICAL AND EXPERIMENTAL STUDIES ON STARTUP STRATEGIES FOR A HEAT-INTEGRATED DISTILLATION COLUMN SYSTEM3.3-5 INTERNALLY HEAT-INTEGRATED DISTILLATION COLUMNS: A REVIEW6-25 AN ENGINEERING ANALYSIS OF CAPACITY IMPROVEMENT IN FLUE GAS DESULFURIZATION PLANT6-26 ANALYSIS OF SEPARATION OF WATER-METHANOL-FORMALDEHYDE MIXTURE6-27 MINIMUM ENERGY AND ENTROPY REQUIREMENTS IN MULTICOMPONENT DISTILLATION

Topic 3.4 Control / Dynamics3.4-1 MODEL PREDICTIVE CONTROL OF INTEGRATED UNIT OPERATIONS CONTROL OF A DIVIDED WALL COLUMN3.4-2 SIMULATION AND EXPERIMENTAL ANALYSIS OF OPERATIONAL FAILURES IN A METHANOL - WATER DISTILLATION COLUMN3.4-3 MODEL-BASED DESIGN, CONTROL AND OPTIMISATION OF CATALYTIC DISTILLATION PROCESSES6-28 OPTIMISATION, DYNAMICS AND CONTROL OF A COMPLETE AZEOTROPIC DISTILLATION: NEW STRATEGIES AND STABILITY CONSIDERATIONS

Topic 4 Integrated Processes4-1 DEVELOPMENT AND ECONOMIC EVALUATION OF A REACTIVE DISTILLATION PROCESS FOR SILANE PRODUCTION4-2 SEPARATION OF OLEFIN ISOMERS WITH REACTIVE EXTRACTIVE DISTILLATION4-3 TRANSESTERIFICATION PROCESSES BY COMBINATION OF REACTIVE DISTILLATION AND PERVAPORATION4-4 INVESTIGATION OF DIFFERENT COLUMN CONFIGURATIONS FOR THE ETHYL ACETATE SYNTHESIS VIA REACTIVE DISTILLATION4-5 SYNTHESIS OF N-HEXYL ACETATE BY REACTIVE DISTILLATION4-6 THERMODYNAMIC ANALYSIS OF THE DEEP HYDRODESULFURIZATION OF DIESEL THROUGH REACTIVE DISTILLATION4-7 DISTILLATION COLUMN WITH REACTIVE PUMP AROUNDS: AN ALTERNATIVE TO REACTIVE DISTILLATION4-8 HYBRID PERVAPORATION-ABSORPTION FOR THE DEHYDRATION OF ORGANICS4-9 NOVEL HYBRID PROCESSES FOR SOLVENT RECOVERY6-29 SCALE-UP OF REACTIVE DISTILLATION COLUMNS WITH CATALYTIC PACKINGS6-30 CONCEPTUAL DESIGN OF REACTIVE DISTILLATION COLUMNS USING STAGE COMPOSITION LINES

Topic 5 Novel Processes5-1 DEVELOPMENT OF A MULTISTAGED FOAM FRACTIONATION COLUMN5-2 OPERATION OF A BATCH DISTILLATION COLUMN WITH A MIDDLE VESSEL: EXPERIMENTAL RESULTS FOR THE SEPARATION OF ZEOTROPIC AND AZEOTROPIC MIXTURES5-3 SIMULTANEOUS OPTIMAL DESIGN AND OPERATION OF MULTIPURPOSE BATCH DISTILLATION COLUMNS5-4 SEPARATION OF TERNARY HETEROAZEOTROPIC MIXTURES IN A CLOSED MULTIVESSEL BATCH DISTILLATION-DECANTER HYBRID5-5 ENTRAINER-ENHANCED REACTIVE DISTILLATION 5-6 NOVEL DISTILLATION CONCEPTS USING ONE-SHELL COLUMNS5-7 INDUSTRIAL APPLICATIONS OF SPINNING CONE COLUMN TECHNOLOGY: A REVIEW6-31 FEASIBILITY OF BATCH EXTRACTIVE DISTILLATION WITH MIDDLE-BOILING ENTRAINER IN RECTIFIER