Ind. Eng. Chem. Res.

2000, 39, 849-854 849

CORRELATIONS

Excess Gibbs Free Energy Model for Calculating the Viscosity of

Binary Liquid Mixtures
Rosana J. Martins,, Ma rcio J. E. de M. Cardoso,*, and Oswaldo E. Barcia
Laborato rio de Fsico-Qumica de Lquidos e Eletroqumica, Departamento de Fsico-Qumica,
Instituto de Qumica, Universidade Federal do Rio de Janeiro, Centro de Tecnologia, Bloco A, sala 408,
Cidade Universita ria, cep. 21949-900, Rio de Janeiro, RJ, Brazil, and Departamento de Fsico-Qumica,
Instituto de Qumica, Universidade Federal Fluminense, Outeiro de Sa o Joa o Batista s/n, cep. 24020-150,
Nitero i, RJ, Brazil

In this article we present a model for correlating dynamic and kinematic viscosities of liquid
mixtures, which is based on Eyrings absolute rate theory for liquid viscosity and the UNIQUAC
equation. The proposed model involves the concept of ideal viscosity and uses the UNIQUAC
Downloaded by AUBURN UNIV on September 3, 2015 | http://pubs.acs.org

equation to represent the deviation from ideal behavior. The expression adopted to describe the
Publication Date (Web): January 29, 2000 | doi: 10.1021/ie990398b

ideal term viscosity has been chosen after a thorough investigation of the performance of different
equations previously proposed in the literature. The correlation results have shown a strong
dependence on the expression used to account for the ideal viscosity contribution. Besides size
and shape parameters, for each pure component, the model requires only two adjustable
parameters per binary system. The binary interaction parameters have been determined by
fitting literature viscosity data. More than 350 binary systems, 4619 viscosity data points at
0.1 MPa, have been correlated using this model. The binary systems investigated are
representative of different types of intermolecular interactions (e.g., nonpolar/nonpolar, nonpolar/
polar, and polar/polar). The calculated values are in good agreement with the experimental ones.
The overall average mean relative standard deviation of the correlations is 1.20%, which is
comparable with those of other correlation models available in the literature.

1. Introduction Most of the methods available in the literature for

Knowledge of the viscosity of pure liquids and respec-
tive mixtures is important for practical and theoretical
purposes. The viscosity of multicomponent liquid mix-
tures is an invaluable type of data for the chemical
engineer in the design and optimization of industrial
processes. Furthermore, the simultaneous investigation
of viscosity and volume effects on mixing can be a
powerful tool for the characterization of the intermo-
lecular interactions present in these mixtures.
Compilations of the viscosities of pure liquids at
different temperatures and pressures, as well as those
of binary liquid mixtures through the whole composition
range, can be found in the literature.1-4 Nevertheless,
the situation changes in the case of multicomponent
liquid mixtures. Over the last years, numerous equa-
tions for liquid mixture viscosity have been proposed
for implementation in computer prediction routines. The
results, in general, are in poor agreement with experi-
mental values, unless pure component and binary
solution viscosity data are taken into account. A recent
review concerning viscosity modeling may be found in
the work of Reid et al.5
* To whom correspondence should be addressed. E-mail:
marcio@iq.ufrj.br.
Universidade Federal do Rio de Janeiro. Fax: (55) (21) 290-
4746. Telephone: (55) (21) 590-9890.
Universidade Federal Fluminense. Fax: (55) (21) 620-7769.
10.1021/ie990398b CCC: $19.00 2000 American Chemical Society
Published on Web 01/29/2000
liquid mixture viscosity calculations are based on the
principle of corresponding states,6-8 the absolute rate
theory of Eyring and co-workers,9 and more recently,
molecular dynamic models.10 The corresponding states
techniques are restricted to a small number of systems
and are not suitable for polar mixtures or mixtures
exhibiting strong associations.5 Careful inspection of the
literature, particularly in the past decade, shows that
use of a viscosity model based on excess Gibbs free
energy and Eyrings absolute rate theory is a good
approach for correlating experimental viscosity data.11-17
Molecular dynamics provides a way to predict the
viscosity of fluid mixtures10,18-20 without making use of
experimental data; nevertheless, the deviations between
calculated and experimental values are, in general,
greater than those obtained with the other types of
models.
As stated previously by Eyring and co-workers,9 there
is a close analogy between the excess free energy of
activation for flow in mixtures and the respective excess
free energy of mixing. Therefore, it seems reasonable
to describe the viscosity of a liquid mixture in terms of
the viscosities of the pure components and some inter-
action parameters characteristic of the mixture.
The aim of this work is to develop a viscosity model
based on Eyrings fluid viscosity approach and the
UNIQUAC equation.21,22 In a subsequent work, its
 850 Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000

extension to multisolvent mixtures and electrolyte gq

systems will be presented.
The basic assumption of our model is the equivalence
between the Gibbs free energy of activation for viscous
flow and the Gibbs free energy of mixing.
2. The Viscosity Model for Liquid Mixtures
The absolute rate theory approach of Eyring provides
the following expression for the viscosity of a liquid
mixture:9
ln(v) ) i xi ln(iv0i ) +
RT
(7)
Extension of the concepts of classical thermodynamics
to the viscous flow behavior of liquid mixtures can be
done by assuming an equivalence between the Gibbs
free energy of activation for flow and the equilibrium
Gibbs free energy of mixing. Taking this direct relation-
ship between the thermodynamics and the transport
property, we decided to use a model based on the concept

( )
hN gq
) exp (1)
v RT
where is the dynamic viscosity of the mixture, h is
Plancks constant, N is Avogadros number, v is the
molar volume of the mixture, gq is the molar Gibbs
free energy of activation for the flow process, and T is
the absolute temperature.
In principle, eq 1 can be used to describe the viscosity
of ideal mixtures as well as that of nonideal systems.
Downloaded by AUBURN UNIV on September 3, 2015 | http://pubs.acs.org
of excess Gibbs free energy to account for the viscosity
deviation from ideal behavior.
The UNIQUAC model,21,22 which has been used for
phase and chemical equilibria predictions, is adopted
to describe the nonideal part of the expression for liquid
mixture viscosity, the second term on the right-hand
side of eqs 4 and 5. The UNIQUAC model is applicable
to multicomponent systems containing polar and non-
polar components as well as those whose molecules
differ considerably in size and shape. As pointed out by
Abrams and Prausnitz,21 the use of some well-defined

For an ideal liquid mixture, in the viscosity sense, eq 1 simplifications reduces the UNIQUAC equation to any
Publication Date (Web): January 29, 2000 | doi: 10.1021/ie990398b

can be written as one of the most popular expressions for the excess Gibbs

( )
q
energy, such as the NRLT and Wilson equations.23
hN gideal Besides its applicability to a large variety of mixtures,
ideal ) exp (2) the UNIQUAC model requires only two adjustable
videal RT
interaction parameters per binary system.
where videal is the molar volume of the ideal mixture in According to the UNIQUAC model,22 the molar excess
the thermodynamic sense. Gibbs free energy, due to short-range interactions, is
The molar Gibbs free energy of activation for flow can the sum of a combinatorial term and a residual term:
be regarded as the sum of an ideal contribution and a
correction or excess term. That is, gE ) gCE + gRE (8)

gq ) gideal
q
+ gqE (3) where the subscript E means excess, and the super-
scripts C and R represent the combinatorial and re-
q
where gideal is the ideal mixture contribution, at the sidual contributions, respectively.
same conditions of temperature, pressure, and composi- The combinatorial contribution takes into account
tion as for the real system (per mole), and gqE is the differences in size and shape of the molecules in the
excess free energy of activation for flow (per mole). mixture, while the residual part of the UNIQUAC model
Combining eqs 1-3 leads to the following expres- accounts for the interactions between the species.21,22
sion: The UNIQUAC molar excess Gibbs free energy is given
by

() ()
gqE
ln(v) ) ln(v)ideal + (4) gE NSOL i z
NSOL i
RT
RT
)
i)1
xi ln
xi
+
2 i)1
qixi ln
i
-
Equation 4 clearly shows that our model considers the NSOL NSOL

iki)
dynamic viscosity of real solutions as the result of an
ideal term contribution and an excess type term. xiqi ln( (9)
i)1 k)1
In the same way, for the kinematic viscosity, we have
where ) /F is the kinematic viscosity and M is the
where NSOL is the number of solvents, qi is the surface
area parameter, i is the surface area fraction, i is the
gqE volume fraction, ki is the interaction parameter be-
ln(M) ) ln(M)ideal + (5)
RT tween species k and i, and z is the coordination number.
For all species the coordination number z is assigned
molar mass of the fluid mixture. the value of 10.
An alternative approach, used by other authors,14,15 A judicious choice of the expression of the ideal term
is to define and of the excess term is of fundamental importance in
order for the model to correlate and predict the viscosi-
q
gideal ties of multicomponent systems.

RT
) i xi ln(iv0i ) - ln(hN) (6) Classical thermodynamics provides an unambiguous
mathematical description of ideal solution properties.23
However, unfortunately, the same does not hold for
which would lead to the following expression for the solution viscosity. There is not an unique equation for
nonideal solution: the ideal viscosity contribution. The most common
 Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000 851

Table 1. Results of the Correlation of the Viscosity of Some Binary Systems According to the Choice of the Ideal Term
Contributiona
MRSD
system ND eq 10 eq 11 eq 13
benzene (1) + toluene (2) 8 0.0003 0.0003 0.0003
n-butyl acetate (1) + n-pentyl acetate (2) 7 0.0014 0.0014 0.0014
toluene (1) + ethylbenzene (2) 7 0.0026 0.0026 0.0026
n-hexane (1) + n-heptane (2) 9 0.0037 0.0037 0.0037
ethyl benzoate (1) + benzene (2) 6 0.0036 0.0031 0.0079
ethyl benzoate (1) + toluene (2) 5 0.0029 0.0030 0.0049
benzyl benzoate (1) + benzene (2) 5 0.0064 0.0242 0.0508
benzyl benzoate (1) + toluene (2) 5 0.0011 0.0164 0.0324
ethyl acetate (1) + benzyl benzoate (2) 10 0.0067 0.0834 0.2041
ethyl acetate (1) + ethyl benzoate (2) 9 0.0044 0.0081 0.0195
ethyl benzoate (1) + benzyl benzoate (2) 10 0.0049 0.0057 0.0186
diethyl ether + phenyl ethyl ether 14 0.0048 0.0066 0.0378
diethyl ether (1) + diphenyl ether (2) 10 0.0070 0.0314 0.2242
phenyl ethyl ether (1) + diphenyl ether (2) 10 0.0053 0.0051 0.0146
1-propanol (1) + n-heptane (2) 24 0.0341 0.0284 0.1248
1-butanol (1) + n-heptane (2) 24 0.0299 0.0428 0.1233
1-pentanol (1) + n-heptane (2) 20 0.0273 0.0443 0.1774
1-hexanol (1) + n-heptane (2) 20 0.0330 0.0531 0.2092
1-octanol + n-heptane (2) 24 0.0199 0.0565 0.2630
1-decanol (1) + n-heptane (2) 24 0.0375 0.0850 0.3985
Downloaded by AUBURN UNIV on September 3, 2015 | http://pubs.acs.org

2-methylpentane (1) + n-hexadecane (2) 9 0.0102 0.0360 0.0595

3-methylpentane (1) + n-hexadecane (2) 9 0.0054 0.0361 0.0491
Publication Date (Web): January 29, 2000 | doi: 10.1021/ie990398b

2,2-dimethylbutane (1) + n-hexadecane (2) 9 0.0061 0.0332 0.0339

2-methylpentane (1) + n-tetradecane (2) 9 0.0053 0.0192 0.0329
2,3-dimethylbutane (1) + n-tetradecane (2) 9 0.0060 0.0199 0.0182
2-butanone (1) + 1-chlorohexane (2) 9 0.0069 0.0068 0.0069
2-butanone (1) + 1-chlorohexadecane (2) 9 0.0041 0.0421 0.0386
ethyl acetate (1) + 1-chlorohexane (2) 9 0.0005 0.0004 0.0006
ethyl acetate (1) + 1-chlorohexadecane (2) 9 0.0007 0.0453 0.0490
a ND, number of experimental data points; MRSD, mean relative standadard deviation.

( ( viscositycal - viscosityexp
))
expressions found in the literature are the following: ND 2 1/2
1
MRSD ) (14)
Kendall and Monroe,24 ND 1 viscosityexp

ln ideal ) i xi ln i (10) where ND is the number of data points and subscripts

cal and exp refer to calculated and experimental values,
respectively.
Bingham,25 Table 1 presents a comparison of the correlational
abilities of eqs 10, 11, and 13 for some binary systems,
1 1
ideal
) i xi (11) which are representative of different types of interac-
tions between solution components. With few excep-
i
tions, the following equation, which uses eq 10 for the
ideal contribution, does better in reproducing the ex-
Cronauer, Rothfus, and Kernmore,26 perimental behavior of the viscosities of the binary
ln ideal ) i xi ln i (12) systems investigated in this work:

()
NSOL NSOL NSOL i
Another is the following equation, which frequently ln(v) ) xi ln(i) + ln( xiv0i ) + xi ln
i
+

()
appears in recent publications on experimental deter- i)1 i)1 i)1
minations of solution viscosities:27,28 z
NSOL i NSOL NSOL

qixi ln - xiqi ln( iki) (15)

ideal ) i xii (13) 2 i)1 i i)1 k)1

where v0i represents the molar volume of the pure

In eqs 10-13, xi is the mole fraction of component i in
the mixture, and i and i are respectively the kinematic
and dynamic viscosities of the ith pure component at
the same temperature and pressure of the mixture.
To obtain the final expression for the multicomponent
liquid mixture viscosity calculation, we have performed
correlations of experimental viscosity data of binary
systems with eqs 10, 11, and 13 together with the
UNIQUAC excess Gibbs free energy equation. The
results for each of the proposed equations were evalu-
ated in terms of the mean relative standard deviation,
MRSD:
component i.
3. Results and Discussion
The viscosity data bank used includes 352 binary
systems at 0.1 MPa.
For each binary system studied, the unknown inter-
action parameters have been estimated by fitting the
experimental viscosity data. The values of volume and
surface area parameters, r and q, for each solvent were
obtained as recommended by Reid et al.5 The pure liquid
viscosity and density data have been taken from the
 852 Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000

Table 2. Correlation Results of the Model, Eq 15,

According to the Type of Mixture Studieda
no. of average
system type systems total ND MRSD (%)
alkanes 50 464 0.58 (3.4)
aromatic 35 245 0.57 (3.7)
alcohols and/or water 92 1618 2.24 (5.2)
substituted aromatics 10 124 0.36 (3.1)
ketones 18 312 0.96 (4.2)
ethers 12 197 0.89 (3.8)
esters 31 406 0.51 (2.7)
amines 13 177 1.00 (4.5)
acids 9 86 1.15 (4.7)
nitroalkanes 1 15 0.46 (4.9)
dimethyl sulfoxide 15 279 2.26 (5.3)
nitriles 16 158 0.96 (3.1)
haloalkanes 50 538 0.93 (4.7)
a N , number of experimental data points; MRSD, mean relative
D
standadard deviation (number in parentheses is the deviation
obtained with GC-UNIMOD17).

same literature reference used for the corresponding

binary system.
The model performance has been evaluated by com-
Downloaded by AUBURN UNIV on September 3, 2015 | http://pubs.acs.org

paring calculated and experimental viscosity values in Figure 1. Correlation of dynamic viscosity of butanoic acid +
Publication Date (Web): January 29, 2000 | doi: 10.1021/ie990398b

terms of the mean relative standard deviation. The nitrobenzene at 298.15 K and 0.1 MPa. (b) Experimental data from
ref 29; (s) the values obtained with eq 15.
objective function used in the determination of the
binary interaction parameters was
ND

F) 1 (viscositycal - viscosityexp)2 (16)

A detailed compilation of the correlation results,

which contains the temperature range, total number of
data points, UNIQUAC interaction parameters, mean
relative standard deviation, and literature source of
experimental data for each binary system studied, is
available as Supporting Information.
In general, our results are of the same order of
magnitude than those of Cao and co-workers,15 who also
fitted their model interaction parameters by means of
binary system viscosity data.
Table 2 summarizes the fitting results according to
the different classes of liquid mixtures investigated. The
classification of the mixtures, in terms of their func-
tional groups, follows the suggestions of Cao et al.17 As
can be seen from Table 2, the binary systems composed
by water or dimethyl sulfoxide are those for which the
present model gave the poorest results. A comparison
of our results to those of the GC-UNIMOD model17 is Figure 2. Viscosity composition curve of 1,2-dichloroethane +
presented in Table 2. The numbers in parentheses in carbon tetrachloride binary system at 303.15 K and 0.1 MPa. (b)
Experimental values from ref 30; (s) the values calculated with
the last column of Table 2 correspond to the deviations eq 15.
in the work of Cao et al.17 As can be seen, the GC-
of the pure component viscosities, as is the case of
UNIMOD model also shows a higher deviation for
ethanol + oleic acid, are quite well represented by our
systems containing alcohols and/or water, as well as for
model. Figure 4 shows the viscosity composition curve
those containing dimethyl sulfoxide. For all types of
of the system methanol + ethanol at different temper-
systems our deviations were significantly smaller than
atures and 0.1 MPa. In this case, again, the results
those obtained with the GC-UNIMOD model. This
obtained with our model are in good agreement with
behavior could be expected because in GC-UNIMOD
the experimental data.
model the interaction parameters were obtained by
The overall average MRSD of our correlation is 1.20%,
fitting only VLE experimental data.
which shows that our model is appropriate to correlate
Figures 1-4 show typical results obtained with our
the viscosity of a large variety of binary liquid systems
model. Figure 1 shows the viscosity composition curve
for temperatures ranging from 283.15 to 328.15 K, and
of a system with a maximum: butanoic acid + nitroben-
at 0.1 MPa.
zene at 298.15 K and 0.1 MPa. Figure 2 presents the
viscosity versus composition behavior of a system with
4. Conclusions
a minimum: 1,2-dichloroethane + carbon tetrachloride
at 303.15 K and 0.1 MPa. As can be seen from Figure A model for liquid mixture dynamic and kinematic
3, binary systems with large differences in the values viscosity is presented and used to correlate the experi-
 Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000 853

The correlation of viscosity data is applicable for the

283.15-328.15 K temperature range, at pressures of 0.1
MPa, and for a large variety of binary systems with
reasonable results. The value of the overall average
mean relative standard deviation, MRSD, is 1.20%. We
have performed some preliminary calculations, with the
binary interaction parameters determined in this work,
for ternary and quaternary liquid systems, which have
shown that the model predicts the experimental viscos-
ity values with a quite reasonable agreement.
In a subsequent work we intend to apply the proposed
model for viscosity calculations of binary and ternary
electrolyte solutions.

Nomenclature
F ) objective function
gq ) molar Gibbs free energy of activation for flow process
h ) Plancks constant
M ) molar weight
MRSD ) mean relative standard deviation
N ) Avogadros number
NSOL ) number of solvents in the mixture
Downloaded by AUBURN UNIV on September 3, 2015 | http://pubs.acs.org

Figure 3. Comparison between experimental and calculated

qi ) surface area parameter
Publication Date (Web): January 29, 2000 | doi: 10.1021/ie990398b

viscosity for the binary system ethanol + oleic acid at 318.15 K

and 0.1 MPa. (b) Experimental values from ref 31; (s) the values ri ) volume parameter
calculated with eq 15. T ) absolute temperature
v ) molar volume of the liquid mixture
v0i ) molar volume of pure liquid i
xi ) mole fraction of component i
z ) coordination number
Greek Letters
) dynamic viscosity
) kinematic viscosity
i ) surface area fraction
i ) volume fraction
ki ) UNIQUAC interaction parameter between species k
and i
Subscripts
cal ) calculated data
E ) excess property
exp ) experimental data
i ) pure component i
Superscripts
q ) activated state
C ) combinatorial
R ) residual
Figure 4. Correlation of the kinematic viscosity of the binary
system methanol + ethanol at different temperatures under 0.1
Acknowledgment
MPa. (b)a 283.15 K, (O)a 293.15 K, (2)b 298.15 K, (0)a 303.15 K,
([)a 313.15 K, and (1)a 323.15 K. The solid lines have the same The authors are grateful to the Brazilian agencies
meaning as in the previous figures. a Experimental data from ref
32. b Experimental data from ref 33.
CNPq, FUJB, CAPES, FINEP, and FAPERJ for finan-
cial support.
mental data for various kinds of binary systems. The
proposed model is based on Eyrings absolute rate theory Supporting Information Available: Table con-
for viscous flow of liquids.9 The resulting equations for taining the binary interaction parameters for each of
liquid mixture viscosity are regarded as a combination the 352 bynary systems investigated in this work as well
of an ideal viscosity with an excess viscosity term. as a list of the original literature source for the
The excess term which appeared in eqs 4 and 5 is experimental viscosity data. This material is available
represented by means of the UNIQUAC excess free free of charge via the Internet at http://pubs.acs.org.
energy model.21,22 The equation for the ideal viscosity
term was chosen after a careful investigation. As Literature Cited
discussed previously, Kendall and Monroes equation24 (1) Riddick, J. A.; Bunger, W. B. Organic Solvents, 3rd ed.;
has been shown to be the most suitable for this purpose. Wiley-Interscience: New York, 1970.
Therefore, eq 15 has been used for liquid mixture (2) CRC Handbook of Chemistry and Physics; Lide, D. R., Ed.;
viscosity calculations. CRC Press Inc.: Boca Raton, FL, 1995.
 854 Ind. Eng. Chem. Res., Vol. 39, No. 3, 2000
(3) Viswanath, D. S.; Natarajan, G. Data Book on the Viscosity
of Liquids; Hemisphere Publishing Corp.: New York, 1989.
(4) Irving, J. B. Viscosity of Binary Liquid Mixtures: The
Effectiveness of Mixture Equations; NEL Report No. 631; National
Engineering Laboratory: East Kilbride, Glasgow, Scotland, 1977.
(5) Reid, R. C.; Prausnitz, J. M.; Poling, B. E. The Properties of
Gases and Liquids, 4th ed.; McGraw-Hill: New York, 1987.
(6) Teja, A. S.; Rice, P. The Measurement and Prediction of the
Viscosities of Some binary Liquid Mixtures Containing n-Hexane.
Chem. Eng. Sci. 1981, 36, 7.
(7) Teja, A. S.; Rice, P. Generalized Corresponding States
Method for the Viscosities of Liquid Mixtures. Ind. Eng. Chem.
Fundam. 1981, 20, 77.
(8) Ely, J. F.; Hanley, H. J. M. Prediction of Transport Proper-
ties. 1. Viscosity of Fluids and Mixtures. Ind. Eng. Chem. Fundam.
1981, 20, 323.
(9) Glasstone, S.; Laidler, K. J.; Eyring, H. The Theory of Rate
Process; McGraw-Hill: New York, 1941.
(10) Wheeler, D. R.; Rowley, R. L. Shear Viscosity of Polar
Liquid Mixtures Via Non-Equilibrium Molecular Dynamics: Wa-
ter, Methanol, and Acetone. Mol. Phys. 1998, 94, 555.
(11) Wei, I. C.; Rowley, R. L. A Local Composition Model for
Multicomponent Liquid Mixture Shear Viscosity. J. Chem. Eng.
Sci. 1985, 40, 401.
(12) Wu, D. T. Prediction of Viscosities of Liquid Mixtures by
a Group Contribution Method. Fluid Phase Equilib. 1986, 30, 149.
Downloaded by AUBURN UNIV on September 3, 2015 | http://pubs.acs.org
(13) Chevalier, J. L.; Petrino, P.; Gaston-Bonhomme, Y. Esti-
Publication Date (Web): January 29, 2000 | doi: 10.1021/ie990398b
mation Method for the Kinematic Viscosity of a Liquid-Phase
Mixture. Chem. Eng. Sci. 1988, 43, 1303.
(14) Schrodt, J. T.; Akel, R. M. Binary Liquid Viscosities and
their Estimation from Classical Thermodynamics. J. Chem. Eng.
Data 1989, 34, 8.
(15) Cao, W.; Fredenslund, Aa.; Rasmussen, P. Statistical
Thermodynamic Model for Viscosity of Pure Liquids and Liquid
Mixtures. Ind. Eng. Chem. Res. 1992, 31, 2603.
(16) Cao, W.; Knudsen, K.; Fredenslund, Aa.; Rasmussen, P.
Simultaneous Correlation of Viscosity and Vapor-Liquid Equilib-
rium Data. Ind. Eng. Chem. Res. 1993, 32, 2077.
(17) Cao, W.; Knudsen, K.; Fredenslund, Aa.; Rasmussen, P.
Group Contribution Viscosity Prediction of Liquid Mixtures Using
UNIFAC-VLE Parameters. Ind. Eng. Chem. Res. 1993, 32, 2088.
(18) Lee, S. H.; Cummings, P. T. Shear Viscosity of Model
Mixtures by Nonequilibrium Molecular Dynamics. II. Effect of
Dipolar Interactions. J. Chem. Phys. 1996, 105, 2044.
(19) Dysthe, D. K.; Fuchs, A. H.; Rousseau, B.; Durandeau, M.
Fluid Transport Properties by Equilibrium Molecular Dynamics.
II. Multicomponent Systems. J. Chem. Phys. 1999, 110, 4060.
(20) Schaink, H. M.; Luo, H.; Hoheisel, C. Molecular Dynamics
Calculation of the Transport Coefficients of Liquid Benzene +
Cyclohexane Mixtures Using Six-Center Lennard-Jones Potentials.
J. Chem. Phys. 1993, 99, 9912.
(21) Abrams, D. S.; Prausnitz, J. M. Statistical Thermodynam-
ics of Liquid Mixtures: A New Expression for the Excess Gibbs
Energy of Partly or Completely Miscible Systems. AIChE J. 1975,
21, 116.
(22) Maurer, G.; Prausnitz, J. M. On the Derivation and
Extension of the UNIQUAC Equation. Fluid Phase Equilib. 1978,
2, 91.
(23) Smith, J. M.; Van Ness, H. C.; Abbott, M. M. Introduction
to Chemical Engineering Thermodynamics; McGraw-Hill: New
York, 1996.
(24) Kendall, J.; Monroe, P. The Viscosity of Liquids. II. The
Viscosity-Composition Curve for Ideal Liquid Mixtures. J. Am.
Chem. Soc. 1917, 39, 1787.
(25) Bingham, E. C. Fluidity and Plasticity; McGraw-Hill: New
York, 1922.
(26) Cronauer, D. C.; Rothfus, R. R.; Kernmore, R. I. Viscosity
and Density of the Ternary Liquid System Acetone-Benzene-
Ethylene Dichloride. J. Chem. Eng. Data 1965, 10, 131.
(27) Aminabhavi, T. M.; Virupakshagouda, B. P. Density,
Viscosity, Refractive Index, and Speed of Sound in Binary Mixtures
of Ethenylbenzene with N,N-Dimethylacetamide, Tetrahydrofu-
ran, N,N-Dimethylformamide, 1,4-Dioxane, Dimethyl Sulfoxide,
Chloroform, Bromoform, and 1-Chloronaphtalene in the Temper-
ature Interval (298.15-308.15) K. J. Chem. Eng. Data 1998, 43,
497.
(28) Aminabhavi, T. M.; Virupakshagouda, B. P. Density,
Viscosity, and Speed of Sound in Binary Mixtures of 1-Chlo-
ronaphtalene with Methanol, Ethanol, Propan-1-ol, Butan-1-ol,
Pentan-1-ol, and Hexan-1-ol in the Temperature Interval (298.15-
308.15) K. J. Chem. Eng. Data 1998, 43, 504.
(29) Ahluwalia, R.; Wanchoo, R. K.; Sharma, S. K.; Vashisht,
J. L. Density, Viscosity and Surface Tension of Binary Liquid
Systems: Ethanoic Acid, Propanoic Acid, and Butanoic Acid with
Nitrobenzene. J. Solution Chem. 1996, 25, 905.
(30) Zhang, H.-L.; Han, S.-J. Viscometric and Volumetric
Studies on Binary Mixtures of 1,2-Dichloroethane and Chlorinated
Methanes or Their Binary Mixtures at 303.15 K. Fluid Phase
Equilib. 1997, 140, 233.
(31) Eduljee, G. H.; Boyes, A. P. Viscosity of Some Binary
Liquid Mixtures of Oleic Acid and Triolein with Selected Solvents.
J. Chem. Eng. Data 1980, 25, 249.
(32) Dizechi, M.; Marshall, E. Viscosity of some Binary and
Ternary Liquid Mixtures. J. Chem. Eng. Data 1982, 27, 358.
(33) Soliman, K.; Marshall, E. Viscosity of Selected Binary,
Ternary, and Quaternary Liquid Mixtures. J. Chem. Eng. Data
1990, 35, 375.
Received for review June 4, 1999
Revised manuscript received November 22, 1999
Accepted December 1, 1999
IE990398B