Accepted Manuscript

Activity coefficients at infinite dilution of organic solutes in the ionic liquid 1-butyl-3-methylimidazolium methyl sulfate

Ming-Lan Ge, Xiao-Mei Deng, Li-Hui Zhang, Jin-Yuan Chen, Jie-Ming Xiong,Wen-Hao Li

PII: S0021-9614(14)00136-0DOI: http://dx.doi.org/10.1016/j.jct.2014.04.020Reference: YJCHT 3921

To appear in: J. Chem. Thermodynamics

Received Date: 6 April 2014Revised Date: 21 April 2014Accepted Date: 24 April 2014

Please cite this article as: M-L. Ge, X-M. Deng, L-H. Zhang, J-Y. Chen, J-M. Xiong, W-H. Li, Activity coefficientsat infinite dilution of organic solutes in the ionic liquid 1-butyl-3-methylimidazolium methyl sulfate, J. Chem.Thermodynamics (2014), doi: http://dx.doi.org/10.1016/j.jct.2014.04.020

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customerswe are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, andreview of the resulting proof before it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

1. The Title: Activity coefficients at infinite dilution of organic solutes in the ionic liquid

1-butyl-3-methylimidazolium methyl sulfate

2. The Authors: Ming-Lan Gea, Xiao-Mei Deng

a, Li-Hui Zhang

a, Jin-Yuan Chen

a, Jie-Ming

Xionga,*

, and Wen-Hao Lib

3. a

Department of Chemical Engineering, Beijing Institute of Petrochemical Technology,

Beijing 102617, People’s Republic of China

b China Petroleum Kelamayi Petrochemical Company, Kelamayi 834003, People’s Republic

of China

*Corresponding author

*E-mail: geminglan@bipt.edu.cn, hygehy@126.com.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

2

Activity coefficients at infinite dilution of organic solutes in the ionic liquid

1-butyl-3-methylimidazolium methyl sulfate

Ming-Lan Gea, Xiao-Mei Deng

a, Li-Hui Zhang

a, Jin-Yuan Chen

a, Jie-Ming Xiong

a,*, and Wen-Hao Li

b

a

Department of Chemical Engineering, Beijing Institute of Petrochemical Technology, Beijing 102617,

People’s Republic of China

b China Petroleum Kelamayi Petrochemical Company, Kelamayi 834003, People’s Republic of China

ABSTRACT: Activity coefficients at infinite dilution (

iγ ) and gas-liquid partition coefficients

( LK ) for organic solutes: alkanes, alkenes, alkyl benzenes, acetonitrile, acetone, tetrahydrofuran,

ethyl acetate, and chloromethanes in the ionic liquid (IL) 1-butyl-3-methylimidazolium methyl

sulfate ([BMIM][CH3SO4]) have been measured by the gas–liquid chromatographic method in

the temperature range of (313.15 to 363.15) K. The values of the partial molar excess enthalpies

at infinite dilution (E,

iH ) were derived from the temperature dependence of the

iγ values. The

entropies (E,

ref iST ) and Gibbs energies (E,

iG ) of organic solutes in [BMIM][CH3SO4] at a

reference temperature Tref =298.15 K were also calculated from the

iγ values. The solubility

parameters of the IL [BMIM][CH3SO4] were also determined by the regular solution theory

(RST). The linear free energy relationship (LFER) analysis of the results was performed to

disclose molecular interactions operating between the IL and the individual solutes.

Keywords: Activity coefficient at infinite dilution, Gas-liquid partition coefficient, Organic

solute, 1-Butyl-3-methylimidazolium methyl sulfate, Solubility parameter, LFER model

1. Introduction

Ionic liquids (ILs) are useful in many chemical applications. They have been used as novel

solvent systems for organic synthesis, separation processes, and electrochemistry [1-5]. For ILs

to be used effectively as solvents, it is essential to know their interaction with different solutes.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

3

Activity coefficients at infinite dilution (

iγ ) and gas-liquid partition coefficients (KL) give a

quantitative measure of interactions between unlike molecules in the absence of solute–solute

interactions, also provide information on the intermolecular energy between ILs and organic

solutes, and can be used to quantify the selectivity and solvent power of ILs [6-9]. Activity

coefficients at infinite dilution have a wide range of applications in the field of chemical

engineering and can be used for the pre-screening of solvents to be used in unit operations such

as extractive distillation and liquid-liquid extraction.

This work continues our studies on the determination of

iγ of various solutes (i) in ILs by the

gas–liquid chromatographic method (GLC) [10-13]. In this paper,

iγ have been measured for

organic solutes: alkanes (hexane, heptane, octane, nonane, decane, cyclohexane,

methylcyclohexane, and 2,2,4-trimethylpentane), alkenes (cyclohexene and styrene), alkyl

benzenes (benzene, toluene, o-xylene, m-xylene, and p-xylene), acetonitrile, acetone,

tetrahydrofuran, ethyl acetate, and chloromethanes in 1-butyl-3-methylimidazolium methyl

sulfate ([BMIM][CH3SO4]) in the temperature range of (313.15 to 363.15) K. The values of the

partial molar excess enthalpies at infinite dilution (E,

iH ) were derived from the temperature

dependence of the

iγ values. The entropies (E,

ref iST ) and Gibbs energies (E,

iG ) of organic

solutes in [BMIM][CH3SO4] at a reference temperature Tref =298.15 K were also calculated from

the

iγ values. The Hildebrand’s solubility parameters of [BMIM][CH3SO4] were calculated as

a function of temperature with the regular solution theory (RST).

Linear free energy relationship (LFER) salvation model of Abraham [14,15] was used to

analyze the obtained experimental data, and the correlative and predictive model was

demonstrated. The correlation parameters were analyzed to understand the interactions that affect

KL, and it is possible to provide valuable information for specific chemical applications.

The chemical structure of [BMIM][CH3SO4] is given in figure 1.

2. Experimental

2.1. Chemicals and materials

The IL [BMIM][CH3SO4] was supplied by Shanghai Chengjie Chemical Co., Ltd. and had a

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

4

purity of >0.99 mass fraction according to manufacturer’s specifications, with the following

certified mass fraction of impurities: w (Cl–) < 5·10

–4, water < 10

–3. Before use, the IL was

subjected to vacuum evaporation at (323 to 333) K over 24 h to remove possible traces of

solvents and moisture. The water mass fraction analyzed by Karl Fischer analysis was less than

4·10–4

. The chemical shift for 1H NMR spectrum (parts per million, D2O) appear as follows: δ

8.596 [s, 1H, H(2)], 7.374 [s, 1H, H(4)], 7.325 [s, 1H, H(5)], 4.111 [m, 2H, NCH2], 3.789 [s, 3H,

NCH3], 3.635 [s, 3H, OCH3], 1.766 [m, 2H, NCH2–CH2], 1.227 [m, 2H, NCH2CH2–CH2], 0.839

[t, 3H, CH3]. The organic solutes were purchased from Beijing Chemical Reagents Company.

Their mass fraction purities were greater than 0.99. The solutes were used without further

purification. The sources and mass fraction purities of materials used are listed in table 1S in

Supplementary Material.

2.2. Apparatus and procedure

The experiments were performed on a SP-3420A gas chromatograph equipped with a thermal

conductivity detector. The column preparation and the packing method used in this work have

been described previously [12,13]. The GC column (stainless steel) with length of 1 m and an

internal diameter of 2 mm was used. Chromosorb P-AW DMCS 80/100 mesh was used as the

solid support and was supplied by SUPELCO. Coating the solid support with [BMIM][CH3SO4]

was performed by dispersing a known mass amount of the Chromosorb in a solution of

[BMIM][CH3SO4] in ethanol followed by evaporation of the solvent in a rotating evaporator. The

Chromosorb was weighed on an electronic balance of precision of ± 0.0001 g before and after the

coating process. The column packing were 48 % mass percent of [BMIM][CH3SO4]. The mass of

the stationary phase [BMIM][CH3SO4] was 0.3794g with a precision of ± 0.0001 g. The

measurements for organic solutes were carried out in the temperature range of (313.15 to 363.15)

K. The column was filled uniformly with the help of an ultrasound vibrator and finally heated

under nitrogen for 8 h at the column temperature of 160 °C. Dry nitrogen was used as the carrier

gas. The flow rate of carrier gas was determined using a calibrated soap bubble flowmeter which

was placed at the outlet after the detector. The flow rate was set for a series of runs and was

allowed to stabilize for at least 15 min before any

iγ determinations were made. The volume of

the samples injected into the GC probes was about (0.05 to 0.5) µL, and the peaks were found to

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

5

be symmetrical, independent of the carrier gas flow rate. The temperature of the GC column was

maintained constant within ± 0.05 K. At a given temperature, each operation was repeated at

least three times to check the reproducibility. The deviation of the retention time of the three

measurements was within ± 0.05 min. The value of the dead time tG was determined with

methane as the nonretainable pure component under the assumption that the effect of the

solubility of methane in the IL was negligible. The measured dead time in the temperature range

has a deviation of ± 0.01 min.

To check the stability of the experimental conditions, such as the possible elution of the

stationary phase by the carrier gas stream, the measurements of retention time were repeated

systematically every (6 to 8) h for hexane and benzene. No change of the retention time was

observed during 80 h of continuous operation.

3. Theory

3.1. Activity coefficients at infinite dilution

In gas-liquid chromatography, the activity coefficient at infinite dilution

iγ were obtained by

the equation proposed by Cruickshank et al [16]. and Everett [17].

o32

20

0Ν

3 2lnlnγ pJ

RT

vBp

RT

vB

pV

RTn iii

iii

i

i

(1)

where iγ is the activity coefficient of solute i at infinite dilution in the stationary phase (3),

0

ip is the vapor pressure of the pure liquid solute i, n3 is the number of moles of the stationary

phase component on the column, and VN is the standardized retention volume obtained by Eq.

(2),

o

0w

f

colGr0

132N 1)()(

p

p

T

TttUJV (2)

where tr denotes the retention time, tG the dead time, U0 the flow rate of the carrier gas, Tcol the

column temperature, Tf the flow-meter temperature, 0

wp the saturation vapor pressure of water

at Tf, and op the pressure at the column outlet.

The second and third terms in equation (1) are correction terms arising from the nonideality of

the mobile gaseous phase and the effect of pressure. Bii is the second virial coefficient of the

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

6

solute, Bi2 is the cross second virial coefficient of the solute (i) with the carrier gas (2), iv is the

liquid molar volume of pure solute, and iv is the partial molar volume of the solute in the

stationary phase (3) at infinite dilution.

For all solutes, values of 0ip were calculated from the Antoine equation, with Antoine

coefficients given by Boublik et al [18]. Molar volumes of solutes vi were estimated using their

experimental densities [19]; partial molar volumes of solutes at infinite dilution

iv have been

assumed to be equal to vi. Bii and Bi2 have been estimated according to the equations suitable for

nonpolar liquids by Tsonopolous’s method [20] with an uncertainty of < ±10 cm3•mol

–1. The

critical parameters needed for the calculations were available from the literature [20]. The cross

critical properties pcij, Tcij, vcij, Zcij, and acentric factor ωij were calculated by using equations

given in the literatures [20,21]. The vapor pressure of the solutes (i) at temperatures of (313.15 to

363.15) K, the critical constants Tc, Pc, Zc, Vc, and acentric factors ω of the solutes and the

carrier gas used in calculation of the virial coefficients are presented in table 2S -3S in the

Supplementary Materials.

The pressure correction term 32J is given by [22]

1/

1/

3

22

oi

3oi3

2

pp

ppJ (3)

where ip and op are the inlet and outlet pressures of the GC column, respectively. The inlet

column pressure ip was determined by inner manometer. Outlet pressure op was kept equal to

atmospheric pressure.

3.2. The regular solution theory and solubility parameters

The activity coefficients at infinite dilution of different solutes in a given solvent can be

used to estimate solubility parameters of the solvents when the solubility parameters of these

solutes are known. These provide a basis for the correlation of the activity coefficients at infinite

dilution of solutes in ILs. The activity coefficient model can be represented by a two-term

equation in which the combinatorial term can be represented by Kikic et al [23] modification to

Flory’s equation, and the residual term is given by the regular solution theory:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

7

rescomb γlnγlnγln iii (4)

3/23/2c o m b )/(1)/l n (γln sisii rrrr (5)

2r e s )δδ)(/(γln siii RTv (6)

where ri and rs are the van der Waals volumes of solute and solvent, respectively; vi is the solute

molar volume; δi and δs are solubility parameters of solute and solvent, respectively. Information

on vi, δi and ri in equations (4) to (7) were obtained from the literature [24]. The ri and δi of the

solutes at T=298.15 K were listed in table 4S in the Supplementary Materials. The van der Waals

volume of the IL was calculated by group contribution method [25]. The solubility parameter of

ILs can be correlated from the experimental ilnγ data. A residual function Yi can be rearranged

from equation (7) according to literature [26].

RTRTRTv

Y si

si

i

ii

22res δδ

δ2δγln

(7)

This equation shows that there is a linear relation between Yi and the solute solubility parameters

δi for a given solvent and temperature T. The value of the solvent solubility parameter δs can be

obtained from the slope of this line. The values of ln γi∞comb

were calculated by equation (5).

According to equation (4), with the ilnγ known (by experimental data), the value of

reslnγi can be calculated, and finally the values of Yi for different solutes in an ionic liquid were

calculated according to equation (7).

3.3. LFER solvation model

Linear free energy relationship (LFER) salvation model was developed by Abraham and

co-workers. This model has been proved to quantify intermolecular solute-stationary phase

interactions governing different chromatographic processes [14], and allows correlating and

predicting various related to solutes from the gas phase to a condensed phase transfer processes.

The representation of the thermodynamic properties is gas-to-liquid partition coefficient

)/( GLL ii ccK , for a volatile solute ( i ) partitioning between an involatile solvent (3) and a carrier

gas (2), KL can be obtained from the following equation [27].

30

3L

γ

ρ

Mp

RTK

ii

(8)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

8

where ρ3 and M3 stand for specific density and molar mass of solvent (IL). 0

ip is the vapor

pressure of the pure liquid solute i. In terms of the LFER theory

lLbBaAsSeEcK Llog (9)

where the capital letters denote the solutes descriptors accounting for respective properties of

solutes, and the lower-case letters are the system constants representing respective contributions

from the IL. The characteristics descriptors of the solutes: E is the solute excess molar refraction

(contribution from interactions with lone pair electrons), S is the solute dipolarity/polarizibility,

A is the solute effective hydrogen bond acidity, B is the solute hydrogen bond basicity, L is the

logarithm of the gas-hexadecane partition coefficient (298 K), which involves the contributions

of solute to the solvent cavity formation and the solute-solvent dispersion interactions. The

system constants are determined by multiple linear regression analysis of experimental log KL

values for a group of solutes of sufficient number and variety to establish the statistical and

chemical validity of the model.

4. Results and discussion

4.1. Activity coefficients at infinite dilution and solubility parameters

Experimental results of

iγ for the solutes in [BMIM][CH3SO4] in the temperature range of

(313.15 to 363.15) K are presented in table 1. The

iγ values for the linear n-alkanes increase

with increasing chain length. The branching of the alkane skeleton (e.g., cyclohexane,

methylcyclohexane, or 2,2,4-trimethylpentane) reduces the values of

iγ in comparison with the

corresponding linear alkanes: hexane, heptane, and octane. The introduction of the double bond

in the six-membered ring (cyclohexene) causes a reduction of

iγ . For the aromatic compounds,

the values of

iγ are distinctly lower in comparison with those of the alkanes, and the values of

iγ increase with increasing size of the alkyl group. The high

iγ values of alkanes indicate low

solubility in the IL and weak solute-IL interactions. Cyclic alkanes, alkenes, and aromatic

molecules interact more strongly with the ILs, as indicated by the lower

iγ values. The smallest

values indicate the stronger interactions between solvent and solute.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

9

The

iγ values of polar solutes was in the order of ethylacetate > tetrahydrofuran > acetone >

acetonitrile. The

iγ values of dichloromethane and trichloromethane are similar, and clearly

lower than that for tetrachloromethane. This behavior indicates that more polar solutes have

better solubilities in the ionic liquid because of the preferred attractive interaction of polar

molecules with the charged ions of the IL and they were prone to strongly retain in the IL

stationary phase.

From the temperature dependence of the activity coefficients at infinite dilution, ilnγ can be

split to its respective enthalpy and entropy terms

R

S

RT

HEi

Ei

i

,,

γln (10)

where R is the gas constant. The temperature dependence of the activity coefficients can be

calculated from equation (11)

/K)(lnγ

T

bai

(11)

Thus the partial molar excess enthalpy E,

iH = Rb and entropy ,E

iS

= Ra at infinite dilution

can be calculated from its slope and intercept, respectively.

The coefficients a and b, the standard deviation σ of the fitted equations, and the values of

iγ at 298.15 K are listed in table 2. The plots of measured ilnγ versus 1/T values and the linear

fit of their data are given in figures. 2 to 5, which show a fairly good fitting quality of equation

(11). The values of E,

iH for the solutes studied are also listed in table 2. For all the 22 solutes,

E,iH all have positive values. This is consistent with the fact that the

iγ values of the solutes

are observed to decrease with an increase in temperature. The limiting partial molar excess Gibbs

energies ii RTG γln

,E of all the studied solutes in [BMIM][CH3SO4] at a reference

temperature 298.15 K are also given in table 2. As seen from table 2, the ,E

iG values of aliphatic

hydrocarbons differs distinctly from that of other solutes. Obviously, the higher the cohesive

energy density of the solute, the more energy has to be spent for breaking the solute-solvent

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

10

interactions during the solution process. In case of aliphatic hydrocarbons this energy penalty is

not compensated by formation of sufficiently strong solute-solvent interactions, which leads to

poor miscibility of the aliphatics with [BMIM][CH3SO4].

The solubility parameter provides a numerical estimation of the degree of interaction between

materials. A large activity coefficient of a solute in a solvent usually means that they have small

mutual solubilities and weak molecular interaction. The linear dependence between Yi and δi at

298.15 K obtained from the

iγ data for [BMIM][CH3SO4] was displayed in figure 6, and the

average solubility parameter δs at 298.15 K is 21.94 MPa1/2

. The solubility parameters of

[BMIM][CH3SO4] from (313.15 to 363.15) K given in table 3, which show that δs approaches to

a constant in the range of temperature studied. The Hildebrand’s solubility parameter is a

measure of solvation capacity of a given solvent. Generally speaking, solute i is expected to be

completely miscible with the solvent, when δi=δs. Therefore, knowing the values of δi and δs

facilitates smart solvent screening and selection.

4.2. LFER solvation model correlation and molecular interaction

The calculation of KL requires the solvent IL density ρ3. The liquid density of

[BMIM][CH3SO4] has been reported in the literature [28], which was calculated from expression

K)/(10772.64119.1)/(ρ 433 Tcmg . The experimental results for the gas-liquid partition

coefficients KL are given in table 5. The LFER model was applied to characterize the nature of

solute interactions with IL [BMIM][CH3SO4]. The experimental KL data of eight alkanols

obtained by Dobryakov [28] was considered, which were methanol, ethanol, 1-propanol,

2-propanol, 1-butanol, 2-butanol, 2-methyl-1-propanol, and 2-methyl-2-propanol. Together with

the KL data in this work, the KL data for 30 organic solutes in the IL [BMIM][CH3SO4] at

temperature T=333.15 K was analyzed and the following equation was obtained shown in

equation (12). Abraham’s solute descriptors used in the LFER solvation model correlation in this

work are listed in table 5S in the Supplementary Materials.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

11

LBA

SEK

)037.0(334.0)173.0(828.0)232.0(043.5

)162.0(225.1)140.0(561.0)132.0(759.0log L

(12)

n=30, R=0.994, R2(adj)=0.986, SD=0.105, F=408.129, p=0.000.

where R is correlation coefficient, R2(adj) is adjusted R square, SD is standard error, F is

covariance ratio and p is significance level. Large correlation coefficients and F values and

relatively small standard error at p=0.000 for equation (12) show an excellent correlation.

In order to examine whether the independent variables in the models interrelate mutually, the

variance inflation factor (VIF), which is a measure of intercorrelation of each independent

descriptor against all other independent descriptor in the regression was introduced. A value of

1.0 indicates no correlation, with values under 5.0 being acceptable. Values over 10.0 indicate an

unstable regression that should be reexamined [29]. The correlation coefficient matrix for

variable and VIF for equation (12) is listed in table 5. The above VIF show that there is no cross

correlation between descriptors.

Figure 7 shows the relationship between the experimental KL and correlated ones by equation

(12). The correlated KL agrees well with the observed. It could be seen that they were well

correlated with each other and mainly spread around a 1:1 straight line. The results indicated that

the new model could successfully calculate the KL of 30 organic solutes in the IL.

Dipolarity/polarizality (s system constants) of [BMIM][CH3SO4] is slightly lower than the

most dipolar/polarizable ionic and nonionic stationary phases. The hydrogen-bond basicity of the

ionic liquid (a system constants) is considerably larger than values obtained for nonionic phases

(0.11 < a < 2.3), and A is the most important descriptor in determining the gas-liquid partition

coefficients. ILs have structural features that would facilitate hydrogen-bond acceptor basicity

interactions (electron-rich oxygen, nitrogen, and fluorine atoms). This is great support for the

idea that the interactions between the -OH group and ionic liquids are very strong.

[BMIM][CH3SO4] also exhibits hydrogen-bond acidic properties as reflected in the non-zero b

system constant.

Considering the magnitudes of its system constants, [BMIM][CH3SO4] can be classified as an

extremely cohesive solvent which possesses a high capacity for participating in lone electron pair

interactions, dipole-type interactions, and hydrogen-bonding type interactions with solutes of

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

12

complementary capabilities. It could be also seen from equation (12), Dipolarity/polarizality (S)

also shows the relative importance in determining KL values and solute-solvent interactions due

to highly polar character caused by coulomb forces acting between the ions in the ILs and lack of

symmetry in cation and anion.

5. Conclusion

In this work, we determined the activity coefficients at infinite dilution of organic solutes with IL

[BMIM][CH3SO4] through GLC measurements. The activity coefficients and the gas-liquid

partition coefficients have been calculated for a series of polar and nonpolar organic solutes at

the temperature range from 313.15 to 363.15 K. The partial molar excess enthalpies at infinite

dilution of the ionic liquid were obtained for the solutes from the temperature dependence of the

experimental activity coefficients at infinite dilution. The solubility parameters of

[BMIM][CH3SO4] were also determined by the regular solution theory and the average solubility

parameter at 298.15 K is 21.94 MPa1/2

. These data are all very important in understanding the

nature of ILs. The results of the LFER prediction of log KL of various solutes indicate that the

model could successfully predict the KL of organic solutes in the IL and disclose molecular

interactions operating between the IL and the individual solutes.

Acknowledgment

This work was supported by Beijing Institute of Petrochemical Technology Undergraduate

Innovative Training Program of Nation Level (Grant No. 2013X00029) and Beijing Municipal

Commission of Education Project (Grant No. KM201210017003).

References

(1) R. D. Rogers, K. R. Seddon, Science. 302 (2003) 792–793.

(2) J. Ranke, S. Stolte, R. Stormann, J. Arning, B. Jastorff, Chem. Rev. 107 (2007) 2183–2206.

(3) A. Heintz, J. Chem. Thermodyn. 37 (2005) 525–535.

(4) C. P. Fredlake, J. M. Crosthwaite, D. G. Hert, S. N. V. K. Aki, J. F. Brennecke, J. Chem. Eng. Data 49

(2004) 954–964.

(5) J. P. Hallett, T. Welton, Chem. Rev. 111 (2011) 3508–3576.

(6) M.-L. Ge, J.-M. Xiong, L.-S. Wang, Chinese Science Bulletin 54 (2009) 2225-2229.

(7) A. Marciniak, Fluid phase equilib. 294 (2010) 213-233.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

13

(8) U. oma s a, M. Królikowska, J. Phys. Chem. B 114 (2010) 8460–8466.

(9) M. Sobota, V. Dohnal, P. Vrbka, J. Phys. Chem. B 113 (2009) 4323-4332.

(10) M.-L. Ge, L.-S. Wang, M.-Y. Li, J.-S. Wu, J. Chem. Eng. Data 52 (2007) 2257–2260.

(11) M.-L. Ge, L.-S. Wang, Q. Zhou, J. Chem. Eng. Data 53 (2008) 1970-1974.

(12) M.-L. Ge, J.-L. Ma, C.-G. Wu, J. Chem. Eng. Data 55 (2010) 1714–1717.

(13) M.-L. Ge, J.-B. Chen, J. Chem. Eng. Data 56 (2011) 3183–3187.

(14) M. H. Abraham, A. Ibrahim, W. E. Acree Jr., Fluid Phase Equilib. 251 (2007) 93-109.

(15) L. M. Sprunger, A. Proctor, W. E. Acree Jr., M. H. Abraham, Fluid Phase Equilib. 265 (2008) 104-111.

(16) A. J. B. Cruickshank, M. L. Windsor, C. L. Young, The Use of Gas–Liquid Chromatography to

Determine Activity Coefficients and Second Virial Coefficients of Mixtures, Proc. R. Soc., London, 1966,

A295, 259–270.

(17) D. H. Everett, Trans. Faraday Soc. 61 (1965) 1637–1645.

(18) T. Boublik, V. Fried, E. Hala, The Vapor Pressure of Pure Substnces, Physical Science Data 17, Elsevier,

Amsterdam, Netherlands, 1984.

(19) H.-Z. Lu, The Handbook of Petrolic Chemical Engineering Data, Chemical Industrial Press, Beijing,

1992.

(20) B.E. Poling, J.M. Prausnitz, J.P. O' Connell, The Properties of Gases and Liquids, McGraw-Hill Book Co,

New York, 2000

(21) J.M. Prausnitz, R.N. Lichtenthaler, E.G. Azevedo, Molecular Thermodynamics of Fluid Phase Eqilibria,

Prentice Hall, New York, 1986.

(22) D. W. Grant, Gas-Liquid Chromatography, Van Nostrand Reinhold, London, 1971.

(23) I. Kikic, P. Alessi, P. Rasmussen, A. Fredenslund, Can. J. Chem. Eng. 58 (1980) 253–258.

(24) C. L. Yaws, Chemical Properties Handbook, McGraw-Hill Book Co, New York, 1999.

(25) A. Bondi, Physical Properties of Molecular Crystals, Liquids and Glasses, Wiley, New York, 1968.

(26) A. L. Revelli, F. Mutelet, J. N. Jaubert, J. Chromatogr. A. 1216 (2009) 4775–4786.

(27) K. Paduszynski, U. Domanska, J. Phys. Chem. B 115 (2011) 8207-8215.

(28) Y. G. Dobryakov, D. Tuma, G. Maurer, J. Chem. Eng. Data 53 (2008) 2154-2162.

(29) G. R. Famini, D. Aguiar, M. A. Payne, R. Rodriquez, L. Y. Wilson, J. Molecular Graphics and Modelling

20 (2002) 277–280.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

14

Appendix A. Supplementary data

TABLE 1S. The sources and mass fraction purities of solutes used

TABLE 2S. The vapor pressure ( 0ip /kPa) of the solutes (i) at temperatures of (313.15 to 363.15) K

TABLE 3S. The critical constants Tc, Pc, Zc, Vc, and acentric factors ω of the solutes and the carrier

gas used in calculation of the virial coefficients

TABLE 4S. The van der Waals volumes (ri) and solubility parameters (δi) of the solutes at T=298.15 K

TABLE 5S. Abraham’s solute descriptors used in the LFER solvation model correlation

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

15

TABLE 1. Experimental activity coefficients at infinite dilution

iγ for various solutes in the ionic

liquid 1-butyl-3-methylimidazolium methyl sulfate ([BMIM][CH3SO4]) at po=101.27kPa and different

temperatures a

T/K

Solutes (i) 313.15 323.15 333.15 343.15 353.15 363.15

Hexane 68.56 61.04 55.16 49.61 44.86 39.85

Heptane 163.07 135.63 116.35 98.95 83.02 72.87

Octane 322.98 273.32 233.72 195.21 159.85 140.62

Nonane 483.43 399.56 350.19 301.48 256.46 218.42

Decane 666.31 587.14 521.70 460.67 398.45 352.83

Cyclohexane 56.06 50.86 47.07 43.91 40.05 36.39

Methylcyclohexane 104.45 93.21 84.46 77.29 71.19 66.47

2,2,4-Ttrimethylpentane 287.38 246.20 208.70 175.66 149.55 126.28

Cyclohexene 15.63 14.35 13.23 12.21 11.48 10.96

Styrene 9.73 9.24 8.77 8.31 7.82 7.36

Benzene 6.08 5.64 5.28 5.02 4.76 4.47

Toluene 15.34 13.82 12.49 11.34 10.16 9.32

o-Xylene 24.28 22.08 20.14 18.19 16.35 14.81

m-Xylene 32.57 28.97 26.02 23.37 20.63 18.60

p-Xylene 30.33 26.98 24.18 21.59 19.05 17.14

Acetonitrile 1.95 1.84 1.73 1.63 1.49 1.38

Acetone 4.31 3.97 3.60 3.25 2.94 2.64

Tetrahydrofuran 5.37 4.95 4.46 4.02 3.65 3.29

Ethyl acetate 12.68 11.59 10.49 9.49 8.41 7.72

Dichloromethane 2.34 2.17 2.06 1.96 1.87 1.74

Trichloromethane 2.12 1.99 1.91 1.80 1.71 1.63

Tetrachloromethane 15.17 13.07 11.70 10.31 9.28 8.34 a Standard uncertainties (u) are as follows: u(po)= ± 0.03 kPa, u(

iγ )=± 5 %, and u(T)= ± 0.02K.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

16

TABLE 2. Coefficients a and b of equation (11), standard deviation σ，

iγ at 298.15 K calculated using

equation (11) , values of the partial molar excess enthalpies at infinite dilution (E,

iH ), entropies

(E,

ref iST ), and Gibbs energies (E,

iG ) of organic solutes in [BMIM][CH3SO4] at a reference temperature

Tref =298.15 Ka

Solute (i) a b/K K15.298γi

E,iH /（kJ·mol

–1）

E,

ref iST /（kJ·mol–1）

E,

iG /（kJ·mol–1） σ

Hexane 0.3553 1214.85 83.9 10.10 -0.88 10.98 0.011

Heptane -0.7799 1840.58 220.0 15.30 1.93 13.37 0.011

Octane -0.3683 1930.77 449.2 16.05 0.91 15.14 0.022

Nonane 0.5290 1770.63 644.0 14.72 -1.31 16.03 0.018

Decane 1.8910 1448.89 855 12.05 -4.69 16.73 0.017

Cyclohexane 0.9766 956.42 65.66 7.95 -2.42 10.37 0.017

Methylcyclohexane 1.3548 1028.75 122.15 8.55 -3.36 11.91 0.018

2,2,4-Ttrimethylpentan

e -0.3070 1875.57 396.90 15.59 0.76 14.83 0.020

Cyclohexene 0.1188 821.81 17.73 6.83 -0.29 7.13 0.016

Styrene 0.2677 631.55 10.87 5.25 -0.66 5.91 0.009

Benzene -0.3786 682.46 6.76 5.67 0.94 4.74 0.006

Toluene -0.9034 1140.05 18.55 9.48 2.24 7.24 0.007

o-Xylene -0.3941 1126.67 29.51 9.37 0.98 8.39 0.014

m-Xylene -0.5785 1274.99 40.36 10.60 1.43 9.17 0.011

p-Xylene -0.7349 1302.07 37.80 10.83 1.82 9.00 0.012

Acetonitrile -1.8336 789.34 2.26 6.56 4.55 2.02 0.019

Acetone -2.0900 1118.10 5.26 9.30 5.18 4.12 0.017

Tetrahydrofuran -1.8805 1120.81 6.55 9.32 4.66 4.66 0.013

Ethyl acetate -1.1125 1148.94 15.50 9.55 2.76 6.79 0.015

Dichloromethane -1.2002 641.19 2.59 5.33 2.98 2.36 0.008

Trichloromethane -1.1494 595.80 2.34 4.95 2.85 2.10 0.006

Tetrachloromethane -1.5952 1349.30 18.73 11.22 3.95 7.26 0.007

a Standard uncertainties (u) are as follows: u(

E,iH )= ± 0.5 kJ·mol

–1, u(

E,

iG )=± 0.5 kJ·mol–1

, and

u( E,

ref iST )=± 0.05 kJ·mol–1

.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

17

TABLE 3. Solubility parameters δs (MPa

1/2) of [BMIM][CH3SO4]

a

T/K Slope Intercept δ from slope δ from intercept R2

313.15 0.0165 0.1983 21.52 22.72 0.986

323.15 0.0161 0.1926 21.59 22.75 0.986

333.15 0.0157 0.1875 21.67 22.79 0.986

343.15 0.0152 0.1826 21.74 22.82 0.986

353.15 0.0149 0.1782 21.84 22.87 0.985

363.15 0.0146 0.1742 21.98 22.93 0.985

aR2 : correlation coefficient

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

18

TABLE 4. Experimental gas-liquid partition coefficients KL for various solutes in the ionic liquid

1-butyl-3-methylimidazolium methyl sulfate ([BMIM][CH3SO4]) at po=101.27kPa and different

temperatures a

T/K

Solutes (i) 313.15 323.15 333.15 343.15 353.15 363.15

Hexane 4.88 3.88 3.11 2.57 2.15 1.87

Heptane 6.20 5.00 4.02 3.35 2.90 2.45

Octane 9.33 7.01 5.36 4.34 3.73 2.99

Nonane 18.35 13.28 9.31 7.07 5.47 4.45

Decane 39.60 25.21 16.67 11.57 8.50 6.30

Cyclohexane 9.03 6.94 5.37 4.22 3.47 2.92

Methylcyclohexane 9.78 7.45 5.75 4.50 3.59 2.87

2,2,4-Ttrimethylpentane 3.35 2.67 2.20 1.76 1.62 1.44

Cyclohexene 36.04 27.24 21.05 16.63 13.19 10.51

Styrene 678.50 435.92 291.55 201.74 144.94 107.12

Benzene 84.33 62.73 47.63 36.45 28.57 23.10

Toluene 103.12 75.38 56.67 43.59 34.82 27.79

o-Xylene 251.59 170.66 119.74 87.65 66.36 51.17

m-Xylene 150.39 105.65 76.11 56.56 43.99 34.34

p-Xylene 155.43 109.36 79.10 59.23 46.18 36.19

Acetonitrile 284.04 205.54 153.40 117.26 94.03 76.51

Acetone 51.30 39.38 31.47 25.84 21.70 18.68

Tetrahydrofuran 57.80 44.14 35.37 28.98 24.09 20.50

Ethyl acetate 39.26 29.02 22.32 17.72 14.65 11.97

Dichloromethane 51.97 40.74 32.03 25.69 20.84 17.64

Trichloromethane 122.63 92.73 70.45 55.85 44.67 36.39

Tetrachloromethane 28.92 23.56 18.96 15.83 13.21 11.25 a Standard uncertainties (u) are as follows: u(po)= ± 0.03 kPa, u(KL)=± 5 %, and u(T)= ± 0.02K.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

19

TABLE 5. The correlation coefficient matrix for the variables and VIF for equation (12)

Variables Correlation matrix

VIF E S A L

E 1.00 2.50

S -0.77 1.00 3.23

A 0.01 -0.12 1.00 1.44

L -0.54 0.60 0.35 1.00 2.24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

20

FIGURE CAPTIONS

FIGURE 1. The chemical structure of [BMIM][CH3SO4]

FIGURE 2. Plot of

iγln vs 1/T for the solutes together with a linear correlation of the data:, hexane;

, heptane; , octane; , nonane; Δ, decane; ×, cyclohexane; , methylcyclohexane; ,

2,2,4-trimethylpentane; — linear fit.

FIGURE 3. Plot of

iγln vs 1/T for the solutes together with a linear correlation of the data: ,

cyclohexene; , styrene. — linear correlation.

FIGURE 4. Plot of

iγln vs 1/T for the solutes together with a linear correlation of the data: , benzene;

, toluene; , o-xylene; , m-xylene; , p-xylene; — linear correlation.

FIGURE 5. Plot of

iγln vs 1/T for the solutes together with a linear correlation of the data: ,

acetonitrile; , acetone; , tetrahydrofuran; , ethylacetate; , dichloromethane; Δ, trichloromethane;

×, tetrachloromethane; —linear correlation.

FIGURE 6. The solubility parameter determination by means of the RST for [BMIM][CH3SO4] at

T=298.15 K

FIGURE 7. Experimental vs. LFER predicted gas-liquid partition coefficients (log KL) of the solutes in

the IL 1-Butyl-3-methylimidazolium methyl sulfate ([BMIM][CH3SO4]) at T=298.15 K

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

21

N N+ -S

O

OCH3O

O

FIGURE 1. The chemical structure of [BMIM][CH3SO4]

2.7 2.8 2.9 3.0 3.1 3.2

3.6

4.0

4.4

4.8

5.2

5.6

6.0

6.4

6.8

ln i

1000K/T

FIGURE 2. Plot of

iγln vs 1/T for the solutes together with a linear correlation of the data:, hexane;

, heptane; , octane; , nonane; Δ, decane; ×, cyclohexane; , methylcyclohexane; ,

2,2,4-trimethylpentane; — linear fit.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

22

2.7 2.8 2.9 3.0 3.1 3.2

2.0

2.2

2.4

2.6

2.8

ln i

1000K/T

FIGURE 3. Plot of

iγln vs 1/T for the solutes together with a linear correlation of the data: ,

cyclohexene; , styrene. — linear correlation.

2.7 2.8 2.9 3.0 3.1 3.21.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

ln i

1000K/T

FIGURE 4. Plot of

iγln vs 1/T for the solutes together with a linear correlation of the data: , benzene;

, toluene; , o-xylene; , m-xylene; , p-xylene; — linear correlation.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

23

2.7 2.8 2.9 3.0 3.1 3.20.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

ln i

1000K/T

FIGURE 5. Plot of

iγln vs 1/T for the solutes together with a linear correlation of the data: ,

acetonitrile; , acetone; , tetrahydrofuran; , ethylacetate; , dichloromethane; Δ, trichloromethane;

×, tetrachloromethane; —linear correlation.

FIGURE 6. The solubility parameter determination by means of the RST for [BMIM][CH3SO4] at

T=298.15 K

14 16 18 20 22 24

0.05

0.10

0.15

0.20

0.25

Y

i (MPa

1/2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

24

0 1 2 30

1

2

3

logK

L

calc

d

logKL

exptl

FIGURE 7. Experimental vs. LFER predicted gas-liquid partition coefficients (log KL) of the solutes in

the IL 1-Butyl-3-methylimidazolium methyl sulfate ([BMIM][CH3SO4]) at T=298.15 K

HIGHLIGHTS

Measurements of activity coefficients at infinite dilution using GLC.

22 Organic solutes in the ionic liquid [BMIM][CH3SO4].

The excess thermodynamic functions, the gas–liquid partition coefficients were calculated.

The solubility parameter of the IL [BMIM][CH3SO4] was determined by the regular solution theory.

The linear free energy relationship (LFER) analysis of the results was performed.