Separation of azeotropic mixtures using protic ionic liquids as extraction solvents

Separation of azeotropic mixtures using protic ionic liquids as extraction solvents

Journal Pre-proof Separation of azeotropic mixtures using protic ionic liquids as extraction solvents Julio E. Sosa, João M.M. Araújo, Eliseo Amado-G...

890KB Sizes 1 Downloads 50 Views

Journal Pre-proof Separation of azeotropic mixtures using protic ionic liquids as extraction solvents

Julio E. Sosa, João M.M. Araújo, Eliseo Amado-González, Ana B. Pereiro PII:

S0167-7322(19)33035-1

DOI:

https://doi.org/10.1016/j.molliq.2019.111733

Reference:

MOLLIQ 111733

To appear in:

Journal of Molecular Liquids

Received date:

28 May 2019

Revised date:

24 July 2019

Accepted date:

10 September 2019

Please cite this article as: J.E. Sosa, J.M.M. Araújo, E. Amado-González, et al., Separation of azeotropic mixtures using protic ionic liquids as extraction solvents, Journal of Molecular Liquids(2019), https://doi.org/10.1016/j.molliq.2019.111733

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof

Separation of Azeotropic Mixtures using Protic Ionic Liquids as Extraction Solvents

LAQV, REQUIMTE, Departamento de Química, Faculdade de Ciências e Tecnologia,

Universidade

Nova

de

Lisboa

(FCT

NOVA),

2829-516

ro

a

of

Julio E. Sosaa,b, João M. M. Araújoa, Eliseo Amado-Gonzálezb and Ana B. Pereiroa,*

Caparica,

Portugal.

Faculty of Basic Sciences, University of Pamplona, Biofuels Lab-IBEAR, Pamplona, Colombia

lP

([email protected], E.A-G.)

re

b

-p

([email protected], J.E.S.; [email protected], J.M.M.A.)

Jo

ur

(+351) 212948318.

na

* Correspondence to: A.B. Pereiro., E-mail: [email protected]; Fax: (+351) 212948550; Tel:

1

Journal Pre-proof Abstract The aim of this work is to evaluate the separation of hydrocarbons (hexane and heptane) from their azeotropic mixtures with ethanol using protic ionic liquid (PIL) as extraction solvents. With this goal in mind, PILs were synthesized and their thermal and physical characterization were carried out. Experimental determination of the phase equilibrium for the ternary systems

of

hydrocarbons + ethanol + PIL at 298.15 K and 101.2 kPa were also carried out in order to evaluate the feasibility of this application. The solute distribution ratio and the selectivity were

ro

also determined to compare the solvent capacity of these PILs. The NRTL equation was used to

-p

correlate the experimental data. Furthermore, this paper provides a comparison of the solvent

re

capacity of these PILs with different extraction solvents (ionic liquids (ILs), ILs mixtures and

lP

deep eutectic solvents) available in the literature. Then, a critical review for the separation of these azeotropic mixtures was carried out using the extraction processes data obtained through

ur

na

the simulation using a conventional software.

processes.

Jo

Keywords: Liquid-liquid equilibria; Protic ionic liquids; Azeotropic mixtures; Extraction

2

Journal Pre-proof 1. Introduction In many areas of industry, the increasing interest related to development of green processes and products makes ionic liquids (ILs) candidates to replace conventional organic solvents in many chemical processes [1]. The possibility of recovering compounds used in industrial processes is not easy, because in many cases the mixtures can form azeotropes [2]. The separation of these azeotropic mixtures into the neat components is needed so that they can be

of

reused, minimizing the costs of industrial processes. Extractive and azeotropic distillation are the

ro

most widely separation processes used to remove one of the components in the azeotropic

-p

system. In this work liquid–liquid separation was selected because it is an advantageous

re

alternative to reduce the energy consumption and the environmental impact of these conventional

lP

processes.

Ionic liquids are widely recognized as green alternatives for engineers not only due to their

na

negligible vapor pressure at room temperature but also for their recyclability [3,4]. Moreover,

ur

ILs can be easily tuned to obtain the appropriated properties for the chemical engineering needs in terms of efficiency and cost of the processes [5]. However, since conventional ILs (aprotic

Jo

ILs) are expensive due to their complex synthesis procedures, the possibility of obtaining ILs from cheap precursors and simple synthesis processes, place the protic ionic liquids (PILs) as a great alternative to traditional ILs [6,7]. These PILs combine excellent properties, such as their high conductivity and exceptional electrochemical properties [7]. Moreover, these compounds present one or more labile hydrogen atoms (in the cationic species) that allow the presence the strong hydrogen bonding interactions. These properties are very interesting in extraction solvents because it can enhance the separation capacity of these compounds. Recently, PILs, like 2-

3

Journal Pre-proof hydroxyethylamonium formate, have received a great attention due to the low costs of synthesis, environmentally friendly nature and suitable solvation properties [8-9]. In petrochemical industry, there are several processes where hydrocarbons (hexane or heptane) and ethanol are present to produce oxygenated additives, reducing the lead in gasoline [10-19]. These compounds form azeotropic mixtures that are conventionally separated using azeotropic distillation [20,21], pervaporation [22], and reverse osmosis [23]. In this paper, PILs

of

based on 2-hydroxyethylamonium cation combined with different carboxylate-based anions were

ro

synthetized and characterized in order to be used as extraction solvents for the removal of

-p

hydrocarbons (hexane or heptane) from its azeotropic mixtures with ethanol. Experimental

re

liquid-liquid equilibrium (LLE) for the ternary systems hydrocarbons + ethanol + PILs were determined as a continuation of our study for the separation of these azeotropic mixtures [10-

lP

19,23]. The PILs capacity to separate these azeotropic mixtures was evaluated using the solute

na

distribution ratio and the selectivity. Furthermore, experimental LLE data were successfully correlated using the NRTL equation [23,24]. This correlation eases the implementation and use

ur

of this model in the simulation studies for extraction processes. Finally, Aspen Hysys software

Jo

was used to optimize the operation conditions for a continuous countercurrent extraction process for the separation of hydrocarbons (hexane or heptane) from its azeotropic mixture with ethanol. The separation capacity of PILs is compared to the capacity of other ILs, ILs mixtures and deep eutectic solvents (DESs) from a critical review of the literature [10-18, 23,25-37]. 2. Experimental Section 2.1. Materials

4

Journal Pre-proof The chemicals and reagents used in this work are shown in Table S1 of Supporting Information (SI) together with the corresponding suppliers and purities. The PILs used in this work were synthesized by a Brønsted-Lowry titration between ethanolamine and the corresponding organic acid (see Table S1 of SI) and were purified as described in literature [8,38]. PILs were dried under vacuum (4 Pa) for at least 48 h with continuous stirring. Karl Fisher (KF) titration (Methonm Ion analysis 831 KF Coulometer) was used to calculate the water

of

content and the results show values less than 1500 ppm. The PILs purities were verified by 1H

ro

NMR (Bruker AVANCE 400 spectrometer). All compounds obtained a purity higher than 98 %

-p

in mass fraction. The chemical structures of these compounds are presented in Table S2 of SI.

re

The 1H NMR characterization obtained for each PIL was the following: [N0002(OH)][HCO2] δH (400 MHz; DMSO) 8.41 (1H, s, H-COO-), 5.97-4.37 (3H, broad signal, NH3+ ), 3.57 (2H, t, N-

lP

CH2-), 2.83 (2H, t, -CH2-O); [N0002(OH)][C2H5CO2] δH (400 MHz; D2O) 3.72 (2H, t, CH2-O), 3.04

na

(2H, t, N-CH2), 2.09 (2H, q, CH3-CH2-COO-), 0.96 (3H, t, CH3-CH2-); [N0002(OH)][C3H7CO2] δH (400 MHz; D2O) 3.74 (2H, t, -CH2-O); 3.06 (2H, t, N-CH2-); 2.07 (2H, t, -CH2-COO-); 1.48 (2H,

Jo

ur

q, CH3-CH2-CH2-); 0.81 (3H, t, CH3-CH2-). 2.2. Thermal Analysis

The decomposition temperatures of the pure [N0002(OH)][HCO2], [N0002(OH)][C2H5CO2] and [N0002(OH)][C3H7CO2] were determined in a thermogravimetric analyser (LABSYS Evo STA). The experiments were performed at atmospheric pressure in platinum pans with heating rates of 1 K·min-1. The onset temperature (Tonset), starting temperature (Tstart) and decomposition temperature (Tdec) were determined and correspond to the temperatures at which the baseline

5

Journal Pre-proof slope changed during heating, the weight loss was less than 1%, and the weight loss was 50%, respectively. Duplicates were done and the uncertainty was better than 2 K. The melting points of the ionic liquids were measured by differential scanning calorimetry using a DSC Q500 (TA Instruments, calibrated with indium standard). Dry nitrogen at a flow rate of about 50 cm3·min-1 was used as the purge gas of the DSC cell. The samples were cooled until 183.15 K, then held during 30 min in isotherm, and heated up to 373.15 K for

of

[N0002(OH)][HCO2], up to 336.15 K for [N0002(OH)][C2H5CO2] and up to 349.15 K for

ro

[N0002(OH)][C3H7CO2]. This procedure was repeated three times at different rates (10 K·min-1, 5

-p

K·min-1 and 1 K·min-1). At the same rate, the results from the second and subsequent cycles were

re

reproducible. The results were analysed using the software Universal Analysis (TA instruments) software. The melting temperature (Tm), solid-solid transitions (Ts) and glass transitions (Tg)

lP

were determined. Duplicates were done and the uncertainty was better than 1 K.

na

2.3. Density, Dynamic Viscosity and Refractive Index

ur

The density and dynamic viscosity of the pure [N0002(OH)][HCO2] were measured in an SVM

Jo

3000 Anton Paar rotational Stabinger viscometer operating at atmospheric pressure in the temperature range from 283.15 K to 343.15 K. Duplicates were done and the reported data are the average value with a maximum relative standard deviation of 1% for viscosity measurements and an uncertainty of the density better than 0.0002 g·cm−3. The global uncertainty of the measurements, considering the purity and the sample handling, is estimated to be 0.1% for the density and 2% for the viscosity. The refractive index of [N0002(OH)][HCO2] was determined using the automatic ABBEMAT 500 Anton Paar refractometer, with an uncertainty in the experimental measurements of ± 4.10-5.

6

Journal Pre-proof Reference fluids (water mili-Q and tetrachlorethylene) were routinely used for the calibration. The verification of the calibration was carried out before each series of measurements in all equipments. 2.4. Liquid-Liquid Equilibrium (LLE) Measurements Ternary LLE data were measured in a glass cell containing a magnetic stirrer with a water

of

jacket connected to a bath controlled at 298.15 K. The temperature in the glass cell was

ro

determined by a thermometer with an uncertainty of ± 0.01 K. For the LLE measurements, 12 ml

-p

of a ternary mixture of known composition was added to this glass cell and was stirred vigorously for 1 h and left to settle for 12 h. Preliminary tests showed that this time was enough

re

to guarantee that the thermodynamic equilibrium was achieved. Then, samples of both layers

lP

were taken with a syringe and the refractive index of these samples were measured. Finally, the

na

compositions of both phases in equilibrium were calculated using the fitting of the refractive index data (calibration curves previously constructed at 298.15 K) with the composition along

ur

the binodal curve using the following equations [10]: (1)

w3=M exp [(N·w10.5) – (O·w13)]

(2)

w3=1- (w1 + w2)

(3)

Jo

nD = A·w1+B·w12+C·w13+D·w14+E·w2+F·w22+G·w23+H·w24+I·w3+J·w32+K·w33+L·w34

where nD corresponds to refractive index, A to O are the adjustable parameters, and w1, w2 and w3 are the mass fraction compositions of hydrocarbon, ethanol and PIL, respectively. The quality of the calibration curves was checked with samples of the binodal curves and the uncertainty of the phase composition is better than 0.005 in mass fraction. These curves were obtained after the determination of the binodal curve. These binodal curves were detected visually using known

7

Journal Pre-proof masses of the three components in a glass cell. Then, ethanol was added slowly until one phase was obtained. All measurements were performed with an uncertainty of ± 10-4 in mass fraction. 2.4. Solid-Liquid Equilibrium (SLE) Measurements The solid-liquid equilibria were determined using a dynamic method with visual detection of the phase transition. Ternary mixtures were prepared using an analytical high-precision balance

of

with ± 0.01 mg resolution in a glass cell containing a magnetic stirrer. This glass cell has a water

ro

jacket connected to a bath which is controlled at 298.15 K (with an accuracy of ± 0.01 K). The

-p

SLE was determined using these known quantities of the three components in the immiscible region where known quantities of ethanol were slowly added maintaining the stirring until the

re

last crystal disappears. The uncertainty of the phase compositions is better than ± 0.006 in mass

lP

fraction.

na

3. Results and Discussion

ur

3.1. Characterization of Protic Ionic Liquids

Jo

The decomposition temperatures and melting points of the pure PILs synthesized in this work determine the range of application of these compounds. The results presented in Table 1 suggest that the increment of the hydrogenated alkyl chain length on the anion causes an increment on the melting temperature. Only [N0002(OH)][HCO2] is liquid at room temperature and presents the highest decomposition temperature (although the differences in this thermal property are not significant). The melting temperatures of [N0002(OH)][C2H5CO2] and [N0002(OH)][C3H7CO2] were not determined because are close or higher than their decomposition temperatures. Finally, the

8

Journal Pre-proof glass transitions were only detected for [N0002(OH)][C2H5CO2] and [N0002(OH)][C3H7CO2] at similar temperatures, close to 208 K.

Table 1 Thermal

propertiesa

of

the

pure

ionic

liquids

[N0002(OH)][HCO2],

Tonset / K

Tdec / K

[N0002(OH)][HCO2]

389

483

486

[N0002(OH)][C2H5CO2]

346

431

[N0002(OH)][C3H7CO2]

359

434

Ts / K

Tg / K

278

192

432

209

436

207

Standard uncertainties of the temperature was ± 2 K.

re

a

Tm /K

-p

Ionic Liquid

of

Tstart / K

ro

[N0002(OH)][C2H5CO2] and [N0002(OH)][C3H7CO2] studied in this work.

lP

Thermophysical properties (density, viscosity and refractive index) of [N0002(OH)][HCO2] were determined between 283.15 and 343.15 K and are reported in Table 2. A comparison with the

na

data available in the literature is illustrated in Fig. S1 of SI [8,9,39-44]. Although there are several experimental data for [N0002(OH)][HCO2], it should be stated that the results are very

ur

different. These differences can be related to the different purities (water or unreacted chemicals)

Jo

of the used ionic liquids. However, the experimental results determined in this work are similar to the reported by Ghatee et al. [41]. These studies are important because there is a lack of accuracy of the most common reported methods and a limited range of temperatures are studied which is a barrier for the industrial application of ionic liquids. Table 2 Density, ρ, dynamic viscosity, η, and refractive index, nD, of the pure [N0002(OH)][HCO2] at 101 kPa.a T/K

ρ/g cm-3

η/mPa·s

nD

283.15

1.2116

485.6

1.47040

9

Journal Pre-proof 1.2090

344.3

1.46941

293.15

1.2065

249.4

1.46841

298.15

1.2041

184.8

1.46743

303.15

1.2017

140.0

1.46645

308.15

1.1994

108.2

1.46549

313.15

1.1971

85.12

1.46454

318.15

1.1948

68.04

1.46357

323.15

1.1925

55.18

1.46260

328.15

1.1902

45.37

1.46166

333.15

1.1882

37.76

338.15

1.1854

31.77

1.45986

343.15

1.1829

27.01

1.45894

-p

ro

of

288.15

a

1.46078

re

Standard uncertainty, u, for pressure, temperature, density, viscosity and refractive index were: u(P) = 0.5 kPa; u(T) = 0.02 K, u(ρ): 0.0002 g·cm−3, u(η): 1% u(nD): 0.00004

lP

3.2. Ternary Diagrams

na

Phase equilibria data for ternary systems hexane (1) + ethanol (2) + PIL (3) or heptane (1) +

ur

ethanol (2) + PIL (3) obtained at 298.15 K are presented in Tables S3 and S4 of SI. The tables include liquid-liquid and solid-liquid equilibrium data, respectively, and these data are also

Jo

plotted in Fig. 1 (comparison of all systems). SLE regions for ternary systems with hexane and heptane are found for the ionic liquids [N0002(OH)][C2H5CO2] and [N0002(OH)][C3H7CO2]. The binodal curves represent the size and shape of the LLE region of these systems. A comparison of these curves shows a slight increase of the immiscible region in the heptane-based systems. The same behaviour is observed when we compare the SLE regions. Moreover, it can be observed that the binodal curve decreases when the length of the alkyl chain in the anion increases. Fig. 1 also represents a clear visualization of the global composition in the different heterogeneous regions. The results show that these PILs are practically immiscible in the hydrocarbons herein

10

Journal Pre-proof tested (hexane and heptane). Furthermore, the composition of ethanol is low in the hydrocarbon rich phase of every studied system. Finally, the positive slopes for the experimental tie lines in all systems suggest that the extraction of ethanol is always possible from their mixture with the

Jo

ur

na

lP

re

-p

ro

of

hydrocarbon in both systems.

11

Journal Pre-proof Ethanol 0.0

Ethanol

a)

1.0

0.1

0.0

0.2

0.6 0.7 0.8 0.9 1.0

1.0

[N0002(OH)][HCO2]

0.0

Heptane 0.0

0.1

0.2

Ethanol

0.0

b)

0.9

0.2

0.8

0.3

-p

0.4

0.6

0.5

0.5 0.4

0.7

0.2

SLE 0.1

1.0

0.0 0.2

0.3

0.4

0.5

0.0 0.1

0.6

0.7

0.8

1.0

0.7

0.4

1.0 [N0002(OH)][C2H5CO2]

0.9

Heptane

0.3

0.1

0.1

0.2

0.4

0.5

0.6

0.4

0.7

0.8

0.9

0.5

0.6

0.7

0.0 0.1

0.8

0.9

f)

0.2

0.8

0.3

0.7

0.4

0.6 0.5 0.4

0.7

0.3

0.8

SLE

0.2

0.9

0002(OH)][C3H7CO2]

[N0002(OH)][C2H5CO2]

0.9

0.6

1.0 [N

1.0

1.0

0.5

0.0 0.3

0.3

Ethanol

0.1

1.0

0.2

0.0

0.0

0.2

SLE 0.2

0.4

1.0

Jo

0.8

0.1

0.5

SLE

0.3

0.9

0.6

0.9

0.4

0.7

0.7

0.8

0.5

0.6

[N0002(OH)][HCO2]

0.8

0.7

0.6

0.5

1.0

e)

0.6

ur

0.8

0.3

0.9

0.9

0.4

c)

0.9

0.2

0.8

1.0

0.1

0.3

na

0.1

Ethanol

Hexane 0.0

lP

0.3

0.8

0.0

0.7

0.5

re

0.6

Hexane

0.6

0.2

0.7

0.9

0.5

Ethanol

1.0

0.1

0.4

ro

0.0

0.3

of

0.9

0.1

1.0

0.0 0.8

0.2

0.9

0.1

0.7

0.3

0.8

0.2

0.6

0.4

0.7

0.3

0.5

0.5

0.6

0.4

0.4

0.6

0.5

0.5

0.3

0.7

0.4

0.6

0.5

0.2

0.8

0.3

0.7

0.4

0.1

0.9

0.2

0.8

0.3

Hexane 0.0

d)

1.0

0.1

0.9

0.1

1.0

Heptane

0.0

0.0 0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

[N0002(OH)][C3H7CO2]

Fig. 1. Triangular phase diagram for ternary system in molar fraction: a) hexane (1) + ethanol (2) + [N0002(OH)][HCO2] (3); b) hexane (1) + ethanol (2) + [N0002(OH)][C2H5CO2] (3); c) hexane (1) + ethanol (2) + [N 0002(OH)][C3H7CO2] (3); d) heptane (1) + ethanol (2) + [N0002(OH)][HCO2] (3); e) heptane (1) + ethanol (2) + [N 0002(OH)][C2H5CO2] (3); f) heptane (1) + ethanol (2) + [N0002(OH)][C3H7CO2] (3) at 298.15 K where (● and ―) are the experimental LLE data; (○ and ― ―) are the NRTL correlation; (● and ―) are the experimental SLE data; and (----) are the experimental data of the binodal curve.

12

Journal Pre-proof 3.3. Distribution Coefficient and Selectivity The distribution coefficient or solute distribution ratio () shows the ratio between the amount of ethanol extracted to the IL rich phases and the amount of remained in the hydrocarbon rich phase. Then, this parameter is related to the solvent capacity of the PILs and determines the amount of compound required for the extraction process. The selectivity, S, the parameter that

of

evaluates the efficiency of the extraction solvent, in this case the extraction of ethanol, from its azeotropic mixture with the inert, in this case hydrocarbon (hexane or heptane). This selectivity

ro

is based on previously calculated  values and provides the number of equilibrium stages needed

-p

in the separation unit (if the selectivity has high values, less equilibrium stages are needed for the

re

separation process). These parameters are widely used in assessing the separation power of

lP

different extraction solvents in liquid-liquid extraction processes and are defined by the

(4) (5)

Jo

ur

na

following expressions:

where x represents the molar fraction, subscripts 1 and 2 are the hydrocarbon and the ethanol, respectively, and superscripts I and II indicate the hydrocarbon rich phase (upper phase) and in the IL rich phase (lower phase), respectively. The results are reported in Table S3 of SI and Fig. 2 depicts the values obtained for the studied ternary systems at 298.15 K. In both cases, the  and S values decrease with the increment of the ethanol concentration in the hydrocarbon rich phase.

13

Journal Pre-proof

15000 S hexane

15000 S heptane

[N0002(OH)][HCO2] [N0002(OH)][C2H5CO2]

12000

[N0002(OH)][HCO2]

12000

[N0002(OH)][C2H5CO2]

[N0002(OH)][C3H7CO2]

[N0002(OH)][C3H7CO2]

9000

9000 800

250

400

125

0 0.00

0.15

0.30

0.45

0 0.00

0.60

800

x2

0.45

0.60

I

800

-p

hexane

heptane

[N0002(OH)][HCO2]

600

600

[N0002(OH)][C2H5CO2]

400

[N0002(OH)][C3H7CO2]

lP

20 15

5 0.15

0.30

0.45

[N0002(OH)][C2H5CO2]

0.60

[N0002(OH)][C3H7CO2]

200 20 15 10 5 0 0.00

0.15

0.30

0.45

0.60

x2I

ur

x2

I

na

10

[N0002(OH)][HCO2]

400

re

200

0 0.00

0.30

ro

of

x2I

0.15

Jo

Fig. 2. Distribution coefficients (β). Selectivity (S) versus molar fraction of ethanol in the hydrocarbon phase for the system hydrocarbon (1) + ethanol (2) + PIL (3) at 298.15 K for : ● [N0002(OH)][HCO2]; ■ [N0002(OH)][C2H5CO2]; ▲ [N0002(OH)][C3H7CO2].

Fig. 2 and Table S3 of SI show not significant difference in these parameters obtained for the systems hydrocarbon + ethanol + PIL meaning that there is no a clear effect due to these anions differences. However, small differences suggest that [N0002(OH)][HCO2] has the best capacity to extract ethanol from its mixture with hydrocarbon. It was found higher differences to compositions lower than 0.1, suggesting better results for [N0002(OH)][HCO2]. These high values may be due to the higher interaction between the –OH group of ethanol with the

14

Journal Pre-proof [N0002(OH)][HCO2]

ionic

liquid.

The

other

two

PILs,

[N0002(OH)][C2H5CO2]

and

[N0002(OH)][C3H7CO2], cannot be used in this range of compositions because they are in the SLE region. 3.4. Correlation using NRTL model In order to perform the simulation studies and process design for the extraction processes, the

of

LLE data must be fitted to a thermodynamic model. Following the methodology widely used for

ro

ILs [16,45], NRTL (Non Random Two Liquid) model was applied to these partially miscible

-p

systems. This model has already demonstrated its ability to satisfactory describe the LLE behaviour of systems involving ILs [16,45]. The third randomness parameter  was optimized

re

and their values are reported in Table 3. The determination of the binary parameters of NRTL

lP

was adjusted to minimize the difference between the experimental and the calculated mole

(6)

ur

na

fraction defined as:

Jo

where: x is the mole fraction; exp and calc are the experimental and calculated values, respectively; subscripts i and l refer to each component and the tie-line, respectively; nc and nt are the total number of components and tie-lines, respectively; and superscripts I and II indicate the hydrocarbon and the IL rich phases, respectively. The NRTL fitting parameters are shown in Table 3 and the consistency of these sets of parameters have been verified (satisfied the Gibbs stability criteria and the isoactivity criterion [46-48]). In this table, the corresponding deviation, RMSD, is also shown which was calculated by applying the following expression: (7)

15

Journal Pre-proof where the subscripts i, l and m are the component, the phase and the tie–line, respectively. The k value indicates the number of experimental tie-lines. The RMSD values are lower than 0.020 for the ternary systems hydrocarbon (1) + ethanol (2) + PIL (3). A comparison between the experimental data and those obtained from the NRTL equation is illustrated in Fig. 1 where the NRTL model describes the liquid-liquid behaviour of

of

the studied mixtures. This paper also provides a critical review on the separation power of the different extraction

ro

solvents used in the literature [10-18, 25-37] for the azeotropic mixtures studied in this work.

-p

The literature review on extraction solvents is summarized in Table S5 of SI where can be

re

verified the great diversity of ILs, deep eutectic solvents (DESs) and mixtures that have been

lP

studied. In order to include these solvents in the simulation studies, new NRTL parameters were calculated for all ternary systems and are summarized in Table S6 of SI. A comparison between

na

the deviations calculated in the literature (see equations in SI) and the RMSDs determine in this

ur

work is plotted in Fig. 3 and listed in Table S6 of SI. In general, the NRTL fitting parameters

Jo

calculated in this work describe better the phase equilibria behaviour of these ternary systems.

16

Journal Pre-proof

Table 3. NRTL parameters obtained for the ternary system hydrocarbon (1) + ethanol (2) + ionic liquid (3). ij

α

RMSD

Hexane (1) + Ethanol (2) + [N0002(OH)][HCO2] (3) 1435.1

189.72

0.10

13

5974.9

-265.59

0.10

23

2291.8

-2291.9

0.10

0.007

of

12

381.23

1022.7

13

4347.1

1124.1

23

-981.15

1815.7

0.28 0.10

0.012

-p

12

ro

Hexane (1) + Ethanol (2) + [N0002(OH)][C2H5CO2] (3)

0.10

916.57

615.79

13

4455.5

-425.63

23

4254.9

lP

12

re

Hexane (1) + Ethanol (2) + [N0002(OH)][C3H7CO2] (3)

-2269.1

0.29 0.09

0.010

0.10

884.55

13

6610.7

23

4574.1

ur

12

na

Heptane (1) + Ethanol (2) + [N0002(OH)][HCO2] (3) -185.72

0.15

-355.59

0.10

-2991.9

0.10

0.004

12 13 23

Jo

Heptane (1) + Ethanol (2) + [N0002(OH)][C2H5CO2] (3)

-48.088

1297.8

0.10

3965.4

1202.1

0.10

218.34

869.62

0.29

0.007

Heptane (1) + Ethanol (2) + [N0002(OH)][C3H7CO2] (3) 12

-2188.7

-175.72

0.28

13

5258.1

-395.59

0.10

23

674.04

-5141.1

0.10

0.005

17

C

1

C Im ][C

0.0

2

[C

C

1

Im 1 SO ][C 4] 1C 1 Im 1P ][C O 4] 2C [C 2P O 2 C [C 4] 1 Im 2C ][C 1 Im ][N 2 SO [C (C 4] 3C F 3S 1 Im O ][N 2) 2] (C F [C 3S O 4C 2) 1 Im 2] [C ][C 4C 1 S [C 1 Im O ][C 4 ] 4C 1 Im 4C ][N 4P [C O (C 6C 4] F 1 Im 3S O ][N 2) 2] (C [C F 3S 6C O 1 Im 2) 2] ][C [C F 3S 6C O 1 Im [C ][N 3 ] 2C (C 1p y] N [N )2 ] [C ( C 3C F 1p 3S y] O [C [N (C 2 )2 ] 4C F 1p y] 3S O [N [C (C 2 )2 ] 2C F 1p 3S yr O ][( 2) N 2] C [C F 4C 3S O 1p yr 2) 2] ][C [C F 3 S 4C O 3] [N 1 pyr ][N 11 14 ] (C [N N (C [N )2 ] F 14 3S 44 ] O [N (C 2 )2 ] 0. 5[ F N 2S O [ 11 N 12 2) 11 (O 2] 63 0. H ] (O ) C 5[ N l + H) ][ B 11 0. F 12 5[ (O 4] 0. C H ] 5[ ) C 2H N l+ 2O 11 1 0. 0. 4] 5[ 67 2(OH C ) ]C [N 3H l+ 11 4O 12 0. (O 4] 5[ H ] ) C C l+ 3H 0. 6O 33 [C 3] 3H 6O 3]

1

[C

1

[C

RMSD

Journal Pre-proof

0.7

a) Literature data This work

0.6

0.5

0.4

f o

0.3

0.2

0.1

l a r P e o r p

n r u

Jo

18

C

1

C Im ][C

0.0

1

RMSD 0.7

[C

Im 1 SO ][C 4] 1C 2C 1P O 1 Im 4] ][C 2C [C 2P O 2C 4] 1 Im [C ][C 2C 2S 1 Im O ][N 4] (C [C F 3C 3S 1 Im O 2) ][N 2] (C F [C 3S O 4C 2) 1 Im 2] [C ][C 4C 1S 1 Im O 4] ][C [C 4C 4C 1 Im 4P O ][N 4] [C (C 6C F 3S 1 Im O ][N 2) 2] (C F 3S O [C 2) 2] 6C 1 Im [C ][P 6C F 1 Im [C ][C 6 ] 2C F 1p 3S y] O [N 3] [C (C F 3C 3S 1p O y] 2) [N 2] (C [C F 3S 4C O 1p 2) yr ][C 2 ] [C F 3S 4C O 1p 3] yr [N ][N 11 (C 14 ] N [N )2 ] 0. (C 67 [N F [C 3S 14 O 2C 0. 44 ] [N 67 2) 1 Im 2] [C (C ][C F 2C 3 S 1 C 1 Im O O ][C 2) 2] 2] +0 1C . 33 O 0. [N 2] 5[ +0 H N .3 4 ]C 11 3[ 12 0. l (O N 67 H H [N ) ]C 4 ][ l+ 11 S C 12 0. (O N 5[ H ] ] ) C C l+ 2H 0. 4O 33 [C 3 ] 3H 6 O 3]

1

[C

1

[C

Journal Pre-proof

b) Literature data This work

0.6

0.5

0.4

f o

0.3

0.2

0.1

l a r P e o r p

Jo

n r u

Fig. 3. Deviation calculated from NRTL correlations for the ternary systems: a) hexane (1) + ethanol (2) + extraction solvent (3); and b) heptane (1) + ethanol (2) + extraction solvent (3).

19

Journal Pre-proof 3.5. Simulation Results The purification of hydrocarbons (hexane or heptane) from its azeotropic mixture by means of an extraction process is carried out in this work. The simulation study is a valuable tool in order to evaluate the possible implementation of this extraction process at industrial scale. Fig. 4 schematically illustrates the flowsheet studied in this work: a countercurrent continuous liquid-

of

liquid extraction column and a solvent recovery stage. In this figure, the liquid-liquid extractor

ro

(with one equilibrium stage at 298.15 K and 101.32 kPa) models the liquid-liquid extraction column and the short-cut distillation models the solvent recovery stage. This extraction process

-p

was optimized via Aspen Hysys V.9 (Aspen Technology Inc., Cambridge, MA, USA). The

re

parameters optimized with the NRTL model at 298.15 K were used to reproduce the phase

lP

behavior of the ternary systems hydrocarbon (1) + ethanol (2) + solvent (3). Operating conditions for the simulation study were selected to decrease the costs while respecting the requirement of

na

an elevated raffinate purity. Both cost and purity increase when the ratio of the solvent/feed flow

ur

rise and the solvent purity increase. If this ratio is minimum, it will be possible to introduce a

Jo

high quantity of azeotropic mixture consuming a minimum of solvent. Besides, the cost of solvent recovery increase when its purity is high. Based on this information, we have selected a solvent/feed ratio lower than 1.5 and a solvent stream of 80%. In these conditions, all solvent streams are liquid and are out of SLE region. Then, the liquid-liquid extractor can be used for the separations of these azeotropes. Then, the solvent recovered in the short-cut distillation unit was always 80%, feeding the liquid-liquid extractor unit. The feed (azeotropic mixture) and solvent compositions were kept constant and the flow rates were optimized to maximize raffinate purity. The Table S7 of SI lists the properties of all streams in the extraction process used to separate ethanol from its azeotropic mixture. Moreover, a comparison between the data obtained for the

20

Journal Pre-proof PILs studied in this work with the other extraction solvents used in the literature [10-18,25-37] is included in Figs. 5 and 6. In general, the ability of these solvents as an azeotrope breaker in extraction processes for the separation of the mixtures hydrocarbon + ethanol has been proved. Taking into account the results, the increment of the alkyl chain on cation and anion increases the purity of raffinate stream (hexane and heptane). These purities obtained in the separation of hexane + ethanol (≥80% in molar fraction) are higher than in the case of heptane + ethanol (see

of

Fig. 5). The high ionicity ionic liquids (HIILs) [10,27] and deep eutectic solvents (DESs)

ro

[18,36,37] are more efficient in the heptane purification. The effect of the ionic liquid’s cation

-p

nature can be studied using the bis(trifluoromelthylsulfonyl)imide-based ionic liquids where the hexane purity in the raffinate stream can be ranked as: [C4C1Im] > [C4C1py] > [N1114]. The best

re

performances in the separations of hydrocarbons from their mixtures with ethanol were achieved

lP

with [C4C1pyr][N(CN)2] in the case of hexane and with [C2C1Im][C1CO2] in the case of heptane.

na

In both separations ILs based on sulphate anion also revealed very good results. Distillation of the extract stream affords the recovery of 80:20 solvent (recycled to the liquid-

ur

liquid extractor) and an ethanol stream containing different alcohol purities which are illustrated

Jo

in Fig. 6. In this steam, similar results are obtained for both azeotropic mixtures hexane + ethanol and heptane + ethanol. In general, almost all extraction solvents achieve ethanol purities greater than 80%. The best ethanol purities were obtained through the DES 0.5 [N1112(OH)]Cl + 0.5 [C2H2O4] for hexane and using [C3C1Im][N(CF3SO2)2] for heptane.

21

of

Journal Pre-proof

Jo

ur

na

lP

re

-p

ro

Fig. 4. Process flowsheet in the purification of hydrocarbons from its azeotropic mixture with ethanol.

22

1

C

0.0

1

Im ][C 1C 1 Im [C ][C 1 SO 4] 2C 1C 1 Im 1 ] [ C PO 4] [C 2C 2P 2C O [C 1 Im 4 2C ][C ] 1 Im [C ][N 2 SO 3C (C 4] 1 Im F ][N 3 SO (C 2 ) [C F 2] 3S 4C O [C 1 Im 2) ][C 2 ] 4C 1 Im 1 S [C ][C O 4] 4C 4C 1 Im [C ] [ N 4 PO 6C 4] (C 1 Im F ][N 3 S (C O2 ) [C F 2] 3S 6C O 1 Im 2) 2] [C ][C F 6 C [C 3S O 1 Im 2C ][N 3 ] 1p (C [C y][ N 3C (C N) 2] 1p F y] 3 [C [N SO 4C (C 2 )2 ] 1p F y] 3S [C O [N 2C (C 2 )2 ] 1p F yr 3S ] O [ C [ N( C 2 )2 ] 4C F 3S 1p O [C yr][ 2 ) C 2] 4C F 1p yr 3 SO [N ][N 3 ] (C 00 02 N (O [N ) H ] ) 00 [H 2 )] 02 C (O O H [N ) ][C 2] 2H 00 02 5 C (O [N H) ][C O2 ] 11 3H 14 ] 7C [ O [N N(C 2] F 14 44 ] 3S 0. O [N 5[ (C 2 ) N [N F 2 ] 11 3S 12 11 O ( O 0. 63 H 2) (O 5[ ) ]C N l + H) ][ 2 ] BF 11 0. 12 5[ ( 0. C 4] 5[ OH) ]C N 2H l 0. + 2O 67 1112( 0 4] . 5[ [N OH C ) ]C 11 3H l+ 12 (O 4O 0. H ) ]C 5[ 4] C l+ 0. 3 H6 O 33 [C 3] 3H 6O 3]

[C

[C

Molar fraction

Journal Pre-proof

1.0 a)

0.8

0.6

f o

0.4

0.2

l a r P e o r p

Jo

n r u

Hexane Ethanol

23

1

[C C

0.0

Heptane Ethanol

1

1

C

1 Im Im ][C 1S [C ][C O 4] 1C 2 [C C1 I 1 P O m 2C ][C 4 ] 1 Im 1 [C ][C CO 2] 2C 2C 1 Im 2 P O [C ][C 4 ] [C 2S 2C 2C O 1 Im 1 Im 4] ] [S [C ][N C 3C (C N ] 1 Im F ][N 3 SO (C 2 ) [C F 2] 3S 4C O [C 1 Im 2 ) ][C 2 ] 4C [C 1 Im 1S 4C ][C O 1 Im 4] [C ][N 4 C 4P 6C ( O C 1 Im F ][N 3 S 4 ] O (C 2 ) F 2 [C 3 S ] O [C 6 C 2) 1I 2 [C 6 C1 I m][ ] PF m 2C ][C 1p 6] [C y][N F3 S O (C 3C 3 1p y] F3 S ] [C [N( O2 ) C 4C F 2] 1p 3 y [C r] SO [C 3 )2 ] 4C F 1p y 3S [ N r] [ O N 3] (C [N 0002 N (O 00 ) H 02 ) ][ H 2] [N (OH C ] ) [C O 00 02 2] 2H (O H 5 C ] ) [C [N O 0. 9 3H 11 2] 0. 14 ] 7C 83 0 [ C O [N [N( [C 2C C 14 1I F 2] 2C 0. 1 Im m][ 44 ][N 3 SO 67 S ( ][C C [ 0. N CF 2 )2 ] 67 C2 C 2 S ]+ 3S O [C 1 I O4 ] + 0.10 2 )2 m 2C ] ][C [ 0 C 1 Im .1 ][C 1 CO 7 [ 2 SO N 4] 0. 1 C 2 ] + H 5[ O 4] 0 [ N 2 ] + .3 [SC 0. 3[ 33 11 N 0. [N 12(O 3 3 NH ] H 11 [ 4 ][C ) ]C N 0. 12(OH l+ H l] 67 4] [N ) ]Cl 0.5 [SC [ + 0. 111 0. C2 H N] 33 2(O [ N H) ] 6 7 [ C 4 O C l+ 2H 3] 0. 1112 33 (OH [N ) ]C 0.33 6 O2 ] [C l+ 11 12 0. 3 H (O 6 H 7 6O ) ]C 3] l + [C 0. 5 H 67 8 O [g ly 3 ] ce ro l]

[C

Molar fraction

Journal Pre-proof

1.0 b)

0.8

0.6

f o

0.4

0.2

l a r P e o r p

Jo

n r u

Fig. 5. Molar fraction of raffinate stream in the steady state calculated from simulations using Aspen Hysys for the purification of: a) hexane; and b) heptane.

24

C

0.0

1

Im ][C 1C 1I [C m][ 1 SO C 4] 2C 1 Im 1 C 1P ][C O 4] [C 2C 2 P O [C 2 C1 I 4] m 2C ][C 1I 2S [C m][ O N 3C (C 4] 1 Im F ][N 3 SO (C 2 ) 2] [C F 3S 4C O [C 1 Im 2) ][C 2 ] 4C 1 Im 1S [C ][C O 4C 4] 1 Im 4C [C 4P ][N O 6C (C 4] 1 Im F ][N 3 S O (C [C F 2 )2 ] 3S 6C O 1 Im 2) ] [C 2 ] [C F [C 6 C1 I 3 SO m 2C ][N 3 ] 1p ( [C y][ N CN 3C (C )2 ] 1p F y] [C [N 3 SO (C 2 )2 4C ] F 1p 3S [C y] [ O N 2C 2) (C 2] 1p F yr ][N 3 SO [C (C 2 )2 ] 4C F 1p yr 3 SO [C ][C 2 ) 4C F 2] 1p yr 3 SO ][N [N 3] (C 0 N [N 002(O )2 ) H 00 ) ][ ] 02 H (O C H O ) ][ C [N 2] 2H 00 02 5 C (O O [N H) ][C 2] 11 3 H 14 ] 7C [ O [N N(C 2] F 14 44 ] 3S O 0. [N 5[ (C 2 )2 ] N [N F3 S 11 12 O (O 11 0. 63 H 2) 5[ ) ]C ( 2 N l + OH) ][ ] 11 B 0. 12 F 0. 5[ (O 5[ H C 4] ) ]C N 2H l+ 0. 11 2O 12 67 0. 4] [N (OH 5[ ) ]C C 11 l+ 3H 12 (O 4O 0. H ) ]C 5[ 4] C l+ 0. 3 H6 33 O [C 3] 3H 6O 3]

1

[C [C

Molar fraction

Journal Pre-proof

1.0 a)

0.8

0.6

f o

0.4

0.2

l a r P e o r p

Jo

n r u

Hexane Ethanol

25

1

[C C

0.0

Heptane Ethanol

1

Im 1C ][C 1 Im ][C 1 SO [C 4] 1C [C 2 C1 I 1 P O m 2C ][C 4 ] 1 Im ][C 1 CO [C 2C 2] 2C 1 Im 2 PO ] [C 4 ] [C [C2 C 2 SO 2C 1 Im 4] 1 ] [C Im] [N [SC 3C N (C 1 Im ] ][N F3 S O (C 2 ) [C F 2] 3S 4C O 1 I [C 2) m 2] ] [C 4 C1 I [C 1 S m 4C O ][C 1I 4] [C m][ 4 C N 6C (C 4 PO 1 Im F 4] ][N 3 SO (C 2 ) 2 [C F3 S ] [C 6 C O2 ) 1 Im 2] [C 6 C ][P 1 Im 2C F ][ 1 [C py][ CF 6 ] N 3 SO 3C (C 1p y] F3 3 ] S [ N [C (C O2 ) 4C F 2] 3S 1p O [C yr] [C 3 )2 ] 4C F 1p y 3S [N r][ O3 ] N [N 000 (C 2( N 00 O )2 ] 02 H )] ( [N OH) ][ [HC C O 00 02 2H 2] (O 5 C H ) ][ [N C O2 ] 0. 3H 11 90 0. 14 ] 83 7C [ [ O [C C2 C [N N(C 2] 1 F 2C 1 Im 44 4 ][ 0. 1 I 3S O 67 m ][S N( ][ C 2 )2 C [ 0. F ] N 67 C2 C C2 S ] + 3 SO [C 1 I O 0 m 4]+ .1 2 )2 ] 2C ][C 0[ 0. 1 Im C ][C 1 CO 17 [ 2S O N 2 ] 1C H 0. O +0 4] 4] 5[ [ [N 2] + .33[ SCN 0. 0. N 33 1112 ] 3 H ( [N OH 3[ N 4 ][C ) ]C 1 H 0. 112( l+ l ] 4 ][ 67 OH SC [N ) ]C 0.5 [ N l C 0. 111 ] 33 2(O + 0 2H [N H) ] .67 4O C [ l + C2 H 3 ] 0. 1112 33 (OH [N ) ]C 0.33 6 O [C 2 ] l+ 11 12 0. 3 H (O 6 H ] 7 6O ) C 3] l + [C 0. 5 H 67 8 O [g ly 3 ] ce ro l]

[C

Molar fraction

Journal Pre-proof

1.0 b)

0.8

0.6

f o

0.4

0.2

l a r P e o r p

Jo

n r u

Fig. 6. Molar fraction of ethanol stream in the steady state calculated from simulations using Aspen Hysys for the purification of: a) hexane; and b) heptane.

26

Journal Pre-proof 4. Conclusion In this work, a series of protic ionic liquids based on 2-hydroxyethylamonium cation were evaluated to extract ethanol from its azeotropic mixture with hydrocarbons (hexane and heptane). These PILs were synthesized and several thermodynamic and thermophysical properties were measured at atmospheric pressure and different temperatures. Experimental LLE data were

of

determined at 298.15 K and these data were used to identify the theoretical operation conditions for extraction process. The corresponding distribution ratios and selectivities were determined.

ro

NRTL model was used to correlate these experimental LLE data. The parameters calculated from

-p

NRTL model allow the implementation and use in simulation studies for the design of extraction

re

process. The design of a countercurrent continuous extraction process including a solvent

lP

recycling stage was optimized using a conventional software. The results obtained with the different extraction solvents used in this work and in the literature demonstrate the efficiency of

na

this extraction process for the purification of hydrocarbons (hexane and heptane) from its

ur

azeotropic mixture with ethanol. Taking into account the raffinate purities obtained in these studies, scaling up for industrial applications can be viable. In this work, liquid-liquid extraction

Jo

process was proposed as alternative to azeotropic or extractive distillation because it can be more environmentally friendly separation process. This separation process is preferred from the economical point of view because the energy requirement is much lower than conventional processes. 5. Acknowledgements Financial support from FCT/MEC (Portugal), through Investigador FCT 2014 (IF/00190/2014 to A.B.P and IF/00210/2014 to J.M.M.A.) and projects PTDC/EQU-EQU/29737/2017, PTDC/QEQ-FTT/3289/2014 and IF/00210/2014/CP1244/CT0003. This work was also supported

27

Journal Pre-proof by the Associate Laboratory for Green Chemistry LAQV (financed by national funds from FCT/MCTES (UID/QUI/50006/2019)) and co-financed by the ERDF under the PT2020 Partnership Agreement (POCI-01-0145-FEDER - 007265). Eliseo Amado-González wants to

Jo

ur

na

lP

re

-p

ro

of

thank the financial support of University of Pamplona, Internal Project 2-2017.

28

Journal Pre-proof 6. References [1]

R. I. Canales, J.F. Brennecke, Comparison of ionic liquids to conventional organic solvents for extraction of aromatics from aliphatics, J. Chem. Eng. Data 61 (2016) 16851699.

[2]

A. B. Pereiro, A. Rodríguez, Mixing properties of binary mixtures presenting azeotropes at several temperatures, J. Chem. Thermodyn. 39 (2007) 1219-1230. M. J. Earle, J. M. S. S. Esperança, M. Gilea, J. N. C. Lopes, L. P. N. Rebelo, J. W.

of

[3]

ro

Magee, J. Widegren, The distillation and volatility of ionic liquids, Nature 439 (2006)

R. D. Rogers, K. R. Seddon, Ionic Liquids - Solvents of the Future, Science 302 (2003)

re

[4]

-p

831-834.

792-794.

J. G. Huddleston, A. E. Visser, W. M. Reichert, H. D. Willauer, G. A. Broker, R. D.

lP

[5]

na

Rogers, Characterization and comparison of hydrophilic and hydrophobic room temperature ionic liquids incorporating the imidazolium cation, Green Chem. 3 (2001)

N. V. Plechkova, K. R. Seddon, Applications of ionic liquids in the chemical industry,

Jo

[6]

ur

156–164.

Chem. Soc. Rev. 37 (2008) 123-150. [7]

J. N. Canongia, J. M. S. S. Esperança, A. Mão de Ferro, A. B. Pereiro, N. V. Plechkova, L. P. N. Rebelo, K. R. Seddon, I. Vázquez-Fernández, Protonic Ammonium Nitrate Ionic Liquids and Their Mixtures: Insights into Their Thermophysical Behaviour, J. Phys. Chem. B 120 (2016) 2397-2406.

[8]

N. Bicak, A new ionic liquid: 2-hydroxy ethylammonium formate, J. Mol. Liq. 116 (2005) 15-18.

29

Journal Pre-proof [9]

S. M. Hosseini, M. M. Alavianmehr, A. Gutiérrez, R. Khalifeh, J. Moghadasi, S. Aparicio, On the properties and structure of 2-hydroxyethylammonium formate ionic liquid, J. Mol. Liq. 249 (2018) 233-244.

[10] F. S. Oliveira, R. Dohrn, A. B. Pereiro, J. Araujo, L. P. Rebelo, I. M. Marrucho, Designing high ionicity ionic liquids based on 1-ethyl-3-methylimidazolium ethyl sulphate for effective azeotrope breaking, Fluid Phase Equilib. 419 (2016) 57-66.

of

[11] A. B. Pereiro, A. Rodriguez, J. Canosa, J. Tojo, Ionic liquid that separates the azeotropic

ro

mixture ethanol HMImPF6, Green Chem. 8 (2006) 307-310.

-p

[12] A. B. Pereiro, A. Rodriguez, Purification of hexane with effective extraction using ionic liquid as solvent, Green Chem. 11 (2009) 346-350.

re

[13] A. B. Pereiro, A. Rodriguez, Separation of ethanol-heptane azeotropic mixtures by

lP

solvent extraction with an ionic liquid, Ind. Eng. Chem. Res. 48 (2009) 1579-1585.

na

[14] A. B. Pereiro, A. Rodríguez, Azeotrope-breaking using [BMIM] [MeSO4] ionic liquid in an extraction column, Sep. Purif. Technol. 62 (2008) 733-738.

ur

[15] A. B. Pereiro, A. Rodríguez, Effective extraction in packed column of ethanol from the

Jo

azeotropic mixture ethanol+hexane with an ionic liquid as solvent, Chem. Eng. J. 153 (2009) 80-85.

[16] A. B. Pereiro, F. J. Deive, J. M. S. S. Esperança, A. Rodríguez, Alkylsulfate-based ionic liquids to separate azeotropic mixtures, Fluid Phase Equilib. 291 (2010) 13-17. [17] P. D. Bastos, F. S. Oliveira, L. P. N. Rebelo, A. B. Pereiro, I. M. Marrucho, Separation of azeotropic

mixtures

using

high

ionicity

ionic

liquids

based

on

1-ethyl-3-

methylimidazolium thiocyanate, Fluid Phase Equilib. 389 (2015) 48-54.

30

Journal Pre-proof [18] F. S. Oliveira, A. B. Pereiro, L. P. N. Rebelo, I. M. Marrucho, Deep eutectic solvents as extraction media for azeotropic mixtures, Green Chem. 15 (2013) 1326-1330. [19] A. B. Pereiro, A. Rodriguez, A study on the liquid-liquid equilibria of 1-alkyl-3methylimidazolium hexafluorophosphate with ethanol and alkanes, Fluid Phase Equilib. 25 (2008) 23-29. [20] L. Laroche, N. Bekiaris, H. W. Andersen, M. Morari, Homogeneous Azeotropic

of

Distillation: Comparing Entrainers. Can, J. Chem. Eng. 69 (1991) 1302–1319.

ro

[21] S. J. Marwil, Separation of hydrocarbon and alcohol azeotropic mixtures by distillation

-p

with anhydrous ammonia, U.S. Patent 4437941, March 1984. [22] T. Okada, T. Matsuura, A study on the pervaporation of ethyl alcohol/heptane mixtures

re

by porous cellulose membranes, Proc. Int. Conf. Pervaporation Processes Chem. Ind. 3rd

lP

(1988) 224–230.

na

[23] M. Laatikainen, M. Lindstrom, Separation of methanol-ethanol and ethanol-normalheptane mixtures by reverse-osmosis and pervaporation. Acta Polytech Scandinavica

ur

Chem. Tech. Metall. 175 (1986) 1-61.

Jo

[24] H. Renon, J. Prausnitz, Local compositions in thermodynamic excess functions for liquid mixtures, AIChE J. 14 (1968) 135-144. [25] F. Cai, M. Zhao, Y. Wang, F. Wang, G. Xiao, Phosphoric-based ionic liquids as solvents to separate the azeotropic mixture of ethanol and hexane, J. Chem. Thermodyn. 81 (2015) 177-183. [26] F. Cai, G. Xiao, Liquid–liquid equilibria for ternary systems ethanol + heptane + phosphoric-based ionic liquids, Fluid Phase Equilib. 386 (2015) 155-161.

31

Journal Pre-proof [27] F. S. Oliveira, R. Dohrn, L. P. Rebelo, I. M. Marrucho, Improving the separation of nheptane

+

ethanol

azeotropic

mixtures

combining

ionic

liquid

1-ethyl-3-

methylimidazolium acetate with different inorganic salts, Ind. Eng. Chem. Res. 55 (2016) 5965-5972. [28] S. Corderí, B. González, Ethanol extraction from its azeotropic mixture with hexane employing different ionic liquids as solvents, J. Chem. Thermodyn. 55 (2012) 138-143. Seoane,

E.

González,

E.

Gonzalez,

1-Alkyl-3-methylimidazolium

of

[29] R.

ro

bis(trifluoromethylsulfonyl)imide ionic liquids as solvents in the separation of azeotropic

-p

mixtures, J. Chem. Thermodyn. 53 (2012) 152-157.

[30] S. Corderí, B. González, N. Calvar, E. Gómez, Ionic liquids as solvents to separate the

re

azeotropic mixture hexane/ethanol, Fluid Phase Equilib. 337 (2013) 11-17.

lP

[31] W. Liu, Y. Ri, K. Ma, X. Xu, Z. Zhu, Y. Wang, Application of 1-hexyl-3-

na

methylimidazolium trifluoromethanesulfonate to the removal of alcohol from mixtures with heptane, Fluid Phase Equilib. 443 (2017) 44-49.

ur

[32] B. González, S. Corderí, A. Santamaria, Application of 1-alkyl-3-methylpyridinium

Jo

bis(trifluoromethylsulfonyl)imide ionic liquids for the ethanol removal from its mixtures with alkanes, J. Chem. Thermodyn. 60 (2013) 9-14. [33] B. González, S. Corderí, Capacity of two 1-butyl-1-methylpyrrolidinium-based ionic liquids for the extraction of ethanol from its mixtures with heptane and hexane, Fluid Phase Equilib. 354 (2013) 89-94. [34] N. M. Aranda, B. González, Cation effect of ammonium imide based ionic liquids in alcohols extraction from alcohol-alkane azeotropic mixtures, J. Chem. Thermodyn. 68 (2014) 32-39.

32

Journal Pre-proof [35] U. Domanska, Z. Zołek-Tryznowska, A. Pobudkowska, Separation of hexane/ethanol mixtures. LLE of ternary systems (ionic liquid or hyperbranched polymer + ethanol + hexane) at T = 298.15 K, J. Chem. Eng. Data 54 (2009) 972-976. [36] A. A. Samarov, M. A. Smirnov, M. P. Sokolova, E. N. Popova, A. M. Toikka, Choline chloride based deep eutectic solvents as extraction media for separation of n-hexane + ethanol mixture, Fluid Phase Equilib. 448 (2017) 123-127.

of

[37] N. R. Rodriguez, B. Molina, M. C. Kroon, Aliphatic + ethanol separation via liquid–

ro

liquid extraction using low transition temperature mixtures as extracting agents, Fluid

-p

Phase Equilib. 394 (2015) 71-82.

[38] B. Janković, N. Manić, R. Buchner, I. P. Korus, A. B. Pereiro, E. A. González, Dielectric

re

spectroscopy and kinetic analysis of nonisothermal decomposition of ionic liquids

lP

derived from organic acid, Thermochim. Acta 672 (2019) 46-52.

na

[39] A. Pinkert, K. L. Ang, K. N. Marsh, S. Pang, Density, viscosity and electrical conductivity of protic alkanolammonium ionic liquids, Phys. Chem. Chem. Phys. 13

ur

(2011) 5136-5143.

Jo

[40] R. Olmos, I. Cota, M. Iglesias, F. Medina, New short aliphatic chain ionic liquids: synthesis, physical properties, and catalytic activity in aldol condensations, J. Phys. Chem. 111 (2007) 12468-12477. [41] M. H. Ghatee, M. Bahrami, N. Khanjari, H. Firouzabadi, Y. Ahmadi, A functionalized high-surface-energy ammonium-based ionic liquid: experimental measurement of viscosity, density, and surface tension of (2-hydroxyethyl)ammonium formate, J. Chem. Eng. Data 57 (2012) 2095-2101.

33

Journal Pre-proof [42] M. Iglesias, A. Torres, R. Gonzalez-Olmos, D. Salvatierra, Effect of temperature on mixing thermodynamics of a new ionic liquid: {2-Hydroxy ethylammonium formate (2HEAF) + short hydroxylic solvents}, J. Chem. Thermodyn. 40 (2008) 119-133. [43] F. M. Mesquita, R. S. Pinheiro, R. S. Aguiar, H. B. Sant’Ana, Measurement of phase equilibria data for the extraction of toluene from alkane using different solvents, Fluid Phase Equilib. 404 (2015) 49-54.

of

[44] X. L. Yuan, S. J. Zhang, X. M. Lu, Hydroxyl ammonium ionic liquids: synthesis,

ro

properties, and solubility of SO2, J. Chem. Eng. Data 52 (2007) 596-599.

-p

[45] A. A. Samarov, M. A. Smirnov, M. P. Sokolova, E. N. Popova, A. M. Toikka, Choline chloride based deep eutectic solvents as extraction media for separation of n-hexane +

re

ethanol mixture, Fluid Phase Equilib. 448 (2017) 123-127.

lP

[46] A. Marcilla, M. M. Olaya, M. D. Serrano, Comments on Liquid-Liquid Equilibrium Data

na

Regression, J. Chem. Eng. Data 52 (2007) 2538-2541. [47] A. Marcilla, J. A. Reyes-Labarta, M. M. Olaya, Should we trust all the published LLE

ur

correlation parameters in phase equilibria? Necessity of their assessment prior to

Jo

publication, Fluid Phase Equilib. 433 (2017) 243-252. [48] I. Díaz, M. Rodríguez, E. J. González, M. González-Miquel, A simple and reliable procedure to accurately estimate NRTL interaction parameters from liquid-liquid equilibrium data, Chem. Eng. Sci. 193 (2019) 370-378.

34

Journal Pre-proof Highlights

ur

na

lP

re

-p

ro

of

The phase equilibria of hydrocarbons + ethanol + PILs systems were measured. The distribution ratio and selectivity were used to compare the solvent capacity. The NRTL equation was used to correlate the experimental data. PILs solvent capacity was compared with different solvents. Extraction processes data were obtained using a conventional software.

Jo

    

35