Liquid–Liquid Extraction of Benzene and Cyclohexane Using

Jun 14, 2018 - In conclusion, the LTTM sulfolane–tetrabutylammonium bromide 5:1 is a promising ... synthesis of cyclohexanol, cyclohexanone, adipyl,...
697 downloads 0 Views 517KB Size
Subscriber access provided by University of South Dakota

Process Engineering

Liquid-Liquid Extraction of Benzene and Cyclohexane Using Sulfolane-Based Low Transition Temperature Mixtures (LTTMs) as Solvents: Experiments and Simulation Shoutao Ma, Jinfang Li, Lumin Li, Xianyong Shang, Shikai Liu, Changyong Xue, and Lanyi Sun Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.8b01524 • Publication Date (Web): 14 Jun 2018 Downloaded from http://pubs.acs.org on June 22, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27 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

Energy & Fuels

Liquid-Liquid Extraction of Benzene and Cyclohexane Using Sulfolane-Based Low Transition Temperature Mixtures (LTTMs) as Solvents: Experiments and Simulation Shoutao Maa, Jinfang Lia, Lumin Lib, Xianyong Shanga, Shikai Liua, Changyong Xuea, Lanyi Suna* a State Key Laboratory of Heavy Oil Processing, College of Chemical Engineering, China University of Petroleum (East China), Qingdao Shandong 266580, China b School of Resources and Chemical Engineering, Sanming University, Sanming Fujian 365004, China

ABSTRACT: The separation of benzene and cyclohexane is considered as one of the most challenging processes in the petrochemical industry. In this paper, low transition temperature mixtures (LTTMs) were used as solvents for the separation of benzene and cyclohexane. The selected LTTMs were sulfolane - tetrabutylammonium bromide 5:1 and ethylene glycol - trimethylamine hydrochloride 5:1, and liquid-liquid equilibrium (LLE) data of benzene-cyclohexane-LTTMs were experimentally determined at 40oC under normal atmosphere. Moreover, the effects of the mole ratio of hydrogen bond donor (HBD) sulfolane and hydrogen bond acceptor (HBA) tetrabutylammonium bromide on extraction performance were also observed based on the LLE data. It is found that when the mole ratio of sulfolane to tetrabutylammonium bromide is 5:1, LTTM has the best extraction performance. In addition, the LLE data of benzene-cyclohexane-LTTMs ternary system were used to fit parameters of the NRTL activity coefficient model. Based on the NRTL model the continuous extraction process was simulated and the operating parameters were obtained, and high product purity (cyclohexane 0.997) and high recovery efficient (cyclohexane 93.28% and benzene 98.25%) can be achieved. In conclusion, the LTTM sulfolane tetrabutylammonium bromide 5:1 is a promising solvent for the extractive separation

ACS Paragon Plus Environment

Energy & Fuels 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

of benzene-cyclohexane mixtures. Keywords: :Liquid-liquid extraction, Low transition temperature mixtures, Benzene, Cyclohexane, Process simulation.

1 INTRODUCTION The separation of aliphatic and aromatic compounds is one of the most challenging and important tasks in petrochemical field1, and benzene and cyclohexane are two representative substances. Cyclohexane is an important organic chemical material which is widely used in the synthesis of cyclohexanol, cyclohexanone, adipyl, hexanediamine, Nylon-6, Nylon-66, polycaprolactam, polyamide fiber and so on. Cyclohexane is less toxic than benzene, and it is an excellent solvent for synthetic rubber, cellulose ethers, resins, oils, asphalt and wax.2,3 With the rapid development of economy, the demand for cyclohexane is increasing. There are two methods of industrial production of cyclohexane: benzene hydrogenation and fractional distillation of petroleum hydrocarbon. And benzene hydrogenation is the main approach for the production of cyclohexane. However, the limitation of reaction equilibrium leads to the impurity of cyclohexane product. Therefore, it is necessary to remove unreacted benzene from the reactor’s effluent stream, which is entrained into the cyclohexane product.4 Since benzene and cyclohexane have very closing boiling points and can even form an azeotrope, traditional distillation method is infeasible for the separation of benzene-cyclohexane mixtures. Therefore, the special methods must be adopted, such as extraction, extractive distillation and azeotropic distillation.5,6 The most attractive separation technique for benzene and cyclohexane is liquid-liquid extraction, especially when the concentration of benzene is less than 20 wt%.4 Liquid-liquid extraction has advantages of low energy consumption due to its mild operating conditions. In addition, the chemical structure and the physical properties of the compounds involved are not affected during the extraction process. However, the key of extraction process is the selection of extractant. A suitable extractant should have

ACS Paragon Plus Environment

Page 2 of 27

Page 3 of 27 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

Energy & Fuels

the advantages of high selectivity and solvent capacity, low volatility, easy regeneration, low viscosity, and high thermal stability.7 Traditional organic solvents are usually selected as extractants, such as sulfolane, furfuryl

alcohol,

ethylene

glycol,

N-methyl

pyrrolidone

(NMP)

and

N-formylmorpholine (NFM). However, these solvents are typically toxic, flammable, volatile and difficult to recover. In order to solve the disadvantages of organic solvents and improve the efficiency of extraction, some scientists begin to focus on the use of mixed solvents. Aspi et. al8 took N,N-dimethylformamide (DMF) + ethanediol as mixed solvents to extract cyclohexane from benzene. It was found that the extraction efficiency of mixed solvents was better than that of single solvent DMF. Dong et al.9 took DMF + ammonium rhodanate (NH4SCN) as mixed solvents to separate the benzene-cyclohexane mixture. The results showed that the selectivity of mixed solvents was 1.88-18.95 and the yield of cyclohexane could reach 94.9%. Though the extraction performance of mixed solvents is better than single solvent, they also have the disadvantages like the traditional organic solvents. Therefore, it is essential to find alternative “green solvents” with the same or better extraction performance.10 Recently, ionic liquids (ILs) have been found to be the promising solvents for extraction because of their high thermal stability, negligible vapor pressure and high solution capacity.11-15 Abu-Eishah et al.16 chose [bmpy][BF4] to separate benzene and cyclohexane, and measured the LLE data at 303.15K under atmospheric pressure. The LLE data were correlated with NRTL model, which showed [bmpy][BF4] performed well in the separation of this mixtures. Lyu et al.17 used COSMO-RS to screen ILs from 12 cations and 22 anions for the separation of benzene and cyclohexane, and found that [C4mim][AlCl4] had a better extraction performance. Song et al.18 selected [C4mim][H2PO4] as solvent for deep desulfurization, and the experiments of IL solubility, liquid-liquid equilibrium, and multistage extraction were carried out to confirm the high performance of [C4mim][H2PO4] for the extractive desulfurization process. Many advantages of ILs make it to be the most attractive solvent. However, it is still a challenge in large-scale industrial application because of its complex ACS Paragon Plus Environment

Energy & Fuels 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

synthesis process, expensive raw materials and unknown toxicity.19-22 In 2003, Abbott et al.23 proposed a new solvent-Deep Eutectic Solvents (DESs) which can exhibit similar characteristic with ILs. DESs have many advantages that ILs do not possess, such as low cost of raw materials, environmental-friendly and facile preparation. However, DESs cannot contain all kinds of these solvents, and many DESs have glass-transition temperature rather than melting points. Thus DESs are also called low transition temperature mixtures (LTTMs).23 Compared with other solvents, LTTMs have an important advantage that their structure can be adjusted according to the selected raw materials and the mole ratio of hydrogen bond donor (HBD) and hydrogen bond acceptor (HBA). Besides that, the properties of LTTMs can be affected by the mole fraction of components and the temperature. In addition, PH and water content can also affect the physical and chemical behavior of LTTMs.24,25 LTTMs also have been studied in the field of liquid-liquid extraction. Rodriguez et al.26 chose glycolic acid - choline chloride 1:1 and lactic acid - choline chloride 2:1 to separate the hexane + ethanol mixture and heptane + ethanol mixtures. Compared with other solvents, the two types of LTTMs have better extraction performance. Hadj-Kali et al.27 took glycolic acid - ethylene glycol and tetraethylammonium p-toluenesulfonate - sulfolane as solvents to separate ethylbenzene and octane. The ternary equilibrium data at 298.15K under the normal atmosphere were obtained, which showed tetraethylammonium p-toluenesulfonate - sulfolane had better extraction performance, high distribution ratio and selectivity. Mulyono et al.10 took DESs tetrabutylammonium bromide - sulfolane as solvent, and measured the LLE data of benzene, toluene, ethylbenzene and m-xylene (BTEX) aromatics with octane. The results showed that DESs had high selectivity and distribution coefficient. Considering the better performance of LTTMs in the field of separating cycloalkanes and aromatics mixtures, the extraction performance of LTTMs with different types and mole ratios of HBD and HBA should be investigated. The objective of this work is to select suitable LTTMs as extractant to separate benzene from mixtures with cyclohexane based on different scales. At first, LLE data ACS Paragon Plus Environment

Page 4 of 27

Page 5 of 27 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

Energy & Fuels

of the ternary systems benzene + cyclohexane +LTTMs are experimental determined, and the extraction performance of LTTMs with different HBDs and different mole ratios of HBD and HBA is studied and discussed. Secondly, the extraction performance of LTTMs are explained by the interaction between molecules with respect to the σ-profile. However, most of the reported ILs and LTTMs cannot provide a high mass-based extraction efficiency due to their higher molecular weight, which may obstruct the applications.17,28,29 Therefore, based on the NRTL model the continuous extraction process is simulated and the operation parameters are optimized through sensitivity analysis. What’s more, since the molar content of the unreacted benzene in mixtures with cyclohexane in practice is lower than 20%, special attention should be paid to the removal of the reactant at low-concentrations of benzene.4,17

2 EXPERIMENT 2.1 Chemicals The chemicals used were benzene, cyclohexane, sulfolane, tetrabutylammonium bromide, ethylene glycol and trimethylamine hydrochloride, which were purchased from Aladdin Industrial Corporation. All chemicals are of high purity (>99 wt%) and used without any further purification. The LTTMs (see Table 1) were prepared according to the method described by Abbott et al.23 The HBA salts were mixed with HBD sulfolane or HBD ethylene glycol with different mole ratios in screw-capped bottles. The bottles were then stirred in an incubating-shaker at the temperature of 100 o

C and the rotational speed of 200 rpm until a clear liquid was formed. Table 1 LTTMs studied in this work. Abbreviation

HBA

HBD

mole ratio

LTTM1

trimethylamine hydrochloride

ethylene glycol

1:5

LTTM2

tetrabutylammonium bromide

sulfolane

1:5

LTTM3

tetrabutylammonium bromide

sulfolane

1:3

LTTM4

tetrabutylammonium bromide

sulfolane

1:7

ACS Paragon Plus Environment

Energy & Fuels 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

2.2 Liquid-Liquid Equilibrium LLE data of the ternary system benzene + cyclohexane + LTTMs were measured at 40 oC under the normal atmosphere. The experimental procedure and instrument were discussed detailly in our current work, and the accuracy of the measuring systems was also evaluated in the literatures.30,31 In this section, a jacked glass cell of 100 cm3 with a magnetic was used as the LLE apparatus, and the experimental apparatus was shown in Fig S1 in the supporting information.30 The temperature of the system was controlled by thermostatic water bath with the uncertainty of around 0.01K, and the gas chromatography (GC) was used to determine the composition of the upper and lower layer under different stirring time and still time. It was found that when the liquid mixtures were continuously agitated with a magnetic for 1 h and then left to settle for 1h, thermodynamic equilibrium can be established completely. Then the samples of the upper and lower layer were taken from the instrument and detected by GC. The samples were calibrated with solutions prepared by gravimetrical standard. FID was used together with PEG-30M capillary column (0.32 mm inner diameter, 30 m length, 0.33 um film thickness). The GC is equipped with a pre-column as the LTTMs cannot be analyzed due to its negligible vapor pressure. The carrier gas is hydrogen with high purity (99.999 wt%), and its flow rate is 30 ml/min. The temperature of the detector and injector were set at 373.15 K. The temperature of column was maintained at 333.15 K. Quantitative results from GC analysis were obtained using area normalization. Finally, the content of LTTMs in each phase can be obtained by subtracting the sum of the measured mass fractions of benzene and cyclohexane from unity.17 The analysis was performed at least three times for each sample with standard deviation less than 0.10%. According to GUM standard32, the uncertainty of two liquid layers compositions was calculated. The absolute deviation of the compositions was estimated to be less than 1.5wt%. Changing compositions of the compounds by Mettler AX205 balance with precision of 0.02mg, and a series of tie-lines data were

ACS Paragon Plus Environment

Page 6 of 27

Page 7 of 27

obtained.

2.3 Data Reliability Verification In this paper, the Othmer-Tobias equation is used to access the reliability and consistency of LLE data.33,34 The data corresponding to the two endpoints of the tie line measured by the experiment is analyzed by the Othmer-Tobias equation as shown in equation (1).  1 − w3II   1 − w2I  ln    = a + b ln  I II  w2   w3  I

(1)

II

Where w2 and w3 are the mass fraction of cyclohexane in the raffinate phase and LTTMs in the extraction phase, respectively. a and b are the parameters of Othmer-Tobias equation. The straight lines plotted by this equation are shown in Fig. 1, the fitting parameters together with corresponding linear regression coefficients R2 close to 1, which indicates the reliability of the experimental LLE data. 1.0 0.5 0.0

ln((1-wI2)/wI2 )

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

Energy & Fuels

-0.5 -1.0

2

LTTM1, R =0.9837 2 LTTM2, R =0.9936 2 LTTM3, R =0.9842 2 LTTM4, R =0.9910

-1.5 -2.0 -2.5

-3.0 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 II

II

ln((1-w3 )/w3 )

Fig. 1 Othmer–Tobias plot of the ternary system benzene+ cyclohexane + LTTMs.

3 EXTRACTION PERFORMANCE In extraction process, the amount of solvent is determined by the distribution coefficient, and the purity of product is determined by the selectivity of solvent. For the separation of benzene and cyclohexane, the solvent should have high solubility for benzene and meanwhile the reasonable separation selectivity. In this section, three

ACS Paragon Plus Environment

Energy & Fuels 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

Page 8 of 27

indicators are used to judge the extraction performance of LTTMs: the mass-based benzene distribution coefficient (βA), the solvent selectivity (S) and the performance index (PI) of solvent. Therefore, the extraction performance of LTTMs which including different types of HBD and HBA was observed. In addition, the effects of mole ratio of HBD to HBA on extraction performance were explored for the LTTM that had higher PI.

3.1 Distribution Coefficient The solute (benzene) distribution coefficient (βA) is defined as the ratio of mass fraction of the solute in the extraction phase to the mass fraction of the solute in the raffinate phase, and the equation is shown as follows:

w1II βA = I w1

(2)

As shown in Table S1 and Table S2, the benzene distribution coefficients in LTTM1 and LTTM2 are lower than 1, which determines much more amount of solvent required for benzene extraction from cyclohexane. Most importantly, it is noticed that sulfolane or ethylene glycol, and hence the LTTMs, is not found in the raffinate phase. This may be attributed to the hydrogen bonding between the HBA and HBD. This is very important from an industrial point of view. Because if pure sulfolane or ethylene glycol is used as a solvent for extraction, its concentration in the raffinate layer may reach 20 wt%, which may cause solvent loss and require additional separation steps.9

3.2 Selectivity The selectivity is the extraction ability of solvent that is the ratio of solute (benzene) distribution coefficient to that of the carrier (cyclohexane). The mass-based selectivity is expressed by equation (3). S=

w1II w2I w1I w2II

ACS Paragon Plus Environment

(3)

Page 9 of 27 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

Energy & Fuels

Where subscripts 1 and 2 represent benzene and cyclohexane, respectively; superscripts II and I represent the extraction and raffinate phases, respectively. The higher of the selectivity of LTTMs is, the better performance of the separation of benzene and cyclohexane is. It is desirable to have LTTMs with a high selectivity to benzene which means that fewer stage are required to remove them. This will reduce the capital cost and energy consumption in the solvent recovering stage. As shown in Table S1 and Table S2, it can be seen that the selectivity of LTTM1 and LTTM2 for benzene is larger than 1, which shows the feasibility of LTTMs used as the solvent. It can be noticed that the solvent selectivity decreases with the increase of the concentration of cyclohexane in the raffinate phase, in most cases.

3.3 Performance Index of Solvent To evaluate the overall extraction efficiency of a solvent, PI is calculated by equation (4).17,35,36 PI = β A × S

(4)

In the following section, the three indicators, which are selectivity, distribution coefficient and performance index, are calculated according to the measured LLE data under different conditions, and thus the extraction performance of different types of LTTMs can be compared quantitatively. As Tables S1-S4 indicate, it can be seen that PI of LTTM2 is much higher than that of other three LTTMs at the low benzene mass fraction, which can demonstrate LTTM2 is the proposing solvent for the separation of benzene-cyclohexane mixtures.

3.4 Effects of Different HBAs and HBDs on Extraction Performance The LLE data of two different LTTMs named LTTM1 and LTTM2 were determined experimentally. The benzene distribution coefficient and selectivity of LTTM1 and LTTM2 are listed in Table S1 and Table S2 of the supporting information,

ACS Paragon Plus Environment

Energy & Fuels

respectively. And the selectivity of LTTMs is plotted together with the benzene mass fraction in the raffinate phase (see Fig. 2). It can be seen that the selectivity of LTTM1 and LTTM2 is larger than 1, which indicates the feasibility of LTTMs used as solvent. The selectivity decreases with the increase of benzene mass fraction in the raffinate phase, which shows that the higher mass fraction of benzene in the raffinate phase is, the worse the selectivity of LTTMs is. This trend is consistent with the results reported earlier in the literature for other systems using different DESs.10 It can be concluded that LTTM1 and LTTM2 have higher selectivity and are more conducive to the extraction of low concentrations benzene from cyclohexane. Moreover, the selectivity of LTTM1 is larger than that of LTTM2. The reason is that the cyclohexane distribution coefficient in LTTM1 is much lower than that in LTTM2 (see Tables S1 and S2), which demonstrates LTTM2 has high capacity for benzene and cyclohexane than that of LTTM1. 30 LTTM1 LTTM2

25 20

S

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

Page 10 of 27

15 10 5 0 0.0

0.1

0.2

0.3

0.4 I

0.5

0.6

0.7

0.8

w1

Fig. 2 Selectivity of LTTM1 and LTTM2.

Fig. 3 shows the relationship between the distribution coefficient and the mass fraction of benzene in the raffinate phase. It can be seen the benzene distribution coefficient is lower than 1, which shows that only a small amount of benzene is extracted in the single-stage extraction process. This is also the reason that more amount of LTTMs is needed in the multi-stage extraction process. However, considering that LTTMs can be reused and cheaper than ILs, the large amount of LTTMs required is not a major problem in the view of industry. It can also be seen LTTM2 has a much higher benzene distribution coefficient than LTTM1. The

ACS Paragon Plus Environment

Page 11 of 27

distribution coefficient of benzene in LTTM1 decreases with the increase of the mass fraction of benzene in the raffinate phase, while the distribution coefficient of benzene in LTTM2 increases. This indicates that high benzene concentration favors its dissolution in LTTM2 and disfavors its dissolution in LTTM1. In addition, as Tables S1 and S2 indicates, the distribution coefficient of cyclohexane in LTTM1 is much lower than that in LTTM2, which demonstrates the capacity of LTTM1 for benzene and cyclohexane is lower than LTTM2. This is also the reason that LTTM1 selectivity for benzene is higher than that of LTTM2. 0.8 LTTM1 LTTM2

0.7 0.6

ßA

0.5 0.4 0.3 0.2 0.1 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

w I1

Fig. 3 Distribution coefficient of benzene in LTTM1 and LTTM2. Fig. 4 shows the relationship between PI and the mass fraction of benzene in the raffinate phase. It is found that the LTTM2 shows a larger PI value over the whole benzene

concentration

region,

which

indicates

LTTM2

sulfolane

-

tetrabutylammonium bromide 5:1 is more suitable for the benzene extraction than LTTM1 ethylene glycol - trimethlyamine hydrochloride 5:1. 10 LTTM1 LTTM2

8 6

PI

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

Energy & Fuels

4 2 0 0.0

0.1

0.2

0.3

0.4 I

0.5

0.6

w1

Fig. 4 PI of LTTM1 and LTTM2.

ACS Paragon Plus Environment

0.7

0.8

Energy & Fuels

3.5 Effects of Mole Ratio of HBD to HBA on Extraction Performance In section 3.4, the extraction performances of two kinds of LTTMs which contain different HBDs and HBAs are studied and found that the PI of LTTM2 containing HBD sulfolane and HBA tetrabutylammonium bromide is better than that of LTTM1 ethylene glycol - trimethlyamine hydrochloride 5:1. Therefore, in this section, the effects of mole ratio of HBD sulfolane to HBA tetrabutylammonium bromide on extraction performance are observed. The LLE data of LTTMs with different mole ratios of HBD to HBA were experimentally determined and shown in Tables S3, S4 and S5 of the supporting information. Fig. 5 shows the relationship between the selectivity of LTTMs with different mole ratios of HBD to HBA and the mass fraction of benzene in the raffinate phase. It indicates that the selectivity of LTTM2 sulfolane tetrabutylammonium bromide 5:1 is larger than that of LTTM3 sulfolane tetrabutylammonium bromide 3:1 and LTTM4 sulfolane - tetrabutylammonium bromide 7:1. Therefore, although sulfolane as an HBD has better selectivity, it has negative effect on the selectivity when sulfolane reaches a certain amount. The reason may be that more cyclohexane is extracted than benzene with the increase of the amount of sulfolane. 18 HBD:HBA=3:1 HBD:HBA=5:1 HBD:HBA=7:1

16 14 12 10

S

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

Page 12 of 27

8 6 4 2 0 0.0

0.1

0.2

0.3

I

0.4

0.5

0.6

0.7

w1

Fig. 5 Selectivity of LTTMs with different molar ratios of HBD to HBA.

The distribution coefficient of LTTMs with different mole ratios of HBD to HBA is also observed and shown in Fig. 6. It can be seen that the distribution coefficient

ACS Paragon Plus Environment

Page 13 of 27

increases with the raise of the mass fraction of benzene in the raffinate phase, while the mole ratio has insignificant effects on the distribution coefficient, and the best mole ratio of sulfolane to tetrabutylammonium bromide is 3:1 for the distribution coefficient. HBD:HBA=3:1 HBD:HBA=5:1 HBD:HBA=7:1

0.56

ßA

0.52

0.48 0.44

0.40 0.0

0.1

0.2

0.3

0.4

0.5

0.7

0.6

w I1

Fig. 6 Distribution coefficient of LTTMs with different mole ratios of HBD to HBA.

Fig. 7 shows the PI of LTTMs with different mole ratios of HBD to HBA. It can be seen that when the mole ratio of sulfolane to tetrabutylammonium bromide is 5:1, the PI of LTTM2 is the best. The reason is that benzene distribution coefficient in LTTM2 is much higher than that in LTTM3 and LTTM4. In summary, when the mole ratio of sulfolane to tetrabutylammonium bromide is 5:1, the PI of LTTM2 is larger than that of the LTTMs with other mole ratios. 10 HBD:HBA=3:1 HBD:HBA=5:1 HBD:HBA=7:1

8 6

PI

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

Energy & Fuels

4 2 0 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

I

w1

Fig. 7 PI of LTTMs with different mole ratios of HBD to HBA.

ACS Paragon Plus Environment

Energy & Fuels 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

4 Sigma-Profile Analysis The extraction performance of LTTMs can be explained by the interaction between molecules with respect to the σ-profile that has been explained in details by Klamt.37 The interaction of molecules can be reflected in the σ-profile, and a higher absolute value of screening charge density (σ) leads to a stronger compound as an HBD or an HBA. Therefore, the difference in the solubility and selectivity of LTTM1 and LTTM2 for benzene and cyclohexane is studied according to the σ-profile since LTTM1 and LTTM2 have different types of HBD and HBA. In COSMO-RS theory, σ-profile obtained from quantum chemical calculations is one of the most important molecule-specific properties.37-39 It is the distribution of the σ of molecule, which is divided into three regions, namely the nonpolar region (−0.0084 e/Å-2 < σ < 0.0084 e/Å-2), the HBD region (σ < −0.0084 e/Å-2) and the HBA region (σ > 0.0084e/Å-2). According to COSMO-RS, a molecule σ-profile in the region of σ > 0.0084 e/Å-2 and σ < −0.0084 e/Å-2 indicates its HBA ability and its HBD ability, respectively.38 Therefore, as seen in Fig. 8, the σ-profile of benzene clearly shows that its charge density results are mainly between -0.009 and +0.009 eÅ-2 which demonstrates that benzene is weak for hydrogen bond interactions, and also can be classified as a nonpolar compound. For benzene, the positive and negative peaks in σ-profile are derived from carbon atoms and the hydrogen in the benzene ring, respectively. The symmetry of these two peaks indicates a strong interaction formed between benzene and itself.38 This is the reason that benzene has a higher boiling point and surface tension. Similar to benzene, the peak of σ-profile of cyclohexane is also approximately symmetric. Hence its nature is similar to that of benzene, and both are nonpolar compounds.

ACS Paragon Plus Environment

Page 14 of 27

Page 15 of 27

35 benzene cyclohexane

30 25 20

p(s )

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

Energy & Fuels

15 10 5 0 -0.03

-0.02

-0.01

0.00

0.01

0.02

0.03

s /(eÅ-2)

Fig. 8 σ-profile of benzene and cyclohexane.

Fig. 9 shows the σ-profile of LTTMs containing different types of HBD and HBA. It can be seen that the peaks of LTTMs are located between -0.02 eÅ-2 and 0.02 eÅ-2. As Fig. 9 indicates, LTTM2 has three peaks and the first peak covers a large area at -0.005 eÅ-2, and it overlaps with the peaks of benzene and cyclohexane. So LTTM2 can interact with benzene and cyclohexane, and that is why it has big solubility for benzene and cyclohexane. Whereas the peak of LTTM1 in this region is 41.5% lower than that of LTTM2, and there is less overlap with the peaks of benzene and cyclohexane. Moreover, the higher peak at -0.005 eÅ-2 for LTTM2 results in a favorable interaction with the right peak of benzene (0.006eÅ-2).Therefore, LTTM1 has lower solubility for benzene and cyclohexane compared with LTTM2. This result is consistent with that the PI of LTTM1 is less than LTTM2. There are slight value of negative screening charge in HBD region (the second peak), leading to considerably few hydrogen bond interaction. The third peak of LTTM2 is in HBA region and does not interaction with the peaks in nonpolar regions. Therefore, the third peak in HBA region has little effect on the extraction performance (and is thus unable to interact with benzene in the nonpolar region).4 In summary, the peaks locating between -0.005 eÅ-2 and 0.005 eÅ-2 of LTTM2 are larger than that of LTTM1, which indicates the interaction between LTTM2 and benzene is stronger than that between LTTM1 and benzene. This may be also the reason that the PI of LTTM2 is larger than that of LTTM1.

ACS Paragon Plus Environment

Energy & Fuels

benzene cyclohexane LTTM2 LTTM1

140 120 100 80

p(s )

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

Page 16 of 27

60 40 20 0 -0.03

-0.02

-0.01

0.00

-2

0.01

0.02

0.03

s /(eÅ )

Fig 9 σ-profile of LTTMs containing different HBDs and HBAs.

5 PROCESS SIMULATION AND ANALYSIS 5.1 NRTL Model The correlation of LLE data is an important step in the study of phase equilibrium, which can be used for process simulation. In this section, the binary interaction parameters of NRTL model were obtained by minimizing the differences between the experimental and calculated equilibrium mass fractions for each of the components over all the experimental tie lines in the ternary system.40,41 The objective function (OF), which describes the differences between the experimental and calculated equilibrium data for each of the components over all the experimental tie lines, is given in equation (5). n

2

3

exp cal OF = ∑∑∑ ( wijk − wijk )

2

(5)

k =1 j =1 i =1

1/2

 n 2 3 ( wexp − wcal )2    ijk ijk RMSD = ∑∑∑  6n  k =1 j =1 i =1 

(6)

Where n is the number of tie lines, wexp is the experimental mass fraction, and wcal is the calculated mass fraction. The subscripts k, j and i refer to the components, the phases and the tie lines, respectively. The root-mean-square deviation (RMSD) values are applied to evaluate the quality of correlation from NRTL model, given by

ACS Paragon Plus Environment

Page 17 of 27 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

Energy & Fuels

equation (6), the corresponding n, wexp, wcal , k, j, i are the same with those in the OF equation. The binary interaction parameters and the RMSD values are shown in Table 2, the mass-based RMSD between the experimental and calculated LLE compositions are evaluated to provide a measure of the correlation accuracy. The calculated RMSD values are between 0.0016 and 0.0097, which demonstrates the NRTL model fits the experimental data well. Table 2 NRTL binary interaction parameters for benzene (1) - cyclohexane (2) - LTTMs (3) system. NRTL parameters

RMSD

i-j

LTTM1

LTTM2

LTTM3

LTTM4

aij

aji

bij/K

bji/K

∂ ij

1-2

33.4155

30.0076

-9534.23

-9922.45

0.1

1-3

43.4127

31.2005

-10000

-10000

0.2

2-3

40.9895

20.3317

-10000

-5041.51

0.1

1-2

-28.3226

40.0277

6993.4833

-10000

0.1

1-3

-17.0623

-37.2826

10000

10000

0.1

2-3

-18.5711

-32.0396

10000

10000

0.1

1-2

9.6511

0.9442

2796.5531

304.34201

0.25

1-3

10.2918

-3.7014

422.5525

1691.0677

0.3

2-3

19.7506

-0.8405

-1769.413

-289.3552

0.1

1-2

-25.3982

-42.2166

10000

10000

0.3

1-3

96.1732

23.0019

-10000

-10000

0.15

2-3

42.7989

-30.4691

-10000

10000

0.17

0.0067

0.0097

0.0016

0.0026

The calculated LLE data using the NRTL model are plotted in Fig. 10. It can be seen that the slope of tie-lines is less than 0, which depicts the selectivity of LTTM2 to benzene is larger than that to cyclohexane. At the same time, the mixing phase is widely separated at the low concentration of benzene, which proves that LTTMs are suitable for the separation of benzene at low concentration. Fig. 10 also shows the ternary diagrams that include the calculated compositions by NRTL correlation. It can

ACS Paragon Plus Environment

Energy & Fuels

be seen that the calculated compositions have good agreement with the experimental data, and the tie-lines of both compositions coincide well in most cases. 0.00 1.0

1.0

ne

ne

cyc

loh

0.4

e

e

cyc

zen

zen 0.4

0.6

0.50

ben

0.6 0.50

Exp NRTL

0.8

0.25

exa

0.25

ex a

Exp NRTL

0.8

loh

0.00

ben

0.75

0.75

0.2

0.2

1.00 0.0

0.2

0.4

0.6

0.8

1.0

0.0 1.00 0.0

0.0 0.2

0.4

0.6

(a) 0.00

0.8

(b) 0.00 1.0

1.0 Exp NRTL

e xa loh

0.6

0.50

cyc

lo h ex a ne c yc

NRTL

0.4

0.75

e zen ben

0.6

0.50

Exp

0.8

0.25

ne

0.8

0.25

0.4

0.75

0.2 1.00 0.0

1.0

LTTM2

LTTM1

e ze n ben

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

Page 18 of 27

0.2

0.4

0.6

0.8

1.0

0.2

0.0 1.00 0.0

0.0 0.2

0.4

0.6

0.8

1.0

LTTM4

LTTM3

(c)

(d)

Fig. 10 Ternary phase diagrams for benzene + hexane + LTTMs at 40oC.

5.2 Extraction Process Analysis In this section, a continuous extraction process for separating benzene from cyclohexane was preliminarily simulated in Aspen Plus V8.4, which using LTTM2 as solvent based on the binary interaction parameters obtained from section 5.1. It is noticed that LTTMs have not been included in the component databases of process simulators. Therefore, it is necessary to create a user-defined LTTMs database in order to allow LTTMs to be selected as components within a process simulation.42,43 Some thermodynamic properties, such as critical properties, boiling point, were

ACS Paragon Plus Environment

Page 19 of 27 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

Energy & Fuels

predicted and required to be implanted in process simulation. For LTTM2 which was selected as the solvent, the critical properties (Tc, Pc and Vc), the normal boiling temperature (Tb), and the acentric factor (ω) were calculated and shown in Table S5 of the supporting information. The flowsheet of extraction process using LTTM2 as solvent is shown in Fig. 11. The mixtures of benzene and cyclohexane are fed at the bottom of extract column, and the solvent LTTM2 is fed at the top of extract column. The flash tank is used for the recovery of LTTM2 since LTTM2 has no vapor pressure. In order to achieve a high cyclohexane product purity of at least 99.50 wt% and meanwhile to ensure a high benzene recovery ratio of about 98.0%.17 The Constraint models of Model Analysis Tools in Aspen Plus are used. The amount of cyclohexane at the top of column is controlled by adjusting the ratio of solvent-to-feed (S/F) in order to ensure the recovery ratio of cyclohexane. Since a small amount of benzene is lost during the flash operation, and it must ensure the final recovery ratio of benzene is greater than 98 wt%. The Constraint model is used to limit the recovery ratio of benzene at the bottom of extraction column. Cyclohexane product

LTTM2

F=6702.5 kg/h XBenzene = 0.003 XCyclohexane = 0.997

F=25500 kg/h

Benzene F=1627.5 kg/h XBenzene = 0.704 XCyclohexane = 0.296

Extractor Benzene + Cyclohexane 40℃ ℃; 1 bar F=8330 kg/h XBenzene = 0.14 XCyclohexane = 0.86

Nstage = 20 P = 1 bar T = 40 oC

Flash P = 1.5 Pa T = 40 oC

Benzene + LTTM2

Recycled LTTM2

Fig. 11 Extraction process for the separation of benzene and cyclohexane using LTTM2.

The relationship between S/F and the number of stages (Nstage) is shown in Fig. 12. It can be seen that S/F reduces with the increase of Nstage at a certain degree of purity and recovery ratio. The reason may be that the increase of Nstage leads to more ACS Paragon Plus Environment

Energy & Fuels

amount of benzene extracted from cyclohexane. Fig. 13-a shows the change of cyclohexane purity with different S/F and Nstage values. It is found that the purity of cyclohexane increases with the increase of S/F and Nstage. Nstage has no obvious effect on the purity of cyclohexane when S/F increases to a certain value. Fig. 13-b shows the relationship between cyclohexane recovery ratio and Nstage. The cyclohexane recovery ratio reduces as S/F increases. The reason may be that the increase of the amount of LTTM2 leads to more benzene extracted, and more cyclohexane is also extracted at the same time. It also can be seen Nstage has insignificant effect on the cyclohexane recovery ratio. Fig. 14 shows the effect of S/F and Nstage on the benzene recovery ratio. It is found that as S/F and Nstage increase, the recovery ratio increases. Similar to Fig. 13-a, when S/F increases to a certain value, Nstage has little effect on the benzene recovery ratio. The main reason is that the selectivity of LTTM2 to benzene is high, and a large amount of solvent will lead more benzene extracted from cyclohexane. Considering the two aspects, one is the changing trend of S/F with Nstage, the other is the energy consumption which is caused by the amount of solvent added, the Nstage is 20 and the solvent flow is 22500kg/h (S/F=3.06), which can meet the purity and the recovery ratio of products. All the operating parameters are all shown in Fig. 11. 7

6

S/F

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

Page 20 of 27

5

4

3 4

8

12

16

20

24

28

Nstage

Fig. 12 Mass ratio of solvent-to-feed (S/F) plotted with the number of stages for the separation demands of cyclohexane mass purity >99.50% and benzene recovery ratio>98%.

ACS Paragon Plus Environment

Page 21 of 27

Mass purity of cyclohexane product

1.000

0.998

0.996 Nstage=16

0.994

Nstage=18 Nstage=20 Nstage=22 Nstage=24

0.992

0.990

2.95

3.00

3.05

3.10

3.15

3.20

3.25

Mass ratio of S/F

a Nstage=16

Rccovery ratio of cyclohexane

0.935

Nstage=18 Nstage=20

0.934

Nstage=22 Nstage=24

0.933 0.932 0.931 0.930

2.95

3.00

3.05

3.10

3.15

3.20

3.25

Mass ratio of S/F

b Fig. 13 Mass purity (a) and recovery ratio (b) of cyclohexane product plotted with the mass ratio of solvent-to-feed (S/F) at different number of stages (Nstage) for the solvent LTTM2. 1.00

Rccovery ratio of benzene

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

Energy & Fuels

0.99

0.98 Nstage=16

0.97

Nstage=18 Nstage=20 Nstage=22

0.96

Nstage=24

0.95

2.95

3.00

3.05

3.10

3.15

3.20

3.25

Mass ratio of S/F

Fig. 14 Recovery ratio of benzene plotted with the mass ratio of solvent-to-feed (S/F) at different number of stages (Nstage) for the solvent LTTM2.

ACS Paragon Plus Environment

Energy & Fuels 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

Page 22 of 27

6 CONCLUSIONS In this paper, LTTM1 ethylene glycol-trimethylamine hydrochloride 5:1 and LTTM2 sulfolane - tetrabutylammonium bromide 5:1 were used as solvents for the separation of benzene and cyclohexane. LLE data of the ternary system benzene-cyclohexane-LTTMs were experimentally determined and the Othmer-Tobias equation was used to assess the reliability and consistency of the obtained experimental tie-lines data. Basing on the LLE data of benzene-cyclohexane-LTTMs, the extraction performances of the two types of LTTMs were compared according to three indicators including selectivity, distribution coefficient and PI. It is found that LTTM2 sulfolane - tetrabutylammonium bromide 5:1 has a higher PI compared to LTTM1. And then the extraction performance was further studied by changing the mole ratio of sulfolane to tetrabutylammonium bromide. It is found that when the mole ratio of HBD to HBA is 5:1, LTTM2 has the best extraction performance. In addition, the NRTL model was applied to correlate the experimental data, and the results show that it has good agreement with the experimental data. Therefore, a continuous extraction process for separating benzene from cyclohexane was preliminarily simulated in Aspen Plus V8.4. The optimal Nstage and S/F are Nstage=20 and S/F=3.06, respectively. For these values, a high cyclohexane product purity of 99.70 wt%, a high cyclohexane recovery ratio of 93.28% and a high benzene recovery ratio of 98.25% can be achieved. It proves that LTTM2 is a very attractive solvent for the separation of benzene-cyclohexane mixtures. ASSOCIATED CONTENT Supporting Information The experimental apparatus (Fig S1), LLE data of benzene + cyclohexane + LTTMs ternary systems (Tables S1-S4) and physical properties of LTTM2 calculated by literatures

(Table

S5)

are

available

free

https://pubs.acs.org/journal/enfuem.

ACS Paragon Plus Environment

of

charge

on

the

Page 23 of 27 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

Energy & Fuels

AUTHOR INFORMATION Corresponding Author Lanyi Sun. Tel.: +86 13854208340. Fax: +86 0532 86981787. E-mail address: [email protected].

Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant: 21676299 and Grant: 21476261) and supported by the Fundamental Research Funds for the Central Universities (Grant: 17CX06025). Finally the authors are grateful to the editor and the anonymous reviewers. NOMENCLATURE Benzene distribution coefficient =βA Deep eutectic solvents=DESs Hydrogen bond acceptor =HBA Hydrogen bond donor =HBD Ionic liquids=ILs Low transition temperature mixtures =LTTMs Liquid-liquid equilibrium=LLE Number of stage=Nstage Performance Index=PI Root-mean-square deviation=RMSD Selectivity=S Solvent-to-feed=S/F Screening charge density=σ

REFERENCE (1) Rodriguez N. R.; Requejo, P. F.; Kroon M. C. Aliphatic–aromatic separation using

ACS Paragon Plus Environment

Energy & Fuels 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

Page 24 of 27

deep eutectic solvents as extracting agents. Ind. Eng. Chem. Res. 2015, 54(45), 11404-11412. (2) Gramajo de Doz, M.B.; Bonatti, C. M.; Sólimo, H.N. Liquid-Liquid equilibria for the quaternary system water+ methyl tert-butyl ether+ benzene+ cyclohexane and its constituent partially miscible ternary systems at 303.15 K. Energ Fuels. 2005, 19(5), 1977-1983. (3) Meindersma G.W.; Podt A. (J.G.); Klaren M.B.; de Haan A.B. Separation of aromatic and aliphatic hydrocarbons with ionic liquids. Chem. Eng. Commun. 2006, 193(11), 1384-1396. (4) Salleh Z.; Wazeer I.; Mulyono S., et al. Efficient removal of benzene from cyclohexane-benzene mixtures using deep eutectic solvents-COSMO-RS screening and experimental validation. J. Chem. Thermodyn. 2017, 104, 33-44. (5) Lei Z.; Chen B.; Zhu J. Extractive distillation with ionic liquids: a review. AIChE J. 2014, 60, 3312-3329. (6) Lei Z.; Chen B.; Ding Z. Special Distillation Processes. Elsevier, 2005. (7) Dong H.; Yang X.; Zhang J. Liquid-liquid equilibria for benzene+ cyclohexane+ N, N-dimethylformamide+ potassium thiocyanate. J. Chem. Eng. Data. 2010, 55(9), 3972-3975. (8) Aspi K. K.; Surana N. M.; Ethirajulu K.; Vennila V. Liquid-liquid equilibria for the cyclohexane + benzene + dimethylformamide + ethylene glycol system. J. Chem. Eng. Data. 1998, 43(6), 925-927. (9) Dong H.; Yang X.; Yue G., et al. Liquid-liquid equilibria for benzene+ cyclohexane+ N, N-dimethylformamide+ ammonium thiocyanate. J. Chem. Eng. Data. 2011, 56(5), 2664-2668. (10) Mulyono S.; Hizaddin H. F.; Alnashef I. M., et al. Separation of BTEX aromatics from n-octane using a (tetrabutylammonium bromide+ sulfolane) deep eutectic solvent–experiments

and

COSMO-RS

prediction.

Rsc

Adv.

2014,

4(34),

17597-17606. (11) Zhou T.; Wang Z.; Ye Y., et al. Deep separation of benzene from cyclohexane by liquid extraction using ionic liquids as the solvent. Ind. Eng. Chem. Res. 2012, 51(15), ACS Paragon Plus Environment

Page 25 of 27 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

Energy & Fuels

5559-5564. (12) García J.; García S.; Torrecilla J. S., et al. Separation of toluene and heptane by liquid-liquid extraction using z-methyl-N-butylpyridinium tetrafluoroborate isomers (z= 2, 3, or 4) at T= 313.2 K. J. Chem. Thermodyn. 2010, 42(8), 1004-1008. (13) García J.; García S.; Torrecilla J. S., et al. Liquid- liquid equilibria for the ternary systems

{heptane+

toluene+

N-butylpyridinium

tetrafluoroborate

or

N-hexylpyridinium tetrafluoroborate} at T= 313.2 K. J. Chem. Eng. Data. 2010, 55(8), 2862-2865. (14) Song Z.; Zhang J.; Zeng Q., et al. Effect of cation alkyl chain length on liquid-liquid equilibria of {ionic liquids+ thiophene+ heptane}: COSMO-RS prediction and experimental verification. Fluid Phase Equilibr. 2016, 425: 244-251. (15) Song Z.; Zeng Q.; Zhang J., et al. Solubility of imidazolium-based ionic liquids in model fuel hydrocarbons: A COSMO-RS and experimental study. J. Mol. Liq. 2016, 224: 544-550. (16) Abu-Eishah S. I.; Dowaidar A. M. Liquid-liquid equilibrium of ternary systems of cyclohexane+ (benzene+ toluene+ ethylbenzene or+ o-xylene) + 4-methyl-N-butyl pyridinium tetrafluoroborate ionic liquid at 303.15 K. J. Chem. Eng. Data. 2008, 53(8), 1708-1712. (17) Lyu Z.; Zhou T.; Chen L., et al. Simulation based ionic liquid screening for benzene–cyclohexane extractive separation. Chem. Eng. Sci. 2014, 113, 45-53. (18) Song Z.; Zhou T.; Zhang J., et al. Screening of ionic liquids for solvent-sensitive extraction-with deep desulfurization as an example. Chem. Eng. Sci. 2015, 129: 69-77. (19) Couling D. J.; Bernot R. J.; Docherty K. M., et al. Assessing the factors responsible for ionic liquid toxicity to aquatic organisms via quantitative structure-property relationship modeling. Green Chem. 2006, 8(1), 82-90. (20) Zhao D.; Liao Y.; Zhang Z. Toxicity of ionic liquids. Clean–soil, air, water, 2007, 35(1), 42-48. (21) Dong K.; Liu X.; Dong H., et al. Multiscale studies on ionic liquids. Chem. Rev. 2017, 117(10), 6636-6695. ACS Paragon Plus Environment

Energy & Fuels 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

(22) Lei Z.; Dai C.; Chen B. Gas solubility in ionic liquids. Chem. Rev. 2014, 114(2), 1289-1326. (23) Abbott A. P.; Capper G.; Davies D. L., et al. Novel solvent properties of choline chloride/urea mixtures. Chem. Commun. 2003, 9(1), 70-71. (24) Francisco M.; van den Bruinhorst A.; Kroon M. C. Low-transition-temperature mixtures (LTTMs): A new generation of designer solvents. Angew. Chem. Int. Edit. 2013, 52(11), 3074-3085. (25) Zhang Q.; Vigier K. D. O.; Royer S., et al. Deep eutectic solvents: syntheses, properties and applications. Chem. Soc. Rev. 2012, 41(21), 7108-7146. (26) Rodriguez N. R.; Molina B. S.; Kroon M. C. Aliphatic+ ethanol separation via liquid–liquid extraction using low transition temperature mixtures as extracting agents. Fluid Phase Equilibr. 2015, 394, 71-82. (27) Hadj-Kali M. K.; Separation of ethylbenzene and n-octane using deep eutectic solvents. Green Process Synth. 2015, 4(2), 117-123. (28) Meindersma G.W.; Hansmeier A.R.; deHaan A.B.D. Ionic liquids for aromatics extraction. Present status and future outlook. Ind. Eng. Chem. Res. 2010, 49(16), 7530-7540. (29) N.R. Rodríguez, A.S. González, P.M. Tijssen, M.C. Kroon, Low transition temperature mixtures (LTTMs) as novel entrainers in extractive distillation. Fluid Phase Equilibr. 2015, 385, 72-78. (30) Dai, F.; Xin, K.; Song, Y., et al. Liquid-liquid equilibria for the ternary system containing 1-Butanol+ methoxy (methoxymethoxy) methane+ water at temperatures of 303.15, 323.15 and 343.15 K. Fluid Phase Equilibr, 2016, 409, 466-471. (31) Ma S.; Yu Q.; Hou Y., et al. Screening monoethanolamine as solvent to extract phenols from alkane. Energ. Fuels. 2017, 31(11), 12997-13009. (32) BIPM, IEC, IFCC, ILAC, ISO, IUPAC, IUPAP and OMIML. Evaluation of measurement data-guide for the expression of uncertainty in measurement. JCGM, 2008, 100. (33) Othmer D.; Tobias P. Liquid-liquid extraction data-the line correlation. Ind. Eng. Chem. 1942, 34, 693-696. ACS Paragon Plus Environment

Page 26 of 27

Page 27 of 27 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

Energy & Fuels

(34) Renon H.; Prausnitz J. M. Local compositions in thermodynamic excess functions for liquid mixtures. AIChE J. 1968, 14(1), 135-144. (35) Anantharaj R.; Banerjee T. COSMO-RS-based screening of ionic liquids as green solvents in denitrification studies. Ind. Eng. Chem. Res. 2010, 49(18), 8705-8725. (36) Anantharaj R.; Banerjee T. COSMO-RS based predictions for the desulphurization of diesel oil using ionic liquids: effect of cation and anion combination. Fuel Process. Technol. 2011, 92 (1), 39-52. (37) Klamt A. COSMO-RS: from quantum chemistry to fluid phase thermodynamics and drug design. Elsevier, 2005. (38) Klamt A.; Eckert F. COSMO-RS: a novel and efficient method for the a priori prediction of thermophysical data of liquids. Fluid Phase Equilibr. 2000, 172(1), 43-72. (39) Mullins E.; Oldland R.; Liu Y.A., et al. Sigma-profile database for using COSMO-based thermodynamic methods. Ind. Eng. Chem. Res. 2006, 45(12), 4389-4415. (40) Chen D. C.; Ye H. Q.; Wu H. Liquid-liquid equilibria of methylcyclohexane– benzene–N-formylmorpholine at several temperatures. Fluid Phase Equilibr. 2007, 255(2), 115-120. (41) Vatani M.; Asghari M.; Vakili-Nezhaad G. Application of genetic algorithm to the calculation of parameters for NRTL and two-suffix margules models in ternary extraction ionic liquid systems. J. Ind. Eng. Chem. 2012, 18(5), 1715-1720. (42) Ma S.; Hou Y.; Sun Y., et al. Simulation and experiment for ethanol dehydration using low transition temperature mixtures (LTTMs) as entrainers. Chem. Eng. Process. 2017, 121, 71-80. (43) Zhu Z.; Ri Y.; Jia H., et al. Process evaluation on the separation of ethyl acetate and ethanol using extractive distillation with ionic liquid. Sep. Purif. Technol. 2017, 181, 44-52.

ACS Paragon Plus Environment