Elsevier

Fluid Phase Equilibria

Volume 462, 25 April 2018, Pages 73-84
Fluid Phase Equilibria

Liquid-liquid equilibria and COSMO-SAC modeling of organic solvent/ionic liquid - hydroxyacetone - water mixtures

https://doi.org/10.1016/j.fluid.2018.01.026Get rights and content

Abstract

In this work conventional organic solvents (ethyl acetate, n-propyl acetate, n-butyl acetate, chloroform) as well as ionic liquids, 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([EMIM][Tf2N]) and 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([BMIM][Tf2N]), were investigated for the extraction of hydroxyacetone from aqueous solution. The liquid–liquid equilibrium (LLE) data for the ternary systems were experimentally determined at T = 298.15 K and atmospheric pressure. In terms of the distribution coefficient and selectivity, [EMIM][Tf2N] and [BMIM][Tf2N] were found to be the most effective solvents for the extraction of hydroxyacetone from aqueous solution, and the composition of ionic liquids in the raffinate phase was found to be negligible. The experimental results were correlated with the NRTL and UNIQUAC models. The root mean square deviations (RMSD) obtained between the NRTL and UNIQUAC correlations for all the systems were in the range of 0.19–5.18%. COSMO-SAC was then used to make a first principles prediction of the phase equilibria of the systems studied. For the organic solvent-based systems, the deviations between predicted and experimental values were in the range of 3.49–8.00% (% RMSD) whereas for the ionic liquid systems, the deviations were in the range of 16.81–18.53%.

Introduction

The changing global climate and the depletion of fossil resources are major issues for the current generation. Consequently, the development of alternative renewable energy resources has been intensely promoted. Compared to other renewable energy sources, biomass has the potential to replace a large fraction of fossil fuels as feedstocks and thus capable of meeting the energy, chemicals and materials requirement of mankind. Biorefineries are manufacturing facilities that produce biofuels and biochemicals from various biomass feedstocks. Conceptually, this is analogous to current petroleum refinery that produces fuels and chemicals from crude oil [[1], [2], [3], [4], [5], [6], [7], [8]].

In order to convert biomass into valuable products within a biorefinery, several processes are needed. These can be divided into four groups: physical, biochemical, thermo-chemical and chemical processes. Among them, fast pyrolysis is a thermo-chemical process for the production of liquid fuel and chemicals from biomass. Bio-oil is the main product of fast pyrolysis and depending on the biomass source, the oxygen content will be in the range of 35–40 wt% due to several oxygen-containing components (acetic acid: 0.5–12 wt%, glycolaldehyde: 0.9–13 wt%, hydroxyacetone: 0.7–7.4 wt %, and furfural alcohol:0.1–5.2 wt %). A higher oxygen content is responsible for a lower heating value of bio-oil. However, the high concentration of oxygen-containing compounds makes bio-oil a good raw material for the production of various chemicals [3,[9], [10], [11], [12], [13], [14], [15], [16]].

The first step in the recovery of chemicals from bio-oil usually involves fractionation using water that results in two fractions-an aqueous top phase enriched in carbohydrate-derived chemicals (acetic acid, glycolaldehyde, hydroxyacetone) and an organic bottom phase containing lignin-containing fractions. Both phases can be processed separately to extract the valuable chemicals. Several studies have been reported for the extraction of bio-oil chemicals including acetic acid [17,18], levoglucosan [19], sugar/sugar derivatives [20] and phenolic compounds [[21], [22], [23], [24]].

Hydroxyacetone (acetol, up to 7 wt% in bio-oil) is an important compound for the chemical industry. It is used for the production of a variety of chemical products such as propylene glycol and acrolein. It is also used in the food industry to give an aroma to foods, in the cosmetic industry as a skin tanning agent, and in the textile industry as a reduced dye [25]. Recent work of Li et al. [26] and Albuquerque et al. [27,28] have developed pathways for the usage of acetol as raw material for the production of diesel fuel and lactic acid respectively. Li et al. [26] have recently developed a new route for the synthesis of renewable diesel or jet fuel range branched alkanes by hydroxyalkylation-alkylation (HAA) of acetol and 2-methylfuranfollowed by hydrodeoxygenation (HDO). Under optimum conditions, 79.1% yield of HAA products was obtained and after the HDO of the HAA products, 79.5% carbon yield for diesel or jet-fuel range alkanes was obtained.

Albuquerque et al. [27] have reported the catalytic synthesis route for the production of lactic acid from acetol aqueous solutions at atmospheric pressure. Complete selectivity (100%) for lactic acid was achieved and the undesired cascade oxidation of lactic to pyruvic acid was hindered. Thus the conversion of acetol to lactic acid is bound to increase the usage of acetol in the near future. Lactic acid is also an industrially important product with wide applications in the food, pharmaceutical, cosmetic, leather, textile and chemical industries. Currently, there is an increased demand for lactic acid as a feedstock for the production of biopolymer poly-lactic acid (PLA), which is a promising biodegradable, biocompatible, and environmentally friendly alternative to plastics derived from petrochemicals [[29], [30], [31], [32], [33]].

The objective of this work is to report data related to the extraction of acetol from aqueous solution using both conventional and ionic liquid solvents. In the initial part of this work, LLE data for the ternary systems of chloroform, ethyl acetate, n-propyl acetate, n-butyl acetate, the ionic liquids[EMIM] [Tf2N], and [BMIM] [Tf2N] each separately with acetol + water mixtures were determined at 298.15 K at atmospheric pressure. The Othmer-Tobias [34] and Hand correlations [35] were used to verify the reliability of the experimental tie-line data. Additionally, the experimental data were correlated with the NRTL [36] and UNIQUAC [37] model. In addition, quantum chemical based conductor like screening model - segment activity coefficient (COSMO-SAC) was used for a first-principles prediction of the tie lines of the above mentioned systems [38,39].

Section snippets

Chemicals and materials

All chemicals used in this work were analytical grade and used without further purification (Table 1). Millipore water was used in all the measurements. The purities of all the materials were checked by density measurement and1H NMR spectroscopy. Analysis of the peaks of 1H NMR spectroscopy indicated negligible impurities. To reduce the water content and volatile compounds in the ILs to negligible values, a vacuum (0.1Pa) for at least 48 h was applied prior to the measurements.

Density measurement

The density of

Activity coefficient models - NRTL and UNIQUAC

The NRTL and UNIQUAC activity models were used to correlate experimental tie-line data. The structural parameters r and q for the compounds studied have been taken from the Aspen Plus process simulation software V8.8 data bank (Aspen Technology Inc., USA) and literature [40,41](Table 4).

The thermodynamic equilibrium condition for a multi-component liquid-liquid system is described by the following expression:γiExiE=γiRxiR(i=1,2,3,)where γi, the activity coefficient of component i in a phase (E

Experimental LLE data

Experimental LLE data for the systems: [EMIM][Tf2N] (1) + acetol(2) + water (3), [BMIM][Tf2N] (1) + acetol (2) + water (3), chloroform (1) + acetol (2) + water (3), ethyl acetate (1) + acetol (2) + water (3),n-propyl acetate (1) + acetol (2) + water (3) and n-butyl acetate (1) + acetol (2) + water (3) are reported in Table 5, Table 6, Table 7, Table 8, Table 9, Table 10. The distribution coefficient (β) and the selectivity (S) were calculated to determine the extraction ability of solvents for

Conclusions

Liquid-liquid equilibria (LLE) for the ternary systems: ethyl acetate (1) + acetol(2) + water (3), n-propyl acetate (1) + acetol (2) + water (3), n-butyl acetate (1) + acetol (2) + water (3), chloroform (1) + acetol (2) + water (3), [EMIM][Tf2N] (1) + acetol (2) + water (3) and [BMIM][Tf2N] (1) + acetol (2) + water (3) were measured at 298.15 K and 0.1 MPa. For the chloroform, [EMIM][Tf2N] and [BMIM][Tf2N] ternary mixtures, the distribution coefficient and selectivity values are greater than

Acknowledgments

The authors gratefully acknowledge CIF (Central Institute Facility) IIT Guwahati for recording the 1H NMR spectra using the 600 MHz NMR. Authors also acknowledge Karl Fisher Titrator (MetroOhm 787 KF Titrino) analysis in the Analytical Laboratory of the Department of Chemical Engineering, IIT Guwahati.

References (48)

  • F. Cherubini

    Energy Convers. Manag.

    (2010)
  • W.M. Budzianowski et al.

    Appl. Energy

    (2016)
  • R. Parajuli et al.

    Renew. Sustain. Energy Rev.

    (2015)
  • S.K. Maity

    Renew. Sustain. Energy Rev.

    (2015)
  • S.K. Maity

    Renew. Sustain. Energy Rev.

    (2015)
  • A. Demirbas

    Appl. Energy

    (2011)
  • S.N. Naik et al.

    Renew. Sustain. Energy Rev.

    (2010)
  • P.S. Nigam et al.

    Prog. Energy Combust. Sci.

    (2011)
  • N.M. Bennett et al.

    Bioresour. Technol.

    (2009)
  • E.M. Albuquerque et al.

    JMolCatal AChem

    (2015)
  • M.A. Abdel-Rahman et al.

    J. Biotechnol.

    (2011)
  • A. Bharti et al.

    Fluid Phase Equil.

    (2015)
  • R.S. Santiago et al.

    Fluid Phase Equil.

    (2010)
  • A. Marcilla et al.

    Fluid Phase Equil.

    (2017)
  • M.F. Demirbas

    Appl. Energy

    (2009)
  • IEA Bioenergy Task 42

    Biorefining, Sustainable and Synergetic Processing of Biomass into Marketable Food & Feed Ingredients, Products (Chemicals, Materials) and Energy (Fuels, Power, Heat), Wageningen, the Netherlands

    (August 2014)
  • B. Kamm et al.

    Biorefineries-industrial Processes and Products: Status Quo and Future Directions

    (2006)
  • A.V. Bridgewater

    Chem. Eng. J.

    (2003)
  • S. Czernik et al.

    Energy Fuels

    (2004)
  • D. Mohan et al.

    Energy Fuels

    (2006)
  • A.V. Bridgewater

    Biomass Bioenergy

    (2012)
  • J.P. Diebold

    A Review of the Chemical and Physical Mechanisms of the Storage Stability of Fast Pyrolysis Bio-oils

    (January 2000)
  • F.H. Mahfud et al.

    Separ. Sci. Technol.

    (2008)
  • C.B. Rasendra et al.

    Chem. Eng. J.

    (2011)
  • Cited by (17)

    • Molecular mechanism, liquid–liquid equilibrium and process design of separating octane-n-butanol system by ionic liquids

      2022, Journal of Molecular Liquids
      Citation Excerpt :

      They found that the COSMO-SAC model can more effectively predict the solubility data. Bharti et al. [23] determined the LLE data for chloroform, n-propyl acetate, ethyl acetate, n-butyl acetate, 1-butyl-3-methylimidazole bis-trifluoromethyl sulfonamide ([BMIM][Tf2N]), and 1-ethyl-3-methylimidazole bis-trifluoromethyl sulfonamide ([EMIM][Tf2N]) with acetyl-water mixtures; they carried out the first-principles-calculation-based prediction of the phase equilibrium for the system by using the COSMO-SAC model. The predicted values were in good agreement with the LLE data for polar organic solvent systems.

    • Experimental measurement, correlation and COSMO-SAC prediction of liquid-liquid equilibrium for MIPK + dimethylphenols + water mixtures

      2020, Journal of Chemical Thermodynamics
      Citation Excerpt :

      Therefore, the COSMO-SAC model provided a way to predict LLE data without experimental data with the help of quantum chemical calculations. This model had been used to predict many phase equilibrium experiments [27–29], and its prediction was pretty well. Therefore, using this model to predict LLE data for extracting dimethylphenols from wastewater was a good choice.

    • Ternary liquid-liquid equilibrium of methanol + isopropyl acetate/methyl methacrylate + 1-methylmidazole hydrogen sulfate at different temperatures and 1 atm

      2019, Journal of Molecular Liquids
      Citation Excerpt :

      In recent years, the quantum chemical based conductor like screening model-segment activity coefficient (COSMO-SAC) is widely used to predict and analyze the phase behavior of LLE containing ILs [23–25]. Bharti et al. [26] made first principles predictions of the tie lines of LLE on the separation of acetol/water using 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide and the result showed the predicted values of COSMO-SAC were in good agreement with the LLE experimental data. Zhou et al. [27] calculated the performance index of thioglycolic acid/water/ILs using the COSMO-SAC and analyzed the relationship between ionic liquids and azeotropes based on the σ-profiles.

    • Liquid-liquid equilibria for azeotropic mixture of methyl tert-butyl ether and methanol with ionic liquids at different temperatures

      2019, Journal of Chemical Thermodynamics
      Citation Excerpt :

      Ionic liquids (ILs), as new extraction agents, are called “designer solvents”, and their structures can be altered to adjusted physical and chemical characteristics for specific applications [16,17]. ILs are widely used in many fields due to their designability and distinct advantages for physical and chemical properties [18–22]. Specifically, ILs show outstanding advantages in liquid-liquid extraction as new extracting agents.

    • The selection and mechanism of composite solvent containing ionic liquid in liquid-liquid microextraction process

      2019, Chemical Engineering and Processing - Process Intensification
      Citation Excerpt :

      3) it needs to have extraction ability for phoxim. Based on the above conditions, 4 traditional extractants–chloroform (CHCl3) [7], chlorobenzene (C6H5Cl), butyl acetate (C4H8O2) and hexyl acetate (C8H16O2) [24]—are selected initially by COSMO-SAC. The results are shown in Table 3.

    View all citing articles on Scopus
    View full text