Vapour–liquid equilibria of the system 1,1,1,2-tetrafluoroethane + monoethylene-glycol dimethylether from 283.15 to 353.15 K: New modified UNIFAC parameters

https://doi.org/10.1016/j.fluid.2008.07.003Get rights and content

Abstract

The vapour–liquid equilibria (p, T, x) of a 1,1,1,2-tetrafluoroethane + ethylene-glycol dimethylether mixture was measured using the static method in the temperature range (283.15–353.15) K in steps of 10 K. The vapour–liquid equilibrium data were correlated by the temperature-dependent five-parameter non-random two-liquid (NRTL) equation. The data obtained in this experiment together with our previous experimental data for the binary system 1,1,1,2-tetrafluoroethane + triethylene-glycol dimethylether were used to calculate the UNIFAC interaction parameters for the CF3 and CH2F groups with the CH3, CH2, CH3O and CH2O groups. The resulting interaction parameters were used to predict the vapour pressures of the various mixtures studied in the literature.

Introduction

In recent years, increasing concern for the environment has sparked interest in the use of absorption heat pumps and absorption chillers for air-conditioning and refrigeration applications. Lithium bromide + water and water + ammonia are the fluids traditionally used in absorption systems, but they have several drawbacks. For example, lithium bromide + water mixture causes corrosion and crystallisation, and water + ammonia requires rectification. Other compounds, both inorganic and organic, have therefore been proposed by various authors. Borde et al. [1] proposed that 1,1,1,2-tetrafluoroethane (also known as R134a or HFC-134a) be used as an alternative refrigerant together with tetraethylene-glycol dimethylether (TEGDME) as an organic absorbent. Polyalkylene-glycols and polyalkylene-glycol dimethylethers have been used as lubricants for R134a in compression systems [2], [3]. Tseregounis and Riley [3] studied the solubility of R134a in glycol-type compounds and showed that R134a is highly soluble in TEGDME. López et al. [4] measured and predicted the solubility of R134a in triethylene-glycol dimethylether (TrEGDME) and in TEGDME at 101.33 kPa in the temperature range (258.15–298.15) K. Machi et al. [5] measured the bubble pressures of the R134a and TrEGDME mixtures in the temperature range (283.15–323.15) K. In a previous study [6], we obtained solubility data on R134a in TrEGDME in the temperature range (283.15–353.15) K. All of these studies have shown that R134a is completely miscible in glycol-type compounds in a wide temperature range, and that mixtures of this kind are noncorrosive and thermally stable.

A great variety of thermodynamic models for fitting and predicting vapour–liquid equilibrium have been developed. Of these models, those based on calculating activity coefficients show good versatility and accuracy. Two of these models have been used in this study. First, the non-random two-liquid (NRTL) model [7] was used to fit the experimental data. Second, the modified UNIFAC group-contribution model [8] was used to predict the vapour–liquid equilibria of R134a and various polyethylene-glycol dimethylether mixtures. Kleiber [9] and Kleiber and Axmann [10] proposed new interaction parameters for certain groups (such as CF3 and CH2F) in the modified UNIFAC model because the prediction of vapour pressure in mixtures containing these groups had been unsatisfactory due to the fact that the fluorine atom causes electronegativity effects on the rest of the groups. With these parameters, the model obtains better vapour–liquid equilibrium results for mixtures with highly electronegative atoms. However, when the modified UNIFAC model with Kleiber and Axmann's parameters [10] is used to predict the vapour–liquid equilibrium of 1,1,1,2-tetrafluoroethane and polyalkylene-glycol dimethylethers, the results are unsatisfactory, as shown by the high deviations obtained by López et al. [4] and various predictions carried out by our laboratory (presented below). This result is probably due to the limited database used by Kleiber and Axmann [10] to calculate the new parameters. In this work, we propose new interaction parameters for the UNIFAC groups involved in 1,1,1,2-tetrafluoroethane and polyethylene-glycol dimethylether mixtures (i.e. CF3, CH2F, CH3, CH2, CH3O and CH2O).

The vapour–liquid equilibria of 1,1,1,2-tetrafluoroethane + ethylene-glycol dimethylether (MEGDME) were measured using a static method in the temperature range 283.15–353.15 K. These experimental data were correlated using the temperature-dependent five-parameter NRTL activity coefficient model. The temperature-dependent group-interaction parameters for the modified UNIFAC model were obtained by regression of our vapour–liquid equilibrium data [6] on R134a + MEGDME and R134a + TrEGDME using the commercial software Aspen Plus [11]. Using these parameters, vapour–liquid equilibria were predicted for the values presented by Borde [1], Lopez et al. [4] and Machi et al. [5] with a lower deviation than that obtained with the previous parameters.

Section snippets

Materials

1,1,1,2-Tetrafluoroethane (CF3CH2F) (DuPont, >99.9%) and ethylene-glycol dimethylether (CH3O(CH2CH2O)CH3) (Aldrich, >99%) were used without further purification. They were carefully degassed by several freezing and thawing cycles. The MEGDME was stored in type-4A molecular sieves before use.

Apparatus and procedure

The vapour–liquid equilibria of the R134a + MEGDME mixture were measured using the static method. The apparatus used has been described in previous studies [12]. It consists of an equilibrium cell, a

Data reduction

The equilibrium composition was calculated based on the initial composition and the experimental pressure and temperature for R134a + MEGDME using Barker's method, as described by Herraiz et al. [14]. This method uses a Redlich–Kister polynomial equation for the excess Gibbs energy of the liquid phase; we used a fourth-degree polynomial. The coefficients were obtained by using Marquardt's non-linear regression program to minimise the objective functionOF=1N1Npicalpiexppiexp2where N is the

Results and discussion

Table 1 shows the experimental vapour pressure data, the liquid and vapour composition calculated using Barker's method, and the activity coefficients and excess Gibbs energy function calculated for the temperature range 283.15–353.15 K. Table 2 shows the NRTL equation parameters. The root-mean-square relative deviation (Eq. (7)) between the experimental and calculated pressure values was 0.7%.RMSD=100×1N1Npicalpiexppiexp2where N is the number of experimental data points and pexp and pcal are

Conclusions

The vapour–liquid equilibria of the 1,1,1,2-tetrafluoroethane (HFC-134a) refrigerant in ethylene-glycol dimethylether were determined for the temperature range (283.15–353.15) K using the static method. The temperature-dependent five-parameter NRTL equation was used to fit all experimental data with satisfactory accuracy. These data indicate that the binary mixture has large negative deviations from Raoult's law. Also the temperature-dependent UNIFAC group-interaction parameters between the

Acknowledgements

Dr. S.K. Chaudhari would like to thank the Spanish Ministry of Science and Technology for the fellowship. We thank Kimikal (Primagaz Group, Almería, Spain) for providing the R134a and the Catalan Goverment for finantial assistance (2005SGR-00944).

References (24)

  • I. Borde et al.

    Int. J. Refrig.

    (1995)
  • E. Preisegger et al.

    Int. J. Refrig.

    (1992)
  • P. Marchi et al.

    J. Chem. Thermodyn.

    (2006)
  • M. Kleiber

    Fluid Phase Equilib.

    (1995)
  • M. Kleiber et al.

    Comput. Chem. Eng.

    (1998)
  • E.R. López et al.

    Fluid Phase Equilib.

    (1997)
  • A. Valtz et al.

    Fluid Phase Equilib.

    (2005)
  • S.I. Tseregounis et al.

    AIChE J.

    (1994)
  • E.R. López et al.

    Ind. Eng. Chem. Res.

    (2004)
  • A. Coronas et al.

    J. Chem. Eng. Data

    (2002)
  • H. Renon et al.

    AIChE J.

    (1968)
  • U. Weidlich et al.

    Ind. Eng. Chem. Res.

    (1987)
  • Cited by (15)

    • Searching for suitable lubricants for low global warming potential refrigerant R513A using molecular-based models: Solubility and performance in refrigeration cycles

      2021, International Journal of Refrigeration
      Citation Excerpt :

      In addition to PECs, the solubility of R1234yf and R134a in PEGDMEs has also been calculated with polar soft-SAFT and results presented here. In particular, the experimental data for systems R1234yf + TrEGDME (Fang et al., 2018), R134a + MEGDME (Chaudhari et al., 2008), R134a + DEGDME (Zehioua et al., 2010), R134a + TrEGDME (Coronas et al., 2002) and R134a + TEGDME (Tseregounis and Riley, 1994) were accurately reproduced fitting a temperature independent binary parameter ξ (see Table 2). Results are plotted in Fig. 4 for R1234yf and Fig. 5 for R134a.

    • Systematic modeling under uncertainty of single, double and triple effect absorption refrigeration processes

      2019, Energy
      Citation Excerpt :

      The non-random two liquid (NRTL) model [10] is the activity coefficient model used most frequently [11–21], [22], possibly because of the widely investigated NH3/H2O system, which has been modelled by the electrolyte-NRTL model [23]. The universal quasi-chemical functional-group activity coefficients (UNIFAC) model [24] has also been considered [25,26] as it enables the modeling of mixture components using functional groups, which makes the model very versatile. In terms of EoS, applications include Soave-Redlich-Kwong (SRK), Peng-Robinson (PR), Patel-Teja (PT) [27], PR with various alpha functions and mixing rules [11,12,18,19,21], as well as soft-SAFT [28], to name but a few.

    • Absorption refrigeration processes with organic working fluid mixtures- a review

      2019, Renewable and Sustainable Energy Reviews
      Citation Excerpt :

      In this respect, the works are discussed below in two separate sections, one including NRTL models and one including all cases other than NRTL, where properties other than equilibrium solubility have also been considered. Chaudhari et al. [52] derive experimental VLE data for the mixture R134a/MEGDME. The data are then correlated using NRTL [53].

    • High-pressure fluid-phase equilibria: Experimental methods and systems investigated (2005-2008)

      2011, Fluid Phase Equilibria
      Citation Excerpt :

      In the following, we summarize some points that are important for experimental methods for phase equilbrium measurement. Almost 800 articles with experimental data on high pressure phase equilibria were found [163–818]. More than 2400 systems have been investigated, from pure-components, binary systems up to complex mixtures with many components.

    View all citing articles on Scopus
    1

    Present address: National Chemical Laboratory, Pune 411008, India.

    View full text