Molecular simulation data for the vapor-liquid phase equilibria of binary mixtures of HFO-1123 with R-32, R-1234yf, R-1234ze(E), R-134a and CO2 and their modelling by the PCP-SAFT equation of state

In this Data in Brief article, we present predictive data for the vapor-liquid equilibria of the binary mixtures of HFO-1123 with R-32, HFO-1234yf, HFO-1234ze(E), R-134a and CO2 from molecular simulation. The VLE in the binary mixtures are then modeled by the PCP-SAFT equation of state. Therefore we determined PCP-SAFT parameters for the pure HFO compounds as well as binary interaction parameters for all mixtures. The simulation data and the PCP-SAFT modelling are discussed in a related research article (Raabe, 2019).


Specifications
The VLE data was acquired by Gibbs Ensemble Monte Carlo (GEMC) simulation using the code TOWHEE [2] and employing force field models from previous work [3e5]. The simulations were performed on the HPC compute server Phoenix of the TU Braunschweig Data format Filtered, analyzed. Experimental factors Simulations were performed in NPT ensemble. Each system consisted of 432 molecules. For each data point, the simulation was equilibrated for 200,000 cycles. The production runs consisted of 300,000 e500.000 cycles.

PCP-SAFT parameters
We here present PCP-SAFT parameters for the pure compounds HFO-1123, -1234yf, and -1234ze(E). Fig. 1 shows a comparison of the correlation of the VLCC and vapor pressure curve of R-1234ze(E), R-1234yf and HFO-1123 by the PCP-SAFT (red line) and available EOS models in REFPROP (dark grey line) as well as with experimental data. For the correlation of the mixture data, we employed PCP-SAFT parameters from literature for the pure compounds R-32, R-134a and CO 2. The parameters are also give in Table 3.
We also provide the fitted interaction parameters k ij for the EOS for the binary mixtures HFO-1123 with R-1234yf, R-1234ze(E), R-32, R-134a, CO 2. The k ij along with the relative average deviations (RAD %) of the pressure and saturated densities, and absolute average deviations (AAD) of the molar vapor Table 2 Data from GEMC simulations for the VLE in binary mixture CO 2 þ HFO-1123: mole fraction in the saturated liquid (x) and vapor phase (y), and saturated densities r L and r V . Values in parentheses denote standard deviations.  Fig. 1. Correlation of the VLCC and vapor pressure curve of R-1234ze(E)¼:, R-1234yf¼C and HFO-1123¼-by the PCP-SAFT (red line) and available EOS models in REFPROP [8](dark grey line). Shown as filled symbols are experimental VLE data for R-1234ze(E) ([7, 9e11]), R-1234yf ( [12]) and HFO-1123 ( [13,14]). Also given are GEMC simulation results @ for HFO-1123 that were employed in the fitting (d, [5]). composition of the correlations from the GEMC simulation results are summarized in Table 4. In [1] we provide depictions of calculated isotherms of all mixtures in comparison with the simulation data and calculations using REFPROP.

Molecular simulation
Predictions for the vapor-liquid equilibria of the binary mixtures were derived by Monte Carlo Gibbs ensemble (GEMC, [15]) simulations in the NPT ensemble using the simulation code TOWHEE [2]. Each system consisted of 432 molecules in total, but depending on the mixture studied, the number of the molecules of both components were varied to yield a feed composition within the two phase region. The Ewald sum technique [16] was employed to deal with the electrostatic interactions with a cut-off radius adjusted to half the box length, whereas the cut-off radius for the Lennard-Jones interactions was set to 12 Å. Standard long-range corrections to the energy and pressure were applied (e.g. Ref. [17]). For each data point, the simulation was equilibrated for 200,000 cycles. The production runs consisted of 300,000e500.000 cycles from which ensemble averages for the compositions and saturated densities of the coexisting phases for the imposed temperature and pressure were determined. Standard deviations of all ensemble averages were calculated by the standard block averaging technique (e.g. Refs. [17,18]).

PCP-SAFT-modelling
For CO 2 , we used the PCP-SAFT parameters proposed by Gross [7] whereas the parameters of R-32 and R-134a were taken from Vin s et al. [6]. In this work, we derived PCP-SAFT parameters for the compounds HFO-1123, R-1234yf and R-1234ze(E). For the tetraflouropropenes R-1234yf and R-1234ze(E), the PCP-SAFT parameters were determined by fitting calculated vapor pressure and liquid densities to experimental data, and we therefore employed the same experimental data set as in the fitting of the PC-SAFT model in our previous work [4]. As experimental data for HFO-1123 in literature are limited [13,14], we also employed our molecular simulation results [5] in the fitting of the EOS parameters for this compound.
To model the refrigerant mixtures, common combining rules for the PCP-SAFT parameters are used that employ an interaction parameter k ij for the interaction energy ε * ij between unlike segments. The interaction parameters k ij were derived by fitting to the GEMC simulation results for the VLE of the binary mixtures presented in this work.

Transparency document
Transparency document associated with this article can be found in the online version at https:// doi.org/10.1016/j.dib.2019.104014.