2017, Vol.46, No.7

An experimental and computational study on the solubility of argon in propan-2-ol at high temperatures and pressures was performed. The following values of the Henry’s law constant for the solution of argon in propan-2-ol were obtained: 58 ± 3 MPa at 480 K, 99 ± 3 MPa at 420 K, and 114 ± 2 MPa at 360 K.

Studying phase equilibria and constructing phase diagrams is not only an important fundamental problem, but also an essential task for chemical engineering and technology.1,2 However, even for simple binary systems the process of collecting and systematising that type of thermophysical data is far from completion. Usually, both experimental and computational approaches are used to obtain the necessary information about the phase behavior of systems under consideration.

Noble gases have found wide application in medicine, environmental chemistry, and lighting. The solubility and phase behavior of noble gases in water and organic solvents was studied rather widely,3 however, for many systems the available information is not complete. The Henry’s law constant for the solution of argon in propan-2-ol was previously reported only in a limited temperature range.47 This communication describes an experimental and computational study on the solubility of argon in propan-2-ol at high temperatures.

The experimental setup for the present gas solubility measurements8 is the same as employed in earlier studies. The following substances were used: argon (CAS number 7440-59-7) supplied by Air Liquide in a gas tank under a pressure of 30 MPa with a volume fraction of 99.9999% and propan-2-ol (CAS number 67-63-0) supplied by Honeywell Riedel–de Haën with a purity of >99.9%. The measuring cell was filled with argon and heated to about 20 K above the desired measuring temperature, and then the desired amount of propan-2-ol was added into it. When the mixture achieved a homogeneous state, the cell was slowly cooled down with the aim to reach the saturated liquid state.9 The experiments were carried out at the temperatures 360, 420, and 480 K. The density of argon and propan-2-ol as well as the saturated vapor pressure of pure propan-2-ol at these temperatures were calculated with equations of state.10,11 The experimental raw data together with the values used in the processing of them are presented in the Supporting Information. The resulting dependence of the phase equilibrium pressure on the mole fraction of argon in liquid propan-2-ol at 360, 420, and 480 K is depicted in Figure 1.

As can be seen, the isotherms may be well approximated by straight lines. The slopes of these lines calculated with the least-squares technique were used to estimate the Henry’s law constant at these temperatures; the estimated values together with their uncertainties are listed in Table 1.

Table
Table 1. Experimental values of the Henry’s law constant for the solution of argon in propan-2-ol at 360, 420, and 480 K12
Table 1. Experimental values of the Henry’s law constant for the solution of argon in propan-2-ol at 360, 420, and 480 K12
T/K 480 420 360
HAr/2-PrOH/MPa 58 ± 3 99 ± 3 114 ± 2

Several approaches have been proposed in the literature13 to obtain the Henry’s law constant on the basis of molecular simulation. The Henry’s law constant is related to the residual chemical potential of argon in propan-2-ol at infinite dilution \(\mu _{\text{Ar}}^{\infty }\).14 The molecular model for argon was developed in Ref. 15 and consists of a single Lennard–Jones site. The parameters ε and σ for argon were estimated to be the following: εAr/kB = 116.79 K, σAr = 3.3952 × 10−10 m, where kB is the Boltzmann constant. The molecular model for propan-2-ol consists of Lennard–Jones united atom sites for the two methyl, the methanetriyl, and the oxygen groups, accounting for repulsion and dispersion. Point charges were located on the methanetriyl and the oxygen Lennard–Jones sites, as well as on the nucleus position of the hydroxy hydrogen. The Coulombic interactions account for both polarity and hydrogen bonding. The interaction between unlike Lennard–Jones sites of two propan-2-ol molecules was defined by the Lorentz–Berthelot combining rule. To describe a binary mixture on the basis of pairwise additive potential models, the binary Lennard–Jones parameters σAB and εAB have to be determined. In this study a modified Lorentz–Berthelot rule was applied that has been discussed in detail elsewhere:16 σAB = 0.5·(σA + σB), \(\varepsilon _{\text{AB}} = \xi \cdot \sqrt{\varepsilon _{\text{A}} \cdot \varepsilon _{\text{B}}} \). The binary interaction parameter ξ can be adjusted to values of the Henry’s law constant. During simulation, the mole fraction of argon in propan-2-ol is exactly zero, as required for infinite dilution because test particles for calculating \(\mu _{\text{Ar}}^{\infty }\) are instantly removed after the potential energy calculation. Simulations were performed at temperatures ranging from 50% to 95% of the critical temperature of propan-2-ol and the according saturated liquid density and vapour pressure of pure propan-2-ol.17 The simulations were carried out using the software ms218 on the “Hazel hen” machine at the High Performance Computing Centre in Stuttgart (hazelhen.hww.de). The state-independent parameter ξ was adjusted such that the differences between the simulation results, the present experimental values and the literature data were minimal. It was found that for the solution of argon in propan-2-ol ξ = 0.964. The simulated values together with their statistical uncertainties are listed in Table 2. Figure 2 shows the good agreement between the simulated and experimental values of this work and literature data.

Table
Table 2. Henry’s law constant for the solution of argon in propan-2-ol at various temperatures from molecular simulation19
Table 2. Henry’s law constant for the solution of argon in propan-2-ol at various temperatures from molecular simulation19
T/K HAr/2-PrOH/MPa
254.1 101.1 ± 0.9
279.6 114.6 ± 0.8
305.0 119.3 ± 0.5
330.4 118.1 ± 0.3
355.8 114.4 ± 0.3
381.2 106.4 ± 0.2
406.6 96.4 ± 0.2
432.0 84.4 ± 0.1
457.5 71.4 ± 0.1
482.9 57.4 ± 0.1

This work was supported by the Ministry of Education and Science of Russian Federation (state contract No. 10.723.2016/ДAAД) and Deutscher Akademischer Austausch Dienst (grant No. 91547204).

Supporting Information is available on http://dx.doi.org/10.1246/cl.170221.

P. A. Nikolaychuk

M. Linnemann

Y. M. Muñoz-Muñoz

E. Baumhögger

J. Vrabec