In an attempt to understand the phase behavior of aqueous hydrogen fluoride, the clustering in the mixture is investigated at the molecular level. The study is performed at the mPW1B95/6-31+G(d,p) level of theory. Several previous studies attempted to describe the dissociation of HF in water, but in this investigation, the focus is only on the association patterns that are present in this binary mixture. A total of 214 optimized geometries of (HF)n(H2O)m clusters, with m + n as high as 8, were investigated. For each cluster combination, several different conformations are investigated, and the preferred conformations are presented. Using multiple linear regressions, the average strengths of the four possible H-bonding interactions are obtained. The strongest H-bond interaction is reported to be the H2O...H-F interaction. The most probable distributions of mixed clusters as a function of composition are also deduced. It is found that the larger (HF)n(H2O)m clusters are favored both energetically and entropically compared to the ones that are of size m + n < or = 3. Also, the clusters with equimolar contributions of HF and H2O are found to have the strongest interactions.

Recently a model has been introduced describing the heat effects of hydrogen fluoride (HF) more accurately than other models available. In this work, we extend this model to predict the coexistence properties of some binary HF mixtures. Twelve HF mixtures are explored initially (R12, R22, R32, R113a, R123, R124, R134a, R142b, R152a, n-propane, HCl, and Cl 2 ) because these are the systems where experimental data are available. Good agreement with experiment, especially in the correlation of miscibility gaps, is realized through the use of two mixing rules, namely, van der Waals (VDW) and Wong-Sandler. We also correlate the binary interaction parameter from the VDW mixing rule using the dipole moment to predict the properties of the HF-R141b system. We also use this model to correlate and predict the properties of an aqueous HF system, including a prediction for a double azeotrope of this system at very dilute concentrations of HF.

This work describes a methodology for the development of a thermodynamic model describing the substances that show strong self- and cross-association interactions. The methodology is fundamentally based on the chemical theory of association interactions. The system used as a case study in this work is a binary mixture containing hydrogen fluoride and water (HF + H2O). Earlier studies have failed to provide a reasonable description of this binary mixture because of the complex association interactions between these compounds, which were not adequately modeled. In this work, the phase behavior of this mixture is understood by exploring these complex association interactions. Pure HF was modeled using 14 different association schemes that allow the formations of different physically meaningful oligomers with different distribution schemes (1−2, 1−6, 1−2−6, etc.), where the 1−2 scheme allows the formation of monomers and dimers and likewise. The parameters for these pure component schemes were obtained by correlating the phase coexistence properties of pure HF and were also used to predict several other pure component properties (ΔH vap, C P , C v , Z, etc.) The dominance of these association patterns and their distribution were understood on the basis of their predictive ability. The pure component association schemes that were developed for HF and water were extended to the binary mixture. The phase coexistence properties were correlated using different association patterns for the pure components with and without considerations for the strong association between them. The significance of these self- and cross-association patterns are studied and understood on the basis of the correlative and the predictive ability of the association schemes. The effect of including the cross-associates that are most likely to be formed in this mixture, from a molecular level hybrid meta-density functional theory study, is also discussed. The methodology described in this work can be utilized to understand and predict the bulk-phase thermodynamic properties of substances that show complex association interactions at a molecular level.

The highly nonideal behavior of hydrogen fluoride (HF) vapor has been considered to be the origin of its numerous vapor phase anomalies. In this work, we report one such potential vapor phase anomaly for HF. For a nonassociating substance like propane, the response functions go through a maximum only once in the supercritical region. However, for HF, when an association model is used to predict the isothermal compressibility (KT), it exhibits a maximum in the supercritical region more than once, and this peak extends well in to the superheated vapor region upon decompression. This theoretical prediction is also supported by two other models recently developed for HF. Note that experimental values of KT for HF have not been reported in the literature so far. Preliminary investigations on this KT maximum for HF have suggested no reentrant spinodal, singularity-free scenario, or any additional first-order phase transition, unlike water, and, also, no lambda (or higher-order phase) transitions, unlike liquid helium. However, this KT peak is similar to the experimentally supported heat capacity (CP) peak of HF which extends into the supercritical and superheated vapor regions. Similar to the CP peak, which is understood based on vapor-phase clustering in HF, we relate KT to the derivatives of enthalpy and entropy of the system. Also, we analyze some of the P-v-T experimental data that are available to provide an overview of the KT behavior in the region of interest, and compare them with the model results. Finally, to explore the effect of including a distribution pattern for the oligomers, we report the results on a model that only includes association. Using this approach, we report KT results with and without a Poisson-type oligomer distribution and show that the KT appears once this distribution scheme is specified.

No abstract