Improved Evaluation of Liquid Densities Using van der Waals Molecular Models

Abstract

A new approach is presented to estimate molar volumes and densities of liquids in ambient conditions from van der Waals models, taking advantage of the correlation between the intermolecular volume and the atomic contributions to the molecular surface area. Using this approach, the role of hydrogen bonds can be quantified. The densities obtained prove remarkably close to the values derived from the ACD group contribution method. However, the present approach requires much less empirical parameters and may be applied to arbitrary H-C-N-O-F-S-Cl-Br compounds. Moreover, it is more reliable than earlier empirical procedures, including quantitative structure property relationships (QSPR). While it does not correct the deficiencies of group contribution methods, it provides a natural approach to introduce temperature effects.

Physics-Based Modeling of Chemical Hazards in a Regulatory Framework: Comparison with Quantitative Structure–Property Relationship (QSPR) Methods for Impact Sensitivities

“…9,10 While the determination of the relevant mechanisms can be extremely difficult for biological activities or toxicological end points, making assumptions regarding the underlying physics, in principle, is easier for the physicochemical properties of chemicals, because they are not dependent on the complex interactions that characterize biological functions. As a matter of fact, such models based on equations derived from physical considerations have been shown to systematically outperform QSPR methods for many properties, including densities 11 and flash point 12 of liquids, melting points 13 and sublimation enthalpies 14 of crystals, Gurney velocities of explosives, 15 and heats of decomposition of organic substances. 16−18 On the other hand, growing safety concerns and recent regulations, including the European Union (EU) REACH legislation, require the evaluation of more-complex physicochemical properties, such as dust explosivity 19 or the impact sensitivity of chemical energetic substances.…”

Inductive modeling of physico-chemical properties: Flash point of alkanes

“…Moreover, they often require extensive experimental data to fit the parameters involved. For simple fluid properties, such as liquid densities, approaches based on physical insight yield better results [9]. On the other hand, straightforward applications of regression techniques to extended datasets fail to take advantage of the fact that even complex prop- * Tel.…”

MATEO: A software package for the molecular design of energetic materials

“…It would require a well-defined procedure to derive atomic volumes from available liquid data. MATEO predictions of liquid densities rely on a very simple assumption concerning the dependence of the atomic molar volume on van der Waals parameters, which was found to work remarkably well [35]. Preliminary unpublished data indicate that even more reliable liquid densities will be obtained on the basis of the ARH approach [33].…”

“…Otherwise, the compound ). Custom components include EEMEO for molecular charge distributions [36], DFT-ENTHALPY for sublimation enthalpies and solid-state formation enthalpies (Section 3.2), ARH [33] and SUBDLIQ [35] for densities, respectively in the crystal and liquid states. is assumed to be a cation, and the sublimation enthalpy of the corresponding nitrate salt is then directly evaluated by the DFT-ENTHALPY module [37].…”

“…The performance of the models implemented in MATEO are relatively well documented in the literature with regard to densities [33,35], formation enthalpies of molecular crystals [36] and salts [37], decomposition temperatures [28] and other properties for which published models are used [27,29]. Thus, the present section focuses on less documented aspects: formation enthalpies of polymers, heats of explosion, and optimization of composite propellants.…”