Condensed-phase properties

The most important molecular interactions of all are those that take place in liquid water. For many years, chemists have worked to model liquid water, using molecular dynamics and Monte Carlo simulations. Until relatively recently, however, all such work was done using effective potentials [4T], designed to reproduce the condensed-phase properties but with no serious claim to represent the tme interactions between a pair of water molecules. 

Statistical mechanics computations are often tacked onto the end of ah initio vibrational frequency calculations for gas-phase properties at low pressure. For condensed-phase properties, often molecular dynamics or Monte Carlo calculations are necessary in order to obtain statistical data. The following are the principles that make this possible. 

Computer simulations therefore have several inter-related objectives. In the long term one would hope that molecular level simulations of structure and bonding in liquid crystal systems would become sufficiently predictive so as to remove the need for costly and time-consuming synthesis of many compounds in order to optimise certain properties. In this way, predictive simulations would become a routine tool in the design of new materials. Predictive, in this sense, refers to calculations without reference to experimental results. Such calculations are said to be from first principles or ab initio. As a step toward this goal, simulations of properties at the molecular level can be used to parametrise interaction potentials for use in the study of phase behaviour and condensed phase properties such as elastic constants, viscosities, molecular diffusion and reorientational motion with maximum specificity to real systems. Another role of ab initio computer simulation lies in its interaction... 

Understanding the condensed-phase properties of HE materials is important for determining stability and performance. Information regarding HE material properties [such as the physical, chemical, and mechanical behaviors of the constituents in plastic-bonded explosive (PBX) formulations] is necessary for efficiently building the next generation of explosives as the quest for more powerful energetic materials (in terms of energy per volume) moves forward.1... 

The quantities that have been presented do effectively characterize the electrostatic potential on a molecular surface. We have shown that a number of macroscopic, condensed-phase properties that depend upon noncovalent interactions can be expressed in terms of some subset of these quantities (frequently... 

This is accordingly a unified approach to representing and predicting condensed phase properties that are determined by noncovalent interactions. We summarize it conceptually as a general interaction properties function (GIPF), Eq. (9) ... 

P. Politzer, J.S. Murray, Computational prediction of condensed phase properties from statistical characterization of molecular surface electrostatic potentials. Fluid Phase Equil. 185, 129-137 (2001)... 

Another reactive force field that is dependent on bond-order was developed by van Duin, Dasgupta, Loran, and Goddard [183] for hydrocarbons. The configurational energy is described as the sum of energy contributions from internal modes as well as non-bonding van der Waals and Coulombic interactions, but the parameters of the functions that describe each contribution is dependent upon the bond order of atoms involved in each description. It is assumed that the bond order between an atom pair is dependent on the interatomic separation. While this model has been used to predict bond dissociation energies, heats of formation and structures of simple hydrocarbons, it was not applied to predict condensed phase properties. However, the form of the potential should allow for condensed phase studies. 

Politzer P, Murray JS (1999) Representation of condensed phase properties in terms of molecular surface electrostatic potentials. Trends Chem. Phys. 7 157... 

Descriptions of classical and quantum mechanical methods for simulating energetic salts are presented. An overview of recent applications of these methods for predictions of gas-and condensed-phase properties, chemical reactivities, and phase transitions is given. The limitations and some suggestions for further developments of the methods are also discussed. 

Volumetric expansion caused by dissolution of CO2 into liquid polymers is accompanied by significant changes in the physical properties of the condensed phase. Properties that change include the melting point, viscosity, interfacial tension, diffusion coefficients, and potentially solubilities of other species and the polarity of the liquid phase. 

Direct comparison of force fields to benchmark-quality CCSD(T) energies is complicated by the fact that most standard, workhorse force fields do not include polarization terms. This leads to errors, but these errors can be partially compensated by other errors. Hence, a force field that compares poorly to CCSD(T) benchmarks for a set of van der Waals dimers may still perform fairly well for condensed-phase properties, due to error cancellation. This is the rationale for obtaining atomic charges in the AMBER force field using restrained electrostatic potential (RESP) fitting (Bayly, 1993) to modest-quality Hartree-Fock/6-31G quantum chemical computations this method tends to overestimate dipole moments, but this is considered beneficial for simulations in water, to approximately cancel errors from neglecting polarization effects... 

We consider a binary two-phase system at temperature T. One phase is a liquid and the other is a solid. Since the effect of pressure on condensed-phase properties is normally negligible at low or moderate pressures, we do not need to specify the pressure. Let component 1 be the liquid solvent and component 2 the solid solute. 

The disadvantage of these generic force fields is that they are usually not sufficiently accurate to predict condensed phase properties reliably. For example, fair agreement between experiment and UFF is a prediction of a bond length within 0.008 nm (33). (The accuracy for main groups elements with UFF is usually better than this.) This is a 4% error for a bond length of 0.2 nm, which translates into ca. 12% error in the density The point is that a 4% error in the structure of an isolated molecule is barely perceptible to the eye, yet when physical properties of condensed phases are predicted for this structure, imfortimately large errors may be encountered. 

---