Interaction potentials for water from accurate cluster calculations
The abundance of water in nature, its function as a universal solvent and its role in many chemical and biological processes that are responsible for sustaining life on earth is the driving force behind the need for understanding its behavior under different conditions, and in various environments. The availability of models that describe the properties of either pure water/ice or its mixtures with a variety of solutes ranging from simple chemical species to complex biological molecules and environmental interfaces is therefore crucial in order to be able to develop predictive paradigms that attempt to model solvation and reaction and transport in aqueous environments. In attempting to develop these models the question naturally arises 'is water different/more complex than other hydrogen bonded liquids'. This proposition has been suggested based on the 'anomalous' behavior of its macroscopic properties such as the density maximum at 4 C, the non-monotonic behavior of its compressibility with temperature, the anomalous behavior of its relaxation time below typical temperatures of the human body, the large value and non-monotonic dependence below 35 C of the specific heat of constant pressure, the smaller than expected value of the coefficient of thermal expansion. This suggestion infers that simple models used to describe the relevant inter- and intra-molecular interactions will not suffice in order to reproduce the behavior of these properties under a wide temperature range. To this end, explicit microscopic level detailed information needs to be incorporated into the models in order to capture the appropriate physics at the molecular level. From the simple model of Bernal and Fowler, which was the first attempt to develop an empirical model for water back in 1933, this process has yielded ca. 50 different models to date. A recent review provides a nearly complete account of this effort coupled to the milestones in the area of molecular simulations such as the first computer simulation of liquid water by Barker and Watts and Rahman and Stillinger, the first parametrization of a pair potential for water from ab-initio calculations by Clementi and co-workers and the first simulation of liquid water from first principles by Parrinello and Carr. Many of the empirical pair potentials for water that are used widely even nowadays were developed in the early 1980's. These early models were mainly parameterized in order to reproduce measured thermodynamic bulk properties due to the fact that molecular level information for small water clusters was limited or even nonexisting at that time. Subsequent attempts have focused in introducing self-consistent polarization as a means of explicitly accounting for the magnitude of the non-additive many-body effects via an induction scheme. Again the lack of accurate experimental or theoretical water cluster energetic information has prevented the assessment of the accuracy of those models.
- Research Organization:
- Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC05-76RL01830
- OSTI ID:
- 894863
- Report Number(s):
- PNNL-SA-45721; KC0301020; TRN: US200702%%439
- Resource Relation:
- Related Information: Intermolecular Forces and Clusters II, Structure and Bonding, 116:119-148
- Country of Publication:
- United States
- Language:
- English
Similar Records
Final Scientific/Technical Report
Testing the Adequacy of Simple Water Models at the Opposite Ends of the Phase Diagram