Skip to main content

Intermolecular Forces

Theory & Computation
Selected Publications

R. Podeszwa, B. M. Rice, and K. Szalewicz, Predicting structure of molecular crystals from first principles, Phys. Rev. Lett. 101, 115503 (2008). [PDF] R. Bukowski, K. Szalewicz, G. C. Groenenboom, and A. van der Avoird, Predictions of the properties of water from first principles, Science 315, 1249 (2007). [URL] K. Szalewicz, K. Patkowski, and B. Jeziorski, Intermolecular interactions via perturbation theory: From diatoms to biomolecules, Structure and Bonding (Berlin) 116, 43 (2005). [PDF]


It is remarkable how broad a range of physical, chemical, and even biological phenomena originates from weak intermolecular interactions (also called van der Waals interactions).  These interactions do not involve forming a chemical bond between the interacting species. Intermolecular interactions (or forces) determine bulk properties of gases and liquids and are responsible for the very existence of molecular liquids and crystals. The knowledge of accurate intermolecular potential energy surfaces (PESs) is necessary to interpret high-resolution spectroscopic and scattering data, including the spectroscopic data coming from planetary atmospheres and the interstellar gas, as well as to construct and tune empirical potentials used in Monte Carlo (MC) or molecular dynamics (MD) bulk simulations.

In recent years, interactions of various molecules with helium became particularly important due to the development of superfluid helium nanodroplet spectroscopy. Weak intermolecular forces are responsible for the biomolecular recognition patterns and the catalytic activity of enzymes, and thus insights into intermolecular PESs are important for drug design. An interesting example of a macroscopic effect of van der Waals forces is given by the recent experimental evidence that Tokay geckos (Gekko gecko) owe their exceptional ability to climb smooth vertical surfaces to the van der Waals attractions between the surface and gecko toe-hairs.

For some purposes, the exact form of the potential energy function is not critical. It is possible to learn a great deal about the general characteristics of gases, liquids, and solids by applying simulation methods with simple potentials. This was the approach used, of necessity, when computers were much less powerful than they are now, but it is still valuable in applications such as protein folding or supercooled liquids and glasses if the objective is to explore generic features rather than to predict the behavior of a particular system.

On the other hand, there are applications where the quality of the potential is paramount, and calculations with an inadequate potential are a waste of effort. One such application is the prediction of crystal structures of molecular solids. Most molecules of any size, including the molecules of pharmaceutical drugs, can crystallize in many different forms, or polymorphs, and it is important for commercial as well as scientific reasons to know which of many hypothetical forms, differing in lattice energy by only a few kJmol–1, are the most stable.

Another example is water. For simulations of liquid water and of biological systems in the presence of water, very simple descriptions are needed because of computational constraints. They usually approximate the many-body effects by modifying the pair potential (the function that describes how the energy of a pair of molecules depends on their relative geometry), typically by enhancing the molecular dipole moment. These models are quite successful in modeling liquid water at ambient temperature and pressure, but give a very poor account of the water dimer, because the modified pair potential is incorrect.

The fundamental reason that the pairwise approximation fails is the existence of pairwise nonadditive interactions.  The interaction between any two molecules leads to a distortion of both molecules, which in turn modifies their interactions with a third molecule. A three-body function is needed to take such effects into account.  Szalewicz group has recently developed a potential derived entirely from first principles that captures  properties of water "across-the-board" by reproducing experimental data from dimer spectra, through properties of small clusters,  up to properties of liquid water and ice.  The potential also supplements experiments:  for example,   for the water dimer it allowed reassignment of spectral lines and provided accurate values for many transitions which could not be measured.  For the liquid water,  this work has shown that the tetrahedral structure of water,  i.e.,  the experimental finding that on the average each water forms four hydrogen bonds,  is mainly due to three-body nonadditive effects.

Research Projects: