Organisers

• Elvira Guardia (Universitat Politècnica de Catalunya, Barcelona, , Spain)
• Marco Masia (Università degli studi di Sassari, Italy)

Supports

   ESF-SimBioMa

   CECAM

Description

A potential energy function has to incorporate information about the system it describes. Much effort has been put into improving intermolecular potentials by refining simple models. This includes the representation of molecules as collections of atom-centered interaction sites with fixed partial charges; since the charges are fixed, there is no explicit treatment of electronic polarization, and intermolecular interactions are treated as pairwise additive. Though the impact of this approximation is diminished through the use of effective pair potentials with enhanced charges, the lack of explicit polarization is physically incorrect [2,3] and is well-known to be problematic for interactions with charge concentrated ions, interactions of ions with pi-electron systems, and even for less obvious cases such as polar solutes in low-dielectric media [4-8]. Nevertheless, after 30 years and universal agreement on the importance of the problem, generally accepted, broadly applicable polarizable force fields have not emerged, multiple treatments of polarizability (inducible dipoles, fluctuating charges, Drude oscillators, etc.) remain under consideration, and simulations of extended inhomogeneous systems with polarizable force fields are still uncommon. Recently, given the urge for more realistic descriptions in computer simulations [5,8-11], and thanks to the increased computational power, the activity in the field is knowing a renovated enthusiasm; studies with polarizable force fields are being performed in different areas of research such as liquid-air interfaces [6,7], water-salt solutions [4,5,12-14], liquid water [14,15] (its properties in ‘exotic’ conditions are still unknown), ionic liquids [16], cell membranes [17] and molten salts [14,18]. Nonetheless, it seems that development and thorough testing of a polarizable force field is still limited to a small niche of researchers, and that its spread to a wider simulation community is somehow damped by myriad applications of nonpolarizable force fields in modeling complex systems. In the medium and long term the impact and prospective capabilities of the simulation work will be affected by the quality of the underlying description of molecular energetics [2,3]. In this atmosphere, it was decided to gather the researchers to focus on current research on polarizability and polarizable force fields and to boost its applications, from molecular biology to medicinal chemistry and materials science.

Scientific Objectives

The goal of the workshop is to provide an opportunity for computational scientists from different branches of science to discuss the state of the art in the field of polarizable force fields. We will invite experts (see preliminar list of acceptances) active in the field, both with theoretical and applied contributions, but we also aim at having a vibrant participation of junior researchers and of newcomers to the field. The discussions throughout the meeting should foster new collaborations and lead to new ideas for advances in this new and exciting area of research. We plan to host 15-20 invited talks of 40 minutes, plus 10-15 contributed short talks, all of them followed by at least 15 minutes of discussion.

The following aspects will be covered:

1- finding a common framework for the polarizable modeling of proteins, interfaces, liquids and surfaces. As mentioned above, different schemes have been developed by the simulation community to include polarization in molecular simulations. Which are the differences in the accuracy among them? Shall we end up using only the best of them for all applications, or shall we deal with one of them depending on the systems to be studied?

2- limits of the polarizable models due the “Pauli effects”. It has been recently found that a multipole description is not enough to take into account electron cloud repulsion at short range [5,8-11]; this effect could be included by making use of damping functions. There is still debate on the functional forms they should have and how to implement them with the Ewald summation technique.

3- strategies to parameterize polarizable force fields. As for nonpolarizable force fields, there is a plenty of techniques to chose the parameters; is there any strategy which have given better results? Which factors should be taken into account for a good parameterization? Is it realistic to think of a ‘universal model’ (for water for example) which reproduces accurately the properties at different states (from gas to solid)?

4- properties of complex systems which can be reproduced only with polarizable force fields. It is common experience that simple point charge models reproduce reasonably some properties; how will the description of the complex systems be improved by using polarizable potentials? Do we get a better estimate of properties, such as, for example, solvation free energies, hydrogen bonding and phase boundaries?

5- the parameterization of “simple” polarizable force fields. Since a vast part of the simulation community makes use of general purpose codes, there should be agreement on the functional forms of polarizable models which would be included in such codes; a convergence of models to simple (few parameters) forms would certainly ease the use of polarizable force fields for a wider community.

6- a preview of future developements. Inclusion of dipolar polarizability represents certainly an improvement of actual models; nevertheless in some case it is not enough, and we should go beyond this description by including higher multipolar effects. When does the dipole polarizability not give a satisfactory description of the interactions? Will it be possible to have a model which accounts also for charge transfer?

References

[1] Science 321, 783 (2008)
[2] Stone, A. J. Science 321, 787 (2008)
[3] Jorgensen, W. L. J. Chem. Theory Comput. (2007) 3, 1877
[4] Guardia, E.; Skarmoutsos, I.; Masia, M. J. Chem. Theory. and Comput. (2009), 5, 1449
[5] Masia, M.; Probst, M.; Rey, R. J. Chem. Phys. (2005) 123, 164505
[6] Jungwirth, P.; Tobias, D. J. Chem. Rev. (2006) 106, 1259
[7] Kuo, I-F. W.; Mundy, C. J. Science (2004) 303, 658
[8] Kaminski, G. A.; Stern, H. A.; Berne, B. J.; Friesner, R. A. J. Phys. Chem. A (2004) 108, 621
[9] Giese, T. J.; York, D. M. J. Chem. Phys. (2005) 123, 164108
[10] Piquemal, J.-P.; Chelli, R.; Procacci, P.; Gresh, N. J. Phys. Chem. A (2007) 111, 8170
[11] Soderhjelm, P.; Ohrn, A.; Ryde, U.; Karlstrom, G. J. Chem. Phys. (2008) 128, 014102
[12] Bucher, D.; Kuyucak, S. J. Phys. Chem. B (2008) 112, 10786
[13] Wick, C. D.; Xantheas, S. S. J. Phys. Chem. B (2009) 113, 4146
[14] Salanne, M.; Vuilleumier, R.; Madden, P. A.; Simon, C.; Turq, P.; Guillot, B. J. Phys.: Condens. Matter (2008) 20, 494207
[15] Paesani, F.; Voth, G. A. J. Phys. Chem. C (2008) 112, 324
[16] Qiao, B.; Krekeler, C.; Berger, R.; Delle Site, L.; Holm, C. J. Phys. Chem. B (2008) 112, 1743
[17] Harder, E.; Mackerell, A. D.; Roux, B. J. Am. Chem. Soc. (2009) 131, 2760
[18] Bitrian, V.; Trullas, J. J. Phys. Chem. B (2008) 112, 1718