Continuum and dipole-lattice models of solvation

Journal of Physical Chemistry B

Volumn 101, Issue 51, 1997, Pages 11254-11264

  Papazyan, Arno ^a   Warshel, Arieh ^a  

^a UNIVERSITY OF SOUTHERN CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

MATHEMATICAL MODELS; NUMERICAL METHODS;

CLAUSIUS MOSSOTTI EQUATION; CONTINUUM DIELECTRIC MODELS; DIPOLE LATTICE; POISSON EQUATION; SOLVATION;

ELECTROSTATICS;

EID: 0031335637     PISSN: 15206106     EISSN: None     Source Type: Journal    
DOI: 10.1021/jp971632j     Document Type: Article

References (57)

1. Warshel, A. Computer Modeling of Chemical Reactions in Enzymes and Solutions; Wiley: New York. 1991.
2. Born, M. Z. Physik 1920, 1, 45.
3. Warshel, A.; Levitt, M. J. Mol. Biol. 1976, 103, 227.
4. Madura, J. D.; Gilson, M. K. Biophys. J. 1993, 64, A220.
5. Zhou, H.-X.; Bagchi, B.; Papazyan, A.; Maroncelli, M. J. Chem. Phys. 1992, 97, 9311.
6. Allen, M. P.; Tildesley, D. J. Computer Simulation of Liquids; Clarendon Press: Oxford, 1987.
7. Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117.
8. Florian, J.; Warshel, A. accepted for publication, J. Phys. Chem. 1997.
9. van Vleck, J. H. J. Chem. Phys. 1937, 5, 320.
10. Papazyan, A.; Maroncelli, M. J. Chem. Phys. 1991, 95, 9219.
11. Bagchi, B.; Vijayadamodar, G. V. J. Chem. Phys. 1993, 98, 3351.
12. Warwicker, J.; Watson, H. C. J. Mol. Biol. 1982, 157, 671.
13. Honig, B.; Sharp, K.; Yang, A. S. J. Phys. Chem. 1993, 97, 1101.
14. Davis, M. E.; McCammon, J. A. J. Comput. Chem. 1989, 10, 386.
15. Fröhlich, H. Theory of Dielectrics: Clarendon Press: Oxford, 1958.
16. Böttcher, C. J. F. Theory of Electric Polarization, 2nd ed.; Elsevier: Amsterdam, 1973.
17. Debye, P. Polar Molecules; The Chemical Catalog Company: New York, 1929.
18. Sauer, J. A. Phys. Rev. 1940, 57, 142.
19. Luttinger, J. M.; Tisza, L. Phys. Rev. 1946, 70, 954.
20. Cohen, M. H.; Keffer, F. Phys. Rev. 1955, 99, 1128.
21. Toupin, R. A.; Lax, M. J. Chem. Phys. 1957, 27. 458.
22. Zwanzig, R. J. Chem. Phys. 1963, 38, 2766.
23. Cole, R. H. Mol. Phys. 1974, 27, 1.
24. Adams, D. J.; McDonald, I. R. Mol. Phys. 1976, 32, 931.
25. Erzan, A.; Stell, G. Phys. Rev. B 1978, 18, 408.
26. Romano, S. Il Nuovo Cimento 1986, 7D, 717.
27. Edelstein, L. L.; Dorogan, I. V.; Simkin, B. Y.; Sheykhet, I. I. J. Mol. Liq. 1992, 51, 139.
28. Zhou, H.-X. J. Phys. Chem. 1993, 97. 4216.
29. Clercx, H. J. H.; Bossis, G. J. Chem. Phvs. 1993, 98, 8284.
30. Ravichandran, S.; Bagchi, B. J. Phys. Chem. 1994, 98, 11242.
31. Papazyan, A.; Maroncelli, M. J. Chem. Phys. 1995, 102, 2888.
32. Coalson, R. D.; Duncan, A. J. Phys. Chem. 1996, 100, 2612.
33. Lee, F. S.; Chu, Z. T.; Warshel, A. J. Comput. Chem. 1993, 14, 161.
34. Warshel, A.; Russell, S. T. Q. Rev. Biophys. 1984, 17, 283.
35. Lipinski, J.; Bartowiak, W. J. Phys. Chem. A 1997, 101, 2159.
36. Sham, Y. Y.; Chu, Z. T.; Warshel, A. J. Phys. Chem. B 1997, 101, 4458.
37. Madura, J. D.; Nakajima, Y.; Hamilton, R. M.; Wierzbicki, A.; Warshel, A. Struct. Chem. 1996, 7, 131.
38. Jensen, G. M.; Warshel, A.; Stephens, P. J. Biochemistry 1994, 33, 10911.
39. Equal polarity does not mean identical solvation because the arrangement of solvent dipoles around the solute, affected by r, can affect solvation energy even though it has no effect on the macroscopic dielectric constant of the pure DL.
40. Alternatively, the system can be viewed in a coarse-grained way, and the polarization can be averaged over a large enough volume element so that the number of dipoles in the volume approaches infinity and the volume element behaves "macroscopically". However, even in this case, as soon as one goes back and uses continuum concepts at microscopic length scales the assumption of infinite dipole density is made, at least implicitly.
41. Skaf, M. S.; Fonseca, T.; Ladanyi, B. M. J. Chem. Phys. 1993, 98, 8929.
42. Luo, H. Q.; Tucker, S. C. J. Am. Chem. Soc. 1995, 117, 11359.
43. Gilson, M. K. Curr. Opin. Struct. Biol. 1995, 5, 216.
44. King, G.; Lee, F. S.; Warshel, A. J. Chem. Phys. 1991, 95, 4366.
45. Note that this field inside the sample is an artificial but necessary consequence of a continuum representation and does not correspond to a microscopically meaningful field such as that on individual dipoles.
46. In addition to providing an objective method of gauging how close a real, discrete solvent is to being a continuum dielectric in the context of solvation, the calculation of microscopic dielectric constants can measure the impact of certain numerical tricks (such as the use of cut-off radii in electrostatic forces during molecular simulations) on the solvation behavior of molecular models. For instance, even if a molecular solvent model with cut-off captures the solvation energetics of a solute, eqs 12 and 16 can, in principle, measure widely different microscopic dielectric constants for the models with and without cut-off and reveal that the solvation energy of the solute is recovered largely through changing the structure of the solvent near the solute rather than by exhibiting a globally correct solvation behavior. Such information can be used to improve existing computational models.
47. Block, H.; Walker, S. M. Chem. Phys. Lett. 1973, 19, 363.
48. 0 in other cases such as solvation of an ion or a dipole (or any solute charge distribution).
49. -5 au that was used for LDLs and PPLs. This higher field strength had to be used in order to obtain reasonable statistics with the thermally fluctuating BDLs and leads to a small but measurable dielectric saturation (i.e., slightly underestimates the dielectric constant). The dielectric saturation effect is corrected by calculating the average polarization of individual dipoles in the sample, calculating the corresponding y value in eq 2, and calculating the linear polarization dipole from the linear limit of eq 2 using the linear polarizability given by eq 3. The total linear (corrected) polarization of the sample is then the total of the average individual linear polarizations.
50. However, it clearly applies to a model that can represent its polarization and solvation behavior, such as an LDL or a PPL.
51. It is important to note that the particular functional form (such as CM or KF relationships) connecting the polarity and the dielectric constant of a material is inconsequential in determining its net dielectric response for a given dielectric constant. The same argument can be made for the solvation response. In addition, the solvation energy of a solute is also influenced by the details of the solvent model in the vicinity of the solute. Since any simplified model (such as a structureless continuum or a dipolar grid) has little pretense of capturing these details, that information about the real system has to be absorbed into the way a cavity is constructed around the solute. We do not address this issue in this article.
52. Kirkwood, J. G. J. Chem. Phys. 1936, 4, 592.
53. We are not implying that such models would be useless. However, it should be recognized that a nonlinear continuum model implies a physical inconsistency. Although such models can be useful, they would nevertheless be ad hoc.
54. Papazyan, A.; Warshel, A. J. Am. Chem. Soc., submitted for publication.
55. Keyes, T. Chem. Phys. Lett. 1980, 76, 369.
56. Note that the averaging is not done for obtaining a converged energy; it is for obtaining the average polarization profile corresponding to the already converged electric potential.
57. 33 repeatedly reproduced small interaction energies which correspond to large effective dielectric constants.