From alper@zohar.bms.com Tue Sep 22 06:45:25 1992 Date: Tue, 22 Sep 92 10:45:25 EDT From: alper@zohar.bms.com (Howard Alper) Subject: dieelctric constants To: chemistry@ccl.net monica(and other interested in the dielectric constant), the computation of the dieelctric constant can be a fairly complicated matter, but for simple dipolar fluids(and polyatomic fluids that to a good approximation can be represented as dipolar fluids) the problem of computing the dielectric constant has been solved in a series of wonderful papers by m. neumann and o. steinhauser in the late 70's and early 80's. a good reference would be: m. neumann, mol. phys. 50, p841(1983), in which a lot of the basic theory is laid out. some other references are: m. neumann, o. steinhauser and g. s. pauley, mol. phys. 52, p97(1984), m. neumann and o. steinhauser, chem. phys. lett. 95(4,5), p417(1983), m. neumann, j. chem. phys. 85, p1567(1986). it is essential in the computation of the dielectric constant to effectively account for the long-range coulombic forces, either through a reaction field or an ewald sum. you may want to check if your program includes these corrections for the long-range electrostatic forces. here's a signpost: when you calculate the so-called GHk factor(which = /Nm**2, where M is the total system diopole moment, N= # molecules, and m=monomer dipole moment), oops that's the Gk factor, look at the value of Gk. if it is somewhat less than 1.0(like Gk=0.15) then your calculation is probably not accounting for the long-range coulombic contributions with a reaction field or ewald sum, and you will have to get or write code to calculate these important interactions. but once you have done that, then the references above will describe how to calculate the dielectric constant. howard alper