[lammps-users] Several questions on command in input file.

Dear all,

I am not sure about my understanding for some commands bellow. Could you give me some advises?

  1. lattice and region. Lattices of style sc, fcc, bcc, and diamond are 3d lattices. The differenc among them are the number of basis atoms. What is the influence on the number of basis atoms ? And for LJ particle, the scale argument stands for reduced density rho*. I feel the reduced density = number of particle / volume. Am I right? And what is the meaning of rho*?

  2. The commands of pair_style and pair_coeff are needed for GB particles. In the pair_style, arguments are g u m and cutoff. The first three are obtained from article and theng =u =m =1. Cutoff just for precision the bigger the better. While in the pair_coeff, the arguments are type name of I,J, e,s,eia, eib, eic, eia, eib, eic,. According to the introduction in the manual and some articles, e relates to interaction between particles and s relates to equilibrium distance. But When I set s=1, two particle with initial xyz (2 0 0) and (1 0 0) will shift to (2.06071 0 0) and (0.939294 0 0) afer minimize computation respectively. I feel the distance 1 should be more stable according to s=1 although the new distance is a little larger than 1. More strange thing is When I set s=10 to the same initial atoms. The stop coordinates are (2.61231 0 0) and (0.387691 0 0). Their distance is much lower than 10 and can not be accepted! I do no know what is the error?

Thanks for any comments.

Lattices are discussed in any solid state textbook.. Rho* is reduced density
in LJ units - you'll have to read up on LJ units, which are really unitless
parameters.

The GB potential will not work with minimization, in that the minimizer
does not alter the orientation of particles. So run a dynamics simulation
to equilibrate instead.

Steve