[gmx-users] some problems by freshman?

[zpxu]

Dear All:
        I am using GROMACS 3.1.4 to simulating 2-walled carbon nanotubes, as:

                  ---------------------
          ---------------------------------------
          ---------------------------------------      -> Inner Layer Tubule
                  ---------------------                -> Outer Layer Tubule
        Intralayer potential I use morse bonds and G96 angles, Interlayer I use L-J(6-12) potential,
and I write files as:
1)nt.gro:
multiwalled nanotube
    400
    1 NT_I  C_I    1   3.294   2.830   2.570   .0000   .0000   .0000
    ....
    1 NT_I  C_I    200   3.294   2.830   2.570   .0000   .0000   .0000
    1 NT_O  C_O    201   3.294   2.830   2.570   .0000   .0000   .0000
    ....
    20 20 50
(I use C_I and C_O to specify the carbon atom in innertube and outertube)
2)nt.top
#include "nt.itp"
[ moleculetype ]
   nt  1
 [ atoms ]
    1  C_I    1 MWNT  C_I    1    .000  12.011
......
(I do not need charge group, so set them from 1~400, because the large number of the same charge group must <64 ?) 
[bonds]
    1    2    3     .141800  478.900000   21.867000
......
[angles]
    1    2   14    2  120.000000  562.200000
 ...... 
[ system ]
  MultiWalled Nanotube
[ molecules ]
 nt 1
3)nt.itp
[ atomtypes ]
    C_I    12.01100    0.000       A   0.298007  2.1961
    C_O  12.01100    0.000       A   0.298007  2.1961
[ nonbond_params ]
  C_I    C_O    1   0.298007 2.1961 
  C_O   C_I    1   0.298007 2.1961 
[ bondtypes ]
C_I    C_I    3   0.1418  478.9  21.867
C_O  C_O   3   0.1418  478.9  21.867
[ angletypes ]
C_I    C_I    C_I    2       120    562.2 
C_O  C_O  C_O   2       120    562.2
4)em.mdp
title               =  nt
constraints         =  none
integrator          =  steep
dt                  =  0.002    ; ps !
nsteps              =  100
nstlist             =  10
ns_type             =  grid
rlist               =  1.0
rvdw                =  1.0
emtol               =  100.0
emstep              =  0.01
--------------------------------------
That's all, but I do not know why it will not converge, and if I run md, if will give bad results.
I am a freshuser of gromacs, If this post abrupt you, I will be very sorry.
thank you.



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20020815/04e9a557/attachment.html>