[gmx-users] SiO2 simulation again.

[Lianqing Zheng]

Dear All:

I am trying to simulate an all-atom (no bonded interaction) system of SiO2
with Coulomb and Buckingham potentials. The following are the .top and
.gro files that I constructed. Do you know if there is any error or flaw
in them? When I ran mdrun, I got an error of "Arithmetic Exception (core
dumped)." Though it could be many reasons, I don't know if I made the
format of top and gro files right with the system. Could you take a
minute to check them?

;topology file
[ defaults ]
2 1
[ atomtypes ]
;name        mass      charge   ptype            c6           c12
    O    15.99940       -1.20        A   	0.0 0.0 0.0
   SI    28.08000        2.40        A   	0.0 0.0 0.0
[ nonbond_params ]
  ; i    j func           a             b	c6
    O    O    2 1.339973E+5	     27.6	1.688507E-2
    O	SI    2 1.737112E+6	  48.7318	1.288457E-2
   SI	SI    2		0.0	      0.0	0.0
[ moleculetype ]
SI 0
[ atoms ]
1	SI	1	SI	SI	1	2.4	28.08000
[ moleculetype ]
O 0
[ atoms ]
1	O	1	O	O	2	-1.2	15.99940
[ system ]
Silica SiO2
[ molecules ]
SI 512
O 1024
; End of top file
__________________________
gro file:
1536
    1    O    O    1  -0.717  -1.254  -1.254 -1.3797 -0.5947 -1.0028
    1   SI   SI    2  -1.344  -1.344  -1.344  0.7731 -0.8680  0.4347
    2   SI   SI    3  -0.985  -0.985  -1.344  0.1876  1.1196  1.3407
    2    O    O    4  -1.075  -0.896  -1.254  2.4244  1.3673  3.6743
    3    O    O    5  -0.896  -1.075  -1.254 -1.3595 -1.2375 -1.6633
    3   SI   SI    6  -0.985  -1.344  -0.985  1.6699 -1.3636 -0.0800
    4    O    O    7  -1.075  -1.254  -0.896  2.3342 -3.2442  0.3516
....
 1023    O    O 1534   1.254   1.254   1.433  0.0548  0.6022  4.3814
  512   SI   SI 1535   1.344   1.344   1.344 -1.3755  1.1702 -0.9112
 1024    O    O 1536   1.433   1.433   1.433 -2.6078 -1.6520  0.2689
2.866400 2.866400 2.866400
(End of gro file)

When using grompp to get .tpr file, I got a warning:

Warning: atom names in fs.top and fs.gro don't match (SI - O)
(more than 20 non-matching atom names)
Warning 1 [file "fs.top", line 32]:
  682 non-matching atom names
  atom names from fs.top will be used
  atom names from fs.gro will be ingnored

Actually, for the most simple all-atom systems, such as liquid argon, how
do we deal with those directives such as [molecules]? I think these two
systems are similar in the view of file format.

I'd appreciate for your great help!

Lianqing