OK... so I got around the "Unresolved atom types" error by generating a custom amber.par:
# !!!NOTE: JUNK DATA!!!
This is the AMBER96 user defined parameter file for NWChem 3.2 and ARGOS 7.0
Electrostatic 1-4 scaling factor 0.833333
Relative dielectric constant 1.000000
Parameters epsilon R*
Atoms
BR 79.90000 0.00000E+00 0.00000E+00 1 1111111111
Q 35 0.00000E+00 0.00000E+00
OQ 16.00000 8.80313E-01 1.72100E-01 1 1111111111
Q 8 4.40157E-01 1.72100E-01
HQ 1.00800 0.00000E+00 0.00000E+00 1 1111111111
Q 1 0.00000E+00 0.00000E+00
CQ 12.01000 4.57729E-01 1.90800E-01 1 1111111111
Q 6 2.28864E-01 1.90800E-01
Cross
Bonds
BR -CQ 0.224041 0.0
CQ -CQ 0.15260 0.0
CQ -HQ 0.10900 0.0
OQ -HQ 0.103079 0.0
Angles
BR -CQ -HQ 1.78568 0.0
BR -CQ -CQ 1.89522 0.0
Proper dihedrals
Atom types
BR 35 0 0 0 0 0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
OQ 8 0 0 0 0 0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
HQ 1 0 0 0 0 0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
CQ 6 0 0 0 0 0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
End
#
This at least let me start the calculation, although setting the VdW parameters for Br to zero is probably the worst approximation possible. I also set all the bond and angle spring constants to zero and didn't go out of my way to make sure that the prepare module figures out my bonding properly (it has trouble with the C-Br bond, FWIW). Is this a safe thing to do?
--craig
|