ERROR: Molecule auto special bond generation overflow (src/molecule.cpp:1542)

LAMMPS LAMMPS Beginners
1

Dear All
I am trying to simulate the Ethylene carbonate molecules using LAMMPS
and I got the error Molecule auto special bond generation overflow.

I have taken the Amber force field file of ethylene carbonate from the literature
For generate the coordinates of molecule structure of Single EC, I used online link DIY-molecules: build your own molecule and below this I putted bond, angle, dihedrals manually
please help me to shortout this problem.

I have shared Files

Input

boundary p p p
units real
atom_style full
bond_style harmonic
angle_style harmonic
dihedral_style harmonic
pair_style lj/cut/coul/long 12
kspace_style pppm 1e-4
special_bonds lj 0.0 1.0 1.0 coul 0.0 0.0 1.0 angle yes dihedral yes

System defination

lattice fcc 1.0
region simbox block -33 33 -17 17 -17 17
create_box 8 simbox bond/types 6 angle/types 8 dihedral/types 8 &
extra/bond/per/atom 5 extra/angle/per/atom 9 extra/special/per/atom 1

include PARM_EC.lammps

molecule ECmol C2H4CO3.txt

create_atoms 0 random 674 4561567 simbox mol ECmol 1475634 overlap 1 maxtry 50

group C2H4CO3 type 1 2 3 4 5
delete_atoms overlap 0.5 C2H4CO3 C2H4CO3 mol yes

minimize 1.0e-4 1.0e-6 100 1000
reset_timestep 0

Equlibiration

fix mynpt all npt temp 300 300 100 iso 1 1 1000

dump mydmp all atom 1000 dump.lammpstrj

variable mytemp equal temp
variable myvol equal vol
fix myat1 all ave/time 10 10 100 v_mytemp file temperature.dat
fix myat2 all ave/time 10 10 100 v_myvol file volume.dat
variable myoxy equal count(C2H4CO3)/10
variable mydensity equal ${myoxy}/v_myvol
fix myat3 all ave/time 10 10 100 v_mydensity file density.dat

thermo 1000
timestep 1.0
run 50000
write_data EC.data

---------Molecules file-------

Ethylenecarbonate. geometry

10 atoms
10 bonds
17 angles
8 dihedrals

Coords

1 1.13260 3.55440 0.00000 # Ox
2 1.13260 2.15440 0.00000 # Cx
3 0.00000 1.33150 0.00000 # OS1
4 2.26520 1.33150 0.00000 # OS2
5 0.43260 0.00000 0.00000 # Ch1
6 1.83260 0.00000 0.30000 # Ch2
7 0.21060 -0.50951 -0.94927 # H
8 -0.14271 -0.62429 0.69948 # H
9 2.05136 -0.30109 1.33513 # H
10 2.39162 -0.76940 -0.25274 # H

Types

1 1 # Ox
2 2 # Cx
3 3 # OS1
4 3 # OS2
5 4 # Ch1
6 4 # Ch2
7 5 # H1
8 5 # H2
9 5 # H3
10 5 # H4

Charges

1 -0.6452 # Ox
2 1.0996 # Cx
3 -0.4684 # OS1
4 -0.4684 # OS2
5 0.0330 # Ch1
6 0.0330 # Ch2
7 0.1041 # H1
8 0.1041 # H2
9 0.1041 # H3
10 0.1041 # H4

Bonds

1 1 1 2 # Ox-Cx
2 2 2 3 # Cx-OS1
3 2 2 4 # Cx-OS2
4 3 3 5 # OS1-Ch1
5 3 4 6 # OS2-Ch2
6 4 5 6 # Ch1-Ch2
7 5 5 7 # Ch1-H1
8 5 5 8 # Ch1-H2
9 5 6 9 # Ch2-H3
10 5 6 10 # Ch2-H4

Angles

1 1 1 2 3 # Ox-Cx-OS1
2 1 1 2 4 # Ox-Cx-OS2
3 2 2 3 5 # Cx-OS1-Ch1
4 2 2 4 6 # Cx-OS2-Ch2
5 3 5 6 4 # Ch2-Ch1-OS1
6 3 6 5 3 # Ch1-Ch2-OS2
7 4 3 2 4 # OS1-Cx-OS2
8 5 3 5 7
9 5 3 5 8
10 5 4 6 9
11 5 4 6 10
12 6 6 5 7
13 6 6 5 8
14 6 5 6 9
15 6 5 6 10
16 7 7 5 8
17 7 9 6 10

Dihedrals

1 1 1 2 3 5
2 2 1 2 4 6
3 3 3 5 6 4
4 4 4 6 5 3
5 5 3 5 6 8
6 6 5 6 4 2
7 7 6 5 3 2
8 8 3 2 4 6

------Force Field------

LAMMPS input parameter script

mass 1 15.9994 # EC Ox
mass 2 12.011 # EC Cx
mass 3 15.9994 # EC Os
mass 4 12.011 # EC Ch
mass 5 1.008 # EC H
mass 6 6.941 # LiPF6 Li
mass 7 30.973762 # LiPF6 P
mass 8 18.998403 # LiPF6 F

pair_coeff 1 1 0.210 2.96 # EC Ox
pair_coeff 2 2 0.105 3.75 # EC Cx
pair_coeff 3 3 0.170 3.00 # EC Os
pair_coeff 4 4 0.066 3.50 # EC Ch
pair_coeff 5 5 0.030 2.50 # EC H
pair_coeff 6 6 0.10314 1.4424 # LiPF6 Li
pair_coeff 7 7 0.13169 3.695 # LiPF6 P
pair_coeff 8 8 0.028716 2.9347 # LiPF6 F

bond_coeff 1 504.4 1.193 # Ox-Cx
bond_coeff 2 401.8 1.360 # Cx-Os
bond_coeff 3 300.5 1.430 # Os-Ch
bond_coeff 4 307.6 1.520 # Ch-Ch
bond_coeff 5 340.4 1.090 # H-Ch
bond_coeff 6 370.8 1.606 # F-P

angle_coeff 1 75.4 124.970 # Ox-Cx-Os
angle_coeff 2 62.0 108.140 # Cx-Os-Ch
angle_coeff 3 67.8 102.050 # Ch-Ch-Os
angle_coeff 4 72.4 110.200 # Os-Cx-Os
angle_coeff 5 50.8 108.600 # Os-Ch-H
angle_coeff 6 46.5 114.030 # Ch-Ch-H
angle_coeff 7 39.3 110.610 # H-Ch-H
angle_coeff 8 139.40 90.00 # F-P-F

dihedral_coeff 1 1.400 1 1 # Ox-Cx-Os-Ch 180.0
dihedral_coeff 2 2.700 1 2 # ox-cx-os-ch 180.0
dihedral_coeff 3 0.144 1 3 # os-ch-ch-os 0.000
dihedral_coeff 4 1.175 1 2 # os-ch-ch-os 0.000
dihedral_coeff 5 0.250 1 1 # os-ch-ch-h 0.000
dihedral_coeff 6 0.800 1 1 # ch-ch-os-cx 180.0
dihedral_coeff 7 0.383 1 3 # ch-ch-os-cx 0.000
dihedral_coeff 8 2.700 1 2 # os-cx-os-ch 180.0

2

The error message is pretty much self-explanatory. Just look at the command I quoted above.

3

I am using dihedral of ethylene carbonate of Amber force field from Masia et. al. 2004 paper.
there are total 7 dihedral type, since some of the same dihedral type having 2 coefficient which is consider as sum of dihedral potential equal to that type of dihedral let say V=V1,n1,phi1 + V2, n2, phi2
when i used dihedral style charmm which doesn’t allow the summation terms.

Dihedral Coeffs charmm

1 0.800 1 180 0 #C2-C2-O2-C1
1 0.383 3 0 0 #C2-C2-O2-C1
2 0.157 3 0 0 #H1-C2-C2-H1
3 0.250 1 0 0 #H1-C2-C2-O2
3 0.000 3 0 0 #H1-C2-C2-O2
4 0.385 3 0 0 #H1-C2-O2-C1
5 1.400 1 180 0 #O1-C1-O2-C2
5 2.700 2 180 0 #O1-C1-O2-C2
6 2.700 2 180 0 #O2-C1-O2-C2
7 1.175 2 0 0 #O2-C2-C2-O2
7 0.144 3 0 0 #O2-C2-C2-O2

it doesnt work and shown error

But When, I have used Fourier style which works well and simulation is running

Dihedral Coeffs #fourier

1 2 0.800 1 180 0.383 3 0 #C2-C2-O2-C1
2 1 0.157 3 0 #H1-C2-C2-H1
3 2 0.250 1 0 0.000 3 0 #H1-C2-C2-O2
4 1 0.385 3 0 #H1-C2-O2-C1
5 2 1.40 1 180 2.70 2 180 #O1-C1-O2-C2
6 1 2.700 2 180 #O2-C1-O2-C2
7 2 0.144 3 0 1.175 2 0 #O2-C2-C2-O2

this is correct way to run the simulation or conversion ?,

4

But you can just use two different dihedral types and define the same dihedral twice once for each type. The result will be the sum of the two. LAMMPS does not check whether each dihedral definition is unique.

You can easily confirm this by setting up some simple tests.

5

Currently I am using PACKMOL and VMD software to generate molecular structure of 1000 EC molecules for LAMMPS input file.
command: when i write topo guessdihedrals on VMD command terminal it writes 7 type of dihedrals but when i gave individual different name of EC atoms(Ox,Os1, Os2, Cx,Ch2,Ch1,H1,H2,H3,H4 ) then total 19 dihedral printed.
but again problem is how we write two dihedral coefficient of same dihedral type
1 1.40 1 180 # Ox-Cx-OS1-Ch1 one more coefficient 2.70 2 180
2 1.40 1 180 # Ox-Cx-OS2-Ch2 one more coefficient 2.70 2 180
3 0.800 1 180 # Cx-OS1-Ch1-Ch2 one more coefficient 0.383 3 0
4 0.385 3 0 # Cx-OS1-Ch1-H1
5 0.385 3 0 # Cx-OS1-Ch1-H2
6 0.800 1 180 # Cx-OS2-Ch2-Ch1 one more coefficient 0.383 3 0
7 0.385 3 0 # Cx-OS2-Ch2-H3
8 0.385 3 0 # Cx-OS2-Ch2-H4
9 2.700 2 180 # OS1-Cx-OS2-Ch2
10 0.144 3 0 # OS1-Ch1-Ch2-OS2 one more coefficient 1.175 2 0
11 0.250 1 0 # OS1-Ch1-Ch2-H3 one more coefficient 0.000 3 0
12 5 0.250 1 0 # OS1-Ch1-Ch2-H4 one more coefficient 0.000 3 0
13 2.700 2 180 # OS2-Cx-OS1-Ch1
14 0.250 1 0 # H1-Ch1-Ch2-OS2 one more coefficient 0.000 3 0
15 0.157 3 0 # H1-Ch1-Ch2-H3
16 0.157 3 0 # H1-Ch1-Ch2-H4
17 0.250 1 0 # H2-Ch1-Ch2-OS4 one more coefficient 0.000 3 0
18 0.157 3 0 # H2-Ch1-Ch2-H3
19 0.157 3 0 # H2-Ch1-Ch2-H4

and one more thing Dihedral Fourier style is same as charmm style except one difference is summation so (can i use Fourier style ?)

6

You need, of course, additional processing of then data. This can be done in VMD with additional scripting (TopoTools has a command to retrieve the list of dihedrals and a command to add dihedrals. So you just the former, loop over the list, and for those that match the type with two definitions, you add a new dihedral with a correspondingly different (symbolic) dihedral type. When exporting to LAMMPS data format, it will be converted to numeric types as needed).

I already gave you a response on that.