lammps-reaxff

Name: lammps-reaxff
Author: jinzhezenggroup

113stars23forks

How to install lammps-reaxff?

jinzhezenggroup/computational-chemistry-agent-skills/lammps-reaxff

$npx -y skills add jinzhezenggroup/computational-chemistry-agent-skills --skill lammps-reaxff

Installs into the current project.

Summary

Run reactive molecular dynamics simulations in LAMMPS with the ReaxFF potential, including preparing input scripts (pair_style reaxff + fix qeq/reaxff), mapping LAMMPS atom types to elements via pair_coeff, choosing ensembles (NVE/NVT/NPT), and adding common ReaxFF diagnostics su

Files · 1

View on GitHub

SKILL.md

1	# LAMMPS + ReaxFF
2
3	Use this skill when the user wants to run molecular dynamics in LAMMPS with a ReaxFF force field, prepare or explain an `input.lammps` file, and set up charge equilibration (QEq) correctly.
4
5	## Agent responsibilities
6
7	1. Confirm the ReaxFF force field file (e.g. `ffield.reax.*`). Do not guess which file is appropriate.
8	- If the user does not have a force field yet, point them to known sources (e.g. LAMMPS `potentials/ffield.reax.*` at https://github.com/lammps/lammps/tree/develop/potentials).
9	1. Confirm the structure/data file (e.g. `data.system`) and the atom type → element mapping needed by `pair_coeff`.
10	1. Ensure the input includes charge handling:
11	- Use a charge-capable atom style, such as `atom_style charge` or `atom_style full`, and ensure charges are initialized either from the data file (with a charge column compatible with the chosen `atom_style`) or via explicit commands (e.g. `set` or equal-style variables). Do not rely on `fix property/atom q` as a substitute for a real charge field used by ReaxFF/QEq.
12	- Add one charge equilibration fix, typically `fix qeq/reaxff`, unless the user explicitly requests otherwise.
13	1. Write the LAMMPS input script yourself; keep examples readable and annotated.
14	1. When possible, validate command availability against LAMMPS docs or local `lmp -h` output before execution.
15	1. Report clearly which command was run, which files were used, and where outputs were written.
16
17	## Minimum information to collect
18
19	Ask only for what is missing:
20
21	- LAMMPS data file path (or structure + how to generate a data file)
22	- ReaxFF force field file path (`ffield.reax...`)
23	- Atom types present and their element mapping (for `pair_coeff * * ffield ...`)
24	- Ensemble (NVE / NVT / NPT)
25	- Temperature, pressure (if NPT), timestep, run length
26	- Execution mode: online provisioning vs user-specified LAMMPS binary
27
28	## Execution mode
29
30	### Online mode (only if internet access + `uv` is available)
31
32	Use:
33
34	```bash
35	uvx --from 'lammps[mpi]' lmp -in input.lammps
36	```
37
38	Notes:
39
40	- If you see `error while loading shared libraries: libmpi.so...`, you likely installed an MPI-linked `lmp` without MPI runtime libraries. Prefer `uvx --from 'lammps[mpi]' ...` (bundles MPI runtime), or load/install MPICH/OpenMPI via system packages/conda/HPC module.
41
42	### Offline mode (common / HPC)
43
44	Do not invent the executable. Ask which command should be used, e.g.:
45
46	- `lmp -in input.lammps`
47	- `mpirun -np 32 lmp_mpi -in input.lammps`
48	- `srun lmp -in input.lammps`
49
50	## Example: annotated NVT input (ReaxFF + QEq)
51
52	See also `assets/input.reaxff.nvt.lammps`.
53
54	```lammps
55	# --------- user knobs ---------
56	variable NSTEPS equal 200000
57	variable THERMO equal 200
58	variable DUMP equal 1000
59
60	variable TEMP equal 300.0
61	variable TAU_T equal 100.0
62
63	# Timestep (fs for units real). For high-T / reactive runs, 0.1 fs is often safer.
64	variable DT equal 0.25
65
66
67	# QEq parameters
68	variable QEQ_EVERY equal 1
69	variable QEQ_TOL equal 1.0e-6
70	variable QEQ_CUTLO equal 0.0
71	variable QEQ_CUTHI equal 10.0
72
73	units real
74	boundary p p p
75	atom_style charge
76
77	read_data data.system
78
79	neighbor 2.0 bin
80	neigh_modify every 1 delay 0 check yes
81
82	# ReaxFF potential
83	pair_style reaxff NULL
84	pair_coeff * * ffield.reax C H O
85
86	# Charge equilibration (required for most ReaxFF parameterizations)
87	fix fqeq all qeq/reaxff ${QEQ_EVERY} ${QEQ_CUTLO} ${QEQ_CUTHI} ${QEQ_TOL} reaxff
88	# (`reaxff` here means QEq parameters are extracted from the ReaxFF force field file.)
89
90	# Thermo and trajectory
91	thermo_style custom step temp pe ke etotal press vol density
92	thermo ${THERMO}
93
94	dump 1 all custom ${DUMP} traj.lammpstrj id type q x y z
95
96	# Dynamics
97	velocity all create ${TEMP} 12345 mom yes rot yes dist gaussian
98	fix fnvt all nvt temp ${TEMP} ${TEMP} ${TAU_T}
99
100	timestep ${DT}
101	run ${NSTEPS}
102	```
103
104	### Notes on the example
105
106	- `units real` is a common choice for ReaxFF (time in fs). Many published ReaxFF workflows use `real`, but the correct choice depends on the parameterization and you

People who liked this also viewed

6 picks

Type

lammps-reaxff

113stars23forks

View on GitHub

How to install lammps-reaxff?

jinzhezenggroup/computational-chemistry-agent-skills/lammps-reaxff

$npx -y skills add jinzhezenggroup/computational-chemistry-agent-skills --skill lammps-reaxff

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# LAMMPS + ReaxFF
2
3	Use this skill when the user wants to run molecular dynamics in LAMMPS with a ReaxFF force field, prepare or explain an `input.lammps` file, and set up charge equilibration (QEq) correctly.
4
5	## Agent responsibilities
6
7	1. Confirm the ReaxFF force field file (e.g. `ffield.reax.*`). Do not guess which file is appropriate.
8	- If the user does not have a force field yet, point them to known sources (e.g. LAMMPS `potentials/ffield.reax.*` at https://github.com/lammps/lammps/tree/develop/potentials).
9	1. Confirm the structure/data file (e.g. `data.system`) and the atom type → element mapping needed by `pair_coeff`.
10	1. Ensure the input includes charge handling:
11	- Use a charge-capable atom style, such as `atom_style charge` or `atom_style full`, and ensure charges are initialized either from the data file (with a charge column compatible with the chosen `atom_style`) or via explicit commands (e.g. `set` or equal-style variables). Do not rely on `fix property/atom q` as a substitute for a real charge field used by ReaxFF/QEq.
12	- Add one charge equilibration fix, typically `fix qeq/reaxff`, unless the user explicitly requests otherwise.
13	1. Write the LAMMPS input script yourself; keep examples readable and annotated.
14	1. When possible, validate command availability against LAMMPS docs or local `lmp -h` output before execution.
15	1. Report clearly which command was run, which files were used, and where outputs were written.
16
17	## Minimum information to collect
18
19	Ask only for what is missing:
20
21	- LAMMPS data file path (or structure + how to generate a data file)
22	- ReaxFF force field file path (`ffield.reax...`)
23	- Atom types present and their element mapping (for `pair_coeff * * ffield ...`)
24	- Ensemble (NVE / NVT / NPT)
25	- Temperature, pressure (if NPT), timestep, run length
26	- Execution mode: online provisioning vs user-specified LAMMPS binary
27
28	## Execution mode
29
30	### Online mode (only if internet access + `uv` is available)
31
32	Use:
33
34	```bash
35	uvx --from 'lammps[mpi]' lmp -in input.lammps
36	```
37
38	Notes:
39
40	- If you see `error while loading shared libraries: libmpi.so...`, you likely installed an MPI-linked `lmp` without MPI runtime libraries. Prefer `uvx --from 'lammps[mpi]' ...` (bundles MPI runtime), or load/install MPICH/OpenMPI via system packages/conda/HPC module.
41
42	### Offline mode (common / HPC)
43
44	Do not invent the executable. Ask which command should be used, e.g.:
45
46	- `lmp -in input.lammps`
47	- `mpirun -np 32 lmp_mpi -in input.lammps`
48	- `srun lmp -in input.lammps`
49
50	## Example: annotated NVT input (ReaxFF + QEq)
51
52	See also `assets/input.reaxff.nvt.lammps`.
53
54	```lammps
55	# --------- user knobs ---------
56	variable NSTEPS equal 200000
57	variable THERMO equal 200
58	variable DUMP equal 1000
59
60	variable TEMP equal 300.0
61	variable TAU_T equal 100.0
62
63	# Timestep (fs for units real). For high-T / reactive runs, 0.1 fs is often safer.
64	variable DT equal 0.25
65
66
67	# QEq parameters
68	variable QEQ_EVERY equal 1
69	variable QEQ_TOL equal 1.0e-6
70	variable QEQ_CUTLO equal 0.0
71	variable QEQ_CUTHI equal 10.0
72
73	units real
74	boundary p p p
75	atom_style charge
76
77	read_data data.system
78
79	neighbor 2.0 bin
80	neigh_modify every 1 delay 0 check yes
81
82	# ReaxFF potential
83	pair_style reaxff NULL
84	pair_coeff * * ffield.reax C H O
85
86	# Charge equilibration (required for most ReaxFF parameterizations)
87	fix fqeq all qeq/reaxff ${QEQ_EVERY} ${QEQ_CUTLO} ${QEQ_CUTHI} ${QEQ_TOL} reaxff
88	# (`reaxff` here means QEq parameters are extracted from the ReaxFF force field file.)
89
90	# Thermo and trajectory
91	thermo_style custom step temp pe ke etotal press vol density
92	thermo ${THERMO}
93
94	dump 1 all custom ${DUMP} traj.lammpstrj id type q x y z
95
96	# Dynamics
97	velocity all create ${TEMP} 12345 mom yes rot yes dist gaussian
98	fix fnvt all nvt temp ${TEMP} ${TEMP} ${TAU_T}
99
100	timestep ${DT}
101	run ${NSTEPS}
102	```
103
104	### Notes on the example
105
106	- `units real` is a common choice for ReaxFF (time in fs). Many published ReaxFF workflows use `real`, but the correct choice depends on the parameterization and you

People who liked this also viewed

6 picks

Type