Using CVFF forcefield with Enhanced Monte Carlo

EMC
1

I am trying to use CVFF interface forcefield with EMC. I have tried the following so far.

  1. created cvff folder in fields/
  2. Copy https://github.com/hendrikheinz/INTERFACE-force-field-and-surface-models/blob/584179265906d93aa40f7d5b275143985871b654/cvff6. _interface_v1_5.frc and cvff_interface_v1_5_templates.dat as cvff.frc cvff_templates.frc
  3. My esh file is as follows.
ITEM    OPTIONS

field           cvff
ntotal          4000
mass            true
replace         true
debug           true
field_debug     true

ITEM    END

# Groups

ITEM    GROUPS

mol1            CCN

ITEM    END

# Clusters

ITEM    CLUSTERS

mol1            mol1 100

ITEM    END
  1. emc.pl file.esh (is successful)
  2. emc_linux_x86_64 build.emc fails with the following error,
(* EMC: Enhanced Monte Carlo simulations *)

version 9.4.4, build Jul 31 2025 20:33:44, date Wed Nov  5 20:40:55 2025

valid until Jul 31, 2026

Info: script v1.0 started at Wed Nov  5 20:40:55 2025
Info: variables = {seed -> -1, ntotal -> 4000, fshape -> 1, output ->
        "polymer", field -> "cvff/cvff", location ->
        "~/software/emc_v9.4.4/field/", nav -> 0.6022141179, temperature ->
        300, radius -> 5, nrelax -> 100, weight_nonbond -> 0.0001, weight_bond
        -> 0.0001, weight_focus -> 1, cutoff -> 9.5, charge_cutoff -> 9.5,
        kappa -> 4, density1 -> 1, lprevious -> 0, lphase -> 0, f_mol1 -> 100,
        chem_mol1 -> "CCN"}
Info: output =
      {detail -> 3, wide -> false, expand -> false, math -> true, reduced ->
        false, info -> true, strict -> true, warning -> true, message -> true,
        debug -> false, exit -> true, ignore -> false}
Info: field = {id -> cvff, mode -> cff, style -> none, name ->
        {"~/software/emc_v9.4.4/field/cvff/cvff.frc",
          "~/software/emc_v9.4.4/field/cvff/cvff_templates.dat"}, compress ->
        false, error -> true, debug -> false, check -> {atomistic -> true,
          charge -> true}}
Info: importing '~/software/emc_v9.4.4/field/cvff/cvff.frc'
Error: core/field/cff/parameters.c:383 FieldCffParametersImport:
       Torsion_1 n out of range.
       Program aborted.

I am bit confused by this error as the .frc file has sufficient torsion_1 parameters. It would be very greatful if someone can point me to the right direction.

2

Dear user,

The reported problem occurs in line 1812 of your cvff.frc, where the n of the torsion_1 contribution is 4. A value higher than 3 is not allowed in CFF force fields.

A bigger problem though is that EMC does not support CVFF since CVFF uses a Lennard-Jones 12-6 potential and mix of Morse and harmonic bonds, rather than a LJ 9-6 potential and (higher) harmonic bonds as is the case for Class II force fields. Is there a reason that you need to use CVFF instead of e.g. PCFF?

3

The main purpose in my mind was to use EMC to generate different initial polymer structures with different force fields for forcefield verification purposes. CVFF happens to be one of the force fields of interests. Currently, it appears that only Material Studio can handle CVFF, and I was looking for free academic software alternatives.

I am also looking for CVFF Class-1 alternative nocross-nomorse.

Please let me know if you have any suggestions, or if I can contribute to the source code.

4

I understand. CVFF is not necessarily one of the best force fields. I think Hendrik Heinz has used it, because it is the only LJ 6-12 in Materials Studio (MS), which he uses to set up and type his systems. We use CHARMM in combination with the parameters as provided by the interface force field (IFF) set of parameters. You can find some of them in $EMC_ROOT/field/charmm/iff/. Typically one uses these force fields per targeted surface. Also, note that the Heinz cvff_template.dat does not necessarily include all the rules to type salts and clays. Heinz typically uses pre-typed unit cells.