The Configuration File¶

The complete 3-pacm-dgeba-epoxy-thermoset.yaml from the depot:

# htpolynet -- Example 3 -- PACM/DGEBA Epoxy Thermoset
#
# Self-contained config: DGE (DGEBA, the active form of bisphenol-A diglycidyl
# ether) and PAC (PACM, 4,4'-methylenebis(cyclohexylamine)) are generated from
# SMILES via the RDKit atom-mapping path.  Run with:
#
#   htpolynet run -diag diagnostics.log 3-pacm-dgeba-epoxy-thermoset.yaml
#
# Cameron F. Abrams -- cfa22@drexel.edu

Title: DGEBA-PACM thermoset
gromacs:
  gmx: gmx
  gmx_options: -quiet -nobackup
  mdrun: gmx mdrun
  mdrun_options:
    gpu_id: 0            # auto-stripped at runtime if gmx is not GPU-capable or no GPU is visible
ambertools:
  charge_method: gas
constituents:
  DGE:
    smiles: "CC(C)(c1ccc(OC[CH:3]([OH:5])[CH3:1])cc1)c1ccc(OC[CH:4]([OH:6])[CH3:2])cc1"
    reactive_atoms: {1: C1, 2: C2, 3: C3, 4: C4, 5: O1, 6: O2}
    count: 200
    symmetry_equivalent_atoms: [[C1, C2], [C3, C4], [O1, O2]]
    stereocenters: [C3]
  PAC:
    # Both amine carbons named C1 and both nitrogens named N1 -- the
    # symmetry_equivalent_atoms entry below tells htpolynet to treat the
    # second copies as C2 / N2.
    smiles: "C1C[CH:1](CCC1CC2CC[CH:2](CC2)[NH2:3])[NH2:4]"
    reactive_atoms: {1: C1, 2: C1, 3: N1, 4: N1}
    count: 100
    symmetry_equivalent_atoms: [[N1, N2], [C1, C2]]
    stereocenters: [C1]
densification:
  initial_density: 300.0  # kg/m3
  equilibration:
    - ensemble: min
    - ensemble: nvt
      temperature: 300
      ps: 10
    - ensemble: npt
      temperature: 300
      pressure: 10
      ps: 300
precure:
  preequilibration:
    ensemble: npt
    temperature: 300        # K
    pressure: 1             # bar
    ps: 200
  anneal:
    ncycles: 2
    initial_temperature: 300
    cycle_segments:
      - T: 300
        ps: 0
      - T: 600
        ps: 20
      - T: 600
        ps: 20
      - T: 300
        ps: 20
      - T: 300
        ps: 20
  postequilibration:
    ensemble: npt
    temperature: 300        # K
    pressure: 1             # bar
    ps: 100
CURE:
  controls:
    search_radius: 0.5  # nm
    radial_increment: 0.25      # nm
    max_iterations: 150
    desired_conversion: 0.95
    late_threshold: 0.85
  drag:
    trigger_distance: 0.6   # nm
    increment: 0.08         # nm
    limit: 0.3              # nm
    equilibration:
      - ensemble: min
      - ensemble: nvt
        temperature: 600
        nsteps: 1000
      - ensemble: npt
        temperature: 600
        pressure: 1
        nsteps: 2000
  relax:
    increment: 0.08         # nm
    equilibration:
      - ensemble: min
      - ensemble: nvt
        temperature: 600
        nsteps: 1000
      - ensemble: npt
        temperature: 600
        pressure: 1
        nsteps: 2000
  equilibrate:
    ensemble: npt
    temperature: 300       # K
    pressure: 1            # bar
    ps: 100
  gromacs:
    rdefault: 0.9          # nm
postcure:
  anneal:
    ncycles: 2
    initial_temperature: 300
    cycle_segments:
      - T: 300
        ps: 0
      - T: 600
        ps: 20
      - T: 600
        ps: 20
      - T: 300
        ps: 20
      - T: 300
        ps: 20
  postequilibration:
    ensemble: npt
    temperature: 300       # K
    pressure: 1            # bar
    ps: 100
reactions:
  - name:        'Primary-to-secondary-amine'
    stage:       cure
    reactants:
      1: PAC
      2: DGE
    product:     PAC~N1-C1~DGE
    probability: 1.0
    atoms:
      A:
        reactant: 1
        resid: 1
        atom: N1
        z: 2
      B:
        reactant: 2
        resid: 1
        atom: C1
        z: 1
    bonds:
      - atoms: [A, B]
        order: 1
  - name:        'Secondary-to-tertiary-amine'
    reactants:
      1: PAC~N1-C1~DGE
      2: DGE
    product:     PAC~N1-C1~DGE-C1~DGE
    stage:       cure
    probability: 0.5
    atoms:
      A:
        reactant: 1
        resid: 1
        atom: N1
        z: 1
      B:
        reactant: 2
        resid: 1
        atom: C1
        z: 1
    bonds:
      - atoms: [A, B]
        order: 1
  - name:        'Oxirane-formation'
    reactants:
      1: DGE
    product:     DGEC
    stage:       cap
    probability: 1.0
    atoms:
      A:
        reactant: 1
        resid: 1
        atom: O1
        z: 1
      B:
        reactant: 1
        resid: 1
        atom: C1
        z: 1
    bonds:
      - atoms: [A, B]
        order: 1

The Title, gromacs, ambertools, densification, precure, CURE, and postcure blocks follow the same structure as example 1. The two interesting items here are constituents (specifically the symmetry_equivalent_atoms and stereocenters keys) and what those mean for the template set htpolynet ends up generating.

`constituents`¶

constituents:
  DGE:
    smiles: "CC(C)(c1ccc(OC[CH:3]([OH:5])[CH3:1])cc1)c1ccc(OC[CH:4]([OH:6])[CH3:2])cc1"
    reactive_atoms: {1: C1, 2: C2, 3: C3, 4: C4, 5: O1, 6: O2}
    count: 200
    symmetry_equivalent_atoms: [[C1, C2], [C3, C4], [O1, O2]]
    stereocenters: [C3]
  PAC:
    smiles: "C1C[CH:1](CCC1CC2CC[CH:2](CC2)[NH2:3])[NH2:4]"
    reactive_atoms: {1: C1, 2: C1, 3: N1, 4: N1}
    count: 100
    symmetry_equivalent_atoms: [[N1, N2], [C1, C2]]
    stereocenters: [C1]

A 2:1 stoichiometric ratio (200 DGE : 100 PAC) gives at most 400 C–N bonds at full conversion.

Reaction expansion via `symmetry_equivalent_atoms`¶

DGEBA has two chemically equivalent reactive carbons (C1, C2), two equivalent stereo carbons (C3, C4), and two equivalent oxirane oxygens (O1, O2); PACM has two equivalent amine nitrogens (N1, N2) and the two carbons they’re attached to (C1, C2). Declaring these pairings via symmetry_equivalent_atoms tells htpolynet to expand each user-supplied reaction into the full set of symmetry-equivalent variants. The two reactions in this YAML expand as follows:

The single primary-to-secondary reaction PAC.N1 + DGE.C1 expands into four: PAC.{N1,N2} × DGE.{C1,C2}, with products PAC~N1-C1~DGE, PAC~N1-C2~DGE, PAC~N2-C1~DGE, PAC~N2-C2~DGE.
Because each of those products is itself a reactant in the secondary-to-tertiary reaction, that reaction expands into eight variants — one per (primary-stage product) × (which DGEBA carbon reacts next). See the diagnostic log at the start of a run for the full list.

So the user wrote 2 cure reactions and 1 cap reaction in the YAML, and htpolynet produces 12 cure templates + 2 cap templates (one per DGEBA carbon).

Diastereomer expansion via `stereocenters`¶

stereocenters: [C3] (for DGE) and stereocenters: [C1] (for PAC) look like a single chirality declaration each — but combined with the symmetry-equivalence above, they expand:

On DGEBA, declaring C3 as chiral implicitly makes C4 chiral too (they’re symmetry-equivalent). C1/C2 are not chiral in the active form (each has two methyl ligands), but they become chiral after one of those methyls forms a C–N bond — so we treat them as potential stereocenters as well, giving 4 stereocenters per DGEBA → 16 diastereomers.
On PACM, C1 chirality plus the C1↔C2 symmetry pairing gives 2 stereocenters → 4 diastereomers.

htpolynet builds the entire diastereomer pool at startup by flipping stereocenters, then samples uniformly from the pool when placing molecules into the initial liquid. This guarantees the starting state is racemic; polymerization cannot then introduce a net handedness as an artefact.

This dance — 2 user reactions → 12 cure templates + 2 cap templates, and 1 + 1 stereocenter declarations → 16 + 4 = 20 diastereomers — explains why the diagnostic log for this example reports 22 molecules detected (5 explicit + 6 implied by stereochemistry + 11 implied by symmetry).

Now we can turn to actually running the build.

The Configuration File¶

constituents¶

Reaction expansion via symmetry_equivalent_atoms¶

Diastereomer expansion via stereocenters¶

`constituents`¶

Reaction expansion via `symmetry_equivalent_atoms`¶

Diastereomer expansion via `stereocenters`¶