Skip to content

xmolout + fort.7: Multi-file Coordination and Property Analysis

In the previous tutorial, we analyzed a single ReaxFF output file (xmolout) and learned how ReaxKit processes trajectories using handlers, analyzers, and workflows.

In this tutorial, we move to the first multi-file workflow: combining geometric information from xmolout with bond- and atom-level properties from fort.7.

This pattern is extremely common in ReaxFF analysis and underpins workflows such as: - coordination analysis, - spatial charge distribution, - bond-order–resolved visualization, - structure–property correlations.

Why do we need both xmolout and fort.7?

ReaxFF outputs intentionally separate information:

xmolout

Contains geometry and structure: - atomic coordinates (x, y, z), - atom types, - box dimensions, - iteration numbers.

It answers:

Where are the trajectories of atoms?

fort.7

Contains chemical and electronic information: - partial charges, - bond orders, - coordination-related quantities, - atom-resolved properties.

It answers:

What are the atoms doing chemically across frames?

So, ReaxKit’s job is to align these two views of the same simulation in a safe, transparent way.

Re-running a multi-file CLI example

A typical command using both files looks like:

reaxkit xmolfort7 plot3d --property charge --frames 0:20:10

This command:

  • reads geometry from xmolout,
  • reads per-atom charge from fort.7,
  • aligns them frame-by-frame and atom-by-atom,
  • produces a 3D scatter plot colored by charge.

How ReaxKit aligns the two files

Frame alignment

  • xmolout defines the master frame index.

  • fort.7 data are queried using the same frame indices.

  • If a frame is missing in one file, it is skipped safely.

Internally, ReaxKit always works in frame space first, then derives iteration or time if requested.

Atom alignment

  • Atom indices are 0-based internally.

  • xmolout provides atom order and coordinates.

  • fort.7 provides values keyed by atom index.

  • ReaxKit matches these explicitly before analysis.

This prevents silent misalignment (a common source of errors in manual scripts).

Explaining the main CLI flags

--property charge specifies which per-atom scalar from fort.7 to use.

Important features:

  • Alias-aware (charge, q, partial_charge all work).
  • Validated against available columns.
  • Canonicalized internally.

This allows you to focus on meaning, not file-specific naming quirks.

--frames 0:20:10 selects which frames to analyze. This slice means:

  • start at frame 0,
  • stop at frame 20,
  • step by 10.

Resulting frames: 0, 10, 20.

This selection applies consistently to both files.

--atoms (optional) allows spatial sub-selection:

  • specific atoms,
  • atom subsets,
  • or all atoms (default).

This is especially useful for:

  • focusing on active sites,
  • analyzing coordination around specific species.

From coordination concepts to plots

Although this workflow does not explicitly compute “coordination numbers,” it provides the building blocks for them:

geometry from xmolout,

bond-order or charge information from fort.7,

spatial projections (3D scatter, 2D heatmaps).

From here, coordination analysis typically involves:

distance cutoffs,

bond-order thresholds,

neighbor counting per atom or per species.

ReaxKit exposes these steps as analyzers so they can be reused across workflows.

Plotting charges against atomic coordinates (sptial distribution of charges)

By getting:

  • geometry from xmolout,
  • bond-order or charge information from fort.7,

spatial projections of charges are possible (3D scatter, 2D heatmaps).

Output locations

As with all workflows, outputs are organized automatically. By default:

reaxkit_outputs/xmol_fort7/3D_scatter/

or

reaxkit_outputs/xmol_fort7/2D_heatmap/

are the output locations for the results.

Each plot:

  • is named by property and frame,
  • is reproducible,
  • can be regenerated with the same command.

This structure is especially helpful when exploring many frames or properties.

  • Always let one file define the frame index (xmolout in here).
  • Use alias names (charge, q) instead of raw column names.
  • Start with sparse frame sampling, then refine.
  • Export intermediate tables before heavy visualization.
  • Treat geometry and chemistry as separate layers.

What you can do next

With xmolout + fort.7 workflows, you can now:

  • compute coordination numbers explicitly,
  • analyze charge transfer vs geometry,
  • visualize bond-order networks,
  • project properties onto 2D planes,
  • combine multiple simulations consistently.

As an example, you may use:

reaxkit video make --folder reaxkit_outputs/xmol_fort7/3D_scatter

to make a gif video out of all frames.

This shows that multiple workflows can be sequenced to:

  1. set-up simulations by generating input files such as control,
  2. execute batch simulations,
  3. perform batch analysis,
  4. generate useful information comparing different simulations with different settings