Skip to content

Combustion of 1,5-Dinitrobiuret (DNB) in the Presence of Nitric Acid Using ReaxFF Molecular Dynamics Simulations

Evidence and attribution

Authority of statements

Prose sections below (Summary, Methods, Findings, etc.) are curated summaries of the publication identified by doi, title, and pdf_path in the front matter above. They are not new primary claims by this wiki.

For definitive numerical values, reaction schemes, and interpretations, use the peer-reviewed article (and optional records under normalized/papers/ when present)—not this page alone.

Summary

ReaxFF reactive MD explores DNB/nitric acid mixtures at 0.5 and 1.0 g/mL to connect atomistic pathways to hypergolic ignition chemistry. Additional QM-driven reparameterization targets DNB dissociation channels and the DCA→DNB formation sequence; certain compositions show a sharp exothermic runaway interpreted as spontaneous ignition (abstract; Introduction; computational methods opening, extract). The Introduction situates 1,5-dinitrobiuret (DNB) as a nitrogen-rich energetic intermediate tied to ionic-liquid hypergolic chemistry with nitric-acid-class oxidizers, motivating atomistic study of DNB + HNO\(_3\) mixtures where experiments are sparse.

Methods

Grounding: papers/Russo_DNB_HNO3_JPCA_2013.pdf; normalized/extracts/2013russo-venue-jp403511q_p1-2.txt (abstract + Computational Methods opening).

1 — MD application (ReaxFF reactive mixtures + temperature ramps)

  • Engine / code: Reactive molecular dynamics with ReaxFF; the article notes the final parameter file can be used with the standalone ReaxFF code, LAMMPS, or the ReaxFF module in ADF (Computational Methods excerpt).
  • System size & composition (mixtures): DNB + HNO\(_3\) mixtures are built at ~0.5 g/mL (18 DNB + 18 HNO\(_3\) in a 2.5 nm × 2.5 nm × 2.5 nm periodic box) and ~1.0 g/mL (37 + 37 in the same box edge length), with additional off-stoichiometric examples started in the Methods text (18 DNB / 180 HNO\(_3\) with enlarged box side lengths 3.65 nm and 2.91 nm as printed in the extract) (Computational Methods excerpt).
  • Boundaries / periodicity: Periodic cubic cells are explicitly described for the mixture preparation excerpt (periodic box) (Computational Methods excerpt).
  • Ensemble / thermostat / timestep / duration (mixtures): After minimization, each mixture system is run for 5 ps NVT equilibration at 500 K using a Berendsen thermostat (500 fs damping), followed by NVE production at 500 K with \(\Delta t = 0.1\) fs, typically 500 ps total (some runs extended/shortened depending on reactivity) (papers/Russo_DNB_HNO3_JPCA_2013.pdf, Simulation Details).
  • Single-molecule temperature ramps: 20 simulations with one DNB in a 2.5 nm periodic box, heated with a Berendsen thermostat to 4000 K target, 500 fs damping, 0.1 fs timestep; terminal NO\(_2\) loss is reported as the dominant first dissociation step in all 20 runs (Computational Methods excerpt).
  • Barostat: N/A — not stated for these ramp/mixture excerpts beyond constant-volume cubic cells.
  • Temperature: 4000 K target in ramp tests; mixture staging temperatures are continued beyond p1–2 in the PDF.
  • Pressure: N/A — not a stated hydrostatic control in the indexed excerpt (density targets are used for mixtures).
  • Electric field: N/A.
  • Replica / enhanced sampling: N/A — 20 independent ramp trajectories are used as replicas for dominant first-step statistics, not enhanced sampling in the umbrella/metadynamics sense.

2 — Force-field training (ReaxFF extension for DNB / HNO\(_3\) chemistry)

  • Parent FF / elements: Extends a previously developed C/H/O/N ReaxFF database for hypergolic systems (Computational Methods excerpt).
  • QM reference: Additional training/optimization uses Jaguar QM with B3LYP and 6-311G++** for reaction energies of single-DNB dissociation channels and for the DCA → DNB formation sequence (Computational Methods excerpt).
  • Training set / targets: Adds reaction energies (isolated reactant/product energy differences) for DNB dissociation pathways and tests the DNB formation pathway energetics vs QM (Computational Methods excerpt + abstract motivation).
  • Optimization: Training/optimization language is used in the excerpt; specific optimizer (GA vs least-squares) is not spelled out on p1–2—see full PDF.
  • Reference data / validation: Post-training, ReaxFF matches QM for the highlighted DNB reactions typically within ~6 kcal/mol, with DCA→DNB steps mostly within ~8 kcal/mol except the final step called out as a small stability deviation (Computational Methods excerpt).

Findings

  • Outcomes & mechanisms: For single-DNB ramps at the stated extreme heating, terminal NO\(_2\) loss is the dominant first dissociation step in all 20 trajectories (Computational Methods excerpt). For DNB + HNO\(_3\) mixtures, the abstract reports that certain compositions can show a very sharp thermal-energy release interpreted as spontaneous ignition / hypergolic-like behavior, with mechanistic discussion in the article body.
  • Comparisons: QM comparisons are shown for trained reaction energies (Figures referenced in excerpt text) and the work positions ReaxFF as reproducing near-QM energetics trends at lower cost (Introduction excerpt).
  • Sensitivity / design levers: Mixture composition and mass density (~0.5 vs ~1.0 g/mL) are explicit knobs in the abstract/Methods opening; ramp tests probe temperature-driven fragmentation.
  • Limitations & outlook: The excerpt notes the final DNB-formation step is slightly less stable vs QM but argues it is unlikely to dominate overall dynamics given large energy releases in earlier steps (Computational Methods excerpt). Atomistic cells are not full engine-scale propellant fluid mechanics.
  • Corpus honesty: Indexed pages include mixture protocol start but truncate before full equilibration/production tables; confirm complete MD staging in pdf_path (and compare 2013russo-venue-research galley variant if auditing formatting differences).

Limitations

Shock/ignition chemistry is extremely sensitive to model and initial geometry; high-temperature ramp protocols are not experimental drop-test conditions.

Relevance to group

Demonstrates iterative ReaxFF extension for nitrogen-rich energetic/ionic-liquid chemistry led from Penn State MNE.

Citations and evidence anchors

  • J. Phys. Chem. A 2013, 117, 9216–9223; DOI 10.1021/jp403511q (extract page 2 footer).
  • Abstract and force-field development paragraph (extract pages 1–2).