Skip to content

ReaxFF reactive force-field modeling of the triple-phase boundary in a solid oxide fuel cell

Evidence and attribution

Maintainer note

Maintainer catalog (SI/galley/proof PDF roles): https://github.com/asepehri93/vanDuinWiki/blob/main/docs/corpus/NON_PRIMARY_ARTICLE_PAPER_SLUGS.md

Authority of statements

Summaries follow the J. Phys. Chem. Lett. letter abstract in the corpus extract. The registered PDF is an author proof; verify pagination against the published issue via doi.

Summary

ReaxFF for Ni/YSZ is built by merging existing YSZ and Ni/C/H descriptions. Reactive MD explores H₂ and C₄H₁₀ feeds at Ni/YSZ triple-phase boundary (TPB) conditions (1250 K and 2000 K in the abstract). Simulations report Ni surface amorphization, interfacial delamination, and coking, aligned with experimental observations, and propose a butane conversion mechanism at the Ni/YSZ interface related to pyrolysis-like steps with product sets consistent with experiment.

Methods

Grounding: papers/Merinov_JPC_Letters_YSZ_Ni_interface_proof.pdf (author proof; verify pagination against the published J. Phys. Chem. Lett. issue) and normalized/extracts/2014merinov-venue-research_p1-2.txt.

1 — MD application (atomistic dynamics)

  • Engine / code: ReaxFF reactive molecular dynamics (RMD) as stated in the letter extract; N/A — MD software package name not stated in the indexed excerpt (often LAMMPS in ReaxFF studies—confirm in the PDF/SI).
  • System size & composition (TPB examples): H₂/Ni/YSZ cell excerpted as 240 Ni, 16 Y, 48 Zr, 148 O with YSZ at ~14 mol % Y₂O₃ [(Y₂O₃)(ZrO₂)₆] and 125 H₂ molecules; C₄H₁₀/Ni/YSZ case uses 20 butane molecules (extract).
  • Boundaries / periodicity: Periodic TPB model (“Figure 1 shows the periodic system…”, extract).
  • Ensemble: NPT equilibration at the target temperature, then NVT production MD on the TPB systems (extract).
  • Timestep: N/A — not stated in 2014merinov-venue-research_p1-2.txt.
  • Duration / stages: ~150 ps at 1250 K for the H₂ case (many reactions observed in that window); ~2 ns at 2000 K for the butane case (extract).
  • Thermostat: N/A — not stated in the indexed excerpt.
  • Barostat: NPT used during equilibration (extract); N/A — barostat type/parameters beyond NPT label not in excerpt.
  • Temperature: 1250 K (H₂/Ni/YSZ, near typical SOFC operation) and 2000 K (C₄H₁₀/Ni/YSZ, elevated to accelerate sampling—authors caution when extrapolating to lower temperature, extract).
  • Pressure: N/A — explicit pressure targets not quoted in the indexed excerpt beyond NPT equilibration mention.
  • Electric field: N/A — not stated.
  • Replica / enhanced sampling: N/A — not stated in the indexed excerpt.

2 — Force-field training (Ni/YSZ ReaxFF assembly)

  • Parent FF / elements: Combine published YSZ and Ni/C/H ReaxFF descriptions; define missing angular terms (e.g., Ni–O–Zr) from combination rules; ignore Ni–Zr and Ni–Y bonded interactions; Ni/Y and Ni/Zr nonbonded parameters from combination rules (extract; see Supporting Information pointer in letter).

Findings

Outcomes and mechanisms

The abstract and extract report Ni surface amorphization, interfacial delamination (decohesion), and coking, phenomena said to align with experiment. For butane, the authors derive a conversion mechanism at the Ni/YSZ interface with pyrolysis-like steps and simulated products matching experiment in their framing.

Comparisons

The letter positions ReaxFF against prior QM validation for YSZ (oxygen-ion diffusion) and Ni/hydrocarbon chemistry blocks, and compares simulation outputs to experimental observations for TPB degradation motifs (abstract/extract).

Sensitivity and design levers

Temperature is used as an explicit lever: 2000 K accelerates sampling versus 1250 K operating-relevant conditions; the extract notes care when mapping high-T observations to lower-T devices.

Limitations and corpus honesty

Registered pdf_path is an author proof; cite figure labels and numbers from the version-of-record PDF when possible. Mechanistic detail beyond the abstract/extract should be checked against the full letter and SI.

Limitations

Proof PDF; high-T reactive simulations are force-field dependent; extrapolation to full SOFC stacks requires continuum and microstructural factors not in the letter.

Relevance to group

van Duin coauthored Caltech MSC SOFC interface study—core oxide / Ni / hydrocarbon reactive chemistry corpus entry.

Citations and evidence anchors

  • https://doi.org/10.1021/jz501891y — J. Phys. Chem. Lett. 5, 4039–4043 (2014).