Skip to content

Defect design of two-dimensional MoS2 structures by using a graphene layer and potato stamp concept

Corpus PDF

The registered file is an ACS proof PDF (Yilmaz_potato_stamp_JPC_2018_proof.pdf). For pagination, final figure numbering, and any publisher corrections, prefer the version-of-record article at the DOI below when available locally.

Summary

The publication develops a conceptual and computational route to engineer sulfur vacancies in monolayer MoS\(_2\) by borrowing a “potato stamp” metaphor from printing: prepare vacancies in graphene, bring that graphene into contact with MoS\(_2\), and allow sulfur to relocate from the MoS\(_2\) surface into graphene vacancies; when the graphene is lifted away, sulfur is removed with it, leaving defective MoS\(_2\) with a designed vacancy population. The work combines nudged elastic band (NEB) and density functional theory (DFT) calculations—implemented with VASP using PAW pseudopotentials, the PBE exchange–correlation functional, and van der Waals corrections as stated in the article—to establish whether S transfer from MoS\(_2\) into graphene monovacancies is thermodynamically favorable relative to alternative placements. The authors then merge MoS\(_2\) and C/H/O ReaxFF parameter subsets and retrain C–S interactions so that ReaxFF molecular dynamics can test whether the stamping idea is dynamically plausible at the atomistic scale. A follow-on demonstration places epoxy species at engineered vacancy sites to probe dissociation chemistry and the exposure of Mo centers that may act in a catalytic role. Adri C. T. van Duin is a coauthor, placing the study in the group’s broader portfolio of 2D materials and reactive force field applications.

Methods

Static QM / NEB (VASP)

DFT uses VASP with PAW potentials, GGA-PBE functional, DFT-D2 dispersion correction, 520 eV cutoff, and 4×4×1 Γ-centered k-point mesh for S/S\(_2\)/S\(_8\) on pristine and monovacancy graphene supercells (72-atom 6×6 sheet excerpt). NEB supplies migration pathways and barriers compared to ReaxFF training targets.

Force-field assembly (ReaxFF)

MoS\(_2\) and C/H/O ReaxFF subsets are merged; C–S cross-terms are reoptimized against the QM binding-energy training set (genetic-algorithm/least-squares workflow per article).

Reactive MD (LAMMPS)

  • Engine / code: LAMMPS molecular dynamics with merged ReaxFF.
  • System size & composition: MoS\(_2\) / graphene interface supercells plus epoxy-functionalized vacancy demonstrations (atom counts in Methods/figures).
  • Boundaries / periodicity: In-plane PBC with vacuum spacing for 2D stacks.
  • Ensemble: NVT canonical trajectories for stamping/dynamics tests unless NVE/NPT substeps are documented—confirm in PDF.
  • Timestep: Femtosecond timestep per Methods.
  • Duration: Equilibration and production segments in ps/ns as tabulated.
  • Thermostat: Nose–Hoover or equivalent thermostat parameters in Methods.
  • Barostat: N/A — hydrostatic barostat not indicated for the cited NVT 2D dynamics unless SI adds NPT—verify locally.
  • Temperature: K setpoints for dynamical tests.
  • Pressure: N/A — no bulk hydrostatic pressure control for the highlighted NVT runs.
  • Electric field: N/A — electric field not applied.
  • Enhanced sampling: N/A — umbrella / metadynamics not indicated.

Findings

The DFT/NEB analysis supports the idea that S atoms can favor relocation from the MoS\(_2\) surface into graphene vacancies, which underpins the proposed mechanism for using graphene as a removable template to pattern sulfur vacancies in MoS\(_2\). ReaxFF MD results are presented as atomistic support for the potato stamp concept, bridging thermodynamic favorability with a dynamical picture of how S can be extracted when graphene is separated. The epoxy studies report dissociation at engineered vacancies and emphasize Mo centers that become accessible for catalytic behavior, connecting defect engineering to subsequent surface chemistry. Taken together, the article positions combined DFT + ReaxFF workflows as practical tools for 2D defect design where both barriers and reactive dynamics matter.

Limitations

The corpus proof PDF may differ in minor formatting from the version of record; numerical barriers, supplementary trajectory details, and exact kinetic claims should be verified against the final article. ReaxFF accuracy for organometallic details and long-timescale defect annealing remains force-field dependent, and the epoxy demonstration is a targeted case study rather than a general synthesis protocol.

Relevance to group

This entry documents van Duin-group participation in 2D defect engineering for TMD systems, combining QM validation with ReaxFF exploration of reactive pathways.

Citations and evidence anchors

  • DOI: 10.1021/acs.jpcc.8b02991.