Skip to content

Atomistic-scale simulations of defect formation in graphene under noble gas ion irradiation

Summary

This corpus PDF is an ACS “Just Accepted” posting of Yoon et al., ACS Nano, DOI 10.1021/acsnano.6b03036. The work pairs ReaxFF molecular dynamics in LAMMPS with helium ion microscopy and aberration-corrected STEM to relate noble-gas ion irradiation of graphene—including post-impact annealing—to defect statistics and nanopore formation. The paper stresses how dose, ion species, and collision cross section steer vacancy-type damage toward coalesced pores, and how Stone–Thrower–Wales (STW) motifs dominate under He\(^+\) whereas monovacancy (MV)-rich statistics appear for heavier noble gases. Electronic stopping is omitted in the nuclear-collision treatment, following cited precedent for graphene.

Methods

MD (ReaxFF, LAMMPS). Simulations use ReaxFF C-2013-based carbon chemistry with graphene–ion repulsion augmented by DFT-fitted terms and a Ziegler–Biersack–Littmark channel as described in Computational methods and details (papers/Yoon_ACSNano_ASAP.pdf); SI holds parameter tables and DFT settings (full functional / basis / k-mesh tables are not transcribed on this page). A periodic graphene supercell about 52 × 40 Ų (thousands of carbon atoms) receives 25 keV He\(^+\), Ne\(^+\), Ar\(^+\), Kr\(^+\) impacts in a central ~30 × 20 Ų window. In-plane PBC apply; edge regions are held at 300 K with a Nosé–Hoover thermostat while the cascade core runs NVE during ballistic impact. Species-dependent timesteps of 0.005–0.02 fs stabilize the cascades. Dose rates of order 10²⁵–10²⁷ ions cm⁻² s⁻¹ (species-dependent) space impacts so each cascade completes before the next. He\(^+\) ladders reach 10¹⁵–10¹⁷ ions cm⁻² cumulative dose, each followed by 1500 K for 25 ps, cool-down to 300 K, then 2000 K for 1.25 ns of reconstruction. Ne\(^+\)/Ar\(^+\)/Kr\(^+\) use 10¹⁴–2×10¹⁵ ions cm⁻², the same 1500 K / 25 ps leg, then 3000 K annealing; the duration of the 3000 K segment is not recovered from the Methods text checked for this note. Barostat, applied electric fields, and replica / enhanced sampling are not used on these fixed-area irradiation legs (cascades plus thermal annealing only). Hydrostatic pressure is not servo-controlled during those legs (fixed in-plane supercell area; N/A — no NPT pressure target in the summarized irradiation protocol).

Force-field training: N/A — not a new ReaxFF fit in the main article; the study applies published C-2013 chemistry with documented DFT/ZBL extensions in SI.

Experiment. HIM (Zeiss ORION NanoFab, ~25–27 kV, 0.191 pA, 0° on suspended CVD graphene) and Nion UltraSTEM MAADF-STEM at 60 kV follow the Experimental methods and details block in the same PDF.

Findings

Post-irradiation annealing drives vacancy-type defects to coalesce into nanopores; heavier ions and higher dose produce larger pores and more amorphized surroundings, tracking STEM trends though some Ne\(^+\) sim/expt pairs differ in dose/energy matching. Defect surveys report ~65% STW character for He\(^+\)-dominated statistics versus ~73% MV prevalence for Ne\(^+\)/Ar\(^+\)/Kr\(^+\) in the conclusions as summarized in the article. Simulated sputtering / removal metrics at 25 keV (~0.03 for He up to ~50% cumulative-style metrics for Kr in the authors’ framing) are contrasted with single-impact literature values, attributing gaps to multi-impact geometry. Simulation–experiment agreement is described as encouraging on dose and morphology while noting impurities, contamination, and possible metal-catalyzed chiseling in experiment. Authored limitations include dose-rate mismatch and using high-T anneals to stand in for room-temperature reconstruction kinetics within MD wall times.

Limitations

Just Accepted PDF: prefer the typeset VOR layout for pagination and final figure quality. This slug’s normalized/extracts slice is boilerplate-heavy; protocol numbers above were taken from pdf_path.

Relevance to group

Duplicate corpus proof of van Duin/ORNL graphene irradiation collaboration.

Citations and evidence anchors

  • DOI: https://doi.org/10.1021/acsnano.6b03036 (papers/Yoon_ACSNano_ASAP.pdf).