Atom vacancies on a carbon nanotube: to what extent can we simulate their effects?
Curators
Duplicate ASAP PDF lineage for the same JCTC article as 2015kroes-venue-ct5b00292; cite that page for integrated protocol text if this proof/header layout is confusing.
Summary¶
Kroes et al. benchmark single- and double-vacancy physics in a (10,0) zigzag carbon nanotube across DFT (PBE, BLYP, partial PBE0 repeats) and several classical potentials (AIREBO, LCBOP, ReaxFF15, Tersoff, with REBO mentioned for SI). Quantities compared include formation energies, relaxed structures, and barriers for reconstruction, migration, and vacancy coalescence. The abstract warns that characterization of these processes differs markedly across methods, motivating cautious use of classical carbon models when building DFT-informed parametrizations. ReaxFF15 is highlighted as a modern reactive carbon parametrization included in the comparison suite.
Methods¶
This ASAP PDF (papers/Kroes_JCTC_2015.pdf) is a second corpus upload of the same JCTC article as 2015kroes-venue-ct5b00292 (DOI 10.1021/acs.jctc.5b00292). The manuscript’s Methods are unchanged: spin-polarized plane-wave DFT in CPMD with PBE and BLYP (selected PBE0 checks), Γ-only sampling, and classical energy benchmarks on (10,0) zigzag single-wall CNT models with single and double vacancies using AIREBO, LCBOPI, ReaxFF15, REBO (mostly SI), and Tersoff (also largely SI), with formation energies referenced to the carbon chemical potential defined in the paper. ReaxFF15 enters through static energies and barriers on the small tetragonal PBC supercells (~359–598 atoms depending on defect), not long finite-temperature ReaxFF MD; the separate ~100,000-atom high-temperature AIREBO MD plus annealing is only used to propose motifs that are then relaxed in DFT. Finite-temperature NVT/NPT production for the small-cell ReaxFF15 benchmarks: N/A — those rows are static energies and barrier workflows, not thermostatted NVT trajectories on the (10,0) defect cells. Hydrostatic pressure targets during those static evaluations: N/A — not reported as NPT pressure control for the ReaxFF15 comparison path (see 2015kroes-venue-ct5b00292 for timestep/thermostat detail on the exploratory AIREBO scouting MD).
Findings¶
- Some long-ranged relaxations require large cells for both DFT and ReaxFF in specific double-vacancy configurations described in the extract.
- The abstract warns that barriers and reconstruction sequences for single- and double-vacancy CNT defects are not consistent across empirical carbon models, so ReaxFF-trained kinetic Monte Carlo or coarse models must inherit this uncertainty unless re-parameterized against QM for the specific defect class.
Limitations¶
- Benchmark scope is a single chirality/diameter nanotube; transferability of conclusions to metallic tubes or wide diameter regimes is not established here.
- ASAP header placeholders appear in the extract; use DOI for bibliographic certainty.
- Curvature strain in small-diameter CNTs can amplify defect energies relative to planar graphene; kinetic models calibrated here should be re-tested on larger-diameter tubes before composite applications.
- Interstitial dopants and endohedral species can pin vacancies differently than pristine (10,0) supercells modeled in the JCTC benchmark.
Relevance to group¶
Adri C. T. van Duin co-authorship ties the benchmark to ReaxFF carbon quality control and cross-method uncertainty quantification important for downstream reactive carbon applications.