Tunable mechanical and thermal properties of one-dimensional carbyne chain: phase transition and microscopic dynamics
Summary¶
One-dimensional sp-hybridized carbon chains—α-polyyne (alternating single/triple bonds) and β-cumulene (repeating double bonds)—are studied with reactive MD using LAMMPS and the literature ReaxFF CHO parameter set (J. Phys. Chem. C). The paper stresses mechanical contrast (bending stiffness, Young’s modulus, tensile response), a thermally driven cumulene→polyyne transition with a critical temperature near 499 K and ultrafast bond rearrangement (~145 fs for the 50-atom chain in the authors’ Fig. 2 trajectory), and thermal transport from Green–Kubo equilibrium MD (cumulene κ reported around 83 W/(m·K) at 480 K in their calculation, roughly twice the polyyne value there, with defective polyyne dropping to about 13% of pristine polyyne κ).
Methods¶
MD application. Simulations use LAMMPS with ReaxFF for carbon/hydrogen chemistry: bond orders (σ, π, π–π contributions) are updated every MD step, and charges follow ReaxFF’s geometry-dependent scheme (J. Phys. Chem. C, Computational Methods). Integration is velocity Verlet with Δt = 0.1 fs. A repeated protocol equilibrates at target T with 100 ps NVT (Nosé–Hoover thermostat) followed by 100 ps NVE to check energy conservation before longer segments. Carbyne models are 50 carbon atoms per chain with periodic boundary conditions along the chain axis and free boundaries transverse to the axis in the thermal-conductivity setup; heat-flux data are accumulated each timestep for 10 ns for Green–Kubo analysis (the article notes PBC along the chain omits explicit length/edge effects in κ). Mechanical loading applies uniaxial strain by end displacements at 2.5×10⁻⁵ Å/fs (strain-rate independence reported between 2.5×10⁻⁶ and 2.5×10⁻⁴ Å/fs); virial stress is reported using a 3.35 Å effective chain diameter (plus 1D force metrics to reduce cross-section ambiguity). Bending analyses give fitted D ≈ 8.5 eV·Å (cumulene) vs 6.7 eV·Å (polyyne) in the small-curvature regime discussed in the text. Phase-transition studies compare slow heating (~0.04 K/fs, near-pristine polyyne) with 2.5 K/fs heating that leaves defective bond populations. Barostat / NPT, applied electric fields, and umbrella or metadynamics are not used in these chain protocols.
Force-field training. N/A — the article applies an existing ReaxFF parametrization (QM-fitted in prior work cited there); it does not document a new optimization campaign.
Static QM / DFT production. N/A — reported observables are from ReaxFF MD; DFT enters only as published benchmarks when the authors compare transition temperatures and bending stiffness to prior first-principles studies.
Findings¶
Outcomes and mechanisms. Near 499 K, cumulene converts to polyyne on a sub-200 fs timescale for the 50-atom model; slow heating from 5 K yields alternating single/triple bonding, whereas 2.5 K/fs ramps (or initializing hot cumulene) introduce nonideal bond-length peaks (~1.45 Å) tied to defective polyyne. Cumulene is stiffer in bending (D about 27% larger than polyyne in the quoted linear E_b vs ρ² regime) and shows ~2× higher thermal conductivity than polyyne in the authors’ Green–Kubo results, with defective bonding collapsing polyyne κ to about 13% of the pristine value. Tensile data for polyyne at 500 K give failure near ~8% strain with ultimate stress ~67 GPa (or ~7.5 nN 1D force) and a Young’s modulus ~1345 GPa from a linear fit to ~2% strain—numbers the authors compare to diamond, graphene, and nanotube benchmarks in the same section.
Comparisons. The article contrasts its 499 K transition temperature with a higher DFT-based literature estimate and lists classical vs quantum statistics, PBC choices, ReaxFF accuracy, and DFT band-gap issues as contributing factors—not as new experiments.
Sensitivity. Thermal conductivity and persistence length depend on temperature (e.g., L_p ~ 33 nm at 300 K in their polymer-physics estimate); mechanical response is insensitive to the tested strain-rate window but strongly sensitive to defect content after fast heating.
Authored limitations / outlook. They flag ReaxFF mismatch to first-principles forces where Peierls distortion is delicate, finite-size/PBC effects in 1D transport, and the need for careful temperature ramps to avoid defective polyyne when targeting ideal alternating bonds.
Limitations¶
ReaxFF accuracy for sp-rich carbon, classical phonon statistics vs quantum modes, PBC choices along the 1D chain, and Green–Kubo finite sampling all affect phase stability and reported κ; the authors explicitly compare their 499 K transition temperature to higher DFT literature values and list multiple methodological reasons (J. Phys. Chem. C discussion section).
Citations and evidence anchors¶
DOI 10.1021/acs.jpcc.5b08026; PDF papers/ReaxFF_others/Liu_carbyne_JPCC_2015.pdf.