Skip to content

Anomalous proton conduction behavior across a nanoporous two-dimensional conjugated aromatic polymer membrane

Aqueous proton transport across a nanoporous two-dimensional conjugated aromatic polymer (2D-CAP) membrane is studied with large-scale ReaxFF reactive molecular dynamics. The computed barrier for proton penetration across 2D-CAP is higher than for a graphtetrayne reference despite larger pores in 2D-CAP, traced to edge hydrogen that stabilizes a local hydrogen-bond network with pore water and slows water mobility, impeding proton conduction.

Summary

The work compares proton penetration energetics and mechanism for 2D-CAP versus graphtetrayne using ReaxFF-based reactive MD. The central result is an anomalously high barrier for 2D-CAP tied to nanopore-edge chemistry (hydrogen at the periphery) rather than pore size alone: those atoms participate in a stable local H-bond network with water in and near the pore, reducing water mobility and hindering proton transport. The study argues that pore-edge decoration must be considered alongside pore size when designing nanoporous 2D proton-selective membranes.

Methods

  • Software / FF: LAMMPS with the CHON-2017_weak ReaxFF parametrization; time step 0.25 fs; Nose-Hoover chain thermostat with 100 fs (NVT) and 1000 fs (NPT) temperature damping as stated in Section 2.2.
  • Ensembles: Metadynamics runs for barrier sampling; additional 500 ps NVT 300 K checks of proton-position statistics; 2 ns NPT (P = 1 atm, T = 300 K) with x and y fixed, then 2 ns NVT 300 K equilibration before production NVT analysis of 2D-CAP vs reference G4 (graphtetrayne).
  • Analysis: Unbiased 2 ns NVT trajectories of proton-membrane distance (Figure 2); metadynamics for penetration barriers; water mobility and H-bond structure at pore edges.

MD application (complementing the bullets). LAMMPS + CHON-2017_weak ReaxFF; 0.25 fs timesteps; Nose–Hoover chains (100 fs coupling in NVT, 1000 fs in some NPT stages per §2.2); 2 ns NPT at 1 atm / 300 K with lateral cell area fixed then 2 ns NVT at 300 K before production NVT 2D-CAP vs G4 (graphtetrayne) comparisons. Replica / umbrella: N/A — not used; metadynamics: yes for barrier sampling of proton permeation. Electric (static) field: N/A — not used. PBC 3D condensed water+membrane supercells (see pdf_path for box dimensions and 2D-CAP and pore H decoration).

Findings

  • The proton penetration barrier for 2D-CAP is about twice that for graphtetrayne, even though 2D-CAP has a larger pore in the comparison presented.
  • The high barrier is attributed to 2D-CAP’s atomic nanopore structure, specifically hydrogen at the pore periphery forming a stable local hydrogen-bond network with water inside or near the nanopores.
  • Water molecules participating in that network show reduced mobility, which impedes the proton transport pathway through the pore.
  • The authors conclude that proton penetration across nanoporous 2D materials depends on pore size and pore-edge composition (decorating atoms or functional groups); certain edge hydrogens can further hamper proton conductivity via localized H-bonding to water.

Comparisons: 2D-CAP is contrasted with G4 (graphtetrayne): higher barrier in 2D-CAP despite a larger pore in the same comparison, tracing to pore-edge H (see PCCP text). Sensitivity / parameters: all NVT / NPT stages reported at 300 K and 1 atm (NPT) in the equilibration blocks in Section 2.2. Limitations: ReaxFF-only barriers and H-bond patterns; uncertainty vs experiment / DFT remains (see ## Limitations). Corpus honesty: this ReaxFF_others pdf_path is the author VOR citable PDF for reproduction; definitive numbers from Table / figure captions in that file, not inferred** here.

Limitations

Barrier heights and structural interpretations are force-field–dependent; quantitative agreement with experiment or DFT would require cross-checks beyond the ReaxFF study as reported.

Relevance to group

Illustrates ReaxFF application to 2D nanoporous membranes and aqueous proton transport, relevant to fuel-cell and flow-battery membrane design contexts cited in the introduction.

Citations and evidence anchors

  • DOI: 10.1039/c9cp06372b