Skip to content

Molecular Interactions and Layer Stacking Dictate Covalent Organic Framework Effective Pore Size

Galley PDF

Corpus path Duong_Shin_ACS_AMI_COF_2021_galley.pdf is a galley; prefer the version-of-record PDF at the DOI when available locally. See also docs/corpus/NON_PRIMARY_ARTICLE_PAPER_SLUGS.md context for proof/galley handling.

Summary

Duong et al. combine ReaxFF molecular dynamics with experiments on an imine-linked carboxylated covalent organic framework (C-COF) membrane platform used for organic solvent nanofiltration (OSN). The scientific claim is that effective pore size—what actually governs selectivity—is not fixed by nominal crystal design alone, but emerges from layer stacking, framework flexibility, and molecule–framework interactions in swollen or solvated environments. Reactive MD captures hydrogen-bonding, conformational rearrangements, and pore breathing that static pore metrics omit, while experiments provide permeance/selectivity benchmarks under solvent conditions relevant to OSN. Adri C. T. van Duin is a coauthor, connecting the work to the group’s porous materials and liquid-interface simulation portfolio.

Methods

1 — MD application. ReaxFF reactive MD of a 2D imine-linked carboxylated C-COF in explicit organic solvents, modeling pore morphology, layer stacking/spacing, and solute/solvent structure inside channels (PBC membrane supercells; NVT/NPT choices and 1 fs-class timesteps as in ACS Appl. Mater. Interfaces 2021 Methods). Thermostat class and time constant, NPT Parrinello–Rahman barostat settings (if any), and target temperature in K are given in the article/SI—not transcribed in this short note. Equilibration and multi-ns-scale trajectory lengths (where given) support G(r,t)-like and pore diameter analyses; confirm ps/ns duration in the full text. Electric field, replica exchange, and shock loading—N/A.

2 — Experiments (integrated with MD). OSN (organic solvent nanofiltration) permeation and selectivity data for the same C-COF chemistry supply macroscopic benchmarks; N/A DFT replica in this paper’s main workflow beyond ReaxFF parametrization** context.

3 — Force-field / QM context. The article uses a published/fitted ReaxFF suitable for COF-solvent C/H/O/N-rich chemistry; N/A here to restate the full training set—see JACS-family SI-level detail in the VOR PDF (galley in corpus; confirm table IDs).

4 — Galley. Ingested ACS file is a proof/galley; prefer VOR pagination.

Findings

Conclusion in the abstract. Aligned, crystalline 1D pores can deliver ultrahigh permeance while still allowing tunable selectivity; MD and OSN experiments show effective (not just designed) pore aperture and solvated solute size depend on solvent environment, so stacking/flexure and molecule–framework interactions matter. Comparisons are ReaxFF trajectory-derived pore metrics versus laboratory filtration performance. Sensitivity to solvent and pore morphology (including pore “breathing”/stacking defects) is the mechanistic point. Limitation: a classical reactive FF is not a DFT thermometer; ReaxFF transfer to other COFs is uncertain without retraining. The local galley file limits citation precision—use the VOR PDF (see ## Limitations).

Limitations

Galley formatting may differ slightly from VOR; ReaxFF transferability to other COF chemistries requires case-by-case validation. OSN performance depends on membrane defects and macroscopic flow fields not fully represented in atomistic cells.

Relevance to group

Demonstrates ReaxFF applied to COFs and liquid–membrane transport with experimental grounding.

Citations and evidence anchors

Reader notes (navigation)

  • paper_keywords includes keyword:galley-or-proof-pdf.