Mapping TpPa-1 covalent organic framework (COF) molecular interactions in mixed solvents via atomistic modeling and experimental study
Corpus note
The PDF is a corrected proof; pagination in the header is a placeholder. Record the final DOI from the version of record when available.
Summary¶
Organic solvent nanofiltration separates complex chemical streams using membranes whose performance depends on nanoscale pore topology, framework flexibility, and nonideal sorption in mixed solvents—effects that are difficult to infer from macroscopic permeance alone. This Journal of Membrane Science contribution models the TpPa-1 covalent organic framework in contact with multicomponent organic solvents and compares atomistic predictions of solvent permeance and solute rejection to laboratory OSN measurements. Yun Kyung Shin and Adri C. T. van Duin are coauthors, signaling force-field work consistent with the group’s classical and reactive parameterization practices, though readers must open the Computational Methods section to confirm whether fixed-bond or reactive potentials dominate for this chemistry. The ingested corpus file is a corrected proof; assign the final Digital Object Identifier from the version of record when it appears in publisher metadata.
Methods¶
Membrane experiments (OSN)¶
Polycrystalline TpPa-1 films; permeance + solute rejection in mixed organic solvents (pharma/fine-chemical-relevant conditions).
Atomistic transport models (classical MD)¶
Solvent–COF interactions; permeance normalized to water to bridge continuum tests and periodic cells; software, ensemble, duration, FF (fixed-bond vs reactive) in papers/Barnes_Shin_COF_JMemSci_2024_galley.pdf—not restated here.
MD application (classical, COF, mixed solvents). The Journal of Membrane Science work uses an MD package (e.g. GROMACS or LAMMPS as named in the article) for molecular dynamics of TpPa-1 framework+solvent cells; N/A for exact atom counts, 3D PBC and fixed framework protocols, NVT/NPT choice, 0.25–1 fs time step ranges, and Nosé–Hoover thermostat+NPT Parrinello/Berendsen barostat for chemical potential-matched permeation if used—Methods in the galley/VOR. Temperatures match OSN conditions. N/A—external electric field; N/A—metadynamics; Hydrostatic pressure per NPT when the protocol applies (N/A in NVT-only segments).
Findings¶
The authors report good agreement between modeled organic-solvent permeance/rejection trends and measured OSN data under explicit modeling assumptions. Where mismatch persists, they attribute discrepancies to realistic film microstructure effects such as linear polymer defects, grain boundaries, and interstitial percolation that bypass nominal pore channels—phenomena that a perfectly crystalline through-pore model would omit.
Comparisons and limitations as authored. The permeance and solute rejection story is explicitly compared to experiment; sensitivity to film defects explains residual gaps. The DOI will appear with the final JMS issue; see ## Limitations for proof caveats.
Limitations¶
Proof-stage PDFs can differ in pagination and editorial metadata from the final issue; operators should reconcile DOI, volume, and page numbers before citing locators. Force-field choices for COF–solvent systems should be verified in the Methods section because abstract-level summaries may not state whether bonds are fixed or reactive.
Relevance to group¶
The paper illustrates atomistic digital-twin style workflows for advanced membranes with van Duin-group participation, connecting polymer/COF materials keywords to solvation and transport validation.
MAS / retrieval notes¶
Agents should flag candidate_tags such as COF TpPa-1 and organic-solvent nanofiltration when building facet filters; until the final DOI is registered in frontmatter, retrieval answers should cite the corrected-proof caveat and avoid inventing pagination. Link to reaxff-family only after Methods confirm whether reactive or fixed-bond force fields dominate.