Skip to content

The peculiar size and temperature dependence of water diffusion in carbon nanotubes studied with 2D NMR diffusion–relaxation D–T2eff spectroscopy

Summary

The study combines two-dimensional nuclear magnetic resonance diffusion–relaxation (D–T2eff) spectroscopy in the stray field of a superconducting magnet with molecular dynamics (MD) simulations to resolve how water dynamics inside carbon nanotubes (CNTs) depend on tube diameter (about 1.1–6.0 nm) and temperature (265–305 K). The work reports multiple resolved “tubular” water components with distinct self-diffusion coefficients and highlights a diameter window (about 3.0–4.5 nm) where confined water shows non-Arrhenius, ultrafast diffusion and high fragility. Nanofluidic transport in CNT membranes motivates separating bulk-like vs surface layers of confined water because slip lengths and effective permeability depend on which population dominates signal in NMR.

Methods

  • Experiments: 2D NMR D–T2eff measurements on confined water in CNTs across the stated diameter and temperature ranges; methods and pulse/field details are given in the article and Supporting Information.
  • Simulations: MD simulations of confined water in CNTs used alongside experiment to interpret size- and temperature-dependent dynamics (implementation details, water model, thermostat, and simulation length are specified in the PDF).
  • Stray-field NMR leverages spatially encoded diffusion filters to separate components with distinct D and T2—critical where pore polydispersity would overlap 1D spectra.

Molecular dynamics (complement to experiment). The authors use molecular dynamics in a standard MD package to simulate water confined in cylindrical carbon nanotube models with O(10^2–10^3) atoms per cell (as stated). PBC or finite-length CNTs follow the published setup. Runs use an NVT-type thermostat to hold 265–305 K temperature; N/A — barostat and N/A for NPT GPa pressure in typical confined-water equilibration under fixed volume; N/A — electric field. Timestep in femtoseconds and duration in ns/long ps of equilibration and production are in the article/SI. N/A — metadynamics / replica exchange per abstract scope.

Findings

  • Confined water can separate into two or more dynamical components along the CNT axis, with different self-diffusion coefficients depending on diameter.
  • In a favorable diameter range (about 3.0–4.5 nm), dynamics of water near the CNT center show distinctly non-Arrhenius behavior, with very fast diffusion and extraordinary fragility, matching the experimental emphasis in the abstract.
  • Joint fit of simulation radial density profiles and experimental D–T2 clusters supports assigning fast components to weakly perturbed core water populations in mid-sized tubes.

Limitations

NMR and MD each carry assumptions (surface chemistry of CNTs, water model, accessible time scales); quantitative comparison across all diameters should follow the uncertainties and controls described in the publication. Metallic vs semiconducting CNT chirality and residual catalyst particles can modify confined water dynamics beyond smooth cylinder models. Pore length and entrance effects in membrane samples may couple to NMR exchange times not captured in infinite cylinder idealizations.

Relevance to group

Provides a tightly coupled experiment + MD example of nanoconfined water transport—useful context for interface and confinement problems adjacent to reactive MD work, though the core methodology here is not ReaxFF-centric.

Citations and evidence anchors

Reader notes (navigation)

This is primarily a classical water + NMR paper—use it for nanoconfined transport context adjacent to ReaxFF electrolyte work.