Molecular dynamics (MD) simulations provide a powerful tool for studying ion transport through nanopores and 2D material membranes at atomic resolution. These simulations capture the dynamic interactions between ions, water molecules, and the pore structure, offering insights into selectivity mechanisms, energy barriers, and transport kinetics that are difficult to probe experimentally. The following sections detail key aspects of ion transport as revealed by MD studies, focusing on selectivity, dehydration effects, voltage-driven transport, and electrokinetic phenomena.

Ion selectivity in nanopores arises from a combination of steric, electrostatic, and hydration effects. MD simulations of graphene and hexagonal boron nitride (hBN) membranes demonstrate that narrow pores with sub-nanometer diameters exhibit strong selectivity based on ion size and charge. For example, a pore diameter of 0.6–0.7 nm in graphene allows the passage of K+ ions while excluding larger Na+ ions due to steric hindrance. Electrostatic interactions further enhance selectivity when the pore contains functional groups or partial charges. In biological ion channels like gramicidin A, MD reveals that carbonyl oxygen atoms lining the pore preferentially coordinate dehydrated K+ over Na+ due to optimal charge density matching, a mechanism replicated in synthetic nanopores with tailored chemical modifications.

Dehydration energy barriers play a critical role in ion permeation. As an ion enters a narrow pore, it must shed part of its hydration shell, requiring energy that depends on pore geometry and chemistry. MD simulations quantify these barriers by tracking free energy profiles (potential of mean force) along the transport pathway. For instance, a Na+ ion traversing a MoS2 nanopore experiences a 20–30 kJ/mol barrier due to partial dehydration, while K+ faces a lower barrier (10–15 kJ/mol) owing to its more flexible hydration shell. Functionalizing the pore rim with negatively charged groups can reduce dehydration penalties by stabilizing the ion’s transition state, as shown in simulations of carboxylate-modified carbon nanotubes.

Voltage-driven transport is governed by the interplay between electric forces and stochastic ion-pore interactions. MD studies of applied electric fields across membranes reveal nonlinear current-voltage relationships, with ion conduction occurring in bursts correlated with water dipole reorientation. At high voltages (>0.5 V/nm), simulations predict electroosmotic flow dominating ion transport, where water drags ions through the pore. This effect is pronounced in hydroxylated graphene pores, where surface charges enhance water mobility. Conversely, at low voltages, ion hopping between binding sites controls conductance, as observed in MD models of biological potassium channels.

Concentration polarization near the membrane surface significantly impacts transport efficiency. MD simulations show ion depletion and accumulation zones forming within nanometers of the pore entrance, creating local electric fields that oppose applied voltages. For example, in desalination membranes, Cl- accumulation on the feed side reduces Na+ influx by 40–60% in simulated systems, consistent with experimental observations. The extent of polarization depends on pore density; membranes with sparse nanopores exhibit more severe polarization than those with uniformly distributed pores.

Electrokinetic phenomena, including streaming potentials and current rectification, emerge from asymmetric ion dynamics. MD reveals that conical or charged nanopores rectify ionic current due to uneven energy barriers for forward and backward ion movement. A study of alumina nanopores demonstrated 5:1 rectification ratios when pore asymmetry combined with surface charges favored cation transport in one direction. Similarly, pressure-driven flow induces streaming potentials, with MD predicting 1–10 mV potentials for 100 MPa pressure gradients across silica nanopores, depending on surface charge density.

Bioinspired channel designs leverage MD insights to optimize performance. Synthetic pores mimicking the selectivity filter of K+ channels achieve high K+/Na+ selectivity by replicating the precise arrangement of carbonyl groups that stabilize dehydrated K+. MD-guided modifications, such as adjusting pore diameter or introducing negative charges, further enhance selectivity ratios beyond 10:1. Another example is artificial water channels incorporating amphiphilic structures that facilitate rapid water permeation while excluding ions, as validated by MD simulations showing water flow rates comparable to aquaporins.

In desalination applications, MD has identified trade-offs between permeability and selectivity. Ultrathin graphene membranes with high pore densities achieve simulated salt rejections >90% while maintaining water fluxes orders of magnitude higher than conventional membranes. However, pore alignment is critical; misaligned pores increase ion leakage due to tortuous pathways, as quantified by MD trajectory analysis. Functionalization with nitrogen or fluorine atoms can mitigate leakage by creating energy barriers for ion translocation.

Limitations of MD include timescale constraints (nanoseconds to microseconds) and force field accuracy. Polarizable force fields improve descriptions of ion-water interactions but increase computational cost. Recent advances in machine learning potentials and enhanced sampling methods address these challenges, enabling longer simulations and more accurate free energy calculations. Despite limitations, MD remains indispensable for probing nanoscale transport mechanisms inaccessible to continuum models or experiments.

Key findings from MD studies can inform membrane design rules:
- Pore diameters <0.8 nm enhance selectivity via steric exclusion.
- Charged functional groups compensate dehydration energies.
- Uniform pore distributions minimize concentration polarization.
- Asymmetric geometries enable current rectification.
- Bioinspired chemical patterns replicate natural channel selectivity.

These principles guide the development of next-generation membranes for desalination, energy harvesting, and biomimetic sensors, with MD simulations providing atomic-level validation before experimental fabrication. Future work will explore dynamic pores, multifunctional coatings, and coupled transport processes to achieve advanced performance metrics.