Charged Nanopores Research Articles

PH and Ionic Strength Effects on Amino Acid Transport through Au-Nanotubule Membranes Charged with Self-Assembled Monolayers

The transport of the amino acids tyrosine and phenylalanine through Au-nanotubule membranes with self-assembled monolayers of alkanethiols is theoretically and experimentally studied. The membranes are prepared by electroless deposition of gold on porous polycarbonate track-etched membranes with control of the inner pore size, followed by self-assembly of acid-functionalized thiols (final mean average pore radius: 7 nm). The capability of switchable ion-transport selectivity by external control (e.g., by changing the pH, ionic strength, and amino acid concentration) is discussed. The flux changes with the pH and the ionic strength of the solution clearly show that electrical charges play a key role in the amino acid transport through the nanopores. Remarkably, an uphill transport of the amino acid is observed when a pH difference is imposed in the external solutions. The theoretical approach based on the Nernst−Planck flux equations considers all of the charged species present in the system (the cationic, anionic, and zwitterionic forms of the amino acid, the hydrogen and hydroxide ions, and the two salt ions) and allows a qualitative understanding of the transport phenomena in the charged nanopores.

Molecular Understanding of Osmosis in Semipermeable Membranes

We investigate single-file osmosis of water through a semipermeable membrane with an uncharged, a positively and a negatively charged nanopore. Molecular dynamics simulations indicate that the osmotic flux through a negatively charged pore (J_) is higher compared to the osmotic flux in a positively charged pore (J+) followed by the osmotic flux in the uncharged pore (J(0)), i.e., J_ > J+ > J(0). The molecular mechanisms governing osmosis, steady state osmosis, and the observed osmotic flux dependence on the nanopore charge are explained by computing all the molecular interactions involved and identifying the molecular interactions that play an important role during and after osmosis. This study helps in a fundamental understanding of osmosis and in the design of advanced nanoporous membranes for various applications of osmosis.

Ionic conduction, rectification, and selectivity in single conical nanopores

Modern track-etching methods allow the preparation of membranes containing a single charged conical nanopore that shows high ionic permselectivity due to the electrical interactions of the surface pore charges with the mobile ions in the aqueous solution. The nanopore has potential applications in electrically assisted single-particle detection, analysis, and separation of biomolecules. We present a detailed theoretical and experimental account of the effects of pore radii and electrolyte concentration on the current-voltage and current-concentration curves. The physical model used is based on the Nernst-Planck and Poisson equations. Since the validity of continuum models for the description of ion transport under different voltages and concentrations is recognized as one of the main issues in the modeling of future applications, special attention is paid to the fundamental understanding of the electrical interactions between the nanopore fixed charges and the mobile charges confined in the reduced volume of the inside solution.

Protein transport through gold-coated, charged nanopores: Effects of applied voltage

The flux of bovine serum albumin and bovine hemoglobin through charged nanopores inside polymeric membranes is analysed as a function of the applied voltage to the nanopore surface, the solution ionic strength and pH. The electrostatic interaction of the protein with the nanopore surface gives low transport rates except at the protein isoelectric point and the minimum of the effective, voltage-induced nanopore charge. This electrostatic sieving effect allows for the separation of proteins with similar molecular weights.

Effect of Protein Adsorption and Ionic Strength on the Equilibrium Partition Coefficient of Ionizable Macromolecules in Charged Nanopores

Recent experimental data on the diffusive transport of globular protein molecules through nanopores (D ∼ 20 nm), covered with chemisorbed self-assembled monolayers (SAMs) of carboxylic acid functional groups (Ku, J.-R.; Stroeve, P. Langmuir 2004, 20, 2030), show a strong increase in flux around the isoelectric point of the protein molecule at 10 mM ionic strength (on−off behavior), but not at 100 mM. To understand these observations we calculated the equilibrium partition coefficient (distribution coefficient) of a spherical protein molecule in a nanopore consisting of ionizable carboxylic acid groups as a function of ionic strength, pH, charge and size of the protein and the pore wall. When transport is diffusion-controlled, the flux of macromolecules through porous media is directly proportional to the equilibrium partition coefficient. To calculate the charge on the pore wall, the ionizable character of the carboxylic acid groups is incorporated as well as protein adsorption. At values sufficiently above the isoelectric point of the protein, pI, protein adsorption is zero. However, with decreasing pH, adsorption strongly increases. Therefore, the pore is negatively charged at sufficiently high pH (just as the protein molecule) and positively charged at sufficiently low pH (just as the protein molecule). Consequently, the equilibrium partition coefficient of additional, thus nonadsorbing, protein molecules is strongly pH-dependent with a maximum at pH around pI. The predicted influence of ionic strength on the peak width of the equilibrium partition coefficient is in qualitative agreement with the experiments, suggesting that electrostatic interactions and especially protein adsorption are important determinants of protein transport through charged nanopores.

Nonequilibrium molecular dynamics simulation of electro-osmotic flow in a charged nanopore

Nonequilibrium molecular dynamics simulations were performed for Poiseuille and electro-osmotic flow in a charged cylindrical nanopore. The goal was to examine any deviations from continuum flow behavior and to compare and contrast the Poiseuille and electro-osmotic flow situations. The fluid was composed of cationic counterions and nonpolar monatomic solvent molecules. The cylindrical surface of the pore wall was represented by a stochastic scattering boundary condition. The lack of any surface roughness and the computational efficiency of the fluid model enabled the velocity profile near the wall to be measured at very high spatial resolution. The simulation results indicate that both Poiseuille and electro-osmotic flow conform to continuum transport theories except in the first monolayer of fluid at the pore wall. The apparent viscosity in this region was highly nonuniform and exhibited singularities. Despite this, the viscosity profiles obtained from Poiseuille and electro-osmotic flow were in good mutual agreement at all locations. The singularities were caused by a local maximum in the solvent and counterion velocity profiles occurring at the edge of the first monolayer of liquid. This apparent channeling of fluid near the pore wall has been observed in previous studies of Poiseuille flow. The exact cause is not clear, but it may be due to cooperative transport of the fluid molecules facilitated by two-dimensional ordering at the wall.

Deviation from electroneutrality in cylindrical pores

Grand canonical ensemble Monte Carlo simulations are performed to investigate the concentrations of electrolytes in a uniformly charged cylindrical nanopore. The electrolyte inside the pore is assumed to be in equilibrium with a bulk electrolyte solution. Two models of electrolytes, namely the restricted primitive model (RPM) and the RPM plus neutral hard spheres, denoted as the solvent primitive model (SPM) are investigated. The results of the two models are compared, and they are also compared to the numerical solution of the Poisson–Boltzmann equation. The effect of pore wall charge on non-neutrality is reported for the SPM model.