Electrokinetic flow of an aqueous electrolyte in amorphous silica nanotubes.

We study the pressure-driven flow of aqueous NaCl in amorphous silica nanotubes using nonequilibrium molecular dynamics simulations featuring both polarizable and non-polarizable molecular models. Different pressures, electrolyte concentrations and pore sizes are examined. Our results indicate a flow that deviates considerably from the predictions of Poiseuille fluid mechanics. Due to preferential adsorption of the different ionic species by surface SiO- or SiOH groups, we find that a significant electric current is generated, but with opposite polarities using polarizable vs. fixed charge models for water and ions, emphasizing the need for careful parameterization in such complex systems. We also examine the influence of partial deprotonation of the silica surface, and we find that much more current is generated in a dehydrogenated nanopore, even though the overall efficiency remains low. These findings indicate that different methods of nanopore preparation, which can produce a range of surface properties, should be examined more closely in the related experimental methods to generate electrokinetic current.

Algorithms for GPU-based molecular dynamics simulations of complex fluids: Applications to water, mixtures, and liquid crystals.

A custom code for molecular dynamics simulations has been designed to run on CUDA-enabled NVIDIA graphics processing units (GPUs). The double-precision code simulates multicomponent fluids, with intramolecular and intermolecular forces, coarse-grained and atomistic models, holonomic constraints, Nosé-Hoover thermostats, and the generation of distribution functions. Algorithms to compute Lennard-Jones and Gay-Berne interactions, and the electrostatic force using Ewald summations, are discussed. A neighbor list is introduced to improve scaling with respect to system size. Three test systems are examined: SPC/E water; an n-hexane/2-propanol mixture; and a liquid crystal mesogen, 2-(4-butyloxyphenyl)-5-octyloxypyrimidine. Code performance is analyzed for each system. With one GPU, a 33-119 fold increase in performance is achieved compared with the serial code while the use of two GPUs leads to a 69-287 fold improvement and three GPUs yield a 101-377 fold speedup.

Molecular dynamics simulations to examine structure, energetics, and evaporation/condensation dynamics in small charged clusters of water or methanol containing a single monatomic ion.

We study small clusters of water or methanol containing a single Ca(2+), Na(+), or Cl(-) ion with classical molecular dynamics simulations, using models that incorporate polarizability via the Drude oscillator framework. Evaporation and condensation of solvent from these clusters is examined in two systems, (1) for isolated clusters initially prepared at different temperatures and (2) those with a surrounding inert (Ar) gas of varying temperature. We examine these clusters over a range of sizes, from almost bare ions up to 40 solvent molecules. We report data on the evaporation and condensation of solvent from the clusters and argue that the observed temperature dependence of evaporation in the smallest clusters demonstrates that the presence of heated gas alone cannot, in most cases, solely account for bare ion production in electrospray ionization (ESI), neglecting the key contribution of the electric field. We also present our findings on the structure and energetics of the clusters as a function of size. Our data agree well with the abundant literature on hydrated ion clusters and offer some novel insight into the structure of methanol and ion clusters, especially those with a Cl(-) anion, where we observe the presence of chain-like structures of methanol molecules. Finally, we provide some data on the reparameterizations necessary to simulate ions in methanol using the separately developed Drude oscillator models for methanol and for ions in water.

Characterization of microstructured fibre emitters: in pursuit of improved nano electrospray ionization performance.

Full-dimensional computational fluid dynamics (CFD) simulations are presented for nano electrospray ionization (ESI) with various emitter designs. Our CFD electrohydrodynamic simulations are based on the Taylor-Melcher leaky-dielectric model, and the volume of fluid technique for tracking the fast-changing liquid-gas interface. The numerical method is first validated for a conventional 20 µm inner diameter capillary emitter. The impact of ESI voltage, flow rate, emitter tapering, surface hydrophobicity, and fluid conductivity on the nano-ESI behavior are thoroughly investigated and compared with experiments. Multi-electrospray is further simulated with 2-hole and 3-hole emitters with the latter having a linear or triangular hole arrangement. The simulations predict multi-electrospray behavior in good agreement with laboratory observations.

How are completely desolvated ions produced in electrospray ionization: insights from molecular dynamics simulations.

We apply molecular dynamics (MD) simulations to study the final phase of electrospray ionization (ESI), where an ion loses all of its associated solvent molecules. By applying an electric field to a cluster of H(2)O molecules solvating an ion and including a surrounding gas of varying pressure, we demonstrate that collisions with the gas play a major role in removing this final layer of solvent. We make quantitative predictions of the critical velocity required for the cluster to start losing molecules via collisions with gas and propose that this should be important in real ESI experiments. Such collisions have heretofore not been explicitly considered in discussions of the ESI process.

Solvation of phenylglycine- and leucine-derived chiral stationary phases: molecular dynamics simulation study.

A theoretical study of the solvation of ( R)- N-(3,5-dinitrobenzoyl)phenylglycine- and ( R)- N-(3,5-dinitrobenzoyl)leucine-derived chiral stationary phases (CSPs) is presented. Semiflexible models of the chiral selectors are prepared from B3LYP/6-311G** calculations, and these are used in the molecular dynamics simulations of the corresponding interface. The chiral interface is examined for four solvents: 100% hexane, 90:10 hexane:2-propanol, 80:20 hexane:2-propanol, and 100% 2-propanol. Despite the similarities between phenylglycine and leucine, the interfaces are distinct both in terms of the selector orientations at the surface and in the number of hydrogen bonds formed with 2-propanol. We also find that an increase in alcohol concentration alters the preferred orientations of the selectors.

Solvation of the N-(1-Phenylethyl)-N'-[3-(triethoxysilyl)propyl]-urea chiral stationary phase in mixed alcohol/water solvents.

A theoretical and experimental study of alcohol/water and alcohol/alcohol solvent mixtures near a surface of N-(1-phenylethyl)-N'-[3-(triethoxysilyl)propyl]-urea (PEPU), a Pirkle-type chiral stationary phase, is presented. Molecular dynamics simulations are performed at room temperature for water/methanol, water/1-propanol, water/2-propanol, and methanol/1-propanol solvent mixtures confined between two PEPU surfaces. The interface was also prepared experimentally by attaching the PEPU molecules to atomic force microscopy tips and oxidized Si(111) substrates. Chemical force spectrometric measurements between such PEPU-terminated tips and samples were taken in the solvent mixtures, and the results are compared to the molecular dynamics study. We find that the extent of hydrogen bonding at the surface is the dominant contributor to the measured forces.