Using high resolution inelastic neutron scattering measurements of the phonon density of states of ice, two separate molecular optic bands at about 28 and 37 meV for ice Ih and ice Ic have been observed, which is the direct result of the very high flux of pulsed neutron source and excellent energy resolution of the inelastic instruments on ISIS at the Rutherford–Appleton Laboratory. In order to reproduce the measured phonon density of states, a new lattice dynamic model has been proposed that two interaction strengths among the H bonds in ice Ih (and ice Ic) associated with the proton configurations have to be assumed. These two force constants are randomly distributed in the network of ice, having a ratio of almost 1:2. This model is capable of reproducing almost every aspect of the measured spectra, the two molecular peaks in 28 and 37 meV in particular and may provide insight into the complex nature of H bonding in ice and water. The large differences in the force constants between the strong and weak H bonds in a disordered ice structure as required by the model would affect the physical and chemical properties of ice and water and could have broader implications. In this paper, we illustrate a complete measurement for all possible recovered high-pressure forms of ice, including ice I, II, IX, V, VI, VII, VIII, and high-/low-density amorphous ice. These measurements demonstrate that the two well-separated molecular bands are associated with the local configurations of protons, which have quite different interaction strengths.

Understanding and controlling the transport of water across nanochannels is of great importance for designing novel molecular devices, machines and sensors and has wide applications, including the desalination of seawater. Nanopumps driven by electric or magnetic fields can transport ions and magnetic quanta, but water is charge-neutral and has no magnetic moment. On the basis of molecular dynamics simulations, we propose a design for a molecular water pump. The design uses a combination of charges positioned adjacent to a nanopore and is inspired by the structure of channels in the cellular membrane that conduct water in and out of the cell (aquaporins). The remarkable pumping ability is attributed to the charge dipole-induced ordering of water confined in the nanochannels, where water can be easily driven by external fields in a concerted fashion. These findings may provide possibilities for developing water transport devices that function without osmotic pressure or a hydrostatic pressure gradient.

We present the parametrization of a new polarizable model for water based on Thole’s method [Chem. Phys. 59, 341 (1981)] for predicting molecular polarizabilities using smeared charges and dipoles. The potential is parametrized using first principles ab initio data for the water dimer. Initial benchmarks of the new model include the investigation of the properties of water clusters (n=2–21) and (hexagonal) ice Ih using molecular dynamics simulations. The potential produces energies and nearest-neighbor (H-bonded) oxygen–oxygen distances that agree well with the ab initio results for the small water clusters. The properties of larger clusters with 9–21 water molecules using predicted structures from Wales et al. [Chem. Phys. Lett. 286, 65 (1998)] were also studied in order to identify trends and convergence of structural and electric properties with cluster size. The simulation of ice Ih produces a lattice energy of −65.19 kJ/mol (expt. −58.9 kJ/mol) with an average dipole moment of 2.849 D. The calculated spectrum for the phonon density of states exhibits features that may correspond to the experimentally measured peaks at 28 and 37 meV. The many body contribution to the total energy is found to be close to 31% for both the water clusters and for ice Ih.

No abstract

The selective rate of specific ion transport across nanoporous material is critical to biological and nanofluidic systems. Molecular sieves for ions can be achieved by steric and electrical effects. However, the radii of Na(+) and K(+) are quite similar; they both carry a positive charge, making them difficult to separate. Biological ionic channels contain precisely arranged arrays of amino acids that can efficiently recognize and guide the passage of K(+) or Na(+) across the cell membrane. However, the design of inorganic channels with novel recognition mechanisms that control the ionic selectivity remains a challenge. We present here a design for a controllable ion-selective nanopore (molecular sieve) based on a single-walled carbon nanotube with specially arranged carbonyl oxygen atoms modified inside the nanopore, which was inspired by the structure of potassium channels in membrane spanning proteins (e.g., KcsA). Our molecular dynamics simulations show that the remarkable selectivity is attributed to the hydration structure of Na(+) or K(+) confined in the nanochannels, which can be precisely tuned by different patterns of the carbonyl oxygen atoms. The results also suggest that a confined environment plays a dominant role in the selectivity process. These studies provide a better understanding of the mechanism of ionic selectivity in the KcsA channel and possible technical applications in nanotechnology and biotechnology, including serving as a laboratory-in-nanotube for special chemical interactions and as a high-efficiency nanodevice for purification or desalination of sea and brackish water.

No abstract

Repulsive force between the O-H bonding electrons and the O:H nonbonding pair within hydrogen bond (O-H:O) is an often overlooked interaction which dictates the extraordinary recoverability and sensitivity of water and ice. Here, we present a potential model for this hidden force opposing ice compression of ice VIII-X phase transition based on the density functional theory (DFT) and neutron scattering observations. We consider the H-O bond covalent force, the O:H nonbond dispersion force, and the hidden force to approach equilibrium under compression. Due to the charge polarization within the O:H-O bond, the curvatures of the H-O bond and the O:H nonbond potentials show opposite sign before transition, resulting in the asymmetric relaxation of H-O and O:H (O:H contraction and H-O elongation) and the H+ proton centralization towards phase X. When cross the VIII-X phase boundary, both H-O and O:H contract slightly. The potential model reproduces the VIII-X phase transition as observed in experiment. Development of the potential model may provide a choice for further calculations of water anomalies.

No abstract