A computational investigation of thermodynamics, structure, dynamics and solvation behavior in modified water models

We investigate the properties of geometrically modified water models by performing molecular dynamics simulations of perturbations of the extended simple point charge (SPC/E) model of water over a wide range of temperatures at 1 bar. The geometric modification consists of altering the H-O-H angle in SPC/E. The dipole moment is held constant by altering the O-H bond length, while the electrostatic charges are left unchanged. We find that a H-O-H angle of at least 100 degrees is necessary for the appearance of density anomalies and of solubility extrema with respect to temperature for small apolar solutes. We observe the occurrence of two incompatible types of structural order in these models: Tetrahedral, with waterlike translational order for bent models with H-O-H angles in excess of 100 degrees ; and linear, with Lennard-Jones-like orientationally averaged translational order for smaller H-O-H angles. Increasing the H-O-H angle causes the density to increase, while at the same time shifting waterlike anomalies to progressively higher temperatures. For bent models with H-O-H angle greater than SPC/E's, we observe arrest of translational motion at 300 K (115 degrees) and 330 K (120 degrees).     

On the application of some solvation models to the water dimer

Three different solvation models are applied to a simple molecular system: the water dimer. The first order perturbation scheme of Hoijtink and Jano, Miertus and Kysel implementation of Klopmans' solvaton model and the virtual charge model of Constanciel and Tapia are used to study some aspects of proton potential curves. The calculations indicate that the solvent reorganization energy is important for the realistic reproduction of solvation. The solvent effect in correlation energy is also investigated using a second order perturbative approach. Finally some difficulties found when using the solvaton method are discussed by means of the Hartree-Fock instabilities theory of Paldus and Cizek and Stanton.     

Ion solvation in water clusters

We describe here molecular dynamics computer simulations performed to study the solvation of ions (Cl^? and Na⁺) in water clusters. Our simulations show that the calculated structure and dynamics of the clusters is very sensitive to the potential model which is used to describe the interactions. From the comparison with thermodynamic data and data from the photoelectron spectra we conclude that in Cl^?(H₂O)_n (n≤20) clusters the ion is located on the surface of the cluster.     

Improved methods for semiempirical solvation models

We present improved algorithms for the SMx (x = 1, 1a, 2, 3) solvation models presented previously [see the overview in C. J. Cramer and D. G. Truhlar, J. Comp.-Aided Mol. Design, 6 , 629 (1992)]. These models estimate the free energy of solvation by augmenting a semiempirical Hartree-Fock calculation on the solute with the generalized Born (GB) model for electric polarization of the solvent and a surface tension term based on solvent-accessible surface area. This article presents three improvements in the algorithms used to carry out such calculations, namely (1) an analytical accessible surface area algorithm, (2) a more efficient radial integration scheme for the dielectric screening computation in the GB model, and (3) a damping algorithm for updating the GB contribution to the Fock update during the iterations to achieve a self-consistent field. Improvements (1) and (2) decrease the computer time, and improvement (3) leads to more stable convergence. Improvement (2) removes a small systematic numerical error that was explicitly absorbed into the parameterization in the SMx models. Therefore, we have adjusted the parameters for one of the previous models to yield essentially identical performance as was obtained originally while simultaneously taking advantage of improvement (2). The resulting model is called SM2.1. The fact that we obtain similar results after removing the systematic quadrature bias attests to the robustness of the original parameterization. © 1995 by John Wiley & Sons, Inc.     

Characteristic behavior of short-term dynamics in reorientation for Gay-Berne particles near the nematic-isotropic phase transition temperature

A specific transition behavior was found in the tumbling motion near the nematic-isotropic phase boundary using molecular dynamics simulations of the Gay-Berne mesogenic model under isobaric conditions at a reduced pressure P* of 2.0. The relaxation time for the motion obtained from the second-rank orientational time correlation function and the rotational diffusion coefficient showed a clear jump at the nematic-isotropic phase transition temperature. Regardless of the temperature dependence of the relaxation time, the change in the rotational diffusion coefficient evaluated from the orientational order parameters and the relaxation time agreed qualitatively with that of real mesogens. The rotational viscosity coefficients gamma(1) and gamma(2) were obtained from the simulation data for the relaxation time for the short-term dynamics and for the rotational diffusion coefficients. gamma(1) was proportional to (2), where is the second-rank orientational parameter. Furthermore, the rotational behavior of the model was compared with that of the Debye approximation in the isotropic phase.     

Hydrophobic agglomeration kinetics of fine kaolinite particles in aqueous suspensions

The hydrophobic agglomeration kinetics of kaolinite has been studied through the in-situ turbidity meter measurement system. The effects of surfactant dosage, stirring rate and particle size on the hydrophobic agglomeration dynamics have been investigated. Appropriate surfactant concentrations provided the strongest hydrophobicity, modest stirring rate offered a higher agglomerates formation rate and a lower disruption rate, and smaller particles had lower energy barrier and higher attachment efficiency. The best hydrophobic agglomeration conditions have been determined and a kinetic model of hydrophobic agglomeration has been proposed by experimental results. The kinetic model has been turned out to be fitted to the dynamic process well and can characterize as well as assess the hydrophobic agglomeration kinetics appropriately.     

Continuum solvation models in the linear interaction energy method

The linear interaction energy (LIE) method in combination with two different continuum solvent models has been applied to calculate protein-ligand binding free energies for a set of inhibitors against the malarial aspartic protease plasmepsin II. Ligand-water interaction energies are calculated from both Poisson-Boltzmann (PB) and Generalized Born (GB) continuum models using snapshots from explicit solvent simulations of the ligand and protein-ligand complex. These are compared to explicit solvent calculations, and we find close agreement between the explicit water and PB solvation models. The GB model overestimates the change in solvation energy, and this is caused by consistent underestimation of the effective Born radii in the protein-ligand complex. The explicit solvent LIE calculations and LIE-PB, with our standard parametrization, reproduce absolute experimental binding free energies with an average unsigned error of 0.5 and 0.7 kcal/mol, respectively. The LIE-GB method, however, requires a constant offset to approach the same level of accuracy.     

X-ray spectroscopy of blocked alanine in water solution from supermolecular and supermolecular-continuum solvation models: a first-principles study

The N1s near-edge X-ray absorption fine structure (NEXAFS) and X-ray emission spectra (XES) of blocked alanine in water solution have been investigated at the first-principles level based on cluster models constructed from classical molecular dynamics simulations. The bulk solvent has been described by both supermolecular and combined supermolecular-continuum models. With the former model we show that NEXAFS spectra convergent with respect to system size require at least the inclusion of the second solvation shell and that averaged spectra over several hundreds of snapshots can well represent the statistical effect of different instantaneous configurations of the solvation shells. With the combined model we demonstrate that calculations of a medium-sized peptide-water supermolecule qualitatively predict the NEXAFS spectrum of the solvated peptide even considering a single geometry. Furthermore, sampling over hundreds of snapshots by the combined model, the explicit inclusion of even a few waters yields an averaged spectrum in good quantitative agreement with the discrete model results. In comparison, the XES spectra show little dependence on the structures of either the solvent shell or the peptide itself. The ramifications of these findings are discussed.     

Minimalist explicit solvation models for surface loops in proteins

We have performed molecular dynamics simulations of protein surface loops solvated by explicit water, where a prime focus of the study is the small numbers (e.g., ~100) of explicit water molecules employed. The models include only part of the protein (typically 500 - 1000 atoms), and the water molecules are restricted to a region surrounding the loop. In this study, the number of water molecules (N(w)) is systematically varied, and convergence with large N(w) is monitored to reveal N(w)(min), the minimum number required for the loop to exhibit realistic (fully hydrated) behavior. We have also studied protein surface coverage, as well as diffusion and residence times for water molecules as a function of N(w). A number of other modeling parameters are also tested. These include the number of environmental protein atoms explicitly considered in the model, as well as two ways to constrain the water molecules to the vicinity of the loop (where we find one of these methods to perform better when N(w) is small). The results (for RMSD and its fluctuations for four loops) are further compared to much larger, fully solvated systems (using ~10,000 water molecules under periodic boundary conditions and Ewald electrostatics), and to results for the GBSA implicit solvation model. We find that the loop backbone can stabilize with a surprisingly small number of water molecules (as low as 5 molecules per amino acid residue). The side chains of the loop require somewhat larger N(w), where the atomic fluctuations become too small if N(w) is further reduced. Thus, in general, we find adequate hydration to occur at roughly 12 water molecules per residue. This is an important result, because at this hydration level, computational times are comparable to those required for GBSA. Therefore these "minimalist explicit models" can provide a viable and potentially more accurate alternative. The importance of protein loop modeling is discussed in the context of these, and other, loop models, along with other challenges including the relevance of appropriate free energy simulation methodology for assessment of conformational stability.     

Ionic and dipolar solvation dynamics in liquid water

The solvation time correlation function for solvation in liquid water was measured recently. The solvation was found to be very fast, with a time constant equal to 55 fs. In this article we present theoretical studies on solvation dynamics of ionic and dipolar solutes in liquid water, based on the molecular hydrodynamic approach developed earlier. The molecular hydrodynamic theory can successfully predict the ultrafast dynamics of solvation in liquid water as observed from recent experiments. The present study also reveals some interesting aspects of dipolar solvation dynamics, which differs significantly from that of ionic solvation. Dedicated to Prof. C N R Rao on his 60th birthday     

Role of water structure in thermodynamics of nonpolar solvation

The SSOZ (site-site Ornstein-Zernike) equation with an original closure condition for liquid molecular systems is used to calculate thermodynamic functions of noble gas solvation in water. Water is modeled by two close sets of atom-atomic potential functions. The calculations indicate that the chemical solvation potential is strongly sensitive to water structure. A comparison with experiment is given. Institute of Theoretical and Experimental Biophysics, Russian Academy of Sciences. Translated fromZhurnal Strukturnoi Khimii, Vol. 37, No. 4, pp. 736–741, July–August, 1996. Translated by L. Smolina     

Vibrational relaxation of liquid water in ionic solvation shells

Femtosecond two-color pump–probe spectroscopy is used to measure the vibrational lifetime of the O–H stretch vibration in solutions of KF, NaCl, NaBr, and NaI in HDO:D₂O. We observe a slow component (roughly 2–4 times slower than in HDO:D₂O) in the decay of the absorption change, which is due to O–H groups that are hydrogen bonded to the dissolved anions. The time constant of this slow component depends on the nature of the anions and is observed to decrease with temperature, in contrast with the temperature dependence of the relaxation of the OH stretch vibration in pure HDO:D₂O.     

On the solvation of ions in small water droplets

The solvations of positively and negatively charged model ions in water droplets have been studied using Monte Carlo simulations performed with a polarizable intermolecular potential function model. Special focus has been placed on the position of the ion in the water droplet. It was found that the sign of the ionic charge is of minor importance but an increased ionic charge localizes the ion to the central regions of the droplet, whereas a large polarizability and a large ionic radius favor locations close to the surface of the water droplet.     

A comparative theoretical study of dipeptide solvation in water

Molecular dynamics studies have been performed on the zwitterionic form of the dipeptide glycine-alanine in water, with focus on the solvation and electrostatic properties using a range of theoretical methods, from purely classical force fields, through mixed quantum mechanical/molecular mechanical simulations, to fully quantum mechanical Car-Parrinello calculations. The results of these studies show that the solvation pattern is similar for all methods used for most atoms in the dipeptide, but can differ substantially for some groups; namely the carboxy and aminoterminii, and the backbone amid NH group. This might have implications in other theoretical studies of peptides and proteins with charged -NH(3) (+) and -CO(2) (-) side chains solvated in water. Hybrid quantum mechanical/molecular mechanical simulations successfully reproduce the solvation patterns from the fully quantum mechanical simulations (PACS numbers: 87.14.Ee, 87.15.Aa, 87.15.He, 71.15.Pd).     

Theory for the solvation of nonpolar solutes in water

We recently developed an angle-dependent Wertheim integral equation theory (IET) of the Mercedes-Benz (MB) model of pure water [Silverstein et al., J. Am. Chem. Soc. 120, 3166 (1998)]. Our approach treats explicitly the coupled orientational constraints within water molecules. The analytical theory offers the advantage of being less computationally expensive than Monte Carlo simulations by two orders of magnitude. Here we apply the angle-dependent IET to studying the hydrophobic effect, the transfer of a nonpolar solute into MB water. We find that the theory reproduces the Monte Carlo results qualitatively for cold water and quantitatively for hot water.     

Computational investigation of order, structure, and dynamics in modified water models

Model liquids have been constructed to study the role of local structure in the anomalous properties of liquid water. The intermolecular potentials were modified by increasing the weight of the Lennard-Jones term relative to the electrostatic term in the SPC/E model for water. The resulting family of liquids varies from SPC/E water to a Lennard-Jones-like liquid. Properties were measured as a function of density and temperature. The local structure was described by two order parameters, one measuring the tetrahedral order and the other measuring the translational order. The translational order parameter was found to be large for both tetrahedral and Lennard-Jones liquids, but to go through a minimum as the potentials were modified, demonstrating that the two types of structure are incompatible. Just as in water several properties (e.g., the translational diffusion coefficient, entropy) exhibit anomalous density dependence as a result of the breakdown of local tetrahedrality, we observed nonmonotonic behavior of the translational diffusion constant and reorientational relaxation rate as the fluids were transformed from tetrahedral to Lennard-Jones-like. This is also an indication of the incompatibility between Lennard-Jones and water-like structure.