An Internally Consistent Method for the Molecular Dynamics Simulation of the Surface Tension: Application to Some TIP4P-Type Models of Water

Theoretical and experimental OD-stretch vibrational spectroscopy of heavy water

In view of the current situation in which the OD-stretch vibrational spectra have been scarcely computed with non-polarizable rigid D2O models, we investigate the IR and Raman spectra of D2O by using a newly-reported model TIP4P/2005-HW. From the comparison between the calculations and experimental data, we find the excellent performance of TIP4P/2005-HW for vibrational spectroscopy of D2O in the same manner as TIP4P/2005 for H2O, although one may still conveniently employ an alternative method that regards OH as putative OD to calculate the OD-stretch spectra with similar quality from TIP4P/2005 trajectories. We also demonstrate that the appropriate setting for the spectral simulation of D2O under the time-averaging approximation reflects the slower dynamics (i.e., slower motion of translation and rotation due to the heavier mass and stronger hydrogen bond) of D2O than H2O. Moreover, we show from the theoretical calculations that the established interpretation of the OH-stretch spectra of H2O is finely applicable to the OD-stretch of D2O.

A Deep Potential model for liquid-vapor equilibrium and cavitation rates of water

Computational studies of liquid water and its phase transition into vapor have traditionally been performed using classical water models. Here, we utilize the Deep Potential methodology-a machine learning approach-to study this ubiquitous phase transition, starting from the phase diagram in the liquid-vapor coexistence regime. The machine learning model is trained on ab initio energies and forces based on the SCAN density functional, which has been previously shown to reproduce solid phases and other properties of water. Here, we compute the surface tension, saturation pressure, and enthalpy of vaporization for a range of temperatures spanning from 300 to 600 K and evaluate the Deep Potential model performance against experimental results and the semiempirical TIP4P/2005 classical model. Moreover, by employing the seeding technique, we evaluate the free energy barrier and nucleation rate at negative pressures for the isotherm of 296.4 K. We find that the nucleation rates obtained from the Deep Potential model deviate from those computed for the TIP4P/2005 water model due to an underestimation in the surface tension from the Deep Potential model. From analysis of the seeding simulations, we also evaluate the Tolman length for the Deep Potential water model, which is (0.091 ± 0.008) nm at 296.4 K. Finally, we identify that water molecules display a preferential orientation in the liquid-vapor interface, in which H atoms tend to point toward the vapor phase to maximize the enthalpic gain of interfacial molecules. We find that this behavior is more pronounced for planar interfaces than for the curved interfaces in bubbles. This work represents the first application of Deep Potential models to the study of liquid-vapor coexistence and water cavitation.

Transferability of vibrational spectroscopic map from TIP4P to TIP4P-like water models

We computed the IR, Raman, and sum frequency generation spectra of water in the OH-stretch region by employing the quantum/classical mixed approach that consists of a vibrational spectroscopic map and molecular dynamics (MD) simulation. We carried out the MD simulation with the TIP4P, TIP4P/2005, and TIP4P/Ice models and applied the map designed for TIP4P by Skinner et al. to each MD trajectory. Although the map is not tuned for TIP4P-like models, TIP4P/2005 and TIP4P/Ice provide the best reproduction of the experimental vibrational spectra of liquid water and crystalline ice, respectively. This result demonstrates the transferability of the map from TIP4P to TIP4P/2005 and TIP4P/Ice, meaning that one can choose an appropriate TIP4P-like model to calculate the vibrational spectra of an aqueous system without rebuilding the map.

Free-energy differences of OPC-Water and SPC/HW-Heavy-Water Models using the Bennett Acceptance Ratio

Surface tension, vapor density of OPC-water and SPC/HW-heavy-water models have been estimated at low temperatures using the scaled model. The free-energy difference, -ΔF, of n-molecules and (n-1)-molecules plus a free probe has been calculated using the Bennett acceptance ratio with the aid of Monte-Carlo simulations. Our results show that the relation between the free-energy difference divided by kBT and the number of molecules to the power minus one-third is linear for n>6. Consequently, the surface tension can be extracted from the straight line slope, whereas the vapor density can be extracted from the intercept, which is proportional to the logarithmic ratio of liquid density to that of vapor density. By scaling the free-energy differences, for at least three different temperatures, to TCT−1, we estimated the critical temperature and hence the surface tension and the vapor density at a wide range of temperatures. The free-energy differences have been calculated at 240K, 260K, and 280K for OPC-water, and at 260K, 280K, and 300K for the SPC/HW-heavy water model.

Thermophysical properties of water using reactive force fields

The widescale importance and rich phenomenology of water continue to motivate the development of computational models. ReaxFF force fields incorporate many characteristics desirable for modeling aqueous systems: molecular flexibility, polarization, and chemical reactivity (bond formation and breaking). However, their ability to model the general properties of water has not been evaluated in detail. We present comprehensive benchmarks of the thermophysical properties of water for two ReaxFF models, the water-2017 and CHON-2017_weak force fields. These include structural, electrostatic, vibrational, thermodynamic, coexistence, and transport properties at ambient conditions (300 K and 0.997 g cm^-3) and along the standard pressure (1 bar) isobar. Overall, CHON-2017_weak predicts more accurate thermophysical properties than the water-2017 force field. Based on our results, we recommend potential avenues for improvement: the dipole moment to quadrupole moment ratio, the self-diffusion coefficient, especially for water-2017, and the gas phase vibrational frequencies with the aim to improve the vibrational properties of liquid water.

Unexpected finite size effects in interfacial systems: Why bigger is not always better-Increase in uncertainty of surface tension with bulk phase width

We present an unexpected finite size effect affecting interfacial molecular simulations that is proportional to the width-to-surface-area ratio of the bulk phase L_l/A. This finite size effect has a significant impact on the variance of surface tension values calculated using the virial summation method. A theoretical derivation of the origin of the effect is proposed, giving a new insight into the importance of optimising system dimensions in interfacial simulations. We demonstrate the consequences of this finite size effect via a new way to estimate the surface energetic and entropic properties of simulated air-liquid interfaces. Our method is based on macroscopic thermodynamic theory and involves comparing the internal energies of systems with varying dimensions. We present the testing of these methods using simulations of the TIP4P/2005 water forcefield and a Lennard-Jones fluid model of argon. Finally, we provide suggestions of additional situations, in which this finite size effect is expected to be significant, as well as possible ways to avoid its impact.

Molecular dynamics investigation on adsorption layer of alcohols at the air/brine interface

Alcohols are a significant group of surfactants which have been employed extensively in industry to improve the interfacial effects. Recently, the change in surface potential (ΔV) of two isomeric hexanols, methyl isobutyl carbinol (MIBC) and 1-hexanol, was investigated by using an ionizing (241)Am electrode. It clearly showed the opposite effects between MIBC and 1-hexanol in the interfacial zone: one enhanced the presence of cations, whereas the other enhanced the presence of anions. This study employs molecular dynamics simulation to provide new insights into the interactions between alcohol molecules and ions as well as water at the molecular level. The results qualitatively agreed with the experimental data and verified the significance of MIBC branching structure on the molecular arrangement within the interfacial zone. The results also highlighted the role of the second water layer on the interfacial properties.

Synergistic adsorption of MIBC/CTAB mixture at the air/water interface and applicability of Gibbs adsorption equation

The synergistic adsorption of a binary surfactant mixture was investigated by tensiometry and neutron reflectometry. The results directly contradicted the conventional Gibbs adsorption equation. The accompanied molecular simulation demonstrated a multilayer arrangement at the synergic conditions, with three distinctively oriented water layers. The positive synergism can be explained by considering the relationship between water orientation and surface tension, in a similar manner to Langmuir's proposal in 1920s. In spite of the supporting evidence, the relationship has not been quantified in literature. The molecular orientation and arrangement are not included in the current theoretical framework, which simplifies the adsorbed zone into a single monolayer. A new theoretical framework is needed to properly quantify the interfacial adsorption for the mixed surfactant systems.

COMPUTER SIMULATION OF LIQUID–VAPOR INTERFACIAL TENSION: LENNARD-JONES FLUID AND WATER REVISITED

We review several commonly used simulation methods for computing liquid–vapor surface tension and associated theoretical treatments of the long-range correction for inhomogeneous systems. Prototype model systems considered in this review are the Lennard-Jones (LJ) fluid and the SPC/E model water. In addition, we examine a variety of factors that can affect calculation of the surface tension γ via the mechanical approach (i.e. using either KB or IK method). It is found that for the LJ fluid, the size of simulation box and the number of particles in the system can have notable effects on the computed surface tension. For SPC/E water, the Ewald parameters can influence computed surface tensions (γ) as well, e.g., very small Ewald parameters tend to overestimate γ. It is also found that the IK method consistently gives γ that are 0.6 - 0.9 mN/m greater than γ computed based on the KB method. When computing the first reciprocal–space contribution to the surface tension, the Ghoufi's strategy gives rise to more sensible profile of pressure difference PN(z)-PT(z) than the Alejandre's strategy although both strategies result in nearly the same average surface tension through the integration of PN(z)-PT(z).

The surface tension of TIP4P/2005 water model using the Ewald sums for the dispersion interactions

The liquid-vapor phase equilibria and surface tension of the TIP4P/2005 water model is obtained by using the Ewald summation method to determine the long range Lennard-Jones and electrostatic interactions. The method is implemented in a straightforward manner into standard simulation programs. The computational cost of using Ewald sums in dispersion interactions of water is estimated in direct simulation of interfaces. The results of this work at 300 K show a dramatic change in surface tension with an oscillatory behavior for surface areas smaller than 5x5sigma(2), where sigma is the Lennard-Jones oxygen diameter. The amplitude of such oscillations substantially decreases with temperature. Finite size effects are less important on coexisting densities. Phase equilibria and interfacial properties can be determined using a small number of water molecules; their fluctuations are around the same size of simulation error at all temperatures, even in systems where the interfaces are separated a few molecular diameters only. The difference in surface tension of this work compared to the results of other authors is not significant (on the contrary, there is a good agreement). What should be stressed is the different and more consistent approach to obtain the surface tension using the Ewald sums for dispersion interactions. There are two relevant aspects at the interface: An adsorption of water molecules is observed at small surface areas and its thickness systematically increases with system size.