Why different water models predict different structures under 2D confinement
Related Experiment Video
Updated: Feb 5, 2026

A Protocol for Computer-Based Protein Structure and Function Prediction
Published on: November 3, 2011
James Dix1, Leo Lue2, Paola Carbone1
1School of Chemical Engineering and Analytical Sciences, University of Manchester, Manchester M13 9PL, United Kingdom.
Abstract:
Experiments of nanoconfined water between graphene sheets at high pressure suggest that it forms a square ice structure (Algara-Siller et al., Nature, 2015, 519, 443). Molecular dynamics (MD) simulations have been used to attempt to recreate this structure, but there have been discrepancies in the structure formed by the confined water depending on the simulation set-up that was employed and particularly on the choice of water model. Here, using classical molecular dynamics simulations, we have systematically investigated the effect that three different water models (SPC/E, TIP4P/2005 and TIP5P) have on the structure of water confined between two rigid graphene sheets with a 0.9 nm separation. We show that the TIP4P/2005 and the TIP5P water models form a hexagonal AA-stacked structure, whereas the SPC/E model forms a rhombic AB-stacked structure. Our work demonstrates that the formation of these structures is driven by differences in the strength of hydrogen bonds predicted by the three water models, and that the nature of the graphene/water interaction only mildly affects the phase diagram. Considering the available experimental data and first-principle simulations we conclude that, among the models tested, the TIP4P/2005 and TIP5P force fields are for now the most reliable when simulating water under confinement. © 2018 Wiley Periodicals, Inc.
Related Concept Videos
Predicting Molecular Geometry
States of Water
Water freezes when the intermolecular forces are greater than the kinetic energy. Unlike most other substances, water is less dense in its solid state than in its liquid state. This is because each water molecule can form...
Prediction Intervals
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
Protein and Protein Structure
A protein's shape is critical to its function. For example, an enzyme...
Water and Mineral Acquisition
Structure of Benzene: Kekulé Model
He proposed that benzene has a cyclic structure of six carbon atoms attached to one hydrogen atom each, with three alternating pi bonds.

