Abstract
We present an ensemble of 16 independent first-principles molecular dynamics simulations of water performed using the Strongly Constrained and Appropriately Normed (SCAN) meta-generalized gradient approximation exchange-correlation functional. These simulations were used to compute the structural and electronic properties of liquid water, as well as polarizabilities, Raman and infrared spectra. Overall, we find that the SCAN functional used at a simulation temperature of 330 K provides an accurate description of the structural and electronic properties of water while incurring a moderate computational cost. The availability of an ensemble of independent simulations provides a quantitative estimate of the uncertainty in computed structural and electronic properties. Results are also compared with a similar dataset generated using the Perdew, Burke, and Ernzerhof exchange-correlation functional at a temperature of 400 K. All simulation data and trajectories are available at http://quantum-simulation.org.
Highlights
We found that the SCAN330 simulations had an average of 3.621 with a standard deviation of 0.11, compared to PBE400 which had an average of 3.566 and a standard deviation of 0.12.19 This indicates that using the Strongly Constrained and Appropriately Normed (SCAN) functional at 330 K leads to an overall slightly stronger intermolecular bonding strength than the PBE functional at 400 K
Throughout our analysis, we found that the SCAN meta-generalized gradient approximation (GGA) functional provides a good description of the structural and electronic properties of liquid water at a moderate computational cost
While it is difficult to clearly separate the parts of electronic correlations that are included in either functional, SCAN as a metaGGA has additional flexibility to satisfy constraints in reproducing known exact results
Summary
Molecular dynamics (MD) simulations of water have been the subject of numerous studies for nearly 50 years, beginning in the 1970s with the work of Rahman and Stillinger. Modeling of water using classical molecular dynamics has been refined over time with the development of increasingly accurate force fields, an area of research that is still very active. First-principles quantum mechanical simulations have become more readily available with the degree of computing power available today and with the development of density functional theory (DFT) coupled with ab initio molecular dynamics (MD) simulations.The accuracy of DFT simulations has progressively improved over time as new exchange-correlation functionals have been introduced. The simplest density functional is the local density approximation (LDA) functional that does not accurately describe hydrogen bonds, leading to poor chemical accuracy. The hybrid PBE0 functional was shown to improve structural properties beyond that of PBE, and the addition of van der Waals to PBE0 provides an even greater increase in accuracy.. There is evidence that van der Waals functionals can be used to increase the accuracy of more complicated functionals such as the meta-GGA Strongly Constrained and Appropriately Normed (SCAN) functional (SCAN+rVV10).35 This is not necessarily improving all properties. The importance of nuclear quantum effects (NQEs) in the description of water was discussed as early as 1985.40,41 The ad hoc approach of increasing simulation temperature to account for NQEs has been discussed by Ruiz Pestana et al.42 We adopt this approach in the present work despite its shortcomings, due to the high cost of more accurate treatments of NQEs, and in order to allow for direct comparisons with previous work. Despite the large body of results published on first-principles simulations of water, a quantitative measure of the uncertainty of the results due to the statistical nature of molecular dynamics simulations has rarely been provided
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
