Are all-atom any better than united-atom force fields for the description of liquid properties of alkanes?

  • PDF / 2,765,258 Bytes
  • 17 Pages / 595.224 x 790.955 pts Page_size
  • 78 Downloads / 178 Views

DOWNLOAD

REPORT


ORIGINAL PAPER

Are all-atom any better than united-atom force fields for the description of liquid properties of alkanes? Guilherme C. Q. da Silva1 · Gabriel M. Silva2 · Frederico W. Tavares3 · Felipe P. Fleming2 · Bruno A. C. Horta1 Received: 27 May 2020 / Accepted: 14 September 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020

Abstract Alkanes are a fundamental part in empirical force fields (FF) not only due to their technological relevance, but also due to the prevalence of alkane moieties in organic molecules, e.g., compounds containing a saturated carbon chain. Therefore, a good description of alkane interactions is crucial for determining the quality of a FF. In this study, the performance of 12 empirical force fields (FF) was evaluated in the context of reproducing liquid properties of alkanes. More specifically, n-octane was chosen as a reference compound since it is a liquid in a broad temperature range and it has numerous experimental data for thermodynamic, transport, and structural properties, as well as for their temperature dependencies. A normalized root-meansquare deviation (NRMSD) analysis was used to rank the force fields in their ability to reproduce the experimental data. Five out of the six best force fields considered were united-atom models. The GROMOS force field showed the smallest deviation in terms of NRMSD, followed by TRAPPE-EH, NERD, CHARMM-UA, TRAPPE-UA, and OPLS-UA. This overall better performance of the united-atom force fields indicates that complexity does not always bring quality. Keywords Force field · Molecular dynamics · Alkanes · Octane · Liquid properties

Introduction Classical molecular simulation techniques have been frequently used to describe the behavior of molecular systems, helping to describe natural phenomena and guiding the design of novel materials and compounds. A critical aspect defining the quality of the molecular simulation is the empirical force field (FF) used to describe the interactions between the particles composing the system. The FF encompasses the functional form and its associated parameters, both of which are derived on the basis of the philosophy adopted by the FF developers. Usually, a comparison of different force fields reveals minimal differences in terms of

This article belongs to the Topical Collection XX - Brazilian Symposium of Theoretical Chemistry (SBQT2019) Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00894-020-04548-5) contains supplementary material, which is available to authorized users.  Bruno A. C. Horta

[email protected]

Extended author information available on the last page of the article.

the functional form, but significant differences in terms of the parameters. These differences arise from the choices made during the calibration procedure: system representation (degrees of freedom), molecular systems of interest, target properties, thermodynamic conditions, and optimization strategy. Ultimately, the reliability of a simulation result will depend on how