We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cond-mat.soft

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Condensed Matter > Soft Condensed Matter

Title: Salt parameterization can drastically affect the results from classical atomistic simulations of water desalination by MoS$_2$ nanopores

Abstract: Water scarcity is a reality in our world, and scenarios predicted by leading scientists in this area indicate that it will worsen in the next decades. However, new technologies based in low-cost seawater desalination can prevent the worst scenarios, providing fresh water for humanity. With this goal, membranes based in nanoporous materials have been suggested in recent years. One of the materials suggested is MoS$_2$, and classical Molecular Dynamics (MD) simulation is one of the most powerful tools to explore these nanomaterials. However, distinct Force Fields employed in MD simulations are parameterized based on distinct experimental quantities. In this paper, we compare two models of salt that were build based on distinct properties of water-salt mixtures. One model fits the hydration free energy and lattice properties, the second fits the crystal density and the density and the dielectric constant of water and salt mixtures. To compare the models, MD simulations for salty water flow through two nanopores sizes were used -- one pore big enough to accommodate hydrated ions, and one smaller in which the ion has to dehydrate to enter, and two rigid water models from the TIP4P family -- the TIP4P/2005 and TIP4P/$\epsilon$. Our results indicate that the water permeability and salt rejection by the membrane are more influenced by the salt model than by the water model, especially for the narrow pore. In fact, completely distinct mechanisms were observed, and they are related to the characteristics employed in the ion model parameterization. The results show that not only the water model can influence the outcomes, but the ion model plays a crucial role.
Comments: 15 pages, 17 figures
Subjects: Soft Condensed Matter (cond-mat.soft); Mesoscale and Nanoscale Physics (cond-mat.mes-hall); Computational Physics (physics.comp-ph)
DOI: 10.1039/D0CP00484G
Cite as: arXiv:2002.00882 [cond-mat.soft]
  (or arXiv:2002.00882v1 [cond-mat.soft] for this version)

Submission history

From: João Pedro Kleinubing Abal M.Sc. [view email]
[v1] Mon, 3 Feb 2020 16:49:22 GMT (2729kb,D)

Link back to: arXiv, form interface, contact.