English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Simulation techniques based on accurate and efficient representations ofpotential energy surfaces are urgently needed for the understanding of complexaqueous systems such as solid-liquid interfaces. Here, we present a machinelearning framework that enables the efficient development and validation ofmodels for complex aqueous systems. Instead of trying to deliver aglobally-optimal machine learning potential, we propose to develop modelsapplicable to specific thermodynamic state points in a simple and user-friendlyprocess. After an initial ab initio simulation, a machine learning potential isconstructed with minimum human effort through a data-driven active learningprotocol. Such models can afterwards be applied in exhaustive simulations toprovide reliable answers for the scientific question at hand. We showcase thismethodology on a diverse set of aqueous systems with increasing degrees ofcomplexity. The systems chosen here comprise bulk water with different ions insolution, water on a titanium dioxide surface, as well as water confined innanotubes and between molybdenum disulfide sheets. Highlighting the accuracy ofour approach with respect to the underlying ab initio reference, the resultingmodels are evaluated in detail with an automated validation protocol thatincludes structural and dynamical properties and the precision of the forceprediction of the models. Finally, we demonstrate the capabilities of ourapproach for the description of water on the rutile titanium dioxide (110)surface to analyze the structure and mobility of water on this surface. Suchmachine learning models provide a straightforward and uncomplicated butaccurate extension of simulation time and length scales for complex systems.

Machine learning potentials for complex aqueous systems made simple