Long Molecular Wires and the Auto-ionization of Water

19 September 2024, Version 1
This content is a preprint and has not undergone peer review at the time of posting.

Abstract

Water auto-ionization is critical in a wide range of chemical, biological, physical, and industrial processes. In this work, we describe a series of hitherto unknown collective molecular processes leading to auto-ionization. Specifically, by combining machine-learned interatomic potentials and spectral adaptive biasing force techniques, we determine the relevant free energy landscape of water auto-ionization. At ambient conditions, the free energy profile reveals two distinct saddle points, each leading to the formation of three- and four-member water wires. The wires feature an individual Zundel ion and a proton diffusion-like transition state, respectively. At elevated temperatures, the auto-ionization process exhibits a more concerted hydrogen transfer mechanism and reveals an alternative pathway involving the synchronous diffusion of Zundel ion pairs, with the ion pair corresponding to an energetic local minimum on the free energy surface. These findings help resolve long-standing conflicting views of the mechanism of water auto-ionization and provide new avenues for the study of proton behavior in different aqueous environments.

Keywords

Water
Machine learning
Enhanced sampling
Reaction mechanism

Supplementary materials

Title
Description
Actions
Title
Supporting Information
Description
The document includes details on first principles calculations, MLIP-MD simulations with enhanced sampling, and the training of machine-learned interatomic potentials (MLIPs). Additionally, it covers the generation of free energy surfaces, the calculation of minimum free energy paths and transition states using CI-NEB, and the analysis of water wire distributions. Validation of MLIP calculations is also presented, alongside a calculation of pKw and the free energy profile for the early stages of water autoionization. Finally, it discusses the training data set used for the model.
Actions

Comments

Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.