Session F45: Modeling the electrochemical interface and aqueous solutions I

Addressing electrified water-metal interfaces with Non-Equilibrium Green's Functions

The TranSIESTA method and code [1,2] were developed within the SIESTA project [3] to study problems involving steady-state non-equilibrium problems in nanoscale costrictions, where an external electric bias is applied between the two sides of the constriction, establishing a steady electric current. Non-equilibrium Green's Functions are used there to solve the problema, as they can deal with open, non periodic systems (leads plus nanoconstriction) driven out of equilibrum (external applied bias). This machinery can be also used to study electrified solid/liquid interfaces [4], where an external bias is applied to the solid electrode. Here, one is not concerned with the quantum electronic transport, but with the effect of the external bias on the properties and the chemical reactions induced at the metal/liquid interface. I will show examples of application of this idea, as a proof of concept for future realistic, atomistic first-principles simulations of electrochemical processes.

References:
[1] M. Brandbyge, J.L. Mozos, P. Ordejón, J. Taylor, and K. Stokbro, Phys. Rev. B 65, 165401 (2002).
[2] N. Papior, N. Lorente, T. Frederiksen, A. García, M. Brandbyge, Computer Physics Communications 212, 8 (2017).
[3] J.M. Soler, E. Artacho, J.D. Gale, A. García, J. Junquera, P. Ordejón, D. Sánchez-Portal, J. Phys.: Condens. Matter., 14, 2745 (2002).
[4] L.S. Pedroza, P. Brandimarte, A.R. Rocha, M.V. Fernandez-Serra, Chem. Sci., 62, 9 (2018).   

First principles simulations of electrified silicon/water interfaces

Si-based materials have been used in a myriad of devices, including light-activated systems such as cathodes in photoelectrochemical cells and p-i-n junctions in optoelectronic biomodulators. In both cases, Si surfaces are in contact with water, and the interface is under bias (e.g. under the effect of an electric field). Here we report a study of electrified Si/water interfaces aimed at understanding charge transfer mechanisms between the solid and the liquid, and their influence in determining the performance of optoelectronic devices. In particular, we carried out ab-initio molecular dynamics simulations of the hydrogenated Si(100)/water interface using the Qbox code (http://qboxcode.org), and we investigated the effect of an applied electric field on the band offsets at the interface. We compared our results with patch clamp measurements and we further investigated the modification of the structural properties of water at the interface, induced by an applied bias.   

Ion Pairing in Liquid Water: An Ab Initio Study

Potentials of mean force are widely used to study solvation characteristics as a function of a particular reaction coordinate of interest. In particular, ion-pair association in water is of great importance to many scientific and industrial applications, but is not widely understood. Here we present a detailed study of NaCl solvation properties in the dilute limit. We examine in detail the sources of discrepancies between simulation results, with the objective to separate approximation and statistical errors from physical interpretations. It is found that the choice of water model and simulation time step has a notable effect on the solvation state stability. Moreover, solution density also has a large effect on energetically favorable configurations. We show how the underlying description of liquid water has important consequences on the inter-ionic potential of mean force, and how the solute ions affect the solution's equilibrium properties. Our results shed light into how ions in water affect its structure and correlate with low density and high density liquid fluctuations.   

Integrating First-Principles Simulations with Electrochemical Experiments: Towards a Realistic Description of Aqueous Interfaces

Improved understanding of electrochemical interfaces is critical for a wide variety of emerging applications, such as hydrogen production, supercapacitors and water desalination. In this talk, we will discuss how first-principles simulations can be integrated with in-situ experiments to understand physicochemical properties of several representative electrochemical systems. We will present our studies of aqueous solutions at graphitic interfaces, where we show that structure and electrical response of the interfaces is governed by a complex interplay between bias potential, intrinsic electronic properties of the electrode, and specific ion effects-including ion hydration and charge transfer. In addition, we will discuss how a combination of first-principles molecular dynamics simulations, many-body perturbation theory, and X-ray photoelectron spectroscopy can provide insights into the relationship between interfacial structure, electronic properties of semiconductors and their reactivity in aqueous solutions.

This work was performed under the auspices of the U.S. DOE by LLNL under Contract DE-AC52-07NA27344, and is supported by the U.S. DOE, Office of Energy Efficiency and Renewable Energy, Fuel Cell Technologies Office via HydroGEN consortium   

Stability and reactivity at solid/liquid interfaces studied by ab initio calculations

Processes at solid-liquid interfaces are at the heart of many present day technological challenges related to the improvement of battery materials, electro-catalysis, fuel cells, corrosion and others. Obtaining the microscopic information needed to describe and quantify the underlying fundamental mechanisms is equally challenging to experiment and theory. Density functional theory (DFT) calculations are able to resolve processes at the microscopic scale. However, the modelling of electrochemical systems is particularly challenging. The main reason is the presence of different classes of materials and phenomena such as metal electrodes, liquid water, huge electric fields within the same system.
We discuss how by using DFT calculations for polar ZnO(0001) surfaces Pourbaix diagrams can be constructed. These diagrams provide direct insight into the role the aqueous electrolyte plays in shaping surfaces and the high selectivity of solvation effects [Phys. Rev. Lett. 120, 066101 (2018)]. Going beyond the thermodynamic description, we utilize our novel potentiostat design to study reactions at electrochemical solid/liquid interfaces under controlled bias [Phys. Rev. Lett. 120, 246801 (2018)]. Focusing on one of the most corrosive system under wet corrosion, the new approach is shown to solve a problem that puzzled corrosion scientists for more than 150-years: what is the underlying mechanism of the experimentally observed link between H-evolution under anodic conditions and Mg dissolution.
  

First-principles study of the hydrogen-bonding network in water at the biased electrode interface

A crucial step for the development of next-generation electrochemical devices will be the understanding of the atomic and electronic structures of interfacial waters next to biased electrode surfaces. In this presentation, carrying out first-principles non-equilibrium electronic structure calculation within the multi-space constrained-search density functional theory (MS-DFT) formalism we have recently developed, we study the bias-dependent structural and electronic properties of the hydrogen-bonding network of water molecules at the gold electrode interface. Benchmarking the non-equilibrium force profile of a single water molecule next to the gold electrode obtained from non-equilibrium Green’s function (NEGF) calculations, we confirm the practical equivalence between MS-DFT and DFT-NEGF. We also report the advantages of MS-DFT in view of electrochemical device simulations by providing (1) well-defined binding energies and (2) the electrochemical potential profiles. Analyzing the spatial profiles of the electrochemical potentials or quasi-Fermi levels at the gold-water interface at varying bias voltages, we extract several important insights into the nature of hydrogen bond network of liquid water at the biased electrochemical interfaces.   

Voltage-Dependent Cluster Expansions for Modeling Catalytic Electrochemical Interfaces in Solution Environments

The cluster expansion method provides the framework to represent functions of lattice configurations. We use this method to model energetics of solid-solution interfaces in electrochemical systems. Semi-local density functional theory calculations with implicit solvation are used to obtain effective cluster interactions that include local relaxations and the effects of the solvent. These calculations, in conjunction with electrochemical double layer models, provide the framework for voltage-dependent cluster expansions in a model electrochemical environment. We use these voltage-dependent cluster expansion to model faradaic charging at electrode-electrolyte interfaces, equilibrium configurations of surface alloys, and surface promotion in near-surface alloys. Preliminary voltage-dependent cluster expansion models of skin alloys suggest that amount of catalytically active sites highly sensitive to the applied potential.   

Multiscale modeling of solvation effects in the Oxygen Evolution Reaction on TiO2

Development of sustainable energy generation and storage technologies able to satisfy the needs of modern society represents a major challenge for scientific research. Innovation here essentially rely on hydrogen (H) /oxygen (O) evolution reactions (ER) and /or oxidation of chemical fuels. OER generally represents the bottleneck of (photo)catalytic water splitting, as it requires high overpotentials. This has motivated an impressive search for sustainable high-performance electrocatalysts for OER, which identified TiO₂. Recently the microscopic mechanism of OER at TiO₂–water interface has been deeply studied. In particular, first principles simulations of rutile TiO₂ in explicit water [J. Phys. Chem. C, 123, 18567 (2019)] found the solvent to affect electrostatically the energetics at the interface, rather than to modify the H bond network. Such results pave the way for the use of implicit and hybrid solvation techniques in ab initiosimulations of this system to further elucidate the effect of the solvent on OER, and specifically on its rate limiting step. To this aim, we exploit state-of-the-art implicit solvation schemes for condensed matter simulations, as implemented in the Environ plugin[http://www.quantum-environ.org] for Quantum ESPRESSO[http://www.quantum-espresso.org].   

Ionic Structure in Dense Electrolytes Confined by Interfaces

Recent surface force measurements have shown that the effective force between mica surfaces does not decay as sharply as predicted by mean-field models when the concentration of the confined electrolytes is high (around 2 M for NaCl). Motivated by these experiments, we use molecular dynamics simulations to extract the ionic structure in aqueous electrolytes confined by two interfaces. Ionic density profiles and contact-region densities are extracted for electrolytes confined between uncharged, charged, and polarizable interfaces at different concentrations with implicit-solvent models. Simulation results show that the net charge density in the contact region near charged surfaces exhibits a distinct trend with increase in interface separation for highly concentrated electrolyte systems (around 2 M for model NaCl) compared to electrolytes at low salt concentration. This behavior is further probed by varying ion sizes, interface charge density, ion valency, and interface polarizability, as well as by computing ion-ion pair correlation functions near the interface. Effects of explicit solvent molecules on the observed ionic structure are also discussed.   

The Electrokinetic Transport of Multivalent Electrolytes: The Effect of Charge Inversion

Devices integrated by nanoconduits hold great potential for clinical and biochemical analysis due to amplified sensibility, faster response and increased portability. In nanoconduits, wherein the electrical double layer may occupy a considerable part of the channel, the hydrodynamics of multivalent electrolytes is highly influenced by interfacial electrokinetic phenomena, such as charge inversion (CI). We conduct atomistic simulations of an electrolyte solution which consists of water as solvent, chlorine as co-ion and different counter-ions, i.e., sodium, magnesium and aluminum. We model Electroosmotic (EOF), Poiseuille (PF) and Couette (CF) flow in silica nanochannels to probe the relation between CI and transport properties. In EOF, we observe that changes induced by CI in the electrokinetic driving force at the diffuse layer, significantly alter the velocity distributions. Moreover, cases of CF and PF flow show that the position of the shear plane is significantly altered by the presence of CI. We find that the nanoconfined electrolytes can be modeled as two immiscible fluids with different transport properties with the shear plane as dividing surface.