Liquid/Solid interfaces: Structure and Dynamics from Spectroscopy and Simulations
CECAM-HQ-EPFL, Lausanne, Switzerland
Liquid-solid interfaces play an important role in a number of phenomena encountered in biological, chemical and physical processes. Surface-induced changes of material properties are not only important for the solid support but also for the liquid itself. In particular it has been shown that water at the interface is substantially different from bulk water, even in proximity of apparently inert surface, such as a simple metal . The complex chemistry at solid-liquid interfaces is fundamental to heterogeneous catalysis and electrochemistry and has become especially topical in connection with the search for new materials for energy production. A quite remarkable example is the development of cheap yet efficient solar cells, whose basic components are dyes molecules grafted to the surface of an oxide material and in contact with an electrolytic solution.
In life science, the most important solid-liquid interfaces are the cell-membrane/water interfaces. Phenomena occurring at the surface of phospholipid bilayers control the docking of proteins, the transmission of signals as well as transport of molecules in and out of the cell. More recently the development of bio-compatible materials has lead to research on the interface between bio-compatible material and lipid/proteins in aqueous solution.
Gaining microscopic insight in the processes occurring at solid-liquid interfaces is therefore fundamental to a wide range of disciplines. Recently, a deeper insight has been gained thanks to major advances in both experimental techniques and computer simulations.
On the experimental side, surface specific techniques, such as non-linear optical spectroscopy (Sum Frequency Generation Spectroscopy (SFG) and Second Harmonic Generation (SHG)), surface sensitive X-ray scattering, in situ Scanning tunneling microscopy(STM) and infrared reflection absorption spectroscopy have permitted to gain information on layers of nanometric thickness at the interface. On the other hand it is quite clear that the experiments require theoretical modelling in order to dissect the results and to rationalize the different factors that contribute to the interfacial properties. In this respect molecular dynamics simulations and in particular first principles simulations can prove a major tool to accurately describe both structure and reactivity in a consistent way. Progresses in first principles simulations in the past 3-5 year shave been enormous. In particular efficient treatment of basis set and long range interactions have permitted to extend the simulation to hundred-thousand atoms, which now allows to tackle realistic models for interfaces, maintaining first principles quality. First-principles simulations have been successfully used to describe spectroscopic properties of liquids throughout the past decade providing very good agreements with experiments . This sets the central topic of our workshop which aims to bring together scientists from computational physics and chemistry and experimentalists of interfaces. This workshop will provide the framework for people from both sides to describe the state of-the-art developments in their respective fields. It will give people the possibility to interact and trigger new collaborations between theoreticians and experimentalists. Bringing together communities ranging from solid-state, surface science, liquid-state and biochemistry will permit to exchange ideas and methodologies, with the purpose to create a network of collaborations between these different communities, and therefore provide new advances in the “interfacial domain”. Our aim in organizing this workshop is to achieve a new microscopic understanding of complex phenomena occurring at different solid-liquid interfaces and to answer some of the many open questions that are described below.
Key topics of the workshop will include recent progresses on:
i) vibrational sum frequencies generation, a non linear vibrational spectroscopy technique which selectively addresses interfaces. We will invite pioneering experimentalists and computational scientists who in the last few years have tried to provide a computational approach to the interpretation of these spectra;
ii) structure of water at hydrophilic and hydrophobic surfaces, where metal oxides and membranes are two prime examples we would like to discuss. In this context, we will address the accuracy of the current available methodologies and their transferability;.
iii) biomimetic materials, where the emphasis will be on the interaction between biological molecules and materials.
Structural changes of water properties at interfaces.
The structure of liquids near interfaces has come under increasing scrutiny over the past decade. One motivation has been the need to understand the properties of fluids, both simple and complex, in confined environments where the interfacial fluid properties might dominate. It is only recently that the molecular-scale structures of these interfaces have been directly observed for any system. Examples include the observation of structuring at mineral/water interfaces,[3–6] and simple molecular fluids  and in confined geometries . A general observation from these studies is that the fluid density profile is generally modified in the presence of an interface. Recent results show that this effect can be amplified or suppressed in confinement while shearing the fluid . Water is one of the most interesting fluids because of the influence of hydrogen bonding that results in many anomalies with respect to other simple fluids, and because of the fundamental role it plays in chemical and biological systems. Significant questions remain concerning the nature of the interfacial water structure and its dependence upon the substrate. For instance, studies of conducting surfaces under potentiometric control revealed evidence for “ice-like” layers at room temperature [10–12]. Additional studies have shown various degrees of water ordering at mineral/water interfaces and in the absence of potential control.
Quite recently SFG experiments have also pointed to ’different’ species of water at different interfaces [13–16] The theoretical interpretation of these experiments is still controversial. This is in part due to the challenging accuracy that is required to explain theoretically the features in the spectra. On one hand, quantum mechanical (QM) approaches would appear most suitable, on the other hand the limited statistical sampling which is accessible with such accuracy poses severe limitations to its application. One issue is the infamous signature at 3200cm-1 which seems to characterize several interfaces of water, but with different microscopic interpretations regarding the water properties. Although major advances have been realized by several computational groups in the last years [17–21] no final consensus has emerged. Furthermore, also with the QM realm the accuracy of DFT functional for the description of the interface is still questionable. Open issues concern the description for example of dispersion forces. Experimentalists who are the developers of the SFG methodology are invited to the workshop for a state-of-the-art presentation of the methodologies and ongoing developments and for the purpose of conducting useful discussions with the theoreticians on the future theoretical developments to be conducted and how to organize more efficiently the interplay between experiments and theory in the near future.
Another topic we wish to address is the role of the hydrophobic vs hydrophilic interactions in modulating the water properties [26–28]. The extreme hydrophilic case of charged surfaces is discussed in a very recent paper which suggest that there is a no-ice behavior for water close to highly charged surfaces,  where the water has been shown to display a significantly different structure at the highly charged interfaces in comparison to neutral/slightly charged surfaces.29 Tahara and coworkers suggest for instance that the surface electric field does not induce the ice-like structure even though it enhances orientational order along the field. This is a hot-topic as the ice-like/liquid-like behaviours of interfacial water dominates the debates in the field. A solid in contact with water often reacts, and in the case of metal oxides it develops a net charge because of different protonation states of the surface groups. The same phenomena is well known to occur at a membrane in contact with a physiological solution, where the membrane is responsible for establishing and maintaining a proton gradient between the internal and external part of the cell. The proteins embedded in the membrane and on the membrane surface can have different protonation state and thus determine local changing.
For oxides in contact with water this charging process controls the sorption of ions affecting the chemical reactivity of the metal oxide surface. The surface charge density at a given pH is determined by the acidity of surface groups, by the surface composition and by the electrostatic potential difference across electrical double layers and therefore by the structure of the electrical double layer itself. Recently, we have contributed to the understanding of the surface acidity of a few oxides at the water interfaces using a DFT-based MD simulation approach to calculated the free energies for the insertion/deletion of protons.6,  On the other end in the biological membranes, the charging process is responsible for the membrane potentials, which in turn regulates ions concentration and transport across the membrane. The same issue affecting the electrical double layer in mineral solids are also present for biological membranes. It is notoriously difficult to disentangle all these factors using only experimental data and here the use of modelling and atomistic simulation plays a crucial role [31, 32]. Biomaterial: protein adsorption on solid surfaces. The adsorption of proteins on solid surfaces is of pivotal importance in the fields of bio-sensing, bio engineering, and life sciences. A detailed knowledge and understanding of the forces and the interactions that determine the amount of adsorbed proteins is key to tailoring surfaces for further applications of importance in the “bio-materials” domain. It is therefore important to study the interactions between artificial interfaces and proteins in liquid environments and to determine the relevance and influence of the forces that are involved [33, 34]. Self-assembled monolayers (SAMs) are one class of synthetic surfaces that has attracted a significant amount of interest in this context due to their ability to resist the non-specific adsorption of proteins. The vibrational spectroscopic characterization of biomolecules adsorbed on a surface has gained considerable interest in the recent years. A well-established technique in this regard is infrared reflection-absorption spectroscopy (IRRAS) which allows obtaining high-quality infrared spectra of monolayer of organic molecules deposited on metal surfaces. A characteristic feature of this method is the surface selection rule, which states that only the vibrations with a component of the transition dipole moment aligned perpendicular to the surface plane can interact with the incident light and contribute to the infrared spectrum [35-39].
In this way, IRRAS also provides information about the orientation of the surface adsorbed molecules. The inherent surface selection rule as well as the sensitivity renders this method particularly suitable for the structural and spectroscopic characterization of self-assembled monolayer (SAMs) which have important applications in molecular electronics and catalysis. We will therefore address the most recent advances in the experimental domain on these systems and address the question of what theoretical modelling is the most suited in order to achieve both structural and spectroscopic properties of these big systems at the interfaces. The interplays between first principles simulations, atomistic classical simulations, coarse-grained and more mesoscopic simulations will also be addressed in the discussions.
Marie-Pierre Gaigeot (LAMBE UMR8587 Université d'Evry val d'Essonne, Université Paris-Saclay) - Organiser
Marialore Sulpizi (Johannes Gutenberg University) - Organiser & speaker