What the Diffuse Layer (DL) Reveals in Non-Linear SFG Spectroscopy

Abstract

Following our recent work [Phys. Chem. Chem. Phys. 20:5190–99 (2018)] that provided the means to unambigously define and extract the three water regions at any charged interface (solid–liquid and air–liquid alike), denoted the BIL (Binding Interfacial Layer), DL (Diffuse Layer) and Bulk, and how to calculate their associated non-linear Sum Frequency Generation Spectroscopy (SFG) χ2(ω) spectroscopic contributions from Density Functional Theory (DFT)-based ab initio molecular dynamics simulations (DFT-MD/AIMD), we show here that the χDL2(ω) signal arising from the DL water region carries a wealth of essential information on the microscopic and macroscopic properties of interfaces. We show that the χDL2(ω) signal carries information on the surface potential and surface charge, the isoelectric point, EDL (Electric Double Layer) formation, and the relationship between a nominal electrolyte solution pH and surface hydroxylation state. This work is based on DFT-MD/AIMD simulations on a (0001) α–quartz–water interface and on the air–water interface, with various surface quartz hydroxylation states and various electrolyte concentrations. The conclusions drawn make use of the interplay between experiments and simulations. Most of the properties listed above can now be extracted from experimental χDL2(ω) alone with the protocols given in this work, or by making use of the interplay between experiments and simulations, as described in this work.

Spectroscopic BIL-SFG Invariance Hides the Chaotropic Effect of Protons at the Air-Water Interface

“…Four independent simulations are performed, each separate trajectory starts from a random configuration of liquid water put in contact with the air, and the cation is added randomly at the interface. The electronic set-up follows our previous works [22,31] on air-water interfaces (neat and in presence of electrolytes). The simulations have been performed with the CP2K package [45,46], the BLYP [47,48] functional is adopted in combination with mixed Gaussian-Plane Waves basis sets and GTH pseudopotentials [49].…”

“…Once the interface is charged, the interpretation of the SFG signal starts to be more tricky as this spectroscopy now probes an interfacial region of far larger thickness than the BIL discussed above, [20,21,[28][29][30][31] where the final SFG signal is now a convolution of second (χ 2 (ω)) and third order (χ 3 (ω)) non-linear contributions, plus interferences [20,21,[28][29][30][31]. Tian et al have shown that any charged interface is composed of two distinct interfacial regions that are SFG active [20]: the Binding Interfacial Layer (BIL), i.e., the first water monolayer as shown in our recent works [21,31] (which is the only layer probed at neutral interfaces), and the Diffuse Layer (DL) made of the subsequent bulk water layer(s) reoriented by the surface field (at charged interfaces). The DL is made SFG active because the surface field induced liquid water reorientation breaks the liquid centrosymmetry.…”

“…The DL is made SFG active because the surface field induced liquid water reorientation breaks the liquid centrosymmetry. The final SFG signal at any charged interface is therefore: χ (2) (ω) = χ (2) BIL (ω) + χ (2) DL (ω) = χ (2) BIL (ω) + χ (3) bulk (ω)∆Φ DL (1) where the DL contribution is the product of the third order non-linear susceptibility of bulk water by the electric potential across the DL [20,21,[28][29][30][31].…”

“…However, this is easily achievable from MD simulations as soon as the two BIL and DL water layers can be defined and extracted from the simulations. This is what we have done in the recent past, providing definitions of the BIL and DL layers from the point of view of the water molecules only (definitions based on three structural descriptors of water) at any charged interface (interface between water and solid/or air), hence providing the means to theoretically calculate χ (2) BIL (ω) and χ (2) DL (ω) spectroscopic SFG signals, individually, by including only the identified water molecules that belong to each water layer into the calculation [21,31]. With these two individual signals in hand, one is able to give a clear assignment of the experimental SFG signatures into contributions arising from these distinct water layers.…”

“…As revealed in ref. [31], while the χ (2) DL (ω) signal does not carry relevant information on the water interfacial structure per se, it does carry other highly relevant information, such as the direct knowledge of the electric field across the DL, of the EDL (Electric Double Layer) formation when electrolytes are included in the aqueous bulk, of the direct measure of the aqueous pH and of the surface charge. While the χ (2) DL (ω) signal carries no direct information on the interfacial water structure, its final contribution to the total χ (2) (ω) signal can be non negligible (depending on the surface electric field) and it has to be removed from the total SFG signal in order to reveal the only relevant structural interfacial information to be assigned, and solely contained in the χ (2) BIL (ω).…”