Is the liquid water surface basic or acidic? Macroscopic vs. molecular-scale investigations

摘要
Many physical phenomena are affected by the intrinsic acidity/basicity of the free liquid water surface, yet it remains an active and controversial subject. Macroscopic bubble and droplet experiments have been interpreted to indicate an air–water interface covered with hydroxide, whereas recent molecularscale studies produce the opposite conclusion, viz. that hydroxide is repelled from the interface while hydronium is strongly adsorbed. Here we report results from resonant UV second harmonic generation (SHG) experiments that are best modeled by surface depletion of hydroxide and establish at most a weak surface adsorption. This finding is consistent with our earlier SHG measurements indicating surface enhancement of hydrated protons, as well as with other molecular-scale experiments and simulations, but stands in stark contrast to the results from macroscopic studies. The acidity, or basicity, of aqueous surfaces could strongly influence heterogeneous atmospheric chemical processes, such as aerosol reactions and gas uptake.

SHG方法能够从微观的角度研究界面离子浓度，他们研究组之前的UV SHG研究空气水界面的工作中发现了hydrated protons在表面的富集。但是这和一些宏观研究方法的结果似乎有些矛盾。UV SHG可以直接和溶液中的CTTS共振，Hydroxide exhibits a broad charge-transfer-to-solvent (CTTS) transition centered at 187 nm in the bulk [59]，从而可以研究溶液界面的离子，We have taken a different approach and applied SHG in the UV to directly probe the interfacial ions. At the UV wavelengths used in these experiments, water possesses no resonances, whereas several anions exhibit strong charge-transfer-to-solvent (CTTS) transitions [59].

本文题目是“水的界面是酸性还是碱性？”
At 225 nm, the SHG intensity remains constant, within experimental uncertainty, due to the limited concentration range explored. At 200 nm, however, the SHG intensity increases significantly at high bulk concentration. Since such a bulk concentration-dependence increase is not observed at the non-resonant wavelengths, this constitutes a direct measurement of surface hydroxide ions at these high concentrations. However, this does not in itself imply surface enhancement of hydroxide. As described above, a small fraction of hydroxide will always exist at the interface despite a repulsive Gibbs free energy due to thermal fluctuations.

讨论
Our direct experimental support for weak hydroxide and strong hydronium [16] propensities for the water surface agrees with the other molecular-level investigations, as described above, but are in stark contrast to the interpretations of macroscopic bubble and droplet electrophoresis experiments [20,21,23–27]. Those experiments measure the motion of gas bubbles or oil droplets in aqueous solution when an electric field is applied. The motion of the droplets/bubbles is analyzed in terms of their zeta-potential and is interpreted as a measure of the overall charge of the droplet/bubble. For neutral and basic pH, the derived zeta-potential is negative, implying that the droplets and bubbles carry a negative surface charge, and only at low pH is a positive zeta-potential observed, i.e. the isoelectric point is around pH 3–4. Such findings appear universal for hydrophobic interfaces and have been interpreted as clear evidence for a strong surface enhancement of hydroxide [20,21,23–27]. However, those experiments only show that the bubbles and droplets behave as though they are negatively charged, moving in a solvent bath that would have to be positively charged, and not directly establishing that the outermost liquid layer is covered by hydroxide.