Interfacial Charge DOI: 10.1002/ange.201108228 Why Are Hydrophobic/Water Interfaces Negatively Charged?** Kevin Roger* and Bernard Cabane Interfaces between water and apolar media (gases, liquids, or solids) have a high cost in free energy. Therefore they tend to recombine to reduce the total interfacial area: in water, oil drops coalesce and air bubbles recombine following collision. The metastability of emulsions, foams, and polymer disper- sions is achieved through adsorption of amphiphilic molecules (ionic or non-ionic), macromolecules, or particles, which block the recombination. The mechanisms of this stabilization are well understood. [1] Yet very fine emulsions made of pure oil droplets in pure water have also been found to be metastable in the absence of any added stabilizers. [2–4] According to sum frequency generation (SFG) spectroscopy [5] and electrophoretic mobility measurements, [2, 3] the droplets of these surfactant-free emulsions are ionized and carry a negative electrical charge. Similar results have been reported for the water/air interface. [6–8] Moreover, this negative charge increases rapidly with pH value and therefore with the bulk concentration of hydroxide ions. [2, 4, 9] The most frequent explanation given for these phenom- ena is that hydroxide ions adsorb at hydrophobic/water interfaces. While consistent with the pH signature of these phenomena, this explanation requires high adsorption ener- gies, more than 20 times the thermal energy k B T (about 50 kJ mol  ), [2, 6] and an outstanding selectivity of hydroxide ions over other simple anions [2, 6, 10, 11] that do not adsorb at such interfaces. On the theoretical side, some models attempt to account for this unexpected adsorption, [12–14] while others find no accumulation of hydroxide ions at hydrophobic interfaces; [15] still, other models look for another origin of the surface charge. [16] At present, there is no clear and straightforward understanding of this intriguing phenom- enon. The basic assumption of all previous experimental and theoretical studies has been that these systems have “pristine” oil/water interfaces, that is, oil molecules in contact with water molecules, although the possibility of contamination by anionic, surface-active impurities has been mentioned. [17] This assumption is supported by the use of pure components (99 %) with additional purification, thoroughly cleaned glass- ware and equipment, inert atmospheres, and good reprodu- cibility of the data. Moreover, the same variation with pH value of the surface charge has been found by different research groups with various experimental systems. However, the question remains whether these results leave no other choice but to accept a specific adsorption of hydroxide ions at hydrophobic interfaces? We propose that an alternative mechanism could be a reaction of hydroxide ions with species that are confined to the interface. We conducted systematic experiments to find out whether such alternative explanations could account for all the published experimental results. We used emulsions obtained through a solvent-shifting method. [18–20] Solutions of hexa- decane, of different purities, were prepared at a constant volume fraction of 10 3 in 99.9% pure acetone. They were then mixed with a much larger volume ( 20) of Milli-Q water. The supersaturated solution separated spontaneously, yielding nanometer-sized hexadecane droplets (average diameter approximately 150–200 nm). This process is partic- ularly well suited to the study of the stability of hydrophobic/ water interfaces, because the droplets grow through a recom- bination mechanism that ends when their interfaces acquire a sufficient number of stabilizing ions. Hence, both final size and polydispersity of the nanometer-sized droplets decrease upon increasing the concentration of stabilizing ions. The final sizes of the droplets can also be controlled through the concentration of oil in acetone and the rate of addition of the acetone/oil solution into water. We performed three types of experiments using these emulsions. First, we measured the electrophoretic mobility of oil droplets in emulsions made with different oils, as a function of the pH value of the aqueous phase (Figure 1). In these experiments, the emulsification process was adjusted so that the average droplet size was the same in each emulsion. The only difference between the three emulsions was the compo- sition of the oil. The first emulsion was made with 99% pure hexadecane in water at pH 6, ionic strength I = 10 3 mol L 1 (sodium chloride), and with a mean droplet diameter was 170 nm. Then we varied the pH value and measured the electro- phoretic mobility of the oil droplets, using a Malvern Zetasizer. We found that the pH variation (Figure 1), reproduced the signature of the surface-charging process described in previous studies. [2, 4] The mobility was close to zero at pH 2–3, increased in magnitude until pH 8, and subsequently remained constant. If this variation was due to the adsorption of hydroxide ions, previously used models immediately show that their adsorption free energy would be above 20 k B T . [2, 6] We checked that the addition of other small anions (halides or carboxylates) did not produce any changes in the electrophoretic mobility, as had been reported. [6, 11] Therefore the change in surface charge detected is specific to the concentration of hydroxide ions. [*] K. Roger, Dr. B. Cabane PMMH, CNRS UMR 7636, ESPCI 10 rue Vauquelin, 75231 Paris cedex 05 (France) E-mail: kevin.roger@espci.fr Dr. B. Cabane Theoretical Chemistry, Lund University 222100 Lund (Sweden) [**] This work was supported by ANR 2010 BLAN 942 03 “LimOuzIne”. We thank F. Ganachaud, B. Jçnsson, G. Karlstrçm, and H. Wennerstrçm for useful discussions and input and C. Labbez and D. Bouttes for their help on the pH-dependence calculation of the surface potential. A ngewandte Chemi e 1 Angew. Chem. 2012, 124,1–5  2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim These are not the final page numbers! Ü Ü