Physica A 392 (2013) 567–582 Contents lists available at SciVerse ScienceDirect Physica A journal homepage: www.elsevier.com/locate/physa A model for molecular emulsions: Water and ‘‘weak water’’ mixtures B. Kežić, R. Mazighi, A. Perera ∗ Laboratoire de Physique Théorique de la Matière Condensée (UMR CNRS 7600), Université Pierre et Marie Curie, 4 Place Jussieu, F75252, Paris cedex 05, France article info Article history: Received 12 July 2012 Received in revised form 14 September 2012 Available online 1 November 2012 Keywords: Aqueous mixtures Water models Micro-heterogeneity abstract The SPC/E water model is mixed with three ⟨⟨weaker⟩⟩ versions of it, obtained by reducing the initial partial charges by multiplicative factors of 1/3, 2/3 and 4/5, respectively, while keeping the same diameter as water, and adjusting only the energy parameters such as to keep each neat substance in a dense liquid phase under ambient conditions. These models cover the observed behaviour of many realistic aqueous mixtures, ranging from demixing (the 1/3 model) to fully mixed hydrophobic-like (2/3) or hydrophilic-like (4/5) situations, the latter both showing strong and weak micro-heterogeneity, respectively. The simulations show that micro-segregation arises even when all constituents have the same length scale, under the sole influence of hydrogen bonding interactions. However, this micro-heterogeneity itself introduces a second length scale by producing domain oscillations in the distribution functions in the nanometer range, that can be captured by making a formal analogy with micro-emulsions. This approach explains the origin of the anomalously large Kirkwood–Buff integrals, often obtained in simulations of realistic aqueous mixtures, as a transient behaviour in the domain range. The analogy can be used to calculate the correct integrals by accounting for domain statistics, without the need to perform expensive large scale simulations. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Liquids are fundamentally disordered systems, excluding the large class of liquid crystalline materials. Considered from this sole point of view, water and aqueous mixtures would be on a par with argon or other such simple liquids, as well as their mixtures. However, water is characterised by a strong tetrahedral order, and since this order is not global—such as in crystalline ice, for example—water is considered as a ‘‘complex’’ disordered liquid [1]. Yet, the nature of the order in water, and in particular tetrahedral order, remains a controversial subject. Two approaches seem to be prevalent in this matter. One is based on ideas inspired from the earlier work of Franks [2], that water is patched with domains of locally ordered hydrogen-bonded water molecules, and could even be a binary mixture of an ordered and disordered forms of the same liquid, and that a corresponding liquid–liquid critical point might be hidden in the low-temperature/high-density part of the phase diagram [3]. The other type of approaches infers that most properties of water and mixtures can be explained by the smallness of the water molecule, when compared to most types of molecules, in the sense that size effects are of leading importance [4–7]. In these latter approaches, the hydrogen bonding property has clearly a secondary importance. The role played by the geometry and the magnitude of the tetrahedral interactions in many natural substances such as water and silica, for example, has been studied by Angell and co-workers [8]. The importance of tetrahedrality in the charge distribution of water has been recently examined by Lynden-Bell and co-workers [9–11], where they varied the geometry of the charge distribution while keeping the charges the same. In the present work, we take a similar approach, and examine the relative importance of the magnitude of the charges that produces the hydrogen-bond (H-bond) interaction ∗ Corresponding author. E-mail address: aup@lptmc.jussieu.fr (A. Perera). 0378-4371/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.physa.2012.10.027