Stabilization of Water-in-Water Emulsions by Nanorods Karthik R. Peddireddy, † Taco Nicolai,* ,† Lazhar Benyahia, † and Isabelle Capron ‡ † LUNAM Universite ́ du Maine, IMMM UMR-CNRS 6283, 72085 Cedex 9 Le Mans, France ‡ UR1268 Biopolyme ̀ res, Interactions et Assemblages, INRA, F-44316 Nantes, France ABSTRACT: Water-in-water (W/W) emulsions formed by mixing incompatible water-soluble polymers cannot be stabilized with molecular surfactants. However, they can be stabilized by particles through the so-called Pickering effect. Recently, it was shown that its stabilization can be achieved also with nanoplates. Here, we show for the first time that even nanorods in the form of cellulose nanocrystals (CNCs) can efficiently stabilize W/W emulsions. Static light scattering and confocal microscopy techniques were used to determine the surface coverage by CNCs. In the presence of 50 mM NaCl very weak gels were formed by excess CNCs in the continuous phase. In this way creaming of the dispersed phase could be arrested. The nontoxicity, sustainability, and low cost of CNCs and the abundant availability of cellulose render these nanorods potentially highly suited for preparing W/W emulsions. A water-in-water (W/W) emulsion is commonly formed when two incompatible hydrophilic polymers are mixed in water above certain threshold concentrations with each phase enriched with one of the polymers. 1 W/W emulsions play an important role in different areas such as green chemistry, 2 cell biology, 1b and food. 1c,3 Compared to oil-in-water (O/W) emulsions, W/W emulsions have very low interfacial tensions (1 μN/m to 1000 μN/m) and a large interfacial thickness (at least several nanometers). Therefore, molecular surfactants cannot stabilize these emulsions. Until recently, the only way to avoid macroscopic phase separation was by gelling one or both phases. However, it has recently been shown that colloidal particles can stabilize W/W emulsions by forming a layer at the interface, which reduces the free energy. 4 The stabilization of interfaces with particles is known as the Pickering effect and has been intensively investigated for O/W emulsions. 3,5 The change of the free energy (ΔG) by particle adsorption depends strongly on the radius of the particles (R), the contact angle (θ), and the interfacial tension (γ) of the system π γ θ Δ =− −| | G R (sphere) (1 cos ) 2 2 (1) For example if R = 100 nm, θ = 90°, and γ = 10 μN/m, the reduction in the free energy due to the adsorption of a particle at the interface is 75 kT. This shows that adsorption of the particles can be practically irreversible even for W/W interfaces. Recently, it has been shown that also gibbsite discs with a radius of 170 nm and a thickness of 7 nm adsorb at the interface and can stabilize W/W emulsions. 4d As the thickness of these discs is comparable with the interfacial thickness, the interfacial area covered by the disks is independent of the contact angle. It was shown by Vis et al. 4d that the gain in free energy for the adsorption of a disc with radius R is given by π γ θ Δ =− −| | G R (disc) (1 cos ) 2 (2) Here we address the question whether even nanorods can stabilize W/W emulsions. We investigated this issue using cellulose nanocrystals (CNCs) that have a highly anisotropic rectangular parallelepiped structure with average dimensions 160 nm × 6 nm × 6 nm (see Figure 1a). CNC is a very promising material for applications, especially in the fields of packaging and healthcare due to its biodegradability and nontoxicity. 2 The model W/W emulsion that was used consisted of mixtures of dextran and poly(ethylene oxide) (PEO). Received: December 30, 2015 Accepted: February 5, 2016 Figure 1. (a) Transmission electron microscopy (TEM) image of CNCs. (b) Phase diagram of the polymers used for the experiments adapted from ref 4a. The solid line indicates the binodal, and the dashed lines indicate tie lines. The colored symbols indicate the compositions of the emulsions studied here. Letter pubs.acs.org/macroletters © XXXX American Chemical Society 283 DOI: 10.1021/acsmacrolett.5b00953 ACS Macro Lett. 2016, 5, 283-286