No abstract

International audienceWe study an unstable highly concentrated emulsion of water droplets in oil with a nonionic surfactant. A technique of light diffusion coupled to a rheometer allows simultaneous measurement of average droplet radius R and emulsion shear elastic modulus G' during time. Over the studied range of volume fraction (from 71 to 95%), we show that Princen and Kiss' (J Colloid Interface Sci 112: 427-437, 1986) model does not apply. A dimensional analysis based on the hypothesis of dominant van der Waals forces is proposed for nonionic surfactants, which is in good agreement with experimental data. We also show that the measured average droplet volume increases linearly with time and that the coalescence rate strongly depends on the volume fraction in relation with different topological conformations of droplets

Emulsions are a type of metastable colloid composed of two or more immiscible liquids. These systems are widely used in a variety of applications, such as cosmetics, drug delivery, food, etc. Although there exist theoretical foundations which offer insights into these systems, industry practices often favor empirical methods. In this work a multiscale approximation is used for the study of water-in-oil (W/O) emulsions. This approach allows for the analysis of interrelationships among macroscopic, microscopic, process, and formulation variables. Additionally, the emulsions were modeled with Computational Fluid Dynamics (CFD), which permitted a better understanding of the role process variables plays. It was possible to establish relationships among incorporated energy, elastic modulus, mean droplet diameter, and stability measurements. In addition, differences in impeller geometry were found to have an effect in the aforementioned variables. Finally, the CFD model allowed for the observation of gradients in relative viscosity, droplet diameter, and dispersed phase volume fraction.

Important differences arise when chemical demulsification strategies are implemented for heavy crude oils (°API ∼ 10). Traditional methods for screening and selecting an appropriate demulsifier based on bottle tests and lipophilic–hydrophilic parameters (i.e., HLB, RSN, and so on) tend to be less adequate because of the almost negligible density difference between the oil and the water phases. This situation leads to a detriment of the separated water often mixed with undesired dense-packed layers (DPLs) and emulsion layers. In this work, dehydration of heavy crude oil emulsions from a Colombian oilfield was assessed through the use of a wide range of chemical demulsifiers of different functionalities. Through the use of bottle tests and transmission/backscattering measurements, it was shown that the demulsification mechanisms involved in these limiting cases (low density difference) are different. Hence demulsifiers with functional groups that have traditionally performed very well for lighter oils fail when applied to the heavy crude oil cases. Poly­(ethylene oxide)/poly­(propylene oxide) block copolymer-based products (PEO/PPO) do not seem to have the ability to penetrate the asphaltene network/film at the liquid–liquid interface (separated water, <17%) while the alkylphenol-aldehyde resins seem to prevent the formation of DPLs/emulsion layers possibly through polar interactions, yielding a good quality water phase after separation.

We studied the energy consumption per unit volume during preparation of highly concentrated water-in-oil emulsions in a two-step semibatch process. In particular, we studied the effect of two process variables, the water addition flow rate (Q w ) and the agitation speed (N). The oil used for emulsion preparation was n-dodecane, the surfactant was sorbitan monooleate (Span 80) and deionized water was used for the dispersed phase. The results obtained showed that two steps were required to get complete incorporation of the dispersed phase and a homogeneous and stable gel-emulsion. With the help of independent physical characteristics measurement performed at the end of the preparation process, we established two functional relationships, relating storage modulus (G′) with energy consumption (E v ), (G′ ∝ E v 0.6 ), and average liquid cell size (R m ) of the dispersed phase with energy consumption (E v ), (R m ∝ E V -0.3 ). A structural scaling law can be deduced that relates the elastic modulus to the reciprocal of the square of the average liquid's cell size of the gel-emulsion, corroborated by independent measurements and predicted by recent published models.