Fundamental understanding of electrocoalescence in heavy crude oils

Electrocoalescence is used to improve oil-water separation during crude oil processing. Electric fields can assist merger of small water drops into larger ones that will precipitate faster, reducing retention time in sedimentation tanks. Electrocoalescence can either be applied under stagnant conditions in larger sedimentation tanks, or in compact coalescers where there is turbulence or shear in the liquid. Electrocoalescence is an energy efficient process, and helps reducing consumption of chemical emulsion breakers. Electrocoalescence works by applying an electric field on a water-in-oil emulsion. Water drops will be polarized from the applied field, and ? if they are close ? the polarization will produce a strong attractive force between the drops. Shear flow helps bringing drops close together. In two earlier projects, the influence of both voltage shape and frequency, and liquid flow, was studied experimentally in model oils, and CFD models of drop pairs and emulsion behavior were established. Field-enhanced microscopic surface instabilities preceding drop merger was observed. It is evident that surface properties govern the coalescence mechanism, but a specific coalescence criterion remains in the dark. The surface properties of the water drops will vary with crude types, temperature, application of surfactants as emulsion breakers, etc. The present project aimed on one hand to improve the understanding of the coalescence process in a real, black crude oil, and on the other hand, to develop a practical tool for studying how the electrocoalescence could be optimized for a specific crude oil. At a larger scale, electrocoalescence is explained by fluid dynamics and electrostatic forces, but it is not trivial to model the behavior of a drop pair that meets in a shear flow where electrostatic forces pull the drops towards each other and deform their surfaces. At the microscopic level, drop merger is governed by properties of the surface film surrounding the water drops. This film can be characterized by surface tension and viscoelastic properties. In order for a coalescence criterion to be developed, these properties have to be explained; studied experimentally and modelled. A test rig for studying the surface behavior of free electrically stressed water drops in a stagnant crude oil has been developed. Optical observation within "black" crude oils was facilitated by using a long-distance microscope coupled to a fast near-infrared (NIR) camera with a resolution high enough to observe fast microscopic deformations of the surfaces. When a drop in a crude oil is stressed by an electric field, it will be polarized and the electrostatic pressure will stretch and deform the drop. When the stresses are varied, the dynamic properties of the drop surface can be observed and characterized. This technique ? drop stretching using an electric field ? is considered to be an alternative to, or even superior to, the well-known oscillating pending drop experiments used for characterizing water-oil interfaces. The behavior of electrostatically stressed water drops was modelled using computational fluid dynamics (CFD) and the level-set method. The model described surfactants moving on the surface, varying surface tension and Marangoni forces. Good quantitative agreement was obtained between simulations and experiments for model-oil systems with added surfactant. The most important remaining hurdle is the description of certain surface properties, which are very different in crude oils compared to "pure" fluids. We have obtained important initial results using a multi-scale approach developed in this project. In this approach, molecular-dynamics simulations of model asphaltene molecules are employed to determine interface-model parameters used in the CFD model. A technique for producing water drops in oils has been developed. Drops and drop pairs of a desired size in the 10?100 micrometer range can be made in a reproducible way using an electrostatic technique pulling water drops out of a syringe. This facilitates experiments on drops and drop pair of a desired size. From a more practical angle, there is a need to study coalescence efficiency under realistic conditions, where emulsion breakers can be used, temperature increased, and liquid shear and voltage level and frequency varied to optimize separation. To achieve this, a benchtop batch model coalescer was developed. It is inspired from a rheometer, using a rotating bob in a tube to produce shear, but avoiding all problems from drop charging at metallic electrodes encountered in a standard rheometer. The model coalescer allowed variation of all mentioned parameters. The model coalescer is considered to be a practical tool for industry. One is now in a good position to take on the challenge of developing a coalescence criterion, and industry has access to a novel, practical test technique for optimizing electrocoalescence for realistic systems.

Electrocoalescence depends on the success of getting drops to get close enough in turbulence and shear to allow electrostatic forces to pull the drops closer and finally coalesce. In a crude oil this is often hindered by components that stabilize the surf aces of drops in an emulsion. The processes that facilitate and hinder electrocoalescence in a real and dark crude oil is are not yet uncovered. A new high-frame-rate NIR camera enables observations of drop behavior inside crude oil with sufficient resolu tion to support comparison with mathematical modeling. We will study drop pair behavior in shear flow and electric field in a setup where drop pairs can be released and studied optically in a transparent model system to support mathematical electrohydrody namic models. This will be compared with behavior of water-/ crude emulsions in a high- pressure heated flow. The dynamic behavior of the interfacial layers in water drops of a selection of crudes will be studied in a novel electrostatic "tensiometer" and compared with conventional pendant drop studies to allow for comparison with mathematical models and numerical simulations based on the level-set method. The next step is to further develop the models to handle drop pairs with surface coating. Then the behavior of adjacent drops in an electric field will be studied optically to investigate the behavior leading to or not leading to coalescence and compare with the models. Finally we will investigate possibilities for diagnosing the character of an oil-w ater emulsion (oil conductivity, water cut and drop distributions) using dielectric spectroscopy. The oil conductivity is a parameter governing the time constant of a liquid and it is necessary to optimize the ac AC frequency to be used to get an efficien t electric field distribution inside a coalescer with covered electrodes.