WO2006043819A1

WO2006043819A1 - Method and device for inducing coalescence in emulsions to facilitate subsequent removal of water from the emulsion

Info

Publication number: WO2006043819A1
Application number: PCT/NO2005/000369
Authority: WO
Inventors: Lars Esben Lundgaard; Gunnar Berg
Original assignee: Sintef Energiforskning As
Priority date: 2004-10-08
Filing date: 2005-10-03
Publication date: 2006-04-27
Also published as: NO20044265L; NO20044265D0

Abstract

Method and device for imparting coalescence in water- in-oil emulsions by supplying an electric voltage and thereby facilitate subsequent separation of water drops. A substantially bipolar, square wave shaped AC voltage which is substantially symmetrical around the voltage zero value is supplied to the emulsion. The frequency is adapted to the current electrodes and the current coalescer in a manner that ensures a capacitive voltage distribution.

Description

Method and device for inducing coalescence in emulsions to facilitate subsequent removal of water from the emulsion

The present invention concerns a method for inducing the merging of drops of water in an emulsion (coalescence) by applying an electric voltage across the emulsion to simplify subsequent separation of water from the emulsion. According to another aspect the invention concerns an electrocoalescer for performing the method.

Background

Oil/ water separators are used to separate water from oil continuous emulsions. In dealing with oil continuous emulsions (water-in-oil emulsions) an electric field applied to the emulsion will cause water drops to coalesce. The water drops become larger and therefore settle at a higher rate allowing use of smaller settling tanks.

The first electro-separator was patented by F.G. Cottrell in 1908 (US patent 895,729) for particles in a gas. Later similar techniques have been widely used to separate water from emulsions of crude oil and water. Patents have been issued e.g. to Prevost et al. (US 5,647,981 and 5,643,469). Eow provides an overview of the prior art technology in two articles: J.S. Eow, M. Ghadiri, A.O. Sharif, T. Williams: "Electrostatic enhancement of coalescence of water drops in oil: A review of the current understanding", Chem.Eng.J., 84 (3), 2001, pp 173-192. J.S. Eow, M. Ghadiri: "Electrostatic enhancement of coalescence of water drops in oil: A review of the technology", Chem.Eng.J., 85, 2002, pp 357-368.

US patent 4,804,453 describes an electrical system to achieve coalescence by which a decreasing field from the inlet side towards the outlet side of the coalescer is desired, since such arrangement is believed to enhance the separation of water from the emulsion.

WO 03/049834 correspondingly describes equipment for separation of water from emulsions by coalescence. The particular feature of this design is the method by which the electrode plates in a vessel are connected to a low voltage power supply outside the vessel so that all parts having a high voltage are isolated from the environment in an as compact way as possible.

Until now direct voltage or alternating voltage at 50/60 Hz has mainly been used for coalescence but also sinusoidal alternating voltages at other frequencies and various pulsed direct voltages have been proposed, cf. e.g. US patent 5,385,658. Often electrodes covered with solid insulation have been used (WO 03/049834) to prevent short circuits in the presence of water plugs.

Equipment for coalescence is usually designed as large tanks holding substantially stagnant emulsions. The companies Kvaerner and ABB have created a new generation of electrocoalescers utilizing alternating current and turbulent flow, cf. US patent 6,136,174. When alternating current is applied the drops become small dipoles between which strong forces occur when the shear in the emulsions brings one drop close to another.

The electro coalescence process for a water-in-oil emulsion is not yet completely understood. The force effect on charged particles in an electric field (electrophoresis) is known as well as on polarized particles (dielectrophoresis). The process of coalescence as such - interpreted as the occurring events when two drops merge to become one single larger drop, is not yet explained, though processes like film thinning between drops and chain formation are believed to be involved.

Lundgaard et al have - based on theories from meteorology - developed a model in which the electrocoalescence process is explained as the result of surface instability occurring above a critical field strength for two drops being adjacent to one another (L. Lundgaard, G. Berg, A. Pedersen, P.J. Nilsen: "Electrocoalescence of water drop pairs in oil", 14^th Int. Conference on Dielectric liquids (ICDL), Graz, Austria, July 7-12, 2002). This theory has later been experimentally investigated and it has been discovered that the instability mainly occurs by deformation of the larger drop which has a lower internal pressure. Figure 1 shows how the larger drop is drawn towards the smaller one. The effect is quite similar to what takes place when a single drop bursts in a high electric field as shown by Berg et al (G.Berg, L.Lundgaard, M.Becidan, R.S.Sigmond: "Instability of Electrically stressed water drops in oil", 14^th Int. Conference on Dielectric liquids (ICDL), Graz, Austria, July 7-12, 2002.) cf. Figure 2.

For coalescers applying DC voltage charge decay from charged drops will normally be a problem since the oil phase has a high conductivity. In AC coalescers this is less significant because the voltage induces a polarization within the drops. In "Electrocoalescence of water drop pairs in oil" the relevance of the frequency of the AC voltage for the voltage distribution in an electrocoalescer with insulated electrodes is discussed: At low frequencies and when using DC voltage the voltage distribution is determined by the resistivity of the materials (solid insulation used for electrode covering has high resistivity while an emulsion has low resistivity) and a resistive voltage distribution results. In such a case the voltage is located across the solid insulation and only a minor voltage is located across the gap where it is supposed to act on the emulsion. This reduces the efficiency of the coalescer. When the frequency is increased the voltage distribution is controlled by the capacitance or the permittivity of the materials and a capacitive voltage distribution results. Since the permittivity of a water-in-oil emulsion and the permittivity of an electrode covering insulation is of the same order of magnitude, the voltage over the emulsion will then increase compared to what was the case when the distribution was resistive. With covered electrodes it is thus necessary to use fields that are rapidly changing (high frequency) to obtain a capacitive voltage distribution resulting in strong fields in the emulsion, cf. Figure 3.

The lack of understanding of the coalescence mechanism and the general physics of the phenomena has lead to a number of unresolved questions in relation to which voltage shapes are most effective. An optimum alternating voltage has been sought, and when direct voltage has been used it has sometimes been found necessary to pulse it. Rapidly changing voltages like pulsed DC may provide the same benefits as high frequency AC, but only during the transient. Control of e.g. voltage supply in a coalescer has been made based on the overall efficiency of the process and there has not been any good criteria based on dielectric or surface chemistry analyses of the emulsion to be treated. The consequence the electrocoalescers have generally been of low efficiency.

The objective of the present invention has been to provide a method that allows electro- coalescence for separation of water from water-in-oil emulsions to take place with a higher efficiency so that larger volumes of emulsions may be treated within a certain period of time or so that the equipment may be downsized for a certain required rate of separation.

A further objective of the invention has been to provide a method that is well suited for AC coalescers and particularly of the kind with covered electrodes, thereby avoiding the problems particularly related to resistive voltage distribution.

Finally the method should be technically simple to carry out. The invention

The invention concerns a method as defined by claim 1. According to another aspect the invention concerns an electrocoalescer as defined by claim 7. Preferred embodiments of the invention are disclosed by the dependent claims. The essential feature of the present invention is to utilize an ac voltage shaped as a substantially bipolar pure square wave.

The frequency of the square wave can be adapted to obtain a capacitive voltage distribution and thus full voltage over the emulsion, ensuring a high electric field around the drops in the emulsion. The force or thrust effect on the drops in a neutral emulsion is not dependent on polarity which means that the forces have the same direction and magnitude during positive and negative half cycles. The forces between drops are the same as with a DC voltage of the same amplitude, supplied to emulsion where the oil phase has good insulating properties but without the negative effects of DC voltage with respect to resistive voltage distribution and possibly of charge relaxation.

Below embodiments of the invention will be described in further detail with reference to the enclosed drawings.

Figure 1 is a schematic illustration of two drops in an electric field.

Figure 2a and /2b show how a water drop changes its shape in an electric field.

Figure 3 is a diagram showing the relation between critical voltage and drops radius for a single drop.

Figure 4 is a schematic comparison of conditions between coalescence with sinusoidal AC and AC in the form of square waves.

Figure 1 shows two drops of water in a vertical electric field. The deformation of the drop with larger radius of curvature and less internal pressure is easily observed. If the voltage exceeds a critical limit the deformation of the larger drop will lead to an unstable condition and the larger drop will eventually be "injected" into the smaller one.

Figure 2a shows how a round water drop (uppermost) in a horizontal electrical field will stretch from the induced dipole moment in the water (middle). Beyond a critical limit the drop becomes unstable which is what is desired in a coalescence process.

Figure 2b illustrates how a critical voltage for a drop is reduced with increasing drop radius. The lowermost graph is for DC and square waves and the middle one for 50 Hz AC. It is easily recognized that a square wave shaped voltage supply allows lower field strength for instability than a sinusoidal voltage supply. The same can be observed with respect to critical voltage level for coalescence between two drops.

Figure 3 shows a scheme diagram of the voltage distribution in a coalescer with covered electrodes. The emulsion and the electrode insulation may be described by a capacitance and a resistance in parallel. At low frequency and for DC the voltage distribution will follow the solid-drawn line and there is thus no voltage drop between the two drops. On the other hand, at high frequency AC the voltage distribution will follow the dotted line and the drops will be positioned at different voltage levels.

Figure 4 depicts the conditions for a pair of drops at a critical voltage level for coalescence. A sinusoidal voltage supply will have a field strength above the critical field strength only a part of the time, while a square wave with sufficient amplitude will have a field strength above the critical field strength all the time. The sinusoidal voltage will in addition provide an (unnecessary) over- voltage part of the time.

In practice the critical field strength for coalescence must be determined empirically with the emulsion in question and the actual coalescer to be used.

The advantages of using a bipolar square wave shaped voltage for coalescence are several:

1. Contrary to a sinusoidal voltage the square wave voltage is high all the time so provided it is above the critical level it will be above all the time. Two drops passing close to each other will therefore coalesce contrary to what might be the case when a sinusoidal AC voltage is used and the voltage may be below the critical level the moment the particles pass by each other. See Figure 4.

2. The voltage in the system may be held lower than with a coalescer with a sinusoidal voltage that has to be supplied with a high voltage to obtain a voltage above the critical level most of the time. Generally speaking the higher the RMS value of a voltage with a given peak value is the more efficient the coalescer will be. For a sinusoidal voltage shape the RMS value is approximately 0.7 while for a square wave shaped alternating voltage it is approximately 1. This means that the potential for improvement is about 40 %.

The voltages may conveniently be in the range 1 to 40 kV — of course dependent on dimensions (electrode separation). Frequencies in the range 10 to 10 000 Hz may be convenient. Technically square waves may be realized by means of transformers, rectifiers, and capacitors where the voltage is turned on/ off mechanically or possibly with a solid state switch. Alternatively power electronics may be used. Rise-time for the square wave voltages depends on impedances and capacitances in the circuit. Since the process depends on the voltage level it is not sensitive to changes in rise-time or ripple (deviations from ideal pulse shape). What is important is that the voltage quickly rises to the desired level and mainly stays at that level for the duration of a half-cycle. The term "mainly" in this context is understood to mean that fluctuations in voltage over a longer period - compared to the half-cycle - do not at any time bring the voltage below a value corresponding to a critical electric field level in the emulsion. The necessity of using electrode insulation is related to the risk of a water bridge that forms a short circuit between the electrodes. If no such risk exist the electrode insulation may be omitted. When electrode insulation is used the frequency of the voltage to be applied should be adapted to the conditions. A liquid or an emulsion will have a certain relaxation time for the charge, defined by: τ = ε_oε_rp, where ε_oε_r is permittivity and p is resistivity. The frequency of the supplied voltage should be so high that a half-cycle is short compared to the emulsion time constant. If the efficiency of a coalescer increases with increasing frequency there is no problem in going higher. Presently, though, it seems that the optimal design is to go as high as required to overcome the emulsion time constant but no higher than that. Recent attempts indicate that increasing the frequency further may prevent a strong agitational movement imparted by drops with a net charge, thereby contributing to a more rapid coalescence in emulsions. This effect may imply that a higher frequency may be advantageous.

In general it is simpler to control a design in which the voltages are low and it will therefore normally be convenient to split a liquid volume/ flow cross sectional area in smaller partial gaps that each separately has convenient field strength [Volts/ meter] with a limited supplied voltage. The electrodes may then be connected in parallel. Coalescer electrodes may be localized in tubes with turbulent flow and in tanks where the degree of turbulence is less.

Claims

1. Method for imparting coalescence in oil/ water emulsions by supplying electric voltage to thereby facilitate subsequent separation of water from the emulsion, characterized in supplying a substantially bipolar, square wave shaped AC voltage which is substantially symmetrical around the voltage zero value.

2. Method as claimed in claim 1, characterized in that the frequency of the AC voltage is in the range 10 - 10 000 Hz.

3. Method as claimed in claim 1 or 2, characterized in that the frequency of the AC voltage is in the range 50 - 1 000 Hz.

4. Method as claimed in claim 2 or 3, characterized in that the frequency is adapted to the current electrodes and the current coalescer in a manner in which a mainly capacitive voltage distribution is obtained.

5. Method as claimed in any one of claims 1-4, characterized in that the voltage is in the range 1 - 40 kV.

6. Method as claimed in claim 5, characterized in that the critical field strength is determined for the current case and that a voltage is chosen to obtain said critical field strength with a defined safety margin.

7. Electrocoalescer for separation of water drops from oil/ water emulsions comprising a vessel or tubes comprising at least two electrodes connected to an AC voltage source, characterized in that the voltage source is adapted to supply the emulsion with voltage pulses that are mainly shaped as square waves.

8. Electrocoalescer as claimed in claim 7, characterized in that it comprises covered electrodes to prevent the risk of short-circuits in the form of "water bridges".

9. Electrocoalescer as claimed in claim 7 or 8, characterized in that it is provided with means enabling variations in voltage level and frequency that are supplied to the emulsion.