Ultrasonics - Sonochemistry 57 (2019) 57–61

Contents lists available at ScienceDirect

Ultrasonics - Sonochemistry
journal homepage: www.elsevier.com/locate/ultson

Water-in-oil emulsions separation using an ultrasonic standing wave T

coalescence chamber
Carlos Mario Giraldo Atehortúaa, Nicolás Pérezb, Marco Aurelio Brizzotti Andradec,
⁎
Luiz Octávio Vieira Pereirad, Julio Cezar Adamowskia,
a
University of São Paulo, Department of Mechatronic and Mechanical Systems Engineering, Sao Paulo 05508-900, Brazil
b
Universidad de la República, Facultad de Ingeniería, Montevideo 11300, Uruguay
c
University of São Paulo, Institute of Physics, Rua do Matão, 1371, São Paulo 05508-090, Brazil
d
PETROBRAS/CENPES, Av. Horácio Macedo 950, Ilha do Fundão, Rio de Janeiro 21941-915, Brazil

A R T I C LE I N FO A B S T R A C T

Keywords: The offshore extraction of crude oil produces stable water in oil emulsion. To separate this emulsion into oil and
Ultrasonic coalescer water phases, the oil/water interfacial film is commonly destroyed by the addition of chemical demulsifiers. The
Water-in-oil emulsion use of an ultrasonic standing wave force field could be an alternative to reduce the dosage of chemical de-
Frequency tracking mulsifiers in the coalescence process. In this work, an ultrasonic separator of water in crude oil emulsions is
investigated through the use of a high frequency ultrasonic standing wave coalescence chamber. The coalescing
chamber uses the acoustic radiation force to induce the coalescence of water droplets at the pressure nodes of a
standing wave field. Due to temperature fluctuations, the excitation frequency is controlled to maintain the
resonance in the coalescence chamber and the voltage amplitude is controlled to deliver a given acoustic power.
Experimental tests using standardized emulsions of water in oil were carried out in a laboratory processing plant.
The effects of ultrasound application, flow rate, initial water content, demulsifier dosage and chamber inlet
temperature were analyzed. The results show that the use of the acoustic radiation force improves the emulsion
separation in all the conditions analyzed, when compared with the gravitational separation technique.

1. Introduction of surfactants. When added, they migrate to the water-oil interface,

causing rupture or weakening of the protective film around the dro-
Petroleum production and extraction processes generate stable plets. The demulsifier effectiveness depends on the composition of the
water-in-oil emulsions [1,2]. Under certain circumstances, amounts of oil as well as the water pH value. In the electrostatic separator, the
water approaching 80%–90% may even be reached. Furthermore, demulsification enhancement occurs by the application of an electric
during the transfer from the wellhead to the manifold, the need to re- field, which promotes the contact between the droplets and enhances
lease part of the gas results in substantial pressure reductions through droplets coalescence [9]. Although electrostatic separators are widely
chokes and valves. Intense mixing of oil and water occur, possibly de- used in the oil industry, mature oil fields may produce excessive
creasing the size of the droplets through flow-induced break-ups down quantities of water. Electrostatic separators containing bare metal
to diameters averaging around ten microns [3]. Gravity separators are electrodes are appropriate for water content lower than 20% [10]. The
currently used to conduct the primary oil processing in the first stage, in electrical conductivity between the electrodes increases with the water
which the final separation is produced by difference in the fluid den- content, and may cause a disruptive discharge if the content of water is
sities [4]. In general, the oil to be sent to the refinery must have less too high. A promising approach for water-in-oil emulsions with water
than 1% of water content. To achieve this value, gravity separation is content above 20% is the ultrasonic separation technique, which is not
enhanced with other techniques, such as addition of a chemical de- limited to emulsions having low concentrations of water, as the afore-
mulsifier [5], electrostatic separation [6], mechanical Barrier [7], mentioned electrostatic separators. The application of high frequency
heating [8], among others. Addition of chemical demulsifiers and ultrasonic standing waves has been investigated [11,12], showing that
electrostatic separators are two of the most common techniques used. the technique has potential to break water in oil emulsion in the oil
Chemical demulsifiers can be injected to neutralize the stabilizing effect industry.

⁎
Corresponding author.
E-mail address: [email protected] (J.C. Adamowski).

https://doi.org/10.1016/j.ultsonch.2019.04.043
Received 15 July 2018; Received in revised form 23 April 2019; Accepted 29 April 2019
Available online 29 April 2019
1350-4177/ © 2019 Elsevier B.V. All rights reserved.
C.M.G. Atehortúa, et al. Ultrasonics - Sonochemistry 57 (2019) 57–61

Typical water-in-oil emulsions have water droplets ranging from

less than 1 µm to tens of micrometers in diameter [3,13], with an
average size on the order of 10 µm [14]. The buoyant force and the
gravity force of the water droplets in the emulsion are proportional to
the droplet volume. However, the drag force is proportional to the
droplet radius considering Stoke’s drag for slow displacements; thus, for
small droplet radius, the process is too slow and the emulsion is highly
stable. Ultrasound separation technique makes use of the acoustic ra-
diation force phenomenon to move the droplets toward pressure nodal
or anti-nodal planes. In the water in oil emulsion, the water droplets
immersed in oil have positive contrast factor and thus move toward the
pressure nodes [15]. That effect causes droplets to clump together [16],
increasing their concentration and their probability of collision. The
choice of the frequency is a compromise between the number of nodal
planes and the acoustic attenuation of the emulsion for a given re-
sonating chamber width. High frequency provides a large number of
nodal planes; however, high frequency sound waves are more strongly
attenuated in a fluid medium with distance. An additional reason for
selecting a high frequency is to increase the cavitation threshold in the
emulsion. Cavitation must be avoided to prevent emulsification.
The design of the resonating chamber is a key point in the im-
plementation of the ultrasonic separator. It basically consists of a re-
sonant cavity with attached piezoelectric transducers. The cavity re-
sonant frequency depends on the speed of sound of the emulsion, which
in turn depends on the temperature. If a temperature gradient occurs
from the inlet to the outlet, different sections of the chamber have
different resonant frequencies, limiting the system efficiency. The use of
a single frequency in the entire chamber produces coalescence only in
places where the resonance occurs. This paper presents the design of a
coalescing chamber with multiple ultrasonic transducers operating in-
dependently. Two independent ultrasonic transducers were used along
the resonating chamber in the flow direction. To maintain a standing
wave in the chamber, a frequency tracking strategy is used to overcome
the sound speed variation during operation, mainly due to temperature
variation [17]. Dynamic testing of ultrasonic separation was carried out
in a laboratory processing plant at the Petrobras research center, using
synthetic water in oil emulsions.

2. Ultrasonic coalescing chamber

2.1. Acoustic radiation force Fig. 1. (a) Coalescing chamber operation principle and (b) illustration of the
ultrasonic coalescing chamber.
The acoustic separation of water-in-oil emulsion is based on the
phenomenon of the acoustic radiation force [18]. When a small droplet force [20,21]). This force originates from the wave scattering by the
of volume V is immersed in a fluid medium in the presence of a plane neighboring droplets. The interparticle force between two droplets can
standing wave of pressure amplitude p0 , it experiences a primary be either attractive or repulsive depending on their orientation in re-
acoustic radiation force, given by [19] spect to the standing wave field. Yet when they are located at the
pressure nodes, the interparticle force is attractive. As pointed out by
πV ⎞ 2
F = −⎛⎜ 2⎟
ϕp0 sin (2kx ), Pangu and Feke [15], once the droplets are located at the pressure node
⎝ 2λρ0 c0 ⎠ (1) due to the action of the primary force, secondary forces induce ag-
glomeration of the droplets, thus improving the separation efficiency.
where x is the droplet position in relation to a pressure node, λ is the
acoustic wavelength, k is the wavenumber, and ρ0 and c0 are, respec-
tively, the density and sound velocity of the fluid in which the droplet is 2.2. Ultrasonic coalescing chamber
immersed. In Eq. (1), ϕ is the acoustic contrast factor, defined as [19]

5ρd − 2ρ0 ρ0 c02 The separation of water-oil emulsion in a continuous flow is per-
ϕ= − , formed in an ultrasonic coalescing chamber formed by two ultrasonic
2ρd + ρ0 ρd cd2 (2)
transducers and an opposing planar reflector. The working principle of
where ρd and cd are the droplet density and the sound velocity in the the ultrasonic coalescing chamber is illustrated in Fig. 1a. The emulsion
droplet, respectively. Droplets with positive contrast factor gather at enters the chamber at the upper inlet and then goes through an ultra-
the pressure nodes while droplets with negative contrast factor ag- sonic standing wave field. In this region, the acoustic radiation force
gregate at the pressure antinodes. Water droplets in an oil medium have drives the water droplets towards the pressure nodes, inducing its
a positive contrast factor and, therefore, they agglomerate at the pres- coalescence and thus increasing the average droplet size. The emulsion
sure nodes of the standing wave field. exits the chamber at the bottom outlet and the final separation of water
In addition to the primary force, multiple droplets may also ex- and oil takes place in a separator vessel. The coalescing chamber op-
perience interparticle forces (the so-called secondary acoustic radiation erates at a frequency of approximately 1 MHz, the acoustic cavity has

58
C.M.G. Atehortúa, et al. Ultrasonics - Sonochemistry 57 (2019) 57–61

Fig. 3. Laboratory processing plant arrangement scheme.

obtained by mixing of 70% oil of 29 °API and 30% of oil of 13 °API, in

Fig. 2. Electric conductance curve with the chamber filled with oil and filled volume, which is similar to a Brazilian pre-salt petroleum. The oil has
with air. viscosities of 24 and 15 cP (0.024 and 0.015 Pa·s) at 60 and 70 °C, re-
spectively. In a first step, synthetic water in oil emulsions with water
an interval volume of 264 cm3 and the distance between the transducers contents of 30% and 50% were obtained by mixing water and the oil
and the opposing reflector is 20 mm. blend in a pendulum mixer at 60 cycles per minute for 1 min. After this
To have an ultrasonic standing wave of high pressure amplitude, the procedure, the mixture was processed in a Polytron® PT 3100D
coalescing chamber must operate at resonance. However, two different homogenizer at 10,000 rpm for 3 min to obtain a water in oil emulsion
types of resonance can be distinguished. The first, and the more intense, with water droplets averaging 10 μm (D50). In all the experiments, the
is due to the piezoelectric transducer. This resonance is essentially in- emulsion was prepared just before the start of each experiment.
dependent of the fluid inside de cavity. The other type of resonance is Without the addition of chemical demulsifier, the emulsion can be
produced when the size of the cavity is a multiple of half wavelength. In considered stable, since we could not observe any detectable change of
this case, the resonance is produced in the fluid itself and a stationary the droplet size distribution.
pressure pattern is established inside the cavity. Fig. 2 shows these two The mean water droplet size distribution in emulsion was measured
different types of resonances. Here we can see the electrical con- using a static laser light scattering particle size analyzer (Mastersizer S
ductance with and without oil inside the chamber. particle size stabilized by He-Ne gas lasers, Malvern Instruments Ltd). It
Temperature fluctuations cause the sound velocity to change, was used a drop of emulsion in 50 mL of spindle oil in order to achieve
shifting the resonance frequency of the acoustic cavity. In addition, the between 10% and 30% of obscurescence. The drop size measurement
temperature at the outlet is higher than the temperature at the inlet. was reported by D50, that is the volume median diameter.
Consequently, the resonance frequency at the inlet is significantly dif- The performance of the ultrasonic coalescence chamber to break the
ferent from the resonance frequency at the outlet, and the use of a emulsion was evaluated with five different setups with and without the
single operating frequency in the cavity would produce coalescence application of ultrasound. Tests were performed by varying the initial
only in places where the resonance occurs. To reduce this problem, two water content in the emulsion, the amount of demulsifier (Dissolvan
independent ultrasonic transducers were used. As shown in Fig. 1b, 961), the flow rate and the temperature in the ultrasonic coalescence
each transducer is formed by three lead zirconate titanate (PZT4) pie- chamber. The electric power of the two ultrasonic transducers were
zoelectric ceramics attached to an aluminum layer, and they are excited kept constant at 40 W per transducer in a total power of 80 W. The
independently by an electronic module with a resonance frequency parameters of each test setup are presented in Table 1.
tracking system (see Supplementary material for details of the re-
sonance tracking system). This ensures that regions in front of both 4. Results and discussion
transducers operate at the resonance, with maximum pressure ampli-
tude. After performing the tests of Table 1, the final water content in the
emulsions was obtained by Karl Fischer potentiometric titration using a
Metrohm 841 Titrando equipment. For each test setup, the experiments
3. Experimental setup
are performed with and without ultrasound, a set of six pairs of mea-
surements was obtained over 3 h of continuous operation. The results
Dynamic testing of acoustic separation with synthetic water in oil
for the five test setups are summarized in Table 2 and Fig. 4.
emulsions were carried out in a laboratory processing plant at the
Petrobras research center. A diagram of the hydraulic circuit used in the
Table 1
experiments is presented in Fig. 3. It is composed of an emulsion re- Test setups of water in oil emulsion separation experiments.
servoir tank, a hydraulic pump, an ultrasonic coalescencing chamber, a
gravity separator vessel with a coalesced water drainage, and the oil Test Initial water Amount of Flow rate Temperature [°C]
setup content [mass demulsifier [ppm] [cm3/min]
discharge. The unit has a temperature control system, with electric percent]
heaters in the pipelines and in the separator vessel, as well as a flow
control system. The ultrasonic coalescence chamber is positioned ver- 1 30 25 50 70
tically in the hydraulic circuit to create a downward flow to prevent 2 30 25 100 70
3 30 50 50 70
water accumulation. The static pressure in the hydraulic circuit was
4 30 25 50 60
kept constant at 517 kPa during the tests. 5 50 25 50 70
To perform the tests, an oil blend with API gravity of 24.2 was

59
C.M.G. Atehortúa, et al. Ultrasonics - Sonochemistry 57 (2019) 57–61

Table 2 higher than that of test setup 1. Although the concentration of water
Final water content in the processed emulsion. was higher in test 5, the final water content was lower than that in
Test setup number [Mass percent] test setup 1. The comparison between test 1 and test 5 shows that
the increase in the water content of the emulsion interfered posi-
Ultrasound on Ultrasound off tively with the coalescence of the water droplets, both with ultra-
sound on and off. Probably, a greater number of water droplets fa-
1 2.6 4.0
2 5.0 15.1 cilitate flocculation and subsequent coalescence.
3 1.9 2.5
4 2.4 4.2 5. Conclusions
5 1.5 2.3

The design of the ultrasonic coalescence chamber and its automatic

electronic controller is a crucial step to obtain high efficiency in the
application of electric power to obtain an ultrasonic standing wave. The
main difficulty in the implementation of an ultrasonic demulsifier in the
MHz range is the tuning of the operating frequency. This frequency
must be selected in a resonance of the fluid inside the chamber cavity.
In the present implementation, the use of two different transducers,
with separate frequency control, makes the operation more efficient.
We demonstrated that the use of at least two separate piezoelectric
transducers in the ultrasonic chamber, transversely positioned to the
flow direction, allows the resonance frequency tracking to obtain a
standing wave in the emulsion even when the emulsion composition
and temperature vary. This behavior allows the control of the electrical
power supplied to the ultrasonic transducers. It assures a constant
power transmission to the emulsion in the frequency range
1.09–1.15 MHz, automatically adjusting the frequency and the magni-
tude of the excitation voltages of the two ultrasonic transducers to
obtain a constant output power. In all the experimental tests conducted,
the ultrasonic coalescence technology was observed to have potential
for reducing the oil dewatering time as compared to the gravitational
segregation mechanism. The results of test setup 2 indicates that the
coalescing effect of ultrasonic standing waves can greatly reduce the
Fig. 4. Results of the water separation for the five test setups of Table 1.
residence time of the emulsion if a complementary technique is used,
such as the electrostatic separation, which works very well for small
The results of the five tests show that the application of ultrasonic amounts of water in the emulsion.
standing wave reduces the final amount of water in the emulsion in all
the analyzed cases. Considering test setup 1 as a basis for comparing Acknowledgements
with the other test setups, the following considerations can be made:
This study was partly financed by the Coordenação de
a) Residence time – The influence of the residence time can be observed Aperfeiçoamento de Pessoal de Nível Superior/Coordination for the
in the experiment of test setup 2, which has twice the flow rate of Improvement of Higher Education Personnel – Brasil (CAPES). We
setup 1 (i.e. its residence time is half of test setup 1) and, conse- thank the support received from Petrobras/ANP (grant n.
quently, the amount of water is about the double with ultrasound on 0050.0064340.11.9), FAPESP (grant n. 2014/08185-2) and CNPq.
and almost four times with ultrasound off. Using a 20 kHz acoustic
chamber without flux, Xie and collaborators reported water content Appendix A. Supplementary data
reduction in crude oil emulsions from 40% to 3.8% at 60 °C in
12 min of irradiation time [22]. Although the operation frequencies Supplementary data to this article can be found online at https://
are different, both works showed the coalescing effect of ultrasonic doi.org/10.1016/j.ultsonch.2019.04.043.
standing waves, and our results indicate that ultrasound application
can greatly reduce the residence time of the emulsion if a com- References
plementary technique is used, such as electrostatic separation.
b) Amount of demulsifier – The injection of chemical demulsifier in [1] K. Kumar, A.D. Nikolov, D.T. Wasan, Mechanisms of stabilization of water-in-crude
primary processing tanks varies from 20 to 150 ppm, depending on oil emulsions, Ind. Eng. Chem. Res. 40 (2001) 3009–3014.
[2] J. Sjoblom, N. Aske, I.H. Auflem, O. Brandal, T.E. Havre, O. Sæter, A. Westvik,
the composition of the oil as well as the water pH value. The final E.E. Johnsen, H. Kallevik, Our current understanding of water-in-crude oil emul-
water content of test setup 1 with 25 ppm of demulsifier and ul- sions. Recent characterization techniques and high pressure performance, Adv.
trasound on is similar to that obtained in test setup 3 with 50 ppm of Colloid Interface Sci. 100 (2003) 399–473.
[3] S. Kokal, Crude-oil emulsions: a state-of-the-art review, SPE Prod. Facil. 20 (2005)
demulsifier and ultrasound off. This comparison indicates ultrasonic 5–13.
standing waves can reduce the consumption of chemical demulsifier [4] E.W.M. Hansen, Phenomenological modelling and simulation of fluid flow and se-
in the coalescence process, thus reducing oil production costs. paration behaviour in offshore gravity separators, ASME Publ. PVP 431 (2001)
23–29.
c) Processing temperature – The processing temperature of test setup 4
[5] C.W. Angle, Chemical demulsification of stable crude oil and bitumen emulsions in
was decreased by 10 °C and kept constant at 60 °C during the test. petroleum recovery – a review, in: J. Sjöblom (Ed.), Encyclopedic Handbook of
The water content obtained with ultrasound at 60 °C was almost the Emulsion Technology, Marcel Dekker, New York, 2001, pp. 541–594.
same as in the test at 70 °C. This result shows that lower processing [6] J.S. Eow, M. Ghadiri, Electrostatic enhancement of coalescence of water droplets in
oil – a review of the technology, Chem. Eng. J. 85 (2002) 357–368.
temperature can be used when ultrasound is on. Lower temperature [7] W. Meon, W. Rommel, E. Blass, Plate separators for dispersed liquid-liquid systems:
means less power required for heating the oil. hydrodynamic coalescence model, Chem. Eng. Sci. 48 (1993) 159–168.
d) Initial water content – The initial water content of test setup 5 is 66% [8] A.B. Nordvik, J.L. Simmons, K.R. Bitting, A. Lewis, T. Strøm-Kristiansen, Oil and

60
C.M.G. Atehortúa, et al. Ultrasonics - Sonochemistry 57 (2019) 57–61

water separation in marine oil spill clean-up operations, Spill. Sci. Technol. B 3 emulsions, Chem. Eng. Sci. 59 (2004) 3183–3193.
(1996) 107–122. [16] T. Leong, L. Johansson, P. Juliano, S.L. McArthur, R. Manasseh, Ultrasonic se-
[9] J.S. Eow, M. Ghadiri, A.O. Sharif, T.J. Williams, Electrostatic enhancement of paration of particulate fluids in small and large scale systems: a review, Ind. Eng.
coalescence of water droplets in oil: a review of the current understanding, Chem. Chem. Res. 52 (2013) 16555–16576.
Eng. J. 84 (2001) 173–192. [17] C.M.G. Atehortúa, N. Pérez, M.A.B. Andrade, J.C. Adamowski, L.O.V. Pereira,
[10] T. Frising, C. Noik, C. Dalmazzone, The liquid/liquid sedimentation process: from Design and implementation of the frequency control in an ultrasonic break water-
droplet coalescence to technologically enhanced water/oil emulsion gravity se- in-oil emulsion chamber, Phys. Proc. 70 (2015) 42–45.
parators: a review, J. Dispers. Sci. Technol. 27 (2006) 1035–1057. [18] H. Bruus, Acoustofluidics 7: the acoustic radiation force on small particles, Lab Chip
[11] G. Ye, X. Lu, P. Han, X. Shen, Desalting and dewatering of crude oil in ultrasonic 12 (2012) 1014–1021.
standing wave field, J. Petrol. Sci. Eng. 70 (2010) 144–144. [19] T. Laurell, F. Petersson, A. Nilsson, Chip integrated strategies for acoustic separa-
[12] P. Parvasi, A. Khaje Hesamedini, A. Jahanmiri, M.R. Rahimpour, A comparative tion and manipulation of cells and particles, Chem. Soc. Rev. 36 (2007) 492–506.
study on droplet coalescence in heavy crude oil emulsions subjected to microwave [20] M.A.H. Weiser, R.E. Apfel, E.A. Neppiras, Interparticle forces on red cells in a
and ultrasonic fields, Sep. Sci. Technol. 48 (2013) 1591–1601. standing wave field, Acta Acust. United Acust. 56 (1984) 114–119.
[13] L.L. Schramm, Emulsions: Fundamentals and Applications in the Petroleum [21] G.T. Silva, H. Bruus, Acoustic interaction forces between small particles in an ideal
Industry, American Chemical Society, Washington, D.C, 1992. fluid, Phys. Rev. E 90 (2014) 063007.
[14] A.E. Bresciani, R.M.B. Alves, C.A.O. Nascimento, Coalescence of water droplets in [22] W. Xie, R. Li, X. Lu, P. Han, S. Gu, Acoustically aided coalescence of water droplets
crude oil emulsions: analytical solution, Chem. Eng. Technol. 33 (2010) 237–243. and dehydration of crude oil emulsion, Korean J. Chem. Eng. 32 (2015) 643–649.
[15] G.D. Pangu, D.L. Feke, Acoustically aided separation of oil droplets from aqueous

61