See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/250229116

Experimental investigation of crude oil desalting and

dehydration

Article  in  Chemical Engineering Communications · January 2003

DOI: 10.1080/00986440302094

CITATIONS READS

29 1,726

3 authors, including:

A. Elkamel
University of Waterloo
448 PUBLICATIONS   4,372 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Solar and Dye-Sensitized photocatalysis for Reduction of Cadmium Ion and Oxidation of Diclofenac in Aqueous Solution
View project

Product design of wheat straw polypropylene composite View project

All content following this page was uploaded by A. Elkamel on 07 November 2014.

The user has requested enhancement of the downloaded file.

Chem. Eng. Comm., 190: 65
82, 2003
Copyright # 2003 Taylor & Francis
0098-6445/03 $12.00 + .00
DOI: 10.1080/00986440390171107

EXPERIMENTAL INVESTIGATION OF CRUDE OIL

DESALTING AND DEHYDRATION

M. AL-OTAIBI
Chemical Engineering Department,
College of Engineering and Petroleum ,
Kuwait University, Safat, Kuwait
Kuwait Oil Company, Safat, Kuwait

A. ELKAMEL
T. AL-SAHHAF
A. S. AHMED
Chemical Engineering Department,
College of Engineering and Petroleum ,
Kuwait University, Safat, Kuwait

Oil produced in most oil fields is accompanied by water in the form of an

emulsion that must be treated. In addition, this water normally contains
dissolved salts, principally chlorides of sodium, calcium, and magnesium. If
crude oil is left untreated, when it is processed in a refinery the salt can cause
various operating problems. This paper investigates experimentally the effect
of five factors (gravity settling, chemical treatment, freshwater injection,
heating, and mixing) on the efficiency of the dehydration=desalting process
for a Kuwaiti crude oil and a commercial demulsifier (Servo CC 3408). These
factors are systematically varied and efficiency is analyzed. Two efficiencies
are defined: a Salt Removal (S=R) efficiency and a Water Cut (W=C) dehy-
dration efficiency. The investigation was carried out through changes made to
a single factor at a time as well as multiple variations of factors. Plots, based
on experimental data, showing the variation of the two efficiencies as a
function of the various factors are given. Two main conclusions are drawn for
the system studied. First, excessive amounts of a demulsifying agent had

Received 27 July 2000; in final form 26 May 2001.

Address correspondence to A. Elkamel, Chemical Engineering Department, College of
Engineering and Petroleum, Kuwait University, P. O. Box 5969, Safat 13060, Kuwait.

65
66 M. AL-OTAIBI ET AL.

adverse effects on the desalting=dehydration process. Secondly, the most

important factor that improved both efficiencies (S=R and W=C) was found to
be the settling time. Efficiencies up to 75% were obtained at settling times of
5 min. This factor was simulated in the experimental runs through the use of a
centrifuge. The implication of this finding is that future desalting=dehydration
systems for the oil and demulsifier studied should be based on centrifugal
techniques.

Keywords: Crude oil emulsions; Dehydration=desalting; Crude oil treatment;

Salt removal efficiency; Water cut efficiency

INTRODUCTION
As most oil fields are growing older, resulting in wet crude oil production,
it is necessary to provide desalting=dehydration systems to separate
the oil and water before the oil can be further processed. Oil de-
salting=dehydration is the process of removing water-soluble salts from an
oil stream. With the increasing regulations on effluent water and the ever-
increasing cost of producing a barrel of oil, the use of emulsion-treatment
plants has become an important practice in crude oil processing. Treat-
ment of emulsions has always ranged from simple methods such as
gravity settlement to highly sophisticated methods such as tri-volted
desalting and dehydration systems. The development of desalting systems
has always been evaluated in terms of quantities of salt and water being
removed. In a desalting unit, when crude oil is heated as part of various
desalting=dehydration or refining processes, the water may be driven off
as steam. The salts in the water, however, do not leave with the steam.
They crystallize and may either remain suspended in oil or form scale
within heat-exchange equipment. Entrained salt crystals may deactivate
catalyst beds and plug processing equipment. Because of these potential
problems, refineries usually reduce crude oil salt content to very low
levels prior to processing. To reduce the amount of desalting required at
the refinery, some oil purchasing contracts specify a maximum salt con-
tent as well as a maximum water content. A typical salt specification
would be 10 Pounds per Thousand Barrels (10.0 PTB). To satisfy such
purchase specifications, producers may be required to perform some oil
desalting.
Among the many reasons desalting=dehydration units are installed is
to avoid transporting high viscosity liquids, ‘‘water-in-oil’’ emulsions,
which require more pumping energy. Furthermore, salts and water are
removed in day-to-day operation for three reasons: corrosion, scale
accumulation, and lowering of activity of catalysts.
EXPERIMENTAL INVESTIGATION OF CRUDE OIL 67

Nature of Petroleum Emulsions

For an oil field, the two basic types of emulsions encountered are water-
in-oil and oil-in-water. More than 95% of crude oil emulsions formed in
the oil field are of the water-in-oil type, water being the dispersed phase,
and oil being the continuous phase. Purchased oil must contain less than
2% water. Further reference to ‘‘emulsion’’ in this work implies water-
in-oil type emulsions.
Water-in-oil emulsions contain complex mixtures of organic and
inorganic materials. The compounds that are found along with water and
oil are called emulsifying agents. Those agents are surface-active mate-
rials that tend to stabilize emulsions. These include asphaltenes, resins,
phenols, organic acids, metallic salts, silt, clays, wax, and many others.
Emulsions may be stabilized by the presence of a protective film
around water droplets. Those protective films, created by emulsifying
agents, act as a structural barrier to coalescence of water droplets. To
break or rupture the film that surrounds a water drop, it is necessary to
introduce chemical action and, in many desalting plants, apply heat. The
chemical used to break the film is widely known as a demulsifier.

Literature Survey
Schramm (1992) discussed the basic principles of petroleum emulsions.
The author noted that the factors that affect emulsion stability are:
viscosity, density differential, water percentage, age of emulsion, control
of emulsifying agents, and the agitation control. The author also noted
that the success of chemical treatment methods is dependent upon the
amount of the chemical, heating, and sufficient residence time for settling.
In addition, Schramm described sampling and testing techniques that
assist in characterizing a process stream’s composition and thus in eval-
uating the effectiveness of a particular separation process. Bartley (1982)
described how heavy crudes present unique problems that require
additional design considerations. Chawla (1987) discussed the problem
of wet crude in Kuwait and the need for desalting plants. al-Kandari
(1997) reported that there are several problems generated from producing
wet crude oil streams: corrosion, lowering the activities of catalysts, and
scale development.
There are many recent studies that seek to develop a comprehensive
general framework for removing water from crude oil that requires
additional processing beyond gravitational separation. For the best
desalting operating conditions, Axsia (2000) suggested that there are a
number of important design features to be considered. Recent advances
in the development of efficient desalting=dehydration processes have
made it possible to reach an optimum set of factors. One method is the
68 M. AL-OTAIBI ET AL.

one-stage system where the desalting process is performed in disc-stack

centrifuges that replace high-volume gravity desalters, (Alfa Laval, 2000).
The benefits gained are to be found in the process simplification of
turning the two-stage process into a single-stage process, resulting in less
brackish=freshwater consumption. On the other hand, AGAR (2000)
developed a system of a minimum of two or a maximum of four energy
absorption instruments: three located in the desalting vessel designated
for service and one installed in the feed line. Anon (1983) noted that a
static mixer could improve desalting efficiency. Burris (1978) designed a
technique of using fresh dilution water to remove salt from crude oil.
Baranov et al. (1986) recommended hydrocyclone vessels for use in
preliminary or complete separation of inhomogeneous systems of the
liquid-liquid type, such as water-oil emulsions. Taylor (1996) focused
on the electrically enhanced phase separation of water-in-oil emulsions.
He used AC, DC, pulsed DC, and combinations of these in the separation
of water-oil emulsions.

FACTORS AFFECTING DESALTING=DEHYDRATION

PERFORMANCE
Most activities in oil production result in a well stream comprising a
blend of gas, entrained water, oil, unwanted sediment, and dissolved
salts. The objective of oil desalting is to remove water-soluble salts and
the entrained water, which normally contains dissolved salts. Formation
water flows with crude in two types: free and emulsified. The free water is
not intimately mixed in the crude and is found in larger drops scattered
throughout the oil phase. This kind of water is easy to remove by gravity
oil-water separators, surge tanks (wet tanks), and desalting vessels. On
the other hand, emulsified waters are intimately mixed and found scat-
tered in tiny drops in the oil phase. This kind is hard to remove by simple
settling devices, so further treatment such as chemical injection, fresh-
water dilution, mixing, heating, and electricity are required. The addition
of diluent water, heating, and applying electricity can enhance gravity
separation.
The main objective of a desalting plant is to break the films sur-
rounding the small water droplets, coalescing droplets to form larger
drops, and allowing water drops to settle out during or after coalescing.
The most important variables affecting desalting performance that have
been identified and studied include (1) settling time, (2) chemi-
cal=demulsifier injection, (3) heat, (4) adding freshwater, (5) mixing
(emulsion, chemical, and the freshwater), and (6) electricity. The last
variable (electricity) will not be studied under the scope of this work and
will be the subject of future research. A discussion of the five factors that
will be investigated in this paper is given below:
EXPERIMENTAL INVESTIGATION OF CRUDE OIL 69

1) Settling time: Most desalting=dehydration equipment relies on gravity

to separate water droplets from the continuous oil phase. Gravity
difference is the active element in this process; the produced=forma-
tion water droplets are heavier than the volume of oil they displace.
The produced water usually carries some salts and solids coated
predominantly with a thin film of oil or just freely flowing along the
emulsion stream (Lohne, 1994).
2) Chemical=demulsifier injection: Emulsions can be further treated by
the addition of chemical destabilizers. These surface-active chemi-
cals adsorb to the water-oil interface, rupture the film surrounding
water drops, and displace the emulsifying agents back into the oil.
Breaking the film allows water drops to collide by the natural force
of molecular attraction. Time and turbulence aid diffusion of de-
mulsifiers through the oil to the interface. Experience has shown
that the mechanism of the chemical process is not explicable by any
simple theory. Nevertheless, there is a rule of thumb learned in the
field that states that the lower the water percentage in an emulsion
the more difficult it is to treat.
3) Heating: Heat causes a decrease in viscosity, thickness, and cohesion
of the film surrounding water drops. Heat also reduces the continuous
phase (oil) viscosity, helping water drops to move freely and faster for
coalescing. Controlling the temperature during operations is a very
delicate job. Any excessive heat might lead to evaporation, which
results not only in a loss of oil volume, but also in a reduction in price
because of a decrease in the API gravity.
4) Dilution with freshwater: Salts in emulsion sometimes come in solid
crystalline form. So, the need for freshwater to dissolve these
crystal salts arises and dilution with freshwater has become a ne-
cessity in desalting=dehydration processes. Freshwater is usually
injected before heat exchangers to increase the mixing efficiency and
to prevent scaling inside pipes and heating tubes. Freshwater is
injected so that water drops in emulsions can be washed out and
then drained off, hence the term ‘‘wash water.’’ The quantity=ratio
of freshwater injected depends on the API gravity of the crude, but,
generally, the injection rate is 3
10% of the total crude flow,
(al-Kandari, 1997).
5) Mixing: High shear actions form emulsions. Similarly, when dilution
water (freshwater) is added to an emulsion, one needs to mix them in
order to dissolve the salt crystalline and to aid in coalescing finely
distributed droplets. Mixing works in three steps: (1) helps smaller
drops to join together, (2) mixes chemical=demuslifier with the
emulsion, and (3) breaks the free injected volume of wash water into
emulsion-sized drops and evenly distributes it.
70 M. AL-OTAIBI ET AL.

EXPERIMENTAL
In a desalting=dehydration process, there are several parameters that can
be altered in order to reach an optimum combination of operating con-
ditions. This daily practice that takes place in most desalting=dehydration
plants is the theme of this experimental section. Under this study, five
parameters will be altered to reach the goal of the study, an optimum
combination of parameters. These are: crude temperature ( C), mixing
time (min), settling time (min), chemical=demulsifier dosage (ppm), and
the amount of freshwater added (%) in ratio to that of the wet crude’s
quantity.

MATERIALS
Crude oil, collected from a Kuwaiti oil well, was supplied by Kuwait Oil
Company (KOC). The characteristics of the crude oil are illustrated in
Table I. Brackish water used in the experiment was collected from field
operation of the Kuwait Oil Company. Table II gives its characteristics.
The chemical used as a demulsifier in the experiment is under the trade
name Servo CC 3408 supplied by Servo Delden BV (Netherlands).

PROCEDURE
There are three main variables that were tested in more detail: chemical
dosage (ppm), freshwater addition (%), and the mixing time (min). The
other two variables, temperature and settling time, were determined to be
of secondary priority due to the vast number of studies that have been
conducted on their effects. It is widely known how temperature and
settling time affect oil viscosity and the rate of downward settling,

Table I Characteristics of the Crude Oil Samples

Characteristic Value

Specific gravity (60 =60 ) 0.864

Reid vapor pressure (Psia) 10.5
Pour point ( F) Less than
30
Average API gravity at 60 F 31.7
Viscosity (Cs) at 70 F 17.4
100 F 10.5
130 F 6.79
160 F 4.8
Average sulphur content (% by wt) 2.7
Asphaltenes (wt%) 2.23
EXPERIMENTAL INVESTIGATION OF CRUDE OIL 71

Table II Chemical Analysis of the Used Brackish Water

Specific gravity (60 =60 ) 1.009

Total dissolved solids (TDS), ppm 8900
Oxygen content (max.), ppm 8
Electrical conductivity microomhs=CC 12714
Calcium as Ca, ppm 801
Magnesium as Mg, ppm 450
Total iron as Fe, ppm 0.25
Sodium as Na, ppm 1926
Chloride as Cl, ppm 4045
Sulphate as SO4, ppm 1500
Bicarbonate as HCO3, ppm 285
Fluoride as F, ppm 2.5
Nitrate as NO3, ppm 13.2
Nitrate as NO2, ppm 6
Phosphate as SPO4, ppm 10
Hydrogen sulphide as H2S, ppm

Chloride as Cl2, ppm

Sodium chloride as NaCl, ppm 6665
Silica as SiO2, ppm 30
Carbonate as CO3, ppm

Caustic alkalinity as NaOH, ppm

Total alkalinity as CaCO3, ppm 289

respectively. Therefore, the five variables were divided into two groups: a
two-point and a multipoint variations group. The two-point group
consists of two variables: temperature ( C) and settling time (min). The
multipoint group, on the other hand, consists of three variables: chemical
dosage (ppm), freshwater addition (%), and the mixing time (min). All
factors were changed as follows:

 Two-point variations group:

A. Temperature: High ¼ 70 C, and low ¼ 55 C.
B. Settling time: High ¼ 3 min, and low ¼ 1 min.
 Multipoint variations group:
A. Mixing time (min): 1, 3, 5, 7, and 9.
B. Chemical dosage (ppm): 1, 2, 5, 8, 10, 12, and 15.
C. Dilution water (%): 1, 2, 3, 4, 6, 8, and 10.

The experimental design that was considered in this study required 980
combinations or experiments in order to study the effects of these five
variables. Temperature and settling time were the factors that were found to
be the least varied in real processes. In fact, settling time is determined in
72 M. AL-OTAIBI ET AL.

early stages of design and manufacturing of desalting=dehydration plants,

carried out by the manufacturer based on the customer’s preferences. The
flow rate, however, is a major factor that contributes to settling time and
hence can be varied daily. Nevertheless, this experiment will focus on the
three factors mentioned above. The two values of settling time and tem-
perature used in this experiment were selected from real field experience.
A short field study conducted in the early stages of this experiment revealed
the range of values of temperature and settling time in four design phases
desalting=dehydration processes (KOC, 1987; TPL, 1992).
In carrying out the experiments, the sample was first analyzed and
tested for salt result (S=R) in pounds per thousand barrels (PTB) and
water cut (W=C) in volume percent. The details of these tests, which were
conducted as per KOC standards, are presented elsewhere (Al-Otaibi,
1999).
The diagram in Figure 1 explains the steps followed in carrying
out one cycle of the experiment. First, freshwater was added, followed
by the chemical=demulsifier. The mixture was then heated in a water
bath heater. The heated mixture was then mixed using a mixed speed
multimixer and poured into a 100 mL centrifuge tube and rotated at
1000 rpm. The final step in completing one cycle was to suck out the
top crude volume in the centrifuge tube and test it for salt result (S=R)
and water cut (W=C). The top volume was taken because that is what
happens in real operation processes: the treated crude after mixing and
heating comes out from the top of the desalting vessel. Water volumes,
being settled out here, are not under investigation in this study. The
order, shown in Figure 1, in which the factors were varied was fol-
lowed in that pattern because of the major objective of this study:
mimicing the real process. In a real process, an emulsion is introduced
into the system and is subjected to freshwater injection followed by
chemical dosage. The mixture of emulsion, freshwater, and chemical is
then heated to a certain temperature and mixed. The blend, at the final
stage, enters a vessel where settling takes place for a certain time
period, allowing water and salt to be drained off. The process at the
final stage produces dry or treated crude oil samples that are tested
and analyzed for S=R and W=C.
In each cycle of the experiment, a sample of crude oil to be tested was
taken in a sample tube or graduated cylinder of about 100 mL. Then,
according to the previously set ranges, freshwater and chemical demul-
sifier were added. Crude oil, freshwater, and chemical were next heated
and then mixed for a certain time (min). The mixture was then taken to
a centrifuge where it was rotated for a certain time (min) for settling
purposes. From the top of the centrifuge tube a certain volume of dry
crude, supposedly treated, was withdrawn by a micro mL syringe and
transferred to a test beaker. The S=R test was conducted on a partial
EXPERIMENTAL INVESTIGATION OF CRUDE OIL 73

Figure 1. Steps taken in one experimental run.

volume of that dry crude, about 10 mL. Then, 50 mL was transferred to a

centrifuge for the W=C test.

RESULTS AND DISCUSSION

As mentioned above the efficiency of desalting in each experiment is
determined by measuring the salt result and the water cut. The efficiency
of salt result is defined as one minus the ratio of final salt content divided
by initial salt content. The water cut efficiency is, however, defined as one
minus final water content divided by initial water content. We first con-
sidered changes in a response produced by a change in the level of each
factor. The individual effect of each factor at different levels was cross-
plotted against S=R and W=C efficiencies as shown in Figures 2 through 6.
In addition, we have also considered multiple effects of changing factors
simultaneously. The data collected for both individual effects and gross
effects will be used in a subsequent paper to develop semiempirical
models for correlating W=C and S=R efficiencies.
74 M. AL-OTAIBI ET AL.

Effect of Wash Water Dilution Ratio

At the beginning of water injection, the W=C efficiency was higher than
S=R efficiency as shown in Figure 2. This can be explained by the pre-
sence of some salt in the crystalline form. The free water was being
washed out, hence increasing the W=C efficiency, but leaving salt crys-
talline flowing out with the product crude. As the dilution water injection
increased, more salt dissolved in the water. The efficiency of S=R
improved, crossing the water cut efficiency at 3% water injection rate and
then stepped up as the water rate was increased. The W=C efficiency,
however, started to deteriorate as the water injection rate increased. This
is due to a phenomenon known in the petroleum field as the ‘‘reverse
emulsion’’ incident. It appears that as the continuous phase, being oil
here, gets more diluted, a higher dissolving rate of salt crystalline is
reached, thus improving salt removal efficiency but increasing the water
phase.
To improve the efficiency of W=C, the wash water injection must
be operated at the optimum point. Beyond that point, experience has
shown that excessive water may lead to deterioration in the pH range
of the water volume as a whole. Ranges of pH above or below 7.0
may cause severe problems in emulsion breaking and precipitation of
hydrocarbon solids (e.g., naphthalene) into the water continuous phase
(Agar, 2000). Elgibaly et al. (1999) stated that at lower values of pH
(3
10) of emulsion, surfactants dissolve readily in the oil phase
forming stable water-in-oil emulsions. In contrast, at pH values
11
13, they noted that surfactants dissolve in the aqueous phase,
creating stable oil-in-water emulsions. Furthermore, dilution maximizes
the density difference between water drops and the oil phase. The
dilution mechanism is also used to minimize the viscosity of the oil

Figure 2. Salt result and water cut efficiencies (%) vs. addition of freshwater (initial
S=R ¼ 100.0 PTB; initial W=C ¼ 5.0%; T ¼ 60 C; mixing time ¼ 1 min; settling time ¼ 1
min).
EXPERIMENTAL INVESTIGATION OF CRUDE OIL 75

Figure 3. Salt result and water cut efficiencies (%) vs. chemical dosage (initial S=R ¼ 61.0
PTB; initial W=C ¼ 2.0%; T ¼ 60 C; mixing time ¼ 1 min; settling time ¼ 1.0 min).

Figure 4. Salt result and water cut efficiencies (%) as a function of temperature (initial
S=R ¼ 100 PTB; initial W=C ¼ 3.0%; mixing time ¼ 1 min; settling time ¼ 1 min).

Figure 5. Salt result and water cut efficiencies as a function of mixing time (initial S=R ¼ 100
PTB; initial W=C ¼ 10%; settling time ¼ 1.0 min, T ¼ 55 C; chemical dosage ¼ 1.0 ppm).
76 M. AL-OTAIBI ET AL.

Figure 6. Salt result and water cut efficiencies as a function of settling time (initial
S=R ¼ 90.0 PTB; initial W=C ¼ 8.0%; chemical dosage ¼ 5.0 ppm; T ¼ 55  C; mixing
time ¼ 1.0 min).

phase, thus improving the overall efficiency of a desalting=dehydration

process.

Effect of Chemical Dosage

The second parameter investigated was the effect of the dosage of the
demulsifier (Servo CC 3408) in ppm on the overall efficiency of the
desalting=dehydration process. Figure 3 illustrates the W=C and S=R
efficiencies. As the demulsifier was added, both the S=R and W=C effi-
ciencies increased, reached a maximum, and then decreased at higher
chemical dosage. The demulsifier acts to neutralize the effect of the
emulsifying agents, freeing more water drops from the surrounding
interfacial film. The excessive use of this demulsifier can decrease the
surface tension of water droplets and actually create a more stable
emulsion, which is difficult to treat. It is worth noting here that chemical
injection is usually applied in practice on a trial and error basis. The
operator injects a certain quantity of chemical, sees the effect on effi-
ciencies, and then adjusts the quantity injected accordingly. Figure 3 will
eliminate this trial and error process for the Servo CC 3408 demulsifier
when used with Kuwaiti crude oil. As can be seen, the point at about
8.0 ppm chemical dosage is optimum.

Effect of Heating
Heating is another important factor used in the desalting=dehydration
process. When an emulsion is subjected to heating, the rise of tempera-
ture will increase thermal motions to enhance the collision rate. As the
temperature increases the viscosity (including interfacial viscosity) will be
reduced, thus increasing the likelihood of water drop coalescence. The
results shown in Figure 4 are representative of the effect of temperature
EXPERIMENTAL INVESTIGATION OF CRUDE OIL 77

on the efficiencies of S=R and W=C of an emulsion sample that has an

initial S=R of 100 PTB and initial W=C of 3.0%. Figure 4 shows that the
W=C increased sharply up to a temperature of 60 C, when the emulsion
started to dehydrate due to evaporation. With excessive heating, W=C
and S=R efficiencies became independent of temperature. Any further
heating did not affect the efficiencies. The decrease in S=R efficiency at
the start of the measurement is not actually an indication of an accu-
mulation of quantity of salt. It is rather a temperature-no-effect-on-S=R
region up to a temperature of 60 C. At that temperature some
improvement in S=R efficiency is observed.
It should be noted that the reported results are based on experiments
conducted in an open vessel. Currently most desalting in refineries is done
in closed electrostatic desalters at 100
150 C; no dehydration due to
evaporation occurs in these units.

Effect of MixingT|me
Mixing is used in a desalting=dehydration process to promote further
dispersion of dilution water and demulsifier=chemical with the emul-
sion. It is also used to help smaller water droplets coalesce, enhancing
the S=R efficiency and, in particular, affects the W=C efficiency.
Figure 5 illustrates this effect. Initially, the W=C efficiency increased
slightly above the S=R efficiency with increasing mixing time. Both
efficiencies increased up to the point of 5 min, when W=C efficiency
began to deteriorate. The reason for such deterioration is due to the
phenomenon of emulsion type inversion when an emulsion suddenly
changes type, from water-in-oil to water-in-oil-in-water. This process
was almost invariably observed in the beaker tests after emulsions had
been stirred for a sufficiently long time. The time required for a stable
emulsion to invert depends on the stirring speed, temperature, the
surfactant species, and its concentration (Sun and Shook 1996).
Although many papers and studies have explored this subject, the exact
mechanism of inversion remains unclear.
It can also be observed from Figure 5 that at a long mixing time, as
more salt quantities melt and are collected at the bottom of the test
beaker, the S=R efficiency improves. Elgibaly et al. (1999) also reported
similar observations. TPL (1992) also reported that too low pressure
drops (short mixing time) across the mixing valve of a desal-
ting=dehydration process will cause a decrease in S=R efficiency. That is
due to a less intimate contact between dilution water and crude. In this
case, a high S=R in the treated crude stream can be expected. In contrast,
too high pressure drops (long mixing time) will create a mixture between
water and oil with a very high stability. Under this case, a high W=C in
78 M. AL-OTAIBI ET AL.

the treated crude stream can be expected. So, for the oil and demulsifier
investigated, very short mixing times will improve the W=C efficiency but
not the S=R efficiency. Along the same line, very long mixing times will
improve S=R efficiency but not the W=C efficiency. The latter statement
actually reflects a situation encountered in our experiment, as can be seen
in Figure 5.

Effect of SettlingT|me
The settling time factor was experimentally investigated using a cen-
trifuge at a fixed speed of 1000 rpm. The relative centrifugation force
(g force) was 223.6. The time was varied from 1 to 5 min. Figure 6
shows a dramatic increase of S=R and W=C efficiencies. The settling time
factor was varied here from 1.0 min to 5.0 min. Basically, the two effi-
ciencies follow the same pattern when settling time is applied. The S=R
efficiency shows a sharp increase at the first three levels (1.0
3.0 min) and
then a leveling off towards the end points (4.0
5.0 min). The W=C effi-
ciency increases as a result of the increase of gravity difference between
oil and water. It can also be observed from Figure 6 that at higher settling
times (4.0
5.0 min), the efficiency reached a maximum value of 65.0%.
The asymptotic behavior in W=C efficiency is due to reaching a point at
which the water droplets are too small to settle out. This can be attributed
to their relatively high specific surface. A specific concentration of che-
mical=demulsifier or more mixing times is required for further settling.
This is due to the type of emulsion that is being tested; more than 40.0%
of the water droplets are considered too small to release from the thin
films surrounding them. Those thin films create what is known as a tight
emulsion: an emulsion that contains very small droplets (less than 10.0
microns) that are hard to settle out.

Combined Effects
Figure 7 shows W=C as a function of mixing time at two different tem-
peratures, 55 and 70 C. In this run, the sample initial W=C was 12.0%,
chemical dosage was 1.0 ppm, dilution water injection rate was 1.0%, and
settling time was 1.0 min. The experimental data show the optimum
mixing time with the corresponding temperature. Mixing times in the
range of 12.0
15.0 min at 55 C give a W=C less than 3.0%. The figure
clearly shows that increasing the temperature while fixing the mixing time
would not improve the W=C final value for the oil and demulsifier stu-
died. This is true because at higher temperatures, the decrease in viscosity
is counterbalanced by the effect of chemical=demulsifier on the type of
emulsion.
EXPERIMENTAL INVESTIGATION OF CRUDE OIL 79

When the W=C value is plotted against chemical dosage (ppm),

Figure 8 shows the reverse situation at the same two values of tem-
perature, 55 and 70 C. The mixing time is fixed here at 1 min but the final
values of W=C behaved in an opposite way to those of Figure 7. If an
excess of chemical=demulsifier is used under the lower temperature, 55 C,
the W=C value becomes independent of chemical dosage. In the opinion
of many researchers (e.g., Schramm, 1992) this independent behavior is
due to the development of multilayers of emulsions that are not affected
by any change of chemical dosage.
Figures 9 and 10 show the effect of chemical dosage on S=R and W=C,
respectively. The initial sample had an initial S=R of 100 PTB and an
initial W=C of 8.0%. The mixing time was set at 3 min and the settling
time was 1 min. The temperature was kept at 55 C. As the chemi-
cal=demulsifier was added, it was absorbed in the oil-water interface, and
the interfacial tension decreased. The amount of dilution water added
affects the output for both the W=C and S=R values. These amounts of
dilution water tend to improve the result for both the S=R and W=C. This
is consistent with the results obtained by other researchers (Bartley, 1982).

CONCLUSIONS
Water associated with oil production can cause considerable operational
problems. The removal of this water can be readily accomplished by
using a desalting=dehydration process. The process utilizes a set of
interacting factors to produce the optimum outcome of treated crude oil
with the least S=R and W=C.
Based on the experimental study performed with a Kuwaiti crude oil
and the demulsifier Servo CC 3408, two main conclusions can be drawn.
First, settling time is the main factor that could be utilized to the highest
level in achieving the best operating conditions. Secondly, excessive

Figure 7. Water cut (%) vs. mixing time at two different temperatures (initial
W=C ¼ 12.0%; chemical dosage ¼ 1.0 ppm; dilution water ¼ 1.0; setting time ¼ 1.0 min).
80 M. AL-OTAIBI ET AL.

Figure 8. Water cut vs. chemical dosage at two different temperatures (initial W=C ¼ 8.0%;
mixing time ¼ 1.0 min; settling time ¼ 1.0 min; dilution water ¼ 1.0%).

Figure 9. Salt result vs. chemical dosage at two different dilution waters (initial S=R ¼ 100.0
PTB, initial W=C ¼ 8.0%; mixing time ¼ 3 min; T ¼ 55 C; settling time ¼ 1 min).

Figure 10. Water cut vs. chemical dosage at two different dilution waters (initial S=R ¼ 100
PTB, initial W=C ¼ 8%; mixing time ¼ 3 min; T ¼ 55 C; settling time ¼ 1 min).
EXPERIMENTAL INVESTIGATION OF CRUDE OIL 81

amounts of demulsifier have adverse effects on crude oil desalting. The

former point raises the issue of new technologies applied in the field of
emulsion treatment. As water content increases with crude oil, wellhead
pressures decrease, the lifting costs increase, and profitable fields become
marginal. The traditional settling-tank technique will be inadequate, and
other means of improving production quality become a must. This
indicates that, for the oil and demulsifier studied, future technology of
water-in-oil emulsion treatment should be based on centrifugal techni-
ques. In fact, it is reported by Alfa Laval (2000) that centrifuges are
capable of removing up to five times more water from the washed crude
oil than traditional settling tanks.

ACKNOWLEDGMENTS
The authors would like to thank Kuwait Oil Company for providing
assistance and help in carrying out this work. Discussions with Prof.
A. Elgibaly at the beginning of this project were helpful. The authors
would like to acknowledge partial support of Kuwait University research
administration unit under grant EC072. The support of the graduate
school for the first author is also appreciated.

REFERENCES
Agar. (2000). Agar’s solution to desalting systems. http:==www.agarcorp.com=
AppNote3.htm System 3
Alfa Laval. (2000). Crude oil desalting. http:==www.alfalaval.com=separation=
customers=sebhc1r.htm
Al-Kandari, J. (1997). Desalting crude oil, Engineers, 55, 22
25.
Al-Otaibi, M. (1999). Experimental Investigation of Kuwaiti Crude Oil
Desalting=Dehydration Process, M.S. thesis, Department of Chemical
Engineering, Kuwait University.
Anon, A. (1983). Static mixer improves desalting efficiency, Oil Gas J. 81(42),
128
129.
Axsia. (2000). Oil, gas and water Process treatment. http:==www.offshore-tech-
nology.com=contractors=separation=axsia=Axsia
Bartley, D. (1982). Heavy crude stocks pose desalting problems, Oil Gas J. 80(5),
117
124.
Baranov, D. A., Kutepov, A. M., and Ternovskii, I. G. (1986). Separation of lube
oil emulsions in hydrocyclone vessels, Khimiya i Tekhnologiya Topliv i Masel,
Moscow Institute of Chemical Machinary Construction (MIKHM), No. 3,
16
18.
Burris, D. R. (1978). How to design an efficient desalting system, World Oil,
186(7), 150
156.
 82 M. AL-OTAIBI ET AL.

Chawla, M. L. (1987). Field desalting of wet crude in Kuwait, Paper 15711

presented at SPE Middle East Show, Oct. 7
10, Manama, Bahrain.
Elgibaly, A., Nashawi, I., Tantawy, M., and Elkamel, A. (1999). Flow properties
of weathered crude oils and their emulsions, J. Dispersion Sci. Technol.
20(3), 857
882.
Kuwait Oil Company, Production Personnel Development Division, Production
and Export Operations Group. (1987). Dehydration-Desalting Principles:
A Training Guide.
Lohne, K. (1994). Separation of solids from produced water using hydrocyclone
technology, Trans. IChemE, 72, A (March), 526
539.
Schramm, L. L. ed. (1992). Emulsions Fundamentals and Applications in the Pet-
roleum Industry, 1
45, 51
74, 295
312, 313
338, 341
382, American
Chemical Society, Washington, D.C.
Sun, R. and Shook, C. A. (1996). Inversion of heavy crude oil-in-brine emulsions,
J. Pet. Sci. Eng. 14, 169
182.
TPL Company. (1992). Desalters Phase IV (GCs 1, 3, 6, 8 and 21), Project Report
for Kuwait Oil Company, Contract 92G026.
Taylor, S. E. (1996). Theory and practice of electrically enhanced phase separa-
tion of water-in-oil emulsions, Trans. IChemE, 74, A(July), 526
539.

View publication stats