SPE

Society of Petroleum Engineers

SPE 25346
Emulsion Treatment Program
Roslan M. Soffian and T.L. Niven, Esso Production Malaysia Inc.
Copyright 1993, Society of Petroleum Engineers, Inc.
This paper was prepared for presentation at the SPE Intemational Meeting on Petroleum Engineering held in Singapore, 8-10 February 1993.
This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). The
material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at the SPE
meetings are subject to pUblication review by the Editorial Committees of the Society of Petroleum Engineers. Permission to copy is restricted to an abstract of not
more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgement of where and by whom the paper is presented
Write librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3636 U.S.A. Telex, 730989 SPEDAL.

ABSTRACT

INTRODUCTION

Increasing emphasis has been placed by Esso

Production Malaysia Inc. (EPMI) to address the
potential emulsion concerns in its existing operations
in light of projected increases in water production
offshore and in anticipation of potential impact on its
export crude quality. The' proper approach to
demulsification is often a point of contention, as a
generic solution to oil treatment is relatively difficult
to find. Although oil treating approaches can be
grouped into four main categories, i.e., heat,
chemical, settling time, and agitation, each producing
system is unique and must be reviewed individually to
arrive at its respective solution.

EPMI operates as a contractor to PETRONAS,

Malaysia'S government oil company, under a
Production Sharing Contract (PSC) awarded in 1976 .
This operating area is about 1.4 million acres in size
and is located in the South China Sea 125 miles (200
kilometers) off the east coast of peninsular Malaysia.

This paper will discuss EPMI efforts to identify

methods of resolving its emulsion concerns, with
focus on one of its offshore locations, Guntong
Complex. This paper will discuss the initial
provisions included in the original platform design
basis, tests performed to identify resolution
requirements and studies to determine the optimum
practical facility requirements.
This paper will focus on EPMl's efforts to remove
water from emulsified crudes (forward emulsion
treatment) only. Produced water treatment disposal is
not addressed in this paper.

Illustrations at the end of paper.

EPMI currently has 12 oil producing fields which

together produce the Malaysian crude commonly
referred to as the Tapis Blend. There are a total of 32
platforms, comprising 14 manned and 18 unmanned
platforms. There are 5 main oil gathering lines from
the producing platforms that feed into a central
pumping platform located in the Tapis field (Figure
1). At Tapis pumping platform, these fluids are
commingled and pumped through a 24 inch trunkline
to the onshore Terengganu crude oil terminal (TCOn
at Kerteh for final processing before being delivered
to a nearby PETRONAS refmery or ships via a
marine offloading system.
Since 1978, EPMl's oil production has increased
significantly. EPMI currently produces in excess of
330 kbpd of oil. However, as the fields mature,
EPMI is now experiencing increasing water
production, currently around 90 kbwpd or 20% of
'total fluid production. This water production is
expected to increase in future years. Some of this
water stays in emulsified form. The increasing water
production not only presents challenges in treating
produced water for disposal, but the increasing
emulsified oil production will also affect the ability to
maintain effective pipeline capacity for peak oil rates

EMULSION TREATMENT PROGRAM

and achieve export crode quality. Five of EPMI's

oilfields (currently contributing 35 % of total oil
production) have some wells producing highly
emulsified oil which contribute to difficulties in
treating the overall oil produced offshore without
some form of additional treatment.
The existing two-stage separation trains at Guntong' A
(Figure 2) simply utilize gravity settling as the
method for oil-water separation. No additional
dedicated facilities were installed to handle the crode
emulsions. This basic principle of oil-water
separation is not adequate for crode emulsions
handling at Guntong A where the low process
temperature of the satellite streams are not favorable
and the emulsions are stabilized by wax crystals
which inhibit or slow down the separation process,
until some additional form of treatment is applied.
Guntong, Tabu and Palas (collectively referred to as
"Guntong Complex" herein) crodes have high pour
points and cloud points, close to the existing process
temperatures. Because of the particular emulsion
characteristics of these crodes, water does not readily
separate in the receiver separators at current
temperatures (approximately 30C). In contrast, some
water is being separated in the Guntong A train which
is operating at 73C.

FORWARD EMULSION TREATMENT

Definition Of Emulsion
An emulsion is a mixture of two immiscible liquids,
with one of the liquids dispersed as droplets in the
other, and chemically bound or stabilized by
emulsifying agents. In the oilfield, oil and water are
encountered as the two liquid phases. They generally
form two types of emulsions; the first type is the
forward emulsion, where the water is dispersed
throughout the oil and will not freely separate in a
conventional separator without some form of
treatment. Once the water is removed from the oil,
whether by gravity separation or other form of
treatment, the water usually contains some oil and
requires treatment before it can be discharged into the
sea. In this case, the oil which is dispersed
throughout the water forms the second type of
emulsion referred to as the reverse emulsion.

The quality of the oil is generally characterized by the

amount of basic sediment and water (BS&W) present
in the oil. The oil BS&W content is based on
observations made after the centrifuging of the oil.

SPE25346

Why Treat Emulsions?

The increasing water production not only presents
EPMI challenges in treating water for disposal, but
the increasing emulsified oil production also presents
challenges in meeting the following objectives:
1) To produce from all wells to maximize
recovery without having to shut-in wells
because of emulsion constraints.
2) To remove the produced water offshore in
order to maintain effective crode pipeline
capacity to shore.
3) To ensure the crode meets customer
requirements on export crode quality.
4) To minimize corrosion effects and maintain
the integrity of the pipelines.

Fields With Strong Emulsion Tendency

Figure 3 illustrates the relative locations of the fields
where the produced water has been found to be not
readily separable from the oil under existing process
conditions without additional treatment. Both types of
emulsions are encountered at each field. The Guntong
Complex, Semangkok, and Irong Barat fields
produced oil tends to retain a large percentage of
water in an emulsified state, with minimal water
separation offshore. Historically, the remaining older
fields, Tapis, Bekok, Tinggi, Tiong, and Pulai did
not present significant problems in the separation of
the produced water from their respective oil with
existing processing facilities.
The inability to remove water offshore frequently
resulted in high watercut wells being shut-in to
maintain maximum oil production to shore and
minimize treating difficulties at TCOT. Today, about
16 and 3 wells are selectively shut-in at Guntong and
Tabu, respectively. As part of the efforts to continue
flowing these wells, a well start-up plan has been
implemented to bring on-line the shut-in wells and
ascertain their relative impact on the final export
crude quality achievable at TCOT. Under this plan, a
well is flowed for a maximum period of 12 hours
over a few cycles prior to being shut-in again and
impact at TCOT is then monitored. The flow of the
well is limited to this period to provide TCOT the
capacity to blend 'upsets' with available 'on-spec'
crude in order to maintain export crode quality.

SPE25346

ROSLAN M. SOFFIAN AND T.l. NIVEN

Initial Bottle Tests Indicated Emulsion Concerns

During the development of the design basis for
Guntong A platform, formation of water-in-oil
emulsions was anticipated in view of its satellites'
production which will arrive cold at Guntong A.
Guntong A presently handles its own production as
well as production from four satellites, Guntong B,
Guntong C, Tabu A, and Palas A. These satellite
platforms are wellhead platforms with no processing
facilities. The full wellstream from these platforms
flows to Guntong A where the oil-water-gas are
separated. Subsea pipelines cool these satellite
streams to around 30C prior to arrival at Guntong A.
The type of treatment solution required to handle t~e
anticipated emulsion concerns was not readIly
identified during the design stage for inclusion as part
of the Guntong A original platform installation.
Operational experience with water production was
necessary to substantiate the concerns and identify
appropriate treatment solutions. Since the impact was
not expected to be significant during the first few
years of operation when the water producti~~. is
minimal, the decision was made to defer mitial
pre-investments on the treatment facilities until ~e
emulsion concerns could be confirmed and deSIgn
parameters can be better defined through tests at a
later time in the field's life.
The volume of water production at Guntong Complex
dictated the period when these tests could be
conducted. During the early phase of Guntong
Complex operations when the water production was
relatively low, all the minimal water produced
remained in emulsified form and contributed to the
high crude BS&W levels leaving Guntong A. Bo~tle
tests were conducted to determine the hydration
capability of the Guntong Complex .crude~ to pr~ict
if future increases in water production wIll contmue
to remain in emulsified form (Figure 4). Crude
emulsions were created by mixing crude and
produced water samples at various percent ~atercuts.
The hydration capability of the crude emulSIOns was
determined based on the amount of water that the
crude can 'hold' for a variety of watercuts when
subjected to the the following test conditions:
o
o
o

Existing and elevated process temperatures.

Various retention times.
Centrifuging effect.

The bottle tests provided a relatively simple and

inexpensive method to obtain data to substantiate the
emulsion concerns and provide indications of
potential factors (eg. heat, retention time, chemical,

etc) that would aid emulsion resolution. The tests

were conducted under static conditions, i.e., off-line
from the process.
In summary, the bottle tests provided the following
results:

Tests conducted on the satellite crudes at their

existing 'cold'
(about 30C) process
temperature indicated that the crudes can
hydrate in excess of 40% BS&W. F~r
Guntong A where its process temperature IS
higher (73C), the crude hydrated only as high
as 30% water.

Tests conducted at elevated temperatures

resulted in lower crude BS&W levels. At the
elevated temperatures, water dropout occurs
relatively faster and the crudes hydrate lower
amounts of water.

Some additional water settled out when

increase in retention time alone was provided.
Increasing the retention time in addition to
increase in temperature resulted in greater
amount of water dropout and lower crude
BS&W levels.

In addition to increasing the temperature

and/or retention time, centrifuging (i.e.,
applying enhanced gravity) the crude emulsion
samples further reduced the crude BS&W
content.

Larger amounts of water settled out at higher

watercuts (in excess of 70%) particularly at
higher temperatures.

Information on the crude hydration tendencies as

indicated by the bottle tests were used to predict ~e
emulsified water production forecast and how It
would impact EPMl's overall production levels if !t
was not resolved (Figure 5). Most of the water IS
expected to come from Guntong Complex with the
rest from Semangkok, Irong Barat, and other fields.
This emulsified water, if not removed offshore, will
erode pipeline capacity, reduce overall production
levels and precipitate treating problems at TCOT.
This forecast formed the basis of EPMI's strategy to
resolve the emulsion concern by removing as much
water offshore as possible, with the initial efforts
focussed on Guntong Complex.
For the remaining fields where the emulsified water
production is expected to be relatively low, EPMI is

EMULSION TREATMENT PROGRAM

SPE25346

continuing to evaluate the benefits of providing

treatment capability onshore at TCOT versus other
cost effective means to remove the water offshore.

Guntong A current process temperature of

73C is sufficient for effective use of chemical
to treat the emulsion.

Pilot Test Proaram

For Guntong B/C satellite streams, the

existing temperature of 30C (when it arrives
at Guntong A) was insufficient for effective
use of chemicals to provide full dehydration.
However, the higher process temperature of
45C in the surge vessel, achieved when the
cold Guntong B/C streams mixes with the hot
Guntong A stream, provided a favorable
condition for the chemical to separate most of
the water in the surge vessel.

Tabu emulsion is stable at temperature of

28C with greater than 50% BS&W possible.
Heat was required to break Tabu emulsion.
No chemical was found to be required. Figure
7 illustrates the effect of increasing the
temperature on the amount of water removed.
Minimal or no water settle out between
3G-50oC. Some dehydration is obtained as the
temperature is further increased with full
dehydration achievable at 80C.

The effectiveness of the hydrocyclone to

adequately resolve the emulsions was
dependent on temperature, with improved
performance
at
higher
temperatures.
Resolution of the emulsions to provide
relatively 'clean' oil (i.e., less than 3%
BS&W) was not achievable without creating
produced water reject streams with high oil
concentrations (in excess of 20 % oil by
volume).

Although the bottle tests provided indications of

potential high volume of water to remain in
emulsified form, concern over the representativeness
and repeatability of the bottle test results led to the
decision to carry out further tests, on-line with the
process, to accurately represent actual process
requirements and test potential solutions.
A pilot test was therefore designed to identify
optimum process temperatures, retention time, and
separation efficiency. The pilot test also included the
opportunity to test chemicals before embarking on
full scale trials on the system.

The pilot test skid was designed to include a variety

of equipment to allow all process variables to be
studied (Figure 6). Heat was provided by a hot oil
heat exchanger, the platform test separators provided
retention time and freewater knockout capability, and
chemical injection facilities for adding chemical.
Provision was also made to test the use of a
hydrocyclone in a relatively new service, i.e.,
removing water from oil. Hydrocyclones, which
utilize the application of enhanced gravity, are
normally used to remove oil from water (reverse
emulsion treatment).
The pilot test skid was set up to enable the desired
emulsion stream be selected for testing. The streams
that were tested were the Guntong A wells, the
combined Guntong AlBIC streams, and the Tabu
stream. The Palas stream was not included since the
amount of water from Palas is forecasted to be
relatively low, with peak of less than 2 kbwpd
throughout its field life. Only a side-stream from each
source was flowed through the pilot test skid. The
pilot test skid was commissioned in May 1991 at a
cost of US$ 250K.
The pilot tests identified some important findings
(Figure 7):

Guntong oil can hold as high as 35 % water in

emulsion, consistent with the bottle test
results.

The emulsion is stable at temperatures below

100C and can only be broken with chemicals.

Chemical Trials
Following the pilot test, EPMI conducted further
chemical field trials at Guntong A to verify the
performance of the chemical in the system. Figure 8
illustrates the results when 20 ppm chemical was
injected into the Guntong A stream. With 20 %
BS&W leaving the production header, almost all of
the water was knocked-out in the first stage separator,
with less than 0.5% BS&W leaving the vessel.
When the cold satellite streams were commingled
with the warmer Guntong A crude, some freewater
knockout occured. With commingled temperature of
around 45-50oC in the surge vessel, the best effluent
that could be achieved was 2-3% BS&W, reflecting a
substantial removal of water.

SPE25346

ROSLAN M. SOFFIAN AND T.L. NIVEN

Trea1ment Solutions
Consideration was geven to the installation of
insulated pipelines from both the Guntong and Tabu
satellites to raise the temperature of the crude arriving
at Guntong A and aid in demulsification. However,
this solution was found to be not suitable in view of
the following:
o

The arrival fluid temperature would not meet

the required emulsion 'break' temperature of
80C. This option would also be costly in
view of the long pipeline distance (14
kilometers) between Tabu satellite platforms
and Guntong A.
Although the Guntong B/C satellite streams
were too cold to drop water out in their own
inlet separation, the higher commingled
temperature when they mix with the warm
Guntong A stream was adequate to provide a
favorable temperature for chemical usage.
Tests indicated that the benefit of maintaining
higher crude arrival temperature attainable
through the use of insulated pipelines would
not significantly reduce the required chemical
concentration.

Based on the pilot test findings, plans are underway

to install chemical iItiection facilities at Guntong A.
The Guntong B/C and Tabu receiver vessels have also
been commissioned for three phase separation to
knockout freewater in order to maximize the
commingled temperature at the surge vessel and
optimize the chemical effect.

EPMI has decided not to proceed with a full scale

project to provide the necessary heat to treat Tabu
emulsion offshore because the heat/treat facilities is
costly (US$21M), the added benefit is small and the
amount of retrofit work is very extensive. Since such
a full scale project is not justified, EPMI is evaluating
the use of the hydrocyclone as a "pre-separator" to
remove the "bulk" free water from entering the
pipeline.
CONCLUSIONS
1.

Each producing system is unique and must be

reviewed individually to determine its
appropriate oil emulsion treatment solution.

2.

Pre-investment in emulsion treatment facilities

for inclusion as part of the platform's original
installation can be avoided in many cases
when the water production increases slowly
over the field's life..

3.

Operational experience and tests performed at

different stages on water production is
necessary to substantiate the emulsion
concerns and identify solutions.

4.

Pilot test on-line with the process is required

to
determine
the
actual
treatment
requirements. Bottle tests may not truly reflect
effectiveness of chemicals and heat in actual
flowing conditions.

ACKNOWLEDGEMENT
The authors wish to thank the management of Esso
Production Malaysia Inc. and PETRONAS for
permission to publish this work.

SPE 25346

EMULSION TREATMENT PROGRAM

OIL GATHERING SYSTEM

o CENTRAL 'LA TFORM

@ sATELLITE 'LATFORM
o FUTURE CENTRAL
o FUTURE SATEUITE

GUB

FIGURE 1. 1978 PSC AREA OIL OPERATIONS, OFFSHORE MALAYSIA

aUNTONG " PL. TFORM

GUffTONQA

WElL'

,I

-".,-..........

..

-t

........... -

---

'ALM"

.....--

}--:'::"'--il-~";:i!~~:>
. ~""ij'"~.~"f~~"~"~-"~"::'"::'-=--'::"'---~=fi~"';":'''::::"'':-:::'':-=-....J
.

FIGURE 2. GUNTONG A SEPARATION PROCESS FLOW SCHEME

; I
T~ 1KE~~!'--;;:~:fT"O"---J

.EKOK~ ~
1-.

._.,

S[LIO/

~~

PUlA!

/."
.,.....

....

L _ _. .

EXISTrNO CENTRAl

EMISTINt) SATElliTE
FUTURE' CENTRAL
o FUTURE SI.. TELUTE

FIGURE 3. FIELDS WITH ST"ONG WA TER-IN-OIL EMULSION TENDENCIES

ROSLAN M. SOFFlAN AND T.L. NIVEN

SPE25346

WATER
PRODUCTION 8
IKBWI'D)

..

.T

YEAR

TE~~ln.LO~D

SE'IES M BOTTLE
TESTS CONDUCTED

IlUNTOftQ. TAIllI
ANt1 'AlAt

.HYOQATtQN

INSTALLtD

S UOIES

tl \/SnoOK CllEMlCAL
WfJEcnON FACilITY
1Jt93 JN$tAllATtON
2) fNSUlAT!b PWlINES
lUSS33"" NOT REoUlRED
3) FUllICAt.. PROJECT
(USS21M) FOR l,ABU
NOT JUSftFlEO -lESS
COSTlVALThNATlYE
BEINo PURSUED

nVlAfF'VtlCTTL

C'''AM.''''

TEST FINOttfOS

2) CIlEM'CAl

Sf l D

HEeE REeOMMENDAnoNS:

IV CHEMICAL
T"ALS TO

~~c=TIOATE

PLATFOJt~

2) DHE....INE

PERFORMANCE

TREATMENT
NOTE:
REQ....EMENTS
(-). SHUT..'" OF HIGH WATERCUT WE'llS REDUCES TME ACTUAL TOTAL WATER
,RODUCTION IN COMfl'ARtSON TO TNt FoREeas T RATEs..

FIGURE 4. CHORONOLOGICAL EVENTS OF EFFORTS

'00r-----------------------------,

FIGURE 6. CURRENT FOCUS ON WATER-IN-OIL EMULSION TREATMENT

....OT TUT P.OO.AM;

_
PIIOCUS IlI!QUMtMElfTt
., TtMOPATUIll!IIlI!TPTION TIME/SPAllATION EffICIENCY
_
TUT_TOflEUlTlll4lU

TESTEQ"""" GP_
NEAT P_.--AIlATOIlIIIlYDIlOCYCl.ONE

C_ _ ''''US$UOIC
1'lL0T TEiT ICHE....TIC

'I

NOT 011.. NX

GUAWELLS
QUAlBte MUl

1L-_T_AB_U_ _

--'I---l><I~1 I

CHEMICAL
IN.JtCl"lOft

J
~_.
"

TEITSE'

__ ~_1

:---------- ........

FIGURE 6. FILOT TEST PROGRAM

---~

TESTSEP)

SPE 25346

EMULSION TREATMENT PROGRAM

PILOT TEST FINDINCS

QUNTOHQ:

TABU:

PERCENT
WATER

REMAlMHO

3!'JC, 8SA.W 'OSSfll CHEMIC4L REOurRED

SUFfl:tEHT TE.MP ; CUBIC ADDITIONAL flEA T REQUIRED

OU~

HEAT REQUJJII!D
NO CHEMICAlI
6D% BS&W POSSIBLE

'.0:0

'--..~ ..~I

_ _- - - ' ' - _ - - '_ _- - '_ _--L_....:::==i

co
10
8.
70
eo

10

TEMPRATUllE /'CI

FIGURE 7. PILOT TEST FINDINGS

0.1" 8.&.

.I.W
2-8

cn.l.w

FIGURE .. CHEMICAL FIELD TRIAL