Society of Petroleum Engineers

SPE 27179

Developments in Environmental Protection Related to Produced

Water Treatments and Disposal (Produced Water Re-Injection)
R.C. Evans, Serck Baker Fluid Systems Consultants
SPE Member

Copyright 1994, Society of Petroleum Engineers, Inc.

This paper was prepared for presentation at the Second International Conference on Health, Safety & Environment in Oil & Gas Exploration & Production held in Jakarta, Indonesia,
25-27 January 1994.

This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper,
as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction 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 SPE meetings are subject to publication review by 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 acknowledgment
of where and by whom the paper is presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A. Telex, 163245 SPEUT.

Abstract Subsurface l>roduced water disposal has taken

place fOi' a number of years onshore where a
Produced water re-injection (PWRI) has become an typical stl'3tegy is to pump the produced water,
increasingly important topic in the oil industry in the last after vat'ious levels of treatment, into a high
few years, especially in mature oil pmducing areas such permeability sand or fractured limestone,
as the North Sea, It even has its own acmnym, The However produced water re- injection for
prime causes for this include the increasing volumes of l>ressure maintenance will be a more exacting
pmduced water fmm mature oil fields, the continued dl'ive engineering challenge as operational problems will
to reduce CAPEX and OPEX, and the greater concem have a direct impact on pmduction and the
over environmental protection, In the latter stages of the profitability of the development,
economic life of an oil field up to 90 % of the produced
fluid volume may be water, Detailed planning for the The uncel'tainties and decisions which will have
treatment and disposal of produced water is required at to be addressed for each produced water re-
an early stage in the development of an oil field to avoid injection pmject include; the injection water
it becoming a bottleneck to pmduction, quality specification; contingency for the sUl'face
disl>osai or storage of the produced water should
Among the perceived benefits of produced water injection injection have to periodically halted; the selection
are; a reduction in the overall volumes of water treated and integration of the additional water sOUl'ce
offshore; that the water should not need deaeration if a which may be required dua-ing the early stages of
closed system is maintained; that it could may reduce the production when the volumes of produced watel'
potential operating problems associated with scaling and al'e insufficient fOi' pressure maintenance; and the
reservoir souring; that the injection zones may be able to design of the pmduced water treatment and
accept a poorer water quality than is specified for surface system,
disposal.
Due to this multidisciplinary' nature, a pl'oduced
water re-injection pmject will obviously require
References and illustrations at end of paper the formation of an integrated team of reservoir

707
engineers, production chemists, geologists and process s)'stems installed. The limited experience implies
engineers. that there are still some uncertainties associated
with produced water injection (fig 2) which need
The key factor in the design of the produced water re- to be addressed by co-ordinated research
injection system will be the injection water quality programmes. Particularly important is the
specification, particularly in comparison to the specifying of injection water quality which will be
specification for surface disposal. Considering the a key factor influencing overall system design.
continuing debate, after two decades of experience, on Produced water injection system design needs to
water quality for seawater injection, research into water be optimised for a particular field to maximise
(Iuality specification for produced water re-injection the net pl'esent value (NPV) of the development
should be targeted as a key area of research if produced (fig 3). If the water quality specification is not
water re-injection is to be confidently and successfully stringent enough injection problems may occur.
applied given tbe constraints of particularly offshOl'e If the injection water quality specification is not
production. Each field will require its own studies to stringent enough, money will be wasted on
determine the water quality for produced water re- unnecessary capital and operating expenditure.
injection and the appropriate water treatment system
design, due to the variations in resel'voir geology and the
characteristics of produced fluids. REDUCING THE COSTS OF ENVIRONMENT
COMPLIANCE, AN R&D TARGET FOR THE
INTRODUCTION 1990'S

The study of produced water reinjection is topical at a Operators are re-exammmg their research
time when worldwide a number of legislative and I)rogrammes to ensure their is direct correlation
regulatory initiatives are being considered to protect the with field needs, and that they can make
environment which could significantly effect the economics significant cost saving benefits. Reducing the
of oil and gas exploration and production (Godec & cost of compliance with increasingly stringent
Biglarbigi, 1991). environmental regulations is one of these.
Produced water is an environmental cost centre
The positive forces for produced water injection may be many ol)erators are concerned about. Re-
summarised as; injection is potentiall)' an appealing alternative,
depending on the injection water quality
* Increasing volumes of produced watel' I'equil'ements (Littlejohn, 1992).
* More stringent environmental and
discharge regulations Produced water disposal is one of the significant
* Requirements to reduce capital and environmental issues receiving significant
operating expenditure attention within the oil and gas industry.
* A water source for pressure maintenance Heightened interest within the public and
and EOR regulatory agencies towards environmental issues
has been an impetus for the industry to
The volumes of produced water for a pal'ticular field can scrutinise its wastes and how they are managed.
increase at an accelerating rate (fig I) highlighting the Companies must be concel'ned about complying
need to plan for produced watel' treatment in the ea...y with applicable waste disposal regulations,
stages of reservoir development to avoid it becoming a mmlmlsmg the impact - of wastes on the
bottleneck to profitable production. Water handling environment, and reducing the potential liability
optimisation is now a major consideration for operators. associated with imprope"'y disposed waste. All of
this must be done within certain economic
There is relatively limited operational experience of, in bounda.-ies.
particular, offshore produced water injection. This is
reflected in the scarcity of published matel'ial on operatOi' Eliminating 01' mll1lmlsmg waste generation is
experience, although more and more data is now being becoming mOl'e critical, both envil"Onmentally and
published as more trials are undertaken and mOl'e economically, as a means of reducing waste

708
related liabilities and costs. The re-injection of produced Activity 18 Finalisation Of Overall
water changes it from a waste product to a resource. System Design

PRODUCED WATER INJECTION REOUIRES A

MULTIDISCIPLINARY APPROACH AND AN
INTEGRATED TEAM DISPOSAL REGULATIONS AND
El'ivlRONMENTAL LAW
A produced water re-injection project requires the
integration of oilfield chemistry, geology, reservoir The extent of the application of produced water
engineering, process design and financial appraisals in re-injection will be greatly influenced by trends
order to achieve the most suitable produced water in environmental law and disposal regulations.
injection system for a particular oilfield. This highli~hts
the need for an integrated and multidisciplinary team In the NE Atlantic and Arctic Oceans, the Paris
right from the initial design stages. Some of the wide Convention on the Prevention of Marine
variety of reservoir engineering and process considerations Pollution f.'om Land Based Sources (PARCOM)
for produced water injection systems are illustrated in 1974 applies to discharge from offshore oil and
fIgUre 4. gas installations. The target standard for
produced water discharges is presently set at a
A typical sequence of activities which may be required for mean of 40 mWI total oil based on solvent
a produced water injection project, up to the finalisation extl'8ction and inC.'ared spectroscopy. More
of the system desi~n, is listed below and reveals the recent I)ARCOM discussions have even sug~ested
variety of expertise required. a possible reduction of the 40 mWI target to 30
mWI or even lower. In the UK the Prevention of
Activity 1 Produced Water Characterisation Oil Pollution Act enacts the relevant conditions
Activity 2 Characterisation Of recommended by the Pa.'is Commission with
Supplemental Water Sources reference to operational oily water discharges
Activity 3 Reservoir Characterisation For from offshore oil and ~as operations. In the USA
Water Injection Purposes new Offshore Emuent Guidelines have established
Activity 4 Coreflooding Tests oil and ~rease limits for all produced water
Activity 5 Reservoir Simulation To dischar~es of a monthly avera~e of 29 mWI and a
Evaluate Injectivity daily maximum of 42 mg/l for terl'itorial seas and
Activity 6 Reservoir Simulation To offshore continental shelf locations. The previous
Evaluate Water Production limits were a daily average of 48 mg/I and a
Activity 7 Pilot Process Plant Tests maximum of 72 mW" The new limits are based
Activity 8 Pilot Injection Tests on improved .performances of gas flotation
Activity 9 Scaling Studies technology. The stipulation of no visible surface
Activity 10 Reservoir Souring Studies sheen on the seawater is also made. In the USA
Activity 11 Microbiology and COI-rosion in offshore areas, companies are mandated by the
Studies Minerals Mana~ement Service (MMS) 30 CFR
Activity 12 Determination Of Injection 250 to pl'()\'ide the best available technology
Water Quality (HAT) for the disposal of pl'Oduced water. The
Activity 13 Integration Of Supplemental genel'8l h'end in the USA is to p.'ohibit the
Water Source dischar~e of produced water in shallow coastal
Activity 14 Selection Of Process Packages locations, yet to allow surface discharge in deeper
Activity 15 Chemical T.'eatment Selection waters of .'egions such as the Gulf of Mexico.
Activity 16 Evaluation Of Techniques To Re-injection by all facilities has been considered
Control Water Production by EPA but has not yet been enforced.
Activity 17 Financial Appraisals Includin~
Comparisons With Alternath'es The EPA does not have sufficient data to
Such As Surface Or Subsurface determine if radionuclides in produced water are
Disposal l)()lIutants of concern, and without this data the

709
EPA chose not to stipulate injection. However the EPA marine environment is determined by the
will be collecting this data in the future in order to review following interrelated factors: spreading,
the situation with respect to surface discharges. evaporation, dispersion, emulsification,
dissolution. sedimentation and various
There is a draft EPA guideline for compulsory produced degmdation processes such as photochemical
water injection at coastal locations to meet the zero oxidation and microbiological degradation. The
discharge requirement in those locations. Some of The I'ate of bacterial action depends on the
USA states on the Gulf of Mexico plan to covel' new composition of the oil, the tempemture, nutl"ients
oilfield developments in the OCS under a new pel'mit, the and sul'face m·ea. As the oil is present as small
water quality provisions in which may restrict produced droplets this gl'eatly enhances the rate of
water discharges leading to the need for I'e-injection. degradation.

Oil company personnel may now find themselves faced It is important to note that dispersed
with personal liabilities taking the form of fines 01' hydmcal"l)ons are not the only constituent of
imprisonment. In the USA, at least, individuals may now pmduced watel" which raises environmental
be subject to criminal convictions. There has begun a concern. Dthel" constituents include dissolved oil,
shift from civil to criminal prosecutions concerning the tI"eatment chemicals, heavy metals, radionuclides
breach of environmental regulations. In the USA plant and high total dissolved solids.
managers, foremen and field hands are serving prison
sentences at an increasing rate for wilfully disregarding The recent study by Stephenson (1992) concluded
environmental laws (Griffin et.at., 1993), that in geneml, pmduced watel" is not harmful in
the offshore emit'onment when it is discharged
I)ersonnel training and QA procedm'es will obviously play within current dischat'ge limits and when more
a large part in the minimising of pollution from both than adequate dilution occm"s, The potential fOi'
onshore and offshore installations. Recently BS 7750 has envimnmental damage incI"eases fOI' discharges
been prepared in the UK to specify requirements for the nem"ShOl"e and coastal locations where dilution
development, implementation and maintenance of and watel" cit'culation are I"estl"icted.
environmental management systems aimed at ensuring
compliance with stated envimnmental policy and Excessive amounts of any of the substances
objectives. This standard complements the dl"aft quoted as being in produced water can be
European Community Eco-Audit Regulation (Version 3). harmful to the envimnment. It is the degree of
It also complements BS 5750 as they take pat"allel dilution and dispersal, and the callacity of the
approaches to achieving and demonstmting compliance envit'onment to use, absorb 01' I"endel' hat"mless
with specified I'equirements. Environmental audits and the constituents in the watel" which will control
associated services now play a major role in oilfield the degl"ee of envimnmental pollution caused by
operations. the sul'face disposal of produced water.

Chronic toxicity is more difficult to quantify that

WHAT IS THE ENVIRONMENTAL IMPACT OF lethal efl"ects and should not be neglected. 1<'01'
PRODUCED WATER? chnmic efl"ects the no observed efl"ect
concentmtion (NDEC) is believed to be 1000
The increasing trend of oil pollution from pmduced watel" times less than that for acute efl"ects (Hansen &
in the North Sea is dit'ectly due to the increasing volumes Davis, 1993).
of produced water being disposed of. In 1991
approximately 4,500 tonnes of oil was discharged into the Thus the degl"ee of envit'onmental hat"m caused
North Sea from about 90 oil and gas production by I"esponsible and compliant surface produced
platforms. This is equivalent to mughly 5 % of the cargo water disposal is still a matter of debate, but
of the Braer which ran aground on the coast of Shetland cel"tainly the outcome will not be a relaxing of
in January 1993. Thus current oil dischal1~es of oil in dischat°ge regulations. The opemtol"s are unlikely
produced water is equivalent to, in volume, a BI"aer type to be given the benefit of the doubt.
disaster every 20 years. However, the fate of oil in the

710
At present the regulations and methods for produced a packer. Coning control is performed by
water treatment for surface disposal do not normally adjusting water production rate to oil production
include the removal of treatment chemicals, soluble rate in order to prevent the water cone from
hydrocarbons, heavy metals or radionuclides, which are breaking through the oil into the oil perforations.
potentially the most toxic components of the produced At equilibrium a stable water cone is held -down
water. below the .oil producing perforations (Wojtanowicz
et.al, 1991). There are several potential
WATER PRODUCTION CONTROL advantages of this method,

Methods of controlling water production should be * oil production rate increases

considered as a means of reducing the volumes of water . without water breakthl'ough
produced on topsides facilities, especially as produced * well life extends far beyond its
water is relatively difficult to treat. A recent survey by value without coning control
the Petroleum Engineer International (October 1992) * produced water not so
revealed that some 89 % of the respondents confirmed contaminated with crude oil,
that excessive water production was adversely affecting demulsifiers and other agents
their fields profitability. Of these operators, 74 % have used in production and therefore
used established, proven technology, whilst 47 % had used easiel' to clean for discharge
new, experimental technology over the past 2 years to regulations
reduce water production. Surprisingly 32 % of the * reduction in water oil/ratio
l'espondents indicated that they had done little to reduce * well productivity not reduced due
water production. to water encroachment in near-
well zone.
The methods of water production control may be divided
broadly into those involving well design, and those The use of horizontal-well drainage can be used
incorporating the use of chemicals or solids squeezed into to repress the coning problem, and so water
the formation. production. Horizontal wells result in a lower
watercut than in vertical wells resulting in less
The chemical/solids methods include the ultra-fine watel' production and potentially less water
particle size cement squeeze, the cement squeeze, and the treatment costs, The higher production index of
use of resins, epoxy plastic pills, cross-linked polymers the horizontal well should result in less pressure
and sodium silicate gels. The use of cross-linked drawdown so reducing the risk of fines movement
polymers and sodium silicate gels were given the lowest and sand production.
technical and economic success ratings in the Petroleum
Engineer International survey. In the Brae. field, when prolific producers have
watered out, a reduction in water cut has been
The well completion design methods include the use of achieved by remedial work planned to exploit
horizontal wells, and dual completions with a tailpil)e unswept areas of the reservoir. Water producing
water sink for vertical wells. zones have been isolated (Wolodkiewicz, 1993).

Parameters effecting water coning in vertical wells are the

mobility ratio, oil zone thickness, ratio of gravity forces to INJECTION WATER QUALITY SPECIFICATION
viscous forces, well spacing, .-atio of horizontal WILL BE TilE KEY TO PRODUCED WATER
permeability to vertical permeability, well penetration and INJECTION SYSTEM DESIGN
production rate.
It is not surprising that their is uncertainty over
In the case of the dual completion/tailpipe water sink water quality specifications for produced water
method, the well is dual completed so that the lower injection, as their are still widely conflicting ideas
perforations are placed in the watel' zone, and the water on watel' (Iuality for the more traditional seawater
can be produced concurrently and independently to oil injection. Yet it is the injection water quality
production. The two producing streams are separated by specification which will be one of the most

711
important factors effecting produced water injection It is important to realise that produced water has
system design. orders of magnitude higher concentrations of
total suspended matter than North Sea seawater.
Produced water is a much more complex fluid than Also, unless cooled, the I)roduced water will also
seawater. The levels of total suspended matter are much be significantly warmer that the North Sea
more variable than North Sea seawater which is in fact seawater, reducing the reservoir cooling and the
one of the cleanest water sources for water injection (fig extent of the thermo-elastic fracture.
5). As a consequence it would be very unwise to
extrapolate the results of injecting North Sea seawater to Thermo-elastic fractures are also less likely to
injecting produced water. occur in unconsolidated sands.

Studies need to be initiated on a variety of scales from For cold seawater injection, traditionally lower
that of the pore network to that of the reservoir as a injectivity declines have been recorded for
whole (fig 6). injection wells than were predicted by cOl'eflood
tests or pore size distribution measures such as
A water quality specification for produced water injection the effective hydraulic radius. The existence of
should include specifications on; thermo-elastic fractures has been proven to be
* oil droplet size distribution one of the reasons for this uiscrepancy. Recent
* total oil in water concentration papel's have discussed the existence of fractures
* suspended solids size distribution developed in water injection wells due to the
* total suspended matter cooling of the reservoir rock by the injection
'" cI,emical treatments water. As discussed by Clifford et.ai." 1990, the
... dissolved gases injection of cool surface water into a warm
.. bacteria reservoir leads to cooling and contraction of the
... scale and corrosion control rock around the wellbore. As the cooled rock
pulls inward towards the· upon the upon the
It is only when these parameters can be specified with warm region around it, a tensile component of
confidence, for a particular oil field, that the most cost stress (thermoelastic) builds up within it. This
effective produced water injection system design can be allows fl'actures to propagate at bottom - hole
formulated. An incorrectly specified injection watel' pressUl'es significantly lower than the
quality could be wasteful in one of two respects. If the conventionally estimated fracture gradients. The
specification is too stringent money will be wasted on fracture opens against the direction of least
un-necessary process equipment. If it is too lax, stress, which means it will occupy a vertical
operating costs will rise due to injectivity problems,the plane.
need to retrofit additional process, reservoir pressure
declines and production rates will decline. A water In the case of water injection into unconsolidated
quality specification of just total oil-in-water will reservoirs, if the depth of permeability damage,
certainly be insufficient yet this is often the only pel'meability heterogeneity and radial flow are
parameter recorded for onshore produced water injection taken into account when using coreflood and pore
and during the few offshore produced water injection netwo..k data, estimates of injectivity index can be
trials that have been conducted. made which closely co....espond to those measured
in the field (fig 7). This will be similar to the
It will be important not to repeat the mistake of seawater case of the injection of hot produced water in the
injection of initially setting an ambiguous water quality abscence of fl'actu..es (natu..al, the..mo-elastic or
specification which then became an industry standard by intentially induced). It should be noted that the
default. This involved just quoting a percent removal method of dete..mining a filtration specification
specification which takes no account of the absolute based on the effective hydraulic radius takes no
concentrations, and maximum size, of particles in the. account of the depth of permeability damage, and
injection water. assumes that the damage will occur along the
enti..e length from the injection well to the
production well.

712
Properly designed coreflood tests using randomly distributed.
appropiately designed coreholders can provide valuable =I: It should take into account the full range
data for the design of produced water injection systems of permeability heterogeneities, and
such as the depth and extent of permeability damage should not use average values.
related to water quality. Whilst the overall permeability .. It should take into account, and not
decline for a coreplug may register 50%, permeability guess at, the depth of permeability
decline may be restricted to the first few centimeters of damage,
the coreplug, whilst the downstream section of the plug .. It should interpret the results of linear
experiences relatively little permeability decline. As a coreflood experiments in terms of radial
particle, or oil droplet has to pass through hundreds of flow from an injection well
pore throats even in the first few centimeters of a . Coreflood experiments should be
coreplug, the probability is that if it is not retained then continued until a stable permeability had
it will keep on passing through the coreplug. been reached enabling the actual level of
Permeability decline in the damage zone of a coreplug pel-meability decline to be quantified.
needs to be in excess of 70% for significant injectivity .. It should not extrapolate the results of
declines to be recorded when taking into account radial short term coreflooding tests using
flow. Also it is important to remember that in terms of concepts of filter cake build up on
injection capacity a 1 metre interval with a pel-meability membrane filters when the membrane
of 1 Darcy is equivalent to a 100 metre interval with a filters are not representative of reservoir
permeability of 10 mD. So if the high permeability zones rocks
remain unblocked there will not be significant overall * It should not use simplistic rules of
llermeability declines. This is why it is essential to fully thumb to determine the blockage of pore
consider permeability heterogeneity. throats by suspended solids.
.. It should not express filtration
specifications on a percent removal basis.
MISCONCEPTIONS ON DETERMINING WATER It should not blindly use filtration
QUALITY REQUIREMENTS specifications from previous water
injection projects when the reservoir and
Certainly the specification of water quality for seawater produced water characteristics may be
injection has been plagued by inappropriate techniques considerably different.
and misconceptions which have lead to inadequate =I: It should take into account heterogeneity
definitions. between the reservoir zones in a field
and the consequential different water
The method used to specify the water quality for produced quality requirements, especially if
water injection should include the characteristics listed significant proportions of the recoverable
below; oil is in the low permeability zones.

. It should integrate reservoir engineering, SWEEP E14....l.CIENCY

sedimentology, process design and source water
studies. Thermo-elastic fl'actures have enabled cold
It should not only consider overall injectivity seawater with suspended solids of up to 10 mg/I
decline for setting filtration specifications, but to be injected into relatively high permeability
should also consider important financial I'eservoirs without apparent losses in injectivity.
considerations such as sweep efficiency With respect to sweep efficiency, the high
.. It should consider the effect of reselovoir pel-meability zones will receive the greater
fracturing where applicable volumes of watel' and so the greater degree of
It should take into account the characteristics of cooling. Thus the thermally induced fractures
sedimentary rocks from the pore throat to the may prefel'entially occur in the high permeability
reservoir scale. layers. The fracturing of the high permeability
.. It should not assume that vital attributes of zones may conceal to an even greater degree the
sedimentary rocks, such as pore throat sizes, are injectivity decline in the low permeability zones.

713
Of prime importance in water injection is the sweep exhibit poor vertical distributions of injection
eBiciency. A low overall injectivity decline may conceal water. In the late 1970's treatment of the
high injectivity declines in the low permeability zones. produced water resulted in cleaner water for the
The blockage of the least permeable layers in an injection waterflood and a sharp increase in oil production.
well by suspended matter will reduce the rate of injection Ultimate oil recovery from the oil reservoir is
into that layer, and as a consequence the rate of oil exrcted to be 53.6 million barrels [8.5 million
sweeping in the low permeability zones will also be m ] of oil which is 35% of the original oil-in -
reduced. However if the most permeable layers remain place (OOIP). Waterflooding with cleaner water
unblocked the injectivity will, to a great extent be is expected to increase the final production by
sustained. If the low permeability zones contain significant 20.3 million barrels [3.2 million m 3] of oil, in
proportions of the recoverable oil, the economics of the comparison to primary recovery alone.
field development will be impaired as a consequence. (fig.
9) Thus maintaining a certain overall level of injectivity At the beginning of the waterflood in 1954 the
in a non - homogeneous reservoir should not be the only blockage of the perforations and the formations
criteria of specifying filtration requirements. associated with the injecting unfiltered produced
watel' was not observed as the high permeability
With respect to the North Sea, comparisons may be made layers were still taking the majority of the flow.
between the relatively high permeability Etive formation However over time it was finally noticed that the
and the low permeability Rannoch Formation. In some use of dirty injection water had caused a
reservoirs the Rannoch may contain over 30 % of th~ reduction in the vertical coverage in all
oil-in-place such as in the Cormorant field, thus setting waterfloods in the Inglewood field. The
the injection water quality specification for the lower subsequent of cleaner injection water increased
permeability zone, may be more economically prudent the vertical coverage from approximately 15% to
than considering injectivity decline for injection into. all about 70 %. Overall injection rates increased by
the formations combined. 35%.

Thus the existence of thermo-elastic fractures during the

injection of the very clean, very cold North Sea seawater CONTROL OF DOWNHOLE WATER QUALITY
into consolidated sands, should not be considered as the
passport to the injection of relatively untreated produced Some measurements of downhole water quality in
water which contains orders of magnitude higher levels of injection wells have supposedly revealed levels of
total suspended matter and is also considerably warmel·. total suspended solids more than twice that in
the unfiltered seawater. However it should be
noted that the water samples would have been
AN EXAMPLE OF OPERATIONAL EXPERIENCE OF exposed to the atmosphere prior to analysis which
THE DETRIMENTAL EFFECT ON PRODUCTION IN leads to the question of precipitation of dissolved
HETEROGENEOUS RESERVOIRS OF POOR i.-on. Thel'e is also the question of contamination
INJECTION WATER QUALITY of the sample caused by contact of the wireline
and sampler- with the tubing walls.
Parsons. et.al. (1988) describe practical experience of
cleaner injection water improving sweep efficiency and oil Even if a downhole deterioration of water quality
recovery associated with waterflooding a thick, is occurring it does not mean that this should be
heterogeneous reservoir in the Inglewood field in the Los accepted even though overall levels of injectivity
Angeles Basin. The reservoir is of unconsolidated sand declines are relatively small. This downhole
with a gross thickness of between 300 m and 425 m. deterioration is costing the operator in terms of
Most of the individual sand bodies are less than 1.8 m the corrosion of the downhole tubulars and
thick separated by shales. The coefficient of permeability possible reductions in sweep efficiency, reducing
variation is 0.6. the rates of injection into the low permeability
zones which may contain significant proport!ons
Water injection began in 1954 with unfiltered produced of the recoverable oil.
water, however water injection wells historically tended to

714
 Water injection well steel tubing lined with high density unsuccessful In the case of Wytch Farm where
polyethylene pipe is now being used by some operators the produced water is passed through sand filters,
to control downhole corrosion. The insertion of the the injection of water with average oil-in-water
aDPE adds about 37 % to the cost of the injection concentrations of less than 15 mg/kg and average
tubing. The tubing can be pulled several times without suspended solids of less than 5 mglkg resulted in
damage to the lining and joint seals. no loss in injectivity.

King (1993) reports on PWRI for the Maureen

EXAMPLES OF RECENT NORTH SEA INJECTION Field, again a relatively high permeability
TRIALS reservoir. Produced water has been injected with
oil-in-water concentrations varying between 50
Recent produced water injection trials by North Sea and 725 ppm with no apparent permanent skin
operators have highlighted the different ideas on water damage, Injectivity losses which occurred when
quality specifications and reasons for injectivity high levels of oil- in -water were injected were
impairment. - _What is noticeable is that the injection restored when relatively cleanel' water was
water quality is overwhelmingly reported only in terms of injected.
total oil in water. The size distribution of the total
suspended matter in relation to the reservoir quality The differing reselovoir response in the above
should be the main water quality parameter. examples may be attributed to differing reservoir
characteristics, and different suspended matter
. Simpsonet.aI. (1991) reported no loss of injectivity during size distributions. The water quality specification
a six week trial involving the injection of produced watel' for produced water injection should take into
with an oil-in-water concentration varying from 50 to account the reservoir characteristics (such as
1200 ppm and a suspended solids concentration varying wettability and pe-rmeability), and oil droplet and
from 5 to 50 mg/I. suspenrled solids size distributions and
concentrations in the injected produced water.
Alternatively, Dorrestijn (1991), reported a decline in
injection rate from 16,000 BWPD to 4,000 BWPD over a
25 day period when injecting produced water with an oil- PROCESS OPTIONS FOR PRODUCED WATER
in -water concentration as low as 20 to 40 ppm. The free INJECTION SYSTEMS
oil, in combination with iron oxide and scale, forming a
filter cake on the injection face was thought to be Some of the various lu'ocess packages which may
responsible for the injectivity decline. be included in -a produced water treatment system
are listed below:
I>etersen & Jorgensen (1991) reported no injectivity
decline when injecting produced water with an oil-in- * Plate separators
water concentration of 20 to 40 mg/I. Tests of injecting * Induced Gas Floatation
water direct from the separator were then tried. Injection * Media Filters
pressure increases were reported following a process upset * Coalescers
when produced water with an oil-in-water concentration * Static Hydrocyclones
of up to 3 % was injected. * Rotary Hydrocyclones
Cartridge Filters
Sweeney & Paige (1993) report that in the case of the * Solid/Liquid Hydrocyclones
relatively high permeability Forties field, the injection 01< Membranes
produced water from the degasser with an average oil-
in-water concentration of 300 mg/kg and an avel'age The selection of the process packages will depend
suspended solids concentration of 20 mg/kg resulted in a on; the injection watel' quality specification; the
50% loss in injectivity in comparison to the injectivity charactel'istics of the untreated produced water;
with seawater. The injectivity could be restored to a great the amount of l)roduced fines/solids; surface
extent by the subsequent injection of seawater, although discharge regulations; supplemental water
stimulation using acid or solvents were generally resources required, whether the platform is

715
Roating; production methods; relative capital and Induced gas flotation is often capable of 96 %
operating expenditure, and weight and space restrictions. removal of oil-in-water, reducing inlet
concentrations of greater than 200 mwl to less
An increasingly relevant subject in the design of produced than 20 mwl with the help of approximately 10
water treatment systems is the possibility of the presence ppm chemical. New developments include
of Normally Occurring Radionuclide Material ·(NORM). pressurised flotation units to eliminate emissions
This will obviously effect the design of the system with of volatile organic carbons and new floating
respect to waste disposal, safe operation and skimmer designs to enable their use on floating
decommissioning. production facilities.

During the life of an oiltield the produced water is As plate separators are not particularly effective
unlikely to provide enough water to satisfY all the water at removing droplets smaller than 50 microns it
injection needs. It will have to be supplemented by either seems likely that a minimum of flotation or
seawater or aquifer water for offshore developments. The hydrocyclone packages will be required for most
additional water source may require deaeration.. Severe produced water injection systems, expect pOssibly
scaling problems may also result from the mixing of the for high permeability strata as p'resent in some of
waters. This will also influence the process selection and the North Sea Palaeocene reservoirs. Thus in
process design. The solid/liquid hydrocyclone may be terms of process equipment the produced water
required when signiticant tines and sand production injection system is likely to be at least as
occurs. expensive as system for surface disposal. It also
means that the produced water should also be of
Figure 10 illustrates the relative cost and droplet size sufficient quality for surface discharge should
removal of some of the process options. Basically, the injection have to be periodically halted. This of
tiner the droplet removal the greater the cost. course assumes discharge regulations are not
tightened a factor which will increase the cost-
Hydrocyclones and flotation units have become the effectiveness of PWRI.
workhorses for offshore produced water treatment to meet
existing surface discharge regulations in the North Sea
and in the USA. CHEMICAL TRF.ATMENTS

In the case of hydrocyclones, separation is achieved by The performance of a produced water treatment
accelerated settling (up to 3000 times that of gravitational and injection system may be dependent on a
force) caused by centrifugal action inside the number of chemicals. Chemicals may be required
hydrocyclone. The hydrocyclones can be used for to control bacteria, scaling and corrosion within
solidlliquid separations and liquid/liquid separation. If the system. Chemicals such as polyelectrolytes
the sand is removed prior to deoiling, the wear on the and demulsitiers may be required for the process
liquid/liquid hydrocyclones will be reduced. However the packa~es. An important factor in the design of
specific design of the hydrocyclone and the use of linings the system is the selection of effective and
will minimise erosion. The advantage of removing the oil compatible chemicals. The entire suite of
before the sand is that it reduces the shearing of the oil chemicals used constitute a chemical system
droplets. The solid/liquid and liquid/liquid hydrocyclones which has to be optimised, both in terms of the
will have significantly different geometries. The cone combination of chemicals used and the individual
angle is much less for the Iiquid/liquid hydrocyclone dose rates. Corrosion inhibitors are notorious
enabling more residence time and reducing the shear. for contributing to the formation of fine
Hydrocyclones may operate directly downstream of the emulsions which make separation more difficult.
three phase separator and at pressures of 60 psi upwards.
Alternatively they may be used downstream of a low
pressure separator or surge tank in which case a positive
displacement type pump such as the progressive cavity
pump which can develop the required head without
excessive shear of the dispersed oil droplets.

716
DATA IS REOUIRED ON PROCESS PACKAGE
PERFORMANCE WITH RESPECT TO ABSOLUTE OIL CONCLUSIONS
DROPLET AND PARTICLE SIZE REMOVAL, NOT
.lUST TOTAL OIL-IN-WATER REMOVAL In light of current uncertainty it to would appear
pmdent to reduce risk by targeted research
In the case of produced water injection the performance projects, particularly with. respect to injection
of the process packages in terms of their absolute water quality specification, process plant
particle/oil droplet removal will be vital in terms of performance, on -line monitoring, and injectivity
process selection. tests.

At present most data has been selected in terms of Water injection is key to the financial success of
performance in reducing total oil-in-water. More data most offshore developments, especially in high
needs to be collected on the long term operation of the production cost areas such as the North Sea.
process packages in terms of maintaining a consistent Although a water injection system including
water quality with respect to absolute maximum injection wells may account for only 5% of total
particle/droplet sizes. capital expenditure in an offshore field
development it can double the oil recovered.
Relatively short term upsets in that respect could cause Thus when considering the potential losses and
significant injectivity decline. gains with respect to produced water injection it
would appear, at present, pmdent to err on the
Developments in on -line monitoring systems will be side of caution with respect to injection water
useful in this respect. quality and process plant selection.

Developments in primary separation techniques

should not be overlooked as potential means of
THE USE OF DECISION ANALYSIS TO REDUCE RISK easing the problem of produced water treatment.
AND UNCERTAINTY
REFERENCES
Over and above the initial uncertainty of accurately
predicting water cuts in the early stages of field CALLAGHAN, D, ET.AL. (1990).
development, this paper has raised uncertainty about key Characterisation Of Residual Hydrocarbons In
factors concerning produced water injection such as the Produced Water Discharged From Gas Production
injection water quality specification and the performance Platforms. SPE Paper No. 20881.
of the process packages in term of absolute particle and
oil droplet size removal. CLIFFORD, P.J. ET. AL. 1990 Modelling The
Vertical Confinement of Injection Well Thermal
Decision analysis, and the associated use of decision trees, Fractures. SPE Paper No. 20741.
is a means of highlighting such uncertainties in the
decision making during design, and also of integrating DORRESTIJN, L.W. (1991). Injection of
technical and financial aspects of the project (fig 11). unfiltered Produced Water In the Helder Field.
Water Management Offshore Conference. October
Not only is there technical and economic uncertainty, the 1991.
problems of decision making are compounded by bias.
GOIlEC, M.L & BIGLARBIGI, K (1991).
The use of decision trees forces engineers to quantify the Economic Effects of Environmental Regulations
probability of their advice being wrong. It also enables On Finding And Ileveloping Crude Oil In The
assessment of the impact of that advice being wrong on U.S.A. J.P.T. Jan 91.
the NPV of the oilfield development. As a consequence it
highlights which uncertainties it would be pmdent to GRIFFIN, J.M, ET.AL. (1993). Measuring The
reduce by doing injectivity tests and pilot tests to obtain Environmental Practices Of An Oilfield Unit.
better information on reservoir response. AFS - Texas Chapter 3rd Annual "Produced

717
Water Seminar".

HANSEN & DAVIES. (1993). Review Of Potential

Technologies For The Removal Of Dissolved Components
From Produced Water. Water Management Offshore
Conference. October 1993.

KING (1993). Maureen Produced Water Injection. SPE

26703.

UTILETON, J. (1992) Reservoir Analysis, Marginal

Fields, Environmental Compliance Drive North Sea R &
D. Petroleum Engineer International. September.

PARSONS, A.R. (1988) Waterflooding A thick

Heterogeneous Reservoir in The Los Angeles Basin. A
Case History. J.P.T. Nov.

PARKER, D.H., ET.AL. (1991). Oil Production And

Water Management In Oman. SPE paper No. 23322.

SIMPSON, A. ET.AL. (1991) Advances In Forties Field

Water Injection. SPE Paper No. ~3140.

PETERSON, J.P & JORGENSEN. J. (1991). Produced

Water Disposal Trials. Water Management offshore
Conference. October 1991.

STEPHENSON, M.T. (1992) Components Of Produced

Water: A Compilation of Industry Studies J.P.T. May.

SWEENEY & PAIGE (1993). Produced Water Re-

Injection - Understanding The Problems. Water
Management Offshore Conference. 1993.

WOJTANOWICZ, A.K.,ET.AL (1991). Oilwell Coning

Control Using Dual Completion With Tailpipe Water Sink.
SPE Paper No. 21654.

WOLODKIEWICZ (1993). Arresting Production Decline

In A Mature North Sea Oilfield. SPE 26786.

718
 FIe.. 1.

PRODUCED WATER TREATMENT

EARLY PLANNING REQUIRED TO
AVOID BOTTLENECKS

INCREASING PRODUCED WATER VOLUMES. A N:SEA FIELD

BWPD (THOUSANDS)
3OOr------------------,
250

200

150

100

50

o L...IooooO::::'--_ _- - - ' ----l...- -'-'

1978 1984 1990

UNCERTAINTY AND RISK

UNCERTAINTY

719
 RELATIONSHIP OF OILFIELD DEVELOPMENT
NET PRESENT VALUE TO INJECTION WATER
QUALITY SPECIFICATION

Net
present
value

Increasing stringent injection

water quality specification

PRODUCED WATER INJECTION

SYSTEM CONSIDERATIONS
PROCESS
_ Process selection
...... _ Addition of supplementary I--
PRODUCTION water sources INJECTION
_ Produced fluids - Chemical treatments _ Reservoir
characteristics characteristics
_ 8caling _ Injected fluid
_ Souring characteristics

_ Production • Injection water

quality specification
techniques
_ 8caling
_ Solids production
_ Chemical treatments
_ Corrosion
_ Souring
_ Chemical treatments
_ Corrosion

720
 PRODUCED WATER. MUCH
HIGHER AND MORE VARIABLE
TOTAL SUSPENDED MATTER
THAN N.SEA SEAWATER
SOURCE WATER TOTAL SUSPENDED MATTER COMPARISONS
ISM. MGlL.
60

50

40
-
MINIMUM

MAXIMUM
Ill!!!ilII
AVERAGE

30

20

10

0
N.SEA AQUIFER PRODUCED RIVER

INJECTION WATER QUALITY

SPECIFICATION
* PORE NETWORK
MODELLING
* RESERVOIR
HETEROGENEITY
EVALUATION
* ON-SITE COREFLOOD AT
RESERVOIR CONDITIONS
*QUANTIFICATION OF
DEPTH OF PERMEABILITY
DAMAGE
* THERMO-ELASTIC
FRACTURING STUDIES
* SWEEP EFFICIENCY
EVALUATIONS
* INJECTIVITY INDEX
CALCULATIONS

721
 INJECTIVITY MODELLING AND
INJECTION WATER QUALITY SPECIFICATION

FRACTIONAL INJECTIVITY INDEX CURVES

FRACTIONAL INJECTIVITY INDEX

0.8

0.6 -

0.4 -
H
0.2 -UNIT 3
+UNIT 1
o'--------'--'--'-----'---'----'--'---'---'-----'-----'-----'---'
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

ABSOLUTE FILTRATION. MICRONS

ON-SITE COREFLOOD TESTS TO

QUANTIFY DEPTH AND EXTENT
OF PERMEABILITY DAMAGE
.PERMEABILITY (% OF INITiAl)
120 , - - - - - - - - - - - - - - , "DAMAGED ZONE"

"UNDAMAGED ZONE"
100f-,--_ _- - - - _ - -
80

60

40

20

CUMULATIVE PORE VOLUMES

722
 RELATE INJECTION WATER
QUALITY TO INJECTIVITY INDEX
AND SWEEP EFFICIENCY
PERMEABILITY RELATED TO FLOW CAPACITY & POROSITY
% OF TOTAL

40 • POROSITY
11II FLOW CAPACITY

30

20

10

o
PERMEABILITY
------Ili>

ric.. /0

DROPLET SIZE REMOVAL AND

RELATIVE COST OF PROCESS
OPTIONS
(AFTER PARKER HAL, AND CALLAGHAN HAL.)

I-
en
o
o
ill
>
!;;(
....J
ill
IY

0.1 1 10 100 1000

DROPLET DIAMETER, MICRONS

723
 DECISION ANALYSIS. A
FRAMEWORK FOR DECISION
MAKING UNDER UNCERTAINTY:
NP'
HIGH GOOD SWEEP
INJECTIVITY EFFICIENCY $
DECLINE
FINE $
FILTRATION
$
$
$
COARSE
FILTRATION $

$
LOW
INJECTIVITY
DECLINE $
DECISION III

EVENT
(PROBABILITIES)

724