SPE 68309

Determination of In-Situ Precipitation of Barium Sulphate During Coreflooding

John E. McElhiney, SPE, 20/20 Reservoir Resources and Robert D. Sydansk, SPE, Kurt A. Lintelmann,
William M. Benzel, SPE, and Kimberly B. Davidson, Marathon Oil Company

Copyright 2001, Society of Petroleum Engineers Inc.

reservoir-scale flooding with low sulphate seawater, a
This paper was prepared for presentation at the SPE 3rd International Symposium on Oilfield favorable consequence of the barium sulphate precipitation is
Scale held in Aberdeen, UK, 30–31 January 2001.
that the sulphate scaling potential is substantially reduced at
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
the producing wells. This work represents a first step in
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to generating experimental data that can be input into a
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 mathematical simulator for predicting the propagation rate and
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
the breakthrough time, under reservoir conditions, of the
for commercial purposes without the written consent of the Society of Petroleum Engineers is sulphate ions of the injection water -- both for low and high
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous sulphate content injection waters.
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Introduction
The production problems caused by mineral scale in oil
Abstract production operations have long been known. Among the most
In-situ barium sulphate precipitation was observed during onerous of all scaling problems is that of sulphate scales,
coreflooding experiments conducted at frontal velocities of particularly barium sulphate scale. This is a difficult scaling
0.31 m/day in fired Berea sandstone cores. The precipitate was problem because of the low solubility of barium sulphate in
observed during a laboratory study on the transport of sulphate most fluids and the commensurate low reactivity of most acids
ions contained within the injection water that displaced a with barium sulphate scale. Deposition of barium sulphate into
synthetic formation water containing barium ions. Two a continuous scale surface on production tubulars exposes
separate cores were 100% saturated with formation water very little surface area for treatment by chemicals, and
similar to that of a West African offshore reservoir (240 ppm therefore this scale is almost impossible to remove once it is
barium and 230 ppm strontium) and were respectively flooded deposited. The most popular approach to addressing the
with synthetic seawater containing 2860 ppm sulphate and barium sulphate scale problem has been to retard or prevent
synthetic low-sulphate (desulphated) seawater containing 36 the formation of this scale in the first place. Scale inhibitors
ppm sulphate. Careful measurement and comparison of flood that are periodically pumped down production wells and into
effluent profiles of precipitating and non-precipitating ionic the producing formation for short distances around the
species such as sulphate, barium, strontium, etc. indicates that wellbore have been developed and are widely utilized. The
barium sulphate precipitated in situ during the flooding inhibitor contacts the formation and is adsorbed onto the
experiments. In the case of the seawater flood, approximately reservoir petrofabric. It is later slowly released into the
27% of the barium that was contained in the formation water produced fluids, thereby inhibiting the formation of sulphate
resident in the core was precipitated. In the case of the low scales for some period of time, usually several months. When
sulphate seawater flood where the propensity for barium the inhibitor concentration levels fall too low to be effective,
sulphate precipitation was low, results were less dramatic and the well is again squeezed with chemical and the cycle is
determination of the exact amount of precipitated barium repeated. This technique is widely known as
sulphate was less certain, e.g., approximately 10% of the ‘squeeze inhibition.’
barium was precipitated. Temperature and pressure conditions Squeeze inhibition is effective in a wide variety of
for these coreflooding experiments were ambient temperature situations; however, it has some drawbacks. The squeeze
(~70°F) and atmospheric pressure. The Berea cores utilized chemical, often a phosphonate or high molecular weight
were 5.08 cm in diameter, 15.2 cm long, and averaged 112 md sulphonate, is usually dissolved and diluted in water for
permeability and 17.5% porosity. transport down the wellbore. The water utilized is provided by
Results of this coreflooding study indicate that in-situ a nearby source, either an aquifer or seawater. Because our
precipitation of barium sulphate occurs while even low- discussion is focussed on offshore oilfield seawater injection
sulphate seawater is flooded through core material that is for pressure maintenance or waterflooding, we will confine
saturated with a formation water containing barium. During
2 J.E. MCELHINEY, R.D. SYDANSK, K.A. LINTELMANN, W.M. BENZEL AND K.B. DAVIDSON SPE 68309

our comments to inhibitor chemicals mixed in raw seawater. experienced sulphate scaling problems and production is
Raw seawater contains approximately 2800 mg/L sulphate mature in terms of water-cut. Since the commercial testing at
ions. The actual concentration varies around the world Brae Alpha, several sulphate removal plants have been built
depending upon specific seawater conditions. Since sulphate around the world and several are in design and/or contract
scale is the target of inhibition, the inhibitor in the squeeze stages. The largest of these, a 400,000 BWPD plant5 yet to
material must be concentrated enough to offset the sulphate come onstream, will be at Elf’s Girassol project offshore
ion borne by the raw seawater. The source of barium ions is Angola. This project is a large project with subsea wellheads
the formation water that co-habits the reservoir with the oil located significant distances from the FPSO that will collect
that is to be produced. As the flooded seawater finds its way to the produced oil. With the advent of subsea wellheads,
the producing well, it mixes with the formation water and horizontal laterals, and the complexity of pipework in
sulphate scale is formed. The fact that most inhibitor solutions produced-oil collection systems comes the difficult logistics of
are aqueous-based causes a second, more important, problem. intervening these systems to place squeeze inhibitors. In these
The formation around the producing oil well is highly oil systems, the advantages of desulphation of injected seawater
saturated and when the inhibitor is injected, this saturation is seem obvious. The advantages of sulphate removal by
reversed; i.e., the producing formation becomes highly water- nanofiltration membranes in conjunction with offshore
saturated. This deleterious change in saturation conditions developments should be borne out by the future performance
causes a reduction of oil relative permeability; thus causing a at Girassol.
reduction in oil production rate. As the inhibitor is slowly During original development of the sulphate-removal
desorbed from the formation into the produced stream of oil membrane, the design provided for removal of all but 100
and water, the oil relative permeability usually recovers slowly mg/L of sulphate from raw seawater. The design and
to its original condition before the well was squeezed. performance of the membrane have now evolved to removal
Unfortunately, by the time the oil production has recovered, of all but approximately 40 mg/L sulphate from injected
the inhibitor levels have often fallen to an ineffective level and seawater. The basics of operation of this system are reported
the whole process needs to be repeated, again re-saturating the elsewhere,6 but in essence 75% of the seawater flowing into
near-well region of the oil well with a high water saturation the sulphate-removal plant emerges containing only 40 mg/L
level. There are serious economic implications that result from sulphate. The remaining seawater (25%) contains
both the loss of oil production rate during application of the approximately 11,000 mg/L sulphate and is discarded
squeeze treatment and the slow recovery of oil production rate overboard into the sea or injected into an aquifer or a water
during release of the squeeze inhibitor, but these economics injector far from the producing well. Even at these
will not be discussed herein. substantially lower levels of sulphate, a concern remains that
The effectiveness of squeeze inhibition deteriorates as the eventually the low sulphate seawater (LSSW) may scale a
concentration of barium, strontium and calcium ions increase producing well shut. The converse problem, formation water
in the formation water. In the early 1980s, Marathon Oil with low barium contents of 55 mg/L, has scaled-up laterals in
Company United Kingdom (MOUK) began development of producing wells in other fields7 flooded with seawater.
the Brae Field complex in the North Sea. Barium levels in the If barium sulphate scaling conditions exist at producing
formation water were reported by Hardy et al.1 to be in the 500 wells, then scaling conditions must also be present in the
– 2500 mg/L range in the various producing lenses of the Brae reservoir itself. If so, then precipitation of barium sulphate
reservoir. During that historical period, there were no squeeze scale will occur in situ. The resulting question is how much
inhibitors capable of effective performance at such high potential for sulphate scale formation will remain at the
barium concentration levels. Marathon subsequently producing well when LSSW is utilized as the waterflooding
developed a polyvinyl sulphonate inhibitor (PVS) that was fluid and how much delay in the formation (if ever) of any
effective at the concentrations of barium encountered in the barium sulphate scale will occur at the production well when
Brae Field complex2. Concurrently, Marathon and Dow LSSW is utilized? These questions were the motivation for the
Chemical Company began work on a modified reverse coreflooding experiments described herein.
osmosis membrane that selectively removes the sulphate ion Specifically, the intent of this initial experimental study was
from raw seawater.3 The thinking was that if the sulphate ion to conduct, for the first time to the best of our knowledge,
could be removed from the injected stream of seawater, then laboratory core flooding experiments to quantitate in-situ
sulphate scale would be prevented from formation altogether barium sulphate precipitation in reservoir rock, as a function
since one of the chemical species required would not be of sulphate concentration in the flood brine (e.g., seawater
available for chemical reaction. Subsequently, the membrane injection), when the reservoir brine contains barium ions.
system was pilot tested in the late 1980s on the Brae Alpha Stated another way, the study was specifically intended to
platform where a 120,000 BWPD plant was built for water quantitate the degree of sulphate ion retention and retardation,
injected into the South Brae Field.4 MOUK subsequently due to barium sulphate precipitation, under the experimental
utilized the low sulphate seawater from this plant for flooding conditions of the reported flooding experiments. The ultimate
at the Central Brae Field. The Central Brae Field involves a objective of such a study is to generate experimental data that
subsea completion located several kilometers from the Brae can be input into a computer simulator to quantitatively
Alpha platform. Oil production performance there has not predict the rate of injection-water sulphate-ion bank
SPE 68309 DETERMINATION OF IN-SITU PRECIPITATION OF BARIUM SULPHATE DURING COREFLOODING 3

propagation, under reservoir conditions, through a reservoir performed as follows: 1b) the core was re-flushed with six
that has formation water containing barium ions that will pore volumes of FWWB, 2b) one ml of NSSW was injected to
cause barium sulphate precipitation. In view that only a prevent scaling in the small holes of the distributor plate, 3b)
limited amount of experimental work could be conducted six pore volumes of RSW were displaced through the core and
under this first-time laboratory study, some unanticipated aliquots of fluid were automatically taken from 0.8 pore
experimental outcomes, problems, and challenges were volume to 2.0 pore volumes, 4b) six pore volumes of distilled
encountered. These factors are also discussed in this paper. water were displaced through the core followed by two pore
volumes of methyl alcohol to completely clean the core.
Experimental The objective of the first set of corefloods, 1A and 1B,
The laboratory methodology chosen to investigate the problem involving the first core was to determine the dispersion
of in-situ barium sulphate precipitation is coreflooding characteristics of the core and to measure the loss of sulphate
utilizing laboratory surrogates of raw seawater (RSW) and of due to precipitation when flooding with RSW. In order to
low sulphate water (LSSW) to displace a laboratory surrogate calculate a material balance on the sulphate loss due to barium
of a typical offshore, West African formation water. LSSW is sulphate precipitation from the coreflood 1B, the dispersion
representative of RSW after treatment by a sulphate-removal profile without sulphate loss from the coreflood 1A must be
membrane plant; i.e., the resultant permeate water that is known since it is the baseline. At the outset, we expected the
injected into the oil reservoir. A third brine was also used and sulphate effluent curve from coreflood 1B to lag that of
had the same composition as the raw seawater except with the coreflood 1A since we thought scale precipitation would
sodium sulphate replaced with sodium chloride. We will refer manifest itself much like the in-situ adsorption loss of a
to it as no sulphate seawater (NSSW). We wanted the species. The first set of corefloods, 1A and 1B, was performed
laboratory experiments to reflect as much representative field utilizing raw seawater (RSW) as the floodwater. The second
water chemistry as possible. The compositions of the waters set of corefloods, 2A and 2B, carried out in the second core,
utilized in this study are shown in Table 1. Two formation was performed utilizing low sulphate seawater (LSSW) as the
waters were utilized: one with barium (FWWB) and one that floodwater, but otherwise the second flooding experiment set
did not contain barium (FWNB). The corefloods were exactly replicated the first set of flooding experiments. The
conducted in fired Berea sandstone cores that were 5.08 cm in two extremes of the flooding experiment sets were chosen to
diameter and 152 cm in length. No oil was present in the illustrate the severity of barium sulphate scale precipitation in
cores. The cores were taken in longitudinal sequence from a the case of flooding with raw seawater and to illustrate the
longer Berea core so they were similar in their mineralogy, substantial reduction of sulphate scale precipitation in the case
etc. The core used for coreflood #1 had 108 md permeability. of flooding with low sulphate seawater.
The core used for coreflood #2 had 116 md permeability. Both Details of the analytical procedures are contained
cores had approximately 17.5% porosity. in Appendix I.
Each coreflood was conducted at 0.31 m/day superficial
(Darcy) velocity. The coreflooding apparatus was Scaling Potential of the Selected Mixed Waters
straightforward. A Ruska pump displaced the fluids through The thermodynamics of scaling were first checked to see
the core that was held in a Hassler holder with a pressurized which species would precipitate in situ. The OLI ScaleChem
rubber sleeve surrounding the core. The inlet face of the core software8 was utilized for these calculations. Table 1 gives the
held a metal plate that uniformly distributed the injected fluid composition of the various waters utilized in this work. The
evenly across the face of the core. Back-pressure was supplied scaling index (SI) for raw seawater (RSW) mixed with
to the core by tilting the coreholder in a slightly uphill position formation water containing barium (FWWB) and for low
toward the effluent end of the core. The core was initially sulphate seawater (LSSW) mixed with formation water
heated and dried in preparation for installation in the containing barium (FWWB) is shown in Figure 1. Figure 2
coreholder. A vacuum was drawn on the core overnight to illustrates the amount of scale (in mg/L) formed by mixing the
remove all air from the core. Aliquots (2.5 ml) of the emerging two combinations of waters. No strontium sulphate scale was
flood effluent were taken automatically with a sample- predicted with LSSW mixed with FWWB, and only a slight
collecting carousel and retained for later analysis. amount at mid-range mixing compositions for FWWB and
Two corefloods were performed within each core. The first RSW (less than 25 mg/L between 40 and 80 % RSW in the
coreflood, 1A, was performed as follows: 1a) the core was mixture). We corroborated these results with that of a
saturated with FWNB and approximately six pore volumes professional colleagure.9 Therefore, we assumed that we were
were displaced with FWNB at several different velocities dealing with a barium sulphate dominated scaling environment
while determining liquid permeability, 2a) two pore volumes in the corefloods described herein.
of RSW were displaced through the core and 2.5 ml aliquots
(one twentieth of a pore volume) of fluid were automatically Discussion of Results
taken from 0.8 pore volume to 2.0 pore volumes and retained Raw Seawater: The emerging sulphate effluent profile from
for analysis of sulphate in each aliquot, 3a) the core was the coreflood 1A is shown in Figure 3, as the normalized
flushed with six pore volumes of NSSW to prepare the core concentration (measured effluent concentration, C, divided by
for the next coreflood. The second coreflood, 1B, was the concentration in the floodwater, Co, vs. pore volumes of
4 J.E. MCELHINEY, R.D. SYDANSK, K.A. LINTELMANN, W.M. BENZEL AND K.B. DAVIDSON SPE 68309

effluent. The emerging sulphate effluent profile from from the first set of corefloods were not complete when the
coreflood 1B is shown in Figure 4, along with that from second set of corefloods was commenced.
coreflood 1A. Overlay of these two sulphate effluent profiles The only indication that barium sulphate precipitated in the
illustrates indeterminate, or little measurable, sulphate loss. second set of corefloods is provided by the pore volume delay
This was surprising, but upon re-analysis, we realized that the in the emergence of the sulphate ion in the effluent. Because
sulphate ion is in large surplus over barium in coreflood 1B. of the various complications described in the preceding
Therefore, we should examine the loss of barium since it paragraph, the sulphate analysis from coreflood 2A was not
would be more sensitive than sulphate in illustrating a loss or useable. The only comparator for the sulphate analysis from
lag in its effluent profile. There was no barium in the coreflood 2B is the effluent sulphate analysis of coreflood 1A.
formation water in coreflood 1A since it was used to provide When the normalized concentration of sulphate in the
dispersion characteristics of the core. Calculation of scaling coreflood 2B is compared with the normalized concentration
tendencies for barium and strontium for the temperature and of sulphate from the coreflood 1A, there is a noticeable lag in
pressure conditions of the coreflood indicate little strontium the coreflood 2B concentrations measured in pore volume
precipitation to be expected; therefore strontium could be used time. See Figure 6. This would be expected with a species that
as a non-precipitating tracer in coreflood 1B. Re-analysis of adsorbs or precipitates in situ in the core.
the aliquots for barium and strontium for coreflood 1B and Integrating the difference between the two sulphate effluent
graphing of the effluent profiles for each are shown in curves in Figure 6, the loss in the second coreflood is
Figure 5. estimated to be 285 mg, or about 10% of the sulphate
Integrating between the effluent profiles for barium and originally injected in coreflood 2B. The scaling tendency for
strontium from coreflood 1B indicates that 3,350 mg of the LSSW/FWWB system illustrated in Figure 1 indicates that
barium was precipitated in the coreflood. The formation water the potential for scaling is small to begin with. Perhaps our
in the core held 12,340 mg of barium, therefore approximately expectation of being able to measure losses in this system was
27% of the available barium was precipitated. The amount of unrealistic. An important result of the second set of corefloods
sulphate consumed in precipitating the barium as barium is that it suggests that flooding with desulphated water and the
sulphate, 2350 mg, is insignificant when compared to the total associated barium sulphate precipitation will virtually
amount of sulphate, 156,240 mg, in one pore volume of eliminate the tendency of the system to scale at all. Placing the
floodwater, and indicates why comparison of the sulphate results of the second coreflood in a true reservoir setting
profiles does not yield the sensitivity to sulphate loss that would seem to indicate that if precipitation of barium sulphate
was sought. occurs within the reservoir rock itself, then further protection
(e.g., scale inhibitor squeezes) of the producing wellbores is
Low Sulphate Seawater: In the second set of corefloods, the not required. Breakthrough of the sulphate in the injected
same sequence in flooding was utilized. The raw seawater LSSW will be delayed in time and will occur at very low
(RSW) was replaced with low sulphate seawater (LSSW) as levels later in the producing history of the wells, if at all. This
the displacing floodwater and the same measurements were fact may be further understood by the shift in the peak of the
made for sulphate, barium and strontium in the effluents from curve for precipitated amounts of barium sulphate scale
the two corefloods, 2A and 2B. However, complications arose formed with LSSW illustrated in Figure 2.
in the second set of corefloods that were unanticipated. There The lag in sulphate concentration in the effluent from
was approximately 16 ppm of sulphate in the formation water coreflood 2B when compared to coreflood 1A is analogous to
since the field sample of West African formation water the lag in barium effluent concentration when compared to the
contained some sulphate. This proved to be deleterious to the non-precipitating species strontium in the coreflood 1B. In
analysis in the second set of corefloods because the low both cases, it is the ionic species that is in diminished quantity
sulphate seawater (LSSW) contained only 36 ppm sulphate in the particular coreflood that demonstrates the pore volume
and the analysis for barium sulphate precipitation was lag in the effluent profile. There is of course a slight difference
dependent upon accurately identifying small changes in the in the two cores that affects the dispersivity of the emerging
sulphate ion. This problem, and the fact, that the high chloride effluent. This slight difference in dispersivity can be viewed in
concentrations interfered with sulphate quantification by ion Figure 7 where the strontium effluent concentrations for
chromatography, caused confusion in sorting out the sulphate flooding experiments 1B and 2B are graphed vs. pore volume.
losses. Furthermore, the barium ion is in tenfold molar surplus Reliance upon the quantitative measure of barium sulphate
to the sulphate ion in the second set of corefloods. This latter precipitation in coreflood 2B as being exact may be
factor makes it necessary to derive the determination of inadvisable since results from two different cores were
barium sulphate precipitation losses from analysis of the utilized. The expected characteristic of a lag in pore volume
sulphate effluent curves since the sulphate analysis is now far time of the appearance of the precipitating species seems to be
more sensitive than the barium analysis to an in-situ corroborated however.
precipitation loss. The realization of these factors and the total
dependence upon the sulphate analysis was not apparent at the Scanning Electron Microscope Analysis: In an effort to
outset of the second set of corefloods because the analyses confirm the precipitation of barium sulphate within the core
through the direct observation of precipitated barium sulphate
SPE 68309 DETERMINATION OF IN-SITU PRECIPITATION OF BARIUM SULPHATE DURING COREFLOODING 5

crystals, the scanning electron microscope (SEM) analytical precipitation of barium sulphate occurs when synthetic
technique was employed. The core of corefloods 1A and 1B raw seawater containing sulphate is mixed in situ with a
that involved raw seawater flooding was cut open along its formation water that contains a significant amount of
axial length, broken at several locations to create fresh dissolved barium ions.
surfaces and graphite coated for examination under a SEM 2. During the experimental work of this study, the
with an X-ray spectroscopy attachment. SEM observations on precipitation loss of barium sulphate during raw seawater
the core verified that barium is not present in the matrix of the coreflooding was adequately characterized by
Berea sandstone. measurement of the effluent profiles of sulphate ion,
The barium was observed in three different habits. barium ion, and a tracer ion. In this study, the non-
However, none of the three were identified as barium sulphate. precipitating tracer specie was the strontium ion.
Figure 8 shows a cluster of barium-containing crystals 3. Determination of the amount of barium sulphate
attached to quartz. Kaolinite is present next to the assemblage. precipitation during laboratory coreflooding with low
Because no anion was detected during the SEM X-ray sulphate seawater was much more difficult to
analysis, a reasonable hypothesis is that the crystals are characterize. While precipitation losses were identified,
barium carbonate. This identification was not the measurement techniques utilizing effluent profiles of
confirmed however. sulphate, barium and strontium require improvement
The authors believe that barium sulphate was precipitated during future work so that a higher degree of certainty and
during the flood experiment. There are several possible precision can be established. The addition of a radioactive
reasons why barium sulphate was not detected during the isotope with similar sulphate scale precipitation
analysis. If the precipitation produced micrograins, the characteristics should be investigated as a means of
average grain size might be much smaller than the resolution improving the analyses.
of the SEM employed in this work. Alternatively, the 4. Coreflooding techniques can expected to provide the
precipitation may have occurred along high permeability paths necessary sulphate scale precipitation data for input into
not sampled. Or even more likely, the total amount of barium reservoir simulations that in turn would provide field-
sulphate precipitated is small, in the milligram range as wide estimates of breakthrough times of sulphate in the
discussed above. When spread over the length and pore injected water at production wells. A delay in
surface of the entire core, its concentration is so sparse, it breakthrough times resulting from in-situ precipitation of
simply could not be detected by the SEM technique. barium sulphate shows promise of offering nearly
complete protection from scaling at production wells
Implications Derived from This Study: Results of this when low sulphate seawater is utilized.
coreflooding study strongly suggest that in-situ precipitation
of barium sulphate occurs while sulphate-containing seawater Nomenclature
is flooded through core material that is saturated with a
formation water containing barium. During reservoir-scale C = ionic specie concentration in mg/L
flooding with low sulphate seawater, a favorable consequence Co = initial ionic specie concentration in mg/L
of the barium sulphate precipitation is that the sulphate scaling FWWB = formation water with (containing) barium
potential is reduced at the producing wells. This work FWNB = formation water with no (not containing)
represents a first step in generating experimental data that can barium
be input into a mathematical simulator for predicting the LSSW = low sulphate seawater, i.e. raw seawater
propagation rate and the breakthrough time, under reservoir treated by the membrane process
conditions, of the sulphate ions of the injection water -- both NSSW = LSSW containing no sulphate
for low and high sulphate content injection waters. RSW = raw seawater, i.e. seawater untreated by the
The sulphate scale precipitation data of this coreflooding membrane process
study, or of more sophisticated and advanced similar
coreflooding experiments, could be used as input into Acknowledgments
computer reservoir simulations that would provide
quantitative field-scale estimates of the breakthrough times at The authors gratefully acknowledge the diligence and care
production wells for the sulphate ions that are contained in the provided in this work by two technicians at Marathon Oil
injected waterflood brine. As a result of this study, the Company’s Petroleum Technology Center. The contributions
expected substantial delay in sulphate breakthrough times of Beth Nalty and Katherine Wilson who assisted with
resulting from the in-situ precipitation of barium sulphate for analytical measurements and tabulations is appreciated. A
seawater injection shows promise of offering nearly complete special thanks to Lois Fitzpatrick, who ably assembled the
protection from barium sulphate scaling at production wells manuscript like so many others she has done in her many
when low sulphate seawater injection is utilized. years of service for Marathon’s Petroleum Technology Center.
Finally, we thank Marathon Oil Company for permission to
Conclusions publish this work and The Dow Chemical Company for
1. Laboratory coreflooding has demonstrated that in-situ financial support of the primary author.
6 J.E. MCELHINEY, R.D. SYDANSK, K.A. LINTELMANN, W.M. BENZEL AND K.B. DAVIDSON SPE 68309

References 1.36 mM sodium bicarbonate in deionized distilled water

1. Hardy, J. A., Barthorpe, R. T., Plummer, M. A., and J. S. flowing at 2.0 mL/min. The ions were detected with a Dionex
Rhudy, “Control of Scaling in the South Brae Field”, OTC pulsed electrochemical detector (in the conductivity mode),
7058, presented at the 24th Annual Offshore Technology with background conductivity suppression using an ASRS-I
Conference, Houston, TX, May 4-7, 1992. 4.0 mm anion self-regenerating suppressor in the recycle
2. U. S. Patent, 5092404, “Polyvinyl Sulfonate Scale Inhibitor,”
Falk, D. O., Dormish, F. L., Beazley, P.M., and Thompson, R.
mode. It should be noted that the sulphate quantification under
G., March 3, 1992. these conditions was problematic due to high concentrations
3. U. S. Patent, 4723603, “Preventing Plugging by Insoluble Salts of chloride in the samples. High concentrations of chloride can
in Hydro-Carbon Bearing Formations and Associated Wells”, overload the ion exchange sites in the column packing, which
M. A. Plummer, issued in the U.K. No. 2221700. can result in poor resolution and erroneous conductivity
4. O’Donnell, Keith, “Membrane Technology Works on North Sea measurement. We tested the precision of the sulphate analysis
Platform”, Oil and Gas Journal, Pennwell Publishing, with, and without, chloride removal using a silver impregnated
December 2, 1996. preparative cartridge prior to the analysis. The sulphate values
5. Vu, V. K., Hurtevent, C., and R. A. Davis, “Eliminating the were not significantly different between samples with the
Need for Scale Inhibition Treatments For Elf Exploration
Angola’s Girassol Field”, SPE 60220, 2000 Second
chloride removed from an undiluted water sample and the
International Symposium on Oilfield Scale, Aberdeen, U.K., same samples after a 1:10 or 1:20 dilution. With a
January 26-27, 2000. considerable number of samples to analyze, a compromise
6. Vu, V. K., Latapie, D., and R. A. Davis, “Barite Scale between sensitivity and speed was reached by analyzing all of
Prevention for Elf Angola’s Girassol Field Using Sulphate the samples diluted 1:20 in deionized distilled water.
Removal Technology,” Deep Offshore Technology Conference, The metal ions were analyzed by inductively coupled argon
Stavanger, Norway, November 1999. plasma (ICAP) spectrometry using a Jarrell-Ash ICAP 9000
7. Matharu, A. P., “Strategies for Scale Management in Horizontal instrument. The samples were analyzed using a 1:50 dilution
Wells: Experience in the Alba Field,” IBC Scale Conference, with a scandium internal standard to correct for any variations
Aberdeen U.K., January 1998.
8. OLI Systems, Inc., Morris Plains, NJ, USA.
in the amount of sample introduced to the instrument, and to
9. Personal Communication, Dr. Mason Tomson, Rice University, monitor system performance. ICAP functions by nebulizing
Houston, TX., August, 2000. the liquid sample and heating it in the argon plasma resulting
in emission of unique wavelengths of light energy for each
Appendix I – Analytical Procedures metal. The amount of emitted light is proportional to the
Ion chromatography (IC) was used for the determination of concentration of each metal ion in solution.
sulphate ion concentrations in solution. The columns used for Precision of the sulphate analysis was about ±33 at the 2800
the separation were an AG4A 4.0 mm precolumn and AS4A- ppm level and ±0.38 at the 15 ppm level. The precision of the
SC 4.0 mm analytical column, both from Dionex Corporation. ICAP method for the barium analysis was ±3.8 ppm and the
The mobile phase consisted of 1.44 mM sodium carbonate and strontium analysis ±3.7 ppm at the 220 ppm level.
SPE 68309 DETERMINATION OF IN-SITU PRECIPITATION OF BARIUM SULPHATE DURING COREFLOODING 7

Table 1 – Compositions of the Employed Laboratory Waters

Ionic Formation Formation Raw Low-SO4

Specie Water Water Seawater Seawater
With Ba++ Without Ba++ [RSW] [LSSW]
[FWWB] [FWNB] (mg/L) (mg/L)
(mg/L) (mg/L)

Sodium 37,719 37,719 11,424 10,665

Potassium 46 46 400 190
Calcium 4,000 4,000 435 108
Magnesium 873 873 1,370 475
Barium 240 0 0 0
Strontium 230 230 0 0

Chloride 68,083 67,959 20,635 18,170

Sulphate 16 16 2,860 36

TDS 111,207 110,843 37,124 29,644

3.5

3.0

2.5
BaSO4 SI

2.0

1.5

1.0

0.5 RSW
LSSW

0.0
0 20 40 60 80 100
Percent RSW or LSSW

Fig. 1 – Barium sulphate scaling index.

8 J.E. MCELHINEY, R.D. SYDANSK, K.A. LINTELMANN, W.M. BENZEL AND K.B. DAVIDSON SPE 68309

400

mg/L Predicted BaSO4 Scale

RSW
300 LSSW

200

100

0
P e rc e n t R S W o r L S S W

Fig. 2 – Seawater mixing with FWWB water - 68°°F, 15 psi, scaling tendency calculations - OLI ScaleChem.

1.2

1.0
C/Co Sulphate

0.8

0.6

0.4

0.2

0.0
0.0 0.5 1.0 1.5 2.0 2.5
Pore Volumes Flood Effluent

Fig. 3 – Sulphate effluent concentration from Coreflood 1A.

SPE 68309 DETERMINATION OF IN-SITU PRECIPITATION OF BARIUM SULPHATE DURING COREFLOODING 9

1.2

1.0
C/Co Sulphate
0.8

0.6

C orefloo d 1A
0.4
C orefloo d 1B

0.2

0.0
0.0 0.5 1.0 1.5 2.0 2.5
P o re V o lu m es F lo o d E fflu en t

Fig. 4 – Sulphate effluent concentration from Corefloods 1A and 1B.

1.2

1.0
Barium
Strontium
0.8
C/Co

0.6

0.4

0.2

0.0
0.0 0.5 1.0 1.5 2.0
Pore Volumes Flood Effluent

Fig. 5 – Barium and strontium effluent concentrations from Coreflood 1B.

10 J.E. MCELHINEY, R.D. SYDANSK, K.A. LINTELMANN, W.M. BENZEL AND K.B. DAVIDSON SPE 68309

1.2

1.0
C/Co Sulphate

0.8

0.6

0.4
Coreflood 1A
Coreflood 2B
0.2

0.0
0.0 0.5 1.0 1.5 2.0 2.5
Pore Volumes Flood Effluent

Fig. 6 – Sulphate effluent concentration from Corefloods 1A and 2B.

1.2

1.0
Coreflood 1B
Coreflood 2B
C/Co Strontium

0.8

0.6

0.4

0.2

0.0
0.0 0.5 1.0 1.5 2.0
Pore Volumes Flood Effluent

Fig. 7 – Strontium effluent concentration from Corefloods 1B and 2B.

SPE 68309 DETERMINATION OF IN-SITU PRECIPITATION OF BARIUM SULPHATE DURING COREFLOODING 11

Mag 2000X. Suspected barium carbonate

Magnification 1500X crystals at center, kaolinite plates at upper right.
25.0kV
25.0kV 200
200 nm
nm 25.0kV
25.0kV 200
200 nm
nm

Fig. 8 – Photomicrographs of Barium Crystal Cluster.