-
Mobility control processes inject a low-mobility displac- ing
agent to increase volumetric and displacement sweep efficiency. The
two main techniques are (I) polymer flooding. whereby a small
amount of polymer is added to thicken brine and (2) foam flooding,
through which low mobilities are attained by injecting a stabilized
dispersion of gas in water. Polymer or mobility buffer drives
also
ams icel- foam sol- bili- s in
-situ
minus ROS] since they displace oil only in excess of ROS. The
mobility ratio concept illustrates how lowering the mobility
improves oil recovery.
The mobility ratio, M, between a displacing agent and a
displaced fluid is
M=hDlh,,, ._____.____................ ._.(l)
where AD is the mobility of displacing agent(s) and A,, is the
mobility of displaced fluid(s). Practical use of Eq. 1 requires
specific definitions of the quantities in the numerator and
denominator and of the conditions at which these quantities are
evaluated.
For two-phase waterflooding, common specializations of Eq. 1 are
the endpoint mobility ratio, M. the average mobility ratio, M, and
the shock mobility ratio. M*. Each definition has been used in
particular applications in the are used to displace micellar and
high-pH slugs. Fohave been used or proposed as driving agents for
mlar. solvent, and steam slugs also. For these reasons tlooding
could be discussed easily in the chapters onvent and thermal EOR
processes; however. foam stazation requires surfactants. whose
discussion belongthis chapter.
Low-IFT methods rely on injecting or forming inChapter 47
Chemical Flooding Larry W. Lake, U. of Texas
Introduction Chemical flooding is any isothermal EOR process
whose primary goals are to recover oil by (1) reducing the mo-
bility of the displacing agent (mobility control process), and/or
(2) lowering the oil/water interfacial tension (low- IFT
process).
These classifications do not exclude processes where both
effects are important, such as micellar/polymer (MP) flooding, nor
do they imply that other effects, such as wettabiiity alteration,
extraction, or oil swelling. are not present. a surface-active
agent (surfactant) which lowers oil/water IFT and, ultimately.
residual oil saturation (ROS). Proc- esscs that inject the
surfactant are called MP floods because of the tendency for
surfactants to form micelles in aqueous solutions and the
inevitable need to drive the micellar solution with polymer.
High-pH or alkaline proc- esses produce a surfactant in situ, since
these processes rely on reactions with acidic components of the
crude to generate the surfactant.
This chapter is divided into sections corresponding to mobility
control and low-IFT processes. Each section be- gins with a general
discussion of how the particular mech- anism recovers oil, followed
by material on the individual processes, and concludes with typical
oil displacement re- sults. The chapter concludes by giving
comparative screening parameters for each chemical flooding
process.
Mobility Control Processes Effect of Low Mobility on Oil
Recovery
Mobility control processes are most applicable to reser- voirs
that have substantial movable oil [oil in place (OIP) literature: M
has been used to correlate areal sweep rela- tions, M* is the most
direct indicator of viscous instability, and M is the most widely
cited value in the literature for waterfloods. Z For the special
case of a piston-like displacement all three definitions are
identical. From the general definition of Eq. I, lowering the
mobility of the displacing fluid is equivalent to lowering any of
the mo- bility ratios. This recovers oil by increasing both volu-
metric and displacement sweep efficiency.
Volumetric sweep efficiency, EV. is the PV of a reser- voir
contacted by a displacing agent div,ided by the total PV. Et, is
composed of two parts: areal, E/r, and inva- sion (vertical) sweep
efficiency, El.
-
ing propathey are growing poor oil bility coil conventof an
oth
Displacsweep cfplacementplacementsweep efLeverett ering
onfactors thEl,: howrecovery
Loweriincreahinscicncics. tcrminc in any ooil rccokeivcre
IOUclcncIc\. control peupensc agent.
link subsequently injected polymer molccutes with ones
or Xn. proc- e poly-
EOR the one
ood in- a low- itself: a from
. Many ecreas-
fects ot ter and of the
through
ter that olymer. brine. ly for is the phase.
lly all rations @on. Such tingcrs also form when M< I. but
damped out by the~~~\~cl&le mobility ratio. The viscous fingers
can contribute substantially to
rocovcry because of large-scale bypassing. Mo- ntrol agents
prevent viscous fingering cithcr in ional waterflood (polymer
flooding) or ah a part erwise inherently unstable EOR process.
ement efficiency. El, (local or microscopic ticicncy). is the
volume of oil recovered in a dis- divided by the oil volume just
before the dis- . the displacement having maximum volumetric
ficiency. The classical solution by Buckley and can be used to
describe the effect of mobility low- En. Because of the dependence
on several ere i+ no unique correspondence between M and ever.
lowering M or An results in improved oil
through larger Eo. ng M. then. results in improved oil recovery
by areal. vertical. and displacement sweep effi-
Since it is the products of these factors that dc- overall oil
recovery. this implies that an incrcasc ne may not r-csutt in a
larfc overall incrcaso in ry. particularly if one of the other
efficiencies . Even with the combination of the three effi- the
incremental oil recovered with ;I mobility
rocess must he balanced against the additional
already bound to solid surfaces. 3. The other mode is use as
agents to lower MThe first mode is not truly a chemical
flooding
ess. since the actual oil-displacing agent is not thmer. The
overwhelming majority of polymer projects have been in the third
mode, which is emphasized here.
Fig. 47. I is a schematic of a typical polymer fljection
sequence: a preflush, usually consisting ofsalinity brine: an oil
bank; the polymer solution freshwater buffer to protect the polymer
solutionbackside dilution; and, finally. chase or drive watertimes
the freshwater buffer contains polymer in ding amounts (a grading
or taper) to lessen the efunfavorable mobility ratio between the
chase wathe polymer solution. Because ofthe driving natureprocess.
polymer tloods always arc performed separate sets of in.jection and
production wells.
M is lowered in a polymer tlood by in.jecting wacontains a
high-molecular-weight. water-soluble pSince the water is usually a
dilution of an oillicldinteractions with salinity are important.
particularcertain classes of polymers. Salinity in this
chaptertotal dissolved solids (TDS) content of the aqueous Typical
values are shown in Fig. 47.2. Virtuachemical flooding propertics
dcpcnd on the concent47-2
Fig. 47.1-Schematic
For more &tailed discuxsions of E,, and El XC Chaps. 30. 43.
and 44 and Refs. 1 through 1 I. Both quantities dcpcnd on
throughput and on several fluid. petrophyai- cal. and geometric
factors. usually expressed as dimen- sionlcsa groups. However. E,.(
and El both increase as Mdecrea~-i.e.. as the mobility of the
displacing agent, A,,. decreases.
When the displacement is piston-like in a homogcnc- ous.
isotropic. horizontal medium and M> I (an UI+I\YIU- hlr mobility
ratio). the displacement front forms small perturbations. called
viscous fingers. which grow dur- rcquircd to in.jcct the viscous
mobility control PETROLEUM ENGINEERING HANDBOOK
of polymer flooding
Polymer Flooding
Polymers have been used in oil production in three modes. I.
They have been used as near-well treatments to in-
prove the performance of waler injectors or high-watercut
producers by blocking off high-conductivity Loncs.
2. Polymers also are used as agents that may be cross- linked in
situ to plug high-conductivity zones at depth in the reservoir.
These proccsscs require that polymer be in.jected with an inorganic
metal cation. which will cross- ofspccific ions rather than
salinity only. It is partcularly important to monitor the aqueous
phahcs total divalcnt
-
Fig. 47.2-Salinities from representallve oilfield brines
cation content (hardness) separately. since the latter are
usually more crltical to chemical llood properties than the same
TDS concentration. Fig. 47.2 also shows typical brine
hardncsscs.
Bccauxc of the high molecular weight ( I to 3 million is
typical) only a small amount (typically about 500 ppm) of polJ,mer
will bring about a substantial increase in water viscosity.
Furthcrmorc. several types ofpolymcrs lower mobility by lowcring
water relative pcrmcability (pcrme- ability reduction effect) in
addition to incrcaxing the water viscosity. How polymer lowers
mobility. and the inter- actions with salinity may be qualitatively
illustrated with some di5cusslon of polymer chemistry.
Polymer Types. Several polymers have been considered for polymer
tlooding: xanthan gum, hydrolyzed poly- acrylamide (HPAM).
copolymers of acrylic acid and acrylamide. copolymers ofacrylamide
and 2-acrylamide-2 Imethyl propane sulfonatc (AM/AMPS).
hydrouyethyl- cellulose (HEC). carboxymethylhydroxycthyIccllulose
(CMHEC). polacrylamide (PAM). polyacrylic acid, flu- can. dextran
polycthylencoxide (PEO). and polyvinyl al- cohol (PA). Only the
first three have actually hccn field tested: however. the variety
of entries in this partial list- ing emphasizes that there arc many
potentially suitable chcmlcals. some of which may prove more
cfkctivc than thoxc currently LIW~. Ncvcrthclcsj. virtually all 01
rhe commercially attractive polymers fall into two gcncric classes:
polyacrylamidcs and poly\accharide\ (hiopolyme1
Fig. 47.3B-Molecular structure for polysaccharlde
jbiopolymer).
Polyacrylamides. PAMs are polymers whose mono- meric unit is the
acrylamide molecule. As used in poly- mer flooding, PAMs have
undergone partial hydrolysis. which causes anionic (negatively
charged) carboxyl groups (-COO-) to be scattered along the chain.
The polymers are called partially hydrolyzed polyacrylamidcs
(HPAMs) for this reason. Typical degrees of hydrolysis arc 30% or
more of the acrylamide monomers: hence. the HPAM molecule is quite
negatively charged. which accounts for many of its physical
properties. The viscosity-increasing feature of HPAM lies in its
large molecular weight. which is accentuated by the anionic
repulsion between polymer molecules and also between segments on
the same molecule. The repulsion causes the molecule in solution to
elongate or uncoil and snag on others similarly elongated, an
effect that accentuates the mobility reduction at higher
concentrations.
If the brine salinity and/or hardness arc high. however, this
repulsion is decreased greatly through ionic shield- ing as the
freely rotating carbon/carbon bonds allow the molecule to ball up
or coil. (Fig. 47.3 shows the molccu- lar structure of
hydroxyethylcellulose.) This causes a cor- responding decrease in
the effectiveness of the polymer, since the snagging effect is
rcduccd greatly. Virtually all HPAM properties show a large
sensitivity to salinity and hardness. This is an obstacle to
usiiyg,HPAM in many reservoirs. On the other hand, HPAM 15
inexpensive. rela- tively resistant to bacterial attack. and
exhibits permea- CHEMICAL FLOODING
I
1,000 a0 d3 r @ Oo 0 I L
I )
,oc
TOTAL DISSOLVED SOLIDS, mg/l or xanthan gum ). The remainder of
this discussion deals with these cxclu\ively. Representative
molecular struc- turn\ are given in Fig. 47.3. . 47-3
POLYACRYLAMIDE
fT-J-
i I
I 2 I HYDROLYZED
POLYACAYLAMIDE
Fig. 47.3A-Molecular structure for partially hydrolyzed poly-
acrylamide. bility reduction.
Polysaccharides. A second major class of polyniers arc the
polysaccharidcs. which are formed from the polymers-
-
47-4
Fig. 47.4-Polymer solution viscosity vs. shear rate and polymer
concentration.
zation of saccharide molecules (Fig. 47.3). Polysaccha- rides or
biopolymers are formed from a bacterial fermentation process. This
process leaves substantial debris in the polymer product that must
be removed be- fore the polymer is injected. I6 The polymer is also
sus- ceptible to bacterial attack after it has been introduced into
the reservoir. These disadvantages are offset by the in-
sensitivity of polysaccharide properties to brine salinity and
hardness. The origin of this insensitivity is seen in Fig. 47.3,
which shows the polysaccharide molecule to be relatively nonionic
and, therefore, free of the ionic shielding effects of HPAM.
Polysaccharides are more branched than are HPAMs and the
oxygen-ringed car- bon bond does not rotate fully; hence, the
molecule in- creases brine viscosity by snagging and by adding a
more rigid structure to the solution. Polysaccharides do not ex-
hibit permeability reduction, however.
At the present time HPAM is less expensive per unit amount than
polysaccharides; however, when compared on a unit amount of
mobility reduction, particularly at high salinities, the costs are
close enough so that the preferred polymer for a given application
is site-specific. Histori- cally, HPAMs have been used in about 95%
of the report- ed field polymer floods. I7 Both classes of polymers
tend to degrade chemically at elevated temperatures.
Polymer Properties. Because of the complexity of the subject, a
comprehensive treatment of polymer proper- ties is not possible.
However. qualitative trends, a few quantitative relations and
representative data are presented on these properties: viscosity
relations, non-Newtonian effects, retention, permeability
reduction, and chemical, biological, or mechanical degradation.
Viscosity Relations and Non-Newtonian Effects. Fig. 47.4 shows
polymer solution viscosity. p,,, vs. shear rate measured in a
laboratory viscometer at fixed salinity. I8 At low shear rates, p,,
is independent of shear rate. P(~,,=~,,), and the solution is a
Newtonian fluid. At higher i. p,, decreases, approaching a limiting
(p,, =p, ) value not much greater than the water vis- cosity, p,,.,
at some critical high shear rate (not shown
on Fig. 47.4). A fluid whose viscosity decreases with in-
creasing ? is shear thinning. The shear thinning behavior of the
polymer solution is caused by the uncoiling and PETROLEUM
ENGINEERING HANDBOOK
Fig. 47.5-Polymer solution viscosity vs. shear rate at various
brine salinities.
unsnagging of the polymer chains when they are placed in a shear
flow. Below the critical shear rate the curve is reversible.
Fig. 47.5 shows a viscosity/shear-rate plot at fixed poly- mer
concentration with variable NaCl concentration for an AMPS polymer.
I9 Note the profound sensitivity of the viscosity to salinity: as a
rule of thumb the polymer solu- tion viscosity decreases a factor
of 10 for every factor- of-10 increase in NaCl concentration. The
viscosity of HPAM polymers and HPAM derivatives is even more
sensitive to hardness, but viscosities of polysaccharide so-
lutions are relatively insensitive to both.
The behavior in Figs. 47.4 and 47.5 is favorable be- cause for
the bulk of a reservoirs volume, P is usually low (about 1 to 5
seconds -I), making it possible to at- tain a design M with a
minimal amount of polymer. Near the injection wells, however, i can
be quite high, which causes the polymer injectivity to be greater
than that ex- pected on the basis of ~1 The relative magnitude of
this enhanced injectivity ef ect can be estimated once quan- P
titative definitions of shear rate in permeable media and
shear-rate/viscosity relations are given.
The polymer solution viscosity/shear-rate relationship may be
described by a power law model.
pp =K(iy , . . (2)
where K and n are the power-law coefficient and expo- nent,
respectively. Eq. 2 applies only over a limited range of shear
rates: below some low shear rate the polymer solution viscosity is
constant at pp O, and above the crit- ical shear rate the polymer
solution viscosity is also con- stant p,O. The truncated nature of
the power law is awkward in calculations; hence, another useful
relation- ship is the Meter model. 22
PLp O-P,, m P p = P p m+ .
If L c >
a~,,............
p 1%
where 01 is an empirical coefficient and > ,,? is the
shear
rate at which CL,, is the average of pLr, and p,, -. As with all
polymer properties, all empirical parameters are func- tions of
salinity, hardness, and temperature.
-
CHEMICAL FLOODING
When applied to permeable media flow thcsc gcncral trends and
equations continue to apply. cc,, is usually called the r(17/u.~rr
viscosity and the effective shear rate. in,. . is based on
capillary tube concepts.
ic = u ii
, . . . . . . . . . . . . . . . . . (4)
4&G%
where II,,. = superficial flux of polymer-rich (water)
phase. k,,. = permeability to polymer-rich (water) phase. S,,. =
saturation of polymer-rich (water) phase.
fraction, and $ = porosity of medium, fraction.
Polymer Retention. All polymers experience retention on solid
surfaces because of adsorption or mechanical trapping within a
permeable medium. Polymer retention varies with polymer type,
molecular weight, rock com- position, brine salinity and hardness,
flow rate, and tem- perature. Field-measured values of retention
range from 20 to 400 Ibm polymer/acre-ft bulk volume with desir-
able retention level being less than about 50 lbm/acre-ft.
Retention causes the loss of polymer from solution, which can cause
the mobility control effect to be destroyed-an effect that is
particularly pronounced at low polymer con- centrations. Polymer
retention also causes a delay in the rate of polymer
propagation.
Offsetting the delay caused by retention is an accelera- tion of
the polymer solution through the permeable medi- um. which is
caused by inaccessible PV (IPV). The most common explanation for
IPV is that the smaller portions of the pore space will not allow
polymer molecules to enter because of their size. Thus, a portion
of the total pore space is uninvaded or inaccessible to polymer.
and ac- celerated polymer flow results. Large as they arc, how-
ever, most polymer molecules will fit easily into all but the
smallest pore throats. Hence, a second explanation of IPV ih based
on a wall exclusion effect whereby the polymer molecules aggregate
in the center of a narrow channel. The polymer pore fluid layer
near the wall has a lower viscosity than the fluid in the center.
which causes an apparent fluid slip.
IPV depends on polymer molecular weight, media per- meability,
porosity, and pore size distribution, becoming more pronounced as
molecular weight increases and the ratio of permeability to
porosity (characteristic pore size) decreases. IPV can be 30% of
the total pore space or greater,
Permeability Reduction. As mentioned previously HPAM polymers
can cause lowered mobility through a permeability reduction effect.
This phenomenon has been described through three factors. 24 The
resistuncc~,fuctor. FR, is the ratio of the injectivity of a
single-phase poly- mer solution to that of brine flowing under the
same con- ditions: A,,. k,,/tL,,, FR=-= x,, k,,cl,, . 47-5
where k,, = permeability to polymer solution.
h,, = mobility of water-rich phase. and h,, = mobility of
polymer solution.
For constant-flow-rate experiments. FR is the inverse ra- tio of
pressure drops; for constant-pressure-drop experi- ments, FR is the
ratio of flow rates. F, is an indication of the total
mobility-lowering contribution of a polymer. To describe the
permeability reduction effect alone. a prl-- mwhilit~, rcductiorl
,firctor-. F,.k , is defined as
A final definition is the residual resistctnw jirctor. FRr .
which is the mobility of a brine solution before and after polymer
injection.
FR,-=- . _. (7)
where A,,.,, is the mobility of water-rich phase before polymer
injection and A,,,(, is the mobility of water-rich phase after
polymer injection. F,Q is a measure of the permanence of the
permeability reduction effect caused by the polymer solution. It is
the primary measure of the performance of a channel-blocking
application of poly- mer solutions. For many cases, F,k and FR,-
are nearly equal; however. FR is usually much larger than F,.,: be-
cause the former contains the viscosity-enhancing effect as well as
the permeability-reduction effect.
The most common measure of permeability reduction is F,.k. Frk
is sensitive to polymer type, molecular weight. degree of
hydrolysis, shear rate, and permeable media pore structure.
Polymers that have undergone even a small amount of mechanical
degradation seem to lose most of their permeability reduction
effect. For this reason. qualitative tests based on screen factor
devices are com- mon to estimate polymer quality. F,.k has been
correlat- ed to polymer adsorption and rheological properties by
Hirasaki and Pope.
Chemical and Biological Degradation. The average polymer
molecular weight can be decreased, to the detri- ment of the
overall process, by chemical. biological, or mechanical
degradation. Chemical degradation can be minimized by restricting
polymer usage to low- temperature applications, and by adding
oxygen scavengers (e.g., sodium sulfate or sodium sulfite) to the
polymer solution. Biological degradation can be elim- nated by
adding oxygen scavengers and biocides (e.g.. formaldehyde or
isopropyl alcohol). In fact, nearly all ap- plications contain some
of these chemicals. usually in very small quantities (see Ref. 16
or 26).
Mechanical Degradation. Mechanicaldegradation, on the other
hand, is potentially present under all applica- tions. Mechanical
degradation occurs when polymer so- lutions are exposed to
high-velocity flows. These can be present in surface equipment
(valves. orifices, pumps. or
tubing), downhole conditions (perforations or screens). or in
the sandface itself. Perforated completions. partic- ularly, are a
cause for concern. since large quantities of
-
47-6 HANDBOOK
PLAC
ON P
oil rate be- . Polymer
arrested a The oil rate substantial- l oil recov- e between hich
would
od. Thus. important to an accurate
Burbank is
n more than nsive sur-
vcy by Manning rt rrl. I7 The table emphasizes oil recov-
Fig. 47.6-Tertiary polymer flood response from North Burbank
Unit, Osage County, OK.
polymer solution arc being forced through a number of small
holes. For this reason. most polymer in,jections are done through
openhole or gravel-pack completions. Be- cause flow velocity falls
off quickly with distance from an injector. little mechanical
degradation occurs within the reservoir itself.
All polymers mechanically degrade under high enough flow rates.
However. HPAMs are most susceptible un- der normal operating
conditions. particularly if the sa- linity or hardness ofthe brine
is large. Evidently, the ionic coupling of these anionic molecules
is relatively fragile compared with the polysaccharide chains.
Also. clonga- tional stress is ar destructive to polymer solutions
as is shear stress, although the two generally accompany each
other. Maerker and others have correlated permanent vis- cosity of
a polymer solution loss to an elongational stretch rate times
length product. 273 On a viscosity/shear-rate plot (purely shear
flow), mechanical degradation usually begins at shear rates greater
than the minimum-viscosity critical shear rate.
Field Results. Fig. 47.6 shows a tertiary polymer tlood- ing
field response from the North Burbank unit, Osage
TABLE 47.1-POLYMER
Oil recovery, O/o remaining OIP Oil recovery, STB/lbm Oil
recovery, STBlacre-ft Permeability variation, fraction Mobile oil
saturation, fraction Oil viscosity, cp Resident brine salinity,
g TDSlL Water-to-oil mobility ratio, dimensionless Average
polymer concentration, ppm Temperature, OF
Number of Projects*
50 80 88
118 62
153
10
87
93 172 Average permeability, md Average porosity, fraction
187 193 cry data and screening parameters used for polymer
flooding. Approximately one-third of the statistics arc from
commercial or field-scale floods. The oil recovery statistics in
Table 47. I show average polymer tlood recov- cries of 3.56%
remaining (after watertlood) oil in place. and 2.69 STB of IOR for
each pound of polymer itljcct- cd. with wide variations in both
numbers. The large varia- bility reflects the emerging nature of
polymer tlooding in the previous decades. Considering the STB IOR
per pound of polymer average and the average costs of crude and
polymer, it appears that polymer flooding could be a highly
attractive EOR process. However. such costs al- ways should be
compared on a discounted basis rctlcct- ing the time value of
money. which will decrease the apparent attractlvencss of polymer
flooding because of the decreased injectivity of the polymer
solutions.
Foam Flooding
Gas/liquid foams offer an alternative to polymers for providing
mobility control in chemical floods, and have been both proposed
and field tested as mobility control agents in steamfloods.
Foams are dispersions of gas bubbles in liquids. Gas/liq- uid
dispersions are normally unstable and usually will break in less
than I second. If surfactants are added to the liquid, however,
stability is improved greatly so that some foams can persist
indefinitely. To understand foam
FLOOD STATISTICS
5.86 0.1 51.8 11.05
339 51 3,700 343 115 46 234 85 349 1.5 7,400 720
Mean Minimum Maximum
3.56 0 25.3 2.69 0 36.5
24.0 0 188.7 0.70 0.06 0.96 0.27 0.03 0.51
36 0.072 1,494
40.4 5.0 133.0
Standard Deviation
5.63 4.86
36.65 0.19 0.12
110.2
33.4 PETROLEUM ENGINEERING
ED 40. I 53000 BARRELS
RODUCTION -
EARS
County, OK. The figure shows WOR andfore and immediately after
polymer injectioninjection began in late 1970, which quickly
declining oil rate and an increasing WOR. then resumed its
prepolymer decline but at aly higher level. The polymer oil or
incrementaery (IOR) from a polymer flood is the differencthe
cumulative oil actually produced and that whave been produced by a
continuing waterflofor a technical analysis of the project it is
establish a polymer flood oil rate decline andwaterflood decline
rate. The IOR for North the shaded area in Fig. 47.6.
Table 47. I summarizes other field results o250 polymer tloods
on the basis of a comprehe0.20 0.07 0.38 0.20
-
CHEMICAL FLO
propertie rcqutheir classificatiosurfact~nts 3s
Surfxtant Chcomposed of astrong affinit~~~i.e.. strong
attmamphtphile (has affinity for both oil and vvater) because
47-7
_ g _ o- No+
I 0
A
iacheofthis dual nature. Fig. 47.7 shows the molecular struc-
ture of two common surfactanta (top two panels) and il- lustrates
(lowest panel) a shorthand notation for surfactant monomers: the
monomer is represented by a tadpole sy~dx~l with the nonpolar
portion being the tail and the polar the head. Surfactants are
classified into four groups that depend on their polar portions
(see Table 47.1)3.
Anionic.s. As required by electroneutrality. the an ionic
aurfactant molecule is uncharged with an inorganic metal cation
(usually sodium) associated with the monomer. In an aqueous
solution the ~~dx~~le ioniLcs to free cations and the anionic
monomer. Anionic sur- factants are the most common in EOR because
they are good surfactants. rclativelv resistant to retention.
stable. and relatively inexpensive.
Cationics. In this case the surfactant ~~dxulc con- tains a
cationic hydrophilc and an inorganic anion to bal- ancc the charpc.
Cationic surfactants have little use in EOR bccausc they are
adsorbed highly by the anionic sur- faces of interstitial
clays.
Nonionics. These surfactants do not form ionic bonds, but when
dissolved in aqueous solutions. they cx- hibit surfactant
properties simply by electronegativity con- trasts betvveen their
constituents. Nonionics arc much less sensittve to high salinities
than anionics or cationics.
Anzphoterics. A final class of surfactants are those that
contain aspects of two or more of the previous class- es. For
example. an amphoteric may contain both an an- ionic group and a
nonpolar group. These surfactants have not been used in EOR.
Within any one class there is a huge variety of possible
surfactants. Fig. 47.7 illustrates differences in Iipophile
molecular weight IC t2 for the sodium dodecyl sulfate (SDS) vs. C
th for Texas No. I]. hydrophilc identity (sul- fate vs. sulfonate),
and tail branching (straight chain for SDS vs. two tails for Texas
No. 1) all within the same class of anionic surfactants. In
addition to these. there are variations in the position ofthe
hydrophile attachment and the number of hydrophiles (monosulfonates
vs. disul- fonatcs. for example). Even small variations can change
surfactant properties drastically (e.g.. sulfates tend to be less
thermally stable than sulfonatcs). For details on the
Table 47.2-CLASSIFICATION OF
4-M -eM Anionics Cationics
Sulfonaies Quaternary ammonium AmSulfates Pyridinum Carboxylates
lmidazolrnrum OtPhosphates Piperidinium Others Sulfononium
Compounds Others COMMERCIAL PETROLEUM SULFONATES
0 I
R-S -O-Nat-
A
R - HYDROCARBON GROUP (non - polar)
Fig. 47.7-Typical surfactant molecular structures
effect of structure on surfactant propertics see Refs. 31 and
32.
Most commercial surfactants contain distributions of surfactants
and surfactant types that further add to their complexity. In the
following. distinctions between sur- factant types are ignored by
simply treating the surfac- tant as the tadpole structure of Fig.
47.7.
Foam Stability. The stability of a foam may be under- stood by
viewing the liquid film separating two gas hub- bles in cross
section as in the lower panel of Fig. 47.8. ? The hydrophiles of
the surfactant are oriented into the in- terior of the film and the
lipophiles toward the bulk gas phase. Suppose that some external
force causes the film to thin as in the lower panel. Since
capillary pressure is inversely proportional to interfacial
curvature. the pres- sure in the thinned portion of the film is
lower than in the ad,iacent flat portion. This causes a pressure
differ- cnce within the film. liquid flow. and healing. The pres-
sure in the gas phase is assumed constant because of its relatively
low density if the foam is static. or low vis- cosity if in
motion.
SURFACTANTS AND EXAMPLES
-e+ d- mphoterics Nonionics
nocarboxyltc Alkyl- ids Alkyl-aryl rs Acyl
Acylamindo- Acylaminepolyglycol ethers ODING
lrcc wnc ili5cusvon of surtactunts and ns. Moat of the
discussion applies to MP well.
emistq. A typical xurfactant monomet- is nonpolar portion
(lipophi1c~i.c.. having for oil) and a polar portion (hydrophile-
ity for water): the entire monomer is an
SODIUM DODECYL SULFATE
c: c
/ \,/ \cA,i\p\cAo
TEXAS NO.1 SULFONATE Polyol ethers Alkanolamides Others
-
three permeabilities in Fig. 47.9, but the effect of foam
47-a
SURFACE TENSION
/I----- _ ADSORPTION
/ / /
/ /
/ / I
/ J I
C.Y.C. CONC.
3 Hii
FlllIl
Film 0
Thlnner, Lou adrorptlon, Larger aurfece tencllon,
Contractlon
Fig. 47.8-Upper panel shows surface tension and adsorption of a
surfactant vs. concentration. Lower panel is the Gibbs-Marangoni
effect.
0 km, = 4.41 OMCY
* k,,, * 0.42 DARCY
0 km 0.22 DARCY
\ 0
0.10 !- * A
0 0.01 60
QUA&Y OF F&i, PLRCESOf
Fig. 47.9-Effective permeability-viscosity ratio vs foam quality
for consolidated oorous media.
The upper panel in Fig. 47.8 shows that the gas/liquid surface
tension is a decreasing function of surface adsorp- tion as
required by the Gibbs theory. According to this view the thinned
portion of the film will have less specif- ic adsorption (since the
surface area is locally greater) and greater surface tension. This
locally high surface tension also causes healing.
Clearly. the surface tension at the gas/liquid interfaces play
an important role in film stability. Very low surface tension would
not be favorable; fortunately gas/liquid sur- face tensions are
rarely lower than 20 dynes/cm even with
0 the best foaming agents. In the absence of external forces.
the film is at an equilibrium thickness caused by a bal- ance
between the repulsion forces of the electrical dou- PETROLEUM
ENGINEERING HANDBOOK
ble layer on the interior of the film boundary and the
attractive Van der Waal forces bctwccn the molecules in the film.
If the film becomes substantially smaller than the equilibrium
thickness. the fret energy barrier between the repulsive and
attractive contributions is breached and the film will
collapse.
Such thinning can be caused spontaneously by diffu- sion of the
gas from small to large bubbles and by gravi- ty drainage. Patton
et al. reported on the rate of spontaneous collapse of a large
number of foams as a func- tion of surfactant type. temperature,
and pH. tl The half- life of the foam heights reported in their
static tests ranges from 1 to about 45 minutes. They report that
anionic sur- factants have greater stability than nonionics. and
that the stability of sulfonate foams is affected greatly by water
hardness. Foams were generally more unstable at high temperatures,
and many could be stabilized by adding a second surfactant.
External effects that may cause the foam to collapse are the
presence of a foam breaker (oil or a high clcctrolyte concentration
could do this). local heating. or contact with a hydrophobic
surface.
Foam Physical Properties. Physically, foams arc char- acterized
by three measures.
Qua&. Foam quality, r, is the percentage of the total (bulk)
foam volume that is gas volume. The quality can increase with
increasing temperature and decreasing pres- sure both because the
gas volume can increase. and also because gas dissolved in the bulk
liquid phase can evolve from solution. Foam qualities can bc quite
high, approach- ing 97% in many cases. A foam with quality greater
than 90% is a dry foam.
Texture. This measure is the average bubble size. The texture
determines how the foam will flow through a permeable medium. If
the average bubble size is larger than the average pore diameter.
the foam flow5 as a progression of films separating individual gas
bubbles. Given typical foam textures and pore sizes, this condi-
tion is most nearly realized in permeable flow. particu- larly for
high foam qualities.
Bubble Size Range. Foams with a large distribution range are
more likely to be unstable because of the gas diffusion from large
to small bubbles.
Mobility Lowering. Foams can reduce the mobility of a gas phase
drastically. Fig. 47.9 shows the steady-state mobility of foams of
differing quality in Berea cores at three different permeabilities
as a function of quality. On the extreme right of this figure (r-
100%). the mo- bility should approach the respective air
permeability divided by the air viscosity: this mobility is two to
three factors of 10 greater than any of the experimental points on
the figure. When r-0 the mobility should approach the water
permeability divided by CL,, Thus. the mobili- ty of the foam is
lower than that of either of its consti- tuents alone. The mobility
of the foam decreases with increasing quality until the films
between the gas bubbles begin to break and the foam collapses (not
shown on Fig. 47.9). Foams are effective in reducing the mobility
at all quality is more pronounced at the highest permeability. This
is a consequence of the contrast between the tcxturc of the foam
and the mean pore size of the mcdium.36
-
CHEMICAL FLOODING
The mobility reduction caused by the foam can be viewed as an
increase in a single-phase viscosity or as a decrease in the
gas-phase permeability. Representative data of the second type are
in Fig. 47.10, which shows the gas-phase permeability, both with
and without foam, and
f as saturation plotted against the liquid injection
rate. - Note that the foam causes a great decrease in gas
permeability at the same rate and even at the same gas saturation
compared to the nonfoaming displacement. The analogous analysis
performed on the aqueous-phase rela- tive permeability shows that
neither the gas saturation nor the presence of the foaming agent
affects the aqueous- phase relative permeability. j8
The low foam mobilities in permeable media flow are postulated
to be caused by at least two different mecha- nisms: (I) the
formation of or the increase in a trapped residual gas phase
saturation and (2) a blocking of pore throats caused by the gas
films. From Fig. 47.10 the ef- fect of a trapped gas saturation,
which would lower the gas mobility through a relative-permeability
lowering, is much smaller than the pore-throat-blocking effect. The
trapped gas-phase saturation effect may become impor- tant.
however, during the later stages of a displacement where the lower
pressures could cause more gas to come out of solution.
The mobility reduction of foams, viewed as a viscosit
enhancement. has been studied in capillary tubes. 38
General observations on these data are that foams are generally
shear-thinning fluids whose power law coeffi- cient increases with
the capillary tube radius. Using the- oretical arguments based on a
Newtonian fluid and an inviscid gas, Hirasaki and Lawson showed
that the film thickness of a single moving bubble increases as the
bub- ble velocity to the two-thirds power. 4o Since shear stress in
a capillary is inversely proportional to film thickness. the
apparent viscosity of a foam in a capillary tube decreases with
increasing velocity. Thus, the shear- thinning effect observed in
capillary tubes is actually a consequence of the film thickening as
velocity increases. A second implication in the Hirasaki-Lawson
theory is that foam texture occupies as great an importance in de-
termining rheological behavior as does foam quality.
Field Results. Field tests of foam injection alone have been
scarce. Holm reports on the injection of an air/brine foam into a
single well in the Siggins field. Though there was no measureable
oil response, the mobility to both air and brine were reduced
significantly, and the in- jection profile into the central well
became more uniform.
Low-IFT Processes In addition to stabilizing gas dispersions,
surfactants used in MP flooding or generated in situ can recover
oil by lowering oil/water IFT.
Lowering ROS
The basic tool for illustrating how lowering IFT reduces ROS is
the capillary desaturation curve (CDC) shown in Fig. 47.1 I. The
CDC is a plot of nonwetting- or wetting- phase residual saturation
on the y axis vs. a dimcnsion-
less ratio of viscous to local capillary forces on a logarith-
mic .Y axis. On Fig. 47.1 I, S,,,.,, is the residual oil (assumed
nonwetting), and S,,,. is the irreducible water- (wetting)-phase
saturation. The CDC has been calculat- 47-9
l.T Ii iiiiiii iiiiil II:
-?-d---GAS PERYEABILITY k--t+sRI 3
3
0.1 0 I 2 3 4 5 6 7 8 9 IO II 12 13 14
LIQUID INJECTION RATE, BARRELS PER. DAY/ SO. FT.
Fig. 47.10-Effect of liquid flow rate and gas saturation on gas
permeability with and without foam
Fig. 47.11--Schematic capillary desaturatjon curves
ed theoretically, 4-w but the most common source of this curve
is experimental measurement. . The ratio of vis- cous to local
capillary forces is the capillary number, N,. 1 one form of which
is
N,. =N ,, p ,,,/u,,.
-
flogarithmic .Y axis. a decrease by several factors of IO in N,
is necessary IO significantly change either residual phase IFT
ccausindyne/cmaurfact
An idealized version of a MP flooding sequence is in ry im-
ff).
nge ent the
size ses sur- . in
ent. I al- are
f a MP ch
gn-
in- of ck. saturation. Ofthc three quantities in Eq. 8 only the
an be changed this drastically; a typical value for g good ROS
reduction is in the range of IO
. a value that can be obtained only with a good ant.
Fig. 47.12. The process is applied invariably to tertiafloods
(those producing at high WORs) and is always plemented in the drive
mode (not cyclic or huff n puThe complete process consists of the
following.
Prejlush. A volume of brine whose purpose is to cha(usually
lower) the salinity and hardness of the residbrine so that mixing
with the surfactant will not cause loss of interfacial activity.
Pretlushcs have ranged in from 0 to 100% of the reservoir PV. In
some procesa sacrificial agent is added to lessen the subsequent
factant retention and also to precipitate divalent cation
MP Slug. This volume. ranging from 5 to 40% PVfield
applications. contains the main oil-recovering agthe prituut:\~
surfactant, in concentrations ranging fromto 20 ~0170. Several
other chemicals (cosurfactants. cohols. oil. polymer. biocide, and
oxygen scavcnger) usually necessary for good oil recovery.
LJ40~i~i~ Buffeer. This fluid is a dilute solution
owater-soluble polymer whose purpose is to drive the \lug and
banked-up fluids to the production wells. Muof polymer flooding
technology carries over to dcsiing and implcmcnting the mobility
buffer.-18
Freshwater Buffer. This i\ a volume of brine containg a
concentration of polymer grading between thatthe mobility buffer at
the front end and zero at the ba47-10
Fig. 47.12-Schematic o
originally trapped phase occurs. The shape of the CDC is
determined by the pore geometry of the medium and the wetting
behavior of the two phases. The wetting phase rcquircs a larger N,.
for complete recovery. To a lesser extent. the CDC shape is
affected by the mean pore size and pore size distribution of the
medium. and the initial saturations. Typical N, s for watertlooding
are quite small. which indicates that ROS and S,,, may be assumed
constant for this purpose (XC Fig. 47. I I). Because of the Fig.
47.13-Schematic of the CMC PETROLEUM ENGINEERING HANDBI 30K
an MP flooding process.
IMP Flooding
MP flooding has appeared in the technical literature un- der
many names: detergent. surfactant, low-tension. solu- ble oil.
microemulsion. and chemical flooding. There are also several
company names that imply a specific sequence and type of injected
fluids as well as the specific nature of the oil-recovering MP slug
itself. Though there are differences among company processes. the
common aspects are more numerous and important. The gradual
concentration decrease mitigates the effect of the unfavorable
mobility ratio between the chase water and mobility buffer.
-
CHEMICAL FLOODING
Chase Water. The purpose of the chase water is sirn- ply to
reduce the expense of continually injecting poly- mer. If the
mobility and freshwater buffers have been designed properly. the MP
slug will be dcplcted before it is penetrated,by the chase
water.
Surfactant Solutions. If an anionic surfactant is dissolved in
brine, the surfactant disassociates into a cation and a monomer. If
the surfactant concentration then is increased, the lipophilic
portions of the surfactant begin to associate among themselves to
form aggregates or micelIes contain- ing several monomers each. A
plot of surfactant monomer concentration vs. total surfactant
concentration (Fig. 47.13) is a curve that begins at the origin,
increases with unit slope, then levels off at the critical micelle
concen- tration (CMC). Above the CMC all further increases in
surfactant concentration cause increases only in the micelle
concentration. CMCs are typically quite small (about 10 - to IO p4
mol/L). At nearly all practical concentra- tions for MP flooding
the surfactant is predominantly in the micellc form; hence. the
name micellar/polymer flood- ing. The representations of the
micelles in Fig. 47.13 and elsewhere are schematic. The actual
micelle structures can take on various forms, which can fluctuate
with time.
When this solution contacts an oleic phase (the term olcic
indicates that this phase can contain more than oil) the surfactant
tends to accumulate at the intervening interface. The lipophilic
tail dissolves in the oleic phase, and the hydrophilic in the
aqueous phase. The accumula- tion at the intcrfacc causes the IFT
between the two phases to lower. The extent of the IFT lowering is
proportional to the cxccss surface concentration of the
surf;lctant-the difference between the surface and bulk surfactant
concentration-from Gibbs theory. as was the case in Fig. 47.X. The
surfactant itself and the attending conditions should bc adjusted
to maximize the excess surface con- cent]-ation; however, in doing
so the solubility of the sur- tactant in the bulk oleic and aqueous
phases also is affected. Since this solubility also impinges on the
mutu- al volubility of brine and oil, which also affects IFTs, this
discussion leads naturally to the topic of surfactanti brine/oil
phase behavior. Curiously, and this is true of many micellar
properties. the surfactant concentration it- self plays a rather
minor role in what follows, compared with the temperature. brine
salinity, and hardness.
SurfactantlBrineiOil Phase Behavior. Surfactanti hrinc/oil phase
behavior is illustrated conventionally on a ternary diagram. A
ternary diagram is an equilateral triangle whose apexes represent
pure components. bound- arlcs represent two-component mixtures, and
interior rep- resents three-component mixtures. For complicated
mixtures the three apexes must represent pseudocom- poncnts whose
composition remains constant through- out the diagram. The pressure
and temperature are also fixed. The diagram can represent both the
overall com- position of a surfactantibrineioil mixture. and the
equi- librium composition of each phase if the mixture forms more
than one phase. Tcrnarica and their accompanying definitions are
discussed in Chap. 23.
MP phase behavior is affected strongly by the salinity
of the brine pseudocomponent. Consider the sequence of phase
diagrams (Figs. 47.14 through 47.16) as the brine salinity is
incrcaxcd. The phase behavior about to bc dc- 1 _(
I S WA TEf? EXTERNAL
A
EXCESS hffCROE~LS/ON 011
I
47-11
OVERALL 0 COMPOSITtW
Fig. 47.14-Schematic of low-salinity surfactantlbrineioil phase
behavior.
scribed was presented originally by Winsor and adapted to MP
flooding later.50.5
At low brine salinity, a typical MP surfactant will ex- hibit
good aqueous (water-rich) phase solubility and poor oleic
(oil-rich) phase solubility. Thus, an overall compo- sition near
the brine/oil boundary of the ternary will split into two phases:
an e,xcess oil phase that is essentially pure oil and a
(water-external) rnicroemulsio~~ phase that con- tains brine
surfactant, and some solubilized oil. The solu- bilized oil occurs
when globules of oil occupy the central core of the swollen
micelles. The tie lines within the two- phase region have a
negative slope. This type of phase environment is called variously
a Winsor Type I system. a lower-phase microemulsion (because it is
more dense than the excess oil phase), or a Type II system. The
lat- ter terminology is adopted here-11 means that no more than two
phases can (not necessarily will) form and (-) means that the tie
lines have negative slope (Fig. 47.14). The right plait point in
such a system. PK. usually is lo- cated quite close to the oil
apex. Any overall composi- tion above the binodal curve is single
phase.
For high brine salinities (Fig. 47. IS) electrostatic forces
drastically decrease the surfactants solubility in the aque- ous
phase. An overall composition within the two-phase Iregion now will
split into an UUJ,S.\ hr-i/~c phase and an
-
47-12
SWOLLEN MICELLE
-
Na+ Na+
NO+
COMPOSITION
Fig. 47.15-Schematic of surfactant/brine/oil high-salinity phase
behavior.
(oil-external) microemulsion phase, which contains most of the
surfactant and some solubilized brine. The brine is solubilized
through the formation of inverted swollen micelles (Fig. 47.15)
with brine globules at their cores. The phase environment is a
Winsor Type II system, an upper-phase microemulsion, or a Type II(
+) system. The plait point, PI,. is now close to the brine
apex.
The two extremes presented thus far are roughly mir- ror images;
note that the microemulsion phase is water- continuous in the Type
II( -) systems and oil-continuous in Type II(+) systems. The
induced solubility of oil in a brine-rich phase, Type II( -).
suggests an extraction mechanism in oil recovery. Though extraction
does play some role. it is dwarfed by the IFT effect discussed
later. particularly when phase behavior at intermediate salini-
tics is considered.
At salinities between those of Figs. 47. I4 and 47.15. there is
a continuous change between Type II( -) and II( +) systems and a
third surfactant-rich phase is formed. as shown in Fig. 47.16. An
overall composition within the three-phase region separates into
excess oil and brine
phases as in the Type II(-) and II( +) environments. and into a
microemulsion phase whose composition is repre- sented by an
;U~U~;LUU point. This environment is called PETROLEUM ENGINEERING
HANDBOOK
COMPOSITION
Fig. 47.16-Schematic of surfactant/brine/od phase behavior at
optimal salinity.
a Windsor Type III, a middle-phase microemulsion. or a Type III
system. Above and to the right and left of the three-phase region
are Type II( -) and II( +) lobes wherein two phases will form as
before. Below the three-phase region there is a third two-phase
region (as required by thermodynamics) whose extent is usually so
small that it is neglected. In the three-phase region there are now
two IFTs between the microemulsion and oil, a,,,,, , and the
microemulsion and water, u,~~,,..
Fig. 47.17, a prism diagram, shows the entire progres- sion of
phase environments from Type II( -) to II( +). The Type III region
forms through the splitting of a criti- cal tie line that lies
close to the brine/oil boundary as the salinity increases to C,,
(low effective salinity limit for Type III phase environment). 53 A
second critical tie line also splits at C,, (high effective
salinity limit for Type III phase environment) as salinity is
decreased from a Type II( +) environment. Over the Type III
salinity range the invariant point, M, migrates from near the oil
apex to near the brine apex before disappearing at the appropriate
crit- ical tie lines. The migration of the invariant point
implies
essentially unlimited solubility of brine and oil in a sin- gle
phase, which has generated an Intense research in- terest into the
nature of the Type III microemulsion.
-
I. At high surfactant concentrations and/or at low tempetants L
3observcrystaare deviscoulow- are adble videnseconcethe
melimina
2. approtant anot pment surfacEffortscosurftation
3.
lent ratios also will cause electrolyte interactions with clay
nate for by
ivalcnt in Fig.
rature es of
low. 6x entra- poly- surely
tech- with
posed antiat- sever-
on is can be asure- raturess.59 or even in the presence of pure
surfac- 6. phases other than those on Fig. 47.17 have been ed.
These phases tend to be high-viscosity liquid
ls or other condensed phases. The large viscosities trimental to
oil recovery since they can cause local s instabilities during a
displacement. Frequently, to medium-molecular-weight alcohols
(cosolvents) ded to MP formulations to melt these undesira-
scosities. When the brine contains polymer. a con- d phase can
be observed at low surfactant ntration because of exclusion of the
polymer from icroemulsion phase. Cosurfactants can be used to te
this polymerisurfactant incompatibility. When cosurfactants are
present it is frequently in- priate to lump all of the chemicals
into the surfac- pex of the Fig. 47.17 prism. If the cosurfactants
do
artition with the primary surfactant during a displace- the
benefit of adding the chemical is lost; hence.
tanticosurfactant separation effects are important. to account
for the preferential partitioning of the
minerals through cation exchange. The disproportioeffects of the
salinity and hardness are accounted defining a weighted sum of the
monovalent and dconcentrations as an effective salinity. The C,.s
47. I7 imply effective salinities.
Phase Behavior and IFT. Early MP flooding litecontains
considerable information about the techniqumeasuring IFTs and what
causes them to be IFTs were found to vary with the types and
conction of surfactant, cosurfactant, electrolyte, oil, andmer and
with temperature. However, in what was one of the most significant
advances in all of MPnology, all IFTs were shown to correlate
directly the MP phase behavior. The correlation was prooriginally
by Healy and Reed, theoretically substed by Huh, 6y and since
verified experimentally by al others. I. The practical benefit of
this correlatithat relatively difficult measurements of IFTs
largely supplanted by simple phase behavior meCHEMICAL FLOODING
Fig. 47.17-Prism diagram showing se
Several variables other than salinity can bring about the Fig.
47.17 phase environment shifts. In general, chang- ing any
condition that enhances the surfactants oil solu- bility will cause
the shift from Type Il( -) to II( +). Some of the more important
are: (1) decreasing temperature, 5 (2) increasing surfactant
molecular weight, (3) decreas- ing tail branching, (4) decreasing
oil specific gravity55-57 and (5) increasing concentration of high-
molecular-weight alcohols. 58 Decreasing the surfactants oil
solubility will cause the reverse change. Thus, Fig. 47.17 could be
redrawn with any of the above variables (and several others) on the
base of the prism with the vari- able C,, (effective salinity)
increasing in the direction of increased oil solubility.
Nonideal Effects. In much the same manner as the ideal gas law
approximates the behavior of real gases, Fig. 47. I7 is an
approximation to actual MP phase behavior. Some of the more
important nonidealities are as follows. actant include a quaternary
phase behavior represen- and a pseudophase theory. 62.63
The Type III salinity limits (C,, and C,,) are func- quence of
phase environment transition.
tions of surfactant concentration. This dependency may be
visualized by tilting the vertical triangular planes in Fig. 47.17
about their bases. This dilution effecth.h5 forms the basis for the
salinity requirement diagram de- sign procedure. UJ The dilution
effect is particularly pronounced when the brine contains
significant quanti- ties of divalent ions.
4. The Fig. 47.17 phase-behavior shifts are specific to the
exact ionic composition of the brine. not simply to the total
salinity. For anionic surfactants. other anions in solution have
little effect on the MP phase behavior: how- ever. cations readily
cause phase-environment changes. Divalent cations (calcium and
magnesium are the most
common) are usually 5 to 20 times as potent as monova- lent
cations (usually sodium). Divalents arc usually pres- ent in
oilfield brines in smaller quantities than monovalents as shown in
Fig. 47.2, but their effect is so pronounced that it is necessary,
as a minimum. to separately account for salinity and hardness.
Nonconstant monovalentidiva- ments. Indeed, in the recent
literature, the behavior of IFTs has been inferred by a narrower
subset of phase- behavior studies based on the solubilization
parameter. hs
-
Fig. 47.18--Correlation of solubilization ratios with IFT.
1 4 1 1 I 1 1 I I 1 I L I I 1 I 0 0.4 0.6 1.2 I.6 2.0 2.4 2 .a
3.2
SALINITY, /o NoCI
Fig. 47.19-IFT and solubilization ratios PETROLEUM ENGINEERING
HANDBOOK
To investigate further the relation between IFTs and phase
behavior, let V,,, V,, , and V, be the volume frac- tions of oil,
brine. and surfactant in the microemulsion phase, respectively.
According to Figs. 47. I4 through 47.16, the microemulsion phase is
present at all salini- ties: hence. all three quantities are
well-defined and con- tinuous. Considering the Type II( -) behavior
of Fig. 47.14, for example, V,,, V,,., and V,, are the coordinates
of the microemulsion phase composition on the binodal curve.
Solubilization parameters between the microemulsion- oleic
phases, F,,,, , for Type II( -) and III phase behavior. and between
the microemulsion-aqueous phases. F,,,,,.. for Type II( +) and III
are defined as
F,,,,,=V,,iV, . . . . . . . . . . . . . . . . . . . . . . . . .
. . ..(9a)
and
F,,,,,.=l,,./l,. . . . . (9b)
The IFTs between the corresponding phases, u,,,(, and u ,?lMr
are functions only of F,,, and F ,,,, ~. Fig. 47.18 shows a typical
correlation.
The corresponding behavior of the solubilization pa- rameters
and IFTs are shown in Fig. 47.19 in a differ- ent manner. Consider
a locus at constant oil, brine, and surfactant overall
concentrations in Fig. 47. I7 but with a variable salinity. If the
nonideal effects are unimpor- tant and the locus is at low
surfactant concentration and intermediate brine/oil ratios, (T,~(,
will be defined from low salinity up to C,,, and uInM. from C,,, to
high salini- ties. Both IFTs are the lowest in the three-phase Type
III region, between C,, and C,,, where both solubiliza- tion
parameters are also large. There is, further, a pre- cise salinity
where both IFTs are equal at values low enough ( - 10 -3 dyne/cm)
for good oil recovery. This salinity is the optimal saliniv for
this particular surfac- tant/brine/oil combination and the common
IFT is the op- timal IFT. Optimal salinities have been defined on
the basis of equal IFTs, as in Fig. 47.19, equal solubiliza- tion
parameters, maximum oil recovery in corefloods, and equal contact
angles. 50*71~72 All definitions of optima1 sa- linity give roughly
the same value; hence, since optimal phase behavior salinity is the
same as maximum oil recov- ery salinity, generating an
interfacially active MP slug translates into generating this
optimal salinity in situ in the presence of the surfactant
material.
Generating Optimal Conditions. Historically there have been
three techniques for generating optimal conditions in an MP
displacement.
I. The MP systems optimal salinity can be raised to that of the
resident brine salinity in the candidate reser- voir. This
procedure philosophically is the most satisfy- ing of the three
design procedures given here and usually the most difficult. Though
it has been the sub.ject of in- tensive research, surfactants that
have high optimal sa- linities that are not (at the same time)
thermally unstable at reservoir conditions, excessively retained by
the solid
surfaces, or expensive have not yet been discovered. Field
successes with synthetic surfactants have demonstrated the
technical feasibility of this approach. however. 73 A scc-
-
npretlush could be accomplished during the watertlood prece
3. ly lodisplaoverobuffeon itbe ingradimot crv vanraunccthe pand
ctfcc
Surfone
I, monoinp aand higheclt~dc
with abov
reservoir clays just as inorganic cations do.
- )
s t -
- ding the MP flood. The salinity gradient design attempts to
dynamical-
wer the resident salinity to an optimum during the cement by
sandvviching the MP slug between the ptimal rcsidcnt brine and an
underoptimal mobility- r salinity. hh.67 The \ucccss of this
procedure relies being necessary that only a portion of the MP slug
the active region for good oil rccovcry. For salinity ent tlonds
the salinity of the mobility buffer is the
signiticant factor in bringing about good oil rccov x The
salinity gradient design has several other ad- zes in being
rcstlient to design and process
rtaintics. in providing a favorable cnv~ironmcnt for olymer in
the mobility buffer. minimizing retention. being relatively
indifferent to the surfactant dilution t.
actant Retention. Surfactants are retained through of at least
four mechanisms. On metal oxide surfaces (Fig. 47.20) the
surfactant mer will adsorb physically through hydrogen hond- nd
micelle-like associations with the monomer tails. ionically bond
vvith cationic surface sites (I). At
4. In the presence of oil in a II( +) phase environment the
aurfactant will reside in the oil-external microemul- sion phase.
Because this region is above the optimal salinity, the IFT is
relatively large (Fig. 47.18) and thisphase and its dissolved
surfactant can be trapped. Asimilar phase trapping effect does not
occur in the II( - environment because the aqueous mobility buffer
misci-bly displaces the trapped aqueous-external microemulsion
phase without permanent retention.
Most studies of surfactant retention have not made thepreviously
mentioned mechanistic distinctions; thercforc, which mechanism
predominates in a given application inot obvious. All mechanisms
retain more surfactant ahigh salinity and hardness, which, in turn.
can be attenu-ated by adding cosurfactants. Precipitation and phase
trappin,g,can be eliminated by lowering the mobility buffersalimty,
at which conditions the chemical adsorption mechanism on the
reservoir clays is predominant. There-fore. there should be some
correlation of surfactant rctcn-tion with reservoir clay content.
Fig. 47.2 I is an attemptto make this correlation by plotting
laborator
2 and field
surfactant retention data against clay fraction. ) The
correlation is by no means perfect since it ignores variations
CHEMICAL FLOODING
Fig. 47.20-Schematic of surfacta
ond way to make the optimal salinity of the MP formula- tion
equal to the resident brine salinity is to add cosurfactant.
2. Resident salinity of a candidate reservoir can be lo- wered
to match the MP slugs optimal salinity. This is the main purpose of
the pretlush step illustrated in Fig. 37.12. A successful pretlush
is appealing because. with the resident salmtty lowered. the MP
slug would displace oil wherever it goes in the reservoir.
Preflushes general- ly require quite large volumes to lower the
resident SB- linity significantly owing to mixing effects and
cation exchange. 7.75 With some planning, the function of the r
surtactant concentrations. C,. this association in- ?~ tail-to-tail
interactions with the solution monomers
proportionally greater adsorption (II and Ill). At and e the CMC
(IV), the supply of monomers bccomcs 47-15
t adsorption on metal oxide surface
constant as does the retention (c, is the adsorbed surfac- tant
concentration).
2. In hard brines the prevalence of divalent cations causes the
formation of surfactantidivalent complexes, which have a low
solubility in brine. * Precipitation of this surfactantidivalent
complex will lead to retention. When oil is present this effect is
lessened by the surfac- tams solubility in the oleic phase.
3. At hardness levels somewhat lower than those re- quired for
precipitation, the preferred multivalentisurfac- tmt complex will
be a monovalent cation that can exchange chemically with cations
originally bound to the in MP formulation and clay type
distribution as well as salinity effects. HowevJer. the figure does
capture a gener- al trend that is useful for a first-order estimate
of reten- tion in a given reservoir.
-
47-16
or weak correlation. X3 The strong correlation in Fig. 0 LAB
DATA 0 FIELD DATA
WEIQHT FRACTION CLAYS
Fig. 47.21-Surfactant retention and weight fraction of
clays.
Fig. 47.22-MP production response from Well 12-l. Bell Creek
Pilot
Et? 0.6
0.4 h
0.2
0 0 0.2-0:4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2 0. 3.2
MOBILITY BUFFER SIZE ( VM~ 1
Recovery Efficiency vs. Mobility Buffer Size
Fig. 47.23--Recovery efficiencies from 21 MP field tests
PETROLEUM ENGINEERING HANDBOOK
A useful way to estimate the volume of surfactant re- quired for
an MP slug is through the dimensionless fron- tal advance lag,
FL).
1-o Pr c FD=p----,
4 P,C, (10)
where FD = frontal advance lag, dimensionless, p,. = density of
rock, mass/volume rock,
p\- = density of surfactant slug, mass/volume
solution,
C, = concentration of surfactant, mass
surfactantimass solution, and
6, = retained surfactant concentration, mass
surfactant/mass rock (includes all forms
of surfactants).
F expresses the volume of surfactant retained at its in- jected
concentration as a fraction of the PV. For best sur- factant usage,
the volume of surfactant injected should be large enough to contact
all of the PV, but small enough to prevent excessive production of
the surfactant. There- fore, the MP slug size, V,I,v, should be
equal to or some- what larger than FD.
Field Response. Fig. 47.22 shows the produced fluid analyses of
Well 12- 1 in the Bell Creek (Carter and Pow- der River Counties,
MT) MP flood. This tlood used a high oil content MP slug preceded
by a pretlush that con- tained sodium silicate to lessen surfactant
retention and reduce divalent cation concentration. Well 12-l was a
producer in the center of a unconfined single 40-acre, five- spot
pattern. Further details on the tlood are available in Refs. 47 and
82.
Before MP slug injection in Feb. 1979, Well 12-l was
experiencing low and declining oil cuts. MP oil response beginning
in late 1980 is superimposed on this decline, reaching peak cuts of
about 13% about 6 months later. Note that. just as in polymer
flooding, the pre-MP decline must be clearly established for
accurate evaluation of the MP oil recovery. The surfactant is
preceding the oil in Fig. 47.22 because of an excessively large
content of water-soluble, inactive disulfonate in the MP slug.
Simul- taneous oil and surfactant production is a persistent fea-
ture of field MP floods, probably because of heterogeneities and
dispersive mixing. Other significant features in Fig. 47.22 are the
evident presence of the pretlush preceding the MP slug, inferred
from the maxima in the pH and silicate concentrations. and the very
effi- cient removal of the calcium cations ahead of the sur-
factant.
Fig. 47.23 shows oil recovery efficiency, EK (in- cremental oil
recovered/OIP at start of MP process), from a survey of more than
40 MP field tests correlated as a function of mobility-buffer slug
size. As of the date of the survey there were no commercial
projects reported. Similar analyses on other process variables
showed no 47.23 indicates the importance of mobility control in MP
design. Note also from this figure that ultimate oil recovery
efficiency averages about 30% in field tests.
-
CHEMICAL FLOODING
Performance Prediction. Generally. three things must be achieved
for efficient oil recovery. X4 ( I ) the MP sur- factant slug must
be propagated in an interfacially active mode, (2) enough
surfactant must be injected so that some of it is unretained by the
permeable media surfaces, and (3) the MP displacement must be
designed so that the ac- tive surfactant sweeps a large portion of
the reservoir without excessive dissipation (because of dispersion)
or channeling.
Attaining the first goal is the result of formulation work based
on the phase-behavior concepts discussed previous- ly. The extent
to which the second and third goals are satisfied depends on
prevailing economics, which, in turn, depends on the oil-recovering
ability of the entire proc- ess. The next few paragraphs describe a
simple proce- dure by which oil recovery and oil rate-time curves
may be estimated for an interfacially active MP process. Since
there are innumerable ways in which interfacial activity may be
lost. the procedure is most accurate for processes that clearly
satisfy the first design goal.
Recovery Efficiency. This procedure has two steps: es- timating
the recovery efficiency of an MP flood and then proportioning this
recovery according to in.jectivity and fractional flow to give an
oil rate-time curve. Space con- siderations will limit many of the
details, which may be found in Ref. 80.
The recovery efficiency. ER, of an MP flood is the product of a
volumetric sweep efficiency, E. a displace- ment efficiency. En,
and a mobility buffer efficiency. E MB
ER=EDEVEMB, . . . (11)
Each quantity must be calculated independently. Displacement
Efficiency. The displacement efficien-
cy of an MP flood is the ultimate (time-independent) volume of
oil displaced divided by the volume of oil con- tacted.
S,, ED=1 -p, . .(l2)
S OM
where S,,, and S,,. are the ROS to an MP and water- flood,
respectively. S,,,,. must be known, but S,,, can be obtained from a
large slug (free from the effects of surfactant retention)
laboratory coreflood. Low values of Sor indicate successful
attainment of good interfacial ac- tivity in the MP slug. If
coreflood results arent availa-
ble, S,, may be estimated from a CDC by using a field capillary
number.
N,=0.565q~L,,y,~l(h~), .._.. ..(13)
where N,. = capillary number, dimensionless,
q = injection/production rate, CL 0 = oil viscosity,
h = net thickness, and
A = pattern area. Eq. 13 is in consistent units. For screening
purposes, as- sume a controlling IFT of lop3 dyne/cm [ 1 pN/m]; the
CDC curve chosen to estimate S ,,r should be consistent, 47-17
Fig. 47.24-Relationship between MP volumetric sweep efficien-
cy, heterogeneity, and slug-size retention ratio.
as much as possible, with conditions of the candidate
reservoir.
Volumetric Sweep Efficiency. Volumetric sweep effi- ciency, E v,
is the volume of oil contacted divided by the volume of target oil.
EV is a function of MP slug size, V ,,, , retention, FD, and
heterogeneity based on the Dykstra-Parsons coefficient, KDP. Fig.
47.24 shows this relationship. KDP may be estimated from geologic
study, from matching the previous waterflood, or from core data (a
typical value would be 0.6). The FD is estimated from Eq. IO on the
basis of the retention level, C,, . surfactant slug concentration,
C,, porosity, and rock and fluid den- sities. C, can come from a
laboratory coreflood, Fig. 47.21 (if clay fraction is known), or by
using C, =0.4 mglg as a default. L,,s and C,, are from the proposed
design.
MoKlity Buffeer Efficiency. The mobility buffer effi- ciency,
EBB, is a function of EV and KDP.
E,+,Be=0.71-0.6KDP, . . (l4a)
where EMBr is the mobility buffer efficiency extrapolated to VM~
=0 and VM~ is the mobility buffer volume, frac- tion V,,. Eq. 14a
was obtained from a numerical simu- lation.
The recovery efficiency now may be calculated from Eqs. 1 I
through l4a. The reasonableness of the value may be checked with
Fig. 47.23.
Calculation of q,, vs. t Plot. The production function (oil rate
vs. time) is based on ER and the following proce- dure. The
dimensionless production function is assumed to be triangular with
oil production beginning at oil bank arrival time, increasing
linearly to a peak (maximum) oil cut when the surfactant breaks
through, and decreasing linearly to sweepout time. The triangular
shape is imposed by the reservoir heterogeneity.
The first step is to calculate the dimensionless oil bank and
surfactant breakthrough times for a homogeneous lab- oratory
coreflood. (14b)
-
II I I b I J 0 100 roe 300 400 500 WO 700
TIME. DAYS
Fig. 47.25-Comparison between predicted and observed oil
rate-ttme responses for the Sloss MP pilot.
t~,=l+F,-S,,,. , _. __ (14c)
where
t Lhh = oil bank arrival time, dimensionless.
s ,,,, = oil bank saturation, fraction,
S,,, = initial oil saturation, fraction,
fr,,, = oil bank oil fractional flow. fraction.
f,,, = initial oil fractional flow, fraction. and
tLl\ = surfactant arrival time, dimensionless.
S,,h andf,,,, may be estimated from the oil/water relative
permeability curve as described previously in Refs. 80 and 85. or
from laboratory experiment.
The second step is to correct these values for the heter-
ogeneity of the candidate reservoir by using an effective mobility
ratio. M,. where
(1%
The corrected breakthrough times are now
ttDoh =tDot,lMe . (16a)
and tD,=tDs/Mc. (16b)
where tD(,,, is the corrected oil bank arrival time, dimen-
sionless, tlDr is the corrected surfactant arrival time. PETROLEUM
ENGINEERING HANDBOOK
dimensionloss. and the peak oil cut. ,f&. i,
M,,-M,, tl)r,h c > ! ? .fy,n = tl)1
of,,--I) ,f;,,, (17)
The final step is to convert the dimensionless production
function to oil rate, yo. vs. time. t. This follows from
q,,=qf;, ---- (l&i)
and
t=V,,tD/q, . .(18b)
wheref,, is the oil cut. tD is the dimensionless time. and .f;,
and tn are any points on the triangular oil recovery curve, which
begins at (t,,/, ,f;,;). peaks at (tljA .f;,,,k). and ends at
(tn,t., 0). The dimensionless time at com- plete sweepout, tD,,,,.,
is selected to make the arca un- der thef;, -tlD curve equal to
ER,
tD,,l.=tnoh +2E,$,,,.,,Jf;,,,,, (19)
A comparison of the results of this procedure with the Sloss
field MP pilot is in Fig. 47.25. Details of this match and other
matches are in the original references.
High-pH Processes The final chemical flooding EOR process is
high-pH flooding (Fig. 47.26). As in polymer and MP flooding. there
is usually a brine preflush to precondition the reser- voir, a
finite volume of the oil-displacing chemical, a graded mobility
buffer driving agent. and the entire proc- ess is driven by chase
water. Moreover. for both high- pH and MP flooding the
oil-displacing chemical is a sur- factant; however, for MP flooding
the surfactant is in- jected while in high-pH flooding it is
generated in situ.
High-pH Chemistry
High pHs indicate large concentrations of the hydroxide anions
(OH -). The pH of an ideal aqueous solution is defined as
pH= -log ioc, + , . . (20)
where the concentration of hydrogen ions, CH + , is in mol/L. As
the concentration of OH - is increased, the concentration of Hi
decreases, since the two con- centrations are related through the
dissociation of water,
K,,, = ( OH-)(CH+)
, . . . . . . . . . . . . CH~O
(21)
and the water concentration is nearly constant. These con-
siderations suggest two means for introducing high pHs into a
reservoir: dissociation of a hydroxyl-containing spe- cies or
adding chemicals that preferentially bind hydro- gen ions. Many
chemicals could be used to generate high pH. but the most commonly
used are sodium hydroxide (caustic, NaOH). sodium orthosilicate.
and sodium carbonate (Na2C03). NaOH generates OH by dissociation:
the
-
hOH by itself is not a surfactant. since the absence of a
lipophilic tail makes it exclusively water-soluble. If the
crudeHA,,. phase
HA
The edepenions tion factanmost
If tle sucharafloodinis theto nement,specieHA,, the Kof ac(CO?an
acas low
ies, the distinction among effects becomes important in
3 oil contains an acidic hydrocarbon component, some of this,
HA,,., can partition to the aqueous
where it can react. Xh
,,F-?A,;+H+. _. __ . e-4
xact nature of HA,, is unknown and probably highly dent on crude
oil type. The deficiency of hydrogen in the aqueous phase will
cause the extent of this reac- to be to the right. The anionic
species A,; is a sur- t that can have many of the properties and
enter into of the phenomena described above for MP flooding.
there is no HA,, originally present in the crude. lit- rfactant
can be generated. A useful procedure for
cterizing crudes for their attractiveness to high-pH g is
through the acid number. The acid number
milligrams of potassium hydroxide (KOH) required utralize I gram
of crude oil. To make this measure- the crude is extracted with
water until the acldlc s HA is removed. The aqueous phase
containing , A,;, and H is then brought to pH=7 by adding
OH. For a meaningful value, the crude must be free idic
additives (e.g.. scale inhibitor) and acidic gases
high-pH flooding.
0 I 2 3 4 CHEMICAL FLOODING
Fig. 47.26-Schematic of
latter two through the formation of weakly dissociating acids
(silicic and carbonic acid, respectively) that remove free H ions
from solution. High-pH chemicals gener- ally have been used in
field applications in concentrations ranging up to 5 wt% (in.jected
pHs of 1 I to 13) and with slug sizes up to 2O%PV. The resulting
amounts of chem- icals are quite comparable with the surfactant
usage in MP flooding: however, high-pH chemicals are substan-
tially less costly. This cost advantage must be discounted by the
historically lower oil recoveries in high-pH flooding. or H?S). A
good high-pH flooding crude will have id number of 0.5 mgig or
greater. but acid numbers as 0.2 mgig may be candidates, since only
a small 47-19
igh-pH flooding process
amount of surfactant is required to saturate oil/brine in-
terfaces. Fig. 47.27 presents a histogram of acid num- bers based
on the work of Jennings.87.8x
Displacement Mechanisms
Oil recovery mechanisms in high-pH flooding have been attributed
to eight separate phenomena. 89 This chapter concentrates on only
three: IFT lowering. wettability reversal, and emulsion formation.
The last two mecha- nisms also are present in MP flooding but are
dwarfed by the low-IFT effect. With smaller ultimate oil recover-
ACID INDEX INTERWL (E&Cl4 05mp. KOH/p RAffiE)
Fig. 47.27-Histogram of acid numbers
-
47-20
1 M 0.26 wt. 46 NaCI M 0.50 wt. % NaCI A 1.00 wt. 16 N&I
IO
SOLUSILIZATION
Y
I- \\ \\
z lOOr
\\ \\
E
\:
a
t
lo- -
IO- IO- IO-
WEIGHT % NaOH IC
Fig. 47.28-IFTs for caustic/crude/brine systems
The generated surfactant, A ,, aggregates at oil/water
interfaces, which can lower IFT. 86 In general, such low- ering is
not as pronounced as in MP flooding but, under certain conditions,
can be large enough to produce good oil recovery. Fig. 47.28 shows
IFT measurements of caus- tic solutions against Long Beach crude
oil at various brine salinities. The IFTs are sensitive to both
NaOH concen- tration and salinity, showing minima in the NaOH con-
centration range of 0.01 to 0.1 wt %. The decrease in IFT in these
experiments is limited by the spontaneous emul- sification of the
oil/water mixture when the IFT reaches a minimum.
There are many similarities in the low-IFT effects in MP and
high-pH flooding. The data in Fig. 47.28 show a clear resemblance
to the data in the upper plot of Fig. 47.19 except they are plotted
vs. NaOH concentration (presumed proportional to A, concentration)
instead of salinity. This suggests an optimal salinity of about I
.O wt% NaCl for a 0.03 wt% NaOH solution. Indeed, the work of
Jennings et al. 87 has shown that there is an op- timal NaOH
concentration for a given salinity in oil recov- ery experiments.
Moreover, the presence of the emulsification effect when IFTs are
low is exactly what one would expect from Fig. 47.17 at a
surfactant con- centration above the invariant point surfactant
concentra- tion. This suggests that the data in Fig. 47.28 showing
a Type II( -) phase environment at low NaOH concen- trations are
Type II(+) at high (similar to what would be expected from the
dilution effect in MP flooding). Fur-
ther work is necessary to establish the connection to MP
phase-behavior definitively, since the actual surfactant
concentration A; is likely to be much lower in a high- PETROLEUM
ENGINEERING HANDBOOK
pH system. However, Nelson PI a!. )( show that a cosur- factant
can increase the optimal salinity in a high-pH sys- tem much like
in MP systems.
See Chap. 28 for a discussion of wettability and its ef- fects
on petrophysical properties. Owens and Archery showed that
increasing the water wetness increased ulti- mate oil recovery,
where the wettability was reported as decreasing the water/oil
contact angle measured on polished synthetic surfaces. This has
also been shown by others using high-pH chemicals.9.YX The
increased oil recovery is the result of two mechanisms: a relative
per- meability effect, which causes the mobility ratio of a dis-
placement to decrease, and a shifting of the CDC (see Fig. 47.1
I).
Cooke et al. 94 have reported improved oil recovery with
increased oil wetness. Other data show that oil recov- ery is a
maximum when the wettability of a permeable medium is neither
strongly water- nor oil-wet. )5 Given the latter information, the
important factor may be the change in the wettability rather than
the actual wettabili- ty of the final state of the medium. In the
original wet- ting state of the medium, the nonwetting phase
occupies large pores and the wetting phase small pores. If the wet-
tability of a medium is reversed, there will be nonwet- ting fluid
in small pores and wetting fluid in large pores. The resulting
fluid redistribution, as the phases attempt to return to their
natural state, would make both phases vulnerable to recovery
through viscous forces.
High-pH chemicals can cause improved oil recovery through the
formation of emulsions. The emulsification produces additional oil
in at least two ways: through a mobility ratio lowering since many
of these emulsions have a substantially increased viscosity and
through solubilization and entrainment of oil in a flowing aque-
ous stream. The first mechanism improves displacement and
volumetric sweep as do the mobility control agents discussed
previously. Local formation of highly viscous emulsions should be
discouraged, however, as these would promote viscous fingering from
the less viscous oil-free high-pH solution. The solubilization and
entrain- ment mechanism would be more important when the IFT
between the swollen water phase and the remaining crude is low.
Fig. 47.28 shows that for certain conditions, emul- sification and
low IFTs occur simultaneously. McAuliffe showed that emulsions
injected in a core and those formed in situ give comparable oil
recoveries.96.97
Rock/Fluid Interactions
Interactions of the high-pH chemicals and the permeable media
minerals can cause excessive retardation in the propagation of
these chemicals throughout the permea- ble medium. This chapter
discusses three aspects of rock- fluid interactions: formation of
divalent/hydroxide com- pounds, cation exchange, and mineral
dissolution.
OH - ions themselves are not appreciably bound to the solid
surfaces; however, in the presence of multivalent cations they can
form hydroxyl compounds,
M +-+x(OH -)FiM(OH),,, (23) which, being relatively insoluble,
can precipitate from so- lution. This reaction, in turn, lowers the
pH of the solu- tion, and also can cause formation damage through
pore
-
CHEMICAL FLOODING
blockage and fines migration. The anionic surf&ant spe- cies
A,, can interact with the inorganic cations in solu- tion just as
in MP flooding; however, the interaction with the divalent cations
usually takes precedence, particular- ly in hard brines (see Fig.
47.3) or where there are sub- stantial quantities of soluble
multivalent minerals. Because of these interactions and those
involving the surfactants A ,;. high-pH processes are as sensitive
to brine salinity and hardness as are MP processes.
Other high-pH rock/fluid interactions are intimately as-
sociated with the clay minerals. Clays are hydrous alu-
minum/silicate compounds that occupy the smallest (less than 2
microns) particle size in typical media. Macroscop- ically. clays
occur as segregated streaks of variable degree of continuity
throughout a typical reservoir, or as distrib- uted clays. which
can line pore walls or fill pore throats. Distributed clays are of
most concern here, since these have quite large surface areas (15
to 40 mig clay), and therefore can exhibit considerable reactivity.
9X Chemi- cally, clays can take on a variety of formulas that
differ substantially in their reactivity even though the
differences in their molecular formulas are apparently minor.
The ability of a clay mineral to exchange divalent cat- ions
with an aqueous solution can drastically change the ionic
environment of a solution with which it is in con- tact. Clays have
excess negative charges caused by the substitution of +2-valence
minerals for +3-valence min- erals within the octahedral or
tetrahedral crystal lattice. 99 The cation exchange capacity, Zv,
is a measure of this excess negative charge; typical Zvs are I to
10 meqi 100 g clay for kaolinite and 100 to 180 meq/lOO g clay for
mont- morillonite. These free anionic sites are covered with cat-
ions from the solution, each of which has a specific degree of
selectivity for the particular clay site. In general, H + has high
clay selectivity, and divalent cations are bound more strongly than
are monovalents. This means that the anionic sites can be occupied
predominantly by H+ and/or divalents even when clays are in contact
with rela- tively soft brines. Any subsequent change in the
electro- lyte environment of the contacting solution can cause the
clays to take or give up these cations with a possible detrimental
effect on high-pH (and MP) flooding.
H cations can exchange on the clay sites with the in- jected
sodium according to
clay-HfNa Sclay-NafH. .(24)
where clay represents a mineral exchange site. I) The rcverxiblc
reaction Eq. 24 will clearly cause the H con- centration to
increase with a resulting pH decline. Fig. 47.29 shows the extent
of the OH - retardation caused by cation exchange in laboratory
tloods. Note that many of the lower pHs may require more than 3 PV
of fluid in-jection to attain the injected pH.
Unlike MP flooding. high-pH chemicals can react directly with
clay minerals and the silica substrate to cause consumption of OH ~
ions. The reactions with clays are manifest by the elution of
soluble aluminum and silica spe- cies from core displacements. The
resulting soluble species subsequently can cause precipitates
through hydroxyl reactions as in Eq. 24. lo2 The rate of hydrox- yl
consumption from this slow reaction (cation exchange is generally
fast enough so that local equilibrium applies) 47-21
w
Fig. 47.29-Effluent histories pH from laboratory corefloods. Ex-
perimental curves are solid lines with points, theo- retical
results are dashed lines.
is determined by a dimensionless DamkGhler number,
ND,,
N,, =+k L/u,,., . . .(25)
where k is the reaction rate constant, time - , and L is the
medium length for a first-order reaction. ND, is the ratio of the
reaction rate to the bulk fluid rate. If all con- ditions are
equivalent between a laboratory experiment and a prototype field
flood, ND, clearly will be much larger in the field than in the
laboratory owing to the much larger length scale. A larger ND,
implies more reaction relative to the residence time within the
system. Thus, it follows that the penetration distance-the distance
traveled by full-strength OH - ions-will be considera- bly smaller
in the field than in the laboratory. Bunge and Radke, who
illustrate this with several numerical calcu- lations, caution
against extrapolating laboratory-measured values of OH -
consumption to field cases unless the dis- crepancies in ND, have
been taken into account.
Field Results
High-pH field tests of articular interest include a wetta-
bility reversal test, 99 an emulsion flood,Y6 and a polymer-driven
flood. lo3 Fig. 47.30 shows the produc- tion data from a high-pH
flood conducted in the Whittier field. 04 The crude oil was 20API
with a 40-cp viscosi- ty, and the 0.2 wt% NaOH chemical was
injected as a
0.23.PV slug. There are many features in these data that are
common
to the responses of the other chemical flooding processes in
Figs. 47.6 and 47.22. The oil production rate declines as the total
fluid production increases, indicating a declin- ing oil cut. The
oil rate response to the caustic injection is again superimposed on
the waterflood decline, which is ext