SPE
Society of PetroIeun ErlgOleers of AIME
SPE 12039
Comparison of Sodium Carbonate, Sodium Hydroxide,
and Sodium Orthosilicate for EOR
by J.H. Burk, Stauffer Chemical Co.
Member SPE-AIME
Copyright 1983 Society of Petroleum Engineers of AIME
This pape.rwas presented at the 5~th.Annuar Tec~nical C::onference and Exhibition held in San Francisco, CA, October 5-8,1983. The material is subject
to correction by the author. Permission to copy IS restricted to an abstract of not more than 300 words. Write SPE, 6200 North Central Expressway
'
Drawer 64706, Dallas, Texas 75206 USA. Telex 730989 SPEDAL.
ABSTRACT
Alkaline and alkali/polymer flooding are economic and versatile techniques propo'sed for applica-
tion in enhanced oil recovery (EOR). Recently, the
use of high-pH alkaline chemicals was emphasized, to
obtain low interfacial tensions with the crude oil
and counteract alkali loss by both reservoir rock
and connate water reactions. This experimental study
compares the utility of Na2C03 buffer solutions with
NaOH and Na4Si04, at equal Na20 levels, for alkaline
and alkali/polymer flooding.
The alkalis were found to be equally effective
in reducing interfacial tension with Wilmington,
Ranger-zone crude, and eight other acid crude oils.
Tertiary recovery results from twenty Berea corefloods with the Ranger-zone crude and hard reservoir
brine were equivalent (6-17%) for Na2C03, NaOH, and
Na4Si04' Similar modest recoveries were obtained for
polymer floods using polyacrylamide or xanthan gum.
However, polymer augmentation of alkaline floods
dramatically increased tertiary oil yields to 73-95%.
In addition, alkali was found to improve the injectivity of polymer solutions. The alkali reactions
with sandstone were much less severe with Na2C03
than with NaOH or Na Si0 •
4
4
INTRODUCTION
Over the last ten years, numerous reports have
been published on the theory of alkali interaction
with reservoir rock and fluids and on experimental
work with alkaline flooding .l- 5 The majority of the
alkaline floods in the field, and most laboratory
studies, have been carried out with sodium hydroxide
(NaOH) or sodium orthosilicate (Na4Si04).5 Use of
high pH and high alkali concentrations up to five
weight percent was deemed necessary to obtain low
interfacial tensions ~7ith the acid crude oils,6,7 to
counteract alkali 10s8,8,9,10 and to decrease alkali
retention time in the formation. 3 ,IO
References and illustrations at end of paper.
Sodium carbonate (Na2C03) and other alkaline
buffers which have a lower pH than NaOH or Na4Si04'
have not been studied extensively.ll,12 The use of
all alkali buffer such as Na2C03 could be advantageous, since the buffered slug would be less reeactive with sandstone minerals due to reduced
hydroxyl ion activity. Its lower pH would be maintained over a wide concentration range despite
alkali consumption within the formation. Results
from very early field flooding trials with Na2C03
were not conclusive. 13 However, more recent field
tests with Na2C03 look promising, especially the
successful implementation of a Na2C03/polyacrylamide
polymer flood in the Isenhour, WY field by Belco
Petroleum Corporation. 14 ,15 Use of alkali in conjunction with polymer to obtain low tension and
improved mobility ratio has been shown to increase
tertiary oil yields substantially;16,17,17a more
specifically, a Na2C03/polyacrylamide system was
reported to be very effective in laboratory corefloods. IS
On a contained Na20 basis, Na2C03 is costeffective compared to NaOH or Na4SiOl,; 5 but unlike
the other alkalis, i t is readily available from vast
trona deposits in Wyoming. NaOH and Na4Si04 prices
can fluctuate substantially due to economic conditions because their manufacture is tied to other
processes. 1 9 This experimental study presents a
side-by-side comparison of NaOH and Na4Si04 with
Na2C03 011 an equal contained Na20 basis. Results
from interfacial tension measurements, and alkaline
and alkali/polymer corefloods are reported. The
aspects of divalent ion precipitation, slug injectivity, and alkali retention and consumption are
discussed.
EXPERIMENTAL DETAILS
A fresh sample of crude oil was provided for
this study by Long Beach Oil Development Company
(LBOD). The sample was taken from the XYZ farm,
Long Beach lease, Ranger-zone, Wilmington field. 20
The oil was stored under nitrogen prior to use~
The following crude oil properties were determined:
total acid number, 2.S0 mg KOH/g; density, 929
k~/m3; and viscosity, 66 mPa·s (52 0 C).
COMPARISON OF SODIUM CARBONATE, SODIUM HYDROXIDE AND SODIUM ORTHOSILICATE FOR EOR
Z
The water formulations were prepared according
to the composition of the formation and injection
water provided by LBOD. Laboratory samples of
"Produced Water" contained Z.98% TDS, with 1.8Z% Cl,
1.04% Na, 0.045% Ca and 0.043% Mg as major constituent ions.
The "Fresh Water" samples were prepared
with 0.776% TDS, containing 0.440% Cl and 0.Z95%
pore volumes), a 0.41 PV oil saturation was chosen
for initiation of tertiary flooding in 'the early runs.
However, due to variations in the flooding behavior of
the different cores, attainment of this oil saturation
was found to be unreliable, and a 20:1 water-to-oil
ratio (WOR) was used subsequently to signal the waterflood end point.
Na as major ions. The calcium and magnesium hardness
ion contents in the fresh water were each less than
1 ppm.
SPE 1Z039
After the waterflood, the following injection
sequence was used for the alkaline, polymer and
The 0.5 and 1.0 weight percent NaZO alkaline
solutions (Table 1) were prepared with fresh water
from sodium carbonate (Stauffer Chemical Company
Dense Soda Ash), a 50% sodium hydroxide solution, or
a 10% stock solution of sodium orthosilicate. Zl
The polyacrylamide solutions were prepared from
alkali/polymer floods:
(1)
0.Z5 PV fresh water pre-flush at 0.30 m/day
(1. 0 ft/ day) .
(Z)
1.0 PV chemical slug injection at 0.30 m/day.
(3)
0.10 PV fresh water post-flush at 3.05 m/day.
(4)
3-5 PV produced water post-flush at 3.05 m/day.
a stock solution containing 5000 ppm of American
Cyanamid CYANATROL® 940 in fresh water. The xanthan
gum solutions were made up with Pfizer FLOCoN® 4800C
and fresh water, to yield a ZOOO ppm active solution.
The polymer stock solutions were then either diluted
with fresh water or blended with the alkaline
chemical followed by dilution with the corresponding
In two cases, Steps (3) and (4) were also carried
out using a frontal advance rate of 0.30 m/day to determine the effect of the enhanced post-flush injection rates on oil recovery.
Incremental samples were
alkaline solution to the appropriate Na20 concentra-
collected for Steps (Z) through (4), and the produced
tion. The amount of polymer needed for the solutions
with a target viscosity of 50 mFa's (5Z 0 C) are given
under Results and Discussion (Tables 6 through 8).
pH and alkali contents.
fluids were analyzed for oil cut, calcium, magnesium,
RESULTS AND DISCUSSION
The interfacial tension (1FT) measurements were
carried out on a constant rate (3600 rpm) spinning
Interfacial Tension Measurements
drop interfacial tensiometer for tensions less than
1.0 mN/m, and on a duNouy ring instrument for
tensions greater than 1.0 mN/m. The spinning drop
measurements with the Wilmington, Ranger-zone crude
oil were made at reservoir temperature (5Z 0 C) after
five minutes of oil/alkali contact. The duNuoy ring
measurements were carried out at 21°C.
Each alkali
was tested at concentrations from 0 to Z.O wt% NaZO
in fresh water.
1FT lowering measurements were also
made with eight other California or Mid-Continent
crude oils at ZloC using alkaline solutions in distilled water with 1.0 wt% added NaCl.
The Berea corefloods were carried out with a
positive displacement Ruska pump psed in conjunction
0
Interfacial tension (1FT) measurements between a
crude oil and an alkaline solution have generally been
accepted as a screening tool, to evaluate the enhanced
oil recovery potential of the crude by the
alkali. 6 ,7,2Z 1FT data were collected for the Rangerzone crude with fresh water solutions of NaZC03' NaOH
and Na4Si04 at 5Z o C (Figure 1). All three alkalis
showed very good tension lowering behavior, with 1FT
minima of 0.042 mN/m for Na2C03 at pH 10.4, of 0.058
mN/m for NaOH at pH lZ.5 and of 0.OZ3 mN/m for Na4Si04
at pH 11.9. Observation of a tension minimum with the
NaZC03 solution and this Ranger-zone crude at pH 10.4
is indicative of the presence of surfactant precursor
organic acids with dissociation constants greater than
with a constant temperature chamber held at 52 C.
The cylindrical, unfired cores were 61 cm long and
10- 10 in this crude. 5 ,ZZ,Z3
5.1 cm in diameter, cut from a single block of Berea
Additional 1FT measurements were carried out with
eight California or Mid~Continent crudes in order to
ascertain that the low-pH alkali response is not specific to this Ranger-zone crude. The 1FT minima vs pH
for the crude oils along with their acid numbers are
sandstone, with an average pore volume (PV) of
Z70 mL. The cores were of medium porosit with a
produced water permeability of 0.4-0.6 ~m. The
1
cores were encased in high temperature epoxy resin,
with two internal pressure taps mounted through the
listed in Table Z.
The tests were conducted with
epoxy on the core surface so as to divide the core
sodium carbonate and sodium hydroxide solutions at
into three equal lengths for pressure measurements.
o - Z.O wt% NaZO in distilled water with 1.0 wt% NaCl
at 2l oC. Crudes I, II and VII generally exhibited a
modest alkali response with 1FT minimum values ?D.l
mN/m for NaZC03 and NaOH at pH ~ll.Z. The crude oils
VI and VIII gave slightly lower tension minima (~0.08
mN/m) at a pH range from 10.1 to 11.Z for both
Typically, the core and fluids were brought to
52°C, and the core was saturated with brine by
evacuation and subsequent injection with produced
water.
All fluids were filtered through a 0.8
Millipore filter prior to injectione
~m
The brine was
alkalis. The remaining crudes III, IV and V showed,
in contrast, very good interfacial tension lowering,
then displaced by injection of the Ranger-zone crude
at a rate of 3.05 m/day, until no further water was
produced. This was followed by injection of produced water at 0.30 m/day to simulate primary and
especially with NaZC03 solutions at pH <11. This
limited study thus indicates that crude oils with a
secondary recoveries in a reservoir.
Based on the
flooding characteristics of two cores used for ex-
lowering with alkaline solutions at pH 10 to 11.
tended waterfloods (0.60 m/day injection with 7-8
good alkali response can undergo substantial tension
J. H. BURK
SPE 12039
Laboratory Corefloods with Wilmington, Ranger-Zone
Crude
The coreflood experiments were carried out in
Berea sandstone cores to provide a uniform and re-
producible medium and to allow comparison of this
work with other recent coreflood studies. 5 ,7,lS,2l
Compositional analysis of a representative core
3
teri:iary recovery of 17% obtained in Run 5. The sodium
carbonate floods at 0.5 wt% Na20 (Runs 3 and 4, Table
4) were carried out in duplicate to establish reproduc:lbility, yielding recoveries of 8.S and 11.2%. The
2.4% difference in tertiary recovery would be typical
of the flooding behavior of two different Berea cores.
Overall, tertiary recovery by alkaline flooding appears
to be potentially interesting for field applica-
sample gave 91.5% Si02, 3.Z% AlZ03, 0.9% KZO, 0.7%
Fez03 and 0.4% CaO due to mostly silica and some clay
minerals. The silica was identified as a-SiOZ by
tion.
X-ray diffraction methods, and the Berea sandstone
to explore the ~ffect on ter:iar y :ecov:ry o~ mobility
control alone/ ,17, l7a and 1n conJunct1on w1th low
surface area was measured to be approximately Z mZ/g
(BET-N )'
Z
Two additional sets of corefloods were carried out
tension.17a,1~ For the first set, polyacrylamide or
xanthan gum was employed in fresh water solutions
Large chemical slugs with 1. 0 PV were employed
(Table 6, Runs 11 and 12). In the second set, 0.5 wt%
Na20, Na2C03, NaOH and Na4Si04 solutions, augmented by
polymer, were injected (Tables 7 and S, Runs 13
tems. The alkaline slug concentrations of 0.5 and
through ZO). In both series, sufficient polymer was
1. 0 wt% Na20 (Table 1) were based on the favorable
added to the injected fluids to achieve a 50 mFa's
low tension alkali response exhibited by Na2C03' NaOH Brookfield viscosity slug at 52°C, which afforded a
and Na4Si04 with the Ranger-zone crude in this confavorable mobility ratio with the Wilmington, Rangercentration range (Figure 1). The pH values observed
ZOnE, crude (66 mPa's, 5Z 0 C).
at 0.5 and 1.0 wt% Na20 for the three alkalis are
listed in Table 1.
The tertiary oil yields obtained with the polymer
floods, shown in Table 6, were lS.O% for the polyThe polymers chosen for this evaluation were the acrylamide and 10.5% for xanthan gum systems. Thus,
commercially available CYANATROL® 940 polyacrylamide
recoveries were very similar to the Na2C03 alkaline
(American Cyanamid) and the FLOCoN® 4S00C xanthan gill' flood recoveries (Table 4). The dramatic effect of
biopolymer (Pfizer). Both are recommended for EOR
mobility control with low tension was evident in the
applications by their manufacturers. The polymer
high tertiary recoveries obtained with the alkaline/
concentrations were adjusted to achieve a high and
polymer floods. Tertiary recoveries for alkali/
comparable injected Brookfield viscosity of 50 mPa's, polyacrylamide were 82-95% for NaZC03, S4% for NaOH,
in this work to magnify potential differences in oil
recovery attributable to the various chemical sys-
and associated with this a similar improvement in
and 92% for Na4Si04.
mobility ratio for all polymer and alkali/polymer
floods. 16 ,lS,Z4
eries for alkali/xanthan gum were 7S-8l% for Na2C03'
79% for NaOH and, 73% for Na4Si04 (Tables 7 and S, Runs
13-15 and 17-20). In these studies, the three alkalis
at 0.5 wt% Na20 levels appeared to be equally effec-
Berea Coreflood Oil Recoveries
tive.
Two extended waterfloods, rUn in representative
Berea cores, provided baseline data on flooding
characteristics, oil recoveries, and residual satura-·
tions for comparison with the different chemical
systems. Thus, continuous injection of 7.11 and
7.32 PV of produced water for Runs 1 and 2 (Table 3)
yielded waterflood recoveries of 54.2 and 49.9%,
respectively. Water to oil ratios (WOR) greater than
20: 1 were attained after injection of about 1. 0 to
1.5 PV. The difference in the residual oil saturation (Sor) for these two experiments was small and
consistent with the variations in the initial
saturation (Soi)'
In comparison, tertiary recov-
However, for the polymers, the polyacrylamide
gave consistently higher tertiary oil yields than the
xanthan gum systems. With the exception of Run 16,
which employed a slow 0.30 m/day post-waterflood, all
runs were carried out with the routine 3.05 m/day
post-flush.
The reduced post-flush injection rate had little
effect on tertiary recovery, since most of the oil was
displaced efficiently by the advancing alkali/polymer
front.
These oil recovery data are in general agree-
ment with earlier reports for NaOH, showing the following ordering for Berea corefloods: polymer < alkali
< alkali/polymer. 17 ,17a The cumulative tertiary oil
recoveries from this work for alkali, polyacrylamide,
The results of the alkaline Berea corefloods are
given in Tables 4 and 5. Generally, an increase in
alkali concentration gave increased tertiary recovery. For equal NazO concentrations, NaZC03 and
Na4Si04 (Runs 3,4,5,9 and 10) yielded similar and
higher recoveries than those obtained with the NaOH
floods (Runs 7 and 8). All runs in Tables 4 and 5,
except RUn 6, were carried out with an :i.dentical
fluid injection procedure, employing a 3.05 m/day
post-flush sequence, \\Thich allowed attainment of
nearly residual oil saturations and very high WOR
within a reasonable 'time limit.
For Run 6, the
post-waterflood injection rate was kept at 0.30 m/day
for 3.2 days, resulting in delayed and continuing oil
production when the experiment was terminated.
With
sufficient fluid throughput, the yield from this
Na 2C0 flood would have probably been similar to the
3
and alkali/polyacrylamide floods are depicted in
Figu·res 2,3 and 4 for Na2C03, NaOH and Na4Si04, respectively. The very large tertiary oil yields obtain .. d with the polymer augmented core floods leave
little doubt about the effectiveness and promise of
alkali/polymer flooding.
Chemical Slug Injectivity
In conjunction with ~low rates, differential
pressures were measured across the front, middle and
rear sections of the core during each flood. Resis-
tance factors (RF)24 were calculated from the pressure
and flow data, according to:
COMPARISON OF SODIUM CARBONATE, SODIUM HYDROXIDE AND SODIUM ORTHOSILICATE FOR EOR
4
OH
SPE 12039
(2)
where
The rate of silica dissolution is first order with
respect to [OH-] and increases with increasing pH and
pressure drop at the end of waterflood
pressure drop at a given point during tertiary
flood
flow rate at the end of waterflood
temperature. 9,10 Consistent with this trend, the Si0 2
levels (Table 9) found for the Na2C03 flood (pH 11.2
11.3) were low and in the same range as observed for
the polymer floods containing no alkali. The NaOH
flood (pH 13.2 - 13.5) showed much higher Si02 levels
which rose, as expected, with increasing NaOH concen-
trations (Table 9). The Na4Si04 floods (pH 12.8 13.1) constituted a special case, due to dissolved
silicates in the injection fluids. At a 0.5 wt% Na20
flow rate at the time of P2 measurement
Increased pressures were observed with injection of
all chemical slugs. This can be attributed to
calcium and magnesium carbonate, hydroxide or
silicate precipitation in the core, crude oil emulsi-
fication, and, especially in the case of the polymer
floods, increased flowing viscosity_ Resistance
factors for the different chemical slugs, calculated
from flow rates and pressure differentials at the
front and rear of the cores, showed the following
ordering: polyacrylamide> xanthan gum » alkali/
polymer> alkali. Thus, RF values ranged from 1.5 51 for the polymer floods, and only from 1.4 - 15 for
the alkali/polymer systems. For the alkaline floods,
the RF data were lowest and in the 1.2 - 2.3 range.
The very high pressures observed with the polymer
slug injection are presumably due to resistance to
polymer flow in the presence of high residual oil
saturation; i.e., low tertiary recoveries were observed.
In contrast, the much lower pressures
observed with the alkali/polymer systems are attributed to efficient oil displacement at the front of
the slug, resulting in increased permeability and
improved flow conditions. No substantial difference
was noted among the alkalis themselves, except that
slightly lower pressures were registered with the
alkali systems employing Na Co •
2 3
concentration, produced SiOZ levels were lower than
injected ones, presumably due to Ca and Mg silicate
precipitation in the core. However, at a 1.0 wt% Na20
level, the produced silica concentrations surpassed
injected levels, which is consistent with strong base
behavior.
Alkali loss in the Berea cores due to consumption
and retention of base by the fluids and the sandstone,8,10,25 was measured by titration of incremental
samples of produced fluids to pH 3.2. Generally,
alkali loss was less for the alkaline floods (3-14 meq)
than for the alkali/polymer floods (10-25 meq). Apparently, the alkali augmented with polymer contacts a
greater pore volume and rock surface area than alkali
alone, resulting in increased loss, and in much improved oil recoveries. The results for alkali loss and
breakthrough for representative alkaline floods are
shown in Table 10. The alkali loss data indicate
typical low-pH buffer behavior for the Na2C03 floods
and strong base action by the NaOH and Na4Si04 systems.
Thus, Na2C03 solutions at 0.5 to 1.0 wt% Na20 maintain
essentially constant pH and exhibit similar sandstone
reactivity and OH- loss with increasing concentration.
In contrast, the strong bases, NaOH and Na4Si04' show
a large increase in alkali loss with increasing concentration and hydroxyl ion activity.
Berea Sandstone Reactivity and Alkali Loss
The pH of the fluids produced from the cores was
monitored, and the calcium, magnesium, silica and
alkali contents were determined. Analysis of the
aqueous fluids consistently showed a dramatic decrease in Ca and Mg ion levels, and an increase in
silica levels coincident with alkali breakthrough.
This decrease in Ca and Mg concentration from greater
than 400 ppm into the 1-10 ppm range was observed for
all three alkaline slugs at 0.5 and 1.0 wt% Na20.
The decrease in hardness ion concentration from the
original levels in the produced water is attributed
Alkali breakthrough for the 1.0 PV Na2C03' NaOH
and Na4Si04 slugs was measured from plots of pH vs PV
produced fluids, at the point where the plot formed a
plateau near peak pH. The data in Table 10 show
clearly that alkali breakthrough depends on the Na20
concentration, and it occurs earlier with increased
alkali concentrations. This is consistent with the
chromatographic lag theory proposed for alkaline flooding by deZabala et al. 3 and by Bunge and Radke. lO The
results also show similar retention times for the
alkalis at equivalent NaZO concentrations, despite the
to dilution by the pre-flush and the chemical slug,
much lower pH of the Na2C03 solutions compared to NaOH
and Na4Si04. This finding was not expected; clearly,
produced fluids from the Na2C03' NaOH and Na4Si04
floods are shown in Table 9. Soluble silicates can
CONCLUSIONS
be formed in the alkaline coref1oods from the interaction of base with the Berea sandstone, according to
1.
more work is needed on long term chromatographic beand, more importantly, to precipitation of insoluble
havior of buffer solutions in rock formations. ll ,12
Ca and Mg carbonate, hydroxide or silicate salts in
the core. These short coref1ood studies thus indicate
Overall, these studies with alkaline core floods
that the hardness ions are reduced to low levels by
demonstrate potential benefits of Na C0 solutions
Na C0 , NaOH and Na Si0 • 7
2 3
4
Z 3
4
associated with low sandstone reactivity and buffer
behavior.
The peak silica concentrations observed in the
the equilibria described in Equations (1) and (2).10
(1)
Low tensions required for EOR by alkaline flooding
can be achieved with alkaline solutions at pH <11.
Na2C03, NaOH and Na4Si04 are equally effective in
1FT lowering with Wilmington, Ranger-zone crude
oil, and other crudes with good alkali response.
SPE 12039
2.
J. H. BURK
and Geothermal Chemistry, Stanford, CA (May 2830, 1980). Soc. Pet. Eng. J. (April 1982)
245-258.
Tertiary oil recovery results for Wilmington,
Ranger-zone crude with Na2C03' NaOH or Na4Si04
solutions indicate that these alkalis are
equally effective and high pH conditions are not
necessary for EaR by alkaline flooding.
3.
5
4.
Mungan, N.: "Enhanced Oil Recovery Using Water
as a Driving Fluid", World Oil (June 1981)
209-20, and (July 1981) 181-90.
5.
Mayer, E.H., Berg, R.L., Carmichael, J.D., and
Weinbrandt, R.M.: "Alkaline Injection for
Enhanced Oil Recovery - A Status Report",
J. Pet. Tech. (Jan. 1983) 209-21.
6.
Jennings, H.Y., Jr.,: uA Study of Caustic
Solution - Crude Oil Interfacial Tensions",
Soc. Pet. Eng. J. (June 1975) 197-202.
7.
Campbell, T.C.: "The Role of Alkaline Chemicals
in the Recovery of Low-Gravity Crude Oils",
J. Pet. Tech. (Nov. 1982) 2510-16.
8.
Lieu, V.T., Miller, S.G., and Staphanos, S.J.:
"Long Term Consumption of Caustic and Silicate
Solutions by Petroleum Reservoir Sands ll , presented at the ACS Symposium on Silicate
Chemistry, New York, NY (Aug. 26, 1981).
9.
Sydansk, R.D.: "Elevated-Temperature Caustic/
Sandstone Interaction: Implications for Improving Oil Recovery", Soc. Pet. Eng. J. (Aug.
1982) ')53-62.
10.
Bunge, A.L., and Radke, C.J.: "Migration of
Alkaline Pulses in Reservoir Sands", Soc. Pet.
Eng. J. (Dec. 1982) 998-1012.
11.
Radke, C. J ., and Somerton, W. H. : "Enhanced
Recovery with Mobility and Reactive Tension
Agents", paper B-2 presented at the Fifth
Annual U.S. DOE Symposium, Tulsa, OK (Aug.
22-24,1979).
12.
Chang, H.L.: "Oil Recovery by Alkaline Waterflooding", Canadian Patent No. 1,037,863 (Feb.
24, 1976).
13.
Nutting, P.G.: "Petroleum Recovery by Soda
Process", Oil and Gas J. (1928) !:2, No. 22, 146
and 238.
14"
Cooke, C.E., Jr., Williams, R.E., and Kalodzie,
P.A.: "Oil Recovery by Alkaline Waterflooding",
J. Pet. Tech. (Dec. 1974) 1365-74.
15.
Sloat, is.: "The Isenhour Unit - A Unique
Polymer Augmented Alkaline Flood", SPE 10719,
presented at the SPE/DOE Enhanced Oil Recovery
Symposium, Tulsa, OK (Apr. 4-7, 1982).
16.
Shah, D.O., and Schechter, R.S. (Editors):
"Improved Oil Recovery by Surfactant and
Polymer Flooding '1 , Academic Press, Inc.,
New York (1977).
17.
Mungan, N.: "Enhanced Oil Recovery Using Water
as a Driving Fluid", World Oil (Feb. 1, 1982)
95-106, and (March 1982) 71-75.
Polymer-augmented Na zC0 , NaOH, and Na Si0
3
4
4
floods recover significantly more oil ,73 to 95:%")
than either low tension alkaline floods or
polymer floods (6 to 18%).
4.
5.
Addition of alkali to polymer floods improves
injectivity due to efficient oil displacement
at the front of the slug.
Sodium carbonate solutions are less corrosive
to sandstone than NaOH or Na4Si04.
Na2C03
buffering action can reduce alkali retention
in the rock formation.
NOMENCLATURE
PV
Na20 -
Pore volume
Sodium oxide, yields two chemical
equivalents of NaOH in solution
Chemical flood
Waterflood
Initial oil saturation prior to
waterflood
S
or
Residual oil saturation after
waterflood
Final oil saturation after chemical
flood and post-flush
% WE' Recovery
(I-So/So i) x 100
% Tertiary Recovery -
(l-Sof/Sor) x 100
% Total Recovery
(l-So/SOi) x 100
ACKNOWLEDGMENTS
The interfacial tension measurements and Berea
coreflood experiments were carried out by Exoil
Services, Golden, CO. I would like to acknowledge
the Long Beach Oil Development Company for providing
us with the Wilmington field, Ranger-zon·e crude oil,
and American Cyanamid for granting partial support
for the alkali/polymer studies. I also want to
acknowledge the great help of Jim Ball of Exoil in
the design and evaluation of this work. I thank
Stauffer Chemical Company for granting permission to
publish this paper.
REFERENCES
1.
2.
3.
Jennings, H.Y., Jr., Johnson, C.E., Jr., and
McAuliffe, C.D.: "A Caustic Waterflooding
Process for Heavy Oils", J. Pet. Tech. (Dec.
1974) 1344-51.
Johnson, C.E., Jr.,: "Status of Caustic and
Emulsion Methods", J. Pet. Tech. (Jan. 1976)
85-92.
deZabala, E.F., Vislocky, G.M., Rubin, E., and
Radke, C.J.,: "A Chemical Theory for Linear
Alkaline Flooding", SPE 8997, presented at the
SPE Fifth International Symposium on Oilfield
17a. Radke, C.J.: "Additives for Alkaline Recovery
of Heavy Oil", Proceedings of the 1982 Heavy Oil
Conference (July 1982) 2-19.
6
18.
19.
COMPARISON OF SODIUM CARBONATE, SODIUM HYDROXIDE AND SODIUM ORTHOSILICATE FOR EOR
20.
22.
deZabala, E. F., and Radke, C. J . : "The Role of
Interfacial Resistance~ in Alkaline Waterflooding of Acid Oils", SPE 11213 presented at the
57th Annual Fall Tech. Conf. and Exhibition of
the SPE of AIME, New Orleans, LA (Sept. 26-29,
1982).
23.
Cham, M., and Yen, T.F.: "A Chemical Equilibrium Model for Interfacial Activity of Crude
Oil in Aqueous Alkaline Solution: The Effects
of pH, Alkali and Salt", Can. J. Chem. Eng.
(April 1982) 3Q5~308.
24.
Pye, D.J.: ,'·'Improved Secondary Recovery by
Control of Water Mobility", Trans. AIME (1964)
231, 911.
25.
Somerton, W.H., and Radke, C.J.: "Role of Clays
with Enhanced Recovery of Petroleum", SPE 8845,
presented at the First Joint SPE/DOE Symposium
on EOR Tulsa, OK (April, 20-23, 1980). J. Pet.
Tech. (March 1983) 643-645 •
Ball, J.T., and Stewart, D.P.: "Polymers in
Alkaline Flooding", presented at the l82nd ACS
Nat'l Mtg., New York, NY (Aug 23-28, 1981).
Smock, D.:
"Consider Soda Ash as Caustic Soda
Alternative", Purchasing (June 10, 1982)
94A7-A10.
Carmichael, J.D.:
"Caustic Waterflooding
Demonstration Project, Ranger-Zone, Long Beach
Unit, Wilmington Field, California", prepared
for Energy Research and Development Agency under
contract No. EF-C-03-l396, B-2/l-B-2/l8.
21.
Novosad, Z., and McCaffery, F.G.: "Laboratory
Evaluation of Sodium Hydroxide and Sodium Or thosilicate for Tertiary Oil Recovery in an Alberta
Reservoir", SPE 10734, presented at the 1982
California Regional SPE Meeting, San Francisco,
CA (March 24-26, 1982)
.
TABLE 1
WT% Na 20 vs WT% ALKALI AND pH
WT% ALKALI (I'H)
WT% Na 2Q
~2C03-
SPE 12039
NaOH
~~
0.5
0.85(11.2)
0.65(13.2)
0.74(12.8)
1.0
1.7 (11. 3)
1.3 (13.5)
1.5 (13.1)
TABLE Z
1FT MINIMUM VALUES vs pH
ORIGIN
GRUDE
Acid Number
Mg KOH/g Crude
IFT(mN/m)/pH
NaOH
Na ZC03--
I
N. Texas
<0.005
0.35/11. 5
0.Z2/1Z.1
II
Nebraska
0.007
O.l1/l1.Z
0.094/11. 5
III
New Mexico
0.35
0.016/10.3
0.033/11.0
IV
California
0.51
0.016/10.8
0.062/11.7
V
California
0.98
0.014/10.3
0.015/10.5
Texas
1. 23
0.076/10.1
0.071/9.3
Oklahoma
1.89
0.16/11.5
0.09/11.9
Wyoming
2. L,.2
0.08/10.9
0.08/11.2
VI
VII
VIII
TABLE 3
LONG WATERFLOODS WITH WILMINGTON RANGER-ZONE
CRUDE IN BEREA SANDSTONE CORES
RUN
1 ------
2
PV WF Inj.
7.11
7.32
S
oi
0.651
0.687
Sor
0.298
0.344
% WF Recovery
54.2
49.9
TABLE 5
TABLE 4
Na C0 COREFLOODS WITH WILMINGTON
2 3
NaOH and Na Si0 COREFLOODS WITH
4
4
RANGER-ZONE CRUDE
WILMINGTON RANGER-ZONE CRUDE
RUN
RUN
7
8
3
4
5
6
Cone., Wt% Na 20
0.50
0.50
1.0
1.0
Alkali
PV WF Inj.
0.58
2.99
0.65
1. 69
Cone., Wt% Na 0
2
0.5
1.0
0.5
1.0
PV CF Inj.
1.0
1.0
1.0
1.0
PV WF Inj.
2.05
1. 94
2.73
1. 86
So~.
0.711
0.745
0.725
0.784
PV CF Inj.
1.0
1.0
1.0
1.0
Sor
0.397
0.473
0.379
0.436
So~.
0.741
0.845
0.750
0.818
S
or
0.399
0.446
0.404
0.444
% WF Recovery
44.2
Sof
0.362
% Tertiary Recovery
8.82
% Total Recovery
49.1
36.5
0.420
47.7
0.314
44.4
0.389
% WF Recovery
NaOH
46.2
NaOH
47.2
9
10
Na Si0
4
4
Na Si0
4
4
46.1
11.2
17.2
10.8
S
or
0.377
0.404
0.368
43.6
56.7
50.4
% Tertiary Recovery
5.51
9.42
8.91
% Total Recovery
49.1
52.2
50.9
45.7
0.374
15.8
54.3
TABLE 6
TABLE 7
POLYMER COREFLOODS WITH WILMINGTON RANGER-ZONE CRUDE
Na C0 /POLYMER COREFLOODS WITH WILMINGTON
z 3
RANGER-ZONE CRUDE
RUN
POLYMER
Cone., ppm
11
12
Po1yac*
X-Gum**
3200
1540
RUN
Cone., Wt% Na 0
2
l3
14
15
16
0.5
0.5
0.5
0.5
Po1yac
Po1yac
X-Gum
X-Gum
3853
3828
1700
1700
PV WF lnj.
0.88
1.12
Polymer
PV CF Inj.
1.0
1.0
Polymer Cone., ppm
S .
0.736
0.771
PV WF lnj.
0.95
2.08
1. 05
1. 63
S
0.423
0.410
PV CF Inj.
1.0
1.0
1.0
1.0
S .
0.692
0.784
0.765
0.738
S
or
0.390
0.433
0.424
0.402
01
or
% WF Recovery
SOf
42.5
0.347
46.8
0.367
01
% Tertiary Recovery
18.0
10.5
% Wli' Recovery
% Total Recovery
52.9
52.4
Sof
* Polyacrylamide (CYANATROL® 940 brand)
** Xanthan Gum (FLOCON® 4800C brand)
43.6
0.018
44.8
0.076
44.6
0.082
45.5
0.088
% Tertiary Recovery
95.4
82.4
80.7
78.1
% Total Recovery
97.4
90.3
89.3
88.1
TABLE 8
TABLE 9
NaOH and Na Si0 /POLYMER COREFLOODS WITH
4
4
SILICA CONTENTS IN PRODUCED FLUIDS
WILMINGTON RANGER-ZONE CRUDE
WITH ALKALINE BEREA COREFLOODS
RUN
17
Alkali
Cone., Wt% NaZO
NaOH
0.5
18
19
ZO
NaOH
Na Si0
4
4
Na Si0
4
4
0.5
0.5
0.5
Polymer
Po1yac
X-Gum
Po1yac
X-Gum
Polymer Cone., ppm
3800
1955
3850
1985
PEAK SiO LEVELS
Z
(ppm)
ALKALI
WT% NaZQ
Na ZC0 3
0.5
70
Na C0
Z 3
1.0
100
NaOH
0.5
450
NaOH
1.0
900
PV WF Inj.
1.86
1.18
1. 96
LIZ
PV CF Inj.
1.0
1.0
1.0
1.0
So~.
0.758
0.748
0.769
0.757
Na Si0
4
4
0.5
Z140 (Z4Z5)*
Sor
0.4Z5
0.409
0.419
0.4Z0
Na Si0
4
4
1.0
5460 (4850)*
% WF Recovery
Sof
43.9
0.070
45.3
0.086
45.5
0.033
44.5
0.115
% Tertiary Recovery
83.5
79.0
9Z.1
72.6
% Total Recovery
90.8
88.5
95.7
84.8
*Injected SiO
Z
levels in parentheses.
TABLE 10
ALKALI LOSS AND BREAKTHROUGH
IN BEREA COREFLOODS
ALKALI
WT% NaZQ
MEQALKAU
LOST
ALKALI BREAKTHROl'GH
(PV AFTER INJECTION)
Na C0
Z 3
0.50
4.5
1.16
Na C0
2 3
1.0
4.1
0.86
7.5
1.20
12.9
0.82
NaOH
0.50
NaOH
1.0
)la Si0
4
4
0.50
Na Si0
4
4
1.0
10'.-----------------------------------------,
Z 9
13.
LIZ
0.71
100,-------------------,
Run 13: 0.5 wt% Na20IPolyac
o
NaOH
Run 11: Polyacrylamide
Run 5: 1.0 wt% Na 2 0
Run 4: 0.5 wt% Na20
1.0
CONCENTRATION-Wt% Na20
Fig. 1-IFT Values-Wilmington Ranger-Zone Crude with Alkalls (52" C)
2.0
3.0
4.0
PORE VOLUMES
Fig. 2 - Percent tertiary oil recovery - Na2C03 and polyacrylamide
5.0
100r'------------------------------------------------,
lOOrl-------------------------------------------------,
Run 19: 0.5 wt% Na,O/Polyac
80
Run 17: 0.5 wt% Na,O/Polyac
BOl-
~
>a:
w
>
§a: 60l-
U
w 60
a:
~
~i=
i=
a:
a:
w
I-
I!:!
~ 401-
~ 40
~
:z
::)
u
~
....I
::)
::)
:z
::)
u
~
~
20
RUn 11: Polyacrylamide
20l-
Run 11: Polyacrylamide
Run 10:
Run 8: 1.0 wt% Na,O
to wt% Na,O
Run 9: 0.5 wt% Na,O
Run 7: 0.5 wt% Na,O
1.0
2.0
3.0
4.0
0'
5.0
o
{-k<
t-1.0
2.0
3.0
4.0
PORE VOLUMES
PORE VOLUMES
Fig. 3- Percent tertiary oil recovery- NaOH and polyacrylamide
Fig. 4- Percent tertiary oil recovery- Na4Si04 and polyacrylamide
5.0