SPE 94660

Investigation of a Novel Chemical for Bullhead Water Shutoff Treatments

K. Elewaut, Delft U. of Technology; A. Stavland, RF Rogaland; A. Zaitoun, Inst. Français du Pétrole; Z. Krilov, INA; and
P.L.J. Zitha, SPE, Delft U. of Technology

Copyright 2005, Society of Petroleum Engineers

Therefore bullhead treatments were attempted with some
This paper was prepared for presentation at the SPE European Formation Damage success only in rare cases,4,7 where permeability contrast
Conference held in Scheveningen, The Netherlands, 25-27 May 2005.
between hydrocarbon and water layers was sufficiently large
This paper was selected for presentation by an SPE Program Committee following review of
information contained in a proposal submitted by the author(s). Contents of the paper, as
to ensure very limited chemical penetration in oil layers.
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to To widen the scope of water control – and more specifically
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at of bullhead treatments – a change in prevailing thinking
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
emerged in the last few years. Studies conducted in the context
for commercial purposes without the written consent of the Society of Petroleum Engineers is of the WELGEL project10 established that oil soluble
prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300
words; illustrations may not be copied. The proposal must contain conspicuous chemicals (OSC) of the alkoxy-silane type are viable bullhead
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
water control agents. Since the OSC does not form a gel in the
oil phase, it may be introduced into the formation with little
risk of damaging oil layers. Most of the OSC should be
Abstract
washed out from oil layers when production resumes. In
Traditionally bullhead water control consists of the placement
water-bearing layers, however, the chemical should form a
of water-soluble chemicals without mechanically isolating oil
sufficient amount of gel.
layers. In order to be successful, bullhead treatments should
Tetra-methyl-ortho-silicate (TMOS), with formula
block water-bearing layers without affecting the permeability
Si(OCH3)4, was found to be suitable for this purpose.10
of the oil-bearing layers too much. This paper reports an
Applications of this molecule to form silica gels (sometimes
experimental investigation of an innovative water shutoff
referred to as xeroglass) are described in detail by Brinker and
chemical deemed suitable to accomplish this task. The
Scherer.11 Plazanet and Thomere12 were first to propose a wide
chemical is soluble in oil without any reaction, and is
class of alkoxy-silane compounds, including TMOS, as sand
transferred into and reacts with water to form a stable gel.
consolidation agents. Thompson and Fogler13 reported core
Core flow experiments consisting mostly of the injection of
flow experiments and micromodel studies using TMOS and
either oil or water in cores containing the chemical dissolved
others compounds, for fluid diversion and fluid profile control
in oil were performed to investigate the ability of the gel to
purpose. Grattoni et al.14 and Karmakar et al.15 performed
reduce water permeability while maintaining oil permeability.
more extensive micro-model and test tube studies respectively,
To ensure that the results are relevant for designing field
drawing a phenomenological picture of mass transfer and
applications, the experiments were performed under
gelation process. Castelijns et al.16 reported the first results of
conditions encountered in various sandstone reservoirs in the
a more comprehensive study of the physical-chemical
North Sea, Croatia and other regions. Permeabilities and
mechanisms of OSC mass transfer and reaction in water using
temperature ranging from 100 to 300 mD and 70-900C were
NMR spectroscopy and imaging. Elewaut and Zitha17
therefore considered in this study. The influence of the other
investigated the mass transfer and gelation of OSC in
physical parameters, such as viscosity, concentration of active
Bentheim sandstone cores. Using using pressure data and X-
chemical in oil chemical, injection rate, etc., was also
ray computed tomography (CT) imaging these authors
investigated.
discussed the effects of the OSC on porosity and phase
mobilities.
Introduction
This study aims at gaining greater insight on the behavior of
High water production in oil and gas wells is a longstanding
the OSC in typical sandstone reservoir conditions. We present
problem that is becoming increasingly critical with maturing
a laboratory study where Berea sandstone cores are first
oilfields worldwide.1-3 The most widely used water control
treated with the OSC. Then the cores are flushed with either
techniques rely on hydrophilic polymers4 and crosslinked
oil or brine to check the selective reduction of brine
polymer gels.5-8 Water treatments with these chemicals are
permeability. The experiments address questions relevant for
often accomplished when oil-bearing zones can be
field application and consider permeability and temperature in
mechanically or otherwise physically isolated.9
the ranges 100 to 300 mD and 70-900C, respectively. Typical
Bullhead treatments, i.e. treatments without zonal isolation,
criteria for applying the new concept in high water cut oil
using polymers and gels did not gain wide acceptance in spite
wells are discussed conceptually.
of their potential advantages. This is due to the risk of
blocking hydrocarbon layers along with the water layers.
2 SPE 94660
Experimental
Chemical system. The OSC system used to carry out the
experiments is a commercial grade TMOS, dissolved in
different types of oil. Pure TMOS reacts in water according to
the following two-step process:
Si(OCH3)4 + 4 H2O Æ Si(OH)4 + 4CH3OH (1)
nSi(OH)4 Æ nSiO2 + 2nH2O (2)
where (1) represents hydrolisis to silicic acid and methanol
and (2) condensation-polymerization of silicic acid to silica.
Fluid flow properties are presented in Table 1. The
composition of brine is given in Table 2. Two sets of
experiments were conducted using different oils as OSC
carriers (see Table 3). The first set of experiments used low-
viscosity oil as the carrier of the OSC. But the experimental
procedure used in these experiments suffered from
defficiencies which lead to uncertainties in the results. The
second set of experiments was carried out with higher
viscosity Tellus22 oil. OSC injection rate and total injected
volumes were also varied in these experiments.
Porous media. The experiments were performed using
laminated Berea sandstone cores with length 100 mm,
diameter 25 mm, porosity 18% and pore volume 8.8 mL. The
cores were drilled from a cubic Berea block (30 cm by side) so
that laminae are oriented parallel to core axis. The cores were
used as provided, without further treatment, to represent more
realistically field conditions. Although clay content could be
as large as 7.5%, the risk of clay swelling was minimized by
using synthetic seawater brine (see composition in Table 3)
for the initial core saturation and for the injection tests. The
absolute permeabilities (kabs) were found to vary from sample
to sample in the range 100-300 mD (see Table 5). The kro/krw
ratios ranged between 2 and 4 indicating a mixed-wet to fully
water-wet system.
Set-up and coreholder. The experimental setup is shown in
Figure 1. Two double-effect P-500 Pharmacia pumps are used
to inject the fluids. The coreholder is made of stainless steel
and is formed by a casing with an internal diameter 3.0 cm and
two end pieces. Grooves carved on the inside of the two end
pieces allow for a more uniform distribution of the fluids.
Special glue resistant to high pressures and temperatures is
used to fit the core inside the stainless steel casing. The
temperature during the experiments was maintained at 70-
90ºC by placing the coreholder in an air-circulated oven. The
first experiments (cores 3-6, Table 1) were performed without
pre-heating the fluids at the core inlet. This resulted in a lower
overall temperature in the core and had adverse effects on the
experiments as will be discussed later. In the following
experiments fluids were pre-heated prior to injection using a
heat exchanger with a temperature regulation system.
The inlet and outlet pressures were monitored and recorded
using a personal computer. Two additional pressure sensors
were also connected to the core using orifices at distances over
the core length. However, due to frequent blocking with gel
the pressures measured with these sensors were unreliable and
were therefore discarded A backpressure regulator was
mounted at the core outlet, just in front of the fraction
collector, to prevent the vaporization of fluids in the core.
Vaporization was expected to occur at the temperatures used
in the experiments, due to the presence of methanol produced
by the chemical reaction.
Procedure. Table 1 summarizes the experimental conditions
which were investigated. The experiments using cores 1 and 2
were discarded due to a malfunction of the set-up: the data
collected were meaningless and will not be discussed further.
Nevertheless for consistency the original numbering of the
cores has been maintained. Note also that core 12 proved to be
only slightly affected by gelation after treatment with OSC,
and the decision was made to reuse it in a second gelation
experiment.
As already mentioned, core is initially saturated with a
synthetic brine to full water saturation (Sw = 1) and then the
absolute permeability is determined (step 1).
Next oil is injected (primary drainage) up to connate water
saturation and oil permeability at Sw = Swc is determined (step
2). Following oil injection the core is once again subjected to
brine injection (secondary imbibition) to residual oil saturation
in order to determine water permeability at Sw = 1 – Sor (step
3). Preceding OSC injection the core is then once again
brought to connate water saturation (step 4). Two main
methods of introducing the OSC (step 5 and onward) into the
core were tested:
First series of experiments: two pore volumes of the OSC
mixture are injected, followed by a two-hour shut in period
(steps 5 and 6); then either oil or brine was injected (step 7) to
determine the effect of saturation on gel formation. After this
injection phase the core is shut in for another 20 hours (step 8)
to allow gelation to continue, and finally the new fluid flow
characteristics are determined.
Second series of experiments: the initial fluid flow
characterization phase is followed by injection of four pore
volumes of the OSC mixture and shut in period of two hours
(step 5 and 6). After core shut in, brine is injected, followed by
oil or brine injection.
Results and Discussion
Injection of 2 PV of IsoparH /OSC at low injection rate:
The first series of experiments was carried out with IsoparH,
low viscosity oil, as the OSC carrier. Injection rates for the
oil/OSC mixtures were low (0.2 mL per minute).
The three cores (3 to 5) present low residual resistance factors
both for oil and water indicating that there is little or no
modification of flow properties in the cores. This suggested
little gel formation (if any), and raised doubts whether mass
transfer of TMOS from the oil to the water phase had occurred
at all or in sufficient amount.
The low injection rate, high temperature, long shut in period (2
hours followed by another 20 hours) are believed to be
sufficient to promote mass transfer and gelation. Several
effects were therefore hypothesized to explain the lack of
permeability modification. First, the composition of IsoparH
could be such that it has a great affinity with TMOS and
SPE 94660
therefore mass transfer is too slow to build up significant
amounts of TMOS in the water phase, over the exposure times
used in our experiments. Secondly, the chemical properties of
TMOS and IsoparH could be such that no good mixing occurs.
This would lead to segregation of the two components and
therefore to settling of heavier TMOS at the bottom of the
vessel. The amount of TMOS introduced into the core could
be significantly lower than expected. Finally, in absence of a
pre-heating system at the core inlet, the temperature of the
fluids in the core could be much lower than the nominal value
of 700C this would delay mass transfer and gelation
considerably.
Table 4 shows that oil permeabilities before TMOS injection
are similar to those calculated during TMOS injection based
on the viscosity of pure IsoparH. This indicates that the
viscosity of the injected oil/TMOS mixture differs little from
pure oil viscosity and suggests that oil/TMOS mixture actually
contained significantly less TMOS than expected. This
supports the second hypothesis.
Nevertheless, the overall experimental procedure was
improved not only by using an oil with greater affinity to the
OSC, but also by pre-heating the fluids at the core inlet to
address the third hypothesis.
Injection of 2 PV of Tellus22/OSC at low injection rate:
For the next set of experiments (cores 6 to 8), high-viscosity
oil Tellus22 was used as a carrier for TMOS. Oil/TMOS
mixture was injected at the same low flow rate (0.2 mL per
minute).
The TMOS injection stage yields very different pressure
drop profiles from the previous series of experiments, and the
fact that pressure drop behavior is the same in cores 6 and 8
indicates reproducibility (Figure 2).
There are several possible explanations for the increase in
pressure drop: First of all the tube volume before the core
could be so large that the oil/TMOS mixture does not
penetrate the core for the first 20 minutes. If such, the volume
of tubing in front of the core should be about 4 mL. However,
the total tubing volume in the entire setup is about 1.4 mL
only. This leads us to believe that it is unlikely that the
segment of tubing before the core inlet is responsible for the
increase in pressure drop. A more plausible explanation is the
change of viscosity due to mass transfer.
TMOS has a significantly lower viscosity than Tellus22,
therefore a mixture of TMOS and oil will have a lower
viscosity than oil alone. The viscosities determined from
cores 6 and 8 are respectively 8.4 and 5.7 cP, both values
being greater than the viscosity of pure Tellus22 (4.6 cP at
90ºC). The data seem to point out an extensive mass transfer.
Once mass transfer occurs, the oil viscosity will increase.
Mass transfer also causes an increase in water saturation and
consequently a decrease in oil saturation. These effects
contribute to pressure drop increase. The effect of gelation on
fluid flow in these experiments is significant. Both water and
oil permeabilities were reduced, showing that TMOS did act
as an overall permeability reducer with no selectivity. After
the experiments the core holder was dismantled and extensive
gelation was observed at the core ends. The conclusion from
this observation is that premature OSC gelation occured with
water trapped in the gaps between the end pieces and the core.
3
Injection of 2 PV of Tellus22/OSC at high injection rates:
In the following procedures another strategy was chosen for
gel placement. The oil/TMOS mixture is introduced at high
flow rate (6 mL/min, which is 30 times higher than the flow
rate used in all previous experiments). Core 9 is subjected to
oil flow following shut in, whereas in cores 10, 11 and 12,
brine is injected following shut in. The experiments using core
11 ended with the initial brine backflush (step 7). The same
was done for core 12. The low RRF obtained through gel
treatment led however to the decision to reuse core 12 in the
following series of experiments.
The 30-fold increase in injection rate during introduction of
TMOS to the core has a strong effect on the shape of the
pressure drop curves (Figure 3), which deviate significantly
from those seen in the previous set of experiments (Figure 2).
Low pressure drop indicates no premature gelling during
placement of TMOS in the core. The decline in pressure drop
at the early stage is explained by the replacement of Tellus22
by the lower-viscosity Tellus22/TMOS mixture.
Injection of 4 PV of Tellus22/OSC injection at high flow
rates:
Limited RRF values in the previous experiments led us to
inject four pore volumes (instead of two) of oil/TMOS mixture
in cores 12 to 14.
The pressure drop obtained in cores 12(2), 13 and 14 exhibit
a characteristic trend that was not observed in the previous
experiments, especially after one pore volume. In the first pore
volume the differential pressures decrease until they reach a
minimum value and then increase. This behavior can be
explained by a greater mass transfer of TMOS into the water
phase and its subsequent hydrolysis. The more rapid increase
of the pressure drop in core 12(2) is probably due to the lower
flow rate used in this experiment (2 mL/min instead of 6
mL/min in cores 13 and 14). This gives a longer residence
time to the oil/TMOS mixture and favors mass transfer and
hydrolysis.
The permeability reduction resulting in core 13 and 14 are
by far the most promising. In the first back-flush, core 13
gives a 13-fold reduction of water permeability while core 14
gives a permeability reduction to oil by a factor 2.3. Similar
results have been observed elsewhere18.
Field application recommendations for the OSC
The above results indicate that the OSC reduces selectively
water permeability without affecting oil permeability too
much as requested for bullhead water control treatments. The
experiments also suggest that the OSC can be used a total
blocking agent, with advantages such as the low viscosity
during injection.
Laboratory studies to optimize chemical composition,
treatment parameters and schedule under the specific reservoir
conditions are required. Parameters controlling the effect of
gelation on fluid flow include: oil type used to inject the OSC,
temperature of the system, injection rate of the oil/OSC
mixture, and water saturation of the rock. TMOS placement
4 SPE 94660
conditions have to be carefully designed and adapted to
reservoir and well conditions.
The optimal injection rate for the OSC/oil mixture will both
depend on the oil properties and the reservoir temperature.
Serious consideration needs to be paid to effects of brine
properties such as the salinity and pH. In the past pH was
found to have a significant impact on the morphology of the
gel, where at lower pH values the gel will form a relatively
simple and open network. Higher pH values however will
result in much a denser gel network. The effect of TMOS gel
morphology on fluid flow in porous media needs to be
investigated in more detail.
Conclusions
• The placement of an OSC-based gel system is a sensitive
system depending strongly on the injection rates and
post-placement procedure.
• Our experiments show that, due to the risk of premature
gelling, low flow rates maybe detrimental to TMOS
placement.
• Optimizing the conditions of TMOS placement strategy
gave high permeability reductions to water (RRFw=13)
coupled with low permeability reduction to oil
(RRFo=2.5), as required of a good selective water shutoff
system.
Acknowledements
This work was done as part of theWELGEL II JIP project co-
funded by: ConocoPhillips, ChevronTexaco, Gaz de France,
INA, Shell, Statoil, and PDVSA/INTEVEP. We gratefully
acknowledge Rien Faber (Shell), Birgitte Schilling (Statoil),
Steve Cheung (Chevron Texaco) and Trond Rolfsvaag
(ConocoPhillips) and Thierry Pichery (Gaz de France) for
invaluable suggestions and comments.
References
1. Bailey, B, Crabtree, M., Tyrie, J., Elphick, J., Kuchuk, F.,
Romano, C., Roodhart, L., Water Control, Oilfield Review,
Spring, 30-51 (2000)
2. Lockhart, T.P. and Borgarello, E., Colloid Science in Oil
and Gas Production and Transportation, Prog. Colloid and
Polymer Sci., 109, 49-59 (1998)
3. White, J. L., Goddard, J. E. and Phillips, H. M., Use of
Polymers to Control Water Production in Oil Wells, JPT, 143-
150, February (1973)
4. Zaitoun, A., Kohler, N. and Guerrini, Y., Improved
Polyacrylamide Treatments for Water Control in Producing
Wells, JPT, July, 862-867(1991)
5. Eggert Jr., R.W., Willhite, G.P. and Green D.W.,
Experimental Measurements of the Persistence of the
Permeability Reduction in Porous Media treated with
Xanthan/Cr(III) Gel Systems, Soc. Petrol. Eng. Reservoir Eng.
J., April, 29-35 (1992)
6. Seright, R.S., Reduction of Gas and Water Permeabilities
using Gels, Soc. Petrol. Eng. Production and Facilities, May,
103-108 (1995)
7. Zaitoun, A. and Kohler, N., Modification of Water/Oil and
Water/Gas Relative Permeabilities after Polymer Treatment of
Oil and Gas Wells, In Situ, 13 (1&2), 55-77 (1989)
8. Zitha, P. L. J., Botermans C. W., v.d. Hoek, J. and
Vermolen, F. J., Control of Flow through Porous Media using
Polymer Gels, J. of Applied Physics, 92 (2), 1143-1153 (2002)
9. Seright, R.S., Placement of gels to modify injection profiles,
paper SPE/DOE 17332 presented at the 1988 SPE/DOE
Enhanced Oil Recovery Symposium, Tulsa, OK (1988).
10. Zitha, P.L.J. (Ed.), Well Treatment and Water Shutoff by
Polymer Gels, Delft University Press, Delft, The Netherlands
(2000)
11. Brinker, C.J., Scherer, G.W., Sol-Gel Science, Academic
Press, San Diego CA (1990)
12. Plazanet, V., Thomere, R., patent FR2624198 (1989)
13. Thompson, K.E. and Fogler, H.S., Pore-Level Mechanisms
for Altering Multiphase Permeability using Gels, SPEJ, 2,
350-362 (1997)
14. Grattoni, C.A., Jing, X., Zimmerman, R.W., Wettability
alteration by ageing of a gel placed within a porous medium,
Presented at the 6th International Symposium on Evaluation
of Reservoir Wettability and its Effect on Oil Recovery, 27-28
September 2000, Socorro, New Mexico, USA.
15. Karmakar, G.P., Grattoni, C.A., Zimmerman, R.W.,
Relative permeability modification using an oil-soluble gelant
to control water production, presented at the 2002 SPE Annual
Technical Conference and Exhibition held in San Antonio,
Texas (2002)
16. Castelijns, H.J., Pel, L, Huinink, H.P., Zitha, P.L.J.,
Investigation of Reactive Transport Phenomena for
Modification of two-phase Flow using NMR, Presented at the
SPE 6th European Formation Damage Conference,
Scheveningen, The Netherlands (2005)
17. Elewaut, K., Zitha, P.L.J., Water Shutoff using Oil-Soluble
Chemicals: an X-ray CT Study, presented at 13th European
Symposium on IOR, Budapest, Hungary (2005)
18. Stavland, A., Goonewardene, P.A., Vatne, K.O.
Experimental testing of an oil soluble gel system for water
shutoff, Report RF (2003)
SPE 94660 5

Table 1. Sequence and methodology of experiments

Steps IsoparH Tellus22

3,4,5 6, 8 and 12(1) 7, 9 and 11 12 (2) 13 14

1 Brine injection Brine injection Brine injection Brine injection Brine injection
2 oil injection oil injection oil injection Oil injection Oil injection
3 brine injection brine injection brine injection Brine injection Brine injection
4 oil injection oil injection oil injection Oil injection Oil injection
5 2 PV 20m% TMOS 2 PV 20m% TMOS 2 PV 20m% TMOS 4 PV 20wt% TMOS 4 PV 20wt% TMOS 4 PV 20wt% TMOS
6 2 hr shutin 2 hr shutin 2 hr shutin 2 hr shutin 2 hr shutin 2 hr shutin
7 Brine backflush Brine backflush Oil backflush Brine backflush Brine backflush Oil backflush
8 ca. 20 hr. shutin ca. 20 hr. shutin ca. 20 hr. shutin Oil backflush Oil backflush Brine backflush
9 Brine backflush Brine backflush Oil backflush
10 Oil backflush Oil backflush (not 12) Brine backflush

Table 2. Properties of fluids used in experiments

Viscosity (cP) Density (g/mL)
at 20ºC at 70ºC at 90ºC
Brine 1.26 0.604 0.521 0.9908
IsoparH 1.09 N/A 0.39 0.7294
Tellus22 34.5 6.68 4.6 0.8654

Table 3. Composition of synthetic brine Table 4.Calculated oil permeabilities

g/L ko1 ko2 k o*
NaCl 24.05 core 3 160 176 184
MgCl2.6H2O 10.77 core 4 138 142 153
CaCl2.2H2O 1.64 core 5 174 192 145

Table 5. Calculated permeabilities for each core

Pre-gel Post-gel
Cores Backflush kabs ko kw ko kw kw ko RRFo RRFw RRFo RRFw
3 190 ± 22 160 ± 33 41 ± 5 176 ± 26 28 ± 5 29 ± 1 122 ± 32 1.44 1.41 1.44 1.41
4 114 ± 15 138 ± 20 38 ± 2 142 ± 16 15 ± 4 28 ± 3 118 ± 34 1.20 1.36 1.20 1.36
5 170 ± 20 174 ± 30 22 ± 1 192 ± 43 27 ± 3 23 ± 1 196 ± 22 1.75 0.96 0.98 0.96
6 183 ± 21 186 ± 5 68 ± 3 147 ± 3 7±2 9±1 17 ± 2 8.65 7.56 8.65 7.56
Brine

7 127 ± 10 133 ± 3 43 ± 2 87 ± 1 7 ± 0.4 8 ± 0.7 8 ± 0.2 11.45 8.60 10.88 5.38

8 152 ± 33 105 ± 6 28 ± 1 67 ± 2 5±1 6±1 13 ± 1 5.15 4.67 5.15 4.67
10 309 ± 42 175 ± 5 61 ± 4 131 ± 10 41 ± 9 42 ± 6 72 ± 8 1.80 1.50 1.82 1.45
12(1) 220 ± 24 124 ± 1 34 ± 3 54 ± 5 17 ± 2 18 ± 2 33 ± 1 - - 1.64 1.89
12(2) - - - - - 3 ± 0,1 4 ± 0,1 6.60 4.50 8.25 6.00
13 224 ± 17 143 ± 3 57 ± 2 81 ± 2 - 4 ± 0,2 32 ± 2 2.50 13.00 2.53 14.25
Cores Backflush kabs ko kw ko ko ko kw RRFo RRFw RRFo RRFw
9 141 ± 11 89 ± 6 45 ± 5 40 ± 1 26 ± 2 25 ± 1 16 ± 2 1.60 2.80 1.60 2.81
Oil

11 260 ± 33 157 ± 7 50 ± 6 90 ± 2 29 ± 2 - - - 1.70 - -

14 271 ± 47 171 ± 7 61 ± 1 96 ± 3 - 41 ± 0,2 26 ± 0,1 2.30 2.30 2.34 2.35
6 SPE 94660

Figure 1. Schematic of the experimental setup.

1.4
Core 6
6 Core 8 Core9
Core10
Core11
1.2 Core12(1)
5
Core12(2)
normalized pressure drop

Core13
normalized pressure drop

Core14

4
1

3
0.8

0.6
1

0.4
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 0.5 1 1.5 2 2.5 3 3.5 4
PV's PV

Figure 2. Normalized pressure drop during injection of the oil/TMOS
mixture at low flow rates. The curves are normalized to the initial
pressure drop values. The pressure is initially stable, and then increases.
Figure 3. Normalized pressure drop during injection of the oil/TMOS
mixture at high flow rates. The curves are normalized to the initial
pressure drop values. The pressure drop decreases first to a value
representative of the lower-viscosity oil/TMOS mixture, and then rises
again as mass transfer causes the oil phase to become more viscous and
the water saturation to increase.