SPE 93674
An Experimental Investigation of Reaction-Transport Phenomena During CO2 Injection
P. Egermann, B. Bazin, and O. Vizika, Inst. Français du Pétrole
Copyright 2005, Society of Petroleum Engineers Inc.
This paper was prepared for presentation at the 14th SPE Middle East Oil & Gas Show and
Conference held in Bahrain International Exhibition Centre, Bahrain, 12–15 March 2005.
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
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
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
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.
Abstract
The injection of carbon dioxide in depleted petroleum
reservoirs or in aquifers is a promising way to cope with the
short-medium term issue of greenhouse gas emissions
mitigation. Several coupled physical and chemical processes
may occur during the injection depending on time and location
within the reservoir. Far field regions are facing long term
reaction in a situation where flow of gas and water at a
reduced rate may induce near fluid-rock equilibrium. In
contrast, near well-bore regions are subjected mainly to gas at
a high flow rate where dissolution/reprecipitation phenomena
may increase/decrease drastically the injectivity.
The purpose of this study is to investigate experimentally
the various situations in representative reservoir conditions
with the objective of achieving knowledge and data for future
physical and numerical modeling and reservoir numerical
simulations of CO2 re-injection. Experiments consist in the co-
injection of CO2 and brine in carbonate cores (limestone). The
temperature and pressure conditions are such that the CO2 is in
supercritical state. Results show that the flow rate and the
composition of the fluids initially present in the core play a
major role in the fluid-rock interaction leading to various non-
uniform dissolution facies and in some cases to re-
precipitation and permeability reduction. These phenomena
have been observed and quantified using various non
destructive techniques (NMR, CT-scanner) and chemical
analyses of the producing fluids, leading to a comprehensive
understanding of the coupled mechanisms taking place.
Introduction
This paper is the first step in a laboratory investigation of CO2
sequestration in carbonate reservoirs. It aims at defining
qualitatively the consequences of dissolution/precipitation
mechanisms on the extent of permeability variations. The
coupling of physical and chemical processes occurs in various
situations, in the near well bore or far in the reservoir and has
consequences on the long term viability of CO2 re-injection in
depleted petroleum reservoirs or in aquifers.
Whether the injection of CO2 is leading to precipitation or to
dissolution, the mechanisms that impact the permeability are
still debated. Most of the literature on CO2 injection in the
context of oil recovery deals with the WAG process. A survey
of the results on field applications showed that the injectivity
of CO2 was the first concern of the operators 1. Based on fluid
properties, it is expected a greater CO2 injectivity in
comparison to water injectivity. However, in practice,
abnormalities 2,3,4 were commonly encountered. Well-bore
impairment requiring injection surveillance 5, particularly in
open-hole wells 6 is mentioned.
Laboratory experiments were planned in an attempt to identify
possible mechanisms responsible of injectivity losses 7,8. The
WAG sequence as well as the presence of residual, i.e.
immobile, phases have been pointed out. Two-phase flow and
relative permeability effects were investigated without
considering chemical mechanisms 9,10. More recently, with the
emerging CO2 sequestration technology, studies dealing with
CO2 injection have pointed out the mechanisms of
dissolution/precipitation as controlling well injectivities.
This study differs from others on several points, which are, in
our opinion, basically needed for understanding CO2 flow in
the reservoir. First, temperature and pressure are fundamental
parameters since they affect the phase diagram. Temperature is
also an important factor with regards to reaction kinetics. We
chose to work in a situation where the CO2 is in supercritical
state. Secondly, we want to account for two important
parameters: (1) the flow regime, which varies significantly
around the well and can strongly impact the interactions
between the circulating fluids and the rock11 and (2) the
multiphase flow of the CO2 phase and the water, which occurs
in the field due to the fractional flow. In this later case, the
dissolution products are transported over a certain distance and
precipitation may occur in regions non affected by
dissolution12.
2 SPE 93674
The experiments were conducted with limestone. Limestone is
very reactive with acids. Dissolution figures obtained from
CO2 injection are presenting strong analogies with dissolution
patterns named wormholes occurring when acid is injected in
the wells as a stimulation treatment. Acid injection in
carbonates was investigated in IFP in various conditions and
dissolution patterns were classified according to the
experimental conditions 13,14. We used these results as a
guideline for the definition of the experiments presented here.
The paper is organized as follows. The first section is a short
literature background on dissolution/precipitation mechanisms
in carbonates. We reported experimental results obtained by
previous investigators in relation with CO2 injection in various
conditions of temperature and pressure. We reviewed
dissolution mechanisms occurring in carbonates with HCl.
Although the acidity of the solution is very high in the later
case, some analogies could be raised. The second part
described our experimental device and the conditions of the
experiments. The study is turned to the application, and
conditions are representative of what can be found in the field.
We investigated various brine compositions to evaluate the
onset of the re-precipitation mechanisms and their effect on
the permeability. The third part is devoted to the presentation
of results and to a discussion in light of the preceding analysis
of dissolution/precipitation mechanisms. It gives indications
for future work.
Background on dissolution/precipitation
mechanisms in carbonates.
CO2 dissolves in water forming H2CO3, which then dissociates
to give an acid that reacts with the calcite present as carbonate.
The chemical reactions are the following:
CO 2 + H 2 O ⇔ H 2 CO 3 ⇔ H + + HCO 3−
H + + HCO 3− + CaCO 3 ⇔ Ca 2+ + 2HCO3 −
This set of reactions indicates clearly that acid is formed by
contact of CO2 with water. As long as the CO2 gas is injected
and is contacting the water phase, acid will form. In this
situation, acid concentration is "unlimited": even though acid
is spent by reaction, complete spending will never occur and
there is in the system an almost unlimited source of acidity.
At this step, it is important to note that the dissolution
potential of the system with two phases, i.e. a water phase and
a CO2 phase flowing simultaneously, is very different from the
situation where a water phase, saturated in CO2, is the only
mobile phase. In the first case, as seen before, the dissolution
potential is unlimited whereas in the second case, the acidity is
removed progressively as the dissolution proceeds. Studies
dealing with the injection of water saturated with CO2 are
investigating only one face of the problem.
Bazin and co-workers13,14 have conducted acidizing
experiments using both limestone and dolomite and acid fluids
with a wide concentration range. Acid dissolution of
carbonates is controlled by acid mass transfer kinetics and is
highly flowrate dependent. This coupling of transport and
reaction is a characteristic of carbonate dissolution. It is an
unstable dissolution process leading to different dissolution
regimes, with the so-called wormhole pattern being the most
characteristic. Dissolution figures are controlled by the acid
injection flow rate. Three dissolution regimes have been
identified. At very low flowrate, the acid is spent locally and
the dissolution pattern is compact. When the flow rate
increases, acid is not completely spent locally and remaining
acidity is transported farther. The dissolution regime is called
the wormholing regime. At very high flowrate, the high acid
filtration from the wormhole walls makes the dissolution
pattern very ramified. The dissolution pattern is named
"ramified wormhole" or uniform dissolution.
Reactive transport phenomena during CO2 injection have been
studied both on sandstones 15 and carbonates 16,17. The
experimental results showed that either a permeability
improvement or a permeability impairment can be obtained 3.
These results demonstrate that the injectivity issue during CO2
injection is very case dependent because it is related to the
rock fabric, the brine composition and also the thermodynamic
conditions. Therefore, it is of primary importance to work
under representative conditions (fluids, rock and
pressure/temperature) in order to derive relevant
recommendations for field applications. Available
experimental data sets have been obtained either through
carbonated brine injection or co-injection of CO2 and brine.
Dissolution patterns have been observed in most of these
experiments and also in the field18. Surprisingly, little attention
has being paid to date on the influence of the hydrodynamic
regime on these mechanisms in spite this is a key parameter in
the near wellbore region where the fluid velocity varies a lot
over a short distance 11. This coupling between transport and
reaction is prone to generate specific porosity/permeability
relationships according to the flow regime. These relations are
very important to introduce in numerical model in order to
properly reproduce the pressure field around the well and the
stress variations that can be detrimental for the wellbore
integrity. This work is an experimental contribution to better
understand these mechanisms in the context of the CO2
geological storage.
Reservoir condition experiments
Core selection and preparation
All the experiments were conducted on limestone samples
plugged in the same quarry rock and in neighbor locations.
The typical size of the plugs was 5 cm in diameter and 7 cm in
length. The limestone was selected due to its good level of
homogeneity at the core scale, which is highly desirable when
permeability evolutions have to be compared from one test to
another. The corresponding mercury porosimetry curve and
the NMR T2 response are provided in Figure 1 and Figure 2.
The shape of the mercury porosimetry curve indicates that the
pore throat distribution can be considered as monomodal. The
NMR T2 response also indicates only one main peak located
at 62 ms suggesting that the pore size distribution exhibit also
SPE 93674
a monomodal behavior. The average petrophysical properties
measured on the plugs were found very homogeneous
whatever the plug used (23% porosity and 3-5 md
permeability). The porosity was determined by the weighting
technique and also checked using the amplitude of the NMR
response in order to derive the brine volume in the saturated
samples.
Two different compositions were used to prepare the brine
injected in the experiments (Table 1). The first one
corresponds to seawater. The second one has exactly the same
composition excepted that the Na2SO4 has been removed from
the brine.
Experimental apparatus
The set-up is designed to work under representative
reservoir conditions up to 300 bar and 120°C (Figure 3). It is
mainly composed of a Hassler type core holder and a pumping
system that allows injecting separately or simultaneously CO2
and brine. Because of the corrosive nature of the mixture
injected, the end-pieces of the core holder and the lines are
built in Hastelloy to insure the integrity of the system over
long periods of injection. It enables also to work with high
salinity brine, which is very common in aquifers or
hydrocarbon reservoirs. The brine is injected directly through
an Hastelloy high-pressure pump, whereas the CO2 is pushed
by water injection though a buffer cell. Another buffer cell
equipped with a piston is also available to inject live oil into
the core and to conduct experiments representative of the
initial saturation state in depleted hydrocarbon reservoir. The
core holder and the buffer are located in an oven in order to
work under controlled temperature. Coiled tubing of sufficient
length insures the heating of the fluid injected from outside the
oven. The whole system can be set under pressure using a
back pressure valve. All the data are automatically recorded on
a PC and a sampling device enables to collect fluid fraction
downstream the back pressure valve in order to conduct
chemical analysis of the produced water as a function of time.
Experimental procedures
The procedure detailed hereafter has been defined on the
one hand to inject both CO2 and brine in a controlled manner
and on the other hand to identify the rock/fluid interaction
processes that impact the petrophysical properties using
different non destructive techniques.
Injection control
First of all, a special attention was paid on the inlet core
holder end-piece because it controls the uniformity of the
injection within the whole core section. The first tests were
performed with a conventional design based on uniformly
distributed grooves. The results showed that the interactions
between the fluids and the rock occurred preferentially in the
center of the core section where the fluid injection was
performed in spite of the groves. Therefore, an inlet end-piece
built with a conical shape was preferred, which leads to
uniform injection and associated well-distributed dissolution
patterns over the whole inlet section of the core.
3
When dealing with hydrodynamic parameters, it is also
very important to control the injection rate. The confining
pressure is insured by an elastomer sleeve, which is very
permeable to CO2. When CO2 is flowing in direct contact with
the sleeve, one part of the CO2 dissolves in the elastomer,
which makes the effective CO2 injection rate decreasing along
the core. Consequently, all the plugs were prepared with a non
permeable, temperature resistant lateral coating to prevent the
CO2 from being in direct contact with the sleeve.
Characterization process
A careful characterization process of the plugs was carried
on before and after the coinjection experiment to properly
quantify the impact of the rock/fluid interactions on the
petrophysical properties. Non destructive techniques were
preferred in order to keep the integrity of the sample for
further analysis of the pore structure evolution. The weighting
technique enabled to derive porosity values. The NMR
technique provided a check of the porosity measurement but
also insight about the evolution of the V/S ratio in the pore
structure. The CT scanner was used to explore the pattern of
the dissolution inside the samples.
Injection conditions
All the experiments were conducted by coinjecting both
the CO2 and the brine at 100 bar and 90°C with a flow rate
ratio equal to unity. We explored both the impact of the total
flowrate and the brine composition on the evolution of the
petrophysical parameters. The total flowrate was tested in a
range between 2 cm3/hr and 500 cm3/hr leading to
representative conditions of the far-field reservoir and the
near-wellbore region. The same cumulated volume of fluids
was injected in all the experiments, which represents around
30 PV (Pore Volume) meaning that the samples have been
flushed significantly.
Results
Visual observations
The common feature in all the experiments is that
dissolution was observed at the inlet face. Nevertheless, the
shape and the distribution of these dissolution patterns
appeared to be extremely dependent on the hydrodynamic
regime (total flowrate) but also on the fluid system used (brine
composition). Figure 4 (right) shows that high flowrates favor
wormhole dissolution patterns (numerous pinholes well
distributed on the injection face), whereas low flowrates (left)
drive to compact dissolution patterns (limited number of big
holes). These results are in line with previous works on
acidizing studies 14. These results also demonstrate that the
pore structure evolution is very dependent on the flow regime.
Therefore, a specific porosity/permeability relationship has to
be introduced in the numerical models as a function of the
flow regime to get relevant results.
Figure 5 shows the lateral extension of the wormholes at
high rate with two different brines. The wormhole extension
seems to be favored when sulfates have been removed from
4 SPE 93674
the brine meaning that a precipitation of insoluble salts
probably occurred with the SO42- containing seawater brine.
The porosity profile evolutions are plotted in Figure 6 for
several injection rates. The CT-scanner measurements enable
to evaluate the lateral extension of the dissolution within the
sample but also the amplitude of the porosity variation as a
function of the distance. From the measured data, it is clear
that both of these parameters are strongly related to the
injection rate. The higher the rate, the deeper the dissolution
patterns. The lower the rate, the higher the amplitude of the
porosity variation at the inlet face. These results are in line
with previous works on acidizing. When the injection rate is
high, the wormhole regime is favored, which leads to
numerous, deep dissolution figures. When the injection rate is
low, the compact dissolution process is favored, which
conducts to massive and large dissolution at the inlet face.
The spatial distribution of the dissolution pattern can also
be obtained within the sample through a 3D reconstruction of
the CT-scanner data. An example is provided in Figure 7 for
the case of high injection rate.
Chemical analysis of the produced fluids
The analysis of the produced water is very usefull to
understand the chemical reactions that are taking place during
the experiment. Both the Ca2+ and the SO42- concentrations
were recorded as well as the pH evolution as a function of
time. At low flowrate (Figure 8), the Ca2+ concentration
continuously increases during the experiment from its initial
value around 400 ppm up to 1100 ppm at the end of the
injection. This variation is directly related to the dissolution
patterns of calcite that are observed at the injection face of the
core. The trend is completely different for the SO42-
concentration, which decreases continuously from its initial
value around 2850 ppm down to 2300 ppm. This variation
suggests that some precipitation of anhydrite occurred during
the experiment due to the solubilization of the Ca2+ in presence
of SO42-. The pH follows a decreasing trend, starting from 8.4
down to 7.6. At high flowrate (Figure 9), the general trends
are exactly the same but it is worth mentioning that the curves
do not stabilize at the same level. The Ca2+ concentration
increases up to 1500 ppm (comparing to 1100 ppm) and the
SO42- concentration down to 2700 ppm only (comparing to
2300 ppm). The pH also decreases down to 7.3 comparing to
7.6.
This analysis enables to identify two main phenomena that
have a strong influence on the petrophysical parameters to
measure. First of all, it is confirmed that anhydrite precipitated
during the experiments, when SO42- was present in the brine.
This precipitation process can impact the permeability
drastically by clogging the pore throats. Secondly, the amount
of the precipitated anhydrite strongly depends on the
hydrodynamic regime. When the flowrate is small, the
residence time of the fluids into the porous medium is higher,
which favors the completion of the chemical reactions leading
to more anhydrite precipitation. Therefore, these
measurements demonstrate the importance of the coupling
effect between the geochemical reaction and the flow regime.
Petrophysical characterization
The evolution of the normalized pressure drop as a
function of time showed also a specific trend depending on the
presence of SO42- in the injected brine (Figure 10 and Figure
11). All the pressure drop curves have been normalized by the
pressure drop generated by the fully brine saturated sample for
the same total injection rate in order to plot comparable curves
from one test to another.
When seawater brine has been used, the pressure drop
systematically increased suggesting a continuous plugging of
the porous medium (Figure 11). These observations are in line
with the chemical analysis, which proved the precipitation of
anhydrite. When SO42- was removed, the normalized pressure
drop continuously decreased over a time period corresponding
approximately to 15 PV injected suggesting permeability
improvement (Figure 10). On the same figure we can also
distinguish a sudden, limited in time, increase of the pressure
drop (around 20 PV). It is suspected that this effect results
from solid particle displacement mobilized by the fluids,
which plugged temporarily the porous medium. At the end of
the experiment (30 PV), the pressure drop decreased again and
stabilized at a level consistent with the previous decreasing
trend.
These observations are consistent with the brine
permeability measurements that have been conducted at the
end of the experiment when all the CO2 (gas phase and from
solution) has been removed miscibly through fresh brine
injection. When SO42- was present in the brine, the lower the
rate the higher the permeability impairment. This is also
consistent with the chemical analysis results (Figure 12). The
experiment conducted at high rate with no SO42- in the brine is
the only one that exhibits a permeability improvement
meaning that only dissolution occurred in this case.
The NMR T2 measurements that have been performed
after the pore structure has been modified by the CO2-brine
injection showed a systematic trend when compared with the
initial T2 distribution. All the T2 responses were shifted to
lower values meaning that the ratio V/S decreased
(Volume/Surface). Because the macroscopic porosity
variations were found to be very small, these measurements
result from a surface increase of the pore structure due to the
coinjection experiment. This effect was found to be more
pronounced at low injection rate (T2 divided by a factor two:
Figure 13). It is certainly associated with an increase of the
specific surface due to the dissolution mode or/and the
precipitation process, which is favored at low flowrate.
Discussion
Geochemical calculations
Calculations were run with the geochemical simulator
PHREEQ. They aimed at evaluating the
precipitation/dissolution mechanisms occurring during the
SPE 93674
CO2 flow. Two equilibrium situations in a closed system were
investigated: (1) an equilibrium of CO2 with seawater
corresponding to the injected solution; (2) an equilibrium with
CO2, seawater and calcite representing the ultimate situation,
i.e. at equilibrium. The composition of the solution and the
saturation indices are reported in Table 2. Calculations were
run at 100 bars and 90°C.
The equilibrium of CO2 and seawater resulted in a solution
pH of 3.2 (Table 2 - Step 1). Saturation indices are negative,
indicating a dissolution potential for calcite. The equilibrium
with CO2, seawater and calcite shows a sur-saturation with
respect to calcium sulfate (Table 2 - Step 2). The calcium
concentration increased to 1,610 mg/l due to the dissolution of
calcite. The brine becomes oversaturated with respect to
calcium sulfate with a saturation ratio of 0.20 (in log units).
Precipitation of calcium sulfate is expected. The remaining
calcium and sulfate concentrations are of 1,428 and 2,371 mg/l
respectively. These results are very comparable to the results
obtained at low flow rate where calcium and sulfate
concentrations were of 1,200 and 2,300 mg/l respectively. It is
concluded that calcite dissolution resulted in an increase of the
calcium concentration in seawater and is followed by calcium
sulfate precipitation.
This analysis is also very consistent with the measured
values of pH performed under high and low injection rates.
The pH value was found to be lower (7.3) under high injection
rate meaning that the producing fluids have a higher potential
of interaction with the rock comparing to the low injection rate
case (pH equal to 7.6). It confirms that high injection rate
favors out of equilibrium mechanisms of interaction between
the fluids injected and the rock.
Importance of the multiphase flow
Some experimental observations like the abnormal
pressure drop response, obtained under high injection rate,
strongly suggest that solid particles displacement can occur.
These phenomena can lead to severe permeability impairment
and is certainly activated both by the fragility of the pore
structure consecutively to the dissolution process and by the
simultaneous flow of CO2 and brine, which contributes to
generate higher pressure gradient for a given flowrate.
The simultaneous flow of CO2 and brine is also important
to consider because it limits the access of the reactive brine to
a fraction of the most accessible pore space. Gas CO2 phase,
being non-wetting, cannot access small pores with high
capillary pressure threshold. Therefore, pore structure
modifications by dissolution or precipitation, and the
associated permeability evolutions, occur in a selective way.
Implications for the well injectivity
The results of this experimental study demonstrate that the
injectivity issue is very complex because it results from a
combination between many parameters (the rock fabric, the
fluid compositions, the thermodynamic conditions, the flow
5
regime). The analysis of the experimental data has confirmed
that both precipitation and dissolution can take place in the
core during a given experiment leading to opposite effects on
the permeability evolution. It is interesting to mention that
high rates tend to limit the permeability reduction due to
precipitation, because the residence of the fluids within the
pore space is shorter in this case. It makes the system be far
from the geochemical equilibrium leading to lower amount of
precipitated calcium sulfate. From a practical point of view, it
suggests that severe permeability impairment in the near
wellbore can be avoided in spite of unfavorable geochemical
conditions if the injection rate is high enough to displace the
equilibrium area of precipitation far from the well.
These results also demonstrate that the pore structure
evolution is very dependent on the flow regime. Therefore, a
specific porosity/permeability relationship has to be
introduced in the numerical model as a function of the flow
regime to get relevant results. This is particularly important to
properly reproduce the pressure field around the well, where
the flow rate varies significantly and can induce severe stress
variations that can be detrimental to the wellbore integrity.
Conclusions
An experimental study of CO2 injection in carbonate cores
(limestone) under reservoir conditions has been presented. The
objective was to investigate experimentally the various
situations that may occur during CO2 injection in underground
formations for geological storage purposes. CO2 was in
supercritical state under the applied pressure and temperature
conditions. Various brine compositions have been investigated
to evaluate the onset of dissolution/re-precipitation
mechanisms and their effect on the permeability evolution.
These phenomena have been quantified using various non-
destructive observation techniques for the cores (NMR, CT-
scanner) and chemical analyses of the effluents.
It has been demonstrated that both the flow rate and the
composition of the fluids initially present in the core play a
major role in the rock-fluid interactions.
Dissolution patterns, comparable to the wormholes during
acidizing processes, have been observed. Their density and
extent depend on the flow rate and the brine composition.
• High flow rates give longer dissolution forms, while low
flow rates lead to more compact dissolution.
• Presence of SO42- in the brine favors re-precipitation of
insoluble salts leading to shorter wormholes and to
permeability decrease.
These results have implications in the well injectivity
studies and in the far field modeling. The coupling between
geochemical reactions and flow regime has to be properly
taken into account in the field to optimize the injection
6 SPE 93674
conditions and in the large-scale numerical models to get
relevant results.
• Permeability impairment in the near well bore could be
limited in spite of unfavorable geochemical conditions if
the flow rate is high enough to keep the equilibrium area
for precipitation far from the well.
• Specific porosity/permeability relationships as a function
of the flow regime have to be elaborated and introduced in
the numerical models.
In future work the impact of rock composition and of the
initial pore structure needs also to be investigated.
Acknowledgements
The authors gratefully thank IFP (Institut Français du Pétrole)
for permission to publish these results. The authors also
acknowledge E. Brosse, C. Magnier for the fruitful discussions
and J. Behot, F. Deflandre, C. Fichen, E. Rosenberg and B.
Prempain for their key contributions in the experimental work.
Nomenclature
DP: Differential Pressure
PV: Pore Volume
SI: Saturation Indice
T2: NMR transverse relaxation time
V/S: Volume/Surface ratio
References
1 Grigg, R.B.: "State of the Industry in CO2 Flood", SPE 38
849, SPE Annual Technical Conference, San Antonio, 5-8
October, 1997.
2 Hadlow, R.E.: " Update of Industry Experience with CO2
Injection", SPE 24928 presented at the Annual Technical
Conference, Washington, 4-7 October 1992.
3 Rogers, J.D. and Grigg, R.B.: "A literature Analysis of the
WAG Abnormalities in the CO2 Process", SPE 59329
presented at the SPE/DOE Improved Oil Recovery
Conference, Tulsa, Oklahoma, 3-5 April 2000.
4 Goodrich, J.H.: " Review and Analysis of Past and Ongoing
Carbon Dioxide Injection Field Tests", SPE 8832 presented at
the Symposium on Improved Oil Recovery at Tulsa, April 20-
23, 1980.
5 Flemming E.A, Brown L.M, Cook R.L: " Overview of
Production Engineering Aspects of Operating the Denver Unit
CO2 Flood", SPE/DOE 24157 presented at the 8TH SPE/DOE
Enhanced oil recovery Conference, Tulsa, 22-24 April 1992.
6 Pittaway, K.R. and Runyan, E.E.: "The Ford Geraldine Unit
CO2 Flood: Operating History", SPE 17278 presented at the
Permian Basin Oil and Gas Recovery Conference, Midland,
10-11 March 1988.
7 Kamath, J ., Nakagawa, F.M., Boyer, R.E. and Edwards,
K.A..: "Laboratory Investigation of Injectivity Losses during
WAG in West Texas Dolomites", SPE 39791, Permian Basin
Oil and Gas Conference, 23 - 26 March, 1998.
8 Christman, P.G. , Gorell, S.B.: " Comparison of laboratory
and field observed CO2 tertiary injectivity", SPE 17335, JPT,
February 1990, p.226.
9 Roper, M.K. , Pope, G.A. and Sepehrnoori, K.: " Analysis of
Tertiary Injectivity of Carbon Dioxide", SPE 23974 presented
at the Permian Basin Oil and Gas Recovery Conference held
in Midland, Texas, March 18-20, 1992.
10 Thibeau, S., Barker, J.W. and Morel, D.: Simulation of
Sour gas Injection into Low Permeability Oil reservoirs", SPE
84362 presented at the Annual Technical Conference, Denver,
5-8 October 2003.
11 Dibble, W.E. and Potter J.M.:”Effect of fluid flow rates on
geochemical processes”, SPE 10994, Annual Technical
Conference and Exhibition, September 26-29 1982, New
Orleans, LA.
12 Noh, M., Lake, L.W., Bryant, S.L. and Araque-Martinez,
A.:”Implications of coupling fractional flow and geochemistry
for CO2 injection”, SPE/DOE 89341, Symposium on
Improved Oil Recovery, April 17-21 2004, Tulsa, TX.
13 Bazin, B., Bieber, M.T., Roque, C. and Bouteca, M.:
"Improvement in the Characterization of the Acid Wormholing
by "In Situ" X-Ray CT Visualizations", SPE 31 073 presented
at the symposium on Formation Damage held in Lafayette, 14-
15 February 1996.
14 Bazin, B.: "From Matrix Acidizing to Acid Fracturing: a
Laboratory Evaluation of Acid /Rock Interactions", SPE
49491 presented at the 8th Abu Dhabi International Petroleum
Exhibition and Conference held in Abu Dhabi, U.A.E., 11-14
October 1998.
15 Ross, G.D., Todd, A.C., Tweedie, J.A. and Will A.G.:” The
dissolution effects of CO2-brine systems on the permeability of
U.K. and North Sea calcareous sandstones”, SPE/DOE 10685,
Symposium on Enhanced Oil Recovery, April 4-7 1982, Tulsa,
TX.
16 Grigg, R.B. and Svec, R.K.:”Co-injeted CO2-brine
interactions with Indiana Limestone”, SCA 2003-19, Int.
Symposium of the Society of Core Analysts, September 21-24
2003, Pau, France.
SPE 93674 7

17 Svec, R.K. and Grigg, R.B.:”Physical effects of WAG 18 Mathis, R.L., Sears, S.O.:”Effect of CO2 flooding on
fluids on carbonate core plugs”, SPE 71496, Annual Technical dolomite reservoir rock, Denver Unit, Wasson (San Andres)
Conference and Exhibition, 30 September- 3 October 2001, field, TX”, SPE 13132, Annual Technical Conference and
New Orleans, LA. Exhibition, September 16-19 1984, Houston, TX.
8 SPE 93674

Table 1. Composition of synthetic seawater.

Species concentration mg/l

Na+ 12,158
Ca2+ 401
Mg2+ 1,315
SO42- 2,850
Cl- 22,010

Table 2 - Geochemical Calculations

seawater - PCO2 = 100 bars - T=90°C.

Step 2
Step 1Before After
precipitation precipitation
pH 3.2 4.8 4.8
Ca2+ (mg/l) 415 1,610 1,428
SO42- (mg/l) 2,850 2,850 2,371
SI (CaCO3) -3.50 0.00 0.00
SI (CaSO4) -0.30 0.20 0.00
Saturation Indices (SI) in log units.
Figure 2 : NMR T2 response
1 .E + 0 4

Back pressure
1 .E + 0 3 Bleed off line of valve
the inlet
Brine
pump
1 .E + 0 2
Pc Hg (bar)

Differential pressure
1 .E + 0 1
Sampling device

By-pass
CO2 Oil DP
1 .E + 0 0
1 0 .8 0 .6 0 .4 0 .2 0
1 .E -0 1 water water

1 .E -0 2 CO2
pump
S H g ( f r a c t io n )
Confining
cell

Figure 3 : experimental set-up

Frequency

Low rate High rate

0.01 0.1 1 10
Throat size (microns)
Figure 4 : dissolution patterns (inlet face)

Figure 1 : mercury porosimetry and associated pore

size distribution
SPE 93674 9

3000 8.6
without SO4-- with SO4--
8.4
2500 SO4 --

8.2

2000

[SO4-- ] and [Ca++]

8
pH

pH
1500 7.8

7.6
Figure 5 : wormhole longitudinal extension at high 1000
injection flowrate (250/250 cm3/hr)
Ca ++ 7.4

500
7.2
0.06
0 7
0 5 10 15 20 25 30
0.05 PV injected

Figure 8 : produced water analysis (low flowrate)

Delta porosity (fraction)

0.04 High rate

Intermediate rate
Low rate
0.03

3000 8.6
0.02
SO4 --
8.4
2500
0.01
8.2

2000 pH
[SO4-- ] and [Ca++]

0 8
0 5 10 15 20 25 30

pH
Length into the core (cm) 1500 7.8
Ca ++
Figure 6 : variation of porosity profiles at 3 7.6
injection rates with the seawater 1000

7.4

500
7.2

1 cm 0 7
0 5 10 15 20 25 30 35
PV injected

Figure 9 : produced water analysis (high flowrate)

Figure 7 : 3D wormhole pattern (high injection

flowrate 250/250 cm3/hr and seawater)
10 SPE 93674

2
1.8 Solid particles
1.6 displacement
1.4
Normalized DP

1.2
1
0.8
0.6
0.4
0.2
0
0 10 20 30 40
Total pore volume injected

Figure 10 : normalized pressure drop without SO42-

(high injection rate)
2
1.8
1.6
Figure 13 : NMR response prior and after injection
1.4
(low injection rate)
Normalized DP

1.2
1
0.8
0.6
0.4
0.2
0
0 10 20 30 40
Total pore volume injected

Figure 11 : normalized pressure drop with SO42-

(high injection rate)

1.3
1.2
1.1 Brine with SO4 2-
1 Brine without SO4 2-
K final / K initial

0.9

0.8
0.7
0.6
0.5

0.4
0.3
1 10 100 1000
Total flowrate (cm3/hr)

Figure 12 : permeability variations