See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/283981825

Effect of nanoparticles on clay swelling and migration

Article  in  Egyptian Journal of Petroleum · November 2015

DOI: 10.1016/j.ejpe.2015.10.006

CITATIONS READS

27 665

4 authors, including:

Peyman Pourafshary Shahab Ayatollahi

Nazarbayev University Sharif University of Technology
161 PUBLICATIONS   1,710 CITATIONS    356 PUBLICATIONS   7,193 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

effects of pore space evolution on transport in porous media View project

Asphaltene precipitation under different gas injection scenarios View project

All content following this page was uploaded by Shahab Ayatollahi on 03 February 2016.

The user has requested enhancement of the downloaded file.

Egyptian Journal of Petroleum (2015) 24, 429–437

H O S T E D BY
Egyptian Petroleum Research Institute

Egyptian Journal of Petroleum

www.elsevier.com/locate/egyjp
www.sciencedirect.com

REVIEW

Effect of nanoparticles on clay swelling

and migration
Abdolhamid Sameni a,*, Peyman Pourafshary b, Milad Ghanbarzadeh a,
Shahab Ayatollahi c

a
Institute petroleum of Engineering, University of Tehran, Iran
b
Department of Petroleum and Chemical Engineering, Sultan Qaboos University, Oman
c
Department of Chemical Engineering, Sharif University of Technology, Iran

Received 2 June 2014; revised 4 September 2014; accepted 16 September 2014

Available online 6 November 2015

KEYWORDS Abstract Clay migration/swelling has been widely documented as the main reason leading to oil
Clay stabilizers; recovery impairment. Interactions of clay particles with the medium surface in the presence of per-
Nanoparticles; meating fluid have been recognized as a critical parameter controlling the fate of clay particles.
Swelling index; These interactions are strongly functions of the ionic strength of the permeating fluid. It is widely
Core flood; reported that reducing the salinity of the reservoir environment facilitates the challenges induced by
Micro model clay particles. On the other hand, low salinity reservoir environment has been recognized as a very
favorable condition for oil recovery. Accordingly, one should consider the positive effect of reduc-
ing salinity on oil recovery and its deteriorative effect on clay particles at the same time to improve
oil recovery in a controlled formation damage mode. This experimental work aims to investigate the
potential remedial effect of different metal oxide nanoparticles to treat clay swelling. Several core
flood experiments and micro-model tests have been conducted to achieve the mentioned goal. Fur-
thermore, swelling tests were quantified in terms of swelling indices to explain the effect of nanopar-
ticles on clay swelling. We concluded that although nanoparticles can be used as a permanent
stabilizer to prevent clay migration, they are not able to prevent clay swelling and may also increase
the pressure drop due to fitting between clay crystals and blocking pores.
Ó 2015 The Authors. Production and hosting by Elsevier B.V. on behalf of Egyptian Petroleum Research
Institute. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).

* Corresponding author.
E-mail addresses: sameni90ipe@ut.ac.ir (A. Sameni), Pourafshary@
squ.edu.om (P. Pourafshary), mghanbarzadeh.90@gmail.com
(M. Ghanbarzadeh), dr.ayatollahi@gmail.com (S. Ayatollahi).
Peer review under responsibility of Egyptian Petroleum Research
Institute.
http://dx.doi.org/10.1016/j.ejpe.2015.10.006
1110-0621 Ó 2015 The Authors. Production and hosting by Elsevier B.V. on behalf of Egyptian Petroleum Research Institute.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
430 A. Sameni et al.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430
2. Materials and experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432
2.1. Core flood primary tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432
2.2. Swelling index test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432
2.3. Final core flooding test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
2.4. Micromodel tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
3. Results and discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
3.1. Core flood primary tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
3.2. Swelling index test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436
3.3. Final core flooding test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436
3.4. Micromodel tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436
4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437

1. Introduction between the two layers increases leading to clay swelling.

The swelling process comprises two main stages: the first stage
Formation damage mechanisms affect reservoir permeability is crystalline swelling, which is a reversible process; the
especially in the near wellbore area and may happen during second is osmotic swelling whereby the particles disperse and
different operations such as drilling and production. It has migrate with flow, and is an irreversible process, as shown in
been reported that approximately 97% of all petroleum reser- Fig. 3.
voirs contain clay minerals and they are categorized into swel- Three main mechanisms are used to reduce clay swelling
ling and non-swelling groups [1]. Non-swelling clays such as which are ion exchange, coating of the clay particles by stabi-
kaolinite may migrate during water flooding because of repul- lizers, and modification of surface affinity toward water [2].
sive forces between clays and pore walls and block pores in the These methods can also be categorized into temporary and
reservoir, as schematically shown in Fig. 1a. Swelling clays permanent remedies. Temporary clay stabilizer additives are
such as montmorillonite may swell in contact with the invaded materials that prevent swelling and migration of clays but
water up to 20 times the original volume, as shown in Fig. 1b, are easily removed by the formation-produced fluids following
and may also migrate. Both these effects reduce porosity and the treatment. The most common temporary clay stabilizers
permeability in the formation. are simple inorganic salts such as NaCl, KCl, ammonium chlo-
The reason for this special behavior of the clays is due to ride (NH4Cl), and calcium chloride (CaCl2). The most recent
their unique structures. The crystal structure of swelling clays advances in clay stabilization have been focused on the area
consists of Al–OH or Fe–OH or Mg–OH octahedral, sand- of permanent clay stabilizer additives. The most common per-
wiched by two Si–O tetrahedral layers, as shown in Fig. 2. manent clay stabilizers are quaternary amine polymers. A
These layers are always deficient in positive charges because monomolecular film of these polymers tightly binds with the
of cation substitution. Interlayer cations are required to bal- clay surface by means of cation exchange and is not removed
ance the negative layer charges. When the exchangeable by the produced fluid [3]. Zhou (1995) divided clay stabilizers
cations are hydrated during water injection and water mole- into different classes and described their advantages and disad-
cules enter the space between the structure layers, the distance vantages [4].

(a)

(b)
Figure 1 Schematic of formation damage by clay mechanisms:
(a) migration; (b) swelling. Figure 2 Schematic of clay structure.
Effect of nanoparticles on clay swelling and migration 431

Figure 3 Swelling process.

Figure 4 Schematic of core flooding set-up.

Nanoparticles have been recognized as very efficient agents

to remedy the colloidal particles migration through the med-
ium. Habibi et al. (2013) investigated the effects of various
nanoparticles such as MgO, SiO2 and Al2O3 on reduction of
fines migration. They concluded that the MgO nanoparticle
is an effective agent for controlling fines migration [5]. Ahmadi
et al. (2011) studied the zeta potential effects on fine migration
reduction [6]. Assef et al. (2013) extended the study to include
the performance of nanoparticles in formation damage control
under various PH and salinity conditions [7]. Recently, similar
attempts have been made to control clay swelling with
nanoparticles, with contradictory results. Sensoy et al. (2009)
tried to use nanoparticles to control clay swelling during water
invasion in drilling shale formations. They concluded that
Figure 5 Swelling index.
injection of nanoparticles could not directly solve the swelling
problem and just increased the pressure drop due to pore plug-
ging in the shale formation [8]. On the other hand, Huang and observed that nanoparticles reduce the pressure drop in
(2011) studied the ability of nanoparticles to reduce clay swel- cores [9]. Therefore, the basic question remains: ‘‘Can
ling in proppant application during water flooding of cores nanoparticles permanently reduce clay swelling and
432 A. Sameni et al.

Figure 6 SEM from Nanoparticles: (a) SiO2 (b) MgO (c) Al2O3.

migration?”. In this work, we have applied different experi-

mental approaches to answer this question.
Table 1 Permeability results for different injections.
In the first step, we did several water flooding tests in the
core scale to study the pressure drops due to clay swelling Permeability (mD)
and migration. Then we studied the effects of nanoparticles Fresh water 66
on damage caused by swelling and migration, and conducted KCl 405
core flooding tests. The effects of nanoparticles were also com- Nano fluid 57
pared with those of other common stablizers. Micro-model
experiments were performed to study different processes for
recovering residual oil inside the porous medium during
EOR or stimulation operations to study fluid flow in porous 2.2. Swelling index test
medium. We also did several tests in micro-model scale for bet-
ter visualization of the process of fluid flow.
A series of swelling index tests were performed to study the
performance of our approach to control the swelling of ben-
2. Materials and experiments tonite in comparison with other common methods. The swel-
ling index test is a simple standard test (ASTM D5890-11),
2.1. Core flood primary tests which is used to measure the swelling tendency of clays.
To perform these tests, a 2 g sample of dry and finely
This preliminary phase studied the effect of clay swelling and ground bentonite clay was added to water in a 100 ml
migration on pressure drop during injection. Results were the cylinder, as shown in Fig. 5. The sample was then covered
reference for our proposed method of remediation. In this and protected from disturbances for a period of 16–24 h.
experiment we made a synthetic porous medium from glass After the complete hydration and swelling of the clays, the
beads with US mesh size between 150 and 200 (74–90 lm). increase in the mixture’s volume was measured as an indica-
We also used experimental bentonite as clay (Density: 620 g/l, tor of the swelling.
Swelling Volume: 10–15 cc). The porous medium was prepared For studying the effect of nanoparticle treatment on the
with 90% glass beads and 10% by weight bentonite powder. swelling mechanism of bentonite, three types of nanoparticles,
The porous medium was well homogenized in a core holder MgO, SiO2 and Al2O3 prepared from Nanoshell company,
measuring 3.81 cm in diameter and 11 cm in length. Then we were selected because of their accessibility and previous studies
applied 800 psi overburden pressure to the core. The flooding regarding their effects on fines migration prevention [5]. SEM
system is shown schematically in Fig. 4. pictures of the used nanoparticles are shown in Fig. 6.

Figure 7 Micromodel set-up.

Effect of nanoparticles on clay swelling and migration 433

150–200 US mesh and 30% porosity was prepared for the

experiments. The glass bead and bentonite were mixed and
well packed in the core holder system. First the porous media
was vacuumed and then saturated by fluid injection. The mate-
rials including brine and nanofluid were flooded to the syn-
thetic core. During each experiment the pressure drop along
the core was measured.

2.4. Micromodel tests

In order to visually investigate the effect of nanoparticles, a

low-pressure micromodel was prepared and used as shown in
Fig. 7. All injection experiments were carried out at ambient
temperature under a constant flow rate. The micromodel used
in the experiments was constructed from two Plexiglass slides
Figure 8 Schematic of the apparatus used in the experiments. of roughly the same size. On both slides 10 holes were made
to accommodate screws. Each slide’s thickness was 10 mm
and a Plexiglass plate with a thickness of 1 mm was used to fill
2.3. Final core flooding test the micromodel with glass beads.
Table 1 shows the permeability and porosity of the micro-
A core flooding test was designed to study both phenomena, model. In this table the properties of the micromodel are pre-
swelling and migration simultaneously. The same synthetic sented before various fluid floodings including water and
porous medium with 10% clay and 90% glass beads with stabilizers.

(a) (b)
Figure 9 (a) Improper clay distribution; (b) homogenized clay distribution.

Figure 10 Pressure drop in porous medium during distilled water injection.

434 A. Sameni et al.

Figure 11 Pressure drop in porous medium during brine (2% KCl) injection.

Figure 12 Pressure drop in porous medium during re-injection of distilled water.

Table 2 Swelling index test results. Table 3 SI results for different types and concentrations of
Fluids SI (ml/2 g bn) nanoparticles.

Distilled water 32 Fluids SI Fluids SI Fluids SI

2% KCl solution 8 0.1% 33 0.1% TiO2 35 0.01% Alumina 33
0.1% SiO2 suspension 32 ZnO (size 30 nm)
0.1% MgO suspension 32 0.1% 32 0.01%TiO2 29 0.05% alumina 32
0.1% Al2O3 suspension 32 ZrO2 + 5% LA
0.1% 34 0.1% alumina 33 0.2% alumina 34
CeO2 + 0.4% SDBS
0.1% 31 0.1% alumina 32 0.05% alumina 32
A schematic diagram of the experimental set-up is shown in ZH + 0.5% CTAB (size 8 nm)
Fig. 8. The micromodel set-up was composed of three parts com-
prising fluid injection section, micromodel and imaging system.
A Nexus 6000 syringe pump was used which can provide and then packed into the micromodel. The clay should be quite
injection rates between 0.01 ll=min and 200 ll=min. Pictures homogeneous among the glass beads. Fig. 9a shows incorrect
from the top of the micromodel were taken while it was position of the clay in the micromodel and Fig. 9b shows a
illuminated with UV light. Glass beads with a diameter of homogenized distribution. The micromodel was vacuumed
0.4–0.59 mm were dry-mixed with 10 wt% bentonite clays prior to fluid injection.
Effect of nanoparticles on clay swelling and migration 435

Figure 13 Pressure drop along the core during injection of water and 0.1% MgO nanoparticles.

Figure 14 Radius of nanoparticle charge effect.

3. Results and discussions pressure drop reduced from 16 psi to 2 psi which means that
the clay swelling is controlled by KCl. Similarly, we can
3.1. Core flood primary tests observe two different behaviors of the clays similar to the first
experiment that initially, a pressure drop hump occurs because
of the clay swelling and then it stabilizes because of the migra-
In the first test the prepared porous medium was flooded by dis-
tion of the clays out of the porous medium.
tilled water at the rate of 15 cc/h. Fig. 10 shows the pressure drop
across the core. Flooding was initiated by saturation of the core
by vacuum. Due to swelling of clays in the presence of distilled
water, the saturating procedure failed. Hence, first the porous
media was vacuumed and then saturated by water injection.
Clay behavior can be divided into two sections. In the first
stage, after one pore volume of injection, crystalline swelling
occurs and the distance between bentonite plates increases
which leads to a pressure drop. In the second part, osmosic
swelling starts, the clay structure collapses and plates migrate
through the porous medium. As the clays break through, the
porosity starts to increase and the pressure drop decreases,
as shown in Fig. 10.
In the second test, we measured the pressure drop in the
core during flooding with 2% KCl brine solution as a tempo- Figure 15 Schematic of clay crystals in contact with ions and
rary stabilizer at the rate of 15 cc/h. As shown in Fig. 11, the nanoparticles.
436 A. Sameni et al.

Figure 16 Paraffin injection in different conditions: (a) after distilled water injection, (b) after KCl injection, (c) after MgO nanofluid
injection.

Fig. 12 shows the pressure drop measurements for the

re-injection of distilled water after KCl flooding in the second Table 4 Porous medium properties.
experiment. This figure shows that the remedial action of KCl Porous medium US mesh size (30–40)
is temporary and clays are swelling again.
Permeability without clay (D) 8.37
Porosity without clay (%) 37
3.2. Swelling index test Permeability with clay (D) 2.8
Porosity with clay (%) 14
In the first experiment, we measured the swelling index (SI) of
bentonite (bn) in distilled water and in 2% KCl solution. These
experiments were treated as a reference for comparison with During the contact between low salinity water and clays, as
the nanoparticle-treated cases. Table 2 shows SI results of the concentration of ions between clay sheets is higher than ion
these cases. As can be seen, adding KCl can control the in the bulk, water molecules diffuse into the clay structure
swelling of bentonite. because of the osmotic pressure difference which causes clay
As shown in Table 2, no remedial effect of this new swelling. Increasing the salinity can solve this problem, as it
approach was observed for the swollen clays. also increases the ion concentration in the whole bulk region.
We checked different nanoparticles at different concentra- With nanoparticles, however, change in the charges did not
tions and diameter sizes. Also, we tried to change the surfactant happen in all sections of the bulk water: osmotic behavior
types used for nanofluid preparation. As shown in Table 3, occurred and clays started to swell. Fig. 14 shows this phe-
none of them reduced swelling index (SI) in a noticeable nomenon schematically. The effects of the nanoparticles are
manner. noticeably very close to them. Hence, they cannot affect the
Our preliminary experiments showed that nanoparticles are fluid bulk.
not able to prevent clay swelling at any concentration. As men- Generally as shown in Fig. 15, after the swelling of the
tioned earlier, other research has shown that nanoparticles are clays, the distance between the plates is less than 2 nm.
effective in controlling fines migration during water movement Positive ions can enter between the sheets due to ions size
but for the swelling we have not seen any noticeable effect. and reduce the distance between them thanks to their charges
but our experiments showed that this process does not
3.3. Final core flooding test happen during nanoparticle injection [10]. Nanoparticles are
normally larger than 5 nm in size. Hence, nanoparticles
From the basic tests we concluded that nanoparticles are cannot enter between the sheets to neutralize the negative
unable to reduce swelling but they can stabilize the clay plates charges between them.
from migration and act as a permanent stabilizer.
Fig. 13 shows the pressure drop along the core during water 3.4. Micromodel tests
and 0.1% MgO nanoparticle dispersed in water injection.
It can be seen that a pressure drop equal to 40 psi occurred The porous medium was injected with different fluids such as
along the core, which is much higher than the reference case in fresh water, 2% potassium chloride solution and 0.01% MgO
Fig. 10 which was about 16 psi. This excessive pressure drop nanofluid. After saturation with different fluids, paraffin was
reveals that although the nanoparticles reduce the clays’ migra- flushed through the micromodel to display its advance into the
tion they are not effective at controlling formation damage medium. Fig. 16 shows the open sections of the porous medium
caused by high clay swelling and pore-blocking. In formations to the movement of paraffin. Hence, this figure shows clearly the
containing swelling clays, injection of nanoparticles will sections which are blocked by clay swelling and migration.
increase the pressure drop due to the additional pluggings in The movement of the liquid paraffin into the micromodel
pore throat spaces. Hence, this approach is not recommended pore spaces is clearly influenced by the presence of the clay.
as a remedial method to prevent formation damage caused by Porous media properties are shown in Table 4. Permeability in
clay swelling. the presence of clay was measured by KCl injection. KCl is used
Effect of nanoparticles on clay swelling and migration 437

to prevent clays swelling and migration. During the nanofluid 4. Nanofluid injection decreased the permeability of the por-
injection due to the clays migration and pores plugging, the pres- ous medium due to plugging of the pores.
sure drop along the core was not stable. Hence, it is not possible
to measure the permeability in these cases. Fig. 16a shows the sit-
Acknowledgements
uation after the distilled water injection. It can be seen that the
movement of the paraffin is limited and different sections of
Z. Fakhruiyan, A. Habibi, are acknowledged for their help in
the porous media are damaged because of swollen clay. After
performing some of the experimental work conducted in this
injection of KCl according to Fig. 16b, it can be seen that the
study.
fluid flow in porous medium is more homogeneous and this
treatment controls the swelling process. Fig. 16c shows the per-
formance of the paraffin injection after treatment by nanofluids. References
Flow paths are limited and permeability reduces in all sections of
the medium. This experiment proves that nanoparticles are not [1] K.K. Mohan, R.N. Vaidya, M. Reed, H.S. Fogler, Colloids
Surf. A Physicochem. Eng. Aspects 73 (1993) 237–254.
effective for controlling damage caused by clay swelling. Perme-
[2] A. Zaltoun, N. Berton, SPE Prod. Eng. 7 (2) (1992) 160–166.
ability in all three cases was measured and is shown in Table 4.
[3] S. Berry, J. Boles, H. Brannon, B. Beall, Performance evaluation
Nanofluid injection decreases the permeability of the porous of ionic liquids as a clay stabilizer and shale inhibitor, in: SPE
medium because of plugged pores. International Symposium and Exhibition on Formation
Damage Control, 2008.
4. Conclusions [4] Z. Zhou, W. Gunter, R. Jonasson, Controlling formation
damage using clay stabilizers: a review, in: Annual Technical
Meeting, 1995.
This work aims to investigate nanoparticle treatment of clay [5] A. Habibi, M. Ahmadi, P. Pourafshary, S. Ayatollahi, Y. Al-
swelling and/or migration. The following can be inferred based Wahaibi, SPE J. 18 (2) (2013) 309–318.
on the conducted measurements including swelling index test, [6] M. Ahmadi, A. Habibi, P. Pourafshary, S. Ayatollahi, Zeta
segregation test, core flood test, and micromodel experiments: potential investigation and mathematical modeling of
nanoparticles deposited on the rock surface to reduce fine
1. The swelling index test proved that nanoparticles cannot migration, in: SPE Middle East Oil and Gas Show and
prevent swelling. Nanoparticle suspensions do not act as Conference, 2011.
a single phase fluid to diffuse between the clay crystals as [7] S. Ayatollahi, Y. Assef, P. Pourafshary, Reduction of fine
salt solutions do. The size of the smallest nanoparticles is migration in different pH and salinity conditions using
nanofluid, in: 10th SPE International Conference and
larger than the size of the largest ions and therefore they
Exhibition on European Formation Damage. 2013.
are not able to diffuse into the clay crystals. [8] T. Sensoy, M. Chenevert, M. Sharma, Minimizing water
2. Segregation tests showed that nanoparticles are effective as invasion in shales using nanoparticles, in: SPE Annual
permanent stabilizers to prevent clay migration compared Technical Conference and Exhibition, 2009.
with salt solutions, which are temporary stabilizers. [9] T. Huang, Clay Stabilization with Nanoparticles, Patent
3. Core flooding tests showed that the presence of nanoparti- Application publication, Pub No. US 2012/0178651 A1,
cles cannot decrease the tendency to clay swelling. Hence, 2011.
nanoparticles do not have positive effects in terms of [10] Y. Lu, Study of property of cationic surfactant for preventing
remedying formation damage caused by clay swelling. clay from swelling, in: International Meeting on Petroleum
Our observations support the findings of Sensoy et al. [8]. Engineering, 1988.

View publication stats