University of Groningen
Synthesis and evaluation of novel linear and branched polyacrylamides for enhanced
oil recovery
Wever, Diego-Armando Zacarias
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to
cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2013
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Wever, D-A. Z. (2013). Synthesis and evaluation of novel linear and branched polyacrylamides for
enhanced oil recovery Groningen: s.n.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the
author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the
number of authors shown on this cover page is limited to 10 maximum.
Download date: 31-07-2017
Synthesis and evaluation of novel linear and
branched polyacrylamides for enhanced oil
recovery
by Diego-Armando Zacarías Wever
Copyright © 2013 by Diego-Armando Zacarías Wever. All right reserved.
No part of this book may be reproduced or transmitted in any forms by any means
without permission of the author.
Cover design:
Carlos-Alberto Gregorio Wever and Diego-Armando Zacarías Wever
Printed by:
NetzoDruk, Groningen
ISBN: 978-90-367-6591-6
ISBN: 978-90-367-6592-3 (electronic version)
The work described in this thesis was conducted at the Department of Chemical
Engineering – Product Technology, Faculty of Mathematics and Natural Sciences,
University of Groningen, The Netherlands.
This research project was financially supported by Shell and SNF Floerger through the
Enhanced Oil Recovery program of the Dutch Polymer Institute (DPI), project nr. 716:
Design of new chemical products (polymers and amphiphilics) for EOR.
Synthesis and evaluation of novel linear and
branched polyacrylamides for enhanced oil
recovery
Proefschrift
ter verkrijging van het doctoraat in de
Wiskunde en Natuurwetenschappen
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. E. Sterken,
in het openbaar te verdedigen op
vrijdag 22 november 2013
om 11.00 uur
door
Diego-Armando Zacarías Wever
geboren op 5 november 1984
te Oranjestad, Aruba
Promotores:
Prof. dr. A.A. Broekhuis
Prof. dr. F. Picchioni
Beoordelingscommisie: Prof. dr. K. Loos
Prof. dr. ir. H.J. Heeres
Prof. dr. D. Vlassopoulos
Dedicated to my beloved wife,
The work in this thesis is at best captured by my wife’s words:
“To find the best liquid plastic to get more oil out of the ground”
and my father
Although you are physically
not present anymore,
I know you are watching
and that you are
proud of your son
This page intentionally left blank
Table of contents
1. Introduction................................................................................ 11
1.1. Oil recovery .............................................................................. 12
1.1.1.
Reservoir properties ...................................................................... 14
1.2. Enhanced oil recovery ................................................................ 15
1.2.1.
1.2.2.
Non-thermal ................................................................................. 15
Polymer flooding ........................................................................... 16
1.3. Currently used polymers ............................................................. 19
1.3.1.
1.3.2.
1.3.3.
1.4.
1.5.
1.6.
1.7.
1.8.
1.9.
Polyacrylamide (PAM) .................................................................... 19
Partially hydrolyzed polyacrylamide (HPAM) ..................................... 20
Xanthan gum................................................................................ 24
Possible new polymers for EOR .................................................... 26
Thickening capabilities ............................................................... 32
End-use requirements for polymers in EOR ................................... 35
Aim and scope of this thesis ........................................................ 37
Acknowledgement ..................................................................... 39
References................................................................................ 39
2. Acrylamide homo- and block copolymers by atomic transfer
radical polymerization in water................................................... 47
2.1. Introduction .............................................................................. 48
2.2. Experimental section .................................................................. 50
2.3. Results and discussion ............................................................... 52
2.3.1.
2.3.2.
2.3.3.
2.3.4.
ATRP of acrylamide ....................................................................... 52
Chain extension experiment, two step ............................................. 56
Chain extension experiment, in situ ................................................. 58
Block copolymerization, synthesis of PAM-b-PNIPAM.......................... 58
2.4. Conclusion ................................................................................ 60
2.5. Acknowledgements .................................................................... 60
2.6. References ................................................................................ 60
3. Branched polyacrylamides: Synthesis and effect of molecular
architecture on solution rheology ............................................... 63
3.1.
3.2.
3.3.
3.4.
3.5.
3.6.
Introduction .............................................................................. 64
Experimental section .................................................................. 65
Results and discussion ............................................................... 70
Conclusion ................................................................................ 83
Acknowledgements .................................................................... 84
References ................................................................................ 84
4. Control over the viscoelasticity of aqueous polyacrylamide
solutions by tailoring the polymer architecture .......................... 87
4.1.
4.2.
4.3.
4.4.
4.5.
4.6.
Introduction .............................................................................. 89
Experimental section .................................................................. 90
Results and discussion ............................................................... 94
Conclusion .............................................................................. 110
Acknowledgements .................................................................. 111
References .............................................................................. 111
5. Acrylamide-b-N-isopropylacrylamide block copolymers: Synthesis
by atomic transfer radical polymerization and effect of
hydrophilic-hydrophobic ratio on solution properties ............... 113
5.1. Introduction ............................................................................ 114
5.2. Experimental section ................................................................ 115
5.3. Results and discussion ............................................................. 118
5.3.1.
5.3.2.
5.3.3.
Synthesis of the macroinitiators .................................................... 118
Synthesis of the block copolymers PAM-b-PNIPAM........................... 118
Solution properties of poly(AM-b-NIPAM) ....................................... 120
5.4. Conclusion .............................................................................. 129
5.5. Acknowledgements .................................................................. 130
5.6. References .............................................................................. 130
6. Branched thermoresponsive polymeric materials: Synthesis and
effect of macromolecular structure on solution properties ....... 133
6.1. Introduction ............................................................................ 134
6.2. Experimental section ................................................................ 135
6.3. Results and discussion ............................................................. 140
6.3.1.
6.3.2.
6.3.3.
6.3.4.
6.3.5.
6.3.6.
Macroinitiators ............................................................................
Synthesis of PK30-g-(PAM-b-PNIPAM) ...........................................
Synthesis of PK30-g-(PAM-co-PNIPAM) ..........................................
Solution properties of PK30-gx-(PAMY-b-PAMZ)...............................
Solution properties of PK30-gx-(PAMY-co-PAMZ) .............................
Surface properties .......................................................................
140
141
144
145
148
151
6.4. Conclusion .............................................................................. 152
6.5. Acknowledgements .................................................................. 153
6.6. References .............................................................................. 153
7. Oil recovery using branched copolymers based on acrylamide . 157
7.1. Introduction ............................................................................ 158
7.2. Experimental section ................................................................ 160
7.2.1.
7.2.2.
7.2.3.
7.2.4.
Materials ....................................................................................
Polymer injectivity experiments ....................................................
Oil recovery................................................................................
Characterization..........................................................................
160
163
164
165
7.3. Results and discussion ............................................................. 166
7.3.1.
7.3.2.
Polymer injectivity ...................................................................... 166
Oil recovery................................................................................ 169
7.4. Conclusion .............................................................................. 175
7.5. Acknowledgements .................................................................. 176
7.6. References .............................................................................. 176
8. Towards new polymers for enhanced oil recovery .................... 179
8.1.
8.2.
8.3.
8.4.
Introduction ............................................................................ 181
Thickening capability, comb-shaped PAM .................................... 184
Salt resistance, comb-shaped PAM ............................................. 186
Hydrolysis resistance, comb-shaped PAM .................................... 189
8.4.1.
Results and discussion ................................................................. 191
8.5. Oil recovery, 2D flow-cell .......................................................... 197
8.5.1.
Oil recovery efficiency.................................................................. 197
8.6. Biopolymers for EOR ................................................................ 199
8.6.1.
8.7.
8.8.
8.9.
8.A.
Thickening capability and viscoelasticity ......................................... 199
Conclusion .............................................................................. 202
Acknowledgements .................................................................. 203
References .............................................................................. 204
Appendix 8A ........................................................................... 206
8.A.1.
Experimental section ................................................................... 206
8.B. Appendix 8B ........................................................................... 211
8.B.1.
Experimental section ................................................................... 211
Summary .................................................................................. 213
Samenvatting ............................................................................ 219
Compilacion .............................................................................. 225
Acknowledgements ................................................................... 231
Curriculum vitae ....................................................................... 233
This page intentionally left blank
Chapter 1
Chapter 1
Introduction
Abstract
Current crude oil extraction techniques are briefly introduced along with
enhanced oil recovery (EOR), particularly polymer flooding. The fundamentals
of polymer flooding are explained together with the requirements which the
polymers have to meet to be applied in this technology. An overview of
recent developments in the field of water soluble polymers aimed at
enhancing the solution viscosity is given. The currently polymers are
discussed in terms of their advantages and limitations. Eventually the aim
and scope of this thesis are presented.
Based on: D.A.Z. Wever, F. Picchioni, A.A. Broekhuis. Polymers for enhanced
oil recovery: A paradigm for structure-property relationships in aqueous
solution. Progress in Polymer Science, 2011, 36, 1558-1628.
11
Introduction
1.1. Oil recovery
The population of the world is projected to increase beyond 8 billion
people by 2035.1 In line with this increase and the corresponding raise in the
living standards in the developing countries, the energy consumption is
expected to grow by 34% between 2015 and 2035. 2 The global primary
energy supply in 2010 comprises several different sources (Figure 1.1).3
Nuclear
5,7%
Hydro
2,3%
Biofuels and waste
10%
Other (solar, wind, geothermal, etc)
0,9%
Coal/peat
27,3%
Natural Gas
21,4%
Oil
32,4%
Figure 1.1: World primary energy supply in 20103
Oil covers approximately 30% of the primary energy supply. The increase in
energy consumption will exert a relevant pressure at industrial level towards
a more efficient exploitation of the current sources. New ones, such as
renewables, have not been demonstrated to be reliable yet and as of 2010
account 2013 for 10 % (projected to reach 14% by 2035) of the total world
energy consumption.2 Therefore, to guarantee the supply of energy and
provide a transition period between current sources and the renewables ones,
current sources have to be exploited in a more efficient manner.
Current oil production has reached >90 million barrels per day (bpd). 4 In
the early 1950’s, the so called “peak oil” theory was developed. This
predicted in first instance a peak in oil production followed by a steady
decline. On the other hand, evidence for the underestimation of a field’s
productivity, on which the theory is based, has been recently published.5
Nevertheless, new technologies can increase and/or extend current oil
production. Easily (by current technologies) recoverable oil is running out;
however significant amounts of oil remain in the reservoirs after the
conventional
methods have
been exhausted.5
Therefore, in order to
guarantee its continuing supply, enhanced oil recovery (EOR) has to be
implemented. After the current recovery methods have been depleted
12
Chapter 1
approximately 7.0·1012 barrels6 of oil will remain in the oil fields, which
represents a production of more than 200 years at the current rate. An
overview of the total amount of oil (composed of conventional oil (light oils)
and unconventional oil (heavy oil and tar sands)), the remaining amount of
oil that can be recovered by current techniques (proven conventional
reserves), the remaining unconventional oil that can be recovered by new
techniques (recoverable resources) and what has been recovered so far is
presented in Figure 1.2.5
Figure 1.2: World’s oil resources5
Oil reservoirs can be classified into three main categories according to the
American Petroleum Institute gravity index (API): light oil reservoirs (API >
31), heavy oil reservoirs (API < 22) and tar sands. Light oil reservoirs are the
most common in the world. In light oil reservoirs the oil is embedded in
porous media (Figure 1.3). In most cases gas (small hydrocarbons) is also
present. In addition saline water is also present as an aquifer (water in a
porous media) or as connate water (water present in the pores of
sedimentary rock as they were being formed) in the oil deposit. The amount
of salt (or dissolved solid) is denoted as total dissolved solid (TDS).
The conventional techniques for extracting the oil out of a reservoir
consist of primary and secondary methods. The primary technique uses
natural forces to produce the oil. Three different mechanisms are utilized to
extract the oil: the aquifer drive, the gas cap drive and the gravity flow. The
aquifer drive, according to which the pressure that is exerted on the oil by
the aquifer represents the driving force for extraction, is the most efficient
mechanism. The production of oil leads to a decrease in pressure of the
reservoir, and the aquifer moves towards the production well. The oil cut
13
Introduction
decreases as more and more water is produced along with the oil. The gas
cap drives the oil in a similar fashion as the aquifer drive. Gas production
(along with the oil) is not seen as a disadvantage, since it also can be used
as an energy source. Finally, gravity is the important factor in the gravity
flow, for which the well placement is obviously. The use of this method is
limited and is heavily dependent on the geology of the reservoir. The primary
techniques recover, depending on the oil reservoir, on average between 1525 % of the original oil in place (OOIP).6, 7
The secondary method involves the injection of either water or gas to
increase the pressure in the reservoir, which in turn drives the oil out. After a
given time, the injected water breaks through in the production wells. As the
production well ages, after the water breakthrough, the water cut increases.
The use of the secondary methods enables the extraction of 20-50 % of the
OOIP depending on the reservoir.6, 7
At most 55 % of the OOIP can be recovered (in most cases this value is
much lower) using the primary and secondary techniques. Therefore a large
portion of the OOIP remains embedded in the reservoir. Since the 1970’s,
many different methods have been developed to increase the oil recovery as
a response to the oil crisis.6,
7
These all belong to the category improved oil
recovery (IOR). Improved oil recovery implies improving the oil recovery by
any means6, such as operational strategies. Enhanced oil recovery, a
subgroup of IOR, is different in that the objective is to reduce the oil
saturation below the residual oil saturation (the latter being defined as the oil
saturation after a prolonged waterflood).
1.1.1. Reservoir properties
Oil reservoirs are porous media of which part of the total volume (the
porosity, φ) is occupied by a fluid, either oil or water. The permeability of the
porous media is defined as the ability of a specified fluid to permeate (flow
through) the porous media. The permeability can be determined by Darcy’s
Law (equation 1.1).
(1.1)
= permeability (mD),
viscosity (cP),
= fluid flowrate (ml/s),
= section length (cm),
= pressure drop (bar), = fluid
= cross sectional area (cm2)
The permeability of the porous media of oil reservoirs varies significantly
depending on the type of reservoir. Reservoirs with sandstone (e.g.
Bentheim) as the porous media display permeability values higher than 1000
14
Chapter 1
mD while carbonate porous media display values lower than 10 mD (e.g.
Rote Mainz). Permeability values above 105 mD are considered pervious, 105
– 10 mD are semi-pervious and values lower than 10 mD are impervious.8
Most oil reservoirs contain either sandstone or carbonate as the porous media
and the range of permeability values encompasses the semi-pervious class as
well as a part of the impervious types.
An increase in the viscosity of the fluid is synonymous to a more difficult
flow through the porous media. This will lead to a lower permeability. One
way of increasing the viscosity of the displacing fluid is through the use of
water-soluble polymers (as an Enhanced Oil Recovery agent).9 A layer of
polymer
might
build
precipitation/adsorption),
up
on
thus
the
leading
surface
to
a
of
the
further
rock
(e.g.
decrease
of
by
the
permeability of the porous media.9
1.2. Enhanced oil recovery
Enhanced oil recovery (EOR) involves different techniques that were
developed to extend the oil field’s life. Most of the EOR techniques have been
developed early in the twentieth century as an answer to the low oil
production in combination with the oil crisis. However, the price of a barrel of
oil was at that time much lower (20$ per barrel) compared to the current one
(~100$ per barrel). Given the low oil price at that time, the EOR techniques
were not fully developed and their application was limited to only a few
projects. Nevertheless, much experience has been gained through the EOR
projects. The focus of oil companies has now turned back to EOR because of
the steep increase of the oil prices and the increased demand for oil products.
In addition, the dwindling number of easy recoverable oil reserves is also
crucial in stimulating the development of better EOR techniques. Two main
categories of EOR technology exist; thermal and non-thermal (the focus of
this thesis). For a complete overview of all the techniques the reader is kindly
referred to the literature.6, 7
1.2.1. Non-thermal
Non-thermal methods use either gas or chemicals to improve the
recovery of oil from an oil field. Chemical EOR consist of mainly polymer
flooding and alkaline surfactant polymer (ASP) flooding. Non-thermal EOR is
more suitable for light oils rather than the heavy (viscous) oils. Chemical EOR
has been developed early in the twentieth century and has been implemented
in several oil reservoirs with different (mixed) results. So far, polymer
flooding is by far the most used chemical EOR technique.
15
Introduction
1.2.2. Polymer flooding
The conventional methods (primary and secondary) can at best extract
55% of the OOIP.6 The oil crisis of the 1970’s sparked the efforts in
developing better oil extracting technologies. The development of polymer
flooding also started during this period. Many different polymer flooding
projects have been carried out with mixed results.10 Nonetheless, a
significant number of polymer flooding projects have been recently started in
many different countries11 indicating the maturity of this technique. The
polymer flooding projects have been performed in reservoir temperatures
ranging from 8 – 110 °C, reservoir permeability values in the range 0.6 –
15000 mD, oil viscosity between 0.01 – 1500 cP and resident brine salinities
ranging between 0.3 – 21.3 % TDS.10
When water is used as a secondary recovery method two different problems
arise, a macroscopic one related to the volumetric sweep efficiency (
as
defined in equation 1.2) and a microscopic problem related to the
displacement efficiency of oil. When a water-flood is performed not all of the
OOIP is contacted by the displacing fluid.
(1.2)
The displacement of a viscous fluid (oil) with another (immiscible) less
viscous fluid creates instabilities which lead to viscous fingering. According to
Homsy12 viscous fingering arises for all mixtures of fluids. In porous media
viscous fingering develops due to an increase in the effective permeability of
the porous media to water.13 Due to the viscous fingering, the sweep
efficiency of an oilfield is limited. Early breakthrough (production) of the
displacing fluid (water) typically occurs. As a consequence, large portions of
the oil reservoir are not swept, thus leaving vast quantities of oil behind. In
order to solve this problem and to enhance the oil extraction, the water-oil
mobility ratio has to be improved. This ratio is dependent on the relative
permeability values and the viscosities of both the oil and the water phase.
The water-oil mobility ratio13 ( ) is defined by equation 1.3, with
and
being the permeability of the porous media to water and oil respectively. The
viscosities of the oil and water are represented by
and
.
(1.3)
16
Chapter 1
A high volumetric sweep efficiency is obtained when the water-oil mobility
ratio is less than or equal to unity. At low mobility ratios a piston like sweep
of the reservoir will be obtained while at higher water-oil mobility ratios
viscous fingering will arise. In theory there are several different methods to
affect the water-oil mobility ratio. The permeability of the porous media can
be altered, the viscosity of the oil can be decreased or the viscosity of the
displacement fluid can be increased. In practice only the two latter
techniques are possible. A schematic presentation of the viscous fingering in
a porous medium is given in Figure 1.3. The flow pattern using a polymeric
solution is also displayed in Figure 1.3.
Figure 1.3: Viscous fingering (left) and polymer flow (right) in a porous medium
In most cases the displacement efficiency (
), as defined in equation 1.4 by
assuming constant oil density9, is relatively low because of the viscosity
difference between the displacing fluid (water) and the oil.
(1.4)
The higher the difference in viscosity between the displacing fluid and the oil,
the more inefficient is the displacement. The displacement efficiency can be
improved by using water-soluble polymers since the viscosity of the
displacing fluid is increased and this will recover the oil at a higher rate than
conventional water flooding. In theory, the residual oil saturation cannot be
reduced by polymer flooding; only the time it takes to reach the residual oil
saturation of the reservoir is reduced. However, this has been contested by
several researchers in China. According to a number of studies 14-19 residual
oil can be mobilized by employing visco-elastic materials, i.e. polymers.
Indeed, Lake9 has defined residual oil to be the one remaining behind in a
certain region of a reservoir that has been thoroughly swept with water.
Residual oil is classified into four different types14,
18
: oil film on the rock
17
Introduction
surface (1), oil trapped in dead ends (2), oil (ganglia) in pore throats retained
by capillary forces (3) and oil unswept in micro-scale heterogeneous portions
of the porous media (4). The four different types of residual oil are presented
schematically in Figure 1.4.
Figure 1.4: The different types of residual oil
Several studies15,
16
proposed a mechanism according to which the visco-
elastic properties of the water solution play a crucial role in enhancing the
recovery of the different kinds of oil outlined above. This mechanism is
mainly supported by indirect evidence and mathematical models.17, 18, 20, 21
Water-soluble polymers for EOR applications have been successfully
implemented, mainly in Chinese oilfields.22,
23
The purpose of the water-
soluble polymers in this application is to enhance the rheological properties of
the displacing fluid. Oil production increases with the microscopic sweep of
the reservoir and the displacement efficiency of the oil9 Indeed, the use of
water-soluble polymers improves the water-oil mobility ratio9, and leads to
enhanced oil recovery. However, given the harsh conditions present in most
oil reservoirs, new problems and limitations arise with the use of watersoluble polymers. Besides positively affecting solution rheology, water-soluble
polymers should withstand high salt concentration, the presence of calcium,
18
Chapter 1
high temperatures (> 70 °C) and long injection times (at least 12 months). 9,
24
High salt concentrations reduce the thickening capability of most ionic
water-soluble polymers while the presence of calcium leads to flocculation. 25
New
water-soluble
temperatures26,
27
polymers
were
successfully
tested
at
higher
while associative water-soluble polymers were tested and
showed promising results compared to traditionally used polymers. 28,
14-18,
Several studies
20,
21
29
demonstrated that the oil is produced faster
(compared to water flooding), but also more oil can be recovered.
Independently of the exact displacement mechanism and efficiency, the use
of water-soluble polymers for EOR still constitutes a challenging research field
at both industrial and academic level.
1.3. Currently used polymers
1.3.1. Polyacrylamide (PAM)
Polyacrylamide was the first polymer used as thickening agent for
aqueous solutions. The thickening capability (increase of the corresponding
solution viscosity) of PAM resides mainly in its high molecular weight, which
reaches relatively high values (> 1·106 g/mol). In the general framework of
EOR processes, PAM is mainly used as the reference “model system” for
chemical modification. Many authors have reported different attempts to alter
the chemical structure of PAM or to synthesize new acrylamide-based
copolymers
with
improved
properties,
i.e.
shear
resistance,
brine
compatibility and temperature stability.30-33 The synthesis of the copolymer
N,N-dimethyl
acrylamide with Na-2-acrylamido-2-methylpropanesulfonate
(NNDAM-NaAMPS) has been accomplished and the polymer was tested for its
performance in EOR applications.30,
31
The stability of the polymer at high
temperature was demonstrated by aging at 120 °C for 1 month. 30 By using a
sand pack, an improved performance in terms of EOR for the NNDAMNaAMPS copolymer31 as compared to an unmodified partially hydrolyzed
polyacrylamide, HPAM, was demonstrated. In another example, the oil
recovery
rate
through
the
use
of
starch-graft-poly(acrylamide-co-(2-
acrylamido-2-methylpropanesulfoacid)) was higher compared to HPAM, and
the novel polymer displayed better temperature and shear stability. 32 These
two examples already define a common research theme in the general field
of water-soluble polymers for EOR. That is, a strategic approach involving the
chemical modification of commercial polymers (in this case PAM) to tailor and
improve
the
corresponding
solution
properties
and
eventually
EOR
performance.
19
Introduction
1.3.2. Partially hydrolyzed polyacrylamide (HPAM)
HPAM, by far the most used polymer in EOR applications, is a copolymer
of AM and acrylic acid (AA) obtained by partial hydrolysis of PAM or by
copolymerization of sodium acrylate with acrylamide (AM).34 The chemical
structure of HPAM is provided in Figure 1.5.
Figure 1.5: Chemical structure of HPAM
In most cases the degree of hydrolysis of the acrylamide monomers is
between 25-35%.9, 35 The fact that a relevant fraction of the monomeric units
needs to be hydrolyzed (lower limit of 25 %) is probably related to the
formation of the corresponding salt. According to the general theory of
polyelectrolyte solutions36, the presence of electrostatic charges along a
polymer backbone is responsible for prominent stretching (due to electric
repulsion) of the polymeric chains in water which eventually results in a
viscosity increase compared to the uncharged analogue. On the other hand,
the degree of hydrolysis cannot be too high because the polymer solution will
become too sensitive to salinity and hardness of the brine (electrolytes
present in solution have a “shielding effect” on the electrostatic repulsion).37
Indeed, polyelectrolytes, i.e. polymers bearing charges, show significantly
different rheological behavior compared to their neutral analogues. 38-40 The
thickening capability of HPAM lies in its high molecular weight and also in the
electrostatic
repulsion
between
polymer
coils
and
between
polymeric
segments in the same coil.9 When polyelectrolytes are dissolved in water
containing electrolytes (salts) a reduction in viscosity is observed. 35,
41-43
It
has been demonstrated that the specific viscosity of HPAM solutions depends
on the amount of salt present.44 This effect is attributed to the shielding
effect of the charges9,
42
leading in turn to a reduction in electrostatic
repulsion and consequently to a less significant expansion of the polymer
coils in the solution. This results in a relatively lower hydrodynamic volume,
which is synonymous with a lower viscosity.43 A few decades ago,
20
Chapter 1
substitution of one or both hydrogens on the amide nitrogen with alkyl
groups has been presented as a solution to the salt sensitivity of HPAM 45,
46
,
although the exact reasons for this behavior have not been fully elucidated.
The addition of monovalent NaCl leads to a reduction in the level of
aggregation. However, at higher ionic strengths (higher salt concentration)
the addition of NaCl leads to macroscopic flocculation.47 It has also been
demonstrated that multivalent cations can form polyion-metal complexes that
affect the viscosity of the resulting solution.48-50 The dependence of the self
complexation of HPAM on the Ca2+ concentration and the degree of hydrolysis
of HPAM has been investigated. It was demonstrated48 that depending on the
Ca2+ concentration intra- and inter-chain complexation takes place (Figure
1.6).
Figure 1.6: Complexation behavior of HPAM under different conditions48
Besides the salt dependency, other factors influencing the viscosity of HPAM
solutions are the degree of hydrolysis, solution temperature, molecular
weight, solvent quality and pressure.44
The increase in the viscosity of the HPAM solutions cannot solely be
accounted for by the increase in viscosity of the solvent.51 The intrinsic
viscosity and the radius of gyration are both invariant with pressure, albeit
with a 10% experimental uncertainty.51 In principle, the average dimension
21
Introduction
of the polymer coils does not change while the solvent volume decreases (i.e.
by increasing the concentration). Therefore the volume fraction of the
polymer coil per unit volume of the solvent increases, hence a higher
viscosity.51 Another parameter that affects the solution viscosity of the
polymer solution is shear.52 Under high shear the HPAM polymer chains are
reduced in size due to chain scission, i.e. fragmentation.53 This leads to a
reduction in the solution viscosity.
HPAM is preferred in EOR applications since it can tolerate the high
mechanical forces present during the flooding of a reservoir. In addition,
HPAM is a low cost polymer and is resistant to bacterial attack. 9 Although the
HPAM solutions display pseudo-plastic behavior9, 35, 41, 54, 55 (shear thinning) in
simple viscometers, it has been demonstrated that these solutions show
pseudodilatant56, 57 characteristics (shear thickening) in porous media as well
as in viscometers at relatively high shear rates (e.g.
values higher than 100
-1
s ). Research has demonstrated the presence of a critical shear rate at which
the shear thickening behavior arises in viscometers.41,
42, 54, 55, 58, 59
This
critical value depends on the degree of hydrolysis of the HPAM, the solution
concentration, the temperature, the quality of the solvent and also on the
molecular weight of the polymer.42, 54 An increase in the degree of hydrolysis
leads to an onset of shear thickening at lower shear rates.54 By decreasing
the average molecular weight, an increase in the polymer concentration
results in a higher critical shear rate.54, 55
The aforementioned shear thinning of HPAM solutions below a critical
shear rate arises due to uncoiling of polymer chains and the dissociation of
entanglements between separate polymer coils.9 Stiffening of the polymer
backbone has been suggested as a possible approach to control the
dependency of HPAM polymer solutions on the shear.60 A stiff polymer
displays a lower mobility and therefore the entanglements, related to the
solution viscosity, can be conserved as the shear increases.
The shear thickening behavior has been attributed to changes in the
molecular conformation involving the formation of additional links between
two chains.59 The shear thickening behavior is observed both in laboratory
rheometers54 (in pure water and aqueous salt solutions) and in porous media.
According to several studies the shear thickening behavior in porous media
arises due to coil-stretched transitions of the polymer chains. The structure of
a porous medium can be seen as alternating wide openings and confined
throats through which the polymer coils have to navigate. In the wide
openings the polymer chains attain a coil structure. When these coils then
have to pass through a narrow throat the polymer coils are forced to deform
and stretch (elongational strain41,
22
57, 61
) in order to pass. This successive
Chapter 1
contraction and expansion of the polymer coils leads to pseudo-dilatant
behavior of the polymer solutions.57,
62, 63
This conformational change of the
macromolecules is reversible since it is commonly explained by formation at
macromolecular level of reversible interactions like hydrogen bonding.
Indeed, it is believed that hydrogen bonding arises for HPAM solutions
between the carboxylic functionalities.64 However, this is contested due to
conflicting data59, 65 on similar polymeric solutions (e.g. for dextran). Instead,
aggregation
of
hydrophobic
bonds
has
been
proposed 64,
albeit
in
poly(methacrylic acid), but this has not been confirmed. 66 A schematic
presentation depicting the essential behavior of HPAM solutions in shear flow
has been proposed55 (see Figure 1.7).
Figure 1.7: Schematic presentation of behavior of HPAM coils in shear flow[46]
Another behavior that has been identified for HPAM solutions, which is
important for EOR, is their negative thixotropic (rheopectic) property, i.e. an
increase in viscosity with shear-time at a constant shear rate.41,
Researchers have identified two different types
67
67-70
of rheopectic behavior for
HPAM solutions (Figure 1.8), type I and type II.
Figure 1.8: Type I and II of rheopectic behavior of HPAM solutions67
The type I effect is observed at low shearing and consists in a slow viscosity
increase with shear-time up to an asymptotic value. The type II effect is seen
23
Introduction
at high shear rates and is displayed as a steep viscosity increase after a
given shear-time, followed by pronounced viscosity oscillation.67
1.3.3. Xanthan gum
Xanthan
gum
is
a
polysaccharide,
which
is
fermentation of glucose or fructose by different bacteria.
produced
71
through
The most efficient
xanthan gum producer is the Xanthomonas campestris bacterium.71,
72
The
chemical structure of xanthan gum (Figure 1.9) displays the presence of two
glucose units, two mannose units and one glucuronic acid unit with a molar
ratio of 2.8-2.0-2.0.73
Figure 1.9: Chemical structure of xanthan gum
The backbone of xanthan gum is similar to cellulose. The side chains of the
polymer contain charged moieties, i.e. pyruvate groups, and the polymer is
thus a polyelectrolyte. However the classic polyelectrolyte behavior according
to which the solution viscosity decreases with the addition of salt is not
displayed in this case. The thickening capability of xanthan gum is due to its
high molecular weight, which ranges from 2 - 50 · 106 g/mol73,
74
and in the
rigidity of the polymer chains.
It has been demonstrated that upon addition of salt (mono- or divalent)
the xanthan gum chains undergo a cooperative conformational transition
from a disordered conformation to an ordered and more rigid structure75-78
(Figure 1.10).
The temperature and the ionic strength, i.e. the amount of electrolyte, of
the solution are triggers for the conformational transition. When testing at
24
Chapter 1
low shear, the rheology of the polymer solution is dependent on the
conformation with the disordered conformation displaying higher solution
viscosities.79 Polymeric solutions employing xanthan gum display high
viscosity at low shear rates80 and thus the disordered conformation
predominates at low shear rates. At high shear rates both conformations
display similar rheological behaviors.79 In addition, pseudoplastic behavior is
observed for the polymer solutions.81 Unlike HPAM, xanthan gum displays
good resistance to high temperatures. It was demonstrated that the solution
viscosity of a polymeric solution employing a commercial xanthan gum
remained relatively constant for more than 2 years at 80 °C. 82 Loss of
solution viscosity occurs at temperatures above 100 °C.
Figure 1.10: Conformational transition of xanthan gum
Several studies83-86 have investigated the temperature dependence of the
apparent viscosity of xanthan gum solutions. In order to display resistance to
temperatures up to 90 °C, the conventional understanding for xanthan gum
solutions is that the ionic strength of the solution has to be relatively high.
Another positive property of xanthan gum is its ability to withstand high
shear forces. Unlike HPAM the solution viscosity does not decrease at
relatively high shear stresses.56 Especially the ordered structure, i.e. in the
presence of salt, can withstand high shear forces79 (up to a shear rate of
5000 s-1).
A disadvantage of xanthan gum is its susceptibility to bacterial
degradation. It has been demonstrated that salt tolerant aerobic and
anaerobic microorganisms can degrade the xanthan gum chains which leads
to the loss in solution viscosity.87-90 Biocides are used to suppress the growth
25
Introduction
of the xanthan gum degrading microorganisms. In most cases formaldehyde
is the most efficient biocide.89,
90
However, the use of biocides to protect the
xanthan gum renders the low environmental impact of the polymer at least
debatable.
1.4. Possible new polymers for EOR
The limited number of available commercial polymers currently employed
in EOR has been the subject of recent developments aimed at improving their
performance. Indeed, a relatively new concept has been studied in the last
four decades, and involves the association between hydrophobic groups that
are incorporated in the backbone of the polymers. 91 Through these
associations a higher thickening capability can be achieved compared to the
traditional polymers.91 Several different types of associating polymers have
been studied. These include the hydrophobically modified polyacrylamide
(HMPAM)92, ethoxylated urethane (HEUR)93, hydroxyethylcellulose (HMHEC)94
and alkali-swellable emulsion (HASE)95. Also combinations of associative
polymers with surfactants have been developed for EOR.96 It has been
demonstrated that the addition of small amounts of surfactants can increase
the viscosity of the aqueous solution containing hydrophobically modified
polymers significantly.94
Other polymers that possess interesting properties, such as high
molecular weight and intrinsic viscosity, have been developed for EOR and
are known as "rigid rod" water-soluble polymers.97 One study compared
hydrophobically modified polyacrylamide (HMPAM) with polyacrylamide (PAM)
in a simple core flood test and demonstrated that the residual resistance
factor (RRF, permanent reduction in the permeability of the formation due to
the adsorption of polymeric chains) after the polymer flood is much higher for
the HMPAM compared to PAM.98 All these modification strategies, together
with new kinds of water-soluble systems, have been extensively reported in
the literature and will be discussed in the next paragraph.
As mentioned earlier, a relatively new class of water-soluble polymers is
the one constituted by hydrophobically associative polymers.91 The first
hydrophobically associative polymers were synthesized almost fifty years
ago99,
100
, albeit for a different purpose than EOR. The research on these
polymers has been primarily fueled by the coating industry91, where
improvement in the rheology of the coating systems was required. During the
1980’s, when the oil crisis hit, a lot of research was performed on EOR. From
the many patents101-106 that have been filed during those years, it is evident
26
Chapter 1
that
this
accelerated
the
development
of
hydrophobically
associative
polymers for use in EOR applications.
Hydrophobically associative polymers contain, in most cases, a small
number of hydrophobic groups, i.e. 8-18 carbon atoms moieties107-110,
distributed along the main backbone111-113. These hydrophobic groups can be
distributed randomly or block-like
both ends
108, 126-136
92, 95, 107, 112, 114-125
, and coupled at one or
. Above a given polymer concentration (dependent on the
molecular structure) the hydrophobic groups associate, when the polymer is
dissolved
in
water,
to
intermolecular liaisons).92,
form
hydrophobic
93, 108, 109, 112, 137-140
micro-domains
(intra
or
These lead to an increase in
hydrodynamic volume, which in turn yields a polymer with a much better
thickening (higher viscosity112) capability compared to its non-associative
analogue.92
Depending
on
the
concentration,
intra-
or
intermolecular
associations as schematically illustrated in Figure 1.11, is detected.
Figure 1.11: Intra- and intermolecular associations137
When the hydrophobic elements are distributed in a block-like fashion along
the backbone of a water-soluble copolymer, the intramolecular associations
are stronger compared to randomly or discretely distributed hydrophobic
groups.108, 126
The temperature dependence of the solution viscosity is an interesting
property of hydrophobically modified polymers for EOR applications. It has
long been accepted that increasing the temperature of the polymer solution
will lead to a reduction in viscosity, probably due to the fact that an increase
in temperature implies a decrease of the association strength of the
hydrophobes.107, 138, 141-145 Increasing the temperature of the solution leads to
27
Introduction
a reduction of the solvent viscosity and hence an increase in the mobility of
the polymer chains while the solubility of the polymer will increase with
temperature.
However,
many
different
aqueous
systems
have
been
demonstrated to display an increase in viscosity upon increasing the
temperature.146-159 Indeed, a temperature increase results in a decrease of
the solubility of one of the components (Lower Critical Solution Temperature
[LCST]-groups) of the polymers. These less soluble components selfaggregate with the hydrophobic groups of the polymers, which leads to an
increase in viscosity.124 Several researchers have proposed a concept for
thermo-associative polymers based on the switch, i.e. the transition between
low and high temperature, of the polymers characterized by a lower critical
solution temperature.145,
156, 157
The concept involves a highly water-soluble
polymer containing blocks or side chains of LCST groups. Upon heating of the
polymer solution, these LCST groups will segregate. A schematic illustration
of this behavior has been presented by Hourdet and coworkers 156 and is
depicted in Figure 1.12.
Figure 1.12: Thermal induced microdomains156
Above the critical overlap chain concentration this transition will lead to an
increase in the viscosity of the solution through intermolecular associations.
Fundamental research on different polymers, in binary (polymer-water)
and ternary (polymer-water-surfactant) systems, has been performed using
different techniques which include
114
NMR
,
23
164,
Na-NMR
165
,
19
13
C-NMR160-163 (solution or solid-state), 1H-
F-NMR166,
NMR
self-diffusion131,
potentiometry171-173, Static and Dynamic Laser Light Scattering133,
168, 172-177
28
, UV-Spectroscopy for polymers bearing chromophores
167-170
135, 136, 151,
92, 115, 116, 178-
Chapter 1
186
,
Small-Angle
Neutron
Transfer (NRET) studies
191-194
Scattering
183, 188-190
and surface tension136,
(SANS)187,
Non-Radiative
Energy
, Size Exclusion Chromatography (SEC)175,
138, 167, 168, 170
. Several different associative
hydrophobically modified polymers have been developed which include
polyacrylamides (HMPAM), ethoxylated urethanes (HEUR), alkali swellable
emulsions
(HASE),
and
polysaccharides
(HM-polysaccharides).
Their
synthesis, rheological behavior and adsorption on surfaces has been
thoroughly discussed and the reader is kindly reverted to a recent review
paper195 covering these aspects.
As evident from the discussion195 a lot of research has been performed on
different water-soluble polymers capable of enhancing the viscosity of the
subsequent polymer solution. Although many breakthroughs have been
accomplished in the application for personal care products and the paint
industry, the application of water-soluble polymers for EOR techniques is
limited. With the correct specification of the required product properties for
EOR applications the optimal water-soluble polymer can in principle be
designed. However in order to be successful a toolbox is needed that
correlates the molecular design of the polymer, i.e. its topology and chemical
composition, to the properties that are subsequently obtained and to the
response to external stimuli (i.e. pH, ionic strength, temperature, salt and
surfactant). For EOR applications there are several parameters that can affect
the rheological properties of the polymeric solutions employed: solution pH,
ionic
strength,
temperature,
electrolyte
concentration,
shear
and
the
presence of bacteria. Depending on the polymeric system, these parameters
can have either a positive or negative effect on the rheological properties as
discussed in the following.
Polyelectrolytes, bearing only one charge, loose their solution viscosity as
the concentration of electrolytes, temperature and shear increase. However
they are resistant to changes in the pH of the solution. Zwitterionic polymers
(polyampholytes) are pH(ionic strength)-responsive polymers. The rheology
of these polymers can be tuned to the desired rheological properties by
changing the pH/ionic strength of the solution. In addition the temperature
dependence of the solution viscosity is minimal. Furthermore these polymers
are capable of enhancing the solution viscosity in salt environments up to
relevant concentrations (e.g. values higher than 100000 ppm TDS), which is
common in many oil reservoirs. At high shear though, a loss in solution
viscosity is observed.
Another category of polymers that are resistant to high temperatures are
the LCST polymers. However, these polymers are susceptible to phase
separation, which leads to a significant drop in the solution viscosity.
29
Introduction
Amphiphilic polymers do not bear charges and are therefore resistant to the
presence of electrolytes. Increase in the temperature leads to a loss in the
solution viscosity. The rheology of amphiphilic polymers is dependent on the
applied shear. At high shear rates, the solution viscosity is low and at low
shear the solution viscosity is high. This behavior is reversible, which for EOR
applications should be a beneficial property. Nevertheless, the distribution of
the hydrophobic groups is crucial in obtaining the required properties. If the
hydrophobic moieties are block-like distributed a stronger enhancement of
the solution viscosity is obtained compared to when the situation where the
distribution is random. The hydrophobicity of the groups also affects the
thickening capability of the polymer with higher hydrophobicity groups having
higher thickening capacity.
Combinations of the properties can also be achieved with amphiphilic
polyelectrolytes or polyampholytes. These polymers bear one or two
(different) charges and hydrophobes. The rheological properties of these
polymeric systems can be tailored by careful molecular design, i.e. the ratio
of the different monomers, of the polymers. Dependent on the molecular
design polymers can be obtained, which are pH-responsive, temperature
insensitive or salt resistant. For certain conditions, i.e. low pH and high
electrolyte concentration, carboxylic groups lead to better responsive
polymers compared to when sulphonic groups are used.
By using the above mentioned properties of the different polymeric
systems, a general trend can be identified where different systems can be
used for different applications. All the polymers have advantages and
disadvantages when compared to each other. These are presented in Table
1.1.
As can be observed in Table 1.1 many of the different polymeric systems
have one or two parameters, which affect their properties in a negative
manner. Therefore, it is crucial to correctly formulate the required properties
of a given application in order to design the correct polymeric system.
It is clear that the successful design of new water-soluble polymers for a
given application requires an integral
multiscale
and
multidisciplinary
approach. Proper definition of the required product properties is in this case
crucial. Knowledge of polymer chemical architecture (and thus of the
synthetic methods used) must be conceptually linked to the desired product
application requirements. In this case viscosity measurements under different
shear conditions are of paramount importance and should be ideally
correlated
with
the
“nature”
(i.e.
architecture
composition) of the corresponding water solution.
30
and
overall
chemical
Chapter 1
Table 1.1: Advantages and disadvantages of the different polymeric systems
Parameters
Polymer type
Polyelectrolyte
Polyampholyte
LCST polymer
Amphiphilic
Amphiphilic polyelectrolyte
Amphiphilic polyampholyte
High
pH
shear
+++-
+
++
++++
Ionic
High
strength temperature
High electrolyte
concentration
Bacteria
+
++
+-
-++
+
++
+
+
+
+
+
+
++
++
++
++
++
++
++
++
++
+-
+-
++-
+-
+++-
++
++-/++
++
++
++
--
Backbone type
Acrylamide (AM)
Ethylene oxide (EO)
MMA-MA-EO
Cellulosic
The influence of external parameters (e.g. pH, temperature etc.) on the
rheological behavior must be coupled with an in depth knowledge of the
relationship between the chemical structure and architecture of the polymer
and the rheological behavior. In this respect, an overall correlation cannot be
defined only as a function of the chemical/molecular structure. Rheological
properties will be affected by a combination of external parameters and the
chemical nature and molecular structure of the polymer. For instance, the
rheological properties of an aqueous solution of an amphiphilic polyelectrolyte
are similar to those of an unmodified analogue without amphiphilic moieties.
However, in the presence of salt a markedly different behavior is observed.
The solution viscosity of the unmodified polymer decreases with increasing
salt concentration, whereas the solution viscosity of the aqueous solution
containing the amphiphilic polyelectrolyte is not affected. Another good
example is the thermal performance of amphiphilic polymers: the rheological
properties of an aqueous solution containing the amphiphilic polymer or its
unmodified analogue, e.g. without the NIPAM monomer, are quite similar.
However, when exposing both solutions to higher temperatures significant
differences arise. The effect of temperature on the solution viscosity of an
aqueous solution containing the amphiphilic polymer is limited, whereas the
viscosity of the unmodified analogue changes significantly.
Although there are many different water-soluble polymers capable of
enhancing the solution viscosity, it is important to understand their
differences and analogies. Different polymers display in general differences in
the agglomeration principles governing their water behavior. On a molecular
level the basic principle is indeed quite general: the presence of relatively
weak
inter(macro)molecular
interactions
(e.g.
hydrophobic
association,
31
Introduction
hydrogen bonding etc.) factually, albeit “virtually”, increases the molecular
weight of the polymer coils. As a consequence the solution viscosity
increases. However, a careful balance must be observed here since
predominantly weak interactions (both in terms of strength and number
thereof) do not result in observable rheological differences while excessively
strong ones might compromise the solubility of the system by leading, for
example, to gel formation.
1.5. Thickening capabilities
As mentioned before, the main purpose of water-soluble polymers is to
control the rheological properties of the solution. In all cases an increase in
the solution viscosity is required. The thickening capability of polymers is the
ability to increase the solution viscosity by the addition of a determined
amount of polymer. Polymers with a high thickening capability can increase
the solution viscosity significantly even at relatively low concentration
(typically in the order of few ppm). The mechanism behind the enhancement
of the solution viscosity can be conceptually divided into several different
possibilities (Figure 1.13).
Figure 1.13: Different methods to increase the solution viscosity of aqueous solutions.
Traditionally the two main types of viscosity enhancement using polymers
were to either increase the concentration (Fig. 1.13, A) or the molecular
weight (Fig. 1.13, B) of the polymer. The increase in concentration leads to
32
Chapter 1
more entanglements and thus a higher viscosity. Higher molecular weights
could be obtained when taking into account the continuous improvements in
the different polymerization techniques. With higher molecular weight
polymers less material is required to reach a predetermined viscosity given
the higher hydrodynamic radius of the polymer coils in water. This is testified
by the well-known Mark-Houwink36,
viscosity (
196
(M-H) equation relating the intrinsic
) to the molecular weight ( ).
(1.5)
where
and
are constants characteristic for a given polymer-solvent
system at a specific temperature.
Another way of improving the thickening capability of a polymer is to
introduce charged moieties along the backbone of the polymer (Fig. 1.13, C).
The charged nature of the backbone will lead to electrostatic repulsions thus
increasing the hydrodynamic volume of the polymer coil which is synonymous
to a higher solution viscosity. The persistence length36 ( ) of a polymer chain
(closely related to the radius of gyration,
) is a function of the persistence
length without electrostatic interactions ( ) plus a contribution (
) related to
the electrostatic repulsions (equation 1.6).
(1.6)
where
= the charge,
permittivity of vacuum,
= relative dielectric permittivity,
= Boltzmann’s constant,
Debye screening length, and
= electrical
= temperature,
= the
= the distance between the charges.
Increasing the amount of charges will lead to a lower distance between the
charges ( ) and thus a higher
value. The persistence length ( ) is then
higher leading to a higher thickening capability.
As mentioned before, a relatively new approach is the introduction of
hydrophobic groups (Fig. 1.13, D), either randomly distributed or block
copolymers. In a water solution these hydrophobic groups have the tendency
to aggregate with each other. Depending on the concentration either intra or
intermolecular hydrophobic associations arise (vida supra).
To illustrate the differences between the polymers studied so far, a
comparison between the thickening ability of the polymers at a fixed
concentration (1 wt.%) is presented in Figure 1.14. The thickening ability of
polyacrylamide based polymers lies in either the high molecular weight or
electrostatic repulsions between charged moieties.195 The same applies for
33
Introduction
biopolymers; however chain rigidity also plays an important role, especially in
the case of xanthan gum.195 Hydrophobically modified polymers though,
achieve
high
solution
viscosity
due
to
intermolecular
hydrophobic
interactions.195 The molecular weights of these polymers are usually much
lower than that of PAM based polymers and biopolymers.195
10
10
Viscosity
Molecular weight
9
8
6
5
4
4
2,1
2
2
2
6
6
Molecular weight (x10 g/mol)
Viscosity (Pa.s)
8
0,56
0,2
0,035
0
AM
HP
M
PA
HM
M
PA
UR
HE
SE
HA
0,08
HM
A
PA
HM
u
ell
-C
0
sic
lo
h
nt
Xa
an
gu
m
Figure 1.14: Thickening abilities of different polymers, the solution viscosity (at
=
10 s-1) of the polymer solution (1 wt.%) with corresponding molecular weight.
Another method, as will be presented by the research in this thesis, is the
introduction of branches, as long as the corresponding solutions are not in
the dilute regime. It has been demonstrated that in the melt state, branched
polymers display a higher zero shear rate viscosity (0) compared to their
linear analogues.197-215 This behavior has also been demonstrated in
concentrated solutions.216 Polymer solutions can be classified into three
different states. In the dilute regime the polymer concentration is so low that
no overlap between the polymeric chains is present. In the semi-dilute
regime overlap starts to arise and in the entangled (concentrated solutions)
state no separate polymeric coils can be distinguished. In dilute solutions it
has been demonstrated that the hydrodynamic volume of branched polymers
is lower than their linear analogues.217 Therefore, if no entanglements are
present a lower solution viscosity is observed. It is worth noting that polymer
solutions used for EOR are, in most cases, semi-dilute solutions. Up to date
and to the best of the author’s knowledge, the effect of the macromolecular
34
Chapter 1
architecture (i.e. presence of branches) on the thickening ability of PAMbased polymers in semi-dilute solutions has not yet been reported and
constitutes therefore a relevant novelty aspect of the present work. One
major hurdle that must be overcome in order to demonstrate this is the
controlled polymerization of acrylamide. This is necessary in order to be able
to correctly specify the macromolecular architecture.
1.6. End-use requirements for polymers in EOR
Although many different water soluble polymers have been identified as
possible new chemical agents for polymer flooding it remains crucial that
certain requirements are met by these polymers for them to be successfully
applied in EOR. Indeed, some are trivial such as solubility in water and
capability of increasing the solution viscosity at low concentrations. However,
other requirements are less obvious but yet as important as the obvious
ones. The following list of requirements (arbitrary order) for polymers to be
considered for EOR has been established in cooperation with Shell:
-
Ability to withstand high salt concentration, >20000 ppm total
dissolved solids (TDS)
-
Applicable at high temperature, 80 – 120 °C
-
Hydrolysis resistant
-
Applicable for oil viscosity values of 1 – 200 cP
-
Usable in low permeable reservoirs, 1 – 50 mD
-
Stability of the polymer solution (in terms of solution viscosity) for a
few years
The majority of the oil reservoirs in the world are injected with brine to
increase the recovery rate of the field. However, given the anionic character
of the currently used HPAM, a significant reduction in solution viscosity is
observed when the polymer is dissolved in salt water. Therefore, new
polymers for EOR should be able to resist the presence of salt without a
significant reduction in the solution viscosity. When taking into account
applications where the TDS reaches values above 20000 ppm, a complete
screening of the charges on the anionic HPAM is observed. The solution
viscosity drops significantly, by almost two orders of magnitude.218,
219
Significant progress has been booked in identifying polymers that display
resistance to the presence of salts such as polyampholytes 220-227 (bearing
both positive and negative charges), polyelectrolyte amphiphilic 120, 161, 180, 181,
188, 228-237
, zwitterionic amphiphilics238-241 or amphiphilics107, 108,
114, 137, 138, 143,
35
Introduction
242-254
. Polymers specifically prepared for the use in high salinity oil reservoirs
have also been developed.255,
256
Nevertheless, other limitations arise with
the use of such polymers. Therefore, progress in new polymers capable of
coping with the presence of salt without introducing new limitations is still
required.
Another important parameter of many of the world’s oil reservoirs is the
high temperature (in many cases > 50 °C). The general reduction in solution
viscosity with temperature, due to the increased mobility of the polymeric
chains in solution, is also observed for the currently used EOR polymers.
Nowadays in general, oil reservoirs with a temperature higher than 50 °C are
not considered for polymer flooding. Given the high polymer concentration
required to match the viscosity of the oil (at high temperatures), applying
polymer flooding would not be economically beneficial. Nonetheless, a safe
limit (without chemical degradation) of ≤75 °C has been defined for
polyacrylamides257 and ≤80 °C for xanthan gum84. Significant progress has
been accomplished on thermoresponsive water-soluble polymers.258 A couple
of different monomers, displaying a LCST behavior, have been copolymerized
with acrylamide for EOR applications.255,
259, 260
However, systematic studies
on the effect of the “size” of the LCST on the rheological properties remains
elusive given the difficulties in controlled synthesis of acrylamide based
polymers.
The application of acrylamide based water-soluble polymers in oil
reservoirs with temperatures above 50 °C (and in the presence of a base)
creates a new problem. Hydrolysis (Scheme 1.1) of polyacrylamide is
extensive at elevated temperatures. Due to the more ionic character of the
polymer, this leads initially to an increase in solution viscosity. However, in
the presence of divalent ions, such as Ca2+ and/or Mg2+, precipitation of the
polymer is observed with a significant loss in solution viscosity and
injectivity257.
Scheme 1.1: Alkaline hydrolysis of polyacrylamide
Copolymerization of acrylamide with hydrolysis resistant monomers has been
extensively
36
investigated.
Homopolymers
of
sodium-2-acrylamido-2-
Chapter 1
methylpropane
sulfonate
(NaAMPS)
are
resistant
to
hydrolysis
at
temperatures up to 120 °C.257 However, copolymers of acrylamide and
NaAMPS hydrolyze readily, depending on the molar ratio between acrylamide
and NaAMPS, at these temperatures leading to precipitation in the presence
of divalent ions.257 Another monomer that has been extensively studied to
improve the hydrolysis resistance of the acrylamide-based polymers is
vinylpyrrolidone.34, 261-263 The homopolymer of vinylpyrrolidone is resistant to
hydrolysis, but, more importantly and unlike NaAMPS, the vinylpyrrolidone
moieties also seem to “protect” the acrylamide units from hydrolysis in the
corresponding copolymers.261-263 Nevertheless, one of the major advantages
of acrylamide can still not be overcome by these monomers and that is its
low price. The introduction of hydrolysis resistant moieties increases the price
of the polymer and ultimately changes the whole economic picture of polymer
based flooding.
The application of polymer flooding depends, among other parameters,
on the viscosity of the oil in the reservoir. The viscosity of oil varies
significantly, from water like consistency up to bitumen (tar sands). It is
desirable to be able to apply polymer flooding for oil viscosities up to 200 cP.
The higher the oil viscosity the more polymer is required to match the
viscosity of the displacing fluid (water). The higher the required polymer
concentration, the less attractive (higher polymer costs) the oil reservoir is
for the application of polymer-based floods. Therefore extensive research has
been done195, and is ongoing, to improve the thickening capabilities of water
soluble polymers.
The permeability of a reservoir is, as mentioned before, the ability of a
fluid to pass through the porous media. A great number of oil reservoirs
around the world have porous media whose permeabilities are lower than 50
mD, the so-called carbonate reservoirs. Currently used polymers are high
molecular weight polymers and will block the pores of such low permeable
reservoirs. Adsorption to the rock surface24 and bridging264-266 by the polymer
chains lead to injectivity loss. Lab core flood testing confirms this by showing
a significant increase of the pressure over the core sample increases as more
and more polymer solution is flowed through.265-267
1.7. Aim and scope of this thesis
The challenges facing the supply of energy are briefly discussed and oil
recovery is introduced along with enhanced oil recovery in Chapter 1.
Currently used polymers are discussed along with their benefits and
limitations. The state of the art of polymers for enhanced oil recovery is
37
Introduction
thoroughly reviewed and new possibilities for other (new) polymers that can
improve on the limitations of currently used polymers are discussed. In
addition, the product specifications are identified and serve as a guideline for
the design of new polymers.
Chapter
2
polymerization
reports
(ATRP)
the
of
first
successful
acrylamide
in
atomic
water at
transfer
radical
room temperature.
Polyacrylamide with molecular weights higher than 150 000 g/mol with
dispersities as low as 1.39 can be prepared. Evidence for the “living”
character of the synthetic method is provided by; good concordance between
the theoretical molecular weight and the actual molecular weight, low
dispersities, linear increase in molecular weight with conversion, and
successful chain extension.
The successful ATRP of acrylamide in water is further expanded to
prepare branched polyacrylamide in Chapter 3. Star (4-arm) polyacrylamides
are prepared through the use of a commercial tetra-functional initiator. The
preparation of comb-like (12-arm) polyacrylamides is also discussed where
novel macroinitiators based on alternating aliphatic polyketones are used.
Evidence for the controlled preparation is provided along with preliminary
results on the rheological properties of aqueous solutions containing the
architectural different polyacrylamides.
Architectural different polyacrylamides are prepared in Chapter 4. The
dependence of the rheological properties on the molecular architecture of the
polymer is investigated. The solution viscosity of an aqueous solution is
heavily dependent on the architecture of the polyacrylamide above the critical
overlap concentration. Both an increase and a decrease in the solution
viscosity can be achieved by the introduction of branches, depending on the
number of arms. In addition, the visco-elastic response of an aqueous
solution
containing
the
polyacrylamides
can
also
be
manipulated
by
controlling the number of branches.
In Chapter 5 the synthesis of block copolymers of AM and NIPAM is
reported. PAM-b-PNIPAM block copolymers were prepared through ATRP in
water at room temperature. The block lengths of both moieties were varied in
order to obtain polymers with varying hydrophilic-lyophilic balances (HLB).
The solution properties, i.e. CMC and solution viscosity as a function of
temperature, of these polymers was correlated to the solubility parameter
).
The synthesis of different, both chemically and architecturally, thermoresponsive polymers based on acrylamide is discussed in Chapter 6. The
solution properties of the random and block copolymers are investigated. The
importance of chemical structure on the thermo-responsiveness of the
38
Chapter 1
polymers is demonstrated. In addition, the effect of the chemical structure
(block or random) and molecular architecture on the surface tensions is
discussed. In general, the block copolymers tend to precipitate from the
solution upon heating while the random copolymers stayed in solution. The
strength of the hydrophobic interactions plays a crucial role in the observed
behavior.
In Chapter 7 the flow properties (through porous media) of the different
polymers is evaluated. Initial filter tests are performed in order to predict the
polymers’ ability to permeate sandstone or carbonate core samples. The ease
of passage through filters of different pore size is independent of the number
of branches. In addition, the recovery of oil out of core samples was
evaluated for the different polymers. The efficiency of the oil recovery
depends on both the chemical structure and the architecture of the polymers.
In addition the recovery of residual oil in a two dimensional flow cell was also
investigated. It is demonstrated that both the chemical structure and the
architecture of the polymer employed affects the amount of residual oil that
can be recovered through the use of polymers in EOR.
To conclude the thesis, a thorough discussion on the problems that have
been tackled and (partially) solved through the polymers developed in this
project is presented in Chapter 8. In addition, polymers applied in EOR often
are injected in solutions containing alkali, leading to chemical degradation of
polyacrylamide. Hydrolysis resistant polymers are synthesized through ATRP.
It is demonstrated that branched polymers can be prepared using hydrolysis
resistant moieties. In addition, the resistance of branched polyacrylamide
against
alkaline
hydrolysis
is
improved
compared
to
that
of
linear
polyacrylamide. Finally an overview of biopolymers that might have potential
for application in EOR is presented.
1.8. Acknowledgement
This work is part of the Research Programme of the Dutch Polymer
Institute DPI, Eindhoven, the Netherlands, projectnr. #716.
1.9. References
1. United Nations, Department of Economic and Social Affairs, Population Division
2011.
2. US Energy Information Administration 2011.
3. International Energy Agency IEA 2012.
4. International Energy Agency IEA 2011.
5. Maugeri, L. Belfer Center for Science and International Affairs, Harvard Kennedy
School 2012.
39
Introduction
6. Thomas, S. Oil Gas Sci. Technol. 2008, 1, 9.
7. Simon, R. Phys. Chem. Earth 1981, 447.
8. Bear, J. Dynamics of Fluids in Porous Media; Dover Publications Inc.: United States
of America, 1988; Vol. 2, pp 784.
9. Lake, L. W. Enhanced Oil Recovery; Prentice-Hall Inc.: Englewood Cliffs, NJ, 1989;
Vol. 1, pp 550.
10. Taber, J. J.; Martin, F. D.; Seright, R. S. SPE 1997, SPE 35385.
11. Alvarado, V.; Manrique, E. Energies 2010, 9, 1529.
12. Homsy, G. M. Annu. Rev. Fluid Mech. 1987, 271.
13. Stanislav, J. F. Rheol. Acta 1982, 4-5, 564.
14. Wang, D.; Cheng, J.; Yang, Q.; Gong, W.; Li, Q.; Chen, F. SPE 2000, SPE-63227MS.
15. Xia, H.; Ju, Y.; Kong, F.; Wu, J. SPE 2004, SPE-88456-MS.
16. Xia, H.; Wang, D.; Wang, G.; Wu, J. Petrol. Sci. Technol. 2008, 4, 398.
17. Zhang, L.; Yue, X. J. Cent. South Univ. T. 2008, 84.
18. Zhang, L.; Yue, X.; Guo, F. Pet. Sci. 2008, 1, 56.
19. Zhang, Z.; Li, J.; Zhou, J. Transport Porous Med. 2011, 1, 229.
20. Wang, D.; Xia, H.; Liu, Z.; Anda, Q.; Yang, Q. SPE 2001, SPE-68723-MS.
21. Yin, H.; Wang, D.; Zhong, H. SPE 2006, SPE-101950-MS.
22. Han, D.; Yang, C.; Zhang, Z.; Lou, Z.; Chang, Y. J. Petrol. Sci. Eng. 1999, 1-3,
181.
23. Li, G.; Zhai, L.; Xu, G.; Shen, Q.; Mao, H.; Pei, M. J. Disper. Sci. Technol. 2000, 4,
367.
24. Sorbie, K. S. Polymer-improved oil recovery; CRC Press: Boca Raton, FL, 1991; .
25. Levitt, D. B.; Pope, G. A. SPE 2008, SPE-113845.
26. Gaillard, N.; Giovannetti, B.; Favero, C. SPE 2010, SPE-129756.
27. Wu, Y.; Wang, K. -.; Hu, Z.; Bai, B.; Shuler, P.; Tang, Y. SPE 2009, SPE-124257.
28. Pancharoen, M.; Thiele, M. R.; Kovscek, A. R. SPE 2010, SPE-129910.
29. Buchgraber, M.; Clemens, T.; Castanier, L. M.; Kovscek, A. R. SPE 2009, SPE122400.
30. Sabhapondit, A.; Borthakur, A.; Haque, I. J. Appl. Polym. Sci. 2003, 12, 1869.
31. Sabhapondit, A.; Borthakur, A.; Haque, I. Energ. Fuel 2003, 3, 683.
32. Song, H.; Zhang, S.; Ma, X.; Wang, D.; Yang, J. Carbohyd. Polym. 2007, 1, 189.
33. Vega, I.; Sanchez, L.; D'Accorso, N. React. Funct. Polym. 2008, 1, 233.
34. Morgan, S. E.; Mccormick, C. L. Progr. Polym. Sci. 1990, 1, 103.
35. Borthakur, A.; Rahman, M.; Sarmah, A.; Subrahmanyam, B. Res. Ind. 1995, 2,
90.
36. Stokes, R. J.; Evans, D. F. Fundamentals of interfacial engineering; Wiley-VCH:
New York, 1997; .
37. Shupe, R. D. J. Petrol. Technol. 1981, 8, 1513.
38. Fuoss, R. M. J. Polym. Sci. 1948, 4, 603.
39. Fuoss, R. M.; Strauss, U. P. J. Polym. Sci. 1948, 2, 246.
40. Fuoss, R. M.; Strauss, U. P. J. Polym. Sci. 1948, 4, 602.
41. Aitkadi, A.; Carreau, P. J.; Chauveteau, G. J. Rheol. 1987, 7, 537.
42. Dupuis, D.; Lewandowski, F. Y.; Steiert, P.; Wolff, C. J. Non-Newton. Fluid 1994,
11.
43. Ellwanger, R. E.; Jaeger, D. A.; Barden, R. E. Polym. Bull. 1980, 6-7, 369.
44. Sukpisan, J.; Kanatharana, J.; Sirivat, A.; Wang, S. J. Polym. Sci. Pol. Phys. 1998,
5, 743.
45. Khune, G.; Donaruma, L.; Hatch, M.; Kilmer, N.; Shepitka, S.; Martin, F. J. Appl.
Polym. Sci. 1985, 2, 875.
46. McCormick, C. L.; Neidlinger, H. H.; Hester, R. D.; Wildman, G. C. Surface
phenomena in enhanced oil recovery; Shah, D. O., Ed.; Plenum: New York, 1981;
pp 741.
47. Dautzenberg, H. Macromolecules 1997, 25, 7810.
48. Peng, S.; Wu, C. Macromolecules 1999, 3, 585.
40
Chapter 1
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
Ohmine, I.; Tanaka, T. J. Chem. Phys. 1982, 11, 5725.
BenJar, P. Y.; Wu, Y. S. Polymer 1997, 10, 2557.
Cook, R. L.; King, H. E.; Peiffer, D. G. Macromolecules 1992, 11, 2928.
Martin, F. D.; Hatch M.J.; Shepitka, J. S.; Ward, J. S. SPE 1983, SPE-11786.
Martin, F. D. SPE 1986, SPE-12651.
Lewandowska, K. J. Appl. Polym. Sci. 2007, 4, 2235.
Hu, Y.; Wang, S.; Jamieson, A. Macromolecules 1995, 6, 1847.
Seright, R. S.; Seheult, M.; Talashek, T. SPE 2009, 5, SPE-115142-PA.
Chauveteau, G. SPE 1981, SPE-10060.
Ferguson, J.; Walters, K.; Wolff, C. Rheol. Acta 1990, 6, 571.
Choplin, L.; Sabatie, J. Rheol. Acta 1986, 6, 570.
Shepitka, J. S.; Case, C. E.; Donaruma, L. G.; Hatch, M. J.; Kilmer, N. H.; Khune,
G. D.; Martin, F. D.; Ward, J. S.; Wilson, K. V. J. Appl. Polym. Sci. 1983, 12,
3611.
Durst, F.; Haas, R.; Interthal, W. J. Non-Newton. Fluid 1987, 2, 169.
James, D. F.; Mclaren, D. R. J. Fluid Mech. 1975, AUG26, 733.
Magueur, A.; Moan, M.; Chauveteau, G. Chem. Eng. Commun. 1985, 1-6, 351.
Silberberg, A.; Mijnlief, P. F. J. Polym. Sci. Pol. Phys. 1970, 7, 1089.
Quadrat, O. Adv. Colloid Interf. Sci. 1985, 1, 45.
Ohya, S.; Matsuo, T. J. Colloid Interf. Sci. 1979, 3, 593.
Bradna, P.; Quadrat, O.; Dupuis, D. Colloid Polym. Sci. 1995, 5, 421.
Bradna, P.; Quadrat, O.; Dupuis, D. Colloid Polym. Sci. 1995, 7, 642.
Bradna, P.; Quadrat, O.; Titkova, L.; Dupuis, D. Acta Polym. 1997, 10, 446.
Quadrat, O.; Bradna, P.; Dupuis, D.; Wolff, C. Colloid Polym. Sci. 1992, 11, 1057.
Leela, J. K.; Sharma, G. Bioprocess Eng. 2000, 6, 687.
Nasr, S.; Soudi, M. R.; Haghighi, M. PJBS 2007, 17, 3010.
Garcia-Ochoa, F.; Santos, V. E.; Casas, J. A.; Gomez, E. Biotechnol. Adv. 2000, 7,
549.
Becker, A.; Katzen, F.; Puhler, A.; Ielpi, L. Appl. Microbiol. Biot. 1998, 2, 145.
Morris, E. R.; Rees, D. A.; Young, G.; Walkinshaw, M. D.; Darke, A. J. Mol. Biol.
1977, 1, 1.
Norton, I. T.; Goodall, D. M.; Frangou, S. A.; Morris, E. R.; Rees, D. A. J. Mol. Biol.
1984, 3, 371.
Dentini, M.; Crescenzi, V.; Blasi, D. Int. J. Biol. Macromol. 1984, 2, 93.
Holzwarth, G. Biochemistry 1976, 19, 4333.
Chen, C. S. H.; Sheppard, E. W. Polym. Eng. Sci. 1980, 7, 512.
Yoshida, T.; Tanner, R. D. In Bioproducts and bioprocess. Springer-Verlag: Berlin,
1993; .
Richardson, R. K.; Rossmurphy, S. B. Int. J. Biol. Macromol. 1987, 5, 257.
Kierulf, C.; Sutherland, I. W. Carbohyd. Polym. 1988, 3, 185.
Lambert, F.; Rinaudo, M. Polymer 1985, 10,.
Seright , R. S.; Henrici, B. J. SPE 1990, 1, SPE-14946-PA.
Wellington, S. L. Soc. Petrol. Eng. J. 1983, 6, 901.
Ash, S. G.; Clark-Sturman, A. J.; Calvert, R.; Nisbet, T. M. SPE 1983, SPE-12085.
Sutherland, I. W. J. Appl. Bacteriol. 1982, 3, 385.
Cadmus, M. C.; Jackson, L. K.; Burton, K. A.; Plattner, R. D.; Slodki, M. E. Appl.
Environ. Microb. 1982, 1, 5.
Bragg, J. R.; Maruca, S. D.; Gale, W. W.; Gall, L. S.; Wernau, W. C.; Beck, D.;
Goldman, I. M.; Laskin, A. I.; Naslund, L. A. SPE 1983, SPE-11989.
Hou, C. T.; Barnabe, N.; Greaney, K. J. Ind. Microbiol. 1986, 1, 31.
Taylor, K. C.; Nasr-El-Din, H. A. J. Petrol. Sci. Eng. 1998, 3-4, 265.
Hill, A.; Candau, F.; Selb, J. Macromolecules 1993, 17, 4521.
Annable, T.; Buscall, R.; Ettelaie, R.; Whittlestone, D. J. Rheol. 1993, 4, 695.
Panmai, S.; Prud'homme, R. K.; Peiffer, D. G. Colloid Surface A 1999, 1-2, 3.
English, R. J.; Laurer, J. H.; Spontak, R. J.; Khan, S. A. Ind. Eng. Chem. Res.
2002, 25, 6425.
41
Introduction
96. Biggs, S.; Selb, J.; Candau, F. Langmuir 1992, 3, 838.
97. Sarkar, N.; Kershner, L. D. J. Appl. Polym. Sci. 1996, 2, 393.
98. Maia, A. M. S.; Borsali, R.; Balaban, R. C. Mat. Sci. Eng. C-Bio S. 2009, 2, 505.
99. Dubin, P. L.; Strauss, U. P. J. Phys. Chem. 1970, 14, 2842.
100. Dubin, P.; Strauss, U. P. J. Phys. Chem. 1967, 8, 2757.
101. Schwab, F. C.; Sheppard, E. W.; Chen, C. S. H. US Patent 4110232, 1978.
102. Landoll, L. M. US Patent 4529523, 1984.
103. Bock, J.; Pace, S. J.; Schulz, D. N. US Patent 4709759, 1987.
104. Bock, J.; Siano, D. B.; Pace, S. J. Canada Patent 1300362, 1992.
105. Bock, J.; Valint, P. L.; Pace, S. J. US Patent 4702319, 1987.
106. Evani, S. US Patent 4814096, 1989.
107. Mccormick, C. L.; Nonaka, T.; Johnson, C. B. Polymer 1988, 4, 731.
108. Lara-Ceniceros, A. C.; Rivera-Vallejo, C.; Jimenez-Regalado, E. J. Polym. Bull.
2007, 2, 425.
109. Tam, K. C.; Jenkins, R. D.; Winnik, M. A.; Bassett, D. R. Macromolecules 1998,
13, 4149.
110. Tam, K. C.; Seng, W. P.; Jenkins, R. D.; Bassett, D. R. J. Polym. Sci. Pol. Phys.
2000, 15, 2019.
111. Glass, J. E. In Polymers in aqueous media: performance through association.
Advances in chemistry series 223; American Chemical Society: Washington, DC,
1989; .
112. Feng, Y. J.; Billon, L.; Grassl, B.; Khoukh, A.; Francois, J. Polymer 2002, 7, 2055.
113. Argillier, J. F.; Audibert, A.; Lecourtier, J.; Moan, M.; Rousseau, L. Colloid Surface
A 1996, 3, 247.
114. Volpert, E.; Selb, J.; Candau, F. Macromolecules 1996, 5, 1452.
115. Ezzell, S. A.; Hoyle, C. E.; Creed, D.; Mccormick, C. L. Macromolecules 1992, 7,
1887.
116. Ezzell, S. A.; Mccormick, C. L. Macromolecules 1992, 7, 1881.
117. Klucker, R.; Candau, F.; Schosseler, F. Macromolecules 1995, 19, 6416.
118. Dowling, K. C.; Thomas, J. K. Macromolecules 1990, 4, 1059.
119. Schulz, D. N.; Kaladas, J. J.; Maurer, J. J.; Bock, J.; Pace, S. J.; Schulz, W. W.
Polymer 1987, 12, 2110.
120. Peiffer, D. G. Polymer 1990, 12, 2353.
121. Iliopoulos, I.; Wang, T. K.; Audebert, R. Langmuir 1991, 4, 617.
122. Senan, C.; Meadows, J.; Shone, P. T.; Williams, P. A. Langmuir 1994, 7, 2471.
123. Tanaka, R.; Meadows, J.; Williams, P. A.; Phillips, G. O. Macromolecules 1992, 4,
1304.
124. Winnik, F. M. Macromolecules 1989, 2, 734.
125. Durand, A.; Hourdet, D. Polymer 1999, 17, 4941.
126. Maia, A. M. S.; Costa, M.; Borsali, R.; Garcia, R. B. Macromol. Symp. 2005, 217.
127. Lundberg, D. J.; Brown, R. G.; Glass, J. E.; Eley, R. R. Langmuir 1994, 9, 3027.
128. Winnik, M. A.; Yekta, A. Curr. Opin. Colloid In. 1997, 4, 424.
129. Huldén, M. Colloid Surface A 1994, 3, 263.
130. Xu, B.; Li, L.; Zhang, K.; Macdonald, P. M.; Winnik, M. A.; Jenkins, R.; Bassett,
D.; Wolf, D.; Nuyken, O. Langmuir 1997, 26, 6896.
131. Rao, B.; Uemura, Y.; Dyke, L.; Macdonald, P. M. Macromolecules 1995, 2, 531.
132. Alami, E.; Almgren, M.; Brown, W. Macromolecules 1996, 14, 5026.
133. Alami, E.; Almgren, M.; Brown, W.; François, J. Macromolecules 1996, 6, 2229.
134. Yekta, A.; Duhamel, J.; Brochard, P.; Adiwidjaja, H.; Winnik, M. A.
Macromolecules 1993, 8, 1829.
135. Maechling-Strasser, C.; Clouet, F.; Francois, J. Polymer 1992, 5, 1021.
136. Maechling-Strasser, C.; François, J.; Clouet, F.; Tripette, C. Polymer 1992, 3,
627.
137. Xie, X.; Hogen-Esch, T. E. Macromolecules 1996, 5, 1734.
138. Abu-Sharkh, B. F.; Yahaya, G. O.; Ali, S. A.; Kazi, I. W. J. Appl. Polym. Sci.
2001, 2, 467.
42
Chapter 1
139. Volpert, E.; Selb, J.; Candau, F. Polymer 1998, 5, 1025.
140. Candau, F.; Selb, J. Adv. Colloid Interfac. 1999, 2-3, 149.
141. Yahya, G. O.; Ali, A.; Al-Naafa, M. A.; Hamad, E. Z. J. Appl. Polym. Sci. 1995, 3,
343.
142. Yahya, G. O.; Hamad, E. Z. Polymer 1995, 19, 3705.
143. Kopperud, H. M.; Hansen, F. K.; Nyström, B. Macromol. Chem. Physic. 1998, 11,
2385.
144. Shaikh, S.; Asrof Ali, S.; Hamad, E. Z.; Abu-Sharkh, B. F. Polym. Eng. Sci. 1999,
10, 1962.
145. Hourdet, D.; Gadgil, J. A.; Podhajecka, K.; Badiger, M. V.; Brûlet, A.;
Wadgaonkar, P. P. Macromolecules 2005, 20, 8512.
146. Mortensen, K.; Brown, W.; Jørgensen, E. Macromolecules 1994, 20, 5654.
147. Nishinari, K.; Hofmann, K. E.; Moritaka, H.; Kohyama, K.; Nishinari, N. Macromol.
Chem. Physic. 1997, 4, 1217.
148. Doelker, E. Adv. Polym. Sci. 1993, 198.
149. Sarkar, N. J. Appl. Polym. Sci. 1979, 4, 1073.
150. Yoshioka, H.; Mikami, M.; Mori, Y.; Tsuchida, E. J. Macromol. Sci. Chem. 1994, 1,
113.
151. Nyström, B.; Walderhaug, H.; Hansen, F. K. Faraday Discuss. 1995, 335.
152. Wang, G.; Lindell, K.; Olofsson, G. Macromolecules 1997, 1, 105.
153. Loyen, K.; Iliopoulos, I.; Audebert, R.; Olsson, U. Langmuir 1995, 4, 1053.
154. Sarrazin-Cartalas, A.; Iliopoulos, I.; Audebert, R.; Olsson, U. Langmuir 1994, 5,
1421.
155. Greenhill-Hooper, M. J.; O'Sullivan, T. P.; Wheeler, P. A. J. Colloid Interf. Sci.
1988, 1, 77.
156. Bokias, G.; Hourdet, D.; Iliopoulos, I.; Staikos, G.; Audebert, R. Macromolecules
1997, 26, 8293.
157. Hourdet, D.; L'Alloret, F.; Audebert, R. Polymer 1997, 10, 2535.
158. Hourdet, D.; L'Alloret, F.; Audebert, R. Polymer 1994, 12, 2624.
159. de Vos, S.; Möller, M.; Visscher, K.; Mijnlieff, P. F. Polymer 1994, 12, 2644.
160. Hutchinson, B. H.; McCormick, C. L. Polymer 1986, 4, 623.
161. Magny, B.; Lafuma, F.; Iliopoulos, I. Polymer 1992, 15, 3151.
162. Newman, J. K.; Mccormick, C. L. Polymer 1994, 5, 935.
163. McCormick, C. L.; Chen, G. S.; Hutchinson, B. H. J. Appl. Polym. Sci. 1982, 8,
3103.
164. Newman, J. K.; McCormick, C. L. Macromolecules 1994, 18, 5123.
165. Newman, J. K.; McCormick, C. L. Macromolecules 1994, 18, 5114.
166. Furó, I.; Iliopoulos, I.; Stilbs, P. J. Phys. Chem. B 2000, 3, 485.
167. Walderhaug, H.; Hansen, F. K.; Abrahmsén, S.; Persson, K.; Stilbs, P. J. Phys.
Chem. 1993, 31, 8336.
168. Nyström, B.; Walderhaug, H.; Hansen, F. K. J. Phys. Chem. 1993, 29, 7743.
169. Abrahmsén-Alami, S.; Stilbs, P. J. Phys. Chem. 1994, 25, 6359.
170. Persson, K.; Abrahmsén, S.; Stilbs, P.; Hansen, F. K.; Walderhaug, H. Colloid
Polym. Sci. 1992, 5, 465.
171. McCormick, C. L.; Elliott, D. L. Macromolecules 1986, 3, 542.
172. Wang, C.; Tam, K. C.; Jenkins, R. D. J. Phys. Chem. B 2002, 6, 1195.
173. Wang, C.; Tam, K. C.; Jenkins, R. D.; Bassett, D. R. Phys. Chem. Chem. Phys.
2000, 9, 1967.
174. Dai, S.; Tam, K. C.; Jenkins, R. D.; Bassett, D. R. Macromolecules 2000, 19,
7021.
175. Chassenieux, C.; Nicolai, T.; Durand, D. Macromolecules 1997, 17, 4952.
176. Prochazka, K.; Martin, T. J.; Webber, S. E.; Munk, P. Macromolecules 1996, 20,
6526.
177. Procházka, K.; Martin, T. J.; Munk, P.; Webber, S. E. Macromolecules 1996, 20,
6518.
178. Biggs, S.; Selb, J.; Candau, F. Polymer 1993, 3, 580.
43
Introduction
179. Biggs, S.; Hill, A.; Selb, J.; Candau, F. J. Phys. Chem. 1992, 3, 1505.
180. Branham, K. D.; Davis, D. L.; Middleton, J. C.; McCormick, C. L. Polymer 1994,
20, 4429.
181. Branham, K. D.; Snowden, H. S.; McCormick, C. L. Macromolecules 1996, 1, 254.
182. Branham, K. D.; Shafer, G. S.; Hoyle, C. E.; McCormick, C. L. Macromolecules
1995, 18, 6175.
183. Kramer, M. C.; Steger, J. R.; Hu, Y.; McCormick, C. L. Macromolecules 1996, 6,
1992.
184. Kramer, M. C.; Welch, C. G.; Steger, J. R.; McCormick, C. L. Macromolecules
1995, 15, 5248.
185. Araujo, E.; Rharbi, Y.; Huang, X.; Winnik, M. A.; Bassett, D. R.; Jenkins, R. D.
Langmuir 2000, 23, 8664.
186. Kumacheva, E.; Rharbi, Y.; Winnik, M. A.; Guo, L.; Tam, K. C.; Jenkins, R. D.
Langmuir 1997, 2, 182.
187. Rufier, C.; Collet, A.; Viguier, M.; Oberdisse, J.; Mora, S. Macromolecules 2009,
14, 5226.
188. Smith, G. L.; McCormick, C. L. Macromolecules 2001, 4, 918.
189. Hu, Y.; Kramer, M. C.; Boudreaux, C. J.; McCormick, C. L. Macromolecules 1995,
21, 7100.
190. Hu, Y.; Smith, G. L.; Richardson, M. F.; McCormick, C. L. Macromolecules 1997,
12, 3526.
191. Cathébras, N.; Collet, A.; Viguier, M.; Berret, J. -. Macromolecules 1998, 4,
1305.
192. Kaczmarski, J. P.; Glass, J. E. Macromolecules 1993, 19, 5149.
193. May, R.; Kaczmarski, J. P.; Glass, J. E. Macromolecules 1996, 13, 4745.
194. Zhang, H.; Pan, J.; Hogen-Esch, T. E. Macromolecules 1998, 9, 2815.
195. Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Prog. Polym. Sci. 2011, 1558.
196. Hiemenz, P. C.; Lodge, T. P. Polymer Chemistry; CRC Press, Taylor & Francis
Group: Boca Raton, London, New York, 2007; .
197. Fetters, L. J.; Kiss, A. D.; Pearson, D. S.; Quack, G. F.; Vitus, F. J.
Macromolecules 1993, 4, 647.
198. Kraus, G.; Gruver, J. T. J. Polym. Sci. Part A 1965, 1PA, 105.
199. Mykhaylyk, O. O.; Fernyhough, C. M.; Okura, M.; Fairclough, J. P. A.; Ryan, A. J.;
Graham, R. Eur. Polym. J. 2011, 4, 447.
200. Robertson, C. G.; Roland, C. M.; Paulo, C.; Puskas, J. E. J. Rheol. 2001, 3, 759.
201. Graessley, W. W.; Roovers, J. Macromolecules 1979, 5, 959.
202. Gotsis, A. D.; Zeevenhoven, B. L. F.; Tsenoglou, C. J. J. Rheol. 2004, 4, 895.
203. Auhl, D.; Stange, J.; Munstedt, H.; Krause, B.; Voigt, D.; Lederer, A.; Lappan, U.;
Lunkwitz, K. Macromolecules 2004, 25, 9465.
204. Gabriela, C.; Munstedt, H. J. Rheol. 2003, 3, 619.
205. Gabriel, C.; Munstedt, H. Rheol. Acta 2002, 3, 232.
206. Gabriel, C.; Kokko, E.; Lofgren, B.; Seppala, J.; Munstedt, H. Polymer 2002, 24,
6383.
207. Münstedt, H. Soft Matter 2011, 6, 2273.
208. Frischknecht, A. L.; Milner, S. T.; Pryke, A.; Young, R. N.; Hawkins, R.; McLeish,
T. C. B. Macromolecules 2002, 12, 4801.
209. McCallum, T. J.; Kontopoulou, M.; Park, C. B.; Muliawan, E. B.; Hatzikiriakos, S.
G. Polym. Eng. Sci. 2007, 7, 1133.
210. Islam, M. T.; Juliani; Archer, L. A.; Varshney, S. K. Macromolecules 2001, 18,
6438.
211. Wood-Adams, P. M.; Dealy, J. M. Macromolecules 2000, 20, 7481.
212. Zamponi, M.; Pyckhout-Hintzen, W.; Wischnewski, A.; Monkenbusch, M.; Willner,
L.; Kali, G.; Richter, D. Macromolecules 2010, 1, 518.
213. Lohse, D. J.; Milner, S. T.; Fetters, L. J.; Xenidou, M.; Hadjichristidis, N.;
Mendelson, R. A.; Garcia-Franco, C. A.; Lyon, M. K. Macromolecules 2002, 8,
3066.
44
Chapter 1
214.
215.
216.
217.
218.
219.
220.
221.
222.
223.
224.
225.
226.
227.
228.
229.
230.
231.
232.
233.
234.
235.
236.
237.
238.
239.
240.
241.
242.
243.
244.
245.
246.
247.
248.
249.
250.
251.
252.
253.
254.
255.
Roovers, J.; Graessley, W. W. Macromolecules 1981, 3, 766.
Roovers, J. Macromolecules 1984, 6, 1196.
McLeish, T. Phys. Today 2008, 8, 40.
Burchard, W. Branched Polymers II 1999, 113.
Tam, K.; Tiu, C. Colloid Polym. Sci. 1990, 10, 911.
Lee, S.; Kim, D. H.; Huh, C.; Pope, G. A. SPE 2009, SPE-124798.
McCormick, C.; Salazar, L. Polymer 1992, 21, 4617.
McCormick, C.; Salazar, L. J. Appl. Polym. Sci. 1993, 6, 1115.
McCormick, C.; Salazar, L. Macromolecules 1992, 7, 1896.
McCormick, C.; Johnson, C. Macromolecules 1988, 3, 694.
McCormick, C.; Johnson, C. Macromolecules 1988, 3, 686.
McCormick, C.; Johnson, C. Polymer 1990, 6, 1100.
Kathmann, E.; Davis, D.; McCormick, C. Macromolecules 1994, 12, 3156.
Peiffer, D.; Lundberg, R. Polymer 1985, 7, 1058.
Smith, G. L.; McCormick, C. L. Macromolecules 2001, 16, 5579.
Petit, F.; Iliopoulos, I.; Audebert, R.; Szonyi, S. Langmuir 1997, 16, 4229.
Shedge, A. S.; Lele, A. K.; Wadgaonkar, P. P.; Hourdet, D.; Pcrrin, P.;
Chassenieux, C.; Badiger, M. V. Macromol. Chem. Phys. 2005, 4, 464.
Tomatsu, I.; Hashidzume, A.; Yusa, S.; Morishima, Y. Macromolecules 2005, 18,
7837.
Noda, T.; Hashidzume, A.; Morishima, Y. Macromolecules 2001, 5, 1308.
Noda, T.; Hashidzume, A.; Morishima, Y. Langmuir 2001, 19, 5984.
Mccormick, C. L.; Hoyle, C. E.; Clark, M. D. Polymer 1992, 2, 243.
Zhong, C.; Luo, P.; Ye, Z.; Chen, H. Polym. Bull. 2009, 1, 79.
Zhuang, D. Q.; Da, J. C. A. H.; Zhang, Y. X.; Dieing, R.; Ma, L.; Haeussling, L.
Polym. Adv. Technol. 2001, 11-12, 616.
Chang, Y. H.; Mccormick, C. L. Macromolecules 1993, 22, 6121.
Armentrout, R. S.; McCormick, C. L. Macromolecules 2000, 8, 2944.
Armentrout, R. S.; McCormick, C. L. Macromolecules 2000, 2, 419.
Chang, Y. H.; Mccormick, C. L. Polymer 1994, 16, 3503.
Mccormick, C. L.; Middleton, J. C.; Cummins, D. F. Macromolecules 1992, 4,
1201.
Mccormick, C. L.; Hester, R. D.; Morgan, S. E.; Safieddine, A. M. Macromolecules
1990, 8, 2124.
Mccormick, C. L.; Hester, R. D.; Morgan, S. E.; Safieddine, A. M. Macromolecules
1990, 8, 2132.
Ringsdorf, H.; Venzmer, J.; Winnik, F. M. Macromolecules 1991, 7, 1678.
Shalaby, S. W.; McCormick, C. L.; Butler, G. B. ACS Symp. Ser. 1991, 159.
Xue, W.; Hamley, I. W.; Castelletto, V.; Olmsted, P. D. Eur. Polym. J. 2004, 1,
47.
Yahaya, G. O.; Ahdab, A. A.; Ali, S. A.; Abu-Sharkh, B. F.; Hamad, E. Z. Polymer
2001, 8, 3363.
Jayasimha Reddy, G.; Venkata Naidu, S.; Rami Reddy, A. V. Adv. Polym. Tech.
2006, 1, 41.
Hwang, F. S.; Hogen-Esch, T. E. Macromolecules 1995, 9, 3328.
Zhang, Y. -.; Da, A. -.; Butler, G. B.; Hogen-Esch, T. E. J. Polym. Sci. Pol. Chem.
1992, 7, 1383.
Zhang, H.; Xu, K.; Ai, H.; Chen, D.; Xv, L.; Chen, M. J. Solution Chem. 2008, 8,
1137.
Bastiat, G.; Grassl, B.; François, J. Polym. Int. 2002, 10, 958.
Lara-Ceniceros, A. C.; Rivera-Vallejo, C.; Jiménez-Regalado, E. J. Polym. Bull.
2007, 4, 499.
Jiménez-Regalado, E. J.; Cadenas-Pliego, G.; Pérez-Álvarez, M.; HernándezValdez, Y. Polymer 2004, 6, 1993.
Yu Wang; Zhiyong Lu; Yugui Han; Yujun Feng; Chongli Tang Adv. Mater. Res.
2011, 654.
45
Introduction
256. Wu, Y.; Mahmoudkhani, A.; Watson, P.; Fenderson, T.; Nair, M. SPE 2012, SPE155653-MS.
257. Moradi-Araghi, A.; Doe, P. H. SPE 1987, SPE-13033.
258. Liu, R.; Fraylich, M.; Saunders, B. R. Colloid Polym. Sci. 2009, 6, 627.
259. Chen, Q.; Wang, Y.; Lu, Z.; Feng, Y. Polym. Bull. 2012, 2, 391.
260. Wang, Y.; Feng, Y.; Wang, B.; Lu, Z. J. Appl. Polym. Sci. 2010, 6, 3516.
261. Stahl, G. A.; Schulz, D. N. Water-Soluble Polymers for Petroleum Recovery;
Plenum Press: New York, United States of America, 1988; .
262. Stahl, G. A.; Moradi-Araghi, A.; Doe, P. H. Polym. Mater. Sci. Eng. 1986, 55,
258.
263. Doe, P. H.; Moradi-Araghi, A.; Shaw, J. E.; Stahl, G. A. SPE 1987, SPE-14233.
264. Chauveteau, G.; Denys, K.; Zaitoun, A. SPE 2002, SPE-75183.
265. Zitha, P. L. J.; Botermans, C. W. SPE 1998, SPE-36665.
266. Zitha, P. L. J.; van Os, K. G. S.; Denys, K. F. J. SPE 1998, SPE-39675.
267. Wever, D. A. Z. 2009.
46
Chapter 2
Chapter 2
Acrylamide homo- and block copolymers by atomic transfer
radical polymerization in water
Abstract
Atomic transfer radical polymerization (ATRP) of acrylamide has been
accomplished in aqueous media at room temperature. By using methyl 2chloropropionate
(MeClPr)
as
the
initiator
and
tris[2-
(dimethylamino)ethyl]amine (Me6TREN) / copper halogenide (CuX) as the
catalyst system, different linear polyacrylamides with apparent molecular
weights higher than 150000 g/mol were synthesized with dispersities as low
as 1.39. The molecular weights agreed well with the theoretical ones at
relatively low-medium monomer/initiator ratios (<700:1). Initial chain
extension experiments (isolated macro-initiator) resulted in a polymer with
bimodal distribution. However, in-situ chain extension experiments, carried
out by addition of a second fresh batch of monomer to the reaction mixture,
confirmed the living nature of the polymerization. By adding a fresh batch of
monomer to a linear macro-initiator (Mn = 22780 g/mol, PDI = 1.42) in
solution, an increase in the molecular weight up to 30220 g/mol (PDI = 1.64)
was observed. In addition linear polyacrylamides were used as macroinitiators for the synthesis of block copolymers polyacrylamide-b-poly(Nisopropylacrylamide).
Based on: D.A.Z. Wever, P.Raffa, F. Picchioni, A.A. Broekhuis. Acrylamide
homopolymers and acrylamide-N-isopropylacrylamide block copolymers by
atomic transfer radical polymerization in water. Macromolecules, 2012, 45,
4040-4045.
47
Atomic transfer radical polymerization of acrylamide
2.1. Introduction
Polyacrylamide (PAM) and its derivatives are widely used in cosmetics,
biomedical applications, wastewater treatment, and oil recovery. 1, 2 Although
their synthesis has been extensively studied, the focus now lies mainly on the
control of the polymerization process through living radical polymerization
strategies.3-9 Atomic transfer radical polymerization (ATRP), a living radical
polymerization technique, allows the synthesis of polymers with well-defined
molecular weights and dispersities (PDI<1,5).10,
11
This technique is widely
used for monomers such as (functionalized) styrenes12, (meth)acrylates12
and acrylonitrile13, but its use to polymerize acrylamide and its derivatives is
limited.
Generally speaking, the ATRP of water soluble monomers still represents
a challenge with respect to the control of the polymerization when using
water as the only solvent.14 ATRP of acrylamide (and its derivatives) has in
general been explored in organic solvents (methanol 15, ethanol16, toluene3, 16,
dimethylformamide16 [DMF], 2-propanol17) and mixtures of organic solvents
with water (ethanol-water18,
19
[4-1 and 7-3, v/v], DMF-water20 [range
between 1-1 to 7-3, v/v] and glycerol-water4,
5, 11, 21
[1-1, v/v]). The
problems connected with the use of water for ATRP (vide infra) can be
mitigated by performing the polymerizations in an organic-water mixture at
low (0 °C) temperatures.22 ATRP of acrylamide in water at elevated
temperatures (>80 °C) has also been reported.4,
11, 21
Low dispersity PAM
could be prepared using an activator generated by electron transfer ATRP in
water at room temperature.23 However, the apparent molecular weights were
relatively low (< 6000 g/mol).
Regarding ATRP in water solution, good results in terms of dispersity and
predictability of molecular weight have been published for few systems 24,
25
.
However, several investigations on the ATRP of hydrophilic acrylic monomers
conducted in aqueous solutions showed that the process is difficult to control,
unless the polymerization rate is slowed down by adding a co-solvent
(usually an alcohol) or a Cu(II) salt.14, 26-28
Successful ATRP has been accomplished for several derivatives of
acrylamide3,
15,
16,
dimethylacrylamide,
18
,
such
as
N-hydroxyethylacrylamide,
N-tert-butylacrylamide
and
N,NN-(2-
hydroxypropyl)methacrylamide. To the best of our knowledge, only few
publications4,
using
5, 11, 29
chloro-acetic
mentioned the controlled polymerization of acrylamide
acid,
2-chloropropionamide
(2-Cl-PA)
or
2-
bromopropionamide (2-Br-PA) as initiators and either CuCl / N,N,N,Ntetramethylethylenediamine (TMEDA) or 2,2-bipyridine (bpy) as catalytic
systems. Although the molecular weight of the polyacrylamide increases
48
Chapter 2
linearly with conversion5,
29
, the apparent (determined by gel permeation
chromatography, GPC) molecular weight differed significantly from the
theoretical one.
The ATRP of acrylamide was investigated in more detail using bpy,
pentamethyldiethylenetriamine (PMDETA), hexamethyltriethylenetetraamine
(HMTETA),
TMEDA
or
1,4,8,11-tetramethyl-1,4,8,11-
tetraazacyclotetradecane (Me4Cyclam) as ligands (Figure 2.1).4
Figure 2.1: Chemical structure of the different ligands used in the ATRP of acrylamide
Although the average molecular weight increased with conversion, no
concordance between the theoretical and experimental values was achieved.
Only with the extraneous addition of copper(II) did the theoretical molecular
weight (conversion·initial monomer:initiator ratio) agree well with the actual
one (apparent Mn as measured by GPC), where the role of copper(II) consists
in ensuring a fast deactivation rate in order to achieve relatively low
dispersity values30. Nevertheless the dispersities of the subsequent polymers
were relatively high (PDI ≥ 1.6), indicating a difficult control of the
polymerization. By using PMDETA and a lower temperature (90 °C instead of
130 °C), a reduction of the dispersity to 1.24 was achieved 11; however, when
higher molecular weight (>5000 g/mol) polymers were synthesized by using
the chloride system, the dispersity increased significantly (PDI > 1.6) 11. A
low dispersity linear PAM, whose molecular weight matched the theoretical
one, could be synthesized using the bromide system (and addition of
extraneous Cu(II)Br).11
49
Atomic transfer radical polymerization of acrylamide
ATRP of acrylamide has also been claimed in aqueous media 23; however the
molecular weight again did not match the theoretical one. Terminated
polyacrylamide (loss of the halogen group) has been reported following the
ATRP of AM using 2-Cl-PA / CuCl / Me6TREN as the initiator/catalyst system
in a DMF-water (50-70% DMF by volume) solution.20 Chain extension
experiments failed due to the loss of the halogen group.20
Successful surface initiated ATRP of acrylamide has also been claimed in
DMF using bpy-based copper complexes.31-34 However, it has been concluded
that bpy-based copper complexes fail to initiate the polymerization of
acrylamide.15,
16,
35
In addition, deactivation of the catalyst, through
complexation by acrylamide or polyacrylamide, limits the conversion.
As evident from the above discussion, the ATRP of acrylamide still
constitutes a significant hurdle in the science of living radical polymerization.
ATRP of acrylamide has been accomplished in aqueous media using MeClPr /
Me6TREN / CuCl as the initiation/catalyst system. The molecular weight of the
polymers increased linearly with conversion and the dispersity remained
relatively low. Chain extension experiments confirmed the living nature of the
polymerizations in aqueous media. In addition, well-defined polyacrylamideb-poly(N-isopropylacrylamide) block copolymers were synthesized.
2.2. Experimental section
Chemicals.
Acrylamide
isopropylacrylamide
(NIPAM,
(AM,
electrophoresis
97%),
grade,
≥99%),
N-
tris[2-(dimethylamino)ethyl]amine
(Me6TREN) copper(I) bromide (CuBr, 98%), copper(I) chloride (CuCl, 98%),
glacial acetic acid, ethanol, diethyl ether and methyl 2-chloropropionate
(MeClPr, 97%) were purchased from Sigma Aldrich. CuBr and CuCl were
purified by stirring in glacial acetic acid for at least 5 hours, filtering, and
washing with glacial acetic acid, ethanol and diethyl ether (in that order) and
then dried at reduced pressure. All the other chemicals were reagent grade
and used without further purification.
ATRP of AM in aqueous media. A 250-mL three-necked flask was
charged with all the solid chemicals (CuCl & AM). A magnetic stirrer and
distilled water were added and subsequently degassed by three freeze-pumpthaw cycles and left under nitrogen. The flask was then placed in an oil bath
at 25 °C. Afterwards Me6TREN was added and the mixture was stirred for 10
minutes. The reaction was started by adding the initiator using a syringe. All
the operations were carried out under nitrogen. After the reaction the
mixture was exposed to air and the polymer was precipitated in a tenfold
50
Chapter 2
amount of methanol. The polymer was dried in an oven at 65 °C up to
constant weight. Detailed reaction conditions are summarized in Table 2.1.
Kinetic experiments. Aliquots of the reaction mixture were removed at
different time intervals using a degassed syringe. The aliquots were
immediately frozen in liquid nitrogen. A portion was used for conversion
measurements with GC and the remaining part was diluted with distilled
water and analyzed with GPC (after precipitation).
Chain extension experiments. Two different methods of chain
extension were carried out; two-step or single step in situ chain extension.
For the two-step method, acrylamide was polymerized using the ATRP
method as described earlier. The polymer was isolated, after a 1 hour
reaction, by precipitation in methanol and characterized. A 100-mL threenecked flask was charged with the solid chemicals (macroinitiator (PAM),
CuBr & AM). A magnetic stirrer and distilled water were added and
subsequently the mixture was degassed by three freeze-pump-thaw cycles.
The flask was placed in an oil bath at 25 °C and the reaction was started by
the addition of Me6TREN under nitrogen. The polymerization was continued
for 22 hours. The polymer was then isolated and characterized.
For the in situ method, acrylamide was polymerized using the ATRP method
as described before. After 1 hour of reaction an aliquot was taken for
analysis. After this, a fresh batch of AM/Me6TREN/CuBr was added under
nitrogen. The polymerization with the fresh batch was continued for a further
period of 2 hours after which a sample was taken for analysis.
Block copolymerization, synthesis of PAM-b-PNIPAM. The macroinitiator PAM-Cl was synthesized according to the aforementioned procedure.
To a round bottomed flask 0.42 g (0.0178 mmol) of the macro-initiator was
added along with NIPAM (1 g, 8.4 mmol). Double distilled water was added
and the mixture was degassed by three freeze-pump-thaw cycles followed by
the addition of the catalyst. The flask was placed in a thermostated oil bath
at 25 °C. To start the reaction, the ligand was added. All operations were
carried out under nitrogen.
A sample of the synthesized block copolymer PAM-b-PNIPAM was
thoroughly washed five times with THF. The washed sample was dried in an
oven at 65 °C. A 1H-NMR spectrum was recorded for both the washed and
virgin samples.
Characterization. The acrylamide conversion was measured using Gas
Chromatography (GC). The samples were dissolved in acetone (polymer
precipitates), filtered or decanted and injected on a Hewlett Packard 5890 GC
with an Elite-Wax ETR column.
51
Atomic transfer radical polymerization of acrylamide
Nuclear Magnetic Resonance (NMR) spectra were recorded on a Varian
Mercury Plus 400 MHz spectrometer. For analysis D 2O was used as the
solvent.
Gel permeation chromatography (GPC) analysis of all the water-soluble
samples was performed on a Agilent 1200 system with Polymer Standard
Service (PSS) columns (guard, 104 and 103 Å) with a 50 mM NaNO3 aqueous
solution as the eluent. The columns were operated at 40 °C with a flow-rate
of 1.00 ml/min, and a refractive index (RI) detector (Agilent 1200) was used
at 40 °C. The apparent molecular weights and dispersities were determined
using a polyacrylamide (PAM) based calibration with WinGPC software (PSS).
2.3. Results and discussion
2.3.1. ATRP of acrylamide. The homo- and block-copolymerization
(with NIPAM) experiments of acrylamide were performed according to
Scheme 2.1.
Scheme 2.1: A, Homopolymerization of AM and B, Block copolymerization of AM and
NIPAM
The
parameters
that
varied
were
the
amount
of
solvent
and
the
monomer/initiator/catalyst ratios (Table 2.1). As can be observed, PAM of
relatively high molecular weights, up to 40 000 g/mol, can be prepared with
relatively low dispersities.
The kinetic plot of the disappearance of AM is non-linear (Figure 1.1),
which is in line with earlier publications on the ATRP of AM 4, 11 and derivatives
thereof.3, 4, 11, 15, 16 The kinetics of living radical polymerization can be divided
into the stationary (quasi-equilibrium) state and a state exhibiting a power
52
Chapter 2
law dependence in time of the conversion index36 (ln [
/ ]), a function of
the monomer concentration at any given time t ( ) and at time zero (
).
Table 2.1: Homopolymerization of acrylamide under different conditions
[M]0:[I]0:[CuCl]0:[Me6TREN]0a
Entry
M/water (wt:vol); T;
Time (min)
1
225 : 1 : 2 : 2
2
385 : 1 : 6.0 : 6.0
3
385 : 1 : 1.5 : 1.5
4
470 : 1 : 1.5 : 1.5
5
500 : 1 : 1.5 : 1.5
6
680 : 1 : 1.5 : 1.5
7
870 : 1 : 1.5 : 1.5
8
945 : 1 : 1.5 : 1.5
9
965 : 1 : 1.5 : 1.5
10b
1000 : 1 : 1.5 : 1.5
11
1625 : 1 : 1.5 : 1.5
12
2785 : 1 : 1.5c : 1.5
13
4355 : 1 : 1.5 : 1.5
a. Molar ratio
b. No increase in molecular weight with
c. CuBr was used.
1:4;
1:6;
1:4;
1:6;
1:12;
1:6;
1:6;
1:6;
1:6;
1:15;
1:6;
1:6;
1:6;
Conv (%)
25 °C; 60
25 °C; 60
25 °C; 60
25 °C; 30
25 °C; 60
25 °C; 90
25 °C; 2
25 °C; 3
25 °C; 60
25 °C; 60
25 °C; 60
25 °C; 60
25 °C; 60
Mn,th
48.7
28.2
88.2
69.8
42.2
78.3
47.3
36.8
75.3
52.0
84.7
58.5
69.1
7
7
24
23
14
37
29
24
51
28
97
115
213
719
717
011
269
998
901
284
719
703
762
833
805
852
Mn,GPC
10
11
22
22
16
32
26
25
38
41
68
75
108
230
900
780
863
780
680
260
850
310
970
370
880
800
PDI
1.40
1.42
1.42
1.88
1.56
1.46
1.54
1.57
1.97
2.04
2.05
2.30
increase in conversion.
In the stationary state the conversion index (ln[
/
]) is represented by
equation 2.1.
(2.1)
where
and
indicates the kinetic constant for propagation,
the initiation rate
the termination rate constant.
In the case where initiation doesn’t follow the conventional system and
the starting concentration of radicals equals zero ([X*]0 = 0, [X*]0 being the
radical concentration at time zero) the conversion index (ln[
/
]) is
represented by equation 2.236.
(2.2)
where
is the equilibrium constant in ATRP (
activation rate constant and
=
/
, where
is the
is the deactivation rate constant in ATRP).
In most ATRP systems the kinetics of the reaction crosses over from the
power law dependence to the quasi equilibrium within 1 minute after starting
53
Atomic transfer radical polymerization of acrylamide
the reaction. Using equation 2.2 we modeled the kinetics of the ATRP of AM
(Figure 2.2).
2,0
2
Entry 11, R (model) = 0.99
2
Entry 4, R (model) = 0.92
1,5
1,6
1,0
1,2
Ln (M 0/M)
Ln (M 0/M)
2,0
0,8
0,5
Entry 11
Entry 4
0,4
0,0
0
2
4
6
8
2/3
Time
10
12
14
16
2/3
(min )
0,0
0
10
20
30
40
50
60
Time (min)
Figure 2.2: Kinetic plot for the ATRP of AM (entry 4 and 11, Table 2.1), big plot on a
linear time scale and inset on a scale of time2/3
A straight line should be obtained when the time scale is adjusted to the
exponent (2/3)36. Indeed, a good correlation is obtained (inset, Figure 2.2)
on a timescale of t2/3. Although the non-linearity of the kinetic plot is an
indication of the presence of termination reactions16, given the results of the
chain extension experiments, the non-linearity probably arises due to a
progressive deactivation of the catalyst by complexation with the growing
polyacrylamide chains15, 16. Moreover, the molecular weights increase linearly
with conversion and the Mn values were in good agreement (especially at
medium molecular weights) with the theoretical values (Figure 2.3).
Low molecular weight tailing has been observed in the GPC traces when
attempting to demonstrate the ATRP of DMAA in toluene using the same CuCl
/ Me6TREN / MeClPr initiatior/catalyst system.3 The only difference with the
present system (despite the monomer) is the lower monomer/solvent ratio.3
However, in the present case (Figure 2.4), the low molecular weight tailing in
the GPC traces of the ATRP of acrylamide is not as pronounced as with the
ATRP of DMAA.
54
Chapter 2
3.0
Mn,GPC
2.8
Mn, theoretical
Molecular weight (g/mol)
2.6
PDI
28000
2.4
24000
2.2
2.0
20000
1.8
1.6
16000
1.4
12000
Polydispersity index (PDI)
32000
1.2
2000
0
0 2 4
40
50
60
70
80
90
1.0
100
Conversion (%)
Figure 2.3: The dependence of the Mn and PDI on the conversion of AM (entry 4,
Table 2.1), dotted lines serve as a guide
Conversion = 69,8%
Mn= 22870
PDI = 1.39
PDI = 1.42
rel. RID intensity
Conversion = 49,8%
Mn= 16540
1000
10000
100000
Molecular weight (Mn)
Figure
2.4:
GPC
traces
of
the
PAM
(entry
4,
Table
2.1),
conditions
[AM]0:[MeClPr]0:[CuCl]0:[Me6TREN]0 = 470:1:1.5:1.5; AM:solvent = 1:6 (w/v);
solvent = water; T = 25 °C
55
Atomic transfer radical polymerization of acrylamide
On the other hand, a significant deviation of the molecular weight from the
theoretical one is observed at conversions higher than 70% (Table 2.1),
which is more pronounced when using high monomer to initiator ratios
(entries 6-11) or higher amounts of catalyst/ligand (entries 1 and 2). When
the amount of solvent was increased (entries 5 & 10), the control of the
polymerization was lost, as also reported for the ATRP of NIPAM in
isopropanol.17 Although the molecular weight of the polymer is similar to the
one prepared with a lower amount of solvent (entry 4 & 9), the PDI is
significantly higher, the conversion is limited (similar to other results for
DMAA in toluene3) and no increase in molecular weight with increase in
conversion was observed (data not shown for brevity).
The dispersities of the PAM (entries 1-9, except entry 5) are lower
compared to the ATRP of acrylamide in water and/or water-glycerol mixture
at elevated temperatures4, which to this point constitute the best results on
ATRP of AM in water. In an attempt to prepare higher molecular weight PAM
(entries 11-13 in Table 2.1), higher monomer to initiator ratios were used.
Although higher molecular weight PAM could be prepared, the dispersities of
the polymers are relatively high. In spite of this, the linear increase of the
molecular
weight
polymerization
37
with
conversion
indicates
a
controlled
radical
(except entries 5 & 10). In addition, similarly to entry 4
(Table 2.1), the extent of low molecular weight tailing is not significant
(Figure 2.5). Given the high reaction rate of the catalyst system, the
viscosity of the reaction medium quickly increases (when using high
monomer / initiator ratios) and this might lead to mass transfer limitations.
As commonly accepted, for a successful ATRP, several conditions should
be met. These conditions are30,
38
: low dispersities (1.0 < PDI < 1.5)
throughout the reaction, linear increase of the molecular weight with
conversion and good concordance between the theoretical molecular weights
with the experimental values (and chain extensions4). The present system,
MeClPr / CuCl / Me6TREN in water meets all these parameters, which is in
stark contrast to the ATRP of AM in DMF-water mixture.20 It can be
speculated that the use of DMF-water in conjunction with a halide salt (LiCl,
KCl or NH4Cl) enhances the rate of termination leading to a dead polymer.20
2.3.2. Chain extension experiment, two-step
As mentioned in the experimental section, two different approaches were
tried in extending the PAM chains. Figure 2.6 displays the GPC results of the
two-step approach. First the macro-initiator was prepared by the ATRP of AM.
After a 1 h reaction period, a conversion of 66% (gravimetrically) was
reached yielding the PAM-Cl macro-initiator (Mn = 23490 g/mol, PDI = 1.45).
56
Chapter 2
In the second stage, the same concentration of AM was used. After a 22 h
reaction period a conversion of 14 % (gravimetrically determined) was
reached. The GPC trace of the chain extended macro-initiator is bimodal
(Figure 2.6) with an Mn = 39600 g/mol and a PDI of 5.63. This result clearly
indicates that a portion of the chains cannot be initiated, even with the
principle of halogen exchange39, 40.
Conversion = 84,7%
Mn=68370
PDI = 1.92
PDI = 2.04
rel. RID detector
Conversion = 57,1%
Mn=45810
1000
10000
100000
1000000
Molecular Weight (Mn)
Figure
2.5:
GPC
traces
of
the
PAM
(entry
11,
Table
2.1),
conditions
[AM]0:[MeClPr]0:[CuCl]0:[Me6TREN]0 = 1625:1:1.5:1.5; AM:solvent = 1:6 (w/v);
solvent = water; T = 25 °C
This result is similar to the chain extensions of either a polystyrene or poly(nbutyl acrylate) with methyl methacrylate.41 The poor initiation efficiency of
the macro-initiator leads to the bimodal distribution (Figure 2.6). The
halogen groups on the macro-initiator are secondary halogens -substituted
carbonyl, which are known to have much lower activation rates compared to
their tertiary and bromide analogues.42,
43
This fact explains the difficulty in
activating the PAM macro-initiator. Nevertheless, the bimodal GPC trace
indicates the presence of the halogen group on the macro-initiator. Initial
results on the chain extensions of polystyrene and poly(n-butyl acrylate) with
methyl methacrylate (MMA) displayed bimodal GPC traces.41 The bimodal
GPC traces were attributed to poor initiation efficiency and the problem was
mitigated by using 10 mol% of styrene in the monomer. 41 This is in stark
contrast to the ATRP of AM (loss of halogen group, i.e. dead polymer) in a
57
Atomic transfer radical polymerization of acrylamide
water/DMF
solution
(1:1)
using
2-Cl-PA/Me6TREN/CuCl
as
the
initiator/catalyst system.20
Conversion = 66 %
Mn= 23490
PDI = 1.45
rel. RID intensity
Conversion = 14 %
Mn= 39600
PDI = 5.63
1000
10000
100000
1000000
Molecular weight (Mn)
Figure 2.6: GPC traces for the two-step chain extension experiment
2.3.3. Chain extension experiment, in situ
Figure 2.7 shows the GPC results of the in situ chain extension approach.
As mentioned earlier, the difference here is that the macro-initiator is not
isolated (by precipitation in methanol). After a one hour reaction period, a
conversion of 88.2% was reached yielding the PAM-Cl macro-initiator (Mn =
22780 g/mol [Mn,th = 24011 g/mol], PDI = 1.42). After this, a second batch
containing the same concentration of monomer, catalyst (halogen exchange
principle) and ligand was added. The conversion of AM (second block)
reached 25.5% after 2 h. The chain extended polymer had a Mn of 30220
g/mol and a PDI of 1.64 (Mn,th = 30953). This result reinforces the
aforementioned conclusion that the halogen group is not lost during the ATRP
of AM.
2.3.4. Block copolymerization, synthesis of PAM-b-PNIPAM
As it is known that thermo-responsive44 polymers offer control over
viscosity by temperature variation, the above mentioned polymer has been
functionalized with NIPAM based blocks. Several PAM-b-PNIPAM block
copolymers
were
prepared
according
to
Scheme
2.1B.
These
block
copolymers have a low dispersity (PDI = 1.48) and a monomodal
58
Chapter 2
distribution. For brevity Figure 2.8 displays only the 1H-NMR spectra of one
example of a PAM-b-PNIPAM block copolymer and of the THF washed
equivalent.
Conversion = 25.5%
Mn = 30220
PDI = 1.42
PDI = 1.64
rel. RID intensity
Conversion = 88.2%
Mn = 22780
1000
10000
100000
Molecular weight (Mn)
Figure 2.7: GPC traces for the in situ chain extension experiment
MeOH
PAM-b-PNIPAM
DP = 330-b-35
PAM
PAM-b-PNIPAM
washed with THF
Methyl-groups NIPAM
PAM-b-PNIPAM
10
8
6
4
2
0
ppm
Figure 2.8: NMR spectra of PAM-b-PNIPAM (virgin and THF washed) and PAM
59
Atomic transfer radical polymerization of acrylamide
The conversion of NIPAM was determined by using the ratio between the
resonances
of
AM
and
NIPAM
units.
The
conversion
equaled
5%
corresponding to a degree of polymerization (DP) of 25 and a M n of 2 811
g/mol. In addition, washing with THF did not change the ratio between the
resonances of the AM and NIPAM units. This confirms that the NIPAM units
are covalently linked to the PAM macro-initiator.
2.4. Conclusion
ATRP of
acrylamide
has been accomplished
in water using
the
MeClPr/Me6TREN/CuCl as the initiator/catalyst system. The molecular weights
were in good agreement with the theoretical values. Linear PAM with
apparent molecular weights higher than 150000 g/mol and dispersities as low
as 1.39 could be prepared. Although the dispersities are higher than for ATRP
of styrene and acrylates, both (two-step and in situ) chain extension
experiments proved the living nature of the polymerizations. In addition, the
well-defined block copolymer (PAM-b-PNIPAM, DP 330-b-25) was synthesized
using the linear macro-initiator (PAM-Cl) prepared by the ATRP of AM in
water. The possibility to synthesize well-defined linear homo- and block
copolymers in water solution and under mild conditions can be highly
attractive for industrial applications.
2.5. Acknowledgement
This work is part of the Research Programme of the Dutch Polymer
Institute DPI, Eindhoven, the Netherlands, projectnr. #716.
2.6. References
1. Shalaby W. Shalaby; Charles L. McCormick; George B. Butler Water-Soluble
Polymers: Synthesis, Solution Properties, and Applications; American Chemical
Society: Washington DC, 1991; .
2. Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Progress in Polymer Science 2011,
11, 1558.
3. Neugebauer, D.; Matyjaszewski, K. Macromolecules 2003, 8, 2598.
4. Jewrajka, S. K.; Mandal, B. M. Macromolecules 2003, 2, 311.
5. Jiang, J.; Lu, X.; Lu, Y. Polymer 2008, 7, 1770.
6. Senoo, M.; Kotani, Y.; Kamigaito, M.; Sawamoto, M. Macromolecules 1999, 24,
8005.
7. Donovan, M.; Sanford, T.; Lowe, A.; Sumerlin, B.; Mitsukami, Y.; McCormick, C.
Macromolecules 2002, 12, 4570.
8. Donovan, M. S.; Lowe, A. B.; Sumerlin, B. S.; McCormick, C. L. Macromolecules
2002, 10, 4123.
60
Chapter 2
9. Benoit, D.; Chaplinski, V.; Braslau, R.; Hawker, C. J. J. Am. Chem. Soc. 1999, 16,
3904.
10. Braunecker, W. A.; Matyjaszewski, K. Progress in Polymer Science 2007, 1, 93.
11. Jewrajka, S. K.; Mandal, B. M. Journal of Polymer Science Part A-Polymer
Chemistry 2004, 10, 2483.
12. Coessens, V.; Pintauer, T.; Matyjaszewski, K. Progress in Polymer Science 2001,
3, 337.
13. Matyjaszewski, K.; Jo, S. M.; Paik, H. J.; Gaynor, S. G. Macromolecules 1997, 20,
6398.
14. Iddon, P. D.; Robinson, K. L.; Armes, S. P. Polymer 2004, 3, 759.
15. Teodorescu, M.; Matyjaszewski, K. Macromolecules 1999, 15, 4826.
16. Teodorescu, M.; Matyjaszewski, K. Macromolecular Rapid Communications 2000,
4, 190.
17. Xia, Y.; Yin, X. C.; Burke, N. A. D.; Stover, H. D. H. Macromolecules 2005, 14,
5937.
18. Narumi, A.; Chen, Y.; Sone, M.; Fuchise, K.; Sakai, R.; Satoh, T.; Duan, Q.;
Kawaguchi, S.; Kakuchi, T. Macromolecular Chemistry and Physics 2009, 5, 349.
19. Appel, E. A.; del Barrio, J.; Loh, X. J.; Dyson, J.; Scherman, O. A. Journal of
Polymer Science Part A-Polymer Chemistry 2012, 1, 181.
20. Guha, S. Journal of the Indian Chemical Society 2008, 1, 64.
21. Jewrajka, S. K.; Mandal, B. M. Journal of the Indian Chemical Society 2005, 9,
819.
22. Ye, J.; Narain, R. J Phys Chem B 2009, 3, 676.
23. Tan, Y.; Yang, Q.; Sheng, D.; Su, X.; Xu, K.; Song, C.; Wang, P. E-Polymers
2008, 25.
24. Zeng, F. Q.; Shen, Y. Q.; Zhu, S. P.; Pelton, R. Journal of Polymer Science Part APolymer Chemistry 2000, 20, 3821.
25. Wang, X. S.; Jackson, R. A.; Armes, S. P. Macromolecules 2000, 2, 255.
26. Save, M.; Weaver, J. V. M.; Armes, S. P.; McKenna, P. Macromolecules 2002, 4,
1152.
27. Robinson, K. L.; Khan, M. A.; Banez, M. V. D.; Wang, X. S.; Armes, S. P.
Macromolecules 2001, 10, 3155.
28. Ma, I. Y.; Lobb, E. J.; Billingham, N. C.; Armes, S. P.; Lewis, A. L.; Lloyd, A. W.;
Salvage, J. Macromolecules 2002, 25, 9306.
29. Jiang, J.; Lu, X.; Lu, Y. Journal of Polymer Science Part A-Polymer Chemistry
2007, 17, 3956.
30. Matyjaszewski, K.; Xia, J. H. Chem. Rev. 2001, 9, 2921.
31. Huang, X.; Wirth, M. J. Macromolecules 1999, 5, 1694.
32. Huang, X. Y.; Doneski, L. J.; Wirth, M. J. Anal. Chem. 1998, 19, 4023.
33. Huang, X. Y.; Wirth, M. J. Anal. Chem. 1997, 22, 4577.
34. Cringus-Fundeanu, I.; Luijten, J.; van der Mei, H. C.; Busscher, H. J.; Schouten, A.
J. Langmuir 2007, 9, 5120.
35. Li, D. W.; Brittain, W. J. Macromolecules 1998, 12, 3852.
36. Goto, A.; Fukuda, T. Progress in Polymer Science 2004, 4, 329.
37. Xia, J. H.; Matyjaszewski, K. Macromolecules 1997, 25, 7697.
38. Patten, T. E.; Matyjaszewski, K. Acc. Chem. Res. 1999, 10, 895.
39. Qin, S. H.; Saget, J.; Pyun, J. R.; Jia, S. J.; Kowalewski, T.; Matyjaszewski, K.
Macromolecules 2003, 24, 8969.
40. Tsarevsky, N. V.; Cooper, B. M.; Wojtyna, O. J.; Jahed, N. M.; Gao, H.;
Matyjaszewski, K. Polymer Preprint 2005, 46, 249-250.
41. Mueller, L.; Jakubowski, W.; Tang, W.; Matyjaszewski, K. Macromolecules 2007,
18, 6464.
42. Tang, W.; Kwak, Y.; Braunecker, W.; Tsarevsky, N. V.; Coote, M. L.;
Matyjaszewski, K. J. Am. Chem. Soc. 2008, 32, 10702.
43. Tang, W.; Matyjaszewski, K. Macromolecules 2007, 6, 1858.
44. Liu, R.; Fraylich, M.; Saunders, B. R. Colloid Polym. Sci. 2009, 6, 627.
61
Atomic transfer radical polymerization of acrylamide
This page intentionally left blank
62
Chapter 3
Chapter 3
Branched polyacrylamides:
Synthesis and effect of molecular
architecture on solution rheology
Abstract
Linear,
star
and
comb-shaped
polyacrylamides (PAM)
have
been
prepared by atomic transfer radical polymerization (ATRP) in aqueous media
at room temperature. The influence of the molecular architecture of PAM on
the rheological properties in aqueous solution has been investigated. The
well-known theory of increased entanglement density by branching for
polymers in the melt can also be applied to polymers in semi-dilute water
solutions. We have demonstrated this by investigating the rheological
properties of PAM of similar molecular weights with different molecular
architectures. Interestingly, the solution viscosity of a comb-like PAM is
higher than its linear and star-shaped analogues (both at equal span
molecular weight, Mn,SPAN, and total molecular weight, Mn,tot). In addition to
the pure viscosity, we also demonstrate that the visco-elastic properties of
the polymeric solutions vary as a function of the molecular architecture of the
employed PAM. The elastic response of water solutions containing comb PAM
is more pronounced than for solutions containing either linear or star PAM at
similar Mn,SPAN and Mn,tot. The obtained results pave the way towards
application of these polymeric materials in Enhanced Oil Recovery (EOR).
Based
on:
D.A.Z.
Wever,
F.
Picchioni,
A.A.
Broekhuis.
Branched
polyacrylamides: Synthesis and effect of molecular architecture on solution
rheology. European Polymer Journal, 2013, 49, 3298-3301.
63
Synthesis of branched polyacrylamide
3.1. Introduction
Polyacrylamide (PAM) is a versatile water soluble polymer which is used
in a number of areas such as oil recovery, wastewater treatment, cosmetics
and biomedical applications.1, 2 For most of these applications the function of
the polymer is to increase the solution viscosity or to behave as a flocculating
agent. Looking more closely at the polyacrylamides currently used, one can
observe that in all the applications linear PAM is employed. This is probably
due to the fact that PAMs with different architectures (i.e. other than linear)
are difficult to prepare. The relatively high propagation rate 3 during
polymerization prevents achieving control over the molecular architecture. It
was demonstrated that uncontrolled grafted PAM can be prepared using free
radical polymerization at higher temperatures.4, 5 Alternatively, branched PAM
has been synthesized through the usage of transfer agents.6,
7
Although a
8
high degree of branching could be obtained there is little to no control in the
reaction and thus no control over the molecular architecture of the resulting
polymer.
The difficulties become even more relevant when attempting a controlled
radical polymerization, i.e. when trying to prepare PAM homo- and copolymers with a well-controlled macromolecular architecture. Historically,
controlled polymerization has been achieved by living anionic polymerization,
reversible addition-fragmentation chain transfer (RAFT) or atomic transfer
radical polymerization (ATRP). Unsuccessful controlled radical polymerization
of acrylamide has been reported.9-12 Similar to N-isopropylacrylamide13, living
anionic polymerization cannot be considered given the acidity (pKa ~ 25-26)
of the amide protons of acrylamide. Recently, the controlled preparation of
hyperbranched PAM has been demonstrated by copolymerizing acrylamide
and
N,N-methylenebis(acrylamide)
using
a
semi-batch
RAFT
polymerization.14 However, in order to prepare comb-shaped polymers with
long arms, more specific methodologies15, i.e. “grafting from” (backbone
functionalized with a RAFT agent or radical initiator) or “grafting through”
(through the use of macromonomers), have to be used leading to more
cumbersome and lengthy preparation routes.
ATRP has enabled the synthesis of a variety of molecular architectures of
an even wider variety of different monomers.16 Nevertheless, given the
difficulty for the ATRP of acrylamide, the synthesis of branched PAM in a
controlled fashion has not been reported so far. However, with the recent
accomplishment of ATRP of acrylamide, either in water
mixture
18
17
or a water-alcohol
, controlled polymerization of acrylamide yielding grafted, comb
and star-shaped PAM can be envisaged. Star-shaped PAM can be easily
prepared using the well-known multifunctional initiators widely used for the
64
Chapter 3
preparation of star polystyrenes and polyacrylates19. Other methods aimed at
the synthesis of comb-like structures of different monomers have been
published
20-22
, but are based on multiple and cumbersome synthetic steps to
prepare the appropriate macroinitiators. This paper describes the preparation
of a multifunctional macro-initiator based on aliphatic alternating polyketone
(PK) oligomer. The latter was functionalized through the classic Paal-Knorr
reaction leading to the desired macro-initiator, which was subsequently used
in the ATRP of acrylamide yielding the envisaged comb-like PAM. Linear and
star-shaped polymers were also prepared using the published method. 17 The
rheological properties for these polymers were compared in aqueous
solutions.
In this work, the aim is to (1) synthesize branched (comb) PAM using
novel macro-initiators based on aliphatic perfectly alternating polyketones
and (2) to investigate the effect of the architecture of the polymer on the
aqueous solution rheology. The choice of chemically modified PK (a polymer
of industrial origin with relatively broad molecular weight distribution) as
initiator stems for the future applicability of the proposed method at
industrial level.
3.2. Experimental section
Chemicals. Acrylamide (AM) (electrophoresis grade, ≥99%), PAM (M w =
5-6·106
g/mol),
tris[2-(dimethylamino)ethyl]amine
(Me6TREN),
2,2-
bipyridine (bpy), copper(I) chloride (CuCl, 98%), copper(I) bromide (CuBr,
98%), methyl 2-chloropropionate (MeClPr, 97%), methyl chloroacetate
(MClAc,
99%)
pentaerythritol
tetrakis(2-bromoisobutyrate)
(97%),
3-
chloropropylamine hydrochloride (98%), and sodium hydroxide (pellets) were
purchased from Sigma Aldrich. CuCl and CuBr were purified by stirring in
glacial acetic acid (Aldrich), washing with glacial acetic acid, ethanol and
diethyl ether (in that order) and then dried under vacuum. All solvents were
reagent grade and used without further purification. The alternating
polyketone with 30 mol% ethylene content (PK30, Mn = 2797 g/mol, PDI =
1.74) was synthesized according to the published procedure.23, 24
ATRP of AM in aqueous media using a primary halogen. The
polymerization was performed in analogy with literature 17. A 250 mL threenecked flask was charged with AM (5 g, 70 mmol). A magnetic stirrer and
distilled water were added and subsequently degassed by three freeze-pumpthaw cycles and left under nitrogen. The flask was then placed in a water
bath at 25 °C. Afterwards CuCl (21 mg, 0.21 mmol) and Me 6TREN (48 mg,
0.21 mmol) were added, and the mixture was stirred for 10 min. The
65
Synthesis of branched polyacrylamide
reaction was started by adding MClAc (15 mg, 0.14 mmol) with a syringe. All
the operations were performed under nitrogen. The polymer was isolated by
precipitation in a ten-fold amount of methanol and subsequently dried in an
oven at 65 °C. Aliquots of the reaction mixture were removed at different
time intervals using a degassed syringe and frozen immediately in liquid
nitrogen. AM conversion was determined using a GC and the molecular
weight and distribution were determined by GPC (after precipitation in
methanol).
Synthesis of the macro-initiator. The chemical modification of the
original PK was performed according to the published method 25 (Scheme
3.1). The reactions were performed in a sealed 250 ml round bottom glass
reactor with a reflux condenser, a U-type anchor impeller using an oil bath
for heating.
Scheme 3.1: Synthesis of the macro-initiators
The chloropropylamine hydrochloride (9.89 g) was dissolved in methanol (90
ml) to which an equimolar amount of sodium hydroxide (2.16 g) was added.
After the polyketone (10 g) was preheated to the liquid state at the
employed reaction temperature (100 °C), the amine solution was added drop
wise (with a drop funnel) into the reactor in the first 20 min. The stirring
speed was set at a constant value of 500 RPM. During the reaction, the
mixture of the reactants changed from a slightly yellowish, low viscosity
state, into a highly viscous brown homogeneous paste. The product was
dissolved in chloroform and afterwards washed with demineralized water in a
separation funnel. The polymer was isolated by evaporating the chloroform at
low pressure (100 mbars). The product, a brown powder, was finally freeze
dried and stored at -18 °C until further use. The macro-initiator was
characterized using elemental analysis, 1H-NMR spectroscopy (in chloroform),
66
Chapter 3
and Gel Permeation Chromatography (GPC). The conversion of carbonyl
groups of the polyketone was determined using the following formula:
(3.1)
, being the average number of carbons in n-m (see Scheme
3.1)
, being the average number of carbons in m (see Scheme 3.1)
molecular weight of nitrogen
molecular weight of carbon
The average number of pyrrole units was determined using the conversion of
the carbonyl groups of the polyketone and formula 3.2:
(3.2)
=
the
average
molecular
weight
of
the
parent
(unmodified)
polyketone
= the
average
molecular weight
of the
polyketone
repeating unit
Comb polymerization. A 250-ml three-necked flask was charged with the
macro-initiator (e.g. entry 11: 0.3293 g, 0.117 mmol). Sufficient acetone
(typically 5-10 ml) was added to dissolve the macro-initiator. Demineralized
water (60 ml) and acrylamide (10 g, 140 mmol) were then added to the
solution. Subsequently, the mixture was degassed by three freeze-pumpthaw cycles. A nitrogen atmosphere was maintained throughout the
remainder of the reaction steps. CuBr (27 mg) was then added to the flask
and the mixture stirred for 10 minutes. The flask was then placed in an oil
bath at 25 °C. The reaction was started by the addition of the ligand
(Me6TREN, 34 mg) using a syringe.
After the pre-set reaction time, the
mixture was exposed to air and the polymer was precipitated in a tenfold
amount of methanol. For the higher molecular weight polymers the solution
was first diluted with demineralized water before being precipitated. The
polymer was isolated by filtration and subsequently dried in an oven at 65
°C.
67
Synthesis of branched polyacrylamide
To investigate whether all the initiation sites on polyketone are reactive (for
acrylamide)
a
lower
monomer
to
initiator
ratio
was
chosen.
The
polymerization using PK30-Cl12 as the macro-initiator was analogous to the
comb polymerization described earlier. The chosen monomer to macroinitiator ratio was relatively low (150:1) so that even at a high conversion
only a few acrylamide units are inserted. A sample was taken after 30
minutes and a
1
H-NMR spectrum was recorded. ChemBioDraw Ultra 12.0
(CambridgeSoft) was used to simulate the 1H-NMR spectrum of the macroinitiator with only few acrylamide units attached, and interpretation was
performed according to literature.26
Block copolymerization. The macroinitiator was prepared according to
the aforementioned procedure. A round bottomed three necked flask was
charged with the macroinitiator (3.6 g, 0.006 mmol) and NIPAM (36 g, 318
mmol). Double distilled water was added, and the mixture was degassed by
three freeze-pump-thaw cycles. Afterwards CuBr (4 mg, 0.028 mmol) was
added and the solution was stirred for 10 min. The flask was placed in a
water bath at 25 °C and the reaction was started by adding Me 6TREN (6.5
mg, 0.028 mmol). All the operations were performed under nitrogen. At set
time intervals aliquots were taken and analyzed by 1H-NMR.
Star polymerization. A 250-ml three-necked flask was charged with
AM (e.g. entry 8, Table 3.2: 5.0 g) and the initiator (pentaerythritol
tetrakis(2-bromoisobutyrate), 26 mg). A magnetic stirrer and distilled water
(30 ml) were added and subsequently degassed by three freeze-pump-thaw
cycles. The flask was then placed in an oil bath at 25 °C, CuCl (31 mg) was
added and the mixture was stirred for 10 minutes. The reaction was started
by adding the ligand (Me6TREN, 44 mg) using a syringe. After the reaction
the mixture was exposed to air and the polymer was precipitated in a tenfold
amount of methanol. The polymer was dried in an oven at 65 °C up to
constant weight.
Characterization. The acrylamide conversion was measured by using
Gas Chromatography (GC). The samples (taken from the reaction mixtures)
were dissolved in acetone (polymer precipitates) and injected on a Hewlett
Packard 5890 GC with an Elite-Wax ETR column. The total molecular weight
(Mn,tot) is calculated by using the acrylamide conversion (monomer-initiator
ratio multiplied by the conversion). The span molecular weight (Mn,SPAN) is
calculated using the Mn,tot and is defined as two times the molecular weight of
one arm (star PAM) or two times the molecular weight of one arm plus the
molecular weight of the macro-initiator (comb PAM).
Gas
Chromatography-Mass
Spectrometry
(GC-MS)
was
used
to
investigate the presence of initiator after the ATRP of AM (using 3-chloro-168
Chapter 3
propanol as the initiator). A sample of the reaction mixture was taken and
precipitated in acetone. An acetone sample, containing 1000 ppm of 3chloro-1-propanol, was used as the blank. GC-MS measurements were
performed on a Hewlett Packard (HP) 6890 Series GC system coupled to a HP
6890 Series Mass Selective Detector. The GC was operated splitless and in
order to blow off the solvent a flow of 80 mL/min of Helium was applied 1
minute after injection, the injector temperature was 250 °C, and an injection
volume of 1 l was used. The temperature program for the oven was as
follows: 40 °C for 5 min followed by heating with 10 °C/min to 280 °C.
Helium was used as the carrier gas with a constant flow rate of 0.8 ml/min.
Nuclear Magnetic Resonance (NMR) spectra were recorded on a Varian
Mercury Plus 400 MHz spectrometer. For analysis D 2O was used as the
solvent.
GPC analysis of all the water-soluble samples was performed on a Agilent
1200 system with Polymer Standard Service (PSS) columns (guard, 10 4 and
103 Å) with a 50 mM NaNO3 aqueous solution as the eluent. The columns
were operated at 40 °C with a flow-rate of 1 ml/min, and a refractive index
(RI) detector (Agilent 1200) was used at 40 °C. The apparent molecular
weights and dispersities were determined using a PAM based calibration with
WinGPC software (PSS). The macroinitiators were analyzed by GPC using THF
(used as received) as the eluent with toluene as a flow marker. The analysis
was performed on a Hewlett Packard 1100 system equipped with three PL-gel
3 m MIXED-E columns in series. The columns were operated at 42 °C with a
flow-rate of 1 ml/min, and a GBC LC 1240 RI detector was used at 35 °C.
The apparent molecular weights and dispersities were determined using
polystyrene standards and WinGPC software (PSS).
The particle sizes of the different polymers were measured using a
Brookhaven ZetaPALS zeta potential and particle size analyzer. Dilute
(polymer concentration < 0.1 wt. %) aqueous solutions were prepared and
filtered prior to the measurement. The laser angle for the measurements was
set at 90 ° and a total of 10 runs were performed for each sample (the
reported value is the average).
Elemental
analysis of the
EuroEA3000-CHNOS
analyzer
macroinitiators was performed on the
(EUROVECTOR
Instruments
&
Software).
Approximately 2 mg of each sample is weighed and placed in tin samplecups. The reported values are the average of 2 runs.
Rheological characterization. The aqueous polymeric solutions were
prepared by swelling the polymers in water for one day and afterwards gently
stirring the solution for another day.
69
Synthesis of branched polyacrylamide
Viscometric
measurements
were
performed
on
a
HAAKE
Mars
III
(ThermoScientific) rheometer, equipped with a cone-and-plate geometry
(diameter 60 mm, angle 2°). Flow curves were measured by increasing the
shear stress by regular steps and waiting for equilibrium at each step. The
shear rate ( ) was varied between 0.1 – 1750 s-1. Dynamic measurements
were performed with frequencies ranging between 0.04 – 100 rad/s (i.e.,
6.37·10-3 – 15.92 Hz). It must be noted that all the dynamic measurements
were preceded by an oscillation stress sweep to identify the linear
viscoelastic response of each sample. With this, it was ensured that the
dynamic measurements were conducted in the linear response region of the
samples.
Fluorescence spectroscopy. Fluorescence spectra of the aqueous
polymer solutions were recorded on a Fluorolog 3-22 spectrofluorimeter. The
excitation wavelength was set at 350 nm and the spectra were recorded
between 365 and 600 nm. The slit width of the excitation was 3 nm while
that of the emission was maintained at 2 nm. All the measurements were
performed in demineralized water at 10 °C.
3.3. Results and discussion
Macroinitiator. The synthesis of the macroinitiator was performed
according to the Paal-Knorr reaction of a halogenated primary amine with
aliphatic perfectly alternating polyketones (Scheme 3.1). The conversion of
the reaction was determined using elemental analysis (Table 3.1). Resonance
peaks corresponding to the pyrrole units were observed with
=5.68 ppm and validated by using model compounds.
25
1
H-NMR at
The average
number of pyrrole units equals the number of side chains which is obtained
after the polymerization of acrylamide by ATRP.
Table 3.1: Properties of the macroinitiator and parent polyketone
Sample
(PK00-xa)
Elemental composition
(C : H : N, wt%)
PK30
67.0 : 8.4 : 0
PK30-Cl12, R1 = Cl
64.2 : 7.8 : 4.6
XCO (%)b
Pyrrole
unitsc
Mn,GPC (g/mol)
PDI
-
0
2 797
1.74
55.10
12
2 093
1.96
a. Number indicates the ethylene content (%) and Cl indicates the halogen present
b. The conversion of the carbonyl groups of the polyketone
c. Average number of pyrrole units per chain
The macroinitiator was analyzed by
1
H-NMR (Figure 3.1). As can be
observed, the resonances corresponding to the pyrrole units (a) and the
70
Chapter 3
aliphatic protons of the amine moiety (b-d) appear in the spectrum of the
chemically modified polyketone.
a
c
a
b
d
c
d
PK30-Cl12
b
a
PK30-virgin
7
6
5
4
3
2
1
ppm
Figure 3.1: 1H-NMR spectra of the macroinitiator and the virgin polyketone
The obtained, chemically modified polyketone can be used as macroinitiator
in the ATRP of acrylamide for the preparation of comb-shaped polymers.
ATRP of AM using a primary halogen. The macroinitiator contains
primary halogens. This has mainly to do with better commercial availability of
the corresponding reagent (amino compound in Scheme 3.1) with respect to
ones containing a secondary or tertiary halogen. Despite the reported worse
performance in ATRP for primary halogens with respect to secondary or
tertiary ones27, this choice is driven by the possible future application at
industrial level. However, before proceeding to the ATRP of AM using the
macroinitiator, it is of paramount importance to confirm that primary
halogens can also lead to the ATRP of AM. This is particularly true when
making allowance for the reported lack in initiation efficiency27, which would
lead to the preparation of poorly defined structures. We started by
investigating the controlled nature of the polymerization. Similar to the ATRP
of AM using MeClPr as the initiator17, the reaction kinetics for the
disappearance of AM, using either chloro acetate or the macroinitiator, show
a non-linear relationship (Figure 3.2). It fits the model presented by Goto
and Fukuda28 quite well, thus, indicating that the non-linearity of the plot
71
Synthesis of branched polyacrylamide
stems from the progressive deactivation of the catalyst by complexation with
the growing polyacrylamide. The conversion index (ln[
/
]) is represented
by equation 3.3.
(3.3)
where
is the equilibrium constant in ATRP,
constant,
is
the
termination
concentration at time zero,
rate
is the propagation rate
constant,
is
the
monomer
is the monomer concentration at any time, and
is the initial initiator concentration.
2,0
2,0
2
R (model) = 0.99
2
R (model) = 0.82
1,6
1,6
1,2
1,2
ln (M 0/M)
ln (M 0/M)
Entry 1 (Table 2),
Entry 14 (Table 2),
0,8
0,8
0,4
0,4
0,0
0,0
0
10
20
30
40
50
60
0
2
4
Time (min)
6
8
Time
2/3
10
12
14
2/3
(min )
Figure 3.2: Kinetic plot for the ATRP of AM (entry 1 & 14, Table 3.2), on a linear (A)
time scale, and (B) on a scale of time2/3
Throughout the reaction for the linear PAM, the molecular weight increases
linearly with conversion and the dispersity remains relatively low (PDI < 1.5).
The molecular weight values are close to the theoretical ones (Figure 3.3,
Entry 1). Although the initiation of primary halogen suffers from low
activity27, the combination of a highly active ligand27 (Me6TREN) with water
(known
to
accelerate
ATRP
reactions17)
provides
control
over
the
polymerization of AM. For the branched PAM, the molecular weights differ
from the theoretical values, possibly as a result of the architectural difference
between the standards used for the GPC (all linear polymers) and the
synthesized PAM. Indeed, as the branches increase in size the differences (in
hydrodynamic volume) with a linear polymer increase.29 Nevertheless, the
increase in apparent molecular weight with conversion and the decrease in
72
16
Chapter 3
the PDI (and later on the block copolymerization with NIPAM) provide strong
evidence for the controlled nature of the polymerization.
3,0
3,0x10
4
2,5x10
4
2,8
Mn,GPC
2,6
Mn,theoretical
2,4
PDI
2,0x10
4
1,5x10
4
1,0x10
4
2,2
2,0
1,8
1,6
1,4
1,2
1
5,0x10
0,0
0 2 4 20
30
40
50
60
70
80
90
Polydispersity index (PDI)
Molecular weight (g/mol)
Entry 1, Table 2
1,0
100
Conversion (%)
6,0
Molecular weight (g/mol)
Entry 14, Table 2
2,5x10
5
5,5
Mn,GPC
5,0
Mn,theoretical
2,0x10
5
1,5x10
5
1,0x10
5
5,0x10
4
4,5
PDI
4,0
3,5
3,0
2,5
2,0
1,5
0,0
0
5
10
15
20
25
30
35
40
45
Polydispersity index (PDI)
3,0x10
5
1,0
96 98 100
Conversion (%)
Figure 3.3: Dependence of the Mn and PDI on the conversion of AM, entry 1 & 14
(Table 3.2); dotted lines serve as guides
It is crucial, for determining the architectural purity of the comb-shaped
polymers, to establish the initiation efficiency of the system. This has been
performed via
1
H-NMR for the branched polymer (see below), but also
through the use of a model compound, 3-chloro-1-propanol (entry 2, Table
3.2). It was confirmed with GC-MS (of the reaction mixture) that no initiator
(below the detection level of the GC-MS) was present after the ATRP with
AM. This is strong evidence for high initiation efficiency.
73
Synthesis of branched polyacrylamide
Comb polymerizations. Comb PAM has been prepared according to Scheme
3.2.
Scheme 3.2: Synthesis of the comb PAMs
The presence of many halogen atoms on a relatively short polymeric chain
(Mn of the macro-initiator is 2797 g/mol) might lead to steric hindrance in the
addition of the first AM units to the C-Cl bonds. To determine whether the
PAM chains grow on each halogen of the macroinitiator (PK30-Cl12) a 1H-NMR
spectrum was recorded after the reaction (Figure 3.4).
PK30-Cl
PK30-Cl12-graft-PAM
12
A
B
B
A
PK30-Cl12-graft-PAM
PK30-Cl
4
3
2
12
1
ppm
Figure 3.4: 1H-NMR spectra of the PK30-Cl12 (macro-initiator) and the PAM grafted
product (PK30-Cl12-graft-PAM)
74
Chapter 3
Given the low monomer/macro-initiator ratio (150:1), in theory, only a few
acrylamide units should be present on the polyketone backbone. The
spectrum of the corresponding polymeric material (PK30-Cl12-graft-PAM) is
compared with the one of the corresponding macro-initator (PK30-Cl12),
taken here as reference. The resonance at 3.5 ppm corresponds with the two
-hydrogens next to the chlorine functionality in the PK30-Cl12 macroinitiator. In the spectrum of the product this resonance disappears (at least
within the experimental error of 1H-NMR), thus confirming the reaction on the
halogen. The appearance of the resonance at 4.3 ppm in the product
spectrum, corresponding with the –hydrogen of the chlorine functionality
attached at the acrylamide chain end, further confirms the AM polymerization
at the halogen initiation point. This in combination with the model compound
(entry 2, Table 3.2) confirms that the average number of arms is equal to the
average number of halogens per chain.
Table 3.2: Characteristics of the (co)polymers
Architecture Entry
Linearf
[M]0:[I]0:[CuCl]0:
[Me6TREN]0a
M/s1/s2b
(w:v:v);
T; Time (min)
Conv.
(%)
Mn,tot
Mn,GPC
PDIc Mn,SPAN
1d
2e
479:1:1.5:1.5
9511:1:1.5:1.5
1:6;
1:3;
25 °C; 60
25 °C; 30
76.6
19.1
28 623
129 124
21 100
84 692
1.47 28 623
1.72 129 124
3
4
5
6
7
966:1:1.5:1.5
1 625:1:1.5:1.5
4 354:1:1.5:1.5
8 790:1:1.5:1.5
14 399:1:1.5:1.5
1:6;
1:6;
1:6;
1:6;
1:6;
25
25
25
25
25
°C; 60
°C;120
°C; 60
°C; 25
°C; 15
75.3
84.7
69.1
59.5
50.8
51 703 38 310
97 833 69 100
213 852 108 800
371 752 131 660
519 928 210 200
1.57 51 703
2.18 97 833
2.30 213 852
3.23 371 752
2.25 519 928
1:6;
1:6;
1:6;
25 °C;180
25 °C;180
25 °C;120
77.5
76.4
62.6
108 246 79 680
156 670 107 800
258 567 216 500
2.06 54 123
1.92 78 335
2.01 129 284
77.7
74.8
72.5
47.6
68.8
66 109
127 337
309 507
304 608
587 766
2.86 13 815
2.31 24 020
2.33 54 382
1.88 53 565
1.97 100 758
Star
8
9
10
1 965:1:6.0:6.0
2 884:1:6.0:6.0
5 811:1:6.0:6.0
Combg
11
12
13
14
15
1 197:1:1.5:1.5
2 395:1:1.5:1.5
6 006:1:1.5:1.5
9 003:1:1.5:1.5
12 025:1:1.5:1.5
1:6:1/3;25 °C;
1:6:3.0;25 °C;
1:8:1.5;25 °C;
1:6:1.0;25 °C;
1:6:1/3;25 °C;
60
60
60
60
60
72 020
104 900
206 400
188 800
271 600
a. Molar ratio
b. M/s1/s2 = Monomer / solvent 1 / solvent 2 = Acrylamide / water / acetone
c. The PAM polymers are prepared solely in water (except the comb were some acetone is used as
a cosolvent for the macroinitiator)
d. Initiator = chloro acetate
e. Initiator = 3-chloro-1-propanol
f. Initiator = methyl 2-chloropropionate
g. Comb PAMs with varying arm molecular weight and relatively low dispersities can be readily
prepared by changing the monomer-initiator ratio. The dispersities of the comb PAMs decrease as
the Mn,tot increases.
75
Synthesis of branched polyacrylamide
The 1H-NMR spectrum of the PK30-g-PAM shows that the halogen atoms are
reactive towards AM insertion. This enables the preparation of comb-like
polymers with a controlled number of branches as well as branch length. This
has been achieved by systematically changing the monomer/initiator ratio
(Table 3.2). The characteristics of the corresponding linear and star-shaped
PAM (for comparison of the rheological properties in aqueous solutions) are
also provided in Table 3.2.
Comb copolymerization, synthesis of PK30-g-(PAM-b-PNIPAM).
To further demonstrate the control of the polymerization (i.e. no loss of the
halogen end group), block copolymers of PK30-g-(PAM-b-PNIPAM) were
prepared. The 1H-NMR spectra of samples of the reaction mixture at different
times are displayed in Figure 3.5. As can be observed in Figure 3.5, the
resonance (2) of the methyl groups of NIPAM
increase in relation to the
resonances (1) corresponding to the backbone of the copolymer.
MeOH
1
2
1440 min
480 min
360 min
240 min
120 min
macroinitiator
5
4
3
ppm
2
1
Figure 3.5: 1H-NMR spectra of the block copolymer at different reaction times
76
Chapter 3
The NIPAM blocks increase in size as the reaction proceeds. This is strong
evidence for the controlled character of the reaction.
Rheological properties. Early studies4, 30 on solution properties of long
chain branched PAM demonstrated that the hydrodynamic volume of a
branched PAM is lower than for its linear analogue (of same molecular
weight). A lower hydrodynamic volume is synonymous to a lower solution
viscosity in dilute solutions. The influence of the molecular architecture on
the rheological behavior of polymers has already been investigated for
different polymers, mostly in the melt.
polyisoprenes
polystyrene
31,
39
35, 48, 49
,
polypropylene
36,
31-38
40-42
It was demonstrated that for
,
polyethylene37,
43-47
and
an enhancement of the zero shear rate viscosity (0) can
be achieved by changing the architecture (linear compared to star, long chain
branched, comb, and H-shaped) of the polymers. In particular, several
experiments
31
display an exponential increase in the 0 with an increase in
the arm molecular weight (Mw,arm). At relatively low total molecular weights
(Mw < 10000 g/mol for HDPE
50
, Mw < 100000 g/mol for polybutadienes32, Mw
< 600000 g/mol for polystyrene49) the η0 of the branched (comb, long chain
branched, and H-shaped) polymers is lower compared to their linear
analogue.
However,
as
the
molecular
weight
increases
(above
the
aforementioned values) the η0 of the branched polymers rapidly surpasses
(given its exponential dependence on the Mw,arm) the value of the linear ones.
Solution viscosity. The molecular weight determination with GPC is
based on the hydrodynamic volume. The comparison between linear, star
and comb-shaped PAM at similar Mn,tot (entries 4, 8 and 12 in Table 3.2)
using the GPC data show that the hydrodynamic radius of the comb PAM is
larger. This suggests a more extended nature of the arms of the comb PAM in
water solution. The PAM side chains originate from a small backbone (Mn =
2093 g/mol) and therefore steric hindrance might lead to extended PAM side
arms in comparison to linear PAM. Similar results have been reported for
poly(acrylic acid) grafts on a polydextran backbone.51 When the solution
viscosity is plotted against the polymer concentration (Figure 3.6) a markedly
different behavior can be observed for the branched/comb polymers
compared to their linear analogues.
In Figure 3.6 three different PAM are compared, a linear, a (4-arm) star
and a comb-like (12-arm). The solution viscosity at
= 10 s-1 is similar for all
the polymers at low concentration. As the concentration of the polymeric
solution increases the observed behavior depends on the architecture of the
polymer. The star polymer displays lower solution viscosity compared to their
linear analogue. This can be attributed to the lower hydrodynamic volume of
star polymers.29
77
Synthesis of branched polyacrylamide
Viscosity (Pa.s)
30
A
comb,
linear,
linear,
star,
20
entry 12
entry 5
entry 4
entry 8
10
0
0
2
4
6
8
10
12
14
16
14
16
Concentration (wt%)
Viscosity (Pa.s)
80
B
comb,
linear,
star,
linear,
60
entry 13
entry 6
entry 10
entry 5
40
20
0
0
2
4
6
8
10
12
Concentration (wt%)
Figure 3.6: Variation in the solution viscosity (measured at
= 10 s-1) as a function of
the polymer concentration and molecular weight. A: linear (2), star and a comb PAM at
a Mn,tot ~ 105000 g/mol and B: linear (2), star and a comb PAM at a Mn,tot ~ 230000
g/mol
The higher solution viscosity of the 12-arm comb-like PAM (Figure 3.6 A and
B) can be attributed to its higher Mn,tot (approximately 25% higher [3.6A] or
10% [3.6B]). However, the differences in solution viscosity are too high to be
attributed solely to the higher Mn,tot. To verify this hypothesis two linear PAMs
(entries 5 & 6) with a higher Mn,tot compared to that of the comb PAMs are
also displayed in Figure 3.6 A and B and as can be seen the solution
viscosities of both linear PAMs are lower than that of the comb. Nevertheless,
one would expect the linear polymer to display the highest solution viscosity
given the more compact structures of the star/branched polymers in
78
Chapter 3
solution.29 However, as can be observed, the comb-like PAM displays a
solution viscosity higher than both the linear analogues of similar (and
higher) molecular weight. In the semi-dilute regime entanglements are
present, and therefore melt like rheological properties can be the explanation
for the observed behavior.
The comparison between the polymers at similar Mn,tot is justified for
industrial applications. However, the three architecturally different polymers
can also be compared using a different approach, where the span molecular
weights (Mn,SPAN) of the star/branched polymers are similar to the molecular
weight of the linear one (Figure 3.7).31
0,06
A
linear,
star,
comb,
entry 3
entry 8
entry 13
0,05
60
Viscosity (Pa.s)
Viscosity (Pa.s)
80
40
20
linear,
star,
comb,
entry 3
entry 8
entry 13
0,03
0,02
0,01
0,00
0,0
0
0
2
4
A-Zoom
6
8
10
12
14
16
0,5
Concentration (wt. %)
0,40
linear,
star,
comb,
entry 4
entry 10
entry 15
0,30
Viscosity (Pa.s)
50
1,0
1,5
2,0
2,5
3,0
Concentration (wt. %)
B
60
Viscosity (Pa.s)
0,04
A-Zoom
40
30
20
B-Zoom
linear,
star,
comb,
entry 4
entry 10
entry 15
0,20
0,10
10
0,00
0
0
2
B-Zoom
4
6
8
10
12
14
16
Concentration (wt. %)
Figure 3.7: Viscosity (measured at
0,0
0,5
1,0
1,5
2,0
2,5
Concentration (wt. %)
= 10 s-1) as a function of the polymer
concentration and molecular weight. A; linear, star and a comb PAM with a similar
MN,SPAN (MN,SPAN ~ 52000 g/mol) and A-Zoom; zoom in of the dilute region. B; linear,
star and a comb PAM with a similar MN,SPAN (MN,SPAN ~ 105000 g/mol) and B-Zoom;
zoom in of the dilute region
79
3,0
Synthesis of branched polyacrylamide
As can be observed in Figure 3.7, the increase in solution viscosity with
concentration is dependent on the span molecular weight of the samples and
the molecular architecture. At the lowest molecular weight studied (Figure
3.7A) the solution viscosity of the star polymers increases in a similar fashion
(although slightly more pronounced) as the linear one whereas the comb-like
displays a more pronounced increase towards higher concentrations. At a
higher span molecular weight (Figure 3.7B) both the star and comb-like
polyacrylamides display a more pronounced increase in solution viscosity with
concentration than to the linear one (with similar Mn,SPAN), with the comb-like
one showing the highest viscosity. This is in line with the theory that
stipulates that the η0 increases exponentially with increase in the Mw,arm for
star/branched polymers31 (compared to a power law for linear polymers 52).
The longer the branches are, the more pronounced the differences between
the linear and branched polymers should be. These predictions are based on
experiments
performed
in
the
melt
(i.e.
fully
entangled
chains).
Nevertheless, the general parameters that affect the viscosity can also be
applied to polymers in solutions where entanglements are present.53, 54
As can be observed in Figure 3.7, the solution viscosities of the comb
and star-shaped PAMs at low polymer concentration are close to each other.
As the polymer concentration increases the solution viscosity of the comb
and star PAMs increase more rapidly than the linear PAMs.
Clear differences in the solution viscosity can be observed when
comparing the architecturally different polymers at high concentration, i.e.
above the overlap concentration. However, as can be observed in Figure 3.6
and 3.7, at low polymer concentration the differences are rather small and
therefore difficult to detect. In order to gain deeper insight, dilute polymer
solutions
are
compared,
and
experiments
aimed
at
demonstrating
hydrophobic associations are performed.
In the dilute region of a polymeric solution, where no entanglements are
present, the viscosity can be described using the “free draining” chain model.
The solution viscosity is determined by the solvent viscosity and the excess
viscosity caused by the energy consumption of a tumbling polymer coil under
flow. According to Stokes and Evans55 the excess viscosity of a solution
(containing Nav·C / Mn macromolecules) is:
(3.4)
where  is the solvent viscosity,
degree of polymerization,
is the zero shear rate viscosity,
is the friction factor per segment,
hydrodynamic radius as determined by light scattering measurements,
80
is the
is the
is
Chapter 3
the Avogrado constant,
is the polymer concentration and
is the
molecular weight of the polymer. The viscosities at vanishing shear rate (
are determined from the low-frequency loss moduli.
)
53
Equation 3.4 relates the excess viscosity (
) to the friction factor ( )
per segment. The latter can be easily evaluated (Figure 3.8A) by determining
the slope of the plot of
vs.
linear,
star,
comb,
1.4
entry 1
entry 6
entry 11
B
1.2
1.0
Slope
1000
.
0.8
0.6
0 (mPa.s)
0.4
100
0.2
0.0
A
10
1
1E24
1E25
1E26
2
-1
NpRg[(NavC)/Mw] (nm )
Figure 3.8: Plot of
vs.
(A) and the corresponding friction
factor per segment (B) for a linear, star and comb PAM
The corresponding
values (Figure 3.8B) are clearly not a function of the
molecular architecture since all differences are well within the experimental
error. This is quite important since it strongly suggests that the differences in
the solution viscosities (both at low and higher concentration) cannot be
attributed to differences in the segmental friction factor. The behavior
observed for the star PAM can be then attributed to the increase in
entanglement density as a result of the architecture. The comb PAM however
possesses a hydrophobic backbone and can therefore display hydrophobic
aggregations. Therefore, it is important to investigate whether or not
hydrophobic associations arise in solution. The comb-like PAM possesses
pyrrole units in the backbone making it possible to probe the solution
structure
with
fluorescence
spectroscopy.
The
critical
aggregation
concentration (CAC) can be determined from the corresponding spectra (data
81
Synthesis of branched polyacrylamide
not shown for brevity). The CAC values are 3 wt.% and 2 wt.% for entry 13
and 15 respectively. In Figure 3.7 (A-Zoom & B-Zoom) the upward trend of
the solution viscosity of entries 13 and 15 starts at lower concentrations than
their respective CAC. We can therefore conclude that the higher viscosity of
the comb polymers below the CAC is due to the molecular architecture
(longer relaxation time and thus a higher solution viscosity, similar to the
melt31 compared to a linear polymer) and above the CAC a combination of
the molecular architecture and hydrophobic associations.
Viscoelastic behavior. The elastic response of an aqueous polymeric
solution is dependent on the molecular weight 56, the concentration56 and the
architecture/chemical composition (presence of hydrophobic groups) of the
polymer.56, 57 In Figure 3.9 two different comparisons are presented.
3
10
= G"
= G'
2
10
90
80
1
10
70
-1
10
-2
10
-3
10
-4
} comb, entry 13
} star, entry 8
} linear, entry 3
10
-5
10
-6
10
A
-7
10
Phase angle
G' G" (Pa)
0
10
60
50
30
20
10
0
0,1
1
10
comb, entry 13
star,
entry 8
linear, entry 3
40
B
0,1
100
= G"
= G'
2
10
1
70
-1
10
-2
10
-3
10
-4
10
} comb,
} star,
} linear,
-5
10
10
C
entry 12
entry 8
entry 4
Phase angle
G' G" (Pa)
10
-7
100
80
0
10
10
90
10
-6
1
Frequency (rad/s)
Frequency (rad/s)
60
50
30
20
10
0
0,1
1
10
Frequency (rad/s)
100
comb, entry 12
star,
entry 8
linear, entry 4
40
D
0,1
1
10
Frequency (rad/s)
Figure 3.9: G’ & G” (A) and phase angle (B) as a function of the frequency for a 4arm star, 12-arm comb-like and linear at similar Mn,SPAN and a polymer concentration of
10.71 wt.% and G’ & G” (C) and phase angle (D) as a function of the frequency for a
4-arm star, 12-arm comb-like and linear at similar Mn,tot and a polymer concentration
of 10.71 wt.%
82
100
Chapter 3
The comparison between a linear, star and comb PAM of similar MN,SPAN
demonstrates that the comb PAM exhibits a more pronounced elastic
behavior, especially at low frequency (Figure 3.9B). When comparing a
linear,
4-arm star and comb at similar Mn,tot only a small difference is
observed at low frequency, i.e. a slightly more elastic behavior for the 4-arm
star and comb compared to the linear PAM (Figure 3.9D). However, at
relatively higher frequencies (> 1 rad/s) the differences become more
significant with the star PAM showing the highest elastic behavior (elastic
response 4-arm star > 12-arm comb > linear). The arms of the 12-arm comb
are shorter compared to the arms of the 4-arm star. At higher frequencies
(higher deformations) the disentanglement of the arms will occur more easily
for the comb given its shorter arms. It is also evident (Figure 3.9C) that the
transition from viscous to elastic behavior occurs at lower angular frequency
for the 4 arm star. Similar results were reported for polyethylene in the
melt.47
The model developed for the viscoelasticity of monodisperse comb
polymer melts50 predicts that the highest 0 (in the melt) for comb polymers
is obtained with combs having long arms but few branches (≤ 12). In
addition, an exponential dependence of the 0 on the molecular weight of the
arms is obeyed. The comparison between a regular 3-arm star and combs
polymers (at least the ones included in the comparison in the paper) show
that the 3-arm star possesses the highest 0. However, the model also
predicts that for a specific range of molecular weights (20000 < MW < 80000
g/mol) a comb polymer possessing 6 arms has a higher 0 compared to a 3arm star.50 For polyisoprene the 0 of a 3-arm star is lower than that of a 4arm star.31 Our data suggest that comb polymers in aqueous solution can
have a higher solution viscosity than a 4-arm star.
3.4. Conclusion
The controlled synthesis of linear, star and comb-shaped PAM by ATRP in
water has been achieved. All the initiation sites on the macroinitiator seem to
react during the ATRP, as strongly evidenced by
1
H-NMR and the use of
model compounds. GPC analysis demonstrates that the comb polymers
display a higher hydrodynamic volume in dilute water solution compared to
their linear and star analogues, preliminarily explained by the more extended
nature of the arms in the comb polymers. Rheological measurements in
(semi)dilute water solution demonstrated that the solution viscosity of comblike PAM is higher (whilst maintaining the concentration constant) than its
linear and star-shaped analogues both at equal Mn,SPAN and Mn,tot. In addition
83
Synthesis of branched polyacrylamide
the elastic response of water solution containing the comb-like PAM is more
pronounced than for the linear and star-shaped PAM (both at equal Mn,SPAN
and Mn,tot). The controlled synthesis of PAM with different architectures allows
the manipulation of the rheological properties of aqueous solution thereof. By
simply changing the architecture of the polymer, a significantly different
behavior, i.e. higher solution viscosity and more pronounced elastic response
at equal Mn,SPAN and Mn,tot, is obtained. The obtained results pave the way for
application of these polymeric materials in EOR.
3.5. Acknowledgement
This work is part of the Research Programme of the Dutch Polymer
Institute DPI, Eindhoven, the Netherlands, projectnr. #716.
3.6. References
1. Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Prog. Polym. Sci. 2011, 1558.
2. Shalaby W. Shalaby; Charles L. McCormick; George B. Butler Water-Soluble
Polymers: Synthesis, Solution Properties, and Applications; American Chemical
Society: Washington DC, 1991; .
3. Huang, S.; Lipp, D. W.; Farinato, R. S. Encyclopedia of Polymer Science and
Technology; John Wiley & Sons, Inc.: 2002; .
4. Kulicke, W. M.; Horl, H. H. Colloid Polym. Sci. 1980, 7, 817.
5. Gleason, E. H.; Miller, M. L.; Sheats, G. F. Journal of Polymer Science 1959, 133,
133.
6. Fanood, M. H. R.; George, M. H. Polymer 1988, 1, 128.
7. Fanood, M. H. R.; George, M. H. Polymer 1988, 1, 134.
8. Fanood, M. H. R. Iranian Polymer Journal 1998, 1, 59.
9. Guha, S. Journal of the Indian Chemical Society 2008, 1, 64.
10. Jewrajka, S. K.; Mandal, B. M. Journal of Polymer Science Part A-Polymer
Chemistry 2004, 10, 2483.
11. Jewrajka, S. K.; Mandal, B. M. Macromolecules 2003, 2, 311.
12. Tan, Y.; Yang, Q.; Sheng, D.; Su, X.; Xu, K.; Song, C.; Wang, P. E-Polymers
2008, 25.
13. Ito, M.; Ishizone, T. Journal of Polymer Science Part A-Polymer Chemistry 2006,
16, 4832.
14. Wang, W.; Wang, D.; Li, B.; Zhu, S. Macromolecules 2010, 9, 4062.
15. Boyer, C.; Bulmus, V.; Davis, T. P.; Ladmiral, V.; Liu, J.; Perrier, S. Chem. Rev.
2009, 11, 5402.
16. Matyjaszewski, K.; Xia, J. H. Chem. Rev. 2001, 9, 2921.
17. Wever, D. A. Z.; Raffa, P.; Picchioni, F.; Broekhuis, A. A. Macromolecules 2012,
10, 4040.
18. Appel, E. A.; del Barrio, J.; Loh, X. J.; Dyson, J.; Scherman, O. A. Journal of
Polymer Science Part A-Polymer Chemistry 2012, 1, 181.
19. Matyjaszewski, K.; Miller, P. J.; Pyun, J.; Kickelbick, G.; Diamanti, S.
Macromolecules 1999, 20, 6526.
20. Neugebauer, D.; Zhang, Y.; Pakula, T.; Sheiko, S. S.; Matyjaszewski, K.
Macromolecules 2003, 18, 6746.
21. Qin, S. H.; Matyjaszewski, K.; Xu, H.; Sheiko, S. S. Macromolecules 2003, 3, 605.
84
Chapter 3
22. Borner, H. G.; Beers, K.; Matyjaszewski, K.; Sheiko, S. S.; Moller, M.
Macromolecules 2001, 13, 4375.
23. Drent, E.; Keijsper, J. J. US Pat. 5225523, 1993.
24. Mul, W. P.; Dirkzwager, H.; Broekhuis, A. A.; Heeres, H. J.; van der Linden, A. J.;
Orpen, A. G. Inorg. Chim. Acta 2002, 147.
25. Zhang, Y.; Broekhuis, A. A.; Stuart, M. C. A.; Picchioni, F. J Appl Polym Sci 2008,
1, 262.
26. Silverstein, R. M.; Webster, F. X.; Kiemle, D. J. Spectrometric Identification of
Organic Compounds; John Wiley & Sons Inc.: 2005; , pp 512.
27. Braunecker, W. A.; Matyjaszewski, K. Progress in Polymer Science 2007, 1, 93.
28. Goto, A.; Fukuda, T. Progress in Polymer Science 2004, 4, 329.
29. Burchard, W. Branched Polymers II 1999, 113.
30. Kulicke, W. -.; Kniewske, R.; Klein, J. Progress in Polymer Science 1982, 4, 373.
31. Fetters, L. J.; Kiss, A. D.; Pearson, D. S.; Quack, G. F.; Vitus, F. J. Macromolecules
1993, 4, 647.
32. Kraus, G.; Gruver, J. T. J. Polym. Sci. Part A 1965, 1PA, 105.
33. Mykhaylyk, O. O.; Fernyhough, C. M.; Okura, M.; Fairclough, J. P. A.; Ryan, A. J.;
Graham, R. Eur. Polym. J. 2011, 4, 447.
34. Robertson, C. G.; Roland, C. M.; Paulo, C.; Puskas, J. E. J. Rheol. 2001, 3, 759.
35. Graessley, W. W.; Roovers, J. Macromolecules 1979, 5, 959.
36. Auhl, D.; Stange, J.; Munstedt, H.; Krause, B.; Voigt, D.; Lederer, A.; Lappan, U.;
Lunkwitz, K. Macromolecules 2004, 25, 9465.
37. Gabriel, C.; Munstedt, H. Rheol. Acta 2002, 3, 232.
38. Münstedt, H. Soft Matter 2011, 6, 2273.
39. Frischknecht, A. L.; Milner, S. T.; Pryke, A.; Young, R. N.; Hawkins, R.; McLeish, T.
C. B. Macromolecules 2002, 12, 4801.
40. Gotsis, A. D.; Zeevenhoven, B. L. F.; Tsenoglou, C. J. J. Rheol. 2004, 4, 895.
41. McCallum, T. J.; Kontopoulou, M.; Park, C. B.; Muliawan, E. B.; Hatzikiriakos, S. G.
Polym. Eng. Sci. 2007, 7, 1133.
42. Islam, M. T.; Juliani; Archer, L. A.; Varshney, S. K. Macromolecules 2001, 18,
6438.
43. Gabriela, C.; Munstedt, H. J. Rheol. 2003, 3, 619.
44. Wood-Adams, P. M.; Dealy, J. M. Macromolecules 2000, 20, 7481.
45. Lohse, D. J.; Milner, S. T.; Fetters, L. J.; Xenidou, M.; Hadjichristidis, N.;
Mendelson, R. A.; Garcia-Franco, C. A.; Lyon, M. K. Macromolecules 2002, 8,
3066.
46. Zamponi, M.; Pyckhout-Hintzen, W.; Wischnewski, A.; Monkenbusch, M.; Willner,
L.; Kali, G.; Richter, D. Macromolecules 2010, 1, 518.
47. Gabriel, C.; Kokko, E.; Lofgren, B.; Seppala, J.; Munstedt, H. Polymer 2002, 24,
6383.
48. Roovers, J.; Graessley, W. W. Macromolecules 1981, 3, 766.
49. Roovers, J. Macromolecules 1984, 6, 1196.
50. Inkson, N. J.; Graham, R. S.; McLeish, T. C. B.; Groves, D. J.; Fernyhough, C. M.
Macromolecules 2006, 12, 4217.
51. Kutsevol, N.; Guenet, J. M.; Melnik, N.; Sarazin, D.; Rochas, C. Polymer 2006, 6,.
52. Degennes, P. G. J. Chem. Phys. 1971, 2, 572.
53. Ferry, J. D. Viscoelastic properties of polymers; John Wiley & Sons: New York,
1980; , pp 641.
54. Nielsen, L. E. Polymer rheology; Marcel Dekker Inc.: New York, 1977; , pp 207.
55. RJ, S.; DF, E. Fundamentals of Interfacial Engineering; Wiley-VCH: United States
of America, 1997; , pp 701.
56. Raju, V. R.; Menezes, E. V.; Marin, G.; Graessley, W. W.; Fetters, L. J.
Macromolecules 1981, 6, 1668.
57. Volpert, E.; Selb, J.; Candau, F. Macromolecules 1996, 5, 1452.
85
Synthesis of branched polyacrylamide
This page intentionally left blank
86
Chapter 4
Chapter 4
Control over the viscoelasticity of
aqueous polyacrylamide solutions
by tailoring the polymer
architecture
Abstract
The
controlled
synthesis
of
high
molecular
weight
comb-like
polyacrylamide (PAM) has been accomplished using atomic transfer radical
polymerization (ATRP) of acrylamide (AM) in water at room temperature. The
number and length (molecular weight) of the arms was varied. In addition,
the overall molecular weight of the macromolecule was also varied (i.e.
macromolecules with equal number of longer arms). Halogen-functionalized
aliphatic polyketones acted as macroinitiators in the polymerization. The
obtained branched polymers were used in water solutions to study the effect
of the molecular architecture on the rheological properties. For comparison
purposes, linear PAM was synthesized using the same procedure. The
intrinsic viscosities and light scattering data suggest that the 13- and 17-arm
PAMs are more extended compared to the linear, 4- and 8-arm analogues.
The comparison of linear, 4-, 8-, 12-, 13- and 17-arm PAM in semi-dilute
solutions demonstrated that the 13- and the 17-arm have the highest
solution viscosity at equal molecular weight. Depending on the PAM
molecular weight and concentration, a significant (as much as 5-fold)
increase in solution viscosity (at a shear rate of 10 s-1) is observed. The
elastic response of aqueous solutions containing the polymers critically
depended on the molecular architecture. Both the 4- and 8-arm polymers
displayed a larger phase angle value compared to the linear analogue. The
13- and 17-arm PAMs displayed a lower phase angle than the linear one.
Ultimately, the rheological properties are dependent on the number of arms
present. The combination of a higher hydrodynamic volume and higher
87
Rheological properties of branched polyacrylamides
entanglement density leads to an improved thickening efficiency (for number
of arms (N) ≥ 13). The improved thickening efficiency of the branched (N ≥
13) PAMs makes these polymers highly interesting for application in
Enhanced Oil Recovery.
Based on: D.A.Z. Wever, L.M. Polgar, M.C.A. Stuart, F. Picchioni, A.A.
Broekhuis. Polymer molecular architecture as tool for controlling rheological
properties of aqueous polyacrylamide solutions for enhanced oil recovery.
Industrial
&
Engineering
10.1021/ie403045y.
88
Chemistry
Research,
2013,
DOI:
Chapter 4
4.1. Introduction
Polyacrylamide (PAM) is a versatile industrial polymer that finds use in
wastewater treatment, cosmetics and enhanced oil recovery (EOR) 1. In
particular, the main purpose of using PAM (mostly in water solution) resides
in the corresponding improvement of the rheological properties. Indeed, in
most applications, an enhancement of the solution viscosity is required.
However, in EOR, it has been concluded that, at equal viscosity, the
viscoelasticity of the solution plays a crucial role in ensuring a high oil
recovery2-7. Such rheological behavior arises from the extremely high
molecular weight (typically Mw ≈ 2·107 g/mol) and the ionic character of the
water soluble polymer employed. The presence of electric charges along the
backbone results (in deionized water) in the stretching of the polymer
chains/coils and ultimately in larger viscosity values. In this context, the use
of partially hydrolyzed PAM (HPAM) represents the most popular choice. The
importance
of
the
solution
elastic
response
has
been
supposedly
demonstrated2-7 by comparing a water solution of HPAM and one of glycerin
in flow experiments specifically designed to simulate oil recovery processes.
However, such comparison might be not completely correct since HPAM is a
high molecular weight polyelectrolyte while glycerin a small molecule. Such
difference in structure of the used chemicals as well as of the corresponding
water solution might indeed result in differences also in other properties (e.g.
surface tension between oil and water), thus hindering a direct correlation of
the observed effect and the supposed cause, in this case the elastic behavior
of the water solution. A better comparison would be between polymeric
solutions where the viscoelasticity is systematically changed. However, for
water soluble PAM a systematic change in the elastic response without
affecting other properties (i.e. molecular weight and dispersity) is difficult.
One approach can be the controlled synthesis of PAM. However, the
monomer itself (acrylamide) represents a difficult candidate to polymerize in
a controlled fashion.8
Controlled synthesis of branched PAM has only limitedly been reported in
literature. In the past, high conversion and high temperature in conventional
free radical polymerization was demonstrated to lead to uncontrolled
branched polyacrylamide.9-14 By increasing the reaction temperature (from
room temperature to 90 °C) and the conversion level of acrylamide, more
branches could be obtained.10 The properties of the uncontrolled branched
PAM were evaluated with respect to their ability to perform as flocculants,
and it was concluded that linear PAM performed better than the uncontrolled
branched PAM. This was attributed to the inherent lower hydrodynamic
volume of the branched PAM.13,
14
Nevertheless, given the uncontrolled
89
Rheological properties of branched polyacrylamides
nature of the polymerization procedure, a mixture of products is synthesized
with no well-defined structure. Controlled radical polymerization for the
preparation of hyperbranched PAM has been recently reported. 15 The
hyperbranched
fragmentation
PAMs
chain
were
synthesized
transfer
(RAFT)
using
reversible
polymerization.
addition-
Although
the
polymerization is a controlled one, the branching occurs randomly. 15
Therefore the control in architecture of the PAM is limited and no correlation
between molecular architecture and rheological properties can be obtained.
Recently controlled synthesis of PAM has been reported in water-ethanol
mixtures16 and, by our group, in water.17 In a water-ethanol mixture, linear
PAM (with molecular weights up to >350 000 g/mol and dispersities as low as
1.10) could be synthesized.16 The molecular weights of PAM reached values
>150 000 g/mol (with dispersities as low as 1.39) in water using the same
catalyst/initiation system.17 With the accomplishment of atomic transfer
radical polymerization (ATRP) of acrylamide, the controlled preparation of
branched PAM can be envisaged. This enables the systematic study of the
structure-property relationships of PAM (with different topologies) in water
solutions. The aim of this work is to prepare in a controlled fashion branched
PAM with varying numbers (and molecular weight) of arms and to investigate
the
effect
of
the
architecture
on
the
rheological
properties
of
the
corresponding water solutions. To the best of our knowledge, this represents
an absolute novelty, in terms of synthetic strategy as well as structureproperty relationship, of the present chapter.
4.2. Experimental section
Chemicals. Acrylamide (AM) (electrophoresis grade, ≥99%), PAM (M w =
5-6·106
g/mol),
tris[2-(dimethylamino)ethyl]amine
(Me6TREN),
2,2-
bipyridine (bpy), copper(I) chloride (CuCl, 98%), copper(I) bromide (CuBr,
98%),
methyl
2-chloropropionate
MeClPr,
97%),
3-chloropropylamine
hydrochloride (98%), and sodium hydroxide (pellets) were purchased from
Sigma Aldrich. CuCl and CuBr were purified by stirring in glacial acetic acid
(Aldrich), washing with glacial acetic acid, ethanol and diethyl ether (in that
order) and then dried under vacuum. All solvents were reagent grade and
used without further purification. The alternating polyketones with 30 mol%
ethylene content (PK30, Mn = 2800 g/mol, PDI = 1.74) was synthesized
according to a published procedure.18, 19
Macroinitiators. The PK30 functionalization was performed according
(Scheme 4.1) to the published method.20 The reactions were performed in a
90
Chapter 4
sealed 250 ml round bottom glass reactor with a reflux condenser, a U-type
anchor impeller, and an oil bath for heating.
Scheme 4.1: Synthesis of the macro-initiators
For the preparation of PK30-Cl12 (taken here as representative example), 3chloropropylamine hydrochloride (9.89 g, 53.6 mmol) was dissolved in
methanol (90 ml) to which an equimolar amount of sodium hydroxide (2.15
g, 53.6 mmol) was added. After the polyketone (10 g, 0.076 mol of dicarbonyl units) was preheated to the liquid state at the employed reaction
temperature (100 °C), the amine was added drop wise (with a drop funnel)
into the reactor in the first 20 min. The stirring speed was set at a constant
value of 500 RPM. During the reaction, the mixture of the reactants changed
from the slight yellowish, low viscous state, into a highly viscous brown
homogeneous paste. The product was dissolved in chloroform and afterwards
washed with demineralized water. The two phases (organic & water) were
separated in a separatory funnel. The polymer was isolated by evaporating
the chloroform at reduced pressure at room temperature. The product, a
brown viscous paste (low functionalization degree) or a brown powder (high
functionalization degree), was finally freeze dried and stored at -18 °C until
further use. Some properties of the macro-initiators are given in Table 4.1.
The macro-initiators were characterized using elemental analysis and 1H-NMR
spectroscopy (in chloroform). The conversion of carbonyl groups of the
polyketone was determined using the following formula:
(4.1)
, the average number of carbons in n-m (see Scheme 4.1)
, the average number of carbons in m (see Scheme 4.1)
molecular weight of nitrogen
molecular weight of carbon
91
Rheological properties of branched polyacrylamides
The average number of pyrrole units was determined using the conversion of
the carbonyl groups of the polyketone and formula 4.2:
(4.2)
= the average molecular weight of the parent (unmodified) polyketone
= the average molecular weight of the repeating unit of polyketone
Comb polymerization. A 250-mL three-necked flask was charged with the
macro-initiator. Sufficient acetone (typically 5-10 ml) was added to dissolve
the macro-initiator. Demineralized water and acrylamide were then added to
the solution. Subsequently, the mixture was degassed by three freeze-pumpthaw cycles. A nitrogen atmosphere was maintained throughout the
remainder of the reaction steps. CuX (X= Cl, Br) was then added to the flask
and the mixture stirred for 10 minutes. The flask was then placed in an oil
bath at 25 °C. The reaction was started by the addition of the ligand
(Me6TREN) using a syringe. After the pre-set reaction time, the mixture was
exposed to air and the polymer was precipitated in a tenfold amount of
methanol. For the higher molecular weight polymers the solution was first
diluted with demineralized water before being precipitated. The polymer was
isolated by filtration and subsequently dried in an oven at 65 °C.
As mentioned before, for the comb-shaped PAMs, the length and number
of arms was varied (Figure 4.1).
Figure 4.1: Schematic overview of the different architectures of the comb-shaped
PAMs
92
Chapter 4
Characterization. The acrylamide conversion was measured by using Gas
Chromatography (GC). Several different samples directly taken from the
reaction mixtures were dissolved in acetone (polymer precipitates) and
injected on a Hewlett Packard 5890 GC with an Elite-Wax ETR column. The
overall molecular weight (Mn,overall) is calculated using the acrylamide
conversion (monomer-initiator ratio multiplied by the conversion value). The
span molecular weight (Mn,SPAN) is calculated using the Mn,overall and is defined
as two times the molecular weight of one arm (star PAM) or two times the
molecular weight of one arm plus the molecular weight of the macro-initiator
(comb PAM).
Nuclear Magnetic Resonance (NMR) spectra were recorded on a Varian
Mercury Plus 400 MHz spectrometer. For analysis chloroform was used as the
solvent.
The particle sizes of the different polymers were measured using a
Brookhaven ZetaPALS zeta potential and particle size analyzer. Dilute
(polymer concentration < 0.1 wt. %) aqueous solutions were prepared and
filtered prior to the measurement. The laser angle for the measurements was
set at 90 ° and a total of 10 runs were performed for each sample (the
reported value is the average).
The macroinitiators were analyzed by GPC using THF (used as received)
as the eluent with toluene as a flow marker. The analysis was performed on a
Hewlett Packard 1100 system equipped with three PL-gel 3 m MIXED-E
columns in serie. The columns were operated at 42 °C with a flowrate of 1
ml/min, and a GBC LC 1240 RI detector was used at 35 °C. The apparent
molecular weights and dispersities were determined using polystyrene
standards and WinGPC software (PSS).
Cryo-Transmission Electron Microscopy (cryo-TEM). A drop of the
polymer solution was placed on a glow discharged holey carbon-coated grid.
After blotting away the excess of solution, the grids were rapidly plunged into
liquid ethane. The frozen specimen were mounted in a Gatan (model 626)
cryo-stage and examined in a Philips CM 120 cryo-electron microscope
operating at 120 kV. Micrographs were recorded under low-dose conditions.
Rheological characterization. The aqueous polymeric solutions were
prepared by swelling the polymers in water for one day and afterwards gently
stirring the solution for another day.
Viscometric measurements were performed on a HAAKE Mars III
(ThermoScientific) rheometer, equipped with a cone-and-plate geometry
(diameter 60 mm, angle 2°). Flow curves were measured by increasing the
shear stress by regular steps and waiting for equilibrium at each step. The
shear rate ( ) was varied between 0.1 – 1750 s-1. Dynamic measurements
93
Rheological properties of branched polyacrylamides
were performed with frequencies ranging between 0.04 – 100 rad/s (i.e.,
6.37·10-3 – 15.92 Hz). It must be noted that all the dynamic measurements
were preceded by an oscillation stress sweep to identify the linear
viscoelastic response of each sample and to ensure that the dynamic
measurements were conducted in the linear response region of the samples.
The viscosity function of the different polymeric solutions was modeled
using the Carreau-Yasuda model21, 22 (equation 4.3).
(4.3)
where
is the viscosity,
is the zero shear rate viscosity,
at infinite shear rate,
thinning,
is the viscosity
is the critical shear rate for the onset of shear
is the power law slope and
represents the width between
and the power law region.
4.3. Results and discussion
Macroinitiators. The synthesis of the macroinitiators was performed
according to the Paal-Knorr reaction (Scheme 4.1) of a halogenated primary
amine
with
aliphatic
perfectly
alternating
polyketones.
The
carbonyl
conversion was determined using elemental analysis (Table 4.1).
Table 4.1: Properties of the macro-initiators
Polyketone sample
(PK30-Cla)
PK30 (virgin)
PK30-Cl4
PK30-Cl8
PK30-Cl12
PK30-Cl13
PK30-Cl17
Elemental composition
(C : H : N, wt%)
67.0
58.6
64.0
64.2
62.9
73.7
:
:
:
:
:
:
8.4
7.1
7.9
7.8
7.6
7.9
:
:
:
:
:
:
0
1.6
3.3
4.6
4.9
6.1
XCO (%)b Pyrrole unitsc
18.87
37.21
55.10
61.14
81.27
0
4
8
12
13
17
Mn,GPC
PDI
2
2
2
2
2
2
1.74
2.02
2.01
1.76
1.97
2.18
797
447
244
093
072
117
a. Number indicates the ethylene content (%)
b. Carbonyl groups conversion as define by equation (4.1)
c. Average number of pyrrole units per chain as defined by equation (4.2)
Resonance peaks corresponding to the pyrrole units were observed with 1HNMR spectroscopy at  5.68 ppm while the -, -, and -hydrogens (relative
to the halogen) were detected at  3.51, 1.95, and 3.86 ppm respectively
(Figure 4.2). The formation of the pyrrole units was also previously
demonstrated using model compounds.20 As can be observed in Figure 4.2,
94
Chapter 4
the resonance of the pyrrole, - and -hydrogens (relative to the halogen) all
increase in magnitude with the conversion of the Paal-Knorr reaction.
The obtained, chemically modified polyketones are used as macroinitiators in the ATRP of acrylamide for the preparation of comb-polymers
with different number of side chains.
a
a
b
c
d
c
d
b
a
PK30-Cl17
c
d
b
PK30-Cl13
a
PK30-Cl8
a
PK30-Cl4
a
d
c
d
c
b
b
PK30, virgin
7
6
5
4
3
2
1
ppm
Figure 4.2: H-NMR spectra of virgin polyketone and the macroinitiators at different
conversion levels of the Paal-Knorr reaction
Comb polymerization. The synthesis of the comb-shaped PAM was
performed according to Scheme 4.2. The ratio between the (macro)initiator
and the monomer was varied in order to synthesize comb-shaped and linear
PAM with different molecular weights. Table 4.2 lists the results for the
different polymers prepared. The reaction of all the halogen sites on the
macroinitiator has already been demonstrated (Chapter 3).
As can be observed in Table 4.2, high molecular weight branched PAM
can be synthesized. An increase in the monomer:macroinitiator ratio leads to
higher
average
molecular
weights.
The
conversion
of
acrylamide
is
suppressed when a low amount of the co-solvent (acetone) is used. Similar
results were reported for the ATRP of acrylamide in water-ethanol mixtures,
where
an
optimum
exists
(30:70,
ethanol-water)
for
the
controlled
polymerization.16 In addition, the viscosity of the reaction mixtures increases
rapidly during the polymerization from water to gel-like solid within 15
95
Rheological properties of branched polyacrylamides
minutes. Therefore, from that point on, mass transfer limitations might play
a role in the low conversion of acrylamide.
Scheme 4.2: Synthesis of the comb PAMs
The dispersities of entries 1, 4, 7 and 10 (PDI < 2.5) could be directly
determined by GPC. The molecular weights of the rest of the entries fall
outside the measurable range of the GPC and are therefore difficult to
measure. However, as will be evident later (Figure 4.9 and 4.10), the
dispersities of the higher molecular PAMs are also relatively low. Indeed, the
slopes of G’ and G” in the terminal zone (on a double logarithmic scale) are 2
and 1 respectively, which is in line with other narrow-distributed polymers.23
Table 4.2: Characteristics of the (co)polymers
M/s1/s2a (w:v:v); T;
Time (min)
Architecture
Entry
[M]0:[I]0:[CuCl]0:
[Me6TREN]0
Conv.
(%)
Mn,overall
Linear
1
2
3
14 399:1:1.5:1.5
50 942:1:1.5:1.5
57 654:1:1.5:1.5
1:6 :0
1:6 :0
1:6 :0
;25 °C;15
;25 °C;25
;25 °C;60
50.8
42.1
62.0
519 928
1 524 432
2 540 789
519 928
1 524 423
2 540 789
4-arm
4
5
14 894:1:3.0:3.0
37 707:1:1.5:1.5
1:4 :1/5 ;25 °C;60
1:4 :1/10;25 °C;60
69.9
65.4
743 019
1 613 401
374 307
809 498
8-arm
6
7
10 037:1:3.0:3.0
49 822:1:3.0:3.0
1:4 :1/5 ;25 °C;60
1:4 :1/10;25 °C;60
88.1
48.8
631 116
1 730 784
160 576
435 493
13-arm
8
9b
10
11
12
12 019:1:1.5:1.5
47 610:1:1.5:1.5
100 050:1:3.0:3.0
149 634:1:3.0:3.0
150 084:1:2.0:2.0
1:6 :1/3 ;25 °C;60
1:6 :1/5 ;25 °C;60
1:4 :1/20;25 °C;60
4:15:1/20;25 °C;60
1:4 :1/40;25 °C;60
68.8
35.7
23.8
23.6
32.8
1
1
2
3
13
149 859:1:3.0:3.0
2:15:1/20;25 °C;60
14.8
14
150 174:1:1.5:1.5
1:2 :1/20;25 °C;60
23.8
a. M/s1/s2 = Monomer / solvent 1 / solvent 2 = Acrylamide / water / acetone
b. 12-arm
17-arm
96
587 766
208 130
692 550
510 092
499 094
1 576 493
2 540 500
Mn,SPAN
100
204
263
388
541
758
152
189
965
119
188 267
301 679
Chapter 4
The experimental conditions can be designed in such a way that branched
PAM with similar molecular weights but different number of arms (i.e. shorter
armlength) and relatively low dispersities can be prepared using ATRP in
water (and water-acetone mixtures). This allows the investigation of the
effect of the number of arms on the rheological properties of these polymers
in water solutions.
Cryo-TEM, semi-dilute solutions. Aqueous solutions of the branched
polymers were investigated using cryo-TEM. A typical cryo-TEM picture (of
the branched PAMs) is displayed in Figure 4.3.
Figure 4.3: Cryo-TEM image (scale bar is 100 nm) of branched PAM, entry 10 ([p] =
0.5 wt.%)
The darker spheres with a diameter of on average approximately 5 nm are
assumed to be the polyketone backbone of the branched PAMs (not present
in the cryo-TEM picture of the linear analogue, not shown for brevity). The
average area (in nm2) available for each arm can be computed by dividing
the surface area of the central backbone (based on a sphere with a diameter
of 5 nm) with the number of arms. The surface area available per arm
significantly decreases as the number of arms increases. The decrease in
surface area will lead to an increase in the steric hindrance for the polymeric
arms close to the backbone.
With this, it can be envisaged that if a high number of arms are present
the polymeric arms will be more extended (especially close to the backbone)
compared to a polymer with a lower number of arms. More evidence to
support this hypothesis is provided by the higher values of intrinsic viscosity
97
Rheological properties of branched polyacrylamides
for the polymers with a high (N ≥ 13) number of arms (as will be evident in
the following section). A similar behavior has been observed for PAM grafted
dextran.24
Intrinsic viscosity, effect of the number of arms (at equal overall
molecular weight). The intrinsic viscosity can be used to investigate the
dilute solution properties of the architecturally different polymers. The
intrinsic viscosity of entries 2, 5, 7, 10, and 13 were determined using
Martin’s25 equation:
(4.3)
where
is the specific viscosity,
concentration-plot,
is the slope of the viscosity-
is the polymer concentration and
is the intrinsic
viscosity. The intrinsic viscosity is obtained by extrapolating the plot of the
specific viscosity over concentration as a function of the concentration to
=
0 (Figure 4.4).
As can be observed in Figure 4.4, the intrinsic viscosity is a function of
the degree of branching. The intrinsic viscosities of the linear, 4- and 8-arm
are the same within the experimental error. Remarkably, the intrinsic
viscosities of the 13- and 17-arm PAMs are significantly higher than the
values found for the 8- and 4-arm PAMs. This is strong evidence that the
highly branched PAMs (N ≥ 13) are more extended in solution compared to
the PAMs with a low degree of branching (N ≤ 8).
4
17-arm,
13-arm,
8-arm,
4-arm,
linear,
2
R = 0,995
2
R = 0,996
2
R = 0,991
2
R = 0,994
2
R = 0,999
log(red)
3
2
1
0
0,0
A
0,4
0,8
1,2
1,6
2,0
Concentration (g/dl)
Figure 4.4: (A) Reduced viscosity as a function of the concentration using Martin’s
equation for entries 2, 5, 7, 10, and 13
98
Chapter 4
Intrinsic viscosity ([], dl/g)
30
25
17-arm,
13-arm,
8-arm,
4-arm,
linear,
[] = 28,00 dl/g
[] = 24,08 dl/g
[] = 7,96 dl/g
[] = 6,63 dl/g
[] = 6,14 dl/g
20
B
15
10
5
0
Linear
4-arm
8-arm
13-arm
17-arm
Molecular architecture
Figure 4.4, continued: (B) Intrinsic viscosity for entries 2, 5, 7, 10, and 13
Solution viscosity, effect of the number of arms (at equal polymer
volume
fraction
and
overall
molecular
weight).
The
rheological
comparison between the branched PAM polymers is conveniently carried out
at equal polymer volume fraction (s = c/c*), with c being the polymer
concentration and c* the critical overlap concentration. This can be defined
as26-28:
(4.4)
with
= molecular weight,
gyration,
= hydrodynamic radius,
= radius of
= Avogrado constant.
The radius of gyration (
) of the comb polymers is estimated (Table 4.3)
using the model developed by Daoud and Cotton29 for star shaped polymers
(equation 4.5).
(4.5)
with
= number of monomer units,
parameter,
= number of arms, and
= monomer excluded volume
= length of each monomeric unit.
For the linear polymer, used in the comparisons, the
is found in
literature30, 31 for a similar size (molecular weight) PAM.
99
Rheological properties of branched polyacrylamides
Table 4.3: Properties of the different (co)polymers
Architecture
Entry
Rg, est (nm)
Rh, DLS (nm)
88a
72
57
61
40
48
36
51
60
71
Linear
2
4-arm
5
8-arm
8
13-arm
10
17-arm
14
a. Taken from literature30
In
order
to
carry
out
the
measurements
(wt.%)
5·
0.09
0.17
0.36
0.44
0.96
well
(wt.%)
0.45
0.85
1.80
2.20
4.80
above
the
overlap
concentration, and maintain an equal excluded volume, the comparison of
the architecturally different polymers is carried out at 5 times the
. Higher
values are not tested given the difficulty in measuring the viscosity
accurately for the highly branched PAMs (gelation).
As mentioned before, the comparison between the architecturally
different polymers are performed at equal polymer volume fraction in order
to investigate the effect the branching has on the solution properties of PAM.
The zero shear rate viscosity (0) is determined by oscillation experiments
using equation 4.623:
(4.6)
with G” = loss modulus,
= frequency. The
is plotted against the
molecular architecture in Figure 4.5, at a polymer concentration of 5· . As
can be observed, the results suggest that the number of arms does affect
in that a higher
is obtained with more arms.
The viscoelasticity of the architecturally different polymers at equal
excluded volume (5· ) was evaluated through oscillation measurements
(Figure 4.6). For low number of arms (4 & 8), a lower elastic response,
compared to a linear PAM, is observed. The elastic response is higher
(compared to a linear analogue) when the number of arms is high, i.e. 13 &
17. According to literature for polymer melts32,
33
,
is exponentially
dependent on the molecular weight of the arms and the effect of the number
of arms becomes saturated above 4 arms33. It is also predicted by a model
for comb shaped polymers in the melt34 that the highest
are obtained with
comb polymer having low number of long arms. However, the results in
aqueous solution (Figure 4.5 & 4.6) are not in agreement with these
predictions. The discrepancy might lie in the difference in concentration
regime (melt vs. semi-dilute), and the fact that associations (Chapter 3) can
100
Chapter 4
arise in the aqueous solution due to the hydrophobic backbone. In addition,
unlike in the melt, in water solution hydrogen bonding (between the solvent
and polymer) might play a significant role in rheological properties.
5
10
Entries = 2, 5, 7, 10 & 13
4
10
3
10
2
0 (Pa.s)
10
1
10
0
10
-1
10
-2
10
-3
10
linear
4-arm
8-arm
13-arm
17-arm
Molecular architecture
Figure 4.5: 0 as a function of the molecular architecture at Mn,overall ≈ 1.6 MDa and a
polymer concentration of 5·
3
10
90
2
10
80
1
10
70
G' G" (Pa)
-1
10
A
-2
10
-3
10
-4
10
Filled symbols = G'
Empty symbols = G"
-5
10
-6
= 17-arm
= 13-arm
= 8-arm
= 4-arm
= Linear
Phase angle
0
10
60
50
B
40
30
20
10
17-arm
13-arm
8-arm
4-arm
Linear
0
10
0,1
1
10
100
0,1
1
Frequency (rad/s)
10
100
Frequency (rad/s)
Figure 4.6: The G’ and G” (A) and the phase angle (B) as a function of the frequency
for the different polymers at a polymer concentration of 5·c*
Solution
viscosity,
effect
of
the
number
of
arms
(at
equal
concentration and overall molecular weight). The effect of the number
of arms on the solution viscosity has been evaluated. The solution viscosity
(at
= 10 s-1) as a function of concentration has been measured, while
maintaining the overall molecular weight constant (Figure 4.7).
101
Rheological properties of branched polyacrylamides
50
40
Viscosity (Pa.s)
A
Linear
4-arm
8-arm
12-arm
30
20
10
0
0
1
2
3
4
5
6
7
8
9
Concentration (wt.%)
14
14
Linear
4-arm
8-arm
13-arm
17-arm
8
6
4
2
C
Linear
13-arm
17-arm
12
Viscosity (Pa.s)
10
Viscosity (Pa.s)
16
B
12
10
8
6
4
2
0
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0,0
4.0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
Concentration (wt.%)
Concentration (wt.%)
Figure 4.7: Viscosity (measure at
= 10 s-1) as a function of concentration for
different overall molecular weights (A, 0.6 MDa, B, 1.6 MDa and C, 2.6 MDa)
As can be observed, the solution viscosity of the 13- and 17-arm branched
PAM is systematically the highest at all molecular weights. The PAM polymers
can also be compared as a function of shear rate. In Figure 4.8A, such a
comparison is made where the concentration of the polymer is kept constant
but the number of arms is varied.
The viscosity function in Figure 4.8A is modeled using the the “CarreauYasuda” model21,
22
in order to evaluate the relaxation time ( ). In the melt,
the molecular weight35,
36
, dispersity index35,
(in solution) are parameters that influence
36
, and polymer concentration22
. As can be observed in Figure
4.8B, the molecular architecture has a pronounced effect on the relaxation
time ( ) in that a higher number of branches leads to a higher .
An increase in the relaxation time also affects the extent of shear
thinning behavior. As can be observed in Figure 4.8A, a solution containing
PAM with 13 or 17 arms display the most pronounced shear thinning
behavior.
102
Viscosity (Pa.s)
Chapter 4
10
2
10
1
10
0
10
A
2
17-arm,
R (model) = 0,9996
13-arm,
R (model) = 0,9995
linear,
R (model) = 0,9998
8-arm,
R (model) = 0,9998
4-arm,
R (model) = 0,9998
2
2
2
2
-1
10
-2
10
-1
10
0
10
1
10
2
10
3
-1
Shear rate (, s )
Relaxation time (, s)
10
1
17-arm,  10,89 s.
13-arm,  4,61 s.
linear,  0,70 s.
8-arm,  0,50 s.
4-arm,  0,47 s.
10
B
0
4-arm
8-arm
Linear
13-arm
17-arm
Molecular architecture
Figure 4.8: A; Viscosity function for PAMs (entries 2, 5, 7, 10 and 13; polymer
concentration of 3.85 wt.%), lines correspond to fits of the “Carreau-Yasuda” model
and B; the relaxation time for entries 2, 5, 7, 10 and 13
The onset of non-Newtonian behavior (in this case shear thinning) is also
affected by branching. As can be observed in Figure 4.8A, the critical shear
rate for the onset of shear thinning is lower for the 17 and 13-arm PAMs
compared to their linear analogue. This is confirmed by the value of
shown
for
brevity)
and
in
line
polyisoprenes37 and polybutadienes38,
with
39
earlier
studies
on
(not
branched
, which concluded that the critical
shear rate for the onset of non-Newtonian behavior is reduced upon
branching.
103
Rheological properties of branched polyacrylamides
The fact that the 13- and the 17-arm PAM display the highest solution
viscosity deviates from experimental observations on polyisoprenes in the
melt, where the highest viscosity is obtained for polymers with low number of
arms.33 Theoretical models, for combs34 and stars32 polymers, also predict
the highest viscosity for polymers with few arms in the melt. In the
entangled regime, the reptation of a star-chain is hindered by the arms. The
built-up stress relaxes through arm retraction, which is a much slower
process compared to linear-chain reptations.32,
34
For star polyisoprenes, the
effect of the number of arms (above N > 4) saturates and the molecular
weight of the arms determines the viscosity.33 However, recently it has been
demonstrated that comb like polyethylenes have 0 much higher than their
linear
and
long
chain
branched
analogues.
40
Nevertheless,
these
measurements are performed in the melt and thus the highest possible
“concentration” is measured. In semi-dilute solutions the number of arms
does have an effect on the solution viscosity, in that the increase in the
number of arms means an increase in the segment density and thus higher
viscosity (provided that the comparison is made above the entanglement
critical concentration). However, if a higher solution viscosity is required, the
increase in segment density has to overcome the negative effect that the
reduction in hydrodynamic volume (due to branching27) has on the solution
viscosity.
Viscoelasticity,
effect
of
the
number
of
arms
(at
equal
concentration and overall molecular weight). The effect of the number
of arms on the viscoelasticity of a water solution was probed by oscillation
experiments. The results are displayed in Figure 4.9, where the polymer
concentrations of the solutions were kept constant for each comparison.
Viscoelastic fluids display at low frequencies (i.e. in the terminal zone) a G”
that is directly proportional to the frequency ( ) with a slope of 1 and G’
proportional to
(a slope of 2).23 As can be observed in the Figure 4.9, all
samples display this behavior at low frequencies. The comparison at equal
polymer concentration demonstrates that the 13- and 17-arm PAM display a
more pronounced elastic response (lower phase angle) irrespective of the
molecular weight. However, the results can be masked by the difference in
viscosity; therefore the comparison is also made at equal
concentration).
104
(at different
Chapter 4
90
3
10
A1
80
1
2
10
70
1
G'G" (Pa)
0
10
-1
10
-2
10
-3
10
-4
Filled symbols = G'
Empty symbols = G"
10
-5
10
=
=
=
=
Linear
4-arm
8-arm
12-arm
Phase angle
10
50
40
30
Linear
4-arm
8-arm
12-arm
20
10
A2
0
2
10
90
B1
80
70
1
0
10
-1
10
-2
10
Filled symbols = G'
Empty symbols = G"
-3
= Linear
= 4-arm
= 8-arm
= 13-arm
= 17-arm
Phase angle
10
G'G" (Pa)
60
60
50
40
Linear
4-arm
8-arm
13-arm
17-arm
30
20
10
10
0
2
10
B2
90
C1
80
70
10
Phase angle
G'G" (Pa)
1
0
10
Filled symbols = G'
Empty symbols = G"
-1
10
0,1
1
10
= Linear
= 13-arm
= 17-arm
100
Frequency (rad/s)
60
50
40
30
20
10
Linear
13-arm
17-arm
C2
0
0,1
1
10
100
Frequency (rad/s)
Figure 4.9: A1, G’ and G” of the PAMs with Mtot = 0.6 MDa and A2 their respective
phase angles (polymer concentration = 5.66 wt.%). B1, G’ and G” of the PAMs with
Mtot = 1.6 MDa and B2 their respective phase angles (polymer concentration = 2.91
wt.%). C1, G’ and G” of the PAMs with Mtot = 2.6 MDa and C2 their respective phase
angles (polymer concentration = 1.96 wt.%)
Viscoelasticity, effect of the number of arms (at equal
and overall
molecular weight). The results of the comparison between the different
105
Rheological properties of branched polyacrylamides
PAMs at equal
are displayed in Figure 4.10. The comparison at equal
reveals that the 13- and 17-arm PAMs display lower phase angles at low
frequencies irrespective of the molecular weight.
1
10
90
A1
80
0
10
70
Phase angle
-1
G'G" (Pa)
10
-2
10
-3
10
-4
10
Filled symbols = G'
Empty symbols = G"
-5
10
= Linear
= 4-arm
= 8-arm
= 12-arm
60
50
40
30
Linear
4-arm
8-arm
12-arm
20
10
A2
0
90
B1
80
70
2
1
10
0
Filled symbols = G'
Empty symbols = G"
10
= Linear
= 4-arm
= 8-arm
= 13-arm
= 17-arm
Phase angle
G'G" (Pa)
10
60
50
40
30
20
10
0
Linear
4-arm
8-arm
13-arm
17-arm
B2
90
C1
80
2
10
Phase angle
G'G" (Pa)
70
1
10
0
10
Filled symbols = G'
Empty symbols = G"
= Linear
= 13-arm
= 17-arm
60
50
C2
40
30
Linear
13-arm
17-arm
20
10
-1
10
0,1
1
10
100
Frequency (rad/s)
0
0,1
1
10
Frequency (rad/s)
Figure 4.10: A1, G’ and G” of the PAMs with Mtot = 0.6 MDa and A2 their respective
phase angles (equal ). B1, G’ and G” of the PAMs with Mtot = 1.6 MDa and B2 their
respective phase angles (equal
). C1, G’ and G” of the PAMs with Mtot = 2.6 MDa and
C2 their respective phase angles (equal
106
)
100
Chapter 4
As the frequency is increased (Mn,overall = 0.6 MDa) to above 10 rad/s, the
phase angles of the 4- and 8-arm PAM decreases to lower values than that of
the linear and 12-arm. Given the different concentration required to reach
the same viscosity, the number of polymeric chains in the solution also
differs. For the 4- and 8-arm PAM a concentration of 3.85 and 4.76 wt.%
(respectively) is required. Compared to the linear and 12-arm PAM (polymer
concentration of 2.91 and 1.96 wt.% respectively), more polymeric chains
are present in the 4 and 8-arm solutions. In addition, the length of the arms
of the 4- and 8-arm PAMs are longer than that of the 12-arm. The
combination of longer arms (a higher arm molecular weight leads to a more
pronounced elastic behavior in the melt41) and higher number of polymeric
chains in solution (an increase in the concentration leads to a more
pronounced elastic behavior for polystyrene in chlorinated diphenyl 23,
42
)
might explain the more pronounced elastic behavior of the solutions
containing 4- and 8-arms. Another explanation might be that more arms
leads to more steric hindrance and therefore less hydrophobic associations
between the hydrophobic polyketone backbones. The 4- and 8-arms PAM
supposedly display more hydrophobic associations, given the less steric
hindrance, and more/stronger hydrophobic associations are known to lead to
a more pronounced elastic response.43-46 Nevertheless, further studies
(currently being carried out) are required to fully elucidate the mechanism
behind the observed behavior.
Viscoelasticity, effect of the length of the arms (at equal
concentration and equal number of arms). The effect of the length of the
arms on the viscoelasticity of a water solution was investigated by oscillation
experiments. The results for the 13-arm PAM are displayed in Figure 4.11. As
can be observed in Figure 4.11, the increase in length of the arms leads to an
increase in both the loss and storage modulus. The transition from the
terminal to the plateau zone is shifted to lower frequencies as the arm length
increases (i.e. also the Mn,overall). In addition the plateau zone becomes longer
as the arm length is increased. Both these effects are in line with results on
low dispersity polystyrene (in the melt).23 In the melt constraints, due to
entanglement, cause an increase in the terminal relaxation time and
increases with molecular weight.23 In the 13-arm PAM case, the constraints
arise due to its high molecular weight and architecture. Therefore the
terminal relaxation time increases with increasing arm length. One might
speculate that it should increase more rapidly compared to a linear polymer
(given the higher relaxation time in the melt for branched polymers 34). This
is in line with the higher solution viscosity of the 13-arm branched PAM
compared to its linear analogue. The phase angle decreases as the arm
107
Rheological properties of branched polyacrylamides
length increases. The disentanglement of the overlapping chains becomes
progressively more difficult as the length of the arms increase. Therefore, in
essence, a stiffer solution is obtained as the length of the arms increase.
The dependence of the
in solution on length of the arms is displayed
in Figure 4.11D. As can be observed, the
increases exponentially
(relatively good fit) with the increase in the length of the arms. This matches
the theory in the melt where the same exponential dependency of the
on
33
the arm molecular weight is observed .
2
2
10
1
10
10
1
G'(Pa)
G"(Pa)
10
0
10
: DParm = 3875
-1
10
0
10
: DParm = 3875
-1
10
: DParm = 2715
: DParm = 2715
: DParm = 1830
A
0,1
: DParm = 1830
: DParm = 1415
-2
10
B
: DParm = 690
1
10
100
1
2
Exponential fit, R = 0,94
7
10
6
: DParm = 2715
10
: DParm = 3785
5
10
0 (Pa.s)
Phase angle
10
: DParm = 1830
60
50
40
100
8
: DParm = 690
70
10
Frequency (rad/s)
: DParm = 1415
80
: DParm = 690
0,1
Frequency (rad/s)
90
: DParm = 1415
-2
10
30
4
10
3
10
2
10
20
D
1
10
C
10
0,1
0
10
1
10
Frequency (rad/s)
100
500
1000
1500
2000
2500
3000
3500
DParm
Figure 4.11: The loss (A) and storage (B) modulus, the phase angle (C) as a function
of the frequency of the 13-arm PAM with different length of the arms (polymer
concentration = 2.91 wt.%), and the 0 as a function of the DParm (D)
Schematic model. With the available data on linear and branched PAMs a
conceptual model can be devised (Figure 4.12) for the branched PAMs in
dilute and semi-dilute solutions. The hydrodynamic radius of the branched
PAMs depends on the number of arms. At low number of arms (N ≤ 8) the
108
4000
Chapter 4
hydrodynamic volume is slightly lower compared to that of a linear analogue.
This is in line with the general view of the more compactness of branched
polymers compared to their linear analogues27, which leads to lower
for
the branched polymers in dilute solutions.47 However, for a relatively high
number of arms (N ≥ 13), the low amount of space available for each arm
will lead to an extended configuration for the arms close to the backbone and
possibly in solution.
Figure 4.12: Schematic model of the branched PAMs
Increasing the concentration of the polymer to above the critical overlap
concentration leads to entanglements. When entangled at equal polymer
concentration, the branched PAMs with a higher number of arms (N ≥ 13)
109
Rheological properties of branched polyacrylamides
have a higher entanglement density compared to PAMs with few arms (N ≤
8). The increase in entanglement density leads to a higher solution viscosity.
In addition, above the critical overlap concentration, the rheology of a starlike (compared to a linear analogue) polymeric solution is governed by the
arm retraction, where the arms explore new configurations through retraction
and extension into new directions.32 As this is a much slower process32
compared to the reptation of linear chains48-51, an exponential dependence of
the 0 on the arm molecular weight is observed in the melt.32
For the 13- and 17-arm PAM, the combination of a higher hydrodynamic
volume (due to stretching) and a higher entanglement density leads to an
increase thickening efficiency compared to their linear analogue. In addition,
an increase in entanglement density, leads to a more pronounced shear
thinning behavior. This is in line with the results in Figure 4.8A.
4.4. Conclusion
The
controlled
synthesis
of
branched
high
molecular
weight
polyacrylamides (PAM) with equal overall molecular weight or with equal arm
lengths, through ATRP in water (and acetone as a co-solvent), has been
accomplished. Branched PAMs of 4, 8, 12, 13 and 17 arms have been
synthesized. The effect of the molecular architecture (i.e. number of arms)
on the rheological properties in semi-dilute water solutions (solution viscosity
and viscoelasticity) was investigated. The 13-arm and 17-arm PAM displayed
a higher solution viscosity compared to the linear, 4-arm, and 8-arm
analogues irrespective of the molecular weight. The comparison between the
13-arm PAM and a linear analogue displays an as much as 5-fold increase in
solution viscosity (at a shear rate of 10 s-1). Furthermore, a more
pronounced shear thinning is observed for the 13 and 17-arm PAMs. The
elastic response of the 13- and 17-arm PAM in solution is more pronounced
compared to their linear analogue. The 4- and 8-arm though, display a lower
elastic response compared to their linear analogues. The rheological
properties of the branched PAMs are dependent on the number of arms and
their length. In semi-dilute aqueous solutions, the combination of a higher
hydrodynamic volume and higher entanglement density leads to an improved
thickening efficiency (for N ≥ 13) of the branched PAMs. The manipulation of
the rheological properties of PAM in water through smart architectural design
opens new ways in designing PAM-based materials for new applications
where control in the rheological properties is crucial. The increased
thickening efficiency of the branched PAMs makes these water soluble
polymers highly attractive for applications in EOR.
110
Chapter 4
4.5. Acknowledgement
This work is part of the Research Programme of the Dutch Polymer
Institute DPI, Eindhoven, the Netherlands, projectnr. #716.
4.6. References
1. Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Prog. Polym. Sci. 2011, 1558.
2. Wang, D.; Cheng, J.; Yang, Q.; Gong, W.; Li, Q.; Chen, F. SPE 2000, SPE-63227MS.
3. Xia, H.; Ju, Y.; Kong, F.; Wu, J. SPE 2004, SPE-88456-MS.
4. Xia, H.; Wang, D.; Wang, G.; Wu, J. Petrol. Sci. Technol. 2008, 4, 398.
5. Zhang, L.; Yue, X. J. Cent. South Univ. T. 2008, 84.
6. Zhang, L.; Yue, X.; Guo, F. Pet. Sci. 2008, 1, 56.
7. Zhang, Z.; Li, J.; Zhou, J. Transport Porous Med. 2011, 1, 229.
8. Matyjaszewski, K.; Xia, J. H. Chem. Rev. 2001, 9, 2921.
9. Fanood, M. H. R.; George, M. H. Polymer 1988, 1, 128.
10. Fanood, M. H. R.; George, M. H. Polymer 1988, 1, 134.
11. Fanood, M. H. R. Iranian Polymer Journal 1998, 1, 59.
12. Gleason, E. H.; Miller, M. L.; Sheats, G. F. Journal of Polymer Science 1959, 133,
133.
13. Kulicke, W. M.; Horl, H. H. Colloid Polym. Sci. 1980, 7, 817.
14. Anthony, A. J.; King, P. H.; Randall, C. W. J Appl Polym Sci 1975, 1, 37.
15. Wang, W.; Wang, D.; Li, B.; Zhu, S. Macromolecules 2010, 9, 4062.
16. Appel, E. A.; del Barrio, J.; Loh, X. J.; Dyson, J.; Scherman, O. A. Journal of
Polymer Science Part A-Polymer Chemistry 2012, 1, 181.
17. Wever, D. A. Z.; Raffa, P.; Picchioni, F.; Broekhuis, A. A. Macromolecules 2012,
10, 4040.
18. Drent, E.; Keijsper, J. J. United States of America Patent US 5225523, 1993.
19. Mul, W.; Dirkzwager, H.; Broekhuis, A.; Heeres, H.; van der Linden, A.; Orpen, A.
Inorg. Chim. Acta 2002, 147.
20. Zhang, Y.; Broekhuis, A. A.; Stuart, M. C. A.; Picchioni, F. J Appl Polym Sci 2008,
1, 262.
21. Carreau, P. J. Transactions of the Society of Rheology 1972, 1, 99.
22. Yasuda, K.; Armstrong, R. C.; Cohen, R. E. Rheologica Acta 1981, 2, 163.
23. Ferry, J. D. Viscoelastic properties of polymers; John Wiley & Sons: New York,
1980; , pp 641.
24. Kutsevol, N.; Guenet, J. M.; Melnik, N.; Sarazin, D.; Rochas, C. Polymer 2006, 6,.
25. Abdel-Azim, A. A. A.; Atta, A. M.; Farahat, M. S.; Boutros, W. Y. Polymer 1998,
26, 6827.
26. Coviello, T.; Burchard, W.; Dentini, M.; Crescenzi, V. Macromolecules 1987, 5,
1102.
27. Burchard, W. Branched Polymers II 1999, 113.
28. Stokes, R. J.; Evans, D. F. Fundamentals of interfacial engineering; Wiley-VCH:
New York, 1997; .
29. Daoud, M.; Cotton, J. P. Journal De Physique 1982, 3, 531.
30. Camail, M.; Margaillan, A.; Martin, I. Polym. Int. 2009, 2, 149.
31. Seery, T. A. P.; Yassini, M.; Hogenesch, T. E.; Amis, E. J. Macromolecules 1992,
18, 4784.
32. Milner, S. T.; McLeish, T. C. B. Macromolecules 1997, 7, 2159.
33. Fetters, L. J.; Kiss, A. D.; Pearson, D. S.; Quack, G. F.; Vitus, F. J. Macromolecules
1993, 4, 647.
34. Inkson, N. J.; Graham, R. S.; McLeish, T. C. B.; Groves, D. J.; Fernyhough, C. M.
Macromolecules 2006, 12, 4217.
111
Rheological properties of branched polyacrylamides
35. Kazatchkov, I. B.; Bohnet, N.; Goyal, S. K.; Hatzikiriakos, S. G. Polym. Eng. Sci.
1999, 4, 804.
36. Stadler, F. J.; Piel, C.; Kaschta, J.; Rulhoff, S.; Kaminsky, W.; Muenstedt, H.
Rheologica Acta 2006, 5, 755.
37. Graessley, W. W.; Masuda, T.; Roovers, J. E. L.; Hadjichristidis, N. Macromolecules
1976, 1, 127.
38. Kraus, G.; Gruver, J. T. Journal of Polymer Science Part A-General Papers 1965,
1PA, 105.
39. Kraus, G.; Gruver, J. T. J Appl Polym Sci 1965, 2, 739.
40. Stadler, F. J.; Arikan-Conley, B.; Kaschta, J.; Kaminsky, W.; Muenstedt, H.
Macromolecules 2011, 12, 5053.
41. Watanabe, H. Progress in Polymer Science 1999, 9, 1253.
42. Holmes, L. A.; Kusamizu, S.; Osaki, K.; Ferry, J. D. Journal of Polymer Science Part
A-2-Polymer Physics 1971, 11, 2009.
43. Kujawa, P.; Audibert-Hayet, A.; Selb, J.; Candau, F. Journal of Polymer Science
Part B-Polymer Physics 2004, 9, 1640.
44. Leibler, L.; Rubinstein, M.; Colby, R. H. Macromolecules 1991, 16, 4701.
45. Volpert, E.; Selb, J.; Candau, F. Polymer 1998, 5, 1025.
46. Regalado, E. J.; Selb, J.; Candau, F. Macromolecules 1999, 25, 8580.
47. Ham, J. S. J. Chem. Phys. 1957, 3, 625.
48. Doi, M.; Edwards, S. F. Journal of the Chemical Society-Faraday Transactions Ii
1978, 1789.
49. Doi, M.; Edwards, S. F. Journal of the Chemical Society-Faraday Transactions Ii
1978, 1802.
50. Doi, M.; Edwards, S. F. Journal of the Chemical Society-Faraday Transactions Ii
1978, 1818.
51. Doi, M.; Edwards, S. F. Journal of the Chemical Society-Faraday Transactions Ii
1979, 38.
112
Chapter 5
Chapter 5
Acrylamide-b-N-isopropylacrylamide
block copolymers: Synthesis by atomic
transfer radical polymerization and
effect of hydrophilic-hydrophobic ratio
on solution properties
Abstract
A series of block copolymers of acrylamide and N-isopropylacrylamide
(NIPAM) characterized by different ratios between the length of the two
blocks have been prepared through atomic transfer radical polymerization in
water at room temperature. The solution properties of the block copolymers
were
correlated
to
their
chemical
structure.
The
effect
of
the
hydrophilic/hydrophobic balance on the critical micelle concentration (CMC)
was investigated. The CMC increases at higher values for the solubility
parameter, thus indicating a clear relationship between these two variables.
In addition, the aqueous solution rheology of the block copolymers was
studied to identify the effect of the chemical structure on the thermoresponsiveness of the solutions. An increase in the length of the NIPAM block
leads to a more pronounced increase in the solution viscosity. This is
discussed in the general frame of hydrophobic interactions strength. The
prepared polymers are in principle suitable for applications in many fields,
particularly enhanced oil recovery (EOR).
Based on: D.A.Z. Wever, G. Ramalho, F. Picchioni, A.A. Broekhuis.
Acrylamide-b-N-isopropylacrylamide block copolymers: Synthesis by atomic
transfer radical polymerization in water and the effect of the hydrophilichydrophobic ratio on the solution properties. Journal of Applied Polymer
Science, 2013, DOI: 10.1002/app.39785.
113
PAM-b-PNIPAM block copolymes, synthesis & properties
5.1. Introduction
Acrylamide
based
polymers
have
been
extensively
studied
and
implemented in many different application fields such as, waste water
treatment,
cosmetics
and
oil
recovery.1,
2
Poly[N-isopropylacrylamide]
(PNIPAM) and copolymers containing NIPAM have been extensively studied.3
The unique property of PNIPAM in water, i.e. a transition from hydrophilic to
partially hydrophobic character4 with increasing temperature, can be utilized
to prepare “smart” (responsive to external stimuli, in this case temperature)
polymeric materials. Possible applications include among others, controlled
drug delivery5, 6 and gene therapy7-9.
Controlled polymerization of NIPAM has been accomplished in water 10,
different alcohols11, and different mixtures of organic solvents and water12-14.
Homopolymers of NIPAM will aggregate and form globules, which precipitate
completely out of an aqueous solution if the temperature is increased above
the lower critical solution temperature (LCST).15 This can be a desired
property in an application such as drug delivery. However, as temperature
sensitive rheological modifiers, this is generally an undesired property since it
leads to precipitation from the solution with consequent loss of any
thickening effect. To mitigate this problem, a more hydrophilic monomer can
be copolymerized with NIPAM.3 At temperatures higher than the LCST of the
NIPAM, the latter will induce association of copolymers chains while the
hydrophilic segment of the copolymer will prevent (if it is long enough) the
copolymer from precipitating out of the solution. According to this effect (i.e.
the hydrophilic/hydrophobic balance), the incorporation of acrylamide, as the
hydrophilic moiety, leads to an increase of the LCST, depending on the
amount of acrylamide up to 100 °C.16,
17
However, up to date the
copolymerization of NIPAM with acrylamide has been reported through the
use of free radical polymerization16,
17
or coupling, i.e. grafting onto or
grafting through, of separately prepared polyacrylamide and PNIPAM. 18,
19
Both synthetic pathways allow little, if any, control over the macromolecular
structure and architecture, thus hindering the study of any reliable structureproperty relationships.
In addition, given the hydrophobic character of NIPAM, when the
polymer is dissolved in water a reduction of the surface tension is observed.20
On the other hand, the incorporation of acrylamide, a more hydrophilic
moiety, in the polymer dampens this effect.16 The higher the fraction of
acrylamide in the copolymer, the higher the surface tension of the
corresponding water solution is (closer to the value measured when only pure
PAM is used).16 The combination of these properties (i.e. surface activity and
rheology) renders these polymers very attractive at both academic and
114
Chapter 5
industrial level. However, as anticipated (vide supra), these copolymers are
usually
synthesized
by
free
radical
polymerization
and
thus
random
copolymers, rather than block for which these effects are expected to be
more relevant. In addition, the uncontrolled nature of the polymerization
leads to a broad range of molecular weights and dispersities. These factors
might hinder a deeper understanding of the relationship between the polymer
structure and its solution properties. As a consequence and in order to widen
the range of possible applications, it is crucial that the synthesis of the
copolymers is controlled and that new synthetic strategies are developed for
the synthesis of block-like structures.
The controlled polymerization of acrylamide has been published recently,
both in an alcohol-water mixture21 and, as reported recently by our group, in
water22. In addition the synthesis of the block copolymer poly(acrylamide-bN-isopropylacrylamide) in water was also accomplished.22
In this paper, the controlled synthesis of the block copolymers PAM-bPNIPAM with varying length of the blocks is reported. First the PAM
macroinitiators are prepared and subsequently NIPAM is polymerized on the
macroinitiator
as
polymerization).
blocks
To
the
(demonstrating
best
of
our
the
living
knowledge,
character
this
has
of
not
the
been
accomplished before. The solution properties, i.e. CMC and solution viscosity
as a function of shear rate and temperature, have been measured.
Correlations between the chemical structure and the solution properties are
provided. The
solution properties are
dependent
on the hydrophilic-
hydrophobic ratio of the copolymers. In addition, the surface properties of
the block copolymers depend in a linear fashion on the solubility parameter.
5.2. Experimental section
Chemicals. Chemicals. Acrylamide (AM, electrophoresis grade, ≥99%),
N-isopropylacrylamide
(NIPAM,
97%),
tris[2-(dimethylamino)ethyl]amine
(Me6TREN) copper(I) bromide (CuBr, 98%), copper(I) chloride (CuCl, 98%),
glacial acetic acid, ethanol, diethyl ether and methyl 2-chloropropionate
(MeClPr, 97%) were purchased from Sigma Aldrich. CuBr and CuCl were
purified by stirring in glacial acetic acid for at least 5 hours, filtering, and
washing with glacial acetic acid, ethanol and diethyl ether (in that order) and
then dried at reduced pressure.23 All the other chemicals were reagent grade
and used without further purification.
PAM macroinitiator. The synthesis of the PAM macroinitiator was
performed according to the literature method.22 Detailed reaction conditions
are summarized in Table 1. The volume of water used was kept constant at
115
PAM-b-PNIPAM block copolymes, synthesis & properties
1:6 (w:v) monomer to water ratio. The amount of catalyst used was 1:1.5
(mol:mol) initiator to CuCl and the same applied also for the ligand ratio
(Me6TREN). The reaction temperature was set at 25 °C and the reaction time
was kept constant at one hour (except for the MI-530). The degree of
polymerization (DP) of the macroinitiators was calculated by using the
conversion (measured by GC) and the initial ratio between the monomer and
initiator. The codes for the macroinitiators are defined as PAMX with X
designating the number of AM units.
Block
copolymerization,
synthesis
of
PAM-b-PNIPAM.
The
macroinitiator PAM-Cl was synthesized according to the aforementioned
procedure. An example of a block copolymerization is reported in the
following. 0.5063 g (0.039 mmol) of the macroinitiator was added to a 100
mL round-bottomed flask along with NIPAM (2.1267 g, 18.8 mmol). Thirteen
mL of demineralized water were added and the system stirred until the
contents were dissolved. The mixture was degassed by three freeze–pump–
thaw cycles followed by the addition of 5.8 mg (0.058 mmol) CuCl. The flask
was placed in a thermostated oil bath at 25 °C. To start the reaction, 13.4
mg (0.058 mmol) Me6TREN was added. All operations were carried out under
nitrogen. After 60 minutes, the reaction was stopped by quenching with 87
mL of demineralized water (≈ 1/3 of the reaction volume or more if the
reaction mixture is viscous). The contents were then purified via dialysis
using membrane tubing Spectra/Por® Dialysis Membrane (molecular weight
cut off [MWCO] = 2,000 g/mol). The product was then dried in an oven at 65
°C until constant weight and then grounded. The codes for the blockcopolymers are defined as PAMX-b-PNIPAMY with X and Y designating the
number of AM and NIPAM units respectively.
The degree of polymerization of NIPAM and the conversion of NIPAM is
calculated using the following:
(5.1)
(5.2)
is the number of monomeric units in the PAM macroinitiator and is
obtained from Table 1.
(protons of the polymer-backbone and of the
methyl groups of the NIPAM units) and
(proton on the first carbon next to
the amide of the NIPAM unit) are the areas of the peaks defined in Figure 1.
corresponds to the number of monomeric units in the PNIPAM that is
116
Chapter 5
attached to the PAM macroinitiator.
corresponds to the experimental
initial monomer / initiator ratio.
Characterization. Acrylamide conversion was measured using Gas
Chromatography (GC). One hundred μL of the sample taken from the
acrylamide polymerization flask was dissolved in 17 mL of acetone (polymer
precipitated) and injected on a Hewlett Packard 5890 GC with an Elite-Wax
ETR column.
Proton Nuclear Magnetic Resonance (1H NMR) spectra were recorded on
a Varian Mercury Plus 400 MHz spectrometer using D 2O as the solvent. The
NIPAM conversion was calculated by determining the ratio of the peak areas
of AM units and the NIPAM units.
Surface tension was measured using the pendant drop method on a
LAUDA DROP VOLUME TENSIOMETER TVT 1. A glass micro syringe was
attached to a needle with a capillary radius of 1.055 mm. The temperature of
the water bath was set to 25 °C and the density difference between air and
water was set to 0.997 g/mL. Two sets of three measurements were taken
and then averaged.
Viscosity measurements were performed on a HAAKE MARS molecular
advanced rheometer. The software program used was the HAAKE Rheowin
Job manager. The amount of sample used for each measurement was 2 mL.
Solution viscosity was measured as a function of the shear rate (0.075 s -1 –
1750 s-1, T= 20 °C) and as a function of temperature (shear rate 1.0 s -1, T =
20 °C – 80 °C, 4 °C/min)
The cloud point of the different copolymers was determined by UV-Vis
analysis.
A
temperature
Jasco
V-630
controlled
UV-Vis
spectrophotometer
six-position
sample
holder
equipped
was
with
used.
a
The
transmittance of the polymer solutions ([p] = 2 wt.%) was recorded at 500
nm at a heating rate of 0.2 °C/min from 20 to 70 °C against a reference
sample containing demineralized water.
The
hydrodynamic
radius
was
measured
through
Dynamic
Light
Scattering (DLS). A Brookhaven ZetaPALS Zeta Potential Analyzer was used
with a 659 nm solid-state laser. DLS was performed in dilute aqueous
solution at 20 °C and a scattering angle of 90 °. In total 10 runs were
performed for each sample (at equal polymer concentration, 0.0005 wt.%,
i.e. below the CMC) and the mean and standard deviation are calculated for
size distribution by weight assuming a lognormal distribution using the MAS
OPTION software.
117
PAM-b-PNIPAM block copolymes, synthesis & properties
5.3. Results and discussion
5.3.1. Synthesis of the macroinitiators. The synthesis of the PAM
macroinitiators was performed according to Scheme 5.1A and Table 5.1 using
different molar ratios between the initiator and AM.
Scheme 5.1: A, synthesis of the PAM macroinitiators (MI) and B, synthesis of the
block copolymers PAM-b-PNIPAM
Table 5.1: Synthesis of the PAM macroinitiators
Entry
[M]0:[I]0
M/water (wt:vol); T; Time (min)a
Conv (%)
Mn,thb (g/mol)
DP
14 450
16 660
37 900
200
235
460
PAM200
300 : 1
1:6; 25 °C; 60
68
PAM235
300 : 1
1:6; 25 °C; 60
78
PAM460
680 : 1
1:6; 25 °C; 90
78
a: M = monomer, wt = weight, vol = volume in mL, T = temperature
b: Theoretical molecular weight = [M]0/[I]0 · conv.
As can be observed in Table 5.1, three different macroinitiators were
prepared with molecular weights varying between 14 000 to 38 000 g/mol.
The controlled nature of the polymerization has been reported already. 22
Further evidence for the living/controlled character of the polymerization is
provided by the ability to prepare block copolymers with NIPAM.
5.3.2. Synthesis of the block copolymers PAM-b-PNIPAM
The acrylamide macroinitiators synthesized in Table 5.1 were used as the
initiators in the copolymerisation with NIPAM. A summary of the experimental
conditions applied to synthesize the different copolymers is given in Table
118
Chapter 5
5.2. Besides the monomer to initiator ratio, in one reaction also the scale of
the preparation has been varied (important for further up-scaling).
Table 5.2: Synthesis of the different PAM-b-PNIPAM block copolymers
Entry
[M]0:[I]0
M/water (wt:vol);
T; Time (min)
Conv
(%)a
DP
NIPAM
Mn,1H-NMR
PAM200-b-PNIPAM30
55 : 1
1:6; 25 °C; 60
57
17 600
30
PAM200-b-PNIPAM70
140 : 1
1:6; 25 °C; 60
50
22 150
70
PAM200-b-PNIPAM70
275 : 1
1:6; 25 °C; 60
26c
22 150
70
PAM200-b-PNIPAM90
140 : 1
1:6; 25 °C; 60
66
24 400
90
PAM200-b-PNIPAM155
270 : 1
1:6; 25 °C; 60
57
31 750
155
PAM200-b-PNIPAM185
550 : 1
1:6; 25 °C; 60
34
35 150
185
PAM200-b-PNIPAM650 1115 : 1
1:6; 25 °C; 60
59
87 750
650
PAM235-b-PNIPAM125 2495 : 1
1:6; 25 °C; 160
5
30 850
125
PAM460-b-PNIPAM10
750 : 1
1:6; 25 °C; 60
1
33 800
10
a: The conversion was determined by 1H-NMR
b: Solubility parameter
c: The conversion is low, which might be due to the larger scale of the reaction
DP
PAM
b
(J1/2·cm-3/2)
200
200
200
200
200
200
200
235
460
27.8
26.5
26.5
25.3
24.2
23.3
22.7
26.7
29.0
The largest block copolymer prepared was PAM200-b-PNIPAM650 and the
smallest was PAM200-b-PNIPAM30. PAM460-b-PNIPAM80 was synthesized in
order to have roughly the same total molecular weight as PAM235-bPNIPAM125, even though it contains a different hydrophobic/hydrophilic
ratio. These two polymers are compared (see below) to investigate whether
the effects observed arise from an increase in molecular weight or from the
increase in NIPAM content (i.e. hydrophobic/hydrophilic ratio).
D 2O
1
2
PAM200-b-PNIPAM650
PAM200-b-PNIPAM185
2
2
2
PAM200-b-PNIPAM155
2
PAM200-b-PNIPAM90
2
1
PAM200-b-PNIPAM70
2
PAM200-b-PNIPAM30
PAM200
5
4
3
2
1
ppm
Figure 5.1: 1H-NMR spectra of the block copolymers PAM200-b-PNIPAM(Y) and the
parent macroinitiator
119
PAM-b-PNIPAM block copolymes, synthesis & properties
As mentioned before, the conversions provided in Table 5.2 were determined
using 1H-NMR (Figure 5.1). The conversion can be calculated by comparing
the ratio of the areas of resonances belonging to the protons of the first
carbon of the isopropyl moieties of the polymer (labelled 1) and the ones for
the rest of the protons labelled 2 (Figure 5.1). The 1H-NMR spectra of the
block copolymers (prepared with the macroinitiator PAM-200) are provided in
Figure 5.1.
The resonance labelled as 1 ( 3.9 ppm) represent the hydrogen atom of
the CH group of the isopropyl group of PNIPAM and therefore the intensity of
this resonance (in relation to the resonances labelled 2,  1.2 – 2.5 ppm)
corresponds
to
the
amount
of
PNIPAM
polymerized
on
the
PAM
macroinitiator. The total area of the resonances labelled 2 correspond to the
protons from the backbone of both the PAM and PNIPAM along with the 6
methyl protons of PNIPAM (2× CH3). This area represents a total of 12
protons (9 from PNIPAM and 3 from PAM). Increasing the [M] 0:[I]0 ratio
leads to a higher area of the resonance corresponding to the NIPAM blocks
indicating that longer NIPAM blocks are prepared (Figure 5.1). The 1H-NMR
spectra of the block copolymers agrees with the proposed structures.
5.3.3. Solution properties of poly(AM-b-NIPAM)
Solution viscosity as a function of shear. In Figure 5.2 the viscosity
of the polymer solution (4 wt.% in demineralized water) as a function of the
shear rate is displayed.
The polymers used are characterized by different hydrophilic (AM) /
hydrophobic (NIPAM) ratios. All polymers consisted of a hydrophilic block of
acrylamide (200 acrylamide units) and a hydrophobic block of PNIPAM of
different lengths (and thus different total molecular weight).
At low shear rates a Newtonian plateau is observed, irrelevant of the
length of the polymer or the number of NIPAM units. As the shear rate is
increased (> 100 s-1) shear thinning is observed (for PAM200-b-PNIPAM185
and PAM200-b-PNIPAM650), which is related to the disruption of the
entanglements.24 At higher shear rates (≥ 500 s-1) shear thickening is
visible for the block copolymers containing PNIPAM block below 100 units.
Given the low number of NIPAM units, the copolymer will behave more like
polyacrylamide. Polyacrylamides are known to display shear thickening
behaviour, related to structure formations (associations due to collision of
chains arise25) and chain stretching, above a critical shear rate.25, 26
Figure 5.2 also shows that larger total molecular weights or larger NIPAM
contents of the polymers result in higher starting viscosities of the solutions.
It is unclear however from the results if this is due to the increase in
120
Chapter 5
molecular weight or from the increase in the NIPAM content. As the NIPAM
blocks increase in length so does the solution viscosity. The bulky isopropyl
group of the NIPAM units inhibits the NIPAM blocks from coiling up as much
as the AM units. Therefore as the NIPAM blocks increase in length the
polymeric chain will be more extended. This leads to a higher hydrodynamic
volume and thus a higher solution viscosity.
Viscosity (mPa.s)
100
PAM200-b-PNIPAM650
PAM200-b-PNIPAM185
PAM200-b-PNIPAM155
PAM200-b-PNIPAM90
PAM200-b-PNIPAM70
10
PAM200-b-PNIPAM30
1
1
10
100
1000
-1
Shear rate (, s )
Figure 5.2: Viscosity vs shear rate of the PAMX-b-PNIPAMY series at a polymer
concentration of 4 wt.%
Four different polymers are compared (Figure 5.3) in order to elucidate which
parameter, molecular weight or NIPAM content, has a more pronounced
effect on the solution viscosity.
The PAM-PNIPAM ratio is different for three of the polymers used in the
comparison; however the molecular weights are similar. A polyacrylamide of
similar molecular weight (PAM460) is also included in the comparison. If the
viscosity was solely dependent on the total molecular weight, then the
solution viscosity of the four different solutions should be similar. However,
as can be observed, clear differences can be distinguished. Although the
Mn,tot of PAM460 is larger than that of PAM235-b-PNIPAM125, it displays a
lower solution viscosity. This confirms that the presence of NIPAM in the
polymer has a much greater effect on the viscosity than the molecular
weight. The comparison between PAM235-b-PNIPAM125 and PAM460-bPNIPAM10 further justifies this conclusion, given the lower amount of NIPAM
in the latter polymer. Further evidence for the increase in viscosity with
increase in the NIPAM content can be obtained from the intrinsic viscosity
121
PAM-b-PNIPAM block copolymes, synthesis & properties
([]).The intrinsic viscosity of the four different samples has been determined
by taking the limit (c  0) of the plots of the reduced viscosity as a function
of the concentration (Figure 5.4).
Viscosity (mPa.s)
100
PAM200-b-PNIPAM185
PAM235-b-PNIPAM125
PAM460-b-PNIPAM10
10
PAM460
1
10
100
1000
-1
Shear rate (, s )
Figure 5.3: Solution viscosity vs shear rate for block copolymers of similar Mn,tot but
different PAM-PNIPAM ratios (polymer concentration is 4 wt.%)
7
6
2
[] = 1.13 dl/g
2
[] = 0.93 dl/g
2
[] = 0.84 dl/g
2
[] = 0.71 dl/g
PAM200-b-PNIPAM185,
R = 0.98
PAM235-b-PNIPAM125,
R = 0.99
PAM530-b-PNIPAM10,
R = 0.98
PAM530,
R = 0.99
red (dl/g)
5
4
3
2
1
0
0,0
0,3
0,6
0,9
1,2
1,5
1,8
2,1
2,4
Concentration (g/dl)
Figure 5.4: Reduced viscosity vs concentration for block copolymers of similar M n,tot
but different PAM-PNIPAM ratios
122
Chapter 5
As evident in Figure 5.4, the [] increases with an increase in the NIPAM
content of the copolymers. With these results it can be concluded that the
differences observed in the solution properties of the four different samples
(with similar Mn,tot but different PAM / PNIPAM ratios) arise from the
differences in the chemical structure.
The solution viscosity is also dependent on the hydrodynamic volume of
the polymer chains in solution. DLS measurements demonstrate that the
hydrodynamic volume is dependent on the hydrophobic-hydrophylic ratio
(Table 5.3).
Table 5.3: Properties of the different block copolymers
Entry
PAM530
PAM530-b-PNIPAM10
Rh, DLS (nm)
c*equation 4 (wt.%)
5·c* (wt.%)
57
0.0100
0.0500
116
0.0010
0.0050
PAM235-b-PNIPAM125
99
0.0013
0.0065
PAM200-b-PNIPAM185
130
0.0006
0.0030
However, in order to evaluate what the effect is of the chemical structure on
the rheological properties the comparison of the solution viscosities is
performed at equal excluded volume (s).27 The concentration at which the
polymeric chains start to overlap is defined as
(equation 5.4) if the radius of gyration (
, and can be calculated
) or the hydrodynamic radius (
)
is known.28, 29
(5.3)
(5.4)
with
being the Avogadro constant,
polymer, and
is the molecular weight of the
is the density of the solution. The comparison between the
four different polymers is also done at a concentration of five times the
critical overlap concentration (5· ) in order to have the same excluded
volume, and the results are displayed in Figure 5.5.
The lower solution viscosity of the block copolymers at equal excluded
volume demonstrates the effectiveness of hydrogen bonding to increase the
solution viscosity. The solutions are well above the overlap concentration and
123
PAM-b-PNIPAM block copolymes, synthesis & properties
thus entanglements are present. The shear thinning behaviour observed
(Figure 5.5) is related to the disentanglements of the chains and disruption of
the weak hydrogen bonds. The hydrogen bonding capability of PAM is higher
compared to PNIPAM. However, the hydrodynamic volume of a polymer chain
increases (as evident from the
). Therefore the observed behaviour is a
balance between the reduction in hydrogen bonding interactions and the
increase in hydrodynamic volume. To conclude, the differences observed in
the solution viscosities (Figure 5.3 and 5.5) of the different polymers arise
due to the differences in chemical structure (PAM / PNIPAM ratio).
10
Viscosity (mPa.s)
PAM460
PAM235-b-PNIPAM125
PAM200-b-PNIPAM185
PAM460-b-PNIPAM10
1
1
10
100
1000
-1
Shear rate (, s )
Figure 5.5: Solution viscosity vs shear rate for block copolymers of similar Mn,tot but
different PAM-PNIPAM ratios at the same excluded volume (polymer concentration is
5· )
To the best of our knowledge this constitutes a novel insight into the effect of
different structural parameters (such as hydrophobic/hydrophilic balance and
molecular weight) on the corresponding solution viscosity. Indeed, to date, a
systematic study of the roles that molecular weight, hydrophobic group
content and distribution (within the copolymer sample) play in solution
properties has not yet been reported.30
Solution viscosity as a function of temperature. The viscosity was
measured as a function of the temperature of the solution and the results are
displayed in Figure 5.6. The polymer concentration of the solutions was set at
2 wt. %. All polymers consisted of a hydrophilic block of acrylamide (roughly
124
Chapter 5
14 000 g/mol or 200 acrylamide units) and a hydrophobic block of PNIPAM of
differing length, resulting in polymers with different total molecular weights.
The shear rate during the temperature sweep was fixed at a value of 1.0 s -1.
To illustrate the effect of NIPAM on the behaviour of the block-copolymers in
solution as a function of temperature, the homopolymer PAM460 is also
displayed in Figure 5.6. As can be observed in Figure 5.6, a clear peak in the
viscosity near 32 °C can be distinguished, except for the homopolymer
(PAM460).
800
PAM200-b-PNIPAM185
PAM200-b-PNIPAM70
PAM200-b-PNIPAM155
PAM200-b-PNIPAM90
PAM460
750
Viscosity (mPa.s)
700
300
250
200
150
100
50
0
0
20
25
30
35
40
45
50
55
60
Temperature (°C)
Figure 5.6: Solution viscosity of 4 wt. % polymers solutions vs temperature
The temperature at which an increase in viscosity is observed does not
change with the NIPAM content, and corresponds to the LCST of PNIPAM. As
the temperature increases from 20°C the viscosity slowly decreases before it
significantly increases to a peak near 32 °C. After the peak, the viscosity
decreases rapidly as the temperatures further increases, stabilizing near the
initial viscosities measured before the peak. The same behaviour in the
solution viscosity at temperatures below and near the LCST is also observed
for the homopolymer of N-isopropylacrylamide.31-33 When the temperature of
the polymer solution reaches the LCST, the isopropyl groups of the PNIPAM
blocks are dehydrated and aggregation between the PNIPAM blocks arises.33
The increase in viscosity in that region indicated that some of this association
is intermolecular leading to the observed increase in solution viscosity. The
decrease in viscosity above the LCST is a result of the majority of the chains
125
PAM-b-PNIPAM block copolymes, synthesis & properties
precipitating
into
macromolecular
aggregates31,
33
and
the
decreased
viscosity of the solvent. However the peaks displayed in Figure 5.5 signify a
response of the polymer to changes in temperature. When comparing
PNIPAM to anionic polyacrylamide (HPAM), which has a similar structure 32,
the HPAM follows the well-known trend of decreasing viscosity as a function
of temperature. Therefore the peak exhibited for the PAM-b-PNIPAM block
copolymers is attributed solely to the presence of NIPAM moieties.
Looking more closely to the peaks it is clear that decreasing the NIPAM
content (from 185 to 155 units) resulted in a decrease in the peak viscosity
from above 700 mPa.s to 275 mPa.s respectively. The peak viscosity reduces
further with smaller blocks PNIPAM. The smaller the PNIPAM blocks are,
weaker hydrophobic aggregations arise. In general, the significant increase in
the solution viscosity for hydrophobically associating polymers results from
the intermolecular aggregation between the hydrophobic groups. 1 The
aggregation results in larger hydrodynamic volumes, which in turn, increase
the
viscosity
of
the
solution.
By
increasing
the
shear
rate,
these
intermolecular associations are disrupted resulting in the decrease of the
hydrodynamic volume and therefore the solution viscosity.1
Critical
micelle
concentration
(CMC).
The
critical
micelle
concentrations were measured by plotting the surface tension (against air) of
a polymer at different concentrations (Figure 5.7).
Surface tension (mN/m)
70
65
PAM200-b-PNIPAM650
PAM200-b-PNIPAM185
PAM200-b-PNIPAM155
PAM200-b-PNIPAM90
PAM200-b-PNIPAM30
60
55
50
45
40
1E-7
1E-6
1E-5
1E-4
1E-3
Concentration (M)
Figure 5.7: Surface tension against the polymer concentration of 5 different
copolymers
126
Chapter 5
As can be observed in Figure 5.7, S-shaped curves are obtained, which
correspond to those expected.34 For low polymer concentrations the solutions
move towards the surface tension of demineralized water (measured to be
70.47 mN/m). As the concentration increases, the surface tension reaches a
region where it decreases dramatically. Then at a specific concentration, the
surface tension stops decreasing with a minimum value near 41.5 mN/m.
This specific concentration is known as the critical micelle concentration
(CMC). Remarkably, all the samples display a surface tension close to the
value of pure PNIPAM, albeit with different CMCs (as will become evident
later). This is in stark contrast to random copolymers of AM and NIPAM,
where the final surface tension is a function of the composition of the
copolymer.16 The surface tension for a 50-50 (mol ratio) random copolymer
is 51.0 mN/m, compared to 41.5 mN/m for PAM200-b-PNIPAM185.
Graphically the CMC can be obtained from the plot in Figure 5.6 by
taking the line of best fit in two places and noting the concentration at the
intersection35 (not shown for brevity). As the PNIPAM block length increases
the concentration needed for micelle formation decreases. This is expected as
the larger the PNIPAM blocks are, the larger the effect of its lower
hydrophilicity.16 The order of magnitudes 10-6 and 10-7 M coincide with that
given in literature for amphiphilic block copolymers.36 The formation of
micelles is a result of the concentration of polymer being high enough such
that interaction between the PNIPAM blocks is beneficial. The aggregation of
less hydrophilic blocks result in the formation of a micelle with a hydrophobic
core (PNIPAM) and a hydrophilic corona (PAM) keeping the micelles stable in
the water solution.36
To justify the correlation between the CMC and the PNIPAM content, the
solubility parameter () was plotted as a function of the CMC (Figure 5.8).
The solubility parameter was calculated using a group contribution theory37,
which takes into account the structure of the polymer and the molar % of
each block. This is similar to the hydrophilic-lipophilic balance (HLB), which
calculates the balance based on molecular weight percentage of each block.
For an acrylamide homopolymer the solubility parameter is 29.14 J 1/2·cm-3/2.
For a pure PNIPAM polymer the solubility parameter is 22.07 J 1/2·cm-3/2.
Therefore the copolymers should have decreasing solubility parameters as
the PNIPAM block increase in length.
As shown by Figure 5.8 the CMC increases linearly as a function of the
solubility
parameter.
This
confirms
the
general
trend
for
non-ionic
surfactants38 where the CMC increases as the hydrophilic content increases.
In literature, however the hydrophilic/hydrophobic balances are depicted by
the HLB number and not the solubility parameter.36 The solubility parameter
127
PAM-b-PNIPAM block copolymes, synthesis & properties
takes into account the structure of the each block and their molar ratios and
the HLB number only looks at the molecular mass ratio of each block. As a
result, comparing the solubility parameter with the CMC illustrates a
structure-property relationship for the CMC and surface activity. This enables
the design of block copolymers with predictable surface properties and
renders the laborious measurements obsolete.
CMC (M)
1E-5
1E-6
ln (y) = 0,5604·x - 27,63
2
R = 0.998
0 22
24
26
28
1/2
-3/2
Solubility parameter (J ·cm
30
)
Figure 5.8: The solubility parameters vs. the CMC
Effect of the chemical structure on the cloud point. The cloud point of
four different block copolymers was determined with UV-Vis (Figure 5.9). The
comparison
of
the
different
block
copolymers
demonstrates
that
by
decreasing the length of the NIPAM block an increase in the cloud point can
be obtained. Similar results were obtained for random copolymers of AM and
NIPAM.16 Random copolymers of AM and NIPAM of higher molecular weights
display cloud points that are dependent on the ratio between the two
moieties.39
A decrease in the NIPAM content from 85 to 55 mol% leads to an
increase in the cloud point from 42 to 74 °C.39 As can be observed in Figure
5.8, a decrease in the NIPAM content from 48 to 13 mol% leads to a slight
increase in the cloud point from 32 to 34 °C. Copolymers of NIPAM and AM
with 50 mol% of AM16 (or 40 mol%39) display a cloud point above 100 °C. Of
all the samples tested (AM content varies between 52 and 87 mol%), the
128
Chapter 5
cloud points were all below 35 °C. This significant difference (compared to
the literature) is attributed to the fact that the NIPAM units in the block
copolymers can form a globule more readily compared to that of a random
copolymer. Therefore, the block copolymers can precipitate out of the
solution much easier compared to random copolymers.
100
Absorbance (%)
80
60
PAM200-b-PNIPAM30
PAM200-b-PNIPAM90
PAM200-b-PNIPAM155
PAM200-b-PNIPAM185
40
20
0
0
20
25
30
35
40
45
50
55
60
65
70
75
Temperature (°C)
Figure 5.9: LCST determination by UV-Vis light transmittance ([p] = 2 wt.%)
5.4 Conclusion
Block copolymers of AM and NIPAM have been prepared by ATRP in
water at room temperature. The controlled nature of the polymerization
allowed for the synthesis of block copolymers with varying block lengths of
both monomers. The aqueous solution properties of the block copolymers
were correlated to their chemical structure. The effect of the hydrophobichydrophilic ratio on the LCST, CMC, and solution rheology was investigated. A
clear correlation exists between the solubility parameter and the CMC, the
latter decreasing with the former. The LCST of the block copolymers is
dependent on the balance between the two moieties. The longer the NIPAM
block length, the closer the LCST is to the one of the NIPAM homopolymer.
The solution viscosity is also dependent on the chemical structure. Longer
129
PAM-b-PNIPAM block copolymes, synthesis & properties
blocks of NIPAM lead to a higher solution viscosity, which is related to the
more extended nature of the NIPAM blocks (compared to AM ones).
The correlation between the solubility parameters and the surface
properties of the copolymers offers the possibility of predicting the surface
properties of block copolymers without the need to measure them. These
new insights, coupled with the novelty of the synthetic strategy pave the way
for application of these materials in e.g. EOR, drug delivery and cosmetics.
5.5 Acknowledgements
This work is part of the Research Program of the Dutch Polymer Institute
DPI, Eindhoven, The Netherlands, project #716.
5.6 References
1. Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Progress in Polymer Science 2011,
11, 1558.
2. Shalaby W. Shalaby; Charles L. McCormick; George B. Butler Water-Soluble
Polymers: Synthesis, Solution Properties, and Applications; American Chemical
Society: Washington DC, 1991; .
3. Liu, R.; Fraylich, M.; Saunders, B. R. Colloid Polym. Sci. 2009, 6, 627.
4. Pelton, R. J. Colloid Interface Sci. 2010, 2,.
5. Galaev, I. Y.; Mattiasson, B. Trends Biotechnol. 1999, 8,.
6. Dilgimen, A. S.; Mustafaeva, Z.; Demchenko, M.; Kaneko, T.; Osada, Y.; Mustafaev,
M. Biomaterials 2001, 17,.
7. Hinrichs, W. L. J.; Schuurmans-Nieuwenbroek, N. M. E.; van de Wetering, P.;
Hennink, W. E. J. Controlled Release 1999, 2-3,.
8. Bulmus, V.; Patir, S.; Tuncel, S. A.; Piskin, E. J. Controlled Release 2001, 3,.
9. Dincer, S.; Tuncel, A.; Piskin, E. Macromolecular Chemistry and Physics 2002, 1011,.
10. Millard, P.; Mougin, N. C.; Boker, A.; Muller, A. H. E. Controlling the Fast ATRP of
N-Isopropylacrylamide in Water. In Controlled/Living Radical Polymerization:
Progress in ATRP; Matyjaszewski, K., Ed.; American Chemical Society: 2009; Vol.
1023, pp 127.
11. Xia, Y.; Yin, X. C.; Burke, N. A. D.; Stover, H. D. H. Macromolecules 2005, 14,
5937.
12. Hu, H.; Du, J.; Meng, Q.; Li, Z.; Zhu, X. Chinese Journal of Polymer Science 2008,
2,.
13. Masci, G.; Giacomelli, L.; Crescenzi, V. Macromolecular Rapid Communications
2004, 4,.
14. Ye, J.; Narain, R. J Phys Chem B 2009, 3,.
15. Dimitrov, I.; Trzebicka, B.; Muller, A. H. E.; Dworak, A.; Tsvetanov, C. B. Progress
in Polymer Science 2007, 11, 1275.
16. Zhang, J.; Pelton, R. Journal of Polymer Science Part A-Polymer Chemistry 1999,
13,.
17. Chiklis, C.; Grasshof, J. Journal of Polymer Science Part A-2-Polymer Physics
1970, 9, 1617.
18. Petit, L.; Karakasyan, C.; Pantoustier, N.; Hourdet, D. Polymer 2007, 24,.
19. Portehault, D.; Petit, L.; Hourdet, D. Soft Matter 2010, 10,.
20. Zhang, J.; Pelton, R. Langmuir 1996, 10, 2611.
130
Chapter 5
21. Appel, E. A.; del Barrio, J.; Loh, X. J.; Dyson, J.; Scherman, O. A. Journal of
Polymer Science Part A-Polymer Chemistry 2012, 1,.
22. Wever, D. A. Z.; Raffa, P.; Picchioni, F.; Broekhuis, A. A. Macromolecules 2012,
10, 4040.
23. Neugebauer, D.; Matyjaszewski, K. Macromolecules 2003, 8, 2598.
24. Ferry, J. D. Viscoelastic properties of polymers; John Wiley & Sons: New York,
1980; , pp 641.
25. Dupuis, D.; Lewandowski, F. Y.; Steiert, P.; Wolff, C. J. Non-Newton. Fluid 1994,
11.
26. Hu, Y.; Wang, S.; Jamieson, A. Macromolecules 1995, 6, 1847.
27. Daoud, M.; Cotton, J. P. Journal De Physique 1982, 3, 531.
28. Burchard, W. Branched Polymers II 1999, 113.
29. Coviello, T.; Burchard, W.; Dentini, M.; Crescenzi, V. Macromolecules 1987, 5,
1102.
30. Biggs, S.; Selb, J.; Candau, F. Langmuir 1992, 3, 838.
31. Monteux, C.; Mangeret, R.; Laibe, G.; Freyssingeas, E.; Bergeron, V.; Fuller, G.
Macromolecules 2006, 9, 3408.
32. Tam, K.; Wu, X.; Pelton, R. Journal of Polymer Science Part A-Polymer Chemistry
1993, 4, 963.
33. Tam, K.; Wu, X.; Pelton, R. Polymer 1992, 2, 436.
34. Zhang, J.; Pelton, R. Colloids and Surfaces A-Physicochemical and Engineering
Aspects 1999, 1-3, 111.
35. Egan, R.; Jones, M.; Lehninger, A. J. Biol. Chem. 1976, 14, 4442.
36. Miao, Q.; Jin, Y.; Dong, Y.; Cao, Z.; Zhang, B. Polym. Int. 2010, 8, 1116.
37. van Krevelen, D. W.; te Nijenhuis, K. Propeties of Polymers. Their Correlation with
Chemical Structure; their Numerical Estimation and Prediction from Additive
Group Contributions; Elsevier: Amsterdam, the Netherlands, 2009; , pp 1030.
38. Barakat, Y.; Gendy, T.; Basily, I.; Mohamad, A. British Polymer Journal 1989, 6,
451.
39. Mumick, P. S.; Mccormick, C. L. Polym. Eng. Sci. 1994, 18, 1419.
131
PAM-b-PNIPAM block copolymes, synthesis & properties
This page intentionally left blank
132
Chapter 6
Chapter 6
Branched thermoresponsive polymeric
materials: Synthesis and effect of the
macromolecular structure on the
solution properties
Abstract
A series of comb-like block and random copolymers based on acrylamide
(AM) and N-isopropylacrylamide (NIPAM) have been prepared by atom
transfer radical polymerization (ATRP). The number of side-arms, the length
of the AM and NIPAM blocks as well as the distribution of the two monomers
(block or random) were systematically varied. The aqueous solution
properties, i.e. the solution viscosity as a function of shear rate and
temperature and the critical micelle concentration (CMC) of the different
copolymers were evaluated. Particular emphasis is dedicated to the thermoresponsiveness of the aqueous copolymers solutions as measured by
rheological behavior. The CMC is a function of the molar ratio between the
AM and NIPAM as well as the distribution (block or random). The surface
tension of the block copolymers is close to the value for pure poly(NIPAM),
while that of the random copolymers is a function of the composition. The
block copolymers tend to precipitate from the solution at temperatures above
the Lower Critical Solution Temperature (LCST) of poly(NIPAM), indicating
the formation of strong aggregates. On the other hand, random copolymers
of AM and NIPAM do not precipitate from the solution (up to 80 °C). In
addition, depending on the composition, thermothickening behavior is
observed. Remarkably, the thermothickening behavior is only present at low
shear rates (
≤ 10 s-1). This, in connection with the ease of the synthesis,
makes these copolymers especially interesting for application in Enhanced Oil
Recovery (EOR).
Based on: D.A.Z. Wever, E. Riemsma, F. Picchioni, A.A. Broekhuis. Comb-like
thermoresponsive
polymeric
materials:
Synthesis
and
effect
of
macromolecular structure on solution properties. Polymer, 2013, 54, 54565466.
133
Branched thermosensitive copolymers
6.1. Introduction
Thermoresponsive (or thermosensitive) polymers have been the subject
of extensive research in the past decade due to their unique properties.1
Thermosensitive
polymers
contain
moieties
that
can
undergo
major
conformational transitions with changes in temperature. Generally speaking,
two different types of thermoresponsive polymers are distinguished. The first
type is a polymer for which the solubility in a given solvent improves with an
increase in temperature (upper critical solution temperature, UCST). 2 The
second type displays the exact opposite behavior, with the solubility
decreasing with temperature (LCST).3 Many different application fields have
been suggested for thermoresponsive polymers: drug delivery agents 4-6,
bioengineering6, 7, sensors8, 9, drag reduction10, 11, and enhanced oil recovery
(EOR)12, 13.
The focus has been mainly on poly(N-isopropylacrylamide) (PNIPAM)
because it possesses a sharp1 (i.e. relatively narrow) LCST window of 31-33
°C (independent of the polymer concentration14), which is close to the
temperature of the human body. Many efforts have been spent towards the
development of drug carriers based on PNIPAM.4,
15-18
Current investigations
are mainly focused on manipulating the LCST value. Several molecular
properties of the polymer affect the LCST: the molecular weight 19, 20 (or even
NIPAM oligomers21), the nature of the endgroups21,
structure
1,
10
22
, and the chemical
(i.e. incorporation of other, hydrophobic or hydrophilic,
monomers). The variety of monomers that have been copolymerized
(random, block and graft) with NIPAM is extensive and has been recently
reviewed.1,
10
To elucidate the effect of the different molecular properties on
the LCST, control in the polymerization of NIPAM is desirable. The controlled
polymerization of NIPAM, i.e. control in the molecular weight and the
dispersity index (PDI), has been accomplished by atomic transfer radical
polymerization (ATRP)19,
20,
23-25
, reversible addition-fragmentation chain
transfer (RAFT) polymerization26, 27 and living anionic polymerization28, 29.
Attention has mainly been given to the preparation of thermosensitive
gels based on NIPAM.30-32 When heated above their LCST, the NIPAM
moieties become dehydrated and effectively hydrophobic in nature. This
results in association, formation of thermoreversible aggregates and increase
in viscosity.1 Similar behavior is obtained using NIPAM based copolymers.
The first reports on this feature were on copolymers of acrylic acid (AA) with
NIPAM grafted on the poly(acrylic acid) (PAA) backbone.33,
34
Acrylamide
(AM) and N,N-dimethylacrylamide (DMA) as the hydrophilic block has also
been demonstrated to lead to a thermothickening behavior in water. 13,
Other moieties such as poly(ethylene oxide) (PEO)
134
36, 37
35
and, more recently,
Chapter 6
an AM based macromonomer38,
39
, as the thermosensitive block have been
investigated. Thee thermoviscosifying effect is observed at shear rates up to
= 800 s-1 indicating strong aggregation. In view of possible applications in
EOR, this might lead to the loss of injectivity. Close to the injection well, the
shear rates are high (due to the injection of large volumes through small
pores), and the temperatures are above the LCST of the polymers. The
significant increase in solution viscosity at higher temperatures and shear
rates will require high pumping pressure to enable injection of the polymer
solution. Ideally, the thermoviscosifying polymer should display a relatively
higher solution viscosity (e.g.  > 80 mPa.s) at higher temperatures (e.g. T
> 50 °C) but only at low shear rates ( < 30 s-1).
Despite the relevant number of studies already published, a systematic
investigation
of
the
copolymer
properties
(e.g.
surface
activity
and
rheological behavior in aqueous solutions) as function of the macromolecular
structure has not been yet reported. This is probably related to the difficulties
in achieving control over the co-polymerization process. With the advent of
synthetic strategies for AM based comb-like homo- and block-copolymers of
AM and NIPAM, the effect of the molecular architecture on the solution
properties of branched block-copolymers can now be probed.
Although significant progress has been booked in the synthesis of water
soluble thermothickening, the synthetic methods comprise multiple steps and
are not controlled thus leading to broad molecular weight distributions. In
addition, the thermothickening properties arise also at high shear ( > 800 s1
) rates which might be detrimental for application in EOR. Here we report
the controlled synthesis of branched terpolymers based on an aliphatic
polyketone backbone with a varying number of thermosensitive side chains
made of diblock (AM-b-NIPAM) or random (AM-ran-NIPAM) moieties. The
effect of the chemical structure, i.e. random or block, and the molecular
architecture (varying number of arms) on the solution properties is
presented. To the best our knowledge, this is the first report on a
thermoresponsive polymer that displays a thermothickening behavior only at
low shear rates ( < 30 s-1), which is crucial for application in EOR.
6.2. Experimental section
Chemicals.
Acrylamide
isopropylacrylamide
(Me6TREN)
(NIPAM,
(AM,
electrophoresis
97%),
grade,
≥99%),
N-
tris[2-(dimethylamino)ethyl]amine
copper(I) bromide (CuBr, 98%), glacial acetic acid, ethanol,
chloroform and diethyl ether were purchased from Sigma Aldrich. CuBr was
purified by stirring in glacial acetic acid for at least 5 hours, filtering, and
135
Branched thermosensitive copolymers
washing with glacial acetic acid, ethanol and diethyl ether (in that order) and
then dried at reduced pressure.40
Synthesis of the macro-initiator. The chemical modification of the
original PK was performed according to the published method 41 (Scheme
6.1). The reactions were performed in a sealed 250 ml round bottom glass
reactor with a reflux condenser, a U-type anchor impeller using an oil bath
for heating.
Scheme 6.1: Synthesis of the macro-initiators
The chloropropylamine hydrochloride 14.8 g (0.114 mol) was dissolved in
methanol (50 ml) to which an equimolar amount of sodium hydroxide (4.56
g, 0.114 mol) was added. After the polyketone (15 g, 0.114 mol of
dicarbonyl units) was preheated to the liquid state at the employed reaction
temperature (100 °C), the amine solution was added drop wise (with a drop
funnel) into the reactor in the first 20 min. The stirring speed was set at a
constant value of 500 RPM. During the reaction, the mixture of the reactants
changed from a slightly yellowish, low viscosity state, into a highly viscous
brown homogeneous paste. The product was dissolved in chloroform and the
organic phase was washed afterwards with demineralized water in a
separation funnel. The polymer was isolated by evaporating the chloroform at
low pressure (100 mbars). The product, a brown powder, was finally freeze
dried and stored at -18 °C until further use. The macro-initiator was
characterized using elemental analysis, 1H-NMR spectroscopy (in chloroform),
and Gel Permeation Chromatography (GPC).
The conversion of carbonyl groups of the polyketone was determined
using the following formula:
136
Chapter 6
(6.1)
the average number of carbons in n-m (see Scheme 6.1)
, the average number of carbons in m (see Scheme 6.1)
atomic weight of nitrogen
atomic weight of carbon
The average number of pyrrole units was determined using the conversion of
the carbonyl groups of the polyketone and formula 6.2:
(6.2)
= the average molecular weight of the parent (unmodified) polyketone
= the average molecular weight of the repeating unit of polyketone
Comb polymers preparation. A 250-ml three-necked flask was charged
with the macroinitiator (e.g. entry PK30-g13-(PAM3275), 0.3279 g, 0.117
mmol macroinitiator or 1.521 mmol Cl-groups). Enough acetone (typically 510 ml) was added to dissolve the macro-initiator. Demineralized water (400
ml) and acrylamide (100 g, 1400 mmol) were then added to the solution.
Subsequently, the mixture was degassed by three freeze-pump-thaw cycles.
A nitrogen atmosphere was maintained throughout the remainder of the
reaction steps. CuBr (25 mg, 0.174 mmol) was then added to the flask and
the mixture stirred for 10 minutes. The flask was then placed in an oil bath at
25 °C. The reaction was started by the addition of the ligand (Me6TREN, 40
mg, 0.174 mmol) using a syringe.
After the pre-set reaction time, the
mixture was exposed to air and the polymer was precipitated in a tenfold
amount of methanol. For the higher molecular weight polymers the solution
was first diluted with demineralized water before being precipitated. The
polymer was isolated by filtration and subsequently dried in an oven at 65
°C.
Block Copolymerization. The prepared PK30-g-PAM (vide supra) was
used as macroinitiator for NIPAM polymerization. A round bottomed three
necked flask was charged with the macroinitiator (e.g. entry PK30-g13(PAM3275-b-PNIPAM4425), 4.38 g, 0.0186 mmol macroinitiator or 0.242
mmol Cl-groups) and NIPAM (21.03g, 186 mmol). Double distilled water was
137
Branched thermosensitive copolymers
added, and the mixture was degassed by three freeze-pump-thaw cycles.
Afterwards CuBr (3.5 mg, 0.024 mmol) was added and the solution was
stirred for 10 min. The flask was placed in a water bath at 25 °C and the
reaction was started by adding Me6TREN (5.5 mg, 0.024 mmol). All the
operations were performed under nitrogen. After the reaction, the mixture
was terminated by the addition of demineralized water (80 mL). The polymer
was precipitated in a tenfold amount of methanol and dried in an oven at 65
°C. The polymer was re-dissolved in demineralized water and dialyzed
(Spectra/Por® Dialysis Membrane, molecular weight cut off = 2 000 g/mol)
and subsequently dried in an oven at 65 °C up to constant weight. The codes
for the block copolymers are defined as PK30-gX-(PAMY-b-PNIPAMZ) with X,
Y and Z the designation for the number of arms, number of AM and NIPAM
units respectively.
Random
Copolymerization.
The
polyketone
macroinitiator
was
synthesized according to the aforementioned procedure. A three-necked flask
is
charged
with
the
polyketone
macroinitiator
(e.g.
entry
PK30-g13-
(PAM1405-co-PNIPAM1405), 0.0983g 0.035mmol) and acetone (10 mL).
Hereafter, a magnetic stirrer, AM (15 g, 210 mmol) and NIPAM (23.8 g, 210
mmol) dissolved in demineralized water (150 mL) were added and the
mixture was degassed by three freeze-pump-thaw cycles. The flask was
placed in a thermostated water bath and stirred constantly at 25 °C after
which the CuBr (7.6 mg, 0.053 mmol) was added. The reaction was started
with the addition of Me6TREN (12.2 mg, 0.053 mmol). All operations were
carried out under nitrogen. After the reaction, the mixture was terminated by
the addition of demineralized water (750 mL) and a sample was taken for GC
analysis
(acrylamide
conversion).
The
polymer
solution
was
dialyzed
(Spectra/Por® Dialysis Membrane, molecular weight cut off = 12 000 - 14
000 g/mol) and subsequently dried in an oven at 65 °C up to constant
weight. The codes for the block copolymers are defined as PK30-gA-(PAMBco-PNIPAMC) with A, B and C the designation for the number of arms,
number of AM and NIPAM units respectively.
Characterization. The acrylamide conversion was measured by using
Gas Chromatography (GC). The samples (taken from the reaction mixtures)
were dissolved in acetone (polymer precipitates) and injected on a Hewlett
Packard 5890 GC with an Elite-Wax ETR column. The total molecular weight
(Mn,tot) is calculated by using the acrylamide conversion (monomer-initiator
ratio multiplied by the conversion). The span molecular weight (Mn,SPAN) is
calculated using the Mn,tot and is defined as two times the molecular weight of
one arm (star PAM) or two times the molecular weight of one arm plus the
molecular weight of the macro-initiator (comb PAM).
138
Chapter 6
Nuclear Magnetic Resonance (NMR) spectra were recorded on a Varian
Mercury Plus 400 MHz spectrometer. For analysis D 2O was used as the
solvent. For measurements at higher temperatures, at least 10 minutes was
waited before recording the spectrum.
The macroinitiators were analyzed by GPC using THF (used as received)
as the eluent with toluene as a flow marker. The analysis was performed on a
Hewlett Packard 1100 system equipped with three PL-gel 3 m MIXED-E
columns in series. The columns were operated at 42 °C with a flow-rate of 1
ml/min, and a GBC LC 1240 RI detector was used at 35 °C. The apparent
molecular weights and dispersities were determined using polystyrene
standards and WinGPC software (PSS).
Elemental
analysis of the
macroinitiators was performed on the
EuroEA3000-CHNOS analyzer (EUROVECTOR Instruments and Software).
Approximately 2 mg of each sample is weighed and placed in tin samplecups. The reported values are the average of 2 runs.
Surface tension was measured using the pendant drop method on a
LAUDA DROP VOLUME TENSIOMETER TVT 1. A glass micro syringe was
attached to a needle with a capillary radius of 1.055 mm. The temperature of
the water bath was set to 25 °C and the density difference between air and
water was set to 0.997 g/mL. Two sets of three measurements were taken
and then averaged. Graphically the critical micelle concentration (CMC) can
be obtained from the plot of the surface tension against the concentration by
taking the line of best fit in two places and noting the concentration at the
intersection (not shown for brevity).42
Rheological characterization. The aqueous polymeric solutions were
prepared by swelling the polymers in water for one day and afterwards gently
stirring the solution for another day.
Viscometric measurements were performed on a HAAKE Mars III
(ThermoScientific) rheometer, equipped with a cone-and-plate geometry
(diameter 60 mm, angle 2°). Flow curves were measured by increasing the
shear stress by regular steps and waiting for equilibrium at each step. The
shear rate ( ) was varied between 0.1 – 1750 s-1. Dynamic measurements
were performed with frequencies ranging between 0.04 – 100 rad/s (i.e.,
6.37·10-3 – 15.92 Hz). It must be noted that all the dynamic measurements
were preceded by an oscillation stress sweep to identify the linear
viscoelastic response of each sample. With this, it was ensured that the
dynamic measurements were conducted in the linear response region of the
samples.
139
Branched thermosensitive copolymers
6.3. Results and discussion
6.3.1. Macroinitiators.
The synthesis of the macroinitiators was performed according (Scheme
6.1) to the Paal-Knorr reaction of a halogenated primary amine with aliphatic
perfectly alternating polyketones.41 The carbonyl conversion was determined
using elemental analysis. Resonance peaks corresponding to the pyrrole units
were observed with 1H-NMR spectroscopy at  5.68 ppm while the -, -, and
-hydrogens (relative to the halogen) were detected at  3.51, 1.95, and 3.86
ppm respectively (Figure 6.1). The resonance of the pyrrole as well as those
for the - and -hydrogens (relative to the halogen) all increase with the
conversion of the Paal-Knorr reaction. The obtained, chemically modified
polyketones are used as macrointiators in the synthesis of block or random
comb-copolymers.
Table 6.1: Properties of the macroinitiator and parent polyketone
Polyketone sample
(PK30-Cla)
PK30 (virgin)
PK30-Cl4
PK30-Cl8
PK30-Cl13
Elemental composition
(C : H : N, wt%)
67.0
58.6
64.0
62.9
:
:
:
:
8.4
7.1
7.9
7.6
:
:
:
:
XCO (%)b Pyrrole unitsc
0
1.6
3.3
4.9
18.87
37.21
61.14
0
4
8
13
Mn,GPC
PDI
2
2
2
2
1.74
2.02
2.01
1.97
797
447
244
072
a. Number indicates the ethylene content (%) and Cl indicates the halogen present
b. The conversion of the carbonyl groups of the polyketone
c. Average number of pyrrole units per chain
The properties of the macroinitiators are given in Table 6.1. The molecular
weight (relative to that of polystyrene) of the macroinitiator decreases as the
conversion of the carbonyl groups increases. The decrease is probably due to
the decrease in hydrodynamic volume caused by the formation of the pyrrole
rings in the backbone of the macroinitiator.
140
Chapter 6
a
a
b
c
d
c
d
b
a
PK30-Cl13
d
c
d
c
b
PK30-Cl8
a
PK30-Cl4
a
b
PK30, virgin
7
6
5
4
3
2
1
ppm
Figure 6.1: 1H-NMR spectra of the macroinitiator and the virgin polyketone
6.3.2. Synthesis of PK30-g-(PAM-b-PNIPAM)
The synthesis of the comb graft block-copolymers was performed
according to scheme 6.2A by using different molar ratios between the
macroinitiator and AM (and NIPAM). First the ATRP of AM was carried out by
following a published method.43
Scheme 2: (A) Block copolymerization of AM and NIPAM and (B) Random
copolymerization of AM and NIPAM on the polyketone based macroinitiator
141
Branched thermosensitive copolymers
Afterwards NIPAM was polymerized as the second block and the NIPAM
conversion was determined by
1
H-NMR (Figure 6.2). A summary of the
experimental conditions applied to prepare the different block copolymers as
well as their GPC analysis is given in Table 6.2. As can be observed in Table
6.2, different graft block copolymers can be prepared where the average
number of grafts, the length of the AM and NIPAM blocks can be
systematically varied. The acrylamide conversion is lower for the higher
functionalized
polyketone
macroinitiators.
As
mentioned
before,
the
theoretical maximum number of side groups that can be obtained is 21.
Table 6.2: Synthesis of the different PK30-g-(PAM-b-PNIPAM) block copolymers
[M]0:[I]0:[L]0:[C]0a
Entry
PK30-g4-(PAM7575)
PK30-g4-(PAM7575-b-PNIPAM15)
PK30-g4-(PAM7575-b-PNIPAM1690)
PK30-g8-(PAM7770)
PK30-g8-(PAM7770-b-PNIPAM60)
PK30-g8-(PAM7770-b-PNIPAM480)
PK30-g13-(PAM3275)
PK30-g13-(PAM3275-b-PNIPAM415)
PK30-g13-(PAM3275-b-PNIPAM4425)
PK30-g13-(PAM6140)
PK30-g13-(PAM6140-b-PNIPAM205)
M/water (wt:vol);
T; Time (min)
Conv (%)
AMb NIPAMc
Mn
DP
NIPAM
DP
PAM
10 005:1:1:1.5
8 000:1:1:1.5
3 000:1:1:1.5
1: 6.0;25 °C;
60 75.7
1:17.4;25 °C; 1440
1:46.5;25 °C; 1425
-
0.2
56.3
541 329
0
543 254
15
732 393 1690
7575
7575
7575
10 025:1:1:1.5
100 000:1:1:1.5
8 000:1:1:1.5
1: 6,0;25 °C;
60 77.5
1: 4.2;25 °C; 1405
1:18.6;25 °C; 1440
-
0.06
6.0
554 982
561 705
609 368
0
60
480
7770
7770
7770
12 000:1:1:2.5
10 000:1:1:1.5
10 000:1:1:1.5
12 000:1:1:1.5
4 965:1:1:1.5
1: 4.0;25 °C; 1440 27.3
1:13.3;25 °C; 1230
1:10.9;25 °C; 1425
1: 6.0;25 °C;
60 51.2
1:31.7;25 °C; 1405
-
4.2
44.2
4.1
235
282
736
439
462
504
0
579
415
146 4425
326
0
515
205
3275
3275
3275
6140
6140
a: AM is the monomer ([M]0) for the homopolymers and NIPAM is the monomer ([M] 0) for the block
copolymers
b: The AM conversion was determined by GC
c. The NIPAM conversion was determined by 1H-NMR
142
Chapter 6
The higher the functionalization degree, the more sterically hindered the
macroinitiator is. Therefore, the reactivity of the macroinitiator will decrease
as the number of halogen atoms increases (due to steric hindrance 44),
eventually leading to lower AM conversions. The resonance labelled (Figure
6.2) as a (=3.9 ppm) represent the hydrogen atom of the CH group of the
isopropyl group of the PNIPAM blocks and therefore the size of this resonance
(in relation to the resonances labelled b and c,  in the range 1.2 – 2.5 ppm)
corresponds to the amount of NIPAM polymerized on the PK30-gX-(PAMY)
macroinitiator. The total area of the resonances labelled b and c correspond
to the protons from the backbone of the PK30, PAM and PNIPAM along with
the 6 methyl protons of PNIPAM blocks (2× CH3). The
1
H-NMR data is in
agreement with solubility tests, and confirms the preparation of block
copolymers with different lengths of the blocks.
c
D2O
c a
a
b
c
PK30-g13-(PAM3275-b-PNIPAM4425)
PK30-g13-(PAM3275-b-PNIPAM415)
PK30-g13-(PAM3275)
PK30-g8-(PAM7770-b-PNIPAM480)
PK30-g8-(PAM7770-b-PNIPAM60)
PK30-g8-(PAM7770)
PK30-g4-(PAM7575-b-PNIPAM1690)
PK30-g4-(PAM7575-b-PNIPAM15)
PK30-g4-(PAM7575)
5
4
3
2
1
0
ppm
Figure 6.2: 1H-NMR spectra of the block copolymers PK30-gX-(PAMY-b-PNIPAMZ) and
the parent macroinitiators PK30-gX-(PAMY)
143
Branched thermosensitive copolymers
6.3.3. Synthesis of PK30-g-(PAM-co-PNIPAM)
The synthesis of the comb graft random-copolymers was performed
according to scheme 6.2B using different molar ratios between the
macroinitiator and AM (and NIPAM). Random ATRP of both monomers was
conducted in water. The experimental conditions and GPC data are
summarized in Table 6.3.
Increasing the [M]0:[I]0 yielded, as expected, higher molecular weight
copolymers. However the conversion of the monomers is lower indicating
mass transfer limitations43 due to the significant increase in the viscosity of
the reaction mixture. The monomer conversion also decreases as the number
of arms on the parent macroinitiator increases. This is in line with earlier
results on the polymerization of AM on the same macroinitiator in water
(Chapter 3). The molar ratio between AM and NIPAM of the copolymers was
similar to the molar ratio of the reaction mixture in all the cases, thus
suggesting a random distribution of the units. This is in line with an earlier
report10 on the free radical copolymerization of AM and NIPAM. Two samples
of the reaction mixture, one taken at low conversion and one at high
conversion, displayed the same molar ratio between the two monomeric units
suggesting a perfectly random distribution.10
Table 6.3: Synthesis of the different PK30-g-(PAM-co-PNIPAM) random copolymers
[M]0[n:m]:[I]0a
Entry
M/water (wt:vol),
T, Time (min)
Conv (%)
Mn, GC
AMa NIPAMb
DP
PAM
DP
NIPAM
12 000[1:1]:1 1:4.0;25 °C;
50 000[1:1]:1 1:4.0;25 °C;
100 000[1:1]:1 1:4.0;25 °C;
210 83.6
210 75.3
210 66.8
81.4
76.8
63.6
908 999 5 015
4 885
3 518 808 18 875 19 240
5 970 955 33 395 31 790
12 000[1:1]:1
75 000[1:1]:1
50 000[1:1]:1
100 000[1:1]:1
1:4.0;
1:4.0;
1:4.0;
1:4.0;
25
25
25
25
°C;
°C;
°C;
°C;
210
210
210
185
73.4
29.4
50.3
37.4
74.4
26.0
50.0
35.3
817 851 4 400
4 460
1 945 492 11 125 10 205
2 302 870 12 540 12 475
3 300 961 18 575 17 510
12 000[1:1]:1
100 000[2:1]:1
PK30-g13-(PAM6320-co-PNIPAM6000)
25 000[1:1]:1
PK30-g13-(PAM8375-co-PNIPAM8130)
50 000[1:1]:1
PK30-g13-(PAM9620-co-PNIPAM9620)
75 000[1:1]:1
PK30-g13-(PAM12140-co-PNIPAM11690) 100 000[1:1]:1
1:3.9;
1:4.0;
1:4.0;
1:4.0;
1:4.0;
1:4.0;
25
25
25
25
25
25
°C;
°C;
°C;
°C;
°C;
°C;
210
210
210
210
210
915
23.4
7.7
50.5
33.5
25.5
24.4
23.4
7.6
48.0
32.5
25.5
23.4
258 914 1 405
1 405
651 397 5 135
2 530
127 916 6 320
6 000
515 267 8 375
8 130
772 689 9 620
9 620
183 519 12 140 11 690
PK30-g4-(PAM5015-co-PNIPAM4885)
PK30-g4-(PAM18875-co-PNIPAM19240)
PK30-g4-(PAM33395-co-PNIPAM31790)
PK30-g8-(PAM4400-co-PNIPAM4460)
PK30-g8-(PAM11125-co-PNIPAM10205)
PK30-g8-(PAM12540-co-PNIPAM12475)
PK30-g8-(PAM18575-co-PNIPAM17510)
PK30-g13-(PAM1405-co-PNIPAM1405)
PK30-g13-(PAM5135-co-PNIPAM2530)
1
1
1
2
a: AM is the monomer ([M]0) for the homopolymers, n:m is the molar ratio between AM and NIPAM
b: The AM conversion was determined by GC
c. The NIPAM conversion was determined by 1H-NMR
144
Chapter 6
6.3.4. Solution properties of PK30-gX-(PAMY-b-NIPAMZ)
Solution viscosity as a function of shear. The viscosity of the
polymer dissolved in demineralized water versus shear rate is displayed is
displayed in Figure 6.3. The polymers used are characterized by different
hydrophilic (AM) / hydrophobic (NIPAM) ratios and a different number of
arms. All polymers consisted of a comb-like hydrophilic block of AM and a
hydrophobic block of NIPAM of different average lengths (and thus different
total molecular weight).
Viscosity (Pa.s)
A
10
0
PK30-g4-(PAM7575-b-PNIPAM1690)
PK30-g4-(PAM7575-b-PNIPAM15)
PK30-g4-(PAM7575)
10
-1
1
10
100
1000
-1
10
1
10
0
B
PK30-g8-(PAM7770-b-PNIPAM480)
10
10
Viscosity (Pa.s)
Viscosity (Pa.s)
Shear rate (s )
10
-1
PK30-g13-(PAM3275-b-PNIPAM4425)
PK30-g8-(PAM7770-b-PNIPAM60)
PK30-g13-(PAM3275-b-PNIPAM415)
PK30-g8-(PAM7770)
PK30-g13-(PAM3275)
-1
10
1
C
0
10
100
-1
Shear rate (s )
1000
-2
1
10
100
1000
-1
Shear rate (s )
Figure 6.3: Viscosity vs shear rate of A: PK30-g4-(PAMY-b-PNIPAMZ) at a polymer
concentration of 5 wt.%, B: PK30-g8-(PAMY-b-PNIPAMZ) at a polymer concentration of
4 wt.%, and C: PK30-g13-(PAMY-b-PNIPAMZ) at a polymer concentration of 3 wt.%
As can be observed, the addition of a NIPAM block to the branched
homopolymer leads to an increase in the solution viscosity and a more
pronounced pseudo-plastic behaviour. The increase in solution viscosity is
145
Branched thermosensitive copolymers
related to the increased molecular weight of the polymers and the presence
of the NIPAM blocks (Chapter 5).
Solution viscosity as a function of temperature. The viscosity was
measured as a function of the temperature of the solution and the results are
displayed in Figure 6.4. All polymers consisted of a hydrophilic block of AM
(roughly 525 000 g/mol or 7650 AM units for PK30-g4 and PK30-g8, 235 000
g/mol or 3275 AM units for PK30-g13) and a hydrophobic block of PNIPAM of
differing length, resulting in block copolymers with different total molecular
weights. The shear rate during the temperature sweep was fixed at a value of
30.0 s-1.
2
PK30-g4-(PAM7575-b-PNIPAM1690)
PK30-g4-(PAM7575-b-PNIPAM15)
Viscosity (Pa.s)
PK30-g4-(PAM7575)
1
A
0
20
30
40
50
60
70
80
o
Temperature ( C)
10
2
1
PK30-g13-(PAM3275-b-PNIPAM415)
PK30-g8-(PAM7770-b-PNIPAM60)
PK30-g13-(PAM3275)
Viscosity (Pa.s)
Viscosity (Pa.s)
PK30-g8-(PAM7770)
1
C
PK30-g13-(PAM3725-b-PNIPAM4425)
PK30-g8-(PAM7770-b-PNIPAM480)
10
10
0
-1
B
0
20
30
40
50
60
o
Temperature ( C)
70
80
0
20
30
40
50
60
70
o
Temperature ( C)
Figure 6.4: Viscosity ( = 30 s-1) versus temperature of A: 5 wt.% PK30-g4-(PAMY-bPNIPAMZ), B: 4 wt.% PK30-g8-(PAMY-b-PNIPAMZ), C: 3 wt.% PK30-g13-(PAMY-bPNIPAMZ)
146
80
Chapter 6
To illustrate the effect of NIPAM on the behaviour of the block-copolymers in
solution as a function of temperature, the corresponding homopolymers are
also displayed. The viscosities of the polymer solutions decrease as the
temperature increases (Figure 6.4). The curves for the block copolymers
display a sharp drop at a temperature of approximately 32-34 °C, close to
the LCST of the NIPAM homopolymer. When the temperature of the polymer
solution reaches the LCST, the isopropyl groups of the PNIPAM blocks are
dehydrated and aggregation between the PNIPAM blocks arises. 45 Most of the
copolymers precipitate out of the solution as the temperature is increased to
above 32 °C (Figure 6.5), indicating strong hydrophobic interactions.
Figure 6.5: Precipitation of the block copolymers at temperatures above 32 °C and no
precipitation of the random copolymers at temperatures up to 80 °C
The same behaviour is observed for linear block copolymers of AM and NIPAM
(Chapter 5). Only entries PK30-g4-(PAM7575-b-PNIPAM15) and PK30-g8(PAM7770-b-PNIPAM60) do not precipitate out of the solution. In these
cases, the length of the hydrophilic block appears to be long enough to keep
the block copolymer in solution at temperatures above 32 °C. The
precipitation of the block copolymers is detrimental for possible application in
EOR, as the precipitates will probably block the porous media (reservoir).
The steep increase in solution viscosity of PK30-g13-(PAM3725-bPNIPAM4425) is caused by the formation of gel particles (Figure 6.4C,
picture) in the rheometer leading to a higher friction and thus a higher
apparent viscosity. However, the values for the solution viscosity are not
reliable due to the precipitation of the copolymer. To investigate the
thermoresponsive character of the NIPAM blocks at a molecular level,
1
H-
NMR spectra were recorded at different temperatures (Figure 6.6). The
147
Branched thermosensitive copolymers
resonances (a and c) corresponding to the NIPAM blocks (of the block
copolymer) disappear almost completely above 30 °C. This indicates that the
NIPAM blocks precipitate out of the solution. Similar results have been
obtained for
diblock star copolymers of NIPAM and 2-hydroxyethyl
methacrylate46,
copolymers
of
NIPAM
and
vinyl
laurate47,
miktoarms
multihydrophilic star block copolymers based NIPAM, acrylic acid and vinyl
pyrrolidone48,
and
on
the
hompolymer
of
NIPAM.49
The
precipitates
correspond to the gel particles observed in the rheometer. The comparison to
a random copolymer show that the resonances (a and c) for the random
copolymer are still present, even at 75 °C.
PK30-g13-(PAM3275-b-PNIPAM4425)
D2O
c
c
a
25°C
a
30°C
35°C
50°C
PK30-g13-(PAM1405-co-PNIPAM1405)
c
c
a
25°C
a
5
75°C
4
3
2
1
0
ppm
Figure 6.6: 1H-NMR spectra of entries PK30-g13-(PAM3275-b-PNIPAM4425) and PK30g13-(PAM1405-co-PNIPAM1405) at different temperatures
6.3.5. Solution properties of PK30-gX-(PAMY-co-NIPAMZ)
Solution viscosity as a function of shear. The solution viscosity as a
function of shear rate for some of the random copolymers is displayed in
Figure 6.7A. Increasing the amount of NIPAM in the random copolymer leads
to a reduction in the solution viscosity (Figure 6.7A). The comparison
between the entries PK30-g13-(PAM5135-co-PNIPAM2530) and PK30-g13(PAM6320-co-PNIPAM6000) demonstrates that the incorporation rate of
NIPAM in the random copolymer has a strong effect on the solution viscosity.
The decrease in solution viscosity can be attributed to the reduction in the
strength of the hydrogen bonds that arise in the solution. The copolymer with
148
Chapter 6
a 1-1 molar ratio for AM and NIPAM units is an ideal random copolymer, i.e.
an alternating distribution of the two monomer units. The proximity of the
NIPAM units and the AM units will disrupt the hydrogen bonds. The
copolymer with a 2-1 molar ratio will have a lower degree of disruption due
to a lower number of NIPAM units. In addition, the effective lengths of the
PAM blocks are longer in the latter copolymer and will lead to stronger
interactions in solution. This behaviour resembles that of hydrophobic
interactions,
where
longer
hydrophobic
groups
will
have
stronger
interactions.12
PK30-g13-(PAM5135-co-PNIPAM2530)
Viscosity (Pa.s)
PK30-g13-(PAM6320-co-PNIPAM6000)
10
1
10
0
10
-1
PK30-g13-(PAM1405-co-PNIPAM1405)
A
0,1
1
10
100
1000
-1
Shear rate (s )
10
1
10
0
o
20 C
o
50 C
o
80 C
Viscosity (Pa.s)
PK30-g13-(PAM1405-co-PNIPAM1405)
10
-1
10
-2
0,1
B
1
10
100
1000
-1
Shear rate (s )
Figure 6.7: Viscosity functions of A: different PK30-g13 random copolymers at 20 °C
and B: PK30-g13-(PAM1405-co-PNIPAM1405) at different temperatures
The viscosity function of PK30-g13-(PAM1405-co-PNIPAM1405) at different
temperatures displays a peculiar behaviour (Figure 6.7B). Increasing the
149
Branched thermosensitive copolymers
temperature to 50 °C leads to a reduction of the solution viscosity (at
> 1
s-1), due to the reduced solvent viscosity and lower strength50 of the
hydrogen bonds. However, at low shear rates ( < 0.5 s-1), the solution
viscosity is equal or higher than the values at 20 °C. At higher temperatures
≤ 3 s-1), the solution viscosity is
and low shear rate (T = 80 °C and
significantly higher which points to, weak interactions between the polymer
chains. As the shear rate is increased the aggregates are disrupted and the
solution viscosity reduces to values lower than those at 50 °C.
Solution viscosity as a function of temperature. The viscosity was
measured as a function of the temperature of the solution containing the
random copolymers and the results are displayed in Figure 6.8.
B
A
10
0
Viscosity (Pa.s)
Viscosity (Pa.s)
10
0
=3s
 = 30 s-1
-1
=3s
 = 30 s-1
-1
PK30-g8-(PAM4400-co-PNIPAM4460)
PK30-g4-(PAM5015-co-PNIPAM4885)
10
-1
10
0
20
30
40
50
60
70
-1
0
80
20
30
0
=1s
 = 5 s-1
-1
 = 50 s
-1
70
80
D
10
-1
Viscosity (Pa.s)
Viscosity (Pa.s)
60
C
PK30-g13-(PAM1405-co-PNIPAM1405)
10
50
Temperature ( C)
Temperature ( C)
10
40
o
o
-1
=1s
 = 5 s-1
-1
 = 50 s
-1
=1s
-1
=5s
 = 50 s-1
-1
PK30-g13-(PAM9620-co-PNIPAM9620)
PK30-g13-(PAM3275)
10
-2
10
0
20
30
40
50
60
o
Temperature ( C)
70
80
-2
0
20
30
40
50
60
70
o
Temperature ( C)
Figure 6.8: Viscosity versus temperature of (A): 3 wt.% PK30-g4-(PAM5015-coPNIPAM4885), (B): 3 wt.% PK30-g8-(PAM4400-co-PNIPAM4460), (C):2 wt.% PK30-g13(PAM1405-co-PNIPAM1405) and PK30-g13-(PAM3275), and (D): 1 wt.% PK30-g13(PAM9620-co-PNIPAM9620)
150
80
Chapter 6
The random copolymers display peculiar behavior, as an increase in viscosity
is observed at higher temperatures at low shear rates (
< 5 s-1). At higher
shear rates, this increase can no longer be distinguished. The increase in
solution viscosity is attributed to the formation of aggregates due to the
hydrophobic character of the NIPAM moieties above the LCST (i.e. T > 32
°C). Similar to the block copolymers, the solutions were visually inspected for
precipitation upon heating (Figure 6.5). As can be observed no precipitation
occurs upon heating for 5 minutes till 80 °C. Similar results, i.e. an increase
in solution viscosity at higher temperatures, have been reported on
copolymers of NIPAM with either AM13 or acrylic acid (AA)33, 34.
The random copolymer PK30-g13-(PAM1405-co-PNIPAM1405) and the
homopolymer PK30-g13-(PAM3275), both with a similar molecular weight (Mn
≈ 235 000 g/mol) are compared in Figure 6.8C. As can be observed in the
figure, the difference in solution viscosity increases as the temperature is
increased from 20 to 80 °C. This demonstrates the potential of the random
copolymers for application in EOR, especially for reservoirs where the
temperature exceeds 70 °C (a limit above which usually the currently used
partially hydrolyzed polyacrylamides [HPAM] are not applied).
6.3.6. Surface properties
Critical micelle concentration (CMC). The surface tension (against
air) of the polymer solutions is plotted against the concentration (Figure 6.9)
in order to determine the CMCs. S-shaped curves are obtained, in agreement
with data on similar systems.51 For low polymer concentrations the solutions
move towards the surface tension of demineralized water (70.5 mN/m). As
the polymer concentration increases, the surface tension reaches a regime
where a strong decrease can be observed. Then at a specific concentration,
the surface tension stops decreasing with a minimum value near 45 mN/m
for the block copolymers. This specific concentration is known as the critical
micelle
concentration
(CMC).
The
CMC
can
be
determined
as
the
concentration at the intersection of two lines of best fit at two places of the
plots in Figure 6.942 (not shown for brevity). As the PNIPAM block length
increases the concentration required for micelle formation decreases. This is
clearly demonstrated by the comparison between the entries PK30-g13(PAM3275-b-PNIPAM415)
and
PK30-g13-(PAM3275-b-PNIPAM4425).
The
longer the PNIPAM blocks are, the larger the effect of its lower hydrophilicity
will be.52 The order of magnitudes 10-6 and 10-7 M is in line with literature
reports for amphiphilic block copolymers.53 All the block copolymers display a
surface tension close to the value of pure PNIPAM (42 mN/m), albeit with
different CMCs. This is in line with the results on linear block copolymers of
151
Branched thermosensitive copolymers
AM and NIPAM (Chapter 5). The values of the branched block copolymers is
slightly higher than that of the linear block copolymers; this is most probably
due to the higher molecular weight of the branched block copolymers (higher
molecular weight is known to increase the surface tension54).
75
75
70
65
 (mN/m)
Surface tension (mN/m)
70
65
60
55
PK30-g13-(PAM5135-co-PNIPAM2530)
PK30-g13-(PAM1405-co-PNIPAM1405)
50
60
45
40
1E-9
1E-8
1E-7
1E-6
1E-5
Concentration (mol %)
55
PK30-g13-(PAM3275-b-PNIPAM4425)
50
PK30-g4-(PAM7575-b-PNIPAM1690)
PK30-g13-(PAM3275-b-PNIPAM415)
PK30-g8-(PAM7770-b-PNIPAM480)
45
40
1E-9
PK30-g13-(PAM6140-b-PNIPAM205)
1E-8
1E-7
1E-6
1E-5
1E-4
Concentration (M)
Figure 6.9: Surface tension against the polymer concentration of 5 different block
copolymers and 2 different random copolymers (insert)
The random copolymers display a similar behavior, although the final surface
tension is higher than that of the block copolymers. Depending on the
composition a different final surface tension is obtained. This is in line with
earlier results on random copolymers of AM and NIPAM (prepared by free
radical polymerization).10 The surface tension for the 1-1 (molar ratio)
random copolymer PK30-g13-(PAM1405-co-PNIPAM1405) is 60.0 mN/m which
is higher than that reported52 for a linear random copolymer (54.0 mN/m).
6.4. Conclusion
Different comb-like block and random copolymers based on acrylamide
(AM) and N-isopropylacrylamide (NIPAM) have been prepared by atom
transfer radical polymerization (ATRP) in water at room temperature. The
152
Chapter 6
average number of side-arms, AM and NIPAM block lengths, and the type of
distribution of the two monomers (block or random) were varied. The
aqueous solution properties of the different copolymers were investigated.
Particular emphasis is dedicated to the thermo-responsiveness of aqueous
solutions
containing
the
copolymers.
The
block
copolymers
tend
to
precipitate out of the solution at temperatures above the LCST of PNIPAM,
indicating the formation of strong aggregates.
1
H-NMR confirmed the
precipitation of the block copolymers, where the resonances corresponding to
the isopropyl groups disappear completely above 32 °C. On the other hand,
random copolymers of AM and NIPAM do not precipitate out of the solution
(the isopropyl resonances were still present at 75 °C and their integral
remained
the
same).
In
addition,
depending
on
the
composition,
thermothickening behavior is observed. Reducing the amount of NIPAM (from
50 to 25 mol%) in the random copolymer led to a less pronounced
thermothickening
behavior.
The
increase
in
solution
viscosity
with
temperature is only present at low shear rates ( ≤ 10 s-1).
The CMC is a function of the molar ratio between AM and NIPAM and
their distribution. The surface tension of the block copolymers is close to the
value for pure PNIPAM, while that of the random copolymers is a function of
the composition. The lower surface tension of the solutions might be
beneficial in the recovery of oil. Coupled with the ease of synthesis, the
potential production of polyketones at a commercial scale and the solution
behavior at higher temperatures, the branched random copolymers are
potential candidates for application in EOR.
6.5. Acknowledgements
This work is part of the Research Program of the Dutch Polymer Institute
DPI, Eindhoven, The Netherlands, project #716.
6.6. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
Liu, R.; Fraylich, M.; Saunders, B. R. Colloid Polym. Sci. 2009, 6, 627.
Koyama, M.; Hirano, T.; Ohno, K.; Katsumoto, Y. J Phys Chem B 2008, 35, 10854.
Liu, H. Y.; Zhu, X. X. Polymer 1999, 25, 6985.
Xiaojiao Kang; Ziyong Cheng; Dongmei Yang; Ping'an Ma; Mengmeng Shang; Chong
Peng; Yunlu Dai; Jun Lin Advanced Functional Materials 2012, 7, 1470.
Fundueanu, G.; Constantin, M.; Ascenzi, P. Int. J. Pharm. 2009, 1, 9.
Ward, M. A.; Georgiou, T. K. Polymers 2011, 3, 1215.
Stayton, P. S.; Shimoboji, T.; Long, C.; Chilkoti, A.; Chen, G. H.; Harris, J. M.;
Hoffman, A. S. Nature 1995, 6556, 472.
Akiyama, H.; Tamaoki, N. Macromolecules 2007, 14, 5129.
Hoare, T.; Pelton, R. Macromolecules 2007, 3, 670.
153
Branched thermosensitive copolymers
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
Mumick, P. S.; Mccormick, C. L. Polym. Eng. Sci. 1994, 18, 1419.
Mumick, P. S.; Hester, R. D.; Mccormick, C. L. Polym. Eng. Sci. 1994, 18, 1429.
Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Prog. Polym. Sci. 2011, 1558.
Sudor, J.; Barbier, V.; Thirot, S.; Godfrin, D.; Hourdet, D.; Millequant, R.;
Blanchard, J.; Viovy, J. L. Electrophoresis 2001, 4, 720.
Fujishige, S.; Kubota, K.; Ando, I. J. Phys. Chem. 1989, 8, 3311.
Bae, Y. H.; Okano, T.; Hsu, R.; Kim, S. W. Makromolekulare Chemie-Rapid
Communications 1987, 10, 481.
Coughlan, D. C.; Quilty, F. P.; Corrigan, O. I. J. Controlled Release 2004, 1, 97.
Lue, S. J.; Chen, C.; Shih, C.; Tsai, M.; Kuo, C.; Lai, J. J. Membr. Sci. 2011, 1-2,
330.
Purushotham, S.; Ramanujan, R. V. Acta Biomaterialia 2010, 2, 502.
Xia, Y.; Yin, X. C.; Burke, N. A. D.; Stover, H. D. H. Macromolecules 2005, 14,
5937.
Hu, H.; Du, J.; Meng, Q.; Li, Z.; Zhu, X. Chinese Journal of Polymer Science 2008,
2, 187.
Xia, Y.; Burke, N. A. D.; Stover, H. D. H. Macromolecules 2006, 6, 2275.
Furyk, S.; Zhang, Y. J.; Ortiz-Acosta, D.; Cremer, P. S.; Bergbreiter, D. E. Journal
of Polymer Science Part A-Polymer Chemistry 2006, 4, 1492.
Ye, J.; Narain, R. J Phys Chem B 2009, 3,.
Millard, P.; Mougin, N. C.; Boker, A.; Muller, A. H. E. Controlling the Fast ATRP of
N-Isopropylacrylamide in Water. In Controlled/Living Radical Polymerization:
Progress in ATRP; Matyjaszewski, K., Ed.; American Chemical Society: 2009; Vol.
1023, pp 127.
Masci, G.; Giacomelli, L.; Crescenzi, V. Macromolecular Rapid Communications
2004, 4,.
You, Y.; Zhou, Q.; Manickam, D. S.; Wan, L.; Mao, G.; Oupicky, D.
Macromolecules 2007, 24, 8617.
Convertine, A.; Lokitz, B.; Vasileva, Y.; Myrick, L.; Scales, C.; Lowe, A.;
McCormick, C. Macromolecules 2006, 5, 1724.
Ito, M.; Ishizone, T. Journal of Polymer Science Part A-Polymer Chemistry 2006,
16, 4832.
Aoshima, S.; Kanaoka, S. Wax Crystal Control: Nanocomposites, StimuliResponsive Polymers 2008, 169.
Schild, H. G. Progress in Polymer Science 1992, 2, 163.
Zhang, Y. Q.; Tanaka, T.; Shibayama, M. Nature 1992, 6400, 142.
Yoshida, R.; Uchida, K.; Kaneko, Y.; Sakai, K.; Kikuchi, A.; Sakurai, Y.; Okano, T.
Nature 1995, 6519, 240.
Durand, A.; Hourdet, D. Polymer 1999, 17, 4941.
Durand, A.; Hourdet, D. Macromolecular Chemistry and Physics 2000, 8, 858.
Petit, L.; Karakasyan, C.; Pantoustier, N.; Hourdet, D. Polymer 2007, 24, 7098.
Hourdet, D.; L'Alloret, F.; Audebert, R. Polymer 1997, 10, 2535.
Hourdet, D.; L'Alloret, F.; Audebert, R. Polymer 1994, 12, 2624.
Wang, Y.; Feng, Y.; Wang, B.; Lu, Z. J Appl Polym Sci 2010, 6, 3516.
Yu Wang; Zhiyong Lu; Yugui Han; Yujun Feng; Chongli Tang Adv. Mat. Res. 2011,
654.
Neugebauer, D.; Matyjaszewski, K. Macromolecules 2003, 8, 2598.
Zhang, Y.; Broekhuis, A. A.; Stuart, M. C. A.; Picchioni, F. J Appl Polym Sci 2008,
1, 262.
Egan, R.; Jones, M.; Lehninger, A. J. Biol. Chem. 1976, 14, 4442.
Wever, D. A. Z.; Raffa, P.; Picchioni, F.; Broekhuis, A. A. Macromolecules 2012,
10, 4040.
Sumerlin, B. S.; Neugebauer, D.; Matyjaszewski, K. Macromolecules 2005, 3, 702.
Tam, K.; Wu, X.; Pelton, R. Polymer 1992, 2, 436.
Cao, Z.; Liu, W.; Ye, G.; Zhao, X.; Lin, X.; Gao, P.; Yao, K. Macromolecular
Chemistry and Physics 2006, 24, 2329.
154
Chapter 6
47. Cao, Z. Q.; Liu, W. G.; Gao, P.; Yao, K. D.; Li, H. X.; Wang, G. C. Polymer 2005,
14, 5268.
48. Sun, S.; Wu, P.; Zhang, W.; Zhang, W.; Zhu, X. Soft Matter 2013, 6, 1807.
49. Andersson, M.; Maunu, S. L. Colloid Polym. Sci. 2006, 3, 293.
50. Dougherty, R. C. J. Chem. Phys. 1998, 17, 7372.
51. Zhang, J.; Pelton, R. Colloids and Surfaces A-Physicochemical and Engineering
Aspects 1999, 1-3, 111.
52. Zhang, J.; Pelton, R. Journal of Polymer Science Part A-Polymer Chemistry 1999,
13,.
53. Miao, Q.; Jin, Y.; Dong, Y.; Cao, Z.; Zhang, B. Polym. Int. 2010, 8, 1116.
54. Jalbert, C.; Koberstein, J. T.; Yilgor, I.; Gallagher, P.; Krukonis, V. Macromolecules
1993, 12, 3069.
155
Branched thermosensitive copolymers
This page intentionally left blank
156
Chapter 7
Chapter 7
Oil
recovery
using
branched
copolymers based on acrylamide
Abstract
The oil recovery from core material and a specifically designed flow cell
using
novel
branched
(comb
like)
polyacrylamides
(PAM)
has
been
investigated. The injectivity characteristics of the different branched PAMs
were evaluated by filtration tests and core flow experiments. The number of
arms of the branched PAM has little to no effect on the filterability and
permeation through a porous medium. The 13-arm branched PAM displayed
a higher residual resistance factor (RRF) in Berea sandstone compared to its
linear analogue and to commercial HPAM. In addition, the thickness of the
layer adsorbed at the rock-surface is higher for the branched PAM. Oil
trapped in dead–end pores is modeled using a 2D flow-cell and the effect of
the number of arms on the recovery of residual oil is evaluated. In brine
solutions, the branched PAMs perform equal or better than their linear
analogues in terms of the solution viscosity. The oil recovery of a branched
PAM with a similar molecular weight is 3 times as high as that for the
commercial polymer. The recovery efficiency, evaluated using low permeable
Berea as the porous medium, is significantly improved by using branched
PAM instead of linear ones (5.0 compared to 1.5 % of the OOIP). An
improvement is also observed when using high permeable Bentheim cores as
the porous medium (9.4% compared to 6.0% of the OOIP). The combination
of a higher RRF and a higher oil recovery (in the 2D flow-cell) might explain
the improved performance of the branched PAMs. The high thickening
capability and the low molecular weight of the branched PAMs makes them
suitable for application in enhanced oil recovery (EOR, especially for low
permeable reservoirs).
Based on: D.A.Z. Wever, F. Picchioni, A.A. Broekhuis. Comb-like
polyacrylamides as flooding agent in enhanced oil recovery. Industrial &
Engineering Chemistry Research, 2013, DOI: 10.1021/ie402526k.
157
Oil recovery using branched polyacrylamides
7.1. Introduction
In most oilfields in the world more than half of the original oil in place
(OOIP) remains after primary (aquifer drive, gas cap drive and gravity flow)
and secondary recovery (water injection) methods have been exhausted. 1,
2
Many different techniques have been developed to improve the recovery of
oil after secondary methods. All these so-called EOR methods aim at
increasing the percentage of the OOIP that can be recovered. One such
method involves the use of high molecular weight polymers dissolved in
water, i.e. polymer flooding.1-4 The mobility ratio ( ) between the dispersing
phase (water) and the dispersed phase (oil) is defined5 as:
(7.1)
where
is the water (brine) mobility,
permeability,
is the water viscosity,
oil viscosity. Ideally
displacing front.5,
6
is the oil mobility,
is the water
is the oil permeability, and is the
is equal (or lower) to unity leading to a stable
When
is higher than unity the displacing front will be
unstable, which will lead to viscous fingering.7,
8
This is often observed in
micro-models using only water as the displacing fluid.7 However, in practice,
the low viscosity of water compared to that of the oil leads to values much
higher than unity. The use of polymers leads to an improved mobility (i.e. a
reduction of ) in the oil reservoir by increasing the viscosity of the injected
fluid (water) and by reducing the formation’s permeability (adsorption of the
polymer chains on the surface of the rock).9, 10
The resistant factor (RF) is a measure for the ability of the polymer to
reduce the permeability of the reservoir through both the increase in solution
viscosity and the adsorption of polymeric chains on the surface of the rock.
In single-phase flow experiments and under the same conditions, i.e. equal
flow rates or equal pressure drops, the RF is defined3 as:
(7.2)
where
is the polymer solution permeability and
is the polymer solution
viscosity. To evaluate the permanent reduction in the permeability of the
formation due to the adsorption of polymeric chains the RRF is determined.
The RRF can be calculated through equation 7.3a if the injection rate of the
polymer and the brine solution (after the polymer injection) is kept constant. 3
158
Chapter 7
(7.3a)
where
is the mobility of the brine solution before the polymer solution is
injected and
is the mobility of the brine solution after the polymer
injection. The RRF can also be determined (equation 7.3b) using the
differential pressure during a brine flood before (
) and after a polymer
injection (
).11-13
(7.3b)
Partially hydrolyzed polyacrylamide (HPAM) is the most widely used polymer
to date for polymer flooding in EOR.2,
3
The limitations of HPAM include
among others, the low resistance towards the presence of salts. Salts will
lead to a significant reduction of the solution viscosity and even precipitation
upon interaction with divalent ions can occur. Another limitation of HPAM is
alkaline hydrolysis, which leads to an increased salt sensitivity. Also high
temperatures (T > 70 °C) and high shear forces are detrimental for the
performance of HPAM due to chemical and mechanical (respectively)
degradation of the chains.2,
14
The use of hydrophobically modified polymers
has been offered as a possible replacement for the HPAMs. 2,
15
The presence
of hydrophobic groups will lead to aggregation in semi-dilute water solutions,
thus increasing the solution viscosity.2, 15 Depending on the composition, the
increase in solution viscosity can be greater compared to that of HPAM.2
Actual core flood experiments have demonstrated that the RF and RRF are
both higher for the hydrophobically modified polymer compared to the values
obtained for HPAM.13 Studies have demonstrated that the adsorption of
hydrophobically modified polyacrylamides is significantly higher (i.e. by
development of a thicker polymeric layer on the surface of the rock)
compared to unmodified analogues.16,
17
If the adsorption is high enough
(and thus a high layer thickness), injectivity issues can arise, i.e. plugging,
where an exponential increase in the pressure of the reservoir is observed.
However, no injectivity problems could be detected for a hydrophobically
modified PAM in dilute18 and semi-dilute regimes. The formation of a gel
layer, on the injection side, was observed only below a critical threshold
permeability and/or pore throat radius.19 Nevertheless, with the development
of many new types of water soluble polymers2, it is crucial to investigate
whether these can be successfully injected in core samples if these are to be
applied in EOR.
159
Oil recovery using branched polyacrylamides
In principle, the use of polymers does lead to an increase in the rate of oil
recovery but the residual oil saturation (i.e. oil left behind after an extensive
water flood) is not affected.3 However, in the last decade, many papers20-27
have been published with evidence suggesting that the viscoelastic behavior
might aid in the recovery of residual oil and thus a reduction in the residual
oil saturation. When comparing a glycerin flood with a HPAM flood at equal
solution viscosity a significantly higher oil recovery out of a dead end was
observed for HPAM.23,
25, 26
However, the comparison between glycerin and
HPAM is not completely sound. Glycerin is a small molecule while HPAM is a
long
chain
polyelectrolyte.
Nevertheless,
the
experiments
appear
to
demonstrate that the residual oil saturation can be reduced using polymers
and that not only the viscosity of the displacing fluid is important in
recovering oil. To probe whether the viscoelasticity of the displacing fluid
affects the oil recovery, polymers with similar thickening capabilities but
different elastic response might aid in clarifying this issue.
Previously, we have demonstrated the successful synthesis of branched
PAM (Chapter 3) displaying an improved thickening capability (in the semidilute regime) compared to that of a linear analogue (Chapter 4). The
viscoelastic response of aqueous solutions (in the semi-dilute regime)
containing the different polymers depends on the molecular architecture of
the PAM, i.e. the number of branches (Chapter 4). In addition, the resistance
in terms of solution viscosity and viscoelasticity (Chapter 8) to the presence
of salt is better than that of HPAM given the uncharged character of the
branched PAM. The objective of this chapter is to investigate whether the
molecular architecture of PAM affects the injectivity of the polymer through
porous media and to determine the oil-recovery performance of these novel
materials.
7.2. Experimental section
7.2.1. Materials.
Sodium chloride (NaCl, ≥ 99%) was purchased from Sigma Aldrich.
IsoporeTM membrane (polycarbonate) hydrophilic filters (pore size = 1.2 and
3.0 m, and a diameter of 47 mm) were purchased from Merck Millipore.
Berea and Bentheim (D x L, 5 x 30 cm) sandstone cores were purchased
from Kocurek Industries. Berea and Rote Mainz sandstone plugs (D x L, 2.5 x
5 cm) were kindly supplied by Shell Global Solution International BV. The
crude oil is a medium oil (API gravity equals 27.8) and originates from the
Berkel oil field in the southwest of the Netherlands. The viscosity of the oil is
71 mPa.s at 20 °C. HPAM (Flopaam 3130 S, 25-35 mol% hydrolyzed) and
160
Chapter 7
linear polyacrylamide (FA920MPM and FA920) were kindly provided by SNF
Floeger (France). The linear and branched non-ionic water soluble polymers
used in the core floods, flow cell and filtration tests were previously
synthesized using atomic transfer radical polymerization (Chapters 2, 3 and
4). An overview of the different polymers used is given in Table 7.1.
Table 7.1: Properties of the different polymers
Architecture
Entry
Mn,th (g/mol)
AM (mol %)
AA (mol %)
Linear
LPAM21445a
LPAM35705a
1 524 432
2 540 789
100
100
0
0
Linear
CLPAM63310b
CLPAM84410b
Poly(AM31515-ran-AA13320)c
4 500 000
6 000 000
3 200 000
100
100
70
0
0
30
4-arm
PK30-g4-(PAM22660)d
1 613 401
100
0
8-arm
PK30-g8-(PAM24310)d
1 730 784
100
0
13-arm
PK30-g13-(PAM23775)d
PK30-g13-(PAM35275)d
PK30-g13-(PAM49190)d
1 692 550
2 510 092
3 499 094
100
100
100
0
0
0
17-arm
PK30-g17-(PAM22140)d
PK30-g17-(PAM35700)d
1 576 493
2 540 500
100
100
0
0
a. Linear polyacrylamide prepared through atomic transfer radical polymerization (ATRP)
b. Commercial linear polyacrylamide
c. Commercial linear HPAM
d. Branched polyacrylamide prepared through ATRP
The physical properties of the different cores were determined as follows. The
porosity (%) was determined using the buoyancy method. The bulk volume
and the grain volume of each sample were measured by immersing the dry
sample in mercury, and by immersing the 100% chloroform-saturated
sample in chloroform. For saturation with chloroform the samples were put in
a vacuum vessel. After evacuation, the samples were saturated and,
subsequently, put in a pressure vessel and pressurized up to 30 bars for at
least one hour to dissolve possible trapped air and saturate the micro-pores.
The pore volume (PV) was calculated based on the total volume of the core
and its corresponding porosity. To check for homogeneity of the cores, a XRay Computed Tomography (CT) scan of the core samples was recorded. All
the samples were dried before being analyzed. A Siemens Volume Zoom IV
spiral scanner located at the Shell’s Rock & Fluid Physics laboratory in
Rijswijk was used. The samples were all scanned at the same conditions
(120kV and 90mAs). Each set of scans comprises two orthogonal longitudinal
scans and one radial scan approximately at the center of the plug.
161
Oil recovery using branched polyacrylamides
The average brine permeability was determined by injecting the brine
solution (5000 ppm NaCl) at different flow rates (60, 120, 180, 240 and 300
mL/h) and measuring the pressure drop across the core sample. The average
brine permeability was then calculated using Darcy’s law (equation 4).28
(7.4)
where
= brine permeability (mD),
= the length of the core (cm),
2
the cross-sectional area of the core (cm ),
(mPa.s),
= the flow rate (cm3/s), and
=
= the viscosity of the fluid
= the pressure drop across the
core (atm). The average pore radius for brine flow can be determined using
the brine permeability and the porosity of the core.29 For this equation 7.5 is
used.
(7.5)
where
= the average pore radius for brine flow (m),
2
permeability (m ), and
= the brine
= the porosity (fraction). The properties of the
cores used in the different experiments are listed in Table 7.2.
Table 7.2: Physical properties of the sandstone cores
Core
Berea
1a
Berea
1b
Berea
1c
Berea
2a
Berea
2b
Bentheim
1a
Bentheim
1b
5
2.5
4.91
22.4
5.5
371
3.62
-
5
2.5
4.91
22.4
5.5
246
2.94
-
5
2.5
4.91
22.4
5.5
528
4.31
-
30
5
19.63
19.5
114.9
75
1.74
72.24
30
5
19.63
19.5
114.9
96
1.97
74.85
30
5
19.63
24.0
141.4
2126
8.36
83.39
30
5
19.63
24.0
141.4
2371
8.83
89.05
Property
Length (cm)
Diameter (cm)
Cross-sectional area (cm2)
Porosity (%)
Pore volume (PV, mL)
Brine permeability (mD)
Average pore radius (m)
Oil saturation (%)
The pore throat size distribution of the Berea 1 cores was determined using
the mercury porosimetry technique.30 This technique uses mercury under
pressure to penetrate the pores. The liquid can penetrate smaller pores when
the pressure is increased.
162
Chapter 7
The relation between the pore-throat size and pressure is defined31 by the
Washburn’s equation:
(7.6)
where
= pore radius (m),
= mercury surface tension (mN/m2),
=
contact angle mercury with rock surface. The pore-throat size is inversely
proportional to the pressure applied. The mercury porosimetry method uses a
range of pressures in order to obtain a pore-throat size distribution. The
pressure is step-wise increased and the liquid intrusion (amount) in relation
to the total liquid intrusion represents the fraction of pores with that
particular pore-throat size.
7.2.2. Polymer injectivity experiments
Filtration tests. Filtration tests were performed to evaluate the
permeation of the different polymer solutions through small pores. The
experimental set-up used for the filtration tests is schematically presented in
Figure 7.1 A.
Figure 7.1: Schematic presentation of the experimental set-up for (A) the filtration
tests and (B) the core flood experiments
163
Oil recovery using branched polyacrylamides
The set-up is fitted first with a MilliporeTM polycarbonate filter and
subsequently filled with 250 mL of the polymer solution through the top
opening. All the valves are closed and afterwards the cylinder is pressurized
to 2 bars with compressed air. The bottom valve is open and the weight of
the effluent is measured (in a beaker) as a function of time using a scale until
more than 200 g of solution has passed (the 2 bar pressure is kept constant
throughout the experiment). The effective diameter of the filter is slightly
lower due to the rubber ring that ensures an air tight seal (deff = 41 mm). In
order to evaluate the ease of passage through the filters the filtration ratio
( ) is calculated using equation 7.5.
(7.7)
where t200-t180 = throughput time of 20 g of the solution at the end of the
test and t40-t20 is the throughput time of 20 g at the start of the experiment.
Core floods. The injectivity of the polymers was evaluated by flooding
sandstone cores (2.5 x 5 cm) with the polymer. First the core was fixed in a
core holder and flooded with carbon dioxide (CO2). Afterwards brine (5000
ppm NaCl) was injected at a low rate (linear velocity < 1 foot/day) for at
least 12 hours to be certain that all the remaining CO 2 had dissolved and no
bubbles were present anymore. Afterwards the brine permeability was
determined according to literature.28 The pressure drop was measured with
GS4200-USB digital pressure transducers (ESI Technology Inc.) linked to a
software program. Subsequently a polymer flood was conducted where at
least 20 pore volumes (PV) of the polymer solution was injected. The linear
velocity of the polymer floods was set at 1 foot per day. The pressure was
recorded as a function of time during the polymer flood. A schematic
overview of the experimental set-up is given in Figure 7.1B.
7.2.3. Oil recovery
Flow-cell experiments. A schematic presentation of the flow-cell (with
the dimensions) is given in Figure 2. The flow cell has been adapted from the
original ones presented in literature34 to resemble dead-end pores (Figure
1.4) that are present in oil reservoirs. The bottom part of the flow-cell is
made out of aluminum while the cover is glass. The depth of the chamber
(designated as blue in Figure 2) is set at 0.5 mm. The chamber is first filled
with oil and afterwards flooded with brine or polymer solutions. The linear
velocity was set at 1 foot per day (0.3048 m/day) and is calculated based on
the total volume of the blue areas (Figure 1.4). Each flood (either brine or
polymer) was continued for at least 24 hours.
164
Chapter 7
Figure 7.2: Schematic presentation of the flow-cell (top view)
The oil recovery out of the different cells was visually determined by taking
high definition pictures before (if a water flood preceded a polymer flood) and
after the floods. Analysis (pixel count) of the image using Adobe allows the
calculation of the amount of oil left behind in the flow-cell.
Core flow experiments. The recovery of oil from sandstone cores was
evaluated using 5 x 30 cm sandstone cores. The cores were placed in a core
holder and saturated with CO2. Afterwards brine (30000 ppm NaCl) was
injected at a low linear velocity (i.e.
= 2 feet/day) for at least a 3 hours in
order for the CO2 to dissolve and ascertain a core free of gas bubbles. The
brine permeability was determined by measuring the pressure drop across
the core and the flow rate using Darcy’s law.28 The core was then filled with
oil to connate water saturation and subsequently a water flood was
performed (at least 5 PV).
7.2.4. Characterization
Rheological
properties.
The
aqueous
polymeric
solutions
were
prepared by swelling the polymers in water for one day and followed by
gently stirring the solution for another day.
Viscometric measurements were performed on a HAAKE Mars III
(ThermoScientific) rheometer, equipped with a cone-and-plate geometry
(diameter 60 mm, angle 2°). Flow curves were measured by increasing the
shear stress by regular steps and waiting for equilibrium at each step. The
shear rate ( ) was varied between 0.1 – 1750 s-1. Dynamic measurements
were performed with frequencies ranging between 0.04 – 100 rad/s (i.e.,
6.37·10-3 – 15.92 Hz). It must be noted that all the dynamic measurements
were preceded by an oscillation stress sweep to identify the linear
viscoelastic response of each sample and to ensure that the dynamic
measurements were conducted in the linear response region of the samples.
165
Oil recovery using branched polyacrylamides
7.3. Results and discussion
7.3.1. Polymer injectivity
Filtration tests. The permeation characteristics of the polymers were
evaluated by the filtration test (and flow through small sandstone cores). The
concentration of the polymers for the filtration test was adjusted to give the
same solution viscosity at
= 10 s-1 (i.e.  ≈ 15 mPa.s), since this value is
close to the average shear rate encountered in porous media. 33,
34
The
rheological properties of the polymers used are displayed in Figure 7.3.
As the degree of branching increases it is expected (based on the results
in Chapter 4) that the polymer concentration required in reaching the
solution viscosity of 15 mPa.s will decrease. As can be observed in Figure 7.3
A, the polymer concentration required to reach the set viscosity is lower for
the branched PAMs in comparison to the linear PAMs. The elastic response
(Figure 7.3 B) of the polymeric solutions is more pronounced for the higher
molecular weight polymers and the polymers with a higher degree of
branching.
20
LPAM35705,
PK30-g13-(PAM35275),
[p] = 3100 ppm
[p] = 3000 ppm
PK30-g17-(PAM35700),
[p] = 2900 ppm
CLPAM63310,
CLPAM84410,
[p] = 4900 ppm
[p] = 3250 ppm
15
G'G" (Pa)
Solution viscosity (mPa.s)
25
10
10
1
10
0
90
10
-1
10
-2
10
-3
80
LPAM35705
PK30-g13-(PAM35275)
PK30-g17-(PAM35700)
CLPAM63310
CLPAM84410
70
60
50
G"
40
30

G'
A
5
1
10
100
1000
10
-4
10
-5
B
-1
Shear rate (s )
10
0
0,1
1
10
Frequency (rad/s)
Figure 7.3: (A) Viscosity functions of the different polymeric solutions and (B) G’, G”
and the phase angle () as a function of the frequency
As mentioned before, the injectivity of the polymer solutions was evaluated
first with a filter. The effluent weight against time curves of the different
polymer solutions are displayed in Figure 7.4.
The filtration ratios of all the polymer solutions tested through a 3.0 m
filter were close to unity indicating good injectivity. However, decreasing the
average pore size of the filter to 1.2m led to an increase in the filtration
ratios of all the polymer solutions, with two solutions that did not pass
through the filter. The clarity of the two latter solutions was inferior to the
166
20
Phase angle
30
Chapter 7
former three (pictures not shown for brevity). Debris in the solution can plug
the filter thus hampering the flow of the solution through the filter, and this
might explain the difficulty of passage through the filter for these solutions.
250
250
B
A
200
Weight (g)
Weight (g)
200
150
100
50
0
100
CLPAM63310
FR = 1,09
PK30-g13-(PAM35275)
FR = 1,14
PK30-g17-(PAM35700)
FR = 1,07
LPAM35705
FR = 1,31
CLPAM84410
FR = 1,37
200
300
150
100
50
FR = 1,57
CLPAM63310
FR = 1,33
PK30-g17-(PAM35700)
FR = 2,07
CLPAM84410
FR = did not pass
PK30-g13-(PAM35275)
0
400
LPAM35705
0
1000
2000
Time (s)
FR = did not pass
3000
4000
Time (s)
Figure 7.4: Weight against time curve for the different polymer solutions with the
respective filtration ratios (as computed using equation 7.7) through a filter with an
average pore size of (A) 3.0 m and (B) 1.2m
Injectivity in sandstone cores. In addition to the filtration tests, flow
through small cores was performed to evaluate the injectivity of the different
polymer solutions through low permeable porous media. The rheological
properties of the polymer solutions used are displayed in Figure 7.5.
A
Poly(AM31515-ran-AA13320)
PK30-g13-(PAM35275)
10
3
10
2
10
1
10
0
LPAM35705
10
1
0,1
1
10
100
-1
Shear rate (s )
1000
10
-1
10
-2
10
-3
10
-4
10
-5
90
B
80
70
60
Poly(AM31515-ran-AA13320)
PK30-g13-(PAM35275)
50
LPAM35705
40
30
20
10
0
0,1
1
10
Frequency (rad/s)
Figure 7.5: (A) Viscosity functions for the different polymers used in the small cores
and (B) G’, G” and the phase angle as a function of the frequency for the polymer
solutions
167
Phase angle
2
G'G" (Pa)
Solution viscosity (mPa.s)
10
Oil recovery using branched polyacrylamides
The concentration of the solutions varied depending on the polymer chemical
and molecular structure (Figure 7.5 A). The elasticity of the polymer
solutions are quite similar (Figure 7.5 B). The physical properties of the cores
(2.5 x 5 cm) used are listed in Table 7.2. The homogeneity of the used cores
was confirmed by CT-scans (results not show for brevity). The brine
composition was set at 5000 ppm NaCl. The polymers included in the
evaluation of the injectivity were entries LPAM35705, PK30-g13-(PAM35275),
and Poly(AM31515-ran-AA13320).
The RF for the different polymers increased until a constant value (not
shown for brevity). For all three polymer solutions, the pressure stabilized
within 5 PV indicating good permeation through the porous media. The RRF
was computed through equation 7.3b (Table 7.3). The average absorbed
polymer layer thickness ( ) can be determined using the RRF and equation
7.6.29
(7.8)
where
= absorbed layer thickness (m),
= the average pore radius (m),
= the residual resistant factor. The thickness of the absorbed layer
affects the permeation of the polymer solution, such that the flow is diverted
from high permeable thief zones towards low permeable un-swept areas.3
However, the thickness of the layer cannot indefinitely increase since this can
lead
to
injection
problems
(i.e.
formation
damage
due
to
polymer
retention/adsoprtion13).
The branched PAM leads to a significantly higher RRF compared to that
of its linear analogue and the commercial HPAM (Table 7.3). The differences
can be attributed to the molecular architecture and chemical structure. The
lower absorbed polymer layer thickness of the commercial HPAM (the
presence of charges reduces the extent of adsorption onto a surface 3) leads
to a lower RRF.
Table 7.3: Results of the injectivity experiments
Architecture
Entry
Core sample
[NaCl], ppm
[p], ppm
RRF
e (m)
Linear
LPAM35705
Berea 1b
5000
3200
7
1.13
Linear
Poly(AM31515-co-AA13320)
Berea 1c
5000
2750
2
0.69
13-arm
PK30-g13-(PAM35275)
Berea 1a
5000
3000
23
1.97
168
Chapter 7
For the branched PAM, we envisaged that the interaction between two coils
to be stronger for the branched PAM as compared to the one for a linear
analogue due to the presence of the arms. This leads in turn to a higher
absorbed layer thickness and thus a higher RRF. The absorbed polymer layer
(Table 7.3) varies in thickness from as low as one fortieth (entry
Poly(AM31515-co-AA13320)) of the average pore throat radius (Figure 7.6)
up to as much as one fifteenth (entry PK30-g13-(PAM35275).
0,3
1,0
0,2
0,6
Pore throat size distribution
Cumulative
0,4
0,1
Cumulative fraction
Fraction of total pores
0,8
0,2
0,0
0,0
0,01
0,1
1
10
100
Pore throat diameter (m)
Figure 7.6: Pore throat size distribution of the Berea 1 cores
Although the average pore throat size is a magnitude larger than the
thickness of the absorbed polymer layer, the distribution of the pore throat
size (Figure 7.6) demonstrates that sizes close to the layer thickness are also
present.
7.3.2. Oil recovery
Flow-cell. The recovery of oil out of dead ends was investigated using a
2D flow-cell. The concentration of the polymer was adjusted so that the
viscosity of the solution matched that of the crude oil. The rheological
properties of the polymer solutions used in the comparison are given in
Figure 7.7.
169
Oil recovery using branched polyacrylamides
[p] = 8500 ppm
[p] = 8500 ppm
PK30-g13-(PAM23775),
[p] = 5250 ppm
PK30-g13-(PAM49190),
[p] = 4000 ppm
PK30-g17-(PAM22140),
[p] = 5250 ppm
G'G" (Pa)
Solution viscosity (mPa.s)
LPAM21445,
PK30-g4-(PAM22660),
10
2
10
1
10
0
10
-1
10
-2
10
-3
10
-4
10
-5
90
B
LPAM21445
PK30-g17-(PAM22140)
PK30-g13-(PAM23775)
PK30-g13-(PAM49190)
PK30-g4-(PAM22660)
Poly(AM31515-co-AA13320)
80
70
60
50
G"

G'
10
30
20
10
Poly(AM31515-co-AA13320), [p] = 8700 ppm
10
1
40
Phase angle
A
100
100
1000
0
0,1
-1
Shear rate (, s )
1
10
Frequency (rad/s)
Figure 7.7: (A) Viscosity functions for the different polymers used in the flow-cell and
(B) G’, G” and the phase angle as a function of the frequency for the polymer solutions
The capability of the polymer solution to recover residual oil out of dead ends
is evaluated based on the results for chambers 2 and 3 (Figure 7.2). The
results are presented in Figure 7.8.
[A] Water
[E] PK30-g13-(PAM49190)
[B] LPAM21445
[C] PK30-g4-(PAM22660)
[F] PK30-g17-(PAM22140)
[D] PK30-g13-(PAM23775)
[G] Poly(AM31515-co-AA13320)
Figure 7.8: Oil recovery out of chambers 2 and 3 of the 2D flow-cell using different
polymer solutions (at equal viscosity, i.e.  ≈ 71 mPa.s at = 10 s-1)
The oil recoveries were determined by measuring the black area after a
particular flood and comparing that to the area after the water flood (Table
7.4).
170
Chapter 7
30
30
25
25
20
20
15
15
10
8700
8500
10
8500
5250
5
5250
5
4000
a
a
PA
M
1,
5
3,
5M
D
M
D
a
13
-a
rm
PA
M
17
-a
rm
13
-a
rm
PA
M
1,
5M
D
1,
5M
D
a
PA
M
1,
5M
D
Li
ne
ar
4ar
m
PA
M
M
PA
a
0
3,
5M
D
a
0
H
Oil recovery (%)
3
Polymer concentration ( 10 ppm)
Polymer concentration
Oil recovery
Figure 7.9: Oil recovery out of dead-ends in the flow-cell
The recovery of oil out of the dead ends depends on the polymer used.
Compared to water the polymer solutions labeled [B], [C], [D] and [G] all
improve the oil recovery by roughly 5%. The efficiency (i.e. the oil recovery)
of these polymer solutions are roughly the same, but the polymer
concentration used (to match the viscosity of the oil) is different (Figure 7.9).
The concentration of the highly (N ≥ 13) branched polymers (with a Mn that
is two times lower than the commercial polymer) is much lower than the
commercial one. Nevertheless, the performance (in terms of oil recovery) is
similar. The comparison between the branched polymer ([E] PK30-g13(PAM49190))
and
the
commercial
polymer
([G]
Poly(AM31515-co-
AA13320)), both having a similar molecular weight, demonstrates the
effectiveness of the branched polymers in recovery residual oil. The former
one displays an oil recovery at least twice as high compared to the
commercial polymer.
When looking at the effect of the number of branches (i.e. [C] PK30-g4(PAM22660), [D] PK30-g13-(PAM35275) and [F] PK30-g17-(PAM22140)), the
results suggest that increasing the number of branches improves the oil
recovery. The molecular weight also plays a role in the amount of oil that is
recovery, increasing the molecular weight of the branches leads to a higher
oil
recovery
([D]
PK30-g13-(PAM35275)
compared
to
[E]
PK30-g13-
171
Oil recovery using branched polyacrylamides
(PAM49190)). Another parameter, evidenced by the results, that seems to
affect the oil recovery out of the dead ends is the viscoelasticity. The elastic
response of polymer solutions containing PK30-g13-(PAM35275), PK30-g13(PAM49190),
and
PK30-g17-(PAM22140)
is
slightly
more
pronounced
compared to that of LPAM21445 and PK30-g4-(PAM22660). The oil recovery
of the former ones ([D], [E], and [F]) is higher than for the latter ones ([B]
and [C]), thus supporting the conclusion that the elasticity of the polymer
solution can aid in recovering residual oil.
Low permeable cores. The physical properties of the Berea sandstone
cores (5 x 30 cm) are listed in Table 7.2. The tertiary oil recovery out of the
low permeable cores was investigated using a commercial polymer (entry
poly(AM31515-ran-AA13320))
and
a
branched
PAM
(entry
PK30-g17-
(PAM22140). Both polymer floods were preceded by a waterflood until the
water cut was below 1%. The concentration of the two polymer solutions was
adjusted to give a solution viscosity (at
= 10 s-1, [p] = 5250 ppm (entry
PK30-g17-(PAM22140) and [p] = 8700 ppm (entry poly(AM31515-ranAA13320)) that is equal to that of the crude oil. The rheological properties of
the two polymeric solutions are presented in Figure 7.10. The thickening
capability of the branched PAM is significantly higher than that of the
commercial HPAM in salt solution (Chapter 8), leading to a lower polymer
concentration required to match the oil viscosity.
The extent of pseudoplastic behavior is slightly more pronounced for the
branched PAM compared to the linear HPAM (Figure 7.10 A). Similar behavior
is observed when comparing a linear and a branched PAM of similar total
molecular weight (Chapter 4). The solution viscosity as a function of
temperature differs for the two polymeric solutions, with the branched PAM
displaying
a
higher
resistance
to
temperature
(Figure
7.10
B).
For
comparison purposes, the temperature resistance of a linear PAM with similar
total molecular weight (entry LPAM21445) was also evaluated. The decrease
in the solution viscosity as a function of temperature is similar to that of the
commercial HPAM. This might make the branched polymers better suited for
application in oil reservoirs with higher temperatures (T > 50 °C).
The branched PAM displays a slight more pronounced elastic response
compared to the commercial HPAM (Figure 7.10 C). This might be beneficial
for the recovery of residual oil, since comparisons between a glycerin and a
HPAM flood suggested that the elasticity of the displacing fluid aids in the
recovery of residual oil.20, 21, 23-27
172
Chapter 7
100
A
B
Viscosity (mPa.s)
60
Viscosity (mPa.s)
10
80
2
PK30-g17-(PAM22140)
Poly(AM31515-ran-AA13320)
40
20
PK30-g17-(PAM22140)
Poly(AM31515-ran-AA13320)
LPAM21445
1
10
100
1000
0
20
30
3
10
2
10
1
10
0
10
-1
10
-2
10
-3
10
-4
10
-5
10
-6
50
60
Temperature ( C)
Shear rate (s )
10
40
o
-1
90
C
80
70
60
50
G"
40
G'
30
20
Poly(AM31515-ran-AA13320)
PK30-g17-(PAM22140)
Phase angle
0,1
G'G" (Pa)
10
1
10
0
0,1
1
10
Frequency (rad/s)
Figure 7.10: (A) Viscosity functions for the two polymer solutions used in the coreflow, (B) viscosity as a function of temperature (at
= 10 s-1), and (C) G’, G” and the
phase angle as a function of the frequency for the polymer solutions
The results of the oil recovery out of the low permeable Berea cores
demonstrate that only 1.5% more oil is recovered with the commercial HPAM
polymer (Figure 7.11). When using a solution containing the branched PAM
as the displacing fluid a total oil recovery of 5.0% is realized. The higher oil
recovery is remarkable given that the viscosity of the displacing fluid is equal
for both solutions (Figure 7.10 A). In addition, the polymer concentration of
the branched PAM is also significantly lower (i.e. approximately 40% lower)
than that of the commercial polymer. The improved oil recovery can be
attributed to the slightly more pronounced elastic response (Figure 7.10 C)
and the higher RRF (Table 7.3) of the former.
173
70
Oil recovery using branched polyacrylamides
48
Start polymer injection
44
Oil recovery (%)
40
Extrapolated waterflood
36
32
Waterflood
PK30-g17-(PAM22140)
28
Waterflood
Poly(AM31515-ran-AA13320)
24
0
0
1
2
3
4
Cumulative displacing fluid injected (PV)
Figure 7.11: Oil recovery from low permeable Berea sandstone cores
High permeable cores. The same two polymer solutions were used to
evaluate the oil recovery out of high permeable cores. The physical
properties of the Bentheim sandstone cores (5 x 30 cm) are listed in Table
7.2. The use of the linear commercial HPAM (entry poly(AM31515-ranAA13320)) leads to an increase of 6.0% in the oil recovery (Figure 7.12).
60
Start polymer injection
55
Oil recovery (%)
50
45
Extrapolated waterflood
40
35
Waterflood
PK30-g17-(PAM22140)
30
Water flood
Poly(AM31515-ran-AA13320)
25
20
0
0
1
2
3
4
Cumulative displacing fluid injected (PV)
Figure 7.12: Oil recovery from high permeable Bentheim sandstone cores
174
Chapter 7
The oil recovery using the branched PAM (entry PK30-g17-(PAM22140)
reaches a maximum of 9.4%. The oil recovery of the branched PAM is more
than 50% higher than that of the commercial HPAM, and the concentration
required to achieve this is significantly lower ([p] ≈ 40% lower). This
demonstrates the relevant potential of branched PAMs in EOR applications.
An overview of the results of the oil recovery from the different
sandstone cores is given in Table 7.4.
Table 7.4: Oil recovery from sandstone cores
Entry
Core sample [p], ppm
Poly(AM31515-co-AA13320)
PK30-g17-(PAM22140)
Berea 2b
Berea 2a
Poly(AM31515-co-AA13320) Bentheim 1b
PK30-g17-(PAM22140)
Bentheim 1a
Oil sat.a
Oil sat.,
Oil sat.,
Oil rec.d
∆P, PFe
(% of PV) waterb (%) polymerc (%) (% of OOIP) (bars)
8700
5250
74.85
72.24
44.02
42.27
42.91
38.88
1.48
4.97
1.9
2.6
8700
5250
89.05
83.39
46.02
41.56
40.66
33.70
6.01
9.43
0.1
0.1
a. The oil saturation in beginning (i.e. the OOIP)
b. The oil saturation after the waterflood
c. The oil saturation after the polymerflood
d. The enhanced oil recovery defined as the volume of oil produced by the polymer flood divided by
the total volume of oil originally in place (as percentage)
e. The maximum pressure drop during the polymerflood
The higher oil recovery in the low permeable Berea cores is probably due to
the larger pressure drop (Table 4) caused by the higher hydrodynamic
polymer layer thickness of the branched PAM compared to the commercial
one. In the high permeable cores, the thickness of the polymer layer doesn’t
affect the pressure drop (Table 4). The incremental oil recoveries over the
waterflood by the polymers are closer to each other. The increased value for
the branched PAM might be caused by the slightly higher elasticity of the
polymeric solution, similar to the results in the 2D flow-cell.
7.4. Conclusion
The oil recovery using branched (co)polymers based on acrylamide was
evaluated through core flow experiments and a 2D flow-cell. Experiments
aimed at investigating the injectivity characteristics of the branched AM
based polymers demonstrated that, compared to linear analogues and a
commercial polymer, similar behavior is observed. The branched PAM
displayed a higher RRF and adsorbed polymer layer thickness compared to
that of its linear analogue and the commercial polymer, which can increase
oil recovery due to an improvement in the sweep efficiency. This is attributed
175
Oil recovery using branched polyacrylamides
to the molecular architecture (stronger interaction between two coils due to
the presence of branches) and chemical structure (higher adsorption due to
the absence of charges).
In brine solutions, the branched PAM performs equal or better (at lower
polymer concentration) than their linear analogues in recovering residual oil
simulated by a 2D flow-cell. The incremental oil recovery over a waterflood of
a branched PAM with a similar molecular weight is 3 times as high as that for
the commercial polymer. The oil recovery in low permeable Berea cores is
significantly improved by using branched PAM instead of linear ones (5.0
compared to 1.5 % of the OOIP, i.e. approximately 3 times higher). The oil
recovery in high permeable Bentheim cores is also significantly improved
when using branched PAMs (9.4% compared to 6.0% of the OOIP). The
combination of a higher RRF and a higher oil recovery (in the 2D flow-cell)
might explain the improved performance of the branched PAMs. The high
thickening efficiency of the branched PAMs coupled with their low molecular
weight makes these polymers highly interesting for application in EOR
(especially for low permeable reservoirs).
7.5. Acknowledgements
This work is part of the Research Program of the Dutch Polymer Institute
DPI, Eindhoven, The Netherlands, project #716.
7.6. References
1. Thomas, S. Oil Gas Sci. Technol. 2008, 1, 9.
2. Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Prog. Polym. Sci. 2011, 1558.
3. Lake, L. W. Enhanced Oil Recovery; Prentice-Hall Inc.: Englewood Cliffs, NJ, 1989;
Vol. 1, pp 550.
4. Donaldson, E. C.; Chilingarian, G. V.; Yen, T. F. Enhanced Oil Recovery II, processes
and operations; Elsevier: Amsterdam, The Netherlands, 1989; Vol. 2, pp 604.
5. Stanislav, J. F. Rheol. Acta 1982, 4-5, 564.
6. Pope, G. A.; Bavière, M. Reduction of Capillary Forces by Surfactants. In Basic
Concepts in Enhanced Oil Recovery Processes. Critical reports on Applied
Chemistry; Bavière, M., Ed.; Springer: 1991; Vol. 33, pp 89-122.
7. Buchgraber, M.; Clemens, T.; Castanier, L. M.; Kovscek, A. R. SPE 2009, SPE122400.
8. Homsy, G. M. Annu. Rev. Fluid Mech. 1987, 271.
9. Gogarty, W. B.; Tosch, W. C. J. Pet. Technol. 1968, 12, 1407.
10. Hirasaki, G. J.; Pope, G. A. SPE 1974, SPE-4026-PA.
11. Ali, L.; Barrufet, M. A. Journal of Petroleum Science and Engineering 2001, 1, 1.
12. Melo, M. A.; Silva, I. P. G.; Godoy, G. M. R.; Sanmartim, A. N. SPE 2002, SPE75194-MS.
13. Maia, A. M. S.; Borsali, R.; Balaban, R. C. Mat. Sci. Eng. C-Bio S. 2009, 2, 505.
14. Thomas, A.; Gaillard, N.; Favero, C. Oil Gas Sci Technol 2012, 6, 887.
15. Taylor, K. C.; Nasr-El-Din, H. A. J. Petrol. Sci. Eng. 1998, 3-4, 265.
176
Chapter 7
16. Argillier, J. F.; Audibert, A.; Lecourtier, J.; Moan, M.; Rousseau, L. Colloid Surface
A 1996, 3, 247.
17. Volpert, E.; Selb, J.; Candau, F.; Green, N.; Argillier, J. F.; Audibert, A. Langmuir
1998, 7, 1870.
18. Dupuis, G.; Rousseau, D.; Tabary, R.; Argillier, J. F.; Grassl, B. Oil Gas Sci Technol
2012, 6, 903.
19. Dupuis, G.; Rousseau, D.; Tabary, R.; Grassl, B. Spe Journal 2012, 4, 1196.
20. Zhang, L.; Yue, X.; Guo, F. Pet. Sci. 2008, 1, 56.
21. Zhang, L.; Yue, X. J. Cent. South Univ. T. 2008, 84.
22. Yin, H.; Wang, D.; Zhong, H. SPE 2006, SPE-101950-MS.
23. Xia, H.; Wang, D.; Wang, G.; Wu, J. Petrol. Sci. Technol. 2008, 4, 398.
24. Xia, H.; Ju, Y.; Kong, F.; Wu, J. SPE 2004, SPE-88456-MS.
25. Wang, D.; Xia, H.; Liu, Z.; Anda, Q.; Yang, Q. SPE 2001, SPE-68723-MS.
26. Wang, D.; Cheng, J.; Yang, Q.; Gong, W.; Li, Q.; Chen, F. SPE 2000, SPE-63227MS.
27. Hou, J. R.; Liu, Z. C.; Zhang, S. F.; Yue, X.; Yang, J. Z. Journal of Petroleum
Science and Engineering 2005, 3-4, 219.
28. Takeuchi, S.; Nakashima, S.; Tomiya, A. J. Volcanol. Geotherm. Res. 2008, 2,
329.
29. Zaitoun, A.; Kohler, N. SPE 1988, SPE-18085.
30. Kaneko, K. J. Membr. Sci. 1994, 1-2, 59.
31. Washburn, E. W. Phys. Rev. 1921, 3, 273.
32. Jones, D. M.; Walters, K. Rheologica Acta 1989, 6, 482.
33. Melo, M. A.; Holleben, C. R. C.; Silva, I. P. G.; Correia, A. B.; Silva, G. A.; Rosa, A.
J.; Lins, A. G.; Lima, J. C. SPE 2005, SPE-94898-MS.
34. Niu, Y.; Ouyang, J.; Zhu, Z.; Wang, G.; Sun, G.; Shi, L. SPE 2001, SPE-65378MS.
177
Oil recovery using branched polyacrylamides
This page intentionally left blank
178
Chapter 8
Chapter 8
Towards new polymers for enhanced
oil recovery
Abstract
The progress booked in this project is discussed in terms of the problems
that have been overcome. Control of the acrylamide (AM) polymerization was
accomplished and this allows the preparation of polyacrylamide (PAM) with
variations in its chemical structure and molecular architecture. Branching in
PAM is presented as a new tool to significantly improve the solution viscosity
without changing the chemical structure of the polymer.
The non-ionic nature of the branched PAM renders it insensitive to the
presence of salt. More specifically, the solution viscosity and the elastic
response are not affected by the presence of salt. The rheological properties
of aqueous solutions are maintained when increasing the amount of salt
whereas those of the commercial HPAM are dependent on the salt
concentration.
The AM units in HPAM are also highly susceptible to alkaline hydrolysis at
high temperatures (T > 50 °C). N,N-dimethyl acrylamide (DMA), as a
hydrolysis resistant moiety, is a viable option especially in combination with
the increase thickening efficiency through branching. The oil recovery
efficiency of the branched PDMA polymers is similar to branched PAMs, as
evaluated using a 2D flow-cell that simulates residual oil trapped in deadends. This makes them interesting polymers for application in EOR where
alkaline agents are used to in situ generate surfactants.
The obtained results constitute a breakthrough in the general field of
chemical product design for EOR. However, a further refinement of the used
criteria can be envisaged on the basis of the changing legislation (e.g.
currently in Norway). This implies the use of biologically degradable polymers
for underground injection as fitting a general “sustainability” concept for this
application. In this context, the potential of different biopolymers for EOR
where synthetic polymers cannot be used is briefly discussed. All the
currently investigated biopolymers are based on polysaccharides, albeit with
slightly
different
molecular
structures.
The
thickening
capability
and
179
Outlook on the application of branched (co)polymers in EOR
resistance towards salt and temperature of the biopolymers are a strong
function of the molecular weight and their ability to form helical structures in
aqueous solutions. The elasticity of the biopolymer solutions are a strong
function of the molecular weight of the biopolymer, with the higher molecular
weight ones displaying a more pronounced elastic response.
180
Chapter 8
8.1. Introduction
Water soluble polymers in EOR have been investigated and applied in
different techniques, i.e. Polymer Flooding, Alkaline Surfactant Polymer (ASP)
Flooding, and Surfactant Polymer (SP) Flooding.1 When water soluble
polymers are applied in any of the former techniques, several different
aspects have to be considered. In the context of this dissertation, i.e. EOR,
the important ones (amongst others) are: the solution viscosity as a function
of the polymer concentration, the dependency of the solution viscosity on the
presence of salt (mono- and divalent ions), and the resistance (in terms of
solution viscosity) towards higher temperatures (T > 50 °C).
The dependency of the solution viscosity on the polymer concentration is
in general well known for homopolymers. The rheological properties depend
on the concentration regime in which the solution is. In general four different
concentration regimes can be distinguished (Figure 8.1).
Figure 8.1: Concentration regimes in polymeric solutions
At low polymer concentration, the polymer coils do not overlap and the
solution rheology can be described2 using Einstein’s equation3 for dilute
solutions of spherical particles:
(8.1)
where
= viscosity of the solution,
= viscosity of the solvent and
= the
volume fraction of the particles.
At these low concentrations, the solution viscosity depends solely on the
volume fraction of the spheres, i.e. the polymer coils, and not on their size.
The polymer concentration at which the polymer coils start to touch (overlap)
each other is defined as the first critical overlap concentration (C*).
According to several estimations, the values for C* generally range between
181
Outlook on the application of branched (co)polymers in EOR
0.1 up to 5.0 wt.%.3 When the polymer concentration is increased beyond
C*, overlapping of the polymer coils becomes more prominent and the
solution viscosity increases significantly. The concentration regime starting
from the C* up to a second critical concentration (C**, above which a gel is
formed) is defined as the semi-dilute regime. The viscosity of a polymeric
solution in this regime is governed by the relaxation, i.e. reptation 4, of the
entanglements in response to disturbances caused by deformation forces
(stresses). At higher concentrations (c > C**, polymer concentration [p] ≥
50 wt.%3), the rheological properties resemble those observed for polymers
in the melt state.3,
5, 6
However, when making allowances for the desired
application (i.e. EOR), the increase in the polymer concentration can lead to
significant problems in the reservoir. The propagation of polymer coils
through narrow pore throats “presses” the coils closer to each other. If the
coils are large enough, bridging (Figure 8.2) can arise which leads to
blockage of pores.7-10
Figure 8.2: Bridging in porous media10
An increase in the polymer concentration is also detrimental for the
economics of a flooding project given the scale of such projects (e.g. for the
Marmul field pilot-project ~25 ton/day of dry polymer has been used11, 12).
The molecular weight also affects the rheological properties of water
soluble polymers. In general, the solution viscosity increases with the
molecular weight of the polymer, and the dependency can be described using
the reptation model of de Gennes.4 At equal polymer concentration, an
182
Chapter 8
increase in the molecular weight will lead to an increase in the overlap
density, which in turn leads to longer relaxation times (synonymous of higher
viscosities). Nevertheless, the molecular weight of the polymers cannot be
increased indefinitely without leading to problems. Sensitivity towards
mechanical degradation becomes a significant problem as the molecular
weight of the polymers increases.13-17 In addition, the aforementioned
problem with bridging will be augmented with an increase in the molecular
weight.8,
9
In practice, high molecular weight HPAM leads to filter cake
formation on the surface (face plugging) of cores, especially low permeable
ones, and cannot be applied in a reservoir field with similar rock properties. 18
The introduction of charges (in PAM, the hydrolysis degree) in a polymer
backbone will lead to an increase in the solution viscosity compared to its
uncharged analogue.19 The higher the amount of charged moieties, the more
stretched the polymer coil will be and thus the higher the solution viscosity.
This enables the use of lower molecular weight polymers without jeopardizing
the thickening capability. However, for the polymer to remain insensitive
towards salinity and hardness of the brine, the hydrolysis degree cannot be
too high (e.g. higher than 40 mol%).20 Therefore, although beneficial for the
thickening capability in de-ionized water, the presence of charged moieties
will result in sensitivity issues towards electrolytes. In the presence of
divalent ions (such as Ca2+), even precipitation can arise, due to inter-chain
complexation21, eventually leading to a complete loss of solution viscosity. In
practice the hydrolysis degree is fixed at 30 mol%. Nevertheless, to date
partially hydrolyzed polyacrylamide (HPAM) is the polymer of choice for
chemical EOR, mainly in connection to its relatively low price (2 - 4 €/kg).
This project started with identifying the limitations of the currently used
HPAM in enhanced oil recovery (vida supra). The main objective of this
dissertation was to tackle a couple of these limitations and present new
solutions. Firstly the controlled polymerization of acrylamide (AM) was
accomplished through the use of atomic transfer radical polymerization
(ATRP) in water. In addition, the “living” character of the polymerization
process
offered
the
possibility
of
adding
a
second
block
of
N-
isopropylacrylamide (NIPAM). The controlled polymerization of AM enabled
the design of different molecular architectures of polyacrylamide (PAM).
Subsequently evidence for the increased thickening capability of branched
PAM versus linear PAM could be provided. In addition, the presence of
branches (N > 8) increased the elastic response of aqueous solutions of the
polymers. Given the uncharged nature of the branched PAMs (compared to
the commercial HPAM with ~ 30 mol% of charged moieties), the presence of
183
Outlook on the application of branched (co)polymers in EOR
salt does not influence the solution properties making these polymers
particularly suitable for high salinity reservoirs.
Another limitation related to HPAM is the temperature stability. The use
of copolymers of AM and NIPAM provides polymers that display thermothickening up to 80 °C, and are therefore resistant (in terms of solution
viscosity) to higher temperatures (T > 50 °C). In addition, the oil recovery
efficiency at high temperatures using the thermo-responsive copolymers is
significantly improved compared to a branched PAM analogue.
In the following sections, the branched PAMs prepared in this thesis are
discussed in terms of their rheological properties compared to either linear
PAMs or commercial linear HPAMs. Unresolved problems related to the use of
AM as a monomer are discussed and preliminary results on improvements are
presented. In addition, new preliminary results for other acrylamide-based
materials as well as several different biopolymers are presented in terms of
rheological behavior and oil recovery performance.
8.2. Thickening capability, comb-shaped PAM
The thickening capability of the currently used HPAM is due to its high to
ultra-high molecular weight (3.5 – 20·106 g/mol) and the presence of
charged (25-35 mol%) moieties.19,
22
According to the general theory of
3
polyelectrolyte solutions , the presence of the charged moieties leads to
electrostatic repulsions and subsequently to prominent chain stretching. 19
However, when dissolved in salt solutions the thickening capability is
significantly hampered (due to the electrostatic screening of the charged
moieties). Other ways of increasing the thickening capability of a polymer is
the introduction of hydrophobic moieties that will lead to aggregate
formation.19,
23
In this thesis (chapter 3 & 4, Figure 8.3), a new approach to
improve the thickening capability (in water solutions) of a polymer has been
developed.
The thickening capabilities of the branched PAMs depend on the
functionalization (number of arms) degree. A low number of arms (N ≤ 8)
leads to polymers which display a lower solution viscosity compared to linear
PAMs of equal theoretical overall molecular weight. This is attributed to the
inherent lower hydrodynamic volume of branched polymers.24,
25
For larger
number of arms (N ≥ 12), a higher solution viscosity is found when compared
to linear analogues. The branched PAMs with a relatively high number of
arms (N = 12, 13 and 17) possessed a higher hydrodynamic radius compared
to the branched PAMs with a low (N = 4 and 8) number of arms at equal total
molecular weight.
184
Chapter 8
Figure 8.3: New approach to increase the solution viscosity of aqueous solutions
In Chapter 1, the thickening capabilities of several different water soluble
polymers were plotted against each other in Figure 1.11. With the results of
Chapter 3 & 4 a comparison of the branched PAM with other (nonhydrophobic) AM based polymers is performed and the results are displayed
in Figure 8.4.
As can be observed in Figure 8.4 A, the thickening capability of the HPAM
is the highest of the three included in the comparison. Remarkably, the
thickening capability of the branched PAM (with a lower molecular weight
than that of the linear PAM) is seven times as high as that of the linear PAM.
This demonstrates that the molecular architecture is a strong tool to improve
the thickening capabilities of water-soluble polymers in the concentration
regime useful for EOR. The thickening capabilities of the branched PAMs have
been extensively discussed in Chapters 3 & 4. Here the focus will be on the
salt resistance of the branched PAMs in terms of solution viscosity and the
viscoelastic response of aqueous solutions containing them.
185
Outlook on the application of branched (co)polymers in EOR
A
4,5
Viscosity
Molecular weight
1,0
5
4
0,8
6
3,2
3
0,6
2
0,4
1
0,2
0,0
0
PAM
HPAM
0,5
13-arm PAM
B
Viscosity
Molecular weight
0,4
5
4
Viscosity (Pa.s)
3,2
3,2
3,2
3,2
6
3,5
3,2
0,3
3
0,2
2
0,1
1
0,0
AM
HP
M
Cl
Cl
Cl
Cl
PA
Na
Na
Na
Na
m
,0
r
2
0,5
2,0
3,0
1
-a
13
Molecular weight (x10 g/mol)
Viscosity (Pa.s)
3,5
Molecular weight (x10 g/mol)
1,2
0
Figure 8.4: Thickening abilities of different AM-based polymers, (A) the solution
viscosity (at = 10 s-1) of the polymer solution (1 wt.%) with corresponding molecular
weight and (B) the solution viscosity (at
= 10 s-1) of the polymer solution (0.5 wt.%)
with corresponding molecular weight at different salt (NaCl) concentration for HPAM
and a 13-arm branched PAM (no salt)
8.3. Salt resistance, comb-shaped PAM
The salt sensitivity of HPAM is a well know problem given its ionic
character.1
The
solution
viscosity
decreases
significantly
as
the
salt
concentration increases. Given that in all oil reservoirs brine (salt water) is
used, it is not a problem that can be circumvented by using deionized water.
In addition, in the presence of weak bases (such as sodium carbonate)
hydrolysis of the acrylamide moieties occurs which becomes extensive at
elevated temperatures (T > 60 °C) The injection of non-hydrolyzed PAM,
186
Chapter 8
rather than HPAM, has been proposed as a new method for EOR.26 The nonhydrolyzed PAM will be hydrolyzed in-situ and the viscosity of the solution will
increase. For oil reservoirs where a high amount of salt is present the use of
pristine PAM can represent a good option. However, the low thickening
capability of linear PAM compared to linear HPAM will be detrimental for a
project given the higher amount of linear PAM required to match the viscosity
of the aqueous phase to that of the oil. Therefore, we propose the use of
branched PAM with its better thickening capability compared to linear PAM for
high salinity applications (Figure 8.5).
10
3
PK30-g13-(PAM49225)
Viscosity (mPa.s)
A
PK30-g13-(PAM23810)
Poly(AM88630-ran-AA37470)
Poly(AM56135-ran-AA23730)
Poly(AM31515-ran-AA13320)
10
2
10
1
0
1
2
3
11
12
Concentration NaCl (wt. %)
Viscosity (mPa.s)
B
10
2
PK30-g13-(PAM49225)
Poly(AM88630-ran-AA37470)
PK30-g13-(PAM23810)
Poly(AM31515-ran-AA13320)
10
1
0,1
1
10
100
1000
-1
Shear rate (s )
Figure 8.5: A; the solution viscosity ( = 10 s-1) as a function of the salt (NaCl)
concentration for HPAM and branched PAM, and B; the viscosity functions of 2 charged
HPAMs and 2 uncharged branched PAMs
187
Outlook on the application of branched (co)polymers in EOR
As evident in Figure 8.5 A, the solution viscosity of the uncharged branched
PAMs (PK30-g13-(PAM23810) and PK30-g13-(PAM49225)) is not affected by
the presence of salt (up to 12 wt.% of NaCl). The solution viscosities ( = 10
s-1, [p] = 5000 ppm) of the charged linear HPAM are all higher than the
branched PAMs in de-ionized water. However, as the amount of salt increases
the solution viscosities of the branched PAMs remain constant while that of
the
charged
HPAMs
decreases
significantly.
Remarkably,
the
solution
viscosity of a charged HPAM with a molecular weight between 8 – 10 · 106
g/mol decreases to values lower than that of the PK30-g13-(PAM49225) (Mn ≈
3.5 · 106 g/mol). This demonstrates the suitability in terms of the solution
viscosity of the branched PAMs for application in high salinity environments.
The shear thinning behavior of the aqueous solutions has also been probed.
As can be observed in Figure 8.5 B, this pseudoplasticity of the branched PAM
is similar to that of the charged HPAM with a molecular weight either 2 or 3
times as high as that of the branched PAM. In actual applications the
pseudoplastic behavior is preferred, given that a low viscosity at high shear
rates will require less pumping energy.
Another important parameter identified for an efficient oil recovery is the
viscoelasticity of the aqueous phase.27-34 In Figure 8.6, the viscoelastic
response of aqueous solutions containing either a linear HPAM or a branched
PAM is displayed.
As can be observed in Figure 8.6 A, the viscoelastic response of the
HPAM solution is dependent on the salt concentration. A significant decrease
in the elasticity of the solution can be clearly distinguished as the
concentration of the salt increases. The reduction19 in the hydrodynamic
volume of the polymer coils, due to electrostatic screening, is the accepted
explanation of the observed behaviour.3 The effective size of the polymer in
solution is smaller, and therefore the extent of overlapping is suppressed
which leads to a lower elastic response.
The results for the uncharged branched PAM (Figure 8.6 B) demonstrate
that the elastic response of the aqueous solution is not affected by the
presence of salt.
188
Chapter 8
90
Increasing NaCl
concentration
80
Phase angle
70
60
A
50
40
NaCl concentration = 30000 ppm
NaCl concentration = 20000 ppm
NaCl concentration = 5000 ppm
NaCl concentration = 0 ppm
30
0
0,1
1
10
100
Frequency (rad/s)
90
80
NaCl concentration = 30000 ppm
NaCl concentration = 20000 ppm
NaCl concentration = 5000 ppm
NaCl concentration = 0 ppm
Phase angle
70
60
B
50
40
30
0
0,1
1
10
100
Frequency (rad/s)
Figure 8.6: (A) the viscoelasticity as a function of the salt (NaCl) concentration for
HPAM (Mw = 3.2·106 g/mol, [p] = 1.0 wt.%), and (B) the viscoelasticity as a function
of the salt (NaCl) concentration of a branched PAM (Mw = 1.7·106 g/mol, [p] = 1.0
wt.%)
8.4. Hydrolysis resistance, comb-shaped PAM
The hydrolysis reaction of PAM is a well-known reaction that can be
catalysed either by an acid or a base.35 The hydrolysis reaction (Scheme 8.1)
leads to the formation of ammonia.
In ASP floods most often sodium carbonate is used as the alkali agent.
Therefore, the resistance to base catalysed hydrolysis of PAM is important. In
general there are two stages of the hydrolysis reaction.35 The first one (high
189
Outlook on the application of branched (co)polymers in EOR
rate) reaches hydrolysis degrees up to 40 mol% and is accelerated by
neighbouring carboxylate groups.
Scheme 8.1: Base catalysed hydrolysis of PAM
The second stage displays a ten times lower rate. This is suppressed by the
electrostatic repulsion between the carboxylate groups and the base, and the
increased viscosity due to chain stretching driven by electrostatic repulsion of
the carboxylate groups leads to mass transfer limitations. The parameters
that have been identified to accelerate the hydrolysis rate are high
temperatures, the presence of salts, polymer concentration, and high
base/polymer ratio.35-37 The characteristics of chemical EOR usually are a low
polymer concentration for economic reasons, temperatures above 50 °C
found for many oil reservoirs, and the presence of salts in the water used as
the displacing fluid. Therefore, it is obvious that the challenge to design a
polymer that can resist the base hydrolysis under the conditions in EOR is
important at an industrial level.
The use of other monomeric units that can withstand alkaline hydrolysis
is
a
viable
option.
Investigations
towards
novel
multiblock
co-
and
terpolymers have demonstrated the effectiveness of changing the AM units
into other more resistant moieties.38,
39
Several different acrylamide based
monomers have been investigated as hydrolysis resistant ones (Figure
8.7).40-42
However, the homopolymers of DMA and AM display a markedly different
behaviour under the same conditions. After 50 hours, the hydrolysis degree
of poly(N,N-dimethylacrylamide) (PDMA) is only 2 mol%, while that of PAM
reached a hydrolysis degree of 30 mol% after only 2 minutes.43,
44
The
reactivity of PAM towards alkaline hydrolysis is 500 times higher compared to
that of PDMA and PAAEE.43,
44
The synthesis of the polymers have all been
through free radical polymerization. In order to benefit from the improved
thickening capability of branched polymers compared with linear ones, the
controlled polymerization of the hydrolysis resistant monomers is required.
The
controlled
polymerization
demonstrated already.45-50
190
of
DMA,
NIPAM
and
AAE
has
been
Chapter 8
Figure 8.7: Hydrolysis resistant acrylamide based monomers
8.4.1. Results and discussion
Macroinitiators. The synthesis of the macroinitiators was performed
according to the Paal-Knorr reaction (Scheme 8.2) of a halogenated primary
amine
with
aliphatic
perfectly
alternating
polyketones.
The
carbonyl
conversion was determined using elemental analysis. The characterization of
the macroinitiators has been extensively investigated in Chapter 3 & 4 and
therefore will not be discussed here. The properties of the macroinitiators
used in the synthesis of branched PDMA are listed in Table 8.1.
Table 8.1: Properties of the macro-initiators
Polyketone sample (PK30-Cla)
PK30 (virgin)
PK30-Cl4
PK30-Cl8
PK30-Cl13
Elemental composition
XCO (%)b
(C : H : N, wt%)
67.0
58.6
64.0
62.9
:
:
:
:
8.4
7.1
7.9
7.6
:
:
:
:
0.0
1.6
3.3
4.9
18.87
37.21
61.14
Pyrrole
unitsc
0
4
8
13
Mn,GPC
PDI
2
2
2
2
1.74
2.02
2.01
1.97
797
447
244
072
a. Number indicates the ethylene content (%)
b. The conversion of the carbonyl groups of the polyketone
c. Average number of pyrrole units per chain
191
Outlook on the application of branched (co)polymers in EOR
The obtained, chemically modified polyketones are used as macroinitiators in
the ATRP of DMA for the preparation of comb-shaped polymers with a
different number of side chains. The synthesis of linear and comb-like PDMA
was performed according to Scheme 8.2.
Scheme 8.2: Synthesis of (A) linear PDMA and (B) comb PDMA
Table 8.2: Characteristics of the linear and branched PDMAs
M/s1/s2a (wt:vol:vol);
T; Time (min)
Architecture
Entry
[M]0:[I]0:[CuCl]0:
[Me6TREN]0
Linear
1
2
3
22 919:1:1.5:1.5
51 623:1:1.5:1.5
88 515:1:3.0:3.0
1:5
1:5
1:5
4-arm
4
79 351:1:3.0:3.0
1:4:1/10; 25 °C;
8-arm
5
79 213:1:3.0:3.0
1:4:1/10; 25 °C;
; 25 °C; 180
; 25 °C; 180
; 25 °C; 180
Conv (%)
Mn,tot
Mn,SPAN
58.4
58.2
47.0
1 326 825
2 978 320
4 124 011
1 326 825
2 978 320
4 124 011
60
70.1
5 514 111
2 759 853
60
64.5
5 064 788
13-arm
6
7
8
9
19 969:1:1.5:1.5
49 905:1:1.5:1.5
99 226:1:3.0:3.0
200 000:1:3.0:3.0
1:6:1/6 ; 25 °C; 130
1:5:1/10; 25 °C; 180
1:5:1/20; 25 °C; 150
1:5:1/40; 25 °C; 180
39.9
46.3
49.9
30.3
789 831
2 290 499
4 908 300
6 007 278
17-arm
10
100 030:1:1.5:3.0
1:4:1/20; 25 °C; 180
41.0
4 065 549
1 268 994
b
124
355
757
926
481 097
a. M/s1/s2 = Monomer / solvent 1 / solvent 2 = N,N-dimethylacrylamide / water / acetone
b. Mn,GPC = 771 300 g/mol and the PDI = 1.80 as determined by aqueous GPC
The ratio between the initiator (or the macroinitiator) and the monomer was
varied in order to synthesize linear and comb-shaped PDMA with different
192
310
181
920
993
Chapter 8
molecular weights. The linear polymers were prepared using MClPr as the
initiator while the comb PDMAs the polyketone based macroinitiators were
used. Table 8.2 lists the results for the different polymers prepared.
The polymerization of DMA in water at room temperature using the
polymerization process described in Chapter 2 & 3 allows for the preparation
of linear and branched PDMA with relatively low dispersity indices. The
3
10
2
A
Viscosity (mPa.s)
10
PK30-g13-(PDMA23105), entry 7
PDMA30045,
0,1
1
10
3
10
2
B
PK30-g17-(PDMA41010), entry 10
entry 2
PDMA41600,
10
100
1000
0,1
entry 3
1
10
-1
100
-1
Shear rate (s )
Shear rate (s )
C
Viscosity (mPa.s)
Viscosity (mPa.s)
rheological properties depend on the number of arms (Figure 8.8).
10
2
PK30-g13-(PDMA49515), entry 8
10
PK30-g8-(PDMA51090),
entry 5
PK30-g4-(PDMA55625),
entry 4
1
0,1
1
10
100
1000
-1
Shear rate (s )
Figure 8.8: Viscosity functions of (A) linear and 13-arm branched PDMA of similar
Mn,tot,[p] = 2.0 wt%, (B) linear and 17-arm branched PDMA of similar Mn,tot,[p] = 2.0
wt% and (C) 4-arm, 8-arm and 13-arm branched PDMA, [p] = 1.0 wt%
Increasing the number of arms (from N = 4 to 17) leads to a higher solution
viscosity at equal polymer concentration and molecular weight, similar to the
results obtained for the branched PAMs (Chapter 4). This is evident from the
193
1000
Outlook on the application of branched (co)polymers in EOR
comparison between a 13-arm (entry 7) and a 17-arm (entry 10) branched
PDMAs
with
their
corresponding
linear
analogues
(entries
2
and
3
respectively). The comparison between a 4, 8 and 13-arm branched PDMA
further demonstrates the effect of the number of branches on the solution
viscosity.
The hydrolysis resistance of the linear and branched PDMA were
investigated under conditions resembling those found in actual chemical EOR
(Figure 8.9).
400
Viscosity retention (%)
350
PAM21445,
PAM21445,
NaCl
NaCl-CaCl2
PK30-g13-(PAM23810),
NaCl
PK30-g13-(PAM23810),
NaCl-CaCl2
A
PK30-g13-(PDMA49515), NaCl
PK30-g13-(PDMA49515), NaCl-CaCl2
300
250
200
150
100
0
25
50
75
100
125
150
175
200
150
175
200
Time (hours)
Hydrolysis degree (%)
100
95
PAM21445,
PK30-g13-(PAM23810),
14
B
NaCl
NaCl
12
10
8
6
4
2
0
0
25
50
75
100
125
Time (hours)
Figure 8.9: (A) Solution viscosity (in percentages from the starting value) as a
function of hydrolysis time for a linear and a branched PAM and a branched PDMA,
[p]=5000 ppm, (B); the hydrolysis degree of the linear and branched PAM as a
function of temperature as measured by
194
13
C-NMR-spectroscopy
Chapter 8
The aqueous solution used in EOR usually remains for extensive times in the
reservoir. Periods of several months onshore, and up to more than one year
offshore
have
been
stated.22,
51
The
most
important
parameter
for
maintaining the success of the polymer flood is the solution viscosity of the
aqueous phase.
Upon hydrolysis charged groups are randomly introduced in the polymer.
This will lead to electrostatic repulsion3 thus increasing the hydrodynamic
volume of the coils in solution and this is synonymous to a higher solution
viscosity. The increase in the solution viscosity as a function of the hydrolysis
time is significantly more pronounced for the linear PAM (PAM21445) when
compared to the branched PAM (PK30-g13-(PAM23810) and PDMA (PK30-g13(PDMA48515). The solution viscosity increases by more than 300 % of the
original value, both with and without CaCl 2. This is strong evidence that the
hydrolysis of the linear PAM is extensive (while that of the branched analogue
is not), since in the absence of salt the solution viscosity increases with an
increase in the hydrolysis degree (up to a limiting value). 26 The differences in
the solution viscosities between the presence of NaCl or NaCl-CaCl2 suggest
that either the hydrolysis is suppressed by the presence of CaCl 2 or part of
the polymer precipitates by complex formation with Ca 2+.52 The relatively low
increase in the solution viscosity of the branched PAM suggests a lower
hydrolysis rate compared to the linear analogue. However, the increase in
solution viscosity with an increase in the hydrolysis rate not necessarily has
to be equal for both the linear and the branched PAM. Therefore, the direct
measurement of the hydrolysis degree (by
13
C-NMR) was carried out for the
two samples (Figure 8.9 B). The increase in the hydrolysis degree is similar
during the first couple of hours. After 50 hours the hydrolysis degree of the
linear PAM surpasses that of the branched PAM indicating that the branched
PAM is more resistance to alkaline hydrolysis compared to the linear
analogue. As can be observed in Figure 8.9 A, the change in the solution
viscosity of the branched PDMA is limited. This is strong evidence that the
branched PDMA is resistant towards alkaline hydrolysis, which is in line with
earlier reports.44
Increasing the residence time under the harsh conditions and the salt
concentration leads to a significant increase in the solution viscosity (Figure
8.10) for the linear PAM.
195
Outlook on the application of branched (co)polymers in EOR
Viscosity retention (%)
300
250
PK30-g13-(PAM35275),
NaCl
PK30-g13-(PAM35275),
NaCl-CaCl2
PAM35705,
PAM35705,
NaCl
NaCl-CaCl2
PK30-g13-(PDMA23105), NaCl
PK30-g13-(PDMA23105), NaCl-CaCl2
PDMA30045,
PDMA30045,
200
NaCl
NaCl-CaCl2
150
100
50
A
0
0
10
20
30
40
50
60
Time (days)
110
Viscosity retention (%)
100
90
80
70
60
50
40
0
0
Poly(AM31515-co-AA13320),
Poly(AM31515-co-AA13320),
NaCl
NaCl-CaCl2
PK30-g13-(PDMA23105),
NaCl
PK30-g13-(PDMA23105),
NaCl-CaCl2
PDMA30045,
PDMA30045,
NaCl
NaCl-CaCl2
10
20
30
40
B
50
60
Time (days)
Figure 8.10: (A) Solution viscosity (in percentages from the starting value) as a
function of hydrolysis time for a linear ([p] = 5000 ppm) and a branched PAM ([p] =
4900 ppm) and a linear ([p] = 5900 ppm) and a branched PDMA ([p] = 4500 ppm), at
equal molar concentration, (B) Solution viscosity (in percentages from the starting
value) as a function of the hydrolysis time for a linear ([p] = 10000 ppm) and a
branched PDMA ([p] = 6500 ppm) and a commercial HPAM ([p] = 5500 ppm), at equal
starting solution viscosity (measured at
= 10 s-1)
The branched PAM (with similar molecular weight) displays at first an
increase in the solution viscosity (albeit less pronounced compared to its
linear analogue) and decreases slowly to below the starting viscosity.
The hydrolysis degree of the 62 days samples was determined by
13
C-
NMR as being 38 and 33 mol% for, respectively, the linear and the branched
PAM. The solution viscosity of the linear and branched PDMA is not
196
Chapter 8
significantly affected by the conditions applied, even after more than 60
days. This suggests that little, if any, hydrolysis takes place. This is
confirmed by
13
C-NMR where no carboxylate units could be detected (i.e.
below the detection limit of
13
C-NMR) for the 62 days samples.
The presence of CaCl2 also affected the solution viscosity of the samples;
however the differences between the samples with CaCl 2 and the ones
without were not large. Although the presence of CaCl 2 did not significantly
affect the solution viscosity of the samples, precipitation was observed in the
case of PAM-based polymers (Figure 8.11).
Figure 8.11: Precipitation of the commercial HPAM (with CaCl2) sample after 42 days
The solutions of the linear and branched PDMA stayed clear even after 62
days in the oven, whereas the linear HPAM became more turbid. This
indicates the formation of large aggregates.
8.5. Oil recovery, 2D flow-cell
The efficiency of the branched hydrolysis resistant PDMA in recovering oil
out of dead-ends was evaluated using the flow-cell (Chapter 7). In addition,
the oil recovery at higher temperatures (i.e. T = 70 °C) using the thermoresponsive block copolymers (Chapter 6) was also evaluated.
8.5.1. Oil recovery efficiency
The efficiency in recovering residual oil by branched PDMA (at room
temperature) and branched random copolymer of AM and NIPAM (at room
temperature and 70 °C) has been evaluated (Figure 8.12).
197
Outlook on the application of branched (co)polymers in EOR
[2] Poly(AM31515-ran-AA13320)
[1] Brine (30000 ppm NaCl)
[4] Water, RT
[7] Water, 70 °C
[3] PK30-g13-(PDMA23105)
[p] = 8700 ppm
[p] = 8000 ppm
Residual oil recovery = 7.6 % (± 1.8)
Residual oil recovery = 8.9 % (± 1.8)
[5] PK30-g13-(PAM3275), RT
[6] PK30-g13-(PAM1405-ran-PNIPAM1405)
[p] = 11000 ppm
[p] = 9000 ppm, RT
Residual oil recovery = 4.8 % (±1.9)
Residual oil recovery = 3.6 % (±1.9)
[8] PK30-g13-(PAM3275), 70 °C
[9] PK30-g13-(PAM1405-ran-PNIPAM1405)
[p] = 11000 ppm
Residual oil recovery = 8.6 % (±1.8)
[p] = 9000 ppm, 70 °C
Residual oil recovery = 50.2 % (±1.0)
Figure 8.12: Oil recovery out of dead ends using branched PDMA ([3]) compared to
brine ([1]) and the commercial polymer ([2])at room temperature using crude oil, and
branched copolymers of AM and NIPAM ([6] and [9]) compared to water ([4] and
[7]) and branched PAM of similar molecular weight ([5] and [8]) both at room
temperature and at 70 °C using a mixture of crude oil and cyclo octane (2-1 vol.%)
The efficiency of the recovery of residual oil using the branched PDMA ([3]
PK30-g13-(PDMA23105)) is similar to that of the branched PAM (Chapter 7,
[D] PK30-g13-(PAM35275) and that of the commercial polymer ([2],
poly(AM31515-co-AA13320). However, the polymer concentration required to
match the solution viscosity of the water phase with that of the oil is higher
compared to a branched PAM of the same molecular weight. Nevertheless,
the ability to recover part of the residual oil makes these hydrolysis resistant
branched PMDA polymers potential candidates for EOR where alkali is also
used to generate in situ surfactants.
The residual oil recovery efficiency of the branched thermo-responsive
copolymers is slightly higher compared to a branched PAM ([5] PK30-g13(PAM3275)), similar molecular weight). However, when performing the
198
Chapter 8
comparison at 70 °C, different recovery efficiencies are observed for the
branched PAM and thermo-responsive copolymer. The recovery efficiency of
the branched PAM increases from 4.8 to 8.6 %. This can be attributed to the
improved mobility ratio (equation 1.3, Chapter 1) due to the lower viscosity
of the oil (the decrease in the oil viscosity is more pronounced that the
decrease of the polymer solution viscosity).
When comparing the recovery efficiency of the branched thermoresponsive copolymer a significantly higher efficiency is observed. This
cannot reside only in the decrease of the oil viscosity at higher temperatures.
The higher oil recovery efficiency of the branched copolymer ([9] PK30-g13(PAM1405-co-PNIPAM1405)) at 70 °C is therefore attributed to the increased
solution viscosity (Chapter 6). The mobility ratio is lower than unity (and thus
lower at 70 °C compared to at RT) given the higher solution viscosity, and
thus a better displacement of the oil takes place. However, from a practical
point of view, the polymer concentration can be decreased until the solution
viscosity at 70 °C matches that of the oil.
The increase in oil recovery efficiency at higher temperatures makes
these types of copolymers interesting candidates for application in EOR where
the reservoir temperatures are high (i.e. T ≥ 50 °C).
8.6. Biopolymers for EOR
In certain regions of the world regulations stipulate that if a polymer is
used in recovering oil, it has to be reusable or biodegradable. If a synthetic
polymer is used, the produced mixture of oil and water (containing the
synthetic polymer) has to be separated and the water phase must be reinjected. However, in most field application the polymer that is produced
along with the oil has been either chemically or thermally degraded and
therefore cannot be re-injected.22 Therefore, the use of biopolymers is almost
inevitable and a lot of effort has been put in developing biopolymers for EOR.
Although there are many examples of biopolymers that can be used for EOR,
only xanthan gum has been applied in actual oil reservoirs 22,
53
, although
there are current (pilot) projects under way with other water soluble
biopolymers, such as schizophyllan.
8.6.1. Thickening capability and viscoelasticity
Most of the biopolymers that have been considered so far for EOR are
polysaccharides.22 The ability of these type of polymers to increase the
viscosity of an aqueous solution is based on their high molecular weight and
in some cases the rigidity of the polymeric chains.19 Although there are many
199
Outlook on the application of branched (co)polymers in EOR
different types and sources of biopolymers, not all of them are soluble in cold
water. In many cases boiling water is required before complete dissolution of
the polymeric chains is obtained. From an economical and practical (remote
locations of many oil reservoirs) point of view, the dissolution in cold water is
preferred. Although only xanthan gum has been applied so far in chemical
EOR, there are many other different biopolymers that might be suitable. The
thickening capability54-56 of several different biopolymers is displayed in
Figure 8.13.
7
7
Viscosity
Molecular weight
6
5
5
6
Viscosity (Pa.s)
5
Molecular weight (x10 g/mol)
6
4
4
3
3
2,6
2
2
2
1,5
1,08
1
0,7
0,96
0,66
1
0,09
0
0
)
n
m
se
an
an
an
M)
ca
HM
gu
yll
(L
os
en
ulo
glu
e(
ar
ph
hit
ge
an
ell
se
o
s
u
o
r
a
c
h
C
o
o
t
l
z
l
l
G
rr
le
n
hi
llu
llu
ca
thy
Sc
Xa
Sc
ce
ce
aMe
bd
CM
CM
m
La
m
gu
Figure 8.13: Thickening capabilities (viscosity measured at
= 10 s-1) of different
biopolymers at a polymer concentration of 1 wt.%
As can be observed in Figure 8.12, there are several other biopolymers that
can significantly increase the viscosity of a water solution. Scleroglucan, a 1,3 linked D-glucose with single D-glucose side chains linked -1,6 every
third unit57, has long been seen as a good substitute for xanthan gum 58,
especially
in
oil
reservoirs
where
high
temperature
and
high
salt
concentration (given the non-ionic character of scleroglucan) are found.59, 60
Another biopolymer that has been identified as a suitable biopolymer for
EOR is schizophyllan (chemically the same as scleroglucan).61 This resides
mainly in its ability to increase the solution viscosity even at very low
polymer concentration (i.e. a solution viscosity of 10 mPa.s,
[p] = 200 ppm).
61
= 10 s-1 and a
In addition, the solution viscosity of an aqueous solution
containing schizophyllan only decreases by 10 % when heated up to 130
200
Chapter 8
°C.61 By comparison, an aqueous solution of xanthan gum decreases by 95
%.61
Carboxy-methyl cellulose (CM cellulose) has also been considered as a
good candidate for EOR.53 The addition of carboxy-methyl groups to cellulose
makes the biopolymer soluble in cold water.53 This makes it attractive for
EOR since no specialty dissolution equipment is required. However, given the
ionic character of carboxy-methylcellulose, the solution viscosity is sensitive
to the presence of salt.53 Depending on the molecular weight (low or high [LM
or HM]) of the parent cellulose polymer a different thickening capability is
observed.
Chitosan has also been shown to increase the solution viscosity of a
water solution significantly.55 However, chitosan is only soluble in acidic
media62,
63
(i.e. pH62 < 6.0) which will significantly hamper its application in
EOR. Nevertheless, the high thickening capability of chitosan still makes it an
interesting polymer as a rheology modifier.
Methyl-cellulose, on the other hand, is soluble in cold neutral water.
Although it’s capability to increase the solution viscosity is less than most of
the biopolymers, its low molecular weight might make it suitable for low
permeable reservoirs. A peculiar behaviour of methyl-cellulose is its gelation
(in water solution) upon heating due to hydrophobic associations.64,
65
The
gelation is reversible; upon cooling the aqueous solution will return to its
original state.64,
65
Nevertheless, the hydrophobic character of parts of the
biopolymer might lead to enhanced adsorption (higher resistant factors,
indicating a higher layer thickness) on the rock surface similar to that
observed for hydrophobically modified polymers.23, 66
Guar gum has also been investigated for application in EOR. It is used
already to control the rheological properties of drilling muds. The thickening
capability of guar gum is higher than xanthan gum, but in solution of high
salinity guar gum is highly sensitive towards high temperatures limiting its
application.67
Little effort has been aimed at investigating the viscoelasticity of aqueous
solutions containing biopolymers. Experiments and mathematical models
have demonstrated the importance of the viscoelasticity of the solution on
the recovery of residual oil27-30,
30-34
, although so far no consensus has been
reached. The viscoelasticity of some commercial biopolymers has been
evaluated and the results are displayed in Figure 8.14.
The results suggest that the elastic response of the aqueous solutions
increases as the molecular weight increase. This indicates that the extent of
overlapping is higher for the higher molecular weight polymers. Although the
201
Outlook on the application of branched (co)polymers in EOR
molecular weight of xanthan gum is not as high as that of CM cellulose, its
elastic response is much more pronounced. A possible explanation for this is
the rigidity of the polymeric chains. Xantham gum is known to form helices in
G' G" (Pa)
water solutions.68-71
10
2
10
1
10
0
10
-1
10
-2
10
-3
10
-4
A
= Xanthan gum
= Guar gum
= CM cellulose
= Methyl cellulose
= -Carrageenan
Filled symbols = G'
Empty symbols = G"
0,1
1
10
100
Frequency (rad/s)
Phase angle
90
Methyl cellulose
80
-Carrageenan
70
Guar gum
CM cellulose
Xanthan gum
60
50
B
40
30
20
10
0
0,1
1
10
100
Frequency (rad/s)
Figure 8.14: (A), the loss and elastic modulus as a function of the frequency of
different biopolymers ([p]=1.0 wt.%) and (B), the phase angle as a function of the
frequency of the same biopolymers
8.7. Conclusion
Currently used partially hydrolyzed polyacrylamide (HPAM) in EOR has
several limitations. The main objective of this dissertation was to design new
acrylamide based polymers that provide solutions to the limitation of the
202
Chapter 8
aforementioned polymer. The first hurdle that had to be passed was the
controlled polymerization of AM. This was accomplished through the use of
atomic transfer radical polymerization (ATRP) in water at room temperature.
Furthermore, given the “living” character of the polymerization process a
second block of N-isopropylacrylamide (NIPAM) can be added to the first AM
block. With the accomplishment of controlled polymerization of AM, PAM with
different molecular architecture could be envisaged. This was achieved
through the use of functionalized (with halogens) alternating aliphatic
polyketones. Subsequently, evidence for the increased thickening capability
of bottle-brush PAM compared to a linear analogue was provided. The
presence of branches (N > 8) increased the elastic response of aqueous
solutions of the polymers. Also, given the uncharged nature of the bottlebrush PAMs (compared to the commercial HPAM [~30 mol% charged
moieties]), the presence of salt does not influence the solution properties
making these polymers particularly suitable for high salinity reservoirs.
Another limitation related to HPAM is the temperature stability. Rheological
characterization demonstrated that copolymers of AM and NIPAM display
thermo-thickening behavior up to 80 °C, and are therefore resistant (in terms
of solution viscosity) to higher temperatures (T > 50 °C). The increased oil
recovery efficiency at high temperature (T = 70 °C) demonstrates the
potential of the thermo-responsive polymers for EOR.
Since in many cases alkaline agents are used in combination with
polymers, hydrolysis of the AM units in HPAM is extensive, especially at high
temperatures (T > 50 °C). The use of hydrolysis resistant moieties such as
DMA is promising, more so in combination with the increased thickening
capability through branching. The oil recovery efficiency of the branched
PDMA polymers is similar to branched PAMs, and these are therefore good
candidates for application in EOR where alkaline agents are used to generate
in situ surfactants.
The potential of using biopolymers for EOR where synthetic polymers
cannot be applied is briefly discussed. Most of the investigated biopolymers
are polysaccharides, with the differences being the source and molecular
structure of the polymeric chains. Their thickening capability is a function of
the molecular weight and the resistance towards salt depends on their ability
to form helical structures in aqueous solutions.
8.8. Acknowledgements
This work is part of the Research Program of the Dutch Polymer Institute
DPI, Eindhoven, The Netherlands, project #716.
203
Outlook on the application of branched (co)polymers in EOR
8.9. References
1. Thomas, A.; Gaillard, N.; Favero, C. Oil Gas Sci Technol 2012, 6, 887.
2. Guyot, A.; Chu, F.; Schneider, M.; Graillat, C.; McKenna, T. F. Progress in Polymer
Science 2002, 8, 1573.
3. Stokes, R. J.; Evans, D. F. Fundamentals of interfacial engineering; Wiley-VCH: New
York, 1997; .
4. Degennes, P. G. J. Chem. Phys. 1971, 2, 572.
5. Ferry, J. D. Viscoelastic properties of polymers; John Wiley & Sons: New York, 1980;
, pp 641.
6. Baumgartel, M.; Willenbacher, N. Rheologica Acta 1996, 2, 168.
7. Chauveteau, G.; Denys, K.; Zaitoun, A. SPE 2002, SPE-75183.
8. Zitha, P. L. J.; Botermans, C. W. SPE 1998, SPE-36665.
9. Zitha, P. L. J.; van Os, K. G. S.; Denys, K. F. J. SPE 1998, SPE-39675.
10. Zitha, P. L. J.; Chauveteau, G.; Leger, L. J. Colloid Interface Sci. 2001, 2, 269.
11. Shell Global Solutions International BV 2012.
12. Al-Saadi, F. S.; Amri, B. A.; Nofli, S.; Van Wunnik, J.; Jaspers, H. F.; Harthi, S.;
Shuaili, K.; Cherukupalli, P. K.; Chakravarthi, R. SPE 2012, SPE 154665,.
13. Moreno, R. A.; Muller, A. J.; Saez, A. E. Polymer Bulletin 1996, 5, 663.
14. Maerker, J. M. Society of Petroleum Engineers Journal 1975, 4, 311.
15. Maerker, J. M. Society of Petroleum Engineers Journal 1976, 4, 172.
16. Seright, R. S.; Seheult, M.; Talashek, T. SPE 2009, 5, SPE-115142-PA.
17. Zaitoun, A.; Makakou, P.; Blin, N.; Al-Maamari, R. S.; Al-Hashmi, A. R.; AbdelGoad, M.; Al-Sharji, H. H. Spe Journal 2012, 2, 335.
18. Wever, D. A. Z. 2009.
19. Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Prog. Polym. Sci. 2011, 1558.
20. Shupe, R. D. J. Petrol. Technol. 1981, 8, 1513.
21. Peng, S.; Wu, C. Macromolecules 1999, 3, 585.
22. Lake, L. W. Enhanced Oil Recovery; Prentice-Hall Inc.: Englewood Cliffs, NJ, 1989;
Vol. 1, pp 550.
23. Taylor, K. C.; Nasr-El-Din, H. A. J. Petrol. Sci. Eng. 1998, 3-4, 265.
24. Anthony, A. J.; King, P. H.; Randall, C. W. J Appl Polym Sci 1975, 1, 37.
25. Kulicke, W. M.; Horl, H. H. Colloid Polym. Sci. 1980, 7, 817.
26. Levitt, D. B.; Pope, G. A.; Jouenne, S. Spe Reservoir Evaluation & Engineering
2011, 3, 281.
27. Wang, D.; Cheng, J.; Yang, Q.; Gong, W.; Li, Q.; Chen, F. SPE 2000, SPE-63227MS.
28. Wang, D.; Xia, H.; Liu, Z.; Anda, Q.; Yang, Q. SPE 2001, SPE-68723-MS.
29. Yin, H.; Wang, D.; Zhong, H. SPE 2006, SPE-101950-MS.
30. Xia, H.; Wang, D.; Wang, G.; Wu, J. Petrol. Sci. Technol. 2008, 4, 398.
31. Hou, J. R.; Liu, Z. C.; Zhang, S. F.; Yue, X.; Yang, J. Z. Journal of Petroleum
Science and Engineering 2005, 3-4, 219.
32. Zhang, L.; Yue, X.; Guo, F. Pet. Sci. 2008, 1, 56.
33. Xia, H.; Ju, Y.; Kong, F.; Wu, J. SPE 2004, SPE-88456-MS.
34. Zhang, L.; Yue, X. J. Cent. South Univ. T. 2008, 84.
35. Kurenkov, V. F.; Hartan, H. G.; Lobanov, F. I. Russian Journal of Applied Chemistry
2001, 4, 543.
36. Maurer, J. J.; Harvey, G. D.; Klemann, L. P. Abstracts of Papers of the American
Chemical Society 1986, 152.
37. Went, P. M.; Evans, R.; Napper, D. H. Journal of Polymer Science Part C-Polymer
Symposium 1975, 49, 159.
38. Gaillard, N.; Giovannetti, B.; Favero, C. SPE 2010, SPE-129756.
39. Vermolen, E. C. M.; van Haasterecht, M. J. T.; Masalmeh, S. K.; Faber, M. J.;
Boersma, D. M.; Gruenenfelder, M. SPE 2011, SPE-141497-MS,.
40. Gelfi, C.; Debesi, P.; Alloni, A.; Righetti, P. G. J. Chromatogr. 1992, 1-2, 333.
41. Miertus, S.; Righetti, P. G.; Chiari, M. Electrophoresis 1994, 8-9, 1104.
204
Chapter 8
42. SimoAlfonso, E.; Gelfi, C.; Lucisano, M.; Righetti, P. G. Journal of Chromatography
a 1996, 1-2, 255.
43. Righetti, P. G.; Gelfi, C. Anal. Biochem. 1997, 2, 195.
44. Righetti, P. G.; Chiari, M.; Nesi, M.; Caglio, S. J. Chromatogr. 1993, 2, 165.
45. Narumi, A.; Chen, Y.; Sone, M.; Fuchise, K.; Sakai, R.; Satoh, T.; Duan, Q.;
Kawaguchi, S.; Kakuchi, T. Macromolecular Chemistry and Physics 2009, 5, 349.
46. Neugebauer, D.; Matyjaszewski, K. Macromolecules 2003, 8, 2598.
47. Kim, K.; Kim, J.; Jo, W. Polymer 2005, 9, 2836.
48. Mueller, A. H. E.; Millard, P.; Mougin, N. C.; Boeker, A. Abstracts of Papers of the
American Chemical Society 2008.
49. Xia, Y.; Burke, N. A. D.; Stover, H. D. H. Macromolecules 2006, 6, 2275.
50. Xia, Y.; Yin, X. C.; Burke, N. A. D.; Stover, H. D. H. Macromolecules 2005, 14,
5937.
51. Sorbie, K. S. Polymer-improved oil recovery; CRC Press: Boca Raton, FL, 1991; .
52. Moradi-Araghi, A.; Doe, P. H. SPE 1987, SPE-13033-PA.
53. Donaldson, E. C.; Chilingarian, G. V.; Yen, T. F. Enhanced Oil Recovery II,
processes and operations; Elsevier: Amsterdam, The Netherlands, 1989; Vol. 2,
pp 604.
54. Enomoto, H.; Einaga, Y.; Teramoto, A. Macromolecules 1984, 8, 1573.
55. Desbrieres, J. Biomacromolecules 2002, 2, 342.
56. Tipvarakarnkoon, T.; Senge, B. Ann T Nord Rheol Soc 2008, 16,.
57. Grassi, M.; Lapasin, R.; Pricl, S. Carbohydr. Polym. 1996, 2, 169.
58. Dixon, B. Bio-Technology 1985, 7, 601.
59. Kalpakci, B.; Jeans, Y. T.; Magri, N. F.; Padolewski, J. P. SPE 1990, SPE-20237MS,.
60. Kulawardana, E. U.; Koh, H.; Kim, D. H.; Liyanage, P. J.; Upamali, K. A. N.; Huh,
C.; Weerasooriya, U.; Pope, G. A. SPE 2012, SPE-154294-MS,.
61. Leonhardt, B. SPE 2011.
62. Pillai, C. K. S.; Paul, W.; Sharma, C. P. Progress in Polymer Science 2009, 7, 641.
63. Badawy, M. E. I.; Rabea, E. I. International Journal of Carbohydrate Chemistry
2011, 460381 (29 pp.).
64. Desbrieres, J.; Hirrien, M.; Ross-Murphy, S. B. Polymer 2000, 7, 2451.
65. Kobayashi, K.; Huang, C. I.; Lodge, T. P. Macromolecules 1999, 21, 7070.
66. Volpert, E.; Selb, J.; Candau, F.; Green, N.; Argillier, J. F.; Audibert, A. Langmuir
1998, 7, 1870.
67. Davison, P.; Mentzer, E. SPE 1982, SPE-9300-PA.
68. Dentini, M.; Crescenzi, V.; Blasi, D. Int. J. Biol. Macromol. 1984, 2, 93.
69. Holzwarth, G. Biochemistry 1976, 19, 4333.
70. Morris, E. R.; Rees, D. A.; Young, G.; Walkinshaw, M. D.; Darke, A. J. Mol. Biol.
1977, 1, 1.
71. Norton, I. T.; Goodall, D. M.; Frangou, S. A.; Morris, E. R.; Rees, D. A. J. Mol. Biol.
1984, 3, 371.
72. Parker, W. O.; Lezzi, A. Polymer 1993, 23, 4913.
73. Jones, D. M.; Walters, K. Rheologica Acta 1989, 6, 482.
205
Outlook on the application of branched (co)polymers in EOR
Appendix 8.A
8.A.1. Experimental section
Chemicals. N,N-dimethylacrylamide (DMA, ≥99%), copper(I) bromide
(CuBr, 98%), copper(I) chloride (CuCl, 98%), methyl 2-chloropropionate
(MeClPr, 97%), sodium chloride (NaCl, ≥99.5%), glacial acetic acid, ethanol,
and diethyl ether were purchased from Sigma Aldrich. Calcium chloride
dihydrate (CaCl2 · 2 H2O, 99%) and sodium bicarbonate (NaHCO3, ≥99%)
were purchased from Merck. CuBr & CuCl were purified by stirring in glacial
acetic acid for at least 5 hours, filtering, and washing with glacial acetic acid,
ethanol and diethyl ether (in that order) and then dried at reduced
pressure.46 All the other chemicals were reagent grade and used without
further purification.
Linear polymerization. A 250-mL three-necked flask was charged with
demineralized water and DMA. Subsequently, the mixture was degassed by
three freeze-pump-thaw cycles. A nitrogen atmosphere was maintained
throughout the remainder of the reaction steps. CuCl and the ligand
(Me6TREN) were then added to the flask and the mixture was stirred for 10
minutes. The flask was then placed in an oil bath at 25 °C. The reaction was
started by the addition of the initiator (MeClPr) using a syringe.
After the
pre-set reaction time, the mixture was exposed to air and milli-Q water was
added. The contents were then purified via dialysis using membrane tubing
Spectra/Por® Dialysis Membrane (molecular weight cut off [MWCO] = 12 000
– 14 000 g/mol). The product was then dried in an oven at 65 °C until
constant weight and then ground.
Macroinitiators. The PK30 functionalization was performed according
(Scheme 8.A.1) to the published method. The reactions were performed in a
sealed 250 ml round bottom glass reactor with a reflux condenser, a U-type
anchor impeller, and an oil bath for heating.
For the preparation of PK30-Cl12 (taken here as an example) 3chloropropylamine hydrochloride (9.89 g, 53.6 mmol) was dissolved in
methanol (90 ml) to which an equimolar amount of sodium hydroxide (2.15
g, 53.6 mmol) was added. After the polyketone (10 g, 76 mmol of dicarbonyl
units) was
preheated
to
the
liquid
state
at the
employed
reaction
temperature (100 °C), the amine was added drop wise (with a drop funnel)
into the reactor in the first 20 min. The stirring speed was set at a constant
value of 500 RPM. During the reaction, the mixture of the reactants changed
from the slight yellowish, low viscous state, into a highly viscous brown
homogeneous paste. The product was dissolved in chloroform and afterwards
washed with demineralized water. The two phases (organic & water) were
206
Chapter 8
separated in a separatory funnel. The polymer was isolated by evaporating
the chloroform at reduced pressure at room temperature. The product, a
brown viscous paste (low degree of functionalization) or a brown powder
(high degree of functionalization), was finally freeze dried and stored at -18
°C until further use. Some properties of the macro-initiators are given in
Table 1. The macro-initiators were characterized using elemental analysis and
1
H-NMR spectroscopy (in chloroform).
Scheme 8.A.1: Synthesis of the macro-initiators
The conversion of carbonyl groups of the polyketone was determined using
the following formula:
(8.A.1)
, is the average number of carbons in n-m (see Scheme 8.2)
, is the average number of carbons in m (see Scheme 8.2)
molecular weight of nitrogen
molecular weight of carbon
The number of pyrrole units was determined using the conversion of the
carbonyl groups of the polyketone and formula 2:
(8.A.2)
= the average molecular weight of the parent (unmodified) polyketone
= the average molecular weight of the repeating unit of polyketone
207
Outlook on the application of branched (co)polymers in EOR
Comb polymerization. A 250-mL three-necked flask was charged with the
macro-initiator. Enough acetone (typically 5-10 ml) was added to dissolve the
macro-initiator. Demineralized water and DMA were then added to the
solution. Subsequently, the mixture was degassed by three freeze-pumpthaw cycles. A nitrogen atmosphere was maintained
throughout the
remainder of the reaction steps. CuBr was then added to the flask and the
mixture stirred for 10 minutes. The flask was then placed in an oil bath at 25
°C. The reaction was started by the addition of the ligand (Me6TREN) using a
syringe. After the pre-set reaction time, the mixture was exposed to air and
the mixture was diluted with demineralized water. The reaction mixture was
then purified via dialysis using membrane tubing Spectra/Por® Dialysis
Membrane (molecular weight cut off [MWCO] = 12 000 – 14 000 g/mol). The
product was then dried in an oven at 65 °C until constant weight and then
grounded.
Characterization. The DMA conversion was measured by using Gas
Chromatography (GC). The samples were injected on a Hewlett Packard 5890
GC with an Elite-Wax ETR column with pentadecane as an internal standard.
The total molecular weight (Mn,tot) is calculated using the DMA conversion
(monomer-initiator ratio multiply by the conversion). The span molecular
weight (Mn,SPAN) is calculated using the Mn,tot and is defined as two times the
molecular weight of one arm plus the molecular weight of the macro-initiator
(comb PDMA).
Gel permeation chromatography (GPC) analysis of one (entry 6, Mn,th
falls in the range of the calibration curve of the GPC while the Mn,th of the rest
of the entries are all higher than the range) of the water-soluble samples was
performed on a Agilent 1200 system with Polymer Standard Service (PSS)
columns (guard, 104 and 103 Å) with a 50 mM NaNO3 aqueous solution as
the eluent. The columns were operated at 40 °C with a flow-rate of 1.00
ml/min, and a refractive index (RI) detector (Agilent 1200) was used at 40
°C. The apparent molecular weights and dispersities were determined using a
polyacrylamide (PAM) based calibration with WinGPC software (PSS).
Carbon (13) nuclear magnetic resonance (13C-NMR) spectroscopy was
performed on a Varian Mercury Plus 500 MHz spectrometer. For analysis D 2O
was used as the solvent. The delay time was set at 2s and at least 10000
scans were performed (overnight). The polymer samples were swelled for 1
day and stirred for another day at room temperature. In order to obtain a
high signal to noise ratio, a high polymer concentration was used. The
hydrolysis degree was determined through the integration method reported
in literature.72
208
Chapter 8
Rheological characterization. The aqueous polymeric solutions were
prepared by swelling the polymers in water for one day and afterwards gently
stirring the solution for another day. Viscometric measurements were
performed on a HAAKE Mars III (ThermoScientific) rheometer, equipped with
a cone-and-plate geometry (diameter 60 mm, angle 2°). Flow curves were
measured by increasing the shear stress by regular steps and waiting for
equilibrium at each step. The shear rate ( ) was varied between 0.1 – 1750
s-1. Dynamic measurements were performed with frequencies ranging
between 0.04 – 100 rad/s (i.e., 6.37·10-3 – 15.92 Hz). It must be noted that
all the dynamic measurements were preceded by an oscillation stress sweep
to identify the linear viscoelastic response of each sample and to ensure that
the dynamic measurements were conducted in the linear response region of
the samples.
Alkaline hydrolysis. Stock solutions of the different polymers were
prepared by swelling the polymers for a day in the alkali-salt mixtures and
gently stirring for another day. The polymer concentration was set at 5000
ppm. NaHCO3 was used as the alkali agent and the concentration was fixed at
3000 ppm. One solution further contained 5000 ppm NaCl and the other one
contained 4925 ppm NaCl and 75 ppm CaCl 2. The solutions were divided into
8 different vials (sealed) and placed in an oven at 70 °C. At set time intervals
a sample vial was removed from the oven and cooled to room temperature.
The viscosity function of the sample was then recorded. Both solutions were
evaluated for a total of 192 hours. The viscosity retention was evaluated for
the samples using equation 8.A.3:
(8.A.3)
where
= the solution viscosity (measured at
polymer samples and
= 10 s-1) of the virgin
= the solution viscosity (measured at
= 10 s-1) of
the polymer sample treated for the specified number of days.
Four other solutions were prepared in order to evaluate for longer
periods. Two sets of comparison were performed. In the first one, the
polymer concentration (in terms of monomeric moles) was set equal between
the different polymers and the other one the solution viscosity (at = 10 s-1)
was kept equal. Again NaHCO3 was used as the alkali agent and the
concentration was fixed at 3000 ppm. One solution further contained 10000
ppm NaCl and the other one contained 9850 ppm NaCl and 150 ppm CaCl 2.
The solutions were divided into 8 different vials (sealed) and placed in an
oven at 70 °C. At set time intervals a sample vial was removed from the
209
Outlook on the application of branched (co)polymers in EOR
oven and cooled to room temperature. The viscosity function of the sample
was then recorded. Both solutions were evaluated for more than 63 days.
210
Chapter 8
Appendix 8.B
8.B.1. Experimental section
Chemicals. Cyclo octane (≥ 99,5 %), guar gum, methyl cellulose,
xanthan gum and sodium carboxy methyl cellulose were purchased from
Sigma Aldrich. The crude oil is a medium oil (API gravity equals 27.8) and
originates from the Berkel oil field in the southwest of the Netherlands. The
viscosity of the oil is 71 mPa.s at 20 °C. The branched non-ionic water
soluble (co)polymers used in the flow cell were previously synthesized using
atomic transfer radical polymerization (Chapters 6, and section 8.4).
Solution preparation. The polymeric solutions were prepared by
swelling the polymers for at least 12 hours in demineralized water and
subsequently stirred for another 12 hours.
Rheological
characterization.
Viscometric
measurements
were
performed on a HAAKE Mars III (ThermoScientific) rheometer, equipped with
a cone-and-plate geometry (diameter 60 mm, angle 2°). Flow curves were
measured by increasing the shear stress by regular steps and waiting for
equilibrium at each step. The shear rate was varied between 0.1 – 1750 s-1.
Dynamic measurements were performed with frequencies ranging between
0.04 – 100 rad/s (i.e. 6.37·10-3 – 15.92 Hz). It must be noted that all the
dynamic measurements were preceded by an oscillation stress sweep to
identify the linear viscoelastic response of each sample and to ensure that
the dynamic measurements were conducted in the linear response region of
the samples.
Flow-cell experiments. A schematic presentation of the flow-cell (with
the dimensions) is given in Figure 7.2. The flow cell has been adapted from
the original ones presented in literature73 to resemble dead ends (Figure 1.4)
that are present in oil reservoirs. The bottom part of the flow-cell is made out
of aluminum while the cover is glass. The depth of the chamber (designated
as blue in Figure 7.2) is set at 0.5 mm. The chamber is first filled with oil and
afterwards flooded with water or polymer solutions. For the branched PDMA
crude oil was used, and for the copolymers of AM and NIPAM a 1-2
(volume%) mixture of cyclo-octane and crude oil was used ( = 17 mPa.s).
The linear velocity was set at 1 foot per day. Each flood (either water or
polymer) was continued for at 24 hours at room temperature (RT). The oil
recovery out of the different cells was visually determined by taking high
definition pictures (before and after the floods). Analysis (pixel count) of the
image using Adobe allows the calculation of the amount of oil left behind in
the flow-cell.
211
Outlook on the application of branched (co)polymers in EOR
This page intentionally left blank
212
Summary
Summary
The increase in the world population and the corresponding raise of the living
standards in the developing countries will lead to a significant increase in the
demand for energy. The world energy consumption is set to increase by 34%
between 2015 and 2035. This increase will exert a significant pressure on
exploiting the current resources more efficiently. Of the total world energy
consumption, more than 80% comprises fossil fuel resources (coal/peat,
natural gas and crude oil) with crude oil accounting for a little over 30%. The
current oil recovery rate is at most 50% for light oils, being defined by the
American Petroleum Institute (API) as having an API value higher than 31.
For medium oil (API < 27) the rate drops to 30% and for heavy oils (API <
22) and tar sands other methods of recovery have to be used.
Enhanced oil recovery (EOR) aims at recovering another part of the oil
that remains behind in a reservoir after conventional, i.e. primary and
secondary, methods have been exhausted. Many different EOR methods
exist, but the focus of this thesis lies on the chemical methods where a water
soluble polymer is used. The purpose of using polymers is to improve the
displacement efficiency of the water phase that is injected into a reservoir.
Currently, partially hydrolyzed polyacrylamide (HPAM) is the choice for
almost all the projects. HPAM (Figure S.1) is a high molecular weight linear
charged polymer.
Figure S.1: Chemical structure of HPAM
The ability of HPAM to increase the viscosity of a water solution originates
from its high molecular weight and chain expansion due to the presence of
charges along the backbone. Another method for increasing the thickening
capability of a polymer is the introduction of hydrophobic groups (Chapter 1).
The rheological properties are significantly affected by this addition, and can
be beneficial in oil recovery. Nevertheless, the molecular architecture of the
main part of the polymer is still a linear chain based on acrylamide (AM). In
this thesis the focus is to investigate whether the molecular architecture can
be used as a tool to control the rheological properties of water solutions. This
required the controlled polymerization of AM. This was accomplished through
213
Summary
the atomic transfer radical polymerization (ATRP) of AM in water at room
temperature (Chapter 2, Figure S.2).
Figure S.2: Controlled polymerization of AM in water
Water soluble methyl 2-chloropropionate was used as the initiator, and a
complex of tris[2-(dimethylamino)ethyl]-amine (Me6TREN) and copper
halogenide (CuX) as the catalyst system. Linear polyacrylamides (PAM) with
molecular weights higher than 1.5 · 105 g/mol and dispersities as low as 1.39
were successfully synthesized. The living nature of the polymerization was
demonstrated by chain extension experiments and the polymerization of a
second block of N-isopropylacrylamide (NIPAM) on a PAM macroinitiator.
With this advent of controlled polymerization of AM more complex
structures such as star and comb-like PAM can be envisaged. Controlled
synthesis of star and comb-shaped branched PAM was accomplished in
Chapter 3. They were synthesized in water at room temperature. Star-like
PAMs were prepared using a commercial initiator, while comb-shaped PAM
polymers were prepared by starting each targeted polymerization with a
novel multi-functional macroinitiator based on alternating aliphatic
polyketones. The rheological properties of aqueous solutions of these PAMs
with their different molecular architectures demonstrated the importance of
branching (Figure S.3).
Figure S.3: Synthesis and the effect of branching on solution rheology
Control over the solution viscosity can be obtained by tailoring the molecular
architecture of the polymers. In Chapter 4, we present the dedicated
molecular design of PAMs as a novel pathway to manipulate the rheological
properties of their aqueous solutions. Comb-shaped branched PAMs were
214
Summary
prepared through ATRP of AM with water as the solvent. The polymers were
prepared by starting each targeted polymerization with the novel multifunctional macroinitiator (based on aliphatic polyketones, Chapter 3). The
number and length of the arms were varied and the rheological properties of
the PAM solutions were investigated (Figure S.4).
Figure S.4: Synthesis of different comb-shaped PAM and manipulation of the
rheological properties through molecular design
It was shown that both the viscosity and the elastic response of the solutions
can be manipulated by tailoring the molecular architecture of the polymers,
i.e. both properties can be steered by the number and length of the arms in
the branched PAM.
Since most oil reservoir possess high temperatures (T > 50 °C)
thermosensitive polymers are good candidates. In Chapter 5 the synthesis of
block copolymers PAM-b-PNIPAM characterized by different ratios between
the lengths of the two blocks is described. The solution properties
demonstrate that the incorporation of NIPAM units will lead to
thermoresponsive behavior. The ratio between the lengths of the two blocks
determines the rheological and surface properties. Increasing the length of
the AM block leads to higher critical micelle concentrations (CMC) but the
surface tension of the solutions approaches the value of the pure PNIPAM,
albeit at different CMCs. In addition, the “dilution” of the block copolymer
with AM does not influence the lower critical solution temperature (LCST) of
the block copolymers. A clear correlation exists between the solubility
parameter and CMC (Figure S.5).
Figure S.5: Correlation between the solubility parameter and the CMC
215
Summary
The thermoresponsive nature of the linear block copolymers was utilized in
designing branched thermo-thickening PAMs in Chapter 6. Comb-shaped
copolymers of AM and NIPAM were synthesized in water at room
temperature. Block copolymers were prepared, where the number of arms
and the length of the blocks were varied. In addition, random copolymers
were prepared where again the number of arms and the molar incorporation
ratio of AM and NIPAM were varied. The rheological properties in semi-dilute
aqueous solutions were investigated as a function of the temperature. The
block copolymers precipitate upon heating to above the LCST of the NIPAM
homopolymer while the random copolymers do not (Figure S.6).
Figure S.6: Thermoresponsive block and random copolymers based on AM and NIPAM
Interestingly, the random branched copolymers display thermo-thickening
behavior only at low shear forces. This is the first report presenting thermothickening behavior of copolymers of AM and NIPAM only at low deformation
forces. In EOR, where the thermo-thickening behavior can be beneficial, it is
important that the thermo-thickening behavior only arises deep inside the
reservoir (low shear rates) and not close to the injection wells (high shear
rates).
The potential of the branched PAMs in improving oil recovery was
evaluated in Chapter 7. The injectivity characteristics of the branched PAMs
were evaluated using filtration tests and sandstone plugs. Higher residual
resistant factors (RRF) and adsorbed polymer layer thickness were observed
for the branched PAMs compared to their linear analogues (and commercial
HPAM). The oil recovery in a 2D flow-cell, low permeable Berea and high
permeable Bentheim sandstone cores was investigated. The higher oil
recovery in the 2D flow-cell by the branched PAMs appears to be caused by
their more pronounced elastic response. In addition, the oil recovery in the
Berea and Bentheim sandstone cores is also improved by using the branched
PAMs. This appeared to be caused by the higher thickness of the adsorbed
polymer layer which led to a higher pressure drop during the polymer flood in
the low permeable Berea sandstone cores.
Chapter 8 presents an overview of the progress booked in this
dissertation. The capability of PAM to increase the solution viscosity can be
216
Summary
improved by different techniques. In this thesis, the introduction of branches
was suggested as a new approach (Figure S.7).
Figure S.7: Novel approach to enhance the thickening capability of PAM
In addition, the solution properties of the branched PAMs were investigated in
brine (salt water). The rheological properties of the branched PAMs in brine
are significantly improved compared to commercial HPAM. Improvement in
the hydrolysis resistance, without jeopardizing the thickening capability, was
accomplished by the utilization of N-substituted derivatives of AM, i.e. N,Ndimethylacrylamide. The oil recovery in the 2D flow-cell using branched
poly(N,N-dimethylacrylamide) indicated similar efficiencies compared to
branched PAM.
Preliminary results on oil recovery in the 2D flow-cell using the thermothickening comb-shaped PAMs developed in Chapter 6 demonstrated their
potential for application in EOR at high temperatures (T ≥ 70 °C).
Additionally, in Chapter 8, a preliminary evaluation of the rheological
properties of different biopolymers is also provided.
217
Summary
This page intentionally left blank
218
Samenvatting
Samenvatting
De toename van de wereldbevolking en de daarmee samenhangende
verhoging van de levensstandaard in derde wereld landen zal de vraag naar
energie enorm doen toenemen. Het wereldwijde energieverbruik zal tussen
2015 en 2035 met 34% stijgen. Deze toename zal een significante druk
zetten op het exploiteren van bestaande bronnen op een meer efficiënte
manier. Meer dan 80% van het huidige wereldwijde energieverbruik bestaat
uit fossiele energiebronnen (kool/veen, aardgas en aardolie), waarvan
aardolie 30% omvat. De huidige oliewinning in gemiddeld percentage voor
één bron is hoogstens 50% voor lichte olie, welke gedefinieerd is door de
Amerikaanse Petroleum Instituut (API) als olie met een API waarde hoger
dan 31. Voor middelmatige olie (API < 27) daalt het percentage tot 30% en
voor zware olie (API < 22) en teerzanden moeten andere winningsmethoden
gebruikt worden.
Het doel van verbeterde olie extractie (EOR) is om nog een deel van de
olie die achterblijft te winnen nadat conventionele, d.w.z. primaire en
secundaire, methodes uitgeput zijn. Er bestaan verschillende EOR methodes,
maar deze dissertatie richt zich op de chemische methodes waarin een water
oplosbare polymeer wordt gebruikt. Het doel van het gebruik van polymeren
is om de verplaatsingsefficiëntie te verbeteren van de water fase die
geïnjecteerd wordt in een reservoir.
Momenteel is gedeeltelijk gehydrolyseerde polyacrylamide (HPAM) de
keus voor bijna alle projecten. HPAM (Figuur S.1) is een lineair geladen
polymeer met een hoog moleculair gewicht.
Figuur S.1: Chemische structuur van HPAM
Het vermogen van HPAM om de viscositeit van een waterige oplossing te
verhogen komt voort uit het hoge molecuul gewicht en het ontvouwen van de
polymeerketens door de aanwezigheid van geladen groepen.
Een andere methode om viscositeitsverhoging te bereiken door middel
van een polymeer is de toevoeging van hydrofobe groepen (Hoofdstuk 1). De
reologische eigenschappen worden sterk beïnvloed door deze toevoeging, en
dit kan gunstig zijn voor oliewinning. Toch is de moleculaire architectuur van
het hoofddeel van het polymeer nog steeds een lineaire keten gebaseerd op
219
Samenvatting
acrylamide (AM). Het doel van het onderzoek beschreven in deze dissertatie
is om te onderzoeken of de moleculaire architectuur gebruikt kan worden als
een gereedschap om controle uit te oefenen op de reologische eigenschappen
van waterige polymeeroplossingen. Dit vergt de gecontroleerde polymerisatie
van AM. Dit is bereikt door middel van levende radicale polymerisatie (Atomic
Transfer Radical Polymerization; ATRP) van AM in water (Hoofdstuk 2, Figuur
S.2).
Figuur S.2: Gecontroleerde polymerisatie van AM in water
Water oplosbaar methyl 2-chloorpropionaat is gebruikt als de initiator, en een
complex van tris[2-(dimethylamino)ethyl]-amine (Me6TREN) en koper
halogeen (CuX) als het katalysator systeem. Lineair polyacrylamides (PAM)
met molecuul gewichten hoger dan 1.5 · 10 5 g/mol en dispersiteiten van 1.39
werden gesynthetiseerd. Het levende karakter van de polymerisatie werd
aangetoond door keten extensie experimenten en de polymerisatie van een
tweede blok van N-isopropylacrylamide (NIPAM) op een PAM macroinitiator.
Door gebruik te maken van deze gecontroleerde polymerisatie van AM
zijn meer complexe structuren denkbaar zoals ster en kamvormige PAM.
Deze gecontroleerde syntheses zijn beschreven in Hoofdstuk 3. Sterachtige
PAMs zijn gesynthetiseerd met behulp van een commerciële initiator, terwijl
kamvormige PAM polymeren zijn gemaakt door elke gerichte polymerisatie te
starten met een nieuwe multifunctionele macroinitiator gebaseerd op
alternerend alifatisch polyketon.
De reologische eigenschappen van waterige oplossingen van deze PAMs
met verschillende moleculaire architecturen toonden het belang aan van
vertakkingen (Figuur S.3).
Figuur S.3: Synthese en het effect van vertakkingen op oplossingsreologie
220
Samenvatting
Door het afstemmen van de moleculaire architectuur van de polymeren bleek
het mogelijk om de viscositeit van de waterige oplossing te controleren. In
Hoofdstuk 4 presenteren wij het gerichte ontwerp van PAMs als een nieuwe
route om de reologische eigenschappen van hun oplossing te manipuleren.
Kamachtig vertakte PAMs werden verkregen door middel van ATRP van AM.
De polymeren werden gesynthetiseerd door de polymerisaties te beginnen
met een nieuwe specifieke multifunctionele macroinitiator (gebaseerd op
alifatische polyketon, Hoofdstuk 3). Het aantal en de lengte van de armen
werd gevarieerd en de reologische eigenschappen van de PAM oplossingen
werden onderzocht (Figuur S.4).
Figuur S.4: Synthese van verschillende kamachtige PAM en de manipulatie van de
reologische eigenschappen door middel van moleculair ontwerp
Er is aangetoond dat zowel de viscositeit als de elasticiteit van de oplossingen
gemanipuleerd kunnen worden door middel van de moleculaire architectuur,
d.w.z. beide eigenschappen kunnen gestuurd worden door het aantal en de
lengte van de armen in de vertakte PAM.
Aangezien de meeste olie reservoirs hoge temperaturen hebben (T > 50
°C), zijn warmtegevoelige polymeren ook goede kandidaten. In Hoofdstuk 5
wordt de synthese van blok copolymeren PAM-b-PNIPAM, gekenmerkt door
verschillende verhoudingen tussen de lengte van de twee blokken,
beschreven (Figuur S.5).
Figuur S.5: Correlatie tussen de oplosbaarheidsparameter en de CMC
Oplossingseigenschappen tonen aan dat de toevoeging van NIPAM eenheden
leidt tot warmte responsief gedrag. De verhouding tussen de lengte van de
221
Samenvatting
twee blokken bepaalt de reologische en de oppervlakte eigenschappen. Een
verhoging in de lengte van de AM blok leidt tot hogere kritische micel
concentraties (CMC) maar de oppervlaktespanning van de oplossingen
bereikt de waarde voor puur PNIPAM, zij het bij verschillende CMCs.
Daarnaast beïnvloedt de “verdunning” van het blok copolymeer met AM niet
de lage kritische oplossing temperatuur (LCST) van de blok copolymeren. Er
bestaat een duidelijke correlatie tussen de oplosbaarheidsparameter voor het
polymeer en de CMC (Figuur S.5).
Het warmte responsieve karakter van de lineaire blok copolymeren is
gebruikt bij het ontwerpen van vertakte warmtegevoelige PAMs in Hoofdstuk
6. Kamvormige blok en random co-polymeren van AM en NIPAM zijn
gesynthetiseerd, waarbij het aantal en de lengte van de blokken is
gevarieerd. Ook de verhouding tussen de AM en NIPAM eenheden is
gevarieerd. De reologische eigenschappen in “semi-dilute” waterige
oplossingen zijn onderzocht als functie van de temperatuur. De blok
copolymeren precipiteren bij het verhitten tot boven de LCST van het
homopolymeer van NIPAM terwijl de copolymeren dit gedrag niet vertoonden
(Figuur S.6).
Figuur S.6: Warmte responsieve blok en willekeurig verdeeld copolymeren gebaseerd
op AM en NIPAM
Interessant is het feit dat de copolymeren alleen bij lage afschuifspanningen
warmte verdikkend gedrag vertonen. Dit is de eerste keer dat aangetoond is
dat copolymeren van AM en NIPAM alleen bij lage vervormingskrachten
warmte verdikkend gedrag vertonen. In EOR, waar het warmte verdikkend
gedrag gunstig kan zijn, is het belangrijk dat dit alleen optreedt in het olie
reservoir (lage afschuifspanningen) en niet dicht bij de injectieputten (hoge
afschuifspanningen).
Het potentieel van de vertakte PAMs in het verbeteren van de
oliewinning is geëvalueerd in Hoofdstuk 7. Het gedrag bij de injectie van de
vertakte PAMs werd onderzocht door middel van filtratie testen en zandsteen
kernen. Het gebruik van vertakte PAMs leidde tot hogere residuale resistentie
factoren (RRF) en hogere polymeer absorptie in vergelijking tot lineaire
systemen (commerciële HPAM). Olie extractie is onderzocht door gebruik te
maken van een 2D stroom-cel, en kernen van laag permeabel Berea en hoog
222
Samenvatting
permeabel Bentheim zandsteen. De hogere oliewinning gevonden voor de
vertakte PAMs in een 2D stroom-cel lijkt veroorzaakt te worden door de
hogere elasticiteit van de oplossingen. Ook in de Berea en Bentheim
zandsteen kernen werd er meer olie gewonnen door gebruik te maken van de
vertakte PAMs. Dit wordt hoogstwaarschijnlijk veroorzaakt door de sterkere
absorptie van het polymeer wat vervolgens leidt tot een hogere drukval over
de laag permeabel Berea zandsteen kernen tijdens de injectie van het
polymeer.
In Hoofdstuk 8 is een overzicht gepresenteerd van de vooruitgang die
geboekt is gedurende dit onderzoek. De capaciteit van PAM om de viscositeit
van een water oplossing te verhogen kan op verschillende manieren
verbeterd worden. In dit proefschrift werd de toevoeging van vertakking als
nieuwe methode uitgewerkt en aangetoond. (Figuur S.7).
Figuur S.7: Nieuwe methode (E) om de verdikkingscapaciteit van PAM te verbeteren
Daarnaast werden de oplossingseigenschappen van de vertakte PAMs
onderzocht in pekel (zoutwater). De reologische eigenschappen van de
vertakte PAMs in zoutwater zijn duidelijk verbeterd in vergelijking met
commerciële HPAM. Door gebruik te maken van N-gesubstitueerde AM
derivaten (zoals N,N-dimethylacrylamide) kon de resistentie tegen de
hydrolyserende werking van alkali verbeterd worden zonder de
verdikkingscapaciteiten te veranderen. Olie extractie in een 2D stroom-cel
mbv vertakte poly(N,N-dimethylacrylamide) gaf een efficiëntie die
vergelijkbaar is met die gevonden voor vertakte PAM.
Voorlopige resultaten van de oliewinning in de 2D stroom-cel met
warmte verdikkende kamvormige PAMs (ontwikkeld in Hoofdstuk 6) toonde
hun potentiele toepassing aan in EOR bij hoge temperaturen (T ≥ 70 °C).
Daarnaast wordt er in Hoofdstuk 8 een voorlopige evaluatie van de
reologische eigenschappen van verschillende biopolymeren gegeven.
223
Samenvatting
This page intentionally left blank
224
Compilacion
COMPILACION
E crecemento di populacion mundial y e aumento corespondiente di e
standard di bida den e paisnan den desaroyo lo ocasiona un aumento
significante den e demanda pa energia. E consumo di energia mundial lo bay
aumenta cu 34 % entre 2015 y 2035. E crecemento aki lo eherce un presion
significante riba e sistema con pa explota e recursonan existente mas
eficientemente. Di e total di consumo mundial di energia, mas di 80% ta
alcapara energia fossiel (carbon, gas natural y petroleo crudo) y petroleo
crudo den un cantidad di un tiki mas cu 30%. Actualmente ta logra recobra
petroleo na un promedio di 50% pa azeta fini; locual e instituto Americano di
petroleo (API) ta defini di tin un API balor di 31. Pa loke ta azeta medio (API
< 27) e promedio ta baha te na 30% y pa azeta pisa (API < 22) y santo yena
cu asfalt mester uza otro sistema pa logra recobra eseynan.
“Enhanced Oil Recovery” (EOR) tin como meta pa recobra un otro parti di
azeta cu ta resta den e reserva despues cu sistemanan convencional,
primario y secundario, a keda totalmente explota. Ta existi hopi metodo di
EOR, pero e enfasis di e tesis aki ta cay riba e esunnan cu ta uza un polymer
cu ta los op den awa. E proposito di uza un polymer ta pa mehora e eficiencia
di e fluho di e awa cu ta inyecta den e reserva. Actualmente “partially
hydrolyzed polyacrylamide” (HPAM) ta e escogencia mas uza pa tur proyecto.
HPAM (Figura S.1) ta un polymer linea cu un peso molecular halto y carga cu
coriente.
Figura S.1: Structura kimico di HPAM
E habilidad di HPAM pa aumenta e viscosidad di un solucion a base di awa ta
origina fei e peso molecular halto y e expansion di e cadena pa motibo cu tin
carga electrico na su base (lomba). Un otro metodo pa aumenta e capacidad
di un polymer pa haci e solucion mas diki ta pa introduci gruponan
hydrofobico (Capitulo 1). E propiedadnan rheological ta keda afecta
significantemente pa e agregacion aki, y por ta beneficioso den recobra
azeta. No obstante cu e arkitectura molecular di e parti mas importante di e
polymer ta keda un cadena linea basa riba acrylamide (AM). Den e tesis aki e
enfoke ta pa investiga si e arkitectura molecular por keda uza como un
225
Compilacion
artefacto pa controla e propiedadnan rheological di solucionnan di awa. Esaki
ta rekeri un polymerisacion controla di AM. A logra esaki pa medio di e
polymerisacion “Atomic Transfer Radical Polymerization” (ATRP) di AM den
awa na un temperatura di ambiente (Capitulo 2, Figura S.2).
Figura S.2: Polymerisacion controla di AM den awa
A uza e componente cu ta disolve den awa “methyl 2chloropropionate” como
e iniciado mas un compleho di “tris[2-(dimethylamino)ethyl]-amine”
(Me6TREN) y koper halogemide (CuX) como e sistema catalisado.
“Polyacrylamides” (PAM) den linea conteniendo moleculenan cu un peso
molecular mas grandi cu 1.5 · 105 g/mol y distribucion mas abou cu 1.39 a
keda sintetisa cu exito. E naturalesa bibo di polymerisacion a keda demostra
cu experimentonan di extension di e cadenanan y e polymerisacion di un di
dos bloki di “N-isopropylacrylamide” (NIPAM) riba e base di un PAM makroiniciado.
Cu e yegada di polymerisacion controla di AM por visualisa mas structura
compleho di e PAM manera den forma di strea y di peña. Sintesis controla di
PAM den forma di strea y peña a keda realisa den Capitulo 3. A sintetisa nan
den awa na un temperatura di ambiente. E PAM den un forma di strea a keda
prepara uzando un iniciado comercial, mientras e PAM den forma di peña a
keda prepara cuminsando cada polymerisacion cu un multi-funsional makroiniciado nobo basa riba “alternating aliphatic polyketone”. E propiedadnan
rheologico di solucionnan di e PAMnan aki cu structuranan (arkitectura)
diferente ta demostra e importancia di e takinan.
Figura S.3: Sintesis y e efecto di takinan riba e rheologia den solucion
226
Compilacion
E control riba e viscosidad di e structuranan ta keda obteni dor di sigui
ahusta e arkitectura molecular di e polymernan. Den Capitulo 4 nos ta
presenta cu hopi dedicacion e diseño molecular di PAMnan como un caminda
nobo pa manipula e propiedadnan reologico di e solucionnan. E tipo di
PAMnan cu forma di peña a keda prepara pa medio di ATRP di AM den awa.
Ta prepara e polymernan door di cuminsa cada polymersacion cu e multifunctional macro-inisiado nobo (a base di e “alternating aliphatic
polyketones”, Capitulo 3). E cantidad y e grandura di e takinan a keda varia y
a investiga e propiedadnan rheoligico di e PAMnan den solucion.
Figura S.4: Sintesis di diferente PAM cu forma di un peña i manipulacion di e
propiedadnan rheologico atraves di diseño molecular
A keda demostra cu e viscosidad y e respons elastico di e solucionnan por
wordo manipula traves di enlarga e arkitectura molecular di e polymernan, cu
otro palabra tur dos medidanan por wordo guia traves di e cantidad y e
largura di e takinan di e PAM.
Como cu mayoria di reserva di petroleo (azeta) tin un temperatura halto
(T > 50ᵒ C) e polymernan sensitivo pa temperatura ta bon candidato pa
esaki. Den Capitulo 5 ta describi e sintesis di copolymernan bloki PAM-bPNIPAM caracterisa pa diferente proporcion entre e largura di e dos blokinan.
E propiedadnan di e solucion ta demostra cu dor di incorpora NIPAMunidadnan lo conduci na un comportacion cu ta responde pa cambionan den
temperatura. E corelacion di e largura di e dos blokinan ta determina e
propiedadnan rheologico y di superficie. Si bo aumente e largura di e AM
bloki bo ta haya un concentracion di micelle (CMC) mas halto pero e tension
riba e superficie di e solucion ta yega e nivel di e PNIPAM puro, aunke cu e
CMCnan ta diferente. Acerca bo por bisa cu dilucion di e copolymer bloki cu
AM no ta influencia e temperatura abou mas critico di e solucion (LCST) di e
copolymer bloki. Entre e parametro di solubilidad y e CMC tin un corelacion
bon cla (Figura S.5).
E caracter di ta sensativo pa temperatura di copolymernan bloki den
forma linea a keda utilisa pa diseña e polymernan di e PAMnan cu taki cu
tambe ta sensitivo na temperatura den Capitulo 6. E tipo di copoyimer cu un
forma di un peña di AM y NIPAM a keda sintetisa den awa na un temperatura
di ambiente. A prepara blok copolymernan, caminda e cantidad di takinan y e
227
Compilacion
largura di e blokinan ta varia. Adicionalmente di polymer cu un distribucion
accidental di e dos unidadnan a keda prepara caminda atrobe a varia e
cantidad di e takinan y e ratio molecular di AM y NIPAM. A investiga e
propiedadnan rheologico den solucionnan cu poco material na diferente
temperatura.
Figura S.5: Corelacion entre e parameter di solubilidad y e CMC
E copolymer bloki ta sali for di e solucion despues di keint’e riba e LCST di e
NIPAM homopolymer mientras cu e copolymer cu distribucion accidental no
(Figura S.6).
Figura S.6: Copolymernan cu un distribucion di bloki of accidental cu ta sensitivo na
temperatura a base di AM y NIPAM
Hopi interesante ta e hecho cu e copolymernan cu un distribucion accidental
ta desplega un conducta di haci e solucion mas diki solamente ora e forsanan
ta suak. Esaki ta e prome reportahe desplegando comportacion caminda e
copolymer di AM y NIPAM ta haci un solucion mas diki solamente bou di
forzanan swak. Den EOR, caminda e sensitividat na temperatura por ta
beneficioso, ta hopi importante cu e comportacion sosode no mas na un nivel
hopi profundo den e reserva y no pega banda di e luga di inyecta.
A evalua e potencial di e PAMnan cu taki pa mehora e proceso pa recobra
petroleo den Capitulo 7. E caracteristicanan di inyeccion di e PAMnan cu taki
a ser investiga husando testnan di filtracion y plugnan di santo. A observa
factornan residual di resistencia (RRF) y mas halto y capanan mas diki di
polymer absorba pa e polymernan di PAM cu taki compara cu e polymernan
linea (y e HPAM comercial). A investiga e recuperacion di petroleo den un 2D
228
Compilacion
“flow cell”, un Berea cu un permeabilidad abou y un Bentheim cu un
permeabilidat halto (Berea y Bentheim ta santo compacta). Ta recobra mas
petroleo den e 2D “flow cell” cu e PAMnan cu taki pa motibo di nan respons
mas elastic. Adicionalmente por mehora e recuperacion di petroleo den
piedranan di santo Berea y Bentheim huzando e PAMnan cu taki. Esaki ta
sosode aparentemente pa motibo di e capanan absorba mas diki di
polymernan cual ta percura pa un presion mas halto durante e testnan den e
piedra di santo Berea cu un permeabilidad abou.
Capitulo 8 ta duna un bista di e progreso logra den e trayecto aki. E
capacidad di PAM pa aumenta e viscosidat di e solucion por keda mehora a
traves di diferente tecnica. Den e tesis aki a sugeri e uzo di PAM cu taki como
un punto di salida nobo (Figura S.7).
Figura S.7: Un punto di salida nobo (E) pa mehora e capacidad di PAM pa hisa e
viscosidad di un solucion
Adicionalmente a studia e propiedadnan di e solucionnan di e PAMnan cu taki
den awa salo. E propiedadnan rheoligico cu e PAMnan cu taki den awa salo ta
significantemente miho compara cu PAMnan comercial y e HPAMnan
comercial. A logra mehora e resistencia pa “hydrolysis” sin cu a daña e
capacidad di e poyimernan pa hisa e viscosidat di un solucion pa medio di uzo
di un N–sustituto deriva di AM, “N,N-dimethylacrylamide” (DMA).
Recuperacion di petroleo den un 2D flow cell huando e PDMAnan cu taki ta
demostra eficiencia similar cu PAMnan cu taki.
Resultadonan preliminar den e proceso di recobra petroleo di 2D flow cell
husando e tipo di PAM cu forma di un peña cu e sensibilidad pa temperatura
desaroya den Capitulo 6 ta demostra nan potencial pa aplica den EOR na un
temperatura halto (T> 70 °C). Adicionalmente, den Capitulo 8, ta duna un
evaluacion preliminar di e propiedadhnan reologico di e diferente biopolymernan.
229
Compilacion
This page intentionally left blank
230
Acknowledgements
Acknowledgements
The journey in Groningen has come to an end, at least the one as a PhDstudent. During my four-year research project many individuals have
assisted, helped and supported me when I was in need. Here I would like to
express my most sincere gratitude to these people.
First and foremost, I would like to thank my first promotor, prof. dr. A.A.
(Ton) Broekhuis. Since the day we met to discuss a master project, he has
been a constant source for guidance, advice, and encouragement. The
greatest thing he ever did for me was his continuous believe in me as a
scientist. He significantly improved my qualities in scientific investigations
and working with different people.
To my second promoter, prof. dr. F. (Francesco) Picchioni, I express my
deepest gratitude for his tireless help and support. His enthusiasm and
inspiration have pushed me to complete the thesis on time. The
transformation from a boss to a friend, over the years, has amazed me and is
the reason why I enjoyed working with him.
The financial support of the DPI, Shell and SNF (through project nr. 716)
is greatly acknowledged. Without this financial support my PhD-project would
simply be impossible.
During my PhD I received many suggestions and feedback on the
research by different people working at DPI, Shell and SNF. I would like to
take the opportunity and thank Jan Stamhuis, Nicolas Gaillard, Marc
Gruenenfelder, Jacques Kieffer, Cedric Favero, Rien Faber, Martin Buijse,
Esther Vermolen, Ibrahim Al-Qarshubi, and Bart Wassing.
I would also like to thank the members of the reading committee, prof.
dr. ir. H.J. Heeres, prof. dr. K.U. Loos, and prof. dr. D. Vlassopoulos for
reading and evaluating my thesis. In addition I also thank them for their
valuable comments and suggestions for improvements.
My work in the lab would be impossible without the help of Anne
Appeldoorn, Marcel de Vries, and Erwin Wilbers. My sincere thanks go to you
for helping me with the different experimental set-ups. I would like to thank
you not only for helping me but also for teaching me the skills that you have.
In my opinion, we (as the Department of Chemical Engineering) are
privileged by having you guys as support for experimental work. Thank you
also for introducing me to the survival adventures. I thank Jan Henk
Marsman and Leon Rohrbach for their assistance with the analytical
equipment. I thank Hans van der Velde for the many elemental analyses. I
also thank dr. M.C.A. Stuart for the cryo-TEM analyses. I am grateful for the
help of Marya van der Duin with all the paper work and also for organizing
231
Acknowledgements
(together with the technical guys) the lovely lab field trips (lab-onions) and
Christmas lunches.
My time in Groningen was filled with fun colleagues. I would like to thank
the people in the Department of Chemical Engineering; Bilal Niazi, Patrizio
Rafa (also thanks for being my paranymph and reading/correcting my
thesis), María Jesus Ortiz Iniesta (I thank you also for being my paranymph
and reading/correcting my thesis), Sjoerd van der Knoop, Nidal Hammoud
Hassan, Claudio Toncelli, Teddy Buntara, Henk van de Bovenkamp, Jelle
Wildschut, C.B. Rasrenda, Louis Daniel, Agnes Ardiyanti, Muhammad Iqbal,
Valeriya Zarubina, Zheng Zhang, Arjan Kloekhorst, Martijn Beljaars, Eric
Benjamins, Cynthia Herder, Jan Willem Miel, Anna Piskun, Rodrigo and
Esteban Araya Hermosilla, Hans Heeres and Joost van Bennekom.
The frustrations build up during the week can at best be coped with
through the Friday afternoon drinks. I would like to thank Patrizio Raffa,
Marta Martinez, Raquel Travieso Puente (also for being my tennis buddy),
María Jesus Ortiz Iniesta and Sébastien Perdriau for making the borrels fun.
In addition, I would like to thank Mathijs Hoekstra, Maarten Sorgdrager,
Sebastiaan Wiering, Bilal Niazi, and Wolter Stam for the fun times we had
playing squash together in Squadraat team 1.
The work contained in this thesis could not have been done without the
help of my students. I would like to take the opportunity and thank Piter
Brandenburg, Sjoerd van der Kuijk, Graham Ramalho, Lorenzo Massimo
Polgar, Erik Riemsma, Herman van Niekerk, Thom Stokman, Dennis van der
Meulen, Lisselore Kolk, Maarten van der Vegte, Bernard Niemeijer, Martien
Jalink, and Lars Kloekke for their commitment to this project. The many
hours spent on discussing many facets of the research have had a significant
impact on the outcome of the project.
Finally, I would like to thank my family for their continuous support
throughout my PhD. Special thanks to Carlos Alberto Gregorio Wever for his
help in designing the cover of my thesis. I also thank my mother for helping
me translating my summary to Papiamento. Tamara Mesker-Wever, I simply
don’t have enough words to explain what you have done for me during my
years in Groningen. The unconditional love, encouragement, trust and so
forth significantly helped me in achieving the goals I set at the start. I will
forever be in debt for your support.
Diego-Armando Zacarías Wever
Groningen, September 2013
232
List of publications
List of publications
Patent
a.
D.A.Z. Wever, F. Picchioni, A.A. Broekhuis. Enhanced oil recovery using
polyacrylamides, EP2604636, 2013.
b.
D.A.Z. Wever, F. Picchioni, A.A. Broekhuis. Enhanced oil recovery using novel
polyacrylamides, WO2013087214, 2013.
Peer-reviewed journal

D.A.Z. Wever, F. Picchioni, A.A. Broekhuis. Polymers for enhanced oil recovery:
A paradigm for structure–property relationship in aqueous solution, Progress in
Polymer Science, 2011, 36, 1558-1628.

D.A.Z. Wever, P. Raffa, F. Picchioni, A.A. Broekhuis. Acrylamide homopolymers
and acrylamide-N-isopropylacrylamide block copolymers by atomic transfer
radical polymerization in water. Macromolecules, 2012, 45, 4040-4045.

D.A.Z.
Wever,
F.
Picchioni,
A.A.
Broekhuis.
Branched
polyacrylamides:
Synthesis and effect of molecular architecture on solution rheology. European
Polymer Journal, 2013, 49, 3289-3301.

D.A.Z.
Wever,
E.
Riemsma,
F.
Picchioni,
A.A.
Broekhuis.
Comb-like
thermoresponsive polymeric materials: Synthesis and effect of the (macro)
molecular structure on the solution properties. Polymer, 2013, 54, 5456-5466.

D.A.Z. Wever, G. Ramalho, F. Picchioni, A.A. Broekhuis. Acrylamide-b-Nisopropylacrylamide block copolymers: Synthesis by atomic transfer radical
polymerization in water and the effect of the hydrophilic-hydrophobic ratio on
the solution properties. Journal of Applied Polymer Science, 2013, DOI:
10.1002/app.39785.

D.A.Z. Wever, L.M. Polgar, M.C.A. Stuart, F. Picchioni, A.A.Broekhuis. Polymer
molecular architecture as tool for controlling rheological properties of aqueous
polyacrylamide solutions for enhanced oil recovery. Industrial & Engineering
Chemistry Research, 2013, DOI: 10.1021/ie403045y.

D.A.Z. Wever, F. Picchioni, A.A. Broekhuis. Comb-like polyacrylamides as
flooding agent in enhanced oil recovery. Industrial & Engineering Chemistry
Research, 2013, DOI: 10.1021/ie402526k.
233
List of publications

D.A.Z. Wever, P. Raffa, A.A. Broekhuis, F. Picchioni. Efficient molecular and
architectural design of polymers for enhanced oil recovery. 2013, to be
submitted.
Poster presentation

D.A.Z. Wever, P. Raffa, F. Picchioni, A.A. Broekhuis. Water-soluble polymers for
enhanced oil recovery. 2012, DPI annual meeting, awarded 2nd poster prize.
Other publications

R. Manurung, D.A.Z. Wever, J. Wildschut, R.H. Venderbosch, H. Hidayat, J.E.G.
van Dam, E.J. Leijenhorst, A.A. Broekhuis, H.J. Heeres. Valorisation of Jatropha
curcas L. plant parts: Nut shell conversion to fast pyrolysis oil. Food and
Bioproducts Processing, 2009, 87, 187-196.

D.A.Z. Wever, H.J. Heeres, A.A. Broekhuis. Characterization of Physic nut
(Jatropha curcas L.) shells. Biomass and Bioenergy, 2012, 37, 177-187.

D.A.Z. Wever, H.J. Heeres, A.A. Broekhuis. Investigation on the structure of
Physic nut (Jatropha curcas L.) shell: Potential as a new resource for wood
composites. 2013, to be submitted.

P. Raffa, P. Brandenburg, D.A.Z. Wever, A.A. Broekhuis, F. Picchioni.
Polystyrene-Poly(sodium methacrylate) amphiphilic block copolymers by ATRP:
effect of structure, pH and ionic strength on rheology of aqueous solutions.
Macromolecules, 2013, 46, 7106-7111.

P. Raffa, D.A.Z. Wever, F. Picchioni, A.A. Broekhuis. Polymeric surfactants:
synthesis, properties and applications. Chemical Reviews, 2013, accepted.

L.M. Polgar, D.A.Z. Wever, C. Toncelli, H. Lentzakis, A.D. Gotsis, D.
Vlassopoulos, A.A. Broekhuis, F. Picchioni. The melt rheology of a new type of
asymmetric polymer star. 2013, to be submitted.
234