University of Groningen Synthesis and evaluation of novel ......Dedicated to my beloved wife, The work in this thesis is at best captured by my wife’s words: “To find the best

University of Groningen

Synthesis and evaluation of novel linear and branched polyacrylamides for enhanced oilrecoveryWever, Diego-Armando Zacarias

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher'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 forenhanced oil recovery. s.n.

CopyrightOther 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 theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 08-03-2021

https://research.rug.nl/en/publications/synthesis-and-evaluation-of-novel-linear-and-branched-polyacrylamides-for-enhanced-oil-recovery(5f6d44f5-a430-4f6e-b215-52b07c2ff735).html

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. Non-thermal ................................................................................. 15

1.2.2. Polymer flooding ........................................................................... 16

1.3. Currently used polymers ............................................................. 19

1.3.1. Polyacrylamide (PAM) .................................................................... 19

1.3.2. Partially hydrolyzed polyacrylamide (HPAM) ..................................... 20

1.3.3. Xanthan gum ................................................................................ 24

1.4. Possible new polymers for EOR .................................................... 26

1.5. Thickening capabilities ............................................................... 32

1.6. End-use requirements for polymers in EOR ................................... 35

1.7. Aim and scope of this thesis ........................................................ 37

1.8. Acknowledgement ..................................................................... 39

1.9. 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. ATRP of acrylamide ....................................................................... 52

2.3.2. Chain extension experiment, two step ............................................. 56

2.3.3. Chain extension experiment, in situ ................................................. 58

2.3.4. 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. Introduction .............................................................................. 64



3.4. Conclusion ................................................................................ 83

3.5. Acknowledgements .................................................................... 84

3.6. References ................................................................................ 84

4. Control over the viscoelasticity of aqueous polyacrylamide

solutions by tailoring the polymer architecture .......................... 87

4.1. Introduction .............................................................................. 89



4.4. Conclusion .............................................................................. 110

4.5. Acknowledgements .................................................................. 111

4.6. 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. Synthesis of the macroinitiators .................................................... 118

5.3.2. Synthesis of the block copolymers PAM-b-PNIPAM ........................... 118

5.3.3. Solution properties of poly(AM-b-NIPAM) ....................................... 120

5.4. Conclusion .............................................................................. 129


5.6. References .............................................................................. 130

6. Branched thermoresponsive polymeric materials: Synthesis and

effect of macromolecular structure on solution properties ....... 133

6.1. Introduction ............................................................................ 134



6.3.1. Macroinitiators ............................................................................ 140

6.3.2. Synthesis of PK30-g-(PAM-b-PNIPAM) ........................................... 141

6.3.3. Synthesis of PK30-g-(PAM-co-PNIPAM) .......................................... 144

6.3.4. Solution properties of PK30-gx-(PAMY-b-PAMZ) ............................... 145

6.3.5. Solution properties of PK30-gx-(PAMY-co-PAMZ) ............................. 148

6.3.6. Surface properties ....................................................................... 151

6.4. Conclusion .............................................................................. 152


6.6. References .............................................................................. 153

7. Oil recovery using branched copolymers based on acrylamide . 157

7.1. Introduction ............................................................................ 158


7.2.1. Materials .................................................................................... 160

7.2.2. Polymer injectivity experiments .................................................... 163

7.2.3. Oil recovery ................................................................................ 164

7.2.4. Characterization .......................................................................... 165


7.3.1. Polymer injectivity ...................................................................... 166

7.3.2. Oil recovery ................................................................................ 169

7.4. Conclusion .............................................................................. 175


7.6. References .............................................................................. 176

8. Towards new polymers for enhanced oil recovery .................... 179

8.1. Introduction ............................................................................ 181

8.2. Thickening capability, comb-shaped PAM .................................... 184

8.3. Salt resistance, comb-shaped PAM ............................................. 186

8.4. 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. Thickening capability and viscoelasticity ......................................... 199

8.7. Conclusion .............................................................................. 202


8.9. References .............................................................................. 204

8.A. 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


Chapter 1

11

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.

Introduction

12

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

Coal/peat

27,3%

Other (solar, wind, geothermal, etc)

0,9%

Biofuels and waste

10%Hydro

2,3%Nuclear

5,7%

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

Chapter 1

13

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

Introduction

14

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 15-

25 % 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), = fluid flowrate (ml/s), = pressure drop (bar), = fluid

viscosity (cP), = section length (cm), = 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

Chapter 1

15

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 up on the surface of the rock (e.g. by

precipitation/adsorption), thus leading to a further decrease of 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.

Introduction

16

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)

Chapter 1

17

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 studies14-19 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

Introduction

18

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 water-

soluble polymers. Besides positively affecting solution rheology, water-soluble

polymers should withstand high salt concentration, the presence of calcium,

Chapter 1

19

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 polymers were successfully tested at higher

temperatures26, 27 while associative water-soluble polymers were tested and

showed promising results compared to traditionally used polymers.28, 29

Several studies14-18, 20, 21 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 NNDAM-

NaAMPS 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.

Introduction

20

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,

Chapter 1

21

substitution of one or both hydrogens on the amide nitrogen with alkyl

groups has been presented as a solution to the salt sensitivity of HPAM45, 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

Introduction

22

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

s-1). 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, 57, 61) in order to pass. This successive

Chapter 1

23

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 proposed64, 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, 67-70

Researchers have identified two different types67 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

Introduction

24

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 produced through

fermentation of glucose or fructose by different bacteria.71 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

Chapter 1

25

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

Introduction

26

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

Chapter 1

27

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 92, 95, 107, 112, 114-125, and coupled at one or

both ends 108, 126-136. Above a given polymer concentration (dependent on the

molecular structure) the hydrophobic groups associate, when the polymer is

dissolved in water, to form hydrophobic micro-domains (intra or

intermolecular liaisons).92, 93, 108, 109, 112, 137-140 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

Introduction

28

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 self-

aggregate 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 coworkers156 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 13C-NMR160-163 (solution or solid-state), 1H-

NMR114, 23Na-NMR164, 165, 19F-NMR166, NMR self-diffusion131, 167-170

potentiometry171-173, Static and Dynamic Laser Light Scattering133, 135, 136, 151,

168, 172-177, UV-Spectroscopy for polymers bearing chromophores92, 115, 116, 178-

Chapter 1

29

186, Small-Angle Neutron Scattering (SANS)187, Non-Radiative Energy

Transfer (NRET) studies183, 188-190, Size Exclusion Chromatography (SEC)175,

191-194 and surface tension136, 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.

Introduction

30

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 and overall chemical

composition) of the corresponding water solution.

Chapter 1

31

Table 1.1: Advantages and disadvantages of the different polymeric systems

Parameters

Polymer type

High

shear pH

Ionic

strength

High

temperature

High electrolyte

concentration Bacteria

Polyelectrolyte - + + - -- ++

Polyampholyte - ++ + + ++ ++

LCST polymer - + ++ ++

Amphiphilic +- + - ++ ++

Amphiphilic polyelectrolyte +- + + - ++ ++

Amphiphilic polyampholyte +- ++ + +- ++ ++

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,

Introduction

32

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

Chapter 1

33

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, 196 (M-H) equation relating the intrinsic

viscosity ( ) 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, = relative dielectric permittivity, = electrical

permittivity of vacuum, = Boltzmann’s constant, = temperature, = the

Debye screening length, and = 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

Introduction

34

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

9

5

2,1

0,0350,2 0,08

0,56

2

HPAMPAM

HMPAM

HEURHASE

HMPAA

HM-C

ellulo

sic

Xanthan g

um

0

2

4

6

8

10 Viscosity

Vis

cosity (

Pa

.s)

0

2

4

6

8

10

Molecular weight

Mo

lecu

lar

we

igh

t (x

10

6 g

/mo

l)

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

Chapter 1

35

architecture (i.e. presence of branches) on the thickening ability of PAM-

based 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 polyampholytes220-227 (bearing

both positive and negative charges), polyelectrolyte amphiphilic120, 161, 180, 181,

188, 228-237, zwitterionic amphiphilics238-241 or amphiphilics107, 108, 114, 137, 138, 143,

Introduction

36

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 investigated. Homopolymers of sodium-2-acrylamido-2-

Chapter 1

37

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

Introduction

38

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 reports the first successful atomic transfer radical

polymerization (ATRP) of acrylamide in water at 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, thermo-

responsive 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

Chapter 1

39

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.

Introduction

40

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-63227-

MS. 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, SPE-

122400. 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.

Chapter 1

41

49. Ohmine, I.; Tanaka, T. J. Chem. Phys. 1982, 11, 5725. 50. BenJar, P. Y.; Wu, Y. S. Polymer 1997, 10, 2557. 51. Cook, R. L.; King, H. E.; Peiffer, D. G. Macromolecules 1992, 11, 2928. 52. Martin, F. D.; Hatch M.J.; Shepitka, J. S.; Ward, J. S. SPE 1983, SPE-11786. 53. Martin, F. D. SPE 1986, SPE-12651. 54. Lewandowska, K. J. Appl. Polym. Sci. 2007, 4, 2235. 55. Hu, Y.; Wang, S.; Jamieson, A. Macromolecules 1995, 6, 1847. 56. Seright, R. S.; Seheult, M.; Talashek, T. SPE 2009, 5, SPE-115142-PA. 57. Chauveteau, G. SPE 1981, SPE-10060. 58. Ferguson, J.; Walters, K.; Wolff, C. Rheol. Acta 1990, 6, 571. 59. Choplin, L.; Sabatie, J. Rheol. Acta 1986, 6, 570. 60. 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.

61. Durst, F.; Haas, R.; Interthal, W. J. Non-Newton. Fluid 1987, 2, 169. 62. James, D. F.; Mclaren, D. R. J. Fluid Mech. 1975, AUG26, 733. 63. Magueur, A.; Moan, M.; Chauveteau, G. Chem. Eng. Commun. 1985, 1-6, 351. 64. Silberberg, A.; Mijnlief, P. F. J. Polym. Sci. Pol. Phys. 1970, 7, 1089. 65. Quadrat, O. Adv. Colloid Interf. Sci. 1985, 1, 45. 66. Ohya, S.; Matsuo, T. J. Colloid Interf. Sci. 1979, 3, 593. 67. Bradna, P.; Quadrat, O.; Dupuis, D. Colloid Polym. Sci. 1995, 5, 421. 68. Bradna, P.; Quadrat, O.; Dupuis, D. Colloid Polym. Sci. 1995, 7, 642. 69. Bradna, P.; Quadrat, O.; Titkova, L.; Dupuis, D. Acta Polym. 1997, 10, 446. 70. Quadrat, O.; Bradna, P.; Dupuis, D.; Wolff, C. Colloid Polym. Sci. 1992, 11, 1057. 71. Leela, J. K.; Sharma, G. Bioprocess Eng. 2000, 6, 687. 72. Nasr, S.; Soudi, M. R.; Haghighi, M. PJBS 2007, 17, 3010. 73. Garcia-Ochoa, F.; Santos, V. E.; Casas, J. A.; Gomez, E. Biotechnol. Adv. 2000, 7,

549. 74. Becker, A.; Katzen, F.; Puhler, A.; Ielpi, L. Appl. Microbiol. Biot. 1998, 2, 145. 75. Morris, E. R.; Rees, D. A.; Young, G.; Walkinshaw, M. D.; Darke, A. J. Mol. Biol.

1977, 1, 1. 76. Norton, I. T.; Goodall, D. M.; Frangou, S. A.; Morris, E. R.; Rees, D. A. J. Mol. Biol.

1984, 3, 371. 77. Dentini, M.; Crescenzi, V.; Blasi, D. Int. J. Biol. Macromol. 1984, 2, 93. 78. Holzwarth, G. Biochemistry 1976, 19, 4333. 79. Chen, C. S. H.; Sheppard, E. W. Polym. Eng. Sci. 1980, 7, 512. 80. Yoshida, T.; Tanner, R. D. In Bioproducts and bioprocess. Springer-Verlag: Berlin,

1993; .

81. Richardson, R. K.; Rossmurphy, S. B. Int. J. Biol. Macromol. 1987, 5, 257. 82. Kierulf, C.; Sutherland, I. W. Carbohyd. Polym. 1988, 3, 185. 83. Lambert, F.; Rinaudo, M. Polymer 1985, 10,. 84. Seright , R. S.; Henrici, B. J. SPE 1990, 1, SPE-14946-PA. 85. Wellington, S. L. Soc. Petrol. Eng. J. 1983, 6, 901. 86. Ash, S. G.; Clark-Sturman, A. J.; Calvert, R.; Nisbet, T. M. SPE 1983, SPE-12085. 87. Sutherland, I. W. J. Appl. Bacteriol. 1982, 3, 385. 88. Cadmus, M. C.; Jackson, L. K.; Burton, K. A.; Plattner, R. D.; Slodki, M. E. Appl.

Environ. Microb. 1982, 1, 5. 89. 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. 90. Hou, C. T.; Barnabe, N.; Greaney, K. J. Ind. Microbiol. 1986, 1, 31. 91. Taylor, K. C.; Nasr-El-Din, H. A. J. Petrol. Sci. Eng. 1998, 3-4, 265. 92. Hill, A.; Candau, F.; Selb, J. Macromolecules 1993, 17, 4521. 93. Annable, T.; Buscall, R.; Ettelaie, R.; Whittlestone, D. J. Rheol. 1993, 4, 695. 94. Panmai, S.; Prud'homme, R. K.; Peiffer, D. G. Colloid Surface A 1999, 1-2, 3. 95. English, R. J.; Laurer, J. H.; Spontak, R. J.; Khan, S. A. Ind. Eng. Chem. Res.

2002, 25, 6425.

Introduction

42

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.

Chapter 1

43

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.

Introduction

44

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.

Chapter 1

45

214. Roovers, J.; Graessley, W. W. Macromolecules 1981, 3, 766. 215. Roovers, J. Macromolecules 1984, 6, 1196. 216. McLeish, T. Phys. Today 2008, 8, 40. 217. Burchard, W. Branched Polymers II 1999, 113. 218. Tam, K.; Tiu, C. Colloid Polym. Sci. 1990, 10, 911. 219. Lee, S.; Kim, D. H.; Huh, C.; Pope, G. A. SPE 2009, SPE-124798. 220. McCormick, C.; Salazar, L. Polymer 1992, 21, 4617. 221. McCormick, C.; Salazar, L. J. Appl. Polym. Sci. 1993, 6, 1115. 222. McCormick, C.; Salazar, L. Macromolecules 1992, 7, 1896. 223. McCormick, C.; Johnson, C. Macromolecules 1988, 3, 694. 224. McCormick, C.; Johnson, C. Macromolecules 1988, 3, 686. 225. McCormick, C.; Johnson, C. Polymer 1990, 6, 1100. 226. Kathmann, E.; Davis, D.; McCormick, C. Macromolecules 1994, 12, 3156. 227. Peiffer, D.; Lundberg, R. Polymer 1985, 7, 1058. 228. Smith, G. L.; McCormick, C. L. Macromolecules 2001, 16, 5579. 229. Petit, F.; Iliopoulos, I.; Audebert, R.; Szonyi, S. Langmuir 1997, 16, 4229. 230. Shedge, A. S.; Lele, A. K.; Wadgaonkar, P. P.; Hourdet, D.; Pcrrin, P.;

Chassenieux, C.; Badiger, M. V. Macromol. Chem. Phys. 2005, 4, 464. 231. Tomatsu, I.; Hashidzume, A.; Yusa, S.; Morishima, Y. Macromolecules 2005, 18,

7837. 232. Noda, T.; Hashidzume, A.; Morishima, Y. Macromolecules 2001, 5, 1308. 233. Noda, T.; Hashidzume, A.; Morishima, Y. Langmuir 2001, 19, 5984. 234. Mccormick, C. L.; Hoyle, C. E.; Clark, M. D. Polymer 1992, 2, 243. 235. Zhong, C.; Luo, P.; Ye, Z.; Chen, H. Polym. Bull. 2009, 1, 79. 236. Zhuang, D. Q.; Da, J. C. A. H.; Zhang, Y. X.; Dieing, R.; Ma, L.; Haeussling, L.

Polym. Adv. Technol. 2001, 11-12, 616. 237. Chang, Y. H.; Mccormick, C. L. Macromolecules 1993, 22, 6121. 238. Armentrout, R. S.; McCormick, C. L. Macromolecules 2000, 8, 2944. 239. Armentrout, R. S.; McCormick, C. L. Macromolecules 2000, 2, 419. 240. Chang, Y. H.; Mccormick, C. L. Polymer 1994, 16, 3503. 241. Mccormick, C. L.; Middleton, J. C.; Cummins, D. F. Macromolecules 1992, 4,

1201. 242. Mccormick, C. L.; Hester, R. D.; Morgan, S. E.; Safieddine, A. M. Macromolecules

1990, 8, 2124. 243. Mccormick, C. L.; Hester, R. D.; Morgan, S. E.; Safieddine, A. M. Macromolecules

1990, 8, 2132. 244. Ringsdorf, H.; Venzmer, J.; Winnik, F. M. Macromolecules 1991, 7, 1678. 245. Shalaby, S. W.; McCormick, C. L.; Butler, G. B. ACS Symp. Ser. 1991, 159.

246. Xue, W.; Hamley, I. W.; Castelletto, V.; Olmsted, P. D. Eur. Polym. J. 2004, 1, 47.

247. Yahaya, G. O.; Ahdab, A. A.; Ali, S. A.; Abu-Sharkh, B. F.; Hamad, E. Z. Polymer 2001, 8, 3363.

248. Jayasimha Reddy, G.; Venkata Naidu, S.; Rami Reddy, A. V. Adv. Polym. Tech. 2006, 1, 41.

249. Hwang, F. S.; Hogen-Esch, T. E. Macromolecules 1995, 9, 3328. 250. Zhang, Y. -.; Da, A. -.; Butler, G. B.; Hogen-Esch, T. E. J. Polym. Sci. Pol. Chem.

1992, 7, 1383. 251. Zhang, H.; Xu, K.; Ai, H.; Chen, D.; Xv, L.; Chen, M. J. Solution Chem. 2008, 8,

1137. 252. Bastiat, G.; Grassl, B.; François, J. Polym. Int. 2002, 10, 958. 253. Lara-Ceniceros, A. C.; Rivera-Vallejo, C.; Jiménez-Regalado, E. J. Polym. Bull.

2007, 4, 499. 254. Jiménez-Regalado, E. J.; Cadenas-Pliego, G.; Pérez-Álvarez, M.; Hernández-

Valdez, Y. Polymer 2004, 6, 1993. 255. Yu Wang; Zhiyong Lu; Yugui Han; Yujun Feng; Chongli Tang Adv. Mater. Res.

2011, 654.

Introduction

46

256. Wu, Y.; Mahmoudkhani, A.; Watson, P.; Fenderson, T.; Nair, M. SPE 2012, SPE-155653-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.

Chapter 2

47

Chapter 2

Acrylamide homo- and block co-

polymers 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 2-

chloropropionate (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 macro-

initiators for the synthesis of block copolymers polyacrylamide-b-poly(N-

isopropylacrylamide).

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.

Atomic transfer radical polymerization of acrylamide

48

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 (methanol15, 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 systems24, 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, 18, such as N-hydroxyethylacrylamide, N,N-

dimethylacrylamide, N-tert-butylacrylamide and N-(2-

hydroxypropyl)methacrylamide. To the best of our knowledge, only few

publications4, 5, 11, 29 mentioned the controlled polymerization of acrylamide

using chloro-acetic acid, 2-chloropropionamide (2-Cl-PA) or 2-

bromopropionamide (2-Br-PA) as initiators and either CuCl / N,N,N,N-

tetramethylethylenediamine (TMEDA) or 2,2-bipyridine (bpy) as catalytic

systems. Although the molecular weight of the polyacrylamide increases

Chapter 2

49

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 achieved11; 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


50

ATRP of acrylamide has also been claimed in aqueous media23; 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 polyacrylamide-

b-poly(N-isopropylacrylamide) block copolymers were synthesized.

2.2. Experimental section

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. 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-pump-

thaw 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

Chapter 2

51

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 three-

necked 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 macro-

initiator 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.


52

Nuclear Magnetic Resonance (NMR) spectra were recorded on a Varian

Mercury Plus 400 MHz spectrometer. For analysis D2O 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 AM4, 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

Chapter 2

53

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

Entry [M]0:[I]0:[CuCl]0:[Me6TREN]0a M/water (wt:vol); T;

Time (min) Conv (%) Mn,th Mn,GPC PDI

1 225 : 1 : 2 : 2 1:4; 25 °C; 60 48.7 7 719 10 230

2 385 : 1 : 6.0 : 6.0 1:6; 25 °C; 60 28.2 7 717 11 900 1.40

3 385 : 1 : 1.5 : 1.5 1:4; 25 °C; 60 88.2 24 011 22 780 1.42

4 470 : 1 : 1.5 : 1.5 1:6; 25 °C; 30 69.8 23 269 22 863 1.42

5 500 : 1 : 1.5 : 1.5 1:12; 25 °C; 60 42.2 14 998 16 780 1.88

6 680 : 1 : 1.5 : 1.5 1:6; 25 °C; 90 78.3 37 901 32 680 1.56

7 870 : 1 : 1.5 : 1.5 1:6; 25 °C; 2 47.3 29 284 26 260 1.46

8 945 : 1 : 1.5 : 1.5 1:6; 25 °C; 3 36.8 24 719 25 850 1.54

9 965 : 1 : 1.5 : 1.5 1:6; 25 °C; 60 75.3 51 703 38 310 1.57

10b 1000 : 1 : 1.5 : 1.5 1:15; 25 °C; 60 52.0 28 762 41 970 1.97

11 1625 : 1 : 1.5 : 1.5 1:6; 25 °C; 60 84.7 97 833 68 370 2.04

12 2785 : 1 : 1.5c : 1.5 1:6; 25 °C; 60 58.5 115 805 75 880 2.05

13 4355 : 1 : 1.5 : 1.5 1:6; 25 °C; 60 69.1 213 852 108 800 2.30

a. Molar ratio

b. No increase in molecular weight with increase in conversion.

c. CuBr was used.

In the stationary state the conversion index (ln[ / ]) is represented by

equation 2.1.

(2.1)

where indicates the kinetic constant for propagation, the initiation rate

and 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 ( = / , where is the

activation rate constant and 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


54

the reaction. Using equation 2.2 we modeled the kinetics of the ATRP of AM

(Figure 2.2).

0 10 20 30 40 50 60

0,0

0,5

1,0

1,5

2,0

0 2 4 6 8 10 12 14 16

0,0

0,4

0,8

1,2

1,6

2,0

Ln

(M

0/M

)

Time2/3

(min2/3

)

Entry 11

Entry 4

Entry 11, R2 (model) = 0.99

Entry 4, R2 (model) = 0.92

Ln

(M

0/M

)

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.

Chapter 2

55

0 2 4 40 50 60 70 80 90 100

0

2000

12000

16000

20000

24000

28000

32000

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

Mn,GPC

Mn, theoretical

Mo

lec

ula

r w

eig

ht

(g/m

ol)

Conversion (%)

PDI

Po

lyd

isp

ers

ity

in

de

x (

PD

I)

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

1000 10000 100000

Conversion = 49,8%

Mn= 16540

PDI = 1.39

rel. R

ID in

ten

sity

Molecular weight (Mn)

Conversion = 69,8%

Mn= 22870

PDI = 1.42

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


56

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 with conversion indicates a controlled radical

polymerization37 (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).

Chapter 2

57

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.

1000 10000 100000 1000000

Conversion = 84,7%

Mn=68370

PDI = 2.04

rel. R

ID d

ete

cto

r

Molecular Weight (Mn)

Conversion = 57,1%

Mn=45810

PDI = 1.92

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(n-

butyl 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


58

water/DMF solution (1:1) using 2-Cl-PA/Me6TREN/CuCl as the

initiator/catalyst system.20

1000 10000 100000 1000000

Conversion = 14 %

Mn= 39600

PDI = 5.63

Conversion = 66 %

Mn= 23490

PDI = 1.45

rel. R

ID in

ten

sity


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

Chapter 2

59

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.

1000 10000 100000

Conversion = 25.5%

Mn = 30220

PDI = 1.64

rel. R

ID in

ten

sity


Conversion = 88.2%

Mn = 22780

PDI = 1.42

Figure 2.7: GPC traces for the in situ chain extension experiment

0246810

PAM-b-PNIPAM

washed with THF

ppm

PAM

PAM-b-PNIPAM

PAM-b-PNIPAM

DP = 330-b-35

MeOH

Methyl-groups NIPAM

Figure 2.8: NMR spectra of PAM-b-PNIPAM (virgin and THF washed) and PAM


60

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 Mn 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.




2.6. 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.

Chapter 2

61

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 A-

Polymer 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.


62


Chapter 3

63

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.

Synthesis of branched polyacrylamide

64

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 rate3 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

high degree of branching could be obtained8 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 co-

polymers 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 17 or a water-alcohol

mixture 18, 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

Chapter 3

65

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.


Chemicals. Acrylamide (AM) (electrophoresis grade, ≥99%), PAM (Mw =

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 literature17. A 250 mL three-

necked flask was charged with AM (5 g, 70 mmol). A magnetic stirrer and

distilled water were added and subsequently degassed by three freeze-pump-

thaw 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 Me6TREN (48 mg,

0.21 mmol) were added, and the mixture was stirred for 10 min. The


66

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 method25 (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),

Chapter 3

67

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-pump-

thaw 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


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.


68

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 macro-

initiator 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 1H-NMR spectrum was recorded. ChemBioDraw Ultra 12.0

(CambridgeSoft) was used to simulate the 1H-NMR spectrum of the macro-

initiator 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 Me6TREN (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-1-

Chapter 3

69

propanol as the initiator). A sample of the reaction mixture was taken and

precipitated in acetone. An acetone sample, containing 1000 ppm of 3-

chloro-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.



solvent.

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 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 macroinitiators was performed on the

EuroEA3000-CHNOS analyzer (EUROVECTOR Instruments & Software).

Approximately 2 mg of each sample is weighed and placed in tin sample-

cups. 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.


70

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.


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 1H-NMR at

=5.68 ppm and validated by using model compounds.25 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%) XCO (%)b

Pyrrole

unitsc Mn,GPC (g/mol) PDI

PK30 67.0 : 8.4 : 0 - 0 2 797 1.74

PK30-Cl12, R1 = Cl 64.2 : 7.8 : 4.6 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 1H-NMR (Figure 3.1). As can be

observed, the resonances corresponding to the pyrrole units (a) and the

Chapter 3

71

aliphatic protons of the amine moiety (b-d) appear in the spectrum of the

chemically modified polyketone.

7 6 5 4 3 2 1

d

cb

aa

PK30-virgin

ppm

PK30-Cl12

ab

c

d

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


72

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, is the propagation rate

constant, is the termination rate constant, is the monomer

concentration at time zero, is the monomer concentration at any time, and

is the initial initiator concentration.

0 10 20 30 40 50 60

0,0

0,4

0,8

1,2

1,6

2,0

Entry 1 (Table 2), R2 (model) = 0.99

Entry 14 (Table 2), R2 (model) = 0.82

ln (

M0/M

)

Time (min)

0 2 4 6 8 10 12 14 16

0,0

0,4

0,8

1,2

1,6

2,0

ln (

M0/M

)

Time2/3

(min2/3

)

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

Chapter 3

73

the PDI (and later on the block copolymerization with NIPAM) provide strong

evidence for the controlled nature of the polymerization.

0 2 4 20 30 40 50 60 70 80 90 100

0,05,0x10

1

1,0x104

1,5x104

2,0x104

2,5x104

3,0x104

Mn,GPC

Mn,theoretical

Mo

lec

ula

r w

eig

ht

(g/m

ol)

0 2 4 20 30 40 50 60 70 80 90 100

1,0

1,2

1,4

1,6

1,8

2,0

2,2

2,4

2,6

2,8

3,0

PDI

Po

lyd

isp

ers

ity

in

de

x (

PD

I)

Conversion (%)

Entry 1, Table 2

0 5 10 15 20 25 30 35 40 45 96 98 100

0,0

5,0x104

1,0x105

1,5x105

2,0x105

2,5x105

3,0x105

Mn,GPC

Mn,theoretical

Entry 14, Table 2

Mo

lec

ula

r w

eig

ht

(g/m

ol)

Conversion (%)

1,0

1,5

2,0

2,5

3,0

3,5

4,0

4,5

5,0

5,5

6,0

PDI

Po

lyd

isp

ers

ity

in

de

x (

PD

I)

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 1H-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.


74

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).

4 3 2 1

PK30-Cl12

-graft-PAMPK30-Cl12

A

BB

PK30-Cl12

-graft-PAM

ppm

PK30-Cl12

A

Figure 3.4: 1H-NMR spectra of the PK30-Cl12 (macro-initiator) and the PAM grafted

product (PK30-Cl12-graft-PAM)

Chapter 3

75

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 macro-

initiator. 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 [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

Linearf

1d 479:1:1.5:1.5 1:6; 25 °C; 60 76.6 28 623 21 100 1.47 28 623

2e 9511:1:1.5:1.5 1:3; 25 °C; 30 19.1 129 124 84 692 1.72 129 124

3 966:1:1.5:1.5 1:6; 25 °C; 60 75.3 51 703 38 310 1.57 51 703

4 1 625:1:1.5:1.5 1:6; 25 °C;120 84.7 97 833 69 100 2.18 97 833

5 4 354:1:1.5:1.5 1:6; 25 °C; 60 69.1 213 852 108 800 2.30 213 852

6 8 790:1:1.5:1.5 1:6; 25 °C; 25 59.5 371 752 131 660 3.23 371 752

7 14 399:1:1.5:1.5 1:6; 25 °C; 15 50.8 519 928 210 200 2.25 519 928

Star

8 1 965:1:6.0:6.0 1:6; 25 °C;180 77.5 108 246 79 680 2.06 54 123

9 2 884:1:6.0:6.0 1:6; 25 °C;180 76.4 156 670 107 800 1.92 78 335

10 5 811:1:6.0:6.0 1:6; 25 °C;120 62.6 258 567 216 500 2.01 129 284

Combg

11 1 197:1:1.5:1.5 1:6:1/3;25 °C; 60 77.7 66 109 72 020 2.86 13 815

12 2 395:1:1.5:1.5 1:6:3.0;25 °C; 60 74.8 127 337 104 900 2.31 24 020

13 6 006:1:1.5:1.5 1:8:1.5;25 °C; 60 72.5 309 507 206 400 2.33 54 382

14 9 003:1:1.5:1.5 1:6:1.0;25 °C; 60 47.6 304 608 188 800 1.88 53 565

15 12 025:1:1.5:1.5 1:6:1/3;25 °C; 60 68.8 587 766 271 600 1.97 100 758

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.


76

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.

5 4 3 2 1

2

macroinitiator

1440 min

480 min

360 min

240 min

ppm

120 min

MeOH

1

Figure 3.5: 1H-NMR spectra of the block copolymer at different reaction times

Chapter 3

77

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. 31-38 It was demonstrated that for

polyisoprenes31, 39, polypropylene36, 40-42, polyethylene37, 43-47 and

polystyrene35, 48, 49 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


78

0 2 4 6 8 10 12 14 16

0

10

20

30A

Vis

co

sit

y (

Pa

.s)

Concentration (wt%)

comb, entry 12

linear, entry 5

linear, entry 4

star, entry 8

0 2 4 6 8 10 12 14 16

0

20

40

60

80B

comb, entry 13

linear, entry 6

star, entry 10

linear, entry 5

Vis

co

sit

y (

Pa

.s)

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

Chapter 3

79

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 2 4 6 8 10 12 14 16

0

20

40

60

80

A

A-Zoom

linear, entry 3

star, entry 8

comb, entry 13

Vis

co

sit

y (

Pa

.s)

Concentration (wt. %)

0,0 0,5 1,0 1,5 2,0 2,5 3,0

0,00

0,01

0,02

0,03

0,04

0,05

0,06

linear, entry 3

star, entry 8

comb, entry 13

A-Zoom

Vis

co

sit

y (

Pa

.s)


0 2 4 6 8 10 12 14 16

0

10

20

30

40

50

60

linear, entry 4

star, entry 10

comb, entry 15

B

Vis

co

sit

y (

Pa

.s)

Concentration (wt. %)B-Zoom

0,0 0,5 1,0 1,5 2,0 2,5 3,0

0,00

0,10

0,20

0,30

0,40

linear, entry 4

star, entry 10

comb, entry 15

Vis

co

sit

y (

Pa

.s)


B-Zoom

Figure 3.7: Viscosity (measured at = 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


80

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 polymers52).

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, is the zero shear rate viscosity, is the

degree of polymerization, is the friction factor per segment, is the

hydrodynamic radius as determined by light scattering measurements, is

Chapter 3

81

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. .

1E24 1E25 1E26

1

10

100

1000

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Slo

pe

linear, entry 1

star, entry 6

comb, entry 11

B

A

0 (

mP

a.s

)

NpR

2

g[(N

avC)/M

w] (nm

-1)

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


82

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 weight56, 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.

0,1 1 10 10010

-7

10-6

10-5

10-4

10-3

10-2

10-1

100

101

102

103

A

G' G

" (

Pa

)

Frequency (rad/s)

} comb, entry 13

star, entry 8

linear, entry 3

}

}

= G"

= G'

0,1 1 10 1000

10

20

30

40

50

60

70

80

90

B

comb, entry 13

star, entry 8

linear, entry 3

Ph

as

e a

ng

le

Frequency (rad/s)

0,1 1 10 10010

-7

10-6

10-5

10-4

10-3

10-2

10-1

100

101

102

C

G' G

" (

Pa

)

Frequency (rad/s)

} comb, entry 12

star, entry 8

linear, entry 4

}

}

= G"

= G'

0,1 1 10 1000

10

20

30

40

50

60

70

80

90

comb, entry 12

star, entry 8

linear, entry 4

D

Ph

as

e a

ng

le

Frequency (rad/s)

Figure 3.9: G’ & G” (A) and phase angle (B) as a function of the frequency for a 4-

arm 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.%

Chapter 3

83

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 3-

arm star.50 For polyisoprene the 0 of a 3-arm star is lower than that of a 4-

arm 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 1H-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 comb-

like 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


84

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.




3.6. 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.

Chapter 3

85

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.


86


Chapter 4

87

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

Rheological properties of branched polyacrylamides

88

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 Chemistry Research, 2013, DOI:

10.1021/ie403045y.

Chapter 4

89

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


90

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 PAMs were synthesized using reversible addition-

fragmentation chain transfer (RAFT) polymerization. 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 structure-

property relationship, of the present chapter.


Chemicals. Acrylamide (AM) (electrophoresis grade, ≥99%), PAM (Mw =

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

Chapter 4

91

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 representative example), 3-

chloropropylamine 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 di-

carbonyl 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)




92



(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-pump-


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

Chapter 4

93

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).


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.










94




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, is the viscosity

at infinite shear rate, is the critical shear rate for the onset of shear

thinning, is the power law slope and represents the width between

and the power law region.


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)


(C : H : N, wt%) XCO (%)b Pyrrole unitsc Mn,GPC PDI

PK30 (virgin) 67.0 : 8.4 : 0 - 0 2 797 1.74

PK30-Cl4 58.6 : 7.1 : 1.6 18.87 4 2 447 2.02

PK30-Cl8 64.0 : 7.9 : 3.3 37.21 8 2 244 2.01

PK30-Cl12 64.2 : 7.8 : 4.6 55.10 12 2 093 1.76

PK30-Cl13 62.9 : 7.6 : 4.9 61.14 13 2 072 1.97

PK30-Cl17 73.7 : 7.9 : 6.1 81.27 17 2 117 2.18

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 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 4.2). The formation of the pyrrole units was also previously

demonstrated using model compounds.20 As can be observed in Figure 4.2,

Chapter 4

95

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 macro-

initiators in the ATRP of acrylamide for the preparation of comb-polymers

with different number of side chains.

7 6 5 4 3 2 1

c

d

b

aa

c

d

ba

PK30-Cl17

ppm

PK30, virgin

PK30-Cl4

PK30-Cl8

PK30-Cl13

d

d

d

b

b

ba

a

a

c

c

c

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


96

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


[Me6TREN]0

M/s1/s2a (w:v:v); T;

Time (min)

Conv.

(%) Mn,overall Mn,SPAN

Linear

1 14 399:1:1.5:1.5 1:6 :0 ;25 °C;15 50.8 519 928 519 928

2 50 942:1:1.5:1.5 1:6 :0 ;25 °C;25 42.1 1 524 432 1 524 423

3 57 654:1:1.5:1.5 1:6 :0 ;25 °C;60 62.0 2 540 789 2 540 789

4-arm 4 14 894:1:3.0:3.0 1:4 :1/5 ;25 °C;60 69.9 743 019 374 307

5 37 707:1:1.5:1.5 1:4 :1/10;25 °C;60 65.4 1 613 401 809 498

8-arm 6 10 037:1:3.0:3.0 1:4 :1/5 ;25 °C;60 88.1 631 116 160 576

7 49 822:1:3.0:3.0 1:4 :1/10;25 °C;60 48.8 1 730 784 435 493

13-arm

8 12 019:1:1.5:1.5 1:6 :1/3 ;25 °C;60 68.8 587 766 100 758

9b 47 610:1:1.5:1.5 1:6 :1/5 ;25 °C;60 35.7 1 208 130 204 152

10 100 050:1:3.0:3.0 1:4 :1/20;25 °C;60 23.8 1 692 550 263 189

11 149 634:1:3.0:3.0 4:15:1/20;25 °C;60 23.6 2 510 092 388 965

12 150 084:1:2.0:2.0 1:4 :1/40;25 °C;60 32.8 3 499 094 541 119

17-arm 13 149 859:1:3.0:3.0 2:15:1/20;25 °C;60 14.8 1 576 493 188 267

14 150 174:1:1.5:1.5 1:2 :1/20;25 °C;60 23.8 2 540 500 301 679

a. M/s1/s2 = Monomer / solvent 1 / solvent 2 = Acrylamide / water / acetone

b. 12-arm

Chapter 4

97

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


98

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, is the slope of the viscosity-

concentration-plot, 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).

0,0 0,4 0,8 1,2 1,6 2,0

0

1

2

3

4

A

17-arm, R2 = 0,995

13-arm, R2 = 0,996

8-arm, R2 = 0,991

4-arm, R2 = 0,994

linear, R2 = 0,999

log(

red)

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

Chapter 4

99

Linear 4-arm 8-arm 13-arm 17-arm

0

5

10

15

20

25

3017-arm, [] = 28,00 dl/g

13-arm, [] = 24,08 dl/g

8-arm, [] = 7,96 dl/g

4-arm, [] = 6,63 dl/g

linear, [] = 6,14 dl/g

Intr

insic

vis

co

sity (

[],

dl/g

)

Molecular architecture

B

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, = hydrodynamic radius, = radius of

gyration, = 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, = monomer excluded volume

parameter, = number of arms, and = 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.


100

Table 4.3: Properties of the different (co)polymers

Architecture Entry Rg, est (nm) Rh, DLS (nm) (wt.%) 5·

(wt.%)

Linear 2 88a 48 0.09 0.45

4-arm 5 72 36 0.17 0.85

8-arm 8 57 51 0.36 1.80

13-arm 10 61 60 0.44 2.20

17-arm 14 40 71 0.96 4.80

a. Taken from literature30

In order to carry out the measurements well 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

Chapter 4

101

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.

linear 4-arm 8-arm 13-arm 17-arm

10-3

10-2

10-1

100

101

102

103

104

105

0 (

Pa

.s)


Entries = 2, 5, 7, 10 & 13

Figure 4.5: 0 as a function of the molecular architecture at Mn,overall ≈ 1.6 MDa and a

polymer concentration of 5·

0,1 1 10 10010

-6

10-5

10-4

10-3

10-2

10-1

100

101

102

103

A

G' G

" (P

a)

Frequency (rad/s)

Filled symbols = G'

Empty symbols = G"

= 17-arm

= 13-arm

= 8-arm

= 4-arm

= Linear

0,1 1 10 100

0

10

20

30

40

50

60

70

80

90

B

17-arm

13-arm

8-arm

4-arm

Linear

Ph

ase

an

gle

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).


102

0 1 2 3 4 5 6 7 8 9

0

10

20

30

40

50

Vis

cosity (

Pa.s

)

Concentration (wt.%)

Linear

4-arm

8-arm

12-arm

A

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

0

2

4

6

8

10

12

14

B

Vis

cosity (

Pa.s

)


Linear

4-arm

8-arm

13-arm

17-arm

0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0

0

2

4

6

8

10

12

14

16

CV

iscosity (

Pa.s

)


Linear

13-arm

17-arm

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 “Carreau-

Yasuda” model21, 22 in order to evaluate the relaxation time ( ). In the melt,

the molecular weight35, 36, dispersity index35, 36, and polymer concentration22

(in solution) are parameters that influence . 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.

Chapter 4

103

10-2

10-1

100

101

102

103

10-1

100

101

102

A

17-arm, R2(model) = 0,9996

13-arm, R2(model) = 0,9995

linear, R2(model) = 0,9998

8-arm, R2(model) = 0,9998

4-arm, R2(model) = 0,9998

Vis

co

sity (

Pa

.s)

Shear rate (, s-1)

4-arm 8-arm Linear 13-arm 17-arm

100

101

B

17-arm, 10,89 s.

13-arm, 4,61 s.

linear, 0,70 s.

8-arm, 0,50 s.

4-arm, 0,47 s.

Rela

xa

tion t

ime

(, s)


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 (not

shown for brevity) and in line with earlier studies on branched

polyisoprenes37 and polybutadienes38, 39, which concluded that the critical

shear rate for the onset of non-Newtonian behavior is reduced upon

branching.


104

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 (at different

concentration).

Chapter 4

105

10-5

10-4

10-3

10-2

10-1

100

101

102

103

G'G

" (P

a)

= Linear

= 4-arm

= 8-arm

= 12-arm

Filled symbols = G'

Empty symbols = G"

A11

0

10

20

30

40

50

60

70

80

90

A2

Pha

se a

ngle

Linear

4-arm

8-arm

12-arm

10-3

10-2

10-1

100

101

102

B1

G'G

" (P

a)

Filled symbols = G'

Empty symbols = G"

= Linear

= 4-arm

= 8-arm

= 13-arm

= 17-arm 0

10

20

30

40

50

60

70

80

90

Linear

4-arm

8-arm

13-arm

17-arm B2

Ph

ase a

ngle

0,1 1 10 100

10-1

100

101

102 C1

G'G

" (P

a)

Frequency (rad/s)

= Linear

= 13-arm

= 17-armFilled symbols = G'

Empty symbols = G"

0,1 1 10 100

0

10

20

30

40

50

60

70

80

90

Linear

13-arm

17-arm C2

Ph

ase a

ngle

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


106

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.

10-5

10-4

10-3

10-2

10-1

100

101 A1

G'G

" (P

a)

= Linear

= 4-arm

= 8-arm

= 12-arm

Filled symbols = G'

Empty symbols = G"

0

10

20

30

40

50

60

70

80

90

A2

Linear

4-arm

8-arm

12-armP

ha

se

an

gle

100

101

102

B1

G'G

" (P

a)

Filled symbols = G'

Empty symbols = G"

= Linear

= 4-arm

= 8-arm

= 13-arm

= 17-arm 0

10

20

30

40

50

60

70

80

90

Linear

4-arm

8-arm

13-arm

17-arm B2

Ph

ase a

ngle

0,1 1 10 10010

-1

100

101

102

C1

G'G

" (P

a)

Frequency (rad/s)

Filled symbols = G'

Empty symbols = G"

= Linear

= 13-arm

= 17-arm

0,1 1 10 100

0

10

20

30

40

50

60

70

80

90

Linear

13-arm

17-arm

C2

Ph

ase a

ngle

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 )

Chapter 4

107

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 diphenyl23, 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 polymers34). 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


108

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

the arm molecular weight is observed33.

0,1 1 10 100

10-2

10-1

100

101

102

A

G"(

Pa

)

Frequency (rad/s)

: DParm

= 3875

: DParm

= 2715

: DParm

= 1830

: DParm

= 1415

: DParm

= 690

0,1 1 10 100

10-2

10-1

100

101

102

: DParm

= 3875

: DParm

= 2715

: DParm

= 1830

: DParm

= 1415

: DParm

= 690B

G'(P

a)

Frequency (rad/s)

0,1 1 10 100

10

20

30

40

50

60

70

80

90

C

Ph

ase

an

gle

Frequency (rad/s)

: DParm

= 690

: DParm

= 1415

: DParm

= 1830

: DParm

= 2715

: DParm

= 3785

500 1000 1500 2000 2500 3000 3500 400010

0

101

102

103

104

105

106

107

108

D

0 (

Pa

.s)

DParm

Exponential fit, R2 = 0,94

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

Chapter 4

109

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)


110

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 star-

like (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.

Chapter 4

111




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-63227-

MS. 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.


112

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



1979, 38.

Chapter 5

113

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 thermo-

responsiveness 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 hydrophilic-

hydrophobic ratio on the solution properties. Journal of Applied Polymer

Science, 2013, DOI: 10.1002/app.39785.

PAM-b-PNIPAM block copolymes, synthesis & properties

114

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 water10,

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 structure-

property 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

Chapter 5

115

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-b-

N-isopropylacrylamide) in water was also accomplished.22

In this paper, the controlled synthesis of the block copolymers PAM-b-

PNIPAM with varying length of the blocks is reported. First the PAM

macroinitiators are prepared and subsequently NIPAM is polymerized on the

macroinitiator as blocks (demonstrating the living character of the

polymerization). To the best of our knowledge, this has not 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.


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



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


116

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 block-

copolymers 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

Chapter 5

117

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 D2O 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 Jasco V-630 UV-Vis spectrophotometer equipped with a

temperature controlled six-position sample holder was used. 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.


118


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,th

b (g/mol) DP

PAM200 300 : 1 1:6; 25 °C; 60 68 14 450 200

PAM235 300 : 1 1:6; 25 °C; 60 78 16 660 235

PAM460 680 : 1 1:6; 25 °C; 90 78 37 900 460

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

Chapter 5

119

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 Mn,1

H-NMR DP

NIPAM

DP

PAM

b

(J1/2·cm-3/2)

PAM200-b-PNIPAM30 55 : 1 1:6; 25 °C; 60 57 17 600 30 200 27.8

PAM200-b-PNIPAM70 140 : 1 1:6; 25 °C; 60 50 22 150 70 200 26.5

PAM200-b-PNIPAM70 275 : 1 1:6; 25 °C; 60 26c 22 150 70 200 26.5

PAM200-b-PNIPAM90 140 : 1 1:6; 25 °C; 60 66 24 400 90 200 25.3

PAM200-b-PNIPAM155 270 : 1 1:6; 25 °C; 60 57 31 750 155 200 24.2

PAM200-b-PNIPAM185 550 : 1 1:6; 25 °C; 60 34 35 150 185 200 23.3

PAM200-b-PNIPAM650 1115 : 1 1:6; 25 °C; 60 59 87 750 650 200 22.7

PAM235-b-PNIPAM125 2495 : 1 1:6; 25 °C; 160 5 30 850 125 235 26.7

PAM460-b-PNIPAM10 750 : 1 1:6; 25 °C; 60 1 33 800 10 460 29.0

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

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-b-

PNIPAM125, 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).

5 4 3 2 1

21

1

2

22

2

2

PAM200-b-PNIPAM650

PAM200-b-PNIPAM185

PAM200-b-PNIPAM155

PAM200-b-PNIPAM90

PAM200-b-PNIPAM70

PAM200-b-PNIPAM30

ppm

PAM200

2

D2O

Figure 5.1: 1H-NMR spectra of the block copolymers PAM200-b-PNIPAM(Y) and the

parent macroinitiator


120

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

Chapter 5

121

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.

1 10 100 1000

1

10

100

PAM200-b-PNIPAM30

PAM200-b-PNIPAM70PAM200-b-PNIPAM90

PAM200-b-PNIPAM155

PAM200-b-PNIPAM185

PAM200-b-PNIPAM650

Vis

cosity (

mP

a.s

)

Shear rate (, s-1)

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-b-

PNIPAM10 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


122

([]).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).

1 10 100 1000

10

100

PAM460

Vis

cosity (

mP

a.s

)

Shear rate (, s-1)

PAM200-b-PNIPAM185

PAM235-b-PNIPAM125

PAM460-b-PNIPAM10

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.%)

0,0 0,3 0,6 0,9 1,2 1,5 1,8 2,1 2,4

0

1

2

3

4

5

6

7PAM200-b-PNIPAM185, R

2= 0.98 [] = 1.13 dl/g

PAM235-b-PNIPAM125, R2= 0.99 [] = 0.93 dl/g

PAM530-b-PNIPAM10, R2= 0.98 [] = 0.84 dl/g

PAM530, R2= 0.99 [] = 0.71 dl/g

re

d (

dl/g

)

Concentration (g/dl)

Figure 5.4: Reduced viscosity vs concentration for block copolymers of similar Mn,tot

but different PAM-PNIPAM ratios

Chapter 5

123

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 Rh, DLS (nm) c*equation 4 (wt.%) 5·c* (wt.%)

PAM530 57 0.0100 0.0500

PAM530-b-PNIPAM10 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 , and can be calculated

(equation 5.4) if the radius of gyration ( ) or the hydrodynamic radius ( )

is known.28, 29

(5.3)

(5.4)

with being the Avogadro constant, is the molecular weight of the

polymer, and 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


124

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).

1 10 100 1000

1

10

Vis

co

sity (

mP

a.s

)

Shear rate (, s-1)

PAM460

PAM460-b-PNIPAM10

PAM235-b-PNIPAM125

PAM200-b-PNIPAM185

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

Chapter 5

125

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).

0 20 25 30 35 40 45 50 55 60

0

50

100

150

200

250

300

700

750

800

Vis

cosity (

mP

a.s

)

Temperature (°C)

PAM200-b-PNIPAM185

PAM200-b-PNIPAM70

PAM200-b-PNIPAM155

PAM200-b-PNIPAM90

PAM460

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


126

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 structure32,

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).

1E-7 1E-6 1E-5 1E-4 1E-3

40

45

50

55

60

65

70

Su

rfa

ce

te

nsio

n (

mN

/m)

Concentration (M)

PAM200-b-PNIPAM650

PAM200-b-PNIPAM185

PAM200-b-PNIPAM155

PAM200-b-PNIPAM90

PAM200-b-PNIPAM30

Figure 5.7: Surface tension against the polymer concentration of 5 different

copolymers

Chapter 5

127

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 J1/2·cm-3/2.

For a pure PNIPAM polymer the solubility parameter is 22.07 J1/2·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


128

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.

0 22 24 26 28 30

1E-6

1E-5

CM

C (

M)

Solubility parameter (J1/2

·cm-3/2

)

ln (y) = 0,5604·x - 27,63

R2 = 0.998

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

Chapter 5

129

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.

0 20 25 30 35 40 45 50 55 60 65 70 75

0

20

40

60

80

100

Ab

so

rba

nce

(%

)

Temperature (°C)

PAM200-b-PNIPAM30

PAM200-b-PNIPAM90

PAM200-b-PNIPAM155

PAM200-b-PNIPAM185

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 hydrophobic-

hydrophilic 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


130

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, 10-

11,. 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.

Chapter 5

131

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.


132


Chapter 6

133

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 thermo-

responsiveness 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, 5456-

5466.

Branched thermosensitive copolymers

134

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 agents4-6,

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 weight19, 20 (or even

NIPAM oligomers21), the nature of the endgroups21, 22, and the chemical

structure1, 10 (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, 35

Other moieties such as poly(ethylene oxide) (PEO)36, 37 and, more recently,

Chapter 6

135

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 s-

1) 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.


Chemicals. Acrylamide (AM, electrophoresis grade, ≥99%), N-

isopropylacrylamide (NIPAM, 97%), tris[2-(dimethylamino)ethyl]amine

(Me6TREN) copper(I) bromide (CuBr, 98%), glacial acetic acid, ethanol,

chloroform and diethyl ether were purchased from Sigma Aldrich. CuBr was



136


then dried at reduced pressure.40

Synthesis of the macro-initiator. The chemical modification of the

original PK was performed according to the published method41 (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.


The chloropropylamine hydrochloride 14.8 g (0.114 mol) was dissolved in


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:

Chapter 6

137

(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



(6.2)



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 5-

10 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


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


138

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-(PAMB-

co-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).

Chapter 6

139



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 sample-

cups. 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












viscoelastic response of each sample. With this, it was ensured that the

dynamic measurements were conducted in the linear response region of the

samples.


140


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)


(C : H : N, wt%) XCO (%)b Pyrrole unitsc Mn,GPC PDI

PK30 (virgin) 67.0 : 8.4 : 0 - 0 2 797 1.74

PK30-Cl4 58.6 : 7.1 : 1.6 18.87 4 2 447 2.02

PK30-Cl8 64.0 : 7.9 : 3.3 37.21 8 2 244 2.01

PK30-Cl13 62.9 : 7.6 : 4.9 61.14 13 2 072 1.97

a. Number indicates the ethylene content (%) and Cl indicates the halogen present



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.

Chapter 6

141

7 6 5 4 3 2 1

cb

aa

c

d

ppm

PK30, virgin

PK30-Cl4

PK30-Cl8

PK30-Cl13

d

d

d

b

b

ba

a

a

c

c

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


142

Afterwards NIPAM was polymerized as the second block and the NIPAM

conversion was determined by 1H-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

Entry [M]0:[I]0:[L]0:[C]0a

M/water (wt:vol);

T; Time (min)

Conv (%) Mn

DP

NIPAM

DP

PAM AMb NIPAMc

PK30-g4-(PAM7575) 10 005:1:1:1.5 1: 6.0;25 °C; 60 75.7 - 541 329 0 7575

PK30-g4-(PAM7575-b-PNIPAM15) 8 000:1:1:1.5 1:17.4;25 °C; 1440 - 0.2 543 254 15 7575

PK30-g4-(PAM7575-b-PNIPAM1690) 3 000:1:1:1.5 1:46.5;25 °C; 1425 - 56.3 732 393 1690 7575

PK30-g8-(PAM7770) 10 025:1:1:1.5 1: 6,0;25 °C; 60 77.5 - 554 982 0 7770

PK30-g8-(PAM7770-b-PNIPAM60) 100 000:1:1:1.5 1: 4.2;25 °C; 1405 - 0.06 561 705 60 7770

PK30-g8-(PAM7770-b-PNIPAM480) 8 000:1:1:1.5 1:18.6;25 °C; 1440 - 6.0 609 368 480 7770

PK30-g13-(PAM3275) 12 000:1:1:2.5 1: 4.0;25 °C; 1440 27.3 - 235 504 0 3275

PK30-g13-(PAM3275-b-PNIPAM415) 10 000:1:1:1.5 1:13.3;25 °C; 1230 - 4.2 282 579 415 3275

PK30-g13-(PAM3275-b-PNIPAM4425) 10 000:1:1:1.5 1:10.9;25 °C; 1425 - 44.2 736 146 4425 3275

PK30-g13-(PAM6140) 12 000:1:1:1.5 1: 6.0;25 °C; 60 51.2 - 439 326 0 6140

PK30-g13-(PAM6140-b-PNIPAM205) 4 965:1:1:1.5 1:31.7;25 °C; 1405 - 4.1 462 515 205 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

Chapter 6

143

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 hindrance44),

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 1H-NMR data is in

agreement with solubility tests, and confirms the preparation of block

copolymers with different lengths of the blocks.

5 4 3 2 1 0

D2O

b

c

ac

c

PK30-g13

-(PAM3275-b-PNIPAM4425)

PK30-g13


PK30-g13

-(PAM3275)

PK30-g8-(PAM7770-b-PNIPAM480)

PK30-g8-(PAM7770)




ppm

PK30-g4-(PAM7575)

a

Figure 6.2: 1H-NMR spectra of the block copolymers PK30-gX-(PAMY-b-PNIPAMZ) and

the parent macroinitiators PK30-gX-(PAMY)


144

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

Entry [M]0[n:m]:[I]0a

M/water (wt:vol),

T, Time (min)

Conv (%)

Mn, GC DP

PAM

DP

NIPAM AMa NIPAMb

PK30-g4-(PAM5015-co-PNIPAM4885) 12 000[1:1]:1 1:4.0;25 °C; 210 83.6 81.4 908 999 5 015 4 885

PK30-g4-(PAM18875-co-PNIPAM19240) 50 000[1:1]:1 1:4.0;25 °C; 210 75.3 76.8 3 518 808 18 875 19 240

PK30-g4-(PAM33395-co-PNIPAM31790) 100 000[1:1]:1 1:4.0;25 °C; 210 66.8 63.6 5 970 955 33 395 31 790

PK30-g8-(PAM4400-co-PNIPAM4460) 12 000[1:1]:1 1:4.0; 25 °C; 210 73.4 74.4 817 851 4 400 4 460

PK30-g8-(PAM11125-co-PNIPAM10205) 75 000[1:1]:1 1:4.0; 25 °C; 210 29.4 26.0 1 945 492 11 125 10 205









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

Chapter 6

145

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).

1 10 100 100010

-1

100

Vis

co

sit

y (

Pa

.s)

Shear rate (s-1)



PK30-g4-(PAM7575)

A

1 10 100 100010

-1

100

101 B

Vis

co

sit

y (

Pa

.s)

Shear rate (s-1)



PK30-g8-(PAM7770)

1 10 100 100010

-2

10-1

100 C

Vis

co

sit

y (

Pa

.s)

Shear rate (s-1)

PK30-g13


PK30-g13


PK30-g13

-(PAM3275)

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


146

related to the increased molecular weight of the polymers and the presence

of the NIPAM blocks (Chapter 5).


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.

0 20 30 40 50 60 70 80

1

2

A



PK30-g4-(PAM7575)

Vis

cosity (

Pa.s

)

Temperature (oC)

0 20 30 40 50 60 70 80

1

2

Vis

cosity (

Pa.s

)

Temperature (oC)



PK30-g8-(PAM7770)

B

0 20 30 40 50 60 70 80

10-1

100

101

C

Vis

cosity (

Pa.s

)

Temperature (oC)

PK30-g13


PK30-g13


PK30-g13

-(PAM3275)

Figure 6.4: Viscosity ( = 30 s-1) versus temperature of A: 5 wt.% PK30-g4-(PAMY-b-

PNIPAMZ), B: 4 wt.% PK30-g8-(PAMY-b-PNIPAMZ), C: 3 wt.% PK30-g13-(PAMY-b-

PNIPAMZ)

Chapter 6

147

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-b-

PNIPAM4425) 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, 1H-

NMR spectra were recorded at different temperatures (Figure 6.6). The


148

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.

5 4 3 2 1 0

cc

cc

a

a

a

D2O

a

PK30-g13-(PAM1405-co-PNIPAM1405)

ppm


75°C

25°C

50°C

35°C

30°C

25°C

Figure 6.6: 1H-NMR spectra of entries PK30-g13-(PAM3275-b-PNIPAM4425) and PK30-

g13-(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

Chapter 6

149

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

0,1 1 10 100 1000

10-1

100

101

Vis

cosity (

Pa.s

)

Shear rate (s-1)

PK30-g13

-(PAM5135-co-PNIPAM2530)

PK30-g13


PK30-g13


A

0,1 1 10 100 1000

10-2

10-1

100

101

B

PK30-g13


Vis

co

sity (

Pa

.s)

Shear rate (s-1)

20 oC

50 oC

80 oC

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


150

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

and low shear rate (T = 80 °C and ≤ 3 s-1), the solution viscosity is

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.


measured as a function of the temperature of the solution containing the

random copolymers and the results are displayed in Figure 6.8.

0 20 30 40 50 60 70 8010

-1

100

A


Vis

cosity (

Pa.s

)

Temperature (oC)

= 3 s-1

= 30 s-1

0 20 30 40 50 60 70 8010

-1

100

BV

isco

sity (

Pa

.s)

Temperature (oC)


= 3 s-1

= 30 s-1

0 20 30 40 50 60 70 8010

-2

10-1

100

= 1 s-1

= 5 s-1

= 50 s-1

PK30-g13

-(PAM3275)

Vis

cosity (

Pa.s

)

Temperature (oC)

= 1 s-1

= 5 s-1

= 50 s-1

PK30-g13


C

0 20 30 40 50 60 70 8010

-2

10-1

D

PK30-g13


= 1 s-1

= 5 s-1

= 50 s-1V

iscosity (

Pa.s

)

Temperature (oC)

Figure 6.8: Viscosity versus temperature of (A): 3 wt.% PK30-g4-(PAM5015-co-

PNIPAM4885), (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)

Chapter 6

151

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


152

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).

1E-9 1E-8 1E-7 1E-6 1E-5 1E-4

40

45

50

55

60

65

70

75

1E-9 1E-8 1E-7 1E-6 1E-5

40

45

50

55

60

65

70

75

PK30-g13


PK30-g13


(

mN

/m)

Concentration (mol %)

Su

rfa

ce

te

nsio

n (

mN

/m)

Concentration (M)

PK30-g13



PK30-g13



PK30-g13


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

Chapter 6

153

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. 1H-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



6.6. References

1. Liu, R.; Fraylich, M.; Saunders, B. R. Colloid Polym. Sci. 2009, 6, 627. 2. Koyama, M.; Hirano, T.; Ohno, K.; Katsumoto, Y. J Phys Chem B 2008, 35, 10854. 3. Liu, H. Y.; Zhu, X. X. Polymer 1999, 25, 6985. 4. Xiaojiao Kang; Ziyong Cheng; Dongmei Yang; Ping'an Ma; Mengmeng Shang; Chong

Peng; Yunlu Dai; Jun Lin Advanced Functional Materials 2012, 7, 1470. 5. Fundueanu, G.; Constantin, M.; Ascenzi, P. Int. J. Pharm. 2009, 1, 9. 6. Ward, M. A.; Georgiou, T. K. Polymers 2011, 3, 1215. 7. Stayton, P. S.; Shimoboji, T.; Long, C.; Chilkoti, A.; Chen, G. H.; Harris, J. M.;

Hoffman, A. S. Nature 1995, 6556, 472. 8. Akiyama, H.; Tamaoki, N. Macromolecules 2007, 14, 5129. 9. Hoare, T.; Pelton, R. Macromolecules 2007, 3, 670.


154

10. Mumick, P. S.; Mccormick, C. L. Polym. Eng. Sci. 1994, 18, 1419. 11. Mumick, P. S.; Hester, R. D.; Mccormick, C. L. Polym. Eng. Sci. 1994, 18, 1429. 12. Wever, D. A. Z.; Picchioni, F.; Broekhuis, A. A. Prog. Polym. Sci. 2011, 1558. 13. Sudor, J.; Barbier, V.; Thirot, S.; Godfrin, D.; Hourdet, D.; Millequant, R.;

Blanchard, J.; Viovy, J. L. Electrophoresis 2001, 4, 720. 14. Fujishige, S.; Kubota, K.; Ando, I. J. Phys. Chem. 1989, 8, 3311. 15. Bae, Y. H.; Okano, T.; Hsu, R.; Kim, S. W. Makromolekulare Chemie-Rapid

Communications 1987, 10, 481. 16. Coughlan, D. C.; Quilty, F. P.; Corrigan, O. I. J. Controlled Release 2004, 1, 97. 17. Lue, S. J.; Chen, C.; Shih, C.; Tsai, M.; Kuo, C.; Lai, J. J. Membr. Sci. 2011, 1-2,

330. 18. Purushotham, S.; Ramanujan, R. V. Acta Biomaterialia 2010, 2, 502. 19. Xia, Y.; Yin, X. C.; Burke, N. A. D.; Stover, H. D. H. Macromolecules 2005, 14,

5937. 20. Hu, H.; Du, J.; Meng, Q.; Li, Z.; Zhu, X. Chinese Journal of Polymer Science 2008,

2, 187. 21. Xia, Y.; Burke, N. A. D.; Stover, H. D. H. Macromolecules 2006, 6, 2275. 22. 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. 23. Ye, J.; Narain, R. J Phys Chem B 2009, 3,. 24. 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.

25. Masci, G.; Giacomelli, L.; Crescenzi, V. Macromolecular Rapid Communications 2004, 4,.

26. You, Y.; Zhou, Q.; Manickam, D. S.; Wan, L.; Mao, G.; Oupicky, D. Macromolecules 2007, 24, 8617.

27. Convertine, A.; Lokitz, B.; Vasileva, Y.; Myrick, L.; Scales, C.; Lowe, A.; McCormick, C. Macromolecules 2006, 5, 1724.

28. Ito, M.; Ishizone, T. Journal of Polymer Science Part A-Polymer Chemistry 2006, 16, 4832.

29. Aoshima, S.; Kanaoka, S. Wax Crystal Control: Nanocomposites, Stimuli-Responsive Polymers 2008, 169.

30. Schild, H. G. Progress in Polymer Science 1992, 2, 163. 31. Zhang, Y. Q.; Tanaka, T.; Shibayama, M. Nature 1992, 6400, 142. 32. Yoshida, R.; Uchida, K.; Kaneko, Y.; Sakai, K.; Kikuchi, A.; Sakurai, Y.; Okano, T.

Nature 1995, 6519, 240.

33. Durand, A.; Hourdet, D. Polymer 1999, 17, 4941. 34. Durand, A.; Hourdet, D. Macromolecular Chemistry and Physics 2000, 8, 858. 35. Petit, L.; Karakasyan, C.; Pantoustier, N.; Hourdet, D. Polymer 2007, 24, 7098. 36. Hourdet, D.; L'Alloret, F.; Audebert, R. Polymer 1997, 10, 2535. 37. Hourdet, D.; L'Alloret, F.; Audebert, R. Polymer 1994, 12, 2624. 38. Wang, Y.; Feng, Y.; Wang, B.; Lu, Z. J Appl Polym Sci 2010, 6, 3516. 39. Yu Wang; Zhiyong Lu; Yugui Han; Yujun Feng; Chongli Tang Adv. Mat. Res. 2011,

654. 40. Neugebauer, D.; Matyjaszewski, K. Macromolecules 2003, 8, 2598. 41. Zhang, Y.; Broekhuis, A. A.; Stuart, M. C. A.; Picchioni, F. J Appl Polym Sci 2008,

1, 262. 42. Egan, R.; Jones, M.; Lehninger, A. J. Biol. Chem. 1976, 14, 4442. 43. Wever, D. A. Z.; Raffa, P.; Picchioni, F.; Broekhuis, A. A. Macromolecules 2012,

10, 4040. 44. Sumerlin, B. S.; Neugebauer, D.; Matyjaszewski, K. Macromolecules 2005, 3, 702. 45. Tam, K.; Wu, X.; Pelton, R. Polymer 1992, 2, 436. 46. Cao, Z.; Liu, W.; Ye, G.; Zhao, X.; Lin, X.; Gao, P.; Yao, K. Macromolecular

Chemistry and Physics 2006, 24, 2329.

Chapter 6

155

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.


156


Chapter 7

157

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.

Oil recovery using branched polyacrylamides

158

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, is the oil mobility, is the water

permeability, is the water viscosity, is the oil permeability, and is the

oil viscosity. Ideally is equal (or lower) to unity leading to a stable

displacing front.5, 6 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

Chapter 7

159

(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.


160

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 semi-

dilute 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.


7.2.1. 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

Chapter 7

161

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 1 524 432 100 0

LPAM35705a 2 540 789 100 0

Linear

CLPAM63310b 4 500 000 100 0

CLPAM84410b 6 000 000 100 0

Poly(AM31515-ran-AA13320)c 3 200 000 70 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 1 692 550 100 0

PK30-g13-(PAM35275)d 2 510 092 100 0

PK30-g13-(PAM49190)d 3 499 094 100 0

17-arm PK30-g17-(PAM22140)d 1 576 493 100 0

PK30-g17-(PAM35700)d 2 540 500 100 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 X-

Ray 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.


162

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), =

the cross-sectional area of the core (cm2), = the viscosity of the fluid

(mPa.s), = the flow rate (cm3/s), and = 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), = the brine

permeability (m2), and = 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

Property

Length (cm) 5 5 5 30 30 30 30

Diameter (cm) 2.5 2.5 2.5 5 5 5 5

Cross-sectional area (cm2) 4.91 4.91 4.91 19.63 19.63 19.63 19.63

Porosity (%) 22.4 22.4 22.4 19.5 19.5 24.0 24.0

Pore volume (PV, mL) 5.5 5.5 5.5 114.9 114.9 141.4 141.4

Brine permeability (mD) 371 246 528 75 96 2126 2371

Average pore radius (m) 3.62 2.94 4.31 1.74 1.97 8.36 8.83

Oil saturation (%) - - - 72.24 74.85 83.39 89.05

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.

Chapter 7

163

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


164

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 CO2 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.

Chapter 7

165

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.









viscoelastic response of each sample and to ensure that the dynamic

measurements were conducted in the linear response region of the samples.


166


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.

1 10 100 1000

5

10

15

20

25

30

So

lutio

n v

isco

sity (

mP

a.s

)

Shear rate (s-1)

LPAM35705, [p] = 3100 ppm

PK30-g13

-(PAM35275), [p] = 3000 ppm

PK30-g17

-(PAM35700), [p] = 2900 ppm

CLPAM63310, [p] = 4900 ppm

CLPAM84410, [p] = 3250 ppm

A

0,1 1 1010

-5

10-4

10-3

10-2

10-1

100

101

B

G'G

" (P

a)

Frequency (rad/s)

G'

G"

0

10

20

30

40

50

60

70

80

90

CLPAM84410CLPAM63310

PK30-g17

-(PAM35700)

PK30-g13

-(PAM35275)

LPAM35705

Ph

ase

an

gle

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 1.2m 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

Chapter 7

167

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.

100 200 300 400

0

50

100

150

200

250

A

Weig

ht (g

)

Time (s)

CLPAM63310 FR = 1,09

PK30-g13

-(PAM35275) FR = 1,14

PK30-g17

-(PAM35700) FR = 1,07

LPAM35705 FR = 1,31

CLPAM84410 FR = 1,37

0 1000 2000 3000 4000

0

50

100

150

200

250

B

We

igh

t (g

)

Time (s)

LPAM35705 FR = 1,57

CLPAM63310 FR = 1,33

PK30-g17

-(PAM35700) FR = 2,07

CLPAM84410 FR = did not pass

PK30-g13

-(PAM35275) FR = did not pass

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) 1.2m

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.

0,1 1 10 100 1000

101

102

Solu

tion

vis

cosity (

mP

a.s

)

Shear rate (s-1)

Poly(AM31515-ran-AA13320)

PK30-g13

-(PAM35275)

LPAM35705

A

0,1 1 1010

-5

10-4

10-3

10-2

10-1

100

101

102

103

G'G

" (P

a)

Frequency (rad/s)


PK30-g13

-(PAM35275)

LPAM35705

0

10

20

30

40

50

60

70

80

90

B

Ph

ase

an

gle

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


168

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 surface3) 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

Chapter 7

169

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).

0,01 0,1 1 10 100

0,0

0,1

0,2

0,3

Fra

ctio

n o

f to

tal p

ore

s

Pore throat diameter (m)

Pore throat size distribution

0,0

0,2

0,4

0,6

0,8

1,0

Cu

mu

lative

fra

ctio

n

Cumulative

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.


170

1 10 100 1000

10

100

Solu

tion

vis

cosity (

mP

a.s

)

Shear rate (, s-1)

LPAM21445, [p] = 8500 ppm

PK30-g4-(PAM22660), [p] = 8500 ppm

PK30-g13

-(PAM23775), [p] = 5250 ppm

PK30-g13

-(PAM49190), [p] = 4000 ppm

PK30-g17

-(PAM22140), [p] = 5250 ppm

Poly(AM31515-co-AA13320), [p] = 8700 ppm

A

0,1 1 10

10-5

10-4

10-3

10-2

10-1

100

101

102

G'G

" (P

a)

Frequency (rad/s)

PK30-g4-(PAM22660)

PK30-g13

-(PAM49190)

PK30-g13

-(PAM23775)

PK30-g17

-(PAM22140)

LPAM21445

Poly(AM31515-co-AA13320)

B

0

10

20

30

40

50

60

70

80

90

Phase

an

gle

G"

G'

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 [B] LPAM21445 [C] PK30-g4-(PAM22660) [D] PK30-g13-(PAM23775)

[E] PK30-g13-(PAM49190) [F] PK30-g17-(PAM22140) [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).

Chapter 7

171

8700 8500 8500

5250 5250

4000

HPAM

3,5

MDa

Line

ar P

AM

1,5

MDa

4-ar

m P

AM

1,5

MDa

13-a

rm P

AM

1,5

MDa

17-a

rm P

AM

1,5

MDa

13-a

rm P

AM

3,5

MDa

0

5

10

15

20

25

30 Polymer concentration

Po

lym

er

co

nce

ntr

atio

n ( 1

03 p

pm

)

0

5

10

15

20

25

30

Oil recovery

Oil

reco

ve

ry (

%)

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-


172

(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-ran-

AA13320)) 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

Chapter 7

173

0,1 1 10 100 100010

1

102

Vis

co

sity (

mP

a.s

)

Shear rate (s-1)

PK30-g17

-(PAM22140)


A

0 20 30 40 50 60 70

20

40

60

80

100

B

Vis

cosity (

mP

a.s

)

Temperature (oC)

PK30-g17

-(PAM22140)


LPAM21445

0,1 1 1010

-6

10-5

10-4

10-3

10-2

10-1

100

101

102

103


PK30-g17

-(PAM22140)

C

G'G

" (P

a)

Frequency (rad/s)

G"

G'

0

10

20

30

40

50

60

70

80

90

Ph

ase

an

gle

Figure 7.10: (A) Viscosity functions for the two polymer solutions used in the core-

flow, (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.


174

0 1 2 3 4

024

28

32

36

40

44

48

Oil

recove

ry (

%)

Cumulative displacing fluid injected (PV)

Waterflood

PK30-g17

-(PAM22140)

Waterflood


Start polymer injection

Extrapolated waterflood

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-ran-

AA13320)) leads to an increase of 6.0% in the oil recovery (Figure 7.12).

0 1 2 3 4

020

25

30

35

40

45

50

55

60

Oil

recovery

(%

)

Cumulative displacing fluid injected (PV)

Waterflood

PK30-g17

-(PAM22140)

Water flood


Start polymer injection

Extrapolated waterflood

Figure 7.12: Oil recovery from high permeable Bentheim sandstone cores

Chapter 7

175

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 Oil sat.a

(% of PV)

Oil sat.,

waterb (%)

Oil sat.,

polymerc (%)

Oil rec.d

(% of OOIP)

∆P, PFe

(bars)

Poly(AM31515-co-AA13320) Berea 2b 8700 74.85 44.02 42.91 1.48 1.9

PK30-g17-(PAM22140) Berea 2a 5250 72.24 42.27 38.88 4.97 2.6

Poly(AM31515-co-AA13320) Bentheim 1b 8700 89.05 46.02 40.66 6.01 0.1

PK30-g17-(PAM22140) Bentheim 1a 5250 83.39 41.56 33.70 9.43 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


176

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).




7.6. 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, SPE-122400.

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, SPE-

75194-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.

Chapter 7

177

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-63227-

MS. 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-65378-

MS.


178


Chapter 8

179

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 dead-

ends. 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

Outlook on the application of branched (co)polymers in EOR

180

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.

Chapter 8

181

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


182

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. reptation4, 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

Chapter 8

183

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


184

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 thermo-

thickening 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

polyelectrolyte solutions3, 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.

Chapter 8

185

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 (non-

hydrophobic) 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.


186

4,5

3,2

3,5

PAM HPAM 13-arm PAM

0,0

0,2

0,4

0,6

0,8

1,0

1,2 Viscosity

Vis

co

sity (

Pa

.s)

0

1

2

3

4

5

Molecular weightA

Mo

lecu

lar

we

igh

t (x

10

6 g

/mo

l)

3,2 3,2 3,2 3,2 3,2

3,5

HPAM

0,5 N

aCl

2,0 N

aCl

3,0 N

aCl

12,0 N

aCl

13-arm

PAM

0,0

0,1

0,2

0,3

0,4

0,5 Viscosity

Vis

cosity (

Pa.s

)

B

0

1

2

3

4

5

Molecular weight

Mo

lecula

r w

eig

ht (x

10

6 g

/mol)

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,

Chapter 8

187

rather than HPAM, has been proposed as a new method for EOR.26 The non-

hydrolyzed 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).

0 1 2 3 11 1210

1

102

103

A PK30-g

13-(PAM49225)

PK30-g13

-(PAM23810)




Vis

cosity (

mP

a.s

)

Concentration NaCl (wt. %)

0,1 1 10 100 1000

101

102

Vis

cosity (

mP

a.s

)

Shear rate (s-1)

PK30-g13

-(PAM49225)


PK30-g13

-(PAM23810)


B

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


188

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.

Chapter 8

189

0,1 1 10 100

0

30

40

50

60

70

80

90

Phase a

ngle

Frequency (rad/s)

NaCl concentration = 30000 ppm




Increasing NaCl

concentration

A

0,1 1 10 100

0

30

40

50

60

70

80

90

B

Phase a

ngle

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


190

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 of DMA, NIPAM and AAE has been

demonstrated already.45-50

Chapter 8

191

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) Elemental composition

(C : H : N, wt%) XCO (%)b Pyrrole

unitsc Mn,GPC PDI

PK30 (virgin) 67.0 : 8.4 : 0.0 - 0 2 797 1.74

PK30-Cl4 58.6 : 7.1 : 1.6 18.87 4 2 447 2.02

PK30-Cl8 64.0 : 7.9 : 3.3 37.21 8 2 244 2.01

PK30-Cl13 62.9 : 7.6 : 4.9 61.14 13 2 072 1.97

a. Number indicates the ethylene content (%)




192

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


[Me6TREN]0

M/s1/s2a (wt:vol:vol);

T; Time (min) Conv (%) Mn,tot Mn,SPAN

Linear

1 22 919:1:1.5:1.5 1:5 ; 25 °C; 180 58.4 1 326 825 1 326 825

2 51 623:1:1.5:1.5 1:5 ; 25 °C; 180 58.2 2 978 320 2 978 320

3 88 515:1:3.0:3.0 1:5 ; 25 °C; 180 47.0 4 124 011 4 124 011

4-arm 4 79 351:1:3.0:3.0 1:4:1/10; 25 °C; 60 70.1 5 514 111 2 759 853

8-arm 5 79 213:1:3.0:3.0 1:4:1/10; 25 °C; 60 64.5 5 064 788 1 268 994

13-arm

6 19 969:1:1.5:1.5 1:6:1/6 ; 25 °C; 130 39.9 789 831b 124 310

7 49 905:1:1.5:1.5 1:5:1/10; 25 °C; 180 46.3 2 290 499 355 181

8 99 226:1:3.0:3.0 1:5:1/20; 25 °C; 150 49.9 4 908 300 757 920

9 200 000:1:3.0:3.0 1:5:1/40; 25 °C; 180 30.3 6 007 278 926 993

17-arm 10 100 030:1:1.5:3.0 1:4:1/20; 25 °C; 180 41.0 4 065 549 481 097

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

Chapter 8

193

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

rheological properties depend on the number of arms (Figure 8.8).

0,1 1 10 100 1000

102

103

A

Vis

co

sity (

mP

a.s

)

Shear rate (s-1)

PK30-g13

-(PDMA23105), entry 7

PDMA30045, entry 2

0,1 1 10 100 1000

102

103

B

Vis

co

sity (

mP

a.s

)

Shear rate (s-1)

PK30-g17


PDMA41600, entry 3

0,1 1 10 100 100010

1

102

Vis

co

sity (

mP

a.s

)

Shear rate (s-1)

PK30-g13


PK30-g8-(PDMA51090), entry 5

PK30-g4-(PDMA55625), entry 4

C

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


194

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).

0 25 50 75 100 125 150 175 200

100

150

200

250

300

350

400

Vis

co

sity r

ete

ntio

n (

%)

Time (hours)

PAM21445, NaCl

PAM21445, NaCl-CaCl2

PK30-g13

-(PAM23810), NaCl

PK30-g13

-(PAM23810), NaCl-CaCl2

PK30-g13

-(PDMA49515), NaCl

PK30-g13

-(PDMA49515), NaCl-CaCl2

A

0 25 50 75 100 125 150 175 200

0

2

4

6

8

10

12

14

95100

B

Hydro

lysis

deg

ree (

%)

Time (hours)

PAM21445, NaCl

PK30-g13

-(PAM23810), NaCl

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 13C-NMR-spectroscopy

Chapter 8

195

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 CaCl2. 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 CaCl2 or part of

the polymer precipitates by complex formation with Ca2+.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 13C-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.


196

0 10 20 30 40 50 60

0

50

100

150

200

250

300

A

Vis

cosity r

ete

ntion

(%

)

Time (days)

PK30-g13

-(PAM35275), NaCl

PK30-g13

-(PAM35275), NaCl-CaCl2

PAM35705, NaCl

PAM35705, NaCl-CaCl2

PK30-g13

-(PDMA23105), NaCl

PK30-g13


PDMA30045, NaCl

PDMA30045, NaCl-CaCl2

0 10 20 30 40 50 60

0

40

50

60

70

80

90

100

110

Vis

cosity r

ete

ntion

(%

)

Time (days)

Poly(AM31515-co-AA13320), NaCl

Poly(AM31515-co-AA13320), NaCl-CaCl2

PK30-g13

-(PDMA23105), NaCl

PK30-g13


PDMA30045, NaCl

PDMA30045, NaCl-CaCl2 B

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 13C-

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

Chapter 8

197

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 13C-NMR where no carboxylate units could be detected (i.e.

below the detection limit of 13C-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 CaCl2 and the ones

without were not large. Although the presence of CaCl2 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 thermo-

responsive 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).


198

[1] Brine (30000 ppm NaCl) [2] Poly(AM31515-ran-AA13320)

[p] = 8700 ppm

Residual oil recovery = 7.6 % (± 1.8)

[3] PK30-g13-(PDMA23105)

[p] = 8000 ppm

Residual oil recovery = 8.9 % (± 1.8)

[4] Water, RT

[5] PK30-g13-(PAM3275), RT

[p] = 11000 ppm

Residual oil recovery = 4.8 % (±1.9)

[6] PK30-g13-(PAM1405-ran-PNIPAM1405)

[p] = 9000 ppm, RT


[7] Water, 70 °C

[8] PK30-g13-(PAM3275), 70 °C

[p] = 11000 ppm


[9] PK30-g13-(PAM1405-ran-PNIPAM1405)

[p] = 9000 ppm, 70 °C


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

Chapter 8

199

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 thermo-

responsive 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 re-

injected. 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 reservoirs22, 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


200

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.

2

0,09

0,7

1,5

1,08

0,66

2,6

5

0,96

Xanthan g

um

Meth

yl ce

llulose

CM ce

llulose

(LM

)

CM ce

llulose

(HM

)

Chitosa

n

Lambda-c

arrageenan

Guar gum

Sclero

glucan

Schizo

phyllan

0

1

2

3

4

5

6

7

Viscosity

Vis

co

sity (

Pa

.s)

0

1

2

3

4

5

6

7

Molecular weight

Mo

lecu

lar

we

igh

t (x

10

6 g

/mo

l)

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 gum58,

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, = 10 s-1 and a

[p] = 200 ppm).61 In addition, the solution viscosity of an aqueous solution

containing schizophyllan only decreases by 10 % when heated up to 130

Chapter 8

201

°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


202

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

water solutions.68-71

0,1 1 10 100

10-4

10-3

10-2

10-1

100

101

102

A

G' G

" (P

a)

Frequency (rad/s)

Filled symbols = G'

Empty symbols = G"

= Xanthan gum

= Guar gum

= CM cellulose

= Methyl cellulose

= -Carrageenan

0,1 1 10 100

0

10

20

30

40

50

60

70

80

90

B

Ph

ase

an

gle

Frequency (rad/s)

Methyl cellulose

-Carrageenan

Guar gum

CM cellulose

Xanthan gum

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

Chapter 8

203

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 bottle-

brush 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.





204

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.; Abdel-

Goad, 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-63227-

MS. 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.

Chapter 8

205

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-20237-

MS,. 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.


206

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) 3-

chloropropylamine hydrochloride (9.89 g, 53.6 mmol) was dissolved in


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

Chapter 8

207

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 1H-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)



The number of pyrrole units was determined using the conversion of the

carbonyl groups of the polyketone and formula 2:

(8.A.2)




208

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-pump-


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 D2O

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

Chapter 8

209



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 CaCl2. 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 = 10 s-1) of the virgin

polymer samples and = 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 CaCl2.

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


210

oven and cooled to room temperature. The viscosity function of the sample

was then recorded. Both solutions were evaluated for more than 63 days.

Chapter 8

211

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.


212


Summary

213

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

Summary

214

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

Summary

215

prepared through ATRP of AM with water as the solvent. The polymers were

prepared by starting each targeted polymerization with the novel multi-

functional 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

Summary

216

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 thermo-

thickening 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

Summary

217

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,N-

dimethylacrylamide. 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 thermo-

thickening 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.

Summary

218


Samenvatting

219

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

Samenvatting

220

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 · 105 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

Samenvatting

221

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

Samenvatting

222

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

Samenvatting

223

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.

Samenvatting

224


Compilacion

225

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

Compilacion

226

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 makro-

iniciado.

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 makro-

iniciado 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

Compilacion

227

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 multi-

functional 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-b-

PNIPAM caracterisa pa diferente proporcion entre e largura di e dos blokinan.

E propiedadnan di e solucion ta demostra cu dor di incorpora NIPAM-

unidadnan 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

Compilacion

228

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

Compilacion

229

“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 bio-

polymernan.

Compilacion

230


Acknowledgements

231

Acknowledgements

The journey in Groningen has come to an end, at least the one as a PhD-

student. 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

Acknowledgements

232

(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

List of publications

233


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-N-

isopropylacrylamide 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.


234

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.

Documents

University of Groningen Synthesis and evaluation of novel ......Dedicated to my beloved wife, The work in this thesis is at best captured by my wife’s words: “To find the best