Synthesis of Hyperbranched Polymers and Analysis of their Solution Behaviour

Abstract.

Contents

Abstract………………………………………………………………………………2

Contents……………………………………………………………………………..3

1 Introduction…………………………………………………………………………5

1. Hyperbranched Polymer……………………………………………………..5
2. Free Radical Polymerisation …………………………………………………8
3. Reversible Addition Fragmentation Chain Transfer (RAFT)………………..10
4. Amphiphiles…………………………………………………………………..13
5. Statistical Copolymers………………………………………………………..14
6. Block Copolymers…………………………………………………………..14
7. Structural Characterisation…………………………………………………..16
   1.      Introduction……………………………………………………….16
   2.      Gel Permeation Chromatography (GPC)…………………………16
   3.      Dynamic Light Scattering…………………………………………16
   4.      Small Angle X-Ray Scattering (SAXS)…………………………..17
   5.      Rheology………………………………………………………….20
   6.      Zeta Potential……………………………………………………..20
8. General Aims………………………………………………………………..21

2. Results and Discussion…………………………………………………………….21
   1. Kinetic Studies on Linear Poly (Butyl Methacrylate)……………………….21
   2. Synthesis of Hyperbranched Poly (Butyl Methacrylate)……………………32
   3. End Group Modification of the Hyperbranched Poly (Butyl Methacrylate)..34
   4. Hyperbranched Poly (Butyl Methacrylate-Stat-Methacrylic Acid)…………36
   5. Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic Acid)..42
   6. Dynamic Light Scattering……………………………………………………46
   7. Small Angle X-Ray Scattering of Dilute Solutions…………………………50
   8. Zeta Potential………………………………………………………………..57
   9. Rheology…………………………………………………………………….58
3. Conclusions………………………………………………………………………..60
4. Experimental………………………………………………………………………63
   1. Materials…………………………………………………………………….63
   2. Synthesis of PETTC…………………………………………………………63
   3. Synthesis of Linear Poly (Butyl Methacrylate)……………………………..64
   4. Synthesis of Hyperbranched Poly (Butyl Methacrylate)……………………65
   5. Synthesis of Hyperbranched Poly (Butyl Methacrylate-Stat-Methacrylic acid)…………………………………………………………………………66
   6. Synthesis of Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic acid)…………………………………………………………..67
   7. Hyperbranched Poly (Butyl Methacrylate) End Group Modification………68
   8. Solution Preparation………………………………………………………..68
   9. Alkylation…………………………………………………………………..69
   10.             Instrumentation……………………………………………………….69
       1. Gel Permeation Chromatography………………………………..69
       2. Dynamic Light Scattering……………………………………….70
       3. ¹H NMR…………………………………………………………70
       4. Rheology…………………………………………………………70
       5. Zeta Potential……………………………………………………71
       6. Small Angle X-ray Scattering……………………………………71
       7. Solid State Infrared Spectroscopy……………………………….71
5. References………………………………………………………………………..71
6. Appendix………………………………………………………………………….76

1 Introduction

1.1 Hyperbranched Polymers

Hyperbranched polymers (Fig. 1) are macromolecules made from an interlinking network of branching polymer chains. Hyperbranched polymers have unique physical and chemical properties, meaning they have many potential applications ranging from drug-delivery to water purification.{Gao, 2004, Hyperbranched polymers: from synthesis to applications}^[1][2][3] These interesting properties are due to the densely-branched structure and the large number of reactive end groups.^[4]

The history of hyperbranched polymers can be traced back to Berzelius near the end of the 19^th century where he made a resin out of tartaric acid and glycerol.^[5] This was followed my Watson Smith reporting the reaction of phthalic anhydride and phthalic acid.^[5] It was also reported by Kienle and Hovey that the reaction of glycerol and phthalic anhydride continued until gelation suggesting a polymer network was forming.^[5] Later on Flory showed that it was possible to synthesise highly branched polymers without gelation, through the use of a monomer having one functional group (A) and two or more different functional groups (B) capable of reacting with the first functional group.^[6] The name hyperbranched polymers was first used in 1988 by Kim and Webster where they made a soluble hyperbranched polyethylene which wasn’t monodisperse.^[7]

A structure related to hyperbranched polymers is a dendrimer, which are defined as being monodisperse perfectly branched polymers compared to the normal polydisperse hyperbranched polymers.^[4] The difference in structures is due to them being synthesised differently, with dendrimers requiring multiple highly controlled steps and hyperbranched polymers just requiring a single less controlled step.^[4] Therefore, hyperbranched polymers are usually simpler to make in industry compared to dendrimers.

Figure 2. A picture of a dendrimer.^[8]

One of the key properties of hyperbranched polymers is the low solution viscosity, which is even present when high molecular weight hyperbranched polymers are made.^[4] This low solution viscosity is believed to be due the hyperbranched polymers being less able to undergo chain entanglement due to entanglement length, so they form individual globular structures.^[9] Most hyperbranched polymers are chemically stable as the chemical bonds forming the structure are held deep inside the hyperbranched core and they are shielded by the functional end groups.^[4]

To form hyperbranched polymers instead of a linear polymer, branching points must be added in the synthesis. Branching points are units with at least three reactive ends which cause the branching of the polymer chain. This causes multiple polymers chains to join and form the network structure. However if too much of the branching unit is added then the branching continues until a polymer gel is formed.^[10] A polymer gel is a continuous network of polymer chains and it is insoluble. With hyperbranched polymers the monomer units determine the solubility of the polymer in solution.^[11] With monomers having hydrophilic groups on them being more soluble in water, and monomer units with more hydrophobic units not being soluble in water.

The most common method of making hyperbranched is poly condensation of an AB_n monomer, however a more cost-effective method is the Strathclyde route.^[12] The Strathclyde method is a cheap and easy way of making hyperbranched polymers.^[13] It involves using the right ratio of branching agent to chain transfer agent but it yields polymers with broad molecular weight distributions and uncontrolled degree of branching. ^[14]

One method of synthesising hyperbranched polymers is the use of multifunctional monomers to make crosslinked polymers.^[15] This process only requires a small amount of the multifunctional monomer to produce cross linked networks, however if the number of cross links per chain exceed one then gelation occurs. This gel formation can be supressed however with the use of chain transfer.^[16] This chain transfer can be performed using a controlled radical polymerisation.^[17]

1.2 Free Radical Polymerisation

Free radical polymerisation contains three key reaction steps (scheme 1).^[18] The first step is initiation; this involves the homolytic cleavage of a covalent bond. This homolytic cleavage can be caused by heat, UV and visible light, ionizing light, redox reagents and electricity.^[19] The free radical formed in the initiation step goes onto perform the second step which is propagation. Propagation is where the free radical adds to a monomer and this addition of monomer continues until termination of the radical takes place, which is the third step. Termination occurs when two radical species interact and either undergoes combination or disproportionation. Combination occurs when two radical chain ends couple together and form a single polymer chain. Disproportion occurs by one radical end transferring hydrogen to the other, forming two polymer chains with one being unsaturated.

Scheme 1. Mechanism of free radical polymerisation. Where I is the initiator, R is the radical species formed by initiation and M is a monomer unit.

Radical polymerisations are convenient for use in industry as they require relatively undemanding conditions and can be carried out in water.^[20] The process can also be used for a wide range of monomers. However, as the rate of initiation is the rate limiting step it is hard to control the molecular weight of the polymers formed. This is due to the propagation and termination step being faster than the initiation step so the polymers grow at different times with the initial polymers formed being longer due to more monomer being present.^[19]

Some of the key limitations of free radical polymerisation are that there is little control over the molar mass distribution due to the termination reactions of the growing radicals being diffusion controlled. Also, it is not possible to synthesise block copolymers due to the life time of the propagating chain being very short.

To control the spread of the molecular weight living radical polymerisation is used.^[20] Living polymerisation is polymerisation with the absence of chain termination and chain transfer reactions.^[21] This works by the radical species being reversibly deactivated and forming a dormant species. This allows the polymers to grow at a similar rate as there is equal chance of them being in the dormant state, and with the dormant state there is a lower radical concentration which means a lower rate of termination. This allows the polymers to form with relatively low poly-dispersity as the rate of growth of the polymer chains stay relatively constant through-out. This is show in Fig. 3 where polymers grown from free radical polymerisation have a greater molecular weight distribution.

Figure 3. Graph comparing molecular weight distributions of conventional free radical (left) against controlled living (right).^[22]

1.3 Reversible Addition Fragmentation Chain Transfer (RAFT)

Reversible addition fragmentation chain transfer is an example of a controlled polymerisation technique.^[23] It works by reversibly deactivating the radical, this prevents them terminating. This is done by species that reversibly react with the propagating radicals via chain transfer. Therefore the majority of propagating chains are in the dormant state.^[24]

Scheme 2. RAFT mechanism

There is an equilibrium between the active and dormant species which is called the RAFT main equilibrium (scheme 2), this allows all the polymer chains to have the same chance of growth. The RAFT polymerisation is performed by a macromonomer, this has a Z and R group. Propagating polymer radicals react with the RAFT macromonomer to form the intermediate shown in scheme 2. The R group acts as a homolytic leaving group, this means that it will easily be released forming the propagating radical species.^[25] For it to be a good homolytic leaving group it needs to have good radical stability. However just being a stable radical isn’t all that is required as it also needs to have a similar or better leaving group ability compared to the monomer.^[24] The Z group is important as it affects the rates of the addition of propagating radicals and the fragmentation of the intermediate radicals. It achieves this by controlling the reactivity of the C=S bond, which influences the rate of radical addition and fragmentation.

RAFT reduces the amount of termination by reducing the concentration of propagating radical polymer chains by causing them to be reversibly capped. This lowers the rate of termination due to there being less chance of radicals colliding and terminating. This is due to the rate of termination being proportional to the concentration of propagating radical squared. However, this also lowers the rate of propagation as it is also proportional to the amount of propagating radical polymer chains present. This means the rate of propagation is slower and more controlled, but the rate of termination is slowed more due to it being effected more by the change in concentration of the propagating radical.^[24]

The monomers used for RAFT polymerisation can be classified into two different groups. The first group is a less activated monomer (LAM) which have a double bond adjacent to a saturated carbon, an oxygen or nitrogen lone pair or the heteroatom of a hetero-aromatic ring.^[24] These LAMs are highly reactive to radical addition, so less active RAFT agents are required. The second group is a more activated monomer (MAM) and the double bond is conjugated to a carboxylic acid, an aromatic ring or a nitrile.^[25] Therefore these monomers are less reactive in radical addition so a more reactive radical agent is required.

The advantages of RAFT are that it provides controlled molecular weight polymers that have a very narrow polydispersity. RAFT can also be used to make block copolymers and other complex morphologies. Furthermore, it is compatible with a wide range of monomers and doesn’t require the protection of functional groups.^[23]

RAFT polymerisation can be used to form crosslinked and hyperbranched polymers. RAFT is prioritised over free radical polymerisation as it shows a delay in the onset of gelation and the molecular weight can be shown to grow at a more steady rate.^[26] A soluble hyperbranched species can also be formed if the ratio of crosslinker concentration to transfer agent is less than 1.^{[10, 15]}

1.4 Amphiphiles

An amphiphilic compound contains both hydrophilic and lipophilic properties. These properties mean that the compound acts differently depending on the solvent system.^[27] This allows amphiphiles to undergo self-assembly into different structures based on the surrounding solvent. In a water based solvent the hydrophobic regions try to avoid contact with the water which is called the hydrophobic effect. This effect leads to the molecule assembling into spherical or elongated tube like structures to form a hydrophobic core which is protected from the water by a shell of hydrophilic units. This process has an enthalpic and entropic contribution.^[28] The enthalpic contribution is due to the favourable interaction between the hydrocarbon chains. The entropic contribution is due to the layout of water molecules in solution. With spherical structures such as micelles, a shell of loosely associated water molecules is formed but with the free hydrocarbon chains the water molecules have more order, due to hydrocarbon chains producing a more locally ordered structure.^[28]

The self-assembly process can be expanded to polymers with amphiphilic polymers. These amphiphilic polymers must contain monomers which have hydrophobic groups and monomers which have hydrophilic groups. This allows them to have interesting properties which would be useful for applications such as paints, coatings, drugs, water treatment and personal care goods.^[29]

Hyperbranched polymers can also undergo self-assembly in certain solvents, depending on the functional groups present on the end of the chains.^{[14, 30]} Therefore changing the functional groups present will affect the solubility of the hyperbranched polymer.^[31]

1.5 Statistical Copolymers

One method of producing an amphiphilic polymer is to synthesise a statistical copolymer. A statistical copolymer consists of two or more monomer units joined together in a random order, which is entirely based on the probability of finding a monomer unit, this is effected by the mole fraction of each monomer.^[32] This allows the polymer to have varying properties depending on the ratio of each monomer present. Therefore, altering the composition of the polymer allows the properties to be tuned for its function.

This can be related to the idea of amphiphilicity if the polymer is made up of one monomer which is hydrophilic and one monomer which is hydrophobic. So, this means the polymer can be used in the same way as amphiphiles but with tuneable properties.

Statistical copolymers in aqueous media can form very complex structures due to the hydrophobic and hydrophilic interactions being spread across the chain instead of being localised like block copolymers. ^[33] The hydrophobic units usually want to associate due to hydrophobic attractions in water, whereas the ionic hydrophilic units usually want to be far apart from each other.^[34] This means the polymers usually aggregate together to form more favourable structures such as micelles.^[35] An example of how a statistical copolymer could act in water is the flower like micelle model where the hydrophilic groups loop out from the central core, protecting the inner hydrophobic core.^[33]

1.6 Block Copolymers

Block copolymers are made up of two or more units of homopolymers bonded together with a covalent bond at their terminal site. This difference in structure gives them different structures when compared with the random ordering of a statistical copolymer or a homopolymer.^[36]

As the different polymer units are separated it allows the formation of polymers with changing properties along the chain. This idea can be further expanded by forming multiblock copolymers with monomers having a range of different properties. This allows for precision controlled applications based on how the block copolymers self-assemble. However nature has shown a greater control in the formation of sequence controlled polymer with proteins which have a defined primary structure which controls self-assembly in the body with the tertiary structure. ^[37]

The common method of synthesis of block copolymer is to synthesise a polymer with one monomer then purify it to remove any unreacted monomers. After purification, a second block is grown onto the polymer chain and this can be repeated to grow larger alternating chains.^[38] Another method is to use living controlled polymerisation, this would involve adding one monomer until it has been fully converted to polymer and then adding a second monomer.^[23]

In aqueous solutions diblock copolymers usually form spherical micelles with the core containing the hydrophobic units and an outer shell of hydrophilic units.^[33] However to get the block polymers to dissolve in water usually requires them to first be dissolved in a solvent of intermediate polarity such as an alcohol and then slowly dispersed into water.^[39] The polymer micelles can be formed in aqueous solutions when the polymer concentration exceeds the critical micelle concentration.^[40] This means the structure of the micelles can be altered by changing the ratio of the different polymer blocks. The block copolymers would associate in solution to form micelle like structures to protect the hydrophobic core.^[41]

1.7 Structural Characterisation

1.7.1 Introduction

To find the structures of hyperbranched copolymers a variety of different characterisation techniques can be used which are described in this section.

1.7.2 Gel Permeation Chromatography

Gel permeation chromatography is a method of polymer analysis which provides the poly dispersity of the sample.^{[42, 43]} The relative molecular weight can also be calculated by using known standards which are similar to the polymer being analysed. It works by separating compounds based on their hydrodynamic volume, with the higher volume compounds eluting first.^[44] The separation is caused by the porous support in the column with the smaller compounds enter the pores and taking longer to elute^[43]. The eluting polymer is detected usually by a refractive index detector; this provides information on changes in concentration. This change in concentration is plotted against the time to show when the polymer leaves the column which can be compared to the calibration to find the molecular weight.

1.7.3 Dynamic Light Scattering

Dynamic light scattering is used to determine the size of particles in suspension in a solution. It works by light hitting small particles and being scattered in all directions. The intensity of this scattered light fluctuates over time, this is due to the particles undergoing Brownian motion. Brownian motion is where particles are moving due to random collisions with the liquid molecules surrounding the particle. This Brownian motion causes the particles to move around, with smaller particles moving quicker. This allows the size of the particle to be related to the speed by using the Stokes-Einstein equation.^[45] The scattered light also undergoes either constructive or destructive interference and this provides the intensity. However, as the particles are moving the constructive and destructive regions change in intensity. This allows the change in intensity from the initial point to be used to produce a correlation function which records the change in correlation of the intensity against time. This change in correlation can be used to find the size of particles as larger particles move more slowly the drop in the correlation function will be slower. This can be used to find the intensity at certain diameters, allowing an intensity distribution to be plotted.^[45] This is also gives the z average which is the average diameter from the intensity of the scattered light.

The intensity distribution can also be converted into the volume distribution by using Mie theory. The Mie theory assumes that the particle being analysed is spherical.^[45] The volume distribution shows how much a particle of a certain diameter scatters, whereas the intensity distribution gives how much light is scattered by particles in a diameter range. The volume distribution is also less reliable because any error present in the intensity distribution will mean there is a large error in the volume function as it is from the intensity function.

1.7.4 Small Angle X-Ray Scattering

Small angle X-ray scattering (SAXS) is a structural characterisation method used to determine averaged particle sizes and shapes.^[45] It is used in a lot of cases to study the structure of soft matter. SAXS is a very useful method as it is non-destructive, accurate and usually requires a very small amount of sample preparation. It also has a broad range of applications such as in polymers, colloids, biological materials, pharmaceuticals and many more.

X-rays can interact with matter causing either absorption or scattering. Absorption is where the atom absorbs the X-ray photon to remove an electron, this leaves a hole in the electronic structure of the atom. This causes rearrangement of the remaining electrons to fill the hole causing the release of X-ray radiation but at a different wavelength to the original X-ray. Scattering is the second interaction which can occur in two different ways, which are Compton scattering and Rayleigh or Thomson scattering.^[46] Compton scattering is where the photon collides with an electron and bounces back.^[47] The photon also loses some of its energy to the electron. The scattered radiation doesn’t provide any structural information as there is no particular phase relationship with the incident radiation. However, with Rayleigh or Thomson scattering there is no energy loss, but the electrons start oscillating at the same frequency as the original X-ray radiation. This means the electrons emit radiation at this frequency. This allows coherent scattering patterns to form which provide structural information.

Therefore, SAXS works by irradiating a sample with X-rays. This causes the atoms to in the sample to scatter the radiation in all directions. This provides background radiation which is shown to be almost constant at small angles. The particles in the sample also produce scattering as long as the particles are made of a different material or density compared to the solvent and are a similar size to the X-ray radiation. The difference in the scattering pattern of the background and the sample is then found, giving the scattering pattern caused by the particles.^[45]

The oscillations of the scattering patterns are related to the shape of the particle being analysed. This is called the form factor. For this to be found to sample must be dilute so that the particles don’t interact with each other. This means that the waves scattered by the different particles lack phase coherence. Therefore, the experimental scattering pattern is the form factor multiplied by the number of particles in the X-ray beam. Also, the observed scattering pattern only corresponds to the form factor of one particle if the particles are identical (monodisperse). However, in polydisperse samples containing a range of different particle sizes and shapes, the form factors of all the particle sizes are summed up to obtain the average scattering pattern.^[45]

To interpret the data, we need to consider the equation for the intensity change between the background and the sample. This is summarised in equation 1. With K being a constant made up of the particle contrast, volume and concentration. The other terms are the form factor P(q) and the structure factor S(q). The structure factor is based on the particle-particle interactions, however if the solution is dilute there are no particle-particle interactions so it will be equal to 1. Therefore, in a dilute solution the change in Intensity is only related to the change in the form factor.

∆Iq=K Pq Sq                                                                               (1)

To describe the scattering pattern various structural models are used. The simplest of these models is the sphere model. This model assumes the micelles is a spherical particle composed of homogeneously distributed matter. This gives the amplitude of a scattering object (F) as equation 2, with form factor being described in equation 3. The terms in the equation are the volume of the particle (V), the difference in the electron density between the sphere and the background environment (Δξ), the form factor in the case of a homogenous spherical particle(P(q,R)), the magnitude of the scattering vector (q) and the radius of the particle (R).^[45]

Fq=V Δξ Pq,R                                       2

Pq,R= 3[sin⁡qR-qRcosqR](qR)3                           (3)

The radius of gyration can also be calculated by using the Guiniers law to fit the scattering intensity at low q (eqn. 4).^[48] With G being the Guinier prefactor, R_g the radius of gyration and q being the scattering vector. This equation is then used to fit the higher q region to create the Unified Guinier/ power law equation (eqn. 5).^{[49] [50]}Where B is a prefactor which is specific to the type of power-law scattering as determined by the regime in which P falls. For Porod’s law P=4 for hard spheres and for Gaussian polymers P=2.

Iq≈G e-Rg2g23                                                 (4)

Iq≈G e-Rg2g23+Ber fqRg63qP                                    (5)

1.7.5 Rheology

Rheology is the study of how a material reacts when a force is applied. This is mainly focused on viscoelastic materials which can act as both solids and liquids.^[18] In the case of this report it is used to find how changing the concentration of the polymer solution affects the solution viscosity. To do this oscillatory experiments were performed to find the complex viscosity. First the linear viscoelastic region has to be found. This is a region where there is a linear trend between stress and strain.^[51] This was found by varying strain and recording the stress value. After this an angular frequency sweep was conducted at constant temperature and strain to find the complex viscosity.

1.7.6 Zeta Potential

Zeta potential is used to find the surface charge. This is an important property as it can determine the interactions between the particles. This means it can determine dispersion characteristics such as flocculation, viscosity, film forming characteristics and dispersion stability. The surface charge isn’t possible to measure directly but it is possible to find the zeta potential. The zeta potential is caused by the electrical double layer at the interface between the particle and the solvent. This means it provides information on the electrical interactions between the particles. This allows the determination of the functional groups present on the surface of the polymer.^[52]

The zeta potential is found from the measured mobility using the Ohshima relation to find function f₁ (eqn. 6).^[53] The terms of this equation are κ which is the inverse of the debye length and a which is the radius found by SAXS. The f₁ function is then used to find the zeta potential by using the henry equation (eqn. 7).^[53] The terms of this equation are u_e is the mobility, ε_rs is the relative permittivity of the electrolyte solution, ε₀ is the electric permittivity of vacuum, η is the dynamic viscosity of the solution and ζ is the zeta potential. This can then be used to find the charge by using the linearized Poisson-Boltzmann equation (eqn. 8).^[54] The terms for this equation are Z which is the charge, λ_B which is the Bjerrum length, e which is the elementary charge and k_B­which is the Boltzmann constant.

f1κa=1+121+2.5κa1+2 exp-κa-3            (6)

ue=23εrsε0ηζf1κa                              (7)

Z=a(1+κa)λBeζkBT                               (8)

1.8 General Aims

The aims of this project are to investigate the self-assembly in aqueous solutions of a range of hyperbranched copolymers. This involved synthesising a variety of hyperbranched polymer and copolymers via reversible addition-fragmentation chain transfer polymerisation. The solution structures of dilute polymer solutions were then analysed via a range of different analytical methods including small angle X-ray scattering and rheology. This data will then be used to find out how changing the types and compositions of hyperbranched copolymers affects the solution properties.

2 Results and Discussion

2.1 Kinetic Studies on Linear Poly (Butyl Methacrylate)

Linear polymerisations of butyl methacrylate (BMA) were performed to find the ideal conditions for high monomer to polymer conversion. Having a high conversion of monomer to polymer is important as it requires less purification and it gives more control on the length of the polymer chains formed. The following polymerisation was performed via reversible addition fragmentation chain transfer (RAFT). Raft was used as it produces polymers with a narrow polydispersity (PDI) and a controlled molecular weight. Six different conditions were tested and the monomer conversion was recorded at different time intervals during the polymerisation. The conditions varied were temperature, weight percent of the reaction, amount of RAFT agent, the solvent and initiator. The RAFT agent used was 4-cyano-4-(2-phenylethanesulfanylthiocarbonyl) sulfanylpentanoic acid (PETTC). The conversion was calculated by using ¹H NMR which showed a monomer peak around 6 ppm which decreased in size throughout the reaction and a peak around 4 ppm which was constant throughout the reaction, as it was present in both polymer and monomer. The general polymerisation to make linear poly (butyl methacrylate) is shown in scheme 3.

Scheme 3. Synthesis of linear poly (butyl methacrylate)

The first set of conditions tested was a temperature of 70 ºC, a concentration of 20 wt%, in isopropyl alcohol (IPA) and using 4,4’-azobis (4-cyanovaleric acid) (ACVA). The conversion of monomer to polymer was found to be initially quick but then it started to plateau after around 24 hours (Figure 4). However, a high conversion of 94 % was achieved after being left to react for 47 hours. The polymer made was analysed by gel permeation chromatography (GPC) and it showed a dispersity of 1.11 (fig. 16). This low value is expected with RAFT polymerisation due to it being a controlled polymerisation with most chains having an equal chance of growth.

Figure 5. The pseudo first order graph for the first set of polymerisation conditions.

This shows that too reach high conversion a long polymerisation is required of over 48 hours. This may be due to the butyl methacrylate having a slow propagation rate and the dormant chain being more stable. Furthermore, as the rate of propagation is proportional to the concentration of monomer, when less monomer is left the rate of propagation decreases. The pseudo first order conversion graph (fig. 5) allows the different rates to be compared quantitatively.

The second system involved increasing the concentration of the solution from 20 weight percent to 30 weight percent. Changing the weight percent changes the ratio of solvent to solid in the reaction, therefore increasing the weight percent means the monomer is more concentrated which makes the monomers more likely to interact. This change was shown to cause a higher conversion of around 95 % after 24 hours (Fig. 6). However, this conversion is still lower than required and leaving it for another 21 hours shows an increase to 98 % conversion which is high but still slow. The pseudo first order graph (fig. 7) shows that there was an increase in the rate of reaction as the gradient has increased. This shows that increasing the weight percent increases the rate of reaction. The GPC again shows a low dispersity of 1.14 which shows changing the weight percent didn’t really affect the spread of molecular weights of the polymers formed (fig. 16).

Figure 6. The conversion of monomer BMA to polymer at 70 ºC, at 30 weight percent in IPA. The final conversion reached is 98% at around 45 hours.

Figure 7. The pseudo first order graph for the second polymerisation conditions.

The third system involved changing the solvent system to toluene from IPA, the radical initiator also needed to be changed as ACVA isn’t soluble in toluene so 2,2’-azobis (2-methylpropionitrile) (AIBN) was used instead. This showed an increase in the initial rate of polymerisation however the reaction seemed to plateau more quickly and over a period of 24 hours a lower conversion was reached in comparison to the previous reaction (fig. 8). The pseudo first order graph (fig. 9) shows that the overall rate has increased but not by much however, the initial rate of reaction is shown to be a lot quicker. The greater initial rate of conversion may have been due to the toluene being a better solvent with the monomer being able to move around more freely, or it may have been due to AIBN being a better radical initiator than ACVA. The GPC (fig. 16) showed a low dispersity of 1.15, which shows that changing the solvent and initiator used didn’t affect the dispersity as it still followed the expected dispersity of RAFT reactions.

Figure 8. The conversion of monomer BMA to polymer at 70 ºC, at 30 weight percent in toluene. The final conversion reached is 93% at around 24 hours.

Figure 9. The pseudo first order graph for the third reaction conditions with the initial and overall rate labelled separately.

The fourth system involved keeping the toluene solvent environment and performing the reaction at a higher temperature of 80 ⁰C. This reaction showed an even faster initial rate of polymerisation but again the rate of conversion started to plateau early than the reactions performed in IPA and it didn’t achieve as high conversion (fig. 10). The pseudo first order graph (fig. 11) shows that again the overall rate doesn’t increase by much but the initial rate shows a large increase. The higher initial rate may have been caused by the increase in the temperature which may have caused the initiator to thermally decompose quicker so more chains can be activated. Additionally, the monomer and propagating radical chain have more kinetic energy so they are more likely to collide. The increase in temperature was shown to not affect the dispersity as it was still low at 1.14 (fig. 16).

Figure 10. The conversion of monomer BMA to polymer at 80 ºC, at 30 weight percent in toluene. The final conversion reached is 91% at around 20 hours.

Figure 11. The pseudo first order graph for the fourth reaction conditions with the initial and overall rate labelled separately.

The fifth system involved increasing the concentration of the reaction from 30 to 40 wt %. This reaction also showed an increase in the initial rate of reaction when compared with the previous reactions (fig. 12). However, it still had the same problem of plateauing too early to achieve a high enough conversion. The pseudo first order graph (fig. 13) also shows that as before the initial rate seems to be a lot higher compared to the overall rate. This may be something to do with the solvent being toluene as this difference in rate of reaction only seems to be present in the reaction systems containing toluene. The GPC again showed a low dispersity of 1.15 (fig. 16).

Figure 12. The conversion of monomer BMA to polymer at 80 ºC, at 40 weight percent in toluene. The final conversion reached is 94% at around 26 hours.

Figure 13. The pseudo first order graph for the fifth reaction conditions with the initial and overall rate labelled separately.

One final reaction was performed with linear poly (butyl methacrylate) but this one involved aiming for a lower degree of polymerisation. This involves using more RAFT agent which means more polymer chains are present, so more chains should be growing meaning the rate of conversion should be higher. The rest of the conditions were the same as the previous linear conditions. This gave a high conversion of 98% at 28 hours in and it was shown to have a quick initial rate (fig. 14). The pseudo first order graph (fig. 15) also supports this as it shows a large initial rate compared to the other reaction systems and it slows down throughout the reaction. The GPC showed a higher dispersity of 1.57, which may have been due to there being more chains growing at once and the stirring not being efficient enough to allow for equal growth chance.

Figure 14. The conversion of monomer BMA to polymer at 80 ºC, at 40 weight percent in toluene with an increased amount of RAFT agent. The final conversion reached is 98% at around 28 hours.

Figure 15. The pseudo first order graph for the sixth reaction conditions with the initial and overall rate labelled separately.

Using these experiments the conditions for the hyperbranched polymerisation to achieve a high conversion were decided. Initially this involved using toluene as the solvent with AIBN at 80 ⁰C with a weight percent of 30%. However, this was changed due to the toluene being hard to precipitate the hyperbranched polymer from, due to it involving multiple steps compared to IPA. Furthermore, when performing a reaction to synthesise a statistical copolymer the polymer formed a gel suggesting that either the initial rate of reaction is too fast causing a polymer network to form or that the polymer based upon methacrylic acid isn’t very soluble in toluene. This meant that the solvent system was changed back to isopropyl alcohol due to it providing the product in a clean powder form after one precipitation step. The final selected conditions to achieve a high conversion for hyperbranched polymers was 20 weight percent in isopropyl alcohol at 75 °C with ACVA as the initiator.

Figure 16. The gel permeation chromatography data for the linear polymers.

The gel permeation chromatography data (fig. 16) of the linear polymers shows the first five to all have a similar molecular weight which is expected as they had the same ratio of RAFT agent to monomer. The sixth polymer system is shown to have a small molecular weight which is expected due to there being more chains formed caused by the increase in the amount of RAFT agent. They all shows a small poly dispersity range which is also expected from RAFT polymerisation due to it being a controlled polymerisation technique.

2.2 Synthesis of Hyperbranched Poly (Butyl Methacrylate)

Hyperbranched poly (butyl methacrylate) was synthesised using the previously selected conditions of 20 wt % in isopropyl alcohol at 75 °C. The method of synthesis was the same as for the linear poly (butyl methacrylate) but with the addition of the cross-linker ethylene glycol dimethacrylate (EGDMA) to cause the branching (Scheme 3). This showed a high conversion of monomer to polymer of around 98%. The pseudo first order graph also shows that it had a high rate of reaction (fig. 17).

Scheme 3. Synthesis of hyperbranched poly (butyl methacrylate)

Figure 17. The pseudo first order graph for the synthesis of hyperbrancehd poly (butyl methacrylate).

To confirm that branching had occurred polymers were analysed by GPC which showed a high dispersity (fig. 18) of 3.44 which is expected for a hyperbranched. A high dispersity is expected due to the branching being uncontrolled which means the molecular weight of the hyperbranched polymers formed are less controlled. Therefore, the GPC trace confirms that the polymer synthesised isn’t purely linear and some branching is present. There is also a lower molecular single peak showing there is a small amount of monomer present.

Figure 18. Gel permeation chromotography trace of hyperbranched poly (butyl methacrylate).

2.3 End Group Modification of the Hyperbranched Poly (Butyl Methacrylate)

The synthesised hyperbranched polymers should have polymers chains with one end of a chain having a carboxylic acid unit from the initiation due to the RAFT mechanism adding the R group of the RAFT agent, and the other end being the Z group of the RAFT agent as shown in figure 19. As the polymers need to have hydrophilic groups to be soluble in water, more chains ending in carboxylic acid will help them dissolve. This means that replacing the chain ends capped with the phenyl rings with a carboxylic acid will make it more soluble in water.

Figure 19. A diagram of the hyperbranched polymer formed to label where the chain ends originate from in the original RAFT agent.

To achieve this substitution, the end group modification method demonstrated by Perrier was used.^[55] This involved dissolving the hyperbranched polymer in IPA and adding a 30-fold excess of ACVA to the polymer and heating for 3 hours under nitrogen (scheme 4). This should cause the triothiocarbonate end of the RAFT agent to be removed and replaced with the 4-cyanovaleric acid.

Scheme 4. End group modification of hyperbranched poly (butyl methacrylate).

The loss of the trithiocarbonate was initially confirmed to have taken place due to the colour change of the polymer from yellow to white (fig. 20). This is due to the trithiocarbonate absorbing visible light causing it to become coloured, however when it is removed the polymer becomes colourless.

Figure 20. A sample of the hyperbranched poly (butyl methacrylate) before (left) and after (right) undergoing the end group modification.

Also, a proton NMR was performed on the sample before and after the reaction and it showed a decrease in the size of the aromatic peak. This proton NMR confirms the loss of the aromatic end group, however there is still a small amount of the aromatic region present showing not all the ends are changed.

This reaction was later shown to be necessary because when preparing the water based solutions for analysis of the original hyperbranched poly (butyl methacrylate), it would precipitate when water was added. However, the acid functionalised polymer remained soluble when water was added showing there are more hydrophilic groups present.

The GPC chromatogram matches the chromatogram of the original hyperbranched polymer it was made from, this shows the branching and general structure of the polymer have been left unchanged. This is useful as it allows a normally water insoluble polymer to become soluble in water with no change to core structure of the polymer (fig. 21).

Figure 21. GPC of the original hyperbranched poly (butyl methacrylate) (orange) overlaid with the end group modified version (blue).

2.4 Hyperbranched Poly (Butyl Methacrylate-Stat-Methacrylic Acid)

The next step involved the synthesis of statistical copolymers of methacrylic acid and butyl methacrylate. They were synthesised in a similar way to the hyperbranched poly (butyl methacrylate) but methacrylic acid was added at the start (scheme 5). The amounts of methacrylic acid and butyl methacrylate were varied for each of the three polymers made, with the ratio of BMA to MAA being 7:3, 8:2 and 9:1. Samples were taken throughout the reaction and analysed by NMR to calculate the conversion of the BMA monomer to polymer. The pseudo first order reaction graphs show that the rate of reaction of the methacrylic acid is slower than the rate of reaction of the butyl methacrylate (fig. 22). This may be due to there being a lower concentration of MAA compared to the concentration of BMA. The conversion of BMA was 95 % which was higher than the conversion of the MAA which was 93%. This is again probably due to there being more BMA than MAA so it may be easier for it to react.

Scheme 5. Synthesis of poly(BMA-stat-MAA)

Figure 22. The pseudo first order graph for the conversion of BMA (blue) and MAA (orange) in the synthesis of poly(BMA-stat-MAA).

Samples of the statistical copolymer made were then alkylated by reacting them with a base to remove the hydrogen from the carboxylic acid and then adding benzyl bromide which reacts with the carboxylic acid end replacing the acidic proton and forming an ester. This was done so that ¹H NMR’s could be run to find the amount of MAA to BMA as the hydrogens in the aromatic region should be related to the number of the MAA units. The ¹H NMR’s gave the actual compositions of the polymers, which is shown in table 1.

	Expected ratio		Ratio from NMR		Degree of polymerisation
Polymer	BMA	MAA	BMA	MAA	BMA	MAA
1	0.80	0.20	0.78	0.22	63.0	17.3
2	0.70	0.30	0.72	0.28	80.1	31.3
3	0.90	0.10	0.90	0.10	81.0	8.6
4	0.70	0.30	0.69	0.31	63.2	27.8

    Table 1. The expected and actual compositions, and the degree of polymerisations found from the ­¹H NMR’s of the alkylated statistical copolymers.

Table 1 shows that the actual compositions are very similar to the compositions that were expected showing that the polymers were successfully synthesised with the correct compositions.

The success of the alkylation was further confirmed to have worked by running a solid-state IR of the sample before and after alkylation as it shows the loss of the carboxylic acid. The loss is shown by the change in transmittance at the 3200 cm^-1 region of the IR which corresponds to the carboxylic acid part (fig. 23).

Figure 23. Solid state IR of statistical copolymer 4 before (blue) and after (orange) undergoing alkylation.

Also, this allowed a GPC to be run due to the GPC used containing a styrene-divinylbenzene column which the methacrylic acid would have stuck to if it wasn’t alkylated. The broad peak shows that the polymer is hyperbranched because if it was a single chain you would expect a single peak with low poly dispersity due to it being made by RAFT (fig. 24). There was also a thin peak at a lower molecular weight which would have been formed by some linear polymerisations taking place instead of branching.

Figure 24. GPC of statistical copolymer 4

2.5 Hyperbranched Poly(Butyl Methacrylate)-Block-Poly(Methacrylic Acid)

Block copolymers were synthesised by using the previously synthesised hyperbranched poly (BMA) and adding a certain number of units of MAA to the end of the chain (scheme 6). Five, ten and twenty methacrylic acid units were added to the end of the hyperbranched poly (BMA).

Scheme 6. Synthesis of hyperbranched poly(BMA)-block-poly(MAA).

The purpose of this was to see how increasing the number of hydrophilic units affected how they self-assembled in an aqueous environment. The conversion of the methacrylic acid was monitored throughout the reaction by ¹H NMR spectroscopy. The conversion of MAA was shown to be about 81 % and the initial rate of conversion appears to be higher than the overall rate of conversion. This may be due the concentration of the MAA being low to begin with and then dropping quickly meaning the rate plateaus.

Figure 25. The pseudo first order graph of the synthesis of block copolymer 1 for the conversion of MAA.

Also, another block copolymer was synthesised by adding five methacrylic acid units to a different sized core to see how it affected particle size. This would allow it to be compared with the other sized core with five added methacrylic acid units.

Samples of each block copolymer were alkylated to remove the carboxylic acid part so that the copolymer could be run down the GPC. This confirms that the copolymer synthesised was a hyperbranched polymer (fig. 26) as there isn’t just a single peak present as would have been expected if it was a linear polymer. This also confirms a large polydispersity which is expected from hyperbranched polymers due to their synthesis being relatively uncontrolled. Comparing the GPC data from the original hyperbranched core with the block copolymer, it can be seem that they have a similar molecular weight and a similar pattern which is expected as they are very similar structures.

Figure 26. GPC of the original hyperbranched poly(BMA) (orange) and the block copolymer made from it (blue).

Two of the block copolymers were purified by precipitating them into a mixture of methanol and water, whereas the other two were precipitated into petroleum ether. This was due to the block copolymers with five methacrylic acid units added, being soluble in petroleum ether. The ones precipitated in the water methanol mix still had water present in them before alkylation so the water would have just destroyed the alkylating agent. This meant they couldn’t be alkylated so GPC data couldn’t be gathered for one of them, however as the core was already analysed by GPC it is known that they are hyperbranched.

An NMR of the alkylated block copolymer was run and compared with the original block copolymer to find out the actual composition. This was done by using the ratio of the aromatic peaks in the NMR and the peaks corresponding to the butyl methacrylate.

	Expected ratio		Ratio from NMR		Degree of polymerisation
Polymer	BMA	MAA	BMA	MAA	BMA	MAA
1	0.91	0.20	0.92	0.08	48.6	4.2
2	0.83	0.30	0.84	0.16	48.6	9.3
3	0.71	0.10	0.72	0.28	48.6	18.9
4	0.93	0.30	0.93	0.07	62.7	4.6

Table 2. The expected and actual compositions, and the degree of polymerisations found from the ­¹H NMR’s of the alkylated block copolymers.

The compositions in table 2 show that the compositions from ¹H NMR is very similar to the expected ratio. This shows that the block polymerisation was successful and the right amount of MAA units were added to most of the polymers.

2.6 Dynamic Light Scattering

The polymers synthesised were made up to a concentration of 50 weight percent in IPA with a small amount of triethanolamine to neutralise the carboxylic acid and then slowly diluted down to 0.5 weight percent by the addition of water. DLS measurements were then run on filtered samples of these.

	Degree of polymerisation
Polymer	BMA	MAA	Z-average / nm
Stat
1	63.0	17.3	11.8
2	80.1	31.3	11.9
3	81.0	8.6	48.0
4	63.2	27.8	14.7
Block
1	48.6	4.2	62.8
2	48.6	9.3	31.9
3	48.6	18.9	58.1
4	62.7	4.6	24.9

Table 3. The DLS data showing the Z average of the different copolymers

The z average diameter was used to compare the sizes of the copolymers formed. The size of the statistical copolymers was shown to increase with the z average of the 90:10 (BMA:MAA) being the largest and the 70:30 and 80:20 copolymers having a lower z average. This may be due to the 90:10 copolymer having only a small amount of methacrylic acid compared to the butyl methacrylate so it may be more likely to form larger aggregates to keep more of the hydrophobic butyl methacrylate from touching the water by forming an inner core containing IPA. Whereas the 80:20 and 70:30 have more methacrylic so they either form smaller aggregates or may just stay as single unimolecular micelles. The idea of aggregates forming is supported by the intensity data due to it showing a range of differently sized particles (fig. 27). However, when comparing the volume data (fig. 28) there is shown to be one main peak for each copolymer system with the two 70:30 ones being very similar, 80:20 ones being slightly larger and the 90:10 one being the largest. However, as the volume data is found from the intensity data it isn’t as reliable.

Figure 27. DLS intensity distribution data for statistical copolymer 1 (blue), 2 (red), 3 (green) and 4 (purple).

Figure 28. DLS volume distribution data for statistical copolymer 1 (blue), 2 (red), 3 (green) and 4 (purple).

The z-averages for the block data show an increase in the size of the hydrodynamic radius apart from the first point when more methacrylic acid units are added. The first point shows a larger than expected radius, this may be due to the formation of aggregated structures due to it not containing enough methacrylic acid units in one polymer to stabilise to core against the water. However, block copolymer 2 and 3 show a linear increase when more methacrylic acid is added. The intensity distribution (fig. 29) shows that there seems to be two different particle sizes in solution due to all of the intensity distributions showing two separate peaks. This may be due to them forming either uni-micelle structures or forming larger aggregates. However, the volume distribution (fig. 30) shows that there is one particle size which dominates. This data also shows a linear trend with the block copolymers with more MAA units added having a larger hydrodynamic radius. But the volume data is less reliable as it is found from the intensity data.

Figure 29. DLS intensity distribution data for block copolymer 1 (purple), 2 (red), 3 (green) and 4 (blue).

Figure 30. DLS volume distribution data for block copolymer 1 (purple), 2 (red), 3 (green) and 4 (blue).

The DLS data for the end group modified hyperbranched poly (BMA) shows a single peak in the intensity graph. The z-average is also shown to be 79.8 nm, this is shown to be larger than the sizes found for the copolymers. This suggests that the polymer particles aggregate and form a larger structure, this may be due to there being less hydrophilic groups so it is hard it to form a uni-micelle like structure. Therefore, it may arrange around the IPA in the aqueous solution to protect the hydrophobic core. The volume distribution also shows a single peak which backs up the idea of there being only one particle size present.

Figure 31. DLS data for end group modified hyperbranched poly(BMA) with intensity distribution (left) and volume distribution (right).

2.7 Small Angle X-Ray Scattering of Dilute Solutions

Polymer solutions were made up to 50 wt % in IPA with TEA to neutralise the carboxylic acid parts. Then they were diluted down to 1 wt % by the slow addition of water. SAXS analysis was then performed on 3 of the statistical copolymer and 3 of the block copolymers. The scattering patterns found were all fit the spherical micelle model, this confirms that they all formed spherical self-assemblies.

	Degree of polymerisation
Polymer	BMA	MAA	Radius / Å
Stat
1	63.0	17.3	58.31
2	80.1	31.3	49.45
3	81.0	8.6	103.49
4	63.2	27.8	–
Block
1	48.6	4.2	–
2	48.6	9.3	102.37
3	48.6	18.9	106.24
4	62.7	4.6	114.70

Table 4. The radius found by SAXS analysis.

The statistical copolymers were found to show a trend of increasing size when a greater percentage butyl methacrylate was present. This may also be due to the methacrylic acid parts of the polymer being solvated so not being shown as part of the radius found via SAXS. This also matches the DLS data with the 90:10 copolymer having a lot larger radius compared to the 80:20 and 70:30. This may further support the idea of aggregates being more common when les methacrylic acid units are present. Also, the SAXS patterns (fig. 32) for the 80:20 and 70:30 copolymers are shown to be very similar whereas the 90:10 pattern is shown to be different with the drop being at a lower q value.

Figure 32. SAXS patterns of aqueous solution of statistical copolymers; where statistical copolymer 1 is blue, 2 is red but multiplied by ten so it can be overlaid and 3 is green and multiplied by 1000.

The block copolymers were found to show similar radius sizes and SAXS patterns (fig. 33) which is expected due to the X-Ray scattering being due to the butyl methacrylate core and not being affected by the solvated methacrylic acid units. This is supported by radius of block copolymer 4 which was synthesised from a butyl methacrylate core which was slightly larger than the rest, shows a slightly larger radius. The SAXS patterns (fig. 33) are shown to be almost identical showing that the butyl methacrylate cores are very similar.

Figure 33. SAXS patterns of aqueous solution of block copolymers; where block copolymer 4 is blue, 2 is red but multiplied by 100 so it can be overlaid and 3 is orange and multiplied by 10000.

The end group modified hyperbranched polymer couldn’t be run in the SAXS for the aqueous solution due to it having particles that were too large so they couldn’t be detected.

Polymer solutions were also made up at 1 wt % in tetrahydrofuran (THF). This was performed due to THF being a good solvent for the copolymer which would allow the radius of gyration (R_g­) to be observed. This allowed a model to be fit to the scattering pattern which could be used to find the R_g values (Table 5).

	Degree of polymerisation
Polymer	BMA	MAA	Rg / Å
Stat
1	63.0	17.3	92.89
2	80.1	31.3	65.58
3	81.0	8.6	60.12
4	63.2	27.8	–
Block
1	48.6	4.2	–
2	48.6	9.3	73.20
3	48.6	18.9	99.94
4	62.7	4.6	97.30

Table 6. The radius of gyration found by SAXS analysis of THF.

The data in table 6 shows that for the statistical copolymers the radius of gyration seems to be larger when an intermediate amount of MAA is present as the 80:20 polymer is larger than both the 70:30 and 90:10 polymer. When compared with the aqueous SAXS data it shows that for statistical copolymer 1 and 2 that the size is smaller in water which suggests that they self-assemble to smaller structures when in an aqueous environment. However, for statistical copolymer 3 the size in water is larger which supports the idea of the formation of an aggregated structure forming. The SAXS patterns for the statistical copolymers show they all have a similar shape due to the patterns being very similar (Fig. 34).

Figure 34. SAXS patterns of THF solution of statistical copolymers; where statistical copolymer 1 is blue, 2 is orange but multiplied by 10 so it can be overlaid and 3 is green and multiplied by 100.

The data in table 6 shows that for the block copolymers the copolymers with a greater amount of MAA show a larger radius of gyration. Also, when more BMA is present the size of the radius of gyration also increases when comparing block copolymer 4 to block copolymer 2. This doesn’t really match the aqueous SAXS data however that is probably due to the solvated MAA not showing up in the SAXS pattern. The SAXS patterns are shown to be very similar suggesting a similar shape (Fig. 35).

Figure 35. SAXS patterns of THF solution of block copolymers; where block copolymer 2 is blue, 3 is orange but multiplied by 10 so it can be overlaid and 4 is green and multiplied by 100.

The end group modified hyperbranched polymer was able to be run due to the particle size being smaller as it was fully dissolved. This gave a radius of gyration value of 52 Å which is a lot smaller than the size found by DLS for the aqueous solutions. Therefore, this supports the idea of the formation of a larger aggregated structure forming. The SAXS pattern is shown in figure 36.

Figure 36. SAXS pattern of THF solution of the end group modified hyperbranched polymer.

2.8 Zeta Potential

Polymers solutions were made up to 50 wt % in IPA with some TEA to induce solubility in water, and then slowly diluted down to 1 wt % with water. These solutions were then diluted down to 0.25 wt % by the addition of a potassium chloride salt solution. The zeta-potential was then measured and for the statistical copolymers a range of values from -55 mV to -73 mV was found (table 6). This shows that the structures formed in solution are relatively stable. Using the radius found from the SAXS data the charge was calculated and this was shown to have no overall trend. This was shown as there is no relationship between the charge and the amount of methacrylic acid units present. However as statistical copolymer 3 has the largest SAXS radius and it was suggested to be due to aggregation, this could support why the charge is higher as it could be explained by the formation of an aggregated structure.

	Degree of polymerisation
Polymer	BMA	MAA	Zeta potential / mV	Charge
Stat
1	63.0	17.3	-73.36	-36.14
2	80.1	31.3	-55.61	-22.03
3	81.0	8.6	-66.31	-73.33
4	63.2	27.8	–	–
Block
1	48.6	4.2	–	–
2	48.6	9.3	-60.85	-66.21
3	48.6	18.9	-67.26	-77.32
4	62.7	4.6	-70.89	-91.40

Table 6. Zeta potential and charge data for the copolymers.

2.9 Rheology

Polymer solutions were made up by dissolving a hyperbranched statistical copolymer of methacrylic acid and butyl methacrylate in a ratio of 30:70 (MAA:BMA), in IPA to 50 wt % with a small amount of triethanolamine (TEA) to neutralise the carboxylic acid units in the methacrylic acid. Then water was slowly added to form polymer solutions with a weight percent of 50, 40, 35, 30, 25, 20 and 10%.

Oscillatory rheology experiments were carried out on these solutions to determine how changing the concentration affects the solution viscosity. A strain sweep was conducted first on each solution to determine the region of linear viscoelasticity. After finding this strain, an angular frequency sweep was performed at constant strain and temperature to find the complex viscosity. The identified complex viscosity was plotted against the polymer concentration and it was also compared to previously identified data for a linear statistical polymer (how reference Toms data) which had the same ratio of BMA to MAA (fig. 37).

Figure 37. A graph describing the complex viscosity of the polymer solutions in a ratio of water to IPA against it concentration; where the hyperbranched polymer is blue and the analogous linear polymer is orange.

The lower concentrations of 10 and 20 wt % show a very low complex viscosity. This may be due to the hyperbranched polymers acting as uni-micelles, with the hyperbranched polymers forming single units. This could be due to the high water content meaning that the polymers find it easier to stabilise themselves instead of forming larger units. Compared to the linear data the complex viscosity of the hyperbranched polymers are shown to be lower, this is expected due to hyperbranched polymers having a lower viscosity due to less chain entanglement.

The data then shows a large increase in viscosity when going to 25 wt % this may be due to the micelles not being fully formed at this concentration. This matches with the peak in the linear data, however there is a large error for this point which may be why it is out of place.

There is a peak viscosity at 35 wt %, this may have been caused by the polymers going from fully dissolved in IPA to being partially dissolved in water. However, as the majority of the polymer is made of hydrophobic BMA it would have preferred to remain in the IPA. Therefore, it may have formed links with other regions of IPA, forming an interlocking network with other hyperbranched polymers, which would have increased the viscosity.

The viscosity drops at 40 wt % which is probably due to most of the polymers being fully dissolved in IPA and far apart. There is then an increase in viscosity to 50 wt % which is expected as it is a high weight percent so it would be very viscous and there is no water present.

3 Conclusions

In conclusion six linear polymers were successfully synthesised via RAFT polymerisation using different reaction conditions to discover the best conditions for high conversion of monomer to polymer. The conditions were found to be a temperature of 75 °C, a weight percent of 20, using IPA as the solvent, ACVA as the initiator and leaving it to polymerise for 24 hours. These linear polymers were analysed via GPC which showed they all had a low polydispersity. A hyperbranched polymer of BMA was successfully synthesised by using the reaction system previously selected conditions. This was proved to be branched via a GPC chromatogram showing a large polydispersity which is due to the uncontrolled nature of the branching. The hyperbranched polymer was then modified by changing the end groups to more hydrophilic end groups so it would be more soluble in water. This was shown to be successful due to the colour of the polymer changing from yellow to white and the NMR showing a decrease in the aromatic region which was on the end group that was replaced.

Four statistical copolymers of MAA and BMA were synthesised. By alkylating the MAA it was possible to calculate the composition via the ¹H NMR. The copolymer compositions of BMA:MAA was shown to be 78:22, 72:28, 90:10 and 69:31. They were also confirmed to be branched by GPC analysis showing a high poly dispersity. Four block copolymers were synthesised by adding MAA units to a hyperbranched poly(BMA) core. By alkylating the MAA units the number of added MAA units was shown to be 4.2, 9.3 and 18.9 for one BMA core and 4.6 MAA units were added to a different BMA core. The GPC showed that they were still hyperbranched as they had similar chromatograms to the original hyperbranched BMA.

The statistical copolymers, block copolymers and end group modified polymer were made into solutions by making up 50 wt % solutions in IPA and then slowly diluting the solution down with the addition of deionised water. DLS was conducted on these solutions and for the statistical copolymers it was shown that the size was higher with the 90:10 polymer suggesting the formation of an aggregated species, whereas the 78:22, 69:31 and 72:28 polymers showed a similar size suggesting the formation of a uni-micelle structure. The DLS data for the block copolymers showed a similar trend with the lower MAA content polymers having a larger size than expected. However, their sizes are larger than the statistical polymers suggesting more interaction with the water forming a larger hydrodynamic size. The DLS data for the end functionalised hyperbranched poly (BMA) showed a single particle size which was larger than the other polymers, this suggests that the polymer particles aggregated together. SAXS was conducted on dilute polymer solutions of the block and statistical copolymers, this gave the radius of the unsolvated part and scattering patterns. The scattering patterns all fitted the spherical model showing all the particles were spherical. The statistical copolymers SAXS radius data showed an increase in particle size when there was a lower ratio of MAA present, this may suggest aggregation being more common when less MAA is present. The block copolymer SAXS radius data shows that the two made from the same core had a similar radius which is expected due to the SAXS data not showing the solvated parts which would have been the MAA. Also, the block copolymer with a slightly larger core was shown to have a slightly larger SAXS radius suggesting a trend between the BMA core and the radius. SAXS measurements were carried out on 1 wt % THF solutions this confirmed all the polymer were spherical and the radius of gyration was found. For the statistical copolymers the radius of gyration of the 80:20 copolymer was shown to be larger with the 70:30 and 90:10 being smaller. Furthermore, when compared with the SAXS radius from the aqueous data the 70:30 and 80:20 polymer have a smaller radius suggesting they form more compact uni-micelles, whereas the 90:10 polymer has a larger radius supporting the idea of it forming aggregated structures. The THF SAXS data for the block copolymers shows that increasing the amount of MAA or BMA increases the radius of gyration. Zeta potential data was also conducted on the dilute polymer solutions of the block and statistical copolymers. For the statistical copolymers, there was shown to be no trend between the amount of MAA units and charge. However, the charge followed a similar trend to the size found by SAXS and DLS with the larger statistical copolymer having the most negative charge. For the block copolymers, there was shown to be no trend between MAA amount and charge.

Rheology measurements were carried out on the statistical copolymer with the 69:31 ratio and solution of 50, 40, 35, 30, 25, 20 and 10 wt % was made up in the same way as before. These showed an increase as the wt % increased which was expected however there was a peak in viscosity at 35 wt %. This peak may have been caused by the formation of a network of polymer micelles in solution, and when the IPA concentration is increased these fully dissolve lowering the solution viscosity.

This work could be further expanded by producing more copolymers of different ratios and checking if they show a trend. Furthermore rheology measurements could be performed on statistical copolymers with different BMA:MAA ratios to see if they have a similar trend in viscosity. The rheology data could also be expanded to block copolymers to see if they show a similar trend to statistical copolymers. SAXS measurements could also be carried out on a range of different aqueous wt % solutions to monitor the self-assembly of the hyperbranched polymers in solution.

4 Experimental

4.1 Materials

Butyl methacrylate (99%) and methacrylic acid (99%) were purchased from Sigma-Aldrich (UK) and the inhibitor was removed by passing them through alumina before use. Ethylene glycol dimethacrylate was purchased from Sigma-Aldrich (UK) and filtered with alumina to remove the inhibitor before use. Carbon disulfide, 2-phenylethanethiol, sodium hydride, triethanolamine, benzyl bromide, caesium carbonate and other reagents were all purchased from Sigma-Aldrich (UK) and used without purification. Solvents isopropyl alcohol, methanol, diethyl ether, toluene, petroleum ether, ethyl acetate, dimethylformamide and d⁶-acetone were purchased from Sigma-Aldrich and used as received. Deuterated chloroform and HPLC grade THF were purchased from VWR (UK). Deionised water was obtained using an Elga Elgastat Option 3A water purifier.

4.2 Synthesis of PETTC

2-Phenylethanethiol (4 g, 28.9 mmol) was added over ten minutes to a stirred solution of sodium hydride (60% in oil) (1.32 g, 33.0 mmol) in diethyl ether (75 ml) at a temperature between 5-10 ^oC. Hydrogen was evolved and the solution became a thick white slurry of sodium phenylethanethiolate which was stirred for 30 minutes. The reaction mixture was cooled to 0°C and carbon disulphide (2.28 g, 30 mmol) was added portion wise. The yellow precipitate of sodium 2-phenylethanetrithiocarbonate was filtered off. The yellow precipitate was then dissolved in diethyl ether (100 ml) and treated with solid iodine (5 g, 20 mmol). The solution was stirred for 1 hour and then the white solid sodium iodide was removed by filtration leaving a dark orange filtrate. The filtrate was washed with aqueous sodium thiosulfate (7.5 g in 300 ml of water) to remove excess iodine. The solution was dried using anhydrous sodium sulfate and solvent was removed under vacuum leaving a residue of bis-(2-phenyl ethane sulfanyl thiocarbonyl)-disulfide (3.6 g, 8.45 mmol, 58.5 %) as a yellow solid.

4.4’-azobis(4-cyanopentanoic acid) (ACVA) (3.62 g, 12.9 mmol) was added to a solution of bis-(2-phenyl ethane sulfanyl thiocarbonyl)-disulfide (3.6 g, 8.45 mmol) in ethyl acetate (120 ml) and degassed under nitrogen for 30 minutes. The solution was refluxed under a dry nitrogen atmosphere for 18 hours. The reaction mixture was then cooled to room temperature and the volatiles were removed under vacuum leaving an orange oil. This oil was washed with water (5 × 40 ml). The organic layer was dried using magnesium sulfate and the solvent was removed under vacuum. The product was recrystallized from ethyl acetate to yield 4-cyano-4-(2-phenylethanesulfanylthiocarbonate)sulfanyl pentanoic acid (PETTC) as orange crystals (0.414 g, 1.22 mmol, 8.4 %). ¹H NMR (400 MHz, CDCl₃, 298K) δ(ppm) = 1.9 (2, 3H, -CH₃), 2.3-2.6 (m, 2H, -CH₂), 2.7 (t, 2H, -CH₂), 3.0 (t, 2H, -CH₂), 3.6 (t, 2H, -CH₂), 7.2-7.4 (m, 5H, aromatic).

4.3 Synthesis of Linear Poly (Butyl Methacrylate)

See table 7 for reagent quantities. The procedure describes the quantities, solvent, initiator, time and temperature to synthesise linear polymer 1.

			Solvent	Initiator
Polymer	BMA / g	PETTC / g	IPA / g	ACVA / g	Temperature / °C	Time / h
1	5.1	0.059	25.80	0.0128	70	47
2	5.0	0.058	11.83	0.0156	70	45
			Toluene / g	AIBN / g
3	5.1	0.056	11.82	0.0096	70	24
4	5.1	0.057	11.82	0.0092	80	20
5	5.2	0.058	7.83	0.0093	80	26
6	5.1	0.241	12.45	0.0389	80	28

Table 7. The reagent quantities and conditions used to synthesise linear P(BMA).

4-cyano-4-(2-phenylethanesulfanylthiocarbonate)sulfanyl pentanoic acid (PETTC) (0.0585 g, 0.17 mmol) (to precise) was added to a 100 cm³ round bottom flask along with butyl methacrylate (5.1 g, 0.036 mol). ACVA (0.0128 g, 0.046 mmol) was added to the reaction mixture followed by isopropyl alcohol (25.8 g) to make a 20 wt% solution. The flask was sealed and degassed for 30 minutes with nitrogen. The reaction mixture was stirred and heated at 70°C for 48 hours. After this time the solution was taken off and precipitated in methanol (900 ml). The white/pale yellow solid was collected via vacuum filtration and dried in a vacuum dessicator. The product was a white/pale yellow solid of poly(butyl methacrylate) (1.5859 g). (final BMA conversion = 94%). ¹H NMR (400 MHz, CDCl₃, 298K) δ(ppm) = 1.0 (t, 3H, -CH₃), 1.0 (s, 3H, -CH_­3), 1.5 (s, 2H, -CH₂), 1.7 (s, 2H, -CH₂), 1.9-2.1 (m, 2H, CH₂), 3.0 (water), 4.0 (s, 2H, CH₂).

4.4 Synthesis of Hyperbranched Poly (Butyl Methacrylate)

See table 8 for reagent quantities. The procedure describes the quantities, solvent, initiator, weight percent, time and conversion to synthesise hyperbranched polymer 2.

			Solvent	Initiator
Polymer	BMA / g	PETTC / g	Toluene / g	AIBN / g	EGDMA / g	Time / h
1	5	0.2388	12.54	0.0389	0.120	27
			IPA / g	ACVA / g
2	5	0.2398	21.74	0.0676	0.120	23
3	5	0.2395	21.72	0.0653	0.120	27

Table 8. The reagent quantities and conditions used to synthesise HB P(BMA).

4-cyano-4-(2-phenylethanesulfanylthiocarbonate)sulfanyl pentanoic acid (PETTC) (0.2398 g, 0.706 mmol) was added to a 50 cm³ round bottom flask along with butyl methacrylate (5.0 g, 0.035 mol) and ethylene glycol dimethacrylate (0.126 g, 0.634 mmol). ACVA (0.067 g, 0.239 mmol) was added to the reaction mixture followed by isopropyl alcohol (21.7 g) to make a 20 wt% solution. The flask was sealed and degassed for 30 minutes with nitrogen. The reaction mixture was stirred and heated at 75°C for 27 hours. After this time the solution was taken off and precipitated in a mix methanol (540 ml) and water (360 ml). The white/pale yellow solid was collected via vacuum filtration and dried in a vacuum desiccator. The product was a white/pale yellow solid of hyperbranched poly(butyl methacrylate) (2.7842 g). (final BMA conversion = 94%). ¹H NMR (400 MHz, CDCl₃, 298K) δ(ppm) = 1.0 (t, 3H, -CH₃), 1.0 (s, 3H, -CH_­3), 1.5 (s, 2H, -CH₂), 1.7 (s, 2H, -CH₂), 1.9-2.1 (m, 2H, CH₂), 3.0 (water), 4.0 (s, 2H, CH₂).

4.5 Synthesis of Hyperbranched Poly (Butyl Methacrylate-Stat-Methacrylic Acid)

See table 9 for reagent quantities all reactions done at 75 °C for 24 hours. The procedure describes the quantities, solvent, initiator, weight percent, time and conversion to synthesise statistical polymer 1.

Polymer	BMA / g	MAA / g	IPA / g	ACVA / g	PETTC / g	EGDMA / g
1	1.74	0.26	8.63	0.0235	0.0842	0.051
2	3.97	1.03	21.64	0.0624	0.2252	0.119
3	4.68	0.32	21.49	0.0568	0.2064	0.109
4	7.94	2.06	43.26	0.1272	0.4588	0.237

Table 9. The reagent quantities used to synthesise HB P(BMA)-stat-P(MAA).

4-cyano-4-(2-phenylethanesulfanylthiocarbonate)sulfanyl pentanoic acid (PETTC) (0.045 g, 0.229 mmol) was added to a 25 cm³ round bottom flask along with butyl methacrylate (1.737 g, 0.014 mol), methacrylic acid (0.263 g, 0.003 mol) and ethylene glycol dimethacrylate (0.045 g, 0.634 mmol). ACVA (0.0238 g, 0.085 mmol) was added to the reaction mixture followed by isopropyl alcohol (8.623 g) to make a 20 wt% solution. The flask was sealed and degassed for 30 minutes with nitrogen. The reaction mixture was stirred and heated at 75°C for 24 hours. After this time the solution was taken off and precipitated in a mix methanol (540 ml) and water (360 ml). The white/pale yellow solid was collected via vacuum filtration and dried in a vacuum desiccator. The product was a white/pale yellow solid of hyperbranched poly(butyl methacrylate-stat-methacrylic acid) (1.802 g). (final BMA conversion = 94.8%). ¹H NMR (400 MHz, CDCl₃, 298K) δ(ppm) = 1.0 (m, 3H, -CH₃), 1.0 (m, 3H, -CH_­3), 1.0 (m, 3H, CH₂), 1.5 (s, 2H, -CH₂), 1.7 (s, 2H, -CH₂), 1.9-2.1 (m, 2H, CH₂), 1.9-2.1 (m, 2H, CH₂), 4.0 (s, 2H, CH₂).

4.6 Synthesis of Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic Acid)

See table 10 for reagent quantities all reactions done at 75 °C for 24 hours. The procedure describes the quantities, solvent, initiator, weight percent, time and conversion to synthesise block polymer 1.

Polymer	HB BMA / g	MAA / g	ACVA / g	IPA / g
1	1.0	0.0465	0.0104	4.23
2	1.0	0.1189	0.0128	6.41
3	1.0	0.2376	0.0127	7.09
4	1.5	0.8914	0.0215	9.11

Table 10. The reagent quantities used to synthesise HB P(BMA)-block-P(MAA).

The previously synthesised hyperbranched poly(butyl methacrylate) (1 g) was added to 25 cm³ round bottom flask along with methacrylic acid (0.0465 g, 0.541 mmol). ACVA (0.0101 g, 0.036 mmol) was added to the reaction mixture followed by isopropyl alcohol (5.988 g) to make a 15 wt% solution. The flask was sealed and degassed for 30 minutes with nitrogen. The reaction mixture was stirred and heated at 75°C for 24 hours. After this time the solution was taken off and precipitated in a mix methanol (200 ml) and water (700 ml). ). The white powdery solid was collected via vacuum filtration and dried in a vacuum desiccator. The product was a white solid of hyperbranched poly(butyl methacrylate)-block-poly(methacrylic acid) (0.696 g). (final MAA conversion = 80%). ¹H NMR (400 MHz, CDCl₃, 298K) δ(ppm) = 1.0 (m, 3H, -CH₃), 1.0 (m, 3H, -CH_­3), 1.0 (m, 3H, CH₂), 1.5 (s, 2H, -CH₂), 1.7 (s, 2H, -CH₂), 1.9-2.1 (m, 2H, CH₂), 1.9-2.1 (m, 2H, CH₂), 4.0 (s, 2H, CH₂).

4.7 Hyperbranched Poly (Butyl Methacrylate) End Group Modification^[55]

The previously synthesised hyperbranched poly(butyl methacrylate) (1 g) was added to a 50 cm³ round bottom flask along with ACVA (0.909 g, 0.003 mol) followed by isopropyl alcohol (17.19 g) to make a 10 wt% solution. The flask was sealed and degassed for 30 minutes with nitrogen. The reaction mixture was stirred and heated at 75°C for 3 hours. After this time the solution was taken off and precipitated in a mix methanol (540 ml) and water (360 ml). ). The white powdery solid was collected via vacuum filtration and dried in a vacuum desiccator. The product was a white solid of hyperbranched poly(butyl methacrylate (0.691 g). ¹H NMR (400 MHz, CDCl₃, 298K) δ(ppm) = 1.0 (t, 3H, -CH₃), 1.0 (s, 3H, -CH_­3), 1.5 (s, 2H, -CH₂), 1.7 (s, 2H, -CH₂), 1.9-2.1 (m, 2H, CH₂), 3.0 (water), 4.0 (s, 2H, CH₂).

4.8 Solution Preparation

Triethanolamine (1.1 eq w.r.t. MAA content) was diluted in isopropyl alcohol. The solution was added to hyperbranched copolymer to form a 50 wt % solution in IPA in a sample vial. This solution was left until fully dissolved. Water was then added dropwise to the solution to form a 40 wt % solution and then it was left until it fully dissolved. This process was repeated to dilute the solution down to 40, 35, 30, 25, 20 and 10 wt %. Then 0.5 g of the 10 wt % solution was added to another sample tube and diluted with water to form a 1 wt % solution and a 0.5 wt % solution.

4.9 Alkylation

See table 11 for reagent quantities. The procedure describes the quantities to synthesise alkylated stat 3 polymer.

Polymer	Polymer / g	Ce2CO3 / g	BzBr / g	DMF / g
Stat
1	0.3	0.182	0.091	3.0
2	0.5	0.043	0.239	4.5
3	0.4	0.153	0.077	5.0
4	0.5	0.300	0.015	5.0
Block
1	0.1	0.023	0.010	0.9
2	0.1	0.056	0.030	1.0
3	0.2	0.176	0.090	1.8
4	0.3	0.084	0.040	2.7

Table 11. The reagent quantities used to alkylate the MAA units on the copolymers.

Statistical copolymer 3 (0.5 g) was added to a 50 ml round bottom flask with dimethylformamide (4.5 ml), benzyl bromide (0.239 g, 1.40 mmol) and caesium carbonate (0.426 g, 1.31 mmol). The reaction mixture was then left to stir under nitrogen for 24 hours. After 24 hours it was filtered and then the filtrate was evaporated under vacuum to leave behind the alkylated product (0.361 g).

4.10 Instrumentation

4.10.1 Gel Permeation Chromatography

Molecular weight distributions were analysed by gel permeation chromatography (GPC) using a DMF eluent containing 0.1 % LiBr with a Agilent 1260 infinity LC system. Two 5 µm (30 cm) mixed-C columns were used in order to separate molecular weights. The detector of the GPC system was a refractive index detector. The mobile phase flowed at a rate of 1.0 mL min^-1. All the sample were calibrated with a set of near monodisperse poly(methyl methacrylate) standards. The sample are filtered through a 45 µm PTFE filter to remove any undissolved material before injection.

4.10.2 Dynamic Light Scattering

Dynamic light scattering measurements were performed using a Zetasizer NanoZS instrument from Malvern Instruments UK, with a fixed scattering angle of 173°. Aqueous polymer dispersion was diluted down to 0.5 wt% with water and light scattering studies were performed at 25 °C. The intensity-average diameter, polydispersity and volume average diameter of the copolymers particles were calculated by cumulant analysis of the experimental correlation function using Dispersion Technology Software version 6.20. Three runs each of a duration of 120 seconds were averaged to give one data point. The measurements were recorded using disposable cuvettes.

4.10.3 ¹H NMR

¹H NMR spectra were recorded using solvents of deuterated acetone ((CD₃)₂CO) and deuterated chloroform (CDCl₃) using a Bruker AV1-400 or AV3HD-400 Mhz spectrometer. Typically, 64 scans were averaged per spectrum.

4.10.4 Rheology

An Anton Paar MCR 502 rheometer operating at 20 °C equipped with cone and plate geometry (50 mm diameter, angle 2°) and a solvent trap was used to collect all rheology data. Strain sweeps were performed between 0.1 and 20% strain, to find the linear viscoelastic region for the measurements. The complex viscosity was measured when performing an Angular frequency sweep between 500 and 0.1 rads^-1 at a constant strain of 10%. The solvent trap was filled with IPA/water in the same ratio as that of the sample to prevent evaporation of the IPA/water solvent in the copolymer solutions.

4.10.5 Zeta Potential

Zeta potential was performed using the same apparatus as the dynamic light scattering. The measurements were performed on 1 wt% solutions of water which was diluted down to 0.25 wt% by the addition of potassium chloride (1 mM). The measurements were performed at 25 °C in disposable capillary zeta cells.

4.10.6 Small Angle X-ray Scattering

SAXS patterns were collected using a modified Bruker AXS Nanostar instrument equipped with a 2D Hi-STAR multi-wire gas detector at a camera length of 1.46 m. The patterns were collected over a scattering vector, q range of 0.0008 nm^-1 to 0.020 nm^-1 using Cu Kα X-ray radiation (λ = 0.154 nm). The length of the scattering vector, q is given by where θ is half the scattering angle.

4.10.7 Solid State Infrared Spectroscopy

Solid state infrared spectroscopy measurements were recorded on a Thermo Scientific Nicolet Is10 ATR. Between a wavenumber of 400-4000 cm^-1. This was done with 124 scans.

5. References

Page numbers for some references

[1] C. Gao and D. Yan, Progress in Polymer Science 2004, 29, 183-275.

[2] Y. Zheng, S. Li, Z. Weng and C. Gao, Chemical Society Reviews 2015, 44, 4091-4130.

[3] C. Yates and W. Hayes, European Polymer Journal 2004, 40, 1257-1281.

[4] B. Pettersson, Pigment & Resin Technology 1996, 25, 4-14.

[5] R. Kienle and A. Hovey, Journal of the American Chemical Society 1929, 51, 509-519.

[6] P. FLORY, Journal of the American Chemical Society 1952, 74, 2718-2723.

[7] Y. KIM and O. WEBSTER, Abstracts of Papers of the American Chemical Society 1988, 196, 104-POLY; Y. KIM and O. WEBSTER, Journal of the American Chemical Society 1990, 112, 4592-4593.

[8] O. Flomenbom, R. Amir, D. Shabat and J. Klafter, Journal of Luminescence 2005, 111, 315-325.

[9] C. Schubert, C. Osterwinter, C. Tonhauser, M. Schomer, D. Wilms, H. Frey and C. Friedrich, Macromolecules 2016, 49, 8722-8737.

[10] N. O’Brien, A. McKee, D. Sherrington, A. Slark and A. Titterton, Polymer 2000, 41, 6027-6031.

[11] S. Liang, X. Li, W. Wang, B. Li and S. Zhu, Macromolecules 2016, 49, 752-759.

[12] A. Slark, D. Sherrington, A. Titterton and I. Martin, Journal of Materials Chemistry 2003, 13, 2711-2720; P. Costello, I. Martin, A. Slark, D. Sherrington and A. Titterton, Polymer 2002, 43, 245-254.

[13] F. Isaure, P. Cormack, S. Graham, D. Sherrington, S. Armes and V. Butun, Chemical Communications 2004, 1138-1139.

[14] T. Das, S. Sengupta, U. Ghorai, A. Dey and A. Bandyopadhyay, Rsc Advances 2015, 5, 102932-102941.

[15] R. England and S. Rimmer, Polymer Chemistry 2010, 1, 1533-1544.

[16] L. Jiang, W. Huang, X. Xue, H. Yang, B. Jiang, D. Zhang, J. Fang, J. Chen, Y. Yang, G. Zhai, L. Kong and S. Wang, Macromolecules 2012, 45, 4092-4100.

[17] M. Yan, Y. Huang, M. Lu, F. Y. Lin, N. B. Hernández and E. W. Cochran, Biomacromolecules 2016, 17, 2701-2709.

[18] C. E. Carraher, Introduction to polymer chemistry, Taylor & Francis, Boca Raton, 2007, p.

[19] F. Ehlers, J. Barth, P. Vana, B. Tang, N. Tsarevsky and B. Sumerlin, Fundamentals of Controlled/living Radical Polymerization 2013, 1-59.

[20] G. Moad and D. H. Solomon, The chemistry of radical polymerization, Elsevier, Amsterdam, 2006, p. xxvi, 639p.

[21] R. Storey, B. Tang, N. Tsarevsky and B. Sumerlin, Fundamentals of Controlled/living Radical Polymerization 2013, 60-77.

[22] G. Moad, E. Rizzardo and S. Thang, Accounts of Chemical Research 2008, 41, 1133-1142.

[23] J. Chiefari, Y. Chong, F. Ercole, J. Krstina, J. Jeffery, T. Le, R. Mayadunne, G. Meijs, C. Moad, G. Moad, E. Rizzardo and S. Thang, Macromolecules 1998, 31, 5559-5562.

[24] G. Moad, E. Rizzardo, S. Thang, B. Tang, N. Tsarevsky and B. Sumerlin, Fundamentals of Controlled/living Radical Polymerization 2013, 205-249.

[25] D. Keddie, G. Moad, E. Rizzardo and S. Thang, Macromolecules 2012, 45, 5321-5342.

[26] G. Moad, Polymer International 2015, 64, 15-24.

[27] I. W. Hamley, Introduction to soft matter : polymers, colloids, amphiphiles and liquid crystals, Wiley, Chichester, 2000, p. 300p.

[28] I. W. Hamley, Introduction to soft matter – revised edition : synthetic and biological self-assembling materials, Wiley ;

John Wiley [distributor], Hoboken, N.J.

Chichester, 2007, p. 328 p.

[29] Y. Morishima, S. Nomura, T. Ikeda, M. Seki and M. Kamachi, Macromolecules 1995, 28, 2874-2881; Y. Hu, G. Smith, M. Richardson and C. McCormick, Macromolecules 1997, 30, 3526-3537.

[30] F. Hatton, P. Chambon, T. McDonald, A. Owen and S. Rannard, Chemical Science 2014, 5, 1844-1853.

[31] H. Tai, C. Duvall, A. Hoffman, P. Stayton and W. Wang, Macromolecular Materials and Engineering 2012, 297, 1175-1183.

[32] K. Davis and K. Matyjaszewski, Statistical, Gradient, Block and Graft Copolymers By Controlled/living Radical Polymerizations 2002, 159, 1-169.

[33] T. Kawata, A. Hashidzume and T. Sato, Macromolecules 2007, 40, 1174-1180.

[34] A. Hashidzume, A. Kawaguchi, A. Tagawa, K. Hyoda and T. Sato, Macromolecules 2006, 39, 1135-1143.

[35] T. Mori, A. Hashidzume and T. Sato, Polymer Journal 2009, 41, 189-194.

[36] E. B. Bradford and L. D. McKeever, Progress in Polymer Science 1971, 3, 129-143.

[37] J. Lutz, Polymer Chemistry 2010, 1, 55-62.

[38] M. Hillmyer, Current Opinion in Solid State & Materials Science 1999, 4, 559-564.

[39] T. Cosgrove, V. Rodin, M. Murray and R. Buscall, Journal of Polymer Research 2007, 14, 167-174.

[40] J. Liu, C. Uhlir, P. Shah, F. Sun and J. Stansbury, Rsc Advances 2016, 6, 64791-64798.

[41] P. Alexandridis, Current Opinion in Colloid & Interface Science 1996, 1, 490-501.

[42] D. Aaserud, L. Prokai and W. Simonsick, Analytical Chemistry 1999, 71, 4793-4799.

[43] L. Hagel, A. Hartmann and K. Lund, J Chromatogr 1993, 641, 63-70.

[44] B. Pirok, P. Breuer, S. Hoppe, M. Chitty, E. Welch, T. Farkas, S. van der Wal, R. Peters and P. Schoenmakers, Journal of Chromatography a 2017, 1486, 96-102.

[45] R. Borsali and R. Pecora, Soft matter characterization, Springer, New York, 2008, p. 2 v. (xxxvi, 1453 p.).

[46] K. CLAYS and A. PERSOONS, Physical Review Letters 1991, 66, 2980-2983.

[47] D. CROMER, Journal of Chemical Physics 1969, 50, 4857-&; P. EISENBERGER and P. PLATZMAN, Physical Review a-General Physics 1970, 2, 415-+.

[48] G. Beaucage, Journal of Applied Crystallography 1996, 29, 134-146.

[49] G. Beaucage, Journal of Applied Crystallography 1995, 28, 717-728.

[50] G. BEAUCAGE and D. SCHAEFER, Journal of Non-Crystalline Solids 1994, 172, 797-805.

[51] D. De Kee and K. Wissbrun, Physics Today 1998, 51, 24-29.

[52] M. Abd Elrehim, B. Voit, B. Bruchmann, K. Eichhorn, K. Grundke and C. Bellmann, Journal of Polymer Science Part a-Polymer Chemistry 2005, 43, 3376-3393.

[53] A. Delgado, E. Gonzalez-Caballero, R. Hunter, L. Koopal and J. Lyklema, Pure and Applied Chemistry 2005, 77, 1753-1805.

[54] M. Hsu, E. Dufresne and D. Weitz, Langmuir 2005, 21, 4881-4887.

[55] S. Perrier, P. Takolpuckdee and C. Mars, Macromolecules 2005, 38, 2033-2036.

6. Appendix

6.1 ¹H NMR Spectra

6.1.1 PETTC

6.1.2 Linear Poly (Butyl Methacrylate)

6.1.3 Hyperbranched Poly (Butyl Methacrylate)

6.1.4 Hyperbranched Poly (Butyl Methacrylate-stat-Methacrylic acid)

6.1.5 Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic acid)

6.1.6 End Group Modified Hyperbranched Poly (Butyl Methacrylate) before and after

6.2 Solid State Infrared Spectroscopy

6.2.1 Comparison of original and Alkylated Hyperbranched Poly (Butyl Methacrylate-stat-Methacrylic acid)

6.2.2 Comparison of original and Alkylated Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic acid)

6.3 Gel Permeation Chromatography

6.3.1 Hyperbranched Poly (Butyl Methacrylate-stat-Methacrylic acid)

6.3.2 Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic acid)

6.1.1 PETTC

6.1.2 Linear Poly (Butyl Methacrylate)

6.1.3 Hyperbranched Poly (Butyl Methacrylate)

6.1.4 Hyperbranched Poly (Butyl Methacrylate-stat-Methacrylic acid)

6.1.5 Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic acid)

6.1.6 End Group Modified Hyperbranched Poly (Butyl Methacrylate) before and after

6.2.1 Comparison of original and Alkylated Hyperbranched Poly (Butyl Methacrylate-stat-Methacrylic acid)

6.2.2 Comparison of original and Alkylated Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic acid)

6.3.1 Hyperbranched Poly (Butyl Methacrylate-stat-Methacrylic acid)

6.3.2 Hyperbranched Poly (Butyl Methacrylate)-Block-Poly (Methacrylic acid)