We study the long time behavior of a system of two or three Brownian hard balls living in the Euclidean space of dimension at least two, submitted to a mutual attraction and to elastic collisions.

Starch is a water-insoluble glucose biopolymer used as an energy cache in plants and is synthesized and degraded in a diurnal cycle. Reversible phosphorylation of starch granules regulates the solubility and, consequentially, the bioavailability of starch glucans to degradative enzymes. Glucan phosphatases release phosphate from starch glucans and their activity is essential to the proper diurnal metabolism of starch. Previously, the structural basis of glucan phosphatase activity was entirely unknown. The work in this dissertation outlines the structural mechanism of activity of two plant glucan phosphatases called Starch EXcess4 (SEX4) and Like Sex Four2 (LSF2). The crystal structures of SEX4 and LSF2 were determined with and without phosphoglucan ligands bound, revealing the basis of their interaction with an endogenous substrate. The data show that SEX4 and LSF2 interact with starch glucans via distinctive mechanisms. SEX4 binds glucan chains via an aromatic-rich pocket spanning its Carbohydrate Binding Module (CBM) and catalytic Dual Specificity Phosphatase (DSP) domains. Conversely, LSF2 lacks a CBM and, instead, binds glucans at two non-catalytic surface-binding sites that are located distally from its active site. In addition, it was previously reported that SEX4 and LSF2 act upon distinct phospho-glucan substrates: SEX4 preferentially dephosphorylates the C6-position of starch glucans and LSF2 exclusively dephosphorylates the C3- position. The data herein reveal that SEX4 and LSF2 contain differences in their active site topology that serve to position the glucan chain in opposite orientations, therefore accounting for the differences in substrate specificity. Using these insights, SEX4 was engineered with reversed substrate specificity, i.e. preferential C3-specific activity. Previous work has established the interaction between phosphatases and protein, lipid, and nucleic acids; however, the current study represents the first insights into phosphatase interaction with carbohydrate substrates. In addition, the insights gained provide a model that will be used in future studies with the mammalian glucan phosphatase laforin, which is linked to neurodegeneration and the fatal epileptic disorder Laforaʼs Disease.

This thesis is based on promising computing paradigm of reversible logic which generates unique outputs out of the inputs and. Reversible logic circuits maintain one-to-one mapping inside of the inputs and the outputs. Compared to the traditional irreversible computation, reversible logic circuit has the advantage that it successfully avoids the information loss during computations. Also, reversible logic is useful to design ultra-low-power nanocomputing circuits, circuits for quantum computing, and the nanocircuits that are testable in nature. Reversible computing circuits require the ancilla inputs and the garbage outputs. Ancilla input is the constant input in reversible circuits. Garbage output is the output for maintaining the reversibility of the reversible logic but is not any of the primary inputs nor a useful bit. An efficient reversible circuit will have the minimal number of garbage and ancilla bits. Barrel shifter is one of main computing systems having applications in high speed digital signal processing, oating-point arithmetic, FPGA, and Center Processing Unit (CPU). It can operate the function of shifting or rotation for multiple bits in only one clock cycle. The goal of this thesis is to design barrel shifters based on the reversible computing that are optimized in terms of the number of ancilla and garbage bits. In order to achieve this goal, a new Super Conservative Reversible Logic Gate (SCRL gate) has been used. The SCRL gate has 1 control input depending on the value of which it can swap any two n-1 data inputs. We proved that the SCRL gate is superior to the existing conservative reversible Fredkin gate. This thesis develops 5 design methodologies for reversible barrel shifters using SCRL gates that are primarily optimized with the criteria of the number of ancilla and garbage bits. The five proposed methodologies consist of reversible right rotator, reversible logical right shifter, reversible arithmetic right shifter, reversible universal right shifter and reversible universal bidirectional shifter. The proposed reversible barrel shifter design is compared with the existing works in literature and have shown improvement ranging from 8.5% to 92% by the number of garbage and ancilla bits. The SCRL gate and design methodologies of reversible barrel shifter are mapped in Quantum Dot Cellular Automata (QCA) computing. It is illustrated that the SCRL-based designs of reversible barrel shifters have less QCA cost (cost in terms of number of inverters and majority voters) compared to the Fredkin gate- based designs of reversible barrel shifters.

Green chemistry principles served as a guide for three industrially-relevant projects. In the first project, silylamines were applied as reversible ionic liquids for carbon dioxide capture from post-combustion flue gas streams. The effect of silylamine structure was thoroughly researched to develop a comprehensive library of silylamines and an accompanying set of structure-property relationships. The proposed solvent systems have the potential to present significant energy savings, as design has focused on their use in a non-aqueous, solvent-free environment. The second project also dealt extensively with carbon dioxide capture, as a reversible, in-situ protecting group for amines. Three strategies for the reversible protection of amines using carbon dioxide were developed and evaluated. Further, a chemoselective reaction was performed using carbon dioxide to protect a reactive amine and consequentially direct reactivity elsewhere within the same molecule. The carbon dioxide-protection technology developed has significant impact in multi-step industrial syntheses, as reversible, in-situ protection with carbon dioxide could eliminate the need for separate protection and deprotection unit operations. Lastly, a study was performed on the thermal degradation and stabilization of PVC in the presence of both plasticizers and thermal stabilizers. The study combined both model compound experiments as well as work with bulk PVC blends to gain a holistic understanding of the processes that take place during the degradation and stabilization of PVC. A bio-based plasticizer was investigated as a replacement for petroleum-based phthalate plasticizers. Additionally, two novel thermal stabilizers for PVC were presented and evaluated.

The shortcomings of current antidepressant agents prompts the design of novel multimodal antidepressants and the identification of new antidepressant targets, especially those located at sub-cellular level. Such antidepressants should possess improved response rates as well as safety profiles. Methylene blue (MB) is reported to possess diverse pharmacological actions and is attracting increasing attention for the treatment of a variety of disorders including Alzheimer’s disease, bipolar disorder, anxiety and depression. MB acts on both monoamine oxidase (MAO) and the nitric oxide (NO)-cGMP pathway, and possesses antidepressant activity in rodents. The principal goal of this study was to design a close structural analogue of MB and to evaluate the effects of these structural changes on MAO inhibition, a well-known antidepressant target. Furthermore, MAO inhibition is also responsible for cardiovascular toxicity in clinically used MAOI inhibitors. For this purpose we investigated the antidepressant properties of the synthetic MB analogue (ethyl-thioniniumchloride; ETC) as well as azure B, the major metabolite of MB, in the forced swim test (FST). ETC was synthesized with a high degree of purity from diethyl-p-phenylenediamine with 6% yield. ETC was firstly evaluated as a potential inhibitor of recombinant human MAO-A and MAO-B. Azure B and ETC were evaluated over a dosage range of 4-30 mg/kg for antidepressant-like activity in the acute FST in rats, and the results were compared to those obtained with saline, imipramine (15 mg/kg) and MB (15 mg/kg) treated rats. Locomotor activity was evaluated to ensure that changes in swim motivation are based on antidepressant response and not due to an indirect effect of the drug on locomotor activity. The results document that ETC inhibits MAO-A and MAO-B with IC50 values of 0.51 μM and 0.592 μM, respectively. Furthermore, ETC inhibits MAO-A and MAO-B reversibly, while the mode of inhibition is most likely competitive. In the acute FST, azure B and ETC were more effective than imipramine and MB in reversing immobility, without inducing locomotor effects. Azure B and ETC increased swimming behaviour during acute treatment, which is indicative of enhanced serotonergic neurotransmission. Azure B and ETC did not affect noradrenergicmediated climbing behaviour. These results suggest that azure B may be a contributor to the antidepressant effect of MB, and acts via increasing serotonergic transmission. Secondly, small structural changes made to MB do not abolish its antidepressant effect even though ETC is a less potent MAO-A inhibitor than MB. / MSc (Pharmaceutical Chemistry), North-West University, Potchefstroom Campus, 2014

Monoamine oxidase (MAO) is responsible for dopamine catabolism in the brain and therefore is especially important in the treatment of Parkinson's disease (PD). MAO–B inhibition provides symptomatic relief by indirectly elevating dopamine levels in the PD brain. PD is caused by the loss of dopaminergic neurons in the substantia nigra and the formation of proteinaceous structures in the brain. The cause of idiopathic PD is unknown, but one theory states that reactive oxygen species (ROS), partly derived from the catalytic cycle of MAO, may be to blame for damaging dopaminergic neurons. Since MAO inhibitors may reduce the MAO–catalyzed production of ROS, these compounds may protect dopaminergic neurons against degeneration in PD. It is commonly accepted that by the time PD symptoms manifest, about 80% of striatal dopamine has been lost. MAO is present as two subtypes in the human brain, namely MAO–A and MAO–B. MAOs are found mainly attached to the mitochondrial membrane and is responsible for the oxidative deamination of various monoamines, including dopamine. MAO is a dimeric enzyme which operates in conjunction with a co–factor, flavin adenine dinucleotide (FAD), to which it is covalently bound. The flavin is in a bent conformation, which assists the catalytic activity of MAO. As mentioned above, the catalytic action of MAO also produces harmful substances such as hydrogen peroxide, ammonia, aldehydes and may also increase the levels of hydroxyl radicals. In the healthy brain, these substances are metabolized rapidly, but the PD brain may exhibit reduced clearance of these species. Thus the inhibition of MAOs may be beneficial to the PD sufferer as it indirectly increases dopamine levels in the brain and may also slow the formation of harmful substances. MAO inhibitors, of the MAO–A type, were first used as anti–depressants. It was these drugs that first prompted researchers to explore MAO inhibitors as novel anti–parkinsonian drugs, as MAO–A inhibition slows the degradation of dopamine. Two types of inhibition modes exist, irreversible and reversible inhibition. Irreversible inhibitors do not allow for competition with the substrate and inactivate the enzyme permanently. Selegiline, a propargyl amine derivative, is an example of an irreversible MAO–B selective inhibitor. The major disadvantage of irreversible inhibitors is that after terminating treatment, recovery of the enzyme activity may require several weeks, since the turnover rate for the biosynthesis of MAO in the human brain may be as much as 40 days. Reversible inhibitors have better safety profiles since they allow for competition with the substrate. (E)–8–(3–Chlorostyryl)caffeine (CSC) is an example of a reversible inhibitor of MAO–B and is also an antagonist of the adenosine A2A receptor. Since antagonism of A2A receptors also produces an antiparkinsonian effect, dual acting compounds such as CSC, which block both the A2A receptors and MAO–B, may have an enhanced therapeutic potential in PD therapy. Current PD therapy available only treats the symptoms of PD and do not halt or slow the progression of the neurodegenerative processes. There therefore exists the need for the development of antiparkinsonian drugs with neuroprotective effects. Since both MAO–B inhibitors and A2A receptor antagonists are reported to possess protective effects in PD and PD animal models, dual acting drugs, that antagonize A2A receptors and inhibit MAO–B, may be candidates for neuroprotection. Using the structure of CSC as lead, we investigate in the current study, the possibility that aminocaffeines may also possess potent MAO–B inhibitory properties. The structures of the aminocaffeine derivatives that were investigated bear close structural resemblance to CSC as well as to a series of alkyloxycaffeine analogues that was recently found to be potent MAO inhibitors. This study therefore further explores the structural requirements of caffeine derivatives to act as MAO inhibitors by examining the possibility that aminocaffeine derivatives may be MAO inhibitors. Such compounds may act as lead compounds for the development of improved PD therapy. In this study, a series of 8–aminocaffeine derivatives were synthesized and evaluated as inhibitors of human MAO–A and B. For this purpose, 8–chlorocaffeine was reacted with the appropriate amine at high temperatures to produce the desired 8–aminocaffeine derivatives. The inhibitory activities of the compounds were determined towards recombinant human MAO–A and B and expressed as IC50 values. The results showed that human MAO–B was most potently inhibited by 8–[methyl(4–phenylbutyl)amino]caffeine with an IC50 value of 2.97 ?M. Human MAO–A was most potently inhibited by 8–[2–(3–chlorophenyl)–ethylamino]caffeine with an IC50 value of 5.78 ?M. It was found that methylation of the amine group at C8 of the caffeine ring increases inhibition but also selectivity towards MAO–B inhibition. For example, 8–[4–(phenylbutylamino)]caffeine inhibits MAO–B with an IC50 value of 7.56 ?M whereas 8–[methyl(4–phenylbutyl)amino]–caffeine has an increased inhibition potency of 2.97 ?M. The selectivity for MAO–B inhibition also increases over MAO–A when the C8 amine is methylated. It was found that the aminocaffeine derivatives bind reversibly to both enzyme isoforms and the mode of inhibition is competitive for MAO–B. From these results it can be concluded that although the 8–aminocaffeine derivatives are only moderately potent MAO–B inhibitors, they may act as lead compounds for the design of more potent reversible MAO inhibitors. Docking studies revealed that the 8–aminocaffeine and 8–[(methyl)amino]caffeine derivatives traverse both the entrance and substrate cavities of the MAO–B enzyme, with the caffeinyl moiety oriented towards the FAD co–factor while the amino–side chain protrudes into the entrance cavity. / Thesis (M.Sc. (Pharmaceutical Chemistry))--North-West University, Potchefstroom Campus, 2012.

Le développement croissant de systèmes de stockages connectés au réseau électrique se voit stimulé par les problématiques de gestion de réseau liés entre autres à l'intégration croissante de technologies de production renouvelables. L'accumulation hydropneumatique d'énergie apparait comme une solution propre et économiquement intéressante dans le panel des technologies existantes. Cette étude analyse une configuration d'accumulation air-eau sans séparation, en cycle fermé et utilisant une pompe/turbine rotodynamique. Les principales problématiques en relation à ces choix techniques sont : les phénomènes de transfert de masse et de chaleur à l'interface air-eau, les caractéristiques de fonctionnement d'une machine hydraulique rotodynamique en régime d'opération variable et le rendement d'accumulation du système. Ces aspects sont traités ici par des approches locales dans certains cas, puis globales afin de pouvoir proposer un modèle de comportement dynamique du système de stockage pouvant faire l'objet de simulations rapides. Les aspects de modélisation ont été élaborés en parallèle au développement d'un banc d'essais de 45 kW réalisé dans le cadre de ce travail. Les observations expérimentales obtenues sur ce banc sont comparées aux résultats du modèle qui rend bien compte de la dynamique et de l'état énergétique du système. Des études d'amélioration du système sont engagées à la fin du document. Elles concernent d'une part la question de la flexibilité de fourniture de puissance et d'autre part le rendement d'accumulation du réservoir de stockage.

Monoamine oxidase (MAO) is responsible for dopamine catabolism in the brain and therefore is especially important in the treatment of Parkinson's disease (PD). MAO–B inhibition provides symptomatic relief by indirectly elevating dopamine levels in the PD brain. PD is caused by the loss of dopaminergic neurons in the substantia nigra and the formation of proteinaceous structures in the brain. The cause of idiopathic PD is unknown, but one theory states that reactive oxygen species (ROS), partly derived from the catalytic cycle of MAO, may be to blame for damaging dopaminergic neurons. Since MAO inhibitors may reduce the MAO–catalyzed production of ROS, these compounds may protect dopaminergic neurons against degeneration in PD. It is commonly accepted that by the time PD symptoms manifest, about 80% of striatal dopamine has been lost. MAO is present as two subtypes in the human brain, namely MAO–A and MAO–B. MAOs are found mainly attached to the mitochondrial membrane and is responsible for the oxidative deamination of various monoamines, including dopamine. MAO is a dimeric enzyme which operates in conjunction with a co–factor, flavin adenine dinucleotide (FAD), to which it is covalently bound. The flavin is in a bent conformation, which assists the catalytic activity of MAO. As mentioned above, the catalytic action of MAO also produces harmful substances such as hydrogen peroxide, ammonia, aldehydes and may also increase the levels of hydroxyl radicals. In the healthy brain, these substances are metabolized rapidly, but the PD brain may exhibit reduced clearance of these species. Thus the inhibition of MAOs may be beneficial to the PD sufferer as it indirectly increases dopamine levels in the brain and may also slow the formation of harmful substances. MAO inhibitors, of the MAO–A type, were first used as anti–depressants. It was these drugs that first prompted researchers to explore MAO inhibitors as novel anti–parkinsonian drugs, as MAO–A inhibition slows the degradation of dopamine. Two types of inhibition modes exist, irreversible and reversible inhibition. Irreversible inhibitors do not allow for competition with the substrate and inactivate the enzyme permanently. Selegiline, a propargyl amine derivative, is an example of an irreversible MAO–B selective inhibitor. The major disadvantage of irreversible inhibitors is that after terminating treatment, recovery of the enzyme activity may require several weeks, since the turnover rate for the biosynthesis of MAO in the human brain may be as much as 40 days. Reversible inhibitors have better safety profiles since they allow for competition with the substrate. (E)–8–(3–Chlorostyryl)caffeine (CSC) is an example of a reversible inhibitor of MAO–B and is also an antagonist of the adenosine A2A receptor. Since antagonism of A2A receptors also produces an antiparkinsonian effect, dual acting compounds such as CSC, which block both the A2A receptors and MAO–B, may have an enhanced therapeutic potential in PD therapy. Current PD therapy available only treats the symptoms of PD and do not halt or slow the progression of the neurodegenerative processes. There therefore exists the need for the development of antiparkinsonian drugs with neuroprotective effects. Since both MAO–B inhibitors and A2A receptor antagonists are reported to possess protective effects in PD and PD animal models, dual acting drugs, that antagonize A2A receptors and inhibit MAO–B, may be candidates for neuroprotection. Using the structure of CSC as lead, we investigate in the current study, the possibility that aminocaffeines may also possess potent MAO–B inhibitory properties. The structures of the aminocaffeine derivatives that were investigated bear close structural resemblance to CSC as well as to a series of alkyloxycaffeine analogues that was recently found to be potent MAO inhibitors. This study therefore further explores the structural requirements of caffeine derivatives to act as MAO inhibitors by examining the possibility that aminocaffeine derivatives may be MAO inhibitors. Such compounds may act as lead compounds for the development of improved PD therapy. In this study, a series of 8–aminocaffeine derivatives were synthesized and evaluated as inhibitors of human MAO–A and B. For this purpose, 8–chlorocaffeine was reacted with the appropriate amine at high temperatures to produce the desired 8–aminocaffeine derivatives. The inhibitory activities of the compounds were determined towards recombinant human MAO–A and B and expressed as IC50 values. The results showed that human MAO–B was most potently inhibited by 8–[methyl(4–phenylbutyl)amino]caffeine with an IC50 value of 2.97 ?M. Human MAO–A was most potently inhibited by 8–[2–(3–chlorophenyl)–ethylamino]caffeine with an IC50 value of 5.78 ?M. It was found that methylation of the amine group at C8 of the caffeine ring increases inhibition but also selectivity towards MAO–B inhibition. For example, 8–[4–(phenylbutylamino)]caffeine inhibits MAO–B with an IC50 value of 7.56 ?M whereas 8–[methyl(4–phenylbutyl)amino]–caffeine has an increased inhibition potency of 2.97 ?M. The selectivity for MAO–B inhibition also increases over MAO–A when the C8 amine is methylated. It was found that the aminocaffeine derivatives bind reversibly to both enzyme isoforms and the mode of inhibition is competitive for MAO–B. From these results it can be concluded that although the 8–aminocaffeine derivatives are only moderately potent MAO–B inhibitors, they may act as lead compounds for the design of more potent reversible MAO inhibitors. Docking studies revealed that the 8–aminocaffeine and 8–[(methyl)amino]caffeine derivatives traverse both the entrance and substrate cavities of the MAO–B enzyme, with the caffeinyl moiety oriented towards the FAD co–factor while the amino–side chain protrudes into the entrance cavity. / Thesis (M.Sc. (Pharmaceutical Chemistry))--North-West University, Potchefstroom Campus, 2012.

Free-radical polymerisation (FRP) has been one of the most important techniques for producing materials used in a very wide variety of applications and has enhanced the lives of millions of people around the world. However, for many years a number of fundamental questions regarding the key kinetic processes involved in FRP have remained unresolved. In particular, an accurate description of the mechanism for diffusion-controlled bimolecular termination has proven elusive. As a result, conventional modelling tools for FRP have often proven unreliable. The aim of this thesis, therefore, was to accurately study the evolution of the bimolecular termination rate coefficient during free radical polymerisation using a new and more accurate methodology based on ‘controlled/living’ reversible addition-fragmentation chain transfer (RAFT) polymerisation. This was undertaken in order to develop a more precise understanding of bimolecular termination and thereby develop a more reliable modeling approach capable of predicting the rates of reaction and evolution of molecular weight distributions for a wide range of experimental conditions and a wide range of functional monomers. The RAFT-CLD-T (RAFT Chain-Length-Dependent Termination) Method was used to determine accurate values for the conversion and chain-length-dependent termination rate coefficient, kti,i(x), as a function of various parameters. These parameters included the chain size, i, polymer concentration (or conversion, x), chain length size distribution and chain architecture/structure. The accuracy of the RAFT-CLD-T Method was crucial to this work, therefore, an important part of this thesis was devoted to evaluating the reliability of this technique. Below 5 % conversion and above 80 % conversion the method was found to be unreliable due to the effects of chain-length-dependent propagation, high PDI’s and short-long termination. However, between 5 % and 80 % conversion it was found that the method is extremely robust and a series of easy-to-use experimental guidelines were determined for accurately applying the RAFT-CLD-T Method. The effects of chain size, chain size distribution, solution polymer concentration, and matrix architecture were examined for the RAFT-mediated polymerisations of methyl methacrylate (MMA), styrene (STY) and methyl acrylate (MA). It was found that four distinct scaling regimes of termination are observed: (1) a ‘short’ chain dilute solution regime, (2) a ‘long’ chain dilute solution regime, (3) a semi-dilute solution regime and (4) a concentrated solution regime. In dilute polymer solutions, chain-length-dependent power law exponents, ’s, determined during the polymerisation of MMA, STY and MA (where kti,i(x)  i-) indicated that termination follows two major scaling regimes with exponents of approximately ~0.5 to 0.6 for ‘short’ chains and and ~0.12 to 0.16 for ‘long’ chains. Importantly, these exponents are in excellent agreement with theoretical predictions for translational and segmental diffusion-controlled termination, respectively. At increasing polymer concentrations, kti,i(x) falls rapidly coinciding with the onset of the gel effect. By comparing results from the RAFT-mediated polymerisations of MMA, STY, MA, and vinyl acetate (VAc) with theoretical models, we found that the onset of the gel effect coincided closely with the theoretical onset of chain overlap. Considerable uncertainty has plagued the evaluation of this phenomenon, but using a difunctional RAFT agent we showed this uncertainty arises from the influence of broad MWD’s on chain overlap and short-long termination. Finally, critical tests of this theory involving the bimolecular termination of linear radicals in solutions of star polymer confirmed that the gel effect coincided with chain overlap. Beyond the gel effect termination slows enormously, passing through the ‘semi-dilute solution’ regime and into the ‘concentrated solution’. In semi-dilute solution, theoretical predictions based on scaling theory (i.e. the ‘blob’ model) were in excellent agreement with results for the polymerisation of PSTY in linear and star polymer solutions, indicating that the solvent quality diminished both with increasing chain length and through the addition of a star polymer matrix. In concentrated solutions, the chain-length-dependent power law exponent increased linearly with conversion. For example, for MMA the chain length dependence of kt in the gel regime scaled as gel = 1.8x + 0.056, suggesting that reptation alone does not describe termination in the concentrated solution. Values of gel for PSTY, MA, and VAc were in similar agreement, indicating that a mechanism intermediate between unentangled and entangled semi-dilute scaling laws applies in the concentrated solution regime. Interestingly, gel values for these monomers were found to decrease with increasing chain flexibility in the order gel(MMA)> gel(STY)> gel(VAc)> gel(MA), suggesting matrix mobility is rate determining in concentrated solutions. Similarly, gel values were also larger in star polymer solutions, coinciding with decreasing matrix mobility. Thus, although it has been commonly believed that polymer chains diffuse via reptation above the gel effect, these results show that this only occurs for solutions containing rigid and/or highly immobile macromolecules and in very high concentrations. To describe these behaviours, a semi-empirical ‘composite kt model’ was also developed to describe kti,i(x) as a function of i and x up to high conversions. We showed that the model is very simple to implement and accurate for modelling a wide range of functional monomers and experimental conditions. In particular, we showed the method was accurate for modelling RAFT-mediated polymerisations of a very wide range of monomers (MA, MMA, and PSTY) and was even accurate for modelling conventional FRP’s. Thus, the model provides a simple, flexible and accurate method for predicting the rate of reaction and evolution of molecular weight distributions across a range of experimental conditions based on accurate kti,i(x) values.

Free-radical polymerisation (FRP) has been one of the most important techniques for producing materials used in a very wide variety of applications and has enhanced the lives of millions of people around the world. However, for many years a number of fundamental questions regarding the key kinetic processes involved in FRP have remained unresolved. In particular, an accurate description of the mechanism for diffusion-controlled bimolecular termination has proven elusive. As a result, conventional modelling tools for FRP have often proven unreliable. The aim of this thesis, therefore, was to accurately study the evolution of the bimolecular termination rate coefficient during free radical polymerisation using a new and more accurate methodology based on ‘controlled/living’ reversible addition-fragmentation chain transfer (RAFT) polymerisation. This was undertaken in order to develop a more precise understanding of bimolecular termination and thereby develop a more reliable modeling approach capable of predicting the rates of reaction and evolution of molecular weight distributions for a wide range of experimental conditions and a wide range of functional monomers. The RAFT-CLD-T (RAFT Chain-Length-Dependent Termination) Method was used to determine accurate values for the conversion and chain-length-dependent termination rate coefficient, kti,i(x), as a function of various parameters. These parameters included the chain size, i, polymer concentration (or conversion, x), chain length size distribution and chain architecture/structure. The accuracy of the RAFT-CLD-T Method was crucial to this work, therefore, an important part of this thesis was devoted to evaluating the reliability of this technique. Below 5 % conversion and above 80 % conversion the method was found to be unreliable due to the effects of chain-length-dependent propagation, high PDI’s and short-long termination. However, between 5 % and 80 % conversion it was found that the method is extremely robust and a series of easy-to-use experimental guidelines were determined for accurately applying the RAFT-CLD-T Method. The effects of chain size, chain size distribution, solution polymer concentration, and matrix architecture were examined for the RAFT-mediated polymerisations of methyl methacrylate (MMA), styrene (STY) and methyl acrylate (MA). It was found that four distinct scaling regimes of termination are observed: (1) a ‘short’ chain dilute solution regime, (2) a ‘long’ chain dilute solution regime, (3) a semi-dilute solution regime and (4) a concentrated solution regime. In dilute polymer solutions, chain-length-dependent power law exponents, ’s, determined during the polymerisation of MMA, STY and MA (where kti,i(x)  i-) indicated that termination follows two major scaling regimes with exponents of approximately ~0.5 to 0.6 for ‘short’ chains and and ~0.12 to 0.16 for ‘long’ chains. Importantly, these exponents are in excellent agreement with theoretical predictions for translational and segmental diffusion-controlled termination, respectively. At increasing polymer concentrations, kti,i(x) falls rapidly coinciding with the onset of the gel effect. By comparing results from the RAFT-mediated polymerisations of MMA, STY, MA, and vinyl acetate (VAc) with theoretical models, we found that the onset of the gel effect coincided closely with the theoretical onset of chain overlap. Considerable uncertainty has plagued the evaluation of this phenomenon, but using a difunctional RAFT agent we showed this uncertainty arises from the influence of broad MWD’s on chain overlap and short-long termination. Finally, critical tests of this theory involving the bimolecular termination of linear radicals in solutions of star polymer confirmed that the gel effect coincided with chain overlap. Beyond the gel effect termination slows enormously, passing through the ‘semi-dilute solution’ regime and into the ‘concentrated solution’. In semi-dilute solution, theoretical predictions based on scaling theory (i.e. the ‘blob’ model) were in excellent agreement with results for the polymerisation of PSTY in linear and star polymer solutions, indicating that the solvent quality diminished both with increasing chain length and through the addition of a star polymer matrix. In concentrated solutions, the chain-length-dependent power law exponent increased linearly with conversion. For example, for MMA the chain length dependence of kt in the gel regime scaled as gel = 1.8x + 0.056, suggesting that reptation alone does not describe termination in the concentrated solution. Values of gel for PSTY, MA, and VAc were in similar agreement, indicating that a mechanism intermediate between unentangled and entangled semi-dilute scaling laws applies in the concentrated solution regime. Interestingly, gel values for these monomers were found to decrease with increasing chain flexibility in the order gel(MMA)> gel(STY)> gel(VAc)> gel(MA), suggesting matrix mobility is rate determining in concentrated solutions. Similarly, gel values were also larger in star polymer solutions, coinciding with decreasing matrix mobility. Thus, although it has been commonly believed that polymer chains diffuse via reptation above the gel effect, these results show that this only occurs for solutions containing rigid and/or highly immobile macromolecules and in very high concentrations. To describe these behaviours, a semi-empirical ‘composite kt model’ was also developed to describe kti,i(x) as a function of i and x up to high conversions. We showed that the model is very simple to implement and accurate for modelling a wide range of functional monomers and experimental conditions. In particular, we showed the method was accurate for modelling RAFT-mediated polymerisations of a very wide range of monomers (MA, MMA, and PSTY) and was even accurate for modelling conventional FRP’s. Thus, the model provides a simple, flexible and accurate method for predicting the rate of reaction and evolution of molecular weight distributions across a range of experimental conditions based on accurate kti,i(x) values.