Search

The topic of this thesis is the synthesis of Taxane derivatives using the asymmetric conjugate addition of alkenes. Taxol is an important anti-cancer drug and owing to its complex structure, remains a subject of fascination by chemists. The asymmetric conjugate addition of alkenes is a power transformation and holds promise in synthesising complex molecules such as Taxol. In Chapter 1, we prove that this is a feasible undertaking by synthesising the C-ring of the Taxol core containing the C8 stereocenter. A complex nucleophile containing a bromide and substituted alkene was successfully synthesised on decagram scale and underwent asymmetric conjugate addition to give this intermediate in 95% yield and 88% ee. In Chapter 2, the trapping of zirconium enolates generated from asymmetric conjugate additions was investigated. Several electrophiles were screened and it was found that only formaldehyde or the Vilsmeier-Haack reagent would give rise to any trapped products. The Vilsmeier-Haack reagent, gave β-chloroaldehydes in modest yields (between 41-60%) and high enantioselectivities (up to 98% ee) for adducts containing quaternary stereocenters. The synthetic utility of these compounds was also demonstrated by synthesising derivatives (Grignard addition and Suzuki-Miyaura couplings). In Chapter 3, we envisged using an unprecedented transannular electrocyclic reaction as a key step to simultaneously form the A and B rings. The precursor 10-membered ring was synthesised by cross metathesis of an intermediate prepared in chapter 2, with a geminal diboron butadiene followed by ring closing Suzuki coupling. Ultimately, we were unsuccessful in carrying out the desired electrocyclic reaction. We discuss possible reasons as to why this was the case. In Chapter 4, the type-II intramolecular Diels-Alder reaction is reviewed in the context of Taxane syntheses and then applied to our model substrate. Using this reaction, we were able to achieve our original objective by synthesising the Taxol core in five steps from simple starting materials in 11% yield, 92% ee and 1:1 d.r. at C1.

This thesis presents a functional analysis of some multi-bar linkage mechanisms in animals. In particular it investigates the effect of multi-bar linkage mechanisms on the locomotion efficiency of animals. Three case studies are presented. The first case study involves modeling the deployment of the jaw mechanism in the sling jaw wrasse. Numerical modeling of the energy required to move in water shows that the jaw mechanism can signifIcantly reduce the energy cost of locomotion. The second case study involves mode ling the effect of the pantograph mechanism in bird wings on the inertial power required for flight. Energy modeling shows that the pantograph mechanism has a small but significant advantageous effect on locomotion by reducing the inertia of wings. The energy modeling also shows that significant aerodynamic braking occurs during flapping. The third case study involves modeling the benefIts of the linkage mechanism inside insect wings. This study shows there are significant benefits in having 4-bar mechanisms because the mechanisms provide compliance that enables elastic5-storage to take place during wing flapping. The case studies show that linkage mechanisms enable complex mechanical functions to take place that lead to signifIcant energy savings. A common factor in all the case studies was that the actuator is optimally located where mass and space is not at a premium. The potential for biomimicry is discussed for each case study. The sling wrasse jaw has potential for harvesting and cleaning applications. The bird pantograph mechanism and insect wing mechanism has potential for micro air vehicle design and robotics design. There are similarities between multi-bar mechanisms in engineering and nature in terms of types of mechanism. However, there are significant differences in the detail implementation. Engineering mechanisms have pinned joints whereas in nature there are pivot points. Engineering mechanisms tend to be planar whereas in nature they are 3-dimensional. Engineering mechanisms tend to be fully constrained whereas nature's mechanisms tend to be under constrained.

Motorway merging has long been regarded as a major source of conflicts and congestion on motorways. Traditional studies of merging behaviour are based on gap acceptance models developed mainly for urban intersections, which tend to oversimplify the very complex dynamic interactive merging behaviour involved. It is believed that this research represents the first comprehensive investigation and modelling of dynamic merging interactions at motorway on-ramps. Emphasis has been given to improving the modelling of merging behaviour and in particular to capture the cooperation between the merging and motorway traffic. This research has developed a feasible integrated microscopic simulation framework to model the interactions among traffic in motorway merging sections. This has been achieved by developing an integrated model (MergeSim) consisting of two sub-models working in tandem: a car-following and a merging model. By assuming different reaction times for different driver states (alert, non-alert and close-following), the new car-following model is shown to be able to capture traffic breakdown, hysteresis, shockwave propagations and close-following situations. The merging model is developed to capture both the acceleration and gap acceptance behaviour of the merging traffic, and the cooperative behaviour of the motorway traffic. The merging model is composed of several sub models: for the traffic in the motorway nearside lane, there is a cooperation model to simulate the cooperative lane-changing and courtesy yielding behaviour and the interactions with the merging traffic; for the merging traffic in the acceleration lane, there are models such as acceleration model, gap selection model, gap acceptance model and a merge model. Sensitivity tests have shown that the integrated model can reasonably replicate all relevant behaviour of individual drivers in merging areas such as normal carfollowing, close-following, cooperative lane-changing, courtesy yielding and gap acceptance. The sensitivity tests on the different merging lengths showed that increased length might reduce merging failures (i. e. the occurrence that the merging driver fails to move into the motorway before reaching the end of the acceleration lane). It can be explained that more merging traffic can successfully take the following gaps with increased merging lengths, which has implications for the geometric configuration of the acceleration lane. The study also established a general calibration and validation framework designed for real-world applications in highway networks using the most readily available traffic surveillance data, the loop detector data. Currently no commonly agreed bench-marking procedure exists (Brockfeld et al., 2005), and this framework has the advantage that the concept and the proposed methodology are suitable for general application to other micro-simulation models using detector data sets. In conclusion, the integrated simulation model (MergeSim) can reliably be used as a tool for further studies and investigations into the effectiveness of techniques related to motorway merging operations.

Protein and oil are the major chemical constituents of soybean seed that affect the quality of soyfood, feed, and oil products. Therefore, soybean cultivars with high protein and/or high oil are desirable for the soyfood and feed markets. Use of molecular-marker-assisted selection will facilitate the breeding process of such cultivar development. The objectives of this research were to identify new quantitative trait loci (QTL) and confirm previously reported QTL associated with seed protein and oil content by using simple sequence repeat (SSR) markers and single nucleotide polymorphism (SNP) markers. Two recombinant inbred line (RIL) populations consisting of 242 individuals from R05-1415 (high protein/low oil) x R05-638 (low protein/high oil) (population 1) and 214 individuals from R05-4256 (high oil/low protein) x V97-1346 (low oil/high protein) (population 2) were used in QTL mapping. F2 plants from the mapping populations were used for SSR/SNP genotyping. In the marker screening, 120 out of 626 SSR and 1652 out of 5361 SNP markers were polymorphic. The RILs from both populations were grown in a randomized complete block design in Argentina in 2010, Stuttgart and Keiser, AR in 2011 and 2012. Seed from F2:3, F2:4 and F2:5 lines were tested for protein and oil content by using near infrared transmittance technique based on 13% moisture. Protein and oil content in both RIL populations exhibited a typical normal distribution. Single marker analysis (SMA) and composite interval mapping (CIM) revealed two novel oil QTL on chromosomes 14 and 6 in population 2 which account for 17% and 13% of the oil content variation, respectively. A minor protein QTL was confirmed on chromosome 14. One major QTL with large effect was confirmed on chromosome 20 across genetic populations, locations, and years; this QTL has opposite effects on seed protein and oil content. Eight new SNP markers flanking this QTL region on chromosome 20 were identified in population 2. These new and confirmed QTL along with linked molecular markers for seed protein and oil content can be used for marker-assisted selection for seed composition improvement in soybean breeding programs.

With the success of Revenue Management (RM) techniques over the past three decades in various segments of the service industry, many manufacturing firms have started exploring innovative RM technologies to improve their profits. This dissertation studies RM for make-to-order (MTO) and make-to-stock (MTS) systems. We start with a problem faced by a MTO firm that has the ability to reject or accept the order and set prices and lead-times to influence demands. The firm is confronted with the problem to decide, which orders to accept or reject and trade-off the price, lead-time and potential for increased demand against capacity constraints, in order to maximize the total profits in a finite planning horizon with deterministic demands. We develop a mathematical model for this problem. Through numerical analysis, we present insights regarding the benefits of price customization and lead-time flexibilities in various demand scenarios. However, the demands of MTO firms are always hard to be predicted in most situations. We further study the above problem under the stochastic demands, with the objective to maximize the long-run average profit. We model the problem as a Semi-Markov Decision Problem (SMDP) and develop a reinforcement learning (RL) algorithm-Q-learning algorithm (QLA), in which a decision agent is assigned to the machine and improves the accuracy of its action-selection decisions via a “learning" process. Numerical experiment shows the superior performance of the QLA. Finally, we consider a problem in a MTS production system consists of a single machine in which the demands and the processing times for N types of products are random. The problem is to decide when, what, and how much to produce so that the long-run average profit. We develop a mathematical model and propose two RL algorithms for real-time decision-making. Specifically, one is a Q-learning algorithm for Semi-Markov decision process (QLS) and another is a Q-learning algorithm with a learning-improvement heuristic (QLIH) to further improve the performance of QLS. We compare the performance of QLS and QLIH with a benchmarking Brownian policy and the first-come-first-serve policy. The numerical results show that QLIH outperforms QLS and both benchmarking policies.