Novel, high molecular weight, high glass transition temperature, photocrosslinkable polyimide and poly(imide siloxane) homo- and segmented copolymers were prepared and characterized. The polyimides were synthesized by the classical two step method of first preparing soluble poly(amic acid) prepolymers by the reaction of various aromatic dianhydrides with aromatic diamines. The siloxane modified copolymers were synthesized by reacting single or mixed components of the aromatic dianhydrides with a mixture of aromatic amine and bis(3-aminopropyl) end blocked polydimethyl siloxane oligomers in a cosolvent system of tetrahydrofuran and N-methyl-2-pyrrolidinone. These dysfunctional aminopropyl terminated siloxane oligomers were prepared through an anionic ring opening equilibration polymerization of octamethylcyclotetrasiloxane with bis(3—aminopropyl) tetramethyldisiloxane in the presence of siloxanolate catalyst. Soluble fully imidized polyimides were obtained by use of a solution imidization procedure which utilized a cosolvent system of N-methyl-2-pyrrolidinone and N-cyclohexyl-2-pyrrolidone at temperatures of approximately 170°C. The fully imidized polyimides were soluble in a variety of solvents. The homo- and copolymers have been characterized for compositional analysis by FT-IR and proton NMR spectroscopy. All polymers were characterized for their thermal properties by differential scanning calorimetry, dynamic mechanical thermal analysis and thermogravimetric analysis. All homoand copolymers possessed excellent thermal characteristics and good mechanical properties. The photosensitive properties of the polyimide and poly(imide siloxane) homo- and copolymers were investigated at the UV wavelengths of 313nm and 365nm. The photosensitivities were found to depend on both the amount of benzylic methyl substituted diamine incorporated into the polyimide backbone, and the amount of aromatic ketone concentration. High concentrations of fluorinated (6F) dianhydride were also desirable. Incorporation of the polydimethylsiloxane segments into the polyimide decreased the optical density without decreasing the photosensitivity and therefore desirably allowed thicker films to be crosslinked at lower exposure doses. The adhesion of the siloxane modified polyimides to the silicon wafers was increased with significantly increasing siloxane content and at 20 weight percent, eliminated the need for conventional coupling agents. / Ph. D.

A numerical-perturbation approach is used to study modal interactions in the dynamic response of infinitely long circular cylindrical shells to an external harmonic excitation. The excitation frequency is near the linear natural frequency of the breathing mode (i.e., primary resonance of the breathing mode) and the linear natural frequency of the breathing mode is approximately twice that of a flexural mode (i.e., two-to-one internal or autoparametric resonance). The method of multiple-time scales is used to derive a set of autonomous first-order nonlinear differential equations that describe the modulation of the amplitudes and phases of the interacting modes. The same approach is used to study the axisymmetric dynamic response of spherical shells to a radial harmonic excitation having a frequency near one of the linear natural frequencies of a flexural mode (i.e., primary resonance of a flexural mode) and in the presence of a two-to-one internal resonance between the excited mode and a lower flexural mode. The modulation equations derived for infinitely long circular cylindrical shells and for axisymmetric spherical shells are scaled to the same form and their fixed points, periodic solutions, and chaotic solutions are studied as the amplitude or the frequency of excitation varies. As the excitation amplitude varies, the fixed-point solutions of the modulation equations exhibit the jump and saturation phenomena. They also undergo supercritical and subcritical Hopf bifurcations as the frequency or the amplitude of excitation varies. Between the two Hopf-bifurcation frequencies, the fixed-point solutions are unstable and limit cycles exist. Some limit cycles experience symmetry-breaking (pitchfork) bifurcation followed by an infinite cascade of period-doubling bifurcations culminating in chaos. Other limit cycles lose stability through cyclic-fold bifurcations causing a transition to chaos. The same procedure is used to study the nonlinear dynamic response of infinitely long circular cylindrical shells to a subharmonic excitation of order one-haIf of the breathing mode in the presence of a two-to-one internal resonance. The force-response curves exhibit saturation, jumps, and Hopf bifurcations. They also show that the shell does not respond until a certain threshold level of excitation is exceeded. The frequency-response curves exhibit jumps and pitchfork and Hopf bifurcations. For certain parameters and excitation frequencies between the Hopf-bifurcation values, limit-cycle solutions of the modulation equations are found. As the excitation frequency changes, the limit cycles deform and lose their stability through either pitchfork or cyclic-foId (saddle-node) bifurcations. Some of these saddIe-node bifurcations cause a transition to chaos. The pitchfork bifurcations break the symmetry of the limit cycles. Period-three motions are observed over a narrow range of excitation frequencies. Lastly, a computerized symbolic manipulator is used to analyze the dynamic response of an infinitely long circular cylindrical shell to radial harmonic excitations. The excitation frequency is near the linear natural frequency of a flexural mode (i.e., primary resonance of a flexural mode). Due to the complete circular symmetry of the shell, each natural frequency corresponds to two orthogonal mode shapes. The mode with the same spatial variation as the excitation is called the driven mode and the other mode is called the companion mode. Modal interactions between the driven mode and the companion mode are studied. The steady-state response of the shell can involve either the driven mode alone (single-mode response) or both the driven and companion modes (two-mode response). The frequency-response curve exhibits jumps and Hopf bifurcations. Between the Hopf·bifurcation frequencies, the modulation equations exhibit multiple limit-cycle solutions. As the excitation frequency varies, these limit cycles go through either saddle-node collisions or incomplete sequences of period-doubling bifurcations. Some of the saddle-node bifurcations result in the birth of limit cycles and some result in transition to chaos. / Ph. D.

New flux-split algorithms are developed for high velocity, high-temperature flow situations, when finite-rate chemistry and non-equilibrium thermodynamics greatly affect the physics of the problem. Two flux-vector-split algorithms, of the Steger-Warming and of the Van Leer type, and one flux-difference-split algorithm of the Roe type are established and utilized for the accurate numerical simulation of flows with dissociation, ionization, and combustion phenomena. Several thermodynamic models are used, including a simplified vibrational non-equilibrium model and an equilibrium model based upon refined statistical mechanics properties. The framework provided is flexible enough to accommodate virtually any chemical model and a wide range of non-equilibrium, multi-temperature thermodynamic models. A theoretical study of the main features of flows with free electrons, for conditions that require the use of two translational temperatures in the thermal model, is developed. Interesting and unexpected results are obtained, because acoustic wave speeds of the symmetric form u±α no longer appear. A simple but powerful asymptotic analysis is developed which allows the establishment of the fundamental gas-dynamic properties of flows with multiple translational temperatures. The new algorithms developed demonstrate their accuracy and robustness for challenging flow problems. The influence of several assumptions on the chemical and thermal behavior of the flows is investigated, and a comparison with results obtained using different numerical approaches, in particular spectral methods, is provided, and proves to be favorable to the present techniques. Other calculations in one and two space dimensions indicate large sensitivities with respect to chemical and thermodynamic modeling. The algorithms developed are of sufficient generality to begin to examine these effects in detail. Preliminary numerical simulations are performed using elementary modeling of transport phenomena. / Ph. D.

The genes coding for mannitol-1-phosphate dehydrogenases (mtlD) from Aspergillus parasiticus and Eschericia coli were cloned and sequenced. The two coding regions were highly homologous and the identity was 88.7% at the amino acid level and 81.6% at the nucleotide level. The two genes translate into polypeptides of equal numbers (382) of amino acids with M, of 40,880 and 41,221, respectively. The possible NAD binding sites were identified for both enzymes in the N-terminal regions according to the consensus sequence fingerprint. The C-terminal regions of both enzymes were similar in sequence to the kinase domain of human liver or -rat liver fructose-6-phosphate-2- kinase:fructose·2,6-bisphosphatase, suggesting that the C-terminal regions are involved in fructose-6-phosphate binding. This conclusion was further supported by site-specitic mutagenesis experiments near the 3’ end of the A. parasiticus gene. A modified A. parasiticus mtlD gene directed the expression, in E. coli, of an enzyme in which amino acid residues 362-369 were altered and amino acid residues 370-382 were deleted with respect to the wild type enzyme. This enzyme exhibited 15% of wild type activity with a 3-fold increase in K_m for fructose-6-phosphate. In the 5’ upstream region of the A. parasiticus mtlD gene, no sequence was found which is similar to the consensus sequences derived for either procaryotic or higher eucaryotic gene promoters; however, inverted repeats were identified, which may be important for regulation of gene expression. A sequence similar to the Shine-Dalgarno sequence was found preceding the translation start codon of the A. parasiticus mtlD gene, which is important for its expression in E. coli. In the 3’ downstream region of the A. parasiticus mtlD gene, an additional open reading frame was found, which translated into a polypeptide of 153 amino acids with M, of 17,111. This polypeptide was identified using maxicell experiments. / Ph. D.

A procedure to generate an approximate bifurcation diagram for a single-degree-of-freedom system in a selected parameter space is developed. The procedure is based on the application of Floquet analysis to determine the stability of second-order perturbation approximations of the solutions of the system in the neighborhoods of specific resonances. As a control parameter is varied, a combination of elementary concepts of bifurcation theory and the proposed method are used to detect the first bifurcation from the periodic solutions and hence infer the qualitative changes that the system experiences. Codimension-one bifurcations are investigated in a two-dimensional parameter space composed of the amplitude and frequency of the excitation. The behavior of a softening Duffing oscillator is analyzed under external and parametric excitation. The dynamics of a ship rolling in waves is also considered and three types of excitations are treated: external, parametric, and a combination of both. Analog- and digital-computer simulations are used to verify the accuracy of the analytical predictions. It is found that the predictions based on the first bifurcation of the analytical solution give a good estimate of the actual behavior of the system. The stability regions of the solutions near each of the resonances display a self-similar structure in the parameter space. The physical implications of these bifurcation patterns are important for the prediction of the capsizing of ships. The dangerous regions of the parameter space where capsizing might occur are identified for a given system. Capsizing is found to occur via two distinct scenarios: one evolving from a large oscillation through a disappearance of a chaotic attractor (crises) and a second, potentially more dangerous, developing from a small oscillation through a sudden tangent instability. These scenarios agree with previous experimental studies. / Ph. D.

Liquefaction ls a phenomenon where a saturated soil can temporarily lose its shear strength during an earthquake as a result of the development of excess pore pressures. For the past 25 years since Iiquefaction phenomenon was first explained, it was thought to be mainly a problem with clean sand, and most of the research has focused on these soils. However, as case history information has come to light, it has become apparent that silty sands are commonly involved, and in some cases even silts. This has generated a need for knowledge about the response of silty sands and silts under seismic loading. Related to this issue is the question of how best to determine the Iiquefaction resistance of these soils in a practical setting. This research has the objectives of providing an understanding of the behavior of saturated silty sands under seismic loading, and developing a rational basis for the use of the Cone Penetration Test (CPT) to predict Iiquefaction resistance in these materials. The study is primarily experimental, relying on laboratory and field testing and the use of a unique, large scale calibration chamber. The calibration chamber allows the field environment to be duplicated in the laboratory where conditions can be closely controlled and accurately defined. One of the first problems to be overcome in the research was to determine how to prepare specimens of silty sands that would reasonably duplicate field conditions in both the small scale of the conventional laboratory tests, and the large scale of the calibration chamber. Out of four different methods explored, consolidation from a slurry proved to be best. Two silty sands were located which had the desired characteristics for the study. Field work, involving both the Standard Penetration Test (SPT) and CPT was done as part of this investigation. The behavior of the silty sands were determined in the laboratory from monotonic and cyclic loading tests. The test results show that the effect of fines is to reduce the cone penetration resistance, but not to affect the liquefaction resistance. The steady state shear strength of the soils seems to be correlated to the cone tip resistance, however, this correlation shows a higher steady state shear strength than those back figured from case history data. The results were also used to define state parameters for both of the soils tested. The state parameter was found to be a reliable index to the liquefaction potential and further study in this area is recommended. / Ph. D.

Demand for new underground transportation systems and utility networks has increased the use of tunneling in soft ground. Many of these tunnels have to be constructed in difficult soil conditions, with strict constraints on ground movement control. Technological advances, such as the pressurized shield or the New Austrian Tunneling Method (NATM), have, to some extent, overcome these difficulties. But the complex interaction between tunneling procedure, ground response, and liner support is still not fully understood. In this dissertation, the three aspects of tunneling, face stability, liner design, and ground surface settlement are analyzed for conditions that might be experienced on current projects. The study is intended to clarify some of the phenomena associated with the use of advanced tunneling techniques in soft grounds, and help improve the current design practice. The NATM generally uses "hand-mining" equipment for excavation, and shotcrete as temporary support of the tunnel wall. The amount and timing of support is optimized by continuously adapting the construction procedure to the conditions found at the tunnel face. In the present study, the applications of the finite element method to tunneling are reviewed, and it is used to model NATM tunneling projects. Using parametric studies, a simplified design method is proposed which allows an estimate of the liner forces and settlements associated with NATM tunneling to be obtained. Pressurized shields are used in soils with little to zero stand-up time to support the tunnel face during excavation. In this work, the face stability of shield tunnels in cohesionless soils is examined using limit analysis principles. Upper bound estimates of the critical face pressure are found in good agreement with results from centrifuge model tests. Empirical correlations for settlement estimates are re-examined, in view of case history data for shield driven tunnels. The ground movements observed on the F3 and F4 contracts of the Washington Metro are analyzed. Earth pressure balance shields were used on these projects. It is shown that difliculties were common in mixed face conditions, unless adequate techniques were used to prevent ground collapse to occur. / Ph. D.

The speB gene of E. coli encodes agmatine ureohydrolase (AUH). AUH catalyses the hydrolysis of agmatine to urea and putrescine in a polyamine biosynthetic pathway. The plasmid pKA5, derived from an E. coli genomic library, was the source of a 2.97 kb restriction fragment containing the speB gene. Sequencing of this fragment revealed three intact open reading frames, ORF1 and ORF2 on one strand and ORF3 on the opposite strand, as well as a truncated open reading frame, ORF4, which terminated 92 kb upstream from ORF3. ORF2 and ORF3 were convergent, and overlapped by 85% of their sequence. ORF1 and ORF3 were separated by a sequence of two imperfect repeats containing four palindromes, three of which were overlapping. ORF3 represented the coding sequence of the speB gene. Two transcripts were detected from the speB gene: a shorter transcript, initiated 101 bp upstream from ORF3, and a polycistronic message, coding for ORF3 and ORF4. The short transcript was abundantly expressed when ORF4 sequences were deleted, but when ORF4 and its upstream sequences were present, the polycistronic message predominated and the amount of the monocistronic message was drastically reduced. The promoter producing the shorter transcript required only a -12 TATACT sequence for activity. Deletion of a 460 bp fragment comprising the 5'-region of ORF1 from a plasmid containing ORF1, ORF2 and speB reduced the activity of AUH by 83%. This fragment contained two divergently oriented promoters. The presence of ORF1 did not stimulate ß-galactosidase encoded by the speB promoter fused to lacΖ. Agmatine induced transcription from speB but not from the ORF4 nor the ORF1 promoters. cAMP caused an 88% reduction in the AUH activity of wild type E. coli K-12 but had no effect on the activity of plasmid encoded AUH. The activity of neither the speB nor the ORF4 promoters fused to lacΖ or phoA were influenced by cAMP; in contrast, the lacZ promoter fused to lacZ or phoA was stimulated by cAMP. Thus, the role of cAMP and CRP on speB expression is indirect and limited to a single copy state. / Ph. D.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

No description available.