Search

Waterside corrosion and associated hydrogen generation remain a major limiting factor for the use of zirconium alloys as cladding materials under severe fuel duty conditions. Although it has been observed that minor alloying additions might significantly affect the corrosion response of Zr alloys, the origins for these differences remain unclear. To develop alloys with greater corrosion resistance, it is important to understand the correlations between alloying additions, oxide structure and the oxidation behavior of zirconium alloys. Three alloys were examined: pure Zr (0.009 wt.% Fe), Zircaloy-4 (1.32 wt.% Sn, 0.19 wt.% Fe, and 0.09 wt.% Cr) , and a ZrFeCr alloy (0.38 wt.% Fe and 0.22 wt.% Cr). All three alloys were corroded in autoclaves in 360°C water. Pure Zr loses protectiveness relatively early (

Both frequentist and Bayesian approaches have been used to characterize population pharmacokinetics and pharmacodynamics(PK/PD) models. These methods focus on estimating the population parameters and assessing the association between the characteristics of PK/PD and the subject covariates. In this work, we propose a Dirichlet process mixture model to classify the patients based on their individualized pharmacokinetic and pharmacodynamic profiles. Then we can predict the new patients dose-response curves given their concentration-time profiles. Additionally, we implement a modern Markov Chain Monte Carlo algorithm for sampling inference of parameters. The detailed sampling procedures as well as the results are discussed in a simulation data and a real data example. We also evaluate an approximate solution of a system of nonlinear differential equations from Euler's method and compare the results with a general numerical solver, ode from R package, deSolve.

Elastic words are those whose length can vary between monosyllabic and disyllabic, without changing the meaning. Though elastic words have known to be many in Chinese, it is still not clear how many words are elastic. In addition, there is no consensus on the motivation of creating elastic words. This dissertation offers a complete annotation of elastic words in modern Standard Chinese and sample annotations of Middle Chinese, and investigates why elastic words are created. Specifically, it examines four properties of elastic words focusing on the homophone-avoidance theory and the prosody theory. The former, by far the most popular one, proposes that disyllabic words are created to reduce homophony and avoid ambiguity after massive syllable loss. In contrast, the prosody theory proposes that elastic words are created because disyllabic words are needed in prosodically strong positions, due to the requirement of Foot Binarity. First, a study examines the relation between homophony and elastic words, based on a complete length elasticity annotation of Modern Chinese Dictionary (2005). Results show that there is no correlation between homophony and elastic words. The second study examines the effect of word category on elastic words in modern Standard Chinese. Results show that (i) half of words in Chinese lexicon are elastic; (ii) content words have higher percentage of elastic words than function words. The third study examines the historical development of elastic words, with a focus on Middle Chinese, especially Tang poems. Results show that there are many elastic words in Middle Chinese, similar to that in Modern Chinese. The fourth study examines word length in Chinese dialects, focusing on Mandarin and Cantonese. Results show that they have similar percentages of disyllabic words and that the size of syllable inventory has no effect on word length. Various evidence consistently points to the conclusion that the prosody theory offers a better explanation of why elastic words are created in Chinese, despite of the fact that the homophone-avoidance theory seems quite intuitive and natural. In other words, elastic words are created to fulfill prosodic requirement rather than to compensate for syllable loss or an increase in homophony.

With increasing areal density in magnetic recording systems, perpendicular recording has successfully replaced longitudinal recording to mitigate the superparamagnetic limit. The extensive theoretical and experimental research associated with perpendicular magnetic recording media has contributed significantly to improving magnetic recording performance. Micromagnetic studies on perpendicular recording media, including aspects of the design of hybrid soft underlayers, media noise properties, inter-grain exchange characterization and ultra-high density bit patterned media recording, are presented in this dissertation. To improve the writability of recording media, one needs to reduce the head-tokeeper spacing while maintaining a good texture growth for the recording layer. A hybrid soft underlayer, consisting of a thin crystalline soft underlayer stacked above a non-magnetic seed layer and a conventional amorphous soft underlayer, provides an alternative approach for reducing the effective head-to-keeper spacing in perpendicular recording. Micromagnetic simulations indicate that the media using a hybrid soft underlayer helps enhance the effective field and the field gradient in comparison with conventional media that uses only an amorphous soft underlayer. The hybrid soft underlayer can support a thicker non-magnetic seed layer yet achieve an equivalent or better effective field and field gradient. A noise plateau for intermediate recording densities is observed for a recording layer of typical magnetization. Medium noise characteristics and transition jitter in perpendicular magnetic recording are explored using micromagnetic simulation. The plateau is replaced by a normal linear dependence of noise on recording density for a low magnetization recording layer. We show analytically that a source of the plateau is similar to that producing the Non-Linear-Transition-Shift of signal. In particular, magnetostatic effects are predicted to produce positive correlation of jitter and thus negative correlation of noise at the densities associated with the plateau. One focus for developing perpendicular recording media is on how to extract intergranular exchange coupling and intrinsic anisotropy field dispersion. A micromagnetic numerical technique is developed to effectively separate the effects of intergranular exchange coupling and anisotropy dispersion by finding their correlation to differentiated M-H curves with different initial magnetization states, even in the presence of thermal fluctuation. The validity of this method is investigated with a series of intergranular exchange couplings and anisotropy dispersions for different media thickness. This characterization method allows for an experimental measurement employing a vibrating sample magnetometer (VSM). Bit patterned media have been suggested to extend areal density beyond 1 Tbit/in2. The feasibility of 4 Tit/in2 bit patterned recording is determined by aspects of write head design and media fabrication, and is estimated by the bit error rate. Micromagnetic specifications including 2.3:1 BAR bit patterned exchange coupled composite media, trailing shield, and side shields are proposed to meet the requirement of 3×10-4 bit error rate, 4 nm fly height, 5% switching field distribution, 5% timing and 5% jitter errors for 4 Tbit/in2 bit patterned recording. Demagnetizing field distribution is examined by studying the shielding effect of the side shields on the stray field from the neighboring dots. For recording self-assembled bit-patterned media, the head design writes two staggered tracks in a single pass and has maximum perpendicular field gradients of 580 Oe/nm along the down-track direction and 476 Oe/nm along the crosstrack direction. The geometry demanded by self-assembly reduces recording density to 2.9 Tbits/in2.

Multiphase interleaving buck converters are widely used in today's industrial point-of-load (POL) converters, especially the microprocessor voltage regulators (VRs). The issue of today's multiphase interleaving buck converters is the conflict between the high efficiency and the fast transient in the phase inductor design. In 2000, P. Wong proposed the multiphase coupledinductor buck converter to solve this issue. With the phase inductors coupled together, the coupled-inductor worked as a nonlinear inductor due to the phase-shifted switching network, and the coupled-inductor has different equivalent inductances during steady-state and transient. One the one hand, the steady state inductance is increased due to coupling and the efficiency of the multiphase coupled-inductor buck converter is increased; on the other hand, the transient inductance is reduced and the transient performance of the multiphase coupled-inductor buck is improved. After that, many researches have investigated the multiphase coupled-inductor buck converters in different aspects. However, there are still many challenges in this area: the comprehensive analysis of the converter, the alternative coupled inductor structures with the good performance, the current sensing of converter and the light-load efficiency improvement. They are investigated in this dissertation. The comprehensive analysis of the multiphase coupled-inductor buck converter is investigated. The n-phase (n>2) coupled-inductor buck converter with the duty cycle D>1/n hasn't been analyzed before. In this dissertation, the multiphase coupled-inductor buck converter is systematically analyzed for any phase number and any duty cycle condition. The asymmetric multiphase coupled-inductor buck converter is also analyzed. The existing coupled-inductor has a long winding path issue. In low-voltage, high-current applications, the short winding path is preferred because the winding loss dominates the inductor total loss and a short winding path can greatly reduce the winding loss. To solve this long winding path issue, several twisted-core coupled-inductors are proposed. The twisted-core coupled-inductor has such a severe 3D fringing effect that the conventional reluctance modeling method gives a poor result, unacceptable from the design point of view. By applying and extending Sullivan's space cutting method to the twisted core coupled inductor, a precise reluctance model of the twisted-core coupled-inductor is proposed. The reluctance model gives designers the intuition of the twisted-core coupled-inductors and facilitates the design of the twisted-core coupled-inductors. The design using this reluctance model shows good correlation between the design requirement and the design result. The developed space cutting method can also be used in other complex magnetic structures with the strong fringing effect. Today, more and more POL converters are integrated and the bottleneck of the integrated POL converters is the large inductor size. Different coupled-inductor structures are proposed to reduce the large inductor size and to improve the power density of the integrated POL converter. The investigation is based on the low temperature co-fire ceramic (LTCC) process. It is found that the side-by-side-winding coupled-inductor structure achieves a smaller footprint and size. With the two-segment B-H curve approximation, the proposed coupled-inductor structure can be easily modeled and designed. The designed coupled-inductor prototype reduces the magnetic size by half. Accordingly, the LTCC integrated coupled-inductor POL converter doubles the power density compared to its non-coupled-inductor POL counterpart and an amazing 500W/in³ power density is achieved. In a multiphase coupled-inductor converter, there are several coupled-inductor setups. For example, for a six-phase coupled-inductor converter, three two-phase coupled inductors, two three-phase coupled-inductors and one six-phase coupled inductors can be used. Different coupled-inductor setups are investigated and it is found that there is a diminishing return effect for both the steady-state efficiency improvement and the transient performance improvement when the coupling phase number increases. The conventional DCR current sensing method is a very popular current sensing method for today's multiphase non-coupled-inductor buck converters. Unfortunately, this current sensing method doesn't work for the multiphase coupled-inductor buck converter. To solve this issue, two novel DCR current sensing methods are proposed for the multiphase coupled-inductor buck converter. Although the multiphase coupled-inductor buck converters have shown a lot of benefits, they have a low efficiency under light-load working in DCM. Since the DCM operation of the multiphase coupled-inductor buck converter has never been investigated, they are analyzed in detail and the reason for the low efficiency is identified. It is found that there are more-than-one DCM modes for the multiphase coupled-inductor buck converter: DCM1, DCM2 …, and DCMn. In the DCM2, DCM3 …, and DCMn modes, the phase-currents reach zero-current more-than-once during one switching period, which causes the low efficiency of the multiphase coupledinductor buck converter in the light load. With the understanding of the low efficiency issue, the burst-in-DCM1-mode control method is proposed to improve the light load efficiency of the multiphase coupled-inductor buck converter. Experimental results prove the proposed solution.