No description available.

Probabilistic graphical models like Bayesian Networks (BNs) are powerful artificial-intelligence formalisms, with similarities to cognition and higher order reasoning in the human brain. These models have been, to great success, applied to several challenging real-world applications. Use of these formalisms to a greater set of applications is impeded by the limitations of the currently used software-based implementations. New emerging-technology based circuit paradigms which leverage physical equivalence, i.e., operating directly on probabilities vs. introducing layers of abstraction, promise orders of magnitude increase in performance and efficiency of BN implementations, enabling networks with millions of random variables. While majority of applications with small network size (100s of nodes) require only single digit precision for accurate results, applications with larger size (1000s to millions of nodes) require higher precision computation. We introduce a new BN integrated circuit fabric based on mixed-signal magneto-electric circuits which perform probabilistic computations based on the principle of approximate computation. Precision scaling in this fabric is logarithmic in area vs. linear in prior directions. Results show 33x area benefit for a 0.001 precision compared to prior direction, while maintaining three orders of magnitude performance benefits vs. 100-core processor implementations.

In recent years, spin relaxation, injection, and manipulation in semiconductors have attracted considerable interest because of several potential applications in "spintronic" devices and the necessity to understand and control spin-based phenomena. In light of the growing interest in spin-related phenomena and devices, there is now renewed interest in the science and engineering of narrow gap semiconductors (NGS). NGS based heterostructures are particularly interesting for spintronic applications due to their large spin-orbit coupling, which leads to considerable zero-field spin splitting. NGS are also candidates for electronic applications, such as high-speed and low-power microprocessors; as reported recently by Intel. Furthermore, as switching rates in electronic devices are pushed to even higher frequencies, it is important to understand dynamics in semiconductors such as NGS on femtosecond time-scales. In this thesis, time-resolved studies of magnetic and non-magnetic NGS using ultrafast-laser spectroscopy techniques such as pump-probe spectroscopy and magneto-optical Kerr/Faraday effect, are reported. Our samples include: InSb-based quantum wells with different confinement potentials; InMnSb films, the newest III-V ferromagnetic semiconductors; and InAs films. The samples for these studies have been provided by the groups of Prof. Santos at the University of Oklahoma, Prof. Furdyna at the University of Notre Dame, and Prof. Guido at Virginia Tech. The objectives in this thesis have been to: a) understand charge/spin dynamics in NGS with novel confinement potentials, b) probe the effect of magnetic impurities on the spin/charge dynamics, and c) develop concepts for spin based device applications. Several specific questions and concepts have been addressed including: the effect of large spin-orbit interaction in NGS on the dynamics, how large Rashba spin splitting in these materials affect the spin coherence life time, and carrier/spin dynamics in ferromagnetic semiconductor structures. / Ph. D.

This research undertakes the problem of vibration control of vehicular and structural systems using intelligent materials and controllable devices. Advanced modeling tools validated with experimental test data are developed to help with understanding the fundamentals as well as advanced and novel applications of smart and conventional suspension systems. The project can be divided into two major parts. The first part is focused on development of novel smart suspensions using Magneto-Rheological (MR) fluids in unique configurations in order to improve efficiency, controllability, and safety of today's vehicles. In this part of the research, attention is paid to fundamentals as well as advanced applications of MR technology. Extensive rheological studies, both theoretical and experimental, are performed to understand the basic behaviors of MR fluids as complex non-Newtonian fluids in novel applications. Using the knowledge obtained from fundamental studies of MR fluids, unique application concepts are investigated that lead to design, development, and experimental testing of two new classes of smart devices: MR Hybrid Dampers and MR Squeeze Mounts. Multiple generations of these devices are built and tested as proof of concept prototypes. Advanced physics-based mathematical models are developed for these devices. Experimental test data are used to validate the models and great agreement is obtained. The models are used as design tools at preliminary as well as detailed design stages of device development. The significant finding in this part of the research is that MR fluids can deliver a much larger window of controllable force in squeeze mode compared to shear and valve modes which can be used in various applications. The second part of the research is devoted to the development of innovative design tools for suspension design and tuning. Various components of suspension systems are studied and modeled using a new physics-based modeling approach. The component of main interest is the shim stack assembly in hydraulic dampers which is modeled using energy and variational methods. A major finding is that the shims should be modeled individually in order to represent the sliding effects properly when the shim stack is deflected. Next, the individual component models are integrated into a full suspension model. This model is then used as a tool for suspension design, synthesis, and tuning. Using this design tool, suspension engineers in manufacturing companies and other industrial sections can easily perform parametric studies without the need to carry out time consuming and expensive field and laboratory tests. / Ph. D.

This thesis describes the design, construction and optimisation of two compact setups to produce ⁸⁷Rb Bose-Einstein condensates and dual ⁷Li-⁸⁷Rb Magneto- Optical Traps (MOTs). The motivation for compact systems is to have simplified systems to cool the atoms. The first experimental setup is based on a single pyrex glass cell without the need for atom chips. Fast evaporation will be achieved in a hybrid trap comprising of a magnetic quadrupole trap and an optical dipole trap created by a Nd:YVO4 laser and with future plans of using a Spatial Light Modulator (SLM). To enhance an efficient and rapid evaporation, we have investigated Light-Induced Atomic Desorption (LIAD) to modulate the Rb partial pressure during the cooling and trapping stage. With this technique, a ⁸⁷Rb MOT of 7 x 10⁷ atoms was loaded by shining violet light from a LED source into the glass cell, whose walls are coated with rubidium atoms. The atoms were then cooled by optical molasses and then loaded into a magnetic trap where lifetime measurements demonstrated that LIAD improves on magnetically-trapped atoms loaded from constant background pressure by a factor of six. This is quite encouraging and opens the possibility to do a rapid evaporation. In a second experiment, we have designed a compact system based on a stainless steel chamber to trap either ⁷Li or ⁶Li atoms in a MOT loaded from alkali-metal dispensers without the need of conventional oven-Zeeman slower. This setup can also load ⁸⁷Rb atoms, allowing future projects to simultaneously produce degenerate quantum gases of bosonic ⁸⁷Rb and fermionic ⁶Li atoms.

No description available.

No description available.

Previous studies have demonstrated the sensitivity of the amplitude of the microwave radiation-induced magnetoresistance oscillations to the microwave polarization. These studies have also shown that there exists a phase shift in the linear polarization angle dependence. But the physical origin of this phase shift is still unclear. Therefore, the first part of this dissertation analyzes the phase shift by averaging over other small contributions, when those contributions are smaller than experimental uncertainties. The analysis indicates nontrivial frequency dependence of the phase shift. The second part of the dissertation continues the study of the phase shift and the results suggest that the specimen exhibits only one preferred radiation orientation for different Hall-bar sections. The third part of the dissertation summarizes our study of the Hall and longitudinal resistance oscillations induced by microwave frequency and dc bias at low filling factors. Here, the phase of these resistance oscillations depends on the contact pair on the device, and the period of oscillations appears to be inversely proportional to radiation frequency.

Research in future generation computing is focused on reducing energy dissipation while maintaining the switching speed in a binary operation to continue the current trend of increasing transistor-density according to Moore’s law. Unlike charge-based CMOS technology, spin-based nanomagnetic technology, based on switching bistable magnetization of single domain shape-anisotropic nanomagnets, has the potential to achieve ultralow energy dissipation due to the fact that no charge motion is directly involved in switching. However, switching of magnetization has not been any less dissipative than switching transistors because most magnet switching schemes involve generating a current to produce a magnetic field, or spin transfer torque or domain wall motion to switch magnetization. Current-induced switching invariably dissipates an exorbitant amount of energy in the switching circuit that nullifies any energy advantage that a magnet may have over a transistor. Magnetoelastic switching (switching the magnetization of a magnetostrictive magnet with voltage generated stress) is an unusual switching paradigm where the dissipation turns out to be merely few hundred kT per switching event – several orders of magnitude less than that encountered in current-based switching. A fundamental obstacle, though, is to deterministically switch the magnetization of a nanomagnet between two stable states that are mutually anti-parallel with stress alone. In this work, I have investigated ways to mitigate this problem. One popular approach to flip the magnetizations of a nanomagnet is to pass a spin polarized current through it that transfers spin angular moment from the current to the electrons in the magnet, thereby switching their spins and ultimately the magnet’s magnetization. This approach – known as spin transfer torque (STT) – is very dissipative because of the enormous current densities needed to switch magnets, We, therefore, devised a mixed mode technique to switch magnetization with a combination of STT and stress to gain both energy efficiency from stress and deterministic 180o switching from STT. This approach reduces the total energy dissipation by roughly one order of magnitude. We then extended this idea to find a way to deterministically flip magnetization with stress alone. Sequentially applying stresses along two skewed axes, a complete 180o switching can be achieved. These results have been verified with stochastic Landau-Lifshitz-Gilbert simulation in the presence of thermal noise. The 180o switching makes it possible to develop a genre of magneto-elastic memory where bits are written entirely with voltage generated stress with no current flow. They are extremely energy-efficient. In addition to memory devices, a universal NAND logic device has been proposed which satisfies all the essential characteristics of a Boolean logic gate. It is non-volatile unlike transistor based logic gates in the sense that that gate can process binary inputs and store the output (result) in the magnetization states of magnets, thereby doubling as both logic and memory. Such dual role elements can spawn non-traditional non-von-Neumann architectures without the processor and memory partition that reduces energy efficiency and introduces additional errors. A bit comparator is also designed, which happens to be all straintronic, yet reconfigurable. Moreover, a straintronic spin neuron is designed for neural computing architecture that dissipates orders of magnitude less energy than its CMOS based counterparts. Finally, an experiment has been performed to demonstrate a complete 180o switching of magnetization in a shape anisotropic magnetostrictive Co nanomagnet using voltage generated stress. The device is synthesized with nano-fabrication techniques namely electron beam lithography, electron beam evaporation, and lift off. The experimental results vindicate our proposal of applying sequential stress along two skewed axes to reverse magnetization with stress and therefore, provide a firm footing to magneto-elastic memory technology.

Cette thèse traite de la compensation magnétique de pesanteur. Tout d’abord, des expériences de lévitation magnétique de fluides sont interprétées à l’aide d’un potentiel magnéto-gravitationnel SL. Puis, l’utilisation d’une méthode générale d’analyse de la force magnétique grâce aux harmoniques du champ magnétique est développée. Elle souligne l’importance et le rôle de chacun des trois premiers harmoniques du champ magnétique sur les configurations de forces résultantes inhérentes à la compensation magnétique de pesanteur. En géométrie cylindrique (invariante par translation) diverses combinaisons de forces d’origines magnétique, gravitationnelle et centrifuge offrent des perspectives nouvelles pour la lévitation magnétique. Une combinaison judicieuse des forces magnétiques et centrifuges permet de compenser exactement la pesanteur sur des matériaux diamagnétiques. En géométrie axisymétrique (invariante par rotation), le dimensionnement de stations de lévitation d’oxygène, techniquement réalisables (NbTi@4,2K), est présenté. Ces stations permettent de léviter des volumes d’oxygène supérieurs à 1 litre avec des inhomogénéités inférieures à 1%. La constitution de ces stations rend possible les variations spatiales et temporelles des configurations d’accélérations résultantes. Enfin, la compensation magnétique dynamique de gravité, à l’aide d’une station de lévitation réelle, est étudiée afin de simuler des phases d’accélération ou de décélération d’engins spatiaux / The thesis deals with magnetic gravity compensation. First of all magnetic levitation experiments are explained with the help of a magneto-gravitational potential SL. Next, a general analysis method of the magnetic force is developed which employs magnetic field harmonics. The method underlines both the significance and role of the first three magnetic field harmonics on the resulting forces inherent in magnetic gravity compensation. In cylindrical geometry – with translational invariance – various combination of magnetic, gravitational and centrifugal forces open new possibilities for the magnetic levitation. A suitable combination of both magnetic and centrifugal forces allows exactly compensating gravity on diamagnetic materials. In axisymmetric geometry – with rotational invariance – designs of feasible oxygen magnetic levitation stations are introduced (NbTi@4,2K). Levitation of oxygen volumes more than one litre with inhomogeneities less than 1% can be accomplished within these magnetic levitation facilities. The constitution of the stations makes possible both spatial and temporal variations of the resulting acceleration configurations. At last the dynamic magnetic compensation of gravity with a real coil system is studied so as to simulate both acceleration and deceleration of spaceships