This dissertation focuses on the thorough characterization of nanoscale interface regions - so-called domain walls - between structurally, electronically and magnetically homogeneous areas or domains in BiFeO3 thin films. These domain walls exhibit intriguing properties not found in the domains themselves and were determined using a combination of nanoscale characterization techniques such as transmission electron microscopy, atomic force microscopy, piezo-response force microscopy, scanning tunneling microscopy, X-ray absorption spectroscopy and photoemission electron microscopy. Enhanced net magnetic moments are observed at certain type of domain

walls (109◦) in BiFeO3, although the adjacent domain areas are antiferromagnetically ordered with only a very small net magnetic moment. The same type of the domain walls is found to be conducting at a semiconductor level while the domains are insulating and exhibit robust ferroelectricity. 2-D arrays of these domain walls exhibit intriguing magnetotransport behavior: when both current path and magnetic field are oriented parallel to the domain walls, about 60% of negative magnetoresistance is observed at low temperature (~10 K), which can be well modeled with electron variable hopping. Moreover, in highly strained BiFeO3 thin films electrically controllable magnetic moments are found localized in a highly distorted rhombohedral-like phase confined between the adjacent tetragonal-like phase. Those magnetic moments can be turned on and off by applying an electric field.

As predicted by the International Roadmap for Semiconductors, III-V metal-oxide semiconductor field-effect transistors(MOSFETs) are the prime candidates for the 7-nm node and beyond owing to their high mobilities. However, several challenges need to be overcome before III-V materials can replace silicon in extremely scaled devices. As the size of semiconductor devices enters into the nanoscale regime, it is important to understand quantum mechanical effect in electron transport because quantum mechanical effects such as microscopic scattering, quantum mechanical tunneling, carrier confinement and interference effects play a crucial role. We have investigated quantum transport properties of III-V and silicon in nano-scale devices using the non-equilibrium Green’s function (NEGF) formalism coupled with a 20 orbital sp3d5s*-SO tight-binding model. A mode space NEGF approach combined with the full-band WKB model has been introduced to calculate band-to-band tunneling current under the valence band where an atomistic full-band NEGF approach provides a zero current. Self-energy functions for carrier scattering have been implemented within the mode space NEGF formalism to instigate the influence of microscopic scattering effects on quantum transport. This approach has been benchmarked by comparing the scattering rates obtained from Fermi’s golden rule which is widely used in traditional semi-classical calculations. The scattering model is used to study the role of phonon scattering and surface roughness scattering on the carrier transport characteristics of Si and III-V channel devices. In the presence of the electron-phonon interactions, the drain current decreases compared with its ballistic limit, and the current reduction ratio increase as the channel length increases. Effects of indium mole fraction and source/drain doping density (NSD) on the performance of nanoscale III-V MOSFETs are explored and their performance is compared with that of Si transistors. For III-V's, we have found an optimum indium mole fraction and NSD that maximize the Ion/Ioff ratio and Ion by balancing injection velocity and short channel effect.

Novel phenomena and functionalities at epitaxial complex oxide heterostructures have been attracting huge scientific attention because of the intriguing fundamental physics as well as potential for technological applications that they embody. Essentially, charge and spin reconstruction at the interface can lead to exotic properties, which are completely different from those inherent to the individual materials, for example, a conductive interface between two insulating materials and interface ferromagnetism in the proximity of an antiferromagnet. The interplay between charge and spin degrees of freedom can be particularly intriguing, leading to a fascinating realm, called multiferroic. In this dissertation, a systematic study is performed on the electronic (charge) and magnetic (spin) interaction/reconstruction across the interface of an all-oxide model heterostructure system consisting of the ferromagnet (FM) La$_{0.7}$Sr$_{0.3}$MnO$_3$ (LSMO) and the multiferroic (ferroelectric and antiferromagnetic) BiFeO$_3$ (BFO). The study demonstrates two pathways of using these exotic interfacial properties to control bulk properties, both the ferroelectricity in BFO and ferromagnetism in LSMO.

The journey starts with the growth of high-quality BFO/LSMO heterostructures with unit-cell precision control using reflection high-energy electron diffraction combined with pulsed-laser deposition, providing an important platform for the investigation of electronic and magnetic coupling phenomena across the interface. First, we have observed a novel consequence of the interface electronic interaction due to the so-called ``polar discontinuity'', namely, a built-in electrostatic potential accumulates across the heterointerface, and provides deterministic control of ferroelectric polarization states in thin films. This observation suggests a strong, delocalized effect with important implications for future electronics based on such materials. Secondly, we have revealed a strong magnetic coupling at this interface, manifested in the form of an enhanced coercive field as well as a significant exchange-bias coupling. Based on our x-ray magnetic circular dichroism studies, the origin of the exchange-bias coupling is attributed to a novel ferromagnetic state formed in the antiferromagnetic BFO sublattice at the interface with LSMO. Thirdly, using a field effect geometry, we have proposed a pathway to use an electric field to control the magnetism in LSMO in which the ground state of the interfacial ferromagnetic state is strongly correlated with the ferroelectric polarization. Magnetotransport measurements clearly demonstrate a reversible switch/control between two distinct exchange-bias states by isothermally switching the ferroelectric polarization of BFO. This is an important step towards controlling magnetization with the electric field, which may enable a new class of electrically controllable spintronic devices and provide a new basis for producing electrically controllable spin-polarized currents. Finally, combining experimental results with first-principle and phenomenological model calculations, a microscopic model has been proposed to understand the underlying physics of the magnetoelectric coupling, providing further insights on achieving the electric-field control of magnetism.

In summary, our studies on the interfacial electronic and magnetic properties at BFO/LSMO heterointerfaces have revealed a strong interplay between the charge, spin, orbital and lattice degrees of freedom at the interface, which will have important implications for a new pathway to use the interface properties to control bulk functionalities (ferroelectric polarization and ferromagnetic magnetization in this study). Such couplings at the interface may be extended to other oxides and will bring into play remarkable physical concepts to this developing field of complex oxide heterointerfaces.

Multiferroic magnetoelectric (ME) materials, which simultaneously show ferroelectricity and magnetic ordering, have attracted a huge scientific attention due to both the intriguing fundamental importance and the great technological potential, as a result of the coupling between the dual ferroic orderings. However there are only a few single-phase multiferroic materials in nature and the performance is limited by the low ordering temperature, the small polarization/magnetization or the weak coupling efficiency. Another promising pathway to engineer ME coupling is through designing heterostructures. The current studies of ME coupling (electric field control of magnetism) in heterostructures can be divided into 3 routes: (1) Using piezoelectric effects to change the strain (lattice); (2) Using ferroelectric polarization to tune the carriers (charge); (3) Using multiferroic materials to manipulate the moments by magnetic coupling (spin). Previous studies of an all-oxide model heterostructure system that consists of the ferromagnet La0.7Sr0.3MnO3 (LSMO) and the BiFeO3 (BFO) reveals an interesting charge and spin interactions at the interface. Following the same route, the first part of the dissertation focuses on two different pathways to improve the ME coupling in manganite/BFO model system. Chapter 3 demonstrates that the ME coupling in LSMO/BFO is dramatically different by changing the atomic stack-ing sequence at the interface. In chapter 4, another model system La0.5Ca0.5MnO3 (LCMO), which is at the ferromagnetic/antiferromagnetic phase boundary, is studied instead of LSMO to explore the limit within this route. The results complementarily suggest the importance of a more comprehensive design rule at the atomic scale in order to achieve a better ME coupling efficiency.

Furthermore, although the coupling between the lattice, charge and spin has been intensively studied in terms of ME coupling, the orbital degree of freedom has been neglected so far. In principle, orbital is strongly coupled to lattice. Therefore a large ME coupling could be expected if a strong spin-orbit interaction could be established. The second part of the dissertation concentrates on pursuing this new ideal through two different pathways. Chapter 5 presents a systematic study of the in-situ strain effect on the magnetic and orbital orderings of a model system Nd0.5Sr0.5MnO3. The results demonstrate the close correlation between the orbital ordering parameter and the strain. However the magnetic ordering parameter is less sensitive to strain in this system. With the implication from chapter 5, 5d transition metal oxide SrIrO3(SIO), which hosts a strong intrinsic spin-orbit coupling due to the large atomic number, is studied in chapter 6. Superlattice LSMO/SIO shows an interesting novel magnetic state with a large orbital momentum in the nominally paramagnetic SIO, which is likely to be a very promising candidate to engineer the ME coupling at the interface.

In summary, our studies on engineering the ME effects in heterostructures have revealed two key implications: 1. in the traditional routes that have been extensively studied, further improvements are possible by carefully engineering the interface; 2. the coupling between spin and orbital degrees of freedom is likely to be another promising route to investigate the ME coupling in heterostructures.

This dissertation presents a growth and characterization study of MoS2 thin films

deposited by pulsed laser deposition, with specific interest in demonstrating precise

thickness control down to a single monolayer and high film uniformity over large

areas (0.25 cm2). Limits to traditional methods for scaling field effect based switching

devices have necessitated the development of alternative device architectures.

Devices utilizing two-dimensional materials such as MoS2 have demonstrated

improved electrostatic integrity at small scales while tunneling field effect transistors

have demonstrated a pathway toward reduced power consumption. However,

drawbacks to previously explored synthesis techniques have limited the applicability

of MoS2 for these applications. Toward this goal, the work presented establishes

repeatable growth conditions to fabricate highly crystalline and uniform MoS2 thin

films over technologically significant dimensions and establishes the efficacy of pulsed

laser deposition as a suitable synthesis technique. Films grown herein are found to be

P-type with significant defect densities suggesting avenues for future research.

This dissertation describes the materials science of memristive switching in titanium dioxide. It discusses the strucutral changes of the oxide that take place and provides a phenomenological model for the dynamical behavior during switching. Further, the dissertation describes a two-state variable nanoscale device based on metal insulator transitions and memristive switching. It concludes with a discussion of various applications of titanium oxide memristive devices.

In recent years, the escalating demand for computing has necessitated the discovery of novel materials to power the next generation of logic and memory devices. Among these materials, ferroelectric hafnia (Hafnium Zirconium Oxide or HZO) stands out as a promising candidate. This thesis delves into the multifaceted realm of thin film ferroelectric HZO, spanning from deposition processes to the analysis of switching and fatigue behaviors. The study utilizes two distinct material systems: HZO/pyrochlore for fundamental investigations and HZO/La0.67Sr0.33MnO3 for practical applications, including Ferroelectric Tunnel Junctions (FTJ), enabling a comprehensive exploration.Chapter 3 focuses on epitaxial HZO with pyrochlore electrodes, revealing strain's role in stabilizing HZO's polar orthorhombic phase. It notes the high switching field in pure epitaxial HZO, hints at an order-disorder phase transition, and introduces a single-crystal ferroelectric epitaxial HZO system. Chapter 4 delves into the fatigue bahavior of hafnia, showing oxygen intercalation and defect chemistry's impact on stabilizing the orthorhombic phase, crucial for endurance behavior. Chapter 5 highlights Tunnel Electro-Resistance effects in HZO-based Ferroelectric Tunnel Junctions and the scale-free property of ferroelectric HZO, promising for high-density memory applications. In summary, this thesis advances HZO-based ferroelectric materials' understanding, laying the groundwork for future research in advanced computing and memory systems.

Central to materials science is the concept that functional properties are derived from structure. By understanding the structure of a material system from the atomic scale to the mesoscale we can understand the emergence of novel phenomena and further tailor these properties for practical applications. Here, I focus on the structure-property relations of ferroelectric vortices formed in superlattices PbTiO3/SrTiO3 with layer thicknesses of only about 20 unit cells. These polar vortices represent the first discovery of a ferroelectric topological texture and were only discovered less than a decade ago. While this discovery has led to an explosion of research on the vortices, there has been a significant lack of work relating to the finer details of the polar vortices. Therefore, in my work, I set about understanding the vortex structure and ordering at various length scales to further understand ferroelectricity and chirality within the system. I first examine the atomic scale structure from the shape of an individual to a vortex to their formation of a one-dimensional lattice. From there I examine domain formation within the system in particular the various domain variants and domain walls present. By understanding the vortex ordering and domain formation I then demonstrate the existence of macroscopic ferroelectricity within the system by scanning probe measurements, electrical characterization, and non-linear optical techniques. I demonstrate how vortex off-centering leads to ferroelectric behavior with multiple switching events that stem from the out-of-plane movement of the vortices. These switching events are then harnessed to demonstrate the non-destructive readout of the ferroelectric state. I then examine ferroelectricity along the axis of the vortices, revealing ferroelectric behavior remarkably different from the ferroelectricity referenced earlier. Despite the fact that both polarization components stem from the same vortex off-centering, I demonstrate the difference in their switching dynamics is due to unconstrained vs. constrained nucleation and growth. Furthermore, I demonstrate that the slower kinetics from constrained growth can be harnessed for a multi-state or analog ferroelectric memory device. Finally, I delve into how the vortex structure also displays structural chirality that is derived from the structure of an individual vortex alongside a secondary chirality originating from the vortex lattice configuration. Overall, my dissertation provides a new, in-depth understanding of the vortex system and demonstrates how their organization leads to emergent ferroelectricity, chirality, collective dynamics, and more.

This dissertation presents a systematic study on spin-orbit torque generation and manipulation in all oxide model systems consisting of ferromagnet La0.7Sr0.3MnO3, SrRuO3, multiferroic BiFeO3 and spin-orbit coupled SrIrO3. Efficient spin-orbit torque generation is crucial in the field of spintronics for the applications of spintronics-based memory and logic devices that requires the writing of magnetic states (i.e., switching the magnetization orientations). The development in efficient spin-orbit torque generation is staggered by the scarce of material candidates and the lack of extra degrees of freedom for spin current manipulation. The primary aim of this dissertation is to demonstrate how complex oxides can be utilized to generate spin currents efficiently and provide a viable pathway to control spin currents. In this dissertation, I will identify a complex oxide model system for spin current generation, with specific interest to investigate and understand the influence of electronic interaction enabled by the epitaxial interface on the spin-orbit torque generation. I will demonstrate that an epitaxial interface is able to facilitate spin current transmission and enable electronic interaction in the heterostructure; and the heterostructure interface, which is overlooked in the field of spintronics, is of vital importance for enhancing spin-orbit torque efficiency. Furthermore, the capability of BiFeO3 of transmitting spin current via antiferromagnet magnons is demonstrated with various techniques. Two pathways-modifying interface termination and using electric fields to control the ferroelectric polarization in BiFeO3, which in turn changes the spin current transmission, are demonstrated. Strikingly, bistable high/low spin transmission states are observed when reversing the ferroelectric polarization of BiFeO3, suggesting a coupling between spin propagation and ferroelectric polarization. In summary, the work presented in this dissertation reveals the important role that epitaxial interface plays in spin-orbit torque generation and demonstrates that multiferroic BiFeO3 is a good candidate for spin current transmission and manipulation. Additionally, the ferroelectric polarization controllable spin propagation in multiferroics offers an alternative to realize spintronics-based memory devices and the introduction of multiferroic materials to the field of spintronics opens up a new avenue for electrical control of spin currents.