Solution-processed optoelectronic materials offer a route to low cost photodetectors, large area solar cells, and integrated optical sources. While significant progress has been reported in organic and polymer spin-cast optoelectronics, colloidal quantum dots offer a distinct further advantage -- the convenient tuning of absorption onset via the quantum size effect. Electronic transport has recently been enhanced in size effect tuned colloidal quantum dot films using ligand exchange, resulting in ultrasensitive photodetectors in both visible and infrared wavelengths. Solid-film ligand exchange, however, generally results in rough film morphologies that are incompatible with high uniformity image sensors. Here, we report a new route to visible-wavelength spin-cast lead sulfide (PbS) nanocrystal photoconductive photodetectors with a sub 1% roughness, compared to the ~10% roughness obtained using previously reported approaches. The new procedure yields devices that exhibit 10 A/W responsivities and reveals an added significant advantage: when illumination conditions change, the photodetectors respond with a single time constant of 20 ms. This compares very favorably to the multi second and multi-time-constant response of previously reported PbS-nanocrystal photoconductive photodetectors.

Nanowires are sensitive to external influences such as surface charges or external electric fields. An Atomic Force Microsope (AFM) is modified to perform back gating and tip gating measurements in order to understand the interaction between an external field, and surface charge and nanowire conductance. A 2D finite element method (FEM) model is developed to simulate the measured conductance. The model shows that surface states play a critical role in determining nanowire conductance. A 3D FEM model is developed to examine the influence of the AFM tip on the lateral resolution of the AFM tip in the electrostatic measurement. The radius of the AFM tip determines the lateral resolution of the tip. However, carrier concentration in the nanowire establishes a lower limit on the lateral resolution, for small tip radii. These results enable one to optimize Scanning Probe Microscopy experiments as well as inform sample preparation for nanowire characterization.

Experiments and analysis in this thesis advance the understanding of critical issues in the carrier transport properties of InAs and InAs/GaAs core/shell heterostructure nanowires (diameter 30-60 nm) grown by molecular beam epitaxy. Effects of robust sub-band quantization structure on the gate-voltage dependence of conductance are observed up to 77 K in a single InAs nanowire with diameter 34\pm2 nm. Electronic field effect mobility at 300 K and 30 K are typically 2000-4000 cm^2V^-1s^-1 and 10000-20000 cm^2V^-1s^-1. Strain induced by lattice mismatch in epitaxial core/shell InAs/GaAs heterostructure nanowires is found to relax by formation of dislocations, correlated with nearly one order of magnitude suppression of room temperature field effect mobility compared with bare InAs nanowires. The carrier transport properties of Mn-doped ZnO nanowires were also investigated, where despite the large bandgap, conductivity is not thermally activated, and carrier mobility is consistent with strong degeneracy of the electron gas at 10 K. A novel method was developed providing the first experimental characterization of the quasi-equilibrium gate-voltage dependent surface potential in nanowire field-effect transistors, based on statistics of charging/discharging of a single Coulomb impurity evident in a random telegraph signal, which succeeds in nanostructures with tiny (attofarad) gate capacitance, where similar capacitance-voltage methods are challenging or impossible. We find that the evolution of channel potential with gate voltage is suppressed in the transistor's accumulation regime due to the screening effects of surface states with D_ss=1-2\times10^{12} cm^-2eV^-1. The gate voltage dependence of the random telegraph signals were used as a novel probe to spectroscopically study strong carrier reflection by single Coulomb impurities in nanowires. Reflection probabilities R=0.98-0.999 approach unity for an electron gas with density n=30-10 /micron in 30 nm diameter, 1 micron long InAs nanowires at 30 K. Results were compared with microscopic theory of electron scattering by Coulomb impurities in nanowires with dielectric confinement, i.e low dielectric constant surroundings. The latter, which is known to enhance the bare Coulomb interaction and excitonic binding energy, is an essential ingredient for the strong scattering in this regime, and in small diameter nanowires causes a breakdown in linear screening. Extending this, we show that InAs nanowires can operate is extremely sensitive charge sensors with sensitivity 60 micro eHz^-1/2 at high temperatures (200 K), a combination of characteristics that is not achieved by existing technology. Strong electrostatic coupling of a single charge to the conducting electron gas in the nanowire is enabled by miniaturization of nanowire diameter, operation in a regime of carrier density where the electronic screening length exceeds the nanowire diameter, and dielectric confinement. Finally, single ZnSe nanowire photodetectors are fabricated and studied. Peak responsivity at 2.0 V bias is 20 A/W at room temperature, similar to that of the best epitaxial ZnSe photodetectors. The high responsivity is due to a photoconductive gain g=500, the ratio of carrier lifetime to carrier transit time. The former is enhanced at room temperature due to rapid selective trapping of one species of excited carriers by surface states.

Thermoelectric modules are an important alternative to heat engines in the harvesting of waste heat. Electrical-thermal analogues are often employed when studying heat conduction and this analogue can be extended to develop an equivalent circuit for thermoelectric effects. For the primarily one-dimensional problem of thermoelectricity, the equations can be discretized to create a simple mathematical model. In this document, such a model is developed from first principles and show that the electro-thermal coupling is properly in- corporated. The results of simulations using the model are then presented and validated experimentally. Furthermore, in one possible application of thermoelectric modules, a self-contained cooling unit with an integrated thermoelectric generator is designed. By performing fluid dynamics simulations on a fan and heat sink model, the geometry and operating conditions can be optimized and the start-up and transient characteristics are studied.

Graphene is a single atomic layer of graphite and has many extraordinary properties. Many graphene based applications have been proposed in recent years and the need of a time domain simulation tool for studying graphene based devices emerges. This thesis focuses on developing a simulation framework for graphene based devices using finite-difference time-domain (FDTD) method. Formulation for a perfectly matched layer (PML) for the sub-cell FDTD method for thin dispersive layers has been derived and implemented. Such a PML is useful when thin layers extend to the boundaries of the computational domain. Using the sub-cell PML formulation to model the graphene thin layers significantly reduces the computational cost compared to using the conventional FDTD. The proposed formulation is accompanied by detailed validation and error analysis studies. Several graphene applications are simulated using the new framework and the results show good agreement with the respective analytical models.

We present FusionSim, a modeling framework capable of cycle-accurate simulation of a complete x86-based computer system with (a) a CPU and a GPU on the same die, and (b) a CPU and a GPU connected as separate components. We use FusionSim to characterize the performance of the Rodinia benchmarks on fused and discrete systems. We demonstrate that the speed-up due to fusion is highly correlated with the input data size. We demonstrate that for benchmarks that benefit most from fusion, a 9.72x speed-up is possible for small problem sizes. This speedup reduces to 1.84x with medium problem sizes. We study a software-managed coherence solution for the fused system. We find that it imposes a minor performance overhead of 2% for most benchmarks. Finally, we develop an analytical model for the performance benefit that is to be expected from fusion and show that FusionSim follows the predicted performance trend.

We present FusionSim, a modeling framework capable of cycle-accurate simulation of a complete x86-based computer system with (a) a CPU and a GPU on the same die, and (b) a CPU and a GPU connected as separate components. We use FusionSim to characterize the performance of the Rodinia benchmarks on fused and discrete systems. We demonstrate that the speed-up due to fusion is highly correlated with the input data size. We demonstrate that for benchmarks that benefit most from fusion, a 9.72x speed-up is possible for small problem sizes. This speedup reduces to 1.84x with medium problem sizes. We study a software-managed coherence solution for the fused system. We find that it imposes a minor performance overhead of 2% for most benchmarks. Finally, we develop an analytical model for the performance benefit that is to be expected from fusion and show that FusionSim follows the predicted performance trend.

Emotion states greatly influence many areas in our daily lives, such as: learning, decision making and interaction with others. Therefore, the ability to detect and recognize one’s emotional states is essential in intelligence Human Machine Interaction (HMI). In this thesis, a pattern classification framework was developed to sense and communicate emo- tion changes expressed by the Central Nervous System (CNS) through the use of EEG signals. More specifically, an EEG-based subject-dependent affect recognition system was developed to quantitatively measure and categorize three affect states: Positively excited, neutral and negatively excited. Several existing feature extraction algorithms and classifiers were researched, analyzed and evaluated through a series of classification simulations using a publicly available emotion-based EEG database. Simulation results were presented followed by an interpretation discussion. The findings in this thesis can be useful for the design of affect sensitive applications such as augmented means of communication for severely disabled people that cannot directly express their emotions. Furthermore, we have shown that with significantly reduced number of channels, classification rates maintained a level that is feasible for emotion recognition. Thus current HMI paradigms to integrate consumer electronics such as smart hand-held device with commercially available EEG headsets is promising and will significantly broaden the application cases.

Emotion states greatly influence many areas in our daily lives, such as: learning, decision making and interaction with others. Therefore, the ability to detect and recognize one’s emotional states is essential in intelligence Human Machine Interaction (HMI). In this thesis, a pattern classification framework was developed to sense and communicate emo- tion changes expressed by the Central Nervous System (CNS) through the use of EEG signals. More specifically, an EEG-based subject-dependent affect recognition system was developed to quantitatively measure and categorize three affect states: Positively excited, neutral and negatively excited. Several existing feature extraction algorithms and classifiers were researched, analyzed and evaluated through a series of classification simulations using a publicly available emotion-based EEG database. Simulation results were presented followed by an interpretation discussion. The findings in this thesis can be useful for the design of affect sensitive applications such as augmented means of communication for severely disabled people that cannot directly express their emotions. Furthermore, we have shown that with significantly reduced number of channels, classification rates maintained a level that is feasible for emotion recognition. Thus current HMI paradigms to integrate consumer electronics such as smart hand-held device with commercially available EEG headsets is promising and will significantly broaden the application cases.

This thesis introduces a new approach to compilation for FPGAs, which we call direct synthesis. We take a technology-mapped circuit netlist and directly map it into a pre-placed and routed FPGA overlay. Solving this problem may help to address the increasing portion of compile time that is attributed to placement and routing, and the tremendous amount of area and energy consumed by the highly flexible FPGA routing network. This thesis presents a direct synthesis algorithm and an algorithm for generating the pre-placed and routed FPGA overlays. Using the direct synthesis flow which we have designed, we can successfully map circuits less than 100 BLEs in size, after modest modi cations to the architecture of the FPGA overlay circuit. While we show that direct synthesis problem is challenging, further architectural modi cations are proposed which can allow the direct synthesis of larger circuits to succeed.