Resistive random-access memory (RRAM) or memristor, is an emerging technology used in neuromorphic computing to exceed the traditional von Neumann obstacle by merging the processing and memory units. Two-dimensional (2D) materials with non-volatile switching behavior can be used as the switching layer of RRAMs, exhibiting superior behavior compared to conventional oxide-based RRAMs. The use of 2D materials allows scaling the resistive switching layer thickness to sub-nanometer dimensions enabling devices to operate with low switching voltages and high programming speeds, offering large improvements in efficiency and performance as well as ultra-dense integration. This dissertation presents an extensive study of linear and logistic regression algorithms implemented with 1-transistor-1-resistor (1T1R) memristor crossbars arrays. For this task, a simulation platform is used that wraps circuit-level simulations of 1T1R crossbars and physics-based model of RRAM to elucidate the impact of device variability on algorithm accuracy, convergence rate, and precision. Moreover, a smart pulsing strategy is proposed for the practical implementation of synaptic weight updates that can accelerate training in real crossbar architectures. Next, this dissertation reports on the hardware implementation of analog dot-product operation on arrays of 2D hexagonal boron nitride (h-BN) memristors. This extends beyond previous work that studied isolated device characteristics towards the application of analog neural network accelerators based on 2D memristor arrays. The wafer-level fabrication of the memristor arrays is enabled by large-area transfer of CVD-grown few-layer h-BN films. The dot-product operation shows excellent linearity and repeatability, with low read energy consumption, with minimal error and deviation over various measurement cycles. Moreover, the successful implementation of a stochastic linear and logistic regression algorithm in 2D h-BN memristor hardware is presented for the classification of noisy images. Additionally, the electrical performance of novel 2D h-BN memristor for SNN applications is extensively investigated. Then, using the experimental behavior of the h-BN memristor as the artificial synapse, an unsupervised spiking neural network (SNN) is simulated for the image classification task. A novel and simple Spike-Timing-Dependent-Plasticity (STDP)-based dropout technique is presented to enhance the recognition task of the h-BN memristor-based SNN.

Linear bipolar circuits, designed with bipolar junction transistors (BJTs), are particularly vulnerable to the effects of space radiation. These circuits, which are usually commercial off-the-shelf (COTS) components, typically exhibit Enhanced Low Dose Rate Sensitivity (ELDRS), which is characterized by the enhancement of degradation when parts are exposed to radiation at low dose rates as compared to high dose rates. This phenomenon poses significant problems for the qualification of bipolar parts for use in low dose rate environments, such as most Earth orbits. ELDRS in BJTs has been well-documented in ground-based experiments; however, the effects of low dose rate irradiation on bipolar transistors manufactured in an integrated linear process had never been characterized in space - until the ELDRS experiment was launched in June 2019. The ELDRS instrument measures changes in the active collector and base currents in 24 lateral PNP (LPNP) BJTs on eight packaged die (three BJTs per die). Sixteen of the 24 BJTs are gated, while eight are standard, un-gated LPNPs. Device Under Test (DUT) and measurement variables include oxide thickness, passivation layer, packaging conditions, and gate voltage. This thesis reports the results obtained after more than 20 months of space flight in a highly elliptical Earth orbit. These results demonstrate that this category of bipolar devices is susceptible to low dose rate exposures and therefore exhibits the ELDRS effect in an actual space environment. This thesis also assess the impact of packaging variables on radiation response and examines one of the major causes behind radiation degradation, interface traps. An understanding of radiation effects in real space environments is critical for future missions that use these low-cost COTS bipolar technologies, making these results highly relevant for the satellite community.

Programmable Metallization Cell (PMC) devices are, in essence, redox-based

solid-state resistive switching devices that rely on ion transport through a solid electrolyte (SE) layer from anode to cathode. Analysis and modeling of the effect of different fabrication and processing parameter/conditions on PMC devices are crucial for future electronics. Furthermore, this work is even more significant for devices utilizing back-end- of-line (BEOL) compatible materials such as Cu, W, their oxides and SiOx as these devices offer cost effectiveness thanks to their inherent foundry-ready nature. In this dissertation, effect of annealing conditions and cathode material on the performance of Cu-SiOx vertical devices is investigated which shows that W-based devices have much lower forming voltage and initial resistance values. Also, higher annealing temperatures first lead to an increase in forming voltage from 400 °C to 500 °C, then a drastic decrease at 550 °C due to Cu island formation at the Cu/SiOx interface. Next, the characterization and modeling of the bilayer Cu2O/Cu-WO3 obtained by annealing the deposited Cu/WO3 stacks in air at BEOL-compatible temperatures is presented that display unique characteristics for lateral PMC devices. First, thin film oxidation kinetics of Cu is studied which show a parabolic relationship with annealing time and an activation energy of 0.70 eV. Grown Cu2O shows a cauliflower-like morphology where feature size on the surface increase with annealing time and temperature. Then, diffusion kinetics of Cu in WO3 is examined where the activation energy of diffusion of Cu into WO3 is calculated to be 0.74 eV. Cu was found to form clusters in the WO3 host which was revealed by imaging. Moreover, using the oxidation and diffusion analyses, a Matlab model is established for modeling the bilayer for process and annealing-condition optimization. The model is built to produce the resulting Cu2O thickness and Cu concentration in Cu-WO3. Additionally, material characterization, preliminary electrical results along with modeling of lateral PMC devices utilizing the bilayer is also demonstrated. By tuning the process parameters such as deposited Cu thickness and annealing conditions, a low-resistive Cu2O layer was achieved which dramatically enhanced the electrodeposition growth rate for lateral PMC devices.

ABSTRACT

Programmable metallization cell (PMC) technology uses the mechanism of metal ion transport in solid electrolytes and electrochemical redox reactions to form metallic electrodeposits. When a positive bias is applied from anode to cathode, atoms at the anode are oxidized to ions and dissolve in the solid electrolyte. They travel to the cathode under the influence of an electric field, where they are reduced to form electrodeposits. These electrodeposits are filamentary in nature and grow in different patterns. Devices that make use of the principle of filament growth have applications in memory, RF switching, and hardware security.

The solid electrolyte under investigation is tungsten trioxide with copper deposited on top. For a standard PMC, these layers are heated in a convection oven to dope the electrolyte. Once the heating process is completed, electrodes are deposited on top of the electrolyte and biased to grow the filaments. What is investigated is the rate of dendritic growth to applied field on the PMC and the composition of the electrolyte. Also investigated are modified three-terminal PMC capacitance change devices. These devices have a buried sensing electrode that senses the increasing capacitance as the filaments grow and increase the upper electrode area.

The rate of dendritic growth in the tungsten trioxide and copper electrolyte of different chemistries and applied field to the PMC devices is the important parameter. The rate of dendritic growth is related to the change of capacitance. Through sensing the change in capacitance over time the modified PMC device will function as an odometer device that can be attached to chips. The attachment of these devices to chips, help in preventing illegal recycling of old chips by marking those chips as old. This will prevent would-be attackers from inserting modified chips in systems that will enable them to by-pass any software security precautions.

This work investigates the effects of ionizing radiation and displacement damage on the retention of state, DC programming, and neuromorphic pulsed programming of Ag-Ge30Se70 conductive bridging random access memory (CBRAM) devices. The results show that CBRAM devices are susceptible to both environments. An observable degradation in electrical response due to total ionizing dose (TID) is shown during neuromorphic pulsed programming at TID below 1 Mrad using Cobalt-60. DC cycling in a 14 MeV neutron environment showed a collapse of the high resistance state (HRS) and low resistance state (LRS) programming window after a fluence of 4.9x10^{12} n/cm^2, demonstrating the CBRAM can fail in a displacement damage environment. Heavy ion exposure during retention testing and DC cycling, showed that failures to programming occurred at approximately the same threshold, indicating that the failure mechanism for the two types of tests may be the same. The dose received due to ionizing electronic interactions and non-ionizing kinetic interactions, was calculated for each ion species at the fluence of failure. TID values appear to be the most correlated, indicating that TID effects may be the dominate failure mechanism in a combined environment, though it is currently unclear as to how the displacement damage also contributes to the response. An analysis of material effects due to TID has indicated that radiation damage can limit the migration of Ag+ ions. The reduction in ion current density can explain several of the effects observed in CBRAM while in the LRS.

High-k dielectrics have been employed in the metal-oxide semiconductor field effect transistors (MOSFETs) since 45 nm technology node. In this MOSFET industry, Moore’s law projects the feature size of MOSFET scales half within every 18 months. Such scaling down theory has not only led to the physical limit of manufacturing but also raised the reliability issues in MOSFETs. After the incorporation of HfO2 based high-k dielectrics, the stacked oxides based gate insulator is facing rather challenging reliability issues due to the vulnerable HfO2 layer, ultra-thin interfacial SiO2 layer, and even messy interface between SiO2 and HfO2. Bias temperature instabilities (BTI), hot channel electrons injections (HCI), stress-induced leakage current (SILC), and time dependent dielectric breakdown (TDDB) are the four most prominent reliability challenges impacting the lifetime of the chips under use.

In order to fully understand the origins that could potentially challenge the reliability of the MOSFETs the defects induced aging and breakdown of the high-k dielectrics have been profoundly investigated here. BTI aging has been investigated to be related to charging effects from the bulk oxide traps and generations of Si-H bonds related interface traps. CVS and RVS induced dielectric breakdown studies have been performed and investigated. The breakdown process is regarded to be related to oxygen vacancies generations triggered by hot hole injections from anode. Post breakdown conduction study in the RRAM devices have shown irreversible characteristics of the dielectrics, although the resistance could be switched into high resistance state.

A modeling platform for predicting total ionizing dose (TID) and dose rate response of commercial commercial-off-the-shelf (COTS) linear bipolar circuits and technologies is introduced. Tasks associated with the modeling platform involve the development of model to predict the excess current response in a bipolar transistor given inputs of interface (NIT) and oxide defects (NOT) which are caused by ionizing radiation exposure. Existing models that attempt to predict this excess base current response are derived and discussed in detail. An improved model is proposed which modifies the existing model and incorporates the impact of charged interface trap defects on radiation-induced excess base current. The improved accuracy of the new model in predicting excess base current response in lateral PNP (LPNP) is then verified with Technology Computer Aided Design (TCAD) simulations. Finally, experimental data and compared with the improved and existing model calculations.

The Programmable Metallization Cell (PMC) is a novel solid-state resistive switching technology. It has a simple metal-insulator-metal “MIM” structure with one metal being electrochemically active (Cu) and the other one being inert (Pt or W), an insulating film (silica) acts as solid electrolyte for ion transport is sandwiched between these two electrodes. PMC’s resistance can be altered by an external electrical stimulus. The change of resistance is attributed to the formation or dissolution of Cu metal filament(s) within the silica layer which is associated with electrochemical redox reactions and ion transportation. In this dissertation, a comprehensive study of microfabrication method and its impacts on performance of PMC device is demonstrated, gamma-ray total ionizing dose (TID) impacts on device reliability is investigated, and the materials properties of doped/undoped silica switching layers are illuminated by impedance spectroscopy (IS). Due to the inherent CMOS compatibility, Cu-silica PMCs have great potential to be adopted in many emerging technologies, such as non-volatile storage cells and selector cells in ultra-dense 3D crosspoint memories, as well as electronic synapses in brain-inspired neuromorphic computing. Cu-silica PMC device performance for these applications is also assessed in this dissertation.

High speed image sensors are used as a diagnostic tool to analyze high speed processes for industrial, automotive, defense and biomedical application. The high fame rate of these sensors, capture a series of images that enables the viewer to understand and analyze the high speed phenomena. However, the pixel readout circuits designed for these sensors with a high frame rate (100fps to 1 Mfps) have a very low fill factor which are less than 58%. For high speed operation, the exposure time is less and (or) the light intensity incident on the image sensor is less. This makes it difficult for the sensor to detect faint light signals and gives a lower limit on the signal levels being detected by the sensor. Moreover, the leakage paths in the pixel readout circuit also sets a limit on the signal level being detected. Therefore, the fill factor of the pixel should be maximized and the leakage currents in the readout circuits should be minimized.

This thesis work presents the design of the pixel readout circuit suitable for high speed and low light imaging application. The circuit is an improvement to the 6T pixel readout architecture. The designed readout circuit minimizes the leakage currents in the circuit and detects light producing a signal level of 350µV at the cathode of the photodiode. A novel layout technique is used for the pixel, which improves the fill factor of the pixel to 64.625%. The read out circuit designed is an integral part of high speed image sensor, which is fabricated using a 0.18 µm CMOS technology with the die size of 3.1mm x 3.4 mm, the pixel size of 20µm x 20 µm, number of pixel of 96 x 96 and four 10-bit pipelined ADC’s. The image sensor achieves a high frame rate of 10508 fps and readout speed of 96 M pixels / sec.

As photons, electrons, and neutrons traverse a medium, they impart their energy in ways that are analytically difficult to describe. Monte Carlo methods provide valuable insight into understanding this behavior, especially when the radiation source or environment is too complex to simplify. This research investigates simulating various radiation sources using the Monte Carlo N-Particle (MCNP) transport code, characterizing their impact on various materials, and comparing the simulation results to general theory and measurements.

A total of five sources were of interest: two photon sources of different incident particle energies (3.83 eV and 1.25 MeV), two electron sources also of different energies (30 keV and 100 keV), and a californium-252 (Cf-252) spontaneous fission neutron source. Lateral and vertical programmable metallization cells (PMCs) were developed by other researchers for exposure to these photon and electron sources, so simplified PMC models were implemented in MCNP to estimate the doses and fluences. Dose rates measured around the neutron source and the predicted maximum activity of activation foils exposed to the neutrons were determined using MCNP and compared to experimental results obtained from gamma-ray spectroscopy.

The analytical fluence calculations for the photon and electron cases agreed with MCNP results, and differences are due to MCNP considering particle movements that hand calculations do not. Doses for the photon cases agreed between the analytical and simulated results, while the electron cases differed by a factor of up to 4.8. Physical dose rate measurements taken from the neutron source agreed with MCNP within the 10% tolerance of the measurement device. The activity results had a percent error of up to 50%, which suggests a need to further evaluate the spectroscopy setup.