Membrane separation is a field of both industrial and academic importance. Current technology is largely based on polymeric materials, and to a less extent other inorganic materials such as ceramics and metals. While developments in materials properties and membrane structures are constantly evolving, there are two challenges that need to be circumvented for better performance, i.e. the control over the pore structure and the chemical flexibility in modifying pore surface. “Bottom-up” approach to construct composite membranes using nanotubes in polymeric matrix is an effective route in fabricating membranes with well-defined architecture and tunable pore surface chemistry. This dissertation focuses on characterization and evaluation of cyclic peptide nanotubes (CPNs), a natural protein channel mimetic, in constructing sub-nanometer composite membranes with a cylinder-forming block copolymer (BCP) matrix in thin films. The fundamental understanding of the self-assembly of the CPNs from the building blocks establishes the foundation in utilizing the unique feature of CPNs to ensure precise structural control over the dimensions of the 1D nanotubes. The knowledge gained from the co-assembly of CPNs and BCP matrix in thin films allows further processing of the nanotubes to form well-aligned transport channels, establishing the guidelines in fabricating sub-nanometer porous membranes with and without surface chemistry modification. By identifying the key parameters in the membrane fabrication processes, design features for creating high-performance CPN based membranes can be determined and expanded. This indeed provides many exciting opportunities in developing new composite membranes with superior separation performances.

The self-assembly of cyclic peptide (CP) subunits forming high aspect ratio nanotubes is driven by strong intermolecular hydrogen bonding. To modulate and tune the growth of CPNs, polymers are conjugated to the exterior of the peptide subunits, resulting in the formation of polymer covered-CPNs (pc-CPNs). Due to the restriction of intermolecular hydrogen bonding, the conjugated polymer chains enter a confined space set by the hydrogen bonding distance. The entropic penalty associated with deforming the conjugated polymers serves as an opposing force destabilizes nanotube structure, while the enthalpic hydrogen bonding drives the nanotube formation. A delicate balance between the enthalpic driving force and the entropic destabilizing force enables one to modulate the growth of the nanotubes. Thus, the dimensions of the resultant pc-CPNs can be supervised simply by regulating the extent of the entropic penalty from the conjugated polymer chains.

In co-assembling CPNs and BCP matrix in thin films, both thermodynamic and kinetic parameters are critical to ensure homogeneous thin film morphology with well-aligned CPN channels at the center of the cylindrical microdomains of the BCP oriented normal to the substrate surface. The balance between the enthalpic interactions between the pc-CPNs and BCP and the entropic cost of polymer chain deformation gives rise to only one nanotube in the cylindrical microdomain. Due to the dynamic nature of CPN formation, preaggregation of the nanotubes causes defects of lay-down nanotubes at the membrane surface, hence compromising membrane quality and integrity. As a result, controlling the kinetic pathway of the co-assembly process is vital to fabricate high quality membranes for separation. Two simple approaches targeting two separate aggregation contributors have been developed to effectively prevent preaggregation of CPNs, resulting in high quality membranes suitable for molecular separation.

With the advancement in incorporating functional groups to the constituting peptide subunits, the interior surface of the CPNs can be further functionalized. Membranes have been fabricated using both the unmodified and modified CPNs, in which gas separation of CO2/CH4 mixture and hydronium ion transport were performed. In general, the incorporation of the CPNs improves the overall performance of the membranes, likely by providing additional pathways for the permeating molecules. Differences in the separation behaviors of the regular CPNs and the methyl-modified CPNs are observed for both gas separation and hydronium ion transport, where higher selectivity for CO2 over CH4 is seen for the methyl-modified CPNs. The local dipole interactions with CO2 molecules as well as the reduction in pore size are speculated to induce the differences in the performances of unmodified and unmodified CPNs.

These studies indeed establish the foundation in fabricating sub-nanometer porous membranes using self-assembled CPNs and BCP matrix in thin films. A delicate balance between the enthalpic and entropic contributions results in precise control over the structures of the nanotubes and the membranes. This unique “bottom-up” strategy demonstrates to be an effective platform in constructing new family of membranes for chemical separations.

In the 21st Century, with the increasing severity of the energy crisis and environmental pollution caused by the combustion of fossil fuels, human society is eager to find more sustainable and renewable resources. Over the past decades, many kinds of renewable energy, such as solar, wind, tidal, and geothermal, have been extensively used to replace fossil fuels. However, these renewable energies share drawbacks of discontinuity and instability. Thus, a system that can realize energy storage and conversion is strongly required. The regenerative electric power has been regarded as the most promising solution considering the reliability of energy supply, which is being widely researched.

Many energy storage systems, like batteries and electrochemical capacitors (ECs), have been hitherto achieved. However, due to the higher requirements of future systems ranging from portable electronics to hybrid electric vehicles and large industrial equipment, much more work still needs to be done to improve the power and energy density of energy storage systems. Thus, lithium-ion batteries (LIBs), characterized by high energy density and cycle stability, become the best possibility to make a breakthrough.

For improving the performance of lithium-ion batteries, the main challenge lies in the kinetics of the transport of electrons and ions, which dominates the rate performance of the devices, especially for high-loading batteries. In terms of the electrode design, massive efforts have been made to improve the transport kinetics, such as decreasing the size of the active materials to shorten the diffusion length of lithium-ions and adding conductive agents to improve the electronic conductivity. At the same time, the transport kinetics in electrolyte also plays an important role.

In lithium-ion batteries, the anions in the electrolytes generally do not participate in the lithiation reactions but exhibit higher mobility than lithium-ions, resulting in a low Li+ transference number (tLi+) of ~ 0.3. A low tLi+ gives rise to concentration polarization, reduces energy efficiency, and causes side reactions and joule heating, which shorten the cycling life, especially under fast charging/discharging condition. Although extensive efforts have been devoted to optimizing the composition of the electrolyte, simultaneously achieving high Li+ conductivity and Li+ transference number remains challenging.

In this dissertation, we designed and synthesized metal-organic frameworks (MOFs)-based electrolytes as anion-sorbent to increase the transference number and facilitate lithium-ion transport.

To find proper MOFs, three post-synthetic strategies focusing on metal cluster, ligand and pore structure, respectively, were used to prepare five typical MOFs. By heat treatment, the open metal sites (OMSs) in MOFs could be exposed to complex with the anions in solvent-filled pore channels, liberating Li+ mobility and affording high Li+ conductivity. With increased OMSs density, the transference number will increase while the electrochemical stability will decrease. By ligand modification and host-guest encapsulation, MOFs can provide free Li+ as anions, which present an increased transference number but inferior conductivity limited by the dissociation degree of the lithium-ions. Specifically, two MOFs structure, MOF-808 (tLi+ = 0.79) and UiO-66 (tLi+ = 0.62) are selected with high surface area and abundant OMSs, which can provide fast and efficient transport of lithium-ions.

For MOF-808, characterized for high OMSs density, a novel MOFs-based single-ion conducting electrolyte (SICE), 808-LiClO4, with both high transference number (0.77) and high conductivity (0.5 mS cm-1) was developed. An in-depth understanding of the mechanism of dehydration and anion adsorption process was explored to verify the significance of OMSs. This SICE can efficiently depress the generation of dendrite. A typical LFP|Li lithium metal battery (LMB) with 88% capacity retention after 400 cycles was achieved with 808-LiClO4 as electrolyte, which provides insight into the exploration of SICE for LMBs and next-generation battery devices.

For UiO-66, characterized for extremely electrochemical stability, a novel composite separators containing UiO-66 particles and PVA were fabricated by electrospinning process, producing non-woven fibrous mats with highly tunable pore size and structure. The electrospun separators show outstanding wettability and thermostability. Meanwhile, incorporating the MOF particles alleviates the decomposition of electrolytes, enhances the electrode reaction kinetics, and reduces the interface resistance between the electrolytes and electrodes. Primarily, they are the first reported separators that can increase ionic conductivity (from 0.7 mS cm–1 to 2.9 mS cm–1) and transference number (from 0.37 to 0.59) simultaneously. With the electrospun separators, the high-loading NCM|graphite cell (loading NCM: 20mg cm-2) with 73% capacity retention after 1000 cycles at high-rate (1C) was achieved.

Implementation of such MOFs-based electrolyte leads to dramatically improved power output and extended cycling lifetime, providing a new route towards better-performance lithium-ion batteries.

In this dissertation, two sections were included to present my work. The first section concerns about soft underlayers (SULs) in magnetic recording media, and the other section introduces electrostatic discharge (ESD) protection devices.

Hard disk drives (HDDs) are still being widely used in consumer electronics applications to store information. Due to the scaling down, higher capacity with smaller size HDDs are highly desired in market. Recently, perpendicular recording caused more attention to overcome some of the potential problems with longitudinal recording. The progress from the materials and techniques may make perpendicular recording more competitive. In this dissertation, a novel amorphous SUL CoFeTaZr is discussed for its superior magnetic properties.

ESD is one of the most important reliability problems for ultra-scaled integrated circuit (IC). Transient high voltage (up to tens of kilovolts) and current generated by ESD in a very short time could introduce very high electric field and current across semiconductor devices, which may result in dielectric damage or melting of semiconductors and contacts, which may affect circuit performance, shorten product life time and increase manufacturing and assembling cost. Therefore, it's necessary to design a protection circuit that discharges the ESD. As the scaling down continues, ESD protection design is facing severe challenges.

This dissertation presented two novel ESD protection structures. One is a new nano crossbar array ESD protection device consists of the SixOyNz composite, as dielectric layer between two metal electrodes. This device has extremely-low leakage current, fast response, good uniformity and robust ESD protection, ideal for mobile applications where power consumption is a great concern. The other one is a new non-volatile memory based ESD protection structure, two structures were introduced. One is with layers of nano crystal dots inside the floating gate film, the other one is with nitride floating gate layer in an MOSFET. Both structures has ultra-low leakage current, wide range of triggering voltage, ideal for multiple power domains.

In the last part, design and characterization of shallow trench isolation and poly-gated diode based ESD structures in a foundry 28 nm technology were reported. ESD-protected monitor circuits were used to evaluate chip-level ESD protection.

Short-term forecasts of heavy precipitation are critical for regional flood control operations, particularly in the Western U.S. where atmospheric rivers can be predicted reliably days in advance. However, spatial bias in these forecasts may reduce their utility for risk-averse system operations, where false negatives could be especially costly. Here we investigate whether deep learning methods can leverage spatial patterns in precipitation forecasts to (1) improve the skill of predicting heavy precipitation events in a target region at lead times from 1-14 days, and (2) balance the tradeoff between the rate of false negatives and false positives (misses and false alarms) by modifying the discrimination threshold of the classifiers. This approach is demonstrated for the Sacramento River Basin, California, using the Global Ensemble Forecast System (GEFS) v2 precipitation fields as input to convolutional neural networks. Results show that the deep learning models do not significantly improve the overall skill (F1 score) relative to the bias-corrected ensemble mean GEFS forecast. However, the models often correct missed predictions from GEFS by compensating for spatial biases at longer lead times. Additionally, the deep learning models provide the ability to adjust the rate of false negatives based on a desired level of risk aversion, with the tradeoff of increasing the false positive rate. Finally, analysis of the network activations (saliency) indicates spatial patterns consistent with physical understanding of atmospheric river events in this region, lending additional confidence in the ability of the model to support system operations.

Nature Products Research (NPR) has a long history of revealing bioactive constituents of natural origin, both as single drug leads within modern western medicine and as mixtures of bioactive constituents enriching traditional medicines. Identifying bioactive constituents in complex mixtures such as those obtained from extracting marine algae has been relying on multidisciplinary techniques, such as bioactivity-guided or spectroscopic-guided fractionation and purification. In this regard, milestones of scientific achievements of NPR have been hailed by applying novel technologies, such as improved separation or purification, spectroscopic hardware with detection limits of natural abundance, software algorithms for accelerating data collecting and processing, and high-throughput screening.

In most NPR, the characterization of novel compounds as well as the dereplication of known compounds entails the collection and analysis of NMR spectra. This involves the running of 1D and 2D NMR spectroscopic experiments for the purpose of partial structure construction, assemblage and relative stereochemistry determination. As exciting advancements in the rapid genetic and proteomic approaches have made their way into NPR, conventional NMR practices have become one of several bottlenecks in the characterization and dereplication of new compounds. In regard to this challenge, we leveraged the advantages of Non Uniform Sampling Nuclear Magnetic Resonance (NUS NMR) and Artificial Intelligence (AI) to create Small Molecule Accurate Recognition Technology (SMART) as a tool to speed up marine natural products discovery. Fast NMR techniques like NUS NMR have the potential to further reduce detection limits while maintaining the same sampling time and quality. Next, we applied over 4000 experimental Heteronuclear Single Quantum Correlation (HSQC) spectra for the AI training. The outcome is that the AI algorithm provided us with structurally insightful AI embedding maps with nodes and clusters representing correlations of related families of natural products. By testing different HSQC spectra using this algorithm, we can greatly accelerate the rate of known compound identification as well as rapidly generating hypotheses about the relationship of new molecules to those used for the training - based entirely on their NMR properties. Specifically, the 2D NMR spectra of a series of unknown compounds isolated from two different marine cyanobacteria were recognized by the SMART belonging to a specific class of marine depsipeptides.