The goal of this project is to determine the characteristics of silicon dioxide thin films, deposited on silicon substrates using PECVD (plasma enhanced chemical vapor deposition) techniques (a method used to deposit thin films onto the substrate), and to relate those characteristics to the deposition parameters and to the film's effectiveness as a dielectric in a metal-insulator-metal capacitor (MIMCAP) structure. MIMCAPs are an important electrical element for integrated circuits (IC), Microelectromechanical Systems (MEMS) (i.e. sensors), and high-performance circuit boards. Throughout the duration of the project, process parameters are varied and the effects of those changes on the resultant film properties are noted. By noticing the general thin film trends as a function of process parameter, thin film characteristics are tuned to the desired levels. A combination of ellipsometry and profilometry are used to characterize the deposited thin films. These techniques allow us to quantify a variety of process and resulting film properties, including deposition rate, internal stress, and thickness uniformity. The wet etch rate (the rate of film etching in a specified chemical solution) of our films is measured by observing the film thickness before and after immersion in a buffered oxide etchant bath for a fixed time period. The primary thin film recipes resulting from this project will include a low deposition rate and a high deposition rate oxide recipe for use at either 125°C or at 350°C. The recipes generated as a result of this project will be available to WNF users and will be utilized for projects and applications outside the scope of MIMCAP structures as well.

Radio frequency integrated circuits (RFIC) are major components in enabling wireless communication. The RF inductor is integral to these applications but has traditionally suffered from low quality factors (i.e dissipative losses) and thus has had lower performance than off-chip discrete inductor components. Adding permalloy (80% Ni, 20% Fe), a soft-magnetic material, in integrated inductors enhances inductance and results in a higher quality factor, with frequencies exceeding 1GHz possible than is currently achievable with standard silicon processing. This makes it possible to use on-chip inductors for higher performance in RFICs such as cell phone RF transceivers that may otherwise require off-chip discrete inductors. Fabricating integrated inductors provides major advantages in that they would reduce component area and power consumptions of RFICs. In this work, integrated inductors are fabricated using a silicon-based integrated passive device technology that increases the copper interconnect thickness thereby dramatically reducing the parasitic resistance losses experienced when making inductors using semiconductor microfabrication processes. The inductor is formed by using multilayered thick resist patterned by ultraviolet light to form a plating mold. The mold is filled via copper electroplating and intentionally overplated beyond the top of the mold. Excess copper is then polished back. This process yields very low surface roughness of the copper and maintains the critical dimensions of the electrical traces, both of which are critical to RF circuit performance.  Additional process work involves electroplating thin film permalloy and evaluating its impact on alloy composition and RF circuit performance. To overcome eddy current losses and the skin depth effect, permalloy will be patterned in strips and dots with varying thicknesses. Dynamic testing with poled and unpoled electroplated permalloy will also be conducted to determine if these inductors can be integrated with complementary metal-oxide semiconductor (CMOS) compatible circuits and maintain RFIC performance.

Fires cause thousands of deaths and billions of dollars in damage each year in the US alone; CERBERUS (Community Extinguishing Robot for the Brisk and Expedited Restoration of Unexcited Space) is an autonomous firefighting robot designed to reduce human interaction with fires and significantly reducing transit time. CERBERUS does this by functioning without any human interaction and being located inside a building before the fire even starts. CERBERUS navigates a single floor of a building and attempts to douse the fire using an on-board portable fire extinguisher. After extinguishant release CERBERUS remains in position to aid firefighters in locating the source of the fire. Visual and audio alarms allow for building occupants to be alerted to CERBERUS’s presence, and are the main system in place for firefighters to locate CERBERUS. Planned incremental improvements include multi-level functionality or opening of blocked passageways.

Traditional imaging methods fall flat when dealing with objects embedded in heavily scattering media because information necessary to image formation is lost in the scattered imaging signal. The nitrogen-vacancy (NV) center in fluorescent nanodiamonds (FNDs) is one promising system to realize high-resolution imaging through heavily scattering media. A novel high-resolution imaging method that is largely impervious to scattering difficulties can be achieved by exploiting the magnetic-field dependence of the NV center’s fluorescence. A single NV center is a magnetic quantum system with spin energy level splitting that depends on the value of the magnetic field at the location of the NV center. Using optically detected magnetic resonance (ODMR), we can determine the splitting between two spin energy levels and thus the value of the magnetic field at the FND. It follows that if FNDs are placed in a magnetic field whose value is known at all points, the FND positions can be determined through application of ODMR, potentially allowing for spatial resolution in the nanoscale. In this work, we characterize the optical and spin properties of FNDs created with a very high density of NV centers in order to determine their suitability for diffusive imaging. Optical brightness, NV center density, and the sensitivity of the fluorescence dependence on magnetic field are reported. Using these properties, we estimate the expected performance of a diffusive imaging microscope based on the FNDs.

Targeted reinnervation is a surgical technique that instills a sense of touch in an amputee's non-existent, "phantom" limb. This allows a person living with amputation to feel sensations such as pressure through their prosthetic device, offering intuitive and high performance movement. However, current methods to map the vast cutaneous information at the targeted sites are manually conducted, and thus are imprecise and tedious. This work offers an automated method to robotically map these sensations, removing the potential bias and inaccuracies from manual mapping. The presented work is a package of scripts that detect target sites and commands a surgical robot, Raven-II, to methodically explore the thresholds and locations of sensations at the target sites. Using a custom-made monofilament attachment to the Raven-II, robot trajectory testing was conducted with randomized target locations (including sham trials) on varying curvatures. Future tests include testing with human subjects to compare precision with manual sensory mapping. Quantifying and providing high resolution sensory input is a remarkable contribution. Even a single point of cutaneous feedback allows a person to intuitively confirm grasp of an object, and adjust pressure accordingly. This automated mapping process is truly exciting, as it enables improved multi-disciplinary methods in systems with advanced signal processing--not only for prosthetics.

In synthetic biology, developing an accurate means of controlling the transcription of several genes at once is critical for creating programmable biological systems. Our research is focused on using guide RNAs (gRNA) in a CRISPR-based system as transcription factors to control and regulate gene expression. Using these components, we seek to develop a novel process by which circuits involving multiple genes may be precisely tuned. The bacterial CRISPR mechanism recognizes invading viral DNA using complementary gRNA and cleaves it with the Cas9 endonuclease protein. By engineering a gRNA to be complementary to a specific DNA sequence, and using a nuclease-deficient Cas9, we can target specific loci in any given genome. Our method capitalizes on the specificity and simplicity of gRNAs and utilizes them as transcription factors in S. cerevisiae. In gene circuits based on CRISPR, the quantity of a gRNA reflects the expression of the corresponding circuit component. The ability to measure the quantity of gRNAs is an effective method of testing the efficacy of a gRNA-based genetic circuit, so we have focused on gRNA quantification. By isolating RNA, creating cDNA via reverse transcription, and measuring target cDNAs using real-time quantitative PCR, we are able to calculate the original concentration of gRNAs. This information is then used to direct the design of our gRNA-based circuits. Progress towards development of a new, simple and specific means of tuning genetic circuits may open a broad range of possibilities for engineering complex yet predictable circuit behavior, and provide a greater degree of programmability to genetic systems.

Saccharomyces cerevisiae haploids can mate to form diploid cells. These cells contain the genomes of both parent haploids. Therefore, engineering haploids with different auxotrophic or fluorescent markers and mating them produces a diploid which contains both of the haploids' respective markers. This can allow for the differentiation of haploid and diploid cells. The yeast mating pathway has been proposed for a variety of applications including diagnostics. Production of a rapid visual output for yeast mating would benefit this tool. The purpose of this project is to visualize a yeast mating without use of specialized equipment or waiting for colonies to form on a plate. To achieve this, each haploid mating type, MATa and MATα, is engineered with sections of a colorimetric reporter such that individually they do not give any output but upon mating and complementation of the full reporter system the diploid alone produces a visual output. Reporters used include the enzymatic reporter gene lacZ from Escherichia coli, the red pigment forming adenine biosynthetic pathway native to S. cerevisiae, and the purple pigment forming violacein biosynthetic pathway from Chromobacterium violaceum. 

Due to the uncertainty and variability of renewable energy resources, energy storage technologies (ES) have become an increasingly important option to enhance system flexibility and, thus, maintain its reliability. However, realizing the benefits of ES is highly dependent on the system specific environment in which the technology is introduced and the manner in which the device is operated. This work focused on studying how ES can be integrated with the existing system, manipulating equations to represent the operation of ES on the system, and using General Algebraic Modeling Software (GAMS) to simulate the optimal operation of ES in power systems. Building upon a unit-commitment generation model, this work presents simulations on a test system that included large wind generation, transmission constraints, and an optimal deployment of ES devices. The goal was to validate the results of the specific ES deployment by simulating the system with real-time wind data as opposed to the daily forecasted amount. By controlling the amounts of dispatchable generation and ES, simulating the system with real-time wind data reflects the actual power balance decisions that occur after the daily commitment schedules have been optimized. The impact of the ES deployment was determined by comparing these simulation results to those when no storage was on the system.  The results confirmed that increasing the available use of ES devices in real-time power balancing can provide for corresponding decreases in wind curtailment and system operation costs.

Virtual Reality (VR) is a computer-simulated environment that can simulate physical presence, often through sight, in places in the real world or in the virtual world. Within virtual reality, dizziness is common due to the delayed response of graphics that appear on the screen. This delayed response is due to the slow speed of rendering of the graphics, and the layers of software/hardware these images have to go through to finally appear on the screen. Our research project focused on rendering and generating these graphics for virtual reality applications at the bare-metal with a real-time constraint to speed up the entire process thereby decreasing the effects of dizziness in the virtual world. The open-source Graphical Processing Unit (GPU) in the Raspberry Pi allowed us to achieve the maximum capabilities to optimize certain aspects of the project to determine the best possible route to a real-time graphics library for virtual reality applications.

The purpose of this research is to develop high quality, high-frequency passive components for use in filtering circuits for radio frequency communication systems. An example of this would be circuits used in communications satellites. For these critical applications, the performance and robustness of the system is paramount. To address these needs, traditional circuit boards are not suitable so high-performance silicon-based interposers for communicating between discreet microchips are being developed. This project focuses on the development of a dual-damascene processes for making high quality interconnect networks using stepper lithography, oxide etching, pulsed chemical vapor deposition (pulsed-CVD), electroplating, and chemical mechanical polishing (CMP). The first layer is comprised of a deposited silicon dioxide layer that is patterned and etched to create the pattern of the wiring interconnect and passive devices like capacitors, inductors, and resistors. Next, a tantalum nitride diffusion barrier and a copper seed layer are deposited using pulsed PECVD. The patterns are filled using Cu electroplating, and the excess material is polished to create a flat uniform surface. This process, pioneered by IBM, is called a damascene process because it is analogous to inlaid metal artwork techniques used in the Middle Ages that originated or were mastered in Damascus. Next, another layer of oxide is deposited which serves as a dielectric and etched in order to create a vertical interconnect (via) between two conductive layers. Afterwards, the procedure is repeated to create the via and second copper layer – a process that is referred to as dual damascene because both layers are formed simultaneously. Using these techniques, we will fabricate passive devices that will be tested and used to create a Process Design Kit (PDK) for the Washington Nanofabrication Facility (WNF) with eventual development of more advanced high frequency passive components for research applications and a potential foundry process flow.

Microorganisms outnumber human cells ten to one in the average person. This microbial community interacts extensively with the human body, with both positive and negative results. My research could allow for a more targeted approach to bacterial infections rather than the commonly used general antibiotics, which eliminate both the beneficial and harmful bacteria in the body, which has been proven to cause deleterious health effects. I have been working on achieving targeted infection of the bacterium Escherichia coli by the virus M13 bacteriophage. This would allow for the virus to infect only very specific bacteria and deliver DNA encoded messages without affecting any other cells. We plan to achieve this by attaching peptides to the coat of the M13 bacteriophage and attaching a complementary protein to the target E. coli’s F-pilus, a projection on the surface of bacteria. Using a virus as a delivery mechanism allows for variability in the genetic message, such as a variety of anti-microbial proteins. The applications of this research lie in many fields, including medicine, with its applications in antibiotics and bioengineering to build more complex cellular systems.

Imposed Dynamo Current Drive (IDCD) is a theoretical and well-grounded mechanism for circulating a confined and sustained Spheromak (smoke-ring shaped) plasma. Current drive mechanisms in general have proven crucial in most attempts at controlled fusion, but existing techniques—such as radio-frequency current drive or neutral beam injection—are limited in their effectiveness; additionally, most research focuses on Tokamak reactor configurations. Imposed Dynamo Current Drive grew from previous successes in Steady Inductive Helicity Injection (SIHI), refined under the HIT-SI group at the University of Washington. HIT-SI tests formation and sustainment of a Spheromak in both Deuterium (heavy Hydrogen) and Helium plasmas, and in this context Imposed Dynamo methods were first considered for driving current without disrupting confinement. IDCD posits that perturbations along otherwise closed magnetic flux surfaces within the plasma-confining volume can induce a plasma current through electron drag, and was first theorized in 2012 with additional refinements in 2014 to account for a previously unnoticed SIHI frequency dependence. This study aims to validate and refine the basis of IDCD through comparison of measured electromagnetic impedance across helical magnetic field injectors ('helicity injectors') to predicted and simulated behaviors. Past work validated the overall profile and frequency dependence of IDCD predictions, as well as demonstrated that impedance across a three-injector configuration is lower than the impedance across a two-injector configuration. Present work focuses on distilling information from finite element simulations to be compared with IDCD predictions, and on addressing whether or not relative phase differences between helicity injectors contributes to overall impedance behavior. Future work aims to synthesize the results considered here in a revision of Imposed Dynamo theory by more accurately determining empirical fitting constants and establishing definitively whether or not additional dependencies, such as phase difference, contribute to overall impedance profiles.

Our lab is investigating the possibility of a magnetic dipole orientation sensor that is compatible with biological samples. Utilizing a thin near-surface layer of nitrogen-vacancy centers in diamond, we are able to optically image small magnetic field disturbances. This has allowed us to detect and locate magnetic nanoparticles on the diamond surface under ambient conditions. In this work, we examine the feasibility of using a similar sensor to detect the magnetic orientation of the particles, in addition to their location. I have developed software to simulate the imaging system, taking into account factors such as the size and material of the particles, as well as the optical properties of the sensing surface. I will present simulation results indicating that orientation measurements should be possible, as well as initial experimental results for 50 nm and 500 nm magnetic particles that were deposited on a diamond surface. The ability to detect the orientation of biologically compatible magnetic nanoparticles will likely be useful in the study of a wide variety of microbiological processes, enabling the tracking of complex motion on the nano-scale.

The optical spintronics and sensing lab researches the fundamental properties of defects in semiconductors for quantum-enabled technologies. The lab is extending its measurement capabilities to the wide bandgap material, ZnO, which requires a near ultraviolet laser source. Here we will use second harmonic generation (SHG), a common type of nonlinear optical interaction, to realize a laser at 369 nm from incident laser light at 738 nm, a wavelength readily available in the laboratory. The SHG laser source consists of a birefringent crystal chosen from BBO (Beta Barium Borate), BIBO (Bismuth Borate), and LBO (Lithium Triborate), all of which are common non-linear optical materials that are often used for SHG. Design parameters include the crystal’s nonlinear susceptibility, its thickness, and cut angle. The cut angle is determined by phase matching between the fundamental wavelength and second harmonic to optimize the SHG efficiency. Birefringence exhibited in these crystals allows us to achieve this phase matching because we are able to find a propagation direction through which both wavelengths experience a same refractive index. The crystal design and efforts toward experimentally realizing the SHG source will be presented.

An automated wave plate rotator (AWPR) is an optical device that allows a user to remotely and precisely control the polarization of a laser beam. AWPR's are used in optical research studies of materials by providing optical intensity or polarization control. Commercial AWPR's are expensive and here I present the design and implementation of a low-cost alternative that utilizes a programmable microcontroller, stepper motor, and a few electronic components. The AWPR will be remotely controlled via Matlab and will include features such as home detection and position feedback. Results on position precision and accuracy will be reported.

The nitrogen-vacancy (NV) center is a defect in diamond that has promising applications in quantum information processing based on quantum entanglement. A critical requirement for quantum entanglement is that separate NV centers have identical optical properties. Here we present the design and implementation of an optical characterization system for NV centers called photoluminescence excitation (PLE) spectroscopy. While scanning a NV center with light from a tunable-frequency red laser, we detect and analyze the intensity of the resulting photoluminescence. The suitability of this system for characterizing NV centers, resonantly exciting the centers, and providing feedback to stabilize the NV optical properties will be discussed.