Intellectual Merit: The two terminal nanoelectronic memristor devices offer an attractive solution for addressing the needs of high-density non-volatile data storage beyond the fundamental limits of Complementary Metal Oxide Semiconductor (CMOS) technology. The unique charge-flux characteristics of memristor devices offers applications in several key areas including high?density and high-performance digital data storage, digital computing, analog non-volatile memory, adaptive neural circuits, and energy-efficient massively parallel neuromorphic computing. This BRIGE project experimentally investigates memristive devices with Hafnium dioxide (HfO2) based switching layer and metal electrodes such as Tungsten (W), Nickel (Ni), Titanium Nitride (TiN), and Ruthenium (Ru). The research develops a fundamental knowledge about the charge transport mechanisms responsible for the switching behavior and how it depends on several process and device parameters such as the composition and thickness of the switching layer, electrodes, and oxygen vacancies concentration in the switching layer. Finally, the project seeks to demonstrate the application of these devices for high-density and high-performance digital and analog memory by systematically studying the switching speed, non-volatile resistive states, reliability, durability, and data-loss effects.

Broader Impacts: The project on memristive devices is expected to have significant impact in the areas of consumer electronic products, energy- efficient extreme-scale computing, and development of self-learning adaptive circuits. The research will be tightly integrated with education at graduate, undergraduate, and high-school levels. The graduate and undergraduate students will gain significant experience in the areas of fabrication and electrical characterization of nanoelectronic devices, which will prepare them to address the challenges of nanoelectronics in 21st century. A course in the area of emerging memory devices will be developed by integrating lectures with hands-on laboratory experience for graduate and undergraduate students. An educational module with live demonstrations to teach the basics of semiconductor devices will be developed for the high school students. Participation in engineering research will be broadened at all levels through activities including research experience for undergraduates, and outreach activities focused on increasing the participation of underrepresented groups in engineering through summer camp for the high school girls, lab experience for the high school girls, active mentoring, and recruitment of the underrepresented students.

Project Report

Nanoelectronic memristive-Resistive Random Access Memory (ReRAM) devices have gathered significant research attention as non-volatile memory devices that can be used for high-density data storage in future electronic products such as smart-phones, laptops, cameras, embedded microprocessors, and solid-state drives. When compared to the conventional memory devices such as DRAM, Flash, and SRAM, ReRAM devices offer potential for tremendous scalability of feature sizes and operation at ultra-low voltages. However, there are several challenges that need to be addressed to take the complete benefit of this enabling technology. This project focused on addressing these issues. Firstly, the project focused on understanding the mechanism of switching and charge transport in ReRAM devices. Device models were developed based on experimental device data that could be used in ReRAM based circuit simulations. Secondly, the project focused on studying the avenues to reduce the switching energies for these devices. A novel multi-step forming technique was investigated that allowed a controlled formation of filament in ReRAM devices by minimizing the current overshoot. Using this electroforming technique, the devices showed switching at lower power and demonstrated much higher endurance over conventional methods of electroforming. Thirdly, possibility of achieving multi-level-cell (MLC) in ReRAM devices were investigated. It was observed that MLC in ReRAM devices can be achieved by using different reset voltages. However, all states were not stable and reliable. In particular, the intermediate states were prone to fail over time under constant voltage stress. Fourth, new materials based ReRAM devices were investigated and process integration techniques to integrate these devices in crossbar matrix was developed. We observed that the interfacial layer and electrodes play an important role in governing the switching properties of ReRAM devices. Fifth, issues of sneak current in ReRAM based crossbar network was investigated via simulation based studies. Role of bidirectional diode reverse saturation current and junction capacitance was studied. In addition, role of parasitic capacitances in ReRAM based crossbar network was also studied. These studies indicated that while read access time for an individual ReRAM device may be very fast, the coupling capacitances can limit ReRAM read performance in high density crossbar networks. In addition, the transient current due to the charging of parasitic capacitances in ReRAM based high density crossbar network can be significantly high which can destabilize the filament in ReRAM devices and cause variability. To achieve education and broadening participation goals, major outcomes included establishing successful collaboration with Toledo Public Schools (TPS) and Saint Ursula Academy (girl’s high-school in Toledo, Ohio). A 4-6 weeks summer internship opportunities for four high-school students from minority community was created in PI’s lab. PI and her group participated in workshops such as Women in Science Day of Meeting (WISDOM), Excelling in Engineering (Excel), distance learning course to train high school teachers in rural Ohio, Latino Youth Summit of Ohio (LYSO). In addition, PI and her students developed several hand-on modules for high-school students to generate their interests and curiosity in science and engineering. The PI involved 5 undergraduate students in research on this project. Out of these students, 3 students have been women students in engineering.

Project Start
Project End
Budget Start
2011-10-01
Budget End
2014-09-30
Support Year
Fiscal Year
2011
Total Cost
$233,089
Indirect Cost
Name
University of Toledo
Department
Type
DUNS #
City
Toledo
State
OH
Country
United States
Zip Code
43606