Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Document Type

Campus-Only Access for One (1) Year

Degree Name

Doctor of Philosophy (PhD)

Degree Program

Mechanical Engineering

Year Degree Awarded


Month Degree Awarded


First Advisor

Stephen S. Nonnenmann

Subject Categories

Materials Science and Engineering | Nanoscience and Nanotechnology | Semiconductor and Optical Materials


Novel nonvolatile memory technologies garner intense research interest as conventional

ash devices approach their physical limit. Memristors, often comprising an

insulating thin lm between two metal electrodes to constitute a class of two-terminal

devices, enable a variety of important large data storage and data-driven computing

applications. In addition to nonvolatile behavior, other features such as high scalability,

low power consumption, and sub-nanosecond response times make memristors

among the most attractive candidate systems. Their strength in electronic storage

relies on the unique properties of the tunable variations in resistance induced from

the accumulation of charged defects based on the applied bias history.

Metal oxides serve as the most common \storage" materials, demonstrating advantages

including simple fabrication, high reliability, and fast operation speeds. While

the basic working concepts and the underlying conduction mechanisms have been

established through combined experimental and simulation studies, the role of metalinsulator

interface, which acts as the crux of coupled electronic-ionic interactions,

has not been fully understood. Continuous scaling, for the purpose of high density

memories, also requires a detailed understanding of the switching behavior and transport

mechanism. Other technical challenges include the development of innovative,

low-cost fabrication methods that e ectively enable high-performance structures as

an alternative to complicated process modules. Stable retention and endurance of

the switching characteristics, as well as uniformity of the switching parameters to

ensure a valid program/read operation also represent signi cant challenges. Studies

in device and materials optimization remain in the formative stages, and thus motivate

this work to drive progress in the most attractive areas, including size dependent

behavior and switching performance of memristors.

This collection of work aims to correlate resistive switching within metal oxide

based memristors with the fundamental physical mechanisms and material properties

on a highly localized scale. Chapter 3 relates the device size and the resulting performance

matrix of memory cells in the rst step towards fully understanding the scaling

projection and reliability issues that a ect nanoscale architectures. Chapter 4 demonstrates

a convective self-assembly, transferable approach that enables the fabrication

of highly-controlled nanoribbon comprising solution-processed nanocrystals, providing

multiple degrees of freedom for understanding the interfacial memristive behavior

of functional oxide nanostructures. As a powerful tool in the study of resistive switching,

conductive AFM probes the homogeneity of the charge transport properties, thus

o ering electrical information by locally applied bias when it is placed in direct contact

with desired regime. Finally we also focus on the improving the cycle-to-cycle uniformity

by embedding nanostructure into conventional metal-insulator-metal (MIM)

geometry in Chapter 5. This improvement is attributed to the concentration of electric

eld when metal nanoislands are inserted into the oxide lm matrix. The details

of this work will highlight the tunable and optimizable template-driven method that

can be applied on any memristive systems, yielding a superior uniformity of operating

voltage and resistance states.

In summary, this thesis promotes the development of novel, high-performance

metal oxide based memristors enabled by the availability of new, nanostructured materials

and innovations in device structure engineering. The switching performance,

underlying mechanisms, area/defect concentration e ects, development of solutionprocessed

nanocrystals assemblies and chemistries, and highly enhanced uniformity

in memristors are addressed by combining systematic deposition approaches with the

advanced nanoscopic observation of the conducting lament, leading to the strongest

competitor among future nonvolatile memory solution.

Available for download on Sunday, September 01, 2019