| dc.description.abstract |
Oxide materials are widely used in current research due to their exceptional intrinsic
properties, such as ferroelectricity, magnetism, and superconductivity. However, the
strong bonding between oxide thin films and their rigid substrates limits flexibility,
restricting their use in applications needing adaptable materials. In the last decade,
advancements in freestanding membrane techniques have enabled the detachment of
oxide thin films from their bulk substrates, providing greater flexibility and allowing for
enhanced control over their properties. This thesis aims to leverage membrane techniques
to bypass the boundaries imposed by epitaxial growth.
In this thesis, first, extensive research has been conducted on the fabrication of
membranes from various materials, showing that the success of membrane transfer de pends on optimizing deposition conditions, buffer layer thickness, and buffer layer ma terial selection, which are crucial for advancing the technique. In this work, oxide films
are grown using pulsed laser deposition (PLD), with the growth monitored in-situ us ing reflection high-energy electron diffraction (RHEED), and their crystalline structures
characterized by X-ray diffraction (XRD).
The study then investigates the weak bonding between SrTiO3 and LaAlO3 mem branes and sapphire substrates. By heating the membranes to high temperatures—below
their melting points—thermally unstable oxide membranes self-assemble into crystalline
nanostructures. Scanning transmission electron microscopy (STEM) analysis confirms
that these nanostructures exhibit a highly crystalline nature, with well-defined facets
and uniform chemical distribution. This level of quality and precision surpasses what
can be achieved through conventional lithography techniques such as e-beam and optical
lithography.
The thesis then introduces the concept of the "vector substrate," which overcomes
conventional limitations of substrate design. In typical epitaxial growth, only the surface
area of the substrate (around 10 nm thick) directly influences the growth, while the bulk
substrate provides mundane properties such as mechanical support and thermal stability.
In contrast, the vector substrate concept uses an oxide membrane as a template noted
as a "vector" for subsequent growth, expanding the possible choice of substrates. This
is achieved by transferring oxide membranes from reusable substrates onto substrates of
choice. This approach is especially useful when conventional substrates are expensive
or hard to get, for example, bicrystal substrates. The feasibility of this technique is
demonstrated in this thesis by successfully fabricating Josephson junctions on bicrystal
vector substrate, created by transferring bicrystalline SrTiO3 membranes on sapphire
substrates.
A major challenge in traditional epitaxy is the requirement for structural symmetry
and lattice constant matching between the film and substrate, which severely limits ma terial combinations. This challenge is also overcome by utilizing membrane techniques
and post-transfer treatments. This work achieves atomically clean interfaces between
threefold symmetric sapphire and fourfold symmetric SrTiO3, which is epitaxially for bidden, featuring a novel moiré-type reconstruction. This paves the way for the creation
of novel heterostructures. These studies overcome the limitations of traditional epitaxy
by utilizing membrane techniques, unlocking new possibilities for material growth and
expanding their practical applications. The thesis concludes with a discussion of future
directions and potential applications of these innovative approaches. |
en |
