The semiconductor industry is responsible for much of the world's extraordinary economic expansion over the past fifty years. Continuing the pace of growth and innovation will require new materials, such as crystalline perovskite films that are monolithically integrated with silicon, to be discovered, and efficient processes for their manufacture to be developed. These new material systems present an ideal platform to explore the fundamental materials physics and have numerous potential technology applications. This research develops a chemical route to the growth of these perovskite materials that will enable their insertion into technology applications in the commercial sector. The research combines an interdisciplinary team using a synergistic combination of epitaxial growth, ab initio theory, and in situ characterization as it develops a fundamental framework for growth of perovskite materials. The program partners university researchers with a technology leader in embedded semiconductors to broaden the student experience as they are exposed to problem definition that keeps the end goal of developing a viable technology front and center. The outreach program is aimed at attracting female high-school students to physical sciences and engineering; in collaboration with the physics instructors in local high schools, the students spend summers in research groups at the University of Texas at Austin and participate in "real science" in a supportive environment.
TECHNICAL DETAILS: Monolithically integrated hybrid oxide/semiconductor systems employing crystalline perovskite layers present an ideal platform to explore the fundamental materials physics determining electronic and magnetic properties of multiferroic heterostructures, and when integrated with semiconductors, potentially have applications in advanced electronics, hyperspectral sensors, and persistent surveillance and radar technologies. Discovery of materials and their fundamental properties will rely on molecular beam epitaxy (MBE) growth of limited volumes of materials; however, chemical routes, such as atomic layer deposition (ALD) need to be explored in parallel to enable lower cost manufacturing routes, growth over large area substrates, and potentially easier insertion of multifunctional oxide technology applications into the commercial sector. The research develops a fundamental framework for the growth of homo- and heteroepitaxial perovskite films that is built on theory and experimental validation, develops chemical routes centered on ALD, and explores the defect nature of the films and interfaces through spectroscopic and diffraction techniques. The focus is on ALD; MBE is used to complement the samples grown by ALD and to prepare surfaces upon which to initiate ALD or follow the detailed steps in layer growth during ALD. Studies explore growth of SrTiO3, LaAlO3 and (Ba,Sr)TiO3 films and heterostructures. Students use an integrated facility that permits in situ growth by MBE and ALD, and in situ characterization using scanning probes, and electron, X-ray and photon spectroscopies.
The semiconductor industry is responsible for much of the world’s extraordinary economic expansion over the past fifty years. Continuing the pace of growth and innovation will require discovery of new materials, such as crystalline oxide films that are integrated with silicon. This research project worked on both the discovery of the crystalline oxide films and new processes to manufacture these films. Potential applications include low power electronics, sensors, radar, and hydrogen production from water. The research involved a team of students predicting the properties of oxides and how they might form, and then using these predictions to guide the experiments into oxide film growth. The program partnered university researchers with a technology leader in semiconductors, GlobalFoundries, to broaden the student experience. Most microelectronics devices are grown on wafers of crystalline silicon that have a very specific arrangement of silicon atoms. For metal oxides integrated with silicon the oxide atoms must be organized into a single crystal layer that exactly matches with the atoms of the silicon on which the oxide is grown. The atoms in silicon form covalent bonds that share electrons between adjacent silicon atoms. While the atoms in an oxide form ionic bonds based on the electrostatic attraction between the positively charged metal ions and the negatively charged oxygen ions. Therefore, special steps are needed to transition from covalent silicon atoms to ionic metal oxide ions while also keeping all the atoms precisely where they need to be. A few metal oxides are known to perform this transition, such as the perovskite metal oxide strontium titanate (SrTiO3) that consists of layers of positively charged strontium ions bonded to oxygen ions and positively charged titanium ions bonded to oxygen ions. The positive and negative charges have to balance in the crystal and alternating layers of SrO and TiO2 form. Once the transition layer is grown on silicon, additional layers of strontium titanate or crystalline layers of different metal oxides can be grown on the template layer. The studies explored two ways to grow metal oxide films. For each growth method, atomic-scale control of thickness and composition is critical since the oxide films grown on silicon are typically around 10 nm thick. As a reference, the average width of a human hair is around 100,000 nm. The first growth method evaporated pure metal sources in a very controlled manner to regulate the precise delivery of the metal atoms needed for crystal growth. This approach is known as molecular beam epitaxy and it allowed the discovery of what films can be grown on silicon and on the template layer. The studies also explored the growth of metal oxide films using metalorganic molecules that react with the crystal surface. This alternative, chemical approach is known as atomic layer deposition and it promises a scalable manufacturing route to the integrated metal oxide films. Chemical routes to the growth of single crystal strontium titanate, lanthanum aluminate, cobalt-doped strontium titanate, and heterostructures of cobalt oxide and titanium dioxide on strontium titanate were discovered. For each case, the films were grown on silicon that was capped with a template layer of crystalline strontium titanate, which was grown using molecular beam epitaxy. Importantly, the research has revealed a minimum of four unit cells (approximately 1.5 nm) of strontium titanate are required to enable crystal growth during atomic layer deposition. The research has also shown single crystal metal oxide layers can be integrated with silicon without the formation of an insulating silicon-oxide interlayer between the silicon surface and the strontium titanate transition layer because atomic layer deposition is performed at low temperatures.
