Novel chlorine-based chemistry and implementation hardware for the growth of lithium niobate and related complex metal oxides

Interest in oxide related research has increased as standard oxides reach their operational limits and new classes of devices are imagined that can only be realized through the use of man-made compounds. Many of these devices require high quality films in order to reach their highest potential. Molecular beam epitaxy (MBE), a known leader for film quality in traditional semiconductors, is poised to become a key producer of high quality oxides if system limitations can be overcome. The concepts presented herein provide the groundwork for leaping beyond those limitations and placing MBE squarely in the center of global research and production of complex oxides. One of the most promising oxides is lithium niobate, LiNbO3, which can potentially deliver novel electronic, optic, and hybrid devices not currently possible.;Growing lithium niobate and other complex oxides using MBE is difficult. The high temperature refractory metals required, precision flux control, and corrosive and oxidizing environments are not the traditional domain of MBE. However, these are not insurmountable difficulties. The next several chapters describe novel chemistries and technologies that can overcome these difficulties.;First, refractory metals that normally require high temperature sources such as electron beam evaporators can be delivered to the substrate through a novel use of low temperature chloride compounds such as niobium (V) chloride. This chloride chemistry allows standard and low temperature sources such as standard effusion cells to deliver refractory and other high temperature materials to the substrate. The need for high temperature sources, especially those that are incompatible with corrosive environments such as electron beam evaporators, is eliminated.;Second, a precision, vapor-phase source and control system is prototyped for these chloride compounds which can achieve improved flux accuracy and allow for greater expandability of a standard MBE to support many sources. The chloride sources use high vapor pressure materials to achieve their low temperature operation. However, this makes them sensitive to temperature changes and leads to flux drift. The vapor-phase source removes the material from inside the main MBE chamber to a remote reservoir to eliminate thermal drifts. The source is controlled solely by a valved mass flow controller and injected into the chamber via a single nozzle. The design uses a manifold at the nozzle which permits attaching multiple sources to the single nozzle saving other main chamber ports.;Third, a novel method of measuring flux with spontaneous ionization current has been developed. This design utilizes a low noise design to measure femtoamp currents generated as an evaporant spontaneously ionizes. The fraction of atoms that ionize can be predicted from the physical properties of the material. Combining the measured current with this predicted fraction of ions has the potential for directly counting the atoms evaporated. Though initial tests show asymptotic behavior at low temperatures for some materials, the currents are still measurable for a range of evaporation rates. The current measurement can be used to directly control evaporation from an effusion cell. The design is sensitive enough to detect outgassing of the cell and cell "spitting" or other non-idealities. Monitoring these non-idealities can help improve other processes by ensuring the cell is fully outgassed and stable. Interest in this design has spawned patents and licensing negotiations for commercial production.;Finally, a miniaturized RF induction cell prototype is shown that can eliminate the need for incandescent filaments in an oxide based MBE. The RF cell has the potential to increase reliability of MBEs for oxide work and achieve higher operating temperatures without the need for densely wound incandescent filaments or electron beam sources.