Critical Issues of Complex, Epitaxial Oxide Growth and Integration with Silicon by Molecular Beam Epitaxy

Open Access
- Author:
- Lettieri, James
- Graduate Program:
- Materials
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- July 23, 2002
- Committee Members:
- Xiaoxing Xi, Committee Member
Venkatraman Gopalan, Committee Member
Thomas E Mallouk, Committee Member
Darrell Schlom, Committee Chair/Co-Chair - Keywords:
- oxides
silicon
molecular beam epitaxy - Abstract:
- Molecular beam epitaxy was used to grow epitaxial oxides on silicon substrates using a systematic approach beginning with an understanding of system oxidation kinetics and thermodynamics and involving a strategy of moving from simple binary oxides to more complex structures. The growth of BaO, SrO, EuO, and SrTiO3 are discussed with a focus on the general theme of integration of functional, epitaxial oxides into a silicon environment. Oxidation studies of various metal systems relevant for oxide on silicon epitaxy and integration are reported. Experimental results demonstrate dramatically different oxidation behavior depending on elemental species and the catalytic nature of an alkaline earth metal at small doping concentrations to enable the full oxidation of the poorly oxidizing metals at oxygen pressures significantly lower than during deposition of the pure metal alone. Results from the deposition of Sr, Ba, Ti, La, Eu, Gd, and Al and codeposited combinations of the various elements are presented. The critical aspects of the growth of alkaline earth oxides on silicon are explained in detail. The step by step transition from the silicon to the alkaline earth oxide mediated by the formation of an interfacial silicide layer as described through reflection high energy electron diffraction (RHEED) is presented and used as a means to understand issues related to interface stability, oxidation, structural, and strain considerations for each stage of the growth. High quality, commensurate alkaline earth oxides (BaO, SrO, and (Ba,Sr)O) are grown on silicon at room temperature and PO2 background ~ 3 ¥ 10-8 Torr, taking advantage of the favorable oxidation kinetics of the alkaline earth metals and anomalously low temperatures necessary for epitaxial growth. The growth of alkaline earth oxide and rare earth earth oxide solid solutions and rare earth oxides (EuO) are described. RHEED data demonstrates the ability to create metastable solid solutions. The first reported epitaxial EuO (a silicon compatible ferromagnet) on silicon is reported, enabled by the use of a thin buffer layer (13 Å) of SrO. X-ray diffraction and RHEED data indicate single phase, single domain films with tremendous potential for spintronics applications. Using a strategy of transition from simple structures to the more complex, the growth of a perovskite (SrTiO3) on silicon is demonstated. Growth of a structurally optimized perovskite structure entails the transformation of a thin interfacial alkaline earth oxide layer into the initial perovskite cells. Such a transformation is facilitated by the strong tendency of the system to form the equilibrium structure under silicon friendly growth conditions (low temperature and low oxidant pressure). Optimized SrTiO3 and La-doped SrTiO3 on silicon are used to integrate a piezoelectric relevant for microelectromechanical systems (MEMS) applications and a ferroelectric relevant for a ferroelectric random access memory (FRAM) architecture. A d33 (piezoelectric coefficient) value of over 400 pm/V under bias is measured for the piezoelectric (Pb(Mn1/3Nb2/3)O3 – PbTiO3) and a remanent polarization of 25 mC/cm2 and fatigue free behavior (>1012 cycles) for a low temperature (450 °C) deposited ferroelectric (Pb(Zr,Ti)O3) is obtained. Initial work concerning the growth of even more complex structures such as conducting and ferroelectric superlattices are described. Short period superlattices of LaTiO3 and SrTiO3 are successfully grown on silicon with a high degree of crystalline quality. Future directions for more complex oxide integration and nonoxide integration are proposed.