This thesis is a sandwich thesis composed of three papers that are published in refereed journals or conferences. The first paper is a systematic experimental study conducted to identify the first phase to form during cobalt germanidation. Hexagonal β-Co5Ge3 was the first phase to form at temperatures as low as 227°C followed by monoclinic CoGe as the second phase at the same temperature. We also report for the first time that both phases that formed were highly ordered partial epitaxial crystal orientations suggesting that both of those low-temperature phases could potentially serve as high quality contacts for germanium based devices with a very low thermal budget which is advantageous for the process design. Those results contributed to a better understanding of cobalt germanidation leading to the first multiphase technology computer aided design model presented in the second paper. This kinetic model for cobalt germanide growth can predict the resulting phase based on anneal time, temperature, and ambient. The model has been calibrated to experimental results. This predictive model can help in the design of cobalt germanide contacts with low resistance and can serve as a general modeling framework for multiphase solid state reaction binary systems. A comprehensive survey of the experimental results for formation of cobalt germanides is discussed and the data are reconciled in the third paper. Factors affecting the resulting phases and their quality are identified and some optimum choices for the experimental parameters are pointed based on the survey. The role of germanium crystal orientation in ohmic and Schottky properties of the contact is analyzed. Fermi level pinning plays a role mainly on metal/(100) n-type Ge interfaces and its role is minimal on p-type Ge and other crystalline orientations. Schottky Barrier Heights for cobalt germanide contacts reported in the literature are surveyed. Crystalline cobalt germanides, forming when Co is deposited at high temperatures, are expected to have lower interface resistivities compared to those reported. The work is important because contact resistance has become one of the most important factors in advanced complementary metal oxide semiconductor (CMOS) technology and advanced devices already include germanium (Ge) in the source/drain regions of devices. It is also important because heating at the interface due to contact resistance is one of the key challenges in power devices and cobalt germanide can be used both for Si and Ge based devices as well as for gallium nitride (GaN) devices. The latter application is possible because cobalt germanide is lattice-matched to GaN. / Thesis / Doctor of Philosophy (PhD) / The main goal of this thesis is to create predictive empirical, mathematical, and physical models to help the designer of the semiconductor process technology to design high quality electric contacts, namely cobalt germanides, to their semiconductor devices, germanium based. The choice of cobalt germanides is motivated by their expected superior quality given the possibility of growing them in crystalline form. We settled a theoretical and experimental controversy regarding the first phase to form by conducting experiments demonstrating that low-temperature forming cobalt germanide phases are highly ordered and could serve as high quality contacts. A predictive physical based mathematical model was developed to assist the designer in obtaining the desired cobalt germanide phase for its needed electrical properties by design. Factors affecting the quality of the germanide were identified based on an extensive survey and the optimum choices for the parameters to obtain high quality contact were pointed.
| Identifer | oai:union.ndltd.org:mcmaster.ca/oai:macsphere.mcmaster.ca:11375/24363 |
| Date | January 2019 |
| Creators | Rabie, Mohamed |
| Contributors | Haddara, Yaser, Electrical and Computer Engineering |
| Source Sets | McMaster University |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0019 seconds