The physical-chemical properties of several interfacial systems of technological relevance are investigated, having as a common goal the elucidation of strategies towards their atomic- and molecular-level control. Such systems can be classified in three groups: (i) ultra-thin films deposited using metalorganic precursors, (ii) metalorganic monolayers on silicon, and (iii) amine-functionalized silicon surfaces. Experimental, theoretical and chemometric methods are conveniently combined to gain a solid understanding of these systems.
The ultra-thin films under investigation are titanium carbonitride (TiNC) and hafnium oxide (HfO2). Since these films may serve as substrates for deposition of other materials in circuit components, their surface chemistry needs to be understood and controlled in order to facilitate further deposition steps. The surface of a TiCN film is transformed to titanium nitride (TiN) through nitridation with ammonia; this compositional change can be reversed by the partial decomposition of ethylene molecules on the surface. The surface reactivity is observed to depend on the film composition, and therefore the method described above serves to reversibly tune the reactivity of Ti-based films. As for HfO2 films, it is found that the deposition temperature affects the degree of crystallinity of the films, which in turn affects their surface chemistry. Thus, together with a control of the composition, it is found that the reactivity of a film can be controlled precisely by controlling the crystallinity.
The investigation of metalorganic monolayers on silicon surfaces was motivated by the need for understanding the first steps of metalorganic-based deposition of films, which is usually characterized by a heavy presence of contaminants that degrade the film properties. Through a combination of vibrational (infrared) spectroscopy and theoretical methods, a feasible pathway for the adsorption and decomposition of Ti[N(CH3)2]4 is found. This pathway starts with the ligand-mediated attachment of the precursor (through a N atom), followed by dissociation of a metal-ligand bond. In addition, the C-H bond is broken, possibly forming Si-C bonds and causing carbon incorporation. This model is found to be rather robust and to adequately describe other types of metalorganic precursors. It allows establishing a generalized model able to explain the success or failure of a metalorganic precursor chemistry for film deposition.
Finally, amine-functionalized silicon surfaces are considered as prototypical systems where the spatial distribution of adsorbates and the control over the reactivity of surface sites can be investigated. The spatial distribution of molecules is investigated at the atomic level by considering the saturation of a Si(100) surface with NH3. It is found that the distribution of (Si)NH2 species can be controlled thermally and, more importantly, that during thermal decomposition N inserts into the substrate in manners that minimize the arising strain. When the surface is covered with NH 3 or with organic amines, its chemical behavior is determined by the basicity of the molecule functionalizing the surface. The precise tuning of the reactivity (basicity) of surface sites opens the doors for highly controllable, selective reactions.
Although these results are obtained from rather fundamental grounds, their interpretation is often translated into manners in which technological applications can be improved. Further directions worth exploring emanated from this work are outlined and discussed. Ultimately, this work intends to highlight the current importance of surface physical chemistry in the continuous development of modern society through the improvement of its technology.
