Our group’s current research efforts are largely devoted to the study of advanced electronic materials processing issues, particularly those related to thin film deposition (the group operates one Molecular Beam Epitaxy (MBE) and six chemical vapor deposition (CVD) systems. In one system we have coupled a Raman spectrometer to a CVD reactor, which can be x-y-z translated to measure gas phase composition and temperature profiles. Raman scattering and LIF, along with reactor modeling, are being used to quantitatively study homogeneous thermal decomposition mechanisms of organometallic precursors. These reaction mechanisms and rate constants are then used to optimize reactor designs and operating conditions. Current studies are focused on the thermal decomposition of column II and III alkyls.

Two of the CVD reactors (commercial metalorganic and hydride vapor phase epitaxy) are devoted to the growth of GaN and related materials. These wide-bandgap materials are of interest for visible and UV light emitting devices as well as high-temperature and high-power applications. Current projects include growth of thick GaN on Si, nucleation and growth of GaN and InN nanorods, and exploring the growth characteristics and properties of InN.

The performance and functionality of integrated circuits (IC) have continuously improved over the last three decades in part through reduction in the physical sizes of features. This reduction has motivated the use of copper metallization schemes to increase conductivity, but brings the need for barrier layers to prevent Cu diffusion into the underlying Si or interlevel dielectric. An NSF Collaborative Research in Chemistry program is supporting work in the development of new precursors to deposit barrier layers (e.g., TaN, WN, LaB₆). The industry roadmap requires the barrier film thickness to be ~10% of the minimum feature size, with the next node at 40 nm. To meet conformality requirements at this node, atomic layer deposition (ALD) methods are being studied, including understanding self-limiting adsorption.

The quality of bulk crystals grown from the melt is largely controlled by buoyancy driven flows. Unfortunately, there are very few techniques to visualize flows in opaque, high temperature, and low Pr number semiconductor melts. In a NASA funded program, we are using YSZ solid-state electrochemical cells, placed along the walls of the liquid metal container, to introduce, extract, and monitor dilute concentrations of dissolved atomic oxygen at the sub-ppm level to visualize flow. Coupling the measured values of the concentration of this tracer species as a function of time and location with a detailed model allows the flow dynamics to be estimated. In addition, we are constructing an integrated microsensor system to study short -dimension dynamics in small systems (e.g. drops).

We are participating in collaboration with the Electrical and Materials Sciences departments on research to reduce the costs of manufacturing photovoltaic solar cells based on Cu(In,Ga)Se₂(CIGS) thin film technology. This interdisciplinary program aims to probe fundamental issues such as reaction pathways, point defect chemistry, and phase equilibria, while exploring alternative processes such as MBE, rapid thermal processing, laser annealing, and ALD. Another program is focused on demonstrating tandem solar cells. In this approach, photon-energy in the different wavelength regions of the solar spectrum are more efficiently converted into electricity by using a stack of single-junction solar cells. Our vision is a tandem structure consisting of a CIS (1.04 eV) bottom cell and a CGS (1.68 eV) top cell. In other research, the use of In_xGa_1-xN solid solutions are being tested as possible PV materials, and organic PV cells are attempting to be integrated with inorganic ones.

The solution to many of the problems in the processing of advanced materials is aided by knowledge of the phase diagram and thermochemistry of these materials. Our group routinely measures component activities in liquid and solid solutions with solid state galvanic cells. This data along with other available data is then critically assessed and solution model parameters estimated to predict multicomponent phase diagrams and compute complex reaction equilibria. We also routinely perform molecular simulations in experimentally difficult systems.