Environmentally benign semiconductor processing for dielectric etch

Semiconductor processing requires intensive usage of chemicals, electricity, and water. Such intensive resource usage leaves a large impact on the environment. For instance, in Silicon Valley, the semiconductor industry is responsible for 80% of the hazardous waste sites contaminated enough to require government assistance. Research on environmentally benign semiconductor processing is needed to reduce the environmental impact of the semiconductor industry. The focus of this dissertation is on the environmental impact of one aspect of semiconductor processing: patterning of dielectric materials.; Plasma etching of silicon dioxide emits perfluorocarbons (PFCs) gases, like C₂F₆ and CF₄, into the atmosphere. These gases are super global warming/greenhouse gases because of their extremely long atmospheric lifetimes and excellent infrared absorption properties. We developed the first inductively coupled plasma (ICP) abatement device for destroying PFCs downstream of a plasma etcher. Destruction efficiencies of 99% and 94% can be obtained for the above mentioned PFCs, by using O ₂ as an additive gas. Our results have lead to extensive modeling in academia as well as commercialization of the ICP abatement system.; Dielectric patterning of hi-k materials for future device technology brings different environment challenges. The uncertainty of the hi-k material selection and the patterning method need to be addressed. We have evaluated the environmental impact of three different dielectric patterning methods (plasma etch, wet etch and chemical-mechanical polishing), as well as, the transistor device performances associated with the patterning methods.; Plasma etching was found to be the most environmentally benign patterning method, which also gives the best device performance. However, the environmental concern for plasma etching is the possibility of cross-contamination from low volatility etch by-products. Therefore, mass transfer in a plasma etcher for a promising hi-k dielectric material, ZrO₂, was studied. A novel cross-contamination sampling technique was developed, along with a mass transfer model.