PITA Fiscal Year 2007 Projects

Product and Process Design and Optimization

Error Control in Distributed Simulation of Complex Chemical Processes: Application to Carbothermic Aluminum Production
The objective is to develop and test a novel architecture for asynchronous, multi-scale simulation with error control. The method we develop relies on tessellation (“gridding”) of physical space and computational architectures which map the simulation problem onto cluster and/or distributed computing systems. The broader objective is to use the simulation tool to aid in pilot plant design and scale-up of a new carbothermic reduction process for aluminum production. Aluminum is easy to recycle and lightweight, it has excellent processing capability and it is resistant corrosion. The carobthermic process promises to reduce capital and operating expenses for aluminum production by about 20% relative to the Hall-Heroult process.