Modeling of nickel deposition in a rotating disk CVD reactor
Ardham, Vikram Reddy
MetadataShow full item record
The Mond process is a major technique of producing pure nickel metal across the globe, the other major technique being electro-refining. Although this process has been practiced for over a century, many questions underlying the chemistry of the process still remain unanswered. The decomposition of nickel carbonyl to nickel and carbon monoxide is a significant step in the process, and we are trying to understand this step in detail. The decomposition reaction is supposed to occur by losing one carbonyl at a time resulting in pure nickel formation at the end. However, the exact mechanism is not understood and there is no reliable data on the kinetics of removal of each carbon monoxide. We have used theoretical quantum chemistry calculations to understand the mechanism; as nickel carbonyl is very poisonous this is arguably the best approach in terms of safety and costs. Transition metals are notoriously difficult to treat using classic computational methods. Therefore, to study gas phase reactions we have employed recently-developed hybrid computational quantum chemistry methods including ccCA-TM and G4/G4 (MP2) which are computationally expensive but perform quite well for these systems. For the solid-gas surface reactions, we relied mostly on the available literature supplemented by a few DFT calculations to develop a kinetic model. The chemical kinetic model is then used in a reactor-scale model of CVD of nickel from nickel carbonyl in a rotating disk reactor setup. This is a powerful research tool because it provides a simplified flow field in which the thickness of the boundary layer above the deposition substrate can be controlled. Here, a model has been developed using available CFD packages including Flow3D and ANSYS FLUENT and finally a simple 1D model using SPIN CHEMKIN. Though the model is yet to be validated by comparison to results from a reactor being built at Vale's research laboratories, it predicts very well the various deposition trends and concentration profiles above the deposition surface. After a validated model has been developed, this model can be used to study the effects of sulfur additives in the precursor and effects of substrate crystal structure and orientation. Moreover, the final design with optimal parameters in addition to the knowledge of chemistry acts as a solid bench mark to build models for new large scale industrial reactors.