Doctor of Philosophy (PhD)


Chemical Engineering

Document Type



Nanofabrication is state of the art technology. Various chemical, mechanical, biochemical and semiconductor products have characteristics controlled by the nanostructures of the surface and interphase. Surface microscopic imaging is generally used to capture different surface features. By properly analyzing the surface image, valuable information regarding manufacturing process and product performance can be extracted. While microscopy measurements can offer very accurate qualitative information about surface features, for many applications, it is critical to obtain a quantitative description of the surface morphology. Various statistical features can be used to characterize the surface in quantitative way. Such an analysis can be done by the multi-resolution capabilities of wavelet transforms (WT). A multi-scale molecular simulation can help to investigate the physical and chemical mechanism in manufacturing process. Multiresolution characterization was performed on the model structure to compare with image analysis. In our research, we have used a soft polymeric surface used in microfabrication application and a hard surface used for catalysis, and applied multiresolution characterization for surface feature extraction and multiscale modeling for optimizing system variables to get desired surface characteristics. In microfabrication, the efficiency of the product reduced by line-edge roughness (LER) created on the polymer surface. Off-line LER characterization is usually based on the top-down SEM image. We have shown a wavelet based segmentation method for edge searching region. There was no external decision involved in the wavelet based edge detection and characterization. Ab-initio atomistic based simulations are generally used for polymer material design in atomic scale. For mesoscale modeling we use the coarse graining of the molecules and use the Flory-Huggins mean field interaction parameters of the clusters of atoms or molecules obtained from ab-initio simulations. In our research we have used coarse grained lattice based important sampling Monte Carlo (MC) and kinetic Monte Carlo (kMC) methods for mesoscale simulation. We have identified the phase separation by spinodal decomposition resulting in the formation of LER. The kinetics of the process is found and the process variables are identified that can reduce the roughness. Surface of a transition metal have been analyzed in a similar way for enhanced catalytic performance.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Romagnoli, Jose A