Date of Award

Spring 1-1-2011

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Electrical, Computer & Energy Engineering

First Advisor

Alan R. Mickelson

Second Advisor

Won Park

Third Advisor

Garret Moddel

Abstract

The distinct behavior of structures of noble metals with nano-dimensions from that of the bulk metal in visible electromagnetic spectrum has provided access to exquisite characteristics and applications in the field of plasmonics. Plasmonics is a field of study of the interfaces of such metals and dielectrics. Embedding nanoparticles of noble metals in dielectric materials to achieve and tune plasmonic features to regions of spectrum, where naturally available materials do not exhibit such phenomenon, has been a subject of great attention for researchers. In this thesis, I present the design and fabrication of coatings composed of thin metal and dielectric layers as a solution to tune the plasmonic features to other visible wavelengths and to near infrared regions and vary the optical properties of the coatings. Pushing down the limits on thickness of the layers far below the subwavelength dimensions has become a necessity to able to tune the plasmonic properties into the near infrared spectrum and beyond. This necessity has forced us to closely study the effects of roughness and continuity of the layers. These effects of pushing the layers to thinner dimensions on the optical properties will be presented and discussed with the aid of topographical nanoscopy images. Theoretical computations of transmission spectra of these coatings are accomplished using Maxwell Garnett approximation and a comparison with empirical results is presented. First thin film flat layers are demonstrated to show the passive tuning of dielectric function, while supporting them with experimental results and theoretical simulations. In later part, thickness of the metal layers is decreased to limits where the effects of roughness and continuity of layers play a substantial role. The surfaces of layers are characterized in detail. A new efficient statistical model is developed that is built on the distribution of size and shapes of particles involved in percolation. This model is used to study the tuning abilities of these layers.

Share

COinS