Experiment-Driven Modeling of Plasmonic NanostructuresPublic Deposited
Plasmonic nanostructures can confine light at their surface in the form of surface plasmon polaritons (SPPs) or localized surface plasmons (LSPs) depending on their geometry. SPPs are excited on nano- and micropatterned surfaces, where the typical feature size is on the order of the wavelength of light. LSPs, on the other hand, can be excited on nanoparticles much smaller than the diffraction limit. In both cases, far-field optical measurements are used to infer the excited plasmonic modes, and theoretical models are used to verify those results. Typically, these theoretical models are tailored to match the experimental nanostructures in order to explain observed phenomena. In this thesis, I explore incorporating components of experimental procedures into the models to increase the accuracy of the simulated result, and to inform the design of future experiments. First, I examine SPPs on nanostructured metal films in the form of low-symmetry moiré plasmonic crystals. I created a general Bragg model to understand and predict the excited SPP modes in moiré plasmonic crystals based on the nanolithography masks used in their fabrication. This model makes use of experimental parameters such as periodicity, azimuthal rotation, and number of sequential exposures to predict the energies of excited SPP modes and the opening of plasmonic band gaps. The model is further expanded to apply to multiscale gratings, which have patterns that contain hierarchical periodicities: a sub-micron primary periodicity, and microscale superperiodicity. A new set of rules was established to determine how superlattice SPPs are excited, and informed development of a new fabrication technique to create superlattices with multiple primary periodicities that absorb light over a wider spectral range than other plasmonic structures. The second half of the thesis is based on development of finite-difference time-domain (FDTD) simulations of plasmonic nanoparticles. I created a new technique to model pyramidal bowtie nanoparticle dimers based on the experimental fabrication procedure. This model was used to sweep various experimental parameters to identify their effect on the LSP resonance of the bowties. Analyzing the near-field distribution around these particles revealed the origin of a miscategorized LSP mode to be an out-of-plane dipole. Finally, I developed a finite-difference time-domain model that simulates the images generated by differential interference contrast (DIC) microscopy of gold nanorods. I discovered that the image contrast of gold nanorods is dependent on the wavelength of incident light relative to the LSP resonance wavelength. Incorporating experimental parameters into the DIC model allowed me to find a correlation between the electric near-field and far-field image contrast, uncovering the origin of this wavelength dependence. Additionally, the simulated DIC image patterns aid in breaking the angular degeneracy associated with the rotation of symmetric nanorods and can be used as training data for future machine learning algorithms to predict the size, shape, and orientation of nanoparticles from far-field images alone.