Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Nanoscale Science and Engineering


Nanoscale Engineering

Content Description

1 online resource (xxxiv, 310 pages) : color illustrations.

Dissertation/Thesis Chair

Robert Geer

Committee Members

Alain Diebold, Nathaniel Cady, Gregory Denbeaux, Nicholas Fahrenkopf


Computational Analysis, Design, Lumerical, Photonics, Simulation, Testing, Integrated optics, Metal oxide semiconductors, Complementary, Nanophotonics, Silicon, Integrated circuits, Optical wave guides

Subject Categories

Nanoscience and Nanotechnology | Physics


The success of Si as a platform for photonic devices and the associated availabilityof wafer-scale, ultra-high resolution lithography for Si CMOS has helped lead to the rapid advance of Si-based integrated photonics manufacturing over the past decade. This evolution is nearing the point of integration of Si-based photonics together with Si-CMOS for compact, high speed, high bandwidth, and cost-effective devices. However, due to the sensitive nature of passive and active photonic devices, variations inherent in wafer-based fabrication processes can lead to unacceptable levels of performance variation both within a give die and across a given wafer. Fully understanding the role of variation in affecting integrated photonic device performance is an important component of scaling the design and manufacturing infrastructure for photonic integrated circuits. Specifically, understanding the key factors that drive the deviation of integrated photonic device performance from designed performance is essential to effectively manage the impact of variation and variability. To address these issues, dissertation research is proposed to develop a set of tools and designs to study and understand the impact of interfacial roughness, wafer thickness variation, feature size variation, and selected feature defects on the optical mode propagation in a set of integrated photonic devices and structures for intra-die, and die to die, and, possibly, batch to batch analysis. This variation analysis can provide meaningful information to manufacturers and designers to improve limitations on device variational control and establish solutions to address the challenges that multi-scale variation poses to the integrated photonics industry.