Surface plasmons are of interest for various applications, including optical interconnects and devices, light sources, nanolithography, biosensors, solar cells, and negative-refraction prisms or superlenses. Some of the most important applications are SP-based optical interconnects and devices, which offer the potential of realizing integrated optical nanocircuitry due to the subwavelength confinement and the slow-wave nature of SPs. The fundamental building element of these applications is the plasmonic waveguide. Among the family of various plasmonic waveguides, the metal-insulator-metal waveguide has superior lateral confinement because of the relatively shallow field penetration into the metal claddings (about a skin depth -- usually tens of nanometers). Such subwavelength confinement cannot be achieved by conventional dielectric optical waveguides. However, the loss in the MIM waveguide is substantial due to the strong absorption of metal in the visible or near-infrared spectrum. Therefore, the design, simulation, and measurement of the loss in the MIM waveguide are critically important in the development of SP-based nanocircuitry. Surface plasmons (SPs) are of interest for various applications, including optical interconnects and devices, light sources, nanolithography, biosensors, solar cells, and negative-refraction prisms or superlenses. Some of the most important applications are SP-based optical interconnects and devices, which offer the potential of realizing integrated optical nanocircuitry due to the subwavelength confinement and the slow-wave nature of SPs. The fundamental building element of these applications is the plasmonic waveguide. Among the family of various plasmonic waveguides, the metal-insulator-metal (MIM) waveguide has superior lateral confinement because of the relatively shallow field penetration into the metal claddings (about a skin depth -- usually tens of nanometers). Such subwavelength confinement cannot be achieved by conventional dielectric optical waveguides. However, the loss in the MIM waveguide is substantial due to the strong absorption of metal in the visible or near-infrared spectrum. Therefore, the design, simulation, and measurement of the loss in the MIM waveguide are critically important in the development of SP-based nanocircuitry. Owing to the subwavelength sizes of MIM waveguides, the excitation of an MIM plasmonic mode typically requires end-fire coupling with tapered fibers or waveguides. Further, the conventional loss measurements require the usage of a near-field scanning optical microscopy (NSOM) or multiple waveguide samples with various length scales; however, the two aforementioned techniques are both complicated and have issues of sensitivity to uncontrollable environmental factors or variations in coupling strength, respectively. These experimental challenges have been a primary reason for the slow experimental development of the MIM waveguide and device. The research in this thesis focuses on the development of the transverse transmission/reflection (TTR) method, which is a more reliable, accurate, and straightforward method of characterizing the plasmonic modes in the MIM waveguide. The theory of the TTR method, which incorporates an attenuated total reflection (ATR) configuration, is developed based on the transmission matrix formulation. A methodology for obtaining the propagation constant and attenuation coefficient of a plasmonic mode in an MIM waveguide is illustrated. Using the Metricon Prism Coupler, the TTR method is experimentally applied to planar, single-mode MIM (Au-SiO$_2$-Au) waveguides with various core thicknesses at $lambda=1550$ nm. The experimental results are in very good agreement with the theoretical results. It is also shown experimentally that the TTR method is robust against difficult-to-quantify parameters such as the metal cladding thickness and the air gap thickness between the prism and the waveguide. As a result, the TTR method can be readily applied by using other similar ATR or prism-coupler configurations, without concern for the sensitivity issues caused by the subtle differences between various configurations. Moreover, the TTR method is also experimentally applied to planar, multimode MIM waveguides. Multimode MIM waveguides, which have larger core sizes, may be of interest for applications in low-loss interconnects or tapered end-couplers. Thanks to the superior angular selectivity of the ATR configuration, the TTR method is capable of detecting the propagation constant and attenuation coefficient of each mode. To the best of the author's knowledge, this is the first time the propagation constant of each mode in a multimode MIM waveguide has been individually measured. Also, to the best of the author's knowledge, this is the first time the attenuation coefficient of each mode in a multimode MIM waveguide has been individually measured. The TTR method is proved to be a reliable, accurate, and straightforward approach to characterize plasmonic modes in MIM waveguides. Future research will target the extension of the TTR method to 2D MIM waveguides, asymmetric MIM waveguides, and inclusion of scattering loss. Taking full advantage of the TTR method, the development of plasmonic devices can be potentially accelerated. The theory of the TTR method, which incorporates an attenuated total reflection (ATR) configuration, is developed based on the transmission matrix formulation. A methodology for obtaining the propagation constant and attenuation coefficient of a plasmonic mode in an MIM waveguide is illustrated. Using the Metricon Prism Coupler, the TTR method is experimentally applied to planar, single-mode MIM (Au-SiO$_2$-Au) waveguides with various core thicknesses at $lambda=1550$ nm. The experimental results are in very good agreement with the theoretical results. It is also shown experimentally that the TTR method is robust against difficult-to-quantify parameters such as the metal cladding thickness and the air gap thickness between the prism and the waveguide. As a result, the TTR method can be readily applied by using other similar ATR or prism-coupler configurations, without concern for the sensitivity issues caused by the subtle differences between various configurations. Moreover, the TTR method is also experimentally applied to planar, multimode MIM waveguides. Multimode MIM waveguides, which have larger core sizes, may be of interest for applications in low-loss interconnects or tapered end-couplers. Thanks to the superior angular selectivity of the ATR configuration, the TTR method is capable of detecting the propagation constant and attenuation coefficient of each mode. To the best of the author's knowledge, this is the first time the propagation constant of each mode in a multimode MIM waveguide has been individually measured. Also, to the best of the author's knowledge, this is the first time the attenuation coefficient of each mode in a multimode MIM waveguide has been individually measured. The TTR method is proved to be a reliable, accurate, and straightforward approach to characterize plasmonic modes in MIM waveguides. Future research will target the extension of the TTR method to 2D MIM waveguides, asymmetric MIM waveguides, and inclusion of scattering loss. Taking full advantage of the TTR method, the development of plasmonic devices can be potentially accelerated.

Thesis (M.S.)--Oregon State University, 2010. / Printout. Includes bibliographical references (leaves 97-106). Also available on the World Wide Web.

Thesis (Ph. D.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2004. / Rhodes, William, Committee Member ; Brooke, Martin, Committee Chair ; Chang, G.K., Committee Member ; Hasler, Paul, Committee Member ; Kohl, Paul, Committee Member. Includes bibliographical references (leaves 133-139).

Thesis (Ph. D.)--School of Electrical and Computer Engineering, Georgia Institute of Technology, 2004. Directed by James D. Meindl. / Vita. Includes bibliographical references (leaves 289-297).

Thesis (Ph. D.)--University of Texas at Austin, 2002. / Vita. Includes bibliographical references. Available also from UMI Company.

Thesis (Ph.D)--Electrical and Computer Engineering, Georgia Institute of Technology, 2009. / Committee Chair: Madhavan Swaminathan; Committee Member: Andrew E. Peterson; Committee Member: Emmanouil M. Tentzeris; Committee Member: Hao-Min Zhou; Committee Member: Saibal Mukhopadhyay. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Improvements for Chip-Chip Interconnects and MEMS Packaging Through Materials and Processing Research Erdal Uzunlar 129 Pages Directed by Dr. Paul A. Kohl The work presented in this dissertation focuses on improvements for ever-evolving modern microelectronic technology. Specifically, three topics were investigated in this work: electroless copper deposition on printed wiring boards (PWBs), polymer-based air-gap microelectromechanical systems (MEMS) packaging technology, and thermal stability enhancement in sacrificial polymers, such as poly(propylene carbonate) (PPC). In the electroless copper deposition study, Ag-based catalysts were identified as a low-cost and equally active alternative to expensive Pd-based catalysts. Hot H2SO4 treatment of PWBs was found as a non-roughening surface treatment method to minimize electrical losses. In MEMS packaging study, a sacrificial polymer-based air-gap packaging technique was improved in terms of identification and simplification of air-gap formation process options, optimization of thermal treatment steps, assessing air-gap formation performance, and analyzing the chemical composition of residue. It was found that non-photosensitive PPC leaves less residue, and creates more reliable air-gaps. The mechanical strength of air-gaps was found to come from residual stress in benzocyclobutene (BCB) caps. In thermal stability of PPC study, the mechanism of thermal stability increase on copper (Cu) surfaces was found as the complex formation between Cu(I) and iodonium of the photoacid generator (PAG), leading to hindrance of acid formation by PAG and restriction of acid-catalyzed decomposition of PPC.

The scaling required to accommodate faster chip performance in microelectronic devices has necessitated a reduction in the dimensions of copper interconnects at the back end of the line. The constant downscaling of copper interconnects has resulted in changes to the microstructure, and these variations are known to impact electrical resistivity and reliability issues in interconnects. In this work, a novel electron diffraction technique called Diffraction Scanning Transmission Electron Microscopy (D-STEM) has been developed and coupled with precession electron microscopy to obtain quantitative local texture information in damascene copper lines (1.8 \mu m to 70 nm in width) with a spatial resolution of less than 5 nm. Misorientation and trace analysis has been performed to investigate the the grain boundary distribution in these lines. The results reveal strong variations in texture and grain boundary distribution of the copper lines upon downscaling. 1.8 \mu m wide lines exhibit strong <111> normal texture and comprise large bamboo-type grains. Upon downscaling to 180 nm, a {111} <110> biaxial texture has been observed. In contrast, narrower lines of widths 120 nm and 70 nm reveal sidewall growth of {111} grains and a dominant <110> normal texture. The fraction of coherent twin boundaries also reduces with decreasing line width. The microstructure changes from bamboo-type in wider lines to one comprising clusters of small grains separated by high angle boundaries in the vicinity of large grains. The evolution of such a microstructure has been discussed in terms of overall energy minimization and dimensional constraints. Finite element analysis has been performed to correlate misorientations between grains and local thermal stresses associated with stress migration. Effect of variations in the copper interconnect microstructure on electromigration flux divergence has also been discussed. / text

Not available / text

As the gate density increases in microelectronic devices, the interconnect delay or RC response also increases and has become the limiting delay to faster devices. In order to decrease the RC time delay, a new metallization scheme has been chosen by the semiconductor industry. Copper has replaced aluminum as the metal lines and new low-k dielectric materials are being developed to replace silicon dioxide. A promising low-k material is porous organosilicate glass or p-OSG. The p-OSG film is a hybrid material where the silicon dioxide backbone is terminated with methyl or hydrogen, reducing the dielectric constant and creating mechanically weak films that are prone to fracture. A few methods of improving the mechanical properties of p-OSG films have been attempted-- exposing the film to hydrogen plasma, electron beam curing, and ultra-violet light curing. Hydrogen plasma and electron-beam curing suffer from a lack of specificity and can cause charging damage to the gates. Therefore, ultra-violet light curing (UV curing) is preferable. The effect of UV curing on an ultra-low-k, k~2.5, p-OSG film is studied in this dissertation. Changes in the molecular structure were measured with Fourier Transform Infrared Spectroscopy and X-ray Photoelectron Spectroscopy. The evolution of the molecular structure with UV curing was correlated with material and fracture properties. The material properties were film shrinkage, densification, and an increase in dielectric constant. From the changes in molecular structure and material properties, a set of condensation reactions with UV light are predicted. The connectivity of the film increases with the condensation reactions and, therefore, the fracture toughness should also increase. The effect of UV curing on the critical and sub-critical fracture toughness was also studied. The critical fracture toughness was measured at four different mode-mixes-- zero, 15°, 32°, and 42°. It was found that the critical fracture toughness increases with UV exposure for all mode mixes. The sub-critical fracture toughness was measured in Mode I and found to be insensitive to UV cure. A simple reaction rate model is used to explain the difference in critical and sub-critical fracture toughness. / text