京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第18778号 / 理博第4036号 / 新制||理||1581(附属図書館) / 31729 / 京都大学大学院理学研究科物理学・宇宙物理学専攻 / (主査)准教授 松田 和博, 教授 八尾 誠, 教授 田中 耕一郎 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第18979号 / 工博第4021号 / 新制||工||1619(附属図書館) / 31930 / 京都大学大学院工学研究科マイクロエンジニアリング専攻 / (主査)教授 鈴木 基史, 教授 木村 健二, 教授 蓮尾 昌裕 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

No description available.

Metal nanostructures have been known for a long time to exhibit optical resonances via localized surface plasmons. The high electric fields in close proximity to the metal surface have prospects to dramatically change the dynamics of electronic transitions, such as an enhanced spontaneous decay rate of a single emitter. However, there have been two major issues which impede advances in the experimental realization of enhanced light-matter interaction. (i) The fabrication of high-quality resonant structures requires state-of-the-art patterning techniques in combination with superior materials. (ii) The tiny extension of the optical near-field requires precise control of the single emitter with respect to the nanostructure. This work demonstrates a solution to these problems by combining scanning probe and optical confocal microscopy. Here, a novel type of scanning probe is introduced which features a tip composed of the edge of a single crystalline gold sheet. The patterning via focused ion beam milling makes it possible to introduce a plasmonic nanoresonator directly at the apex of the tip. Numerical simulations demonstrate that the optical properties of this kind of scanning probe are ideal to analyze light-matter interaction. Detailed experimental studies investigate the coupling mechanism between a localized plasmon and single colloidal quantum dots by dynamically changing coupling strength via their spatial separation. The results have shown that weak interaction affects the shape of the fluorescence spectrum as well as the polarization. For the best probes it has been found that it is possible to reach the strong coupling regime at the single emitter level at room temperature. The resulting analysis of the experimental data and the proposed theoretical models has revealed the differences between the established far-field coupling and near-field coupling. It has been found that the broad bandwidth of plasmonic resonances are able to establish coherent coupling to multiple transitions simultaneously giving rise to an enhanced effective coupling strength. It has also been found that the current model to numerically calculate the effective mode volume is inaccurate in case of mesoscopic emitters and strong coupling. Finally, light-matter interaction is investigated by the means of a quantum-dot-decorated microtubule which is traversing a localized nearfield by gliding on kinesin proteins. This biological transport mechanism allows the parallel probing of a meta-surface with nm-precision. The results that have been put forward throughout this work have shed new light on the understanding of plasmonic light-matter interaction and might trigger ideas on how to more efficiently combine the power of localized electric fields and novel excitonic materials. / Metallische Nanostrukturen sind seit langer Zeit bekannt dafür optische Resonanzen durch lokalisierte Oberflächenplasmonen zu zeigen. Hohe elektrische Felder in direkter Nähe zur Metalloberfläche versprechen dramatische Dynamikänderungen von elektrischen Übergängen wie z.B. die gesteigerte spontane Zerfallsrate eines Einzelemitters. Es gibt jedoch zwei maßgebliche Gründe warum die Fortschritte der experimentellen Realisierung von Licht-Materie Wechselwirkung ausgebremst wird. (i) Die Herstellung von qualitativ hochwertigen resonanten Strukturen benötigt modernste Strukturierungsmethoden sowie die bestmöglichen Materialeigenschaften. (ii) Die winzigen Dimensionen von optischen Nahfeldern erfordern eine präzise Kontrolle des Einzelemitters im Bezug zur Nanostruktur. Diese Arbeit löst diese Probleme durch die Kombination eines Rasterkraftmikroskops mit einem optischen Konfokalmikroskop. Dabei wird eine neuartige Rastersonde vorgestellt welche eine Spitze aufweist die aus der Ecke einer monokristallinen Goldflocke besteht. Die Strukturierung mittels eines fokussierten Ionenstrahls ermöglicht es einen plasmonischen Nanoresonator direkt an der Spitze der Sonde herzustellen. Numerische Simulationen haben gezeigt, dass die optischen Eigenschaften für diese Art von Sonde ideal sind um Licht-Materie Wechselwirkung zu untersuchen. Die hier gezeigten experimentellen Studien haben den Kopplungsmechanismus zwischen lokalisierten Plasmonen und einzelnen kolloidalen Quantenpunkten untersucht indem die Kopplungstärke dynamisch über den Abstand kontrolliert wurde. Die Ergebnisse haben gezeigt, dass schwache Wechselwirkung einen Einfluss auf die Form des Fluoreszenzspektrums als auch auf die Polarisation hat. Die besten Sonden haben gezeigt, dass es möglich ist starke Wechselwirkung mit Einzelemittern bei Raumtemperatur zu erreichen. Die resultierende Analyse der experimentellen Daten und die aufgestellten theoretischen Modelle haben die Unterschiede zwischen der etablierten Fernfeldkopplung und der Nahfeldkopplung aufgezeigt. Dabei wurde beobachtet, dass die große Bandbreite von plasmonischen Resonanzen es möglich macht kohärent mit mehreren Übergängen gleichzeitig zu koppeln und dabei die effektive Kopplungsstärke zu höhen. Es wurde auch festgestellt, dass das aktuelle Model zur numerischen Beschreibung von effektiven Modenvolumen Ungenauigkeiten bei mesoskopischen Emittern und starker Wechselwirkung aufzeigt. Zuletzt wird die Licht-Materie Wechselwirkung mittels Quantenpunkt-bestückten Mikrotubuli untersucht, die auf Kinesin Proteinen durch ein lokalisiertes Nahfeld gleiten. Dieses biologische Transportsystem erlaubt es eine Meta-Oberfläche mit nm-Präzision parallel zu untersuchen. Die Ergebnisse, die diese Arbeit hervorgebracht hat, wirft neues Licht auf das Verständnis von plasmonischer Licht-Materie Wechselwirkung und könnte als Grundlage dienen neue Ideen zu entwickeln um effizienter die Stärke von lokalisierten elektrischen Felder und neuartiger exzitonischer Materialien zu kombinieren.

Graphene nanoribbons, graphene based optical sensors, and grating based plasmonics are explored experimentally. Graphene nanoribbons exhibit highly insulating states that may allow for graphene based digital applications. We investigate the sensitivity of these states to local charged impurities in ultra high vacuum. We look into the possibility of isolating two-dimensional films of H-BN and BSCCO, and test for any interesting phenomena. We also assess graphene*s applicability for optical sensing by implementing a new style of spectral detector. Utilizing surface plasmon excitations nearby a graphene field-effect transistor we are able to produce a detector with wavelength sensitivity and selectivity in the visible range. Finally, we study another plasmonic phenomenon, and observe the resonant enhancement of diffraction into a symmetry-prohibited order in silver gratings.

The current thrust towards developing silicon compatible integrated nanophotonic devices is driven by need to overcome critical challenges in electronic circuit technology related to information bandwidth and thermal management. Surface plasmon nanophotonics represents a hybrid technology at the interface of optics and electronics that could address several of the existing challenges. Surface plasmons are electronic charge density waves that can occur at a metal-dielectric interface at optical and infrared frequencies. Numerous plasmon based integrated optical devices such as waveguides, splitters, resonators and multimode interference devices have been developed, however no standard integrated device for coupling light into nanoscale optical circuits exists. In this thesis we experimentally and theoretically investigate the excitation of propagating surface plasmons via resonant metal nanoparticle arrays placed in close proximity to a metal surface. It is shown that this approach can lead to compact plasmon excitation devices. Full-field electromagnetic simulations of the optical illumination of metal nanoparticle arrays near a metal film reveal the presence of individual nanoparticle resonances and collective grating-like resonances related to propagating surface plasmons within the periodic array structure. Strong near-field coupling between the nanoparticle and grating resonances is observed, and is successfully described by a coupled oscillator model. Numerical simulations of the effect of nanoparticle size and shape on the excitation and dissipation of surface plasmons reveal that the optimum particle volume for efficient surface plasmon excitation depends sensitively on the particle shape. This observation is quantitatively explained in terms of the shape-dependent optical cross-section of the nanoparticles. iv Reflection measurements on nanoparticle arrays fabricated using electron-beam lithography confirm the predicted particle-grating interaction. An unexpected polarizationdependent splitting of the film-mediated collective resonance is successfully attributed to the existence of out-of plane polarization modes of the metal nanoparticles. In order to distinguish between the excitation of propagating surface plasmons and localized nanoparticle plasmons, spectrally resolved leakage radiation measurements are presented. Based on these measurements, a universally applicable method for measuring the wavelength dependent efficiency of coupling free-space radiation into guided surface plasmon modes on thin films is developed. Finally, it is shown that the resonantly enhanced near-field coupling the nanoparticles and the propagating surface plasmons can lead to optimized coupler device dimensions well below 10 m.

Cholera toxin (CT), secreted from Vibrio cholerae, causes a massive fluid and electrolyte efflux in the small intestine that results in life-threatening diarrhea and dehydration which impacts 3-5 million people per year. CT is secreted into the intestinal lumen but acts within the cytosol of intestinal epithelial cells. CT is an AB5 toxin that has a catalytic A1 subunit and a cell binding B subunit. CT moves from the cell surface to the endoplasmic reticulum (ER) by retrograde transport. Much of the toxin is transported to the lysosomes for degradation, but a secondary pool of toxin is diverted to the Golgi apparatus and then to the ER. Here the A1 subunit detaches from the rest of the toxin and enters the cytosol. The disordered conformation of free CTA1 facilitates toxin export to the cytosol by activating a quality control mechanism known as ER-associated degradation. The return to a folded structure in the cytosol allows CTA1 to attain an active conformation for modification of its Gsα target through ADP-ribosylation. This modification locks the protein in an active state which stimulates adenylate cyclase and leads to elevated levels of cAMP. A chloride channel located in the apical enterocyte membrane opens in response to signaling events induced by these elevated cAMP levels. The osmotic movement of water into the intestinal lumen that results from the chloride efflux produces the characteristic profuse watery diarrhea that is seen in intoxicated individuals. The current model of intoxication proposes only one molecule of cytosolic toxin is required to affect host cells, making therapeutic treatment nearly impossible. However, based on emerging evidence, we hypothesize a threshold quantity of toxin must be present within the cytosol of the target cell in order to elicit a cytopathic effect. Using the method of surface plasmon resonance along with toxicity assays, I have, for the first time, directly measured the efficiency of toxin delivery to the cytosol and correlated the levels of cytosolic toxin to toxin iv activity. I have shown CTA1 delivery from the cell surface to the cytosol is an inefficient process with only 2.3 % of the surface bound CTA1 appearing in the cytosol after 2 hours of intoxication. I have also determined and a cytosolic quantity of more than approximately .05ng of cytosolic CTA1 must be reached in order to elicit a cytopathic effect. Furthermore, CTA1 must be continually delivered from the cell surface to the cytosol in order to overcome the constant proteasome-mediated clearance of cytosolic toxin. When toxin delivery to the cytosol was blocked, this allowed the host cell to de-activate Gs, lower cAMP levels, and recover from intoxication. Our work thus indicates it is possible to treat cholera even after the onset of disease. These findings challenge the idea of irreversible cellular toxicity and open the possibility of postintoxication treatment options.

The observation of voltage-tunable plasmon resonances in the terahertz range in two dimensional electron gas (2-deg) of a high electron mobility transistor (HEMT) fabricated from the InGaAs/InP and AlGaN/GaN materials systems is reported. The devices were fabricated from a commercial HEMT wafer by depositing source and drain contacts using standard photolithography process and a semi-transparent gate contact that consisted of a 0.5 [micro]m period transmission grating formed by electron-beam lithography. Narrow-band resonant absorption of THz radiation was observed in transmission in the frequency range 10-100 cm-1. The resonance frequency depends on the gate voltage-tuned sheet-charge density of the 2deg. The fundamental and higher resonant harmonics were observed to shift towards lower frequencies with the implementation of negative gate bias. The theory of interaction of sub millimeter waves with 2deg through corrugated structure on top has been applied to calculate and understand the phenomena of resonant plasmon excitations. The observed separation of resonance fundamental from its harmonics and their shift with gate bias follows theory, although the absolute frequencies are lower by about a factor of 2-3 in InGaAs/InP system. However, calculated values match much better with AlGaN/GaN system.

The glycoprotein CD36 is expressed in the plasma membrane (PM) of many cell types that surround or contact arteries, including macrophages, myocytes, and endothelial cells. CD36 binds oxidized low density lipoprotein (oxLDL), which promotes atherosclerosis, and fatty acids (FA), which promotes their cellular uptake. To gain insights into the molecular mechanisms of uptake, HEK293 cells expressing CD36 were studied by cell biological and fluorescence methods. To test our hypothesis that the PM is not an impermeable barrier to FA and that FA move into cells by diffusion via their uncharged form, we first applied biophysical fluorescence spectroscopy to directly measure transmembrane FA movement and membrane fluidity. Expression of CD36 in HEK293 cells did not increase either transport across the PM or the fluidity of the PM compared to HEK293 cells without CD36; however, CD36 enhanced intracellular FA esterification. Furthermore, the widely used “inhibitors” of FA transport did not alter either the rapid FA transmembrane diffusion in HEK293 cells or diffusion in control experiments with protein-free phospholipid bilayers. To gain new insights into the physiological relevance of FA binding to CD36, we applied surface plasmon resonance (SPR) to quantify FA and oxLDL binding to the ectodomain of CD36. Structurally distinct FA [saturated, monounsaturated (cis and trans), polyunsaturated, ω-3, ω-6, and oxidized FA] were pulsed in a solubilized form (bound to methyl-β-cyclodextrin) across SPR channels, generating real-time association and dissociation binding curves. With the exception of the oxidized FA hydroxyoctadecadienoic acid (HODE), all FA tested bound to CD36 with rapid association and dissociation kinetics similar to human serum albumin. In addition, FA increased oxLDL binding to CD36. To investigate whether FA affect CD36-mediated oxLDL uptake in live cells, we monitored fluorescent oxLDL (Dii-oxLDL) uptake using confocal microscopy. Addition of exogenous FA to serum-free media enhanced dose-dependent oxLDL uptake. Exceptions were ω-3 FA, which bound to CD36, and HODE, which did not bind to CD36, demonstrating FA structure-specific effects on a major function of CD36 and a new mechanistic link between atherosclerosis and high levels of FA in obese and Type-II diabetic individuals.

Non-equilibrium systems have no set method of analysis, and a wide array of dynamics can be present in such systems. In this work we present three very different non-equilibrium models, inspired by biological systems and phenomena, that we analyze through computational means to showcase both the range of dynamics encompassed by these systems, as well as various techniques used to analyze them. The first system we model is a surface plasmon resonance (SPR) cell, a device used to determine the binding rates between various species of chemicals. We simulate the SPR cell and compare these computational results with a mean-field approximation, and find that such a simplification fails for a wide range of reaction rates that have been observed between different species of chemicals. Specifically, the mean-field approximation places limits on the possible resolution of the measured rates, and such an analysis fails to capture very fast dynamics between chemicals. The second system we analyzed is an avalanching neural network that models cascading neural activity seen in monkeys, rats, and humans. We used a model devised by Lombardi, Herrmann, de Arcangelis et al. to simulate this system and characterized its behavior as the fraction of inhibitory neurons was changed. At low fractions of inhibitory neurons we observed epileptic-like behavior in the system, as well as extended tails in the avalanche strength and duration distributions, which dominate the system in this regime. We also observed how the connectivity of these networks evolved under the effects of different inhibitory fractions, and found the high fractions of inhibitory neurons cause networks to evolve more sparsely, while networks with low fractions maintain their initial connectivity. We demonstrated two strategies to control the extreme avalanches present at low inhibitory fractions through either the random or targeted disabling of neurons. The final system we present is a sparsely encoding convolutional neural network, a computational system inspired by the human visual cortex that has been engineered to reconstruct images inputted into the network using a series of "patterns" learned from previous images as basis elements. The network attempts to do so "sparsely," so that the fewest number of neurons are used. Such systems are often used for denoising tasks, where noisy or fragmented images are reconstructed. We observed a minimum in this denoising error as the fraction of active neurons was varied, and observed the depth and location of this minimum to obey finite-size scaling laws that suggest the system is undergoing a second-order phase transition. We can use these finite-size scaling relations to further optimize this system by tuning it to the critical point for any given system size. / Doctor of Philosophy / Non-equilibrium systems have no set method of analysis, and a wide array of dynamics can be present in such systems. In this work we present three very different non-equilibrium models, inspired by biological systems and phenomena, that we analyze through computational means to showcase both the range of dynamics encompassed by these systems, as well as various techniques used to analyze them. The first system we model is a surface plasmon resonance (SPR) cell, a device used to determine the binding rates between various species of chemicals. We simulate the SPR cell and compare these computational results with a mean-field approximation, and find that such a simplification fails for a wide range of reaction rates that have been observed between different species of chemicals. Specifically, the mean-field approximation places limits on the possible resolution of the measured rates, and such an analysis fails to capture very fast dynamics between chemicals. The second system we analyzed is an avalanching neural network that models cascading neural activity seen in monkeys, rats, and humans. We used a model devised by Lombardi, Herrmann, de Arcangelis et al. to simulate this system and characterized its behavior as the fraction of inhibitory neurons was changed. At low fractions of inhibitory neurons we observed epileptic-like behavior in the system, as well as extended tails in the avalanche strength and duration distributions, which dominate the system in this regime. We also observed how the connectivity of these networks evolved under the effects of different inhibitory fractions, and found the high fractions of inhibitory neurons cause networks to evolve more sparsely, while networks with low fractions maintain their initial connectivity. We demonstrated two strategies to control the extreme avalanches present at low inhibitory fractions through either the random or targeted disabling of neurons. The final system we present is a sparsely encoding convolutional neural network, a computational system inspired by the human visual cortex that has been engineered to reconstruct images inputted into the network using a series of “patterns” learned from previous images as basis elements. The network attempts to do so “sparsely,” so that the fewest number of neurons are used. Such systems are often used for denoising tasks, where noisy or fragmented images are reconstructed. We observed a minimum in this denoising error as the fraction of active neurons was varied, and observed the depth and location of this minimum to obey finite-size scaling laws that suggest the system is undergoing a second-order phase transition. We can use these finite-size scaling relations to further optimize this system by tuning it to the critical point for any given system size.