UBIR Repository (Univ. at Buffalo)
Not a member yet
5736 research outputs found
Sort by
Linear and Bottlebrush Block Copolymers with Highly Incompatible Segments
No full textPh.D.The rapid development of miniaturized electronic devices has motivated the creation of methods for the production of nanomaterials with sub-10 nm features. Block copolymer self-assembly is a highly effective bottom-up approach to generating nanostructured materials. Two major challenges stand in the way of using block copolymer for the fabrication of patterns with ultrasmall pitch sizes: (1) the access to polymers with high segmental interaction parameter that drives phase separation at very small chain lengths, and (2) methods of orientating block copolymer morphology perpendicular to the substrate in thin films. The research presented in this thesis explores new chemistries and polymer architectures to address the above mentioned challenges...**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Magnetic Proximity Effects in Hybrid Graphene/Ferromagnetic Systems
No full textPh.D.In this thesis, we investigate quantum transport in graphene under nonequilibrium conditions, focusing both on the behavior exhibited by native graphene, and by heterostructures of graphene and a ferromagnetic metal. Differential conductance mapping is used to probe the quan-tum corrections to the conductance of graphene at low temperatures (3-50 K). Weak-localization (WL) effects, which depend strongly on temperature, are observed as a result of coherent transport in graphene devices, realized on both SiO2 and hexagonal bornon nitride(h-BN) substrates. The WL is manifested in bare graphene samples as a dip around zero bias in the differential conductance map. The application of a bias voltage leads to dephasing of the carriers and suppression of this WL. By plotting the bias-induced change of differential con-ductance as a function of the dimensionless voltage (eV/kBT), we find that all data collapse onto a universal, temperature-independent form. According to this, the linear conductance remains approximately unchanged for voltages eV < kBT, before crossing over to a logarithmic function of V at larger voltages, reflecting the quenching of the quantum corrections. By implementing graphene/ferromagnetic hybrid structures in a "polarizer-analyzer" geometry, we map the differential conductance with and without the ferromagnetic polarizer included in the current path. Symmetry breaking spin-orbit coupling is induced in graphene charge carriers by the ferro-magnetic polarizer, and is maintained through several micrometers of transport within the graphene channel. This leads to the manifestation of weak anti-localization away from the polarizer. This effect, which depends strongly and non-monotonically on carrier concentration, is ob-served as an increase in the conductance of the analyzer, emerging as structured peaks near zero-bias in the conductance map. These results therefore demonstrate the manner in which transport in graphene may be strongly impacted by non-locally induced spin-orbit coupling.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Data-Driven Modeling of Processes and Properties in Additive Manufacturing
No full textPh.D.Additive manufacturing (AM) was conceptualized as a tool for rapid-prototyping and visualization, but it soon emerged as a competitive manufacturing process. AM's main attraction is the potential to manufacture parts of complex geometries through a simple and highly repetitive process of layer-by-layer deposition. Although the process is repetitive and fully automated, the interactions between the layers during the deposition process are tightly coupled and far from being well quantified. Computational models of the processes are critically needed to unravel these interactions. However, the current state-of-the-art physics-based models are computationally demanding and cannot be used for any realistic optimization or process control. To address the high-cost challenge, we built a number of surrogate models (SM) to predict the output of the printing process at a significantly reduced computational cost. We use the physics-based models to generate the data, to train the SM models, and to inform the feature engineering design. The surrogate models (SM) serve as a proxy to the physics-based and experimental models to significantly lower the cost of prediction while providing high accuracy. We introduced a unique geometry featurization that is one of the key insights from this work. Rather than directly using the part geometry, we use the gcode and translate it into a set of features (local distances from heat sources, e.g., heated depositions, and sinks, e.g., cooling surfaces). This set of features is used as an input for the SM. Moreover, we leveraged the analytical solution to the moving heat source model to determine the heat influence zone (HIZ). The size of HIZ allows deciding a priori what should be the cardinality of the distance sets. We showed that for fused filament fabrication, the size of HIZ is small; thus, the number of input parameters for the SM is small as well. In this thesis, we build several surrogate models by increasing the input complexity of the geometry, printing path, as well as the output dimensionality of consolidation degree and thermal time series profiles. Through a comprehensive set of tests and analysis, we demonstrate the high predictive power and low computational cost of the models. Specifically, we demonstrated the capabilities of our models to predict the thermal history (time series data) and consolidation degree (scalar) for points at the interfaces between roads, with acceptable accuracy (error below 10.0%) in real-time. With such performance, the models open the possibility of optimization as well as process planning, and in-situ monitoring for closed-loop control.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Single-Priority Designing: The Impacts of Resilient, Affordable, and Lifespan Design Approaches in Urban Infill Housing
No full textMARCH/MUPObjectives: The goal of this thesis was to develop a deeper perspective on the impacts of prioritizing resilient, affordable, and lifespan (RAL) design values in detached, single-family houses. This process developed a basis through which to analyze tradeoffs between competing RAL metrics. Methods: First, the salient literature surrounding the methods of implementing RAL values was examined. Next, an annotated critique was conducted of a baseline design that attempted to equally balance these three values in one design. This critique facilitated the creation of three single-priority designs that examined how the individual prioritization of RAL in separate schemes would impact the design of houses. From these designs, adaptations to Buffalo's Green Code were considered through the lens of their abilities to enhance RAL within the single-priority designs. Findings: This type of prioritization allows single-house values to supersede neighborhood values. It would have a significant impact on the relationship between the form and function of house design, the lifestyle of residents through deviation from normative practices, and neighborhoods. Furthermore, although the Green Code has the potential to act as a balancing agent at the neighborhood scale to counteract single-value prioritization at the house scale, it is limited in its ability to control the finer details of design. Conclusions: This thesis engaged in an experiment that created visions of the extremities of single-focus methods of design, in the hopes of providing context for future work to consider the tradeoffs that designers make when values compete. In understanding the impacts of the upper limits of their work, designers can better work within them.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Revisiting the Matilda Effect in Communication Journals: A Citation Analysis of Journal of Applied Communication Research, Communication Education, and Communication Theory
No full textM.A.The Matilda Effect states biases exist toward female scientists in science and one form of bias is published articles with female first authors are cited less often than male authors. In order to revisit the Matilda Effect in communication, the paper reviewed citation patterns for articles (N=708) published in Journal of Applied Communication Research (JACR), Communication Education (CE) and Communication Theory (CT) in 10 years (2002-2006, 2012-2016). This analysis failed to find support for the Matilda Effect when analyses cover all three publications. There were no significant differences in citation numbers and self-citation numbers under the effect of sex after controlling published year in JACR and CE, but with CT, there was a Matilda Effect that articles with male first authors were cited more often than female first authors. An important finding is that journal significantly influenced the citations. Specifically, articles in CT were either cited or self-cited more often than those in JACR and CE. However, the proportion of female first authors did not increase over time, which is inconsistent with the results in previous studies.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
A Comparison of Acuros® XB Algorithm to AAA in Treatment Plans with Various Prostheses Present
No full textM.S.Two commonly used algorithms in radiotherapy dose calculations are Acuros® XB and Anisotropic Analytical Algorithm. These algorithms both attempt to calculate the dose in various mediums in the body accurately but do so in different ways. AXB is considered the superior algorithm because of its ability to model and calculate the dose better especially in areas of great inhomogenities, such as with metal present. The goal of this study was to create phantoms to mimic various situations where metal is commonly present in the body and determine to optimal assigned densities of the high-density materials for each case to more accurately model the protheses. The optimal densities were used in the treatment plan to compare the two algorithms' accuracy. Prosthetic devices (two hip, one knee, and dental implant set) were utilized to develop patient-like phantoms and CT image sets of each phantom was acquired. The phantoms were contoured in the treatment planning system and the prostheses were divided into components and assigned specific materials and densities. The treatment plans were calculated using both AAA and AXB algorithms and an iterative process to determine the optimal densities was performed. For the first hip prosthesis phantom two plans were generated, one with a single direct field and the other was a VMAT treatment plan that mimicked a clinical patient plan. The measured data was acquired using Sun Nuclear Corporation's ArcCHECK® and analyzed with the company's software, SNC Patient. The second hip prosthesis phantom utilized Sun Nuclear Corporation's MapCHECK® 2 and SNC Patient Software and two pieces of Gafchromic film to acquire the measured data. The knee and dental phantoms dose were measured with Gafchromic film and all film results were analyzed using Varian Medical Systems' DoseLab Pro software. The optimal assigned materials and densities were determined for each prosthesis set-up. The two metal-on-polyethylene hip prostheses phantoms yielded the same optimal assigned materials and densities, despite the varying sizes of the prostheses. The femoral head ("ball") optimal assigned material and density was Stainless Steel (7.24 g/cm3), the acetabular component ("socket") was assigned Aluminum (2.6113 g/cm3), and the femoral stem was divided and contoured into a "neck" and "body" region. The optimal material was Titanium Alloy, but the assigned densities differed. The "neck" region's optimal density was 4.31g/cm3 and the "body" was 4.4001g/cm3. The knee prosthesis was contoured into two regions: the metal and the plastic liner. The optimal material for the metal was Titanium Alloy (4.6144g/cm3) and the plastic liner was assigned PMMA (1.19g/cm3). The dental fillings were assigned Stainless Steel and a density of 8.0 g/cm3.The two algorithms yielded comparable pass rate results but typically AXB algorithm more accurately modeled the shape of the prostheses. The algorithm pass rates depended on how accurately the high-density materials were modeled in the treatment planning system. The accuracy of a treatment plan can be improved by contouring a prosthesis into regions to reflect the actual materials present and by assigning the appropriate densities to each region.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
The Role of Institutional Ownership in Firms' Leverage and Cost of Financing
No full textPh.D.This thesis has three essays, which are empirical studies at the intersection of institutional investors, fixed incomes, and firms' leverage and financing cost. The topic focuses on the incentives and constraints institutional investors face and the impact these investors have on security valuation, asset allocation, and firms' decision-making, such as those pertaining to capital structure or investment decisions. The analysis provides empirical evidence suggesting that institutional investors impact companies' decision making and asset allocation as well as financial market stability by being a source of uncertainty in the supply of capital, reducing agency problems, and generating spillover collateral and counterparty risk. In this dissertation, I describe all of these projects in detail. The first chapter examines how trading behaviors among institutional bond investors affect the cost of debt. Firms with a more significant percentage of long-term investors have lower debt costs. By contrast, high short-term ownership results in the uncertainty of capital supply and high debt cost. These findings are not driven by investors' or bonds' characteristics after using the investors' funding structure as an instrument. The results suggest that short-term investors' capital uncertainty results in fragility problems in the corporate bond market. Conversely, long-term investors play an important role in enhancing corporate governance, thereby lowering the cost of debt. The second chapter studies the role of stockholders and bondholders' renegotiation frictions in influencing firms' financial leverage. The result shows that firms tend to be overleveraged when shareholders have greater strategic advantages, and the problem of over-leverage weakens when institutional bond ownership is more concentrated. The effects of renegotiation frictions are not linear with the leverage ratio. Moreover, these effects are more pronounced when firms are riskier and have weaker governance. Consistent with the prediction of agency theory, renegotiation frictions play a critical role in determining leverage when the risk of managerial fraud is high or shareholder-creditor conflicts are more severe. The third chapter examines the effects of collateral quality, and counterparty risk on repo terms and yield spreads in the secondary MBS market. The paper uses the triparty deal-level repo and time-series aggregate data and finds that collateral and counterparty risks have significant effects on repo spreads and haircuts in the repo market. Importantly, risks in the repo market have a spillover effect on yield spreads of MBS in the cash market. Borrower-lender relationship mitigates the impact during the financial crisis. The Fed's QE policies reduce repo and MBS spreads through counterparty and collateral risk channels by helping attenuate both risks in funding markets.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Perceptions of Adversary Preferences for Accommodation and Coercion and Their Effect on International Conflict
No full textPh.D.Is war unavoidable? It is no mystery that the issues nations confront each other over sometimes result in violence and unnecessary human suffering, while others are resolved peacefully through diplomacy. The question as to why one and not the other has always remained at the forefront of international relations research. In fact, it was in the ashes of two world wars, of which destroyed over 100 million lives, where this inquiry earnestly began. Like so many before, this dissertation seeks to contribute to our collective knowledge – even if incrementally – as to why nations may traverse farther onto the path of war. I contend that the foreign policy decisions of nations seized by international disputes are influenced, in part, by perceptions of how inclined their adversaries are in resolving interstate issues through accommodation or coercion. Specifically, I hypothesize that nations rely on information derived from their adversaries' conflict history and public statements to respond in accordance with a strategy of reciprocation. This translates to nations pursuing more accommodative policies – like appeals for negotiation, mediation, or adjudication – when adversaries covey a willingness for such reconciliation processes, and more coercive policies – like the issuing of military threats, the imposition of economic sanctions, or the use of armed force – when adversaries project a predilection towards bullying. Through a survey experiment and statistical analysis of real conflict data, I provide evidence to support my theory.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Load Resistance of Monolithic Zirconia Screw-Retained Implant-Supported Single Crowns with Straight and Angulated Screw Channel Designs
No full textM.S.Recently popular angulated screw channel (ASC) crowns provide a prosthetic alternative to cement retention when implant trajectory is incompatible with traditionally designed screw- retained crowns. This study aimed to evaluate the fracture toughness of monolithic zirconia screw-retained implant-supported single crowns bonded to Ti bases with straight and angulated screw channel designs. A total of 30 maxillary left central incisor crowns were fabricated from monolithic gradiated translucency zirconia. Crowns were bonded to one of three corresponding Ti base interfaces (n = 10); 2 straight channel Ti bases (Straumann Variobase and Dess Ti Base) and 1 Ti base indicated for ASC crowns (Dess AngleBase). Mechanical loads up to 700 N were applied to the palatal aspect of assembled restorations at a 35° angle and force-displacement data were recorded. Excluding one instance of early failure, 100% of crowns remained intact through loading up to 700 N. Visible plastic deformation of the screw and Ti base was identified at this level of loading. At a load of 200 N, which is within the average clinical range for the anterior maxilla maximum force, the resistance to deflection of samples was not significantly different across all categories of Ti base (P > .05). In this study, incorporating an ASC into monolithic zirconia crowns did not impact fracture incidence at physiologic loading limit of 700 N when compared to similar restorations with traditional straight screw channels. The metallic components of screw-retained crowns appear to deform at loads below those where zirconia fracture is expected using the design studied.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
A Structural Study of Network Coding Through the Lens of Group Theory
No full textPh.D.Network coding is a network communication scheme which deploys encoding at intermediate nodes, and can significantly improve the network throughput. In this thesis, we study the network coding paradigm through the lens of algebraic group theory. Specifically, we study the notion of "group network codes" and in particular their performance in the context of the "edge removal problem". Group network codes are a special family of network coding schemes that are associated with an algebraic (group) structure. Group network codes generalize linear codes, however, unlike linear codes, are known to be (approximately) optimal in the sense that any network coding scheme can be approximated by a group network coding scheme. While group network codes offer an algebraically structured alternative to general coding schemes, not much is known on this special family of codes. In this thesis, we seek to better understand group network codes. Specifically, we present certain structural properties of group network codes (e.g., their error probability), alternative formalizations for Abelian group network codes (that resemble the operational formalization of linear codes), and we study group network codes in the context of the edge removal problem. The edge removal problem is a fundamental problem in network coding, which studies the impact on achievable source rates (capacity region) when a communication edge is removed from a given network. The edge removal problem is understood on a handful of special network instances and network coding schemes, however, similar to the explicit characterization of the network coding capacity region for general networks, which remains widely open, the edge removal problem in full generality is not well understood and is open as well. The optimality of group network codes gives rise to the possibility that it is sufficient to understand the edge removal problem when restricted to the context of group network codes in order to understand the edge removal problem in full generality. In this thesis, we study the edge removal problem in the context of group network codes. We solve the edge removal problem for Abelian group network codes (a class of group codes defined with Abelian groups), we show that solving the edge removal problem reduces to the problem of modifying given coding schemes in such a way that preserves rate while adding an algebraic group structure to very localized parts of the network, and extend these results in an axiomatic manner beyond group codes to more general families of encoding functions. The results presented in this thesis appear in part in: 1. Wei, Fei, and Michael Langberg. "The Effect of Removing a Network Communication Edge: Group Network Codes." 55th Annual Allerton Conference on Communication, Control, and Computing (Allerton). IEEE, 2017.2. Wei, Fei, Michael Langberg, and Michelle Effros. "A Local Perspective on the Edge Removal Problem." 2019 IEEE International Symposium on Information Theory (ISIT). IEEE, 2019.3. Wei, Fei, Michael Langberg, and Michelle Effros. "Towards an Operational Definition of Group Network Codes." arXiv preprint arXiv:2002.00781 (2020).**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*