113588 research outputs found
Sort by
The Influence of Size, Geometry, and Temperature on Deformation Mechanisms in Laser-Powder Bed Fusion Alloys
No full textThis study presents a comprehensive experimental investigation into the effects of size, geometry, and temperature on the mechanical behavior and deformation mechanisms of additively manufactured (AM) metal alloys produced via laser powder bed fusion (L-PBF). The work focuses on four major alloy systems, Ti-6Al-4V, Haynes 214, Haynes 282, and GRX-810, to elucidate how geometric scaling and temperature jointly influence strength, ductility, and the underlying microstructural processes governing deformation and failure in L-PBF metals. Through a combination of quasi-static mechanical testing, surface topography analysis, X-ray computed tomography (XCT), and electron microscopy (SEM/EBSD/EDS), the influence of intrinsic microstructural features such as porosity, grain morphology, and crystallographic texture on mechanical response was systematically characterized across multiple sample sizes and temperature regimes. In the first segment of this work, the mechanical response of L-PBF Ti-6Al-4V was investigated as a function of specimen thickness, geometry, and test temperature. These findings underscore the importance of accurately capturing surface topography and cross-sectional area variations in thin-wall structures to ensure reliable stress calculations and model calibration. Results revealed that thin-wall flat samples exhibited enhanced elongation but reduced strain hardening at room temperature, with this trend reversing at 250 °C and 450 °C. Round samples showed no change in ductility at ambient temperature; however, they experienced more pronounced ductility losses compared to flat samples as the sample diameter decreased. The increase in strain hardening with temperature, independent of geometry or sample size, is presumed to be attributed to the reduction in the critical resolved shear stress (CRSS) of basal slip systems, which promotes increased dislocation entanglement. The second phase of this study extended the investigation to the high-temperature L-PBF nickel-based superalloy Haynes 214. Here, strong size dependence was observed in the plastic flow behavior, particularly through the activation of the Portevin–Le Châtelier (PLC) effect and dynamic strain aging (DSA) at 650 °C. Thinner specimens exhibited increased serration frequency and reduced critical strain for onset, reflecting the sensitivity of DSA to grain boundary volume and solute diffusion kinetics. A pronounced ductility loss was observed between 600 °C and 870 °C, associated with the transition from transgranular plasticity to intergranular grain-boundary cracking driven by carbide segregation and oxidation. Above 870 °C, partial recovery of ductility occurred due to the onset of Orowan bypass mechanisms and the partial dissolution of the strengthening γ′ phase. Building upon these findings, the final stage of this work investigated temperature-dependent deformation behavior in two additional high-temperature L-PBF superalloys: Haynes 282 and GRX-810. Comparative analysis revealed that L-PBF Haynes 282 maintains higher tensile strength and work-hardening capacity than L-PBF Haynes 214 and L-PBF GRX-810 as temperatures increase up to 870 °C, owing to its leaner and thermally stable γ′ fraction. At 1093°C, L-PBF GRX-810 demonstrated superior high-temperature strength and ductility retention due to the stabilizing effect of its yttria-based oxide dispersoids and late-stage deformation twinning, respectively. Despite higher bulk defect densities, including voids and microcracks, GRX-810 exhibited minimal degradation in mechanical properties, indicating that the observed porosity had negligible influence on quasi-static performance. Across all alloys, the severity of the PLC effect was found to increase as sample size decreases at 650°C testing temperatures. The collective findings of this work highlight that the influence of size and geometry on mechanical performance cannot be universally generalized across AM alloys but must instead be understood within the context of specific microstructures, thermal histories, and alloys. The results provide a robust experimental foundation for advancing microstructure-sensitive constitutive models, including physics-informed and machine-learning–assisted crystal plasticity frameworks. Ultimately, this work enhances the fundamental understanding of how thin-wall AM structures deform and fail across a wide temperature range, guiding the optimization of L-PBF process parameters, alloy design strategies, and component qualification protocols for high-performance applications in aerospace, propulsion, and energy systems
Bridging the Theory and Practice of Interactive Imitation Learning
No full textImitation learning (IL) is a general learning paradigm for tackling sequential decision-making problems, leveraging offline expert demonstrations, interactive annotations, or both, with the aim of learning a policy that has performance competitive with the expert, using as few annotations as possible. This thesis focuses on interactive imitation learning, bridging theory and practice by developing provably efficient algorithms together with practical methods. Our contributions, organized by Chapters 4, 5, 6, and 7, are as follows: • First, under the realizable setting, we revisit a recent conclusion that Behavior Cloning (BC)—which relies solely on offline demonstrations—cannot be improved in general, and show that when annotation cost is measured per state, algorithms with interactive annotations provably outperform BC with improvements by up to a horizon factor. • Beyond the realizability assumption, we propose oracle-efficient algorithms for restricted policy classes that achieve improved sample and interaction-round complexity guarantees, complemented by a separate computational lower bound. • Next, we address computational efficiency for general policy classes by designing a new oracle-efficient algorithm and a practical variant, both achieving preferable performance
on continuous control benchmarks. • Finally, we introduce hybrid imitation learning (HyIL), where the learner leverages both offline demonstrations and interactive queries, and propose an efficient algorithm with both theoretical guarantees and empirical gains over purely offline or interactive methods. Together, these contributions substantially advance the statistical efficiency and computational tractability of interactive imitation learning, laying a strong foundation for both theoretical progress and practical impact
Balancing Innovation and Integrity: Australia's AI Ethics and Trust Regulation in Global Context [Article]
No full textArticleAs artificial intelligence (AI) becomes increasingly embedded in society, ensuring its ethical use and public trust is a global imperative. This paper critically
examines Australia’s approach to regulating AI ethics and trust, comparing it with frameworks in the United States, the United Kingdom, and the European Union.
While these jurisdictions adopt varied strategies—ranging from risk-based and sector-specific to principle-driven models—Australia relies primarily on voluntary
standards, such as the AI Ethics Principles and the Voluntary AI Safety Standard. Despite their intent, these frameworks lack enforceability, leading to inconsistent adoption and limited accountability. The paper highlights key ethical challenges, including privacy breaches, algorithmic bias, and the absence of legal safeguards in high-risk AI applications. It argues that Australia’s current regulatory landscape is insufficient to address the rapid evolution of AI technologies. To bridge this gap, the authors propose a meta-regulation approach—one that integrates legal oversight with organizational self-regulation, fostering both innovation and ethical responsibility. This model offers a flexible yet accountable framework for embedding ethical principles into AI development and deployment. The paper concludes by emphasizing the need for Australia to adopt a more robust, enforceable, and adaptive regulatory strategy to ensure trustworthy AI.This material published in Arizona Journal of International and Comparative Law is made available by the James E. Rogers College of Law, the Daniel F. Cracchiolo Law Library, and the University of Arizona Libraries. If you have questions, please contact the AJICL Editorial Board at http://arizonajournal.org/contact-us/
Filling the A.I. Gap: How Domestic and International Law Fails to Protect Artificial Intelligence Whistleblowers [Note]
No full textNoteAs artificial intelligence (A.I.) development accelerates beyond the reach of current regulatory frameworks, whistleblowers in the A.I. sector, particularly those employed by privately held firms, face a dangerous legal void. This Note identifies a critical regulatory shortfall, termed the “A.I. Gap,” where employees seeking to expose unsafe but not explicitly illegal A.I. practices are left unprotected under both U.S. and EU law. Through a detailed analysis of high-profile whistleblower cases, including the 2024 “Right to Warn” letter and disclosures by former OpenAI and Microsoft employees, the Note demonstrates how existing laws, such as the Dodd-Frank Act, the False Claims Act, and the EU Whistleblower Directive, fail to protect individuals who raise concerns about speculative or ethical A.I. risks. The Note also examines how non-disclosure agreements (NDAs) are strategically used to suppress internal dissent and limit legal recourse. Ultimately, this Note proposes a multi-step reform framework to protect AI whistleblowers across internal, governmental, and post-disclosure stages, emphasizing the need for confidential, responsive, and independent reporting channels; statutory redefinition of whistleblowing to include risk-related concerns; and robust anti-retaliation safeguards. Without these reforms, the public remains vulnerable to unaccountable A.I. development practices and the individuals best positioned to expose them remain silenced.This material published in Arizona Journal of International and Comparative Law is made available by the James E. Rogers College of Law, the Daniel F. Cracchiolo Law Library, and the University of Arizona Libraries. If you have questions, please contact the AJICL Editorial Board at http://arizonajournal.org/contact-us/
Building the Future of Law Libraries: Artificial Intelligence, Opportunities, and Advancement
No full textWhite paperThe Future of Law Libraries initiative convened six regional roundtables on Artificial Intelligence & the Future of Law Libraries with experts from academic, court, firm, and government law libraries, as well as allied professions, using scenario-building methodology to examine how AI is reshaping legal education, work, and systems and what law libraries must do to lead that change. The common message: legal information professionals must take an active, coordinated role in AI policy, training, and infrastructure or risk being sidelined as legal information vendors and non-library actors set the agenda.
This white paper distills convergent themes and proposes collaborative directions. It explores three recommendations that sprang from the roundtables: 1) create a centralized AI organization, 2) develop tiered training for legal information professionals, and 3) establish a shared knowledge hub. If we are successful in this next stage, we will have coordinated advocacy and standards, a workforce with more advanced skills, and an open, authoritative, dynamic, centralized repository. We will be convening teams to push these recommendations forward and we provide a link in the Call to Action section for our colleagues to join this effort.This item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at [email protected]
Soil Health: Sources Utilized in Plant Nitrogen Uptake
No full textThis article in the VegIPM Newsletter (Vol. 16, No. 24) explains that crops mainly rely on inorganic nitrogen and that organic forms play only a minor role in plant uptake.Documents in the Arizona Pest Management Center collection are made available by the Arizona Pest Management Center (APMC) and the University Libraries at the University of Arizona. For more information about items in this collection, please contact https://acis.cals.arizona.edu/about-us/arizona-pest-management-center
Mechanisms of Asthma Protection by Microbial Agents
No full textAsthma is the most common chronic disease of childhood and global prevalence has increased significantly in recent decades, coinciding with a shift in exposure to microbes present in the environment. Our lab previously showed that airway administration of microbial agents, such as Amish farm dust extracts (AFDE) or the bacterial lysate OM-85 (Broncho-Vaxom), protects against experimental asthma and has profound transcriptional effects in an ovalbumin murine model of experimental asthma. However, transcriptional mechanisms underlying the asthma protective effects of AFDE and OM-85 require further investigation. The work described in this dissertation sought to characterize the cellular and transcriptional modifications induced in the lung by AFDE and OM-85 using an Alternaria alternata extract-induced experimental asthma mouse model. Considering the phenotypic similarities, a comparison was performed to determine whether AFDE and OM-85 confer asthma protection by similar changes in transcription. Lung function and bronchoalveolar lavage (BAL) cellularity were measured in BALB/c wild-type-mice treated intranasally (i.n.) with either AFDE or OM-85 in the presence or absence of Alternaria extracts. Lung tissue collected from these mice was used to perform flow cytometry, bulk RNA-seq, and single nucleus (sn)RNA-seq to profile and compare the responses to AFDE and OM-85. This work showed that the protection conferred by AFDE and OM-85 involves the differential expression of several shared genes. Both microbial agents suppress airway hyperresponsiveness (AHR) and eosinophilia, downregulate genes involved in type-2 inflammation, and induce a strong, activated, polyclonal B cell signal. Ongoing studies in B cell deficient mice seek to determine whether B cells are required for the asthma protective effects of AFDE
Numerical Investigation of the Linear Stability Theory of Compressible Boundary Layers
No full textHigh resolution compressible Linear Stability Analysis (LST) was performed on compressible boundary layers to investigate the the primary instability for mach numbers (Me) from 0.1 to 15 chosen to capture all possible flow regimes with the objective to understanding the effect of mach numbers on maximum spatial amplification rates (-alpha) and N - Factors (N) and its associated effects. The simulations are modelled based on the report on Linear Stability Theory (LST) Mack, 1984, spatial theory is utilized as it is close to the real world cases. The primary instability was comprehensively investigated on a 7 degree half angle straight cone with Adiabatic and Isothermal wall conditions with Re = 11 x 10^6 (1/m). First, similarity solution solver was used to generate the base flow and then high resolution Linear Stability Solver (Haas, 2020) was utilized to obtain the spatial amplification rates and amplitude ratios. Similar to the Mack 1984's work, the influence of axisymmetric second mode and oblique mode such as first and third-mode on spatial amplification rates and amplitude ratios was explored, revealing that first mode is dominant in the early mach numbers, third mode shows its presence around Me = 9, axisymmetric second mode is dominant after Me = 4. Effect of mach numbers on wave-angles (psi) revealed that obliqueness of first mode and third mode. Furthermore, the influence of wall cooling and heating across different flow regimes showed that wall heatingdestabilizes the first mode, contrary to that, heating stabilizes the second mode. The effect of Mach numbers on viscous instability were examined, both first and second modes exhibit a destabilizing effect with regard to local Reynolds numbers
Rethinking Interpersonal Dependence
No full textAs finite and social beings we depend on one another for survival, flourishing, and for the mundane forms of assistance that allow us to move through our days. Understanding the resulting “web of social interconnection” is an essential part of the care ethical project. In this dissertation, I critique previous attempts to theorize this “web”, offer a novel account of interpersonal dependence, and explore the nature of the dependence involved in caring and loving relationships.In Chapter 1, “Dependency Relations Are Not (Necessarily) Need-Meeting Relations,” I provide negative arguments against the way dependency has traditionally been theorized in care theory – namely, as a need-meeting relationship.
In Chapter 2, “Depending on Others,” I offer a positive account of dependence itself. I analyze dependency as a relation in which someone normatively expects another person to perform work, and that second person countenances their expectations. I also address an obvious objection: what about the dependency of, for instance, newborn infants, who don’t appear capable of normatively expecting work of others? On a practice-based view of action, infants can expect insofar as they are meaningfully treated as participants in cross-cultural parenting practices.
Chapter 3, “Love, Fairness, and Sharing a Life,” concerns how dependency work is distributed in loving partnerships. I precisify and challenge the idea that considerations of fairness are out of place in relations of love. I argue that love and fairness are integrated in the sense that partners who “share a life” are only able to perform particular “relationally participatory” acts (loving and expressing love) if their actions are sensitive to fairness.
Chapter 4, “Rethinking Dependence and Care,” I reject Care Monism, the view that all idealized dependency relations are caring. Embracing Pluralism about dependency ideals allows us to explain the value of non-caring and non-intimate relations of “help”, and to grant disabled people greater control over the meaning of their relationships with personal assistants
Least Restrictive Environment: Beyond the Law – Policy, Bias, and the Power To Decide How White, Female LEA Representatives Interpret and Make Placement Decisions in Arizona Schools
No full textThis qualitative study investigates how Local Education Agency (LEA) representatives,predominantly white female educators, interpret and implement the Least Restrictive
Environment (LRE) from Individuals with Disabilities Education Act (IDEA) when making
placement decisions for Black, Indigenous, and People of Color (BIPOC) students in Arizona
schools. This study uses systems theory and social constructionism as theoretical frameworks to
examine the connections among vague legal mandates, institutional constraints, resource
disparities, and personal bias in shaping educational placements. Through semi-structured
interviews and vignette analysis, findings reveal that decisions are often influenced more by
district policy, systemic pressures, and personal perspectives than student-centered equity
concerns. Participants’ understanding of LRE differed, with many reverting back to compliance-
driven approaches that perpetuate exclusionary practices and reinforce racialized outcomes. The
research outlines how linguistic considerations, cultural expectations, and legal ambiguities
shape special education placements that disproportionately affect students from historically
underrepresented group. The study calls for more coherent policy guidance, equity-focused
professional development, and a critical reimagining of inclusive education that foregrounds the
lived experiences of underrepresented students. These findings have important implications for
legal reform, educational leadership, and advocacy seeking to dismantle systemic barriers to
equitable placement in special education.Release after 12/01/202