VTechWorks
VTechWorks provides global access to Virginia Tech scholarship, including journal articles, books, theses, dissertations, conference papers, slide presentations, technical reports, working papers, administrative documents, videos, images, and more by faculty, students, and staff. Faculty can deposit items to VTechWorks from Elements, including journal articles covered by the University open access policy. Email vtechworks@vt.edu for help.
Open Access Policy
Virginia Tech's open access policy enables researchers to deposit the accepted version of scholarly articles with no embargo.
Theses and Dissertations
Virginia Tech was first in the world to require ETDs in 1997, and continues to add scans of older theses and dissertations.
Open Textbooks
More than 50 freely available and openly licensed textbooks are among our most downloaded items.
Communities in VTechWorks
Select a community to browse its collections.
Recent Submissions
Molecular Structure and Thermodynamics of CO2 and Water Adsorption on Mica
Aybar, Mert (Virginia Tech, 2025-12-12)
The adsorption of CO2 and water on clay surfaces plays a key role in applications such as gas storage in saline aquifers and depleted hydrocarbon reservoirs but is not yet fully understood. Here, the adsorption of CO2 and water vapor is studied using Grand Canonical Monte Carlo and molecular dynamics simulations. At a bulk pressure of 100 bar, pure CO2 adsorbs strongly on mica and forms extensive layers next to it. CO2 adsorption is lowered substantially if introducing water vapor above mica and is largely eliminated when the relative humidity (RH) approaches about 60%. When pure water vapor is introduced above a mica surface, a sub-nm thick liquid water film develops on it to form apparent liquid-solid and liquid-vapor interfaces simultaneously. Using the ITIM (Identification of the Truly Interfacial Molecules) analysis, how individual water layers develops in this film as RH increases is delineated. It was discovered that the water film is spatially heterogeneous, and the true liquid-vapor interface emerges only at an RH of 60-80%. Introducing 100-bar CO2 into the water vapor above the mica surface modulates water adsorption nonlinearly: at RH = 0.01%, the water adsorption is reduced by ≈30%; as RH increases, the reduction is weakened, and eventually, enhancement of water adsorption by about 7% occurs at RH = 90%. These variations are attributed to the interplay of film thinning by high-pressure CO2, competition of mica surface sites by CO2 molecules, and the energetic and entropic stabilization of interfacial water by CO2 molecules.
This work has important implications for subsurface energy and environmental technologies. For example, it indicates that, in depleted unconventional reservoirs, assuming a completely dry environment can grossly overestimate the CO2 storage contributed by adsorption on mineral surfaces. Furthermore, it suggests that the swelling of clay in very low relative humidity can be suppressed by the displacement of interstitial water by high-pressure CO2, which can compromise the mechanical integrity of caprocks in underground hydrogen storage sites employing CO2 as cushion gas. These implications warrant experimental studies in the future.
Adversarial Déjà Vu: Jailbreak Dictionary Learning for Stronger Generalization to Unseen Attacks
Dabas, Mahavir (Virginia Tech, 2025-11-19)
Large language models (LLMs) have achieved remarkable performance across a wide range of tasks, yet they remain vulnerable to jailbreak attacks that bypass safety guardrails and elicit harmful behavior. Defending against such attacks is particularly challenging when adversaries introduce novel strategies that differ from those observed during training or redteaming. Existing alignment and adversarial training approaches often struggle to generalize to these unseen attacks, as they are fundamentally limited by the distributions of prompts and behaviors present in their training data. This thesis addresses this challenge by rethinking jailbreak robustness through a compositional and data-centric lens. We introduce the Adversarial Déjà Vu hypothesis, which posits that ostensibly novel jailbreak attacks are rarely new in kind, but instead arise from recombinations of a finite set of recurring adversarial skills. To investigate this hypothesis, we conduct a large-scale temporal study of jailbreak attacks spanning multiple years and develop an automated pipeline to extract adversarial skills from attack prompts. To reduce redundancy while preserving explanatory power, we compress these skills into a compact Jailbreak Dictionary using sparse dictionary learning, yielding a set of interpretable adversarial skill primitives. Through temporal cutoff experiments, we demonstrate that attacks released after a given cutoff can be effectively explained as sparse compositions of skill primitives learned from earlier attacks, providing empirical support for the compositional structure underlying jailbreak evolution. Building on this insight, we propose Adversarial Skill Compositional Training (ASCoT), a training paradigm that improves robustness to unseen attacks by explicitly training models on diverse compositions of adversarial skill primitives rather than on isolated jailbreak instances. ASCoT expands coverage of the adversarial skill space while maintaining fixed data scale, enabling principled generalization to novel attack compositions. Empirical evaluations across multiple open-weight language models and a broad suite of jailbreak benchmarks show that ASCoT substantially reduces harmful behavior on unseen attacks, including multi-turn jailbreaks, while preserving general capabilities and avoiding excessive over-refusal. Collectively, this thesis reframes jailbreak robustness as a problem of compositional generalization over adversarial skills rather than memorization of attack instances. By uncovering the reusable structure of jailbreak strategies and leveraging it for data-centric training, this work provides both conceptual insight into the evolution of adversarial attacks and practical defenses for building safer and more robust language models.
Optimal Control in Aircraft Trajectory Planning
Kim, Heejin S. (Virginia Tech, 2025-12-03)
Trajectory planning for autonomous and piloted aircraft must satisfy not only geometric constraints but also full six-degree-of-freedom dynamics, actuator limits, and mission- or safety-driven performance objectives. This thesis investigates optimal control as a refinement tool for trajectory-planning problems that are typically addressed using geometric or search-based methods. Two case studies are considered. The first addresses loss-of-thrust contingency landing for a fixed-wing aircraft, formulating a nonlinear optimal control problem that minimizes a population-density-weighted ground risk metric, with and without an explicit time penalty, subject to a six-degree-of-freedom Cessna 182 model and operational constraints. Initial trajectories are provided by discrete search and Dubins paths, and the refined solutions reduce risk and/or flight time while remaining dynamically feasible and trackable in closed-loop simulation. The second case study develops an energy-optimal trajectory generation framework for a Lift+Cruise QuadPlane small uncrewed aircraft system. Starting from spline-based reference paths, a six-degree-of-freedom QuadPlane model and power-based cost function are used to compute dynamically feasible trajectories that include hover--to--cruise mode transitions and multi-waypoint maneuvers. Results show consistent reductions in power and energy consumption, smoother attitude and angular-rate histories, and explicit characterization of trade-offs among energy use, maneuver aggressiveness, and operational constraints. Collectively, these case studies demonstrate how optimal control can bridge fast geometric planning and high-fidelity dynamics, supporting preflight design of risk-aware and energy-efficient trajectories for both contingency and nominal operations.
New focus on cardiac voltage-gated sodium channel β1 and β1B: Novel targets for treating and understanding arrhythmias?
Williams, Zachary J.; Payne, Laura Beth; Wu, Xiaobo; Gourdie, Robert G. (Elsevier, 2025-01)
Voltage-gated sodium channels (VGSCs) are transmembrane protein complexes that are vital to the generation and propagation of action potentials in nerve and muscle fibers. The canonical VGSC is generally conceived as a heterotrimeric complex formed by 2 classes of membrane-spanning subunit: an α-subunit (pore forming) and 2 β-subunits (non–pore forming). NaV1.5 is the main sodium channel α-subunit of mammalian ventricle, with lower amounts of other α-subunits, including NaV1.6, being present. There are 4 β-subunits (β1–β4) encoded by 4 genes (SCN1B–SCN4B), each of which is expressed in cardiac tissues. Recent studies suggest that in addition to assignments in channel gating and trafficking, products of Scn1b may have novel roles in conduction of action potential in the heart and intracellular signaling. This includes evidence that the β-subunit extracellular amino-terminal domain facilitates adhesive interactions in intercalated discs and that its carboxyl-terminal region is a substrate for a regulated intramembrane proteolysis (RIP) signaling pathway, with a carboxyl-terminal peptide generated by β1 RIP trafficked to the nucleus and altering transcription of various genes, including NaV1.5. In addition to β1, the Scn1b gene encodes for an alternative splice variant, β1B, which contains an identical extracellular adhesion domain to β1 but has a unique carboxyl-terminus. Although β1B is generally understood to be a secreted variant, evidence indicates that when co-expressed with NaV1.5, it is maintained at the cell membrane, suggesting potential unique roles for this understudied protein. In this review, we focus on what is known of the 2 β-subunit variants encoded by Scn1b in heart, with particular focus on recent findings and the questions raised by this new information. We also explore data that indicate β1 and β1B may be attractive targets for novel antiarrhythmic therapeutics.
Gap junctional and ephaptic coupling in cardiac electrical propagation: homocellular and heterocellular perspectives
Wu, Xiaobo; Payne, Laura Beth; Gourdie, Robert G. (Wiley, 2025-05-31)
Electrical communication in the heart is crucial for maintaining normal cardiac function. Traditionally, gap junctional coupling between cardiomyocytes has been accepted as the primary mechanism governing electrical propagation in the heart. However, numerous studies have demonstrated that gap junctions are also present between different cell types in heterocellular structures and disruption of such gap junctional coupling can be associated with cardiac dysfunction. In addition to gap junctional coupling, ephaptic coupling has been proposed as another mechanism for electrical communication between cardiomyocytes. Reducing ephaptic coupling has been shown to have negative impacts on cardiac conduction. While the existence of ephaptic coupling between different types of cardiac cell is under investigation, a recent study suggests that ephaptic coupling at heterocellular contacts between cardiomyocytes and fibroblasts may provide a proarrhythmic substrate in cardiac disease. In this review, we examine the current literature on electrical communication in the heart, including gap junctional and ephaptic coupling in homocellular and heterocellular contexts. Further, we offer a perspective on gaps in knowledge and opportunities for further advancing our understanding of electrical coupling mechanisms in action potential propagation in the heart. (Figure presented.).