Dissertation Defense Announcements

This dissertation explores the perception of social skills in structured virtual interviews, with a focus on how verbal, nonverbal, and vocal signals influence Social Skills Perception (SSP). The study explores the impact of specific behaviors on observers' perceptions of social skills and why particular behaviors are associated with low social skills perception scores. A central finding is the "just right" effect, which reveals that both excessive and insufficient displays of certain behaviors can negatively affect SSP, emphasizing the importance of balanced social skills demonstration in structured interviews. The findings contribute to social skills literature by building on the Heggestad et al. (2023) Social Skills Framework and highlight the complex dynamics of SSP in professional settings, particularly in structured interviews. This study suggests that such interviews might not fully capture exceptional social skills, offering insights into interview practices and evaluating social competencies in the workplace. Future research is encouraged to explore SSP across various contexts and cultures to deepen our understanding of these phenomena.

Energy storage plays a crucial role in addressing the growing demand for energy and electricity while simultaneously reducing greenhouse gas emissions. Most of the power infrastructure in the U.S. heavily relies on water-cooling technology, leading to significant freshwater withdrawals. To mitigate high water withdrawal rates and the thermal pollution of water sources, an alternative solution involves implementing dry cooling towers (DCT) or air-cooled condensers (ACC). However, the effectiveness of dry cooling techniques depends on the dry bulb temperature of the ambient cooling air, resulting in a plant performance penalty equivalent to approximately a 2%-point efficiency loss compared to wet cooling.
The current research focuses on designing a cost-effective latent heat cold thermal energy storage (LCTES) system to enhance the performance of DCT/ACC during the summer months. This is achieved by storing cold energy during the nighttime in inexpensive materials like phase change materials (PCM), such as CaCl2 hexahydrate or CC6. To guide the LCTES design, a numerical analysis of the melting and solidification processes of PCM within the tube array was conducted. Transient two-dimensional Navier-Stokes equations and a Realizable k-ɛ turbulence model were used to predict fluid flow and heat transfer in LCTES heat storage modules. The enthalpy-porosity technique was employed to model PCM melting and solidification.
The numerical results show excellent agreement with experimentally obtained values. The resulting design successfully met the predefined performance criteria, achieving a cooling effect of 4 °C for a four-hour duration while maintaining a pressure drop of less than 100 Pa. The proposed prototype-scale tube array design can efficiently cool the incoming ambient air, and PCM in the LCTES can be fully frozen overnight. The energy storage density of the system falls within the range of 22 to 27 kWh/m3, with the maximum energy efficiency reaching around 75% during the system charging and discharging processes. Apart from its primary focus on coal power plant dry cooling technology, the suggested concept can also be used for industrial, commercial, and residential applications, including concentrated solar power (CSP).

Graph coloring is commonly used to schedule computations on parallel systems. Given a good estimation of the computational requirement for each task, one can refine the model by adding a weight to each vertex. Instead of coloring each vertex with a single color, the problem is to color each vertex with an interval of colors.
Stencil graphs appear naturally in the parallelization of applications, where the location of an object in a space affects the state of neighboring objects. Rectilinear decompositions of a space generate conflict graphs that are 9-pt stencils for 2D problems and 27-pt stencils for 3D problems. We show that the 5-pt stencil and 7-pt stencil relaxations of the problem can be solved in polynomial time. We prove that the decision problem on 27-pt stencil is NP-Complete. We evaluate the effectiveness of several different algorithms experimentally.
Executing graph algorithms in a parallel or distributed context is a challenging problem. It is possible that the algorithm picks a partial order with long chains, which limits its utility to parallel applications. We investigate how distributed dataflow graph algorithms obtain a partial order and how one could favor orders with shorter long chains. We study the behavior of these different algorithms on randomly generated RMAT graphs and real-world graphs. We show that our ordering methods can significantly reduce the length of the longest chain.

Growth mechanism study is important to achieve high-quality materials growth using more convenient approaches and to realize controlled growth with compositional and structural tunability. In vapor-based deposition, vapor-solid (VS) and vapor-liquid-solid (VLS) processes have been two classic mechanisms for the growth of micro- and nano-scale structures. The VS process is a non-catalyst growth controlled by vapor supersaturation, while the VLS process is a catalyst-assisted growth initiated and guided by eutectic particles. This research reported a co-growth of three-dimensional (3D) Si crystals with two-dimensional (2D) Si2Te3 crystals and explored its growth mechanism. Chemical vapor deposition (CVD) method has been employed using Te and Si powders as the source materials in the presence of a Cu-coated Si substrate. The growth mechanism study reveals that the Te source plays two different roles in the growth. First, it serves as a reactant which vaporizes and reacts with Si powders to yield 2D Si2Te3 growth via the VS mechanism. A unique “liquid epitaxial growth” was discovered that Te droplets formed prior to the Si2Te3 growth could promote a quasi-epitaxial growth of Si2Te3 crystals on a lattice mismatch substrate. Second, the Te serves as an unstable catalyst for the 3D Si growth. The Si growth is promoted by the synergistic effects of Cu and Te: (1) Cu as a stable catalyst facilitates the reaction of Te vapor species with the Si substrate, forming ternary Te-Cu-Si eutectics; (2) due to the instability of Te, the Te-Cu-Si eutectic particles evaporate and release Si vapor as the precursor for the VS growth of Si crystals. This intermediate process is dubbed as a vapor-liquid-vapor (VLV) process which provides a new approach for the material growth with lower growth temperature, lower cost, and higher compatibility for device fabrications.