Engineering topics
Zhang, Wei
Publications and source records attributed to Zhang, Wei.
Steam-Induced Coarsening of Single-Unit-Cell MFI Zeolite Nanosheets and Its Effect on External Surface Brønsted Acid Catalysis
Commonly used methods to assess crystallinity, micro-/mesoporosity, Brønsted acid site density and distribution (in micro- vs. mesopores), and catalytic activity suggest nearly invariant structure and function for aluminosilicate zeolite MFI two-dimensional nanosheets before and after superheated steam treatment. Yet, pronounced reaction rate decrease for benzyl alcohol alkylation with mesitylene, a reaction that cannot take place in the zeolite micropores, is observed. Transmission electron microscopy images reveal pronounced changes in nanosheet thickness, aspect ratio and roughness indicating that nanosheet coarsening and the associated changes in the external (mesoporous) surface structure are responsible for the changes in the external surface catalytic activity. Finally, superheated steam treatment of hierarchical zeolites can be used to alter nanosheet morphology and regulate external surface catalytic activity while preserving micro- and mesoporosity, and micropore reaction rates.
Fine-tuned order-order phase transitions in giant surfactants via interfacial engineering
Not Available
Kinetic pathways of ionic transport in fast-charging lithium titanate
Fast-charging batteries typically use electrodes capable of accommodating lithium continuously by means of solid-solution transformation because they have few kinetic barriers apart from ionic diffusion. One exception is lithium titanate (Li 4 Ti 5 O 12 ), an anode exhibiting extraordinary rate capability apparently inconsistent with its two-phase reaction and slow Li diffusion in both phases. Through real-time tracking of Li+ migration using operando electron energy-loss spectroscopy, we reveal that facile transport in Li 4+x Ti 5 O 12 is enabled by kinetic pathways comprising distorted Li polyhedra in metastable intermediates along two-phase boundaries. Our work demonstrates that high-rate capability may be enabled by accessing the energy landscape above the ground state, which may have fundamentally different kinetic mechanisms from the ground-state macroscopic phases. This insight should present new opportunities in searching for high-rate electrode materials.
Modularly Constructed Polyhedral Oligomeric Silsesquioxane-Based Giant Molecules for Unconventional Nanostructure Fabrication
Controlled assembly of nanoscale building blocks is a promising approach to obtain functional materials with unique properties. In this paper, we report a way to manipulate the supramolecular structures of giant molecules based on discotic triangle cores and isobutyl polyhedral oligomeric silsesquioxanes (BPOSS) nanoparticles (NPs). It is found that depending upon the number of BPOSS at the periphery of the discotic cores, the packing of these nanoscale components (discotic core and POSS) could be manipulated into either cylindrical or Frank–Kasper (F–K) A15 ($Pm\overline 3n$) phases. The formation of these supramolecular nanostructures is mandated by the balance between the stacking of the discotic cores and the steric hindrance effect of the BPOSS NPs. This strategy to manipulate the packing of nanoscale building blocks for different supramolecular nanostructures including the fabrication of cylindrical structures and A15 ($Pm\overline 3n$) phases may be extended to other nanoscale building blocks for future development of materials with complex structures as well as tailored functionalities and properties.
Bonding similarities and differences between Y–Sb–Te and Sc–Sb–Te phase-change memory materials
The scandium (Sc) – alloyed Sb 2 Te 3 phase-change alloy has recently been found to enable ultrafast crystal nucleation due to the formation of Sc-stabilized octahedral motifs in the amorphous phase, rendering cache-type phase-change memory feasible. When yttrium (Y) is added, however, non-octahedral bonding patterns form in the amorphous Sb 2 Te 3 -based network even though Y has a valence electron configuration similar to that of Sc and also forms perfect octahedral bonding environments with tellurium in the YTe crystal. Here we elucidate the origin of this difference between Sc and Y, by carrying out thorough ab initio simulations and orbital-based bonding analyses on amorphous Y–Sb–Te and Sc–Sb–Te compounds. We also demonstrate how the smooth overlap of atomic positions (SOAP) similarity kernel can be used to quantify the structural similarity of local motifs in the amorphous phase with respect to various crystalline yttrium and scandium tellurides, both in the nearest-neighbor shell and beyond. We discover that the bonding contrast of Y- and Sc-centered structural motifs in amorphous Sb 2 Te 3 stems from their parent crystals at high Te concentrations. The larger atomic radius of Y and the weaker charge transfer when bonded with Te is found to allow more Te neighbors and cause a more open bonding environment, leading to higher coordination numbers and non-octahedral environments. We discuss the implications of the different local environments for practical applications in memory devices.
Filament-assisted mode conversion in magnetized plasmas
At density filaments in magnetized plasmas, electromagnetic waves are guided along the filament ("wave-filament bound states"). Several wave-filament bound states exist beyond those predicted. Here, the new bound states occur under experimentally relevant conditions, and are especially common under typical NSTX conditions, where they are a plausible mechanism for the experimentally observed power losses in the edge plasma.
Wake Structure Induced by Seal Whisker: Effects of Angle of Incidence
Seal whiskers have been found to produce unique wake flow structures that minimize self-induced vibration and reduce drag. The cause of these wake features are due to the peculiar three-dimensional morphology of the whisker surface. The whisker can be defined as an elliptical cross section with variation in the major and minor axis of the ellipse along the length of the whisker as well as rotation of the elliptical plane with respect to the whisker axis, angle of incidence. While the effects of several dominant parameters of the whisker morphology have been studied, the effect of the rotation of the elliptical plane has not been well understood. This paper will examine the influence of the angle of incidence on the wake flow structure through series of water channel studies. Several models of whisker-like geometries will be tested which isolate the rotation angle as the only variation between models. The wake structure behind each seal whisker model will be measured through particle image velocimetry (PIV). The different models’ wake structures will be compared identifying the effect of angle of incidence on the wake structure. Angle of incidence was found to influence the wake structure through reorganization of velocity patterns, reduction of recovery length and modification of magnitude of Tu. These results help provide a more complete understanding of the seal whisker geometry relationship to wake structure and can provide insight into design practices for application of whisker geometery to various engineering problems.
Computational Process Modeling for Additive Manufacturing (OSU)
Powder-Bed Additive Manufacturing (AM) through Direct Metal Laser Sintering (DMLS) or Selective Laser Melting (SLM) is being used by NASA and the Aerospace industry to "print" parts that traditionally are very complex, high cost, or long schedule lead items. The process spreads a thin layer of metal powder over a build platform, then melts the powder in a series of welds in a desired shape. The next layer of powder is applied, and the process is repeated until layer-by-layer, a very complex part can be built. This reduces cost and schedule by eliminating very complex tooling and processes traditionally used in aerospace component manufacturing. To use the process to print end-use items, NASA seeks to understand SLM material well enough to develop a method of qualifying parts for space flight operation. Traditionally, a new material process takes many years and high investment to generate statistical databases and experiential knowledge, but computational modeling can truncate the schedule and cost -many experiments can be run quickly in a model, which would take years and a high material cost to run empirically. This project seeks to optimize material build parameters with reduced time and cost through modeling.
Computational Process Modeling for Additive Manufacturing
Computational Process and Material Modeling of Powder Bed additive manufacturing of IN 718. Optimize material build parameters with reduced time and cost through modeling. Increase understanding of build properties. Increase reliability of builds. Decrease time to adoption of process for critical hardware. Potential to decrease post-build heat treatments. Conduct single-track and coupon builds at various build parameters. Record build parameter information and QM Meltpool data. Refine Applied Optimization powder bed AM process model using data. Report thermal modeling results. Conduct metallography of build samples. Calibrate STK models using metallography findings. Run STK models using AO thermal profiles and report STK modeling results. Validate modeling with additional build. Photodiode Intensity measurements highly linear with power input. Melt Pool Intensity highly correlated to Melt Pool Size. Melt Pool size and intensity increase with power. Applied Optimization will use data to develop powder bed additive manufacturing process model.
IPM10: Computational Process Modeling for Additive Manufacturing (OSU)
No abstract available
Parametric Studies of Flow Separation using Air Injection
Boundary Layer separation causes the airfoil to stall and therefore imposes dramatic performance degradation on the airfoil. In recent years, flow separation control has been one of the active research areas in the field of aerodynamics due to its promising performance improvements on the lifting device. These active flow separation control techniques include steady and unsteady air injection as well as suction on the airfoil surface etc. This paper will be focusing on the steady and unsteady air injection on the airfoil. Although wind tunnel experiments revealed that the performance improvements on the airfoil using injection techniques, the details of how the key variables such as air injection slot geometry and air injection angle etc impact the effectiveness of flow separation control via air injection has not been studied. A parametric study of both steady and unsteady air injection active flow control will be the main objective for this summer. For steady injection, the key variables include the slot geometry, orientation, spacing, air injection velocity as well as the injection angle. For unsteady injection, the injection frequency will also be investigated. Key metrics such as lift coefficient, drag coefficient, total pressure loss and total injection mass will be used to measure the effectiveness of the control technique. A design of experiments using the Box-Behnken Design is set up in order to determine how each of the variables affects each of the key metrics. Design of experiment is used so that the number of experimental runs will be at minimum and still be able to predict which variables are the key contributors to the responses. The experiments will then be conducted in the 1ft by 1ft wind tunnel according to the design of experiment settings. The data obtained from the experiments will be imported into JMP, statistical software, to generate sets of response surface equations which represent the statistical empirical model for each of the metrics as a function of the key variables. Next, the variables such as the slot geometry can be optimized using the build-in optimizer within JMP. Finally, a wind tunnel testing will be conducted using the optimized slot geometry and other key variables to verify the empirical statistical model. The long term goal for this effort is to assess the impacts of active flow control using air injection at system level as one of the task plan included in the NASAs URETI program with Georgia Institute of Technology.
Reinforcement learning in scheduling
The goal of this research is to apply reinforcement learning methods to real-world problems like scheduling. In this preliminary paper, we show that learning to solve scheduling problems such as the Space Shuttle Payload Processing and the Automatic Guided Vehicle (AGV) scheduling can be usefully studied in the reinforcement learning framework. We discuss some of the special challenges posed by the scheduling domain to these methods and propose some possible solutions we plan to implement.