Xerox Engineering Research Fellows
2020 Research Opportunities
Research Projects: Static and Dynamic Structural Analysis of Heritage Masonry Buildings
My research and teaching interests are in computational solid and structural mechanics, in the development of engineering practices in antiquity, and in the study of heritage buildings in earthquake-prone areas. Ongoing research projects open to qualified undergraduates are in the structural analysis of historic monumental masonry buildings subjected to earthquake loading.
My principal collaborators are University of Rochester Professor Chris Muir, ME, and Professor Michael Jarvis, History; and Professor Christopher DeCorse, Anthropology, Syracuse University.
Engineering undergraduates participating in these projects are trained in the application of fundamental modeling techniques widely used for research and product development in many areas of modern engineering: solid modeling reconstruction of complex geometries, 3D FEM linear and nonlinear analysis, kinematic limit analysis simulating 2D and 3D collapse mechanism. Students will be encouraged – and guided - to submit conference and journal papers based on their research results.
Research Project #1: Seismic Vulnerability of the Elmina Castle
Determine the constructions history, assess the damage state, and determine the seismic vulnerability of the Elmina Castle, Ghana (1482, Portuguese, Dutch, English). This project requires participating in the 2020 Summer Field School in Engineering and Archaeology of Heritage Buildings of West African, May 24 – June 27, in Ghana. Built in 1482 by the Portuguese Crown, St. Jorge Castle at Elmina is the oldest permanent structure introduced by Europeans in Sub-Saharan Africa. Recognized as a UNESCO World Heritage Site, the Elmina Castle is a monument of extraordinary importance for understanding four centuries of interactions between West Africa, Europe, and the Americas beginning with the late 15th century and culminating with the Atlantic Slave Trade of the 17th and 18th centuries. Furthermore, due to its meticulous planning and continuous restoration, the building itself is the best-preserved and most complete example of European late-medieval masonry construction transplanted in Sub-Saharan Africa. Besides the survey and solid modeling reconstruction of the entire castle, engineering students are involved in the structural evaluation and seismic analysis of distinct building elements within the castle, such as curtain walls, brick vaulted rooms, and towers. This work involves applying nonlinear static and dynamic finite element models (FEM) to complex 3D problems in structural mechanics characterized by quasi-brittle materials.
Research Project #2: Structural Response and Conditions for Structural Collapse of the Frigidarium of the Bath of Diocletian
Determine the structural response and the conditions for structural collapse for the Frigidarium of the Bath of Diocletian in Rome (298-305 AD) subjected to a horizontal acceleration. This is a gigantic vaulted structure built of unreinforced pozzolanic concrete on which we have already done extensive linear and nonlinear finite element modeling (FEM). We have also applied kinematic limit analysis based on the damaged configurations predicted by the nonlinear FEM models. Since 2010, the Xerox Fellowship has supported six students working on the static and dynamic analysis of Roman concrete vaults, leading to several journal and conference (national and international) publications.
Research Project #3: Structural Response and Conditions for Seismic Collapse for Concrete Vaulted Structures Built by the Maya in the Puuc Region of Yucatan, Mexico
Determine the structural response and the conditions for seismic collapse for concrete vaulted structures built by the Maya in the Puuc region of Yucatan, Mexico (Late Classic, ~1,000 AD). Puuc architecture is characterized by the usage of excellent lime concrete with mechanical and physical properties similar to modern Portland concrete. Maya vaulted structures – ranging from relatively small halls inside temples or palaces to large gateway arches – are often incorrectly assumed to behave structurally as corbelled arches, held in equilibrium by superimposed projecting stone blocks. In reality, due to their inner solid concrete core, Maya vaults behave structurally like an elastic continuum, quite similar to Roman concrete vaults. We are aiming at investigating the static and dynamic (seismic) response and failure mechanisms of typical Puuc vaults using FEM models and kinematic limit analysis.
Research Project: Mechano-transduction of the Inner Ear Sensory Organ
We study the mechano-transduction of the inner ear. In the cochlea, mammalian hearing organ, mechanical stimuli (sounds) are encoded to neural signals. The identification of mechanical properties of cochlear sensory cells and tissues is crucial to better understand how we hear (or fail to hear). To measure the mechanical properties, we need to apply calibrated pressures in the order of milli Pascals and measure displacements in nanometers at the frequency of up to tens of kHz. Students will participate in measuring mechanical responses of artificial and biological micro structures in a micro-fluidic device. Through this project, students will learn how the principles of acoustics, fluid dynamics, solid mechanics and vibrations are applied to micro-mechanical experiments with biological tissues. Students will gain experiences with vibration measurements, imaging and data acquisition devices.
Research Project: Oceanic Circulation Using Satellites
There are two components that regulate Earth's climate. These are fluid systems: (1) the atmosphere and (2) the ocean, which can transport heat from the equator toward the poles, thereby maintaining a habitable planet. This project involves working with satellite observations of oceanic flow that have been revolutionizing research in oceanic fluid dynamics over the past decade. The student will have the opportunity to learn and work with Python and/or Matlab, learn about and work with satellite datasets of the global ocean that is collected by NASA, ESA and other space agencies. The primary project objective is to implement novel analysis and diagnostics that our group has been developing on this data to understand some of the intriguing aspects of oceanic flow that is currently the subject of active research in the oceanography community. There is also the opportunity to port this data into different formats, so that it can be incorporated into the "Science on the Sphere" repository.
Research Project: #1 Nanomechanical Deformation of Solids D2/T2 at Cryogenic Temperatures
This project intends to examine the plastic deformation of solid hydrogen (D2/T2) at cryogenic temperatures near the triple point (about 19 K), for application in the optimization of capsule fuel targets for inertial confinement fusion. When isomechanical dimensionless groups for D2 are compared to other face-centered-cubic (FCC) and hexagonal-close-packed (HCP) materials, D2 behaves like the FCC and HCP counterparts in just about all aspects of their mechanical behavior, including elastic anisotropy, activation energy, and diffusion. The goal here is to identify the nucleation and dislocation multiplication mechanisms, as they are driven by the constraint of the plastic shell around D2/T2 and the imposed subcooling. We need to study the nanomechanics of dislocation/defect generation in HCP D2/T2, by using molecular dynamics simulations. We specifically are interested in understanding how defects generate and grow during subcooling and growth of ice layer. The students will investigate the effect of cooling rate and identify the role of dislocation activities on the grain boundary formation and phase transformation phenomena during the subcooling process. Through this project, students will learn about the fundamentals of phase transformation, solid mechanics and plastic deformation at the atomistic level. The students acquire skills on using the software LAMMPS for molecular dynamics simulation and they often need to write their own codes for analyzing the raw data.
Research Project: #2 Capturing Nanoscale Lattice Variations by Deep Learning of Synthetic X-Ray Diffraction Data
At extraordinary high pressures, like those at the laboratory for laser energetics (LLE), materials can undergo extreme real-time structural changes. For example, Aluminum can change from facecentered cubic to hexagonal closest packed to body-centered cubic and even more complex structures at higher pressures. New phase nucleation/transformation processes are poorly understood at rapid (nanoseconds or less) timescales due to temporal and spatial limitations of conventional pump-probe measurements. X-ray crystallography is currently the most successful method for determining the real-time structural changes in materials. The technique consists of shining a bright, collimated x-ray beam through a material of interest; detailed information about structural order is then inferred from the far-field pattern of scattered rays. The scattering images contain visual features, such as rings, spots, and halos, which encode detailed information about the size, orientation, and packing of atoms, molecules, and nanoscale domain. However, timeresolved diffraction images are data-intensive, include many uncertainties and strongly depend on the accuracy of the hand-crafted features, and hence depend on the researcher's skill to infer structural changes by digging those data. Thus, the goal of this research is to develop automated deep learning techniques to mine such information-rich data to filter and detect lattice-level mechanisms responsible for phase transformation and deformation. Through this project, student will exploit large-scale non-equilibrium molecular dynamics simulations to generate synthetic xray diffraction data. Students will learn about the fundamental concepts of solid mechanics, probability theory, and essentials in neural network architecture. The students will acquire skills on preparing the training data, using the deep learning tools for implementing the model, training model on GPUs and using visualization methods for interpreting the data.