ABSTRACTS for CLASS of 2026 SENIOR PROJECTS

as of October 1, 2026

These abstracts describe work that students PLAN to do in their projects; the abstracts for their final papers will generally be very different.

Nathan Alemu with Prof. Xuan Gao

P-type 2D InSe Field Effect Transistors

This project will investigate p-type behavior in two-dimensional indium selenide (InSe) field-effect transistors. The primary goal is to determine and characterize the conditions under which InSe exhibits p-type semiconductor behavior by studying the electrical transport properties of InSe devices. Measurements of device characteristics will be used to examine how charge carriers respond to applied voltages and to identify evidence of hole-dominated conduction. This work will provide a better understanding of the electronic properties of InSe and its potential use in semiconductor devices. 

Duncan Cayenne-McCall with Prof. Corbin Covault

Searching for discrepancies in Auger@TA event reconstruction methods

The goal of the Auger@TA experiment is to test Auger detector hardware at the Telescope Array in the northern hemisphere to better understand the discrepancies between the two. 

We aim to reproduce Auger@TA event calculations using the Auger derived central data acquisition system (CDAS), and compare these with values determined by the Auger Offline analysis system. We plan to compare energy, direction and other reconstructed event parameters.  One of the goals of the project will be to include results from the SSD scintillator detectors that have been recently deployed onto the Auger@TA stations in the field. We additionally want to check for self-consistency of results in the Auger analysis chain before we compare our results to those obtained independently by our collaborators on the Telescope Array (TA) experiment.

Ege Dalcan with Prof. Corbin Covault

Using a Muon Telescope to Increase the Resolution of a Single Photomultiplier Tube Triggered Cherenkov Particle Detector

Water-based Cherenkov detectors, like those used in the Pierre Auger Observatory, typically utilize coincidences between multiple photomultiplier tubes (PMTs) to filter out instrumentation noise and accurately resolve the muon energy peak from the detector’s muon histogram. However, in small or single-PMT detectors, the muon peak can get buried under the noise and have a distinct shelf-like look. This lack of resolution of the muon peak prevents precise fitting and detector calibration. This research will investigate the benefits of installing a muon telescope interposing the Cherenkov detector. By using the coincidence signal of the muon telescope to trigger the Cherenkov detector, we aim to increase the resolution of the muon histogram required for calibration.

Phoebe Elliot with Prof. Hari Padma

Machine learning identification of weak spectral features in RIXS experiments

Resonant inelastic X-ray scattering (RIXS), is a powerful spectroscopic technique to probe quantum materials.RIXS can detect collective excitations and ordered phases across the charge, spin, and orbital sectors, allowing for the reconstruction of effective Hamiltonians and identification of the interplay between these coupled degrees of freedom. However, much of this information is found in weak spectral features, a problem further exacerbated by the long acquisition times necessary for high-energy-resolution experiments. In this project, I will develop a deep learning neural network to identify weak spectral features in noisy RIXS data. The model will be trained on RIXS datasets from the PI’s prior work as well as publicly available datasets from published manuscripts. Our work will allow for more rapid analysis of experimental data and pave the way towards real-time steering of experiments. 

Vivian Ford with Prof. Idit Zehavi

The Influence of Splashback Galaxies on Angular Conformity

Galactic angular conformity claims that the satellite galaxies along the major axis of a central galaxy are more likely to be quenched than along the minor axis. This angular dependence has been studied in both observations and simulations, and has recently been shown to extend beyond the halo to galaxies outside the virial radius. There are two competing explanations for the cause of angular conformity. The first is internal AGN (active galactic nucleus) Feedback: the supermassive black hole at the center of the halo could be releasing large amounts of energy, causing anisotropic quenching. The second is external infall from filaments, the proposal that quenched satellite galaxies accumulate along cosmic filaments aligned with the central galaxy’s major axis. We propose to study the contribution of splashbacks to the angular conformity signal. Splashback galaxies are satellite galaxies that have already passed through the host halo and are moving back outwards. We will use TNG data, a cosmological hydrodynamical simulation, to model the relationship between angular conformity and splashback galaxies.

James Gomez Faulk with Prof. Corbin Covault

Improving Recorded VEM Value for Calibration of Ultra High Energy Cosmic Ray Data

The Auger@TA project uses Water Cherenkov Detectors (WCDs) to record Ultra High Energy Cosmic Ray (UHECR) detections with the aim of investigating the difference in results between experiments run in the Northern and Southern hemispheres. The primary signal used by Auger@TA to calibrate the WCD data is the Vertical Equivalent Muon (VEM), a charge signal from the photomultiplier tube (PMT) that results from Cherenkov light produced by a muon passing vertically through the detector. In practice, the VEM is determined as the peak of a statistical accumulation of single muon signals, referred to as the muon histogram. However, the presence of an underlying spectrum of non-linear noise in the muon histogram data can shift the peak, causing a misreported VEM and a miscalibration of WCD data. Using noise data recorded in the lab, the intention of my project is to create a method to filter the underlying noise data from the muon histogram to more accurately record the VEM.  I will then use the VEM value to calibrate the UHECR data recorded by our in-house particle detector as well as currently operating detectors deployed in the field as part of the Auger@TA experiment.

Seth Gratz with Prof. Jesse Berezovsky

Exploring Higher-Order Correlations in Musical Rhythm Using Statistical Mechanics

Rhythmic patterns in music can be represented by audible events occupying discrete bins in time which can be modelled with statistical physics. A mean field model has been developed where the probability of a rhythmic event depends on the probability of the occupancy of its surrounding time bins. However, in music, notes are not arranged in time independently of musical context. Instead, they occupy time with respect to previous notes, patterns, and musical emphasis. We use statistical physics as a map to develop a model of musical rhythm that incorporates these higher-order correlations through conditional probabilities governing the occupation of successive time bins. This may allow the model to describe empirically observed rhythms that are not explained by the mean field model, such as preferences for certain types of syncopation. By comparing the model to actual music compositions, we can investigate whether statistical mechanics can be used to describe the underlying structure of musical rhythm.

Mathew Gummere with Prof. Michael Martens

Developing a Course: AI/ML in Physics

This project develops the curriculum and materials for a new CWRU course in the Physics department that explores the intersection of AI and physics. The goal is to teach students what tools exist, how they’re applied, and how to select the right one for a given task. I’ll build the course firstly by surveying students and faculty and by collecting real experimental data and examples from faculty here, so the exercises may come from problems people in the department are actually working on. This will be augmented by other real data sets for students to work with outside of CWRU. A particular emphasis will be placed on “tool-leveling,” meaning selecting a tool that’s appropriately powerful for the job eg.: can a clustering algorithm handle this, or does the problem actually require a more expensive, black-box LLM? The course will cover a range of tools likely including but not limited to: regressions, neural networks, diffusion models, pattern recognition, and transformers. Each unit will include an introduction to the tool, a set of physics data to analyze, exercises using Python, and an examination of what the tool produced.

Julia Indyk with Prof. Michael Hinczewski

 

Yuelin Jin with Prof. Xuan Gao

Top Gated 2D Semiconductor Devices on Flexible Substrates

This project focuses on developing top gated 2D semiconductor devices on flexible substrates. The main goal is to replace the traditional rigid silicon substrate with a softer material so that the device can bend while still working properly. A challenge in the fabrication of gate-controlled semiconductor devices on flexible substrates is that the flexible substrate may not function well as a gate as conventional silicon substrates. To address this, the project will explore adding an extra gate on top of the device to enable device control and help maintain normal gate performance in devices like field-effect transistors.

Tanmay Kapoor with Prof. Lydia Kisley

Determination of Water Coordination in EF Hands of Lanmodulin Using Luminescence

Rare earth elements (REEs) have unique properties making them useful in electronics, healthcare, automotive, and other industries. Separation methods which isolate REEs from mined or recycled sources are crucial due to the high demand and supply risk of REEs. Lanmodulin (LanM) is a protein first discovered in the M. extorquens bacterium. LanM proteins exhibit extremely tight binding with REEs as proven by low (picomolar level) dissociation constants. LanM consists of four domains known as “EF hands” which are responsible for the specific binding with REEs. Amino acids, the building blocks of these EF hands, can be bioengineered to recreate sections of proteins as short peptides. Likewise, peptides derived by the EF hand sequences of LanM protein are known to selectively bind to REEs and understanding their binding mechanisms allows us to more adeptly apply these materials in REE separation applications. Previous studies determined water coordination as a major entropic contribution to peptide-REE binding affinities. Yet, the number of water molecules present in LanM peptide-REE complexes remains largely unknown. Our research aims to use fluorescence spectroscopy to measure the number of water molecules coordinated in the bound peptide-REE complexes for four REEs: europium, terbium, samarium, and dysprosium. Using a tryptophan antennae incorporated into the LanM peptides, we can amplify the luminescence of these complexes and use the excited-state decay times to determine the number of water molecules coordinated within these complexes. The findings will overall be used to better understand the thermodynamic contributions of binding affinities of REEs-LanM peptide complexes and provide insights into developing REE extractants with better performance.

Ian Lape with Prof. Shulei Zhang

Magnon-Plasmon Hybridization and Quantum Altermagnets

Collective excitations such as magnons (spin waves) and plasmons (collective electron oscillations) play a central role in modern condensed matter physics. In conventional materials, these excitations typically occur at very different energy scales and therefore interact only weakly. A recently identified class of magnetic materials known as altermagnets may fundamentally change this picture. Although altermagnets have zero net magnetization, their crystal symmetry generates unusual momentum-dependent spin splitting of electronic states, potentially enabling special spin-polarized plasmons whose energy scales overlap with magnons. We develop effective theoretical and computational models for understanding when and how magnons and spin plasmons may hybridize in altermagnetic systems.

Chloe Meyer with Prof. Sudha Chakrapani (Pharmacology)

Structural Studies of 5-HT3AR Intermediate States

5-hydroxytryptamine type 3 receptors (5-HT3R) are pentameric ligand-gated ion channels (pLGICs) that mediate fast excitatory neurotransmission in the gut and CNS. Hyperactivity of 5-HT3AR is associated with chemotherapy-induced nausea and vomiting, irritable bowel syndrome (IBS), and neuropsychiatric conditions like depression and anxiety. Setrons are competitive 5-HT3 antagonists that target the full pentamer and are widely used to treat chemotherapy-induced vomiting. However, adverse side effects like severe constipation limit their use in treatment of IBS and motivate an interest in drug design with partial agonists that may have fewer systemic side effects. Existing 5-HT3AR structures resolved with traditional Cryo-EM or X-ray crystallography describe apo, agonist-bound, and antagonist-bound states captured at equilibrium or steady-state populations on the minutes timescale. A millisecond resolution is required to visualize dynamic transition states, leaving the structural intermediates associated with rapid conformational changes that underlie channel gating, ion selectivity, and permeation uncharacterized. In this project, we use various biophysical approaches to resolve the structural description of 5-HT3AR in its activation intermediate states, elucidating the structural details of events that occur on the millisecond timescale.

Kazutada Nakao with Prof. Christopher Wirth (Engineering)

Effect of Cap Thickness on Dynamics of Electric Field Mediated Janus Particles Near a Boundary

Micro-scale active matter systems are of great interest in developing micro-cargo transport, drug delivery, and environmental applications. Active platinum-polystyrene Janus particles  self-propel near a boundary and can be studied as a model active matter system. One way to propel the particles is by applying an AC electric field (E = 83 kV/m, f = 0.1-2000 kHz) between two conductive boundaries, causing the particles to move via induced charge electrophoresis. We aim to understand how changing the cap thickness (7-40 nm) of the Janus particles will affect their behavior, such as frequency and voltage dependent speed and particle-particle interactions, like chain formation. We hypothesize that an increase cap thicknesses will have increased gravitational torque effects, leading to slower particle speeds. Optical microscopy is used to record the trajectories of the particles, which is used to find mean squared displacement and extract particle speed information. Agent-based simulations are leveraged to model particle-particle interactions along with effects of gravitational torque in viscosity-dominated transport, such as chain formation.

Rafaella Ortiz Cardenas with Prof. John Ruhl

Mechanical Modeling and Testing of Plastic Vacuum Windows for Millimeter-Wave Cameras

Cosmic Microwave Background (CMB) telescopes use vacuum windows to separate the cryogenic receiver systems from the external environment while allowing millimeter-wave radiation to reach the detector. As CMB instruments increase in size, larger-diameter windows are required. Increasing the window thickness for larger apertures improves structural stability, but also increases the absorptive losses and in-band emission, therefore decreasing instrument sensitivity. This project investigates the mechanics involved in the design of large-aperture High Density and Ultra-High Molecular Weight Polyethylene vacuum windows for future CMB cameras. The properties of the materials will be characterized and tested. Then, finite element analysis will be simulated to evaluate stress and deformation with different geometries. Experimental tests of deflection versus applied pressure over the window aperture and burst pressure will be compared with the simulation results to improve the accuracy of the models and determine the appropriate designs. The results will inform the choice of safe thickness for a given aperture size of these materials for future CMB cameras.

Manav Patel with Prof. Benjamin Monreal

Toward Direct Exoplanet Imaging: Simulation of a Proposed Exoplanet Telescope

There have been many proposed ground telescopes that could theoretically be used to image exoplanets. However cost has quickly become a limiting factor for most due to the large circular mirrors they require. Dr. Monreal proposed a cheaper alternative that would consist of a long rectangular mirror and a curved mirror that may be able to image exoplanets directly with a cheaper support structure. However, questions were raised about the actual imaging ability of such a telescope. This capstone project will be focused on simulating this new telescope idea in Ansys Zemax and creating simulated images to see if this telescope idea is worth investigating further.

Rohan Rajappan with Prof. Steve Hostler (Engineering)

Build and Test of a Liquid Rocket Vehicle and Propulsion System

This senior project focuses on the manufacturing, assembly, and test of a liquid-propelled rocket vehicle and propulsion system based on Half Cat Rocketry’s Mojave Sphinx architecture. This project would be the first time a complete vehicle is being constructed at Case Western Reserve University. As such, this project addresses a gap within the Case Liquid Propulsion Team, which has been primarily focused on propulsion systems rather than integrated systems. Topics covered consist of fluid and nozzle architecture, manual machining, hardware integration, and static test-fire preparations. Ultimately, this project will help establish a foundation for integration to support future liquid rocketry at CWRU.

Alexander Rollins 

 

Robbie Shepherd with Prof. Johanna Nagy

Developing Cryogenic Testing Capabilities for Cosmic Microwave Background Experiments 

 

Tejasvin Shrikanth with Prof. Lydia Kisley and Vignesh Venkataramani

Quantifying Expansion Anisotropy in Osmotic versus Tensile Expansion Microscopy: A Comparison of MAGNIFY and TExM

Expansion microscopy achieves super-resolution imaging by embedding a specimen in a swellable hydrogel and physically enlarging it, so that features below the diffraction limit become separable under an ordinary light microscope. This process relies on the assumption that the sample expands isotropically, preserving structure and shape so that nanoscale dimensions can be scaled directly by the expansion factor. Two distinct approaches can achieve this expansion: osmotic swelling (as used in protocols such as MAGNIFY), which relies on charge repulsion and osmotic pressure upon water hydration; and tensile expansion (TExM), which mechanically stretches a hydrogel using a custom expansion device. This project quantifies the magnitude and spatial structure of anisotropy in both osmotic and tensile expansion. Two-photon polymerized fiducial markers, printed using direct laser writing (Nanoscribe), are embedded in the hydrogel prior to gelation to provide a reference of known geometry against which local deformation is measured before and after expansion. The pattern utilizes ArUco identification markers so that position, rotation, and distortion can be resolved computationally rather than through manual landmark placement. The recovered marker coordinates are compared against an ideal similarity transform, with the resulting deviations yielding a map of local and large-scale anisotropy. These deviations are analyzed for magnitude, direction, and spatial correlation across the expansion process. Furthermore, this analysis is applied across variations in gelation chemistry, sample preparation, and tissue homogenization to evaluate how each factor influences anisotropy, ultimately contrasting the mechanical fidelity of osmotic swelling with tensile expansion.

Wanrou Sun with Prof. Michael Kesden (UT Dallas)

Correlation between Binary Black Hole Spin Precession and Remnant Black Hole Properties

One of the strangest predictions of Einstein’s theory of gravity is the existence of gravitational waves, which are ripples in the fabric of spacetime. Gravitational waves are emitted during black hole collisions, which happen in three stages: inspiral, merger, and ringdown. During inspiral, misalignment of the black holes’ spins causes precession and nutation of the orbital angular momentum about the total angular momentum. This phenomenon, known as spin precession, is described using five phenomenological parameters: precession amplitude, precession frequency, nutation amplitude, nutation frequency, and the variation in precession frequency. Following the final stage of the collision, the no-hair theorem states that the remnant black hole can be fully described by its mass, spin, and recoil or kick velocity. Using the Python module PRECESSION, we simulate binary black hole systems to investigate the correlation between the precession parameters and the properties of the final black hole.

Hazel Treumann with Robert Brown and Robert Deissler

A Unified Approach to Multiple Disease Components with Crystalline Character

      Crystals are associated with a variety of diseases, including malaria, gout, pseudogout, and kidney stones. In this project, we apply an established magneto-optical device to a unified study of these crystals. The magneto-optical device measures changes in light transmission through a sample as a magnetic field is applied, providing information about the crystals’ optical and magnetic properties. By comparing the absorption and scattering responses of these different crystals, we hope to find commonalities and general rules for the absorption and scattering cross sections. This approach could provide a foundation for using magneto-optics to identify crystalline components in disease processes, which may provide information for the development of new healthcare techniques.

Modeste Whitaker with Prof. Cyrus Taylor