High-Energy Physics Parallel Talks

The spatial distribution of neutrons (neutron flux) is the most important quantity in nuclear reactor monitoring and diagnostics because it is proportional to local power density and fuel burnup. While traditional pressurized water reactors rely on detector placements throughout the core to monitor the spatially resolved flux distribution, new approaches must be designed for next-generation reactors whose compact designs and high temperatures render invasive core penetrations infeasible. To address this technology gap, a new framework called Green’s Kernel Reconstruction from Accessible Ex-Core Measurements (GRACE) has been formulated at Argonne National Laboratory. The aim of GRACE is to reconstruct the flux distribution throughout the reactor during real-time operation using ex-core detectors in conjunction with a machine-learned Green’s function and spatially resolved effective nuclear data derived a priori from high-fidelity reactor physics simulations.

This is accomplished by assuming the monoenergetic neutron diffusion (MND) equation as its governing model due to its favorable mathematical properties. In the MND, effective 1-group nuclear data forms the set of coefficients. The MND accuracy in describing reactor-wide neutron transport phenomena must be studied. The present work evaluates the variation of effective nuclear data as a function of control rod positioning and fuel burnup in the reference JSI TRIGA Mark-II reactor via high-fidelity Monte Carlo calculations using the OpenMC software. A general upper bound on effective nuclear data variation throughout a test reactor’s core is established at approximately 10%. In response, interpolation of this data across the reactor’s parameter space is implemented and analyzed.

Monte Carlo simulations give access to Euclidean-time correlation functions that can be used to study large quantum systems, including field theories like quantum chromodynamics (QCD). Lattice QCD provides a method of theoretical calculations for the strong subnuclear force.  Such simulations necessarily include some level of statistical uncertainty. Meanwhile, the physical requirement of causality implies stringent self-consistency conditions that these correlation functions must satisfy within the spectral decomposition. The constraints turn out to have a convex semi-definite geometric structure, which can be leveraged using established techniques from convex optimization. This line of thinking gives a different perspective on established results, including the success of fits to the truncated spectral decomposition. My talk and poster will describe ongoing work to study the number of states accessible within a given noisy correlation function.

With an expanding infrastructure at Very Low Earth Orbit (VLEO) altitudes, constraining the effects and interactions of ionospheric plasma has become a critical area of study. Joule heating, the result of collisional friction between plasma and neutral molecules, is a primary driver of energy and momentum exchange in the upper atmosphere that significantly influences our ability to monitor and track satellites in VLEO. However, due to a lack of observational data at lower orbital altitudes where it is zoned most intensely, the spatial and temporal scales of Joule heating are largely unknown. While advances in satellite engineering have made VLEO constellations dedicated to atmospheric observation feasible in the near future, there is a lack of study on the optimal orbital architectures required for such missions. We present an Observing System Simulation Experiment (OSSE) to evaluate high-latitude VLEO in situ Joule heating observations at 250 km altitude. Using the Coupled Thermosphere Ionosphere Plasmasphere Electrodynamics (CTIPe) model as a synthetic “ground-truth”, and NASA’s Community Coordinated Modeling Center’s (CCMC) Kamodo software to simulate satellite flythroughs, we analyze two distinct architectures across a range of scales: a linear “string-of-pearls” train and a 2D triangular formation. Our results demonstrate that logarithmically-spaced linear formations effectively capture multi-scale structures, while triangular formations capture 2D horizontal gradients yet exhibit a strict resolution threshold with increasing circumradii. Ultimately, our results indicate the critical importance of constellation geometries in achieving effective resolutions able to capture dynamic plasma structures in VLEO, and highlight the distinct sensitivity of the upper atmosphere.

In conjunction with NIST and the Ricochet particle physics experiment, this work addresses the physical challenge of tracking resonator frequency shifts that occur in response to applied magnetic flux. In this setup, particle detector signals produce magnetic flux differences that are encoded as frequency shifts within 2D resonators, which are located on a superconducting chip at the bottom of a dilution refrigerator.

Monitoring these frequency shifts across multiple parallel channels allows for the accurate readout of complex physics detector arrays. We developed a passive tone tracking system with a field-programmable gate array (FPGA) development board, the ZCU 216, to process these multiplexed signals.

By generating multiple frequency modulated waves in the megahertz regime, merging them with a polyphase filter bank, and then upconverting them to the gigahertz range, we follow these frequency shifts with high precision. Flux responses are converted to time-domain radio-frequency profiles using a flux ramp. With numerically controlled local oscillators on the ZCU 216, these radio frequency profiles can be generated in the megahertz regime with high frequency resolution.

Our results demonstrate successful simultaneous multi-tone generation, accurate frequency profiles, and passive tone tracking. This framework establishes a pathway to time-multiplex the architecture to support up to 64 tones for high-density detector arrays.

While the traditional approach to computing scattering amplitudes relies on Feynman diagrams, scalability quickly becomes an issue as the number of interacting particles increases. Therefore, newer approaches to calculating these amplitudes have been developed using Lie algebras. Using these algebraic methods, a closed-form expression for the one-loop gluon amplitude has been proposed and successfully verified against the standard one-loop single-minus four-gluon amplitude. In my research, I aim to extend this verification to five interacting gluons. I also speculate that the same symmetry used in the algebraic treatment of the four-gluon amplitude cannot be directly applied to the five-gluon case due to additional momentum-conservation constraints; however, a similar approach appears plausible. Finally, we outline our progress toward completing the full (n=5) verification.

My team does research on Liquid Argon Time Projection Chambers (LArTPCs), which are used in particle physics experiments to make precise measurements of particles, such as neutrinos. Neutrinos are neutral particles produced by a variety of cosmic and nuclear events, various experiments are active or being planned to study neutrinos in detail, such as the Deep Underground Neutrino Experiment (DUNE). Working in Professor Mike Mooney’s research group, we are investigating the effectiveness of Radon-220 injections for the calibration of LArTPC detectors. This research is vital for investigating methods of calibration for the DUNE detectors which will be huge (about 70ktons of Argon) and deep enough underground such that cosmic rays cannot be used for calibration. In our lab, at CSU we operate a small-scale LArTPC known as Colorado Argon Test Stand (CATS) which is one of the only such setups in the US, run by a university. With CATS we have demonstrated successful radon injection, and are currently analyzing data with the plan to publish a paper to showcase our findings and data soon. My involvement with CATS included assisting in detector assembly and operations as well as data analysis. My analysis is focused on identifying unstable components of the Radon-220 decay chain, like Bismuth-212 to Polonium-212, which is a prominent signal and is very useful for calibrating a detector. This project has given me an amazing opportunity to make significant contributions to huge particle physics experiments like DUNE, as well as learning skills in statistics, detector R&D, data collection, and data analysis.

Whistler-mode waves have been observed at Venus for 50 years. On Earth, whistler waves are one of three key signatures of lightning, as well as optical flashes and broadband radio emissions. However, optical flashes and broadband radio emissions are not observed at Venus, leaving whistler-mode waves as the primary evidence for lighting. Multiple observations of whistler waves at Venus led to the hypothesis that, like Earth, it may also produce lightning. On its mission to the Sun, Parker Solar Probe (PSP) passed Venus seven times during its Venus Gravity Assists (VGA’s). PSP offered a unique opportunity to analyze the origins of the whistler-mode waves at Venus. A previous study analyzed observed whistler-mode waves on VGA 4, finding that the waves were propagating planetward, a characteristic inconsistent with lightning. The on-board FIELDS search-coil magnetometer (SCM) and electric field power spectral data allow for a detailed analysis of the wave characteristics and Poynting vector directions. In this study, the FIELDS SCM and electric field data are used to identify and analyze whistler-mode waves during the seventh VGA of the mission and calculate the Poynting vector directions. The goal is to support or refute findings from VGA 4. Observations of the whistler-mode waves during VGA 7 indicate that planetward propagation is a recurring feature of the near-Venus space, strongly suggesting that the generation of whistler-mode waves at Venus can be caused by mechanisms other than lightning. These findings suggest that, unlike Earth, detecting lightning on Venus may require surface observation.

The Belle 2 experiment, coupled to the SuperKEKB collider in Tsukuba, Japan, aims to search for physics beyond the standard model using high precision measurements. One of the areas of interest is observing new particles that may be discreetly present in decays from tau leptons, exploiting the large number of tau leptons produced inSuperKEKB. The τ → ηπν decay has never been observed and it is expected to occur 1 out of every 10^5 tau decays;  it is a candidate for making the new physics visible due to its heavy suppression in the Standard Model. One of the main challenges for the discovery of τ → ηπν is the large number of background events: there are many non-suppressed tau decays that resemble τ → ηπν which makes correctly identifying this decay channel difficult. This project aims to increase the sensitivity of τ → ηπν in Belle II by testing various machine learning algorithms, discriminating signal events from background. The algorithm will focus on reconstructing the decay η → γγ, combined with a charged pion, supported by machine learning algorithms that will discriminate between these rare processes. The goal is to compare these models against each other in their ability to increase the significance of the signal as well as find the optimal variables to use as selection criteria to minimize the amount of background present. These results will aid future measurements that could lead to a discovery concerning New Physics.