Geometry Modeling

To answer design questions for the proposed PROSPECT-II detector, I built a parameterized model of it in GEANT4. Simple options allow the users to construct any array of optical segments they wish. Easily modifiable geometrical parameters are given to accommodate different design parameters, such as window spacing and PMT housing thickness, with the detector resizing itself seamlessly to fit any user options.

Optical Simulation

The optical model was then fine-tuned using data to produce a realistic detector response. A curved PMT surface (blue) records every photon that touches it. The scintillator, reflector panels, light guides, and other detector components are all modeled to reflect their real-life optical parameters. The model is highly optimized and can simulate upwards of 50,000 photons per second on a single thread, even on personal machines.

Analysis Framework

Any capable simulation framework also requires a robust analysis pipeline. Detailed results from the optical model can be output either as ROOT or CSV files. For those more comfortable with Python, I wrote analysis code utilizing Pandas to quickly study the impact of design decisions on adverse effects like crosstalk. I also have ROOT code that performs similar functions as well as predicting the detector response with histogram look-up tables.

Neutrino Directionality

An experimentalist's PhD isn't complete without analyzing real data, so my next project was getting to work on reconstructing the average antineutrino direction from the HFIR (High Flux Isotope Reactor) core at Oak Ridge National Laboratory. The PROSPECT collaboration has been able to produce what we believe is the world's most precise reconstruction of reactor antineutrino direction, despite the significant challenge of missing segments in the PROSPECT-I detector. The paper has been submitted to PRD, and is currently undergoing review. Here's the arxiv link if you want to read through it!

(Results of plot subject to change)

Extracting Systematics

Similarly, no experimental paper is complete without a thorough exploration of possible systematic biases. Using background radioactivity present in the PROSPECT-I detector, I verified our directionality methodology while also calculating systematic uncertainties that might be present. The ²¹⁴Bi→²¹⁴Po decay results from the decay chain of the ubiqituous gas, ²²²Rn. Applying the same directionality techniques to this decay allows us to measure the precision of our position reconstruction.