Deep learning optimal molecular scintillators for dark matter direct detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cook, Cameron, Blanco, Carlos, Smirnov, Juri
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918251033788416
author Cook, Cameron
Blanco, Carlos
Smirnov, Juri
author_facet Cook, Cameron
Blanco, Carlos
Smirnov, Juri
contents Direct searches for sub-GeV dark matter are limited by the intrinsic quantum properties of the target material. In this proof-of-concept study, we argue that this problem is particularly well suited for machine learning. We demonstrate that a simple neural architecture consisting of a variational autoencoder and a multi-layer perceptron can efficiently generate unique molecules with desired properties. In specific, the energy threshold and signal (quantum) efficiency determine the minimum mass and cross section to which a detector can be sensitive. Organic molecules present a particularly interesting class of materials with intrinsically anisotropic electronic responses and $\mathcal{O}$(few) eV excitation energies. However, the space of possible organic compounds is intractably large, which makes traditional database screening challenging. We adopt excitation energies and proxy transition matrix elements as target properties learned by our network. Our model is able to generate molecules that are not in even the most expansive quantum chemistry databases and predict their relevant properties for high-throughput and efficient screening. Following a massive generation of novel molecules, we use clustering analysis to identify some of the most promising molecular structures that optimise the desired molecular properties for dark matter detection.
format Preprint
id arxiv_https___arxiv_org_abs_2501_00091
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Deep learning optimal molecular scintillators for dark matter direct detection
Cook, Cameron
Blanco, Carlos
Smirnov, Juri
High Energy Physics - Phenomenology
Materials Science
High Energy Physics - Experiment
High Energy Physics - Theory
Instrumentation and Detectors
Direct searches for sub-GeV dark matter are limited by the intrinsic quantum properties of the target material. In this proof-of-concept study, we argue that this problem is particularly well suited for machine learning. We demonstrate that a simple neural architecture consisting of a variational autoencoder and a multi-layer perceptron can efficiently generate unique molecules with desired properties. In specific, the energy threshold and signal (quantum) efficiency determine the minimum mass and cross section to which a detector can be sensitive. Organic molecules present a particularly interesting class of materials with intrinsically anisotropic electronic responses and $\mathcal{O}$(few) eV excitation energies. However, the space of possible organic compounds is intractably large, which makes traditional database screening challenging. We adopt excitation energies and proxy transition matrix elements as target properties learned by our network. Our model is able to generate molecules that are not in even the most expansive quantum chemistry databases and predict their relevant properties for high-throughput and efficient screening. Following a massive generation of novel molecules, we use clustering analysis to identify some of the most promising molecular structures that optimise the desired molecular properties for dark matter detection.
title Deep learning optimal molecular scintillators for dark matter direct detection
topic High Energy Physics - Phenomenology
Materials Science
High Energy Physics - Experiment
High Energy Physics - Theory
Instrumentation and Detectors
url https://arxiv.org/abs/2501.00091