Variance reduction techniques for Monte Carlo simulations of photon transport in liquid xenon detectors

Abstract

This work is an attempt to improve simulations of background signals in a liquid xenon dark matter experiment. In an experiment such as XENON1T one aims to measure a Dark matter signal. Since such signals have not been recorded, and therefore an effort is made to increase the sensitivity of the experiment. One strategy is by scaling up the experi- ments and thereby increase the active volume of the experiment. With larger detectors, simulations of background signals become more inefficient and will be computationally expensive. However, with advance variance reduction techniques, it is possible to improve the simulation efficiency by three orders of magnitude. This project is centred around one of such a method and provides insight into the working of the method as well as the validation thereof. Chapter 1 presents an introduction into dark matter and the experimental evidence to support the dark matter hypothesis. Chapter 2 describes the XENON dark matter experiment and the working thereof in more detail. Chapter 3 provides a description of gamma-ray transport and interactions in matter. Chapter 4 introduces an MC method to simulate gamma-rays in a liquid xenon vessel. In chapter 5, a paper is presented that is the result of this research and gives an accurate description of the accelerated MC techniques. Lastly, a conclusion and discussion are provided in chapter 6.