Exotic Long - Lived Particles

A search for hadronising long-lived massive particles at the Large Hadron Collider is conducted with the ATLAS detector. No excess events are found. Based on statistical analysis, upper limits on the production cross section are observed to be between $0.01$ pb and $0.006$ pb for colour octet particles (gluinos) with masses ranging from $300 \ \mathrm{GeV/c}^2$ to $1400 \ \mathrm{GeV/c}^2$, and $0.01$ pb to $0.004$ pb for colour triplet particles (stops and sbottoms) with masses ranging from $200 \ \mathrm{GeV/c}^2$ to $900 \ \mathrm{GeV/c}^2$. In the context of Supersymmetry with decoupled sfermion and sboson sectors (Split-SUSY), this gives a lower limit on the gluino mass of $989 \ \mathrm{GeV/c}^2$, and $683 \ \mathrm{GeV/c}^2$ for the stop mass and $618 \ \mathrm{GeV/c}^2$ for the sbottom mass. In addition, a new method is presented that improves the speed ($\beta$) estimation for long-lived particles in the ATLAS tile calorimeter with a factor of $7$ improvement in resolution at low-$\beta$ and a factor of $2$ at high-$\beta$. An additional advantage of the new method is that the $\beta$-response is flat within the range of acceptance ($0.2 < \beta \leq 0.9$) with an average $\beta$-resolution of $2.2 \%$. The resolution of the speed estimate using the ATLAS pixel tracker has been improved by a factor of $3$ at low $\beta$ to a factor of $2$ at higher $\beta$. The new method also introduces a flat response compared with previous methods. Furthermore, a simulation of magnetically charged monopoles is conducted. Based on simulation, magnetic monopoles with Dirac charges $g_D > 10$ will predominantly be trapped in the LHC beam-pipe if produced within ATLAS. Two regions are identified as optimal for the extraction of the beam-pipe in a magnetometer-based search. Finally, a contribution has been made to the proposed IceCube low-energy extension called PINGU which will improve the neutrino sensitivity to $E_{\nu} \sim 1 \ \mathrm{GeV}$. Motivated by indirect searches for Dark Matter annihilation in the sun, a likelihood based reconstruction method is developed that allows neutrino interaction vertex, energy, and direction estimation at this energy. The method provides a fast estimate based on a newly developed parametric model and multi-dimensional nested sampling. In addition, two neutrino flavour detection algorithms are developed which allow separation of charge-current muon events from other neutrino events.