Open Positions
Doctoral / Post-doc Positions
Current vacancies can be found here.
If you are interested, please contact Prof. Dr. K. Nikolopoulos(konstantinos.nikolopoulos"AT"uni-hamburg.de)
Bachelor / Master Theses
We offer a broad range of topics for Bachelor and Master theses throughout the year. Please contact us directly at Prof. Dr. K. Nikolopoulos(konstantinos.nikolopoulos"AT"uni-hamburg.de).
Current Master Thesis Topics:
[M] Exploring Sensitivity to H→Za with Hadronic Z and a Decays in Vector Boson Fusion Production
This thesis explores a search strategy for the fully hadronic final state of exotic Higgs decays H→Za in events where the Higgs is produced via vector boson fusion (VBF). The VBF topology, with two energetic forward jets and a large rapidity separation, provides an efficient trigger handle and strong background suppression, enabling sensitivity to hadronic decays that are otherwise difficult to record. The Z boson is reconstructed in its hadronic decay, and the light pseudoscalar a is also targeted in hadronic decay modes, leading to a characteristic event with two forward VBF jets plus central jets from the decaying bosons. The main challenge is the overwhelming QCD multijet background. The project will develop reconstruction and classification techniques to identify the Z and a candidates using jet kinematics and substructure, exploring resolved and merged regimes. Machine learning methods can be used to improve the assignment of jets to the Z and a candidates, and to enhance signal-to-background separation while keeping the invariant mass distributions suitable for robust signal extraction.
[M] Identification of hadronically decaying scalar bosons produced from Higgs boson decays in association with a Z boson using machine learning
This thesis investigates the identification of hadronically decaying scalar bosons produced in Higgs boson decays in events with associated ZH production. The Z boson provides a clean tag, while the scalar is reconstructed from its hadronic decay products, forming one or two jets depending on the event kinematics. The main challenge is separating this signal from large Standard Model backgrounds such as Z plus jets and top quark production. Machine learning classifiers are developed using event kinematics, jet substructure, and flavour tagging information. Different model choices are compared and validated with studies of performance, calibration, and stability in control regions, including checks for mass sculpting.
[M] Charm tagging algorithms exploiting exclusive decays of charged D* mesons
This project explores charm tagging algorithms that explicitly exploit the exclusive decay chain of charged D* mesons decaying to neutral D mesons and charged pions. The key handle is the charged “slow pion”, a low transverse momentum track relative to the jet axis that lies very close to the jet direction. This signature appears for prompt charm production in c jets, while in b jets it can arise from a BC cascade and is typically more displaced because the b hadron decays before producing the charm hadron. The project evaluates how the slow pion information can be integrated into a broader charm tagging strategy using modern machine learning, keeping the approach open to different model classes such as boosted decision trees, multilayer perceptrons, or graph-based methods.
[M] Simulation of gas properties and detector response for the Spherical Proportional Counter
The Spherical Proportional Counter (SPC) is a gaseous detector used for rare-event searches, in particular for the detection of low-mass Dark Matter. This master's thesis will focus on detailed simulations of gas properties and detector response for gas mixtures relevant to SPC operation, providing results that will inform the DarkSPHERE experiment, a next-generation Dark Matter search project, and guide realistic choices of gas mixtures as well as possible alternatives for optimal detector performance. The detector performance is strongly influenced by the properties of the gas mixture, including electron transport, attachment, diffusion, and gas amplification. Electron transport parameters will be calculated using Magboltz, while signal formation and gas amplification will be simulated using Garfield++. Realistic electric field configurations of the detector will be modelled using COMSOL or Gmsh to provide accurate input for the Garfield++ simulations. The dependence of detector performance on gas composition, pressure, electric field, and gas purity will be studied. The simulation results will be compared with available experimental data, and the impact of different gas mixtures on key detector parameters such as gain stability, energy resolution, and low-energy sensitivity will be evaluated.
[M] Data Analysis with the Proto-0 Demonstrator for the DarkSide-20k Experiment
The DarkSide-20k experiment, currently under construction at Italy’s Gran Sasso underground
laboratory, is set to become the world’s largest direct search experiment for Dark Matter. By
using ultra-pure underground argon as a target and advanced silicon photomultipliers for signal
detection, it will probe previously unexplored regions of the WIMP parameter space. This project
focuses on the Proto-0 demonstrator, a smaller-scale version of the full detector designed to test
and optimise data acquisition and analysis strategies, currently operating at the University of
Naples in Italy. The work involves hands-on analysis of experimental signals, optimisation of
analysis techniques, and contributions to the preparation for future detector operation. The
project offers the chance to collaborate with the University of Naples and an international team
of DarkSide-20k researchers, providing valuable experience in a large-scale, cutting-edge
physics experiment.
Current Bachelor Thesis Topics:
[B] Background Modelling for Higgs To Charm Searches in the Vector Boson Fusion Topology
This project studies data driven background estimation for an invariant mass search for H(cc) in the VBF topology. A measurement of H(cc) directly probes the Higgs coupling to charm quarks and helps complete the picture of Higgs Yukawa interactions. The VBF signature, with two forward jets and a large rapidity gap, improves sensitivity by suppressing backgrounds and enabling clean event categorisation. Even so, the dominant limitation remains the large, smoothly falling QCD continuum under the Higgs peak, so a reliable background model is essential for an unbiased and precise result. The study compares three approaches, a standard functional form fit in sidebands using established parametric families, a functional decomposition using smooth bases such as Bernstein polynomials or regularised splines with controlled complexity, and Gaussian process modelling that interpolates from the sidebands into the signal region with a principled uncertainty estimate, optionally as a smooth residual on top of a simple baseline.
[B] Improved calibration of CMS b-tagging algorithms with multi-variate identification of b-jets from top quark pair decays in dilepton final states
This thesis targets the b tagging calibration workflow in CMS using dilepton ttbar events, where a multivariate classifier assigns the two b jets from the ttbar decay. The main task is a systematic hyperparameter study of the current BDT, optimising jet assignment performance and stability, supported by standard validation and closure tests. In addition, the thesis will investigate an alternative model, for example a neural network trained on the same inputs, and compare it to the BDT in terms of efficiency, purity, robustness, and practical usability. As an optional extension, the project can explore whether the jet assignment score can be decorrelated from chosen ttbar system observables to reduce kinematic dependencies while preserving discrimination.
[B] Installation of a laser light source on the Spherical Proportional Counter and detector calibration
The Spherical Proportional Counter (SPC) is a gaseous detector used for rare-event searches, in particular for the detection of low-mass Dark Matter. During long-term operation, detector performance can be affected by gas contamination due to outgassing from detector materials, leading to a reduction of the effective gain through electron attachment to electronegative impurities such as oxygen. To monitor and study this effect, a pulsed ultraviolet laser system can be used. The fifth harmonic of an Nd:YAG laser (213 nm) induces photoelectric emission of single electrons from the aluminium vessel, producing well-defined signals suitable for low-energy calibration. The Bachelor's thesis will focus on the integration of the laser system via an optical feedthrough, the verification of signal detection in coincidence with a reference detector, and the calibration of the SPC response in the low-energy regime.
[B] Simulation, construction and characterisation of an ACHINOS sensor for the Spherical Proportional Counter
The Spherical Proportional Counter (SPC) is a gaseous detector used for rare-event searches, in particular for the direct detection of low-mass Dark Matter. The ACHINOS sensor, located at the centre of the detector, is responsible for signal formation and readout. The Bachelor's thesis will focus on detector simulations to study the dependence of the detector performance on construction parameters such as anode size and distance from the central electrode. Based on the simulation results, an 11-anode ACHINOS sensor will be constructed in a clean-room environment. The performance of the sensor will be characterised using radioactive sources, and the response of the individual anodes will be calibrated, including data analysis of the acquired measurements.