Experimental Particle Physics: Research at the frontier of human knowledge, high technology, integration into a truly international community of scientists. Interested? The Heidelberg ATLAS group offers opportunities for students at various levels. Post-Docs and group members will be happy to talk to you.
Below, you can find some examples for topics of bachelor and master theses. For more information and other actual topics for Bachelor, Master or PhD theses please contact:
Please also have a look at the offers of our graduate schools!
The ability of the ATLAS detector to record events with two energetic jets is severely limited by the output bandwidth of the trigger system. As a consequence the sensitivity of the detector to mediator particles to a ‘Dark Sector’ with low masses is greatly reduced. To overcome this challenge, ATLAS data is recorded at high rate but with a much reduced set of detector information. This project, which can also be expanded into a Masters topic, is to understand the calibration of jets in this so-called trigger-level analysis and to develop new approaches to model the invariant mass spectrum for di-jet events in order to search for new mediator particles.
Unsupervised machine learning can be a very powerful tool for particle physics. Traditionally, searches for new physics at the ATLAS detector are optimized to enhance the experimental signatures for a benchmark set of theory models beyond the Standard Model. Machine learning allows us to break from this paradigm in that we can search the data for anomalous events that do not appear ’Standard Model-like’. This master thesis would involve the implementation of unsupervised machine learning on ATLAS data.
In addition to the Higgs discovery, the ATLAS detector at the LHC was designed to search for signatures of new phenomena beyond the Standard Model, such as Supersymmetry, extra dimensions and Dark Matter. During the first run of the LHC at a center of mass energy of 8 TeV, no hints of new physics have been seen. Also in the second run with a center of mass energy of 13 TeV no new physics has yet to be discovered. We are seeking a master student interested in developing a new innovative method to exploit the current data in order to search for new physics with high precision. This work would include studying the ratio between the 8 TeV and 13 TeV data sets in order to reduce the experimental uncertainties and to quantify the sensitivity of this approach to different new physics models.
In international co-operation, the KIP ATLAS group has developed and built the Level-1 Calorimeter trigger (L1Calo) of the ATLAS detector at the LHC accelerator, and the group has participated in the very successful operation during the first two LHC runs in 2009-2012 (run-1) and 2015-2018 (run-2). During the ongoing second long shutdown of the LHC, both the LHC and the experiments are preparing the machine and detectors for the higher luminosity and increased centre of mass energy expected for the upcoming LHC run-3, starting in 2021. For the ATLAS detector, a major upgrade of the trigger system is ongoing, involving the replacement of large parts of the front end electronics of the Liquid-Argon Calorimeter. The correct functionality of the electronics feeding the ATLAS Level-1 Calorimeter trigger needs to be validated as soon as the new hardware comes into operation. For this test runs will be taken throughout 2019, in which charge is injected into the calorimeter's electronics chain, mimicking the signals of particle collisions. These runs needs to be analysed with regard to the validation of correct cable mappings, absolute as well as correlated noise performance, signal timing, energy calibration, analogue saturation effects, etc. (Valid till mid 2019)
The search for final states with three bosons offers the possibility to access processes where four vector bosons couple, the so called quartic couplings. Since these processes are very rare, many of them are only accessible due to the large amount of data recorded by the LHC experiments. Furthermore a search for anomalous quartic couplings, i.e. couplings which are forbidden, e.g. ZZγγ, offers a powerful tool in the search for new physics beyond the Standard Model. Here we are interested in final states with two Z bosons and one photon. The goal is to analyse the full ATLAS run-2 data set and to observe this process for the first time. In order to separate the signal events from background machine learning techniques will be applied. The work is embedded in a group which already analyses a similar topology with one Z boson and two photons.
