
Dr. Susanne Westhoff
Assistant Professor High Energy Physics
This is our weekly seminar on particle physics at the intersection of theory and experiment. It takes place during term time on Tuesdays from 13:30 - 14:30 in the Hilbertruimte, HG 02.802.
We discuss topics related to our research in a relaxed setting, with lots of time for discussion. If you would like to give a talk, please contact me: susanne.westhoff@ru.nl.
The unique design of the LHCb detector, with its flexible trigger and precision vertex tracker, offers the possibility to search for long-lived particles with low masses and short lifetimes, in complementarity with other general-purpose detectors at the LHC. In this talk, I will discuss the searches that have been performed at LHCb and their future prospects. Although searches for low-mass LLPs decaying hadronically are particularly challenging, they can benefit from the upgrade of the LHCb’s online trigger. I will also discuss suggested phenomenological models for these searches, as well as the opportunities available during the LHC Run 3.
B meson decays are important players in the search for physics beyond the Standard Model (SM) of particle physics. Linked to the antimatter-matter asymmetry in the universe, a key interest in this is also improving the understanding of the mechanism of CP violation within the SM. The large amount of data gathered by the B factories and LHCb allows testing the SM with an unprecedented precision, probing scales much higher than the reach of direct searches at the LHC. In this sense, B mesons serve as a Telescope in the search for new interactions and particles. To do so, requires precise and reliable theoretical predictions. In this talk, I will present some of the challenges and new ideas to push the theoretical precision up. Specifically, I will focus on the determination of the CKM element Vcb, Vub and how to handle non-leptonic decays.
Since the discovery of the Higgs boson in 2012 at CERN, particle physicists have studied its properties with higher and higher precision. If any significant deviation from the Standard Model predictions is present in the experimental data, it would be an important hint of new physics. To perform these studies, we use the Higgs decay into two photons, which, despite the very low branching ratio, represents one of the cleanest channels experimentally. In this talk, I will present the most recent measurements - and their interpretation - of the Higgs production cross sections in the diphoton decay channel using proton-proton collisions recorded by the ATLAS detector at the LHC.
In many extensions of the Standard Model, the electroweak phase transition is first order. Such a phase transition proceeds via the formation and collision of bubbles. The bubble collisions can source a stochastic gravitational wave background signal, with characteristic frequency right in the sensitivity band of LISA. We can thus use data from gravitational wave experiments to probe physics beyond the standard model. In this talk, I will focus on the contribution to the gravitational wave signal from sound waves that get formed in the interactions between the plasma and the bubble walls. I will discuss the parameters that describe the phase transition, and argue that the speed of sound plays a significant role. To quantify the importance of this effect, I will discuss the use of effective field theories for an accurate description of the phase transition parameters. I will then present the results of a recent accurate computation of the sound speed in a representative model.
In PTOLEMY, we aim to develop a detector for the Cosmic Neutrino Background. This is done measuring the endpoint of the electron energy spectrum of tritium decay in great detail. Current experiments are reaching the limits of energy resolution for measurements with molecular tritium, and to improve the resolution even further we want to work with atomic tritium, which we will attach to a graphene base. Currently we are testing hydrogenation of graphene samples to gain understanding of this process before using tritium. In this talk I will introduce the PTOLEMY experiment and show the contribution of our group at Radboud.
It is starting to become apparent that physicists can use Large Language Models (like the model underlying ChatGPT) for a variety of tasks, such as helping with coding simulations, accurately summarising papers, helping to frame experimental designs, aiding hypothesis generation, or simply asking information retrieval questions. In our project we are evaluating the capabilities and limitations of these large models in the context of how they might increase or extend scientific understanding. This includes questions such as how to conceptualise/test possible discoveries made by these models, what are good ways to use these tools to increase the efficiency and effectiveness of scientific research, and whether these models at some point might contain scientific understanding in some meaningful way such that they might be able to translate this understanding onto human scientists. We are currently creating a framework useful towards generating understanding tests, which can be used to measure the degree of scientific understanding of agents. This can serve as a benchmark for different models, but it can also serve to establish whether a human can increase their understanding after interacting with these models.
In our efforts to ensure that no hint of physics beyond the Standard Model (BSM) eludes us, the possibility and exotic signatures of long-lived BSM particles (LLPs) have recently attracted the attention of the HEP community. We discuss our studies exploring and addressing the various challenges faced from the very beginning at the triggering stage to dedicated search analyses. We also plan to discuss the multitude of new proposals and experiments coming up to explore the lifetime frontier.
Neutrino telescope experiments are rapidly becoming more competitive in indirect detection searches for dark matter. Neutrino signals arising from dark matter annihilations are typically assumed to originate from the hadronisation and decay of Standard Model particles. I will discuss a supersymmetric model, the BLSSMIS, that can simultaneously obey current experimental limits while still providing a potentially observable non-standard neutrino spectrum from dark matter annihilation.
The (phenomenological) minimal supersymmetric SM predicts the existence of a supersymmetric partner particle to the tau; the supersymmetric tau particle, or stau for short. In my master’s internship I simulate the ATLAS detector at CERN to find possible signatures of stau pair production. In particular I try to make the argument for a lower luminosity LHC run to better observe or exclude the existence of staus in a certain mass range.
Despite ongoing experimental and theoretical efforts, the nature of DM remains elusive. Historically, most attention has been on weakly interacting particles (WIMPs), whose parameter space is increasingly constrained. This has caused the community to consider alternatives to the standard DM paradigm. One such alternative is strongly interactive particles (SIMPs). Moreover collisionless DM is in tension with cosmological observations, in the literature known as the cusp-core problem. A velocity-dependent self-interaction in the dark sector alleviates this problem. In this talk I will show that scattering via a massive resonant particle can provide the correct velocity dependence, and apply this in a dark sector consisting of dark pions and photons. I will then show the requirements for reproducing the correct relic density in a SIMP setup, as well as constraints on the kinetic mixing between the dark photon and standard model photons. I will conclude with the validity of the model.
Machine Learning techniques are widely used for the analysis of collider events, especially for jet tagging (distinguishing jets originally hadronized from gluons, t-quark, b-quark, light quarks, W, etc.). It is also used for event classification between selected models (normally SM vs. one single BSM). We tried for years to extend the usage of ML techniques in a more general case, i.e. the signal could be a mix of various parameters of one single model, a mix of models, or even unknown models. Finally, we hope we can find some general methods to detect any anomalies in the real experimental data.
The dark photon is a well-motivated new particle, arising from a renormalizable interaction with the photon field. This so-called vector portal can lead to a dark sector, which can contain candidates for dark matter. This dark photon can be long-lived for suitable values of the parameters and, therefore, escape searches using prompt signals. We propose a different strategy, namely displaced vertex search, that is viable at e^+ e^- colliders such as Belle II that can probe these long-lived particles. We found that Belle II has excellent sensitivity to such dark photon signals and can probe them in an unexplored region of the parameter space using already collected data. For those who have heard my talk before, to give you something new, I will go into more details on how I model the background for the search.