NSF-BSF: Study of Radiation Hard Materials and AI Analysis Methods for LHC Reaction Plane Detectors

Each year, the experiments at the Large Hadron Collider, LHC, at the European Laboratory for Nuclear and Particle Physics, CERN, in Geneva Switzerland collect data from high energy collisions of lead ions (Pb-Pb collisions). The colliding nuclei consist of protons and neutrons, which in turn are composite objects of quarks. For example, a proton contains two up-quarks and one down-quark.

Quarks are bound together in protons and neutrons through the strong nuclear force, mathematically described by Quantum Chromodynamics (QCD). The force is mediated through the exchange of gluons as force carrier particles. In the overlap region of the nuclear collision, protons and neutrons are destroyed and leave behind the so-called quark-gluon plasma (QGP), a form of pure quark matter that existed in the microseconds immediately following the Big Bang at the beginning of the Universe.

The reaction products of Pb-Pb collisions in the CERN detectors (ATLAS, CMS, ALICE and LHCb) often produce QGP. The observation of the final state particles makes it possible to characterize the properties of the QGP and in this way increases our understanding of the fundamental matter in the early universe. The research funded through this grant will result in radiation resistant materials and machine learning algorithms that are needed to use the ATLAS Zero Degree Calorimeter (ZDC) and the ATLAS Reaction Plane Detector (RPD) to determine the geometry of the Pb-Pb collision system.

The researchers supported by this award will measure both, the magnitude of the overlap of the colliding nuclei as well as the orientation of the nuclear collision system with respect to the ATLAS detector systems. The RPD and ZDC are located in a radiation environment that inflicts severe radiation damage to materials. For example, most optical materials will lose their light transmission after just a few weeks of usage.

This group will test different advanced fused silica materials for use as optical media in the active detection elements of RPDs and ZDCs. They will also study different radiation resistant photomultiplier tubes that will be used to readout the Cherenkov created in the active detection elements of the RPDs and ZDCs. Furthermore, the group will develop advanced machine learning algorithms that will be used to analyze the RPD data and to extract the orientation angle of the Pb-Pb collisions system with respect to the ATLAS detector.

The advanced radiation resistant materials and the machine learning algorithms developed for ATLAS will be also used by CMS ZDCs and RPDs and for future experiments at the Electron Ion Collider, EIC, at Brookhaven National Laboratory, BNL, on Long Island, NY. In the laboratories at UIUC and Ben Gurion University more than 12 graduate and undergraduate students will work on the R&D for the RPDs and ZDCs.

In addition, the laboratories host undergraduate summer students from 4-year colleges and high school student interns. Students will be trained in the Physics of the QGP, instrumentation for high radiation environments, advanced machine learning algorithms and collaborative research in large, international collaborations. The radiation-resistant detectors and materials developed in this project are expected to have applications in other fields, such as dose monitoring in radiation oncology, use for space-based instrumentation and radiation monitoring in response to nuclear accidents.

The project will use unique irradiation facilities at the Soreq Nuclear Research Center in Israel and in the LHC tunnel at CERN that have been used to identify radiation-hard Cherenkov radiator materials for the present ATLAS and CMS ZDCs and RPDs. The project also will make use of state-of-the-art materials research facilities, such as MRL at UIUC and laboratories at Ben Gurion University.

The work on Machine Learning Algorithms (MLAs) will start from the current ATLAS RPD MLAs and expand to include data from the ZDCs in order to characterize the event geometry of heavy ion collisions. The MLA applications will be developed and tested on DeltaAI at UIUC NCSA, the most performant GPU computing resource in NSF’s portfolio. The unique radiation facilities and the project expertise with MLAs will then be used to evaluate technology choices and materials for EIC forward detectors, for the ePIC electromagnetic calorimeter.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.