Adverse Effects of Radiation on AFP Sub-Detector at LHC

The ATLAS Forward Proton (AFP) sub-detector and the adverse effects of radiation from Large Hadron Collider (LHC) collimators during p-p collisions at 13.6 TeV RAD 13 Conference Marko Milovanovic on behalf of ATLAS Forward Detectors 19.06.2025.

AFP location in P1/LHC > AFP configuration: AFP stations are located at ~205 m and ~217 m on both sides of the ATLAS Interaction Point (IP). The (NEAR) stations located closer to the interaction point contain Silicon Tracking detectors/planes (SiT), while (FAR) ones at 217 m are also equipped with Time-of-Flight (ToF) devices. At 212 m there is a patch-panel (PP) installed housing electronics services for module voltage regulation and data acquisition. Each station s detectors are mounted on a flange, housed inside a Roman Pot, which moves closer to the beam upon declaration of Stable Beams. The purpose of the AFP tracking system is to measure the trajectory of protons deflected during a proton-proton interaction & collect data as much as possible for physics analysis. Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 2

Proximity to Target Collimator Longitudinal 6 (TCL6) issue > During Run-3, AFP experienced an increased number of equipment damage/failure due to increasing beam intensities/integrated luminosity, also owing to its position within the LHC ring. > Geometry considerations: AFP FAR station is placed just ~1m upstream of TCL6 => high radiation levels next to TCL6 stem from small aperture. TCL6 protects downstream magnets & detectors by absorbing secondary particles emerging from IP1&5, reducing radiation damage and heat load on superconducting magnets, preventing quenches. > Impact on AFP: Electronics: mostly in FAR stations & patch-panel suffering from high radiation damage - many components had to be replaced and SiT calibrations/tuning were not possible to be performed. Operation: more issues expected during runs, causing dead time Personnel: almost impossible to perform any intervention on FAR stations due to extremely high radiation doses, reaching even ~1mSv/h (CERN default max allowed dose/person/day: 50 Sv!) Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 3 LSS1L: 900 Sv/h (dose rate at 40cm)

Consequences on AFP operation (2024 example) > Started observing issues early on: Electronics breaking down due to high radiation. Especially in vicinity of FAR stations and patch panels. Many Single Event Upsets (SEUs) observed/experienced. > SiT HV (I-V) scans performed at the beginning of the run (07/04) and after a month (10/05): Results show significant breakdown voltage deterioration, indicating suffering from severe radiation damage. > ToF mostly with module failures, requesting interventions every possible short access available. > Resulted in data-taking inefficiency and increased acquired radiation doses to AFP personnel. Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 4

Proposed and implemented solutions > Collimator settings: Relax TCL6 settings to favorable conditions => agreed in 2023, helping tremendously, however NOT in 2024 due to conflicting constraints (R2E in other experiments)! Thus, all the equipment had to go through extensive refurbishment during YETS 23/24. > Shielding: FAR stations: not really possible due to constraints imposed by vacuum system support and reserved volumes for handling. Only a small shielding wall in the available volume was installed in 2024 however with no substantial impact. PP: more volume available, therefore a shielding wall was able to be designed and installed during EYETS* 2024/25. FLUKA simulation results indicate reduction in: o Dose: inside the shielding bunker by a factor of ~3! o Particle fluences: by a factor of 2! Consequences: o FAR: ToF detector/crate in 2025 removed from the tunnel. o PP: access to components inside the PP limited/more difficult. *EYETS - End-of-YEar Technical Stop Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 5

Proposed and implemented solutions > Upgrade of ALARA Level & permanent IMPACT introduction Since at average of 400 s/h dose rate near FAR stations, CERN Radiation Protection unit suggested to increase the default implemented ALARA (As Low As Reasonably Achievable) Level 1 (up to 50/100 Sv) to Level 2 (up to 1 mSv) for AFP interventions in order to relax time constraints. o At this rate, one person can work up to 10 min at this location! Permanent/yearly IMPACT (Intervention Management Planning and Coordination Tool) was also introduced in order to minimize approval time for accessing LHC during short interventions and better follow up of individual doses throughout the year. This also involved appropriate preparation of Work Dose Planning (WDP) document. > Radiation monitoring: Deployment of battery powered online radiation monitors (BatMons) on the racks and patch panels on both sides of IP1 to follow up on the experienced radiation levels and shielding effect. Measurements indicated expected TID (Total Ionising Dose) levels: ~100 ~200 Gy, & HEH (High Energy Hadronic) fluences: 1011 1012 HEH/cm2. Occurrence of radiation-induced failures deemed critical, both in terms of lifetime degradation and very high probability of SEEs (Single Event Effects)! AFP locations (tunnel areas in general) are NOT safe for electronics. Fun fact: Deployed BatMons at FAR stations died a few months after installation Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 6

Further issues and improvements > Radioactive transport Since the beginning of Run3, stricter rules were introduced about transport of radioactive material => no self-transport possible* In order to have (sometimes irradiated) spares promptly available, there was a need to organize a storage area in a close vicinity of tunnel elevator. Buffer zone not an option as it was not meant as storage. Recently obtained a radiation supervised lab nearby (SR1) which can be used for this purpose. Fun fact: recently removed ToF crates were still too hot for transport back to our standard radiation supervised lab, having 8 Sv/h (Side-C) and 10 Sv/h (Side-A) at 40cm, greatly surpassing radiation supervised lab limit of 3 Sv/h! Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 7

Further issues and (potential) improvements > SiT radiation damage mitigation The Silicon tracker is based on a 3D silicon sensor bump-bonded onto FEi4b ASIC (ATLAS Insertable B-layer (IBL) based), thus both sensor and the chip suffer from radiation-induced damage. Apart from regular/monthly threshold re-tuning/calibration, a known mechanism to improve performance of both is through thermal annealing. Usually during Technical Stops (TS) or Machine Developments (MD), AFP uses the downtime opportunity to anneal the SiT modules. This greatly helps with lowering the leakage current, breakdown voltage and full depletion voltage of the sensor, while also reducing ASIC s LV current as well (TID bump). Constant degradation with irradiation is however inevitable. Occasionally, when AFP would experience an SEU in the Powering controller card circuit (PP), one whole arm/side would not be able to be powered. In such case, no need to insert the pots towards the LHC beam on the faulty side, in order to preserve them from unnecessary exposure to much high radiation during stable beams. SiT Temperature AFP is NOT able to be operated during PbPb data-taking, however this year it will participate in the Oxygen campaign. Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 8

Conclusions > AFP is exposed to extreme radiation due to proximity to LHC collimator TCL6, leading to severe operational challenges. > Frequent failures observed in ToF crates, patch panel electronics, and SiT modules due to Single Event Upsets (SEUs) and cumulative radiation damage. > Refurbishment efforts and shielding during YETS brought partial improvements, but limitations in space and design prevent full protection, especially at FAR stations. > Upgraded safety protocols: ALARA Level 2 and permanent IMPACT implemented to enable interventions more manageable under high-radiation conditions. > Radiation monitoring with BatMons confirms high TID and HEH fluences, underlining unsuitability of tunnel areas for long-term electronics operation. > Logistical challenges in transporting irradiated materials further complicate maintenance during short accesses. > Thermal annealing and insertion control help mitigate SiT degradation but nothing prevents it entirely of course. Therefore, change of modules every, or every second year is mandatory. > The continued operation of AFP is feasible, but demands significant maintenance effort and personnel resources. Sustained innovation in radiation mitigation and system design is essential to ensure efficient data collection is maintained and not severely affected by these challenges. Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 9

Backup Thank you for your attention! Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 10

Projected radiation levels in LSS1 Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 11

Projected radiation levels in LSS1 Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 12

Update from R2E-MCWG on BatMon measurements Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 13

Conclusions from R2E-MCWG on BatMon measurements > The small shielding walls were to remain until end of AFP program (at the time seemed most likely until end of 2024) and it was just struggling to survive until this point. Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 14

FLUKA results in more details: Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 15

HV comparison (September 2024) Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 16

AFP optics dependent various diffractive patterns Marko Milovanovic | RAD 13 Conference | 16 June 2025 | Page 17