Measurement of SiPM Dark Currents and Annealing Recovery for Fluences Expected in ePIC Calorimeters at the Electron-Ion Collider
Jiajun Huang, Sean Preins, Ryan Tsiao, Miguel Rodriguez, Barak Schmookler, Miguel Arratia · 2025
Read Measurement of SiPM Dark Currents and Annealing Recovery for Fluences Expected in ePIC Calorimeters at the Electron-Ion Collider by Jiajun Huang, Sean Preins, Ryan Tsiao, Miguel Rodriguez, Barak Schmookler, Miguel Arratia (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-14622.txt.
Introduction
Jiajun Huang and co-authors measure radiation damage in silicon light sensors in this 2025 paper. These sensors, called SiPMs, will read out all calorimeters in the ePIC experiment at the Electron-Ion Collider. Their dark current grows as radiation harms the silicon.
The team exposed the chosen sensor models to a proton beam in California. Doses spanned the full range expected across the detector's spots. Dark current was tracked by dose and by applied voltage. Heating the sensors, called annealing, was tested as a repair step. The results form a reference for designing the calorimeters and planning their operation under radiation load.
Catalog record
| CREATOR | Jiajun Huang, Sean Preins, Ryan Tsiao, Miguel Rodriguez, Barak Schmookler, Miguel Arratia |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 17 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-14622
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Measurement of SiPM Dark Currents and Annealing Recovery for Fluences Expected in ePIC Calorimeters at the Electron-Ion Collider Jiajun Huang, Sean Preins, Ryan Tsiao, Miguel Rodriguez, Barak Schmookler, Miguel Arratia Abstract Silicon photomultipliers (SiPMs) will be used to read out all calorimeters in the ePIC experiment at the Electron-Ion Collider (EIC). A thorough characterization of the radiation damage expected for SiPMs under anticipated EIC fluences is essential for accurate simulations, detector design, and effective operational strategies. In this study, we evaluate radiation damage for the specific SiPM models chosen for ePIC across the complete fluence range anticipated at the EIC, 10 8 10^{8} to 10 12 10^{12} 1-MeV n eq n_{\mathrm{eq}} /cm 2 per year, depending on the calorimeter location. The SiPMs were irradiated using a 64 MeV proton beam provided by the University of California, Davis 76" Cyclotron. We measured the SiPM dark-current as a function of fluence and bias voltage and investigated the effectiveness of high-temperature annealing to recover radiation damage. These results provide a comprehensive reference for the design, simulation, and operational planning of all ePIC calorimeter systems. 1 Introduction Silicon photomultipliers (SiPMs) are increasingly used in particle and nuclear physics experiments Simon:2018xzl . The ePIC detector at the future Electron-Ion Collider Accardi:2012qut , ABDULKHALEK2022122447 , incorporating concepts from Refs. Adkins:2022jfp , ATHENA:2022hxb , will be among the first collider experiments using SiPMs from day one. Approximately \mathscr{O}(1) million SiPMs will be used to read out all calorimeter systems Bock:2022lwp , as well as additional subsystems such as Cherenkov imaging detectors. While SiPMs offer numerous advantages over traditional photosensor technologies Simon:2018xzl , their susceptibility to radiation damage Garutti:2018hfu must be thoroughly understood. Such understanding is critical to inform detector design, perform realistic simulations, and develop robust operational strategies for experiments at colliders, particularly those using proton or nuclear beams like the EIC. The nominal luminosity of the EIC depends on the operating center-of-mass energy, peaking at 10^{34} cm -2 s -1 ABDULKHALEK2022122447 . Consequently, calorimeters will experience annual fluences between 10^{8} and 10^{12} 1-MeV neutron equivalent per cm 2 ( n_{\mathrm{eq}} /cm 2 ), depending on their location ePICFluence . Although this fluence generally produces moderate damage to SiPMs, characterized mainly by increased dark current rather than other performance degradations Garutti:2018hfu , the damage is amplified by the anticipated room-temperature operating conditions of most ePIC calorimeters. Dark current increases noise and may become problematic for system operation and temperature control, as well as affect the calorimeter cell thresholds, ultimately degrading physics performance. Several previous studies in the literature have examined SiPM radiation damage (see, e.g., Ref. Garutti:2018hfu ), yet a detailed, model-specific characterization across the complete EIC-relevant fluence range is still lacking. Additionally, it is important to understand the potential recovery of SiPM performance through high-temperature annealing Garutti:2018hfu , a factor crucial for detectors subjected to the highest radiation levels. At the EIC, the highest fluences ( 10^{11} to 10^{12} 1-MeV n_{\mathrm{eq}} /cm 2 per year) are expected for subdetectors nearest the beam pipe along the hadron-beam direction, specifically the innermost section ( r<50 cm) of the forward HCAL Bock:2022lwp , the calorimeter Insert Arratia:2022quz , and the Zero-degree calorimeter Milton:2024bqv . All these subsystems will operate at room temperature. By comparison, the SiPM-on-tile section of the CMS High-Granularity Calorimeter will be exposed to a maximum fluence of approximately 8\times 10^{13} 1-MeV n_{\mathrm{eq}} /cm 2 over its 10-year lifetime, but it operates at -30^{\circ} C to suppress dark current by a factor of 30 CMS:2017jpq . Consequently, one year of CMS HGCAL operation at the HL-LHC will yield dark current levels roughly equivalent to those experienced by the aforementioned EIC calorimeters. In either case, operating these detectors will likely push the limits regarding noise levels, underscoring the need for accurate measurements of the expected dark current and the recovery achieved through annealing, which is a mitigation option built into the design of the calorimeter insert Arratia:2022quz and the ZDC Milton:2024bqv . This work provides a comprehensive, unified reference on SiPM dark current induced by radiation damage relevant to the ePIC calorimeter systems. Section 2 describes the experimental setup, Section 3 presents the results, and Section 4 summarizes the conclusions. 2 Experimental Setup 2.1 Irradiation facility and SiPM models tested All SiPMs studied in this work were irradiated at the University of California, Davis, using the 76-inch Isochronous Cyclotron, which delivers a proton beam with a kinetic energy of 64 MeV. The cyclotron beam fluence is relatively uniform within a radial distribution of 3 cm DavisCyclotron . The fluence is measured to better than 10 \% with an ionization chamber. For reference, a 64 MeV proton beam induces approximately 50% more damage compared to a 1 MeV neutron beam AKKERMAN2001301 . \frac{K_{64MeVp^{+}}}{K_{1MeVn}}=\frac{NIEL_{64MeVp^{+}}}{NIEL_{1MeVn}}=\frac{\Phi_{1MeVn}}{\Phi_{64MeVp^{+}}}\approx 1.5\thinspace, (1) where K is the damage factor, NIEL is the non-ionizing energy loss, and \Phi is the equivalent fluence. The SiPM models irradiated and characterized in this study are manufactured by Hamamatsu: S14160-6050HS, S14160-6015PS, S14160-3015PS, S14160-1315PS, and S13360-6050VE. A metal stand holding a metal mesh covering the beam exit window was used, with the SiPMs attached to the mesh by aluminum tape. This allowed the SiPMs to be positioned approximately 15 cm from the center of the exit window, as shown in Fig. 1 . The tape containing a group of SiPMs was replaced after irradiation at each of the fluence levels: 10^{8} , 10^{9} , 10^{10} , 10^{11} , 10^{12} , and 10^{13} protons/cm 2 . Figure 1 : Figure (a) (top) illustrates the laser alignment of the SiPM tape placed on the mesh, centered directly in front of the cyclotron beam exit. Figure (b) (bottom) shows various types of SiPMs positioned within a 3 cm radius for irradiation. The number of units irradiated for each model at a given fluence point as well as the corresponding ePIC calorimeter is detailed in Table 1 . Models of SiPMs 10^{8} N {}_{p^{+}} 10^{9} N {}_{p^{+}} 10^{10} N {}_{p^{+}} 10^{11} N {}_{p^{+}} 10^{12} N {}_{p^{+}} 10^{13} N {}_{p^{+}} ePIC Detector Usage S14160 1315PS 1 3 3 3 3 2 nHCAL, pHCal S14160 3015PS 1 2 2 3 3 1 nEMCAL , bHCAL, pHCal(Insert), ZDC S14160 6015PS 1 1 1 2 2 1 nEMCAL, bEMCAL, pEMCAL S14160 6050HS 2 4 4 4 4 2 bEMCAL , pEMCAL, pHCAL (Insert), ZDC S13360 6050VE 2 2 2 2 2 0 bEMCAL Table 1 : Number of SiPMs irradiated at each fluence level. All SiPM models are manufactured by Hamamatsu. The calorimeter nomenclature is as follows: nHCAL, bHCAL, and pHCAL denote the electron-going endcap, barrel, and hadron-going endcap hadronic calorimeters, respectively. Similarly, nEMCAL, bEMCAL, and pEMCAL refer to the electron-going endcap, barrel, and hadron-going endcap electromagnetic calorimeters. ZDC stands for Zero Degree Calorimeter. The term “Insert" refers to the region of the pHCAL covering the pseudorapidity range 3.0<\eta<4.0 with respect to the proton beam. 2.2 Characterization Setup To prevent unintended annealing caused by soldering SiPMs directly onto measurement devices, a contact-based electrical connection setup using pogo pins was implemented, as illustrated in Fig. 2 . This setup features a hexagonal PCB designed to connect SiPM anodes and cathodes via pogo pins. It is secured with 3D-printed holders, and a custom cable with BNC and 3-pin connectors is used to interface