Low dose gamma irradiation study of ATLAS ITk MD8 diodes

Silicon strip detectors developed for the Inner Tracker (ITk) of the ATLAS experiment will operate in a harsh radiation environment of the HL-LHC accelerator. The ITk is thus designed to endure a total fluence of 1.6E15 1MeV n_eq/cm2 and a total ionizing dose (TID) of 66 Mrad in the strip detector region. A radiation-hard n^+-in-p technology is implemented in the ITk strip sensors. To achieve the required radiation hardness, extensive irradiation studies were conducted during sensor development, primarily performed up to the maximal expected total fluence and TID to ensure a full functionality of the detector at its end-of-life. These studies included irradiations of sensors with various particle types and energies, including the Co60 gamma-rays. Our previous results obtained for gamma-irradiated diodes and strip sensors indicate a linear increase of bulk current with TID, while the surface current saturates at the lowest TID levels checked (66 Mrad), preventing a determination of the exact TID for which the observed saturation occurs. This work presents the results coming from irradiations by Co60 gamma-rays to multiple low TIDs, ranging from 0.5 to 100 krad. The detailed study of total, bulk, and surface currents of diodes explores an unknown dependence of surface current on the TID, annealing, and temperature. Additionally, the effect of the p-stop implant between the bias and the guard ring of measured samples is shown. The observations are relevant for the initial operations of the new ATLAS tracker.

💡 Research Summary

This paper presents a systematic study of low‑dose gamma irradiation effects on n⁺‑in‑p MD8 diodes that are representative of the silicon strip sensors to be used in the ATLAS Inner Tracker (ITk) for the High‑Luminosity LHC. The authors irradiated two diode designs – a standard MD8 diode and a variant (MD8p) that incorporates a p‑stop implant between the bias ring and guard ring – with 60Co gamma rays at total ionizing doses (TID) ranging from 0.5 krad to 100 krad. The irradiation was performed in two campaigns with dose rates of 1.60 krad min⁻¹ (up to 8 krad) and 8.5 krad min⁻¹ (10–100 krad), and the temperature was kept below 35 °C during exposure. After irradiation, the samples were stored at below –20 °C to prevent uncontrolled annealing.

The electrical characterization was carried out with a probe station at a controlled temperature of 20 ± 0.1 °C and relative humidity below 1 %. The authors measured the full I‑V curves and, crucially, separated the total leakage current (I_TOT) into bulk leakage (I_BULK) – the current flowing through the active volume bounded by the guard ring – and surface leakage (I_SURF) – the current flowing from the guard ring to the edge, which is dominated by surface effects. Measurements were performed at reverse bias up to 300 V, which corresponds to full depletion of the 320 µm thick silicon.

Key findings are as follows:

1. Dominance of Surface Current at Low Doses – Even at the lowest investigated dose (0.5 krad), the surface component rises sharply and becomes the dominant contribution to the total leakage current. In contrast, the bulk current remains essentially constant across the entire dose range up to 100 krad. This confirms that the n⁺‑in‑p technology is highly tolerant to non‑ionizing energy loss (NIEL) damage, while ionizing damage in the SiO₂ and Si‑SiO₂ interface drives the leakage increase.

2. Absence of Surface‑Current Saturation – Earlier work reported saturation of the surface current at about 66 Mrad, but the present study shows no sign of saturation up to 100 krad. The authors therefore infer that the saturation threshold lies somewhere between 100 krad and the previously observed 66 Mrad. This has practical implications for the early operation of the ITk, where surface leakage could continue to increase before reaching a plateau.

3. Effect of the p‑stop Implant – The MD8p diodes, which contain a p‑stop implant, exhibit slightly higher absolute currents than the plain MD8 devices. The p‑stop does not reduce the magnitude of the surface leakage; rather, it provides a well‑defined separation between bulk and edge currents, facilitating more accurate current decomposition.

4. Annealing Behaviour – Two annealing regimes were explored. Isothermal annealing at 60 °C for 80 minutes caused a modest increase in both bulk and surface currents, especially at bias voltages near full depletion. Longer isothermal anneals (up to 1280 minutes) showed a saturation of the increase after roughly 640 minutes. In contrast, stepwise isochronal annealing from 80 °C to 300 °C (20‑minute steps) produced a pronounced reduction of both current components at temperatures above 100 °C, ultimately restoring the leakage currents to their pre‑irradiation levels. This demonstrates that gamma‑induced defects in both the bulk silicon and the SiO₂/Si interface can be fully annealed at sufficiently high temperatures.

5. Temperature Dependence and Activation Energy – The authors measured leakage currents over a temperature range of –50 °C to +20 °C for three MD8p diodes irradiated to 10, 50, and 100 krad. All data were fitted to the standard semiconductor leakage expression I(T)=A·T²·exp(–E_A/2kT). The fitted activation energies for total, bulk, and surface currents were statistically indistinguishable, with an average value of E_A ≈ 1.20 eV (±0.01 eV). This value aligns with previously reported activation energies for silicon bulk generation‑recombination currents, indicating that the temperature dependence of the surface leakage is governed by the same fundamental processes as the bulk leakage, despite the different physical origins (oxide traps versus bulk defects).

6. Experimental Robustness – The dose uncertainty was kept below 5 %, and the I‑V measurements were repeated after each annealing step, ensuring reproducibility. The use of both MD8 and MD8p structures allowed the authors to cross‑validate the current separation methodology.

Conclusions – The study provides a detailed quantitative picture of how low‑dose gamma irradiation influences leakage currents in ATLAS ITk‑type diodes. The surface leakage dominates the total current increase, shows no saturation up to 100 krad, and can be fully mitigated by high‑temperature annealing (>100 °C). The bulk leakage remains stable across the examined dose range, confirming the radiation hardness of the n⁺‑in‑p bulk material. The identical activation energies for bulk and surface components simplify modeling of temperature‑dependent leakage in detector simulations. These results are directly relevant for the commissioning and early operation of the ATLAS ITk, informing decisions on biasing, cooling, and possible annealing procedures to maintain optimal detector performance throughout its lifetime.