Tobias Bus, Benjamin Linnik, Michael Deveaux, Ali Yazgili ADvanced MOnolithic Sensors for News on radiation hardness studies for the CBM MVD Tobias Bus, Benjamin Linnik, Michael Deveaux, Ali Yazgili Institut für Kernphysik, Goethe Universität Frankfurt am Main 26.02.2018 Ladies and gentleman, my name is Tobias Bus and ill give you a short talk about radiation hardness of MAPS for the cbm mvd.

Leakage currents in MAPS +3.3V Reset +3.3V Output SiO2 SiO2 SiO2 N++ N++ P+ N+ Ileak 50µm P- 15µm Lets have a look at the sensor. The active volume is in the middle. Above is the higher doped P-Well, which has N implantations, that create depleted areas at the pn-junctions. The diode here is collecting charge from the active volume. By charge I mean signal electrons as well as leakage current. So – what do we know about leakage current? P+ Leakage current: Adds shot noise May clear the signal from the pixel (more later) Tobias Bus, DPG 2018, Bochum

Radiation tolerance: 5 ∗10 14 𝑛 𝑒𝑞 𝑐𝑚² PIPPER-2: „Fully depleted“ MAPS with AC- coupling Last year we tested some monolithic active pixel sensors, that were manufactured in a tower jazz process. An improvement of this process is the AC coupling which allows us to apply a higher voltage to the collecting diode, which increases the depleted volume of the sensor. The advantage of fully depleted sensors is a higher radiation tolerance, but on the other hand they struggle with effects of higher leakage currents, which could be a thing for the PIPPER-2 sensor as we can see on this plot: The sensor works fine, if it is heavily cooled. With higher teamperatures only a fraction of the original charge is collected. This could be a cause of charge clearing induced by high leakage currents. Signal amplitude highly relies on temperature. Charge clearing due to high leakage current? Tobias Bus, DPG 2018, Bochum

Leakage current – What do we know? Fully depleted PIN-diodes: Which volume? Δ𝐼=𝛼⋅ Φ 𝑒𝑞 ⋅𝑉 𝛼≈4⋅ 10 −17 𝐴/𝑐𝑚 G. Lindström, NIM-A 512.1 (2004): 30-43 Leakage current: scales with volume and radiation dose. does not depend on doping. What do we know? Earlier studies have shown, that for fully depleted sensors the leakage currents scales linearally with the radiation dose and the volume and does not depend on doping. Is it the same for the PIPPER-2 sensor and if yes, which volume is meant in the formula? Is it the active – or just the depleted volume? Does the fully depleted sensor has a higher leakage current than the not fully depleted one? So it has to be checked which volume causes the leakage current. Known to hold for depleted sensors. MAPS are not (always) depleted. Open issue: Valid for MAPS? Measure leakage current Tobias Bus, DPG 2018, Bochum

Open issue: How to measure depleted volume Seed (1 Pixel) Ndepl X-rays Depleted Summed (up to second next neighbor) Nepi Depletedvolume: 𝑉 𝑑𝑒𝑝𝑙 ≈ 𝑁 𝑑𝑒𝑝𝑙 𝑁 𝑒𝑝𝑖 ⋅ 𝑉 𝑝𝑖𝑥𝑒𝑙 Known from design Therefore studies with older Mimosa19 sensors were made, because they have 3T pixels, which allow us to directly measure the leakage current and they are not fully depleted. This comes in handy for us because just a fraction of the photons convert in the depleted volume and deposite its charge there. The charge of all other hits is shared by several adjacent diodes. To determine the just mentioned fraction one can devide the number of entries in this peak where all signal electrons are gathered by the entries of all hits Hierf+ür wurden Studien an veralteten Mi-19 Sensoren durchgeführt. Diese Sensoren eigenen sich besonder für eine solche Studie, da sie 3T Pixel enthalten und dadruch der Leckstorm direkt gemessen werden kann. Außerdem sind die Sensoren NICHT voll depletiert, sodass sich das depletierte Volumen vom aktiven Volumen deutlich unterscheidet. Daher kann man prüfen, welches der beiden Columina letzzendlich den Leckstrom ausmacht. Tobias Bus, DPG 2018, Bochum

Leakage current in MAPS? Literature 𝛼≈4⋅ 10 −17 𝐴/𝑐𝑚 Observed 𝛼≈3.4⋅ 10 −17 𝐴/𝑐𝑚 MIMOSA-19 Non-depleted sensor Preliminary Mismatch compatible with room temperature annealing. Ali Yazgili, bachelor thesis, in progress Depleting also increases the leakage current Depleted volume was used. Epitaxial layer delivers mostly no leakage current. Leakage current seems to scale with depleted volume. Tobias Bus, DPG 2018, Bochum

Leakage current in SB pixels Visible signal Qs Qphy Signal charge [a.u.] Lost Visible signal Qs Qphy High leakage current Lost Requirementsfor good (>90%) signal: 𝜏 𝑚𝑖𝑛 >10 𝑡 𝑖𝑛𝑡 =50µs => 𝐼 𝐿𝑚𝑎𝑥 ≈500 fA t [readout cycles] Collected charge Leakage current 𝑄 𝑆 = 𝑄 𝑃ℎ𝑦 ⋅ exp − 𝑡 𝜏 with 𝜏≈ 𝑘 𝐵 𝑇 𝐶 𝑒 ⋅ 1 𝐼 𝑙𝑒𝑎𝑘 with 𝐶≈10 fF Clearing of the pixel is accelerated by radiation induced leakage current. To obtain good signal leakage current must remain limited

Relevant for depleted sensors? Estimate highest acceptable leakage current, e.g. for MAPS with 5µs integration time Requiredforgood>90% signal: 𝜏 𝑚𝑖𝑛 >10 𝑡 𝑖𝑛𝑡 =50µs => 𝐼 𝐿𝑚𝑎𝑥 ≈500 fA± a factor More depletion (full matrix) Comparison with data: Ileak of heavily irradiated sensor exceeds limit. Cooling helps. 500 fA/pixel CBM-MVD PIPPER-II, IPHC Strasbourg Unsere Messungen haben gezeigt, dass diese Bedingung nur erfüllt ist, wenn wir die stark depletierten sensoren (zeige auf den den >20 V Bereich), wenn extrem stark gekühlt wird. In diesem Plot nur bis -15 Grad. Extreme Unterschiede im Leckstrom in Abhängigkeit von der Bestrahlung uind Temperatur. Austoben. Min 2 Min. Conclusion: Depletion… Improves charge collection in heavily irradiated sensors. Increases leakage current of heavily irridiated sensors. => Fast readout and/or cooling is needed to exploit advantages. Tobias Bus, DPG 2018, Bochum