New developments in 3D-trench electrode sensors
Gian-Franco Dalla Betta · 2025
Read New developments in 3D-trench electrode sensors by Gian-Franco Dalla Betta (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-2503-04272.txt.
Introduction
Jixing Ye and co-authors push a fast particle sensor toward production scale. Their 2025 paper reports new batches of 3D-trench electrode sensors, silicon detectors built for the brutal radiation and pile-up of future collider runs, where they must timestamp particles to about ten picoseconds.
The trench design keeps electric fields even, which protects timing after heavy radiation damage. The problem has been yield: too few good sensors per wafer for large detectors. This batch adds process fixes and a dashed-trench variant. Bench tests showed low leakage and high breakdown voltage in most devices, and the new variant yielded better than the standard one. The authors present the run as a step toward sensors large enough for real experiments.
Catalog record
| CREATOR | Gian-Franco Dalla Betta |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 10 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-04272
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
New developments in 3D-trench electrode sensors Gian-Franco Dalla Betta Abstract Future high-luminosity hadron collider experiments feature unprecedented levels of event pile-up and extreme radiation environments, calling for sensors capable of 4D tracking, even after significant radiation damage. To this purpose, 3D sensors represent a viable solution, since they provide excellent radiation tolerance and very good temporal resolution. In particular, owing to the uniform electric field and weighting field distributions, 3D-trench electrode sensors from the INFN TIMESPOT project have shown a temporal resolution of ∼ \sim 10 ps after irradiation fluences up to 1 × \times 10 17 1-Mev n eq /cm 2 . In spite of the excellent performance of these sensors, 3D-trench pixel technology is not yet fully established and the fabrication yield is not yet adequate for the production of large size pixel sensors. To improve the potential of the 3D-trench concept for large-area sensors, a new batch of sensors was designed at the University of Trento and fabricated at FBK, as part of the AIDA Innova project. Besides introducing some process improvements, this batch includes two different sensor variants: the standard one with continuous ohmic trenches, and a modified one with dashed ohmic trenches. On-wafer electrical test results show that most of the sensors have low leakage current and high breakdown voltage. Moreover, the fabrication yield for the new design variant is higher than that of the standard design. 1 Introduction The success of the Large Hadron Collider (LHC) paved the way for the upgrade to the High Luminosity LHC (HL-LHC), which has the ability to facilitate detailed studies of known phenomena, as well as the observation of very rare new events [ 1 ] . To advance the measurements to the next level of precision and increase the probability of discovering new physics particles, a successor to the HL-LHC, such as the Future Circular Hadron Collider (FCC), is necessary [ 2 , 3 ] . However, the massive event pile-up and extreme radiation environments that accompany such experiments will require sensors with 4D tracking capabilities to disentangle different vertices, even after fluences up to \sim 10 17 1-MeV n eq /cm 2 . Though remarkable temporal resolutions ( \sim 30 ps) have been achieved in state-of-the-art Ultra-Fast Silicon Detectors (UFSDs) [ 4 ] , further sensor engineering is required to enhance their radiation hardness. 3D sensors, on the other hand, are known for their extreme radiation hardness due to the short inter-electrode distance. Moreover, temporal resolutions comparable to UFSDs have been reported for 3D sensors with columnar electrodes [ 5 , 6 ] , while even better performances have been found when the columnar electrodes are replaced with trenches [ 7 ] . Most importantly, studies show that 3D sensors with trench electrodes can maintain their high temporal resolution with slightly decreased efficiency after radiation damage up to 1 \times 10 17 1-MeV n eq /cm 2 [ 8 ] . Despite the extraordinary performances in terms of temporal resolution and radiation hardness, the fabrication technology of 3D-trench electrode sensors is still in development. Defects in the long, continuous ohmic trench electrodes present in the original design [ 9 ] were observed in the TIMESPOT batches, that limited the fabrication yield. To address this issue, besides process improvements, a new design variant incorporating small gaps in the ohmic trenches was proposed. Results of TCAD and Monte-Carlo simulation show comparable temporal resolution and charge collection efficiency (CCE) of the standard and new design [ 10 ] . Following the proof-of-concept study, a new batch of sensors including both designs was fabricated at FBK as part of the AIDA Innova project. In this paper, detailed information about this batch are presented, covering aspects from layout design to reticle arrangements. On-wafer electrical tests are also reported to provide insight into the yield of the two designs. 2 AIDA Innova 3D-trench electrode sensors 2.1 Device description The fabrication inherited the single-sided technology that has been developed for small pitch 3D pixel sensors, in which stepper lithography is used to enhance the definition of layout details [ 11 ] . As indicated by Figure 1 (left), the active bulk has a thickness of 150 \mu m and is bonded to a highly doped p-type support wafer. Though all electrodes are fabricated from the front side, the ohmic trenches are made passing through the entire active volume in order to allow direct bias from the backside. The junction electrodes, on the other hand, are kept at a safe distance from the backside to prevent early breakdown. Moreover, a uniform p-spray layer is present on the front surface to isolate the pixels. Figure 1 (right) shows the pixel layout of the standard (STD) and dashed (DSH) sensors. Figure 1 : Schematic cross-section (left) and pixel layouts (right) of 3D-trench sensors. 2.2 Reticle arrangement To explore the full potential of the reticle and understand the impact of sensor size on the yield, its layout is designed to accommodate sensors of various array sizes: 32 \times 32, 64 \times 64, and 128 \times 128. The larger sensors are intended for future Read-Out Chips (ROCs) such as the IGNITE (INFN) and Picopix (CERN), whereas the smaller ones are compatible with existing ROCs from the TIMESPOT project. The baseline pixel size is 55 \times 55 \mu m 2 ; smaller pixels - 42 \times 42 \mu m 2 - have been introduced in test structures to start exploring the impact of smaller inter-electrode spacings on the performance. As can be seen from Figure 2 , each reticle includes 6 STD and 6 DSH sensors with an array size of 64 \times 64, located on the left side; while on the right side sit 3 STD and 3 DSH sensors with the size of 32 \times 32. Every reticle also hosts 1 STD and 1 DSH sensors with the size of 128 \times 128. Technological test structures are located at the top right of the reticle, which enables investigation of the fabrication processes. The block of test structures located nearby includes both STD and DSH sensors with different pixel sizes, configured as groups of individual pixels, strips, and diodes. It should be stressed that, compared to the TIMESPOT batches that only had 11 reticle shots per wafer, this batch was processed with an enhanced technology to improve the yield and increase the device density, for which two options were implemented: 18 or 29 reticle shots per wafer. It is worth noting that for both wafer layout configurations the wafer bow and warp were found to be remarkably small, below 20 \mu m in most case and below 60 \mu m in the three worst wafers. Figure 2 : (Left) Photograph of a processed wafer with 29 reticle shots and (right) layout of the reticle. 3 Electrical characterization On-wafer measurements were performed at room temperature and in dark conditions using the automatic probe station available at FBK, which can provide a maximum bias voltage of 100 V. The current-voltage (I-V) curves of all pixel sensors were tested using a temporary metal [ 12 ] . As an example, Figure 3 shows the I-V curves of two groups of ten 128 \times 128 sensors for both designs. It can be