Speaker
Description
The ARCADIA Main Demonstrator 3 (MD3), developed by the ARCADIA INFN collaboration, is a fully Depleted Monolithic Active Pixel Sensor in the LFoundry 110nm CIS technology. It features a custom backside process that allows for the full depletion of the high-resistivity substrate.
The first test beam on the MD3 was performed at the Fermilab Test Beam Facility in July 2024 with a 120 GeV proton beam, with a custom-made telescope constituted by two MD3 tracking planes and one MD3 Device Under Test.
This work presents the testbeam results, focusing on the tracking performance in terms of efficiency and spatial resolution.
Summary (500 words)
The ARCADIA INFN collaboration developed a Fully Depleted Monolithic Active Pixel Sensor as a technology demonstrator, using the LFoundry 110nm CIS technology. The whole high-resistivity substrate can be depleted, thanks to a custom backside process that allows a uniform electric field distribution inside the sensing volume. Several technology demonstrators have been developed with an overall active thickness of 50, 100 and 200 μm, which makes them suitable for both charged particles and X-rays detection.
The pixel array has an area of 1.3 x 1.3 cm^2, with a 25 μm pixel pitch. The pixel output is digital, the readout architecture is event-driven and it can handle a rate up to 100 MHz/cm^2. Τhe chip design has been optimized for very low power consumption (10-30 mW/cm^2, depending on the event rate). These features make it suitable for experiments at future colliders, such as the FCC (Future Circular Collider), as well as space and medical applications.
The characterization of the ARCADIA Main Demonstrator (MD3) with 200 μm active thickness has been carried out with table-top experimental setups, including X-ray and radioactive sources.
During July 2024, the first test beam on the ARCADIA MD3 was performed with a 120 GeV proton beam at the Fermilab Test Beam Facility, using a custom-made telescope constituted by two ARCADIA tracking planes and one ARCADIA Device Under Test (DUT), as shown in Figure 1.
The data acquisition of the three planes is managed by one single commercial FPGA board. No scintillator-based trigger system is included in the telescope. For this reason, the data acquisition of the three planes was synchronized by sending a common signal to reset chip timestamps. Given the event-driven readout and the absence of an external trigger, clusters on the three planes are correlated in time, considering a 10 μs window. Starting from the time-correlated events, tracking, alignment, and DUT residual analysis are then performed.
This contribution presents an overview of the testbeam results, focusing on the tracking performance as a function of different front-end electronics parameters, such as the pixel threshold.
The average cluster multiplicity and extension along rows and columns can be seen in Figure 2. For the whole threshold scan range, the tracking efficiency is above 99.5%, and the width of the DUT residual distribution is around 5 μm. An example of DUT residual distribution for the default threshold value (VCASN = 5) is shown in Figure 3.