Abstract
High performance DMAPS (Depleted Monolithic Active Pixel Sensor) for hadrontherapy. "This project aims to significantly improve the capabilities of the pixel tracker in a fixed target experiment, particularly in terms of the amount of data that can be collected for the same amount of time and spatial resolution, which for obvious statistical reasons allows for greater accuracy on the measurements, in order to improve the knowledge of the fragmentation cross section used for hadrotherapy. We will replace the sensors used till now in the vertex of the FOOT experiment with those of the next 'generation' (the MIMOSIS sensor), with integration times approximately 40 times shorter in order improve the overall capabilities of the experiment. The sensor is currently available, and the techniques for using it are already fully in the hands of the involved teams. Finally, the new Vertex Detector that will be constructed using the new sensors should also be able to cope with measurements for much heavier nuclei such as Fe56, which is of considerable importance for the studies of radio protection in space, fundamental for the planning of future deep space missions." "The main feature that distinguishes the MIMOSIS sensor from the sensors currently used in FOOT is the temporal resolution of 5 μs with respect to the present 185.6 μs frame readout time. Therefore, the MIMOSIS sensor is about 37 times “faster”. Moreover, the digital reading logic based on the priority encoder and ""elastic"" reading buffers allow a capacity of 0.7 Mhits/s × mm2 at peak and ability to handle larger fluctuations in the data stream. The new MIMOSIS sensor has a larger active area of 31.5 x 13.5 mm2 with respect to the previous sensors of 19.2 x 19.8 mm2. To cover at least the current angular phase space, two sensors for each plane are required. The two sensors will be glued on the two sides of the printed circuit board, overlapping on a central line in correspondence of the hole center provided in the same printed circuit board. The PCB hole will be open below the entire active area of the two sensors to minimize the material to be crossed by particles. The sensors will however, for the same reason, be thinned to a thickness of 50 μm. In this geometrical configuration the acceptance of the new VTX detector would be increased to 31.5 mm compared to the current 19 mm. In the other direction the size of the sensitive zone would increase, assuming an overlap of the active area of about 1 mm, to about 26 mm (12.5+12.5 mm+1mm of superimposition). Moreover, the increased pixel size, from a pitch of 20.7 μm to 26.88 μm in one direction and 30.24 μm in the other, will produce a reduction in the spatial resolution of the tracking, which is currently of the order of a few micrometres."
Results achieved
: Charged particle therapy (hadrontherapy) is increasingly used to treat deep-seated tumors thanks to the Bragg peak, which concentrates the radiation dose within the tumor while minimizing damage to healthy tissues. Nevertheless, secondary radiation produced before and beyond the Bragg peak remains a major source of uncertainty because of the limited knowledge of nuclear fragmentation cross sections. The FOOT experiment addresses this issue by measuring charged fragments produced in both inverse and direct kinematics, requiring tracking detectors with micrometric spatial resolution and excellent timing performance. Within this framework, the PRIN project aimed to upgrade the FOOT Vertex Detector by replacing the current MIMOSA-28 sensors with the new MIMOSIS-2.1 Monolithic Active Pixel Sensors (MAPS), developed for the CBM experiment at FAIR. Compared with the present detector, the new sensors reduce the integration time from about 186 μs to 5 μs while preserving excellent spatial resolution and significantly improving radiation tolerance. These features allow higher acquisition rates and better tracking performance, increasing the statistics achievable during beam measurements. The upgraded Vertex Detector consist of four tracking planes equipped with two MIMOSIS-2.1 chips each. The sensors provide spatial resolutions of about 10 μm, time resolutions of about 5 μs and high radiation hardness, making them suitable for the experimental conditions of FOOT. Each pair of sensors is mounted on a dedicated proximity board connected to a custom DAQ board and a DE10-Nano FPGA system. Data are transmitted through configurable LVDS serial lines and stored in memory until an external trigger identifies the events to be recorded and transferred to the acquisition server. The FPGA also distributes the clock and manages the sensor configuration through the I²C communication protocol. The MIMOSIS chips continuously transmit data using a data-push architecture with a global-shutter readout. Communication and configuration are performed through the I²C protocol operating at 400 kHz, allowing the control of operating mode, thresholds and active output lines. The FPGA continuously receives the data stream while only the frames contained within a programmable trigger window are stored after an external trigger. This strategy limits the amount of recorded data while preserving all information associated with beam events, making the acquisition compatible with the available data bandwidth. The first characterization of the system was carried out at the University of Bologna using a MIMOSIS-1 prototype mounted on a proximity board connected to the DAQ electronics. The setup reproduced the final acquisition chain and allowed both slow-control operations and high-speed data readout to be validated. Two main measurements were performed. The first consisted of a threshold scan to determine the optimal operating threshold for each sensor submatrix by fitting the response with a sigmoid function and extracting the pedestal and noise parameters. The selected operating threshold efficiently suppresses dark-count noise while maintaining full detector sensitivity. The second measurement verified the correct decoding of pixel addresses using a radioactive source. Without a collimator the expected radial hit distribution around the source position was observed, whereas the addition of a collimator produced a well-defined spot whose dimensions matched the irradiated area, confirming the correct operation of the readout chain and hit reconstruction. The final validation of the detector was carried out at the Beam Test Facility of the INFN Laboratori Nazionali di Frascati using the complete readout architecture. The setup consisted of six tracking layers, each equipped with two MIMOSIS sensors read out by independent DE10-Nano boards, reproducing the configuration foreseen for the upgraded FOOT Vertex Detector. The detector was exposed to a 450 MeV electron beam to evaluate the stability of the acquisition system and the performance of the sensors under realistic operating conditions. All chips operated correctly throughout the data-taking period, demonstrating the reliability of both the electronics and the data acquisition chain. Beam profiles reconstructed from the recorded hits confirmed the correct synchronization of the sensors and the expected detector response. Data analysis is still in progress; however, preliminary results indicate an overall detection efficiency exceeding 97%, confirming that the MIMOSIS sensors satisfy the requirements of the FOOT experiment. These results validate the adopted readout architecture and represent an important milestone toward the installation of the upgraded Vertex Detector. The activities performed within this project demonstrated the successful integration and characterization of the new MIMOSIS MAPS sensors for the upgrade of the FOOT Vertex Detector. Laboratory measurements validated the detector configuration, communication protocol, threshold optimization procedure and hit reconstruction, while beam tests at Frascati confirmed the stability of the complete acquisition system under realistic operating conditions. The excellent preliminary efficiency, together with the improved timing performance and radiation hardness of the MIMOSIS sensors, represents a significant improvement over the current detector. The upgraded system enables higher acquisition rates and more precise tracking, providing better statistics and contributing to more accurate measurements of nuclear fragmentation cross sections relevant to hadrontherapy.Project details
Unibo Team Leader: Sara Valentinetti
Unibo involved Department/s:
Dipartimento di Fisica e Astronomia "Augusto Righi"
Coordinator:
INFN-Istituto Nazionale di Fisica Nucleare(Italy)
Total Unibo Contribution: Euro (EUR) 82.522,00
Project Duration in months: 24
Start Date:
28/09/2023
End Date:
28/02/2026
