Towards a New Generation of Monolithic Active Pixel Sensors

TL;DR

This paper introduces a new generation of Monolithic Active Pixel Sensors (MAPS) in 65 nm CMOS technology, emphasizing higher circuit density, reduced material budget, and suitability for future high-energy physics applications.

Contribution

It presents the development and testing of a novel MAPS sensor with small collection electrodes and integrated front-end amplifiers, demonstrating improved performance in beam tests.

Findings

Successful fabrication and testing of first test chips at MAMI

High signal-to-noise ratio achieved with small collection electrodes

Potential for use in future Higgs factories and beam telescopes

Abstract

A new generation of Monolithic Active Pixel Sensors (MAPS), produced in a 65 nm CMOS imaging process, promises higher densities of on-chip circuits and, for a given pixel size, more sophisticated in-pixel logic compared to larger feature size processes. MAPS are a cost-effective alternative to hybrid pixel sensors since flip-chip bonding is not required. In addition, they allow for significant reductions of the material budget of detector systems, due to the smaller physical thicknesses of the active sensor and the absence of a separate readout chip. The TANGERINE project develops a sensor suitable for future Higgs factories as well as for a beam telescope to be used at beam-test facilities. The sensors will have small collection electrodes (order of $\mu$m) to maximize the signal-to-noise ratio, which makes it possible to minimize power dissipation in the circuitry. The first batch of test chips, featuring full front-end amplifiers with Krummenacher feedback, was produced and tested at the Mainzer Mikrotron (MAMI) at the end of 2021. MAMI provides an electron beam with currents up to 100 $\mu$A and an energy of 855 MeV. The analog output signal of the test chips was recorded with a high bandwidth oscilloscope and used to study the charge-sensitive amplifier of the chips in terms of waveform analysis. A beam telescope was used as a reference system to allow for track-based analysis of the recorded data.