Detector Physics and Readout Electronics
Particle detectors are our "eyes" to observe the tinyest objects of our universe at the femtoscale. In our group, we develop fast and high-precision detectors to detect charged-particle tracks. So-called Micropattern Gas Detectors (MPGD) make use of gas as active medium and microscopic structures created by photo-lithography and etching techniques to amplify and read out the electronic signal. The two most prominent variants of MPGD are the Gas Electron Multiplier (GEM) and Micromegas, both of which are used in our research. Time Projection Chambers (TPC) using GEMs as amplification structure are a new and powerful technique, developed in our group, to record 3-D movies of particle collisions. Modern readout electronics is an integral part of every particle detector. Signals created in the detector need to be further amplified, processed and digitized by high-speed application-specific integrated circuits (ASICs). In our group, we perform research on understanding, adapting and improving the performance of MPGD for very high particle rates, as they are found at particle accelerators. The detectors are assembled in clean rooms, tested in the laboratory and in particle beams, and are then installed and operated in particle physics experiments. The new Forschungs- und Technologiezentrum Detektorphysik (FTD) provides all the high-tech equipment for our research.
Micropattern Gas Detectors
Micro Pattern Gas Detectors (MPGD) are a key technology for radiation detection in particle physics experiments, especially when high rates, large active areas or volumes, low material budget, radiation hardness and cost-effective solutions with good spatial resolution are required. COMPASS can be considered a pioneer in the application of GEM detectors. A set of 22 large-size triple-GEM detectors was constructed and installed under the leadership of B. Ketzer between the years 2000 and 2002 and has been operated by the group since then. For the physics program with hadron beams, GEM detectors with pixel readout were developed , which feature a very high rate capability and an extremely small material budget.
For AMBER, new GEM detectors are currently being developed. The new detectors should combine the known assets of GEM detectors with new features like trigger-less readout, minimal inefficient regions over a size of the order of square meters. Current projects include the quality control, assembly and characterization of these detectors as well as analysis of their performance in the beam. In cooperation with RD51 at CERN, several detectors of this kind will be built.
In addition, we investigate the behavior of MPGD at high rates at the microscopic level using simulations and measurements. An example is the investigation of charging-up phenomena in GEMs. Due to the exposed polyimide part of GEMs, charged particles get adsorbed there and change the electric field locally. Understanding this effect is important for the operation of GEM-based gaseous detectors in general.
Time Projection Chambers
TPCs are large-volume drift detectors that basically consist of a gas-filled vessel. Applying a precise drift field, and measuring the drift time of ionization electrons, 3-D pictures of particle collisions can be made. The drawback of TPCs has always been that they could take only about 1'000 such pictures per second, a rate that limits the application in high-luminosity environments as the LHC. The group of B. Ketzer has shown that this constraint can be overcome by using GEM detectors as amplification stage. The first large-size GEM-based TPC was applied in the FOPI experiment at GSI.
The technology was then further developed and adapted for the upgrade of the world's largest TPC at the ALICE detector at the CERN Large Hadron Collider (LHC). Our group - together with many other groups all over the world - took part in the upgrade. With the new technology, the ALICE TPC will be continuously recording heavy-ion collisions happening at a rate of about 50'000 per second, producing several TB of data per second. Current research in the group focuses on understanding and optimizing the performance of the new detector.
A similar detector, albeit at a smaller scale, is currently under study for CBELSA/TAPS, a local experiment on baryon spectroscopy at the ELSA accelerator. Currently, we study methods to measure distortions of the electron drift using a UV laser system, and perform Monte-Carlo simulations to assess the gain in physics performance with such a new detector.
Readout Electronics
In order to read out the signals induced in thousands of strips or pixels of particle detectors, modern electronics is applied. Very often, Application-Specific Integrated Circuits (ASICs) are directly connected to the readout electrodes of the detector. These highly integrated micro-chips perform first processing as preamplification and shaping of the signal. In the next step, the analog signal is digitized and processed further before it is transfered to the computer for storage. To control the ASICs and steer and further process the data flow, modern Field Programmable Gate Arrays (FPGA) are used. They are able to process many digital signals in parallel with complex operations and integrate dedicated memory blocks, Ethernet stacks, CPUs and recently even cores optimised for algorithms of Artificial Intelligence (AI). Those devices are traditionally programmed in Hardware Description Language (HDL), but recently programming in high level languages (High Level Synthesis HLS) is possible.
In our group, we use several ASICs to read out our detectors, depending on the purpose. The APV25 chip, originally developed for the CMS silicon tracker, is used to read out GEM detectors for COMPASS and NA64. We are currently preparing a new version of front-end cards and readout Analog-to-Digital Converters (ADCs) in cooperation with TU Munich. The components are assembled in-house, making use of equipment in the FTD as a pick & place machine, a wire bonding machine, etc. Each electronic card needs to be tested using a dedicated setup before it is connected to the detector.
The AFTER-T2K ASIC has been developed for the T2K experiment in Japan. It was adopted for the FOPI TPC at TU Munich and is also used for small test TPCs in our lab. In contrast to the APV25, this chip can store and send out up to 512 samples in time per channel, allowing us to record a full drift frame of a TPC.
While both the APV25 and the AFTER-T2K ASICs need an external trigger signal in order to be read out, future experiments with very high data rates and/or complicated event topologies like AMBER will operate in a free-running or self-triggered mode. Such a mode of operation is much more powerful in terms of physics potential, but requires a more intelligent front-end electronics, which can detect and send out signals without an external trigger. This is currently one of the main topics related to electronics development in our group. The topic is pursued in a Marie-Curie project funded by the EU and led my M. Lupberger. We are evaluating the novel TIGER and VMM ASICs to equip future detectors. For the VMM, we apply the RD51 collaboration general Scalable Readout System. For the TIGER, we collaborate with the group of INFN Torino. A final decision on which ASIC shall be employed at AMBER needs to be taken soon and we are comparing the two ASICs. Then, a dedicated AMBER front-end board will have to be delveloped to interface to the general AMBER DAQ. This requires hardware, software and FPGA firmware developments as well as intense testing of the electronics.
CERN's second largest accelerator, the so-called Super Proton Synchrotron SPS, not only serves as a pre-accelerator for the LHC, but its beams are also used to produce secondary beams guided towards CERN's North Area fixed-target experiments. In hadron-mode these beams consist mainly of pions.
AMBER, the successor experiment of COMPASS in the North Area, is heavily interested in physics with kaon- and antiproton-beams during a large part of its physics programme. For this it is necessary to filter out the other particle types so that the majority of beam particles are of the type of interest.
The so-called radio-frequency separation technique (p. 102) provides a method to clean the beam from unwanted particles (so mainly pions). Here, one makes use of the fact that the different particle species have different velocities at a given momentum. In a first RF-cavity (cavities provide a time- and position-dependent electric field) all particles get transversely deflected depending on their arrival time in the cavity. A second cavity is placed in a certain distance behind the first one, and their relative phase is tuned such that the wanted particles receive a net deflection while the others are not kicked. With a beam dump located behind the second cavity one filters out all types that have no net-deflection. Finally, the beam mainly consists of one species.
At the moment, we are in a phase where the feasibility of such a filtering technique is investigated. Therefore, it is necessary to develop and analyze different beam optics and cavity settings with simulation codes like MAD-X and BDSIM. If there is a positive outcome for the feasibility, the next step would be a conceptual design of such an RF separated beam. These studies are mainly performed at CERN and in cooperation wit the Beams Departement - Experimental Areas (BE-EA).