Periodic Reporting for period 1 - GaGARin (Advanced Simulation Techniques for Gaseous Detectors: Application on Spherical Proportional Counters)
Periodo di rendicontazione: 2022-04-01 al 2024-03-31
Gaseous particle detectors are a commonly deployed technology in particle physics and beyond due to their flexibility, scalability, and cost compared to alternatives. They play an important role in many fundamental physics experiments from small-scale detectors such as the tabletop sized MIGDAL experiment all the way to large-scale experiments such as the ATLAS and CMS experiments at the LHC.
The path to detailed understanding of gaseous detectors is through detailed computer simulations. The goal of GaGARin was to bring together existing simulation technologies into a powerful, flexible, and fully validated simulation framework for gaseous detectors. To do this the project combined two simulation toolkits, Geant4, which is primarily concerned with the interactions of particles in matter and widely used in fundamental research, biology, and industry, and Garfield++, a simulation toolkit primarily dedicated to the study of gaseous based detectors. The combination of these toolkits, along with others such as finite element modelling software, allows the framework delivered by GaGARin to be used for a wide range of exciting studies, such as optimising micro-pattern gas detectors, performing neutron spectroscopy, and searching for dark matter (DM).
GaGARin is made more powerful with hardware acceleration technologies, namely graphics processing units (GPUs). GPUs were originally designed to render computer graphics, such as those in video games, however, more recently they have been found to be extremely powerful in other fields such as scientific computing and artificial intelligence. To take advantage of this a major objective of GaGARin was to add GPU support to a core algorithm of Garfield++, which simulates particle movement in the detector and their collisions with the atoms of the gas. This will enable advanced computations to be performed and existing computations to run much faster, allowing them to converge on results more quickly.
The final research objective of GaGARin was to use the simulation framework that was developed, including GPU support, in a variety of settings, such as rare-event searches, detector optimisation and neutron spectroscopy. The simulation framework was used in each of these settings, with simulations performed by the GaGARin project being used towards the design of the DarkSPHERE dark matter search experiment, in the MIGDAL experiment aiming for the unambiguous observation of the Migdal effect, in direct DM searches by the NEWS-G collaboration; and in neutron spectroscopy with spherical proportional counters.
The path to detailed understanding of gaseous detectors is through detailed computer simulations. The goal of GaGARin was to bring together existing simulation technologies into a powerful, flexible, and fully validated simulation framework for gaseous detectors. To do this the project combined two simulation toolkits, Geant4, which is primarily concerned with the interactions of particles in matter and widely used in fundamental research, biology, and industry, and Garfield++, a simulation toolkit primarily dedicated to the study of gaseous based detectors. The combination of these toolkits, along with others such as finite element modelling software, allows the framework delivered by GaGARin to be used for a wide range of exciting studies, such as optimising micro-pattern gas detectors, performing neutron spectroscopy, and searching for dark matter (DM).
GaGARin is made more powerful with hardware acceleration technologies, namely graphics processing units (GPUs). GPUs were originally designed to render computer graphics, such as those in video games, however, more recently they have been found to be extremely powerful in other fields such as scientific computing and artificial intelligence. To take advantage of this a major objective of GaGARin was to add GPU support to a core algorithm of Garfield++, which simulates particle movement in the detector and their collisions with the atoms of the gas. This will enable advanced computations to be performed and existing computations to run much faster, allowing them to converge on results more quickly.
The final research objective of GaGARin was to use the simulation framework that was developed, including GPU support, in a variety of settings, such as rare-event searches, detector optimisation and neutron spectroscopy. The simulation framework was used in each of these settings, with simulations performed by the GaGARin project being used towards the design of the DarkSPHERE dark matter search experiment, in the MIGDAL experiment aiming for the unambiguous observation of the Migdal effect, in direct DM searches by the NEWS-G collaboration; and in neutron spectroscopy with spherical proportional counters.
To achieve the research objectives of GaGARin five work packages (WPs) were defined.
WP1 was to develop the framework combining the strengths of Geant4 and Garfield++. A powerful and flexible simulation framework was created and designed in such a way that several different programs could be used to describe the detector using finite element methods, including Elmer which has the advantage of being open source and free to use. To ensure that the framework was user friendly, feedback from undergraduate students and scientific collaborators was incorporated, leading to new ways to configure the parameters of the detector being simulated in a simple manner and to submit simulations to be run on computer clusters.
WP2 rewrote key Garfield++ algorithms to run on GPUs using CUDA. The GPU versions of the algorithms were tested to ensure consistency with the CPU versions, and performance was benchmarked across various CPU and GPU models. High-performance GPUs can simulate electron avalanches in gaseous detectors orders of magnitudes faster than CPUs.
WP3 was to apply the simulation framework developed in WP1 in research and development of micro-pattern gaseous detectors. This work was performed in the context of the MIGDAL experiment which uses micro-pattern gaseous detectors called gas electron multipliers. The simulation framework was used to study the optimal operating voltages for the detector that give the experiment the greatest chance of success. The MIGDAL experiment took data during summer of 2023 and winter of 2024
WP4 was to apply the simulation framework of WP1 in rare-event searches. The simulation framework was used in two experiments, the MIGDAL experiment, which is looking for the very rare Migdal effect, and the NEWS-G experiment, which is searching for DM. MIGDAL used the simulations produced as part of GaGARin to model the expected processes that would be observed in the detector and to discriminate events produced by the Migdal effect from those produced by background processes. NEWS-G used the simulations to help calculate the experiments efficiency to detect DM interactions for different mass hypotheses, an important input for DM searches. The framework also played an important role in the design of the DarkSPHERE experiment, which is the next-generation DM experiment of the NEWS-G collaboration. In particular it played an important role in a breakthrough in spherical proportional counter instrumentation, namely the individual anode readout of an ACHINOS sensor.
Finally, WP5 was to apply the simulation framework in the context of neutron spectroscopy. Simulations were performed to compare with data taken as part of the project. Data were recorded using a spherical proportional counter filled with nitrogen, detecting both thermal and fast neutrons from an Americium-Beryllium source. The simulation was important in understanding the data and led to new insights. Further data were recorded at the MC40 cyclotron at the University of Birmingham and again compared with simulation.
The simulation framework was effective across all areas and contributed to several publications. Regular updates on the project's status were delivered to the NEWS-G and MIGDAL experiments and shared at conferences, workshops, and seminars.
WP1 was to develop the framework combining the strengths of Geant4 and Garfield++. A powerful and flexible simulation framework was created and designed in such a way that several different programs could be used to describe the detector using finite element methods, including Elmer which has the advantage of being open source and free to use. To ensure that the framework was user friendly, feedback from undergraduate students and scientific collaborators was incorporated, leading to new ways to configure the parameters of the detector being simulated in a simple manner and to submit simulations to be run on computer clusters.
WP2 rewrote key Garfield++ algorithms to run on GPUs using CUDA. The GPU versions of the algorithms were tested to ensure consistency with the CPU versions, and performance was benchmarked across various CPU and GPU models. High-performance GPUs can simulate electron avalanches in gaseous detectors orders of magnitudes faster than CPUs.
WP3 was to apply the simulation framework developed in WP1 in research and development of micro-pattern gaseous detectors. This work was performed in the context of the MIGDAL experiment which uses micro-pattern gaseous detectors called gas electron multipliers. The simulation framework was used to study the optimal operating voltages for the detector that give the experiment the greatest chance of success. The MIGDAL experiment took data during summer of 2023 and winter of 2024
WP4 was to apply the simulation framework of WP1 in rare-event searches. The simulation framework was used in two experiments, the MIGDAL experiment, which is looking for the very rare Migdal effect, and the NEWS-G experiment, which is searching for DM. MIGDAL used the simulations produced as part of GaGARin to model the expected processes that would be observed in the detector and to discriminate events produced by the Migdal effect from those produced by background processes. NEWS-G used the simulations to help calculate the experiments efficiency to detect DM interactions for different mass hypotheses, an important input for DM searches. The framework also played an important role in the design of the DarkSPHERE experiment, which is the next-generation DM experiment of the NEWS-G collaboration. In particular it played an important role in a breakthrough in spherical proportional counter instrumentation, namely the individual anode readout of an ACHINOS sensor.
Finally, WP5 was to apply the simulation framework in the context of neutron spectroscopy. Simulations were performed to compare with data taken as part of the project. Data were recorded using a spherical proportional counter filled with nitrogen, detecting both thermal and fast neutrons from an Americium-Beryllium source. The simulation was important in understanding the data and led to new insights. Further data were recorded at the MC40 cyclotron at the University of Birmingham and again compared with simulation.
The simulation framework was effective across all areas and contributed to several publications. Regular updates on the project's status were delivered to the NEWS-G and MIGDAL experiments and shared at conferences, workshops, and seminars.
The simulation framework and the associated GPU acceleration of Garfield++ that has been delivered enables more realistic simulations to be performed. Simulations that may have taken a full day can now performed in a matter of minutes on appropriate GPU hardware that were deemed previously not to be practically attainable.
GaGARin's contributions to neutron spectroscopy open the exciting prospect of affordable and safe neutron spectrometers being deployed in both academic and industrial settings. Even more performant neutron spectroscopy, with improved particle identification, may be possible when combined with another of GaGARin's key outputs, the individual anode-readout ACHINOS.
GaGARin will have a long-lasting legacy as it played an important role in the design of the DarkSPHERE experiment, which is proposed to be installed at Boulby Underground Laboratory.
Finally, the profile of the Birmingham Gaseous Detector Laboratory has been strengthened by GaGARin, through peer-reviewed publications and presentations at national and international conferences, workshops, and seminars.
GaGARin's contributions to neutron spectroscopy open the exciting prospect of affordable and safe neutron spectrometers being deployed in both academic and industrial settings. Even more performant neutron spectroscopy, with improved particle identification, may be possible when combined with another of GaGARin's key outputs, the individual anode-readout ACHINOS.
GaGARin will have a long-lasting legacy as it played an important role in the design of the DarkSPHERE experiment, which is proposed to be installed at Boulby Underground Laboratory.
Finally, the profile of the Birmingham Gaseous Detector Laboratory has been strengthened by GaGARin, through peer-reviewed publications and presentations at national and international conferences, workshops, and seminars.