Skip to main content

an Innovative Negative Ion TIme projection chamber for Underground dark Matter searches

Periodic Reporting for period 1 - INITIUM (an Innovative Negative Ion TIme projection chamber for Underground dark Matter searches)

Reporting period: 2019-03-01 to 2020-08-31

INITIUM goal is to boost the development of gaseous Time Projection Chamber (TPC),in particular in direct Dark Matter (DM) searches. The presence of DM is nowadays an established, yet still mysterious paradigm: deciphering its essence is one of the most compelling tasks for fundamental physics. The determination of the incoming DM direction can provide an astrophysical correlation that offers an unique key for unambiguous identification of DM. TPCs are now mature technology to aim at developing a ton-scale experiment and can provide the best observables for a directional DM searches, able to record not only recoil energies as conventional experiments but also track direction and dE/dx topology. INITIUM focuses on the development of a 1 m3 Negative Ion TPC (NITPC) with Gas Electron Multipliers (GEMs) and optical readout with sCMOS cameras and PMTs at Laboratori Nazionali del Gran Sasso (LNGS). The combination of 3D optical sCMOS + PMT readout with reduced diffusion and fiducialization offered by the negative ion drift (NID) can offer superior means for discrimination of electron recoils (ERs) from nuclear recoils (NRs) and directionality down to keV energies.
Given the project R&D, we produced two prototypes to focus each on different sets of studies: MANGO, 10x10 cm2 readout, 5 cm drift, 1 sCMOS+1 PMT (Fig1), and LIME, 33x33 cm2 readout, 50 cm drift, 1 sCMOS+4 PMTs (Fig2). MANGO will study innovative gas mixtures and GEM structures for improved light yield and reduced diffusion, and LIME (identical to one of the 18 INITIUM modules, Fig3) will verify on realistic dimensions underground INITIUM expected performances, testing materials and construction techniques. Given the challenge of NID, we started with He:CF4 to first characterise the prototypes and better understand means for improvement before adding SF6. The sCMOS Orca Fusion purchased (to be later used in the final 1 m3 TPC) improves over Orca Flash thanks to reduced noise and larger quantum efficiency. For MANGO we assembled an innovative field cage with a PVC thin resistive foil (10^11 Ohm/resistivity), a new approach recently tested by NEWAGE. This solution can give better continuous surface approximation and reduce internal radioactivity. Preliminary tests showed results consistent with conventional field cages. With MANGO we demonstrated for the first time electroluminescence in He:CF4. We placed a mesh 3 mm beyond the last GEM to create a 11.3 kV/cm electric field that accelerates electrons to produce additional light from the gas. We observed a factor 3 enhancement of the light (consistent with expectation), without increase of electric gain, nor deterioration of energy resolution. For LIME, following the experience from CMS, we stretched GEMs on a PMMA frame rather than on CERN PCBs to minimise the radioactivity from this item. LIME field cage is made of 6 mm × 2 mm Cu rings supported by a PMMA comb at 1 cm pitch.LIME energy resolution at 5.9 keVee is 15% along the full 50 cm length.We manufactured a second field cage for LIME with the resistive foil, to be tested soon.LIME characteristics allow a preliminary but already significant measurement of underground LNGS neutron flux. With LNGS Services, we identified LIME underground location (Fig4 left) and are working on its setup (Fig4 right).We are also setting up a lab in overground LNGS for MANGO R&D studies and LIME support.COVID19 emergency significantly slowed work with prototypes and setups at LNGS, but we hope to finalise overground lab in Fall 2020 and to bring LIME underground by first semester of 2021.
We developed detector simulation from engineering design to study external and internal backgrounds and optimise shielding and materials.We studied external gammas and neutrons effect with different Cu, Pb and H2O shielding, and the goal of <10^4 gamma/year in 0-20 keV and <1 neutron/year.We identified a compact expensive solution with (going inwards) 50 cm H2O+15 cm archeological Pb+5 cm Cu and a more economic one with 2 m H2O+5 cm Cu, worked out through cost-benefit study starting from realistic cost estimates.Preliminary INITIUM layout is shown in Fig3, marginally modified from proposal design to be better able to mitigate possible contingencies.Each endcap is equipped with 9 modules with LIME design.DAQ employs CameraLink for sCMOS and custom digitization boards for PMTs.Gas system shown in Fig5 continuously flow at 100 l/h and recirculate gas through purification elements removing contaminants, with exhaust gas stored for proper external treatment.For both DAQ and gas system, we purchased components to test on LIME before finalise them on the 1 m3 TPC.Thanks to LNGS Services, we performed gamma radioactivity measurements of INITIUM main internal components (reported in Fig6 together with performances measurements of sCMOS obtained with LEMON prototype).We measured large 40K contamination in glass objective and are working with producing companies to develop these with alternative materials, and large radioactivity in GEM frames (reason why we stretched them on PMMA for LIME).We are working on sCMOS and GEM radioactivity reduction in close contact with the companies and CERN.We purchased a non-working sCMOS to measure its internal parts to single out the main radioactivity contributions and develop reduction solutions.
We developed simulation to reproduce expected signals in sCMOS sensors.Given the known difficulties in simulating NID, we started from He:CF4 to add SF6 later on.We generate NR with SRIM and ER with Geant4, and apply Garfield to get effective gas ionisation.Electron diffusion, gas light yield and sCMOS sensor sensitivity and noise are parametrically simulated from our measurements with LEMON.While we are working on a systematic crosscheck, 5.48 MeV simulated He recoils shows same range of experimental measurement with 241Am.We developed the iDBSCAN track reconstruction and identification algorithm based on DBSCAN with multiple iteration to select track with different ionization patterns.We studied its performances on ER from 55Fe and NR induced by AmBe source and obtained ER rejection at 5.9 keVee of 10^{-3} and of the sCMOS noise of 10^{-4}$, with 40% NR efficiency.While these results are already encouraging, rejection can significantly be improved with a more sophisticated multivariate approach, on which we are working.
-We obtained the first experimental evidence of electroluminescence in He:CF4, opening new windows for any gas detector He:CF4 based and optically read out thanks to the possibility of increasing light yield without the need for additional electric gain.
-Following the work by NEWAGE, we manufactures a 50 cm field cage with an innovative resistive foil.This can result in a significant advancement of TPCs for any field where low radioactivity and/or low material budget is needed.
-We started for the first time a systematic evaluation of sCMOS radioactivity and of its internal components and development of low-radioactivity objectives.This is very important for any sCMOS application when low radioactivity budget is required.
-We managed to stretch 33 x 33 cm2 GEMs on low radioactivity PMMA rather than on CERN PCBs, known to be significantly radioactive, a very interesting development for GEM based gas detectors requiring low radioactivity budget.
The further results expected until the end of the project will follow the WPs and developments discussed in the proposal.