The Full Scale Prototype of the Cylindrical-GEM as Inner Tracker in Kloe2 G.Bencivenni, S.Cerioni, D.Domenici, M.Gatta, S.Lauciani, G.Pileggi, M.Pistilli, Laboratori Nazionali di Frascati - INFN 1
The Kloe detector Kloe2 highlights Motivation for an Inner Tracker Cylindrical GEM prototype: Small size results Building procedure Cathode foils GEM foils Anode read-out foils Electrodes exctraction Vertical insertion system Detector sealing HV and FEE boards Stretching system Preliminary results Conclusions Overview 2
The KLOE detector Drift Chamber Calorimeter Multi-purpose detector optimized for K long physics Huge, transparent Drift Chamber in 5.2 kgauss field of a SC coil Carbon fiber walls, 55000 stereo wires, 2 m radius, 4 m long, He/iC 4 H 10 gas mixture Momentum resolution: σ(p T )/p T ~ 0.4% Pb-Scintillating Fiber Calorimeter with excellent timing performance 24 barrel modules, 4 m long and C-shaped End-Caps for 98% solid angle coverage Time resolution: σ T = 54 ps / E(GeV) 50 ps Energy resolution: σ E /E = 5.7% / E(GeV) 3
Kloe2 Highlights Kloe2 will exploit the capability of DaΦne with the new Crab Waist interaction region (test starts in November 07 see Murtas poster N15-223) 50 fb -1 are planned to be delivered in 3 4 years at the Φ peak energy for a vast physics program: significant sensitivities for the study of K S η and η rare decays, neutral kaon interferometry, lepton universality test will be reached Expect to start experimental program before 2010 The apparatus will run with the present sub-detectors (Drift Chamber and Calorimeter), a reduced Magnetic field to enhance sensitivity for low-energy tracks, and some important hardware upgrades: new read-out of Calorimeter new small angle Calorimeter and QCAL new γγ Tagger new ultra-light Inner Tracker based on GEM technology 4
Inner Tracker Available space for the insertion of the Inner Tracker inside the Drift Chamber 5
Inner Tracker geometry 5 independent Tracking Layers (L1-L5) L1 radius 15 cm (limited by K L K S interference) L5 radius 25 cm (limited by DC dimensions) DC wall 25 cm 15 cm 60 cm DC wall Could cope with a different IR 6
IT simulation results Simulation results for a π track from K S ππ Δx @pca Δz @pca Δp x @pca Δx @vertex IT 0.6 mm 0.9 mm 1.2 MeV/c 1.9 mm No IT 1.7 mm 2.2 mm 1.6 MeV/c 4.9 mm pca: point of closest approach C1: Δx@pca: difference of x coord. between the pca of track wrt the vertex and the vertex C2: Δz@pca: difference of z coord. between the pca of track wrt the vertex and the vertex C3: Δp x @pca: difference of momentum p x between the track at pca to the vertex and the vertex C4: Δx@vtx: sigma of the difference of x coord. between the reconstructed vertex and MC vertex 7
IT motivations and requirements Optimization for the physics coming from the interaction region: fine vertex reconstruction of K S and η decay products Detector requirements are: a) σ rφ ~ 200 µm; σ z ~ 500 µm point space resolution b) 5 khz/cm 2 rate capability c) Very low material budget < 1.5% X 0 (large MS effect on low-energy tracks) A GEM Detector allows to easily fulfill the requirements a) and b) With the technology of Fully Cylindrical GEM without internal mechanical supports an ultra-light detector can be built to fulfill the requirement c) 8
Cylindrical Triple-GEM Detector Read-out A Cylindrical Triple-GEM detector is obtained inserting one into the other the 5 cylindrical electrodes 2 mm 2 mm 2 mm 3 mm Anode GEM 3 GEM 2 GEM 1 Cathode Peculiar features of the C-GEM Ultra-light: anode/read-out and cathode are realized with the same raw material of the GEM foils, for a total material budget of ~3 X 0 for a Triple-GEM Dead zone free: inside the active area. All the fiberglass support mechanics is placed at the edges of the cylinder. The structural rigidity is obtained by stretching the detector 9
Small Size Prototype In 2006 we built the first C-GEM prototype using GEM foils from LHCb ( ~90mm, L ~250mm) Manual insertion of electrodes First application of the vacuum bag technique Final chamber in the support system, stretched at 700 g/cm and tested with X-rays in current mode up to gain of 10 4 without discharges 10
Full size L1 C-GEM prototype At the beginning of 2007 the construction of a Full Size Prototype has started The dimensions are similar to those of the IT Layer1: same diameter 300 mm Ø, but reduced 352 mm active length All electrodes obtained as a join of 3 foils (planar gluing) GEM size: 450x1000 mm 2 (single foil 450x333 mm 2 ) Anode readout with 1538 strips and 650 µm pitch (only rφ coordinate) FEE based on CARIOCA-GEM chip (amplificator/shaper/discriminator) Fiberglass support mechanics out of the active region (Permaglass) 11
C-GEM building procedure 2 1 1. An epoxy glue (Araldite 2011) is distributed along one edge of the GEM (~3 mm) 3. The cylinder is enveloped in a vacuum bag. Vacuum is applied with a Venturi system, providing a uniform pressure of 1 kg/cm 2 2. The GEM foil is rolled on a Aluminum mould coated with a 400 µm thick machined Teflon film for a non-stick, low-friction surface 4. The result is a perfect cylindrical GEM 12 3 4 With the same procedure Anode and Cathode are obtained
Cylindrical Cathode 3 foils joined together Permaglass annular flanges outside the active area support the detector 13
Cylindrical GEMs Foils are preliminary tested in a humidity controlled box 20 independent HV sectors ~50 cm 2 area 3 foils are planary glued with the vacuum tecnhique 3 mm overlap region 352 mm 960 mm 14
Anode read-out signal induction area with strips at 650 µm pitch FEE bonding flaps 500 µm pitch Detail of the read-out flaps to bond FEE 15
HV Resistors After the cylindrical gluing GEM are partially extracted from the mould HV 1MΩ resistors are soldered on the 60 GEM sectors 16
Electrode Extraction PVC exctraction ring cathode TEFLON coated mould 17
GEM1 Vertical Insertion System The Cathode is fixed to the bottom Al plate The other electrodes are fixed to the top plate and are pulled down slowly with a very precise linear bearing equipment cathode 18
glue is dispensed just before the foils are closed Detector Sealing Once the detector is fully assembled the insertion system can be rotated to easily seal both the sides detector is sealed with epoxy glue 19
HV Distribution boards Drift connection HV and read-out on the same side 6 connections for GEMs supplied by independent HV channels 3 boards are mounted, one for each independent GEM foil 20
Front-End Electronics FEE FEE for for Full Full Size Size Prototype based on on CARIOCA-GEM the 8 channel CARIOCA-GEM chip chip (LHCb Design Muon of a Motherboard System), uses bond 2 boards: to the detector with ZIF connectors carrying a Motherboard I/O signals bond and to controls the detector with ZIF connectors hosting regulators Design of and a Daughterboard I/O connectors hosting the chip a 16 channels Daughterboard hosting 2 CARIOCA chips FEE for final Inner Tracker based on dedicated 64 channels ASIC chip (GASTONE*) with serial read-out, uses 2 boards the same Motherboard for CARIOCA-GEM can be used a new 64 channels Daughterboard hosting the GASTONE chip * Collaboration with INFN-Bari 21
Prototype view with CARIOCA solution Each FEE slot hosts 4 CARIOCA-GEM chips, for 32 channels readout Output signal connections Front-End faraday-cage Kapton bonding flaps for input signal connection (ZIF connector) 22
Support and Stretching system The detector is finally mounted on a support system and longitudinally stretched with a tension of ~200 g/cm It can be easily and safely transported on a test-beam Load cell 6 gas inlets/outlets 23
Preliminary Test Results The CGEM has been flushed with Ar/CO2 (70/30) gas mixture at 286 cc/min and operated with the following parameters: Drift/Et1/Et2/I = 1.5/3/3/4 kv/cm G1/G2/G3 = 370/360/320 Estimated GAIN ~ 3000 Currents on each electrode < 10 na No discharges observed 25 october 2007 Next steps: Gas tightness test with gas humidity measurement Mechanical stretching test Mechanical strength test of the gluings Cosmic ray test Test beam at BTF-LNF with 500 MeV electrons in Dec07 24
Conclusions The Kloe detector is ready to take up the challenge given by the next 50 fb -1 delivered by the new DaΦne Hardware upgrades are foreseen to reduce systematics and open new physics channels A new Inner Tracker will be inserted to improve the vertex reconstruction for the physics near the IR (mainly K S and η) The IT is based on the innovative technology of fully Cylindrical GEM detectors (C-GEM) The Full size prototype, successfully built and preliminary tested, proves the feasibility of such a novel design The positive result opens the way for a new and competitive category of Ultra-light Micro-Pattern Gas Vertex Detectors 25
SPARES 26
Roll-in Proposal The Kloe roll-in will be performed in 2 steps: Step 0 : take place at end of 2008. Roll-in of the present detector with the minimal upgrades for a reliable and efficient run Step 1: take place possibly at end 2009. Implementation of all the hardware upgrades aiming to a long period data taking 27
IT Read-out scheme The cylindrical Anode is segmented with a 2-D set of X-V strips providing point space measurement X strips read-out from above V strips read-out from below and partially from above Xpitch = Vpitch = 650 µm XV angle = 40 Resolutions: σ rφ ~ 190 µm; σ z ~ 370 µm X and V strips are engraved on the same kapton plane V strips are connected with electrical vias 650 40 28 650
Test results: electron transparency Typical curves for electron transparency vs drift, transfer and induction fields are the fingerprint of a GEM Results are in perfect agreement with those obtained with planar triple-gem operated with the same gas mixture Is=anode current Id=cathode current I3d=current on the bottom side of the third GEM 29
Simulation of the GEM glued junction line the ~3mm overlap kapton region has been simulated (Maxwell+Garfield) the effect is a distortion of the field caused by space-charge electrons are anyway focused towards the multiplication holes of the GEM the efficiency for a MIP crossing @90 the middle of the overlap slightly drops from 100% to ~98% 30
Anode Assemby 31