Performance of a LargeArea GEM Detector Prototype for

Performance of a Large-Area GEM Detector Prototype for the Upgrade of the CMS Muon Endcap System Vallary Bhopatkar M. Hohlmann, M. Phipps, J. Twigger, A. Zhang Dept. of Physics and Space Sciences, Florida Institute of Technology (for the CMS GEM Collaboration) 11/13/2014

Outline Ø GEMs for Future CMS Muon Endcap Upgrade Ø Construction of GE 1/1 Prototype Large-Area GEM Detector Ø FNAL Test Beam Oct 2013: Setup and Measurements Ø Detector Performance in FNAL Test Beam § Uniformity, Strip Multiplicity, Efficiency Ø Tracking Results § Correlation of GE 1/1 Detector with Trackers, Resolution Ø Correction for Non-Linear Strip Response Ø Efficiency and Resolution for Emulated VFAT Binary Hit Input using APV Analog Readout Ø Summary and Future Plans 11/13/2014

GEMs for Future CMS Muon Endcap Upgrade CMS Upgrade for LS 2 Muon Endcap Station 1 ers GE 1/1 long (1. 5<|η|<2. 2) and short (1. 6<|η|<2. 2) version § 36 superchambers in each endcap (Refer to Patrizia Barria’s talk (N 44 -4) for an overview of the CMS muon endcap upgrade) 11/13/2014 ng or t lo sh 100 superchambers (trapezoidal triple-GEM detectors) with rch § 10 o pe Introducing GE 1/1 in high-η region 1. 5<|η|<2. 2 su § am b Ø During Long shut down 2 (LS-2)

Top view of GEM+ CSC system in CMS GEMs for Future CMS Muon Endcap Upgrade Why GEMs? Ø High gain (~104) Ø Sustains a high rate (~MHz/cm 2) Ø Precise tracking and time resolution (~ns) Ø High spatial resolution Ø CSC + GEM -> accurate measurement of muon bending angle Ø Discriminate lower p. T muon from higher p. T Ø Helps to lower global muon trigger rate 11/13/2014 IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar

Construction of a Large-Area GEM Detector GE 1/1 Prototype III at Florida Tech Total assembly time 3 Hrs 40 mins with 2 people Step I GEM foils assembly Stack of GEM foils with inner frames ready to place on drift electrode Step III Close the GEM detector with readout board Step II Stretch GEM foils with inner and outer frames 8 -sector with 384 radial readout strips each (10 -15 cm long) Ø Internal gap configuration of the detector: 3/1/2/1 mm Finished GEM detector w/ APV frontend hybrids connected Eta 8 7 6 5 4 3 2 1 Ø GEM foils produced by single-mask etching technique at CERN Ø Active area: approximately 99 × (28 - 45) cm 2 11/13/2014 1 D readout board with 3, 072 radial strips connected through 24 Panasonic connectors to 24 APV hybrids

FNAL Test Beam Oct 2013 Setup and Measurements CMS GE 1/1 -III Detector Trackers ØGas mixture used in all detectors: Ar/CO 2 70: 30 ØBeam Energies: 1. 2. Mixed hadrons: 32 Ge. V Proton: 120 Ge. V ØThree 10 cm × 10 cm & one 50 cm × 50 cm GEM trackers with 2 D Trackers readout area @ 4200 V ØDAQ with RD 51 SRS ØGE 1/1 -III detector tests: 1. High voltage scan from 2900 V to 3350 V 2. Position scan at 3250 V in 3 positions: Upper row APV Middle row APV Lower row APV 11/13/2014

Basic Performance Characteristics of Detector from FNAL Test Beam Data Cluster Charge Distribution Cluster Size at 3250 V Cluster Size vs. High Voltage Mean strip multiplicity increases with voltage Cluster charge distribution at 3250 V fitted with Landau function Charge Uniformity Ø Average strip multiplicity at operating voltage is 2. 4 strips Detection Efficiency with Different Cuts on Pedestal Widths Ø At 3250 V, most probable values for all 3 APV position are used to determine the charge uniformity across the detector At 3250 V 11/13/2014 Ø Detection efficiency measured with this detector is 97. 08% (with 5 -sigma cut on pedestal width). Detection Efficiency is 97. 08%

Tracking Results Reference Tracker and Correlations with CMS GE 1/1 Detector Beam profiles with first tracker 11/13/2014 Tracker 4 Exclusive residuals σ = 75 µrad Tracker 3 Reference detector resolutions in φ Tracker 2 Inclusive residuals σ = 21 µrad 120 Ge. V proton Tracker 1 Residuals of Tracker 1 in φ 32 Ge. V mixed-hadron Correlation of GE 1/1 detector hits with hits in first tracker detector:

Tracking Results Resolutions for CMS GEM Detector in Eta Sector 5 Inclusive Residual σ = 111 μrad in φ Exclusive Residual σ = 138 μrad in φ Resolution of the detector: σanalog = 124 µrad (27% of strip pitch) 11/13/2014

Non-Linear Strip Response Correction Scatter Plots of Exclusive Residual vs. ηN using combined HV scan data 2 -strips cluster η 2 fitted with 10 -degree polynomial function 11/13/2014 3 -strips cluster η 3 fitted with serpentine function IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar

Non-Linear Strip Response Correction For N= 2 and N =3 Before correction Resolutions for HV Scan After correction Resolutions for Position Scan Overall resolution is improved by ≤ 8% after correcting the non-linear strip response 11/13/2014 IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar

Emulated VFAT Binary Readout Results Using APV Analog Readout • • • CMS upgrade electronics proposes VFAT 3 that collects data using binary output VFAT threshold reconstructed using following conversions 1 ADC count = 0. 03172 f. C and 1 VFAT unit = 0. 08 f. C Efficiency is plotted for three threshold values 10 VFAT, 12 VFAT, and 15 VFAT units Detection efficiency using different VFAT thresholds Detection efficiency is 97% 11/13/2014 IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar

Emulated VFAT Binary Readout Results Using APV Analog Readout Resolutions for CMS GEM Detector in Eta Sector 5 Inclusive Residual σ = 123 μrad in φ Exclusive Residual σ = 153 μrad in φ The resolution of the GE 1/1 detector is again obtained using geometric mean of inclusive and exclusive residuals Resolution of the detector σbinary = 137 µrad (corresponds to σ = 255 µm in azimuthal direction) To be compared with expected resolution using strip pitch: pitch/√ 12 = 131 µrad 11/13/2014 IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar

Summary Ø The beam test at FNAL was successful as the performance of the GE 1/1 prototype-III large-area detector meets the expectation for the CMS upgrade. Ø Detection efficiency of the detector is 97% for both pulse height readout and emulated VFAT binary readout in eta sector 5. Ø We observed non-uniformity in all three APV positions due to slight bending of drift and readout boards. This is being improved in new GE 1/1 prototype detectors developed this year. Ø The spatial resolution improves with high voltage and gives the best value of 124 μrad using full pulse height readout, when the detector is operated at efficiency plateau. Ø Resolution of GE 1/1 detector is improved by ≤ 8% after correcting the non-linear strip response. Ø Resolution using emulated VFAT binary method is 137 µrad, which is consistent with pitch/√ 12 = 131 µrad 11/13/2014

Thank You! 11/13/2014 IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar

Backup Slides 11/13/2014 IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar

Gas Electron Multiplier Detector Ø Micro Pattern Gas Detector (MPGD) Ø GEM foil is a kapton foil coated with copper on both sides that has an array of holes (typically 140µm pitch) Ø High voltage is applied across foils, which creates avalanche of electrons through holes Ø Provides good efficiency and spatial resolution 70µm 140µm Source: http: //gdd. web. cern. ch/GDD/ Ø Typical gas gain of 104 with gas mixture of Ar/CO 2 Triple GEM configuration: Most popular and reliable Murtas, F. : “Development of a gaseous detector based on Gas Electron Multiplier (GEM) Technology 11/13/2014 Source: http: //www. flc. desy. de/tpc/basicsgem. php

Efficiency Measurement • Studied the hit distribution, charge distribution and cluster size for HV scan and Position scan • Evaluates the efficiency from cluster multiplicity(CM) Where, N 1: No. of events with CM≥ 1 for given sector N: Total no. of events N 2: sum of the no. of events with CM≥ 1 for other sectors 11/13/2014 IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar

Tracking • 11/13/2014 IEEE NSS MIC Conference 2014, Seattle, WA - Vallary Bhopatkar