Investigation of GEM space point resolution for a TPC tracker Dean Karlen / Carleton University Carleton GEM group: Bob Carnegie, Madhu Dixit, Jacques Dubeau, Dean Karlen, Hans Mes, Morley O'Neill, Ernie Neuheimer, Anna Kristofferson, Jeff Mottershead http://www.physics.carleton.ca/~karlen/gem
Outline! What is a GEM?! Why use GEMs in a TPC?! Point resolution studies at Carleton! experimental setup hexagonal pads! data and results Ar CO 2 and P1! simulation work! Future plans! Summary 26 October 2 Dean Karlen / Carleton University 2
Gas Electron Multiplier (GEM) E drift =.2 1. kv/cm E transfer 3 kv/cm E induction 3.5 kv/cm V 4 V V 4 V Readout pads 26 October 2 Dean Karlen / Carleton University 3
GEM foils and pads for this study fabricated at the CERN PCB workshop GEM foil GEM pads 5 µm 2.5 mm pitch hexagons 14 µm 26 October 2 Dean Karlen / Carleton University 4
Using a GEM in a TPC Conventional TPC readout GEM TPC readout 26 October 2 Dean Karlen / Carleton University 5
Potential advantages for GEM readout! Improved space point resolution! E B and track angle systematics suppressed! Improved two particle separation power! r -φ : signals distributed over smaller area! z : faster induction pulses (v e > v ion )! Natural ion feedback suppression! no gating required (non-triggered expt.)! Less mass in TPC endcap! no wires held under tension 26 October 2 Dean Karlen / Carleton University 6
Point resolution studies at Carleton x-ray tube pin hole GEM cell 2D micrometer stage 26 October 2 Dean Karlen / Carleton University 7
Details! x-ray mean energy: 4.5 kev! pinhole diameter: 5 µm! Gas: Ar CO 2 (~9:1) / P1 : Ar CH 4 (9:1)! pre-amps:! fast Lecroy HQV 81 with Ar CO 2! slower ALEPH TPC pre-amp with P1! readout:! two 4-channel digital scopes (9 bit ADC)! 5 MHz sampling for HQV 81! 125 MHz sampling for ALEPH preamps 26 October 2 Dean Karlen / Carleton University 8
-.1 -.2 -.1 An event -.5 -.1 -.15 -.2-5 x 1.2.4.6.8.1 x 1-5.2.4.6.8.1 3 2 1 8 7 -.25 -.5 -.75 -.1 Ar CO 2 (9:1) HQV81 preamps.2.4.6.8.1 x 1-5 -.2.2.4.6.8.1 x 1-5 4 6 -.1 -.2 -.5 -.1 -.15 -.5 -.1.2.4.6.8.1-5 -.15 x 1.2.4.6.8.1 x 1-5.2.4.6.8.1 x 1-5 26 October 2 Dean Karlen / Carleton University 9
Analysis Direct pulse Induced pulse direct charge deduced from tail peak value used 26 October 2 Dean Karlen / Carleton University 1
Localization using charge sharing PAD 2 PAD 8! assume electron cloud is 2D Gaussian y v x! charge fraction is given by integral over pad area! 1 to 1 mapping between v and (f 1,f 8 ) PAD 1! One free parameter cloud size 26 October 2 Dean Karlen / Carleton University 11
Determining cloud size Ar CO 2 (~9:1) P1: Ar CH 4 (9:1) charge fraction in central pad 1.8.6.4 σ x = 37 ± 3 µm.2.5.6.7.8.9 1 1.1 1.2 1.3 1.4 y coordinate of collimator (mm) σ x = 53 ± 3 µm 26 October 2 Dean Karlen / Carleton University 12
Position from pad fractions! figure: 1 to 1 mapping from (x,y) to (f 1,f 8 )! invert the mapping to determine (x,y) from (f 1,f 8 ) y coordinate (mm) 2.2 2 1.8 1.6 1.4 1.2 1.8.6 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 Pad 1 fraction 9 Pad 8 fraction -.8 -.6 -.4 -.2.2.4.6.8 x coordinate (mm) 26 October 2 Dean Karlen / Carleton University 13
Charge sharing result Ar CO 2 6 5 4 x =.95 mm σ x =.43 mm (x,y) col = (-.1,1.143) mm ID Entries Mean RMS 11 846 -.9321E-1.6551E-1 96.18 / 58 Constant 55.89 Mean -.9525E-1 Sigma.4284E-1 5 4 y = 1.129 mm σ y =.61 mm ID Entries Mean RMS 12 846 1.126.7586E-1 51.72 / 55 Constant 46.62 Mean 1.129 Sigma.6122E-1 3 3 2 2 1 1 -.4 -.2.2.8 1 1.2 1.4 1.6 x coordinate y coordinate 26 October 2 Dean Karlen / Carleton University 14
Charge sharing result Ar CO 2 xmeas - xtrue (mm).1.5 -.5 -.1 -.15 -.2.8 1 1.2 1.4 1.6.1.5 -.5 -.1 -.15 -.2 First data from Aug 28 Last data from Aug 25.8 1 1.2 1.4 1.6 ymeas - ytrue (mm).2.15.1.5 -.5 -.1.8 1 1.2 1.4 1.6.2.15.1.5 -.5 -.1 y collimator (mm) First data from Aug 28 Last data from Aug 25.8 1 1.2 1.4 1.6 26 October 2 Dean Karlen / Carleton University 15
Charge sharing result P1! standard deviation again 5-6 µm not diffusion limited x meas x true (mm).1.5 -.5 -.1 -.15 -.2.5 1 1.5 y meas y true (mm).2.15.1.5 -.5 -.1 y collimator setting (mm).5 1 1.5! increase in σ y due to sensitivity of algorithm to cloud size 26 October 2 Dean Karlen / Carleton University 16
Localization from charge sharing! Method only works in regions where significant charge is deposited on 3 pads Ar CO 2 P1 Charge sharing regions Charge sharing region 26 October 2 Dean Karlen / Carleton University 17
Localization from induced pulses! The amplitude of the induced pulse from a charge cloud that falls on a neighbouring pad depends on the distance between the cloud and pad. distance to pad centre (mm) 4 3.75 Pad 2 Ar CO 2 Pad 8 3.5 Pad 3 3.25 3 2.75 Pad 7 Pad 4 Pad 6 2.5 2.25 2 1.75.2.3.4.5.6.7.8.9.1.11 4 3.75 3.5 3.25 3 2.75 2.5 2.25 2 P1 induced pulse / total charge pulse 1.75.2.3.4.5.6.7.8.9.1.11 26 October 2 Dean Karlen / Carleton University 18 Pad 2 Pad 8 Pad 3 Pad 7 Pad 4 Pad 6
Example! collimator location: (-.1,1.143) 3 2 1 -. 1 -. 2.2.4.6. 8.1 x 1-5 -1 -. 1 -. 2.2.4.6.8.1 x 1-5 -.5 -.1 -.15 -.2.2.4.6.8.1 x 1-5 -2-3 -3-2 -1 1 2 3 -.5 -.1 -.15.2.4.6.8.1 x 1-5 26 October 2 Dean Karlen / Carleton University 19
Induction signal results Ar CO 2.1.1 y meas y true (mm) x meas x true (mm).5 -.5 -.1 -.15 -.2.2.15.1.5 -.5.5 1 1.5.5 -.5 -.1 -.15 -.2.2.15.1.5 -.5 First data from Aug 28.5 1 1.5 Last data from Aug 25 σ x 5 8 µm σ y 6 1 µm σ smallest near centre of central pad biases likely due to cross-talk -.1 -.1.5 1 1.5.5 1 1.5 y collimator position (mm) 26 October 2 Dean Karlen / Carleton University 2
Induction signals results P1.1.2 x meas x true (mm).5 -.5 -.1 -.15 y meas y true (mm).15.1.5 -.5 -.2.5 1 1.5 -.1.5 1 1.5 y collimator position (mm) σ x 5 8 µm σ y 9 1 µm 26 October 2 Dean Karlen / Carleton University 21
Simulation work! Java simulation of GEM is in development! arbitrary GEM structure can be defined interactively! keeps track of electron distributions on pads! Comparison with P1 data:! predicts average cloud width of.5 mm compared to observed value of.56 mm! confirms that diffusion contribution to measurement spread less than 25 µm! spread in measurements due primarily to x-ray spot size 26 October 2 Dean Karlen / Carleton University 22
Simulation program 26 October 2 Dean Karlen / Carleton University 23
Future plans! Further studies with hexagonal pads! begin studies of track resolution! optimize GEM pad geometry for tracking! rectangles stretched along r direction (?)! small (15 cm) TPC under construction! FADC readout in development! eventually a large scale TPC test in B field 26 October 2 Dean Karlen / Carleton University 24
Summary! Interesting new results:! relatively large pads (2.5 mm diameter) can give excellent space point resolution (of order 5 µm) by using charge sharing! large pads are necessary for TPC application to keep the electronics channel count to a reasonable level! induction pulses (previously neglected) can provide similar resolution, but is more challenging! recovers lost resolution for clusters without charge sharing! GEM readout for TPC looks promising! 26 October 2 Dean Karlen / Carleton University 25