CMS Tracker studies Daniel Pitzl, DESY Present CMS silicon tracker Design Material budget Upgrade phase I: 4 layer pixel 5 layer pixel? Resolution studies with broken line fits
CMS Si Tracker 2
Phase I upgrade Alessia Tricomi Tracker Upgrade Sep 2009 http://indico.cern.ch/conf erencedisplay.py? confid=47293 3
CMS Si Tracker The TIB consists of four concentric cylinders placed at radii of 255.0 mm, 339.0 mm, 418.5 mm, and 498.0 mm, respectively, and extending from 700 mm to +700 mm along the z axis. The two innermost layers have double sided modules while the outer two layers have single sided modules. TIB uses 320µm thick silicon micro strip sensors. The strip pitch is 80µm on layers 1 and 2 and 120µm on layers 3 and 4 in the TIB, leading to a single point resolution of 23µm and 35µm, respectively. The TOB has 6 layers of rods with average radii of 608, 692, 780, 868, 960, and 1080 mm. The TOB extends in z between ±118cm. TOB uses 500µm thick micro strip sensors with strip pitches of 183µm on the first 4 layers and 122µm on layers 5 and 6. It provides another 6 r measurements with single point resolution of 53mm and 35µm, respectively. CMS detector, JINST 3 (2008) S08004 A Satpathy, J. Phys. Conf. Ser. 110 (2008) 092026 4
reconstructed photon conversions Nancy Marinelli, http://arxiv.org/abs/0710.2818v1 5
CMS tracker material All trackers Barrel pixel Upgrade: factor 2 less in center factor 4 less in endcaps pixel note 2009 6
CMS tracker resolutions CMS detector, JINST 3 (2008) S08004 7
Broken line fit Track model: broken helix from V. Blobel, implemented by C. Kleinwort. Takes multiple scattering and intrinsic resolution into account. Covariance matrix of helix curvature and impact parameter to the beam is calculated. Volker Blobel, Nuclear Instruments and Methods A566 ( 2006) 14 17 8
CMS tracker momentum resolution broken line simulation agrees with CMS paper for present. 9
CMS impact parameter resolution broken line simulation agrees with CMS paper for present. 10
CMS pixel upgrade: 4 layers S. König 11
CMS pixel upgrade: 4 layers 2 identical half-shells. 1184 modules (79M pixels) (1.6 present barrel) R1 = 29 mm, 96 modules (if R=22 mm beam pipe is possible) PSI R2 = 68 mm, 224 modules PSI R3 = 109 mm, 352 modules DESY R4 = 160 mm, 512 modules Italy 12
CMS pixel tracker resolution study Pixel stand alone tracking for the high level trigger. present: R X0 mm % BP 29 0.5 4 layers: R X0 mm % BP 25 0.5 4 thinner: R X0 mm % BP 25 0.5 4 small BP: R X0 mm % BP 22 0.5 L1 44 L2 73 L3 102 L1 39 L2 68 L3 109 L4 160 L1 39 L2 68 L3 109 L4 160 L1 29 L2 68 L3 109 L4 160 2.5 2.5 2.5 2.5 2.5 2.5 2.5 13 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
CMS Pixel charge sharing W. Erdmann 14
CMS pixel momentum resolution RBP = 29 mm RBP = 25 mm (pt) / pt = 6.4% 2.4% pt [GeV] RBP = 22 mm (pt) / pt = 2.5% 0.4% pt [GeV] 15
CMS pixel impact parameter resolution RBP = 29 mm RBP = 25 mm (dca) = 160 µm / pt 72 µm RBP = 22 mm (dca) = 62 µm / pt 20 µm 16
Radiation damage RBP = 22 mm 17
Pixel tracker design study vary R2 for a 3 layer detector R2 = (R1+R3)/2 is optimal R2 1.1(R1+R3)/2 is optimal 18
Pixel tracker design study vary R3 for a 3 layer detector pt resolution improves like 1/(R3 R1)2 increasing R3 improves dca resolution! 19
Pixel tracker design study vary R1 for a 3 layer detector pt resolution degrades like (R3 R1)2 reducing R1 improves dca resolution 20
Pixel tracker design study vary R2 and R3 for a 4 layer detector pt optimum: R2 = 70, R3 = 115 mm dca optimum: R2 = 85, R3 = 115 mm 21
track seeding 3 of 4 3 of 3 Alessia Tricomi CMS Tracker week Oct 2009 http://indico.cern.ch/conferencedisplay.py?confid=47301 22
B tagging ttbar no pile up present upgrade at uds = 0.01 b from 0.58 to 0.64 Alessia Tricomi Tracker Upgrade Sep 2009 http://indico.cern.ch/conferencedisplay.py?confid=47293 23
LHC beam size = E s transverse beam size s optical beta function, * = at IP = N / dynamic emittance = E /m p boost factor N = 3.75 µm rad normalized emittance https://edms.cern.ch/file/445830/5/vol_1_chapter_2.pdf Ep [TeV] 3.5 5 5 7.5 β* [m] 2 2 1 0.5 σ [μm] 44.8 37.5 26.5 15.3 24
ALICE beam pipe Smaller beam pipe diameter for better c and b tagging desired: From R=2.9 cm to R=2.0 or even R=1.3 cm like CDF. Thinner beam pipe desired: From 0.8 to 0.4 mm Be. Thinner pixel layers. ~9 month shutdown. Discussed in LEB group: postpone. 25
Summary The CMS tracker resolution can be understood analytically using the broken track formalism. The pixel stand alone track resolution can be improved substantially by adding a 4th layer. Important for the High Level Trigger? Layer placement: Inner layer as close as possible to the beam: smaller beam pipe! Outer layer as far out as affordable: lever arm. 2nd and 3rd layer near the middle, for tracking redundancy. 2nd layer not too close to the 1st layer (only adds multiple scattering). A 5th layer in the middle adds redundancy and material: low priority? Impact of a 4th layer and a smaller beam pipe on physics analyses to be studied: Better b tagging performance? 26