Full Silicon Tracking Studies for CEPC

Transcription

1 Full Silicon Tracking Studies for CEPC Weiming Yao (IHEP/LBNL) for Silicon Tracking Study Group CEPC-SppC Study Group Meeting, September 2-26, Beihang University cepc/cepc twiki/index.php/pure Silicon Detector /

2 Outline Introduction. Silicon tracker designs and their performances. Detector simulation and reconstruction. Comparing with CEPC V performance. Conclusion 2/

3 Introduction CEPC full silicon tracker has been implemented in Mokka (Chengdong). Based on CEPC V silicon elements, we replace TPC with additional SIT layers and FTD endcaps. The advantage is to recycle the ILD silicon tracking, which seems work out of box. The current design means to prove a principle. But, it would allow us to re-optimize the design and improve tracking. ILC SID vs CEPC: B=T.T, r max =.2.4m, Barrel strip single double sided. /

4 Full Silicon Tracker Concept We compared the tracking performance of several design options using a toy MC. CEPCSID geometry r (m) eta =. eta =. eta = 2. eta =. eta = X/X Material vs. eta for test layouts z = cm z = cm z = + cm z (m) Eta Figure: CEPC full silicon tracker with rad length up to 7% 4/

5 Excellent Resolutions The expected resolutions from toy simulation is better than ILC SID. pt resolution vs. pt for test layout d resolution vs. pt for test layout z resolution vs. pt for test layout pt sigma /pt [%] e e e+ 8 SID 8 CEPCSID 2 SID 2 CEPCSID d resolution [microns] SID 8 CEPCSID 2 SID 2 CEPCSID z resolution [microns] SID 8 CEPCSID 2 SID 2 CEPCSID GeV/c GeV/c GeV/c Figure: Resolutions for /pt, d, and z. /

6 Full Detector Simulation and Reconstruction Generated single muon in CEPC full silicon. Reconstructed using Marlin Silicon only. Modifying pattern recognition to use more silicon layers. R (mm) z (mm) (a) CEPC V(Default) R (mm) 2.. silicon only -2-2 z (mm) (b) CEPC full silicon Default Silicon 2 Nhit (c) Number of hits 6/

7 Tracking Efficiencies Requiring P T >. GeV and.8 < θ < Efficiency is bit lower in barrel and endcap overlap region for full silicon, which could be improved. Note the plots made after a quick bug fix in FTD geometry, but it requires more study. Eff. Default Fixed Eff. Default Fixed Eff. Default Fixed. Silicon Fixed. Silicon Fixed. Silicon Fixed Theta Phi Figure: Efficiencies vs pt, theta and phi 7/

8 Resolution The pt resolution seems comparable. d(/) Resolution -.9 Default Fixed Barrel.8 Silicon Fixed d(/) Resolution Default Fixed Endcap 2. Silicon Fixed Figure: resolution in Barrel and Endcap regions 8/

9 d Resolution d resolution is quite similar in barrel, bit worse in endcap. D Resolution(mm) - 9 Default Fixed Barrel 8 Silicon Fixed D Resolution(mm) Default Fixed Endcap 2 Silicon Fixed Figure: d resolution in Barrel and Endcap regions 9/

10 z Resolution z resolution is quite similar in barrel, bit worse in endcap. Z Resolution(mm) Default Fixed Barrel 6 Silicon Fixed Z Resolution(mm) 6 Default Fixed Endcap Silicon Fixed Figure: z resolution in Barrel and Endcap regions /

11 d, z, and Omega Pulls The pull width seems off at.8, which is sensitive to materia and resolution in simulation and reconstruction. Events 2 2 Constant.64e+4 ± 7.8e+ Mean ±.286 Sigma.826 ±.2 Events 2 2 Constant.8e+4 ± 6.87e+ Mean.64 ±. Sigma.864 ±.2 Events 2 2 Constant.62e+4 ± 7.8e+ Mean ±.294 Sigma.84 ± D_pull - - Z_pull - - Omega_pull Figure: Track parameter pull for d, z, and omega /

12 Phi and Theta Pulls The pull width also seems low, which needs to be understood. Events Constant 2.6e+4 ±.4e+2 Mean -.2 ±.9 Sigma.7 ± Phi_pull Events Constant.988e+4 ± 9.49e+ Mean.6928 ±.2289 Sigma.626 ± Theta_pull Figure: Track parameter pull for phi and theta 2/

13 Conclusion The concept of full silicon tracker has been implemented and seems working. Its single particle performance is comparable to CEPC V, meeting the physics requirements. There are rooms for improvement, especially for improving silicon clustering and fitting. Allow us to start reoptimizing its design and improving silicon tracking for CDR. /