A Diamond Micro-strip Electron Detector for Compton Polarimetry. Amrendra Narayan Mississippi State University on behalf of Hall C Compton Team

Transcription

1 A Diamnd Micr-strip Electrn Detectr fr Cmptn Plarimetry Amrendra Narayan Mississippi State University n behalf f Hall C Cmptn Team

2 Outline Qweak Plarimetry requirements Hall C Cmptn Overview Electrn detectr Data Acquisitin Analysis apprach Preliminary results PAVI 11 Mississippi State University 2

3 Qweak and Plarimetry The Qweak experiment aims t measure the weak charge f the prtn with a precisin f 4.1%, by measuring the parity vilating asymmetry in plarized e-p elastic scattering with a precisin f 2.5% Qweak Errr Budget Uncertainty δapv/apv δqw/qw Statistical (~2500 hurs at 150 μa) 2.1% 3.2% Systematic: 2.6% Hadrnic structure uncertainties 1.5% Beam plarimetry 1.0% 1.5% Effective Q 2 determinatin 0.5% 1.0% Backgrunds 0.5% 0.7% Helicity-crrelated beam prperties 0.5% 0.7% Ttal: 2.5% 4.1% Qweak talk: Katherine Myers, Sept 8 ne f the largest experimental cntributin t the errr budget The Hall-C Mller plarimeter is the highest precisin plarimeter at JLab, hwever it is peridic, invasive and perates nly at lw currents.. The new Cmptn plarimeter is cntinuus, nn invasive and can perate at high currents. PAVI 11 Mississippi State University 3

4 Overview: Cmptn Layut Parameter Value Beam Energy 1.16 GeV Laser Wavelength 532 nm Chicane bend angle 10.1 deg Electrn free drift distance 1.6 m Max. Electrn Displacement 17 mm Cmptn edge energy 46 MeV PAVI 11 Mississippi State University 4

5 Overview: Laser Table Installed laser table Test setup fr Fabry-Pert cavity Installed laser table Phtn target at center f chicane is Cherent Verdi 10W laser lcked t lw gain Fabry-Pert cavity Pwer in the cavity is ~ 1kW laser plarizatin > 99% lw reflectivity mirrr in Fabry-Pert cavity allws rbust measurement f laser plarizatin curtesy: Dnald Jnes, University f Virginia PAVI 11 5

6 Overview: γ-detectr Integrated Yield (a.u.) Integrated yield (a.u.) Laser cycled n and ff with a perid f ~ 140 s γ - detectr signal is integrated with n threshld t eliminate sensitivity t gain drift Tried CsI crystal at first but fund that phsphrescence with ms t secnd timescales diluted ur measurement Currently using PbW crystal detectr. Less energy reslutin but fr signal integratin this is nt an issue. Achieving <1% statistical uncertainty in a few hurs. Laser On Laser Off Time (secnds) Fit t Yield frm CsI during Beam Trip Time Cnstants 0.2s (20%) 1.3s (12%) 6.3s (68%) Time (a.u.) PAVI 11 curtesy: Dnald Jnes, University f Virginia 6

7 Overview: e-detectr Thrugh the γ-detectr and e-detectr we have tw independent measurements having different uncertainties hence being a gd crss-check n each ther We use diamnd micr-strip detectr fr detecting the Cmptn scattered electrns We have 4 planes f the detectr t allw cincidence measurements PAVI 11 Mississippi State University 7 e

8 e-detectr: wrking The detectr uses Diamnd which is artificially grwn using Chemical Vapr Depsitin Hw des it wrk as a detectr? Operatin f Diamnd Detectrs Bias vltage ~ 1000 V, Charge cllectin distance (d) ~ 250 μm PAVI 11 Mississippi State University 8

9 Diamnd micr-strip detectrs alumina (ceramic) used fr carrier bard metallizatin (n diamnd) dne with TiPtAu detectr dimensins : 21 mm x 21 mm detectr thickness: 500 µm each detectr plate has 96 strips strip pitch is 200 µm. PAVI 11 Mississippi State University 9

10 e-detectr: installatin PAVI 11 Mississippi State University 10

11 e-detectr: installed upstream view dwnstream view PAVI 11 Mississippi State University 11

12 e-detectr DAQ Diamnd micr - strip detectrs Amplificatin, shaping and digitizatin f the signal Trigger prcessed using FPGA based v1495 This is the first Diamnd micr-strip detectr t be used as a tracking device in an experiment PAVI 11 Mississippi State University 12

13 e-detectr DAQ Diamnd micr - strip detectrs Gain : 200 mv (10x10 3 ) x (1.6x10-19 ) = 120 mv / fc Amplificatin, shaping and digitizatin f the signal QWAD bards custm made by TRIUMF Trigger prcessed using FPGA based v1495 test input after charge integratin & amplificatin digital utput ~ 1 mv ~ 200 mv LVDS ut PAVI 11 Mississippi State University 13

14 e-detectr DAQ Diamnd micr - strip detectrs Amplificatin, shaping and digitizatin f the signal Trigger prcessed using FPGA * based v1495 v1495 : CAEN general purpse lgic mdules. The mdule was prgrammed fr trigger generatin and data readut using VHDL * Field Prgrammable Gate Array PAVI 11 Mississippi State University 14

15 e-detectr DAQ : schematic typically ~9000 electrns Cnditins the signal t avid verlap f successive events, hence aviding false triggers amplifies, shapes and digitizes the detectr pulse Decides if a pulse crrespnds t a real event Based n the decisin f Trigger Lgic, the crrespnding signal is accepted/rejected at the P. buffer Trigger Lgic charge pulse frm detectr 0.56 m flex cable Digitizatin f signal (QWAD) Signal Recnditining Pattern Buffer Read ut Delay Line ~ 60 m shielded twisted pair cable Delays the input signal fr the trigger decisin t be made. The Pattern Buffer acts as a temprary strage t allw the slw Readut prcess thrugh the VME backplane detectr Qweak Amplifier Discriminatr FPGA bard CPU PAVI 11 Mississippi State University 15

16 e-detectr DAQ: Trigger signals frm tw cnsecutive strips in a detectr plane Plane 1 Plane 2 Plane 3 Trigger cnditin: (2 in 4 planes) The fundamental repetitive lgic unit Plane 4 t suppress backgrund, we require a cincidence between multiple planes default trigger is hits n 2 ut f 4 planes we lcalize the trigger in a single detectr plane t 4 cnsecutive strips PAVI 11 Mississippi State University 16

17 charge nrmalized hits plane 2 Charge nrmalized strip hit Beam Current: 150 μa plane 3 plane 4 Red: Laser On Black: Laser Off Green: Backgrund subtracted we had 3 active planes during Qweak Run 1 Cmptn edge at Strip 50 Strip number PAVI 11 Mississippi State University 17

18 Asymmetry (A) Asymmetry Laser n : Laser ff :: 2 : 1 where and Laser OFF Run time : 84 min Run # : Beam : 150 μa IHWP : in Thery Plarizatin: 90.4 % +/- 0.7% using Plane-2 Strip number Distance frm beam (mm) PAVI 11 Mississippi State University 18

19 we knw: e-beam energy, magnet field map, detectr lcatin & gemetry Calculating Plarizatin deviatin frm nminal e-beam mmentum f Cmptn recil e - Cmptn asymmetry is precisely knwn frm QED as a functin f mmentum Fitting this theretical asymmetry t the measured asymmetry gives us the beam plarizatin PAVI 11 Mississippi State University 19

20 Asymmetry (A) Calculating Plarizatin The Cmptn edge fr the theretical Cmptn asymmetry is fixed at 17.6 mm frm the beam (based n knwn beam parameters and detectr gemetry) Plarizatin is btained by perfrming a tw parameter fit with plarizatin and effective pitch where Plane 2 Plane 3 Plane 4 Thery Plarizatin: 90.4 % +/- 0.7% effective pitch parameter fit Distance frm beam (mm) PAVI 11 Mississippi State University 20

21 time in min Plarizatin(%) Preliminary Plarizatin Errr (%) PRELIMINARY Spt change reactivatin 42 days f Cmptn data each pint represents a ~ 1 hr run nly Partial systematic errr (due t strip pitch) included the dtted vertical lines represent spt changes n the phtcathde the dashed vertical line represents Re-activatin n an average the beam current was ~ 160 μa Run Number 42 days Apr 1, 2011 May 12, 2011 PAVI 11 Mississippi State University 21

22 Plarizatin(%) Plarizatin(%) Quantum efficiency Zming int a regin f cnsecutive spt changes: Plarizatin was fund t drp significantly befre the spt mve Time (a.u) Quantum Efficiency (%) PAVI 11 Mississippi State University 22

23 Systematic errrs Errr Cntributin (~)Value Due t detectr strip size 0.2 % Detectr gemetry 0.15 % Difference between planes 0.2 % Magnetic field? Beam & Laser Psitin? Dead time? TBD? Laser Plarizatin /- 0.4% (verall) Ttal 0.50 % We dn t expect the unknwn in the abve table t be very large PAVI 11 Mississippi State University 23

24 Summary Accmplished: This is the first Diamnd micr-strip detectr t be used as a tracking device in an experiment Despite several challenges psed by the electrnic nise envirnment, leading t strict trigger cnditin, we achieved the design gal f < 1% statistical uncertainty and prjected lw systematic Next: In ur preparatin fr Qweak run-2, We have 4 active planes (already installed) Adapting frm experiences f run-1, we are using mre niserbust electrnics, with a better cntrl ver signal crrelatins in adjacent channels. All set t prvide an independent abslute plarizatin measurement fr Hall-C beam PAVI 11 Mississippi State University 24

25 Cmptn Team Institutins invlved: 1. Cllege f William and Mary (γ - detectr) 2. Jeffersn Lab (all subsystems) 3. Mississippi State University (e - detectr) 4. MIT Bates (magnets, vacuum can, detectr hlder, previus CsI crystal fr γ - detectr) 5. TRIUMF (Qweak Amplifier Discriminatr bards) 6. University f Manitba (e - detectr) 7. University f Virginia (laser and γ - detectr) 8. University f Winnipeg (e - detectr) 9. Yerevan Physics Institute (γ - detectr and help with e - detectr) Alphabetical rder This wrk was supprted by U.S. DE, Grant Number: DE-FG02-07ER41528 PAVI 11 Mississippi State University 25

26 Thanks Qweak Cllabratin, July Cllege f William and Mary This presentatin was made pssible due t significant cntributin frm Dipangkar Dutta, Vladas Tvaskis and Dnald Jnes The authr can be cntacted at narayan@jlab.rg PAVI 11 Mississippi State University 26

27 Extras PAVI 11 Mississippi State University 27

28 charge nrmalized laser ff hits charge nrmalized laser ff hits Backgrund subtractin N N N N Laser On Laser Off Laser On Laser Off N N N N Laser On Laser Off Laser On Laser Off Time Time Time Time Laser On Laser Off Laser On Laser Off / Time / Time / Time / Time Beam Beam Off Beam Off Off Beam Off N N N N Beam Beam Off Beam Off Off Beam Off befre electrnic nise subtractin After electrnic nise subtractin Run # Beam ON Laser Off Run # Beam ON Laser Off strip number strip number PAVI 11 Mississippi State University 28

29 why diamnd? Prperty Silicn Diamnd Band Gap (ev) Electrn/Hle mbility (cm 2 /Vs) 1450/ /1600 Saturatin velcity (cm/s) 0.8x10 7 2x10 7 Breakdwn field (V/m) 3x x10 7 Dielectric Cnstant Displacement energy (ev) Lw leakage current, shrt nise Fast signal cllectin Lw capacitance, nise Radiatin hardness e-h creatin energy (ev) Av. e-h pairs per MIP per micrn Charge cllectin distance (micrn) full ~250 Smaller signal PAVI 11 Mississippi State University 29

30 2 parameter fit x = (strip # f edge - strip # f histgrammed bin)*strip_pitch x = (strip # f edge - strip # f histgrammed bin)*strip_pitch * P2 PAVI 11 Mississippi State University 30

31 Input Recnditining Stage Input frm Detectr Disable Trigger width Rising Edge Detectin Pipeline width Pulse Width Adjustment f Trigger utput (cunting circuit) Pulse Width Adjustment f Pipeline Delay (cunting circuit) Input t Trigger Lgic Circuit Input t Pipeline Delay Circuit Hld Off Hld Off decisin (cunting circuit) PAVI 11 Mississippi State University 31

32 DAQ We accumulate the cunts in the detectr ver a given Helicity windw and read it ut at the end f the Helicity windw Our Helicity reversal rate is ~ 960 Hz wait time fr Helicity stabilizatin ~ 76 μs PAVI 11 Mississippi State University 32