1. Low Mass Lepton Pairs in PHENIX Craig Woody BNL Upgrades Plenary Meeting June 14, 2002

2. C.Woody, Upgrades Plenary Meeting, 6/14/02 2 Issues for the TPC/HBD PHENIX Upgrade Need to prepare PHENIX LOI Input to RHIC II CD-0 document (draft due 7/31/02) BNL R&D proposal (due 9/20/02)  Physics Issues Measurement of low mass electron pairs and vector mesons Weizmann HBD proposal (June 2001) lays out a strategy for Dalitz rejection needed to do this, but we need to state the physics case for making this measurement, including issues of rates, triggering, etc. (physics discussed by Itzhak in several talks, Tony’s thoughts on trigger…) Providing a new inner tracking detector for PHENIX Need to make the case for improving global PHENIX tracking with an inner tracking detector (to do this right requires a detailed Monte Carlo)  R&D Issues

3. C.Woody, Upgrades Plenary Meeting, 6/14/02 3 Low Mass Electron Pairs at the SPS No enhancement in pp and pA collisions Strong enhancement of low-mass e + e - pairs in A-A collisions (wrt to expected yield from known sources) Enhancement factor (.25 <m<.7GeV/c 2 ) : 2.6 ± 0.5 (stat) ± 0.6 (syst)

4. C.Woody, Upgrades Plenary Meeting, 6/14/02 4 Interpretation of Low Mass Pairs 1. Thermal radiation from the hadron gas  annihilation :  +  -   *  e + e - 2. In-medium r broadening Rapp, Wambach et.al. 3. Dropping of the r mass G.E.Brown et.al.

5. C.Woody, Upgrades Plenary Meeting, 6/14/02 5 Low Mass Electron Pairs at RHIC Dilepton spectra from central Au-Au collisions at full RHIC energy R..Rapp Y. Akiba This calculation is being repeated using EXODUS, PYTHIA and PISA, including all backgrounds and a realistic field map C.Aidala, K.Ozawa and R.Averbeck

6. C.Woody, Upgrades Plenary Meeting, 6/14/02 6 TPC tracking coverage Df=2p -1.0<|h|<1.0 dp/p ~ 2% Inner Coil creates a “field free” (∫Bdl=0) region inside the Central Magnet PHENIX Inner Detector

7. C.Woody, Upgrades Plenary Meeting, 6/14/02 7 GEANT Simulation of TPC/HBD 100 MeV e - 20 cm 55 cm 70 cm CsI Photocathode C.Aidala

8. C.Woody, Upgrades Plenary Meeting, 6/14/02 8 Field map, inner region (r < 80 cm) C.Aidala

9. C.Woody, Upgrades Plenary Meeting, 6/14/02 9 Effects of the Non-Uniform Magnetic Field 50 MeV/c electron  = 85 o “Loopers” Low momentum tracks can be turned around by the outer field C.Aidala

10. C.Woody, Upgrades Plenary Meeting, 6/14/02 10 Occupancy of HBD (1% Si) Averaged over 200 events No SiSi No field (Tracking to 1 MeV) 19.550.1 No field (Tracking to 10 MeV) 17.543.9 No field (Tracking to 20 MeV) 16.342.3 Field (Tracking to 1 MeV) 28.459.8 Field (Tracking to 10 MeV) 25.356.5 Field (Tracking to 20 MeV) 22.651.6 4 cm C.Aidala

11. C.Woody, Upgrades Plenary Meeting, 6/14/02 11 Radiography of Photon Conversions C.Aidala All conversions Conversions from primary photons only Integrated over 2 p, e + or e - having at least one hit in TPC R (cm) Z(cm) R (cm) Z(cm)

12. C.Woody, Upgrades Plenary Meeting, 6/14/02 12 e+e+ e-e- TPC/HBD e-e- e+e+ p p p V0V0 measured in outer PHENIX detectors (P e > 200 MeV/c) Operate PHENIX with low inner B field to optimize measurement of low momentum tracks Identify signal electrons (low mass pairs, r,w,f, …) with p>200 MeV in outer PHENIX detectors Identify low momentum electrons (p<200 MeV) using dE/dx from TPC and/or Cherenkov light in HBD Calculate effective mass (or opening angle) between all opposite sign tracks identified as electrons ( e electron > 0.9, p rej > 1:200) Reject pair if mass < 130 MeV (or q < 200 mrad) Must provide sufficient Dalitz rejection (>90%) while preserving the true signal Strategy for Low Mass Pair Measurement

13. C.Woody, Upgrades Plenary Meeting, 6/14/02 13 Vector meson Survival Probability    J/  Y Survival probability of  is 70% for dN/dy = 650, when we keep 90% Dalitz rejection ratio. K.Ozawa (U.Tokyo)

14. C.Woody, Upgrades Plenary Meeting, 6/14/02 14 The TPC as an Inner Tracking Detector in PHENIX Expect d p/p ~ 2% (300 m m or better space points, ~35 pad layers) Would provide tracking resolution comparable with the silicon tracker over 2 p in azimuth and | h | < 1 Tracking through the central field (in normal running conditions) would give better rejection against false high p T tracks - second independent momentum measurement - can observe decays, conversions, etc… Tracking through the highly non-uniform “field free region” would give better association of Cherenkov “blobs” with electrons in HBD - field would be optimized to measure low momentum tracks - could make effective mass cut rather than just opening angle cut - provides dE/dx information for e/ p separation for p< 200 MeV/c

15. C.Woody, Upgrades Plenary Meeting, 6/14/02 15 Interplay Between the TPC/HBD and the New Proposed Silicon Vertex & Tracking Detector The silicon vertex and tracking detector potentially adds a significant amount of material in front of the HBD which will increase the background for the electron pair measurement Can we provide the same level of tracking with the TPC with less mass, use it in combination with the HBD for the electron pair measurement, and get by with only a silicon vertex detector ? Must agree on geometrical boundaries for both detectors (e.g.r min ) Can the silicon strip/pad tracking detector work for heavy ions ?

16. C.Woody, Upgrades Plenary Meeting, 6/14/02 16 Rates for Vector Mesons AuxAu Min. Bias: dN( p 0 )/dy=80, dN( r 0 )/dy=16, r 0 -> e + e - = 4.5x10 -5 PHENIX acceptance for f -> e + e - ~ 3.5% (assume same for r 0 ) N( r 0 ) = 2.5 x 10 -5 per min bias event 240 m b -1 => 1.74x10 9 min. bias events => N( r 0 ) = 44,000 (c.f. 8,000 J/ y ) Archive rate: 60 MB/s, 20% m.b.,180 kB/m.b. evt => 66 Hz m.b. archive rate 8 weeks at 25% overall duty factor => 8x10 7 m.b. events => N( r 0 ) = 2000 Assume Level 2 trigger using HBD => 8 kHz to EvB 8 weeks at 25% overall duty factor => 9.7x10 9 events to EvB => N( r 0 ) = 240,000 With Level 1 trigger -> 7x10 10 m.b.events => N( r 0 ) = 1.74x10 6 T.Frawley UGW 5/7/02

17. C.Woody, Upgrades Plenary Meeting, 6/14/02 17 R&D Issues for the PHENIX TPC/HBD Study performance of GEM detectors - current study (B. Yu) has shown some effects which need to be understood - how to get potentially excellent spatial resolution for the TPC - need to study and learn to build GEMs with CsI photocathodes - is the HBD really “hadron blind” ? Studies with fast drift gases (CF 4,CH 4,…) - Measure drift velocities, drift lengths, diffusion parameters, dE/dx, etc - Measure optical transmission extending into the VUV (B. Azmoun) Electronics development A significant effort will be required to develop readout electronics for the TPC - new front end ASIC design - new FEM which will include zero suppression (same for silicon) Electronics for the HBD may be simpler, but will also require significant development - lower noise due to smaller primary signal, larger pads,… - must be low mass (part will sit in the PHENIX acceptance)

18. C.Woody, Upgrades Plenary Meeting, 6/14/02 18 Time Scale and Cost TPC and HBD are separate features of the same detector However, if we want both, they must be integrated from the beginning Implementation could be staged, starting with HBD - different time scales (driven by development of TPC readout electronics) Prototype HBD in 2004 ? Upgrade to TPC by 2006-2007 Costs - TPC electronics development could cost several $M - Total cost of 3-5 $M is not out of line with NSAC plan - R&D is presently being funded out of joint LDRD with STAR (looks good to continue on for next year) - Needs additional DOE detector R&D money in FY03 (BNL proposal 9/02) - Likely no DOE construction money until FY04

19. C.Woody, Upgrades Plenary Meeting, 6/14/02 19 Additional Slides

20. C.Woody, Upgrades Plenary Meeting, 6/14/02 20 Rate Issues for TPC Track density in pp dN ch /dy = 2.6 L x s = 2 x 10 32 x 60 mb ( S s = 500 GeV) = 12 MHz (min bias) 12 MHz => 12 events/ m sec => ~50 m.b. events per drift time 12 MHz x 2.6 trks x 1.5 = 47 x 10 6 trks/sec = ~ 47 trks/ m sec Au-Au L ~ 8 x 10 27 cm -2 s -1 O-O L ~ 1.6 x 10 29 cm -2 s -1 p-p L ~ 2 x 10 32 cm -2 s -1 (possibly -> 4 x 10 33 cm -2 s -1 ) Track density in HI dN ch /dy = 150 (min bias) L x s = 8 x 10 27 x 7.2 b = 58 kHz 58kHz x 150 trks x 1.5 = 13 x 10 6 trks/sec = 13 trks/ m sec 47 trks/ m sec x 4 m sec = 188 trks 188 trks x 35 pts(max)/trk x 12 bytes/pt ~ 80 kB

21. C.Woody, Upgrades Plenary Meeting, 6/14/02 21 Number of channels A plane = p (55 2 - 20 2 ) = 8247 cm 2 A pad = 1.0 x 0.2 = 0.2 cm 2 N pad = 8247/0.2 = 41,233 x 2 = ~ 80K chs Readout Speed in PHENIX Buffer size < 40 beam clock ticks 40 x 100 ns (10 MHz) = 4 msec (EMCAL pushes this to 6.4 msec) Readout time < 40 msec (25 KHz in DCM) 800 Kb => 6.4 x 10 6 bits/40 x 10 -6 sec = 1.6 x 10 11 bits/sec = 160 Gbits/sec (20 Gb/sec) => 160 1Gbit fibers (TEC has 128) => 40 DCMs (TEC has 28) Data Volume 4 msec/20 ns => 200 time samples (8 bits) 80K x 200b ~ 16 Mb Assume zero suppression of 1/20 => 800 Kb (actual hit rate gives 80 Kb => 1/200) Trigger Expect actual trigger rate in pp to be ~10 4 Hz at L = 2 x 10 32 (trigger rate of interesting events including W,Z,charm is only ~ 10 3 Hz) If data volume after final zero suppression is 80 Kb, then 80 Kb x 10 4 /sec => 800 Mb/sec => need Level 3 trigger Data Volume and Readout Speed for TPC

22. C.Woody, Upgrades Plenary Meeting, 6/14/02 22 GEM Detector Issues CsI photocathode deposited on outer GEM foil Multistage GEM used to detect single photoelectons Is the HBD really “hadron blind” GEM can give excellent position resolution (~ 100 mm), but requires large number of readout channels for TPC Needs to spread charge out on readout plane

23. C.Woody, Upgrades Plenary Meeting, 6/14/02 23 GEM Detector Studies Gas Electron Multiplier 10x10 cm 2 GEM from CERN B.Yu

24. C.Woody, Upgrades Plenary Meeting, 6/14/02 24 TPC Drift Cell Constructing drift stack similar to one being tested for LEGS LEGS PHENIX / STAR

25. C.Woody, Upgrades Plenary Meeting, 6/14/02 25 Gas Transparency and Photocathode Studies in the VUV VUV Spectrometer CF 4 Absorbance of CF 4 Wavelength (angstroms) VUV measurement (normalized) Typical R1259 99.997 purity CsI Quantum Efficiency B.Azmoun

26. C.Woody, Upgrades Plenary Meeting, 6/14/02 26 TPC Readout Electronics Readout Pads D R ~ 1 cm f ~ 2 mm P.O’Connor