The WSU LC R

1
The WSU LC RD program
Rene Bellwied, Dave Cinabro, Vladimir Rykov
Wayne State University

• Who are we ?
• What have we done ?
• What would we like to do ?
• Hardware and Software

2
The WSU LC RD program

• Mix of NSF funded HE group and DoE funded Nuclear
  group
• Interest application of Silicon technologies to
  large area solid state tracking.
• Group was funded by Prescott Committee and NSF in
  the past two years to conduct LC RD. Vladimir
  was partially funded by this grant.
• Nuclear group designed, constructed, installed
  and operates the STAR-SVT (7 Million project, 50
  people from 9 institutions, project started in
  1993 and was completed in 2001, Rene Bellwied was
  project leader throughout this time.
  Collaborating institutions BNL, LBNL, Ohio
  State, University of Texas in Austin, Sao Paulo,
  Dubna, Protvino, Warsaw University)

3
The SVT in STAR
Construction in progress
Connecting components
4
The SVT in STAR (Feb.2001)
The final device.
and all its connections
5
STAR-SVT characteristics

• 216 wafers (bi-directional drift) 432 hybrids
• 3 barrels, r 5, 10, 15 cm, 103,680 channels,
  13,271,040 pixels
• 6 by 6 cm active area max. 3 cm drift, 3 mm
  (inactive) guard area
• max. HV 1500 V, max. drift time 5 ms, (TPC
  drift time 50 ms)
• anode pitch 250 mm, cathode pitch 150 mm
• SVT cost 7M for 0.7m2 of silicon (3 year RD, 5
  year construction)
• Radiation length 1.4 per layer
• 0.3 silicon, 0.5 FEE (Front End Electronics),
• 0.6 cooling and support. Beryllium support
  structure.
• FEE placed beside wafers. Water cooling.

6
SDDs 3-d measuring devices(a solid state TPC)
7
A typical pattern on a hybrid for a
central Au-Au event

• central event inner layer 15 hits/hybrid
  (middle 8 hits, outer 5 hits)
• overall track multiplicity 1000/event

8
Typical SDD Resolution
9
Wafers B and T dependence

• Operated at B6T in E896 at the AGS. B fields
  parallel to drift increase the resistance and
  slow the drift velocity.
• The detectors work well up to 50oC but are also
  very T-dependent. T-variations of 0.10C cause a
  10 drift velocity variation
• Detectors are operated at room temperature in
  STAR.
• We monitor these effect via MOS charge injectors

10
Present status of technology

• STAR (completed in 2001)
• 4in. NTD material, 3 kWcm, 280 mm thick, 6.3 by
  6.3 cm area
• 250 mm readout pitch, 61,440 pixels per detector
• SINTEF produced 250 good wafers (70 yield)
• ALICE (to be completed in 2006)
• 6in. NTD material, 2 kWcm, 280 mm thick, 280 mm
  pitch
• CANBERRA produced around 100 prototypes, good
  yield
• Future (NLC)
• 6in. NTD, 150 micron thick, any pitch between
  200-400 mm
• 10 by 10 cm wafer

11
Silicon detector option for LCD
Central tracker Silicon Drift DetectorsFive
layers Radiation length / layer 0.5
sigma_rphi 7 mm, sigma_rz 10
mm             Layer Radii    Half-lengths
            -----------    ------------
             20.00 cm      26.67 cm
             46.25 cm        61.67 cm
             72.50 cm        96.67 cm
             98.75 cm       131.67 cm
            125.00 cm       166.67 cm 56 m2
Silicon Wafer size 10 by 10 cm of Wafers
6000 (incl. spares) of Channels 4,404,480
channels (260 mm pitch)        
12
Tracking efficiencies LD vs. SD

• Tracking efficiencies
• For 100 hit efficiency (95.30.13)
• For 98 hit efficiency (94.50.14)
• For 90 hit efficiency (89.50.20)
• ? LD ?

• ? SD ?
• Tracking efficiencies
• For 100 hit efficiency (97.30.10)
• For 98 hit efficiency (96.60.12)
• For 90 hit efficiency (92.70.16)

13
Momentum studies (LD / SD)
LD ?
SD ?
log10(Pt, GeV/c)
log10(Pt, GeV/c)
14
Missing and ghost energies

• For hit efficiency 100
• Missing energy (11.70.6) GeV
• (7.10.3)
• Ghost energy (19.60.8) GeV
• (13.10.6)
• ? LD ?

• ? SD ?
• For hit efficiency 100
• Missing energy (5.70.4) GeV
• (3.30.2)
• Ghost energy (4.80.4) GeV
• (2.90.2)

15
Preliminary conclusions

• Momentum resolution
• The SD option has slightly better resolution at
  high momentum and slightly worse resolution at
  low momentum compared to LD
• With the existing 3d tracking and pattern
  recognition software (Mike Ronan et al.) the SD
  option has a slight advantage in tracking
  efficiency, shows less missing and ghost energy,
  and less ghost tracks)

16
Track Timing at ee- Linear Collider with the
Silicon Drift Detector Main TrackerR. Bellwied,
D. Cinabro, V. L. RykovWayne State
University,Detroit, Michigan
Chicago LC Workshop, Chicago, Illinois, January
7-9, 2002 V. L.
Rykov, Wayne State University
17
Conclusion of track-timing study(hep-ex/0202030,
submitted to NIM)
 

• It is shown that, with the SDD based central
  Main Tracker for the detector at ee- Linear
  Collider, the track selection and timing is
  possible at the nanosecond and even
  sub-nanosecond level.
• This means that, even at the NLC and/or JLC with
  the bunch spacing at 1.4 ns, each high-PT track
  can be assigned to a particular bunch crossing at
  the confidential level of up to 2?.
• For the considered here 5-layer central Main
  Tracker, it is suggested to make layers 1, 2, 3
  and 5 drifting along z-axis, but layer 4 drifting
  along the azimuth (?-axis) with effectively no
  negative impact on the trackers momentum
  resolution. In other words, all the above is just
  for free with the SDD Main Tracker.

Chicago LC Workshop, Chicago, Illinois, January
7-9, 2002 V. L.
Rykov, Wayne State University
18
RD for Large Tracker Application

• Improve position resolution to 5mm
• Decrease anode pitch from 250 to 100mm.
• Stiffen resistor chain and drift faster.
• Improve radiation length
• Reduce wafer thickness from 300mm to 150mm
• Move FEE to edges or change from hybrid to SVX
• Air cooling vs. water cooling
• Use 6in instead of 4in Silicon wafers to reduce
  channels.
• More extensive radiation damage studies.
• Detectors/FEE can withstand around 100 krad (g,n)
  
• PASA is BIPOLAR (intrinsically rad. hard.)
  
• SCA can be produced in rad. hard process.

19
WSU RD interests

• Main goal develop either full scale tracker or
  intermediate tracking layer on the basis of
  Silicon Drift technology.
• Projects
• Hardware
• 1.) design new prototype drift detector layout
  (incl. frontend stage) optimized for LC use
  (i.e. larger detector, higher pitch, higher
  voltage, less power consumption)
• 2.) collaborate with BNL on prototype production
  of wafer and frontend chip
• Software
• 1.) optimize 3d tracking code for solid state
  tracker, compare performance to gas detector and
  other silicon technologies
• 2.) write slow simulator for detector response
  and apply STAR tracking and pattern recognition
• 3.) find unique drift detector applications (e.g.
  track timing)

20
WSU proposal for the next 3 years (50 K per
project per year)

• In collaboration with the Instrumentation
  division at BNL
• 1.) design and produce a prototype batch (20)
  of new, optimized Silicon drift detectors. The
  proposed major changes compared to the old STAR
  design are
• a.) increase the detector size by using six inch
  rather than four inch wafers
• b.) increase the readout pitch in order to
  reduce the channel count
• c.)  thin the wafer from 300 micron to 150
  micron
• d.) operate wafers at higher voltage (up to 2500
  V) to accommodate new drift length
• 2.) design and produce a new prototype of a CMOS
  based frontend chip.
• a.)    use deep sub-micron technology to improve
  radiation hardness
• b.)    reduce power consumption to allow
  air-cooling of the detector
• c.)    potentially include the ADC stage into
  the PASA/SCA design
• d.)    test tape automated bonding rather than
  wire-bonding 

21
WSU proposal (cont.)

• 3.) we also propose to investigate a design for
  the mechanical support of the Silicon ladders
  based on a design used for the Silicon Strip
  detector layer in STAR.  
• 4.) software efforts
• a.)continue our comparative study of the
  performance of a Silicon drift detector based
  main tracker with the existing tracking and
  pattern recognition code...
• b.)provide a full GEANT based geometry
  definition of our proposed tracker before the
  fall of 2002.
• c.)port a detector response code from STAR into
  the LC simulation framework.
• d.)adapt a code recently written by a WSU led
  software group for STAR which allows track
  matching between the two main tracking detectors
  in STAR and the electro-magnetic calorimeter in
  STAR. An integrated tracking code (IT) can be
  applied to the SD design in order to
  simultaneously analyze the information from the
  vertex detector, the main tracker and the
  calorimeter.

22
Whats next for SDD ?

• The project has to grow, we need more groups
  interested in SDD (as of now only WSU and BNL
  expressed some interest).
• Prototype detectors for use in test setups at
  universities or other National Labs are available
  through WSU/BNL.
• People with mask design skills could work on new
  prototype layouts.
• The wafer and frontend electronics RD could be
  split in two projects.
• What should the frontend be DSM-CMOS, bipolar,
  different chips for different stages or single
  chip, implanted or wire-bonded ?
• Readout electronics and DAQ integration have not
  been addressed at all.
• Software development and simulations needs a lot
  more manpower. Talk to us if youre interested
  (bellwied_at_physics.wayne.edu)
• Check out the web at http//rhic15.physics.wayne.
  edu/bellwied/nlc

23
Simulation framework
B 5 T
B 3 T
144 layers
24
Simulation Studies

• Momentum resolution
• Present 20 mm pos.res., 1.5 rad.length/layer,
  Beampipe wall thickness 2 mm
• Future 5 mm pos.res., 0.5 rad.length/layer,
  Beampipe wall thickness 0.5 mm
• Two Track Resolution.
• Present 500 mm
• Future 200 mm

25
Track time stamping with the SDD (intrinsic)

• Correct timing
• Hit positions are determined correctly, and
  fit to a track with a good ?2.
• Wrong timing with some hit SDDs drifting in the
  opposite directions to the others (probability
  15/16th93.75)
• Hit positions are determined incorrectly, and
  do not fit to a track, i.e. ?2 is bad.
• Wrong timing with all hit SDDs drifting the same
  direction (probability1/16th6.25)
• Hit positions are determined incorrectly, but
  still fit to a shifted track with a good ?2.

Chicago LC Workshop, Chicago, Illinois, January
7-9, 2002 V. L.
Rykov, Wayne State University
26
Simulation Studies (cont.)

• Momentum resolution
• Modify Position Resolution
• Modify Radiation length Si thickness,
  Electronics
• Modify Beam Pipe Wall Thickness

27
Track time stamping with the TPC

• It is recognized that, if the time stamping for
  the tracks in the TPC or SDD is not done, it
  could seriously impact the detector performance,
  particularly its missing mass resolution.
• Sorting out tracks, using the Main Tracker only
  is always the most desirable option.
• The suggested solution for the TPC was to place,
  at some TPC depth, fast intermediate tracker,
  made from scintillating fibers and/or silicon
  intermediate tracking layer inside the TPC.
• (Physics Resource Book, 2001)

Chicago LC Workshop, Chicago, Illinois, January
7-9, 2002 V. L.
Rykov, Wayne State University
28
Various drift axis combinations in MT layers
Impact on PT-resolution ?anode 7 ?m ?drift
10 ?m

• In the options z?z?z (the best for time
  stamping), momentum resolution at high PT
  deteriorates by 10, compared to zzzzz (the best
  for PT resolution).
• There is virtually no worsening of momentum
  resolution for zzz?z drift.

Chicago LC Workshop, Chicago, Illinois, January
7-9, 2002 V. L.
Rykov, Wayne State University