D0 Silicon Tracker Replacement for Run 2b SMTII 20042007

1
D0 Silicon Tracker Replacement for Run 2bSMT-II
2004-2007

• Tianchi Zhao
• May 1, 2001
• Tevatron Collider Run 2b
• D0 Silicon Tracker for Run 2a
• Radiation Damage and Radiation hard Materials
• MRI Proposal and Responsibilities
• D0 Silicon Tracker for Run 2b

2
Tevatron Operation
Run 2a (2001-2003) 2 fb-1
36x36 (396 ns bunch spacing) until reaches 1032
cm-2sec-1 switch to
140x100 (132 nsec bunch spacing) Peak
luminosity 2x1032 cm-2sec-1 Run
2b (2004-2006) 13 fb-1
150 ?Rad crossing angle Increase antiproton
intensity by 2-3 Peak luminosity 5x1032
cm-2sec-1 TeV33 (2007 and beyond) 10
fb-1 per year Peak luminosity 1033
cm-2sec-1
3
(No Transcript)
4
D0 Run 2A Silicon Tracker
6 Barrels Double-sided Single sided
12 F-Disks double-sided
H-Disks double-sided
H-Disks double-sided
5
D0 Run 2a Silicon Barrel Side-View
Layer 1-4
Inner radius 27 mm Outer radius 94 mm
Barrel length 762 mm 4 layres - 8 overlapping
sublayers
6
D0 Run 2a Barrel Silicon Tracker
7
D0 Run 2a Barrel Silicon Tracker Summary
Layer 1 Outer 2 barrels 24 single sided
sensors Axial Inner 4 barrels 48
double sided sensors Axial 2o Stereo Layer
3 Outer 2 barrels 48 single sided sensors
Axial Inner 4 barrels 96 double
sided sensors Axial 2o Stereo Layer 2 All
6 barrels 72 double sided sensors Axial
90o Stereo Layer 4 All 6 barrels 144
double sided sensor Axial 90o Stereo Total
number of sensors 432 Total number of
channels 387,120 Inner 4 barrels 8
measurements -gt 4 axial,
two 2o Stereo, two 90o Stereo Outer 2 barrels
6 measurements -gt 4
axial, two 90o Stereo
8
Operating voltage limited by breakdown of
integrated capacitors
9
D0 Silicon Tracker Replacement for Run 2b
D0 Silicon Track was conceived in early 1990s
(VFD ltlt 100 V) It was designed for 3 fb-1
Decision made to replace the entire silicon
tracker after Run 2a Two inner layers
expected to die some time during Run 2b
Partial replacement is too difficult to do
Time required for a partial replacement it much
longer than the 6 month shut down
between Run 2a and Run 2b SMT-II project has
started at D0
10
Symptoms of Radiation Damaged
Detectors P-type doping level increases as
radiation level increases Bulk type inversion (n
? p) at 0.4 - 0.6 Mrad Depletion voltage
increases after inversion (10s V ? 100s
V) Leakage current linearly with radiation
-gt increased shot noise -gt increased heat
load Reduced inter-strip isolation and charge
collection -gt signal reduction -gt deep
level traps -gt under depletion Micro
discharges Localized radiation damage
noisy strips
11
Radiation Damage Machanizm Surface damage to
peak at 100 kRad and effects insignificant
SiO2 layer Positive ion accumulation (VFD
increase by 10-20V) SiO2-Si interfaces
new energy levels in band gap Dominate
effect Bulk damage due to energetic particles
displacing lattice silicon atoms Detailed
physics processes for bulk damages are not well
understood Doping changes due to nuclear
reactions examples 30Si n ? 31Si ? ?
31P e- n-type 28Si(n,p) ?
28Al (ET 4 MeV) p-type
28Si(n,d) ? 27Al (ET 4 MeV)
p-type 28Si(n,?) ? 25Mg (ET 4 MeV)
p-type Insignificant compared to dislocation
defects
12
(No Transcript)
13

Depletion voltage as a function of radiation
fluence
Neff (2 ?si VFD /qw2)
Doping type inversion n --gt P
14
Radiation Hard Silicon Material and Process

• Oxygen diffused FZ silicon Most promising
• Typical level of oxygen doping in FZ silicon
  1015/cm3
• Oxygenation can go up to 1018/cm3
• Effective doping level still 1012/cm3
• Other materials
• Carbon diffused FZ silicon (diffusion too
  slow)
• Tin diffused FZ silicon
• Low resistivity FZ silicon
• Epitaxy silicon film
• Best radiation resistance may be expected for n
  readout strip on
• p-type or n-type bulk
• Depletion region starts from the n strip
  side
• Works better if partially depleted

15
Calculated oxygen diffusion profile at 1150 oC
Si diffusion coefficiency _at_ 2.25 x 10-10 cm2s-1
16
Comparison Standard and Oxygenated Silicon
Initial effective n-type doping 1-2 x 1012/cm3
(Note Oxygen doping level up to
1018/cm3) Doping type inversion at particle
fluence 2.5 x 1013 cm-2 (0.6 Mrad) Rate of
p-type doping increase much slower for oxygenated
bulk
17
More on Oxygenated Silicon O2
oxgenation technique was pioneered by a BNL group
in 1992 No effects was found after irradiated
with 1 MeV neutrons In 1999, RD-48 at CERN found
x2 improvement for protons Further
investigations made by the BNL group
(IEEE NS Vol. 47, No. 6, p1892, Dec, 2000) High
O2 concentration Oxygenation at High T at 1200
oC for 9 days Thermal donors added ?
925 ?cm Sample 921 Cool
down from 550 oC to 350 oC in 3 hours No
termal doner ? 2300 ?cm
Sample 903 Cool down from 550 oC to 350
oC in 5 minutes Low O2 concentration 1100 oC
for 6 hours Sample 923 No
thermal donor effects (low O2 level)
18

• BNL Study Results
• Initial VFD similar for 923 and 903
• Initial VFD of 921 is 2 times higher
• VFD increases more slowly for 903 after
  inversion High O2 level helps
• 921 appears more resistant to radiation than
  Sample 903. But VFD higher

923 Low O2 level 903 High O2 level
without TD 921 High O2 level with TD
19
Cooling and Radiation Damage Low
temperature operation reduces diode leakage
current ATLAS silicon _at_ -7 oC results a x10
leakage current reduction Detector must be kept
at low temperature all the time even
during shutdown and detector maintains To avoid
reverse annealing At room temperature,
Neff rises after radiation No reverse
annealing if below 5-10 oC
20
Radiation Level for Silicon Microstrip Detectors
at Tevatron
For best impact resolution, silicon trackers must
start as close as possible to the beam
have minimum amount dead material before the
first measurement Current D0 barrel layer-1
silicon starts at r 27 mm CDF in Run 2a has
added a layer-00 at rmin 13.5 mm D0 is
planning to do the same in Run 2b At r 13.5
mm 6x1013 charged-particles/cm2/fb-1 ? 1.5
MRad/fb-1 The inner most layer can get 15-20
Mrad in Run 2b
21
New D0 Run 2b Silicon Tracker SMT-II
Design Constraints
Beam pipe parameters Radius of central
section 11 mm (?)
(CDF beryllium beam pipe _at_ 0.25M) Expand to
radius starting at z ? 100 mm
(?) Flange radius
20 - 22.5 mm (?) Total length of
beam pipe 1520 mm
(?) SMT-II Parameters 13 mm lt R lt 160 mm
Beam pipe r 11 mm
CFT ID 180 mm
Minimum active length 1.2 m (? lt 2)
Maximum installable length as one piece
1320 mm
22
Problems of D0 Run 2a Silicon Tracker
Serious delays, cost overruns caused because of
Complicated design Too many different
sensor types and hybrids Double sided
axial, 2o 90o stereo, different width
Single sided different width F, H disks
(?15o) Vendor technical difficulties and
delays double-sided sensors
Readout chips and hybrids Low mass
cables More than 1 year after the original
schedule and after the start of 36x36 collision,
only 23 are cabled up and only a few pieces are
readout
23
Layout of D0 Run 2a Silicon Tracker
24
Silicon Tracker End View(Concept)
New SMT-II ID 25 mm OD 320 mm SMT-I
Barrel ID 54 mm OD 200 mm
Outer constrained to fit inside fiber tracker L0
sensors mounted on new 1 OD beam pipe
25
Expected SMT II Performance
26
Sensor Choice for SMV-II
Single sided sensors only
Commodity products that can be purchased
Radiation hard Much less problems than
double-sided Two single sided bonded together
for 2D readout Oxgenated n-type FZ silicon
bulk p strips on n-type bulk Positive high
voltage on back plane (n side) Back plane AC
coupled to ground for signals Break down gt 1000
V for inner layers
27
L2-5 SensorHybrid Modules

• Two 12 cm long, 5 chip wide sensors with 55 mm
  pitch
• Flex circuit hybrid laminated on substrate wire
  bonds directly connect sensor to SVX4 readout
  chips
• Do not bond hybrids on silicon as done for SMT-1
• Stereo sensors mounted on opposite side of
  silicon support

28
Stave Support Structures

• Modules are mounted on Stave support structures
  that run the full length of the tracker
• Cooling, low-mass cables run along length of
  stave
• Two different module designs are being
  investigated

29
Mechanical and Cooling Issues
Stave support (no beryllium bulkheads) Strong
outer carbon fiber composite shell Thin inner
carbon fiber composit shell Cooling techniques
(baseline water plus glycol) Inner layer
requries -10 to -20 oC Engineering team at
Fermilab SiDet actively working
30
Front End Chip SVX4
SVXIIe used in D0 SMT-I SVXIII used in CDF Run
2a Silicon Tracker SVXIIe and SVXIII use
1.2 ?m process that is now obsolete Develop SVX4
based on CDFs SVXIII design LBL/Fermilab
Use 0.25 ?m process that is extremely
radiation resistant DC voltage reduced from
2.5 V to 1.25 V Chip will not get much
smaller because analog part dominates
31
(No Transcript)
32
(No Transcript)
33
Total silicon 7 m2
Total number of channels 947k (SMT-I 793 k)
34
Tracker Parameters

• Plan to gang two sensors per readout channel for
  L2-5
• Maximum of 948 readout channels (no ganging of
  L0-1)
• Fits within current DAQ (960 readout channels
  available)

35
MRI Proposal to NSF Development of a Silicon
Vertex Detector for the Higgs Search at the
Tevatron Collider Fresno State UIC (C. Gerber
PI) KU (A.Bean Project director) KState (R.
Demina PI) Michigan State Stony Brook Brown (R.
Partridge PI) Washington (G. Watts, H.J.Lubatti
PI) 1.96(NSF)0.4(foreign) 0.42(match)
2.78M Covers silicon, chips, electronics
and testing for L0-4 and mechanics for L0-1
If successful funding starts 8/01
36
Institutional Responsibilities and founding
Brown IC testing and module
assembly 327 k Fresno State
Hybrid procurement and testing 179
k Kansas U Hybrid testing
404 k UIC Module
assembly 200 k Kansas
State Sensor procurement and testing
320 k Stony Brook Sensor procurement and
testing 303 k MSU
Assembly fixture procurement 60 k U
Washington Inner layer design and testing
165 k
37
(No Transcript)
38
PAC (4/20/01) has advised D0 and CDF to adapt a
common design for Run 2b silicon tracker
Run 2A CDF Layer-00, 0, 1 Layer-00 is installed
on beam pine
L00
CDF Inner Layers L00 Rmin 13.5 mm L0 Rmin
25.4 mm L1 Rmin 41.2 mm
L0
L1
39
CDF Layer-00 Construction
40
Sensor width 8.4 mm and 14.6 mm, Active
length 78.4 Implant strip pitch 25 ?m
Implant strip widths of 8
?m Readout pitch of 50 ?m Two sensors are wire
bonded together to form pairs Six pairs total
along length 960 mm Hybrids mounted at the
ends of the array Fine-pitch kapton cables carry
signals from each of the six pairs of sensors
CDF layer-00 Laddle
41
Assembly CDF layer-00
42
Fine Picth Low Mass Cables for CDF layer-00
43
Hybrids of CDF Layer-00 at the end
44
Completed CDF layer-00 Silicon Tracker
45
Insert CDF Layer-00 in to SVXII
46
PAC (4/20/01) has advised D0 and CDF to adapt a
common design for their Run 2b silicon tracker
Current D0 Plan for inner layers
Accept CDF inner layer design for Run 2b
Improved oxygenated single sided sensors High
resolution stereo view for inner layers
was not in D0 NSF proposal High resolution
stereo (90o) for L0 (was not in D0 NSF
proposal) L0 uses 1 and 2 chip wide sensors
for axial measurements (only 2 chip wide
sensors in D0 NSF proposal) L1 uses 2
chip wide sensors CDF wants to
separate inner layers (L0-1-3 r lt 6cm)
from 3 outer layers D0 current
plan is to build L0-1 on beam pipe

47
Hybrid and Cable for Inner
Layers Move hybrids to two ends of the barrel
for inner L0 and L1 Reduce sensor damage
during construction Ease ladder construction
Ease cooling requirement Bring signals from
microstrips to hybrids by fine pitch low mass
cables Copper and gold traces on kapton
50 ?m pitch and fanned out to 100 ?m pitch to
reduce capacitance CDF has received new sample
cables from Keycom (Japan) They plan to visit
Keycom later this year
48
CDF Run 2b
D0 Run 2b Silicon Tracker Design
Concepts -Barrel Construction-
Long barrel, no disks Simplify construction
Interaction region length reduced to ?
12 cm Do not use short barrel construction
SMV-I has 6 short barrels Each barrel has two
sensor wire bonded together Hybrids glued on
top of the silicon sensors Use long ladder
construction as done by CMS Long staves (1200
mm for layer-5) Sensors, suport
frames, integrated cooling channels, hybrids
49
A Possible Fall Back Scenario What
if Run 2b is on schedule but SMV-II is
late? Replace existing beam pipe with a thinner
one as planned Build L0 quickly Install layer-0
to substitute dead layer-1,2 Do the full
replacement at a later stage
50
Current D0 Plan for outer layers
Reject CDF outer layer approach
CDF D0 rmax
20 cm rmax 16 cm 9
layers (3 at 40ltrlt60) 6 layers
planned No hybrid in tracking volume
Outer layers still has
hybrids in
tracking volume 100 ?m readout pitch for L3-5
55 ?m readout pitch for L2-5
My opinion D0 should accept CDFs 100 ?m
readout pitch for the outer layers and also move
hybrids to end end of the barrel using pine pitch
cables. Some mechanical aspects will have to be
different 320 mm OD (D0) vs 400 mm OD
(CDF) 1320 mm installable length limit for
D0
51
Construction Schedule
Detector construction cannot start before SVX4
chips are available SVX4 chips available
Feb. 2003 ? Hybrid construction and testing
after Feb. 2003 Ladder (stave) construction
after hybrids are available Burn-in and
laser probing Partial detector test (10
test) Final detector assembly and test
Current estimated completion
schedule Construction completed
Sept. 24 2004 Installation completed
Dec.24 2004
52
Status
D0 SMT-II is still in early planning stage. Many
important parameters and approaches have not been
determined. D0 has an engineering team at
Fermilab start working on mechanical issues D0
SMT-II project management is taking shape
CDF is has the layout and basic approaches more
or less determined CDF is moving ahead building
prototypes SVX4 is based on CDFs current SVX3
chip design CDF is pushing to start
construction at beginning of next year
53
Pixel Option for Layer 0 50 ?m x
400 ?m pixel detectors as used in CMS and
ATLAS Very attractive because capability of
pixels Cost, schedule and man power issues are
difficult at this time Will be a mandate for
TeV33 (2007 and beyond)
54
Layer 0 Radiation Specs for 15 fb-1 _at_CDF L00
Rmin 13.5 cm 8x1014 charged-particles/cm2 ?
20 Mrad _at_ Rmin 16.5 cm 5x1014
charged-particles/cm2 ? 13 Mrad VFD 500 V
at end of Run 2b
General Specifications for D0 L0-L1 Axial Sensors
o 8 cm long o 128 or 256 strips o 25 mm strip
pitch, 50 mm readout pitch o 1.28 cm active
width o number of devices 25620 spares
310 o tentative schedule by Apr 2002
55
Specifications for D0 L0-L1 type sensors
(continued)
Ø Junction breakdown gt
700V Ø Micro-discharge breakdown gt 700V
Ø Total detector current lt 100
nA/cm2 (at VFD 10V) Ø Total detector current at
700V lt 4 mA Ø Interstrip resistance (DC)
gt 2G? Ø Total interstrip capacitance
lt 1.2 pF/cm Ø Not working strips
lt 1
56
Specifications for D0 L0-L1
Sensors Ø Biasing scheme polyresistors on
both ends Ø Poly resistor values
4.5 ? 0.5 M? Ø Passivation
SiO2 0.5 ?m thick Ø Implant
strip width 8 ?m Ø Metal
strips Al, AC
coupled over the p-implant Ø Al strip width
7 10 ?m Ø Al strip
thickness gt 1 ?m
Ø Al strip resistivity
lt 20 ?cm Ø Coupling capacitance
gt 10 pF/cm Ø Coupling capacitor breakdown
gt 100V (for L0 type only, safety
against occasional beam losses)