Transcript Development of Silicon Detectors

WHI - Project Review 2002
- Halbleiterlabor (HLL) Projects at HLL
Overview (list of main projects)
CCD Development for ROSITA
SDD Application (art analysis)
DEPFET pixels for TESLA and XEUS
Requirements for vertex detectors
Technology development, design, simulation, prototyping
Comparison to other vertex detector concepts
Summary
R. H. Richter - WHI Project Review Dec, 17th 2002
Projects at HLL (1) – High Energy Physics
ATLAS SCT Strip detectors
Requirements: Single sided strips (p in n), radiation hardness up to 3x10 14 neq/cm²
Concept (p in n), design, prototyping: HLL
Production: Hamamatsu, CiS (Erfurt)
Status: Series production (finished by Dec 2002), Acceptance tests at HLL
ATLAS Pixel sensors
Requirements: Pixel size 50x450 µm²,
extremely radiation hard, 1x1015 neq/cm²!!
Concept, technology and design: HLL
Production: CiS (Erfurt), Tesla (Czech Republic)?
Status: Start of series production, HLL (support)
Linear Collider (TESLA): Thin and fast Active Pixel Sensors (see below)
CAST
Start of development at HLL: 2002
End: 2005
Aim: Search for solar axions
by use of an X-ray telescope equipped with PN-CCD (see talk by R. Kotthaus)
R. H. Richter - WHI Project Review Dec, 17th 2002
Projects at HLL (2) - Astrophysics
XMM
Launch of the satellite: 1999
Aim: Study of galactic and extragalactic X-ray sources
Detector requirements: energy res., position res., time res.
Instrument: PN-CCD
ROSITA
Start of the development: 2000
Scheduled Mission: 2007 - 2009
Aim: Orbit scan 0.5~keV to 15~keV
Instrument: Optimized PN-CCD with frame store region
XEUS
Start of the development: 1996
Scheduled start: 2015
Aim: See XMM. But increased sensitivity by a factor 200, improved angular
resolution, extended energy range
Instrument: Optimized PN-CCD with frame store region OR Active Pixel
Detector (DEPFET)
R. H. Richter - WHI Project Review Dec, 17th 2002
ROSITA
Modified Repetition of ABRIXAS
Orbit scan: 0.5~keV to 15~keV
To be installed on ISS
Advanced PN-CCDs (separated image and
storage areas)
Prototypes produced in new laboratory
R. H. Richter - WHI Project Review Dec, 17th 2002
CDD Development for ROSITA and XEUS
Fast transfer of signals into the frame store
Slow (low noise) read out from the frame store
Out of time event probability: 0.2% (factor 30
lower than at XMM)
Improvement of Charge Transfer Effciency (CTE)
Charge transfer loss is
by a factor of 13 smaller
than that of the XMM-CCD
No Titanium contamination
in the new CCDs by using
of HE-implantation instead
of an epitaxial layer.
Low energy resolution
Al-K line 76eV FWHM
C-K line 82eV FWHM
Achieved by an improved entrance window (use of <100> instead of <111>
crystal orientation) and by a better electronic noise 3.5 e- rms (XMM: 4.5 e-)
Compact X-ray fluorescence spectrometer
Manuscript: Faust I
by Johann Wolfgang v. Goethe
Investigated at
Bundesanstalt für Materialprüfung
From the composition of the ink they
concluded that parts of Faust I were
corrected at a time when Goethe already
worked on Faust II.
Röntec-Spectrometer equipped with
a Silicon Drift Chamber
R. H. Richter - WHI Project Review Dec, 17th 2002
Silicon detectors for LC vertex detector (TESLA)
Collaboration with Unversity of Bonn (N. Wermes)
WHI-HLL: Sensor
Bonn: Read out and steering chips
Synergy with XEUS Project (MPE)
R. H. Richter - WHI Project Review Dec, 17th 2002
TESLA Vertexdetector
Options:
CCD
MAPS
HAPS
DEPFET
Total > 500 MPixel (bei 25x25 µm Pixelgröße)
(read out speed in 50 MHz)
R. H. Richter - WHI Project Review Dec, 17th 2002
Layer
Module size
No. Of
modules
I
13 x 100 mm
1x8
II
22 x 125 mm
2x8
III
22 x 125 mm
2 x 12
IV
22 x 125 mm
2 x 16
V
22 x 125 mm
2 x 20
Detector requirements for LC
TESLA:
• high position resolution (vertex reconstruction,
momentum resolution)
• low radiation length of inner layers
• low power consumption
(500MPixel + cooling additional material not
allowed)
• high readout out speed for background
suppression
• radiation tolerant
• pixel size (20-30 µm)2
5(+)10/p sin3/2θ µm
• sensor thickness d=50µm
 0.1% X0 per layer
( layer I @ r=13mm )
• DEPFET: Pmean< 1W
operation @ 300 K
• 50MHz, read out speed
occupancy < 1%
• 100-200krad (5 years)
5 x 109 neq/cm2
R. H. Richter - WHI Project Review Dec, 17th 2002
Module concept with DEPFETs
Auslesechips
520 x 4000 pixel
DEPFET-Matrix
(25 x 25µm Pixel)
Steuerchips
Auslesechips
• Sensor area thinned down to 50 µm
• Remaining frame for mechanical stability
carrying readout and steering chips
R. H. Richter - WHI Project Review Dec, 17th 2002
DEPFET-Principle
Radiation
source
top gate
n+
p+
p-channel
internal gate
+
bulk
p+
n+
-n - +
-+
-+
-
drain
potential via axis
top-gate / rear contact
~1mm
~300 mm
totally depleted
n--substrate
potential minimum
for electrons
p+
rear contact
V
FET integrated on high ohmic n-bulk
Advantages:
of thethe
charge
at the
position of collection
Collection Amplification
of electrons within
internal
gate
=> no transfer loss
ModulationFull
of the
FET
current by the signal charge!
bulk
sensitivity
Non structured thin entrance window (backside)
Very low input capacitance => very low noise
R. H. Richter - WHI Project Review Dec, 17th 2002
Excellent noise values measured on single pixels
Ka
6000
5000
# Zähler
4000
3000
2000
Escape - Peak
Kb
1000
0
2
4
6
Energie [keV]
55Fe-spectra
@ 300K
ENC = 4.8 +/- 0.1 eR. H. Richter - WHI Project Review Dec, 17th 2002
BioScope - imaging of tracer-marked bio-medical samples
(P. Klein and W. Neeser)
Noise: ca. 70 ENC @ 300K
Slow operation (old technology)
Large arrays are impossible
(JFET => VP variations)
Large cell size
Rectangular DEPFET pixel detector
MOS transistor instead of JFET
A pixel size of ca. 20 x 20 µm² is
achievable using 3µm minimum
feature size.
R. H. Richter - WHI Project Review Dec, 17th 2002
DEPFET pixel matrix
Low power consumption
Fast random access to
specific array regions
- Read filled cells of a row
- Clear the internal gates
of the row completely
- Read empty cells
R. H. Richter - WHI Project Review Dec, 17th 2002
imaging spectroscopy
purpose
particle tracking
7.68 x 7.68 cm²
1024 x 1024 pixels
detector format
1.3 x 10 cm² (x 8)
520 x 4000 pixels (x 8)
1 Mpix
2.1 Mpix (x8)
75 µm
pixel size
25 µm
300 ... 500 µm
thickness
50 µm
4 el. ENC
noise
~ 100 el. ENC
1.2 msec
2.5 µsec
readout time
/ detector
/ row
50 µsec
20 nsec
DEPFET 6” -Technology
Double poly / double aluminum process
on high ohmic n- substrate
along p-channel
perpendicular to channel (with clear)
R. H. Richter - WHI Project Review Dec, 17th 2002
Pixel prototype production (6“ wafer)
for XEUS and LC (TESLA)
Aim: Select design options for an optimized array operation
(no charge loss, high gain, low noise, good clear operation)
On base of these results => production of full size sensors
Many test arrays
- Circular and linear DEPFETS
up to 128 x 128 pixels
minimum pixel size about 30 x 30 µm²
- variety of special test structures
Production will be finished in spring
R. H. Richter - WHI Project Review Dec, 17th 2002
Potential during collection - 3D Poisson equation (Poseidon)
(50µm thick Si, NB=1013cm-3,VBack=-20V)
Depth 10µm
Depth
Depth1µm
4µm
7µm
External (internal) Gates
Drain
n+ clear contacts
Sources
Cell size 36 x 27 µm²
R. H. Richter - WHI Project Review Dec, 17th 2002
Potential distribution during Reading
Back contact
Internal Gate
2D dynamic simulation
along the channel
ID adjusted to 100µA
(W/L =18µm/5µm)
Vinternal Gate ca. 3V
Localized charge generation
simulates a hit
Source
Drain
R. H. Richter - WHI Project Review Dec, 17th 2002
DEPFET simulation – TeSCA (2D, time dependent)
hit response to a generation of 1600 electron-hole pairs
Current production status
Pixel array section – Design with clockable clear gate
1 Pixel cell
Drain
Gate
Status:
- Poly I and II ok
- Implantations (N-Side) ok
- P-Side Processing started
8 lithographic steps ready
To do:
- P-Side
diode / entrance window
- Contact openings
- Metal 1 FS/BS => Measurements
- Metal 2
Clear
4 / 6 mask steps until März `03 / Juni `03
Clear
gate
Source
R. H. Richter - WHI Project Review Dec, 17th 2002
Crossing polysilicon lines
Problems with demolished polysilicon lines
and bad polyI/polyII insulation
Solved now
R. H. Richter - WHI Project Review Dec, 17th 2002
Processing thin detectors
- the Idea -
R. H. Richter - WHI Project Review Dec, 17th 2002
Detector thinning – first results
Thickness of detector region : 50µm
of frame : 350µm
Size: 8cm x 1cm
Wafer bonding – MPI f. Festkörperstrukturphysik, Halle
Wafer grinding – SICO GmbH, Jena
Anisotropic etching – CiS gGmbH Erfurt, MPI Halbleiterlabor Munich
R. H. Richter - WHI Project Review Dec, 17th 2002
1,5mm
Read out chip – test submission (Marcel Trimpl - Bonn)
- fast current read out -
4 mm
TSMC 0,25 µm process (ca. 60 000 transistors)
contains all important parts of the design
Measurements: Very encouring results with nearly TESLA requirements !
R. H. Richter - WHI Project Review Dec, 17th 2002
Performance estimation of
TESLA vertex detector candidates
Resolution
5(+)10/p
sin3/2θ µm
CCD
Similar as
used in SLD
HAPS
4.2(+)4.0/p
sin3/2θ µm
+
++
R&D
7µm (-)
Hybrid APS
MAPS
Monolith.APS
CMOS
Microelectr.
DEPFET
Material
budget
≤ 0.1% X0/l.
2µm (+++)
But at
50MHz ?
Like CCD
++
Read out
Speed
(50 MHz)
O?
R&D !!
--
++
+
O?
R&D
Power
consumpt.
R&D !!
+
+?
R&D
R&D
Radiation
tolerance
Ionisation, n
R&D
wellengineered
product –
enough
Potential?
++
Back up
solution
+?
Large area
device
Composition
of dice ?
+?
?
-+?
R&D
++
+?
R. H. Richter - WHI Project Review Dec, 17th 2002
Remarks
R&D
Be patient
for
6-12 months

9th EUROPEAN SYMPOSIUM ON SEMICONDUCTOR DETECTORS
New Developments on Radiation Detectors
took place at Schloss Elmau, June 23 - 27, 2002
R. H. Richter - WHI Project Review Dec, 17th 2002
Summary
o
o
Our part in ATLAS is almost done
Future projects: LC (TESLA), ROSITA, XEUS – encouraging CCD results
o
DEPFET is promising detector candidate for future HE and astrophysics
experiments. Key features: low noise, full bulk sensitivity, no charge transfer loss,
low power consumption, random access within an array
A new DEPFET 6 inch technology (2 poly/ 2 aluminum) was developed for large
arrays and high speed operation.
DEPFET Prototype production has been started and will be finished in spring ’03.
Read out electronic studies are very encouraging.
A concept for merging the DEPFET technology with a thinning technology is
proposed.
o
o
o
o
Plans for 2003:
Measurement and analysis of the prototype production
New Drift Chamber and CCD submissions
R. H. Richter - WHI Project Review Dec, 17th 2002