Transcript twepp07_pozzati

Università di Pavia
Dipartimento di Elettronica,
I-27100 Pavia, Italy
1
MAPS in 130 nm triple well CMOS technology
for HEP applications
INFN
Sezione di Pavia
I-27100 Pavia, Italy
2
Enrico Pozzati1,2, Massimo Manghisoni2,3, Lodovico Ratti1,2,
Università di Bergamo
Dipartimento di Ingegneria
Industriale, I-24044
Dalmine (BG), Italy
3
Valerio Re2,3, Valeria Speziali1,2, Gianluca Traversi2,3
Introduction
In this work deep N-well CMOS monolithic active pixel sensors (DNW-MAPS) are presented as an alternative approach to signal processing in pixellated detectors
for high energy physics experiments. Based on different resolution constraints, some prototype MAPS, suitable for applications requiring different pitch, have been
developed and fabricated in 130 nm triple well CMOS technologies. This work presents experimental results from the characterization of some test structures
together with TCAD and Monte Carlo simulations intended to study the device properties in terms of charge diffusion and charge sharing among pixels.
Deep N-well pixel sensor concept
Apsel2 laser source tests
A deep junction N-well is used as the
collecting electrode.
The impact of front-end electronics on the
pixel fill-factor can be limited by placing
the NMOS devices belonging to the
processor analog section inside the deep
N-well structure.
Efficiency loss due to the presence of Ntype diffusions housing P-type transistors
might not be significant whether the deep
N-well is comparatively larger.
Prototype chips
A low power InGaAs/GaAlAs/GaAs laser source
has
been
employed
for
experimental
characterization of the Apsel2T chip 3×3
matrix (λ = 1060 nm).
The chip has been back-illuminated in order to
avoid reflection from the die surface.
Electrical pulses with an energy close to 200 fJ
are required to emulate a MIP at the die
surface (substrate thickness = 254 µm).
(*)
Matrix scan tests fearure 961 laser position
points with a 5 µm step along both X and Y
directions.
Apsel family chips
Charge collected by the 3×3 matrix as
a function of the laser spot position.
Readout chain includes charge preamplifier,
shaping stage, threshold discriminator and latch.
Charge sensitivity [mV/fC]
Peaking time can be programmed to assume
one of the following values: 0.5 µs, 1 µs, 2 µs.
PIXEL
The pitch is about 50 µm.
Central pixel
equipped
with a 60 fF
injection
capacitor
SDR0 chip
Readout chain includes charge
threshold discriminator and latch.
preamplifier,
The pitch is about 25 µm and is suitable for
applications at the International Linear Collider
experiments.
Signal processing at pixel level include sparsified
data readout.
Charge collected by the central pixel in the 3×3
matrix as a function of the laser spot position.
1_1
614
1_2
610
1_3
540
2_1
573
2_2
576
2_3
500
3_1
565
3_2
600
3_3
534
Apsel2-geometry physical simulation
SDR0-geometry physical simulation
Monte Carlo simulation assumptions
80 electrons for each micron are generated
uniformly along a linear track normal to the
device surface. Gaussian distribution in the
plane orthogonal to the track with σ=0.5 µm.
Simulated volume is 230×230×80 µm3.
Electron lifetime has been taken into account
(9.2 µs @ NA = 1015 cm-3 ).
(**)
TCAD simulated device. 36 collision points have
been considered with a 5 µm step for both X and
Y axes.
a)
b)
Charge collected by the central pixel in the 3×3
SDR0 matrix according to TCAD simulations (a)
and Monte Carlo simulations (b).
Monte Carlo simulation results. Charge collected
by the 3×3 matrix as a function of the incident
particle position.
Monte
Carlo
TCAD
Charge collected by
the central pixel
1011
1060
Charge collected by
the 3×3 matrix
1619
Matrix loss (%) with
respect to the max
value (**)
48
Laser
Monte
Carlo
Charge collected by
the central pixel
1398
1512
Charge collected by
the 3×3 matrix
2283
2343
31
31
Matrix loss (%) with
respect to the max
value (*)
Comparison between experimental and simulated
results for the Apsel2 MAPS.
(*)
Monte Carlo simulation results. Charge
collected by the Apsel 3×3 matrix is
displayed as a function of the position of
the MIP collision point.
Conclusions
1566
A laser source has been employed to characterize the Apsel geometry in terms of charge
diffusion and charge sharing among pixels.
49
A Monte Carlo code have been developed to simulate random walk of minority carriers in
an undepleted detector substrate.
Comparison between Monte Carlo and TCAD
simulation results for the SDR0 MAPS.
Experimental results on the 3×3 matrix of the Apsel2T chip can be reproduced in Monte
Carlo simulations. A good agreement was found between TCAD and Monte Carlo results in
the SDR0 geometry simulation.
Further investigation on the MAPS properties will be performed in the next months with
the experimental characterization of the SDR0 prototype chip.