Journal of Instrumentation

PAPER You may also like

- RADIOACTIVE POSITRON EMITTER
Characterisation of planar sensors for the inner PRODUCTION BY ENERGETIC ALPHA
PARTICLES IN SOLAR FLARES
tracker of the CMS experiment R. J. Murphy, B. Kozlovsky and G. H.
Share

- Dynamic high energy density plasma

To cite this article: M. Hajheidari and on behalf of the CMS Tracker group 2022 JINST 17 C09002 environments at the National Ignition
Facility for nuclear science research
Ch J Cerjan, L Bernstein, L Berzak
Hopkins et al.

- ASCOT simulations of 14 MeV neutron

View the article online for updates and enhancements. rates in W7-X: effect of magnetic
configuration
J Kontula, J P Koschinsky, S Äkäslompolo
et al.

This content was downloaded from IP address 111.68.99.155 on 09/11/2023 at 01:19

 Published by IOP Publishing for Sissa Medialab
Received: November 25, 2021
Accepted: July 13, 2022
Published: September 6, 2022

12th International Conference on Position Sensitive Detectors

12–17 September, 2021
Birmingham, U.K.

Characterisation of planar sensors for the inner tracker

2022 JINST 17 C09002

of the CMS experiment

M. Hajheidari on behalf of the CMS Tracker group

University of Hamburg, Institute for Experimental Physics,
Luruper Chaussee 149, 22761 Hamburg, Germany

E-mail: mohammadtaghi.hajheidari@desy.de

Abstract: The Compact Muon Solenoid (CMS) experiment is expected to collect an inte-
grated luminosity of 3000 fb−1 during the High Luminosity phase of the Large Hadron Collider
(HL-LHC). This scenario comes with a high number of collisions per bunch crossing, and in turn,
a high level of radiation for the innermost layer of the CMS tracker. Simulations estimate a 1 MeV
neutron equivalent fluence, Φeq, of 2.3 × 1016 cm−2 at a distance of 2.8 cm from the collision
point. The inner tracker of the CMS detector is required to withstand this range of fluence and
maintain tracking performance. Planar pixel sensors with an active thickness of 150 μm and pixel
sizes of 25 × 100 μm2 or 50 × 50 μm2 have been produced by Hamamatsu Photonics (HPK) and
Fondazione Bruno Kessler (FBK). The sensors were bump bonded to the RD53A readout chip
prototype. The sensor-chip modules were irradiated with 23 MeV protons to the 1 MeV neutron
equivalent fluence of up to 2.4 × 1016 cm−2 at the Zyklotron AG (ZAG).
Non-irradiated and irradiated modules were tested in the DESY II beam test facility. The spatial
resolution as a function of the incidence angle and hit efficiency as a function of the bias voltage
of the sensors were determined from these measurements. It is shown that for the highest fluence,
the planar modules still reach 98% hit efficiency at bias voltages below 800 V.

Keywords: Particle tracking detectors (Solid-state detectors); Pixelated detectors and associated
VLSI electronics; Radiation-hard detectors

© 2022 IOP Publishing Ltd and Sissa Medialab https://doi.org/10.1088/1748-0221/17/09/C09002

Contents

1 Introduction 1

2 Sensor design and experimental setup 2

3 Test beam characterisation 3

2022 JINST 17 C09002

4 Spatial resolution of the non-irradiated sensor 3

5 Hit efficiency of irradiated sensors 4

6 Summary 4

1 Introduction

In the next long shut-down, the Large Hadron Collider (LHC) will be upgraded to the High Lumi-
nosity LHC. In this scenario, the instantaneous luminosity is projected to reach 5 × 1034 cm−2 s−1,
or 7.5 × 1034 cm−2 s−1 in the ultimate luminosity scenario. This translates to an average of
140 (200) proton-proton interactions per bunch crossing. The integrated luminosity of the CMS
experiment is expected to reach 3000(4000) fb−1 after 10 years [1].
The Inner Tracker (IT) of the CMS detector will be upgraded to work efficiently in the new
radiation environment [2]. The pixel size will be reduced by a factor of 6 with respect to the current
detector. Moreover, the thickness of the sensors used in IT will be reduced from 280 μm to 150 μm
in order to decrease the drift distance of the charge carriers and minimise trapping effects.
There are two sets of requirements for the sensors used in the Inner Tracker: after installation,
√
a hit efficiency above 99% and a point resolution smaller than the binary resolution, i.e. pitch/ 12,
is required, while at the end of the lifetime, a hit efficiency above 98% for the sensors used in
the innermost layer (L1) and above 99% for layers 2 to 4 is required. These efficiencies need to
be achieved at a bias voltage below 800 V. The spatial resolution at the end of the lifetime is
expected to be at the level of the binary resolution. For the CMS IT, hybrid silicon modules are
being investigated for the HL-LHC. Planar and 3D sensors with pixel sizes of 25 × 100 μm2 and
50 × 50 μm2 are candidates for the IT. Results characterising 3D sensors were published in [3]
and are not discussed further in this paper. The results of the test-beam characterisation of planar
sensors before and after irradiation are reported in this paper.

–1–
2 Sensor design and experimental setup

In this work, planar p-type sensors produced by Hamamatsu (HPK) and Fondazione Bruno Kessler
(FBK) are investigated. The sensors have an active thickness of 150 μm with a pixel size of
25 × 100 μm2 . The schematics of HPK sensors for two design variants are shown in figure 1.

2022 JINST 17 C09002

Figure 1. Sensor design variants: (a) standard and (b) bricked designs with bitten implant. In these sketches,
purple circles show the opening for passivation, orange squares are the contact, and, green area is the n+
implant, and the red lines are the p-stop layer.

The sensors are bump bonded to the RD53A readout chip [4]. The chip has a tunable threshold
for hit detection, measuring Time over Threshold (ToT) with 4 bit resolution at 40 MHz. For the
CMS Inner Tracker, the linear front end has been chosen. Sensors were irradiated at ZAG with
23 MeV protons up to a 1 MeV neutron equivalent fluence of 2.4 × 1016 cm−2 . To calculate the
equivalent fluence, a hardness factor of κ = 2 is used.
The beam test characterisation was performed at the DESY II facility [5] using an electron
beam with an energy of 5.2 GeV. For track reconstruction, an EUDET-type telescope with six
planes was used. Each plane of the telescope contains a MIMOSA 26 sensor with 1152 columns
and 576 rows and a pixel size of 18.4 × 18.4 μm2 [6]. The readout of the telescope is binary and the
estimated intrinsic resolution of each plane is 3.2 ± 0.9 μm. The integration time of the MIMOSA
sensors is relatively long (115 μs). In order to select the tracks in time with the readout cycle of
the Device Under Test (DUT), a CMS phase-1 pixel module is used as a timing layer [7]. Figure 2
shows a sketch of the test beam setup.

Figure 2. Sketch of the setup for the test beam measurements.

–2–
3 Test beam characterisation

Two important parameters, which are determined from the test beam measurements are the hit
efficiency and the spatial resolution. The hit efficiency of the sensor is defined as:
Nht
= (3.1)
Nt
N t is the total number of reconstructed in-time tracks within the DUT fiducial region and N ht is
the number of those tracks with a hit in the DUT. The spatial resolution is determined by obtaining
the residual distribution. In order to obtain this distribution, one needs to calculate ΔxDUT as:

2022 JINST 17 C09002

ΔxDUT = xDUT − xtele (3.2)

xtele is the projected hit position using the telescope and xDUT is the reconstructed hit position
by DUT. The resolution of the sensor is related to the width of the residual distribution and is
calculated: 
σDUT = σm 2 − σtele 2 (3.3)
σ m is the width of the measured residual distribution (i.e. ΔxDUT ) and σ tele is the estimated
telescope resolution at the DUT surface. The reconstructed position in the DUT, i.e. xDUT, is
obtained by the Center of Gravity (CoG) algorithm [8] using the ToT values. Therefore, one
expects a better resolution as the charge sharing between pixels increases. This is obtained either
by the Lorentz drift due to magnetic field in the experiment, or by rotating the sensor in the test
beam setup.

4 Spatial resolution of the non-irradiated sensor

The spatial resolution of two non-irradiated sensors with standard and bricked designs (shown in
figure 1) have been measured at various angles of beam incidence. The turn angle is a rotation
around the long pitch (y-axis) so that the charge sharing is along the x-axis (see figure 1). The bias
voltage of the sensors was set at 120 V. Figure 3 shows the spatial resolution in x- and y-directions
as a function of turn angle. The telescope resolution, σ tele in equation (3.3),was estimated directly
from the residual distribution obtained from the difference between upstream and downstream
triplets and subtracted from measured value (σ m in equation (3.3)).

Figure 3. Spatial resolution in x (a) and y (b) as a function of rotation angle around the y-axis.

–3–
 One can see that both designs show similar values for the resolution in the x-axis where the
best value of 2 μm is reached at the optimum angle of 9◦ (≈atan ( thickness
pitch
)). Along the long pixel
side, however, the two sensors have different resolutions. For the standard design, σ y does not
√
change as a function of turn angle and stays at the level of the binary resolution (100 μm/ 12).
For the bricked design, σ y reaches 12 μm which is slightly better than the binary resolution for a
pixel size of 50 μm. This can be understood by looking at the pixel cell of the bricked module
(figure 1(b)) where the effective pitch size is 50 μm instead of 100 μm.

5 Hit efficiency of irradiated sensors

2022 JINST 17 C09002

The hit efficiency of several irradiated modules was measured as a function of bias voltage. The
sensors were placed inside a cold box and cooled down using a circulation chiller and Peltier
elements to minimise the leakage current. The temperature of the chiller liquid was set to −35 ◦ C.
For track reconstruction, only the first arm of the telescope (upstream triplet in figure 2) was used.
The threshold of the RD53A chip was set in the range of 1100 – 1300 electrons. Pixels with an
occupancy more than 1% of the number of triggers were masked as noisy during the offline analysis
and not taken into the account in calculation of hit efficiency. The measurements were performed
at normal incidence.
Figure 4 shows the hit efficiency of irradiated modules as a function of bias voltage. Two
modules irradiated at the highest fluences, i.e. FBK at 2.4 × 1016 cm−2 and HPK at 2.0 × 1016 cm−2
reach 98% efficiency at a bias voltage below 800 V and meet the requirement for the layer 1. The
other three modules irradiated at lower fluences reach 99% efficiency and are qualified for layer 2.
The leakage current of the sensors was monitored during the measurement. It was observed that
for all modules the leakage current stays below the limit of 760 μA (10 nA/pixel) [4].

Figure 4. Hit efficiency of irradiated planar sensors as a function of bias voltage.

6 Summary

Non-irradiated and irradiated planar pixel sensors for the CMS Inner Tracker were measured in
the DESY II beam test facility. The hit efficiency and spatial resolution of these sensors was
determined from these measurements. For both non-irradiated sensors, a minimum resolution of

–4–
2 μm was achieved along the short pitch axis at the optimum angle. For the bricked design, the
resolution along the long pitch axis reached a minimum value of 12 μm. The hit efficiency of
irradiated planar modules was measured as a function of bias voltage at normal incidence. The
results show planar pixel sensors meet the requirements defined for layers 1 and 2 of the Inner
Tracker.

References

[1] CMS collaboration, The CMS experiment at the CERN LHC, 2008 JINST 3 S08004.

2022 JINST 17 C09002

[2] CMS collaboration, The Phase-2 Upgrade of the CMS Tracker, CERN-LHCC 014 (2017).
[3] A.G. Alonso et al., Test beam characterisation of irradiated 3D pixel sensors, 2020 JINST 15 C03017.
[4] M.G. Sciveres, The RD53A Integrated Circuit, CERN-RD53-PUB-17-001 (2017).
[5] R. Diener et al., The DESY II test beam facility, Nucl. Instrum. Meth. A 922 (2016) 265.
[6] H. Jansen et al., Performance of the EUDET-type beam telescopes, EPJ Tech. Instrum. 3 (2016) 7.
[7] W. Adam et al., The CMS phase-1 pixel detector upgrade, 2021 JINST 16 P02027.
[8] R. Turchetta, Spatial resolution of silicon microstrip detectors, Nucl. Instrum. Meth. A 335 (1993) 44.

–5–