Prepared for submission to JINST
Construction and commissioning of CMS CE prototype
silicon modules
B. Acar, G. Adamov, C. Adloff, S. Afanasiev , N. Akchurin, B. Akgün, M. Alhusseini, J. Alison,
G. Altopp, M. Alyari, S. An, S. Anagul, I. Andreev, M. Andrews, P. Aspell, I.A. Atakisi,
O. Bach, A. Baden, G. Bakas, A. Bakshi, P. Bargassa, D. Barney, E. Becheva, P. Behera,
A. Belloni, T. Bergauer, M. Besancon, S. Bhattacharya, S. Bhattacharya, D. Bhowmik,
P. Bloch, A. Bodek, M. Bonanomi, A. Bonnemaison, S. Bonomally, J. Borg, F. Bouyjou,
D. Braga, J. Brashear, E. Brondolin, P. Bryant, J. Bueghly, B. Bilki, B. Burkle, A. Butler-Nalin,
S. Callier, D. Calvet, X. Cao, B. Caraway, S. Caregari, L. Ceard, Y. C. Cekmecelioglu,
G. Cerminara, N. Charitonidis, R. Chatterjee, Y. M. Chen, Z. Chen, K. y. Cheng,
S. Chernichenko, H. Cheung, C. H. Chien, S. Choudhury, D. Čoko, G. Collura, F. Couderc,
I. Dumanoglu, D. Dannheim, P. Dauncey, A. David, G. Davies, E. Day, P. DeBarbaro, F. De
Guio, C. de La Taille, M. De Silva, P. Debbins, E. Delagnes, J. M. Deltoro, G. Derylo,
P.G. Dias de Almeida, D. Diaz, P. Dinaucourt, J. Dittmann, M. Dragicevic, S. Dugad, V. Dutta,
S. Dutta, J. Eckdahl, T. K. Edberg, M. El Berni, S. C. Eno, Yu. Ershov, P. Everaerts, S. Extier,
F. Fahim, C. Fallon, B. A. Fontana Santos Alves, E. Frahm, G. Franzoni, J. Freeman,
T. French, E. Gurpinar Guler, Y. Guler, M. Gagnan, P. Gandhi, S. Ganjour, A. Garcia-Bellido,
Z. Gecse, Y. Geerebaert, H. Gerwig, O. Gevin, W. Gilbert, A. Gilbert, K. Gill, C. Gingu,
S. Gninenko, A. Golunov, I. Golutvin, T. Gonzalez, N. Gorbounov , L. Gouskos, Y. Gu,
F. Guilloux, E. Gülmez, M. Hammer, A. Harilal, K. Hatakeyama, A. Heering, V. Hegde,
U. Heintz, V. Hinger, N. Hinton, J. Hirschauer, J. Hoff, W. S. Hou, C. Isik, J. Incandela, S. Jain,
H. R. Jheng, U. Joshi, O. Kara, V. Kachanov, A. Kalinin, R. Kameshwar, A. Kaminskiy,
A. Karneyeu, O. Kaya, M. Kaya, A. Khukhunaishvili, S. Kim, K. Koetz, T. Kolberg, A. Kristić,
M. Krohn, K. Krüger, N. Kulagin, S. Kulis, S. Kunori, C. M. Kuo, V. Kuryatkov, S. Kyre,
O. K. Köseyan, Y. Lai, K. Lamichhane, G. Landsberg, J. Langford, M. Y. Lee, A. Levin, A. Li,
B. Li, J.-H. Li, H. Liao, D. Lincoln, L. Linssen, R. Lipton, Y. Liu, A. Lobanov, R. S. Lu,
I. Lysova, A. M. Magnan, F. Magniette, A. A. Maier, A. Malakhov, I. Mandjavize, M. Mannelli,
J. Mans, A. Marchioro, A. Martelli, P. Masterson, B. Meng, T. Mengke, A. Mestvirishvili,
I. Mirza, S. Moccia, I. Morrissey, T. Mudholkar, J. Musić, I. Musienko, S. Nabili, A. Nagar,
A. Nikitenko, D. Noonan, M. Noy, K. Nurdan, C. Ochando, B. Odegard, N. Odell, Y. Onel,
W. Ortez, J. Ozegović, L. Pacheco Rodriguez, E. Paganis, D. Pagenkopf, V. Palladino,
S. Pandey, F. Pantaleo, C. Papageorgakis, I. Papakrivopoulos, J. Parshook, N. Pastika,
M. Paulini, P. Paulitsch, T. Peltola, R. Pereira Gomes, H. Perkins, P. Petiot, F. Pitters,
F. Pitters, H. Prosper, M. Prvan, I. Puljak, T. Quast, R. Quinn, M. Quinnan, K. Rapacz, L. Raux,
G. Reichenbach, M. Reinecke, M. Revering, A. Rodriguez, T. Romanteau, A. Rose, M. Rovere,
A. Roy, P. Rubinov, R. Rusack, A. E. Simsek, U. Sozbilir, O. M. Sahin, A. Sanchez,
arXiv:2012.06336v1 [physics.ins-det] 10 Dec 2020
R. Saradhy, T. Sarkar, M. A. Sarkisla, J. B. Sauvan, I. Schmidt, M. Schmitt, E. Scott, C. Seez,
F. Sefkow, S. Sharma, I. Shein, A. Shenai, R. Shukla, E. Sicking, P. Sieberer, Y. Sirois,
V. Smirnov, E. Spencer, A. Steen, J. Strait, T. Strebler, N. Strobbe, J. W. Su, E. Sukhov,
L. Sun, M. Sun, C. Syal, B. Tali, U. G. Tok, A. Kayis Topaksu, C. L. Tan, I. Tastan, T. Tatli,
R. Thaus, S. Tekten, D. Thienpont, T. Pierre-Emile, E. Tiras, M. Titov, D. Tlisov, J. Troska,
Z. Tsamalaidze, G. Tsipolitis, A. Tsirou, N. Tyurin, S. Undleeb, D. Urbanski, V. Ustinov,
A. Uzunian, M. van de Klundert, J. Varela, M. Velasco, M. Vicente Barreto Pinto, P. M. da
Silva, T. Virdee, R. Vizinho de Oliveira, J. Voelker, E. Voirin, Z. Wang, X. Wang, F. Wang,
M. Wayne, S. N. Webb, M. Weinberg, A. Whitbeck, D. White, R. Wickwire, J. S. Wilson,
H. Y. Wu, L. Wu, C. H Yeh, R. Yohay, G. B. Yu, S. S. Yu, D. Yu, F. Yumiceva, A. Zacharopoulou,
N. Zamiatin, A. Zarubin, S. Zenz, H. Zhang, J. Zhang
E-mail: arnaud.steen@cern.ch, bora.akgun@cern.ch
Abstract: As part of its HL-LHC upgrade program, the CMS Collaboration is developing a High
Granularity Calorimeter (CE) to replace the existing endcap calorimeters. The CE is a sampling
calorimeter with unprecedented transverse and longitudinal readout for both electromagnetic (CE-
E) and hadronic (CE-H) compartments. The calorimeter will be built with ∼30,000 hexagonal
silicon modules. Prototype modules have been constructed with 6-inch hexagonal silicon sensors
with cell areas of 1.1 cm
2
, and the SKIROC2-CMS readout ASIC. Beam tests of different sampling
configurations were conducted with the prototype modules at DESY and CERN in 2017 and 2018.
This paper describes the construction and commissioning of the CE calorimeter prototype, the
silicon modules used in the construction, their basic performance, and the methods used for their
calibration.
Contents
1 Introduction 1
2 Module components, construction and testing 3
2.1 Module construction 3
2.2 Silicon sensors 6
2.3 Front-end electronics 7
2.4 Module testing 8
2.4.1 IV tests 8
2.4.2 Tests on the front-end electronics 9
3 Final beam test setup 11
3.1 CERN-SPS H2 beam line instrumentation and trigger system 12
3.2 Silicon electromagnetic calorimeter prototype 12
3.3 Silicon hadronic calorimeter prototype 14
4 Beam test performance 16
4.1 Event building and analysis procedures 16
4.2 Pedestals and noise: calculation and stability 17
4.3 Signal reconstruction 19
5 Channel-to-channel response equalization and gain linearization 22
5.1 Channel-to-channel response equalization 22
5.2 Gain linearization 27
5.2.1 Gain linearization using beam-test data 28
5.2.2 Gain linearization using charge injection 29
6 Conclusion 33
1 Introduction
The CERN High-Luminosity LHC (HL-LHC) will operate with a higher instantaneous luminosity
than the CERN LHC and is expected to record ten times more data. The increase in the instantaneous
luminosity is a challenge for detector design, due to the needs for increased radiation tolerance and
for the mitigation of effects due to overlapping events (pile-up), which is expected to be as high
as 200 collisions per bunch crossing. To cope with these conditions, the CMS Collaboration
has undertaken an extensive R&D program to upgrade many parts of the detector, including the
replacement of the calorimeter endcaps [1]. There are two key requirements that the new endcap
calorimeters must meet. Firstly, they should maintain acceptable performance after the delivery
– 1 –
of the expected HL-LHC integrated luminosity (3000 fb
–1
), when the total neutron fluence in the
innermost region will be 10
16
n
eq
/cm
2
and the total ionizing dose will be 2 MGy. Secondly, the
detector needs to have ∼50 ps timing resolution to mitigate the pile-up.
The CE [2], shown schematically in Figure 1, is a high granularity sampling calorimeter with
50 active layers and more than 6 million channels. Silicon modules with a hexagonal sensor,
an absorber plate, and readout electronics will be the building blocks of the calorimeter. In the
electromagnetic section (CE-E) of the calorimeter, silicon modules will be interleaved with lead
and copper absorbers. The silicon sensors will be segmented into hexagonal cells with an area
of approximately 1.1 cm
2
. In the innermost region the segmentation will instead result in cells
with an area of 0.5 cm
2
, where the fluence will be highest. The hadronic section (CE-H) will also
use silicon sensors in the region where the radiation is high, and plastic scintillator tiles readout
by on-tile silicon photomultipliers (SiPM) where it is low. The main absorber of the hadronic
calorimeter will be steel. The full calorimeter will be inside a cold volume kept at -30
◦
C to reduce
the dark currents in the silicon sensors and the SiPMs. This highly-segmented calorimeter will
provide transverse, longitudinal and precision timing information on showers that will be essential
for pile-up mitigation, event reconstruction, and analysis.
~2.3 m
Electromagnetic calorimeter (CE-E): Si, Cu/CuW/Pb absorbers, 28 layers, 25.5 X
0
& ~1.7 𝛌
Hadronic calorimeter (CE-H): Si & scintillator, steel absorbers, 22 layers, ~9.5 𝛌
CE-E
CE-H
Silicon
Scintillator
~2 m
Figure 1: Schematic view and key parameters of the CMS High Granularity Calorimeter Endcap.
Several tests of calorimeters built with prototype silicon modules have taken place in beams
at CERN, Fermilab and DESY. The goals for these tests were to validate the basic design of the
CE, to study the calorimetric performance of a silicon-based calorimeter, and to compare the
Geant4 simulation [3] of the calorimeter with experimental data. The first prototypes of hexagonal
– 2 –
silicon modules were tested in beams at CERN and Fermilab in 2016, with up to 16 silicon
modules, equipped with the SKIROC2 ASIC [4]. Despite the limited number of silicon modules,
encouraging results were achieved in terms of energy resolution, and there was good agreement
with a Geant4 simulation of the detector [5]. In addition, during a separate beam test in 2016, the
timing performance of sets of non-irradiated and irradiated silicon diodes were evaluated. Their
measured timing resolution was about tens of picoseconds [5].
In October 2018, a two-week beam test, at the H2 beam line of the CERN Super Proton
Synchrotron (SPS), was conducted with a calorimeter built with 94 prototype silicon modules, that
were equipped with a new version of the readout ASIC, the SKIROC2-CMS [6]. The response
of the calorimeter was measured with beams of charged hadrons, electrons and muons that had
momenta from 20 to 300 GeV/c.
This paper describes the construction and commissioning of CMS CE prototype silicon mod-
ules and their assembly into the prototype calorimeter. Section 2 describes the silicon module
components, and their assembly and testing. Section 3 describes the setup used during the 2018
beam test. Section 4 shows the performance of the prototype modules, and in section 5 the calibra-
tion procedures used, including channel-to-channel response equalization, and the gain linearization
are presented.
2 Module components, construction and testing
The building block of the CE is the silicon module. It consists of a baseplate for mechanical support,
a silicon sensor, and a printed circuit board (PCB) with embedded electronics. The construction
procedure for prototype modules used in the 2016 beam tests is documented in [5]. In 2018, the
semi-automated module assemlby process was demonstrated with the construction of 94 modules
equipped with 6-inch silicon sensor, reaching the targeted production rate of 6 modules per day.
The limiting factor was the time for glue curing at room temperature. The assembly and testing
procedures, established at the University of California, Santa Barabara (UCSB) pilot centre, were
tested and improved during the prototype module production.
The CE detector will have approximately 30,000 silicon modules built with 8-inch silicon
sensors. For their assembly, it is planned that there will be up to six module assembly centres
(MACs). The planned production rate is 24 modules per day at each MAC during the construction
phase. The assembly and testing procedures developed for the 6-inch modules are now being applied
to the 8-inch modules, which is the baseline design of the CE detector [2].
2.1 Module construction
The construction of a silicon module is shown schematicaly in Figure 2. They consist of: a baseplate
in copper or copper-tungsten, a 100 µm thick gold-plated Kapton™ sheet, a hexagonal silicon
sensor, and a printed circuit board, called ‘hexaboard’, holding four readout ASICs. The baseplate
provides mechanical support and thermal conductivity between the module and the cooling layer,
as described in Section 3. The baseplate flatness is within 100 µm, and the thickness tolerance is
less than 30 µm. Two different baseplate materials are used: copper for CE-H and copper-tungsten
(25% Cu and 75% W) for CE-E prototypes. In the CE-E, the denser baseplate is used to increase
compactness of the calorimeter. The gold plated Kapton™ sheet, epoxied onto the baseplate, serves
– 3 –
