=Paper=
{{Paper
|id=Vol-2507/448-452-paper-83
|storemode=property
|title=Performance of the CMS Precision Proton Spectrometer During LHC Run2 and Its Upgrades for Run 3
|pdfUrl=https://ceur-ws.org/Vol-2507/448-452-paper-83.pdf
|volume=Vol-2507
|authors=Fabrizio Ferro
}}
==Performance of the CMS Precision Proton Spectrometer During LHC Run2 and Its Upgrades for Run 3==
https://ceur-ws.org/Vol-2507/448-452-paper-83.pdf
Proceedings of the 27th International Symposium Nuclear Electronics and Computing (NEC’2019)
Budva, Becici, Montenegro, September 30 – October 4, 2019
PERFORMANCE OF THE CMS PRECISION PROTON
SPECTROMETER DURING LHC RUN2 AND ITS UPGRADES
FOR RUN3
F. Ferro1,a On behalf on the CMS and TOTEM Collaborations
1
I.N.F.N. - Genova, Via Dodecaneso 33, 16146 Genova, Italy
E-mail: a fabrizio.ferro@ge.infn.it
The CMS Precision Proton Spectrometer (PPS) consists of silicon tracking stations as well as timing
detectors to measure both the position and direction of protons and their time-of-flight with high
precision. Special devices called Roman Pots are used to insert the detectors inside the LHC beam pipe
to allow the detection of scattered protons close to the beam itself. They are located at around 200 m
from the interaction point in the very forward region on both sides of the CMS experiment. The tracking
system consists of 3D pixel silicon detectors while the timing system is made of diamond pixel detectors
and Ultra Fast Silicon Detectors. PPS has taken data at high luminosity while fully integrated into the
CMS experiment. The total data collected correspond to around 100 fb-1 during the LHC Run2. In this
presentation, the PPS detector operation, commissioning and performance are discussed, as well as the
upgrades foreseen for Run3.
Keywords: LHC, CMS, TOTEM, spectrometer, proton, tracking, timing
Fabrizio Ferro
Copyright © 2019 for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
448
Proceedings of the 27th International Symposium Nuclear Electronics and Computing (NEC’2019)
Budva, Becici, Montenegro, September 30 – October 4, 2019
1. Introduction
The Precision Proton Spectrometer (PPS) detector system has been installed and integrated into
the CMS experiment during the LHC Run 2 data taking period. It is a joint project by the CMS and
TOTEM collaborations with the capability of measuring protons scattered at very small angles, operating
at high instantaneous luminosity [1]. The scattered protons remain inside the beam pipe, displaced from
the central beam orbit, and can be measured by detectors placed inside movable beam pipe insertions,
called Roman Pots (RP), that approach the beam down to a few mm. The idea is to measure the
displacement of the scattered proton with respect to the beam by means of tracking detectors and their
time-of-flight by means of timing detectors. The displacement measurement is converted into that of the
proton momentum loss via the precise knowledge of the accelerator optics. The measurement of the
TOF provides an estimation of the longitudinal position of the proton-proton interaction and can
therefore be used to dramatically reduce the background induced by the pile-up. The PPS detectors have
collected data corresponding to an integrated luminosity larger than 100 fb−1 during the LHC Run 2
(2016–2018). The PPS detector allows the study of central exclusive production (CEP), i.e. the process
pp→p(∗)+X+p(∗), by detecting at least one of the outgoing protons. In the CEP process, that may occur
through hadronic interaction or by photon-photon fusion, one or both protons may dissociate into a low-
mass state (p∗). When both protons dissociate no signal is detected in PPS. The central system X may
consist of W or Z boson pairs, photon or lepton pairs, high-pT jet production, and in general states with
JPC=0++ (also 2++ in case of fusion).
2. The PPS experimental layout
Figure 1 shows the layout of the RP stations installed at around 210 m from the CMS interaction
point (IP5), along the beam line in LHC sector 56. A symmetric configuration is installed in LHC sector
45. The stations are comprised of RP’s that approach the beam vertically from the top and bottom, and
horizontally. During standard machine operation, scattered protons feature a large displacement in the
horizontal direction and a small vertical displacement at the RP positions. The horizontal RP’s are hence
used. The vertical RP’s are used in special configurations of the machine and in low intensity proton-
proton fills for the calibration and alignment of the detectors. Each detector arm consists of two RP’s
instrumented with silicon tracking detectors that measure the transverse displacement of protons with
respect to the beam, and one RP station with timing detectors to measure their time-of-flight. The
measurement of the proton TOF aims to reject background combinations with a proton from a pileup
interaction, or a beam-halo particle. Three RP stations per arm were equipped during LHC Run2 with
detectors in the horizontal pot. The configuration changed each year, reaching in 2018 the final
Figure 1. Schematic layout of the beam line between the interaction point and the location of the RP stations in the LHC
(one side only).
(nominal) layout. The start of the PPS program, initially foreseen for 2017, was advanced to 2016; that
year, the existing TOTEM Si strip detectors were used for tracking, while no timing detectors were
installed. The Si strips, originally designed for low luminosity runs, were not able to sustain the high
radiation dose taken during nominal runs, therefore only ~15 fb−1 of data could be written on disk. They
were however sufficient to prove that PPS can actually operate and take data valid for physics analysis.
449
Proceedings of the 27th International Symposium Nuclear Electronics and Computing (NEC’2019)
Budva, Becici, Montenegro, September 30 – October 4, 2019
In 2017 an intermediate detector configuration was installed. One of the stations used for tracking and
equipped with Si strip detectors was kept, while the other one was equipped with 3D pixel sensors [2].
Moreover, a special station between the previous two was equipped with timing detectors: diamonds [3]
and Ultra Fast Silicon Detectors (UFSD) [4]. Finally, in 2018, both tracking stations were equipped with
3D silicon pixels and the timing station with diamond detectors. The details of the detectors used by PPS
in the Run 2 are described in the following sections, focusing on the layout and performances of 3D
pixels, for the tracking, and of the diamonds, for the timing [5, 6, 7].
3. The PPS tracking detector
Silicon strip sensors with a reduced insensitive region at the edge facing the beam have initially
been used. A RP can host 10 silicon strip sensor planes, half at a +45◦ angle and half at a −45◦ angle with
respect to the bottom of the RP. These sensors cannot sustain a large radiation dose and cannot identify
multiple tracks in the same event. For this reason they have been gradually replaced by new 3D silicon
pixel sensors, with one RP station during the 2017 data taking run, and both RP tracking stations in
2018. Each RP hosts 6 3D pixel sensor planes that are read by 4 or 6 PSI46dig read out chips (ROC) [8],
depending on the sensor size. The data collected by the ROC’s are collected and serialized by a token
bit manager (TBM) and sent to the front end (FE) electronics via an interface card (RPix portcard) that
transforms the digital electrical signal into optical. The FE cards (FED for data and FEC for the controls)
are the same as the ones used for the Phase-1 CMS Pixel Tracker [9]. The pixel dimensions are 150x100
µm2 and provide a spatial resolution x~17 µm, mainly due to the charge sharing between adjacent
pixels. Because of the peculiarity of the shape of the signal, concentrated close to the beam region, see
Figure 2 (left), the detector irradiation is dramatically non-uniform, as well as the correlated damage in
the ROC’s that are bump bonded directly on the sensor. As a consequence, the signal generated in the
few most damaged pixels drifts in time more quickly than in the rest of the plane, causing a delay with
the rest of the pixels that may become comparable to the bunch crossing time, jeopardizing the
association of the hits to the event. In order to cope with this issue, when it was clear that the accumulated
dose was creating the problem mentioned above, the pixel RP’s have been slightly raised in position,
thus moving the signal in a less damaged region of the detector. Apart from this problem, the detector
performance was very high, with a ~98% average efficiency and more than 99% efficiency in most of
the detector (see Figure 2 (right)).
Figure 2. Left: impact point of the reconstructed tracks in a RP equipped with 3D pixels. Right: efficiency of a single pixel
plane at the beginning of the data taking period in 2017.
450
Proceedings of the 27th International Symposium Nuclear Electronics and Computing (NEC’2019)
Budva, Becici, Montenegro, September 30 – October 4, 2019
4. The PPS timing detector
The timing detectors consisted of single-crystal CVD diamond sensor planes, with one plane of
ultra-fast silicon sensors during the 2017 data taking, while in 2018 only diamond sensors were used. In
the latter configuration, each timing RP was equipped with two planes of single diamond and two planes
of double diamond detectors; in the latter, two diamond sensors were bonded to the same read out
channel in order to double the amplitude of the signal. The signal is amplified and shaped a first time
inside the RP itself, then it is sent to a NINO [1] chip that works as an amplifier and discriminator and
finally to an HPTDC [1] that converts the time-over-threshold into a digital measurement. An accurate
calibration of the detectors was needed to provide a definite estimation of the timing measurement
uncertainty and a test beam campaign was carried out after the data taking to assess the damages due to
the radiation. The test beam campaign confirmed that the timing resolution provided by the double
diamond sensor itself is of the order of ~50 ps and that the radiation hardness of the detector is actually
very good, with more than 94% efficiency after a year of data taking at the LHC. The evolution of the
time resolution coming from the analysis of the LHC data is shown in Figure 3: a general degradation
of the resolution by 20-50% is clearly visible and is due to the radiation damage of the sensor and of the
readout electronics close to the beam.
Figure 3. Evolution during 2018 data taking of the single channel timing resolution (left) and of the average time resolution
from the track reconstruction in each of the timing RP (right).
5. Physics results
Central exclusive production of lepton pairs has been observed for the first time at the LHC in
proton-proton collisions at s=13 TeV, with data collected in 2016 corresponding to an integrated
luminosity of 9.4 fb-1 [10]. The process considered is pp→p+l+l-+p(∗), where one proton may dissociate
into a low-mass state and l=e,µ. Events are selected with m(l+l-) > 110 GeV/c2. A scattered proton is
measured in one or both RP stations in either detector arm, in LHC sectors 45 or 56. The main
background sources are Drell-Yan and double-dissociation events, combined with a proton from a pileup
interaction or a beam-halo particle. They are suppressed by requiring that no extra tracks be present at
the l+l- vertex, and the two leptons be back-to-back in the transverse plane. Additionally, the
reconstructed proton momentum loss (RP) is required to agree within 2 with the value calculated from
the two leptons. Figure 4 shows the correlation of (l+l-) versus (RP), for protons measured in the
detectors in sector 45 (left arm) and 56 (right arm). The regions outside acceptance for both near and far
RPs in a detector arm, and only for the near RP, are shown. Events with matching and non-matching
values of (l+l-) and (RP), and those for which (l+l-) falls below the (RP) acceptance, are indicated
separately. A total of 12 matching events are observed in the µ+µ- channel, and 8 events in the e+e-
451
Proceedings of the 27th International Symposium Nuclear Electronics and Computing (NEC’2019)
Budva, Becici, Montenegro, September 30 – October 4, 2019
channel, with a combined significance over the background estimate of 5.1. No events are observed
with matching protons simultaneously in both arms, consistent with expectations of exclusive production
of lepton pairs within the detector acceptance for these data.
Figure 4. Correlations between the proton fractional momentum loss reconstructed indirectly from the two leptons (l+l-) and
from the scattered protons (RP) in sector 45 (left) and 56 (right) of the LHC.
6. Plans and prospects for Run 3
While Run 2 data are still being analyzed, the PPS group is preparing the new detectors for Run
3, that is scheduled to start in 2021. A new RP per side will be installed to add an additional set of timing
detectors. Hence, the configuration for Run 3 will consist of two tracking and two timing RP’s per side,
equipped with 3D silicon pixels and double diamond detectors, respectively. The new tracking detectors
will be bonded to PROC600 readout chips [11] that will replace the PSI46dig’s. Moreover, the detector
packages will be equipped with a remotely controlled motor that will be able to change their position
inside the RP and therefore spatially distribute the radiation damage and reduce its impact. New double
diamond timing detectors are being built and extensive optimization of the electronics is being carried
out, aiming at a significant improvement of the timing resolution. The goal of PPS is to take data during
the whole Run 3: a rich physics program lies ahead.
References
[1] CMS and TOTEM Collaborations, CERN-LHCC-2014-021.
[2] S.I. Parker et al., Nucl. Instr. and Meth. A 395 (1997) 328.
[3] G. Antchevet al. [TOTEM Collaboration], JINST12(2017) no.03, P03007
[4] V. Sola et al., JINST12(2017) no.02, C0207
[5] CMS Collaboration, CERN-CMS-DP-2019-036
[6] CMS Collaboration, CERN-CMS-DP-2019-034
[7] CMS Collaboration, CERN-CMS-DP-2018-056
[8] H. Chr. Kastli, Nucl. Instr. and Meth. A 731 (2013) 88.
[9] CMS Collaboration, CERN-LHCC-2012-016; CMS-TDR-11, 7 September 2012 (2012).
[10] A. M. Sirunyanet al. [CMS and TOTEM Collaborations], JHEP1807(2018) 153
[11] A. Starodumov et al., JINST 12 (2018) C01078
452