The CMS Precision Proton Spectrometer (PPS) Past, Present, and Future Jonathan Hollar (LIP) LIP Seminar, Jan. 21, 2021 1 PPS: The Precision Proton Spectrometer • A series of very small detectors (few cm2 active area) Roman Pot stations Set of movable detectors to approach the beam. • Located very far from the central CMS detector (~210m)RP for tracking at the stations CERN LHC RP for timing stations RF shield Each unit includes 3 RPs (1 horizontal and 2 New cylindrical design to host larger detectors vertical for alignment runs only). and reduce the impedance and increase available vacuum compatibility in terms of outgassing, and particle shower development have to be taken space. Detector Only one inserted into account in the geometrical designpackage and in the choice of materials (Section 3.4). for high lumi runs Top • • The timing RPs are equipped with a 300 μm thick Approachingwindow very towards close the beam.to Horizontal the LHC beam• The thickness (~2 mm) is required to compensate the pressure inside movablegradient "Roman on the larger window. • No vertical stations needed because the alignment is Bottom Pot” vesselsdone in byevery propagating LHC tracks from the tracking stations. BeamThe tracking RPs are equipped withfill a thin window 150 μm thick towards the beam. The CT-PPS RPs are inserted at ~15σ from the LHC beam in standard high luminosity fills. 26 2 Figure 42: Top: drawings of the cylindrical detector housing for the new RPs designed to accomodate timing detectors. Bottom: the manufactured pots. Figure 43: Dimensions of the cylindrical RP. After considering various options and after an iterative optimisation, the following design has been adopted for the new RPs (Figures 42 and 43)[6]. The volume housing the detectors will have a cylindrical 50 3 Physics Perspectives In most of the pp collisions at the LHC, the scattered protons dissociate after exchanging gluons or quarks. However, in the case of colourless exchanges, photon (γ) exchange for electromagnetic interactions or pomeron (IP) exchange for strong interactions, interacting protons could emerge intact. The intact protons, having lost a fraction ⇠ = ∆p/p of their longitudinal momentum, are deflected from the proton bunch by the LHC magnets and measured in the PPS detector with unprecedented resolution. In hard scattering events (involving production of high pT particles), particles produced at the interaction point can be measured and reconstructed by the CMS de- tector. Combining the information from PPS with that from the central CMS detector allows the Physics of PPS study of hard interactions in CEP processes. A schematic diagram for a CEP process is depicted in Fig. 1. p1 p10 • In a special class of LHC collisions, the protons γ, IP stay intact and scatter in the far forward direction X • γγ or multiple-gluon (“Pomeron”) exchanges γ, IP p2 p20 • Detecting the protons provides a new powerful toolFigure to study 1: Schematic γγ collisions diagram for at centralhundreds exclusive of production, pp pXp. ! GeV to TeV scales For Runs 2 and 3 of the LHC, the range of invariant mass of the system “X” extends from ap- • Operation at high luminosity => sensitivityproximately to a wide 350 GeV range to 2 TeV, of rare when and both protonsnew processes can be detected (“double-arm” measurement), and above 50 GeV when only one proton is detected by PPS (“single-arm” measurement). In the former case, full event reconstruction is made possible by matching the proton kinematics to the kinematics of “X” measured by the central detector. As the exclusive standard model processes Electroweak and Standard Model under investigation have small cross sections at high masses, mX, (more details in Section 3.2), indirect newmost measurements physics will be limitedDirect by statistical new uncertainties physics with the full Run 2+3 data set. Physics In addition, searches for new phenomena will benefit from the higher integrated luminosities of the HL-LHC. In channels where the kinematicsearches matching is not sufficient to adequately suppress Anomalousbackgrounds couplings from in uncorrelated pileup protons, precision timing detectors can be used to match γγ→ll the longitudinal vertex position of the system “X” with the protons, as described in Section 8.1.4. γγ→gauge bosons (WW,ZZ,γγ), Resonances, missing Dijet production As discussed in the following chapters, the combination of all four locations under study for HL- γγ→top quarks, mass searches… … LHC detectors (at 196, 220, 234, and 420 m on each side of IP5, see Appendix B.2 for a configuration γγ→tauoverview) leptons… would cover an extended mass range from approximately 50 GeV to 2.7 TeV when both protons can be detected. In addition to the increased integrated3 luminosity, this will allow an expanded set of physics topics to be studied, in comparison to Runs 2 and 3. The increased upper mass range would increase the acceptance for both direct and indirect searches for beyond standard model (BSM) physics. The reduced lower mass limit, associated with a station at 420 m from IP5, would significantly enhance the acceptance for all SM processes and, in particular, Higgs production, as well as for the production of feebly coupled BSM resonances (such as e.g. light axion-like particles). A few examples are illustrated in this section, using generator-level Monte Carlo simulations with the PPS acceptance calculated for the HL-LHC beam and optics parameters (Version 1.3 [9]) averaged over the fill (i.e. the middle of the “luminosity levelling trajectory”; see Section 5.2 for the running scenario). All studies shown in this section have been performed for the vertical beam crossing at IP5, in which the beams cross in the y z plane rather than the x z plane, as now officially decided [8]. − − 9 Reconstructing protons in PPS (in a nutshell) Proton reconstruction needs: Proton kinematics: - RP alignment: mostly (x,y) shifts. X - Optics calibration: transport matrix Overview describing the beam optics constrained Proton reconstruction infrom space… data. Optical functions such as dispersion Dx("), etc. PPS (“PrecisionAcceptance for two tagged protons Proton from MX ~ 400 Spectrometer”) GeV – 2 TeV, depending on machine optics. - Alignment & calibrations computed partially The groupfrom• Protonslow-luminosity is in chargeare bentfill at the through of start providing of datathe LHC calibrated forward taking, with otherwise the same conditions Near •Far accelerator magnetic fields, and arrive at the • Near-beam tracking andp timing detectors, as in high-luminosity operation. proton objectsdetectors to CMS analyzers RP IP5 ► p* RP - Global alignment calculatedRP for each fill; 210 housed in moveable “Roman Pot” installations 220 tracks movements of detectors,timing or the beam. far in the LHC beam-line ◄ RP near IP Hit mapReconstructing (x,y): - Optics• kinematics computedDetectors220 for measure a of set protonsof principal x,y position which stay intact after interaction: far crossing angles, and interpolated for RP • ~210-220 m from the CMS interaction⇠ point intermediate values. 210 LHC magnet lattice : fractional energy loss near • Then “reverse engineer” to find the fractional (optics) t - Two reconstruction flavours: single-RP • Detectors must be moved to within: 4-momentum ~2mm from a single transfermomentum PPS detector loss station(ξ) and with 4-momentum transfer of the LHC beam at top energies - extreme worse resolutionsquared and (t) globalof the multi-RP proton at the collision point ✓: proton scatteringreconstruction, angle with lower at efficiency, IP notably constraints on control/safety systems when SiStrip detectors are used (2017). CMS Week – February 2020 • Requires very detailed understanding!5 of • Selects intact protons from “exclusive”Proton variablesLHC magnetic constrain fields kinematics (“optics”), and (M, Y) of centrally alignment of the Roman Pots interactions• For signal pp→ events,pXp kinematicsproduced are closed system X, measured by CMS • Closed kinematics• ξ values for are correctly related toreconstructed the invariant events mass and rapidity of the central system “X” • Depending• onMx final acceptance state “X”: roughly 300-2000 GeV 4 • Reconstruct the full 13 TeV collisionYoung energy POG: less than a year in operation, CT-PPS detector • • Reconstruct everything and searchstarted for leftover data taking “missing at mass” the beginning (similar idea of as the recoil Run 2 analysis in e+e-) 1 I Over 100 fb− of data collected 2 I Analyses started/ongoing within SMP, TOP, EXO – more to come! 2 Low pileup fill & CT-PPS timing detectors … and in time ��Pile-up����� + background������� and �� rejection:����� Timing • Main interest for CT-PPS: collect sample for first � � ��� = ��� � � • attempts at calibrating forward proton timing ��� ��� A major background is due to responding to timing resolutions of 10 ps (top) and 30 ps (bottom) are shown for signal and background “pileup”detectors protons, (diamonds coming + from ultra-fast silicon) events. other collisions in the same γγ→WW, both protons from signal Inclusive WW background 1 1 ���� simulation simulation -t [ns] � ⋅ Δt -t [ns] 0.8 B 0.8 � ⋅ ΔtF Δ CMS-TOTEM Δ CMS-TOTEM LHC bunch crossing 0.6ToF 0.6ToF Vertex • Principle - if both protons stay intact, Δt(protons) PPS0.4 ToF Backward 0.4 PPS ToF Forward 0.2 0.2 measured by CT-PPS is correlated with vertex z • 0 0 Pileupposition as large measured as 50-60 inin central tracker/pixels -0.2 -0.2 � � LHC Run 2 -0.4 � Δ�� + � Δ�� =-0.4��� + ��� -0.6 -0.6 � � � -0.8 ���� = � Δ�� − -0.8��� = ��� − � Δ�� = Δ�� − Δ�� -1 -1 2 -25 -20 -15 -10 -5 0 5 10 15 20 25 -25 -20 -15 -10 -5 0 5 10 15 20 25 • Needs detailed understanding of HPTDC Leading lepton vertex z position [cm] Leading lepton vertex z position [cm] • Mitigated by precisely D. Di Croce, 2015 measuringsettings, the arrival channel time timing of offsets, thresholds, etc. γγ→WW, both protons from signal Inclusive WW background 1 1 simulation simulation 0.8-t [ns] 0.8-t [ns] HEP2016Δ - Valparaiso - A.