A Novel Calorimeter for HL-LHC and Beyond CMS Endcap High Granularity Calorimeter for Phase II"

A Novel Calorimeter for HL-LHC and Beyond CMS Endcap High Granularity Calorimeter for Phase II" Work also supported by ERC Grant 670406 - Novel Calorimetry! T. S. Virdee,! Imperial College London! Oxford Jan16! 1!

Discovering the Higgs boson" The Nobel Prize in Physics 2013 was awarded jointly to" François Englert and Peter W. Higgs " "for the theoretical discovery of a mechanism that contributes to our understanding of the origin of mass of subatomic particles, and which recently was confirmed through the discovery of the predicted fundamental particle,! by the ATLAS and CMS experiments at CERN's Large Hadron Collider"! 1/19/16! Physics Days 15 Helsinki tsv! 2!

SM Higgs Boson as a Physics Benchmark For Detector Design Natural Width 0.01 1 10 100 GeV Transparency! from the 90 s! tsv Apr 2015! Theory does not predict m H" The favourable decay modes change with mass" Zumino Memorial! 3!

1990 Aachen: Search for H 2γ" Aachen! 1990! Oxford Jan16! 4!

Instrument Quality and Capability Matters Discovery of the Higgs Boson" Physics Driving the Design! Measure the energies of photons from! a decay of the Higgs boson! to a precision of 0.5%.! Idea (1993 few yellowish cm 3 samples)" R&D (1993-1998: improve rad. hardness: purity, stoechiometry, defects)" Prototyping (1994-2001: large matrices in test beams, monitoring)" Mass manufacture (1997-2008: increase production, QC)" Systems Integration (2001-2008: tooling, assembly)" Installation and Commissioning (2007-2008)" Collision Data Taking (2009 onwards)" Δt ~ 20 " years "!!!" " Discovery of a new heavy boson (2012)" 5! Cambridge13!

Instrument Quality and Capability Matters Discovery of the Higgs Boson" Jul 2007! Infineon Oct15 tsv! 6!

The Higgs boson in CMS" CMS: H γγ Channel" CMS: H Z 4l Channel" T. Virdee, MA Apr15! 7!

Moving Forward Should we really expect new physics?" Ample observational evidence for physics Beyond the SM" Neutrino mass (oscillations)! Dark Matter! ν µ ν e! ν τ 2015! The lightness of the Higgs boson?! Matter-antimatter asymmetry! How did it arise?! Oxford Jan16! 8!

Physics Outlook: Questions for the LHC" 1. SM contains too many apparently arbitrary features - presumably these should become clearer as we make progress towards a unified theory.! 2. Clarify the e-w symmetry breaking sector" SM has an unproven element: the generation of mass! Higgs mechanism ->? or other physics?! Answer will be found at LHC energies" e.g. why M γ = 0! M W, M Z ~ 100,000 MeV!! Transparency from the early 90 s! 3. SM gives nonsense at LHC energies" Probability of some processes becomes greater than 1!! Nature s slap on the wrist!! Higgs mechanism provides a possible solution! 4. Identify particles that make up Dark Matter! Even if the Higgs boson is found all is not completely well with SM alone:! next question is Why is (Higgs) mass so low?! If a new symmetry (Supersymmetry) is the answer, it must show up at O(1TeV)" 5. Search for new physics at the TeV scale SM is logically incomplete does not incorporate gravity! Superstring theory dramatic concepts: supersymmetry, extra space-time 1/19/16! Physics Days 15 Helsinki tsv! 9! dimensions?!

Reminder that LHC is a Discovery Machine Search for γγ Resonances" T. Virdee, MA Apr15! 10!

Reminder that LHC is a Discovery Machine Search for γγ Resonances" Local σ Global σ Mass (GeV)! Γ (GeV) nwa! ATLAS" CMS (8+13)" 3.6! 2.0! 750! ~25! 3.0! 1.7! 750! ~25! T. Virdee, MA Apr15! 11!

2013: European Strategy " for Particle Physics " The Highest Priority" The Strategy in HEP" 2014 US P5: The Highest Priority! Oxford Jan16! 12!

SLHC Started a Long Time Ago" Oxford Jan16! 13!

What makes it worthwhile to run longer an HEP experiment? " 1. Higher centre-of-mass Energy! 2. Higher Integrated Luminosity! 3. Qualitatively better detectors! Oxford Jan16! 14!

LHC roadmap to achieve full potential" Prudently design for 5000 fb -1! Oxford Jan16! 15!

Recent LHCC and RRB Meetings: The Outcome" Based on the LHCC and USG findings and their subsequent endorsement by the Cern RB, the CERN Management is the RRB agreed with the following statement: The RRB considers the Step1 of the approval process for the Phase II Upgrades for the ATLAS and CMS experiments successfully completed. A scale of funding between the full funding and the intermediate scenario seems to meet the performance requirements. The CERN Management, supported by the findings of the LHCC and the UCG, considers the funding scenario realistic. The experiments are therefore encouraged to proceed to the next step of the Phase II upgrades S. Bertolucci! CMS reference design = 265 MCHF CORE (ATLAS = 275 MCHF)! 16! Oxford Jan16!

The Current CMS Detector!" 3000 scientists from 40 countries! Scintillating Crystals! Silicon Tracker! Gas ionization chambers! Originally designed for 500 fb -1! Brass plastic scintillator! Oxford Jan16! 17!

CMS Reference Upgrades for Phase II (TP)" Trigger/HLT/DAQ" Track information in Trigger (hardware)! Trigger latency 12.5 µs - output rate 750 khz! HLT output 7.5 khz! New Endcap Calorimeters " Rad. tolerant - increased transverse and longitudinal segmentation - intrinsic precise timing capability! Barrel EM calorimeter " New FE/BE electronics! Lower operating temperature (8 )! Muon systems" New DT & CSC FE/BE electronics! Complete RPC coverage 1.5 < η < 2.4! Muon tagging 2.4 < η < 3! Beam radiation and luminosity! Common systems &infrastructure! New Tracker " Rad. tolerant - increased granularity - lighter! 40 MHz selective readout (Pt 2 GeV) in Outer Tracker for Trigger! Extended coverage to η 3.8! Oxford Jan16! 18!

PbWO 4 Endcap ECAL will need Replacing" Due to Radiation Damage of PbWO 4 by hadrons" Oxford Jan16! 19!

Endcap HCAL will need Replacing" HCAL with PM readout: in situ measurements! Due to Radiation Damage of plastic scintillator by e.m. dose" Does replacement of PMs by higher q.e. SiPM s give enough margin up to LS3! Oxford Jan16! 20!

CMS Endcap Calorimetry needs Replacing! Start by considering the endcap region as a blank canvas! and ask! After the discovery of the Higgs boson! what are the physics priorities for the endcaps?! Oxford Jan16! 21!

Runs 2 and 3" Looking Ahead to Phase 1" 2009 2010 2011 Run 1 2012 ~25 fb -1 2013 2014 2015 2016 2017 2018 2025... LS1 Phase 0 Run 2 ~75-100 fb -1 LS2 2019 Phase I 2020 2021 Run 3 2022 ~350 fb -1 2023 2024 LS3 Phase II 2035? ~3000 fb -1 Conduct detailed studies of the properties of the found Higgs boson. Run II will produce > 5M Higgs bosons! Search for exotic decays of Higgs boson?! Exp. W Z g b t ATLAS [8,13] [6, 8] [7, 8] [8, 11] N/a [20, 22] [13, 18] CMS [5, 7] [4, 6] [4, 6] [6, 8] [10, 13] [14, 15] [6, 8] Couplings Precision ~ 5-15%" 300 fb -1! Search for new physics: resonances, supersymmetry, exotica, yet unknown." It is conceivable that we find new heavy particle(s) in Phase 1.! Look for deviations from the standard model precision SM measurements (e.g. tens of millions of top pairs produced/yr)! T. Virdee, MA Apr15! 22!

Looking Further Ahead to Phase 2 (HL-LHC)" Topmost Priority exploitation of the full potential of the LHC" High luminosity upgrade of the machine and detectors with a view to collecting ten times more data than in the initial design 2009 2010 2011 Run 1 2012 ~25 fb -1 2013 2014 2015 2016 2017 2018 2025... LS1 Phase 0 Run 2 ~75-100 fb -1 LS2 2019 Phase I 2020 2021 Run 3 2022 ~350 fb -1 2023 2024 LS3 2035? ~3000 fb -1 Conduct detailed studies of the properties of the found Higgs boson. " How much does it contribute to restoring unitarity in VBF (closure test of SM), exotic decays, rare decays (e.g. H µµ)! LHC HL-LHC - a Higgs factory! 100M produced 3ab -1! Phase II L(fb 1 ) Exp. W Z g b t Z 300 ATLAS [8,13] [6, 8] [7, 8] [8, 11] N/a [20, 22] [13, 18] [78, 79] [21, 23] CMS [5, 7] [4, 6] [4, 6] [6, 8] [10, 13] [14, 15] [6, 8] [41, 41] [23, 23] 3000 ATLAS [5, 9] [4, 6] [4, 6] [5, 7] N/a [8, 10] [10, 15] [29, 30] [8, 11] CMS [2, 5] [2, 5] [2, 4] [3, 5] [4, 7] [7, 10] [2, 5] [10, 12] [8, 8] Couplings Precision ~ 2-10%, H self coupling ~30% (needs study)! Continue searching for new physics. If new physics has been found by the end of Phase 1, associated particle(s) will be heavy. Then conduct detailed studies in Phase 3 (HL-LHC).! tsv Apr 2015! Zumino Memorial! 23!

Physics Case at HL-LHC" Physics Drivers for the Design of Endcap Calorimeter Upgrade" Some Examples:! 1. Physics with isolated objects (e, γ, µ, τ..) as today (low thresholds)! 2. Physics with VBF/VBS tags! 3. Physics using boosted objects (jet substructure in boosted W, Z, tops)! 4. QCD studies of quark v/s gluon jets! ALL at 140-200 events pileup! Take Example 2 and look in some detail:! - VBF H(ττ), H(inv, e.g. DM), Dark Matter, SUSY production (may be the only way to discover at LHC moderately heavy charginos /neutralinos)! - VBS: closure test of SM, bread and butter SM physics! Physics Challenges! Trigger (especially L1) e.g. VBF jets without condition on the rest of the event, Vertex ID, PU jet ID/rejection, timing, pileup suppression,..! Oxford Jan16! 24!

Event Display of VBF Jets (VBF H γγ)" Oxford Jan16! 25!

Event Display of VBF Jets (VBF H γγ)" Pileup <140>! Oxford Jan16! 26!

Event Display of VBF Jets (VBF H γγ)" Standalone simulation: Taking Slices through ECAL section! ΔR~0.2! Oxford Jan16! 27!

Event Display of VBF Jets (VBF H γγ)" Standalone simulation: Taking Slices through Si- HCAL section! Oxford Jan16! 28!

Time is short!" Look for a proven and adequately radiation hard active material! allowing a dense, good energy resolution combined e.m./ hadronic calorimeter, with a small R M, good two-shower (em and hadronic) shower separation,! with high lateral and longitudinal granularity! Explore a silicon based sampling calorimeter! (absorber material W, Pb, brass/fe)! Oxford Jan16! 29!

Choice of Silicon for Endcap Calorimetry" Recent advances in silicon detectors (where signal generation is well understood under irradiation), in electronics and data transmission suggest the possibility of their use in high granularity electromagnetic and hadronic calorimetry at the LHC." The obvious questions that arise when considering such a silicon sampling calorimeter are:! Cost of Si sensors (and overall cost)" HGC uses much simpler design than for strip trackers, cost basis has been established through a price enquiry from a reputable vendor.! Radiation tolerance of sensors and electronics! wide range of tests carried out to levels corresponding to 1.5 3000 fb -1! Level-1 trigger when using very high granularity" a new challenge ( but already can be faced with current technology)! Engineering challenge" integration, construction, assembly, thermal management, connectivity/ services,.! All of the above will benefit from synergies with the Tracker! Oxford Jan16! 30!

Si Sensors Radiation Tolerance" Complemented charged hadron irradiation with neutron irradiation! Circles indicate fluence for 3000 fb -1 and at 600V bias! Oxford Jan16! 31!

Si Sensors: Leakage Current" Oxford Jan16! 32!

Radiation Levels and Design Implication" 300µm! 200µm! 100µm! Oxford Jan16! 33!

CMS Endcap Calorimeter Needs to be Replaced" Oxford Jan16! 34!

Calorimeter Design" Construction:! Hexagonal Si-sensors built into modules.! Modules with a W/Cu backing plate and PCB readout board.! Modules mounted on copper cooling plates to make wedge-shaped cassettes.! Cassettes inserted into absorber structures at integration site (CERN)! Key parameters:! 593 m 2 of silicon! 6M ch, 0.5 or 1 cm 2 cell-size! 21,660 modules (8 or 2x6 sensors)! 92,000 front-end ASICS.! Power at end of life 115 kw.! System Divided into three separate parts:!!ee Silicon with tungsten absorber 28 sampling layers 25 X o + ~1.3 λ!!fh Silicon with brass absorber 12 sampling layers 3.5 λ!!bh Scintillator with brass absorber 11 layers 5.5 λ! EE and FH are maintained at 30 o C. BH is at room temperature.! Oxford Jan16! 35!

Performance studies in standalone setup" Our simulation benchmarked against published CALICE test-beam results! Electrons" Target 1% constant term (em)! Energy Resolution v/s Si thickness! em shower energy containment! Oxford Jan16! 36!

Performance studies in standalone setup" Our simulation benchmarked against published CALICE test-beam results! Pions" Energy Resolution: CALICE! Energy Resolution: HGC! 10% improvement after! software compensation! Oxford Jan16! 37!

Mechanical Structure: TP Design" Tungsten plates embedded in carbon fibre structure! Oxford Jan16! 38!

Calorimeter Design: Cassettes, Structure" CALICE Prototype! C-fibre structure with embedded W plates! HGC: 30 Sectors! Sensors mounted on both sides of Cassette! HGC: ECAL Structure! 3 wheels rotated wrt each other glued on intermediate disks and mounted on an inner cone! Cassette! Alveolar! Oxford Jan16! 39!

HGCAL : BH Structure" Oxford Jan16! 40!

Silicon Modules: TP Design" First modules will soon be made for test beams in 2016 Design as close to the final one as possible! Oxford Jan16! 41!

HGC Silicon Sensors" 6 or 8 wafer diameter (8 preferred)! n-on-p or p-on-n decide after irradiation tests & cost/benefit analysis! Sensors of three (active) thicknesses 100 µm, 200 µm, 300 µm (deep diffusion process or physical wafer thickness)! 2 different designs 128 and 256 channels (6 sensors), cell size ~ 0.5 cm 2, ~ 1 cm 2 (cell capacitance limited to about 60 pf)! DC coupled structure (leakage current compensating electronics)! Hexagonal shape for optimal use of wafer surface and to be placed like tiles! Total area of silicon ~ 600 m 2 " Received first 20 (of ~200) 6 hex p-on-n sensors (HPK) before Xmas. " for Test Beam prototypes" Also pursuing sensor development with Infineon and NOVATI in synergy with Tracker! Oxford Jan16! 42!

Si HGC Sensor development" Very productive meeting at Infineon Villach site (Austria) on October 28" High level attendance from Infineon included CEO & CTO of Infineon Villach as well as senior management representative from Infineon headquarters in Munich! An 8 HGC n-on-p sensors submission is under preparation! Oxford Jan16! 43!

HGCAL Test Beam Modules" Modified 8 wire-bonding jig to accommodate 6 prototypes" Oxford Jan16! 44!

Thermal Mockups and Tests" Module FEA! Thermally conductive epoxy between chips and PCB, regular epoxy on all other adhesive layers:! Maximum temperature (on PCB): -0.6 C! Maximum Sensor temperature: -28.5 C! Thermal gradient across Sensor: +1.3 C! Cassettes FEA! Goal: ΔT ~ 1-2 K! 6mm Cu plate 1 pipe uniform heat load! ΔT ~ 0.9K (over the cassette)! Cooling Tube: OD-4.8mm, ID-3.2mm, Length - 5.9 m, mass flow: 2.0 gm/sec,! T max -28.00C, T min -28-86C.! Sensor temperature! Oxford Jan16! 45!

Front-end Electronics: Specifications" Major challenge is large dynamic range @ low power and low noise (radiation hardness to 200 Mrads, 1.5 10 16 n eq /cm 2 )! Front-End Chip Requirements" Noise ~ 2 000e -" Including Sensor I leak noise (end of HL-LHC I leak max ~ 10 ua)! Shaping Time 10 ~ 20ns" Dynamic range ~ 10pC*" ~ 3 000MIP in 300um Silicon! Power ~ 10mW / channel" On-detector electronics ~ Σ100kW for ~ 6M channels! Precision timing" 50ps for cells with Q deposited > 60fC (~20ps per shower)! * Dynamic range of 10 pc (3000 mips @ 300um, 4500 mips @ 200 um, 9000 mips @ 100 um) roughly constant in E T. Increased energy compensates for smaller Si thickness! Oxford Jan16! 46!

f.e. ASIC - Design" Performance deduced using SPICE simulation! Into SARs ADC (<100fC)! ToT! 6pC! σ T0 ~ 50ps! T 0! 130 nm technology - incorporating! 10-bit SARs ADC: Existing and tested design! 12-bit TDC: Existing and tested design! Oxford Jan16! 47!

f.e. ASIC - Design" Performance deduced using SPICE simulation! Submit test vehicles in 130nm Mar16! Submit V1 ASIC Mar 2017! Oxford Jan16! 48!

Timing: e.m. and hadronic showers" e.m. intrinsic shower timing! Hadron shower timing! Oxford Jan16! 49!

Angular Resolution: e.m. showers" Oxford Jan16! 50!

HGC f.e. Electronics: Conceptual Architecture" Data Transfer capacity matched to 120um Si / 512 Channel Modules in high η region (high radiation forces use of Electrical Links)! For 300um Si / 256 Channel Modules (the bulk of the surface) option to combine data from adjacent Modules onto a single Optical Link! Lossless compression ( a la Huffman coding) used to transfer data out from the modules! TP Block Diagram! Oxford Jan16! 51!

Data Transfer for DAQ @ 1 MHz" An example of lossless compression" Oxford Jan16! 53!

Data Transfer for L1 @ 40 MHz " An example scheme" Oxford Jan16! 54!

Mip Calibration" Occupancy (> 0.5 mip)! <200 evts>! EE! FH! Layer No.! Sensor intercalibration will be tracked and maintained using mip signals in any triggered event! Targeting 0.5% contribution to constant term -> 3% inter-cell calibration! Oxford Jan16! 55!

Mip Inter-calibration of each cell" Charge calibration circuit! 2 overlapping ranges! Targeting 0.5% contribution to constant term -> 3% inter-cell calibration" mip tracking" e.g. ±1 layer MIP tracking! Need 1.5M events (even possible in HLT farm) to reach 3% precision with noise = 0.4 MIP, <NPU> = 140" Special small area pads for large S/N" Oxford Jan16! 56!

Level-1 Trigger: TP Design" The design of the following trigger architecture is based on existing or near-existing technology! FPGAs with similar capabilities as next generation Xilinx Virtex Ultrascale! 10 Gbps links (lpgbt) will be used for the transfer of data between the detector and the trigger system (mild radiation area)! 20 Gbps will be available for communication between off-detector trigger modules (no radiation)! The architecture is similar to the Phase-1 Stage-2 calorimeter trigger architecture! Two processing layers! The first one with a regional view of the detector! The second one with a global view of the detector! Time multiplexing to concentrate data in the second layer! Oxford Jan16! 57!

Level-1 Trigger" Group 4 cells to get a trigger cell" Simple data compression " Full resolution data" Each module produces up to 6 Gb/s! MP7" Oxford Jan16! 58!

Performance: Level-1 Trigger" e.m. Trigger Endcap region" Single em object! double em object! Oxford Jan16! 59!

Performance: Level-1 Trigger" Jet-Trigger Endcap region! Considerable power lies in the selection of events with difficult signatures! e.g. selection at L1 of VBF topologies without any requirement in the central region.! single jet! jet! Oxford Jan16! 60!

All cold Endcap" Oxford Jan16! 61!

BH: Scintillator Tiles: TP Design" Oxford Jan16! 62!

BH: A Possible Scintillator Readout Scheme" Inspired by CALICE Work" Oxford Jan16! 63!

BH: A Possible Scintillator Readout Scheme" Oxford Jan16! 64!

Integration: All-Cold HGCAL Cold-Warm Transition" Oxford Jan16! 65!

HGC Test Beam Calo Modules" Module! FNAL Test-Beam! Hanging File Calorimeter design! Oxford Jan16! 66!

HGC Test Beams" Q1-Q4 2016: Prototype beam tests with SKIROC electronics" February 2016: First fully functional Si-HGC modules equipped with existing SKIROC chips" Spring 2016: test beams at FNAL, with Si-HGC EE slice equipped with SKIROC chips" Commission test beam systems, especially read-out of module equipped with SKIROC chips! Enable tests of EM calorimetric response! Prepare for more detailed studies using modules equipped with SKIROC-CMS chips! Fall 2016: test beams at CERN, with Si-HGC EE and FH slice equipped with SKIROC-CMS chips" Enable detailed studies of calorimetric response and performance of baseline ASIC architecture and variants! Aim to include timing studies of EM and Hadronic shower evolution with ~50ps calorimeter cell timing resolution! Oxford Jan16! 67!

Event Display: VBF Jets" Oxford Jan16! 68!

Event Display: VBF Jets" Oxford Jan16! 69!

Performance Studies" Much work was carried out in the last year or so. " Most of it is described in the TP and the Scoping Document" The HGCAL simulation and reconstruction software has to be revised! Oxford Jan16! 70!

CMS Upgrades Schedule" Timeline for Sensors! Prototyping/freeze sensors specifications up to Q1 2018! Production Q2 2020 to Q4 2022! Timeline for Modules! Prototyping/freeze sensors specifications up to Q1 2019! Production Q3 2020 to Q1 2023! Timeline for f.e. ASICs! V1 submission Q1 2017! V2 submission Q2 2018 (+ if needed final one in Q1-2020)! Oxford Jan16! 71!

Summary I" The physics program in the HL_LHC phase will have a different emphasis now that we have found a Higgs boson" This will require precision and finesse across the broadest possible range of signatures! Jets, especially VBF jets, tau-jets, boosted jets (jet substructure) etc. will play an important role as well as traditional channels involving photons, electrons and muons such as H γγ and H 4 electrons! A design of an integrated sampling em and hadronic Endcap Calorimeter Upgrade has been established" Potentially allows a 5D reconstruction of showers/jets! The potential of high resolution timing, appears naturally in the f.e. design. It s use is being investigated more deeply.! Oxford Jan16! 72!

Summary II" The HGC+BH will be a powerful upgrade of CMS endcap calorimetry. The technology chosen should be able to tolerate up to 5000 fb -1 whilst retaining almost full performance." The challenges ahead lie in engineering (and prototyping). Construction will be launched after all final components/system are in hand and tested. Schedule has >1 year of contingency before first HL- LHC beams (2026)." High granularity calorimetry offers excellent potential and performance improvement for the physics at the HL-LHC." Oxford Jan16! 73!