Upgrade of the LHCb Calorimeter system Yu Guz

Upgrade of the LHCb Calorimeter system Yu. Guz (IHEP Protvino) on behalf of the LHCb collaboration

Another LHCb talk at CHEF 2017: • Operation and performance of the LHCb calorimeter system – Maximilien Chefdeville 02 -Oct Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 2

The LHCb detector 2008 JINST 3 S 08005 Main subdetectors: • Vertex Locator (Ve. Lo): a silicon strip detector surrounding the IP • warm magnet, ~4 Tm; • two RICH detectors • electromagnetic calorimeter with preshower • hadron calorimeter • tracker stations (inner: silicon; outer: straw) • muon identification system Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 3

The LHCb Calorimetry System MUON SYSTEM HCAL beam PS/SPD ECAL More details: talk of Maximilien Chefdeville, this conference. Ø solid angle coverage 300 x 250 mrad Ø distance from IP: ~12. 5 m Ø four subdetectors: SPD, PS, ECAL, HCAL Ø based on scint. /WLS technique, light readout with PMT Ø provides: § L 0 trigger on high p. T e±, π0, γ, hadron § precise energy measurement of e± and γ § particle identification: e±/γ/hadron; contributes to Muon ID (HCAL). Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 4

The LHCb trigger: Run I & Run II ØHardware Level-0 trigger § SPD multiplicity CALORIMETRY § search for a highest ET object: • ET (e±/γ) > 2. 7 Ge. V CALORIMETRY • ET (hadr. ) > 3. 6 Ge. V CALORIMETRY • p. T (μ) > 1. 4 Ge. V/c MUON ID system § up to 1 MHz output -> adjustable thresholds to match beam conditions ØSoftware High Level Trigger (HLT) § ~26000 CPU cores ØStorage rate: 12. 5 k. Hz L 0 CALO: 60% of the L 0 bandwidth Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 5

Performance in physics analyses Many physics analyses rely on CALO performance (here are examples from Run I publications). B 0 → K*γ 5279± 93 ev J/ψ → µ+µη‘ → B 0→K*γ, BS→φγ: CP asymmetries are sensitive to NP contribution; the results agree with SM Yu. Guz CHEF 2017 J/ψ → µ+µη → π+ π- π0 B decays with π0 , η’ in final state: Radiative decays of B-mesons: 2011 data analyzed (1/fb-1) [NP B 867 (2013) 1]: π+π-η(γγ) study of decays B 0 (Bs )→J/ψη (η’) [JHEP 01 (2015) 024]: first observation of B 0→J/ψη’; BR of the four decays and their ratios; η – η’ mixing angle from R(η’) first observation of B 0→J/ψω, Bs→η’η’, Bs→ψ(2 S)η’. Upgrade of the LHCb Calorimeter system 6

LHCb upgrade In 2010 -2012, LHCb got 3. 2 fb-1 at 7 and 8 Te. V; ~5 fb-1 @13 Te. V expected by the end of 2018 Excellent operation at luminosity of ~2· 1032 cm-2 s-1. There is strong physics case to continue the flavor physics programme. Upgrade – for work at higher luminosities: 2· 1033 (Phase 1) and 2· 1034 (Phase 2) cm-2 s-1. Plan to get ~50 and ~300 fb-1, to probe NP at % and ‰ level. LHCb Upgrade Eo. I: CERN-LHCC-2008 -007 LHCb Upgrade Lo. I: CERN-LHCC-2011 -001 LHCb Upgrade Framework TDR: CERN-LHCC-2012 -007 LHCb Phase-2 Upgrade Eo. I: CERN-LHCC-2017 -003 Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 7

LHCb Upgrade(s) timeline LHCb Present state running at L=(3 -4)4· 1032/cm 2/s LHCb Upgrade 1 a L-->2· 1033/cm 2/s fully confirmed, work ongoing LHCb Upgrade 1 a continue at L=2· 1033/cm 2/s LHCb Upgrade 2 L-->2· 1033/cm 2/s Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 8

LHCb Upgrade – trigger organization for higher luminosity

LHCb upgrade With the present trigger organization, 1 MHz L 0 limit saturates the physics processes yield with increasing luminosity (especially L 0 Hadron)! The hadron trigger selects b-events, but not particular final state. At high lumi, each event contain many pp-interactions, so L 0 Hadron selectivity suffers from pileup. We have to stay within 1 MHz (450 k. Hz for hadrons) by means of increasing the p. T threshold; but this just reduces rate, the fraction of events of interest remains the same, thus their rate saturates. NB The Dimuon trigger is more selective for J/ψϕ, and saturates later than the Hadronic one Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 10

The LHCb trigger: Upgrade Solution for LHCb Upgrade: ØNo Level-0 trigger any more! ØEvent selection will be based of full event reconstruction Ø A reduced analog of L 0, LLT, is foreseen for the first year of upgrade operation (commissioning) The trigger & DAQ scheme re-organization is the first step for any LHCb upgrade, absolutely inevitable. The electronics has to be rebuilt for all subdetectors. Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 11

LHCb CALO: ageing effects

Radiation environment @ CALO G. Corti, M. Karacson, LHCB-PUB-2017 -013 For ECM = 8 Te. V 14 Te. V: factor of ~ 1. 7 for same integrated luminosity (coordinate dependent) Dose map @ ECAL front, corresponding to the 2012 FLUKA simulation (2. 21 fb-1, ECM = 8 Te. V) Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 13

Radiation environment @ CALO Instrumented area: |x|>32 cm Dose profiles at the depth of EM shower max. Rapid decrease with |x|. Dose profiles along z at different x positions. For 14 Te. V operation, cells closest to the beam pipe receives per 1 fb -1 (numbers valid within ± 50%): ECAL front – 1. 6 k. Gy; HCAL front – 0. 5 k. Gy ; ECAL, z of EM shower max – 3 k. Gy; HCAL, z of hadronic shower max – 0. 8 k. Gy ECAL back – 1. 2 k. Gy; HCAL back – 0. 05 k. Gy; Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 14

ECAL radiation hardness Irradiation of Inner modules in the LHC tunnel 137 Cs source scan along the module Two ECAL modules were installed in 2009 in the LHC tunnel upstream of the LHCb interaction point, close to the beam pipe. The dose is measured at the module surface. Sizeable uncertainty (~ x 1. 5 – x 2) in obtaining the dose in plastic inside the module The module may be considered operational till ~3 -4 Mrad at shower maximum: – 10… 13 fb-1 @ 14 Te. V for the innermost cells (32 cm from the beam) – 25… 30 fb-1 @ 14 Te. V at 48 cm from the beam – … Other irradiation tests were done at the CERN PS Irrad facility, the results are consistent Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 15

HCAL radiation degradation The HCAL radiation degradation can be evaluated in situ from the 137 Cs calibration data. The hadronic shower maximum lays within the tile row 0 (ECAL is ~1. 2 λI); the dose in the row 5 is much less. The light yield degradation in a tile row #i is approximately: Conclusion on HCAL central modules ageing: • the light yield degradation is moderate for LHCb Run I and Run II, even in the centre; • we may expect faster degradation for Run III, at 2· 1033 cm-2 s-1; • we should not worry about it, because HCAL will be used only during first year(s) of Run III (LLT) Yu. Guz CHEF 2017 LS 1 + beam commissioning Upgrade of the LHCb Calorimeter system 16

PMT ageing HCAL ECAL Significant DC anode current in HCAL and ECAL PMTs occurs during operation. Leads to degradation (dynode system), which is compensated by increasing HV. HCAL: up to 150 C·fb; ECAL: up to 30 C·fb not irradiated 20 k. Gy PMT entrance window (HAMAMATSU UV glass) degradation after 20 k. Gy at hadron beam (and 2 days annealing) is ~5% at Y 11 emission peak. Cockcroft-Walton (CW) HV generator boards mounted at each PMT remain operational till ~ 15 k. Gy PMTs and CW boards will be regularly replaced as needed Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 17

LHCb CALO Upgrade (Phase I)

LHCb CALO Upgrade Phase-1 LHCb PID Upgrade TDR, CERN-LHCC-2013 -022 Luminosity 2· 1033 cm-2 s-1 (~5. 5 pp interactions per event): Ø present ECAL and HCAL will be kept Ø PS and SPD will be removed: no need for particle ID in L 0 Ø The Frontend electronics will be rebuilt to cope with new DAQ & Trigger Ø The PMT gain will be reduced by factor of 5, to reduce PMT degradation • PMT linearity: OK within required dynamic range Ø to compensate, the FE gain will be increased x 5 Ø low noise frontend ASIC (ICECAL) developed Ø detector maintenance should follow radiation degradation of detector components: • regular replacement of degraded parts (PMTs / Cockcroft-Walton HV boards) • LS 3: replacement of ECAL Inner modules Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 19

LHCb CALO Upgrade: FEB • • • number of channels: 32 input impedance: 50 Ω integration time: 25 ns sensitivity: 4 f. C / ADC count dynamic range: 12 bit dynamic pedestal subtraction timing: individual per channel non-linearity: <1% spill-over: <1% cross-talk: <0. 5% Analog processing CALO FEB prototype Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 20

LHCb CALO Upgrade: LS 2 activities Ø PRS/SPD dismantling Ø New DAQ infrastructure installation Ø ECAL/HCAL electronics replacement PRS/SPD installation, 2006 Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 21

CALO Upgrade Phase 1 b & 2

LHCb CALO: LS 3 activity LHCb PID Upgrade TDR, CERN-LHCC-2013 -022 Originally, it was planned to replace inner ECAL modules damaged by radiation (we have 32 spares) – quite significant amount of work. Lift for central columns of modules beam plug (lead) However at 2· 1033 cm-2 s-1 we can expect ECAL performance degradation at the centre due to insufficient granularity, and can use LS 3 not only replace central cells, but to improve ECAL. Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 23

LHCb CALO Upgrade Phase 1 b LHCb Phase-2 Upgrade Eo. I, CERN-LHCC-2017 -003 Phase 1 b LHCb Upgrade (~2025, LS 3): consolidation and enhancement. LHCb will continue running at L=2· 1033 cm-2 s-1, to reach ~50 fb-1. By LS 3, HCAL will not be necessary any more (no hardware EThadron selection) and will be dismounted, in favor of MUON system additional muon filters, or whatever else they find necessary to install ECAL: at L=2· 1033 cm-2 s-1, occupancy will be too high not only in the centre • instead of replacing only very central modules, it is suggested to enhance granularity by replacing all modules in the horizontal band • high granularity small Moliere radius Ø W based modules in the centre ? Ø plastic scintillator will work for Run IV Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 24

LHCb CALO Upgrade Phase-2 LHCb Phase-2 Upgrade Eo. I, CERN-LHCC-2017 -003 Phase 2 (~2031): the goal is a detector which is able to take data at L=2· 1034 cm-2 s-1, up to ~300 fb-1: up to 100 Mrad (hadrons >20 Me. V: ~1015/cm 2) zone >5 Mrad (plastic scint. not applicable) ECAL: migration to new technology at the centre (5 -10% of total surface). ECAL front Granularity should be adjusted for the rest, according to the occupancy! A viable solution can be Compact LYSO/W Shashlik originally developed for CMS • sampling W-based • low Moliere radius, high granularity possible • rad hard up to required doses / fluences Po. S(ICHEP 2016)239 Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 25

LHCb CALO Upgrade Phase-2 LHCb Phase-2 Upgrade Eo. I, CERN-LHCC-2017 -003 Issues, needs and challenges: 1. Fast timing detector for EM showers, in order to help assigning photons to correct PVs • ~20 ps time resolution over the whole ECAL surface (48 m 2) • can be either dedicated 4 D Si pixel detector layer(s) embedded into the Shashlik structure, or a separate “Preshower” detector (with converter) in front of ECAL • it will help also to improve coordinate resolution (important for di-photon mass reconstruction) 2. super rad-hard photodetector • PMTs • Si. PM (better time resolution) at low temperature (50 -100 K) 3. try to use Ga. GG instead of LYSO (talk of M. Korjik, Thursday) • less expensive (Gadolinium vs Lutetium) • equally rad hard (or better) • but: emission peak at 520 nm, appropriate WLS and Ga. GG samples from photodetector to be identified FOMOS-Materials (RU) 4. … Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 26

Conclusions ● The Calorimetry system of LHCb is running successfully since 2009, demonstrating quite good detector performance. ● Upgrade for higher luminosity (2· 1033 cm-2 s-1) during LS 2 is under preparation. The Calorimetry system will be subject to the following modifications: § The Preshower, SPD and lead converter will be removed § All the Front End electronics of HCAL and ECAL will be rebuilt. § The gain of all the PMTs will be reduced by factor of 5, with corresponding increase of sensitivity of input amplifiers of Front End Boards ● Further LHCb upgrades, Phase 1(b) and 2, are now under consideration, for LHC LS 3 and LS 4. The final goal is to have a detector running at L=2· 1034 cm-2 s-1. ● HCAL will be removed ● ECAL should be rebuilt using a new technology. ● a timing detector with few*10 ps precision should be introduced Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 27

Thank you!

Backup

PS / SPD Preshower detector: two planes of scintillator tiles, with 1. 5 cm thick lead plane between them. Size and segmentation: matches ECAL. The scintillator tiles are 15 mm thick. The light is captured and re-emitted by WLS fiber (3. 5 loops) glued in a deep groove machined at the surface of the tile. Light readout: multi-anode PMT. LED # cells Inner Middle Outer 2 tiles 12 x 12 cm 2 144 tiles 4 x 4 cm 264 tiles 6 x 6 cm 16 SPD Lead PS 64 -ch MAPMT VFE board The scintillation light is captured by WLS fibers glued into the tiles, and transported via clear fibers to 64 -channel HAMAMATSU multi -anode PMT R 7600 -00 -M 64 MOD. Both PS and SPD are equipped with LED monitoring system Nphe/MIP The light yield of all 12032 cells measured on cosmics at production: ~ 25+-12 ph. el. / MIP HV setting: uniform, ~700 -800 V Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 30

ECAL Inner 9 cells Middle 4 cells Outer 1 cell PMT CW base Shashlik technology • 4 mm thick scintillator tiles and 2 mm thick lead plates, ~25 X 0 (1. 1 λI); Moliere radius ~ 36 mm; • modules 121. 2 x 121. 2 mm 2, 66 Pb +67 scintillator tiles; Average performance figures from • Segmentation: 3 zones 3 module types, Inner (9 cells beam test (there is slight difference between zones): per module), Middle (4), Outer (1). Total of 3312 modules, 6016 cells, (7. 7 x 6. 3) m 2, ~100 tons. Light yield: ~ 3000 ph. el. / Ge. V • Light readout: PMT R-7899 -20, HAMAMATSU. HV supply: individual Cockcroft-Walton circuit at each PMT. Yu. Guz CHEF 2017 Energy resolution: Upgrade of the LHCb Calorimeter system 31

HCAL spacers beam scintillators row 0 row 1 row 2 beam row 3 row 4 row 5 master plate PMT Pipes of the source movement system An HCAL module with optics assembled light guide WLS fibers HCAL module: self-supporting structure containing either 16 outer or 8 outer + 32 inner cells. HCAL consists of 52 modules, 9. 5 tons each. • Tilecal technology (originally developed for ATLAS). • The volume ratio Scint: Fe ~ 3: 16. • Instrumented depth: 1. 2 m, 6 tile rows, ~5. 6λI • Outer cells: 26 x 26 cm 2, Inner : 13 x 13 cm 2 (half tiles) • Total of 1488 cells, 2 x 26 modules, 500 tons Performance from the beam test: • energy resolution • light yield 105± 10 ph. el. / Ge. V • Scintillator, fibers, PMTs, LED system: similar to ECAL • built-in 137 Cs calibration system Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 32

LHCb upgrade In 2010 -2012, LHCb got 3. 2 fb-1 at 7 and 8 Te. V; ~5 fb-1 @13 Te. V are expected by 2018 There is strong physics case to continue the flavor physics programme. Upgrade – to be able to get ~50 (Phase-1) and 300 (Phase-2) fb-1. ECAL related studies LHCb Upgrade Eo. I: CERN-LHCC-2008 -007 LHCb Upgrade Lo. I: CERN-LHCC-2011 -001 LHCb Upgrade Framework TDR: CERN-LHCC-2012 -007 LHCb Phase-2 Upgrade Eo. I: CERN-LHCC-2017 -003 Yu. Guz CHEF 2017 Upgrade of the LHCb Calorimeter system 33