Long Lived Particles in LHCb Upgrade II Elena DallOcco on behalf - - PowerPoint PPT Presentation

SLIDE 1

Long Lived Particles in LHCb Upgrade II

Elena Dall’Occo

• n behalf of the LHCb collaboration

HL/HE LHC Meeting Fermilab 04-06/04/2018

SLIDE 2

2

LLPs at LHCb

HL/HE LHC - 04/04/2018 Elena Dall’Occo

LHCb MS

Cherenkov drift tube pixel silicon strip ECAL HCAL muon

CMS

Int.J.Mod.Phys. A 30, 1530022 (2015)

<latexit sha1_base64="Vo3mpBtWeRIcrqoQfThuRVFr7tw=">AB/HicbVDLSgNBEOyNrxhf8XHTw2AQPIVdUdSDEPTiMQHXBJKwzE5mkyEzs8vMrBD8Fe8eFDxKn6HN29+ipPHQRMLGoqbrq7woQzbVz3y8nMzS8sLmWXcyura+sb+c2tWx2nilCfxDxWtRBrypmkvmG01qiKBYhp9WwezX0q3dUaRbLG9NLaFPgtmQRI9hYKcjvNDRrCxwYdIGOT1BDCZToIF9wi+4IaJZ4E1Io7X1UvgGgHOQ/G62YpIJKQzjWu65iWn2sTKMcDrINVJNE0y6uE3rlkosqG72R9cP0IFVWiKlS1p0Ej9PdHQueCG2nwKajp72h+J9XT0101uwzmaSGSjJeFKUcmRgNo0AtpigxvGcJorZWxHpYIWJsYHlbAje9MuzxD8qnhe9ig3jEsbIwi7swyF4cAoluIYy+EDgHh7hGV6cB+fJeXexq0ZzKzDX/gvP8A/NOWDA=</latexit><latexit sha1_base64="t+NoqN5q2F05YupKi0Qig+ofUN8=">AB/HicbVDLSgMxFM3UV62v8bFTJFgEV2VGFHUhFN24bMGxhc4wZNJMG5pkhiQj1FJc+R9uXKi4lX6HO7/BnzB9LR64MLhnHu5954oZVRpx/m0cjOzc/ML+cXC0vLK6pq9vnGjkxi4uGEJbIeIUYFcTVDNSTyVBPGKkFnUuh37tlkhFE3GtuykJOGoJGlOMtJFCe8tXtMVRqOE5PDqGvuQwVaFdErOCPAvcSekWN4ZVL8edgeV0P7wmwnOBEaM6RUw3VSHfSQ1BQz0i/4mSIpwh3UIg1DBeJEBb3R9X24b5QmjBNpSmg4Un9O9BXqsj08mRbqtpbyj+5zUyHZ8GPSrSTBOBx4vijEGdwGEUsEklwZp1DUFYUnMrxG0kEdYmsIJwZ1+S/xDktnJbdqwrgAY+TBNtgDB8AFJ6AMrkAFeACDO/AInsGLdW89Wa/W27g1Z01mNsEvWO/f2bKXcg=</latexit><latexit sha1_base64="t+NoqN5q2F05YupKi0Qig+ofUN8=">AB/HicbVDLSgMxFM3UV62v8bFTJFgEV2VGFHUhFN24bMGxhc4wZNJMG5pkhiQj1FJc+R9uXKi4lX6HO7/BnzB9LR64MLhnHu5954oZVRpx/m0cjOzc/ML+cXC0vLK6pq9vnGjkxi4uGEJbIeIUYFcTVDNSTyVBPGKkFnUuh37tlkhFE3GtuykJOGoJGlOMtJFCe8tXtMVRqOE5PDqGvuQwVaFdErOCPAvcSekWN4ZVL8edgeV0P7wmwnOBEaM6RUw3VSHfSQ1BQz0i/4mSIpwh3UIg1DBeJEBb3R9X24b5QmjBNpSmg4Un9O9BXqsj08mRbqtpbyj+5zUyHZ8GPSrSTBOBx4vijEGdwGEUsEklwZp1DUFYUnMrxG0kEdYmsIJwZ1+S/xDktnJbdqwrgAY+TBNtgDB8AFJ6AMrkAFeACDO/AInsGLdW89Wa/W27g1Z01mNsEvWO/f2bKXcg=</latexit><latexit sha1_base64="KO6mHdpVv2ZRs8Tjq8FnmfCblUA=">AB/HicbVBNS8NAEJ3Ur1q/4sfNy2IRPJVEFPUgFL14rGBsoQlhs92S3eTsLsRaij+FS8eVLz6Q7z5b9y2OWjrg4HezPMzItSzpR2nG+rtLC4tLxSXq2srW9sbtnbO/cqyShHkl4IlsRVpSzmHqaU5bqaRYRJw2o8H12G8+UKlYEt/pYUoDgXsx6zKCtZFCe89XrCdwqNElOjlFvhQoVaFdWrOBGieuAWpQoFGaH/5nYRkgsacKxU23VSHeRYakY4HVX8TNEUkwHu0bahMRZUBfnk+hE6NEoHdRNpKtZov6eyLFQaigi0ymw7qtZbyz+57Uz3T0PchanmaYxmS7qZhzpBI2jQB0mKdF8aAgmkplbEeljiYk2gVMCO7sy/PEO65d1Nxbp1q/KtIowz4cwBG4cAZ1uIEGeEDgEZ7hFd6sJ+vFerc+pq0lq5jZhT+wPn8A7x6Txw=</latexit> <latexit sha1_base64="/qnqePH96WhGzAE8Z3JVLm/OwSc=">AB+3icbVBNSwMxEJ31s9avao9eQktBEOquIOpBKHrxWMHaQncp2TbhibZJckKS6l/RQ8eVLz6R7z5b0w/Dtr6YODx3gwz8KEM21c9tZWl5ZXVvPbeQ3t7Z3dgt7+/c6ThWhDRLzWLVCrClnkjYM5y2EkWxCDlthoPrsd98oEqzWN6ZLKGBwD3JIkawsVKnUPQ16wncEcCXSK3eor8SqdQdqvuBGiReDNSrpX8o2cAqHcKX343Jqmg0hCOtW57bmKCIVaGEU5HeT/VNMFkgHu0banEgupgODl+hCpW6aIoVrakQRP198QC60zEdpOgU1fz3tj8T+vnZroPBgymaSGSjJdFKUcmRiNk0BdpigxPLME8XsrYj0scLE2LzyNgRv/uVF0jipXlS9WxvGFUyRgwMowSF4cAY1uIE6NIBABk/wCm/Oo/PivDsf09YlZzZThD9wPn8AVRGUSA=</latexit><latexit sha1_base64="9s8vpsgODPjH6KyBZEC10gJ+uQM=">AB+3icbVDLSsNAFJ3UV62vaJduhpaCIMREHUhFN24rGBsoQlhMp20Q2eSMDMRQqhf4d6NCxW3/oi7/o3Tx0JbD1w4nHMv94TpoxKZdtjo7Syura+Ud6sbG3v7O6Z+wcPMskEJi5OWCI6IZKE0Zi4ipGOqkgiIeMtMPhzcRvPxIhaRLfqzwlPkf9mEYUI6WlwKx6kvY5CvgJh1fQts6g1wjMum3ZU8Bl4sxJvVnzjp/HzbwVmN9eL8EZJ7HCDEnZdexU+QUSimJGRhUvkyRFeIj6pKtpjDiRfjE9fgQbWunBKBG6YgWn6u+JAnEpcx7qTo7UQC56E/E/r5up6MIvaJxmisR4tijKGFQJnCQBe1QrFiuCcKC6lshHiCBsNJ5VXQIzuLy8Q9tS4t506HcQ1mKINDUANHwAHnoAluQu4AIMcvIA38G48Ga/Gh/E5ay0Z85kq+APj6wdc9ZXO</latexit><latexit sha1_base64="9s8vpsgODPjH6KyBZEC10gJ+uQM=">AB+3icbVDLSsNAFJ3UV62vaJduhpaCIMREHUhFN24rGBsoQlhMp20Q2eSMDMRQqhf4d6NCxW3/oi7/o3Tx0JbD1w4nHMv94TpoxKZdtjo7Syura+Ud6sbG3v7O6Z+wcPMskEJi5OWCI6IZKE0Zi4ipGOqkgiIeMtMPhzcRvPxIhaRLfqzwlPkf9mEYUI6WlwKx6kvY5CvgJh1fQts6g1wjMum3ZU8Bl4sxJvVnzjp/HzbwVmN9eL8EZJ7HCDEnZdexU+QUSimJGRhUvkyRFeIj6pKtpjDiRfjE9fgQbWunBKBG6YgWn6u+JAnEpcx7qTo7UQC56E/E/r5up6MIvaJxmisR4tijKGFQJnCQBe1QrFiuCcKC6lshHiCBsNJ5VXQIzuLy8Q9tS4t506HcQ1mKINDUANHwAHnoAluQu4AIMcvIA38G48Ga/Gh/E5ay0Z85kq+APj6wdc9ZXO</latexit><latexit sha1_base64="XOLSF1NK8QXEQ+y+MJ+wE4fsoDU=">AB+3icbVBNS8NAEN3Ur1q/qj16WSwFTzERD0IRS8eKxhbaELYbDft0t1N2N0IdS/4sWDilf/iDf/jds2B219MPB4b4aZeVHKqNKO821VlbX1jeqm7Wt7Z3dvfr+wYNKMomJhxOWyF6EFGFUE9TzUgvlQTxiJFuNL6Z+t1HIhVNxL3OUxJwNBQ0phpI4X1hq/okKOQn3B4BR37DPqtsN50bGcGuEzckjRBiU5Y/IHCc4ERozpFTfdVIdFEhqihmZ1PxMkRThMRqSvqECcaKCYnb8BLaMoBxIk0JDWfq74kCcaVyHplOjvRILXpT8T+vn+n4IioSDNBJ4vijMGdQKnScABlQRrlhuCsKTmVohHSCKsTV41E4K7+PIy8U7tS9u9c5rt6zKNKjgER+AYuOActMEt6APYJCDZ/AK3qwn68V6tz7mrRWrnGmAP7A+fwBEG5K/</latexit>

LHCb
 2<η<5

• unique acceptance: 2 < η < 5
• low pile-up (~1-2 visible interaction)
• excellent vertex resolution (στ~45 fs for Bs0)
• excellent mass resolution (0.5% in µµ)

Int.J.Mod.Phys. A 30,1530022 (2015) JINST 10 (2015) P06013

• good jet reconstruction
• energy resolution ~10% for jets with

pT > 10 GeV

• b(c) tagging efficiency ~65%(25%)

for 0.3% light-parton contamination

SLIDE 3

3

LLPs at LHCb

HL/HE LHC - 04/04/2018 Elena Dall’Occo

• calibration + reconstruction in real time
• very soft triggers:

at hardware level (L0):

• muons with pT > 1.5 GeV
• calo deposits with ET > 3 GeV

at software level (HLT):

• topological triggers on detached vertices
• PID and jets in trigger
• excellent for light dimuons

(prompt and detached) new µµ turbo trigger with online muon id requirement (no pre-scale)

Many searches for LLPs performed and ongoing!

3

10

4

10

5

10

2

10

3

10

4

10

5

10

6

10

7

10

m(µµ) [ MeV ] Candidates / σ[m(µµ)]/ 2

LHCb

√s = 13 TeV µ+µ− µ±µ± prompt-like sample pT(µ) > 1 GeV, p(µ) > 20 GeV

• Phys. Rev. Lett. 120, 061801 (2018)

see Mike’s talk

SLIDE 4

4

LHCb Upgrades

HL/HE LHC - 04/04/2018 Elena Dall’Occo

2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2030 Run 3 Run 4 Run 2

EYETS

LS2

Injector upgrades

LS3

HL-LHC Installation,ATLAS/ CMS Phase 2 upgrades

LHCb Upgrade I

LHCb Upgrade I Installation starts

LHCb

LHCb Upgrade I(b): Incremental improvements/prototype detectors 2028 2028

L = 2 x 1033 Lint ~ 50 fb-1 Lint ~ 8 fb-1

LHCb Upgrade I installation starts LHCb Upgrade I(b): incremental improvements/prototype detectors

LHCb LHCb Upgrade I

~1 visible interactions ~5 visible interactions

Upgrade I

• no hardware trigger, full software trigger with readout of the full event at 40 MHz
• improved vertex resolution
• momentum resolution of the tracking system ~10-20% better
• reduced ghost rate

will extend LHCb reach for LLPs!

SLIDE 5

5

LHCb Upgrades

HL/HE LHC - 04/04/2018 Elena Dall’Occo

2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2030 Run 3 Run 4 Run 2

EYETS

LS2

Injector upgrades

LS3

HL-LHC Installation,ATLAS/ CMS Phase 2 upgrades

LHCb Upgrade I

LHCb Upgrade I Installation starts

LHCb

LHCb Upgrade I(b): Incremental improvements/prototype detectors 2028 2028

L = 2 x 1033 Lint ~ 50 fb-1 Lint ~ 8 fb-1

LHCb Upgrade I installation starts LHCb Upgrade I(b): incremental improvements/prototype detectors

LHCb LHCb Upgrade I

~1 visible interactions ~5 visible interactions

2031 2032 2033 2034 2035 2040 2030 Run 5 Run 6 LS4

LHCb Upgrade II

LS5

LHCb Upgrade II Run 4 and Run 5 2030 ++ Accumulate 300 fb-1 Pixel timing mandatory

L = 1-2 x 1034 Lint ~ 300 fb-1

LHCb Upgrade II

Upgrade II (wrt Upgrade I)

• 10x higher vertex multiplicity
• 10x higher particle multiplicity
• 10x higher radiation damage

~50 visible interactions

SLIDE 6

Detector improvements for LLPs?

6

Elena Dall’Occo HL/HE LHC - 04/04/2018

SLIDE 7

7

LHCb Upgrade II

HL/HE LHC - 04/04/2018 Elena Dall’Occo

How LHCb might look like…

• silicon and scintillating fibres

based tracking detectors

• new sub-detectors (magnet

stations, TORCH)

• high granularity ECAL with

extended dynamic range

• HCAL removed

Tracking system and trigger crucial for LLP searches! UT micro strips VELO pixels inner/middle/

• uter tracker

magnet side stations timing plane (TORCH)

CERN-LHCC-2017-003

see Matt’s talk tomorrow

SLIDE 8

8

VELO

Elena Dall’Occo

VELO plays a fundamental role in LLP searches:

• reconstruction of primary and secondary

vertices

• part of the tracking system
• decay length accessible by LHCb ~20 cm

(decay in the VELO)

HL/HE LHC - 04/04/2018

in upgrade II pixel timing becomes essential to reduce vertex mis-association in the high pile up environment!

unambiguous flight distance

e mary x:

unambiguous flight distance ambiguous flight distance

mary ces:

ambiguous flight distance

t0, t0 + Δt

clear assignment

+

clear assignment

single PV two PVs two PVs + timing

t0 t0 + Δt

SLIDE 9

1 10

2

10

3

10

[mm] z

500 − 500 1000

[mm] r (signed)

20 − 10 − 10 20

LHCb

9

VELO Material

Elena Dall’Occo

…a material map of the VELO is essential to reduce the background in LLP searches!

• beam-gas (helium) collisions
• material interaction along the full

length of the VELO

• secondary interactions of hadrons

used to map the material

• analysis performed for Run1 and Run2
• method already applied successfully on a LLP analysis

(Phys. Rev. Lett. 120, 061801 (2018))

…repeat for next runs?

HL/HE LHC - 04/04/2018

RF box at ~5 mm from the beam:

• < 5 mm: background dominated

by heavy flavour

• > 5 mm: background mainly from

material interaction

• the removal of the RF foil is under

study for upgrade II

• it would reduce the background

and improve the IP resolution

• if not feasible it is crucial to veto

the material interactions…

CERN-LHCb-DP-2018-002

SLIDE 10

10

VELO

Elena Dall’Occo

VELO Upgrade I performance at HL-LHC luminosity:

• ghost rate explodes
• tracking efficiency decreases
• spatial resolution degrades

HL/HE LHC - 04/04/2018

potential design improvements to recover performance losses:

• smaller pixels and thinner silicon

sensor

scenario1

• removal of RF box (biggest

contributor to material budget)

scenario2

Pseudorapidity

2 4 6

Ghost Fraction

0.1 0.2 0.3 0.4 0.5 0.6

• 1

s

• 2

cm

33

10 × 2

• 1

s

• 2

cm

34

10 × 2

Pseudorapidity ] c

• 1

[GeV

T

p 1/

1 2 3

m] µ resolution [

3D

IP

10 20 30 40 50 60 70 80 90 100 LHCb simulation

• 1

s

• 2

cm

33

10 × 2

• 1

s

• 2

cm

34

10 × 2

] c

• 1

[GeV

T

p 1/

1 2 3

m] µ resolution [

3D

IP

10 20 30 40 50 60 70 80 90 100 LHCb simulation

Scenario 1 Scenario 2

Pseudorapidity

2 4 6

Ghost Fraction

0.02 0.04 0.06 0.08 0.1 0.12 0.14

Scenario 1 Scenario 2

most likely to get additional gain (timing for reconstruction? timing for extrapolation?)

CERN-LHCC-2017-003

SLIDE 11

11

Tracks in LHCb Now

Elena Dall’Occo

Long tracks

• excellent spatial resolution close

to PV

• excellent momentum resolution in

magnitude and direction

Downstream tracks

• reconstruction of daughters of LLPs

that decay beyond VELO acceptance (20 cm<SV<200 cm)

• good momentum resolution

Upstream tracks

• reconstruction of charged particles

bent out of the acceptance

• excellent directional resolution
• approximate estimation of the

momentum possible

VELO track Downstream track Long track Upstream track T track VELO UT T1 T2 T3

most relevant tracks for LLP searches

HL/HE LHC - 04/04/2018

SLIDE 12

12

Tracks in LHCb Upgrade II

Elena Dall’Occo

Downstream tracks

• downstream timing with TORCH (Time

Of Internally Reflected CHerenkov light)

• target: precision per track ~15 ps
• track matching will improve with

consequent suppression of ghosts

Upstream tracks

• tracking acceptance can be

significantly increased for soft tracks by tracking stations on the magnet

• the momentum resolution will

improve

• a large fraction of track now lost will

be recovered

HL/HE LHC - 04/04/2018

D* → D0π+

• it will allow to timestamp LLP

decaying after the VELO and match it to the correct vertex

• it will provide particle id to lower

momenta through the ToF

CERN-LHCC-2017-003

SLIDE 13

13

Trigger

Elena Dall’Occo HL/HE LHC - 04/04/2018

LH

same strategy as Upgrade I:

• no hardware trigger
• triggerless readout at the LHC bunch

crossing rate

• fully software-based trigger with no

further offline processing

• signal classification at the trigger level

what can be further improved?

• studies ongoing on adding downstream

reconstruction at HLT1(‘retina’ scheme with dedicated processors in the DAQ)

EPJ Web Conf. 127 (2016) 00005

• more flexibility
• higher efficiency for low mass searches

and electron modes

• possibility to access lower jet masses

(emerging jets, jet substructures…)

CERN-LHCC-2017-003

SLIDE 14

What sensitivity can we reach with HL-LHC?

14

Elena Dall’Occo HL/HE LHC - 04/04/2018

SLIDE 15

15

LLPs Decaying to Jet Pairs

Elena Dall’Occo HL/HE LHC - 04/04/2018

• search for hidden sector LLP decays via

SM Higgs portal

• mass range: 25-50 GeV
• lifetime range: 2-500 ps
• dataset: run I (2 fb-1)
• signature: single displaced vertex with 2

associated jets

strategy

• trigger on displaced vertex
• requirements on jet pointing and material

interaction veto to reduce main backgrounds:

• vertex from heavy flavour decay or material

interaction

• SM dijet events
• fit of the di-jet mass in bins of lateral

displacement Rxy

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 0.4 <

xy < 1.0 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 1.0 <

xy < 1.5 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 1.5 <

xy < 2.0 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 2.0 <

xy < 3.0 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 3.0 <

xy < 5.0 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 5.0 mm <

xy

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 0.4 <

xy < 1.0 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 1.0 <

xy < 1.5 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 1.5 <

xy < 2.0 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 2.0 <

xy < 3.0 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 3.0 <

xy < 5.0 mm

s = 8 TeV

LHCb

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 5.0 mm <

xy

s = 8 TeV

LHCb

20 40 60 80 Mass (GeV/c2) 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 3.0 < Rxy < 5.0 mm

LHCb

πV πV b b b b H0

LHCb acceptance for 
 all 4 jets is only few %

in most cases only 1 πv within LHCb acceptance

• Eur. Phys. J. C77 (2017) 812

20 40 60 80 Dijet mass [GeV/c2] 10−1 1 10 102 103 104 Candidates / (2 GeV/c2) 3.0 < Rxy < 5.0 mm s = 8 TeV

LHCb

BR=1 best fit background

SLIDE 16

LLPs Decaying to Jet Pairs

Elena Dall’Occo . − . − . − . [m] . − . − .

• .
• .
• .
• .

(/SM

→) ( → vv)

.

LHCb

.

= , TeV

. v = GeV/ . v = GeV/ . v = GeV/ . v = GeV/ HL/HE LHC - 04/04/2018

16

upper limits set on SM-Higgs BR to dark pions

• Eur. Phys. J. C77 (2017) 812

SLIDE 17

10−4 10−3 10−2 10−1 100 101 102

πV cτ [m]

10 20 30 40 50 60 70 80

mπV [GeV/c2] Regions where B(H0 → πV πV ) > 50% is excluded at 95% CL ATLAS 20.3 fb−1 at 8 TeV LHCb 2.0 fb−1 at 7-8 TeV CMS 18.5 fb−1 at 8 TeV

LLPs Decaying to Jet Pairs

Elena Dall’Occo HL/HE LHC - 04/04/2018

17

. − . − . − . [m] . − . − .

• .
• .
• .
• .

(/SM

→) ( → vv)

.

LHCb

.

= , TeV

. v = GeV/ . v = GeV/ . v = GeV/ . v = GeV/

BR > 50%

upper limits set on SM-Higgs BR to dark pions

competitive and complementary limits to ATLAS and CMS!

what about HL-LHC?

pushing to low mass and lifetime

• Eur. Phys. J. C77 (2017) 812

SLIDE 18

10−4 10−3 10−2 10−1 100 101

πV cτ [m]

10 20 30 40 50 60 70

mπV [GeV/c2]

ATLAS dominated CMS dominated LHCb jet substructure

LHCb preliminary Regions where B(H0 → πV πV ) > 50% is excluded at 95% CL LHCb 300 fb−1 LHCb published Run 1

LLPs Decaying to Jet Pairs

Elena Dall’Occo HL/HE LHC - 04/04/2018

18

extrapolation to 300 fb-1

• signal and background scaled to 14 TeV
• conservative assumptions on detector performance (trigger, material interaction,

jet reco)

• optimistic assumptions on the effect of pile-up

10−4 10−3 10−2 10−1 100 101

πV cτ [m]

10 20 30 40 50 60 70

mπV [GeV/c2]

ATLAS dominated CMS dominated LHCb jet substructure

LHCb preliminary Different constraints on B(H0 → πV πV ) at 95% CL with 300 fb−1 at LHCb B > 50% excl. B > 5% excl. B > 2% excl. B > 1% excl.

SLIDE 19

19

LLP Decaying Semileptonically

Elena Dall’Occo HL/HE LHC - 04/04/2018

• 2 approaches:
• RPV mSUGRA neutralino as

benchmark mass range: 23-198 GeV

• simplified topologies, less model

dependent mass range: 25-50 GeV

4 production mechanisms considered

˜ χ ˜ χ

PA “˜ g” ˜ χ PB

h0 ˜ χ ˜ χ

PC

¯ ˜ q ˜ q ˜ χ ¯ q q ˜ χ

PD

• lifetime range: 5-100 ps
• dataset: run I (1+2 fb-1)
• signature: single displaced vertex

with several tracks and a high pT µ

• background dominated by bb
• search for massive LLP decaying semileptonically into SM particles

EPJC (2017) 77:224

pT

1 10 102 103 104

Data 7 TeV LV38 10ps LV38 50ps bb sim.

–

Entries/(0.5 mm) [mm]

5 10

LLP Rxy

15 20

e) LHCb

SLIDE 20

]

2

c LLP mass [GeV/

20 40 60 80

)

2

c Entries/(1.5 GeV/

1 −

10 1 10

Data 8 TeV Fit: total background signal c)

LHCb

20

LLP Decaying Semileptonically

Elena Dall’Occo HL/HE LHC - 04/04/2018

38 GeV, 5 ps

strategy

• trigger on µ + displaced vertex
• exploit µ isolation to define a signal and a

control region enhanced in background

• simultaneous fit of the LLP candidate mass

in the 2 regions to extract number of candidates

no significant excess observed

EPJC (2017) 77:224 double LLP from 125 GeV Higgs

LLP mass Cross-section [pb] [GeV/c2]

10-1 1 10 20 40 60

τLLP = 50 ps LHCb

√s =8 TeV RPV mSUGRA

LLP mass Cross-section [pb] [GeV/c2]

10-1 1 10 50 100 150 200

τLLP = 5 ps m˜

q = 1300 GeV

/c2 LHCb

√s =8 TeV

SLIDE 21

21

LLP Decaying Semileptonically

Elena Dall’Occo HL/HE LHC - 04/04/2018

10−4 10−3 10−2 10−1

˜ χ0

1 cτ [m]

10 20 30 40 50 60 70 80

mχ0

1 [GeV/c2]

LHCb preliminary Different constraints on B(H0 → ˜ χ0

1˜

χ0

1) at 95% CL with Run 1 data at LHCb

B > 50% excl. B > 25% excl. B > 10% excl. B > 5% excl.

strategy

• trigger on µ + displaced vertex
• exploit µ isolation to define a signal and a

control region enhanced in background

• simultaneous fit of the LLP candidate mass

in the 2 regions to extract number of candidates

EPJC (2017) 77:224 double LLP from 125 GeV Higgs

LLP mass Cross-section [pb] [GeV/c2]

10-1 1 10 20 40 60

τLLP = 50 ps LHCb

√s =8 TeV

]

2

c LLP mass [GeV/

20 40 60 80

)

2

c Entries/(1.5 GeV/

1 −

10 1 10

Data 8 TeV Fit: total background signal c)

LHCb

38 GeV, 5 ps

SLIDE 22

10−4 10−3 10−2 10−1

˜ χ0

1 cτ [m]

10 20 30 40 50 60 70 80

mχ0

1 [GeV/c2]

LHCb reachable LHCb jet substructure

LHCb preliminary Regions where B(H0 → ˜ χ0

1˜

χ0

1) > 5% is excluded at 95% CL

LHCb 300 fb−1 LHCb published Run 1

10−4 10−3 10−2 10−1

˜ χ0

1 cτ [m]

10 20 30 40 50 60 70 80

mχ0

1 [GeV/c2]

LHCb reachable LHCb jet substructure

LHCb preliminary Different constraints on B(H0 → ˜ χ0

1˜

χ0

1) at 95% CL with 300 fb−1 at LHCb

B > 5.0% excl. B > 1.0% excl. B > 0.5% excl. B > 0.1% excl.

Elena Dall’Occo HL/HE LHC - 04/04/2018

22

extrapolation to 300 fb-1

• signal and background scaled to 14 TeV
• conservative assumptions on detector performance (trigger, material interaction,

jet reco)

• optimistic assumptions on the effect of pile-up

LLP Decaying Semileptonically

SLIDE 23

23

Summary

Elena Dall’Occo

• 10x luminosity
• significant changes in the detector to cope with the challenging conditions:

many proposals/ideas! (3rd Workshop on LHCb Upgrade II)

• LHCb reach for LLPs significantly improves beyond luminosity scaling: pushing

towards low masses and low lifetimes

• proposal for a new dedicated detector (CODEX-b) to further extend physics

reach and capabilities for LLPs (see Mike’s talk)

• unique coverage complementary to ATLAS and CMS
• many inputs coming from the theory community:
• confining Hidden Valley arXiv:1708.05389
• soft bombs JHEP 08 (2017) 076
• rare Z decays to a hidden sector arXiv:1710.07635

With the upgrade II we have…

…growing interest for LLPs at LHCb!

HL/HE LHC - 04/04/2018

SLIDE 24

Back Up

24

SLIDE 25

0.5 1.0 2.0 5.0

mϕ (GeV)

10−12 10−10 10−8 10−6 10−4

sin2 θ

CODEX-b SHiP CHARM LHCb, 3 fb−1 LHCb, 300 fb−1

• 25

CODEX-b

Elena Dall’Occo HL/HE LHC - 04/04/2018

x

ϕ

SM SM

CODEX-b box UXA shield shield veto IP8 Pb shield DELPHI

General strategy: Look for decays-in-flight of LLPs from IP8

strategy: look for decays-in-flight of LLPs from IP8 Five benchmark LLP scenarios:

• Massive dark photon γd (kinetic mixing)
• Light O(GeV) scalar φ (Higgs mixing)
• Heavy neutral lepton (mixing with the

active neutrinos)

• h → dark glueballs (twin Higgs mixing

portal)

• QCD coupled ALP ma < 3π (diphoton

channel!)

arXiv:1708.09395

SLIDE 26

26

Extrapolation to HL-LHC

Elena Dall’Occo HL/HE LHC - 04/04/2018

assumptions: the pile-up won't affect much neither the jet reconstruction neither the backgrounds For jet reconstruction, we think we can have it under control by either:

• applying the same ML techniques to mitigate the pile up already used by

ATLAS/CMS

• removing neutrals from jets (resolution would degrade but this is done already for

Turbo jets) For the backgrounds, for LLPs there are two main sources, heavy flavour and material interactions:

• heavy flavour should be about the same
• material interactions: the understanding of the VELO material significantly

increased in the last year, and hopefully this will be also the case in the HL-LHC. If RF foil thinner or even completely removed, background drastically reduced

SLIDE 27

27

Turbo

Elena Dall’Occo HL/HE LHC - 04/04/2018

• nly exclusive decays

(nothing else saved )

• new intermediate

solution

• trigger candidate +

subset of reconstruction saved

• full event reconstruction can

be persisted

• variables such as isolation,
• bjects for jet reconstruction

can be saved

SLIDE 28

28

Confining Hidden Valley Model

LHC-LLP - 20/10/2017 Elena Dall’Occo

heavy flavour decay channel di-muon channel

arXiv:1708.05389

q

¯ q qv ¯ qv Z0

c ¯ c

D+ D−

πv πv πv πv

µ− µ+ qv ¯ qv Z0 q ¯ q µ− µ+

ηv ωv ωv ωv

potential better sensitivity than ATLAS/CMS

signature: 2 vertices with large separation from PV and significant separation between each other signature: a displaced vertex well separated from beamline

SLIDE 29

29

Soft Bombs

LHC-LLP - 20/10/2017 Elena Dall’Occo

JHEP 08 (2017) 076

Image by Matt Strassler

• large multiplicity of soft particles
• spherical event shape
• multiplicity scales linearly with energy
• might be signature of strongly-

coupled hidden valleys

“given that LHCb will eventually operate fully in the trigger-less mode, it will have unique sensitivity to soft signatures of new physics”

SLIDE 30

30

Rare Z Decays to a Hidden Sector

LHC-LLP - 20/10/2017 Elena Dall’Occo

arXiv:1710.07635

101 102 mA [GeV] 10−4 10−3 10−2 10−1 ε LHCb BaBar DY14 300 fb−1 DY14 3 ab−1 EWPO 20 30 50 mhD = 300 fb−1 3 ab−1

LHCb 15 fb−1

αD = 0.1

• FIG. 4: Sensitivity projections for the Z → hDA0 → 4` + X search for integrated luminosity of 300 fb1 (upper dark

lines) and 3000 fb1 (lower faint lines) at √s = 13 TeV and ↵D = 0.1. The dotted green line shows the projected LHCb sensitivity with 15 fb1 from Ref. [29], while the dashed lines show the projections for the proposed Drell-Yan search from Ref. [32]. Notation and existing bounds are the same as in Fig. 3.

• rare decays of the Standard Model Z boson are powerful probes of hidden sectors
• Z decays into hidden-sector particles typically give rise to large multiplicities of soft

particles