Constraining h s s at lepton colliders Matthias Schla ff er - - PowerPoint PPT Presentation

Constraining h → s¯ s at lepton colliders

Matthias Schlaffer

Weizmann Institute of Science based on ongoing work with:

• J. Duarte-Campderros,
• G. Perez, A. Soffer

GGI, Florence August 2018

Gauge boson masses

Higgs is main source of electroweak symmetry breaking!

Parameter value 1 − 0.5 − 0.5 1 1.5 2 2.5 3 3.5 4 bb

µ

τ τ

µ

WW

µ

ZZ

µ

γ γ

µ

Run 1 LHC CMS and ATLAS ATLAS+CMS ATLAS CMS σ 1 ± σ 2 ±

µX = σBRX|meas.

σBRX|SM

[1606.02266]

Higgs couples to gauge bosons as expected

Matthias Schlaffer 1

What about fermion masses?

SM: economic solution, Higgs does it! h ⇒ v ⇒ mψ ∝ y

Particle mass [GeV]

1 −

10 1 10

2

10

v

V

m

V

κ

• r

v

F

m

F

κ

4 −

10

3 −

10

2 −

10

1 −

10 1 W t Z b µ τ ATLAS+CMS SM Higgs boson ] fit ε [M, 68% CL 95% CL

Run 1 LHC CMS and ATLAS

[1606.02266]

Matthias Schlaffer 2

However

Large mass hierarchy

| | | | | | | | | |

u p d

• w

n c h a r m s t r a n g e t

• p

b

• t

t

• m

e l e c t r

• n

m u

• n

t a u H i g g s 106 107 108 109 1010 1011 1012 Mass [eV]

[Thanks to Gilad Perez]

Matthias Schlaffer 3

Measurement of Yukawa couplings

Does the SM Higgs generate fermion masses? µX = prod × BR(h → X) SM × BRSM(h → X) > tth, h → ⌧⌧, h → bb > 5 (X ) > h → µµ: µµµ < 2.8 at 95 % CL

[ATLAS: 1705.04582, CMS: 1807.06325]

Lighter fermions even less constrained!

Matthias Schlaffer 4

Difficulties

i) small branching ratio

[GeV]

H

M

120 121 122 123 124 125 126 127 128 129 130

Branching Ratio

• 4

10

• 3

10

• 2

10

• 1

10 1

LHC HIGGS XS WG 2016

b b τ τ µ µ c c gg γ γ ZZ WW γ Z

[LHCHXSWG]

| | | | | | | | | | u p d

• w

n c h a r m s t r a n g e t

• p

b

• t

t

• m

e l e c t r

• n

m u

• n

t a u H i g g s 106 107 108 109 1010 1011 1012 Mass [eV] Matthias Schlaffer 5

Difficulties

i) small branching ratio

[GeV]

H

M

120 121 122 123 124 125 126 127 128 129 130

Branching Ratio

• 4

10

• 3

10

• 2

10

• 1

10 1

LHC HIGGS XS WG 2016

b b τ τ µ µ c c gg γ γ ZZ WW γ Z

[LHCHXSWG]

| | | | | | | | | | u p d

• w

n c h a r m s t r a n g e t

• p

b

• t

t

• m

e l e c t r

• n

m u

• n

t a u H i g g s 106 107 108 109 1010 1011 1012 Mass [eV]

ii) difficult final state for quarks > quarks appear as jets > large background > hard to distinguish Nevertheless: h → cc will be measured at % level at FCC-ee [1310.8361] What about strange?

Matthias Schlaffer 5

Exclusive decay h → φγ [1306.5770, 1406.1722]

• h
• ¯

s s +

• h
• ¯

s s > Clean decay: BR((s¯ s) → K+(u¯ s) + K−(¯ us)) ≈ 50% > BUT: BR(h → ) ≈ 2 × 10−6 [1505.03870] > compare BR(h → s¯ s) ≈ 2 × 10−4 > only weak limit at future (hadron) colliders [1406.1722] > current limit: BR(h → ) < 4.8 × 10−4 [1712.02758] Ideas to use differential distributions [see e.g. 1606.09253,

1606.09621, 1609.06592, 1611.05463]

Matthias Schlaffer 6

Brute force method

Alternative ansatz: > FCC-ee will produce O(106 − 107) Higgses via e− e+ Z∗ Z h > O(200 − 2000) of which decay into strange quarks > tag strange jets > Done before in Z → s¯ s

– Measurement of the strange quark forward backward asymmetry around the Z0 peak [DELPHI Collaboration, Eur.Phys.J. C14 (2000)] – Light quark fragmentation in polarized Z0 decays [SLD Collaboration, Nucl.Phys.Proc.Suppl. 96 (2001)]

Matthias Schlaffer 7

Setup and assumptions

data ⇒ kinematic separation cut&count, BDT,...

h → jj

• ther bkg

s-tagger ⇒ limit

Setup and assumptions

data ⇒ kinematic separation cut&count, BDT,...

h → jj

• ther bkg

s-tagger ⇒ limit Part I: > Clean sample with 107 Higgses > Only background other Higgs decays (h → gg, bb, cc) > We know which jets originate from the Higgs decay > Generate and shower with PYTHIA > No detector simulation

Matthias Schlaffer 8

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum.

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum. In which kaons can a s quark hadronize? K± K0

S

K0

L

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum. In which kaons can a s quark hadronize? K± vis. 1/6 inv.1/3

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum. In which kaons can a s quark hadronize? K± vis. 1/6 inv.1/3 K± vis. 1/6 inv.1/3 CC/NC/NN=9/6/1

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum. Charged kaon reconstruction: > stable on detector scales > tracking efficiency 95% > Particle ID ⇡± K±

some observable 2 bench marks e.g.: > no ID > ✏K = 95% ✏π = 12%

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum. Charged kaon reconstruction: > stable on detector scales > tracking efficiency 95% > Particle ID

[TopLC17 talk by Kurata]

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum. Charged kaon reconstruction: > stable on detector scales > tracking efficiency 95% > Particle ID

[TopLC17 talk by Kurata]

0.1 0.5 1 5 10 50 100 0.0 0.5 1.0 1.5 2.0 2.5 3.0 p [GeV]

• dE/dx resolution

10% 7% 6% 5% 4%

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum. Neutral K0

s reconstruction:

> Decay length ∼ 80 cm > Needs to decay to ⇡± within 5 mm < R < 1 m > reconstruction efficiency 80%

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum. Neutral K0

s reconstruction:

> Decay length ∼ 80 cm > Needs to decay to ⇡± within 5 mm < R < 1 m > reconstruction efficiency 80% jet 1 K+ K− K0

s

K+ jet 2 K+ K0

s

softer

> Keep hardest pair of kaons with charge sum |q1 + q2| < 2 > Split into CC, NC and NN channel

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum.

| [mm] |d 0.01 0.02 0.03 A.U. 50 100

s s b b c c gg WW* s s b b c c gg WW* s s b b c c gg WW* s s b b c c gg WW* s s b b c c gg WW*

Preliminary Impact parameter

> straight extrapolation

• f

tracks > no vertexing > O(60 − 80%) of kaon can- didates in b-jets stem from b-decays > O(40%) of kaon candidates in c-jets stem from c-decays > smearing according to mo- mentum and angle > 5 µm uncertainty on IP

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum.

s g

DK±(x, M2

Z)

[adapted from 0803.2768]

Fragmentation functions

Matthias Schlaffer 9

Setup for an s-tagger

Ansatz: s-jets dominantly contain a prompt kaon that carries a large fraction of the jet momentum.

[GeV]

||

p 5 10 15 20 A.U. 0.05 0.1 0.15

s s b b c c gg WW* s s b b c c gg WW* s s b b c c gg WW* s s b b c c gg WW* s s b b c c gg WW*

Preliminary parallel kaon momentum

> candidates from non-s-jets are soft, especially in g-jets > here: p(jet) ≈ 60 GeV

Matthias Schlaffer 9

s-tagging performance in CC channel

> impact parameter d0 < 15µm

2 4 6 8 10 12 14 10-5 10-4 10-3 10-2 10-1 100 p||

cut [GeV]

B

ss gg bb cc uu dd W

no particle ID Preliminary

Matthias Schlaffer 10

s-tagging performance in CC channel

> impact parameter d0 < 15µm

2 4 6 8 10 12 14 10-5 10-4 10-3 10-2 10-1 100 p||

cut [GeV]

B

ss gg bb cc uu dd W

with particle ID: ✏K = 95%, ✏π = 12% Preliminary

Matthias Schlaffer 10

Number of events

> impact parameter d0 < 15µm

5 10 15 20 10-8 10-6 10-4 10-2 p||

cut [GeV]

BR · no particle ID

ss gg bb cc uu dd W

Preliminary

Matthias Schlaffer 11

Number of events

> impact parameter d0 < 15µm

5 10 15 20 10-8 10-6 10-4 10-2 p||

cut [GeV]

BR · with particle ID: ✏K = 95%, ✏π = 12%

ss gg bb cc uu dd W

Preliminary

Matthias Schlaffer 11

Results part I

5 10 15 20 0.0 0.5 1.0 1.5 2.0 2.5 p||

cut [GeV]

Z0

107 Higgs bosons CC NC d0

cut=15m, K= 95%

d0

cut=15m, K=100%

d0

cut=24m, K= 95%

Preliminary

> strange Yukawa within reach of FCC-ee! > Improvements possible with vertexing

Matthias Schlaffer 12

Realistic Collider

Existing studies for h → bb, cc, gg: > Cut&Count: mh = 120 GeV [1207.0300] > Cut&Count+BDT [Talk by Yu Bai @ CEPC meeting] Assumptions: > h⌫⌫ final state (don’t consider h`` or hjj) > Only CC channel (no combination with NC) > Non-h → jj flavor decomposition as in BDT study flavor W bb uu dd cc ss gg relative abundance 65.3 9.8 6.1 6.0 6.4 6.0 0.2 > ✏W from ee → WW > Only statistical uncertainty

Matthias Schlaffer 13

Results part II

data ⇒ kinematic separation cut&count, BDT,...

h → jj

x

• ther bkg

y s-tagger ⇒ limit x-Axis: #(h → q¯ q) BR(h → q¯ q) ≈ #(h → gg) BR(h → g¯ g) ≈ Lh✏hjj y-Axis: # of all events that are not e+e− → ⌫⌫h, h → jj y-Axis: ≈ L P

bkg

bkg✏bkg

hjj

For each point (x,y) find best cut values to minimize upper limit

Matthias Schlaffer 14

Results part II

x-Axis: #(h → q¯ q) BR(h → q¯ q) ≈ #(h → gg) BR(h → g¯ g) y-Axis: # of all events that are not e+e− → ⌫⌫h, h → jj

102 103 104 105 106 107 # h → jj events 102 103 104 105 106 107 # reducible background events Cut&Count BDT L = 50 ab−1 L = 5 ab−1 L = 250 fb−1 1 2 5 10 20 50 100 200 500 95% CL on µ

Upper limit on µ p|| Preliminary

Matthias Schlaffer 15

Results part II

x-Axis: #(h → q¯ q) BR(h → q¯ q) ≈ #(h → gg) BR(h → g¯ g) y-Axis: # of all events that are not e+e− → ⌫⌫h, h → jj

102 103 104 105 106 107 # h → jj events 102 103 104 105 106 107 # reducible background events Cut&Count BDT L = 50 ab−1 L = 5 ab−1 L = 250 fb−1 0.95 0.96 0.97 0.98 0.99 1.0 K±|best

Best choice of PID p|| Preliminary

Matthias Schlaffer 15

Results part II

x-Axis: #(h → q¯ q) BR(h → q¯ q) ≈ #(h → gg) BR(h → g¯ g) y-Axis: # of all events that are not e+e− → ⌫⌫h, h → jj

102 103 104 105 106 107 # h → jj events 102 103 104 105 106 107 # reducible background events Cut&Count BDT L = 50 ab−1 L = 5 ab−1 L = 250 fb−1 14 15 16 17 18 19 20 21 22 23 24 dcut

0 |best [µm]

Best choice of d0 cut p|| Preliminary

Matthias Schlaffer 15

Results part II

x-Axis: #(h → q¯ q) BR(h → q¯ q) ≈ #(h → gg) BR(h → g¯ g) y-Axis: # of all events that are not e+e− → ⌫⌫h, h → jj

102 103 104 105 106 107 # h → jj events 102 103 104 105 106 107 # reducible background events Cut&Count BDT L = 50 ab−1 L = 5 ab−1 L = 250 fb−1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 pcut

|| |best [GeV]

Best choice of p|| cut Preliminary

Matthias Schlaffer 15

Results part II

x-Axis: #(h → q¯ q) BR(h → q¯ q) ≈ #(h → gg) BR(h → g¯ g) y-Axis: # of all events that are not e+e− → ⌫⌫h, h → jj

102 103 104 105 106 107 # h → jj events 102 103 104 105 106 107 # reducible background events Cut&Count BDT L = 50 ab−1 L = 5 ab−1 L = 250 fb−1 1 2 5 10 20 50 100 200 500 95% CL on µ

Upper limit on µ p|| Preliminary

Matthias Schlaffer 15

Conclusion

> s-tagger in the context of h → s¯ s > Proof-of-concept, can be improved > with ≈ 50 ab−1 (FCCee): µs . 5 > with ≈ 250 fb−1 (ILC): µs . O(50) > applicable to other searches with s-jets (up to some modifications)

Thank You

Matthias Schlaffer 16

BACKUP

Matthias Schlaffer 17

non-Higgs background [1207.0300]

> M2

recoil = s + m2 Z − 2EZ

√s independent of Higgs decay > signal-background separation in h → bb, cc, gg > with mh = 120 GeV > Z → invisible (20%):

– ✏h = 33% – S/B = 0.58

> Z → ee (3.4%):

– ✏h = 38% – S/B = 0.74

> Z → µµ (3.4%):

– ✏h = 47% – S/B = 1.4

> Z → hadrons (70%):

– ✏h = 26% – S/B = 0.08 [0902.3035]

Matthias Schlaffer 18

background flavor

ee → WW (ZZ, Z) ⌫⌫h (ZZ, Z) ⌧⌧h qqh WW fs ⌧⌫qq0 ⌫⌫dd non-jj ⌫⌫uu bb non-jj µ⌫qq0

• rel. [%]

47.1 18.0 13.7 12.2 2.7 2.5 2.0

Matthias Schlaffer 19