higgs portal at future colliders · higgs portal dark matter model p. ko, j. li. arxiv:1610.03997 l...
TRANSCRIPT
Characterizing DM properties with (non)minimalHiggs portal at future colliders
Pyungwon Ko
Korea Institute for Advanced Study
Based on B. Dutta, T. Kamon, P. Ko, Jinmian Li, arXiv: 1705.02149, 1712.05123
Aug. 28, 2018
COSMO 2018
IBS, Daejon, Korea
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 1 / 24
Outline
1 Generalities
2 Status of LHC searches for Higgs portal DM models
3 Prospects of searches at future colliders500 GeV ILC100 TeV FCC-pp/SppC
4 Conclusion
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 2 / 24
Crossing Relations and WIMP Detections
Crossing sym is an exact sym of a single Lagrangian in QFT
It should be applied to a model Lagrangian which is valid in allthe relevant kinematic region
One can not apply it to EFT or Simplified Models
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 3 / 24
Effective scalar operator
Limits after LHC run-1, red: mono-jet; green:one-lepton; yellow: two-leptons; blue: multi-jets.
σEFT/FT(pp→ χχ)@LHC
Region I: EFT limit is valid.
Region II: Resonant enhancementin FT.
Region III: σEFT ∝ 1m2med
VS
phase space factor in FT
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 4 / 24
Scalar mediated simplified model
J. Brooke, M. Buckley, et. al. arXiv:1603.07739
LH 3 −gχHχ̄χ−∑f
gvyf√2Hf̄f
LA 3 −igχAχ̄γ5χ−∑f
igvyf√2Af̄γ5f
Projected CMS invisible Higgs search in VBF
Constraint on scalar is weaker thanpseudoscalar, due to smaller cross section and
a softer EmissT .
Requires sizeable coupling for detection(gv = gχ = g), may not realistic, e.g.Γ > mH/A.
SM gauge symmetry is not conserved if H/Ais an singlet.
φ · f̄f = φ · (f̄LfR + f̄RfL)
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 5 / 24
Higgs portal dark matter model
P. Ko, J. Li. arXiv:1610.03997
LFDM 3 −gχSχ̄χ− λHSH†HS2 − µ30S
− µ1SH†H − µ2
3!S3 − λS
4!S4
LintFDM 3 −(H1 cosα+H2 sinα)
∑f
mf
vhf̄f
+ gχ(H1 sinα−H2 cosα) χ̄χ
gχ = 3, sinα = 0.3, mχ = 80 GeV(Throughout the talk. )
Production limited by SM Higgs-fermioncouplings.
Recast of the CMS invisible Higgs search inMono-jet.
Interference between two mediators (reducethe sensitivity at low MH2
).
All points way below the current LHCsensitivity.
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 6 / 24
Simplified vs. UV complete modelsS. Baek, P. Ko, M.H. Park, W.I. Park, C. Yu, arXiv:1506.06556
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 7 / 24
Outline
1 Generalities
2 Status of LHC searches for Higgs portal DM models
3 Prospects of searches at future colliders500 GeV ILC100 TeV FCC-pp/SppC
4 Conclusion
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 8 / 24
Minimal gauge invariant models: FDM, VDM and SDM
Lagrangians: Ko et al, arXiv:1112.1847 (SFDM), 1212.2131 (VDM)
LFDM = LSM + χ̄(i/∂ −mχ − gχS)χ + 12∂µS∂
µS − 12m2
0S2 − λHSH†HS2
−µ30S − µ1SH
†H − µ23!S3 − λS
4!S4
LVDM = − 14VµνV
µν +DµΦ†DµΦ− λΦ(Φ†Φ−v2φ2
)2 − λHΦ(H†H −v2h2
)(Φ†Φ−v2φ2
)
LSDM = 12∂µS∂
µS − 12m2
0S2 − λHSH†HS2 − λS
4!S4
Interaction Lagrangians:
LintFDM = −(H1 cosα +H2 sinα)
[∑f
mfvh
f̄f −2m2
Wvh
W+µ W
−µ −m2Zvh
ZµZµ
]+gχ(H1 sinα−H2 cosα) χ̄χ
LintVDM = −(H1 cosα +H2 sinα)
[∑f
mfvh
f̄f −2m2
Wvh
W+µ W
−µ −m2Zvh
ZµZµ
]− 1
2gVmV (H1 sinα−H2 cosα) VµV
µ
LintSDM = h [
∑f
mfvh
f̄f −2m2
Wvh
W+µ W
−µ −m2Zvh
ZµZµ] + λHSvh hS
2
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 9 / 24
Higgs portal DM within EFT
The models for the fermion and vector DM are not good, since
they are not renormalizablenot gauge invariant for vector DM casethus they violate unitarity
One cannot apply crossing relations for fermion or veector DM
Need to work in (the simplest possible) UV completions
In fact they are not even well-defined EFTs
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 10 / 24
General feature
HiDM
,t ≡ m2
DD
dσSDM
dt∝ σh
∗SDM × |
1
t−m2h
+ imhΓh|2,
dσFDM
dt∝ σh
∗FDM × |
1
t−m2H1
+ imH1ΓH1
−1
t−m2H2
+ imH2ΓH2
|2 ·(2t− 8m
2χ
),
dσVDM
dt∝ σh
∗VDM × |
1
t−m2H1
+ imH1ΓH1
−1
t−m2H2
+ imH2ΓH2
|2 ·(
2 +(t− 2m2
D)2
4m4V
).
[GeV]DDm0 50 100 150 200 250 300 350 400 450 500
Eve
nt F
ract
ion
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0.22
0.24
SM
SDM
VDMs
FDM200
FDM300
FDM400
FDM500
=3χParton, g
200 300 400 500 600 700 800
DDm
0.01
0.1
1
Eve
nt F
ract
ion
=3χ
Parton, g
SDM
FDM 200 300 500
VDM 200 300 500
=3χ
Parton, g
200 300 400 500 600 700 800 [GeV]DDm
1−10
1
10
(V/F
) ra
tio
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 11 / 24
DM production at the ILC
The dominant DM production process:
e+e− → Z(→ ff) H(∗)1,2 (→ DD)
DM pair four-momentum:PµDD = Pµ
e++ Pµ
e− − PµZ = (
√s− EZ ,−~pZ)
DM pair invariant mass:m2DD = s+m2
Z − 2EZ√s
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 12 / 24
Features of DM production at the ILC
[GeV]DDm0 50 100 150 200 250 300 350 400 450 500
Eve
nt F
ract
ion
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0.22
0.24
SM
SDM
VDMs
FDM200
FDM300
FDM400
FDM500
=3χParton, g
[GeV]DDm0 50 100 150 200 250 300 350 400 450 500
Eve
nt F
ract
ion
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
SM
SDM
VDMs
FDM200
FDM300
FDM400
FDM500
=3χ
Delphes, g
Dominant background processes:
e−
e+
νe
ν̄e
Zj
j
e−
e+
Z
Z
ν̄e
νe
j
j
e−
e+
W
νe
ν̄e
Z
ν̄e
j
j
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 13 / 24
Discovery prospects of the hadronic channel (FDM)
Preselection cuts:
Lepton veto
Exactly two jets
EmissT > 50 GeV
Boosted decision tree analysis with inputs:
mDD, pT (j1), pT (Z), EmissT , ∆φmin, pT (j2), mjj
10−3
10−2
10−1
100
−0.8 −0.6 −0.4 −0.2 0 0.2 0.410−3
10−2
10−1
100
−0.8 −0.6 −0.4 −0.2 0 0.2 0.4
NS/N
B
BDT Cut
FDM200FDM300FDM400FDM500
NS/N
B
BDT Cut
FDM200FDM300FDM400FDM500
0
2
4
6
8
10
12
14
−0.8 −0.6 −0.4 −0.2 0 0.2 0.40
2
4
6
8
10
12
14
−0.8 −0.6 −0.4 −0.2 0 0.2 0.4
Significance
BDT Cut
L =1000 fb−1
FDM200FDM300FDM400FDM500
Significance
BDT Cut
L =1000 fb−1
FDM200FDM300FDM400FDM500
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 14 / 24
Discovery prospects of the hadronic channel (FDM)
FDM200 FDM300 FDM400 FDM500
σ0 [fb] 1.643 0.9214 0.4221 0.2526
εpre 0.796 0.717 0.655 0.698
BDT 0.3615 0.2132 0.1929 0.2129
NS/1000 fb−1 697.8 410.5 148 102
NB/1000 fb−1 2248.5 11453.5 12736 10898
NS/√NS +NB 12.85 3.769 1.31 0.97
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 15 / 24
Outline
1 Generalities
2 Status of LHC searches for Higgs portal DM models
3 Prospects of searches at future colliders500 GeV ILC100 TeV FCC-pp/SppC
4 Conclusion
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 16 / 24
Signal: mono-jet?
Tmisp
0 100 200 300 400 500 600 700 800
Eve
nt F
ract
ion
0
0.05
0.1
0.15
0.2
0.25
0.3
Scalar
minΓFermion
minΓ×Fermion 20
minΓVector
minΓ×Vector 20
CMS mono-jet search,
arXiv:1703.01651
CMS tt+DM search in dileptonic
channel,
arXiv:1711.00752
M. R Buckley et.al.;
U. Haisch et.al.;
F.Boudjema et.al.;
J. Ellis et.al. ...
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 17 / 24
Signal: mono-jet?
Tmisp
0 100 200 300 400 500 600 700 800
Eve
nt F
ract
ion
0
0.05
0.1
0.15
0.2
0.25
0.3
Scalar
minΓFermion
minΓ×Fermion 20
minΓVector
minΓ×Vector 20
CMS mono-jet search,
arXiv:1703.01651
CMS tt+DM search in dileptonic
channel,
arXiv:1711.00752
M. R Buckley et.al.;
U. Haisch et.al.;
F.Boudjema et.al.;
J. Ellis et.al. ...
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 17 / 24
ttH channel is promising channel
The observed (expected) significance of 5.8 (4.9) stan-dard deviations
ttH cross section is enhanced more significantly withincreasing collision energy.
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 18 / 24
Signal and background
The dominant DM production process:
q
q̄
g
t̄
t
D
Dhi
g
g
hi
t
t̄
D
D
g
g
t̄
t
D
Dhi
Dominant background processes:
Cross section (NLO)
t̄t 1316.5 pbt̄tW 20.5 pbt̄tZ 64.2 pb
W`W`b 128.4 pb
Signal benchmark points:
t`t`+ DM
FDM200 34.2 fbFDM300 18.7 fbFDM400 14.8 fbFDM500 12.5 fb
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 19 / 24
Features of DM spin
200 300 400 500 600 700 800
DDm
0.01
0.1
1
Eve
nt F
ract
ion
=3χ
Parton, g
SDM
FDM 200 300 500
VDM 200 300 500
=3χ
Parton, g
200 300 400 500 600 700 800 [GeV]DDm
1−10
1
10
(V/F
) ra
tio
1− 0.8− 0.6− 0.4− 0.2− 0 0.2 0.4 0.6 0.8 1)llθcos(
0.1
Eve
nt F
ract
ion
=3χ
Parton, g
SDM
FDM200
VDM200
DM300
DM400
DM500
=3χ
Parton, g
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 20 / 24
Analysis strategy
Preselection: Exactly two opposite sign lepton and at least one bjet in the final state.
m`` /∈ [85, 95] GeV.
0 100 200 300 400 500 600 700 [GeV]T
missE
0
10
20
30
40
50
60
70
80
90
Cro
ss s
ectio
n [p
b]
=3χ
Delphes(After preselection), g
SM
SDM
FDM 200 300 400 500
VDM 200 300 400 500
=3χ
Delphes(After preselection), g
0 50 100 150 200 250 300(l,l) [GeV]T2m
0
10
20
30
40
50
60
70
80
Cro
ss s
ectio
n [p
b]
=3χ
Delphes(After preselection), g
SM
SDM
FDM 200 300 400 500
VDM 200 300 400 500
=3χ
Delphes(After preselection), g
EmissT > 150 GeV.
mT2(l, l) > 150 GeV.
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 21 / 24
Cuts flow for SM processes and Signals
Backgrounds t̄t t̄tW t̄tZ W`W`bCross section 1316.5 pb 20.5 pb 64.2 pb 128.4 pb
Presections 63.76 pb 351.8 fb 1.9 pb 25.4 pb
m`` /∈ [85, 95] GeV 59.8 pb 330.4 fb 1.05 pb 23.9 pbEmissT > 150 GeV 17.76 pb 69.61 fb 261.14 fb 3.5 pbmT2(l, l) > 150 GeV 23.83 fb 1.92 fb 32.1 fb 38.0 fb
Signals FDM200 FDM300 FDM400 FDM500Cross section 34.2 fb 18.7 fb 14.8 fb 12.5 fb
Presections 7.86 fb 3.99 fb 3.05 fb 2.55 fb
m`` /∈ [85, 95] GeV 7.47 fb 3.82 fb 2.92 fb 2.44 fbEmissT > 150 GeV 4.17 fb 2.44 fb 1.93 fb 1.63 fbmT2(l, l) > 150 GeV 0.87 fb 0.62 fb 0.54 fb 0.47 fb
L95% [fb−1] 305 602 793 1047
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 22 / 24
Spin characterization (EmissT and cos(θ``))
Two dimensional binned log-likelihood test: L(data|Hα) =∏i,j
tnijij e−tij
nij !
σsys. = 0.5%
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 23 / 24
Conclusion
The gauge invariant Higgs portal DM models for FDM and VDMrequire at least two mediators, while that of SDM only need one.
At the ILC, mH2 . 300 GeV can be probed at more than 3-σ level.
At the 100 TeV p-p collider, discovery/exclusion can be made withan integrated luminosity less than 1 ab−1 given a 1% systematicuncertainty, while the spin discrimination require integratedluminosity of a few O(10) ab−1 given a 0.5% systematicuncertainty.
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 24 / 24
Backup.
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 24 / 24
Benchmark points
The relevant parameters in FDM for collider search:gχ = 3, sinα = 0.3, mχ = 80 GeV and mH2
= (200, 300, 400, 500) GeV.
ΓFDMmin (H2) = Γ(H2 → χχ) + Γ(H2 → WW/ZZ) + Γ(H2 → ff)
= cos2α · g2
χ
mH2
8π(1−
4m2χ
m2H2
)3/2
+ sin2α ·
Gµm3H2
16√
2πδV
√√√√1− 4m2V
m2H2
(1− 4m2V
m2H2
+ 12m4V
m4H2
)
+ sin2α · (
mf
v)2 3mH2
8π(1−
4m2f
m2H2
)3/2
,
where f is the SM fermion, V = Z,W and δV = 1(2) for Z(W±).
Parameters for the vector DM production are chosen accordingly:sinα = 0.3, mV = 80 GeV and gV is chosen such that the total decaywidth of H2 is the same as benchmark points of FDM.
mH2[GeV] 200 300 400 500
Γmin(H2) [GeV] 14.2 60.1 103.0 144.5gV 3.53 3.07 2.37 1.91
Fix mS = 80 GeV and take appropriate λHS such that the productioncross section of the signal process is the same with that in the FDM.
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 24 / 24
Discovery prospects of the hadronic channel
Kinematic distributions:
(j1) [GeV]T
p0 50 100 150 200 250
Eve
nt F
ract
ion
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
SM
SDM
VDMs
FDM200
FDM300
FDM400
FDM500
(Z) [GeV]T
p0 50 100 150 200 250
Eve
nt F
ract
ion
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
SM
SDM
VDMs
FDM200
FDM300
FDM400
FDM500
[GeV]missTE
0 50 100 150 200 250
Eve
nt F
ract
ion
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05SM
SDM
VDMs
FDM200
FDM300
FDM400
FDM500
minφ ∆0 0.5 1 1.5 2 2.5 3
Eve
nt F
ract
ion
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
SM
SDM
VDMs
FDM200
FDM300
FDM400
FDM500
∆φmin
= mini=1,2
∆φ(pmissT , p(ji))
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 24 / 24
Spin characterization
The same preselection and BDT cuts as used for FDM the benchmark pointFDM200 (FDM300) are applied to the corresponding benchmark point SDM200(SDM300) and VDM200 (VDM300).
SDM200 SDM300 VDM200 VDM300
σ0 [fb] 1.643 0.9214 1.734 0.8674
εpre 0.7875 0.7875 0.801 0.711
NS/1000 fb−1 447 322.3 726 363.5
S 4.4 3.3 0.59 0.44
[GeV]χχm160 170 180 190 200 210 220 230 240 250
Eve
nt N
umbe
r
200
400
600
800
1000
1200
SDM+SM
VDM+SM
FDM+SM
=200 GeV)2H
After BDT cut (m
[GeV]χχm160 180 200 220 240 260 280 300 320 340 360
Eve
nt N
umbe
r
1000
2000
3000
4000
5000
6000
SDM+SM
VDM+SM
FDM+SM
=300 GeV)2H
After BDT cut (m
SDM: δχ2
=
5∑i=1
(NFDM+SMi −NSDM+SM
i√NFDM+SMi
)2
VDM: S = |NFDMS −NVDM
S |/√NB
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 24 / 24
Spin characterization
Using only the distribution of EmissT :H0 is the FDM + SM, Hα can be VDM/SDM + SM
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 24 / 24
Discovery prospects
200 300 400 500 600 700 [GeV]T
missE
0
5
10
15
20
25
30
35
Cro
ss s
ectio
n [fb
]
=3χ
Delphes(After cuts), g
SM
SDM
FDM 200 300 400 500
VDM 200 300 400 500
=3χ
Delphes(After cuts), g
Binned log-likehood analysis:
L(data|Hα) =∏i
tnii e−ti
ni!, Q = −2 log
(L(data|Hα)
L(data|H0)
).
σsys. = 1%
Pyungwon Ko (KIAS) Higgs portal DMs at colliders ICHEP2018 24 / 24