Available on CMS information server CMS CR -2017/133

The Compact Muon Solenoid Experiment

Mailing address: CMS CERN, CH-1211 GENEVA 23, Switzerland

Conference Report09 May 2017 (v3, 29 May 2017)

Search for the dark matter mediator with 13 TeVdata

Raffaele Angelo Gerosa for the ”Exotica, all searches expt SUSY”, ATLAS and CMS collaborations.

Abstract

Includes low mass dijet, low mass dileptons, 3rd generation, etc.

Presented at Moriond/EW2017 52nd Rencontres de Moriond on Electroweak Interactions and Unified Theories

Search for dark matter mediators at the LHC

Raffaele Gerosaon behalf of the ATLAS and CMS collaborations

University California San Diego (UCSD)9500 Gilman Drive, 0354 La Jolla (CA), United States (US)

Results of searches for dark matter mediators decaying either to invisible or visible particlesare presented looking at proton-proton (pp) collisions collected by the ATLAS and the CMSexperiments at the LHC, focusing mostly on 13 TeV data. No significant deviations from theStandard Model (SM) predictions have been observed across several search channels, probingeither spin-1 vector/axial-vector dark matter mediators or spin-0 scalar/pseudoscalar ones.

1 Introduction

Although the SM describes remarkably well everything we see in collider experiments, it is notan entirely satisfactory theory to explain all the phenomena we see in nature. Astrophysicalobservations have provided evidences for the existence of Dark Matter (DM) in the universe 1,2.The most current model assumes that DM candidates are weakly interactive massive particles(WIMPs). If such particles exist, direct pair production of WIMPs may occur in TeV-scalecollisions at the CERN LHC 3.

2 Direct searches for dark matter mediators

If DM particles are produced in pp collisions, they would not generate directly observable signalsin the detector. However, if they recoil against an object “X” radiated from the initial state,they may produce a large transverse momentum imbalance (Emiss

T ) in the event. These eventtopologies are usually called “mono-X” signatures, where “X” might be a hadronic jet, a photon,a weak boson, etc. These signatures represent the main candles used to search for direct pro-duction of DM in particle colliders. They are sensitive to a large range of DM masses, providingcomplementary information to direct and indirect DM searches. The interaction between SMand DM particles is mediated by a bosonic mediator, which mass scale could be accessible bythe LHC collisions. Simplified DM models are used as benchmarks 4,5,6,7, where the DM particleis assumed to be a Dirac fermion with mass mDM, which couples to SM quarks via a spin-1mediator, such as a Z’, or a spin-0 particle with unknown mass mmed.

The channel which provides the strongest bounds on both spin-1 and spin-0 mediators is the socalled “monojet” analysis, which consists in a search for pair-produced DM particles in associ-ation with high pT jets. Both ATLAS 8 and CMS 9 collaborations have performed this searchusing 2.3 fb−1 and 12.9 fb−1 of

√s = 13 TeV data, respectively 10,11. Candidate events are

accepted after requiring a high energetic jet, significant missing transverse energy and no ad-ditional leptonic activity in the detector. The main backgrounds, coming from Z → νν andW → `ν+jets, are modelled from dedicated control regions in data. No excess above the SMbackground prediction has been observed, thus exclusion limits at 95% CL are set on the ratioof the measured signal cross section to the predicted one.Fig. 1 (left) shows the exclusion contours as a function of the mediator mass, mmed, and theDM mass, mDM, assuming an axial-vector mediator with gq = 0.25 and gDM = 1. Mediatormasses up to 1.95 TeV and DM masses up to 550 GeV are excluded. In contrast, Fig. 1 (right)reports the exclusion sensitivity as a function of the mediator mass for a pseudoscalar mediatorwith gq = gDM = 1 and mDM = 1 GeV. In this case, the CMS monojet result is compared withalternative search channels, where a spin-0 mediator may be produced in association with a bbor a tt pair. Mediator masses up to 430 GeV are excluded at 95% CL.

Figure 1 – Exclusion limits at 95% CL on the signal strength µ = σ/σth assuming an axial-vector mediator (left)or a pseudoscalar one (right)11,12. In the axial-vector case, limits are shown in the mmed−mDM plane for gq = 0.25and gDM = 1, where the observed (expected) excluded area is the one below the solid (dotted) red (blue) contour.In the pseudscalar case, limits are shown as a function of mmed for gq = gDM = 1 and mDM = 1 GeV.

Moreover, the SM Higgs boson might play the role of a scalar dark matter mediator 13,14,15,16

since it couples to every particles with mass. Both ATLAS and CMS collaborations have per-formed direct searches for invisible decays of the Higgs boson, exploring different productionmechanisms: gluon fusion (monojet search), vector boson fusion, associated production withweak bosons (mono-V searches) or with tt pairs. Eventually, a combination of these searchesat√s = 7, 8 TeV performed by the ATLAS collaboration yields an upper limit of 25% on

the Higgs boson invisible branching fraction B(h → inv) 17. Similarly, the CMS collaborationperformed a combination using data collected at

√s = 7, 8, 13 TeV, which sets an upper bound

on B(h→ inv) of 24% 18.Fig. 2 (left) shows the CMS profile likelihood ratios as a function of B(h→ inv) for both partialcombinations of individual channels and the full combination. Upper limits on B(h→ inv) areinterpreted in the context of the Higgs portal model of DM interactions, where a hidden sectorprovides a stable DM candidate with tree-level couplings to the SM Higgs boson. Direct detec-tion experiments are sensitive to the elastic scattering between DM particles and nuclei throughHiggs boson exchange, producing nuclear recoil signatures which can be interpreted in terms ofDM-nucleon interaction cross section. When the DM candidate is lighter than mh/2, the limiton invisible Higgs decay width (Γinv) can be translated into a spin-independent DM-nucleon

cross section bound. Fig. 2 (right) shows the 90% CL upper limit on the spin-independent DM-nucleon cross section from the ATLAS collaboration, assuming mh = 125 GeV, for the scalar,majorana and vector DM particle scenarios.

inv)→B(H0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

q

0

1

2

3

4

5

6

7

8

9

10

CombinedqqH-taggedVH-taggedggH-tagged

ObservedExpected

(13 TeV)-1 (8 TeV) + 2.3 fb-1 (7 TeV) + 19.7 fb-14.9 fb

CMS

WIMP mass [GeV]1 10 210 310

]2W

IMP

-nuc

leon

cro

ss s

ectio

n [c

m

57−10

55−10

53−10

51−10

49−10

47−10

45−10

43−10

41−10

39−10

DAMA/LIBRA (99.7% CL)CRESST II (95% CL)CDMS SI (95% CL)CoGeNT (99% CL)CRESST II (90% CL)SuperCDMS (90% CL)XENON100 (90% CL)LUX (90% CL)

Scalar WIMPMajorana WIMPVector WIMP

ATLAS

Higgs portal model:ATLAS 90% CL in

-1 = 7 TeV, 4.5-4.7 fbs-1 = 8 TeV, 20.3 fbs

Vis. & inv. Higgs boson decay channels]

inv, BRγZκ, γκ, gκ, µκ, τκ, bκ, tκ, Zκ, Wκ[

<0.22 at 90% CLinv

assumption: BRW,ZκNo

Figure 2 – Left: CMS profile likelihood ratios as a function of B(h→ inv) assuming SM production cross sectionsof the Higgs boson, where the solid curves represent the observation in data, while the dashed ones are theexpectation from the background only hypothesis 18. The likelihood scans are shown for the full combination aswell as for the individual channels targeting different Higgs production modes. Right: ATLAS upper limit at 90%CL on the WIMP-nucleon cross section in the context of the Higgs portal model vs mDM

17. Limits are reportedseparately for a scalar, majorana or vector WIMP and they are determined from the 90% CL bound of 22% onB(h→ inv). ATLAS results are compared to several exclusion contours from direct detection experiments.

3 Indirect searches for dark matter mediators

A large variety of BSM models predict new narrow resonances which can be observed alongthe dijet mass spectrum, mjj . One of the most popular benchmark for the dijet search isrepresented by the Z’ leptophobic model 19,20, which predicts a spin-1 resonance coupled to SMquarks via a universal coupling constant, gq. If the Z’ is allowed to decay also to invisibleparticles, i.e. gDM 6= 0, this benchmark becomes equivalent to the simplified DM model for spin-1 DM mediators. Therefore, dijet measurements may allow to further constrain the existence ofDM particles.Both ATLAS and CMS collaboration have recently released measurements of the dijet spectrumbased on 2016 data 21,22 corresponding to 37 fb−1 and 35.9 fb−1, respectively. Due to a limitedamount of bandwidth available at the trigger level, these searches can only explore the dijetspectrum for mjj > 1.25 TeV. No excess over the predicted background has been observed.Fig. 3 (left) shows the ATLAS 95% CL exclusion limit on the quark coupling gg as a functionof the resonance mass, mZ′ , for the leptophobic Z’ model. In addition, both collaborationshave also performed a trigger object based analysis to search for resonances with mass in therange between 0.6 and 1.1 TeV 22,23. Eventually, to test the existence of resonances lighterthan 600 GeV, ATLAS looked at events where the mediator is boosted in the transverse plane,recoiling against a high-pT ISR photon or jet 24. This allows to probe resonances with massbetween 0.25 and 1 TeV for the X + γ search and between 300 and 600 GeV for the X + j case.Finally, another limitation occurs while going even lower in mass: for boosted resonances lighterthan ≈ 300 GeV, the jets arising from the decay overlap in the detector so that jet substructuretechniques are necessary to reconstruct signal events. CMS performed a search for boosted lowmass Z’ resonances probing, for the first time at hadron colliders, the mass region between 100and 300 GeV, using 2.7 fb−1 of 13 TeV data 25. Results are shown in Fig. 3 (right).

[TeV]Z’m

1.5 2 2.5 3 3.5

qg

0.05

0.1

0.15

0.2

0.25

0.3

-1 = 13 TeV, 37.0 fbs ATLAS

Observed 95% CL upper limit

Expected 95% CL upper limit

Z' mass (GeV)80 100 200 300 400 500

Bco

uplin

g, g

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5CMSPreliminary (13 TeV)-12.7 fb

expectedobserved

σexpected 2σexpected 1

UA2CDF Run 1CDF Run 2

-1 20.3 fb⊕ATLAS 13 )γ (ISR -1 3.2 fb⊕ATLAS 13

(TLA)-1 3.2 fb⊕ATLAS 13 (Scouting)-1 18.8 fb⊕CMS 8

Figure 3 – 95% CL exclusion limits on the quark coupling, gq, for the Z’ leptophobic model as a function of theresonance mass, mZ′ . Left: latest result from the ATLAS high mass analysis, mjj > 1.5 TeV 21. Right: exclusionfrom the CMS boosted dijet search in the mjj range between 100 and 300 GeV, where gq = gB/6

25. The exclusionapplies up to gq ≈ 0.5 when the narrow width approximation breaks down.

This set of complementary approaches allow to search for new physics in dijet events over a broadmass range, from 100 GeV up to 4 TeV. Dijet limits are used to constrain DM simplified modelswith a leptophobic vector and axial-vector mediators that couples to quarks and DM particles.Fig. 4 left (right) shows the exclusion contours in the plane defined by the mediator mass, mmed,and the DM one, mDM, for a vector (axial-vector) mediator assuming gq = 0.25 and gDM = 1.For this particular choice of couplings, the dijet analysis shows stronger sensitivity than directDM searches, excluding mediator masses between 0.5 and 2.5 TeV independently of mDM. Theexclusion power of the dijet search on mmed increases with mDM because the branching fractionin qq increases with mDM. When 2 ·mDM > mmed, the mediator cannot decay to DM particles,thus DM limits becomes equivalent to the leptophobic Z’ ones with gq = 0.25. In addition, when

mDM = 0, dijet DM bounds corresponds the ones on the Z’ mass at gq/√

1 + 16/3 ·Nf ≈ 0.18.

In contrast, direct searches have better sensitivity only for light mediators, which cannot beprobed by the dijet analysis.

Mediator mass [GeV]0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

[GeV

]D

Mm

0

200

400

600

800

1000

1200

1400

1600

CMS 95% CL[EXO-16-056]Dijet

[EXO-16-030]Boosted dijet

[EXO-16-037]qq

DM + j/V

[EXO-16-039]γDM +

[EXO-16-038]llDM + Z

DM

= 2 x m

MedM

= 1.0DM

g

= 0.25q

g

CMS 95% CL[EXO-16-056]Dijet

[EXO-16-030]Boosted dijet

[EXO-16-037]qq

DM + j/V

[EXO-16-039]γDM +

[EXO-16-038]llDM + Z

Dirac DMVector mediator

Preliminary CMS (13 TeV)-1 & 36 fb-1, 27 fb-113 fb

0.12≥ 2

hcΩ [EXO-16-056] Dijet

[EXO-16-030] Boosted Dijet

[EXO-16-037] qq

DM + j/V

[EXO-16-039] γDM +

[EXO-16-038] llDM + Z

Figure 4 – The 95% CL observed excluded regions for a vector (left) and axial-vector (right) mediator in the planeof mmed−mDM coming from the latest CMS and ATLAS direct (mono-X) and indirect (dijet) searches22,27. LHCresults are compared to the constraints from the cosmological DM relic density reported in gray. Exclusions arecomputed for a Dirac DM particle, a universal coupling gq = 0.25 and a DM coupling gDM = 1.

It should also be noted that the excluded region strongly depends on the chosen coupling andmodel scenario. For small values of gq, the Z’ production cross section decreases while thebranching ratio into DM increases, thus the dijet exclusion sensitivity decreases more rapidlythen the one from direct searches.

4 Comparison with direct and indirect detection experiments

The limits from ATLAS and CMS experiments, obtained using the simplified DM models, maybe compared to the ones from direct and indirect DM detection experiments, which are usuallyexpressed as bounds on the DM-nucleon scattering cross section. The approach outlined inRef. 26 is used to translate the exclusion contours in the mmed − mDM plane into limits onthe spin-dependent/spin-independent DM-nuclean cross section as a function of mDM. Resultsfrom CMS and ATLAS collaborations are shown in Fig. 5 for the vector (left) and axial-vector(right) mediators. When compared to direct detection experiments, limits from the LHC arethe strongest one for DM masses smaller than 5 GeV for the vector case. For an axial-vectormediator, LHC limits are the strongest one over the whole mDM range.

[GeV]DMm1 10 210 310

]2 [c

mD

M-n

ucle

onS

Iσ

47−10

46−10

45−10

44−10

43−10

42−10

41−10

40−10

39−10

38−10

37−10

36−10

35−10

Moriond 2017

arXiv:1509.02448CDMSlite

arXiv:1509.01515Cresst

arXiv:1607.07400PandaX

arXiv:608.07648LUX

[EXO-16-038] llCMS DM+Z

[EXO-16-039] γCMS DM+

[EXO-16-037] qq

CMS DM+j/V[EXO-16-056]CMS Dijet

[EXO-16-030]CMS Boosted dijet

Excluded at 90% CL

Preliminary CMS (13 TeV)-1 & 36 fb-1 & 27 fb-113 fb

Figure 5 – Excluded regions at 90% CL on the spin-independent (left) and spin-dependent (right) DM-nucleoncross section as a function of mDM as obtained from the latest CMS22 and ATLAS27 direct (mono-X) and indirect(dijet) searches. The exclusion contours from LHC measurements are compared with limits from CDMSlite, Cresst,PandaX and LUX for the spin-independent case, while only LUX is shown in the spin-dependent one.

References

1. J. L. Feng, “Dark Matter Candidates from Particle Physics and Methods of Detection”,Ann. Rev. Astron. Astrophys. 48 (2010) 495.

2. T. A. Porter, R. P. Johnson and P. W. Graham, “Dark Matter Searches with AstroparticleData”, Ann. Rev. Astron. Astrophys. 49, 155 (2011).

3. M. Beltran, D. Hooper, E. W. Kolb, Z. A. C. Krusberg and T. M. P. Tait, “Maverick darkmatter at colliders”, JHEP 1009, 037 (2010).

4. P. Harris, V. V. Khoze, M. Spannowsky and C. Williams, “Constraining Dark Sectors atColliders: Beyond the Effective Theory Approach”, Phys. Rev. D 91, 055009 (2015).

5. M. R. Buckley, D. Feld and D. Goncalves, “Scalar Simplified Models for Dark Matter”,Phys. Rev. D 91, 015017 (2015).

6. P. Harris, V. V. Khoze, M. Spannowsky and C. Williams, “Closing up on Dark Sectors atColliders: from 14 to 100 TeV”, Phys. Rev. D 93, no. 5, 054030 (2016).

7. J. Abdallah et al., “Simplified Models for Dark Matter Searches at the LHC”, Phys. DarkUniv. 9-10, 8 (2015).

8. ATLAS Collaboration, “The ATLAS Experiment at the CERN Large Hadron Collider”,JINST 3, S08003 (2008).

9. CMS Collaboration, “The CMS Experiment at the CERN LHC”, JINST 3, S08004 (2008).10. ATLAS Collaboration, “Search for new phenomena in final states with an energetic jet

and large missing transverse momentum in pp collisions at√s = 13 TeV using the ATLAS

detector”, Phys. Rev. D 94, no. 3, 032005 (2016).11. CMS Collaboration, “Search for dark matter produced with an energetic jet or a hadron-

ically decaying W or Z boson at√s = 13 TeV”, arXiv:1703.01651.

12. CMS Collaboration, “Search for dark matter in association with a top quark pair atsqrt(s)=13 TeV”,CMS-PAS-EXO-16-005, https://cds.cern.ch/record/2204933.

13. S. Baek, P. Ko, W. I. Park and E. Senaha, “Higgs Portal Vector Dark Matter : Revisited”,JHEP 1305, 036 (2013).

14. A. Djouadi, O. Lebedev, Y. Mambrini and J. Quevillon, “Implications of LHC searchesfor Higgs portal dark matter”, Phys. Lett. B 709, 65 (2012).

15. A. Djouadi, A. Falkowski, Y. Mambrini and J. Quevillon, “Direct Detection of Higgs PortalDark Matter at the LHC”, Eur. Phys. J. C 73, no. 6, 2455 (2013).

16. A. Beniwal, F. Rajec, C. Savage, P. Scott, C. Weniger, M. White and A. G. Williams,“Combined analysis of effective Higgs portal dark matter models”, Phys. Rev. D 93, no.11, 115016 (2016).

17. ATLAS Collaboration, “Constraints on new phenomena via Higgs boson couplings andinvisible decays with the ATLAS detector”, JHEP 1511, 206 (2015).

18. CMS Collaboration, “Searches for invisible decays of the Higgs boson in pp collisions atsqrt(s) = 7, 8, and 13 TeV”, JHEP 1702, 135 (2017).

19. E. Eichten, I. Hinchliffe, K. D. Lane and C. Quigg, “Super Collider Physics”, Rev. Mod.Phys. 56, 579 (1984).

20. B. A. Dobrescu and F. Yu, “Coupling-mass mapping of dijet peak searches”, Phys. Rev.D 88, no. 3, 035021 (2013).

21. ATLAS Collaboration, “Search for new phenomena in dijet events using 37 fb−1 of ppcollision data collected at

√s =13 TeV with the ATLAS detector”, arXiv:1703.09127.

22. CMS Collaboration, “Searches for dijet resonances in pp collisions at√s = 13 TeV

using data collected in 2016.”, CMS-PAS-EXO-16-056, https://cds.cern.ch/record/

2256873.23. ATLAS Collaboration, “Search for light dijet resonances with the ATLAS detector using

a Trigger-Level Analysis in LHC pp collisions at√s = 13 TeV”, ATLAS-CONF-2016-030,

http://cds.cern.ch/record/2161135.24. ATLAS Collaboration, “Search for new light resonances decaying to jet pairs and produced

in association with a photon or a jet in proton-proton collisions at√s = 13 TeV with the

ATLAS detector”, ATLAS-CONF-2016-070, http://cds.cern.ch/record/2206221.25. CMS Collaboration, “Search for light vector resonances decaying to quarks at 13 TeV”,

CMS-PAS-EXO-16-030, https://cds.cern.ch/record/2202715.26. A. Boveia et al., “Recommendations on presenting LHC searches for missing transverse

energy signals using simplified s-channel models of dark matter”, arXiv:1603.04156.27. ATLAS Collaboration, https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/

CombinedSummaryPlots/EXOTICS/index.html