quest for dark matter at colliders and...

Greg Landsberg Moriond 2019 Joint Gravity/QCD Session ❖ March 27, 2019

QUEST FOR DARK MATTER AT COLLIDERS AND BEYOND

I always wanted to be a gravitational physicist...

Gravity for Masses


Gravity for Masses

Masses for Gravity


Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Gravity vs QCD✦ It's easy to tell a Moriond Gravity person from a

Moriond QCD person on the slopes...

�3

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019



�3

Gradient decent vs. scattering!

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019



�3

Gradient decent vs. scattering!You mainly care about one force; we mostly care about the other three!

I'll try to convince you that they are at least as interesting as the one you study...

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Stairway to Heaven✦ Is there place for dark matter in this picture?

�4

Dark Matter from Cosmos

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Dunkle Materie✦ Fritz Zwicky was first to suggest the possibility of DM in the 1930-

ies by observing the Coma cluster of galaxies and comparing the velocities to the ones expected from the virial theorem:

✦ Found that galaxies move too fast to remain bound unless there is additional invisible mass (DM)

✦ N.B. Similar technique as used by Le Verrier in 1846 to predict the existence of Neptune from the distortion of the Uranus orbit

�6Graciela Gelmini-UCLA

‘Weighing’ galaxy clusters?

In 1933 Fritz Zwicky found thefirst indication of the DM.Used the “Virial Theorem”

Example: for planetsGM⊙m

r= mv2

|Gravitational Potential Energy|=2×Kinetic Energy

in the Coma Cluster: found its galaxies move too fast to remain boundedby the visible mass only)

Later: also gas in clusters moves too fast (is too hot - as measured inX-rays) to remain in it, unless there them DM.

PASI 2012, Buenos Aires, March 8, 2012 14

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Galaxy Rotation Curves ✦ In the 1970-ies, Vera Rubin has

measured rotation curves in a number of spiral galaxies, only to find flat distribution, indicative of hidden mass

�7

v ⇠ 1/pr

Graciela Gelmini-UCLA

Start by ‘weighing’ the Sun

GM⊙m

r2= m

v2

r⇒ v =

!GM⊙r

Rotation curve

⇒M⊙ = 1.9889× 1030 kg

PASI 2012, Buenos Aires, March 8, 2012 10

v ⇠ 1/pr

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Gravitational Lensing✦ One can also infer about

the distribution of dark matter mass using weak or strong lensing

�8

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Other Observations✦ Observation of the Bullet cluster by the Chandra X-ray

satellite showed that hot matter is left behind the clouds of dark matter (reconstructed via gravitational lensing)

�9

DM

DMGas Gas

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Abundance Measurements✦ The leading model of

cosmology is ΛCDM, favored by cosmic microwave background measurements, baryon acoustic oscillations, gravitational lensing, and other observations

✦ Latest CMB data dominated by the Planck satellite results, suggest the following components of the energy balance in the universe

�10

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

WIMP Miracle✦ With the development of SUSY in the 1970-ies, it was

quickly realized that a neutral, weakly interacting particle with the mass of ~100 GeV can naturally serve as a DM candidate

✦ This is known as "WIMP Miracle" and is one of three miracles of SUSY (the other two being a natural solution to the hierarchy problem and the unification of couplings)

✦ As the universe expands and cools down, WIMPs thermally decouple (i.e., can't annihilate any longer), giving fixed DM abundance:

�11

⌦DMh2 =0.2⇥ 10�9GeV�2

h�vi

h�vi ⇠ 10�9GeV�2

⌦DM ⇠ 0.2(weak crosssection)

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Relic Abundance✦ When DM is in thermal

equilibrium w/ matter, both the formation and annihilation are possible and equal:

✦ As the universe cools down, the formation is not possible, and only annihilation remains:

✦ As the universe further expands, the annihilation is no longer possible: DM freezes out

✦ This DM is non-relativistic (cold) with typical velocity ~300 km/s�12

��! qq

�� ! qq

24 April 14 Feng 9

FREEZE OUT

(1) Assume a new heavy particle X is initially in thermal equilibrium:

XX ↔⎯ qq

(2) Universe cools:

XX ⎯ qq

(3) Universe expands:

XX ⎯ qq

→←/

→←//Zeldovich et al. (1960s)

(1)

(2)

(3)

Increasingannihilationstrength↓

Feng, ARAA (2010)

⌦DMh2 = 0.12

Cold

Hot

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Nota Bene✦ While WIMPs are the most sought and theoretically preferred DM

candidates, they are not the only known possibility: ๏ Axions (very light pseudoscalar particles proposed to solve the "strong

CP" problem) could also serve as a DM candidate ๏ Dark sector (DM has complicated structure and is found in the dark

sector, which only communicates to SM particles weakly) ๏ Compact astronomical objects and primordial particles, such as

super-heavy monopoles, could also be DM candidates ✦ Finally, while a single DM source is economic, the nature may have

opted for a more complex solution, and there may be several sources of DM

✦ In what follows I will nevertheless focus on the WIMP DM, predicted not only by SUSY, but also by other new physics models

✦ Generally speaking, any model that predicts a weakly interacting massive particle, whose stability must be ensured by a certain symmetry (e.g., R-parity in SUSY), is a good DM candidate

�13

Slid

e

SIMPs/ELDERS

UltralightDarkMa5er

Muong-2

Small-ScaleStructure

Microlensing

DarkSectorCandidates,Anomalies,andSearchTechniques

HiddenSectorDarkMa5er

SmallExperiments:CoherentFieldSearches,DirectDetecIon,NuclearandAtomicPhysics,Accelerators

GeV TeVkeVeVneVfeVzeV MeVaeV peV µeV meV PeV 30M�

WIMPsQCDAxion

≈


≈

Beryllium-8

BlackHoles

HiddenThermalRelics/WIMPlessDM

AsymmetricDM

Freeze-InDM

Pre-InflaIonaryAxion

Post-InflaIonaryAxion

FIG. 1: Mass ranges for dark matter and mediator particle candidates, experimental anomalies,and search techniques described in this document. All mass ranges are merely representative; fordetails, see the text. The QCD axion mass upper bound is set by supernova constraints, andmay be significantly raised by astrophysical uncertainties. Axion-like dark matter may also havelower masses than depicted. Ultralight Dark Matter and Hidden Sector Dark Matter are broadframeworks. Mass ranges corresponding to various production mechanisms within each frameworkare shown and are discussed in Sec. II. The Beryllium-8, muon (g � 2), and small-scale structureanomalies are described in VII. The search techniques of Coherent Field Searches, Direct Detection,and Accelerators are described in Secs. V, IV, and VI, respectively, and Nuclear and Atomic Physicsand Microlensing searches are described in Sec. VII.

II. SCIENCE CASE FOR A PROGRAM OF SMALL EXPERIMENTS

Given the wide range of possible dark matter candidates, it is useful to focus the searchfor dark matter by putting it in the context of what is known about our cosmological historyand the interactions of the Standard Model, by posing questions like: What is the (particlephysics) origin of the dark matter particles’ mass? What is the (cosmological) origin ofthe abundance of dark matter seen today? How do dark matter particles interact, bothwith one another and with the constituents of familiar matter? And what other observableconsequences might we expect from this physics, in addition to the existence of dark matter?Might existing observations or theoretical puzzles be closely tied to the physics of darkmatter? These questions have many possible answers — indeed, this is one reason why

13

Gre

g La

ndsb

erg

- DM

at C

ollid

ers

and

Beyo

nd -

Mor

iond

201

9

Dark Matter Landscape�14

Battaglieri et al., arXiv:1707.04591

Slid

e

SIMPs/ELDERS

UltralightDarkMa5er

Muong-2

Small-ScaleStructure

Microlensing

DarkSectorCandidates,Anomalies,andSearchTechniques

HiddenSectorDarkMa5er

SmallExperiments:CoherentFieldSearches,DirectDetecIon,NuclearandAtomicPhysics,Accelerators


WIMPsQCDAxion

≈


≈

Beryllium-8

BlackHoles

HiddenThermalRelics/WIMPlessDM

AsymmetricDM

Freeze-InDM

Pre-InflaIonaryAxion

Post-InflaIonaryAxion

FIG. 1: Mass ranges for dark matter and mediator particle candidates, experimental anomalies,and search techniques described in this document. All mass ranges are merely representative; fordetails, see the text. The QCD axion mass upper bound is set by supernova constraints, andmay be significantly raised by astrophysical uncertainties. Axion-like dark matter may also havelower masses than depicted. Ultralight Dark Matter and Hidden Sector Dark Matter are broadframeworks. Mass ranges corresponding to various production mechanisms within each frameworkare shown and are discussed in Sec. II. The Beryllium-8, muon (g � 2), and small-scale structureanomalies are described in VII. The search techniques of Coherent Field Searches, Direct Detection,and Accelerators are described in Secs. V, IV, and VI, respectively, and Nuclear and Atomic Physicsand Microlensing searches are described in Sec. VII.

II. SCIENCE CASE FOR A PROGRAM OF SMALL EXPERIMENTS

Given the wide range of possible dark matter candidates, it is useful to focus the searchfor dark matter by putting it in the context of what is known about our cosmological historyand the interactions of the Standard Model, by posing questions like: What is the (particlephysics) origin of the dark matter particles’ mass? What is the (cosmological) origin ofthe abundance of dark matter seen today? How do dark matter particles interact, bothwith one another and with the constituents of familiar matter? And what other observableconsequences might we expect from this physics, in addition to the existence of dark matter?Might existing observations or theoretical puzzles be closely tied to the physics of darkmatter? These questions have many possible answers — indeed, this is one reason why

13

Gre

g La

ndsb

erg

- DM

at C

ollid

ers

and

Beyo

nd -

Mor

iond

201

9

Dark Matter Landscape�14

Battaglieri et al., arXiv:1707.04591

{ This talk

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Known Unknowns✦ While the true origin of DM is unknown, several

things about DM are well understood ✦ Assuming that DM has particle origin we know that:

๏ It has to be a neutral particle ๏ It's unlikely that it carries color (strong interactions) ๏ It must be stable on a cosmological timescale ๏ It must have the right abundance, which sets

constraints on its decay channels, couplings, and mass ✤ For example, ordinary neutrinos can't be a sole source of

DM, despite having mass ✤ In order to get the right abundance, DM must be able to

interact with the SM particles, which is achieved via a "mediator" particle coupled to both SM species and DM�15

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Searching for Dark Matter✦ A clever way of searching for DM is to look for a recoil of DM from a

heavy nucleus (scattering again!) ✦ These types of experiments use very cold targets and are conducted

deep underground to shield cosmic rays ✦ The sensitivity drops dramatically for a light DM, as there is not much

of a recoil! ✦ Latest generation of experiments uses liquid

uses Xe as the target (LUX, Xenon1T, PandaX, LZ)

๏ As the dominant Xe isotope has spin 0, these experiments are sensitive mainly to spin-independent (SI) scattering and benefit from a resonantenhancement ~A2

๏ Other targets are used to probespin-dependent (SD) scattering (C4F10, CF3Br, etc)

�16

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Direct Detection Experiments�17

SuperCDMS

CoGeNT

Slid

e

FIG. 8: Left: Constraints and projections (90% c.l.) for the DM-nucleon scattering crosssection. Thick gray lines are current world-leading constraints [108, 116, 129, 130]. Projections areshown with solid/dashed/dotted lines indicating a short/medium/long timescale, respectively, withthe same meaning as in Fig. 6. Blue lines denote the DoE G2 experiment projections. Yellow regiondenotes the WIMP-discovery limit from [131] extended to lower masses for He-based experiments.Right: As in left plot, but focused on the 100 MeV to 10 GeV DM mass range.

� � �� -��

��-��

��-��

��-��

��-��

��-��

��-��

��-�

��-�

��-�

��-�

��

��

�� [��/��]

��-��[�� ]

��-��[��]

��

��

��

��-��

��-��

��-��

��

��-��

�

�� -��

ν��

ν��

FIG. 9: Constraints from direct-detection experiments (solid lines), colliders and indirect detection(labelled, dashed), and projections for new experiments (labelled, dashed/dotted lines) for thespin-dependent scattering cross section for protons or neutrons o↵ nuclei. Constraintsare shown from PICO-60 [116], LUX [132], PICO-2L [133], PICO-60 CF3I [134], and IceCube [135].Projections from PICO (proton) and LZ (neutron) are also shown [115]. The expected backgroundfrom atmospheric, supernova and solar neutrinos in both xenon and C3F8 is shown by the shadedregions [131].

52

Gre

g La

ndsb

erg

- DM

at C

ollid

ers

and

Beyo

nd -

Mor

iond

201

9

Direct Detection Results✦ State of the art results dominated by the latest Xenon 1T result,

with LUX & PandaX-II being close second/third (SI) and PICO (SD) ✦ Large neutrino detectors (IceCube, SuperK) are also sensitive to

DM via annihilation inside the Sun or (very slow) decay ✦ N.B.: SD limits are ~6 orders of magnitude weaker than SI ones

�18

� �� -��

��-��

��-��

��-��

��-��

��-��

��-��

��-��

��-��

��-�

��-�

��-�

��-�

�� [��/��]

��-��σ �

�[�� ]

��-��σ �

�[��]

��

��-��

��

�

��-��

��-��

��

��-��

��

��

��-�

��-�� -�

��

��-��

��

��

��-��

��

��

��

��

SD LimitsSI Limits

Neutrino floor

Neutrino floor

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

1 10 100 1000 10410!5010!4910!4810!4710!4610!4510!4410!4310!4210!4110!4010!39

10!1410!1310!1210!1110!1010!910!810!710!610!510!410!3

WIMP Mass !GeV"c2#

WIMP!nucleoncrosssection!cm2 #

WIMP!nucleoncrosssection!pb#

8BNeutrinos

Atmospheric and DSNB Neutrinos

CDMS II Ge (2009)

Xenon100 (2012)

CRESST

CoGeNT(2012)

CDMS Si(2013)

EDELWEISS (2011)

DAMA SIMPLE (2012)

ZEPLIN-III (2012)COUPP (2012)

SuperCDMS Soudan Low ThresholdXENON 10 S2 (2013)

CDMS-II Ge Low Threshold (2011)

SuperCDMS Soudan

Xenon1T

LZ

LUX (2013)

DarkSide G2

DarkSide 50

DEAP3600

PICO250-CF3I

PICO250-C3F8

7BeNeutrinos

NEUTRINO C OHER ENT SCATTERING

NEUTRINO COHERENT SCATTERING

CDMSlite

(2013)

SuperCDMS SNOLABLUX 300-day

SuperCDMS SNOLAB

(Green&ovals)&Asymmetric&DM&&(Violet&oval)&Magne7c&DM&(Blue&oval)&Extra&dimensions&&(Red&circle)&SUSY&MSSM&&&&&&MSSM:&Pure&Higgsino&&&&&&&MSSM:&A&funnel&&&&&&MSSM:&BinoEstop&coannihila7on&&&&&&MSSM:&BinoEsquark&coannihila7on&&

Future of Direct Detection✦ Next generation of DD experiments will reach the "ν floor"

�19

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Indirect Detection✦ Look for DM annihilation into SM particles,

further decaying to lighter species (bb, ττ, ...) ๏ Every time a π0 is created, it decays into

two photons - look for an excess of γ rays in the sky

✦ Number of ground-based (HESS, Magic, Veritas) and space-based (Fermi-LAT) instruments

๏ Look for γ ray emission from the center of galaxy, cluster of galaxies, dwarf galaxies, etc.

๏ Certain hints (galactic center, dwarfgalaxies) have been reported, but no definitive observation (yet)

๏ Could also look for positron emission (AMS, Pamela)

๏ Results are reported in terms ofvelocity-averaged annihilationcross section

�20an improvement in analysis methods and indeed the recently presented PASS8 results arealready excluding WIMPS below the thermal cross-section up to 100 GeV. The next majorstep forward will be the Cherenkov Telescope Array, CTA, an array of about 80 telescopeswith a factor 10 improved sensitivity and an extended energy range from about 10 GeV to40 TeV [33]. A prediction more or less based on scaling performances of IACTs arrived atthe conclusion that the thermal WIMP cross-section could be probed at masses between 100GeV and 10 TeV [34, 35] . For CTA, systematics on the cosmic ray background estimate(ratio between o↵ and on region acceptance) and irreducible di↵use emission can becomecritical in the limit of large event statistics. An attempt to account for both has beenpresented in [36] with somewhat dimmed expectations. Also pair conversion telescopesare planned, foremost GAMMA-400 [37], the DArk Matter Particle Explorer (DAMPE)[38] and the High Energy cosmic Radiation Detector (HERD) [39]. Both have in commonvery deep calorimeters, enabling energy resolution of the order of 1 %. The latter has anenvisaged e↵ective area of about twice the Fermi-LAT, and main progress can be expectedin the area of spectral feature detection. e.g. [20].

Figure 2: Left panel: current most relevant constraints on annihilation cross-section versusmass for quark-channels. Fermi-LAT dwarfs [26], HESS dwarf galaxies [31], HESS halo [28],VERITAS Segue1 [30], MAGIC Segue1 [29]. Right panel: Constraints in the next decade:VERITAS 1000h DM programme [32], HESSII (my guess), Fermi-LAT dwarf and CTA [35]and CTA including systematic uncertainties [36]. Model points, taken from [47], (as inother figures) represent MSSM-7 with red colour: consistent with cosmological dark matterdensity, blue colour: annihilation cross-section allowed to be smaller.

3 Charged cosmic-ray probe: a clear signal – but of what?

The main signature for DM in charged cosmic rays are in the anti-proton and positronchannel. Anti-particles are very rarely produced in secondary processes and even a smalladdition of anti-particles produced in WIMP annihilation could give rise to a detectablesignal, revealing itself as a rise in the positron to electron or antiproton to proton ratio,conveniently the ratio is taken to cancel acceptance systematics which should a↵ect particles

5

Thermal relic cross section

Mono-mania, orLHC as a Dark Matter Factory

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Make It, Break It, or Shake It!✦ There are three main approaches to detect DM:

๏ DM-nucleon scattering (direct detection) ๏ Annihilation (Indirect detection) ๏ Pair production at colliders

✦ All three processes are nothing but topological permutations of one and the same Feynman diagram: ๏ But: how to trigger on a pair of

DM particles at colliders? ๏ Initial-state radiation (ISR: g, γ,

W/Z, H, …) to rescue! ✦ Original idea - to use the ISR - appeared

nine years ago: ๏ Beltran, Hooper, Kolb, Krusberg, and Tait,

“Maverick Dark Matter at Colliders” arXiv:1002.4137 (306 citations)

�22

PRODUCTION OF DARK MATTER AT CMS

• Search%for%evidence%of%pair[produc=on%of%Dark%MaAer%par=cles%(χ)

• Dark%MaAer%produc=on%gives%missing%transverse%energy%(MET)

• Photons%(or%jets%from%a%gluon)%can%be%radiated%from%quarks,%giving%monophoton%(or%monojet)%plus%MET

3

4

q

q

�

�

Figure 1: Dark matter production in association with a single jet in a hadron collider.

3.1. Comparing Various Mono-Jet Analyses

Dark matter pair production through a diagram like figure 1 is one of the leading channelsfor dark matter searches at hadron colliders [3, 4]. The signal would manifest itself as an excessof jets plus missing energy (j + /ET ) events over the Standard Model background, which consistsmainly of (Z � ⇥⇥)+ j and (W � ⌅inv⇥)+ j final states. In the latter case the charged lepton ⌅ islost, as indicated by the superscript “inv”. Experimental studies of j + /ET final states have beenperformed by CDF [22], CMS [23] and ATLAS [24, 25], mostly in the context of Extra Dimensions.

Our analysis will, for the most part, be based on the ATLAS search [25] which looked for mono-jets in 1 fb�1 of data, although we will also compare to the earlier CMS analysis [23], which used36 pb�1 of integrated luminosity. The ATLAS search contains three separate analyses based onsuccessively harder pT cuts, the major selection criteria from each analysis that we apply in ouranalysis are given below.3

LowPT Selection requires /ET > 120 GeV, one jet with pT (j1) > 120 GeV, |�(j1)| < 2, and eventsare vetoed if they contain a second jet with pT (j2) > 30 GeV and |�(j2)| < 4.5.

HighPT Selection requires /ET > 220 GeV, one jet with pT (j1) > 250 GeV, |�(j1)| < 2, and eventsare vetoed if there is a second jet with |�(j2)| < 4.5 and with either pT (j2) > 60 GeV or�⇤(j2, /ET ) < 0.5. Any further jets with |�(j2)| < 4.5 must have pT (j3) < 30 GeV.

veryHighPT Selection requires /ET > 300 GeV, one jet with pT (j1) > 350 GeV, |�(j1)| < 2, andevents are vetoed if there is a second jet with |�(j2)| < 4.5 and with either pT (j2) > 60 GeVor �⇤(j2, /ET ) < 0.5. Any further jets with |�(j2)| < 4.5 must have pT (j3) < 30 GeV.

In all cases events are vetoed if they contain any hard leptons, defined for electrons as |�(e)| < 2.47and pT (e) > 20 GeV and for muons as |�(µ)| < 2.4 and pT (µ) > 10 GeV.

The cuts used by CMS are similar to those of the LowPT ATLAS analysis. Mono-jet eventsare selected by requiring /ET > 150 GeV and one jet with pT (j1) > 110 GeV and pseudo-rapidity|�(j1)| < 2.4. A second jet with pT (j2) > 30 GeV is allowed if the azimuthal angle it forms withthe leading jet is �⇤(j1, j2) < 2.0 radians. Events with more than two jets with pT > 30 GeV arevetoed, as are events containing charged leptons with pT > 10 GeV. The number of expected andobserved events in the various searches is shown in table I.

3 Both ATLAS and CMS impose additional isolation cuts, which we do not mimic in our analysis for simplicity andsince they would not have a large impact on our results.

4

q

q

�

�

Figure 1: Dark matter production in association with a single jet in a hadron collider.

3.1. Comparing Various Mono-Jet Analyses

Dark matter pair production through a diagram like figure 1 is one of the leading channelsfor dark matter searches at hadron colliders [3, 4]. The signal would manifest itself as an excessof jets plus missing energy (j + /ET ) events over the Standard Model background, which consistsmainly of (Z � ⇥⇥)+ j and (W � ⌅inv⇥)+ j final states. In the latter case the charged lepton ⌅ islost, as indicated by the superscript “inv”. Experimental studies of j + /ET final states have beenperformed by CDF [22], CMS [23] and ATLAS [24, 25], mostly in the context of Extra Dimensions.

Our analysis will, for the most part, be based on the ATLAS search [25] which looked for mono-jets in 1 fb�1 of data, although we will also compare to the earlier CMS analysis [23], which used36 pb�1 of integrated luminosity. The ATLAS search contains three separate analyses based onsuccessively harder pT cuts, the major selection criteria from each analysis that we apply in ouranalysis are given below.3

LowPT Selection requires /ET > 120 GeV, one jet with pT (j1) > 120 GeV, |�(j1)| < 2, and eventsare vetoed if they contain a second jet with pT (j2) > 30 GeV and |�(j2)| < 4.5.

HighPT Selection requires /ET > 220 GeV, one jet with pT (j1) > 250 GeV, |�(j1)| < 2, and eventsare vetoed if there is a second jet with |�(j2)| < 4.5 and with either pT (j2) > 60 GeV or�⇤(j2, /ET ) < 0.5. Any further jets with |�(j2)| < 4.5 must have pT (j3) < 30 GeV.

veryHighPT Selection requires /ET > 300 GeV, one jet with pT (j1) > 350 GeV, |�(j1)| < 2, andevents are vetoed if there is a second jet with |�(j2)| < 4.5 and with either pT (j2) > 60 GeVor �⇤(j2, /ET ) < 0.5. Any further jets with |�(j2)| < 4.5 must have pT (j3) < 30 GeV.

In all cases events are vetoed if they contain any hard leptons, defined for electrons as |�(e)| < 2.47and pT (e) > 20 GeV and for muons as |�(µ)| < 2.4 and pT (µ) > 10 GeV.

The cuts used by CMS are similar to those of the LowPT ATLAS analysis. Mono-jet eventsare selected by requiring /ET > 150 GeV and one jet with pT (j1) > 110 GeV and pseudo-rapidity|�(j1)| < 2.4. A second jet with pT (j2) > 30 GeV is allowed if the azimuthal angle it forms withthe leading jet is �⇤(j1, j2) < 2.0 radians. Events with more than two jets with pT > 30 GeV arevetoed, as are events containing charged leptons with pT > 10 GeV. The number of expected andobserved events in the various searches is shown in table I.

3 Both ATLAS and CMS impose additional isolation cuts, which we do not mimic in our analysis for simplicity andsince they would not have a large impact on our results.

Direct Detection (t-channel) Collider Searches (s-channel)

Monophoton + MET Monojet + MET

χ

χ

q

q

M

Dire

ct

dete

ctio

n

Freeze-out,indirect detection

Collider production

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Mono-Mania✦ More phenomenological follow-ups sparked interest from the collider

community: ๏ Goodman et al, arXiv:1005.1286, 366 citations ๏ Bai, Fox, Harnik, arXiv:1005.3797, 371 citations ๏ Goodman et al, arXiv:1008.1783, 595 citations ๏ Fox, Harnik, Kopp, Tsai, arXiv:1103.0240, 245 citations - LEP reinterpretation ๏ Fox, Harnik, Kopp, Tsai, arXiv:1109.4398, 435 citations - LHC case

✦ The first experimental search came from the CDF collaboration: ๏ Mono-top, arXiv:1202.5653, 31 citations ๏ Monojets, arXiv:1203.0742, 87 citations

✦ Was quickly followed and superseded by ATLAS and CMS: ๏ CMS, Monophotons, arXiv:1204.0821, 192 citations ๏ CMS, Monojets, arXiv:1206.5663, 276 citations ๏ ATLAS, Monophotons, arXiv:1209.4625, 172 citations ๏ ATLAS, Monojets, arXiv:1210.4491, 253 citations

✦ ...and then the hell broke loose with dozens of other mono-X searches

�23

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Effective Field Theory✦ Effective Field Theory (EFT) is a convenient

simplified description of a complicated process via effective operator (and effective coupling)

✦ Most well-known example: Fermi theory of muon decay

�24

Muon Decay µ−→ e

−+ νµ + νe

Muon decay has been studied for decades and continues to be studied today as a

means for testing the Standard Model and search for evidence of New Physics

W

1 : µ

3 : νµ

2 : νe

4 : e

• Again, working in the limit of q2 ≪ M2W , the amplitude is

M =g2

w

8M2W

ˆ

u3γµ(1 − γ5)u1

˜ ˆ

u4γµ(1 − γ5)v2

˜

• This is identical to the amplitude in the previous example, except for u2 → v2,

but both either spinors give us /p2 in the trace (sincemν = 0)

Physics 506A 14 - Weak interactions Page 11

Muon Decay II

• We have the identical spin-averaged squared matrix element as e ν scattering

D

|M|2E

=2g4

w

M4W

(p1 · p2)(p3 · p4)

• Since the kinematics of µ decay are different from those of e νµ scattering, we

will need to start our work here.

• In the muon rest frame, (p1 · p2) = mµE2, and

(p3 · p4) =ˆ

(p3 + p4)2 − p23 − p2

4

˜

/2

=ˆ

(p1 − p2)2 − 0 − 0˜

/2

=ˆ

p21 + p2

2 − 2p1 · p2

˜

/2

= mµ(mµ − 2E2)/2

• The spin-averaged squared matrix element simplifies toD

|M|2E

=g4

w

M4W

m2µE2(mµ − 2E2)


Muon Decay Rate

• Integrating over the electron energy, we obtain the muon decay rate:

Γµ =

„

gw

MW

«4 m2µ

2(4π)3

Z mµ/2

0

E2

„

1 −4E

3mµ

«

dE

=1

6144π3

„

gw

MW

«4

m5µ

• In the limit of q2 ≪ M2W , our results always depend on the ratio of gw and

MW , and not the two constants separately.

Hence we can define the Fermi coupling constantGF , by GF =√

2g2w

8M2

W

• This allows us to write the muon lifetime as

τµ =192π3

G2F m5

µ


Muon Decay Rate


Γµ =

„

gw

MW

«4 m2µ

2(4π)3

Z mµ/2

0

E2

„

1 −4E

3mµ

«

dE

=1

6144π3

„

gw

MW

«4

m5µ




2g2w

8M2

W


τµ =192π3

G2F m5

µ


Muon Decay Rate


Γµ =

„

gw

MW

«4 m2µ

2(4π)3

Z mµ/2

0

E2

„

1 −4E

3mµ

«

dE

=1

6144π3

„

gw

MW

«4

m5µ




2g2w

8M2

W


τµ =192π3

G2F m5

µ


GF

✦ But, just like applying Fermi theory to describe W production at the LHC would fail completely, one has to be very careful about applicability of the effective theory for DM searches

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Dark Matter Interactions✦ Early DM searches: EFT based

๏ Since then understood the fundamental limitations of EFT and largely moved to simplified models of DM

✦ Moving away from EFT allows for a more fair LHC vs. DD/ID experiments comparison and emphasizes the complementarity of these three approaches

�25

χ

χ

q

q

M

Dire

ct

dete

ctio

n

Freeze-out,indirect detection

Collider production

gχ gqmχ

Fundamentally 4D problem!

Monojets: the Classics

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Monojet Searches✦ Monojet analysis is a classical search for a number

of new physics phenomena ๏ Smoking gun signature for supersymmetry, large extra

dimensions, dark matter production, ... ๏ Was pursued since early 1980s

✦ The signature is deceptively simple, yet it's not ๏ Backgrounds from instrumental effects ๏ Irreducible Z(νν)+jet background ๏ Reducible backgrounds from jet mismeasurements and

W+jets with a lost lepton ✦ Number of techniques have been developed since

the first search by UA1; will show the state-of-the-art results from CMS

�27

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Monojet Searches✦ We've come a long way since Carlo Rubbia's first

attempt!

�28

[PL, 139B, 115 (1984)]

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Monojet Searches✦ We've come a long way since Carlo Rubbia's first

attempt!

�28

[PL, 139B, 115 (1984)]

W → τν → hadrons + neutrinos Z(νν)+jet

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

18 6 Results and interpretation

Table 5: Expected event yields in each pmissT bin for various background processes in the mono-

V signal region. The background yields and the corresponding uncertainties are obtained afterperforming a combined fit to data in all the control samples, but excluding data in the signalregion. The other backgrounds include QCD multijet and g+jets processes. The expectedsignal contribution for a 2 TeV axial-vector mediator decaying to 1 GeV DM particles and theobserved event yields in the mono-V signal region are also reported.

pmissT (GeV) Signal Z(nn)+jets W(nn)+jets Top quark Diboson Other Total bkg. Data250-300 11.7± 0.6 5300± 170 3390± 120 553± 54 396± 69 128± 25 9770± 290 9929300-350 15.7± 0.7 3720± 98 1823± 53 257± 27 261± 46 79.8± 13 6140± 140 6057350-400 11.8± 0.6 1911± 59 808± 28 101± 12 134± 25 25.0± 4.8 2982± 79 3041400-500 15.8± 0.7 1468± 45 521± 15 48.8± 5.7 107± 20 20.0± 3.6 2165± 55 2131500-600 8.59± 0.56 388± 18 103.0± 5.1 10.7± 1.9 33.8± 7.0 1.76± 0.53 537± 23 521600-750 7.04± 0.47 151.0± 9.9 33.4± 2.3 1.9± 1.1 20.2± 4.5 1.05± 0.25 208± 11 225>750 4.48± 0.40 37.7± 3.7 7.09± 0.69 0.28± 0.25 10.2± 2.3 0.06± 0.03 55.3± 4.6 61

Even

ts /

GeV

2−10

1−10

1

10

210

310

410

510

610 (13 TeV)-135.9 fb

CMSmonojet

Data

inv.→H(125)

= 2.0 TeVmed

Axial-vector, m

)+jetsννZ()+jetsνW(l

WW/WZ/ZZ

Top quark+jetsγ(ll), γZ/

QCD

Dat

a / P

red.

0.80.9

11.11.2

[GeV]missT

p400 600 800 1000 1200 1400

Unc

.(D

ata-

Pred

.)

2−02

Even

ts /

GeV

2−10

1−10

1

10

210

310

410

(13 TeV)-135.9 fb

CMSmono-V

Data

inv.→H(125)

= 2.0 TeVmed

Axial-vector, m

)+jetsννZ()+jetsνW(l

WW/WZ/ZZ

Top quark+jetsγ(ll), γZ/

QCD

Dat

a / P

red.

0.8

1

1.2

[GeV]missT

p300 400 500 600 700 800 900 1000

Unc

.(D

ata-

Pred

.)

2−02

Figure 9: Observed pmissT distribution in the monojet (left) and mono-V (right) signal regions

compared with the post-fit background expectations for various SM processes. The last bin in-cludes all events with p

missT > 1250 (750) GeV for the monojet (mono-V) category. The expected

background distributions are evaluated after performing a combined fit to the data in all thecontrol samples, as well as in the signal region. The fit is performed assuming the absence ofany signal. Expected signal distributions for the 125 GeV Higgs boson decaying exclusivelyto invisible particles, and a 2 TeV axial-vector mediator decaying to 1 GeV DM particles, areoverlaid. The description of the lower panels is the same as in Fig. 5.

The results for vector, axial-vector, and pseudoscalar mediators are compared to constraintsfrom the observed cosmological relic density of DM as determined from measurements ofthe cosmic microwave background by the Planck satellite experiment [97]. The expected DM

Fast-Forward 30 Years✦ State-of-the-art analysis, which employs multiple control regions

and the latest theory calculations of NLO EW and QCD corrections

�29

CM

S ar

Xiv:

1712

.023

452016

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

A Monojet Event�30

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

DM Interpretation✦ Present the limits in terms of constraints on the mediator

vs. DM particle masses for fixed value of couplings ๏ Convention: gq = 0.25; gDM = 1

�31

6.1 Dark matter interpretation 19

Figure 10: Exclusion limits at 95%CL on µ = s/sth in the mmed–mDM plane assuming vector(left) and axial-vector (right) mediators. The solid (dotted) red (black) line shows the contourfor the observed (expected) exclusion. The solid contours around the observed limit and thedashed contours around the expected limit represent one standard deviation due to theoreticaluncertainties in the signal cross section and the combination of the statistical and experimentalsystematic uncertainties, respectively. Constraints from the Planck satellite experiment [97] areshown as dark blue contours; in the shaded area DM is overabundant.

[GeV]medm0 100 200 300 400 500 600

theo

ryσ/

σ95

% C

L up

per l

imit

on

1

2

3

4

5

6 (13 TeV)-135.9 fb

CMS

Median expected

68% expected

95% expected

Observed

= 1µ

= 1DM

= 1, gq

= 1 GeV gDM

Scalar med, Dirac DM, m

Figure 11: Expected (dotted black line) and observed (solid black line) 95%CL upper limitson the signal strength µ = s/sth as a function of the mediator mass for the scalar mediators(left) for mDM = 1 GeV. The horizontal red line denotes µ = 1. Exclusion limits at 95%CLon µ = s/sth in the mmed–mDM plane assuming pseudoscalar mediators (right). The solid(dashed) red (back) line shows the contours for the observed (expected) exclusion. Constraintsfrom the Planck satellite experiment [97] are shown with the dark blue contours; in the shadedarea DM is overabundant.

abundance is estimated, separately for each model, using the thermal freeze-out mechanismimplemented in the MADDM [98] framework and compared to the observed cold DM densityWch

2 = 0.12 [99], where Wc is the DM relic abundance and h is the Hubble constant.

2016

2016

CMS arXiv:1712.02345Vector mediators Axial-vector mediators

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

DM Interpretation✦ (Pseudo)Scalar mediators are harder, because of

the assumed Yukawa couplings (don't couple to light quarks; need top quark loop in production)

�32


Figure 10: Exclusion limits at 95%CL on µ = s/sth in the mmed–mDM plane assuming vector(left) and axial-vector (right) mediators. The solid (dotted) red (black) line shows the contourfor the observed (expected) exclusion. The solid contours around the observed limit and thedashed contours around the expected limit represent one standard deviation due to theoreticaluncertainties in the signal cross section and the combination of the statistical and experimentalsystematic uncertainties, respectively. Constraints from the Planck satellite experiment [97] areshown as dark blue contours; in the shaded area DM is overabundant.

[GeV]medm0 100 200 300 400 500 600

theo

ryσ/

σ95

% C

L up

per l

imit

on

1

2

3

4

5

6 (13 TeV)-135.9 fb

CMS

Median expected

68% expected

95% expected

Observed

= 1µ

= 1DM

= 1, gq

= 1 GeV gDM

Scalar med, Dirac DM, m

Figure 11: Expected (dotted black line) and observed (solid black line) 95%CL upper limitson the signal strength µ = s/sth as a function of the mediator mass for the scalar mediators(left) for mDM = 1 GeV. The horizontal red line denotes µ = 1. Exclusion limits at 95%CLon µ = s/sth in the mmed–mDM plane assuming pseudoscalar mediators (right). The solid(dashed) red (back) line shows the contours for the observed (expected) exclusion. Constraintsfrom the Planck satellite experiment [97] are shown with the dark blue contours; in the shadedarea DM is overabundant.

abundance is estimated, separately for each model, using the thermal freeze-out mechanismimplemented in the MADDM [98] framework and compared to the observed cold DM densityWch

2 = 0.12 [99], where Wc is the DM relic abundance and h is the Hubble constant.

2016

2016C

MS

arXi

v:17

12.0

2345

Scalar mediators Pseudoscalar mediators

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2


[GeV]DMm1 10 210 310

]2 [c

mD

M-n

ucle

onSI

σ

47−10

45−10

43−10

41−10

39−10

37−10

35−10

33−10

31−10

29−10

27−10

25−10

CMS exp. 90% CL CMS obs. 90% CLLUX CDMSLiteXenon-1T CRESST-IIPandaX-II

(13 TeV)-135.9 fb

CMS = 1

DM = 0.25, g

qVector med, Dirac DM, g

[GeV]DMm1 10 210 310

]2 [c

mDM

-pro

ton

SDσ

46−10

44−10

42−10

40−10

38−10

36−10

34−10

32−10

30−10

28−10

CMS exp. 90% CL CMS obs. 90% CL

PICO-60 Picasso

bIceCube b tIceCube t

bSuper-K b

(13 TeV)-135.9 fb

CMS = 1

DM = 0.25, g

qAxial med, Dirac DM, g

Figure 13: Exclusion limits at 90%CL in the mDM vs. sSI/SD plane for vector (left) and axial-vector (right) mediator models. The solid red (dotted black) line shows the contour for theobserved (expected) exclusion in this search. Limits from CDMSLite [102], LUX [103], XENON-1T [104], PANDAX-II [105], and CRESST-II [106] are shown for the vector mediator. Limitsfrom Picasso [107], PICO-60 [108], IceCube [109], and Super-Kamiokande [110] are shown forthe axial-vector mediator.

[GeV]DMm10 210

/s)

3 (c

m〉

vσ〈

31−10

30−10

29−10

28−10

27−10

26−10

25−10

CMS exp. 90% CL

CMS obs. 90% CL

FermiLAT

(13 TeV)-135.9 fb

CMSPseudoscalar med, Dirac DM

= 1DM

= 1, gq

g

Figure 14: For the pseudoscalar mediator, limits are compared to the the velocity averaged DMannihilation cross section upper limits from Fermi-LAT [101]. There are no comparable limitsfrom direct detection experiments, as the scattering cross section between DM particles and SMquarks is suppressed at nonrelativistic velocities for a pseudoscalar mediator [111, 112].

Complementarity w/ DD✦ Collider experiments win over direct detection in SD case

(axial-vector mediator) up to mDM ~ 500 GeV and in the SI case (vector mediator) for very light DM (mDM < 5 GeV)

�33

2016 2016

CM

S ar

Xiv:

1712

.023

45

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019


[GeV]DMm1 10 210 310

]2 [c

mD

M-n

ucle

onSI

σ

47−10

45−10

43−10

41−10

39−10

37−10

35−10

33−10

31−10

29−10

27−10

25−10

CMS exp. 90% CL CMS obs. 90% CLLUX CDMSLiteXenon-1T CRESST-IIPandaX-II

(13 TeV)-135.9 fb

CMS = 1

DM = 0.25, g

qVector med, Dirac DM, g

[GeV]DMm1 10 210 310

]2 [c

mDM

-pro

ton

SDσ

46−10

44−10

42−10

40−10

38−10

36−10

34−10

32−10

30−10

28−10

CMS exp. 90% CL CMS obs. 90% CL

PICO-60 Picasso

bIceCube b tIceCube t

bSuper-K b

(13 TeV)-135.9 fb

CMS = 1

DM = 0.25, g

qAxial med, Dirac DM, g

Figure 13: Exclusion limits at 90%CL in the mDM vs. sSI/SD plane for vector (left) and axial-vector (right) mediator models. The solid red (dotted black) line shows the contour for theobserved (expected) exclusion in this search. Limits from CDMSLite [102], LUX [103], XENON-1T [104], PANDAX-II [105], and CRESST-II [106] are shown for the vector mediator. Limitsfrom Picasso [107], PICO-60 [108], IceCube [109], and Super-Kamiokande [110] are shown forthe axial-vector mediator.

[GeV]DMm10 210

/s)

3 (c

m〉

vσ〈

31−10

30−10

29−10

28−10

27−10

26−10

25−10

CMS exp. 90% CL

CMS obs. 90% CL

FermiLAT

(13 TeV)-135.9 fb

CMSPseudoscalar med, Dirac DM

= 1DM

= 1, gq

g

Figure 14: For the pseudoscalar mediator, limits are compared to the the velocity averaged DMannihilation cross section upper limits from Fermi-LAT [101]. There are no comparable limitsfrom direct detection experiments, as the scattering cross section between DM particles and SMquarks is suppressed at nonrelativistic velocities for a pseudoscalar mediator [111, 112].

Complementarity w/ ID✦ For a pseudoscalar mediator, the nucleon scattering cross

section is velocity suppressed due to the because of the following factor in the matrix element:

✦ Given v ~ 10-3, the sensitivity of DD experiments vanishes ✦ The collider results can be compared w/ ID experiments and

are more stringent for DM masses below ~150 GeV

�34

2016

CM

S ar

Xiv:

1712

.023

45

M ⇠ (~vf � ~vi) · ~�n<latexit sha1_base64="M8s3AZqoltlL5dc1N6tB/7/vPhU=">AAACHXicdVBPS8MwHE397/w39eglOAQ9WNpurXoTvXgRJjgV1jLSLJ1hSVuSdDDKvogXv4oXD4p48CJ+G9Otgoo+CLy893skvxemjEplWR/G1PTM7Nz8wmJlaXllda26vnElk0xg0sIJS8RNiCRhNCYtRRUjN6kgiIeMXIf908K/HhAhaRJfqmFKAo56MY0oRkpLnWoj9zFi8HwEfUk53PUHBMNBJ4L7sKR0z8fdRBU3PdLjqKNjNcu0G55z5EHLdF2vbhfEcV3H86BtWmPUQIlmp/rmdxOccRIrzJCUbdtKVZAjoShmZFTxM0lShPuoR9qaxogTGeTj7UZwRytdGCVCn1jBsfo9kSMu5ZCHepIjdSt/e4X4l9fOVHQY5DROM0ViPHkoyhhUCSyqgl0qCFZsqAnCguq/QnyLBMJKF1rRJXxtCv8nV45p103nolE7PinrWABbYBvsAhscgGNwBpqgBTC4Aw/gCTwb98aj8WK8TkanjDKzCX7AeP8EHLKhYg==</latexit>

Beyond Monojets

Slid

e

21

renormalizable models predicting a vectorial DM candidate require an extended dark Higgssector, which may lead to modifications of kinematic distributions assumed for the invisibleHiggs boson signal, such interpretation is not provided in the context of this Letter. Figure 10(right) shows the 90% CL upper limits on the spin-independent DM-nucleon scattering crosssection as a function of mc, for both the scalar and the fermion DM scenarios. These limits arecomputed at 90% CL so that they can be compared with those from direct detection experimentssuch as XENON1T [79], LUX [80], PandaX-II [81], CDMSlite [82], CRESST-II [83], and CDEX-10 [84] which provide the strongest constraints in the mc range probed by this search. In thecontext of Higgs-portal models, the result presented in this Letter provides the most stringentlimits for mc smaller than 18 (7) GeV, assuming a fermion (scalar) DM candidate.

Vκ0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4

Fκ

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5 (13 TeV)-1 (8 TeV) + 38.2 fb-1 (7 TeV) + 19.7 fb-14.9 fb

CMS

LHC best fit68% CL95% CLSM production

SMσ

inv)

/→

B(H

× σ

95%

CL

uppe

r lim

it on

[GeV]χm1 10 210 310

]2 [c

mDM

-nuc

leon

SIσ

47−10

46−10

45−10

44−10

43−10

42−10

41−10

40−10

39−10

38−10

37−10

90% CL limits inv) < 0.16→B(H

Fermion DMScalar DM

Direct detectionXENON-1TLUXPandaX-IICDMSLiteCRESST-IICDEX-10

(13 TeV)-1 (8 TeV) + 38.2 fb-1 (7 TeV) + 19.7 fb-14.9 fb

CMS

Higgs-portal models

Figure 10: On the left, observed 95% CL upper limits on (s/sSM)B(H ! inv) for a Higgsboson with a mass of 125.09 GeV, whose production cross section varies as a function of thecoupling modifiers kV and kF. Their best estimate, along with the 68% and 95% CL contoursfrom Ref. [4], are also reported. The SM prediction corresponds to kV = kF = 1. On the right,90% CL upper limits on the spin-independent DM-nucleon scattering cross section in Higgs-portal models, assuming a scalar (solid orange) or fermion (dashed red) DM candidate. Limitsare computed as a function of mc and are compared to those from the XENON1T [79], LUX [80],PandaX-II [81], CDMSlite [82], CRESST-II [83], and CDEX-10 [84] experiments.

10 SummaryA search for invisible decays of a Higgs boson is presented using proton-proton (pp) collisiondata at a center-of-mass energy

ps = 13 TeV, collected by the CMS experiment in 2016 and

corresponding to an integrated luminosity of 35.9 fb�1. The search targets events in which aHiggs boson is produced through vector boson fusion (VBF). The data are found to be consis-tent with the predicted standard model (SM) backgrounds. An observed (expected) upper limitof 0.33 (0.25) is set, at 95% confidence level (CL), on the branching fraction of the Higgs bosondecay to invisible particles, B(H ! inv), by means of a binned likelihood fit to the dijet massdistribution. In addition, upper limits are set on the product of the cross section and branchingfraction of an SM-like Higgs boson, with mass ranging between 110 and 1000 GeV.

A combination of CMS searches for the Higgs boson decaying to invisible particles, using ppcollision data collected at

ps = 7, 8, and 13 TeV (2015 and 2016), is also presented. The com-

bination includes searches targeting Higgs boson production via VBF, in association with avector boson (with hadronic decays of the W boson and hadronic or leptonic decays of the Z

Gre

g La

ndsb

erg

- DM

at C

ollid

ers

and

Beyo

nd -

Mor

iond

201

9

H(inv.) Limis✦ Higgs boson could couple to DM and decay invisibly

if it's mass mDM < mH/2 = 62.5 GeV ๏ ATLAS: B(H → inv.) < 0.26 (0.17) @ 95% CL ๏ CMS: B(H → inv.) < 0.19 (0.15) @ 95% CL

�36

CMS arXiv:1809.05937

2011-16

ATLAS-CONF-2018-054

✦ Can use dedicated searches for H(inv.) or reinterpret monojet results

✦ Best sensitivity is achieved in combination

✦ Results are competitive with DD experiments for mDM < 10 GeV

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Direct SUSY Searches✦ Since SUSY gives a perfect WIMP candidate, direct searches for neutralino

production set limit on DM ✦ Depending on the nature of the neutralino (wino, bino, Higgsino), limits are still

not very strong, particularly if superpartners of leptons are heavy

�37

https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/CombinedSummaryPlots/SUSY/

) [GeV]20χ∼,

1±χ∼m(

200 400 600 800 1000 1200

) [G

eV]

0 1χ∼m

(

0

100

200

300

400

500

600

700 2lν∼ / Ll

~arXiv:1509.07152ATLAS-CONF-2019-008

τ 2τν∼ / Lτ

∼

arXiv:1407.0350arXiv:1708.07875WW 2larXiv:1403.5294ATLAS-CONF-2019-008

via −1χ∼ +1χ

∼

2l+3lν∼ / Ll~

arXiv:1509.07152arXiv:1803.02762WZ 2l+3larXiv:1403.5294arXiv:1712.08119arXiv:1803.02762arXiv:1806.02293

±l±+lγγWh lbb+2jbb+l arxiv:1812.09432

via02χ∼ ±1χ

∼

τ 2τν∼ / Lτ

∼

arXiv:1708.07875

via02χ∼ ±1χ

∼/ −1χ∼ +1χ

∼

Observed limits at 95% CL PreliminaryATLAS -1=8,13 TeV, 20.3-139 fbs March 2019

) ]20χ∼,

1±χ∼ ) + m(

10χ∼ [ m( 2

1 ) = ν∼/ Lτ∼/ Ll

~m(

)1

0χ∼

) = m

( 2

0χ∼

m(

2012-2016

https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/CombinedSummaryPlots/SUSY/

Slid

e G

reg

Land

sber

g - N

ew Id

eas

in B

SM S

earc

hes

- Bra

ndei

s - F

all 2

018

Supercemetery?✦ At first glance, the TeV scale SUSY is simply not there...

�38

Model Signature!L dt [fb−1] Mass limit Reference

Incl

usi

veS

ea

rch

es

3rd

gen.

squark

sdir

ect

pro

duct

ion

EW

dir

ect

Lo

ng

-liv

ed

pa

rtic

les

RP

V

qq, q→qχ01

0 e, µ 2-6 jets EmissT 36.1 m(χ

01)<100 GeV 1712.023321.55q [2×, 8× Degen.] 0.9q [2×, 8× Degen.]

mono-jet 1-3 jets EmissT 36.1 m(q)-m(χ

01)=5 GeV 1711.033010.71q [1×, 8× Degen.] 0.43q [1×, 8× Degen.]

gg, g→qqχ01


01)<200 GeV 1712.023322.0g

m(χ01)=900 GeV 1712.023320.95-1.6gg Forbidden

gg, g→qq(ℓℓ)χ01

3 e, µ 4 jets 36.1 m(χ01)<800 GeV 1706.037311.85g

ee, µµ 2 jets EmissT 36.1 m(g)-m(χ

01 )=50 GeV 1805.113811.2g

gg, g→qqWZχ01


01) <400 GeV 1708.027941.8g

3 e, µ 4 jets 36.1 m(g)-m(χ01)=200 GeV 1706.037310.98g

gg, g→ttχ01

0-1 e, µ 3 b EmissT 79.8 m(χ

01)<200 GeV ATLAS-CONF-2018-0412.25g

3 e, µ 4 jets 36.1 m(g)-m(χ01)=300 GeV 1706.037311.25g

b1b1, b1→bχ01/tχ

±1

Multiple 36.1 m(χ01)=300 GeV, BR(bχ

01)=1 1708.09266, 1711.033010.9b1b1 Forbidden

Multiple 36.1 m(χ01)=300 GeV, BR(bχ

01)=BR(tχ

±1 )=0.5 1708.092660.58-0.82b1b1 Forbidden

Multiple 36.1 m(χ01)=200 GeV, m(χ

±1 )=300 GeV, BR(tχ

±1 )=1 1706.037310.7b1b1 Forbidden

b1b1, b1→bχ02 → bhχ

01

0 e, µ 6 b EmissT 139 ∆m(χ

02 , χ

01)=130 GeV, m(χ

01)=100 GeV SUSY-2018-310.23-1.35b1b1 Forbidden

∆m(χ02 , χ

01)=130 GeV, m(χ

01)=0 GeV SUSY-2018-310.23-0.48b1b1

t1 t1, t1→Wbχ01 or tχ

01

0-2 e, µ 0-2 jets/1-2 b EmissT 36.1 m(χ

01)=1 GeV 1506.08616, 1709.04183, 1711.115201.0t1

t1 t1, Well-Tempered LSP Multiple 36.1 m(χ01)=150 GeV, m(χ

±1 )-m(χ

01)=5 GeV, t1 ≈ tL 1709.04183, 1711.115200.48-0.84t1t1

t1 t1, t1→τ1bν, τ1→τG 1 τ + 1 e,µ,τ 2 jets/1 b EmissT 36.1 m(τ1)=800 GeV 1803.101781.16t1

t1 t1, t1→cχ01 / cc, c→cχ

01

0 e, µ 2 c EmissT 36.1 m(χ

01)=0 GeV 1805.016490.85c

m(t1,c)-m(χ01 )=50 GeV 1805.016490.46t1

0 e, µ mono-jet EmissT 36.1 m(t1,c)-m(χ

01)=5 GeV 1711.033010.43t1

t2 t2, t2→t1 + h 1-2 e, µ 4 b EmissT 36.1 m(χ

01)=0 GeV, m(t1)-m(χ

01)= 180 GeV 1706.039860.32-0.88t2

χ±1χ0

2 via WZ 2-3 e, µ EmissT 36.1 m(χ

01)=0 1403.5294, 1806.022930.6χ±

1 /χ0

2ee, µµ ≥ 1 Emiss

T 36.1 m(χ±1 )-m(χ

01)=10 GeV 1712.081190.17χ±

1 /χ0

2

χ±1χ∓

1 via WW 2 e, µ EmissT 139 m(χ

01)=0 ATLAS-CONF-2019-0080.42χ±

1

χ±1χ0

2 via Wh 0-1 e, µ 2 b EmissT 36.1 m(χ

01)=0 1812.094320.68χ±

1 /χ0

2

χ±1χ∓

1 via ℓL/ν 2 e, µ EmissT 139 m(ℓ,ν)=0.5(m(χ

±1 )+m(χ

01)) ATLAS-CONF-2019-0081.0χ±

1

χ±1χ∓

1 /χ02, χ

+

1→τ1ν(τν), χ02→τ1τ(νν) 2 τ Emiss

T 36.1 m(χ01)=0, m(τ, ν)=0.5(m(χ

±1 )+m(χ

01)) 1708.078750.76χ±

1 /χ0

2

m(χ±1 )-m(χ

01 )=100 GeV, m(τ, ν)=0.5(m(χ

±1 )+m(χ

01)) 1708.078750.22χ±

1 /χ0

2

ℓL,R ℓL,R, ℓ→ℓχ01

2 e, µ 0 jets EmissT 139 m(χ

01)=0 ATLAS-CONF-2019-0080.7ℓ

2 e, µ ≥ 1 EmissT 36.1 m(ℓ)-m(χ

01 )=5 GeV 1712.081190.18ℓ

HH, H→hG/ZG 0 e, µ ≥ 3 b EmissT 36.1 BR(χ

01 → hG)=1 1806.040300.29-0.88H 0.13-0.23H

4 e, µ 0 jets EmissT 36.1 BR(χ

01 → ZG)=1 1804.036020.3H

Direct χ+

1χ−

1 prod., long-lived χ±1 Disapp. trk 1 jet Emiss

T 36.1 Pure Wino 1712.021180.46χ±1

Pure Higgsino ATL-PHYS-PUB-2017-0190.15χ±1

Stable g R-hadron Multiple 36.1 1902.01636,1808.040952.0g

Metastable g R-hadron, g→qqχ01

Multiple 36.1 m(χ01)=100 GeV 1710.04901,1808.040952.4g [τ( g) =10 ns, 0.2 ns] 2.05g [τ( g) =10 ns, 0.2 ns]

LFV pp→ντ + X, ντ→eµ/eτ/µτ eµ,eτ,µτ 3.2 λ′311=0.11, λ132/133/233=0.07 1607.080791.9ντ

χ±1χ∓

1 /χ02 → WW/Zℓℓℓℓνν 4 e, µ 0 jets Emiss

T 36.1 m(χ01)=100 GeV 1804.036021.33χ±

1 /χ0

2 [λi33 ! 0, λ12k ! 0] 0.82χ±1 /χ

0

2 [λi33 ! 0, λ12k ! 0]

gg, g→qqχ01, χ

01 → qqq 4-5 large-R jets 36.1 Large λ′′

112 1804.035681.9g [m(χ0

1)=200 GeV, 1100 GeV] 1.3g [m(χ0

1)=200 GeV, 1100 GeV]Multiple 36.1 m(χ

01)=200 GeV, bino-like ATLAS-CONF-2018-0032.0g [λ′′

112=2e-4, 2e-5] 1.05g [λ′′

112=2e-4, 2e-5]

tt, t→tχ01, χ

01 → tbs Multiple 36.1 m(χ

01)=200 GeV, bino-like ATLAS-CONF-2018-0031.05g [λ′′

323=2e-4, 1e-2] 0.55g [λ′′

323=2e-4, 1e-2]

t1 t1, t1→bs 2 jets + 2 b 36.7 1710.071710.61t1 [qq, bs] 0.42t1 [qq, bs]

t1 t1, t1→qℓ 2 e, µ 2 b 36.1 BR(t1→be/bµ)>20% 1710.055440.4-1.45t1

1 µ DV 136 BR(t1→qµ)=100%, cosθt=1 ATLAS-CONF-2019-0061.6t1 [1e-10< λ′23k<1e-8, 3e-10< λ′

23k<3e-9] 1.0t1 [1e-10< λ′

23k<1e-8, 3e-10< λ′

23k<3e-9]

Mass scale [TeV]10−1 1

ATLAS SUSY Searches* - 95% CL Lower LimitsMarch 2019

ATLAS Preliminary√

s = 13 TeV

*Only a selection of the available mass limits on new states orphenomena is shown. Many of the limits are based onsimplified models, c.f. refs. for the assumptions made.

Slid

e G

reg

Land

sber

g - S

USY

Sea

rche

s at

CM

S - C

orfu

201

7

But: Read the Fine Print!✦ Keep in mind that:

๏ Searches typically assume 100% branching fraction in a particular channel

๏ Many searches assume mass degeneracy between various SUSY particles, e.g., squarks of different generation

๏ Interpretation is usually done via simplified model framework, not in the full model�39

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

500 600 1000 2000 3000 4000 [GeV]Z'm

2−10

1−10

1

10

210

310

410

thσ/

σ

combination )bh(bσ1±combination )ττh(

σ2±combination )γγh(

h(ZZ) h(WW)

CMS Preliminary

Z'-2HDM, Dirac DM=100 GeVχ

=300 GeV, mAm=1β=1, tan

χ=0.8, g

Z'g

H=m±H=mAm

(13 TeV)-135.9 fb

Solid (dashed) lines:observed (expected) 95% CL limits

1000 1500 2000 2500 3000

500

600

700

800

900

1000

ATLAS Preliminaryps = 13 TeV, 79.8 fb

�1

h(bb) + Emiss

T: Z’+2HDM simplified model

tan � = 1, gZ = 0.8, m� = 100 GeV, mH = mH± = 300 GeV

Figure 8: Exclusion contours for the Z 0-2HDM scenario in the (mZ0,mA) plane for tan � = 1, gZ0 = 0.8, and m� =100 GeV. The observed limits (solid line) are consistent with the expectation under the SM-only hypothesis (denselydashed line) within uncertainties (filled band). Observed limits from previous ATLAS results at

ps = 13 TeV

(dash-dotted line [19]) are also shown.

0

2

4

6

8

10

12

Expecte

dupper

lim

it

onµ

(95%

CL)

ATLAS Preliminaryps = 13 TeV, 79.8 fb

�1

h(bb) + Emiss

T: Z’+2HDM simplified model

tan� = 1, gZ = 0.8, m� = 100 GeV, mA = 500 GeV

VR track jets 2b

(±1� and ±2�)

FR track jets 2b

(scaled)

1000 1500 2000 2500 3000

mZ 0 [GeV]

1

2

3

µlim

it

FR

/V

R

Figure 9: Comparison of the expected upper limits on the signal strength µ for the analysis using variable-radius(VR) track-jets (dashed line) against the previous iteration of the analysis performed with fixed-radius (FR) track-jets(dash-dotted line) with two b-tagged jet and scaled to 79.8 fb�1, for fixed mA = 500 GeV and di�erent values ofmZ0 of the Z 0-2HDM benchmark model. Other di�erences between the two analyses include the suppression of themultijet background using the object-based Emiss

T significance, reduced uncertainties from the MC statistics, and theimprove calibration of the b-tagging e�ciency in the VR analysis. The lower panel is the ratio of the upper limits,showing a significant improvement in the high mZ0 region.

21

Mono-Higgs Production�40

gZ' = 0.8, tanβ = 1

CM

S PA

S EX

O-1

8-01

1

2 3 Data and Simulated Samples

Figure 1: Feynman diagrams for the benchmark DM signal models: baryonic Z0 (left) and

2HDM (right).

A fit-based analysis similar to that of the SM h! gg search is used to estimate the signal yield.In addition to a high-p

missT category, a lower p

missT category is also considered in order to be

sensitive to possible signals with less pmissT .

2 The CMS Detector

The central feature of the CMS detector is a superconducting solenoid, of 6 m internal di-ameter, providing an axial magnetic field of 3.8 T along the beam direction. Within the su-per conducting solenoid volume are a silicon pixel and strip tracker, a lead-tungstate crystalelectromagnetic calorimeter (ECAL), and a brass and scintillator hadron calorimeter (HCAL).Charged particle trajectories are measured by the silicon pixel and strip tracker system, cover-ing 0 f 2p in azimuth and |h| < 2.5, where the pseudorapidity is h = � ln (tan q/2), andq is the polar angle with respect to the counterclockwise-beam direction. Muons are measuredin gas-ionization detectors embedded in the steel return yoke. Extensive forward calorimetrycomplements the coverage provided by the barrel and endcap detectors. The electromagneticcalorimeter, which surrounds the tracker volume, consists of 75,848 lead-tungstate crystals thatprovide coverage in pseudorapidity |h| < 1.479 in the barrel region (EB) and 1.479 < |h| < 3.0in two endcap regions (EE). The EB modules are arranged in projective towers. A preshowerdetector consisting of two planes of silicon sensors interleaved with a total of three X0 of leadis located in front of the EE. In the region |h| < 1.74, the HCAL cells have widths of 0.087 inpseudorapidity and azimuth (f). In the ( h, f ) plane, and for |h| < 1.48, the HCAL cells map onto 5x5 ECAL crystal arrays to form calorimeter towers projecting radially outwards from closeto the nominal interaction point. At larger values of |h|, the size of the towers increases and thematching ECAL arrays contain fewer crystals. Within each tower, the energy deposits in ECALand HCAL cells are summed to define the calorimeter tower energies, subsequently used toprovide the energies and directions of hadronic jets (highly collimated showers of particles). Amore detailed description of the CMS detector can be found in Ref. [11].

3 Data and Simulated Samples

The data considered in this analysis correspond to an integrated luminosity of 35.9 fb�1 col-lected by the CMS experiment at the CERN LHC in 2016 at

ps = 13 TeV. Diphoton triggers

with asymmetric transverse energy thresholds (30/18 GeV) were used to select events. Theanalyzed sample fulfills standard data quality criteria for all components of the CMS detector.

The analysis is optimized using fully simulated samples of the dark matter associated pro-duction with a Higgs boson in 2HDM and Baryonic Z’ (Z0

B) models [10]. 2HDM signals are

2016

ATLAS-CONF-2018-039

2015

2016

2017

gq = 0.25, g𝛘 = 1

✦ Mono-Higgs searches in the latest addition in the context of certain vector mediator (Z') models

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Mono-H Event Display�41

Searches for the Mediator

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Search for the Mediator✦ One doesn’t need to produce DM at the LHC to

look for a mediator ✦ Even if it's too heavy to be produced

or too light to be observed, one canfocus on the mediator particle, whichvery well could be within the reach of the LHC!

✦ Since the mediator is coupled to the initial state, one could look for dijet decays of the mediator by "recycling" dijet resonance searches ๏ Also possible to reinterpret dilepton searches if the

mediator couples to leptons in addition to quarks

�43

11 1 1 1

�

� q

q

gl

l

� �

�

�

q qq

q

Z

q

q

q

q

q

q(a) (b) (c) (d) (e)

q

q

RR⇤ R(⇤)

g� gq

R

R⇤

g�

g�

gq

gq gqgq

gq gq

Figure 1. The processes considered in this work in terms of visible sector quarks (q, q), DSPs (�, �)and the on-shell (o↵-shell) mediator particle R (R⇤). The various process are: (a) DM annihilationwhich sets the relic abundance, (b) DM scattering in direct detection experiments, (c) monojetsignatures, in this case due to initial state radiation of a gluon, (d) LHC Dijet resonance signaturespurely through mediator-quark couplings and (e) dijet associated production.

in order to avoid overstating the strength of direct detection limits. This approach

leads to a compelling interplay between the di↵erent DM detection techniques and

will lead us to conclude that the LHC monojets, LHC dijets and direct detection

strategies each has a unique foothold in the search for DSPs.

In figure 1 we sketch the setup for a dark sector theory involving a DSP � and a

mediator between the visible sector and the dark sector R, together with the detection

processes considered in this work. We denote the couplings between the mediator and

the visible sector quarks (the DSP) with gq (g�). For the purposes of exploring the broad

phenomenology of this dark sector and the general interplay between the di↵erent probes let

us combine the two couplings into an e↵ective DSP-SM coupling g =p

gq g� and consider

the e↵ect of varying the coupling g. The local density of DSPs in the Milky Way ⇢ is

proportional to the DSP relic abundance from thermal freeze-out ⌦DSP, which scales as

the inverse of the annihilation cross section, i.e. ⇢ / ⌦DSP / g�4. Any cross section

involving interactions between the visible sector and the DSP, such as collider production

and direct detection, will scale as � / g4 [1, 28–31] (assuming an o↵-shell mediator). Thus,

broadly speaking, the rate of events in di↵erent DM probes have very di↵erent scaling with

couplings if a standard thermal history is assumed. They are:

• Collider searches for missing energy: Rate / � / g4 .

• Direct detection: Rate / (� ⇥ ⇢) / g0 .

• Indirect detection: Rate / (� ⇥ ⇢2) / g�4 .

Furthermore, resonance searches at colliders typically depend on the production cross sec-

tion for the resonance, �R, multiplied with the branching ratio into the final state under

consideration. If the (on-shell) mediator has a large branching into light quarks we hence

obtain the final important signature

• Collider searches for dijet resonances: Rate / �R / g2q .

This simple consideration demonstrates that, assuming a standard thermal history and con-

sidering the specific phenomenology of the mediator, these four di↵erent detection strate-

gies are parametrically complementary. In essence, large couplings imply large collider

– 3 –

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

ATLA

S-C

ON

F-20

19-0

07

Dijet Resonance Searches�44

2016

Low-mass trigger-level

✦ Standard search to do at any new energy ✦ Recent addition to the dijet search portfolio:

๏ Scouting (trigger-level) analysis based on low-threshold triggers writing only very limited information about the event

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

ATLA

S-C

ON

F-20

19-0

07

Dijet Resonance Searches�44

2016

Low-mass trigger-level

ATLA

S ar

Xiv:

1804

.034

96E

vents

/ B

in

410

510

610

710

810

[GeV]m500 600 700 800 9001000 2000

2−

02

[GeV]jjm500 600 700 800 9001000 2000

2−

02

Sig

nifi

cance

ATLAS

, |y*| < 0.3-1Data, 3.6 fbBackground fitBumpHunter interval

x 500σZ’, = 0.1

q = 550 GeV, gZ’m

, |y*| < 0.6-1Data, 29.3 fbBackground fitBumpHunter interval

x 500σZ’, = 0.1

q = 750 GeV, gZ’m

-value = 0.6pBH

-value = 0.42p 2χ

-value = 0.44pBH

-value = 0.13p 2χ

Figure 4: The reconstructed dijet mass distribution (filled points) for events in the |y⇤ | < 0.3 and |y⇤ | < 0.6 signalregions. The solid lines depict the background estimate obtained by a sliding-window fit. Overall agreementbetween the background estimate and the data is quantified by the �2 p-value. The most discrepant localized excessin either signal region identified by the BumpHunter algorithm is indicated by the vertical lines. The open pointsshow two possible signal models. The lower panels show the bin-by-bin significances of di�erences between thedata and the background estimate, considering only statistical uncertainties.

6 Results and limits

Figure 4 shows the invariant mass distributions for dijet events in each signal region including the resultsfrom the sliding-window background estimates. The global �2 p-value is 0.13 in the |y⇤ | < 0.6 signalselection and 0.42 in the |y⇤ | < 0.3 signal selection, indicating the data agrees well with the backgroundestimate. The most discrepant interval identified by the BumpHunter algorithm [30, 31] is 889–1007 GeVfor events with |y⇤ | < 0.6. Accounting for statistical uncertainties only, the probability of observinga deviation at least as significant as that observed in data, anywhere in the distribution, is 0.44 andcorresponds to significance of 0.16 �. Thus, there is no evidence of any localized excess.

Limits are set on both a leptophobic Z 0 simplified dark-matter model [32] and a generic Gaussian model.The Z 0 simplified model assumes axial-vector couplings to SM quarks and to a Dirac fermion dark-mattercandidate. No interference with the SM is simulated. Signal samples were generated so that the decay rateof the Z 0 into dark-matter particles is negligible and the dijet production rate and resonance width dependonly on the coupling of the Z 0 to quarks, gq, and the mass of the resonance, mZ0 [9]. The model’s matrixelements were calculated in M��G�� 5 [33] and parton showering was performed in P�� 8 [34].The width of a Z 0 resonance with gq = 0.10, including parton shower and detector resolution e�ects, isapproximately 7%. Limits are set on the cross-section, �, times acceptance, A, times branching ratio, B,of the model, and then displayed in the (gq,mZ0) plane.3 The acceptance for a mass of 550 GeV is 20%for a Z 0 simplified model with gq = 0.10 for the |y⇤ | < 0.3 signal selection, and 41% for a signal of massequal to 750 GeV for the |y⇤ | < 0.6 signal selection.

Limits are also set on a generic model where the signal is modeled as a Gaussian contribution to theobserved mj j distribution. For a given mean mass, mG , four di�erent Gaussian widths are considered: awidth equal to the simulated mass resolution (which ranges between 4% and 6%), and the fixed fractions5%, 7% and 10% of mG . As the width increases, the expected signal contribution is distributed across

3 Limits on the coupling are obtained accounting for the scaling of the signal cross-section with g2q .

7

✦ Standard search to do at any new energy ✦ Recent addition to the dijet search portfolio:

๏ Scouting (trigger-level) analysis based on low-threshold triggers writing only very limited information about the event

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Dijet Event Display�45

M = 8 TeV

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Mediators: Paradigm Shift✦ At first, we were looking at the highest masses, which opened up due to

the record-high machine energy ✦ These are low-background searches, but only sensitive to large couplings ✦ Last few years marked a shift in paradigm: we are going for high-

background, experimentally challenging searches for low couplings and low masses - something that earlier machines may have missed!

�46

Mass

Cou

plin

g

"Stairway to Hell"

Early LHC Searches

Nowadays

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Trijets as Dijet Proxies✦ Usually, initial-state radiation (ISR) creates

difficulties at the LHC, as it pollutes final states we are interested in with extra jets

✦ However, it could also be our best friend: ๏ It gives the possibility to trigger on an event when

everything else fails - perfect for low-mass final states (or DM!)

๏ While one has to pay a price for an energetic ISR jet (or a photon) , it's a great (only?) way to trigger...

�47

q

q_

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Boost or Bust✦ Typical trigger threshold on an ISR jet is ~500 GeV ✦ If we want to extend the dijet search to even

lower masses than the scouting technique allows, we typically have a boosted topology

�48

Z’

Boosted jets: Increasing transverse momentum, pT

•••• 𝑍’

𝒁’: Boosted Dijet + ISR

Small-radius jets Large-radius jet

q’

q

arXiv:1801.08769 (sub. to Phys. Lett. B)

Lorentz boost (ɣ)

𝛂 ≈ 2/ɣ

ɣ ~ pTISR/m(Z'); for m(Z') ~ 100 GeV, ɣ ~ 5, and 𝛂 ~ 0.5: reconstructed as a single jet

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Jet Substructure Techniques�49

18 6 Performance in data and systematic uncertainties

Pruned jet mass (GeV)40 50 60 70 80 90 100 110 120 130

Even

ts /

( 5 G

eV )

50

100

150

200

250 POWHEG+PYTHIA6tt Single t

WW/WZ/ZZ W+jets MG+PYTHIA6

Data MC total fit

Data total fit MC bkg fit

Data bkg fit

νµ →W

< 0.51τ/2τ

(8 TeV)-119.7 fb

CMS


Even

ts /

( 5 G

eV )

20

40

60

80

100



Data MC total fit


Data bkg fit

νµ →W

> 0.51τ/2τ

(8 TeV)-119.7 fb

CMS


Even

ts /

( 5 G

eV )

20

40

60

80

100

120

140



Data MC total fit


Data bkg fit

ν e→W

< 0.51τ/2τ

(8 TeV)-119.7 fb

CMS


Even

ts /

( 5 G

eV )

10

20

30

40

50

60


WW/WZ/ZZ W+jets MG+PYHTIA6

Data MC total fit


Data bkg fit

ν e→W

> 0.51τ/2τ

(8 TeV)-119.7 fb

CMS

Figure 11: Pruned jet mass distribution in the tt control sample that (left column) pass and(right column) fail the t2/t1 < 0.5 selection for the (upper row) muon, and for the (lower row)electron channels. The result of the fit to data and simulation are shown, respectively, by thesolid and long-dashed line and the background components of the fit are shown as dashed-dotted and short-dashed line.

✦ In the past 10 years, we saw significant theoretical and experimental developments in identifying jet with substructure

✦ These involve several steps: ๏ Jet grooming - removing soft,

wide-angle radiation that artificially increases the jet invariant mass

๏ Jet substructure: how likely isthat a large-radius jet consists of N subjets

๏ Jet mass measurement - after grooming and determining that the jet has a substructure, jet invariant mass becomes a powerful discriminant to look for resonances decaying into two jets

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Low-Mass Dijets�50

for details) are sensitive to di�erent mass regimes as well as coupling values. The boosted dijet+ISRanalysis has the best reach for low masses, excluding Z

0

A mediator masses between 100 GeV and 220 GeV.Two new interpretations, for the dibjet and tt resonance analyses, are presented for these models. The dibjet(tt resonance) analysis places constraints on Z

0

A mediators with masses between 500 GeV and 2.5 (2) TeV,in the same region of sensitivity of the dijet, TLA dijet, and boosted dijet+ISR analyses.

To illustrate the complementarity of the searches [60, 62], three di�erent coupling scenarios are alsoconsidered in the interpretation of the results:

Scenario 1 gq = 0.25, g` = 0, g� = 1 (leptophobic Z0

V/Z 0

A);

Scenario 2 gq = 0.1, g` = 0.01, g� = 1 (leptophilic Z0

V);


A).

In particular, the lower lepton coupling value is set to highlight the dilepton search sensitivity even for verysmall values of this parameter.

The exclusions from the resonance searches (dijet, dibjet, dilepton) in the (mZ0

V/A,m�) plane are derived

from the limits placed on resonances reconstructed with a Gaussian shape, while the limits from the

100 200 1000 2000 [GeV]

AZ'm0.030.040.05

0.1

0.2

0.30.40.5

1qg

(Preliminary)

(Preliminary)Jet trigger

Di-b-jet trigger

| < 0.612|y*

| < 0.312|y*

Phys. Rev. D 91, 052007 (2015)

1−20.3 fb

arXiv: 1801.08769

1−36.1 fb

ATLAS-CONF-2016-070

1−Preliminary, 15.5 fb

ATLAS-CONF-2016-070


Phys. Rev. D 98 (2018) 032016

1−24.3 & 36.1 fb

Phys. Rev. Lett. 121 (2018) 081801

1−3.6 & 29.7 fb

Eur. Phys. J. C 78 (2018) 565

1−36.1 fb

Phys. Rev. D 96, 052004 (2017)

1−37.0 fb

Phys. Rev. D 96, 052004 (2017)

1−37.0 fb

ATLAS

Axial vector mediatorDirac DM

= 1.0χ

= 10 TeV, gχm

1− = 13 TeV, 3.6-37.0 fbs

Dijet 8 TeV

Boosted dijet + ISR

)γResolved dijet + ISR (

Resolved dijet + ISR (j)

Dibjet

Dijet TLA

resonancestt

Dijet

Dijet angular

95% CL upper limitsObservedExpected

=0.07Z'/mΓ

=0.1Z'/mΓ

=0.15Z'/mΓ

Figure 10: Dijet search contours for 95% CL upper limits on the coupling gq as a function of the resonance mass mZ0

Afor the leptophobic axial-vector Z

0

A model. The expected limits from each search are indicated by dotted lines. TheTLA dijet analysis has two parts, employing di�erent datasets with di�erent selections in the rapidity di�erence y⇤ asindicated. The yellow contour shows the results of the dijet search using 20.3 fb�1 of 8 TeV data. Coupling valuesabove the solid lines are excluded, as long as the signals are narrow enough to be detected using these searches.The TLA dijet search with |y⇤ | < 0.6 is sensitive up to �/mZ0 = 7%, the TLA dijet with |y⇤ | < 0.3 and dijet + ISRsearches are sensitive up to �/mZ0 = 10%, and the dijet and dibjet searches are sensitive up to �/mZ0 = 15%. Thedijet angular analysis is sensitive up to �/mZ0 = 50%. No limitation in sensitivity arises from large width resonancesin the tt resonance analysis. Benchmark width lines are indicated in the canvas. The �/mZ0 = 50% lies beyond thecanvas borders.

27

ATLAS arXiv:1804.03496

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019


for details) are sensitive to di�erent mass regimes as well as coupling values. The boosted dijet+ISRanalysis has the best reach for low masses, excluding Z

0

A mediator masses between 100 GeV and 220 GeV.Two new interpretations, for the dibjet and tt resonance analyses, are presented for these models. The dibjet(tt resonance) analysis places constraints on Z

0

A mediators with masses between 500 GeV and 2.5 (2) TeV,in the same region of sensitivity of the dijet, TLA dijet, and boosted dijet+ISR analyses.

To illustrate the complementarity of the searches [60, 62], three di�erent coupling scenarios are alsoconsidered in the interpretation of the results:

Scenario 1 gq = 0.25, g` = 0, g� = 1 (leptophobic Z0

V/Z 0

A);


V);


A).

In particular, the lower lepton coupling value is set to highlight the dilepton search sensitivity even for verysmall values of this parameter.

The exclusions from the resonance searches (dijet, dibjet, dilepton) in the (mZ0

V/A,m�) plane are derived

from the limits placed on resonances reconstructed with a Gaussian shape, while the limits from the

100 200 1000 2000 [GeV]

AZ'm0.030.040.05

0.1

0.2

0.30.40.5

1qg

(Preliminary)

(Preliminary)Jet trigger

Di-b-jet trigger

| < 0.612|y*

| < 0.312|y*

Phys. Rev. D 91, 052007 (2015)

1−20.3 fb

arXiv: 1801.08769

1−36.1 fb

ATLAS-CONF-2016-070


ATLAS-CONF-2016-070


Phys. Rev. D 98 (2018) 032016

1−24.3 & 36.1 fb

Phys. Rev. Lett. 121 (2018) 081801

1−3.6 & 29.7 fb

Eur. Phys. J. C 78 (2018) 565

1−36.1 fb

Phys. Rev. D 96, 052004 (2017)

1−37.0 fb

Phys. Rev. D 96, 052004 (2017)

1−37.0 fb

ATLAS

Axial vector mediatorDirac DM

= 1.0χ

= 10 TeV, gχm

1− = 13 TeV, 3.6-37.0 fbs

Dijet 8 TeV

Boosted dijet + ISR

)γResolved dijet + ISR (

Resolved dijet + ISR (j)

Dibjet

Dijet TLA

resonancestt

Dijet

Dijet angular

95% CL upper limitsObservedExpected

=0.07Z'/mΓ

=0.1Z'/mΓ

=0.15Z'/mΓ

Figure 10: Dijet search contours for 95% CL upper limits on the coupling gq as a function of the resonance mass mZ0

Afor the leptophobic axial-vector Z

0

A model. The expected limits from each search are indicated by dotted lines. TheTLA dijet analysis has two parts, employing di�erent datasets with di�erent selections in the rapidity di�erence y⇤ asindicated. The yellow contour shows the results of the dijet search using 20.3 fb�1 of 8 TeV data. Coupling valuesabove the solid lines are excluded, as long as the signals are narrow enough to be detected using these searches.The TLA dijet search with |y⇤ | < 0.6 is sensitive up to �/mZ0 = 7%, the TLA dijet with |y⇤ | < 0.3 and dijet + ISRsearches are sensitive up to �/mZ0 = 10%, and the dijet and dibjet searches are sensitive up to �/mZ0 = 15%. Thedijet angular analysis is sensitive up to �/mZ0 = 50%. No limitation in sensitivity arises from large width resonancesin the tt resonance analysis. Benchmark width lines are indicated in the canvas. The �/mZ0 = 50% lies beyond thecanvas borders.

27

ATLAS arXiv:1804.03496

EM coupling

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019


8 10 20 30 100 200 1000 2000 [GeV]Z'M

1−10

1

qg'

5% = Z'M / Z'Γ

10% = Z'M / Z'Γ

30% = Z'M / Z'Γ

50% = Z'M / Z'Γ

100% = Z'M / Z'Γ

qq→Z'

95% CL exclusions

~100% < Z'M / Z'Γ[arXiv:1801.08769]ATLAS Boosted Dijet, 13 TeV

[EXO-16-046], 13 TeVχCMS Dijet

[arXiv:1901.10917], 13 TeVγATLAS Dijet+ISR

~30% < Z'M / Z'Γ[arXiv:1901.10917]

, 13 TeVγATLAS bb+ISR

[arXiv:1806.00843]CMS Broad Dijet, 13 TeV

[arXiv:1804.03496]ATLAS Dijet TLA, 13 TeV

~10% < Z'M / Z'Γ[arXiv:1703.09127]ATLAS Dijet, 13 TeV

[arXiv:1604.08907]CMS Dijet, 8 TeV

[arXiv:1802.06149]CMS Dijet b tagged, 8 TeV

[arXiv:1806.00843]CMS Dijet Scouting '16, 13 TeV

[arXiv:1806.00843]CMS Dijet '16, 13 TeV

[arXiv:1710.00159]CMS Boosted Dijet, 13 TeV

[EXO-17-026]CMS Dijet '16+'17, 13 TeV

[EXO-17-027], 13 TeVγCMS Boosted Dijet+

[Nucl. Phys. B 400, 3 (1993)]UA2

[arXiv:hep-ex/9702004]CDF Run1

[arXiv:hep-ex/9702004]CDF Run1

[arXiv:1404.3947])Z'/MZ'ΓZ width (all

[arXiv:1404.3947])Z'/M

Z'Γ width (all Υ

CMS Preliminary Moriond 2019

EM coupling

https://twiki.cern.ch/twiki/bin/view/CMSPublic/SummaryPlotsEXO13TeV#Dijet_summary_plots

https://twiki.cern.ch/twiki/bin/view/CMSPublic/SummaryPlotsEXO13TeV#Dijet_summary_plots

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

0 0.5 1 1.5 2 2.5 3 3.5 [TeV]

VZ'm

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

[TeV

]χ

m

ATLAS

= 1χ

= 0, gl

= 0.25, gq

gVector mediator, Dirac DM

All limits at 95% CL

Dijet

DijetPRD 96, 052004 (2017)

-1 = 13 TeV, 37.0 fbsDijet

PRL 121 (2018) 0818016

-1 = 13 TeV, 29.3 fbsDijet TLA

Preliminary ATLAS-CONF-2016-070

-1 = 13 TeV, 15.5 fbsDijet + ISR

resonance

tt

EPJC 78 (2018) 565

-1 = 13 TeV, 36.1 fbs resonancett

Dibjet

PRD 98 (2018) 032016

-1 = 13 TeV, 36.1 fbsDibjet

+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393

-1 = 13 TeV, 36.1 fbs γ+missTE

JHEP 1801 (2018) 126

-1 = 13 TeV, 36.1 fbs+jet missTE

PLB 776 (2017) 318

-1 = 13 TeV, 36.1 fbs+Z(ll) missTE

JHEP 10 (2018) 180

-1 = 13 TeV, 36.1 fbs+V(had) missTE

VZ'

= mχ m×2

= 0.122hcΩ

Thermal Relic

(a)

0 0.5 1 1.5 2 2.5 3 3.5 [TeV]

VZ'm

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

[TeV

]χ

m

ATLAS

= 1χ

= 0.01, gl

= 0.1, gq



Dile

pton

JHEP 10 (2017) 182

-1 = 13 TeV, 36.1 fbsDilepton

Dije

t

DijetPRD 96, 052004 (2017)


PRL 121 (2018) 0818016


+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


VZ'

= mχ m×2

= 0.122hcΩ

Thermal R

elic

(b)

Figure 11: Regions in a (mediator-mass, DM-mass) plane excluded at 95% CL by dijet, dilepton and EmissT +X

searches, for leptophobic (a) or leptophilic (b) vector mediator simplified models described in Section 2.1.1. Theexclusions are computed for a DM coupling g�, quark coupling gq , universal to all flavours, and lepton coupling g`as indicated in each case. Dashed curves labelled “thermal relic” correspond to combinations of DM and mediatormass values that are consistent with a DM density of ⌦h

2 = 0.12 and a standard thermal history as computed inMadDM [62, 214]. Above the curve in (a) annihilation processes described by the simplified model deplete ⌦h

2 tobelow 0.12. In (b), this occurs between the two dashed curves. The dotted line indicates the kinematic thresholdwhere the mediator can decay on-shell into DM.

28

Big Picture: Spin 1�51

✦ Analogous limits for axial vector mediators ๏ The complementarity depends significantly on the coupling values!

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

0 0.5 1 1.5 2 2.5 3 3.5 [TeV]

VZ'm

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

[TeV

]χ

m

ATLAS

= 1χ

= 0, gl

= 0.25, gq



Dijet

DijetPRD 96, 052004 (2017)


PRL 121 (2018) 0818016




resonance

tt

EPJC 78 (2018) 565


Dibjet

PRD 98 (2018) 032016


+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


PLB 776 (2017) 318


JHEP 10 (2018) 180


VZ'

= mχ m×2

= 0.122hcΩ

Thermal Relic

(a)

0 0.5 1 1.5 2 2.5 3 3.5 [TeV]

VZ'm

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

[TeV

]χ

m

ATLAS

= 1χ

= 0.01, gl

= 0.1, gq



Dile

pton

JHEP 10 (2017) 182


Dije

t

DijetPRD 96, 052004 (2017)


PRL 121 (2018) 0818016


+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


VZ'

= mχ m×2

= 0.122hcΩ

Thermal R

elic

(b)





28


✦ Analogous limits for axial vector mediators ๏ The complementarity depends significantly on the coupling values!

0 0.5 1 1.5 2 2.5 3 3.5 [TeV]

VZ'm

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

[TeV

]χ

m

ATLAS

= 1χ

= 0, gl

= 0.25, gq



Dijet

DijetPRD 96, 052004 (2017)


PRL 121 (2018) 0818016




resonance

tt

EPJC 78 (2018) 565


Dibjet

PRD 98 (2018) 032016


+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


PLB 776 (2017) 318


JHEP 10 (2018) 180


VZ'

= mχ m×2

= 0.122hcΩ

Thermal Relic

(a)

0 0.5 1 1.5 2 2.5 3 3.5 [TeV]

VZ'm

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

[TeV

]χ

m

ATLAS

= 1χ

= 0.01, gl

= 0.1, gq



Dile

pton

JHEP 10 (2017) 182


Dije

t

DijetPRD 96, 052004 (2017)


PRL 121 (2018) 0818016


+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


VZ'

= mχ m×2

= 0.122hcΩ

Thermal R

elic

(b)





28

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Comparison w/ DD✦ Nice complementarity with DD experiments

�52 1 10 210 310 [GeV]χm

48−10

47−10

46−10

45−10

44−10

43−10

42−10

41−10

40−10

39−10

38−10

37−10]2-n

ucle

on) [

cmχ (

SIσ

ATLAS

= 1χ

= 0, gl

= 0.25, gq


ATLAS limits at 95% CL, direct detection limits at 90% CL

Dijet

Dijet

PRD 96, 052004 (2017)


PRL 121 (2018) 0818016



-1 = 13 TeV, 15.5 fbsDijet + ISR tt resonance

EPJC 78 (2018) 565-1 = 13 TeV, 36.1 fbs

tt resonance

Dibjet PRD 98 (2018) 032016

-1 = 13 TeV, 36.1 fbs

Dibjet

+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


PLB 776 (2017) 318


JHEP 10 (2018) 180


CRESST III

arXiv:1711.07692CRESST III

XENON1T

arXiv:1805.12562XENON1T

PandaX

PRL 117, 121303 (2016)PandaX

DarkSide-50

arXiv:1802.06994DarkSide-50

LUX

Phys. Rev. Lett. 116, 161302 (2016)PRL 118, 021303 (2017)LUX

(a)

1 10 210 310 [GeV]χm

46−10

45−10

44−10

43−10

42−10

41−10

40−10

39−10

38−10

37−10]2-p

roto

n) [c

mχ (

SDσ

ATLAS

= 1χ

= 0.1, gl

= 0.1, gq

gAxial-vector mediator, Dirac DM


Dilepton

JHEP 10 (2017) 182-1 = 13 TeV, 36.1 fbs

Dilepton

Dijet

Dijet

PRD 96, 052004 (2017)


PRL 121 (2018) 0818016


+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


8F3PICO-60 C

PRL 118, 251301 (2017)8F3PICO-60 C

(b)

Figure 13: A comparison of the inferred limits with the constraints from direct detection experiments on (a) thespin-dependent WIMP–proton scattering cross-section in the context of the vector leptophobic model and (b) thespin-independent WIMP–nucleon scattering cross-section in the context of the axial-vector leptophilic model. Theresults from this analysis, excluding the region inside or to the left of the contour, are compared with limits fromdirect detection experiments. ATLAS limits are shown at 95% CL and direct detection limits at 90% CL. ATLASsearches and direct detection experiments exclude the shaded areas. Exclusions beyond the canvas are not implied forthe ATLAS results. The dijet and E

missT +X exclusion regions represent the union of exclusions from all analyses of

that type.

31

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Comparison w/ DD✦ Nice complementarity with DD experiments

�52 1 10 210 310 [GeV]χm

48−10

47−10

46−10

45−10

44−10

43−10

42−10

41−10

40−10

39−10

38−10

37−10]2-n

ucle

on) [

cmχ (

SIσ

ATLAS

= 1χ

= 0, gl

= 0.25, gq



Dijet

Dijet

PRD 96, 052004 (2017)


PRL 121 (2018) 0818016




EPJC 78 (2018) 565-1 = 13 TeV, 36.1 fbs

tt resonance

Dibjet PRD 98 (2018) 032016

-1 = 13 TeV, 36.1 fbs

Dibjet

+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


PLB 776 (2017) 318


JHEP 10 (2018) 180


CRESST III


XENON1T


PandaX

PRL 117, 121303 (2016)PandaX

DarkSide-50


LUX


(a)

1 10 210 310 [GeV]χm

46−10

45−10

44−10

43−10

42−10

41−10

40−10

39−10

38−10

37−10]2-p

roto

n) [c

mχ (

SDσ

ATLAS

= 1χ

= 0.1, gl

= 0.1, gq



Dilepton

JHEP 10 (2017) 182-1 = 13 TeV, 36.1 fbs

Dilepton

Dijet

Dijet

PRD 96, 052004 (2017)


PRL 121 (2018) 0818016


+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


8F3PICO-60 C

PRL 118, 251301 (2017)8F3PICO-60 C

(b)



that type.

31

1 10 210 310 [GeV]χm

48−10

47−10

46−10

45−10

44−10

43−10

42−10

41−10

40−10

39−10

38−10

37−10]2-n

ucle

on) [

cmχ (

SIσ

ATLAS

= 1χ

= 0, gl

= 0.25, gq



Dijet

Dijet

PRD 96, 052004 (2017)


PRL 121 (2018) 0818016




EPJC 78 (2018) 565-1 = 13 TeV, 36.1 fbs

tt resonance

Dibjet PRD 98 (2018) 032016

-1 = 13 TeV, 36.1 fbs

Dibjet

+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


PLB 776 (2017) 318


JHEP 10 (2018) 180


CRESST III


XENON1T


PandaX

PRL 117, 121303 (2016)PandaX

DarkSide-50


LUX


(a)

1 10 210 310 [GeV]χm

46−10

45−10

44−10

43−10

42−10

41−10

40−10

39−10

38−10

37−10]2-p

roto

n) [c

mχ (

SDσ

ATLAS

= 1χ

= 0.1, gl

= 0.1, gq



Dilepton

JHEP 10 (2017) 182-1 = 13 TeV, 36.1 fbs

Dilepton

Dijet

Dijet

PRD 96, 052004 (2017)


PRL 121 (2018) 0818016


+XmissTE

+XmissTE

Eur. Phys. J. C 77 (2017) 393


JHEP 1801 (2018) 126


8F3PICO-60 C

PRL 118, 251301 (2017)8F3PICO-60 C

(b)



that type.

31

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019


✦ For the first time started probing spin-0 mediators

2016

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

DM: Quo Vadis?✦ We hope that the multi-prong attack on the DM will

bear fruit ๏ DD experiments are moving to a multi-ton mass range

and will soon reach the neutrino floor ๏ Many new ideas of non-WIMP and light DM searches

(electron recoil, CCD detectors, etc.) ๏ New generation of axion searches curbing the available

parameter space ๏ New indirect detection experiments coming online, and

exciting hints from the present ones are being followed up

๏ Collider experiments are processing large amount of Run 2 data and are moving toward more sophisticated models of dark matter, including hidden-sector searches

�54

Slid

e G

reg

Land

sber

g - D

M a

t Col

lider

s an

d Be

yond

- M

orio

nd 2

019

Conclusions✦ There is an overwhelming evidence that dark matter exists

and outweighs ordinary matter by a factor of five ๏ While there is no guarantee that dark matter has particle physics

origin, this is certainly a compelling possibility ๏ If exist, dark matter particles could be produced at the LHC,

leading to exciting mono-X signatures ๏ In addition, LHC can look directly for the dark matter mediators

✦ A lot of theoretical and experimental progress in the past few years - an exciting and rapidly developing subject

✦ LHC searches are complementary to both direct and indirect detection and offer unique sensitivity for pseudoscalar mediator and/or very light dark matter particles

✦ Many searches ongoing with the large data sets accumulated in the last few years - please stay tuned!

✦ Gravity may be not the only force the dark matter feels!

�55

Thank You!

quest for dark matter at colliders and...

Documents