ATL-PHYS-PROC-2014-142 23September2014
Nuclear Physics B Proceedings Supplement 00 (2014) 1–3
Nuclear Physics B Proceedings Supplement
ATLAS Sensitivity to WIMP Dark Matter in the Monojet Topology at
√ s = 14 TeV
Steven Schramm (on behalf of the ATLAS Collaboration)
Department of Physics, University of Toronto, Toronto, Ontario, Canada M5S1A7
Abstract
Searches for monojet plus missing transverse momentum signatures are sensitive to new phenomena involving invisible particles, such as the pair-production of dark matter, one particularly well motivated possibility. We report on the expected sensitivity to dark matter in monojet signatures at an upgraded LHC with the ATLAS detector. The effective field theory models typically used for monojet dark matter interpretations have validity limitations. These are addressed both through applying additional constraints, and through a first look at the use of simplified models of dark matter pair-production.
Keywords: Dark Matter, WIMPs, monojet, mono-jet, ATLAS
1. Introduction
One of the most sensitive channels for generic Dark Matter (DM) searches at the LHC is the monojet plus missing transverse momentum (EmissT ) topology. Such searches benefit immensely from the increased centre of mass energy, thus it is important to understand what can be expected at the upgraded LHC, both in the coming year and in the more distant future. Increased collision energies also lead to more questions on the validity of the DM models used, thus it is vital to address the topic when communicating with the larger DM community.
The ATLAS Collaboration [1] has carried out such stud- ies, and has published a note with both initial sensitivity projections and validity considerations [2].
2. Simulation and event selection
While it is now known that at least the first year of upgraded LHC data will be at 13 TeV, this was not known at the time of these studies, and thus only a cen- tre of mass energy of 14 TeV was considered. Multi- ple luminosity scenarios and associated pileup condi- tions are studied in order to cover different expected
data taking periods of interest for both the LHC and High Luminosity LHC (HL-LHC). The luminosity sce- narios considered and their associated pileup conditions are 5 fb−1 (hµi=60, first months), 25 fb−1(hµi=60, first year), 300 fb−1 (hµi=60, LHC program), and 3000 fb−1 (hµi=140, HL-LHC program), wherehµiis the average number of collisions per bunch crossing. The scenario of 20 fb−1athµi=20 for 8 TeV is studied as a reference.
Where available, 14 TeV samples produced with the full ATLAS detector simulation inGEANT4[3] are used for backgrounds. Otherwise, scale factors derived using 8 TeV MC are applied to kinematically similar MC sam- ples to estimate missing backgrounds, such asW →µν with the muon added to the EmissT forZ →ννestimation.
The D5 Effective Field Theory (EFT) operator [4] and theZ0simplified model [5] are used as benchmarks for DM sensitivity. All signal samples for both models were produced with the full ATLAS simulation at 14 TeV.
The event selection extends previous 8 TeV searches [6], requiring a central leading jet of pT >300 GeV, an optional second jet of pT >50 GeV, EmissT >400 GeV, no leptons, and events where jets and EmissT are not aligned. Two additional signal regions are defined by increasing the EmissT cut to 600 and 800 GeV.
Steven Schramm (on behalf of the ATLAS Collaboration)/Nuclear Physics B Proceedings Supplement 00 (2014) 1–3 2
threshold [GeV]
miss
ET
[GeV] *Suppression Scale M
0 500 1000 1500 2000 2500 3000 3500
14TeV, 25 fb-1 8TeV, 20 fb-1 5% syst
=50 GeV mχ
=400 GeV mχ
Simulation Preliminary ATLAS
400 600 800
< 4π gDM gSM <
π
Figure 1: Expected sensitivity with a flat 5% systematic for 50 and 400 GeV DM after 1 year of 8 and 14 TeV data [2].
3. Expected sensitivity
Two simple uncertainty parametrizations are consid- ered for the expected scenarios of early data conditions and the ultimate reach of the LHC/HL-LHC, chosen to be flat systematics of 5% and 1% respectively. With increasing luminosity, the 5% systematic quickly be- comes the limitation, thus improving detector perfor- mance and the understanding of jets, EmissT , and other physical objects will be crucial. Improved sensitiv- ity in the 300 and 3000 fb−1scenarios can possibly be achieved by further increasing the EmissT threshold [2].
3.1. Projected limits
Moving to an upgraded LHC increases EFT limits by about a factor of 2 with the first year of luminosity, as seen in Figure 1. Another factor of 2 is expected after the full HL-LHC run, as per Figure 2.
3.2. Discovery potential
The upgraded LHC provides a 5σDM discovery po- tential of 1.5 TeV after one year of data [2]. An ultimate reach of 2.6 TeV is expected, as per Figure 3.
3.3. Simplified models
After one year of data, the upgraded LHC increases sensitivity to the pair-production of DM via aZ0 me- diator by an average of a factor of 2 [2], similar to the gain seen for the EFT model. The resulting limits on the interaction scale can be seen in Figure 4.
threshold [GeV]
miss
ET
[GeV] *Suppression Scale M
0 1000 2000 3000 4000 5000
3000 fb-1
300 fb-1 mχ=50GeV
Simulation Preliminary ATLAS
=14 TeV s 1% syst
400 600 800
< 4π gDM gSM <
π
Figure 2: Expected sensitivity with a flat 1% systematic for 50 GeV DM after the full LHC and HL-LHC programs [2].
[TeV]
M*
1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 3.2
]σsignificance [
0 2 4 6 8 10 12 14 16 18 20
discovery σ 5
evidence σ 3
= 50 GeV D5, mχ
Simulation Preliminary ATLAS
=14 TeV
s
∫
Ldt=3000fb-1 π < 4 gDM gSM <π
5% systematic 1% systematic
Figure 3: Expected discovery potential for 50 GeV DM after the full HL-LHC run, with both 1% and 5% flat systematics [2].
[TeV]
Mmed
10-1 1 10
M* [TeV]
0 1 2 3 4 5
0.1 0.2
0.5 1
2
5
4π
=50GeV mχ
=400GeV mχ med/3
Γ=M
=50GeV, mχ
π med/8 Γ=M
=50GeV, mχ
med/3 Γ=M
=400GeV, mχ
/8π
=Mmed
=400GeV, Γ mχ
contours gχ gq
non-perturbative regime EFT limits
>400 GeV
miss
=14TeV, ET
s
Simulation Preliminary ATLAS
Ldt=25fb-1
∫
Figure 4: Expected sensitivity to aZ0simplified model of DM pair- production, after one year of 14 TeV data with 5% systematics [2].
Steven Schramm (on behalf of the ATLAS Collaboration)/Nuclear Physics B Proceedings Supplement 00 (2014) 1–3 3
gDM
gSM
1 1.5 2 2.5 3
[%]tot medMR
0 20 40 60 80 100
> 400 GeV miss ET
> 600 GeV miss ET
> 800 GeV miss ET
ATLASSimulation Preliminary
= 14 TeV
s -1
L = 25 fb
∫
= 50 GeV mχ
Figure 5: Fraction of valid EFT events expected after one year of 14 TeV data with a flat 5% systematic for 50 GeV DM [2].
4. Model validity
The validity of the EFT model used has recently been called into question [7, 8, 9]. For an EFT to be valid, the momentrum transfer Qtrmust be much less than the mass of the mediator Mmed, thus forbidding on-shell production. While the assumption of an EFT makes sense at direct detection experiments, it is less clear that this holds in energetic LHC collisions. A minimal va- lidity constraint for the D5 operator was imposed, where gSM(gDM) is the coupling of the integrated-out mediator to SM (DM) particles, and M∗is the suppression scale:
Qtr <Mmed= √
gSMgDMM∗
Without knowledge of the complete theory, the cou- pling values remain unknown. In order to determine the dependence of the fraction of valid events (RtotM
med) passing this criterion, a scan over the couplings is per- formed, as shown for 50 GeV DM in Figure 5. The va- lidity of this scenario is seen to be maximal for coupling values of √
gSMgDM & 2, thus clarifying the valid re-
gions of parameter space. The nominal limits M∗expare then rescaled by this validity fraction as detailed in Ref- erence [2], giving valid limits M∗valid. The rescaled limits for 50 GeV DM are shown in Figure 6 with the nominal limits overlaid for comparison.
Note that there is a strong dependence of the momen- tum transfer and thus the validity on the mass of the DM, where higher masses are less valid [2].
5. Conclusions
These studies highlight the importance of the first year of data to the sensitivity of monojet searches for DM at the upgraded LHC. Within the first year, the ex- pected improvement in the sensitivity with respect to
gDM
gSM
1 1.5 2 2.5 3
M* [TeV]
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
alid M*v > 400 GeV miss
ET M*exp
> 600 GeV miss ET
> 800 GeV miss ET
ATLASSimulation Preliminary
= 14 TeV
s -1
L = 25 fb
∫
= 50 GeV mχ
Figure 6: Expected EFT limits after one year of 14 TeV data with a flat 5% systematic for 50 GeV DM. Dashed lines are nominal limits, solid lines are rescaled to account for the fraction of valid events [2].
8 TeV is about a factor of 2 for both EFTs and sim- plified models. The following years to the end of the HL-LHC program are expected to gain another factor of 2, although further studies of higher EmissT thresholds may improve the sensitivity.
The validity of the EFT models typically used in monojet searches has also been studied for the D5 op- erator, and has been found to be partially valid for low masses and more questionable for heavier masses. In all cases, the unknown couplings between the mediator and both the Standard Model and DM particles plays a large part in the validity. A more comprehensive look into the remaining operators has been left to future studies.
References
[1] ATLAS Collaboration, The ATLAS Experiment at the CERN Large Hadron Collider, JINST 3 (2008) S08003.
[2] ATLAS Collaboration, Sensitivity to WIMP Dark Matter in the Final States Containing Jets and Missing Transverse Momentum with the ATLAS Detector at 14 TeV LHC (ATL-PHYS-PUB- 2014-007, http://cdsweb.cern.ch/record/1708859).
[3] Agostinelli, S., et al., GEANT4: A simulation toolkit, Nucl. In- strum. Meth. A506.
[4] J. Goodman, et al., Constraints on Dark Matter from Colliders, Phys.Rev. D82 (2010) 116010.
[5] P. J. Fox, et al., Missing Energy Signatures of Dark Matter at the LHC, Phys.Rev. D85 (2012) 056011.
[6] ATLAS Collaboration, Search for New Phenomena in Monojet plus Missing Transverse Momentum Final States using 10fb−1 of pp Collisions at √
s=8 TeV with the ATLAS detector at the LHC (ATLAS-CONF-2012-147, https://cds.cern.ch/record/1493486).
[7] G. Busoni, et al., On the Validity of the Effective Field Theory for Dark Matter Searches at the LHC, Phys. Lett. B728.
[8] G. Busoni, et al., On the Validity of the Effective Field Theory for Dark Matter Searches at the LHC, Part II: Complete Analysis for thes-channel, JCAP 1406 (2014) 060.
[9] G. Busoni, et al., On the Validity of the Effective Field Theory for Dark Matter Searches at the LHC Part III: Analysis for the t-channel, arXiv:1405.3101.