The milliQan experiment: search for milli-charged particles at the LHC
aa r X i v : . [ phy s i c s . i n s - d e t ] A ug The milliQan experiment: search for milli-chargedparticles at the LHC
Haitham Zaraket ∗† Lebanese University, Faculty of Sciences, Multidisciplinary Physics LabE-mail: [email protected]
Charge quantization has always been enigmatic. Theoretically, Millicharged particles can be ananswer. The use of existing detectors, without affecting their initial mandate, is a very promisinglow cost new physics detector for millicharged particles. The Milliqan collaboration has installeda 1/100th version of the full detector at LHC point 5 in CMS. Data collected in 2018 is now underinvestigation/analysis. Some results are presented in this paper.
European Physical Society Conference on High Energy Physics - EPS-HEP2019 -10-17 July, 2019Ghent, Belgium ∗ Speaker. † Work partially supported by the Lebanese University Grant. On behalf of the MilliQan Collaboration. c (cid:13) Copyright owned by the author(s) under the terms of the Creative CommonsAttribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). https://pos.sissa.it/ illiqan update
Haitham Zaraket
1. Why milli-charged?
Charge quantization is a remarkable feature of the standard model (SM). Stable particles withcharge < 0.3e (quarks already known to have charges of 1/3e and 2/3e) are not part of the predictedand detected flora of particles in the SM. Millicharge particles, if they exist, are invisible to currentLHC detection system due to their low charge. The low charge implies low energy deposits (singlephotons) in exiting particle detectors and are highly affected by particle background.The search for millicharged particle is related to complementary searches at LHC (not coveredby ATLAS/CMS searches) for hidden/dark sector. For example massless dark photons are relatedin several models to the existence of such millicharged particles and can have distinctive signatureat the LHC. From the astrophysical side, recent excitement has shown that milli-charged darkmatter at the 1% mass-density level is a leading explanation [1] for the EDGES 21-cm result [2].Constraints from astrophysical observation on the small couplings to the Standard Model can beused to guide our search for millicharged particles (for instance [3]).Millicharged particles can arise in several models. Among the simplest scenarios one canimagine, a simple extension to the Standard Model can be made by just adding a new U(1) gaugesymmetry. The coupling to the standard model can then be made by adding a kinetic mixing termamong the gauge invariant field tensors F µν (for U EM ( ) ) and B µν (for the additional U ′ ( ) ) L = L SM − κ B µν F µν − B µν B µν . (1.1)A new fermion fermion is then added with coupling to U ′ ( ) : L = L SM − κ B µν F µν − B µν B µν + ¯ ψ (cid:0) i / ∂ − e ′ / B − m (cid:1) ψ . (1.2)The above theory is equivalent to the theory where we can make the change of field B µ → B µ + κ A µ . Hence eliminating the coupling between the photon field A µ and the dark photon ( B µ ), butgenerating a new charge to the new fermion κ e ′ (the millicharged particle mCP): L = L SM − B µν B µν + ¯ ψ (cid:0) i / ∂ − κ e ′ / A − e ′ / B − m (cid:1) ψ . (1.3)Models with additional symmetry groups can also be used to generate millicharged particles. Ad-ditional massive bosons ( Z ′ ) can be considered beside dark photons.The main millicharged particle production mechanisms are QCD inspired processes. In ppcollisions, besides Drell-Yan, we can list processes with η , η ′ , ρ and J / ψ decay. The detectioncan be either through the decay of the mCP or by scattering, energy loss.
2. Detector proposal
There is a long history of direct and indirect searches for mCPs. For mCPs with mass belowthat of the electron one finds strong bounds from astrophysical observations and cosmologicalmodels. Laboratory tests can be highlighted ranging from indirect decay of ortho-positronium tothat of accelerator beam experiment (SLAC for instance [4]).The challenge in detecting a low charge particle comes from the fact that we have a lowerionization energy. A detector with large depth is needed for the particle to traverse. Segmentation1 illiqan update
Haitham Zaraket is needed to be sure that the passing particle is an ionizing particle. The idea of the MilliQandetector came in 2014 [5] to add detector sensitive to milli-charged particles produced in LHCcollisions with: • a charge down to 10 − e, hence an energy loss of 10 − lower than that of an e-charged particleis expected. Hence the need for long, sensitive active length to see a single photo-electron. • The proposal was a 1 m x 1 m x 3 m scintillator plus photomultiplier Tube (PMT) array,pointing back to interaction point, in well shielded area near CMS or ATLAS. • With triple coincidence, dominant random background should be controlled.In 2016 a Letter Of Intent (LOI) was published [6] identifying the location in CMS point 5 and afirst Full detector simulation was made.
3. Status of Demonstrator
In the fall of 2017 a 1% scale "demonstrator" of the proposed detector was installed in CMSdrainage gallery at 33 m from the CMS IP. The main mission of the demonstrator was to studythe feasibility of the experiment, focusing on understanding various background sources such asradioactivity of materials, PMT dark current (main challenge for single PE), cosmic rays, and beaminduced backgrounds. Being around 70 m underground a natural, but not, complete shield fromcosmic muons is expected. The expected sensitivity is shown in Figure 1. An updated sensitivitycurve taking into account a realistic model of the detector and its installation position besides fullsimulation of background is under investigation by the collaboration.
Figure 1:
The simulated sensitivity with 15 ns coincidence interval with three layers of the detector.
The initial design respected the LOI proposal by implementing three layers of 2x3 scintillatorplus PMT. Several PMT species were used. Events are triggered if there is a signal above the triggerthreshold in three trigger groups within a window of 100 ns. Each trigger group contains twochannels and the trigger group for each channel is given by floor(channel/2). The channel mappingensures adjacent bars are in the same trigger group. The trigger thresholds varied during data taking2 illiqan update
Haitham Zaraket to satisfy rate requirements. Scintillator slabs and lead bricks were added to analyze backgroundand Tag through-going particles or to shield radiation. There are also several hodoscope packscomposed of small arrays of 0 . ×
18 inch rectangular pieces of plastic scintillator readout viaSiPMs attached at one end. These provide finer grained position information that allows crudetracking through the device (see Figure 2). As a sign of readiness for full detector installation the
Figure 2:
The initial installation of the 1% of the MilliQan detector. used support structure was designed to hold the full detector. Besides, a full mechanical designwas made with a modular strategy (see Figure 3) to simplify the upgrade and maintenance of thedetector, keeping our promise of not affecting any activity of CMS.
Figure 3:
The proposed Modular mechanical design of the full MilliQan detector.
4. Highlights from Demonstrator Data
In this section we give some highlights of the analysis of the data taken during the last year.3 illiqan update
Haitham Zaraket
Full analysis of the result will be published later by the collaboration.In 2018 run a very successful 37 fb − was done with around more than 2000 h trigger live-time. A very valuable experience in operating the detector was performed. In December 2018 twoadditional channels were added to the design as a fourth layer. The run and collected data wasshown to be a powerful resource to study and optimize the performance of final detector. A firsttest for the correct alignment of the detector was made by comparing the rate of through-goingparticles during a fill and comparing it with the luminosity time constant (14 h in this case) asshown in Figure 4. Figure 4:
The luminosity measured by CMS compared to the rate of through-going particles through theMilliQan detector.
Cosmic Muons and Muons from CMS were detected and the data taken was validated withthe simulated GEANT efficiency. An in-situ N PE per charge calibration was done. It should bementioned that the calibration of the N PE per unit charge incident on the detector is crucial fordetermining the lowest charge to which milliQan can be sensitive. To achieve this, we first computethe number of PEs for cosmic muons incident on the demonstrator. The value for N PE is extractedby dividing the pulse area of cosmic muons by the pulse area of a single PE obtained from delayedscintillation PEs. The method of using delayed scintillation PEs to measure the SPE response wasvalidated using an LED bench measurement. Extrapolating the N PE to fractional charges by scalingby Q an estimation of N PE = Q ∼ × − e . Triggering is another importantmission of the prototype to be tested and validated.The total background as a function of the minimal N PE per charge in the event can be evaluatedusing data taking periods in which there are no collisions and scaled to the total length of thedata taking with collisions. To study further correlated backgrounds we used the ABCD method[A: pointing path with coincidence less than a threshold (15 ns or 30 ns), B: non pointing pathwith coincidence less than the threshold, C: non pointing path with coincidence greater than thethreshold, D: pointing path with coincidence greater than the threshold). As can be shown inthe closure test shown in Figure 5 perfect matching is observed between data and simulation. InDecember 2018 a fourth layer was added to the installed demonstrator. The data and the simulation4 illiqan update Haitham Zaraket has shown a reduction of the background by two orders of magnitude compared to three layersdesign as can be seen in Figure 5.
Highlights from Demonstrator Data “see” muons from CMS IP
Validated GEANT efficiency in situ N PE /Q calibration from cosmics Further validated simulation
Established ABCD method to estimate backgrounds
Observed correlated backgrounds
Changed to 4-layer design
Reduced background by ~2 orders magnitude compared to 3-layers
Event display of a muon from CMS PE SPE
100 200 300 400 500 600 70020406080100120140
Fill Lumi [1/pb] N t h r ou g h g o i n g p a r t i c l e s Measured rate — -1 Rate from simulation — -1
10 min nPE − − − − E v en t s / h r Figure 5:
The ABCD closure test with three and four layers comparing data and simulation.
5. Summary
The search for millicharged particles using existing accelerators facilities represents a verypromising path. The feasibility prototype for MilliQan (the demonstrator) has operate successfullythe last year with very positive signs for full detector installation. With the data collected triggering,time and charge calibration were possible. A clear understanding of background was performed.
References [1] Julian B. Munoz, Abraham Loeb,
Insights on Dark Matter from Hydrogen during Cosmic Dawn ,Nature (2018) 557, 684 [arXiv:1802.10094v2].[2] J. D. Bowman, A. E. E. Rogers, R. A. Monsalve, T. J. Mozdzen and N. Mahesh,
An absorption profilecentred at 78 megahertz in the sky-averaged spectrum , Nature (2018) 555, 67-70.[3] Jae Hyeok Chang, Rouven Essig, Samuel D. McDermott,
Supernova 1987A constraints on sub-GeVdark sectors, millicharged particles, the QCD axion, and an axion-like particle , J. High Energ. Phys.(2018) 2018: 51.[4] A. A. Prin et al.,
Search for Millicharged Particles at SLAC , Phys. Rev. Lett. (1998) 81, 1175-1178.[5] Andrew Haas, Christopher S. Hill, Eder Izaguirre, Itay Yavin,
Looking for milli-charged particleswith a new experiment at the LHC , Physics Letters B (2015), 746, 117-120 [arXiv:1410.6816].[6] Austin Ball et al,
A Letter of Intent to Install a milli-charged Particle Detector at LHC P5 , (2016)[arXiV:1607.04669]., (2016)[arXiV:1607.04669].