1. PADME project at DA  NE BTF Mauro Raggi Laboratori Nazionali di Frascati PADME kick-off Meeting Frascati, 20-21 April 2015 PADME website http://www.lnf.infn.it/acceleratori/padme/index.html

2. Outline  P ositron A nnihilation into D ark M atter E xperiment PADME proposal  Beam conditions and the Target  The electromagnetic calorimeter  The dipole magnet  The spectrometer  The Vacuum chamber  Analysis technique for annihilation production  Signal selection criteria  Positron flux measurement  Limit evaluation  Conclusion and prospects M. Raggi PADME kick-off MeeringFrascati 20-21/04/152

3.  The simplest hidden sector model just introduces one extra U(1) gauge symmetry and a corresponding gauge boson: the “dark photon" or U boson.  Two type of interactions with SM particles should be considered  As in QED, this will generate new interactions of the type:  Not all the SM particles need to be charged under this new symmetry  In the most general case q f is different in between leptons and quarks and can even be 0 for quarks. ( P. Fayet, Phys. Lett. B 675, 267 (2009).)  The coupling constant and the charges can be generated effectively through the kinetic mixing between the QED and the new U(1) gauge bosons  In this case q f is just proportional to electric charge and it is equal for both quarks and leptons. The simplest dark sector model M. Raggi PADME kick-off Meering A’  ee ee Frascati 20-21/04/153

4. Dark photon searches in the world M. Raggi PADME kick-off Meering Status: publishing, approved, proposals BelleII SuperKEKB BaBar @ PEP-II JLAB: APEX HPS, DarkLight BDX A1 @ MAMI WASA @ COSY HADES @ GSI KLOE2 @ DAFNE PADME @ BTF VEPP-III ATLAS, CMS @ LHC P-348 @ SPS SHIP @ SPS NA48/2 Mu3e @PSI Cornell Phenix @RHIC Frascati 20-21/04/154

5. Dark photon searches status  Favored parameters values explaining g-2 (green band)  A’-boson light 10-100 MeV  Status of dark photon searches  Beam dump experiments (grey)  e + e  appearance after a dump  Fixed target  Peak search over QED BG  Mesons decays  Peaks in M(e + e - ) or M(  +  )  Indirect exclusion from g e -2 g  -2  Recent tight limit in red filled area M. Raggi PADME kick-off Meering  Many different techniques, assumptions on dark photon interaction models  Kinetic mixing, decay to electrons, no dark sector particles arXiv:1412.0018v2 Frascati 20-21/04/155

6. Status  q ≠0 and A’→e + e  M. Raggi PADME kick-off Meering g-2 muon band excluded by recent NA48/2 measurement Frascati 20-21/04/156

7. Status A’→  decays M. Raggi PADME kick-off MeeringFrascati 20-21/04/157  At present there are no MI experimental limit for the A’  invisible decay  Just a never published ArXiv 0808.0017 by Babar ‘08 with very limited sensitivity on  2 (  3S   U assumes coupling to quarks!)  Indirect limit from K + →   (assumes coupling to quarks!) arXiv:1309.5084v1 Babar ’08 ArXiv 0808.0017

8. Why dark photon invisible decays? M. Raggi PADME kick-off Meering  The invisible search technique remove any assumption except coupling to leptons  A’ increase its capability of having escaped detection so far  No data in the minimal assumptions  Even if pure leptophilic models are less natural W. J. Marciano et al. “If, instead, the A’ decays primarily into invisible light particles (e.g. a pair of dark matter particles with mass < m A’ /2), that change would essentially negate all the bounds except those coming from anomalous magnetic moments” W. J. Marciano et al. arXiv:1402.3620v2 Frascati 20-21/04/158 PhysRevD.89.095006

9. Invisible dark photon and kaons M. Raggi PADME kick-off Meering In models assuming that the dark photon couples to SM through kinetic mixing  q ≠0 K ± →  ± can be used to constrain K ± →  ±  ’ Depending on how the model is build the limit can change significantly for example allowing the coupling to Z. Z d =A’ for Marciano! Frascati 20-21/04/159 PhysRevD.89.095006

10. The PADME approach  At present all experimental results rely on at least one of the following model dependent assumptions:  A’ decays to e + e  (visible decay assumption BR(A’→e + e  = 1)  A’ couples with the same strength to all fermions (  q =  l ) (kinetic mixing)  In the most general scenario (PADME)  A’ can decay to dark sector particles lighter than the A’ BR(A’→e + e  <<1)  Dump and meson decay experiment only limit  2 BR(A’→e + e  <<1)  A’ can couple to quark with a coupling constant smaller  l or even 0  Suppressed or no production at hadronic machines and in mesons decays  PADME aims to detect A’ produced in e + e  annihilation and decaying into invisible by searching for missing mass in e + e  →  ’  A’→   No assumption on the A’ decays products and coupling to quarks  Only minimal assumption: A’ bosons couples to leptons  PAMDE will limits the coupling of any new light particle produced in e + e  collision (scalars (H d ), vectors (A’ and Z d )) M. Raggi PADME kick-off MeeringFrascati 20-21/04/1510

11. DA  NE Beam Test Facility (BTF) M. Raggi PADME kick-off Meering NIM A 515 (2003) 524–542 electronspositrons Maximum energy [MeV]750 (1050) MeV550 (800) MeV Linac energy spread0.5%1% Typical Charge [nC]2 nC0.85 nC Bunch length [ns]1.5 - 40 Linac Repetition rate1-50 Hz Typical emittance [mm mrad]1~10 Beam spot  [mm] 1 mm Beam divergence1-1.5 mrad  Longer Duty Cycle  Standard BTF duty cycle = 50*10 ns = 5x10 -7 s  Already obtained upgrade 50*40ns= 20x10 -7 s work in progress to exceed 100 ns  Energy upgrade possible in 2017.  The accessible M A’ region is limited by beam energy  Region from 0-22 MeV can be explored with 550 MeV e + beam Frascati 20-21/04/1511

12. BTF beam summary  Energy spread  p/p ~1%  Beam spot: 1 – 2 mm RMS  Divergence: 1 – 1.5 mrad  Effect of multiple scattering and Bremsstrahlung on the Beryllium exit window and in air has to be considered  Both size and divergence depend on the optics  Beam position: 0.25 mm RMS  Pulse duration: 1.5 – 40 ns  10 ns during collider operations Frascati 20-21/04/15M. Raggi PADME kick-off Meering12 Beam E spread Measurement of the beam E spread Nucl. Instrum. Meth. A718 (2013) 107–109 Beam spot size Beam spot center

13. The PADME experiment M. Raggi PADME kick-off Meering 1.75 m  10 3 -10 4 e + on target per bunch at 50 bunch/s (10 13 -10 14 e + /year)  Basic detector components:  Active 50  m diamond target  GEM based magnetic spectrometer ~1m length  Conventional 0.6T magnet  15 cm radius cylindrical crystal calorimeter with 1x1x20 cm 3 crystals Frascati 20-21/04/1513

14. The PADME experiment M. Raggi PADME kick-off Meering By C. Capoccia LNF SPAS Frascati 20-21/04/1514

15. The PADME diamond target M. Raggi PADME kick-off Meering  First BTF test-beam with polycrystalline diamonds: 1. Two 500  m thick and 4 metal strips: 6.5 mm long and 1.5 mm pitch 2. 300  m thick 40 graphitized strips 3 mm long, 100  m width, and 170  m pitch 3. 50  m thick, 2×2cm 2 sample for first PADME prototype 4. 50  m thick 5×5mm 2 sample for BTF beam diagnostics with Silver Paint 1. 2. 3. 4. 50  m, silver painted, Estimated CCD=10-20  m Main result of feasibility of 50  m sensors already established Frascati 20-21/04/1515

16. A possible analyzing magnet for PADME M. Raggi PADME kick-off Meering 52 cm 116 cm 11 to 20 cm gap Frascati 20-21/04/1516 Available at CERN magnet division

17. A possible analyzing magnet for PADME M. Raggi PADME kick-off Meering Tapered poles 42 kW 16 kW 95 kW Frascati 20-21/04/1517

18. PADME vacuum vessel study M. Raggi PADME kick-off Meering Frascati, Servizio Vuoto V. Lollo, S. Bini Al 2219 T851 or AL6082 T6 2 mm side walls 4 mm ribs Different possibilities under study to minimize the material thickness C. Capoccia LNF SPAS Frascati 20-21/04/1518

19. PAMDE spectrometer M. Raggi PADME kick-off Meering There is the possibility of having a spectrometer outside the vacuum: Impact on the PADME-visible experiment to be understood Inside Outside vacuum vessel trackers vacuum vessel trackers Frascati 20-21/04/1519 In what follow we will use a simplified version of the spectrometer just made of scintillators that is not used for measuring the momentum

20. The electromagnetic calorimeter  Cylindrical shape: radius 150 mm, depth of 200 mm  Inner hole 4 cm radius  Active volume 13120 cm 3 total of 656 crystals 10x10x200 cm 3  Material LSO(Ce): high LY, high  small X 0 and R M, short  decay  Material BGO: high LY, high  small X 0 and R M, long  decay, (free form L3?)  Expected performance:   (E)/E =1.1%/√E 0.4%/E 1.2% superB calorimeter test at BTF [NIM A 718 (2013) 107–109]   (  ) = 3 mm/1.75 m < 2 mrad  Angular acceptance 1.5-5 degrees M. Raggi PADME kick-off Meering 30 cm Frascati 20-21/04/1520

21. PADME geant4 simulation M. Raggi PADME kick-off MeeringFrascati 20-21/04/1521

22. Search in annihilation production M. Raggi PADME kick-off MeeringFrascati 20-21/04/1522

23. Experimental technique  Search for the process: e + e  →  A’ on target e  at rest electrons  (10 4 550 MeV e + )/bunch beam on a 50  m diamond target with 50 bunch/s  Collect 4x10 13 e+ on target in each year of data taking period at BTF (60% eff.)  Measure in the ECal the E  and  angle wrt to beam direction  Compute the M miss 2 = (P 4 e   P 4 beam  P 4  ) 2  P 4 e   =(0,0,0,m e ) and P 4 beam  =(0,0,550,sqrt(550 2 + m e 2 ) ) M. Raggi PADME kick-off Meering  A’ 550 MeV e + ECal P 4 beam Spectrometer C Target M miss 2 Spectrometer Frascati 20-21/04/1523

24. Invisible signal selection Selection cuts  Only one cluster in Ecal  Rejects e + e  →  e + e  →  final states  5 cm < R Cl < 13 cm  Improve shower containment  (E)/E  Positron veto: no tracks in the spectrometer in ±2 ns  Reject BG from bremsstrahlung identifying primary positrons  Photon veto: no g with E  >50 MeV in time in ±1ns in the SAC  Cluster energy within: E min (M A’ ) < E Cl < E max (M A’ ) MeV  Removes low energy bremsstrahlung photons and piled up clusters  Missing mass in the region: M miss 2 ±  M miss 2 M. Raggi PADME kick-off MeeringFrascati 20-21/04/1524

25. Main background sources  Geant4 simulation accounts for:  Bremsstrahlung, 2 photon annihilation, Ionization processes, Bhabha and Moller scattering, and production ofδ-rays.  Custom treatment of e + e  →  (  ) using CalcHep generator. M. Raggi PADME kick-off Meering ee ee A’  Frascati 20-21/04/1525 Signal: e + e    +missing mass (A’)   ee ee ee ee   ee ee    +1 electron +1  +2  Backgrounds

26. Bremsstrahlung background M. Raggi PADME kick-off MeeringFrascati 20-21/04/1526 Present inefficiency is due to particles remaining into the beam spot due to low energy of the radiated gamma. Improving with an imaging veto?

27. Rejecting 3  veto Frascati 20-21/04/15M. Raggi PADME kick-off Meering27 Low mass region is less suppressed due to the request of ESAC>50MeV Are the missing photon at large angle? Can we have a veto also there? 3  no veto 3  veto 3  background

28. The  normalization selection  Number of calorimeter clusters = 2  Cluster energy: 100MeV<E cl <400 MeV  Cluster radial position 5 cm <R Cl < 13 cm   invariant mass 20 MeV < M  < 26 MeV M. Raggi PADME kick-off MeeringFrascati 20-21/04/1528 Used to measure the beam flux, in alternative to the diamond

29. PADME invisible sensitivity estimate  Based on 2.5x10 10 fully GEANT4 simulated e + on target events  Number of BG events is extrapolated to 1x10 13 electrons on target  Using N(A’  )=  (N BG )   enhancement factor  M A’  A’  )/  ) with  =1 M. Raggi PADME kick-off Meering M. Raggi, V. Kozhuharov Advances in High Energy Physics Vol. 2014 ID 959802, Frascati 20-21/04/1529 PADME invisible exclusion

30. Limiting factors to sensitivity  The limiting factor to the present sensitivity are:  The 50Hz repetition rate of the linac (coupling)  The energy of the linac (mass)  The efficiency of the positron veto due to beam spot and energy spread  Gamma veto for low energy.  Possible improvements  Smaller beam energy spread (0.25% is not out of reach)  Smaller spot size (0.5mm is in within reach)  Energy up to 1GeV is technically possible Frascati 20-21/04/15M. Raggi PADME kick-off Meering30

31. PADME project plans  Project has been presented as a “What Next” Project in INFN CSN1  The project has received positive comments form CSN1 referees  Proposal for R&D financing will be discussed in the July CSN1 meeting  Proto collaboration formed including  LNF, Rome1, Lecce and Sofia university  6 weeks test beam time asked at DA  NE BTF in 2015  Study the prototype of BGO calorimeter solution (L3 crystals)  Test diamond target prototypes  Study the maximum beam current per bunch and beam spot size  Optimize beam characteristics for PADME operation bunch length, number of particle per bunch, background, beam positioning stability M. Raggi PADME kick-off MeeringFrascati 20-21/04/1531

32. Conclusions and plans  An experiment running at DA  NE BTF sensitive to both A’→invisible and A’→e + e  decays has been proposed to INFN CSN1  Exclusion limit in  2 down to 1-2 ・ 10 -6 can be achieved in invisible decays with the present BTF beam parameters in the region M A’ 2- 22 MeV (28 with e + energy 750 MeV)  M. Raggi and V. Kozhuharov, Advances in High Energy Physics Vol. 2014 ID 959802,  Possible reduction of beam energy spread 0.25% beam spot to 0.5x0.5 mm 2 and bunch width to the level of 500 ns will be desirable to back ground reduction and  2 sensitivity region improvements. M. Raggi PADME kick-off MeeringFrascati 20-21/04/1532

33. 33 Welcome to the dark side

34. SPARE SLIDES M. Raggi PADME kick-off MeeringFrascati 20-21/04/1534

35. MC calorimeter performance  Missing mass resolution in agreement with toy MC using   (E)/E =1.1%/√E 0.4%/E 1.2% [NIM A 718 (2013) 107–109]  Differences are ~ 10%  Resolution is the result of combination of angular resolution energy resolution and angle energy correlation due to production M. Raggi PADME kick-off Meering Mmiss 2 (MeV) Frascati 20-21/04/1535

36. Possible BTF upgrades  Energy upgrades up to 1.2 GeV electrons  Proposal to reach >800 MeV energy for positrons (see V. Buonomo BTF user workshop)  Longer Duty Cycle  Standard BTF duty cycle = 50*10 ns = 5x10 -7 s  Already obtained upgrade 50*40ns= 20x10 -7 s (Thanks to BTF team)  Any increase of duty cycle increase linearly experiment statistics  Collimation system  Assure better beam definition for positrons beam  Maximum current in BTF hall  Limited by radio protection to 6.2x10 8 per bunch for long term operation  Can reach >3x10 10 particle per bunch after proper screening M. Raggi PADME kick-off Meering See recent BTF user workshop for details at: https://agenda.infn.it/conferenceOtherViews.py?view=standard&confId=7359 Frascati 20-21/04/1536

37. Indirect limits M. Raggi PADME kick-off Meering Phys.Rev.D80:095002,2009 Phys.Rev.Lett.106:080801,2011 α −1 = 137.035 999 037 (91) However this is based on a single measurement with drastically improved precision |a e th – a e exp | = (1.06 ± 0.82) × 10 −12 2  limit from new  Phys Rev D 86, 095029 (2012) Frascati 20-21/04/1537

38. Cross section enhancement M. Raggi PADME kick-off MeeringFrascati 20-21/04/1538