1. John Jaros for the Heavy Photon Search Collaboration Dark Interactions June 11, 2014 HPS: The Heavy Photon Search Experiment at Jefferson Laboratory John Jaros for the HPS Collaboration Dark Interactions BNL June 11, 2014

2. A new U(1) gauge boson A’ may mediate Dark Interactions * New U(1)’s are expected in many BSM theories * Kinetic mixing with the SM photon couples the A’ to SM hypercharge * Heavy photons could explain experimental anomalies in particle physics and astrophysics * A’ is characterized by its mass m A’, coupling to charge  e, and coupling to dark charge,  D. HPS is a new, special purpose experiment at JLAB which will search for heavy photons over a wide range of masses and couplings in unexplored parameter space. Dark Interactions with Heavy Photons HPS HPS Dark Interactions 20142

3. First constraints on hidden sector photons followed from the re- interpretation of beam dump experiments and the measurements of the electron and muon anomalous magnetic moments. The first new searches for Heavy Photons have employed existing data sets, at the e + e - colliders with KLOE and BABAR and elsewhere. Bjorken, Essig, Schuster, and Toro (BEST) taught us: The first new experiments searching for Heavy Photons, at Mainz and JLAB, have used existing two-arm spectrometers. HPS is a new, special purpose experiment, dedicated to searching for an A’. We chose to go after the unique territory with  << 10 -3 which is accessible with a vertex detector. BEST’s suggestion: “Fixed-target experiments are ideally suited for discovering new MeV-GeV mass U(1) gauge bosons through their kinetic mixing with the photon” HPS and the Search for Heavy Photons Phys. Rev. D80, 2009,075018 3

4. Searching for an A’ with Small Couplings Small couplings mean very few events. Need lots of luminosity. Lots of lum means lots of background, low S/B QED tridents, an irreducible physics background, overwhelm A’ production. But small couplings also make the A’ long-lived. A secondary vertex signature powerfully discriminates against the prompt trident background. It’s all in the tails. The A’ decay length signal is in the tails of the prompt trident signal. HPS must understand and control the tails of the trident vertex distribution. Full simulation confirms this is possible. HPS Dark Interactions 2014 Decay Length c  Vertex Distribution Tridents 10–15 mm cut 4

5. A’ kinematics  very forward production HPS Design Choices HPS Dark Interactions 2014 E A’  E beam  A’  0  decay = m A’ /E A’ Want  m/m ~ 1% for bump hunt Want  z ~ 1mm Vertexing A’ decays requires detectors close to the target. Invariant mass is an essential signature, so good momentum/mass resolution is also required. Vertexing and bump hunting need tracking and a magnet. Trigger with a high rate, rad hard EM Calorimeter Placed downstream of the magnet, it can ID e + and e -. HPS opts for large forward acceptance/moderate currents. This requires placing sensors as close as possible to the beam. e + and e - Entering ECal Beam’s Eye View 5

6. Controlling Beam Backgrounds With sensors close to the beam (just ½ mm for the first Si sensor ), background control, radiation damage, and beam stability become critical. Constraints * Avoid Multiple Coulomb Scattered (MCS) beam (the background for HPS) * Avoid the “sheet of flame”, the beam electrons which have radiated, lost energy, and been deflected in the horizontal plane by the magnet * Avoid beam gas interactions. * Avoid errant beam motions. Design Solutions * Split the detectors top-bottom to avoid the beam and the “sheet of flame” * Run the tracker in vacuum to eliminate beam gas interactions * Tightly collimate the incident beam. photons Top View: Analyzing Magnet MCS beam e- “sheet of flame” target B  HPS Dark Interactions 2014 6

7. Small A’ cross-sections need high luminosity. High luminosity generates huge numbers of tridents and accidentals Spread the data collection out maximally in time * Maximize accelerator duty cycle CEBAF12 runs at 500 MHz! Go Fast! * Choose rad hard detectors with short response times Pulse lengths in the SVT and Ecal are ~ 60ns * Use high rate DAQ and high rate Trigger SVT readout at 40 MHz (APV25 + SLAC ATCA) Ecal readout at 250 MHz (JLAB FADC250) 50 kHz Trigger CEBAF12 E = up to 12 GeV High currents  100  A Continuous! 500 MHz  (e - + W  W + A’ + e - ) 7

8. HPS Concept Magnet chicane in Hall B at JLAB. Dipole analyzing magnet in the middle. 6-layer Silicon Vertex Tracker, split top-bottom and residing in vacuum, measures momentum and decay vertices. 442 crystal PbWO 4 electromagnetic calorimeter, also split top-bottom, sits behind the tracker, triggers on e+e- pairs, and identifies electrons. HPS Dark Interactions 2014 1 m e-e- Target Si Vertex Tracker ECal 8

9. HPS Apparatus: Beamline HPS will reside in the Hall B alcove, directly behind the general purpose CLAS12 detector, and before the Hall B dump. CEBAF6 beams were very small (  y < 40  m), stable, with very little halo. CEBAF12 beams for HPS are expected to be the same. An upstream protection collimator, with a 3 mm vertical aperture, will protect the silicon from any beam motion associated with beam trips. HPS Dark Interactions 2014  y =20  m halo<10 -5 Hall B Alcove Beam Scan 9

10. HPS Apparatus: SVT HPS Dark Interactions 2014 SVT Design: Six layers of Si modules, split top-bottom, each with two sensors, one axial and the other at small angle stereo, in vacuum. Each sensor is a 4x10 cm Hamamatsu microstrip detector with 60  m sense pitch. Fast Readout: CMS APV25 chip provides 40 MHz continuous amplitude sampling, with 3  sec latency. Digitizing electronics and power distribution also reside in vacuum. Power and control in/data out through vacuum feedthroughs. Electronics and sensors are cooled < 0 0 C to remove heat and boost radiation hardness. Precision Movers position layers 1-3 close to the beam, do wire scans, and insert targets as needed. Beam’s Eye View positioning lever electronics Cooling Movable target 10

11. HPS Apparatus: SVT 2012 Test Run at JLAB provided proof of principle Track reconstruction, good momentum resolution, and vertexing demonstrated HPS Dark Interactions 2014 mip signal S/N ~ 25  t = 2ns Impact parameter 11

12. HPS Apparatus: ECal Ecal consists of top and bottom modules, each arranged in 5 layers with 442 lead-tungstate (PbW0 4 ) crystals in all Crystals are readout with APDS and preamplifiers Data are recorded in 250 MHz JLAB FADCs A thermal enclosure holds crystal temperature constant to ~1 0 F to stabilize gains. Ecal is downstream of SVT & magnet PbWO 4 crystal with APD and preamp Crystals are arrayed above and below the Ecal vacuum chamber e-e- e-e- 12

13. Ecal/trigger worked well in the HPS Test Run. Several upgrades (larger APDs, improved preamps, and light monitoring system) will boost performance and improve reliability. Trigger rates measured in Test Run agreed well with EGS simulation, giving confidence that rates will be as expected in HPS. Ecal Test Run Performance Color shows average crystal PH over Face of ECal Horizontal Crystal Number Vertical Crystal Number Dead Zone Missing Channels HPS Dark Interactions 2014 13

14. HPS Reach in 2015 HPS Dark Interactions 2014 HPS 22 4.5  1 wk 1.1 GeV 1 wk 2.2 GeV 2 wk 4.4 GeV 14

15. HPS Schedule Summer 2014 Completing detector construction Fall 2014 Installation and commissioning on beamline December 2014 First Electron Beams in HPS Spring 2015 Re-commission and Physics Runs at 1.1 and 2.2 GeV Fall 2015 Physics Run at 4.4 GeV TBD 2016 Off for CLAS12 Commissioning 2017-2019? Additional Running at 2.2, 4.4, and 6.6 GeV TBD HPS Dark Interactions 201415

16. HPS Collaboration JLAB + SLAC + FNAL + IPNO Orsay + INFN Genova + Universities (+ New Collaborators at Glasgow and INFN Catania, Torino, Sassari, Roma) HPS Dark Interactions 201416

17. Conclusions HPS is a new experiment at JLAB, dedicated to searching for heavy photons with masses 10-200 MeV and couplings 10 -3 <  < 10 -5 in unexplored regions of parameter space. HPS uses a large acceptance forward spectrometer, operating close to the incident electron beam. It depends on the accelerators’ ~100% duty cycle and very high rate electronics and DAQ to integrate large luminosities in this environment. Invariant mass and vertexing signatures let HPS achieve sensitivity to very small values of the A’ coupling. Using invariant mass alone, HPS covers  2 > few x 10 -7 for 10 < m A’ < 200 MeV. HPS is completing construction this summer, and will be installed in Hall B at JLAB this fall. First data taking is scheduled for Spring 2015. Besides heavy photons, HPS is sensitive to other possible hidden sector particles, and is capable of discovering True Muonium, the µ + µ - atom. HPS Dark Interactions 201417

18. Backup Slides HPS Dark Interactions 201418

19. Simulated Vertexing Performance HPS Dark Interactions 2014 Accurate knowledge of SVT occupancy gives us confidence that stand alone pattern recognition will work in the presence of realistic backgrounds. Simulated tracking efficiency is ~ 98% with beam backgrounds included. Only 5% of tracks have miss-hits, which can cause vertex tails, and spoil reach. Track quality, vertex quality, and trajectory cuts nearly eliminate vertex tails. Before quality cuts After quality cuts Tracks with miss-hits make tails m A’ = 200 MeV E beam = 5.5 GeV Vertex Distribution along the beam direction, Z v 19

20. Trigger Rates HPS Dark Interactions 2014 Accurate knowledge of electron backgrounds allows reliable estimates of trigger rates. Use full Monte Carlo with backgrounds. 20

21. High Rate DAQ SVT DAQ uses SLAC ATCA-based architecture * Sensor hybrids pipeline data at 40 MHz and send trigger-selected data to COB for digitization, thresholds, and formatting. COB transfers formatted data to JLAB DAQ. * Record data up to 16kHz in pipeline mode. Will push this up to 50 kHz with upgrades.. * One ATCA crate with 2 COBs handled the full HPS Test Run SVT (20 modules, ~10k channels). Ecal DAQ and Trigger * Data recorded in 250 MHz JLAB FADC. PH and time transferred every 8ns to Trigger Processors. * Trigger sent to SVT DAQ and FADC for data transfer. * Ecal FADC and DAQ can trigger and record data up to 50 kHz. HPS Dark Interactions 2014 Cluster on Board (COB) Ecal DAQ/Trigger 21