Cherenkov Radiation and RICH Detectors

Transcription

1 Cherenkov Radiation and RICH Detectors Neville Harnew * HT 2016 Basic expression of Ch radiation History of Ch radiation and the RICH What is a RICH? Physics for which you need a RICH Ingredients of a RICH Illustrative RICHes geometry, radiators, photodetectors DELPHI BaBar DIRC LHCb Large scale RICH systems: Super-K Amanda / Icecube Pierre Auger Epilogue: a very quick look at transition radiation 1 * With thanks to Guy Wilkinson

2 History of Cherenkov Radiation Discovery : Pavel Cherenkov 1936 Radiation seen when uranyl salts exposed to radium source. Explanation: Tamm and Frank 1937 Experimental exploitation in HEP pioneered by Cherenkov himself Cherenkov: (Cherenkov, Tamm, Frank: Nobel Prize 1958) 2

3 Cherenkov Radiation in a Nutshell Occurs when the speed of a charged particle is greater than the speed of light in the medium Both a threshold and thereafter, an Fundamental Cherenkov relation: angular dependence up to saturation (β=1) Frank-Tamm relation: N γ = no. of photons; E = Photon energy; L=radiator length; Z = particle charge Determination of N γ requires an integration over energy range acceptance 3 So number of photons will increase with velocity (up to saturation)

4 RICH : Ring Imaging Cherenkov Fathers of the RICH Cherenkov : 1936 discovery Arthur Roberts: first to propose exploiting ϑ c Tom Ypsilantis: driving force behind practical RICH

5 What is a RICH? Measurement of β from RICH, together with p, from tracking system, allows mass, and hence PID to be determined. This is an excellent way of separating π from kaons and protons The simplest way to exploit Cherenkov radiation is to choose n such that heavy particles do not emit light. Threshold Cherenkov counter (not a RICH!) But if we want to do better, or if momentum is far from monochromatic, then we need to measure ϑ C. We have to image the ring. This is a RICH 5

6 Experiments which need Hadron ID B physics CP violation studies Hadron spectroscopy/exotic searches Large volume neutrino detectors (special case see later) Many other experiment would benefit (eg ATLAS/CMS), but they often have other priorities for the space! 6

7 B (& D) Physics Requirements for PID B physics CP violation experiments perform exclusive reconstruction of final states, with & without kaons (and protons). Hadron PID mandatory. eg. selection of B h + h - - below is an LHCb simulation study Another example: kaon flavour tagging essential in CP measurements 7

8 B (& D) Physics Requirements for PID Suppress background from combinatorics. PID allows much cleaner reconstruction of charm (and B) mesons D φ π Without PID With PID CLEO III D s φ π 8

9 Ingredients of a RICH We need a radiator, a mirror and a photon detector. Why do we need mirrors? 1. Often to take light out of detector acceptance 2. Mirrors focus light. Without mirrors we would have a splodge. But sometimes it is true we can survive without mirrors 9

10 Proximity Focusing : no mirrors If radiator is sufficiently narrow in extent, then light which emerges will be a ring, rather than a splodge. 16 cm Photon Detector Pure N 2 Charged Particle Cherenkov Photon LiF Radiator CLEO III RICH 20 µm wires CH +TEA 4 CaF Window mm Reinforcing G10 Rod Fibreglass Siderail This is proximity focusing. Works with solid and liquid radiators, where we get plenty of photoelectrons. Photon Detectors LiF Radiators 82 cm 101 cm 10

11 Considerations in Building a RICH Want to optimise ring resolution. Ring resolution determines how far in momentum we have PID. σ σ Contributions to ϑ C : Emission point error (how well the focusing works) Detection point error (the spatial resolution of the photodetector) Chromatic error see next slide. N pe optimised through radiator choice & length, and photodetector performance. 11

12 Chromatic Dispersion A significant source of uncertainty in the Cherenkov angle determination is that of chromatic dispersion. Refractive index varies with wavelength/photon energy, and hence so will ϑ c Control this effect by limiting wavelength. Do this with choice of photodetector technology &/or filters. Visible light preferable to U-V, although many U-V RICHes in use! 12

13 An untraviolet RICH: DELPHI ( ) OMEGA, DELPHI & SLD were the first experiments to use RICHes in anger. Provided PID from low p up to 25 GeV or so. 13

14 DELPHI RICH the principle Very nice! 14

15 DELPHI RICH the reality 3 fluid system, mirrors and HV shoehorned into 70cm (inaccessible) gap Gas radiator C 5 F 12 Mirrors Photosensitive (TMAE) drift volume MWPCs Liquid radiator C 6 F 14 AAgghhhh, the horror! 15

16 BaBar Detector of Internally Reflected Cherenkov Light (DIRC) DIRC radiators cover: 94% azimuth, 83% c.m. polar angle 16

17 BaBar DIRC BaBar PID requirements: π-k separation up to 4 GeV Low track density Must be fast (bx every 4 ns) Severe space and material constraints PMTs provide adequate resolution DIRC an ideal solution! 17

18 DIRC RECONSTRUCTION DIRC Ring images: reflection ambiguities up to 16 (θ c, φ c ) ambiguities per PMT hit, Cherenkov ring images are distorted: complex, disjoint images 18

19 DIRC Performance N pe and resolution π-k separation vs momentum σ( θ c, ) = 2.4 mrad 19

20 LHCb: a two RICH (3 radiator) detector 20

21 Why RICH? Why 3 radiators? Physics requirements, and kinematics of b production, mean there is a big range in the momentum of the hadrons we wish to identify Momentum Only suitable PID technique is RICH. Even then, three radiators are needed: Low p: aerogel ; central p: C 4 F 10 ; high p: CF 4. Span 2<p<100 GeV/c 21

22 LHCb RICH 1: a two-in-one detector Aerogel Spherical Mirror, tilted at 0.3 rad to focus rings and to keep photodetectors outside acceptance Gas volume Flat Mirror Photodetectors, enclosed in magnetic shielding box (not shown) to protect vs fringe field of dipole magnet 22

23 Aerogel as a RICH Radiator Aerogel: a low density form a quartz. Refractive index Well suited to low momentum π-k separation. Used extensively in threshold counters, but not in RICHes C=0.006 μm 4 cm -1 4 cm thick tile The problem of Rayleigh scattering: T = A e (- C t / λ 4 ) Scattering of photons limits transmission at low wavelength. Scattered photons background. Aim for as clear samples as possible, which means low values of C (= clarity coefficient ) 23

24 LHCb RICH: Hybrid Photo-Diodes (HPDs) Good single photon efficiency (~30% max) Sensitivity in visible Capacity to cover large area (several m 2 ) Good spatial resolution (order mm 2 ) High rate capabilities 24

25 Particle ID reconstruction Global event likelihood algorithm (particle ID algorithm fits the event as a whole in both RICHes). Likelihood function includes expected contributions from signal plus background for every pixel. IoP Tyndel-fest 1 st July 2011, RAL Neville Harnew 25

26 Cherenkov angle vs momentum Using isolated rings IoP Tyndel-fest 1 st July 2011, RAL Neville Harnew 26

27 PID calibration samples D from D* Κ s Λ Calibration data give unique π/k/p samples allow PID performance in efficiency and purity to be evaluated with data IoP Tyndel-fest 1 st July 2011, RAL Neville Harnew 27

28 Large water volume neutrino detectors Examples: SNO Super-Kamiokande 50 k ton H km underground Cherenkov rings are an ideal technique for detecting ν μ, e 28

29 Cherenkov Rings in Super-K Cherenkov light a perfect signature of a neutrino interaction ν e in water. μ e No momentum measurement, so PID performed from sharpness of ring. Timing response of PMTs necessary to determine particle direction. 29

30 Oh dear, oh dear! 30

31 Antartica 31

32 Ice Cube: a Cherenkov Counter for Studying High Energy Neutrinos 32

33 33

34 Hit Multiplicity Energy Measurement Upgoing muons from H.E. neutrino interactions. E µ =10 TeV, 90 hits E µ =6 PeV, 1000 hits 34

35 Pierre Auger Search for Ultra High Energy Cosmic Rays Knee Fluorescence detectors? Water Cherenkov Detectors 35

36 Communications antenna GPS antenna Electronics enclosure Solar panels Battery box 3 nine inch photomultiplier tubes Plastic tank with 12 tons of water 36

37 RICH Conclusions In the olden days, RICHes were tricky! Very often this technique is criticised as being too difficult and not reliable. We admit that in some senses this is true But today we know how to build them: B Physics Experiments Spectroscopy Atmospheric neutrinos Tom Ypsilantis Cherenkov detection is a vital tool in armoury of experimental HEP: RICH detectors, neutrino detectors, cosmic rays 37

38 Epilogue: a very superficial look at transition radiation detectors What is transition radiation and what is its use in HEP? Brief example: ATLAS TRT For more information: Boris Dolgoshein, Transition Radiation Detectors, Nuclear Instruments and Methods A326 (1993)

39 Basics of Transition Radiation Transition radiation emitted when particle moves across interface of 2 media with different dielectric constants (predicted in 1946 by Ginzburg and Frank) Consider ultra-relativistic particle passing through thin foil of material (1) in environment of material (2), then differential distribution of radiation is: E TR is energy of radiation; ω is angular frequency; ω p is plasma frequency ; Θ is angle of emission. Φ 1 is phase angle, due to interference between boundaries. Characteristics of transition radiation: 1) Forward peaked 2) X-rays 3) Total energy radiated proportional to γ! 39

40 Transition Radiation & HEP Applications Dependence on γ makes TR an attractive method of PID, particularly for discriminating between electrons and hadrons. Used for non-destructive electron identification (cf. calorimeter). Works over wide momentum range. Experimental challenges: 1) Radiation very feeble. So require many foils (usually lithium or polyethelene) 2) Forward peaking means that almost always X-rays and primary particle are seen by same detector. Generally one detects particle de/dx and TR together. So must distinguish sum of energy from de/dx alone, or look for clusters specifically associated with absorption of the X-rays. 40

41 ATLAS Transition Radiation Tracker Part of the ATLAS Inner Detector. Provides combined tracking, with standalone pattern recognition and electron identification. Layers of xenon-filled straw tubes interleaved with polypropylene foils. Can suppress pions by a factor of about 100, for 90% electron eff. 41

42 SPARE SLIDES 42

43 TMAE / What was tough about DELPHI What is this TMAE stuff? Tetrakis dimethylamine ethylene Glows sickly green on contact with oxygen Photo-sensitive ( photo-ionizing ) in the UV Some challenges of the DELPHI RICH Very limited and inaccessible space for 3 fluid system and mirrors. Need to keep gas radiator at 40 degrees to stop it condensing, whereas TMAE kept at 28 degrees (to optimise absorption length) Liberated photoelectrons which have to drift 1.5 m (TPC technique) Unsurprisingly, took a long time to commission. Not a plug and play detector. Many crises along the way. But when working it worked a treat! 43

44 DIRC RECONSTRUCTION Time information provides powerful tool to reject accelerator and event related background. Calculate expected arrival time of Cherenkov photon based on track TOF photon propagation in radiator bar and in water σ( t) = 1.7 nsec t: difference between measured and expected arrival time ± 300 nsec trigger window ± 8 nsec t window (~ background hits/event) (1-2 background hits/sector/event) t (nsec) 44

45 Belle Cherenkov Counter: not strictly a RICH! Cherenkov technique allows hadron PID even when ring not imaged merely look for presence/absence of light. No light means heavy particle. ( veto or threshold mode) Amount of light seen can still be exploited. This a viable approach if we do not need to cover a large range in p! 45

46 HPDs and Testbeam Results 46

47 Photodetectors for Ice Cube optical sensor 10 inch Hamamatsu R-7081 records timestamps digitizes waveforms penetrator pressure sphere optical gel HV board flasher board DOM main board delay board PMT mu metal cage transmits to surface at request via digital communications can do local coincidence triggering design requirement Noise rate ~1 khz SN monitoring within our Galaxy 47

48 AMS TRD Purpose: to look for positrons and suppress proton background by factor of To be achieved by combined TRD / ECAL system. 20 layers of polypropylene radiator and proportional straw tubes (Xe) 90% e-id efficiency for 0.1% contamination 48