Status of the Photomultipiers tests for COMPASS Recoil Detector

1
Status of the Photomultipiers tests for COMPASS
Recoil Detector

• Tomasz Szczesniak
• Soltan Institute for Nuclear Studies,
• PL 05-400 Otwock-Swierk, Poland

2
Plan of this presentation

• Description of scintillators layout
• Selection of photomultipliers (PMTs)
• Measurements
• photoelectron yield
• time jitter
• time resolution
• Results
• XP20D0
• XP45D2
• Future tests

3
1. Description of scintillators layout

• Detector consists of 2 barrels of 24
  scintillators each
• A barrel
• - thickness 0.4 cm
• - length 2.84 m
• - median width 6.6 cm
• - distance to target 25 cm
• B barrel
• - thickness 5 cm
• - length 4 m
• - median width 29 cm
• - distance to target 110 cm

Both ends of scintillator are connected to light
guides and next read out by photomultipliers
4
2. Selection of photomultipliers

• Crucial properties
• High quantum efficiency
• Good photoelectron collection efficiency
• Low time jitter
• Best possible time resolution
• Detector A 2 diameter
• Detector B 5 diameter

5
Proposed photomultipliers

• For A scintillators (internal detector)
• Photonis XP20D0
• This is modificated version of standard XP2020
• - 8 stages instead of 12
• XP20D0 with 10 stages will be also discussed
  because of better amplification.
• - the screening grid at the anode.
• Due to the screening grid, a time resolution
  measured with various scintillators is improved
  by 1.15 1.2 factor, as compared to the standard
  PMT.
• (see M. Moszynski, Prospects for new fast
  photomultipliers NIM A337(1993)154)

6
Proposed photomultipliers

• For B scintillators (external detector)
• Photonis XP45D2
• This is modificated version of standard XP4512
• - with the screening grid at the anode
• A comparative study of the XP4512 with similar
  products of Hamamatsu and Electron Tubes showed
  the best photoelectron collection in XP4512.
• (see M. Moszynski, G.J. Costa, G. Guillaume, B.
  Heusch, A. Huck, S. Mouatassim, "Comparative
  study of 130 mm diameter photomultipliers for
  neutron detectors", Nucl. Instr. and Meth.
  A308(1991)97)

7
Commonly used anode construction

• The anode built as a grid is placed inside the
  last dynode
• Advantages
• - Low time-of-flight of electrons from last
  dynode to the anode
• - Good charge collection at the anode

8
Screening grid at the anode

• Disadvantage of commonly used anode construction
• The anode signal consists of two two components
• Main one due to the collection of the electrons
  from the last dynode
• Parasitic (shifted in time) induced in the anode
  by the electrons travelling from the penultimate
  dynode to the last dynode

Solution screening grid
9
3. Measurements

• Photoelectron yield (Bertolaccini et al. method)
• The number of photoelectrons per energy unit
  (Nphe) is determine by comparison of the peak
  position due to the single photoelectrons
  (PP1phe) with the characteristic point at the
  energy spectrum (PPE), i.e. full energy peak
• Nphe (PPE gain1phe) / (PP1phe gainE) / E
• Where
• gain1phe - value of the gain for single
  photoelectron
• gainE - value of the gain for the tested
  gamma-rays
• E - energy of characteristic point MeV

10
1phe peak

• Measured with XP4512
• Amplification gain 500
• 4.3 plastic coupled to PMT

• Measured with XP2020
• Amplification gain 2000
• No crystal coupled to PMT

11
137Cs energy spectrum

• Measured with XP2020
• Amplification gain 5
• With 2 diameter NaI(Tl) coupled to PMT

• Measured with XP4512
• Amplification gain 2.5
• With 4.3 plastic coupled to PMT

Nphe 5700 phe/MeV
Nphe 2000 phe/MeV
12
Measurements

• Time jitter
• variations of transit time of electrons
  between the cathode and the first dynode.
• Two components
• A chromatic component due to the spread of
  initial velocities of electrons originating from
  the same point.
• A geometric component due to the different
  lengths of primary paths of electrons emitted
  from different points on the cathode.

13
Measurements

• Transit time is determined by measuring the
  interval between signals known to be synchronous
  with light pulses at the cathode (LED in our
  case) and resulting anode pulses.
• Differences due to geometric component can be
  measured by focusing the light pulses on
  different parts of the cathode and noting the
  corresponding transit times.

14
Time jitter for 2 PMT without the screening grid
FWHM (wholePC) 506 ps
15
Time jitter for 2 PMT with the screening grid
FWHM (wholePC) 521 ps
16
Measurements

• Time resolution
• CFD Ortec 935, optimized for each combination of
  PMT and scintillator
• Time To Pulse Height Converter Ortec 457
• Slow-fast arrangement
• Time spectra measured in relation to the
  truncated BaF2 crystal coupled to the XP20Y0Q/DA.
  Its contribution to time resolution was 90?4 ps
  for 60Co and 128?4 ps for 511 keV annhilation
  quanta (22Na).

17
4.Results

• XP20D0
• recently tested with LSO and LaBr3
  scintillators.

18
Results
19
Results
20
Results
21
Results
22
Results

• A qualitative comparison, different crystals, not
  the same adjustment of timing
• A screening grid and a very high blue sensitivity
  of XP20D0 are more important, for a rather slow
  crystals as LSO and LaBr3, than the superior time
  jitter of R5320.
• The study showed the very good timing
  capabilities of the XP20D0 PMTs
• The importance of the screening grid at the anode
  and a superior quantum efficiency were pointed
  out.

23
Results

• XP45D2
• Photonis developed a modified version of 5
  diameter XP4512 with the screening grid.

24
Results
25
Results
26
Results
27
Results

• Problems with XP45D2 with the grid.
• Phe number slightly lower
• Time resolution for Co-60 worse
• Time resolution for Na-22 slightly better
• Limited dynamic range of the anode signals,
  timing affected by the space charge effect
•  
• The tested PMT has a very high gain. HV 1100
  1200 V. Typical interstage voltage corresponds to
  50 V only
• Problems of XP45D2 will be discussed by Marek
  with Photonis in Beaune in
• this week.

28
5. Future tests

• Complete tests with light guides
• Smaller light guide for the internal detector
  will be send to us together with a small piece of
  the plastic scintillator (5 cm long for example)
• It will allow us a testing of the transmission of
  the light guide using 5" diameter PMT by coupling
  the small piece of plastic directly to the PMT
  and via the light guide
• In this way we can look also for a degradation of
  other parameters as energy spectra and time
  spectra
• Comparison of Reflector VM2000 film and Teflon
  tape