Quartz Fiber Calorimetry and PPAC for IP Beam Instrumentation for LC

1
Quartz Fiber Calorimetry and PPAC for IP Beam
Instrumentation for LC

• Y. Onel (Iowa)
• E. Norbeck (Iowa)
• D. Winn (Fairfield)
• ALCPG - Victoria Linear Collider Workshop
• July 28-31, 2004

2
Quartz Fiber Calorimetry
3
Quartz Calorimeter Features

• The detector is intrinsically radiation hard at
  the required level (hundreds of MRads)
• The detector, for all practical purposes, is
  sensitive to the electromagnetic shower
  components (?M)
• It is based on Cherenkov radiation and is
  extremely fast (lt 10 ns)
• Low but sufficient light yield (lt1 pe/GeV)
• The effects of induced radioactivity and neutron
  flux to a great extend are eliminated from the
  signal
• Neutron production is considerably reduced
  (high-Z vs low-Z)
• The detector is relatively short
• The detector is perfectly hermetic

4
Cherenkov Light Generation

• When high energy charged particles traverses
  dielectric media, a coherent wavefront is emitted
  by the excited atoms at a fixed angle ? called
  Cherenkov light.
• Light is generated by Cherenkov effect in quartz
  fibers
• Sensitive to relativistic charged particles
  (Compton electrons...)
• d2N/dxd?2?? q2(sin2?c / ?2)
• (2?? q2/ ?2 )1-1/?2n2
• ? min 1/n
• Emin 200 KeV
• Amount of collected light depends on the angle
  between the particle path and the fiber axis

5
Iowa-Fairfield-ORNL-Tennessee-Mississippi
6
PPP-I Schematic View
7
PPP-I
ROBox ( Light Guides) R6425 PMTs
Fiber Bundles (EM, HAD and TC) 300-micron core QP
LED, Laser and PIN PDs
Iron Absorber (9.5 ?I)
Ferrules
Radioactive Source Tubes
3 x 3 Tower structure (6 cm x 6 cm)
8
Previous Experimental Data on Photodetectors by
HF Group
R6427
9
HF Pulse Shape
10
Spatial Uniformity w/ e- beam
11
Spatial Uniformity w/ ?- beam
12
PPP-I Response to 100 GeV e- and 225 GeV ?-
13
Energy Response Linearity
HF PPP1 responds linearly within 1 to electrons
in the energy range tested (6 200 GeV). The
?- response is highly nonlinear.
14
Energy Resolution
Energy resolution of a calorimeter is
parameterized as (?/E)2 (a/?E)2 b2 a/?E
sampling term Characterizes the statistical
fluctuations in signal generating processes.
b Constant term Responsible for the
imperfections of the calorimeter, signal
collection non-uniformity, calibration errors and
leakage from the calorimeter.
15
HF Wedge
16
First HF End Completed
17
First HF End Completed
18
Quartz Fiber Radiation Damage

• Major radiation induced absorption bands for
  Si-core fibers are grouped as
• Prominent absorption band in 600-630 nm due to
  non-bridging-oxygen hole centers (NBOHC). NBOHC
  is a molecular structure where Si atom is bonded
  to four Oxygens and one of them carries an
  unpaired e-, ?Si-O
• NBOHCs have a luminescence band at around 670nm
• The origin of the NBOHC is the conversion of
  paired hydroxyl (OH-) groups into peroxy linkages
  during the plasma deposition of F-doped cladding.
  The peroxy linkages serve as NBOHC precursors by
  breaking the O-O bond.
• E color center One of the most studied defects
  in SiO2, ,?Si. Has an absorption peak at 212nm
  and luminescence at 450nm. Produced in glasses by
  energetic irradiation and during fiber drawing
  process.
• Attenuation tail extending to near-UV has
  several origins. Strongset from Cl impurities.
• Different color centers may interact with each
  other and may display different characteristics
  when irradiated.

19
Motivation for Neutron Radiation Tests of Quartz
Fibers

• Scientific literature about optical
  characteristics of Quartz fibers is generally in
  the infrared band (800nm, 1300nm, 1550nm studied
  a lot)
• Many of these studies conducted by ? or e-
  irradiation
• Our studies concentrated on 325-800 nm range
• PMT sensitivity 400-500 nm
• Two experiments were carried out
• UTR-10 , 10 kWatt Reactor _at_ ISU, Ames
• MGC-20E cyclotron of ATOMKI in Hungary

20
ISU Reactor Test Setup

• Fibers were subject to ?-rays, fission spectrum
  neutrons and thermal neutrons
• ?-rays uniform, fast-slow neutrons position
  dependent
• Total ? dose 22 kRad (measured with commercial
  dosimeter)
• Neutron Flux 1.3x1010 n/cm2/s/kW
• Integrated neutron fluence at the end of
  experiment 1x1015 n/cm2
• Reactor power altered periodically

21
ATOMKI Cyclotron Neutron Source Test Setup

• 18 MeV proton incident on 3 mm thick Be-target
  to generate neutrons, ltEgt3.7 MeV
• Eneutron ranged up to 20 MeV
• 25.3 hours of operation, total neutron fluence
  ?1.02 x 1015 n/cm2 18
• Average nfluence at the cylinder ? 0.6 x 1015
  n/cm2
• During Irradiation, the dose rate was constant ?
  1.1 x 1010 n/cm2/sec 18

22
Xe Lamp Spectrum
23
Analysis

• FSHA After 1015 n/cm2, 1 dB/m attenuation in
  blue-visible optical region that matches the
  sensitivity of the PMTs used in HF detector.
• All Si-core fibers tend to recover to varying
  degrees, ? ? 103-104seconds.
• Importance of in situ optical measurements is
  manifest by the recovery data presented. This is
  particularly important for the calibration of the
  detectors. A(?) A(?0) 10/L log Iirr(?) /
  I0(?)
• A(?0) attenuation of fiber prior to irradiation
• L length of the fiber (4 meters in our
  case)
• I spectral intensities
• Second term represents the irradiation induced
  loss.
• 325-800nm range is covered
• The intensities were binned in 25nm intervals and
  average values were used in calculations and
  figures.

24
Experimental Area _at_ CERN in LIL

• Motorized support.
• Moved from the beam during stop.
• Dose rate 600rad/s.
• Beam perpendicular to fluorescent screen.
• There is effectively 5.5 cm iron in front of
  fiber. Fiber embedded inside the iron.
• Iron block placed _at_ 8 slope with respect to the
  beam.
• Beam scanning of 8 cm on fluorescent screen
  irradiates 100 cm fiber length.
• Fiber placed _at_ max of dE/dx of EM shower.
• Dosimeters were installed behind iron absorber in
  the same place with fibers.
• Iowa group has tested fibers at LIL CERN 500 MeV
  electrons NIM A490 (2002) 444

25
Experimental Setup

• Measurements were done In situ.
• Spectrometer, light source and PC were kept in
  temperature stabilized place.
• Irradiation place was at room temperature.

26
Sample Spectra (Before After Irradiation)
AFTER IRRADIATION (54MRad)
BEFORE IRRADIATION
UV light absorbed by long fibers. Total decrease
almost for all wavelengths. Deep around 610
nm. Least effect between 700 and 800 nm.
27
Attenuation
A(?, D) - (10/L)log10I(?, D)/I(?,0)

• Obtained using previous two spectrum.
• There is no transmission below 350 nm.
• Relatively bigger attenuation at 610 nm.
• No effect between 700 and 800 nm in our
  measurement precision.
• Relative deep around 450 nm.

54 MRad
28
Parameterization for Irradiation
A(?, D) - (10/L)log10I(?, D)/I(?,0)
Attenuation A(?,D) ?(?)D?(?) Power
law parameterization I(?,D)/I(?,0)
exp-4.343L?(?)(D/Ds)?(?) Fit function D
dose I(?,0)
reference spectrum taken before irradiation, at
D 0 I(?,D) spectrum taken at
dose D L length of the
fiber Ds 100 Mrad scale factor ?(?)
corresponds attenuation _at_ 100 Mrad with our
parameterizations. lt?(450)gt 1.52 0.02 dB/m
lt (610)gt 6.08 0.04 dB/m
29
Recovery Studies
A(?,t) / A(?, tirr) 1/1?(?)(t/tirr-1)?(?), t
gt tirr

• When beam stops fibers start to recover.
• Continue to take data after turning of the beam
  without touching the fibers.
• Recovery is faster _at_ 450 nm then 610 nm.

450 nm
610 nm
30
Future RADDAM

• We will test special quartz fibers with quartz
  cladding. These fibers are Silica/Silica,
  High-OH, UV enhanced, QQ (Quartz core/ Quartz
  cladding) with different type of buffer materials
  (Acrylic, Polymide, Aluminum) with different
  diameters (300, 600, and 800 micron)
• Fibers will be given 5 x 1017 n/cm2, about 20
  Grad(neutrons with energy gt 0.1 MeV)in IPNS
  (Intense Pulsed Neutron Source)at Argonne
  National Laboratory
• The range of 10-50 Grad will also be available at
  this facility.
• We will test the induced attenuation vs
  wavelength, transmission of Xe light in the
  350-800 nm range after irradiation. Also measure
  the tensile strength before and after the
  irradiation.

31
Iowa/Fairfield/Adana
32
Cleaved Fibers
33
QQ fiber Transmission Measurements

• Transmission of Xe light through QQ fibers before
  radiation
• Measured at Iowas HEP lab using
  micro-spectrometer

34
PPAC for LC Calorimetry
35
Typical low-pressure PPAC

• Two flat plates
• Separated by 2 mm
• Filled with 10 torr isobutane
• MIPs often leave no signal
• 700 V between plates
• Timing resolution better than 300 ps
• Used with 50 MeV/nucleon heavy ions

36
Single Pixel PPAC For Test With High-Energy
Electrons at JLab or SLAC

• Gap 1.0 mm
• Cathode 2X0 8.26 mm of tantalum
• Area of anode is 0.25cm2

37
Signal from a PPAC pixel
FWHM is 1.3 ns
Single peak with considerable noise. The noise is
large because of the small size of the signal
using our 137Cs source. With the much larger
signals from high-energy electrons, the noise
will be negligible.
38
Electron and positive-ion currents
The electrons are collected in less than 1 ns.
It is the moving electrons that generate the
signal that is measured. The current from the
slow moving positive ions is smaller by a factor
of a thousand. We have looked at the
positive-ion signal using special electronics and
find that it lasts for about a microsecond.
39
Four equal bunches separated by 1.4 ns
Each bunch gives a signal with 1.3 ns FWHM
40
PPAC for Hadronic Calorimeter

• Three flat plates, separated by 2 mm
• Middle plate at high voltage
• Outer plates hold atmospheric pressure
• Filled with 10-40 torr of a suitable gas
• Gas flows in one side and out the other
• Timing resolution better than 300 ps
• Plate composition chosen to maximize signal,
  i.e. maximize conversion of softphotons to
  electrons

41
PPAC energy resolution
Poor for single heavy ion Current per mm2 is
huge! Same size signal from shower should have
good resolution. Measure resolution with double
PPAC Look at ratio between two sides
42
Iowa PPAC - a radiation hard detector
Double PPAC for testing energy and time
resolution. The PPAC detector concept can be
developed as a candidate for the luminosity
monitor.
43
Tests with double PPAC

• Test with EM showers using 80 ps bunches of 7 GeV
  electrons from the Advanced Photon Source, at
  Argonne National Laboratory
• Planned test with low energy hadron showers using
  the 120 GeV proton test beam at Fermilab

44
PPAC Test at ANL

• IOWA double PPAC was tested for energy and time
  resolution with electron showers from the
  Advanced Photon Source (APS) at Argonne National
  Laboratory.
• The booster ring of the APS puts out 76 ps
  bunches of 7 GeV positrons at the rate of two per
  second, with 3.6 x 1010 positrons in each bunch.
• In normal operation the positrons are injected
  into the main storage ring where they are used to
  produce synchrotron radiation.
• There are maintenance and development periods
  during which the beam is directed into a beam
  dump. We set up our equipment next to the beam
  line just in front the beam dump.
• The entire beam bunch has an energy of 2.5 x 1020
  eV, or 2.5 x 108 TeV, much more than we needed.

45
Results of PPAC Test at ANL

• To make use of this beam we placed the PPAC close
  to the beam line where it would be exposed to
  showers generated by the outer halo of the beam
  striking the beam pipe. Because of the small
  angle between the positrons and the wall of the
  beam pipe, the wall acted as an absorber with a
  thickness of several centimeters. The showers
  were developed in this absorber.
• We expected the time resolution between the front
  and back PPACs to be less than 300 ps. What we
  found was 3 ns (FWHM). This is still a fast
  signal even though it is an order of magnitude
  slower than expected.
• The poorer than expected time resolution was
  caused by noise that required the discriminator
  levels to be set high in order to eliminate
  spurious events.
• One source of noise was caused by the necessity,
  because of safety regulations, to have the power
  for the preamps near the beam line come from a
  wall plug in the beam tunnel while the rest of
  the electronics was powered from a wall plug in
  the floor above.

46
Timing resolution

• A lower limit on the expected timing resolution
  was measured by cross connecting timing signals
  from alpha particles. The results are shown in
  the next slide.

47
Timing resolution
48
Energy Resolution Data of PPAC Test at ANL
Ratio Efront to Eback is constant to within 2
49
PPAC Test at Iowa for Electronics
PPAC with Alpha source
50
PPAC Test at Iowa for Electronics
51
PPAC Test at ANL
PPAC under beam line to beam dump
52
CONCLUSIONS
PPACs for sampling calorimeters

• Can be made radiation hard.
• Have good energy resolution.
• Are fastsubnano-second time resolution.
• Can be made to reject background.

53
BACK UP SLIDES
54
No Texas tower effect
With above-atmosphere hydrocarbon gasoccasional
proton from n-p scatteringgives huge signal. In
PPAC, proton hits wall at almost full
energy.PPAC signal mostly from low-energy
electrons. We will test this with detailed
simulations.
55
Background reduction
The background can be reduced by subdividing the
detector into small sectors. One such design has
a single plate at high voltage and the grounded
plate divided into small segments, of perhaps 1.
cm2. With such a small area the plate spacing
would be reduced to 1 mm, which provides the
additional benefit of a faster signal. Of course,
the subdivision comes at the price of additional
electronics to measure the signal size of each
segment.