General radiation detection

concepts

Myroslav Kavatsyuk
KVI – Center for Advanced Radiation Technology
[email protected]
 A generic radiation detector
detector sensitive volume
quantum of
radiation

assumptions:
• individual radiation quanta are detected
• all interactions per quantum are so fast they can be considered instantaneous
• some or all energy is deposited, resulting in ± charge carrier pairs or light
• + and/or – charge carrier, or light is collected
• per quantum, a pulse is generated: pulse mode operation

pulse shape depends on detector type

and eventual electronics

2
Information contained in detector pulses
energy related to:
amplitude
integral

time pick-off level

t0
“arrival time”

count rate (pulses/second)  radiation intensity

3
 Energy spectrum (pulse height spectrum)

pulses are digitized (ADC) energy calibration:

 intensity vs. channel number energy vs. channel number
 most often very linear

intensity vs. energy

4
 Energy resolution: definition
Full width at half maximum: FWHM

G.F. Knoll, Radiation Detection

and Measurement, 3 rd Edition

• FWHM can be determined experimentally for any peak

• sometimes FWTM (FW tenth M) provides useful information

5
 Energy resolution: origin
1. detector performance
– electronic noise
– drift, stability

2. statistical variation in number of primary “excitations”

E e: energy required for a primary excitation
– # of primary excitations N = E: energy deposited by radiation quantum
e

– if excitations are independent: Poisson statistics

Δ N =√ N
Δ E FWHM ∝µ √ E
Δ E FWHM 1
∝µ
E √E
1% FWHM requires N = 55 000

6
 Intensity
energy spectrum results from a “counting” experiment, not a “scanning” experiment
 intensity from energy spectrum is: NOT height, amplitude
BUT integral (sum over channels)

e.g. peak intensity

K.S. Krane, Introductory

Nuclear Physics, 1 st Edition

7
 Detector efficiency

source detector

G.F. Knoll, Radiation Detection

and Measurement, 3 rd Edition geometry
dependence
• absolute efficiency ( abs): number of pulses detected
strong
number of quanta emitted
• intrinsic efficiency ( intr): number of pulses detected
weak
number of quanta incident on detector
• solid angle ( )

isotropic sources: e abs=e int r 

point source: =
1
2( ( ))
1−cos atan
a
d
8
 Pick of the time information
Leading-edge discrimination

“arrival time” for a pulse with

different amplitude will have
different offset with respect to
the true start of the signal
development

True start of a
signal
development time pick-off level

t0, ”arrival time”

Simple method; result depends on pulse amplitude; can be corrected upon

measurement of the pulse amplitude 9
 Pick of the time information
Constant-fraction discrimination (CFD)

Time stamp should be set when

signal rises to a certain fraction
of the pulse amplitude

True start of a
signal
development threshold

t0, ”arrival time”

10
Pick of the time information
Constant-fraction discrimination (CFD)

Time stamp should be set when

signal rises to a certain fraction
of the pulse amplitude

Specially formed signal can

guarantee constant position of
the time stamp with respect to
the true start of the pulse

CFD method yields stable results, however

might have worse performance due to the
noise influence

11
 Timing
• arrival time of a quantum of radiation in the detector ?
• timing requires different pulse handling than energy determination
• different time pick-off methods
• often, optimal timing is incompatible with optimal energy resolution

ΔV ΔV • low noise
Δ t= • fast detector
dV /dt • “bright” detector

Δt
12
 Timing resolution
measurement of time differences
– between detector & reference signal
– between 2 detectors
coincidence measurement

G.F. Knoll, Radiation Detection

and Measurement, 3 rd Edition

13
Timing resolution: a real-life example

• crystal: 3x3x5 mm3

• bare LaBr3:Ce(5%)
• LYSO

• Hamamatsu SiPM
S10362-33-050c

• spectralon reflective material

• Agilent DC282 digitizer

• 8 GS/s
• 700 MHz anti-aliasing filter

• extra amplification timing channel

14
 World-record timing resolution
511 keV coincidence resolving time (CRT) (FWHM)
LaBr3:Ce(5%) 95 ps
LYSO:Ce 138 ps

LaBr3

spectra shift
20 mm 20 mm as Δt=2 Δx/c
D.R. Schaart et al., Phys. Med. Biol. 55 (2010) N179
R.Vinke et al., 2009 IEEE Nucl. Sci. Symp. Conf. Rec. M06-2
S.Seifert et al., 2009 IEEE Nucl. Sci. Symp. Conf. Rec. J01-4 15
 Complex detection systems

Complex detection systems are required to identify detected particles

or measure all their properties.

Examples:

Tracking detectors in magnetic field EM calorimeter

 Identification of charged particles
Requirements:
●
Active volume of the detector is position sensitive
→ possible to “record” particles tracks
●
Particle looses only fraction of its energy in the active material
●
Detector is placed inside of a magnetic field

Example of recorded tracks

Energy loss in the detector

Curvature of tracks
→ momentum measurement
 Calorimetry

High-energy photons/electrons generate shower of EM particles:

●
In order to measure energy of initial particle all secondary
secondary particles have to be detected and their energy
added up → Calorimeter

In most experiments it is important to measure precisely direction of

the initial particle → large detector with single active volume will
have no position sensitivity

Solution: detector volume has to be segmented

 Calorimetry

Solution → detector volume has to be segmented:

●
Energy information → sum of energy deposited in fired volumes
●
Position of primary interaction → weighted average of positions of fired
volumes:
weights are calculated using deposited energy : W i = W0 + ln(Ei/E)
Ei – energy deposited in a single active volume
E – energy deposited by the complete shower
 Calorimetry: energy resolution

Radiation length mean distance

over which the electron energy
is reduced to 1/e of its original
value due to radiation loss only

Ec is the electron energy at

which the cross section
of bremsstrahlung becomes
equal to that of pure ionization
 Calorimetry: energy resolution
The number of electrons and
positrons in a shower
produced by a photon with a
given energy fluctuates
statistically

n ~ E/Ec

Therefore, relative energy

resolution (relative width of the
distribution) will be proportional
to:

or
 Calorimetry: energy resolution
The complete form of the relative
energy resolution is:

a – constant term (calibration of

detector, looses of shower
particles)
b – statistical term

c – noise of readout electronics