Nuclear electronics

391 Pages·1974·114.881 MB·English

by  Nicholson, P.W.

Contents
,
f •
•
•
•
I
•
,
1 An Outline of Detection Methods 1 •
•
• • • • •
1.1 Interaction of nuclear radiation with matter 1
• • •
•
1.1.1 Introduction
1
• • • • • • •
1.1.2
Heavy charged particles 1
• • • • •
1.1.3 Electrons and positrons 3
• • • • •
x-
1.1.4 and y-rays 4
• • • • • • •
1.1.5 Neutron's 8
• • • • • • • •
1.2
Statistics of the production of ionization. 11
• • •
1.3 Gas filled detectors 13
• • • • • • •
1.3.1 Ionization chambers 13
• • • • • •
1.3.2 Proportional counters . 17
• • • • •
1.3.3 Geiger countt;rs 19
• • • • • • •
1.4 20
Semiconductor detectors
• • • • • •
1.4.1 20
Introduction
• • • • • • •
1.4.2 Homogeneous (bulk) detectors 23
• • • •
1.4.3 24
PN junction detectors .
• • • • •
1.4.4 24
Surface junction detectors
• • • • •
1.4.5 25
Lithium drifted detectors
• • • • •
1.4.6 Thin, totally depleted dE/dx detectors 26
• • •
1.5 27
Scintillation counters
• • • • • • •
,
1.5.1 27
Introduction
• • • • • • •
1.5.2 28
Organic scintillators
• • • • • •
1.5.3 30
Inorganic scintilla tors
• • • • • •
1.5.4 31
Photomultiplier tubes
•
• • • • • •
1.5.5 , 33
Response to different particle energies and types
•
35
1.6
Other detection methods
, •
• • • • •
1.6.1 35
Cerenkov detectors
• • • 1 • • • • •
1.6.2 36
Spark chambers .
r. •
, • • • • •
•
III
Contents •
Xl
3.3.4 Gere-Miller restorer . 118
• • • • •
I 3.3.5 Other methods . 120
• • • • • •
Resolution in Spectroscopy Systems . 122
· 4
• • • • •
-
4.1 Noise. 122
• • • • • • • • •
4.1.1 Statistical nature of noise 122
•
• • • •
•
4.1.2 Johnson or thermal noise 126
. • • • • •
4.1.3 Current shot noise 127
• • • • • •
4.1.4 Flicker or excess noise. 128
• • • • •
4.2 Signal to noise ratio, fwhm and the equivalent noise charge
. (ENC) 128
• • • • • • • • •
•
4.3 Optimization of signal to noise ratio 130
• • • •
4.3.1 General case 130
• • • • • • •
4.3.2 Radiation detector systems 133
• • • • •
4.3.3 Optimization under additional constraints 136
• •
4.4 Signal to noise ratio for practical systems 139
• • •
4.4.1 CR-RC shaping. 139
• • • • • •
141
4.4.2 Various systems .
• • • • • •
. 149
4.5 Effects of pile up .
• • • • • • •
149
4.5.1 Overloading
• • • • • • •
4.5.2 Estimation of loss of resolution 150
• • • •
•
4.5.3 Effect of AC coupling and finite detector load 154
•
156
4.6 Effect of baseline restoration on signal to noise ratio .
•
4.7 Sampling methods, time variant and pulse shaping 158
~on-linear
•
5 Amplifiers 166
• • • • • • • • • •
•
166
5.1 Introduction
• • • • • • • •
5.2 The field effect transistor 166
•
• • • • • •
5.2.1 Properties of the FET • 166
- • • • • • •
5.2.2 Noise in the FET 168
• • • • • •
•• •
5.2.3 Equivalent circuit for detector and preamplifier 170
•
5.3 Preamplifier configurations 172
• • • • • •
5.3.1 Conditions at the input 172
• • • • •
173
5.3.2 Charge sensitive configuration
• • • •
175
5.3.3 Current sensitive configuration
• • • •
5.4 175
Semiconductor detector preamplifiers
• •
• •
175
5.4.1
General considerations
• • • • •
179
5.4.2
Preamplifier circuits
• • • • • •
190
5.4.3
Circuits not using FETs
• • • • •
5.5 190
Requirements on main amplifiers for spectroscopy
•
•
190
5.5.1 Summary of requirements
•
• • • •
191
5.5.2 Bandwidth requirements
•
• • •
•
Contents •••
Xlii
287
7.7 Time to amplitude converters (TACs)
~ • • • •
•
7.7.1 Pulse overlap type 287
I
• • • • • •
7.7.2 Start-stop type 288
• • • • • • •
7.7.3 291
Compensation of amplitude time walk in T ACs
•
7.8 Linear gates 292
• • • • • • • •
7.9 Pile up rejector circuits 296
• • • • • •
303
Mu1tichannel Pulse Height Analysers
8
•
• • • • •
-
8.1 Components of multichannel analysers (MCAs) 303
• •
8.1.1 Analogue to digital converter (ADC) 303
• • •
8.1.2 Storage of digital data. 304
• • • • •
8.1.3 Ancillary circuits 305
• • • • • •
8.2 ADCs 307
• • • • • • • • •
8.2.1 Stacked discriminator type 307
• • • • •
8.2.2 Wilkinson type 309
• • • • • • •
8.2.3 Successive approximation type 310
• • • •
8.2.4 Hybrid types 312
• • • • • • •
8.3 Speed and precision of ADCs 313
• • • • •
8.3.1 Relative speed and precision of the different types 313
•
8.3.2 Sliding scale method of channel smoothing 315
• •
8.4 Design examples of ADCs 317
• • • • • •
•
8.4.1 Wilkinson type 317
• • • • • • •
8.4.2 Successive approximation type 321
• • • •
8.4.3 Dual ramp Wilkinson . . 325
• • • • •
8.5 328
Spectral stabilization
• • • • • • •
8.6 332
Additional facilities on MCAs
• • • • •
,
9 Mu1tiparameter and Computer Analysis 337
• • • •
9.1 Multiparameter analysis 337
• • • • • •
9.1.1 Introduction 337
• • • • • • •
9.1.2 341
Optical display
• • • • • • •
9.2 342
Partial storage of selected descriptors
• • • •
9.2.1 342
Associative storage
• • • • • •
•
9.2.2 345
Associative zone storage
• • • .' •
.
9.2.3 . 347
Transformation method •
• • • • •
9.2.4 347
Use of contents-addressable memory
• • •
9.3
349
Large computer systems
• • • • • •
9.4
Data highways 354
• • • • • • • •
•
358
Appendix A Transform Methods in Transient Analysis .
• •
• 358
A.l The general differential equation. .
•
• •
•
•
I
An Outline o f Detection
Methods
1.1 Interaction of nuclear radiation with matter
1.1.1 Introduction
Nearly all detection methods, the Cerenkov detector being a notable excep
tion, make use of the ionization or excitation produced in a detection
medium as a result of the absorption of all or part of the energy of the nuclear
particle. In the case of charged particles, ionization and excitation is pro
duced directly by the interaction of the electromagnetic field of the particle
with the electrons of the detection medium and the resultant ionization and
excitation is distributed as a track centred on the track of the particle. In the
case of uncharged particles such as X- and y-ray photons, the particle must
first undergo some process, such as a photoelectric or Compton interaction,
which transfers all or part of its energy to an electron which in turn produces
the track of ionization or excitation. Similarly neutrons must undergo an
interaction, such as a collision with a nucleus, the charged product of which
then produces the ionization or excitation. In consideration of their inter
action with the detection medium the commoner nuclear particles may be
divided into four groups:
•
(i) heavy charged particles (protons, alpha particles, heavy ions, mesons)
(ii) electrons and positrons
(iii) X-rays and y-rays
(iv) neutrons
1.1.2 Heavy charged particles
A heavy charged particle entering the detection medium loses its energy by
a succession of interactions, mainly between its electromagnetic field and
~ha~ of electrons in the medium, resulting in electronic excitation and
IOnization. According to classical mechanics the maximum energy that
Emnx
1
Nuclear Electro .
2
nlC!
s can acquire in a collision with a particle of m
an eI ec t ron of mas I» ass AI
0
and energy E is given by
_ 4moME _ 4
1»0
+
Ero •• - (mo M)2 - E M
(1.1)
•
The quantum mechanical treatment shows that there is a small probabTt
maxi~~
of the electron acquiring an energy slightly higher than the classical y
The more energetic of the electrons, often termed 'delta' rays cam
E n
max·
!
Si NE 102A
NE 102A Ge Si G
/ e
1mmr-----------r----------T.r-+--f-~~,
E
E
~
N
o 0
a particles
em
-IOOIJ m - 100 \0
w I'-
z •
U
••
-
tf)
.-
•
0
(f)
.-
~
0
.c-
.,
IOjJm 10 em
~ a .
c:
'"
o
particles air c
Cl:
protons 0
oc
OIr .
1 IJ m' -:-__________ __ 1 em
___,_'::_----___:_:!
"--~~
0 ·1 1 10 100
Energy. MeV
Figure 1.1 Range of ex particles and protons in various detector media.
Data taken from Goulding4 for Si and Ge, Evans2 for air, and Nuclear
Enterprises Ltd3 for plastic scintillator type NE I02A
govern~ ~he
themselves produce further ionization ; it is this factor that
colhs~ot
width of the track of ionization. Since the fraction of energy lost per
of the particle, given by (1.1), is small, the resulting deflections of the partIe:
lar~s
path are small and the track is comparatively straight. Owing to the
~f ~~:).
number collisions necessary to bring the particle to. re.st, the
~
comparatIVely well defined (typically with a standard deviatIOn of 1
8
' , ' known
IOC
F
or a given particle type and energy if the range R d'uJll
1 IS
" ' , !lie I
medIUm of density P and atomic weight AI' then the range R2 In a
I
AtJ Outline of Detection Methods 3
\ .
of density P2 and atomic weight A2 may be estimated by a simple empirical
J
relation known as the Bragg-Kleeman rule:
(~
(1.2)
-
.
For mixtures and compounds the same rule can be used 'if an effective
atomic weight A is used where
j j
JA= n A +n A +···
2 2
(1.3)
+ + ...
j
njJA n2JA2
where n n are the atomic fractions of the constituent elements of
l ' ,·.·
2
atomic weights AI ' A Figure 1.1 shows the ranges of protons and
2' ....
alpha particles calculated for various detection media.
1.1.3 Electrons Gnd positrons
For energies up to 10 MeV, electrons lose their energy to the detection
medium mainly by excitation and ionization of the eledrons of the medium,
as in the case of heavy charged particles. For higher energy electrons the
loss of energy as bremsstrahlung becomes increasingly important and the
intensity of this varies as Z2 where Z is the atomic number of the medium.
Thus, for example, 9 MeV elec,trons in lead lose as much energy due to
bremsstrahlung as due to ionization.
In absorption due to ionization, owing to the lower mass of the electron
compared to that of a heavy charged particle, a much greater fraction of the
particle energy is transferred to the absorber electron in each collision so
that the delta rays are correspondingly more energetic and are capable of
causing considerable secondary ionization at a greater distance ' from the
original particle track. For a 0·1 MeV electron in air about two-thirds of the
total ionization is due to secondary ionization.
Owing to the large energy loss per collision, the path of the particle shows
[<;on: . ble deflections and the range is not wen defined. Since an electron
at a much higher velocity than a heavy charged particle of the same
, it spends less time in the vicinity of the absorber atoms and so the
of energy loss and the density of ionization are correspondingly less.
or this reason the range of electrons.is much greater thilll that of heavy
. of the same energy. Figure 1.2 shows the range (essentially an
'oxi'r n figure) for several detector media. For various medium atomic
absorbers the range, for a given energy, expressed in terms of mass
• •
umt area of the absorber, is approximately constant provided brems-
. does not accout for significant energy loss. Finally for beta rays
havIng a maximum energy of E the nomogram in Figure 1.3 gives the
a .
max' . .
pproxImate percentage transmission for absorbers havmg dIfferent mass
,
Nuclear Ele
I
4 cr~~
th absorber is interposed between Source and d
't ea when e , If ' h' k etect
per um ar ) or where the source Itse IS t IC and causin
Q
(externa,1 absorber , g sell
absorptiOn, es differing only slightly from those of electro
't ns have rang ' , ns
ij
a~nihilate
POSI ro When brought to rest a PO,slt,ron w.IlI, With ;
the same energy, nd emits the charactenstIc anmhilatIon radiatio
' b 'ng electron a , , n
0
neigh oun h of energy moe?- (511 keY), mutually opposite directionl
two y-rays, eac III
100r-----r---~--1
10
,
NaI._~
Range 1
mm
0,1
,
O'OI~~-'--'-~----,L1---~10
0 ,01 0 ,1
Electron energy. MeV
Figure 1.2 Range of electrons in various detector media,
Data taken from Goulding4 for Ge and Si. Nel,ms5 fo~ N~;
,
and Nuclear Enterprises Ltd3 for NE I02A plastIc sCIntIllat
1,1.4 X- and )I-rays
X- and )I-rays lose energy in matter by three main processes :
(i) photoelectric effect
(ii) Compton effect
(iii) pair production d t al
In the photoelectric effect, alJ the energy of the photon is transferre
Os j!
I
h
e ectron, t e on'g m, al photon disappearing in the process, Th e proces
-
120
050
100 75
-
90 060
70
r
- -
80 65
90 070
60
-
70
60 - 50 80 0·80
70 0·90
40
50
60 1·0
-
30
- 50
40
- 1·2
40
20
r
-
30 " 1-4
- .--
0
-
r
-" c. 1·6
.-0 iii
10 20
c.
1·8
20 .0
~
0 r "
-
18 - V> 2·0
-.0
0
16 - " "
I 5 10 - 2·2
~
Q)
14 - Q) >< 2-4
(j) w
- 2·6
12
2·8
10 3-0
-
35
8 -
- 4·0
6 - •
-
4 -
Figure 1.3 Nomogram for the absorption of
fi
continuous rays according to Radiochemical
Centre, Amersham6 reproduced with per
,
mission. A straight edge placed between the
fi
absorber thickness and the maximum energy
enables the percentage absorption to be read
ofT on one of the centre scales. The external
absorption scale refers to an absorber inter
pOsed between Source and detector, and the
self-absorption scale refers to the case of a thick
Source with uniformly distributed radioactivity
6 Nuclear Electronic!
most likely 'to occur with a K electron, in which case the energy imparted tQ
the photoelectron by a photon of energy hv would be
hv - E~ (1.4)
where is the binding energy of the K electron. This is immediately fol.
E~
lowed by de-excitation of the atom with the emission of one or more X-rayS
of total energy These X-rays may escape from the detection medium bUi
E~.
due to their low energy, are likely to be reabsorbed in other photoelectric
processes. Thus the K X-ray may interact with an L electron of another atoll!
which in turn would de-excite and give an L X-ray. This in turn would
probably be reabsorbed with the eventual emission of an X-ray of still lower
energy. Thus after this rapid sequence of events a whole family of photo.
electrons are produced whose combined energy is very nearly equal to the
original photon energy hv. Of course any of the family of X-rays may escape
the detector medium or may undergo a Compton process rather than a
photoelectric process. However, owing to the decreasing energies of the
sequence of X-rays, the probability rapidly decreases and even the escape
of the K X-ray is an event rare enough to be usually unimportant.
In the Compton effect only part of the photon energy is transferred to an
electron and the photon is deflected and decreased in energy. The angle at
which the photon is scattered can take on any value and the energy imparted
to the electron correspondingly has a continuous distribution from zero to
a maximum of E The value of E (which corresponds to the case of the
• ' •
m x m x
photon being back-scattered) is given by
E 2(hv)2 (1.5)
+
m c lhv
= 2
m.x
o
Figure 1.4 shows the distribution of energy imparted to the electron by
gamma rays of three different energies.
Pair prodl!ction can occur if the gamma ray has an energy exceeding 211lo(
(1'02 MeV). In this case the original photon disappears and the electron
positron pair carries away the excess energy hv - 2m c2 The electron and
o .
positron lose energy through collisions and eventually come to rest. The
positron then annihilates \'Iith a neighbouring electron and emits the .
teristic annihilation radiation, a pair of y-rays each of energy m c 511
2 =
.(
o
that are emitted in opposite directions. One or both of these may be
absorbed by photoelectric or Compton processes so that the total
imparted to detector medium can vary widely. It is usually possible,
to see three major components in the energy imparted to the medium
2
hv, hv - moc and hv - 2mo2 corresponding to the total absorption
both, one, or neither of the annihilation y-rays.
The relative magnitude of the Compton, photoelectric and pair
processes will depend on the energy of the photon and the nature of tile
I An Outline of Detection Methods
7
t!ien medium. If the photon traverses a thickness dx of the medium we will
<define dx as the total probability of any of the three absorption processes
~T
o€ curring, where is known as the total coefficient of absorption or macro
~T
sGepic cross section. In this case a beam of photons of intensity, 10 after
traversing an absorber thickness of x will have a primary beam intensity of
Since the total probability of absorption dx in a thickness dx is
Jioe - ~TX. ~T
~
...
I I I I I
~ -
160
f-
c
-e
c·
u
-'" c- .
140
~
"'E ~ hl/=0·511 MeV
u 120 c- -
~
6
-
-
__ 100 c-
o
-
•
'"
.-
§ 80 -
-
c •
. r ......
>-
-
-
l:' 60
hl/= 1·2 MeV
'"
-
c -
...
'"
- -
'" 40
c.
c - -
.o-
hl/=2'76 MeV
t; 20)1 0.. /
3l
-
'"
.'.".
o
o 05 1·0 1·5 2·0 2•5
U
Electron energy in MeV
Figure 1.4 Differential cross ~ection fo: the
Compton effect. The ordinate proportIOn~1
IS
to the number of Compton electrons per UnIt
electron energy interval and the data are for
three different y-ray energies. (Davisson and
7
Evans
)
toile sum of the individual probabilities for photoelectric, Compton and pair
production processes we may write
+ +
~T ~PhOIO ~compl ~p.ir
(1.6)
=
' .
For the photoelectric process with ')I-rays of energy above the K shell
binding energy the magnitude of ~PhOIO is given approximately by
5
~PhoIO oc NZ (hV)-3'5 (1.7)
where the absorber has N atoms per unit volume of atomic number Z. Thus
toNe use of a high atomic number absorber and of low energy photons greatly
increases the probability of photoelectric interaction.