89% found this document useful (9 votes)
17K views52 pagesKnoll Solutions Manual 3rd Ed
The solution of radiation detection and measurement for all question .
Uploaded by
Tamara AbedCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
89% found this document useful (9 votes)
17K views52 pagesKnoll Solutions Manual 3rd Ed
The solution of radiation detection and measurement for all question .
Uploaded by
Tamara AbedCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 52
SOLUTIONS TO PROBLEMS
from
Radiation Detection and Measurement,
Third Edition, by Glenn F. Knoll
Published in 2000 by
John Wiley and Sons, New York
These solutions have been prepared with the assistance of a number of past and
present students from the Department of Nuclear Engineering and Radiological Sciences
at the University of Michigan, with major contributions from Paul Nowak, Richard
Kruger, Jim Fox, and Shana Browde. The Mathematica solutions in Chapter 17 were
provided by Prof. David Wehe.
Glenn F. Knoll
Professor of Nuclear Engineering and Radiological Sciences
University of Michigan
‘Ann Arbor, Michigan
October 1988, December 1990, and November, 2000 1.
12,
1-4,
1-5.
1-6.
17,
1-8.
1-9,
CHAPTER 1
Line: a,c,d,e,f,andi. Continuous: b, g, and h.
Mshell,
Momentum conservation: pq =P, (ni Tepresents product nucleus)
p= mv = (2mE)°5
(mE), = 2mE),
Energy conservation: EytE, = Q
Eyt(mg/t,) Ey =Q
Eq = Q(1+(mg/m,)) = 5.50 MeV/(1+4/206)
Ey, = 5.395 MeV
2=he/E
A= (4.135 x 105 eV-s) (3.0 x 108 my/s)/(195 x 103 eV)
= 0.0636 x 101 m = 6.36 pm
2BSY — 17gn + 18S
Q=M@SU) c2—M(N7Sn) c2-M(8Sn) c2
M@SU) = 235.0439 M(117Sn) = 116.9029 M(118Sn) = 117.9016 amu
Mc? =931.5M (MeV)
Q=223.0 MeV
Sp. Ac. = (In2- Av) (ty M) = (In 2)(6.02 x 1023 mol-1) / (3.866 x 108 s)(3 g/mol)
= 3.60 104Ba/g = 9.73 kCi/g
E=qV=(2e)(3MV)=6MeV
‘Heavy water contains deuterium. For 7H(y,n), Q = -2.224 MeV.
Therefore, the minimum gamma ray energy is 2.224 MeV.
7H 43H — 4He+In
Assume tritium nucleus is at rest, 21.
2-2.
23.
2-4,
energy conservation: E, +Ey.=Ey+Q or
Pa?/2my + Pye*/2mMyje = Pe?/2my + Q
‘Multiplying through by 2m, and making the approximations my, = 4m, and my = 2M,
Pr? + Pue’/4 = pa’/2 + 2m,Q
For the neutrons emitted in the forward direction (0°), the 4He nucleus must be emitted at
180° to the incident deuteron's direction.
Momentum conservation: py + Pyte = Pa
Eliminating py, and solving yields: p, = (pa + [6p42 + 40m, QI°5 )/5
Inserting pq = [2mgEg]°- = [2 (2)(1.008 amu) (0.150 MeV)]°5 = 0.778
and Q= 17.6 MeV leads to py = 5.497
E, = py2/2m, = (5.497)2/ 2 (1.008) = 14.99 MeV
CHAPTER 2
T=1.2x10-7R (my/E)5 s
my=4amu E=5MeV
T=1.073x10-7Rs
Rgi= (5 mg/em?) / (2330 mg/em3) = 2.1 x 103 em=2.1 x 105 m
Tsi=2.3x 102s or 2.3 ps
Ry ~ (0.9 mg/em2) / (0.08988 mg/cm?) = 0.1 m
Ty= 11x 108s or Ins
R= 210 um for 5 MeV protons in Si. Therefore, the range remaining
after 100 jum is (210 - 100) or 110 pm. Corresponding energy remaining = 3.2 MeV
‘The expected energy loss is given by AE = (-dE/dx)t
From Fig. 2-10, (-dE/dx)/p for a 1 MeV ot particle is approximately 380 MeV-cm?/g.
Pau = 19.32 g/m? t=5x10-4+em
Using these values, AE =
.67 MeV. Therefore, the a particle loses all its energy.
According to the figure, (pR)s; = 0.5 g/em?, Using the scaling law,
p, PSR sinlAn (5 wkm’ (2698 aoe
OT PanfAs (2.699 glem') [78.09 2-6.
2-8.
2.9.
2-10.
2-12.
With hv = 1 MeV, @ = 90°, and m,c? =
hv
).511 MeV, the formula
hy’ =.
1+ me (1 cos 6)
yields hv’ = 0.338 MeV
Probability of photoelectric absorption per atom = t = const. x Z"/(Ey35), For n= 4.5,
as
4 =0.00
a
Compton scattering: a, b, andd
Photoelectric absorpti
Pair production: e
©
a) mean free path = 4 = 1/1 = 1/{(up) p] = 1/[(0.06 cm?/g) (3.67 g/em®)] = 4.54 cm
b) prob. of photoelectric absorption = 1 - expl-tt]
= 1 —exp[ — (0.01 cm/g) (3.67 g/cm?) (1 cm)]
= 0.036
Sce text.
‘With wy = 2/18, wo = 16/18, and the given attenuation coefficients,
the equation
P.-L,
yields (1i/p)¢ = 0.153 cm?/g. The mean free path is given by
2.= fp = Walp) pl = 1/[(0.153 em?/g) (1 gem?)] = 6.52 em
Ey = (5x 106 eV) (1.602 x 10-19 J/eV) = 8.01 x 10-13 J
N = 1 J/g = 1.248 x 10!
P= (# electrons per unit time) (energy per electron)
= (100 x 10-6 C/s) (106 eV/e-) (1.602 x 10-19 H/eV) (1 €7/1.602 x 10-9 C) = 100W 2-13. With the exposure rate const. = 13.2, ct = 108 mCi, and d = 500 cm,
the equation
X=r,&
ag
yields an exposure rate of 52.8 mR/hr.
2-14, E=mC,AT > AT=E(mC,)
E/m = (10 x 10-9 rad/hr) (100 erg/g-rad) (10-7 J/erg)
Cp = 4.187 J/g
AT = 2.39 x 10-8 °C/hr
2-15, From Chapter 1 (p. 20), we know that 2.3 x 10° neutrons/s are produced per 1g of
252Cf and that average neutron energy is between 0.5 and 1 MeV. So from fig.
2.22b, he = 5 x 10"! Sv-cm’,
He= he
N__ Gulg)(2.3x10'neuts/s- 4g)(8-3600s
brad 4x(500cm)*
He = (6.33x10%em™)(5x10"'Sv-cm?)(100x10°mren/Sv) = 0.317 mrem
® = 6.33x10'neuts/om?
CHAPTER 3
3-1, To first approximation, the sample variance is independent of the size of the sample.
Therefore, there should be no significant change in its value.
3-2, Using the binomial distribution, with success defined as "heads",
p=05 x=8 n=12
PQx)= * (1-p)"* = 0.1208
™ mE P
3-3, Success = counting a male.
mean = p n= (0.75) (15) = 11.25
variance = n p (I-p) = (15) (0.75) (1 - 0.75) = 2.81
standard deviation = [variance]?5 = 1.68
—— eee 34,
3-6.
3-8.
3.9.
3-10.
Suocess = rolling a six
p=6 x=0 n=10
Using the binomial distribution, P(0) = 0.1615.
a) Success = an error
p=1/60 n=250
mean = pn=4.17
standard deviation = [n p (1-p)]95 = 2.02
Using Poisson distribution, standard deviation = 2.04,
b) p=1/60 n=100 x=0
Using Poisson distribution, the result is P(0) = €~1000 = 0,189,
a,b, and f.
net = (source + bkgd) ~ bkgd = 561-410 = 151
net = [ (Source + bkgd) + bkgd JS = [561 + 410]95 = 31.2
count rate = [(source + bkgd) ~ bkgdV time = (846 -73)/(10 min) = 77.3 min-t
count rae = [ (Source + bkgd) + bkgd J?5 / time = [ 846 +73 ]05 / (10 min) = 3.03 min!
Using the data from prob. 3-8, optimal allocation is given by
seB
Since Ts,p + Ty = 20 min, Ts, = 15.45 min, and Tp = 4.55 min,
With this time allocation, the standard deviation in the net source counting rate is
sB, B 46, 73 at
NV Tea Ty V 13a 5a = 266mi0
‘The improvement factor is 3.03/2.66 = 1.14.
As a fraction, o = [1/N]°5 = [1/ST]°5, where N is the number of counts, S is the count
rate, and T is the time interval. Since the count rate is constant,
[02 T = [02 Tl, or
Ty = Tp [69/6]? = (10 min) [2.8/1.0] = 78.4 min
‘The additional time = 78.4 ~10 = 68.4 min 3-11.
3-12.
3-13.
3-14,
3-15.
3-16.
a) If $ >> B, the figure of merit 1/T = €2S. Therefore, double the source signal (S).
b) IfS <0) = 1 - P(O) = 1 - (BT)? exp(-BT) / 0! = 1 —exp(-BT) = 0.99
BT =~In 0.01 = 4.605
T = 4,605/B = 4.605/(2.87 min’) = 1.60 min
NAlTq ~Nog/Mog = 45.1 min’!
Ta = Np /Tp —Nog/Tpg = 353.4 min!
R= tpt, = 7.84
To calculate Gp, use R = [Np/Tp — Nug/Togl/INa/Ta —Nog/Tgl-
Application of the error propagation formula yields og = 0.634,
ty = 845/30 min) = 28.17 min! rg = 80 - 28.17 = 51.83 min"!
if Ty—Np/Tp
Applying the error propagation formula, Os = [Ny/(Ty)? + Np/(Tg)? 15,
Og = [rp/Ty + Np/(Tp)? ]°5 = [80/T + 845/(30)? ]-5
0s = 0.03 rs = 0.03 (51.833) = 1.555
Eliminating os and solving, T = 54.1 min.
T=
a) t=0.25 [2+ 7242452412] =22
y2 = (5-1)(22)/28 = 3.14
Using Chi-square table, p ~ 0.5, which indicates a good fit to the Poisson distribution.
'b) Gmean = [mean/n]®5 = [28/5]°5 = 2.37
©) Since each of the sets is comparable to the original one,
sts 02 = 28/5 =5.6
4) mean = [mean/N]5 = [28/(5)(30)]°5 =
43
For this data set, mean
1616.1, s?= 3197, x2 = 21.22,
Using Chi-square table, p = 0.5. They are consistent with the Poisson distribution. 3-17.
3-18.
3-19.
3-20.
= 1, ~tp =2162.4— 2081.5 = 80.9 mint
€=Ny/Ta-Np/Tp
Applying the error propagation formula and substituting Na for (6,4)? and Np for (6p)? :
Ge = [NA(Ta)? + Np/(Tp)2I°5 = [ry/Ty + ty/T pS
6, = [(2162.4 min+)/(10 min) + (2081.5 min)/(20 min)}°5
6,=17.9
Since the results differ by 80.9/17.9 = 4.5 6, the difference in the rates is apparently real.
A,=A,R where R=NJN,
(nJAy? = GaJAs? + (ORR?
(02)? = (,05/3.50)2 + (op/R)?
(Op/R)? = 1.959 x 104 = (Gyy/N,)? + (n/N)? = IN, + IN,
= 2 (1/1000 s(T2)]
T = 204s
i Ty = (1683/10 min) — $0 min) = 118.3 mint
1) = Tiq—Ty = (914/10 min) — 50 min = 41.4 min“
0%(r,) = 0%(r,) + 0%(r,) = [(1683)!7/10)]2 + 0 = 16.83
0%) = O%(ry) + O%(r,) = [(14) 2/10)? +0 = 9.14
R= tyr, = 118.3/41.4 = 2.857
(Op/R)? = [OlFp)/t2]? + [olny]? = 9.14/(41.4)? + 16.83/(118,3)2 = 6.535 x 10-3
Half-life = T =0.693yInR = 0.693(24 h)/in 2.857 = 15.84h
AT/AR. = [ -0.693+-(1/R)V/(In RY? = (-0.693)(24 h)/(2.857)(1.050)2 = 5.284 bt
OAT) = (dT/AR)?6°(R) = (5.284 hol)? (6.535 x 10-3)(2.857)?
O(T) = 1.22h
Assume rp has no associated uncertainty
mp = ms Tolts
=100 ig (13834/(3 min) — 281 min-1/91396/(10 min) ~ 281 min-1]
= 48.88 pg
Replacing rq by No/Ty and rs by Ne/Ts, and applying the error propagation formula yields
6=0.473 ug 3-21.
3-22,
3-23.
Using the equation N = No exp[-At),
N,
1
heath
2
Applying the error propagation formula, and substituting No for (0,)? and N; for (6,)?
yields
1+e*
or0.5At = 1+e-At_ Solving this equation iteratively, t = 2.218/A.
Wy = expt)
t= Info] / 1
Applying the error propagation formula and setting
(6)? =1= Ip expt) (Ip and t have 0 variance),
(6)? = exp(ut)/[H? Ig]
To minimize the error,
do, texp(it) _ 2 exp(ut)
du wh wh
Solving, = 2/t = 2/(1 cm) =2.cm-1
‘The gamma ray energy corresponding to this attenuation in aluminum is about 35 keV.
1=P() +P(>0)
POR) = (ang)™ €x(Kayg) / x!
Xavg = Mean number of events in 10 ms interval
= (10 ms) (100 s+) = 1
1)? exp(-1) /0! = 0.368
1-P(0) = 0.632
PO)
P(>0) 3-24.
3-25
4-1.
42,
43.
4-4,
45,
(00 counts/min)(30 min) = 3000
a) Expected number of background counts =
Then:
oy, = 13000 = 54.8
1, = 1.645264, =127.5
b) From Eq. 3.67,
Np=4.6530, +2.706= 257.6
From Eq. 3.68,
Np 257.6
ae = —__2576
eT (0.853)(0.15)(30min)(60 .)
= 1.12/s or 1.12 Bq
a) Fora fixed frequency f, T = 1/f, mean waiting time is (1/2)'
b) Assuming a Poisson random process, = 1/r = 1/(30min)~
=30min.
CHAPTER 4
Vinax = O/C = (10 €-) (1.602 x 10-1 C/e~) / (100 x 10-1? F) = 1.602 mV
See text
See text
RC = (10 000 ohms) (300 x 10-1? F) =3 x 10-6 s = 3000 ns >> t,
Large collection circuit time constant.
N INTEGRAL CcuIESe
oul CURVE 46.
47.
48.
49.
a)
a N
iv 4H iv 4H
b)
GN
oH N
iv H iv 4
°)
a N
L
isv 4 isv H
Resolution of the peaks requires that the energies be separated by at least 1 FWHM.
Let FWHM = E)—E, = 490 - 435 = 55 keV.
For Ey, R= FWHM, = 55/435 = 12.6%
For Ey, R = FWHMEE; = 55/490 = 11.2%.
Taking the smaller value, R must be smaller than 11.2%.
R
N
35 [F/N]OS
5.5225 F/R? = (5.5225)(0.1)/(0.005)? = 22 090
2 3
FWHM yg) =f FWHM ag + FWHM nis = oy (0.02H,)” + (0.048)
FWHM yea = 0.0447 H,
R=FWHMgyeran / Ho = 4.47% 4-10.
4-11.
4-12,
Q. = 2n(1 — d[d? + a2}0-5) = 2n(1 — 20/1202 + 52}°5) = 0.1876 steradians
P=Q/4n (assuming all space is randomly sampled)
= 2n(1 — d/fd? + a2)9S)/4n = (1 — d/{d? + a7] y/2 = (1 — 1/1 + (a/d)?J5)/2
afd = tan(0.59/2) = 4.36 x 10-3
P=4,76 x 10-6
N=S€j, Q/4n
S = 0.8)(20 x 108 s-1)(100 s) = 1.6 x 108
1.6 x 106)(0.12)(0.1876)/4n = 2866 counts
.28 x 10-2 min-)
(mt)
Solving fort, t= mr! — nl. Writing this result for the two observations and eliminating +
yields
1. = (In 2)ftyp = (In 2)/54 min =
nonparalyzabl a
Letting n, = nget and solving for ng yields
1 -expQ) _ 1 ~ expt (0.01284) (49)) _ 91 gg99
Ti T I
mm, 131340 93384
‘aralyzable model =m =n exp(-nt)
Solving for ~t, —t = (In{m/n})/n. Writing this result for the two
observations and eliminating + yields
Ij) _ nom)
n 7,
0 1
Letting nj = nge-t and solving for ng yields
ate in(m,) + In(m,)
m= exp fi
1-e"
Inserting the appropriate values yields no = 201800. 414,
415.
416.
4-17.
5-1.
The dead time loss in A = n—ma, and dead time loss in B = n-mg.
We require n - mg =2(n—m,). Solving for n, n= 2m,—mg.
ma=n/{14nt,] and mg =n/{1+ntg]. Substituting these expressions
into the expression for n and again solving for n yields
12
ny
Letting tq = 30 10-6 and ty = 100 10-6 s yields n = 13 333 s+.
‘With m, =m, = 10000 s+ and mj, = 19 000 s+, the equation
sgn Ma My (MyM) (Myy—M)
mm, my
yields t= 5.26 ps.
Paralyzable model: m =n exp(-nt), or 0 = n exp(-nt)—m
Using the Newton-Raphson method, the latter representation yields
Agy—m explny A]
Int
Letting m= 107 s-! and t = 1.5 x 10-6 s, an initial guess of 0 yields n = 119,700 s! and
an initial guess of 107 yields n = 1,996,000 s*.
Bhyax = te]
T= UltMpax €] = 1/((50 000 s~) (2.718)] = 7.36 ps.
CHAPTER 5
# electrons = E/W = (5.5 x 105 eV)/(42.7 eV/ion pair) = 1.29 x 105
Q= (# charge cartiers)(charge per carrier)
Q= (1.29 x 105)(1.602 x 1019 Cle’) = 2.06 x 1014 C
1 = (# a's per second)(charge per «) = (300 s-1)(2.06 x 10-4 C) = 6.19 pA 5-2.
53.
5-4,
5-5.
y= 0.04935 + 04116x
1 (A/pa)
0.02 0.08 = 0.08. 0.08. 0.10 0.12
WV (volts)
Saturation current is obtained by extrapolating to 1/V value of zero:
1 = 0.04935, 1 = 20.26 pA
For 0.5% drop in this value, Ay = .005 (,04935) = 2.468 x 104
Ax = Ay/slope = 2.468 x 104 /.04116 = 5.995 x 103
x = O+Ax = 5.995 x 103
V = Ux = 1/5.995x 103 = 167V
Av =aQqc
exposure = AQ/M = AV CM
At STP, M = (1.293 kg/m?) V = (1.293 kg/m) (50 x 10 m3) = 6.465 x 10° kg
exposure = (5V)(75 x 10-12 F)/(6.465 x 10° kg)
= 5.80 wCikg or 22.5 mR.
EJp = (1000 V / 0.05 m)/(1 atm) = 2x10'V/m-atm.
Using Fig. 5.2, v = 0.36x10° mis.
te = d/v = (0.05 m)/(0.36x10° m/s) = 1.39 ps
AQ/At = CAV/At = (250 x 1012 F)(150 V)/(1800 s) = 20.8 pA 5-6. Using the Si curve in Fig. 2-14, , (pR)s; = 0.25 g/em?,
RsPsifAne (0.25 elem) 2897 _ aa
. =211.an
PrinfAg (1.204 10° glem®) 28.09
5-7. Secondary electrons created by the 5 MeV gamma rays must not reach either electrode. For
minimum electrode spacing, assume the electrons are created exactly in between the parallel
electrodes. For 5 MeV gamma rays, typical secondary electron ranges are on the order of
several meters. Consequently, the electrode spacing would have to be larger than several
‘meters for compensation to be preserved.
R,
58. 1=Ey_, &e/W
T= (49 x 103 eV/B)(150 x 103 B sec-!)(1.602 x 10-19 C/e-/(27 eV/e)
= 43.6 pA
59, I=RM
At STP, M = (1.293 kg/m®)(I liter)(10°3 miter) = 1.293 x 10-3 kg/m
1 = (35.8 x 10°12 C/kg-s)(1.293 x 10-3 kg/m?)
= 4.63 x 10-4 A
5-10. V = [Mge/C] [x/d]
V = [(1000 €°)(1.602 x 1019 C/e-)/(150 x 1012 F)] [2/5]
=0.427 pV
ite air surroundings
Tonization
current
\
vacuum surroundings
3 gem? Wall thickness
a) According to Table 5-2, electronic equilibrium is reached for a wall thickness of
043 gem,
b) Ina vacuum, no secondary electrons enter the chamber, so there is inadequate
compensation using thin walls. This reduces the measured ion current compared
with air surroundings. 5-12. Using Table 5-2, 10 MeV gamma rays require a wall thickness of 4.9 g/em?,
t= (4.9 g/em?)/(2.699 g/cm?) = 1.82 em
5-13, Is=RM
M = (1.293 kg/m?) V [P/Pg] (TyT]
M = (1,293 kg/m?) (2.5 x 10-3 m3) [3/1] [273/373] = 7.10 x 10°3 kg.
Tg = (100 x 10-12 C/kg:s) (7.10 x 10°3 kg) = 0.71 pA
5-14, Exposure = AQ/M = AV C/M.
M = (1.293 kg/m) [Vol] [P/Pg] [T/T]
M = (1.293 kg/m) [10 x 106 m3] [1 atny/1 atm] [273 K/293 K] = 1.205 x 105 kg
Exposure = (50 x 103 V) (20 x 102) / (1.205 x 105 kg)
Exposure = 8.30 x 108 C/kg = 0.32 mR
5-15. First portion: Inge/dC] (vt + v-)t_ > V’ = [nge/dC] (vt + v-)
Second portion: V = [nge/dC] (vit +x) > V‘ = [nge/dC] (w+)
R= (slope of first portion)/(slope of second portion) = 1 + v/v*
rE and viepteE
+b/t = 1+ 1000 = 1000
CHAPTER 6
6-1. n= E/W = (108 eV)/(26.2 eV/ion pair) = 3.82 x 104 ion pairs
Gy = [F n]°5 = [(0.17) (3.82 x 104)]5 = 80.6 (or 0.21% of n)
6-2. Airis not used in proportional counters because it has a substantial electron attachment
coefficient, so that gas multiplication cannot occur under normal conditions. Air can be
used in ion chambers because either free electrons or negative ions may then be collected.
6-3. a) Letting M = 1000, a= 0.003 cm, b= 1 cm, p= 1 atm, AV = 23.6 eV,
and K = 4,8 x 104 V/cmatm, the equation
ee
Ina) AV" Kpain(b/ay
InM
yields V = 1793 volts.
a 6-4.
65.
6-6.
6-8.
6-9.
'b) Letting a =0.006 and V = 1793, the equation yields M = 7.52, so that the
multiplication decreases by a factor of 1000/7.52 = 133.
c) Letting a = 0.003, b= 2, and V = 1793, the equation yields M = 192, so that the
multiplication decreases by a factor of 100/192 = 5.2.
Letting a = 0.005 cm, b = 5 cm, p = 1 atm, V = 2000 V, AV = 23.6 eV, and K = 4.8 x 104
‘V/em-atm, the equation
Vv In2 ae
M= Tom av" Kpaln@a)
yields M = 4.93.
V=QIC= [ng eMI/C = [{E/W}eMI/C = EeM/WC
(5 x 106 eV) (1.602 x 10-19 C) (4.93) / [(26 eV) (500 x 10°12 F)]
V=0.304mV
‘The key is to make more than one measurement of output pulse amplitude for separate
values of deposited energy that differ by known ratios. If the amplitude remains
proportional to deposited energy, the detector is operating in the proportional region.
Alpha particle sources often provide monoenergetic particles whose range is less than the
dimensions of the chamber. Because of this, the proportional counter can record each
alpha particle that enters the active volume with almost 100% efficiency, and the output
pulse amplitude will be proportional to the particle energy.
In contrast, the range of beta particles exceeds the chamber dimensions. ‘The number of
ion pairs, and hence the pulse height, is only proportional the the small fraction of particle
‘energy lost in the chamber. This fraction decreases as the beta particle energy increases,
because the rate of energy loss, dE/dX, decreases as the beta particle energy increases.
‘Therefore, the higher the energy of the beta particle, the smaller the deposited energy, and
the smaller the pulse height.
Q=CV=ngM + M=CV/{noe]
1g = E/W = (50 x 10? eV)/(26.2 eV) = 1908
M = (200 x 1012 F)(10? V)/{(1908)(1.602 x 1019 Cy] = 6540
E=VArin(b/a)] > — t¢= VE In(b/a)]
Tc = (2000V)/[(106V/m) In(2/0.003)] = 0.0308 cm
‘Vonatt region! Viotal = [% (Fc)? h] / [xe b? h] = [ro/b]? = [0.0308/2]? = 0.0237%
The field tube prevents gas multiplication from occurring near the ends of the counter,
where the electric field is distorted. 6-10.
6-11.
Tl.
7-2.
73.
‘The discontinuities are absorption edges corresponding to the binding energies of electrons
in K, L, and M shells in the absorber atoms. If the gamma energy is above an absorption
edge, photoelectric absorption can occur. If the gamma energy is slightly below the edge,
this proces isnot energetically possible, andthe total interaction probability drops
ply.
‘The fluctuation in the size of individual avalanches is avoided, and the statistical limit on
energy resolution is lower.
CHAPTER 7
‘The quench gas has the lower ionization potential in order to favor charge transfer
collisions in which the positive ions of the primary fill gas are neutralized, and positive
ions of the quench gas are collected.
‘The starting voltage reflects the point at which the avalanche size is sufficient to assure
spread of the Geiger discharge. ‘The avalanche size is proportional to the multiplication
factor M, so the starting voltage should vary inversely with M. From the Diethom
equation:
a
~In@a) AV Kpaln(b/a)
a) Doubling the diameter of the anode wire (a) will slightly increase the first factor, but
will sharply decrease the last factor. Therefore, M decreases, and Vian increases.
b) Doubling the fill gas pressure (p) will decrease the last factor. Therefore, M decreases,
and Vyiar increases.
c) Consider the change from P-5 to P-10 gas. This results in increased values for both K
and AV. This results in a lower value for M and, therefore, a higher value of Veter
Increasing the voltage results in a higher initial electric field, so that a greater buildup of
space charge is required to reduce the field below its critical value. Consequently, the
pulse height increases.
To reach the Geiger region, (ng)'p 2 1.
(1) =3 M
M2 1/3p = 1/3 x 105) = 3,33 x 104
‘Using the equation and data in prob. 6-3 a), setting M = 3.33 x 10 and iterating, the
result is V= 2166 volts, TS.
7-6.
a) Proportional; The pulse height varies as the avalanche amplitude which, in turn,
depends on voltage in an approximately exponential manner.
Geiger: The pulse amplitude corresponds to the number of ion pairs at the point at
which the accumulated positive space charge is sufficient to reduce the electric field below
its critical value. ‘This number will increase in approximate proportion to the original
electric field or linearly with the applied voltage.
) Proportional: The quench gas must absorb UV photons.
Geiger; The quench gas must pick up positive charges from the original positive ions
through charge transfer collisions.
©) Proportional: Because heavy charged particles tend to deposit all of their energy, and
electrons only part of theirs, the two radiations can be separated by their different pulse
heights.
Geiger; No differentiation can be achieved, because pulse height is independent of
particle type and energy.
4) Proportional; The maximum counting rate is often set by pulse pile-up. ‘The minimum
pulse shaping time (that will minimize pile-up) is limited by the finite rise time of the
pulses.
Geiger: The maximum counting rate is limited by the long dead time of the tube itself.
©) Proportional: Gamma rays produce very small amplitude pulses and are often below
the discrimination level.
Geiger: Counting efficiency is a few percent due primarily to the liberation of
secondary electrons from the detector walls.
a) The response to gamma rays is due to interactions in the solid wall of the detector.
‘These interactions result in the liberation of secondary electrons that ionize the gas. The
free electrons so formed subsequently drift to the anode wire and create an avalanche. The
efficiency for counting gamma rays depends on a) the probability that the incident gamma
ray interacts in the wall to produce a secondary electron, and b) the probability that the
secondary electron reaches the fill gas before the end of its track. Process a) generally
increases with wall thickness. Process b) is initially a weak function of wall thickness, but
decreases when the distance of the average point of interaction from the inner wall exceeds
the secondary electron range. With large wall thickness, attenuation of the incident gamma
rays ultimately decreases the efficiency.
) tu is typically 1-2 mm, and is approximately equal to the maximum secondary electron
range.
If another event occurs during a dead time, it can trigger another Geiger discharge, but of
smaller then normal amplitude. Thus it normally will not be counted, but will extend the
time needed to clear the tube of excess positive charges. The tube will then behave more
closely as a paralyzable detector. 7-8. The critical radius for avalanche formation is obtained using Eq = W/lrax In(b/a)], 80
ert = W/E it In(b/a)] = 1500 V/f(2 x 106 V/m)(In(2/0,005)] = 1.25 x 102m
drift distance = d = b — ra, = 2.0 — 0.0125 = 1.9875 em
tein = (drift distance)/(drift velocity) = d/[HcE.vg/P] = Pd/ME wg
b
c 1 f vo In(b/r,.) 1500V___In(2,/0.0125)
vs r. T In(b/a) b= Tai, In(b/a) (2.0 - 0.0125) cm In(2./0.005)
rit it
E jug = 639 V/em
(0.5 atm) (2.0 —0.0125)em / (1.5 x 10-4 m? atmn/s-V) (639 V/cm) (10-4 m?/em?)
04 ms
7-9. For the non-paralyzable case with m
I-mt 1-977
1
= Y= =2857/s.
7 350us /
fase with m=")
For the paralyzable case with m %
Solving:
= 102 _ 0.693 _roso/s.
Solving: m= =
7-10 a) Setrt=0.05 since rt << 1
2s
——*___ = 2207 number of intervals sampled
(900+ 0.3+ 6.0)us
=0.0213 or 2. 13% BL.
82.
8-4.
85.
CHAPTER 8
int = (energy converted to light) / (total energy) =n E, /E
E, = he/A = (4.135 x 10-5 eV-s) (3.00 x 108 m/s) / (447 x 10-9 m) = 2.775 eV
© cint = (20300 photons) (2.775 eV/photon) / (106 eV) = 5.63%
1() = Ip exp-ve]
y(t) = JT.dt = [gt] [1 — expla}
y(W/y(ee) = 1 — expl-t/t] = 0.99
t= tn 100 = (230 ns) (4.605) = 1.059 ps
‘Maximum brightness (photons/time) occurs at time = 0, and is proportional to scintillation
efficiency and inversely proportional to decay time.
Ratio = [E,cigi(Nal)/ €gciy (anth)][t(anth)/ t (Nal)] = [230/100}{30ns/230ns] = 0.30
‘Thus, even though the light yield from anthracene is less, its maximum brightness is.
‘greater than that for Nal(T1),
Organics generally have faster response (although several fast inorganics are in use).
Inorganics have higher light output, are more linear, and have higher detection efficiencies
for high energy gamma rays.
‘The cost depends on the application, but generally favors plastic or liquid organics for
large volume detectors.
Activators are used in inorganic sciritillators in order to increase the probability of visible
photon emission. Transitions across the band gap of the parent crystal tend to yield
photons with energies higher than for visible light, and these photons can be reabsorbed,
Activators introduce intermediate energy levels and photons that can be transmitted
efficiently through the crystal. They are not required in organics because the scintillation
process arises from transitions within the energy level structure of individual molecules.
Nal(T]) has the highest scintillation yield when measured using standard PM tubes, but
CsI(TI) has a higher yield when measured using photodiodes with broad spectral response.
‘The scintillation process in organics arises from transitions within the energy level
structure of a single molecule, and can be observed from a given molecular species
independent of its physical state. Inorganic scintillators, however, require a regular
crystalline lattice as a basis for the scintillation process.
6, = sin“! [ngie/Mplastic] = sin} [1.0/1.58] = 39.27°
an %
foo J 0 sind
é
fraction of light that escapes =" —° _____ = [cos(0) — cos(@,)] = 0.226
Qn 8-9. # scintillations/beta particle = (scint eff.)(Ep )/ Ephoion = (Scint eff. (Ep Y/ {he/A]
= (0.13) (108 eV) / [(1.24x 10-6 eV-m)/(410 x 10-9 m)] = 43300
2 2
Rrrupit _ (0.15 cm) 5
Qigaeg = = EO = 5.63 x 10
act An deigy 4 (10 cm)
photons received per flash = (# scintillations/beta particle)(Qje,) = 2.4
Flash is not visible.
8-10. a) Since a 1 MeV electron is a “minimum ionizing particle” (see Fig 2.1), its
energy loss is about 2 MeV per g/cm’ in light materials (see p. 32).
Energy loss AE. ud 1.032. .0.3mm
elem om
AE = 60keV
b) From 8.5, assume a light yield of 10 photons/keV.
Therefore, light yield = 600 photons.
c) From Eq. 8.14,
1 9.
=a-—%)=0.0285
2°" Tse
= 0.606
Im
Transmission over Im = exp(->™)
m
‘Number of transmitted photons = (600)(0.0285)(0.606) =10.36.
CHAPTER 9
9-1. Ey =hefh = (1.24 x 10-6 eV-m) / (1.5 eV) = 827 nm
9-2. Vavg = 0.5 [2Ep/m]!? = 0.5 [2(150 eV)(1.602 x 10-19 YeV) /9.11 x 10-3! kg]!
= 3.63 x 106 m/s
t= x/Vavg = 12 103 m/ 3.63 x 106 m/s = 3.30 ns
9-3. e- gain = 106 = 55 — 8 = 10 > E(e*) = 200 eV — V = 200 volts per stage.
Overall voltage = (200V/stage)(6 stages) = 1200 V. 9-4, Emission rate of e- = Tia/ gain = [2x 10-9 A / 108 ] (e-/ 1.602x 10-9 C)
= 12,500 e/s
9-5. Gain = BN = [(Vs)0-5]N = [(V/N)25)N = [(V/10)95]!9 = 10-6 VS
AGain/AV = 6 x 10-6 V5
AV = AGain / 6 x 10-6 V5 = (0.01 Gain) / 6x 10-6 V5 = 0.01 V/6 = 0.01 (108 volts) /6
AV = 1.67 volts
96. vip ee ry
1-0 ©
‘To find the time at which the pulse reaches its maximum, differentiate V() with respect to
time, and set equal to zero. Solving for this time, trax = [1/(@-A)]in(O/A). Inserting this
value for t gives the maximum amplitude for a given time constant (1/8).
‘The umplitude reached by the pulse using an infinite time constant is just Q/C.
Note that 4 = 1/230ns = 4.35 x 103 ns.
1/0 (ns) tax (M8) Vtmaxd (QC)
10 328 0.038
100 147 229
1000 439 645
7970 840 900
‘The last entry was found by trial and error to give the value 0.9 in the last column,
9-7. V@=0
in general, i() = CV’+ VR
for0 I=CV'+W/R > V(t)=RI(1-e*RC)
forT V()=RI(1—eTRC) eRe
Sketches:
RC>>T RC<> t, the maximum pulse height V = Q/C
1 = E-tscin/hv = (1.2x10° €V)(0.12/(3 eV/photon) = 4.8x10* photons
Q= n(light coll. eff:(photo cath. eff.)(1" dynode eff.)(gain)ge
= (4.8x10°)(0.7)(0.2)(0.8)(100,000)(1.62x10" C) = 8.71x10"' C
max. pulse height = Q/C = (8.71x10"! Cy(10" F) = 0.871 V
Microchannel plate PM tubes show excellent timing properties. The total electron transit
time through a channel is a few ns, compared with 20 — 30 ns for conventional multiplier
structures. ‘The spread in transit time, which determines timing performance, is about
100 ps, a factor of two or three better than conventional PM tubes.
1. = he/E = (1.24 x 10-6 eV-m)((1.11 eV) = 1.12 pm
Energy deposition rate = (5x10° eV)(10°) = 5x10"? eV/s
1.240%10° €V-M _ 9 gs eviph
Average photon energy =
420x10" m
Photon emission rate = 2%10° €V/s%0.03 _ 5 98.10" pis
2.95 eV/ph
(5.08 x 10!° ph/s)(0.80)(0.75 e/ph)(1.6 x 10°? Cle’)
4.88 x 10° Cis
4.88 0A
Assuming 3.62 eV/e-h pair in Si (Table 11.1),
V = (5000)(3.62 V) = 18.1 kV
CHAPTER 10
1+ (E/mge?) (1 cos 8) 1+ (E/mgc*)
Solving for E,
MeV
T-O5MeV/0511 MeV
= 23.23 MeV 10-2.
10-3.
10-4,
10-5.
10-6.
10-7.
10-8.
Using the formula for Compton scattering, with E = 2 MeV and @ = 309, E’= 1,312 MeV
Using the formula again, with E = 1.312 MeV and @ = 609, E’= 0.574 MeV
The energy deposited = 2 - 0.574 = 1.426 MeV
This result is independent of the sequence of scattering.
For any event, maximum energy deposition occurs for @ = 180°, so that
e-—2
1+ QE/m,c’)
A.1 MeV photon will yield a 0.204 MeV photon, which in tum will yield a 0.113 MeV
photon. The energy deposited = 1—0.113 = 0.887 MeV.
For Nal, the index of refraction n = c/v = 1.85
v= (3.00 x 10! cmys)/1.85 = 1.62 x 1010 cm/s
t=d/e = (3 cm)/(1.62 x 10! cm/s) = 1.85 x 10-10 s = 0.185 ns
This is very small compared with the 230 ns scintillation decay time. ‘Thus, the light from
both events sum,
Ina detector of realistic size, the photopeak will be enhanced by some histories that begin
with Compton scattering or pair production, but for which the full energy is eventually
absorbed. ‘Thus, the peak-to-total ratio will be larger than 1/23.
R=K/E}*S
K=Rey [Ec]? = (7%) [0.662 MeV]95 = 5.695
Rya = 5.695 / [By]°5 = 5.695 /[1.28 MeV]95 = 5.03%
€4c= [# of pulses recorded] / [# of pulses incident] = 1 - exp[-(up)pt]
€ = 1 - expl-{0.955 em?/g) (3.667 g/cm?) (0.5 cm)] = 0.826 = 82.6%
photofraction = {area under photopeak] / [total area] = jp /€
Eip = €j (Photofraction) = (0.826) (40%) = 33.1%
a) For 1.17 MeV gamma rays and @ = 1, the Compton scattering formula yields
B’=0.21 MeV
‘The energy of the Compton edge = 1.17 - 0.21 = 0.96 MeV 10-9.
10-10.
10-11.
10-12.
b) Calculation of backscatter again makes use of © = 7 in the Compton scattering formula.
‘The results are E’(1 MeV) = 0.204 MeV
E’2 MeV) = 0.227 MeV
E’ MeV) = 0.235 MeV
Intrinsic peak efficiency: a and c. Energy resolution: b, f, and h
RC = (105 Q) (1040 F) = 105 s
+ for Nal(T!) =230x 109s
Since RC >> +, the maximum pulse height V = Q/C
N= E€gzin/hv = (105 eV)(.12)/(3 eV/photon) = 40 000 photons
=n (light coll. eff.) (photo cath. eff.) (Ist dynode eff.) (gain) q.
Q= (4.0 x 104) (0.5) (0.2) (0.8) (2.51%) (1.602 x 10-19 C) = 4.89 x 10-12 C
max. pulse height = Q/C = (4.89 x 10-12 C)/(10-10 F) = 48.9 mV
a) [counts sumJ/[counts Y, peak] = Ny2/fNy - Nyal = [Ny/Ny2 - 1
=[(S ey; AB / (Se €2y1 y2 OP} - 1
(1/ eny2 Qe} — 14
= [ery2Qq/ 01 —
Q=2n(1- = - an(a- = 0.663
e +e Vie?
Q, = Q/4x =0.0528. Given e, = 0.3 and y= 1,
{counts sumJ/[counts 7; peak] = 0.0161.
b) Nig = Seep; y2 QP = (105s) (0.5) (0.3) (1) (1) (0.0528)? = 41.8 st
Tey =1yt hy + ryTTy (assuming rt <1)
= 2 ryty = 2t (GQpe,) (SQrep) = 23 x 10° s)(105/3)? (.0528)? (0.5)(0.3)
ere
Rp = 06 g/cm?
p =3.667 g/cm?
2 R=0.164em
Veo escape = Med? hyea = 1 [2.54 em — 0.164 em]? [5.08 cm —2 (0.164 cm)] = 84.3 cm?
?h =m (2.54 cm)? (5.08 cm) = 102.96 cm?
Vescape /Veyi = [Vey ~ Vio scape] / Vey = 0.181 10-13.
10-14.
10-15.
10-16.
11-2,
8) The source emits positrons, which produce annihilation radiation.
b) If the energy of the primary gamma rays is high, pair production within surrounding
‘materials can give a significant yield of annihilation radiation.
FWHM = R Hp = (0.085) (662 keV) = 56.3 keV
FWHM = 2.356
0 = 56.3 keV/2.35 = 23.9 keV
Low atomic number scintillator materials are used for electron spectroscopy because the
probabilities of backscattering and bremsstrahlung, which are undesirable processes,
Increase with increasing atomic number.
High atomic number scintillator materials are used for gamma ray spectroscopy because
the probability of photoelectric absorption increases rapidly with atomic number.
‘The major advantage is higher efficiency due to avoidance of sample self-aborption and
attenuation by detector entrance windows.
CHAPTER 11
Doping becomes significant when the dopant concentration is comparable with the intrinsic
cartier concentration. From Table 11-1,
For Si: 1.5 x 1010 fem? carrier concentration
4.96 x 1022 atoms/em?
impurity level ~ 1.5 x 101° / 4.96 x 1022 atoms/em? = 3.0 x 10-13 = .0003 ppb
For Ge: impurity level = 2.4 x 10!3/ 4.41 x 10% = 5.4 x 10-10 = 0,54 ppb
# carriers in silicon / # carriers in air = [ Ep / Wsi] / [Ep / Weg] = Waie / Ws
= (35 eV/e~) / (3.62 eV/e- hole pair) = 9.67
ng = E / Wg; = (105 eV)/(3.76 eV/e~ hole pair) = 26,600 e- hole pairs
variance = Fng = (~ 0.1) (26000) = 2600 11-8.
11-9,
11-10.
A.
p(T) = CT3? exp(-E, / 2kT)
(300 K) = C (300!) exp(- (0.665 eV)/[2 (8.62 x 10-5 eV/K) (300 K)]) = 0.0135 C
Interpolating to find E, at 77 K,
E,(77 K) = 0.746 eV + (0.665 eV — 0.746 eV)(77/300) = 0.7252
O77 K) = C (77°) exp(— (0.7252 eV)/{2 (8.62 x 10-5 eV/K) (77 K)]}) = 1.27 x 1021. C
Rate is reduced by p(300 K)/p(77 K) = 1.07 x 1019
The positive terminal is connected to the n side, since then one is attempting to draw
electrons across the junction from the p side, where they are the minority carrier.
[rv / [2VNe
a d=\/ Ce °F :
(The above relationships are valid only for partially depleted configurations. For fully
depleted configurations, see discussion in text.)
Since the dopant concentration N appears in the denominator of Eq. 11-18, the
depletion depth is maximized by keeping its value small. This can only be
accomplished by starting with material of high purity or high resistivity.
Bias voltage ~ 35 volts,
At perpendicular incidence, 461 = K(Ep - AE), where K is the multichannel analyzer
calibration in channels/unit energy.
At35°: 455 = K(Ep — AE / cos 35°) = K(Ep — 1.221 Fy)
‘Subtracting second line from first: 6=K(O.221AEy) or ABy=27.2/K
‘Substituting this result in the first line: 461=KE 27.2. or K=488.2/Ey
Thus: AEg = 27.2 E/ 488.2 = (27.2)(5.486 MeV) / 488.2 = 305 keV
‘The capacitance of the detector increases with surface area, adding to the noise figure at the
input of the preamplifier. Also, detectors with larger area also have the potential for larger
Teakage current at the edges of the junction.
Veat = 107 ems for both holes and electrons
Jv =0.01 em/ 107 cm/s = 1 ns 11-12. Vyutse = QC
CHK (Vop95
Vout = QVop)5 Mc
AVisne AV,
= 05 P= 0.5 (5%) = 2.5%
pulse op
11-13. Serious changes appear at 10"! afem?,
Alpha flux ~ Activity/(4nd2) = (10 x 108 o/s) / (4n(10 cm)2) = 7958 o/em?s
t= (10! a/em?) / (7958 afem?-s) = 1.26 x 107s
=145d
11-14. Calibration is (5.486 x 106 eV — 0J/[116-0 ch] = 47300 eV/ch
Pulse height defect = true energy — apparent energy
7 = 21.0 x 106 eV — (47300 eV/ch) (402)
= 1.99 MeV
11-15.
aN/GE =
bog
2.5 5 E(MeV)
‘Note that the alpha particles will have a dE/dx value that increases with decreasing energy,
so they will lose less than half their initial energy over the first half of their range.
Energy loss straggling will add width to the energy distributions in cases b and c.
11-16. Will use Eq 11.6.
First find “2 at 77K = (0.0253 Vy( 2.) = 6.65 x 10° V
e 293
=
a= PHBE = [6952107 2G) _ 5 65.103 = 36m
cE 10°V/em
(Note that this value is independent of the assumed mobility since the product of
the diffusion coefficient and the drift time does not change.) 12-1,
12-2,
12-3.
12-4,
12-5.
CHAPTER 12
35 (F/N]°S
IfF=F/2, then Ryey = 2.35 [F/2N]05
Raoy /R = 2-05 = 0.707
E=V/d = Vy/th = Vo/ be
From Fig 11-2, gq, = 107 cm/s for both electrons and holes.
AtTTK, Hy = 4.2 x 104 cm?/V-s and j1, = 3.6 x 104 cm?/Vs.
Using the smaller value of j1,
V = Ved / He = Var d/ He = (107 ern/s) (1 em) / (3.6 x 104 cm?/V-s) = 278 V
Assume all holes and electrons must travel 10 mm to be collected.
‘Want maximum transit time <.001 T
(1 cm) / (107 cnys)(103) $T or T2 1045
W2=W2+W2+W2
Assume charge collection is virtually complete, so W,? = 0.
W2=W2+W2 = (2.35) Fee +W2
W2 = (2.35)? (0.08) (2.96 eV) (140 103 eV) + (1.2 x 103 eV)?
W, = 1.274 keV (or 0.91%)
Solving for E in the formula
2
2E/mye
E,=E (5)
1+ 2B/mg¢
= 1.16 MeV
(with mpc? = 0.511 MeV) we find that E = 1.375 MeV.
‘The location of the Compton edge (1.16 MeV) is detector independent, because it is due to
interactions with free electrons.
Escape peaks are more prominent because of the greater transparency of germanium to
annihilation radiation (due both to its lower atomic number and typically smaller detector
volume). 12-6,
12-7.
12-8,
13-1.
13-2,
13-3.
13-4,
13-5.
Absolute efficiency of 7.62 x 7.62 cm Nal at 25 cm for 1.333 MeV = 1.2 x 10° (standard)
Eat = (1.2 x 10-4) (25 om/40 em)? = 4.69 x 10+
Ge = 0.40 Eat
T= Ege @ = (0.40) (4.69 x 10-4) (1.5 x 105 s~
= 28.151
W2 = W,? + W,2 + W.2 = W4? = (2.35)? FeE.
W2 = (2.35)? (0.08) (2.96 eV) (0.662 x 105 eV)
W, = 930.4 eV (or 0.14%)
E=2.10 MeV —2 (0.511 MeV) = 1.078 MeV (double escape peak)
CHAPTER 13
a) The fluorescent yield in Si is smaller than that in Ge.
b) The incident radiation penetrates to a greater depth (photoelectric cross section in Si is,
small).
©), Si X-rays have low energies (1.8 keV) compared to Ge (11 keV), so their probability
of escaping the detector is relatively small.
tex 5
V=pe=pV/x (use p for holes, because they are less mobile than electrons)
cs te = 2/ MV = (04 cm)? / (1.1 x 104 em? /V's) (2000 V) = 7.27 ns
We should be able to resolve two peaks separated by 1 FWHM.
CuK, =8.028 keV CuKgy = 8.048 keV Zn Ky =8.616 keV Zn Ky, = 8.639 keV
Must be able to resolve Cu Kg; from Zn Kga, or AE = 8.616-8.048 = 0.568 keV.
R = AEE, = 2 (0.568) / (8.048+8.616) = 6.82%
Low energy X-rays fail to penetrate the window materials and the Si dead layer.
N
Lip Qn
Ep = 4TNy/SyQ
‘Sy=Ny/ £494 = (146835 counts) / (15 min)(0.83) = 11794 min?
Sx = 0.58638, = 6915 min?
Q= Ald? = B em?)/(10 em)? = 0.030
Ep = 4x (730 counts / 60 min) / (6915 min-!)(0.030) = 73.7% 13-6. Wp=2.35VeFE
For an assumed F = 0.11 and € = 3.76 eV in silicon:
Wp = 2.35y(3.76eV)(0.11)(59.5x10'eV) = 369 eV
In germanium, assume values of F = 0.08 and € = 2.96 eV:
W, =2.35y{(2.96eV)(0.08)(59.5x10eV) =279 eV
‘The result expected for a silicon drift detector is identical to that shown
above for a Si (Li) detector since € and F are characteristics of the detector material.
13-7. The following attenuation coefficients at 662 keV were obtained from the code
PHOTCOEF. (Very rough estimates could be obtained from Fig 13.22)
Interaction coefficients
Material Photoelectric ‘Compton
emg" em emg? em™
Si 9.1603x10° | 2.1258x10% | 7.6740x107 | _1.7084x10"
Ge 1.7794x10° | _9.4752x10°_ | 6.7619x107 | 3.6007x10"
Ca 7342310" _|_6.3511x107 | _6.5317x10" | 5.6499x10
Te 9.2840x10° | _5.7932x107_[_6.2296«107 _|_3.8873x10
Hg 4.0326%10 5.4626x10" 6.0549x107 | 8.2020x10"
1 1o171x107 | 5.0145x107 | 6.382110" _ | 3.6007x10"
CAT (8-606 gem") | 8.3746x10° 5.9750x10" |" 6.371110" _|_3.8609x107
Hel @-64pem') | 2.3483x107 | _1.5029x10" | 6.237710" | 3.9921x10™
Hee Hye + Hoon
He
Foo _ Foto
‘Material Tin Foto Fo
Si 3.89 om 0.0012 0.9988
Ge 1.88) 0.0256 0.9744
Cae 159 0.116 0.884
gk. 1.26 0.274 0.726 13-8.
14-1,
14-2,
14-3.
14-4,
At 10 keV in Si, linear attenuation coefficient for photoelectric absorption is about
80 cm" and Compton contribution is negligible (see Fig. 12.22).
Mean free path:
1
pele
5 =0.0125em
# 80cm
‘Thus a thickness of 300 jum absorbs nearly all incident X-rays.
a) Readout time = 2° _ 9.6555
eH:
<. Exposure time = (20)(0.655 s) = 13.1 s
In this time, want only 5% of pixels hit,
“na = (0.951 256)1256) _ 95015
131s
b) Inone interaction, deposit
= OY __9.76x10° electrons
3.62eV /e™
Since multiple hits are rare, this is minimum storage requirement.
c) Max leakage charge per pixel = (0.1)(2.76x10" e*) = 276 electrons
(276)(1.6 x10" C)(256)*
Max leakage current =
(3.1)
=22x107A
CHAPTER 14
Such detectors if operated in the ionization region produce very small amplitude pulses that
are less convenient to process than those produced in the proportional region. If operated
in the Geiger region, all information on pulse amplitude is lost, and background or gamma
ray induced events no longer can be discarded by amplitude selection.
VengeM/C, ng = E/W = (0.764 x 108 eV)/ (42.7 eVipair) = 17900
V = (17 900)(1.602 x 10-19 C)(1000) / (100 x 10-1? F) = 28.7 mV
increase in efficiency ~ (final concentration)/(initial concentration) = 96 /19.8
e= 1—expl-2,L] = 1 -expl-No,L]
N= 0.96 (n/V) = 0.96 Ay PRT
.96 (6.02 x 10% g-mol-!)(80 x 10 Pa) /(8.314 J/g-mol-K)(300 K)
85 x 105 m3
3840 barns = 3840 x 10-28 m?
—expl- (1.85 x 1025 mr)(3840 x 10-28 m2)(0.10 m)] = 0.51 145.
14-6,
14-7,
148.
14.9,
14-10.
14-11,
15-1,
He-3 counters can be operated at much higher pressures and are therefore preferred for
those applications in which maximum detection efficiency is important.
‘The continua caused by the wall effect are expected to be rectangular shaped only in planar
geometry. For tubes of small diameter, the curvature of the wall causes some particles that
would otherwise deposit all their energy to deposit only part of it. The effect is more
noticable for the alpha particle than for the lithium particle continuum because the range of
the alpha particle is larger.
For interactions in the thin boron layer, there are two possibilities:
1) The alpha particle escapes the wall with no energy loss. 1.47 MeV is deposited.
2) The Li-7 nucleus escapes the wall with no energy loss. 0.84 MeV is deposited,
Thus the pulse height spectrum should consist of two simple peaks of equal areas located
at these two energies.
Using data from Table 8.3 for excitation by fast electrons (will overestimate the
actual light yield for heavy charged particles):
11,000 photons/MeV in Li(Eu), 3500 photons/MeV in Li glass,
‘The Q-value for the °Li(n,a)°H reaction is 4.78 MeV.
For Lil(Eu): N = (4.78 MeV) x (11000 photons/MeV) = 52600 photons
For Li glass: N = (4.78 MeV) x (3500 photons/MeV) = 16700 photons
Once the thickness of the deposit exteeds the range of the fission fragments in the deposit
material, the fragments can no longer escape from the regions farthest from the deposit
surface.
Number of beta particles = I/¢ = (10-9 C/s) / (1.602 x 10-19 C/e-) = 6.24 x 10° e-/s
N(®/No = expl-Ga6t] = expl-(150 x 10-2 em?)(3.0 x 1083 enr2s-1)(1.577 x 107 s)]
=0931 The sensitivity thus drops by 6.9%.
CHAPTER 15
Neglect absorption of neutrons in I, and assume 100% enrichment of SLi.
e=1—exp[-Nogt]
N = (6.02 x 10 nucl/mole)(4.08 g/om?)/(132.9 g/mole) = 1.85 x 10%/cm?
1MeV: = 1 —expl—(1.85 x 1022 cm?) (0.25 x 10-2 cm?) (0.4 em)] = 0.18%
Thermal: = 1 -exp[- (1.85 x 1022 cm’) (940 x 10-24 cm2) (0.4 cm)] ~ 100% 15-2.
15-3,
15-4,
15-5.
a) At small diameters, the neutrons may not be moderated enough to be detected with good
efficiency.
b) At large diameters, the number of collisions in the moderator will tend to increase,
leading to a lower value of the most probable energy when the neutron reaches the
detector. However, the probability tht an incident heutron even reaches the detector will
decrease as the moderator is made thicker.
©) As the detector becomes a smaller fraction of the total volume of the system, there will
be a lower probability that a typical neutron path will intersect the detector before escaping
from the surface of the moderator. Futhermore, a neutron may be absorbed within the
‘moderator before it has a chance of reaching the detector, because Gis larger at lower
neutron energies.
Counts Te
Sphere Diameter
Because virtually all reaction products are fully absorbed, the pulse height spectrum will be
a simple peak centered at an energy of 4.78 MeV.
Neglecting slowing down time, t = d/v = (10 cm) / (2.2 x 105 m/s) = 45 us. This time is
long compared with electron drift times in the BF; tube (few microseconds), so the time
delay between neutron incidence and measurement of the leading edge is controlled by the
neutron diffusion time.
energy: Eg +Ep=E,+Q — po4/2mg + pr/2mp=p,2/2m,+Q (Pi =0)
Multiplying through by m, and making the approximations m, = mg/4 = my/3,
Pafl4 + Pr°/3 = PyPt 2m,Q
Momentum conservation: pa+Pr=Pa (Pri= 0)
Eliminating p, and solving for pr yields
6p, +288 pe+672m,Q
P=
14
Use the negative sign, because pr is negative. 15-6,
15-7.
15-8.
Inserting p, = [2mE,]°5 = [2 (1.009 amu) (3 MeV)]05 = 2.46 and Q= 4.78 MeV,
99
Ex = pr2/2my = (3.992) 2 (3 x 1.009) = 2.63 MeV
Ey =-Ep +E, +Q=-2.63 +3 +4,78 = 5.15 MeV
3He+!n 33H + 1p
Assume helium nucleus is at rest.
Energy conservation: E,+Ep=E,+Q or
Pp?/2my + py?/2my = p,?/2m, + Q
“Multiplying through by m, and making the approximations mp = m, = m/3,
Pp? + Pr¥3 = p,? + 2m,Q
‘The maximum proton energy occurs when the proton is emitted in the forward direction
(0) and the triton is emitted at 180° to the incident neutron's direction,
‘Mornentum conservation: pp +Pr=Pa
Eliminating py and solving for pp yields
2p, + 4/36 pe +96 Q
g
Inserting p, = [2mE,]°5 = [2 (1.009 amu) (1.5 MeV)]?4 = 1.74 and Q = 0.765 MeV,
Pp= 213 :
E, = p,2/2m, = (2.132)/2 (1.009) = 2.24 MeV
‘The epithermal peak corresponds to the detection of incident neutrons which have been
reduced to the thermal range by moderation in external materials. All such neutron
interactions deposit an energy equal to the Q-value of 764 keV.
‘This type of detector makes use of elastic scattering. For CH,,
[1-expl-y,,5, +Noo.) d]
y= 15 bams G¢=4.5bams d=Scm
For an ideal gas at 1 atmosphere and 300K, N = Ayn/V = AyP/RT, or
N = (6.02 x 10% g-mol-!) (1 atm) / (82.06 cm} -atrm/K/g-mol) (300 K)
45 x 1019 em,
Na=4N Nc=N
Inserting these values, € = 0.73%. 15-9,
15-10.
15-11,
Er +En=Eno (eq.1)
Ep = [4A/(1+A)?] (Cos0g)? Ey, (A~1)
Eg = (cosOg)? Eno (eq. 2)
Pp Sing + Pp Sin®,.¢=0 (conservation of y-momentum)
(x sing)? = (Pye sind,
(x)? (1 ~ (C088g)*) = pe sin,
Substituting p? = 2mE, and noting that mg = my
Eg (1 — (cosOg)* ) = Ey, (1 - (cos, 9)? ) (eq. 3)
Using eq. 1 to eliminate Ex and eq. 2 to eliminate cos®g, and then solving for Ey yields
Ens Eno (C090 ,,0)?
E,,,= 1 MeV (cos 40)? = 0.587 MeV
t= xWv = x/[2E,, fm]
t= (0.03 m)/ [2 (0.587 x 10 eV) (1.602 x 10-19 J/eV)/(1.675 x 10-27 kg)]95 =
Events cannot be resolved.
83 ns
Scattering from hydrogen is isotropic in the center-of-mass system for these energies, so
‘we would expect each neutron group to generate an idealized rectangular response function
extending from zero to the full neutron energy. If each neutron group is equally intense,
each rectangle will be of equal area. ‘The measured specrum will then be the superposition
of these three rectangles:
aN
a
15 150 300
Energy (keV)
From the solution shown for Problem 15-9:
ER +Ent=Eno
Eg = EyoC0s?O and Egy = Eo C0S* Ons
Thus: cos? @ + cos? @,.¢ = 1
0s? Og = cos? On
Cos OR = cOsO,¢ 0 Og = 90°— Ong 15-12,
15-13.
15-14,
15-15.
15-16.
16-1.
Letting a = 0.005 cm, b = 2 cm, p = 0.75 atm, V = 2000 V, AV = 36.5 eV, and
K = 6.9 x 104 V/em-atm, the equation
ven v
“In@a) AV" Kpainiblay
yields M = 5.03,
V=QIC= [ng e MYC = [{E/W}eM] /C = EeM / [WC]
‘V = (105 eV) (1.602 x 1019 C) (5.03) / {(29 eV) (60 x 10? F)}
V=0.463 mv
Ep = [4A/(1+A)?] (cos0,)? Ey
At 900, Ep =0.
Elmax = (4A/(1+A)?] E,, = [4 (28)/(29)?] (1 MeV) = 0.133 MeV
Scattering from hydrogen at 5 MeV is isotropic in the center-of-mass system, so that o(8)
does not change with © and is equal to the constant 6, /4n. Scattering from helium is not
isotropic in the center-of-mass system,
‘The radiator thickness must be kept small in order to avoid appreciable energy loss of the
recoil protons before they leave the radiator.
‘Number of B-10 nuclei per unit volume:
N= 0523x193 852i Mole 6. 99¢19% WUCLB 9 jog mucl B-10
gsciut em’ 10.8gB mole nucl B
= 5.68 x 10° B-10 nuclei/em*
Mean free path:
—————EeEE EEE
= No (5.68x10" /cm')(3840x10 cm?)
+ Average time to capture:
A _ 0.458em
124 = DSN _ 7 08;
Vv 2200m/s—
This time neglects the time spent by the neutron in the fast and epithermal energy
ranges.
1458cm
CHAPTER 16
t= dWv = (15 m)/(0.659)(3.0 x 108 m/s) = 75.9 ns 16-2.
16-3.
16-4,
16-5.
16-6.
16-7.
16-8.
‘By monitoring the reflection conditions on an oscilloscope, the termination resistance is
varied until no reflections occur. At that point, the characteristic impedance of the cable is
‘equal to the termination resistance value.
4) rise time = 0.50 us
transit time = (20 m)/(0.659)(3.0 x 108 m/s) = 0.10 ps
rise time >> transit time, so pulse is slow, and no termination is necessary.
b) rise time = 10 ns
transit time = (10 m)/(0.84)(3.0 x 108 m/s) = 39.7 ns
rise time << transit time, so pulse is fast, and termination may be necessary.
a) Use RG-S8C/U with characteristic Z of 50 Q (or any other 50 Q cable). Use a shunt
resistance at receiving end of 50 Q.
b) Amplitude that appears in second component after proper termination is V,/2= 2.5 V.
Ry =Ro (@—- D(a + 1) = 40.912
Ry = Ro (20)(02 - 1) = 10.100
Find equivalent impedance looking into left terminal if loads 1 and 2 are 50 Q.
3R=Ry
R= 50/3 = 1662
Find time needed for pulse to decay to 1% of its amplitude.
AW =AcexpCyt)
t=~tln (A(O/A,) = -50 is 1n(0.01)
=230 ps
V=E (1 ~exp (k))
RC integrator: 16-9,
16-10.
16-11.
The integrating factor is p = exp(/ dt /t) = exp (\/t). Multiplying the equation by p yields
dE XPV) EB
a |] C1 ~exp-vk) expr).
Integrating yields
Eu, x(t) = E exp(tét) — SS exp(tk) exp(v/t) + C.
Att=0, Eyy:=0. Therefore, C= Et/(k-1). Dividing through by exp(t/1) yields
Et
= exp(-t/k) + c=<
exp(/t)
Ey =E- zo
Want Boy = 0.5 Big
Eou/Ein = JAI sin Qn ft +6), so A] =0.5
lal =(.+ @y/pr9s=05
£=0.577 f;
f, = 1/{2nt] = 1/[2nRC] = 1/[2x(500 2)(500 x 10-12 F)] = 6.37 x 105 Hz
f = (0.577)(6.37 x 105 Hz) = 3.68 x 105 Hz = 368 kHz
For this case, Eau = E [Vt] exp(-t/t). To find the amplitude of the pulse, find the
maximum. Taking the first derivative and setting it equal to 0, and dividing by E,
O= [1/t] expt) - [ve] exp(-v/t)
Simplifying yields t= *. Therefore
out =E /e = (1 V)/(2.718)
368 V
Input to RC stage is E expl-t/ty]. The equation to solve is
dB 1
1
et Bo Eel]
a
The integrating factor is p = exp(/ dt /%) = exp (Wt). Multiplying the equation by p,
noting the left side is an exact differential, and integrating yields
Uh
expl-t/t,] explvt,] +C
a 16-12.
16-13.
16-14,
|. Therefore, C = -Et,/[t — to]. Inserting this result and simplifying
[expl-v1,1 -expl-ve,]]
Bipolar pulses minimize baseline shift at high counting rates.
Rise time corresponds to the time over which the charge produced by the detector is
integrated across the capacitance of the collection circuit (ie. the the charge collection time
of the detector).
voltage |
10
<—>
oT ]
+ wpe b —_ >
A.=Ay
Ay =0.5 (V)(WV/10) = V2W/(20)
Ay = (10 — V)(L) + 2(0.5)(10 — V)(W/2)(1 — [V/10]) = (10 — V) [L + (W/2) (10 - V)/10]
‘Equating these expressions and simplifying yields
V=[10L + SW/L+W]
shift = 10-V = SW/[L + W] = SWE
a) W=5x10%s f=100s1 shift=2.5mV
b) W=5x10-%s f=50000s4 shift=1.25V 16-15.
17-1.
17-2,
17-3.
Input Output
Vit) vo.
200 300ns
0 200 ns 0 100
Down and back propagation time through the cable
= (20 m) / (0.659)(3.0 x 108 mys) = 100 ns.
CHAPTER 17
Let V' represent the voltage across the input terminals to the operational amplifier.
Then Voy: =-AV'. The voltage drop across Ry is Viq — V" , across Ry is V'~ Vou
Since the currents through both resistors must be the same, we use Ohm's law to equate
them: (Vig - VIR, = (V'~Vow/ Ra-
Solving for V' and equating to—V,y/A_ (from the first equation), we obtain:
Vou/ Vin = AR2/{Ry ~(A+1)Rj]. Now under the conditions that (A+1) >> Ry/R, and
‘A>> 1 this expression simplifies to:
Vou! Vin —Ro/Ry
From the diagram, the voltage across C; is just Viq. The voltage across Cris (Vin ~ Vou):
‘The input charge Q is divided between these two capacitances, so:
Q = Q#Q, = CGViq + C(Vin— Vou). Combining with Voyy= —AVin » we obtain:
-AQIIC, + (AH)Cq. Now with the assumption that A >> (C,+ Cp/ Cy
-ac;
From a signal-to-noise standpoint, itis always preferable to minimize the capacitive
loading on the preamplifier, and therefore long interconnecting cables between the detector
and preamplifier should be avoided. 17-4.
17-5.
17-6.
17-7.
17-8.
17-9.
17-10.
17-11.
set to LLD
Integral
Disc.
linear shaped Anti-
pulse] coin.
Integral
Disc.
—
seto ULD
The signal will not be as steady; its fractional standard deviation will increase.
a) long shaping time b) monopolar _¢) without active baseline restoration
For nonparalyzable systems (from Eq. 17-17):
f, = exp(-nt)/ (nt +1)
= expl-(25000/s)(4 x 10 s)] /[(25000/s)(4 x 106s) + 1] = 0.823
f (lost) = 1-f, = 0.177
For paralyzable systems (from Eq. 17-18):
f = exp(-2nt) = 0.819
f (ost) = 1-f, = 0.181
Leading edge triggering
a) CR-RC-CR double differentiating network _b) Double delay line shaping
Coincidence unit
Since the input pulses are uncorrelated, the anticoincidence output rate in the absence of
chance coincidences should simply be r,+12 . ‘The (first order) chance coincidence rate
will be given by 2tryr. Thus the observed output rate should be ry+ 1 —2tryr9 17-12.
17-13.
17.14,
17-15.
8) rq=2t fy This equation can be used if t, 7, and rp are measured,
b) The chance coincidence rate can be determined directly by temporarily inserting a large
delay in either branch of the system so that the true coincidence peak occurs well away
from the acceptance time window of the coincidence unit.
‘Typically, the velocity of propagation in a coaxial cable is 5.1 ns/m.
1= tv = (100 ns) / (5.1 ns/m) = 19.8 m
B
2.50 MeV By
Y By
1.33 1.33
117
Mo nH) fe
0 0
Assuming all nuclear levels are short-lived, there will be true coincidences between Y} and
‘Ya for the case at the left, but none will be observed for the case at the right.
1, =e1e)8 Tey = 200yf
Ryley = € 1828/21,"
Note that e;=n/S and e=ry/S
Thus ¥,/rq,= 1/248
a) Changing the solid angle does not change S or t, so it has no effect.
) Increasing the amount of source material increases S, so the ratio decreases. However,
decreasing the source size too much will result in poor statistics.
c) Increasing the resolving time t decreases the ratio.
4) Increasing the window widths from a small value will at first have no effect on the
ratio, since it is effectively a change in e, and €2. Eventually, however, increasing the
window width will no longer add true events, but will admit some pulses not associated
with the radiation of interest. ‘Thus the ratio will then decrease, since the apparent
singles rates continue to increase while the true rate does not. 17-16. a) 2¢=(125-25ns) ». t=50ns
'b) The fall width (at its base) = 140 ~ 110 ns = 30 ns
©) fey = 2tyry
ener
2er2= Ty
T= [ty / 21)
=[(5 1) /2 (50x 10-9 s)]°5 = 7070 st
17-17. 3 stages of 4 bits each requires:
3 x 2* = 48 comparators
4 stages of 3 bits each:
4x2} = 32 comparators
Fewer stages will result in lower latency time.
17-18. Mathematica Program:
Fora step function with V_i=
different digital filters:
Clear[v)
for i= 0, and 0 elsewhere, apply the following four
We define our input function with starting index 1 instead of 0:
Vid ts; ied
Vij Os; cd
‘V[-2]
o
Our filter looks like (note that indices start with 1 in Mathematica):
h= (0.25, 0.25, 0.25, 0.25)
{0.25, 0.25, 0.25, 0.25)
Lay}
0.25
The filter output is just the dot product of the filter and the input vectors over the
length of the filter. We add one to the index so the outcome[0] correctly corresponds
to t=0. 17-18, (continued)
wii : Table(Vii + 3], (4, Length(h] )]
outcome[j_] := N[Dot[vv[j], Reverse(h]}]
outcome [0]
:
Map[outcome, Table[i, {i, -5, 5}]]
{0., 0., 0.25, 0.5, 0.75, 1., 1.,1.,1.,1.,1.}
ListPlot(%, PlotJoined + True]
7
o8
0.6
O48
‘Changing the filter code to that of a "derivative" and reversing the code order:
he (1, -1, 0, 0)
{1, -1, 0, 0)
ListHictMiplotore, Tele, (4, -5, 5}]], Plchbired + False, Plctstyle > PoireSizo(0.(5]]
oe
0.6 17-18. (conclusion)
(c) Here is a "differentiator" with memory:
h= (0.5, 0.5, -0.5, -0.5}
{0.5, 0.5, -0.5, -0.5}
empracencsar-ymm ‘Bbie[i, (1, -5, 5}]], Pictbined + False, Mctstyle + Poirtsize(0.03)]
0.8
o-oo 0 e 0-@-©
4
- Graphics +
(d) Here is the exponential integrator filter:
h = Table{Exp(-(i - 1) /2), (4, 5})
i ES ne = 7
{u. = }
io ca
stelceMpfatame, BE, (5, 5)}}, Pctbined + Riles, Pctsyle ~ PoireSize(0.025)]
.
2 .
.
as
+ Graphics - 18-1,
18-2,
18-3.
18-4,
18-5,
18-6.
18-7,
CHAPTER 18
pulse height channels = 5 channels / R = 5/0,003 =
Zero offset will not affect peak spacing.
4p = A; (Gain 1/Gain 2) = 24(750/1000) = 18 channels
t.=N/f (Nis the channel number)
‘The maximum value will occur for the largest pulses, which will be stored in channel 2048.
£= (2048)/(25 x 10-6 s) = 81.9 MHz
2N = 4096 + N=12
‘The dead time models in Chapter 4 assume a fixed dead time. A Wilkinson-type ADC has a
variable dead time which is proportional to the channel number in which the pulse is stored.
a) T=N/f +B = (300)/(80 x 108 s-!) +2.5 x 10-6 s = 6.25 ps
b) T= N/f+B = (220)/(80 x 108 s+) + 2.5 x 10-6 s = 5.25 ps
Simplify by assuming all pulses are of average size.
m = n/(1 +nt) = S000/s/[1 + (5000)(5.25 x 10)] = 4870 /s
Fractional dead time = mt = 2.6%
Repeating for n= 5 x 104/s gives m=39600/s and mt = 20.8%
©) Live time = (actual time)(live fraction)
Actual time = (live time) / (1 ~ dead fraction)
In the case of n= 5000/s, actual time = 10 min/(1~.026) = 10.27 min
For n=50000/s, actual time = 10 min/(1 ~.208) = 12.63 min
Because the pulses are periodic, the frequency dependence of dead time losses will
show discontinuities when the period T equals an integral number of dead time values 1.
For T>t, or f< 11.1 kHz, there are no losses of the periodic pulses. The dead
fraction of the MCA (for randomly occuring pulses) is simply ft and ranges up to 100%
at 11.1 kHz,
For T
