Chapter IV:

Interactions of photons with matter

1
 Contents of the chapter

• Introduction

• Compton effect

• Photoelectric effect

• Pair creation

• Attenuation coefficient

2
 Basic considerations

• Photons are classified according to their mode of origin:

– ° rays accompany nuclear transitions E° = hº = Ei-Ef (E° > 100 keV)
– Bremsstrahlung (continuous X rays) result from a charged particle
acceleration
– Characteristic X rays are emitted in atomic transitions of bound
electrons between the K, L, M, … shells in atoms (EX < 100 keV)
• Momentum with p = E°/c = ħk and k the
wavenumber
• Photons interact with matter in a single event → between two
events they have no interaction with matter (unlike the
charged particles via the Coulomb force)
• Photons are indirectly ionizing radiations
3
 Different interactions of ° with matter (1)

For 1 keV < E° < 1 GeV → Classification of Fano: 4 types of

interactions and 3 consequences of the interaction → 12
theoretical processes are possible (even if a few of them are
extremely rare or were never observed)

Type of interaction Effect of the interaction

1: Atomic electron A: Coherent diffusion
2: Atomic nucleus B: Incoherent diffusion
3: Electric field of the nuclei
C:Ttotal absorption
and atomic electrons
4: Mesonic field of atomic
nuclei

4
 Different interactions of ° with matter (2)

• Only 3 effects are dominating →

1. Compton Effect (1B): the photon is scattered by a free or weakly bound
electron → the sum of the scattered photon energy and of the electron
kinetic energy is equal to the energy of the incident photon
2. Photoelectric effect (1C): The photon is absorbed by the electronic
system (atom) → it gives all his energy → an atomic electron is emitted
out of the atom with a kinetic energy equal to the energy of the
photon minus the binding energy of the atomic electron
3. Pair creation (3C): In the electric field of a nucleus or of an electron →
the photon disappears and an electron-positron pair appears
• 2 other processes can also play a role →
1. Rayleigh scattering (1A): the photon is scattered without energy loss
by an electronic system (atom)
2. Photodisintegration of the nucleus (2C): the photon is absorbed by the
nucleus and a particle is emitted (°, ®, p, n, …)
5
Correspondence atomic shell ↔ electronic configuration

Couche atomique Configuration e-

6
 Examples of binding energies

7
Energies are in keV
 Remark

• Photon cannot be absorbed by free electron (and then gives it

all its energy)
• Conservation of the energy and of the momentum with hº0 the
energy of the photon and m, E et p, the mass, the total energy
and the momentum of the electron→

• This implies E = pc+mc2 and by definition → E2 = p2c2+m2c4 →

only possible for p = hº0 = 0 → must be rejetcted

8
 Compton effect

• Scattering by a free electron (in the range of energy where

Compton effect occurs → electrons are considered as free →
when this approximation is no more true → photoelectric
effect is dominating)
• The photon gives a part of its energy to an electron

9
 History of Compton effect: Thomson

• Thomson classically calculates (1906) the scattering cross section of

an electromagnetic wave by a free electron
• Hypothese: Due to the force caused by the electric field of the wave
→ oscillation of the electron → this oscillation has same frequency as
the wave and has the same direction as the electric field → electric
dipole → the electron irradiate → diffusion of the incident wave in a
continuous way
• For a non-polarized incident wave → Thomson calculates the
diffusion cross section d¾0 in solid angle d (with µ the angle formed
with the direction of the incident wave) →

with re = e2/4¼²0mc2 = 2.8 10-15 m, the classique radius of the electron

10
 Demonstration of Thomson (1)

• Let’s consider an electromagnetic wave with frequency º that

interacts with a free electron (mass m, charge -e) → the electron
undergoes a force F due to the incident electric field:

with the direction of polarization

• The motion equation of the electron is →

• In the dipolar approximation the emitted power per unit of solid

angle is (differential equation of Larmor - see electromagnetism
teaching) →

11
with £ the angle between the polarization direction and the observer
 Demonstration of Thomson (2)

• From equation of motion we directly obtain h a2 i, the mean

square acceleration →

• The differential power becomes →

• We consider now the modulus of the Poynting vector that is for

the energy flux by second (see you know what…) →

12
 Demonstration of Thomson (3)

• The differential cross section is obtained with →

• If we consider an non-polarized incident wave → we have to

work out the average of £ →

with µ, the scattering angle

• We thus obtain the equation of Thomson →

13
 Total cross section of Thomson

• By integration over angles → total scattering cross section of

Thomson →

• In his demonstration → the incident wave is scattered in a

continuous way

14
 History of Compton effect: Compton

• Measurement of Compton (1922) → the scattered wave has

not a continuum spectrum but follows the relation →

with ¸0 and ¸1 the wavelengths of incident and scattered

photons and h/mc = ¸C, the wavelength of Compton
• This expression shows that the wavelength displacement does
not depend on Z and on the incident wavelength and that the
energy and momentum lost by the photon goes to only one
electron

15
 Demonstration of the expression of Compton (1)

• In the laboratory frame → conservation law of the four-vector

energy-momentum before and after the scattering →

16
 Demonstration of the expression of Compton (2)

• Conservation laws →

• With E2 = p2c2+m2c4 →

• And with:

17
 Relations between energies and angles (1)

• We consider E0 = the incident photon energy, E1 = the scattered

photon energy, T = E0- E1 = the kinetic energy given to the
electron, µ = the photon diffusion angle, Á = the angle
between the trajectory of the electron and the photon initial
direction and ® = E0 /mc2 = the ratio between incident photon
energy et the electron mass energy →

18
 Relations between energies and angles (2)
• As 0 < µ < ¼ → (E1)min = E0/(1+2®) and (E1)max = E0 µ
0°
5°

10°

20°

30°

60°
90°
120°
150°
180°

511 keV

255 keV →
backscattering
19
Relations between energies and angles (3)

20
 Remarks on the energy of the scattered photon

• The energy modification for the photon depends on the

incident photon energy
• For small E0 → small energy loss for the photon (for any µ)
• For E0 ↗ → the variation of the energy of the scattered photon
as a function of the angle becomes important
• For 90° → E1 always < 511 keV (= mc2)
• For 180° → E1 always < 255 keV (= mc2/2) → backscattered
photon → backscattering peak in ° spectra

21
 Relations between energies and angles (4)
• We have 0 < Á < ¼/2 → Tmin = 0 and Tmax = E0/[1+(1/2®)]

T = E0

• For large E0 → Tmax ≈ E0 - 255 keV (backscattered photon) 22

 Relations between energies and angles (5)
Á

23
Angular-differential cross section for Compton effect (1)
• Equation of Klein-Nishina (quantum electrodynamics): valid
for free electrons at rest
• The angular-differential scattering cross section of a non-
polarized photon in the solid angle d around the direction
making an angle µ with the initial direction of the photon is
given by →

with re, the classical radius of the electron

• Remark → no dependence on Z

24
 Angular-differential cross section (2)

For ® ¿ → E1 ≈ E0 → d¾ ≈ d¾0 → we obtain the Thomson cross

section 25
Angular-differential cross section (3)

26
 Energy-differential cross section (1)

• From the angular-differential cross section → Energy-

differential cross sections →

27
 Energy-differential cross section (2)

T = Tmax
d¾/dT (unité arbitraire)

28
 Total cross section for the Compton effect (1)

• After integration of the Klein-Nishina cross section over

angles →

• For ® ¿ → ¾ ≈ ¾0 = 8¼re2/3 (Thomson cross section)

σ (Thompson unit)

29
 Total cross section for the Compton effect (2)

• For ® À → ¾ → (ln ®)/® → Compton cross section ↘ when

photon energy ↗
• Indeed we observe for ® À → angular distribution
predominant in µ = 0 → no diffusion → no energy transfer →
no effect

• The atomic cross section a¾ = Z¾ is thus / Z

30
 Collision = Scattering + Absorption
• The cross section ¾ represents the scattering probability → a part of
the energy is scattered and the other one is given to the e- (absorbed)
• To characterize this aspect → we define a scattered cross section ¾s
and an absorption cross section ¾a with ¾ = ¾s + ¾a

31
 Coherent and incoherent scattering (1)

• For E0 ¿ → we cannot consider that e- are free and at rest →

scattering by the whole electronic system (atom)
• If the atom stays in its initial state → the energy of the photon
does not change but it changes its direction (the atom takes the
momentum difference) → Rayleigh scattering (coherent
scattering)
• If the atom changes its state → the photon loss energy →
incoherent scattering
• For large energies → incoherent scattering = Compton scattering

32
 Coherent and incoherent scattering (2)

• Approximation for coherent scattering → the electronic system

corresponds to a system with charge Ze and mass Zm → the
Rayleigh scattering cross section is →

• In reality → the structure of the electronic cloud implies a

decrease of the scattering → introduction of the factor of
atomic structure F →

33
 Coherent and incoherent scattering (3)

• For incoherent scattering → modification of the Klein-Nishina

cross section by the function of incoherent scattering S → S
considers the fact that the electrons of the atom are bonded →
the photon can be incapable of ejecting an electron from the
atom →

• For large energies → S → 1 → Compton cross section

34
 Rayleigh
Coherent and incoherent scattering (4)

incoherent
=
Compton

incoherent

Remark → coherent scattering ↗ for Z ↗ 35

Comparison incoherent - Compton

36
 Photoelectric effect

• The photoelectric effect is a process in which an incident

photon interacts with an atom and an electron is emitted
(process correctly explained by Einstein in 1905)
• This process of photon capture by an atom with an electron
excited in a continuous state is the inverse process of
spontaneous emission of a photon by an excited atom

37
 Energy conservation

• There are absorption of a photon with energy hº0 that is completely

absorbed by an atom and consequent emission of an electron (called
photoelectron) with a kinetic energy T out of an atomic shell
characterized by a binding energy Bi (i = K, LI, LII, LIII, …) → by
neglecting the recoil energy of the nucleus (due to ≠ of mass) →

• The energy conservation implies hº0 > Bi

• When hº0 ↗ the probability of photoelectric effect ↘ since the
behaviour of the electron more and more approaches the one of a
free electron (and the absorption by a free e- is impossible)
• The more bound electrons (shell K) have the largest probability to
absorb the photon (with always the condition hº0 > BK) 38
 Bi-2 (keV2)
Binding energy

For Z > 30 → binding energies approximatively follow Bi = ai(Z-

ci)2 (with ai and ci constant for each shell) 39
 Cross section (1)

• The cross section per atom a¿ can be decomposed into a sum of

partial cross sections (a¿i) corresponding to the emission of an
electron from a given shell i →

• The calculation of a¿K has been done for a hydrogen-like atom

in the Born approximation using a plane wave as wave function
for the emitted electron
• We suppose hº0 ¿ mc2 (non-relativistic approximation) and
hº0 À BK (interaction between the nucleus and the electron
neglected)
40
 Cross section (2)

• We find (with ®, the fine structure constant and ¾0, the

Thomson cross section) →

• As re /a0 = ®2 (with a0, the Bohr radius) →

• And as, for a hydrogen-like atom, BK = ®2Z 2mec2/2 →

41
 Cross section (3)

• Factor a0/Z → approximatively = to the atom size

• Variation with hº0-7/2 for the energy
• When hº0 ≈ BK → Born approximation no more valid →
introduction of a correction term f(») →

• For » → 0 (i.e. hº0 À BK) → f(») → 1 → previous situation

• For » → 1 (i.e. hº0 ≈ BK) → f(») →

42
 Cross section (4)

• The other partial cross sections have the same behaviour as a¿K
→ we generally write the total cross section as→

with n varying between 4 and 4.6 and k varying between 1 and

3 (C is a constant)
• For hº0 / 0.1 MeV (most important energy range for the
photoelectric effect) → n ≈ 4 and k ≈ 3
• For hº0 ' 1 MeV → n ≈ 4.5 and k ≈ 1

43
 Variation of a¿ with hº0

hº0-3

hº0-2

hº0-1

44
Variation of a¿ with Z

45
 Cross section: Example (1)

BL1(Pb) = 15.9 keV

BL2(Pb) = 15.2 keV
BL3(Pb) = 13.0 keV

BK(Pb) = 88 keV

For E > 88 keV → the 2 e- of the K shell contribute for 3/4 of the cross section (by
46
comparison to the 80 other e-) ↔ large importance of Bi in a¿
 Cross section: Example (2)

Comparison theory ↔ experiment

47
 Angular distribution of the photoelectrons (1)

• In non-relativistic case → the differential cross section da¿/d is

/ f(µ) →

with ¯ = v/c, the relative velocity of the photoelectron and µ, the

emission angle of the photoelectron relatively to the initial
direction of the photon
• Cross section = 0 in the direction of the incident photon (µ = 0) →
the electron aims to be emitted in the direction of the electric
field of the electromagnetic wave
• For E photons ↗ → more and more electrons are ejected in the
forward direction
48
 Angular distribution of the photoelectrons (2)

• In the relativistic case →

• In both cases → For photons energy ↗ → more and more

electrons are ejected in the forward direction

49
Angular distribution of the photoelectrons (3)

50
 Bipartition angle

• Bipartition angle µb: angle for which 1/2 of the photoelectrons

are emitted in the forward direction inside a cone with half-
angle smaller than µb

• For instance → for hº0 = 0.5 MeV → 1/2 of the photoelectrons

are emitted inside a cone with half-angle ' 30°
51
 Consecutive phenomena
After a photoelectric effect → hole in an inner-shell →
electronic rearrangement → emission of a X-ray
(fluorescence) or an Auger electron → definition of the
fluorescence yield ωi (photon emission probability after a
transition to the shell i) Auger electron
→ E=BK-BL-BL

X ray→ E=BK-BL

52
K-shell fluorescence yield

53
 Pair production

• The photon is completely absorbed and its place appears a

electron-positron pair. This process only takes place in the field
of a nucleus or of an electron (more rarely)

• For the pair production in the electric field of an atomic

electron → triplet production (a part of the energy is
transferred to the initial electron)
• What is the minimal energy of the incident photon to have
pair production?
54
 Conservation laws (1)
• We work initially in the frame of the « target » particle with
mass M → this particle is at rest →

• After interaction → we work in the center of mass frame →

55
 Conservation laws (2)

• After the interaction → center of mass frame →

• We note Ttot = Te + Tp + TC
• By conservation of the invariant P2 = (E/c)2+p2 →

56
 Conservation laws (3)

• The minimum energy hº0,min is obtained by equalizing both

expressions and by considering the kinetic energy Ttot = 0 →

• In the nucleus field → M À m → hº0,min = 2mc2

• In the electron field → M = m → hº0,min = 4mc2

• Remark → It is possible to have pair production in the electron

field for photon energy between 2mc2 and 4mc2 because the
atom can take a part of initial momentum → however the
probability of this process is extremely weak
57
 Pair production cross section in a nucleus field (1)

• Pair creation occurs inside the electronic cloud → the screening

effect due to atomic electrons is important
• Cross sections calculations made by Bethe and Heitler (1934)
• In the nucleus field → attractive force for the electron and
repulsive force the positron → their energy distributions are
different
• However → weak effect (< 0.0075 £ Z MeV) → can be
neglected → the differential cross section for the creation of an
electron with kinetic energy T− is equal to the one of creation
of a positron with kinetic energy T+ = hº0−2mc −T− and is
symmetric with respect to the mean energy →

58
 Pair production cross section in a nucleus field (2)

• The energy-differential cross section is →

with ¾p = ®re2 = 5.80 £ 10-32 m2

• It can be written →

with x = T+ /(hº0-2mec2)

59
Results for P(x,hº0,Z)

60
 Comments on the function P(x,hº0,Z)

• The function is symmetric

• The function P does not depend a lot on the atomic number Z

→ the cross section is thus proportional to Z2

• The function P varies slowly with the energy hº0 of the photon

• The shapes of various curves are similar

• For 0.2 < x < 0.8 → P is approximately constant

61
 Total cross section

• The total cross section is given by integration on T+ →

with h P i, the mean value of P

• h P i does not depend a lot on Z and is slowly increasing with
hº0 → becomes constant for large energies (> 100 MeV) due to
the screening of the nuclear field by the atomic electrons
62
 Fuction h P i

• Line: nucleus field

• Dash: electron field 63
 Cross section for triplet production

• Very complex calculations

• It is possible to show →

• And thus →

with C, parameter dependent only on hº0 such as C → 1 for hº0

→ 1 and ↗ slowly for hº0 ↘ (C ≈ 2 for hº0 = 5 MeV)
• The triplet production contributes little to the total cross
section except for media with small Z (1% for Pb and 5-10% for
Z » 10) 64
 Function h P itriplet

• Line: nucleus field

• Dash: electron field 65
 e- - e+ direction of emission

• For hº0 quite larger than the energy threshold, electrons and
positrons are emitted in forward direction
• Mean emission angle (relatively to the direction of the photon
is roughly (radians) →

• Example: For hº0 = 5 MeV → h T i = 1.989 MeV and h µ i = 0.26

radians ' 15°
66
 Consecutive phenomenon to the pair production

• First, positron is slowing down in the medium (large cross

section for Coulombian interactions)

• Second, annihilation of the positron when it is (quasi) at rest

with an electron at rest in the medium

• After the annihilation → two photons of 511 keV energy are

emitted with an angle of 180° between them (conservation
laws)

67
 Photodisintegration of nuclei

• The photon is absorbed by an atom and a particle is emitted.

• That particle can be a photon or a light particle: p, n, α,…

• This interaction is possible when the photon energy is larger

to the threshold energy of the process (between 8 and 20
MeV)

68
 Comparison of various effects

Photoelectrique: complex

Z effect + C factor

Thomson
Z2 effect

Z effect

69
Comparison of the three dominant effects

70
 Attenuation coefficients (1)

• The 3 main photon interaction processes in matter have been

characterized by their atomic scattering cross section :
– Photoelectric effect: a¿
– Compton effect: a¾ = Z¾
– pair creation: aκ
• As other processes play a negligible role in our energy range →
the total atomic cross section a ¹ is →

71
 Attenuation coefficients (2)

• As previously seen → in a thin target (with atomic density N) of width

dx, the scattering probability for 1 photon is a ¹Ndx
• For a monoenergetic beam of I photons (k) per time unit → the
collision rate is Ia ¹Ndx
• The variation dI of the intensity after crossing the target is (by
assuming that each collision implies a loss in the beam → all
scattering are absorbing) → dI = -Ia ¹Ndx
• For a thick target (width l) and an initial beam ? to the target with I0
particles → the intensity after the target is →

• ¹ = a¹N: Linear attenuation coefficient (unit: m-1) → allows to the

evaluate the scattering rate

72
 Remark on experimental conditions

To check this exponential equation → a particular geometry is

needed → narrow beam geometry that prevents deflected
primaries and secondaries to reach the detector

« large distance »
« large distance »

uniform

Narrow beam geometry

93
 Narrow beam geometry: characteristics

• Large distance between the source and the attenuator →

particles perpendicular to the attenuator
• Large distance between attenuator and detector → each
particle deflected in the attenuator will miss the detector
(intensity of the primary beam in the detector independent of
the distance of the attenuator ↔ intensity of the deflected
primaries and of the secondaries ↘ with the square of this
distance) → the relative intensity of the primary beam ↗ with
this distance
• The beam is collimated → it uniformly covers the detector → ↘
of the number of deflected primaries and of secondaries
generated inside the attenuator
94
 Narrow beam geometry: shielding

• The shield around the attenuator stops all incident radiations

except those passing through the aperture

• The shield around the detector stops all radiations except

passing through the aperture (µ ≈ 0°) → Pb for X-rays or °
(advantage: small thickness)

755
Example of attenuation experiment

Transmission of °
from 60Co (1.17 and
1.33 MeV) though a
Cu target

76
 Alternative coefficients

• We van write (with M, the molar masse of the medium, ½ its

density and NA the Avogadro number):

• (½l): Area density (unit: kg m-2)

• ¹/½: Mass attenuation coefficient (unit: m2 kg-1) →

• ¸ = 1/¹: Mean free path (unit: m) → mean distance travelled

by a photon between two collisions
• ½/¹: Mass attenuation length (unit: kg m-2)
77
 Mass attenuation coefficient (1)

• ¹/½: Mass attenuation coefficient (unit: m2kg-1) → ratio of

dI/I by ½dl with dI/I, the fraction of indirectly ionizing
radiations which undergo interactions along the distance dl
travelled inside a medium of density ½
• Global coefficient global → takes into account the interactions
of particles in matter regardless of the nature of the
interaction
• The mass coefficients are directly proportional to the cross
section and do not depend on the physical nature of the
target → these coefficients are displayed in databases

78
 Mass attenuation coefficient (2)

• In a medium with various atom species → the interaction

probability is the sum of interaction probabilities with each
atom specie (since the molecular binding energies are weak
compared to energies of ° rays)

• The total mass attenuation coefficient is given by →

with wi, the mass fractions of the various atom species

79
Attenuation coefficients: example of lead

80
 Attenuation coefficients: practical examples

• 1 MeV photons in air: μ/ρ=0.064 cm2/g with ½(air)=0.001205

g/cm3 → ¹=7.71 10-5 cm-1 → after 1m → I/I0= 99.2%
• 10 keV photons in air: μ/ρ=5.1 cm2/g with ½(air)=0.001205
g/cm3 → ¹=6.15 10-3 cm-1 → after 1m → I/I0= 54.1%
• 1 MeV photons in lead: μ/ρ=0.070 cm2/g with ½(lead)=11.35
g/cm3 → ¹=7.95 10-1 cm-1 → after 1m → I/I0≈0%
→ after 1cm → I/I0≈45.2%
• 10 keV photons in lead: μ/ρ=130.6 cm2/g with ½(lead)=11.35
g/cm3 → ¹=1.48 103 cm-1 → after 1cm → I/I0≈0%

http://www.nist.gov/pml/data/xraycoef/index.cfm

81
 Mass energy-transfer coefficient (1)

• The mass attenuation coefficient ¹/½ is a measurement of the

mean number of interactions between a photon and matter →
it allows to evaluate frequency of collisions
• For frequent applications → important parameter is the energy
transferred « locally » in the medium i.e. the energy
transferred to electrons → effects of photons in matter are due
(almost) exclusively to electrons → see « Radiation protection »
• Definition of another quantity more adapted to this aspect →
mass energy-transfer coefficient ¹tr/½

82
 Mass energy-transfer coefficient (2)
• ¹tr/½: Mass energy-transfer coefficient (unit: m2kg-1) → quotient of
dEtr/(EN) (with E the energy of all particles excluding rest energy) by
½dl where dEtr/(EN) is the fraction of energy of the incident particles
transformed in kinetic energy of charged particles by interactions in a
depth dl of the medium of density ρ → also: ¹tr= (Etr/E)¹
• Also defined as →

with fi, the fractions of photon energy transferred to kinetic energy of

charged particles for all processes

83
 Fractions of energy transferred

• Photoelectric effect →

with EX, the mean energy of fluorescence photons

• Compton effect →

withh E1 i, the mean energy of scattered photon → remark:

formally X-rays have to be considered → practically they can be
neglected
• Pair production (in the field of nucleus and of electron) →

84
 Mass energy-absorption coefficient

• A part of kinetic energy of charged particles set in motion can

be absorbed no locally → a part of the energy can be lost in
radiative processes (especially Bremsstrahlung but also in-flight
annihilation or fluorescence radiations)
• ¹en/½: Mass energy-absorption coefficient (unit: m2kg-1) →
product of the mass energy-transfer coefficient by (1-g), with g
the fraction of energy lost on average in radiative processes as
the charged particles slow to rest in the material
• g is specific to the material

85
Comparison ¹tr ↔ ¹en (1)

Significant difference only for high

energies of the ° rays → when the
charged particles produced by the
interaction have enough energy to be
characterized by an important
Bremsstrahlung (especially for high Z
materials)

86
Comparison ¹tr ↔ ¹en (2)

87
 Mass absorption coefficient

• ¹a/½: Mass absorption coefficient (unit: m2kg-1) → coefficient

for which only we suppose that only scattered photons
(coherently or incoherently) take energy away

• Coefficient rarely used (never!)

• Finally →

88
Schematic overview of the coefficients

89
Example of the application of these coefficients

Energy deposited inside water

by ° rays from a 60Co point-
source put at the center of a
water sphere as a function of
the distance between the
source and the detector

90