An Introduction to

Accelerator Physics
C. Zhang

12th International Conference on Mechanical Engineering Design of

Synchrotron Radiation Equipment and Instrumentation (MEDSI2023)
November 7- 10, 2023 • Beijing
2

Particle Accelerators
 Accelerator physics and Engineering
Cryogenics
RF System
Magnets

Microwave Accelerator Power Supply

Physics

Control &
Instruments
Vacuum
Radiation
Protection
Mecha. Enginee.
 Preface
Accelerator physics is to study particle beams, their
motion in environments of accelerators, involving
external electromagnetic fields and their interactions.
Accelerator physics evolves classical mechanics,
electromagnetism, statistical physics, and quantum
physics.
Basic knowledge of the beam physics is briefly
introduced in this talk for the engineers who are
working in the particle accelerator related fields，
especially synchrotron radiation facilities.
4
Basic Concepts An Introduction to
Accelerator Physics
Transverse Motion
Longitudinal Motion
Collective Effects
Synchrotron Radiation
Free Electron Lasers (FEL)
5
1. Basic Concepts
Constants & Relations
Motion in E-M Fields
Linear accelerators
Synchrotrons
6
 1.1 Constants & Relations
Speed of light c = 2.99792458×1010 cm/sec
Planck constant H = 6.626075×10-34 J⋅s
Electron charge e = 1.6021773×10-19 Coulumbs
Electron volts 1eV = 1.6021773×10-19 Joule
Energy and rest mass 1eV/c2=1.78×10-36 kg
 Electron m0, e= 0.51099906 MeV/c2 = 9.1093897×10-31 kg
 Proton m0, p= 938.2723 MeV/c2 = 1.6726231×10-27 kg
7
 Some Basic Relations
Relativistic energy E=mc2=m0γc2

Relativistic momentum p=mv=m0γβc β=v/c γ = (1-β 2)-1/2

Energy - momentum E2/c2=p2+m02c2

relationship Ultra-relativistic case: β≈ 1, E≈ pc
Kinetic energy T = E-m0c2 = m0c2 (γ −1)
 
Equation of motion under dp = f ⇒ m d (γ v ) = q(E + v × B )
Lorentz force dt
0
dt

8
 1.2 Motion in E-M Fields
Particle motion is governed by Lorentz force:

dp
dt
[  
= q E+v×B ]

2 2 dE 2 dp
2 2 4
E = p c + m0 c ⇒ E =c p ⋅
dt dt
2 2
dE qc  qc  
( )
  
⇒ = p⋅ E + v × B = p⋅ E
dt E E
A magnetic field does not change a particle′s energy.
Only an electric field can do this.
9
 1.3 Linear Accelerators
Simplest example is a vacuum chamber with one
or more DC accelerating structures with the
electric field aligned in the direction of motion.
 Beam energy is limited to a few MeV, due to maximum
electric field in material.
HEPS linac, IHEP To achieve higher energies, the electric fields are
alternating at RF cavities in linacs.
 Avoids expensive bending magnets;
 No energy loss due to synchrotron radiation;
 But requires many structures for limited accelerating
gradient;
SNS Linac, Oak Ridge  A long accelerator is needed for a high energy linac.
10
Accelerating structures: Travelling wave & Standing wave
Structure 1:
Travelling wave structure: particles keep in phase with
the accelerating waveform.
Phase velocity in the waveguide is greater than c and
needs to be reduced to the particle velocity with a series of
irises inside the tube whose polarity changes with time.
The structure should bretty the same as electrons at
3 MeV are already at 99% of speed of light.
Travelling wave structure is often applied in electron
linacs .

Structure 2:
Standing wave structure: A series of drift tubes
alternately connected to high frequency oscillator.
Particles accelerated in gaps, drift inside tubes .
For constant frequency generator, drift tubes
increase in length as velocity increases.
Beam has pulsed structure.
 1.4 Synchrotrons
The principle of synchrotrons is to vary B-field in
p(t)c
time to match increase in energy and keep revolution B(t) =
radius constant and synchronism between frf and fc. Zeρ
Magnetic field produced by bending magnets (dipoles), f rf ( t ) = h ⋅ f c
increases linearly with momentum.
p E
Bρ = ≈ so E [GeV] ≈ 0.3 B [T] ρ [m]
e ce
Large radius for high energy synchrotrons is requried
for the practical limitations for magnetic fields..
e.g. LHC: E = 7 TeV, B = 8.36 T, ρ = 2.7 km
12
 Types of Synchrotrons
Conventional synchrotrons: slow or rapid cycling machine used for booster injector
or fixed target experiments.
Storage rings: accumulate particles and keep circulating for long periods; used
for powerful machines such as synchrotron radiation facilities.
Colliders: two beams circulating in opposite directions for collision to reach high
center mass energy.

HEPS Booster SSRF Storage Ring BEPCII Storage Rings

13
 2. Transverse Motion
Motion Description Lattice
Bending Dispersion
Focusing Orbit Distortion
Hill’s Equation Coupling
Phase Space Non-linearity
14
 2.1 Motion Description

s − longitudinal direction Tangential to beam direction

x – horizontal direction
Transverse directions
y – vertical direction
15
 2.2 Bending
By increasing energy E (hence p) and B together in p
ρ=
a synchrotron, it is possible to maintain a constant qB
radius and accelerate a beam of particles.
In a synchrotron, the confining magnetic field
comes from a system of several magnetic
dipoles forming a closed arc.
Dipoles are mounted apart, separated by
HEPS dipole in test
straight sections with vacuum chambers
including equipment for focusing, acceleration,
injection, insertion devices, beam monitors,
and vacuum pumps, etc. HEPS 7BA cell

16
 2.3 Focusing
A sequence of focusing-defocusing fields provides
a strong focusing force.
Quadrupoles focus horizontally, defocus
vertically or vice versa. Forces are linearly B
y
proportional to displacement from axis.
x
A succession of opposed elements enable particles
to follow stable trajectories, making small
(betatron) oscillations about the design orbit.

HEPS quadrupole
17
 2.4 Hill’s Equation
Hill’s Equation:
x′′ + k x ( s ) x = 0, y′′ + k y ( s ) y = 0
Equations of motion in accelerator components：
 Drift section: x′′ = 0, y′′ = 0 𝟏𝟏 𝒍𝒍
M(s /s0 )= � �
𝟎𝟎 𝟏𝟏
 Dipole: 1
x′′ + 2
x = 0, y′′ = 0
ρ
 Quadrupole: x′′ + k x = 0, y′′ − k y = 0
 Sextupole : x ′′ + ks ( x 2 − y 2 ) = 0, y′′ − 2ksxy = 0 18
 Solution of the Hill’s Equation
Hill’s equation β (s) – envelope function
It is with linear-periodic coefficients is behavior of the machine
2
d u
2
+ k ( s )u = 0 ε − Emittance
ds
k( s) = −
1 dB y is behavior of the beam
where Bρ dx
Physical meaning (H and V planes)
and u denotes x and y
Like restoring constant in simple  Beam envelope
harmonic motion, the solution is uenv( s ) = εβ ( s )
u( s ) = εβ ( s ) sin(φ ( s ) + φ0 )  Maximum excursions
ds
where φ ( s ) = ∫ is betatron phase u′( s ) = ε / β ( s )
β ( s)
19
 2.5 Phase Space x´ x´
Under linear forces, any particle moves
on an ellipse in phase space (x, x´).
x x
Ellipse rotates in accelerator lattice, but
its area is preserved:
General equation of ellipse is
Emittance β x′ 2 + 2α x x′ + γ x 2 = ε
α, β, γ are functions of distance s
(Twiss parameters), and ε is a constant.
Area of ellipse = πε.
For non-linear beams one can use 95%
emittance ellipse or rms emittance
Acceptance: Ax,y > εx,y
2
 X2  Y 2  ε rms = x2 x′ 2 − xx′
Ax =  
 , Ay =  
 
 β x  max  β y  max (statistical definition)
20
 2.6 Lattice
Lattice is defined as the pattern of focusing magnets, bending magnets
and straight sections connecting in between;
It has a strong influence on aperture of vacuum chambers and thus other
systems of the accelerator.

21
 Matrix formalism
As a consequence of the linearity of Hill’s equations, we can describe the
evolution of the trajectories in a lattice by means of linear transformations

In terms of the amplitude and phase function the transfer matrix is

 β ( s) 
 (cos ∆φ + α 0 sin ∆φ ) β ( s )β 0 sin ∆φ 
 β 0 
M ( s2 / s1 ) =  
(α ( s ) − α 0 ) cos ∆φ + (1 + α ( s )α 0 ) sin ∆φ β0
−

[cos ∆φ − α ( s ) sin ∆φ ]
 β ( s )β 0 β ( s) 
where α = dβ /ds，γ = (1+α2)/β .
For a periodic machine the transfer matrix over a lattice period reduces to

22
 Matrix formalism
As a consequence of the linearity of Hill’s equations, we can describe the
evolution of the trajectories in a lattice by means of linear transformations

In terms of the amplitude and phase function the transfer matrix is

 β ( s) 
 (cos ∆φ + α 0 sin ∆φ ) β ( s )β 0 sin ∆φ 
β 
M(s/s0) ==  0

(α ( s ) − α 0 ) cos ∆φ + (1 + α ( s )α 0 ) sin ∆φ β0
−

[cos ∆φ − α ( s ) sin ∆φ ]
 β ( s )β 0 β ( s) 
where α = dβ /ds，γ = (1+α2)/β .
For a periodic machine the transfer matrix over a lattice period reduces to

23
 Example: FODO Lattice
The matrix for one period
between mid-planes of F magnet s
in thin lens approximation is
 1 0  1 L  1 0  1 L  1 0
M x, y =      
  1/ 2 f 1  0 1  ± 1 / f 1  0 1   1 / 2 f 1
 1 − L2 / 2 f 2 2 L(1 ± L / 2 f  L2
=  2 2 2

 µ x , y = cos −1 (1 − 2
)
 − L / 2 f (1  L / 2 f ) 1 − L / 2 f  2f
 cos µ + α sin µ β sin µ  1 ± sin( µ / 2) 1
=   β x, y = 2L =
 γ sin µ cos µ − α sin µ  sin µ γ x, y
α x, y = 0 Qx,y = (Nc⋅µx,y/2π)
24
 2.7 Dispersion
The dispersion has its origin that a particle of higher momentum is deflected
through a less angle in a bending magnet.
It was shown that the equations of motion of a
charged particle is a linear Hill’s equation
 1  1 ∆p
 
x' '+  2 − k ( s )  x =
ρ  ρ p0
The solutions of the equation can be written in
terms of the optics functions
∆p
x( s ) = εβ ( s ) cos(φ ( s ) − φ0 ) + D( s )
p0
β ( s) β (s )
D= ∫ cos[πν − φ ( s ) − φ ( s )]ds
2 sin πν ρ(s )
25
 Chromaticity
Chromaticity is defined as betatron tune changes with momentum spread:

The chromaticity has its origin that a particle of higher

momentum is focused less in a quadrupole magnet.
The Hill’s equation is written as:
where u denotes x or y, and ∆k = -k⋅∆p/p0
The chromaticity of the synchrotron is derived as:
Non-zero chromaticity will make tune spread for momentum spread, and
negative chromaticity will cause head-tail instability when γ>γt.
Sextupoles are used for correct chromaticity, while the nonlinear effects need to
be mitigated by detail simulation to obtain desired dynamic aperture. 26
 2.8 Orbit Distortion
Dipole errors may cause orbit distortion of particle beams.

Element Source Field Error Deflection Direction

Field error 〈∆B/B 〉 θ⋅〈∆B/B 〉 x
Dipole (angle=θ )
Tilt 〈∆θz 〉 θ⋅〈∆θz 〉 y
Quadrupole (Kl) Displacement 〈∆x, ∆y 〉 Kl⋅〈∆x, ∆y 〉 x, y

Similar to dispersion case, the equation of motion is written as

1 ∆B ( s )
y' '+ ky =
Bρ B
The solutions are y( s ) = εβ ( s ) cos(φ ( s ) − φ0 ) + ycod ( s )
β ( s) ∆B ( s ) β ( s )
∫ Bρ ( s ) cos[πν − φ ( s ) − φ ( s )]ds
Measured by BPM’s
yCOD =
2 sin πν Corrected by dipoles
27
 2.9 Linear Coupling
Coupling is the phenomena that energy exchange between two oscillators.
In accelerators, the horizontal and vertical
motions are also coupled:
x ′′ + k x x = −( k + M ′ / 2) y − My′
y′′ + k y y = − ( k − M ′ / 2) x − Mx ′
Where
1  ∂B x ∂B y  Bs
k=−  −  , M =−

2 Bρ  ∂x ∂y  0 Bρ

x-y coupling can be compensated with skew

quadrupoles and anti-solenoids. εy ( ∆ν ) 2
κ= =
ε x (ν x − ν y )2 + 2( ∆ν )2
It can be measured with tune approaching.
28
 2.10 Non-linearity
There are higher order terms in the expansion of magnetic field of magnets:
1 ∂ nBy
𝑴𝑴 kn = n Normal multipoles
𝒌𝒌𝒏𝒏 (𝒔𝒔) + 𝒊𝒊𝒋𝒋𝒏𝒏 (𝒔𝒔) B0 ρ 0 ∂x
𝑩𝑩𝒚𝒚 + 𝒊𝒊𝑩𝑩𝒙𝒙 = 𝑩𝑩𝟎𝟎 𝝆𝝆𝟎𝟎 �� (𝒙𝒙 + 𝒊𝒊𝒊𝒊)𝒏𝒏 � n
( 0,0 )
𝒏𝒏! 1 ∂ Bx
𝒏𝒏=𝟏𝟏 jn =
B0 ρ 0 ∂x n
Skew multipoles
( 0,0 )

The Hill’s equations acquire additional nonlinear terms:

d2x  1   M k n ( s ) + ijn ( s ) 
2
+  2 − k1 ( s )  x = Re  ∑ n
( x + iz ) 
ds  ρ ( s)   n= 2 n! 
d 2z  M k n ( s ) + ijn ( s ) 
2
+ k1 ( s ) z = − Im  ∑ n
( x + iz ) 
ds  n= 2 n! 
There is no analytical solution available in general, and the equations have to
be solved by tracking or perturbative analysis.
29
3. Longitudinal Motion
RF Cavities & Acceleration
Synchrotron Oscillation
Bunches and Buckets
γ-Transition
30
 3.1 RF Cavities & Acceleration
RF cavities are necessary condition for beam
acceleration.
Both linear and circular accelerators use
electromagnetic fields oscillating in resonant
∆E = qV0 sin ϕ s
cavities to apply the accelerating force.
In linacs – particles follow straight path
through series of cavities.
In circular accelerators – particles follow
circular path in B-field and particles return
to same accelerating cavity each time around.
31
3.2 Synchrotron Oscillation
This is a rigid pendulum
2πV0 hηf 0 2
φ = − 2
(sin φ − sin φ s )
E0 β γ
For small amplitudes
2
2πV0 h η f 0  + Ωs 2 ∆φ = 0
φ + ∆ φ = φ
E0 β 2γ
Synchrotron frequency
ηhV0 cos φ s
fs = 2
f0
2πE0 β γ
Synchrotron tune
fs ηhV0 cos φ s
νs = =
f 2πE0 β 2γ
32
 3.3 Bunches and Buckets
For large amplitudes
2
Ω
φ = − s
(sin φ − sin φ s )
cos φ s
Integrated becomes an invariant
φ 2 Ω 2s
− (cos φ + φ sin φ s ) = const .
2 cos φ s
The equation of each separatrix is
φ 2 Ω 2s Ω 2s
− (cos φ + φ sin φ s ) = − [cos(π − φs ) + (π − φs )sin φs ].
2 cos φ s cos φ s
And the half height is
12
 eV0 
(∆E E s )max = ±β  G (φ s ) G (φ s ) = [2 cos φ s − (π − 2φ s ) sin φ s ]
 πhηE s 
Bucket size should be larger than bunch size [∆E/Es and ∆φ (∆τ, ∆z)]  high V0
33
 4. Collective Effects
Space Charge
Wake Fields and Impedance
Beam Instabilities
Beam-beam Effects

34
 4.1 Space Charge
A charged particle in a beam undergoes electromagnetic field from other particles
η…charge density in C/m3 Electric Magnetic
λ... constant line charge π a2η → →
→ η B = Bφ E = Er
I…constant current λβc = π a2ηβc → →
div E =
a…beam radius curl B = μ0 J ε0 Current density (βcη)
→ → → → → → →
∫∫∫ div E dV = ∫∫ E d S ∫ B r dφ = ∫∫ curl B d A
Volume element Apply these integrals over

cylinder radius r length l cross section radius r

s η
r 2 πl = E r 2rππ Bφ 2rπ = μ0 r 2 πβcη
y
ε0
I
Bφ Er
 I r I r
dφ Er = Bφ =
rφ 2πε0 βc a 2 2πε0c 2 a 2
x

dA a cross
s section For β 1 (γ>>1) FE=qEr =-FB = -q⋅Bφ⋅v so ΣF=0
35
 4.2 Wake Fields and Impedance
Wake field will be driven by a beam when there is a discontinuity in the
vacuum chamber:


F⊥ ( r , θ , z ) = − eI mWm ( z )mr m −1 ( rˆ cos m θ − θˆ sin m θ )
F|| ( r , θ , z ) = − eI mWm′ ( z )r m cos m θ
Impedances are just Fourier transforms of wake functions:
⊥ i dz − iωz / v || ∞ dz − iωz / v
Zm (ω ) = ∫
v/c v
e Wm ( z ) , Z m (ω ) = ∫
− ∞ v
e Wm′ ( z )

Impedance Z = Zr + iZi V acts back on the beam

Induced voltage V ~ Iw Z = –IB Z  Instabilities
36
 4.3 Instabilities
Robinson instability
 Caused by fundamental cavity mode
 Cure: Beam will be stabilized by properly
tuning the cavity (frf < hf0 for γ>γt).
Head-tail Instability
 Caused by transverse wake field
 Cure: Beam will be stabilized by properly setting chromaticity (ξ> 0 for γ>γt).
Strong head-tail Instability
 Caused by transverse wake as counterpart of beam break-up in a linac.
 Cure: control of broadband transverse impedance.
Potential-well distortion and microwave instability
 Cure: control of broadband longitudinal impedance.
Ion trapping and fast ion instability – better vacuum and proper bunch pattern
Coupled bunch instability
 Multi-bunch effects caused by long-range transverse & longitudinal wakes
 Cure: control of narrowband impedance and applying feedback systems.
37
5. Synchrotron Radiation

Principle of synchrotron radiation

Radiation damping
Radiation fluctuation
Synchrotron radiation sources
38
5.1 Principle of Synchrotron radiation
The electromagnetic radiation emitted when the charged particles
are accelerated (deflected).
The particle loses the energy of e 2γ 4
U0 =
U0 in one turn: 3ε 0 ρ

E (GeV )4
For electrons: U 0 ( keV ) = 88.46
ρ (m )
e.g. LEP2: E=100GeV，ρ = 3100 m, U0=1.85 GeV
For protons: E (GeV )4
U 0 ( keV ) = 7.78 × 10 −12
ρ (m )
e.g. LHC: E=7000GeV，ρ= 2804 m, U0= 6.66 keV
 5.2 Radiation damping
In storage rings, synchrotron radiation loss is
compensated by the RF fields:
U 0 = eV0 sin(ϕ s )
d 2ε 2 dε 2
This will cause radiation damping: 2
+ + ω sε = 0
dt τ s dt
Longitudinal: higher the energy of particle more the
SR loses;
Vertical: radiated momentum is with P⊥, while RF
compensates P|| ;
Horizontal: After the photon emission δxβ + δxε = 0

40
 5.3 Radiation fluctuation
If it is with pure damping, the emittance would be getting to zero in x, y and s
directions.
Diffusion effect off
However, the radiated energy is emitted in quanta: each
quantum carries an energy u = ħω;
The emission process is instantaneous and the time of
emission of individual quanta is statistically independent;
Radiation damping combined with radiation fluctuation
determine the equilibrium beam distribution and therefore Diffusion effect on
finite emittance, beam size, energy spread and bunch length.

55  γ 2 H / ρ
3
 55  
1/ 2
αc
εx = σE =   ⋅γ σz = σE
32 3 mc J x 1 / ρ 2
 64 3 ρmc  Ωs

Jx is damping partition numbers, H is dispersion invariant: H ( s ) = γD x 2 + 2αD x D x '+ β D x '2

41
https://lightsources.org/

5.4 Synchrotron radiation facilities

There are more than 50 synchrotron radiation facilities in the world
(operational or under construction)
 Parameters Design

The High Energy

Circumference（m） 1360.4
Lattice type 48×7BA

Photon Source (HEPS)

Beam energy（GeV） 6
Beam current（mA） 200
Emittance（pm·rad） 60
Brightness（phs/s/mm2/mrad2/0.1%BW） >1×1022

Under construction in Beijing

Huairou Science Park
(2019-2025)
43
5. Free Electron Laser (FEL)

From traditional laser to FEL

From SR to FEL

FEL’s in the world

44
 5.1 Traditional laser and FEL
FEL
beam

electron
beam used
beam

“Bound” Free Electron

Electron Lasers Lasers

Traditional laser gains energy from energy level transition in the

“bound” electrons in atoms.
FEL converts the energy of free electrons of accelerators into
laser power.
FEL’s may provide high power lasers from THz to X-rays.
45
 6.2 From SR to FEL
Resonant condition

Gain length

Synchrotron radiation emitted from the stored electrons in a storage

ring is continuous spectrum, high average power, high repetition rate,
high brightness (1023) and with multi-beam ports.
FEL generates from undulators radiation is advantage in its
monochromaticity, coherence, short pulses (~fs), high pulse power and
even higher brightness (1033).
SR and FEL are complementary in scientific applications.
46
5.3 XFELs in the World

SXFEL 2020

Eero-XFEL Swiss-FEL
2017 2018

47
Shanghai HIgh repetition rate XFEL and Extreme light facility（SHINE）

Electron energy: 8 GeV

Photon energy : 0.4-25 keV
Pulse dureton : 1-100 fs
Repetition rate: 1 MHz
Total length :3.1 km
ca 30m underground
SXFEL

48
 Summary
As tools of discovery, particle accelerators have been
rapidly developed since 1930’s.
The purpose of accelerator physics is to study behaviors
of particle beams.
The accelerator physics and technology are closely related
in design, construction and performance of accelerators.
Vigorously growing of synchrotron radiation facilities
and free electron lasers has also promoted the
development of accelerator physics and technology.
 Bibliography
1. E. Wilson, An Introduction to Particle Accelerators
2. D. Edwards & M. Syphers: An Introduction to the Physics of
High Energy Accelerators
3. S.Y. Lee: Accelerator Physics
4. A.W. Chao and M.Tigner, Handbook of Accelerator Physics and
Engineering
5. M. Conte & W. MacKay: An Introduction to the Physics of
Particle Accelerators
6. M. Livingston & J. Blewett: Particle Accelerators
7. M. Reiser: Theory and Design of Charged Particle Beams
50
Thank You for
Attention