Fundamentals of Accelerator Physics and Technology (Presentation Slides) (2006) WW

Fundamental Accelerator Theory, Simulations and Measurement Lab Arizona State University, Phoenix January 16-27, 2006
Simulation and Measurement Lab

Overview of simulation software,
Overview of simulation software,
measurement equipment
measurement equipment
and assignments
and assignments
Soren Soren Prestemon Prestemon
Program layout
Program layout
Simulation and
Measurements
Soren Prestemon
Three simulation and three measurement assignments
Three simulation and three measurement assignments
Final project: Design an accelerator
Final project: Design an accelerator
(for high energy physics or for synchrotron radiation)
(for high energy physics or for synchrotron radiation)
2 Days!
2 Days!
Simulations:
1. Dipole/quadrupole fields
Code Magnet/Poisson
2. Cavity modeling
Code Superfish
3. Synchrotron radiation
Code SynRad
Measurements:
1. Dipole or quadrupole fields
Hall probe
2. Microwave cavity
Network analyzer
3. Insertion device
Hall probe
+
+
+
2 Days
2 Days
2 Days
Note: the simulation and measurement lab is 30% of your grade
Workgroups
Workgroups
Simulation and
Measurements
Soren Prestemon
Eight teams
Eight teams
labwork
labwork
is a group effort!
is a group effort!
Labwork is designed as 2-day exercise (lab+simulation)
Each exercise will require a written report
9Lab reports for the first week (2 reports) are due Monday, Feb. 23;
9Lab reports for the second week are due before the final exam
The last 2-day exercise will consist of the design of a storage ring
1 1a-M
2 1a-C
3 1b-M
4 1b-C
5 2-M
6 2-C
7 3-M
8 3-C
Key
Day\Group # 1 2 3 4 5 6 7 8
1/16/2005
1/17/2005 1 2 3 4 5 6 7 8
1/18/2006 2 1 4 3 6 5 8 7
1/19/2006 5 6 7 8 1 2 3 4
1/20/2006 6 5 8 7 2 1 4 3
1/23/2006* 7 8 5 6 7 8 5 6
1/24/2006 8 7 6 5 8 7 6 5
1/25/2006
1/26/2006
1/27/2006*
* previous lab assignements due
Design of a Storage Ring
Lab and Calculations Overview
1a
1b
2
3
4
Dipole Field measurements
Quadrupole field measurements
Microwave cavity measurements
Dipole/Quadrupole calculations
Dipole/Quadrupole calculations
Cavity simulation
Design of a storage ring
Insertion device measurments Insertion device radiation properties
Computer lab Measurement lab
Lab groups
Lab groups
Course
Introduction
Jeffrey Ball, ORNL, ae
Molly Scannell, BNL, bs
Khalimov Mirkomal, Arifov Institue of Electronics, gs
Anthony Bavuso, Jefferson Lab, bsee
Hannes Bartosik, Tech. University of Vienna, gs
Jonah Weber, LBNL, bsee
Danielle Sanzone, SLAC, bap
Ekaterina Danilova, ORNL, ms
Nicola Pozzobon, University of Pisa , gs
Heritier Makamizile Mbo, Congo-Fr. Electric Company, gs
William Chase, Brookhaven National Lab, bsp
Lukas Jagerhofer, Tech. University of Vienna, gs
Lynn Garren, Fermilab, phd
Paul Cummings, Embry-Riddle Aeronautical U., ug
Muhammad Jamil, Konkuk University, gs
Everette R. Martin, Fermilab, bsee
Laurie Elizabeth (Elisa) Dowell, Naval Research Lab, msrs
Kristine Ferrone, Brookhaven National Laboratory, bsa
Andres Gomez Alonso, U. Politecnica de Catalunya & CERN, gs
Sammie Garvin, George Mason University, bs
Robert Hensley, Embry-Riddle Aeronautical University, ug
Artur Paytyan, Fermilab, ms
Mohammad Adil Khan, Kyungpook National University, gs
Valentina Previtali, Genova University & CERN, gs
Eric Tse, SLAC, bap
3
2
8
4
7
1
6
5
Introduction to lab exercises:
Dipole/
Dipole/
quadrupole
quadrupole
exercise
exercise
Simulation and
Measurements
Soren Prestemon
Learn about design and purpose of magnetic elements
Learn about design and purpose of magnetic elements
The magnetic elements form the backbone of a storage ring:
The magnetic elements form the backbone of a storage ring:
Dipoles steer the beam, bending it on a circular path;

Dipoles steer the beam, bending it on a circular path;
Quadrupoles
Quadrupoles
focus the beam, essential to keep the electrons
focus the beam, essential to keep the electrons
from diverging
from diverging
Simulations:
Dipole and quadrupole magnetic
system design
Goal: understand excitation curves,
saturation effects, purpose of
poles and yokes, field harmonics,
beam steering and focusing using
the code Magnet
Measurements:
Dipole or quadrupole field measurements
Goal: Learn to assemble a model magnet
and a Hall probe; measure spatial field
profiles, excitation curves; understand
current and current density, integrated
fields, and beam steering or focusing
as a function of current
Accelerating cavity exercise
Simulation and
Measurements
Soren Prestemon
Learn about RF cavity design and measurements
Learn about RF cavity design and measurements
Accelerating cavities serve:
Accelerating cavities serve:
in booster rings, to bring electrons up to nominal

in booster rings, to bring electrons up to nominal
operating energy,
operating energy,
in the main ring to compensate for radiation energy

in the main ring to compensate for radiation energy
loss (e.g. synchrotron radiation)
loss (e.g. synchrotron radiation)
Simulations:
RF cavity simulation
Goal: understand electric and
magnetic fields that can exist in a
cavity; find resonant modes and
frequencies, distinguish modes
from harmonics
Measurements:
Pillbox cavity measurements
Goal: Estimate base resonant frequency
from geometry of a cavity; understand
fields excited by two antennas
installed in the device; measure
resonant curve and impact of
geometry on cavity tune
Simulation and
Measurements
Soren Prestemon
Learn about insertion devices for synchrotron radiation
Learn about insertion devices for synchrotron radiation
Insertion devices characterize third generation SR sources:
Insertion devices characterize third generation SR sources:
Provide intense SR tailored to specific scientific experiments

Provide intense SR tailored to specific scientific experiments
They also serve in damping rings to reduce
They also serve in damping rings to reduce
emittance
emittance
Simulations:
Synchrotron radiation generation
Goal: understand the radiation
characteristics associated with
bending magnets, wigglers, and
undulators, including spectral
range and photon beam
polarization
Measurements:
Field measurements and characterization
of a wiggler magnet
Goal: Using a Hall probe, understand
field characteristics and resulting
radiation properties of a wiggler;
evaluate impact of a wiggler on the
electron beam (e.g. steering and
focusing)
Storage ring design
Storage ring design
Simulation and
Measurements
Soren Prestemon
Use knowledge learned during the course to design a ring
Use knowledge learned during the course to design a ring
There are a large number of design choices in the design of a
There are a large number of design choices in the design of a
storage ring; the scientific purpose serves to dictate the
storage ring; the scientific purpose serves to dictate the
optimal parameter choices
optimal parameter choices
Design process:
Choose to design either a SR storage ring or a HEP machine
Choose basic design point: electron or proton energy, ring diameter
Choose realistic ring elements (dipole field strengths, quadrupole fields, etc)
Incorporate elements discussed during the course (particle sources,
accelerating structures, chromatic aberration correction, vacuum
components, etc)
Simulate performance using program BeamOptics: plot betatron functions,
perform particle tracking, modify design to yield reasonable lifetime
Goals
Goals
The goal of the lab is to provide a hands-on
appreciation for the design, fabrication, and
measurements associated with some key
accelerator components
The lab questions should be addressed as well as
possible, but above all we want you to understand the
issues involved
You are free to do further measurements, play with
design modifications, etc
An additional computer exercise (extra credit!) on
quadrupole focusing can be addressed in your spare
time you would find it useful for the final project
and ask questions!
1
Lecture No. 10
Lecture No. 10
Injection & Extraction
Fernando Fernando Sannibale Sannibale
2
Introduction
Introduction
F. Sannibale
Every accelerator complex Every accelerator complex
that includes a storage ring that includes a storage ring
requires an injection system. requires an injection system.
With increasing complexity, With increasing complexity,
when more than one ring is when more than one ring is
present, extraction systems present, extraction systems
make their appearance and make their appearance and
more injection systems are more injection systems are
added. added.
CERN CERN
Systems used for injection Systems used for injection
can be used as well for extraction can be used as well for extraction
by simple mirror reflection. by simple mirror reflection.
In fixed target experiments, In fixed target experiments,
where the beam is extracted and where the beam is extracted and
sent to a target, the extraction sent to a target, the extraction
systems can assume different systems can assume different
characteristics. characteristics.
3
The Typical Injection Section
The Typical Injection Section
F. Sannibale
matching elements matching elements
Septa magnets Septa magnets
Fast kicker Fast kicker
Storage Ring Storage Ring
Transferline Transferline
Matching Matching: for an efficient injection, the optical functions in the trans : for an efficient injection, the optical functions in the transfer fer- -line line
and in the storage ring must be the same at the injection point. and in the storage ring must be the same at the injection point.
Septum Magnet Septum Magnet: Special magnet with a thin wall that allows to place the : Special magnet with a thin wall that allows to place the
magnet close to the storage ring orbit. Can operate in DC or in magnet close to the storage ring orbit. Can operate in DC or in pulsed mode. pulsed mode.
Fast Kicker Fast Kicker: It is the pulsed element that gives the final kick that puts t : It is the pulsed element that gives the final kick that puts the he
injected beam on the storage ring orbit. Its pulse must last for injected beam on the storage ring orbit. Its pulse must last for less than a less than a
ring revolution period for avoiding kick the beam again. ring revolution period for avoiding kick the beam again.
In some injection schemes, a slow orbit bump localized in the s In some injection schemes, a slow orbit bump localized in the septum eptum
region, brings the beam closer to the septum wall. region, brings the beam closer to the septum wall.
4
Injection Matching
Injection Matching
F. Sannibale
Matching is fundamental for protons and ions Matching is fundamental for protons and ions, because of the , because of the
absence of damping. absence of damping.
Any injection error is immediately exported to the storage ring Any injection error is immediately exported to the storage ring
generating generating emittance emittance growth. growth.
Matching is less critical for electrons and positrons because da Matching is less critical for electrons and positrons because damping mping
washes washes- -out any injection error within the acceptance of the storage rin out any injection error within the acceptance of the storage ring g
By matching at the injection point the optical and the dispersio By matching at the injection point the optical and the dispersion n
functions between the functions between the transferline transferline and the storage ring, one ensures a and the storage ring, one ensures a
smooth transition for the beam from the injector to the storag smooth transition for the beam from the injector to the storage ring. e ring.
Anyway, for some specific application a good matching is importa Anyway, for some specific application a good matching is important also nt also
for electrons and positrons. For example, for electrons and positrons. For example, top top- -off operation off operation (quasi (quasi- -
continuous injection) in synchrotron light sources and lepton continuous injection) in synchrotron light sources and lepton colliders colliders
requires a good matching in order to minimize the perturbation ( requires a good matching in order to minimize the perturbation (noise) noise)
that the injection transient can generate during users data tak that the injection transient can generate during users data taking. ing.
5
Septum Magnet
Septum Magnet
F. Sannibale
DA NE Frascati
Special care in the design must be Special care in the design must be
used for avoiding field leakage that used for avoiding field leakage that
will affect the beam orbit will affect the beam orbit
Septa can be in or out of Septa can be in or out of
vacuum, DC or pulsed. vacuum, DC or pulsed.
6
DANE
Krasnykh SLAC
Fast Kickers
Fast Kickers
F. Sannibale
Normally, the kicker pulse duration must be shorter than a revol Normally, the kicker pulse duration must be shorter than a revolution ution
period. Typical kickers pulses have ~ 100 ns duration. New machi period. Typical kickers pulses have ~ 100 ns duration. New machines nes
such as the ILC and other special applications are asking for such as the ILC and other special applications are asking for
challenging kickers with few ns pulse duration. challenging kickers with few ns pulse duration.
LNLS-Campinas
SPEAR 3
or discrete elements Pulse
Forming Network (PFN)
Distributed
PFN
7
Single Turn Injection
Single Turn Injection
F. Sannibale
The transfer matrix from the septum to the kicker position is gi The transfer matrix from the septum to the kicker position is given by: ven by:
Kicker Kicker
S
e
p
t
u
m
S
e
p
t
u
m
= /2 = /2 = /2 = /2 = /2 = /2 = /2 = /2
Optimum phase advance Optimum phase advance
( )
( ) ( )
( )

+ +
+
=
S
S
K
K
S
S K
S K S K
S K S
S
K
K
K
x
x
a
a
x
x

sin cos
cos sin 1
sin sin cos
By imposing that By imposing that
at the kicker at the kicker x x
K K
= 0 = 0

tan
cot
=
+
=
S
S
S
S
x x
x x
S S
is defined by the required stay clear at the septum position. is defined by the required stay clear at the septum position.
Now using these values at the septum, one Now using these values at the septum, one
can calculate the angle at the kicker position: can calculate the angle at the kicker position:

tan
sin
1
= =
S
K S
K
x
x
In order to place the beam on the ring reference orbit, the an In order to place the beam on the ring reference orbit, the angular kick gular kick
must be equal to must be equal to . . Note that the minimum Note that the minimum is obtained when is obtained when = = /2 /2
x x
S S
8
Stacking Beam
Stacking Beam
F. Sannibale
If multiple injection in the same bucket is required If multiple injection in the same bucket is required ( (stacking stacking), the ), the
previous scheme can modified as follows: previous scheme can modified as follows:

betatron betatron phase advance phase advance
Kicker 1 Kicker 1 Kicker 2 Kicker 2
/2 /2 /2 /2 /2 /2 /2 /2 phase advance phase advance
The new kicker pre The new kicker pre- -kicks the kicks the stored stored beam so that when it will pass beam so that when it will pass
through Kicker 1 (simultaneously with the injected bunch) it wil through Kicker 1 (simultaneously with the injected bunch) it will be placed l be placed
back on the nominal closed orbit. back on the nominal closed orbit.
9
Off
Off
-
-
Energy
Energy
Injection Scheme
Injection Scheme
F. Sannibale
By using a lattice with nonzero horizontal dispersion in the se By using a lattice with nonzero horizontal dispersion in the septum area a ptum area a
special injection scheme can be used. special injection scheme can be used.
No fast kicker is required. Because of the damping requirement, No fast kicker is required. Because of the damping requirement, such a such a
scheme can be used only for electrons and positrons. scheme can be used only for electrons and positrons.
The injected beam will move on dispersion orbits for few dampin The injected beam will move on dispersion orbits for few damping times g times
until the radiation damping will bring it at the nominal energy until the radiation damping will bring it at the nominal energy merging with merging with
the stored beam. the stored beam.
S
e
p
t
u
m
S
e
p
t
u
m
S
x
On On- -energy beam trajectory energy beam trajectory
Inj Inj
x
p
p
=
0
0
A particular off A particular off- -energy beam trajectory energy beam trajectory
Inj
x
S Inj
x x
Several rings use or successfully tested the off Several rings use or successfully tested the off- -energy injection schemes. energy injection schemes.
For example APS, HERA, CESR, For example APS, HERA, CESR,
10
Electrons, Protons
Electrons, Protons
and Heavy Ions
and Heavy Ions
F. Sannibale
There is a fundamental difference between the two cases of ligh There is a fundamental difference between the two cases of light particles t particles
(electron and positrons) and of heavier ones (proton, ions, ...) (electron and positrons) and of heavier ones (proton, ions, ...). .
Stored Beam
Injected Beam
Electrons and positrons Electrons and positrons benefit from benefit from synchrotron radiation damping synchrotron radiation damping. The . The
injected beam within few damping times oscillates down to the st injected beam within few damping times oscillates down to the stored beam ored beam
merging with it. merging with it. Liouville Liouville theorem is not violated because synchrotron theorem is not violated because synchrotron
radiation is a non radiation is a non- -Hamiltonian phenomenon. The equilibrium distribution of Hamiltonian phenomenon. The equilibrium distribution of
the stored beam is usually the stored beam is usually gaussian gaussian. .
For For protons and ions protons and ions Liouville Liouville theorem does not allow the merging of the theorem does not allow the merging of the
beams and one needs to use special schemes that allow to fill th beams and one needs to use special schemes that allow to fill the whole e whole
available transverse acceptance. The final distribution is irre available transverse acceptance. The final distribution is irregular. gular.
11
Multiturn
Multiturn
Injection
Injection
for Protons and Ions
for Protons and Ions
F. Sannibale
The typical scheme uses a The typical scheme uses a fast orbit bump fast orbit bump in the septum area in in the septum area in
combination with a lattice tuned on a combination with a lattice tuned on a resonant resonant betatron betatron tune tune: :
New beam is injected in the same bucket every revolution period New beam is injected in the same bucket every revolution period. .
The bump is properly shifted between the injected bunches. The bump is properly shifted between the injected bunches.
While the resonant tune generates a constant rotation in the While the resonant tune generates a constant rotation in the
transverse phase space. transverse phase space.
By selecting the proper combination of tune and speed of the By selecting the proper combination of tune and speed of the bump, it bump, it
is possible to fill the phase space in an almost uniform way (be is possible to fill the phase space in an almost uniform way (beam am
painting). painting).
For example, for a fractional For example, for a fractional
horizontal tune of 0.25: horizontal tune of 0.25:
The first of such schemes was The first of such schemes was
used in 1953 at the used in 1953 at the
COSMOTRON Cyclotron in COSMOTRON Cyclotron in
Brookhaven Brookhaven
x
x
x
x
1
x
x
1
2
x
x
1
3
2
x
x
1
4
3
2
x
x
1
5
4
3
2
x
x
1
2 6
5
4
3
x
x
1
2
3
4
5
6
Longitudinal painting is also possible by properly changing the Longitudinal painting is also possible by properly changing the energy energy
of the injected beam every injection cycle. of the injected beam every injection cycle.
12
Charge Exchange Injection
Charge Exchange Injection
F. Sannibale
Stored protons Stored protons
Stripping foil Stripping foil Stripping foil Stripping foil
H H
- -
stripped protons stripped protons
Stripping foil Stripping foil
Bump Magnet Bump Magnet
On 1963, G.I. On 1963, G.I. Budker Budker and G.I. and G.I. Dimov Dimov at at Novosibirsk Novosibirsk conceived a new conceived a new
multiple injection scheme involving H multiple injection scheme involving H
- -
ions: ions:
The stripping foil must be very efficient in removing the elect The stripping foil must be very efficient in removing the electrons from rons from
the negative ions (~99%) and at the same time the perturbation t the negative ions (~99%) and at the same time the perturbation the foil he foil
induces on the stored proton beam should be as small as possible induces on the stored proton beam should be as small as possible. .
Carbon and aluminum ~ 10 to 100 Carbon and aluminum ~ 10 to 100 m thickness are typically used. m thickness are typically used.
Stripping is a Stripping is a- -non Hamiltonian phenomenon so non Hamiltonian phenomenon so Liouville Liouville theorem does theorem does
not apply and multiple injections on the same phase space area a not apply and multiple injections on the same phase space area are re
allowed. allowed.
13
Exploiting
Exploiting
Betatron
Betatron
Resonances
Resonances
F. Sannibale
By setting the lattice on tune By setting the lattice on tune
resonances resonances, one can generate , one can generate
islands in the phase space. islands in the phase space.
The example shows a 4 The example shows a 4
th th
order order
resonance resonance
Simulation by M. Giovannozzi
In fact, In fact, resonances resonances and and
phase space islands have phase space islands have
been efficiently exploited for been efficiently exploited for
both injecting or extracting both injecting or extracting
the beam the beam
Resonances Resonances are usually are usually
dangerous and carefully dangerous and carefully
avoided in designing and avoided in designing and
operating a storage ring. operating a storage ring.
Anyway, there are exceptions. Anyway, there are exceptions.
14
Resonant Schemes
Resonant Schemes
F. Sannibale

Simulation by M. Giovannozzi
The 4 The 4
th th
resonance shown in the resonance shown in the
previous viewgraph can be used for a previous viewgraph can be used for a
resonant extraction scheme for resonant extraction scheme for
example. example.
The simulation shows a slow bump The simulation shows a slow bump
bringing the beam close to the septum bringing the beam close to the septum
wall and a fast kick extracting the wall and a fast kick extracting the
beam from the storage ring to other beam from the storage ring to other
side of the septum wall into the side of the septum wall into the
transferline transferline. .
Schemes like this are studied for Schemes like this are studied for multi multi- -turn extraction turn extraction as required from as required from
some fixed target applications. In the example, because of the 0 some fixed target applications. In the example, because of the 0.25 tune, .25 tune,
the islands rotate 90 deg on every turn and are extracted in fou the islands rotate 90 deg on every turn and are extracted in four turns. r turns.
The central core is then extracted by the fast kicker. The central core is then extracted by the fast kicker.
Injection schemes exploiting Injection schemes exploiting betatron betatron resonances resonances, have been , have been
successfully tested in few storage rings. In the AURORA ring fro successfully tested in few storage rings. In the AURORA ring from m
Sumitomo for example, they use a half Sumitomo for example, they use a half- -integer resonance scheme. integer resonance scheme.
15
References
References
F. Sannibale
G.H. Rees G.H. Rees, Injection, CAS , Injection, CAS - - 5th General accelerator physics course, 5th General accelerator physics course, CERN 94 CERN 94- -01 01
Rende Rende Steerenberg Steerenberg, talk at AXEL , talk at AXEL 2005, March 17, 2005. 2005, March 17, 2005.
16
Possible
Possible
Homeworks
Homeworks
F. Sannibale
In designing a single turn injection system, we have the choice In designing a single turn injection system, we have the choice of of
locating our kicker at two different positions, say A and B. At locating our kicker at two different positions, say A and B. At A the A the
betatron betatron phase advance from the septum position is phase advance from the septum position is /2 and the /2 and the
horizontal beta function is 3m. AT B the phase advance is 5/8 horizontal beta function is 3m. AT B the phase advance is 5/8 but the but the
beta function is 7 m. Where would you place the kicker? If the r beta function is 7 m. Where would you place the kicker? If the required equired
stay clear at septum imposes a distance between the injected bea stay clear at septum imposes a distance between the injected beam m
trajectory and the closed orbit of 2 cm, what is the required an trajectory and the closed orbit of 2 cm, what is the required angular kicker gular kicker
for storing the beam? The beta function at the septum is 5 m. for storing the beam? The beta function at the septum is 5 m.
If we want to upgrade the above system for stacking the beam by If we want to upgrade the above system for stacking the beam by
placing a second kicker in the mirror symmetry point with respec placing a second kicker in the mirror symmetry point with respect of the t of the
septum magnet. What would it be the required angular kicker in t septum magnet. What would it be the required angular kicker in the he
second kicker? second kicker?
If the RF frequency in our ring is 500 MHz and the harmonic num If the RF frequency in our ring is 500 MHz and the harmonic number is ber is
400, what will be the maximum pulse duration that the kicker pul 400, what will be the maximum pulse duration that the kicker pulse can se can
have? have?
1
Lecture No. 11
Lecture No. 11
Lifetime in Storage Rings
Lifetime in Storage Rings
2
Introduction
Introduction
Lifetime
F. Sannibale
Particles are lost in accelerators because of the accelerator f Particles are lost in accelerators because of the accelerator finite aperture. inite aperture.
The limiting aperture in accelerators can be The limiting aperture in accelerators can be physical physical or or dynamic dynamic. .
The vacuum chamber defines the physical aperture, while transver The vacuum chamber defines the physical aperture, while transverse and se and
momentum acceptances of the accelerator define the dynamical one momentum acceptances of the accelerator define the dynamical one. .
Processes important for the lifetime include: Processes important for the lifetime include: elastic and inelastic residual elastic and inelastic residual
gas scattering gas scattering, , scattering with the other particles in the beam scattering with the other particles in the beam, , quantum quantum
lifetime for electrons and positrons lifetime for electrons and positrons, , tune tune resonances resonances, ,
Many processes can excite particles on orbits larger than the n Many processes can excite particles on orbits larger than the nominal. ominal.
If the displacement in the new orbit is larger than the aperture If the displacement in the new orbit is larger than the aperture, the particle is , the particle is
obviously lost. obviously lost.
For most applications beam needs to be stored for as long as po For most applications beam needs to be stored for as long as possible, so it ssible, so it
is very important to contain the above effects within acceptable is very important to contain the above effects within acceptable values. values.
Such a requirement has important consequences on the design con Such a requirement has important consequences on the design constraints. straints.
For example, limiting the effects of the residual gas scattering For example, limiting the effects of the residual gas scattering pushes pushes
towards ultra high vacuum technologies. towards ultra high vacuum technologies.
Damping plays a major role in the electron/positron case. For p Damping plays a major role in the electron/positron case. For protons and rotons and
heavy ions, lifetime is usually much longer but any perturbation heavy ions, lifetime is usually much longer but any perturbation will will
progressively build progressively build- -up and generate losses. up and generate losses.
3
The Concept of Lifetime
The Concept of Lifetime
Lifetime
F. Sannibale
By defining the By defining the lifetime lifetime as: as:
From the last equation, one can see that the lifetime is define From the last equation, one can see that the lifetime is defined as the time d as the time
required for the beam to reduce its number of particles to required for the beam to reduce its number of particles to 1/ 1/e e of the initial of the initial
value. value.
Lifetime due to the individual effects (gas, Lifetime due to the individual effects (gas, Touschek Touschek, ) can be similarly , ) can be similarly
defined. The total lifetime will be then obtained by summing the defined. The total lifetime will be then obtained by summing the individual individual
contributions: contributions:
....
1 1 1 1
3 2 1
+ + + =

With this definition, the problem of calculating the lifetime i With this definition, the problem of calculating the lifetime is reduced to the s reduced to the
evaluation of the single lifetime components. evaluation of the single lifetime components.
( ) constant with dt t N dN =
t
e N N

=
0
1
=
In a loss process, the number of particles lost at the time In a loss process, the number of particles lost at the time t t is proportional to is proportional to
the number of particles present in the beam at the time the number of particles present in the beam at the time t t: :
4
Is the Constant Lifetime
Is the Constant Lifetime
Model Accurate?
Model Accurate?
Lifetime
F. Sannibale
In fact, in most of the electron storage rings the lifetime act In fact, in most of the electron storage rings the lifetime actually depends on ually depends on
current. current.
Additionally, synchrotron radiation by hitting the vacuum chamb Additionally, synchrotron radiation by hitting the vacuum chamber transfers er transfers
the energy required to the molecules trapped in the vacuum chamb the energy required to the molecules trapped in the vacuum chamber wall to er wall to
be released ( be released (gas gas desorption desorption). ).
Because of this, for higher stored currents, the synchrotron ra Because of this, for higher stored currents, the synchrotron radiation diation
intensity increases generating more intensity increases generating more desorption desorption and increasing the pressure and increasing the pressure
in the vacuum chamber ( in the vacuum chamber (dynamic pressure dynamic pressure). This will increase the scattering ). This will increase the scattering
of the beam with the residual gas, with a consequent reduction o of the beam with the residual gas, with a consequent reduction of the beam f the beam
lifetime. lifetime.
The previous model, where the lifetime was assumed constant, is The previous model, where the lifetime was assumed constant, is often too often too
simple for describing the case of real accelerators. simple for describing the case of real accelerators.
Anyway, for reasonably small variations of the current, the con Anyway, for reasonably small variations of the current, the constant lifetime stant lifetime
assumption is locally valid and it is widely used. assumption is locally valid and it is widely used.
In fact, the In fact, the Touschek Touschek effect effect (discussed later), whose contribution dominates (discussed later), whose contribution dominates
the losses in many of the present electron accelerators, depends the losses in many of the present electron accelerators, depends on current. on current.
When the stored current decreases with time, the losses due to When the stored current decreases with time, the losses due to Touschek Touschek
decrease as well and the lifetime increases. decrease as well and the lifetime increases.
5
Example of Lifetime in Real
Example of Lifetime in Real
Accelerators
Accelerators
Lifetime
F. Sannibale
ALS ALS
DA DA NE NE
Electrons Positrons
HERA HERA
Photon Photon- -Factory Factory
6
Dynamic Aperture and
Dynamic Aperture and
Momentum Acceptance
Momentum Acceptance
Lifetime
F. Sannibale
Quite often in existing storage rings, the aperture is not limi Quite often in existing storage rings, the aperture is not limited by the ted by the
vacuum chamber size. vacuum chamber size.
Cornell
ILC-DR
In fact, In fact, nonlinearities nonlinearities in the fields of the magnets create resonance in the fields of the magnets create resonance
islands in the phase space that can capture particles with lar islands in the phase space that can capture particles with large amplitude ge amplitude
orbits and bring them in collision with the vacuum chamber. orbits and bring them in collision with the vacuum chamber.
This effect creates a virtual aperture for This effect creates a virtual aperture for
the machine which is usually referred as the the machine which is usually referred as the
dynamic aperture dynamic aperture
Due to their strong nonlinear nature, Due to their strong nonlinear nature,
dynamic apertures can be calculated only dynamic apertures can be calculated only
numerically. numerically.
In the longitudinal plane, the In the longitudinal plane, the momentum acceptance momentum acceptance is limited by is limited by
the the size of the RF bucket size of the RF bucket or by the or by the dynamic aperture for the off dynamic aperture for the off- -
momentum particles momentum particles. In fact, off . In fact, off- -energy particles in energy particles in dispersive dispersive
regions can hit the dynamic aperture of the ring even if their regions can hit the dynamic aperture of the ring even if their
momentum difference is still within the limits of the RF accepta momentum difference is still within the limits of the RF acceptance. nce.
7
Cross Section
Cross Section
of a Scattering Event
of a Scattering Event
Lifetime
F. Sannibale
In scattering processes it is useful to define as In scattering processes it is useful to define as cross section cross section the the
event rate per unit incident flux and per target particle event rate per unit incident flux and per target particle. .
For an interaction with cross section For an interaction with cross section , the number of events per second , the number of events per second
(event rate), in the rest frame of the particle group 2 for exam (event rate), in the rest frame of the particle group 2 for example, is given ple, is given
by: by:
( )( )
IR IR
V v n n V n v n N
dt
dN
N
1 2 1 2 1 1 2 1
= = = =
where where n n
1 1
and and n n
2 2
are the densities of the two groups of particles are the densities of the two groups of particles v v
1 1
is the is the
velocity of the particle group 1 and velocity of the particle group 1 and V V
IR IR
is the volume of the region where is the volume of the region where
the two particles interact. the two particles interact.
The equation above applies for uniform densities The equation above applies for uniform densities n n
1 1
, , n n
2 2
and constant and constant . .
For the more general case where these quantities For the more general case where these quantities depend on position, the depend on position, the
above expression must be replaced with: above expression must be replaced with:
=
IR
V
dz dy dx n n v N
2 1 1

Let us consider two groups of particles. Particles in the same Let us consider two groups of particles. Particles in the same group have group have
same momentum and are distributed in uniform spatial distributio same momentum and are distributed in uniform spatial distributions. ns.
8
Incident positive
particles
Gas Lifetime:
Gas Lifetime:
Elastic Scattering
Elastic Scattering
Lifetime
F. Sannibale
When a charged particle passes close to a residual gas molecule When a charged particle passes close to a residual gas molecule, it is , it is
deflected by the electric field of the molecule nucleus. deflected by the electric field of the molecule nucleus.
This phenomenon is a particular case of This phenomenon is a particular case of
Coulomb scattering Coulomb scattering and it is usually referred as and it is usually referred as
Rutherford scattering Rutherford scattering, after the name of the , after the name of the
English scientist that first discovered it in 1911. English scientist that first discovered it in 1911.
Rutherford experiments were quite important Rutherford experiments were quite important
because proved that the atom mass is not because proved that the atom mass is not
uniformly distributed (Thomson model) but uniformly distributed (Thomson model) but
instead concentrated in a very small positively instead concentrated in a very small positively
charged part of the atom, the nucleus. charged part of the atom, the nucleus.
The equation of motion for the problem can be solved showing th The equation of motion for the problem can be solved showing that the at the
trajectories of the scattered particles are hyperbolae ( trajectories of the scattered particles are hyperbolae (Kepler Kepler problem). problem).
In the process, the incident particle does not loose energy, so In the process, the incident particle does not loose energy, so this kind of this kind of
scattering is referred as scattering is referred as elastic elastic. .
In a storage ring, when a beam particle scatters with a residua In a storage ring, when a beam particle scatters with a residual gas molecule l gas molecule
it undergoes to it undergoes to betatron betatron oscillations. If the oscillation amplitude is larger than oscillations. If the oscillation amplitude is larger than
the ring acceptance the particle is lost. the ring acceptance the particle is lost.
9
Rutherford Scattering
Rutherford Scattering
Cross Section
Cross Section
Lifetime
F. Sannibale
Rutherford calculated the differential cross section for the el Rutherford calculated the differential cross section for the elastic scattering astic scattering
of a charged particle with a nucleus: of a charged particle with a nucleus:
( )
( )
] [
2 sin
1
2
4
1
4
2
2
2
0
MKS
p c
Ze Z
d
d
Inc R

|
|
.
|
\
|
=
where where Z Z
Inc Inc
e e is the charge of the incident particle, is the charge of the incident particle, Ze Ze is the charge of the is the charge of the
nucleus, nucleus, c c and and p p are the velocity and the momentum of the incident particles are the velocity and the momentum of the incident particles
and and is the scattering angle. is the scattering angle. is the solid angle. is the solid angle.
In deriving the previous equation, screening effects of the ato In deriving the previous equation, screening effects of the atom electrons m electrons
and direct inelastic scattering with the atoms electrons were ne and direct inelastic scattering with the atoms electrons were neglected glected
because small. Nucleus recoil has been neglected as well. because small. Nucleus recoil has been neglected as well.
x
z
y
For small angles, the screening from the molecule electrons mus For small angles, the screening from the molecule electrons must be taken t be taken
into account and for large scattering the nucleus finite size mu into account and for large scattering the nucleus finite size must be st be
considered. considered.
10
Gas Lifetime:
Gas Lifetime:
Elastic Scattering
Elastic Scattering
Lifetime
F. Sannibale
In the case of a beam of In the case of a beam of N N particles scattering on a residual gas molecule in particles scattering on a residual gas molecule in
a storage ring, the accelerator aperture will limit the scattere a storage ring, the accelerator aperture will limit the scattered angle to some d angle to some
value value
MAX MAX
. For scattered angles larger than . For scattered angles larger than
MAX MAX
the particle will be lost. the particle will be lost.
If If n n is the gas molecule density, is the gas molecule density, A A
T T
the beam transverse size, the beam transverse size, L L the ring the ring
length, length, T T the revolution period and the revolution period and c c the beam velocity, then: the beam velocity, then:
R molecules particles beam
Gas
N
dt
dN
=
By using the definition of cross section, the rate of losses is By using the definition of cross section, the rate of losses is given by: given by:
L
c
A
N
T A
N
T T
particles beam
= =
L nA N
T molecules
=

d
d
d
d d
d
d
MAX
R
Lost
R
R
sin
2
0

=
=
( )

|
|
.
|
\
|
=

MAX
d
p c
Ze Z c N n
dt
dN
Inc
Gas
2 sin
sin
4 2
4
2
2
2
0
( )
( )
] [
2 tan
1
4
2
2
2
2
0
MKS scattering elastic gas for rate Loss
p c
Ze Z c N n
dt
dN
MAX
Inc
Gas

|
|
.
|
\
|
=
11
Gas Lifetime:
Gas Lifetime:
Elastic Scattering
Elastic Scattering
Lifetime
F. Sannibale
760
] [
0
Torr
P
n M n =
The number of molecules for cubic centimeter, for a gas at 0 The number of molecules for cubic centimeter, for a gas at 0
o o
C C and at 760 and at 760
Torr Torr, is given by the , is given by the Loschmidt's Loschmidt's constant constant n n
0 0
= 2.68675 = 2.68675 x x 10 10
25 25
m m
- -3 3
( )
A
T
Inc
Torr
Gas
p c
Ze Z
P
N c Mn
dt
dN

4
760
4
2
2
] [
2
0
0
|
|
.
|
\
|

If we assume that our gas is composed by If we assume that our gas is composed by
M M- -atomic molecules and that its pressure is atomic molecules and that its pressure is
P P, then the density of the gas is: , then the density of the gas is:
For a ring with acceptance For a ring with acceptance
A A
and for small and for small , ,
the maximum scattering angle at the scattering the maximum scattering angle at the scattering
point is: point is:
( )
( ) s
s
T
A
MAX
=
For an For an estimate estimate, we can replace , we can replace
T T
with its with its
average value along the ring: average value along the ring:
T
A
MAX
=
( )
Gas
t N N = exp
0
] [
4 760
2
2
0
2
0
] [
MKS
Ze Z
p c
n M c P
T
A
Inc Torr
Gas
|
|
.
|
\
|
Which used with the previous results gives: Which used with the previous results gives:
That integrated: That integrated:
with: with:
12
Gas Lifetime:
Gas Lifetime:
Inelastic Scattering
Inelastic Scattering
Lifetime
F. Sannibale
In the inelastic scattering, the incident particles In the inelastic scattering, the incident particles lose energy lose energy during the during the
process. process.
We can distinguish two main phenomena: We can distinguish two main phenomena:
Gas Gas bremsstrahlung bremsstrahlung: the incident particle is : the incident particle is
deflected by the molecule nucleus and because of deflected by the molecule nucleus and because of
the transverse acceleration, radiates a photon. the transverse acceleration, radiates a photon.
This effect is important for relativistic particles This effect is important for relativistic particles
Atom excitation Atom excitation: the interaction brings the atom : the interaction brings the atom
to ionization or into an excited state. to ionization or into an excited state.
The effect is important for non relativistic particles The effect is important for non relativistic particles
For both the processes, if the amount of lost energy is beyond For both the processes, if the amount of lost energy is beyond the the
momentum acceptance of the ring the particle is lost. momentum acceptance of the ring the particle is lost.
The lifetime contribution due to inelastic scattering is calcul The lifetime contribution due to inelastic scattering is calculated following ated following
the same steps used for the elastic case, replacing the cross se the same steps used for the elastic case, replacing the cross section for the ction for the
elastic scattering with the sum of the two cross elastic scattering with the sum of the two cross- -section terms for the inelastic section terms for the inelastic
case. case.
13
Gas
Gas
Bremsstrahlung
Bremsstrahlung
Lifetime
F. Sannibale
Bremsstrahlung Bremsstrahlung in accelerators is important for in accelerators is important for
relativistic electrons and positrons. relativistic electrons and positrons.
( ) ( ) ( ) [ ]
|
|
.
|
\
|
+
|
|
.
|
\
|
+
|
|
.
|
\
|
=
1 1 1 2
0
1 1
2
0 0
2
0
2
1
3
2
ln
3
4
1
3
4

E
Z
E E
r Z
d
d
B
( )
=
0 0
2
0
3 1
1
100
E E
c m
Z
functions screening
energy particle E
particle the by lost energy energy photon
constant structure fine
hc
e
radius electron classical m r

= =
=

2 1
0
0
2
15
0
,
137
1
4
10 818 . 2

The differential cross The differential cross- -section was first section was first
calculated by calculated by Bethe Bethe and and Heitler Heitler: :
Where: Where:
with: with:
For high relativistic electrons, the screening is maximum and t For high relativistic electrons, the screening is maximum and the cross he cross- -
section becomes ( section becomes (complete screening case complete screening case): ):
|
|
.
|
\
|
+
|
|
.
|
\
|
+
|
|
.
|
\
|
=
0
2
0 0
2
0
2
1
9
1
ln
3
1
209 . 5 1
3
4 4
E
Z
E E
r Z
d
d
B

nucleus
photon
14
Gas
Gas
Bremsstrahlung
Bremsstrahlung
Lifetime
F. Sannibale
After some manipulations, a simple expression for the lifetime After some manipulations, a simple expression for the lifetime contribution contribution
due to gas due to gas bremsstrahlung bremsstrahlung can be obtained: can be obtained:
|
|
.
|
\
|
=
0
ln
3
4 1
E
E
L
c
A
R Brem
Here, Here, E E
A A
is the is the energy acceptance energy acceptance of the storage ring. of the storage ring.
L L
R R
is is radiation length radiation length of the gas and is defined as the length required to of the gas and is defined as the length required to
the particle to lose (1 the particle to lose (1 - - e e
- -1 1
) of its energy when traveling trough the gas. ) of its energy when traveling trough the gas.
|
.
|
\
|
+
3 1
2 2
0
183
ln 4
9
2 1
Z
n Z r
L
R
In a real accelerator, the residual gas is a combination of dif In a real accelerator, the residual gas is a combination of different ferent
molecular species. Anyway, it turns out that the average molecular species. Anyway, it turns out that the average < <Z Z
2 2
> > over the over the
different species is ~ 50 which is approximately the value for n different species is ~ 50 which is approximately the value for nitrogen. itrogen.
This allow to write with good approximation: This allow to write with good approximation:
( )
] [ 0
] [
1
ln
14 . 153
nTorr A
hours
Brem
P E E

15
Atomic Excitation
Atomic Excitation
Lifetime
F. Sannibale
The differential cross The differential cross- -section for the atomic excitation contribution is very section for the atomic excitation contribution is very
similar to the one for the similar to the one for the bremsstrahlung bremsstrahlung: :
( ) ( ) ( ) [ ]
|
|
.
|
\
|
+
|
|
.
|
\
|
+
|
|
.
|
\
|
=
2 1 2 2
0
2 1
2
0 0
2
0 .
1
3
2
ln
3
8
1
3
4

E
Z
E E
r Z
d
d
Exc
( )
=
0 0
2
0
3 2
2
100
E E
c m
Z
functions screening
2 1
,
with different: with different: and: and:
For the electron accelerator case, these differences make the c For the electron accelerator case, these differences make the cross ross- -section section
for atomic excitation much smaller than the for atomic excitation much smaller than the bremsstrahlung bremsstrahlung one. one.
2
Z of instead Z
4 8 of instead
For extremely relativistic particles, the complete screen case For extremely relativistic particles, the complete screen case gives: gives:
|
|
.
|
\
|
+
|
|
.
|
\
|
+
|
|
.
|
\
|
=
0
2
0 0
2
0
1
9
1
ln
3
2
085 . 7 1
3
4 4
E
Z
E E
Zr
d
d
Exc

16
Gas Lifetime: Example of
Gas Lifetime: Example of
Vacuum Requirements
Vacuum Requirements
Lifetime
F. Sannibale
We already saw that for We already saw that for electrons electrons in the approximation of in the approximation of < <Z Z
2 2
> ~ 50, > ~ 50, we we
have for the gas have for the gas bremsstrahlung bremsstrahlung lifetime: lifetime:
In the same approximation, the inelastic gas scattering lifetim In the same approximation, the inelastic gas scattering lifetime becomes: e becomes:
( )
] [ 0
] [
1
ln
14 . 153
nTorr A
hours
Brem
P E E

[ ]
] [
] [
] [
] [
2
0
25 . 10
m
T
m
A
nTorr
GeV
hours
Gas
P
E
Evaluating these expressions for the typical electron ring case Evaluating these expressions for the typical electron ring case, one finds , one finds
that the requirement on vacuum is for that the requirement on vacuum is for dinamic dinamic pressures of the order of the pressures of the order of the
nTorr nTorr. .
17
Touschek
Touschek
Effect
Effect
Lifetime
F. Sannibale
The The Touschek Touschek is the dominant effect limiting the lifetime in many of is the dominant effect limiting the lifetime in many of
the modern electrons storage rings. the modern electrons storage rings.
Particles in the bunch are subjected to Particles in the bunch are subjected to betatron betatron oscillations. oscillations.
Coulomb scattering between the particles can transfer transverse Coulomb scattering between the particles can transfer transverse
momentum to the longitudinal plane. momentum to the longitudinal plane.
If this extra momentum brings the two scattered particles beyon If this extra momentum brings the two scattered particles beyond the d the
momentum acceptance of the ring, then the particles are lost. momentum acceptance of the ring, then the particles are lost.
This process is usually referred as the This process is usually referred as the Touschek Touschek
effect effect after the Austrian scientist that discovered it. after the Austrian scientist that discovered it.
The first observation was done in The first observation was done in
the early 60s in the early 60s in Frascati Frascati at ADA, at ADA,
the electron the electron- -positron accelerator positron accelerator
conceived by conceived by Touschek Touschek and the and the
first ever built. first ever built.
1921 1921- -1978 1978
ADA ADA
18
Touschek
Touschek
Effect
Effect
Lifetime
F. Sannibale
The effect can be properly investigated in the center of mass The effect can be properly investigated in the center of mass system (CMS), system (CMS),
where the particles are non where the particles are non- -relativistic. relativistic.
2 4 2
2
0 .
sin
3
sin
4 4r
d
d
Tousc
In this frame, the Coulomb scattering between two particles of In this frame, the Coulomb scattering between two particles of the same the same
specie and with equal but opposite momentum specie and with equal but opposite momentum p p, is regulated by the , is regulated by the Mller Mller
differential cross differential cross- -sections: sections:
velocity CMS c
cos p p
S

In the CMS, the longitudinal component of the In the CMS, the longitudinal component of the
momentum due to the scattering is: momentum due to the scattering is:
Which in the laboratory system becomes: Which in the laboratory system becomes:

cos ~ p p E
c
p p
S S S
=
|
.
|
\
|
=
where the ~ sign is a good approximation because the particles a where the ~ sign is a good approximation because the particles are non re non- -
relativistic in the CMS. relativistic in the CMS.
The last equation shows how the momentum transfer in the labora The last equation shows how the momentum transfer in the laboratory tory
system is system is amplified by a factor amplified by a factor . .
p
s
y
x

cos sin cos
=
d d d sin =
19
Touschek
Touschek
Effect
Effect
Lifetime
F. Sannibale
If If p p
s s
is larger than the momentum acceptance is larger than the momentum acceptance p p
A A
, both the scattered , both the scattered
particles are lost. And the condition for losing a particle beco particles are lost. And the condition for losing a particle becomes: mes:
The The Mller Mller cross cross- -section can now be section can now be
integrated within this limits obtaining: integrated within this limits obtaining:
>
p
p
A
cos
After some additional algebra and assuming After some additional algebra and assuming gaussian gaussian distributions, we distributions, we
finally obtain the finally obtain the Touschek Touschek lifetime for a lifetime for a flat beam flat beam: :
( ) ( )
( )
T
A X S Y X
Tousch
C
p p
N c r
2
0
2 3 3
2
0
.
1
4
1

=
+ =

ln 1
1 8
2 4
2
0
.
r
Tousch
( ) 5 . 1
78 . 1
1
ln ~ 1 ln
2
1 1
2

|
|
.
|
\
|
|
|
.
|
\
|
=

T
u
T T
T T
du e
u u
u
C
T

2
0
|
|
.
|
\
|
=
X
A
T
p
p

with with
and and
A similar equation can be obtained for the case of round beams. A similar equation can be obtained for the case of round beams.
and where the approximate expression can be used for and where the approximate expression can be used for
T T
< 1. < 1.
20
Minimizing
Minimizing
Touschek
Touschek
Losses
Losses
Lifetime
F. Sannibale
In many electron (and positron) storage rings, high current and In many electron (and positron) storage rings, high current and small small
emittances emittances are usually required. This makes of the are usually required. This makes of the Touschek Touschek effect the major effect the major
responsible for particle losses in such rings. responsible for particle losses in such rings.
( ) ( )
( )
T
A X S Y X
Tousch
C
p p
N c r
2
0
2 3 3
2
0
.
1
4
1
=
In most In most colliders colliders, the energy is usually fixed, larger , the energy is usually fixed, larger emittances emittances (and thus (and thus
larger beam sizes) are welcome while the current should be as hi larger beam sizes) are welcome while the current should be as high as gh as
possible. Short bunches are preferred (hourglass effect). possible. Short bunches are preferred (hourglass effect).
Acceptance
2
0
2
RF V
p
p
RF
A
Depending on the application, a tradeoff between the different Depending on the application, a tradeoff between the different requirements requirements
must be defined. must be defined.
In synchrotron light sources, higher beam energy and longer bun In synchrotron light sources, higher beam energy and longer bunches ches
(harmonic cavities) can be used, while (harmonic cavities) can be used, while emittances emittances (and beam sizes) must be (and beam sizes) must be
small and the current must be high. small and the current must be high.
For all applications, the momentum For all applications, the momentum
acceptance must be maximized acceptance must be maximized
Dynamic aperture Dynamic aperture
21
Quantum Lifetime
Quantum Lifetime
Lifetime
F. Sannibale
At a fixed observation point along a storage ring, the transvers At a fixed observation point along a storage ring, the transverse motion of a e motion of a
particle is sampled as a pure sinusoidal oscillation: particle is sampled as a pure sinusoidal oscillation:
( ) y x T t a x
T
T T
, sin = + =

T
a
Usually, tunes are chosen in order to avoid Usually, tunes are chosen in order to avoid resonances resonances. In such a situation at . In such a situation at
a fixed a fixed azimuthal azimuthal position, a particle turn after turn sweeps all possible position, a particle turn after turn sweeps all possible
positions between the envelope: positions between the envelope:
In the presence of synchrotron radiation, photon emission random In the presence of synchrotron radiation, photon emission randomly changes ly changes
the invariant the invariant a a and consequently changes the trajectory envelope as well. and consequently changes the trajectory envelope as well.
Cumulative photon emission can bring the particle envelope beyon Cumulative photon emission can bring the particle envelope beyond the ring d the ring
acceptance in some acceptance in some azimuthal azimuthal point and the particle is lost. point and the particle is lost.
The explained loss mechanism is responsible for the so The explained loss mechanism is responsible for the so- -called called transverse transverse
quantum lifetime. quantum lifetime.
Similar arguments apply also for the longitudinal plane and the Similar arguments apply also for the longitudinal plane and the longitudinal longitudinal
quantum lifetime quantum lifetime can be defined as well. can be defined as well.
22
Quantum Lifetime
Quantum Lifetime
Lifetime
F. Sannibale
Quantum lifetime was first estimated by Quantum lifetime was first estimated by Bruck Bruck and Sands: and Sands:
( )
lifetime quantum Transverse
y x T A
A
T T
T
T
D Q
T T
, 2 exp
2 2
2
2
=

( )
lifetime quantum al Longitudin
E
E A D Q
L L
2 2
2 exp
eV E
U
V e
E h
E J
U
V e
F
E h
E J E
RF
C
L RF
C
L
E
A
8
1
0 1
0
0 1
0
2
2
10 08 . 1
2

|
|
.
|
\
|

|
|
.
|
\
|
=

y x T
E
where
E
T T T T
,
2
0
2
=
|
|
.
|
\
|
+ =

For an For an iso iso- -magnetic ring: magnetic ring:
Transverse quantum lifetime sets the minimum requirement for th Transverse quantum lifetime sets the minimum requirement for the e
transverse aperture, while the longitudinal one defines the mini transverse aperture, while the longitudinal one defines the minimum mum
momentum acceptance necessary from the lifetime point of view. momentum acceptance necessary from the lifetime point of view.
Quantum lifetime very strongly Quantum lifetime very strongly
depends on the ratio between depends on the ratio between
acceptance and acceptance and rms rms size. size.
Values for this ratio of 6 or little larger Values for this ratio of 6 or little larger
are usually required. are usually required.
23
Tune
Tune
Resonances
Resonances
Lifetime
F. Sannibale
Tune Tune resonances resonances are carefully avoided in designing storage rings. are carefully avoided in designing storage rings.
In fact, particles trapped in a resonance can be quickly lost. In fact, particles trapped in a resonance can be quickly lost.
Lower order Lower order resonances resonances are usually more dangerous. are usually more dangerous.
(<~ 12 (<~ 12
th th
for protons and <~ 4 for protons and <~ 4
th th
for electrons) for electrons)
ntegers i i m l and order resonance m l i m l
y x
, , + = +
ALS ALS
Anyway, imperfections, nonlinear effects and Anyway, imperfections, nonlinear effects and
phenomena associated with momentum phenomena associated with momentum
diffusion can bring the particle on a resonance. diffusion can bring the particle on a resonance.
Examples of common tune shift effects in Examples of common tune shift effects in
storage rings: storage rings: non linear non linear multipole multipole terms in terms in
magnetic fields magnetic fields (tune shift on amplitude), (tune shift on amplitude),
beam beam- -beam effects during collision beam effects during collision, ,
wakefields wakefields (tune shift on current), (tune shift on current), inelastic inelastic
gas scattering combined with nonzero gas scattering combined with nonzero
chromaticity chromaticity, ,
The working point in the tune plane must be carefully selected The working point in the tune plane must be carefully selected in order to in order to
minimize the impact of all such effects. minimize the impact of all such effects.
24
Remarks on the Lifetime for
Remarks on the Lifetime for
Protons and Heavy Ions
Protons and Heavy Ions
Lifetime
F. Sannibale
In protons and heavy ions storage rings no damping is present. In protons and heavy ions storage rings no damping is present. As a As a
consequence any perturbation on the particle trajectories builds consequence any perturbation on the particle trajectories builds- -up and can up and can
eventually lead to the loss of the particle. eventually lead to the loss of the particle.
Other effects such as Other effects such as elastic gas scattering elastic gas scattering, ,
molecule excitation molecule excitation, , fluctuations in the fluctuations in the
magnetic and RF fields magnetic and RF fields, , Coulomb scattering Coulomb scattering
(intra (intra- -beam scattering) beam scattering), , , ,
add up to generate a lifetime of the order of add up to generate a lifetime of the order of
hundreds of hours typically. hundreds of hours typically.
On the other hand, important loss mechanisms On the other hand, important loss mechanisms
for electrons, become negligible for protons. for electrons, become negligible for protons.
These include for example, These include for example, Touschek Touschek and gas and gas
bremsstrahlung bremsstrahlung scattering. scattering.
Quite often in Quite often in colliders colliders, the interaction between the colliding beams, the so , the interaction between the colliding beams, the so- -
called called beam beam- -beam effect beam effect, becomes the main mechanism of losses. , becomes the main mechanism of losses.
25
References
References
Lifetime
F. Sannibale
A. A. Wrulich Wrulich, Single Beam Lifetime, , Single Beam Lifetime,
CAS CAS - - 5th General accelerator physics course, CERN 94 5th General accelerator physics course, CERN 94- -01 01
M. Sands M. Sands, The Physics of Electron Storage Rings. An Introduction, , The Physics of Electron Storage Rings. An Introduction,
SLAC Report 121 UC SLAC Report 121 UC- -28 (ACC) (1970) 28 (ACC) (1970)
26
Possible Homework
Possible Homework
Lifetime
F. Sannibale
Show how different lifetime contributions add up together for t Show how different lifetime contributions add up together for the total he total
lifetime. Explain the physical meaning of the constant lifetime. Explain the physical meaning of the constant (the inverse of the (the inverse of the
lifetime). lifetime).
Estimate the lifetime for the DA Estimate the lifetime for the DA NE electron beam. Use the line in the NE electron beam. Use the line in the
lifetime plot in the examples viewgraph. lifetime plot in the examples viewgraph.
Calculate the number of molecules per cm Calculate the number of molecules per cm
3 3
for a gas of N for a gas of N
2 2
at the pressure of at the pressure of
1 1 nTorr nTorr. .
Estimate the lifetime due to elastic gas scattering for a 1.9 Estimate the lifetime due to elastic gas scattering for a 1.9 GeV GeV electron electron
beam at a pressure of 1 beam at a pressure of 1 nTorr nTorr. Assume that the gas is mainly N . Assume that the gas is mainly N
2 2
(Z=7), that (Z=7), that
the average ring beta function is 1.5 m and that the ring accept the average ring beta function is 1.5 m and that the ring acceptance is 10 ance is 10
- -6 6
m. m.
Remember that Remember that
0 0
=8.8543 =8.8543 x x 10 10
- -12 12
F m F m
- -1 1
. .
For the same ring of the previous problem, calculate the lifeti For the same ring of the previous problem, calculate the lifetime due to gas me due to gas
bremsstrahlung bremsstrahlung for the case of 1% relative momentum acceptance. for the case of 1% relative momentum acceptance.
For the same ring, calculate also the For the same ring, calculate also the Touschek Touschek lifetime for a bunch current lifetime for a bunch current
of 10 of 10 mA mA and average and average rms rms beam sizes of 100 beam sizes of 100 m, 10 m, 10 m and 1 cm, for x, y and m and 1 cm, for x, y and
s respectively. Assume an average s respectively. Assume an average rms rms value for x of 60 value for x of 60 rad rad. .
Finally, estimate for the same ring the longitudinal quantum li Finally, estimate for the same ring the longitudinal quantum lifetime when fetime when
the longitudinal damping time is 10 ms. the longitudinal damping time is 10 ms.
27
Spherical Coordinates
Spherical Coordinates
And Solid Angle
And Solid Angle
Lifetime
F. Sannibale
x
z
y

cos
sin sin
cos sin
r z
r y
r x
=
=
=
d d dr r dz dy dx sin
2
=
d d d sin =

4 sin
0
2
0
= =

d d d
1
Lecture No. 12
Lecture No. 12
Collective effects.
Collective effects.
Single and
Single and
Multibunch
Multibunch
Instabilities
Instabilities
2
Introduction
Introduction
Collective Effects
F. Sannibale
Charged particles in a beam mutually interact and interact with Charged particles in a beam mutually interact and interact with the the
conductive walls of the vacuum chamber. conductive walls of the vacuum chamber.
These effects are usually referred as These effects are usually referred as collective effects collective effects. .
For example, we already saw that the For example, we already saw that the Touschek Touschek effect can generate losses effect can generate losses
in electron storage rings. Other examples include space charge l in electron storage rings. Other examples include space charge limiting the imiting the
minimum minimum emittance emittance and the maximum current in proton storage rings, beam and the maximum current in proton storage rings, beam- -
beam effects reducing the luminosity performance in beam effects reducing the luminosity performance in colliders colliders, ,
Collective effects play a major role, quite often limiting the Collective effects play a major role, quite often limiting the final final
performance of an accelerator. performance of an accelerator.
In designing high performance accelerators, collective effects In designing high performance accelerators, collective effects need to be need to be
carefully taken into account and solutions for minimizing these carefully taken into account and solutions for minimizing these effects need effects need
to be adopted. to be adopted.
Additionally, collective effects make particles within the bunc Additionally, collective effects make particles within the bunch and h and
between bunches communicate, allowing for single bunch and between bunches communicate, allowing for single bunch and multibunch multibunch
instabilities. instabilities.
Solutions can be Solutions can be passive passive, when in the design phase the parameters are , when in the design phase the parameters are
chosen in order to contain collective effects, or chosen in order to contain collective effects, or active active where the accelerator where the accelerator
operates above instability threshold but operates above instability threshold but feedback feedback systems damp the systems damp the
instabilities down. instabilities down.
3
Space
Space
-
-
Charge
Charge
Collective Effects
F. Sannibale
We already saw how Coulomb scattering between two particles in t We already saw how Coulomb scattering between two particles in the he
beam can generate particle losses by the beam can generate particle losses by the Touschek Touschek Effect. Effect.
But Coulomb interaction is also responsible for the so But Coulomb interaction is also responsible for the so- -called called space space
charge charge effect. In this case, the generic particle in the bunch experie effect. In this case, the generic particle in the bunch experiences nces
the the collective collective Coulomb force due to the field generated by the charge of Coulomb force due to the field generated by the charge of
all the other particles in the bunch. all the other particles in the bunch.
Such fields, referred also as Such fields, referred also as self self- -fields fields, are quite nonlinear and their , are quite nonlinear and their
evaluation usually requires numerical techniques. evaluation usually requires numerical techniques.
Anyway, by using the proper approximation, it is possible to obt Anyway, by using the proper approximation, it is possible to obtain ain
analytical solutions that gives us some useful insights on the e analytical solutions that gives us some useful insights on the effect. ffect.
4
Teng
Teng
Solution
Solution
and the
and the
Gaussian
Gaussian
case
case
Collective Effects
F. Sannibale
( ) ( )
,
1
,
1
0 0
y
b a b
E x
b a a
E
y x
+
=
+
=

By assuming a By assuming a continuous continuous (non (non- -bunched) beam with bunched) beam with constant linear charge constant linear charge
density density and with a and with a stationary uniform elliptical stationary uniform elliptical distribution in the transverse distribution in the transverse
plane, plane, Teng Teng in 1960 found the following expression for the fields in 1960 found the following expression for the fields inside inside the the
beam: beam:
( ) ( )
x
b a a
c
B y
b a b
c
B
y x
+
=
+
=

0 0
,
For a more realistic For a more realistic gaussian gaussian distribution in the transverse plane distribution in the transverse plane
and for and for x x << <<
x x
and and y y << <<
y y
: :
( ) ( )
,
2
1
,
2
1
0 0
y E x E
y x y
y
y x x
x

+
=
+
=
( ) ( )
x
c
B y
c
B
y x x
y
y x y
x

+
=
+
=
2
,
2
0 0
For both cases the fields scale linearly with For both cases the fields scale linearly with x x and and y y, and: , and:
, ,
x y y x
E
c
B E
c
B

= =
with with a a and and b b the ellipse half the ellipse half- -axes, and the beam moving along axes, and the beam moving along z z with velocity with velocity c c. .
5
Effects of the Space
Effects of the Space
-
-
Charge
Charge
Collective Effects
F. Sannibale
( ) B v E q F

+ =
Such space charge fields exert forces on the beam particles, wh Such space charge fields exert forces on the beam particles, whose ose
intensities are given by the intensities are given by the Lorentz Lorentz Force: Force:
( ) ( ) ( )x qE cB E q F
x y x x
2 2
1 1 = =
The last equations show that the The last equations show that the space charge forces become negligible for space charge forces become negligible for
relativistic beams relativistic beams. .
Such a situation generates a Such a situation generates a betatron betatron tune shift with current tune shift with current for the particles for the particles
in the core of the beam. in the core of the beam.
And using the And using the
previous relations: previous relations:
( ) ( ) ( ) y qE cB E q F
y x y y
2 2
1 1 = + =
They also show that for the non They also show that for the non- -relativistic beam, the forces are repulsive relativistic beam, the forces are repulsive
and proportional to the distance from the beam center. and proportional to the distance from the beam center.
This is equivalent to a defocusing This is equivalent to a defocusing quadrupole quadrupole in both planes with strength in both planes with strength
proportional to the current in the beam. proportional to the current in the beam.
For the non For the non- -core particles the linear dependence of the force breaks and core particles the linear dependence of the force breaks and
numerical calculations are required for the evaluation of the sp numerical calculations are required for the evaluation of the space charge ace charge
effects. effects.
6
Vacuum Chamber Effects:
Vacuum Chamber Effects:
Image Charge
Image Charge
Collective Effects
F. Sannibale
Particle beams requires ultra high vacuum pressures, that can b Particle beams requires ultra high vacuum pressures, that can be achieved e achieved
inside special metallic vessels called inside special metallic vessels called vacuum chambers vacuum chambers. .
For the Maxwell equations, the electric field associated with t For the Maxwell equations, the electric field associated with the particle he particle
beam, must terminate perpendicularly on the chamber beam, must terminate perpendicularly on the chamber equipotential equipotential
conductive walls. conductive walls.
In the lab system the beam electromagnetic field of a relativis In the lab system the beam electromagnetic field of a relativistic particle is tic particle is
transversely confined within an angle of ~ transversely confined within an angle of ~ 1/ 1/ (where (where is the particle energy is the particle energy
in rest mass units). in rest mass units).
This boundary conditions requires that the same amount of charg This boundary conditions requires that the same amount of charge but with e but with
opposite sign, travels on the vacuum chamber together with the opposite sign, travels on the vacuum chamber together with the beam. Such beam. Such
charge is referred as the charge is referred as the image charge image charge. .
Negative Negative
Charged Beam Charged Beam
1
Negative Negative
Negative Negative
7
Vacuum Chamber Wake Fields
Vacuum Chamber Wake Fields
Collective Effects
F. Sannibale
The beam and its electromagnetic field travel inside the vacuum The beam and its electromagnetic field travel inside the vacuum chamber chamber
while the image charge travels while the image charge travels on on the chamber itself. the chamber itself.
Any variation on the chamber profile, on the chamber material, Any variation on the chamber profile, on the chamber material, or on the or on the
material properties breaks this configuration. material properties breaks this configuration.
The result is that the beam loses a (usually small) part of it The result is that the beam loses a (usually small) part of it is energy that is energy that
feeds the electromagnetic fields that remain after the passage o feeds the electromagnetic fields that remain after the passage of the beam. f the beam.
Such fields are referred as Such fields are referred as wake fields wake fields. .
Vacuum chamber wake fields generated by beam particles, mainly Vacuum chamber wake fields generated by beam particles, mainly affect affect
trailing particles and in the case of ultra trailing particles and in the case of ultra- -relativistic beams can relativistic beams can only only affect affect
trailing particles. trailing particles.
8
Wake Fields and Instabilities
Wake Fields and Instabilities
Collective Effects
F. Sannibale
Wake fields are transient effects, they are generated during th Wake fields are transient effects, they are generated during the beam e beam
passage and then last for a finite amount of time that depends o passage and then last for a finite amount of time that depends on the n the
particular wake and on the geometry of the vacuum chamber. particular wake and on the geometry of the vacuum chamber.
If the wake field lasts for the duration of a bunch (hundreds o If the wake field lasts for the duration of a bunch (hundreds of f ps ps typically), typically),
particles in the bunch tail can interact with the wakes due to t particles in the bunch tail can interact with the wakes due to the particles in he particles in
the head and the head and single bunch instabilities single bunch instabilities can be triggered (distortion of the can be triggered (distortion of the
longitudinal distribution, bunch lengthening, ). longitudinal distribution, bunch lengthening, ).
If the wake field lasts longer, for example for the distance in If the wake field lasts longer, for example for the distance in time between time between
bunches (several ns typically), wakes from leading bunches can i bunches (several ns typically), wakes from leading bunches can interact with nteract with
following bunches and potentially generate following bunches and potentially generate multi multi- -bunch or coupled bunch bunch or coupled bunch
instabilities instabilities. .
9
Wake Potentials
Wake Potentials
Collective Effects
F. Sannibale
For this reason, in investigating longitudinal wake fields we c For this reason, in investigating longitudinal wake fields we consider onsider only the only the
electric component of the wake fields electric component of the wake fields. .
In practical cases, wake potentials are integrated over some fi In practical cases, wake potentials are integrated over some finite length. The nite length. The
length of the integration path must be longer than the wake fiel length of the integration path must be longer than the wake field maximum d maximum
extension. extension.
( ) ( ) , , , , , ,

= s d t t r r s E t t r r V
lead trail trail lead W lead trail trail lead W

It is often convenient to deal with wake It is often convenient to deal with wake
potentials instead of wake fields. The potentials instead of wake fields. The wake wake
potential potential is defined as the energy variation is defined as the energy variation
induced by the wake field of the leading induced by the wake field of the leading
particle on the unit charge trailing particle. particle on the unit charge trailing particle.
Wake fields effects can be divided into longitudinal and transv Wake fields effects can be divided into longitudinal and transversal. In the ersal. In the
longitudinal case the wakes affect the energy of the particles, longitudinal case the wakes affect the energy of the particles, while in the while in the
transverse case is their transverse momentum to be affected. transverse case is their transverse momentum to be affected.
x
y
s
lead
r
trail
r
lead
q
lead lead
t v s =
trail trail
t v s =
trail
q
In our results, In our results, s s = = vt vt with with v v constant, this is a very good approximation for constant, this is a very good approximation for
relativistic particles but it is also a reasonable assumption fo relativistic particles but it is also a reasonable assumption for the cases where r the cases where
the wake induced energy variation is small respect to the partic the wake induced energy variation is small respect to the particle energy . le energy .
10
Wake Functions
Wake Functions
Collective Effects
F. Sannibale
The The wake function wake function is instead defined as the energy variation induced by the is instead defined as the energy variation induced by the
wake field of a unit charge leading particle on the unit charge wake field of a unit charge leading particle on the unit charge trailing particle. trailing particle.
( )
( )
lead
lead trail trail lead W
lead trail trail lead
q
t t r r V
t t r r W

=
, ,
, ,

The The total wake potential total wake potential for a bunch with charge distribution for a bunch with charge distribution i i with: with:
( ) Nq t d r d t r i =

,
is given by: is given by:
( ) ( ) ( )
= t d r d t r i t t r r W t r V
trail trail trail trail

, , ,
,
The total wake potential gives the energy variation that the tr The total wake potential gives the energy variation that the trailing particle ailing particle
experiences due to the wakes of the whole bunch. experiences due to the wakes of the whole bunch.
Very often in real accelerators, we deal with distributions tha Very often in real accelerators, we deal with distributions that live in t live in
the neighbor of the bunch center. In this case, it is sufficient the neighbor of the bunch center. In this case, it is sufficient to use the to use the
wakes on axis that can be obtained by setting wakes on axis that can be obtained by setting r r and and r r
trail trail
= 0 = 0 in the in the
previous expressions ( previous expressions (monopole wake monopole wake approximation). approximation).
11
Coupling Impedance
Coupling Impedance
Collective Effects
F. Sannibale
The wake function represents the interaction of the beam with t The wake function represents the interaction of the beam with the external he external
environment in the environment in the time domain time domain. .
( ) ( ) t t with d e r r W r r Z
trail
j
trail trail
= =

, , , ,

As for other phenomena, the equivalent As for other phenomena, the equivalent frequency domain frequency domain analysis can be analysis can be
very useful giving a different insight and additional interpreta very useful giving a different insight and additional interpretations. tions.
The frequency domain alter ego of the wake function is the The frequency domain alter ego of the wake function is the coupling coupling
impedance impedance, measured in Ohm and defined as the , measured in Ohm and defined as the Fourier transform of the Fourier transform of the
wake function wake function: :
If If I I is the Fourier transform of the charge distribution, the Fourie is the Fourier transform of the charge distribution, the Fourier transform r transform
of the total induced voltage is simply given by: of the total induced voltage is simply given by:
( ) ( ) ( ) , , , , ,
~
r I r r Z r r V
trail trail

=
And the time domain expression can be obtained by the inverse F And the time domain expression can be obtained by the inverse Fourier ourier
transform: transform:
( ) ( )

d e r r V r r V
j
trail trail
, ,
~
2
1
, ,

12
Interpretation of the
Interpretation of the
Coupling Impedance
Coupling Impedance
Collective Effects
F. Sannibale
The coupling impedance is a complex quantity with real and imag The coupling impedance is a complex quantity with real and imaginary parts: inary parts:
( ) ( ) ( ) , , , , , ,
trail j trail R trail
r r Z j r r Z r r Z

+ =
There is a strong analogy between wake field and electronic cir There is a strong analogy between wake field and electronic circuit theories. cuit theories.
This can be exploited and wakes can be represented by equivalent This can be exploited and wakes can be represented by equivalent circuits. circuits.
i
B C L R
Analogously to the circuit case, the resistive part of the coup Analogously to the circuit case, the resistive part of the coupling impedance ling impedance
is responsible for the beam losses, while the imaginary part def is responsible for the beam losses, while the imaginary part defines the ines the
phase relation between the beam excitation and the wake potentia phase relation between the beam excitation and the wake potential. l.
( )
L
C
R Q
LC
jQ
R
Z
R
R
R
= =
|
|
.
|
\
|
+
= ,
1
,
1
For example, the impedance of a parallel RLC circuit is often a For example, the impedance of a parallel RLC circuit is often associated to ssociated to
the impedance of the so the impedance of the so- -called called high order modes high order modes (HOM), single resonance (HOM), single resonance
wakes in the vacuum chamber. wakes in the vacuum chamber.
( )
( )
( )
>

<
=

0
1 4
4 1 1 sin
4 1 1 cos
0 0
2
2
2
2

Q
Q
Q
C
e W
R
R
Q
R
13
Narrow
Narrow
-
-
band and Broad
band and Broad
-
-
band
band
Coupling Impedances
Coupling Impedances
Collective Effects
F. Sannibale
Using this RLC model the Using this RLC model the HOMs HOMs can be classified in two main categories. can be classified in two main categories.
Narrow Narrow- -band impedances band impedances. These modes are characterized by relatively high . These modes are characterized by relatively high Q Q
and their spectrum is narrow. The associated wake last for a rel and their spectrum is narrow. The associated wake last for a relatively long time atively long time
making this modes important for making this modes important for multibunch multibunch instabilities. instabilities.
Broad Broad- -band impedances band impedances. These modes are characterized by a low . These modes are characterized by a low Q Q and their and their
spectrum is broader. The associated wake last for a relatively s spectrum is broader. The associated wake last for a relatively short time making hort time making
this modes important only for single bunch instabilities. this modes important only for single bunch instabilities.
14
Transverse Wake Fields Case
Transverse Wake Fields Case
Collective Effects
F. Sannibale
A similar approach and definitions can be used for the transver A similar approach and definitions can be used for the transverse wake se wake
case. case.
The The transverse wake function transverse wake function defines the defines the transverse momentum kick per transverse momentum kick per
unit leading charge and unit trailing charge unit leading charge and unit trailing charge due to the wake fields. due to the wake fields.
Transverse wake fields are excited when the beam passes out of c Transverse wake fields are excited when the beam passes out of center enter. If . If
the displacement is small enough only the the displacement is small enough only the dipole dipole term proportional to the term proportional to the
displacement is important. In such a situation, the displacement is important. In such a situation, the transverse dipole wake transverse dipole wake
function, function, defined as the transverse wake function for unit displacement, defined as the transverse wake function for unit displacement, can can
be used. be used.
Longitudinal and transverse wakes are representation of the sam Longitudinal and transverse wakes are representation of the same 3D wake e 3D wake
field and are linked each other by the Maxwell equations. field and are linked each other by the Maxwell equations.
The so The so- -called called Panofsky Panofsky- -Wenzel relations Wenzel relations allow to calculate one wake allow to calculate one wake
component when the other is known. component when the other is known.
The The transverse coupling impedance transverse coupling impedance is defined as the Fourier transform of is defined as the Fourier transform of
the transverse wake function times the transverse wake function times j . j .
15
Impedance of Accelerators
Collective Effects
F. Sannibale
In a real accelerators, the vacuum chamber has a very complex s In a real accelerators, the vacuum chamber has a very complex shape and hape and
includes many components that can potentially have trapped HOM includes many components that can potentially have trapped HOM. .
In summary, when the beam transits along the vacuum chamber it In summary, when the beam transits along the vacuum chamber it excites excites
wake fields. These can be classified in three main categories: wake fields. These can be classified in three main categories:
Anyway, not all the wakes excited by the beam can be trapped in Anyway, not all the wakes excited by the beam can be trapped in the the
vacuum chamber. In fact, for a given vacuum chamber geometry, it vacuum chamber. In fact, for a given vacuum chamber geometry, it exist a exist a
cutoff frequency cutoff frequency such that modes with frequency above cutoff propagates such that modes with frequency above cutoff propagates
along the chamber: along the chamber:
- - wake fields that travels with the beam (such as the space charg wake fields that travels with the beam (such as the space charge); e);
- - wake fields that are localized in some resonant structure in th wake fields that are localized in some resonant structure in the e
vacuum chamber (narrow and broad band HOM); vacuum chamber (narrow and broad band HOM);
- - high frequency wakes, above the vacuum chamber cutoff, that high frequency wakes, above the vacuum chamber cutoff, that
propagates along the vacuum chamber. This last category does propagates along the vacuum chamber. This last category does
not generate any net interaction with the beam unless they are not generate any net interaction with the beam unless they are
synchronous with the beam itself. synchronous with the beam itself.
size transverse chamber a
a
c
f
Cutoff

16
Collective Effects
F. Sannibale
Narrow band impedances, such as the ones due to the RF cavities Narrow band impedances, such as the ones due to the RF cavities and and
potentially to other components such as kickers and similar stru potentially to other components such as kickers and similar structures, are ctures, are
well localized and are usually treated independently. well localized and are usually treated independently.
In the case of broad band impedances, the contributions of the In the case of broad band impedances, the contributions of the single single
components (bellows, vacuum ports, transitions, diagnostics, ) components (bellows, vacuum ports, transitions, diagnostics, ) are are
individually calculated (mainly numerically) and/or measured and individually calculated (mainly numerically) and/or measured and then added then added
together. together.
Most of the circular accelerators with relatively long bunches Most of the circular accelerators with relatively long bunches (greater than (greater than
few tens of few tens of ps ps rms rms) present a total coupling impedance which is mainly ) present a total coupling impedance which is mainly
inductive. Scaled models, such as the inductive. Scaled models, such as the broadband resonator broadband resonator (parallel RLC (parallel RLC
with with Q Q ~ 1 ~ 1 and and 2 2
R R
~ ~ f f
Cutoff Cutoff
) or such as the empirical ) or such as the empirical SPEAR scaling SPEAR scaling model, model,
are often used to represent with some success the total broadban are often used to represent with some success the total broadband d
impedance of such accelerators. impedance of such accelerators.
Concerning the wake fields that propagates with the beam, we al Concerning the wake fields that propagates with the beam, we already ready
glanced on the space charge wake. Another example of such a ki glanced on the space charge wake. Another example of such a kind of nd of
wake is the one represented by the wake is the one represented by the resistive wall impedance resistive wall impedance, where the finite , where the finite
resistivity resistivity of the vacuum chamber walls generates a wake that can be often of the vacuum chamber walls generates a wake that can be often
expressed as the impedance of a high expressed as the impedance of a high- -frequency broad frequency broad- -band resonator. band resonator.
17
The Synchrotron Radiation
The Synchrotron Radiation
Wake Field
Wake Field
Collective Effects
F. Sannibale
The wake field due to synchrotron radiation, belongs to the cat The wake field due to synchrotron radiation, belongs to the category of the egory of the
wakes that propagates with the beam. wakes that propagates with the beam.
r
e
-
E
f
SP Emission Cone
Such a wake is important only for the relativistic particle cas Such a wake is important only for the relativistic particle case. e.
Relativistic particles on a curved trajectory emit synchrotron Relativistic particles on a curved trajectory emit synchrotron radiation (SR). radiation (SR).
The SR fields propagates in a cone of emission centered on the t The SR fields propagates in a cone of emission centered on the tangent to angent to
the beam trajectory at the emission point and with ~ the beam trajectory at the emission point and with ~ 1/ 1/ aperture. aperture.
The fields propagate at the speed of light, while the particles The fields propagate at the speed of light, while the particles move on the move on the
curved trajectory. For this reason, even if the particles are re curved trajectory. For this reason, even if the particles are relativistic the lativistic the
projection of their speed on the tangent direction is smaller th projection of their speed on the tangent direction is smaller than an c c. .
In other words, the SR wake field due to a particle in the tail In other words, the SR wake field due to a particle in the tail of the bunch of the bunch
can reach and interact with a particle in the head! can reach and interact with a particle in the head!
This is exact the opposite of what happens with vacuum chamber w This is exact the opposite of what happens with vacuum chamber wakes. akes.
18
Single Bunch Effects
Single Bunch Effects
Collective Effects
F. Sannibale
Broadband impedances have important effects on accelerators. Broadband impedances have important effects on accelerators.
In electron storage rings in the presence of radiation damping, In electron storage rings in the presence of radiation damping, the the
equilibrium distributions at low current are usually equilibrium distributions at low current are usually gaussian gaussian. By increasing . By increasing
the current per bunch, the wakes become stronger and can generat the current per bunch, the wakes become stronger and can generate non e non
gaussian gaussian equilibrium distributions. equilibrium distributions.
In In linacs linacs and in heavy particle accelerators, broad band impedances can and in heavy particle accelerators, broad band impedances can
generate generate emittance emittance and energy spread growth. and energy spread growth.
In all accelerators, if the current per bunch is increased furth In all accelerators, if the current per bunch is increased further, the wakes er, the wakes
can become strong enough to generate single bunch instabilities can become strong enough to generate single bunch instabilities that can that can
severely change the characteristics of the bunch and/or generate severely change the characteristics of the bunch and/or generate particle particle
losses. losses.
In what follows, some examples (not a complete review!) of such In what follows, some examples (not a complete review!) of such cases will cases will
be given. be given.
19
Potential Well Distortion:
The SR Wake Case
The SR Wake Case
Collective Effects
F. Sannibale
The example concerns the case of the synchrotron radiation (SR) The example concerns the case of the synchrotron radiation (SR) wake in wake in
electron storage rings. electron storage rings.
The SR wake due to the particles in the tail of The SR wake due to the particles in the tail of
the bunch interacts with particles in the head by the bunch interacts with particles in the head by
changing their energy. changing their energy.
This generates a distortion of the normally parabolic RF potent This generates a distortion of the normally parabolic RF potential ( ial (potential potential
well distortion well distortion). In this situation, the bunch is forced to a new equilibrium w ). In this situation, the bunch is forced to a new equilibrium with ith
a non a non- -gaussian gaussian longitudinal distribution. longitudinal distribution.
When a storage ring is tuned for short bunches When a storage ring is tuned for short bunches
(~ few (~ few ps ps rms rms), the SR becomes the dominant ), the SR becomes the dominant
wake. wake.
Head Tail
nominal bunch
distribution
SR wake
accelerates
bunch head
Total potential wake
(gaussian case)
20
The General Case
The General Case
Collective Effects
F. Sannibale
The potential well distortion mechanism, shown for the case of The potential well distortion mechanism, shown for the case of the SR wake, the SR wake,
is actually quite general and common to all kind of wakes in ele is actually quite general and common to all kind of wakes in electron rings. ctron rings.
Remembering that wakes can be represented by the real and imagi Remembering that wakes can be represented by the real and imaginary part nary part
of the coupling impedance, some common rules can be derived. of the coupling impedance, some common rules can be derived.
The The real (resistive) part real (resistive) part of the coupling of the coupling
impedance generates impedance generates asymmetric asymmetric
distortions distortions and and lengthening lengthening of the bunch of the bunch
distribution. The distribution. The bunch center of mass bunch center of mass
moves towards a different RF phase moves towards a different RF phase to to
compensate for the wake induced compensate for the wake induced
energy losses. energy losses.
The The imaginary (reactive) part imaginary (reactive) part of the of the
coupling impedance generates coupling impedance generates
symmetric distortions symmetric distortions of the bunch of the bunch
distribution. The bunch distribution. The bunch center of mass center of mass
does not move does not move (no energy losses). It (no energy losses). It
generates generates bunch lengthening or bunch lengthening or
shortening shortening. .
L
1
L
3
L
2
L
3
> L
2
> L
1
Z = j L
C
> 0
R
3
> R
2
> R
1
R
1
R
2
R
3
Z = R
21
Single Bunch Instabilities:
The SR Wake Case
The SR Wake Case
Collective Effects
F. Sannibale
Simulated instability showing Simulated instability showing
the the microbunching microbunching. .
Courtesy of Marco Courtesy of Marco Venturini Venturini
In an electron storage ring, if the current per In an electron storage ring, if the current per
bunch is above a specific threshold, the SR bunch is above a specific threshold, the SR
wake can drive a wake can drive a microbunching microbunching instability instability in in
the electron bunch. the electron bunch.
10.5mA
28.8mA
100ms 80 60 40 20 0
Time (msec)
40.0mA
10 10 mA mA
29 29 mA mA
40 40 mA mA
Time ( Time (msec msec) )
B
o
l
o
m
e
t
e
r

s
i
g
n
a
l

(
V
)
B
o
l
o
m
e
t
e
r

s
i
g
n
a
l

(
V
)
ALS Data ALS Data
( ) ( ) ] [ 2
0
3 1 2 3 3 2 1 2 1
0
Units MKS r C c e m A
q
=
( )
3 2
2 9
2 3 6 11
2 3
2 1
0
2 1
cos
1
s
C
s
RF
b
J
V f h
A I
>
The SR wake becomes strong enough to create The SR wake becomes strong enough to create
temporary micro temporary micro- -structures in the bunch that structures in the bunch that
radiates strong bursts of coherent radiates strong bursts of coherent
synchrotron radiation in the far synchrotron radiation in the far- -infrared. infrared.
22
Beam Break Up
Beam Break Up
Collective Effects
F. Sannibale
When a bunch enters off When a bunch enters off- -axis in a axis in a linac linac structure it excites transverse structure it excites transverse
wakes. wakes.
If the impedance associated with the wake is broad If the impedance associated with the wake is broad- -band, the head of the band, the head of the
bunch can excite the wakes that will deflect the tail of the bun bunch can excite the wakes that will deflect the tail of the bunch. ch.
The effect was first observed in 1966 at SLAC in the 2 miles lo The effect was first observed in 1966 at SLAC in the 2 miles long ng linac linac of the of the
SLC (Stanford Linear SLC (Stanford Linear Collider Collider) and was responsible for luminosity limitation. ) and was responsible for luminosity limitation.
In long high current/bunch In long high current/bunch linacs linacs the effect can build up and the bunch can the effect can build up and the bunch can
be distorted into a banana like shape. This effect is known as be distorted into a banana like shape. This effect is known as single single- -bunch bunch
beam break up beam break up (SBBU). (SBBU).
t
0
t
1
t
2
t
3
t
4
t
5
t
6
t
0
t
1
t
2
t
3
t
4
t
5
t
6
23
Horizontal size in
a dispersive region
Microwave Instability
Microwave Instability
Collective Effects
F. Sannibale
The total broad band impedance of a storage ring is also respon The total broad band impedance of a storage ring is also responsible of another sible of another
longitudinal single bunch instability known as the longitudinal single bunch instability known as the microwave instability microwave instability. .
When the current per bunch is larger than the instability thres When the current per bunch is larger than the instability threshold: hold:
The net effect on the bunch is an The net effect on the bunch is an increase of the energy spread increase of the energy spread above threshold above threshold
with a consequent with a consequent increase of the bunch length increase of the bunch length and of the beam and of the beam transverse size transverse size
in in dispersive dispersive regions. regions.
( )
n Z e
E E
I
E C
peak
//
2
0 0
2
>
the single particles get excited by the wakes on exponentially g the single particles get excited by the wakes on exponentially growing rowing
longitudinal oscillations. Because non longitudinal oscillations. Because non- -linearities linearities, the oscillation frequency , the oscillation frequency
changes with amplitude limiting the maximum amplitude and in mos changes with amplitude limiting the maximum amplitude and in most of the cases t of the cases
no particle loss happens. no particle loss happens.
24
Multi Bunch Instabilities
Multi Bunch Instabilities
Collective Effects
F. Sannibale
In the case of narrow In the case of narrow- -band impedances the wake generated by one bunch band impedances the wake generated by one bunch
can last long enough to interfere with other bunches or with the can last long enough to interfere with other bunches or with the bunch itself bunch itself
in subsequent turns. In this situation multi in subsequent turns. In this situation multi- -bunch instabilities can be excited. bunch instabilities can be excited.
High current accelerators are carefully designed in order to mi High current accelerators are carefully designed in order to minimize broad nimize broad
band and narrow band impedances. Anyway, even in the best concei band and narrow band impedances. Anyway, even in the best conceived ved
accelerator, the impedance accelerator, the impedance cannot vanish cannot vanish and there will be always a current and there will be always a current
threshold above which the beam will become unstable. threshold above which the beam will become unstable.
If the accelerator is required to operate above the instability If the accelerator is required to operate above the instability threshold, threshold, active active
feedback systems feedback systems are necessary for damping down the instabilities. are necessary for damping down the instabilities.
Despite these difficulties, properly designed accelerators with Despite these difficulties, properly designed accelerators with low overall low overall
broad broad- -band impedance, carefully damped band impedance, carefully damped HOMs HOMs and active longitudinal and and active longitudinal and
transverse bunch by bunch feedbacks achieved very remarkable res transverse bunch by bunch feedbacks achieved very remarkable results. ults.
Currents of few Amps have been stored in electron and positron m Currents of few Amps have been stored in electron and positron machines achines
(PEP 2, KEK (PEP 2, KEK- -B, DA B, DA NE, ) and of many tens of NE, ) and of many tens of mA mA in proton machines (SPS, in proton machines (SPS,
TEVATRON, HERA, ). TEVATRON, HERA, ).
25
Coupled Bunches Modes
Coupled Bunches Modes
Collective Effects
F. Sannibale
n = 1 n = 1
n = 2 n = 2
n = 3 n = 3
n = 4 n = 4
From Dan Russell's
Multiple DOF Systems
26
Multi
Multi
-
-
Bunch Instabilities
Bunch Instabilities
Mechanism
Mechanism
Collective Effects
F. Sannibale
( )
[ ] ( )
[ ] ( ) damping radiation
D n n
t
n n
t e t
D n
+ =
+

0
Im
Re sin
By using the model of coupled harmonic oscillators, every mode By using the model of coupled harmonic oscillators, every mode can be can be
characterized by a complex frequency characterized by a complex frequency and by the equation of a damped and by the equation of a damped
oscillator: oscillator:
The oscillation becomes unstable (anti The oscillation becomes unstable (anti- -damping) when: damping) when:
( ) always
D D
0 0 ] Im[ > < +
Wakes fields produce a shift of the imaginary part of the frequ Wakes fields produce a shift of the imaginary part of the frequency: ency:
( )
n
S
C
B n
Z
E
e
I
] Im[
Depending on the signs of the momentum compaction and of the Depending on the signs of the momentum compaction and of the
impedance, some modes can become unstable when the current per b impedance, some modes can become unstable when the current per bunch is unch is
increased. increased.
Feedback systems increase Feedback systems increase
D D
so that to increase the threshold for the so that to increase the threshold for the
instabilities. instabilities.
27
Good Wake Fields

Good Wake Fields
Collective Effects
F. Sannibale
Wakefield Wakefield- -based acceleration schemes. based acceleration schemes.
Strong R&D and very promising results. Strong R&D and very promising results.
Wake fields are commonly exploited in Wake fields are commonly exploited in
diagnostic systems used for the diagnostic systems used for the
characterization of the beam properties. characterization of the beam properties.
Bunches in electron storage rings with longitudinal Bunches in electron storage rings with longitudinal
distribution asymmetrically distorted by wake distribution asymmetrically distorted by wake- -fields emit fields emit
coherent synchrotron radiation at much higher coherent synchrotron radiation at much higher
frequencies than bunches with nominal frequencies than bunches with nominal gaussian gaussian
distribution. This can be exploited for designing far distribution. This can be exploited for designing far- -
infrared synchrotron light sources with revolutionary infrared synchrotron light sources with revolutionary
performances. performances.
Bad Wake God Wakes
Not all of these evanescent and ghostly wakes are bad in acce Not all of these evanescent and ghostly wakes are bad in accelerator lerator
applications. In fact, there are few examples were wakes play a applications. In fact, there are few examples were wakes play a positive role: positive role:
Laser
Gas
Gas jet nozzle
e
-
bunch
Plasma
channel
THz
Radiation
LOASIS LOASIS
28
References
References
Collective Effects
F. Sannibale
L. Palumbo, V. G. L. Palumbo, V. G. Vaccaro Vaccaro, M. , M. Zobov Zobov, Wake fields and impedances, CERN , Wake fields and impedances, CERN- -95 95- -06 06
A. A. Chao Chao, Physics of Collective Beam Instabilities in High Energy , Physics of Collective Beam Instabilities in High Energy
Accelerators, Wiley Accelerators, Wiley- -Interscience Interscience Pub. (1993). Pub. (1993).
A. A. Chao Chao, M. , M. Tigner Tigner, Handbook of Accelerator Physics and Engineering, , Handbook of Accelerator Physics and Engineering,
Word Scientific Pub. (1998). Word Scientific Pub. (1998).
29
Possible Homework
Possible Homework
Collective Effects
F. Sannibale
Calculate, the repulsive force that a proton with a 100 Calculate, the repulsive force that a proton with a 100 m displacement from m displacement from
the beam center will experience due to the space charge from the the beam center will experience due to the space charge from the other other
protons in the beam. The beam has a circular profile with protons in the beam. The beam has a circular profile with rms rms size of 2 mm size of 2 mm
and an energy of 2 and an energy of 2 GeV GeV. The linear charge density is of 0.7 . The linear charge density is of 0.7 nC/m nC/m. Estimate if . Estimate if
the effect on the particle integrated over one turn is significa the effect on the particle integrated over one turn is significant or not (the nt or not (the
ring length is 100 m). Compare with the case of an electron beam ring length is 100 m). Compare with the case of an electron beam with the with the
same characteristics. same characteristics.
In principle, a particle accelerator built in the space (orbiti In principle, a particle accelerator built in the space (orbiting around the ng around the
earth for example) could be built without a vacuum chamber. Will earth for example) could be built without a vacuum chamber. Will the the
particles in such an accelerator be subjected to any wake field? particles in such an accelerator be subjected to any wake field? Please Please
explain your answer. explain your answer.
1
Lecture No. 13
Lecture No. 13
Real Accelerators.
Real Accelerators.
Errors and Diagnostics.
Errors and Diagnostics.
2
Introduction
Introduction
Real Accelerators
Errors & Diagnostics
F. Sannibale
In designing and constructing an accelerator, physicists and en In designing and constructing an accelerator, physicists and engineers do gineers do
their best in making a perfect job and in foreseeing any possibl their best in making a perfect job and in foreseeing any possible operation e operation
mode for their device. mode for their device.
In this lecture, we will briefly introduce the more typical (an In this lecture, we will briefly introduce the more typical (and predictable) d predictable)
errors affecting real accelerators. We will also discuss with so errors affecting real accelerators. We will also discuss with some more me more
details examples (not a complete list) of diagnostic systems and details examples (not a complete list) of diagnostic systems and beam beam
measurements used for correcting for those errors. measurements used for correcting for those errors.
In most of the cases, the In most of the cases, the ideal machine ideal machine remains just a concept and one has remains just a concept and one has
to deal with more real objects where construction tolerances and to deal with more real objects where construction tolerances and unpredicted unpredicted
phenomena generate effects that need to be measured and correcte phenomena generate effects that need to be measured and corrected. d.
3
Magnet Misalignment Effects
Magnet Misalignment Effects
Real Accelerators
F. Sannibale
A A multipole multipole of order of order n n, with a , with a tilt error tilt error will will
present present a skew component of order a skew component of order n n with with
amplitude proportional to the tilt angle. amplitude proportional to the tilt angle.
A A multipole multipole of order of order n n, with a , with a displacement displacement
error error will present will present all the all the multipolar multipolar
components with order components with order i = 1, 2, , n i = 1, 2, , n - - 1 1. . x
y
x
4
Dipole Error
Dipole Error
Component Effect
Component Effect
Real Accelerators
F. Sannibale
If we have dipolar errors in If we have dipolar errors in N N different ring locations, they generate different ring locations, they generate N N kicks kicks

i i
and a total orbit distortion given by: and a total orbit distortion given by:
( )
( )
( )
( ) ( ) ( ) [ ] y x w s s s
s
s w
N
i
i w i i w
w
w
, cos
sin 2
1
= + =

=

Because of radiation damping, positrons and electrons converge Because of radiation damping, positrons and electrons converge into the into the
distorted orbit in roughly a damping time. Protons and heavier p distorted orbit in roughly a damping time. Protons and heavier particles articles
oscillates around the distorted closed orbit without converging oscillates around the distorted closed orbit without converging into it. into it.
In the case of a single kick at the position In the case of a single kick at the position s s, the displacement induced the , the displacement induced the
kick at the same point kick at the same point s s is given by: is given by:
( ) ( ) y x w s s w
w w
, cot
2
1
= =
If a If a corrector or steering magnet corrector or steering magnet (small dipole magnet capable of generating (small dipole magnet capable of generating
a kick a kick ) has a ) has a beam position monitor beam position monitor (BPM) nearby, by kicking the beam and (BPM) nearby, by kicking the beam and
using the previous relation, the beta function at that point can using the previous relation, the beta function at that point can be measured. be measured.
We saw that a displacement error in a magnet generates a We saw that a displacement error in a magnet generates a dipole dipole
component component in its center. This term induces a beam in its center. This term induces a beam orbit distortion. orbit distortion.
Betatron phases
Note that for integer tunes no closed orbit exists. Note that for integer tunes no closed orbit exists.
5
Orbit Correction Basics
Orbit Correction Basics
Real Accelerators
F. Sannibale
Or in matrix representation, when: Or in matrix representation, when:
( )
( ) ( ) [ ] N j s s s
s
u
M
i
i j i i
j
j
..., , 2 , 1 cos ) (
sin 2
) (
1
= + =

=

By using By using M M correctors magnets, we can find a set of kicks that cancels the correctors magnets, we can find a set of kicks that cancels the
displacement of the beam at the BPM positions. This is obtained displacement of the beam at the BPM positions. This is obtained when: when:
{ }
N N
u u u ..., , ,
2 1
= u
The kicks that need to be applied to the steering magnets for co The kicks that need to be applied to the steering magnets for correcting the rrecting the
closed orbit distortion, can be obtained by inverting the previo closed orbit distortion, can be obtained by inverting the previous equation: us equation:
By measuring the orbit distortion in By measuring the orbit distortion in N N BPMs BPMs along the ring, we find the set of along the ring, we find the set of
displacements: displacements:
( )
( ) ( ) [ ]

+ = =
i j
i j
i j M N
s s
s s
M with cos
sin 2
) ( ) (
M u
N M
u M
1
=
The elements of the The elements of the response matrix response matrix M M, can be calculated from the machine , can be calculated from the machine
model, or measured by individually exciting each of the correcto model, or measured by individually exciting each of the correctors and rs and
measuring the induced displacement in each of the measuring the induced displacement in each of the BPMs BPMs. .
6
Quadrupole
Quadrupole
Error
Error
Component Effect
Component Effect
Real Accelerators
F. Sannibale
Quadrupole Quadrupole error components (gradient errors) can be due to misalignment error components (gradient errors) can be due to misalignment
of higher order of higher order multipolar multipolar magnet ( magnet (sextupoles sextupoles, , octupoles octupoles, ) or due to error , ) or due to error
in the current in the current- -strength calibration of strength calibration of quadrupole quadrupole magnets. magnets.
strength quadrupole
p
Gq
k where L k
Q Q
= =
0
2
Gradient errors generate a Gradient errors generate a betatron betatron tune shift tune shift equal to: equal to:
L L is the is the quadrupole quadrupole magnetic length, magnetic length, G G is its gradient and is its gradient and q q and and p p
0 0
are the are the
particle charge and momentum respectively. The previous equation particle charge and momentum respectively. The previous equation can be can be
used for measuring the beta function at the used for measuring the beta function at the quadrupole quadrupole position, when the position, when the
tune shift for small change of the magnet strength is measured. tune shift for small change of the magnet strength is measured.
It can be shown that It can be shown that quadrupole quadrupole gradient errors makes the gradient errors makes the half half- -integer integer
resonance unstable resonance unstable. .
Tilt errors in Tilt errors in quadrupole quadrupole magnets generate magnets generate coupling coupling between the vertical and between the vertical and
the horizontal planes. On the other hand, on purpose tilted the horizontal planes. On the other hand, on purpose tilted quadrupoles quadrupoles
( (skew skew quadrupoles quadrupoles) can be used for compensating the coupling due to lattice ) can be used for compensating the coupling due to lattice
non non- -linearities linearities. .
7
ALS ALS
Non
Non
-
-
ideal Magnets:
ideal Magnets:
Multipolar
Multipolar
Terms
Terms
Real Accelerators
F. Sannibale
Good design and construction can minimize but not Good design and construction can minimize but not
cancel the cancel the multipolar multipolar field presence. Additionally, in most field presence. Additionally, in most
of storage rings of storage rings sextupole sextupole (and sometimes (and sometimes octupole octupole) )
magnets are added on purpose for the compensation of magnets are added on purpose for the compensation of
chromatic effects and for improving the dynamic aperture. chromatic effects and for improving the dynamic aperture.
Multipolar Multipolar field components field components
introduce non introduce non- -linearities linearities that that
generate a shift in the generate a shift in the betatron betatron
frequency for large amplitude frequency for large amplitude
oscillations ( oscillations (tune shift on amplitude tune shift on amplitude). ).
These tune shifts can bring particles on tune These tune shifts can bring particles on tune resonances resonances
generating particle losses ( generating particle losses (dynamic aperture dynamic aperture). ).
Simplified geometries, imperfections and mechanical tolerances Simplified geometries, imperfections and mechanical tolerances in the in the
design and construction of accelerator magnets, populates the ac design and construction of accelerator magnets, populates the accelerators celerators
with a plethora of with a plethora of higher order higher order multipolar multipolar terms terms. .
On the other hand, these frequency shifts generate On the other hand, these frequency shifts generate de de- -coherence coherence in the in the
oscillations with a damping effect on instabilities ( oscillations with a damping effect on instabilities (Landau damping Landau damping). ).
8
SOLEIL SOLEIL
Power Supply Fluctuations
Power Supply Fluctuations
Real Accelerators
F. Sannibale
Fluctuations in the power supply current Fluctuations in the power supply current of the accelerator magnets can of the accelerator magnets can
limit the performance of an accelerator. limit the performance of an accelerator.
Jitter in the dipole magnet power supplies generates fluctuatio Jitter in the dipole magnet power supplies generates fluctuations in the ns in the
beam energy inducing jitter in the tunes and orbit fluctuations beam energy inducing jitter in the tunes and orbit fluctuations as well. as well.
Jitter on Jitter on quadrupole quadrupole magnet power supplies generate magnet power supplies generate betatron betatron tune tune
fluctuations that can bring particles on tune fluctuations that can bring particles on tune resonances resonances and generate particle and generate particle
losses. losses.
Any power supply fluctuation will be transferred to beam (ampli Any power supply fluctuation will be transferred to beam (amplified in the fied in the
case of strong focusing machines) affecting the ultimate perform case of strong focusing machines) affecting the ultimate performance of the ance of the
accelerators. accelerators.
Power supply stability requirements Power supply stability requirements
strongly depend on which part of the strongly depend on which part of the
accelerator the magnet is located. accelerator the magnet is located.
Typical relative stability requirements range Typical relative stability requirements range
from few units of 10 from few units of 10
- -3 3
for beam transfer for beam transfer- -lines lines
power supplies to about 10 power supplies to about 10
- -5 5
for the case of for the case of
storage ring power supplies. storage ring power supplies.
9
Other Perturbations
Other Perturbations
Real Accelerators
F. Sannibale
Several other perturbations can affect the proper operation of Several other perturbations can affect the proper operation of an an
accelerator. accelerator.
Last but not least, accelerators are designed for specific appl Last but not least, accelerators are designed for specific applications that ications that
often require detectors using high magnetic fields. often require detectors using high magnetic fields.
This is the typical case for high energy physics experiments in This is the typical case for high energy physics experiments in colliders colliders, or of , or of
insertion devices for radiation production in light sources. The insertion devices for radiation production in light sources. These fields if not se fields if not
compensated can have a strong impact on the accelerator performa compensated can have a strong impact on the accelerator performance. nce.
Very large accelerators are sensitive to the earth magnetic fie Very large accelerators are sensitive to the earth magnetic field, to the moon ld, to the moon
phases, to neighbor railway stations, phases, to neighbor railway stations,
All accelerators are sensitive to environmental fields and vari All accelerators are sensitive to environmental fields and variables: stray ables: stray
magnetic fields due to equipment or to high power electric cable magnetic fields due to equipment or to high power electric cables, presence s, presence
of other accelerators, temperature variations, fluctuations of t of other accelerators, temperature variations, fluctuations of the main AC he main AC
power, ground motion, vibrations, power, ground motion, vibrations,
In order to minimize and compensate for the effects due to all In order to minimize and compensate for the effects due to all these these
perturbations and errors, an efficient beam diagnostics system n perturbations and errors, an efficient beam diagnostics system need to be eed to be
used. used.
10
Electromagnetic
Electromagnetic
Beam Position Monitors
Real Accelerators
F. Sannibale
11
a
y
A
V
B
V
Electromagnetic
Electromagnetic
Real Accelerators
F. Sannibale
In electromagnetic In electromagnetic BPMs BPMs, the image charge in an , the image charge in an
electrode is proportional to the beam current and to electrode is proportional to the beam current and to
the angle included between the beam and the the angle included between the beam and the
electrode extremes: electrode extremes:
I G V
I G V
B
A
=
=
) ( ) ( y a y a w But + = =
y a
w
I G V
y a
w
I G V
B
A
+
=

=
y I
y a
w G
V V
B A

=
2 2
2

a
y
V V
V V
B A
B A

=
+
B A
B A
V V
V V a
y
+

2
a I
y a
w G
V V
B A
2 2
2

= +
In addition to this geometric In addition to this geometric

effect, the field lines tend to effect, the field lines tend to
cluster closely in the region of cluster closely in the region of
the nearest electrode (the the nearest electrode (the E E field field
must be perpendicular to the must be perpendicular to the
walls). For this geometry, this walls). For this geometry, this
gives an additional factor two: gives an additional factor two:
12
Button Type
Button Type
BPMs
BPMs
Real Accelerators
F. Sannibale
! boh b a = +
SOR SOR- -Ring Ring
D C B A
D C B A
D C B A
D B C A
V V V V
V V V V
K y
V V V V
V V V V
K x
+ + +
+ +
=
+ + +
+ +
=
) ( ) (
,
) ( ) (
APS APS
A
V
B
V
C
V
D
V
Capacitive type (derivative response), low coupling impedance, Capacitive type (derivative response), low coupling impedance, relatively relatively
low sensitivity, best for storage rings. low sensitivity, best for storage rings.
PEP II PEP II
DELTA DELTA
Typical Typical
geometry used geometry used
in the presence in the presence
of synchrotron of synchrotron
radiation. radiation.
13
Stripline
Stripline
Electrode
Electrode
Real Accelerators
F. Sannibale
Transmission line type, relatively high beam impedance, high se Transmission line type, relatively high beam impedance, high sensitivity, nsitivity,
directionality capability, best for directionality capability, best for linacs linacs and and transferlines transferlines. .
FNAL FNAL
Injector Injector
c L 2
0 2 0 1
0 Z or R and Z R Usually
L L
= =
1 L
R
2 L
R
0
Z
Out
V
L
1 L
R
2 L
R
0
Z
Out
V
L
c L t 2 =
1 L
R
2 L
R
0
Z
Out
V
L
c L t =
Wall
I
Blue color: inverted polarity pulse Blue color: inverted polarity pulse
No signal on No signal on R R
L L2 2
! !
1 L
R
2 L
R
0
Z
Wall
I
Out
V
L
0 = t
Vacuum
Chamber
Stripline
Electrode
14
Stripline
Stripline
BPM
BPM
Real Accelerators
F. Sannibale
Stripline Stripline structures are also widely used as the kicker in transverse a structures are also widely used as the kicker in transverse and nd
longitudinal feedback systems. longitudinal feedback systems.
SLAC SLAC
LCLS LCLS
Out
V
Out
V
FNAL Injector FNAL Injector
HERA HERA Stripline Stripline BPM BPM
SPRING 8 Kicker SPRING 8 Kicker
15
Other
Other
BPMs
BPMs
Real Accelerators
F. Sannibale
Photon Photon - - BPM BPM
y
e-
e-
SR
Beam
Blades
FMB FMB
BESSY II, BESSY II,
ALS, ALS,
SLS, SLS,
LNLS LNLS
The intensity of the modes in the resonant The intensity of the modes in the resonant
structure is proportional to the beam offset structure is proportional to the beam offset
In In resonant resonant BPMs BPMs the the
beam excites modes in beam excites modes in
resonant structures resonant structures
TTF BPM TTF BPM
16
SNS SNS
Beam Profile Monitors:
Beam Profile Monitors:
Wire Scanners
Wire Scanners
Real Accelerators
F. Sannibale
A moveable wire scans the beam transversally. A moveable wire scans the beam transversally.
Oxford Oxford- -Danfisik Danfisik
KEK KEK
The interaction between the beam The interaction between the beam
and the wire generates (by ionization, and the wire generates (by ionization,
bremsstrahlung bremsstrahlung, atomic excitation, ) , atomic excitation, )
a shower of secondary emission a shower of secondary emission
particles proportional to the number particles proportional to the number
of beam particles hitting the wire. of beam particles hitting the wire.
The secondary particles (mainly electrons The secondary particles (mainly electrons
and photons) are detected and the beam and photons) are detected and the beam
transverse profile can be reconstructed. transverse profile can be reconstructed.
The wire material can be a metal, carbon, or The wire material can be a metal, carbon, or a laser beam (Compton scattering, a laser beam (Compton scattering,
neutralization) neutralization)
35
30
25
20
15
10
5
0
W
i
r
e

s
i
g
n
a
l

(
m
V
)
-20 -10 0 10 20
Actuator position (mm)
600
500
400
300
200
100
0
L
a
s
e
r

n
o
t
c
h

(
m
i
c
r
o
V
)
BNL BNL- -SNS SNS
H H
- -
beam beam
profile profile
17
Virtual Photons
Virtual Photons
Real Accelerators
F. Sannibale
In the Particle Sources lecture, we already saw that accordin In the Particle Sources lecture, we already saw that according to quantum g to quantum
field theory, a photon with a large enough energy can oscillate field theory, a photon with a large enough energy can oscillate between the between the
states of virtual electron states of virtual electron- -positron pair and of real photon. positron pair and of real photon.
e
The opposite is also true. An electron moving in The opposite is also true. An electron moving in
the free space can be considered as surrounded the free space can be considered as surrounded
by a cloud of by a cloud of virtual photons virtual photons that appear and that appear and
disappear and that indissolubly travel with it. disappear and that indissolubly travel with it.
Nevertheless, in particular situations, the electron can receiv Nevertheless, in particular situations, the electron can receive a kick that e a kick that
separates it from the photons that become real. separates it from the photons that become real.
when the electron moves on a curved trajectory, the transverse when the electron moves on a curved trajectory, the transverse
acceleration induces the separation. This is the case of acceleration induces the separation. This is the case of
synchrotron radiation synchrotron radiation. .
when a relativistic electron moves inside a media and the speed when a relativistic electron moves inside a media and the speed
of light in the media is smaller than the particle velocity, the of light in the media is smaller than the particle velocity, then the n the
separation can happen. This is the case of the separation can happen. This is the case of the Cerenkov Cerenkov radiation. radiation.
when a relativistic electron moves inside a non when a relativistic electron moves inside a non- -homogeneous homogeneous
media, then the separation can happen. This is the case of the media, then the separation can happen. This is the case of the
transition (diffraction) radiation transition (diffraction) radiation. .
18
Photon Based
Photon Based
Beam Profile Monitors
Beam Profile Monitors
Real Accelerators
F. Sannibale
Photon diagnostics exploiting all the described emission mechan Photon diagnostics exploiting all the described emission mechanisms are isms are
widely used for measuring the transverse and longitudinal profil widely used for measuring the transverse and longitudinal profiles of es of
relativistic beams. relativistic beams.
A. Lumpkin A. Lumpkin
In fact, the spatial distribution of the photons reproduces exa In fact, the spatial distribution of the photons reproduces exactly the ctly the
particle distribution of the beam and can be conveniently used f particle distribution of the beam and can be conveniently used for the or the
characterization of the beam. characterization of the beam.
Monitors exploiting transition and Monitors exploiting transition and Cerenkov Cerenkov radiation are relatively invasive radiation are relatively invasive
and are mainly used in single pass or few and are mainly used in single pass or few- -turns accelerators. turns accelerators.
The angular distribution of the photons depends on several beam The angular distribution of the photons depends on several beam
parameters. This fact can be exploited for the measurements of q parameters. This fact can be exploited for the measurements of quantities uantities
other than the beam distribution as well. other than the beam distribution as well.
19
Synchrotron Radiation
Monitors
Monitors
Real Accelerators
F. Sannibale
Synchrotron radiation, very abundant in electron and positron a Synchrotron radiation, very abundant in electron and positron accelerators and ccelerators and
present in very high energy proton storage rings, is widely used present in very high energy proton storage rings, is widely used for transverse for transverse
and longitudinal beam profile measurements. and longitudinal beam profile measurements.
One of the appealing features of such One of the appealing features of such
diagnostic systems is that they are non diagnostic systems is that they are non- -
invasive. invasive.
ALS ALS
The resolution of these monitors are The resolution of these monitors are
limited by the geometry of the system and limited by the geometry of the system and
by the radiation diffraction. by the radiation diffraction.
The geometric limitation requires The geometric limitation requires
small aperture systems while the small aperture systems while the
diffraction term requires large diffraction term requires large
apertures and shorter photon apertures and shorter photon
wavelengths. Tradeoff solutions must wavelengths. Tradeoff solutions must
be adopted. be adopted.
Typical resolutions in electron Typical resolutions in electron
storage rings using hard x storage rings using hard x- -ray ray
photons range between few and photons range between few and
tens of microns. tens of microns.
20
Other Beam Profile Monitors
Other Beam Profile Monitors
Real Accelerators
F. Sannibale
The simplest beam profile monitor is probably the one using The simplest beam profile monitor is probably the one using fluorescent fluorescent
screens screens intercepting the beam. intercepting the beam.
The beam particles hitting the screen material excite the atoms The beam particles hitting the screen material excite the atoms that that
subsequently radiate a photon in the visible range when decaying subsequently radiate a photon in the visible range when decaying back to the back to the
ground state. ground state.
The resulting image of the beam on the screen is then viewed by The resulting image of the beam on the screen is then viewed by a a ccd ccd
camera and eventually digitized by a frame grabber for further a camera and eventually digitized by a frame grabber for further analysis. nalysis.
Such monitors are destructive and typically are used only in be Such monitors are destructive and typically are used only in beam am
transferlines transferlines. .
Another category of beam profile monitors are the Another category of beam profile monitors are the ionization chambers ionization chambers. .
In this monitor, a gas in a dedicated portion of the vacuum cha In this monitor, a gas in a dedicated portion of the vacuum chamber is mber is
ionized by the passage of the beam. ionized by the passage of the beam.
Depending on the scheme used, either the electrons or the ionize Depending on the scheme used, either the electrons or the ionized atoms can d atoms can
be detected for the beam profile reconstruction. Time of flight be detected for the beam profile reconstruction. Time of flight analysis of the analysis of the
ionized particles are usually necessary. ionized particles are usually necessary.
Because of their Because of their perturbative perturbative nature, these monitors are mainly used in nature, these monitors are mainly used in
single pass accelerators. single pass accelerators.
21
Image from
Max Planck Institute web site
Photocathode
STREAK CAMERA Scheme STREAK CAMERA Scheme
Photon
Photon
-
-
Based
Based
Longitudinal Profile Monitors
Longitudinal Profile Monitors
Real Accelerators
F. Sannibale
In photon In photon- -based longitudinal beam profile monitors, detectors such as based longitudinal beam profile monitors, detectors such as
streak cameras streak cameras, , fast photodiodes and fast photodiodes and photomultipliers photomultipliers are used. are used.
In the streak camera case, time resolution of several hundreds o In the streak camera case, time resolution of several hundreds of f fs fs can be can be
achieved. achieved.
APS APS
LEP LEP
ESRF ESRF
Streak cameras with an additional couple of Streak cameras with an additional couple of
sweeping electrodes (orthogonal to the other sweeping electrodes (orthogonal to the other
one) have single bunch one) have single bunch- -single turn capabilities single turn capabilities
and can be used for the characterization of and can be used for the characterization of
single and single and multibunch multibunch intabililities intabililities. .
22
More on Longitudinal
More on Longitudinal
Profile Monitors
Profile Monitors
Real Accelerators
F. Sannibale
Femtoseconds Femtoseconds resolution resolution
(or even smaller) can be (or even smaller) can be
achieved by achieved by interferometric interferometric
techniques involving techniques involving
coherent light in the Far coherent light in the Far- -IR IR
(coherent synchrotron (coherent synchrotron
radiation, coherent transition radiation, coherent transition
radiation, ) or by electro radiation, ) or by electro- -
optic techniques using non optic techniques using non- -
linear crystals and laser linear crystals and laser
probing. probing.
For relatively long bunches ~ 100 For relatively long bunches ~ 100 ps ps or longer or longer
electromagnetic pickups can be efficiently electromagnetic pickups can be efficiently
used. used.
DA DA NE NE
Accumulator Accumulator
In this example, the beam inside the DA In this example, the beam inside the DA NE NE
Accumulator (~ 150 Accumulator (~ 150 ps ps rms rms) is measured by ) is measured by
using the signal from a using the signal from a stripline stripline. .
LOASIS LOASIS - - LBL LBL
LOASIS LOASIS - - LBL LBL
23
Current Monitors: Faraday Cup
Current Monitors: Faraday Cup
and Current Transformers
and Current Transformers
Real Accelerators
F. Sannibale
Current transformers Current transformers are used for measuring the AC component of the are used for measuring the AC component of the
beam current: beam current:
L
R
for I
N
R
I
R L i
R L i
N
R
V
beam beam OUT
>>
+
=
1
beam
I
N
OUT
V
R
BERGOZ BERGOZ
Conceptually, the Conceptually, the Faraday cup Faraday cup is the simplest among the current monitors. is the simplest among the current monitors.
Electrostatic fields with the proper sign can be Electrostatic fields with the proper sign can be
added in order to avoid that primary and added in order to avoid that primary and
secondary (emission) charged particles can secondary (emission) charged particles can
leave the cup affecting the measurement. leave the cup affecting the measurement.
For short bunches, if the shape of the bunch needs to be measur For short bunches, if the shape of the bunch needs to be measured as well, ed as well,
the FC has to be designed as a transmission line in order to pre the FC has to be designed as a transmission line in order to present a good sent a good
high frequency response. high frequency response.
beam
I
OUT
V
core
ty permeabili High
Out
V
Cup electrode
24
Beam Current Monitors:
The DC Current Transformer
The DC Current Transformer
Real Accelerators
F. Sannibale
For measuring the average beam current (DC component), the For measuring the average beam current (DC component), the parametric parametric
current transformer current transformer or or DC current transformer DC current transformer (DCCT) is used: (DCCT) is used:
The DCCT uses two high The DCCT uses two high
permeability cores driven to permeability cores driven to
saturation by a low frequency saturation by a low frequency
current modulation. current modulation.
Demodulator
beam
I
Out
V
. 2 har
nd
Hz 200 ~
The signals from two secondary coils The signals from two secondary coils
of the cores are mutually subtracted. of the cores are mutually subtracted.
Because of the non Because of the non- -linear linear
magnetization curve of the core magnetization curve of the core
material, this difference signal is zero material, this difference signal is zero
only when the beam current is zero. only when the beam current is zero.
In the presence of beam current this difference signal is non In the presence of beam current this difference signal is non- -zero and in zero and in
particular shows a second harmonic component. particular shows a second harmonic component.
A current proportional to the amplitude of this component is fe A current proportional to the amplitude of this component is fed back into a d back into a
third coil in order to compensate for the beam current and to ma third coil in order to compensate for the beam current and to make the ke the
difference signal zero. difference signal zero.
At equilibrium, the current flowing in this third coil is equal At equilibrium, the current flowing in this third coil is equal in amplitude to in amplitude to
the beam current but opposite in sign. the beam current but opposite in sign.
25
DA DA NE NE- -LINAC LINAC
Z Z = = R R
Wall Current Monitors.
Wall Current Monitors.
Real Accelerators
F. Sannibale
The band The band- -width of such monitors is limited to width of such monitors is limited to
few GHz. few GHz.
Additionally, in the described configuration Additionally, in the described configuration
they can radiate and/or pick they can radiate and/or pick- -up high frequency up high frequency
electromagnetic noise. electromagnetic noise.
OUT
V
beam
I
Z
Chamber Vacuum
beam OUT
I Z V =
FNAL FNAL
Electrons Electrons
Positrons Positrons
For limiting such a noise, a metallic For limiting such a noise, a metallic
shield loaded with ferrites (inductive shield loaded with ferrites (inductive
loading) can be used. loading) can be used.
Ferrites
Shield Metallic
26
Beam Spectrum
Beam Spectrum
Real Accelerators
F. Sannibale
In the presence of In the presence of betatron betatron and/or synchrotron oscillations, sidebands and/or synchrotron oscillations, sidebands
around each of the revolution harmonics will appear at the frequ around each of the revolution harmonics will appear at the frequencies: encies:
By detecting the passage of a By detecting the passage of a
particle at a fixed particle at a fixed azimuthal azimuthal
position the following time domain position the following time domain
signal can be observed. signal can be observed.
By Fourier transforming or by By Fourier transforming or by
using a spectrum analyzer, the using a spectrum analyzer, the
same signal in the frequency same signal in the frequency
domain will appear as: domain will appear as:
0
3T
0
4T
0
0
T
0
2T
t
I
L
v
T
f = =
0
0
1
Revolution harmonics Revolution harmonics
0
3 f
0
4 f
0
0
f
0
2 f
f
df
dP
s y x w Q of part fractional nf f
w S
, ,
0
= =
In the case of a multi In the case of a multi- -particle beam, because of the non particle beam, because of the non- -zero momentum zero momentum
spread and machine non spread and machine non- -linearities linearities, the particles have slightly different , the particles have slightly different
oscillation frequencies. As a consequence the spectral lines wil oscillation frequencies. As a consequence the spectral lines will show a finite l show a finite
thickness. thickness.
Instability Instability
sidebands sidebands
27
Schottky
Schottky
Noise Monitors
Noise Monitors
Real Accelerators
F. Sannibale
Schottky Schottky noise cannot be used in noise cannot be used in
electron and positron machines electron and positron machines
because in those accelerators, the because in those accelerators, the
noise due to synchrotron radiation noise due to synchrotron radiation
quantum fluctuations is strong and quantum fluctuations is strong and
covers the covers the Schottky Schottky noise. noise.
FNAL FNAL
By using resonant electromagnetic pick By using resonant electromagnetic pick- -ups ups
(cavity or (cavity or waveguide waveguide) the signal resulting from ) the signal resulting from
the motion of all the particles can be detected. the motion of all the particles can be detected.
In fact, In fact, Schottky Schottky noise monitors are actually the main non noise monitors are actually the main non- -invasive invasive
diagnostic tool used in heavy particle storage rings. Quantities diagnostic tool used in heavy particle storage rings. Quantities that can be that can be
measured include longitudinal and transverse tunes, momentum spr measured include longitudinal and transverse tunes, momentum spread and ead and
beam current. beam current.
Because the motion of the particles is essentially Because the motion of the particles is essentially
independent, such a signal appears as a noise and independent, such a signal appears as a noise and
it is usually referred as the it is usually referred as the Schottky Schottky noise noise (SN). (SN).
SN find applications in beam diagnostics SN find applications in beam diagnostics
FNAL FNAL
Synchrotron
sidebands
Revolution
harmonic
28
Tune Measurement
Tune Measurement
Real Accelerators
F. Sannibale
In electron and positron machines, in order to measure the In electron and positron machines, in order to measure the betatron betatron tunes tunes
in the absence of instabilities, coherent beam oscillations need in the absence of instabilities, coherent beam oscillations need to be excited. to be excited.
Synchrotron tune can be measured by modulating the RF phase or Synchrotron tune can be measured by modulating the RF phase or
amplitude and by measuring the induced sidebands using the sum s amplitude and by measuring the induced sidebands using the sum signal ignal
from a pick from a pick- -up.The up.The same detection part of the same detection part of the betatron betatron tune measurement tune measurement
system can be used. system can be used.
DA DA NE NE
Horizontal
Vertical
0
KICKER
BAND-PASS
FILTER
0
0 0
BEAM
POSITION
MONITOR
AMPLITUDE
A
PHASE
NETWORK
ANALYZER
HP 4195 A
RATIO B/A
RF OUT B
100 W
CLASS A
100 W
CLASS A
29
Beam Characterization:
Beam Characterization:
Chromaticity Measurement
Chromaticity Measurement
Real Accelerators
F. Sannibale
Chromaticity is measured in storage rings by changing the moment Chromaticity is measured in storage rings by changing the momentum um
(energy) of the beam and by recording the induced tune variation (energy) of the beam and by recording the induced tune variations. s.
The beam momentum is usually changed by varying the RF frequency The beam momentum is usually changed by varying the RF frequency. .
In this way, the revolution period is modified and the particles In this way, the revolution period is modified and the particles are forced into are forced into
trajectories with different curvature in the dipole magnets. Thi trajectories with different curvature in the dipole magnets. This can happen s can happen
only if the particles change their momentum. only if the particles change their momentum.
RF
RF
C C C C
f
f
f
f
T
T
L
L
p
p
=

1 1 1 1
0
0
0 0 0
y x w
p p
Q
w
w
,
0
=
=
RHIC RHIC
Fits
The experimental The experimental
data are fitted by a data are fitted by a
polynomial function. polynomial function.
The fitting function The fitting function
calculated at the calculated at the
nominal momentum nominal momentum
gives the linear gives the linear
chromaticity. chromaticity.
30
Orbit Measurements
Orbit Measurements
Real Accelerators
F. Sannibale
In most of acceleration applications, the beam orbit needs to b In most of acceleration applications, the beam orbit needs to be very stable. e very stable.
In In colliders colliders, counter , counter- -rotating beams with transverse size in the nanometer scale rotating beams with transverse size in the nanometer scale
need to overlap for collision, while in light sources orbit stab need to overlap for collision, while in light sources orbit stability requirements are ility requirements are
often on the order of a micron. often on the order of a micron.
We saw how accelerator imperfections can generate orbit distort We saw how accelerator imperfections can generate orbit distortions. ions.
Such orbits need to be carefully Such orbits need to be carefully measured measured and and corrected corrected. .
In addition, orbit feedbacks are often used to ensure the requir In addition, orbit feedbacks are often used to ensure the required stability. ed stability.
In circular machines, the transverse beam trajectory can be app In circular machines, the transverse beam trajectory can be approximated by a roximated by a
sinusoid oscillating at the sinusoid oscillating at the betatron betatron frequency. frequency.
Nyquist Nyquist theorem states that we need to sample the orbit in a number of theorem states that we need to sample the orbit in a number of positions positions
at least twice the at least twice the betatron betatron tune number. With some contingency, at least four tune number. With some contingency, at least four
BPMs BPMs per 2 per 2 betatron betatron phase advance are used in circular and linear accelerators. phase advance are used in circular and linear accelerators.
Absolute orbit measurements suffer of Absolute orbit measurements suffer of
accuracy limitations. In fact, the actual accuracy limitations. In fact, the actual
center of magnets and center of magnets and BPMs BPMs is not exactly is not exactly
known. known.
Measured closed orbits are often referred Measured closed orbits are often referred
to a golden orbit, which is usually to a golden orbit, which is usually
obtained by the obtained by the beam beam- -based based alignment of alignment of
the beam to the center of the the beam to the center of the quadrupoles quadrupoles. .
NSRL NSRL - - Hefei Hefei
31
Momentum and Momentum
Momentum and Momentum
Spread Measurement
Spread Measurement
Real Accelerators
F. Sannibale
In In linacs linacs and and transferlines transferlines the momentum and momentum the momentum and momentum
spread are mainly measured by spectrometer systems. spread are mainly measured by spectrometer systems.
The beam enters in the field of a dipole magnet The beam enters in the field of a dipole magnet
where particles with different where particles with different momenta momenta follows follows
different trajectories. different trajectories.
0
p p
BLUE
<
0
p p
RED
>
Detector
Bend
The particle position is then measured on a detector The particle position is then measured on a detector
downstream the magnet. downstream the magnet.
The spectrometer resolution is limited by the intrinsic beam si The spectrometer resolution is limited by the intrinsic beam size at the ze at the
detector plane and by field non detector plane and by field non- -linearities linearities. .
DA DA NE NE
Secondary emission
hodoscope
Spectrometer
control window
DA DA NE NE- -Linac Linac
32
Momentum Spread &
Momentum Spread &
Emittance
Emittance
Measurement in Rings
Measurement in Rings
Real Accelerators
F. Sannibale
In electron and positron storage rings the equilibrium beam In electron and positron storage rings the equilibrium beam emittance emittance and and
the momentum spread can be measured by the combined measurement the momentum spread can be measured by the combined measurement of at of at
least two transverse beam profiles at two different ring locatio least two transverse beam profiles at two different ring locations. ns.
2 , 1
1
1
2 / 1
2
=
+
+
= index system i
p
x
p
xi xi i rms

2
1 2
2
2 1
2
1
2
2
2
2
2
1
1
x x x x
x rms x rms
x
x x

=
+
=
2
1 2
2
2 1
2
2
1 1
2
2
2
x x x x
x rms x rms
p
x x
p

: emittance
p: momentum
: beta function
: dispersion
: emittance
ratio
x
rms
: rms beam
size
The beam size at a particular The beam size at a particular
azhimutal azhimutal position is given by: position is given by:
If the beam size is measured in two different points of the rin If the beam size is measured in two different points of the ring and the g and the
optical functions at such points are known, then: optical functions at such points are known, then:
33
Emittance
Emittance
Measurement
Measurement
In
In
Linacs
Linacs
and
and
Tranferlines
Tranferlines
Real Accelerators
F. Sannibale
A popular technique for measuring the A popular technique for measuring the emittance emittance in in linacs linacs or or transferlines transferlines
uses the so uses the so- -called called three gradient method three gradient method. .
The gradient (focusing The gradient (focusing- -defocusing strength) defocusing strength)
of a of a quadrupole quadrupole is changed and the related is changed and the related
transverse beam profiles are recorded by a transverse beam profiles are recorded by a
detector downstream the detector downstream the quadrupole quadrupole. .
Quadrupole
Beam profile
monitor
The measurement requires a minimum of 3 different The measurement requires a minimum of 3 different quadrupole quadrupole gradients but gradients but
the accuracy can be improved if more points are taken. the accuracy can be improved if more points are taken.
The beam size at the detector is defined by the The beam size at the detector is defined by the
beam beam emittance emittance and by the local beta function. and by the local beta function.
The The emittance emittance is an invariant while the beta is an invariant while the beta
changes with the changing changes with the changing quadrupole quadrupole gradient. gradient.
An analytical expression linking the transverse An analytical expression linking the transverse
profiles with the beam profiles with the beam emittance emittance can be derived can be derived
and used for fitting the experimental data. and used for fitting the experimental data.
From the fit, the values for the From the fit, the values for the emittance emittance and and
for the optical functions at the for the optical functions at the quadrupole quadrupole
position can be finally extracted. position can be finally extracted.
DA DA NE NE
LINAC LINAC
34
More Measurements
More Measurements
Real Accelerators
F. Sannibale
Lifetime measurements Lifetime measurements: by current monitors, by beam loss monitors, by : by current monitors, by beam loss monitors, by
modifying beam parameters in order to discriminate among differe modifying beam parameters in order to discriminate among different nt
contributions, contributions,
Coupling impedance measurements Coupling impedance measurements: by characterizing the tune shift on : by characterizing the tune shift on
current, by measuring instability thresholds, by energy spread current, by measuring instability thresholds, by energy spread
measurements, measurements,
Optical function measurements Optical function measurements: by single : by single quadrupole quadrupole gradient perturbation, gradient perturbation,
by phase advance between position monitors, by response matrix, by phase advance between position monitors, by response matrix, by energy by energy
momentum variation for dispersion function measurement, momentum variation for dispersion function measurement,
Momentum acceptance measurements Momentum acceptance measurements: by changing the particle momentum : by changing the particle momentum
in combination with lifetime measurements, by modifying accelera in combination with lifetime measurements, by modifying accelerator tor
parameters for discriminating among different contributions, parameters for discriminating among different contributions,
Non Non- -linearities linearities and dynamic aperture measurements and dynamic aperture measurements: by kicking the beam : by kicking the beam
transversely and characterizing the tune shift on amplitude, by transversely and characterizing the tune shift on amplitude, by frequency frequency
map analysis, by de map analysis, by de- -coherence measurements, coherence measurements,
Transverse coupling measurements Transverse coupling measurements: by transverse beam profile monitors, : by transverse beam profile monitors,
by response matrix, by closest tune approach, by response matrix, by closest tune approach,

35
References
References
Real Accelerators
F. Sannibale
M. M. Serio Serio, , Diagnostica Diagnostica e e misure misure, Seminars on DA , Seminars on DA NE, February 2000. NE, February 2000.
M. M. Zolotorev Zolotorev, Radiation and Acceleration, tutorial, LBL 2005. , Radiation and Acceleration, tutorial, LBL 2005.
R. R. Littauer Littauer, Beam Instrumentation, Proc. Physics of High Energy Particle , Beam Instrumentation, Proc. Physics of High Energy Particle
Accelerators (Stanford, CA, 1982); AIP Conf. Accelerators (Stanford, CA, 1982); AIP Conf. Porc Porc. 105 (1982) 869. . 105 (1982) 869.
M. M. Minty Minty, Diagnostics, CAS Synchrotron Radiation and Free , Diagnostics, CAS Synchrotron Radiation and Free- -Electron Electron
Lasers, Lasers, Brunnen Brunnen, Switzerland, July 2003. , Switzerland, July 2003.
36
Possible Homework
Possible Homework
Real Accelerators
F. Sannibale
A misaligned A misaligned quadrupole quadrupole in an electron storage rings with 5.2 horizontal in an electron storage rings with 5.2 horizontal
tune generates a horizontal closed orbit distortion of 2 mm at i tune generates a horizontal closed orbit distortion of 2 mm at its own ts own
position. Calculate the kick that a corrector magnet inside the position. Calculate the kick that a corrector magnet inside the quadrupole quadrupole
needs to apply for correcting the orbit. The horizontal beta at needs to apply for correcting the orbit. The horizontal beta at the the quadrupole quadrupole
is 3 m. is 3 m.
Describe the shape of the pulse from a matched Describe the shape of the pulse from a matched stripline stripline of 5 cm length of 5 cm length
detecting a uniform distributed beam with 2 ns total length. detecting a uniform distributed beam with 2 ns total length.
Define the electronic circuit equivalent to a resistive wall cu Define the electronic circuit equivalent to a resistive wall current monitor rrent monitor
with a ferrite loaded shield. Calculate the frequency response o with a ferrite loaded shield. Calculate the frequency response of such a f such a
monitor. monitor.
Calculate the length of the detector of the FNAL Injector Calculate the length of the detector of the FNAL Injector stripline stripline in the in the
figure on the figure on the stripline stripline electrode viewgraph. electrode viewgraph.
1
Lecture No. 14
Lecture No. 14
Light Sources.
Light Sources.
Brightness and Insertion Devices
Brightness and Insertion Devices
Thanks to Herman Thanks to Herman Winick Winick and David Robin and David Robin
2
Introduction
Introduction
Light Sources
Brightness & ID
F. Sannibale
However other researchers soon realized that synchrotron However other researchers soon realized that synchrotron
radiation was the brightest source of infrared, ultraviolet, and radiation was the brightest source of infrared, ultraviolet, and x x- -
rays, and that could be very useful for studying matter on the rays, and that could be very useful for studying matter on the
scale of scale of atoms and molecules atoms and molecules. .
Electron accelerators were initially developed to probe Electron accelerators were initially developed to probe elementary elementary
( (subnuclear subnuclear) particles ) particles for the study of the fundamental nature of matter, for the study of the fundamental nature of matter,
space, time, and energy. space, time, and energy.
In the earlier times, synchrotron In the earlier times, synchrotron
radiation was just considered as a radiation was just considered as a
waste product waste product limiting the limiting the
performance achievable with lepton performance achievable with lepton
machines. machines.
The first time synchrotron radiation was observed in an accelera The first time synchrotron radiation was observed in an accelerator was in tor was in
1947 from the 70 1947 from the 70 MeV MeV electron beam at the General Electric Synchrotron in electron beam at the General Electric Synchrotron in
New York State. New York State.
3
Interaction of
Interaction of
Photons with Matter
Photons with Matter
Light Sources
Brightness & ID
F. Sannibale
Radiography Radiography
Diffraction Diffraction
Photoelectric Effect Photoelectric Effect Compton Scattering Compton Scattering
4
What is Synchrotron Radiation?
What is Synchrotron Radiation?
Light Sources
Brightness & ID
F. Sannibale
We already showed that We already showed that
synchrotron radiation is synchrotron radiation is
electromagnetic radiation emitted electromagnetic radiation emitted
when charged particles are when charged particles are radially radially
accelerated accelerated (move on a curved (move on a curved
path). path).
Electrons Electrons accelerating accelerating by by
running up and down in a running up and down in a
radio antenna emit radio radio antenna emit radio
waves (long wavelength waves (long wavelength
electromagnetic waves) electromagnetic waves)
Both cases are manifestation of the same physical phenomenon: Both cases are manifestation of the same physical phenomenon:
Charged particles radiate when accelerated Charged particles radiate when accelerated. .
5
Why Do Particles Radiate
Why Do Particles Radiate
under Acceleration?
under Acceleration?
Real Accelerators
F. Sannibale
We already saw that according to quantum field We already saw that according to quantum field
theory, a particle moving in the free space can be theory, a particle moving in the free space can be
considered as surrounded by a cloud of considered as surrounded by a cloud of virtual virtual
photons photons that appear and disappear and that that appear and disappear and that
indissolubly travel with it. indissolubly travel with it.
e
When When accelerated accelerated, the particle receives a kick that can separate it from the , the particle receives a kick that can separate it from the
photons that become real and independently observable. photons that become real and independently observable.
In the field of the magnets in a In the field of the magnets in a
synchrotron, charged particles synchrotron, charged particles
moves on a curved trajectory. The moves on a curved trajectory. The
transverse acceleration, if strong transverse acceleration, if strong
enough, allows for the separation enough, allows for the separation
and and synchrotron radiation synchrotron radiation is is
generated. generated.
Lighter particles are easier to accelerate and radiate photons Lighter particles are easier to accelerate and radiate photons more more
efficiently than heavier particles. efficiently than heavier particles.
6
The Classical Picture
The Classical Picture
Light Sources
Brightness & ID
F. Sannibale
Historically, the whole theory was developed well before quantu Historically, the whole theory was developed well before quantum m
mechanics was even conceived: mechanics was even conceived:
The description of synchrotron radiation presented in the previ The description of synchrotron radiation presented in the previous ous
viewgraph made use of quantum field theory. viewgraph made use of quantum field theory.
Power Larmor
a
c
q
P
2
3
0
2
6
=
- - in in 1897 Joseph 1897 Joseph Larmor Larmor derived the derived the
expression for the instantaneous total expression for the instantaneous total
power radiated by an accelerated charged power radiated by an accelerated charged
particle. particle.
- - and in and in 1898 Alfred 1898 Alfred Lienard Lienard
(before the relativity theory!) (before the relativity theory!)
extended extended Larmors Larmors result to result to
the case of a relativistic the case of a relativistic
particle undergoing particle undergoing
centripetal acceleration in a centripetal acceleration in a
circular trajectory circular trajectory
1857 1857- -1942 1942
1869 1869- -1958 1958
7
Longitudinal vs.
Longitudinal vs.
Transverse Acceleration
Transverse Acceleration
Light Sources
Brightness & ID
F. Sannibale
2
2
3 2
0 0
2
6

dt
d
c m
q
P
p
2
||
3 2
0 0
2
||
6

=
dt
d
c m
q
P
p
negligible!
( )
radius curvature q
c
P = =
2
4
2
0
6
Radiated power for transverse acceleration Radiated power for transverse acceleration increases dramatically with increases dramatically with
energy energy. This sets a practical limit for the maximum energy obtainable . This sets a practical limit for the maximum energy obtainable with a with a
storage ring, but makes the construction of synchrotron light so storage ring, but makes the construction of synchrotron light sources urces
extremely appealing! extremely appealing!
8
The Brightness
The Brightness
of a Light Source
of a Light Source
Light Sources
Brightness & ID
F. Sannibale
Flux Flux = # of photons in given = # of photons in given / /
sec sec
Brightness Brightness = # of photons in given = # of photons in given / /
sec, mrad sec, mrad , mrad , mrad , mm , mm
2 2
= =
d S d Brightness
d
N d
Flux
In one of the previous lectures, we already dealt with the conc In one of the previous lectures, we already dealt with the concept of ept of brightness brightness
and showed how this quantity is the one of the main parameters f and showed how this quantity is the one of the main parameters for the or the
characterization of a particle source. characterization of a particle source.
We remind that We remind that brightness is defined brightness is defined as the density of particle on the 6 as the density of particle on the 6- -D phase D phase
space space. .
The same definition applies to the photon case The same definition applies to the photon case, just taking into account that , just taking into account that
photons are bosons and that the photons are bosons and that the Pauli Pauli exclusion principle does not apply. exclusion principle does not apply.
This is an important advantage because, at least from the point This is an important advantage because, at least from the point of view of of view of
quantum mechanics, no limitation to achievable photon brightness quantum mechanics, no limitation to achievable photon brightness exists. exists.
From the above definitions, one can see that for a given flux, From the above definitions, one can see that for a given flux, sources with a sources with a
smaller smaller emittance emittance will have a larger brightness. will have a larger brightness.
9
How Bright is a
How Bright is a
Synchrotron Light Source?
Synchrotron Light Source?
Light Sources
Brightness & ID
F. Sannibale
10
Angular Distribution
Angular Distribution
Light Sources
Brightness & ID
F. Sannibale
Radiation becomes more focused at higher energies. Radiation becomes more focused at higher energies.
Cone aperture Cone aperture
~ 1/ ~ 1/
11
Time Compression
Time Compression
Light Sources
Brightness & ID
F. Sannibale
12
Typical Band
Typical Band
-
-
Width
Width
Of Synchrotron Light
Of Synchrotron Light
Light Sources
Brightness & ID
F. Sannibale
t
=
1
Example for an electron ring with 1.9 Example for an electron ring with 1.9 GeV GeV and with a bending radius of 5 m: and with a bending radius of 5 m:
1 18 19
10 1 . 3 10 2 . 3 7 . 2

s s t mm l
nm
f
c
Hz f
MAX
MIN MAX
61 . 0 10 9 . 4
2
17
=
1
2
2
= l
Very broad band! Very broad band!
13
Electromagnetic Spectrum
Electromagnetic Spectrum
Light Sources
Brightness & ID
F. Sannibale
Coherent Synchrotron Radiation Coherent Synchrotron Radiation
THz THz Synchrotron Light Sources Synchrotron Light Sources
Synchrotron Light Sources Synchrotron Light Sources
14
Polarization
Polarization
Light Sources
Brightness & ID
F. Sannibale
15
What Properties Make Synchrotron
What Properties Make Synchrotron
Radiation so Useful?
Radiation so Useful?
Light Sources
Brightness & ID
F. Sannibale
High brightness and flux High brightness and flux
SR offers many characteristics of visible lasers
SR offers many characteristics of visible lasers
but into the x
but into the x
-
-
ray regime!
ray regime!
Wide energy spectrum Wide energy spectrum
Highly polarized and short pulses Highly polarized and short pulses
Recapitulating the main properties of synchrotron radiation: Recapitulating the main properties of synchrotron radiation:
Partial coherence Partial coherence
High Stability High Stability
16
How to Exploit
How to Exploit
Light Sources
Brightness & ID
F. Sannibale
Modern synchrotron light sources are accelerators optimized for Modern synchrotron light sources are accelerators optimized for the the
production of synchrotron radiation. production of synchrotron radiation.
ALS LINAC ALS LINAC
ALS ALS
ALS Booster ALS Booster
CIRCE CIRCE
ESRF ESRF- - France France
SPRING 8 SPRING 8
Japan Japan
APS APS - - USA USA
17
Applications
Applications
Light Sources
Brightness & ID
F. Sannibale
Medicine
Medicine
Biology
Biology
Chemistry
Chemistry
Material Science
Material Science
Environmental Science
Environmental Science
and much more

and much more
18
Materials Science
Materials Science
Light Sources
Brightness & ID
F. Sannibale
Using SR to learn Using SR to learn
how high how high
temperature temperature
superconductors superconductors
work work
Using SR to make Using SR to make
miniature mechanical and miniature mechanical and
electromechanical devices electromechanical devices
Visualizing magnetic Visualizing magnetic
bits on a computer bits on a computer
hard drive hard drive
Understanding Understanding
how debris causes how debris causes
damage to aircraft damage to aircraft
turbines turbines
19
Next generation
Next generation
of semiconductors
of semiconductors
Light Sources
Brightness & ID
F. Sannibale
EUV Lithography EUV Lithography
ALS ALS
20
Chemistry and Biology
Chemistry and Biology
Light Sources
Brightness & ID
F. Sannibale
Studying Studying
Anthrax Toxin Anthrax Toxin
components to components to
develop develop
treatment in the treatment in the
advanced stages advanced stages
of infection. of infection.
Measuring very low levels of mercury in Measuring very low levels of mercury in
fish and determining its chemical form. fish and determining its chemical form.
Cholera toxin attacking a gut cell Cholera toxin attacking a gut cell
21
Protein Crystallography
Protein Crystallography
Light Sources
Brightness & ID
F. Sannibale
Leukemia Leukemia
Drug Design Drug Design
GLEEVEC GLEEVEC
Understanding how proteins are made Understanding how proteins are made
Ribosomes Ribosomes make the stuff of life. They are make the stuff of life. They are
the protein factories in every living the protein factories in every living
creature, and they churn out all proteins creature, and they churn out all proteins
ranging from bacterial toxins to human ranging from bacterial toxins to human
digestive enzymes digestive enzymes
22
Cellular Imaging
Cellular Imaging
Light Sources
Brightness & ID
F. Sannibale
This is an image taken with the x This is an image taken with the x- -
ray microscope of a malaria ray microscope of a malaria- -
infected blood cell. Researchers at infected blood cell. Researchers at
Berkeley Lab use pictures like this Berkeley Lab use pictures like this
to analyze what makes the malaria to analyze what makes the malaria- -
infected blood cells stick to the infected blood cells stick to the
blood vessels. blood vessels.
23
Biomedicine
Biomedicine
Light Sources
Brightness & ID
F. Sannibale
Image of a human coronary artery Image of a human coronary artery
taken with synchrotron radiation at taken with synchrotron radiation at
SSRL SSRL
before estrogen loss after estrogen loss
Studies of osteoporosis at SSRL Studies of osteoporosis at SSRL
These studies make use of the These studies make use of the
penetrating power of X penetrating power of X- -rays, rays,
rather than their short rather than their short
wavelength wavelength
24
Art & Archaeology
Art & Archaeology
Light Sources
Brightness & ID
F. Sannibale
Sulfuric acid causing the Sulfuric acid causing the
decay of the decay of the Vasa Vasa, the , the
Swedish warship which sank Swedish warship which sank
in Stockholm harbor in 1628 in Stockholm harbor in 1628
Virgin, Child, and Saint John Virgin, Child, and Saint John A A
renaissance panel painting by Jacopo renaissance panel painting by Jacopo
Sellaio Sellaio or or Filippino Filippino Lippi Lippi being restored at being restored at
the Cantor Art Center the Cantor Art Center
25
X
X
-
-
rays have enabled
rays have enabled
seminal scientific discoveries
seminal scientific discoveries
Light Sources
Brightness & ID
F. Sannibale
18 Nobel Prizes
Based on X-ray
Work
Chemistry
1936: Peter Debye
1962: Max Purutz and Sir John
Kendrew
1976 William Lipscomb
1985 Herbert Hauptman and Jerome
Karle
1988 Johann Deisenhofer, Robert
Huber and Hartmut Michel
1997 Paul D. Boyer and John E.
Walker
2003 Peter Agre and Roderick
Mackinnon
Physics
1901 Wilhem Rontgen
1914 Max von Laue
1915 Sir William Bragg and son
1917 Charles Barkla
1924 Karl Siegbahm
1927 Arthur Compton
1981 Kai Siegbahn
Medicine
1946 Hermann Muller
1962 Frances Crick, James Watson
and Maurice Wilkins
1979 Alan Cormack and Godrey
Hounsfield
26
20 000 Users World
20 000 Users World
-
-
Wide
Wide
Light Sources
Brightness & ID
F. Sannibale
54 in operation in 19 countries used by more than 20,000 scient 54 in operation in 19 countries used by more than 20,000 scientists ists
8 in construction 8 in construction
11 in design/planning 11 in design/planning
For a list of SR facilities around the world see
http://ssrl.slac.stanford.edu/SR_SOURCES.HTML
www.sesame.org.jo
27
Steep Growth in Brightness
Steep Growth in Brightness
Light Sources
Brightness & ID
F. Sannibale
Wilhelm Conrad Wilhelm Conrad
Rntgen Rntgen (1845 (1845- -1923) 1923)
1845 1845
28
Growth in X
Growth in X
-
-
ray Brightness Compared
ray Brightness Compared
to Growth in Computing Speed
to Growth in Computing Speed
Light Sources
Brightness & ID
F. Sannibale
One million One billion
A million more
Computing speed
X-rays
10
15
10
14
10
13
10
12
10
11
10
10
10
9
10
8
10
7
10
6
10
5
10
4
10
3
10
2
10
1
10
0
10
-1
10
23
10
22
10
21
10
20
10
19
10
18
10
17
10
16
10
15
10
14
10
13
10
12
10
11
10
10
10
9
10
8
10
-7
1960 1970 1980 1990 2000
Computing speed
X-ray Brightness
X-ray Brightness
Computing speed
29
How to Optimize a
How to Optimize a
Synchrotron Light Source
Synchrotron Light Source
Light Sources
Brightness & ID
F. Sannibale
The ultimate performance parameter of a synchrotron light sourc The ultimate performance parameter of a synchrotron light source is the e is the
brightness. brightness.
The battle for the brightness maximization is fought in two fro The battle for the brightness maximization is fought in two fronts: nts:
In the storage ring, by increasing the current and designing ne In the storage ring, by increasing the current and designing new w
lattices capable of smaller lattices capable of smaller emittances emittances. Current of hundreds of . Current of hundreds of mA mA and and
lattices with ~ 1 nm lattices with ~ 1 nm emittance emittance are presently used. are presently used.
In the ring elements where the synchrotron radiation is actuall In the ring elements where the synchrotron radiation is actually y
generated: dipole magnets and insertion devices. And this is whe generated: dipole magnets and insertion devices. And this is where re
spectacular improvements have been achieved! spectacular improvements have been achieved!
Light sources are usually classified for increasing brightness Light sources are usually classified for increasing brightness as: as:
1 1
st st
generation generation: x : x- -ray tubes. ray tubes.
2 2
nd nd
generation generation: parasitic synchrotron radiation sources from : parasitic synchrotron radiation sources from
dipoles in dipoles in colliders colliders. .
3 3
rd rd
generation generation: dedicated storage rings with insertion devices : dedicated storage rings with insertion devices
4 4
th th
generation generation: free electron lasers : free electron lasers
30
How Synchrotron Radiation
How Synchrotron Radiation
is Generated in Storage Rings
is Generated in Storage Rings
Light Sources
Brightness & ID
F. Sannibale
Continuous spectrum characterized Continuous spectrum characterized
by by
c c
= critical energy = critical energy

c c
(keV (keV) = 0.665 B(T)E ) = 0.665 B(T)E
2 2
(GeV) (GeV)
For example For example: :
for B = 1.35 T E = 2 for B = 1.35 T E = 2 GeV GeV
c c
= 3.6keV = 3.6keV
+ harmonics at higher energy + harmonics at higher energy
Quasi Quasi- -monochromatic spectrum monochromatic spectrum with with
peaks at lower energy than a wiggler peaks at lower energy than a wiggler
K = where is the angle in each is the angle in each
pole pole
1
=
u
2
2
(1 + ) ~ (fundamental) (fundamental)
K
2
2
2
1
(keV) =
0.95 E
2
(GeV)
K
2
u
(cm)
(1 + )
2
bending magnet
t
1
t
2
Bending Magnet A Sweeping Searchlight
Wiggler Incoherent Superposition
Dipoles
t
3 t
4
(10-100)
1
t
5
1
Undulator Coherent Interference
( N)
1
t
1
t
2
Dipoles
t
3 t
4
(10-100)
1
t
5
1
( N)
1
wiggler - incoherent superposition
t
1
t
2
Dipoles
t
3 t
4
(10-100)
1
t
5
1
( N)
1
undulator - coherent interference
31
Bend Magnet
Bend Magnet
Light Sources
Brightness & ID
F. Sannibale
C shaped for C shaped for
allowing to the allowing to the
radiation to exit radiation to exit
Normal Normal- -Conductive Conductive
~ 1.5 T Max ~ 1.5 T Max
32
Bend Magnet Synchrotron
Bend Magnet Synchrotron
Radiation Spectrum
Radiation Spectrum
Light Sources
Brightness & ID
F. Sannibale
Universal function Universal function Spectrum: Spectrum:
Critical frequency Critical frequency
33
Dipoles for Hard X
Dipoles for Hard X
-
-
rays
rays
Light Sources
Brightness & ID
F. Sannibale
At the Advanced Light Source three of the existing thirty six 1. At the Advanced Light Source three of the existing thirty six 1.3 T 3 T
dipoles were replaced by three dipoles were replaced by three 5 T superconducting dipoles 5 T superconducting dipoles
( (superbends superbends). ).
Superbend Superbend
without cryostat without cryostat
Superbend Superbend
with cryostat with cryostat
300.0
B

(
T
)
-180.0 -60.0 60.0 180.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
distance along beam (mm)
-300.0
0.0
Superbend Superbend
magnetic field magnetic field
34
Bending Magnet Bending Magnet
Spectrum Energy Dependency
Spectrum Energy Dependency
Light Sources
Brightness & ID
F. Sannibale
Remark: The Remark: The
distribution for longer distribution for longer
wavelengths does not wavelengths does not
depend on energy. depend on energy.
35
Planar
Planar
Undulators
Undulators
Light Sources
Brightness & ID
F. Sannibale
Particle Particle
trajectory trajectory
Permanent Magnets Permanent Magnets
Invented by Invented by
Klaus Klaus Halbach Halbach
1924 1924- -2000 2000
36
Undulator
Undulator
Radiation
Radiation
Light Sources
Brightness & ID
F. Sannibale
Photons emitted by different poles interfere transforming the Photons emitted by different poles interfere transforming the
continuous dipole continuous dipole- -like spectrum into a discrete spectrum like spectrum into a discrete spectrum
The interference condition requires that, while traveling along The interference condition requires that, while traveling along one period of one period of
the the undulator undulator, the electrons slip by one radiation wavelength with respect to , the electrons slip by one radiation wavelength with respect to
the (faster) photon. the (faster) photon.
37
From
From
Undulator
Undulator
Radiation
Radiation
to Wiggler Radiation
to Wiggler Radiation
Light Sources
Brightness & ID
F. Sannibale
The spectrum of the The spectrum of the undulator undulator radiation: radiation:
pole each in angle bending the is where K
U

2
=
harmonic
K
st U
1
2
1
2
2
2
1

+ =
depends strongly on the depends strongly on the

strength parameter strength parameter K K: :
eB
c m
0
= Remembering that: Remembering that:

One can see that One can see that K K is is
proportional to the field proportional to the field B B: :
In a permanent magnet In a permanent magnet undulator undulator, , B B and consequently and consequently K K
can be modified by changing the gap height. The larger the can be modified by changing the gap height. The larger the
gap the lower the field. gap the lower the field.
When When B B is increased, both is increased, both K K and the and the
wiggling inside the wiggling inside the undulator undulator increase as increase as
well well. With the larger wiggling, the overlap . With the larger wiggling, the overlap
between the radiated field ( between the radiated field (1/ 1/ cone) cone)
decreases and the interference is reduced. decreases and the interference is reduced.
For For K K >> 1 >> 1 no interference is present and no interference is present and
the the undulator undulator presents the continuum presents the continuum
spectrum typical of the spectrum typical of the wiggler wiggler. .
Gap Gap
B
c m
e
K
U
0
2
38
Elliptically Polarizing
Elliptically Polarizing
Undulators
Undulators
Light Sources
Brightness & ID
F. Sannibale
ALS EPU50 (1998) ALS EPU50 (1998)
Pure permanent Pure permanent
magnet technology, magnet technology,
Elliptically polarizing Elliptically polarizing
capability. capability.
Such a device allows Such a device allows
for the complete control for the complete control
of the polarization from of the polarization from
linear in to elliptical. linear in to elliptical.
The arrays of The arrays of
permanent magnets permanent magnets
can be mechanically can be mechanically
shifted modifying shifted modifying
the polarization of the polarization of
the radiated light. the radiated light.
39
Future of Synchrotron Radiation
Future of Synchrotron Radiation
Light Sources
Brightness & ID
F. Sannibale
Shorter Pulse Lengths Shorter Pulse Lengths
- - Femto Femto (10 (10
- -12 12
) and ) and Attosecond Attosecond (10 (10
- -15 15
) )
Higher Brightness Higher Brightness
- - Free Electron Lasers Free Electron Lasers
Terahertz (T Terahertz (T- -rays) rays)
- - Coherent Synchrotron Radiation Coherent Synchrotron Radiation
(?)
(?)
40
Free Electron Laser Basics
Free Electron Laser Basics
Light Sources
Brightness & ID
F. Sannibale
In free electron lasers (FEL), a relativistic electron beam and In free electron lasers (FEL), a relativistic electron beam and a laser overlap and a laser overlap and
travel simultaneously inside an travel simultaneously inside an undulator undulator. .
Oscillating through the Oscillating through the undulator undulator, the electron bunch interacts with the laser and , the electron bunch interacts with the laser and
in a minor way with its own electromagnetic field created via sp in a minor way with its own electromagnetic field created via spontaneous ontaneous
emission. Depending on the relative phase between radiation and emission. Depending on the relative phase between radiation and electron electron
oscillation, electrons experience either a deceleration or accel oscillation, electrons experience either a deceleration or acceleration. eration.
The laser is tuned at the frequency of one The laser is tuned at the frequency of one
of the of the undulator undulator harmonics. The whole harmonics. The whole
undulator undulator is included inside an optical cavity is included inside an optical cavity
composed by two reflecting mirrors located composed by two reflecting mirrors located
at the two at the two undulator undulator extremes. extremes.
In such a schemes the laser beam In such a schemes the laser beam
bounces many times back and forward bounces many times back and forward
inside the cavity and has multiple inside the cavity and has multiple
interactions with the electron beam. interactions with the electron beam.
Through this interaction a longitudinal fine structure, the so Through this interaction a longitudinal fine structure, the so called called micro micro- -
bunching bunching, is established which amplifies the electromagnetic field at th , is established which amplifies the electromagnetic field at the laser e laser
frequency. frequency.
41
The SASE FEL Scheme
The SASE FEL Scheme
Light Sources
Brightness & ID
F. Sannibale
In the self In the self- -amplified spontaneous emission (SASE) FEL, there is no laser and amplified spontaneous emission (SASE) FEL, there is no laser and the the
electron beam interacts only with its own spontaneous emission. electron beam interacts only with its own spontaneous emission.
In order to obtain a large gain in the SASE In order to obtain a large gain in the SASE
scheme, a long and expensive scheme, a long and expensive undulator undulator is is
required. In a conventional FEL the required. In a conventional FEL the
undulator undulator is much shorter because the is much shorter because the
laser beam is re laser beam is re- -circulated many times circulated many times
inside the cavity. Unfortunately, the highest inside the cavity. Unfortunately, the highest
frequency achievable with such a frequency achievable with such a
configuration is limited to the near configuration is limited to the near- -UV UV
because of the absence of efficient large because of the absence of efficient large
incidence angle mirrors for shorter incidence angle mirrors for shorter
wavelengths. wavelengths.
Saturation Saturation
For such a scheme to work, one has to guarantee a good electron For such a scheme to work, one has to guarantee a good electron beam quality beam quality
and a sufficient overlap between radiation pulse and electron bu and a sufficient overlap between radiation pulse and electron bunch along the nch along the
undulator undulator. To achieve that, one needs a low . To achieve that, one needs a low emittance emittance, low energy spread electron , low energy spread electron
beam with an extremely high charge density in conjunction with a beam with an extremely high charge density in conjunction with a very precise very precise
magnetic field and accurate beam steering through the magnetic field and accurate beam steering through the undulator undulator. .
42
X X- -FEL FEL
(Some) FEL Projects
(Some) FEL Projects
Light Sources
Brightness & ID
F. Sannibale
43
References
References
Light Sources
Brightness & ID
F. Sannibale
Technical Design Report (TDR) for TESLA, Technical Design Report (TDR) for TESLA, Part V Part V The X The X- -ray free electron laser ray free electron laser
44
Possible Problems
Possible Problems
Light Sources
Brightness & ID
F. Sannibale
Calculate the critical energy in Calculate the critical energy in eV eV for the ALS for the ALS superbends superbends knowing that the knowing that the
electron beam energy is 1.9 electron beam energy is 1.9 GeV GeV, the field is 5 T and the total deflection angle , the field is 5 T and the total deflection angle
for the magnet is 10 deg. Remember that the photon energy is giv for the magnet is 10 deg. Remember that the photon energy is given by en by hf hf
(with (with h h the Planck constant, 6.626068 10 the Planck constant, 6.626068 10
- -34 34
m m
2 2
kg / s, and kg / s, and f f the photon the photon
frequency) frequency)
Always for the ALS case, calculate the critical energy for the Always for the ALS case, calculate the critical energy for the normal bends normal bends
knowing that the bending radius is 4.957 m and the total deflect knowing that the bending radius is 4.957 m and the total deflection angle for ion angle for
the magnet is 10 deg. the magnet is 10 deg.
Using the universal spectrum for the bending magnet radiation, Using the universal spectrum for the bending magnet radiation, calculate for calculate for
both the above cases, the maximum radiated power in 0.1% bandwid both the above cases, the maximum radiated power in 0.1% bandwidth when th when
400 400 mA mA electrons are stored ( the ring length is 197 m). Indicate at w electrons are stored ( the ring length is 197 m). Indicate at which hich
photon energy is the maximum located. photon energy is the maximum located.
1
Lecture No. 15
Lecture No. 15
Colliders
Colliders
and Luminosity
and Luminosity
2
Introduction
Introduction
Colliders &
Luminosity
F. Sannibale
Colliders Colliders for high energy particle physics experiments are surely one of for high energy particle physics experiments are surely one of the the
most important application of particle accelerators. most important application of particle accelerators.
Actually, the developments in particle accelerators and of eleme Actually, the developments in particle accelerators and of elementary particle ntary particle
physics probably represent one of the best examples of synergy b physics probably represent one of the best examples of synergy between etween
different physics disciplines. different physics disciplines.
Colliders Colliders can be characterized by the different nature of the colliding can be characterized by the different nature of the colliding
particles (leptons or hadrons) and by the different acceleration particles (leptons or hadrons) and by the different acceleration scheme used scheme used
(linear or circular) (linear or circular)
In existing lepton In existing lepton colliders colliders, electrons collide with positrons and a significant , electrons collide with positrons and a significant
R&D is undergoing for the definition of a possible scheme for a R&D is undergoing for the definition of a possible scheme for a muon muon
collider collider. .
Hadron Hadron colliders colliders include, protons colliding with protons or anti include, protons colliding with protons or anti- -protons and protons and
heavy ion heavy ion colliders colliders. .
Higher collision energies can be achieved with Higher collision energies can be achieved with hadron hadron colliders colliders but cleaner but cleaner
measurements can be done with lepton measurements can be done with lepton colliders colliders. .
3
Introduction
Introduction
Colliders &
Luminosity
F. Sannibale
In electron In electron- -positron collisions the particles annihilate and all the energy positron collisions the particles annihilate and all the energy at at
the center of mass system is available for the generation of ele the center of mass system is available for the generation of elementary mentary
particles. particles.
Such particle generation can happen only if it exists a particle Such particle generation can happen only if it exists a particle with rest mass with rest mass
energy equal to the collision energy at the center of mass syste energy equal to the collision energy at the center of mass system. m.
The energy of the colliding beams can be tuned on the rest mass The energy of the colliding beams can be tuned on the rest mass energy of a energy of a
known particle for studying its properties, or can be scanned fo known particle for studying its properties, or can be scanned for the research r the research
of unknown particles. of unknown particles.
In In hadron hadron colliders colliders, the quarks in the hadrons interact during the collision , the quarks in the hadrons interact during the collision
and generate other particles. Because each and generate other particles. Because each hadron hadron is a combination of three is a combination of three
quarks, quarks, simultanoeus simultanoeus generation of different particles is possible. generation of different particles is possible.
Most of the particles generated during a collisions usually have Most of the particles generated during a collisions usually have a short a short
lifetime and decay in other particles. Particles detectors are d lifetime and decay in other particles. Particles detectors are designed in order esigned in order
to measure the particle itself when possible or to measure the p to measure the particle itself when possible or to measure the particles articles
generated during the decay. generated during the decay.
4
High Energy Physics Detectors
High Energy Physics Detectors
Colliders &
Luminosity
F. Sannibale
CDF CDF
CDF CDF
KLOE KLOE
KLOE KLOE
5
Energy at the
Energy at the
Center of Mass System
Center of Mass System
Colliders &
Luminosity
F. Sannibale
Two particles have equal rest mass m
0
.
Center of Mass Frame (CMF): Velocities are equal and
opposite, total energy is E
cm
.
The quantity is invariant.
In the CMF, we have
While in the LF:
And after some algebra we can obtain for relativistic particles:
2 1
2 E E E
cm

( )
2 1
2 2
0 2 1
2
1
2
1
2
2 1
~ ~
2 2
~ ~
2
~ ~ ~ ~
P P c m P P P P P P + = + + = +
( )
2 2
2
2 1
c E P P
CM
= +
( )
2
2 1
P P +
( ) p c E P
CM
, 2
1
=
( ) p c E P
CM
= , 2
2
Laboratory frame (LF):
( )
1 1 1
, 2
~
p c E P =
( )
2 2 2
, 2
~
p c E P =
6
Interaction Region
Interaction Region
Reference Frame
Reference Frame
Colliders &
Luminosity
F. Sannibale
{ x, y, z } Lab. Reference Frame
P
1
P
2
s
+
s
-
y vertical axis
x horizontal axis
IP
z longitud. axis
crossing angle
7
Basic Definitions
Basic Definitions
Colliders &
Luminosity
F. Sannibale
Luminosity:
Luminosity:
Event Rate
Event Rate
for a
for a
Unit Cross Section Event
Unit Cross Section Event
Event Rate
Event Rate
per Unit Incident Flux
per Unit Incident Flux
per Target Particle
per Target Particle
Cross Section:
Cross Section:
8
Counter Rotating
Counter Rotating
Beams Case
Beams Case
Colliders &
Luminosity
F. Sannibale
L = == = 2v f
R

dxdydz dt n
+ ++ +
x, y, z, t ( (( ( ) )) ) n

x, y, z, t ( (( ( ) )) )
n

x, y, z, t ( (( ( ) )) )
dxdy dz

n

x, y, z, t ( (( ( ) )) ) = == = N

Head Head- -on Collision on Collision
Counter Counter- -rotating Beams rotating Beams
with Longitudinal Speed with Longitudinal Speed
v
v
-
-
-
-
Single Bunch Single Bunch
-
-
Revolution Frequency Revolution Frequency f f
R R
-
-
9
Gaussian
Gaussian
Beam
Beam
Single Bunch Luminosity
Colliders &
Luminosity
F. Sannibale
n

x, y, z, t ( (( ( ) )) ) = == = N

e

x
2
2
x
2

y
2
2
y
2

z vt ( (( ( ) )) )
2
2
z
2
2 ( (( (
) )) )
3
2

x

y

z
n
+ ++ +
x, y, z, t ( (( ( ) )) ) = == = N
+ ++ +
e

x
2
2
x+ ++ +
2

y
2
2
y+ ++ +
2

z + ++ +vt ( (( ( ) )) )
2
2
z+ ++ +
2
2 ( (( (
) )) )
3
2

x+ ++ +

y+ ++ +

z+ ++ +

x
,
y
constants
L = == = f
R
N
+ ++ +
N

2
x+ ++ +
2
+ ++ +
x
2
( (( ( ) )) )

y+ ++ +
2
+ ++ +
y
2
( (( ( ) )) )

x+ ++ +
= == =
x

y+ ++ +
= == =
y
L = == = f
R
N
+ ++ +
N

4
x

y
10
Two Main Effects
Two Main Effects
Colliders &
Luminosity
F. Sannibale
Charge Related Effects:
Charge Related Effects:
Or
Or
beam
beam
-
-
beam
beam
effects.
effects.
Charge plays a major role,
Charge plays a major role,
limiting the achievable
limiting the achievable
luminosity in most of
luminosity in most of
storage ring
storage ring
colliders
colliders
.
.
How the geometry of
How the geometry of
the interaction point (IP)
the interaction point (IP)
and the size of the
and the size of the
beams affect luminosity
beams affect luminosity
Geometric Effects:
Geometric Effects:
11
Geometric Luminosity:
Geometric Luminosity:
Collider
Collider
Parameters
Parameters
Colliders &
Luminosity
F. Sannibale

y

= == =

1+ ++ +

y

x

= == =
1
1+ ++ +

x

x
= == =
y
= == = 0
L = == = f
R
N
+ ++ +
N

1+ ++ +
( (( ( ) )) )
4
x

y

f
R
N
+ ++ +
N

4
x

y

L = == = f
R
N
+ ++ +
N

4
x

y
* * * *
12
Geometric Luminosity
Geometric Luminosity
Colliders &
Luminosity
F. Sannibale
L = 2 f
R
v
c
cos c
2
v
2
sin
2

dxdydzdt f
+
f
f
+
=
N
+
e

x cos z sin ( )
2
2
x+
2
x, z ( )

y
2
2
y+
2
x ,z ( )

x sin + z cos vt ( )
2
2
z+
2
2 ( )
3
2

x+
x, z ( )
y +
x, z ( )
z+
k +
2
=
k +
2
1+
x sin + z cos w
+
( )
k+
2

k = x, y
f
=
N
e

x x ( ) cos+ z sin [ ]
2
2
x
2
x, z ( )

y y ( )
2
2
y
2
x ,z ( )

x x ( )sin + z cos +v t t ( ) [ ]
2
2
z
2
2 ( )
3
2

x
x, z ( )
y
x, z ( )
z
k
2
=
k
2
1 +
x sin + z cos w
( )
k
2

k = x, y
Very low currents Very low currents
Negligible Negligible
beam beam- -beam effects beam effects
Crossing angle Crossing angle
Horizontal & vertical Horizontal & vertical
Offset Offset x x and and y y
IR position IR position t t
Different beta Different beta
star for the two star for the two
beams beams
13
Beam
Beam
-
-
Beam Effects
Beam Effects
Colliders &
Luminosity
F. Sannibale
y
z
For a For a gaussian gaussian charge distribution: charge distribution:
y
=
2Nr
e

y
exp

x
2
2
x
2
+
w

y
2
2
y
2
+
w
|
\
|
.
|
2
y
2
+
w ( )
3
2
2
x
2
+
w
( )
1
2
dw
x
=
2Nr
e

x
exp

x
2
2
x
2
+
w

y
2
2
y
2
+
w
|
\
|
.
|
2
y
2
+
w ( )
1
2
2
x
2
+
w ( )
3
2
dw
[m] y
03
03
-100 m 100 m
10 mrad
-10 mrad
x = 0
N = 8.9 10
10
y
=20 m
E= 510 MeV
y
y
14
Linear Beam
Linear Beam
-
-
Beam
Beam
Colliders &
Luminosity
F. Sannibale
Q
F
k
1 0
1
f
k
1
|
\

|
.
| k = x, y
Focusing Focusing Quadrupole Quadrupole (thin lens): (thin lens):
Beam Beam- -Beam Beam
Deflection Deflection
(off (off- -center particles) center particles)
Linear Beam Linear Beam- -Beam Tune Shift Beam Tune Shift
y
+
=
N
r
e
y
+
2
y
+
x
( )
= Q
y

x
+
=
N
r
e
x
+
2
x
+
x
( )
= Q
x
y <<
y
x <<
x
* *
y
1
f
y
y f
y
=
2Nr
e
y

x
+
y
( )

x
1
f
x
x f
x
=
2Nr
e
x

x
+
y
( )
-1
-1
* *
*
* * *
15
4
4.5
5
5.5
6
4 4.5 5 5.5 6
Q
y
Q
x
Choice of the Working Point
Choice of the Working Point
Colliders &
Luminosity
F. Sannibale
m + n = resonance order
mQ
x
+ nQ
y
= p m, n, p N
Tune Tune Resonances Resonances
Working Point Working Point
Beam Beam- -Beam Tune Beam Tune
Footprint Footprint
16
Nonlinear Beam
Nonlinear Beam
-
-
Beam
Beam
Colliders &
Luminosity
F. Sannibale
Nonlinear Beam Nonlinear Beam- -beam Kick beam Kick
Synchro Synchro- -betatron betatron Effects Effects
Radiation Damping Radiation Damping
Quantum Fluctuations in Quantum Fluctuations in
Synchrotron Radiation Emission Synchrotron Radiation Emission
Lattice Lattice Nonlinearities Nonlinearities
( (sextupoles sextupoles, higher order , higher order multipoles multipoles) )
RF RF Nonlinearities Nonlinearities
- -
- -
- -
- -
- -
- -
Simulation Codes Simulation Codes
For example: For example:
LIFETRACK by LIFETRACK by Shatilov Shatilov
or or
BBC by Hirata BBC by Hirata
[m] y
03
03
-100 m 100 m
10 mrad
-10 mrad
y
y
No analytical solutions for the non No analytical solutions for the non- -linear beam linear beam- -beam case. beam case.
17
Beam
Beam
-
-
Beam Simulations
Beam Simulations
Colliders &
Luminosity
F. Sannibale
5.14, 5.20
5.14, 5.21
5.14, 5.22
5.15, 5.20 5.15, 5.21 5.15, 5.22
5.16, 5.20 5.16, 5.21
5.16, 5.22
Beam Beam- -beam simulations for beam simulations for
the DA the DA NE NE collider collider. .
Q
x
Q
y
Q Q
x x
= 5.09 = 5.09 Q Q
y y
= 5.07 = 5.07
Q Q
x x
= 5.15 = 5.15 Q Q
y y
= 5.21 = 5.21
Q
x
Q
y
18
Maximum Linear Beam
Maximum Linear Beam
-
-
Beam
Beam
Tune Shift
Tune Shift
Colliders &
Luminosity
F. Sannibale
The beam The beam- -beam interaction actually sets the maximum beam interaction actually sets the maximum
achievable luminosity in practically all the existing achievable luminosity in practically all the existing colliders colliders. .
No consistent and exhaustive theory exists. No consistent and exhaustive theory exists.
Estimate of the maximum achievable linear tune Estimate of the maximum achievable linear tune
shift: shift:
Statistical analysis of the maximum linear tune Statistical analysis of the maximum linear tune
shifts achieved in existing shifts achieved in existing colliders colliders
Simulations Simulations
The linear beam The linear beam- -beam parameter is used a measure of the beam parameter is used a measure of the
strength of the beam strength of the beam- -beam interaction. beam interaction.
19
In Max Linear Tune Shift Regime
In Max Linear Tune Shift Regime
Colliders &
Luminosity
F. Sannibale
N
+
= N
= N
x
=
1
1+

x

y
=

1+

y
L =

r
e
|
\

|
.
|
2
f
R
1+ ( )

2
Equal tune shift Equal tune shift

design design
L = f
R
N
+
N
4
x
y
*
*
w
=
N r
e
2
w

y
+
x
( )
w = x, y
*
* *
L =

r
e
|
\

|
.
|
2
f
R
1+
1
y
+

x
|
\

|
.
|
2
y
*
*
=
x
=
y
=
20
Low Beta Scheme
Low Beta Scheme
Colliders &
Luminosity
F. Sannibale
-5
0
5
10
15
20
25
30
35
-3 -2 -1 0 1 2 3
z (m)
x
(m)
Few centimeters vertical beta @ IP are routinely obtained. Few centimeters vertical beta @ IP are routinely obtained.
21
Hourglass Effect
Hourglass Effect
Colliders &
Luminosity
F. Sannibale
In a low beta insertion, the beta function between the IP and th In a low beta insertion, the beta function between the IP and the first e first
quadrupole quadrupole is given by: is given by:
( )
|
|
.
|
\
|
+ = =
2
2
1
y
y x y x y
z
z

( )
|
|
.
|
\
|
+ =
2
1
y
y y
z
z

The The hourglass effect hourglass effect can can
limit the luminosity limit the luminosity
achievable in low beta achievable in low beta
schemes. schemes.
The use of short bunches The use of short bunches
can reduce the effect on can reduce the effect on
luminosity. luminosity.
y
(m)
-5
0
5
10
15
20
25
30
35
-3 -2 -1 0 1 2 3
z (m)
Hourglass Hourglass
shaped beam shaped beam
L =

r
e
|
\

|
.
|
2
f
R
1+ ( )

2
22
Low Beta Scheme Implications
Low Beta Scheme Implications
Colliders &
Luminosity
F. Sannibale
Large Vertical Beta Functions in D Quads @ IR Large Vertical Beta Functions in D Quads @ IR
Larger Negative Values of Vertical Larger Negative Values of Vertical Chromaticities Chromaticities
Stronger Correcting Stronger Correcting Sextupoles Sextupoles
Smaller Dynamic Aperture Smaller Dynamic Aperture
Decrease of Beam Lifetime Decrease of Beam Lifetime
Short Bunches for Minimizing the Hourglass Effect Short Bunches for Minimizing the Hourglass Effect
Increase of Increase of Touschek Touschek Effect Effect
Decrease of Beam Lifetime Decrease of Beam Lifetime
Higher Frequency Components in the Beam Spectrum Higher Frequency Components in the Beam Spectrum
Possible Coupling with High Frequency Possible Coupling with High Frequency
Vacuum Chamber Modes: Instabilities Vacuum Chamber Modes: Instabilities
Higher Peak RF Voltages: Larger Number of Cavities Higher Peak RF Voltages: Larger Number of Cavities
RF RF Nonlinearities Nonlinearities, Stronger High Order Modes , Stronger High Order Modes
Coherent Synchrotron Radiation Coherent Synchrotron Radiation
with High Current per Bunch with High Current per Bunch
. . . . . . . . . . . .
23
Round Beam vs Flat Beam
Round Beam vs Flat Beam
Colliders &
Luminosity
F. Sannibale
Round Beam (k~1):
Round Beam (k~1):
A Factor 2 of Luminosity Gain A Factor 2 of Luminosity Gain
Both the Beta Functions @ IP Must Be Small: Both the Beta Functions @ IP Must Be Small:
Technically Difficult to Obtain Technically Difficult to Obtain
Large Negative Large Negative Chromaticities Chromaticities in Both Planes in Both Planes
Strong Strong Sextupole Sextupole Correction Correction
Small Dynamic Aperture Small Dynamic Aperture
Strong Beam Strong Beam- -beam Effects beam Effects
Flat Beam (k<<1):
Flat Beam (k<<1):
A Factor 2 of Luminosity Loss A Factor 2 of Luminosity Loss
Chromaticity Handling not Critical: Chromaticity Handling not Critical:
It is Possible to Arrange the It is Possible to Arrange the
Collider Collider Parameters in Order to Parameters in Order to
Obtain Better Luminosity Performances Obtain Better Luminosity Performances
24
Multibunch
Multibunch
Luminosity
Luminosity
Colliders &
Luminosity
F. Sannibale
L = N
B
L
SB
Large Number of Bunches:
Large Number of Bunches:
Separate Rings Separate Rings
Small Distance Between 2 Small Distance Between 2 adiacent adiacent bunches bunches
Multibunch Multibunch Instabilities Instabilities
Low Impedance Vacuum Chamber Low Impedance Vacuum Chamber
HOM Free Ring Components HOM Free Ring Components
Longitudinal Feedback System Longitudinal Feedback System
Horizontal Crossing Angle @ IP Required Horizontal Crossing Angle @ IP Required
in Order to Avoid Parasitic Crossing in Order to Avoid Parasitic Crossing
Synchro Synchro- -betatron betatron Resonances Resonances
. . . . . . . . . .
Larger Stored Current Larger Stored Current
Vacuum System Limitations Vacuum System Limitations
Large RF Power Large RF Power
Vacuum Chamber Large Heating Load Vacuum Chamber Large Heating Load
. . . . . . . . . .
- -
- -
- -
- -
- -
- -
- -
- -
- -
- -
- -
- -
25
Detector Implications
Detector Implications
Colliders &
Luminosity
F. Sannibale
The The Solenoidal Solenoidal Field Introduces Field Introduces Coupling Coupling
between the Vertical and Horizontal Planes between the Vertical and Horizontal Planes
that Must Be Carefully Corrected. that Must Be Carefully Corrected.
Experimental Requirements Concerning Experimental Requirements Concerning
Solid Angle Stay Clear Forced to Have Solid Angle Stay Clear Forced to Have
Permanent IR Permanent IR Quadrupoles Quadrupoles and a Very and a Very
Reduced Configuration of Beam Reduced Configuration of Beam Dignostics Dignostics
26
Detector Effects Compensation:
Detector Effects Compensation:
DA
DA

NE example
NE example
Colliders &
Luminosity
F. Sannibale
IP
KLOE Solenoid
Compensating Solenoids
IR Quadrupoles
S
=
1
2 B
( )
B
z
z
1
z
2
s ( )ds
Solenoid Frame Rotation Angle: Solenoid Frame Rotation Angle:
B
z
Comp.1
s ( )ds + B
z
KLOE
s ( )ds + B
z
Comp.2
s ( )ds = 0
Field Integral Compensation: Field Integral Compensation:
Rotated IR Rotated IR Quadrupoles Quadrupoles to correct Coupling: to correct Coupling:
n
Q
=
1
2 B ( )
B
z
IP
C
n
s ( )ds n =1, 2, 3 C
n
n th quad center position
B
z
= 0. 6 T B ( )= 1. 70 T m
C
1
= 0. 53 m C
2
= 1. 04 m C
3
=1. 59 m
1
Q
= 5. 35 deg
2
Q
= 10. 5 deg
3
Q
=16.1 deg
27
Integrated Luminosity
Integrated Luminosity
Optimization
Optimization
Colliders &
Luminosity
F. Sannibale
-0.1
0
0.1
0.2
0.3
0.4
0.5
-500 0 500 1000 1500 2000 2500 3000 3500
Average Luminosity vs Run Duration
Complete Injection Scheme
<L>/L0 (beam tau 3000 s Td 1200 s)
<L>/L0 (beam tau 4000 s Td 1200 s)
<L>/L0 (beam tau 5000 s Td 1200 s)
<L>/L0 (beam tau 6000 s Td 1200 s)
<
L
>
/
L
o
Run Duration (s)
-0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
-500 0 500 1000 1500 2000 2500 3000 3500
Average Luminosity vs Run Duration
Current Flip Flop Top up Scheme
<L>/L0 (beam tau 3000 s Td 300 s)
<L>/L0 (beam tau 4000 s Td 300 s)
<L>/L0 (beam tau 5000 s Td 300 s)
<L>/L0 (beam tau 6000 s Td 300 s)
<
L
>
/
L
o
Run Duration (s)
28
Examples of
Examples of
Colliders
Colliders
Colliders &
Luminosity
F. Sannibale
29
Examples of
Examples of
Colliders
Colliders
Colliders &
Luminosity
F. Sannibale
KEK KEK
B B- -Factory Factory
1
Lecture No. 1
Lecture No. 1
Historical Overview,
Accelerator Examples and Applications
Accelerator Examples and Applications
David Robin David Robin
2
Content
Content
Examples & Applications
D. Robin
Why Accelerators
History
Examples and Applications
3
Why Accelerators
Why Accelerators
D. Robin
Particle accelerators are devices producing beams
of energetic particles such as ions and electrons
which are employed for many different purposes.
This includes:
Ultra-precise Electron Microscopy
Creation of New Particles
High Brightness Photon Sources for Material Analysis and
Modification, Spectrometry, .
Ion Implanters, for Surface Modification and for Sterilization and
Polymerization
Radiation Surgery and Therapy of Cancer

4
Ultra
Ultra
-
-
Precise
Precise
Electron Microscopy
Electron Microscopy
D. Robin
Probing particles such as electrons and protons provided by particle
accelerators are required for studies of atomic constituents. The
associated de Broglie wavelength of a probing particle defines the
minimum object size that can be resolved.
15
, where
4 10 eVs (Planck's Constant)
(Particle Momentum)
h
p
h
p
=
=
Resolving Smaller Objects Requires Higher Momentum Probe Particles
Example 1 : An electron with a 1 GeV/c momentum will have a de Broglie
wavelength of 10
-15
m (10
-14
m ~ nucleus size, 10
-15
m ~ proton size).
Example 2 : An electron with a 1 keV/c momentum will have a de Broglie
wavelength of ~ 4.0 x 10
-12
m. A photon with 1 keV energy has a wavelength
= ch/e ~ 1.2 x 10
-9
m. This implies ~ 300 times better resolution and shows why
electron microscopes have much better resolution than optical ones.
5
Ultra
Ultra
-
-
Precise
Precise
Electron Microscopy
Electron Microscopy
D. Robin
Sub-micron objects such as the
constituents of a living cell are
often investigated in electron
microscopes where electrons,
accelerated typically to a few
hundred kilovolts, are used to
hit the objects and scatter from
them.
Quarks and leptons
can be sensed down to
distances of
10
-18
meters by means
of particles from giant
accelerators with
particle energies of
~100 GeV.
Objects with dimensions down to the size of a living cell are
investigated by optical microscopes and those down to
atomic dimensions by electron microscopes.
The living cell is commonly
studied by means of an optical
microscope which receives
scattered photons of visible
light.
To penetrate the interiors of atoms and molecules, it is
necessary to use radiation of a wavelength much smaller than
atomic dimensions.
6
D. Robin
Particles from accelerators colliding with target particles may lead to
the creation of new particles, which acquire their mass from the
collision energy according to the formula E=mc
2
. It is thus by
conversion to mass of excess kinetic energy in a collision that
particles, antiparticles and exotic nuclei can be created.
7
Other Applications
Other Applications
D. Robin
Particle accelerators are not only unique as tools for the exploration
of the subatomic world, but are also used in many different
applications such as material analysis and modification and
spectrometry especially in environmental science.
About half of the world's 15,000 accelerators are used as ion
implanters, for surface modification and for sterilization and
polymerization.
The ionization arising when charged particles are stopped in
matter is often utilized for example in radiation surgery and
therapy of cancer. At hospitals about 5,000 electron
accelerators are used for this purpose.
Accelerators also produce radioactive elements that are used
as tracers in medicine, biology and material science.
Of a great and increasing importance in material science are
ion and electron accelerators that produce abundant numbers
of neutrons and photons (light sources) over a wide range of
energies.
8
What Are Accelerators
What Are Accelerators
Used For
Used For
D. Robin
110 Nuclear and particle physics research
70 Synchrotron radiation sources
20 Hadron therapy
200 Medical isotopes production
5,000 Radiotherapy
1,000 Accelerators in non-nuclear research
1,500 Accelerators in industry
7,000 Ion implanters and surface modifications
Number Category
World wide inventory of accelerators, in total 15,000. The
data have been collected by W. Scarf and W. Wiesczycka
(See U. Amaldi Europhysics News, June 31, 2000)
9
History
History
D. Robin
Cathode Ray Tubes
Late 1800s
Multipole Gaps
Cockcroft Walton (1920)
Time Varying Fields
Ising (1924) and Wideroe
Cyclotron
Lawrence (1930)
Van Der Graff (1930)
Alvarez Linac
McMillan (1946)
Synchrotron
Oliphant (1943)
Synchrocyclotron and Betatron
McMillan and Veksler (1944)
Strong Focusing
Courant and Snyder (1952)
Electrostatic Field Based Electrostatic Field Based
Time Varying Field Based Time Varying Field Based
10
Discovery of X
Discovery of X
-
-
Rays
Rays
D. Robin
Wilhelm Conrad
Rntgen (1845-1923)
Bertha
Rntgens Hand
8 Nov, 1895
Modern
radiograph of a
hand
Crookes Tube
Among the most important discovery for medicine.
From:
hyperphysics.phy-astr.gsu.edu
11
Discovery of Electron
Discovery of Electron
D. Robin
In 1896 Joseph John Thomson investigated the nature of the catho In 1896 Joseph John Thomson investigated the nature of the cathode de
rays which were found to be charged and to have a precise charge rays which were found to be charged and to have a precise charge- -to to- -
mass ratio. This discovery of the first elementary particle, the mass ratio. This discovery of the first elementary particle, the electron, electron,
marks the start of a new era, the electronic age. marks the start of a new era, the electronic age.
Original vacuum tube Original vacuum tube
used by J. J. Thomson used by J. J. Thomson
12
Beyond Cathode Ray Tubes
Beyond Cathode Ray Tubes
D. Robin
Cathode Ray Tubes are Single Gap Devices Cathode Ray Tubes are Single Gap Devices
Small Energy (10s of Small Energy (10s of KeV KeV) )
The existing different types of accelerators beyond Cathode The existing different types of accelerators beyond Cathode
Ray Tubes were invented during a time span of nearly Ray Tubes were invented during a time span of nearly
four decades 1920 four decades 1920 - - 1960 1960
Many of the items mentioned here will be discussed in more Many of the items mentioned here will be discussed in more
detail in Lecture 4. detail in Lecture 4.
13
Potential
Potential
-
-
Drop Accelerators
Drop Accelerators
D. Robin
The first high-voltage particle
accelerator had a potential drop of
the order of 100 kilovolts and was
conceived by and named Cockcroft
Walton Accelerator in 1920.
The most common potential-drop
accelerator in use today is named
after its inventor, the American
Robert Jemison Van de Graaff.
Nowadays most van de Graaff
accelerators are commercial devices
and they are available with terminal
voltages ranging between one and
25 million volts (MV)
One of the biggest tandem One of the biggest tandem
accelerators was used for many accelerators was used for many
years at years at Daresbury Daresbury in the United in the United
Kingdom. Its acceleration tube, Kingdom. Its acceleration tube,
placed vertically, was 42 meters long placed vertically, was 42 meters long
and the centre terminal could hold a and the centre terminal could hold a
potential of up to potential of up to 20 million volts 20 million volts. .
Photo: CCLRC Photo: CCLRC
Potential Drop Accelerators Employ Electrostatic Fields
In comparison the potential in clouds just
before they are discharged by lightning is
about 200 MV.
14
Time
Time
-
-
varying voltage
varying voltage
D. Robin
In 1924, the Sweden G. Ising suggested
that the maximum energy could be
increased by replacing the single gap
holding a DC voltage by placing along a
straight line several hollow cylindrical
electrodes holding pulsed voltages.
The Norwegian Rolf Widere realized
that, if the phase of the alternating
voltage changed by 180 degrees during a
particles trip between gaps, the particle
could gain energy in each gap. Based on
this idea he built a three-stage
accelerator for sodium ions.
The principle of repetitive acceleration conceived in the 1920s is an important
milestone in the quest for higher and higher energies. According to this
principle, acceleration is achieved by means of a time-varying voltage instead of
a static voltage as used in e.g. van de Graaff accelerators.
Ising's first suggestion for a linac
Rolf Wideroe
15
Cyclotron
Cyclotron
D. Robin
The first circular accelerator of practical importance based on The first circular accelerator of practical importance based on the the
principle of repetitive acceleration was the cyclotron, invented principle of repetitive acceleration was the cyclotron, invented by by
Ernest Orlando Lawrence. Ernest Orlando Lawrence.
The inventor of the The inventor of the
cyclotron, Ernest cyclotron, Ernest
Orlando Lawrence Orlando Lawrence
(left), and his student (left), and his student
Edwin Edwin Mattison Mattison
McMillan McMillan
In a cyclotron, the charged particles circulate in a strong magn In a cyclotron, the charged particles circulate in a strong magnetic field etic field
and are accelerated by electric fields in one or more gaps. Afte and are accelerated by electric fields in one or more gaps. After having r having
passed a gap, the particles move inside an electrode and are scr passed a gap, the particles move inside an electrode and are screened eened
from the electric field. When the particles exit from the screen from the electric field. When the particles exit from the screened area ed area
and enter the next gap, the phase of the time and enter the next gap, the phase of the time- -varying voltage has varying voltage has
changed by 180 degrees so that the particles are again accelerat changed by 180 degrees so that the particles are again accelerated. ed.
16
Cyclotron
Cyclotron
D. Robin
In 1938 the first European cyclotron at Collge de France in Paris
accelerated a deuteron beam up to 4 MeV and by hitting a target, an intense
source of neutrons was produced.
The first cyclotron
4.5 diameter (1929).
A serious problem with the early cyclotrons was the energy limit of about
10 MeV for the acceleration of protons. This limit depends on the slowing
down of protons rotating in a constant magnetic field due to their
relativistic increase of mass or equivalent total energy.
Lawrence at the 37 cyclotron (1937) Lawrence at the 37 cyclotron (1937)
17
Synchro
Synchro
-
-
Cyclotron
Cyclotron
D. Robin
To overcome the energy limitation of a To overcome the energy limitation of a
cyclotron, cyclotron, the principle of phase stability the principle of phase stability
was invented and proved in 1944/45. The was invented and proved in 1944/45. The
inventors were Vladimir inventors were Vladimir Losifovich Losifovich
Veksler Veksler and by Edwin and by Edwin Mattison Mattison McMillan, McMillan,
a former student of Lawrence, at the a former student of Lawrence, at the
University of California in Berkeley. University of California in Berkeley.
They showed, independently of each They showed, independently of each
other, that by adjusting the frequency other, that by adjusting the frequency
of the applied voltage to the decreasing of the applied voltage to the decreasing
frequency of the rotating protons, it frequency of the rotating protons, it
was possible to accelerate the protons was possible to accelerate the protons
to several hundred to several hundred MeV MeV. .
The largest synchrocyclotron still in use The largest synchrocyclotron still in use
is located in is located in Gatchina Gatchina outside St outside St
Petersburg and it accelerates protons to Petersburg and it accelerates protons to
a a kinetic energy of 1,000 kinetic energy of 1,000 MeV MeV. The iron . The iron
poles are 6 meters in diameter and the poles are 6 meters in diameter and the
whole accelerator weighs 10,000 tons, a whole accelerator weighs 10,000 tons, a
weight weight comparable to that of the Eiffel comparable to that of the Eiffel
Tower Tower. The energies attained correspond . The energies attained correspond
to that of a proton accelerated in a to that of a proton accelerated in a
potential drop of one billion volts. It is potential drop of one billion volts. It is
used for nuclear physics experiments used for nuclear physics experiments
and medical applications. and medical applications.
18
Sector
Sector
-
-
Focusing Cyclotron
Focusing Cyclotron
D. Robin
In the early 1960s, a new type of In the early 1960s, a new type of
cyclotron, the sector cyclotron, the sector- -focusing cyclotron focusing cyclotron
emerged. Iron sectors were introduced in emerged. Iron sectors were introduced in
the pole gap so that an the pole gap so that an azimuthal azimuthal variation variation
of the magnetic field was obtained. This of the magnetic field was obtained. This
azimuthal azimuthal variation provides a strong variation provides a strong
vertical focusing on the circulating beam vertical focusing on the circulating beam
of ions and it is then not necessary to of ions and it is then not necessary to
have the azimuthally averaged field to have the azimuthally averaged field to
decrease with increasing radius as it has decrease with increasing radius as it has
to do in the conventional cyclotron in to do in the conventional cyclotron in
order to maintain vertical focusing. Hence, order to maintain vertical focusing. Hence,
the average magnetic field as a function of the average magnetic field as a function of
radius, can be increased so that the radius, can be increased so that the
rotation frequency of the ion remains rotation frequency of the ion remains
constant in spite of the increase of mass constant in spite of the increase of mass
of the accelerating ion. of the accelerating ion.
The separated sector cyclotron The separated sector cyclotron
in Vancouver, provides 600 in Vancouver, provides 600 MeV MeV
negative hydrogen ions and negative hydrogen ions and it is it is
the largest of all cyclotrons the largest of all cyclotrons. The . The
picture shows the gap inside picture shows the gap inside
which the ions are accelerated. which the ions are accelerated.
19
Sector
Sector
-
-
Focusing Cyclotron
Focusing Cyclotron
D. Robin
Application of Cyclotrons with Heavy Ions: Application of Cyclotrons with Heavy Ions:
Sector focusing cyclotrons have been very useful for providing l Sector focusing cyclotrons have been very useful for providing low ow- -
energy heavy ions. energy heavy ions.
Using accelerated heavy ions, several new elements have been Using accelerated heavy ions, several new elements have been
discovered first in Berkeley and discovered first in Berkeley and Dubna Dubna and later in Darmstadt. The and later in Darmstadt. The
heaviest element so far discovered, element 110, was first found heaviest element so far discovered, element 110, was first found in in
Darmstadt and the discovery has been confirmed by the groups in Darmstadt and the discovery has been confirmed by the groups in
Dubna Dubna and Berkeley. The research is still intense and element 112 has and Berkeley. The research is still intense and element 112 has
been claimed in Darmstadt, element 114 in been claimed in Darmstadt, element 114 in Dubna Dubna. .
Since the maximum energy in a cyclotron is limited by the streng Since the maximum energy in a cyclotron is limited by the strength of th of
the magnetic field and its radial extension, superconducting wir the magnetic field and its radial extension, superconducting wire coils e coils
are now used instead of conventional copper coils around the iro are now used instead of conventional copper coils around the iron n
poles to provide stronger fields. poles to provide stronger fields.
20
Betatron
Betatron
D. Robin
21
Synchrotron
Synchrotron
D. Robin
The two other types of accelerators based on the principle of re The two other types of accelerators based on the principle of repetitive petitive
acceleration, the synchrotron and the linear accelerator, are im acceleration, the synchrotron and the linear accelerator, are important in portant in
elementary particle physics research, where highest possible par elementary particle physics research, where highest possible particle ticle
energies are needed. energies are needed.
In synchrotrons, the particles are accelerated along a ring In synchrotrons, the particles are accelerated along a ring- -shaped orbit and the shaped orbit and the
magnetic fields, bending the particles, increase with time so th magnetic fields, bending the particles, increase with time so that a at a constant orbit constant orbit
is maintained during the acceleration. is maintained during the acceleration.
The synchrotron The synchrotron
concept seems to concept seems to
have been first have been first
proposed in 1943 proposed in 1943
by the Australian by the Australian
physicist physicist Mark Mark
Oliphant. Oliphant.
22
Weak
Weak
-
-
Focusing Synchrotrons
D. Robin
The first synchrotrons were of the so called The first synchrotrons were of the so called weak weak- -focusing type focusing type. The vertical . The vertical
focusing of the circulating particles was achieved by sloping ma focusing of the circulating particles was achieved by sloping magnetic fields, gnetic fields,
from inwards to outwards radii. At any given moment in time, the from inwards to outwards radii. At any given moment in time, the average average
vertical magnetic field sensed during one particle revolution is vertical magnetic field sensed during one particle revolution is larger for larger for
smaller radii of curvature than for larger ones. smaller radii of curvature than for larger ones.
The first synchrotron of this type was the Cosmotron at the Broo The first synchrotron of this type was the Cosmotron at the Brookhaven khaven
National Laboratory, Long Island. It started operation in 1952 a National Laboratory, Long Island. It started operation in 1952 and nd
provided protons with provided protons with energies up to 3 energies up to 3 GeV GeV. .
In the early 1960s, the worlds highest energy weak In the early 1960s, the worlds highest energy weak- -focusing focusing
synchrotron, the synchrotron, the 12.5 12.5 GeV GeV Zero Gradient Synchrotron Zero Gradient Synchrotron (ZGS) started its (ZGS) started its
operation at the Argonne National Laboratory near Chicago, USA. operation at the Argonne National Laboratory near Chicago, USA.
The The Dubna Dubna synchrotron, the largest of them all with a radius of synchrotron, the largest of them all with a radius of 28 meters 28 meters
and with a weight of the magnet iron of and with a weight of the magnet iron of 36,000 tons 36,000 tons
Cosmotron
Weak Focusing
23
Strong
Strong
-
-
D. Robin
In 1952 Ernest D. Courant, Milton Stanley Livingston In 1952 Ernest D. Courant, Milton Stanley Livingston
and Hartland S. Snyder, proposed a scheme for and Hartland S. Snyder, proposed a scheme for
strong focusing of a circulating particle beam so strong focusing of a circulating particle beam so
that its size can be made smaller than that in a weak that its size can be made smaller than that in a weak- -
focusing synchrotron. focusing synchrotron.
In this scheme, the bending magnets are made In this scheme, the bending magnets are made
to have alternating magnetic field gradients; to have alternating magnetic field gradients;
after a magnet with an axial field component after a magnet with an axial field component
decreasing with increasing radius follows one decreasing with increasing radius follows one
with a component increasing with increasing with a component increasing with increasing
radius and so on. radius and so on.
Thanks to the strong focusing, the magnet Thanks to the strong focusing, the magnet
apertures can be made smaller and therefore apertures can be made smaller and therefore
much less iron is needed than for a weak much less iron is needed than for a weak- -
focusing synchrotron of comparable energy. focusing synchrotron of comparable energy.
The first alternating The first alternating- -gradient synchrotron gradient synchrotron
accelerated electrons to 1.5 accelerated electrons to 1.5 GeV GeV. It was built at . It was built at
Cornell University, Ithaca, N.Y. and was Cornell University, Ithaca, N.Y. and was
completed in 1954. completed in 1954.
Size comparison between the
Cosmotron's weak-focusing
magnet (L) and the AGS
alternating gradient focusing
magnets
24
Strong
Strong
-
-
D. Robin
Soon after the invention of the principle of alternating Soon after the invention of the principle of alternating- -gradient focusing, the gradient focusing, the
construction of two nearly identical very large synchrotrons, wh construction of two nearly identical very large synchrotrons, which are ich are still in still in
operation operation, started at the European CERN laboratory in Geneva and the , started at the European CERN laboratory in Geneva and the
Brookhaven National Laboratory on Long Island in New York. Brookhaven National Laboratory on Long Island in New York. At CERN protons At CERN protons
are accelerated to 28 are accelerated to 28 GeV GeV and at Brookhaven to 33 and at Brookhaven to 33 GeV GeV. The CERN proton . The CERN proton
synchrotron (PS) started operation in 1959 and the Brookhaven Al synchrotron (PS) started operation in 1959 and the Brookhaven Alternating ternating
Gradient Synchrotron (AGS) in 1960. Gradient Synchrotron (AGS) in 1960.
Brookhaven AGS CERN PS
25
The Largest Synchrotron
The Largest Synchrotron
D. Robin
Inside the 6.9 km long tunnel of the Inside the 6.9 km long tunnel of the
CERN 450 CERN 450 GeV GeV super proton super proton
synchrotron. The blue magnets synchrotron. The blue magnets
focus, and the red magnets bend the focus, and the red magnets bend the
particles. particles.
Photo: Photo: Cern Cern
Aerial view of the CERN laboratory situated between Aerial view of the CERN laboratory situated between
Geneva airport and the Geneva airport and the Jura Jura mountains. The circles mountains. The circles
indicate the locations of the SPS and LEP accelerators indicate the locations of the SPS and LEP accelerators
placed in underground tunnels. After the LEP placed in underground tunnels. After the LEP
accelerator has stopped operation at the end of the accelerator has stopped operation at the end of the
year 2000, it was dismounted and the large year 2000, it was dismounted and the large Hadron Hadron
Collider Collider (LHC) is currently being installed in the 27 km (LHC) is currently being installed in the 27 km
long tunnel. long tunnel.
Photo: CERN Photo: CERN
26
Linear Accelerators
Linear Accelerators
D. Robin
After After Ising Ising and and Wideroe Wideroe scheme, an improved version of a linear accelerator was scheme, an improved version of a linear accelerator was
conceived in 1946 by Luis Walter Alvarez who generated the radio conceived in 1946 by Luis Walter Alvarez who generated the radio frequency frequency
voltage differently; standing radio voltage differently; standing radio- -frequency waves inside cylindrical cavities. frequency waves inside cylindrical cavities.
These so called Alvarez structures are still used for ion and pr These so called Alvarez structures are still used for ion and proton acceleration. oton acceleration.
Alvarez was awarded the 1968 Nobel Prize in Physics for his deci Alvarez was awarded the 1968 Nobel Prize in Physics for his decisive sive
contributions to elementary particle physics. contributions to elementary particle physics.
KEK
A big boost to the development of linear accelerators came when A big boost to the development of linear accelerators came when Hansen and Hansen and
the Varian brothers (1937) developed the first klystron (frequen the Varian brothers (1937) developed the first klystron (frequencies up to 10 GHz) cies up to 10 GHz)
an efficient and high power source of radio frequency. an efficient and high power source of radio frequency.
In parallel, newer and more efficient RF structures were obtain In parallel, newer and more efficient RF structures were obtained by coupling ed by coupling
together many pillbox together many pillbox- -like cavities. like cavities.
Stanford Linear Accelerator Stanford Linear Accelerator
Very high energy accelerators became a feasible reality and sev Very high energy accelerators became a feasible reality and several machines eral machines
where constructed. where constructed.
27
Accelerator
Accelerator
-
-
Based
Based
Light Sources
Light Sources
D. Robin
Modern synchrotron light sources are accelerators optimized for Modern synchrotron light sources are accelerators optimized for the the
production of synchrotron radiation. production of synchrotron radiation.
ESRF ESRF- - France France
APS APS - - USA USA
More in Lecture 14! More in Lecture 14!
28
Colliding Beams
Colliding Beams
D. Robin
In the continuous race for higher energies, required in the sear In the continuous race for higher energies, required in the search for ch for
undiscovered heavy particles and for the exploration of smaller undiscovered heavy particles and for the exploration of smaller distances, particle distances, particle
colliders colliders have been found to be superior to other types of accelerator have been found to be superior to other types of accelerator- -based based
experiments (fixed target). experiments (fixed target).
The first electron The first electron- -positron positron collider collider in operation was ADA in 1960 in in operation was ADA in 1960 in Frascati Frascati. .
This little storage ring with a little more than 1 m diameter wa This little storage ring with a little more than 1 m diameter was conceived s conceived
and designed by Bruno and designed by Bruno Toshek Toshek and operated at 250 and operated at 250 MeV MeV. ADA was the . ADA was the
proof of principle that allowed to set the theoretical and exper proof of principle that allowed to set the theoretical and experimental basis imental basis
for the later construction of accelerators such as the LEP at CE for the later construction of accelerators such as the LEP at CERN with RN with
almost 27 Km circumference. almost 27 Km circumference.
At the same time, pioneering work on how to collide two beams of At the same time, pioneering work on how to collide two beams of electrons electrons
circulating in two synchrotrons was done in Novosibirsk at the circulating in two synchrotrons was done in Novosibirsk at the Budker Budker
institute. institute.
The first The first collider collider to be used for experiments was the intersecting storage to be used for experiments was the intersecting storage
rings (ISR), used at CERN from 1971 to 1983. rings (ISR), used at CERN from 1971 to 1983.
Several Nobel prizes were assigned for results obtained by accel Several Nobel prizes were assigned for results obtained by accelerators (B. erators (B.
Richter and S. Ting 1976, C. Richter and S. Ting 1976, C. Rubbia Rubbia and S. van and S. van der der Meer Meer 1984) 1984)
29
Other Accelerators
Other Accelerators
D. Robin
The continuous electron beam facility The continuous electron beam facility
(CEBAF) at the Jefferson Laboratory, (CEBAF) at the Jefferson Laboratory,
Virginia, USA, accelerates electrons Virginia, USA, accelerates electrons
up to 6 up to 6 GeV GeV in a race in a race- -track track microtron microtron
with a circumference of 1.4 km. with a circumference of 1.4 km.
Acceleration takes place in 338 Acceleration takes place in 338
hollow shells (cavities) placed in the hollow shells (cavities) placed in the
straight sections inside straight sections inside cryomodules cryomodules
and the beam is bent 180 degrees in and the beam is bent 180 degrees in
five different arcs. During the first five different arcs. During the first
revolution, the electrons move in the revolution, the electrons move in the
upper arcs, they descend upper arcs, they descend
successively and after five successively and after five
revolutions of acceleration they have revolutions of acceleration they have
reached the bottom arcs. Experiments reached the bottom arcs. Experiments
are situated in three different halls, A, are situated in three different halls, A,
B and C. In the future, a new hall D B and C. In the future, a new hall D
will be added and the energy will be will be added and the energy will be
increased to 12 increased to 12 GeV GeV. .
Illustration: DOE/Jefferson Lab. Illustration: DOE/Jefferson Lab.
Superconducting RF Superconducting RF
30
Other Accelerators
Other Accelerators
D. Robin
Cooler Storage Rings Cooler Storage Rings
Cooling a circulating particle beam means reducing the momentum Cooling a circulating particle beam means reducing the momentum
spreads and the transverse dimensions of the beam. spreads and the transverse dimensions of the beam.
Electron cooling was invented in Novosibirsk in the late 1970s a Electron cooling was invented in Novosibirsk in the late 1970s and nd
Electron cooling is useful for improving the quality of beams of Electron cooling is useful for improving the quality of beams of
protons, antiprotons and ions protons, antiprotons and ions
Meson Factories Meson Factories
During the 1960s, three accelerators were built to provide inten During the 1960s, three accelerators were built to provide intense se
fluxes of beams of medium fluxes of beams of medium- -energy, several hundred energy, several hundred MeV MeV, charged p , charged p- -
mesons. mesons.
Neutron Sources Neutron Sources
When a high When a high- -energy proton penetrates a target of heavy material such energy proton penetrates a target of heavy material such
as lead, tungsten or uranium, numerous neutrons are knocked out. as lead, tungsten or uranium, numerous neutrons are knocked out. For For
example, one proton of 800 example, one proton of 800 MeV MeV stopped in a target of uranium gives stopped in a target of uranium gives
rise to about 30 neutrons on the average. rise to about 30 neutrons on the average.
At present, the most powerful pulsed neutron source is located a At present, the most powerful pulsed neutron source is located at the t the
Rutherford Appleton Laboratory near Oxford, U.K., where a 70 Rutherford Appleton Laboratory near Oxford, U.K., where a 70 MeV MeV
linear accelerator is the injector to a synchrotron that provide linear accelerator is the injector to a synchrotron that provides protons s protons
of 800 of 800 MeV MeV with an intensity of 200 microamperes with an intensity of 200 microamperes
31
Conclusion
Conclusion
D. Robin
Exponential growth of energy with time
Exponential growth of energy with time
Increase of the energy by an order of magnitude every 6 Increase of the energy by an order of magnitude every 6- -10 years 10 years
Every new idea evolves up to a point of saturation and than is Every new idea evolves up to a point of saturation and than is
replaced by another new idea replaced by another new idea
Energy is not the only interesting parameters where
Energy is not the only interesting parameters where
there has been phenomenal improvements
there has been phenomenal improvements
Exponential growth in Brightness (for example) of 13 orders of Exponential growth in Brightness (for example) of 13 orders of
magnetude magnetude in only 40 years! in only 40 years!
With clever new ideas these advances will surely
With clever new ideas these advances will surely
continue into the future!
continue into the future!
32
Thanks
Thanks
D. Robin
Wish to thank Y.
Wish to thank Y.
Papaphilippou
Papaphilippou
and
and
N.Catalan
N.Catalan
-
-
Lasheras
Lasheras
for sharing the
for sharing the
tranparencies
tranparencies
that they used in the
USPAS, Cornell University, Ithaca, NY 20th June
1st
1st
July 2005
July 2005
Wish to acknowledge the web based article Accelerators
Wish to acknowledge the web based article Accelerators
and Nobel Laureates by Sven
and Nobel Laureates by Sven
Kullander
Kullander
which can be
which can be
viewed at
viewed at
http://
http://
nobelprize.org/physics/articles/kullander
nobelprize.org/physics/articles/kullander
/
/
1
Lecture No. 2
Lecture No. 2
Maxwell Equations
Maxwell Equations
and Special Relativity in Accelerators
and Special Relativity in Accelerators
2
Need to know how particles will move in the
presence of electric and magnetic fields. Present
a basic review of classical physics*
Equations of Motion
Calculations of the Fields
Special Relativity
Give a couple examples
* Will ignore quantum mechanical effects for now
Motivation
Motivation
Maxwell Equations
& Special Relativity
D. Robin
3
Feynman Lectures by R. Feynman, R.
Leighton, and M Sands
Introduction to Electrodynamics by D.
Griffith
Spacetime Physics by E. Taylor and J.
Archibald
Particle Accelerator Physics, Basic
Principles and Linear Beam Dynamics by
H. Wiedemann
References
References
Maxwell Equations
D. Robin
4
a m F
r
r
=
Newtons Laws of Motion
( ) B v E q F
r
r
r r
+ =
Lorentz Force Equation Force on a charged
particle traveling with velocity, v, in the presence
of an electric, E, or magnetic, B, field
To determine the particle motion one needs to
know the electric and magnetic fields
Maxwells Equations
Equations of Motion
Equations of Motion
Maxwell Equations
D. Robin
5
( ) da A d A =

I. Divergence Theorem
II. Curl Theorem
( ) dI A da A =

Fundamental Theorems
Fundamental Theorems
Maxwell Equations
D. Robin
6
Vectorial
Vectorial
Algebra
Algebra
Changing the
Particle Energy
F. Sannibale
Curl Theorem ( Curl Theorem (Stokes Stokes Theorem) Theorem)
Divergence Theorem Divergence Theorem
Volume Integral Volume Integral
Surface Integral (Flux) Surface Integral (Flux)
Line Integral ( Line Integral (Circuitation Circuitation) )

=
V S
dV F dS n F
( )

=
S l
dS n F l d F
( ) F F = 0
( ) u u = 0
u F F = = 0
or or
( (F F is conservative if curl is conservative if curl F F is zero) is zero)
( ) ( ) F F F F =
2
7
o
E
=
I. Gauss Law
(Flux of E through a closed surface) = (Charge inside/
0
)
0 = B
II. (No Name)
(Flux of B through a closed surface) = 0
2 2 12
Nm / C 10 85 . 8

=
o
permittivity of free space

Maxwells Equations
Maxwells Equations
Maxwell Equations
D. Robin
8
t
B
E
=
III. Faradays Law
(Line Integral of E around a loop) =
-d/dt(Flux of B through the loop)
0 0 0
E
B j
t

= +
IV. (Amperes Law modified by Maxwell)

(Integral of B around a loop) = (Current through the loop)/
0
+d/dt(Flux of E through the loop)
7 2
4 10 / A
o
N

=
permeability of free space
Maxwells Equations
Maxwells Equations
Maxwell Equations
D. Robin
9
2
2
0 0
2
E
E
t

=
In free space combining Gauss Law and Faradays Law

2
2
0 0
2
B
B
t

=
In free space combining Amperes Law and the last Maxwell Equations
2
2
2 2
1
where is the velocity of the wave
f
f v
v t
Equation of a wave in three dimensions is

Maxwells Equations and Light
Maxwell Equations
D. Robin
10
From Maxwell Equations
to Wave Equation
to Wave Equation
Changing the
Particle Energy
F. Sannibale
11
2
2
2 2
1
where is the velocity of the wave
f
f v
v t
Equation of a wave in three dimensions is

8
0 0
1
3.00 10 / v m s

= =
Maxwells equations imply that empty space supports the propagation of
electromagnetic waves traveling at the speed of light
Perhaps Light is an electromagnetic wave
We can scarcely avoid the inference that light consists in the transverse
undulations of the same medium which is the cause of electric and magnetic
phenomena - Maxwell
Maxwell Equations
D. Robin
12
0
0 0 0
1
0
E dS dV
B dS
dB
E dl dS
dt
dE
B dl j dS dS
dt

=
=
=
= +

r r
r r
r
r r r
r
r r r r
r
Maxwells Equations in Integral Form
Maxwells Equations in Integral Form
Maxwell Equations
D. Robin
13
1.The principle of relativity. The laws of
physics apply in all inertial reference systems.
2. The universal Speed of light. The speed of
light in vacuum is the same for all inertial
observers, regardless of the motion of the
source.
Special Relativity
Special Relativity
Maxwell Equations
D. Robin
14
c
v
c
v
=
2
2
1
1
where v is the velocity of the particle and
c is the velocity of light
v m p = : momentum particle where m is the rest mass of
the particle
( ) B v E q
dt
dp
F
r
r
r r
+ = =
Lorentz
Lorentz
Factor,
Factor,
Maxwell Equations
D. Robin
15
Rest Energy, E
o
: E
o
= mc
2
Total Energy, E : E = mc
2
Momentum, p : p = mv
2
2 2 2
0
p
c E E
= +
Energy and Momentum
Energy and Momentum
Maxwell Equations
D. Robin
16
Two particles have equal rest mass m
0
.
Laboratory Frame (LF): one particle at rest, total energy is E.
Centre of Mass Frame (CMF): Velocities are equal and
opposite, total energy is E
cm
.
The quantity is invariant.
In the CMF, we have
In general
In the LF, we have and
And finally
Lorentz
Lorentz
Particle collisions
Particle collisions
Maxwell Equations
D. Robin
17
Two inertial frames moving with respect to each other with
velocity,v
( )
=
=
=
=
c
xv
t t
z z
y y
vt x x
2
'
'
'
'
Lorentz
Lorentz
Transformation
Transformation
Maxwell Equations
D. Robin
18
Two celebrated consequences of the transformation are
Time dilation and Lorentz contraction
Time dilation. A clock in the primed frame located at x = vt will
show a time dilation, t = 1/
Lorentz contraction. An object in the primed frame with
length L along the x axis and is at rest in the primed frame will
be of length L = L/ in the unprimed frame
Time Dilation and
Time Dilation and
Lorentz
Lorentz
Contraction
Contraction
Maxwell Equations
D. Robin
19
( )
( )
B E E
B E E
E E
y z z
z y y
x x
v
v
+ =
=
=
'
'
'
+ =
=
E B B
E B B
B B
y z z
z y y
x x
c
v
c
v
2
'
2
'
'
Lorentz
Lorentz
Transformation of
Transformation of
Electric and Magnetic Fields
Electric and Magnetic Fields
Maxwell Equations
D. Robin
20
Thanks
Thanks
D. Robin
Wish to thank Y.
Wish to thank Y.
Papaphilippou
Papaphilippou
and
and
N.Catalan
N.Catalan
-
-
Lasheras
Lasheras
for sharing the
for sharing the
tranparencies
tranparencies
1st
1st
July 2005
July 2005
21
L2 Possible Homework
Maxwell Equations
D. Robin
Problem 1. Problem 1.
Protons are accelerated to a kinetic energy of 200 Protons are accelerated to a kinetic energy of 200 MeV MeV at the end of the at the end of the Fermilab Fermilab
Alvarez linear accelerator. Calculate their total energy, their Alvarez linear accelerator. Calculate their total energy, their momentum and their momentum and their
velocity in units of the velocity of light. velocity in units of the velocity of light.
A charge A charge pion pion has a rest energy of 139.568 has a rest energy of 139.568 MeV MeV and a mean life time of and a mean life time of = 26.029 = 26.029
nsec nsec in its rest frame. What are the in its rest frame. What are the pion pion life times, if accelerated to a kinetic life times, if accelerated to a kinetic
energy of 20 energy of 20 MeV MeV? And 100 ? And 100 MeV MeV? A ? A pion pion beam decays exponentially like beam decays exponentially like e e
- - /t /t
. At . At
what distance from the source will the what distance from the source will the pion pion beam intensity have fallen to 50%, if beam intensity have fallen to 50%, if
the kinetic energy is 20 the kinetic energy is 20 MeV MeV? Or 100 ? Or 100 MeV MeV? ?
A positron beam accelerated to 50 A positron beam accelerated to 50 GeV GeV in the in the linac linac hits a fixed hydrogen target. hits a fixed hydrogen target.
What is the available energy from a collision with a target elec What is the available energy from a collision with a target electron assumed to be tron assumed to be
at rest? Compare this available energy with that obtained in a l at rest? Compare this available energy with that obtained in a linear inear collider collider where where
electrons and positrons from two similar electrons and positrons from two similar linacs linacs collide head on at the same collide head on at the same
energy. energy.
Rest energy of an electron = 0.511 Rest energy of an electron = 0.511 MeV MeV
Rest energy of a proton = 936 Rest energy of a proton = 936 MeV MeV
22
Maxwell Equations
D. Robin
Show that a function satisfying f(x,t)= Show that a function satisfying f(x,t)=f(x f(x- -vt vt) automatically ) automatically
satisfies the wave equation. satisfies the wave equation.
A A muon muon has a rest mass of 105.7MeV and a lifetime at rest of has a rest mass of 105.7MeV and a lifetime at rest of
2.2e 2.2e- -6 s. 6 s.
Consider a Consider a muon muon traveling at 0.9c with respect to the lab traveling at 0.9c with respect to the lab
frame. What is its lifetime? How far does the frame. What is its lifetime? How far does the muon muon travel? travel?
How does this compare to the distance it would travel if How does this compare to the distance it would travel if
there were no time dilation? there were no time dilation?
Consider a Consider a muon muon accelerated to 1GeV. accelerated to 1GeV.
What is its velocity? How long does it live? What is its velocity? How long does it live?
For a non For a non- -relativistic charge moving in the z direction, relativistic charge moving in the z direction,
calculate the general particle trajectory when subjected to a calculate the general particle trajectory when subjected to a
field field B B
x x
= =B B
z z
=0, and B =0, and B
y y
=sin(2 =sin(2 z/ z/ ) for 0<z<z ) for 0<z<z
0 0
, and B , and B
y y
=0 =0
elsewhere. elsewhere.
1
Lecture No. 3
Lecture No. 3
Charged Particle Optics. Matrix
Charged Particle Optics. Matrix
Representation of the Accelerator Elements
Representation of the Accelerator Elements
2
What are the Optics?
Magnet Definitions
Magnet Functions
Particle motion in accelerator
Coordinate system
Beam guidance
Dipoles
Beam focusing
Quadrupoles
Hills equations and Transport Matrices
Matrix formalism
Drift
Thin lens
Quadrupoles
Dipoles
Sector magnets
Rectangular magnets
Doublet
FODO
Outline
Outline
Single Particle Dynamics
Matrix Representation
D. Robin
3
The Optics are the distribution of elements
(typically magnetic or electrostatic) that
guide and focus the beam - sometimes
referred to as the lattice.
Focusing Elements
Bend Element
D. Robin
4
Choice of the design depends upon the
goal of the accelerator
Small spot size
High brightness
Small divergence
Obey certain physical constraints (building
or tunnel)
Optics or Lattice Design
Optics or Lattice Design
Determines the Beam Properties
Determines the Beam Properties
D. Robin
5
The motion of each charged particle is determined by the
electric and magnetic forces that it encounters as it orbits
the ring:
Lorentz Force
( ),
is the relativistic mass of the particle,
is the charge of the particle,
is the velocity of the particle,
is the acceleration of the part
F ma e E v B
m
e
v
a
= = +
icle,
is the electric field and,
is the magnetic field.
E
B
Equations of Motion
Equations of Motion
D. Robin
6
1. Given an existing lattice, determine the
properties of the beam.
2. For a desired set of beam properties,
determine the design of the lattice.
The first problem is in principle straight-forward
to solve.
The second problem is not straight-forward a
bit of an art.
Two Problems
Two Problems
(Inverse Problems)
(Inverse Problems)
D. Robin
7
Quadrupoles
Sextupoles
Dipoles
Magnets to Guide
Magnets to Guide
and Focus the Beam
and Focus the Beam
D. Robin
8
2n-pole:
dipole quadrupole sextupole octupole
n: 1 2 3 4
Normal: gap appears at the horizontal plane
Skew: rotate around beam axis by /2n angle
Symmetry: rotating around beam axis by /n angle, the field
is reversed (polarity flipped)
N
S
N
S
S
N
N
S S
S
N N
N
N
N
N
S
S
S
S
Magnet Definitions
Magnet Definitions
D. Robin
9
There are several magnet types that are
used in storage rings:
Dipoles used for guiding
B
x
= 0
B
y
= B
o
Quadrupoles used for focussing
B
x
= Ky
B
y
= Kx
Sextupoles used for chromatic
correction
B
x
= 2Sxy
B
y
= S(x
2
y
2
)
Typical Magnet Types
Typical Magnet Types
D. Robin
N
S
N
S
S
N
N
S S
S
N N
10
Functions of the
Functions of the
Magnetic Elements
Magnetic Elements
D. Robin
N
S
N
S
S
N
N
S S
S
N N
11
In the Lorentz Force Equation as written below, the
dependent variable is time, t
Lorentz Force
( ),
is the relativistic mass of the particle,
is the charge of the particle,
is the velocity of the particle,
is the acceleration of the part
F ma e E v B
m
e
v
a
= = +
icle,
is the electric field and,
is the magnetic field.
E
B
Dependent Variable
Dependent Variable
D. Robin
12
Consider a storage ring for
particles with energy E
with N dipoles of length l
The bending angle is
The bending radius is
SNS ring dipole
The integrated dipole strength will be
By fixing the dipole field, the dipole length is imposed and vice versa
The highest the field, shorter or smaller number of dipoles can be used
Ring circumference (cost) is influenced by the field choice
Dipoles
Dipoles
D. Robin
13
Magnetic element that deflects the beam by an angle proportional to the
distance from its centre (equivalent to ray optics) provides focusing.
For a focal length f the deflection angle is
A magnetic element with length l and with a gradient g has a field
so that the deflection angle is
y
f
focal point
The normalised focusing strength
In more practical units, for Z=1
The focal length becomes
and the deflection angle is
Focusing Elements
Focusing Elements
D. Robin
14
Quadrupoles are focusing in one plane and
defocusing in the other
The field is
The resulting force
Need to alternate focusing and defocusing in
order to control the beam, i.e. alternating
gradient focusing
From optics we know that a combination of two
lenses with focal lengths f1 and f2 separated by
a distance d
If f
1
= -f
2
, there is a net focusing effect, i.e.
v
F
B
F
B
v
Quadrupoles
Quadrupoles
D. Robin
15
Coordinate system used to describe the motion is
usually locally Cartesian or cylindrical
Typically the coordinate system chosen is the one that
allows the easiest field representation
y x
s
Change dependent variable from time, t, to longitudinal
position, s
Coordinate System
Coordinate System
D. Robin
16
Integrate through the elements
Use the following coordinates*
*Note sometimes one uses canonical momentum
rather than x and y
0
, ' , , ' , ,
dx dy p L
x x y y
ds ds p L

= = = =
y x
s
Integrate
Integrate
D. Robin
17
The equations of motion within an element is
The fields have to be defined
General equations of motion
General equations of motion
D. Robin
18
The equations become
Inhomogeneous equations with s-dependent coefficients
Note that the term 1/
2
corresponds to the dipole weak
focusing
The term P/(P) is present for off-momentum particles
Equations of motion
Equations of motion
Linear fields
Linear fields
D. Robin
19
Solutions are combination of the ones from the
homogeneous and inhomogeneous equations
Consider particles with the design momentum.
The equations of motion become
with
Hills equations of linear transverse particle motion
Linear equations with s-dependent coefficients (harmonic oscillator
with time dependent frequency)
In a ring or in transport line with symmetries, coefficients are
periodic
Not feasible to get analytical solutions for all accelerator
George Hill
Hills equations
Hills equations
D. Robin
20
u
u
Consider K(s) = k
0
= constant
Equations of harmonic oscillator
with solution
with
for k
0
> 0
for k
0
< 0
Note that the solution can be written in matrix form
Harmonic oscillator
Harmonic oscillator
spring
spring
D. Robin
21
General transfer matrix from s
0
to s
Note that
which is always true for conservative systems
Note also that
The accelerator can be build by a series of matrix multiplications
from s
0
to s
1
from s
0
to s
2
from s
0
to s
3
from s
0
to s
n
S
0
S
1
S
2
S
3
S
n-1
S
n
Matrix formalism
Matrix formalism
D. Robin
22
System with mirror symmetry
S
System with normal symmetry
S
Symmetric lines
Symmetric lines
D. Robin
23
to get a total 4x4 matrix
Combine the matrices for each plane
4x4 Matrices
4x4 Matrices
D. Robin
24
0 L
x
x
xL
s
Consider a drift (no magnetic elements) of length L=s-s
0
Position changes if there is a slope. Slope remains unchanged
Transfer matrix of a drift
Transfer matrix of a drift
D. Robin
25
0 f
x
x
x
x
0
f
Consider a lens with focal length f
Slope diminishes (focusing) or increases (defocusing). Position remains
unchanged
Focusing
Focusing
-
-
defocusing thin lens
defocusing thin lens
D. Robin
26
Consider a quadrupole magnet of length L.
The field is
with normalized quadrupole gradient (in m
-2
)
The transport through a quadrupole is
x
x
0 L s
Quadrupole
Quadrupole
D. Robin
27
Consider a dipole of length L. By setting in the focusing quadrupole matrix
the transfer matrix for a sector dipole becomes
with a bending radius
In the non-deflecting plane
This is a hard-edge model. In fact, there is some edge focusing in the
vertical plane
L
Sector Dipole
Sector Dipole
D. Robin
28
Consider a rectangular dipole of length L. At each edge, the deflecting
angle is
It acts as a thin defocusing lens with focal length
The transfer matrix is with
For <<1, =/2.
In deflecting plane (like drift) in non-deflecting plane (like sector)
Rectangular Dipole
Rectangular Dipole
D. Robin
29
x
L
Consider a quadrupole doublet,
i.e. two quadrupoles with focal
lengths f
1
and f
2
separated by a
distance L.
In thin lens approximation the
transport matrix is
with the total focal length
Setting f
1
= - f
2
= f
Alternating gradient focusing seems overall focusing
This is only valid in thin lens approximation!!!
Quadrupole
Quadrupole
Doublet
Doublet
and AG Focusing
and AG Focusing
D. Robin
30
Consider a defocusing quadrupole
sandwiched by two focusing
quadrupoles with focal lengths f.
The symmetric transfer matrix from
center to center of focusing quads
with the transfer matrices
The total transfer matrix is
L L
FODO Cell
FODO Cell
D. Robin
31
From Gauss law of magnetostatics, we construct a vector potential
Assuming a 2D field in x and y, the vector potential has only one
component A
s
The Amperes law in vacuum (inside the beam pipe)
Using the previous equations one finds the conditions which are
Riemann conditions of an analytic function.
There exist a complex potential of z = x+iy with a
power series expansion convergent in a circle with radius
|z| = r
c
(distance from iron yoke)
x
y
iron
r
c
Magnetic
Magnetic
Multipole
Multipole
Expansion
Expansion
D. Robin
32
From the complex potential we can derive the fields
Setting we have
Define normalized units
on a reference radius, 10
-4
of the main field to get
Note: n=n-1 the American convention
Magnetic
Magnetic
Multipole
Multipole
Expansion
Expansion
D. Robin
33
Thanks
Thanks
D. Robin
Wish to thank Y.
Wish to thank Y.
Papaphilippou
Papaphilippou
and
and
N.Catalan
N.Catalan
-
-
Lasheras
Lasheras
for sharing the
for sharing the
tranparencies
tranparencies
1st
1st
July 2005
July 2005
34
D. Robin
Derive the thin matrix representation for a focusing Derive the thin matrix representation for a focusing quadrupole quadrupole starting from the starting from the
thick element matrix. Hint: calculate the limit for the matrix thick element matrix. Hint: calculate the limit for the matrix when the when the quadrupole quadrupole
length approaches zero while the integrated magnetic field is ke length approaches zero while the integrated magnetic field is kept constant. pt constant.
Suppose that a particle traverses, first, a thin focusing lens Suppose that a particle traverses, first, a thin focusing lens with a focal length with a focal length
F; second, a drift of length L; third, a thin defocusing lens wi F; second, a drift of length L; third, a thin defocusing lens with focal length F; th focal length F;
and, fourth, another drift of length L. Calculate the matrix for and, fourth, another drift of length L. Calculate the matrix for this cell. this cell.
Consider a system made up of two thin lenses each of focal leng Consider a system made up of two thin lenses each of focal length th F F, one , one
focusing and one defocusing, separated by a distance focusing and one defocusing, separated by a distance L L. Show that the system . Show that the system
is focusing if is focusing if |F|>L |F|>L. .
1
Lecture No. 4
Lecture No. 4
Changing the Particle Energy
Changing the Particle Energy
2
Why Accelerate Particles?
Why Accelerate Particles?
Changing the
Particle Energy
F. Sannibale
In both the relativistic and non In both the relativistic and non- -relativistic case accelerating relativistic case accelerating
a particle means to modify (increase) its energy. a particle means to modify (increase) its energy.
In In colliders colliders tuning the tuning the c.m c.m. energy on resonance allows . energy on resonance allows
to create new particles to create new particles
In most of accelerator applications the particle energy In most of accelerator applications the particle energy
is one of the fundamental design parameter and tuning knob: is one of the fundamental design parameter and tuning knob:
In light sources the energy defines the spectrum of the In light sources the energy defines the spectrum of the
emitted radiation emitted radiation
The energy defines the penetration depth of a particle The energy defines the penetration depth of a particle
inside materials (cancer therapy, ) inside materials (cancer therapy, )
In relativistic particles storage rings the energy losses need In relativistic particles storage rings the energy losses need
to be restored in order to keep the particles stored. to be restored in order to keep the particles stored.
3
How to Accelerate Particles?
How to Accelerate Particles?
Changing the
Particle Energy
F. Sannibale
Charged particles: Charged particles: Electric fields mainly. Electric fields mainly.
Neutral particles Neutral particles can be accelerated by: can be accelerated by:
scattering, scattering, spallation spallation
Much more efficient and much more controllable Much more efficient and much more controllable
Large number of schemes and techniques used to Large number of schemes and techniques used to
generate the required electric fields. generate the required electric fields.
Continuous R&D going on Continuous R&D going on
4
Electromagnetic Fields
Electromagnetic Fields
Changing the
Particle Energy
F. Sannibale
Maxwell Equations in vacuum (SI Units Maxwell Equations in vacuum (SI Units differential form): differential form):
dt
B d
E =
dt
E d
J B
0 0 0
+ =
0
= E
0 = B
Time variable Time variable
magnetic fields magnetic fields
are are always always
associated with associated with
electric fields electric fields
(and vice versa) (and vice versa)
Coulombs or Gauss law for electricity Coulombs or Gauss law for electricity
Gauss law for magnetism Gauss law for magnetism
Faradays law Faradays law
Amperes law Amperes law
1831 1831- -1879 1879
5
Lorentz
Lorentz
Equation
Equation
Changing the
Particle Energy
F. Sannibale
B fields can change the trajectory of a particle B fields can change the trajectory of a particle
But But cannot cannot do do work work and thus change its energy and thus change its energy
( ) B v E q F + =
E q F =
l d F W
= ( )

+ = l d B v q l d E q
l d E q W
=
1853 1853- -1928 1928
6
Electric Field Representation
Electric Field Representation
Changing the
Particle Energy
F. Sannibale
( )
=
n
ks t n i
no
e E E
0
( )
( ) ( ) [ ] ks t i ks t E e E E
o
ks t i
o
+ = =

sin cos
( )
( )
=
ks t i
e f dt E

2
1
We will use in our calculations this representation. We will use in our calculations this representation.
Such a representation is quite general. Such a representation is quite general.
In fact, arbitrary electric fields can be represented as: In fact, arbitrary electric fields can be represented as:
Plane wave representation: Plane wave representation:
Periodic Case Periodic Case Non Non- -periodic Case periodic Case
7
From DC to .
From DC to .
Changing the
Particle Energy
F. Sannibale
( ) ks t i
e E E

=

0
In actual accelerators we often deal with a single frequency: In actual accelerators we often deal with a single frequency:
0 2
Hz
3
10 10 2
Hz
11 6
10 10 2
Hz
18 12
10 10 2
Electrostatic Accelerators Electrostatic Accelerators
Induction, Induction, Betatrons Betatrons
Radio Frequency (RF) Radio Frequency (RF)
accelerators accelerators
Laser Laser ponderomotive ponderomotive accel accel. .
Present dominant technology Present dominant technology
8
Electrostatic Accelerators:
The Simplest Scheme
The Simplest Scheme
Changing the
Particle Energy
F. Sannibale
Still one of the most used schemes for electron sources Still one of the most used schemes for electron sources
-
-
-
-
-
-
Cathode
Anode
E
HV
V q W =
HV
V
Budker Budker Institute Institute
Diode Pierce
Geometry
9
The
The
Cockcroft
Cockcroft
-
-
Walton Scheme
Walton Scheme
Changing the
Particle Energy
F. Sannibale
James James Cockcroft Cockcroft and Ernest Walton in and Ernest Walton in
1932 accelerated protons to 800 1932 accelerated protons to 800 keV keV
and produced fission of Lithium in and produced fission of Lithium in
Helium Helium (Nobel Prize 1951) (Nobel Prize 1951)
He Li p 2 +
FERMILAB FERMILAB
Still used as the first Still used as the first
accelerator stage for protons accelerator stage for protons
and ions and ions
V = V
0
sin( t)
2 V
0
4 V
0
V
Out
= 2 V
0
N
CELLS
V
Out
CASCADE GENERATOR (1914) CASCADE GENERATOR (1914)
10
The Van de
The Van de
Graaff
Graaff
Changing the
Particle Energy
F. Sannibale
7MV in 1933 7MV in 1933
~ 20 MV nowadays ~ 20 MV nowadays
The needle transmits the charge to The needle transmits the charge to
the belt by glow discharge and/or the belt by glow discharge and/or
field emission field emission
The electric field inside the sphere The electric field inside the sphere
is zero permitting the passage of the is zero permitting the passage of the
charge from the belt to the sphere charge from the belt to the sphere
The maximum voltage is limited by The maximum voltage is limited by
voltage breakdown. Inert gasses voltage breakdown. Inert gasses
( (Freon Freon, SF6) help. , SF6) help.
11
Van de
Van de
Graaff
Graaff
Accelerator:
Accelerator:
Applications
Applications
Changing the
Particle Energy
F. Sannibale
Tandem Scheme Tandem Scheme
Negative ions (H Negative ions (H
- -
for example) are for example) are
created and accelerated through the created and accelerated through the
first stage first stage
At the end of the first stage the At the end of the first stage the
electrons are stripped out from the electrons are stripped out from the
ions (by a gas target for example) ions (by a gas target for example)
In the second stage the positive ions In the second stage the positive ions
(protons in our example) are (protons in our example) are
accelerated. The net energy gain is accelerated. The net energy gain is
twice twice the voltage of the Van de the voltage of the Van de Graaff Graaff
+
+
+
+
+
+ +
1
st
Stage
2
nd
Stage
12
Low Frequency Accelerators:
Low Frequency Accelerators:
Induction
Induction
Linacs
Linacs
&
&
Betatrons
Betatrons
Changing the
Particle Energy
F. Sannibale
dt
B d
E =
Induction Accelerators allows for very high currents (~ 1kA) at Induction Accelerators allows for very high currents (~ 1kA) at
relatively moderate energies (few relatively moderate energies (few MeV MeV) )
Beam
( ) t I I sin
0
=
Induction
Linac
First Induction Accelerators in ~ 1935 First Induction Accelerators in ~ 1935
Betatron: ring for beta particles (electrons)
Hz 60 50 2
13
EM Fields in Free Space:
The Wave Equation
The Wave Equation
Changing the
Particle Energy
F. Sannibale
z y x i
t
E
c z
E
y
E
x
E
E
i i i i
i
, ,
1
2
2
2 2
2
2
2
2
2
2
=
=
From Maxwell equations it is possible to derive for free space From Maxwell equations it is possible to derive for free space
electromagnetic waves: electromagnetic waves:
2
2
2
2
2
2
2
:
z y x
Laplacian
=
( )
( ) ( )
( )
( ) ( )
( ) [ ]
kz t i kz t i
B
kz t i kz t i
F z z
e e A e e A y x E E
+ +
+ =

,
0
For accelerating particles we need For accelerating particles we need
a non a non- -zero field component in the zero field component in the
z direction (for example) z direction (for example)
2
2
2
2
1
t
E
c
E
z
z
=
Forward wave Forward wave
Backward wave Backward wave
And we will look for a solution in the shape: And we will look for a solution in the shape:
14
Solution in Cylindrical Coordinates
Solution in Cylindrical Coordinates
Changing the
Particle Energy
F. Sannibale
0
1 1
0
2
2
2
2
0
2
2
0
2
0
2
=
|
|
.
|
\
|
+
z
z z z
E k
c
E
r r
E
r r
E
( )
ikt t i
z z
e e r E E

=

,
0
2
2
2
2
1
t
E
c
E
z
z
=
2
2
2
2
2 2
2
2
1 1
z r r r r
( ) ( )

in
z z
e r E r E

=
0 0
~
,
0
~
~
1
~
0
2
2
2
2
2
0
2
0
2
=
|
|
.
|
\
|
+
z
z z
E
r
n
k
c r
E
r r
E
( ) ( ) r k B r k A E
C n C n z
Y J
~
0
+ =
2
2
2
2
k
c
k
C
=

The typical RF accelerating structures present axial symmetry. The typical RF accelerating structures present axial symmetry.
The natural coordinates for this case are the cylindrical coordi The natural coordinates for this case are the cylindrical coordinates. nates.
Again, because of the axial symmetry it is convenient to assume Again, because of the axial symmetry it is convenient to assume that that
the the azimuthal azimuthal component of the field has periodicity component of the field has periodicity n n
Which has as general solution: Which has as general solution:
where where
cutoff cutoff
wavenumber wavenumber
15
Wave Equation General Solution
Wave Equation General Solution
Changing the
Particle Energy
F. Sannibale
( ) ( ) [ ]
ikz t i in
C n C n z
e e e r k B r k A E

+ =

Y J
( ) ( )
( ) ( )
( )
( ) ( )
( ) [ ]
kz t i kz t i
B
kz t i kz t i
F C n z
e e A e e A n r k E
+ +
+ =

cos J
( ) kind the of functions Bessel x
st
n
1 J
( ) kind the of functions Bessel x
nd
n
2 Y
And analogously for the longitudinal component of the magnetic f And analogously for the longitudinal component of the magnetic field: ield:
By using these expressions and the Maxwell curl equations it is By using these expressions and the Maxwell curl equations it is
possible to derive similar expressions for possible to derive similar expressions for E E
r r
, , E E

, , , , B B
r r
and and B B

( ) ( )
( ) ( )
( )
( ) ( )
( ) [ ]
kz t i kz t i
B
kz t i kz t i
F C n z
e e C e e C n r k B
+ +
+ =

cos J
16
Boundary Conditions & Classification
Boundary Conditions & Classification
Changing the
Particle Energy
F. Sannibale
The general solution can now be applied to real geometries. The general solution can now be applied to real geometries.
By imposing the values that the fields assume at the boundaries By imposing the values that the fields assume at the boundaries
( (boundary conditions boundary conditions), the values of the constants in the general ), the values of the constants in the general
solution can be evaluated and the solution for the specific solution can be evaluated and the solution for the specific
geometry is found. geometry is found.
It is useful to classify the possible solutions in the It is useful to classify the possible solutions in the
following categories: following categories:
TM modes TM modes: where the magnetic field is in : where the magnetic field is in
the transverse plane the transverse plane
TE modes TE modes: where the electric field is in : where the electric field is in
the transverse plane the transverse plane
TEM modes TEM modes: where both the electric and : where both the electric and
magnetic fields are in the transverse plane magnetic fields are in the transverse plane
TM TMn n

n n
r r
n n
z z
n n

: periodicity in : periodicity in
n n
r r
: periodicity in r : periodicity in r
n n
z z
: periodicity in z : periodicity in z
Good for Good for
Acceleration! Acceleration!
17
The Cutoff Frequency
The Cutoff Frequency
Changing the
Particle Energy
F. Sannibale
2
2
2
2
C
k
c
k =

From the definition of cutoff From the definition of cutoff
wavenumber wavenumber: :
( )
( ) kz t i
F C z
e A r k E

=

0
J
C C
ck =
Cutoff (angular) frequency Cutoff (angular) frequency
By defining: By defining:
propagates wave the real is 0
2
> > k k
C

lly exponentia decreases and
propagate not does wave the
immaginary is 0
2
< < k k
C

z k
C
on depend not does and propagates not does wave the 0 = =
18
Pill Box Cavity Example
Pill Box Cavity Example
Changing the
Particle Energy
F. Sannibale
E
B
beam
a
L
Pill Box Boundary Conditions: Pill Box Boundary Conditions:
( ) ( ) 0 = = = = a r E a r E
z
( ) ( ) 0 0 0 = = = = z E z E
r
( ) ( ) 0 = = = = L z E L z E
r
( ) ( ) t r k E E
C C
TM
z
cos J
0 0
010
=
For example, for the TM
010
mode:
0
010 010
= =
TM
z
TM
z
E z E
a k
C
405 . 2 =
( ) ( ) t r k E
k c
B
C C
C
TM
sin J
1 0
2
010
=
a c f
C C
2 405 . 2 2 = =
mm a MHz f 5 . 229 500 : Example
0
= =
TM
modes
19
Phase Velocity & Group Velocity:
Phase Velocity & Group Velocity:
Definitions
Definitions
Changing the
Particle Energy
F. Sannibale
Phase Velocity Phase Velocity
( ) kz t E E
z
= cos
0
0 = =
dt
dz
k
dt
d
kz t =
k
v
P
=
dk
d
v
G
=
( ) ( ) ( ) ( )
c
c
c c k c
k
v
C C C
P
>
= =
2 2 2
2
2
1

2 2
k k c
C
+ =
c
k k
c
k k
ck
dk
d
v
C C
G
<
+
=
+
= =
2 2 2 2
1
For propagating waves For propagating waves v v

P P
> > c c No acceleration is possible! No acceleration is possible!
Group Velocity Group Velocity
20
Phase Velocity & Group Velocity
Phase Velocity & Group Velocity
Physical Meaning
Physical Meaning
Changing the
Particle Energy
F. Sannibale
v v
P P
is the speed at which the phase propagates or at is the speed at which the phase propagates or at
which the wave propagates rigidly which the wave propagates rigidly
( ) kz t E E
z
= cos
0
( ) ( ) [ ] ( ) ( ) [ ] ( ) ( ) kz t kz t z k k t z k k t = + + + cos cos 2 cos cos
c
k
v
P
> =
c
dk
d
v
G
< =

v v
G G
is the speed at which the energy propagates or is the speed at which the energy propagates or
a variation of the wave envelope propagates a variation of the wave envelope propagates
For example: wave beating. For example: wave beating.
k
v
P
=
dk
d
k
v
P

21
RF Accelerators:
RF Accelerators:
Wideroe
Wideroe
and Alvarez Schemes
and Alvarez Schemes
Changing the
Particle Energy
F. Sannibale
In 1946 Alvarez overcame to the In 1946 Alvarez overcame to the
inconvenient by including the inconvenient by including the Wideroe Wideroe
structure inside a large metallic tube structure inside a large metallic tube
forming an efficient cavity where the forming an efficient cavity where the
fields were confined. fields were confined.
In 1925 In 1925- -28 28 Ising Ising and and Wideroe Wideroe conceived the first linear accelerator ( conceived the first linear accelerator (linac linac). The ). The
revolutionary device was based on the revolutionary device was based on the drift tubes scheme drift tubes scheme. .
RF i i
T v L
2
1
Synchronicity condition: Synchronicity condition:

At high frequency the At high frequency the Wideroe Wideroe scheme becomes scheme becomes lossy lossy due to electromagnetic due to electromagnetic
radiation. radiation.
During the decelerating half period of the RF, the beam is shiel During the decelerating half period of the RF, the beam is shielded inside the ded inside the
conductive tubes. conductive tubes.
200 MHz RF 200 MHz RF
source from radars source from radars
The Alvarez structures are still widely used as pre The Alvarez structures are still widely used as pre- -accelerator for protons and accelerator for protons and
ions. The particles at few hundred ions. The particles at few hundred keV keV from a from a Cockcroft Cockcroft- -Walton for example, are Walton for example, are
accelerated to few hundred accelerated to few hundred MeV MeV. .
22
Accelerating Structure Evolution
Accelerating Structure Evolution
Changing the
Particle Energy
F. Sannibale
By loading the cylindrical structure by disks By loading the cylindrical structure by disks v v
P P
can be reduced down to can be reduced down to
match the speed of the particle for an efficient acceleration match the speed of the particle for an efficient acceleration
KEK
PARTICLE
v
P
v
- - Traveling wave, constant impedance: Traveling wave, constant impedance: electric field decreases exponentially electric field decreases exponentially
with length. The irises have constant radius. with length. The irises have constant radius.
- - Traveling wave, constant gradient: Traveling wave, constant gradient: the electric field is constant along the the electric field is constant along the
structure. The iris have decreasing radius. structure. The iris have decreasing radius.
- - Standing wave: Standing wave: no wave propagation. The beam transit time in the cell must no wave propagation. The beam transit time in the cell must
be much smaller than the wave period for efficient acceleration be much smaller than the wave period for efficient acceleration
- - Normal conductive or Super conductive Normal conductive or Super conductive
L L- -band ~ 1.5 GHz, S band ~ 1.5 GHz, S- -band ~ 3 GHz, X band ~ 3 GHz, X- -band ~ 11GHz, . band ~ 11GHz, .
Newer and more efficient RF structures were obtained by coupling Newer and more efficient RF structures were obtained by coupling together together
many pillbox many pillbox- -like cavities. like cavities.
23
Energy Gain in RF Structures
Energy Gain in RF Structures
Changing the
Particle Energy
F. Sannibale
Synchronism: Synchronism:
s s KIN
e LP r q E
cos 1
2
0
=
s s KIN
e
LP r q E
cos
1
2
0
=
2
length th cavity wi a For
4
4
sin
0
RF
particle
RF
particle
RF
s KIN
L
v
v
P R q E

=
|
|
.
|
\
|
=
Standing Standing
Wave Wave
Traveling Wave Traveling Wave
Constant Impedance Constant Impedance
Traveling Wave Traveling Wave
Constant Gradient Constant Gradient
PARTICLE
v
P
v
TW: TW:
t
E
z
SW: SW:
( ) ( ) t E r E
C
TM
z
cos 0
0
010
= =
Transit Transit
time factor time factor
24
Frequency Frequency
Preference Preference
Frequency Frequency
Scaling Scaling
PARAMETER PARAMETER
Scaling with Frequency
Scaling with Frequency
Changing the
Particle Energy
F. Sannibale
0
LP r E
s KIN

The table is incomplete. For an exhaustive list see for example: The table is incomplete. For an exhaustive list see for example:
The Stanford Two The Stanford Two- -Mile Accelerator Mile Accelerator (Chapter 6), R.B. Neal Editor (1968). (Chapter 6), R.B. Neal Editor (1968).
Higher Higher
frequencies frequencies
pushes to pushes to
high energy high energy
applications applications
Lower Lower
frequencies frequencies
pushes to pushes to
high current high current
applications applications
Choosing the RF frequency is a very critical step in designing a Choosing the RF frequency is a very critical step in designing an n
accelerator. Several parameters depend in opposite way on the accelerator. Several parameters depend in opposite way on the
frequency and the best match to the application will be a frequency and the best match to the application will be a
frequency that trading between contrasting requirements: frequency that trading between contrasting requirements:
High High

- - 2 2
RF energy stored in the structure RF energy stored in the structure
High High

1/2 1/2
Maximum possible electric field strength (empirical) Maximum possible electric field strength (empirical)
High High

- - 1/2 1/2
Total RF peak power ( Total RF peak power (P P
0 0
) )
High High

1/2 1/2
Shunt impedance per unit length ( Shunt impedance per unit length (r r
s s
) )
Frequency Frequency
Frequency Frequency
Scaling Scaling
PARAMETER PARAMETER
Low Low

- - 1 1
Diameter of beam aperture Diameter of beam aperture
Low Low

- - 2 2
Maximum RF power available for single source Maximum RF power available for single source
Low Low

1/2 1/2
Beam Loading ( Beam Loading (- -dV/di dV/di) )
Low Low

- - 1/2 1/2
Peak beam current at maximum conversion efficiency Peak beam current at maximum conversion efficiency
High High

- - 2 2
RF energy stored in the structure RF energy stored in the structure
High High

1/2 1/2
Maximum possible electric field strength (empirical) Maximum possible electric field strength (empirical)
High High

- - 1/2 1/2
Total RF peak power ( Total RF peak power (P P
0 0
) )
High High

1/2 1/2
Shunt impedance per unit length ( Shunt impedance per unit length (r r
s s
) )
Frequency Frequency
Frequency Frequency
Scaling Scaling
PARAMETER PARAMETER
25
Large and Small
Large and Small
Linear Accelerators
Linear Accelerators
Changing the
Particle Energy
F. Sannibale
The combined development and optimization of the described RF s The combined development and optimization of the described RF structures tructures
jointly to the development of powerful and efficient RF sources jointly to the development of powerful and efficient RF sources (klystrons), (klystrons),
permitted the ambitious design and construction of large and ver permitted the ambitious design and construction of large and very high y high
energy linear accelerators. energy linear accelerators.
In the International Linear In the International Linear Collider Collider (ILC) (ILC)
project, electron and positron project, electron and positron linacs linacs longer than longer than
30 Km and with energies over 500 30 Km and with energies over 500 GeV GeV are are
under consideration. under consideration.
Above all, we want to mention the 3 Above all, we want to mention the 3- -km linear accelerator that started km linear accelerator that started
operating in 1966 at the Stanford Linear Accelerator Center and operating in 1966 at the Stanford Linear Accelerator Center and that is that is
capable of accelerating electron and positrons up to more than 5 capable of accelerating electron and positrons up to more than 50 0 GeV GeV, with , with
an average gradient in the RF structure of ~ 17 an average gradient in the RF structure of ~ 17 MeV/m MeV/m. .
At the same time, much smaller At the same time, much smaller linacs linacs from few from few MeV MeV to few hundred to few hundred MeV MeV are are
the backbone of the injector in most existing electron acceler the backbone of the injector in most existing electron accelerators. ators.
R&D on higher frequency RF structures is R&D on higher frequency RF structures is
demonstrating gradients larger than 100 demonstrating gradients larger than 100 MeV/m MeV/m. .
26
Linear Accelerators
Linear Accelerators
vs. Circular Accelerators
vs. Circular Accelerators
Changing the
Particle Energy
F. Sannibale
RF cavity technology allowed the development of both linear and RF cavity technology allowed the development of both linear and
circular accelerators. circular accelerators.
The main advantage of circular accelerators is that a single ca The main advantage of circular accelerators is that a single cavity, vity,
where the beam passes many times guided by the confinement actio where the beam passes many times guided by the confinement action n
of magnetic fields, is capable of very high energy acceleration. of magnetic fields, is capable of very high energy acceleration. This is This is
a very efficient scheme where only a relatively small amount of a very efficient scheme where only a relatively small amount of RF RF
power is required. power is required.
Unfortunately, for light particles the emission of synchrotron Unfortunately, for light particles the emission of synchrotron radiation radiation
can limit the maximum energy achievable (~ 100 can limit the maximum energy achievable (~ 100 GeV GeV for electrons). for electrons).
In general, circular accelerators are more efficient with heavy In general, circular accelerators are more efficient with heavy
particles and medium energy electrons, while linear accelerators particles and medium energy electrons, while linear accelerators are are
preferred with high energy electrons. preferred with high energy electrons.
Efficiency is not all. For example, circular machines usually s Efficiency is not all. For example, circular machines usually show how
more stable beam characteristics while the beam more stable beam characteristics while the beam emittance emittance can be can be
(maintained) smaller in linear accelerators. Different applicati (maintained) smaller in linear accelerators. Different applications can ons can
find their best match in either one or the other schemes. find their best match in either one or the other schemes.
27
Cyclotron and
Cyclotron and
Synchro
Synchro
-
-
cyclotron
cyclotron
Changing the
Particle Energy
F. Sannibale
1939 Nobel Prize
Proton Source
Uniform
Magnetic
Field
Accelerated
Protons
Electric
Field
c v for
eB
m
veB
mv
veB
p
v
r
T
R
<< = = = =
2 2 2 2
For non For non- -relativistic particles relativistic particles
the revolution period the revolution period
does not depend on energy does not depend on energy
If the RF frequency is equal to the particles revolution freque If the RF frequency is equal to the particles revolution frequency synchronicity is ncy synchronicity is
obtained and acceleration is achieved. obtained and acceleration is achieved.
The The synchro synchro- -cyclotron is a variation that allows acceleration also of relati cyclotron is a variation that allows acceleration also of relativistic vistic
particles. The RF frequency is dynamically changed to match particles. The RF frequency is dynamically changed to match
the changing revolution frequency of the particle the changing revolution frequency of the particle
The first cyclotron
4.5 diameter (1929).
In an uniform magnetic field: In an uniform magnetic field:
In 1946 Lawrence built in Berkeley the 184 In 1946 Lawrence built in Berkeley the 184 synchro synchro- -cyclotron with an orbit radius cyclotron with an orbit radius
of 2.337 m and capable of 350 of 2.337 m and capable of 350 MeV MeV protons. The largest cyclotron still in operation is protons. The largest cyclotron still in operation is
in in Gatchina Gatchina and accelerates protons to up 1 and accelerates protons to up 1 GeV GeV for nuclear physics experiments. for nuclear physics experiments.
E. O. Lawrence
28
Microtrons
Microtrons
Changing the
Particle Energy
F. Sannibale
integer an is n n =
Synchronicity condition Synchronicity condition
(energy gain per turn) (energy gain per turn)
0.511 0.511 MeV MeV for electrons for electrons
938 938 MeV MeV for protons for protons
Useful only for accelerating electrons. Useful only for accelerating electrons.
The maximum energy is~ 30 The maximum energy is~ 30 MeV MeV (limited by the magnet size) (limited by the magnet size)
B
magnet
RF
cavity
magnetic
shield
Electron
source
Veksler Veksler
1944 1944
Moroz Moroz and Roberts 1958 and Roberts 1958
29
Synchrotrons &
Synchrotrons &
Storage Rings
Storage Rings
Changing the
Particle Energy
F. Sannibale
APS APS
Achieving higher energies in cyclotrons Achieving higher energies in cyclotrons
requires very large magnets. requires very large magnets.
Above ~ 400 Above ~ 400 MeV MeV the realization of cyclotrons the realization of cyclotrons
becomes inconvenient and expensive becomes inconvenient and expensive
A storage ring is a synchrotron were the particles are not acce A storage ring is a synchrotron were the particles are not accelerated lerated
but just stored at a fixed energy for a relatively long time. but just stored at a fixed energy for a relatively long time.
Colliders Colliders, synchrotron light sources, , synchrotron light sources,
on accelerati during change must
beam ic relativist - non for
2
0
0
RF
RF
f
m
B e Z
h f h f = =

The synchronicity condition is given by: The synchronicity condition is given by:
Synchrotrons have achieved energy as high as Synchrotrons have achieved energy as high as
100 100 GeV GeV for electrons and 1000 for electrons and 1000 GeV GeV for protons for protons
In a synchrotron the radius is fixed and all the In a synchrotron the radius is fixed and all the
fields can be confined only around the orbit. fields can be confined only around the orbit.

to scale must constant
0
= = B
ZeB
c m
R
30
RF Sources: The Klystron
RF Sources: The Klystron
Changing the
Particle Energy
F. Sannibale
Invented by Hansen Invented by Hansen
and the Varian brothers and the Varian brothers
in 1937 in 1937
Very powerful source Very powerful source
from ~ 100 MHz to more from ~ 100 MHz to more
than 10 GHz than 10 GHz
Widely used in all kinds of accelerators Widely used in all kinds of accelerators
31
R&D on Future
R&D on Future
Accelerator Schemes
Accelerator Schemes
Changing the
Particle Energy
F. Sannibale
The R&D on new acceleration techniques is extremely active and The R&D on new acceleration techniques is extremely active and addressed addressed
towards a very large variety of new accelerating techniques. Res towards a very large variety of new accelerating techniques. Results are ults are
promising especially from the accelerating gradient point of vie promising especially from the accelerating gradient point of view were extremely w were extremely
high values have been already obtained. high values have been already obtained.
Among the techniques under study, Among the techniques under study,
here we want to mention as an here we want to mention as an
example, the one based on the so example, the one based on the so- -
called called laser laser wakefield wakefield acceleration acceleration. .
Laser
Gas
Gas jet nozzle
e
-
bunch
Plasma
channel
LOASIS LOASIS
LBL LBL
A high intensity laser is focused on an A high intensity laser is focused on an
atomic gas jet. atomic gas jet.
The laser ionizes most of the atoms The laser ionizes most of the atoms
creating a plasma and also stimulating a creating a plasma and also stimulating a
resonant motion of the electrons in the resonant motion of the electrons in the
plasma. plasma.
This electron motion breaks the charge balance inside the very This electron motion breaks the charge balance inside the very dense plasma dense plasma
inducing extremely high gradients in the plasma area surrounding inducing extremely high gradients in the plasma area surrounding the laser. the laser.
Electrons in the plasma can find the right phase and can be acce Electrons in the plasma can find the right phase and can be accelerated to high lerated to high
energies. Gradients of many tens of energies. Gradients of many tens of GeV/m GeV/m in few mm have been already in few mm have been already
demonstrated. demonstrated.
32
The Cyclotron:
The Cyclotron:
Different Points of View
Different Points of View
Changing the
Particle Energy
F. Sannibale
By Dave Judd and By Dave Judd and Ronn Ronn MacKenzie MacKenzie
From LBNL Image Library From LBNL Image Library
Collection Collection
the operator
33
Possible Homework
Possible Homework
Changing the
Particle Energy
F. Sannibale
Prove that the magnetic field cannot do work Prove that the magnetic field cannot do work
Explain how the cascade generator in the Explain how the cascade generator in the Cockcroft Cockcroft- -Walton Walton
schemes works schemes works
Explain why the electric field inside the Van de Explain why the electric field inside the Van de Graaff Graaff sphere sphere
is zero is zero
Show that for free space electromagnetic waves E and B vectors a Show that for free space electromagnetic waves E and B vectors are mutually re mutually
orthogonal and have no component in the wave propagation directi orthogonal and have no component in the wave propagation direction. on.
Show also the relation between their modules. Show also the relation between their modules.
Calculate the maximum energy gain in Calculate the maximum energy gain in MeV MeV vs. input power in MW for a 3.048 vs. input power in MW for a 3.048
m long constant gradient accelerating structure with shunt imped m long constant gradient accelerating structure with shunt impedance for unit ance for unit
length length r r
s s
= 53 M = 53 M /m and attenuation factor /m and attenuation factor = 0.57. Calculate also the power = 0.57. Calculate also the power
required for accelerating relativistic electrons to 60 required for accelerating relativistic electrons to 60 MeV MeV
Derive the expression for the transit time factor for a pillbox Derive the expression for the transit time factor for a pillbox resonating in its resonating in its
TM TM
010 010
mode mode
Calculate the internal diameter of the external pipe of a 200 M Calculate the internal diameter of the external pipe of a 200 MHz Alvarez Hz Alvarez
structure operating in the TM010 mode. structure operating in the TM010 mode.
34
Vectorial
Vectorial
Algebra
Algebra
Changing the
Particle Energy
F. Sannibale
Curl Theorem ( Curl Theorem (Stokes Stokes Theorem) Theorem)
Divergence Theorem Divergence Theorem
Volume Integral Volume Integral
Surface Integral (Flux) Surface Integral (Flux)
Line Integral ( Line Integral (Circuitation Circuitation) )

=
V S
dV F dS n F
( )

=
S l
dS n F l d F
( ) F F = 0
( ) u u = 0
u F F = = 0
or or
( (F F is conservative if curl is conservative if curl F F is zero) is zero)
( ) ( ) F F F F =
2
35
to Wave Equation
to Wave Equation
Changing the
Particle Energy
F. Sannibale
1
Lecture No. 5
Lecture No. 5
Phase Space Representation.
Phase Space Representation.
Ensemble of Particles,
Ensemble of Particles,
Emittance
Emittance
.
.
2
From Single Particle to a Beam
From Single Particle to a Beam
Phase Space
& Emittance
F. Sannibale
The number of particles per bunch in most accelerators is The number of particles per bunch in most accelerators is
typically included between 10 typically included between 10
5 5
to 10 to 10
13 13
. .
Integrating the particle motion for such a large number of Integrating the particle motion for such a large number of
particles along accelerators with length ranging from few particles along accelerators with length ranging from few
meters up to tens of kilometers can prove to be a tough task. meters up to tens of kilometers can prove to be a tough task.
Fortunately, Fortunately, statistical mechanics statistical mechanics gives us very developed gives us very developed
tools for representing and dealing with sets of large number tools for representing and dealing with sets of large number
of particles. of particles.
Quite often, the statistical approach can give us elegant and Quite often, the statistical approach can give us elegant and
powerful insights on properties that could be hard to extract powerful insights on properties that could be hard to extract
by approaching the set using single particle techniques. by approaching the set using single particle techniques.
3
A Convenient Reference Frame
A Convenient Reference Frame
Phase Space
& Emittance
F. Sannibale
In accelerators we are interested in studying particles along t In accelerators we are interested in studying particles along their heir
trajectory. A natural choice is to refer all the particles relat trajectory. A natural choice is to refer all the particles relatively to a ively to a
reference trajectory reference trajectory . .
In each point of this trajectory we can define In each point of this trajectory we can define
(for example) a Cartesian frame moving with (for example) a Cartesian frame moving with
the reference particle . the reference particle .
In this frame the reference particle is always In this frame the reference particle is always
at the origin and its momentum is always at the origin and its momentum is always
parallel to the direction of the parallel to the direction of the z z axis. axis.
The coordinates { The coordinates {x x, , y y, , z z} for an arbitrary particle represent its } for an arbitrary particle represent its
displacement relatively to the reference particle along the thre displacement relatively to the reference particle along the three e
directions. directions.
In the lab frame the particle moves on the curvilinear coordina In the lab frame the particle moves on the curvilinear coordinate te s s
with speed with speed ds ds/ /dt dt. .
Such a trajectory is assumed to be the solution of the Such a trajectory is assumed to be the solution of the Lorentz Lorentz equation equation
for the particle with the nominal parameters (reference particle for the particle with the nominal parameters (reference particle). ).
z
y
x
Reference Reference
trajectory trajectory
y x z =
4
Phase Space Representation
Phase Space Representation
Phase Space
& Emittance
F. Sannibale
In relativistic classical mechanics, the motion of a single In relativistic classical mechanics, the motion of a single
particle is totally defined when, at a given instant particle is totally defined when, at a given instant t, t, the the
position position r r and the momentum and the momentum p p of the particle are given of the particle are given
together with the forces acting on that position. together with the forces acting on that position.
z z y y x x r
i i i i
+ + = z p y p x p p
zi yi xi
+ + =
z F y F x F F
z y x
+ + =
It is quite convenient to use the so It is quite convenient to use the so- -called called phase space phase space
representation, a 6 representation, a 6- -D space where the D space where the i i
th th
particle assumes the particle assumes the
coordinates: coordinates:
{ }
zi i yi i xi i i
p z p y p x P , , , , ,
In most accelerator physics calculations, the three planes can b In most accelerator physics calculations, the three planes can be e
considered with very good approximation as decoupled. considered with very good approximation as decoupled.
In this situation, it is possible and convenient to study the pa In this situation, it is possible and convenient to study the particle rticle
evolution independently in each of the planes: evolution independently in each of the planes:
{ }
xi i
p x , { }
yi i
p y , { }
zi i
p z ,
5
Particles Systems & Ensembles
Particles Systems & Ensembles
Phase Space
& Emittance
F. Sannibale
w
p
w
( )
w
p w P ,
1
w
p
w
w
p
w
The phase space can now be used for representing particles: The phase space can now be used for representing particles:
The set of possible states for a system of The set of possible states for a system of N N particles particles
is referred as is referred as ensemble ensemble in statistical mechanics. in statistical mechanics.
In the statistical approach, the particles lose their In the statistical approach, the particles lose their
individuality. The properties of the whole system individuality. The properties of the whole system
as individual entity are now studied. as individual entity are now studied.
Important properties of the density functions can now be derived Important properties of the density functions can now be derived. .
Under particular circumstances, such properties allow to calcul Under particular circumstances, such properties allow to calculate the ate the
time evolution of the particle system without going through the time evolution of the particle system without going through the
integration of the motion for each single particle. integration of the motion for each single particle.
( )
z y x z y x D
dp dz dp dy dp dx p z p y p x f , , , , ,
6
( ) z y x w dp dw p w f
w w D
, , ,
2
=
The above expressions indicate the number of particles contained The above expressions indicate the number of particles contained in the in the
elementary volume of phase space for the 6D and 2D cases respect elementary volume of phase space for the 6D and 2D cases respectively. ively.
z y x w , , =
z y x w N dp dw f
w D
, ,
2
= =
N dp dz dp dy dp dx f
z y x D
=
6
The system is fully represented by the density of particles The system is fully represented by the density of particles f f
6 6D D
and and f f
2 2D D
: :
6
Hamiltonian Systems
Hamiltonian Systems
Phase Space
& Emittance
F. Sannibale
Non Non- -Hamiltonian Forces: Hamiltonian Forces:
Stochastic processes (collisions, quantum emission, diffusion, Stochastic processes (collisions, quantum emission, diffusion, ) )
Inelastic processes (ionization, fusion, fission, annihilation, Inelastic processes (ionization, fusion, fission, annihilation, ) )
Dissipative forces (viscosity, friction, ) Dissipative forces (viscosity, friction, )
A system of variables A system of variables q q (generalized coordinates) and (generalized coordinates) and p p (generalized (generalized
momenta momenta) is Hamiltonian when exists a function ) is Hamiltonian when exists a function
H H( (q q, , p p, , t t) that allows to describe the evolution of the system by: ) that allows to describe the evolution of the system by:
The function The function H H is called is called Hamiltonian Hamiltonian and and q q and and p p are referred as are referred as
canonical conjugate variables. canonical conjugate variables.
i
i
p
H
dt
dq
=
i
i
q
H
dt
dp
=
In the particular case that In the particular case that q q are the usual spatial coordinates { are the usual spatial coordinates {x x, , y y, , z z} and } and
p their conjugate p their conjugate momenta momenta { {p p
x x
, , p p
y y
, , p p
z z
}, }, H H coincides with the total energy of coincides with the total energy of
the system: the system:
Energy Kinetic Energy Potential + = + = T U H
{ }
{ } , ,...., ,
, ,...., ,
2 1
2 1
N
N
p p p p
q q q q
7
The Continuity Equation
The Continuity Equation
Phase Space
& Emittance
F. Sannibale
If there is a flow of matter going inside the If there is a flow of matter going inside the
volume then the density inside the volume volume then the density inside the volume
must increase in order to must increase in order to conserve the mass conserve the mass. .
( ) dz dy dx dV volume the in mass dz dy dx t z y x = , , ,
dt dS n v dS dt v dm
n
= =
dS n v
dt
dm
=

=
S
dS n v
dt
dM
=
V
dV M

=
S V
dS n v dV
dt
d

=
V S
dV F dS n F

=
V S
dV v dS n v

=
V V
dV v dV
dt
d

0 = +
v
t
n
v
n v v
n
=
dS
The The continuity equation continuity equation is a consequence is a consequence
of the conservation law of the conservation law
Let the density Let the density : :
But it is also true that: But it is also true that:
8
The
The
Liouville
Liouville
Theorem
Theorem
Phase Space
& Emittance
F. Sannibale
1809-1882
( ) { }
x x D
p x v p x f

, and , Let
2
0 = +
v
t
( ) ( )
x
x
f
p
p
f p
p
f
x
x
f
t
f
p
f p
x
f x
t
f
v f
t
f
D
x
x
D x
x
D D D
x
D x D D
D
D
= +

2 2
2 2 2 2 2 2
2
2
0
2
2
2
2 2 2
=

x
D
x
D
x
x
D D
p x
H
f
p x
H
f
p
p
f
x
x
f

t d
df
p
p
f
x
x
f
t
f
v f
t
f
D
x
x
D D D
D
D 2 2 2 2
2
2
=
= +

0
2
=
t d
df
D
Let us use the continuity equation with our phase space distribu Let us use the continuity equation with our phase space distributions. tions.
For simplicity we will use the 2D distribution, but the same exa For simplicity we will use the 2D distribution, but the same exact results ct results
apply to the more general 6D case. apply to the more general 6D case.
But our system is Hamiltonian But our system is Hamiltonian
Liouville Liouville Theorem Theorem: The phase space density for : The phase space density for
a Hamiltonian system is an invariant of the a Hamiltonian system is an invariant of the
motion. Or equivalently, the phase space motion. Or equivalently, the phase space
volume occupied by the system is conserved. volume occupied by the system is conserved.
9
Decoupling the Problem:
Decoupling the Problem:
the Longitudinal Phase Space
the Longitudinal Phase Space
Phase Space
& Emittance
F. Sannibale
0
0
E
E E
=
In most of existing accelerators the phase space planes are wea In most of existing accelerators the phase space planes are weakly kly
coupled. In particular, we can treat the longitudinal plane inde coupled. In particular, we can treat the longitudinal plane independently pendently
from the transverse one in the large majority of the cases. from the transverse one in the large majority of the cases.
In the longitudinal plane we apply our electric fields for acce In the longitudinal plane we apply our electric fields for accelerating the lerating the
particles and changing their energy. particles and changing their energy.
It becomes natural to use It becomes natural to use energy energy as one of the longitudinal plane as one of the longitudinal plane
variable together with its canonical conjugate variable together with its canonical conjugate time time. .
The effects of the weak coupling can be then investigated as a The effects of the weak coupling can be then investigated as a
perturbation of the uncoupled case. perturbation of the uncoupled case.
In accelerator physics, the In accelerator physics, the relative energy variation relative energy variation and the and the relative relative
time distance time distance with respect to a reference particle are often used: with respect to a reference particle are often used:
0
t t =
According to According to Liouville Liouville, in the presence of Hamiltonian forces, the area , in the presence of Hamiltonian forces, the area
occupied by the beam in the longitudinal phase space is conserve occupied by the beam in the longitudinal phase space is conserved. d.
More in Lecture 8. More in Lecture 8.
10
The Transverse Phase Space
The Transverse Phase Space
Phase Space
& Emittance
F. Sannibale
ds
dx
x p
xi
=
For the transverse planes { For the transverse planes {x x, , p p

x x
} and { } and {y y, , p p
y y
}, it is usually used a slightly }, it is usually used a slightly
modified phase space where the momentum components are replaced modified phase space where the momentum components are replaced by: by:
ds
dy
y p
yi
=
The physical meaning of the new variables: The physical meaning of the new variables:
x c m
ds
dx
v m
dt
dx
m p
s x
= = =
0 0 0

y
ds
dy
y tan = =
w
s
PROJECTION WS
p
w
x
ds
dx
x tan = =
y c m p
y
=
0

( )
2 1
2
1
= = and
c
v
where
s
Note that Note that x x and and p p
x x
are canonical conjugate variables while are canonical conjugate variables while x x and and x x are are
not unless there is no acceleration ( not unless there is no acceleration ( and and constant) constant)
y x w , =
The relation between this new variables and the The relation between this new variables and the
momentum (when momentum (when B B
z z
= 0) is: = 0) is:
11
Definition of
Definition of
Emittance
Emittance
Phase Space
& Emittance
F. Sannibale
We will consider the decoupled case and use the We will consider the decoupled case and use the
{ { w w, , w w } plane where } plane where w w can be either can be either x x or or y y. .
x
x
We define as We define as emittance emittance the phase space area the phase space area
occupied by the system of particles, divided by occupied by the system of particles, divided by
y x w
A
w w
w
, = =
As we previously shown, As we previously shown, x x and and y y are conjugate to are conjugate to x x and and y y when when B B
z z
= 0 = 0
and in absence of acceleration. In this case, we can immediately and in absence of acceleration. In this case, we can immediately apply apply
the the Liouville Liouville theorem and state that for such a system the theorem and state that for such a system the emittance emittance is is
an invariant of the motion an invariant of the motion. .
This specific case is actually extremely important. This specific case is actually extremely important.
In fact, for most of the elements in a beam In fact, for most of the elements in a beam transferline transferline, such as dipoles, , such as dipoles,
quadrupoles quadrupoles, , sextupoles sextupoles, , the above conditions apply and the , , the above conditions apply and the
emittance emittance is conserved. is conserved.
12
Emittance
Emittance
Conservation in
Conservation in
The Presence of
The Presence of
B
B
z z
Phase Space
& Emittance
F. Sannibale
When the When the B B
z z
component of the magnetic field is present ( component of the magnetic field is present (solenoidal solenoidal
lenses for example), the transverse planes become coupled and th lenses for example), the transverse planes become coupled and the e
phase space area occupied by the system in each of the transvers phase space area occupied by the system in each of the transverse e
planes is not conserved anymore. planes is not conserved anymore.
Anyway in this situation, the Anyway in this situation, the Liouville Liouville theorem still applies to the 4D theorem still applies to the 4D
transverse phase space where the transverse phase space where the ipervolume ipervolume occupied by our system is occupied by our system is
still a motion invariant. still a motion invariant.
Actually, if we rotate the spatial reference frame around the Actually, if we rotate the spatial reference frame around the z z axis by the axis by the
Larmor Larmor frequency frequency
L L
= = qB qB
z z
/ / 2 2 m m
0 0
, then the planes become decoupled and , then the planes become decoupled and
the phase space area in each of the planes is conserved again. the phase space area in each of the planes is conserved again.
z
x
13
Emittance
Emittance
Conservation in
Conservation in
The Presence of Acceleration
The Presence of Acceleration
Phase Space
& Emittance
F. Sannibale
When the particles in a beam undergo to acceleration, When the particles in a beam undergo to acceleration, and and change change
and the variables and the variables x x and and x x are not canonical anymore. are not canonical anymore. Liouville Liouville theorem theorem
does not apply and the does not apply and the emittance emittance is not conserved. is not conserved.
The last expression tells us that the quantity The last expression tells us that the quantity is a system invariant is a system invariant
during acceleration. By defining the during acceleration. By defining the normalized normalized emittance emittance: :
c m
p
p
p
y
y
z
y
0
0
tan

= = =
c m
p
p
p
y
y
z
y
0 0 0
0
0
0
0 0
tan

= = =

0 0
0
=
y
y
0 0
case in this that shown be can It
y
y
y
y
0 0 0 y y
=
0 y
p
0 z
p
0
0
p 0 y
p
0
2
0 0
2
0
2
0
2
2
2
z z
p
c m T T
c Tm T
p
+
+
=
p
Accelerated by E
z
Accelerated by E
z
y x w
w w n
, = =
We can say that the We can say that the normalized normalized emittance emittance is conserved during acceleration is conserved during acceleration. .
The acceleration couples the longitudinal plane with the transve The acceleration couples the longitudinal plane with the transverse one: rse one:
the 6D the 6D emittance emittance is still conserved but the transverse is not. is still conserved but the transverse is not.
14
The Real Beam Case:
The Real Beam Case:
r.m.s
r.m.s
Emittance
Emittance
Phase Space
& Emittance
F. Sannibale
2
2 2
x x x x
rms
=
( )
( )

=
=
x d dx x x f
x d dx x x f x
N
x
x
D
D
N
n
n
,
,
2
2
2
1
2
2
( )
( )
=
=
x d dx x x f
x d dx x x f x
N
x
x
D
D
N
n
n
,
,
2
2
2
1
2
2
( )
( )
=
=
x d dx x x f
x d dx x x f x x
N
x x
x x
D
D
N
n
n n
,
,
2
2
1
x
x
rms
rms rms rms
x x
x x
x
x
x
x
2
2
2
2
2
For a real beam composed by For a real beam composed by N N particles we can calculate the second particles we can calculate the second
order statistical moments of their phase space distribution: order statistical moments of their phase space distribution:
And define the And define the rms rms emittance emittance as the quantity: as the quantity:
This is equivalent to associate to the real This is equivalent to associate to the real
beam an beam an equivalent or phase ellipse equivalent or phase ellipse in the in the
phase space with area phase space with area
rms rms
and equation: and equation:
2
x
2
x
15
0 = t
w
w
w
w
0
1
> = t t
w
w
1 2
t t t >> =
Nonlinear Forces
Nonlinear Forces
and
and
Filamentation
Filamentation
Phase Space
& Emittance
F. Sannibale
In the case of a Hamiltonian system, as a consequence of the In the case of a Hamiltonian system, as a consequence of the Liouville Liouville
Theorem the Theorem the emittance emittance is conserved is conserved
This is true even when the forces acting are on the system are This is true even when the forces acting are on the system are
nonlinear nonlinear (space charge, nonlinear magnetic and/or electric (space charge, nonlinear magnetic and/or electric fileds fileds, ) , )
This is This is not not true in the case of the true in the case of the rms rms emittance emittance. .
In the presence of nonlinear forces the In the presence of nonlinear forces the rms rms emittance emittance is not conserved is not conserved
Example: Example: filamentation filamentation. Particles with different phase space coordinates, . Particles with different phase space coordinates,
because of the nonlinear forces, move with different phase space because of the nonlinear forces, move with different phase space velocity velocity
But the But the rms rms emittance emittance calculated for increasing times calculated for increasing times increases increases. .
The The emittance emittance according to according to Liouville Liouville is still conserved. is still conserved.
16
The
The
Twiss
Twiss
Parameters
Parameters
Phase Space
& Emittance
F. Sannibale
We saw that a beam with arbitrary phase space We saw that a beam with arbitrary phase space
distribution can be represented by an equivalent distribution can be represented by an equivalent
ellipse with area equal to the ellipse with area equal to the rms rms emittance emittance divided divided
by by . and with equation: . and with equation:
y x w w w w w
w Tw Tw Tw
, 2
2 2
= =
+ +
1 with
2
=
Tw Tw Tw

The status of the beam at a given moment is totally defined when The status of the beam at a given moment is totally defined when the the
emittance emittance and two of the and two of the Twiss Twiss parameters are known. parameters are known.
x
x
y x w w w
w w
w
w
w
w
w
w w rms w
, 2
2
2
2
2
= =

A convenient representation for this ellipse, often used in acce A convenient representation for this ellipse, often used in accelerator lerator
physics, is the one by the so physics, is the one by the so- -called called Twiss Twiss Parameters Parameters
T T
, ,
T T
and and
T T
: :
By comparing the two ellipse equations, we can derive: By comparing the two ellipse equations, we can derive:
y x w w w w w
w Tw w Tw w Tw
,
2 2
= = = =
17
Propagating the
Propagating the
Twiss
Twiss
Parameters
Parameters
Phase Space
& Emittance
F. Sannibale
When the beam propagates along the When the beam propagates along the beamline beamline, the eccentricity and the , the eccentricity and the
orientation of the equivalent ellipse change while the area rema orientation of the equivalent ellipse change while the area remains ins
constant ( constant (Liouville Liouville theorem). In other words, the theorem). In other words, the Twiss Twiss parameters parameters
change along the line according to the action of the line elemen change along the line according to the action of the line elements. ts.
x
x
The single particle matrix formalism can now be extended to the The single particle matrix formalism can now be extended to the Twiss Twiss
parameters. For example for a drift of length parameters. For example for a drift of length L L in the horizontal plane: in the horizontal plane:
0
0 0
0
0
1 0
1
x x
x L x x
x
x L
x
x
=
+ =
|
|
.
|
\
|
|
|
.
|
\
|
=
|
|
.
|
\
|
( )
( )
0 0
2
0 0 0 0
2
0
2
0 0
2
0
2 2
0
2
0 0
2
2
x x x L x x L x x x
x x
x x L x L x x L x x
+ = + =
=
+ + = + =
x
x DRIFT
DRIFT
x
x F QUAD
F QUAD
x
x

0 0
0
0 0
2
0
2
T T T
T T
T T T T
L
L L
=
=
+ =
|
|
|
.
|
\
|
|
|
|
.
|
\
|
=
|
|
|
.
|
\
|
0
0
0
2
1 0
0 1 0
2 1
T
T
T
T
T
T
L
L L
18
Getting Familiar with the
Getting Familiar with the
Concept of
Concept of
Emittance
Emittance
Phase Space
& Emittance
F. Sannibale
A couple of examples: A couple of examples:
Propagation of beams with Propagation of beams with
different different emittance emittance through a through a
FODO lattice FODO lattice
Propagation of beams with Propagation of beams with
different different emittance emittance through a through a
drift space drift space
19
The Concept of Acceptance
The Concept of Acceptance
Phase Space
& Emittance
F. Sannibale
Example: Acceptance of a slit Example: Acceptance of a slit
y
y
-h/2
h/d
-h/d
-h/2
h
d
Electron
Trajectories
Matched beam
emittance
Acceptance at
the slit entrance
Unmatched
beam emittance
20
The
The
Emittance
Emittance
Role
Role
In Accelerator Applications
In Accelerator Applications
Phase Space
& Emittance
F. Sannibale
The The emittance emittance is a fundamental parameter in most is a fundamental parameter in most
accelerator applications. accelerator applications.
Free electron lasers (FEL): Free electron lasers (FEL): Intensity of the radiation Intensity of the radiation
strongly depends on strongly depends on emittance emittance. The smaller the better . The smaller the better
Synchrotron light sources: Synchrotron light sources: smaller smaller emittances emittances gives gives
higher brightness higher brightness
Colliders Colliders: : higher higher emittances emittances give higher luminosity give higher luminosity
(in beam (in beam- -beam limited regime) beam limited regime)
Electron microscopes: Electron microscopes: High resolution requires lower High resolution requires lower
emittances emittances

21
Possible Homework
Possible Homework
Phase Space
& Emittance
F. Sannibale
Calculate the Calculate the Twiss Twiss parameter transport matrix for both planes parameter transport matrix for both planes
of a focusing of a focusing quadrupole quadrupole in the thin lens approximation. in the thin lens approximation.
Prove the relation Prove the relation / /
0 0
= =y y/ /y y
0 0
, where , where and and
0 0
are the vertical are the vertical
emittance emittance after and before acceleration by a field after and before acceleration by a field E E
z z
, and , and y y and and
y y
0 0
are the divergences after and before acceleration. are the divergences after and before acceleration.
Tip: use the definition of Tip: use the definition of rms rms emittance emittance
1
Lectures No. 6 & 7
Lectures No. 6 & 7
Beam Optical Functions
Beam Optical Functions
&
&
Betatron
Betatron
Motion.
Motion.
2
Want to touch on a number of concepts including:
Weak Focusing
Betatron Tune
Strong Focusing
Closed Orbit
One-Turn Matrix
Twiss Parameters and Phase Advance
Dispersion
Momentum Compaction
Chromaticity
Concepts
Concepts
Optical Functions
& Betatron Motion
D. Robin
3
Weak Focusing
V. Veksler and E. M.
McMillan around 1945
Strong Focusing
Christofilos (1950),
Courant, Livingston, and
Snyder (1952)
Historical Prospective
Historical Prospective
Optical Functions
& Betatron Motion
D. Robin
4
Weak
Weak
-
-
Optical Functions
& Betatron Motion
D. Robin
The first synchrotrons were of the so called weak-
focusing type.
The vertical focusing of the circulating
particles was achieved by sloping magnetic
fields, from inwards to outwards radii.
At any given moment in time, the average
vertical magnetic field sensed during one
particle revolution is larger for smaller radii of
curvature than for larger ones.
Weak Focusing
5
Uniform field is focusing in the radial plane but not in the
vertical plane
Stability of transverse oscillations
Stability of transverse oscillations
Optical Functions
& Betatron Motion
D. Robin
6
Stability in BOTH PLANES requires that 0<n<1
Vertical focusing is achieved at the expense of horizontal
focusing
r dr
B dB
n
/
/
=
Focusing in both planes if field lines bend outward
Weak focusing
Weak focusing
Optical Functions
& Betatron Motion
D. Robin
7
The number of oscillations about the design orbit in one turn
design orbit
design orbit
Tune
Tune
Optical Functions
& Betatron Motion
D. Robin
8
Expressing these results in terms of derivatives measured along the
equilibrium orbit
( )
orbit design the respect to with derivative a is ' where
0 ' ' , 0
1
' '
2
0
2
0
= + =
+
R
ny
y
R
x n
x
Stability requires that 0<n<1
The particle will oscillate about the design trajectory with the number of
oscillations in one turn being
ly vertical n
radially n - 1
The number of oscillations in one turn is termed the tune of the ring.
For stable oscillations the tune is less than one in both planes.
Weak focusing
Weak focusing
Optical Functions
& Betatron Motion
D. Robin
9
Disadvantage
Tune is small (less than 1)
As the design energy increased so does the circumference of the
orbit
As the energy increases the required magnetic aperture increases
for a given angular deflection
Because the focusing is weak the maximum radial displacement is
proportional to the radius of the machine.
The result is that the scale of the magnetic components of a high
energy synchrotron become unreasonably large and costly
Disadvantages of weak focusing
Optical Functions
& Betatron Motion
D. Robin
10
Weak
Weak
-
-
Optical Functions
& Betatron Motion
D. Robin
The first synchrotron of this type was the Cosmotron at the
Brookhaven National Laboratory, Long Island. It started operation
in 1952 and provided protons with energies up to 3 GeV.
In the early 1960s, the worlds highest energy weak-focusing
synchrotron, the 12.5 GeV Zero Gradient Synchrotron (ZGS)
started its operation at the Argonne National Laboratory near
Chicago, USA.
The Dubna synchrotron, the largest of them all with a radius of 28
meters and with a weight of the magnet iron of 36,000 tons
Cosmotron
11
Solution
Strong focusing
Use strong focusing and defocusing elements
(|n| >>1)
Optical Functions
& Betatron Motion
D. Robin
12
One would like the restoring force on a particle displaced from the
design trajectory to be as strong as possible.
ALS Bend (n~25)
In a strong focusing lattice there is a
sequence of elements that are either
strongly focusing or defocusing.
The overall lattice is stable
In a strong focusing lattice the
displacement of the trajectory does
not scale with energy of the machine
The tune is a measure of the amount
of net focusing.
Strong Focusing
Strong Focusing
Optical Functions
& Betatron Motion
D. Robin
13
Strong
Strong
-
-
Optical Functions
& Betatron Motion
D. Robin
In 1952 Ernest D. Courant, Milton Stanley Livingston
and Hartland S. Snyder, proposed a scheme for
strong focusing of a circulating particle beam so
that its size can be made smaller than that in a weak-
focusing synchrotron.
In this scheme, the bending magnets are made
to have alternating magnetic field gradients;
after a magnet with an axial field component
decreasing with increasing radius follows one
with a component increasing with increasing
radius and so on.
Thanks to the strong focusing, the magnet
apertures can be made smaller and therefore
much less iron is needed than for a weak-
focusing synchrotron of comparable energy.
The first alternating-gradient synchrotron
accelerated electrons to 1.5 GeV. It was built at
Cornell University, Ithaca, N.Y. and was
completed in 1954.
Size comparison between the
Cosmotron's weak-focusing
magnet (L) and the AGS
alternating gradient focusing
magnets
14
Describing the Motion
In principle knowing both the magnetic lattice and the
initial coordinates of the particles in the particle beam is all
one needs to determine where all the particles will be in
some future time.
Ray-tracing each particle is a very time consuming
especially for a storage ring where the particles go around
for billions of turns.
Can do much more
Want to understand the characteristics of the ring Maps
Strong
Strong
-
-
Optical Functions
& Betatron Motion
D. Robin
15
What Can We Learn?
What Can We Learn?
Optical Functions
& Betatron Motion
D. Robin
Some parts of the ring the beam is large and in others it is small
The particles oscillate around the ring a number of times
16
Tune is the number of oscillations that a particle
makes about the design trajectory
Design orbit
On-momentum
particle trajectory
Tune
Tune
Optical Functions
& Betatron Motion
D. Robin
17
Use a map as a function to project a particles
initial position to its final position.
A matrix is a linear map
One-turn maps project project the particles
position one turn later
What is a Map?
What is a Map?
Optical Functions
& Betatron Motion
D. Robin
x
x
y
y
initial
x
x
y
y
final
MAP
18
Begin with equations of motion Lorentz force
Change dependent variable from time to
longitudinal position
Integrate particle around the ring and find the
closed orbit
Generate a one-turn map around the closed orbit
Analyze and track the map around the ring
Generating a Map
Generating a Map
Optical Functions
& Betatron Motion
D. Robin
19
A closed orbit is defined as an orbit on which a
particle circulates around the ring arriving with the
same position and momentum that it began.
In every working story ring there exists at least one
closed orbit.
Closed orbit
Find the Closed Orbit
Find the Closed Orbit
Optical Functions
& Betatron Motion
D. Robin
20
A one-turn map maps a set of initial coordinates of a
particle to the final coordinates, one-turn later.
The map can be calculated by taking orbits that have a
slight deviation from the closed orbit and tracking
them around the ring.
Closed orbit
( ) ( )
( ) ( )
, ,
' '
' '
, ,
' ' +...
'
' ' +...
'
f f
f i i i co i i co
i i
f f
f i i i co i i co
i i
dx dx
x x x x x x
dx dx
dx dx
x x x x x x
dx dx
= + +
= + +
Generate a one
Generate a one
-
-
turn Map
turn Map
Around the Closed Orbit
Around the Closed Orbit
Optical Functions
& Betatron Motion
D. Robin
21
There are two approaches to introduce the motion of
particles in a storage ring
1. The traditional way in which one begins with Hills
equation, defines beta functions and dispersion,
and how they are generated and propagate,
2. The way that our computer models actually do it
I will begin with the first way
Two approaches
Two approaches
Optical Functions
& Betatron Motion
D. Robin
22
( )
( )
''
''
0
0
x
x
x K s x
y K s y
+ =
+ =
Piecewise Focusing
Piecewise Focusing
Optical Functions
& Betatron Motion
D. Robin
Assume that in a strong focusing synchrotron synchrotron the focusing varies
piecewise around the ring
s
23
Illustration in the simple case of Hills Equation on-
energy
Analytically solve the equations of motion
Generate map
Analyze map
In a storage ring
with periodic
solutions
( )
( )
''
''
0
0
x
x
x K s x
y K s y
+ =
+ =
Case of Hills Equation
Case of Hills Equation
Optical Functions
& Betatron Motion
D. Robin
24
Phase Function
Phase Function
Optical Functions
& Betatron Motion
D. Robin
Solution of the second condition
If we select the integration constant to be 1:
then
' ' ''
'
0
const

+ =
=
0
( ) (0)
( )
s
ds
s
s

= +
Knowledge of the function (s) along the line allows

to compute the phase function
'
1 =
25
Twiss
Twiss
Parameters
Parameters
Optical Functions
& Betatron Motion
D. Robin
Define the Betatron or twiss or lattice functions
(Courant-Snyder parameters)
26
Eliminating the angles by the position and slope we define
the Courant-Snyder invariant
This is an ellipse in phase space with area
The twiss functions have a geometric
meaning
The beam envelope is
The beam divergence
Courant
Courant
-
-
Snyder invariant
Snyder invariant
Optical Functions
& Betatron Motion
D. Robin
27
The Beam Behavior
The Beam Behavior
Optical Functions
& Betatron Motion
D. Robin
Meaning of Beam Envelope and Beta Function and Emittance
Area of ellipse the same everywere (emittance)
Orientation and shape of the ellipse different everywhere (beta and alpha function)
28
( )
''
0 u k s u + =
Solution of Hills Equation
Solution of Hills Equation
Optical Functions
& Betatron Motion
D. Robin
The general solution of
Can be written as
( ) ( ) cos( ( ) (0)) u s s s =
There are two conditions that are obtained
2
'' 2 ' 2 2
1 1
'
0
2 2
k

+ =

' ' ''
0 + =
29
ELSA
ELSA
Optical Functions
& Betatron Motion
D. Robin
30
Example from ELSA
Example from ELSA
Optical Functions
& Betatron Motion
D. Robin
31
How to Compute
How to Compute
Twiss
Twiss
Parameters at one point
Parameters at one point
Optical Functions
& Betatron Motion
D. Robin
Steps
1. Compute the one turn transfer matrix
2. Extract the twiss parameters and tunes
32
One can write the linear transformation,
R
one-turn
, between one point in the
storage ring (i) to the same point one
turn later
1
' ' ' '
where =
' '
i i
one turn
x C S x
x C S x
C S
R
C S
+

=

i
One Turn Transfer Matrix
One Turn Transfer Matrix
Optical Functions
& Betatron Motion
D. Robin
33
The one turn matrix (the first order term of the map) can
be written
Where , , are called the Twiss parameters
and the betatron tune, = /(2*)
For long term stability is real
|TR(R)|= |2cos |<2
cos sin sin

' ' sin cos sin
one turn
C S
R
C S

+

= =

2
2
1
'
,
=
+
=
One turn matrix
One turn matrix
Optical Functions
& Betatron Motion
D. Robin
34
One can diagonalize the one-turn matrix, R
This separates all the global properties of the matrix into N and the
local properties into A.
In the case of an uncoupled matrix the position of the particle each
turn in x-x phase space will lie on an ellipse. At different points in the
ring the ellipse will have the same area but a different orientation.
1
one turn
one turn
N AR A
=
x
x
x
x
Computation of
Computation of
beta
beta
-
-
functions and tunes
functions and tunes
Optical Functions
& Betatron Motion
D. Robin
35
The eigen-frequencies are the tunes. A contains information about the
beam envelope. In the case of an uncoupled matrix one can write A
and R in the following way:
The beta-functions can be propagated from one position in the ring to
another by tracking A using the transfer map between the initial point
the final point
This is basically how our computer models do it.
1
one turn
one turn
N AR A
=
1
0
0
1
cos sin cos sin sin
sin cos sin cos sin

+

=

f fi i
A R A =
Computation of
Computation of
beta
beta
-
-
functions and tunes
functions and tunes
Optical Functions
& Betatron Motion
D. Robin
36
Transport of the twiss parameters in terms of the transfer
matrix elements
Transfer matrix can be expressed in terms of the twiss
parameters and phase advances
2 2
2 2
2
1
2
' '
' ' '
' '
f i
C CS S
CC C S SS
C C S S

= +

( )
( )
1
cos sin sin
sin cos cos sin
f
fi i fi f i fi
i
fi
i f i f
i
fi fi fi f fi
f
f i f i
R

+

=

+

+

Transport of the beam ellipse
Optical Functions
& Betatron Motion
D. Robin
37
Assume that the energy is fixed no cavity or damping
Find the closed orbit for a particle with slightly
different energy than the nominal particle. The
dispersion is the difference in closed orbit between
them normalized by the relative momentum
difference
p/p = 0
p/p > 0
' ' ' '
,
,
x y
x y
p p
x D y D
p p
p p
x D y D
p p

= =

= =
Dispersion
Dispersion
Optical Functions
& Betatron Motion
D. Robin
38
Dispersion, D, is the change in closed orbit as a function
of energy
Dispersion
E/E = 0
E/E > 0
x
E
x D
E
=
0 0 1
'
' ' ' '
x
x
f i
x C S D x
x C S D x

=

Dispersion
Dispersion
Optical Functions
& Betatron Motion
D. Robin
39
Dispersion is the distance between the
design on-energy particle and the design off
energy particle divided by the relative
difference in energy spread between the two.
' '
x
x
p
x D
p
p
x D
p
=
Dispersive Systems
Dispersive Systems
Optical Functions
& Betatron Motion
D. Robin
40
Momentum compaction, , is the change in the closed
orbit length as a function of momentum.
E/E = 0
E/E > 0
L p
L p

=
Momentum Compaction
Momentum Compaction
Optical Functions
& Betatron Motion
D. Robin
0
0
L
x
D
ds
=

41
Off-momentum particles are not oscillating around
design orbit, but around chromatic closed orbit
Distance from the design orbit depends linearly with
momentum spread and dispersion
Design orbit
Design orbit
On-momentum
particle trajectory
Off-momentum
particle trajectory
Chromatic close orbit
Chromatic Closed Orbit
Chromatic Closed Orbit
Optical Functions
& Betatron Motion
D. Robin
42
Focal length of the lens is dependent upon energy
Larger energy particles have longer focal lengths
Chromatic
Chromatic
Aberation
Aberation
Optical Functions
& Betatron Motion
D. Robin
43
By including dispersion and sextupoles it is possible to
compensate (to first order) for chromatic aberrations
The sextupole gives a position dependent
Quadrupole
B
x
= 2Sxy
B
y
= S(x
2
y
2
)
Chromatic Aberration
Chromatic Aberration
Correction
Correction
Optical Functions
& Betatron Motion
D. Robin
44
No dispersion or dispersion slope at
the beginning and end of the line
Achromatic Transport
Optical Functions
& Betatron Motion
D. Robin
45
No dispersion or dispersion slope at the end of the
line
Dispersion is negative in the central bends (cuts the
corner)
Isochronous Transport
Isochronous Transport
Optical Functions
& Betatron Motion
D. Robin
46
No dispersion or dispersion slope at the end of the line
Dispersion is positive in the central bend but the central
bend is inverted
Isochronous and
Isochronous and
Optical Functions
& Betatron Motion
D. Robin
47
In an linear uncoupled machine the turn-by-turn positions
and angles of the particle motion will lie on an ellipse
0
0 0
'
( ) ( ) cos( ( ) )
( ) cos( ( ) ) sin( ( ) )
( ) ( )
x s s s
x s s s
s s

= +
= + +
2 2
2
' '
Area of the ellipse, :
x xx x
= + +
Beam Ellipse
Beam Ellipse
Optical Functions
& Betatron Motion
D. Robin
48
Beam ellipse matrix
Transformation of the beam ellipse matrix
x
x
beam

,
, ,
,
T
x x
x i f
beam f beam i
x i f
R R

=

Optical Functions
& Betatron Motion
D. Robin
49
Transport of the
Transport of the
Beam Ellipse
Beam Ellipse
Optical Functions
& Betatron Motion
D. Robin
50
Transport of the twiss parameters in terms of the transfer
matrix elements
Transfer matrix can be expressed in terms of the twiss
parameters and phase advances
2 2
2 2
2
1
2
' '
' ' '
' '
f i
C CS S
CC C S SS
C C S S

= +

( )
( )
1
cos sin sin
sin cos cos sin
f
fi i fi f i fi
i
fi
i f i f
i
fi fi fi f fi
f
f i f i
R

+

=

+

+

Optical Functions
& Betatron Motion
D. Robin
51
This approach provides some insights but is limited
Begin with on-energy no coupling case. The beam is
transversely focused by quadrupole magnets. The
horizontal linear equation of motion is
2
2
3 356
( ) ,
where , with
( )
being the pole tip field
the pole-tip radius, and
[T-m] . [GeV/c]
T
T
d x
k s x
ds
B
k
B a
B
a
B p
=
=
First approach
First approach
traditional one
traditional one
Optical Functions
& Betatron Motion
D. Robin
52
The solution can be parameterized by a psuedo-
harmonic oscillation of the form
0
0 0
'
,
( ) ( ) cos( ( ) )
( ) cos( ( ) ) sin( ( ) )
( ) ( )
where ( ) is the beta function,
( ) is the alpha function,
( ) is the betatron phase, and
x y
x s s s
x s s s
s s
s
s
s
= +
= + +
is an action variable
0
s
ds
Hills equation
Hills equation
Optical Functions
& Betatron Motion
D. Robin
53
L6 & L7 Possible Homework
Optical Functions
& Betatron Motion
D. Robin
At the At the azimuthal azimuthal position position s s in an proton storage ring, the in an proton storage ring, the Twiss Twiss parameters are parameters are

x x
=10 m, =10 m,
y y
=3 m, and =3 m, and
x x
= =
y y
=0. If the beam =0. If the beam emittance emittance is 10 nm for the horizontal is 10 nm for the horizontal
plane and 1 nm for the vertical one and the dispersion function plane and 1 nm for the vertical one and the dispersion function at that location at that location
is zero for both planes, what is the is zero for both planes, what is the rms rms beam size (beam envelope) and the beam size (beam envelope) and the rms rms
beam divergence for both planes at the location beam divergence for both planes at the location s s? What will be the case for an ? What will be the case for an
electron beam? electron beam?
Explain what the dispersion function represent in a storage rin Explain what the dispersion function represent in a storage ring. Explain what is g. Explain what is
the difference between dispersion and chromaticity. the difference between dispersion and chromaticity.
Explain the difference between an Explain the difference between an achromat achromat cell and an cell and an isochronous isochronous one. one.
In the horizontal direction, the one In the horizontal direction, the one- -turn transfer matrix (map) for a storage turn transfer matrix (map) for a storage
ring is: ring is:
Is the Is the emittance emittance preserved? preserved?
Is the motion stable Is the motion stable

7 . 0 05 . 0
1 5 . 1
54
Optical Functions
& Betatron Motion
D. Robin
1. Show that there are two conditions that can be derived
relating
2.
( )
''
0 u k s u + =
( ) ( ) cos( ( ) (0)) u s s s =
( ), ( ) s s
x
x
Sketch the phase space ellipse at these locations
Focusing quad
Beam envelope
1
Lecture No. 8
Lecture No. 8
Longitudinal Dynamics in Storage Rings
Longitudinal Dynamics in Storage Rings
2
B
y
Path Length Dependence
On Trajectory
On Trajectory
Storage Rings
Longitudinal Dynamics
F. Sannibale
In the example (sector bending magnet) In the example (sector bending magnet) L L > > L L
0 0
so that so that
C C
> 0 > 0
Higher energy particles will leave the magnet later. Higher energy particles will leave the magnet later.
z z
B q
c m
qB
p
0

= =
2
0
2
0
2
0
2
0
1
c m
E
c m
E c m
c m
W
+ =
+
= =
p > p
0
C
electrons for E
MeV
511 . 0 1
] [
+
protons for E
GeV
938 . 0 1
] [
+
B and A between length Trajectory
0
= L
C and A between length Trajectory = L
0
0
0
0
p
p p
L
L L
0 0
p
p
L
L
C
where where
C C
is constant is constant
p
0
A
B
O

0 0 0
1
E
E
p
p
L
L
For
C C
>>
3
on Velocity
on Velocity
Storage Rings
F. Sannibale
Consider two particles with different momentum on parallel traje Consider two particles with different momentum on parallel trajectories: ctories:
p p p + =
0 1
L
1
L
0
( ) ct L ct L
0 0 0 1
= + =
0 0
0 1
0

=
L
L L
L
L
( ) = = =
3
0 0 0
c m c m p c m p
2
0
p
p
0
2
0
1
p
p
L
L
=
This path length dependence on momentum applies everywhere, This path length dependence on momentum applies everywhere,
also in straight trajectories. also in straight trajectories.
At a given instant At a given instant t t: :
But But: :
The effect quickly vanishes for relativistic particles. The effect quickly vanishes for relativistic particles.
Higher momentum particles precede the ones with lower momentum. Higher momentum particles precede the ones with lower momentum.
4
Total Path Length
Total Path Length
Dependence on Momentum
Dependence on Momentum
Storage Rings
F. Sannibale
0
s s s =
0 0
2
0
1
p
p
p
p
L
s
C C
|
|
.
|
\
|
=
In this reference frame we can combine the previous results and In this reference frame we can combine the previous results and
obtain for the obtain for the path length dependence on momentum path length dependence on momentum: :
We define as the We define as the reference orbit reference orbit the trajectory of length the trajectory of length L L
0 0
that the that the
reference particle reference particle with nominal energy with nominal energy E E
0 0
describes between describes between A A and and B B. .
The position The position s s of a generic particle will be referred to of a generic particle will be referred to s s
0 0
, the position of , the position of
the reference particle on the reference orbit the reference particle on the reference orbit: :
Lets consider a particle moving in a region in the Lets consider a particle moving in a region in the
presence of electric and magnetic fields. Under the presence of electric and magnetic fields. Under the
action of such fields, the particle will define a action of such fields, the particle will define a
trajectory of length trajectory of length L L between the points between the points A A and and B B. .
ds
A
B
particle reference the precedes particle the s for 0 <
Where the constant Where the constant
C C
= =
- -2 2

C C
is is
called the called the momentum compaction momentum compaction
5
Energy Variation
Energy Variation
Storage Rings
F. Sannibale
( ) qV s d t r E q E
L
F
= =
,
0
We define as We define as V V the voltage gain the voltage gain for the particle. for the particle.
V V depends only on the particle trajectory and includes the contri depends only on the particle trajectory and includes the contribution of bution of
every electric field present in the area (RF fields, space charg every electric field present in the area (RF fields, space charge fields, e fields,
fields due to the interaction with the vacuum chamber, ) fields due to the interaction with the vacuum chamber, )
The particle can also experience The particle can also experience energy variations energy variations U U( (E E) that depend also ) that depend also
on its energy on its energy, as for the case of the radiation emitted by a particle under , as for the case of the radiation emitted by a particle under
acceleration (synchrotron radiation when the acceleration is tra acceleration (synchrotron radiation when the acceleration is transverse). nsverse).
ds
E(r, t)
A
B
The energy gain for a particle that moves from The energy gain for a particle that moves from A A to to B B is given by: is given by:
The total energy variation will be given by the sum of the two The total energy variation will be given by the sum of the two terms: terms:
( ) E U qV E
T
+ =
6
The Rate of
The Rate of
Change of Energy
Change of Energy
Storage Rings
F. Sannibale
( ) ( ) ( )
0 0 0
E U s qV s E
T
+ =
The energy variation for the reference particle is given by: The energy variation for the reference particle is given by:
For particle with energy For particle with energy E = E E = E
0 0
+ + E E and orbit position and orbit position s = s s = s
0 0
+ + s s: :
( ) ( ) ( ) ( ) ( ) E
dE
dU
E U s
ds
dV
q s qV E E U s s qV s E
E s
T
+ + + + + + =
0 0
0 0 0 0
Where the last expression holds for the case where Where the last expression holds for the case where
s << L s << L
0 0
(reference orbit length) and (reference orbit length) and E << E E << E
0 0
. .
In this approximation we can express the average rate of change In this approximation we can express the average rate of change of of
the energy respect to the reference particle energy by: the energy respect to the reference particle energy by:
( ) ( )
0
0
T
s E s E
dt
E d
T T

|
|
.
|
\
|
+
E
dE
dU
s
ds
dV
q
T dt
E d
E s
0 0
0
1
particle reference the of velocity c
B and A between orbit reference the of lenght L
with
c
L
T where
=
=
=
0
0
0
0
0

7
Towards the Longitudinal
Towards the Longitudinal
Motion Equation
Motion Equation
Storage Rings
F. Sannibale
In the present approximation of small In the present approximation of small s s and and E E, the average rate of change , the average rate of change
of the particle position respect to the reference particle posit of the particle position respect to the reference particle position is: ion is:
0 0 0
1
p
p
T L
s
dt
d
C
c
E
p
c
dE
dp
0

=
dt
E d
p dt
s d
C

=
0
2
2
0 0 0 0
0 0
0
0 0 0 0 0 0
2
2
1 1
E
C
s
C
E
C
s
C
dE
dU
T p
p
c s
ds
dV
T
q
p
E
dE
dU
T p
s
ds
dV
T
q
p dt
s d
= =

|
|
.
|
\
|
+
E
dE
dU
s
ds
dV
q
T dt
E d
E s
0 0
0
1
dt
s d
dE
dU
T
s
ds
dV
T
q
p dt
s d
E s
C

+ =
0 0
0 0 0
2
2
1
0
0
p
p
c
dt
s d
C

dt
p d
p
c
dt
s d
C

=
0
0
2
2

But: But:
and remembering: and remembering:
8
The Longitudinal
The Longitudinal
Motion Equation
Motion Equation
Storage Rings
F. Sannibale
0 2
2
2
2
= +
s
dt
s d
dt
s d
D
( )
dt
s d p
t E
C
0
Finally, by defining the quantities: Finally, by defining the quantities:
We obtain the equations of motion for the longitudinal plane: We obtain the equations of motion for the longitudinal plane:
0
0
E E
L s
<<
<<
0
0 0
2
1
s
C
ds
dV
T
q
p
=
0
0
2
1
E
D
dE
dU
T
=
We will study the case of storage rings where We will study the case of storage rings where dV dV/ /ds ds is mainly due to the is mainly due to the
RF system used for restoring the energy lost per turn by the bea RF system used for restoring the energy lost per turn by the beam. m.
9
The Damped Oscillator
The Damped Oscillator
Equation
Equation
Storage Rings
F. Sannibale
0 2
2
2
2
= +
s
dt
s d
dt
s d
D
This expression is the well known This expression is the well known
damped harmonic oscillator equation, damped harmonic oscillator equation,
which has the general solution: which has the general solution:
( ) ( )
t i t i t
Be Ae e t s
D

+

n oscillatio damped anti 0
n oscillatio damped 0
<
>
D
D
motion unstable 0
n oscillatio stable 0
2
2
<
>
( )
t i t i t
Be Ae e
D

+
( )
t i t i t
Be Ae e
D

+
t
D
e

0 0
2
> > and
D
The stable solution represents an oscillation with frequency The stable solution represents an oscillation with frequency 2 2 and with and with
exponentially decreasing amplitude. exponentially decreasing amplitude.
10
Damping in the Case of
Damping in the Case of
Storage Rings
Storage Rings
Storage Rings
F. Sannibale
The case of damped oscillations is exactly what we want for sto The case of damped oscillations is exactly what we want for storing ring
particles in a storage ring. particles in a storage ring.
0 >
D
0
0
2
1
E
D
dE
dU
T
=
0
0
<
E
dE
dU
The synchrotron radiation (SR) emitted when particles are on a The synchrotron radiation (SR) emitted when particles are on a curved curved
trajectory satisfies the condition. The SR power scales as: trajectory satisfies the condition. The SR power scales as:
( ) ( ) radius trajectory P dt dU
SR
= =
2
2
2 2
4
1
Typically, synchrotron radiation damping is very efficient in e Typically, synchrotron radiation damping is very efficient in electron lectron
storage rings and negligible in proton machines. storage rings and negligible in proton machines.
The The damping time damping time 1/ 1/
D D
(~ ms for e (~ ms for e
- -
, ~ 13 hours LHC at 7 , ~ 13 hours LHC at 7 TeV TeV) is usually ) is usually
much larger than the period of the longitudinal oscillations much larger than the period of the longitudinal oscillations 1/2 1/2 (~ (~ s). s).
This implies that the damping term can be neglected when calcula This implies that the damping term can be neglected when calculating the ting the
particle motion for particle motion for t t << << 1/ 1/
D D
: :
0
2
2
2
= +
s
dt
s d
Harmonic oscillator equation Harmonic oscillator equation
11
RF
T
Synchronicity
Synchronicity
in Storage Rings
in Storage Rings
Storage Rings
F. Sannibale
Lets consider a storage ring with reference trajectory of lengt Lets consider a storage ring with reference trajectory of length h L L
0 0
: :
( ) ( ) t V t V
RF RF
sin
=
c
L
T
0
0
=
RF
T
3 3
0
= = h T T
RF
RF
Cavity
h
f
f hT T
RF
RF
= =
0 0
Synchronicity Condition Synchronicity Condition
RF RF
RF
f
T
2 1
= =
number harmonic
h
the
called is integer The
12
The Synchrotron
The Synchrotron
Frequency and Tune
Frequency and Tune
Storage Rings
F. Sannibale
In a storage ring the at equilibrium: In a storage ring the at equilibrium:
t c s
0
=
0
0 0
2
1
s
C
ds
dV
T
q
p
=
( ) ( ) ( ) t h V t V t V
RF RF 0
sin
sin
= =
( )
0
0
0
0
cos
1
0 0
t
c
V h
dt
dV
c ds
dV
RF
t s
= =
( )
frequency n synchrotro
c
V h
p
q
s
C
cos
2
0 0
2
0
2
=
phase s synchronou t
RF s
=
0

tune n synchrotro
S
0

=
( ) ( ) ( ) 0 sin
0 0 0
= = + U V q E U s qV
s
V q
U
s
sin
0
=
Where Where U U
0 0
is the energy lost per turn and is the energy lost per turn and V V is integrated over turn. is integrated over turn.
For our storage ring: For our storage ring:
13
The Synchrotron
The Synchrotron
Oscillations
Oscillations
Storage Rings
F. Sannibale
0
2
2
2
= +
s
dt
s d
( )
dt
s d p
t E
C
0
( ) + = t s s cos
<<
D
( )
= t
p
s E
C
sin
0
p c E =
0
t
RF
=
s
Phase = : t c s
0
=
RF
c s
0
=
( ) + = t cos
0
:
p
p
Deviation Momentum Relative

=
( )

= t
h
C
sin
0
If If Additionally: Additionally:
Synchrotron Oscillations Synchrotron Oscillations
A different set of variables: A different set of variables:
For For s << L s << L
0 0
and and E << E E << E
0 0
. .
14
The Longitudinal
The Longitudinal
Phase space
Phase space
Storage Rings
F. Sannibale
( ) + = t cos
( )

= t
h
C
sin
0
1

2
0 2
2
2
=
|
|
.
|
\
|
C
h
0 >
C
C
h
We just found: We just found:

This equation represents an ellipse in the This equation represents an ellipse in the
longitudinal phase space { longitudinal phase space { , , } }
0 >
C
0 >
D
( )

+ =

t e
t
D
cos
( )

=

t e
h
t
C
D
sin
0
With damping: With damping:
In rings with negligible synchrotron radiation (or with negligib In rings with negligible synchrotron radiation (or with negligible le
non non- -Hamiltonian forces, the longitudinal Hamiltonian forces, the longitudinal emittance emittance is conserved. is conserved.
This is the case for heavy ion and for most proton machines. This is the case for heavy ion and for most proton machines.
15
Phase Stability
Phase Stability
Storage Rings
F. Sannibale
We showed that for the synchronous phase: We showed that for the synchronous phase:
( ) ( ) t V t V
RF RF
sin
=
t
V
|
|
.
|
\
|
=
V q
U
S
arcsin
0 1
V q
U
S
sin
0
=
|
|
.
|
\
|
=
V q
U
or
S
arcsin
0 2

0 0
2
0 0
1
p
p
p
p
L
s
T
t
C C
|
|
.
|
\
|
=
But But
unstable stable For
S S C
2 1
, 0 >
stable unstable For
S S C
2 1
, 0 <
For positive charge particles: For positive charge particles:
For negative charge For negative charge
particles all the phases particles all the phases
are shifted by are shifted by . .
2
1
1
|
|
.
|
\
|
=
C
TR
Crossing the transition energy Crossing the transition energy

during energy ramping requires during energy ramping requires
a phase jump of ~ a phase jump of ~
We define as We define as transition transition
energy energy the energy at the energy at
which which
C C
changes sign. changes sign.
16

S S
0 0 or or
Large Amplitude
Large Amplitude
Oscillations
Oscillations
Storage Rings
F. Sannibale
( ) ( ) ( ) ( )
V q V q
d
dV
q qV V q qV E
S S S S T
S

sin
+ = + + = + =
So far we have used the So far we have used the small oscillation approximation small oscillation approximation where: where:
In the more general case of larger phase oscillations: In the more general case of larger phase oscillations:
( ) ( ) ( ) + + =
S S T
V q qV E sin
Stable and unstable orbits exist. The two regions are separated Stable and unstable orbits exist. The two regions are separated by a special by a special
trajectory called trajectory called separatrix separatrix
Larger amplitude orbits have smaller synchrotron frequencies Larger amplitude orbits have smaller synchrotron frequencies

S S
= = 0 0 or or
Separatrices
RF Buckets
And by Numerical integration: And by Numerical integration:
For larger amplitudes, trajectories in the phase space are not For larger amplitudes, trajectories in the phase space are not ellipsis ellipsis
anymore. anymore.
17
Momentum Acceptance
Momentum Acceptance
Storage Rings
F. Sannibale

S S
0 0 or or
The RF bucket is the area of the longitudinal phase space where The RF bucket is the area of the longitudinal phase space where a particle a particle
orbit is stable orbit is stable
0
2
0
2
p c h
V q
p
p
C
ACC

=
|
|
.
|
\
|

S S
= = 0 0 or or
RF Buckets
The The momentum acceptance momentum acceptance is defined as the maximum momentum that a is defined as the maximum momentum that a
particle on a stable orbit can have. particle on a stable orbit can have.
( p/p
0
)
ACC
.
( )
0
2
0
2
2 p c h
V q
Q
Q F
p
p
C
ACC

=
|
|
.
|
\
|
( )
|
|
.
|
\
|
=
Q
Q Q F
1
arccos 1 2
2
factor voltage Over
U
V q
Q
s 0
sin
1
= =
18
Bunch Length
Bunch Length
Storage Rings
F. Sannibale
In electron storage rings, the statistical emission of synchrot In electron storage rings, the statistical emission of synchrotron radiation ron radiation
photons generates photons generates gaussian gaussian bunches. bunches.
The over voltage The over voltage Q Q is usually large so that the core of the bunch lives in is usually large so that the core of the bunch lives in
the small oscillation region of the bucket. The equation of moti the small oscillation region of the bucket. The equation of motion in the on in the
phase space are elliptical: phase space are elliptical:
1

2
0 2
2
2
=
|
|
.
|
\
|
C
h
0
0

p
p c
s
h
C C

If If
p p
/ /p p
0 0
is the is the rms rms relative momentum spread relative momentum spread of the of the gaussian gaussian distribution, distribution,
then the then the rms rms bunch length bunch length is given by: is given by:
( )
0
2
0
0 0
3
0
cos
2 p
V f h
p
q
c
p
c
p
S
C
p
C
S
In the case of heavy ions and of most of protons machines, the In the case of heavy ions and of most of protons machines, the whole RF whole RF
bucket is usually filled with particles. The bunch length bucket is usually filled with particles. The bunch length l l is then proportional is then proportional
to the difference between the two extreme phases of the to the difference between the two extreme phases of the separatrix separatrix: :
( ) 2
1 2 RF
l =
19
Effects of the
Effects of the
Storage Rings
F. Sannibale
A charged particle when accelerated radiates. A charged particle when accelerated radiates.
( )
2
4
3
2
0
3
2
E
c m
r c
P
dt
dU
e
SR
= =
In high energy storage rings transverse acceleration induces si In high energy storage rings transverse acceleration induces significant gnificant
radiation (synchrotron radiation) while longitudinal acceleratio radiation (synchrotron radiation) while longitudinal acceleration generates n generates
negligible radiation ( negligible radiation (1/ 1/
2 2
). ).
turn per lost energy dt P U
finite
SR
0
( ) [ ]
planes all in damping
dt E P
dE
d
T dE
dU
T
DY DX
SR
E
D

,
2
1
2
1
0
0 0
0
= =
emittances and spread momentum m equilibriu
p
Y X
p

,
0
Synchrotron radiation plays a major role in the dynamics of an Synchrotron radiation plays a major role in the dynamics of an electron electron
storage ring storage ring
radius electron classical r
e

curvature trajectory
20
Energy Lost per Turn
Energy Lost per Turn
Storage Rings
F. Sannibale
( )
2
4
3
2
0
3
2
E
c m
r c
P
dt
dU
e
SR
= =
c
ds
dt ct ct s = =
For relativistic electrons: For relativistic electrons:
turn per lost energy dt P U
finite
SR
0
( )

= =

finite
e
finite
SR
ds
c m
E r
ds P
c
U
2 3
2
0
4
0
0
3
2 1
In the case of dipole magnets with constant radius In the case of dipole magnets with constant radius ( (iso iso- -magnetic magnetic case): case):
( )

4
0
3
2
0
0
3
4 E
c m
r
U
e
=
The average radiated power is given by: The average radiated power is given by:
( )
nce circumfere ring L
L
E
c m
r c
T
U
P
e
SR
= =
4
0
3
2
0
0
0
3
4
21
Damping Coefficients
Damping Coefficients
Storage Rings
F. Sannibale
By performing the calculation one obtains: By performing the calculation one obtains:
( )
2
4
3
2
0
3
2
E
c m
r c
P
dt
dU
e
SR
= = ( ) [ ]
= = dt E P
dE
d
T dE
dU
T
SR
E
D 0
0 0
2
1
2
1
0
( ) D
E T
U
D
+ = 2
2
0 0
0
Where Where D D depends on the lattice parameters. depends on the lattice parameters.
For the For the iso iso- -magnetic separate function magnetic separate function case: case:

2
L
D
C
=
Analogously, for the transverse plane: Analogously, for the transverse plane:
( ) D
E T
U
X
= 1
2
0 0
0
and and
0 0
0
2 E T
U
Y
=
Sometimes the Sometimes the partition numbers partition numbers are used: are used:
1 1 2 = = + =
Y X S
J D J D J
with with
= 4
i
J
22
Quantum Nature of
Quantum Nature of
Storage Rings
F. Sannibale
We saw that synchrotron radiation induces damping in all the We saw that synchrotron radiation induces damping in all the
planes. planes.
Because of that, one would expect that all the particles should Because of that, one would expect that all the particles should
collapse in a single point. collapse in a single point.
This This does not does not happen because of the happen because of the quantum nature of quantum nature of
synchrotron radiation synchrotron radiation. .
In fact, photons are randomly emitted in quanta of discrete In fact, photons are randomly emitted in quanta of discrete
energy and every time a photon is emitted the parent electron energy and every time a photon is emitted the parent electron
undergoes to a jump in energy. undergoes to a jump in energy.
Such a process perturbs the electron trajectories exciting Such a process perturbs the electron trajectories exciting
oscillations in all the planes. oscillations in all the planes.
These oscillations grow until reaching These oscillations grow until reaching equilibrium equilibrium when balanced when balanced
by the radiation damping. by the radiation damping.
23
Emittance
Emittance
and
and
Momentum Spread
Momentum Spread
Storage Rings
F. Sannibale
At equilibrium the momentum spread is given by: At equilibrium the momentum spread is given by:
m C where
ds
ds
J
C
p
q
S
q p
13
2
3
2
0
2
0
10 84 . 3
1
1
= =
|
|
.
|
\
|

case magnetic iso
J
C
p
S
q p
=
|
|
.
|
\
|

2
0
2
0
For the horizontal For the horizontal emittance emittance at equilibrium: at equilibrium:
( )

+ +
=
T T T
s H 2
2 2
=
ds
ds H
J
C
X
q
2
3
2
0
1
where: where:
In the vertical plane, when no vertical bend is present, the sy In the vertical plane, when no vertical bend is present, the synchrotron nchrotron
radiation contribution to the equilibrium radiation contribution to the equilibrium emittance emittance is very small and the is very small and the
vertical vertical emittance emittance is defined by machine imperfections and is defined by machine imperfections and nonlinearities nonlinearities
that couple the horizontal and vertical planes: that couple the horizontal and vertical planes:
1
1
1 +
=
+
=
X Y
and factor coupling with
24
Time Scale in
Time Scale in
Storage Rings
Storage Rings
Storage Rings
F. Sannibale
At this point we have discussed the motion of a particle in an At this point we have discussed the motion of a particle in an
accelerator for all the planes. accelerator for all the planes.
Damping: Damping: several ms for electrons, ~ infinity for heavier particles several ms for electrons, ~ infinity for heavier particles
It can be helpful remarking the time scale for the different It can be helpful remarking the time scale for the different
phenomena governing the particle dynamics. phenomena governing the particle dynamics.
Betatron Betatron oscillations: oscillations: ~ tens of ns ~ tens of ns
Synchrotron oscillations: Synchrotron oscillations: ~ tens of ~ tens of s s
Revolution period: Revolution period: ~ hundreds of ns to ~ hundreds of ns to s s
25
Possible Homework
Possible Homework
Storage Rings
F. Sannibale
Calculate the synchrotron frequency and tune for the ALS when t Calculate the synchrotron frequency and tune for the ALS when the ring he ring
is operating in the following configuration: RF = 500 MHz, harmo is operating in the following configuration: RF = 500 MHz, harmonic nic
number = 328, E = 1.9 number = 328, E = 1.9 GeV GeV, momentum compaction = 0.00137, energy lost , momentum compaction = 0.00137, energy lost
per turn = 279 per turn = 279 keV keV, peak RF voltage = 1.3 MV. , peak RF voltage = 1.3 MV.
Calculate the ratio between the synchrotron radiation power rad Calculate the ratio between the synchrotron radiation power radiated by iated by
a particle in the Large a particle in the Large Hadron Hadron Collider Collider (LHC), the proton (LHC), the proton collider collider at CERN, at CERN,
and the one radiated by a particle in the Advanced Light Source and the one radiated by a particle in the Advanced Light Source (ALS), the (ALS), the
electron storage ring in Berkeley. The magnet bending radius is electron storage ring in Berkeley. The magnet bending radius is ~2810 m ~2810 m
and ~5 m and the particle energy is 7000 and ~5 m and the particle energy is 7000 GeV GeV and 1.9 and 1.9 GeV GeV for the LHC and for the LHC and
the ALS respectively. (Remember that the electron mass is 9.1095 the ALS respectively. (Remember that the electron mass is 9.1095 10 10
- -31 31
Kg Kg
while the proton one is 1.6726 10 while the proton one is 1.6726 10
- -27 27
Kg) Kg)
Calculate the general solution for the damped harmonic oscillat Calculate the general solution for the damped harmonic oscillator or
equation equation
Calculate the momentum acceptance for the ALS ring. Compare it Calculate the momentum acceptance for the ALS ring. Compare it with with
the acceptance value that the ring would have for zero synchrono the acceptance value that the ring would have for zero synchronous us
phase. phase.
1
Lecture No. 9
Lecture No. 9
Particle Sources
Particle Sources
2
Lecture Outline
Lecture Outline
Particle Sources
F. Sannibale
Electron Sources Electron Sources
Basic information. A brief review and some glossary. Basic information. A brief review and some glossary.
How to extract electrons. How to extract electrons.
Characteristics of an electron source. Characteristics of an electron source.
Examples of existing sources. Examples of existing sources.
Performance limiting factors. Performance limiting factors.
An example of a new source scheme. An example of a new source scheme.
Neutron Sources Neutron Sources
Protons and Heavy Ions Sources Protons and Heavy Ions Sources
Anti Anti- -particles Sources particles Sources
3
Electron Story
Electron Story
Particle Sources
F. Sannibale
From the Greek LEKTRON that means Amber.
Fundamental particle: lightest lepton.
m = 9.1095 10
-31
kg or 9.1095 10
-28
g
(1837 times lighter than a proton)
e = 1.6022 10
-19
C or 4.803 10
-10
esu
Discovered by
J.J. Thomson in 1897
Cathode Ray Tube
For the first time it was proved that the atom is not indivisible
and that is composed by more fundamental components.
4
Where Electrons Can Be
Where Electrons Can Be
Found and Produced
Found and Produced
Particle Sources
F. Sannibale
Atom
Solid
Plasma
Pair formation
Heavier Particle Decay Product
Widely used in sources for:
accelerators, microscopes,
technological applications, ...
Commonly used in
positrons and anti-protons
sources
Radiation source
Very high gradient
accelerators research.
Proton and heavy ions sources
Low current, high quality beams,
microscopes, electron holography,
inverse photoemission, ...

5
Two Families of Particles:
Two Families of Particles:
Fermions and Bosons
Fermions and Bosons
Particle Sources
F. Sannibale
In quantum physics, all particles can be divided into two main categories according to
their spin.
Particles with half-integer spin are called fermions, those with integer spin are called
bosons.
Extremely important difference: only fermions, follow the Pauli exclusion principle:
No two fermions may occupy the same state.
As a consequence, when fermions are introduced into a system, they will occupy higher
energy levels when the lower ones are filled up.
On the contrary, bosons will all occupy the lower energy level allowed by the system
Because of the Pauli principle, the two particle categories follow different energy
distributions:
( )
1
1
=
kT E
BE
Ae
E f
Bose-Einstein Distribution:
photons, mesons
Bosons
( )
( )
1
1
+
=
kT E E
FD
F
e
E f
Fermi-Dirac Distribution:
electrons, protons, neutrons,...
Fermions
6
The
The
Fermi
Fermi
Energy
Energy
Particle Sources
F. Sannibale
We are interested to the case where the system of fermions is a solid with its electrons.
The E
F
value is a property of the particular material. Example: E
F
for copper is 7 eV.
( )
1
1
=
kT E
BE
e
E f
Bose-Einstein Distribution for Bosons
( )
( )
eV 7
1
1
=
+
=
F
kT E E
FD
E
e
E f
F
( ) T E f
F FD
= 2 1
Fermi-Dirac Distribution for Fermions
Definition : In a system of fermions
the Fermi energy E
F
is the energy
of the highest occupied state at
zero temperature.
We will deal only with electron sources.
Being electrons fermions (spin 1/2)
we will concentrate our attention in the
Fermi-Dirac distribution
7
Solids and Work Function
Solids and Work Function
Particle Sources
F. Sannibale
empty band
populated states
at T = 0 K
Solid
E
F
Vacuum (free electrons)
Single
Atom
}
states where electrons
are bound to nuclei
}
States where electrons
are bound to the solid,
not to the single nucleus.
State
Energy
m
o
r
e

b
o
u
n
d

Solid
band
band
The band is a range of
energy with a very fine
discrete structure (states).
Practically a continuum
Ionization
Energy
Definition: the work function W
F
is the energy needed to bring an
electron from the Fermi level to
the vacuum level
(a point at infinite distance away
outside the surface).
Work
Function
Vacuum
Last
populated state
Single Atom
Empty state
Example: for Copper (Cu)
E
I
= 7.7 eV
W
F
= 4.7 eV
8
Insulators and Conductors
Insulators and Conductors
Particle Sources
F. Sannibale
Definition 1: In solids, the valence band
is the band that at T = 0 K, is occupied by
the highest energy electrons.
Definition 2: The conduction band
is the higher energy band above the
valence band.
INSULATORS. At T = 0 K:
The valence and the conduction bands are separated
by a gap with no allowed energy states.
The valence band is completely filled with electrons.
The conduction band is totally empty.
CONDUCTORS. At T = 0 K:
The valence and the conduction bands overlap. The same
band is now at the same time of valence and of conduction.
The energy states in such resulting band are only partially
filled.
Conduction Band
Valence Band
Energy of electrons
Gap
a. Insulator
Conduction Band
Valence Band
Energy of electrons
b. Conductor
E
F
9
The Conduction Phenomenon
The Conduction Phenomenon
Particle Sources
F. Sannibale
V
L
E = V
/
L
Solid
l
e
-
l
L
V
l E E Variation Energy = = =
Conduction Band
(Empty)
Valence Band
(Full)
a. Insulator
E
Energy of electrons
No energy state
available in the gap.
Energy of electrons
b. Conductor
E
F
E
Empty energy
states are now
available.
No conduction!
Conduction!
10
At room temperature,
T ~ 300 K
Semiconductors:
Semiconductors:
a Special Kind Of Insulator
a Special Kind Of Insulator
Particle Sources
F. Sannibale
A semiconductor is an insulator with a relatively small gap between the
valence and conduction bands.
The gap is small enough that at room temperature (T ~ 300K), such a
phase transition has already happened.
Above absolute zero (T = 0K), the atoms in a crystal (solid) start vibrating.
As a result, some electrons scatter with the atoms gaining extra energy
(the larger is T, the larger is the extra energy).
For a high enough temperature, in some insulators this extra energy can be larger than
the gap and electrons in the valence band are allowed to go in the conduction band.
As a consequence, such a solid undergoes to a
phase transition from insulator to conductor when the temperature is increased!
Electrons in
Conduction Band
Silicon,
Germanium,
.
11
Thermionic
Thermionic
Emission
Emission
in Conductors
in Conductors
Particle Sources
F. Sannibale
f
FD
(E)
Energy
E
F
E
V
Vacuum
At room
temperature,
T ~ 300 K
W
F
1
.
0
Conduction Band
Energy
E
F
E
V
Vacuum
At high
temperature,
T ~ 1,500 K
W
F
1
.
0
f
FD
(E)
Conduction Band
Emitted
Electron
Owen Richardson received a Nobel prize in 1928 "for his work on the thermionic
phenomenon and especially for the discovery of the law named after him".
Thermionic emission was initially reported in 1873 by Guthrie in Britain.
12
The Photoelectric Effect
The Photoelectric Effect
Particle Sources
F. Sannibale
f
FD
(E)
Energy
E
F
E
V
Vacuum
W
F
1
.
0
Conduction Band
e
-
h
h E Energy Photon
ph
= =
photon frequency
Planck Constant= 6.626068 10
-34
m
2
kg / s
F ph
W E If
Albert Einstein received the 1921 prize in 1922 for work that
he did between 1905 and 1911 on the Photoelectric Effect.
F ph e
W E T =
Max Planck received the 1919 Nobel for the

development of the Quantum Theory of the photon.
13
Field Emission
Field Emission
Particle Sources
F. Sannibale
Field emission was first observed in 1897 by Robert Williams Wood.
But only in 1928, Fowler and Nordheim gave the first theoretical
description of the phenomenon. It was one of the first application of the
quantum mechanics theory.
Quantum tunneling is the quantum-mechanical effect of transitioning through
a classically-forbidden energy state.
r
e
U
p
2
0
4
1
= r E e
constant = E
Vacuum
e
-
Vacuum
e
-
e
-
Tunneling
Vacuum
14
Secondary Emission
Secondary Emission
Particle Sources
F. Sannibale
Primary Particle Primary Particle
Secondary Electrons Secondary Electrons
Primary Particles Primary Particles: photons, : photons,
electrons, protons, neutrons, electrons, protons, neutrons,
ions, ... ions, ...
Physical Processes Physical Processes: ionization, : ionization,
elastic scattering, Auger elastic scattering, Auger
Electrons, Electrons, photoelectric effect, photoelectric effect,
bremsstrahlung bremsstrahlung and pair and pair
formation, formation, Compton scattering, ... Compton scattering, ...
15
Electron Gun Schematic
Electron Gun Schematic
Particle Sources
F. Sannibale
Electron
Generator
Electron
Generator
Accelerating and
Focusing Section
Accelerating and
Focusing Section
Thermionic Thermionic
Photoelectric Photoelectric
Field emission Field emission
Secondary emission Secondary emission
.... ....
Focusing Focusing: usually by : usually by
stationary magnetic fields. stationary magnetic fields.
Acceleration: Acceleration: by electric fields. by electric fields.
Electrostatic Electrostatic
Pulsed Pulsed
Radio Frequency Radio Frequency
Wakefields Wakefields
Application
Application
Higher gradients Higher gradients
16
Electron Source
Electron Source
Main Parameters
Main Parameters
Particle Sources
F. Sannibale
Current: Current:
Average: from Average: from pA pA to several tens of A. to several tens of A.
Peak: Peak: from from A A to thousand of A. to thousand of A.
Energy: Energy: from few from few eV eV to several to several MeV MeV
Energy Spread: Energy Spread: from ~ 0.1 from ~ 0.1 eV eV and up. and up.
Pulse Length: Pulse Length: from from hundreds of hundreds of fs fs to seconds. to seconds.
Single electron. Single electron.
Polarization: Polarization: orientation of the electron spin orientation of the electron spin
Time Structure: Time Structure:
DC DC
Pulsed: Pulsed: from single shot to hundreds of kHz from single shot to hundreds of kHz
CW: CW: from hundreds of MHz to several GHz from hundreds of MHz to several GHz
17
The
The
Emittance
Emittance
:
:
An Important Gun Parameter
An Important Gun Parameter
Particle Sources
F. Sannibale
Emittance Emittance: : volume of the phase space volume of the phase space
occupied by the particles of the beam occupied by the particles of the beam
Liouville Liouville Theorem: Theorem: in a Hamiltonian system (non in a Hamiltonian system (non- -
dissipative system) the dissipative system) the emittance emittance is conserved is conserved
x
dz
dx
x =
1
p
2
p
x
dz
dx
x =
x
dz
dx
x =
2
2 2
x x x x
rms

=
effective effective ( (rms rms) ) Emittance Emittance: :
Smaller Smaller emittance emittance are usually preferred. are usually preferred.
It is very easy to increase the It is very easy to increase the emittance emittance, but very , but very
hard to decrease it! hard to decrease it!
18
Brightness and
Brightness and
Degeneracy Factor
Degeneracy Factor
Particle Sources
F. Sannibale
Short pulses, low energy spread, small Short pulses, low energy spread, small emittances emittances, high , high
current densities, all lead to a current densities, all lead to a high degeneracy factor. high degeneracy factor.
Brightness: Brightness: phase space density of particles. I.e. number of phase space density of particles. I.e. number of
particles per unit of phase space volume. particles per unit of phase space volume.
This can be interpreted as the fact that the phase space volume This can be interpreted as the fact that the phase space volume occupied by occupied by
a particle is given by: a particle is given by: ( (
c c
/2 /2 ) )
3 3
= = elementary phase space volume elementary phase space volume
Heisenberg Heisenberg uncertainty principle: uncertainty principle: it is impossible to determine with precision and
simultaneously, the position and the momentum of a particle.
electrons for pm mc h wavelength Compton
z y x w
c
c w
426 . 2
, , 4
= =
=

Degeneracy Factor, Degeneracy Factor, : : brightness in units brightness in units
of elementary phase space volume. of elementary phase space volume.
Number of particles per elementary volume. Number of particles per elementary volume.
Because of the Because of the Pauli Pauli exclusion principle the exclusion principle the limit value of limit value of is: is:
infinity for bosons and infinity for bosons and 1 for non polarized fermions 1 for non polarized fermions. .
Applied to emittances:
19
Examples of Electron Guns
Examples of Electron Guns
Particle Sources
F. Sannibale
20
More Examples
More Examples
Particle Sources
F. Sannibale
21
Performances Limitations
Performances Limitations
Particle Sources
F. Sannibale
RF Guns. RF Guns.
Repetition Rate. Heat load in the RF structures limits. Repetition Rate. Heat load in the RF structures limits.
Max electric field. Field emission limits. Dark current. Max electric field. Field emission limits. Dark current.
High power High power thermionic thermionic guns. guns.
Average Current. Limits in the cathodes current density. Average Current. Limits in the cathodes current density.
Cathode lifetime. Cathode lifetime.
Cathode thermal Cathode thermal emittance emittance limit limit
Field emission guns. Field emission guns.
Max electric field at the tip. Limits in the minimum size of th Max electric field at the tip. Limits in the minimum size of the tip. e tip.
Intrinsic low average current. Intrinsic low average current.
Secondary Emission Gun. Secondary Emission Gun.
Low current densities. Low current densities.
High energy spread. High energy spread.
22
The Ultimate Limit
The Ultimate Limit
Particle Sources
F. Sannibale
Emittance Emittance
Energy spread Energy spread
Brightness Brightness
Degeneracy Degeneracy
factor factor
Thermionic Thermionic: : ~ 10 ~ 10
- -14 14
SEM: SEM: ~ 10 ~ 10
- -14 14
Photo Photo- -RF guns: RF guns: ~ 10 ~ 10
- -12 12
Field emission: Field emission: ~ 10 ~ 10
- -
5 5
The degeneracy factor inside a metal cathode is ~ 1 The degeneracy factor inside a metal cathode is ~ 1
How do we loose all of that How do we loose all of that
?
?
Extraction Mechanism Extraction Mechanism
Coulomb interaction Coulomb interaction
(space charge) (space charge)
Practically, most of the edge applications (accelerators, free Practically, most of the edge applications (accelerators, free
electron lasers, microscopes, inverse photoemission, ...) are electron lasers, microscopes, inverse photoemission, ...) are
limited by the performance of the electron gun in: limited by the performance of the electron gun in:
!!!
!!!
23
A High Degeneracy
A High Degeneracy
Electron Source
Electron Source
Particle Sources
F. Sannibale
24
Fundamental Concepts
Fundamental Concepts
Particle Sources
F. Sannibale
1) Electron Excitation. In the region of well defined and controlled volume (defined by
the overlap of the lasers) we ionize on average one alkali atom per laser pulse. The
electron in the excited atom will have a total energy close to zero and will start to drift
away from the ion.
2) Waiting Period. After the laser pulse, we wait the time necessary for the electron to
go far enough from the ion loosing most of its kinetic energy and we apply a short
pulsed voltage to extract the electron from the ionization region.
3) Electron Acceleration. In this step, we accelerate the electron up to the energy
required by the considered application.
4) Ion Clearing. After the electron acceleration, we apply a cleaning field in order to
remove the residual ion before the beginning of the following cycle. In this way it is
avoided that the residual ion will interact with the electron produced in the next pulse.
The application of all such concepts allows to eliminate the Coulomb interaction
between electrons (a single electron per cycle is produced) and to properly control the
interaction between the electron and ions (parent and residual ones).
The degeneracy factor for this source is expected to be: ~ 10
-2
25
Protons and Ions Sources
Protons and Ions Sources
Particle Sources
F. Sannibale
In most protons and ion sources a gas of neutral atoms In most protons and ion sources a gas of neutral atoms
or molecules is heated into a or molecules is heated into a plasma plasma state were ions state were ions
and electrons are dissociated and move independently and electrons are dissociated and move independently
as free particles. as free particles.
Heating mechanism can be of various kind: Heating mechanism can be of various kind:
thermal, electrical, or light (ultraviolet light thermal, electrical, or light (ultraviolet light
or intense visible light from a laser). or intense visible light from a laser).
In a source, the ions are then extracted from the In a source, the ions are then extracted from the
plasma and accelerated. plasma and accelerated.
Neutral gas of practically any specie of atom can be Neutral gas of practically any specie of atom can be
produced and used in sources. For example, neutral gas produced and used in sources. For example, neutral gas
of metals can be obtained by heating the solid element of metals can be obtained by heating the solid element
inside ovens inside ovens
26
Discharge Based
Discharge Based
Ions Sources
Ions Sources
Particle Sources
F. Sannibale
In In Penning Penning discharge sources, discharge sources, Magnetrons Magnetrons and and Plasmatrons Plasmatrons a high a high
voltage discharge (arc) in 0.001 to 1 voltage discharge (arc) in 0.001 to 1 Torr Torr pressure is used for generating pressure is used for generating
the plasma. the plasma.
Figure by C.E. Hill
CERN
The ions then diffuse out from an aperture on the plasma chamber The ions then diffuse out from an aperture on the plasma chamber
and are accelerated by the voltage between cathode and anode. and are accelerated by the voltage between cathode and anode.
The arc electric field The arc electric field
accelerates the electrons accelerates the electrons
and a magnetic field and a magnetic field
makes them move on makes them move on
spiraling orbits inside the spiraling orbits inside the
plasma ionizing more plasma ionizing more
atom along their atom along their
trajectory. trajectory.
27
ECR Sources
ECR Sources
Particle Sources
F. Sannibale
Electron Cyclotron Resonance (ECR) sources Electron Cyclotron Resonance (ECR) sources, ,
exploit this mechanism. exploit this mechanism.
In such sources there is no cathode and the In such sources there is no cathode and the
average lifetime and reliability is very good. average lifetime and reliability is very good.
Non Non- -relativistic particles in a constant magnetic field move relativistic particles in a constant magnetic field move
on a circular trajectory at a on a circular trajectory at a constant revolution frequency constant revolution frequency
independently from their energy (cyclotron principle): independently from their energy (cyclotron principle):
Figure by NSCL- Michigan State University
m
eB
=
0
Lets consider a plasma immersed in a Lets consider a plasma immersed in a

solenoidal solenoidal field. Applying an electromagnetic field. Applying an electromagnetic
field with frequency field with frequency
0 0
, the electrons in the , the electrons in the
plasma will resonate at their cyclotron plasma will resonate at their cyclotron
frequency gaining energy from the field. frequency gaining energy from the field.
The electrons will describe spiraling orbits The electrons will describe spiraling orbits
with increasing radius and ionizing additional with increasing radius and ionizing additional
atoms along their path. atoms along their path.
28
Negative Ion Sources
Negative Ion Sources
Particle Sources
F. Sannibale
The same schemes described for positive ions generation can be The same schemes described for positive ions generation can be used for used for
generating negative ions, (penning, magnetrons, ). generating negative ions, (penning, magnetrons, ).
In a In a surface source surface source, ~ a mono , ~ a mono- -layer of Cesium on the source surface strongly layer of Cesium on the source surface strongly
increase the production of H increase the production of H
- -
. Collision of the plasma particles with the Cs . Collision of the plasma particles with the Cs
surface generates surface generates desorption desorption of ions including the desired H of ions including the desired H- -. .
The physical processes in negative ion sources are still poorly The physical processes in negative ion sources are still poorly understood understood
but three types of source are generally recognized; surface, vol but three types of source are generally recognized; surface, volume and ume and
charge exchange. charge exchange.
Negative ions find very important applications in Tandems and i Negative ions find very important applications in Tandems and in injecting n injecting
into accumulator rings by stripping the charge: the process is n into accumulator rings by stripping the charge: the process is non on- -
hamiltonian hamiltonian and the beam and the beam emittance emittance can be reduced. can be reduced.
In In Volume Sources Volume Sources, scattering between the gas , scattering between the gas molecules can generate molecules can generate
negative ions. For example, measurements of H negative ions. For example, measurements of H
- -
ions in large ions in large- -volume, low volume, low- -
pressure hydrogen discharges indicated densities which were much pressure hydrogen discharges indicated densities which were much larger larger
than those predicted by theory. than those predicted by theory.
Double charge exchange Double charge exchange of positive (or neutral) ion beams on alkali metal of positive (or neutral) ion beams on alkali metal
vapor targets was once a favored method of negative ion producti vapor targets was once a favored method of negative ion production. They on. They
are not very efficient in producing high energy H are not very efficient in producing high energy H
- -
but are still very useful for but are still very useful for
producing exotic species of negative ions. producing exotic species of negative ions.
29
Positron Sources
Positron Sources
Particle Sources
F. Sannibale
From quantum field theory, a photon with energy From quantum field theory, a photon with energy
larger than twice the electron rest mass (~1.02 larger than twice the electron rest mass (~1.02 MeV MeV) )
oscillates between the phase of photon and the oscillates between the phase of photon and the
one of one of pair of virtual electron and positron pair of virtual electron and positron. This . This
virtual particles live for an extremely short time for virtual particles live for an extremely short time for
then recombining back into the original photon then recombining back into the original photon
ready for a new cycle to start again. ready for a new cycle to start again.
This is a consequence of the This is a consequence of the Heisemberg Heisemberg indeterminacy principle indeterminacy principle
and these virtual particles cannot be directly detected. and these virtual particles cannot be directly detected.
Anyway, if the photon during this virtual Anyway, if the photon during this virtual
particle phase passes close to an atom nucleus, particle phase passes close to an atom nucleus,
the interaction between the nucleus fields and the interaction between the nucleus fields and
the pair will allow for the virtual particles to the pair will allow for the virtual particles to
become real and separate from each other. become real and separate from each other.
30
Positron Sources
Positron Sources
Particle Sources
F. Sannibale
In existing positron sources, the high energy photons required In existing positron sources, the high energy photons required for pair for pair
production are generated by first impinging a high current elect production are generated by first impinging a high current electron ron
beam on a high Z metallic target. beam on a high Z metallic target.
1 mm r.m.s. Beam size @ the PC
100 mA
Positron current at the
Linac end (550 MeV)
250 MeV Energy at the PC
> 4 A
Current at the positron
converter (PC)
The electrons penetrating the material are deflected by the nuc The electrons penetrating the material are deflected by the nuclei fields lei fields
and radiate high energy photons. These photons interact with the and radiate high energy photons. These photons interact with the nuclei nuclei
finally generating the pairs. finally generating the pairs.
The newborn positrons leaving the target are separated from the The newborn positrons leaving the target are separated from the
electrons, captured and accelerated to higher energies in a dedi electrons, captured and accelerated to higher energies in a dedicated cated
section of the section of the linac linac optimized for the task. optimized for the task.
DA DA NE NE
The The Frascati Frascati - -factory factory
31
Antiproton Sources
Antiproton Sources
Particle Sources
F. Sannibale
Figure from CERN web site
Existing sources of antiprotons ( Existing sources of antiprotons (Fermilab Fermilab and CERN) exploit the and CERN) exploit the
proton proton- -antiproton pair production mechanism when high energy protons antiproton pair production mechanism when high energy protons
scatters on the nuclei of a metallic target generating pairs. scatters on the nuclei of a metallic target generating pairs.
Production rate is very Production rate is very
small: small:
~ 10 ~ 10
- -5 5
antiproton/proton antiproton/proton
~ 10 ~ 10
11 11
antiproton/hour antiproton/hour
32
Neutron Sources
Neutron Sources
Particle Sources
F. Sannibale
An alternative scheme An alternative scheme fo fo generating neutrons generating neutrons
is given by the so is given by the so- -called called spallation spallation neutron neutron
source source, where a high energy , where a high energy- -high power high power
accelerator produces pulsed neutron beams by accelerator produces pulsed neutron beams by
bombarding a target with intense proton beams. bombarding a target with intense proton beams.
The more efficient neutron sources are nuclear reactors. The more efficient neutron sources are nuclear reactors.
However, they cannot be built because international treaties However, they cannot be built because international treaties
prohibits civilian use of highly enriched uranium U prohibits civilian use of highly enriched uranium U
235 235
. .
1 1 GeV GeV Protons at target Protons at target
1.4 MW Proton Power at the Target 1.4 MW Proton Power at the Target
24 kJ/pulse 24 kJ/pulse
1.5 x 10 1.5 x 10
14 14
protons /pulse protons /pulse
> 1.5 G$ > 1.5 G$
33
Possible Homework
Possible Homework
Particle Sources
F. Sannibale
Calculate the RF frequency for an ECR H Calculate the RF frequency for an ECR H
+ +
source with a source with a
solenoidal solenoidal field of 0.5 T. field of 0.5 T.
Calculate the minimum energy in Calculate the minimum energy in eV eV units that a photon units that a photon
should have to potentially generate a proton should have to potentially generate a proton- -antiproton pair. antiproton pair.
Estimate the number of hours required to store 100 Estimate the number of hours required to store 100 mA mA of of
antiprotons in the antiprotons in the Tevatron Tevatron at the at the Fermilab Fermilab. The ring circumference . The ring circumference
is ~ 6400 m, the beam energy is 980 is ~ 6400 m, the beam energy is 980 GeV GeV. Assume an injection rate . Assume an injection rate
of about 6.5 x 10 of about 6.5 x 10
11 11
antiprotons/hour. Remember that the antiproton antiprotons/hour. Remember that the antiproton
mass is ~ 1.6726 x 10 mass is ~ 1.6726 x 10
- -27 27
kg. kg.
General introduction to the labs
General introduction to the labs
+
+
Homework
Homework
Soren Soren Prestemon Prestemon
Two
Two
-
-
dimensional static fields
Simulation and
Measurements
Soren Prestemon
Magnetic field described in terms of potentials: Magnetic field described in terms of potentials:
In free-space, the magnetic field can be determined from a scalar potential V:
( )
( )
results from 0 and identity 0, B V B V V = = =
G G
and from a vector potential A:
( ) ( )
results from 0 and identity 0, B A B A A = = =
G G G G G
In two dimensions, A is a scalar and we can use complex notation In two dimensions, A is a scalar and we can use complex notation: :
( )
*
x y
dF d
B iB B i i A iV
dz dz
= = = +
F is an analytic function in a good F is an analytic function in a good- -field region, and can be expanded as a field region, and can be expanded as a
Taylors Taylors series for |z|<|z series for |z|<|z
0 0
|=r |=r
0 0
: :
( )
0
( ) ( )
i n
n
n
F z F x iy F re c z
= + = =
( , ) ( , ) ( , )
z x iy
df f f
i
dz x y
f x y u x y iv x y
= +

= =

= +
u v
x y
u v
y x

=

=

( )
1
*
1
( )
n
n
n
B z i nc z

=

Cauchy
Riemann
1
0
0
1
0
0
n n n
n n
n
n n
n
c i
n
a r
B
n
b r
B

= +
=
=
( )
1
*
0
1
0
( )
n
n n
n
z
B z B a ib
r

= +

Note: the coefficients are a function of

the reference radius and the
characteristic field. By tradition:
a
n
: skew coefficients
b
n
: normal coefficients
Note: other definitions of a
n
and b
n
exist!
Two
Two
-
-
Simulation and
Measurements
Soren Prestemon
Magnetic field described in terms of Magnetic field described in terms of multipoles multipoles: :
( )( )
( )
( )
0
2 2 2 2
0 1 1
0 0
2 2 2 2
0 1 1
0 0
( , ) ( , ) ( , )
( , ) ...
2
( , ) ...
2
n
n
n
F x y A x y iV x y i x iy
b a
A x y B b x a y x y xy
r r
a b
V x y B a x b y x y xy
r r

= + = + +

= + + + +

= + + +

Example: V describes geometry
of magnetized surfaces to yield
a multipole field;
for a pure normal dipole:
only b
1
non-zero
b
1
y=+/-V
0
( )( ) ( )
( )
0 0
0 0
1
0
0 0
1
0
( , ) ( , ) ( , )
( , ) cos( ) sin( )
( , ) cos( ) sin( )
n
n
i
n n
n n
n
n n
n
n
n n
n
F x y A x y iV x y i x iy i re
b a r
A r B r n n
n n r
a b r
V r B r n n
n n r
= + = + + = +

= +

= +

In cylindrical coordinates:
V
0
-V
0
g
0
2
y
V
B
g
=
x
y
Example: A can be used to find
current distribution on a cylinder
to yield a given multipole field
(see homework);
for a pure normal dipole:
only b1 non-zero
Things to consider
For iron-dominated magnets, Amperes law
provides much insight:
When possible, the simulations should correspond
to the geometry used in the measurements
Lab grades will be based on your understanding
of the technical issues; if some measurements do
not look clean, discuss possible reasons and
how the measurements could be improved with
appropriate equipment
General issues
General issues
Simulation and
Measurements
Soren Prestemon
The purpose of the labs and simulations is to provide insight in The purpose of the labs and simulations is to provide insight into to
hardware and design software associated with accelerators hardware and design software associated with accelerators
each report should include discussion of issues each report should include discussion of issues
Working as a group
Read through the days lab/simulation
assignment and plan your approach for
efficiency
Proceed with the experiments in a safe and
orderly manner
Be sure power supplies are off when connecting
leads and/or performing assembly
Be careful with heavy/clumsy magnet parts (in
particular the dipole measurement)
The Hall probes are sensitive, expensive
equipment; handle with care. You can tape them
to glass rods, etc, but do not allow the tape to
adhere to the probe itself.
Clean up your experimental area after finishing
the days lab; messy areas will be noted!
Remember that the homework is due after both
the lab and simulation component are finished;
plan your days accordingly
=
g
0
2
In air: 1
In iron: 1 0
iron
iron
H dl I Hg H dl
B H
H dl
= = +
=
=
>>

G G G G
G G
v
Magnetostatics, ,e.g. dipole and quadrupole magnetic
fields, can be calculated using the code Poisson. Example
scripts (*.am) are available. These are evaluated by right-
click,run-autofish). The resulting *.T35 files can be
plotted (double-click) or data evaluated (right-click,
Interpolate). Multipole data is available in the
OUTPOI.txt file.
The program SynRad provides radiation properties for a
wide variety of storage rings and SR sources, including
brightness, flux, and power calculations. New storage ring
and source parameters can also be analyzed.
The program BeamOptics_APS.exe provides a forum for
the design of a storage ring. The lattices of existing rings
can be reviewed, or a new ring can be built from the
ground up. This program will serve as the foundation for
the final project: Storage Ring Design.
Software
Software
Simulation and
Measurements
Soren Prestemon
The software is generally intuitive, and is best learned by tria The software is generally intuitive, and is best learned by trial. Input files l. Input files
are available that can be copied and modified as needed for the are available that can be copied and modified as needed for the
problems. problems.
Software:
Resonant frequencies of pillbox cavities are purely a
function of the geometry. The code TMModeFreq.tcl is
in your folder and will evaluate these frequencies.
Wideroe drift tube lengths can be calculated using the
script calcDriftTubeLengths.tcl, given linac rf frequency,
initial beam energy, energy gain per gap, and number of
drift tubes,
Cavity fields can be calculated and plotted using the
script runUrmel.tcl, which reads a cavity description file
*.urmi.
FODO cells can be analyzed with the script
FODOcells.tcl, which plots Twiss functions ( and
functions) and matched and mismatched trajectories.
Necktie diagrams, which describe FODO cell stability
regions, can be generated with the script
NecktieDiagrams.tcl. Either the trace of the transfer
matrix or the phase advance can be plotted.
Software location:
Please use your c:\studenti directory for all calculations. If you plan to modify an input file, make a
copy with a new name, so that the original can be reviewed in case of problems during running.
Example input files are available in the folder C:\LANL\Examples.
The code SynRad is located in the folder C:\Program Files\Stanford University\SynRad.
The code BeamOptics_APS.exe is located in C:\Program Files\APS\Beam Optics (APS version).
1. Show that in free-space, the complex function B* is analytic (hint: a complex function is analytic if it
satisfies Cauchy-Riemann)
2. Find the 3
rd
order (I.e. sextupole) terms in the expansion of A and V in Cartesian coordinates
3. Field about an infinite line-current
a) Find the field B
r
about an infinite line current on a circle of radius r
b) Find the field B
y
(x,y=0) for an infinite line located at (x
0
=0,y
0
)
4. The vector potential of a line current located at is given by
What is the multipole field associated with the current distribution on the circle r=r
0
defined by
Hint: replace I in the expression for A with I() and integrate from 0 to 2
5. Assume I
0
=10kA, r
0
=0.01meters, and m=1. What is the field (Bx,By) at (x=0.005,y=0).
ote:
0
(permeability of free space) = 4 10
-7
Homework
Homework
Simulation and
Measurements
Soren Prestemon
0
( , ) r
( )
0
1
0
1
( , ) ( )
2
n
n
I r
A r Cos n
n r

=

=

( )
0 0
( , ) I r r I Cos m = =
( , )
( , )
z
A r
B r
r

Fundamentals of Accelerator Physics and Technology (Presentation Slides) (2006) WW

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fundamentals of Accelerator Physics and Technology (Presentation Slides) (2006) WW

Uploaded by

Copyright:

Available Formats

Fundamental Accelerator Theory, Simulations and Measurement Lab Arizona State University, Phoenix January 16-27, 2006

Simulation and Measurement Lab

Dipoles steer the beam, bending it on a circular path;

in booster rings, to bring electrons up to nominal

in the main ring to compensate for radiation energy

Provide intense SR tailored to specific scientific experiments

Good Wake Fields

In addition to this geometric In addition to this geometric

and much more

depends strongly on the depends strongly on the

= Remembering that: Remembering that:

Equal tune shift Equal tune shift

permittivity of free space

IV. (Amperes Law modified by Maxwell)

In free space combining Gauss Law and Faradays Law

Equation of a wave in three dimensions is

Equation of a wave in three dimensions is

For propagating waves For propagating waves v v

Synchronicity condition: Synchronicity condition:

For the transverse planes { For the transverse planes {x x, , p p

Knowledge of the function (s) along the line allows

We just found: We just found:

Crossing the transition energy Crossing the transition energy

Max Planck received the 1919 Nobel for the

Lets consider a plasma immersed in a Lets consider a plasma immersed in a

Note: the coefficients are a function of

You might also like