Linear Accelerator Technology

David Alesini
(INFN-LNF, Frascati, Rome, Italy)
 ACKNOWLEDGEMENTS
Several pictures, schemes, images and plots have been taken from papers and presentations reported at
the end of the presentation.

I would like to acknowledge all the following authors:

Hans Weise, Sergey Belomestnykh, Dinh Nguyen, John Lewellen, Leanne Duffy, Gianluigi Ciovati,
Jean Delayen, S. Di Mitri, R. Carter, G. Bisoffi, B. Aune, J. Sekutowicz, H. Safa, D. Proch, H.
Padamsee, R. Parodi, E. Jensen, Paolo Michelato, Terry Garvey, Yujong Kim, S. Saitiniyazi, M.
Mayierjiang, M. Titberidze, T. Andrews, C. Eckman, Roger M. Jones, T. Inagaki, T. Shintake, F.
Löhl, J. Alex, H. Blumer, M. Bopp, H. Braun, A. Citterio, U. Ellenberger, H. Fitze, H. Joehri, T. Kleeb,
L. Paly, J.Y. Raguin, L. Schulz, R. Zennaro, C. Zumbach. Detlef Reschke, David Dowell, K. Smolenski,
I. Bazarov, B. Dunham, H. Li, Y. Li, X. Liu, D. Ouzounov, C. Sinclair
WHAT DOES IT MEANS LINAC TECHNOLOGY? …For FEL and ERL…
Electron
sources

NC Accelerating
structures
Waveguide
components

RF sources Fabrication
techniques

Power Cryostat for SC

distribution structures

But also: quadrupoles,

magnets, vacuum,
beam diagnostics SC Accelerating Tuning, RF
devices,… structures measurements
 ACCELERATING CAVITIES
P RF in Ez Direction of propagation
To accelerate charged particles, the RF wave must
have an electric field along the direction of
propagation of the particle. There are basically two
possibilities: z
1-Using standing wave (SW) TM010-like modes in a P RF in P RF out
resonant cavity (or multiple resonant cavities) in Ez
which the beam is synchronous with the resonating P RF TW
field;

2-Using a travelling wave (TW) disk loaded structure z

operating on the TM01-like mode in which the RF
wave is co-propagating with the beam with a phase
velocity equal to the beam velocity (c for e-). ⇒The structures are powered by RF generators (typically klystrons).

⇒The cavities (and the related LINAC technology) can be of

different material:
• copper for normal conducting (NC, typically TW) cavities;
• Niobium for superconducting cavities (SC, typically SW);

⇒We can choose between NC or the SC technology depending on

the required performances in term of:
• accelerating gradient (MV/m);
• RF pulse length (how many bunches we can contemporary
accelerate);
• Duty cycle: pulsed operation (i.e. 10-100 Hz) or continuous wave
(CW) operation;
• Average beam current.
SW CAVITIES
 SW CAVITIES PARAMETERS: Vacc, Pdiss, W
To compare different technologies is necessary to define
some parameter that characterize each accelerating Ez
structure.

ACCELERATING VOLTAGE
We suppose that the cavities are
[
E z ( z, t ) = Re E z ( z ) e jωRF t ]
powered at a constant frequency fRF.
The maximum energy gain of a particle
crossing the cavity at a velocity v (∼c for z
electrons) is obtained integrating the B E
time-varying accelerating field sampled
by the charge along the trajectory:
z
jωRF
Vacc = ∫ E ( z) e
cavity
z
v
dz

MODE TM010
DISSIPATED POWER
Real cavities have losses. STORED ENERGY
Surface currents (related to the surface magnetic field
) “sees” a surface resistance and dissipate The total energy stored in the cavity:
energy, so that a certain amount of RF power must be 644 energy density
47444 8
provided from the outside to keep the accelerating field
1 r 2 1 r 2
at the desired level. The total dissipated power is: W = ∫  ε E + µ H  dV
power density cavity 4 4 
6474 8 volume
1
Pdiss = ∫ Rs H tan 2
dS NC cavity (Cu 3 mΩ at 1 GHz)
cavity
2 SC cavity (Nb at 2 K 10 nΩ at 1 GHz)
wall
 SW CAVITIES PARAMETERS: R, Q, R/Q
ACCELERATING VOLTAGE (Vacc) DISSIPATED POWER (Pdiss) STORED ENERGY (W)

SHUNT IMPEDANCE QUALITY FACTOR

W
The shunt impedance is the parameter that qualifies Q = ω RF NC cavity Q∼104
Pdiss SC cavity Q∼1010
the efficiency of an accelerating mode. The highest
is its value, the larger is the obtainable accelerating
voltage for a given power. Traditionally, it is the
quantity to optimize in order to maximize the
accelerating field for a given dissipated power:
2
R/Q
R=
V acc
[Ω ] 2
R Vacc
2 Pdiss =
Q 2ωRFW
NC cavity R∼1MΩ SC cavity R∼1TΩ The R/Q is a pure geometric
qualification factor. It does not depend
on the cavity wall conductivity. R/Q of a
single cell is of the order of 100.
 SW CAVITIES : EQUIVALENT CIRCUIT AND BANDWIDTH
The previous quantities plays crucial roles in the evaluation of the cavity Pin=1 MW
performances. Let us consider the case of a cavity powered by a source (klystron) at R/Q=100
a constant frequency in CW and at a fixed power (Pin). β=1 (no reflections, Pdiss=Pin)
fres=1 GHz

The reachable Vacc

for a given power is
Frequency domain
proportional to √Q R
Vacc = 2 RPdiss = 2 QPin ∝ Q
Q 

BANDWIDTH

3dB

Modulator cavity
waveguide
and klystron ∆f RF 3 dB

 ∆f
∆f RF 1  RF 3 dB NC = 100kHz
= ⇒
3 dB

f RF Q  ∆f RF 3 dB SC < 1Hz

Pin
 SW CAVITIES : FILLING TIME AND DISSIPATED POWER
Let us now consider the case of a cavity powered by a source (klystron) in pulsed mode at a frequency fRF=fres. Let as
calculate the power we need from the klystron (and the dissipated one) to obtain a given accelerating voltage
Time domain

One needs several filling

times to completely fill
the cavity

FILLING TIME
2Q
τF =
ωRF

Pdiss ∝1/Q
2
Vacc 1 1
Pin = ∝
R Q Q
2 
The reachable Vacc for a given power is Q 
proportional to √Q but, on the other hand,
the filling time is ∝Q
τF NC
≈ µs Prefl to the generator (it
that has to be
τF SC
> 100ms protected!)
 SW CAVITIES : RF STRUCTURE AND BEAM STRUCTURE
The “beam structure” in a LINAC (or ERL) is directly related to the “RF structure”. There are basically two possible type of
operations:
• CW (continuous wave) ⇒ allow, in principle, to operate with a continuous beam
• PULSED OPARATION ⇒ there are RF pulses at a certain repetition rate (Duty Cycle (DC)=pulsed width/period)

RF power
RF pulses

t
Amplitude 103-108 RF periods
Bunch spacing

Because of the very low power dissipation and low RF power required to achieve a certain Vacc, the SC structures allow
operation at very high Duty Cycle (DC) up to a CW operation.
 Pin MULTI-CELL SW CAVITIES Pin

R .

• In a multi-cell structure there is one RF input coupler. As a

consequence the total number of RF sources is reduced, with a
nR
simplification of the layout and reduction of the costs;

• The shunt impedance is n time the impedance of a single cavity

• They are more complicated to fabricate than single cell cavities;

• The fields of adjacent cells couple through the cell irises and/or Ez
through properly designed coupling slots.

• The N-cell structure behaves like a system composed by N coupled

oscillators with N coupled RF modes. The modes are characterized by z
a cell-to-cell phase advance given by:
nπ
∆φ n = n = 0,1, ..., N − 1
N −1
• The most efficient mode (and generally used) is the π mode.

• Field amplitude variation from cell to cell should be small for maximum
acceleration efficiency⇒ necessity of tuning

• It is possible to demonstrate that over a certain number of cavities

(>10) the overlap between adjacent modes can be a problem from the
tunability and operational point of view.
TW CAVITIES
 TW CAVITIES: BASICS
In TW structures an e.m. wave with Ez≠0 travel together with the beam in a special guide in which the phase velocity of
the wave matches the particle velocity (v). In this case the beam absorbs energy from the wave and it is continuously
accelerated.

CIRCULAR WAVEGUIDE IRIS LOADED STRUCTURE

In order to slow-down the wave

ω
ω ω phase velocity, iris-loaded
By solving the wave equation it periodic structure are used. The
turns out that an e.m. wave field in this kind of structures is
propagating in a constant cross that of a special wave travelling
section waveguide will never be
synchronous with a particle
ωRF ωRF
within a spatial periodic profile.
The structure can be designed to
beam since the phase velocity is have the phase velocity equal to
always larger than the speed of the speed of the particles. This
light c. The first propagating allows acceleration over large
mode with Ez≠0 is the TM01 mode distances.
β β
Periodic in z
β* β* 2π/D
p  of period D
E z TM = E 0 J 0  01 r  cos(ω RF t − βz )
 a  = E P (r, z ) cos(ω RF t − βz )
01
Ez TM 01− like

MODE TM01 MODE TM01-like

 TW CAVITIES PARAMETERS: r, α, vg
Similarly to the SW cavities it is possible to define some figure of merit of the TW structures

Shunt impedance per unit length.

Similarly to SW structures the
higher is r, the higher the available
accelerating field for a given RF
power.

D z
jω RF
Vz = ∫ Ez ⋅ e c
dz single cell accelerati ng voltage 2
Field attenuation constant:
E because of the wall dissipation,
r=
0
acc
V
E acc = z average accelerati ng field in the cell p diss the RF power flux and the
D
accelerating field decrease along
1  pdiss
Re E × H  ⋅ zˆdS
*
Pin = ∫
Section
2  
average input power (flux power)
α= the structure.
1 2
2 Pin
Pdiss = Rs ∫ H tan dS average dissipated power in the cell
2 cavity Group velocity: the velocity of the
P
wall
v g = in energy flow in the structure (∼1-
Pdiss w
p diss = average dissipated power per unit length 2% of c).
D
w
Q = ω RF
energy density
644 47444 8
1 r 2
W = ∫  ε E + µ H  dV
1 r 2 Working mode: defined as the
stored energy in the cell pdiss
cavity   phase advance of the fundamental
4 4
∆φ = β D
volume

W harmonic over a period D. For

w= average stored energy per unit length
D several reasons the most common
π/3
mode is the 2π
 TW CAVITIES: EQUIVALENT CIRCUIT AND τF
In a TW structure, the RF power enters into the cavity through an
input coupler, flows (travels) through the cavity in the same
direction as the beam and an output coupler at the end of the
structure is connected to a matched power load.
If there is no beam, the input power reduced by the cavity losses
goes to the power load where it is dissipated.
In the presence of a large beam current, however, a fraction of the
TW power is transferred to the beam.

In a purely periodic structure,

made by a sequence of identical
cells (also called “constant
impedance structure”), α does
not depend on z and both the RF
power flux and the intensity of the
accelerating field decay
exponentially along the structure :
E z (z ) = E 0 e −αz
The filling time is the time High group velocities allow reducing the
necessary to propagate the RF duration of the RF pulse powering the
wave-front from the input to structure. However since:
the end of the section of length
L is: L Low group velocity is
τF = preferable to increase the vg =
Pin
vg effective accelerating field w
Differently from SW cavities for a given power flowing w ∝ E2
after one filling time the cavity in the structure.
L is completely full of energy
 TW CAVITIES: PERFORMANCES (1/2)
Just as an example we can consider a C-band (5.712 GHz) r=82 [MΩ/m]
accelerating cavity of 2 m long made in copper. α=0.36 [1/m]
vg/c=1.7%
τF=400 ns (very short if compared to SW!)

Field attenuation
due to the
copper
dissipations

τF
Output power (dissipated into RF STRUCTURE AND BEAM STRUCTURE
Input power the RF load): it is not convenient
to have very long RF structures TW structures have very short filling time (<1µ µs) and allow
because their efficiency operation in pulsed mode with high peak power (tens of MW
decreases over a certain length per structure) and relatively high accelerating field (>50-100
(2-3 m depending on the MV/m), with short RF pulses (1 µs) and low repetition rate
operating frequency). (10-100 Hz) and low DC (10-3-10-2 %) in single or few bunches
 TW CAVITIES: PERFORMANCES (2/2)
If we compare the performances of this copper structure with the same cavity made on
a superconducting material it is quite easy to understand that it is not convenient to
use TW SC structures:
r=82 [MΩ/m]⇒[TΩ/m]
⇒we do not gain in term of Eacc as we do for SW structures and as a consequence we α=0.36 [1/m]⇒∼0
do not gain in term of Vacc⇒ direct consequence of the TW mechanism (no field build vg/c=1.7%
up effects!) τF=400 ns
⇒for a short structure all power is dissipated into the RF load

⇒It is, in principle possible to gain with TW SC if we increase the length of the structure
and, as a consequence, the RF pulse length but we have problems of available power
sources (high power/long RF pulses) and cavity construction
 TW CAVITIES: CONSTANT GRADIENT STRUCTURES
It is possible to demonstrate that, in order to keep the
accelerating field constant along the structure, the iris
apertures have to decrease along the structure in
such a way that the field attenuation is compensated
by the increase of the stored energy (with consequent
decrease of the group velocity).

In general the constant gradient structures are more

efficient than constant impedance ones, because of
the more uniform distribution of the RF power along
them.
MATERIAL
 NORMAL CONDUCTING (NC) MATERIAL: COPPER
power density
6474 8
1
Pdiss = ∫ Rs H tan 2
dS
cavity
2
E0 σ
wall Rs vs RF FREQUENCY
conductor
The microwave surface resistance of a normal metals is expressed by:
z
πf RF µ 0 1 1 Skin depth: penetration of
δ
Rs = = δ = the EM field ans surface
σ σδ πf RF µ 0σ currents inside the metal

For copper: σ=5.7x108 S/m⇒Rs(@1 GHz)≅3mΩ, δ<1 µm

Rs AT LOW TEMPERATURE ANOMALUS SKIN EFFECT
At low temperature and at high frequency, NC material there is a mechanism called
“anomalous skin effect” that increases the conductivity with respect to the DC case. For
copper, as example, at microwave frequencies and cryogenic temperatures, one can see
that, although the DC conductivity increases by a factor 100, the anomalous skin effect
allows only a decrease of a factor 6 in the surface resistance. This shows that it is
definitely not convenient to cool an NC metal to cryogenic temperatures.
COPPER
The most widely used NC metal for RF structures is OFHC
copper (Oxigen free high conductivity) for several reasons:

1) Easy to machine (good achievable roughness at the few

nm level)
2) Easy to braze/weld
3) Easy to find at relatively low cost
4) Very good electrical (and thermal) conductivity
5) Low SEY (multipacting fenomena)
6) Good performances at high accelerating gradient
 SUPERCONDUCTING (SC) MATERIAL: NIOBIUM (Nb)
⇒The SC was discovered in 1911.
⇒ For SC elements at T<Tc in DC regime the resistance is 0.
⇒ In AC (RF) regime the surface resistance of a SC is always larger
than 0 (even if very small if compared to NC element).
⇒For frequencies below 10 GHz (and T<Tc/2) the experimental data
are well described by the empirical relation:
exponential decrease with
Depends on the material temperature (high
frequency cavity >1 GHz

ω2 −α
Tc have to be cooled to

Rs = A + Rres
T reduce the dissipation)
e
T
Square dependence
with frequency Two fluid model or Residual resistance (typically 5-20 nΩ) this term
BCS resistance RBCS dominate the low frequency (10-150 MHz)
resonators. Caused by: magnetic flux trapped in at
NIOBIUM cooldown, surface contaminations, defects,…

The most common material for SC cavities is Nb because:

• Nb has a relatively high transition temperature (Tc=9.25 K).
• SC can be destroyed by magnetic field grater than a critical
field Hc ⇒ Pure Nb has a relatively high critical magnetic field
Hc=170-180 mT.
• It is chemically inert
• It can be machined and deep-drawn
• It is available as bulk and sheet material in any size, fabricated
by forging and rolling
• Large grain sizes (often favoured) obtained by e-beam melting
Instead of bulk or sheet, it can also be coated (e.g. by
sputtering) on Cu
• Other advantages: thermal stability, material cost, possible
optimisation of Rs
 PARAMETERS SCALING WITH FREQUENCY
We can analyze how all parameters (r, Q) scale with frequency and what are the advantages or disadvantages in accelerate
with low or high frequencies cavities.

parameter NC SC ⇒r/Q increases at high frequency

Rs ∝ f1/2 ∝ f2
⇒for NC structures also r increases and this push to
Q ∝ f-1/2 ∝ f-2 adopt higher frequencies
r ∝ f1/2 ∝ f-1 ⇒for SC structures the power losses increases with f2
r/Q ∝f and, as a consequence, r scales with 1/f this push to
adopt lower frequencies
w// ∝ f2 Wakefield intensity:
w┴ ∝ f3 related to BD issues ⇒On the other hand at very high frequencies (>10 GHz)
power sources are less available

⇒Beam interaction (wakefield) became more critical at

high frequency
For FEL and ERL basically:
Compromise ⇒Cavity fabrication at very high frequency requires
SW SC: 500 MHz-1500 MHz between several higher precision but, on the other hand, at low
TW NC: 3 GHz-6 GHz requirements frequencies one needs more material and larger
SW NC: 0.5 GHz-3 GHz machines

⇒short bunches are easier with higher f

LINAC TECHNOLOGY:
NC TW CAVITIES
 NC TW STRUCTURES: ACCELERATING CELLS
⇒Copper structures with many cells (hundred), and input coupler,
and an output coupler connected to an RF load;

⇒The structures operate typically:

with short RF pulses (∼0.5-2µ) in single bunch (or few bunches)
at high peak power (∼50 MW)
at high accelerating field (20-40 MV/m)
on the 2π/3 mode
in pulsed mode at a low rep. rate (10-100 Hz) and low DC
in S or C band (3 GHz, 6 GHz)

⇒the TW cells are optimized to have high shunt impedance, low filling
time and the most important role is played by the iris dimensions.

⇒Cooling pipes are inserted or brazed around the cells to guarantee

the temperature stability of the structures avoiding detuning of the
structure under high power feeding .
 NC TW STRUCTURES: COUPLERS
The structures are fed by waveguides. The coupler,
realized by a slot in the waveguide, matches the TE10 mode
of the waveguide with the traveling wave mode (TM01-like).

J-type couplers or integrated splitters allow compensating

the dipole kick in the coupling cells.

Input Output
TW cells coupler
coupler

Rounded shapes in the

couplers (low magnetic filed)
allows to reduce the pulsed
heating.

Race track profiles allow to

compensate the quadrupole
distortions of the field in the
coupling cells
 NC TW STRUCTURES: FABRICATION
The cells and couplers are fabricated with
milling machines and lathes starting from
OFHC forged or laminated copper with
precisions that can be of the order of few um
and surface roughness <50 nm.

The cells are then piled

up and brazed together
in vacuum or hydrogen
furnace using different
alloys at different
temperatures (700-1000
C) and/or in different
steps.
 NC TW STRUCTURES: TUNING
To compensate deformations and imperfections that can also occur
during the brazing process, tuning is often necessary. The standard
method is to measure the field inside by a perturbation technique (Steele
method) and to “tune” the phase advance per cell to the correct value by
deforming the outer volume of the cells with deformation tuners.
Deformation of tuner
the volume

Metallic or
dielectric bead

NA
NC TW STRUCTURES: RF WAVEGUIDE NETWORK AND POWER SOURCES
TW structures require high peak power pulsed sources. To this purpose klystron+RF compression systems (SLED) are usually
adopted

Courtesy T. Inagaki et al.

LINAC TECHNOLOGY:
SC SW CAVITIES
 SC SW STRUCTURES: ACCELERATING CELLS
Typically the SC SW structures are:

⇒ Single and multi cell structures (up to ∼10)

⇒ Operating on the π mode
⇒ The irises have an elliptical shape to:

• minimize Esurf/Eacc (and then the electrons field emission)

• minimize Bsurf/Eacc (break-down of superconductivity for
Nb is 170-190 mT)
• suppress multipacting
• increase the machinability and cleanability

Also for this type of cavities the iris radius play a

fundamental role in the design of the structures
5

4.5
Bpeak/Eacc
Bpeak/Eacc [mT/(MV/m)]

3.5
(R/Q) [kΩ /m],
Epeak/Eacc,

3
Bsurf mT
= 4.2 2.5
Epeak/Eacc
E acc MV m 2

1.5
For TESLA cavities and (R/Q)
1
then theoretically the
0.5
maximum achievable Eacc
26 28 30 32 34 36 38 40
is about 40-45 MV/m ri [mm]
 SC SW STRUCTURES: POWER COUPLERS
RF input
Coaxial-type electric couplers have the widest applications, because magnetic coupling with
waveguides or loops can create hot spots in the cavities with additional design complications

Vacuum barriers (windows). They prevent contamination

QEXT tunability. For many
of the SC structure. Obviously these barrier are necessary
accelerators it is necessary to tune
also in normal conducting accelerators but the demand
the coupling changing the
on the quality of the vacuum and reliability of the
penetration of the antenna in the
windows are less stringent. The failure of a window in
pipe.
superconducting accelerator can necessitate very costly
and lengthy in repart. They are made, in general, in Al2O3.
Ceramic material have a SEY that stimulates the
multipacting activity. Ti-coating can reduce this
phenomena.

Thermal barrier: The RF power

must be fed into the cold
superconducting cavity and in the
coupler we cross the boundary
between the room temperature and
the low-temperature environment
 SC SW STRUCTURES: HOM DAMPERS
SC cavities can be used to accelerate train of bunches. As bunch traverses a
cavity, it deposits electromagnetic energy on Higher order modes (HOM)
described in terms of long range wakefields. Subsequent bunches (or the
same bunch in several turns like in ERLs) may be affected by these fields
causing instabilities and additional heating of accelerator components.

Several approaches are used:

• Loop couplers (several per cavity

for different modes/orientations)
• Waveguide dampers
• Beam pipe absorbers (ferrite or
ceramic)
 SC SW STRUCTURES: FABRICATION
Nb is available as bulk and sheet material in any size,
fabricated by forging and rolling. High Purity Nb is made
by electron beam melting under good vacuum.
The most common fabrication techniques for the cavities
are to deep draw or spin half-cells.
Alternative techniques are: hydroforming, spinning an
entire cavity out of single sheet or tube and Nb sputtering

After forming the parts are electron beam welded

together

CAVITY TREATMENT
The cavity treatment after the welding is quite complicated
and require several steps between:
• buffered chemical polishing (BCP), electropolishing and
etching to remove surface damaged layers of the order of 100
µm
• rinsed with ultraclean water also at highpressure (100 bar)
• Thermal treatments up to >1000 C to diffuse H2 out of the
material increasing the Nb purity (RRR)
• high-temperature treatment with Ti getter (post-purification)
• RF tuning
 SC SW STRUCTURES: CRYOMODULE
The cavity is immersed in a liquid helium bath, which is pumped to
remove helium vapor boil-off as well as to reduce the bath temperature.
An RF input coupler and other penetrations create “spurious” sources of
heat losses.
Proper design methods must be used (material choice, heat intercepts,
etc.)
The cold portions of the cryomodule need to be extremely well
insulated, which is best accomplished by a vacuum vessel surrounding
the helium vessel and all ancillary cold components

European XFEL
 REAL PERFORMANCES OF SC CAVITIES: Q vs Eacc (1/2)
Usually the performances of SC cavities are
analyzed by plotting dependence of their quality
factor on the accelerating field.
There are several mechanisms responsible for
additional losses under high power.

The measured surface resistivity is larger than predicted

by BCS theory. Causes are:
• magnetic flux trapped in at cool down
• dielectric surface contaminations (chemical
residues, dust,…)
• NC defects and inclusions
• surface imperfections
• hydrogen precipitates

Multipacting=multiple impact electron amplification (MP) is a resonant

process, when a large number of electrons build up under influence of RF field
(input couplers, cavities, etc.). It needs two conditions:
• electron synchronization with RF field
• electron multiplication via secondary electron emission (SEY).
MP was an early limitation of SRF cavities’ performance.
It was overcome by adopting spherical/elliptical cell shapes.
In severe cases MP may cause quench and limit the cavity field.
Also the RF conditioning can reduce the MP
 REAL PERFORMANCES OF SC CAVITIES: Q vs Eacc (2/2)
Thermal breakdown occurs when the heat
generated at the hot spot is larger than that
can be transferred to the helium bath
causing T>Tc and, as a consequence,
“quench” of the superconducting state

Exponential increase of losses due to acceleration of Field

Emitted electrons. Associated with production of X-rays and
dark current.
The main cause of FE is particulate contamination.
FE can be prevented by proper surface preparation and
contamination control.
It is possible to reduce using High-power Pulsed Processing
(HPP) and/or Helium processing.
 SC SW STRUCTURES: RF FEEDING SYSTEM

The requirements on the stability of the accelerating field in

a superconducting acceleration structure are comparable to
those in a normal-conducting cavity. However, the nature
and magnitude of the perturbations to be controlled are
rather different. Superconducting cavities possess a very
narrow bandwidth and are therefore highly susceptible to
mechanical perturbations. Significant phase and amplitude
errors are induced by the resulting frequency variations.
Perturbations can be excited by mechanical vibrations
(microphonics), changes in helium pressure and level, or
Lorentz forces. Slow changes in frequency, on the time scale
of minutes or longer, are corrected by a frequency tuner,
while faster changes are counteracted by an amplitude and
phase modulation of the incident rf power.
ELECTRON SOURCES
 NC SW STRUCTURES: RF PHOTO-GUNS
RF guns are used in the first stage of electron beam generation in FEL and
acceleration.
• Multi cell: typically 2-3 cells
• SW π mode cavities
• operate in the range of 60-120 MV/m cathode peak accelerating field
with up to 10 MW input power.
• Typically in L-band- S-band (3-1 GHz) at 10-100 Hz.
• Single or multi bunch (L-band)
• Different type of cathodes (copper,…)

cathode

The electrons are emitted on

the cathode through a laser
that hit the surface. They are
then accelerated trough the
electric field that has a
longitudinal component on
axis TM010.
 RF PHOTO-GUNS: EXAMPLES

PITZ L-band Gun

Frequency = 1,300 MHz
Gradient = up to 60 MV/m
Exit energy = 6.5 MeV
Rep. rate 10 Hz
Cs2Te photocathode
RF pulse length ∼1 ms
800 bunches per macropulse
LCLS I Normalized rms emittance
Frequency = 2,856 MHz 1 nC 0.70 mm⋅mrad
Gradient = 120 MV/m 0.1 nC 0.21 mm⋅mrad
Exit energy = 6 MeV
Copper photocathode
RF pulse length ∼2 µs
Bunch repetition rate = 120 Hz
Norm. rms emittance
0.4 mm⋅mrad at 250 pC
 RF PHOTO-GUNS: RF WAVEGUIDE DISTRIBUTION

Ceramic
DC1 window
KLY+ Isolator/
MOD circulator

DC2
Ion pump

protect the
klystron from
reflected power.
SW structures Gun probe
always reflect Isolate the vacuum of
power (during RF the waveguides from
transients) that of the gun
 DC PHOTO-GUNS
DC photoguns can be used as electron sources for high average current accelerator (CW, ERL).
In this case the cathode of GaAs(Cs) is used in a DC system. Average currents up to 100 mA can
be achieved.

Cornell DC gun
Gradient = 5 – 10 MV/m
Gun exit energy = 0.35 MeV
GaAs and K2CsSb photocathodes
Bunch repetition rate = 1300 MHz
Norm. rms emittance = 0.5/0.3 µm at 80 pC
Average current = 65 mA (at 50 pC)
The Cornell Photoemission Gun.
POWER SOURCES AND
POWER DISTRIBUTION
 Klystron
RF SOURCES
• Intensity modulation of
DC beam by cavity
• Output cavity
• Both ppulsed and CW
• Typical 0.3-30 GHz
• High power >50 MW´s
• High gain (>40dB)

Inductive Output Tube (IOT)

• Intensity modulation of
DC beam by control grid
• Typical up to 2 GHz
• Higher efficiency than
klystron
• Moderate power (<100
kW)

• Combines power of many transistors

Solid state amplifier • Soft failure mode (single module failure does not
cause failure of the amplifier, just reduction of the
output power)
• Typical up to 2 GHz
• Efficiency is approaching and even exceeding that
of vacuum tubes
• Moderate high power
 WAVEGUIDE RF COMPONENTS
Isolators/circulators

The circulator is a passive non-reciprocal device with 3

or more ports, and protect (isolate) the RF power High power RF loads
sources from microwave power reflected back from a
non-ideal loads. This is possible due to unique magnetic
properties of ferrites that, when properly magnetized,
introduce different phase shift for electromagnetic
Ceramic windows
waves traveling in opposite directions.

Directional
couplers

Waveguide
bend Waveguide
pumping port
EXAMPLES
EXAMPLE: SWISSFEL LINAC (PSI)
 EXAMPLES: JAPANESE XFEL (SPRING-8)

Thermionic gun
The most remarkable feature of
the injector is employing a
thermoionic gun (500 keV CeB6
single crystal). RF beam
manipulations with multi-stage RF
cavities are the necessary to not
degrade the initial emittance.
Damped C-band cavities
 EXAMPLES: EUROPEAN XFEL

Frequency = 1.3 GHz

Gradient = up to 60 MV/m
Exit energy = 6.5 MeV
Cs2Te photocathode
NEW TECHNOLOGIES
 NEW TECHNOLOGY BASED ON CLAMPING FOR HIGH GRADIENT RF
PHOTOGUN
⇒The new SPARC GUN and the ⇒The guns have been tested at
ELI-NP one have been realized high power and with beam
without brazing using a novel ⇒The results demonstrate the use
process developed at LNF-INFN of this novel technique for a high
involving the use of special brightness photoinjectors and can
vacuum/RF gaskets. be extended to more genral RF
structures
⇒The new technique could be
applied to other RF structures
(S-Band, X band,…) and more
high power tests (in X band as
example) would be very useful
to understand the criticalities
and the limits of this new
technology.
 DAMPED/HIGH GRADIENT/HIGH REPETITION RATE C-BAND
ACCELERATING STRUCTURES FOR THE ELI-NP LINAC
⇒ The linac energy booster of the European ELI-NP proposal
foresees the use of 12, 1.8 m long, travelling wave C-Band
structures.
⇒ Because of the multi-bunch operation, the structures integrate
a very effective dipole HOM damping system to avoid beam
break-up (BBU).
⇒ An optimization of the electromagnetic and mechanical design
has been done to simplify the fabrication and to reduce their
cost.
⇒ The high power test on the first full scale structure shown the
feasibility of the 33 MV/m, 100 Hz, long RF pulse operation
 REFERENCES
Hans Weise, DESY, The Electron Accelerator of the European XFEL, presentation at the European XFEL User Meeting 2011, Hamburg, Germany

Sergey Belomestnykh, Brookhaven National Laboratory, Principles of RF superconductivity, Presentation at the USPAS school, Durham, NC, 2013

Dinh Nguyen, John Lewellen and Leanne Duffy, Los Alamos National Laboratory, RF Linac for High-Gain FEL, USPAS School, 2014

Gianluigi Ciovati, Thomas Jefferson National Accelerator Facility, presentations at the USPAS School, 2015.

Jean Delayen, Thomas Jefferson National Accelerator Facility, presentations at the USPAS school, 2015

S. Di Mitri, RF Technology, USPAS school 2015

R. Carter, Review of RF power sources for particle accelerators, CAS School on Radiofrequency engineering, CERN-2005-003, 2005 and RF power
generation, CAS School on RF for accelerators, Ebeltoft, Denmark, 2010, CERN–2011–007

G. Bisoffi, Superconducting Cavities, CAS School on Radiofrequency engineering, SeeimCERN-2005-003, 2005

B. Aune et al., Superconducting TESLA cavities, PRST-AB, VOLUME 3, 092001 (2000)

J. Sekutowicz, Desy, Superconducting Linear Accelerator for the European XFEL, International Workshop on X-ray Diagnostics and Scientific
Application of the European XFELRyn, Poland, 14-17 February 2010 and Superconducting Cavities, CAS on RF for Accelerators, Ebeltoft, Denmark,
8-18 June, 2010.

H. Safa, Surface effects in SCRF cavity, CAS School on Superconductivuty and Cryogenics for accelerators and detectors, CERN-2004-008

D. Proch, RF cavity fabrication, CAS School on Superconductivuty and Cryogenics for accelerators and detectors, CERN-2004-008
H. Padamsee, Designing superconducting cavities for accelerators, CAS School on Superconductivuty and Cryogenics for accelerators and
detectors, CERN-2004-008 and Design Topics for Superconducting RF Cavities and Ancillaries, CAS School on Superconductivity for Accelerators,
Erice 2013, CERN–2014–005

R. Parodi, Couplers and HOM dampers, CAS School on Superconductivuty and Cryogenics for accelerators and detectors, CERN-2004-008

E. Jensen, CERN, Cavity basics, CAS School on RF for accelerators, Ebeltoft, Denmark, 2010, CERN–2011–007 and RF Principles and TM Mode
Cavity, SRF 2015, Whistler, 2015

Paolo Michelato, INFN Milano – LASA, Cavity Processing: EP/BCP, heat treatments, baking and clean room techniques, SRF13 Tutorials,
September 2013.

Terry Garvey, The SwissFEL Linac, Presentation at the John Adams Institute, Oxford University, 8th May, 2014.

Yujong Kim†, S. Saitiniyazi, M. Mayierjiang, M. Titberidze, T. Andrews, and C. Eckman, Performance Comparison of S-band, C-band, and X-band RF
Linac based XFELs, ICFA FLS2012 Workshop, Newport News, USA

Roger M. Jones, Wakefield suppression in high gradient linacs for lepton linear colliders, PRST-AB, 12, 104801 (2009)

T. Inagaki#, K. Shirasawa, T. Sakurai, C. Kondo, T. Ohshima, Y. Otake, and T. Shintake, OPERATION STATUS OF C-BAND HIGH-GRADIENT
ACCELERATOR FOR XFEL/SPRING-8 (SACLA), Proceedings of IPAC2011, San Sebastián, Spain.

K. Togawa,* T. Shintake, T. Inagaki, K. Onoe,† and T. Tanaka, CeB6 electron gun for low-emittance injector, PRST-AB 10, 020703 (2007)

T. Inagaki,* C. Kondo, H. Maesaka, T. Ohshima, Y. Otake, T. Sakurai, K. Shirasawa,† and T. Shintake, High-gradient C-band linac for a compact x-ray
free-electron laser facility, PRST-AB 17, 080702 (2014)

F. Löhl, J. Alex, H. Blumer, M. Bopp, H. Braun, A. Citterio, U. Ellenberger, H. Fitze, H. Joehri, T. Kleeb, L. Paly, J.-Y. Raguin,
L. Schulz, R. Zennaro, C. Zumbach, Status of the SwissFEL C-band Linac, FEL 2014 Conference, August 25-29, 2014, Basel, Switzerland

David Dowell, USPAS school, High Brightness Electron Injectors for Light Sources, 2010

K. Smolenski, I. Bazarov, B. Dunham, H. Li, Y. Li, X. Liu, D. Ouzounov, C. Sinclair, Design and Performance of the Cornell ERL DC Photoemission
Gun, AIP Conf. Proc. 1149, 1077 (2009);