scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

PARMELA simulations of a PWT photoinjector

12 May 2003-Vol. 3, pp 2126-2128
TL;DR: In this paper, a 10+2(1/2)-cell PWT photoinjector can achieve an emittance of 0.56 mm-mrad for a 1 nC bunch charge at a peak field of 55 MV/m.
Abstract: Conventional rf guns such as the BNL/SLAC/UCLA 1.6-cell gun require a high operating peak field typically above 120 MV/m. By contrast, the PWT gun can operate at a much lower peak field with excellent beam properties. A 10+2(1/2)-cell PWT photoinjector can achieve an emittance of 0.56 mm-mrad for a 1 nC bunch charge at a peak field of 55 MV/m; and 1.04 mm-mrad for 2 nC at 60 MV/m. By operating the PWT gun at a low peak field, dark current production from back bombardment of electrons emitted elsewhere in the rf cavity is mitigated as the number of electrons which could back stream to hit the photocathode and the cathode holder is significantly curtailed. The quantum efficiency and lifetime of a semiconductor photocathode, such as GaAs are significantly improved if the surface of the cathode is subjected to a lower peak field.

Summary (2 min read)

PWT PHOTOINJECTOR

  • A compact, 10-30 MeV, photoelectron linac using the Plane Wave Transformer (PWT) design has been under development at DULY Research [1] for wideranging applications in research, medicine and industry.
  • The PWT photoinjector integrates a photocathode directly into a multicell, standing-wave, π-mode linac, in which the open cells in a large vacuum tank are strongly coupled.
  • The beam is focused with emittance compensating solenoids or permanent magnets, and achieves high current with low emittance, producing extremely high brightness.
  • A higher frequency (X-band) version of the PWT would produce an even brighter beam.
  • In addition, the defocusing kick which the beam receives at the exit of the last cell occurs at a significantly higher energy than for a 1.6-cell gun, resulting in less emittance growth.

EMITTANCE CALCULATIONS

  • The authors have performed improved, detailed beam dynamics simulations of the PWT photoinjector using the Los Alamos gun code PARMELA, Version 3.
  • With refined space charge mesh and careful optimization, the performance of the PWT gun was evaluated for three illustrative cases.
  • For a given peak field, the initial injection phase and the solenoidal field are adjusted to give the lowest normalized transverse emittance.
  • They both can achieve very low emittance.

Electron Backstreaming from the PWT Iris to the Photocathode

  • Since electrons have a smaller mass and are hence more mobile, fieldemitted electrons in the rf cavity can potentially reach the cathode, damaging the activated layer.
  • The model includes a drift tube at the center of the endplate.
  • The maximum number of electrons that can reach the cathode occurs at such locations when the rf field magnitude exceeds a threshold value and when the initial rf phase at the iris at the time the electrons are emitted is near zero.
  • Figure 5b is an example of snapshots of electrons emitted from the surface of the first iris and backstream toward the cathode plane under the rf and magnetic fields in the PWT.
  • Since field emission is likely to be most abundant when the surface field is at its maximum (i.e. 90° rf phase); by operating the PWT at a peak field much lower than conventional rf guns, it is expected that dark current emission from the GaAs and its support structure would be considerably mitigated.

Back Bombardment of Electrons Emitted from the Cathode Holder

  • Another PARMELA simulation was performed to evaluate the possible back bombardment of electrons which were first emitted near the PWT end plate and cathode support.
  • At very low peak fields electrons would oscillate near the cathode surface.
  • The operating peak field chosen for the proposed polarized PWT gun appears to be optimum in that it is low enough to discourage back bombardment of electrons from the iris, and yet high enough to avoid back bombardment of electrons from the cathode holder.
  • During rf conditioning, however, it is important that the GaAs cathode is not exposed to low fields.
  • For that reason, the authors will be using a sacrificial cathode during rf conditioning and replacing it with a freshly activated cathode via the load lock.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

PARMELA SIMULATIONS OF A PWT PHOTOINJECTOR*
Y. Luo, D. Newsham, D. Yu, DULY Research Inc., Rancho Palos Verdes, CA 90275
J. Clendenin, Stanford Linear Accelerator Center, Menlo Park, CA 94025
Abstract
Conventional rf guns such as the BNL/SLAC/UCLA
1.6-cell gun require a high operating peak field typically
above 120 MV/m. By contrast, the PWT gun can operate
at a much lower peak field with excellent beam
properties. A 10+2(1/2)-cell PWT photoinjector can
achieve an emittance of 0.56 mm-mrad for a 1 nC bunch
charge at a peak field of 55 MV/m; and 1.04 mm-mrad for
2 nC at 60 MV/m. By operating the PWT gun at a low
peak field, dark current production from back
bombardment of electrons emitted elsewhere in the rf
cavity is mitigated as the number of electrons which could
back stream to hit the photocathode and the cathode
holder is significantly curtailed. The quantum efficiency
and lifetime of a semiconductor photocathode, such as
GaAs are significantly improved if the surface of the
cathode is subjected to a lower peak field
PWT PHOTOINJECTOR
A compact, 10-30 MeV, photoelectron linac using the
Plane Wave Transformer (PWT) design (Figure 1) has
been under development at DULY Research [1] for wide-
ranging applications in research, medicine and industry.
The PWT photoinjector integrates a photocathode directly
into a multicell, standing-wave, π-mode linac, in which
the open cells in a large vacuum tank are strongly
coupled.
Figure 1: Schematic of a PWT photoinjector.
The beam is focused with emittance compensating
solenoids or permanent magnets, and achieves high
current with low emittance, producing extremely high
brightness.
An S-band PWT prototype was fabricated by DULY
Research and installed at UCLA [1]. A higher frequency
(X-band) version of the PWT would produce an even
brighter beam. In a current DULY/SLAC collaboration,
an S-band PWT gun is being developed into a polarized rf
gun capable of high vacuum in the 10
-11
Torr range to
support a long GaAs cathode life and a high quantum
efficiency [2].
With an integrated photocathode, the PWT has a simple
system design which eliminates a separate accelerating
section with the associated focusing optics and additional
rf feeds as required by a conventional 1.6-cell gun in
order to achieve the same energy gain. In addition, the
defocusing kick which the beam receives at the exit of the
last cell occurs at a significantly higher energy than for a
1.6-cell gun, resulting in less emittance growth.
EMITTANCE CALCULATIONS
We have performed improved, detailed beam dynamics
simulations of the PWT photoinjector using the Los
Alamos gun code PARMELA, Version 3. With refined
space charge mesh and careful optimization, the
performance of the PWT gun was evaluated for three
illustrative cases. Optimized results of the simulations are
shown in Table 1, and in Figure 2.
Table1: Parameters of 0.1 nC, 1 nC and 2 nC operations.
Charge per Bunch (nC) 0.1 1.0 2.0
Frequency (MHz) 2856
Energy (MeV) 8 17 17.5
Normalized RMS Emittance
(mm-mrad, no thermal emit.)
0.36 0.56 1.04
Energy Spread (%) 1.4 1.0 1.3
Bunch Length (rms, ps) 2.2 3.6 3.2
Peak Current (A) 13 80 180
Linac Length (cm) 58
Beam Size (rms, mm) 1.2 .85 1.62
Peak Magnetic Field (Gauss) 875 1694 1620
Peak Electric Field (MV/m) 27 55 60
Peak Brightness (10
14
A/m
2
-
rad
2
)
2.0 5.1 3.3
For a given charge per bunch, the initial beam size and
bunch length (both flat top) are varied for each case and
are optimized for space charge minimization. For a given
peak field, the initial injection phase and the solenoidal
field are adjusted to give the lowest normalized transverse
emittance. A low field of 27 MV/m is sufficient for
acceleration of a low emittance beam with a charge of 0.1
nC (see Table 1). The 1-nC case uses a peak field of 55
MV/m, and represents a typical FEL injector (Figure 2).
The 2-nC case uses a peak field of 60 MV/m, and
represents an injector under NLC-like conditions.
Shown in Table 1 are the calculated normalized
emittance of 0.56 mm-mrad (1 nC) and 1.04 mm-mrad (2
nC), which are lower than earlier simulation results [1].
Thermal emittance is not included. From Figure 2, it is
interesting to note that the energy spread and bunch length
decrease with drift distance. It appears that a negative
___________________________________________
* Work supported by DOE SBIR grant no. DE-FG0398ER82566.
SLAC-PUB-10084
July 2003
Work supported in part by the Department of Energy contract DE-AC03-76SF00515.
Presented at the Particle Accelerator Conference (PAC 03), Portland OR., 5/12/2003 - 5/16/2003

chirp is introduced by the gun, in this case allowing for
longitudinal focusing of the bunch. In all cases, the beam
brightness, defined as twice the peak current divided by
the transverse emittance squared, well exceeds 10
14
A/(m-
rad)². From Table 1 and Figure 2, it is clear that the
integrated PWT design produces excellent beam quality
over a wide range of input parameters.
Figure 2: PWT beam parameters for 1 nC bunch charge.
Figure 2 indicates that the transverse emittance and
beam size go through their respective minima and then
grow as the beam continues on after leaving the PWT.
This is a result of the energy (about 17 MeV) not being
high enough to overcome space charge effects. The beam
must be matched into an additional accelerator. The
standard way to do this with a 1.6-cell gun is to position
the input of the accelerator at a laminar beam waist. The
emittance will then be preserved if the proper accelerating
gradient is chosen. Two solutions have been found with
1.6-cell guns for achieving an emittance minimum after
acceleration. The first is to adjust the gun solenoid so the
waist coincides with the emittance minimum without the
accelerator. Utilizing the fact that the emittance undergoes
a plasma oscillation, the second solution is to adjust the
solenoid so that the waist coincides with the immediately
following emittance maximum. These strategies can also
be applied to the PWT.
A peak field of 55 MV/m in an S-band PWT gun is
considerably lower than an operating peak field of 120-
140 MV/m in a conventional 1.6-cell gun. The lower
peak field at the photocathode is particularly beneficial for
the survivability of an activated GaAs cathode in a
polarized rf gun, and also, as we shall see below, for
mitigation of dark current and electrical breakdown due to
field emission and electron back bombardment.
For comparison with the 1.6-cell gun [3], we also
calculated the tranverse emittance for a PWT gun in
which the length of the first cell is 0.625 of a normal cell
of a half wavelength, instead of a 0.5 cell as in the present
design. Figure 3 shows the normalized transverse
emittance versus magnetic field (varying the injection
phase at a given peak field) for these two cases.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1450 1650 1850 2050 2250 2450 2650 2850 3050
Magnetic Field (Gauss)
Transverse Emittance
(mm-mrad)
0.625 cell 0.5 cell
Figure 3: Emittance vs magnetic field for 1 nC bunch.
The solid diamonds are for a 0.625 first cell at a peak field
of 60, 80, 100, 120, 140 and 160 MV/m, and the open
circles are for a 0.5 first cell at 45, 55, 73, 91, 109, 127
and 145 MV/m.
The simulations show that an S-band, π-mode,
integrated PWT photoinjector with the first cavity of 0.5
cell is better than the first cavity of 0.625 cell. They both
can achieve very low emittance. But performance is
different for these two cases. For the 0.5 cell case, the
normalized transverse emittance is 1.08 mm-mrad at 45
MV/m (Bz = 1.5 kG). It decreases to 0.56 mm-mrad as
the peak field increases to 55 MV/m (Bz = 1.7 kG). From
73 to 145 MV/m, the emittance stays very close to 0.4
mm-mrad. For the 0.625 cell case, the emittance
decreases as the field gradient increases, but always
remains higher than the 0.5 cell case. The normalized
transverse emittance is 1.6 mm-mrad at 60 MV/m (Bz =
1.8 kG) and is 0.63 mm-mrad at 160 MV/m (Bz = 3.0
kG). Unlike the BNL/SLAC/UCLA 1.6-cell gun, the
PWT electron beam is only minimally affected by the exit
rf kick.
ELECTRON BACKBOMBARDMENT
SIMULATIONS
Electron Backstreaming from the PWT Iris to
the Photocathode
A GaAs photocathode is prone to dark current emission
when bombarded by electrons or ions. Since electrons
have a smaller mass and are hence more mobile, field-
emitted electrons in the rf cavity can potentially reach the
cathode, damaging the activated layer. We have

performed PARMELA simulations to assess the effects of
electron backstreaming from the first PWT iris to the
photocathode. A thin ring electron bunch, initially with
essentially zero velocity, is placed at a given location on
the surface of the iris. The model includes a drift tube at
the center of the endplate. The aperture of the drift tube is
chosen to be that of the photocathode and its support plug
(0.5 inch). Field emitted electrons move under the rf field
and external magnetic field of the PWT. The number of
electrons which can reach the photocathode depends on
the initial location of the ring bunch, the rf phase and the
peak rf field. At certain locations the rf field on the PWT
iris surface may be as high as 1.17 times the maximum
peak field on axis. The maximum number of electrons
that can reach the cathode occurs at such locations when
the rf field magnitude exceeds a threshold value and when
the initial rf phase at the iris at the time the electrons are
emitted is near zero. Figure 4 shows the dependence of
the field threshold on the iris location at which electrons
are emitted. Figure 5b is an example of snapshots of
electrons emitted from the surface of the first iris and
backstream toward the cathode plane under the rf and
magnetic fields in the PWT. No electrons would reach the
cathode below the field threshold. The beneficial effect
of operating the PWT at a lower peak field (55 MV/m)
than conventional rf guns (>120 MV/m) is evident from
Figure 4.
Figure 4: Threshold peak field for electrons emitted from
the first PWT iris reaching the cathode surface. The insert
shows the iris geometry.
In the PWT gun, backstreaming electrons emitted at an
initial rf phase of 90° from the first iris never reach the
cathode at a peak field (on axis) of 55 MV/m; and at an
initial phase of 0°, only a small fraction of the electrons
would reach the cathode. By contrast if the gun were to
operate at a peak field (120MV/m) comparable to that in a
conventional, BNL-type, 1.6-cell gun, then a large
number of electrons emitted from the iris surface would
hit the cathode. Since field emission is likely to be most
abundant when the surface field is at its maximum (i.e.
90° rf phase); by operating the PWT at a peak field much
lower than conventional rf guns, it is expected that dark
current emission from the GaAs and its support structure
would be considerably mitigated.
Back Bombardment of Electrons Emitted from
the Cathode Holder
Another PARMELA simulation was performed to
evaluate the possible back bombardment of electrons
which were first emitted near the PWT end plate and
cathode support. The model puts a ring of electrons
initially around the cathode and tracks their trajectories
under the rf field and the solenoidal field. At a
sufficiently high peak electric field around 15 MV/m or
higher, all electrons emitted near the cathode would either
go through the first iris or hit the disk (see Figure 5a). At
very low peak fields electrons would oscillate near the
cathode surface. The low field resonance-like condition
could send electrons back to the GaAs cathode surface
and contaminate it. The lower the peak field, the longer
the “trapped” electrons would stay near the cathode
surface. Therefore in order to prevent electrons emitted
from the cathode or the cathode holder from hitting the
photocathode in an rf gun, the operating voltage should
not be too low. The operating peak field chosen for the
proposed polarized PWT gun appears to be optimum in
that it is low enough to discourage back bombardment of
electrons from the iris, and yet high enough to avoid back
bombardment of electrons from the cathode holder.
During rf conditioning, however, it is important that the
GaAs cathode is not exposed to low fields. For that
reason, we will be using a sacrificial cathode during rf
conditioning and replacing it with a freshly activated
cathode via the load lock.
a) b)
C
A
T
H
O
D
E
C
A
T
H
O
D
E
F
I
R
S
T
I
R
I
S
Figure 5: Examples of snapshots (projected onto an x-y
plane) of backstreaming electrons emitted (a) from the
cathode holder, and (b) from the first iris.
REFERENCES
[1] D. Yu, Proc. 2nd ICFA Adv. Accelerator Workshop
on the Physics of High Brightness Beams, UCLA,
November 9-12, 1999, p.585.
[2] D. Yu et al., WPAB043 in these Proceedings.
[3] M. Ferrario et al., Proc. 2nd ICFA Adv. Accel.
Workshop on the Physics of High Brightness Beams,
Nov. 1999, p.534.
Citations
More filters
20 Jan 2016
TL;DR: In this paper, analisi di spettroscopia di fotoemissione a raggi x (XPS) was used to correlate le proprieta' foto-emissive osservate alla composizione chimica dei fotocatodi and per studiare i processi di deterioramento del catodo.
Abstract: Acceleratori di particelle di nuova generazione, come CLIC (Compact LInear Collider), necessitano di fasci di elettroni ad elevata brillanza che potrebbero essere generati attraverso l'uso di un fotoiniettore (sorgente di fotoelettroni attivata da un fascio laser). Fotocatodi di Cs2Te illuminati con fasci laser nell'ultravioletto (UV) sono attualmente utilizzati in diversi fotoiniettori, ma le specifiche tecniche delle sorgenti di elettroni per acceleratori innovativi come CLIC sono significativamente piu' difficili da realizzare. In particolare, il requisito piu' stringente per il drive beam di CLIC e' quello di riuscire ad ottenere allo stesso tempo impulsi di elettroni ad elevata carica (8.4 nC) ed alta frequenza di ripetizione (500 MHz), treni di impulsi di lunga durata (140 us) e fotocatodi con una vita utile ragionevolmente estesa. In particolare l'energia del treno di impulsi laser nel UV, per una tale estesione temporale, e' al momento limitata a causa del deterioramento del fascio laser durante il processo di generazione di quarta armonica. L'utilizzo della seconda armonica (fascio laser nel verde), a condizione che questa sia accoppiata ad un fotocatodo con una soglia di fotoemissione bassa, rappresenterebbe una soluzione a questa limitazione. L'antimoniuro di Cesio (Cs3Sb), essendo un materiale con buone proprieta' fotoemissive nelle lunghezze d'onda del visibile, risulta essere un materiale di interesse in questo ambito. L'obiettivo principale di questa tesi di dottorato e' quello di fornire uno studio sperimentale approfondito dell'uso di tale materiale per applicazioni in fotoiniettori. Le attivita' sperimentali volte a studiare la possibilita' di produrre catodi Cs3Sb di elevata qualita', comprendono la produzione, attraverso tecniche di deposizione di film sottili, e la caratterizzazione delle proprieta' fotoemissive in una linea di accelerazione DC. Le prestazioni dei fotocatodi di Cs-Sb sono state studiate all'interno di un fotoiniettore a radiofrequenza, permettendo cosi' di caratterizzarne le funzionalita' piu' rilevanti per queste applicazioni. Analisi di spettroscopia di fotoemissione a raggi x (XPS) sono state svolte per correlare le proprieta' fotoemissive osservate alla composizione chimica dei fotocatodi e per studiare i processi di deterioramento del catodo, a cui esso e' soggetto durante il funzionamento del fotoiniettore. Inoltre, e' stato condotto un confronto con le prestazioni raggiunte da catodi di Cs2Te. Questo lavoro di tesi fornisce una dettagliata caratterizzazione delle prestazioni di fotocatodi in Cs-Sb e rappresenta percio' un importante contributo allo sviluppo di fotocatodi per applicazioni in fotoiniettori a radiofrequenza ed elevata carica.

9 citations


Cites background from "PARMELA simulations of a PWT photoi..."

  • ...In addition, field emitted electrons in the RF cavity can potentially hit the cathode (97, 98, 99)....

    [...]

Proceedings ArticleDOI
16 May 2005
TL;DR: In this article, the projected normalized transverse rms emittance of a ∼ 10MeV polarized beam is an order of magnitude lower than that calculated for a TESLA pre-accelerator utilizing dc gun and bunch compression.
Abstract: Plane-wave-transformer (PWT) photoinjectors being developed by DULY Research Inc. are capable of operation in ultra high vacuum and moderate field gradient. Expected performance of a proposed L-band polarized electron PWT injector for the International Linear Collider (ILC) is evaluated in this paper. The projected normalized transverse rms emittance of a ∼ 10MeV polarized beam is an order of magnitude lower than that calculated for a TESLA pre-accelerator utilizing dc gun and bunch compression.
Frequently Asked Questions (1)
Q1. What are the contributions mentioned in the paper "Parmela simulations of a pwt photoinjector*" ?

In this paper, a 10+2 ( 1/2 ) -cell PWT photoinjector was used to achieve an emittance of 0.56 mm-mrad for a 1 nC bunch charge at a peak field of 55 MV/m.