LEAP
Energy Gain
Gain From
From Laser
Laser
LEAP Phase
Phase II,
II, Net
Net Energy
Fields
In Vacuum
Vacuum
Fields In
Christopher
Eric R.
R. Colby,
Colby, Tomas
Tomas Plettner*
Plettner*
Christopher D.
D. Barnes,
Barnes, Eric
Stanford
Accelerator Center,
Center, Menlo
Menlo Park,
Park, CA
CA 94025
94025
Stanford Linear
Linear
Accelerator
*
Stanford
University,
Stanford,
CA
94305
*Stanford University, Stanford, CA 94305
Abstract.
Acceleration Program
Program (LEAP)
(LEAP) seeks
seeks to
to modulate
modulate the
the
Abstract. The
The current
current Laser
Laser Electron
Electron Acceleration
energy
of the
the electrons
electrons with
with aa copropagating
copropagating pair
pair of
of crossed
crossed
energy of
of an
an electron
electron bunch
bunch by
by interaction
interaction of
laser
an optical
optical injector
injector design
design for
for aa LEAP
LEAP cell
cellso
sothat
thatititcan
canbe
be
laser beams
beams at
at 800
800 nm.
nm. We
We present
present an
used
an electron
electron bunch.
bunch. Unique
Unique features
features of
of the
the design
designare
arediscussed
discussed
used to
to give
give net
net energy
energy gain
gain to
to an
which
and which
which will
will also
also provide
provide aa robust
robust signature
signature for
for the
the
which will
will allow
allow this
this net
net energy
energy gain
gain and
LEAP
LEAP interaction.
interaction.
INTRODUCTION
INTRODUCTION
Modern
of magnitude
magnitude greater
greater than
than those
those
Modern Terawatt
Terawatt lasers
lasers have
have fluences
fluences 13
13 orders
orders of
produced
by
SLAC
klystrons
and
other
radiowave
sources.
This
indicates
great
produced by SLAC klystrons and other radiowave sources. This indicates great
promise
for
the
production
of
intense
electric
fields
and
for
high
gradient
acceleration.
promise for the production of intense electric fields and for high gradient acceleration.
Many
these large
large fields
fields directly
directly in
in aa dielectric
dielectric structure
structure have
havebeen
been
Many schemes
schemes for
for using
using these
proposed
[1,2].
STELLA
and
STELLA-II
at
Brookhaven
have
also
demonstrated
proposed [1,2]. STELLA and STELLA-II at Brookhaven have also demonstrated
acceleration
Free Electron
Electron Lasers
Lasers [3].
[3]. The
The LEAP
LEAP program,
program,
acceleration of
of electrons
electrons using
using Inverse
Inverse Free
aa collaboration
between
Stanford
University
and
SLAC,
seeks
to
accelerate
electrons
collaboration between Stanford University and SLAC, seeks to accelerate electrons
with
the
electric
field
of
intense
laser
light
in
vacuum
by
crossing
two
lasers
with
with the electric field of intense laser light in vacuum by crossing two lasers with
opposite
component to
to the
the field
field propagating
propagating
opposite phase
phase such
such that
that there
there is
is aa longitudinal
longitudinal component
with
with the
the electrons
electrons [4].
[4].
aceeleratQr
cell
11 cm
cm
FIGURE 1. The LEAP Cell as used
FIGURE
used in recent
recent experimental
experimental runs.
runs.
CP647, Advanced Accelerator Concepts: Tenth Workshop, edited by C. E. Clayton and P. Muggli
© 2002 American Institute of Physics 0-7354-0102-0/02/$19.00
294
In
energy of
of electrons
electrons in
in aa bunch
bunch
In Phase
Phase II of
of LEAP,
LEAP, we
we seek
seek to
to modulate
modulate the
the energy
produced
involves aa single
single interaction
interaction
produced by
by aa conventional
conventional rf
rf accelerator
accelerator [5,6].
[5,6]. This
This involves
between
all phases
phases of
of
between the
the laser
laser beams
beams and
and the
the electrons.
electrons. Because
Because the
the electrons
electrons cover
cover all
the
the laser
laser field,
field, some
some particles
particles will
will be
be accelerated
accelerated and
and others
others decelerated.
decelerated.
At
University, the
the laser
laser produces
produces
At the
the SCA-FEL
SCA-FEL center
center located
located at
at Stanford
Stanford University,
regeneratively
picoseconds. The
The peak
peak
regeneratively amplified
amplified 800
800 nm
nm radiation
radiation with
with 11 mJ
mJ in
in two
two picoseconds.
electric
field
is
1
GV/m,
and
we
have
a
crossing
angle
between
the
two
laser
beams
of
electric field is 1 GV/m, and we have a crossing angle between the two laser beams of
32
mrad.
Because
the
laser
phase
velocity
is
greater
than
c,
the
30
MeV
electrons
slip
32 mrad. Because the laser phase velocity is greater than c, the 30 MeV electrons slip
by
in aa LEAP
LEAP cell.
cell.
by π71 radians
radians in
in aa distance
distance of
of 1.5
1.5 mm,
mm, the
the length
length of
of the
the interaction
interaction in
Energy
Energy modulation
modulation of
of 20
20 keV
keV is
is thus
thus achieved
achieved with
with one
one cell.
cell.
For
we will
will bunch
bunch
For Phase
Phase II,
II, we
we will
will move
move to
to the
the NLCTA
NLCTA facility
facility at
at SLAC.
SLAC. There
There we
the
electrons
at
our
optical
frequency
in
order
to
give
net
energy
to
the
electrons.
This
the electrons at our optical frequency in order to give net energy to the electrons. This
article
presents
the
results
of
detailed
simulations
showing
how
to
create
and
preserve
article presents the results of detailed simulations showing how to create and preserve
femtosecond
to accelerate.
accelerate.
femtosecond structure
structure on
on our
our electron
electron beam
beam for
for long
long enough
enough to
Demonstration
of
an
optical
injector
and
net
energy
gain
is
our
goal.
Demonstration of an optical injector and net energy gain is our goal.
TABLE
TABLE 1.
1. HEPL
HEPL and
and NLCTA
NLCTA Electron
Electron Beam
Beam Parameters.
Parameters.
Parameter
HEPL
Parameter
HEPL Value
Value
Electron
30
Electron Energy
Energy
30 MeV
MeV
Energy
15-20
keV
Energy Spread
Spread
15-20keV
Beam
1I ppC
Beam Charge
Charge
C
Normalized
88 π7i mm⋅mrad
Normalized Emittances
Emittances
mm- mrad
Laser
Laser Power
Power (Wiggler)
(Wiggler)
Laser
20
Laser Power
Power (LEAP
(LEAP Cell)
Cell)
20 MW
MW
NLCTA Value
Value
NLCTA
60 MeV
MeV
60
20 keV
20keV
50 pC
pC
50
mm⋅mrad
22 TCπ mmmrad
40 MW
MW
40
50 MW
MW
50
PROPOSED
PROPOSED EXPERIMENTAL
EXPERIMENTAL SETUP
SETUP
Below
proposed experiment,
showing the
the three
three
Below is
is aa schematic
schematic of
of the
the layout
layout for
for our
our proposed
experiment, showing
main
modulation, second
second is
is aa chicane
chicane for
for
main components.
components. First
First is
is an
an IFEL
IFEL for
for energy
energy modulation,
optical
bunch
formation,
and
last
is
the
LEAP
cell
where
we
accelerate:
optical bunch formation, and last is the LEAP cell where we accelerate:
50 mrad
IFEL
Compressor Chicane
LEAP Cell
FIGURE
wiggler period
period is
is 1.8
normalized strength
is 0.455.
0.455.
FIGURE 2.
2. Schematic
Schematic of
of Phase
Phase II.
II. The
The wiggler
1.8 cm
cm and
and normalized
strength is
The
The Compressor
Compressor Chicane
Chicane deflection
deflection angle
angle is
is 50
50 mrad.
mrad.
295
The
Laser (IFEL)
(IFEL) modulates
modulates the
the energy
energy of
of the
the incoming
incoming
The Inverse
Inverse Free
Free Electron
Electron Laser
electron
bunch
at
the
wavelength
of
the
drive
laser,
here
800
nm.
After
the
IFEL
electron bunch at the wavelength of the drive laser, here 800 nm. After the IFEL
modulates
the
energy
of
the
electrons,
travel
through
the
Compressor
Chicane
modulates the energy of the electrons, travel through the Compressor Chicane
converts
into phase
phase bunching.
bunching.
converts the
the energy
energy modulation
modulation into
As
our
original
electron
bunches
from
the X-band
X-band accelerator
accelerator are
are 11 picosecond
picosecond
As our original electron bunches from the
long,
there
are
expected
to
be
approximately
300
optical
bunches
produced.
The
long, there are expected to be approximately 300 optical bunches produced. The
second
laser
pulse
goes
to
the
LEAP
cell
and
accelerates
this
bunch
train.
second laser pulse goes to the LEAP cell and accelerates this bunch train.
To
our design,
design, we
we used
used aa start-to-finish
start-to-finish simulation
simulation of
of the
the
To gain
gain insight
insight and
and refine
refine our
NLCTA
required for
for Phase
Phase II.
II.
NLCTA facility,
facility, including proposed additions to itit required
SIMULATION
We
gun through
through the
the first
first accelerator
accelerator section,
section, as
as itit
We used
used PARMELA [7] from the gun
provides
highly relativistic.
relativistic. For
For increased
increased speed,
speed,
provides accuracy when the electrons are not highly
ELEGANT
of the
the linac
linac to
to the
the LEAP
LEAP interaction
interaction area.
area.
ELEGANT [8] was used to model the rest of
This
LEAP cell
cell region
region to
to the
the end.
end. To
Tomodel
model
This article
article focuses
focuses on the simulations of the LEAP
the
with aa few
few modifications to
to allow
allow itit to
to share
share
the IFEL,
IFEL, GENESIS
GENESIS 1.3 [9] was used, with
data
tracking FEL
FEL code,
code, where
where transverse
transverse
data with
with ELEGANT. This code is a particle tracking
variables
and the longitudinal
longitudinal motions
motions are
are given
givenby
by
variables are pushed using transfer matrices and
fourth
standard FEL
FEL equations.
equations.
fourth order
order Runge-Kutta
Runge-Kutta solutions to the standard
After
through the
the small
small chicane
chicane was
was
After the
the IFEL, the propagation to the LEAP cell through
again modeled
modeled with ELEGANT. For the LEAP interaction
again
interaction and
and calculation
calculation of
of
spectrometer images, a MATLAB code was used.
spectrometer
Figure 33 shows
shows longitudinal
longitudinal phase
phase spaces
Figure
spaces as
as the
the electron
electron bunch
bunch propagates
propagates through
through
our experiment
experiment from
from the
the end
end of
of the
the NLCTA
NLCTA main
our
main accelerator.
accelerator. The
The first
first plot
plot shows
showsthe
the
electron bunch
bunch produced
produced by
by the
the NLCTA,
NLCTA, the
electron
the second
second shows
shows one
one cycle
cycle of
of energy
energy
modulation after
after the
the IFEL.
IFEL. The
The third
modulation
third phase
phase space
space plot
plot shows
shows one
one optical
optical bunch
bunch after
after
the chicane,
chicane, and
and the
the last
last phase
phase space
space is
is what
what is
the
is expected
expected after
after the
the LEAP
LEAP interaction.
interaction.
Gompressor Chicane
LEAP Cell
FIGURE
3.
Phase
space
progression
through
LEAP
Phase
II
FIGURE 3. Phase space progression through LEAP Phase IIexperiment
experiment
296
Design Considerations
Because bunching requires that electrons only change phase by as much as one
quarter of a wavelength longitudinally, the bunch structure must be preserved to a
tolerance of better than 200 nanometers. Path length differences represent a
significant condition, in that particles taking extreme trajectories through a focusing
system will travel further than the reference particle by more than 200 nanometers
unless care is taken.
The desire to minimize path length differences requires that the entire experimental
setup be placed near the final focus used to make the electron beam narrow enough to
pass through the -10 |i slit of the LEAP cell.
Thus the IFEL must be as short as practical, while giving sufficient energy
modulation to bunch. As the intrinsic energy spread of the beam is of order 20 keV,
one can achieve bunching with energy modulations in the IFEL that are not much
greater than this value. Our 40 MW of laser power in a 3 period wiggler gives about
70 keV of energy modulation peak-to-peak. Although larger energy modulation
allows stronger bunching, it also makes detection of the LEAP interaction's
accelerating effect hard to see, driving the design to smaller modulation.
For velocity bunching, these small energy spreads are not sufficient to bunch the
electrons in less than 2 meters of drift distance, because the particles are highly
relativistic. Fortunately, use of a chicane allows bunching to occur in a much shorter
distance. With a deflection of 50 mrad, the electrons achieve maximum bunching in a
chicane 12 cm long. Thus, even allowing several centimeters of open space for beam
diagnostics, one can bunch the electrons optically in a distance of only 20 cm.
For comparison, STELLA's second stage gives large energy gain, so they can see
signal even after inducing large energy spreads to bunch the beam with only a drift
space. Also, the bunching at the CC>2 wavelength has a much larger longitudinal scale,
so the transverse path effects of refocusing the beam are much less of a concern.
ANTICIPATED RESULTS
As we cannot observe longitudinal phase spaces directly, we simulate what can be
observed in the real experiment, namely spectra of the electron bunches. Because the
optical frequency bunching is not perfect, significant numbers of electrons will be
decelerated even as the majority gain energy.
Below is a simulated scan where the relative phase between the IFEL and LEAP
cell is varied over 5 periods. Averaged spectra are taken at 90 degree intervals
showing the different effects on the electron beam as the relative phase changes.
297
FIGURE 4. Simulated phase scan with jitter covering five cycles of relative phase between IFEL and
LEAP cell. On the right are averaged spectra taken each 90 degrees apart, showing the distinctive
behavior which provides the robust signature we seek.
These electron bunch spectra change markedly as we scan the relative laser phase
future
between the IFEL and LEAP cell. This behavior, although not ideal for a future
practical accelerator, gives a specific signature, allowing ready detection of even a
weak LEAP interaction. For a practical accelerator, a superior bunching system would
involve larger energy modulations. The current goal is to demonstrate optical
cell's effect unambiguously.
bunching at 800 nm, and to detect the LEAP cell’s
Practical Issues
Based upon experience from Phase I, spatial and temporal overlap of the laser and
1-2 picoseconds
picoseconds long
long and
and will
will be
be
electron pulses is a significant concern. Both are 1-2
100 microns in RMS transversely.
less than 100
Care must also be taken to control the phase of the laser beams driving the IFEL
and LEAP cells. We will use feedback with delay lines and piezo actuators to ensure
repeatable relative laser phase.
directly affects
affects
Intrinsic energy spread of the electron beam is a consideration, as itit directly
the bunching. The photoinjector proposed for the NLCTA does not have sufficient
stability, so we use a chicane and energy collimator to constrain the energy spread and
jitter. Studies indicate that as few as 5% of our linac pulses will make it to the LEAP
cell, so data taking will be slow.
298
Fortunately, there are several parameters which have relaxed tolerances. The
energy of the electron beam, and equivalently the IFEL wiggler's strength, can vary by
several percent without significantly degrading the electron bunching. The chicane
similarly can have strength errors of up to 5%. Since the IFEL and chicane will be
made with permanent magnets, this insensitivity is helpful.
Such things as electron focusing and position errors will affect the brightness of our
images or the data taking rate, but do not affect the signature shapes for which we are
searching.
CONCLUSION
LEAP Phase II presents a number of technical challenges, but is very promising. It
will demonstrate optical injection for future laser based accelerators, and will show
energy gain to electrons in a vacuum using laser radiation.
ACKNOWLEDGMENTS
The authors wish to thank Dennis Palmer for help modeling the ORION gun and Xband accelerator fields which will be used at the NLCTA. We also thank Wayne
Kimura and Jim Spencer for fruitful discussions of permanent magnet designs for the
IFEL and chicane. This work is supported by the U.S. Department of Energy, under
contracts DE-AC03-76SF00515 and DE-FG03-97ER41043-II.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
Pantell, R.H., and Piestrup, M.A., Appl Phys. Letters 32, 781-783 (1978).
Edighoffer, J.A., and Pantell, R.H., J. Applied Physics 50, 6120-6122 (1979).
Kimura, W.D., et, al, Phys. Rev. Letters, Vol. 86 No. 18, 4041-4143 (2001).
Esarey, E., Sprangle, P., and Krall, J., Physical Review E, Vol. 52 No. 5, 5443-5453 (1995).
Huang, Y.C., et. al, Nucl. Instr. Methods A 407, 316-321 (1998).
Plettner, T., et. al., PAC 2001 Conference Proceedings, MOPC003, pp. 108-110 (2001).
Crandall, K, and Young, L., "PARMELA: Phase and Radial Motion in Electron Linear
Accelerators." Private communication.
8. Borland, M.D., "ELEGANT: ELEctron Generation ANd Tracking." Private communication.
9. Reiche, S., Nucl. Instrum. Methods A 429, 243-248 (1999).
299