PHYSICAL REVIEW SPECIAL TOPICS - ACCELERATORS AND BEAMS 12, 063501 (2009)
Radial focusing and energy compression of a laser-produced proton beam
by a synchronous rf field
Masahiro Ikegami,1 Shu Nakamura,2,* Yoshihisa Iwashita,1 Toshiyuki Shirai,1 Hikaru Souda,1 Yujiro Tajima,1
Mikio Tanabe,1 Hiromu Tongu,1 Hiroyuki Itoh,1 Hiroki Shintaku,1 Atsushi Yamazaki,1 Hiroyuki Daido,2 Akifumi Yogo,2
Satoshi Orimo,2 Michiaki Mori,2 Mamiko Nishiuchi,2 Koichi Ogura,2 Akito Sagisaka,2 Alexander S. Pirozhkov,2
Hiromitsu Kiriyama,2 Shyuhei Kanazawa,2 Shuji Kondo,2 Yoichi Yamamoto,2 Takuya Shimomura,2 Manabu Tanoue,2
Yoshimoto Nakai,2 Atsushi Akutsu,2 Sergei V. Bulanov,2 Toyoaki Kimura,2 Yuji Oishi,3 Koshichi Nemoto,3
Toshiki Tajima,2 and Akira Noda1
1
2
Institute for Chemical Research, Kyoto University, Gokasho, Uji, Kyoto, 611-0011, Japan
Advanced Photon Research Center and Photo Medical Research Center, Japan Atomic Energy Agency, 8-1-7 Umemi-dai,
Kizu, Kyoto 619-0215, Japan
3
Central Research Institute of Electric Power Industry, 2-6-1 Nagasaka, Yokosuka, Kanagawa 240-0196, Japan
(Received 15 June 2007; revised manuscript received 15 October 2008; published 30 June 2009)
The dynamics of a MeV laser-produced proton beam affected by a radio frequency (rf) electric field has
been studied. The proton beam was emitted normal to the rear surface of a thin polyimide target irradiated
with an ultrashort pulsed laser with a power density of 4 1018 W=cm2 . The energy spread was
compressed to less than 11% at the full width at half maximum (FWHM) by an rf field. Focusing and
defocusing effects of the transverse direction were also observed. These effects were analyzed and
reproduced by Monte Carlo simulations. The simulation results show that the transversely focused protons
had a broad continuous spectrum, while the peaks in the proton spectrum were defocused. Based on this
new information, we propose that elimination of the continuous energy component of laser-produced
protons is possible by utilizing a focal length difference between the continuous spectral protons and the
protons included in the spectral peak.
DOI: 10.1103/PhysRevSTAB.12.063501
PACS numbers: 41.85. p, 41.75.Jv, 52.38.Kd
I. INTRODUCTION
An increase in the acceleration gradient of chargedparticle beams has long been required for fundamental
researches, such as particle and nuclear physics, in order
to obtain high-energy particles within a limited length of
the accelerators. From the point of view of such applications as hadron therapy and material treatment, the downsizing of ion accelerators by using a high acceleration
gradient is also important. The emission of energetic ions
from a plasma created by a high-intensity short-pulse laser
has been reported [1,2]. Recent experiments have shown
that a multi-MeV ion beam was stably generated from a
laser-irradiated thin foil [3–12]. Laser-plasma ion acceleration is one of the candidates of a compact accelerator
[13]. For actual use, monochromatization of the broad
energy spectrum should be developed. Recently, several
approaches to create quasimonoenergetic ion beams have
been tested in several laboratories [14–17] which, however,
seems to be difficult for such an operation with a moderate
repetition rate (up to 10 Hz, for example) with good
reproducibility. In order to produce monoenergetic ion
beams with high repetition-rate operation, we have pro*Present address: High Energy Accelerator Research
Organization (KEK), 1-1 Oho, Tsukuba, Ibaraki 305-0801,
Japan.
1098-4402=09=12(6)=063501(6)
posed the combination of laser-plasma acceleration with a
phase-rotation method by a synchronous rf electric field
[18–21].
Phase rotation is a method in which the configuration of
the particle distribution in longitudinal phase space is
rotated by an rf electric field so as to realize a small energy
spread. Laser-produced ions are emitted within a very short
time duration ( 1 ps) from the plasma created by an
ultrashort laser pulse with a duration shorter than 100 fs.
Since the ions have a variety of energy, after the passage of
a certain drift distance ( 1 m in our experiment), ions
with higher energy arrive at the gap of the rf cavity earlier,
and those with a lower energy arrive later. If the phase of
the rf field is adjusted so that the earlier arrived ions are
decelerated and the ions arriving later are accelerated, we
can collect the ions to a certain energy region.
In the present work, the dynamics of laser-produced
proton beams (with a broad energy spectrum) was studied
by applying an rf electric field in both experiments and
(computer situations) calculations.
II. EXPERIMENTAL SETUP
The present experiment was carried out at the Kansai
Photon Science Institute, Japan Atomic Energy Agency,
using a Ti:sapphire laser system, called J-KAREN [22],
with the setup shown in Fig. 1. In this experiment, the peak
063501-1
Ó 2009 The American Physical Society
Phys. Rev. ST Accel. Beams 12, 063501 (2009)
MASAHIRO IKEGAMI et al.
Off-axis parabolic mirror
45 deg
1.125 m
TOF detector
0.12 m
1.138 m
Tape target
Laser
pulse
Aperture: Rf-cavity for
phase rotation
5 mm
diameter
Sweeping
magnets
FIG. 1. (Color) Schematic view of the experimental setup of
phase rotation. The protons emitted normal to the rear surface of
the tape target are injected into the rf cavity. The flight time of
protons with and without phase rotation is measured by a TOF
detector. The sweeping magnets are used to remove electrons so
as to suppress the background in the scintillation counter of the
TOF detector.
power and the pulse duration were up to 18 TW and 25 fs,
respectively [23]. The repetition rate of the laser was
10 Hz. The laser pulse was focused by an off-axis parabolic
mirror with a focal length of 325 mm, and the focused
energy was 0.45 J on a 7:5 m-thick polyimide tape target
giving an intensity of 4 1018 W=cm2 on the target. The
laser pulse was incident on the target surface with an angle
of 45 degrees, as shown in Fig. 1. The tape target was rolled
up so as to be irradiated on a fresh surface during continuous irradiation. The protons were emitted normal to the
rear surface of the target. The emission time of the protons
was estimated to be less than 1 ps. The protons which
were emitted from the target had a Gaussian distribution in
the transverse direction [24,25], and their divergence solid
angle to the target normal was 3:8 10 1 sr in 2, where
represents the standard deviation. The protons passed
through an aperture with a 5 mm diameter before injection
to a phase-rotation rf cavity. The solid angle of the aperture
was 9:1 10 5 sr. In this small solid angle, the radial
distribution of the protons was almost uniform. If the
protons produced at the target had the above-mentioned
divergence solid angle (3:8 10 1 sr), about 0.4% of them
could pass through the aperture. At the rf cavity, the faster
protons were decelerated and the slower protons accelerated. In order to develop a time spread of the beam required
for phase rotation, the protons must pass through a certain
distance. Therefore, the rf cavity was set at 1.125 m downstream from the target (Fig. 1). The rf cavity had a hole to
pass through a beam of which the inner diameter was
50 mm. The solid angle of this hole was larger than the
5 mm diameter aperture. Therefore, all of the protons that
passed through the aperture could be accepted.
The rf cavity is composed of a quarter-wave length
resonator. The cavity has two gaps. The gap size and the
voltage amplitude are 20 mm and 80 kV, respectively. The
frequency of the rf is 82 MHz. To measure the rf phase and
the gap voltage, the rf signal is monitored through the
pickup port of the rf cavity. The rf voltage applied to the
cavity is synchronized with the pulsed laser so that the rf
electric field could be applied with a fixed phase relation to
the pulsed laser in order to efficiently compress the energy
spread of the laser-produced protons.
Seed pulses to enter into Ti:sapphire amplifiers are
generated by a mode-locked oscillator having a frequency
of 82 MHz. Among these 82 MHz pulse trains, laser pulses
are selected at a 10 Hz repetition rate, and then further
amplified. When we use an 82 MHz oscillator of the seed
pulses as the master oscillator for the rf, the phase relation
between the time of laser irradiation to the target which is
the time of proton pulse duration and the rf phase will be
well defined. Note that laser pulse width as well as proton
pulse width is almost negligibly short compared with the
period of rf. The source signal of the rf from the master
oscillator is picked up by a PIN photodiode and amplified
by three-stage rf amplifiers up to 30 kW and fed to the rf
cavity. Although only one cycle of the rf electric field is
sufficient for the purpose of an energy compression of a
pulsed beam generated by each laser pulse, rf power is fed
with the duration of up to 1 ms ( 105 periods) into the
cavity to realize stable operation. This rather long duration
of rf power feeding is also necessary to cover the filling
time to build up a sufficient rf voltage (stored energy) in the
cavity before the arrival of protons. The timing jitter between the laser pulse and the rf electric field is negligible.
This rather long pulsed operation of the rf cavity (duty
factor of up to 1%) contributes to the reduction of the
power dissipation and heat generation at the cavity wall
to a tolerable level.
III. ENERGY SPECTRUM
The energy of the protons is measured with a time-offlight (TOF) detector [26,27]. The solid angle of the TOF
detector is 1:7 10 4 steradian (sr) to the tape target in the
setup shown in Fig. 1. This is larger than the solid angle of
the 5 mm diameter aperture. Therefore, all of the protons
that passed through the aperture could be detected in
the present experiment. The time-of-flight signal of
protons measured by the TOF detector is shown in
Figs. 2(a) and 2(b) by solid lines. The vertical axis corresponds to the number of protons that entered into the
aperture of the detector. In Fig. 2(a), the solid gray line is
the TOF spectrum of laser-produced protons averaged over
ten shots in the case without an rf field. The black solid line
is the TOF spectrum averaged over ten shots when the rf
field for phase rotation is applied. It is found that the
distribution of the TOF signals is strongly modulated by
phase rotation. In Fig. 2(b), each solid line represents a
spectrum shot by shot. The structure of the TOF distribution is reproduced precisely for every laser shot. In the
absence of an rf field, the horizontal axis can easily be
converted to the energy of a proton, because proton velocity is constant during the flight from the target to the
detector. The energy spectrum converted from the mea-
063501-2
Proton number
(counts/(shot 100keV 1.7 10-4 sr))
RADIAL FOCUSING AND ENERGY COMPRESSION OF A . . .
Phys. Rev. ST Accel. Beams 12, 063501 (2009)
4
1.4 10
4
1.2 10
(a)
4
Without rf field (experiment)
Phase rotation (experiment)
Phase rotation (simulation)
(b)
1 10
3
8 10
3
6 10
3
4 10
3
2 10
0
0.12
0.14
0.16
0.18
0.2
0.22 0.12
Proton number
(counts/(shot 100keV 1.7 10-4 sr))
Time of flight of protons (µs)
0.16
0.14
0.18
0.2
0.22
Time of flight of protons (µs)
4
5 10
4
Without rf field (experiment)
After phase rotation (simulation)
4 10
(c)
4
3 10
4
2 10
4
1 10
0
0.4
0.6
0.8
1
1.2
1.4
Energy (MeV)
FIG. 2. (Color) (a) TOF signal with and without the phase rotation. The horizontal axis is the arrival time of protons from the foil target
to the TOF detector. Part (b) shows the TOF signal shot by shot. Although the number of protons is different shot by shot, the global
structure of the distribution is reproduced precisely. (c) Energy spectra corresponding to the TOF spectra of (a).
sured TOF signal is shown by the gray solid line in
Fig. 2(c). When the protons are accelerated or decelerated
by the rf cavity, the velocities change during the flight.
Therefore, in the case with phase rotation, the TOF does
not directly give the correct proton energy.
In order to obtain the exact kinetic energy of the phaserotated protons, we should calculate the velocity change at
the rf gap. Those protons having the energy spectrum of the
gray solid line of Fig. 2(c) are assumed to pass through the
rf field of the cavity. The voltage, frequency, and phase of
the cavity are set at the same condition as in the experiment. Then, from the calculation, the energy spectrum after
the phase rotation is evaluated, as shown in Fig. 2(c) by the
black solid line. The energy spread of the spectral peak at
1.05 MeV is 11% at FWHM (including two partial peaks).
In this peak, 104 protons are estimated to be included.
These correspond to 0.07% of protons produced at the
target. Calculation of the flight time to the TOF detector
with use of the simulated proton energy spectrum gives us
a TOF spectrum similar to the experimental one, as shown
by the dotted line in Fig. 2(a). This result shows the
relevance of our calculation. It should be noted that the
TOF method smooths out the fine structure of the original
energy distributions because it cannot take into account the
velocity change at the rf cavity. Therefore, in order to
obtain the exact energy spectrum of the phase-rotated
beam, we should calculate the energy spectrum from the
observed TOF spectrum by the above procedure. The
energy spectrum after phase rotation includes several spec-
tral peaks, as shown in Fig. 2(c). The proton intensity
increases by more than a factor of 2 in each peak after
phase rotation. Each peak can be separated by an analyzing
magnet put downstream of the cavity. Then, we obtain a
quasimonoenergetic proton beam.
In the following, we quantitatively show the effect of the
phase rotation on the energy spectrum. The interval of each
peak in the energy spectrum is decided by the rf frequency
and the distance between a cavity and a target, as follows.
The energy gain of a charged particle that passes through
the rf gap can be written as
E ¼ qVT cosð þ 0 Þ;
(1)
where q, V, and T are the charge, voltage amplitude, and
transit time factor, respectively [28]. The notation is the
phase of the rf when a proton reaches the rf gap which is
given by ¼ 2fL=v, where f, L, and v are the frequency of the rf, the distance from the target to the rf gap,
and the velocity of protons, respectively. The notation 0 is
the initial rf phase. For simplicity, we consider that the rf
initial phase, 0 , is set at 0:5. Then, protons that arrive at
the gap within the phase n < < ðn þ 1Þ gain energy
(accelerated), and those within ðn þ 1Þ < < ðn þ 2Þ
lose energy (decelerated), where n is a positive integer. As
a result, protons are collected to velocities of
vn ¼
fL
;
n
n ¼ 1; 2 . . .
(2)
each of which is the central velocity of the nth peak created
063501-3
Phys. Rev. ST Accel. Beams 12, 063501 (2009)
MASAHIRO IKEGAMI et al.
in the energy spectrum. In the present experiment, the
energy peak at 1.05 MeV corresponds to the 6th peak.
The phases of adjacent peaks should satisfy the relation
nþ1 n ¼ 2fL=vnþ1 2fL=vn ¼ 2, where n
is the rf phase corresponding to the nth peak. The velocity
difference, v ¼ vn vnþ1 , of adjacent peaks indicates
the velocity range collected to the nth spectral peak. The
acceptable velocity range of the nth spectral peak is written
as
vn
v
:
¼
vn þ fL
vn
(3)
Therefore, if the rf frequency f or distance from the
target to the rf gap L is reduced, the velocity range collected to a peak is increased.
IV. TRANSVERSE FOCUSING AND DEFOCUSING
Control of the transverse profile of the protons accelerated in a laser plasma is described in Ref. [16] based on a
destructive method that is not suitable for repetitive operation. Generally, it is known whether charged particles that
pass through an rf gap receive a transverse kick, depending
on the rf phase [28]. In our experiment, we found that the
radial component of the rf electric field clearly gives focus
or defocus effects on the motion of the protons in the
transverse direction. The transverse profile of the proton
beam is detected at 1.736 and 2.365 m downstream of the
tape target (the proton source) by using a nuclear track
detector, CR-39 [29]. Since CR-39 detection has sensitivity
only to heavy particles (protons, ions, and neutral atoms), it
is suitable for proton detection under heavy backgrounds of
laser light, laser-generated hard x rays, and electrons. The
surface of the CR-39 is covered with a thin foil so that only
specific protons higher than a certain threshold energy are
detected. By changing the thickness of the foil carefully,
we can choose the energy range of the detected protons. In
the present experiment, the thickness of the foil is adjusted
so that protons higher than 0.78 MeV also can be detected.
The cross section of the proton beam detected by the CR39 is shown in the upper pictures of Figs. 3(a)–3(c). When
we do not apply phase rotation, the radial distribution of
protons measured at 1.736 m downstream from the tape
target becomes as shown in Fig. 3(a). It is found that the
divergence angle of the proton beam is 5.4 mrad, which is
determined by the diameter of the aperture in Fig. 1. In the
cross section of the beam, the proton distribution is almost
uniform. When the rf field for phase rotation is switched
on, the radial distribution drastically changed as shown in
Fig. 3(b). This is detected at the same position as shown in
Fig. 3(a). Both focusing and defocusing components are
visible. When the phase-rotated beam is observed further
downstream (2.365 m) from the tape target, the inner
component becomes small and dense. Therefore, we could
judge that this component is focusing. On the other hand,
the outer components are defocused. It is a matter of
FIG. 3. Comparison of the transverse profile of the beam. (a)–
(c) (upper pictures) are the profile detected in the experiment.
(d)–(f) (lower pictures) are the profile obtained by Monte Carlo
simulation with the same condition to the experiment. Parts (a),
(b), (d), and (e) are distributions at 1.736 m downstream from the
tape target the proton source. Parts (a) and (d) are the profile
without phase rotation, (b) and (e) are with phase rotation. Parts
(c) and (f) are measured at 2.365 m downstream from the tape
target with phase rotation. In the experiment a mesh was inserted
between the target and the rf cavity to measure the divergence.
From (c), a diameter of the defocusing component is estimated to
be about 40 mm at 2.365 m downstream from the target. The
TOF detector is put at 2.383 m downstream from the target, and
it has an aperture of 42 mm. Therefore, a significant part of the
focused protons was detected by the TOF detector, even when
the rf cavity was switched on in the present experiment.
concern whether the protons that have a specified energy
are selectively focused, or not. In the present measurement
we could not distinguish the energy between the focusing
and defocusing components, while the threshold of the
energy could be identified. Identification of the energy of
the focused and defocused protons gives important information for understanding the dynamics of phase-rotated
beams.
In order to clarify such a problem, we carry out a Monte
Carlo simulation assuming a time-dependent rf field for
phase rotation. Protons are tracked in six-dimensional
phase space with the method of transfer matrix [30]. By
using this simulation, we could identify the energy of
protons that have various radial positions. In order to
evaluate the justification of the simulation, we first compared the transverse projection of protons, under the same
condition as in the experiment. A radially uniform proton
beam having the energy spectrum as shown by the solid
gray line in Fig. 2(c) was assumed to pass through the rf
gap with a divergence angle of 5.4 mrad. As shown in the
lower pictures of Figs. 3(d)–3(f), the simulation results are
qualitatively consistent with the experimental results.
By using this simulation method, we can discuss the
relation between the radial distribution and the energy
spectrum. The transverse projection of protons of the en-
063501-4
RADIAL FOCUSING AND ENERGY COMPRESSION OF A . . .
ergy between 0.85 and 0.99 MeV, which corresponds to the
valley of the energy spectrum, was investigated by the
simulation. As shown in Fig. 4(b), in this energy region,
only the focusing component is included. The radial distribution of protons between an energy of 0.99 and
1.11 MeV, which corresponds to the peak of the energy
spectrum, is shown in Fig. 4(c). The distribution has two
spatial components: one is defocused, and the other seems
to be more focused. For the other energy peaks, similar
results were obtained. This means that the focused protons
Y (mm)
4 10
Phase rotation(PR)
PR+post-processing
(a)
4
3 10
0
-20
4
2 10
4
1 10
0
10
-10
0.4
1
0.8
Energy (MeV)
0.6
1.2
Y (mm)
Proton number
(counts/(shot 100keV 1.7 10 -4 sr))
20 (b)
4
20 (c)
10
0
-10
-20
-20 -10 0 10 20
X (mm)
Energy gain ∆E
∆E
Defocus
Velocity of ions (v)
Transverse effect
kr
(d)
Focus
collected to the
Transverse effect energy peak
(e)
Focusing elements
Phase rotation cavity
beam stopper
foucused beam with continuous spectrum
mono-energetic diverged beam
FIG. 4. (Color) (a) Simulated energy spectra of phase-rotated
beam. The transverse profile is also simulated by the Monte
Carlo method. Parts (b) and (c) show the simulated transverse
profile of phase-rotated proton beam included in the valley of the
energy spectrum (between 0.85 and 0.99 MeV) and that in the
energy peak (between 0.99 and 1.11 MeV). Parts (b) and (c) are
calculated at 2.365 m downstream of the tape target (the proton
source) with the same condition as Fig. 3. Part (d) is a schematic
illustration of the relation between the strength of the effects of
energy compression and transverse focusing. The transverse axis
is written with the velocity. (e) Setup of postprocessing used to
remove protons with a continuous spectrum. If the phase-rotated
beam passes through such a system, the energy spectrum is
expected to change as shown in (a).
Phys. Rev. ST Accel. Beams 12, 063501 (2009)
are distributed over the whole energy spectrum, while the
defocused protons are collected in the energy spectral
peak. Such a property of a phase-rotated beam can be
explained as follows. The focusing strength in the transverse direction (kr ) (this corresponds to the reciprocal of
the focal length) is given by the following formula [30]:
kr ¼
qVT sinð þ 0 Þ
;
mv2 2
(4)
where m, v, and
are the mass, velocity, and Lorentz
factor of the proton, respectively. T is the transit time factor
used in Eq. (1). is the wavelength of the rf. is the phase
of the rf which is identical to that of Eq. (1); ¼ 2fL=v.
The positive sign of kr means a defocusing effect, and the
negative sign of kr means focusing. We consider Eqs. (1)
and (4) as being a function of velocity, v. Then, for the
present experimental parameters, the trigonometric functions (cosine and sine) are dominant for the determination
of the signs and absolute values of E and kr in the
velocity region of the present laser-produced protons
( 1:5 MeV). In Fig. 4(d), we illustrate the transverse
[kr ; Eq. (4)] and longitudinal [E; Eq. (3)] effects as
functions of the velocity of a proton over a period of the
rf. In this figure, the transverse effect of the phase rotation
can be divided into focusing and defocusing phases. Since
the laser-produced protons arrive in the phase-rotation
cavity with a variety of velocities, the protons are injected
into all of the phases. We consider the protons that have
velocities included in the right half of Fig. 4(d) (this
velocity region corresponds to a transverse defocusing
phase). Since these protons also have an energy spread,
their distribution can be illustrated as the parallelogram
shown in the v E space of Fig. 4(d). In this region the
faster protons lose energy and the slower ones gain energy.
As a result, the protons at the transversely defocusing
phase are compressed in energy and a spectral peak is
created. On the other hand, the protons that have velocities
corresponding to the left half of Fig. 4(d) are focused in the
transverse direction. In this region the faster protons gain
energy and the slower protons lose energy. This means that
the total proton energy spread increases. The net result is
that after the phase rotation the protons belonging to the
continuous part of the energy spectrum have been radially
focused, while those in the spectral peaks defocused. The
results of Figs. 3(b), 3(c), 3(e), and 3(f) and of Figs. 4(b)
and 4(c) can be explained by this interpretation.
If we utilize such a property of phase-rotated protons
and ions, we should cut the tail of each quasimonoenergetic peak in the energy spectrum. After phase rotation,
focused protons (ions) always have a continuous energy
spectrum. Fortunately, these protons (ions) are focused
within a small radius near the focal point, as shown in
Figs. 4(b) and 4(c). Therefore, if the focused components
are removed by a beam stopper at the focal point, those
protons (ions) with a continuous energy spectrum can be
063501-5
Phys. Rev. ST Accel. Beams 12, 063501 (2009)
MASAHIRO IKEGAMI et al.
removed. Utilizing this feature by the setup shown in
Fig. 4(e), we can select only the energy peaks. In order
to evaluate the efficiency of this method, a Monte Carlo
simulation assuming the same conditions as the previous
simulation was performed. A beam stopper of 5 mm in
radius was placed at 1.5 m downstream of the cavity to
remove the focused component. The distance of 1.5 m
corresponds to the focal length of the focusing component.
According to the simulation, the intensity of the spectral
peak is slightly reduced after this postprocessing. On the
other hand, the protons that are included in the valley of the
energy spectrum are completely removed [Fig. 4(a)].
V. CONCLUSIONS
We successfully created quasimonoenergetic peaks in
the energy spectrum of MeV proton beams by combining
laser-plasma acceleration and energy modulation by using
a phase-rotation cavity. We found that 0.07% of protons
produced at the target were collected to the highest quasimonoenergetic peak. This method has a capability to create
monoenergetic beams with a relatively high repetition rate.
We also showed that the phase-rotation changes the radial
distribution of laser-produced protons, depending on their
position in the energy spectrum. The protons included in
the energy peak were defocused, and the almost focused
protons belonged to a continuous energy spectrum.
Utilizing the difference in the focal length between the
quasimonoenergetic protons and those having a continuous
energy spectrum, we could remove the continuous energy
component. A series of postprocessing for the laserproduced protons shown in this paper can open up the
feasibility of improving quality and reproducibility of the
laser-produced proton and ion beams for practical applications [31].
ACKNOWLEDGMENTS
This work was partly supported by ‘‘Advanced Compact
Accelerator Development Project’’ which belongs to the
Ministry of Education, Culture, Sports, Science and
Technology (MEXT) of Japanese government, and it was
also partly supported by the Grant-in-Aid for the Global
COE Program ‘‘The Next Generation of Physics, Spun
from Universality and Emergence’’ of the MEXT of
Japan. This work was also partly supported by the
Special Coordination Fund (SCF) for Promoting Science
and Technology commissioned by the MEXT of Japan.
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
063501-6
A. P. Fews et al., Phys. Rev. Lett. 73, 1801 (1994).
F. N. Beg et al., Phys. Plasmas 4, 447 (1997).
E. L. Clark et al., Phys. Rev. Lett. 84, 670 (2000).
S. P. Hatchett et al., Phys. Plasmas 7, 2076 (2000).
A. Maksimchuk, Phys. Rev. Lett. 84, 4108 (2000).
Y. Murakami et al., Phys. Plasmas 8, 4138 (2001).
K. Krushelnick et al., IEEE Trans. Plasma Sci. 28, 1184
(2000).
K. Nemoto et al., Appl. Phys. Lett. 78, 595 (2001).
H. Schwoerer, P. Gibbon, S. Düsterer, R. Behrens, C.
Ziener, C. Reich, and R. Sauerbrey, Phys. Rev. Lett. 86,
2317 (2001).
K. Matsukado et al., Phys. Rev. Lett. 91, 215001 (2003).
A. Fukumi et al., Phys. Plasmas 12, 100701 (2005).
M. Nishiuchi et al., Phys. Plasmas 15, 053104 (2008).
D. Habs, G. Pretzler, A. Pukhov, and J. Meyer-ter-vehn,
Prog. Part. Nucl. Phys. 46, 375 (2001).
B. M. Hegelich et al., Nature (London) 439, 441 (2006).
H. Schwoerer et al., Nature (London) 439, 445 (2006).
T. Toncian et al., Science 312, 410 (2006).
S. Tar-Avetisyan et al., Phys. Rev. Lett. 96, 145006
(2006).
A. Noda et al., in Proceedings of the 7th Linear
Accelerator Conference, LINAC 2002, Vienna (EPS,
Geneva, 2002), p. 644.
A. Noda et al., Laser Phys. 16, 647 (2006).
S. Nakamura et al., Jpn. J. Appl. Phys. 46, L717 (2007).
A. Noda et al., Int. J. Mod. Phys. B 21, 319 (2007).
H. Kiriyama et al., Opt. Lett. 33 645 (2008).
M. Mori et al., IEEE Trans. Plasma Sci. 36, 1872 (2008).
A. Yogo et al., Phys. Rev. E 77, 016401 (2008).
M. Nishiuchi et al., Appl. Phys. Lett. 94, 061107 (2009).
S. Nakamura et al., Jpn. J. Appl. Phys. 45, L913 (2006).
A. Yogo et al., Phys. Plasmas 14, 043104 (2007).
S. Y. Lee, Accelerator Physics (World Scientific,
Singapore, 1999), p. 374.
A. P. Fews and D. L. Henshaw, Nucl. Instrum. Methods
Phys. Res. 197, 517 (1982); A. P. Fews, Nucl. Instrum.
Methods Phys. Res., Sect. B 71, 465 (1992).
K. R. Crandall and D. P. Rusthoi, Los Alamos National
Laboratory Report No. LA-UR-90-4146, p. 17.
U. Linz and J. Alonso, Phys. Rev. ST Accel. Beams 10,
094801 (2007).