Advanced laser particle accelerator ... - Library Without Walls

LA-UR-

J()- 05/?~

Approved for public relea se; distribution is unlimited.

Title:

Advanced Laser Particle Accelerator Development at LANL: From Fast Ignition to Radiation Oncology

Author(s):

K. A. Flippo, S. A. Gaillard, T. Kluge, M. Bussmann, D. T . Offermann,J. A. Cobble, M. J. Schmitt, T. Bartal, F. N. Beg, B. Gall, D. C. Gautier,M . Geissel, T . J. T. Kwan, G. Korgan, S. Kovaleski, T. Lockard, S. Malekos,D . S. Montgomery, M. Schollmeier,Y. Sentoku, and T . E. Cowan

Intended for:

MC Workshop 2010 Proceedings of the AlP

,,;QAlamos NATIONAL LABORATORY -~~

EST.1943 - - -

Los Alamos National Laboratory , an aHirmative action/equal opportunity employer, is operated by the Los Alamos National Security, LLC for the National Nuclear Security Administration of the U.S. Department of Energy under contract DE-AC52-06NA25396. By acceptance of this article, the publisher recognizes that the U.S. Government retains a nonexclusive, royalty -free license to publish or reproduce the published form of this contribution , or to allow others to do so , for U.S . Government purposes. Los Alamos National Laboratory requests that the publisher identify this article as work performed under the auspices of the U.S. Department of Energy . Los Alamos National Laboratory strongly supports academic freedom and a researcher's right to publish; as an institution, however, the Laboratory does not endorse the viewpoint of a publication or guarantee its techn ical correctness. Form 836 (7/06)

Advanced Laser Particle Accelerator Development at LANL: From Fast Ignition to Radiation Oncology a

A. Flippoa, S. A. Gaillard b, T. Kluge b, M. Bussmann b, D. T. Offelmann , a d 1. A. Cobble a, M. 1. Schmite, T. Bartal\ F. N. Beg\ B. Gall , D. C. Gautier , g f g d a M. Geissel\ T. 1. T. Kwan , G. Korgan , S. Kovaleski , T. Lockard , S. Malekos , f b D. S. Montgomerl, M. Schollmeie{,Y. Sentoku , and T. E. Cowan K.

a Los Alamos National Laboratory, PO BOX 1663, Los Alamos, NM 87545 ForschungsZentrum Dresden-RossendOlf, Bautzner Landstr. 400,01328 Dresden, GERMANY University O;:Calijornia, San Diego, Mechanical and Aerospace Engineering Department, La Jolla, CA 92038 University of Missouri, Electrical and Computer Engineering, Columbia MO 65211 e Sandia National Laboratories, PO Box 5800, Albuquerque, NM 87185 J University of Nevada, Physics, Reno, NV 89557 g Nanolabz, 661 Sierra Rose Dr. , Reno, NV 89511 b

c

Abstract. Laser-plasma accelerated ion and electron beam sources are an emerging field with vast prospects. and promise many superior applications in a variety of fields such as hadron cancer therapy, compact radioisotope generation, table-top nuclear physics, laboratory astrophysics, nuclear forensics, waste h'3nsmutation, SN M detection, and inertial fusion energy. LANL is engaged in several projects seeking to develop compact high current and high energy ion and electron sources. We are especially interested in two specific applications: ion fast ignition/capsule perturbation and radiation oncology in conjunction with our partners at the ForschungsZenrrum Oresden-Rossendorf (FZO). Laser-tobcam conversion efficiencies of over 10% are needed for practical applications, and we have already shown inhcrcnt etliciencies of >5% from flat foils, on Trident using only a 5th of the intensity and energy of the Nova Petawatt. With clever target designs, like srructured curved cone targets, we have also been able to achieve major ion energy gains, leading to the highest energy laser-accelerated proton beams in the world. These new target designs promise to help usher in the next generation of particle sources realizing the potential of laser-accelerated beams.

Keywords: laser-acceleration, laser-plasma interactions, plasma based accelerators, ion acceleration, fast ignition, hadron therapy, dircct laser light acceleration, target normal sheath acceleration PACS: 41.75.Jv, 52.38.Kd, 52.38.Ph, 52.57.Kk, 52.65.Rr, 52.65.Ww

MOTIVATION AND INTRODUCTION The LANL advanced laser-based ion beam program is focused on improving laser-ion sources [1,2,3,4] for applications of interest in science and technology, for example the study of astrophysical phenomena [5], nuclear reactions and isotope production [6,7], nuclear material detection and forensics [8], ion fast ignition [9] inertial confinement fusion [10 , 11,12,13], the radiography of dense objects, transient electric and magnetic fields [14,15], and as an injector for accelerators [16]; as well as in radiation oncology using proton or ion therapy [17]. Until recently the highest maximum proton energy of 58 Me V [1] was obtained 10 years ago using the LLNL Nova PW laser with - 450 J of laser energy, - 2.6x I 0 20 W/cm 2 of laser intensity, and solid targets, via the TNSA mechanism p 8]. Since then, and for an entire decade , that record using flat-foil (FF) targets, or any other target type or geometry, though equaled at lower intensity [19], had never been surpassed [20], even though some of the data has been acquired with significantly higher laser energy of up to 1000 J, and/or intensity of up to 2o 2 3x 10 W/cm • Increasing the laser to ion conversion efficiency is critical for keeping laser energy low in applications like Ion Fast Ignition Fusion [9] and maximizing the energy of laser-accelerated proton beams is of

considerable interest to many applications, and in particular for radiation oncology via proton therapy [17, 21, 22, 23,24]. The therapeutic window for cancer treatment starts at 62 MeV for eye cancer (for a depth of30 mm) [25], and extends towards 225 MeV (for a depth of 30 cm) for deep-seated tumors [17]. For the first time new experiments have reproducibly broken the 60 Me V empirical barrier [20] using flat-top cone (FTC) targets at modest laser energy (80 J) and intensity (1.3 x 1020 W/cm2), and in which a new mechanism responsible for the TNSA electron population is identified. Conventional wisdom suggests that, with.in a TNSA-like acceleration mechanism, for which one usually assumes an isothermal expansion of the electrons in vacuum [26], increasing the electron temperature will increase the maximum ion energies. Target geometries proposed to do so include reduced-mass targets (RMTs) [27], stacked foils [28] and cone targets [29,30,31].

LANL MICRO-CONE EXPERIMENTS: BOOSTING PROTON ENERGIES Using LANL's TRIDENT laser we have previously shown an increase in the maximum proton energy from 19 MeV with FF target to over 30 Me V with FTC targets at 20 J of laser energy [32] (see Figure 1(a)). This increase was partially attributed to optical collection (i.e. microfocusing) at the cone tip, yielding an enhancement in laser intensity [29]; and partially to better laser absorption. On that experiment however, no diagnostics of the hotelectron distribution inside the cone was available. Follow-on experiments at Trident using 80 J at intrinsic contrast (i .e. IO. R) did not show an increase in proton energy [19]. This was identified to be due to a large preplasma filling the cone, causing the laser to be absorbed far from the flat top, therefore preventing the hot electrons from efficiently reaching the flat top [19,33]. The new experiments were performed at 80 J, but at a much higher laser contrast (i.e . > 10-10) [34] , which minimized the preformed plasma. The proton energies were observed to be significantly enhanced under the specific condition that the laser grazes along the cone wall of a wide-neck FTC. The proton energy was increased to 67.5 MeV, as compared to 50 MeV using a FF. Prior simulations focused on a well-aligned laser interacting with conical targets [29,32]. But now we have, for the first time, systematically studied, via simulations, compared to experiments, the local interaction of the laser with the flat top as well as with the cone wall, yielding a novel understanding of laser-accelerated protons using cone-shaped targets[ 35]. This may lead to a new path for even higher proton energy in the future, using simpler and better optimized targets.

(b) 1.E+12 .-----r----..

ct>FF '1'1

l.E+ll

rl

~ Q/

::e ...... ...

-+1.E+l0

r2

Q/

.c

E l.E+09

'fo01

..

I---+--+--t-r--+--;--+--!--i

IC-« t----+,EoO'

:l

Z

-+--+--+-.-t--t---t-----i

6-?----+~-I-_-I--.......~-+---I-.....,....j_~

o

~

•

»

~

Proton Energy (MeV]

~

~

~

g 1.E-HlS

2 0.

l.E ..07 l.E"06

o

20 40 Proton Energy [MeV]

60

FIGURE 1. (a) The increase in proton energy for a FTC target ( e and e ) at 20 J on the TRIDNET at intrinsic contrast, 10. 8, as compared to a FF ( . ) with a simulated FTC spectra (0 ). (b) Increase in proton energy for FTCs (t and t ) compared to FFs (t) on TRIDENT at 80 J and high contrast, 10. 10. (b inset) copper Ka x-ray images of FTC r I and n , an asymmetric interaction.

Figure I (b inset) shows copper K-a x-ray images of FTC r I and fl, showing the asymmetric nature of the laser interacting with the cone walls which is related to where the laser light is absorbed as described in Ref 33. These observations were the impetus for running simulations using the PICLS code [36] at grazing incidence. These new simulations, shown in Figure 2, show an increase in the electron-current density and an increase in the laser focal intensity at the flat-top for the case of laser grazing. The increased intensity due to the grazing along the wall to the

flat-top goes from an ao=13.3 (aligned on axis with no offset) to an ao=18.8 (aligned on axis with a grazing offset along the wall) at the flat-top of the cone. The is equivalent to a ponderomotive temperature increase from 3.9 MeV to 6.2 MeV, but the simulations show that the temperature in the first case is 4.1 MeV and in the grazing case is 9.2 MeV, which cannot be accounted for by the pure increase in intensity. Upon further analysis, simulations show that the electrons bunch along the grazing section of the wall. When the laser pulse is close to the cone wall, it can penetrate into cone wall plasma up to a skin length beyond the overcritical density border. From there the electric field, which is almost normal to the wall, can extract electrons and pull them into the vacuum. If the laser intensity is high enough, the vxB force becomes strong enough to convert the transverse electron momentum into forward motion fast enough for the electron to approach the phase velocity of the laser before it would be pushed back into the cone wall by the next laser phase. Now, the electron remains in phase with the laser and its transverse momentum is continuously converted into forward momentum. This extraction and acceleration of electrons causes a continuous return-current inside the cone wall close to the surface, which gives rise to a quasi-static magnetic field (Figure 2 (a)). This field prevents the extracted electrons from reentering the overcritical plasma and being decelerated again, so that the electrons are guided along the surface [29,37]. Hence, the laser can continuously interact with the electrons and continue to accelerate them. During this continuous acceleration the electrons can gain energies well exceeding the ponderomotive energy.

FIGURE 2. PleLS simulation, 140 fs after the laser hits the flat top for a 15 flm offset case (the laser is grazing the cone wall horizontally), showing (a) Bz , (b) E and (inset) current density (color scale is in log-scale). Simulations by T. Kluge at FZD.

LANL ION SOURCE DEVELOPMENT ON OMEGA EP LANL has been leading a series of ion generation experiments on Omega EP to develop high-energy-content laser-driven carbon ion beam sources for dynamic defect studies of imploding fuel capsules. LANL has developed a new suite of diagnostics: The Thomson Parabola Ion Energy analyzer (TPIE) [38], the Target Heating Verdi laser (THVL) [39], and the Li/F activation pack for the LNLL PFP-II diagnostic [40] , for these experiments. Initial data on the performance of Omega EP in terms of ion energies from dielectric targets (CVD diamond) and conductors (Cu) have been gathered using the PFP-II loaded with Radiochromic film as the primary diagnostic. F,igure 3 compares the beam quality and spectra from CVD diamond (a) and copper (b) targets at 15 micron thicknesses. The beam profile from the 15 micron target has a well defined outer edge of the proton beam and a higher temperature (Maxwellian fit) to the beam, 5.58 MeV versus the beam from the copper target, which had a large defuse beam, and an inner darker beam, with a characteristic temperature of3.79 MeV. Overall the beam from the copper target contained more protons in the 4-36 Me V range, but fewer total ions in the higher end of the spectrum, 15-36 MeV. Approximately 2.04% and 1.81 % respectively of the laser energy was converted into protons from the CVD and copper targets. Planar foil targets of various thicknesses along with hemisphere targets (800 urn diameter, 35 urn thick) are shown in Figure 4, where one can see that the proton beam energy from planar targets peaks around 45 MeV at a target thickness of about 24 microns for an EP pulse length of lOps and a pulse energy of ~ 1000 J, and falls off for thicker targets. This is a very different behavior from other lasers [21,41], were there is a gentle ramp from increasing from thicker to thinner targets, and a sharp drop at some minimum thickness, Figure 4( b),(c),(d). This indicates that the lOps regime of the OMEGA EP lasers is significantly different than the sub I ps regime of other short-pulse laser systems. The hemi-shell targets were used to show ion beam focusing. The focal spot size was varied on the hemi, to find a dependence on focal spot to ion yield and focal position The preliminary results for protons are promising,

CVD Target Beam Profile at ( b) 15 Jim 965J, 10ps, 3.8x10'11 W/cm2

15 Jim Cu Target Beam Profile at

(a)

982J, 10ps, 4.1 x1019 W/cm2

t-I) .

t1J . 23.8 MeV

HO · ' 1l3 MeV

1() 6 M" V

1-0- 284 Mev

HD · 14.9 MeV

Ill :=~ : ~~ IO

1.81'" CE

. .

---

HO· 36.0 M.V

HD - 23 .8111eV

HO - 18.3 MeV

,.

2.C4'l1 CE

.-

HO- 36.0 M.V

"

HO- 28.4 MeV

,

FIGURE 3: Compariso n of 15 pm CYD diamond target (a) with J5 pm copper ta rget (b). The panels s how data from the RCF film stack with energies associated for each layer, and a proton spectrum fitting the data in the lower right of each panel.

(a)

50

l

• Cu Flat 2-"2 mm

50 - - - - - - - - ---1 y = 1.22x • 17.2 W=0.9802

• _

Cu Flat 500"2 urn CuRat1~2um

RMT

• eve AaI2"'2 mrn

>' 11130

Ma"'"

~

.

• eve Rat 1,42 mm

K

Aug

[20

CVD Hornis 800

W

micron~

-------- - - - - - - - - - - ------i

10

-Linear (CVD FbI 2"2 mm~)

0 0

10

5

15

20

25

30

35

Target Thickness[umj

(c)

(b) LLNL Callisto a •

•

~

16

~

14 12 10 8

~

~

t:i .,\

UJ

0

-4....0..-

---_

o

2

40

e

Targel Thickness

80 ( ~m)

o 10 0

120

•

,Io~dGr\I

- l.niar(C,,".c.~MvrflJ

P~

.... .N O.e

COfr"",f'd M~¥

- - - -l.JMM I CofTeded ..:..l)

I" ii, T

Ifl "'-

I

6 4

0

L.crwer "'ectio"....!!~t

:2 0

-

(d)LULl100TW

UMich T'

C~t ~foI R~M·.1df

!

~ K-

----

;

~~-

,~-