L.O. Silva-Particle Acceleration & Relativistic Astrophysics in the Lab

8/4/2019 L.O. Silva-Particle Acceleration & Relativistic Astrophysics in the Lab

1/45

Particle acceleration &relativistic astrophysics in the lab

G

oLP/IF

PN

Institu

to

Superior

T

cnico

L. O. Silva | ELI Scientific Challenges, April 26 2010


2/45


Acknowledgments

S. F. Martins, F. Fiza, J. Vieira, J. Martins,M. Marti, R. A. Fonseca

Work in collaboration with:

F. Tsung, J. Tonge, J. May, W. B. Mori (UCLA)

Simulation results obtained at epp and IST Clusters (IST),

Dawson Cluster (UCLA), Franklin (NERSC), Intrepid(Argonne), and Jugene (FZ Jlich)


3/45


Contents

Laser e- acceleration

Boosted frame simulations 10 PW laser

High brilliance betatron radiation

Gamma ray beams

Relativistic beams for astrophysics

Relativistic shocks

Conclusions


4/45


Contents




Gamma ray beams


Relativistic shocks

Conclusions


5/45


Particle accelerators: rich science and applications

Adapted from Tom Katsouleas (Duke)

LargeVerified Standard Model of ParticlePhysics

W, Z bosons

Quarks, gluons and quark-gluonplasma

Asymmetry of matter and anti-matter

In pursuit of the Higgs boson

CompactMedicine

cancer therapy, imaging

Industry

lithography

Light sources (synchrotrons)bio imaging

condensed matter science

International Linear Collider


6/45


Particle accelerators

High Energy

HighLuminosity

High BeamQuality

Low Cost

L = fN2

/4xy

/ 0.1% 10% n yy < 1 mm mrad

[event rate]

[low energy spread] [low emittance]

[1/10 of 1010!/TeV]Gradients > 100 MeV/m Efficiency > few %


7/45


Recent achievements in laser-plasma accelerators

Guiding in plasma channels and ~1 GeV e- beams

Leemans et al, Nat. Phys. (2006)

Self-guided propagation and self-injected electrons (>1 GeV)

Hafz et al, Nat. Photonics (2008); Kneip et al, PRL (2009); Froula et al, PRL (2009);Clayton et al, submitted [1.4 GeV in 1.3 cm @ 1018 cm-3]

Controlled all-optical injection of monoenergetic electron beams

Faure et al, Nature (2006)

Beam loading in nonlinear wakes

Tzoufras et al (2008); Rechatin et al (2009)

Intense incoherent radiation (betatron x-rays & undulator radiation)

Rousse et al, PRL (2004); Kneip et al(submitted); Froula et al(in preparation) &H.-P Schlenvoigt et al, Nat. Phys. (2004); M. Fuchs et al, Nat. Phys. (2009)


8/45


Recent progress has put plasma accelerationat the forefront of Science


9/45


Tech developments have triggered recent progresses

09 Peak laser intensity ~ 1022 W/cm2 09 Peak computing power > 1 Tflop/s

Mourou, Tajima, Bulanov (2006) Source: top500.org


10/45

New Features in v2.0

Bessel Beams

Binary Collision Module

Tunnel (ADK) and Impact Ionization

Dynamic Load Balancing PML absorbing BC

Optimized higher order splines

Parallel I/O (HDF5)

Boosted frame in 1/2/3D

osiris framework

Massivelly Parallel, Fully RelativisticParticle-in-Cell (PIC) Code

Visualization and Data Analysis Infrastructure

Developed by the osiris.consortium! UCLA + IST

OSIRIS 2.0

Ricardo Fonseca: [email protected] Tsung:[email protected]

http://cfp.ist.utl.pt/golp/epp/http://exodus.physics.ucla.edu/ L. O. Silva | ELI Scientific Challenges, April 26 2010


11/45


OSIRIS strong scaling up to ~300k CPUs

! Spatial domain decomposition! Local field solver! Minimal communication! Dynamic Load Balancing

Optimize scalability and tap new hardwares

New hardware features

SIMD units

tailored code already in production

GPUs

CUDA development (test PIC code)

PowerXCell

1

10

100

104

105

Spe

ed

up

CPUs

JUGENEGermany


12/45


Can LWFA reach the energy frontierwith the next generation of lasers?

!! ! ! ! !! ! ! ! ! ! !

! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !

!

! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !

! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !

! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !

! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !

! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !

! ! ! ! ! ! !

! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

! ! ! ! ! ! ! ! ! ! ! ! ! !!

!"#$%%&'()"

*+,-.)/&0."1234

!""#$%&'()*

+,-'/#56(%(789+--: .;0.

!""#$%&'()+

+,-'/#56(%(789+--:.;0.

!""#$%& '(),

+,-'/#56(%(789+--:.;0.

-".%&'()/+01!+?@/&0%$6/

9+--:.;0.

*2-345"67/+01

!+?@/&0%$6/

,%&"5%&0..;0.

*2(?=@AB#

*+?=*CB#

*+?=*CB#

>DEFDG$%G#D$H$%#$IJ#%&B'6D5D$H

K0"$D6)FH#D6#

% 'GL)M

'66%5%&'$D"G

N;O=N&'H

E%G%&'$D"G

-5'#(')FH#D6#

3PQ

*2?

C2?

,22?

+0+22(< =*7>?=*2B#8))5D6'$D"G#

E0(%/#;%&)F@$(0/G$#&%H0&'/)$&%"#$/6#/"I

!%'(=)J5#%#.D$6FH'&L

C2?

! ! ! ! ! ! ! ! ! ! ! ! !

!!

!!!!

!

!!!

!!!

!! ! ! !!!!!!!!!!!

!!

!!!!


13/45


Limits to energy gain in LWFA

E= eEzLacc

Diffractionlaser pulse diffracts inscale ofZr (Rayleigh length) ~ few mm

Depletionlaser pulse looses its energy to the plasma in Ldeplfor small a0, Ldepl >> Ldph ; for a0 > 1, Ldepl ~ Ldph

TK & JMD 83

Recenin using

laser-p

ticles to high e

nergie

to whichlinear a

ccelerators

1limited.

The beat-wave ac

celerator2

proposedby Dawso

n and Tajima to

excite larg

tude electrostat

ic plasma waves

which can accele

ra

particles.' Whe

reas particles i

n the beat-wave

accel-

erator can g

ain onlya finite

amount of energy

before

they become out

of phasewith the

beat wave, by i

ntro-

ducing aperpend

icular magnetic

field the partic

les

are deflected acros

s the wave front

therebypreventi

them from outrun

ning thewave. T

he particles may

b

accelerated to a

rbitrarily high

energy as they r

id

across the wave'

fronts like surfers

cuttingacross

face ofan ocean

wave (see Fig. 1

).

-B AB, z

/ yFig. 1 An

electrontrapped

by a pot

ing at Vph sees

an electric

which accelerates

it across

Sugiharaand Midzu

no3 and

shown that class

ical particles

dicularly propag

ating electros

ted until they d

e-trap near t

is letter we con

sider t

when theE x B ve

i.e., E> B

comp

Dephasingelectrons overtake accelerating structurein Ldph ~ 10 cm/n0 [1016 cm-3]

v ~ vgroup laser

v ~ c


14/45

Blow-out regime of laser wakefield acceleration



15/45


Match laser spot size tobubble radius

Linear focusingforce

Electric fields created by laser pulse

Phenomenological theory based on physical picture

W. Lu et al. PR-STAB (2007)

Longitudinal

Ezmaxa0

kpR kpW0 = 2 a0

Letch > Ld

cFWHM > 2R/3Ld

2

3

20

2p

R

Matched laser parameters

Transverse

kpR 2a0

Linear acceleratinggradient

For maximum energy gain:trapped e- dephasing before pump depletion

Letch c2

0/2

pFWHM


16/45


W0 =3

2cFWHM

np[1018 cm3] 3.71

a30

P[TW]

0[m]

0.8

2

FWHM[fs] 53.22

0[m]

0.8

2/3

[J]

a20

1/3

Lacc[cm] 14.09[J]

a30

q[nC] 0.17

0[m]

0.8

2/3([J] a0)

1/3

E[GeV] 3

[J]

a20

0.8

0[m]

2/3

Different regimes for the LWFA

* S. Gordienko and A. Pukhov PoP (2005)** W. Lu et al. PR-STAB (2007)

Main goalMaximize

electron energy

Efficiency

0.52/a019%

Typical a0 2nc/np (nc/np)

1/5 2 3

Maximum electron energy


17/45


Parameter range for 250 J laser system

Laser

Plasma

e- Bunch

a0

Spot [m]

Duration [fs]

Density [cm-3]

Length [cm]

Energy [GeV]

Charge [nC]

* S. Gordienko and A. Pukhov PoP (2005)** W. Lu et al. PR-STAB (2007)

Simulationtime [days in

512CPUs]


18/45


Boosted Frames in LWFA simulations

L

aboratoryFrame Resolution gains

Particles

Time steps

Total

Time step (1 + )

Resolution (1 + )

Plasma

contraction

Total time (1 +

)

2(1 + )2BoostedFrame

J.-L. Vay, PRL 98, 130405 (2007)


19/45


+3GeV self-injection in strongly nonlinear regimeExtreme blowout a0=53

Laboratory frame3000x256x256 cells

~109 particles105 timesteps

x2

[m]

ElectronEnergy[GeV]

90

60

30

0

3

2

1

0

x1 [ m]

806040200

x1 [ m]

806040200

x1 [ m]

806040200

x1 [ m]

80604020010

-3

10-1

e-density[1

.5e19cm

]-3

101

10

1

S.F. Martins et al, Nature Physics (2010)


20/45


~300x fasterthan lab simulation

ElectronEnergy

Lab[GeV]

2

6

10

14

130001250012000

600050004000

x2

Boost[m]

x1 Boost [m]

x1

Boost [m]

x1

Boost [m]

x2

Boost [m]

100

200

300

400

400

200

200

300

300

2400

0

1200

Density

e-Energy

Lab[GeV]

-3

Density

[2.7e17cm

]-1

80

40

0

5

3

1

10

-2

10

x3

Boost [m]

Laser

pulse

Accelerating

electron beam

+10GeV self-injection in nonlinear regimeControlled self-guided a0=5.8

Boosted frame7000x256x256 cells

~109 particles3x104 timesteps

=10


21/45


10

20

30

40

0 1 2 3 4 5

Ez

[GV/cm]

x1

Boost [m]

0.1

0.0

-0.1

6000400020000

Distance [m]

Energy[GeV]

+40GeV with externally injected beamsChannel guided a0=2

Boosted frame8000x128x128 cells

~5x108 particles2x105 timesteps

=10

~300x fasterthan lab simulation


22/45


Regime comparison and fine-tuning

7000x256x256 cells~109 particles

3x104 timesteps=10


23/45

Particle accelerators & plasmas

High Energy

HighLuminosity

High BeamQuality

PolarizedBeams

Low Cost



24/45


Contents




Gamma ray beams


Relativistic shocks

Conclusions


25/45


Applications for LWFA beams

LINAC

undulator

100m

1km

1cm

E > 10 GeV

B ~ 1 T

LightSources(FEL)

Plasm

abased

Ultra shortaccelerating

structure

Undulator likemotion in ion

channel


26/45


Betatron radiation in plasma wakefield acceleration

Ultra shortaccelerating

structure

S. Wang et al, PRL (April 2002)

I ~ 1019 photons/s 0.1 % BW mm2 mrad2@ 6 keV


27/45

Scaling for multi-PW lasers (300 J class lasers)

Wigglerstrength

CriticalFrequency

RadiatedPower

Radiated

Energy

* W. Lu et al. PR-STAB (2007)** E. Esarey et al. PRE (2002)

3a2

r/c

remec22

a2

Ne/3

Power lacc/c

kpr0

0/p

= p/2

kpr03/2

p/0 0

= r0/W0

26([300 J])1/6(0[m]/a0)1/12

580([300 J]/a20)1/3(0[m])

1/60

2.8 107([300 J])5/6(0[m])5/12a1

/6

0 0

2.52([300 J])1/3a4/30

(0[m])1/3kW

32([300 J])2/3a5/30

(0[m])1/6GW

372([300 J])4/3a2 30

(0[m])1/3J

53 2([300 J])5/3a1/30

(0[m])1/6mJ

...

...

...



28/45


Particle tracking in OSIRIS

Technically challenging

Subset of ~103 particles in ~109

Storing information for every particlenot feasible

104 iter. " 109 part. ~ 500 TB

Relevant physics associatedwith small subset of particles

Record detailed 7D phase-spaceof interesting particles

follow interestingparticles

tag all particles


29/45

>10GeV simulation for next generation lasers

Total radiated energy ~ 80 mJ

Typical photon energy ~ 0.1-1 MeV

Typical # of photons ~ a few 1011

Beam divergence ~ Kwiggler/ ~ 2 - 5 mrad

Ebeam ~ 10 GeV

nplasma ~ 1017 cm-3

r0 ~ rbeam ~ 5 m

+10 GeV blowout LWFA stage

Radiated Power

P =1

12

e2

c34

pr2

02

Typical Frequency

c =3

4c2r0

2

p

Kwiggler ~ 52

Full Particle Tracking & Radiation



30/45


Collimated radiation beam for >12 GeV e- beam

measured beam angulardivergence (/2):in x2 = 0.64 mradin x3 = 0.61 mrad

26.5 cm 25.8 cm


31/45


X-rays to -rays betatron radiation

spectral intensity at center point

zoom


32/45


Contents




Gamma ray beams


Relativistic shocks

Conclusions


33/45

Gamma Ray Bursters

Relativistic colliding flows present in many astro scenarios

N. Gehrels, L. Piro and P. J. T. Leonard, Scientific American, Dec. 2002, p. 89



34/45

Plasma instabilities critical to shock formation and field structure

B-fields generated by currentfilamentation/Weibel in GRBs[Medvedev & Loeb, Gruzinov & Waxman, 99, Silva et al, 03]

Fields in relativistic shocks

are mediated by Weibel/current

filamentation generated fields

[Spiktovsky 08, Martins 09]

Shockfront

|B|2

density



35/45

Fields in shock Fermi acceleration B-field generation/amplification

Ab initio Fermi acceleration

determined by structure of the fields

in the shock front

[Spitkovsky 08, Martins et al, 09]

B-field amplification in

upstream region via non-

resonant Bell instability[Bell 04]

-

-



36/45


37/45

z [m]x [m]

100

100

150

150

50

50

0

40

80

y [m]

150 0 20 40 60 8010050

150

100

50

y[m]

150

100

50

y[m]

x [m] z [m]

Density[2.7e17cm-3]

3.0

2.0

1.0

0.0

MagneticField[T]

20

10

0

-10

-20

100

10

1

Density

Magnetic Field

Beam is charge neutral = no blow-out

ElectronsPositrons

Beam filamentation and B-field generation



38/45

Standard

5x beam temperature

10x plasma density

2 4 6 8 10

Distance x [cm]

Log

(

10

B

B

0

2 4 6 8Distance x [10 c/ ]p

3

-0

-1

-2

-3

-4

-5

WeibelinstabilityLinear stage

Beam length< p

Transition from purely transverse to mixed mode

Beam filamentation and B-field generation

nb=nplasma

Beam length~ p



39/45

Radiation is direct evidence for scattering in Weibel turbulence

n

v

t

d2I

dd=

e2

4c

n [(n )]

(1 n )2ei(tn.r(t)/c)

dt

2

100 101 102 103 104

Var1

106

107

108

Selected

Va

riables

!/!p

P(!)

!!!!

-1

!!-2

Synthetic radiation spectrum

dI/d



40/45

H. Takabe

H. Takabe et al, PPCF 50, 124057 (2008)

Formation and propagationof Weibel mediated

collisionless shocks

Recent developments Youichi Sakawa et al, HEDLA (2010), submitted to PRL (2009)


41/45

Numerical Parameters

x kp = 0.5 - 1.5

z kp = 0.5 - 1.5

Particles per cell = 64 # particles = 5x109

# time steps = 105

Ignitionlaser

0 = 1m

I0 = 5x1019 - 5x1021 Wcm-2

plane polarized

56 m x 16 m

ne0 = 100 nc

mi/me = 3672 (D+)

Ti0 = Te0 = 100 eV

Physical Parameters

Laser

Plasma

F. Fiza et al, in preparation (2010)L. O. Silva | ELI Scientific Challenges, April 26 2010


42/45

Magnetic field

Electron density

Ion phase-space

Current filamentation instabilityleads to thermalization/slow down

of incident beam

reflected ions

shock front

hot region

strong mass build-up/compression

~ 20 c/pi



43/45

electron density

vreflected ions = 0.2 cvshock= 0.1 c

vhole boring = 0.07 c

laser

vshock

vreflected ionsvhole boring

n2

n1

hb =

nc

2ne

Zm

M

I2

1.37 1018

1/2= 0.07*

shock =(1 + add)

2d 1

1 + d + ad(2

d 1)

0.1n2

n1

=add + 1

ad 1 3

**

~ 100 c/pi



44/45


Contents




Gamma ray beams


Relativistic shocks

Conclusions


45/45


Conclusions

+ 40 GeV beams with 250 J laser system

New laser systems in the 10 PW range able to explore full range of LWFAscenarios

From high charge multi-GeV beams to externally injected 40 GeV beams

High brilliance gamma-ray beams from betatron radiation

Betatron radiation generated can reach novel parameters not achievable withother machines

Gamma-ray beams from betatron radiation with ~ 10 GeV beams

Relativistic beams/flows for astrophysics

Beams appropriate to probe basic processes relevant in relativisticastrophysics

Ability to drive relativistic shocks in the laboratory

Documents

L.O. Silva-Particle Acceleration & Relativistic Astrophysics in the Lab