x-ray light sources driven by laser plasma accelerators

LLNL-PRES-742828 This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. Lawrence Livermore National Security, LLC

X-ray light sources driven by laser plasma accelerators for

high energy density science experiments

Félicie Albert, PhDPhysicist

Deputy Director, LLNL High Energy Density Science center

FESAC Meeting – August 31st 2021

Collaborators

1

M. Sinclair, K.A. Marsh, C.E. Clayton, and C. Joshi (UCLA)

E. Galtier, P. Heimann, E. Granados, H. J. Lee, B. Nagler, A. Fry (LCLS)

W. Schumaker, F. Fiuza, E. Gamboa, L. Fletcher, S.H Glenzer (SLAC )

P. M. King, I. Pagano, M. C. Downer, B. M. Hegelich (UT Austin)

A. Ravasio, F. Condamine, M. Koenig (LULI)

A.G.R. Thomas, C. Kuranz

ï100 0 100 200 300 400 5000.8

1

1.2

1.4

1.6

pixel number

inte

nsity

a.u

.

RAL 2014 [email protected]

very preliminary analysis of shock data

• Lineout clearly reveals shock front – and possibly the density difference??

unshocked

shocked

B. Barbrel, J. Gaudin, F. Dorchies(CELIA)

M. Berboucha, S. Mangles, J. Woods, K. Powder, N. Lopes, E. Hill, S. Rose, Z. Najmudin (Imperial College London)

D. Kraus, R.W. Falcone, C. Geddes (UCB/LBNL)

P. Zeitoun (LOA)

J. L. Shaw, D. H. Froula (LLE)

N. Lemos, P. M. King, B. B. Pollock, C. Goyon, A. Pak, J. E. Ralph, Y. Ping, A. Fernandez-Panella, S. Hau-Riege, T. Ogitsu, A. Saunders, J. D. Moody


§ X-rays: a powerful tool for high energy density science experiments

§ Using high power lasers to generate laser plasma accelerators and x-rays

§ Two applications of x-rays from laser plasma accelerators- Imaging complex high energy density science experiments

- Understanding electron-ion equilibration in warm dense matter

§ Conclusion

Outline

2

FESAC Meeting – August 31st 2021 3

At LLNL we use the National Ignition Facility (NIF) and concentrate its 192 beams into a mm3


Such experiments create extreme, transient conditions of temperature and pressure that are hard to diagnose

4

100 million degrees

20x the density of lead


Many High Energy Density Science experiments rely on x-ray backlighters with unique properties

5

Bremsstrahlung Titan – 10 ps

Broadband emissionOMEGA – 100 ps

Line emissionNIF – 1 ns

Sandia Z – 3 nsÊ

Ê

Ê

1 2 5 10 20 50 100

105

107

109

1011

1013

Photon energy @keVD

Phot

onsêe

VêSrêps

X-ray sources for HEDS experiments Barrios et al, HEDP 9, 626 (2013)

- Radiography, X-ray diffraction

Ping et al 84, RSI 123105 (2013)

- X-ray absorption spectroscopy

Bailey et al, Nature 517, 56 (2015)

- X-ray opacity

Jarrott et al, POP 21 031201 (2014)


Many High Energy Density Science experiments rely on x-ray backlighters with unique properties

6

Barrios et al, HEDP 9, 626 (2013)

- Radiography, X-ray diffraction

Ping et al 84, RSI 123105 (2013)

- X-ray absorption spectroscopy

Bailey et al, Nature 517, 56 (2015)

- X-ray opacity

Jarrott et al, POP 21 031201 (2014)

This work – x-rays driven by laser wakefield acceleration

X-ray sources for HEDS experiments

Sandia Z – 3 ns

Broadband emissionOMEGA – 100 ps

Betatron radiationps

Line emissionNIF – 1 ns

Bremsstrahlung Titan – 10 ps

Ê

Ê

Ê

1 2 5 10 20 50 100

105

107

109

1011

1013

Photon energy @keVD

Phot

onsêe

VêSrêps


Broadband (keV – MeV)

Ultrafast (fs – ps)

Collimated (mrad)

Small source size (µm)

Synchronized with drive laser (ns, fs, or XFEL)

Unique properties of sources driven by LWFA can enable applications

7

Shock physicsPhase contrast imaging of laser driven shocks

Hydrodynamic instabilities motionµm resolution imaging without motion blur

Radiography of dense targetsMeV x-rays with small source size

OpacityBroadbad backlighter over 100’s of eV

Electron-ion equilibration/warm dense matterX-ray absorption spectroscopy with sub ps resolution

Unique properties Applications

TargetDrive laser

X-rayprobe


Broadband (keV – MeV)

Ultrafast (fs – ps)

Collimated (mrad)

Small source size (µm)

Synchronized with drive laser (ns, fs, or XFEL)

Unique properties of sources driven by LWFA can enable applications

8

Shock physicsPhase contrast imaging of laser driven shocks

Hydrodynamic instabilities motionµm resolution imaging without motion blur

Radiography of dense targetsMeV x-rays with small source size

OpacityBroadbad backlighter over 100’s of eV

Electron-ion equilibration/warm dense matterX-ray absorption spectroscopy with sub ps resolution

Unique properties Applications

TargetDrive laser

X-rayprobe


X-ray sources with MeV photons and <10 µm resolution are required to understand some of the experiments done at the NIF

9

X-ray source

Double shell implosion, 1800 g/cc

100

µm

1 MeV radiography

0.2 MeV radiography

Fully implodedcapsule


We use ”pump-probe” experiments and x-ray measurement techniques to understand these conditions

10

Wavelength

Inte

nsity

X-rays in X-rays out

Inte

nsity

Wavelengthsample

X-ray absorption spectroscopy


Inte

nsity

Wavelength

We use ”pump-probe” experiments and x-ray measurement techniques to understand these conditions

11

sample

Laser in at Time 0 (T0)

Wavelength

Inte

nsity

X-rays outX-rays in atT0+T


4th generation x-ray light sources used for scientific applications could be used but are billion dollar-scale national facilities

12

Synchrotron: APS

Argonne Nat. Lab., IL

X-ray free electron laser: LCLS

SLAC, CA

3 km


Laser plasma accelerators offer a compact alternative to these big machines

13

Laser plasma accelerator

LLNL1 meter

Accelerating electrical field is 1000 times stronger than in a regular accelerator

SLAC, CA

3 km

X-ray free electron laser: LCLS


Using high power lasers to generate laser plasma accelerators and x-rays


An intense laser pulses drives electron plasma waves

15

Wake behind a boat

60 µm

Plasma wave behind a laser

Nuno Lemos, LLNL


Laser-produced plasmas can naturally sustain large acceleration gradients which makes laser plasma accelerators 1000 x smaller

16

100 MV/m 100 GV/m

E0 =mcω p

eω p =

nee2

mε0

ne =1018 cm−3 → E0 = 96 GV/mPlasma frequencyAcceleration gradient

SRF Cavity Gas cell – laser plasma

FESAC Meeting – August 31st 2021 17F. Albert et al, Plasma Phys. Control. Fusion (2014)

Laser pulse

Electron plasma wave

Trapped Electron

Betatron X-ray beam


Laser wakefield acceleration can produce x-rays using several processes

18

e- X-rays

Betatron x-ray radiation

Laser photon

Scattered photonCompton scattering

+Bremsstrahlung

Gamma-ray photon

Nucleus

1

2

3

e-

e-

LaserGasne ~ 1019 cm-3

Nozzle


Nozzle


Nozzle

Low Z foil

High-Z foil

Ex ~ 𝛾!𝑛"𝑟#

Ex ~ 4𝛾! EL

Ex ~ 𝛾

keV

keV - MeV

MeV


Most of these sources are typically produced with ultrashort laser pulses in the blowout regime (cτ ~ lp/2)

19

ne

0

e-de

nsity

cτ

lp=c/ωp~ 1/ne1/2

Condition to be in the blowout regime cτ ~ 1/ne1/2 30 fs ne~ 1019 cm-3


Self modulated laser wakefield acceleration is easier to achieve with picosecond scale lasers (cτ >> lp)

20

Condition to be in the self-modulated regime cτ >>~ 1/ne1/2 1 ps ne~ 1019 cm-3

e-de

nsity

ne

0

cτ

lp=c/ωp~ 1/ne1/2


High charge, relativistic electron beams are accelerated through self-modulated laser wakefield acceleration

21

Electron plasma wave

Laser pulse envelope

cτLaser300 µm (1 ps)

lpLaserI>1018 W/cm2

Gasne ~ 1019 cm-3

Nozzle

lp = c/ωp~ 1/ne1/2

10 µm

Creation of an electron plasma wave (EPW)

1 Raman forward and self-modulation instabilities

Wave breaking traps electrons in EPW potential

ω0=ωs +/-mωpk0 = ks +/-mkp

2 3

Laser Intensity

Electron density


Imaging complex high energy density science experiments


Our project is developing laser plasma accelerators on large kJ-class picosecond lasers

23

TitanCompton scattering/Bremsstahlung

TitanSMLWFA electronsBetatron radiation

Titan, LLNL OMEGA-EP, LLE NIF – ARC, LLNL LMJ – PETAL, CEA

OMEGA-EPSMLWFA electrons

NIF - ARCSMLWFA electrons

TitanRadiography

LMJ – PETAL SMLWFA electrons

NIF - ARCX-ray sources

OMEGA-EPX-ray sources

OMEGA-EPRadiography


Laser wakefield – betatron experiments – Titan LLNL

Titan Laser150 J0.7 ps

Target3 mm He jet

ne = 1019 cm-3

86 % in 28 µmI = 5x1018 W/cm2

A. Saunders

W. Schumaker

C. Goyon J. Shaw

N. Lemos

F. Albert

B. PollockS. Andrews

self-modulated regime

cτ

lp=c/ωp~ 1/ne1/2


We have developed a platform to produce x-rays in the self modulated laser wakefield acceleration regime

25F. Albert et. al, Phys. Rev. Lett. (2013), Phys. Rev. Lett. (2017)

MeV

70

90

150

1000 800 nm

…

200 600 1000

0

500

1000

1500

2000200 600 1000

0

500

1000

1500

2000

200 600 1000

0

500

1000

1500

2000200 600 1000

0

500

1000

1500

2000

200 600 1000

0

500

1000

1500

2000200 600 1000

0

500

1000

1500

2000

200 600 1000

0

500

1000

1500

2000200 600 1000

0

500

1000

1500

2000

Interferometer Electronspectrometer

Optical Spectrometer

Filters

Detector(CCD, Image plates, stepwedge filter)

→B


Electrons accelerated in the SMLWFA regime produce betatron x-rays

26

2 4 6 8 10 12

0.05

0.10

0.15

Filter

Yie

ld@N

orm

aliz

edD

1

2

3

4

56

7

8

9

10

111213



27

2 4 6 8 10 12

0.05

0.10

0.15

Filter

Yie

ld@N

orm

aliz

edD

2 4 6 8 10 12

0.05

0.10

0.15

Filter

Yie

ld@N

orm

aliz

edD

10 20 30 40 50

0.02

0.05

0.10

0.20

0.50

X-ray energy @keVD

X-

ray

inte

nsity@ar

b.un

itD

X =

Ec = 20 keV

of x rays through elements of the system (Al, Mylarwindows) and the calibrated image plates’ absorptionand efficiency [24]. For photon energies between 1 and30 keV, we utilize the filter wheel. Assuming that thebetatron motion of the electrons dominates the observedx-ray emission in this range, we consider an intensitydistribution per unit photon energy dE and solid angle dΩas a function of the photon energy E of the form:

d2IdEdΩ

∝!EEc

"2

K22=3½E=Ec"; ð1Þ

which is valid for betatron x rays on axis [25]. Here, Ec isthe critical energy of the betatron spectrum, and K2=3 is amodified Bessel function. The distribution function iscalculated through the different filters of the wheel andintegrated to obtain the corresponding signal that it wouldyield on the image plate. The filters are sufficiently thin toneglect the effects of scattering for our range of energies.Both the experimental and theoretical data are normalizedso that the sum of the signals of the filters is equal to 1. Thedata are analyzed through a least squares fitting method byminimizing the number

PiðDi − TiÞ2, where Di and Ti

are, respectively, the measured and calculated normalizedsignals for each filter. One example is shown in Fig. 3(a)for a0 ¼ 3.05 and ne ¼ 1019 cm−3. Here, the best fit isobtained for Ec ¼ 10 keV. In our experimental conditions,the highest critical energy Ec ¼ 20 keV was measured fora0¼ 3.02 and ne ¼ 1.3 × 1019 cm−3. By differentiating thesignal obtained in the iron-chromium Ross pair (filters 6and 5, see image of Fig. 1), we can deduce the x-ray photonyield Nx at 6.5&0.5keV. At constant electron density ne¼1.3×1019cm−3, it goes from Nx¼3×108photons=eVSr fora0 ¼ 1.44 to Nx ¼ 1.45 109 photons=eVSr for a0 ¼ 3.02.A sharp stainless-steel edge placed 22 cm from the sourcecasts a clear shadow on the first image plate detector,indicating that for energies below 30 keV, the main sourceof x rays originates at the gas jet, consistent with betatronemission. We do not expect any significant hard x-ray

bremsstrahlung emission from the very underdense plasma.The measured 1=e2 source diameter has an upper valueof 35 μm.To quantify the x-ray spectrum at photon energies

between 10 and 500 keV, we use the stacked image platespectrometer. In addition to the betatron spectrumdescribed by Eq. (1), we assume an additional high-energyphoton background so that the total number of photons perunit energy on axis is:

dNx

dE∝

1

E

!EEc

"2

K22=3½E=Ec" þ A exp½−E=ET "; ð2Þ

where ET is the temperature of the exponentially decayingbremsstrahlung spectrum and A its amplitude relative tothe betatron spectrum. We propagate Eq. (2) through thedifferent materials of the experiment and through thecalibrated stacked image plate spectrometer [23,26].The number RT;i ¼ PT;0=PT;i is calculated, where PT;0 isthe total theoretical yield in the first plate (plate C0 in Fig. 1)and PT;i the total theoretical yield in subsequent plates fori ¼ 1∶7. These values are compared to the experimentalresults RE;i ¼ PE;0=PE;i to minimize the residue

PiðRE;i −

RT;iÞ2 by varying the parameters ET and A. The betatroncritical photon energy is set at Ec ¼ 10 keV in agreementwith the Ross pair filters measurements. The best fit [inset ofFig. 3(a) with the experimental data] is obtained for ET ¼200 keV andA ¼ 0.00014. The residue is higher by a factorof 10 if we fit using only betatron or bremsstrahlungdistributions separately. We deduce that the total x-ray yieldobserved in our experiment and shown in Fig. 3(b) is acombination of betatron radiation (dominant up to 40 keV)and bremsstrahlung (dominant above 40 keV). The brems-strahlung, inevitable whenever relativistic electrons areproduced, is likely due to lower energy (<500 keV) elec-trons being strongly deflected by the magnet onto the wallsof the target chamber.To explain the observed betatron x-ray spectra, we

performed 2D PIC simulations with OSIRIS for a varietyof conditions [27]. We illustrate the salient observationsfrom one simulation that uses an a0 ¼ 3, τ ¼ 0.7 ps, λ0 ¼1.053 μm laser pulse focused to a spot size of 15 μm(1=e2 intensity radius) into a 200 μm density up ramp. Thepulse duration and a0 were chosen to match the exper-imental values, and the spot size matches the value obtainedfrom the Gaussian fit (1=e2 intensity radius) of themeasured spot. The pulse then propagates through a3 mm-long fully ionized helium plasma of constant electrondensity ne ¼ 1 × 1019 cm−3. The simulation utilizes amoving window with box dimensions of 500 μm in thelongitudinal (laser propagation) direction and 150 μm inthe transverse direction. The corresponding resolutions are,respectively, 60 and 7.2 cells per λ0. To calculate betatronx-ray emission in these conditions, we select 750 randomelectrons in energy to match the overall spectrum[Fig. 4(c)]. The simulation is run again while also tracking

1 5 10 50 100

105

106

107

108

109

1010

X ray Energy keV

Phot

ons

eVSr

2 4 6 8 10

0.05

0.10

0.15

0.20

0.25

Filter numberX

ray

inte

nsity

norm

aliz

ed

Betatron Ec=10 keV

Bremsstrahlung T = 200 keV

1 2 3 4 5 6 70.4

0.5

0.6

0.7

0.8

0.9

Plate

Rat

ioPl

ate

Plat

e0 (a) (b)

FIG. 3. (a) Normalized x-ray yield through filters of Fig. 1 (reddots) for a0 ¼ 3.05 and ne ¼ 1019 cm−3 and critical energy fitscalculated with Eq. (1), with Ec ¼ 5 keV, 10 keV, and 15 keV(solid, dashed, and dotted lines). Inset: stacked image plate dataRE;i (red dots) and fit RT;i for a photon distribution [Eq. (2)] withEc ¼ 10 keV, A ¼ 0.000 14, and T ¼ 200 keV. (b) Deducedbetatron and bremsstrahlung spectra (see text for details).

PRL 118, 134801 (2017) P HY S I CA L R EV I EW LE T T ER Sweek ending

31 MARCH 2017

134801-3



28

10 20 30 40 50

0.02

0.05

0.10

0.20

0.50

X-ray energy @keVD

X-

ray

inte

nsity@ar

b.un

itD Ec = 10 keV

2 4 6 8 10 12

0.05

0.10

0.15

Filter

Yie

ld@N

orm

aliz

edD


d2IdEdΩ

∝!EEc

"2

K22=3½E=Ec"; ð1Þ






dNx

dE∝

1

E

!EEc

"2



PiðRE;i −



1 5 10 50 100

105

106

107

108

109

1010

X ray Energy keV

Phot

ons

eVSr

2 4 6 8 10

0.05

0.10

0.15

0.20

0.25

Filter numberX

ray

inte

nsity

norm

aliz

ed

Betatron Ec=10 keV


1 2 3 4 5 6 70.4

0.5

0.6

0.7

0.8

0.9

Plate

Rat

ioPl

ate

Plat

e0 (a) (b)



31 MARCH 2017

134801-3

X =



29

Ec = 5 keV

10 20 30 40 50

0.02

0.05

0.10

0.20

0.50

X-ray energy @keVD

X-

ray

inte

nsity@ar

b.un

itD

2 4 6 8 10 12

0.05

0.10

0.15

0.20

Filter

Yie

ld@N

orm

aliz

edD


d2IdEdΩ

∝!EEc

"2

K22=3½E=Ec"; ð1Þ






dNx

dE∝

1

E

!EEc

"2



PiðRE;i −



1 5 10 50 100

105

106

107

108

109

1010

X ray Energy keV

Phot

ons

eVSr

2 4 6 8 10

0.05

0.10

0.15

0.20

0.25

Filter numberX

ray

inte

nsity

norm

aliz

ed

Betatron Ec=10 keV


1 2 3 4 5 6 70.4

0.5

0.6

0.7

0.8

0.9

Plate

Rat

ioPl

ate

Plat

e0 (a) (b)



31 MARCH 2017

134801-3

X =

P.M. King et al, RSI (2019)



30

2 4 6 8 10 12

0.05

0.10

0.15

0.20

Filter

Yie

ld@N

orm

aliz

edD

10 20 30 40 50

0.02

0.05

0.10

0.20

0.50

X-ray energy @keVD

X-

ray

inte

nsity@ar

b.un

itD Ec = 10 keV

X =

Best fit for Ec = 10 keV +/- 2 keV (least squares fit) – 109 photons/eV/Sr


Betatron x-rays have critical energies of 10-40 keV

31F. Albert et. al, Phys. Rev. Lett. (2017)

Measured/calculated x-ray spectrum

Ec = 40 keV

Ec = 10 keV

Noise level

Betatron - ExperimentPIC simulation


Optimized betatron radiation produces the most photons for energies <40 keV

32

Betatron, Ec = 10 keV

ne = 1.5 x 1019 cm-3

Elaser = 150 Ja0 ~ 3


Nozzle

𝒇 𝑬 ~ 𝑬𝑬𝒄

𝟐𝑲𝟐/𝟑𝟐 𝑬

𝑬𝒄

F. Albert et. al, Phys. Rev. Lett. (2017)


Compton scattering allows for increased photon flux up to a few 100 keV

33N. Lemos et. al, Phys. Plasmas (2019)

Compton scattering𝒇 𝑬 ∝ 𝑬𝒙𝒑 − 𝑬

𝑻𝟏+𝑬𝒙𝒑 − 𝑬

𝑻𝟐T1 = 36 keV (Filter wheel)

ne = 4 x 1018 cm-3


Ross pairs


Nozzle

CH 100 µm


Compton scattering allows for increased photon flux up to a few 100 keV

34N. Lemos et. al, Phys. Plasmas (2019)

Compton scattering𝒇 𝑬 ∝ 𝑬𝒙𝒑 − 𝑬

𝑻𝟏+𝑬𝒙𝒑 − 𝑬

𝑻𝟐T2 = 78 keV (Step wedge)

ne = 4 x 1018 cm-3


Step wedge LaserGasne ~ 1019 cm-3

Nozzle

CH 100 µm


LWFA-driven bremsstrahlung produces the most photons at MeV energies

35N. Lemos et. al, PPCF (2018)

LWFA-driven bremsstrahlung𝒇(𝑬) ∝ 𝑬𝒙𝒑[−𝑬/𝑻]

T = 838 keV (Step wedge)Step wedge


Nozzle

ne = 4 x 1018 cm-3


W 100 µm


We can control the x-ray flux and energy by combining several processes

36


Nozzle


Nozzle


Nozzle

CH 100 µm

CH 100 µm

W250 µm

CH 100 µm

W50 µm

T 71 keV

T 331 keV

T 641 keV

N. Lemos et. al, In preparation


Spectral and flux tuning allows for optimized radiography applications

37N. Lemos et. al, In preparation

Half hohlraum – 30 µm Au

W ballr = 400 µm


Nozzle


Nozzle


NozzleW ball areal density

~ 0.7 g/cm2

T 71 keV

T 331 keV

T 641 keV


We can reproduce radiographs of test objects using the x-ray ray tracing code HADES

38I. Pagano et. al, In preparation

4 cmExperimental Radiograph Simulated Radiograph Comparison

ExperimentHADES


Nozzle

Ta250 µm


SM-LWFA driven x-ray source shows 1.4x higher radiography SNR for the same conditions

39I. Pagano et. al, In preparation

Areal density = 7.6 g/cm2


Nozzle

Ta1 mm

Laser

Nozzle

Ta1 mm


Understanding electron-ion equilibration in warm dense matter


Betatron x-ray source development at LCLS-MEC

41

cτ

lp=c/ωp~ 1/ne1/2

Blowoutregime


Application: detection of nonthermal meltingin SiO2

42

Non thermal transition

non-equilibriumstructure A

energy deposition

Thermalization

0 10-100fs 10-100ps

non-equilibriumstructure B

thermalizedstructure B


Absorption spectroscopy of SiO2 a the O K-edge (535 eV)

43

Si 2pSi 2s

Si 1s

O 1s

O 2s

Conduction band

Valence band

9 eV

535 eV

1848.6 eV

SiO2 300 K

510 520 530 540 550 560

0

0.2

0.4

0.6

0.8

1

Photon Energy @eVDAbs

orba

nce

O K-edge


No absorption of x-ray probe photons belowO K-edge energy

44

Si 2pSi 2s

Si 1s

O 1s

O 2s

Conduction band

Valence band

9 eV

535 eV

1848.6 eV

SiO2 300 K

510 520 530 540 550 560

0

0.2

0.4

0.6

0.8

1


orba

nceBetatron photon

<535 eV

O K-edge


Sharp transition corresponds to strong absorption of x-ray photons for energies above the O K-edge

45

Si 2pSi 2s

Si 1s

O 1s

O 2s

Conduction band

Valence band

9 eV

535 eV

1848.6 eV

SiO2 300 K

510 520 530 540 550 560

0

0.2

0.4

0.6

0.8

1


orba

nceBetatron photon

>535 eV

O K-edge


Multiphoton absorption causes electrons to cross the bandgap and leave vacancies in the valence band

46

Si 2pSi 2s

Si 1s

O 1s

O 2s

Conduction band

Valence band

9 eV

535 eV

1848.6 eV

SiO2 10,000 KHeatingOptical laser1015 W/cm2


1s-valence band transitions are now authorized: strong absorption peak 9 eV below the edge

47

Si 2pSi 2s

Si 1s

O 1s

O 2s

Conduction band

9 eV

535 eV

1848.6 eV

SiO2 10,000 K

510 520 530 540 550 560

0

0.2

0.4

0.6

0.8

1


orba

nce

9 eV

O K-edge

Betatron photon

Valence band

HeatingOptical laser1015 W/cm2


Defect states also allow absorption within bandgap upon heating, K-edge is broadened and red shifted

48

Conduction band

Si 2pSi 2s

Si 1s

O 1s

O 2s

<9 eV

535 eV

1848.6 eV

SiO2 10,000 K O K-edge

Betatron photon

Valence band

510 520 530 540 550 560

0

0.2

0.4

0.6

0.8

1


orba

nce

ColdWarm



Defect states also allow absorption within bandgap upon heating, K-edge is broadened and red shifted

49

Conduction band

Si 2pSi 2s

Si 1s

O 1s

O 2s

<9 eV

535 eV

1848.6 eV

SiO2 10,000 K O K-edge

Betatron photon

Valence band

510 520 530 540 550 560

0

0.2

0.4

0.6

0.8

1


orba

nce

ColdWarm



We have demonstrated the use of betatron x-rays as a tool for absorption spectroscopy

50

MeV

70

90

150

Electronspectrometer

Ellipsoidal mirrorX-ray CCD200 nm SiO2

Energy (eV)

700450100 µm

0

0.4

0.8

1.2

1.6

450 500 550 600 650

Cold XANES spectra registeredduring the delay scan [smooth 7 eV]

(normalized over [550-575 eV])

Mean am-SiO2 (11 spectra = 330 shots)Mean cr-SiO2 (2 spectra = 60 shots)

Abso

rban

ce

Energy (eV)

Not the same slopefor cr-SiO2(higher density ?)(substrate ?)

These averageswill be resp. usedto estimate thepre-edge integral

450 550500 600 650

0

Abso

rban

ce

0.4

0.8

1.2

1.6

X-ray energy (eV)

Amorphous SiO2 330 shotsCrystalline SiO2 60 shots

Cold absorption spectra

→B


We have demonstrated the use of betatron x-rays as a tool for absorption spectroscopy

51

MeV

70

90

150



Energy (eV)

700450100 µm

Cold/warm absorption spectra

Pump1015 W/cm2

0

0.4

0.8

1.2

1.6

450 500 550 600 650

Cold / Hot XANES spectra averageover all the delay scan [smooth 7 eV]

(normalized over [550-575 eV])am-SiO2 sample only

Mean cold 330 shotsMean hot 330 shots

Abso

rban

ce

Energy (eV)

Integration rangeof each hot - cold avg= [500 - 545] eV

Alternatively= [500 - 535] eV

Pre-edge

450X-ray energy (eV)

550500 600 650

0

Abso

rban

ce

0.4

0.8

1.2

1.6 Mean cold SiO2 330 shotsMean hot SiO2 330 shots

→B


We have demonstrated betatron x-rays absorption spectroscopy with sub ps resolution

52

MeV

70

90

150



Energy (eV)

700450100 µm

Cold/warm absorption spectra

Pump1015 W/cm2

→B

3. Time-resolved XANES (4/5)!

• Pre-edge level extracted from integration of “hot” – “cold” averaged"– Pre-edge time evolution versus delay gives an upper limit for temporal resolution"

! Just considering date on am-SiO2 … but the run #568 seems to be aberrant"! Without it, the best fit gives a temporal resolution = 0.48 ± 0.13 ps rms"

-2

0

2

4

6

8

10

-2 0 2 4 6 8 10 12

Delay scan on am-SiO2 sampleXANES spectra normalized

then substracted to avg coldand integrated over ≠ spectral range

Cold over [500-545] eVCold over [500-535] eVHot over [500-545] eVHot over [500-535] eV

Inte

grat

ed s

pect

ra h

ot -

avg

cold

(eV)

Delay (ps)

run #568

Error estimated from cold data :0.8 rms over [500-545]0.6 rms over [500-535]=> reported on error bars

y = 6*(0.5+0.5*erf((m0-m1)/m...ErrorValue

0.21992-0.0030769m1 0.403530.96311m2

NA13.571ChisqNA0.8086R

Erf fit gives :- "t0" = 0.0 ± 0.2 ps (good sync.)- temp. res. = 0.96 ± 0.40 ps (0.67 ± 0.28 ps rms) (1.60 ± 0.66 ps FWHM)

-2

0

2

4

6

8

10

-2 0 2 4 6 8 10 12

Delay scan on am-SiO2 sampleXANES spectra normalized

then substracted to avg coldand integrated over ≠ spectral range

Cold Integr. [500-545eV] (eV)Cold Integr. [500-535eV] (eV)Hot Integr. [500-545eV] (eV)Hot Integr. [500-535eV] (eV)

Inte

grat

ed s

pect

ra h

ot -

avg

cold

(eV)

Delay (ps)

y = 6*(0.5+0.5*erf((m0-m1)/m...ErrorValue

0.09553-0.017005m1 0.179090.68532m2

NA2.9108ChisqNA0.95702R

Without #568, erf fit gives :- "t0" = 0.0 ± 0.1 ps (good sync.)- temp. res. = 0.68 ± 0.18 ps (0.48 ± 0.13 ps rms) (1.13 ± 0.30 ps FWHM)

Delay (ps)In

tegr

ated

spec

tra

hot –

aver

age

cold

(eV)

Integrated 500 – 545 eVIntegrated 500 – 535 eV

-2 0 2 4 6 8 10 12

0

2

4

6

8

10

12


We still have a lot of ongoing exciting projects

53

New acceleration mechanisms

-2 mm

keV-MeV sources and applications

Platform development on larger HEDS lasers

LaserNetUS experiments using betatron source3

75 100 125 150 175 200-100

-50

0

50

100

[mra

d]

0

1

2

pC/M

eV/m

rad

75 100 125 150 175 200-100

-50

0

50

100

[mra

d]

-1

0

1

pC/M

eV/m

rad

75 100 125 150 175 200-100

-50

0

50

100

[mra

d]

0

1

2

pC/M

eV/m

rad

75 100 125 150 175 200-100

-50

0

50

100

[mra

d]

-1

0

1

pC/M

eV/m

rad

100

101

102

100

101

10-1

(A)

(B)

×10

×20

Laser Polarization

Laser Polarizationx

y

y

x

FIG. 2: Measured electron energy spectra for a plasmawith electron density ne = 3⇥ 1017 cm�3 dispersed(A) perpendicular and (B) parallel to the linear laserpolarization direction. The contrast is adjusted and aline-out along the dashed line is plotted (solid line) toemphasize the forking feature in the dispersed electronprofile. The signal above 100 MeV is multiplied by 10and 20 in (A) and (B), respectively, to emphasize the

spectrum at high energy. The dashed black lineindicates the acceptance aperture of the magnet. Notethat (A) and (B) were taken on two di↵erent shots with

similar laser energies.

ing structure is seen, and the FWHM beam divergenceis instead 21 mrad [red curve in Fig. 2(B)] at the sameenergy. The elliptical beam profile of the electrons shownin Fig. 1(C) gives the overall full-angle divergence at half-maximum charge of the electron beam in the two planesas 47 and 27 mrad in the x and y directions, respectively,consistent with the dispersed spectra.The forking structure gives clear evidence that elec-

trons above 60 MeV are gaining some or most of theirenergy by the DLA process [22, 27]. Electrons acceler-ated mainly through DLA generally exhibit higher energyand greater divergence along the laser polarization di-rection compared to electrons accelerated predominantlythrough SM-LWFA. This larger divergence is evident inthe forking structure seen only for high-energy electronsdispersed perpendicular to the laser polarization, as inFig. 2(A).To discern the relative contribution of the various

mechanisms to the final energy of the electrons, we simu-lated the full acceleration process with particle trackingusing the quasi-3D algorithm of the Osiris PIC simula-tion framework [31, 32] for laser and plasma parameterssimilar to those used in the experiment (see supplemen-tal material). This algorithm allows us to unambiguouslydetermine the work done by the longitudinal field of theplasma wave (Ez,m=0), as well as the transverse (Ex,m=1)and longitudinal (Ez,m=1) fields of the laser. This hasallowed us to more correctly determine the overall DLA

contribution. Here m = 0 and m = 1 refer to the cylin-drical modes corresponding predominantly to the wakeand the laser, respectively.

FIG. 3: Snapshot of the electron density profile after4.64 mm of propagation (left to right) through theplasma; z and x are the longitudinal and transversedirections, respectively. Also shown are the m = 0

longitudinal electric field (SM-LWFA) overlaid in redand the m = 1 transverse electric field envelope (DLA)in green. The dashed green line shows the vacuum laserfield envelope at the focus. The tracked electrons, with

x positions given by their radial distance (onlyhalf-space is shown), indicate where in space eachacceleration mechanism is dominant. The charge

density has been integrated in ✓.

Figure 3 shows the envelope of the transverse laser fieldEx,m=1 (green), the plasma density (blue) and the on-axis longitudinal electric field of the plasma wave Ez,m=0

(red) 4.64 mm into the plasma. Clear modulation of boththe laser envelope and plasma waves is evident. How-ever, a hydrodynamic channel is not fully formed (notshown) within the laser pulse; the ion density remainsabove 0.9n0 across the first bucket (potential well) andabove 0.7n0 where the laser field is of significant ampli-tude, where n0 is the initial plasma density. The wave-length of the plasma wave is increased for the first threebuckets by strong beam loading, but for subsequent buck-ets it is close to 2⇡c/!p. The plasma electrons trappedby the plasma wave are color-coded to indicate whichacceleration mechanism is at work (see subsequent para-graphs). The accelerated electrons group together in thelater plasma buckets, where they gain energy predomi-nantly by interacting with the longitudinal field of thewave associated with SM-LWFA. However, the electronstrapped in the front three buckets of the wake gain netenergy predominantly through the DLA process as theyinteract with the peak-intensity portion of the laser pulse.Relativistic self-focusing helps to maintain the peak in-tensity of the laser pulse (see dashed green line).To quantify the contribution of each acceleration mech-

anism (i.e., SM-LWFA and DLA), we use electron track-ing in Osiris to calculate the work done on each electronby the di↵erent spatial components of mode 0 (wake) and

• Role of Direct Laser Acceleration in Self-modulated LWFA*

• 3D OSIRIS PIC simulations confirm observation (UCLA collaboration)

*P.M. King et al, PRAB (2021)

• Betatron, Bremsstrahlung, Compton sources for radiography applications

• LaserNetUS experiment at Texas Petawatt on radiography

650 700 750Energy [eV]

0.35

0.4

0.45

0.5

0.55

0.6

0.65

0.7

0.7550 nm Fe - transmission

Run 213 coldRun 217 coldRun 222 coldCold averagecxro

LaserGas

Nozzle

LaserGas

Nozzle

LaserGas

Nozzle

CH + W (50 µm)

T 71 keV

T 331 keV

T 641 keV

CH + W (250 µm)

N. Lemos (in prep)2 new students: B. Pagano (UT Austin) and A. Aghedo (FAMU)

• 150 MeV, 700 nC beams at OMEGA EP*

• 150 MeV > µC at NIF- ARC• Development of new targets

and diagnostics

0

100

-100

[mrad]

W-NEPPS

Gas Tubee beam

• Study of warm dense iron with XANES

• Phase contrast imaging of laser-driven shocks in water

*J.L. Shaw et al, Sc. Rep (2021)

*M. Berboucha, E. Galtier et al**C. Kuranz et al


Conclusions and future work

54

§ We have demonstrated the production of novel x-ray sources from laser-plasma accelerators on several laser facilities

§ They are broadband (keV - MeV), ultrafast (fs - ps), small source size (µm), collimated (mrad), synchronized with drive laser

§ They enable new applications— Study of ultrafast non-thermal melting in SiO2— Radiography of dense objects— Phase contrast imaging of laser-driven shocks and hydrodynamic instabilities— Study of opacity in HED matter

§ Future work and challenges— Improving sources stability and flux— Applications from proof-of-principle to practical— LWFA sources as probes for HED science experiments, single shot and rep-rate

N. Lemos et al, PPCF 58 034108 (2016)F. Albert et al, PRL 118 134801 (2017)F. Albert et al, POP 25 056706 (2018)N. Lemos et al, PPCF 60, 054008 (2018) P. M. King et. al, Rev. Sc. Instr. 90, 033503 (2019)F. Albert et al, Nuclear Fusion, 59, 032003 (2019)P. M. King et al, PRAB, 24, 011302 (2021)J.L. Shaw et al, Sci. Rep, 11 7498 (2021)N. Lemos et. al, PRL (in preparation)


Access to this type of research will be facilitated by networks

55

Founded 2018

x-ray light sources driven by laser plasma accelerators

Documents