Matthias Uiberacker

Brijuni Conference, 31. August 2006, Brijuni Islands, Croatia

Sampling electron dynamics in atoms in real time with sub-femtosecond resolution

Prof. Ferenc Krausz

Dept. f. Physik, Ludwig-Maximilians-Universität München, Germany

Max-Planck-Institut für QuantenoptikGarching, Germany

Institut für PhotonikTechnische Universität WienWien, Austria

attosecond physics aims at gaining insight into the motion of electrons on atomic scales

&

of electronic motion in atoms, molecules, solids and plasmas

&

of electronic motion in atoms, molecules, solids and plasmas

0.00.0 0.10.1 0.20.2 0.30.3 0.40.4 [nanometers][nanometers]

e-e-

e-e-

direct controldirect control real-time observation real-time observation

time (s)

10-18 10-15 10-12

10-15

10-12

10-9

space (m)

molecules & solids

atoms

mic

ros

cop

y, d

iffr

acti

on

the microcosm: imaging in space and time

nuclearstructure &dynamics

attosecondmetrology

attosecondmetrology

femtosecondmetrology

atomsatoms

molecules

electronselectrons in

in

attophysicsattophysics

high–speed photography of microscopic processes:

time–resolved pump-probe spectroscopy

a sampling system with sub-fs resolution

what to do?

utilizing pump/probe techniques

ultrashort visible laser pulses are close to the wavecycle-limit of pulse duration (1-3fs).

..but, can be used to produce shorter (sub-fs) xuv pulses (high harmonic generation)

efficiency for 2 sub-fs xuv pulses is not sufficient yet

pump pulse and probe pulse need to be short enough to freeze the motion of electrons

using the electric field of laser pulses for probing with sub-fs resolution

= a(t)cos(ωLt + φ)

a(t)a(t)

T0 2.5 fsT0 2.5 fs

T0 /4 625 as (@ 0 0.75 µm)T0 /4 625 as (@ 0 0.75 µm)

E(t)

requires stabilization and control of carrier-envelope phase in combination with a weak sub-fs xuv pulse ->

pump/probe measurements with sub-fs resolution!

waveform-controlled few-cycle light opens the door to steering & capturing electrons on an attosecond timescale

A. Baltuska, T. Udem, M. Uiberacker, M. Hentschel, E. Goulielmakis, C. Gohle, R. Holzwarth, V. Yakovlev, A. Scrinzi, T. W. Hänsch, F. Krausz, Nature 421, 611 (2003)

ultrabroadband dispersioncontrol with chirped multilayers

R. Szipöcs, K. Ferencz, Ch. Spielmann, F. KrauszOpt. Lett. 19, 201 (1994)

stabilization of the frequency comb of a mode-locked laser

T. W. Hänsch et al., 1997, 1999H. Telle et al. Appl. Phys. B 69, 327 (1999)D. Jones et al., Science 288, 635 (2000)

Ne gasNe gas

few-femtosecond,few-cyclelaser pulse

λL 750 nmTp = 5 - 6 fsWp = 0.2 - 0.4 mJ

few-femtosecond,few-cyclelaser pulse

λL 750 nmTp = 5 - 6 fsWp = 0.2 - 0.4 mJ

Drescher et al., Science 291, 1923 (2001)

Hentschel et al., Nature 414, 509 (2001)

Kienberger et al., Science 297, 1144 (2002)

phase-stabilizedelectric fieldphase-stabilizedelectric field

xuv/x-ray radiation from strongly driven atoms

3D-solution of the Schrödinger equation for hydrogen: Armin Scrinzi (TU Vienna)

steering bound electrons with controlled light fields: the generation of a sub-femtosecond pulse

x(t)

EELL((tt))

EELL((tt))

EELL((tt))

EELL((tt))

xuv/x-ray radiation from strongly driven atoms

6060 7070 8080 9090 100100 11011000

10001000

Inte

ns

ity

[c

ou

nts

]In

ten

sit

y [

co

un

ts]

Photon energy [eV]Photon energy [eV]

xuv-filterxuv-filter

sub-femtosecond xuv/x-ray pulse generation

Ne gasNe gas

few-femtosecond,few-cyclelaser pulse

λL 750 nmTp = 5 - 7 fsWp = 0.3 - 0.5 mJ

few-femtosecond,few-cyclelaser pulse

λL 750 nmTp = 5 - 7 fsWp = 0.3 - 0.5 mJ

6060 7070 8080 9090 100100 11011000

10001000

Inte

ns

ity

[c

ou

nts

]In

ten

sit

y [

co

un

ts]

Photon energy [eV]Photon energy [eV]

time-of-lightelectron or ionspectrometer

time-of-lightelectron or ionspectrometer

atomicgas

atomicgas

near- diffraction-limitedxuv/soft-x-ray beamnear- diffraction-limitedxuv/soft-x-ray beam

xuv and laser pulse act on target particlesxuv and laser pulse

act on target particles

Drescher et al., Science 291, 1923 (2001)

Hentschel et al., Nature 414, 509 (2001)

Kienberger et al., Science 297, 1144 (2002)

attosecond pulse generation and detection

by irradiation with xuv-light pulses by irradiation with xuv-light pulses

core-levelphoto-

emission

+1

core-levelphoto-

emission

+1

Augerdecay

+2

Augerdecay

+2

photo-emission

&shake up

+1

photo-emission

&shake up

+1

Augerdecay

&shake up

+2

Augerdecay

&shake up

+2

kin.

ene

rgy

kin.

ene

rgy

00

core orbitalcore orbital

occupiedvalenceoccupiedvalence

unocc.valenceunocc.valence

triggering electronic transitions inside atoms

valencephoto-

emission

+1

valencephoto-

emission

+1final

charge statefinal

charge state

bind

. en

ergy

bind

. en

ergy

Exuv (t)

subsequentsubsequentcore-holecore-hole

by means of strong-field-induced free-free transitions: streaking by means of strong-field-induced free-free transitions: streaking

kin.

ene

rgy

kin.

ene

rgy

bind

. en

ergy

bind

. en

ergy core-level

photo-emission

+1

core-levelphoto-

emission

+1

Augerdecay

+2

Augerdecay

+2

photo-emission

&shake up

+1

photo-emission

&shake up

+1

Augerdecay

&shake up

+2

Augerdecay

&shake up

+2

00

occupiedvalenceoccupiedvalence

unocc.valenceunocc.valence

EL(t)

probing electronic transitions inside atoms

Exuv (t)

attosecondstreak camera

core orbitalcore orbital

valencephoto-

emission

+1

valencephoto-

emission

+1final

charge statefinal

charge state

resolution ~ 100 femtoseconds

electron-optical streak camera

D. J. Bradley et al., Opt. Commun. 2, 391 (1971)M. Y. Schelev et al., Appl. Phys. Lett. 18, 354 (1971)

eAL(t) eAL(t)

mapping time to momentum

electron release timeelectron release time

Δp(t7)Δp(t7)

Δp(t6)Δp(t6)

Δp(t5)Δp(t5)

Δp(t3)Δp(t3)

Δp(t2)Δp(t2)

Δp(t1)Δp(t1)

00

momentumchange alongthe EL vector

momentumchange alongthe EL vector

-500 as-500 as 00 500 as500 as

laser electric fieldlaser electric field

t7t7t1t1 t2t2 t3t3 t4t4 t5t5 t6t6

optical-field-driven streak camera optical-field-driven streak camera J. Itatani et al., Phys. Rev. Lett. 88, 173903 (2002)M. Kitzler et al., Phys. Rev. Lett. 88, 173904 (2002)

timetime

time-dependent

electron emission

time-dependent

electron emission

electronmomentumdistribution

electronmomentumdistribution

attosecond streak camera: complete measurement of a few-cycle light wave and a sub-fs xuv pulse

ele

ctr

on

co

un

ts /

bin

ele

ctr

on

co

un

ts /

bin

∆W(t) eAL(t)∆W(t) eAL(t)

6565

7575

8585

ele

ctr

on

kin

eti

c

en

erg

y [

eV

]

ele

ctr

on

kin

eti

c

en

erg

y [

eV

]

delay [fs]delay [fs]00 44-4-4 88

xuv pulsexuv

pulse

time [fs]time [fs]

inte

ns

ity

[a

rb.

u.]

inte

ns

ity

[a

rb.

u.]

00

11

ins

tan

tan

eo

us

e

ne

rgy

sh

ift

[eV

]in

sta

nta

ne

ou

s

en

erg

y s

hif

t [e

V]

9292

9393

9494

9595

9696

9797

-0.4-0.4 -0.2-0.2 0.00.0 0.20.2 -0.4-0.4

xuv = 250as

measurementsimulation

6060 7070 8080 9090 100100

final electron energy, Wf [eV]final electron energy, Wf [eV]

6060 7070 8080 9090 10010000

100100

6060 7070 8080 9090 100100

In the absenceof the laser field

single 250-attosecondxuv pulse @ 95 eV single 250-attosecondxuv pulse @ 95 eV

-6

6

3

0

-3

lig

ht

ele

ctr

ic f

ield

, E

L(t

) [1

07

V/c

m]

LL

( )( )

dA tE t

dt

-20-20

-10-10

00

1010

2020

v

ec

tor

po

ten

tia

l, -

AL(t

)

ve

cto

r p

ote

nti

al,

-A

L(t

) [f

sM

V/c

m]

[fs

MV

/cm

]

E. Goulielmakis et al., Science 305, (2004)

by means of strong-field-induced free-free transitions: streaking by means of strong-field-induced free-free transitions: streaking

kin.

ene

rgy

kin.

ene

rgy

bind

. en

ergy

bind

. en

ergy core-level

photo-emission

+1

core-levelphoto-

emission

+1

Augerdecay

+2

Augerdecay

+2

photo-emission

&shake up

+1

photo-emission

&shake up

+1

Augerdecay

&shake up

+2

Augerdecay

&shake up

+2

00

occupiedvalenceoccupiedvalence

unocc.valenceunocc.valence

EL(t)

probing electronic transitions inside atoms

attosecondstreaking spectroscopy

EXUV (t) core orbitalcore orbital

valencephoto-

emission

+1

valencephoto-

emission

+1final

charge statefinal

charge state

streaked electron spectra following core-hole excitation in krypton by a sub-fs xuv pulse

M. Drescher et al., Nature 419, 803 (2002)

tracing core-hole decay directly in time lifetime of M-shell (3d) vacancy in Krypton: h = 7.91 fs tracing core-hole decay directly in time lifetime of M-shell (3d) vacancy in Krypton: h = 7.91 fs

by means of strong-field-induced bound-free transitions: tunneling by means of strong-field-induced bound-free transitions: tunneling

kin.

ene

rgy

kin.

ene

rgy

bind

. en

ergy

bind

. en

ergy core-level

photo-emission

+1

core-levelphoto-

emission

+1

Augerdecay

+2

Augerdecay

+2

photo-emission

&shake up

+1

photo-emission

&shake up

+1

Augerdecay

&shake up

+2

Augerdecay

&shake up

+2

00

occupiedvalenceoccupiedvalence

unocc.valenceunocc.valence

EL(t)

probing electronic transitions inside atoms

EXUV (t) core orbitalcore orbital

valencephoto-

emission

+1

valencephoto-

emission

+1final

charge statefinal

charge state

attosecondtunneling spectroscopy

multiphoton versus tunneling ionization: the Keldysh theory

Keldysh, L.V., Sov. Phys. JETP 20, 1307 (1965)

multiphoton ionization: due to absorption of many photons

tunneling ionization: due to suppression of Coulomb potential

electron wave-packetsemitted within a time t shorter than the half-period of the laser

effective Coulomb barrier

tunneling

electron emission within a time mp shorter than the pulse duration

L

bL 2

eE

mW

Keldyshparameter:

> 1 < 1

time evolution of probing – ionization with a few-cycle pulse

atomic/molecular

target

bind

. en

ergy

bind

. en

ergy

occupiedvalenceoccupiedvalence

unocc.valenceunocc.valence

core orbitalcore orbital

kin.

ene

rgy

kin.

ene

rgy

level 1level 2

low E0high E0

level 2level 1

level 1

attosecond tunneling spectroscopy

first experiments in neon and xenon

testing the sub-fs resolution with neon

XUV pulse

NIR pulse

A. Kikas et al., J. of Electr. Spectr. and Rel. Phen. 77, 241-266 (1996).

95.2 % 4.8 %

92.8 % 7.2 %2.4 %

steps are visible -> sub-fs resolution is valid(signal/noise has to be improved)

2.4 %

Ne2+ versus delay time

..to be published

xenon energy levels – illustration of dynamics

XUV pulse

NIR pulse

8.9 %

78 %

3.3 %

9.7 %

Auger2Auger1

NIR-DI

NIR-I

Xe4+ versus time

Xe3+ versus time

A1 = 6.0 0.7 fs

A2 = 30.8 1.4 fs

A1

F. Penent et al., Phys. Rev. Lett. 95, 083002 (2005).

resolving electron dynamics in xenon

0 50 100 150 200 250

0

10

20

30

40

Delay time

Xe4+

Double exponential fit

Cou

nts

A1 = 6.0 0.7 fs

A2 = 30.8 1.4 fs

this experiment: time-integral frequency-resolved

experiments:

..to be published

F. Penent, Phys. Rev. Lett. 95, 083002 (2005).

A1 (4d3/2) = 6.3 0.2 fsA1 (4d5/2) = 5.9 0.2 fs

A2 > 23 fs

xuv optics & atomic spectroscopy:Th. Uphues, U. Kleineberg, U. Heinzmann

Univ. Bielefeld, Germany M. Drescher

Univ. Hamburg, DESY, Germanylight phase control:

Ch. Gole, R. Holzwarth, T. Udem, T. W. Hänsch Univ. Munich - MPQ Garching, Germany& measurement:

G. Paulus, H. Walther A&M Univ. Texas, USA / MPQ Garching Ch. Lemell, J. Burgdörfer, A. Scrinzi Vienna Univ. Techn., Austriametrology: P. B. Corkum, M. Yu. Ivanov NRC Canada, Ottawa, Canadamolecular spectroscopy: M. Kling, M. Vrakking AMOLF, Amsterdam, Netherlands M. Lezius, K. Kompa MPQ Garching, Germany

xuv optics & atomic spectroscopy:Th. Uphues, U. Kleineberg, U. Heinzmann

Univ. Bielefeld, Germany M. Drescher

Univ. Hamburg, DESY, Germanylight phase control:

Ch. Gole, R. Holzwarth, T. Udem, T. W. Hänsch Univ. Munich - MPQ Garching, Germany& measurement:

G. Paulus, H. Walther A&M Univ. Texas, USA / MPQ Garching Ch. Lemell, J. Burgdörfer, A. Scrinzi Vienna Univ. Techn., Austriametrology: P. B. Corkum, M. Yu. Ivanov NRC Canada, Ottawa, Canadamolecular spectroscopy: M. Kling, M. Vrakking AMOLF, Amsterdam, Netherlands M. Lezius, K. Kompa MPQ Garching, Germany

postdoctoral:

A. Apolonski

A. Baltuska

A. Cavalieri

T. Fuji

E. Goulielmakis

R. Kienberger

J. Seres

M. Uiberacker

V. Yakovlev

PhD candidates:

N. Ishii

T. Metzger

J. Rauschenberger

M. Schultze

C. Theisset

A. Verhoef

graphics: Barbara Ferus

postdoctoral:

A. Apolonski

A. Baltuska

A. Cavalieri

T. Fuji

E. Goulielmakis

R. Kienberger

J. Seres

M. Uiberacker

V. Yakovlev

PhD candidates:

N. Ishii

T. Metzger

J. Rauschenberger

M. Schultze

C. Theisset

A. Verhoef

graphics: Barbara Ferus

coworkers & collaborators

End

1 ionization of D2

2 recollisional excitation

3 formation of a coherent superposition (1sg

+,2pu+)

state in D2+

1 ionization of D2

2 recollisional excitation

3 formation of a coherent superposition (1sg

+,2pu+)

state in D2+

2pσu+2pσu+

may attosecond control of electronic motion in chemical bonds affect the outcome of molecular dynamics?

D+

rightright+D D

D+

leftleft + D D

YES: direction of emission of D+ is controlled by light waveform YES: direction of emission of D+ is controlled by light waveform

D2+D2+

1sσg+1sσg+

D2D2

e-e-

RR

1

23

time [fs]time [fs]

EL(t

)E

L(t

)

asy

mm

etry

lef

t/ri

gh

tas

ym

met

ry l

eft/

rig

ht

-5-5 00 55 1010 1515

0.50.5

00

-0.5-0.5

phase-controlledfew-cycle wave

M. Kling et al., Science 312, 246 (2006)

time

right

left

00

M. Kling et al., Science 312, 246 (2006)

time

right

left

M. Kling et al., Science 312, 246 (2006)