37
Angle Angle - - resolved photoemission resolved photoemission spectroscopy (ARPES) spectroscopy (ARPES) Experimental study of the electronic structure Experimental study of the electronic structure of strongly of strongly - - correlated electron systems correlated electron systems Andr Andr é é s s Felipe Felipe Santander Santander - - Syro Syro Centre de Centre de Spectrom Spectrom é é trie trie Nucl Nucl é é aire aire et de et de Spectrom Spectrom é é trie trie de Masse de Masse Universit Universit é é Paris Paris - - Sud Sud OXITRONICS OXITRONICS OXITRONICS

Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

  • Upload
    others

  • View
    46

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

AngleAngle--resolved photoemission resolved photoemission spectroscopy (ARPES)spectroscopy (ARPES)

Experimental study of the electronic structure Experimental study of the electronic structure of stronglyof strongly--correlated electron systemscorrelated electron systems

AndrAndrééss Felipe Felipe SantanderSantander--SyroSyro

Centre de Centre de SpectromSpectroméétrietrie NuclNuclééaireaireet de et de SpectromSpectroméétrietrie de Massede Masse

UniversitUniversitéé ParisParis--SudSud

OXITRONICSOXITRONICSOXITRONICS

Page 2: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

C. Bareille, F. Fortuna (CSNSM – Orsay)F. Boariu, M. Klein, A. Nüber, F. Reinert (Uni‐Würzburg)P. Lejay (Institut Néel – Grenoble)O. Copie, M. Bibes, N. Reyren, A. Barthélémy (UMPhy – Thales)T. Kondo (Ames‐Lab)F. Bertran, A. Nicolaou, A. Taleb‐Ibrahimi, P. Le Fèvre (SOLEIL)Y. Apertet, P. Lecoeur (IEF ‐ Orsay) S. Pailhès (CEA ‐ Saclay)X. G. Qiu (Chinese Academy of Sciences)G. Herranz (ICMAB ‐ Spain)R. Weht (CNEA ‐ Argentina)M. J. Rozenberg, M. Gabay (LPS – Orsay)

THANKS!THANKS!

Page 3: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

1D

2D

3D

Many properties of solids are determined by electrons within a narrow energy slice (~kBT) around EF (dc conductivity, magnetism, superconductivity…)

Fermi Surface

Condensed (crystalline) matter in a nutshell

Room temperature: kBT = 25 meVAdapted from A. Damascelli’s Exciting-2003 lecture and E. Rotenberg’s lecture

Allowed electronic statesRepeated-zone scheme

-kF kF

Page 4: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

ARPES principle

F. Reinert and S. Hüfner, New Journal of Physics 7, 97 (2005)

θsin2 kinmE=||kh

Bkin EWhE −−= ν

x

y

z

e-θ

ϕ

Detector

Sample

Kinetic energy analyzer

, Av

νh

Simple example:Cu(111) surface-state

kx

ky

E

Best for 2D or 1D systems

Page 5: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

ARPES with He-lampWürzburg University (DE)

(F. Reinert’s team)

ARPES with synchrotron radiationSynchrotron Radiation Center

University of Wisconsin – Madison (USA)

Page 6: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

a = 4.129 Åc = 9.58 Å

Data: E. Hassinger et al., PRB B 77, 115117 (2008).Figure: A. Villaume et al., PRB 78, 012504 (2008).

Hidden‐orderLong‐moment AFM

Example of state-of-the art ARPES: The hidden-order transition in URu2i2

[001]

X

P

Y

N

Z

[100]

[010]

ΔΓ Σ

Λ

Σ

Page 7: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

0.40.0-0.4k|| (Å

-1)0.40.0-0.4

k|| (Å-1)

0.40.0-0.4k|| (Å

-1)0.40.0

k|| (Å-1)

-80

-60

-40

-20

0

EB (m

eV)

Low

HighT = 26 K T = 18 K = THO T = 13 K T = 10 K

ARPES view across the transition

Spectra along (110)

AFSS, M. Klein, F. L. Boariu, et al., Nature Physics 5, 637 ‐ 641 (2009)

Page 8: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Photoemission in the N-electron system: theoretical snapshot

Photoemission intensity given by Fermi’s golden rule:

( ) ( )νδπε hEEHI Ni

Nf

Ni

(N)int

Nf −−ΨΨ∝

22,h

k

NiΨ N-electron ground state (energy Ei

N) of unperturbed Hamiltonian

NfΨ Excited state (energy Ef

N) of unperturbed Hamiltonian: N-1 electrons in the solid + 1 free electron of energy ε and momentum k

Matrix element:• k-conservation • Symmetries

Energy conservation

Page 9: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

• true in the UV for electrons in the bulk• not necessarily true at the surface

indirect transitions interfering with bulk transitions asymmetric line-shapes

[ ] pAA(r)ppA(r) ⋅≈⋅+⋅≈−

mce

mceH e

int 21

Light-matter interaction: perturbation theory + dipole approximation

Dipole approximation: A constant over atomic dimensions

[ ] 0, ≈⋅∇= AAp hi

Page 10: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Electron escape to the surface

ARPES is a surface technique: one needs clean surfaces + work under ultra-high vacuum (better than 10-9 Torr).

λesc (Ekin)= Electron escape depth(average distance traveled by electron without inelastic scattering)

Page 11: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

SUDDEN APPROXIMATIONThe ejected electron should be “fast enough” to neglect its

interaction with the hole left behind

Ekin must be large hν must be large

Final state = Plane wave in the vacuum

Furthermore, one has to make sure that the photoemission process itself does not modify

the electronic structure of the material…

Page 12: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Photoemission: many-body effectselectron-electron interaction

Adapted from R. Claessen’s Cargese-2005 lectures

“loss” of kinetic photoelectron energy due to excitation energy stored in the remaining interacting system

electron-phonon interaction

• Shift in energy of the main photoemission line• Broadening of main line (life-time of excited

interacting system)• Transfer of spectral-weight to higher binding

energies (excitations of the remaining interacting system)

Page 13: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Spectra analysis

1 , =≡ hhωBE

( ) ( ) ( ) ( )ωωνω ,,,, 0 kAkk AfII =

( ) ( )( )[ ] ( )[ ]22 ,,

,1,ωωεω

ωπ

ωkk

kkk Σ′′+Σ′−−

Σ′′−=A

Σ ′′Σ′ Energy renormalization

Lifetime of dressed e-Many-body physics

A(k,ω) = Probability of adding or removing one electron at (k,ω)

Page 14: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Interaction effects on ARPES spectra

( ) ( ) ( ) ( )ωωνω ,,,, 0 kAkk AfII =

( ) ( )( )[ ] ( )[ ]22 ,,

,1,ωωεω

ωπ

ωkk

kkk Σ′′+Σ′−−

Σ′′−=A

A(k,ω) = Probability of adding or removing one electron at (k,ω)

Binding energy EB

Σ’’(k,ω)(life-time)

Σ’(k,ω)(renormalization)

EF

εk(ω) I(k,ω)

EF

εk

k

Page 15: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Many-body physics – Effects of the interactions on the band structure:

Example of surface states of Mo(110)

T. Valla et al., PRL 83, 2085 (1999)

Mo(110) band along Mo(110) band along ΓΓNNT = 70 KT = 70 K

Γee ~ ω2

Γe-ph Eliashberg

Γe-imp = const

Page 16: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Polarization = LVSlits = V

Polarization = LHSlits = V

Polarization = LHSlits = H

xy

z

dxy orbital

ARPES: Effects of orbital symmetries

( )22

,N

iNffiM Ψ⋅Ψ∝ pAk

even

Page 17: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

-0.4 0.0 0.4 ky (Å

-1)

-0.3

-0.2

-0.1

0.0

E - E

F (e

V)

10K LH

-0.4 0.0 0.4 ky (Å

-1)

-0.3

-0.2

-0.1

0.0

E - E

F (e

V)

10K LV

x

y

z

x

y

z

Polar = LVSlits = V

Polar = LHSlits = V

Example: subbands with orbital order at the surface of SrTiO3

Page 18: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Summary of ARPES techniqueWhen it works…

ARPES is a powerful technique for the study of the electronic structure of complex systems

Rapidly developing…

Detailed band structures and Fermi surfacesFermi velocity and effective massGapsMany-body effects

Spin-resolved ARPESTime-resolved ARPESMicro-ARPESLaser-ARPES using UV/X-ray lasers (HHG in noble gases)Ultra-high resolution ARPES using TOF detector

Page 19: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Superficially metallicA 2D electron gas with universal subbands at the 

surface of SrTiO3

Superficially metallicSuperficially metallicA 2D electron gasA 2D electron gas with universal with universal subbandssubbands at the at the 

surface of SrTiOsurface of SrTiO33

Nature Nature 469469, 189, 189‐‐193 (2011).193 (2011).

Page 20: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Preliminaries: A 2D metal that spontaneously forms at the interface between

two insulators

Page 21: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

HighHigh‐‐mobility 2DEG at the LaAlOmobility 2DEG at the LaAlO33/SrTiO/SrTiO33 interfaceinterface

A. Ohtomo and H. Y. Hwang, Nature 427, 423 (2004).

LaAlO3: Band insulatorΔ = 5.6 eV

SrTiO3: Band insulator(quantum paraelectric)

Δ = 3.2 eV

(

(

(

(

)-

)+

)0

)0

Page 22: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

KT2D

( ) νzcC VVT −∝

Electric field control of the LAO/STO interface ground stateElectric field control of the LAO/STO interface ground state

A. D. Caviglia et al., Nature 456, 624 (2008).

SC

AFM

NFL

HeavyFL

Heavy fermion physicsT

Pressure

SCAFM

FL

NFL

PG

Cuprate superconductivityT

Doping

Page 23: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Why so important?Why so important?2DEGs (e.g. at semiconductor interfaces) are the first step in the 

design of useful micro‐electronic devices, like FETs.

SrTiO3 is the preferred substrate to grow thin films of transition‐metal oxides. 

Transition‐metal oxides have properties that surpass those of semiconductors: high‐temperature superconductivity, colossal magnetoresistance, high thermoelectric power with good conductivity, multiferroic behavior, …

Possibility of combining two or more of these functionalities in a single device through (multi)‐thin films of transition metal oxides. A 2DEG at the interface provides a conduction channel.

Controllable systems to study fundamental aspects of the physics of transition‐metal oxides, which are often “strongly‐correlated electron systems”.

Page 24: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

HeteropolarHeteropolar interfaces based on STO:interfaces based on STO:OPEN QUESTIONSOPEN QUESTIONS

What is the origin of the metallic electron gas?

What is its thickness?

What is its dimensionality? (2D, 3D; i.e., can dgas < λF)

What is (are) the mechanisms responsible for (super)conductivity and high mobility?  

Page 25: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

M. Baslestic et al., Nature Materials 2008, doi: 10.1038/nmat2223

Extension of 2DEG at LAO/STO interfaceExtension of 2DEG at LAO/STO interface

Page 26: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Metallic 2DEG can also be induced by Metallic 2DEG can also be induced by electrostatic doping of the pure STO surfaceelectrostatic doping of the pure STO surface

K. Ueno et al., Nature Materials 7, 855 (2008)

Page 27: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

• Oxygen vacancies

• Polar catastrophe (build‐up of electric field due to the polar LaO layer)

• Electronic reconstruction (charge leakage between LaTiO3 and SrTiO3)

• Atomic diffusion

• Surface reconstruction to avoid polar catastrophe

Scenarios invokedScenarios invoked

Page 28: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Surprise Surprise …… surprisesurprise!!

A 2DEG spontaneously forms at the vacuum-cleaved surface of SrTiO3

Subband structure, carrier density, confinement size similar to those found in other STO-related interfaces

For this 2DEG, the mechanism is well understood door wide open to generate novel 2DEGs in other oxides!

Page 29: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

n3D = 1018 cm-3 1019 cm-3 1020 cm-3 1021 cm-3

2π/a

2π/a

n2D = 1014 cm-2

ky

Δ

Γ

Spin-orbit + tetra + ortho + …

E1(dxy)

E1(dxz / yz)

E2(dxz / yz)

-V0 ky

E

EF

Γ

z

-V0

n=1n=2

EF

eFL≈L

SrTiOSrTiO33: bulk : bulk vsvs 2D2D‐‐confined electronic structureconfined electronic structure

dxydyz

dxz

ky

E

EF

Γ

xy

z

light

heavy

AFSS et al, Nature 469, 189‐193 (2011).

Page 30: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

-0.4 0.0 0.4 ky (Å

-1)

-0.3

-0.2

-0.1

0.0

E - E

F (e

V)

n3D < 1013 cm-3

10K LH

-0.4 0.0 0.4 ky (Å

-1)

-0.3

-0.2

-0.1

0.0

E - E

F (e

V)

10K LV

Metallic 2DEG at the surface of insulating SrTiOMetallic 2DEG at the surface of insulating SrTiO33

x

y

z

x

y

z

Polar = LVSlits = V

Polar = LHSlits = V

AFSS et al, Nature 469, 189‐193 (2011)

Page 31: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

-0.4 0.0 0.4 ky (Å

-1)-0.4 0.0 0.4

ky (Å-1)

n3D < 1013 cm-3 n3D ~ 1018 cm-3 n3D ~ 1020 cm-3

20K 20K

-0.4 0.0 0.4 ky (Å

-1)

-0.3

-0.2

-0.1

0.0

E - E

F (e

V)

10K LV

SrTiOSrTiO33: universal electronic structure at the surface: universal electronic structure at the surface

AFSS et al, Nature 469, 189‐193 (2011)

Page 32: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 ky (Å

-1)

-0.3

-0.2

-0.1

0.0

E - E

F (e

V)

min

max

10K LH

BΓ102

10K LV

SrTiOSrTiO33: summary of : summary of subbandssubbandsfor the surface 2DEGfor the surface 2DEG

mlight = 0.7me mheavy ≈ 20me

AFSS et al, Nature 469, 189‐193 (2011)

Page 33: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

0π/a

2π/a

π/a

-π/a

0π/a

2π/a

-π/a

-2π/a

2π/a

3π/a

4π/a

π/a

2π/a

3π/a

4π/a

k z

k y

kx

min

max

-0.8

-0.4

0.0

0.4

0.8

ky (

Å-1

)

-2.0 -1.0 kx (Å

-1)

-π/a

π/a

Γ102 Γ112Fermi surface and Fermi surface and density of carriersdensity of carriers

n2D-Total = 1.9 – 2.4x1014 cm-2 ≈ 0.29 – 0.35 e/a2

22 2πF

DAn = AF = Fermi surface area

AFSS et al, Nature 469, 189‐193 (2011)

Page 34: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

ΔE = E2(dxz) – E1(dxz) = 120 meVmL

* = 0.7me mH* ≈ 20me

F ≈ 83 MV/m

E1(dxy) = -210 meV V0 ≈ 260 meV

eFL ≈ ΔE L ≈ 14.5 Å ≈ 4 u.c.eFLmax = V0 Lmax ≈ 31 Å ≈ 8 u.c.

4 u.c. ≤ L ≤ 8 u.c.

Confinement potential and size of 2DEG: Confinement potential and size of 2DEG: 

z

n = 1

n = 2

E

E = 0

-V0

eFL≈

Lmax

( ) eFzVzV +−= 0

3/23/12

0 41

23

2 ⎥⎦

⎤⎢⎣

⎡⎟⎠⎞

⎜⎝⎛ −⎟

⎠⎞

⎜⎝⎛

⎟⎟⎠

⎞⎜⎜⎝

⎛+−= ∗ eFn

mVE

zn

πh

T. Ando, A. B. Fowler, and F. Stern, Rev. Mod. Phys. 54, 437-672 (1982).

Page 35: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

( )

( )BFA

F

FdFne F

D

+=

′′= ∫1

2 0 02

ε

εε

• K. Ueno et al., Nature Mater. 7, 855 (2008). • R. C Neville, B. Hoeneisen, and C. A. Mead. J. Appl. Phys. 43, 2124 (1972).

F [V/m]A(4K) = 4.097 x 10-5

B(4K) = 4.907 x 10-10 m/V

n2D ≈ 0.25 e/a2

Carrier density from confinement potential:Carrier density from confinement potential:

Page 36: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

Y. Aiura et al., Surface Science 515, 61-74 (2002)

Valence band bending due to O‐vacancies at the vacuum‐fractured surface of STO

Valence band bending due to OValence band bending due to O‐‐vacancies at the vacancies at the vacuumvacuum‐‐fractured surface of STOfractured surface of STO

Page 37: Angle-resolved photoemission spectroscopy (ARPES)ngsces2011.cesga.es/wp-content/uploads/2011/08/santander.pdfAngle-resolved photoemission spectroscopy (ARPES) Experimental study of

ConclusionsConclusions

Observation of subband structure with orbital ordering at the bare surface of STO.

Universal metallic 2DEG. Exists even at the surface of non-doped, bulk-insulating STO!

Characteristics of this 2DEG in quantitative agreement with 2DEGs in STO-based interfaces/heterostructures.

Origin of the observed 2DEG: oxygen vacancies.