Exciton dynamics and valley-contrasting properties in ... · Exciton dynamics and...

Exciton dynamics and valley-contrasting properties in heterostructures based on

atomically-thin semiconductors

Stéphane BERCIAUD IPCMS - CNRS - Université de Strasbourg - France

[email protected]

Workshop on Chiral Optical Modes, Aussois, Nov. 27, 2018

Tuesday March 19, 2018

Guillaume Froehlicher PhD student (2013-16)

Etienne Lorchat PhD student (2015-)

Acknowledgements

Stefano Azzini, Thibault Chervy Thomas Ebbesen, Cyriaque Genet

Cédric Robert, Delphine Lagarde, Xavier Marie

Takashi Taniguchi Kenji Watanabe

ISIS

Ajayan, Kim, Banerjee - Physics Today (2016)

2H-TMD (semiconductors) M= Mo, W X= S, Se, Te

M

X

X

C

Graphene (semimetal)

M

X

X

C1.7 1.8 1.9 2.0 2.1 2.2

0

1

360 370 380 390 400 410 420 430

0

1

Lum

inescence inte

nsity (

a.

u.)

Photon Energy (eV)

1L

3L

Bulk

E1

2g

Bulk MoS2

24.7 cm-1

3L MoS2

23.2 cm-1

Ram

an Inte

nsity (

a.

u.)

Raman Shift (cm-1)

1L MoS2

18.7 cm-1

A1g

10µm

20 µm 20 µm

a) Optical image b) Luminescence Image

MonolayerMoS2

BulkMoS2

c) d)

Entering “Flatland”

Direct bandgap emission

Helicity-dependent valley addressability

Ajayan, Kim, Banerjee - Physics Today (2016)

2H-TMD (semiconductors) M= Mo, W X= S, Se, Te

M

X

X

C

Graphene (semimetal)

M

X

X

C

Entering “Flatland”

Direct bandgap emission


0.9 1.0 1.1 1.2

5L

6L

7L

bulk

Ph

oto

lum

. (u

.a.)

Energy (eV)

1L

2L

3L

4L

MoTe2

PRB 2016

laser

Gr

BN

TMD

BN

Gr

WSe2

s± s±

interlayer charge and exciton transport

optoelectronics valleytronics nanomechanics

< 1 nm

|1, 𝒌

Massicotte et al., Nat Nano 2016 Mak et al., Science 2014 De Alba et al., Nat. Nano 2016

room T excitonic effects spin-valley locked properties

2D Materials: a unique toolkit for fundamental and applied physics

Nanophotonics, single photon emission

materials engineering (phase, anisotropy, strain)

electron-phonon coupling

K K’

MoX2 s+ s-

K K’

WX2 s+ s-

Spin-valley locking in monolayer TMDs

Broken inversion symmetry + strong spin orbit coupling

E

k K K’

WX2 s+ s-

K K’

MoX2 s+ s-

Valley-polarized excitons...and their coherent superpositions

G. Wang et al. RMP 2018 Z. Ye et al., Nat. Phys. 2016

s-

s+

MoS2

T=14KExc: s+

1.0

0.5

0.0

r

1.70 1.80 1.90 2.00 1.80 1.90 2.00Photon energy (eV) Photon energy (eV)

Ph

oto

lum

(arb

.)

c

2.0

MoS2 T=14 K, exc: s+, 1.96 eV

s+

s+

s-

MoS2

T=14K Exc: s-

1.0 0.5 0.0

r

1.70 1.80 1.90 2.00 1.80 1.90 2.00 Photon energy (eV) Photon energy (eV)

P

ho

tolu

m (

arb

.)

Fragile Valley Contrasts V

alle

y Po

lari

zati

on

(a.

u.)

c MoS2

s+

s-

MoS2

T=14KExc: s-

1.0

0.5

0.0

r

1.70 1.80 1.90 2.00 1.80 1.90 2.00Photon energy (eV) Photon energy (eV)

Ph

oto

lum

(arb

.)

K.F. Mak et al., Nat. Nano 2012

• r up to ~100 % at low T (MoS2) • g up to 60 % at low T (MoS2)

• Valley contrasts are lost at room T How to protect/tailor valley contrasts? C. R. Zhu et al., PRB 2014

r (g) degrees of circular (linear) polarization

𝜌 =𝜌0

1 + 2𝜏𝑋 𝜏𝑆

Outline

+_ Graphene/TMD heterostructures as a 2D optoelectronic building block

Room temperature valley polarization and coherence

in TMD/Graphene heterostructures

Room temperature Chiral coupling of valley excitons with spin-momentum locked

surface plasmons

Surface plasmons (SPs)

Strong interaction with light

+_< 1 nm

Semiconducting 2H-TMDs

M = Mo, W X = S, Se, Te

MX2

2D Matter meets 2D Light



ISIS (Strasbourg) z

x

y

kz

ETM

SE

2

FIG. 2. (a) White light (WL) microscope image of the sample

and photoluminescence (PL) image of the same region under480nm excitat ion. (b) ...

tering) Under our condit ions we reach a chiral separat ion

efficiency of 15%, limit due to the finite spin-valley con-

t rast in WS2. This value could potent ially exceed 50%

by using curent ly available preparat ion schemes for the

TMD material (cite records of CPL).

Paragraph 4: Our system consists of a hole array plas-

monic resonator covered by a monolayer of WS2 (see

Fig. 2(a)). We begin by sput tering 200nm of gold on

a glass subst rate coated by a 5nm thick Chromium ad-

hesion layer. The plasmonic hole array is milled through

the metallic layers using a Carl Zeiss Auriga FIB sys-

tem. A period ⇤ = 480nm is chosen to generate a plas-

monic mode resonant with the absorpt ion energy of the

A-exciton of WS2 at 2.01eV. We then spin-coat on the

metal a 5nm thick PMMA (polymethyl methacrylate)

layer, on top of which the mechanically exfoliated WS2

flake is t ransferred.

Paragraph 5: An SEM image of the bare plasmonic

grat ing is shown in Fig. 2(b). It consists of rotat -

ing rectangular nano-apertures (160nm x 90nm) with

an orbital period equal to 6 t imes the grat ing period.

Similar arrangements of anisotropic apertures have al-

ready been demonst rated to allow for spin-dependent

surface plasmon launching (cite Shuang Zhang, Has-

man, Capasso,...). In our case, the local orientat ion

of the anisot ropy axis of the nanoaperture is given by

φ = ⇡x/ N ⇤, where N is the number of grat ing pe-

riods needed to obtain a ⇡ -rotat ion. Therefore, light

scat tered from the array acquires a geometric phase of

φg = 2⇡σx/ N ⇤, where σ = ± 1 correspond respect ively

to right and left -handed circular polarizat ions [18,19]??.

Di↵erent iat ing thisphasewith respect to thex coordinate

gives a rotat ion-induced momentum kg = 2σ⇡ / N ⇤which

adds up to the usual grat ing law, giving the plamonic

resonant condit ion: k SP P = k / / ± nG x ± mG y + kgx, as

sketched in Fig. 2(c). The rout ing act ion of such a spin-

FIG. 3. Angle-resolved reflect ivity spect rum of the samplein circular (a) right and (b) left polarizat ion. Angle-resolved

SHG spect rum for circular (c) left and (d) right excitat ion at1.01eV.

orbit grat ing under monochromat ic excitat ion is demon-

st rated in real space using a leakage microscope [? ]. We

clearly observe the unidirect ional launching of plasmonic

waves according to the spin of the incident light beam.

... Simplify this paragraph...

Paragraph 6: The measured angle-resolved white light

reflectometry of the whole structure (Fig. 2(a)) is shown

in Fig. 3 for left and right analyzed circular polariza-

t ions. In each case, two st rongly dispersing branches are

observed, featuring a clear ant i-crossing behaviour. This

demonst rates that the system is in the st rong coupling

regime, with the two branches corresponding to upper

and lower polaritonic states. This is further confirmed

by a coupled oscillators fit to the dispersions, giving a

Rabi split t ing of 40meV. The two dispersion diagrams

show a clear mirror symmetry with respect to the normal

incident (kx = 0) axis for the two opposite handedness,

clearly demonstrat ing the capability of our st ructure as

a spin-determined polaritons launcher. ...Cooperativity

factor C = 0.8...

Paragraph 7: Whereas this represents an interest ing

plat form for general polaritonic systems, here we invest i-

gate the interplay between this opt ical chirality and the

valley-cont rast ing chirality of theWS2 monolayer. A first

demonst rat ion of such an interplay is found in the res-

onant second harmonic response of the coupled system.

Indeed, WS2 valley t ransit ions have been shown to give

a high spin cont rast in circularly polarized second har-

monic generat ion (SHG), with select ion rules opposite to

those of linear absorpt ion [? ]. The angle resolved reso-

nant SHG for left and right circularly polarized excitat ion

(Fig. 3(c) and (d) respect ively) display lobes that corre-

. . . . x

Helicity-dependent directional SP launching



+_< 1 nm



MX2




ISIS (Strasbourg) z

x

y

kz

ETM

SE


-kSP

+kSP



+_< 1 nm



MX2

Tailored control of the valley pseudospin using TMD-SP architectures





z

x

y

kz

ETM

SE

Room temperature chiral coupling between valley excitons and helicity-momentum locked surface plasmons

T. Chervy*, S. Azzini*, et al., ACS Photonics 5, 1281 2018 (Arxiv 1701.07972) See also K. Kuipers’ “intermezzo”

-kSP

+kSP


Coll: T. Chervy, S. Azzini, J.A. Hutchinson, T.W. Ebbesen, C. Genet (ISIS, Uni. Strasbourg) Y. Gorodetzki (Ariel, IL), S. Wang (Eindhoven, NL)

Chiral coupling: 1L-TMD exciton with spin-orbit SP mode

Angle-resolved WL absorption (1-R ~ A) – CPL analysis

Anticrossing

Strong-coupling FOM

Resonant condition TMD excitons / SP modes

2

FIG. 2. (a) White light (WL) microscope image of the sample

and photoluminescence (PL) image of the same region under480nm excitat ion. (b) ...

tering) Under our condit ions we reach a chiral separat ion

efficiency of 15%, limit due to the finite spin-valley con-

t rast in WS2. This value could potent ially exceed 50%

by using curent ly available preparat ion schemes for the

TMD material (cite records of CPL).

Paragraph 4: Our system consists of a hole array plas-

monic resonator covered by a monolayer of WS2 (see

Fig. 2(a)). We begin by sput tering 200nm of gold on

a glass subst rate coated by a 5nm thick Chromium ad-

hesion layer. The plasmonic hole array is milled through

the metallic layers using a Carl Zeiss Auriga FIB sys-

tem. A period ⇤ = 480nm is chosen to generate a plas-

monic mode resonant with the absorpt ion energy of the

A-exciton of WS2 at 2.01eV. We then spin-coat on the

metal a 5nm thick PMMA (polymethyl methacrylate)

layer, on top of which the mechanically exfoliated WS2

flake is t ransferred.

Paragraph 5: An SEM image of the bare plasmonic

grat ing is shown in Fig. 2(b). It consists of rotat -

ing rectangular nano-apertures (160nm x 90nm) with

an orbital period equal to 6 t imes the grat ing period.

Similar arrangements of anisot ropic apertures have al-

ready been demonstrated to allow for spin-dependent

surface plasmon launching (cite Shuang Zhang, Has-

man, Capasso,...). In our case, the local orientat ion

of the anisot ropy axis of the nanoaperture is given by

φ = ⇡ x/ N ⇤, where N is the number of grat ing pe-

riods needed to obtain a ⇡ -rotat ion. Therefore, light

scat tered from the array acquires a geometric phase of

φg = 2⇡σx/ N ⇤, where σ = ± 1 correspond respect ively

to right and left -handed circular polarizat ions [18,19]??.

Di↵erent iat ing thisphasewith respect to thex coordinate

gives a rotat ion-induced momentum kg = 2σ⇡ / N ⇤which

adds up to the usual grat ing law, giving the plamonic

resonant condit ion: k SP P = k / / ± nG x ± mG y + kgx, as

sketched in Fig. 2(c). The rout ing act ion of such a spin-

FIG. 3. Angle-resolved reflect ivity spect rum of the samplein circular (a) right and (b) left polarizat ion. Angle-resolved

SHG spect rum for circular (c) left and (d) right excitat ion at1.01eV.

orbit grat ing under monochromat ic excitat ion is demon-

st rated in real space using a leakage microscope [? ]. We

clearly observe the unidirect ional launching of plasmonic

waves according to the spin of the incident light beam.

... Simpli fy this paragraph...

Paragraph 6: The measured angle-resolved white light

reflectometry of the whole st ructure (Fig. 2(a)) is shown

in Fig. 3 for left and right analyzed circular polariza-

t ions. In each case, two st rongly dispersing branches are

observed, featuring a clear ant i-crossing behaviour. This

demonst rates that the system is in the st rong coupling

regime, with the two branches corresponding to upper

and lower polaritonic states. This is further confirmed

by a coupled oscillators fit to the dispersions, giving a

Rabi split t ing of 40meV. The two dispersion diagrams

show a clear mirror symmet ry with respect to the normal

incident (kx = 0) axis for the two opposite handedness,

clearly demonst rat ing the capability of our st ructure as

a spin-determined polaritons launcher. ...Cooperativity

factor C = 0.8...

Paragraph 7: Whereas this represents an interest ing

plat form for general polaritonic systems, here we invest i-

gate the interplay between this opt ical chirality and the

valley-cont rast ing chirality of theWS2 monolayer. A first

demonst rat ion of such an interplay is found in the res-

onant second harmonic response of the coupled system.

Indeed, WS2 valley t ransit ions have been shown to give

a high spin cont rast in circularly polarized second har-

monic generat ion (SHG), with select ion rules opposite to

those of linear absorpt ion [? ]. The angle resolved reso-

nant SHG for left and right circularly polarized excitat ion

(Fig. 3(c) and (d) respect ively) display lobes that corre-

Grating period L=480 nm

T. Chervy*, S. Azzini* et al., ACS Photonics 2018

Room temperature chiralitons

K

K’

“Chiralitons”

• Plasmon sorter efficiency: 15% Net chiraliton flow: 6 % 40 %

Recent related results: Spin-momentum locking: TU Delft (Science 2018), UT Austin (arXiv:1801.06543) Valley polaritons: U. Sheffield, Nat. Photon. 2018, Northwestern U. Nat Photon 2018, Würzburg PRB2017

• PL polarization tomography on a linear basis

Measured as the m11 Mueller matrix element

• Chiralitonic valley coherence

Coherent superposition of counterpropagating chiralitons

No valley contrasts in bare TMD Cavity protection mechanism? Designing new 2D building blocks for chiral optics.

(𝑆1|𝑇𝑀 − 𝑆1|𝑇𝐸)/2 = 𝑚11~ 5 − 8 %

Outlook:

Outline





surface plasmons

• Graphene: 2D semi-metallic channel Quasi-transparent (~2% absorption per layer) High carrier mobility and large carrier density Gate-tunable properties

• TMD: 2D semiconducting channel Strong light-matter interaction Broadband absorption and tunable emission

De Fazio et al., ACS Nano 2016

0

1

2

3

1.5 2.0 2.50.0

0.5

1.0

PL

In

ten

sity (

arb

. u

nits)

R

/R

Energy (eV)

1L MoSe2

300 K

A B

Gr/TMD heterostructures: why the interest?

R, PL, Raman Tunable excitation

Electrical Readout

•Gr-TMD heterostructures:

Strong interlayer coupling (J. He et al., Nat Comm. 2014)

Photogating/photodetection (W. Zhang et al., Sci. Rep. 2014)

ps-range photoresponse (M. Massicotte et al., Nat Nano 2016)


VB

TMD Thickness (nm)

Res

po

nse

rat

e (s

-1)

Res

po

nse

tim

e

M. Massicotte et al., Nat Nano 2016 ICFO




Low efficiency with monolayer TMD Energy transfer?

+

_

+

_

_

++

_

_

++

_

+

_

+

_

TMD Graphene

(a) Interlayer Electron Transfer

(b) Interlayer Hole Transfer

(c) Dexter-type Interlayer Energy Transfer

(d) Förster-type Interlayer Energy Transfer

Charge and energy transfer mechanisms

Probed in the steady state (Raman) Balanced electron and hole transfer

Determine exciton dynamics Relative efficiencies?

Our Experimental Approach

Understanding near-field interactions (IET vs ICT)

Implications for optoelectronic devices

PL (TMD) Raman (Gr, TMD)

Tunable excitation

+_

Gr

TMD

PL (TMD) Raman (Gr, TMD)

Tunable excitation

+_

Electrical Readout Gr

TMD

< 1 nm

Photolum. Raman

Tunable excitation

+_< 1 nm

Gr

TMD

Atomic Force Microscopy

MoSe2

Gr

Gr/MoSe2

Smooth Gr/MoSe2 domains

AFM

0.0 0.5 1.0 1.5

0

1

2

3

He

igth

(n

m)

Distance (µm)

2

0.0 0.5 1.0 1.5

0

1

2

3

0.6

He

igth

(n

m)

Distance (µm)

2

1.3 1.4 1.5 1.6 1.7 1.8 1.9

0.000

0.003

PL inte

nsity (

arb

. units)

Photon energy (eV)

1.3 1.4 1.5 1.6 1.7 1.8 1.9

0

1

PL inte

nsity (

arb

. units)

Photon energy (eV)1.3 1.4 1.5 1.6 1.7 1.8 1.9

0

1

PL

in

ten

sity (

arb

. u

nits)

Photon energy (eV)

1.3 1.4 1.5 1.6 1.7 1.8 1.9

0

1

PL inte

nsity (

arb

. units)

Photon energy (eV)

5 µm

Bare MoSe2

2 nm 0.6 nm

Decoupled Gr/MoSe2 Coupled Gr/MoSe2

Strong PL Quenching ~ 300

Photoluminescence mapping

1020 1021 1022 1023

0.01

0.1

1

I PL/

ph

oto

ns (

arb

units)

photons (cm-2 s-1)

Exciton: dynamics: PL vs photons

• PL saturation on bare and decoupled MoSe2

Bare MoSe2

Decoupled Gr/MoSe2

Normalized PL: IPL

photons

• PL saturation on bare and decoupled MoSe2: Exciton-Exciton Annihilation (EEA)

EEA: N. Kumar et al., PRB 89, 125427 (2014), S. Mouri et al., PRB 90, 155449 (2014), D. Sun et al., Nano Lett. 14, 5625 (2014)

S. Mouri et al., PRB 2014

1020 1021 1022 1023

0.01

0.1

1

I PL/

pho

ton

s (

arb

units)

photons (cm-2 s-1)

Bare MoSe2

Decoupled Gr/MoSe2

Exciton dynamics: PL vs photons

1020

1021

1022

1023

0.01

0.1

1

I PL/

photo

ns (

arb

units)

photons

(cm-2 s

-1)


• No PL saturation on Gr/MoSe2 and IPL(B) ~ IPL(A)


Exciton-Exciton Annihilation (EEA)

Bare MoSe2

Decoupled Gr/MoSe2

Coupled Gr/MoSe2

1020

1021

1022

1023

0.01

0.1

1

I PL/

ph

oto

ns (

arb

units)

photons

(cm-2 s

-1)

• No PL saturation on Gr/MoSe2

Bare MoSe2

Decoupled Gr/MoSe2

Coupled Gr/MoSe2

0 2 4 6 8 10 1210

-3

10-2

10-1

100

Bare MoSe2

Decoupled SLG/MoSe2

Coupled SLG/MoSe2

PL C

ounts

(arb

. units)

Time Delay (ns)

IRF



Drastic reduction of the excitonic lifetime (< 1 ps)

Fast interlayer Energy Transfer?


G. Froehlicher, E. Lorchat, S. Berciaud, Phys. Rev. X 8, 011007 (2018) Fast TRPL at INSA Toulouse

1020

1021

1022

1023

0.01

0.1

1

I PL/

ph

oto

ns (

arb

units)

photons

(cm-2 s

-1)


Bare MoSe2

Decoupled Gr/MoSe2

Coupled Gr/MoSe2



Drastic reduction of the excitonic lifetime (< 1 ps)

Fast interlayer Energy Transfer?


G. Froehlicher, E. Lorchat, S. Berciaud, Phys. Rev. X 8, 011007 (2018)

-10 0 10 20 30 40

10-2

10-1

100 Data

Fit

IRF

PL c

ounts

(arb

. units)

Time Delay (ps)

Fast TRPL at INSA Toulouse

Outline





surface plasmons

SOC

+K −K

Back to valley contrasts

Photonic state Excitonic state

𝜎± |±1

1

2 𝜎+ ± |𝜎−

1

2 +1 ± |−1

ΓTMD ≪ ΓS ⇒ 0% valley polarization

SOC

+K −K

Photonic state Excitonic state

𝜎± |±1

1

2 𝜎+ ± |𝜎−

1

2 +1 ± |−1

ΓTMD ≪ ΓS ⇒ 0 % valley polarization

ΓTMD/Gr ≪ ΓS ⇒ finite valley polarization

Back to valley contrasts

Our experimental approach

BN-encapsulated WS2-Gr heterostructure

Mueller polarimetry setup

Mueller Polarimetry

m11,22 = valley coherence m33 = valley polarization

Degrees of linear and circular polarization

mi0: birefringence

m0i: dichroisim

Experimental “sanity check” : m11 = m22 and mij,i j= 0

Mueller Polarimetry in WS2/Gr

Laser @600 nm

Mapping valley contrasts

45% valley polarization, 30% valley coherence @Room Temperature

E.Lorchat, S.Azzini, T.Chervy et al., ArXiv:1804.06725 (doi: 10.1021/acsphotonics.8b01306)

2.07 2.06 2.05 2.042.07 2.06 2.05 2.04

(b)

*

s+s+

s+s-

PL In

ten

sity (

arb

. u

nits)

598 601 604 607

0.0

0.5

1.0(d)

*

m3

3

Wavelength (nm)

*

*

XX

XY

PL In

ten

sity (

arb

. u

nits)

(a)

598 601 604 607

0.0

0.5

1.0(c)

m1

1

Wavelength (nm)

Energy (eV)Energy (eV)

Record valley coherence up to 60 % (10 %) in BN-capped WS2/Gr (BN capped WS2)

Large valley contrasts in WS2/Gr (20 K)

Val

ley

Co

he

ren

ce

Val

ley

Pola

riza

tio

n

F. Cadiz et al. PRX 2017 (valley coherence of 50% in BN-capped MoS2)

E.Lorchat, S.Azzini, T.Chervy et al., ArXiv:1804.06725 (doi: 10.1021/acsphotonics.8b01306)

*Raman features

Conclusion and outlook

2D matter meets 2D light: many games to play at the interface between (chiral) nanophotonics, condensed matter physics and materials science Interfacing 2D materials with chiral plasmonic resonators Probing dark excitonic states using surface plasmons Electrical and electromechanical control of chiral coupling Charge and energy transfer mechanism(s) Intervalley scattering mechanisms in TMD/Gr vs bare TMD

Acknowledgements

Stefano Azzini Thibault Chervy

Yuri Gorodetzki (Ariel, IL) Shaojun Wang (Eindhoven, NL)

James Hutchinson Thomas Ebbesen Cyriaque Genet

Cédric Robert, Delphine Lagarde, Xavier Marie

Takashi Taniguchi Kenji Watanabe

ISIS

Funding:

Etienne Lorchat Guillaume Froehlicher Luis Parra-Lopez Florentin Fabre StNano Clean room

Supplementary Slides Net charge transfer in TMD-Gr heterostructures General slides on heterostructures and on energy transfer

How about photoinduced charge transfer?

Raman Scattering

Tunable excitation

+_

…study of net (and slow) eletron transfer from TMD to Graohene (cf PRX 2018). This effect is likely extrinsic (not observable if we replace SiO2 by hBN) but is of interest for photodetection/photogating (see next slide).


W. Zhang et al., Sci. Rep. 2014 KAUST




1200 1400 1600 2600 28000

1

2

3

4

5

(no) D mode

2L sus

1L sus

G mode

C

ou

nts

(a

rb.

un

its)

Raman Shift (cm-1)

2D mode

Raman spectroscopy: a quantitative probe of doping

0 5 10 15 20 25

0

5

10

15 electrons

holes

Sample 1

Sample 2

Sample 3

Sample 4

Sample 5

G (

cm

-1)

G (cm

-1)

0 5 10 15 20 25

-15

-10

-5

0

5

10

15

Sample 1

Sample 2

Sample 3

Sample 4

Sample 5

2

D -

0 2D(c

m-1)

G-

0

G(cm

-1)

strain

2.22

D

G

55.02

D

G

h+

2.02

D

G

e-

2D

Data: G. Froehicher & SB, PRB 2015. See also: A. Das et al., Nat. Nano 2008 S. Pisana et al., Nat Mater 2007, J. Yan et al., PRL 2007,…

1590 2640 2680 2720

0

1

2

3

4

52D

Ra

ma

n in

ten

sity (

arb

. u

nits)

Raman shift (cm-1)

G

1590 2640 2680 2720

0

1

2

3

Ra

ma

n in

ten

sity (

arb

. u

nits)

Raman shift (cm-1)

2DG

1590 2640 2680 2720

0

1

2

3

4

5

Ra

ma

n in

ten

sity (

arb

. u

nits)

Raman shift (cm-1)

G 2D

1590 2640 2680 2720

0

1

Ra

ma

n in

ten

sity (

arb

. u

nits)

Raman shift (cm-1)

5 µm Reference on SiO2

Bare MoSe2

Coupled Gr/MoSe2

Decoupled Gr/MoSe2

ph increases

Eex = 2.33 eV

Raman response vs photon flux (1)

1020

1021

1022

1023

456789

13141516171819

G (

cm

-1)

ph

(cm-2

s-1)

1020

1021

1022

1023

1584

1586

1588

1590

1592 SLG/SiO2

Decoupled SLG/MoSe2

Coupled SLG/MoSe2

G (

cm

-1)

ph

(cm-2

s-1)

G-mode frequency G-mode FWHM

Clear signatures of a net photoinduced charge transfer


Clear signatures of a net photoinduced charge transfer


1584 1586 1588 1590 15924

6

8

10

12

14

16

18

20

G (

cm

-1)

G (cm

-1)

increasing ph

Width-Frequency Correlation

Correlation between the 2D- and G-mode frequencies

• Comparison with Gr/MoS2

W. Zhang et al., Scientific Reports 4, 3826 (2014)

electrical measurements

𝑒− transfer

0 5 10 15 20 25

-15

-10

-5

0

5

10

15

Sample 1

Sample 2

Sample 3

Sample 4

Sample 5

2D

-

0 2D(c

m-1)

G-

0

G(cm

-1)

55.02

G

D

h+

2.02

G

D

e-

holes

electrons Evidence for TMD Gr electron transfer

1584 1586 1588 1590 1592

2674

2676

2692

2

D (

cm

-1)

G (cm-1)

0.11

increasing ph

𝑒− transfer

-300 -200 -100 0 100 200 3001582

1584

1586

1588

1590

1592

1594

G (

cm

-1)

EF (meV)

-300 -200 -100 0 100 200 300 400

-5.5 -2.4 -0.6 0.0 0.6 2.4 5.5 9.8

1584

1586

1588

1590

1592

1594

1596

1598

1600

1602

n (x1012

cm-2)

EF (meV)

wG (

cm

-1)

-2

0

2

4

6

8

10

12

14

DG

G (

cm

-1)

G. Froehlicher and S. Berciaud, PRB 91, 205413 (2015) G. Froehlicher, E. Lorchat, S. Berciaud, Phys. Rev. X 8, 011007 (2018)

Transfered electron density nG

Quantifying photoinduced doping

-300 -200 -100 0 100 200 300

4

6

8

10

12

14

16

G (

cm

-1)

EF (meV)

Evidence for net photoinduced electrpon transfer Extrinsic effect – slow dynamics

0 2 4 6 8 10 12

-200

-100

0

100

200

300

sample 1

sample 2

sample 3

EF (

me

V)

ph (x1023 cm-2 s-1)

1020 1021 1022 1023 1024

100

150

200

250

300

0.6

1.4

2.4

3.8

5.5

n (

x1

01

2 c

m-2

)

S5

EG

rF

(m

ev)

ph (cm-2 s-1)

S4

0 1 2 3 4 5 6100

150

200

250

300

0.6

1.4

2.4

3.8

5.5

nG (

x10

12 c

m-2

)

Air

EF (

meV

)

ph (x1023 cm-2 s-1)

Reproducibility and environmental effects

0 1 2 3 4 5 6100

150

200

250

300

0.6

1.4

2.4

3.8

5.5

nG (

x10

12 c

m-2

)

Air

EF (

meV

)

ph (x1023 cm-2 s-1)

Vacuum

Electron trapping by molecular adsorbates in air Direct saturation under vacuum

1020 1021 1022 1023 1024

-200

-100

0

100

200

300

-2.4

-0.6

0

0.6

2.4

5.5

n (

x1

01

2 c

m-2

)

S3

S2

EG

rF

(m

eV

)

ph (cm-2 s-1)

S1

1020 1021 1022 1023 1024

100

150

200

250

300

0.6

1.4

2.4

3.8

5.5

n (

x1

01

2 c

m-2

)

S5

EG

rF

(m

ev)

ph (cm-2 s-1)

S4

Several Gr/MoSe2/SiO2 samples Ambient AIR

Gr/MoSe2/SiO2 vs MoSe2/Gr/SiO2

Air (full) vs vacuum (open)

Microscopic mechanism (in vacuum)

MoSe2 graphene

van der Waals gap

−

+

e−

h+

𝐸F,eM

𝐸F,hM

𝐸FiM

Light on, t=0

MoSe2 graphene

van der Waals gap

−

+

e−

h+

𝐸FiM

Light on, t∞

• n+nG=p • Saturation for n Γe = p Γh

• Independent of fphoton

𝐸FGr

𝐸FGr

+ +

Balanced electron and hole currents at saturation Intrinsic mechanism? Optical determination of band offsets?

Recap: Fermi level and PL intensity

Exciton dynamics in MoSe2 is largely independent of the electron and hole transfer efficiencies

Strong hint for dominant energy transfer

Partial conclusion

Strong interlayer coupling in Gr/TMD heterostructures

Saturation of the net electron transfer

Highly efficient (sub)-picosecond energy transfer

Förster vs Dexter energy transfer?

Electrical control of charge and energy transfer

Implications for optoelectronics and optospintronics

More info: G. Froehlicher, E. Lorchat, S. Berciaud, Phys. Rev. X 8, 011007 (2018)

van der Waals Heterostructures

Haigh, Gorbachev et al., Nature Materials 2012 Manchester Group

Castellanos-Gomez et al. 2D Materials 1 011002 (2014)

No dangling bounds No lattice mismatch issues Rotational degree of freedom • 2010 : Graphene on hBN • 2017 : wet or dry transfer, pick up and lift,… • Numerous possibilities!

C-H Lee et al. Nat. Nano (2014) (Columbia)

Atomically thin p-n junctions

52

Graphène

hBN

MoS2

WSe2

-3

-4

-5

-6

--7

Gra

ph

en

e

Mo

S 2

Mo

Se2

Mo

Te2

WS 2

WSe

2

TMD: Y. Liang et al., APL 103, 42106 (2013), M. Ugeda et al., Nat. Mater. 5, 1091 (2014) Graphene: Y.-J. Yu et al., Nano Lett. 9, 3430 (2008)

-4.57

-3.89 -3.74

-4.09

-3.61

-4.29

Ene

rgy

(eV

)

TMD/TMD: Type II band alignment TMD/Graphene: Photoinduced TMD Gr e and h transfer

Band offsets in 2D materials

Microscopic Mechanism


MoSe2 graphene

Vacuum level

𝜒M

𝑊G

van der Waals gap

𝐸FGr

𝑒−

𝐸FM

MoSe2 graphene

van der Waals gap

𝐸FM

In the dark

Before contact: n-doped MoSe2

Weakly doped graphene

After contact (in the dark): Neutral MoSe2

n-doped graphene

𝐸FGr

X

E

Outlook: low-temperature photoluminescence

1.4 1.5 1.6 1.7

0

1

2

3

4

1.4 1.5 1.6 1.7

1.4 1.5 1.6 1.710-1

100

101

102

103

104

1.4 1.5 1.6 1.7

PL

In

ten

sity (

arb

. u

nits)

Defect states Trion

ExcitonMoSe2 MoSe2/Gr

Exciton

4K4K

x10

PL

In

ten

sity (

arb

. u

nits)

Energy (eV)

Defect states Trion

Exciton

MoSe2 MoSe2/Gr

Exciton

4K4K

Energy (eV)

: Exciton

: Trion

No trion emission!

Förster energy transfer: near field dipole-diople interaction

𝐄𝐷 =1

4𝜋𝜀0𝑘2 𝐫˄𝝁𝐷 ˄𝝁𝐷

𝑒𝑖𝑘𝑟

𝑟2+

3𝐫 𝐫. 𝝁𝐷

𝑟2− 𝝁𝐷

1

𝑟3−

𝑖𝑘

𝑟2𝑒𝑖𝑘𝑟

Th. Förster Annalen der Physik 437, 55 (1948) Novotny, Hecht Principles of Nano Optics, Ch 8 R0 : Förster radius

Förster and Dexter energy transfer

Coulomb (FRET) term ‘Long’ range (power law) Implies spectral overlap

Exchange (Dexter) term Short range (exponential, idem CT) Implies overlap of molecular orbitals

1 2 Förster 1946-51

Dexter 1953

A Govorov, PL Hernandez Martinez, HV Demir Understanding and Modeling Förster-type Resonance Energy Transfer (FRET), Springer 2016

5 µm0

1

Ph

oto

n E

mis

sio

n R

ate

(arb

.)

Quartz

Graphene

Elaser Elum

GrapheneNPL

Elaser Elum

Quartz

Energy transfer in low-dimensional heterostructures

Z. Chen et al., ACS Nano 4, 2964 (2010) (QD-Gr) F. Federspiel et al., Nano Lett. 15, 1252 (2015) (QD-Gr, NPL-Gr) L. Gaudreau et al., Nano Lett. 13, 2030 (2013) (Dye-Gr)

Graphene

Quartz

D. Kozawa et al., Nano Lett. 16, 4087 (2016)

Nanoscale emitter/graphene

C. Roquelet et al., APL 97, 141918 (2010)

Molecules/nanotubes

TMD/TMD

nanoplatelets

Highly Efficient Förster-type Energy transfer

Interlayer energy transfer has been largely overlooked in TMD/Gr

© F. Vialla

quartz

0 1 10 100

0.1

1

gT~1/d

4

2D-graphene model

MgO thickness (nm)

Decay R

ate

(n

s-1)

CVD graphene

Quartz

MgO

0 1 10 100

0.1

1

gT~1/d

4

2D-graphene model

MgO thickness (nm)

De

cay R

ate

(n

s-1)

CVD graphene

Quartz

MgO

2D Materials at IPCMS

• Optical spectroscopy

• Phonons, excitons and their coupling(s)

IPCMS-Strasbourg

-50 0 50 100150200250300

0

12L MoTe

2oMX

iMX

oX

iX

Breathing

Ram

an In

ten

sity (

arb

. un

its)

Raman Shift (cm-1)

Shear

168 172 176

2L

5L

4L

3L

1L oX modeR

am

an Inte

nsity (

arb

. units)

Raman Shift (cm-1)

0 2 4 6 8 10 12

169

170

171

172

173

174

Fre

qu

en

cy (

cm

-1)

Number of Layers

oX

Mo Te

Interlayer interactions: Davydov splitting and unified description of the phonon modes

2H TMD - Froehlicher et al., Nano Lett. 2015, Miranda et al., Nano Lett. 2017. Coll: L. Wirtz group, Uni. Luxembourg

1T’ TMD - Lorchat et al. ACS Nano 2016 (ReS2 and ReSe2)

See also Froehlicher et al., J. Raman Spec. 2018 (special issue on 2D materials)

+_

Exciton dynamics and valley-contrasting properties in ... · Exciton dynamics and...

Documents

Transcript of Exciton dynamics and valley-contrasting properties in ... · Exciton dynamics and...