The Raman Fingerprint of Graphene - The Nanotube...

CAMBRIDGE UNIVERSITYDEPARTMENT OF ENGINEERING

ELECTRONIC DEVICES AND MATERIALS GROUP

The Raman Fingerprint of GrapheneA. C. Ferrari1, J. C. Meyer2, V. Scardaci1, C. Casiraghi1,

M. Lazzeri3, F. Mauri3, S. Piscanec1, D. Jiang4, K. S.Novoselov4, S. Roth2, A. K. Geim4

1Department of Engineering, University of Cambridge, Cambridge, UK2Max Planck Institute for Solid State Research, Stuttgart 70569, Germany3Institut de Mineralogie et de Physique des Milieux Condenses, Paris, France4Department of Physics and Astronomy, University of Manchester, UK



NanoTube-Evolution-Nagano 06



NanoTube-Evolution-Nagano 06

“Cut…. Get graphene”Eklund-Sensei NT06

“Press (50GPa)…Get Diamond-like Carbon!”

S. Saito-Sensei CCNT06



Graphene

•Electron transport described by the (relativistic-like) Dirac equation Access to the rich and subtle physics of quantum electrodynamics in a relatively simple condensed matter experiment

•Scalability of graphene devices to true nanometre dimensions makes it a promising candidate for future electronic applications, because of its ballistic transport at room temperature combined with chemical and mechanical stability.

•Graphene is the two-dimensional (2d) building block for carbon allotropes of every other dimensionality

….big hype recently…



Transistor: Graphene Ribbon



Drawing:(micro) mechanical cleavage of graphite

How to Make Graphene?


ELECTRONIC DEVICES AND MATERIALS GROUPindividual atomic sheets: do they exist?

How to Make Graphene?

GRAPHITE ISSTRONGLY LAYERED

SLICE DOWN TO ONE ATOMIC PLANE


ELECTRONIC DEVICES AND MATERIALS GROUP1 μm

SEM

1 μm

0Å 9Å 13Å

Free-Standing GrapheneAFM

1 μm

single layer of atoms visible by “naked” eyeonly on 300 nm SiO2

OPTICS

Key: Visual Identification



HoweverNeed Extremely

Good Eye To Spot!!!

Single Layer



TwoLayers



One Comment…

However…Low yield, messy, not scalable

Better to grow graphenedirectly on substrate

This can be done…But…not the subject of this talk

Mechanical cleavage is nice and simple



Another Comment

AFM thickness of single layer is0.5-1.5 nm! Due to chemical contrast

We want to be 150% sure

TEM



Free-Hanging graphene sheets

500nm

1 layer of graphene !

J. C. Meyer



Preparation1. Graphene sheet on substrate

2µm

2. Metal grid patterned ontothe flake

10µm



Preparation

3. Etching of substrate

10µmFlake remainsin metal grid

4. TEM and electron diffraction analysis

500nm



Electron diffraction:

Highly crystalline samples

Number of layers?Diffraction tilt series!



Sheets fold back at the edges, and sometimes show a wrinkle within the sheet. HRTEM analysis of the folding allows to verify the layer count.

2nm

2nm

One-layer graphene



Two-layer graphene

2nm



Two-Layer Graphene

2 Layers

Stacking A/B



8 layers



We Need High Throughput Non Destructive

Quick Substrate Independent

Identification Technique

Raman Spectroscopy



1500 2000 2500 3000

Inte

nsity

Raman shift (cm-1)

514nm

Graphene

Graphite



2500 2600 2700 2800 2900 30000

10000

20000

30000

40000

Graphene

Inte

nsity

(a. u

.)

Raman shift (cm-1)

514 nm

Graphite G’ Peak

(2D peak)

Clear Fingerprint



1800 2400

Inte

nsity

(A.U

.)

Raman Shift (cm-1)

Graphene

1200 1300 1400 1500 1600 1700 1800

Inte

nsity

(A.U

.)

Raman Shift (cm-1)



1800 2400

Inte

nsity

(A.U

.)

Raman Shift (cm-1)

Graphene

1200 1300 1400 1500 1600 1700 1800

Inte

nsity

(A.U

.)

Raman Shift (cm-1)

D peak intensityNOT related to Number of Layers

Disorder (in the widest possible meaning)

See Tuinstra Koening 1970…



1500 2000 2500 3000

Inte

nsity

(A.U

.)

Raman Shift (cm-1)

Graphene

On Substrate

Suspended



1500 2000 2500 3000

Inte

nsity

(A.U

.)

Raman Shift (cm-1)

2-Layers

514 nm

2600 2640 2680 2720 2760 2800

Inte

nsity

(A.U

.)

Raman Shift (cm-1)



1500 1550 1600 16501500 1550 1600 1650

10 layers

5 layers

Inte

nsity

(a. u

.)

Raman shift (cm-1)

1 layer

2 layers

graphite

Raman shift (cm-1)

514nm 633 nm

Slight Upshift ~ 5cm-1



2600 2700 28002600 2700 2800

10 layers

5 layers

2 layers

Inte

nsity

(a. u

.)

Raman shift (cm-1)

1 layer

graphite

514nm

633 nm

Raman shift (cm-1)



2550 2600 2650 2700 2750 28000

5000

10000

15000

200002D2B2D2A

2D1A

Inte

nsity

(A.U

.)

Raman Shift (cm-1)

E) 2 Layer

514.5 nm

633 nm

2D1B

1300 1350 14000

500

1000

1500

2000

2500Edge Graphite

D2

Inte

nsity

(A. U

.)

Raman Shift (cm-1)

D

D) 514.5 nm

D1

Edge 1 Layer



The Origin of G’ (2D) Peak

BUT it is the second order of D peak

Nothing to do with G peak

Named G’ since is one of the 2 biggest peaks in graphite

D forbidden in perfect crystal By Raman Fundamental Selection Rule q~0

However 2nd order always allowed: q+(-q)=0



D Peak Dispersion Vidano et al. (1981)

Pocsik et al. (1998)



Active by double resonance(Baranov 1988, Thomsen-Reich 2000)

D peak comes from LO phonons(Ferrari Robertson 2000)

Strongly dispersive due to Kohn Anomaly at K(Piscanec et al. 2004)



Second order no defect scattering necessary

However…

In principle ALL phonons activeBUT

Double resonant phonons enhanced due to resonance and strong

electron-phonon coupling



Previous double resonance modelspredict multiple D peaks for graphene

Double structure of 2D peak in graphitenever explained

Traditional interpretation (1980)2 Maxima in graphite

Phonon Density of States at K and MWRONG since 2D disperses with excitation

KEY: Evolution of Electron Bands with number of layers

in contrast with experiments



Three Possible Processes

However: only 1 contributes



Three Possible Processes

However: only 1 contributes

Small:Trigonal Warping

Forbidden:EPC=0

OK!



Trigonal Warping Effect

• q > K Strong EPC and large portion of the phase-space • q < K Strong EPC but small portion of the phase-space • q ~ K EPC~0

Adapted from: Kurti et al., Phys. Rev. B 65 165433 (2002)



1 Component D and 2D peaks

2500 2600 2700 2800 2900 3000

Graphene

Inte

nsity

(a. u

.)Raman shift (cm-1)

514 nm2D

1 3 0 0 1 3 5 0 1 4 0 0

Inte

nsity

(A. U

.)

R a m a n S h if t (c m -1 )

5 1 4 .5 n m

E d g e 1 L a y e r

D



0.8 0.7 0.6 0.5

1300

1320

1340

1360

1380

1400

1420

Μ Κ Γ

q = (ξ,0,0)

ve Vector (2π/a0)

Two-layer GrapheneTwo possibilities:

1) Phonon Splitting

2) Band splitting

Phonon SplittingK-M is Minor

PRL 93, 185503 (2004)



Band Splitting Main Effect

2550 2600 2650 2700 2750 28000

5000

10000

15000

200002D2B2D2A

2D1A

Inte

nsity

(A.U

.)

Raman Shift (cm-1)

2 Layer

514.5 nm

633 nm

2D1B

4 components2 Most intense



Origin of Small Upshift of G peak

1 5 0 0 1 5 5 0 1 6 0 0 1 6 5 0

Inte

nsity

(a. u

.)

R a m a n sh ift (cm -1 )

1 la ye r

g ra p h ite

-0.1 0.0 0.11570

1580

1590

1600

q = (0,ξ,0)

Γ

~5 cm-1 Upshift

IRRaman GrapheneRaman Graphite

PRL 93, 185503 (2004)

Kohn Anomaly



Phonon-Linewidths and EPCIn a perfect crystal, phonon linewidths determined byInteraction with other elementary excitations:

EPCan γγγ +=γan : anharmonic contribution, due to interactionWith other phonons. Determined by anharmonicterms in interatomic potential.

γEPC: interaction with electron-hole pairs. Determinedby EPC and present in systems with null electron gap



Γ-E2gLO Phonon Decay Processes

From the Fermi Golden Rule:

2

2)(2 β

γ Γ∝

−Γ

EPCEPCE LO

g



Γ-E2gLO : Graphite Raman G Peak

1000 1200 1400 1600 1800 2000

Inte

nsity

(A. U

.)

Raman Shift (cm-1)

FWHM(G)= 13.5 cm-1

Single Crystal Graphite

No D Peak

No extra broadening due to disorder

No FWHM(G) increase with temperature

γan≤1.5 cm-1

(spectrometer resolution)

EPC(Γ)=45.5 (eV/Ą)2



And… Single layer graphene…

1200 1400 1600 1800 2000

FWHM(G)= 14 cm-1

Inte

nsity

(A.U

.)

Raman Shift (cm-1)

Similar EPC

PRB 73, 155426 (2006)



Implications for NanotubesSingle 2D peak graphene ⇒

Single 2D peak in Single Wall CNT

Curvature and confinement give diameter dependence

2D(SWNT)~2D (graphene)- A/d

2D position in Graphite should not be used to scale

Distribution of SWNTs of different diameters, distribution of 2D peaks



What about Multi-Wall?

First approximation each wall gives a 2D peak

DWNT two 2D peaks (inner and outer wall)

HOWEVER, inter-wall interactions Can change simple picture

Further splitting, Less peaks!

Details to follow…



ConclusionsIdentified unique features of Raman spectrum, which fingerprintsgraphene amongst all other carbon allotropes.

The Raman spectrum evolution with increasing number of layers reflects the evolution of the electronic structure and electron-phonon interactions

Raman spectroscopy is a quick, high-throughput, non-destructive technique for the unambiguous identification of graphene layers.

Raman+Graphene is Good Fun!



Funding:• EU FAMOUS & CANAPE project• Marie Curie Fellowship IHP-HPMT-CT-2000-00209• The Royal Society• EPSRC GR/S97620/01•The Leverhulme Trust

CPU:• HPCF, Cambridge UK• IDRIS, Orsay France



Reference

The Raman Fingerprint of Graphene - The Nanotube...

Documents

Transcript of The Raman Fingerprint of Graphene - The Nanotube...