Epitaxial Graphene A new electronic material Walt de Heer

Transcription

1 Epitaxial Graphene A new electronic material Walt de Heer School of Physics Georgia Institute of Technology

2 Georgia Tech W. A. de Heer, C. Berger,P. N. First,E. Conrad X. Wu, M. Sprinkle, M. Ruan,, Y. Hu,, G. Rutter,, L. Miller, K. Kubista, J. Hass,, N.Sharma, NIST J. Stroscio,, and others (NIST) CEA P. Soukiassian Soleil A. Taleb-Ibrahimi A. Tejeda CNRS M. Potemski, G. Martinez, C. Faugeras (other collaborators will be acknowleged later)

3 The History of Graphene Graphitic layers on transition metals, carbides known since the early 70s First identification of: monolayer of graphite single-crystal plane two-dimensional graphite The breakthrough: Gateable graphenes Emphasis on transport SiC Van Bommel, Surf. Sci. (1975) LaB6 Oshima Appl Phys (1977) Pt Zi-Du Surface Science (1987) Ni Rosei PRB(1983) Ir Kholin Surf Sci (1984) Re Gall Sov Phys Sol State (1985) TaC Aizawa PRL 1990 TiC Nagashima, Surf Sci (1993) Ru Marchini (2007) WC TaC, HfC, SiC Forbeaux (1998) SiO 2 Novoselov Nature (2004) (Thin graphite) SiC Berger J. Chem Phys (2004) (Epitaxial graphiene) SiO 2 Novoselov Nature (2005) (Exfoliated graphene) SiO 2 Zhang Nature (2005) (Exfoliated graphene)

4 Graphene was experimentally well-known as a 2D crystal! See, for example 3.2. Monolayer graphite Thin Solid Films 266 ( 1995) N.R. Gall, E.V. Rut kov, A.Ya. Tontegode Graphite films of monolayer thickness form on the surface of many metals (Ir, Re, MO, Pt, Ni, Rh). Monolayer graphite films preserve their individuality as two-dimensional crystals on the surface of the metals [ 61] Valence saturation of monolayer graphite films leads to their catalytic passivity and to a weak bonding only by Van der Waals forces [9]. many atoms (Cs, K, Ba, C, Pt, Si, Au, etc.) [6,19] and even molecules (C,) [20] can intercalate into MGF, penetrating between the graphite layer and the metal surface [6,19,21].

5 So why did it take so long for graphene to catch on? Almost NOBODY CARED!

6 Epitaxial graphene is not an isolated single graphene sheet. However, it is easily made and it exhibits several graphene properties more clearly than exfoliated graphene!

7 Graphene Properties Scalability Mobilities >10 5 cm 2 /Vsec Doping < /cm 2 Berry s phase of ITRS 2007 UHV emerging Furnace material and research devices Tape graphene SiC Si-face C-face Si-face C-Face termination C-face Landau Level E B Weak anti-localization Gapless Linear Dispersion Si-Face termination??

8 Production and Structure of Multilayered Epitaxial Graphene

9 Epitaxial Graphene on SiC C-Face termination (0001) ML n~10 9 /cm 2 Graphene layers SiC (0001) SiC 1-5 ML n~10 12 /cm 2 SiC E E Si-Face termination

10 AFM: C-face MEG C-face, HV (~10-5 Torr) RF induction furnace ~1450 C, 7 min. See M. Ruan W26.011

11 Graphene Growth Gambaz et al, Carbon 46, 841 (2008). Furnace Growth C-face ~ 0.2nm ~ 0.02nm 9 x 9 m Exfoliaded 400 x 400nm At least one sheet continuously covers the entire surface

12 Important notes on furnace grown epitaxial graphene. 1. The GIT furnace grown graphene crystals are exceptionally well-formed compared to UHV grown material with is of poor quality 2. At least the top layer is continuous over the entire surface making MEG graphene crystals by far the largest quasi 2D crystals known. 3. The number of layers varies (at most) by about 1 layer in well-made samples. 4. The interface layer is n doped and probably more disordered than the other layers.

13 20 layers 2D width: 22 cm -1 No D band 2D-Peak cm -1 Lorentzian fit 23 cm -1 Raman Spectroscopy G-Peak cm -1 A. C. Ferrari et.al. PRL 97, (2006)

14 Stacking Si face: Bernal (AB) C-face Rotational stacking (Multilayered epitaxial graphene)

15 Graphene/SiC Commensurability (LEED) SiC bulk Graphene Diffuse Graphene ring SiC bulk (6 3 x 6 3) SiC R30º Si-Face UHV grown 69.1eV C-Face furnace grown 72.2eV Graphene Surface X-ray Diffraction R-2.204º R2.204º R30º

16 Graphene/Graphene Commensurability (aligned with SiC) 30 0 (aligned with SiC) º = graphene-graphene commensurate cell 2.2º = graphene-sic commensurate cell 8.9Å Si-face Graphite (AB stacked) C-face (rotated phases)

17 STM evidence for rotated phases ( 13 x 13) G R46.1º 8.9 Å 5 Å Sample bias: -800 mv Tunneling current: 100 pa 2 sheets with a relative rotation of J. Hass, et al. Phys. Rev. Lett (2008)

18 Substrate induced rotations SiC Å X 160Å data ( )R ( )R10.9 ( )R (relative to SiC) (relative to SiC)

19 20nmX20nm Miller et al Science, In print 3.8nmX3.8nm 47nmX47nm

20 Rotational Domain Boundaries Atomically flat and continuous across boundary 400 nm 50 pm Joseph A. Stroscio

21 Rotational Domain Boundaries 20 nm 8 nm Joseph A. Stroscio joseph.stroscio@nist.gov

22 Rotational domains (LEEM) 10 m 72.2eV Ellipsometery thickness map: 10±1 layer E.Conrad, M.Sprinkle

23 The Dirac cone

24 Graphene band structure Dirac Point H = v F ˆ p ˆ = ( ˆ x, ˆ y, ˆ x = 0 1 ; 1 0 y = 0 i ; i 0 z = z ) Velocity is constant E =±v F p Neutrino-like dispersion

25 Substrate-induced band gap in Si-face EG Zhou, Gweon, Fedorov, First, de Heer, Lee, Guinea,. Castro Neto, Lanzara Nature Materials 6, (2007) n=1 n=2 n=3 n= Graphene on Si-face: gap is observed; Gap closes as the number of layers increases.

26 Rotations preserve sublattice symmetry Isolated graphene sheet Graphite bilayer R30/R2 fault pair

27 T=6K; near K-point ARPES: MEG (C-face) E D - E F ~ 20 mev ± 10meV n~ 10 9 /cm 2 (essentially undoped) Each rotated sheet displays linear dispersion!

28 Layer Interactions (MEG) ~40meV?? F These two sheets are ~3-4Å apart v (right) 0.99 F (ave.) = 0.87 ± m/s v (left) = 0.81 ±.02 F (graphite) = m/s k(a -1 )

29 Spectro-microscopy of single and multi-layer graphene supported by a weakly interacting substrate Knox, Wang, Morgante, Cvetko, Locatelli, Onur Mentes, Angel Ni, Philip Kim, Osgood ARPES of exfoliated graphene: does it have a Dirac point?

30 Scanning tunneling spectroscopy of Landau levels Observing the Quantization of Zero Mass Carriers in (Multilayer Epitaxial ) Graphene Science In Press D.L. Miller, K. D. Kubista, G. M. Rutter, M.Ruan, Walt A. de Heer, P. N. First, J. A. Stroscio (NIST, GIT)

31 Previous STS Measurements on Graphite Surfaces Complex spectra Mixture of peaks of linear and nonlinear in B T. Matsui et al. PRL (2005) G. Li and E. Andrei Nature Phys. (2007) Courtesy Joe Stroscio

32 Landau Levels in Graphene Density of states vs E, B Magnetic Field B=8T E = n eh% c n B ± n 2 sgn( ) 2 n=0, 1... Graphene heb E ( 1/2) n = n+ n 0 * m Standard 2DEG Energy (mev) Courtesy P.N.FIrst

33 STS Multilayer Epitaxial Graphene: Landau Levels

34 E n E 0 (mev) LL 0 shift c* = 1.13 x 106 m/s

35 Landau Level Fit T=4K Simple sum of Voigt functions (Gaussian, Lorentzian convolution) Gaussian: 2.8 mev (instrument function + thermal broadening; fixed) LL 0 Lorentzian: 1.5 mev (0.4 ps lifetime: lower limit to momentum relaxation time)

36 Tunneling Magnetoconductance Oscillations (~SdH oscillations, but not restricted to E F ) Density of states vs E, B Magnetic Field Energy (mev) Courtesy P.N.FIrst

37 Tunneling Magnetoconductance Oscillations Analogous to SdH oscillations

38 E vs B Fit with v= 1.07*10^6,

39 Landau Index n Fan Plots n B -1 (T -1 ) E = n eh% c n B ± 2 sgn( ) 2 n=0, 1... LL index vs 1/B for different energies

40 Dispersion E(k) c* = (1.070 ± 0.007) x 10 6 Slight asymmetry 2 E = sgn( n) 2 eh c% n B n=0, ± 1... = c% n h k

41 ARPES: C-face MEG vf = 0.87 ± m/s ED - EF ~ 20 ± 10 mev

42 Landau level spectroscopy

43 Courtesy C. Faugeras

44

45 Bands in a magnetic field : Landau level Energy quantization 2 r c n F r c =p/eb F =2 /k F B Normal electrons: quadratic E(k) B B Dirac particles: linear E(k) Energy Energy 3 n E F Magnetic field E n = (n ) heb m * Magnetic field E n =± 2ehc 2 Bn

46 Landau Spectra of C-face Graphene (B) line Transition energy (mev) E F E n =± 2ehc 2 Bn Transmission layers 9-10 layers 50 layers B=1.5T 1.4T 1.5T 1.5T 0.8 HOPG ~ μm B(T 1/2 ) Wavenumber (cm) -1 M. Potemski, G.Martinez, CNRS-LCMI

47 IR cyclotron resonance spectroscopy E F PRL accepted >250,000

48 IR cyclotron resonance spectroscopy LL s visible at RT below 1T.

49 MEG Landau Levels Temperature Independent LL s. Temperature Independent widths Boltzmann population of levels Weak Electron-Phonon Coupling: >250,000 at RT

50 Transport

51 Epitaxial Graphene on 4H-SiC SiC C-Face termination SiC C-face Multilayer Epitaxial Graphene (MEG) ML Si-face Ultrathin Graphite 1-5 ML Si-Face termination n~10 9 /cm 2 Graphene layers Uncharged MEG n~10 12 /cm 2 SiC E Charged interface graphene layer C-face

52 Magneto-transport of 1-2 graphene layers on Si-face Hall bar 400X800 m Si-face, UHV grown, =1000 Rxx ( /sq) 0.3-4K Rxy ( ) Landau index (n) Landau plot n s = cm -2 (same as from xy ) 1/B n (T -1 ) Berry s phase= Rxy ( ) n=3 xy =1.8k Berger et al. J. Chem. Phys. 2004

53 T=4,6, 9, 15, 35 and 58 K; -9 T B 9 T. C-face Hall bar 1 mx200 nm *=27000cm 2 /Vs. Landau levels: E n (B)= 2neBv 2 0 ) Confinement: E n (W)=n v 0 /W Confined Landau levels: E n (B,W) [E n (W) 4 +E n (B) 4 ] 1/4 Berry s phase=

54 Magnetotransport: graphene like; Berry s phase = (No QHE) Hall bar C-face R ( ) R xx ( /sq) 1 m x 6.5 m R= 502 /sq = cm 2 /Vs Landau level index 1/B (T -1 ) R/R=4% Field (T) 100 mk Shubnikov de Haas oscillations Landau level index 1/B (T -1 ) Solid State Com. 2007,Grenoble/GIT collaboration

55 Weak anti-localization (WAL) 4.2 K WAL Theory WL 4.2 K C-face Hall bar

56 Magnetotransport of ~1 graphene sheet on C-face Cross etched in a flat terrace xy 2.5 m (k ) μ 9000 cm 2 /Vs n s = cm -2 xx xx (k ) Field(T)

57 Anomalous Conductance Transition A reversible, reproducible, drop in the conductivity is observed at 200K. The resistance is at its theoretical minimum. Transport is phase coherent over the entire structure (0.5X5 μm). Resistance is at its theoretical minumum (no boundary scattering!) Oscillations periodic in the magnetic field are seen Temperature (K) Could this be the Hoffstadter butterfly? (Moire with a 20 nm lattice constant)

58 Devices

59 The Dirac point E F E D xx (Resistance)(k ) K -5 Tesla Negative Hall (holes) Dirac point Positive Hall (electrons) 5 0 xy (Hall)(k ) 1.5 mx12 m Top gated Hall bar, Si face G D S G V g (V) -5

60 Top and side gated FETs Si-face Top gate g1 Top gated FET g2 Resistance (k /sq) x12.5 m C-face Top gate 3.5x12.5 m s Side gated FET g1 g3 d C-face Side gate d x1 m Gate Voltage (V) The first epitaxial graphene transistors S g2

61 Production process SiC blank Hydrogen etch Device Integration Pattern

62 Final Device Geometry SiC Pt Source Al gate (next level) L=10um W=5um Gr Al SiC Pt Drain Al HfO 2 SiC Graphene Device description and cross section Nominal device Graphene/SiC active layer (C-side), L = 10um, W = 5 um, ridge parallel, 50nm HfO 2 dielectric, Al gate Microscope image shown before gate lift-off J. Kedzierski, P. L. Hsu, P. Hea ley, P. W. Wyatt, C. L. Keast, M. Sprinkle, C. Berger, W. A. de Heer, Epitaxial graphene transistors on SIC substrates,ieee T Electron Dev 55, 2078 (2008).

63 Set of Identical Devices (Si-face) d (1/Ohms) 1.4m 1.2m 1.0m GraphC W2 Si T HfO2 = 40nm W=5μm, L=10μm Minimum conductivity 130uS to 250uS Field Effect Mobility values cm 2 /Vs I on /I off ~ V g (V) Drain current vs. gate voltage at V d = 0.5V J. Kedzierski, P. L. Hsu, P. Hea ley, P. W. Wyatt, C. L. Keast, M. Sprinkle, C. Berger, W. A. de Heer, Epitaxial graphene transistors on SIC substrates,ieee T Electron Dev 55, 2078 (2008).

64 Lithography 98 ribbons per 3.5 x 4.5 mm chip Hall bars and 2-point devices