Tailoring the electronic properties of low-dimensional carbon hybrids Thomas Pichler University of Vienna Faculty of Physics Tokyo 5.10.2012
Acknowledgements Vienna University group: C. Kramberger, A. Grüneis, P. Ayala,K. de Blauwe, X. Liu, R. Pfeiffer, J. Chacon, G. Ruiz, A. Briones, L. Shi, M. Sauer, P. Rohringer, F. Simon (now Budapest), A. Chernov, H. Kuzmany IFW-Dresden: M. Rümmeli, M. Knupfer, B. Büchner Surrey University: H. Shiozawa (now Vienna Univ.), R. Silva Synchrotron: BESSY II: D. Batchelor, P. Hoffmann, J. Fink, R. Follath, S. Krause ELETTRA: P. Lacovig, M. Dalmiglio, S. Lizzit, A. Goldoni Theory: Donestia San Sebastian: A. Rubio, C. Allicante, D. Mowbray Lille Univ.: L. Wirtz, Samples: AIST: H. Kataura Tokyo Metrop. Univ.: K. Yanagi IFW-Dresden: M.H. R. Nagoya Univ: Y. Miyata, R. Kitaura, H. Shinohara Funding: DFG, FWF, APART, EU
OUTLINE Introduction/Motivation/Experimental/Applications Part 1: Electronic structure of 1D nanostructures: Single walled carbon nanotubes (SWCNT): a) Pristine SWCNT: nature of the metallic ground state b) Electronics and optics with functionalized SWCNT: Intercalation, substitution/attachment and filling Part 2: Electronic structure of 2D nanostructures: Graphene systems: a) Graphite revisited: A Key to Graphene and SWCNT b) GIC KC 8 revisited: A key to Graphene c) Graphane, n-doped graphane: Tuneable gap, acceptor level d) Electron dispersion, electron phonon coupling Summary and outlook
TODAY Introduction/Motivation/Experimental/Applications Part 1: Electronic structure of 1D nanostructures: Single walled carbon nanotubes (SWCNT): a) Pristine SWCNT: nature of the metallic ground state b) Electronics and optics with functionalized SWCNT: Intercalation, substitution/attachment and filling Part 2: Electronic structure of 2D nanostructures: Graphene systems: a) Graphite revisited: A Key to Graphene and SWCNT b) GIC KC 8 revisited: A key to graphene c) Graphane, n-doped graphane: Tuneable gap, acceptor level d) Electron dispersion, electron phonon coupling Summary and outlook
Low dimensional quantum solids? All solids are of course quantum systems 1D and 2D quantum solids: - Mesoscopic systems with size quantization - Quantum effects already at room temperature - Correlation effects crucial: electron-hole, electron-electron, electron-phonon - Instabilities in 1D and 2D systems! Molecular carbon nanostructures are archetypical examples of perfect 1D and 2D solids
Dimensionality of sp 2 bonded carbon allotropes Graphite, Mittelalter 3D Carbon nanotubes Iijima, 1991 1D Fullerene 0D Graphene 2D A. Geim, K. Novoselov, 2004 Nobel prize physics, 2010 R. Curl, H. Kroto, R. Smalley, 1985 Nobel prize chemistry, 1996
But also.. Diamond Graphite more? Carbon nanowire Carbyne A. Hirsch et al. Nature Materials 9, 868 (2010)
Motivation: Why all carbon nanostructures? Advantages of carbon: 1. cheap, light 2. structural properties: hard, stiff, flexible, ductile,. 3. biocompatible 4. electronic properties: insulating semiconducting semimetallic metallic zero gap semiconductor diamond fullerenes, graphite, nanotubes nanotubes graphene Tunable by functionalisation
Electronic propertieres: Special properties of nanotubes Geometry of SWCNT : 2/3 are semiconducting 1/3 metallic conductivity tunable by functionalisation Temendous current carrying capacity: 1 billion A/cm 2 no electromigration Excellent heat conductor: twice as good as diamond Mechanical properties: Strength at least 100 times higher than steel Youngs modulus 1 Tpa Flexible ultimate carbon fibers Geometry: High aspect ratio of 1000 to 10000, 1 nm diameter High surface area
Electronic structure of different carbon allotropes Two 1s core electrons and 4 valence electrons (two 2s and two 2p) σ σ σ σ 2p 2p 2p z π π 2s sp 3 σ σ 2s sp 2 π σ π σ sp 3 Diamond sp 2 graphite nanotubes fullerenes conj. polymers σ-bonds: W~50 ev π-bonds W~ 20 ev σ bond π-orbital: W~ 0.5 ev
Introduction: formation of single-wall carbon nanotubes SWCNTs Hamada vector graphene sheet r r r C = na1 + ma2 = ( n, m) C (n,0) Φ A (n,m) d (n,0) zigzag Φ=0 a1 o (n,n) C O(A) (n,n) armchair Φ=30 a 2 The diameter of SWCNT (n,m): d n,m = C /π=a(n 2 +m 2 +mn) 1/2 /π a= a 1 = a 2 =2.49 Å (n,m) chiral e.g. (10,10) d=1.37nm 0<Φ<30
Introduction: electronic properties of SWCNTs periodic boundaries 2πd k n =2π n (k n -quantization) k K 1/d K metal 1/d K semiconductor 9 (10,10) 9 E F E F E(K) (ev) 6 3 0-3 EF 6 3 0-3 Density of states 0 (10,10) K Γ K M K -6-6 (17,0) -9-9 0.0 0.5 1.0 0.0 0.5 1.0 ka/π ka/π (17,0) 0-3 -2-1 0 1 2 3 E(k) (ev) E g ~1/d Electronic properties are determined by the structure of SWCNT using tight binding approximation
What are we doing on nanotubes..? Synthesis and purification of SWCNT and DWCNT. high vacuum CVD! Electronic/optical properties: EELS, XAS,ARPES,Raman, PL,OAS! What happens upon functionalization? Charge transfer vs. hybridisation: bonding environment, doping level, Fermi level shift and changes in the DOS. Nature of the metallic ground state: Normal Fermi liquid or a Tomonaga-Luttinger liquid? Transition 1D 3D metal? Electronic structure of functionalized SW(DW)CNT from high energy and optical spectroscopy
SWCNT from laser ablation/ arc discharge Purification and opening: HNO 3, H 2 O 2, Heat treatment SWCNT yield and diameter distribution vs. T, p, laser pulse, gas, catalyst, semiconducting metallic Optimised: 70 wt% SWCNT, d=1.2-1.4 nm SWCNT with no catalyst and carbon impurities, d=1.4 +/- 0.12 nm
Optical response of type separated bulk SWCNT Y. Miyata et al., J. Phys. Chem. C 2008, 112, 3591 (2008)
Samples Semiconducting Metal S S* 3,4 3,4 DOS M S * 2,3 S 2 2 M * S S * 2,3 1 1 M M * 1 1-2 -1 0 1 2 Energy (ev) diameter 1.37 ± 0.08nm Miyata et al. J Phys. Chem.C 112, 13187 (2008).
Gain from High Energy Spectroscopy Occupied Density of States Photoemission Spectroscopy Unoccupied Density of States X-ray absorption and EELS Matrix element weighted DOS Valence Band Conduction Band (site selective) (site selective for RESPES) Bonding environment Charge Transfer Basic Correlation Effects Bonding environment Charge Transfer Basic Correlation Effects Core Hole Effects!!!
Core Level High resolution XPS METALLICITY MIXED SWCNT Normalized Intensity 0,4eV 285.2 284.8 284.4 284.0 Binding Energy (ev) Kramberger et al. PRB 75 (2007) 235437 -Doniac-Sunjic lineshape -Voigtian (symetric) M SC
High resolution XPS of metallicity sorted SWCNT Graphite 0,32 ev Doniach Sunjic Voigtian Core hole screening not visibly affected in metallic samples Work functions: different in bulk metallic SWCNTs sample P.Ayala et. al. PRB 80 (2009) 205427
k What is the electronic behavior of the 1 D system?? σ* π* π σ 0 D E F π* π 2 D C1s C 1s
High Energy Spectroscopy studies on sp 2 hybridized C structures VB-PES Conduction Band-XAS Normalized Intensity σ π C 60 SWCNTs Graphite 10 8 6 4 2 0 Energy (ev) k Matrix element weighted DOS of the valence band 0 D 1 D 2 D Normalized Intensity π σ 285 290 295 Energy (ev) Site selective weighted projected DOS of the conduction band
Quasi 0D : Fullerenes + resol. broad. Measured VB-PES HOMO 0D + resol. broad. Measured LUMO XAS σ* π* π Calc. DOS Calc. DOS σ 6 5 4 3 2 1 0 Binding Energy (ev) 283 284 285 286 287 288 Energy (ev) C1s Quantum Chemical calculations of molecular orbitals match experimental DOS small core hole effects in XAS T.Pichler et al. PRL 78(1997) 4249
2D: Graphite-Graphene VB-PES 2D XAS π π* E F π TB-DOS of the valence band matches experimental DOS But: strong core hole effects in XAS of the conduction band C 1s Bruhwiler et al.prl 76 (1996)1761
C 1s XAS: Conduction band of metallicity mixed SWCNT bundles E F π* π Normalized Intensity π 285 290 Energy (ev) SWCNT HOPG 282 283 284 285 286 σ Energy (ev) Site selective weighted projected DOS of the conduction band Fine structure in π* resonance: S 1 *,S 2 *, S 3 *, and M 1 *. But: exitonic effects for the π* along tube axis like in graphite S 1 * S 2 * M 1 * S 3 * C1s BE M 1 S 2 S 1 S 1* S 2 * -1.0-0.5 0.0 0.5 1.0 Energy (ev) M 1 * vhs C. Kramberger et al. PRB, 75, 235437 (2007)
vhs in DOS of conduction band: XAS fine structure E F π* π C 1s Normalized Intensity a) DOS π M * M * 2,3 M* 4 1 Normalized Intensity 0 1 2 3 0 1 2 3 Energy (ev) Energy (ev) b) DOS π S * S * S * S * S * 4 1 2 3 5 284 285 286 287 284 285 286 287 Energy (ev) Energy (ev) Strong excitonic effect on π* resonance (parallell), vhs only weakly affected P. Ayala et al. PRB80, 205427 (2009)
Valence band of a metallicity mixed SWCNT bucky paper High resolution angle integrated photoemission at T=35 K: Intensity (arb. units) σ M 1 S2 1.5 1.0 0.5 0.0 Binding energy (ev) π S 1 M 1 S 2 S 1 S 1* S * 2-1.0-0.5 0.0 0.5 1.0 Energy (ev) M 1 * 10 8 6 4 2 0 Binding energy (ev) vhs of tubes: S 1, S 2, M 1 good agreement with TB calculations e.g. H. Ishii et al, Nature, 426, 540 (2003), H. Rauf et al.
Diameter cumulative DOS in VB photoemission response a) DOS b) DOS Intensity (cps) π 4 2 0-2 -4 4 2 0-2 -4 Energy (ev) Energy (ev) M 2,3 M 1 E F Intensity (cps) π S 3,4,5 S 2 E F 0.5 0.0 Energy (ev) 4 3 2 1 0 4 3 2 1 0 Binding energy (ev) Binding energy (ev) Very good agreement between PES peaks and diameter cumulative DOS! Additional life time broadening in experiment! Chemical potential for semiconducting tubes shifted by 0.1 ev! No Fermi edge in metallic SWCNT S 1 P. Ayala et al. PRB80, 205427 (2009)
Scaling in 1D and 3D metals: gain from photoemission Photoemission intensity: I(E) ~ n(e) f(e) 3D metal: Fermi-liquid PES shows Fermi edge f(e) = 1 exp[(e-e f )/kt] + 1 1D metal: Tomonaga-Luttinger liquid 1D paramagnetic metal correlated electron state interaction parameter g n(e) ~ const, α=0, g=1 1 I(E) T = 0 E α E f α=0 g<< 1 repulsive Coulomb interaction g>1 attractive interaction power law dependence of n(e): n(e) ~ E α For SWCNT: α=(g+1/g 2)/8 =0.3 0.5
Metallic tubes in metallicity mixed samples Intensity (arb. units) T = 35K M 1 log Intensity (arb. units) α = 0.43 0.2 0.02 log Binding energy (ev) S 2 Metallic SWCNT in a bundle of SWCNT have the renormalization of a Tomonaga- Luttinger liquid 1.5 1.0 0.5 0.0 Binding energy (ev) α = 0.43, g=0.18 S 1 semiconducting metallic good agreement with theoretical predictions and results from transport measurements H. Ishii et al, Nature, 426, 540 (2003); H. Rauf et al, PRL 93, 096805 (2004)
A bundle of undoped metallic SWCNT Intensity (arb. units) log Intensity (arb. units) α = 0.37 1 0.1 0.01 log BE (ev) M1 1.5 1.0 0.5 0.0 Binding energy (ev) No Fermi liquid! No Fermi edge for metallic SWCNT! Undoped purely metallic SWCNT bundles are still 1D metals Still van der Waals interaction of the tubes in a bundle! Implications on maximum conductivity in e.g. transparent electrodes P. Ayala et al. PRB80, 205427 (2009)
Intercalation Alkali-metal intercalation Sapphire film doping (10,10) UHV vapor getter e - DOS (ev -1 C -1 ) 0.1 e - 0.0-2 -1 0 1 2 Energy (ev) in-situ doping: UHV evaporation (5x10-10 mbar) Na, K, Rb, Cs and Ba SAES getters
INTERCALATION: What happens with doping? doping SC M
INTERCALATION: What happens with doping? doping SC M Is a bundle of doped metallic tubes a normal Fermi liquid or a Luttinger liquid?
Transition from a TLL to a FL Crossover to a Fermi liquid Intensity (arb. units) pristine 1:20 metal T=35K α = 0 α = 0.43 0,10 0,05 0,00-0,05 Binding energy (ev) Dependence of the power law scaling factor alpha from the degree of doping. H. Rauf et al, PRL 93, 096805 (2004)
Where is the transition from a TLL to a FL? Intensity (arb. units) pristine metal 1:20 T=35K α = 0 α = 0.43 Intensity (arb. units) log Intensity (arb. units) M1 α = 0.37 1 0,1 0,01 log BE (ev) TLL? 0,10 0,05 0,00-0,05 Binding energy (ev) FL 1,5 1,0 0,5 0,0 Binding energy (ev)
Rigid band shift in metallicity mixed samples M 1 Intensity (arb. units) S 2 S 1 Intensity (arb. units) α = 0.43 T=35K T=35K α ~ 0 1.5 1.0 0.5 0.0 Binding energy (ev) 1 0.1 Binding energy (ev) M 1 S 2 S 1 S M * 1* S * 2 1 S 1, S 2 and M 1 shift to higher binding energies power law behavior at first stable abrupt change at K:C = 1:125, E(S 1 ) = 0.37eV -1.0-0.5 0.0 0.5 1.0
Doping dependence of the LL parameter α in mixed samples 0.5 1 3 4 1.0 α 0.4 0.3 0.2 0.1 2 o E s1 =0.3 ev 5 6 E s1 = 0.37 ev Energy (ev) 0.5 0.0-0.5 6 4 2 7 5 3 1 0.5 0.0-0.5 0.0 0.000 0.005 0.010 K / C 7-1.0 Power law scaling vs. intercalation: α= const. up to K/C=1/500, E S1 =0.25 ev α=0, g=1 above K/C=1/150, E S1 >0.35 ev a bundle of only metallic tubes is a normal Fermi liquid! H. Rauf et al. PRL 93, 096805 (2004); C. Kramberger et al. PRB 79, 195442(2009)
Substitution: Heteronanotubes TEM of DWBN-NT SWCNT B-SWNT, BC 3 NT B,N Substitution C B BCN-SWNT, MWBNT E. Borowiak-Palen et al. Chem.Comm.1, 82 (2003), CPL, 387, 516 (2003) G. Fuentes et al. PRB 67, 35429 (2003), PRB 69, 245403 (2004), P.Ayala, et al., Reviews of Modern Physics (2010).
Filling of the inner space of SWCNTs with molecules Modified electronic properties? What happens upon the formation of the peapods? Charge transfer, chemical bonding or orbital hybridisation? Conversion to double-wall carbon nanotubes (DWCNT)?? Filling@SWCNT so-called peapod? Filling Potential applications Carrier tuning of SWCNT, protection for molecules and even drugs, stabilizing reactive compounds and nanocrystals, chemical reaction inside a 1D confined nano reaction tube, 1D spinchain for spin electronics,
Examples Fullerenes@SWCNT C 60 @SWCNT : 1D C 60 chain inside SWCNT (TEM,Raman) Gd@C 82 @SWCNT : Band gap modulation (STM,STS) Dy 3 N@C 80 @SWCNT : Charge transfer (PES) Conversion to DWCNT(TEM, Raman)! C 60 @SWCNT Organic molecules @SWCNT TCNQ, TDAE, Anthracene@SWCNT, Metallocenes p- and n- type doping
Estimation of the local filling ratio by TEM Peapods: e.g. C 60 filled SWCNT >90 % filling
Bulk filling ratio, structure in peapods π* C 60 t 1u t 1g C1s Maximum filling ratio with C 60 : 100% no charge transfer to C 60 C 60 - C 60 distance 9.7 Å no C 60 polymer in SWCNT X. Liu et al., Phys. Rev. B 65, 45419 (2002), Synth. Met. 135-136, 715 (2003)
Ferrocene filling of SWCNT and transformation to DWCNT High filling factor after long time annealing in at 150 C
Metallocene filling and chemistry inside SWCNT Metallocenes : TM(C 5 H 5 ) 2, RE(C 5 H 5 ) 3 Metallocene consists of H M C TM: Transition metal atoms RE: Rare-earth metal atoms Planar aromatic ligand: C 5 H 5 0.4 nm Electronic & magnetic & chemical properties varying with metal atoms. ~ 1.4 nm 0.6 nm M = Fe : Ferrocene Most stable metallocene Suitable size to be filled inside SWCNT!!!
Metallocene filling: Filling factor d = 1.4 nm 0.755 nm Assuming a 1D ferrocene chain inside SWCNT (d =1.4 nm) with adjacent ferrocene distance ~ 0.755 nm Fe/C ratio ~ 0.003 Filling factor ~ 35 %
Nanochemistry in 1D: Gain from TEM High resolution Overview 50 nm Ferrocene Iron carbide Iron oxide nanocrystals H. Shiozawa et al. Adv. Mat. (2008)
Ferrocene filled SWCNT and growth of DWCNT Intensity (arb. units) Normalized intensity Fe2p 1/2 Fe2p 3/2 C1s x20 x1 720 710 300 290 Binding energy (ev) M 1 S2 S 1 Ferrocene B Normalized intensity FeCp 2 @NT 600 C, 2h. 1150 C, 1h. Normalized intensity SWCNT FeCp 2 @NT 600 C, 2h. 2.0 1.5 1.0 0.5 0.0 Binding energy (ev) A 725 720 715 710 705 Binding energy (ev) Charge transfer E F =0.1 ev, ~0.2 e - per ferrocene Power law behavior α =0.55 growth of inner tubes at ~ 600 C 1150 C, 1h. 150 200 250 300 350 Raman shift (cm -1 ) H. Shiozawa et al. Adv. Mat. (2008);PRB (2008)
DWCNT/Inner tubes: Gain from high resolution Raman (6,5) (6,4) Normalized intensity (14,6) (16,3) (13,6) (17,3),(15.6) (14,7),(18,1) (16,2), (14,5), (17,0) (13,8),(16,4) (12,9),(11,10) (17,2),(15,5),(18,0) (11,9),(15,4),(10,10) Normalized intensity (14,6),(16,3),(13,7) (17,1),(12,8) Ferrocene S FeCp 2 @NT, 1150 C, 1h. b C 60 @NT, 1150 C, 15h. (6,5) (6,4) C 60 290 300 310 320 330 340 350 360 Raman shift (cm -1 ) Catalytic growth of inner DWCNT differs from non-catalytic! 0.60 0.65 0.70 0.75 Diameter difference (nm) H. Shiozawa et al. Adv. Mat. (2008)
Temperature induced chemical reaction in a SWCNT nanoreactor 1. Ferrocene filling 2. Catalytic growth from Iron carbide particles 3. Iron release via defects Iron oxide nanocrystals MD simulation from S. Maruyama H. Shiozawa et al. Adv. Mat. (2008); PRB (2008)
Ferrocene filling of metallicity sorted SWCNT M. Sauer et al. submitted
PL of ferrocene filled Hipco SWCNT Pristine Opened Filled X. Liu et al. Advanced Functional Materials DOI: 10.1002/adfm.201200224 (2012)
PL of ferrocene filled Hipco SWCNT X. Liu et al. Advanced Functional Materials DOI: 10.1002/adfm.201200224 (2012)
Summary for carbon nanostructures Graphene and SWCNT are ideal 1D and 2D objects Outstanding mechanical and electronic properties Ideal systems to study basic correlation effects even at room temperature!!!!!! Tuneable electronic properties and interactions!!! Interesting for both basic research and applications
.. and now finally to something completely different... It s..
My research group in Vienna: LOW DIMENSIONAL QUANTUM SOLIDS The beauty of instabilities!!! WANTED: PhD students, postdoc eager to unravel the equilibrium ground state dressed by correlations!
Gd@C 82 @SWCNTs Gd@SWCNTs Gd@C 82 @SWCNTs undergo structural transformation upon heat treatment, yielding 1D Gd encapsulated nanowires. P.Ayala et al., Materials Express 1,30 (2011).
Gd@C 82 @SWCNTs Gd@SWCNTs XAS response in the 3d levels of Gd and RESPES. Stronger resonance for Gd nanowires smaller hybridisation P.Ayala et al.,, Materials Express 1,30 (2011).
Combinational doping: Intercalation of peapods Peapods: e.g. C 60 filled SWCNT Filling/Intercalation Sapphire film UHV vapor K,C 60 in-situ filling: UHV evaporation (5x10-10 mbar) K, C 60 at 120/400 C Equilibration 12 h at RT/350 C Annealing at T=650 C
Electronic structure of C 60 peapods: Predictions Calculated DOS Influence of more conduction channels Que PRB 66, 193405 C 60 -LUMO 1/g p2 = 2/g t2-1 α t = 0.46 g t = 0.18 C 60 g p = 0.13, α p = 0.73 Okada et al. PRB 67, 205411;PRL 86, 3835 C 60 -LUMO states near the Fermi level, 4 bands noticeable increase of α expected
Electronic structure of C 60 peapods Intensity (arb. units) σ HOMO-3,4 pristine SWCNT pristine peapods h g +g g =3.5 ev π h u =2.3 ev Intensity (arb. units) log Intensity (arb. units) M 1 S 2 S1 α = 0.55 1 0.1 log Binding energy (ev) 1.5 1.0 0.5 0.0 Binding energy (ev) 9 8 7 6 5 4 3 2 1 0 Binding energy (ev) t 1u C 60 derived molecular orbitals (Mos) observed Binding energies as in bulk C 60, 100% filling Power law α=0-48-0.55 basically no increase!!! -1.0-0.5 0.0 0.5 1.0 H. Rauf et al., PRB 72, 245411 (2005)
Combinational doping: Intercalation of C 60 peapods 1470-6.5 cm -1 /x Competitive charge transfer to SWCNT and C 60 peas! Position A g (2) ( cm -1 ) 1460 1450 1440 1430 Raman: A g (2) of C 60 -x : - charge transfer up to C 60-6 - no line phases: 0.8 <x< 5.3 average charge transfer - charge transfer induced polymerization -6 e 1420 0 1 2 3 4 5 6 (C 60 ) -x -6 K + T. Pichler et al., PRL 87, 267401 (2001), PRB 67, 125416 (2003).
Archetypical 1D system: New electronic, optical and structural properties: Why carbon nanotubes? - Ideal 1D quantumwires - Hard, ductile Nanowires - Molecular Magnets, Spintronics - Protection for reactive elements and molecules - Gas storage, gas sensing - FET, VIAS, optoelectronic - Sources for field emission - Organic superconductors - Li-Ion batteries - Sensors, actuators, NT-yarns, composits, drug delivery.. Ideal tools for nanoelectronics/optics/mechanics/(bio)chemistry