Supporting Information

Transcription

1 Supporting nforation Ultrafast Bidirectional Charge Transport and Electron Decoherence At Molecule/Surfaces nterfaces: A Coparison of Gold, Graphene and Graphene Nanoribbon Surfaces Olgun Adak 1, Gregor Kladnik 2,3, Gregor Bavdek 4, Albano Cossaro 5, Alberto Morgante* 3,5, Dean Cvetko 2,5*, Latha Venkataraan 1,6* 1 Departent of Applied Physics and Applied Matheatics, Colubia University, New York, NY 2 Faculty of Matheatics and Physics, University of Ljubljana, Ljubljana, Slovenia 3 Departent of Physics, University of Trieste, Trieste, taly 4 Faculty of Education University of Ljubljana, Ljubljana, Slovenia 5 CNR-OM Laboratorio Nazionale TASC, Basovizza SS-14, k 163.5, Trieste, taly 6 Departent of Cheistry, Colubia University eail: cvetko@io.cnr.it, lv2117@colubia.edu, organte@io.cnr.it Contents: 1. Measureent Details and Saple Preparation 2. XPS and NEXAFS Measureents 3. DFT Calculations for NEXAFS spectra 4. The RPES Measureents 5. The RPES Core-Hole-Clock Method with Charge Transfer fro the Substrate 6. References 1

2 1. Measureent Details and Saple Preparation The experients were conducted at the ALOSA bealine of the Elettra Synchrotron in Trieste. The easureent chaber was aintained at an ultrahigh-vacuu at pressures of bar. 1-3 The 4,4 -bipyridine (BP) olecules were purchased fro Siga-Aldrich (purity >99%) and used without further purification. The olecules were placed in a Pyrex cell and connected to the pre-chaber through a leak valve. The Au(111) surface was prepared by cycles of Ar+ sputtering and annealing at 400 K. X- ray photoeission spectroscopy (XPS) was used to check for any cheical ipurities (O, N, and C). The BP cell was heated to several tens of degrees above roo teperature and BP was leaked into a pre-chaber to aintain a partial BP pressure of 10 7 bar for 4-5 inutes. The saple was cooled to -65 C for the ultilayer phase, and it was kept at roo teperature for the onolayer phase. The epitaxial graphene on Ni(111) surface was prepared by repeatedly sputtering (Ar+, 2keV) and annealing (900K) the Ni(111) surface. XPS and Heliu ato scattering were used to check for the presence of any cheical ipurities and to probe the surface order. Epitaxial graphene was prepared via catalytic dissociation of ethylene on Ni(111) surface. An ethylene partial pressure of 10-6 bar was aintained for 60 inutes while keeping the surface at 850K. 4 The graphene fil was probed using XPS and ultraviolet photoeission spectroscopy with He line at 40.8 ev. Exaination of the graphene π-band botto, close to Γ, was used to identify the graphene layer as epitaxial graphene. 4 The epitaxial graphene/ni(111) saple was exposed to abient conditions before being transferred to the ALOSA chaber. The saple was annealed 2

3 in the ALOSA chaber at 500 K to recover the pristine graphene. XPS, valance band photoeission spectroscopy and near edge X-ray absorption fine structure (NEXAFS) spectroscopy easureents were further perfored to characterize the epitaxial graphene/ni(111) saple. To create the BP self-liiting onolayer fil, a partial BP pressure of 10 7 bar was aintained in the pre-chaber for 3 inutes, while the saple was cooled to - 50 C. Graphene nanoribbons (GNR) were fored depositing 10,10 -dibroo-9,9 -biantryl (AOKBO, 98+% purity) on the cleaned Au(111) saple aintained at 200C to ensure saturating the Au(111) surface with the GNR precursor. This surface was then annealed to 400C to create GNRs as described previously. 5 XPS and NEXAFS easureents were perfored to characterize the GNR fil on the Au(111) surface. To create the BP onolayer fil, a partial BP pressure of 10 7 bar was aintained in the chaber for 5 inutes while the saple was cooled to -45 C. 2. X-ray Photoeission Spectroscopy and Near Edge X-ray Absorption Fine Structure Spectroscopy Measureents X-ray photoeission spectroscopy (XPS) easureents were perfored at the ALOSA bealine with the X-ray bea at grazing-incidence (4 ) to the saple. Photoelectrons fro the saple were collected at the noral to the surface using a heispherical electron analyzer with an acceptance angle of 2, and an overall energy resolution of ~ 0.2 ev. The energy scale for XPS spectra was calibrated by aligning the Au 4f 7/2 peak to a binding energy of 84.0 ev for the Au(111) and GNR on Au(111) easureents. For the epitaxial graphene on Ni(111), XPS spectra were aligned to the Feri level. 3

4 Near edge X-ray absorption fine structure (NEXAFS) easureents were perfored on the nitrogen K-edge by sweeping the incident photon energy fro 396 ev to 420 ev in steps of 0.1 ev. The photon incidence angle was set to 6. Spectra were acquired using a channeltron detector with a wide acceptance angle in the partial electron yield ode and a high pass filter with cutoff energy set to 370 ev. The photon flux was onitored on the last optical eleent along the bea path. The saple noral was oriented either parallel (p-pol) or perpendicular (s-pol) to the light polarization (s-pol). The relative intensity of the NEXAFS signal in s-pol and p-pol for the N1s to LUMO transition was used to obtain the orientation of the aroatic ring relative to the surface. The angle θ of the ring to the surface is deterined as tan! = 2!! /!! where s and p are the intensities of the LUMO NEXAFS peak fro the s-pol and p-pol spectra respectively. 3. DFT Calculations for NEXAFS Spectra We calculate the nitrogen K-edge spectra using GPAW, a grid-based real-space projectoraugented-wave (PAW) code, with the BLYP exchange-correlation functional 6, 7. For these siulations, isolated olecules were first relaxed to their optiized geoetries. Default grid spacings and convergence thresholds were eployed. All NEXAFS calculations were perfored using the half-core-hole approxiation 8. The absolute energy scale was deterined by perforing a delta Kohn-Sha calculation and shifting the calculated spectru using the calculated total energy difference between the ground state and the first core excited state. 4

5 S Figure 1. The easured agic-angle N K-edge NEXAFS spectru (blue) for bipyridine on Au(111) copared with the calculated absorption spectra (dashed red) in the transition potential approxiation. The calculated peaks with the highest transition probabilities are indicated and assigned accordingly. 4. The RPES Measureents The RPES spectra at the nitrogen K-edge were obtained by taking XPS scans spanning 0-60 ev binding energy range with the photon energy tuned between 394 ev and 423 ev in steps of 0.2 ev and 0.5 ev for all surfaces studied. RPES easureents for charge transfer tie calculations were perfored in agic angle conditions, with the light polarization and the electron analyzer at 54.7 with respect to the surface noral. This yielded a RPES signal that was independent of the olecular orientation on the saple. For the angle-dependent RPES easureents, the angle between the photon polarization and surface noral set to the values stated in the ain text while the electron analyzer was always collinear with the photon polarization. These RPES data were then noralized and analyzed following previously published procedures. 9, 10 Briefly, 5

6 the photon flux on the last optical eleent along the bea path was easured and used to noralize and calibrate the RPES scans for any fluctuations in photon intensity and energy. The energy scale for XPS spectru was calibrated as detailed above. The non-resonant photoeission spectru was obtained fro the XPS scan at 395 ev and subtracted fro the entire RPES spectra. The RPES line scans were then further noralized by the overall Auger intensity which scales with the absorption cross-section at a given photon energy. S Figure 2. Nitrogen K-edge RPES aps of (A) BP ultilayer on Au(111) (B) BP onolayer on Au(111), (C) BP onolayer on epitaxial graphene and (D) BP onolayer on GNR. The LUMO* resonance scans are taken along the dashed black lines. The dashed orange lines indicate the energy used for the scans above the ionization edge. 6

7 n S Figure 2, we show nitrogen K-edge RPES aps of a BP ultilayer on Au(111) and a BP onolayer on Au(111), epitaxial graphene and GNR. We see the N1s to LUMO transition signature around photon energy of 399 ev. Fro these aps, we obtain the line scans that are shown in Figure 3 of the ain text. The core-hole clock ethod described below is then used to deterine the charge transfer tie and the fraction of the LUMO* that drops below the Feri level in the onolayer systes. 5. The RPES Core-Hole-Clock Method with Charge Transfer fro the Substrate n this section, we present the extension of the standard core-hole clock ethod 11 to obtain the bidirectional charge transfer tie and the fraction of the LUMO* that drops below the Feri level fro the RPES easureents. n the core-hole clock ethod, a core electron is photoexcited to the LUMO, leaving a core-hole on the olecule; and the energy distribution of the subsequent electron eission is easured. The process relevant to the core-hole clock ethod is the filling of the core-hole and the subsequent eission of an electron, leaving the LUMO epty; this is denoted as the participator decay. n a coupled syste, the participator decay gets quenched if the charge transfer to the substrate occurs within the core-hole life-tie. Therefore, by coparing the participator decay intensity in the isolated and the coupled syste, the charge transfer tie in the coupled syste can be deterined as has been done before. 11 n the easureents presented here, when a core-hole is present, the LUMO* is partially below the Feri level and can thus get occupied due to a charge transfer fro the substrate. 7

8 To deterine the bi-directional charge transfer tie, one needs to consider two types of decay process that will contribute to the participator decay intensity. First, there is a participator decay that occurs fro the LUMO* electron. Additionally, there is an Auger type participator decay (fro the fraction of the LUMO* that is below Feri) that can occur after charge transfer fro the LUMO* to the substrate. Denoting the participator decay intensity as o in the isolated syste and as c in the coupled syste, we can get a relation between o and c as: = τ + x CT CH c o o τch + τct τch + τct τ (1) where τ CH is the core-hole lifetie, τ CT is the charge transfer tie, x is the fraction of the LUMO* below the Feri level. The first ter in Eq. 1 is the standard expression that has been used before and is attributed to the photoeission type participator decay. 11 The second ter is due to the Auger type participator decay that can occur only if the LUMO* is partly below Feri and charge transfer fro the LUMO* to the substrate has occurred. The process giving rise to the second ter in Eq. 1 is identical to the Auger type participator decay occurring after the charge transfer fro the substrate to LUMO* following an excitation of a core electron above the ionization edge. We denote this process as the super-participator decay and its intensity ( s ) is given by s τ CH = ox τ + τ CH BT. (2) Here, τ BT is the bi-directional charge transfer tie. 8

9 Since the charge transfer tie in both directions is equal, Eq. 1 and 2 can be solved to find x and τ CT which yields τ CT = τ CH f (1 β) 1 f (1 β). (3) f β x = (4) 1 f (1 β) where f is c / o and β is s / c. The equations above are valid when the light polarization is perpendicular to the nodal plane of the nodal plane of the olecular π-syste. Therefore, we need to account for the fact that the olecular π-syste akes a sall angle with the substrate, as deterined fro NEXAFS easureents presented in the ain text. We also need to consider the light polarization when coparing the Auger type super-participator intensities with photoeission type participator intensity since there is a clear angular dependence in the latter. The photoeission type participator decay intensity is proportional to (ε n) 2 where ε is the polarization unit vector and n is the vector noral to the nodal plane of the olecular π- syste. We calculate the (ε n) 2 as a function of ε and n assuing that there is no aziuthal dependence on the olecular orientation on the surface. We use ε = sin( θ) x+ cos( θ) z (5) n = sinα cos(φ π / 2) x + sinα sin(φ π / 2) y + cosα z (6) 9

10 where θ is the polar angle between the surface noral and the polarization unit vector, α is the polar angle between the surface noral and the noral to the olecular π-syste, and φ is the aziuthal angle of the olecular π-syste. Averaging (ε n) 2 over φ, and using Eq. 5 and Eq. 6 we get: sin θsin α = + α (7) ( ε n) cos θ cos Since easureents are ade at the agic angle (θ =54.7 ), cos 2 θ=sin 2 θ/2, and thus the average (ε n) 2 is independent of α. We obtain: τ c = CT o + o τ CH +τ CT cos 2 θ x τ CH (8) τ CH +τ CT s = o cos 2 θ x τ CH τ CH +τ BT (9) where yields, c, s and o are the easured intensities at the agic angle. Solving Eq. 8 and Eq. 9 τ CT = τ CH f (1 β) 1 f (1 β) (10) x = cos2 θ f β 1 f (1 β). (11) where f is c / o and β is s / c. 10

11 6. References 1. Floreano, L.; Cossaro, A.; Gotter, R.; Verdini, A.; Bavdek, G.; Evangelista, F.; Ruocco, A.; Morgante, A.; Cvetko, D. J. Phys. Che. C, 2008, 112, (29), Cvetko, D.; Lausi, A.; Morgante, A.; Toasini, F.; Prince, K. C.; Sastry, M. Measureent Science & Technology, 1992, 3, (10), Vilercati, P.; Cvetko, D.; Cossaro, A.; Morgante, A. Surf. Sci., 2009, 603, (10-12), Patera, L. L.; Africh, C.; Weatherup, R. S.; Blue, R.; Bhardwaj, S.; Castellarin-Cudia, C.; Knop- Gericke, A.; Schloegl, R.; Coelli, G.; Hofann, S.; Cepek, C. Acs Nano, 2013, 7, (9), Batra, A.; Cvetko, D.; Kladnik, G.; Adak, O.; Cardoso, C.; Ferretti, A.; Prezzi, D.; Molinari, E.; Morgante, A.; Venkataraan, L. Cheical Science, 2014, 5, Enkovaara, J.; Rostgaard, C.; Mortensen, J. J.; Chen, J.; Dulak, M.; Ferrighi, L.; Gavnholt, J.; Glinsvad, C.; Haikola, V.; Hansen, H. A.; Kristoffersen, H. H.; Kuisa, M.; Larsen, A. H.; Lehtovaara, L.; Ljungberg, M.; Lopez-Acevedo, O.; Moses, P. G.; Ojanen, J.; Olsen, T.; Petzold, V.; Roero, N. A.; Staushol-Moller, J.; Strange, M.; Tritsaris, G. A.; Vanin, M.; Walter, M.; Haer, B.; Hakkinen, H.; Madsen, G. K. H.; Nieinen, R. M.; Norskov, J.; Puska, M.; Rantala, T. T.; Schiotz, J.; Thygesen, K. S.; Jacobsen, K. W. J. Phys.:Cond. Mat., 2010, 22, (25). 7. Ljungberg, M. P.; Mortensen, J. J.; Pettersson, L. G. M. Journal of Electron Spectroscopy and Related Phenoena, 2011, 184, (8-10), Cavalleri, M.; Odelius, M.; Nordlund, D.; Nilsson, A.; Pettersson, L. G. M. Phys. Che. Che. Phys., 2005, 7, (15), Batra, A.; Kladnik, G.; Vazquez, H.; Meisner, J. S.; Floreano, L.; Nuckolls, C.; Cvetko, D.; Morgante, A.; Venkataraan, L. Nat. Coun., 2012, 3, Kladnik, G.; Cvetko, D.; Batra, A.; Dell'Angela, M.; Cossaro, A.; Kaenetska, M.; Venkataraan, L.; Morgante, A. J. Phys. Che. C, 2013, 117, (32), Bruhwiler, P. A.; Karis, O.; Mårtensson, N. Rev. Mod. Phys., 2002, 74, (3),