First principles Simulations for fullerenes and Atomic Force Microscopy: Reactivity and Manipulation at the atomic scale.

Transcription

1 First principles Simulations for fullerenes and Atomic Force Microscopy: Reactivity and Manipulation at the atomic scale. Rubén Pérez Nanomechanics & SPM Theory Group Departamento de Física Teórica de la Materia Condensada Projects supported by RES in BSC, October 2008

2 Outline 1. Fullerenes from Aromatic Precursors by Surface-catalysed Cyclodehydrogenation Nature 454, 865 (2008) 2. Complex Patterning by Vertical Interchange Atom Manipulation Using Atomic Force Microscopy Science 322, 413 (2008)

3 Fullerenes from Aromatic Precursors by Surface-catalysed Cyclodehydrogenation Nature 454, 865 (2008) Ab initio Simulations: UAM: G. Biddau, M. Basanta, J. Ortega, R. Perez Surface Science: ICMM (CSIC): G. Otero,C. Sanchez-Sanchez, R. Caillard, M.F. Lopez, C. Rogero, F.J. Palomares, J. Mendez, J.A. Martin-Gago Chemistry: ICMM (CSIC): B. Gomez-Lor ICIQ: N. Cabello, A.M. Echevarren

4 Objective and motivations OBJECTIVE: Study and efficient synthesis fullerenes and triazafullerenes C57N3 C60 MOTIVATIONS: microelectronics, superconductivity, corrosion resistence, non linear optics, organic ferromagnetism... Available synthesis methods: Fullerenes: - Graphite Vaporization (uncontrolled) - Through dehydrogenation1 (low efficiency) 1] Scotts et al., Science 295, (2002) Triazafulleres: none

5 The Process VACUUM THERMAL EVAPORATION Cyclodehydrogenation FULLERENE AND TRIAZAFULLERENES Adsorption process ANNEALING (750 K) Efficiency ~100%

6 Experimental and Theoretical evidences PAHs DFT computational details: 1) C60H30 2) C57N3H33 Both Local Orbital (FIREBALL) and plane wave (CASTEP) abinitio methods have been used in LDA and GGA approximation Ideal conditions: Both are markedly twisted, Δz: 0.08 Å (polyarene 1), 0.09 Å (heteropolyarene 2).Twisting angle: 20.5º (1) and 30.5º (2) 1 2

7 Experimental and Theoretical evidences PAHs: Adsorption Experimental STM molecular shape: three wings (molecules do not fragment and maintain almost planar configuration) lobes (DFT calculations associate to external ring).

8 Experimental and Theoretical evidences PAHs: Adsorption Theoretical: As from STM images, energy landscape is complicated, many possible adsorption configurations. Strong variation of molecules height (Δz varies from 0.08 Ang to 0.02 Ang) Tendency to maxime the number of C-Pt bonds.

9 Experimental and Theoretical evidences Apparent height of 0.14 nm Base length of 2.2 nm No dependence on bias voltage. DFT Molecular Orbitals, confirm STM height and lengths. Molecular Orbitals of PAHs strongly hybridize with surface, resulting in a broad DOS at Fermi level ~11 Angstrom ~1.3 Angstrom

10 Experimental and Theoretical evidences ANNEALING 750 K After annealing, molecules have: height is 0.4 nm and diameter of 1.35 nm The low height indicates a strong covalent adsorption. Are they really fullerenes (triazafullerene)? 1) These values are compatible with other fullerenes adsorption datas on Pt, Si and Au. 2) We adsorbed in same experimental condition, commercial fullerenes and no difference are observed. For triazafullerene, XPS 1 peak, shows 3)

11 Experimental and Theoretical evidences After annealing, molecules have: height is 0.4 nm and diameter of 1.35 nm The low height indicates a strong covalent adsorption. Are they really fullerenes (triazafullerene)? 1)These values are compatible with other fullerenes adsorption datas on Pt, Si and Au. 2)We adsorbed in same experimental condition, commercial fullerenes and no difference are observed. 3)For triazafullerene, XPS 1 peak, shows

12 Experimental and Theoretical evidences After annealing, molecules have: height is 0.4 nm and diameter of 1.35 nm The low height indicates a strong covalent adsorption. Are they really fullerenes (triazafullerene)? 1) These values are compatible with other fullerenes adsorption datas on Pt, Si and Au. 2) We adsorbed in same experimental condition, commercial fullerenes (2) and no difference are observed. For triazafullerene, XPS 1 peak, shows 3)

13 Experimental and Theoretical evidences After annealing, molecules have: height is 0.4 nm and diameter of 1.35 nm The low height indicates a strong covalent adsorption. Are they really fullerenes (triazafullerene)? 1) These values are compatible with other fullerenes adsorption datas on Pt, Si and Au. 2) We adsorbed in same experimental condition, commercial fullerenes and no differences are observed. 3) For triazafullerene, XPS 1st peak, shows the2 nitrogen bonded with nd carbon with sp hydridization, the 2 can be attributed to nitrogen of the carbon cage interacting with platinum surface. DFT calculations confirm this orientation.

14 Annealing and Cyclodehydrogenation MASS SPECTROMETRY ON DEUTERATED PRECURSORS Thermal annealing (750 K) produces a serie of dehydrogenation reactions. To exclude that activation is due to the tip, dehydrogenation signal in a mass spectrometer has been detected as a function of annealing temperature. To avoid artefacts of H desorption from apparatus, deuterated precursors have been synthesized. The mass spectrum shows: 1) No molecular shape modification dehydrogenation under 400 K or 2) Between 400 and 600 K we could have partial dehydrogenation as asymmetrical molecules are observed 3) Rapid desorption about 700K 4) Formation of HD from H2 and D2 is enhanced by the strong interaction molecule-surface

15 Annealing and Cyclodehydrogenation DFT CHARACTERIZATION Adsorbed dehydrogenation DFT calculations are computationally impossible, by the way the theoretical calculation for the free molecules provide evidence that dehydrogenation induces a strong tendency towards cyclization. 1) Removal of central H atoms, triggers structural and electronical change to induce wings to approach each other to form a hydrogenated semi-cage 2) A NEB systematic study of fully dehydrogenated mocule via ab-initio methods, shows that the different possible paths from planar precursors to fullerene cage, has barries less than 0.3 ev

16 Annealing and Cyclodehydrogenation DFT CHARACTERIZATION Adsorbed dehydrogenation DFT calculations are computationally impossible, by the way the theoretical calculation for the free molecules provide evidence that dehydrogenation induces a strong tendency towards cyclization. 1) Removal of central H atoms, triggers structural and electronical change to induce wings to approach each other to form a hydrogenated semi-cage 2) A NEB systematic study of fully dehydrogenated mocule via ab-initio methods, shows that the different possible paths from planar precursors to fullerene cage, have barriers smaller than 0.3 ev

17 Annealing and Cyclodehydrogenation

18 The Process

19 Conclusions and Future Developments OUR GOALS: In this work we have shown a very efficient synthesis process for fullerenes and triazafullerenes We have characterized adsorptions of PAHs, fullerenes and triazafullerenes in different adsorption sites, providing the first ab-initio calculations for all the systems studied. We have described the molecular orbital shape of molecules, characterizing the structural (height) and electronical contributions to DOS. Theoretical and experimental STM images are in good agreements Fullerenes from Aromatic Precursors by Surface-catalysed Cyclodehydrogenation NEXT: 454, 865 (June 2008) A wide range of possible heterofullerene and endohedral fullerenes could be obtained Surface catalysed cyclodehydrogenation could be proposed for other metallic surfaces Nature

20 Complex Patterning by Vertical Interchange Atom Manipulation Using Atomic Force Microscopy Science 322, 413 (2008) Ab initio Simulations: UAM: P. Pou, R. Perez FZU(Prague): P. Jelinek AFM Experiments: Osaka University:Y. Sugimoto, M. Abe, S. Morita NIMS (Tsukuba, Japan): O. Custance

21 ATOMIC FORCE MICROSCOPY (AFM) G. Binnig, C. Gerber & C. Quate, PRL 56 (1986) 930 2nd most cited PRL: citations!!!

22 Dynamic AFM

23 Dynamic AFM: Our Goal Why changes observed in the dynamic properties of a vibrating cantilever with a tip that interacts with a surface make possible to: AM-dAFM Obtain molecular resolution images of biological samples in ambient conditions. Resolve atomic-scale defects in UHV. FM-dAFM R. García and R. Pérez, Surf. Sci. Rep. 47, 197 (2002)

24 Tip-sample Interaction: FV + FvdW + Fchem Fchem wfs overlap Sensitivity to Short-Range Forces? Exp: A = 340 Å!!!, R = 40 Å Weak singularity at turning points!!!!

25 Dynamic Force Spectroscopy: Access to Fts Ad2 Ad1 Re1 Ad2 Ad1 H3 Ad2 Ad1 Inversion algorithms 0,4 Short-range Force (nn) 0,2 U. Durig, APL 76, 1203 (2000) 0,0-0,2-0,4 F.J. Giessibl, APL 78, 123 (2001) -0,6-0,8 Exp. Adatom 2 Exp. Restatom 1 Exp. H3-1,0-1,2-1,4-1,6-1, Tip-surface Distance (Å) 8 9 J. E. Sader & S. P. Jarvis, APL 84, 1801 SR(2004). forces amenable to ab initio calculations

26 Recent developments in FM-AFM N. Oyabu et al. Phys. Rev. Lett. 96, (2006). Ad2 1.4 Disspation (ev per Cycle) 1. DISSIPATION: Characterizing the tip structure and identifying a dissipation channel due to single atomic contact adhesion. 1.6 Ad1 Ad1 Re1 Re2 H3 Ad2 1.2 Ad1 Ad Exp. Adatom 2 Exp. Restatom 1 Exp. H Tip-surface Distance (Å) 2. IMAGING: changes in topography: access to the real surface structure? Y. Sugimoto et al Phys. Rev. B 73, (2006) fn m -7.3 fn m -8.4 fn m 3. CHEMICAL IDENTIFICATION: based on the relative interaction ratio of the maximum attractive force measured by dynamic force spectroscopy Y. Sugimoto et al Nature 446, 64 (2007).

27 Recent developments in FM-AFM Understanding RT DFM-based single-atom manipulation 4. LATERAL MANIPULATION: Si vacancy on Si(111)-7x7 (tip assisted thermal hopping) Y. Sugimoto et al. Phys. Rev. Lett. 98, (2007). 5. VERTICAL MANIPULATION: Tip/sample exchange of atoms on Sn/Si(111) Y. Sugimoto et al., Science 322, 413 (2008).

28 Interchange vertical manipulation: Si Sn Tip positioning over a selected Si atom Sn Si Sn f=-7.3 Hz Atom tracking technique Lateral precision: ± 0.1 Å M. Abe, Y. Sugimoto, O. Custance, and S. Morita, Appl. Phys. Lett. 87 (2005) M. Abe, Y. Sugimoto, O. Custance, and S. Morita, Nanotechnology 16 (2005) 3029.

29 Interchange vertical manipulation: Si Sn Tip approach toward the Si atom Sn Sn Si Sn SiSn Sn f=-7.3 Hz Ref special shape jump Z [Å] γ [ fn m] f [Hz] - 10 Typical f-z curve

30 Interchange vertical manipulation: Si Sn Tip retraction from the surface Sn Sn Si Sn SiSn Sn f=-7.3 Hz f=-7.3 Hz The Si atom on the surface was special shape jump Z [Å] γ [ fn m] - 10 f [Hz] Sn Sn Sn interchanged with a Sn atom at the tip apex. The contrast of the images before and after manipulation are almost same.

31 Reproducibility Si Sn f [Hz] Z [Å] Sn Si γ [ fn m] f [Hz] - 15 γ [ fn m] Z [Å] Si Sn Z [Å] Sn Si Z [Å] γ [ fn m] - 30 f [Hz] f [Hz] - 15 γ [ fn m] -5-10

32 Mechanism: Vertical Interchange Vertical manipulation as a combination of mechanical and thermal process. Formation of a characteristic dimer structure along deformation path (energetically stable). Atomic rearrangements reflects in energy & forces discontinuities. Local character of the tip-sample deformation. Temperature effect not included in the simulation.

33 Mechanism: Energy Barriers Explore possible dimer rearrangement due to thermal & mechanical movement Complicated configuration space; only limited configurational space explored Estimated energy barriers ~ 0.4 ev (Nudged elastic band calculation). Dependence on the tip-sample distance, tip elasticity & structure, temperature...

34 Nanostructuring by interchange vertical manipulation Atomic dip-pen nanolithography

35 Atomic dip-pen nanolithography 11 Interchange vertical manipulations + 1 Interchange lateral manipulation The construction time was reduced to 1.5 hours. Sn Si

36 Conclusions Single-atom manipulations: atomistic insight into these processes. Lateral Si-vacancy manipulation Significant reduction of activation energy due to the tip proximity Operating at the attractive force regime Y. Sugimoto et al, Phys. Rev. Lett. 98, (2007) Vertical Si/Sn-exchange manipulation New manipulation method: Interchange vertical manipulation Characteristic mechanical deformation: the hybrid tip-surface dimer structure Operating at the repulsive force regime Combination of mechanical and thermal processes Y. Sugimoto et al, Science 322, 413 (2008).

37 Acknowledgements & Funding MEC (Spain): Projects MAT , NAN C10-07 EU FP-6: STREP project FORCETOOL (NMP4-CT ) RES (Computing): Magerit (CesViMa) & Mare Nostrum (BSC)