Fluorine-programmed nanozipping to tailored nanographenes on rutile TiO2 surfaces

See allHide authors and affiliations

Science  04 Jan 2019:
Vol. 363, Issue 6422, pp. 57-60
DOI: 10.1126/science.aav4954

Nanographenes on oxides

The growth of nanographene islands and ribbons on metal surfaces can be accomplished on single-crystal metal surfaces through carbon-carbon coupling reactions, but the surfaces of oxides do not assist these reactions. Kolmer et al. show that fluorinated aryl groups can be coupled to form nanographenes on the rutile surface of titanium oxide. The fluorine substitution of the aryl groups was selected so that as the carbon-fluorine bonds were thermally activated, a stepwise process sequentially added aromatic rings around a central aryl group until it was completely substituted.

Science, this issue p. 57


The rational synthesis of nanographenes and carbon nanoribbons directly on nonmetallic surfaces has been an elusive goal for a long time. We report that activation of the carbon (C)–fluorine (F) bond is a reliable and versatile tool enabling intramolecular aryl-aryl coupling directly on metal oxide surfaces. A challenging multistep transformation enabled by C–F bond activation led to a dominolike coupling that yielded tailored nanographenes directly on the rutile titania surface. Because of efficient regioselective zipping, we obtained the target nanographenes from flexible precursors. Fluorine positions in the precursor structure unambiguously dictated the running of the “zipping program,” resulting in the rolling up of oligophenylene chains. The high efficiency of the hydrogen fluoride zipping makes our approach attractive for the rational synthesis of nanographenes and nanoribbons directly on insulating and semiconducting surfaces.

Carbon-based nanostructures synthesized through rational surface-assisted C–C coupling on single-crystal metal surfaces (1, 2) include individual isomers of fullerenes (3, 4) and fullerene fragments (5, 6), the chirality-pure carbon nanotubes (7), atomically precise nanographenes (NGs) (810), and graphene nanoribbons (GNRs) with a well-defined periphery (6, 1115). The consecutive, thermally triggered cyclodehydrogenation of the polycyclic aromatic hydrocarbon (PAH) precursor bearing required C–C connectivity to the target nanostructure represents the key transformation in the on-surface synthesis strategy. The synthesis of hexabenzo[bc,ef,hi,kl,no,qr]coronene (HBC) by Weiss et al. showed that this step can be realized efficiently under ultrahigh vacuum (UHV) conditions on atomically clean metal surfaces (16) (Fig. 1A). The catalytic activity of the metals substantially reduced the activation barrier of the cyclization (8). In 2010, Cai et al. applied a similar strategy to form atomically precise GNRs on the Au(111) surface (11) (Fig. 1B). This discovery paved the way toward the fabrication of complex molecular nanoarchitectures on selected noble metal surfaces (17). However, for most practical applications, a carbon-based nanostructure must be transferred onto insulating or semiconducting surfaces (18, 19).

Fig. 1 Selected examples of on-surface syntheses of NGs and nanoribbons by a bottom-up approach.

(A) First on-surface synthesis of NG HBC (16); (B) rational synthesis of GNRs on Au(111) (11); (C) attempts to perform cyclodehydrogenation on a metal oxide surface (20); (D) first rational on-surface synthesis of NGs on a nonmetallic surface (this work).

An attractive yet challenging way to tackle this problem is the controlled synthesis of carbon nanostructures directly on technologically relevant nonmetallic substrates, such as metal oxide surfaces (2022). However, all reported attempts to perform the cyclization on metal oxides have been unsuccessful because of the lack of catalytic activity in the cyclodehydrogenation process (20) (Fig. 1C). The cyclization of PAH precursors on such surfaces requires high temperatures, which leads to a loss of selectivity. Thus, the rational synthesis of tailored carbon-based nanoarchitectures on metal oxide surfaces requires the development of an alternative cyclization technique. Previously, we have found that intramolecular aryl-aryl coupling can be effectively realized through C–F bond activation on γ-Al2O3 under relatively mild conditions (2327). Further exploration revealed that metal oxides of III and IV groups also displayed activity in cyclodehydrofluorination at elevated temperatures (28). Among them, bulk powders of titanium dioxide activated C–F bonds at 570 K, which made it an attractive candidate for on-surface investigations performed under UHV conditions.

We present the rational on-surface synthesis of NGs on a semiconducting rutile TiO2(011) surface through dominolike HF zipping of programmed fluoroarene precursors (Fig. 1D). The high potential of the approach was demonstrated by a challenging transformation consisting of the formal rolling up of the linear oligophenylene chain around a phenyl moiety, yielding NG HBC (Fig. 1D). In contrast to the commonly used rigid design of precursors, our approach allows the regioselectivity of the cyclization to be unambiguously programmed by F atom positions, providing sufficient flexibility in the design of precursor molecules.

To investigate the HF-zipping process on a rutile surface, two model NGs—namely, DBPP (dibenzo[ij,rst]phenanthro[9,10,1,2-defg]pentaphene) and HBC—were chosen as target compounds (Fig. 2, A and B). DBPP, which can be considered an ultrashort armchair GNR (6-AGNR), represents the model for the on-surface synthesis of GNRs, and HBC is one of the smallest and best-characterized NGs (2933). Both NGs possess easy-to-recognize geometries in scanning tunneling microscopy (STM) images. The required specially “programmed” fluorinated oligophenylenes P1 and P2 were obtained by multistep organic synthesis (for details, see the supplementary materials). The key feature of the cyclodehydrofluorination is the “switchable” activity of the C–F bond. Only C–F bonds with close proximity to a C–H bond displayed activity, whereas peripheral C–F remained completely intact. This reactivity enabled tandem cyclization via HF elimination in a truly dominolike fashion, because each subsequent cyclization step led to the “activation” of one new C–F bond (26). The active H–F pairs for the HF-zipping concept are shown schematically in Fig. 2, A and B.

Fig. 2 HF elimination–based zipping of precursors.

Schemes of (A) DBPP and (B) HBC formation concepts. (C) Three-dimensional (3D) visualization of an STM image (+2 V; 10 pA) of P2 deposited on rutile (011)-(2×1) at RT and annealed at 570 K for 10 min. Note different configurations of molecules. (D) 3D STM image (+2.5 V; 50 pA) obtained after sample annealing at 670 K. Single HBC molecules are adsorbed on newly formed ad-islands with different surface reconstruction (yellow-green). (Inset) STM image (+2 V, 10 pA, 2.5 nm by 2.5 nm) of a single HBC molecule. Scale bars in (C) and (D) are 2 nm.

All on-surface experiments, including low-temperature STM, x-ray photoemission spectroscopy (XPS), and mass spectrometry (MS) studies, were performed in situ under UHV conditions. We deposited precursor molecules by using standard Knudsen cells on the (2×1) reconstructed (011) face of the rutile TiO2 (for details, see the supplementary materials). To thermally induce the transformation of P1, we started with submonolayer deposition of precursor molecules on the substrate kept at room temperature (RT) and then heated the substrate to ~570 K (bulk activation temperature) for 10 min. Under these mild conditions, most P1 molecules desorbed from the surface, leaving almost bare surface terraces with no clear evidence of successful HF elimination (see the supplementary materials). At a higher annealing temperature of ~670 K, particularly flat molecules with the specific rhomboid shape expected for DBPP were observed in our STM experiment. However, because of the appreciable thermal desorption of P1, DBPP molecules were adsorbed only on chemically active sites (step edges or domain boundaries), which complicated the accurate interpretation of their nonuniform STM contrast (see the supplementary materials).

With the larger precursor P2, after deposition onto an RT substrate, P2 molecules were found mostly in globular form on reconstructed terraces of rutile (011) with STM. However, after annealing at ~570 K, the molecules remained on the surface (Fig. 2C). Moreover, we observed a general change in P2 appearance and observed elongated geometries with lengths up to ~2.5 nm. These geometries correspond to different possible configurations of P2 on the surface (see detailed analysis in the supplementary materials), consistent with the expected flexibility of the precursors. This observation points out that the globular geometry, the favorable gas-phase configuration of P2 preserved after deposition, was only metastable on the surface.

To induce the transformation of P2, we heated the rutile (011) substrate to ~670 K for 10 min. Although most of the molecules were found on step edges, some molecules appeared directly on newly formed ad-islands that allowed their high-quality STM visualization (Fig. 2D). The round shape and the size (~1.6-nm full width at half maximum, as shown by STM image cross sections in Fig. 3, A and B) of the molecules are in agreement with the expected HBC model (Fig. 3C), our unoccupied-states STM image simulations (see the supplementary materials), and previous reports (29, 30, 32). The presence of HBC molecules demonstrated the feasibility of the HF-zipping strategy on the rutile surface. The programmed P2 precursor was transformed via five consecutive cyclodehydrofluorination steps followed by the final cyclodehydrogenation into the target NG molecule.

Fig. 3 HF versus H2 elimination.

(A) STM image (+2 V, 50 pA) presenting molecules of HBC and H2-HBC (arrows) formed from P2 after annealing at 670 K. (B) Cross sections along purple and green lines in (A). (C and D) Optimized geometries of (C) HBC and (D) H2-HBC molecules. Gray and white circles correspond to carbon and hydrogen atoms, respectively. (E) High-resolution STM image (+2.7 V, 200 pA) of HBC and H2-HBC on newly formed reconstruction of rutile (011). Scale bars in (C) to (E) are 1 nm.

Among molecules found on ad-islands, some possessed an additional protrusion located beside the center of the molecule, as marked by arrows in Fig. 3A. We attributed these images to molecules of helicenelike H2-HBC, presented in Fig. 3D, in which the final cyclodehydrogenation step was not accomplished. The presence of two H atoms in the fjord region caused deviation from planar geometry, with two corresponding benzene rings oriented out of plane. The high-resolution STM image of an HBC and H2-HBC pair in Fig. 3E confirms the expected position of the protrusion on the molecules. The uncopied-states STM contrast of H2-HBC is also reproduced in our STM image simulation (see the supplementary materials), which is in good agreement with the experimental image. The formation of the H2-HBC intermediate after only the first five cyclization reactions activated via F was rather unexpected, because the last cyclization step should occur spontaneously given the high strain in the fjord region. However, the observation of the H2-HBC intermediate indicates that, on the TiO2 surface, the cyclodehydrofluorination process had a markedly lower activation barrier than cyclodehydrogenation. A similar strategy was in fact completely inefficient on metal surfaces. As we found for the Au(111) surface, catalytic activation of the competitive cyclodehydrogenation processes resulted in undefined molecular structures formed from P2 after their annealing (see results and discussion in the supplementary materials). Thus, metal oxide surface assistance in the cyclodehydrofluorination is the key aspect of the successful HF zipping (28). Although the cyclization mechanism is not fully understood, the most probable scenario includes the C–F bond polarization by the active Ti center, allowing synchronous Friedel-Crafts–like arylation (34, 35).

The STM data show that the HF-zipping strategy was efficient, as many single-target molecules were found locally on the surface. Other than H2-HBC and HBC, no other molecular species were observed on the ad-islands where high-quality STM imaging was possible. To shed more light on the global outcome of the thermally induced HF elimination reaction, we combined the STM data with the XPS chemical analysis. The C1s core-level photoemission spectrum of P2 molecules deposited on the rutile (011) surface at RT (Fig. 4A) showed an asymmetric signal composed of two peaks separated by ~2 eV in binding energy (EB), which correspond to C–C (red line, EB = 284.4 eV) and C–F (green line, EB = 286.3 eV) contributions (36, 37). Figure 4B presents the F1s core-level region. For RT deposition, the single peak observed at EB = 686.8 eV is related to an organofluorine component (13, 36, 37). The absence of a peak at ~684.5 eV suggests the lack of a F–Ti component (38). Thus, the XPS analysis confirmed that the C–F bonds in the precursor molecules were intact after deposition on the surface at RT.

Fig. 4 Chemical analysis.

(A) XPS C1s core-level spectra measured for P2 deposited on rutile (011) at RT (top) and after annealing to 670 K (bottom). arb., arbitrary. (B) XPS F1s core-level spectra measured for P2 deposited on rutile (011) at RT and after annealing to 670 K. The arrow marks the energy of the eventual F–Ti component. (C) Mass spectrometer signals of a TiOF2 molecule (m/z = 101.8) registered during controlled heating of bare and P2-covered rutile TiO2(011) from RT to around 770 K. The main fraction of thermally desorbed TiOF2 molecules from the P2-covered surface is registered for temperatures exceeding 670 K. (D) High-resolution STM image (+2.7 V, 200 pA) presenting new reconstruction of the ad-island with apparent c(2×1) symmetry. (E) Calculated structure of the hydroxylated rutile TiO2(011)-(1×1) surface forming (2×1) reconstruction. Gray, red, and white circles correspond to titanium, oxygen, and hydrogen atoms, respectively. (F) Unoccupied-states STM image simulation of the structure presented in (E). Scale bars in (D) to (F) are 0.5 nm. Purple rectangles mark corresponding unit cells.

After annealing at 670 K for 10 min, the XPS signal from the C1s region consisted of only a single peak at EB = 284.5 eV, corresponding to a C–C component (Fig. 4A), confirming global scission of C–F bonds in P2 molecules caused by efficient HF elimination. The corresponding STM images (Fig. 3, A and E) show that the flat-lying HBC molecules were found on the newly formed reconstructed areas of the surface, forming ad-islands attached to TiO2(011)-(2×1) step edges or domain boundaries. These structures were not observed in a control experiment, where we directly deposited HBC molecules on the rutile (011) and annealed the sample to 670 K (see the supplementary materials). Moreover, other halogenoarenes with bromine (2022) or iodine (21) substituents annealed under similar conditions at rutile titania surfaces did not produce the observed ad-islands.

An obvious hypothesis is that F atoms are building blocks in these new structures. However, Fig. 4B shows that after annealing, the XPS signal from the F1s region was strongly reduced, consisted of only organic fluorine (<10% of the initial intensity), and provided no indication of a F–Ti component at ~684.5 eV (arrow). Lack of F at the surface region motivated us to look for fluorine-containing molecules desorbing from the rutile (011) surface after the HF-zipping reaction by monitoring corresponding MS signals during controlled sample heating from RT up to 770 K. We observed signals only for molecules of TiOF2 (Fig. 4C) and TiOF (see the supplementary materials). Despite the sharp peak located at ~500 K, the main fraction of TiOF2 desorbed when the temperature exceeded 670 K, consistent with our previous experimental observations.

Desorption of TiOF2 points to strong Ti–F covalent bonding of F to the surface during the HF zipping, which may indirectly explain the observed surface reconstruction under newly formed HBC molecules. Adsorbates strongly interacting with the surface are known to induce changes in rutile titania reconstructions, as previously reported for (110) (39) and (011) (40, 41) faces. Recent work by Balajka et al. (41) reported a water-induced reconstruction of a rutile (011) surface to form a bulk-terminated (1×1) face covered with a hydroxyl group overlayer forming (2×1) reconstruction. In the case of a HF elimination reaction, the desorption of TiOF2 species also formally leads to the formation of water molecules, TiO2 + 2HF → TiOF2 + H2O, which then could dissociate and locally create the hydroxyl-rich reconstruction. Close inspection of the ad-island structure performed by high-resolution STM characterization (Fig. 4D) shows bright protrusions forming rows along the [01-1] direction, with neighboring ones shifted by half of the lattice, giving apparent pseudohexagonal c(2×1) symmetry. Results of our density functional theory calculations shown in Fig. 4E present the exact structure of the ad-islands with four hydroxyl groups (two dissociated water molecules) per (2×1) unit cell. Corresponding unoccupied-states STM image simulation (Fig. 4F) confirmed that for high positive bias voltages, this (2×1) surface reconstruction gave the apparent c(2×1) symmetry observed in the experimental data. These results support the rutile titania surface participation in intermediate states of the cyclodehydrofluorination and reflect its crucial role in this process. Our results provide a pathway toward custom-designed sp2 carbon–based nanostructure formation by direct on-surface synthesis methods on technologically relevant semiconducting or insulating surfaces.

Supplementary Materials

Materials and Methods

Supplementary Text

Figs. S1 to S34

Table S1

References (4245)

References and Notes

Acknowledgments: Funding: The research was supported by the Polish Ministry of Science and Higher Education, contract no. 0341/IP3/2016/74. R.Z. acknowledges support received from the National Science Center, Poland (2017/24/T/ST5/00262). A.K.S. and K.A. thank the Deutsche Forschungsgemeinschaft (DFG-SFB 953 “Synthetic Carbon Allotropes” project A6, and AM407) for financial support. Work was partially conducted at the Center for Nanophase Materials Sciences (CNMS), which is a DOE Office of Science User Facility. The research was partially supported from basic and statutory funds of the Jagiellonian University in Krakow provided by the Polish Ministry of Science and Higher Education. Author contributions: K.A. and M.K. conceived the project. K.A. and A.K.S. carried out the precursor synthesis and analysis. M.K. and R.Z. conducted the on-surface synthesis and low-temperature STM, XPS, and MS analyses with support from L.Z., S.G., and M.S. M.E. conducted the computations. M.K. and K.A. prepared the manuscript with feedback from all other authors. Competing interests: None declared. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper or the supplementary materials.
View Abstract

Stay Connected to Science

Navigate This Article