On-surface synthesis of a nitrogen-embedded buckybowl with inverse Stone–Thrower–Wales topology

Curved π-conjugated polycyclic aromatic hydrocarbons, buckybowls, constitute an important class of materials with wide applications in materials science. Heteroatom doping of buckybowls is a viable route to tune their intrinsic physicochemical properties. However, synthesis of heteroatom-doped buckybowls is a challenging task. We report on a combined in-solution and on-surface synthetic strategy toward the fabrication of a buckybowl containing two fused nitrogen-doped pentagonal rings. We employ ultra-high-resolution scanning tunneling microscopy and spectroscopy, in combination with density functional theory calculations to characterize the final compound. The buckybowl contains a unique combination of non-hexagonal rings at its core, identified as the inverse Stone–Thrower–Wales topology, resulting in a distinctive bowl-opening-down conformation of the buckybowl on the surface. Our controlled design of non-alternant, heteroatom-doped polycyclic aromatic frameworks with established bottom-up fabrication techniques opens new opportunities in the synthesis of carbon nanostructures with the perspective of engineering properties of graphene-based devices.


General chemical methods
All reagents and solvents were purchased from commercial sources and were used as received unless otherwise noted. Reagent grade solvents (CH 2 Cl 2 , hexanes) were distilled prior to use. DMF was dried over magnesium sulfate, then distilled and stored under argon. Transformations with moisture and oxygen sensitive compounds were performed under a stream of argon. The reaction progress was monitored by means of thin layer chromatography (TLC), which was performed on aluminium foil plates, covered with Silica gel 60 F 254 (Merck) or Aluminium oxide 60 F 254 (neutral, Merck). Products purification was done by means of column chromatography with Kieselgel 60 (Merck) or Aluminium oxide (Fluka). Occasionally, dry column vacuum chromatography (DCVC) for purification of products obtained was performed using Silica gel Type D 5F. The identity and purity of prepared compounds were proved by 1 H NMR and 13 C NMR spectrometry as well as by MS-spectrometry (via EI-MS or ESI-MS). NMR spectra were measured on Bruker AM 500 MHz, Bruker AM 600 MHz, Varian 600 MHz, Varian 400 MHz or Varian 200 MHz instruments with TMS as internal standard. All chemical shifts are given in ppm. All melting points for crystalline products were measured with automated melting point apparatus EZ-MELT and were given without correction. The absorbance and fluorescence spectra were measured in dichloromethane on Perkin -Elmer Lambda 25 UV/VIS and Hitachi F-7000 respectively.

General procedure for the synthesis of Tetrarylpyrroloyrroles (4a-c)
In a 25 mL round-bottom flask equipped with a reflux condenser and magnetic stir bar, 6 mL glacial acetic acid was placed followed by the addition of aldehyde (6 mmol), 2-aminobiphenyl (6 mmol), and TsOH (0.6 mmol). The mixture was stirred at 90 °C for 30 min. After that time butane-2,3dione (3 mmol) was slowly added via syringe and the resulting mixture was stirred at 90 °C for 3 h. The reaction mixture was then cooled to room temperature. The precipitate of the obtained dye was then filtered off and washed with cooled glacial acetic acid. Recrystallization from AcOEt and drying under vacuum afforded pure product.

General procedure for the synthesis of π-expanded Tetraarylpyrrolopyrroles (5a-c)
To a 20mL sealed tube equipped with magnetic stir bar (flushed with argon prior to use), 8 mL of dry toluene and 0.5 mmol of adequate TAPP were placed. To a dissolved substrate 11 mg (0.05 mmol) of Pd(OAc) 2 , 391 mg (1.2 mmol) of Cs 2 CO 3 and 30 mg (0.11 mmol) of PPh 3 were added.
Reaction was conducted at 120 °C for 3 hours. Then after cooling 8 mL of water was added and resulting mixture was stirred for another 15 min. Two phases were separated, water phase was extracted with ethyl acetate (3 × 10 mL). Organic phases were combined and dried, solvent was evaporated and crude product was washed with copious amount of diethyl ether.

General procedure for the synthesis of 6a
To a 50 mL round-bottom flask, flushed with argon prior to use, and equipped with magnetic stir bar and septum, 10 mL of dry methylene chloride and 0.2 mmol of 5a were placed. To suspended substrate, 4 mmol of iron(III) chloride dissolved in 6 mL of dry nitromethane were added via syringe. Reaction was conducted at room temperature for 30 min. Then 12 mL of water was added, and resulting mixture was stirred for another 15 min. Two phases were separated, water phase was extracted with methylene chloride (3 × 15 mL). Organic phases were combined and dried, solvent was evaporated and crude product was purified by means of flash column chromatography.

General procedure for the synthesis of 6b-c
To a 50 mL round-bottom flask, flushed with argon prior to use, and equipped with magnetic stir bar and septum, 15 mL of dry dichloroethane and 0.2 mmol of adequate π-exp TAPP 5 were placed. To dissolved substrate, 4 mmol of iron(III) chloride dissolved in 6 mL of dry nitromethane were added via syringe. Reaction was conducted at 80 °C for 16 hours. Then 20 mL of water was added and resulting mixture was stirred for another 15 minutes. Two phases were separated, water phase was extracted with methylene chloride (3 x 15 mL). Organic phases were combined and dried, solvent was evaporated and crude product was purified by means of flash column chromatography.

On-surface reaction of 6a on Ag(111)
Since complete ring closure of 6a to form 8 is not observed on the Au(111) surface, we tried to induce surface-assisted ring closure on Ag(111), which is a more catalytically active substrate. After sublimation of precursor 6a under ultra-high vacuum (UHV) conditions onto an atomically clean Ag(111) surface held at room temperature, large-scale STM images ( Supplementary Fig. 19a) show the presence of long self-assembled chains and honeycomb islands of 6a. Supplementary  Figures 19b and 19c show high-resolution STM images of the honeycomb islands and chains, respectively. After annealing to 300°C, large-scale STM images ( Supplementary Fig. 19d) reveal the presence of mostly chains of 6a composed of only a few units, few half-closed species (7a) and some unknown species. Supplementary Figure 19e shows these structures in detail. Few units corresponding to 7a are highlighted with red circles. In addition, the unknown species (highlighted with a white circle) presents a much lower apparent height compared to 6a/7a. The identity of these species is not clear, and we propose that they represent decomposed molecules.

On-surface reaction of 6a on Cu(111)
After sublimation of precursor 6a under UHV conditions onto an atomically clean Cu(111) surface held at room temperature, large-scale STM images ( Supplementary Fig. 20a) show the presence of self-assembled linear chains presenting a rod-like appearance, as well as individual molecules. Supplementary Figure 20b shows a high-resolution STM image of the self-assembled chains. After annealing to 260°C, intermolecular covalent coupling and decomposition dominate, and ill-defined clusters prevail on the surface. We did not find either 7a or 8 on the surface (Supplementary Fig.  20c). We conclude that the Cu(111) surface is too active, and promotes decomposition of the molecules.

Bowl-opening-up species -8'
Interestingly, we find a minor amount of a molecular species which first appears after the annealing step to 270°C. Referred to as 8', its appearance changes with the tunneling bias voltage. Notably, at positive bias values, the molecule presents a characteristic appearance wherein the periphery of the molecule shows a uniform contrast with the center appearing noticeably dark (Supplementary Fig.  22b and 22c). Similar characteristic feature in STM images of surface-adsorbed buckybowls was reported which corresponded to a bowl-opening-up configuration of the respective buckybowls 1,2 . Furthermore, through UHR-STM images of 8' (Supplementary Fig. 22d and 22e), we clearly resolve the peripheral hexagonal rings, with considerable loss of contrast on the inner hexagonal and heptagonal rings, and no contrast on the pentagonal rings. These observations strongly hint towards 8' being the bowl-opening-up variant of 8. To aid our understanding of the identity of 8', we constructed the structure of the bowl-opening-up variant of 8 on Au (111) (shown in Supplementary Fig. 22f and 22g). The corresponding DFT simulated STM images for this conformation ( Supplementary Fig. 22h and 22i) show striking agreement with the experimental STM images at both polarities of the tunneling bias voltage, thus verifying our speculation that 8' is indeed the bowl-opening-up variant of 8. The factors stabilizing this conformation on the surface are presently not known.  Fig. 22) suggests that the empirical van der Waals parameter employed in our study is stronger than the optimum, but nevertheless provides an overall faithful and correct trend. Two different starting geometries (i.e. bowl-opening-up and tilted bowlopening-up) were chosen for the geometry optimization procedures in order to make sure that the choice of initial geometry does not critically influence the final optimized result. Note that while the bowl-inversion process in the gas-phase has a finite energy barrier of 0.35 eV ( Supplementary Fig. 25), geometry optimization procedures on Au(111) resulting in a final bowlopening-down geometry from initial bowl-opening-up geometries show no energy barriers. This highlights the crucial role of molecule-substrate interaction in determining the conformation of 8 on the surface.

Supplementary
of 8 to adopt a bowl-opening-down conformation on the surface. We note that since the bowlopening-up conformation is not a stable conformation on the surface ( Supplementary Fig. 26), it is not possible to directly compare the relative change in vdW energy between the two conformations on the surface. Our qualitative argument serves to illustrate that the driving force for 8 to adopt a bowl-opening-down conformation is the maximization of vdW interaction between 8 and the underlying Au(111) substrate.