Combinatorial design of molecular seeds for chirality-controlled synthesis of single-walled carbon nanotubes

The chirality-controlled synthesis of single-walled carbon nanotubes (SWCNTs) is a major challenge facing current nanomaterials science. The surface-assisted bottom-up fabrication from unimolecular CNT seeds (precursors), which unambiguously predefine the chirality of the tube during the growth, appears to be the most promising approach. This strategy opens a venue towards controlled synthesis of CNTs of virtually any possible chirality by applying properly designed precursor molecules. However, synthetic access to the required precursor molecules remains practically unexplored because of their complex structure. Here, we report a general strategy for the synthesis of molecular seeds for the controlled growth of SWCNTs possessing virtually any desired chirality by combinatorial multi-segmental assembly. The suggested combinatorial approach allows facile assembly of complex CNT precursors (with up to 100 carbon atoms immobilized at strictly predefined positions) just in one single step from complementary segments. The feasibility of the approach is demonstrated on the synthesis of the precursor molecules for 21 different SWCNT chiralities utilizing just three relatively simple building blocks.


Supplementary Methods
All chemicals were purchased from Sigma-Aldrich or ChemPUR and used without any further purification. Reactions that require an inert atmosphere were degassed by three cycles (3x 1 min) of sonication under membrane-pump vacuum, followed by the exchange of the atmosphere with nitrogen or argon.
APPI mass spectra were recorded on a quadrupole time-of-flight (QqToF) mass spectrometer, the Bruker MAXIS. MALDI-TOF mass spectra were recorded with a Shimadzu Biotech Axima Confidence and a Bruker Reflex III.

STM experiments
The STM experiments were performed in ultrahigh vacuum conditions (base pressure 10 -11 mbar) using a preparation chamber linked to a ScientaOmicron LT-STM/AFM operating at 5K. Platinum (111) crystals (MaTeck GmbH) were prepared by cycles of Ar+ sputtering (1keV) and thermal annealing (1100K), followed by a final flash above 1300K for 1 min. The precursor molecules were thermally evaporated (500°C) using a 6-fold organic evaporator (Mantis GmbH), while the substrate was held at room temperature (or higher as indicated). Images were acquired with ScientaOmicron's Matrix software and subsequently analyzed with the MatrixFileReader XOP package (bytephysics) for Igor Pro (Wavemetrics).

Experimental Part
1-bromo-4-methyl-naphthalene (S1) 1-Methylnaphthaline (172 mL; 175 g; 1.23 mol) were solved in CHCl3 (1 L). While cooling the solution in a water bath 63.1 mL of bromine (1.23 mol) were slowly added. The solution was then refluxed and monitored by TLC. After cooling the solution to room temperature, the solution was then twice washed with a NaOH solution (1M). Then the organic layer was dried over Na2SO4 and the solvent was evaporated. The resulting brown oil was then twice vacuum-distilled at 180 °C and 70 mbar, resulting in 247 g (1.12 mmol, 91 %) of the pure yellow oil. The measured 1 H-and 13 C-NMR corresponds to the reference compound.

1-bromo-4-(bromomethyl)naphthalene (S2)
S2 was prepared by modified procedure from: M. Carreno et al., J. Org. Chem. 1999, 64, 1387-1390.0 g (23.0 mmol) of 1-bromo-4-methyl-naphthalene were dissolved in 30 mL of distilled DCM, then 4.5 g (25.3 mmol) and a catalytic amount of DBPO were added. The mixture was refluxed for 2 h (TLC monitoring). After cooling the solution to room temperature it was plugged through silica with hexane/DCM (2:1) as eluent. After evaporating the solvent the 5.1 g (17.0 mmol, 75 %) of the product were used without further purification in the next step. A small amount was purified for NMR analysis.

2-(4-bromo-1-naphthyl)acetic acid (S4)
1.0 g (4.1 mmol) of S3 were dissolved in 30 mL of acetic acid, then 10 mL of 10M sulfuric acid were slowly added under stirring. After leaving the solution for 12 h under reflux, the mixture was cooled to room temperature and diluted with 200 mL of water. After 2 h the white precipitate was filtered off, washed with water and dried under vacuum. 1.0 g (3.8 mmol, 96 %).

5-bromo-2H-acenaphthylen-1-one (S5)
1.0 g (3.8 mmol) of 2-(4-bromo-1-naphthyl) acetic acid were dissolved in 5 mL of thionylchloride and reflux for 90 min under nitrogen atmosphere. The excess of thionylchloride was removed under vacuum. The resulting oil was dissolved in 5 mL of DCM and cooled down to 0 °C (nitrogen atmosphere), before 1.0 g (2 eq, 7.6 mmol) of AlCl3 was slowly added. The mixture was stirred for 1 h at 0 °C and then heated to reflux for 15 min. After cooling to room temperature, the mixture was slowly poured to a suspension of 100 g of ice and 10 mL of hydrochloric acid. After warming up to room temperature the product was extracted with DCM (3x50 mL). The combined organic layers were dried of Na2SO4 and the solvent was evaporated. Crude product was purified by flash column chromatography on silica, using hexane/ DCM (1:1) as an eluent. 608.0 mg (2.5 mmol, 65 %) of the desired ketone were obtained.  78, 133.57, 132.73, 131.54, 129.98, 129.07, 127.89, 127.52, 127.29, 125.28, 121.40, 38.43.

6-methylbenzo[g]chrysene (S12)
4.37 g (14.9 mmol) of S11, 4.15 g I2 (1.1 eq, 16,3 mmol) und 10.4 mL propyleneoxide (10.0 eq, 149 mmol) were dissolved in 400 mL of cyclohexane and left at room temperature, under stirring in an UV-reactor (using a 400 Watt medium pressure mercury lamp). After competition (TLC monitoring) the solution was washed with sodium thiosulfate solution. The organic layer was dried over Na2SO4, before removing the solvent which resulted in 3.94 g (13.5 mmol, 90 %) of S12. The compound was used in the next step without further separation. A small amount of S12 were purified by flash chromatography for characterisation.

2-(bromomethyl)dibenzo[c,g]chrysene (S19)
350 mg (1.023 mmol) of S18, 200 mg of NBS (1.1 eq, 1.125 mmol) and 50 mg of DBPO (0.207 mmol) were dissolved in 30 mL of CCl4 and the mixture was refluxed for 2 h (TLC monitoring). After cooling the solution to room temperature it was plugged through silica (washed with toluene), and evaporated. The resulting 267 mg (0.634 mol, 62 %) of the S19 were used without further purification in the next step.

2-(dibenzo[c,g]chrysen-2-yl)acetic acid (S21)
0.7 g (1.9 mmol) of S20 were dissolved in 30 mL of acetic acid and 10 mL of 10M H2SO4 were added. After leaving the solution for 12 h under reflux, then cooling it to room temperature the solution was added to 200ml of water and leading to white precipitation. The suspension was left for 2 h. The precipitate was filtered off, washed with water and dried under vacuum. 692 mg (1.8 mmol), yield: 94 %. The compound was used in the next step as obtained.

Benzo[g]indeno[1,7-bc]chrysen-9(8H)-one (S22)
10.0 mg (26 µmol) of S21 and one drop of DMF were added to 5mL of DCM under nitrogen atmosphere and the reaction mixture was cooled to 0°C. After adding 0.1 mL of C2O2Cl2 the reaction mixture was heated under reflux for one h. After cooling the solution to room temperature and evaporating the solvent, the resulting oil was dissolved in 5mL of DCM and cooled down to 0°C, before 7.1 mg (2eq, 52 µmol) of AlCl3 were slowly added in small portions. After stirring of the reaction mixture at 0°C for one h. The solution was refluxed for 15 min and cooled to room temperature and poured into 10 g of ice containing 1 mL of concentrated HCl. After warming up to room temperature the reaction mixture was extracted with DCM (3x10 mL). The combined organic layers were dried of Na2SO4 and the solvent was evaporated. After a purification by flash column chromatography (toluene), 6.5 mg (18 µmol, 68 %) of the ketone S22 were obtained.

General procedure (combinatorial assembly of trimers)
50 mg (0.05-0.06 mmol) of the corresponding ketone (homotrimer synthesis) or mixture of two ketones in 1:1 or 1:2 ratio (combinatorial assembly) were dissolved in 1,5 mL of o-DCB directly in a glass ampule. After adding of 0,1 mL (0,94 mmol) of TiCl4 under argon atmosphere, the reaction mixture was freeze, degassed and the ampule was sealed under vacuum. The ampule was heated for 48 h at 150 °C. After cooling to room temperature the ampule was open and the reaction mixture was slowly poured into 20 mL of acetone and left for several h. The precipitate containing di-and trimers was filtered off (filtrate contains higher macrocycles and oligomers). The precipitate was dissolved in toluene and plugged through silica (toluene) to remove polar side products. Fraction containing trimers (MS analysis) was used further for HPLC separation. The overall yield of trimers was around 40-70% depending on segments and ratio used. The segments used and yields are summarized in Supplementary Fig 1 and  As an alternative to the cyclotrimerisation of cyclic ketones, a route based on Suzuki coupling was tested.
89mg (0.12 mmol) tribromodecacyclen, and 13mg (0.5 eq, 0.06mmol) biphenyl-boronic acid and 43mg (1.5 eq, 0.18mmol) anthracene boronic acid were dissolved in 36 mL of DMF/toluene (2:1) mixture. The respective tribromodecacyclen was obtained through a cyclotrimerisation of S5 according general trimerization procedure. The reaction mixture was three degassed and heated for 9 h under nitrogen at 80°C. After cooling the solution to room temperature 50 mL water were added and the solution was extracted with toluene. The organic layers were combined and dried over Na2SO4 and then filtrated through silica gel. The following HPLC analysis showed the formation of the desired products along formation of several sideproducts, which were identified as mono-and bis-functionalized decacyclenes.
Since the separation of differently substituted decacyclens appears to be very difficult, no further effort was done to improve the reaction conditions.

Cyclotetramerization
The tetramers were obtained using previously described protocol (Sumy et al., Chem. Eur. J. 2016, 22(14), 4709-4712). Briefly: The mixture of TiCl4 (1 mL) in ODCB (10 mL) was heated to reflux under nitrogen atmosphere. Cyclic ketone (1mmol) dissolved in ODCB was added dropwise to the refluxing mixture. After consumption of the educt the hot reaction mixture was slowly poured into crushed ice / HCl mixture and extracted with DCM. Due to our experience with the trimerization for the tetramerization of A and B a 1:2 ratio was chosen.
Tables containing the calculated exact masses and formulas of all combinations for tri-, tetra-, pentaand hexamers of the displayed segments A, B and C (combinations consisting of all three monomers are excluded). Comments: reciprocal space/lattice showed at least one twin component, which was excluded as much as possible for refinement weaker data set absolute structure could not get determined because of disorder and twinning situation; racemic twin refinement was implemented: BASF = -0.3 (19)