Supramolecular Nanopatterns of Molecular Spoked Wheels with Orthogonal Pillars: The Observation of a Fullerene Haze

Abstract Molecular spoked wheels with intraannular functionalizable pillars are synthesized in a modular approach. The functionalities at their ends are variable, and a propargyl alcohol, a [6,6]‐phenyl‐C61‐butyrate, and a perylene monoimide are investigated. All compounds form two‐dimensional crystals on highly oriented pyrolytic graphite at the solid–liquid interface. As determined by submolecularly resolved scanning tunneling microscopy, the pillars adopt equilibrium distances of 6.0 nm. The fullerene has a residual mobility, limited by the length of the flexible connector unit. The experimental results are supported and rationalized by molecular dynamics simulations. These also show that, in contrast, the more rigidly attached perylene monoimide units remain oriented along the surface normal and maintain a smallest distance of 2 nm above the graphite substrate. The robust packing concept also holds for cocrystals with molecular hexagons that expand the pillar–pillar distances by 15 % and block unspecific intercalation.


Contents
2.5 Bias voltage dependence of STM images of 3 S10 2.6 Topgraphies of 1 and 3 S12 2.7 Overview STM image of the binary mixture of 1 and 4 S13 2.8 Overview STM image of the binary mixture of 2 and 4 S14 2.9 Overview STM image of the binary mixture of 3 and 4 S15 3 Synthesis S16 3.1 Synthesis of 1, 2, and 3 S16 3. 1 General information

Materials and equipment
Reagents were purchased at reagent grade from commercial sources and used without further purification. All air-sensitive reactions were carried out using standard Schlenk techniques under argon. diisopropylsilyl]acetylene (CPDiPS-acetylene) and [(3-cyanopropyl)dimethylsilyl]acetylene (CPDMS-acetylene) were synthesized according to literature procedures described in [S1]. Reaction solvents (THF, piperidine, dichloromethane, pyridine, triethylamine, toluene) were dried, distilled, and stored under argon according to standard methods; workup solvents were either used in "p.a." quality or purified by distillation (dichloromethane, cyclohexane). Prior to characterization and further processing, all solids and oils were dried at r.t. under vacuum. 1 H and 13 C NMR spectra were recorded on a Bruker Avance I 300 MHz, Bruker Avance I 400 MHz, Bruker Avance III HD 500 MHz Prodigy and Bruker Avance III HD 700 MHz Cryo (300.1, 400.1, 500.1 and 700.1 MHz for 1 H and 75.5, 100.6, 125.8 and 176.0 MHz for 13 C). Chemical shifts are given in parts per million (ppm) referenced to residual 1 H or 13 C signals in deuterated solvents. All NMR spectra were recorded at r.t. unless otherwise described. Mass spectra were measured on a Finnigan ThermoQuest MAT 95 XL (EI-MS), a Sektorfeldgerät MAT 90 (EI-MS), a Bruker Daltonics micrOTOF-Q (ESI-MS, APCI), a Thermo Fisher Scientific Orbitrap XL mass spectrometer (ESI-MS), a Bruker Daltronics autoflex TOF/TOF (MALDI-MS; matrix material: DCTB, no salts added) and an ultrafleXtreme TOF/TOF of the Bruker Daltonik company (MALDI-MS; matrix material: DCTB, no salts added). m/z peaks smaller than 10 % (compared to the basis peak) are not reported. Thin layer chromatography was conducted on silica gel coated aluminium plates (Macherey-Nagel, Alugram SIL G/UV254, 0.25 mm coating with fluorescence indicator). Silica gel Kieselgel 60 (Merck, 0.040-0.063 mm) was used as the stationary phase for column chromatography. UV/vis absorption spectra were recorded on a Perkin Elmer Lambda 18 and fluorescence emission spectra on a Perkin Elmer LS-50B spectrophotometer using 10 mm quartz cuvettes. Microwave assisted reactions were performed in a CEM Discover Labmate instrument (maximal power: 300 W). Melting points were measured using an optical microscope equipped with a heating table (Leica DMLB, Leica LMW, Testo 965).

STM experiment
Scanning tunneling microscopy (STM) was performed under ambient conditions (r.t.) at the solution/solid interface, using 1,2,4-trichlorobenzene (TCB) as solvent and highly oriented pyrolytic graphite (HOPG) as substrate. In a typical experiment, 0.2 μL of a 1  10 -7 M to 3  10 -5 M solution of the compound(s) of interest was dropped onto a freshly cleaved HOPG substrate at r.t. or at elevated temperature (80 °C), kept at this temperature for 10 s to 20 s, and allowed to cool to r.t. before the STM measurements were performed with the tip immersed into the solution. Bias voltages between -1.4 V and +1.1 V and tunneling current set points in the range of 9 pA to 55 pA were applied to image the supramolecular adlayers shown here. The experimental setup consists of an Agilent 5500 scanning probe microscope that is placed on a Halcyonics actively isolated microscopy workstation. It is acoustically shielded with a home-built box. Scissors cut Pt/Ir (80/20) tips were used and further modified after approach by applying short voltage pulses until the desired resolution was achieved. HOPG was obtained from TipsNano (via Anfatec) in ZYB-SS and DS quality. All STM images (unless otherwise noted) were calibrated by subsequent immediate acquisition of an additional image at reduced bias voltage, therefore the atomic lattice of the HOPG surface is observed which is used as a calibration grid. Data processing, also for image calibration, was performed using the SPIP 5 (Image Metrology) software package. (Supra-) molecular modelling was performed using Wavefunction Spartan '16 and '18. Equilibrium geometries of the backbone structures were obtained using molecular mechanics (based on the Merck Molecular Force Field (MMFF)) and a graphene monolayer with fixed atom positions as interaction partner. Alkoxy sidechains were subsequently attached with the alkoxy-backbone angles as observed by STM.

Computational details
For the optimized input structures (see section 1.3.), geometry optimizations were performed at the GFN-FF level of theory. The resulting structures served as input for subsequent molecular dynamics (MD) simulations. MD simulations with GFN-FF were carried out for 1 ns at room temperature (298 K) employing the implicit GBSA(THF) solvation model. A time step of 2 fs at an increased hydrogen mass of 4 amu was chosen. All calculations were performed with the freely available xtb 6.4.0 program packages with default convergence criteria 10 -7 E h for energies and 10 -5 E h • Bohr -1 for gradients. All calculations were performed on Intel © Xeon E5-2660 v4 @ 2.00 GHz machines. The GFN-FF optimized structures (xyz-files) and the MD trajectories (mp4-files) for 2 and 3 are as separated files parts of the Supporting Information.

S5
2 Additional STM images 2.1 Overview STM image of 1 Self-assembled monolayers (SAMs) of 1 at the solid/liquid interface of HOPG and a solution of 1 in 1,2,4-trichlorobenzene (TCB) are investigated by STM (Figure 3a/h/k/l, Main Text, and Figure S1). At a concentration of c = 5 × 10 -7 M, a chiral honeycomb pattern with domain sizes > 140 2 nm 2 ( Figure S1) is observed. Lattice defects include missing molecules in the packing (arrow 1 in Figure S1) and unspecifically adsorbed molecules in the hexagonal nanopores (e.g. arrows 2 and 3 in Figure S1). The latter move within these nanopores as a result of translational degrees of freedom and are therefore observed as diffuse contrast features. Figure S1. Overview STM image of 1 at the solid/liquid interface of HOPG and a solution of 1 using TCB as a solvent. Image parameters: 140 × 140 nm 2 (internal scanner calibration), c = 5 × 10 -7 M, thermally annealed for 20 s to 80 °C, V S = -1.4 V, I t = 17 pA. Arrow 1 indicates missing molecules, arrows 2 and 3 mark examples of unspecifically adsorbed molecules of 1 located in the intermolecular nanopores.

Overview STM image of 2
SAMs of 2 are investigated at the solid/liquid interface of HOPG and a solution of 2 in TCB by STM ( Figure 3b-d/i/m, Main Text, and Figure S2). 2 is comparable to 1 in backbone size, shape, and alkoxy side chain substitution, although carrying a sterically demanding fullerene derivative substituent at its pillar unit. When a 1 × 10 -7 M solution of 2 is applied to a freshly cleaved HOPG surface, a chiral honeycomb packing with domains exceeding 90 2 nm 2 lateral size ( Figure S2) is observed, however with alike defects compared to 1. These include missing molecules within the packing (e.g. arrow 1 in Figure  S2). Moreover, one molecule in Figure S2 lacks a fullerene substituent (arrow 2). This also proves the distinguishability of three-dimensional substituents at the given imaging parameters and gives a hint on the isomorphism of 1 and 2. Figure S2. Overview STM image of 2 at the solid/liquid interface of HOPG and a solution of 2 using TCB as a solvent. Image parameters: 89 × 89 nm 2 (internal scanner calibration), c = 1 × 10 -7 M, thermally annealed for 20 s to 80°C, V S = -0.80 V, I t = 55 pA. Arrow 1 indicates a packing defect (missing molecule), and arrow 2 indicates a single molecule lacking the fullerene substituent.
SAMs 3 at the solid/liquid interface of HOPG and a solution of 3 in TCB are investigated by STM ( Figure  3e-g/j/n in the Main Text, and Figure S3). At concentrations of c = 1 × 10 -5 M to 5 × 10 -7 M in the supernatant solution phase, a chiral honeycomb pattern is observed, and at 1 × 10 -6 M, representatively, a domain size of > 72 2 nm 2 is found. As for 1 and 2, a contrast variation in one of the intermolecular nanopores (arrow 1 in Figure S3) is observed, attributed to an unspecifically adsorbed molecule. In the center of each medium bright backbone, depending on the exact bias voltage, V S , a small dot-shaped contrast feature is observed, attributed to the three-dimensional (3D) substituent. Figure S3. Overview STM image of 3 at the solid/liquid interface of HOPG and a solution of 3 using TCB as a solvent. Image parameters: 72 × 72 nm 2 (internal scanner calibration), c = 1 × 10 -6 M, thermally annealed for 20 s to 80°C, V S = -0.8 V, I t = 24 pA. Arrow 1 indicates an unspecifically adsorbed molecule.

Bias voltage dependence of STM images of 2
In the STM images shown in Figure 3b-d, Main Text, we have shown that the bias voltage does not significantly affect the visibility of the fullerene unit that is connected to the pillar unit of the MSW backbone as a butanoate. More specifically, the fullerene appears as interrupted, maximally bright scan lines superimposed to the MSW backbones, independent on whether a substrate bias voltage of V S = -0.80 V, +1.10 V, or +0.50 V (Figures 3b-d, Main Text, respectively, and Figure S2) is applied. The HOMO and LUMO of both the MSW and the fullerene units are electronically decoupled and separated around the voltage drop, i.e. the connection line of the two Fermi levels of both substrate and tip, E F sub and E F tip . Moreover, the exact energetic positions of HOMOs and LUMOs obtained for isolated molecules in the gas phase vary substantially when a molecule is electronically coupled to a solid substrate, and levels are broadened. Generally, MOs located in the conduction region, which is the energetic region between both Fermi energies, lead to an increase of the tunneling current as compared to vacuum. Based on these assumptions, we propose the following energetic scenario that can explain the observed image contrast. Nominally, at a moderate negative sample bias voltage ( Figure S4a and STM image in Figure 3b, Main Text, and Figure S2), the HOMO of the MSW and the fullerene should be energetically located in the conduction region, thus both molecule parts should be visible and appear bright, as they both should contribute to the tunneling current. Moreover, due to the flexible nature of the connection, the fullerene can be closer to either substrate or tip, thus the exact energetic location of its HOMO can alter, which is shown by the grey tilted bar in Figure S4a. This should, however, have no influence on its general visibility in this case. More specifically, the three-dimensional extension of 2 leads to a dominance of the fullerene once it is spatially located between the tip and the MSW and substrate at a given time, rendering as brightly appearing scan lines. However, thermal and tip-induced motion may lead its invisibility at other times, rendering as darker scan lines (of the MSW backbone, only). We accordingly refer to the random observation of bright scanlines between otherwise dark scanlines as "fullerene haze". A similar behavior holds for a moderate positive sample bias voltage ( Figure S4b, and STM image in Figure 3c, Main Text), with inverse energetic scenario. At small positive sample bias voltage ( Figure S4c, and STM image in Figure  3d, Main Text), the HOMO of the MSW should energetically be located outside of the conduction region, therefore the MSW should not contribute to the overall tunneling current. However, while the actual image (  The energetic location of the HOMO of the fullerene unit varies with its spatial location, which can be closer to either substrate or tip, which is indicated by tilted grey lines. S10

Bias voltage dependence of STM images of 3
In the STM images shown in Figure 3e-g (Main Text) and Figure S3, we observed a significant bias voltage dependence of the STM images of 3. At moderate negative substrate bias voltage (of V S = -0.61 V, cf. schematic drawing in Figure S5a and STM image in Figure 3e, Main Text), all molecule parts (i.e. the MSW backbone, the pillar unit carrying the perylene monoimide (PMI) unit, as well as the alkoxy side chains interdigitating intermolecularly on the HOPG substrate) are clearly visible. From this we conclude that the HOMOs of both MSW backbone and PMI/pillar units should be energetically located in the conduction region, contribute to the image contrast, and therefore might be visible by STM. The scenario should change at larger negative sample bias voltage V S , here defined as a voltage closer to zero (of V S = -0.50 V, cf. schematic drawing in Figure S5b and STM image in Figure 3f, Main Text): In this case, the HOMO of the 3D unit (i.e. PMI/pillar) may be energetically located outside (or, below) the conduction region. Consequently, the HOMO of the MSW backbone is the only orbital energetically located within the conduction region, so that only this molecule part renders visible (as seen in Figure 3e, Main Text). At small positive bias voltage (of V S = +0.50 V, cf. schematic drawing in Figure S5c and STM image in Figure 3g, Main Text), the 3D unit dominates the image contrast, as its HOMO is energetically located in the conduction region, while the HOMO of the MSW backbone is not, and is consequently invisible (or barely visible) in the STM image. The heights of 1 and 3 measured in the STM experiments are results of the topographic heights of the adsorbed species on the HOPG surface as well as their electronic properties translating into different electrical conductivities (or tunneling resistivities), which are also bias voltage dependent. Figures S6a  and b are reprints of the STM images provided in Figure 3a and e (Main Text). In Figure S6c and d we provide topography cross sections, i.e. apparent relative heights that are referenced to the solvent covered HOPG surface defined as zero, through the solid white lines in Figures S6a and b. The data are compared to the topographic heights obtained from the molecular models shown in Figure 2 (Main Text).

Synthesis of 5
Coupling of 2,7-dibromocarbazole with 12 and subsequent reduction of the NO 2 -group followed by a Sandmeyer-like reaction yielded 15 which was coupled with CPDMS acetylene via Sonogashira reaction leading to the "T"-shaped building block 16. Twofold Sonogashira reaction with acetylene 17 followed by deprotection of the CPDMS-group yielded after deprotection of the CPDMS group the Mshaped building block 5.

Synthesis of 6a and 6b
Fourfold iodination of tetraphenylmethane in para position followed by a statistical Sonogashira reaction with propargyl alcohol lead to the central unit 6a, while coupling with 26 lead to 6b, respectively.
Analytical data coincide with those in ref [S4].

30
Copper powder (2.46 g, 38.71 mmol) was activated at 80 °C for 24 h in 3-picoline (16 mL) under moisture exclusion (drying tube, CaCl 2 ). 29 (2.16 g, 3.77 mmol) was added to the reaction mixture and stirred at 175 °C for 24 h. After cooling to r.t., CHCl 3 and aq. HCl (3.2 M) were added. The aqueous phase was extracted two times with CHCl 3 . The combined organic phase was washed with water and dried over Na 2 SO 4 . Removal of the solvent and purification via column chromatography (DCM:PE = 2:1, R f = 0.6) yielded pure 30 (818 mg, 1.62 mmol, 43 %) as a red solid.