An Azobenzene-Clamped Bichromophore

Abstract

An azo-clamped nanoscale bichromophoric cyclophane is synthesized by the intramolecular Pd(II)-catalyzed coupling of the corresponding bisacetylenic precursor. The two azo moieties in the latter can adopt cis and trans configurations. Thin-layer chromatography shows only two spots, and by scanning tunneling microscopy the trans/trans and cis/cis isomers are found. The final cyclophane does not show any switching behavior at all, but dense and wide structures are visualized after adsorption to highly oriented pyrolytic graphite. Photophysical investigations of the cyclophane show that most of the fluorescence is quenched, most likely due to the azo clamp. However, bright molecules show nearly perfect single-photon emission, meaning that efficient energy transfer between the two chromophores takes place within the molecule.

Key words

cyclophanes - azo-switches - phenylene–ethynylene–butadiynylenes - scanning tunneling microscopy - single-molecule spectroscopy

Introduction

π-Conjugated materials, low-molecular-weight compounds as well as polymers have interesting optical and electronic properties, making them active components in organic electronics like organic light-emitting diodes, organic solar cells, and organic field-effect transistors. Further improvement of the devices requires a deeper understanding of the interaction of the molecular segments in the ground state and in the excited state, both within one and between two or more molecular species.[1]–[4] The latter is of outstanding importance since devices, even though generally containing the electroactive materials in form of a thin film, have thousands of billions of single molecules.

Modern synthetic methodologies as well as sophisticated spectroscopic techniques, also on the single-molecule level, are used to make a large variety of elaborated conjugated model compounds available now and allow studying their fundamental optical behavior in conjugated molecules.[5]–[8] Single-molecule fluorescence spectroscopy allows the investigation of individual molecules and gives detailed information on the fluorescence energy, the vibronic structure as well as life-time and polarization, information that is averaged when bulk measurements are performed. However, studying discrete aggregates, e.g. dimers or trimers of oligomers, rather seems like an oxymoron: to perform the single-molecule measurement, the analyte molecule is diluted in a solvent to single-molecule concentration (~10⁻¹² M) and mixed with a polymethyl methacrylate (PMMA)/toluene solution. This solution is then dynamically spin-coated, resulting in a film thickness of roughly 100 nm, in which the separated single molecules can be investigated. It is clear that most molecules are isolated in the matrix, and even if a few dimers or larger aggregates might be observed, the units comprising these are randomly oriented with respect to each other. A way out of this dilemma is defined multichromophoric aggregates, where not only the number of individual chromophores but also their relative spacing and orientation can be specified. We recently described a number of clamped oligomers with various lengths and distances.[9]–[13] Specifically, we connected poly(phenylene–ethynylene–butadiynylene)s[14] with defined lengths between 5 and 10 nm at their ends, so that they are held at distances between 0.38 and 2.1 nm, forming shape-persistent macrocycles.[15] The distance is measured at the clamp positions and might differ in the molecule center due to a restricted persistence length of the phenylene–ethynylene–butadiynylene rods.[16] One possibility to rigidify phenylene–ethynylene–butadiynylenes is ladderization.[17] We also synthesized and investigated in detail a trichromophoric oligomer system, in which the spatial separation between the clamped three oligomers is around 0.7 nm.[13] However, in all these cases the distances between the oligomers are, apart from thermal fluctuations, constant. Since the coupling between the oligomers is sensitive to their spacing, we describe here a bichromophore with azo clamps in the hope that a change of the coupling can be observed when a light-driven photoisomerization of the azo moieties of the macrocycle can be realized.[18] Azobenzenes, after their first description in the 1830 s[19] and the determination of their structure 30 years later,[20] have become an important class of industrial colorants and still represent more than half of the industrially produced dyes.[21] A hundred years after their first description, the light-induced isomerization of the azo moiety was reported.[22] This was the starting point for the implementation of azo moieties for light-induced structure and function alteration of linear and cyclic molecules of different sizes and functionalities.[23]

Results and Discussion

The straightforward synthesis of the azo-containing bichromophoric cyclophane 1 presented here is shown in [Scheme 1]. 1 is an azo-analogue of our recently described clamp structure, i.e. 2.[12] 3,3-Diiodoazobenzene (3)[24] was coupled with the monoprotected bisacetylene 4 [25] to give 5 in a yield of about 71% after purification by column chromatography and recycling gel permeation chromatography (rec GPC). However, 5 contains small amounts of the Glaser side product that could not be removed at this stage of the reaction sequence. Nevertheless, after statistical deprotection of 5, the mono-protected bisacetylene 6a (32%) along with the completely deprotected byproduct 6b (29%) were obtained in the pure form. Attempts to cyclodimerize 6b to 1 using CuCl and CuCl₂ in pyridine were not successful and gave only the starting material, acyclic dimer, and unknown byproducts of higher molecular weight. The route of success was first a dimerization of the monoprotected “half-ring” 6a to 7 (48%), deprotection of the acetylenes (81%), and intramolecular ring-closure to 1 (50 – 60%) using Pd(II) and Cu(I) as a catalyst system, I₂ as an oxidant and diisopropylamine as a base in THF. As expected, upon cyclization the hydrodynamic radius of the compound decreases dramatically, leading to a large change in the GPC-determined molecular weight (vs. polystyrene), although only two hydrogen atoms are removed (see the Supporting Information, SI).

Scheme 1 a) Pd(PPh₃)₄, CuI, piperidine, THF, rt, 6 d, 71%; b) TBAF (1 M in THF), THF, H₂O (5 vol%), rt, 5 h, 32%; c) Pd(PPh₃)₂Cl₂, CuI, I₂, HN(iPr)₂, THF, rt, 16 h, 48%; d) TBAF (1 M in THF), THF, rt, 12 h, 81%; e) Pd(PPh₃)₂Cl₂, CuI, I₂, HN(iPr)₂, THF, 50 – 60%.

While the NMR spectrum of the intermediates shows the presence of cis and trans azobenzene units, in the final macrocycle the azobenzene units cannot be switched to the cis isomers, most probably due to steric reasons of the transition state of the isomerization since space-filling models of trans/trans-1 and cis/cis-1 show no significant deformation of the chromophores. Thus, the proposed photoinduced alteration of the chromophore–chromophore distance is not realized in this compound 1.

Although the absence of any switching behavior in a smaller macrocycle has been reported,[26] we expected that the large ring would give the system enough flexibility to allow for a photoinduced change of the azo conformation.[27] Speculations about the large π-system being responsible for the stabilization of the trans-conformation can be discarded, since 6b and 8 undergo reversible trans–cis switching (vide infra).

Self-assembled monolayers of 1 and 8 at the solid/liquid interface of the respective compound in 1,2,4-trichlorobenzene (TCB) and highly oriented pyrolytic graphite (HOPG) were investigated by scanning tunneling microscopy (STM).

At a concentration of 1 of 5 × 10⁻⁶ M in the supernatant liquid phase, 1 forms (after thermal annealing for 20 s to 80 °C, a routine procedure for enhancing self-assembly and packing order) a densely covered self-assembled monolayer with domain sizes of similarly oriented molecules in the range of 20 × 20 nm² (see the overview STM image in the SI). In the high-resolution STM image shown in [Figure 1]a, each molecule of 1 is imaged as a pair of bright lines (attributed to the rigid rod units) connected by two medium-bright regions (attributed to the azobenzene units). The rigid rods neither appear as perfect lines, nor do they all have the same shape. More precisely, wide pairs of rods (as, e.g., those marked by arrow 1 in [Figure 1]a) should provide space for all (or most) of the alkoxy side chains being aligned along the HOPG surface, while tight pairs of rods (e.g., those marked by arrow 2 in [Figure 1]a) might correspond to alkoxy side chains pointing towards the solution phase (see SI). Their random occurrence correlates with some degree of disorder of the otherwise two-dimensionally (2D) crystalline domains, to which a unit cell of a = (5.1 ± 0.2) nm, b = (3.1 ± 0.2) nm, γ (a, b) = (78 ± 2)° and an orthogonal orientation of the backbones, c, to one of the HOPG main axis directions, d, is indexed. An idealized supramolecular model (of a geometry-optimized wide conformer with all hexyloxy side chains adsorbed in parallel to the HOPG surface and along one of its main axis directions, see SI) is shown in [Figure 1]c.

Figure 1 (a, b) Scanning tunneling microscopy images and (c, d) (supra-)molecular models of (a), (c) 1 and (b), (b1), (b2), (d) 8 at the solid/liquid interface of highly oriented pyrolytic graphite (HOPG) and solutions of the respective compounds in 1,2,4-trichlorobenzene. Image parameters: (a) 18 × 18 nm², V _S = −0.8 V, I _t = 55 pA, c = 5 × 10⁻⁶ M, sample thermally annealed for 20 s to 80 °C prior to imaging; unit cell a = (5.1 ± 0.2) nm, b = (3.1 ± 0.2) nm, γ (a, b) = (78 ± 2)°; additional packing parameters: γ (c, d) = (90 ± 3)°, b ∥ d; (b): 30 × 30 nm², V _S = −1.1 V, I _t = 117 pA; c = 1 × 10⁻⁵ M, sample thermally annealed for 20 s to 80 °C prior to imaging; unit cell a = (1.4 ± 0.1) nm; additional packing parameters: γ (c, d) = (83 ± 2)°, γ (a, d) = (7 ± 2)°; (b1), (b2): each 2.1 × 2.1 nm². Red and white (black) lines as well as blue arrows indicate unit cell vectors, HOPG main axis directions, and backbone directions, respectively.

At a concentration of 8 of 10⁻⁵ M in the supernatant liquid phase, 8 covers the surface densely (after thermal annealing for 20 s to 80 °C). An overview STM image (see SI) shows domains of parallel-aligned backbones, however with disorder. In the detailed STM image ([Figure 1]b), one dimensionally (1D) crystalline domains are observed, each consisting of a few molecules. The image region in [Figure 1]b marked by the white dots is translated to the supramolecular model in [Figure 1]d. In this region, five molecules of 8 are aligned in parallel, and a unit vector of a = (1.4 ± 0.1) nm is indexed. Their azobenzene units adopt the trans/trans configuration, and are oriented anti relative to the central rigid rod. Moreover, the rigid rod units at each of the azobenzene units are oriented in the anti-conformation (cf. Figure S8). The distance of the molecules is defined by intermolecularly interdigitating hexyloxy side chains that are aligned along one of the HOPG main axis directions, d. The rigid rods are oriented along c with γ (c, d) = (83 ± 2)° relative to one of the HOPG main axis directions, d, and the unit cell vector, a, is oriented relative to d with γ (a, d) = (7 ± 2)°. In the marked surface region, two more molecules of 8 are shifted in parallel (along c) by half a molecule length relative to the five previously discussed molecules, so that seven trans-azobenzene units (each with two rigid rods in the anti-conformation) form a row. Notably, 8 in the periodic packing adopts a linear shape with two kinks, attributed to the azobenzene units. One example of a trans-azobenzene unit is magnified in [Figure 1]b1 and is clearly distinguishable from the cis-azobenzene unit shown in [Figure 1]b2. Moreover (and despite the thermal annealing procedure), a single molecule of cis/cis-8 is observed (the ends of which are marked by arrows 3/5 and 4/6 in [Figure 1]b/d), where the rigid rods adopt an angle of γ (c′, c′′) = (96 ± 2)°.

These results motivated us to investigate 1, its precursor 8 (containing three phenylene–ethynylene rods of two different lengths, connected by two azobenzene units), and 6b (containing two “short” phenylene–ethynylene rods connected to one azobenzene unit) by means of UV/vis spectroscopy (see SI). When a solution of 6 × 10⁻⁶ M of 1 in toluene was prepared and allowed to stand for 2 h, the UV/vis spectrum showed two maxima at 319 and 421 nm. Irradiation with a light-emitting diode with 410 nm maximum wavelength and a FWHM of 20 nm (see SI) did not lead to any discernible spectral change (see SI). After preparing solutions of 8 and 6b under identical conditions, absorption maxima at 311, 319 and 416 nm (8) as well as 310, 319, and 399 nm (6b) were observed. An irradiation at 410 nm for 5 s (or 310 nm for 60 s, 8 and 6b) led to photostationary states with a minor reduction in absorbance in the spectral range of 300 to 360 nm. Allowing each of the solutions to stand for 5 h in the dark at rt led to a slow increase in absorbance in this region. Azobenzene[21] shows dramatic changes in intensities, and π-extended pyrene derivatives with four azobenzene-ethynyl groups[28] still show significant variations in spectral intensities after irradiation (also for a UV-light-induced back reaction). We attribute the minor intensity changes in 8 and 6b to the fact that their spectra (particularly at λ > 350 nm) are determined predominantly by the absorption of the phenylene–ethynylene chromophores (cf. extinction coefficients in Figure S11). The spectrum of 1 does not show any changes after irradiation. This fact is consistent with its ¹H NMR spectrum, which does not contain any hint for the presence of the cis/cis isomer (Figure S1).

We then investigated macrocycle precursor 8 and the half-ring 6b by 1D and 2D thin-layer chromatography (TLC, see SI). Nominally, we would expect trans/trans-8, trans/cis-8, and cis/cis-8 as well as trans-6b and cis-6b, and each set of compounds should have different R _f values. Handling the compounds and performing the TLC experiments (using Cy : DCM 1 : 2) in complete darkness led to single peaks for both species (R _f (8): 0.33; R _f (6b): 0.46). We conclude that these solutions consist of trans/trans-8 and trans-6b, respectively. Exposing 8 (and 6b) to daylight (either before and during its application to the TLC plate, or after dissolving and depositing the compound and performing a first TLC run in darkness) or 410 nm (while otherwise handling the compound in darkness) led to additional spots (with R _f (8) = 0.21 and R _f (6b) = 0.24). Unexpectedly, a third spot, indicative of a third switching state, was not observed for 8. This absence might be explained by either identical R _f values or concerted trans–cis-isomerization of both azobenzene units after photon absorption. This observation is consistent with the STM experiment described before where trans/trans and cis/cis isomers were observed yet. Exposure of 6b during the TLC run led to smeared peaks, indicative of trans–cis (and cis–trans) isomerization on the TLC timescale.

Ensemble measurements to investigate the photophysical properties of 1 were performed in toluene solution. [Figure 2]a shows the normalized absorption (λ _Max ^Abs ≈ 420 nm) and photoluminescence (PL) spectra (λ _Max ^PL ≈ 470 nm) under excitation at 410 nm (green curves). In order to compare the results with our previously synthesized structures, we also show the spectra of the rigid clamp reference 2 ([Scheme 1]) with a chromophore spacing of approx. 0.7 nm (black).[12] Both the absorption and emission spectra of compound 1 are blue-shifted compared to this earlier bichromophoric structure. The blue-shift in the absorption of 1 presumably originates from the inductive effects of the azobenzene clamps. [Figure 2]b shows the PL decay of 1 (green curve) in toluene solution, measured under pulsed laser excitation using time-correlated single-photon counting (TCSPC) together with the instrument response function (IRF) of the TCSPC system used (gray). By reconvolution of the signal with the IRF, we fit the data using a biexponential function (red) with decay times of τ ₁ = 25 ps and τ ₂ = 648 ps, where the latter value is determined by a tail fit of the decay. This biexponential decay profile of 1 is in stark contrast to that of the rigid clamp reference 2, where a monoexponential decay was observed with a PL lifetime of 660 ps.[12]

Figure 2 Ensemble spectroscopy in toluene solution: (a) normalized absorption (dotted lines) and emission (solid lines) spectra of compound 1 (green) and the bichromophoric system 2 with rigid clamping units and a 0.7 nm interchromophoric spacing (black). (b) PL intensity decay transient of 1 (green) together with the instrument response function (IRF; gray) of the TCSPC system used. The biexponential fit (red) is achieved by reconvolution of the signal with the IRF and by performing a tail fit of the signal to reveal the dominant time constants τ ₁ = (25 ± 1) ps and τ ₂ = (648 ± 2) ps.

As we showed in earlier work, the dependence of the PL spectra and PL lifetime on the interchromophore distance can be described in the framework of H-type electronic coupling of molecular aggregates.[13],[29] As the chromophore separation decreases, the PL spectrum is shifted to lower photon energies and the oscillator strength of the radiative transition decreases. The comparison with the 0.7 nm-spaced rigid clamp reference 2 in [Figure 2] indicates a slightly increased chromophore spacing of the azobenzene bichromophore 1, and hence the PL spectrum of 1 is blue-shifted and the PL lifetime is slightly lowered to τ ₂ = 648 ps with respect to 2. This observation agrees with the STM measurements shown in Figure S4, which suggest a chromophore separation in 1 of between 0.8 and 1.6 nm. In order to identify the origin of the unexpectedly fast PL lifetime component of 1, we measured the PL quantum yield (PLQY) using the commercial dye ATTO 390 as a reference standard and found a value of (4 ± 0.4)%. This value is strongly reduced compared to the previously reported rigid clamp bichromophoric system, where a value of > 60% was measured.[10] Therefore, we conclude that the fast PL decay component of τ ₁ = 25 ps follows a fast nonradiative recombination pathway, which is only present in compound 1 but not in the rigidly clamped reference 2.

This conclusion raises the question of whether the multiexponential fluorescence decay in solution is related to the occurrence of multiple subpopulations or conformations of the molecules, or whether these dynamics can also be observed on the level of single isolated molecules. Single-molecule spectroscopy offers direct access to answering this question. By strongly diluting the analyte into a 2 wt% PMMA–toluene solution, we can immobilize single molecules by spin coating this mixture onto microscope glass cover slips to form a PMMA film of approx. 50 nm thickness. The concentration of the analyte within the PMMA/toluene solution determines the density of immobilized molecules observed in confocal scan images of the microscope as shown in [Figure 3]. We adjusted the analyte concentration of both the previously studied rigid clamp reference 2 and of sample 1 prior to spin coating to the very same concentration and show the resulting confocal microscopy scan images in [Figure 3]. For the same concentration, we find a strongly reduced fluorescence spot density of (0.011 ± 0.004) µm⁻² for compound 1 compared to the rigid clamp structures 2 with a spot density of (0.176 ± 0.008) µm⁻².

Figure 3 Single molecule spectroscopy: (a) and (b) show confocal scan images with dimensions of (20 × 20) µm² of the 0.7 nm rigid clamp structures 2 and azobenzene-clamped bichromophore 1, respectively. From these images, we extract spot densities of (0.176 ± 0.008) µm⁻² and (0.011 ± 0.004) µm⁻², respectively. (c) and (d) show examples of single-molecule PL decays (black) together with monoexponential fits (red lines).

[Figures 3](c, d) shows example TCSPC traces (black curves) of the rigid and azobenzene bichromophoric clamp structures 2 and 1, respectively, together with single-exponential tail fits (red lines). The time constants extracted are stated in the figure. Surprisingly, unlike in the ensemble measurements, the average single-molecule PL lifetimes are almost identical for both samples ((0.83 ± 0.17) ns for 1 vs. (0.89 ± 0.2) ns for 2), even though the lifetimes scatter within each sample. A statistical analysis of this observation is shown in Figure S16 of the SI. The most important observation is that all measured spots of 1 show a single-exponential PL decay without any fast lifetime component, unlike the ensemble measurements in [Figure 2]. We therefore conclude that two subpopulations must be present within the ensemble of compound 1. The first group exhibits photophysical properties comparable to the previously reported rigid clamp reference 2. However, since the density of visible spots is approximately 16 times lower for 1 than for the rigid clamp reference 2, the results indicate that most single molecules of 1 show strong fluorescence quenching, i.e. nonradiative recombination, leading to the low PL lifetime component τ ₁ and the low PLQY in solution. Due to this low brightness, we cannot detect this subpopulation at the single-molecule level. This strong quenching also agrees with the PLQY ratio of the two compounds of approx. 60% to 4% (≈ 15). Since a trans–cis switching of the azo moieties in 1 can be excluded, it is conceivable that the azobenzene groups themselves directly influence the nonradiative decay pathways of the chromophores. On the one hand, such quenching could also conceivably be caused by the flexibility and mobility of the clamp pieces, as has already been observed for molecular rotors such as BODIPY dye molecules.[30],[31] On the other hand, it is also possible that energy or electron transfer occurs to the azobenzene groups, for which low fluorescence quantum yields have already been reported.[32] A possibility to test whether efficient energy transfer can occur within the clamped bichromophore structure 2 is to measure the fluorescence photon statistics of single molecules as discussed in Figure S16 of the SI. For both compound 1 and the analogous rigid clamp structure 2, we find nearly perfect single-photon emission, i.e. so-called photon antibunching, in the single-molecule fluorescence. In a multichromophoric molecule such as 1, this effect can only occur if multiple excitons, which are present at the same time, can diffuse in space and annihilate via bimolecular recombination pathways, i.e. singlet–singlet annihilation.[32] The nearly perfect single-photon emission of 1 shows that efficient energy transfer indeed takes place between the two chromophores within the molecule. It is also conceivable that, under certain circumstances, a transfer process to the azobenzene clamp units might be possible and thereby quench the excited-state population.

Conclusions

An azo-clamped bichromophoric cyclophane was successfully synthesized by the intramolecular Pd(II) and Cu(I) catalyzed coupling of the corresponding bisacetylenic precursor. While a photoinduced switching of the precursors could indeed be observed, most easily by TLC, the final cyclophane does not show any switching behavior. STM investigations prove the overall structure of the cyclophane and show that its open precursor adopts trans/trans and also cis/cis configurations. Photophysical investigations show that most of the fluorescence is quenched, which we attribute to the influence of the azo clamp in the molecule. However, bright molecules show nearly perfect single-photon emission, meaning that efficient energy transfer between the two chromophores takes place within the molecule.

Experimental Section

Reagents and analytics. 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. 2-Bromo-5-iodo hydroquinone (8) was prepared as described in Ref. [33].[33] [(3-Cyanopropyl) diisopropylsilyl] acetylene (CPDiPS-acetylene) and [(3-cyanopropyl dimethylsilyl] acetylene (CPDMS-acetylene) were prepared as described in Ref. [34].[34] Diiodo azobenzene (2) was prepared as described in Ref. [35].[35] Reaction solvents (tetrahydrofuran, 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 rt under vacuum. ¹H and ¹³C NMR spectra were recorded on a Bruker Avance I 300 MHz, a Bruker Avance I 400 MHz, a Bruker Avance III HD 500 MHz Prodigy and a Bruker Avance III HD 700 MHz Cryo (300.1, 400.1, 500.1 and 700.1 MHz for ¹H and 75.5, 100.6, 125.8 and 176.0 MHz for ¹³C). Chemical shifts are given in parts per million (ppm) referenced to residual ¹H or ¹³C signals in deuterated solvents. All NMR spectra were recorded at rt 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 Bruker Daltonics 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). TLC 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.

Gel permeation chromatography (GPC). GPC was performed in THF (HPLC grade, stabilized with 2.5 ppm BHT) at rt. GPC analyses were run on an Agilent Technologies system at a flow rate of 1 mL/min using an IsoPump (G1310 A), a diode array UV detector (G1315B) and PSS columns (Polymer Standards Service; Mainz, Germany; 10², 10³, 10⁵ and 10⁶ Å, 5 µ, 8 × 300 mm). All molecular weights were determined vs. PS calibration (PS standards from PSS, Mainz, Germany).

For the preparative separation, a Shimadzu rec GPC system, equipped with an LC-20 AD pump, an SPD-20 A UV detector and a set of three preparative columns from PSS (either SDV 10³ Å, 5 µ, 20 × 300 mm or SDV preparative linear S, 5 µ, 20 × 300 mm) with precolumn (SDV, 5 µ, 20 × 50 mm) was employed. The system operated at a flow rate of 5 mL/min, THF, 35 °C.

Scanning tunneling microscopy (STM). STM was performed under ambient conditions (rt) at the solution/solid interface, using TCB as the solvent and HOPG as the substrate. In a typical experiment, 0.2 µL of a 5 × 10⁻⁶ M to 1 × 10⁻⁵ M solution of the compound of interest was dropped onto a freshly cleaved HOPG substrate at elevated temperature (80 °C), kept at this temperature for 20 s, and allowed to cool to rt before the STM measurements were performed with the tip immersed into the solution. Bias voltages between −0.8 and −1.2 V and tunneling current set points in the range of 26 – 117 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 a 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. (Supra-)molecular modelling was performed using Wavefunction Spartan ʼ18. Equilibrium geometries shown in Figure S4 were obtained using molecular mechanics (based on the Merck molecular force field) and a graphene monolayer with fixed atom positions as the interaction partner. Equilibrium geometries shown in Figure S7(b – g) were obtained with the same method, however using three different starting geometries. These were manually created to match the shapes observed in the STM image shown in Figure S7a (arrows 1 – 6). Moreover, dihedral angles of the azobenzene units were frozen to obtain the trans or cis isomers. The molecular models shown in [Figure 1]c were obtained from the backbone structure shown in Figure S4(a, b) and subsequently added all-trans-configured alkoxy side chains oriented along the HOPG main axis directions observed in the STM image, and these molecules were used to create the supramolecular model. The supramolecular model shown in [Figure 1]d was obtained in a similar procedure to match the structures observed in the STM image shown in [Figure 1]b.

Optical spectroscopy. UV/vis absorption spectra shown in Figures S9–S11 were recorded on a Perkin Elmer Lambda 18 spectrometer using 10 mm quartz cuvettes.

Ensemble absorption and PL spectra shown in [Figure 2] were recorded by dissolving the analyte in toluene solution and filling this into a 10 mm quartz cuvette (Hellma Analytics, Quartz SUPRASIL). The data were recorded using a Perkin Elmer spectrometer (Lambda 650) for absorption and a Horiba Jobin Yvon Fluoromax 4 for PL. Spectra were background-corrected and normalized.

The PL decay of 1 in toluene solution was measured on an inverted confocal microscope as described elsewhere (Figure S6). A frequency-doubled Ti : sapphire oscillator (Spectra Physics Mai Tai BB) operating at 440 nm and 80 MHz repetition frequency was used for excitation. Using a single-photon counting module (Picoquant-MPD-050-CTB), we recorded the signal over a time period of 5 minutes. The extracted PL lifetimes shown in [Figure 2]b were confirmed by additional measurements using a picosecond streak-camera system (data not shown).

Single-molecule measurements were performed under excitation at a wavelength of 440 nm with a power density of approx. 750 Wcm⁻² using two single-photon detectors (Picoquant-π-SPAD-20) in a Hanbury Brown and Twiss detection geometry.

Synthetic procedures

Synthesis of 5: Under an Ar atmosphere, 3 (36.0 mg, 71 µmol)³⁵, 4 (200.0 mg, 0.18 mmol), Pd(PPh₃)₄ (9.5 mg, 0.008 mmol), and CuI (1.0 mg, 0.004 mmol) in piperidine (15 mL) and THF (5 mL) were stirred at rt for 6 d. Water and CH₂Cl₂ were added, the aqueous phase was extracted with CH₂Cl₂, and the organic phase was washed with aqueous HCl (2 M), water and brine and dried over MgSO₄. After evaporation of the solvent, column chromatographic purification (Cy : DCM = 1 : 1 → 2 : 3; R _f = 0.53 (2 : 3)) gave 5 as a yellow film (139.8 mg, 58 µmol, 71%).

Formula: C₁₅₆H₂₁₆ N₄O₁₂Si₂, molar mass: 2395.63 g/mol.

¹H NMR (500 MHz, CDCl₃, rt) δ [ppm]: 8.10 (t, ⁴ J _HH = 1.8 Hz, 2 H), 7.91 (ddd, ³ J _HH = 8.0 Hz, ⁴ J _HH = 2.0 Hz, ⁴ J _HH = 1.2 Hz, 2 H), 7.65 (dt, ³ J _HH = 7.7 Hz, ⁴ J _HH = 1.3 Hz, 2 H), 7.52 (t, ³ J _HH = 7.8 Hz, 2 H), 7.05 (s, 2 H), 7.04 (s, 2 H), 7.02 (s, 2 H), 7.01 (s, 2 H), 6.96 (s, 2 H), 6.93 (s, 2 H), 4.08 – 4.00 (m, 20 H), 3.96 (t, ³ J _HH = 6.4 Hz, 4 H), 2.44 (t, ³ J _HH = 7.0 Hz, 4 H), 1.94 – 1.81 (m, 28 H), 1.61 – 1.46 (m, 24 H), 1.43 – 1.24 (m, 48 H), 1.15 – 1.06 (m, 28 H), 0.94 – 0.82 (m, 40 H).

¹³C NMR (126 MHz, CDCl₃, rt) δ [ppm]: 154.5, 153.9, 153.7, 153.4, 152.5, 134.2, 129.3, 125.9, 124.7, 123.2, 119.9, 118.0, 117.4, 117.4, 117.3, 116.6, 115.0, 114.8, 114.5, 114.4, 113.8, 113.4, 104.1, 95.3, 94.2, 91.8, 91.8, 91.7, 91.6, 87.1, 70.0, 69.9, 69.9, 69.8, 69.3, 31.8, 31.8, 29.9, 29.6, 29.5, 29.5, 29.5, 27.1, 26.0, 26.0, 25.9, 25.8, 25.8, 22.8, 22.8, 21.5, 20.9, 18.4, 18.2, 14.2, 14.2, 12.0, 9.8, 1.2.

MS (MALDI-TOF, DCTB) m/z: 2393.6 [M]⁺, 2643.7 [M + DCTB]⁺.

Calculated exact mass: 2393.60 g/mol.

Synthesis of 6a and 6b: Under an Ar atmosphere, TBAF (1 M in THF, 0.1 mL, 0.1 mmol) was added to 5 (130.0 mg, 54 µmol) in THF (6 mL) and water (0.24 mL). After 1 h, 1.5 h, and 2.5 h, additional portions of TBAF (each 0.1 mL, 0.1 mmol) were added. After a total reaction time of 5 h, water and CH₂Cl₂ were added, the aqueous phase was extracted with CH₂Cl₂, and the organic phase was washed with water and brine and dried over MgSO₄. After evaporation of the solvent, column chromatographic purification (Cy : DCM = 2 : 3 → 1 : 2, R _f = 0.64 (1 : 2)) and additional purification by rec GPC gave 6a as a yellow film (38.5 mg, 0.02 mmol, 32%). In addition, 6b was also obtained as a yellow film (32.5 mg, 16.0 µmol, 29%) (Cy : DCM = 2 : 3 → 1 : 2, R _f = 0.70 (1 : 2)).

6a: Formula: C₁₄₆H₁₉₇ N₃O₁₂Si, molar mass: 2214.28 g/mol.

¹H NMR (500 MHz, CDCl₃, rt) δ [ppm]: 8.10 (t, ⁴ J _HH = 2.0 Hz, 2 H), 7.92 – 7.89 (m, 2 H), 7.65 (d, ³ J _HH = 7.6 Hz, 2 H), 7.52 (t, ³ J _HH = 7.8 Hz, 2 H), 7.07 – 6.91 (m, 12 H), 4.10 – 3.93 (m, 24 H), 3.34 (s, 1 H), 2.44 (t, ³ J _HH = 7.0 Hz, 2 H), 1.94 – 1.75 (m, 26 H), 1.62 – 1.42 (m, 24 H), 1.42 – 1.28 (m, 48 H), 1.16 – 1.04 (m, 14 H), 0.94 – 0.81 (m, 38 H).

¹³C NMR (126 MHz, CDCl₃, rt) δ [ppm]: 154.5, 154.3, 153.9, 153.7, 153.5, 153.4, 152.5, 134.2, 129.3, 125.9, 124.7, 123.2, 119.9, 118.1, 117.9, 117.4, 117.3, 117.2, 116.6, 114.8, 114.4, 113.8, 113.4, 112.7, 104.1, 95.3, 94.2, 91.7, 91.7, 91.4, 87.1, 82.4, 80.2, 70.0, 69.9, 69.8, 69.8, 69.8, 69.7, 69.3, 31.8, 31.8, 31.8, 31.7, 29.9, 29.5, 29.5, 29.5, 29.4, 29.4, 29.4, 29.3, 26.0, 25.9, 25.9, 25.8, 25.8, 25.8, 25.8, 22.8, 22.8, 22.7, 21.5, 20.9, 18.4, 18.2, 14.2, 14.2, 14.2, 12.0, 9.8.

MS (MALDI-TOF, DCTB) m/z: 2212.4 [M]⁺, 2462.5 [M + DCTB]⁺.

Calculated exact mass: 2212.47 g/mol.

6b: Formula: C₁₃₆H₁₇₈ N₂O₁₂, molar mass: 2032.92 g/mol.

¹H NMR (500 MHz, CDCl₃, rt) δ [ppm]: 8.10 (t, ⁴ J _HH = 1.9 Hz, 2 H), 7.91 (dt, ³ J _HH = 8.0 Hz, ⁴ J _HH = 1.6 Hz, 2 H), 7.65 (dt, ³ J _HH = 7.7 Hz, ⁴ J _HH = 1.5 Hz, 2 H), 7.52 (t, ³ J _HH = 7.8 Hz, 2 H), 7.05 (s, 1 H), 7.04 (s, 1 H), 7.02 (s, 1 H), 7.01 (s, 1 H), 7.00 (s, 1 H), 6.98 (s, 1 H), 4.09 – 3.98 (m, 24 H), 3.34 (s, 2 H), 1.91 – 1.79 (m, 24 H), 1.61 – 1.46 (m, 24 H), 1.43 – 1.30 (m, 24 H), 1.30 – 1.17 (m, 24 H), 0.94 – 0.82 (m, 36 H).

¹³C NMR (126 MHz, CDCl₃, rt) δ [ppm]: 154.3, 153.9, 153.7, 153.7, 153.5, 152.5, 134.2, 129.3, 125.9, 124.7, 123.2, 118.1, 117.4, 117.3, 117.2, 115.1, 114.8, 114.5, 114.4, 113.8, 112.7, 94.2, 91.8, 91.7, 91.7, 91.4, 87.1, 82.4, 80.2, 69.9, 69.9, 69.8, 69.8, 69.8, 32.1, 31.8, 31.8, 31.8, 31.7, 29.9, 29.5, 29.5, 29.5, 29.4, 29.4, 29.3, 26.0, 25.8, 25.8, 25.8, 25.8, 22.8, 22.8, 22.8, 22.7, 14.3, 14.2, 14.2, 14.2.

MS (MALDI-TOF, DCTB) m/z: 2031.3 [M]⁺, 2281.4 [M + DCTB]⁺.

Calculated exact mass: 2031.34 g/mol.

Synthesis of 7: Under an Ar atmosphere, 6a (78.5 mg, 36 µmol), Pd(PPh₃)₂Cl₂ (7.5 mg, 0.01 mmol), CuI (5.1 mg, 0.03 mmol), and I₂ (16.2 mg, 0.06 mmol) in HN(iPr)₂ (5 mL) and THF (7 mL) were stirred at rt for 16 h. Water and CH₂Cl₂ were added, the aqueous phase was extracted with CH₂Cl₂, and the organic phase was washed with aqueous HCl (1 M), water, and brine and dried over MgSO₄. After evaporation of the solvent, column chromatographic purification (Cy : DCM = 2 : 3 → 1 : 2, R _f = 0.45 (1 : 2)) and additional purification by rec GPC gave 7 as an orange film (37.7 mg, 8.5 µmol, 48%).

Formula: C₂₉₂H₃₉₂ N₆O₂₄Si₂, molar mass: 4426.54 g/mol.

¹H NMR (500 MHz, CDCl₃, rt) δ [ppm]: 8.10 (s, 4 H), 7.91 (d, ³ J _HH = 8.0 Hz, 4 H), 7.65 (d, ³ J _HH = 7.6 Hz, 4 H), 7.52 (t, ³ J _HH = 7.8 Hz, 4 H), 7.06 – 6.92 (m, 24 H), 4.09 – 3.94 (m, 48 H), 2.43 (t, ³ J _HH = 6.9 Hz, 4 H), 1.92 – 1.76 (m, 52 H), 1.62 – 1.46 (m, 48 H), 1.41 – 1.28 (m, 96 H), 1.16 – 1.07 (m, 28 H), 0.95 – 0.80 (m, 76 H).

¹³C NMR (126 MHz, CDCl₃, rt) δ [ppm]: 155.2, 154.5, 153.9, 153.7, 153.7, 153.5, 153.4, 152.6, 134.2, 129.3, 125.9, 125.7, 124.7, 123.2, 120.0, 118.0, 118.0, 117.4, 117.3, 117.3, 116.6, 115.6, 115.0, 114.8, 114.5, 114.4, 113.8, 113.4, 95.3, 94.2, 91.8, 91.8, 87.1, 70.0, 69.9, 69.9, 69.8, 69.3, 32.8, 32.1, 31.8, 31.8, 31.8, 31.7, 30.5, 29.9, 29.5, 29.5, 29.5, 29.4, 29.4, 29.3, 26.0, 26.0, 26.0, 25.8, 25.8, 25.8, 22.8, 22.8, 21.5, 20.9, 18.4, 18.2, 14.2, 12.0, 9.8, 1.2.

MS (MALDI-TOF, DCTB) m/z: 4422.9 [M]⁺, 4673.1 [M + DCTB]⁺, 4923.2 [M + 2DCTB]⁺.

Calculated exact mass: 4422.92 g/mol.

Synthesis of 8: Under an Ar atmosphere, TBAF (1 M in THF, 0.07 mL, 0.07 mmol) was added to 7 (50.0 mg, 11 µmol) in THF (7 mL) and stirred at rt for 12 h. Water and CH₂Cl₂ were added, the aqueous phase was extracted with CH₂Cl₂, and the organic phase was washed with water and brine and dried over MgSO₄. After evaporation of the solvent, column chromatographic purification (Cy : DCM = 1 : 1, R _f = 0.58) and additional purification by rec GPC gave 8 as a yellow film (37 mg, 9 µmol, 81%).

Formula: C₂₇₂H₃₅₄ N₄O₂₄, molar mass: 4063.83 g/mol.

¹H NMR (500 MHz, CDCl₃, rt) δ [ppm]: 8.11 – 8.09 (m, 4 H), 7.94 – 7.88 (m, 4 H), 7.67 – 7.64 (m, 4 H), 7.52 (t, ³ J _HH = 7.8 Hz, 4 H), 7.06 – 6.97 (m, 24 H), 4.09 – 3.98 (m, 48 H), 3.34 (s, 2H), 1.92 – 1.77 (m, 48 H), 1.61 – 1.46 (m, 48 H), 1.42 – 1.29 (m, 96 H), 0.96 – 0.85 (m, 72 H).

¹³C NMR (126 MHz, CDCl₃, rt) δ [ppm]: 155.1, 154.3, 153.9, 153.7, 153.7, 153.7, 153.5, 153.5, 153.3, 152.5, 134.2, 130.9, 129.3, 128.8, 125.9, 124.7, 124.3, 123.2, 118.1, 118.0, 117.4, 117.3, 117.2, 117.2, 115.6, 115.1, 114.8, 114.6, 114.5, 114.4, 114.3, 113.8, 112.7, 94.2, 92.4, 91.8, 91.7, 91.7, 91.5, 91.4, 87.1, 82.4, 80.2, 79.8, 79.5, 69.9, 69.9, 69.8, 69.8, 69.8, 69.7, 31.8, 31.8, 31.8, 31.7, 31.7, 29.5, 29.5, 29.4, 29.4, 29.4, 29.3, 29.3, 26.0, 25.8, 25.8, 25.8, 25.8, 25.8, 22.8, 22.8, 22.8, 22.7, 14.2, 14.2, 1.2.

MS (MALDI-TOF, DCTB) m/z: 4060.7 [M]⁺, 4310.8 [M + DCTB]⁺, 4561.0 [M + 2DCTB]⁺.

Calculated exact mass: 4060.66 g/mol.

Synthesis of 1: Under an Ar atmosphere, 8 (10.0 mg, 2.5 µmol) in THF (20 mL) was purged with Ar for 1 h. By using a syringe pump, this solution was slowly added (72 h) to Pd(PPh₃)₂Cl₂ (17.3 mg, 24.6 µmol), CuI (2.3 mg, 12.3 µmol), and I₂ (3.5 mg, 13.8 µmol) in THF (20 mL) and HN(iPr)₂ (15 mL) at 50 °C and then additionally stirred for 72 h. After cooling to rt, water and CH₂Cl₂ were added, the aqueous phase was extracted with CH₂Cl₂, and the organic phase was washed with water and brine and dried over MgSO₄. After evaporation of the solvent, the crude product was dissolved in CH₂Cl₂, filtered through a plug of silica and purified by rec GPC to give 1 as a yellow film (5 – 6 mg, 1.23 – 1.48 µmol, 50 – 60%; variable yields of different reactions).

Formula: C₂₇₂H₃₅₂ N₄O₂₄, molar mass: 4061.81 g/mol.

¹H NMR (700 MHz, CDCl₃, rt) δ [ppm]: 8.18 (d, ⁴ J _HH = 1.9 Hz, 4 H), 7.94 (dt, ³ J _HH = 8.1 Hz, ⁴ J _HH = 1.4 Hz, 4 H), 7.63 (dt, ³ J _HH = 7.5 Hz, ⁴ J _HH = 1.3 Hz, 4 H), 7.52 (t, ³ J _HH = 7.7 Hz, 4 H), 7.05 (s, 4 H), 7.04 (s, 4 H), 7.02 (s, 4 H), 7.01 (s, 4 H), 6.99 (s, 4 H), 6.99 (s, 4 H), 4.08 – 3.96 (m, 48 H), 1.89 – 1.75 (m, 48 H), 1.59 – 1.46 (m, 48 H), 1.44 – 1.26 (m, 96 H), 0.94 – 0.80 (m, 72 H).

¹³C NMR (176 MHz, CDCl₃, rt) δ [ppm]: 155.1, 153.9, 153.7, 153.7, 153.7, 153.5, 152.3, 129.3, 124.8, 118.0, 117.4, 117.3, 117.3, 115.6, 114.8, 114.6, 114.4, 113.9, 112.7, 94.2, 92.5, 92.2, 91.6, 79.6, 69.9, 69.9, 69.9, 69.8, 45.4, 32.8, 31.8, 31.8, 31.7, 31.7, 31.2, 29.9, 29.5, 29.5, 29.4, 29.4, 29.4, 29.3, 26.7, 25.9, 25.9, 25.8, 25.8, 25.8, 25.8, 22.8, 22.8, 22.8, 22.7, 14.2, 14.2, 14.1, 14.1, 1.2.

MS (MALDI-TOF, DCTB) m/z: 4058.7 [M]⁺, 4308.8 [M + DCTB]⁺.

Calculated exact mass: 4058.64 g/mol.

Funding Information

The authors thank the Deutsche Forschungsgemeinschaft for funding through collaborative grant no. 455 731 873, and RTG-2591 “TIDE-Template-designed Organic Electronics.”

Conflict of Interest

The authors declare no conflict of interest.

• References

• 1 Müllen K, Scherf U. Organic Light-Emitting Devices: Synthesis, Properties and Applications. Wiley VCH; Weinheim: 2006
• 2 Klauk H. Organic Electronics: Materials, Manufacturing and Applications/Organic Electronics II: More Materials and Applications. Wiley VCH; Weinheim: 2006. /2012
• 3 Cicoira F, Santato C. Organic Electronics: Emerging Concepts and Technologies.. Wiley VCH; Weinheim: 2013
• 4 Köhler A, Bässler H. Electronic Processes in Organic Semiconductors: An Introduction. Wiley VCH; Weinheim: 2015
• 5 Schwartz BJ. Nat. Mater. 2008; 7: 427
• 6 Scheblykin IG. Nat. Chem. 2013; 5: 903
• 7a Hu ZJ, Haws RT, Fei Z, Boufflet P, Heeney M, Rossky PJ, Vanden Bout DA. PNAS 2017; 114: 5113
• 7b Raithel D, Simine L, Pickel S, Schötz K, Panzer F, Bederschneider S, Schiefer D, Lohwasser R, Köhler J, Thelakkat M, Sommer M, Köhler A, Rossky PJ, Hildner R. PNAS 2018; 115: 2699
• 8a Noriega R, Rivnay J, Vandewal K, Koch FPV, Stingelin N, Smith P, Toney MF, Salleo A. Nat. Mater. 2013; 12: 1038
• 8b Lupton JM. Adv. Mater. 2010; 22: 1689
• 9 Stangl T, Wilhelm P, Schmitz D, Remmerssen K, Henzel S, Jester S-S, Höger S, Vogelsang J, Lupton JM. J. Phys. Chem. Lett. 2015; 6: 1321
• 10 Stangl T, Bange S, Schmitz D, Würsch D, Höger S, Vogelsang J, Lupton JM. J. Am. Chem. Soc. 2013; 135: 78
• 11 Jester S-S, Schmitz D, Eberhagen F, Höger S. Chem. Commun. 2011; 47: 8838
• 12 Liu S, Schmitz D, Jester S-S, Borys NJ, Höger S, Lupton JM. J. Phys. Chem. B 2013; 117: 4197
• 13 Allolio C, Stangl T, Eder T, Schmitz D, Vogelsang J, Höger S, Horinek D, Lupton JM. J. Phys. Chem. B 2018; 122: 6431
• 14a Giesa R. J. Macromol. Sci., Rev. Macromol. Chem. Phys. 1996; C36: 631
• 14b Tour JM. Chem. Rev. 1996; 96: 537
• 14c Bunz UHF. Chem. Rev. 2000; 100: 1605
• 15a Grave C, Schlüter D. Eur. J. Org. Chem. 2002; 3075
• 15b Höger S. Chem. Eur. J. 2004; 10: 1320
• 15c Zhang W, Moore JS. Angew. Chem. Int. Ed. 2006; 45: 4416
• 15d Iyoda M, Yamakawa J, Rahman MJ. Angew. Chem. Int. Ed. 2011; 50: 10522
• 15e Ball M, Zhang B, Zhong Y, Fowler B, Xiao S, Ng F, Steigerwal M, Nuckolls C. Acc. Chem. Res. 2019; 52: 1068
• 16a Jeschke G, Sajid M, Schulte M, Ramezanian N, Volkov A, Zimmermann H, Godt A. J. Am. Chem. Soc. 2010; 132: 10107
• 16b Hinderer F, May R, Jester S-S, Höger S. Macromolecules 2016; 49: 1816
• 17 Meißner SA, Eder T, Keller TJ, Hofmeister DA, Spicher S, Jester S-S, Vogelsang J, Grimme S, Lupton JM, Höger S. Nat. Commun. 2021; 12: 6614
• 18 Müri M, Schuermann KC, De Cola L, Mayor M. Eur. J. Org. Chem. 2009; 2009: 2562
• 19a Mitscherlich E. Ann. Phys. 1834; 108: 225
• 19b Mitscherlich E. Ann. Pharm. 1834; 12: 311
• 20 Glaser C. Z. Chem. 1866; 9: 308
• 21 Zollinger H. Color Chemistry. Synthesis, Properties, and Applications of Organic Dyes and Pigments.. 3rd ed. Weinheim: Wiley-VCH; 2003
• 22 Hartley GS. Nature 1937; 140: 281
• 23a Hamo F, Djedaini-Pilard F, Barbot F, Len C. Tetrahedron 2009; 65: 10105
• 23b Bandara HMD, Burdette SC. Chem. Soc. Rev. 2012; 41: 1809
• 23c Beharry AB, Woolley GA. Chem. Soc. Rev. 2011; 40: 4422
• 23d Merino E, Ribagorda M. Beilstein J. Org. Chem. 2012; 8: 1071
• 23e Wagner-Wysiecka E, Łukasik N, Biernat JF, Luboch E. J. Inclusion Phenom. Macrocyclic Chem. 2018; 90: 189
• 23f Reuter R, Wegner HA. Chem Commun. 2011; 47: 12267
• 24 Lee S, Hua Y, Flood AH. J. Org. Chem. 2014; 79: 8383
• 25 Jester S-S, Schmitz D, Eberhagen F, Höger S. Chem. Commun. 2011; 47: 8838 For an alternative synthetic approach, see the respective section in the Supporting Information of the present publication
• 26 Reuter R, Hostettler N, Neuburger M, Wegner HA. Eur. J. Org. Chem. 2009; 2009: 5647
• 27 Cojal González JD, Iyoda M, Rabe JP. Angew. Chem. Int. Ed. 2018; 57: 17038
• 28 Zeitouny J, Belbakra A, Llanes-Pallas A, Barbieri A, Armaroli N, Bonifazi D. Chem. Commun. 2011; 47: 451
• 29 Eder T, Vogelsang J, Bange S, Remmerssen K, Schmitz D, Jester S-S, Keller TJ, Höger S, Lupton JM. Angew. Chem. Int. Ed. 2019; 58: 18898
• 30 Liu X, Chi W, Qiao Q, Kokate SV, Cabrera EP, Xu Z, Liu X, Chang Y-T. ACS Sens. 2020; 5: 731
• 31 Freixas VM, Wilhelm P, Nelson T, Hinderer F, Höger S, Tretiak S, Lupton JM, Fernandez-Alberti S. J. Phys. Chem. A. 2021; 125: 8404
• 32 Hofkens J, Cotlet M, Vosch T, Tinnefeld P, Weston Kenneth D, Ego C, Grimsdale A, Müllen K, Beljonne D, Brédas JL, Jordens S, Schweitzer G, Sauer M, De Schryver F. PNAS 2003; 100: 13146
• 33 Höger S. Liebigs Ann. Recl. 1997; 1997: 273
• 34a Höger S, Bonrad K. J. Org. Chem. 2000; 65: 2243
• 34b Gaefke G, Höger S. Synthesis 2008; 14: 2155
• 35 Lee S, Hua Y, Flood AH. J. Org. Chem. 2014; 79: 8383

Publication History

Received: 12 July 2022

Accepted after revision: 30 August 2022

Accepted Manuscript online:
12 September 2022

Article published online:
26 October 2022