Heptagon-Containing Saddle-Shaped Nanographenes: Self-Association and Complexation Studies with Polycyclic Aromatic Hydrocarbons and Fullerenes

Key words

nanographenes - contorted aromatics - supramolecular chemistry - host–guest systems - heptacyclic polyarenes - molecular recognition

Introduction

Polycyclic aromatic hydrocarbons (PAHs) have demonstrated to be versatile actors in the field of supramolecular chemistry mainly by virtue of the establishment of π–π and hydrophobic interactions.[1] Their scope ranges from playing the role of host to guest, and even both simultaneously, as in the case of self-association processes. Planar systems are the most broadly explored by far, with examples such as the self-association study of dodecyl-chained hexa-peri-hexabenzocoronene (HBC) derivatives reported by Müllen and coworkers.[2] In this case, the association resulted in an upfield shift of the ¹H NMR signals upon concentration increase, phenomenon which was subsequently proved to be solvent-dependent,[3] evidencing the influence of the solvophobic effect on the association process. As a result of these supramolecular interactions, diverse applications of the self-association of planar PAHs and nanographenes (NGs) were developed encompassing topics such as the formation of discotic liquid crystals,[4] their implementation in photovoltaic systems,[5] [6] supramolecular nanotubes or nanofibers[7] [8] displaying relevant optoelectronic,[9] [10] [11] sensing[12] [13] and spintronic properties,[14] and as organogelators,[15] among others. When it comes to planar PAHs being part of the structure of supramolecular hosts, we find multifarious examples of not only 2D cyclophanes,[16] such as nanohoops encapsulating C₆₀ or C₇₀,[17] [18] [19] [20] [21] but also other kinds of architectures such as metal–organic cages.[22] [23] [24] Likewise, planar PAHs such as pyrene or even coronene have been employed as guests for different molecular receptors,[25] [26] [27] [28] [29] [30] which is also true for positively curved systems, whose most prominent representative is corannulene.[29] [31] [32] [33]

The introduction of a curvature within the structure of PAHs and NGs makes them feature higher solubilities on account of the weakening of the π–π stacking. However, this curvature provides, in contrast, access to better host–guest shape complementarities with other curved systems. Corannulene, for instance, binds to C₆₀ [34] [35] and its incorporation in receptors enables their binding to fullerenes.[36] [37] [38] It is also involved in applications derived from its self-association such as liquid crystals,[39] organogelators,[40] and supramolecular polymer formation.[41] Besides, self-aggregation studies of hydrophilic analogues of corannulene functionalized with nucleosides allowed for an uncommon examination in water media.[42]

As opposed to NGs featuring a bowl-shaped positive curvature, literature related to self-association and host–guest behavior of negatively curved saddle-shaped NGs[43] [44] containing only heptagonal carbocycles as nonhexagonal rings remains almost inexistent. Theoretically, Wheeler and coworkers pointed to a better self-association of [7]circulene amongst their smaller and larger [n]circulene congeners with n = 6–10.[45] Miao and coworkers described an elusive co-crystallization process of a heptacycle-containing NG with C₆₀.[46] In addition to that, in our group we have recently reported the design, synthesis, and use as a selective C₇₀ supramolecular receptor of a cyclophane comprising two heptagon-containing HBC analogues (hept-HBCs).[47]

Here we present the first systematic study of the supramolecular behavior of five differently functionalized saddle-shaped hept-HBCs ([Figure 1], 1–5), by examining their self-association as well as their host–guest abilities towards both planar and curved π-systems.

Figure 1 Structures of the heptagon-containing saddle-shaped HBC derivatives 1–5.

Results and Discussion

Among the collection of hept-HBCs synthesized, the heptagonal carbocycle is either functionalized with carbonyl groups (1, 3, 4), constituting a tropone unit, a methylene (2), or with four additional fused rings extending the π system (5). The periphery of the HBCs was decorated either with tert-butyl or phenyl groups or hydrogen atoms in proximal ([Scheme 1], C_e) or distal positions ([Scheme 1], C_h) with respect to the heptacycle. The keystone in the synthesis of all these contorted analogues, developed in our research group, is based on a Co-catalyzed alkyne cyclotrimerization resulting in the simultaneous formation of both the central benzene and the heptacycle rings, followed by a final Scholl cyclodehydrogenation[48] generating the hept-HBC skeleton.[49]

Scheme 1 a) Synthesis of heptagon-containing nanographenes 2–5. Reagents and conditions: i) Tebbe reagent (0.5 M in toluene), THF, 0 °C to r.t., 2 h, 98%; ii) PPh₃, CBr₄, toluene, reflux, 28 h, 83%; iii) 4-tert-butylphenylboronic acid, Pd(PPh₃)₄, K₂CO₃, toluene, EtOH/H₂O, 100 °C, 20 h, 84% (see the Supporting Information); iv) FeCl₃, 1,2-dichloroethane, CH₃NO₂, 70 °C, 48 h, 96%; v) Pd(PPh₃)₄, K₂CO₃, toluene/H₂O/EtOH, reflux, 16 h, 56% (for 3) ; vi) phenylboronic acid, Pd(PPh₃)₄, K₂CO₃, toluene/H₂O/EtOH, reflux, 20 h, 70% (for 4).[47] b) DFT-optimized structures (ωB97XD/def2SVP in CHCl₃) of: 1 (left), 3 (middle), and 5 (right). H atoms have been omitted for clarity.

Synthesis of compound 1 was achieved following a procedure described in our group.[49] Subsequent Tebbe olefination yielded hept-HBC 2 in 98% yield ([Scheme 1], i). Wittig-like reaction over 1 in the presence of CBr₄ and PPh₃ afforded 1,1-dibromoalkene 6 in 83% yield ([Scheme 1], ii). Subsequent Suzuki cross-coupling reaction with 4-tert-butylphenylboronic acid on intermediate 6 resulted in the dicoupled product in 84% yield (see the Supporting Information), followed by a cyclodehydrogenation reaction using classical FeCl₃ conditions providing extended NG 5 in excellent 96% yield ([Scheme 1], iii–iv).[50] On the other hand, compounds 3 and 4 [47] were successfully synthesized from precursor 7,[51] recently reported by our research group. The presence of two bromine atoms in 7 is an appropriate launch pad for further derivatization, which was indeed leveraged in respective reactions under Suzuki cross-coupling conditions without and with phenylboronic acid, giving rise to 3 and 4 in 56% and 70% yields, respectively ([Scheme 1], v–vi).

The excellent solubility of compounds 1–5 in CDCl₃ allowed for their full characterization by means of both 1D and 2D NMR techniques, which enabled the complete assignment of all signals. These data were further supported by HRMS experiments with exact masses and isotopic distributions confirming the proposed structures (for more details, see the Supporting Information).

Derivatives 1–5 were also studied using UV-vis spectroscopy. Derivatives 1–4 show a similar spectrum with the main absorption features in the 300–400 nm region ([Figures S128], [S130], [S132], and [S134] in the Supporting Information). In all cases, the main band has its maximum at around 350–360 nm and exhibits some vibronic structure, less resolved in the case of 1. Additionally, there is a small band or shoulder centered at 382–391 nm. The position of the substituents does not seem to have much influence on the absorption as the λ_max slightly changes when they are attached in different positions (356 nm for 1, 354 nm for 3). In addition, the nature of the double bond on the heptagonal ring has a slightly more pronounced effect. Replacing the C = O for a C = C group results in a slight hypsochromic shift of the main absorption band (356 nm for 1, 351 nm for 3). On the other hand, the inclusion of aromatic rings as substituents induces an 11 nm bathochromic shift, which can be attributed to some extra delocalization of the π system. Compound 5 has a completely different UV-vis spectrum. It displays a broad absorption between 300 and 450 nm with maxima at 323, 360, and 413 nm and a tail up to ca. 570 nm ([Figure S136]). This absorption at longer wavelengths is in agreement with the more extended π-surface of this system in comparison with compounds 1–4.

Additionally, hept-HBCs 1–5 were investigated theoretically by density functional theory (DFT) computational studies at the ωB97XD/def2SVP or B3LYP/6-31G(d,p) level of calculation in CHCl₃, both delivering similar results (for more details see the Supporting Information). Optimized structures revealed, as expected, a saddle-shape curvature, induced by the presence of the heptacycle in NGs 1–5. Moreover, additional torsion is shown in the case of 5 owing to steric hindrance between the hydrogen atoms in the cove region. Compound 5 shows, besides, larger dimensions than the rest (16.1 × 11.0 vs. 11.4 × 10.5 Å), which might foster more effective complexations with larger π-systems.

Studies to evaluate the self-association equilibria for species 1–5 were conducted via ¹H NMR titrations in CDCl₃ solution at concentrations ranging from 1 to 100 mM. Upon increasing the concentration of the monomers 1–5 during the self-association titrations, an upfield shift is experienced by most of the NG protons (see [Figure 2] and [Figures S37–S56] in the Supporting Information). A monomer–dimer association model (K _d, [Eq. 1])[52] or an indefinite equal K self-association model (K _E, [Eq. 2])[53] was considered due to the unknown nature of the aggregates formed. The constants were determined by a nonlinear least-squares fitting method through [Eq. 1] or [Eq. 2]. The fitting to these models led to association constants summarized in [Table 1], giving rise to values ranging from 1.5 to 24 M⁻¹ according to the monomer–dimer model or from 3.1 to 47.3 M⁻¹ according to the indefinite one. All values are similar except for 4, which stands out among their analogues 1, 2, 3, and 5. This observation matches the augmented π expansion of 4, provided by the appended phenyl rings in C_e positions, which maximizes the π interactions between the two monomers. As a result of these low binding constants, the extent of self-association covered in the titrations is far from being complete and this process is relevant at high concentrations, as shown by the calculated α_agg values (see the Supporting Information).[54] Only compound 4, which establishes stronger interactions, shows a significant degree of self-association at low concentrations.

Figure 2 Self-association experiment of nanographene 1. Aromatic region of the ¹H NMR (400 MHz, CDCl₃, 298 K) spectra of 1 at different concentrations. Inset: nonlinear least-squares fitting of the changes in the δ (400 MHz, CDCl₃, 298 K) of H_h upon concentration change using [Eq. 1] (K _d = 3.0 ± 0.2 M⁻¹). Color coding and labels are defined in [Scheme 1].

Equation 1 C denotes the concentration; δ is the observed chemical shift; δ _m is the chemical shift for the monomer; Δδ (Δδ = δ _d  − δ _m) stands for the change in chemical shift from the monomer to the dimer; and K _d represents the association constant for the dimer formation.[52]

Equation 2 C denotes the concentration; δ is the observed chemical shift; δ _m is the chemical shift for the monomer; Δδ (Δδ = δ _s − δ _m) stands for the change in chemical shift from the monomer to the molecule in the stack; and K _E represents the self-association constant.[53]

Table 1 Self-association constants for NGs 1–5 [a]

NG	1	2	3	4	5
K d (m −1) [b]	3.0 ± 0.2	1.5 ± 0.5	6.0 ± 0.6	24 ± 4[d]	6.7 ± 2.0
K E (m −1) [c]	6.2 ± 0.4	3.1 ± 1.1	12.0 ± 1.2	47.3 ± 8	13.3 ± 4.0

^a Measured by ¹H NMR in CDCl₃ at 298 K.

^b Using [Eq. 1].

^c Using [Eq. 2].

^d From Ref. 47.

[Eq. 1] and [Eq. 2] are equivalent, being K _E = 2K _d. As a result, distinguishing between association models from the experimental ¹H NMR data is not possible[53] and the same fitting was obtained in both cases. Therefore, we cannot unambiguously affirm which of the models describes better the behavior of our system. Nevertheless, an analysis of the expected size of the aggregates formed under this isodesmic model according to [Eq. 3] showed that at 0.1 M, the highest concentration used in this work, the number average size of the assemblies is below 2 for all compounds, except for 4, for which a slightly higher value of ca. 2.7 was calculated (see the Supporting Information, [Table S2]). In this situation, an equal constant isodesmic model predicts that the species present even at a high concentration are mainly monomers and dimers, except again for compound 4, in which higher assemblies can be significantly populated.

Equation 3 N is the number average aggregate size, C denotes the total concentration; K _E represents the self-association constant in the isodesmic model.[54]

Analysis of the extension of the shifting of the ¹H NMR signals evidences that the position where the ^t Bu groups are attached to the hept-HBC core plays a key role in the observed chemical shifts, i.e., when located on the C_h carbon atoms, the major shift was experienced by the H nuclei closer to the tropone moiety (Δδ _Hc = −0.73 ppm; Δδ _Hd = −0.90 ppm; Δδ _He = −0.63 ppm for 3), whilst when attached to the C_e the influence on the chemical shift is higher for the planar part of the molecules (Δδ _Hi = −0.76 ppm; Δδ _Hh = −0.59 ppm; Δδ _Hg = −0.48 ppm for 1). However, the modification of the tropone unit by its conversion into the methylidene or the fused diphenylene motifs caused low impact in the resulting self-association process.

Once inquired into the self-association process of saddle-shaped hept-HBCs 1–5, further investigations on their complexation with a selection of guests of different geometry and electronic nature were accomplished. Among the flat guests, we proposed pyrene and benzo[a]pyrene as planar nonfunctionalized PAHs, naphthalene diimide (NDI) 8 as a π-acceptor,[55] 1,5-dialkoxynaphthalene 9 as a π-donor, and corannulene, C₆₀, and C₇₀ as curved guests ([Figure 3]). The association constants (K _a) between NGs and PAHs or fullerenes were determined by ¹H or ¹³C NMR titrations in CDCl₃ or o-DCB-d ₄ at r.t. The NMR data were analyzed using a nonlinear least-squares curve fitting procedure performed with the online software Bindfit [56] with a 1:1 global fitting model (Nelder–Mead method).[57] Considering the geometry of the hept-HBCs and the guests involved in the binding equilibria, other models such as a 1:2 stoichiometry cannot be ruled out. However, attempts to fit the data to other models proved unsuccessful, as meaningless results were obtained. This is not surprising taking into account the low association observed, which makes binding constants in a 1:2 model very difficult to determine reliably. For this reason a 1:1 model was assumed to obtain an estimation of the binding constants.

Figure 3 Structures of the guests used in this work.

When we glance at K _a results,[58] summarized in [Table 2], we conclude that, in general, the best results for the planar guests were found for the complexes assembled with both the electron acceptor and donor 8 and 9, respectively. On the other hand, the lack of complexation between hosts 3 and 4 and pyrene as a guest, both bearing ^t Bu groups in C_h positions, hints at a disfavored binding due to this structural feature. Nevertheless, recognition of benzo[a]pyrene was of the same magnitude for all hosts, while again for NDI 8, the binding is better for derivatives 1 and 2, with ^t Bu groups in the C_e position. Besides, dialkoxynaphthalene 9 reached the maximum K _a value with guest 5, and, overall, these NGs, except for host 2, are more prone to complex electron-rich PAHs.

Table 2 Association constants (K _a, in M^{− 1}) between hept-HBCs 1–5 and selected guests

Hept-HBC host	Pyrene[a]	Benzo[a] pyrene[a]	NDI (8)[a]	1,5-Dialkoxy naphthalene (9)[a]	Corannulene[a]	C60 [b]	C70 [b]
1	20.9 ± 0.8	4.07 ± 0.08	17.7 ± 0.3	15.9 ± 0.8	13.5 ± 0.7	12.3 ± 0.2	50.2 ± 2.2
2	8.03 ± 0.11	6.05 ± 0.06	21.8 ± 0.7	7.96 ± 0.34	13.4 ± 0.7	15.8 ± 0.2	35.3 ± 1.3
3	<0.01	3.67 ± 0.06	9.40 ± 0.13	36.6 ± 2.6	8.40 ± 0.53	8.13 ± 0.07	28.3 ± 0.7
4	<0.01	4.73 ± 0.12	7.26 ± 0.11	18.4 ± 1.2	<0.01	20.0 ± 0.1[c]	3.75 ± 0.02[c]
5	8.04 ± 0.18	6.48 ± 0.13	17.7 ± 0.2	65.7 ± 2.1	10.9 ± 0.5	18.5 ± 0.3	53.1 ± 2.4

^a Measured by ¹H NMR in CDCl₃.

^b Measured by ¹³C NMR in o-DCB-d ₄.

^c From Ref. 47.

Furthermore, from the variation of the ¹H NMR signals during the titrations, it is inferred that the interactions between hept-HBCs 1–5 with pyrene, benzo[a]pyrene, and NDI 8 ([Figure 4] and [Figures S57–S60], [S67–S72], [S77–S82], [S87–S92], [S97–S102] in the Supporting Information) take place in the more planar part of the hosts. In the case of electron-rich guest 9, the changes on chemical shift are minimal (up to |0.02| ppm), which hampers the clear correlation between shift and host–guest interaction location (see [Figures S63], [S64], [S73], [S74], [S83], [S84], [S93], [S94], [S103], and [S104] in the Supporting Information).

Figure 4 Aromatic region of the ¹H NMR (400 MHz, CDCl₃, 298 K) spectra for the titration of 1 with 8 (0–11.8 equiv). Inset: fitted binding isotherm using a 1:1 association model (K _a = 17.7 ± 0.3 M⁻¹) showing the change in the chemical shift for H_h. Color coding and labels are defined in [Scheme 1].

When we evaluate the interaction with the first curved guest, corannulene, similar K _a values were found for all guests aside from 4, with which no recognition was found (see [Figures S65], [S66], [S75], [S76], [S85], [S86], [S95], [S96], [S105], and [S106] in the Supporting Information). Finally, among the binding abilities of hept-HBCs 1–5 with fullerenes C₆₀ and C₇₀ (see [Figures S107–S122] in the Supporting Information), it is worth mentioning that a K _a of ca. 53 M⁻¹ is observed between π-extended 5 and C₇₀, and, as a trend, there is a clear preference for C₇₀ over C₆₀, except for host 4, conceivably due to the presence of the phenyl rings on position C_e. Last, a comparison between the affinity of hosts 1 and 3 towards C₇₀ points to a preference for the compound with the ^t Bu groups closer to the troponic carbonyl group of the molecule.

Conclusions

Five saddle-shaped hept-HBCs 1–5 were synthesized and fully characterized. Moreover, their self-association properties were proven, finding the position and the nature of the peripheral groups to play an important role in the aggregation of these compounds. On the contrary, the enhancement of the distortion of these contorted HBC derivatives or the modification of the tropone unit revealed to have little effect on the self-association properties. Besides, the complexation ability of contorted NGs 1–5 towards PAHs and fullerenes was demonstrated, with its maximum exponent in the recognition between π-extended host 5 and electron donor guest 9 (K _a = 66 M⁻¹) and fullerene C₇₀ (K _a = 53 M⁻¹). Thus, this study points at a promising future of the supramolecular chemistry of negatively curved hept-HBCs, as confirmed by their self-association and the sensing of PAHs and fullerenes.

Experimental Section

Experimental Details

Unless otherwise noted, commercially available reagents, solvents, and anhydrous solvents were used as purchased without further purification. Anhydrous THF was freshly distilled over Na/benzophenone. Pd(PPh₃)₄,[59] and compounds 1,[49] 4,[47] 7,[51] and 8 [55] were prepared according to literature procedures.

TLC was performed on Merck Silica gel 60 F₂₅₄ aluminum sheets. The TLC plates were stained with potassium permanganate (1% w/v in water) or observed under UV light when applicable. Flash column chromatography was performed with Silica gel 60 (VWR, 40–63 μm).

¹H and ¹³C NMR spectra were recorded at room temperature on a Varian Direct Drive (400 or 500 MHz), Bruker Avance III HD NanoBay (400 MHz), or Bruker Avance Neo (400 or 500 MHz) spectrometers at a constant temperature of 298 K. Chemical shifts are given in ppm and referenced to the signal of the residual protiated solvent (¹H: δ = 7.26 for CDCl₃) or the ¹³C signal of the solvents (¹³C: δ = 77.16 for CDCl₃ or δ = 132.39 for o-DCB-d ₄) or to the signal of the residual TMS (¹H: δ = 0.00). Coupling constant (J) values are given in Hz. Abbreviations indicating multiplicity are as follow: m = multiplet, p = quintet, q = quartet, t = triplet, d = doublet, dd = doublet of doublets, td = triplet of doublets, s = singlet, br = broad. Signals were assigned by means of 2D NMR spectroscopy (COSY, heteronuclear single-quantum correlation spectroscopy, heteronuclear multiple bond correlation spectroscopy).

Electrospray (ESI) HRMS spectra were recorded on a Waters Xevo G2-XS QTOF or on a Bruker Maxis II spectrometer. MALDI mass spectra were recorded on a Bruker Ultraflex III mass spectrometer. IR spectra were recorded with a Perkin–Elmer Spectrum Two FTIR ATR spectrometer.

Self-Association Studies

Solutions of NGs at different concentrations (1–100 mM) were prepared in CDCl₃ using volumetric flasks and volumetric pipettes. The ¹H NMR spectra at each concentration were recorded.

PAH-Binding Studies

For the titrations with PAHs, a solution of the corresponding hept-HBC derivative was prepared in CDCl₃ using a micropipette. Then, the solution of the corresponding PAH was prepared in another vial using the solution of the NG as a solvent in order to maintain a constant concentration of hept-HBC during the titration experiment. The addition of the solution of the PAH to the NG solution (450 μL) was carried out with Hamilton^® syringes typically using the following order: 4 × 3, 2 × 6, 2 × 12, 3 × 24, 3 × 120 μL. After each addition, the solution was shaken for 30 seconds and the ¹H NMR spectrum was recorded.

Fullerene-Binding Studies

For the titrations with fullerene, a solution of the corresponding fullerene was prepared in o-DCB-d ₄ using a micropipette. Then, the solution of the corresponding hept-HBC was prepared in another vial using the solution of the fullerene as a solvent in order to maintain a constant concentration of fullerene during the titration experiment. The addition of the solution of the NG to the fullerene solution (500 μL) was carried out with Hamilton^® syringes typically using the following order: 8 × 16, 2 × 32, 4 × 64 μL (for C₆₀) and 1 × 16, 4 × 32, 4 × 64, 1 × 90 μL (for C₇₀). After each addition, the solution was shaken for 30 seconds and the ¹³C NMR spectrum was recorded.

Computational Methods

DFT theoretical calculations were performed at the B3LYP/6-31G(d,p) or ωB97XD/def2SVP levels for the five heptagon-containing NG analogues using the Gaussian 09 software package.[60] Chloroform was used as a solvent, applying the polarizable continuum model with the integral equation formalism (IEFPCM) implemented in Gaussian 09. Frequency calculations were performed to confirm the optimized structures corresponded to energy minima.

Procedures

Compound 2

To a degassed solution of 1 (50 mg, 0.075 mmol) in freshly distilled anhydrous THF (10 mL), cooled in a water-ice bath, was added the Tebbe reagent (0.5 M in toluene, 0.20 mL, 0.10 mmol). The solution was stirred for 5 min at 0–4 °C and 15 min at r.t. The round-bottom flask was again immersed in a water-ice bath and Tebbe reagent (0.5 M in toluene, 0.20 mL, 0.10 mmol) was added. The solution was stirred for 5 min at 0–4 °C and 15 min at r.t. This operation was repeated another time and the solution was stirred for 1 h at r.t. Subsequently, NaOH_(aq) (1 M; 10 mL) was added to quench the reaction. The resulting mixture was diluted with H₂O (20 mL) and extracted with CH₂Cl₂ (3 × 30 mL). The combined organic phases were dried over Na₂SO₄ and the solvent was evaporated under reduced pressure. The crude material was purified by column chromatography (SiO₂, CH₂Cl₂/hexane 10:90 then 20:80) to yield 2 (49 mg, 98%) as a yellow solid.

¹H NMR (500 MHz, CDCl₃): δ = 8.88–8.84 (m, 6 H, H_c+d+i), 8.79 (m, 4 H, H_f+g), 7.95 (t, J = 7.7 Hz, 2 H, H_h), 7.87 (t, J = 7.6 Hz, 2 H, H_b), 7.68 (d, J = 6.4 Hz, 2 H, H_a), 5.08 (s, 2 H, H_CH2), 1.64 (s, 18 H, H ^t _Bu).

¹³C NMR (126 MHz, CDCl₃): δ = 152.92, 149.80, 143.95, 131.45, 130.57, 130.00, 129.77, 128.96, 128.64, 127.91, 126.97, 125.85, 124.90, 124.28, 123.50, 123.12, 122.53, 121.92, 121.53, 120.70, 120.40, 118.18, 115.28, 35.67, 31.95.

IR (neat): 2956, 1613, 1588, 1462, 1368, 1256, 1078 cm⁻¹.

HRMS (ESI⁺): m/z [M + Na]⁺ calcd for C₅₂H₃₆Na: 683.2715; found: 683.2737.

Compound 6

To a degassed solution of 1 (346 mg, 0.522 mmol) in anhydrous toluene (20 mL) were added PPh₃ (1.30 g, 4.95 mmol) and CBr₄ (865 mg, 2.61 mmol). The suspension was refluxed for 28 h. The solvent was removed under reduced pressure and the crude material was purified by column chromatography (SiO₂, CH₂Cl₂/hexane 20:80) to afford 6 (355 mg, 83%) as a yellow solid.

¹H NMR (500 MHz, CDCl₃): δ = 8.91 (m, 4 H), 8.81 (m, 6 H), 7.94 (m, 4 H), 7.77 (d, J = 7.2 Hz, 2 H), 1.65 (s, 18 H).

¹³C NMR (126 MHz, CDCl₃): δ = 150.01, 147.31, 140.96, 132.14, 130.58, 130.04, 130.02, 128.30, 128.21, 127.80, 127.12, 124.81, 124.63, 123.97, 123.63, 123.09, 122.50, 122.05, 121.67, 120.87, 120.76, 118.11, 90.43, 35.74, 31.97.

IR (neat): 2957, 1675, 1612, 1588, 1463, 1393, 1078, 811 cm⁻¹.

HRMS (MALDI⁺): m/z [M]⁺ calcd for C₅₂H₃₄Br: 816.1022; found: 816.1015.

Compound 5

A degassed solution of S1 (see the Supporting Information) (128 mg, 0.138 mmol) in 1,2-dichloroethane (180 mL) was split in six different 50 mL round-bottom flasks. These solutions were heated to 70 °C and subsequently, in each flask was added a degassed solution of FeCl₃ (75 mg) in dry CH₃NO₂ (500 μL) portionwise. These solutions were further stirred for 48 h at 70 °C. The six resulting mixtures were combined, diluted with CH₂Cl₂ (100 mL) and washed with brine (150 mL). The organic layer was then dried over Na₂SO₄ and the solvent was evaporated under vacuum. The crude material was purified by column chromatography (SiO₂, CH₂Cl₂/hexane 20:80) to give 5 (124 mg, 96%) as an orange solid.

¹H NMR (500 MHz, CDCl₃): δ = 8.94 (d, J = 1.8 Hz, 2 H, H_f), 8.89 (d, J = 7.9 Hz, 2 H, H_i), 8.75 (d, J = 8.0 Hz, 2 H, H_g), 8.42 (d, J = 1.8 Hz, 2 H, H_d), 8.37 (d, J = 2.0 Hz, 2 H, H_j), 8.30 (d, J = 8.6 Hz, 2 H, H_c), 8.20 (m, 4 H, H_b+k), 7.97 (t, J = 7.7 Hz, 2 H, H_h), 7.61 (dd, J = 8.8, 2.0 Hz, 2 H, H_l), 1.66 (s, 18 H, H_n), 1.48 (s, 18 H, H_m).

¹³C NMR (126 MHz, CDCl₃): δ = 149.98, 149.82, 134.53, 133.93, 131.01, 130.76, 130.49, 130.03, 130.00, 129.54, 129.53, 128.82, 128.25, 127.03, 126.68, 126.43, 125.23, 124.74, 124.35, 124.08, 122.48, 122.38, 122.13, 121.38, 121.36, 119.31, 119.16, 118.17, 116.99, 35.73, 35.27, 32.02, 31.62.

IR (neat): 3390 (br), 2957, 2922, 2852, 1612, 1589, 1462, 1363, 1262, 1093, 1024 cm⁻¹.

HRMS (MALDI⁺): m/z [M]⁺ calcd for C₇₂H₅₆: 920.4377; found: 920.4383.

Compound 3

To a degassed solution of 7 (91 mg, 0.11 mmol) in toluene (6 mL) were added Pd(PPh₃)₄ (38 mg, 0.033 mmol), K₂CO₃ (306 mg, 2.22 mmol), and a degassed mixture of EtOH/H₂O (3:1, 4 mL). The mixture was refluxed for 16 h. Subsequently, HCl_(aq) (5%, 30 mL) was added and the resulting mixture was extracted with CH₂Cl₂ (3 × 50 mL). The combined organic phases were dried over Na₂SO₄ and the solvent was removed under reduced pressure. The crude material was purified by column chromatography (SiO₂, CH₂Cl₂/hexane 60:40) to afford 3 (41 mg, 56%) as a yellow solid.

¹H NMR (500 MHz, CDCl₃): δ = 9.13 (d, J = 1.7 Hz, 2 H, H_i), 8.76 (d, J = 1.7 Hz, 2 H, H_g), 8.54 (t, J = 4.9 Hz, 2 H, H_c), 8.49 (d, J = 7.6 Hz, 2 H, H_f), 8.23 (d, J = 7.8 Hz, 2 H, H_d), 7.66 (m, 6 H, H_a+b+e), 1.77 (s, 18 H, H ^t _Bu).

¹³C NMR (126 MHz, CDCl₃): δ = 202.60, 149.97, 142.07, 130.89, 129.90, 129.81, 128.15, 127.24, 127.03, 126.82, 126.54, 124.76, 124.16, 123.87, 123.03, 122.71, 122.59, 120.56, 120.48, 119.45, 118.62, 35.84, 32.11.

IR (neat): ν = 2955, 2924, 2862, 1671, 1609, 1588, 1392, 1362, 1336, 1255 cm⁻¹.

HRMS (ESI⁺): m/z [M + Na]⁺ calcd for C₅₁H₃₄ONa: 685.2507; found: 685.2530; m/z [M + H]⁺ calcd for C₅₁H₃₅O: 663.2688; found: 663.2697.

• References

• 1 Steed JW, Atwood JL. Supramolecular Chemistry, 2nd Edition. John Wiley & Sons; Chichester: 2009
• 2 Wu J, Fechtenkötter A, Gauss J, Watson MD, Kastler M, Fechtenkötter C, Wagner M, Müllen K. J. Am. Chem. Soc. 2004; 126: 11311
• 3 Kastler M, Pisula W, Wasserfallen D, Pakula T, Müllen K. J. Am. Chem. Soc. 2005; 127: 4286
• 4 Herwig P, Kayser CW, Müllen K, Spiess HW. Adv. Mater. 1996; 8: 510
• 5 Schmidt-Mende L, Fechtenkötter A, Müllen K, Moons E, Friend RH, MacKenzie JD. Science 2001; 293: 1119
• 6 Wong WW. H, Subbiah J, Puniredd SR, Purushothaman B, Pisula W, Kirby N, Müllen K, Jones DJ, Holmes AB. J. Mater. Chem. 2012; 22: 21131
• 7 Hill JP, Jin W, Kosaka A, Fukushima T, Ichihara H, Shimomura T, Ito K, Hashizume T, Ishii N, Aida T. Science 2004; 304: 1481
• 8 Kulkarni C, Munirathinam R, George SJ. Chem. Eur. J. 2013; 19: 11270
• 9 Yamamoto Y, Fukushima T, Jin W, Kosaka A, Hara T, Nakamura T, Saeki A, Seki S, Tagawa S, Aida T. Adv. Mater. 2006; 18: 1297
• 10 Yamamoto Y, Fukushima T, Suna Y, Ishii N, Saeki A, Seki S, Tagawa S, Taniguchi M, Kawai T, Aida T. Science 2006; 314: 1761
• 11 Treier M, Liscio A, Mativetsky JM, Kastler M, Müllen K, Palermo V, Samorì P. Nanoscale 2012; 4: 1677
• 12 Mogera U, Gedda M, George SJ, Kulkarni GU. ACS Appl. Mater. Interfaces 2017; 9: 32065
• 13 Mogera U, Sagade AA, George SJ, Kulkarni GU. Sci. Rep. 2014; 4: 4103
• 14 Kulkarni C, Mondal AK, Das TK, Grinbom G, Tassinari F, Mabesoone MF. J, Meijer EW, Naaman R. Adv. Mater. 2020; 32: 1904965
• 15 Ito S, Herwig PT, Böhme T, Rabe JP, Rettig W, Müllen K. J. Am. Chem. Soc. 2000; 122: 7698
• 16 Li G, Matsuno T, Han Y, Phan H, Wu S, Jiang Q, Zou Y, Isobe H, Wu J. Angew. Chem. Int. Ed. 2020; 59: 9727
• 17 Lu D, Zhuang G, Wu H, Wang S, Yang S, Du P. Angew. Chem. Int. Ed. 2017; 56: 158
• 18 Cui S, Zhuang G, Lu D, Huang Q, Jia H, Wang Y, Yang S, Du P. Angew. Chem. Int. Ed. 2018; 57: 9330
• 19 Huang Q, Zhuang G, Jia H, Qian M, Cui S, Yang S, Du P. Angew. Chem. Int. Ed. 2019; 58: 6244
• 20 Lu D, Huang Q, Wang S, Wang J, Huang P, Du P. Front. Chem. 2019; 7: 668
• 21 Xu Y, von Delius M. Angew. Chem. Int. Ed. 2020; 59: 559
• 22 Suzuki K, Takao K, Sato S, Fujita M. J. Am. Chem. Soc. 2010; 132: 2544
• 23 Ronson TK, League AB, Gagliardi L, Cramer CJ, Nitschke JR. J. Am. Chem. Soc. 2014; 136: 15615
• 24 Ronson TK, Meng W, Nitschke JR. J. Am. Chem. Soc. 2017; 139: 9698
• 25 Yamashina M, Tanaka Y, Lavendomme R, Ronson TK, Pittelkow M, Nitschke JR. Nature 2019; 574: 511
• 26 Dale EJ, Vermeulen NA, Juríček M, Barnes JC, Young RM, Wasielewski MR, Stoddart JF. Acc. Chem. Res. 2016; 49: 262
• 27 Liu XT, Wang K, Chang Z, Zhang YH, Xu J, Zhao YS, Bu XH. Angew. Chem. Int. Ed. 2019; 58: 13890
• 28 Lozano D, Álvarez-Yebra R, López-Coll R, Lledó A. Chem. Sci. 2019; 10: 10351
• 29 Ibáñez S, Peris E. Angew. Chem. Int. Ed. 2020; 59: 6860
• 30 Blanco V, García MD, Terenzi A, Pía E, Fernández-Mato A, Peinador C, Quintela JM. Chem. Eur. J. 2010; 16: 12373
• 31 Schmidt BM, Osuga T, Sawada T, Hoshino M, Fujita M. Angew. Chem. Int. Ed. 2016; 55: 1561
• 32 Joshi H, Sreejith S, Dey R, Stuparu MC. RSC Adv. 2016; 6: 110001
• 33 Fan QJ, Lin YJ, Hahn FE, Jin GX. Dalton Trans. 2018; 47: 2240
• 34 Mizyed S, Georghiou PE, Bancu M, Cuadra B, Rai AK, Cheng P, Scott LT. J. Am. Chem. Soc. 2001; 123: 12770
• 35 Georghiou PE, Tran AH, Mizyed S, Bancu M, Scott LT. J. Org. Chem. 2005; 70: 6158
• 36 Sygula A, Fronczek FR, Sygula R, Rabideau PW, Olmstead MM. J. Am. Chem. Soc. 2007; 129: 3842
• 37 Yanney M, Sygula A. Tetrahedron Lett. 2013; 54: 2604
• 38 Barbero H, Ferrero S, Álvarez-Miguel L, Gómez-Iglesias P, Miguel D, Álvarez CM. Chem. Commun. 2016; 52: 12964
• 39 Miyajima D, Tashiro K, Araoka F, Takezoe H, Kim J, Kato K, Takata M, Aida T. J. Am. Chem. Soc. 2009; 131: 44
• 40 Mattarella M, Haberl JM, Ruokolainen J, Landau EM, Mezzenga R, Siegel JS. Chem. Commun. 2013; 49: 7204
• 41 Kang J, Miyajima D, Mori T, Inoue Y, Itoh Y, Aida T. Science 2015; 347: 646
• 42 Mattarella M, Berstis L, Baldridge KK, Siegel JS. Bioconjugate Chem. 2014; 25: 115
• 43 Márquez IR, Castro-Fernández S, Millán A, Campaña AG. Chem. Commun. 2018; 54: 6705
• 44 Pun SH, Miao Q. Acc. Chem. Res. 2018; 51: 1630
• 45 Guan Y, Jones ML, Miller AE, Wheeler SE. Phys. Chem. Chem. Phys. 2017; 19: 18186
• 46 Gu X, Li H, Shan B, Liu Z, Miao Q. Org. Lett. 2017; 19: 2246
• 47 Jiménez VG, David AH. G, Cuerva JM, Blanco V, Campaña AG. Angew. Chem. Int. Ed. 2020; 59: 15124
• 48 Zhai L, Shukla R, Rathore R. Org. Lett. 2009; 11: 3474
• 49 Márquez IR, Fuentes N, Cruz CM, Puente-Muñoz V, Sotorrios L, Marcos ML, Choquesillo-Lazarte D, Biel B, Crovetto L, Gómez-Bengoa E, González MT, Martin R, Cuerva JM, Campaña AG. Chem. Sci. 2017; 8: 1068
• 50 Cheung KY, Xu X, Miao Q. J. Am. Chem. Soc. 2015; 137: 3910
• 51 Castro-Fernández S, Cruz CM, Mariz IF. A, Márquez IR, Jiménez VG, Palomino-Ruiz L, Cuerva JM, Maçôas E, Campaña AG. Angew. Chem. Int. Ed. 2020; 59: 7139
• 52 Chu M, Scioneaux AN, Hartley CS. J. Org. Chem. 2014; 79: 9009
• 53 Martin RB. Chem. Rev. 1996; 96: 3043
• 54 Chen Z, Lohr A, Saha-Möller CR, Würthner F. Chem. Soc. Rev. 2009; 38: 564
• 55 David AH. G, García-Cerezo P, Campaña AG, Santoyo-González F, Blanco V. Chem. Eur. J. 2019; 25: 6170
• 56 http://supramolecular.org/ (January 4, 2021)
• 57 Thordarson P. Chem. Soc. Rev. 2011; 40: 1305
• 58 The K _a values presented have not been corrected to take into account the influence of the self-association of the hept-HBC derivatives or the guests benzo[a]pyrene or 8, which also have some tendency to self-associate. These processes can be included in the binding in some cases assuming the simpler monomer–dimer model to study the self-association and considering that the self-assembled species (i.e., dimers) can also form heteroassemblies as they have two π surfaces available. However, a maximum variation of ca. 10% was observed in the values of the binding constants, so this effect can be omitted in a reasonable approximation
• 59 Coulson DR, Satek LC, Grim SO. Inorg. Synth. 1972; 13: 121
• 60 Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani G, Barone V, Mennucci B, Petersson GA, Nakatsuji H, Caricato M, Li X, Hratchian HP, Izmaylov AF, Bloino J, Zheng G, Sonnenberg JL, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T, Montgomery Jr JA, Peralta JE, Ogliaro F, Bearpark M, Heyd JJ, Brothers E, Kudin KN, Staroverov VN, Keith T, Kobayashi R, Normand J, Raghavachari K, Rendell A, Burant JC, Iyengar SS, Tomasi J, Cossi M, Rega N, Millam JM, Klene M, Knox JE, Cross JB, Bakken V, Adamo C, Jaramillo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C, Ochterski JW, Martin RL, Morokuma K, Zakrzewski VG, Voth GA, Salvador P, Dannenberg JJ, Dapprich S, Daniels AD, Farkas O, Foresman JB, Ortiz JV, Cioslowski J, Fox DJ. Gaussian 09, Revision B.01. Gaussian, Inc.; Wallingford, CT: 2010