Oxidative Cyclodehydrogenation Reactions with Tetraarylporphyrins

: The extension of the aromatic π -system of por-phyrins is a powerful method to alter their optoelectronic properties. Herein, aryl substituents were fused to porphyrin cores by Scholl oxidation reactions that selectively produced mono-and doubly-fused porphyrins in yields of up to 69 %. Several different aryl substituents attached to the porphyrin were investigated with respect to their reactivity under Scholl conditions. The fusion of aromatic units to porphyrin cores has become a flourishing research area that generated a plethora of π - extended porphyrins. [1–3] The extension of the chromophores' π -system represents an efficient tool to alter and therefore tailor the optical properties, e.g., the position of the absorption bands. For that, small polycyclic aromatic hydrocarbons (PAHs) such as benzene [4–11] or anthracene [12–15] but also larger ones such as nanographenes [16,17] or graphene itself [18] were fused once or several times to porphyrins. Typical synthesis procedures either use already π -extended pyrrole building blocks for porphyrin synthesis or fuse the desired aromatic units to pre-formed porphyrins afterwards. [3] Especially the second concept, the post-fusion of aromatics to porphyrins, established itself as the method of choice. Due to the versatility of this approach [19,20] that relies on numerous suitable coupling possibilities, a huge π -extended porphyrin variety has already been prepared. For example, oxidative cyclodehydrogenation reactions, commonly known as Scholl oxidations,

The fused products were fully characterized, i.e., by UV/Vis absorption spectroscopy, which showed drastic changes in the electronic features. Insight into the solid-state behavior was obtained by X-ray crystallography. Our approach represents a novel option for the late-stage functionalization of porphyrinbased compounds.
Recently however, we subjected nickel porphyrins to Scholl conditions that resulted, instead of forming a PAH in the porphyrins' periphery, in a reaction between the aryl substituents and the porphyrin core. [29] Because the properties of the obtained product significantly differed to the ones of common porphyrins, e.g., the UV/Vis absorption features, we were curious whether Scholl oxidations might be a suitable tool for modifying the porphyrins' characteristics. This prompted us to further investigate the reactivity of porphyrins under oxidative aromatic coupling conditions, which will be presented in the following.
The first examples of π-extended porphyrins, in which mesophenyl substituents were fused to porphyrin cores, were prepared by an approach based on palladium catalysis. [7][8][9][10][11] For that, suitable halogenated porphyrins were reacted to the respective π-extended products with up to four fused phenyl rings per porphyrin core. [10,11] Later on, Osuka and co-workers prepared similar π-extended porphyrin motifs, however, via a different route. Their approach showed that 3,5-di-tert-butylphenyl-substituted nickel porphyrins can be transformed to the respective π-extended derivatives under Scholl conditions. [31,32] For example, nickel porphyrins 1 bearing different functional groups at one meso-position were studied with respect to their reactivity behavior under Scholl conditions (Scheme 1). [32] In Scheme 1. Synthesis of doubly-fused porphyrins 2 that were reported by Osuka and co-workers in 2016. [32] R = NO 2 , POPh 2, Bpin. their study, electron withdrawing substituents R attached to the porphyrin core lead to a regioselective formation of 10,12-and 18,20-doubly phenylene-fused nickel porphyrins 2. [32] In our work, we approached the question about the reactivity of nickel porphyrins in Scholl reactions from a different perspective. Instead of studying the influence of one functional group, we were interested in the effect of the aryl substituents themselves. For that, our investigation started with a tetrakis-(3,5-di-tert-butylphenyl) substituted nickel porphyrin 3 that was subjected to oxidative aromatic coupling conditions (Scheme 2). For the Scholl reaction, nickel porphyrin 3 was treated with an excess amount of dry FeCl 3 (8 equiv.) in a CH 3 NO 2 /CH 2 Cl 2 mixture. Under these conditions, the C-C coupling reaction proceeded quickly and already after 1 h a new intensive olive spot was detected via TLC analysis (Scheme 2, left TLC). The reaction was stopped after 3 h and the crude product was purified by silica column chromatography. An olive green solid that was identified as the mono-fused porphyrin 4 was obtained in 69 % yield. During the end of the reaction, we realized that next to 4 another compound started to accumulate within the reaction mixture. We therefore tested different reaction conditions, in which more equivalents of FeCl 3 were added and allowed to react for a longer period of time. And indeed, after a reaction time of 23 h with 16 equiv. of FeCl 3 a new intensive red-brownish spot was detected next to the olive one of 4 via TLC analysis (Scheme 2, right TLC). After silica column chromatography, the red-brownish product was isolated (as well as 4) and identified as the doubly-fused porphyrin 5 (36 % yield). Although theoretically several different doublyfused porphyrins exist, we only detected and isolated a single isomer (5) under the given reaction conditions. These results are in contrast to the previous findings by the Osuka group, [32] in which electron-withdrawing groups were accounted for the formation of doubly-fused isomers like 2 or 5. So far, attempts to synthesize triply-or quadruply-fused porphyrins with the herein reported procedure were unsuccessful.
The fused porphyrin products 4 and 5 were unambiguously identified by NMR and UV/Vis spectroscopic as well as mass spectrometric and X-ray crystallographic techniques. In the NMR spectra, the fusion of an aryl substituent to the adjacentposition is clearly noticeable by an increased number of signals  ( Figure 1). For example, nickel porphyrin 3 shows 9 signals in the 13 C NMR spectrum, whereas the mono-fused nickel porphyrin 4, due to its reduced symmetry, has 48 separate carbon signals (Figure 1b). The fusion of a second bond to the doublyfused system 5, however, increases the symmetry again and therefore less signals with respect to 4 appear in the NMR spectrum ( Figure S10). Single crystals, suitable for X-ray crystallography were obtained for the starting material nickel porphyrin 3 as well as for the mono-and doubly-fused nickel porphyrins 4 and 5, respectively ( Figure 2). The crystals were grown out of CH 2 Cl 2 /MeOH (3,5) or CH 2 Cl 2 /n-hexane/MeOH (4) solution mixtures. Initial attempts to crystallize the fused-porphyrins, however, failed because 4 and 5 showed an increased solubility behavior in organic solvents compared to porphyrin 3. Nevertheless, slow evaporation of the solvent mixture that is accompanied by a gradually increasing concentration yielded crystals with a suitable size and quality for X-ray diffraction experiments. A comparison between the structures of 3, 4 and 5 revealed significant differences in the solid-state characteristics. The molecules of nickel porphyrin 3 are, due to the steric demand of the 3,5-di-tert-butylphenyl substituents, only loosely packed (Figure 2d). The fusion of aryl substituents to porphyrin cores does not only extend the porphyrins' π-system but also reduces its overall steric demand. Hence, more dense packing structures are feasible. Therefore, mono-and doubly-fused porphyrins 4 and 5 feature the formation of dimers in the crystal with distances between the porphyrin cores [33] of 4.22 Å and 3.30 Å for 4 and 5, respectively. The most pronounced interaction within the dimer was observed for the doubly-fused porphyrin 5, in which the two molecules almost completely overlap due to π-π interactions ( Figure 2h). Furthermore, a short Ni-Ni distance of 3.51 Å was found for the dimers of 5. Additionally,  The fusion of one or two aryl substituents to the porphyrin core significantly influences the electronic characteristics. The changes from 3 to the mono-and doubly-fused porphyrins 4 and 5 are recognizable by distinct color shifts from orange (3) to olive-green (4) to red-brownish (5). Additionally, fused nickel porphyrins 4 and 5 feature, compared to 3 as well as to other tetraarylporphyrins, a significantly more intensive color to the human eye. The changes of the electronic properties were studied by UV/Vis absorption spectroscopy (Figure 3). The fusion of one aryl ring to the porphyrin core (4) leads to a drastic decrease of the molar extinction coefficient accompanied by a broadening of the porphyrins' B-band with strong absorption features in the range of 350-500 nm. The fusion of a second aryl ring (5) further broadens the B-band, which is leading almost to an absorption plateau in the range of 350-550 nm. Because molecules 4 and 5 absorb light in such a large window of the visible region their color intensity appears so strong.
After fusing the 3,5-di-tert-butylphenyl substituents once (4) and twice (5) to the porphyrin core we started to test other aryl substituents under Scholl conditions. For that, 4-tert-butylphenyl and 4-bromophenyl substituents as well as mesityl as a control group were chosen and reacted (Table 1). For meaningful results, all reactions were performed under the exact same conditions and tightly followed by TLC analysis. The only variable was the reaction time. As a consequence, the chosen reaction conditions are a compromise that allow suitable reactivity for all porphyrins but do not represent optimized conditions for the individual compounds. The isolated products were identified by NMR spectroscopy and mass spectrometry. Reactivity for all porphyrins under the given conditions was observed, however, at significantly different reaction rates (compare Table 1). The fastest reaction was observed for the 3,5-di-tertbutylphenyl-substituted porphyrin 3, followed by the 4-tertbutylphenyl-substituted porphyrin 6. The 4-bromophenyl-substituted porphyrin 7 showed, similar to 3 and 6, the formation of a fused product that was observed by TLC analysis (olive green spot). The reaction of 7, however, seemed to proceed even slower and less smoothly than the ones of 3 and 6. As a result, significant amounts of by-product were formed during the reaction of 7, which could not be separated from the desired fused product. Hence, no clear characterization of the product and determination of the yield could be performed. Finally, the mesityl control group porphyrin 8, which has no available positions for a fusion reaction, was tested. As expected, no fusion of the aryl substituent occurred but instead, a partial chlorination of the porphyrins was observed by mass spectrometry ( Figure S22). To conclude, the successful preparation of π-extended porphyrins by a fusion reaction of the aryl substituents to the porphyrin cores was presented. Standard FeCl 3 mediated Scholl conditions have proven to be suitable for the fusion of 3,5-ditert-butylphenyl as well as 3-tert-butylphenyl substituents to the respective porphyrin core. Chlorination, which is a typical side reaction, only occurred for the control group, the mesitylsubstituted porphyrin 8 that cannot undergo a fusion reaction.
Finally, we like to emphasize the advantage of the herein presented concept: With our protocol, simple A 4 -symmetric porphyrins, which can be easily prepared in decent yields and quantities, can be transformed to newly π-extended dyes using well-established reaction conditions. Due to this simplicity and the reliance on standard reactions, we think that this concept has a great potential for the preparation and application of exciting new dye materials. Additionally, this protocol is suitable for the late stage modification of many already existing porphyrin arrays. For example, the absorption characteristics of porphyrin-based light absorbing arrays might be improved and therefore higher efficiencies of photovoltaic devices could be achieved. [34][35][36] The only limiting factor is that the substituents and functional groups attached to the porphyrin must be tolerant to Scholl conditions. To fully reveal the potential of this concept, fusing aryl substituents to porphyrins by Scholl reactions are currently under further investigation in our group.
Deposition Numbers 2025048, 2025049, 2025050 and 2025051 contain the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service www.ccdc.cam.ac.uk/structures.