All are aromatic: A 3D globally aromatic cage containing five types of 2D aromatic macrocycles

Abstract

•A fully π-conjugated molecular cage was synthesized in a simple way•Chemical oxidation led to a 3D globally aromatic cage•All five types of 2D macrocycles along the 3D skeleton are Hückel aromatic•A close relation between 2D π-aromaticity and 3D π-aromaticity was established Aromaticity is a vital concept in organic chemistry and π-/σ-conjugated molecules have intrinsic tendency to become aromatic via electron delocalization. Although there have been many studies on the aromaticity in 2D π-conjugated macrocycles, the examples of 3D globally aromatic systems are very limited. Research on 3D π-aromaticity will not only help us to better understand the fundamental chemical principles but also provide a new type of 3D π-conjugated molecules for electronics, spintronics, and quantum information. In this work, a 4-fold symmetric π-conjugated molecular cage 1 was facilely synthesized and two-electron oxidation led to a 3D globally aromatic cage 12+. Our detailed experimental measurements and theoretical analysis reveal that all five types of formally available 2D macrocycles along the 3D π-conjugated skeleton in 12+ are Hückel aromatic. This work discloses the close correlation between 3D global aromaticity and 2D Hückel aromaticity. The studies on three-dimensional (3D) aromaticity have been mainly focused on fullerenes, boron-based deltahedrons/clusters, metal clusters, and polyhedral hydrocarbons, but there is very limited research on the fundamental aromaticity rule for 3D fully π-conjugated molecules. Herein, we report a π-conjugated molecular cage in which two aromatic porphyrin units are bridged by four thiophene-based arms. Two-electron chemical oxidation leads to a 3D globally aromatic cage with a C2 symmetry according to X-ray diffraction, NMR, electronic absorption spectra, and theoretical calculations. Detailed magnetic shielding response analysis along different axes reveals that all the possible five types of two-dimensional (2D) macrocycles in the cage skeleton are aromatic and follow the Hückel rule. The switch from localized aromaticity to global aromaticity upon chemical oxidation is also observed in a tricyclic model compound. This study indicates that 3D global aromaticity in a molecular cage can be explained by 2D Hückel aromaticity of the individual π-conjugated macrocycles. The studies on three-dimensional (3D) aromaticity have been mainly focused on fullerenes, boron-based deltahedrons/clusters, metal clusters, and polyhedral hydrocarbons, but there is very limited research on the fundamental aromaticity rule for 3D fully π-conjugated molecules. Herein, we report a π-conjugated molecular cage in which two aromatic porphyrin units are bridged by four thiophene-based arms. Two-electron chemical oxidation leads to a 3D globally aromatic cage with a C2 symmetry according to X-ray diffraction, NMR, electronic absorption spectra, and theoretical calculations. Detailed magnetic shielding response analysis along different axes reveals that all the possible five types of two-dimensional (2D) macrocycles in the cage skeleton are aromatic and follow the Hückel rule. The switch from localized aromaticity to global aromaticity upon chemical oxidation is also observed in a tricyclic model compound. This study indicates that 3D global aromaticity in a molecular cage can be explained by 2D Hückel aromaticity of the individual π-conjugated macrocycles. Aromaticity is initially confined to planar two-dimensional (2D) π-conjugated monocyclic molecules, which follow [4N+2] Hückle rule.1Hückel E. Quantentheoretische Beiträge zum benzolproblem.Z. Physik. 1931; 70: 204-286Google Scholar, 2Breslow R. Antiaromaticity. Acc. Chem. Res. 1973; 6: 393-398Google Scholar, 3Krygowski T.M. Cyrañski M.K. Czarnocki Z. Häfelinger G. Katritzky A.R. Aromaticity: a theoretical concept of immense practical importance.Tetrahedron. 2000; 56: 1783-1796Google Scholar, 4Sondheimer F. Annulenes. Acc. Chem. Res. 1972; 5: 81-91Google Scholar, 5Cyrański M.K. Energetic aspects of cyclic pi-electron delocalization: evaluation of the methods of estimating aromatic stabilization energies.Chem. Rev. 2005; 105: 3773-3811Google Scholar Over the years, studies have demonstrated that this concept is also applicable for diverse structures in which π- and/or σ-electrons can be efficiently delocalized and result in stabilization. For example, 2D π-conjugated polycyclic molecules could be globally aromatic if there is a dominant [n]annulene-like (n = 4N+2) ring current circuit, as observed in porphyrinoids,6Stępień M. Latos-Grażyński L. Aromaticity and tautomerism in porphyrins and porphyrinoids.in: Krygowski T.M. Cyrański M.K. Aromaticity in Heterocyclic Compounds. Springer, 2008: 82-153Google Scholar, 7Saito S. Osuka A. Expanded porphyrins: intriguing structures, electronic properties, and reactivities.Angew. Chem. Int. Ed. Engl. 2011; 50: 4342-4373Google Scholar, 8Tanaka T. Osuka A. Chemistry of meso-aryl-substituted expanded porphyrins: aromaticity and molecular twist.Chem. Rev. 2017; 117: 2584-2640Google Scholar porphyrin-based nanorings,9Peeks M.D. Claridge T.D.W. Anderson H.L. Aromatic and antiaromatic ring currents in a molecular nanoring.Nature. 2017; 541: 200-203Google Scholar,10Rickhaus M. Jirasek M. Tejerina L. Gotfredsen H. Peeks M.D. Haver R. Jiang H.W. Claridge T.D.W. Anderson H.L. Global aromaticity at the nanoscale.Nat. Chem. 2020; 12: 236-241Google Scholar and non-benzenoid polycyclic hydrocarbons.11Liu C. Sandoval-Salinas M.E. Hong Y. Gopalakrishna T.Y. Phan H. Aratani N. Herng T.S. Ding J. Yamada H. Kim D. et al.Macrocyclic polyradicaloids with unusual super-ring structure and global aromaticity.Chem. 2018; 4: 1586-1595Google Scholar,12Liu C. Ni Y. Lu X. Li G. Wu J. Global aromaticity in macrocyclic polyradicaloids: Hückel’s rule or Baird’s rule?.Acc. Chem. Res. 2019; 52: 2309-2321Google Scholar On the other hand, Möbius twists13Heilbronner E. Hückel molecular orbitals of Möbius-type conformations of annulenes.Tetrahedron Lett. 1964; 5: 1923-1928Google Scholar, 14Ajami D. Oeckler O. Simon A. Herges R. Synthesis of a Möbius aromatic hydrocarbon.Nature. 2003; 426: 819-821Google Scholar, 15Stępień M. Latos-Grażyński L. Sprutta N. Chwalisz P. Szterenberg L. Expanded porphyrin with a split personality: a Hückel–Möbius aromaticity switch.Angew. Chem. Int. Ed. Engl. 2007; 46: 7869-7873Google Scholar, 16Tanaka Y. Saito S. Mori S. Aratani N. Shinokubo H. Shibata N. Higuchi Y. Yoon Z.S. Kim K.S. Noh S.B. et al.Metalation of expanded porphyrins: a chemical trigger used to produce molecular twisting and Möbius aromaticity.Angew. Chem. Int. Ed. Engl. 2008; 47: 681-684Google Scholar, 17Yoon Z.S. Osuka A. Kim D. Möbius aromaticity and antiaromaticity in expanded porphyrins.Nat. Chem. 2009; 1: 113-122Google Scholar, 18Stępień M. Sprutta N. Latos-Grażyński L. Figure eights, Möbius bands, and more: conformation and aromaticity of porphyrinoids.Angew. Chem. Int. Ed. Engl. 2011; 50: 4288-4340Google Scholar, 19Craig D.P. Paddock N.L. A Novel type of aromaticity.Nature. 1958; 181: 1052-1053Google Scholar, 20Zhu C.Q. Xia H.P. Carbolong chemistry: a story of carbon chain ligands and transition metals.Acc. Chem. Res. 2018; 51: 1691-1700Google Scholar and triplet [n]annulenes21Baird N.C. Quantum organic photochemistry. II. Resonance and aromaticity in the lowest 3ππ∗ state of cyclic hydrocarbons.J. Am. Chem. Soc. 2002; 94: 4941-4948Google Scholar, 22Wörner H.J. Merkt F. Photoelectron spectroscopic study of the first singlet and triplet states of the cyclopentadienyl cation.Angew. Chem. Int. Ed. Engl. 2005; 45: 293-296Google Scholar, 23Ottosson H. Organic photochemistry: exciting excited-state aromaticity.Nat. Chem. 2012; 4: 969-971Google Scholar, 24Rosenberg M. Dahlstrand C. Kilså K. Ottosson H. Excited state aromaticity and antiaromaticity: opportunities for photophysical and photochemical rationalizations.Chem. Rev. 2014; 114: 5379-5425Google Scholar, 25Sung Y.M. Yoon M.C. Lim J.M. Rath H. Naoda K. Osuka A. Kim D. Reversal of Hückel (anti)aromaticity in the lowest triplet states of hexaphyrins and spectroscopic evidence for Baird's rule.Nat. Chem. 2015; 7: 418-422Google Scholar follow a [4N] aromaticity rule. The recognition of three-dimensional (3D) aromaticity can be dated back 60 years ago when Lipscomb et al. considered [B12H12]2− as a superaromatic molecule.26Lipscomb W.N. Pitochelli A.R. Hawthorne M.F. Probable structure of the B10H102- ion.J. Am. Chem. Soc. 1959; 81: 5833-5834Google Scholar Later studies showed that closo boron hydrides [BnHn]2− (6≤n≤12)27Schleyer PvR. Subramanian G. Dransfeld A. Decisive evidence for nonclassical bonding in five-vertex closo -boranes, X 2 B 3 H 3, X = N, CH, P, SiH, BH -.J. Am. Chem. Soc. 1996; 118: 9988-9989Google Scholar,28King R.B. Three-dimensional aromaticity in polyhedral boranes and related molecules.Chem. Rev. 2001; 101: 1119-1152Google Scholar and some boron-based clusters29Huang W. Sergeeva A.P. Zhai H.J. Averkiev B.B. Wang L.S. Boldyrev A.I. A concentric planar doubly π-aromatic B192- cluster.Nat. Chem. 2010; 2: 202-206Google Scholar, 30Sergeeva A.P. Popov I.A. Piazza Z.A. Li W.L. Romanescu C. Wang L.S. Boldyrev A.I. Understanding boron through size-selected clusters: structure, chemical bonding, and fluxionality.Acc. Chem. Res. 2014; 47: 1349-1358Google Scholar, 31Zhai H.-J. Zhao Y.-F. Li W.-L. Chen Q. Bai H. Hu H.-S. Piazza Z.-A. Tian W.-J. Lu H.-G. Wu Y.-B. et al.Observation of an all-boron fullerene.Nat. Chem. 2014; 6: 727-731Google Scholar are also aromatic. A correlation between classic 2D polycyclic aromatic hydrocarbons and these deltahedral boron hydrides was established by Poater et al. by showing that both share a common origin regulated by the valence electrons in confined space.32Poater J. Solà M. Viñas C. Teixidor F. π-aromaticity and three-dimensional aromaticity: two sides of the same coin?.Angew. Chem. Int. Ed. Engl. 2014; 53: 12191-12195Google Scholar In 2002, Hirsch et al. presented a 2(N+1)2 spherical aromaticity rule for Ih symmetric fullerenes after considering the similarities between their molecular orbitals and radial-symmetric atomic orbitals.33Hirsch A. Chen Z. Jiao H. Spherical aromaticity in Ih symmetrical fullerenes: the 2(N+1)2 rule.Angew. Chem. Int. Ed. Engl. 2000; 39: 3915-3917Google Scholar, 34Bühl M. Hirsch A. Spherical aromaticity of fullerenes.Chem. Rev. 2001; 101: 1153-1183Google Scholar, 35Reiher M. Hirsch A. From rare gas atoms to fullerenes: spherical aromaticity studied from the point of view of atomic structure theory.Chemistry. 2003; 9: 5442-5452Google Scholar This rule was also found to be applicable to some highly symmetric polyhedral hydrocarbons36Chen Z. King R.B. Spherical aromaticity: recent work on fullerenes, polyhedral boranes, and related structures.Chem. Rev. 2005; 105: 3613-3642Google Scholar,37Bremer M. von Ragué Schleyer P. Schötz K. Kausch M. Schindler M. Four-Center Two-Electron Bonding in a Tetrahedral Topology. Experimental Realization of Three-Dimensional Homoaromaticity in the 1,3-Dehydro-5,7-adamantanediyl Dication.Angew. Chem. Int. Ed. Engl. 1987; 26: 761-763Google Scholar and metal clusters.38Chen Z. Jiao H. Hirsch A. Schleyer P.v.R. Spherical homoaromaticity.Angew. Chem. Int. Ed. Engl. 2002; 41: 4309-4312Google Scholar, 39Johansson M.P. Sundholm D. Vaara J. Au32: A 24-carat golden fullerene.Angew. Chem. Int. Ed. Engl. 2004; 43: 2678-2681Google Scholar, 40Wang J. Jellinek J. Zhao J. Chen Z. King R.B. Schleyer P.v. Hollow cages versus space-filling structures for medium-sized gold clusters: the spherical aromaticity of the Au50 Cage.J. Phys. Chem. A. 2005; 109: 9265-9269Google Scholar It was further theoretically extended to open-shell systems by Solà et al.41Poater J. Solà M. Open-shell spherical aromaticity: the 2N2 + 2N + 1 (with S = N + ½) rule.Chem. Commun. (Camb). 2011; 47: 11647-11649Google Scholar In addition, close stacking between antiaromatic molecules could lead to enhanced aromaticity.42Corminboeuf C. Schleyer Pv Warner P. Are antiaromatic rings stacked face-to-face aromatic?.Org. Lett. 2007; 9: 3263-3266Google Scholar, 43Nozawa R. Tanaka H. Cha W.Y. Hong Y. Hisaki I. Shimizu S. Shin J.Y. Kowalczyk T. Irle S. Kim D. Shinokubo H. Stacked antiaromatic porphyrins.Nat. Commun. 2016; 7: 13620Google Scholar, 44Nozawa R. Kim J. Oh J. Lamping A. Wang Y. Shimizu S. Hisaki I. Kowalczyk T. Fliegl H. Kim D. Shinokubo H. Three-dimensional aromaticity in an antiaromatic cyclophane.Nat. Commun. 2019; 10: 3576Google Scholar, 45Kawashima H. Ukai S. Nozawa R. Fukui N. Fitzsimmons G. Kowalczyk T. Fliegl H. Shinokubo H. Determinant factors of three-dimensional aromaticity in antiaromatic cyclophanes.J. Am. Chem. Soc. 2021; 143: 10676-10685Google Scholar All these studies demonstrate that π-/σ-conjugated molecules or clusters always have the tendency to reach the lowest energy state (aromatic) through electron delocalization. Surprisingly, the investigated 3D aromatic systems cover very little about the fully π-conjugated molecular cages or polyhedrons, although 2D π-aromaticity was widely studied. This is probably because the synthesis of 3D π-conjugated systems with all π-electrons being fully delocalized along the 3D skeleton is a very challenging task. There were indeed a number of 3D π-conjugated molecular cages,46Högberg H.-E. Thulin B. Wennerström O. Bicyclophanehexaene, a new case cyclophane from a sixfold wittig reaction.Tetrahedron Lett. 1977; 18: 931-934Google Scholar, 47Wu Z. Lee S. Moore J.S. Synthesis of three-dimensional nanoscaffolding.J. Am. Chem. Soc. 1992; 114: 8730-8732Google Scholar, 48Kayahara E. Iwamoto T. Takaya H. Suzuki T. Fujitsuka M. Majima T. Yasuda N. Matsuyama N. Seki S. Yamago S. Synthesis and physical properties of a ball-like three-dimensional π-conjugated molecule.Nat. Commun. 2013; 4: 2694Google Scholar, 49Matsui K. Segawa Y. Itami K. All-benzene carbon nanocages: size-selective synthesis, photophysical properties, and crystal structure.J. Am. Chem. Soc. 2014; 136: 16452-16458Google Scholar, 50Song J. Aratani N. Shinokubo H. Osuka A. A porphyrin nanobarrel that encapsulates C60.J. Am. Chem. Soc. 2010; 132: 16356-16357Google Scholar, 51Zhang C. Wang Q. Long H. Zhang W. A highly C70 selective shape-persistent rectangular prism constructed through one-step alkyne metathesis.J. Am. Chem. Soc. 2011; 133: 20995-21001Google Scholar, 52Ke X.S. Kim T. He Q. Lynch V.M. Kim D. Sessler J.L. Three-dimensional fully conjugated carbaporphyrin cage.J. Am. Chem. Soc. 2018; 140: 16455-16459Google Scholar, 53Cha W.Y. Kim T. Ghosh A. Zhang Z. Ke X.S. Ali R. Lynch V.M. Jung J. Kim W. Lee S. et al.Bicyclic Baird-type aromaticity.Nat. Chem. 2017; 9: 1243-1248Google Scholar but the π-electrons are mainly localized in individual benzenoid rings, or only delocalized along one or two macrocycles, not fully delocalized at three dimensions. Our group recently demonstrated that a 3-fold symmetric diradicaloid molecular cage showed 3D global aromaticity at the hexacationic state (c-T126+), and a [6N+2] electron counting rule was proposed for this specific type of D3 symmetric molecular cage.54Ni Y. Gopalakrishna T.Y. Phan H. Kim T. Herng T.S. Han Y. Tao T. Ding J. Kim D. Wu J. 3D global aromaticity in a fully conjugated diradicaloid cage at different oxidation states.Nat. Chem. 2020; 12: 242-248Google Scholar The study also implied that this pseudo-spherical symmetric cage containing 50 delocalized π-electrons could also satisfy Hirsch’s 2(N+1)2 rule (N = 4) in analog to fullerene C6010+. However, the much lower symmetry (D3 symmetry) in comparison with the ideal Ih symmetry for C6010+ deferred a clear conclusion. Therefore, whether the 3D fully delocalized molecular cage follows the 2D Hückel rule for each individual macrocycle or a 3D Hirsch’s 2(N+1)2 rule remains a question. To deeper understand the 3D global π-aromaticity, it is necessary to synthesize more complex molecular cages with higher symmetry. In this context, we designed a 4-fold symmetric molecular cage 1 in which two aromatic porphyrin units are linked by four thiophene-based linkers (Scheme 1). The porphyrin and thiophene rings are chosen because they allow 3D electron delocalization as compared with the aromatic benzene ring-based molecules.6Stępień M. Latos-Grażyński L. Aromaticity and tautomerism in porphyrins and porphyrinoids.in: Krygowski T.M. Cyrański M.K. Aromaticity in Heterocyclic Compounds. Springer, 2008: 82-153Google Scholar, 7Saito S. Osuka A. Expanded porphyrins: intriguing structures, electronic properties, and reactivities.Angew. Chem. Int. Ed. Engl. 2011; 50: 4342-4373Google Scholar, 8Tanaka T. Osuka A. Chemistry of meso-aryl-substituted expanded porphyrins: aromaticity and molecular twist.Chem. Rev. 2017; 117: 2584-2640Google Scholar, 9Peeks M.D. Claridge T.D.W. Anderson H.L. Aromatic and antiaromatic ring currents in a molecular nanoring.Nature. 2017; 541: 200-203Google Scholar, 10Rickhaus M. Jirasek M. Tejerina L. Gotfredsen H. Peeks M.D. Haver R. Jiang H.W. Claridge T.D.W. Anderson H.L. Global aromaticity at the nanoscale.Nat. Chem. 2020; 12: 236-241Google Scholar,54Ni Y. Gopalakrishna T.Y. Phan H. Kim T. Herng T.S. Han Y. Tao T. Ding J. Kim D. Wu J. 3D global aromaticity in a fully conjugated diradicaloid cage at different oxidation states.Nat. Chem. 2020; 12: 242-248Google Scholar Although the neutral cage compound may still possess 2D localized aromaticity of the porphyrin unit, it could be converted into a 3D globally aromatic cage upon chemical oxidation. Indeed, our studies reveal that its dication (12+) shows desired 3D global aromaticity. More importantly, we found that all the individual 2D macrocycles formed by the four arms and the fragments of the porphyrin unit, including the two porphyrin macrocycles, are aromatic and follow Hückel’s [4N+2] aromaticity rule. This indicates a close correlation between 3D global aromaticity and 2D Hückel aromaticity. In this article, we report the detailed synthesis, physical characterization, and analysis on the 3D aromaticity. The molecular cage was synthesized by Ni(COD)2-mediated intermolecular Yamamoto coupling of the intermediate compound 6, followed by oxidative dehydrogenation of the isolated octahydro-cage 7 by 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) in dichloromethane (DCM) (Scheme 1). The key intermediate 6 was prepared by a reaction sequence involving formylation, Alder-Longo condensation, and nickel (II) ion insertion starting from 3. During the Yamamoto coupling reaction, an intramolecularly coupled tricyclic product 8 was also obtained, and subsequent oxidative dehydrogenation gave the fully conjugated macrocycle 2, which can serve as a model compound to understand the switch from localized aromaticity to global aromaticity upon chemical oxidation. In both molecules, bulky 4-tert-butyl-2,6-dimethylphenyl groups are attached onto the most reactive methylene sites. Thus, the final products are soluble and stable and can be purified by routine silica gel column chromatography. The cyclic voltammogram and differential pulse voltammogram of 1 show three reversible oxidation waves with half-wave potential (E1/2ox) at 0.01, 0.41, 1.02 V, and three reversible reduction waves with half-wave potential (E1/2red) at −1.40, −1.67, and −1.86 V (versus Fc+/Fc; Fc: ferrocene) (Figure 1A). Compound 2 also exhibits amphoteric redox behavior, with four reversible oxidation waves (E1/2ox at −0.09, 0.01, 0.28, and 0.45 V) and one reversible reduction wave (E1/2red at −1.44 V) (Figure 1B). Two-electron chemical oxidation of both 1 and 2 by oxidant NO·SbF6 gave the corresponding dications 12+ and 22+ (see evolution of electronic absorption spectra during the chemical titration: Figures S1 and S2), which could be isolated in single-crystal form. No higher-oxidation-state species could be obtained even when excessive oxidant was added. Compound 1 in DCM shows an intense Soret band with a maximum (λabs) at 425 nm, which is typical for porphyrin, and another intense band with λabs at 545 nm, which is reminiscent of that of quinoidal bithiophene moieties (Figure 1C), indicating that the porphyrin units and the four arms are not well conjugated. On the other hand, its dication displays a broad absorption band at near-infrared (NIR) region extending beyond 1,600 nm, indicating effective π-electron delocalization. The existence of an intense band with λabs at 683 nm also implies a possible global aromatic character. Compound 2 shows a similar absorption band to 1, with the lowest-energy band split, presumably due to its more rigid structure (Figure 1D). However, its dication exhibits an intense band even at the NIR region (λabs = 1,068 nm). The X-ray crystallographic analysis of the single crystal of 1 at 100 K reveals a slant cage structure with a center of symmetry (Figure 2A). The two porphyrin rings remain nearly planar and aligned parallel to each other, with a plane-to-plane distance of about 7.8 Å. The sulfur atoms of the middle bithiophene units point inside the cavity, with the distance between the opposite sulfur atoms being 16.95/17.43 Å, which could represent the inner diameter of 1. The distortional angles between the porphyrins and the neighboring thiophene units are about 58.8°, 62.3°, 77.5°, and 89.2°, implying a weak π-conjugation between them. This is further evident from its 1H NMR spectrum in CDCl3 at 298 K (Figure 3A). The resonance for the β-H (proton e) of the porphyrin unit appears at the chemical shift δ = 8.83 ppm, which is typical for aromatic porphyrin. The resonances for the bithiophene units (protons a and b) appear as doublets at δ = 7.00 and 6.33 ppm, respectively, indicating a dominant quinoidal structure. The resonances for the protons c and d on the thiophene rings that are directly linked to the porphyrin unit are broadened and split, with δ at about 6.54 and 7.02 ppm, which can be explained by the slow dynamic rotation of the thiophene ring as previously observed in another analogous 2D system.55Ren L. Gopalakrishna T.Y. Park I.H. Han Y. Wu J. Porphyrin/quinoidal-bithiophene-based macrocycles and their dications: template-free synthesis and global aromaticity.Angew. Chem. Int. Ed. Engl. 2020; 59: 2230-2234Google Scholar Accordingly, the peak for the proton e is also broadened and split, but heating the solution of 1 in CDCl2CDCl2 from 298 to 393 K (Figures S3 and S4) did not result in sharpening of the resonances, indicating a substantially high rotation energy barrier.Figure 31H NMR spectra (500 MHz, aromatic region) of 1, 12+, 2, and 22+Show full caption(A) 1 in CDCl3 at 298 K.(B) 12+ in CD2Cl2 at 238 K.(C) 2 in CDCl3 at 298 K.(D) 22+ in CD2Cl2 at 298 K. The top row are the chemical structures with labels (∗∗ is “0” for neutral compounds and “2+” for the dications). The peaks labeled with ∗ arise from the solvent CHCl3 and its satellite peaks. Peaks shaded by purple color represent the protons in the porphyrin ring, whereas peaks shaded by yellow color represent the aromatic protons in the substituents.View Large Image Figure ViewerDownload Hi-res image Download (PPT) (A) 1 in CDCl3 at 298 K. (B) 12+ in CD2Cl2 at 238 K. (C) 2 in CDCl3 at 298 K. (D) 22+ in CD2Cl2 at 298 K. The top row are the chemical structures with labels (∗∗ is “0” for neutral compounds and “2+” for the dications). The peaks labeled with ∗ arise from the solvent CHCl3 and its satellite peaks. Peaks shaded by purple color represent the protons in the porphyrin ring, whereas peaks shaded by yellow color represent the aromatic protons in the substituents. The aromaticity of 1 was further investigated by the anisotropy of the induced current density (ACID),56Geuenich D. Hess K. Köhler F. Herges R. Anisotropy of the induced current density (ACID), a general method to quantify and visualize electronic delocalization.Chem. Rev. 2005; 105: 3758-3772Google Scholar nucleus-independent chemical shift (NICS),57Schleyer P.V.R. Maerker C. Dransfeld A. Jiao H. van Eikema Hommes N.J.R. Nucleus- independent chemical shifts: a simple and efficient aromaticity probe.J. Am. Chem. Soc. 1996; 118: 6317-6318Google Scholar and 3D isochemical shielding surface (ICSS)58Humphrey W. Dalke A. Schulten K. VMD: visual molecular dynamics.J. Mol. Graph. 1996; 14: 27Google Scholar,59Lu T. Chen F. Multiwfn: a multifunctional wavefunction analyzer.J. Comput. Chem. 2012; 33: 580-592Google Scholar calculations. The ACID plot reveals a clockwise diatropic ring current circuit in the porphyrin unit when the magnetic field is perpendicular to the porphyrin plane (along the y axis) (Figure 4A), whereas the ACID plots with the magnetic field directed along the x and z axes do not show obvious ring current flow through the arms (Figures S12 and S13). The NICSiso at the geometric center is calculated to be −4.0 ppm. The 3D ICSS map shows a major magnetic shielding environment above and below the porphyrin units (Figure 4B and S46). All these calculations suggest that 1 only contains two localized aromatic porphyrin macrocycles and is not 3D globally aromatic, which is consistent with the experimental data. The X-ray crystallographic structure of the dication 12+ measured at 100 K shows a dramatic change as compared with the neutral compound (Figure 2B). The molecule has a C2 point of symmetry, with a C2 rotational axis across the two Ni(II) centers. The four thiophene-based arms are labeled as α, β, α′, and β′ (α is equivalent to α′ as β is equivalent to β′) to facilitate discussion. Both porphyrin units now adopt a saddle-shaped conformation, with the four meso-carbons bent in an “up-down-up-down” mode relative to the mean plane. The two downward bent meso-carbons in one porphyrin unit are linked with the two downward bent meso-carbons in the other porphyrin unit through the arms α/α′, whereas the other two upward bent meso-carbons connect with the two upward bent meso-carbons in the other porphyrin through the arms β/β′. Overall, the molecule shows a highly distorted structure with two arms (α/α′) bent down and the other two (β/β′) bent up. Such conformation somewhat releases the strain and facilitates π-electron delocalization at three dimensions. Although the quality of the crystallographic data does not allow reliable bond length analysis, it is clear that the distortional angles between the porphyrins and the neighboring thiophene units (67.2°, 44.9°, 72.2°, and 44.2°) become smaller as compared with 1, indicating more efficient π-orbital overlap. The two porphyrin units are more separated, with a mean distance of about 10.3 Å. At the meanwhile, the lateral diameter of the cavity (16.36 and 16.40 Å) defined by the same method to 1 becomes smaller. Unlike the neutral compound 1, the 1H NMR spectrum of 12+ in CD2Cl2 at 238 K displays two sets of well-resolved signals in accordance with two different halves separated by the middle points of the bithiophene units (Figure 3B; see Figures S6 and S7 for the assignment by 2D COSY (correlated apectroscopy) and ROESY (rotating frame Overhause effect spectroscopy) spectra). The resonances for all the protons on the π-conjugated backbone now appear in a low field (δ = 7.37 ∼ 9.03 ppm) with ten doublet peaks, and in particular, the protons a/a', b/b', c/c′, and d/d' on the thiophene units are all considerably shifted downfield as compared with 1, indicating a global aromatic character. The different chemical shifts for the equivalent protons could be correlated to their different shielding/de-shielding chemical environments, assuming that the whole cage is rigid and 3D aromatic in solution. The effective π-electron delocalization was also clearly shown in the calculated frontier molecular orbital profiles (Figure S51). Calculations provide more insight into the aromaticity of 12+. Interestingly, ACID plots along five different axes (z, x, y, xz, and −xz) all show a dominant clockwise ring current circuit cross the arms and the porphyrin units (Figure 4A). That means, this cage is indeed 3D globally aromatic. The NICS(0) value in the geometric center is calculated to −9.52 ppm, and the NICS scan in the cavity along the x, y, and z axes reveals only negative NICS values (Table S1), indicating an aromatic character. In addition, the 3D ICSS map also demonstrates a more magnetically shielded cavity compared with that in 1 (Figures 4B and S47), further supporting its 3D global aromaticity. To understand this unique 3D aromatic system, the porphyrin units are dissected into four fragments labeled as a/b/c/d for the top one and a'/b'/c'/d' for the bottom one (Figure 5), corresponding to their X-ray structure with a C2 symmetry (Figure 2B, top). Then, five types of π-conjugated macrocycles can be defined: (1) two equivalent macrocycles across α/β′ or α'/β arms (i.e., a-α-a′-β' and c-α′-c′-β); (2) two equivalent macrocycles across α/β or α'/β' arms (i.e., b-β-b′-α and d-β′-d′-α'); (3) two equivalent porphyrin macrocycles (i.e., a-b-c-d and a'-b′-c′-d'); (4) four equivalent macrocycles across α/α' arms (i.e., b-c-α′-c′-b′-α, b-c-α′-d′-a′-α, a-d-α′-d′-a′-α, and a-d-α′-c′-b′-α); and (5) four equivalent macrocycles across β/β' arms (i.e., a-b-β-b'-a'-β′, a-b-β-c'-d'-β′, c-d-β′-d′-c′-β, and c-d-β′-a′-b′-β). Notably, all these five types of macrocycles are aromatic according to the ACID calculations, and one can also draw an [n]annulene-like conjugation pathway with 46, 46, 18, 54, and 54 delocalized π-electrons, respectively (Figures S38 and S39). These 2D canonical forms resonate between each other and form a highly stable 3D aromatic system containing five types of 2D Hückel aromatic macrocycles. The positive charges are also delocalized at three dimensions (Figure S50). The calculated harmonic oscillator measure of aromaticity (HOMA) values of the five types of 2D macrocycles based on the optimized geometries show obvious increase from 1 (0.52 ∼ 0.58) to 12+ (0.64 ∼ 0.68) for the type I, II, IV, and V π-conjugation pathways (Figures S41 and S42), indicating enhanced aromaticity after two-electron oxidation. The HOMA values of the porphyrin unit remain nearly the same (0.77) due to aromatic character in both neutral and dicationic (type III) forms. The π-conjugated skeleton of 12+ has a total of 114 delocalized π electrons (excluding the sulfur atoms, which contribute little to the ring current according to ACID plots) and a much lower symmetry as compared with the Ih symmetric fullerenes; thus, it is clear that it does not follow Hirsch’s 2(N+1)2 spherical aromaticity rule. In our previous studies on the D3 symmetric, 3D globally aromatic cage c-T126+,17Yoon Z.S. Osuka A. Kim D. Möbius aromaticity and antiaromaticity in expanded porphyrins.Nat. Chem. 2009; 1: 113-122Google Scholar all the three macrocycles across the two bridge head carbons and any two of the three arms are actually also 2D Hückel aromatic. Therefore, it seems that to attain 3D global aromaticity in this type of molecular cages, all the formally available 2D π-conjugated macrocycles should be aromatic. ACID calculations of the analogs of 12+ in which one or two arms are replaced by a non-conjugated doubly hydrogenated linker were also conducted, which reveal that the remaining π-conjugated parts remain globally aromatic (Figures S22–S35). This finding further supports that all individual 2D π-conjugated macrocycles in 12+ are aromatic and contribute to the 3D global aromaticity. The switch from porphyrin-based localized aromaticity to global aromaticity along the whole tricyclic system upon chemical oxidation was also observed in the model compound 2. X-ray crystallographic structure of 2 shows a slightly distorted saddle-shaped porphyrin core, and the two fused earring-like units are bent toward the same side relative to the porphyrin mean plane, with a cavity depth of about 2.0 Å (Figure 2C). The distortional angles between the porphyrins and the neighbored thiophene units are 53.4°, 58.8°, 66.8°, and 67.2°, respectively. Upon oxidation to the dication, the two arms are further bent with a deeper cavity (depth: ∼3.3 Å), but the distortional angles between the porphyrin and the arms become smaller (29.2°, 33.3°, 43.7°, and 52.4°), implying better π-conjugation (Figure 2D). The ACID plot of 2 shows a diatropic ring current circuit in the central porphyrin unit and the four thiophene rings directly linked to the porphyrin unit, whereas the bithiophene units are non-aromatic, similar to 1. The NICS value at the center of the two earrings in 2 is calculated to be 2.5 ppm, which should be arisen from the de-shielding effect from the nearby aromatic porphyrin and thiophene rings. The 3D ICSS map also shows a dominant shielding environment above and below the porphyrin unit (Figure 4B). On the other hand, the calculated ACID plot of the 22+ displays a dominant clockwise current circuit through the outer periphery of the π-conjugated skeleton (Figure 4A). However, detailed analysis of the chemical structure suggests that this is actually due to a superposition effect of the ring currents from three aromatic rings, including two aromatic earrings and one aromatic porphyrin unit, which possess a [22] and [18]annulene-like conjugation pathway, respectively (Figure S40). The cancelation of the clockwise ring current flow at the bridges (herein, the two fragments on porphyrin shared with the two earrings) leads to a weak clockwise ring current at the porphyrin unit and an overall high diatropic current density along the outer periphery, which is similar to the tricyclic anthracene molecule. The calculated HOMA values along different π-conjugation pathways reveal a similar enhancement of aromaticity of 22+ compared with the neutral compound 2 (Figures S43 and S44). The calculated NICS value at the two earring centers now becomes −5.9 ppm, indicating a switch from non-aromaticity to aromaticity. The calculated 3D ICSS map of 22+ further supports a magnetic shielding environment for all three cycles (Figure 4B). Therefore, the dication 22+ is also globally aromatic, with all three possible π-conjugated macrocycles being 2D Hückel aromatic. The change of electronic structure and aromaticity from 2 to 22+ was also demonstrated by experiment. The 1H NMR spectrum of 2 in CDCl3 at 298 K (Figure 3C) reveals a typical aromatic porphyrin structure, with resonances for the β-H appearing at the low field (δ = 9.36 and 8.62 ppm for protons e and f, respectively). The four thiophenes directly linked with the porphyrin units are also aromatic (δ = 6.78 and 7.60 ppm for protons c and d, respectively), whereas the quinoidal bithiophene units are non-aromatic (δ = 6.57 and 6.12 ppm for protons a and b, respectively). The peaks for the aromatic protons g and h on the aryl substituents are split (δ = 7.18, 7.14 ppm), indicating that the rotation of the aryl substituents is slow on NMR timescale in this rigid tricyclic molecule. The 1H NMR spectrum of 22+ in CD2Cl2 at 298 K shows that the resonances for the protons f on the porphyrin unit (δ = 10.20 ppm), protons a (δ = 9.10 ppm) and b (δ = 8.55 ppm) on the bithiophene units, and the protons g and h on the aryl substituents (δ = 7.72, 7.63 ppm) are all shifted to lower field as compared with that in 2, which can be explained by the de-shielding effect from the diatropic ring current along the outer periphery. However, the proton e on the porphyrin unit above the two earrings is shifted to higher field (δ = 8.79 ppm) due to the shielding effect from the diatropic ring current. Nevertheless, the chemical shift indicates an overall de-shielding chemical environment, which can be explained by three reasons: (1) the central porphyrin unit is still aromatic; (2) the backbone is non-planar and the proton e is far from the geometric center of the earrings (Figure 2D); and (3) the two positive charges are delocalized (Figure S50), which results in decreased electron density. It is also noted that the resonances for the protons c and d coalesced at room temperature, which is likely due to restricted rotation, as we previously observed in a similar system.55Ren L. Gopalakrishna T.Y. Park I.H. Han Y. Wu J. Porphyrin/quinoidal-bithiophene-based macrocycles and their dications: template-free synthesis and global aromaticity.Angew. Chem. Int. Ed. Engl. 2020; 59: 2230-2234Google Scholar In accordance with the global aromatic character, the compound 22+ in DCM shows an intense band with λabs at about 1,068 nm. In summary, a complex porphyrin and thiophene-based fully π-conjugated molecular cage (1) was synthesized by a facile strategy. A switch from 2D local aromaticity to 3D global aromaticity upon two-electron oxidation was demonstrated in its dication (12+). Importantly, we found that all the five types of π-conjugated 2D macrocycles formed from the constructional fragments are Hückel aromatic. This study, together with our previous analysis on a D3 symmetric cage c-T126+, strongly suggests that the 3D global aromaticity can actually be explained by the 2D Hückel aromaticity of the individual π-conjugated macrocycles that can be drawn in the 3D π-conjugated skeleton. Such a scenario is also observed in the tricyclic model compound 2. All these studies again demonstrate a fundamental chemical principle, that is, the π-conjugated molecules always have a tendency to become aromatic with the lowest energy state. The current C2 symmetric cage 12+ obviously does not follow Hirsch’s 2(N+1)2 spherical aromaticity rule. Continuous searching for highly symmetric, π-conjugated molecular cages is still needed to answer the question.