English

• Fluorescent materials have attracted widespread research interest in recent decades on account of their promising applications in light-emitting diodes, biological sensing, organic electronic devices and so on [1–4]. Nevertheless, conventional organic fluorophores generally showcase weakened or entirely quenched emission in condensed states owing to the aggregation-caused quenching (ACQ) effects, which vastly restricts their applications in necessary high concentration or solid state [5–7]. This puzzle was tactfully tackled when Tang and co-workers reported an inverse phenomenon known as aggregation-induced emission (AIE) [8,9]. Specific fluorophores are almost non-emissive in dilute solution yet display brilliant emission under aggregation states due to the astriction of intramolecular rotation [10,11]. Tetraphenylethylene (TPE) is the most representative AIE-active fluorophore whose free rotation of phenyl rings and twisting of the C=C bond are astricted in aggregative states, thereby suppressing the nonradiative decay and allowing efficient emission [12–14]. Due to the distinct optical behavior and facile modification, TPE fluorogens have been extensively introduced into organic frameworks, polymeric materials, and supramolecular assemblies to regulate their photophysical properties and facilitate the development of various intriguing fluorescent materials [15–20]. Fundamentally, the emission of TPE fluorophores is highly correlated to the degree of conformational distortion [21,22]. However, study on the relationship between fine structures of TPE-based fluorescent materials and the accompanying emissive behaviors represents a significant challenge.

  Coordination-driven self-assembly via the spontaneous formation of metal−ligand bonds can provide a powerful strategy for addressing the scientific issue above [23–26]. In this strategy, the self-assembly between the Lewis-basic donor and Lewis-acidic acceptor can go through efficient error correction and self-repairing process benefiting from their kinetic reversibility, thereby resulting in thermodynamically favorable supramolecular coordination complexes (SCCs) with predictable shapes and sizes [27,28]. As a kind of representative SCCs, discrete metallacages could have the well-defined three-dimensional geometry and conformation through regulating the angularity, size and stoichiometry of the rigid precursors, which are emerged as a promising platform to investigate AIE properties [29,30]. As such, the TPE fluorogens could be incorporated into well-designed metallacages to adjust their topological conformation, therefore leading to variable fluorescent properties. Although a wealth of delicate TPE-based emissive SCCs have been reported, the influence of topological geometries on fluorescent behavior is rarely investigated [31–33]. In coordination-driven self-assembly, subtle change of building blocks generally results in SCCs with different geometries [34,35]. Therefore, we envision that such SCCs can serve as an ideal model system to explore the impacts of fine structures and geometrical effect of TPE-based fluorescent materials on their emissive behaviors.

  Herein, we designed and synthesized two TPE-based metallacages, 1 and 2, with different topological geometries via the coordination-driven self-assembly, and further investigated the emissive behaviors of TPE in the distinguishing fine structures (Fig. 1). In specific, the TPE-based tetrapyridyl ligand 3 was selected to coordinate with the di-Pt(Ⅱ) acceptor 4 in a 3:6 ratio to form a barrel-shaped metallacage 1, and with the tri-Pt(Ⅱ) acceptor 5 in a 6:8 ratio to yield a ball-shaped metallacage 2, respectively. With the same donor, we obtain two metallacages with distinct geometries by varying the metal acceptors. Notably, the different topological conformations lead to variable emissive properties of the two metallacages in spite of possessing identical TPE core.

  Figure 1

  

  Figure 1.  Schematic representation of the formation of the metallacages 1 and 2 with different topological structures and photophysics via coordination-driven self-assembly.

  DownLoad: Full-Size Img PowerPoint

  In this work, we combined the TPE-based tetrapyridyl ligand 3 with the di-Pt(Ⅱ) acceptor 4 and tri-Pt(Ⅱ) acceptor 5, respectively, to fabricate two different TPE-containing metallacages with the barrel-shape or ball-shape (Fig. 1). In detail, by mixing the ligand 3 and di-Pt(Ⅱ) acceptor 4 in a 3:6 ratio at 70 ℃ for 8 h in deuterated DMSO, the barrel-shaped metallacage 1 was obtained in nearly quantitative yield. Meanwhile, the ball-shaped metallacage 2 was prepared via the self-assembly between the ligand 3 and tri-Pt(Ⅱ) acceptor 5 in a 6:8 ratio by following the same synthetic method.

  The formation of these two discrete TPE-containing metallacages was first investigated by NMR analyses (¹H and ³¹P{¹H}) of the reaction mixtures. As shown in Fig. 2c, the ¹H spectrum of the barrel-shaped metallacage 1 exhibited downfield shifts of the characteristic peaks corresponding to aromatic protons (H_{a-f, 1}) of coordinated structures, compared with those of pure ligand 3 (Fig. 2b) and di-Pt(Ⅱ) accepter 4 (Fig. 2d). In addition, the characteristic signals assigned to the aromatic protons (H_a) on the pyridine rings were split into two sets after metal coordination, which was in accordance with the phenomenon in the reported SCC structures [36,37]. Similar trends were also observed in the ¹H spectrum of the ball-shaped metallacage 2, showing the broadening signals with downfield shifts (Figs. 2g, h, and i). The obvious downfield shifts could attribute to the decrease in electron density of original protons brought by the formation of Pt−N coordination bonds. Two-dimensional diffusion-ordered ¹H NMR (DOSY) was also performed to provide more evidence for the formation of the simplex assembled products, which exhibit a nearly single vertical trace in both cases. The weight-average diffusion coefficients (D) were measured to be 8.55 × 10⁻¹¹ m²/s for the barrel-shaped metallacage 1 (Fig. S24 in Supporting information) and 11.7 × 10⁻¹¹ m²/s for the ball-shaped metallacage 2 (Fig. S25 in Supporting information). Moreover, both the ³¹P{¹H} spectra of metallacages, 1 and 2, exhibited the sharp singlets (ca. 16.13 ppm for 1 and 15.11 ppm for 2) with concomitant ¹⁹⁵Pt satellites, which corresponded to a simplex phosphorous environment (Figs. S26 and S27 in Supporting information). The NMR signals of metallacages, 1 and 2, had pronounced upfield shifts of about 5.78 and 6.19 ppm, respectively, relative to those of precursory Pt(Ⅱ) acceptors. The well-defined signals in ¹H and ³¹P{¹H} spectra of these species supported the formation of the single, discrete assemblies with high symmetry.

  Figure 2

  

  Figure 2.  Simplified chemical structures of metallacages 1 (a) and 2 (f). Partial ¹H NMR spectra (400 MHz, DMSO-d₆, 298 K) of ligand 3 (b, g), acceptor 4 (d) and 5 (i) as well as metallacages 1 (c) and 2 (h). ESI-TOF-MS spectra of metallacages 1 (e) and 2 (j).

  DownLoad: Full-Size Img PowerPoint

  The stoichiometry for the formation of discrete metallacages, 1 and 2, was further supported by observations of electrospray ionization time-of-flight mass spectrometry (ESI-TOF-MS). In the ESI-TOF-MS spectrum of metallacage 1 (Fig. 2e), multiple prominent peaks for [M − x(OTf)]^x+ (x = 6−12) were observed distinctly, which were assigned to the intact [3 + 6] assembly with charge states resulted from the loss of OTf⁻ counterions (e.g., peak at m/z 921.2068 corresponding to [M − 9OTf]⁹⁺). Similarly, the stoichiometry of discrete metallacage 2 was also evidenced by ESI-TOF-MS (Fig. 2j), which showed ten peaks corresponding to [M − x(OTf)]^x+ (x = 7−16), including the peak at m/z 1429.1384 for [M − 12OTf]¹²⁺. All the relevant peaks were isotopically resolved and in good agreement with their calculated theoretical distributions, suggesting the molecular structure of two discrete metallacages.

  In consideration of the difficulty in growing single crystals of these two assemblies suitable for X-ray diffraction, molecular simulations were conducted to afford a further understanding of the structural features of two TPE-containing metallacages, 1 and 2. All the geometry optimizations were accomplished by exploiting Gaussian09 software in a semiempirical PM6 method. The molecular simulations demonstrated a well-defined barrel-like structure for 1 with a ca. 1.6 × 2.5 × 1.1 nm cavity (Figs. 3a and b), while the simulated structure of 2 adopted a ball-like shape possessing a ca. 2.0 × 2.0 × 2.0 nm cavity (Figs. 3c and d). It is plausible that the ball-shaped metallacage with a closed-framework exert higher tension on TPE units compared to the barrel-shape structure with an open-framework. Such results indicated the ball-shaped assembly might have a stronger restriction on the freedom and flexibility of TPE ligands than the barrel-shaped structure, thereby leading to the different photophysical properties (vide infra).

  Figure 3

  

  Figure 3.  Ball-and-stick models of the optimized structures of metallacages 1 (a and b) and 2 (c and d). Green balls represent carbon atoms, blue balls represent nitrogen atoms, brown balls represent phosphorus atoms, and gray balls represent platinum atoms.

  DownLoad: Full-Size Img PowerPoint

  To investigate the photophysical behaviors of the two metallacages, we first recorded the absorption and emission profiles in dilute dichloromethane solutions. In the normalized UV–vis absorption spectra (Fig. 4a), the TPE-based ligand 3 exhibited two broad absorption bands centered at ca. 310 and 345 nm, respectively. After metal coordination, the barrel-shaped metallacage 1 only displayed one sharp absorption band at ca. 310 nm, while the broad absorption bands of ball-shaped metallacage 2 were centered at ca. 325 nm and 345 nm. In the corresponding fluorescence spectra (Fig. 4b), ligand 3 was weakly emissive at ca. 525 nm. On the contrary, two metallacages exhibited prominent fluorescence on account of the efficient suppression of TPE units within the rigid metallacage frameworks. Moreover, the emission maximum of barrel-shaped metallacage 1 blueshifted to 515 nm, and the ball-shaped metallacage 2 further blueshifted to 500 nm. According to the TPE-based AIE mechanism, certain wavelength shifts are typically related to specific conformations of the phenyl rings, where a coplanar conformation results in a red shift and a perpendicular conformation leads to a blue shift. The cage tension exerted on TPE units makes the phenyl rings more constrained and perpendicular than the original conformation. It is plausible that the ball-shaped metallacage 2 imposes more constraint on TPE units, thereby causing a more salient blue shift.

  Figure 4

  

  Figure 4.  Absorption (a) and fluorescence (b) spectra of ligand 3, metallacages 1 and 2 in CH₂Cl₂ (λ_ex = 360 nm, c_TPE = 10 µmol/L). In view of the metallacages 1 and 2 possessing different amounts of TPE units in the single structure, all photophysical tests were performed by maintaining the identical concentrations of fluorescent TPE units, namely c_TPE, of metallacages 1 and 2. (c) Quantum yield and maximum emission wavelength of metallacages 1 and 2 in CH₂Cl₂ (λ_ex = 360 nm, c_TPE = 10 µmol/L).

  DownLoad: Full-Size Img PowerPoint

  It is noteworthy that the ball-shaped metallacage 2 exhibited higher fluorescence intensity compared to the barrel-shaped metallacage 1. Therefore, we further demonstrated the quantum yields (Φ_F) of the two metallacages in dilute dichloromethane solutions (Figs. S28 and S29 in Supporting information). As expected, the Φ_F of 1 was only 3.47%, and that of 2 was increased to 8.34% (Fig. 4c). Since the enhanced fluorescence resulted from the restriction of TPE units in essence, the more enhanced emission of ball-shaped metallacage was due to higher cage tension of the sphere geometry.

  To gain further insight into the AIE properties of the two metallacages with different geometries, the fluorescence spectra were recorded in CH₂Cl₂/hexane mixtures (Figs. 5a and b). The addition of hexane into the CH₂Cl₂ solution could reduce the solubility of the metallacage and thus favor the formation of aggregation. With the increase of the hexane content, the fluorescence intensities were chronologically enhanced and finally reached a maximum at 90% hexane content. Such a phenomenon was consistent with classical AIE behavior, indicating the further restriction of the intramolecular motions of the embedded TPE units upon aggregation. Notably, in the aggregated states when the hexane content reached 90%, a 6.7-fold fluorescence enhancement was observed for the barrel-shaped assembly 1 (Fig. 5c), whereas ball-shaped 2 displayed only a 2.3-fold enhancement (Fig. 5d). In other words, the barrel-shaped scaffold showed a higher fluorescence intensity relative to the ball-shaped one in the aggregated states, which is opposite to what was observed in dilute solutions. This is likely due to greater rigidity as well as tension exerted on TPE units of the spherical structure. In contrast, the bulk sphere scaffold hindered the approach of aggregation in concentrated solutions, thereby showing less amplification. In addition, we observed various yet tunable emissions in different solvents due to the synergistic effect of solubility and polarity. The intensity and wavelength of two metallacages can be readily modulated by simply varying the solvents (Fig. S30 in Supporting information).

  Figure 5

  

  Figure 5.  Fluorescence spectra of metallacages 1 (a) and 2 (b) versus hexane fraction in CH₂Cl₂/hexane mixed solutions (λ_ex = 360 nm, c_TPE = 10 µmol/L). Plots of maximum emission intensity of metallacages 1 (c) and 2 (d) versus hexane fraction in CH₂Cl₂/hexane mixed solutions.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, we describe the synthesis of emissive barrel-shaped metallacage 1 and ball-shaped metallacage 2 through the stoichiometric and structural control of the same donor and different acceptors. Two metallacages were fully characterized by multinuclear NMR and ESI-TOF-MS for structural confirmation. Due to the differences in the strength and rigidity of the scaffold, as imposed by their shapes, different fluorescence enhancement behaviors were observed. In particular, the barrel-shaped metallacage 1 showed a weaker fluorescence in dilute solutions relative to that of the ball-shaped 2, while exhibiting a more salient increasing amplitude in the aggregation process. Such a phenomenon demonstrates that the properties of SCCs could be greatly influenced by their geometries, as the more constrained geometry would result in more salient fluorescence. Our work not only enriches the family of light-emitting SCCs, but also enhances the understanding of the mechanism of AIE moieties, thereby providing enlightenment for the novel design of emissive materials.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was financially supported by Interdisciplinary Program of Shanghai Jiao Tong University (No. YG2019QNA16), Shanghai Sailing Program (No. 20YF1422600), Natural Science Foundation of Shanghai (No. 22dz1207603), and National Natural Science Foundation of China (Nos. 32101092 and 21901161).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.108531.

  
  
• 1. [1]

     A. Bessette, G.S. Hanan, Chem. Soc. Rev. 43 (2014) 3342–3405. doi: 10.1039/C3CS60411J

  2. [2]

     V.W.W. Yam, V.K.M. Au, S.Y.L. Leung, Chem. Rev. 115 (2015) 7589–7728. doi: 10.1021/acs.chemrev.5b00074

  3. [3]

     L. Zhang, Y.F. Wang, M. Li, Q.Y. Gao, C.F. Chen, Chin. Chem. Lett. 32 (2021) 740–744. doi: 10.1016/j.cclet.2020.07.041

  4. [4]

     H. Sepehrpour, W. Fu, Y. Sun, P.J. Stang, J. Am. Chem. Soc. 141 (2019) 14005–14020. doi: 10.1021/jacs.9b06222

  5. [5]

     M. Li, S. Jiang, Z. Zhang, et al., CCS Chem. 2 (2020) 337–348. doi: 10.31635/ccschem.020.201900109

  6. [6]

     W. Qian, M. Zuo, P. Niu, X.Y. Hu, L. Wang, Chin. Chem. Lett. 33 (2022) 1975–1978. doi: 10.1016/j.cclet.2021.09.070

  7. [7]

     Y. Hong, J.W.Y. Lam, B.Z. Tang, Chem. Soc. Rev. 40 (2011) 5361–5388. doi: 10.1039/c1cs15113d

  8. [8]

     J. Luo, Z. Xie, J.W.Y. Lam, et al., Chem. Commun. 18 (2001) 1740–1741.

  9. [9]

     J. Mei, N.L. Leung, R.T. Kwok, J.W. Lam, B.Z. Tang, Chem. Rev. 115 (2015) 11718–11940. doi: 10.1021/acs.chemrev.5b00263

  10. [10]

      J. Mei, Y. Hong, J.W.Y. Lam, et al., Adv. Mater. 26 (2014) 5429–5479. doi: 10.1002/adma.201401356

  11. [11]

      Kenry, B.Z. Tang, B. Liu, Chem 6 (2020) 1195–1198. doi: 10.1016/j.chempr.2020.05.018

  12. [12]

      N. Jiang, Y. Wang, A. Qin, J.Z. Sun, Z.T. Ben, Chin. Chem. Lett. 30 (2019) 143–148. doi: 10.1016/j.cclet.2018.10.020

  13. [13]

      M. Zhang, M.L. Saha, M. Wang, et al., J. Am. Chem. Soc. 139 (2017) 5067–5074. doi: 10.1021/jacs.6b12536

  14. [14]

      J. Guan, R. Wei, A. Prlj, et al., Angew. Chem. Int. Ed. 59 (2020) 14903–14909. doi: 10.1002/anie.202004318

  15. [15]

      X. Ji Wang, Z. Li, F. Huang, Adv. Mater. 29 (2017) 1606117. doi: 10.1002/adma.201606117

  16. [16]

      Z. Wei, Z.Y. Gu, R.K. Arvapally, et al., J. Am. Chem. Soc. 136 (2014) 8269–8276. doi: 10.1021/ja5006866

  17. [17]

      X. Yan, T.R. Cook, P. Wang, F. Huang, P.J. Stang, Nat. Chem. 7 (2015) 342–348. doi: 10.1038/nchem.2201

  18. [18]

      M. Zhang, S. Li, X. Yan, et al., Proc. Natl. Acad. Sci. U. S. A. 113 (2016) 11100–11105. doi: 10.1073/pnas.1612898113

  19. [19]

      N.B. Shustova, B.D. McCarthy, M. Dincă, J. Am. Chem. Soc. 133 (2011) 20126–20129. doi: 10.1021/ja209327q

  20. [20]

      J. Dong, P. Shen, S. Ying, et al., Chem. Mater. 32 (2020) 6706–6720. doi: 10.1021/acs.chemmater.0c02277

  21. [21]

      Z. Guo, G. Li, H. Wang, et al., J. Am. Chem. Soc. 143 (2021) 9215–9221. doi: 10.1021/jacs.1c04288

  22. [22]

      Z. Guo, J. Zhao, Y. Liu, et al., Chin. Chem. Lett. 32 (2021) 1691–1695. doi: 10.1016/j.cclet.2020.12.028

  23. [23]

      S. Datta, M.L. Saha, P.J. Stang, Acc. Chem. Res. 51 (2018) 2047–2063. doi: 10.1021/acs.accounts.8b00233

  24. [24]

      T. Islamoglu, S. Goswami, Z. Li, et al., Acc. Chem. Res. 50 (2017) 805–813. doi: 10.1021/acs.accounts.6b00577

  25. [25]

      M. Pan, W.M. Liao, S.Y. Yin, S.S. Sun, C.Y. Su, Chem. Rev. 118 (2018) 8889–8935. doi: 10.1021/acs.chemrev.8b00222

  26. [26]

      Y. Sun, C. Chen, P.J. Stang, Acc. Chem. Res. 52 (2019) 802–817. doi: 10.1021/acs.accounts.8b00663

  27. [27]

      P.J. Stang, B. Olenyuk, Acc. Chem. Res. 30 (1997) 502–518. doi: 10.1021/ar9602011

  28. [28]

      Z. Zhou, X. Yan, M.L. Saha, et al., J. Am. Chem. Soc. 138 (2016) 13131–13134. doi: 10.1021/jacs.6b07173

  29. [29]

      W. Wang, Y.X. Wang, H. -B. Yang, Chem. Soc. Rev. 45 (2016) 2656–2693. doi: 10.1039/C5CS00301F

  30. [30]

      X. Yan, M. Wang, T.R. Cook, et al., J. Am. Chem. Soc. 138 (2016) 4580–4588. doi: 10.1021/jacs.6b00846

  31. [31]

      Y. Li, J. Zhang, H. Li, et al., Adv. Opt. Mater. 8 (2020) 1902190. doi: 10.1002/adom.201902190

  32. [32]

      X. Yan, P. Wei, Y. Liu, et al., J. Am. Chem. Soc. 141 (2019) 9673–9679. doi: 10.1021/jacs.9b03885

  33. [33]

      J. Zhao, Z. Zhou, G. Li, P.J. Stang, X. Yan, Natl. Sci. Rev. 8 (2021) nwab045. doi: 10.1093/nsr/nwab045

  34. [34]

      G. Li, Z. Zhou, C. Yuan, et al., Angew. Chem. Int. Ed. 59 (2020) 10013–10017. doi: 10.1002/anie.202000078

  35. [35]

      Y. Sun, C. Chen, J. Liu, P.J. Stang, Chem. Soc. Rev. 49 (2020) 3889–3919. doi: 10.1039/D0CS00038H

  36. [36]

      X. Yan, H. Wang, C.E. Hauke, et al., J. Am. Chem. Soc. 137 (2015) 15276–15286. doi: 10.1021/jacs.5b10130

  37. [37]

      C. Mu, Z. Zhang, Y. Hou, et al., Angew. Chem. Int. Ed. 60 (2021) 12293–12297. doi: 10.1002/anie.202100463

• 

  Figure 1  Schematic representation of the formation of the metallacages 1 and 2 with different topological structures and photophysics via coordination-driven self-assembly.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Simplified chemical structures of metallacages 1 (a) and 2 (f). Partial ¹H NMR spectra (400 MHz, DMSO-d₆, 298 K) of ligand 3 (b, g), acceptor 4 (d) and 5 (i) as well as metallacages 1 (c) and 2 (h). ESI-TOF-MS spectra of metallacages 1 (e) and 2 (j).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Ball-and-stick models of the optimized structures of metallacages 1 (a and b) and 2 (c and d). Green balls represent carbon atoms, blue balls represent nitrogen atoms, brown balls represent phosphorus atoms, and gray balls represent platinum atoms.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Absorption (a) and fluorescence (b) spectra of ligand 3, metallacages 1 and 2 in CH₂Cl₂ (λ_ex = 360 nm, c_TPE = 10 µmol/L). In view of the metallacages 1 and 2 possessing different amounts of TPE units in the single structure, all photophysical tests were performed by maintaining the identical concentrations of fluorescent TPE units, namely c_TPE, of metallacages 1 and 2. (c) Quantum yield and maximum emission wavelength of metallacages 1 and 2 in CH₂Cl₂ (λ_ex = 360 nm, c_TPE = 10 µmol/L).

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Fluorescence spectra of metallacages 1 (a) and 2 (b) versus hexane fraction in CH₂Cl₂/hexane mixed solutions (λ_ex = 360 nm, c_TPE = 10 µmol/L). Plots of maximum emission intensity of metallacages 1 (c) and 2 (d) versus hexane fraction in CH₂Cl₂/hexane mixed solutions.

  下载: 全尺寸图片 幻灯片
•