English

• Perylene diimides (PDI), a class of organic dye with excellent fluorescence, photo/thermal stability and low cost, have been intensively studied for nearly a century [1–3]. Due to the electron-deficient nature, PDI derivatives can be readily reduced to form radical anions PDI^•–, which exhibit typical strong absorption extending to the near-infrared (NIR) region, making them promising candidates for photothermal conversion applications. However, planar PDI chromophores have a highly strong tendency to self-aggregate by π–π stacking effect, which leads to fluorescence quenching [4,5] and loss of radical species [6,7], greatly deteriorating their photophysical properties. To circumvent the aggregation of PDIs, the most common strategy is the covalent derivatization with bulky substituents. For example, Müllen, Yin and Würthner et al. reported the derivatization with bulky substituents at the bay position of PDI core [8,9], and Würthner, Wong, Haag and Gao et al. introduced large substituents at the imide position [10–12]. However, complete insulation of chromophores is still not quite achieved in concentrated aqueous solutions for PDIs [13], particularly for labile radicals PDI^•– [6,7]. Recently, we have developed an elaborate doubly-strapped structure GBox-1⁴⁺ with the PDI chromophore as central core to realize the complete suppression of chromophore aggregation, and significantly stabilizes PDI radicals to show an excellent NIR photothermal effect [14]. However, the synthesis of GBox-1⁴⁺ involves costly palladium catalysts and tedious procedures, indicating that this promising strategy for constructing non-aggregating doubly-strapped structures still requires further development using more economical reagents and high-yield synthetic methods.

  Besides, the practical applications are unfavorably affected by the instability of radical species under ambient conditions [15–20]. The formation of environmentally stable organic radical species thus remains of great importance [21–25], especially for NIR photothermal conversion [26–28]. Many studies have focused on improving the stability of PDI^•–, such as covalently modifying PDI cores and incorporating PDI^•– into films or supramolecular assemblies [29–32]. Very recently, Yin and co-workers reported a crystalline metal–organic framework as a NIR photothermal material that contains air-stable PDI radical anions generated through photo-induced electron transfer [33]. However, additional electron donors of organic amines and photo-excitation process are still needed in this system to post-modify the hybrid material. Thus, air-stable radical-based PDI materials obtained without additional treatments would be highly desirable. For prokaryotic bacteria, it is known that respiration can generate a transmembrane redox potential, which acts as an energy source for supporting biological functions, bioelectrosynthesis, and microbial fuel cells [34–36]. Originating from this redox potential, bacteria can exhibit obvious reducing ability [37,38], which may be a powerful tool for the in situ fabrication of radical-based materials without additional procedures. Moreover, the unique microenvironment and spatial isolation effect of bacteria may stabilize the traditionally unstable radical species and obviously promote the NIR photothermal behavior of the biomass materials.

  Here we rationally synthesized a water-soluble doubly-strapped dye GBox-3⁴⁺ ("Gemini Box") with a PDI core enclosed by bilateral cationic bipyridinium straps. Thanks to the effectively spatial isolation, the chromophore aggregation has been demonstrated to be completely suppressed even in concentrated aqueous solutions up to 2 mmol/L, thereby preserving the characteristic fluorescence of PDI chromophore. After incubation of bacteria with GBox-3⁴⁺, the radical species PDI^•– have been proved to stably exist in the bacterial composites under ambient conditions, both for aqueous suspension and solid samples. Moreover, the dye-bacterial composites exhibited an high-efficiency NIR photothermal effect with high durability and were further exploited as photothermal agents for seawater desalination.

  Cyclophane GBox-3⁴⁺ was prepared from commercially available reagents in four steps (Fig. 1a). First, the angular triazol-based amine 1 was prepared from a benzonitrile derivative, and the imide condensation reaction of 1 and 1, 6, 7, 12-tetrachloroperylene tetracarboxylic dianhydride was then carried out in propionic acid at 155 ℃ to obtain tetrapodal PDI-1 [39]. After bromination with hydrobromic acid at 110 ℃, PDI-1 was transformed into the precursor PDI-Br. Macrocyclization reaction between PDI-Br and 4, 4′-vinylenedipyridine was straightforward performed in acetonitrile at 60 ℃ for 72 h to produce the corresponding dibromide salt of GBox-3·4Br, which was then transformed into GBox-3·4PF₆ with high purity via anion exchange [40] (the overall cyclization yield is 18%). The water-soluble GBox-3·4Cl was finally obtained by adding excess tetrabutylammonium chloride to acetonitrile solution of GBox-3·4PF₆. Compared to our previously developed double-cavity cyclophane GBox-1⁴⁺, the triazol-based cyclophane in this work was synthesized using cost-effective starting materials through several high-yielding steps, eliminating the need for expensive palladium catalysts. This optimized synthetic process facilitates successful scale-up, paving the way for further practical applications.

  Figure 1

  

  Figure 1.  (a) Synthetic route to GBox-3⁴⁺. (b) Partial ¹H NMR spectrum of GBox-3·4Cl in D₂O. (c) HRMS of GBox-3·4Cl. (d) The optimized structure of GBox-3⁴⁺ at the M06-2X/6-31+G** level.

  DownLoad: Full-Size Img PowerPoint

  The formation of GBox-3⁴⁺ with a highly symmetrical structure was proved by ¹H NMR spectrum (Fig. 1b). The signals of pyridinyl and methylene protons are shifted downfield (Fig. S3 in Supporting information), which is attributed to the formation of pyridinium moieties. High-resolution electrospray ionization mass spectrometry (HRMS) shows intense high-resolution signal at m/z = 345.5617 and 472.4035, which can be assigned to [M-4Cl]⁴⁺ and [M-3Cl]³⁺ species with the loss of four and three anions (Fig. 1c). Attempts to obtain crystals of GBox-3⁴⁺ suitable for single crystal X-ray diffraction analysis were unsuccessful, thus geometry optimization was performed using the M06-2X/6-31+G** method [41]. As expected, GBox-3⁴⁺ has a doubly-strapped structure, and the central core of large PDI chromophore is enclosed by the pyridinium sidewalls on both sides. The centroid distance between PDI and 4, 4′-vinylenedipyridinium units is ca. 6.9 Å (Fig. 1d), which avoids the formation of face-to-face π–π stacking of PDIs.

  The UV–vis spectra of GBox-3·4Cl in water showed a regular vibronic signal pattern with a maximum value at 543 and 505 nm (Fig. 2a), corresponding to 0–0 and 0–1 vibronic transition, respectively [42,43]. With the concentration increased from 0.2 mmol/L to 2 mmol/L, the extinction coefficient and absorption ratio of the two main bands (A_{0, 0}/A_{0, 1}~1.38) remain essentially unchanged. In contrast, the A_{0, 0}/A_{0, 1} of the acyclic reference PDI-ref²⁺ (see Supporting information for details) were eventually reduced from ~1.15 to 0.93 (Fig. 2b and Fig. S15 in Supporting information). Since the reduction of A_{0, 0}/A_{0, 1} value indicates a strong chromophore-chromophore interaction, it can be concluded that the PDI aggregation of GBox-3⁴⁺ is thoroughly eliminated, even in concentrated aqueous solutions up to 2 mmol/L.

  Figure 2

  

  Figure 2.  UV–vis absorption spectra of aqueous solutions of (a) GBox-3⁴⁺ from 0.2 mmol/L to 2 mmol/L and (b) PDI-ref²⁺ from 0.2 mmol/L to 1.4 mmol/L. (c-e) Colocalization images of GBox-3⁴⁺ with LTR for RAW 264.7 cells. Scale bar: 8 µm.

  DownLoad: Full-Size Img PowerPoint

  As anticipated, GBox-3·4Cl exhibited characteristic fluorescence emissions of PDI derivatives centered at around 590 nm over the investigated concentration range of 10–500 µmol/L in water (Fig. S16 in Supporting information) [44], which encourages us to employ GBox-3⁴⁺ as a fluorescent probe for live-cell imaging. The cell counting kit-8 (CCK-8) assay showed that GBox-3⁴⁺ was non-cytotoxic to RAW 264.7 macrophage cells with > 90% cell viability at the tested concentration range of 1–125 µmol/L (Fig. S17 in Supporting information). To elucidate the distribution of GBox-3⁴⁺ in live cells, RAW 264.7 cells were coincubated with GBox-3⁴⁺ (100 µmol/L) and LysoTracker Red (LTR, a lysosome marker). Confocal microscopy images (Figs. 2c-e) displayed that there was a large fluorescence overlay between GBox-3⁴⁺ and LTR with high overlay coefficient of ~0.945 (Fig. S19 in Supporting information), indicating the perfect lysosomal localization of GBox-3⁴⁺.

  These results show that the doubly-strapped PDI chromophore in GBox-3⁴⁺ can be spatially insulated by covalent linkage to avoid intermolecular π-π stacking of chromophores in aqueous solution, in contrast to that of the acyclic reference PDI-ref²⁺. Subsequently, we investigated their redox behaviors. The cyclic voltammograms (CVs) indicated that the first and second reduction peaks of GBox-3⁴⁺ in aqueous solutions are at −0.13 V and −0.38 V, corresponding to the one-electron reduction process of PDI/PDI^•– and PDI^•–/PDI^2-, respectively (Fig. 3a). By contrast, PDI-ref²⁺ has the lower first and second reduction potentials (−0.35 and −0.83 V) [45]. Such a significant alteration of reduction potential has been observed in a supramolecular PDI-cucurbituril system [46], where the spatial isolation effect from host-guest interactions positively shifts the first reduction potential from −0.37 V for free PDI to 0.16 V in water. Additionally, this redox behavior is consistent with the following characterization by UV–vis absorption spectroscopy (Fig. 3b). When excessive 6 equiv. Na₂S₂O₄ as a reducing agent was added to 0.1 mmol/L GBox-3⁴⁺ solution, the characteristic absorption bands of PDI^•– centered at 775 nm and PDI^2- at 642 nm simultaneously occurred. Nevertheless, there was only the PDI^•– signal with low concentration for PDI-ref²⁺ solution under the identical reduction conditions, but no signal of PDI^2- species was observed. This probably can be ascribed to the chromophore aggregation of PDI-ref²⁺ in water, which affected its redox behavior and prevented the further reduction to PDI^2- species. Since the reduced PDI species are easy to be quenched by oxygen, the characteristic absorption bands of PDI^•– and PDI^2- completely disappeared after the 3 h upon exposure to air, whether for GBox-3⁴⁺ or PDI-ref²⁺.

  Figure 3

  

  Figure 3.  (a) Cyclic voltammetry curves of 0.1 mmol/L GBox-3⁴⁺ and PDI-ref²⁺ aqueous solutions and chemical structures of radical anion and dianion. (b) UV–vis absorption spectra when adding excessive Na₂S₂O₄ reducing agent. (c) Electron paramagnetic resonance spectra of water-suspended samples. (d) Photothermal heating curves and (e) infrared thermal imaging under irradiation of 808 nm laser at 1 W/cm². (f) Photothermal cycle curve of water-suspended sample of GBox-3⁴⁺-S. aureus.

  DownLoad: Full-Size Img PowerPoint

  It is known that PDI moieties can be reduced in situ by bacteria with reductive ability, thus GBox-3⁴⁺ and PDI-ref²⁺ were both incubated with a common bacteria S. aureus. The formation of PDI radical species is indicative of dark blue dye-bacteria composites, accompanied by the appearance of NIR absorption and EPR signal. Interestingly, after exposure to air for more than 6 h, the EPR signal with g value of 2.0038 still maintained for the GBox-3⁴⁺-S. aureus suspension [47], whereas the PDI-ref²⁺-S. aureus and GBox-3⁴⁺-Na₂S₂O₄ systems showed negligible EPR signals (Fig. 3c). For another bacteria E. coli, similar phenomena were observed, and only GBox-3⁴⁺-bacteria system exhibited stable EPR signals under ambient conditions (Fig. S31 in Supporting information).

  With air-stable PDI^•– species in hand, the photothermal effect of the bacterial composite extending to the NIR region was thus systematacially studied. The aqueous suspension of 0.1 mmol/L GBox-3⁴⁺-S. aureus can be heated with a temperature rise of 35  ℃ upon 808 nm laser irradiation at a power density of 1 W/cm², showing a good NIR photothermal effect (Fig. 3d and e). By contrast, the solutions of GBox-3⁴⁺-Na₂S₂O₄ and PDI-ref²⁺-S. aureus suspension as the control groups displayed much weaker photothermal effects. The photothermal conversion efficiency of GBox-3⁴⁺-S. aureus system was calculated to be as high as 57.58% (Fig. S30 in Supporting information). The 808 nm Laser with different power densities from 0.50 W/cm² to 2.00 W/cm² was used to irradiate the water-suspended samples of GBox-3⁴⁺-S. aureus, and showed a power dependent photothermal effect. To further assess the photothermal stability of the bacterial composite, the suspension was irradiated by 808 nm laser at 1.0 W/cm² through ten heating/cooling cycles, and there was no obvious substantial performance deterioration after irradiation (Fig. 3f), indicating high light/thermal stability of the bacterial composite in water.

  The solid samples were initially obtained by centrifugation of the water-suspended samples with a dye concentration of 0.2 mmol/L, and then dried under ambient conditions. As shown by image of the scanning electron microscopy (SEM), the aggregated bacteria in the GBox-3⁴⁺-S. aureus solid retained their normal round shape (Fig. 4a), indicating that the binding of dye GBox-3⁴⁺ did not change the bacterial morphology. Similarly, there was an intense EPR signal only for the GBox-3⁴⁺-S. aureus solid, whereas the PDI-ref²⁺-S. aureus and GBox-3⁴⁺-Na₂S₂O₄ systems did not show perceptible EPR signals (Fig. 4b). Accordingly, only the GBox-3⁴⁺-S. aureus solid showed an excellent photothermal effect, and the temperature could rise from the initial temperature of 30~120  ℃ within 1 min upon 808 nm laser irradiation at 1.0 W/cm² (Figs. 4c and d). In sharp contrast, the control groups of GBox-3⁴⁺-Na₂S₂O₄ and PDI-ref²⁺-S. aureus samples did not reveal obvious NIR photothermal effects. In addition, the GBox-3⁴⁺-S. aureus solid also exhibited a power-dependent photothermal property by the 808 nm laser irradiation with different power densities from 0.50 to 2.00 W/cm² (Fig. 4e). The curves of ten continuous heating and cooling cycles indicated that the GBox-3⁴⁺-S. aureus solid has a high photothermal durability (Fig. 4f). The solid samples involving other bacteria also exhibited similar behaviors (Figs. S26–S29 in Supporting information).

  Figure 4

  

  Figure 4.  (a) SEM image of GBox-3⁴⁺-S. aureus solid sample (scale bar: 5 µm). (b) Electron paramagnetic resonance spectra of solid samples. (c) Photothermal heating curves and (d) infrared thermal imaging of solid samples under irradiation of 808 nm laser at 1 W/cm². (e) Temperature rises at different NIR laser intensities (W/cm²) and (f) photothermal cycling curve of the GBox-3⁴⁺-S. aureus solid sample (1 W/cm²).

  DownLoad: Full-Size Img PowerPoint

  Based on the above results, we can conclude that the air stability of the radical species relies on the simultaneous existence of the doubly-strapped PDI dye and bacteria. On the one hand, the doubly-strapped structure is a vital factor to stabilize radical species by preventing the chromophore aggregation. On the other, bacteria can provide the reductive microenvironment and the sufficient spatial isolation effect that shield the induced PDI radical species from oxygen in the air.

  Owing to the remarkable photothermal properties, particularly its absorption bands extending into the NIR region, the solid sample GBox-3⁴⁺-S. aureus was further examined as a photothermal agent for seawater desalination under simulated solar conditions (Fig. 5a) [48]. A 10 mg sample of GBox-3⁴⁺-S. aureus powder was uniformly dispersed onto a polytetrafluoroethylene (PTFE) membrane with a 2.0 mm pore size. This prepared membrane was then floated on water and directly utilized in subsequent experiments. The temperature/time profiles of the GBox-3⁴⁺-S. aureus-PTFE membrane versus the PTFE substrate alone under solar illumination are presented in Fig. 5b. The PTFE blank reached an equilibrium temperature of 43 ℃, indicating minimal contribution to photothermal conversion. In contrast, the GBox-3⁴⁺-S. aureus-PTFE membrane achieved a significantly higher equilibrium temperature of 128 ℃, reflecting its rapid and sustained temperature increase. To investigate its solar photothermal conversion capacity, the GBox-3⁴⁺-S. aureus-PTFE membrane was employed as a heat source for water evaporation under solar irradiation (5 suns, 5 kW/m²) [49]. A notable temperature rise was observed, accompanied by visible water vapor production, with the membrane reaching a final equilibrium temperature of 110  ℃. Conversely, the PTFE membrane alone only reached 51  ℃ (Fig. 5c). The water evaporation capacity of the GBox-3⁴⁺-S. aureus-PTFE membrane was further assessed by monitoring the change in water mass over 1 h of sunlight exposure. The membrane exhibited a water evaporation rate of 5.59 kg m^-2 h^-1, corresponding to a solar conversion efficiency of 70.37% under 5 suns irradiation (Fig. 5d) [50]. In contrast, the PTFE membrane without GBox-3⁴⁺-S. aureus displayed a much lower evaporation rate of 1.06 kg m^-2 h^-1 and an efficiency of 13.3% (Table S1 in Supporting information). These findings highlight the high light-to-water evaporation efficiency of the GBox-3⁴⁺-S. aureus membrane.

  Figure 5

  

  Figure 5.  (a) The simulated setup for recording the change of water mass under solar irradiation. (b) Temperature change of GBox-3⁴⁺-S. aureus-PTFE under 5 sun irradiation (5 kW/m²). The insets are IR thermal images of GBox-3⁴⁺-S. aureus-PTFE under these condistions. (c) Temperature change of the membranes floating on water under 5 sun irradiation. (d) Water evaporation curves of the membranes under 5 sun irradiation.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, the rational design of the water-soluble PDI cyclophane GBox-3⁴⁺ effectively expands the applications of PDI as fluorophores in aqueous environments and as NIR absorbers for photothermal conversion. The incorporation of double-sided cationic molecular straps ensures complete suppression of chromophore aggregation, even in concentrated aqueous solutions up to 2 mmol/L, allowing GBox-3⁴⁺ as a live-cell fluorescent probe. The stable existence of the radical species PDI^•– in GBox-3⁴⁺-bacteria composites can be observed under ambient conditions, and the stabilization of PDI^•– can be contributed to the doubly-strapped PDI structure and bacterial microenvironment. The resultant dye-bacterial composites exhibit a high-efficiency NIR photothermal effect with high durability, making them promising agents for seawater desalination. This work highlights the potential of rational molecular design and the innovative use of bacterial microenvironments in the development of advanced 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.

  CRediT authorship contribution statement

  Jingjing Zhang: Writing – original draft, Investigation, Data curation. Fei Yang: Validation, Funding acquisition, Data curation. Liying Zhang: Investigation, Data curation. Ran Li: Investigation, Data curation. Guo Wang: Validation, Software. Yanqing Xu: Writing – review & editing, Supervision, Funding acquisition. Wei Wei: Writing – review & editing, Funding acquisition, Conceptualization.

  Acknowledgments

  This work was supported by the Beijing Natural Science Foundation (Nos. 2242004 and 2232027), the National Natural Science Foundation of China (No. 22171021), and the China Postdoctoral Science Foundation (No. 2023M730245).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110627.

  
  
• 1. [1]

     S. Chen, P. Slattum, C. Wang, L. Zang, Chem. Rev. 115 (2015) 11967–11998. doi: 10.1021/acs.chemrev.5b00312

  2. [2]

     C.W. Struijk, A.B. Sieval, J.E. Dakhorst, et al., J. Am. Chem. Soc. 122 (2000) 11057–11066.

  3. [3]

     F. Würthner, C.R. Saha-Möller, B. Fimmel, et al., Chem. Rev. 116 (2016) 962–1052. doi: 10.1021/acs.chemrev.5b00188

  4. [4]

     D. Görl, X. Zhang, F. Würthner, Angew. Chem. Int. Ed. 51 (2012) 6328–6348. doi: 10.1002/anie.201108690

  5. [5]

     K. Fu, G. Liu, ACS Nano 18 (2024) 2279–2289. doi: 10.1021/acsnano.3c10151

  6. [6]

     R.O. Marcon, S. Brochsztain, J. Phys. Chem. A 113 (2009) 1747–1752. doi: 10.1021/jp808383e

  7. [7]

     Z. Cheng, F. Xing, Y. Bai, et al., Asian J. Org. Chem. 6 (2017) 1612–1619. doi: 10.1002/ajoc.201700334

  8. [8]

     M. Sun, K. Müllen, M. Yin, Chem. Soc. Rev. 45 (2016) 1513–1528.

  9. [9]

     D. Schmidt, M. Stolte, J. Süβ, et al., Angew. Chem. Int. Ed. 58 (2019) 13385–13389. doi: 10.1002/anie.201907618

  10. [10]

      B. Zhang, H. Soleimaninejad, D.J. Jones, et al., Chem. Mater. 29 (2017) 8395–8403. doi: 10.1021/acs.chemmater.7b02968

  11. [11]

      T. Heek, C. Fasting, C. Rest, et al., Chem. Commun. 46 (2010) 1884–1886. doi: 10.1039/b923806a

  12. [12]

      B. Gao, H. Li, H. Liu, et al., Chem. Commun. 47 (2011) 3894–3896. doi: 10.1039/c1cc00058f

  13. [13]

      S. Rehm, V. Stepanenko, X. Zhang, T.H. Rehm, F. Würthner, Chem. Eur. J. 16 (2010) 3372–3382. doi: 10.1002/chem.200902839

  14. [14]

      F. Yang, R. Li, W. Wei, et al., Angew. Chem. Int. Ed. 61 (2022) e202202491.

  15. [15]

      D. Schmidt, D. Bialas, F. Würthner, Angew. Chem. Int. Ed. 54 (2015) 3611–3614. doi: 10.1002/anie.201408067

  16. [16]

      V.V. Roznyatovskiy, D.M. Gardner, S.W. Eaton, M.R. Wasielewski, Org. Lett. 16 (2014) 696–699. doi: 10.1021/ol403736m

  17. [17]

      R. Kang, R. Miao, Y. Qi, et al., Chem. Commun. 53 (2017) 10018–10021.

  18. [18]

      S. Kumar, P. Mukhopadhyay, Green Chem. 20 (2018) 4620–4628. doi: 10.1039/c8gc01614c

  19. [19]

      J.E. Bullock, M.T. Vagnini, C. Ramanan, J. Phys. Chem. B 114 (2010) 1794–1802. doi: 10.1021/jp908679c

  20. [20]

      S. Kumar, M.R. Ajayakumar, G. Hundal, P. Mukhopadhyay, J. Am. Chem. Soc. 136 (2014) 12004–12010. doi: 10.1021/ja504903j

  21. [21]

      C. Shu, Z. Yang, A. Rajca, Chem. Rev. 123 (2023) 11954–12003. doi: 10.1021/acs.chemrev.3c00406

  22. [22]

      J. Yang, X. Tian, L. Yuan, et al., Chin. Chem. Lett. 35 (2024) 109745.

  23. [23]

      L. Ji, J. Shi, J. Wei, T. Yu, W. Huang, Adv. Mater. 32 (2020) 1908015.

  24. [24]

      Z. Chen, W. Li, M. Sabuj, et al., Nat. Commun. 12 (2021) 5889.

  25. [25]

      C. Chen, Q. Liang, Z. Chen, et al., Angew. Chem. Int. Ed. 60 (2021) 26718–26724. doi: 10.1002/anie.202110441

  26. [26]

      Z. Wang, J. Zhou, Y. Zhang, W. Zhu, Y. Li, Angew. Chem. Int. Ed. 61 (2022) e202113653.

  27. [27]

      S. Wang, S. Li, J. Xiong, et al., Chem. Commun. 56 (2020) 7399. doi: 10.1039/d0cc02193h

  28. [28]

      Z. Chen, Y. Li, F. Huang, Chem 7 (2021) 288–332.

  29. [29]

      Y. Jiao, K. Liu, G. Wang, Y. Wang, X. Zhang, Chem. Sci. 6 (2015) 3975–3980.

  30. [30]

      R.O. Marcon, S. Brochsztain, Langmuir 23 (2007) 11972–11976. doi: 10.1021/la702642h

  31. [31]

      M.A. Iron, R. Cohen, B. Rybtchinski, J. Phys. Chem. A 115 (2011) 2047–2056. doi: 10.1021/jp1107284

  32. [32]

      E. Shirman, A. Ustinov, N. Ben-Shitrit, et al., J. Phys. Chem. B 112 (2008) 8855–8858. doi: 10.1021/jp8029743

  33. [33]

      B. Lü, Y. Chen, P. Li, K. Müllen, M. Yin, Nat. Commun. 10 (2019) 767.

  34. [34]

      J.M. Benarroch, M. Asally, Trends Microbiol. 28 (2020) 304–314.

  35. [35]

      Y. Su, S. Cestellos-Blanco, J.M. Kim, et al., Joule 4 (2020) 800–811.

  36. [36]

      S. You, M. Ma, W. Wang, et al., Adv. Energy Mater. 7 (2017) 1601 -364.

  37. [37]

      R. Qi, H. Zhao, X. Zhou, et al., Angew. Chem. Int. Ed. 60 (2021) 5759–5765. doi: 10.1002/anie.202015247

  38. [38]

      Y. Yang, P. He, Y. Wang, et al., Angew. Chem. Int. Ed. 56 (2017) 16239–16242. doi: 10.1002/anie.201708971

  39. [39]

      G. Rauch, S. Höger, Chem. Commun. 50 (2014) 5659–5661. doi: 10.1039/c4cc02124j

  40. [40]

      M. Mahl, K. Shoyama, A.M. Krause, D. Schmidt, F. Würthner, Angew. Chem. Int. Ed. 59 (2020) 13401–13405. doi: 10.1002/anie.202004965

  41. [41]

      Y. Zhao, D.G. Truhlar, Theor. Chem. Acc. 120 (2008) 215–241. doi: 10.1007/s00214-007-0310-x

  42. [42]

      P. Spenst, F. Würthner, Angew. Chem. Int. Ed. 54 (2015) 10165–10168. doi: 10.1002/anie.201503542

  43. [43]

      J.M. Giaimo, J.V. Lockard, L.E. Sinks, et al., J. Phys. Chem. A 112 (2008) 2322–2330. doi: 10.1021/jp710847q

  44. [44]

      F. Yang, M.M. Zhen, S.S. Wang, et al., Chin. Chem. Lett. 33 (2022) 5088–5091.

  45. [45]

      L. Cui, Y. Jiao, A. Wang, et al., Chem. Commun. 54 (2018) 2208–2211. doi: 10.1039/c8cc00177d

  46. [46]

      Y. Yang, Z. An, Polym. Chem. 10 (2019) 2801–2811. doi: 10.1039/c9py00393b

  47. [47]

      Y. Jiao, K. Liu, G. Wang, Y. Wang, X. Zhang, Chem. Sci. 6 (2015) 3975–3980.

  48. [48]

      F. Yang, Y. Li, K. Huang, W. Wei, Y. Xu, New J. Chem. 48 (2024) 2901–2906. doi: 10.1039/d3nj05714c

  49. [49]

      M. Shen, X. Zhao, L. Han, et al., Chem. Eur. J. 28 (2022) e202104137.

  50. [50]

      L. Bian, L. Jia, Y. Zhou, et al., Mater. Today Commun. 33 (2022) 104337.

• 

  Figure 1  (a) Synthetic route to GBox-3⁴⁺. (b) Partial ¹H NMR spectrum of GBox-3·4Cl in D₂O. (c) HRMS of GBox-3·4Cl. (d) The optimized structure of GBox-3⁴⁺ at the M06-2X/6-31+G** level.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  UV–vis absorption spectra of aqueous solutions of (a) GBox-3⁴⁺ from 0.2 mmol/L to 2 mmol/L and (b) PDI-ref²⁺ from 0.2 mmol/L to 1.4 mmol/L. (c-e) Colocalization images of GBox-3⁴⁺ with LTR for RAW 264.7 cells. Scale bar: 8 µm.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Cyclic voltammetry curves of 0.1 mmol/L GBox-3⁴⁺ and PDI-ref²⁺ aqueous solutions and chemical structures of radical anion and dianion. (b) UV–vis absorption spectra when adding excessive Na₂S₂O₄ reducing agent. (c) Electron paramagnetic resonance spectra of water-suspended samples. (d) Photothermal heating curves and (e) infrared thermal imaging under irradiation of 808 nm laser at 1 W/cm². (f) Photothermal cycle curve of water-suspended sample of GBox-3⁴⁺-S. aureus.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) SEM image of GBox-3⁴⁺-S. aureus solid sample (scale bar: 5 µm). (b) Electron paramagnetic resonance spectra of solid samples. (c) Photothermal heating curves and (d) infrared thermal imaging of solid samples under irradiation of 808 nm laser at 1 W/cm². (e) Temperature rises at different NIR laser intensities (W/cm²) and (f) photothermal cycling curve of the GBox-3⁴⁺-S. aureus solid sample (1 W/cm²).

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) The simulated setup for recording the change of water mass under solar irradiation. (b) Temperature change of GBox-3⁴⁺-S. aureus-PTFE under 5 sun irradiation (5 kW/m²). The insets are IR thermal images of GBox-3⁴⁺-S. aureus-PTFE under these condistions. (c) Temperature change of the membranes floating on water under 5 sun irradiation. (d) Water evaporation curves of the membranes under 5 sun irradiation.

  下载: 全尺寸图片 幻灯片
•