Wavelength-dependent tri-state photoswitching in crystals via synergistic dimer confinement and charge-transfer stacking

Abstract

Anthracene (An) is a well-known photoactive unit that undergoes [4+4] dimerization upon UV excitation or a hetero-Diels–Alder reaction with singlet oxygen to form epoxide under visible light. Despite this dual reactivity, there is no single material exhibiting both photoresponses simultaneously. Herein, by utilizing an electron-deficient metal–organic framework (MOF) host, we obtain an An-dimer-encapsulated MOF cocrystal enabling wavelength-dependent tri-state photoswitching. Specifically, the strontium-containing MOF with preorganized arrays of naphthalene diimide (NDI) dimers and suitable cavity size of 10.3 Å is found to incorporate an An dimer in each cavity, forming extended crystalline π stacks with an A–D–D–A motif. Detailed structure–property analysis reveals the significant impact of dimer formation and charge-transfer stacking on the photophysical properties and photoactivity of the Sr-NDI@An cocrystal. Remarkably, Sr-NDI@An exhibits three distinct wavelength-selective photoresponses: (i) UV-triggered [4+4] dimerization of guest An, (ii) red-light-induced hetero-Diels–Alder reaction between singlet oxygen and An, and (iii) near-infrared-light-driven radical generation. This conceptual work opens the door toward the facile modulation of aromatic stacking for desirable complex functions.

---

New concepts

Most of the photoresponsive materials can only undergo one transformation toward light excitation, which hardly meets the requirement for advanced smart materials for applications in secured information transaction, high-capacity data storage, and biomedical microrobotics. Herein, utilizing an electron-deficient metal–organic framework (MOF) as a host, we obtain the An dimer encapsulated MOF cocrystal for wavelength-dependent tri-state photoswitching. Specifically, the strontium-containing MOFs with preorganized arrays of naphthalene diimide (NDI) dimers and a suitable cavity size of 10.30 Å are found to incorporate An dimer in each cavity to achieve the fabrication of extended crystalline π stacks with A–D–D–A motif. Spectral evidence verified that three photoreactions occurred for this single crystal: UV triggered [4+4] dimerization of guest An, red light-excited hetero-Diels–Alder reaction between ¹O₂ and An, and near-infrared light (NIR) driven radical generation. The synergy of CT interaction and dimer formation is essential for this unprecedented three-way photoresponsiveness. Our conceptual work opens up opportunities for the goal-oriented design of functional materials exploiting the rational packing of donors and acceptors, which might contribute to the development of smart materials for applications in product authentication, next-generation opto-electronics, and microrobotics for engineering or biomedicine.

1. Introduction

Stimulus-responsive materials are important models for understanding dynamic behaviors in biological systems such as signal transduction.^1–3 Among the stimuli, light is highly desirable for manipulating material properties in a clean, efficient and precise manner.⁴ Nowadays, light-responsive materials typically consist of photoactive units that can undergo a single photoreaction such as isomerization, cycloaddition, or radical generation.^5–7 Developing wavelength-dependent multi-photoresponsive materials remains challenging yet crucial for advanced applications such as secure information transfer, high-capacity data storage, and precision medicine.^8,9

Anthracene (An) is a good candidate for constructing multiple photoresponsive materials in view of its varied photoreactions under distinct conditions. Firstly, the photodimerization of An to dianthracene represents the oldest known [4+4] photocycloaddition reaction for wide applications in the synthesis of two-dimensional polymers,¹⁰ photomechanical grippers of soft robotics,¹¹ and the chirality transfer from helical metal nanostructures to organic molecules.¹² Normally, UV excitation of wavelength longer than 320 nm is required to initiate the [4+4] photodimerization without decomposing the product dianthracene. In the meantime, the distance between the two participating An molecules needs to obey Schmidt's rule, implying that it should be less than 4.2 Å. Therefore, most [4+4] photodimerization reactions occur in the solution state to ensure the collision between An molecules. For rare solid-state examples, the confinement of multiple An molecules within host materials^13,14 or strong π-interactions between An derivatives^15,16 are essential preconditions. Importantly, the photodimerization of An is reversible with high-energy UV irradiation (λ = 250–280 nm) or thermal treatment (≈200 °C), enabling dynamic behaviors in smart materials and optical writing/erasing cycles.

Besides the [4+4] photodimerization, the [4+2] hetero-Diels–Alder cycloaddition between singlet oxygen (¹O₂) and An is another widely utilized photoreaction of An. Typically, a photosensitizer is required for the light-induced generation of ¹O₂. In previous reports methylene blue (MB)^17,18 or porphyrin derivatives^19,20 have been constantly utilized as photosensitizers under red-light excitation (λ = 620–660 nm). To date, this An-involving [4+2] cycloaddition was only reported in solution, which might originate from the requirement for an additional photosensitizer and the efficient diffusion of reactive species ¹O₂.

Therefore, the key to furnishing the target material possessing a dual photoactivity for [4+4] photodimerization and [4+2] hetero-Diels–Alder reaction, is simultaneously satisfying the close alignment of the An dimer and the efficient generation of ¹O₂. As mentioned above, the confinement of the An dimer in the host materials is a feasible approach to realize [4+4] photodimerization. However, no study has unambiguously determined the spatial arrangement of An dimers within a single host lattice to enable efficient [4+4] photodimerization. Existing host–guest systems for this reaction rely on computational models for An distribution.^13,14 The other challenge is the generation of ¹O₂ within An-dimer-encapsulated host. The photosensitizer MB or porphyrin act as a guest rather than a structural component of porous materials. Additionally, for a porphyrin-based host, the interaction between porphyrin and An is too weak. Alternatively, ¹O₂ may be generated via charge transfer (CT) systems involving An, bypassing traditional photosensitization. Recently, through space CT (TSCT) has been suggested as a workable mechanism to render a donor–acceptor (D–A) system with thermally activated delayed fluorescence (TADF).^21,22 The CT excited singlet state (¹CT) can undergo intersystem crossing (ISC) to reach the triplet state (³CT), which can transfer energy to ambient oxygen, giving ¹O₂. Critically, An is an electron-rich aromatic molecule, which easily forms CT complexes with electron-deficient acceptors such as naphthalene diimide (NDI).²³

To enable simultaneous incorporation of anthracene dimer and CT pair within a single solid, organic D–A cocrystals represent a promising solution. Ordered arrays of D–A pairs are characteristic for organic cocrystals, and the interchromophoric electronic coupling D–A pairs contribute to varied properties including wavelength-tunable luminescence,²⁴ multi-photon harvesting,²⁵ and spin qutrit.²⁶ Nevertheless, diverse weak interactions between D–A pairs often lead to polymorphism and stoichiomorphs for a given D–A pair.^22,27 For precisely controlling the arrangement of an An dimer and the CT pair, the template effect is crucial. Stoddart utilized the supramolecular engineering strategy to select a porphyrin cage with the pore size of 10.4 Å, in the range of 9.9 to 10.5 Å for hosting two guest molecules at suitable π–π stacking distances (3.2–3.5 Å).²⁸ This molecular container forms a 1 : 2 host–guest complex with perylene to enable red-to-blue triplet–triplet annihilation upconversion (TTA-UC) in solution. Klajn reported a cage with triimidazolylbenzene (TImB) as the four panels to encapsulate the dimer of an An derivative in its cavity, whereas for different An derivatives, the rate for [4+4] cycloaddition was accelerated or hampered by the “topochemical postulate”.²⁹ These studies highlight that the host with a suitable cavity size and the appropriate orientation of An dimer are all important to achieve our goal. Metal–organic frameworks (MOFs) are promising candidates to fulfil these requirements in view of their porous nature, varied topology and designable skeletons.³⁰ Notably, D–A host–guest MOFs with tris(4-pyridyl)triazine (TPT) as the panel ligand³¹ incorporate donors such as An, coronene,³² dibenzothiophene, and N-phenylcarbazole,²¹ demonstrating tunable CT emission and TADF.²¹ NDI is another widely used acceptor in MOF cocrystals.^33–35 For these NDI-based MOF cocrystals, donors with small π-surfaces such as derivatives of benzene and naphthalene were found to be encapsulated efficiently. However, such systems have exclusively featured 1 : 1 guest–cavity ratios, with functionality largely limited to luminescence rather than photoresponsive behavior.

Previously, utilizing NDI-based tetracarboxylic acid ligand H₄BINDI and strontium ion (Sr²⁺), we obtained extended π-stacked assemblies between pyrene (Py) and NDI with the A–D–D–A pattern.³⁶ This represents the first report of the dimer encapsulation of polyaromatic hydrocarbon (PAH) in MOF cocrystals. Later, Zhang utilized the same MOF host Sr-NDI to form cocrystals with 9,10-dimethylanthracene, enabling visible-light triggered ¹O₂ generation for dye degradation.³⁷ Complementarily, covalent-linked NDI-An dyads have demonstrated efficient ¹O₂ production via spin orbit charge transfer ISC.³⁸ Thus, we propose that the MOF cocrystal between Sr-NDI and An might achieve dual photoreactivity: [4+4] photodimerization and [4+2] hetero-Diels–Alder cycloaddition under selective photoexcitation. Furthermore, NDI-based MOFs are established photochromic materials due to the photogenerated NDI radical anion, serving as another possible photoreaction for Sr-NDI cocrystals.^39,40 As shown in Fig. 1, the An dimer is encapsulated in the cavity of Sr-NDI to give the cocrystal Sr-NDI@An. Spectral evidence verified that three photoreactions occurred for this single crystal: UV-triggered [4+4] dimerization of guest An, red-light-induced hetero-Diels–Alder reaction between ¹O₂ and An, and near-infrared (NIR)-driven radical generation. The unprecedented tri-state photoswitching originates from the synergy of CT interaction and dimer confinement. These findings establish a design paradigm for precisely engineering functional π-units, while elucidating fundamental principles for integrating complex functionalities in supramolecular systems.

Fig. 1 Schematic diagram showing the unique stacking patterns of electron acceptor and anthracene dimers for wavelength-dependent photoreactions including UV (365 nm) induced [4+4] cycloaddition, red light-excited (635 nm) hetero-Diels–Alder reaction, and NIR-driven (808 nm) radical generation.

2. Results and discussion

2.1. Synthesis and characterization of MOF cocrystals Sr-NDI@An

Sr-NDI is synthesized via the solvothermal reaction between the metal salts and the four-carboxylate NDI ligand H₄BINDI⁴⁰ whereas the host–guest MOF cocrystal Sr-NDI@An is obtained via addition of anthracene to the reaction mixture before heating (ESI†). As shown in Fig. 2, Sr-NDI is a two-fold interpenetrated MOF with the neighbouring NDI units aligned in an orthogonal manner and a center-to-center distance of 3.48 Å. A cavity of 10.30 Å exists between the ordered arrays of NDI dimers, which allows for the encapsulation of two guest molecules with tight π–π interactions. As expected, single-crystal X-ray diffraction (SCXRD) analysis revealed that two An molecules (Fig. 2) entered the void of Sr-NDI, giving the desirable stacking pattern of A–D–D–A. Specifically, Fig. 2 highlights the intensive π interactions within Sr-NDI@An. Close π–π contacts of 3.39 Å are observed between the ligand NDI and the guest An, suggesting strong charge transfer (CT) interactions. The dihedral angle between the transition dipoles of NDI and An is measured to be 57.5° (Fig. S1, ESI†), further supporting the orbital overlap between the D–A (An–NDI) pair.²³ In the meantime, the two An guests reside in the cavity with a core-to-core distance of 3.49 Å, which is far away from the upper limit (∼4.2 Å) for Schmidt's rule of cycloaddition. The two An aromatic planes are almost perpendicular to each other at an angle of 88.7°, while the two An rings are inclined towards each other at an angle of 12.6°. These observations suggest that the π-orbital overlapping is allowed for the included An dimer, which will contribute to excimer emission and photodimerization.¹³

Fig. 2 Crystal structures of the host Sr-NDI and the MOF cocrystal Sr-NDI@An. The π interactions are highlighted (a) along with the spatial arrangement of the An dimer (b).

Consequently, the distance between NDI dimers is 10.17 Å for Sr-NDI@An, nearly identical to that in Sr-NDI (10.30 Å). This close alignment of PAHs has been extensively utilized to study the smallest dye aggregate, revealing photophysical properties distinct from monomers.⁴¹ However, direct observation of An dimer in the extended crystalline π-stacks has not been reported before. Moreover, the space group of Sr-NDI@An is still tetragonal I41/a, identical to the pristine Sr-NDI (Table S1, ESI†).³⁶ As exhibited in Fig. S2 (ESI†), the coordination mode of Sr²⁺ remains unchanged upon An inclusion. Besides, the Sr-oxygen bond lengths show fluctuations of less than 0.02 Å, which explains the similar SCXRD parameters between Sr-NDI and Sr-NDI@An (Table S1, ESI†).

Powder X-ray diffraction (PXRD) results of Sr-NDI@An are well matched to the simulated one (Fig. S3, ESI†), indicating the phase purity of the obtained MOF cocrystal. There are no noticeable peak shifts between Sr-NDI@An and Sr-NDI (Fig. S4, ESI†), indicating their similar skeletons. Thermogravimetric analysis (TGA) of Sr-NDI@An (Fig. S5, ESI†) demonstrates this cocrystal is stable up to 520 °C, with a residual skeleton mass fraction of 74%. Compared to the 82% residue for Sr-NDI at 550 °C, this confirms consumption of the encapsulated An dimer between 180 and 500 °C, consistent with anthraquinone-incorporated Sr-NDI.^{42 1}H NMR analysis of acid-digested MOF cocrystals (see the ESI† for details) indicates that the cavity loading ratio for Sr-NDI@An is around 14% (Fig. S6, ESI†). The above-mentioned data confirmed the successful encapsulation of the An dimer in Sr-NDI, for which the unique DA stack with close π-interactions are promising for the [4+4] cycloaddition between An dimers and the hetero Diels–Alder reaction involving singlet oxygen (¹O₂).

2.2. Photophysical properties of Sr-NDI@An

The encapsulation of the An dimer with Sr-NDI endows the host with distinct photophysical properties (Fig. 3). Initially, Sr-NDI is a colourless block crystal, which turns to light-purple diamond-like crystal after An incorporation (Fig. S7, ESI†). This colour darkening is assignable to the CT interaction between the electron-deficient NDI ligand and the electron-rich An guest.³³ Accordingly, the solid-state UV-vis absorption spectrum of Sr-NDI@An demonstrates that except for the absorption peak of the NDI core at approximately 400 nm, a relatively wide absorption band emerges, centred at 605 nm (Fig. 3a). Notably, the absorption tail of Sr-NDI@An extends to nearly 830 nm, suggesting photoactivity spanning UV to near-infrared (NIR) regions—enabling wavelength-dependent photoresponses. NDI-based MOFs are established semiconductors,³⁵ and thus we calculate the bandgap value of Sr-NDI@An using the absorption spectrum.⁴³ As shown in Fig. 3b, the CT interaction and the continuous π stack within Sr-NDI@An lead to the decrease of the bandgap from 2.92 eV to 1.72 eV. This result reveals that the electron transfer and communication is largely promoted in the Sr-NDI@An cocrystal than in the empty host Sr-NDI, which is beneficial for applications such as photocatalysis and photovoltaic device.^44,45 In fact, there is facile electron transfer between NDI and the guest An. The fluorescence intensity of An in DMF undergoes a consecutive decrease to nearly zero upon the titration of H₄BINDI (Fig. 3c), originating from the electron transfer from excited An to H₄BINDI.⁴⁶ For Sr-NDI@An, no detectable fluorescence is observed in the solid state, assignable to the generation of radical ion pairs in the presence of light.³³ In this case, Sr-NDI@An is not suitable for luminescence-related applications but might be ideal for light-to-chemical/heat conversion. In contrast, the DMF suspension of Sr-NDI@An (50 μg mL⁻¹) exhibits a broad featureless emission spectrum with a small emission peak at 528 nm (Fig. 3d), which is a characteristic peak rarely detected for the An dimer.⁴⁷ These results collectively confirm dimeric An incorporation. Electrochemical impedance spectroscopy (EIS) indicates significantly reduced charge-transfer resistance in Sr-NDI@An versus Sr-NDI (Fig. 3e), suggesting that Sr-NDI@An is a better candidate for charge transportation in photocatalysis or optoelectronics than the original host.^36,48 The bigger photocurrent of Sr-NDI@An than Sr-NDI further verifies that more pairs of electron and hole are generated upon 450 nm excitation in cocrystals.

Fig. 3 Photophysical properties of Sr-NDI@An. (a) Solid-state UV-vis absorption spectra of Sr-NDI and Sr-NDI@An. (b) Calculated energy gaps of Sr-NDI and Sr-NDI@An. (c) Fluorescence change of An (30 μM) upon the titration of H₄BINDI in DMF solution. (d) Fluorescence spectrum of the DMF suspension of Sr-NDI@An. (e) Electrochemical impedance spectroscopy (EIS) of Sr-NDI and Sr-NDI@An. (f) Photocurrent of Sr-NDI and Sr-NDI@An under 450 nm irradiation.

2.3. Wavelength-dependent tri-state photoswitching of Sr-NDI@An

The photodimerization of anthracene derivatives is a classical photoreaction with a strict requirement for the alignment of the anthracene core.^12,49 In the case of Sr-NDI@An, the guest anthracene adopts a dimeric configuration with the core-to-core distance of 3.49 Å, enabling [4+4] cycloaddition.¹⁶ Upon UV irradiation (365 nm, 20 min), acid-digested Sr-NDI@An exhibits a new ¹H NMR peak at 4.65 ppm (Fig. S8, ESI†), confirming dianthracene formation.⁴⁷ This new peak undergoes a continuous growth along with extending the UV irradiation time, further verifying the UV-enabled photodimerization of anthracene inside Sr-NDI@An (Fig. 4a). For the first two hours, the product yield for dianthracene is 33%, which reaches 64% at the fourth hour. However, after 6 hours, the [4+4] cycloaddition yield plateaus at 74%, indicating that the reaction reaches equilibrium where a maximum of 74% of incorporated An guests undergo cycloaddition. More interestingly, for the 4 hour UV-irradiated Sr-NDI@An, heating at 120 °C for two hours leads to the consumption of the dianthracene (Fig. S9, ESI†). This result is consistent with the reversible nature of the [4+4] cycloaddition of An, verifying that the reversibility is conserved in the confined space of Sr-NDI. Although the [4+4] cycloaddition of An is an old photoreaction which has been extensively investigated even in single crystals,^27,50 this cycloaddition within cocrystal Sr-NDI@An is still the first example for An as the guest in MOF pores or in cocrystals with clear structure information between An dimers. This is because the dimer encapsulation of An is rarely reported in MOF or cocrystals, and the few examples could only obtain the relative positions between An guests from indirect evidence.^13,14 We have tried to obtain the SCXRD of UV-irradiated Sr-NDI@An whereas the crystal quality is too poor to reveal the generation of dianthracene. This might be ascribed to the photodimerization-induced crystal lattice change.⁵⁰

Fig. 4 Wavelength-dependent photoresponsive behaviors of Sr-NDI@An. (a) ¹H NMR monitoring of the [4+4] cycloaddition of Sr-NDI@An with UV irradiation (365 nm). (b) Absorption change of the diphenylisobenzofuran (DPBF) in the presence of Sr-NDI@An and red excitation (635 nm). The inset is the fitting line for the generation rate of singlet oxygen. (c) Electron paramagnetic resonance (EPR) spectra of Sr-NDI@An before and after NIR illumination (808 nm) as well as the following treatments. (d) PXRD patterns of Sr-NDI@An under different treatments. (e) The possible energy diagram for three photoreactions of Sr-NDI@An, UV induced [4+4] cycloaddition, red light-excited hetero-Diels–Alder reaction, and NIR-driven radical generation. S₁ represents the lowest-lying singlet excited state. ¹CT is the singlet excited charge transfer state, while ³CT is the triplet charge transfer state. ISC represents intersystem crossing. CS is short for charge separated state.

Additionally, CT interactions between NDI and An endow Sr-NDI@An with red-light photoresponsiveness. Under 635 nm excitation, the ¹H NMR of Sr-NDI@An exhibits a distinct peak at 8.23 ppm (Fig. S10, ESI†), which should be the signal for the epoxide of An after the hetero-Diels–Alder reaction between singlet oxygen and An.¹⁸ Prolonged red-light irradiation progressively intensifies this peak while diminishing the signal at 7.46 ppm (Fig. S11, ESI†), confirming An consumption. This [4+2] cycloaddition involving anthracene typically relies on additional triplet photosensitizer because ¹O₂ is the crucial active species.^18,19 For Sr-NDI@An, the red-light enabled generation of ¹O₂ was verified by the obvious absorption decrease of diphenylisobenzofuran (DPBF) during the 635 nm excitation (Fig. 4b) as well as the EPR signal enhancement for the TEMP-trapped ¹O₂ adduct (Fig. S12, ESI†).³⁷ Moreover, the ¹O₂ generation rate of Sr-NDI@An was calculated as 0.022. The mechanism for the production of ¹O₂ by Sr-NDI@An likely stems from the CT interaction between NDI and An, which was reported to facilitate the intersystem crossing (ISC) from the singlet CT state (¹CT) to the triplet CT state (³CT) via the through-space CT (TSCT) or spin–orbit charge transfer (SOCT) mechanism of the D–A system.^21,22,38 For these two mechanisms, the interaction between D–A pairs plays a key role. The dihedral angle of 57.5° between NDI and An in Sr-NDI@An supports the electron communication or orbital coupling between D–A pairs, enabling efficient generation of triplet state.²³ After 6 hours of irradiation, the epoxide yield is calculated as 72%, suggesting efficient ¹O₂ diffusion inside Sr-NDI@An for effective hetero-Diels–Alder reaction (Fig. S11, ESI†). To further verify the importance of ¹O₂ for this Diels–Alder cycloaddition, we conducted the red-light excitation of Sr-NDI@An under N₂. After two hours of 635 nm irradiation, only a trace product of Diels–Alder reaction was detected, which is significantly reduced compared to that under ambient air (Fig. S13, ESI†). These results collectively verify that the triplet state is involved in this [4+2] Diels–Alder cycloaddition.

Interestingly, the compact stacking of An and NDI enables exhibition of near-infrared photoresponsiveness by Sr-NDI@An. After only 10 seconds of 808 nm irradiation (0.4 W cm⁻²), the initial transparent light-purple Sr-NDI@An crystals transform to opaque dark solids. The EPR spectra show a sharp increase of the radical intensity with a g value of 2.002 (Fig. 4c), suggesting that upon NIR excitation, charge separation within NDI and An occurs to generate long-lived radical species.⁵¹ This radical generation was also supported by the weaker EPR signal of the irradiated Sr-NDI@An after 5 days, whereas the small variation between air-exposed and nitrogen-protected samples suggests the high oxygen-tolerance of the generated radical pairs (Fig. 4c). The stability of NDI radical anion of the ligand H₄BINDI has been reported to enhance after incorporation into MOF Sr-NDI, as the NDI units adopt an orthogonal arrangement, suppressing radical quenching via transport mechanism.⁴⁰ In our MOF cocrystal Sr-NDI@An, the nearest NDI distance (3.55 Å) exceeds that in Sr-NDI (3.48 Å). This increased inter-NDI distance, combined with the orthogonal arrangement of NDI units, inhibits radical quenching via the transport mechanism, thereby stabilizing photo-generated NDI radical anions within Sr-NDI@An.²³ Moreover, the oxygen-blocking effect of the crystalline skeleton and the guest-filled pores endows the radical pairs in Sr-NDI@An with high oxygen tolerance.

These distinct photoresponses of Sr-NDI@An to different wavelengths are confirmed to arise without structural degradation, since post-treatment PXRD patterns and TGA curves remain nearly identical to pristine samples (Fig. 4d and Fig. S14, ESI†). This result is consistent with the low loading ratio of An (14%) in Sr-NDI@An and our analysis for the light-induced product. The reason for the three photoreactions of Sr-NDI@An toward distinct light excitation originates from its special A–D–D–A stacking pattern and the different energies of the respective photon (Fig. 4e). UV photon exhibits the highest energy to excite the guest An of Sr-NDI@An to its singlet excited state (S₁), which then reacts with the nearby An to give the product of [4+4] cycloaddition. Prior studies have demonstrated Schmidt's rule for the close alignment of An units for [4+4] cycloaddition.^16,29,52 For Sr-NDI@An, the core-to-core distance between the encapsulated An dimer is 3.49 Å, satisfying the Schmidt's rule. To further emphasize the significance of the dimer encapsulation, we synthesize the control MOF cocrystal Ca-NDI@An, which possesses an isoreticular framework to Sr-NDI@An but accommodates only one An guest in its cavity (Fig. S15, ESI†). No photodimerization product is detected in Ca-NDI@An after 3 hours of UV irradiation (Fig. S16, ESI†). This result provides direct experimental evidence that the dimer arrangement is crucial for [4+4] photodimerization within MOF cavities. In contrast, the red light can only excite the cocrystal Sr-NDI@An to its ¹CT, which decays to the triplet state (³CT) to activate oxygen for producing ¹O₂. Subsequently, the Diels–Alder reaction between ¹O₂ and An is allowed to give the epoxide. Furthermore, the NIR photon could excite the cocrystal Sr-NDI@An to the charge separation (CS) state, leading to the generation of radical ion pairs, which can be utilized as photothermal agents or spin converters.⁵³ Notably, the charge-transfer stacking is essential for red light-driven Diels–Alder reactions and NIR-induced radical generation, given that Sr-NDI exhibits no absorption beyond 600 nm (Fig. 3a). DFT calculations of Sr-NDI@An reveal that the HOMOs are concentrated on the electron-donating An guests while the LUMOs are located on the ligand NDI, verifying the strong CT interaction between NDI and An within Sr-NDI@An (Fig. S17, ESI†). Therefore, synergy between dimer encapsulation and charge-transfer stacking is required for Sr-NDI@An to achieve tri-state photoswitching. This MOF cocrystal represents the first crystalline material integrating [4+4] cycloaddition, [4+2] cycloaddition, and radical generation functionalities. Its multimodal photoreactivity provides a platform to advance stimulus-responsive materials in optical writing, data storage, smart displays, and sophisticated information transfer.

3. Conclusions

In sum, this work establishes an efficient strategy for precise arrangement of electron donors and acceptors in crystalline solids to achieve targeted photoactivities. By employing a metal–organic framework (MOF) with cavity dimensions tailored for π-stacked dimer encapsulation, we demonstrate an anthracene dimer inclusion enabling solid-state [4+4] cycloaddition. In the meantime, the MOF skeleton is electron-deficient NDI, which can form D–A pairs with An to generate ¹O₂ via intersystem crossing from the photoexcited CT state. This reactive oxygen species then reacts with proximate anthracene units to undergo hetero-Diels–Alder reaction. Moreover, the emergent CT band confers the MOF cocrystal NIR photoresponsiveness, yielding stable radical ion pairs. This synergy of CT interactions and dimer encapsulation creates a wavelength-selective nanoreactor—unprecedented in solid-state host–guest systems or stimulus-responsive materials. Given the modularity of MOF architectures and tunable photoactive units, this conceptual advance enables goal-oriented design of functional materials through rational donor/acceptor packing, overcoming current synthetic limitations. Such wavelength-discriminated photoresponsiveness in a singular material platform may pioneer smart materials for product authentication, secure communication, next-generation opto-electronics, and microrobotics for engineering or biomedicine.

Author contributions

Z. Huang conducted the majority of the experiments and ran all the data analyses. J. Wei and X.-Y. Wang obtained the single-crystal data and gave the final crystal structure. Z. Huang prepared the ESI.† X.-Y. Wang and S. Xu contributed to a part of characterization experiments. L. Huang co-wrote the manuscript and gave useful advice. L. Zeng conceived and supervised this project and wrote the manuscript. All authors contributed to the revision of the final manuscript and ESI.†

Conflicts of interest

The authors declare no competing interests.

Data availability

All data pertaining to this manuscript are available within the main text and ESI.† All relevant data underlying the results of this study are available from the corresponding author upon request.

Acknowledgements

The authors gratefully acknowledge funding support from the National Natural Science Foundation of China (NSFC) (No. 22371135), the Natural Science Foundation of Tianjin (No. 24JCQNJC01870), and the Postdoctoral Fellowship Program of CPSF (GZB20230315, 2024T170436 and 2024M751518).

References

1. B. M. Vickerman, E. M. Zywot, T. K. Tarrant and D. S. Lawrence, Nat. Rev. Chem., 2021, 5, 816–834.
2. X. Zhang, S. Rehm, M. M. Safont-Sempere and F. Wurthner, Nat. Chem., 2009, 1, 623–629.
3. C. Ni, D. Chen, Y. Yin, X. Wen, X. Chen, C. Yang, G. Chen, Z. Sun, J. Wen, Y. Jiao, C. Wang, N. Wang, X. Kong, S. Deng, Y. Shen, R. Xiao, X. Jin, J. Li, X. Kong, Q. Zhao and T. Xie, Nature, 2023, 622, 748–753.
4. H. Jeong, W. Park, D. H. Kim and K. Na, Adv. Drug Delivery Rev., 2021, 177, 113954.
5. Y. X. Shi, W. H. Zhang, B. F. Abrahams, P. Braunstein and J. P. Lang, Angew. Chem., Int. Ed., 2019, 58, 9453–9458.
6. R. Haldar, L. Heinke and C. Woll, Adv. Mater., 2020, 32, e1905227.
7. Z. Chang, D.-H. Yang, J. Xu, T.-L. Hu and X.-H. Bu, Adv. Mater., 2015, 27, 5432–5441.
8. H. Xiong, Y. Xu, B. Kim, H. Rha, B. Zhang, M. Li, G.-F. Yang and J. S. Kim, Chemistry, 2023, 9, 29–64.
9. S. Panja and D. J. Adams, Chem. Soc. Rev., 2021, 50, 5165–5200.
10. P. Kissel, D. J. Murray, W. J. Wulftange, V. J. Catalano and B. T. King, Nat. Chem., 2014, 6, 774–778.
11. I. Tahir, E. Ahmed, D. P. Karothu, F. Fsehaye, J. Mahmoud Halabi and P. Naumov, J. Am. Chem. Soc., 2024, 146, 30174–30182.
12. X. Wei, J. Liu, G.-J. Xia, J. Deng, P. Sun, J. J. Chruma, W. Wu, C. Yang, Y.-G. Wang and Z. Huang, Nat. Chem., 2020, 12, 551–559.
13. M. Tu, H. Reinsch, S. Rodríguez-Hermida, R. Verbeke, T. Stassin, W. Egger, M. Dickmann, B. Dieu, J. Hofkens, I. F. J. Vankelecom, N. Stock and R. Ameloot, Angew. Chem., Int. Ed., 2019, 58, 2423–2427.
14. X.-Y. Chen, H. Chen, L. Đorđević, Q.-H. Guo, H. Wu, Y. Wang, L. Zhang, Y. Jiao, K. Cai, H. Chen, C. L. Stern, S. I. Stupp, R. Q. Snurr, D. Shen and J. F. Stoddart, J. Am. Chem. Soc., 2021, 143, 9129–9139.
15. S. Chakraborty, H. P. Q. Nguyen, J. Usuba, J. Y. Choi, Z. Sun, C. Raju, G. Sigelmann, Q. Qiu, S. Cho, S. M. Tenney, K. E. Shulenberger, K. Schmidt-Rohr, J. Park and G. G. D. Han, Chemistry, 2024, 10, 3309–3322.
16. X. D. Huang, X. F. Ma and L. M. Zheng, Angew. Chem., Int. Ed., 2023, 62, e202300088.
17. I. Jahovic, Y. Yang, T. K. Ronson and J. R. Nitschke, Angew. Chem., Int. Ed., 2023, 62, e202309589.
18. Y. Yang, T. K. Ronson, D. Hou, J. Zheng, I. Jahovic, K. H. Luo and J. R. Nitschke, J. Am. Chem. Soc., 2023, 145, 19164–19170.
19. J. Y. Zeng, X. S. Wang, Y. D. Qi, Y. Yu, X. Zeng and X. Z. Zhang, Angew. Chem., Int. Ed., 2019, 58, 5692–5696.
20. Z. Zhang, L. Ma, F. Fang, Y. Hou, C. Lu, C. Mu, Y. Zhang, H. Liu, K. Gao, M. Wang, Z. Zhang, X. Li and M. Zhang, JACS Au, 2022, 2, 1479–1487.
21. B. Zhang, J. Xu, C. T. Li, H. L. Huang, M. X. Chen, M. H. Yu, Z. Chang and X. H. Bu, Angew. Chem., Int. Ed., 2023, 62, e202303262.
22. D. Barman, M. Annadhasan, A. P. Bidkar, P. Rajamalli, D. Barman, S. S. Ghosh, R. Chandrasekar and P. K. Iyer, Nat. Commun., 2023, 14, 6648.
23. N. Sikdar, A. Hazra, D. Samanta, R. Haldar and T. K. Maji, Angew. Chem., Int. Ed., 2020, 59, 18479–18484.
24. Q. Liao, Q. Li and Z. Li, Adv. Mater., 2023, 2306617.
25. Y. Wang, H. Wu, P. Li, S. Chen, L. O. Jones, M. A. Mosquera, L. Zhang, K. Cai, H. Chen, X.-Y. Chen, C. L. Stern, M. R. Wasielewski, M. A. Ratner, G. C. Schatz and J. F. Stoddart, Nat. Commun., 2020, 11, 4633.
26. J. R. Palmer, M. L. Williams, R. M. Young, K. R. Peinkofer, B. T. Phelan, M. D. Krzyaniak and M. R. Wasielewski, J. Am. Chem. Soc., 2024, 146, 1089–1099.
27. Akhtaruzzaman, S. Khan, B. Dutta, T. S. Kannan, G. Kumar Kole and M. Hedayetullah Mir, Coord. Chem. Rev., 2023, 483, 215095.
28. H. Chen, I. Roy, M. S. Myong, J. S. W. Seale, K. Cai, Y. Jiao, W. Liu, B. Song, L. Zhang, X. Zhao, Y. Feng, F. Liu, R. M. Young, M. R. Wasielewski and J. F. Stoddart, J. Am. Chem. Soc., 2023, 145, 10061–10070.
29. J. Gemen, M. J. Bialek, M. Kazes, L. J. W. Shimon, M. Feller, S. N. Semenov, Y. Diskin-Posner, D. Oron and R. Klajn, Chemistry, 2022, 8, 2362–2379.
30. X. T. Liu, K. Wang, Z. Chang, Y. H. Zhang, J. Xu, Y. S. Zhao and X. H. Bu, Angew. Chem., Int. Ed., 2019, 58, 13890–13896.
31. M.-H. Yu, X.-T. Liu, B. Space, Z. Chang and X.-H. Bu, Coord. Chem. Rev., 2021, 427, 213518.
32. D. S. Zhang, Q. Gao, Z. Chang, X. T. Liu, B. Zhao, Z. H. Xuan, T. L. Hu, Y. H. Zhang, J. Zhu and X. H. Bu, Adv. Mater., 2018, 30, e1804715.
33. Y. Takashima, V. M. Martinez, S. Furukawa, M. Kondo, S. Shimomura, H. Uehara, M. Nakahama, K. Sugimoto and S. Kitagawa, Nat. Commun., 2011, 2, 168.
34. W. B. Li, Y. Wu, X. F. Zhong, X. H. Chen, G. Liang, J. W. Ye, Z. W. Mo and X. M. Chen, Angew. Chem., Int. Ed., 2023, 62, e202303500.
35. Y. Zhou and L. Han, Coord. Chem. Rev., 2021, 430, 213665.
36. L. Zeng, T. Zhang, R. Liu, W. Tian, K. Wu, J. Zhu, Z. Wang, C. He, J. Feng, X. Guo, A. I. Douka and C. Duan, Nat. Commun., 2023, 14, 4002.
37. K. Liu, L. Wang, S. Li, H. Liu, D. Zhang, M. Jiang, W. Chen, F. Jiao, X. Zhang and W. Hu, Adv. Funct. Mater., 2023, 33, 2306871.
38. X. Chen, A. A. Sukhanov, M. Taddei, B. Dick, J. Zhao, V. K. Voronkova and M. Di Donato, J. Phys. Chem. Lett., 2022, 13, 8740–8748.
39. M. H. You, M. H. Li, H. H. Li, Y. Chen and M. J. Lin, Dalton Trans., 2019, 48, 17381–17387.
40. B. Garai, A. Mallick and R. Banerjee, Chem. Sci., 2016, 7, 2195–2200.
41. D. Bialas, E. Kirchner, M. I. S. Röhr and F. Würthner, J. Am. Chem. Soc., 2021, 143, 4500–4518.
42. Z. Wang, L. Zeng, C. He and C. Duan, ACS Appl. Mater. Interfaces, 2022, 14, 7980–7989.
43. Z. Li, J. Jiao, W. Fu, K. Gao, X. Peng, Z. Wang, H. Zhuo, C. Yang, M. Yang, G. Chang, L. Yang, X. Zheng, Y. Yan, F. Chen, M. Zhang, Z. Meng and X. Shang, Angew. Chem., Int. Ed., 2024, 63, e202412977.
44. Q. Huang, X. Ye, W. Chen, X. Song, Y.-T. Chen, X. Wen, M. Zhang, Y. Wang, S.-L. Chen, L. Dang and M.-D. Li, ACS Energy Lett., 2023, 8, 4179–4185.
45. T. Banerjee, F. Podjaski, J. Kröger, B. P. Biswal and B. V. Lotsch, Nat. Rev. Mater., 2020, 6, 168–190.
46. M. Gutierrez, Y. Zhang and J. C. Tan, Chem. Rev., 2022, 122, 10438–10483.
47. A. Das, A. Danao, S. Banerjee, A. M. Raj, G. Sharma, R. Prabhakar, V. Srinivasan, V. Ramamurthy and P. Sen, J. Am. Chem. Soc., 2021, 143, 2025–2036.
48. A. M. Beiler, B. D. McCarthy, B. A. Johnson and S. Ott, Nat. Commun., 2020, 11, 5719.
49. Y. J. Ma, G. Xiao, X. Fang, T. Chen and D. Yan, Angew. Chem., Int. Ed., 2023, 62, e202217054.
50. S.-L. Huang, T. S. A. Hor and G.-X. Jin, Coord. Chem. Rev., 2017, 346, 112–122.
51. B. Kim, J. Lee, Y. P. Chen, X. Q. Wu, J. Kang, H. Jeong, S. E. Bae, J. R. Li, J. Sung and J. Park, Sci. Adv., 2022, 8, eade1383.
52. M. Tu, H. Reinsch, S. Rodriguez-Hermida, R. Verbeke, T. Stassin, W. Egger, M. Dickmann, B. Dieu, J. Hofkens, I. F. J. Vankelecom, N. Stock and R. Ameloot, Angew. Chem., Int. Ed., 2019, 58, 2423–2427.
53. X. Chen, H. Xie, E. R. Lorenzo, C. J. Zeman, Y. Qi, Z. H. Syed, A. E. B. S. Stone, Y. Wang, S. Goswami, P. Li, T. Islamoglu, E. A. Weiss, J. T. Hupp, G. C. Schatz, M. R. Wasielewski and O. K. Farha, J. Am. Chem. Soc., 2022, 144, 2685–2693.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2418888 and 2419043. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5mh01024a

---