Red-light-excited dynamic near-infrared organic afterglow materials for in vivo bioimaging Authors: Lei Zhou, Jiacheng Yang, Zhenyi He, Zhiqin Wu, Ping Jiang, Jinming Song, Liangwei Ma, He Tian, Xiang Ma Published: 2026 ABSTRACT AbstractExtending the excitation and emission wavelengths into the red or even near-infrared region is a highly challenging yet scientifically valuable research topic in the field of organic afterglow materials. To solve this issue, we put forward a twisted intramolecular charge transfer dopant molecule design strategy, in which long-lived electron-deficient dopant is decorated with electron-rich substituent. In this way, the orbital energy level of the dopant can be lowered while maintaining the compatibility with the host. Benefiting from the twisted molecular conformation and small energy gap, the obtained dopant (CN) shows visible-light-excited afterglow with various performance (decay path, lifetime, emission wavelength) when doped into different matrices. Particularly, the maximum excitation wavelength extends to 567 nm and tails to 700 nm when CN is doped into benzophenone matrix. More importantly, the maximum emission wavelength of the afterglow extends to 725 nm (τ = 67.82 ms). We also successfully apply this material in autofluorescence-free bioimaging. This work provides a viable molecular design strategy for developing red-light-excitable near-infrared afterglow materials and demonstrates their potential for in vivo bioimaging.Subject terms: Optical materials and structures, Optical physics Introduction Organic afterglow materials have received widespread attention due to their characteristic long emission lifetimes1–7. The prolonged emission lifetime of these materials adds a time dimension to traditional optical properties such as emission wavelength, intensity, or polarization8–11. This makes them uniquely advantageous for applications in bioimaging, advanced anti-counterfeiting, information encryption, sensing, and other fields compared to traditional fluorescent materials12–16. Although many high-performance organic afterglow materials have been successfully developed in recent years, most of the existing material systems face the issue of short excitation wavelengths (<450 nm), mainly in the ultraviolet region17–23. In contrast, visible light, especially red light, offers significant advantages in reducing phototoxicity, enhancing photostability, and improving penetration depth24–27. Developing visible-light-excited organic afterglow materials is crucial for expanding their practical applications. Particularly in the field of autofluorescence-free bioimaging and biosensing, the ability to be excited by long wavelength light and emission in the near-infrared region are essential requirements. Molecules designed for visible light excitation (absorption) materials often require extending the conjugation of dyes or constructing donor-acceptor (D-A) molecular systems28–32. However, the frontier molecular orbitals of such compounds are typically π-type, making it difficult to achieve sufficient spin-orbit coupling. To promote the utilization of triplet excitons, heavy atoms like bromine and iodine are often introduced into the molecular framework19,33–38. While these heavy atoms enhance the intersystem crossing (ISC) rate, they also increase the radiative transition rate of the lowest triple state (T1), thereby significantly reducing phosphorescence lifetime. The bi-component doping strategy can bypass the ISC process of the dopant molecules by utilizing interactions between the host and dopant to directly generate the triplet excited state of guest, thereby achieving long-lived afterglow39–41. Therefore, this strategy holds unique advantages in developing long-wavelength-excited afterglow materials. However, matching the host and dopant in bi-component doping materials is often based on semi-empirical trial-and-error methods, lacking effective theoretical guidance, which greatly limits the development of novel organic afterglow materials42,43. Recently, a series of red afterglow materials based on pyrene was developed24. Notably, even after modifying with strong electron-donating structures such as methoxybenzene or N, N-dimethylaniline, the obtained derivatives still interact with the matrix to produce afterglow. These results hint that the grafting of electron-donating group does not change the compatibility with the host. This insight inspired us to utilize existing electron-deficient dopants as parent compounds and modify them with electron-rich substituents, such as N, N-dimethylaniline, to construct twisted intramolecular charge transfer (TICT)-type dopants. And if structural modifications maintain the compatibility with the host, this could significantly broaden the excitation and emission wavelengths of the resulting materials, enabling visible-light-excited, or even red-light-excited afterglow materials. In the preliminary studies, we found electron-deficient compound CK could exhibit decent yellow afterglow after doping into benzophenone (BPO) matrix with an 87.25 ms lifetime. To verify our hypothesis, we grafted N, N-dimethylaniline into CK skeleton to construct a TICT-type dopant (CN, Fig. 1a). The results showed that, compound CN exhibited a clear afterglow centered at 725 nm with a 67.82 ms lifetime when doped into BPO matrix. Importantly, the maximum excitation wavelength extends to 567 nm and tails to 700 nm. Benefiting from flexibility of the molecular skeleton, the modulation of the afterglow emission decay path, lifetime, and wavelength could be achieved by changing the matrices. By adjusting the host materials, the afterglow emission wavelength could be tuned from 625 nm to 725 nm (Fig. 1b, c). We also successfully applied this material in autofluorescence-free bioimaging Fig. 1d. This work provides a viable molecular design strategy for developing red-light-excitable near-infrared afterglow materials suitable for in vivo bioimaging. Schematic illustration of constructing red-light-excited near-infrared (NIR) dynamic afterglow materials and in vivo afterglow bioimaging. a Design principles for guest molecules; b Schematic diagram of dynamic and near-infrared afterglow; c Comparison of emission and excitation wavelength of this work with other previous work based on visible-light-excited afterglow materials (Table S1); d Schematic diagram of in vivo afterglow bioimaging Results The photophysical properties in the solution and single crystal state As shown in Fig. S1, a decent emission centered at 575 nm with yellow afterglow was found after doping compound CK into BPO matrix. The afterglow lifetime was measured to be 87.25 ms. After decorating it with N, N-dimethylaniline, the compound was switched from an ICT-type molecule to a TICT-type molecule (CN, Fig. 2a). As shown in Fig. 2b, CN showed a strong peak around 320 nm (assigned to π → π* transition) and relatively weak peak in the range of 400-600 nm (assigned to charge transfer transition) in the absorption spectra after being dissolved in different solvent. Similar to previous reported TICT compound, the CT absorption band exhibited distinct shift (Δλ ≈ 30 nm) with the change of solvent polarity (detailed information in Table S2). And a much more obvious shift in fluorescence spectra was observed (Δλ ≈ 200 nm) with the polarity change (n-Hexane (HEX), toluene (TOL), dioxane (DIOX), dichloromethane (DCM)). Specifically, compound CN exhibited a near-infrared emission centered at 734 nm in DCM as the solvent. Additionally, a weaker emission peak at 584 nm was observed in DCM as the solvent. This dual emission behavior is also a characteristic feature of TICT molecules, with the short-wavelength emission typically attributable to the planar CT state and the long-wavelength emission to the TICT state. Potential energy surface scanning result strongly supported the TICT nature of CN. As shown in Fig. 2c, the dihedral angle between the naphthalene segment (Acceptor) and N, N-dimethylaniline segment (Donor) in the optimized structure of CN tended to 50° or 130°. While the lowest singlet state (S1) tended to take a more twisty conformation with a near 90° dihedral angle between the donor and acceptor. To further investigate the TICT nature of CN, femtosecond transient absorption (TA) measurement was performed under 360 nm excitation in DCM (Fig. 2d). Through global fitting of the TA data, three luminescent species were identified with lifetimes of 20 ps, 216 ps, and 6.36 ns respectively (Fig. 2e). The second and third components (216 ps/6.36 ns) were found to be nearly identical to the fluorescence lifetime at 734 nm measured by time-correlated single-photon counting (TCSPC), and were thus identified as the planar CT and TICT state (Fig. S2). Through TA spectrum, a short-lived excited-state absorption (ESA) band (550–650 nm) rapidly forms under 360 nm excitation, which can be attributed to the Franck-Condon (FC) state. This evolves into a planar CT state ESA band after 20 ps, then the planar CT state transforms into a long-lived TICT state after 216 ps. Evolution-associated difference spectra (EADS) extracted from the global analysis of femtosecond TA data also exhibited this process (Fig. 2e). These findings collectively validate the TICT nature of CN. Moreover, single-crystal data could also confirm the twisted molecule structure. As shown in Fig. 2a, the dihedral angles between the acceptor and donor in CN single crystals grown from acetonitrile (ACN) and trichloromethane-ethanol (TCM-EtOH) were 121.87° and 58.41°, respectively. It can be observed that these two single crystals exhibit a mirror-image relationship, with the difference in dihedral angles primarily arising from molecular asymmetry and consistent with theoretical calculations. The fluorescence emission of the crystals has a significant redshift compared to the solution state, which may be due to generation of aggregates (Fig. 2f). The characterization of solution-state and single-crystal state properties of CN. a Molecule structure of CN (left), single crystal structures (right) of CN obtained from ACN and TCM-EtOH; b Absorption (dotted line) and PL (solid line) spectra of CN in different solvents (λex = 360 nm, C = 1×10-4 mol/L, Inset: photograph of CN in different solvents under UV light); c Potential energy surface scanning of CN with different dihedral angles; d Femtosecond TA spectra of CN in DCM under 360 nm excitation; e Top and bottom panels indicate population dynamics and EADS obtained through global analysis; f Fluorescence spectra of CN powder and single crystals (λex = 360 nm) The photophysical properties of host–guest doping system To verify our design strategy of dopant molecules, CN was doped into BPO matrix (named as BPO@CN) with a 1000:1 doping mass ratio. As we expected, BPO@CN exhibited a distinct delayed emission around 625 nm with a lifetime of 100.32 ms and a photoluminescence quantum yield (PLQY) of 20.7% (Fig. 3a and Fig. S3). A clear orange-red afterglow lasting for more than 1.0 s could be observed by naked eyes after tuning off the 365 nm light (Fig. 3g), and this afterglow remained observable even in bright surroundings (Fig. S4). Compared with parent compound CK, the wavelength of delayed emission red-shifted about 50 nm. Furthermore, benefiting from the flexible molecule structure, the delayed emission wavelength exhibited a significant red-shift with the change of matrix. As shown in Fig. 3b, c, the delayed emission red-shifted to 651 nm, and to 666 nm with red afterglow when the matrix changed to bis(2-diphenylphosphino-phenyl) ether (OBDP) and 4-dimethylaminopyridine (DMAP) (the obtained materials named as OBDP@CN and DMAP@CN, respectively). This distinct shift (Δλ ≈ 41 nm) should be attributed to the flexible molecule structure of compound CN, as evidenced by the absorption spectra. As shown in Fig. S5, BPO@CN exhibited the maximum absorption wavelength at 517 nm and extended to 650 nm. The absorption spectra exhibited a distinct shift with the matrix changed to OBDP and DMAP. Specially, the maximum absorption wavelength was extended to 560 nm and tailed to 750 nm when CN was doped into DMAP matrix (Fig. S5), which implied a red-light excitability. This longer CT absorption wavelength than the solution state indicated a more twisted molecular configuration, which may result from the strong polarity and rigid micro-environment. Moreover, the fluorescence emission also exhibited a similar red-shift tendency. Among them, DMAP@CN showed the reddest fluorescence emission centered at 666 nm (Table 1). Such a redshift in absorption and emission spectra indicated that compound CN may take different molecule conformations in different matrices. As shown in Fig. 3b, when in OBDP matrix, the maximum delayed emission wavelength exhibited distinct redshift compared with the fluorescence emission. The energy difference between the delayed and fluorescence emission of OBDP@CN was calculated to be 0.08 eV, according to the emission spectra. More importantly, the lifetime of the delayed emission showed a power law decay pattern (Fig. S3b), which is the most typical character of long persistent luminescence (LPL)42,44–47. Furthermore, long-lived radical signals were detected in electron paramagnetic resonance (EPR) experiment (Fig. S6). Therefore, the delayed emission of OBDP@CN could be assigned to the long-lived emission from exciplex. Both temperature-dependent spectra and exciton accumulation phenomena can well support this viewpoint (Figs. S7 and 3e). Similarly, power law decay was found by doping CK into OBDP (Fig. S8). When the matrix was changed to BPO or DMAP, the delay emission, with a lifetime of 100.32 ms and 47.82 ms, respectively (Fig. S3), showed a same spectrum with the fluorescence emission, indicating the delayed emission was likely to be from the singlet excited state. More importantly, as the temperature decreasing, the delayed emission intensity kept decreasing, and the temperature-dependent lifetime decay curve exhibited a non-monotonic trend. As the temperature rising, the lifetime first increased and then decreased (Figs. 3d, f and S9). These phenomena indicated that the afterglow of BPO@CN and DMAP@CN was likely be thermally activated delayed fluorescence (TADF) type afterglow48–51. The photophysical properties of these doping materials. a–c Fluorescence (FL) and delayed (DL) spectra of BPO@CN, OBDP@CN and DMAP@CN (doping ratio 1000:1, λex = 360 nm, delayed time = 0.5 ms); d, f Temperature-dependent delayed spectra of BPO@CN and DMAP@CN; e Delayed spectra of OBDP@CN at different photoactivation durations; g Afterglow pictures of BPO@CN, OBDP@CN and DMAP@CN (DMAP@CN excited by 310 nm and others by 365 nm) Table 1The key photophysical parameters of BPO@CN, DMAP@CN, OBDP@CN (doping ratio = 1000:1)λAbs (nm)λEx (nm)λF (nm)λDF (nm)τ (ms)PLQYBPO@CN527560625625100.3220.7%DMAP@CN51763166666647.8228.0%OBDP@CN560544623651/71.9%“/” indicates that the lifetime of the LPL is not calculated The key photophysical parameters of BPO@CN, DMAP@CN, OBDP@CN (doping ratio = 1000:1) “/” indicates that the lifetime of the LPL is not calculated The visible light excitation capability of these doping materials More importantly, the excitation spectra of all materials extended to visible light region. As shown in Fig. 4a, BPO@CN showed a strong excitation peak in the range of 200–400 nm, which is assigned to the excitation band of BPO. While a relatively weak excitation band was also recorded in the range of 400–600 nm, which is assigned to the excitation of CN. Furthermore, excitation-emission dependent spectra further demonstrated the full-spectrum excitation capability of these materials (Fig. S10). However, under long-wavelength excitation, the delayed emission peak exhibited a redshift compared to that under short-wavelength excitation, which may be attributed to the conformational diversity of the CN molecules. Similar phenomena were also observed for OBDP@CN and DMAP@CN, except that OBDP@CN exhibited relatively strong excitation efficiency in the visible light region (Figs. 4b, S11 and S12). Such long excitation wavelength should originate from the decreased energy gap between the HOMO and LUMO of CN by CT effect. Figure 4c showed the afterglow pictures of BPO@CN, DMAP@CN under different excitation wavelengths. It’s worth noting that, DMAP@CN showed a maximum peak around 631 nm in the excitation spectra and tailed to 650 nm, which is the longest excitation wavelength for afterglow materials (Fig. 1c). The clear red afterglow could even be observed under 630 nm flashlight excitation (Fig. 4c). The visible light excitation capability and intrinsic mechanism of dynamic afterglow. a Excitation spectra of BPO@CN; b Emission-Excitation dependent spectra of OBDP@CN; c Afterglow photographs of BPO@CN, OBDP@CN and DMAP@CN under different excitation wavelength; d, e Femtosecond TA spectra of OBDP and OBDP@CN under 360 nm excitation; f Top and bottom spectra indicate population dynamics and EADS of OBDP@CN obtained through global analysis The intrinsic mechanism of dynamic afterglow Exciplex is the most convincible mechanism for such bi-component afterglow materials. According to this mechanism, it is considered that an exciplex will form after excitation52–54. Then, a long-lived emission is generated either from the exciplex or from the triplet excited state of the dopant. The prerequisite for the generation of exciplex is that the energy levels of the HOMO and LUMO of host and guest should match a type II alignment. However, BPO does not meet this prerequisite (Fig. S13). BPO is a famous phosphor with efficient ISC yield. Our previous work indicated that the interaction between BPO and the dopant occurs at the triplet state55. Therefore, the singly occupied molecular orbitals (SOMOs) of the host material BPO are introduced into the frontier orbital energy diagrams of CN to gain further insight. Based on the T1 energy, the SOMO energy level of BPO can be estimated to be about -3.87 eV. The energy levels of HOMO and SOMO of BPO and HOMO and LUMO of CN match a type-II arrangement, which allows the formation of exciplex56. To further elucidate the mechanism of exciplex formation, femtosecond TA spectra were acquired for both host-guest doped materials (BPO@CN, OBDP@CN, DMAP@CN and PMMA@CN) and pure hosts (BPO, OBDP and DMAP) (Figs. 4d–f and S14–S19). Taking the OBDP@CN doping system as an example, as shown in Fig. 4d, e, neither ground-state bleaching (GSB) signals nor stimulated emission (SE) signals were detected in the TA spectra of either OBDP@CN or OBDP. Through global fitting of TA data, a new luminescent species was observed in OBDP@CN. (Fig. 4f). Both OBDP@CN and OBDP exhibited similar ESA signals in the 600-650 nm range immediately after excitation, attributable to the absorption of the OBDP singlet excited state. In OBDP, the signal at 625 nm decayed over 7 ns. However, in OBDP@CN, the 625 nm signal decayed to zero within 10 ps, followed by the emergence of a new long-lived ESA signal at 562 nm. This can be attributed to the signal of the exciplex. This peak is entirely distinct from the signal peak (542 nm) observed in PMMA@CN (Fig. S15). The population dynamics provided a more intuitive representation of the evolutionary process of the luminescent species (Fig. 4f). These findings indicate complex interactions between OBDP and CN, leading to the generation and evolution of new species which exhibit distinct photophysical properties. Owing to the difficulty in producing a sufficiently uniform and transparent film of DMAP@CN, its TA spectral quality was suboptimal. Nevertheless, we can still observe an emission process analogous to that of OBDP@CN (Fig. S16). Slightly different, BPO and BPO@CN appeared to possess similar TA spectra. However, global fitting analysis revealed a new species generated from the BPO triplet state within the BPO@CN, which can be classified as the exciplex (Figs. S17–S19)55. Fig. S20 showed a simplified diagram of the mechanistic process. It could be inferred from the excitation spectra that the excitation of either host or guest could produce exciplex. For BPO@CN and DMAP@CN, the exciton went from the excited state of the exciplex returned to the singlet or triplet of CN, and was finally emitted as TADF. Whereas in the OBDP@CN system, LPL arose from the charge separation and recombination of exciplex. PLQY and lifetime under different excitation wavelengths further substantiated this conjecture, with PLQY and lifetime remained virtually identical regardless of whether long-wavelength or short-wavelength excitation was employed (Figs. S21, S22 and Table S3). CN exhibited different delayed emissions in various matrices, likely due to its flexible molecular skeleton. This flexibility allows CN to adopt twisted but distinct molecular configurations in different environments, as supported by the absorption spectra. Consequently, the delayed emission wavelength of CN varied across matrices. To further validate the mechanism of dynamic afterglow, two methyl groups were introduced to the compound CN to restrain the rotation of N, N-dimethylaniline group (EN). As shown in Fig. S23, N, N-dimethylaniline group is almost orthogonal to the naphthalene segment in the optimum structure. The dihedral angle between them in the ground state was calculated to be about 89.85°. PES scan result showed compound EN tend to take an orthogonal molecular configuration both in S0 and S1 state (Figs. S24 and S25). N, N-dimethylaniline group was almost unable to twist because of the steric hindrance. Absorption spectra of EN in different solvents could also confirmed this point. Similar to CN, EN showed a strong absorption peak around 320 nm (assigned to π → π* transition) in different solvents (Fig. S26). And the absorbance of the CT absorption band (400–600 nm) significantly decreased compared with compound CN, which may indicate the charge transfer between donor and acceptor was blocked. Compound EN showed a very similar delayed emission around 605 nm when doped into BPO, OBDP or DMAP matrix (Fig. S27 and S28). These very similar delayed emission wavelengths testified the variable delay emission of CN resulted from the molecular structural flexibility. As we mentioned above, TADF is the basis for such rich photophysical properties of CN. To further testify it, compounds CN was doped into polymethyl methacrylate (PMMA) matrix. As shown in Fig. S29, the obtained film (PMMA@CN) showed similar delayed emission spectrum (τ = 0.67 ms) with the fluorescence, and both its luminescence intensity and lifetime exhibited a trend of first increasing and then decreasing with rising temperature (Fig. S30), proving the TADF property of CN57–59. The slight redshift of the delayed emission may result from the restraint of the rotation in the rigid micro-environment. These phenomena jointly prove our hypothesis. Photophysical properties of near-infrared afterglow During our research, we found BPO@CN showed an interesting concentration dependence. As shown in Fig. 5a, BPO@CN showed a maximum emission at 625 nm in the fluorescence spectrum when the doping ratio was set as 5000:1. The emission showed obvious increase and maintained in similar shape with the doping ratio raised to 1000:1. A shoulder peak in the range of 700–800 nm was generated when the ratio was further raised to 100:1. And the emission intensity around 625 nm showed slight decrease compared with 1000:1. The shoulder peak evolved to a distinct emission peak at 725 nm with the ratio increased to 10:1. Similar variation, which further confirmed the TADF nature of the delay emission, could also be observed in the delayed emission spectra (Fig. 5b). Particularly, the maximum excitation wavelength extended to 567 nm and tailed to 700 nm (Fig. 5c). The lifetime decay curve reveals that the lifetime at 725 nm (τ = 67.82 ms, Fig. 5d) exhibited two distinct phases: a rapid decay prior to 0.1 s, followed by a slow decay thereafter. Time-resolved spectra further corroborated this lifetime decay behavior at 725 nm (Fig. 5e). It was crucial to emphasize that the long lifetime observed at 725 nm is genuinely present and does not originate from the 625 nm component. The newly emerged emission peak at 725 nm was speculated to be the emission from J-aggregation, which can be evidenced by single crystal data60. Single-crystal stacking structure of CN indicated the formation of J-aggregates (Fig. 5f), which supports our speculation. Similar delayed emission from J-aggregation could also be observed in the DMAP and OBDP matrices (Fig. S31). Photophysical properties of near-infrared afterglow. a FL and b DL spectra of BPO@CN under different doping ratios; c Excitation spectra of BPO@CN under doping ratio 10:1; d Lifetime decay curves and e Time-resolved spectra of BPO@CN (doping ratio 10:1); f Single crystal stacking diagram of ACN and TCM-EtOH In vivo afterglow bioimaging application Afterglow imaging holds potential for eliminating interference from tissue autofluorescence signals in imaging applications due to its long-lived luminescence properties. Consequently, we formulated the near-infrared afterglow material developed in this study into a contrast agent to validate its afterglow imaging capabilities. The contrast agent was administered to mice via oral gavage. This contrast agent comprised BPO@CN, DSPE-mPEG (molecular weight 2000), gum Arabic, and phosphate-buffered saline. DSPE-mPEG serves as a surfactant, whilst gum Arabic is a plant-derived thickening polysaccharide. Using the bioluminescence mode of an in vivo imaging system (IVIS), we imaged the mouse stomach using the afterglow following red light irradiation. As shown in Fig. S34, the contrast agent sustained detectable afterglow signals for over 15 min after light irradiation cessation. Following administration of the contrast agent, the stomach signals in mice were difficult to visualize effectively in fluorescence mode due to interference from the liver, intestine and tissue autofluorescence. However, in bioluminescence mode, obvious signals originating from the stomach were observed using afterglow imaging, and no effective signal was observed for untreated mice (Fig. 6a, b). To further confirm whether the afterglow signals originated from the stomach, we performed dissections on the mice. As shown in the Fig. 6c, d, fluorescence mode revealed distinct signals in the mouse gastrointestinal tract and liver regions, distinguishing these signals from the contrast agent versus tissue autofluorescence was proved challenging, hindering precise pathological information acquisition. In contrast, afterglow imaging generated distinct signals in the stomach under bioluminescence mode, eliminating fluorescence interference and facilitating enhanced pathological assessment within living organisms. This study demonstrates the potential of visible light-excited near-infrared afterglow contrast agents as highly effective in vivo biosensing tools. They significantly optimize response performance while effectively suppressing autofluorescence interference. In vivo afterglow bioimaging application. In vivo fluorescence and afterglow bioimaging of the a untreated and b oral gavage in mice; In vitro (c) and (d) fluorescence and afterglow bioimaging of mice after anatomy (fluorescence imaging signals displayed in ‘YellowHot’ color table, afterglow imaging signals displayed in “Rainbow” color table) Discussion In conclusion, we proposed a molecular design strategy toward visible-light-excited organic afterglow materials by modifying an electron-deficient dopant with electron-donating groups. By doping the obtained TICT-type dopant (CN) into different matrices, a series of red-light-excited NIR afterglow materials was obtained, with various performance (decay path, lifetime emission, emission wavelength). Such rich photophysical property variations were mainly attributed to the twisted molecular structure of CN. It was worth mentioning that a NIR afterglow (τ = 67.82 ms) centered at 725 nm was obtained. Moreover, the excitation wavelength of the afterglow was extended to 650 nm. By the afterglow imaging technology of utilizing these long-wavelength-excitable near-infrared afterglow materials, the critical challenge of tissue autofluorescence interference in conventional fluorescence imaging was successfully resolved. Materials and methods Synthesis The synthesis route to the target compounds and the experimental details on the synthesis are provided in the Supplementary Information. Characterization 1H NMR and 13C NMR spectra were measured on a Bruker AV-600 spectrometer. The electronic spray ionization (ESI) high-resolution mass spectra were tested on a Waters LCT Premier XE spectrometer. The UV–Vis absorption spectra and PL spectra were performed on a Varian Cray 500 spectrophotometer. Fluorescence, phosphorescence, and lifetime of delayed emission spectra were recorded on an Agilent Cary Eclipse spectrophotometer. Fluorescence lifetimes were measured using the TCSPC system of the FLS1000 fluorescence spectrometer from Edinburgh Instruments, in conjunction with the EPL-375 tunable laser. Absolute PL quantum yields were determined with a spectrometer C11347-11 (Hamamatsu, Japan). Femtosecond transient absorption spectra were acquired by using the Time-Tech Spectra TA100 instrument. Afterglow bioimaging was conducted by using the PerkinElmer IVIS Spectrum CT imaging system. The animal procedures were conducted in strict accordance with the Guidelines for Care and Use of Laboratory Animals of East China University of Science and Technology. The compounds and solvents used in the synthesis were purchased from Adamas-beta. Theoretical methods Density functional theory (DFT) and time-dependent (TD-DFT) calculations were performed with the Gaussian 16 (Revision E.01) software package. The ground-state (S0) geometries were optimized with the B3LYP and 6-311G* basis. The excitation energies in the singlet and triplet states were obtained using TD-DFT method based on the optimized S0 molecule structure. Supplementary information Supplementary information.docx Acknowledgements We gratefully acknowledge the financial support from the National Natural Science Foundation of China (22125803, T2488302, 22305080), the Science and Technology Commission of Shanghai Municipality (24DX1400200), the Guangxi Department of Science and Technology (AA23062016), the Fundamental Research Funds for the Central Universities, and the Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China (JYB2025XDXM404). Author contributions L. Zhou, L. Ma, X. Ma, and H. Tian conceived the project, L. Zhou designed the molecules and conducted the experiment. L. Zhou, L. Ma and X. Ma wrote the manuscript. J. Yang, Z. He, Z. Wu, P. Jiang, J. Song discussed the results and commented on the manuscript. Data availability The authors declare that the data supporting the findings of this study are available within the article and its Supplementary Information file. All data are available from the authors upon reasonable request. Conflict of interest The authors declare no competing interests. Contributor Information Liangwei Ma, Email: liangweima@ecust.edu.cn. Xiang Ma, Email: maxiang@ecust.edu.cn. Supplementary information The online version contains supplementary material available at 10.1038/s41377-026-02340-3. References Associated Data Supplementary Materials Supplementary information.docx Data Availability Statement The authors declare that the data supporting the findings of this study are available within the article and its Supplementary Information file. All data are available from the authors upon reasonable request.