Introduction

Developing green materials with certain advantages and functional characteristics using renewable resources is beneficial for achieving sustainable development goals. Wood as a kind of green material, is a widely used porous material with characteristics such as low density, high modulus, high strength, high toughness, and low thermal conductivity1,2,3,4,5. Due to its distinctive microstructure and composition, transparent wood prepared from wood exhibits good optical properties, high strength-to-weight ratio, and hierarchical structure, making it a promising biomaterial-derived material6. It also possesses excellent properties such as high transparency and good mechanical strength7,8, making it applicable in various fields such as optoelectronic devices9,10,11,12, energy-efficient buildings13,14, thermal energy storage15,16,17, solar cells18,19,20, and UV absorption21,22,23.

Room-temperature phosphorescence (RTP) and organic long persistent luminescence (OLPL) materials are a class of luminescent materials that have longer emission lifetimes than fluorescent materials24,25,26. Conventionally, RTP materials with phosphorescence lifetime (τP) longer than 100 ms are considered as room-temperature afterglow materials27,28,29, because the RTP property of such materials can be readily recorded by inexpensive equipment and even distinguished by cameras and human eyes; these luminescent behaviors enable RTP afterglow materials’ broad applications in areas such as oxygen analysis, mapping, anti-counterfeiting, data encryption, background-free bioimaging, and time-gated optical sensing30,31,32,33,34,35,36.

For transparent wood, the involvement of RTP afterglow property would endow the materials with display, esthetics, personalized customization, and science popularization functions. For RTP materials, the transparent wood matrices mean that excitation light can easily enter to excite the luminescent molecules inside, and then penetrate out to show bright afterglow property. Besides, the introduction of wood into RTP material system can reduce the amount of synthetic polymer used. In addition, the involvement of wood component would provide excellent mechanical property for RTP material system6. Moreover, because of the presence of rich hydrogen bonding and relatively high crystallinity of cellulose fibrils in wood, the wood component can suppress oxygen diffusion to protect organic triplet excited states in RTP material system. Therefore, the combination of transparent wood and RTP afterglow would exhibit broad prospects. There are reported stuides on the versatile fabrication of wood-based RTP materials37,38,39,40, but not on RTP transparent wood. It has been reported that RTP transparent wood showed RTP duration around 200 ms in the literature (Supplementary Discussion)41. It is known that materials’ RTP duration can reach about 5–10 folds of their τP32,34,42,43, so RTP duration around 200 ms would refer to a τP of around several tens of milliseconds. To our knowledge, there have been no reports on the preparation of transparent wood as room-temperature organic afterglow materials with τP > 100 ms, for example, τP of 1 s and even 10 s.

In actuality, it is not only difficult to achieve transparent wood of long τP, but it is also challenging in conventional organic systems to obtain high-performance RTP and OLPL materials with high quantum efficiency or long emission lifetime or both in environmental conditions due to the spin-forbidden nature of intersystem crossing and afterglow quenching in organic systems44,45,46,47,48,49,50. To address this issue, groundbreaking research has demonstrated that design strategies based on heavy atom effects, n-π* transitions, molecular crystallization, control of aggregation, and supramolecular assembly can enhance RTP and organic persistent luminescence efficiency31,32,33,34,35. Additionally, our research group has discovered various organic afterglow materials with excellent photophysical properties through dopant-matrix design strategies50,51,52,53,54,55,56,57. In donor-acceptor systems, appropriate combinations of organic molecules can produce OLPL lasting for several hours through photo-induced charge separation and subsequent delayed charge recombination58,59. Recent studies have shown that these RTP and OLPL systems can maintain afterglow characteristics when exposed to environmental conditions or even dispersed in an aqueous medium60,61,62,63,64,65. The dopant-matrix design strategies allow for flexible selection of luminescent dopants and organic matrices to construct RTP and persistent luminescent materials with different structures and compositions66,67,68,69,70. For example, specific polymer matrices provide a rigid environment that inhibits non-radiative decay of the triplet excited state (knr), thus constructing organic RTP and persistent luminescent materials66,71,72. Poly(methyl methacrylate) (PMMA), a low-cost petroleum-based polymer, is widely used as polymer matrix due to its high transparency, low density, low toxicity, excellent durability, and controllability. Various PMMA-based RTP materials with anti-counterfeiting and data encryption functions have been reported66,71,72,73,74,75,76. The dopant-matrix design strategy allows the polymer component to control the excited state property of the luminescent component, resulting in polymer-based organic RTP and afterglow materials with excellent mechanical properties77,78. Solution casting is the most commonly used technique for preparing polymer-based RTP materials77,78,79,80. However, it requires a large amount of solvent, and the evaporation process is slow, making the preparation process time-consuming. On the other hand, melt casting technology is frequently applied in polymer processing as it avoids the use of solvents81, but it requires high temperatures and also leads to energy waste. When preparing polymer-based RTP materials by melt casting, some of the luminescent dopants may undergo decomposition at relatively high temperatures. Solvent-free and time-saving methods under mild condition via photopolymerization have also been developed for the preparation of polymer-based RTP materials82.

On the basis of the above advancements in the fabrication strategies of RTP and organic afterglow materials, particularly those in polymer-based RTP systems, here we design high-performance RTP transparent wood by simultaneously considering the requirements of material fabrication from both aspects of efficient and long-lived RTP and robust transparent wood. In the process of preparing transparent wood, the impregnated resin requires itself to be transparent without visible light absorption, have a refractive index similar to that of wood fibers, and have good compatibility with mother fibers1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,73. Therefore, PMMA is highly regarded as a suitable resin for fabricating transparent wood. We conceive that if the luminescent dopant, monomer, macromolecular matrix framework, and certain initiators can be converted into polymers at an appropriate temperature, the molecular motion of the dispersed luminescent dopants in the formed polymer can be greatly suppressed. This would establish a solvent-free method for preparing polymer-based RTP materials. In addition, the design of luminescent dopants is essential for the preparation of RTP materials with long afterglow duration and high afterglow efficiency. Luminescent difluoroboron β-diketonate (BF2bdk) compounds have been proven to possess interesting optical properties, including large molar absorption coefficients and high photoluminescence quantum yields (PLQY) in both solution and solid states. They also exhibit a wide range of tunable emission spectra with intramolecular charge transfer (ICT) characteristics. BF2bdk molecules have become excellent building blocks in dopant-matrix systems for designing high-performance and functional afterglow materials. Therefore, in this study, we prepared a room-temperature phosphorescent multifunctional transparent wood based on PMMA/wood. The collaboration of wood and PMMA leads to the enhancement of afterglow property. Besides its sustainable nature, the wood component in the present study not only enhances the mechanical property of PMMA/wood-based materials (stronger than PMMA materials) but also serves as oxygen barrier to protect dopant’s triplet excited states from oxygen quenching (unlike PMMA-based materials, the RTP afterglow property of dopant-PMMA/wood materials can be activated by short-term excitation). This interesting combination of polymer, wood, and room-temperature phosphorescence has resulted in long afterglow emission lifetime, excellent functionality, and important practical significance.

Results

Design and synthesis of BF2bdk luminescent dopants for afterglow material fabrication

It is known that materials’ afterglow brightness at time, t, after ceasing excitation light can be expressed as L(t) ~ εΦe−t/τ, where ε, Φ and τ represent molar absorption coefficient, afterglow efficiency and afterglow lifetime, respectively83,84. L(t) increases with ε, Φ and τ. BF2bdk compounds are selected as luminescent dopants to prepare RTP afterglow materials in the present study because of the following reasons. (1) In the aspect of ε, many BF2bdk molecules possess large ε. Here 2 F and 23 F molecules exhibit ICT characters in their S1 states (Fig. 1) and show large ε around 2 ~ 3 × 104 M−1 cm−1 in UVA region (Supplementary Table 1). (2) For intersystem crossing (ISC), it is noted that the multiple electron-donating functional groups on 2 F and 23 F molecules enrich the types of both singlet and triplet excited states. As a result, some triplet excited states would have different excited state natures from S1 states, which would enhance S1-to-Tn ISC process according to the El-Sayed rule. This receives the support from TD-DFT calculation where multiple ISC channels between S1 and Tn states with moderate or large SOCME values (around or >0.5 cm−1) can be found (Fig. 1 and Supplementary Fig. 1–2). Besides, dipole-dipole interaction between ground-state matrix and S1-state BF2bdk also increase ISC tendency in the system (vide infra). High ISC quantum yield represents a prerequisite for high afterglow efficiency. (3) In the aspect of τP, in the absence of heavy atom effect and strong n-π* transition characters, BF2bdk’s T1 states show small phosphorescence rates (kP) and can exhibit long phosphorescence lifetimes (τP) when nonradiative decay and oxygen quenching are sufficiently suppressed under ambient conditions.

Fig. 1: Structures and properties of BF2bdk.
Fig. 1: Structures and properties of BF2bdk.
Full size image

Isosurface maps of electron-hole density difference of the lowest singlet and triplet excited states of 3 F and 23 F obtained by TD-B3LYP/6-31 g(d,p).

Here the BF2bdk compounds (2 F and 23 F) were synthesized using the Suzuki-Miyaura reaction and our cascade reaction; aromatic intermediates were formed by the reaction of bromo tri-phenylbenzene with 4-methoxyphenylboronic acid, followed by stirring with boron trifluoride etherate (BF3·Et2O) and acetic anhydride at 85 °C for 4 h to yield the final BF2bdk compounds (see Methods and Supplementary Methods). To isolate the BF2bdk compounds, column chromatography was performed on silica gel, followed by recrystallization for purification. The purity of the compound was confirmed by HPLC, and their structures were characterized using 1H NMR, 13C NMR, and 19F NMR spectra (Supplementary Fig. 3-14). Detailed information regarding the synthetic procedures, structural characterization, and purity of the BF2bdk compounds can be found in Supplementary Information.

We characterized compounds 3 F and 23 F, and the results showed that both compounds exhibited high-energy absorption bands at 250–300 nm and low-energy absorption bands at 300–400 nm in dichloromethane solution, as well as similar absorption band shapes and absorption maxima in other common organic solvents according to their UV-vis spectra (Supplementary Fig. 15 and Supplementary Table 1). As supported by TD-DFT calculations, which indicated that compounds 3 F and 23 F could be assigned to ICT transitions from the triphenyl to the dioxaborine group (Fig. 1). However, solutions or powders of 3 F and 23 F do not exhibit afterglow under environmental conditions (Supplementary Fig. 16). This can be attributed to two factors: (a) the intersystem crossing tendency is not strong enough to form sufficient triplets in single-component BF2bdk system; (b) the active intramolecular motion (in solution state) and the electronic coupling between neighboring molecules (in powder) contribute the nonradiative decay of triplet excited states and thus quenching room temperature phosphorescence57,83.

Material fabrication and photophysical measurements

We prepare RTP transparent wood according to the procedure as shown in Fig. 2a; transparent wood, which inherently possesses multifunctionality, can also serve as the polymer matrix for BF2bdk (Fig. 2). Based on our previous research, PMMA was used as the matrix and doped with BF2bdk compounds to design high-performance organic afterglow systems, providing beneficial assistance in their applications50,51,52,53,54,55,56,57,71,80,82. We doped compounds 3 F and 23 F to prepare 3F-PMMA and 23F-PMMA samples, and found that the afterglow duration of the two polymer-based materials can last for 8–9 s at room temperature (Fig. 2c, e). And the τP is between 800 and 1000 ms (Fig. 2h, i, l, m). In order to investigated the triplet excited state properties of compounds 3 F and 23 F when doped in transparent wood matrix, we doped the same concentration of BF2bdk into MMA during the preparation of transparent wood, resulting in transparent wood samples with yellow-green afterglow at room temperature (Fig. 2b, d). The steady-state emission spectrum of the 3F-PMMA/wood sample showed fluorescence bands in the range of 385–500 nm, with a maximum emission wavelength of 414 nm (S1 level, 3.33 eV). The delayed emission spectrum (1 ms delay) displayed phosphorescent bands in the range of 500-650 nm (Fig. 2f), with the maximum emission peak at 524 nm (T1 level, 2.84 eV). The PLQY of the 3F-PMMA/wood sample was 29%. The emission decay curve monitored at 524 nm revealed a τP of 1023.9 ms for the solid sample of 3F-PMMA/wood (Fig. 2g). 23F-PMMA/wood samples exhibit typical RTP performance similar to 3F-PMMA/wood. The steady-state emission spectrum of the 23F-PMMA/wood sample showed the maximum emission spectrum at 408 nm (S1 level, 3.39 eV), while the delayed emission spectrum (1 ms delay) displayed phosphorescent bands in the range of 480-600 nm (Fig. 2j), with the maximum emission peak at 515 nm (T1 level, 2.85 eV). The PLQY of the 23F-PMMA/wood sample reached 35%. The τP of the 23F-PMMA/wood sample was 1035.3 ms (Fig. 2k). Compound 23 F in the PMMA/wood matrix exhibited similar afterglow properties to the 3F-PMMA/wood sample, which is consistent with the observed electron-hole diagram of their T1 states (Fig. 1).

Fig. 2: Preparation and photophysical properties of BF2bdk-PMMA and BF2bdk-PMMA/wood system.
Fig. 2: Preparation and photophysical properties of BF2bdk-PMMA and BF2bdk-PMMA/wood system.
Full size image

a Schematic illustration of the preparation of RTP transparent wood; Photographs of (b) 3F-PMMA/wood, (c) 3F-PMMA, (d) 23F-PMMA/wood and (e) 23FPMMA samples at UV-ON and UV-OFF states; room-temperature steady-state and delayed emission spectra of (f) 3F-PMMA/wood, h 3F-PMMA, j 23F-PMMA and (l) 23F-PMMA; room-temperature phosphorescence decay of (g) 3F-PMMA/wood, (i) 3F-PMMA, (k) 23F-PMMA/wood and (m) 23F-PMMA.

To confirm that both 3 F and 23 F exhibit RTP-type afterglow mechanisms, energy transfer, donor-acceptor afterglow, and impurity-induced mechanism, all of which have the potential to induce afterglow property, have been unambiguously ruled out. (1) Room-temperature afterglow can be induced by excited-state energy transfer from RTP donors to emitting acceptors85,86, and the energy transfer process only occurs when the donor is sufficiently excited. The BF2bdk-PMMA/wood afterglow materials obtained in this study can be excited by 365 nm UV light, whereas PMMA/wood matrix shows negligible UV-vis absorption at 365 nm (Supplementary Fig. 17). Besides, PMMA/wood sample (without BF2bdk) has insignificant room-temperature afterglow upon ceasing 365 nm excitation (Supplementary Fig. 18). Therefore, the possibility of room temperature afterglow induced by energy transfer from the PMMA/wood matrix to BF2bdk can be ruled out. (2) Recent studies have shown that when the T1 energy levels of the organic matrix fall between the S1 and T1 states of the dopant, the T1 state of the organic matrix can act as a bridge for the intersystem crossing (ISC) from singlet to triplet states of the dopant, leading to the occurrence of organic room-temperature afterglow87. The T1 energy levels of PMMA and cellulose and hemicellulose of delignified wood have been estimated from their units (Supplementary Fig. 19 and Supplementary Table 2) to be 5.11 eV, 6.67 eV and 6.87 eV by TD-B3LYP/6-31 G(d, p), respectively (it is difficult for us to obtain PMMA and delignified wood’s T1 levels by experiments due to their insignificant phosphorescence). As a result, the ISC-bridged process is not applicable to the BF2bdk-polymer system in this study, as the T1 level of the PMMA/wood matrix is significantly higher than the S1 and T1 states of the BF2bdk molecules (around 3.35 eV and 2.85 eV obtained from fluorescence and phosphorescence maxima). Additionally, the high T1 level of the organic matrix plays a crucial role in preventing afterglow quenching caused by triplet-to-triplet energy transfer from the luminescent dopant to the organic matrix, enabling the construction of high-performance afterglow materials88. (3) In organic materials, the donor-acceptor mechanism can induce charge separation states, which has been observed in some organic systems58,59. Long-lived luminescence observed in organic systems is attributed to retarded charge recombination in rigid solid matrices. For organic donor-acceptor afterglow systems, the formation of charge separation states through intermolecular charge transfer between donor and acceptor molecules is a prerequisite for long-lasting luminescence. However, in this study, negligible intermolecular charge transfer between the BF2bdk dopant and the organic matrix was observed due to the lower HOMO and higher LUMO levels of the latter (the HOMO/LUMO levels of BF2bdk, PMMA unit and delignified wood unit have been summarized in Supplementary Table 2), excluding the possibility of a long-lasting luminescence mechanism in this system. (4) Recent reports suggest that impurities may also be a possible cause of organic room temperature afterglow89. In this study, careful purification of the BF2bdk compound was carried out using column chromatography, followed by three recrystallizations in spectroscopic grade dichloromethane/hexane to ensure its purity. Additionally, HPLC confirmed their high purity (Supplementary Fig. 20). The obtained high-purity BF2bdk powder did not exhibit afterglow properties at room temperature. However, the afterglow material obtained by doping BF2bdk into a host matrix showed significant afterglow properties under ambient conditions. It is worth noting that the afterglow material obtained at low doping concentrations still exhibited high-performance afterglow characteristics. In the low-concentration afterglow material, even if there are possible impurities, the BF2bdk molecules and any potential impurities should be separated by the organic matrix, so that the charge separation and recombination processes between the BF2bdk molecules and any possible impurities can be statistically negligible. We also compare excitation spectra and UV-vis spectra of the system (Supplementary Fig. 15 and 20). The UV-vis spectra show electronic absorption property of luminescent dopants, while the excitation spectra reveal the absorption property of luminescent component in the system. The match between UV-vis and excitation spectra supports that the luminescent property of the present system comes from BF2bdk dopants rather than impurity. These experimental and analytical results presented in this article exclude the possibility that impurities cause the room-temperature afterglow performance.

We investigated the delayed emission at 77 K and found that low temperature prolongs the phosphorescence lifetime, which is a typical RTP mechanism. At 77 K, the 3F-PMMA/wood and 23F-PMMA/wood materials exhibit a peak at about 408 nm in their steady-state emission, with a delayed emission spectrum (1 ms delay) of 507 nm (Fig. 3a–d), consistent with those at room temperature. So, the room-temperature phosphorescence and delayed emission of BF2bdk-PMMA/wood originate from the slow phosphorescence decay of the T1 state of BF2bdk in the PMMA/wood matrix (Fig. 3e). According to previous studies, the polymer matrix can simultaneously lower kF and increase kISC of the ICT state through dipole-dipole interactions, significantly promoting the ISC of the BF2bdk excited state51,52,56. Here we perform photophysical measurements of BF2bdk in polystyrene (PS), poly(methyl methacrylate) (PMMA), phenyl benzoate (PhB) and 4-methoxybenzophenone (MeOBP) (Supplementary Fig. 21 and 22) and calculate the dipole moments of ground-state matrices and S1-state BF2bdk (Table 1) to investigate the dipole-dipole interaction for ISC enhancement. Steady-state emission spectra (Supplementary Fig. 21) exhibit that the fluorescence maxima (λF) of 3F-matrix samples increase with the dipole moments of organic matrices (Table 1). Delayed emission spectra show less significant change of phosphorescence maxima upon the variation of organic matrices (Supplementary Fig. 21). These suggest that the organic matrices interact with S1-state 3 F molecules via dipole-dipole interaction and thus lower 3 F’s S1 level; the dipole moment of 3 F’s S1 state can be calculated by TD-DFT to be 5.97 Debye. 3 F’s T1 level is found to be insensitive to the medium (Table 1), which obey Kasha’s exciton model90. Consequently, the decrease of singlet-triplet splitting energy (ΔEST) leads to ISC enhancement in organic matrices with larger dipole moments. To further investigate ISC behavior, excited state lifetimes of 3 F’s S1 states (τF) have been measured from fluorescence decay profiles (Supplementary Fig. 22). The average τF has been found to first increase with matrix’s ground-state dipole moments from PS (2.0 ns) through PMMA (2.3 ns) to PhB (3.8 ns) and then decrease to 1.9 ns in MeOBP matrix (Supplementary Fig. 22). According to the reported studies on τF of ICT systems51,52,91,92,93, fluorescence decay rate constant (kF) decreases with the polarity of the medium, which can explain τF increase from PS through PMMA to PhB. In view of the equation τF = 1/(kF + kISC + knr) (kISC and knr represent ISC and nonradiative decay rate constants, respectively), the decrease of τF to 1.9 ns suggest significant enhancement of kISC in MeOBP matrix with large dipole moment (Table 1). Given ΦISC = kISC/(kF + kISC + knr), the simultaneous kISC increase and kF decrease upon dipole-dipole interaction between ground-state matrices and S1-state BF2bdk can significantly boost ΦISC in the RTP system. This is our understanding of the role of organic matrix for emergence of organic afterglow in the present system; recent studies based on ultrafast spectroscopy94 support our proposed dipole effect for ISC enhancement. It should be noted that, despite their strong capability in ΦISC enhancement, PhB and MeOBP are small-molecule matrices that are not suitable for fabricating transparent wood.

Fig. 3: Photophysical properties of RTP transparent wood at 77 K and the proposed RTP mechanism.
Fig. 3: Photophysical properties of RTP transparent wood at 77 K and the proposed RTP mechanism.
Full size image

77 K delayed emission spectra of (a) 3F-PMMA/wood and (c) 23F-PMMA/wood; 77 K emission decay profile of (b) 3F-PMMA/wood and (d) 23F-PMMA/wood; (e) Proposed RTP mechanism of BF2bdk-PMMA/wood materials.

Table 1 Photophysical data of 3 F in different organic matrices by experiment and dipole moment of the ground-state organic matrices by TD-DFT calculation
Full size table

In control experiment, we have prepared RTP wood using natural wood as a template and also performed photophysical studies on the obtained RTP wood. The fluorescence and phosphorescence bands of 23F-PMMA/raw wood (RTP wood, not transparent) are similar to those of 23F-PMMA/wood (i.e. the RTP transparent wood). However, the RTP lifetime of 23F-PMMA/raw wood (568.6 ms) has been found to be shorter than that of 23F-PMMA/wood (1035.3 ms) (Supplementary Fig. 23). UV-Vis spectra collected by an integrating sphere system show that the delignified wood (obtained by NaClO2 treatment) and bleached wood (obtained by Hu’s H2O2/UV method) have insignificant or weaker absorption in the visible region. In contrast, raw wood has significant absorption in the 420-700 nm visible region in its UV-vis spectrum (Supplementary Fig. 24); this visible absorption property is attributed to the chromophore groups in lignin. Given the spectral overlap between 23F-PMMA’s phosphorescence band and raw wood’s absorption band in the visible region, the decrease of RTP lifetime of 23F-PMMA/raw wood (568.6 ms, when compared to 23F-PMMA/wood’s RTP lifetime of 1035.3 ms) can be attributed to excited state energy transfer from 23 F’s T1 states to lignin’s chromophore groups in raw wood86. In the RTP transparent wood system where delignified wood is used, such excited state energy transfer should be insignificant, so the organic triplet excited states keep long phosphorescence lifetimes; this is an additional advantage of using transparent wood as organic matrix for preparing RTP materials.

Structure and property of RTP transparent wood

The characterization of the transparent wood was conducted (Fig. 4). In the FT-IR spectra, the characteristic band at 1505 cm−1 corresponded to the aromatic skeleton vibration of lignin, which is attributed to the presence of phenolic aldehyde hydroxyl groups (Fig. 4j)95. However, in the spectrum of PMMA/wood, the characteristic peaks of lignin disappeared, while the carbonyl (C = O) band of PMMA appeared, further confirming the successful combination of the two components. SEM observation shows the filled and unfilled structures of wood in both longitudinal and cross-sectional directions (Fig. 4a–d), ensuring that the wood is well infiltrated, which is consistent with the reported structure of transparent wood. SEM/elemental mapping exhibits the distribution of fluorine element on the cross section of BF2bdk-PMMA/wood RTP materials (fluorine is exclusively from BF2bdk, Fig. 4e–h), from which one can find BF2bdk molecules are well dispersed in the materials. The loading content of BF2bdk has been measured by UV-vis method based on Lambert-Beer law to be ~0.23 wt% in BF2bdk-PMMA/wood materials (PMMA and BF2bdk in BF2bdk-PMMA/wood materials can be readily dissolved in dichloromethane. BF2bdk has significant absorption in UVA region while PMMA doesn’t). We also perform ultrathin cross-sectional TEM observation and XRD measurement to further study the morphology and structure of natural wood and transparent wood. Before TEM analysis, the samples are stained by uranyl acetate and lead citrate. Figure 4l show the ultrastructure of the wood cell wall, similar to the reported studies96. For the sample of transparent wood (Fig. 4m), the framework structure with higher contrast can be assigned to the wood frame, while the component with lower contrast can be attributed to PMMA within the wood frame. In control experiments, BF2bdk-PMMA materials show room-temperature afterglow after sufficient UV irradiation (Fig. 2c, e), whereas BF2bdk-delignified wood samples show very weak afterglow at room temperature (Supplementary Fig. 25). These suggest that the afterglow property of BF2bdk-PMMA/wood materials mainly originates from the BF2bdk molecules in PMMA domains (Fig. 4i). The rigid microenvironment provided by the glassy polymer matrix (PMMA polymers in transparent wood materials have number-averaged molecular weight (Mn) of 226 kg/mol, weight-averaged molecular weight (Mw) of 523 kg/mol, and a polydispersity index (Mw/Mn) of 2.31 (Supplementary Fig. 26), as measured by GPC in tetrahydrofuran phase using polystyrene standard) can restrict the non-radiative deactivation of the BF2bdk triplet state and to some extent protect the triplet state from oxygen quenching. It is known that PMMA matrix allows oxygen permeation74,75,76,77, especially in the film state. Under ambient conditions, the organic afterglow properties of PMMA-based materials typically cannot be activated by short-term and low-power excitation because oxygen in the PMMA matrix can quench the triplet excited states of the luminescent dopants74,75,76,77. After sufficient irradiation, oxygen in the PMMA matrix can be removed and activated, exhibiting obvious room-temperature phosphorescence. Interestingly, when PMMA is combined with wood, RTP of BF2bdk-PMMA/wood materials can be generated under short-term excitation, compensating for the shortcomings of pure PMMA as a matrix. Cellulose and hemicellulose in the wood, which are rich in hydroxyl groups, can form multiple short and strong hydrogen bonds (confirmed by FT-IR as shown in Fig. 4j) to serve as oxygen barriers; cellulose fibrils have long been known as an excellent oxygen barrier because of hydrogen bonding and relatively high crystallinity97. Fig. 4k show that raw wood, delignified wood, and PMMA/wood show characteristic diffraction peaks at 15.8° and 22.3° in the XRD profiles (λ = 0.154 nm, Cu-Kα radiation), which can be assigned to (101) and (002) planes of type I crystalline cellulose, respectively98. The XRD studies indicate that the delignification and transparent wood production processes didn’t disrupt the crystalline structure of cellulose. These results, together with SEM and TEM observations in Fig. 4, indicate that the PMMA micro-domains are encapsulated by the wood frame (Fig. 4i), and wood frame composed of crystalline cellulose with rich hydrogen bonding can serve as oxygen barrier to suppress oxygen quenching of organic triplets in the transparent wood system, thereby improving afterglow performance. Since the PMMA micro-domains are encapsulated by these wood frame, the oxygen diffusion within the PMMA/wood materials would be largely suppressed. Given the knr + kq values at 77 K are negligibly small, experimental kP values can be estimated from τP at 77 K to be 0.57 and 0.46 s−1 for the 3 F and 23 F systems, respectively. According to the equation τP = 1/(kP + knr + kq), experimental knr + kq values in transparent wood at room temperature can be calculated to be as small as 0.41 and 0.51 s−1 for the 3 F and 23 F systems, respectively. The collaboration of PMMA and wood for the protection of organic triplet excited states lays a solid foundation for the development of room-temperature organic afterglow transparent wood as a biomass-based type of material.

Fig. 4: Structure of transparent wood.
Fig. 4: Structure of transparent wood.
Full size image

Longitudinal-sectional SEM images of (a) raw wood and (b) 23F-PMMA/wood; Cross-sectional SEM images of the (c) raw wood and d 23F-PMMA/wood; (e–h) SEM/elemental mapping of 23F-PMMA/wood (carbon, oxygen and fluorine); (i) Scheme of the morphology and structure of BF2bdk-PMMA/wood; (j) FT-IR of raw wood, delignified wood, PMMA and PMMA/wood; (k) XRD of raw wood, delignified wood and PMMA/wood; Ultrathin cross-sectional TEM images of (l) raw wood and (m) 23F-PMMA/wood.

The transmittance of the PMMA/wood and BF2bdk-PMMA/wood composite materials at 550 nm both were 92% (Fig. 5a) and haze value of these exceeded 86% (Fig. 5b). The high transmittance of PMMA/wood ensured the effective collection of light, while the high haze enlarged the irradiation area of light under the same lighting conditions. Compared to glass materials, which have almost no light scattering ability, PMMA/wood was more flexible in the use of light. This capability was meaningful for the construction of indoor lighting buildings, which could achieve a uniform lighting. Particularly, the high haze obscuration effect becomes prominent when the distance between the object and PMMA/wood is >5 cm (Fig. 5c).

Fig. 5: Property of transparent wood.
Fig. 5: Property of transparent wood.
Full size image

a Transmittance and (b) haze of PMMA/wood, PMMA and raw wood; (c) Photographs of a 7 cm × 7 cm × 0.5 mm transparent wood at a height of 5 cm (left), transparent wood closely fitting the pattern next to it (middle) and closely matches the pattern (right).

The mechanical property of the transparent wood has also been tested. The maximum tensile strength and modulus of elasticity (MOE) of the raw wood in the longitudinal direction were 10.53 MPa and 2289.57 MPa, respectively. The PMMA/wood composite materials exhibited values of 64.92 MPa and 4083.16 MPa, respectively (Fig. 6a, b). In PMMA/wood composite materials, the rigid structural lignin is removed, and the presence of hemicellulose is considered crucial for forming a rigid network of interconnected cells. This is due to the high retention of hemicellulose components in the wood matrix, which cover the surface of cellulose microfibers, improving the bonding between fibers and forming a rigid hydrogen bonding network with cellulose and filled PMMA to support the cell wall1,2,3,4,5,6,7,8,9,10,11,12,13,14,15. According to the reported study of compressed wood prepared in the literature6, the tensile stress of transparent wood can be significantly enhanced by increasing the cellulose volume fraction. Here we also increased the cellulose volume fraction from 5%–20% by wood compressing to obtain a tensile stress of 149 MPa (Supplementary Fig. 27). In addition, PMMA/wood had an initial instantaneous water contact angle of 91° and stabilized after 30 s, indicating its good hydrophobicity (Fig. 6c). Therefore, PMMA/wood is a promising material with strong vitality and good industrial prospects in the field of high-performance polymers. We also tested the water absorption thickness expansion rate of PMMA/wood after soaking in water for 24 h, and almost no deformation occurred, indicating that it has good dimensional stability (Supplementary Table 3). Through thermogravimetric analysis (TGA), it can be seen that transparent wood also has good heat resistance compared to raw wood (Supplementary Fig. 28). Finally, we observed the chemical resistance of three groups of transparent wood samples by treating them in a 5% H2SO4 solution, a 4% NaOH solution for 48 h, and a 3% NaCl aqueous solution for 24 h. Due to the acid, alkali, and salt resistance of PMMA itself, which is filled into the wood frame, it has a good protective effect on wood. As a result, transparent wood didn’t undergo any changes, such as foaming or corrosion (Supplementary Fig. 29). To investigate the durability of the materials, corresponding experiments have been performed: (1) the materials are exposed to a xenon lamp with a power density of about 90 mV/cm2 (to simulate sunlight) for 12 h; (2) the materials are placed in a 65% ± 5% humidity box at 20 ± 2 °C for 12 h; (3) to accelerate the experiment of thermal stability, the materials are placed in an 80 °C oven for 1 h. After these treatments, the mechanical property (tensile strength and MOE) and afterglow property (τP) have been found to show only small change (Supplementary Fig. 30), indicating the durability of the materials. Such kind of materials does indeed have excellent durability as reported in the literature18,98,99.

Fig. 6: Mechanical and hydrophobic properties of transparent wood.
Fig. 6: Mechanical and hydrophobic properties of transparent wood.
Full size image

a Tensile stress and (b) MOE of raw wood and PMMA/wood (standard deviation is given); (c) Water contact angle of PMMA/wood at 0 s, 30 s and 60 s.

Universality of the method

In addition to BF2bdk compounds, other luminescent dopants such as coronene (Cor) with a strong ISC tendency and a small kP value100,101,102 have also been introduced to the transparent wood system. Different from the ICT character of BF2bdk system, coronene’s S1 and T1 states are of localized excitation (LE) character. Unlike many aromatic compounds, coronene molecules of D6h symmetry feature symmetry-forbidden S1-to-S0 transition, exhibit fluorescence lifetime (τF) up to around 100 ns in rigid environment, and possess fluorescence decay rate (kF) on the order of only 106 s−1 103. According to the equation ΦISC = kISC/(kF + kISC + knr), the small kF is conducive for coronene system to obtain a high ΦISC, which represents a prerequisite for bright RTP materials. In addition, coronene’s T1-to-S0 transition is spin-forbidden and symmetry-forbidden, so that its kP is as small as ~0.1 s−1 100,101,102. In this study, coronene was used as luminescent dopant dissolved in MMA to prepare room-temperature organic afterglow transparent wood (Fig. 7). The steady-state emission spectrum of Cor-PMMA/wood showed fluorescence bands in the range of 400–500 nm (Fig. 7f), while the delayed emission spectrum displayed characteristic phosphorescent bands of coronene in the range of 500–650 nm, with the maximum emission at 566 nm. Cor-PMMA and Cor-PMMA/wood exhibited the same trend (Fig. 7d), and in the PMMA/wood matrix, the duration of organic afterglow at room temperature is 20 s, which is longer than what was observed in Cor-PMMA (Fig. 7a, b), while the τP is 0.8 s longer (Fig. 7e, g). When coronene’s T1 states are well protected in PMMA/wood matrix, the luminescent system exhibits bright RTP with τP of 4.1 s (Fig. 7g). In addition, we also perform deuteration on coronene molecules. C-D bond vibration is much weaker than that of C-H bond, so that the intramolecular motion of DCor’s T1 states would be suppressed. The DCor-PMMA/wood RTP materials have been found to show brighter afterglow and longer τP up to 11.9 s (Fig. 7c, f, g).

Fig. 7: Photophysical property of coronene transparent wood systems.
Fig. 7: Photophysical property of coronene transparent wood systems.
Full size image

Photographs of (a) Cor-PMMA, (b) Cor-PMMA/wood and (c) DCor-PMMA/wood materials at UV-ON and UV-OFF states; (d) Room temperature steady-state emission spectra and delayed emission spectra (1 ms delay) of Cor-PMMA; (e) The emission decay curves of Cor-PMMA under ambient conditions monitored at 570 nm; (f) Room temperature steady-state emission spectra and delayed emission spectra (1 ms delay) of Cor-PMMA/wood and DCor-PMMA/wood; (g) The emission decay curves of Cor-PMMA/wood and DCor-PMMA/wood materials under ambient conditions monitored at 570 nm.

More wood species have been tried. We use delignified pine, poplar, ash and beech in the place of delignified balsa wood to prepare BF2bdk-PMMA/wood samples. Despite the obtained samples exhibit similar fluorescence and phosphorescence bands to balsa system, the obtained samples show much lower transparency than balsa system (Supplementary Fig. 31). The reduced transparency should be caused by the increased wood density104; pine, poplar, ash and beech have higher density than balsa wood.

Besides Berglund’s pioneering method for conventional preparation of transparent wood6, we also use Hu’s H2O2/UV method1 to remove lignin’s chromophores and then perform infiltration to prepare dopant-PMMA/wood RTP materials (size, 22 cm × 8 cm), which would be time-saving and cost-effective and can reduce the amount of liquid waste. The transparency and afterglow property are similar to that obtained by conventional method (Supplementary Fig. 32). Due to the limited size of vacuum impregnation setup in our institute, currently we are unable to perform large-scale fabrication. Despite of this, we still propose and illustrate the process for large-scale preparing RTP transparent wood (Supplementary Fig. 33). In the future, we will perform large-scale fabrication for applications. In addition, the preparation of thicker transparent wood (2 mm and 3 mm) has also been carried out (Supplementary Fig. 34).

Material functions

In previously reported studies on the preparation of polymer-based afterglow materials, luminescent objects were manufactured using solution-casting techniques, which required the evaporation of solvents and caused significant environmental harm. In this study, we can create organic afterglow materials with various shapes by designing the shape of the wood (Fig. 8a). Inspired by recent studies using the indirect strategy of excited-state energy transfer27,85,86,105, we used Cor-PMMA/wood as the donor and selected rhodamine 6 G (R6G) as the acceptor to prepare three-component Cor-PMMA/wood-R6G samples, which inherited the strong phosphorescence and longer afterglow duration of the Cor-PMMA/wood donor (observable for up to 21 s with human eyes) (Fig. 8b). The steady-state emission spectrum of the Cor-PMMA/wood-R6G three-component sample showed combined emission from Cor-PMMA/wood and R6G, while the delayed emission spectrum displayed emission at 571 nm and 634 nm (Fig. 8e, f). Although the triplet-to-singlet energy transfer from coronene to R6G is spin-forbidden, resulting in a small rate constant (kET), the small kP and long τP of the coronene triplet state provide sufficient opportunity for triplet-to-singlet energy transfer, thereby achieving excited-state energy transfer from Cor-PMMA to R6G fluorescent dye. In addition, RTP transparent wood can be used as display such as directional indications and safety exits (Fig. 8c and Supplementary Fig. 35a). It is well-known that Night Pearl is beautiful and can be used as a decoration. The RTP afterglow transparent wood can be considered as a biomass-based organic Night Pearl. Cartoon patterns have been produced to show its esthetics function (Fig. 8d and Supplementary Fig. 35b, c). Moreover, in terms of personalized customization, a SIOC logo has been made of RTP transparent wood, where SIOC represents Shanghai Institute of Organic Chemistry (Supplementary Fig. 35d). Furthermore, our research institute held a Public Science Popularization Day event in May 2024 (Fig. 8g), and RTP transparent wood has also been used as a display item to stimulate the interest of primary and secondary school students. These proof-of-demonstration studies indicate the combination of transparent wood and room temperature afterglow has broad prospects. In the aspect of RTP materials, transparent wood matrices not only provide the advantages of sustainability but also provide transparent property for RTP emitters, which allow excellent penetration of excitation lights and emission lights for the emergence of bright afterglow for diverse applications.

Fig. 8: Function of RTP transparent wood.
Fig. 8: Function of RTP transparent wood.
Full size image

a Photographs of the plant-shaped assembly afterglow objects of 23F-PMMA/wood materials; (b) Photographs of Cor-PMMA/wood-R6G three-component materials; (c) Left turn pattern of Cor-PMMA/wood afterglow materials; (d) Rabbit patterns of 3F-PMMA/wood afterglow materials; (e) Room temperature steady-state and delayed emission (1 ms delay) spectra of Cor-PMMA/wood-R6G sample; (f) Room temperature afterglow decay of Cor-PMMA/wood-R6G materials monitored at 571 nm; (g) Transparent wood was exhibited at the Public Science Popularization Day held by our research institute; (h) A commercial flat-panel night reading light and (i) a flat afterglow lamp based on RTP transparent wood on a text; (j) A transparent wood-based afterglow panel excited by UV lamp behind the panel and (k) an opaque woo-based afterglow panel excited by UV lamp behind the panel.

Inspired by the popular flat-panel night reading lights of recent years—characterized by their slim, lightweight design, portability, localized illumination that doesn’t disturb others, and energy saving (Fig. 8h)—we developed a flat afterglow lamp based on transparent wood (Fig. 8i). In dark environments, this transparent wood-based afterglow panel can be placed close to text or images of interest, allowing the viewer to see the illuminated content through the panel, a feature that opaque afterglow panels cannot achieve. Additionally, for afterglow panels intended as both energy-efficient and esthetically pleasing lighting, the excitation light source is generally positioned behind the panel. When transparent wood-based afterglow panels are excited from behind, the front surface can serve as a light source (Fig. 8j). In contrast, opaque afterglow panels emit only faint afterglow from the front after excitation, limiting their functionality for practical illumination (Fig. 8k).

Similar to other related materials, the RTP transparent wood is prone to contamination during use. The surface of most of the reported transparent wood has good or moderate hydrophobicity. Supplementary Table 4 summarizes mechanical strength and water contact angle of the reported transparent wood and other related biomass-based materials. Long-alkyl-chain-containing silane has been reported to modify the surface of transparent wood to enhance its hydrophobicity13,106, which can resist the contamination of sewage but still cannot resist oil or fingerprint. Here we add KY-1203 (a kind of UV-curable perfluoropolyether-containing compounds that also allow thermal initiation polymerization, Shin-Etsu Chemical Co. Ltd.) during the preparation process of transparent wood to endow the materials with anti-smudge property (Fig. 9). The KY-1203 modified DCor-PMMA/wood afterglow materials not only exhibit bright RTP afterglow but also display anti-sewage property (Fig. 9a and Supplementary Fig. 36). The 23F-PMMA/wood (without KY-1203) can be easily contaminated by fingerprints (with oil sludge) (Fig. 9b). In contrast, the KY-1203 modified 23F-PMMA/wood RTP materials exhibit fingerprint removability (Fig. 9c, d). The surface of KY-1203 modified Cor-PMMA/wood shows water and oil repellency (a water contact angle of 108° and oleic acid contact angle of 78°) (Fig. 9e–h), which have been rarely reported in the literature of such type of materials. The anti-smudge function of RTP transparent wood would enrich their application scenarios.

Fig. 9: KY−1203 modified RTP transparent wood with anti-smudge property.
Fig. 9: KY−1203 modified RTP transparent wood with anti-smudge property.
Full size image

a Photographs of KY−1203 modified DCor-PMMA/wood afterglow materials exhibiting the capability of repelling sewage (water droplets with black ink was used to mimic sewage); (b) Photographs of 23F-PMMA/wood after being contaminated by fingerprints (with oil sludge); (c) Photographs of KY−1203 modified 23F-PMMA/wood after being contaminated by fingerprints (with oil sludge) and (d) after cleaning up by wiping; (e) Water contact angle of KY−1203 modified 23F-PMMA/wood and (f) KY−1203 modified Cor-PMMA/wood; (g) Oleic acid contact angle of KY-1203 modified 23F-PMMA/wood and h KY-1203 modified Cor-PMMA/wood.

Discussion

We report a preparation method for organic afterglow transparent wood using a thermal initiated polymerization and dopant-matrix design strategy. By using luminescent dopants, monomers, natural wood, and initiators, BF2bdk-PMMA/wood afterglow materials are directly generated through thermal initiated polymerization. The polymer matrix enhances the intersystem crossing of BF2bdk excited states through dipole-dipole interactions with BF2bdk. The polymer matrix also provides a rigid environment for the BF2bdk triplet to suppress its non-radiative decay and oxygen quenching. A comparison with other reported luminescent materials in terms of their afterglow characteristics, mechanical properties, and potential applications has been made in Supplementary Table 5 (the reported polymer-based and biomass-based afterglow materials that didn’t measure mechanical property have not been included in this table). From Supplementary Table 5, one can find that the DCor-PMMA/wood materials in the present study exhibit the longest RTP lifetimes among the reported studies of such kind of afterglow materials. In addition, the KY-1203 fluorinated dopant-PMMA/wood RTP materials, which exhibit water/oil repellency and fingerprint removability, have been rarely reported in the literature. Moreover, the application of the dopant-PMMA/wood RTP materials in science popularization and other fields has also been explored. The RTP afterglow transparent wood we prepared has potential applications in mechanical properties, hydrophobicity, large-scale applications, as well as anti-counterfeiting materials or photonic devices.

We estimate the cost per ton of dopant-PMMA/wood RTP materials, detailed in Supplementary Table 6. BF2bdk-PMMA/wood and Cor-PMMA/wood can be prepared at cost of ¥0.023 ($0.0032) and ¥0.027 ($0.0037) per gram, respectively. The cost of dopant-PMMA/wood RTP materials is higher than traditional building materials such as glass (0.0010 ~ 0.0015 ¥/g) and stone (0.0010 ~ 0.0030 ¥/g), slight higher than PMMA (0.018 ~ 0.020 ¥/g), but much lower than most of other RTP materials (please find the cost of biomass-based RTP materials in the published review article)107. If the RTP transparent wood is solely used as structural material, the RTP transparent wood is indeed significantly more expensive than conventional wood. Besides its mechanical property, the RTP transparent wood in the present study has excellent afterglow property and shows promising application in display and lighting, esthetics, personalized customization, and science popularization. These features increase the value of RTP transparent wood. During the purchasing process, many consumers prioritize cost-effectiveness rather than focusing solely on price. According to cost analysis in Supplementary Table 6 and literature reports, the cost-effectiveness of our RTP transparent wood is higher than that of other biomass-based RTP materials107.

Methods

Physical measurements and instrumentation

Nuclear magnetic resonance (NMR) spectra were recorded on a JEOL Fourier-transform NMR spectrometer (400 MHz), including 1H NMR, 13C NMR, and 19F NMR. UV-Vis absorption spectra were recorded on a Techcomp UV1050 and Shimadzu UV-3600 spectrophotometer. The steady-state and delayed emission spectra were collected by Hitachi FL-4700 fluorescence spectrometer equipped with chopping systems; the delayed emission spectra were obtained with a delay time of ~1 ms. The excited state decay profiles in millisecond to second region were collected by Hitachi FL-4700 fluorescence spectrometer equipped with chopping systems. The fluorescence decay profiles in nanosecond region were recorded by using time-correlated single photon counting technique (TCSPC) on an Edinburgh FLS1000 fluorescence spectrometer equipped with a picosecond pulsed diode laser. Photoluminescence quantum yield was measured by a Hamamatsu absolute PL quantum yield measurement system based on a standard protocol. Photographs and videos were captured by HUAWEI P40 cameras. Before the capture, samples were irradiated by a 365 nm UV lamp (5 W) for ~5 s at a distance of ~15 cm. The morphologies of the wood were observed using the scanning electron microscopy (SEM) instrument of ZEISS Sigma 300 from Germany with a gold spraying process. Transmission electron microscopy (TEM) is used to observe the microstructure of wood by JEOL JEM-F200 from Japan. X-Ray Diffraction (XRD) is used to characterize the crystalline properties of wood by Rigaku SmartLab SE from Japan. The Fourier transform infrared (FT-IR) spectroscopy technique was also used for analysis. The FT-IR profiles were recorded on a Nicolet iS10 system with the scanning range of 4000-600 cm−1 and the resolution is 4.00 cm−1. Thermogravimetric analysis (TGA) was performed to study thermal stability using 209 F3 Tarsus of Germany. Water contact angles were measured using a contact angle measuring instrument (OCA100, Data physics, Germany). The mechanical performance of the materials was tested using the universal testing machine (S2080, Instron, USA). MOE and tensile forces were also studied based on the GB/T 1928-2009 (General Requirements for Physical and Mechanical Tests of Wood) guidelines. The wood physical and mechanical test methods follow general rules, including ISO 3129 (1975, Wood-Sampling Methods and General Requirements for Physical and Mechanical Testing of Small Clear Wood Specimens), which specifies wood sampling methods and general requirements for physical and mechanical tests. These standards are considered the equivalent national standard (NEQ). The water absorption thickness expansion rate of transparent wood was tested according to the national standard (GB/T 15036-2018).

Synthesis of luminescent compounds via cascade reaction

Into a round bottom flask were added 5’-bromo-1,1’:3’,1”-terphenyl (309.2 mg, 1.0 mmol), 2,3-difluoro-4-methoxyphenylboronic acid (257.4 mg, 1.3 mmol), K2CO3 (138.4 mg, 1.0 mmol), Pd(OAc)2 (11.1 mg, 0.05 mmol), ethanol (5 mL) and deionized water (2 mL). The reaction mixture was heated to 85 °C for 12 h. Then, the reaction mixture was diluted by dichloromethane, and K2CO3 precipitates were removed by suction filtration. The obtained crude product was condensed by rotary evaporation, and then purified by column chromatography over silica gel using petroleum ether/dichloromethane (1:3) as eluent to give colorless transparent oil. The colorless transparent oil was then added to a round bottom flask containing acetic anhydride (2.0 mL) and boron trifluoride diethyl etherate (0.9 mL). The reaction mixture was heated to 85 °C for 6 h and cooled to room temperature. The reaction mixture was extracted with dichloromethane and washed with deionized water. The crude product was dried on anhydrous magnesium sulfate, concentrated by rotary evaporation. The crude product purified by column chromatography over silica gel using petroleum ether/dichloromethane (2:3) to give pale yellow 23 F. 23 F was further purified by three cycles of recrystallization in spectroscopic grade dichloromethane/hexane. 1H NMR (400 MHz, chloroform-d, relative to Me4Si/ppm) δ 8.17, 8.15, 7.83, 7.81, 7.76, 7.72, 7.67, 7.66, 7.50, 7.49, 7.47, 7.43, 7.41, 7.39, 7.25, 7.22, 7.20, 6.88, 6.85, 6.83, 6.61, 3.96, 2.44. 13C NMR (125 MHz, chloroform-d) δ 192.33, 147.88, 142.70, 140.39, 130.16, 129.73, 129.01, 128.16, 127.96, 127.36, 126.67, 125.58, 123.81, 108.44, 97.43, 56.74, 24.84. 19F NMR (376 MHz, chloroform-d, 298 K) δ -125.51, -128.02, -145.57. HRMS m/z found (calcd for C29H21O3BF4K+): 543.1172 (543.1177). The full details for the synthesis of compound 3 F and its structural characterization results are available in the Supplementary Information.

Wood delignification

The chromophore groups in lignin can absorb visible lights and reduce light transmittance of the sample. The delignified process, which can remove the chromophore groups, is applied for the preparation of transparent wood, like Berglund’s study6 and many other reported studies. Balsa wood (Ochroma lagopus, moisture content, 5.42%, purchased from Yidimei franchise store, Hangzhou, China) was delignified by using 2 wt % NaClO2 in CH3COOH (pH = 4.6) at 80 °C. The reaction lasted for about 4 h until the wood became totally white. Bleach the delignified wood in H2O2 (30%) solution at 90 °C for 1 h. The treated samples were washed with ultrapure water to remove the chemical residue. Then, delignified balsa wood chips were stored in ethanol and acetone solvents to maintain the rigidity of its cellulose skeleton. In this study, transparent wood with different sizes and thicknesses of 0.5 mm, 1 mm, 2 mm, and 3 mm have been prepared; the size and thickness of transparent wood have been given in the main text and figure captions. It is known that the delignification process would be influenced by the thickness and size of wood6. Here we apply the frequently used conditions for wood delignification and find that most of the lignin can be removed.

Measurement of wood compositions

The compositions of wood were measured according to Van Soest method108 on a ANKOM 220 fiber analyzer. The detailed experimental procedures have been added in the Supplementary Information. The compositions of raw wood, bleached wood (obtained by Hu’s H2O2/UV method1), delignified wood (obtained by NaClO2 treatment), and the delignified plus bleached wood (obtained by NaClO2 treatment, followed by H2O2 soaking) have been measured to be 59.89% cellulose + 26.53% hemicellulose + 11.88% lignin, 58.26% cellulose + 25.15% hemicellulose + 10.01% lignin, 64.80% cellulose + 20.03% hemicellulose + 1.08% lignin, and 67.43% cellulose + 20.31% hemicellulose + 1.05% lignin, respectively (the raw data has been attached in Supplementary Information); in the literature, it has been reported that during the ADL (acid detergent lignin) step of the Van Soest method, a part of lignin would undergo degradation in 72% sulfuric acid, leading to the underestimation of lignin composition and consequently the overestimation of cellulose composition109. The data of wood compositions indicates that most of the lignin in balsa wood can be effectively removed through NaClO2 treatment, and meanwhile some hemicellulose is also removed, aligning with findings reported in the literature110. In contrast, the bleached wood (obtained by Hu’s H2O2/UV method) shows minimal changes or slight decreases in the contents of cellulose, hemicellulose, and lignin. Notably, in the bleached wood (obtained by Hu’s H2O2/UV method), there is a slight reduction in lignin content, which may be attributed to the degradation of chromophores within the lignin, aligning with the UV-vis results of bleached wood (Supplementary Fig. 24, collected by an integrating sphere system).

Preparation of RTP afterglow transparent wood

To obtain RTP afterglow transparent wood, the filling materials were used to fill the delignified wood. 10 mL MMA doped with 0.3 wt% thermal initiator 2,2’-azobis(2-methylpropionitrile) (AIBN) was firstly pre-polymerized at 70 °C and then mixed with 0.3 wt% 3 F or 23 F by stirring under ultrasound until completely dissolved. Next, the dried delignified wood was immersed in the MMA mixture prepolymer solution and then placed in a vacuum drying oven, where the mixture solution was filled into the delignified wood. Finally, the delignified wood filled with MMA mixture solution was assembled between two glass substrates. It was then wrapped in tin foil, placed in an oven, and polymerized at 70 °C for 4 h, leading to RTP afterglow transparent wood. PMMA-based RTP afterglow materials were prepared by directly mixing MMA solution with AIBN and adding 3 F or 23 F compounds. When preparing anti-smudge transparent wood, KY-1203 was first mixed with MMA at KY-1203/MMA volume ratio of 1/25, and then the other preparation steps are the same as above.