Simple Vanilla Derivatives for Long-Lived Room-Temperature Polymer Phosphorescence as Invisible Security Inks

Developing novel long-lived room-temperature polymer phosphorescence (RTPP) materials could significantly expand their application scope. Herein, a series of RTPP materials based on eight simple vanilla derivatives for security ink application are reported. Attributed to strong mutual hydrogen bonding with polyvinyl alcohol (PVA) matrix, vanilla-doped PVA films exhibit ultralong phosphorescence emission under ambient conditions observed by naked eyes, where methyl vanillate shows the longest emission time up to 7 s. Impressively, when vanilla-doped PVA materials are utilized as invisible security inks, and the inks not only present excellent luminescent emission stability under ambient conditions but also maintain perfect reversibility between room temperature and 65°C for multiple cycles. Owing to the unique RTPP performance, an advanced anticounterfeiting data encoding/reading strategy based on handwriting technology and complex pattern steganography is developed.

Conventional single component RTP emission has normally an emission lifetime over 100 ms and quantum yield (Φp) below 5% under ambient conditions [33,34]. Some recently discovered long-lived RTP systems show Φp of up to 31.2% (2,4,6-trimethoxy-1,3,5-triazine powder) [35] and a lifetime of more than 1,360 ms (isophthalic acid structure) [21,23]. Studies have also been conducted to realize persistent RTP through single component organic materials and doped systems, which include a persistent luminescence material based on the organic photo-induced charge separation system [36], amide derivatives with high quantum yield and long lifetime [33], visible-light-excited organic green phosphorescence emission system [37], and single-crystal emission system from blue to red based on the diphenylsulfone core functionalized with phenoxazine and phenothiazine units [38]. Meanwhile, we recently reported a series of pyrene derivatives doped in polyvinyl alcohol (PVA) matrix to show tunable room-temperature polymer phosphorescence (RTPP) from blue to red [39] and dynamic excitation-dependent polymeric long persistent luminescence systems using polyphosphazenes [40]. Nevertheless, RTPP systems with ultralong lifetime and high quantum yield are still rare.
Although electronic medium has become an indispensable part of our daily life, the paper is still a widely used information storage medium. Therefore, the security requirements of paper information are demanding higher and higher. One of the most popular ways to achieve paper-based secure data recording is to use security ink, where printed information or patterns are visible only under ultraviolet light [41,42]. While a series of photoluminescent material inks such as stimulus-responsive discolored systems, photochromic polymers, metal-free carbon dots, and ion quenched fluorescence switches have been reported [43,44], fluorescent inks are mainly limited to single security with short fluorescence lifetime, which is unable to fulfill higher level security requirements.
Based on the molecular design viewpoint, hydrogen bonding interaction is very important for achieving RTPP by minimizing the nonradiative decay processes [45,46]. Vanilla derivatives can form inter/intramolecular hydrogen bonding in the PVA matrix to suppress nonradiative transition or reduce molecular motions and triplet deactivation process by forming a relatively rigid amorphous environment ( Figure 1). Herein, we tuned methoxy (-OCH 3 ) and carboxylic acid (-COOH) functional groups on vanilla derivatives, where vanilla molecules surrounded by mutual hydrogen bonding interactions in the PVA matrix could lead to suppressed nonradiative transition. To validate this strategy, eight simple vanilla derivatives (M1 to M4 and M1-acid to M4-acid) were doped into the PVA matrix. Obviously, all films (except M2) emit sky blue phosphorescence emission after the removal of 254 nm UV lamp. These films exhibit strong RTPP performance with long lifetime phosphorescence emission, reaching 7 s emission under naked eye observation (Supplementary Figure S1 and Table S1). M4 with the PVA matrix shows the longest phosphorescence lifetime up to 369.8 ms, which is longer than most of the reported RTPP materials. It is noteworthy that the phosphorescent lifetime of M2 with the PVA matrix is 3.9 times higher than its acid form M2-acid in the PVA matrix. These results demonstrate a new method to discover more RTPP materials. Because of the selective emission characteristics of the paper matrix, the visualization for on/off switching of phosphorescence signals could be realized, showing promising potential for information encryption and fingerprint identification.

Results and Discussion
To study the photophysical properties, a series of films were fabricated by a drop-coasting method based on hydrolyzed PVA aqueous solution (30 mg mL -1 ) containing different concentrations of vanilla compounds (0.1, 0.3, 0.5, 1.0, and 3.0 mg mL -1 denote as M-0.1 mg, M-0.3 mg, M-0.5 mg, M-1.0 mg, and M-3.0 mg, respectively). As PVA has a lot of hydroxy groups, these vanilla compounds with hydroxy groups could easily form hydrogen bonding with PVA. The   formed hydrogen bonding interactions provide a relatively rigid environment to suppress nonradiative transition, beneficial to the generation of long-lived phosphorescence emission [18,19]. Thus, hydrogen bonding is the main factor to suppress the total nonradiative decay from triplet excited state based on the intramolecular and intermolecular processes in vanilla-doped PVA films (Figures 2(a) and 2(b) and Supplementary Figure S2).
Among them, M1-M4-doped PVA films have a longer phosphorescence lifetime than that of M-acid-doped ones because of different substitutions. There are two major reasons to consider. (i) Because having more methyl groups would occupy more space and result in the suppression of nonradiative decay [47], M1 shows a longer phosphorescence lifetime than that of M1-acid in the PVA film. The intermolecular motion of the -COOH group is easier than that of -COOCH 3 , which may increase the K nr Phos rate from the triplet excited state in the M-acid series. For example, while the chemical structures of M1 and M1-acid are very similar, the K nr Phos of M1-acid is 3.5 times higher than that of M1 in the PVA film (Supplementary Figure S3). (ii) Energy dissipation in the form of thermal relaxation of the -OCH 3 group is larger than that of the -CH 3 group, which influences the phosphorescence lifetime. For example, M1-doped film (363.8 ms) has a longer phosphorescence lifetime than M2 (282.2 ms), M1-acid-doped film (101.3 ms) has a longer lifetime than M2-acid (72.8 ms), M4-doped film (369.8 ms) has a longer lifetime than M1 (363.8 ms), and M4-acid doped film (105.5 ms) has longer lifetime than M1acid (101.3 ms).
Long-lived phosphorescence decay profiles show monoexponential fitting, indicating that only one triplet emission center exists for these films (Figure 3(a) and Supplementary Figure S10 and S11). Luminescent lifetime and intensity decrease with the increase of temperature from 77 K to 300 K, exhibiting typical RTP emission characteristics (Supplementary Figure S12-S14 and Table S2). For further understanding, the mechanism of the long-lived phosphorescence, the phosphorescence emission lifetime, and quantum yields of these films were recorded under ambient conditions. M4-doped PVA shows the longest phosphorescence lifetime (τ = 369:8 ms) with Φ p of 11.19%, while M2-acid-doped PVA exhibits the shortest phosphorescence lifetime (τ = 72:8 ms) with Φ p of 12.93%. It should be noted that, although M2-doped PVA has lower Φ p than that of M2-acid-doped PVA, the former exhibits 4 folds of lifetime higher than that of the latter. Similarly, M3-doped PVA (τ = 303:8 ms) shows a longer phosphorescence lifetime than that of M3-acid-doped PVA (τ = 223:4 ms), and the phosphorescence lifetime of M4-doped PVA is longer than that of M4-acid-doped PVA. As compared with M4-doped PVA, M1 with one more -OCH 3 group in the PVA matrix presents a longer lifetime, but lower Φ p . The reason may be that the -OCH 3 group rotates more easily than -CH 3 , thus causing higher intermolecular motion. It is very interesting that these pure organic compounds in the PVA matrix show such a long phosphorescence lifetime with high quantum yields, attributed to the strong hydrogen bonding interaction for suppressing nonradiative transition in the doped systems. Taking M1 as an example (Supplementary Figure S2, Figure S3, and Table S1), the K nr phos value of M1 at the crystal state and doped film state is 2258 s -1 and 2.56 s -1 , respectively. For these doped PVA films, the RTPP is dominated by the doping concentration of phosphors. An optimum concentration of vanilla-doped PVA was determined to be 0.3 mg mL -1 of vanilla compounds in the PVA aqueous solution (30 mg mL -1 ). At lower doping concentrations, vanilla compounds could not form enough hydrogen bonding with PVA to suppress the nonradiative transition. When the doping concentration is over 0.5 mg mL -1 , excessive -COOH and/or -OH groups of phosphors cannot fully form hydrogen bonding with the PVA chain, thus increasing the vibration of the PVA system. Therefore, it is generally appreciated that photophysical properties are closely related to the molecular structure of these phosphors.
Interestingly, the phosphorescence property of these vanilla-doped PVA films shows obvious temperature dependence. The major attraction is the significant change of phosphorescence spectra upon naturally cooling from 65°C to room temperature under ambient conditions, where the maximum phosphorescence emission intensity decreases from 3,250 to almost 0 a.u. after 50 min (Figure 3(c)). By heating the samples under the same conditions, obvious phosphorescence intensity could be recovered (Figure 3(d)). In addition, M1-0.3 mg-doped film exhibits almost identical phosphorescence intensity in different environments (i.e., air, argon, and oxygen), revealing the inertness of the triplet excited states to oxygen (Supplementary Figure S15a). However, the phosphorescence emission spectra are sensitive to water molecule (Supplementary Figure S15b). On account of hydrogen bonding interaction in these RTPP systems, the temperature dependence should be caused by moisture in the air, i.e., water could permeate into films to break hydrogen bonding between vanilla molecules and PVA matrix [5]. During the cooling process, hydrogen bonding interactions would be destroyed by the interference of the water molecules, resulting in increased nonradiative transition pathways in doped systems. On the contrary, in the heating process, the evaporation of water molecules from the doping systems leads to the reformation of hydrogen bonding interactions between phosphors and PVA.
To gain a deep insight into the unique phosphorescence, the UV stability experiments of all luminescent films were conducted (Supplementary Figure S16). The maximum emission intensity shows a very little change when the irradiation time increases from 0 to 20 hours, revealing that the UV stability of these long-lived room-temperature polymer phosphorescence films is high. In addition, weak phosphorescence emission of M1-0.3 mg-doped PMMA also proved that a relatively rigid amorphous matrix provided by PVA is necessary for vanilla phosphors (Supplementary Figure S17). Powder XRD studies provide further evidence for the temperature dependence of the films (Supplementary Figure S18). A new diffraction peak at about θ = 22°for M1-0.3 mg PVA film appears under gentle heating (Figure 3(e)), because of the reformation of hydrogen bonding after the water removal. On the contrary, no diffraction peak appears at the same position (Supplementary Figure S18b) during the cooling process. These studies also indicate that more hydrogen bonding interactions are formed with the increase of M1 phosphor concentration in the doped PVA film (Supplementary Figure S18c). Hydrogen bonding interaction between the matrix and phosphors was proven circumstantially by Fourier-transform infrared (FT-IR) spectroscopy. The vibration of the -OH group in pure PVA matrix locates at 3,252 cm -1 , attributed to intermolecular and intramolecular hydrogen bonding interactions (Figure 3(f) and Supplementary Figure S19). The -OH vibration of vanilladoped PVA films shows upward shifts of 6-24 cm -1 (24 cm -1 for the M1-doped PVA film and 6 cm -1 for M2-doped PVA). These results clearly validate that the vanilla  phosphors have a strong ability to form hydrogen bonding interaction with PVA. To further probe into the nature of obvious RTPP properties from these films under ambient conditions, proton nuclear magnetic resonance ( 1 H NMR) spectra were carried out ( Figure 4). Taking M1-acid as an example, its protons are marked as A-D, and -OH of PVA is labelled as E in Figure 4(b). Proton D from -COOH located at 12.5 ppm completely disappears after doping M1-acid with PVA at low concentrations, attributed to D-E type hydrogen bonding. A similar situation occurs to proton A from -OH located at 9.86 ppm, with A-E type hydrogen bonding. Besides, three peaks from isotactic, heterotactic, and syndiotactic structures of the -OH group at 4.3 ppm to 4.6 ppm become a broad peak at 4.5 ppm after forming the M1-acid-3.0 mg PVA film. Upon increasing the doping concentration, protons B and C on the benzene ring gradually appear, along with a red shift. At high concentrations, those initially disappeared proton peaks recover. A possible explanation is that PVA cannot provide enough hydroxy groups to bind with vanilla molecules at high concentrations. 1 H NMR spectra of other doped films show a similar trend (Supplementary Figure S20). These observations prove the formation of strong intermolecular hydrogen bonding between vanilla compounds and PVA matrix.
Unique RTP property and sensitive RTPP character of M1-acid-0.3 mg PVA make it an attractive ink material in advanced anticounterfeiting and information storage since such a long-lived excited state is very difficult to be replicated. The rapid anticounterfeiting processes of phosphorescent ink pen made using M1-acid-0.3 mg PVA are illustrated in Figure 5(a), where the fluorescence and phosphorescence emission on paper could be carefully controlled. As shown from Figure 5(b) and Supplementary Video S1, the "CQUT" letters on the postcard were written just once, and invisible information could be easily read by the naked eye after turning off the UV lamp. Long-lived phosphorescence emission could be recognized more than 3 s after removing the UV lamp, indicating that the RTPP ink can be used to handwrite on the postcard substrate. To further explore more abundant applications, we carried out this on/off visualization with different substrates, including offset papers, parchment, white card papers, kraft papers, and glazed printing papers, marked as I-V, respectively. M1-acid-0.3 mg PVA was used as the RTPP pen ink to paint on these five types of paper substrates (Figure 5(c)   . Surprisingly, the RTPP ink is highly selective to the substrates, showing different RTP intensities and phosphorescence emission time. The longest emission about 5 s is from paper V, and paper I exhibits the shortest RTPP emission (still over 1 s). The emission difference may be due to the different composition and cellulose contents of these papers. The ink on these papers could still be distinguishable regardless of weak blue fluorescence interference from the paper background, showing its high application potential for multiple information anticounterfeiting.
Powder XRD and FT-IR analyses prove different RTPP emissions on these paper substrates, attributed to different hydrogen bonding interactions as previously demonstrated. In powder XRD patterns, the peak of the blank paper V at θ = 28:47°gradually increases to 29:52°after painting with PVA solution, indicating that the hydrogen bonding formation takes effect on paper V. Meanwhile, FT-IR spectra also show the stabilizing process by weakening the vibration of -OH in paper V (Supplementary Figure S21). Offset paper and parchment exhibit a similar stabilization process through hydrogen bonding interaction (Supplementary Figure S22). Atomic force microscopy (AFM) studies also demonstrate no variation of roughness on the surface of encrypted papers (Supplementary Figure S23). The tensile test shows that doped vanilla compounds have a certain effect on the toughness of the films when increasing the doping concentrations (Supplementary Figure S23). The results indicate that the pure PVA film has the highest tensile strength of 25.61 MPa. The tensile strength of the M1-0.3 mg PVA film under the best luminescence intensity is 19.66 MPa, still having a good ductility.
Herein, we show how to use a flat brush to brush the security ink on the papers and refill the ink into a pen. We make exquisite patterns with highly efficient identification and double emission (Figures 5(d) and 5(e) and Supplementary Video S3 and Video S4). The preparation process is very simple and convenient. The results in Figure 5(d) indicate that this kind of long-lived phosphorescence emission ink could be used in a large area, achieving very strong fluorescence and phosphorescence dual emission on the papers. After heating and cooling treatments based on the procedure shown in Figure 5(a) for 50 cycles, excellent luminescence performance could still be maintained, demonstrating the high robustness of the ink. Refilling the RTPP ink into an empty signing pen is highly feasible and reproducible ( Figure 5(e)). We draw a lotus flower picture only once on the white card paper with the RTPP ink pen. What excites us is that the pattern with double emission characteristics drawn by hand is even more exquisite than that drawn by the painting brush, which would lay a foundation for the generalization of the RTPP ink pen in the future.

Conclusion
In conclusion, we have fabricated a series of efficient RTPP systems by doping simple vanilla derivatives into PVA through hydrogen bonding interactions, achieving the maximum phosphorescence lifetime of 369.8 ms and the maximum phosphorescence quantum yield of 14.36%. Because of unique photophysical properties with long-lived phosphorescence emission, these RTPP systems could be employed as security inks. The security inks could be widely utilized on different substrates, with the longest emission over 5 s on the glazed printing paper. The prepared security inks not only maintain excellent luminescent properties under ambient conditions but also exhibit highly identifiable characters after repeated heating/cooling cycles for at least 50 times. Owing to these remarkable features of the RTPP inks, an advanced anticounterfeiting data encoding/reading strategy based on handwriting technology and complex pattern steganography has been proposed for future practical uses.

Preparation of Doped
Matrix. PVA solid (6 g) with hydrolysis degree of 100, 89, and 80 was, respectively, dissolved in deionized water (200 mL) at 95°C for 1 h, which was then filtered to obtain PVA aqueous solution (30 mg mL -1 ) for further use. Poly(methyl methacrylate) solid (3 g) was dissolved in tetrahydrofuran (100 mL) at 65°C for 1 h and then filtered for further use.

Preparation of Vanilla-Doped Films and Phosphorescent
Inks. Firstly, M1, M1-acid, M2, M2-acid, M3, M3-acid, M4, and M4-acid (0.3 mg mL -1 for each molecule) were, respectively, dispersed in eight vials of PVA solutions with the concentration of 30 mg mL -1 , and eight homogeneous solutions were obtained after ultrasonication for 3 h. Secondly, a series of films were fabricated by a drop-coasting method using each aqueous solution (30 mg mL -1 ) containing different concentrations of vanilla compounds. Meanwhile, these eight solutions were, respectively, injected into eight pen cartridges to yield eight security inks. Thirdly, a hand-painted lotus and ink-brushed leaf were prepared by hand painting. Fourthly, after drying in air, the patterns were dried under 65°C in an oven for 1 h.

Data Availability
All data needed to evaluate the conclusions in the paper are present in the paper and the supplementary materials. Additional data related to this paper may be requested from the authors.

Conflicts of Interest
The authors declare no competing financial interests.