Research ArticleChemistry

Engendering persistent organic room temperature phosphorescence by trace ingredient incorporation

See allHide authors and affiliations

Science Advances  07 May 2021:
Vol. 7, no. 19, eabf9668
DOI: 10.1126/sciadv.abf9668


Pure organic persistent room temperature phosphorescence (RTP) has shown great potential in information encryption, optoelectronic devices, and bio-applications. However, trace impurities are generated in synthesis, causing unpredictable effects on the luminescence properties. Here, an impurity is isolated from a pure organic RTP system and structurally characterized that caused an unusual ultralong RTP in matrix even at 0.01 mole percent content. Inspired by this effect, a series of compounds are screened out to form the bicomponent RTP system by the trace ingredient incorporation method. The RTP quantum yields reach as high as 74.2%, and the lifetimes reach up to 430 ms. Flexible application of trace ingredients to construct RTP materials has become an eye-catching strategy with high efficiency, economy, and potential for applications as well as easy preparation.


With unique luminescent properties and low cost, pure organic room temperature phosphorescence (RTP) materials have received broad attention (14). In the past century, researchers believed that it was difficult to use pure organic materials to emit phosphorescence at room temperature because of the difficulty of intersystem crossing (ISC) between singlet and triplet and fast nonradiative transition (5). Recently, pure organic RTP with long lifetime and high efficiency has been realized by several methods, such as crystal or cocrystal packing (69), polymer and host-guest supramolecular systems (1013), and bimolecular charge separation states (14). However, it is still an urgent need for more convenient strategies to construct RTP and a reliable mechanism for afterglow luminescence.

There are reports on the existence of impurities and their impact on some organic RTP systems (15, 16), although the chemical structure and impact mechanism of impurities were not specified because of the difficulties in separation, purification, and structural characterization. Literature shows that Liu and colleagues (17) have recently reported the ultralong RTP of a series of carbazole derivative caused by 1H-benzo[f]indole, an isomer of carbazole, participating in the synthesis process. Because of the low content of 1H-benzo[f]indole in carbazole, its existence has been overlooked in previous reports (18, 19). Because trace impurities may affect the luminescence of materials, organic compounds always need to be carefully purified, despite that few reports focused on impurity research in the RTP system.

In this research, an unusual yellow-green persistent luminescence of 1-(4-bromophenyl)-1H-imidazole (1BBI) powder was observed at room temperature. A trace impurity was observed to play important roles in generating RTP. The impurity was then successfully isolated and identified to be 6-(1H-imidazol-1-yl)-N,N-dimethyl isoquinolin-1-amine (DMIQI). Further studies demonstrated the great influence of trace impurities on RTP emission. The existence of impurities and their impacts on RTP inspired us with an efficient strategy to design a pure organic trace ingredient–mediated bicomponent RTP system. A series of artificially added active ingredients were screened out, working as impurities to engender pure organic RTP. The matrix 1BBI was also extended to a halogen-free imidazole derivative. Bicomponent RTP systems achieved an extra-high quantum yield (QY) up to 74.2%, and their lifetimes varied from 2 to 430 ms. Considering the low content of ingredient (<0.1%) achieving bright RTP, the trace ingredient–mediated bicomponent systems could work as an efficient, economical, and convenient strategy in realizing pure organic RTP of high QY and extra-long lifetime. In addition, the achieved white-light emission showed its excellent application potential as well.


Impurity in RTP organism

The compound 1BBI was synthesized through the substitution reaction of imidazole and 4-bromofluorobenzene followed by purification by flash column chromatography on silica gel. A bright yellow-green afterglow luminescence was observed lasting for more than 2 s. With one more flash column chromatography purification and recrystallization from methanol for several times, the afterglow still existed. However, the emission intensity decreased with more purification, while the emission wavelength and lifetime were kept unchanged (Fig. 1, A and B). Last, 1BBI was purified to almost no photoluminescence after flash column chromatography for twice on silica gel and one-time reversed-phase chromatography on C18-bonded silica gel (Fig. 1D and movie S1). High-performance liquid chromatography (HPLC) analysis showed that the impurity content had been reduced to a very low level after repeated purification (fig. S6). These facts indicated that the RTP in 1BBI was caused by the impurities.

Fig. 1 Phenomena of impurity-induced RTP.

(A) RTP spectra of 1BBI with different purification. Excitation, 365 nm; slit width, 10 nm; voltage, 700 V. a.u., arbitrary units. (B) RTP decay curve of 1BBI with different purification. (C) Structure of DMIQI and 1BBI identified by single-crystal x-ray diffraction. (D) Different luminescent performance between crude and extra-pure 1BBI under 365-nm irradiation. UV, ultraviolet. (E) RTP spectra of 1BBI mixed with various contents of DMIQI. Excitation, 365 nm; slit width, 10 nm; voltage, 650 V. (F) Changes of phosphorescence intensity (value at 550 nm) and lifetime with different molar ratios.

The abnormal RTP could be elucidated only after the structure of the impurity is resolved. About 70-mg impurity was successfully isolated from more than 200-g crude 1BBI products. The nuclear magnetic resonance (NMR) spectra (1H, 13C, 1H-1H correlation spectroscopy (COSY), heteronuclear multiple bond correlation (HMBC), and heteronuclear multiple quantum correlation (HMQC) and high-resolution mass spectrometer (HRMS) helped to infer the structure to be DMIQI. To confirm that structure, we synthesized DMIQI by a different route starting from 6-bromoisoquinolin-1(2H)-one (fig. S4). The NMR and HRMS spectra of synthetic DMIQI agreed with that of isolated DMIQI. The structure of DMIQI was further determined by single-crystal x-ray diffraction and structure analysis (Fig. 1C). To explore the origin of the impurity DMIQI, we placed imidazole, 4-bromofluorobenzene, and 1BBI, respectively, in the same reaction condition [with NaH in N,N′-dimethylformamide (DMF) while heating]. The result showed that no DMIQI was produced, indicating that the impurity was not from the decomposition of reagents or products. The RTP of pure 1BBI would not show up even if placed under ambient conditions for more than 8 months (fig. S7). It was then found that the synthesis of 1BBI using the sublimated purified imidazole could avoid the impurity DMIQI, while the purification of 4-bromofluorobenzene or DMF by vacuum distillation did not affect the harvest of DMIQI. In addition, an intermediate, 4-(1H-imidazol-1-yl)-N,N-dimethylbenzamide, was successfully identified through HPLC-HRMS (fig. S8). In view of this, we proposed that impurities like aminoacetaldehyde in imidazole reacted with the intermediate to produce the new impurity DMIQI (fig. S9).

To demonstrate the influence of trace impurities on RTP emission, we constructed a bicomponent RTP system of DMIQI/1BBI mixture by the solvent evaporation method. Different proportions [0.01 to 5 mole percent (mol %)] of DMIQI in 1BBI were completely dissolved in methanol. Then, the solvent was evaporated under reduced pressure to result in a white crystalline solid. The time-resolved luminescence spectra showed a double-peak emission at 520 and 550 nm (Fig. 1E). With the increase of impurity ratio, the RTP intensity first increased to the highest at 0.5 mol % and then decreased (Fig. 1F), while the RTP lifetime was kept almost unchanged. RTP intensity remained about 30% of the highest even when the proportion dropped to 0.01 mol %. Extra-low level of impurity may also lead to high RTP benefits. All the luminescence decay curves followed the exponential decay. The calculated lifetime of all proportions was 212 ± 14 ms (fig. S10). The pure DMIQI showed a maximum fluorescence emission at 427 nm and undetectable phosphorescence at room temperature. This means that the unusual phosphorescence was generated through the interaction of impurities and matrices.

Design of trace ingredient–mediated bicomponent RTP system

The impurity DMIQI was found by coincidence, but the case of RTP caused by impurities was not uncommon. The 1BBI also showed a strong RTP signal even when synthesized by a completely different route from 4-bromoaniline. The emission spectra and lifetime differed from that of DMIQI/1BBI system (fig. S11). It seemed a feasible strategy to engender effective pure organic RTP through the interaction of trace ingredients and matrices.

It was time consuming and inefficient to separate and identify impurities from synthetic matrix. As an alternative, it was more efficient to find active ingredients to substitute for the impurities to light up RTP in matrix. A total number of six active ingredients (Fig. 2A) were screened out to engender RTP after adding 1BBI. The isomers of matrix were first tested. The 2-(4-bromophenyl)-1H-imidazole (2BBI) and 4-(4-bromophenyl)-1H-imidazole (4BBI) showed bright cyan RTP under 254-nm irradiation when mixed with 1BBI. The RTP QY of 4BBI/1BBI reached as high as 74.2% without deaeration. Another type of active ingredient was N-substituted aminobenzaldehyde derivatives, 4-(dimethylamino)benzaldehyde (DMBA), 4-(pyrrolidin-1-yl)benzaldehyde (PBA), and 9-julolidine carboxaldehyde (JCA). Being added to 1BBI, they could also show obvious cyan RTP peaks at 465, 480, and 495 nm under excitation at 365 nm, respectively. Unexpectedly, the emission could last for more than a second after removing the irradiation light (Fig. 2C). The measured lifetimes were 122, 113, and 143 ms, respectively (Fig. 2B). An analog of DMIQI, 2-(1H-imidazol-1-yl)-N,N-dimethyl quinolin-6-amine (DMQI) was also synthesized to achieve the longest RTP lifetime (τ = 262 ms) of our designed trace ingredient–mediated bicomponent RTP systems, and the RTP emission could reach the yellow region (peak at 560 nm). In general, seven compounds were able to engender RTP in 1BBI matrix. The maximum RTP wavelength ranged from 465 to 560 nm with lifetime ranging from 6.8 to 263 ms. None of these purified ingredients or matrix showed any detectable phosphorescence at room temperature, and their fluorescence has notable differences in terms of wavelength and intensity (figs. S12 to S20).

Fig. 2 Optical characteristics of trace ingredient–mediated bicomponent RTP systems.

(A) Chemical structures of seven selected ingredients. (B) Normalized RTP spectra of trace ingredient–mediated bicomponent RTP systems. (C) RTP decay curve of trace ingredient–mediated bicomponent RTP systems. (D) Afterglow images of trace ingredient–mediated bicomponent RTP systems. The spectra and images of 2BBI/1BBI and 4BBI/1BBI were measured under 254-nm excitation and others under 365 nm.

To explore the structural commonality of active ingredients, we tested a series of compounds with similar structures for RTP after being mixed with the matrix. However, even compounds with similar structures may not generate the same effect, such as 4-(1H-imidazol-1-yl)-N,N-dimethylbenzenamine. Figure S21 provides part of compounds that were tested inactive with 1BBI, which indicated that the matrix material has strong selectivity for active ingredients. Further research is needed to explore the unusual RTP emission in bicomponent systems.

The low-temperature phosphorescence was used to investigate the attribution of RTP. The phosphorescence lifetimes of DMBA/1BBI, PBA/1BBI, and JCA/1BBI were 179, 149, and 278 ms at 77 K, respectively, compared with 124, 109, and 143 ms at room temperature (fig. S22). The increase in lifetime at 77 K verified the luminescence to be phosphorescence. The phosphorescence spectra of pure 1BBI at 77 K showed an emission peak at around 425 nm, which highlighted a big difference with the bicomponent system (fig. S23). The phosphorescence spectra of pure DMBA, PBA, and JCA showed phosphorescence in MeOH solution at 77 K, which were much closer to their RTP than 1BBI (fig. S24). Their low-temperature phosphorescence lifetimes including DMIQI (fig. S25) were in an order of magnitude with that of RTP in bicomponent system. The low-temperature phosphorescence spectra and lifetime illustrated that the RTP in trace ingredient–mediated bicomponent system might emit from the triplet of active ingredients rather than from the matrices.

All the tested x-ray powder diffractometer (XRD) patterns indicated that the ingredient/matrix systems were crystal materials (fig. S26). Compared with the matrix, the XRD patterns were similar to that of pure 1BBI. When taking DMBA/1BBI into further research, three distinct peaks (black arrow) were found differing from the pure DMBA or 1BBI (Fig. 3A). After grinding the sample for 1 hour, these characteristic peaks remained but decreased in intensity. Similarly, the RTP also decreased but remained (Fig. 3B). The melting point of the DMBA/1BBI mixture was significantly lower than that of the pure components by differential scanning calorimetry test (Fig. 3C). This revealed that a cocrystal or solid solution may be formed and play roles in the generation of RTP.

Fig. 3 Crystal analysis of DMBA/1BBI solid.

(A) XRD pattern of DMBA/1BBI before and after grinding. (B) RTP intensity change of DMBA/1BBI before (red line) and after (blue line) grinding. (C) Melting points of the DMBA/1BBI mixture compared with single components.

Halogen-free bicomponent RTP system

The 1BBI is not the only matrix that can conduct the bicomponent RTP. The compound 1-(4-(4H-imidazol-4-yl)phenyl)-1H-imidazole (DIB) was synthesized for the purpose of removing the halogen atom in 1BBI. The carefully purified DIB showed no RTP even in crystal states. Then, the DMQI and DMIQI were selected to mix with DIB to form the halogen-free bicomponent RTP system. As seen from Fig. 4A, the RTP emission spectra of DMQI/DIB and DMIQI/DIB were similar to that of DMQI/1BBI and DMIQI/1BBI, respectively. It reveals that the RTP may come from the triplet ingredients rather than from the matrix. The QYs dropped from 7.6% (DMQI/1BBI) and 8.0% (DMIQI/1BBI) to 6.4% (DMQI/DIB) and 1.4% (DMIQI/DIB) because of the loss of heavy-atom effect. The RTP lifetime of DMQI/DIB was increased to τ = 430 ms compared with τ = 263 ms of DMQI/1BBI. When applied to other active ingredients, all the QYs and lifetimes were decreased at a reasonable level (Table 1 and Fig. 4B).

Fig. 4 Optical characteristics of trace ingredient–mediated bicomponent system using DIB matrix and corresponding mechanism.

(A) Normalized RTP spectra of trace ingredient–mediated bicomponent system using DIB matrix at 1 mol % contents. (B) RTP intensity decay curve of trace ingredient–mediated bicomponent system using DIB matrix at 1 mol % contents. (C) Fluorescence spectra (solid line and left y axis; excitation, 254 nm; slit width, 5 nm; voltage, 400 V) and RTP spectra (dashed line and right y axis; excitation, 254 nm; slit width, 5 nm; voltage, 600 V) of 4BBI/DIB at various proportions. (D) Fluorescence decay of DIB before and after being mixed with 4BBI. (E) Electron and hole motion after photoexcitation. Upon photoexcitation, electrons were transported from HOMO to LUMO of the matrix (I). Then, electrons from HOMO of impurities were transferred to the HOMO of the excited matrix to generate the charge transfer state (II). The resulting electrons and holes diffused in different directions to form free charge carriers (III). Last, nongeminate radiative recombination of free charge carriers generated luminescence (IV). (F) Defect-induced charge recombination. The impurities worked as defects or energy traps and enhanced the charge recombination process. RISC, reverse intersystem crossing; IC, internal conversion.

Table 1 RTP emission peaks, lifetimes, and QYs of trace ingredient–mediated RTP systems.

View this table:

The matrix DIB emits fluorescence peak at 328 nm in solid state. The absolute fluorescence QY was measured as 16.6%. Taking 4BBI/DIB as a model, the fluorescence intensities and QYs of DIB decreased significantly as the content of 4BBI increased (Fig. 4C and table S1). The fluorescence lifetime of 1% 4BBI/DIB was reduced to 1.73 ns from 5.54 ns of pure DIB (Fig. 4D). Meanwhile, the absolute fluorescence QY dropped to 9.4%. All the fluorescence changes of the matrix were consistent with the donor characteristic in the energy transfer system. However, the singlet-singlet energy transfer did not contribute to the formation of triplet states, and the singlet-triplet energy transfer was spin forbidden (20). One probable process was that the donor matrix absorbed first a photon to generate an electron-hole pair. Then, a charge transfer state was formed through electron transfer. The charge transfer state generated triplet excitons through the process of charge separation and charge recombination (14, 15).

Mechanism of trace ingredient–mediated bicomponent RTP

Recently, a great number of pure organic RTP system was achieved on the basis of certain strategies (21). Strategy-based mechanisms have also been proposed to explain the generation of RTP. In the traditional photoluminescence process, RTP is generated by the radiative transition of triplet state, which is generated from an excited singlet state through ISC. Crystal packing (22), polymerization (23), and host-guest interaction (2426) are analyzed to suppress the molecular vibration and oxygen quenching, which competed with the radiative transition. However, vibration restraint and avoiding quenching were not the sole factors for RTP, because not all materials could emit RTP in crystal state in vacuum (27). Other mechanisms for enhancing ISC were proposed by increasing heavy-atom effect, reducing the energy gap between singlet and triplet (ΔEST), and being subjected to the El-Sayed rule (28, 29).

For a multicomponent system, the phosphorescence mechanism would be unusual and more complex. Although the crystallization suppresses the molecular vibration and oxygen quenching, the above mechanism cannot elucidate the RTP of the trace ingredient–mediated bicomponent system. Cocrystallization of two compounds with similar crystal structure could achieve direct heavy-atom effect to enhance the RTP (30, 31). However, the difference between trace ingredient–mediated bicomponent RTP system is that the single component alone from a cocrystalline system can still emit RTP. External heavy-atom effect (32) and intermolecular halogen bond (33) are effective strategies to obtain RTP, which may explain partly the mechanism of trace ingredient–mediated bicomponent RTP. However, RTP can still be obtained in a halogen-free trace ingredient–mediated bicomponent system. Recently, a dopant system was reported to generate charge-separated states and emit RTP through the following charge recombination progress (14, 15).

Here are details of the possible mechanism of trace ingredient–mediated bicomponent RTP system. The matrices and impurities (or active ingredients) in the crystal formed charge transfer states upon photoexcitation. From the excitation spectra of 2BBI/1BBI and 4BBI/1BBI, we can see a weak excitation peak at ~350 nm (figs. S27 and S28), which did not match any absorption spectra of the single component. The fluorescence of DMIQI/1BBI system was red-shifted as the ratio increased (fig. S29). The change of excitation and emission spectra may reveal the formation of charge transfer state. Taking DMIQI/1BBI system as an example, the electron of 1BBI (or DMIQI) jumped from highest occupied molecular orbital (HOMO) to lowest unoccupied molecular orbital (LUMO) and generated an electron-hole pair after absorbing a photon (Fig. 4E). The generated excitons then diffused to the surface of DMIQI and 1BBI, generating the charge transfer state. Subsequently, the electron-hole pair was separated into free charge carriers before relaxing to the ground state. Last, the radiative recombination of free charge carriers or electron-hole pair generated fluorescence and phosphorescence (34). Recombination of free charge carriers generated from different photon has a 25% probability of producing singlet state and 75% triplet state according to spin statistics (35). The impurity DMIQI worked as defects in crystals and promoted defect-induced charge recombination (Fig. 4F) (36). This mechanism provided an efficient way to obtain triplet state by bypassing ISC.

White-light emission by bicomponent RTP system

Trace ingredient–mediated bicomponent RTP system exhibited a variety of luminescence colors, which makes it highly feasible for white-light emission design (37, 38). The DMQI/1BBI (1 mol %) couple presented blue fluorescence peak at 460 nm and yellow RTP peak at 560 nm. Under 280-nm excitation, the blue and yellow emissions achieved a balance to form white-light emission (Fig. 5, A and B). The calculated Commission Internationale de L’Eclairage (CIE) chromaticity coordinate (0.30, 0.33) was located at white region of the CIE 1931 chromaticity diagram (Fig. 5C). A series of proportion between 0.2 and 5 mol % DMQI/1BBI were prepared to optimize the white-light emission. They all exhibited near white-light emission under 280-nm excitation and yellow afterglow after removing the excitation light (fig. S30). The 1 mol % DMQI/1BBI showed the brightest white-light emission among all the tested samples of different proportions.

Fig. 5 White-light emission system.

(A) Steady-state luminescence of 1% DMQI/1BBI under 280-nm excitation. (B) Images of 1% DMQI/1BBI under daylight or 280-nm excitation. (C) CIE coordinate of 1% DMQI/1BBI luminescence.


In this work, a strong impurity-induced RTP with ultralong afterglow was observed in solid 1BBI. Although the content of the impurity DMIQI was very low, enough impurity was successfully isolated to determine its structure. The photoluminescence measurement of DMIQI/1BBI with differing molar contents showed that utilization of trace impurities or active ingredients could form a bicomponent RTP system of high efficiency, economy, and convenient preparation.

Beyond the accidental discovery of impurities from synthesis, an effective screening method was proposed to screen out a series of organic compounds, called ingredients, working similar to and even better than the impurity. Furthermore, the halogen-free bicomponent RTP systems were also formed by alternating matrix with a halogen-free imidazole derivative. Among all constructed bicomponent RTP systems, a high phosphorescence QY of 74.2% was achieved. It is easy for the bicomponent RTP system to realize strong afterglow obvious to the unaided eye with a lifetime up to 430 ms. Further analysis showed that the heavy-atom effect of the matrix can significantly improve the phosphorescent efficiency with limited effect on the lifetime of phosphorescence. The lifetime and wavelength of phosphorescence were more dependent on the optical properties of added active ingredients. Halogen-free ingredients are crucial in obtaining RTP of ultralong afterglow. By combining halogen-containing matrix with halogen-free impurities, the pure organic bicomponent RTP system can achieve both high QYs and ultralong afterglow.

Although no general rules were summarized for designing bicomponent RTP systems from a large number of ingredient/matrix combination tests, important empirical conclusions in our system can still be drawn. The interaction between isomers was most likely to generate bicomponent RTP, and the N-substituted aromatic amines structure proved more beneficial to the persistent RTP. The mutual selectivity between matrix and impurity is the key content to construct the bicomponent system, which needs further research in the future.

In summary, a trace ingredient–mediated bicomponent strategy for pure organic persistent RTP was introduced in this paper inspired by the trace impurity DMIQI engendering unusual RTP in organic compound 1BBI. By simply mixing the matrix with a trace amount of screened active ingredients, a series of trace ingredient–mediated bicomponent RTP systems of high phosphorescence QY and/or ultralong phosphorescence lifetime were achieved with tunable colors. The low-temperature phosphorescence properties, fluorescence lifetime of matrix, and XRD analysis illustrated the corresponding mechanism of this bicomponent RTP involved charge transfer, charge separation, and trap-assisted charge recombination process. This work concerning the great impact of trace impurity or ingredients on RTP will lead to a new understanding of persistent organic RTP. The trace ingredient–mediated strategy will be a facile way to design RTP materials for its high efficiency, color-tunable, low cost, and easy-to-prepare properties.



The compounds 2BBI, 4BBI, DMBA, PBA, and JCA were purchased from commercial source and purified before use. 1BBI, DIB, DMQI, and DMIQI were synthesized as described in the Supplementary Materials, carefully purified before use, and characterized by NMR and HRMS. All other reagents and solvents were obtained commercially and used as supplied without further purification unless specified otherwise.

General information

The ultraviolet-visible (UV-vis) absorption spectra of solid sample were obtained on a PerkinElmer Lambda 950 spectrophotometer. The UV-vis absorption spectra of solution state were obtained on an Agilent Cary 60 spectrophotometer. The steady-state photoluminescence spectra at room temperature were measured in air using a Horiba FluoroMax-4 spectrofluorometer. The time-resolved photoluminescence spectra and lifetime at room temperature were recorded in air on an Agilent Cary Eclipse spectrophotometer. The photoluminescence QYs were measured using a HAMAMATSU absolute PL QY spectrometer (C11347). NMR spectra were measured on a Bruker AV-400 spectrometer and processing on MestReNova (Mestrelab Research, version: 9.0.1) software. The HPLC coupled with HRMS was tested on a Thermo Fisher Scientific Q Exactive Plus spectrometer system. The electron impact HRMS was tested on a Waters GCT Premier spectrometer. X-ray diffraction experiments were carried out on Bruker D8 Venture diffractometer with a PHOTON 100 CMOS area detector using Mo-K radiation from an Incoatec IμS micro source with focusing mirrors. Reversed-phase chromatography was performed on SepaBean machine (Santai Technology Inc., China) equipped with C18-bonded SepaFlash columns.

Isolation of impurity DMIQI from 1BBI

Crude 1BBI was applied standard flash column chromatography on silica gel. The eluent solvent was changed in order from petroleum ether (PE), PE/ethyl acetate (EA) = 10:1, PE/EA = 5:1, PE/EA = 2:1, PE/EA = 1:1, EA, dichloromethane (DCM)/MeOH = 30:1, DCM/MeOH = 10:1 to MeOH at last. The effluent was collected in glass tubes in order. The solution in each tube was concentrated and dropped to the surface of solid 1BBI. After the solvent evaporated naturally, the strong RTP emerged if 1BBI was mixed with the right impurity. Last, the DMIQI showed Rf = 0.62 on a thin-layer chromatography plate using DCM/MeOH = 10:1 as a developing solvent.

Preparation of trace ingredient–mediated bicomponent RTP system

The ingredient-activated bicomponent RTP system was prepared by a simple solvent evaporation method. The selected ingredients (or impurities) and matrices were completely dissolved in MeOH solution at required proportions and evaporated to dryness under reduced pressure. Then, the sample was used directly for further test.

Time-dependent density functional theory calculations

Time-dependent density functional theory calculations were performed on the Gaussian 09E01 program. All the geometries were first optimized with B3LYP/6-311g(d) method, and then the singlet and triplet energies were calculated on the basis of optimized ground-state geometries. Their HOMO and LUMO were shown in fig. S31. The singlet-triplet energy gap (ΔST) and transition were calculated to explore ISC (table S2).


Supplementary material for this article is available at

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial license, which permits use, distribution, and reproduction in any medium, so long as the resultant use is not for commercial advantage and provided the original work is properly cited.


Acknowledgments: We acknowledge Y. Cheng for constructive discussions. We thank Y. Kang and Y. Huang at the Research Center of Analysis and Test of East China University of Science and Technology for the help on solid-state absorbance analysis. Funding: This work was supported by the National Natural Science Foundation of China 21788102, 22020102006, 21722603, and 21871083, the Program of Shanghai Academic/Technology Research Leader 20XD1421300, and the Innovation Program of Shanghai Municipal Education Commission 2017-01-07-00-02-E00010. “Shu Guang” project was supported by the Shanghai Municipal Education Commission and the Shanghai Education Development Foundation 19SG26, the Shanghai Municipal Science and Technology Major Project 2018SHZDZX03, and the Fundamental Research Funds for the Central Universities. Author contributions: Conceptualization: B.D., X.M., and H.T. Methodology: B.D. and L.M. Investigation: B.D., L.M., and Z.H. Supervision: X.M. and H.T. Writing—original draft: B.D. and X.M. Writing—review and editing: B.D., L.M., Z.H., X.M., and H.T. Competing interests: X.M., H.T., and B.D. are inventors on a provisional patent application related to this work that has been filed by the East China University of Science and Technology (application no. 202011207838.X, filed 2 November 2020). The authors declare that they have no other competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.

Stay Connected to Science Advances

Navigate This Article