1. Introduction

Afterglow luminescent materials with a long-lived emission lifetime have received considerable attentions in optoelectronics, advanced security imaging, molecular imaging, and data security [1–4]. Numerous efficient afterglow phenomena have been well exhibited in the inorganic materials. Nevertheless, these afterglow materials normally involve the doping or co-doping with transition metals and rare-earth ions, whose high cost and cytotoxicity is a noticeable concern [5–8]. For the above reasons, the metal-free organic afterglow materials provide a potential alternative and gain the extensive attentions by researchers. However, achieving effective room temperature phosphorescence (RTP) in organic materials is still a great challenge due to their poor intersystem crossing (ISC) and rapid rate of nonradiative deactivation [9–10].

Carbon dots (CDs) as an emerging class of luminescent materials have the huge potential to be developed as the effective metal-free RTP material, because the effective singlet-to-triplet ISC can be activated through the assistant of the functional groups (such as C = N and C = O bonds), or by doping of nitrogen and phosphorus elements, which facilitates the transition of the triplet states excitons [11–18]. So far, effective RTP performances of CDs have been developed through the construction of the polymer structure, or by mixing CDs with different matrices [15–24]. Till date, considerable attention was paid to adjust the RTP emission wavelength, some examples include color-tunable RTP from CDs produced through heating treatment of a series fluorescence CDs with boric acid or melting urea, but only a few studies on the lifetime-tunable of CD-based RTP have been reported [25–27]. However, for their practical application, such as anti-counterfeiting and data encryption, the adjustments of RTP lifetimes are equally important [26]. Thus, it is important to develop new synthetic methods toward CDs with the lifetime-tunable RTP emission.

In this work, we developed a facile method to adjust lifetime and improve emission intensity of RTP CDs with green emitting color. Firstly, the solid-state RTP CD1 with weak green emission was synthesized through the solvothermal reaction of urea in dimethylformamide (DMF), as shown in Fig. 1. And then, inspired by the concept of crosslink-enhanced emission (CEE) effect, through introducing polymer polyvinylpyrrolidone (PVP) in the process of the CD formation (named CD2), the RTP lifetime of CDs increased from 45 to 139 ms, and the emission intensity was also improved. Furthermore, the CD2 were further modified with urea by a heating treatment (uCD2), and the RTP lifetime was further improved, leading to an increase of lifetime from 139 to 550 ms. It has been found that the increase of RTP lifetime benefits from the decreasing vibration and rotation of C = O/C = N bonds at the CDs surface, so as to protect triplet states from quenching. Finally, naked-eye-observable time-resolved anti-counterfeiting application was prepared based on the modified CDs.

 

Fig. 1. The synthesis process for CD1, CD2, and uCD2.

Download Full Size | PDF

2. Results and discussion

From transmission electron microscopy (TEM), the CD1 is found to be nearly monodisperse with an average particle diameter of 2.0 nm (Fig. 2(a)-(b)). The surface groups and chemical compositions of the CD1 are identified by Fourier transform infrared (FT-IR) and X-ray photoelectron spectroscopy (XPS) spectra. As seen in Fig. 2(c), the CD1 displays a broad absorption peaks from 2700 to 3600 cm^-1, attributed to the stretching vibrations of hydroxyl (-OH), amino (-NH₂), and methylene (-CH₂-) groups [28–29]. In addition, the stretching vibrations of C = N at 1665 cm^-1 in the CNH group, and the stretching of C = O at 1636 cm^-1 in the amide group are detected [30]. As shown in Fig. 2(d) and Fig S1 in the Supplement 1, the CD1 mainly contains C, N, and O, elements with compositions of 74.6%, 5.4%, and 20%, respectively [31–33]. The HR XPS spectrum for the C 1s indicates the presence of C-C/C = C (284.6 eV), C-N (285.2 eV), C-O (286.2 eV) and C = O (288.8 eV) bonds in the CDs, [34] and the corresponding spectrum for the N 1s contains two components that can be assigned to C-N (399.6 eV), and N-H (401.2 eV), respectively [27].

 

Fig. 2. (a) TEM image of CD1. (b) The size distribution of CD1 particles. (c-d) The FT-IR and XPS spectrum of CD1. (e) The FL and RTP emission spectra of CD1. (f) Proposed RTP emission processes of CD1 powder.

Download Full Size | PDF

The luminescence performances of the CD1 have been investigated. Fig S2 (black line) shows two strong UV-vis absorption peaks of CD1 water solution, at around 270 nm from π→π* transitions of the aromatic sp² domain of C = C bonds, and the shoulder at 310 nm from the n→π* transition of C = O/C = N bonds [4]. It is found that CD1 powder presents blue fluorescence (FL) emission at around 465 nm under the excitation of 365 nm UV lamp, and a green emitting RTP at around 540 nm can be detected after switch off the lamp (Fig. 2(e)), which is associated to the triplet states of excitons in CD1 that consisting with the literatures (Fig. 2(f)) [35–36]. As shown in Fig S3 and Table S1, the RTP dynamic process can be fitted by a tri-exponential with average decay lifetime of 45 ms. The multiplex dynamics may be due to a wide range of chemical environments for the C = O/C = N bonds on the surface of CDs [20]. To confirm this possible explanation, the phosphorescence excitation (PLE) spectrum of the CD1 was detected under 540 nm emission (Fig S2, red line). In PLE spectrum of CD1 powder, the excitation band at 310 nm nearly overlap with its absorption peak, suggesting the phosphorescence comes from the C = N/C = O bonds of CD1 [4,20].

The current works demonstrate that the polymer-like structure of CDs could decrease vibration and rotation of the excited triplet species, thus suppressing the non-radiative transitions and generate RTP [19]. Inspiring by these works, we introduced PVP in the process of the CD formation. The morphology of CD2 was characterized by the TEM image, as shown in Fig. 3(a). It is found that the CD2 are well dispersed nanoparticles with a mean particle diameter of 2.3 nm. The HRTEM image provided as insets of Fig. 3(a) shows the crystalline part is surrounded by non-crystalline region, which indicates wrapped PVP chains [28]. Figure 3(b) shows the comparison of the RTP emission spectra for CD1 and CD2. The two samples exhibit similar RTP behaviours, but the emission intensity of CD2 is stronger. The decay processes of RTP emission for CD2 was presented in Fig. 3(c). It is found that the lifetime for CD2 increases clearly comparing with that of CD1, from 45 ms in CD1 to 139 ms in CD2. The increased RTP lifetime for the CD2 is probably due to the linkage of PVP polymer chains at the surface of CDs, which can decrease vibration and rotation of the excited triplet species, thus inhibiting non-radiative relaxation channels [19].

 

Fig. 3. (a) TEM image of CD2 (Inset provide the HRTEM image). (b) The RTP emission spectra of CD1 and CD2. (c) RTP lifetime decay curve of CD2, under excitation at 365 nm.

Download Full Size | PDF

The RTP emission intensity of CDs was further enhanced by heating treatment of CDs with urea. As shown in Fig. 4(a), (a) bright green RTP emission was observed from uCD2 powder when the UV light was turned off, with commission International de l’Eclairage (CIE) color coordinates (Fig S4) of (0.329, 0.456), which is consistent with the RTP phenomenon observed by the naked eye. By reference to the previously reported method, the phosphorescence quantum yield (QY) of the uCD2 powder has been calculated with 5.3% [17]. In the X-ray diffraction (XRD) patterns (Fig. 4(b)), the peaks of uCD2 are almost the same as those of the melting urea, indicating that the CDs particles are composited with the urea matrix for CD/urea composite sample. To obtain insights into the RTP properties of the uCD2, their FL and RTP decay spectra were further measured. As expected, it is found that uCD2 exhibit higher RTP emission intensity compared with that of CD2 (Fig. 4(c)). The RTP decay process (Fig. 4(d)) can be fitted by the three-exponential parameters as summarized in Table S1 in the Supplement 1. Compared with CD2, clear increases in the average lifetimes are seen for the uCD2 (increase from 139 to 550 ms). On one hand, it can be attributed to the formation of hydrogen bonds through interactions between the C = O/C = N of CDs and amino groups of melting urea as shown in Fig. 4(e) [20,37]. The formation of hydrogen bonds could rigidify and decrease vibration and rotation of C = O/C = N bonds, inhibiting non-radiative relaxation channels, thus leading to enhancement of the RTP emission [20–25]. On the other hand, the melting urea could act as a rigid matrix, exhibiting strong suppression in the non-radiative pathways, which is similar to the matrix packing mechanism [20]. To further prove the role of the melting urea, we prepared uCD2 powder at different reaction times from 0 to 6 h. Optimal RTP emission is found to be excited for 3 h (Fig S5). At short reaction times (2 h), only a small portion of urea is converted into melting urea, which leads to low emission due to insufficient hydrogen bonds between urea and CDs [20]. At long reaction times (6 h), almost all urea is depleted so that they cannot offer strong rigidity to restrict the vibration, resulting in relatively poor phosphorescence [20].

 

Fig. 4. (a) Photographs of the uCD2 powder taken at the different delay times after UV excitation light (365 nm) has been turned off. (b) The XRD patterns of urea and uCD2. (c) RTP emission spectra of CD2 and uCD2. (d) RTP lifetime decay curve of uCD2, under excitation at 365 nm. (e) Schematic illustration for the possible interactions of CDs surface groups with melting urea.

Download Full Size | PDF

The uCD2 fabricated here are highly stable, and a slight decay is detected of their initial FL intensity, under continuous illumination with a UV light source for 10 h (Fig S6). The excellent photo-stability of CDs and coupled with their low cost and environment-friendliness are compelling for the application in lighting application [38]. According to the literature, in order to achieve the white light emitting color, we developed CDs@EuCl₃ composite through mixing the CDs with EuCl₃ (details are demonstrated in the Experimental section) [39]. The FL spectra of CDs@EuCl₃ composite with different amounts of EuCl₃ and CDs (mass ratios: 50:11, 50:9, 50:6, 50:2, and 50:0) under 380 nm excitation are shown in Fig. 5(a). From the FL spectra, it can be clearly found that the proportion of red emission increases significantly compared to blue emission, and the luminescent colors of the CDs@EuCl₃ composite can be regulated from blue to white in line with different proportions of red and blue emissions (Fig. 5(b)). Importantly, a pure white light emission with CIE coordinates of (0.331, 0.324) and QY of 16.9% was achieved under 380 nm irradiation. Then, we combined the CDs@EuCl₃ sample with a 380 nm UV chip to fabricate a WLED, as shown in Fig. 5(c). This device gives an intense white light with CIE coordinates of (0.325, 0.319), Ra (average CRI value) (72.0) and luminous efficiency (10.6 lm W^-1). The stability of the WLED based on CDs@EuCl3 composite was monitored for 10 h under continuous operation (Fig S7) and a 11% decrease meant nice stability, indicating promising practical lighting applications. In addition, the RTP properties of the CDs powder could be developed for promising applications in anti-counterfeiting [30]. As shown in the bottom of Fig. 5(d), (a) panda image was printed onto a non-luminescent background paper using the uCD2 water solution as ink. After complete drying, a bright blue panda was observed under UV excitation and the green phosphorescence signal of the panda also can be recognized after removing the UV lamp, owing to the ultralong phosphorescence lifetime of the uCD2.

 

Fig. 5. (a) FL emission spectra of CDs@EuCl₃ composites with different mass ratios: 50:11, 50:9, 50:6, 50:2, and 50:0. All samples were excited at 380 nm. (b) CIE chromaticity diagram showing the color coordinates of the samples presented in (a). (c) FL emission spectrum of the CDs@EuCl₃ composites based WLED (inset: Photograph of the WLED operated at 10 mA). (d) Digital images of the sealed panda on a non-luminescent background paper using uCD2-based ink under 365 nm UV lamp on and off, respectively.

Download Full Size | PDF

3. Conclusion

In summary, we presented a facile and cost-effective strategy to increase the lifetime of green RTP CDs through the multi-step modification. Through modulating the surface composition of the CDs with different modifiers, the RTP lifetime remarkably improved, which was attributed to the confinement of vibration and rotation of the triplet states, thus inhibiting non-radiative relaxation channels. Furthermore, the RTP CDs has been demonstrated to be a potential security ink, which could be employed in the fields of anti-counterfeiting. Moreover, benefitting from the high operability of CDs, the RTP CDs have also been successfully applied to light-emitting diode applications. Our results are important for the rational design of CD-based materials to realize efficient RTP emission.

4. Experimental section

4.1 Chemicals

All reagents were used as received without further purification. Dimethylformamide (DMF, 99.5%), citric acid (99.5%) was purchased from Beijing Chemical Works. Urea (99%) was purchased from Xilong Scientific Company Limited. Polyvinylpyrrolidone (PVP), was purchased from Sigma-Aldrich.

4.2 Preparation of CD1, CD2 powder

For CD1: 1 g urea was dissolved in 10 mL DMF solution. Then the well-stirred solution was reacted at 180 °C for 8 h for the formation of CD1. For CD2: 1 g urea and 0.5 g PVP were dissolved in 10 mL DMF solution. Then the well-stirred (1 h) solution was reacted at 180 °C for 8 h for the formation of CD2. After the reaction, the reactors were cooled to room temperature naturally, and then the solution was dialyzed with a dialysis bag. The CDs powder was then obtained by drying at 60 ℃.

4.3 Preparation of uCD2 powder

Firstly, CD2 powder (200 mg) and urea (300 mg) were dissolved in 5 ml water solution in beaker by shaking for 10 min to completely dissolve the urea and CDs. Then the beaker was put into oil bath at 150 °C for 3 h. After the reaction, the uCD2 powders were obtained by ice bath.

4.4 Preparation of CDs@EuCl₃ composite

Firstly, uCD2 powder (100 mg) and EuCl₃ (22, 18, 12, or 4 mg) were dissolved in 10 ml water solution in beakers by shaking for 10 min to completely dissolve the EuCl₃ and CDs. Then the beaker was put into oil bath at 150 °C for 3 h. After the reaction, the CDs@EuCl₃ composite powder were obtained by ice bath.

4.5 Characterization

For the transmission electron microscopy (TEM) and high-resolution microscopy observations were performed with a JEOL-2100F microscope. Ultraviolet-Visible (UV-Vis) absorption spectra were measured with a Shimadzu UV-3101PC UV-Vis scanning spectrophotometer. X-ray photoelectron spectroscopy (XPS) was performed with an ESCALab220i-XL electron spectrometer from VG Scientific. Fourier transform infrared (FT-IR) spectra were obtained on a Nicolet 6700 FT-IR spectrometer. Photoluminescence spectra were recorded at room temperature by using one FLS980 spectrometer (Edinburgh Instruments Ltd). The excitation source was a 450W Xe arc lamp. Phosphorescence emission spectra of CDs were measured by a QE Pro (Ocean Insight). Phosphorescence lifetimes were measured using PLS980. The absolute emission quantum efficiency values were measured at room temperature using a calibrated integrating sphere in FLS980 spectrometer.