Peng
Lin
abcd,
Junpeng
Shi
*abcd,
Lin
Liu
cd,
Yile
Kang
cd,
Liang
Song
cd,
Maochun
Hong
*abcd and
Yun
Zhang
*abcd
aSchool of Rare Earths, University of Science and Technology of China, Hefei 230026, China. E-mail: shijunpeng10@mails.ucas.edu.cn; hmc@fjirsm.ac.cn; Zhangy@fjirsm.ac.cn
bGanjiang Innovation Academy, Chinese Academy of Science, Ganzhou 341000, China
cState Key Laboratory of Structural Chemistry, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou 350002, China
dXiamen Key Laboratory of Rare Earth Photoelectric Functional Materials, Xiamen Institute of Rare Earth Materials, Haixi Institute, Chinese Academy of Sciences, Fuzhou 350002, China
First published on 19th July 2023
Persistent luminescent phosphors (PLPs) have attracted much attention in anti-counterfeiting, biosensing and bioimaging due to their characteristic properties of emitting long persistent luminescence (PersL) after excitation ceases. However, most of the PLPs developed so far have only one PersL emission peak, which limits their application in advanced anti-counterfeiting and precise biosensing. Here, a series of dual-emissive Zn2GeO4:x% Tb3+/y% Bi3+ PLPs was prepared by codoping Tb3+ ions and Bi3+ ions to significantly improve the trap depth and trap density, resulting in a 16.8-fold increase in PersL intensity compared to the Zn2GeO4 host. The photoluminescence (PL) color was adjusted to pale blue under 254 nm UV excitation by trap depth engineering, while the PersL color was green, and the PL color under 365 nm UV excitation was yellow. In addition, the PersL intensity was dramatically increased by 16.8 times by co-doping Tb3+ and Bi3+. Therefore, multi-mode anti-counterfeiting was achieved by multi-stimulus response. Furthermore, the dual-emission band was designed as a ratiometric PersL aptasensor for the detection of mucin 1. The detection linearity ranged from 0.1 to 100 ng mL−1 with a detection limit of 0.036 to 1 ng mL−1. This novel PLP with a dual-emission band expands the application range of PLPs.
However, most of the PLPs developed to date have only one emission band, which severely limits their application prospects. This characteristic poses a challenge in anti-counterfeiting, as only one mode of anti-counterfeiting can hardly meet the requirements of advanced information encryption. Therefore, PLPs are usually hybridized with other luminescent ingredients through complicated design in order to achieve multi-mode anti-counterfeiting.23–25 However, this approach induces difficulties in the fabrication process and information reading due to the multicomponent and labyrinthine pattern involved.26,27 Similarly, the accuracy of solo-emissive PLPs is relatively low when they are used for biosensing due to the lack of reference. The detection signal is strongly correlated with time due to the fact that PersL decays dramatically once the excitation is stopped, resulting in an inevitable detection error when quantifying.28–30 Therefore, there is an urgent need to develop novel dual-emissive PLPs to meet the requirements of multi-mode anti-counterfeiting and high-precision biosensing.
Herein, a series of dual-emissive Zn2GeO4:x% Tb3+/y% Bi3+ PLPs is prepared. The two PL emission bands, attributed to the Zn2GeO4 host, are located in the blue and green regions of the visible spectrum. The trap depth and trap density were increased by co-doping Tb3+ and Bi3+, resulting in a significantly enhanced PersL intensity. By engineering the trap depth, pale blue photoluminescence (PL) was observed under 254 nm UV excitation, while strong PersL green emission was exhibited after the excitation was ceased. In addition, yellow PL emission was observed under 365 nm excitation. Based on the multistimulus-responsive luminescence properties, this PLP was applied for multi-mode anti-counterfeiting. Additionally, this novel dual-emissive PLP was developed as a ratiometric PersL aptasensor for sensitive and accurate detection of tumor marker mucin 1 (MUC1) without background interference. Compared to solo-emissive PLPs, this dual-emissive PLP has obvious advantages in various applications.
Among the samples, Zn2GeO4:8% Tb3+/1% Bi3+ exhibited unique luminescence properties. Under 254 nm excitation, the photoluminescence spectrum showed a broad emission peak at 445 nm, resulting in a pale blue PL color that was visible to the naked eye (Fig. 1a). When the excitation source was changed to 365 nm UV light, the PL color changed to yellow (Fig. 1b), and the photoluminescence spectrum exhibited a broad emission peak at 580 nm. The photoluminescence quantum yields were found to be 15.86% and 8.33%, corresponding to the emission peaks at 445 nm and 580 nm, respectively. Moreover, after the 254 nm UV irradiation, the sample emitted strong persistent green luminescence (Fig. 1c). The PersL spectrum showed a broad peak from 400 nm to 600 nm, peaking at 527 nm. The CIE coordinates of PL under the 254 nm excitation, PL under the 365 nm excitation and PersL were calculated to be (0.18, 0.17), (0.41, 0.44), and (0.23, 0.58), respectively (Fig. 1d), demonstrating the different luminous color responds to excitation at different wavelengths.
The luminescence performance of ZGO was significantly affected by the concentrations of Bi3+ and Tb3+. For PersL, the intensity of PersL without doping in the ZGO host was extremely weak (Fig. 2a). Separate doping of Tb3+ ions and Bi3+ had a negligible effect on the intensity of PersL. However, the Tb3+ and Bi3+ codoping significantly enhanced the PersL of the host and hardly changed the PersL color. The PersL intensity of Zn2GeO4:8% Tb3+/1% Bi3+ was increased by 16.8 times compared to the ZGO host. To reveal the optimal doping concentration for PersL, different concentrations of Tb3+ and Bi3+ were tested (Fig. 2b). When the concentration of Bi3+ ions was fixed at 1%, the PersL intensity gradually increased with the doping concentration of Tb3+ ions increasing to 8%. However, the intensity of PersL decreased when the doping concentration of Tb3+ exceeded 8% due to concentration quenching. Similarly, when the concentration of Tb3+ ions was fixed at 8%, the PersL intensity gradually increased with the doping concentration of Bi3+ increasing to 1%. However, the intensity of PersL decreased when the doping concentration of Bi3+ exceeded 1% due to concentration quenching. Therefore, the optimal doping concentrations for PersL intensity were found to be 8% Tb3+ and 1% Bi3+. And the intensity of PersL remained considerable for 5 hours after the excitation (Fig. 2c).
As to PL emission, the doping concentrations of both Tb3+ and Bi3+ affected the shape of the PL spectra excited at 254 nm (Fig. 2d and e). Two peaks were observed in the PL spectra, which were fitted as band A and band B (Fig. S5†). The luminous intensity ratios of band A/band B at different doping concentrations were analyzed, as shown in Fig. 2f and g. With an increase in Bi3+ concentration, the band A/band B intensity ratio first decreased, then increased, and finally decreased again. With the Tb3+ concentration increase, the band A/band B intensity ratio gradually increased, followed by a decrease. Under the 365 nm UV excitation, a broad emission band was observed from 450 to over 700 nm (Fig. S7†). The characteristic peak of Tb3+ (542 nm), attributed to 5D4 → 7F5, was quenched with increasing Bi3+ concentration, resulting in yellow fluorescence.31
Compared to the ZGO host, with separate doping of Tb3+ and Bi3+, the number and position of peaks of PL curves and PersL curves were slightly changed and could be well fitted as band A and band B (Fig. S5†). Moreover, no lanthanide characteristic narrow peak was observed with Tb3+ doping, and no additional excitation band was observed with Bi3+ doping (Fig. S6†). Therefore, it was reasonably speculated that the PL (λex = 254 nm) and PersL emission bands were attributed to the Zn2GeO4 host instead of Tb3+ ions or Bi3+ ions. To investigate how the dopant concentration affects the luminescence properties, the thermoluminescence (TL) of the samples was studied (Fig. 3a and b). The trap depth was calculated according to the empirical formula E = Tm/500 (E and Tm represent the trap depth and TL peak, respectively).32 Without a dopant, the trap depth of the ZGO host was 0.61 eV (Fig. 3c). The trap depth gradually increased to 0.73 eV, while the Tb3+ concentration increased to 8%. The trap depth decreased when the doping concentration of Tb3+ exceeded 8%. With the increase of Bi3+ doping concentration, the trap depth became deeper from 0.61 eV to 0.75 eV. The deeper trap depth prolonged the PersL emission. In addition to the trap depth, the variation of the doping concentration significantly increased the trap density. The trap density reached its maximum when the doping concentrations of Tb3+ and Bi3+ were 8 and 1%, respectively (Fig. 3c and d). Compared to the ZGO host, the trap density was increased by 33.2 times with the codoping of Tb3+ and Bi3+, leading to a much stronger PersL intensity. Based on the TL results, a plausible PersL emission mechanism relying on the codoping of Tb3+ and Bi3+ was proposed (Fig. 3e).33 Due to the two emission bands in the PL spectrum and the PersL spectrum, we reasonably speculate that there are three defect levels in the forbidden band that are caused by intrinsic defects in the ZGO host. The ground state was caused by Vzn or VGe (zinc vacancy and germanium vacancy). The excited state levels were caused by VO and Zni (oxygen vacancy and zinc interstitial), corresponding to the lower energy level and higher energy level, respectively. Under excitation, electron–hole pairs were generated. The holes shifted from the valence band to the ground state, while the electrons simultaneously transitioned to the conduction band and were captured by the trap. After the cessation of excitation, the captured electrons moved back to the conduction band due to thermalization and then relaxed to the two excited state levels through non-radiative processes. The recombination of electrons and holes resulted in two PersL emission bands. Due to the deeper trap depth and increased trap density, the trap caused by the codoping of Tb3+ and Bi3+ captured more electrons compared to the ZGO host and simultaneously prolonged the electron transition. Therefore, through codoping Tb3+ and Bi3+, the PersL intensity was improved and the PersL duration was prolonged.
Based on the above results, multi-stimulus-responsive, multimode anti-counterfeiting was fabricated by 3D-printing, which is convenient for various application scenarios (Fig. 4a). In detail, photocurable resin was prepared and fabricated to emboss with different patterns. Meanwhile, the PLP powder was mixed with PDMS to form a slurry as ink. The patterns were printed by sealing, and thermocuring was subsequently performed to fix the PLP. The encrypted information was read by UV lamp irradiation. Under 254 nm irradiation, pale blue patterns were observed, while the patterns became green after ceasing irradiation (Fig. 4b). When excited by 365 nm UV light, yellow luminescence of patterns was observed, and no PersL was detected after the excitation. In conclusion, after excitation at different wavelengths, the corresponding color of the pattern was observed, achieving triple-mode anti-counterfeiting.
![]() | ||
Fig. 4 (a) Schematic illustration of the fabrication process of anticounterfeiting patterns and (b) typical example of dynamic multimodal anti-counterfeiting. |
In order to adapt to a wider range of applications, we attempted to reduce the size of Zn2GeO4:5% Tb3+/1% Bi3+ by a hydrothermal method. The nano Zn2GeO4 (nZGO) possessed a nanorod structure of about 50–100 nm in length, as revealed by the transmission electron microscopy (TEM) image (Fig. 5a). The nZGO was found to contain Zn, Ge, O, Tb and Bi, as shown by the EDS spectrum, and all the elements were homogeneously distributed (Fig. 5b and Fig. S8†). The XRD pattern of nZGO matched well with that of a standard pdf card, indicating the pure phase of nZGO (Fig. 5c). Surprisingly, the PersL spectrum of nZGO showed two strong peaks, including band A and band B (Fig. 5d). Furthermore, the decay trends monitored at 455 nm and 540 nm were similar, and the intensity ratio of the two emission bands (I540/I455) remained constant (Fig. 5e). The I540/I455 value changed slightly when the concentration of nZGO in aqueous solution was altered (Fig. 5f). Based on the dual-emissive PersL properties of nZGO, a ratiometric PersL nanoprobe without autofluorescence was designed for the detection of mucin 1 (MUC1), which is a widely known tumor marker that is overexpressed in tumor cells.34,35 The design strategy is shown in Fig. 6a. Dual-emissive PersL nZGO-NH2 was conjugated with a sulfhydryl terminal MUC1 aptamer via a cross-linking reaction by sulfo-SMCC to obtain nZGO-apt. The ATTO425-modified aptamer cDNA was subsequently hybridized with nZGO-apt via base complementary pairing to yield the ratiometric PersL nanoprobe nZGO-ATTO425. The PersL at 455 nm was quenched by Förster resonance energy transfer (FRET) due to the overlap of the absorption of ATTO425 and the emission of nZGO at 455 nm (Fig. S9†). In the presence of MUC1, nZGO-apt preferentially bound to MUC1, leaving behind cDNA-ATTO425, resulting in PersL recovery at 455 nm with MUC1 concentration-independent PersL at 540 nm as an internal reference.
The successful conjugation of nZGO-ATTO425 was confirmed by the obvious zeta potential variation of the samples at each step (Fig. 6b). The hydrodynamic diameter gradually increased from 222.9 ± 4.1 nm (nZGO-NH2) to 317.3 ± 2.5 nm (nZGO-apt), further proving that nZGO-ATTO425 was obtained (Fig. S10†). Besides, the characteristic absorption band at ∼445 nm originating from ATTO425 was observed in nZGO-ATTO425 (Fig. S11†). The conjugation content of ATTO425 was calculated to be 1.6% (wt%) with 56.4% conjugation efficiency based on the concentration standard curve of ATTO425 (Fig. S12 and S13†). MUC1 detection was performed by incubating nZGO-ATTO425 with MUC1, and the change in the PersL ratio (Δ(I540/I455)) was monitored. The maximum change was observed after 1 hour of incubation with MUC1 (Fig. 6c). With MUC1 concentration increasing, the PersL intensity at 455 nm gradually recovered while the PersL intensity at 540 nm remained constant, indicating concentration-dependent MUC1 detection (Fig. 6d). In the MUC1 concentration range from 0.1 ng mL−1 to 100 ng mL−1, the good linear relationship between Δ(I540/I455) and the logarithmic concentration of MUC1 with a determination coefficient (R2) of 0.9945 was proved (Fig. 6f). The limit of detection (LOD) was estimated to be 0.036 ng mL−1 on the basis of the 3σ (standard deviation) rule. The above results indicate that this ratiometric PersL nanoprobe without autofluorescence interference has advantages in a broad detection range, precise detection and a low LOD.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3qi01098h |
This journal is © the Partner Organisations 2023 |