Color manipulation of Bi3+-activated CaZnOS under stress with ultra-high efficiency and low threshold for anticounterfeiting applications

Yun-Ling Yang a, Qian-Li Li ab, Xue-Chun Yang a, Woochul Yang c, Ran An d, Ting Li e, Yu Zhou a, Hong-Wu Zhang f, Jing-Tai Zhao a and Zhi-Jun Zhang *ac
aSchool of Materials Science and Engineering, Shanghai University, Shanghai, 200072, P. R. China. E-mail:
bPanzhihua University, Panzhihua, 617000, P. R. China
cDepartment of Physics, Dongguk University, Pildong-ro, Choong-gu, Seoul, 04620, South Korea
dInstitute of Fluid Physics, CAEP, Mianyang, Sichuan 621900, P. R. China
eSchool of Materials Science and Engineering, Guilin University of Technology, Guilin, 541000, P. R. China
fInstitute of Urban Environment, Chinese Academy of Sciences, Xiamen, 361021, P. R. China

Received 29th November 2019 , Accepted 14th January 2020

First published on 15th January 2020

We report blue to green emission mechanoluminescence (ML) in CaZnOS:Bi3+,Li+with about 50 times greater ML intensity than that of the strong ML material ZnS:Cu+. The ML spectrum showed an intense emission band peaking at 485 nm and a red-shift of about 9 nm on increasing the stress or friction. In addition, the preferable threshold of CaZnOS:Bi3+,Li+ was below 1 N due to its excellent strain sensitivity. The depths of the traps responsible for ML were estimated to be in the range of 0.68–1.2 eV. The as-prepared CaZnOS:Bi3+,Li+ phosphors were successfully used for the fabrication of homemade ink and flexible thin films, which could be further used for anticounterfeiting and stress mapping. These findings reveal its great potential for application in an advanced anticounterfeiting technology and visualization of stress distribution for damage diagnosis due to its ultra-high efficiency, low threshold and color tunability.

1. Introduction

Mechanoluminescent (ML) materials can respond to fracture or deformation of the solid under mechanical stimuli as they possess piezoluminescence (EML), triboluminescence (TL), and fractoluminescence (FML) properties.1,2 This phenomenon is caused by a special luminescence mechanism and offers potential applications in stress sensing, displays, illumination without source and bio-imaging, and has recently attracted tremendous attention.3–5 It eliminates the need for an electron and photon excitation source and just utilizes mechanical energy to emit light. Hence, ML materials can be energy saving and environmentally safe.6,7

In recent years, research on piezoluminescence, which is also termed elastic ML, has made dramatic advances.8 Xu et al. demonstrated in their outstanding works that strong intensity light can be observed by applying stress on ZnS: Mn2+ and SrAl2O4: Eu2+ samples.9,10 Notably, metal-ion-doped ZnS could exhibit strong visible light emission, was highly durable and could achieve multicolor emissions.11,12 It is known that doped transition metal ions, such as Mn2+, Cu+, and rare earth ions serve as activators in most of the ML materials. Among them, some lanthanide ion-activated ML materials, such as SrAl2O4: Eu2+,9,13 LiNbO3: Pr3+[thin space (1/6-em)]14 and Sr3Sn2O7: Sm3+,15 have limited applications due to the high cost of the rare-earth elements and the line emissions originating from the 4f–4f transitions. Furthermore, weak ML intensity (BaZnOS: Mn2+)16 leads to the failure of color manipulation because concentration quenching is prone to occur at high doping concentrations; for example, CaZnOS: Mn2+ could achieve color change from yellow to red with different Mn2+ concentrations from 0.1 to 10 mol%.17 At present, there is no report on bismuth ion-induced ML. Bi is well-known as a green element for its non-toxic and non-carcinogenic properties.18 Furthermore, it is not only much cheaper than rare-earth ions, but also possesses prominent advantages, such as different valence states and broad emission bands in different colors due to its strong interaction with the surrounding lattice. Bi0 or Bi+ ions in some host lattices show infra-red emission at about 1000 to 1600 nm. Bi2+ ions exhibit red emission in the wavelength range of 580 to 700 nm, and Bi3+ ions emit light in the broad 450 to 500 nm band.19–21 In this work, bismuth ion exhibited an interesting luminescence phenomenon by changing the ML color slightly from blue-green to green with increased pressure or friction power. It also had strong ML intensity, which indicated its potential to be used in storage devices, stress sensors and displays. Interestingly, slight color manipulation could be achieved by changing the concentration of the doped Bi3+ ions.

It is well-known that CaZnOS is a promising host lattice and shows piezoelectric performance and good luminescence. Our group has illustrated the crystal structure and the photoluminescence properties of the CaZnOS:Bi3+,Li+ phosphor in 2018.22

In this paper, we have studied the ML properties, including ML intensity, color and sensitivity by varying Bi3+ concentration, as well as piezoluminescence and triboluminescence with different loads and load rates. Moreover, we have investigated the piezoelectric effect and mechanism of ML in the CaZnOS host lattice.

2. Experimental section

2.1 Materials and synthesis

A series of Ca1−2x%ZnOS: x% Bi3+, x% Li+ (x = 0, 0.5, 1, 2, 3, 5 and 7) phosphors were prepared by the solid-state reaction method at a high temperature. In order to maintain the charge balance, Li+ was introduced into CaZnOS. The starting materials, CaCO3 (99.99%), ZnS (99.99%), Bi2O3 (99.99%) and Li2CO3 (99.99%) were weighed, mixed and subsequently ground in an agate mortar. The powder mixtures were fired at 1100 °C for 3 h in a horizontal tube furnace in an Ar atmosphere. Finally, the samples were gradually cooled down to room temperature (RT) in the furnace and ground again in the agate mortar. To characterize ML, cylinder-shaped (25 mm in diameter and 15 mm in thickness) composites were prepared by mixing the as-synthesized particles with an optical epoxy resin. The commercial ZnS phosphor (GGS 42) was purchased from Osram Sylvania Inc. Pre-irradiation (λ = 365 nm) was performed for about 5 minutes for all stress tests, and the samples for ML test were put in a dark environment for more than 4 hours to prevent the influence of afterglow on the ML properties. In addition, the synthesized phosphors were uniformly mixed into a PDMS matrix in a specific weight ratio and made into an elastomer. Finally, a homemade ink was prepared by dispersing the phosphor homogeneously in a certain ratio of glycerin and ethylene glycol and used for printing specific patterns.

2.2 Characterization

The powder X-ray diffraction (XRD) images were collected on a Rigaku D/max 2500 diffractometer (Cu Kα radiation, λ = 1.54178 Å) at 40 kV and 200 mA. All 2θ values ranged from 10 to 80° with a step size of 0.02°. The diffuse reflection spectra of the samples were collected on a HITACHI instrument (U-3900H) in the wavelength region of 250–800 nm. The UV emission spectra and ML spectra were obtained at RT using an ANDOR spectrometer (SR-500i-B1-R) equipped with an optical fiber and CCD camera (CCD-20259), and the intensity of the emitted ML was calibrated by a photomultiplier tube (C13796). The fluorescence decay curves were obtained using Edinburgh Instruments (FLS1000) with a μF2 flash lamp as the excitation resource. Compressive loading at different deformation rates was applied using a SHIMADZU (AGS-X) electronic universal testing machine. In addition, the images of the samples under the compressive load were obtained using a CANON (EOS 70D) digital camera. Thermoluminescence curves were collected on a YULI instrument (TOSL-3DS) in the temperature range from RT to 500 °C at a heating rate of 1 °C s−1 after pre-excitation under 365 nm for about 5 min.

3. Result and discussion

Fig. 1(a) shows the XRD patterns of the CaZnOS: x% Bi3+, x% Li+ (x = 0, 0.5, 1, 2, 3, 5, 7) phosphors prepared by solid-state reaction at a high temperature. It was evident that the XRD patterns agreed well with that of the CaZnOS single crystal, and that the phases were highly pure. Fig. 1(a) (right panel) shows the refined shift of the strongest diffraction peak in the 2θ range of 31° to 32°. The shift of the predominant peak from 31.62° (crystal plane (012)) to 31.36° (crystal plane (004)) can be clearly seen. This means that the preferred orientation of crystal growth was along the (004) plane with increasing Bi3+ concentration. This result is consistent with that reported in the literature.23 As known previously, the ionic radius of Ca2+ (CN = 6, r = 1.00 Å) was smaller than that of Bi3+ (CN = 6, r = 1.03 Å).24 Hence, as shown in Fig. 1(b), the unit cell volume gradually increased with an increase in the Bi3+ doping concentration, which might be attributed to the Bi3+ ions occupying the Ca2+ ionic sites.22Fig. 1(c) shows the crystal structure of CaZnOS. It belonged to the hexagonal system, the space group was P63mc, and the unit cell parameters were a = b = 3.757 Å, c = 11.401 Å, and V = 139.39 Å3. One Zn atom was coordinated with three S atoms and one O atom, forming a tetrahedron, while one Ca atom was bonded to three S atoms and three O atoms to form a distorted octahedron.25 In order to further study the effect of the addition of Bi3+ ions on the crystal structure, XRD Rietveld refinement was performed, and the results are shown in Fig. 1(c). The unit cell parameters a and c increased with increasing Bi3+ ion concentration, with c increasing greater than a (Fig. S1, ESI). This is consistent with the result that the intensity of the diffraction peak of the lattice plane (004) grew stronger. In addition, Fig. S2(a) and (b) (ESI) show the cross-section SEM image of the CaZnOS: 0.5% Bi3+, 0.5% Li+ and CaZnOS: 5% Bi3+, 5% Li+ phosphors embedded in the epoxy resin, and the crystal had changed from a nearly spherical shape to a cuboid.
image file: c9tc06543a-f1.tif
Fig. 1 (a) XRD patterns and enlarged peak shift of CaZnOS: x% Bi3+, x% Li+ (x = 0, 0.5, 1, 2, 3, 5, 7). (b) The relationship between the unit cell volumes of the CaZnOS:Bi3+,Li+ phosphors and Bi3+ concentration. (c) The crystal structure of CaZnOS:Bi3+,Li+ and the Rietveld result of the unit cell parameters.

As shown in Fig. 2(a), the diffuse reflectance spectra of CaZnOS: x% Bi3+, x% Li+ (0 ≤ x ≤ 7) revealed a broad intense absorption band at about 290–330 nm, originating from the valence band to conduction band transition of the CaZnOS host lattice. The optical gaps reduced from 4.08 to 3.37 eV with the Bi3+ concentration increasing from 0 to 7% (Fig. S3, ESI). These values are comparable with those reported by Hintzen et al. and Zhang et al.16,26 The optical gaps were calculated by the Kubelka–Munk function:27

F(R) = K/S = (1 − R)2/2R(1)
here R, S, K are the reflection, scattering coefficient and absorption, respectively. The absorption edge showed a slight red-shift with increasing Bi3+ concentration due to the incorporation of the larger Bi3+ ion into the CaZnOS host lattice, and the result is consistent with the increased unit cell volume. The red-shift of the absorption edge and the decrease in CaZnOS:Bi3+,Li+ band gap with increasing Bi3+ concentration could be ascribed to the spin-exchange interaction between the Bi3+ ions and the band electrons, which give rise to negative and positive corrections to the energy of the conduction and valence bands, respectively, leading to a redshift of the optical bandgap.28 In addition, the broad absorption band at about 370 nm was ascribed to the transition of Bi3+ from 1S03P1.29

image file: c9tc06543a-f2.tif
Fig. 2 (a) Diffuse reflectance spectra of CaZnOS:Bi3+,Li+. (b) Photoluminescence (PL) (λex = 290 nm) and (c) ML (load = 5000 N) spectra of CaZnOS: x% Bi3+, x% Li+ (x = 0.5, 1, 2, 3, 5, 7). (d) Integrated (PL and ML) intensity and the emission wavelength of CaZnOS:Bi3+,Li+ phosphors with different x values.

Fig. 2(b) and (c) show the emission spectra obtained with the excitation wavelength of 366 nm and a load of about 5000 N, respectively. The broad emission band with the peak at about 480 nm in both PL and ML was attributed to the transition of the Bi3+ ions from 3P1,01S0. In addition, it was evident that with increasing Bi3+ concentration, the emission intensity first increased and then decreased. However, the maximum PL and ML emission intensities of the phosphor were observed at x = 3 and x = 5, respectively (Fig. 2(d)), which was mainly caused by the subtle difference in the luminescence mechanisms of PL and ML. The PL and ML intensities decreased at higher Bi concentrations because of the concentration quenching effect. The peak value of the PL spectra initially showed a blue-shift (about 7 nm) on increasing the Bi3+ concentration from 0.5 to 3 mol%, and then, the wavelength was basically unchanged from 3 to 7 mo%, while the peak values of the ML spectra showed a continuous blue-shift (of about 13 nm) with the increase in Bi3+ ion concentration. This was because more electrons were being transferred from 3P1 to the ground state 1S0 with increasing Bi3+ ion concentration, leading to emission at slightly shorter wavelengths.22

Bismuth is known to exist in multiple valence states, including 0, +1, +2, and +3.30–33 Even with the same valence of Bi3+, it has exhibited variable emission colors spanning from ultraviolet to red.34 This phenomenon has been mainly attributed to the naked 6s electrons of the Bi3+ ions being susceptible to the surrounding crystal field.35Fig. 3(a) shows the configurational coordinate diagram of the Bi3+ ions; a Bi3+ ion has a ground state 1S0 from the 6s2 configuration, and four excited states 1P1, 3P2, 3P1, and 3P0 from the 6s26p configuration.36,37 In general, the 1S03P1, 1P1 electron transitions are allowed, while the 1S03P2 transition is spin forbidden, and 1S03P0 is completely forbidden.34 In addition, the energy levels of 3P1 and 3P0 are relatively close, and hence their emissions during the 3P11S0 and 3P01S0 transitions are relatively close.38

image file: c9tc06543a-f3.tif
Fig. 3 (a) The coordinate energy level diagram for luminescence from Bi3+ in CaZnOS. (b) The fluorescence decay curves of CaZnOS: 5% Bi3+, 5% Li+ (λex = 366 nm, λem = 485 nm) from 80 to 480 K. (c) Spectra and images of PL (λex = 290 nm), ML (load = 5000 N), TL (load = 3 N) and afterglow of CaZnOS: 5% Bi3+, 5% Li+. (d) The normalized PL (λex = 290, 366 nm and afterglow), ML (load = 2000, 5000 N and afterglow) and TL (load = 1, 3 N and afterglow) spectra of CaZnOS: 5% Bi3+, 5% Li+. (e) Changes in the normalized ML spectra of CaZnOS: 5% Bi3+, 5% Li+ at various loading rates, and (f) the enlarged ML spectra showing the red-shift.

In order to further study the transitions from the emission spectrum, the decay curves of the CaZnOS: 5% Bi3+, 5% Li+ phosphor with excitation at 366 nm were studied from 80 to 480 K (Fig. 3(b)). The corresponding fluorescence decay curves fitted the double-exponential equation well:39

I(t) = I0 + A1[thin space (1/6-em)]exp(−t/τ1) + A2[thin space (1/6-em)]exp(−t/τ2)(2)
where I and I0 are the luminescence intensities at times t and 0, respectively. t is the time, A1 and A2 are the values for different fitting constants, τ1 and τ2 are the decay times for the exponential components. According to the decay data, the value of average lifetime τ was calculated by the following formula:
τ = (A1τ12 + A2τ22)/(A1τ1 + A2τ2)(3)

Fig. 3(b) shows that the average lifetime of Bi3+ decreased from 66 μs to 44 ns when the temperature increased from 80 to 480 K. It has been reported that the radiative transition of 3P11S0 has a typical decay time of 10−8−10−6 s at high temperatures, while the transition of 3P01S0 has a long decay time of about several microseconds and happens at low temperatures.21,33,40 Therefore, the two transitions 3P11S0 and 3P01S0 may occur in the CaZnOS: 5% Bi3+, 5% Li+ phosphor at room temperature (about 840 ns). The corresponding τ1 and τ2 values are summarized in Table S1 (ESI).

In order to further study the luminescence properties of the CaZnOS:Bi3+,Li+ phosphors under load, we selected the CaZnOS: 5% Bi3+, 5% Li+ sample with the strongest ML intensity for further studies. Fig. 3(c) shows the different emission wavelengths and colors ranging from green to blue based on the methods to used excite the sample. The detailed results are shown in Fig. 3(d), where all the spectra were normalized. It was obvious that the peak value of the PL spectrum showed a red-shift between 290 and 366 nm excitation. In addition, the peak value of the ML and TL spectra presented a red-shift when the load increased from 2000 to 5000 N and the friction power from 1 to 3 N, respectively. However, all the afterglow spectra exhibited a blue-shift after 5 min irrespective of the method used to excite the sample, and Fig. S4 (ESI) shows the persistent decay curve of the CaZnOS: 5% Bi3+, 5% Li+ phosphor.

To further study the changes in the peak position, all the spectra in Fig. 3(d) were decomposed, as shown in Fig. S5 (ESI). The emission spectra showed that the relative intensities of the two fitted peaks of the composition spectrum changed due to the two excitation wavelengths 290 and 366 nm (Fig. S5(a) and (b), ESI). This indicated that different excitation wavelengths can cause changes in the Bi3+ luminescence properties; in this case, owing to the changes in the ratio of the 3P01S0 and 3P11S0 transitions at 2.50 and 2.67 eV, respectively. The afterglow was quite different in the two spectra because the afterglow and the lighting modes were essentially different between the two cases (Fig. S5(c), ESI). Fig. S5(d) and (e) (ESI) show that the ML spectra obtained from 2000 and 5000 N were different. The proportion of the peak value at about 2.47 eV had increased with an increased load. This meant that the dominant transition of Bi3+ ions changed from 3P11S0 to 3P01S0. A possible reason is that as the deformation of the sample increases with an increase in pressure, the crystal field environment of the Bi3+ ions changes significantly, and the luminescence of the Bi3+ ions is highly dependent on the crystal field environment. Triboluminescence is a special form of mechanoluminescence, the specific mechanism of which is still unknown. Since the touch area of friction is very small, it is equivalent to a considerable load at the beginning. Fig. S5(g) and (h) (ESI) show that it was consistent with the results of ML, in which the proportion of the peak value at about 2.49 eV increased with an increased load. As time went by, the energy of the deep traps was released and resulted in an afterglow, which was dominated by the 3P11S0 transition and exhibited a bluer emission.

As shown in Fig. 3(e), the peak value appeared changed when the loading rate increased under 5000 N load. As seen in the enlarged spectra (Fig. 3(f)), the peak position of the emission band exhibited a red-shift of 6 nm when the loading rate increased from 3 to 10 mm min−1, while the peak value no longer changed with a further rise in loading rate from 10 to 20 mm min−1. The reason for the red-shift of the spectrum is the ratio change of the two 3P01S0 and 3P11S0 transitions. This result is consistent with the red-shift of the spectrum with increasing load, as mentioned previously, and the effect of increasing the loading rate was equivalent to that of increased load. However, the peak position did not continue to change when the load reached a certain threshold when the loading rate reached more than 10 mm min−1.

Fig. 4(a) shows the compression ML responses of the typical x-value CaZnOS:Bi3+,Li+ phosphor pellets under different loads from 0 to 2000 N. The ML intensity increased slowly with the x value from 0.5 to 2, and then, the ML intensity was obviously enhanced for the x values 3 and 5, after which, the ML intensity decreased sharply for x = 7. The inset of Fig. 4(a) shows that the sheet with the CaZnOS: 5% Bi3+, 5% Li+ phosphor was much more sensitive to stress than the others. In addition, the ML intensity decreased with an increase in the number of load and release cycles (Fig. 4(b)). However, it is worth noting that the ML intensity could be fully recovered after UV irradiation for about 5 min, and the luminescence intensity was much stronger than that of the commercial powder (ZnS:Cu+) after 3 cycles of load application (Fig. S6, ESI). As shown in Fig. 4(c), the ML intensity of CaZnOS: 5% Bi3+, 5% Li+ was about 50 times higher than that of ZnS:Cu+. The ultra-high ML intensity and low ML threshold value indicated that CaZnOS: 5% Bi3+, 5% Li+ is a potential candidate for stress display or lighting and visualization of stress distribution for damage diagnosis.

image file: c9tc06543a-f4.tif
Fig. 4 (a) Comparison of the ML intensities of CaZnOS: x% Bi3+, x% Li+ (x = 0.5, 1, 2, 3, 5, 7) with applied loads up to 2000 N, and the inset shows the sensitivity levels for the different x values. (b) ML recovery behavior of CaZnOS: 5% Bi3+, 5% Li+ (i) with 2000 N loading for 10 cycles, (ii) ML recovery behavior after UV irradiation for about 5 min. (c) Comparison of ML intensities of the CaZnOS: 5% Bi3+, 5% Li+ and ZnS:Cu+ sheets during in the load range 0 to 2000 N, and the enlarged ML intensity of ZnS:Cu+ is in the inset. (d) The stretch-release ML spectra of CaZnOS: 5% Bi3+, 5% Li+ with a stretch rate of 16 mm s−1 and images of the elastomer.

Additionally, Fig. 4(d) shows the stretch-release ML spectrum of the elastomer with the CaZnOS: 5% Bi3+, 5% Li+ phosphor, wherein the emission peak at 472 nm originating from the 3P1,01S0 transition of Bi3+ is evident. The right graph in Fig. 4(d) shows an intense blue-green light emission under a stretch of 7 N.

To further clarify the relationship between the different traps and ML, the thermoluminescence (ThL) glow curves of CaZnOS: 5% Bi3+, 5% Li+ were measured, as shown in Fig. 5. Two broad peaks at 465 and 488 nm are evident in the right picture of Fig. 5(b). These were attributed to the 3P01S0 and 3P11S0 transitions of the Bi3+ ions, and notably, this result is consistent with the previous results from the decay and ML curves. The ThL curve of CaZnOS: 5% Bi3+, 5% Li+ was decomposed into four peaks at 60, 95, 150, 195 °C by employing a general kinetic order equation (Fig. 5(c)). The fitting results suggested that four different depth traps existed in CaZnOS:Bi3+,Li+, and they possibly originated from the intrinsic defects in the host as the Ca2+ ions in CaZnOS are unequally replaced by the Bi3+ ions. Furthermore, the depths of the four traps were evaluated to be about 0.68, 0.77, 0.91 and 1.2 eV, respectively. Moreover, the trap at 0.68 eV led to long-persistent luminescence.41 They were fitted according to the following equation:42

image file: c9tc06543a-t1.tif(4)
where E is the energy level depth (activation energy) of the trap, S is the frequency factor, b is the kinetic order of the equation, k is the Boltzmann constant, β is the heating rate (1 K s−1), T0 (K) is the initial temperature, and n0 is the concentration of the trapped electrons at T0. The frequency factor S should be in the range of 1010–1017 for reasonable fitting results that can be used for further discussion. The values of the corresponding parameters S, n0, and b are summarized in Table S2 (ESI).

image file: c9tc06543a-f5.tif
Fig. 5 (a) ThL glow 3D curves, and (b) contour plot of the CaZnOS: 5% Bi3+, 5% Li+ sample as a function of temperature and wavelength with a heating rate of 1 °C s−1 and an excitation wavelength of 365 nm. (c) The fitting results of ThL glow at 487.4 nm wavelength. (d) Schematic of the internal electric field generated in CaZnOS. (e) The ML mechanism of Bi3+ ions in CaZnOS.

As mentioned previously, CaZnOS crystallizes in the polar non-centrosymmetric hexagonal space group of P63mc and exhibits piezoelectric performance. It shows significant anisotropy in the c-axis direction and perpendicularity to the c-axis direction. Hence, it naturally exhibits a piezoelectric effect under the influence of external force.43–45Fig. 5(d) shows that when there was no external force, the charge centers of the cation and the anion coincided with each other, whereas when an external force was applied, the charge centers of the cation and anion moved slightly, generating a dipole moment. The superposition of the dipole moments of all the crystal units in the stressed region produced a built-in electric field along the stress direction at the macroscopic level, and the internal electric field caused a change in the local band gap, as shown in Fig. 5(e). The electrons are detrapped near the conduction band, and escape into the conduction band when a load is applied, and when they nonradiatively recombine with the holes, the detrapped electrons are released to the bottom of the conduction band and transfer the energy to the Bi3+ ions.46 Finally, the blue-green emission is observed owing to the transition of Bi3+ from 3P1,01S0. It is easy to notice that the magnitude of strain determines the strength of the internal electric field and the number of electrons detrapped. Therefore, a larger stimulation results in a stronger ML intensity and easier release of the deeper trap energies, which together lead to an increase in the ratio of 3P11S0 transition.

Fig. 6(a) illustrates the photographs of the CaZnOS: 5% Bi3+, 5% Li+ mixture in epoxy resin under 365 nm light excitation and afterglow after about 5, 180, 420, 600 and 900 s, respectively. It is notable that the color changed from cyan-blue to navy blue due to the 3P11S0 and 3P01S0 transitions of Bi3+. Persistent luminescence could be observed by the naked eye for more than 900 s. Inspired by this interesting phenomenon, a “panda” was printed with synthesized ultrafine phosphor, which evenly dispersed in homemade ink (glycerin and ethylene glycol), as a security label. As shown in Fig. 6(b), very shallow marks could be observed under fluorescent light at room temperature, and an intense navy blue afterglow was emitted under 60 °C heating with weak light. In addition, the homemade security label could emit an intense cyan-blue light under 365 nm irradiation and the faint dark blue afterglow could be seen even after the irradiation source was removed. Surprisingly, an intense navy blue afterglow could be observed again at 60 °C. This phenomenon of slight heating causing a remarkable enhancement of the afterglow intensity is consistent with the position of the strongest thermoluminescence peak at 60 °C (Fig. 5). Thermal energy excites the electrons to release energy from shallow traps at 60 °C, leading to the intense navy blue emission. Moreover, the writing characteristics of different people could be distinguished by using the thin film as a stress sensor, which was fabricated by dispersing the phosphor in PDMS, as shown in Fig. 6(c). Based on the handwritten “ML” patterns in the photographs, the corresponding intensity distribution maps were collected. After 5 minutes of de-irradiation, they had a bright afterglow, which vanished completely at 72 minutes. We recorded the stress distribution maps and the photographs of writing both when the PDMS sheet had the afterglow and with no afterglow. Due to the afterglow and the different emission color exhibited by the pattern “ML”, the handwriting pressure appeared more precise and brighter, as on the left side of Fig. 6(c). Therefore, the handwriting habits of signees can be recorded and distinguished using the stress sensor made of CaZnOS:Bi3+,Li+, which shows great potential for applications in stress-based anti-counterfeiting technologies.

image file: c9tc06543a-f6.tif
Fig. 6 (a) Optical images irradiated under 365 nm UV light for 1 min and the afterglow images from 5 to 900 s. (b) Photograph of a “panda” printed with the home-made ink with the CaZnOS:Bi3+,Li+ powder dispersed in glycerin and ethylene glycol. The photograph represents images under ambient light, weak light at 60 °C, UV light (365 nm), and the afterglow at room temperature and 60 °C in darkness, respectively. (c) Demonstration of the recording of methods used and make changes as required beginning signature patterns.

4. Conclusions

A series of CaZnOS: x% Bi3+, x% Li+ phosphors were prepared by a solid-state method at a high temperature. As the Bi3+ ions concentration increased, the unit cell volume increased and the crystals grew in the preferred-orientation. The sample emitted blue-green color at about 485 nm due to the Bi3+ transition of 3P1,01S0 under excitation at 366 nm or external stress. However, the afterglow exhibited a blue color of about 465 nm at 5 min after the removal of the stimuli. In addition, the wavelengths of PL and ML emission peaks exhibited a slight blue-shift of about 7 and 13 nm, respectively, with the increase of Bi3+ ion doping concentration. Furthermore, for CaZnOS: 5% Bi3+, 5% Li+, the decay time was about 840 ns, indicating the existence of two transitions of Bi3+, 3P11S0 and 3P01S0. Further research revealed that the peak position was red-shifted of different degrees when stimulated with greater energy, such as the excitation wavelength of 366 nm, about 5,000 N pressure and 3 N friction. Moreover, the peak position also exhibited a red-shift of about 6 nm with an increasing load rate from 3 to 20 mm min−1. The change of the ratio from 3P11S0 to 3P01S0 was the main cause of this red-shift. The phosphor sheet made with the epoxy resin exhibited a superior sensitivity of about 100 N, and stretching the PDMS sheet with the CaZnOS: 5% Bi3+, 5% Li+ phosphor resulted in bright blue-green light. Moreover, the ML intensity of the CaZnOS: 5% Bi3+, 5% Li+ phosphor was 50 times that of the commercial powder ZnS:Cu+, demonstrating that CaZnOS:Bi3+,Li+ is a potential candidate for pressure sensors, damage diagnosis and display imaging applications. The ThL curves illustrated that there were four types of traps in CaZnOS:Bi3+,Li+, and their activation energies were evaluated to be about 0.66, 0.77, 0.91 and 1.2 eV, respectively. Furthermore, the ML mechanism diagram shows that the magnitude of strain determines the strength of the internal electric field and thus the amount of detrapped electrons, and high strain will cause the energy of deeper traps to be released more easily and increase the ratio of the 3P11S0 transition.

Conflicts of interest

The authors declare no competing financial interests.


This work was supported by the National Natural Science Foundation of China under Grant No. 51772185, 11104298, 51402184, U1832159, 2017C110028, China Postdoctoral Science Foundation (Grant No. 2019M651469) and National Research Foundation of Korea (NRF) grant funded by the Ministry of Education (No. 2016R1D1A1B03933488 and 2016R1A6A1A03012877) and the Korea government (MSIT) (No. 2019R1A2C1002844). This work was supported by Shanghai Synchrotron Radiation Facility.


  1. N. C. Eddingsaas and K. S. Suslick, Nature, 2006, 444, 163 CrossRef CAS PubMed.
  2. A. Feng and P. F. Smet, Materials, 2018, 11, 484 CrossRef PubMed.
  3. X. D. Wang, H. L. Zhang, R. M. Yu, L. Dong, D. F. Peng, A. H. Zhang, Y. Zhang, H. Liu, C. F. Pan and Z. L. Wang, Adv. Mater., 2015, 27, 2324–2331 CrossRef CAS PubMed.
  4. S. M. Jeong, S. Song, K. l. Joo, J. Kim, S. H. Hwang, J. Jeong and H. Kim, Energy Environ. Sci., 2014, 7, 3338–3346 RSC.
  5. N. Terasaki, H. W. Zhang, H. Yamada and C. N. Xu, Chem. Commun., 2011, 47, 8034–8036 RSC.
  6. C. Wu, S. S. Zeng, Z. F. Wang, F. Wang, H. Zhou, J. C. Zhang, Z. P. Ci and L. Y. Sun, Adv. Funct. Mater., 2018, 28, 1803168 CrossRef.
  7. S. M. Jeong, S. Song and H. Kim, Nano Energy, 2016, 21, 154–161 CrossRef CAS.
  8. W. L. Li, Q. Y. Huang, Z. Mao, Q. Li, L. Jiang, Z. L. Xie, R. Xu, Z. Y. Yang, J. Zhao, T. Yu, Y. Zhang, M. P. Aldred and Z. G. Chi, Angew. Chem., Int. Ed., 2018, 57, 12727–12732 CrossRef CAS PubMed.
  9. C. N. Xu, T. Watanabe, M. Akiyama and X. G. Zheng, Appl. Phys. Lett., 1999, 74, 2414–2416 CrossRef CAS.
  10. D. Tu, C. N. Xu, Y. Fujio and A. Yoshida, Light: Sci. Appl., 2015, 4, e356 CrossRef CAS.
  11. S. M. Jeong, S. Song, S. K. Lee and B. Choi, Appl. Phys. Lett., 2013, 102, 051110 CrossRef.
  12. S. M. Jeong, S. Song, S. K. Lee and N. Y. Ha, Adv. Mater., 2013, 25, 6194–6200 CrossRef CAS PubMed.
  13. K. S. Sohn, M. Y. Cho, M. Kim and J. S. Kim, Opt. Express, 2015, 23, 6073–6082 CrossRef CAS PubMed.
  14. D. Tu, C. N. Xu, A. Yoshida, M. Fujihala, J. Hirotsu and X. G. Zheng, Adv. Mater., 2017, 29, 4 Search PubMed.
  15. J. Li, C. N. Xu, T. Dong, X. N. Chai, X. S. Wang, L. S. Liu and E. Kawasaki, Acta Mater., 2018, 145, 462–469 CrossRef CAS.
  16. C. J. Duan, A. C. A. Delsing and H. T. Hintzen, Chem. Mater., 2009, 21, 1010–1016 CrossRef CAS.
  17. J. C. Zhang, L. Z. Zhao, Y. Z. Long, H. D. Zhang, B. Sun, W. P. Han, X. Yan and X. S. Wang, Chem. Mater., 2015, 27, 7481–7489 CrossRef CAS.
  18. R. P. Cao, Y. L. Cao, T. Fu, S. H. Jiang, W. S. Li, Z. Y. Luo and J. W. Fu, J. Alloys Compd., 2016, 661, 77–81 CrossRef CAS.
  19. J. Y. Zheng, M. Y. Peng, F. W. Kang, R. P. Cao, Z. J. Ma, G. P. Dong, J. R. Qiu and S. H. Xu, Opt. Express, 2012, 20, 22569–22578 CrossRef CAS PubMed.
  20. R. P. Cao, M. Y. Peng and J. R. Qiu, Opt. Express, 2012, 20, A977–A983 CrossRef PubMed.
  21. R. P. Cao, T. Fu, H. D. Xu, W. J. Luo, D. D. Peng, Z. Q. Chen and J. W. Fu, J. Alloys Compd., 2016, 674, 51–55 CrossRef CAS.
  22. D. J. Pan, Y. L. Yang, X. C. Yang, T. Xu, Y. Y. Cui, X. Y. Sun, H. Feng, L. Liu, Y. F. Shen, J. T. Zhao and Z. J. Zhang, J. Alloys Compd., 2018, 742, 1037–1045 CrossRef CAS.
  23. D. Tu, C. N. Xu and Y. Fujio, J. Adv. Dielectr., 2014, 4, 1450017 CrossRef CAS.
  24. R. D. Shannon, Acta Crystallogr., 1976, 32, 751–767 CrossRef.
  25. T. Sambrook, C. F. Smura, S. J. Clarke, K. m. Ok and P. S. Halasyamani, Inorg. Chem., 2007, 46, 2571–2574 CrossRef CAS PubMed.
  26. Z. J. Zhang, A. Feng, X. Y. Chen and J. T. Zhao, J. Appl. Phys., 2013, 114, 213518 CrossRef.
  27. J. Zhang, L. Zhang, Z. Tang, Z. Zhang and T. Wang, Ceram. Int., 2003, 29, 583–586 CrossRef CAS.
  28. F. W. Kang, H. S. Zhang, L. Wondracek, X. B. Yang, Y. Zhang, D. Y. Lei and M. Y. Peng, Chem. Mater., 2016, 28, 2692–2703 CrossRef CAS.
  29. Y. L. Yang, Y. Zhou, D. J. Pan, Z. J. Zhang and J. T. Zhao, J. Lumin., 2019, 206, 578–584 CrossRef CAS.
  30. M. Y. Peng, C. Zollfrank and L. Wondraczek, J. Phys.: Condens. Matter, 2009, 21, 285106 CrossRef PubMed.
  31. X. G. Meng, J. G. Qiu, M. Y. Peng, D. P. Chen, Q. Z. Zhao, X. W. Jiang and C. S. Zhu, Opt. Express, 2005, 13, 1628–1634 CrossRef CAS PubMed.
  32. L. Seijo and Z. Barandiaran, Phys. Chem. Chem. Phys., 2014, 16, 17305–17314 RSC.
  33. F. W. Kang, M. Y. Peng, S. H. Xu, Z. J. Ma, G. P. Dong and J. R. Qiu, Eur. J. Inorg. Chem., 2014, 1373–1380 CrossRef CAS.
  34. P. Boutinaud, Inorg. Chem., 2013, 52, 6028–6038 CrossRef CAS PubMed.
  35. P. P. Dang, S. S. Liang, G. G. Li, H. Z. Lian, M. M. Shang and J. Lin, J. Mater. Chem. C, 2018, 6, 9990–9999 RSC.
  36. R. P. Cao, J. L. Zhang, W. D. Wang, Q. L. Hu, W. S. Li, W. Ruan and H. Ao, Luminescence, 2017, 32, 908–912 CrossRef CAS PubMed.
  37. P. P. Dang, S. S. Liang, G. G. Li, Y. We, Z. Y. Cheng, H. Z. Lian, M. M. Shang, A. A. Al Kheraif and J. Lin, Inorg. Chem., 2018, 57, 9251–9259 CrossRef CAS PubMed.
  38. Y. Zorenko, V. Gorbenko, T. Voznyak, V. Jary and M. Nikl, J. Lumin., 2010, 130, 1963–1969 CrossRef CAS.
  39. Y. Wei, G. C. Xing, K. Liu, G. G. Li, P. P. Dang, S. S. Liang, M. Liu, Z. Y. Cheng, D. Y. Jin and J. Lin, Light: Sci. Appl., 2019, 8, 15 CrossRef PubMed.
  40. R. P. Cao, G. J. Quan, Z. H. Shi, Z. Y. Luo, Q. L. Hu and S. L. Guo, J. Lumin., 2017, 181, 332–336 CrossRef CAS.
  41. F. W. Kang, X. B. Yang, M. Y. Peng, L. Wondraczek, Z. J. Ma, Q. Y. Zhang and J. R. Qiu, J. Phys. Chem. C, 2014, 118, 7515–7522 CrossRef CAS.
  42. Y. L. Liu, B. F. Lei and C. S. Shi, Chem. Mater., 2005, 17, 2108–2113 CrossRef CAS.
  43. R. Chen, J. Appl. Phys., 1968, 40, 570–585 CrossRef.
  44. Z. L. Wang and J. H. Song, Science, 2006, 312, 242–246 CrossRef CAS PubMed.
  45. Z. L. Wang, Piezoelectric electronics and piezoelectric optoelectronics, Science Press, CHN, 2012 Search PubMed.
  46. Y. Chen, Y. Zhang, D. Karnaushenko, L. Chen, J. H. Hao, F. Ding and O. G. Schmidt, Adv. Mater., 2017, 29, 1605165 CrossRef PubMed.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c9tc06543a
Yunling Yang and Qianli Li contributed equally.

This journal is © The Royal Society of Chemistry 2020