Introduction

Photon UC in lanthanide-doped nanomaterials has drawn much attention recently due to their superior spectroscopic properties [1, 2]. As the most significant near-infrared (NIR) to visible UC structures, the lanthanide-sensitized UC based on the ET from Yb3+ to Er3+(Tm3+/Ho3+) in β-NaYF4 nanocrystals has been studied intensively owing to its promising applications in color display [3, 4], super-resolution nanoscopy [5, 6], security printing [7, 8], laser materials [9,10,11], and biological luminescent labels [12,13,14]. It is well-known that lanthanide ions have an abundance of 4fN electronic states, which typically generate multi-band emissions [15]. However, multi-band emissions preclude the quantitative imaging of samples targeted with multiple upconverting probes and reduce the sensitivity of imaging [16]. Therefore, some efforts have been made to achieve single-band UC emissions [17,18,19]. For example, the introduction of transition metals (Mn2+, etc.) into Yb/Er co-doped nanomaterials can enhance the R/G (650-to-545 nm) ratio and achieve single-red-band emission because of the strongly ET process between Er3+ and Mn2+ [20,21,22,23,24].

To date, some studies have been performed to investigate the Mn2+-doped Yb/Er nanocrystals for applications in bioimaging [20, 25], sensor [26,27,28], and biomarker detection [29]. In fact, compared to the nanocrystals, microcrystals fascinate more advantages for applications in micro-optoelectronic devices [30], volumetric color display [31, 32], and microlasers [11] based on their high crystallinity and luminescent efficiency [33]. Consequently, it is very important for us to study the UC luminescence properties of the microcrystals. However, most present measurements of the samples were executed by powders or in organic solvents, which can cause severe overheating problems and be influenced by adjacent microcrystals [34]. Thus, exploring the UC luminescence and tunable color from the single microcrystal can effectively avoid the influence of the environment and broaden its further applications in micro-optoelectronic devices.

Moreover, excited by 980 nm CW laser, Yb/Er co-doped materials usually emit red (650 nm) and green (525 and 545 nm) UC emissions, as well as a weaker blue (410 nm) emission. Generally, the red and green UC emissions dominate the spectra, and the blue emission remains relatively weak. In comparison with these three UC emissions, other UC emissions are rarely observed in Yb/Er co-doped materials. Previously, we have observed a newly emerged 560 nm (2H9/2 → 4I13/2) emission from a single β-NaYF4:Yb/Er microcrystal under saturated excitation [35]. As excitation intensity increases, the 560-nm emission increases rapidly and further exceeds the traditional green emission (545 nm). However, for Yb/Er/Mn tri-doped materials, the transition of 2H9/2 → 4I13/2 (560 nm) in Er3+ also acts as an ET channel populating the level 4T1 of Mn2+, which, to the best of our knowledge, has not been reported or explored so far. Thus, for Yb/Er/Mn tri-doped materials, the suppressing of the new green (560 nm) UC emission and the tuning of R/G ratio remain largely unknown. Therefore, as discussed above, exploiting the UC emissions from single β-NaYF4:Yb/Er/Mn microcrystal can help us understand the tuning of new R/G ratio and expand its scope of applications in micro-optoelectronic devices.

In this work, we have synthesized Mn2+-doped β-NaYF4:Yb/Er microcrystals through a simple one-pot hydrothermal method. The UC emission properties and the relevant luminescence color from a single microcrystal were investigated using a high-performance luminescence collection system including an inverted fluorescence microscope with a × 100 objective lens (NA = 1.4). Excited by a 980-nm CW laser, the luminescence color can be tuned from green to red when gradually increasing the doped Mn2+ ions from 0 to 30 mol%. The tuning R/G ratios for traditional 650 to 545 nm and new 650 to 560 nm have been discussed in detail. The mechanism of the tunable UC emission color was also demonstrated based on the ET process between Mn2+ and Er3+.

Methods

Materials

The raw materials were purchased from Aladdin (China): Y2O3 (99.99% metals basis), Yb2O3 (99.99% metals basis), Er2O3 (99.99% metals basis), MnCl2·4H2O (99% metals basis), nitric acid (HNO3, analytical reagent), ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA-2Na, analytical reagent), sodium hydroxide (NaOH, analytical reagent), ammonium fluoride (NH4F, analytical reagent). All the chemicals were used as received directly without further purification.

Synthesis of β-NaYF4 Microcrystals

We synthesized the β-NaYF4:Yb/Er/Mn (20/2/× mol%) microcrystals through a modified hydrothermal method. The Y2O3, Yb2O3, and Er2O3 powders were dissolved in dilute nitrate solution and heated to remove the residual nitrate, yielding a clear solution of Ln (NO3)3 (0.2 M). In a typical procedure, the EDTA-2Na (1 mmol) and NaOH (6 mmol) were mixed with 13.5 mL deionized (DI) water under continuously stirring in a flask yielding a clear solution. Then, 5 mL of MnCl2 (0.2 M) and Ln (NO3)3 (0.2 M) aqueous solutions, 8 mL of NH4F (2.0 M) aqueous solutions, and 7 mL of dilute hydrochloric acid (1 M) were injected into the flask. The mixtures were stirred for 1.5 h and then transferred into a 50 mL Teflon-lined autoclave and heated at 200 °C for 40 h. The as-obtained precipitates were collected by centrifugation, washed with DI water and ethanol for several times, and finally dried in air at 40 °C for 12 h. Microcrystals with different concentrations of Mn2+ can be achieved by varying the volume of MnCl2 aqueous solutions (the total Y3+, Yb3+, Er3+, and Mn2+ ions content were kept constant at 1 mmol).

Physical Characterization

X-ray diffraction (XRD) patterns of the microcrystals were measured using X-ray diffractometer with Cu K radiation at 40 kV and 200 mA (Rigaku). The morphology of the β-NaYF4:Yb/Er/ (20/2/× mol%) microcrystals was characterized by scanning electron microscope (SEM) (S4800, Hitachi).

Photoluminescence Measurements

For photoluminescence experiments, the 980-nm CW laser was introduced into an inverted microscope (Observer A1, Zeiss) and focused on the microcrystals by using a × 100 objective lens (NA = 1.4). The diameter of the excitation spot was estimated to be ~ 2.0 μm. The UC luminescence was collected by the same objective lens and then delivered to a spectrometer (SR-500I-B1, Andor) equipped with a charge-coupled device (CCD) (DU970N, Andor) for analysis. The luminescence color of the single microcrystal was recorded by using a camera (DS-Ri2, Nikon). The UC luminescence lifetime was measured by using a digital oscilloscope (1 GHz, InfiniiVsionDSOX6002A, KEYSIGHT) and a nanosecond pulsed laser (with a pulse duration of 20 ns and a repetition rate of 10 Hz) as the excitation source. All the measurements were performed at room temperature.

Results and Discussion

The typical morphologies of the β-NaYF4:Yb/Er/Mn (20/2/x mol%) microcrystals are characterized by SEM images, as shown in Fig. 1a–e. It indicates that the microcrystals exhibit a pure hexagonal phase of morphology and uniform size with the diameters of ~ 3.5 μm and lengths of ~ 13 μm. Notably, the length of the microcrystals slightly reduces to 10 μm as the doping Mn2+ ions increase to 30 mol%. Figure 1g shows the XRD patterns of the β-NaYF4:Yb/Er microcrystals doped with different concentrations of Mn2+ ions. All the diffraction peaks were in good agreement with the standard hexagonal phase of NaYF4 (JCPDS No. 16-0334). Importantly, as the doping Mn2+ ions increase, the microcrystals still maintain a hexagonal phase and no other impurity peaks are observed. It reveals that the addition of Mn2+ ions has no influence on the morphology and crystal phase of the β-NaYF4 microcrystals. Furthermore, some diffraction peaks firstly shift slightly towards to higher angles as the doping Mn2+ ions gradually increase from 0 to 10 mol%, and then drift back to lower angles when further increasing the Mn2+ ions up to 30 mol%. The results probably indicate that the smaller Mn2+ ions (r = 1.10 Å) mainly occupy the larger Na+ (r = 1.24 Å) sites at low Mn2+ concentrations (less than 10 mol%), and then insert the Y3+ (r = 1.159 Å) sites in NaYF4 host lattice with the Mn2+ ions increasing up to 30 mol% [4, 36, 37]. We have also performed the compositional analysis on the β-NaYF4 microcrystals by EDS, as shown in Fig. 1g–h. The EDS analysis confirms the presence of Na, F, Y, Yb, and Er elements in Mn-free β-NaYF4:Yb/Er microcrystals (Fig. 1g). In comparison, the Mn element is found in β-NaYF4:Yb/Er microcrystals doping with 30 mol% Mn2+ ions (Fig. 1h), indicating the Mn2+ ions are well embedded in the NaYF4 host lattice.

Fig. 1
figure 1

SEM micrographs of the NaYF4:Yb/Er (20/2 mol%) microcrystals doping with a 0, b 5, c 10, d 20 and e 30 mol% Mn2+ ions, respectively. f XRD patterns of the NaYF4 microcrystals doping with different concentrations of Mn2+ ions. EDS analysis of the NaYF4:Yb/Er (20/2 mol%) microcrystals doping with g 0 and h 30 mol% Mn2+ ions, respectively

Figure 2a shows the UC emissions from a single β-NaYF4:Yb/Er (20/2 mol%) microcrystal doped with different amounts of Mn2+ ions under a relatively low excitation intensity (1.59 kW cm−2). The insets show the single microcrystal and its corresponding luminescence color observed from the microscope. Three major emission bands are indexed in the spectra, which are ascribed to the transitions of 2H9/2 → 4I15/2 (410 nm), (2H11/2/4S3/2) → 4I15/2 (525 and 545 nm), and 4F9/2 → 4I15/2 (650 nm) from Er3+, respectively. For the Mn-free single microcrystal, the green (545 nm) emission dominates the emission spectrum, leading to the single microcrystal lighting up with a green luminescence color. With increasing the doping Mn2+ ions, the red (650 nm) emission grows remarkably and gradually exceeds the green emission and dominates the spectrum in the end as the dopant Mn2+ ions reach 30 mol%. Thus, the luminescence color can be tuned from green to yellow and finally becomes red. Figure 2b displays the calculated CIE chromaticity coordinates based on the UC emission spectrums in Fig. 2a. It is obvious that the UC luminescence color changes from green to red with the doping Mn2+ ions increasing from 0 to 30 mol%. As shown in Fig. 2c, it is interesting to observe several new UC emission bands as the excitation intensity increases up to 95.52 kW cm−2. These new UC emissions can be detected both in Mn-free and highly Mn2+-doped microcrystals. As demonstrated in our previous study [35], these new UC emissions originate from the transitions of 4G11/2 → 4I15/2 (382 nm), 4F5/2 → 4I15/2 (457 nm), 2K15/2 → 4I13/2 (472 nm), 4G11/2 → 4I15/2 (506 nm), 2H9/2 → 4I13/2 (560 nm), and 4G11/2 → 4I11/2 (618 nm) in Er3+, respectively. It is noteworthy that the newly emerged UC emissions can be observed regardless of the Mn2+ concentration and the new 560 nm emission is always stronger than the traditional green (545 nm) emission.

Fig. 2
figure 2

a UC emission spectra from a single β-NaYF4:Yb/Er (20/2 mol%) microcrystal doping with 0, 10, and 30 mol% Mn2+ ions under the excitation intensity of 1.59 kW cm−2. b CIE chromaticity coordinates for the UC luminescence of the single β-NaYF4:Yb/Er (20/2 mol%) microcrystals doping with different amounts of Mn2+ ions. c UC emission spectra from a single β-NaYF4:Yb/Er (20/2 mol%) microcrystal doping with 0, 10, and 30 mol% Mn2+ ions under the excitation intensity of 95.52 kW cm−2

To clearly identify these new UC emissions, we demonstrated the UC emissions from the single β-NaYF4:Yb/Er/Mn (20/2/10 mol%) microcrystal under different excitation intensities, as shown in Fig. 3a. At the excitation intensity of 1.59 kW cm−2, the red emission (650 nm) is much stronger than the traditional green emission (545 nm), and the new 560 nm UC emission is lower than the traditional green (545 nm) emission. Furthermore, the UC emissions centered at 382, 506, and 472 nm can be distinguished from the spectra. When increasing the excitation intensity up to 9.55 kW cm−2, the 560-nm emission exceeds the 545 nm and becomes comparable to the red emission (650 nm). Moreover, the 506- and 472-nm emissions become more efficient. If we further increase the excitation intensity up to 31.84 kW cm−2, the 560-nm emission increases dramatically and exceeds the traditional red emission (650 nm). This is different from the previous reports in which doping the Mn2+ ions only facilitated the enhancement red emission and in which the new 560-nm UC emission was not observed. Meanwhile, the newly emerging emission bands at 382, 506, and 472 nm further increase with the excitation intensity rising to 95.52 kW cm−2. In Fig. 3b, we calculated the R/G ratios for a single β-NaYF4:Yb/Er/Mn (20/2/10 mol%) microcrystal under different excitation intensities. The ratio (560 to 545 nm) increases from ~ 0.97 to 1.96 with the excitation intensities varying from 1.59 to 95.52 kW cm2. However, the traditional R/G ratio (650 to 545 nm) rises from 1.27 to 1.72 and the new R/G ratio (650 to 560 nm) decreases from 1.31 to 0.87 as the excitation intensities increase. Figure 3c demonstrates the dependence of the UC emission intensity on the excitation intensity for the single β-NaYF4:Yb/Er/Mn (20/2/10 mol%) microcrystal. Under low power excitation, the slopes of the four UC emissions are all close to ~ 2. Moreover, these slopes become less than 1 under high power excitation, which should be attributed to the saturation effect [38,39,40,41].

Fig. 3
figure 3

a UC emission spectra from a single β-NaYF4:Yb/Er/Mn (20/2/10 mol%) microcrystal irradiated with different excitation intensities. b The UC emission intensity ratios for a single β-NaYF4:Yb/Er/Mn (20/2/10 mol%) microcrystal as a function of the excitation intensities. c The dependence of the UC emission intensity on the excitation intensity for a single β-NaYF4:Yb/Er/Mn (20/2/10 mol%) microcrystal

Figure 4 shows the tunable UC emission intensity ratios for the single β-NaYF4:Yb/Er (20/2 mol%) microcrystal doping with different concentrations of Mn2+ under low and high excitation intensity. The ratio (560 to 545 nm) is less than 1 under low power excitation but becomes larger than 1.5 at high power excitation. At low power excitation (Fig. 4a), the traditional R/G (650 to 545 nm) ratio is basically consistent with the new R/G (650 to 560 nm) ratio. These two ratios (650 to 545 and 650 to 560 nm) start from ~ 0.87 and then gradually increase to ~ 2.7 with the doping Mn2+ ions varying from 0 to 30 mol%. Nevertheless, these two ratios become different under high excitation intensity (Fig. 4b). The traditional ratio (650 to 545 nm) rises from ~ 1.2 to 3.4, whereas the new ratio (650 to 560 nm) increases from 0.66 to 2.15 when the doped Mn2+ ions growing from 0 to 30 mol%. It reveals that the traditional and new R/G ratios exhibit different tunable trends under low and high excitation intensity. The newly emerged 560-nm UC emission changes the tunability of multicolor UC emissions, which is different from the previously reported results [20,21,22,23,24].

Fig. 4
figure 4

The UC emission intensity ratios for a single β-NaYF4:Yb/Er (20/2 mol%) microcrystal doping with different concentrations of Mn2+ ions under the excitation intensity of (a) 1.59 kW cm−2 and (b) 95.52 kW cm−2

To further understand the principle of tuning multicolor in the β-NaYF4:Yb/Er/Mn microcrystals, we examined the energy level diagram for Yb3+, Er3+, and Mn2+ ions. As shown in Fig. 5, the mechanism of the populating processes, UC emissions, nonradiative transitions, and ET processes are also displayed. For β-NaYF4:Yb/Er microcrystals, Yb3+ ions absorb the 980 nm incident light, and then populate the Er3+ ions from the ground state to the excited states through ET processes. Two possible approaches can promote the higher excited levels of Er3+. One is through the green-emitting levels (4S3/2 and 2H11/2) populating the 4G,2K manifold, and the other is through the red-emitting level (4F9/2) populating the level of 2H9/2. Once the excited levels of Er3+ are populated, significant UC emissions will be generated. Thus, the traditional green (545 nm) and red (650 nm) UC emissions can be easily observed, which exhibit highly efficient UC emissions and have been widely studied. Generally, for 560-nm emission (2H9/2 → 4I13/2), the level of 2H9/2 can be populated through the red-emitting level (4F9/2), or through the green-emitting levels (4S3/2, 2H11/2) populating the 4G,2K manifold (then followed nonradiative transition to the level of 2H9/2). Similarly, the 382-, 410-, 457-, 472-, and 506-nm UC emissions are based on the same principle with the population of higher emitting levels of Er3+. Moreover, the 618 nm emission originates from populating the 4G,2K manifold and transition of 4G11/2 → 4I11/2.

Fig. 5
figure 5

Schematic energy level diagrams for Yb3+, Er3+, and Mn2+ under the 980-nm CW laser excitation, the mechanism of the possible ET, non-radiative transitions, and UC emissions are also presented in the diagram

In addition, for Mn2+-doped β-NaYF4:Yb/Er microcrystals, the UC luminescence color can be changed from green to red. As shown in Fig. 5, there are two possible ET routes from Er3+ to Mn2+: one is from the 2H9/2 → 4I13/2 transition of Er3+ to 6A1 → 4T1 transition of Mn2+ (process ET1), and another is from the 4S3/2 → 4I15/2 transition of Er3+ to 6A1 → 4T1 transition of Mn2+ (process ET2). These two processes (ET1 and ET2) would decrease the 560- and 545-nm UC emissions. When the level 4T1 of Mn2+ is populated, the absorbed energy transfers backward from 4T1 → 6A1 transition of Mn2+ to 4I15/2 → 4F9/2 transition of Er3+ (process ET3). This process will promote the population of level 4F9/2 in Er3+ and increase the red (650 nm) UC emission. Therefore, the principle of the tunable color is derived from the nonradiative ET from the levels 2H9/2 and 4S3/2 of Er3+ to the level 4T1 of Mn2+, then followed by back-ET which increases the population of the level 4F9/2 in Er3+, thereby resulting in the enhancement of R/G ratio [20, 22]. The suppression of traditional green emission (545 nm) and enhancement of red emission signify the strong interaction between Er3+ and Mn2+ ions, confirming that their ET processes are significant. In the previous studies, the transition of 2H9/2 → 4I13/2 was considered as a nonradiative transition approach and rarely emitted at 560-nm UC emission. In fact, as Fig. 5 shows, the 560-nm emission transition is also an ET channel from Er3+ to Mn2+. Therefore, the 560-nm UC transition will compete with the ET process (ET1) as the pump power increases. Under lower pump power, the absorbed energy mainly populates the lower excited states of Er3+ and the new green (560 nm) emission is relatively weaker than the traditional green (545 nm) emission, and simultaneously the process ET1 is insufficient. When the pump power is sufficiently high, the higher excited states of Er3+ can be efficiently populated, leading to the competition between the 560-nm emission and process ET1.

We next examined the time-resolved measurements for the β-NaYF4:Yb/Er microcrystals doped with different amounts of Mn2+ ions. Figure 6 shows the decay curves and corresponding lifetimes for the red (650 nm), traditional green (545 nm), and new green (560 nm) UC emissions. It can be found that the lifetime of red emission (650 nm) is the longest among the UC emissions. It reveals that the level 2H9/2 of Er3+ can be significantly populated through the red emitting level (4F9/2). Therefore, we have observed that the 560-nm UC emission becomes more efficient (Figs. 2b and 3a). Notably, the lifetime of the 545- and 560-nm UC emissions tends to decrease as the doping Mn2+ ions increase. In contrast, the lifetime of the red (650 nm) emission proposes a declining trend with the doping Mn2+ ions increase from 0 to 30 mol%. The reason is that more doping Mn2+ ions increase the rate of ET process from Mn2+ to Er3+, leading to more electrons populating the red emitting level (4F9/2) of Er3+. The conversion efficiency of the processes ET1 and ET2 was obtained using the following equation [19, 42]:

$$ \eta =1-\frac{\tau_{\mathrm{Yb}/\mathrm{Er}\left(\mathrm{Mn}\right)}}{\tau_{\mathrm{Yb}/\mathrm{Er}}} $$
(1)
Fig. 6
figure 6

Time-resolved evolutions of the UC emissions from the β-NaYF4:Yb/Er (20/2 mol%) microcrystals doping with different amounts of Mn2+ ions. a (4S3/2) → 4I15/2 (545 nm), b 2H9/2 → 4I13/2 (560 nm), c 4F9/2 → 4I15/2 (650 nm)

where τYb/Er(Mn) and τYb/Er represent the lifetime of β-NaYF4:Yb/Er microcrystals doping with and without Mn2+ ions, respectively. By using the lifetime values from Fig. 6, we can obtain that the efficiency of η1 was approximately 34% and η2 was nearly 41% for the β-NaYF4:Yb/Er microcrystals doped with 30 mol% Mn2+ ions. The results reveal that the processes ET1 and ET2 play an important role in populating the level 4T1 of Mn2+, which leads to an enhancement of the red UC emission based on the process ET3 from Mn2+ to Er3+. It is worth noting that the η1 is less than η2, indicating that the process ET2 is more efficiently than ET1. Therefore, compared with the traditional ratio (650 to 545 nm), the new ratio (650 to 560 nm) remains a lower tunability since the process ET1 is simultaneously acting as radiative transition (560 nm UC emission) and an ET channel from Er3+ to Mn2+.

Conclusion

In conclusion, we have demonstrated the tunable multicolor display from a single Mn2+-doped β-NaYF4:Yb/Er microcrystal by individual optical characterization. The tuning multicolor from green to red was realized in a single β-NaYF4:Yb/Er microcrystal by varying the doping amounts of Mn2+ ions. Under high power excitation, the newly emerged green (560 nm) UC emission modifies the region of tuning R/G ratio. Further investigations suggest that the tunable multicolor not only depends on the doping concentrations of Mn2+ ions, but also relies on the excitation intensities. Our work presents a new approach for understanding the tunable multicolor in Mn2+-doped with Yb3+/Er3+ microcrystals. We believe that the tunable color for the single microcrystal provides potential prospects in both color display and micro-optoelectronic devices.