1 Introduction

Research on trivalent lanthanide-doped upconversion (UC) phosphor materials have attracted broad attention during the past decade for their widespread application in diverse areas, like solid-state lighting [1], solar cells [2,3,4], temperature sensors [5], anti-counterfeiting security inks [6,7,8,9], modern display devices [10], and biomedical imaging [11,12,13]. The lanthanides show characteristic visible emissions because well-shielded 4f electrons show sharp intra-4f transitions [14, 15]. An important criterion of obtaining efficient phosphor lies in careful selection of the host compound. UC luminescence intensity is greatly enhanced by host materials having lower phonon energy as they decrease non-radiative energy transfers [10]. Among numerous inorganic materials, tungstates have drawn significant attraction due to their easy synthesis method [16, 17], chemical stability [18], yield and lower phonon energy [19]. Compared to the sol–gel and co-precipitation method, hydrothermal synthesis uses water as the reaction medium. The synthesis cost is low and reaction conditions are more controllable which cause advantages of producing crystal of better purity and uniform dispersion [8]. For UC phosphors, Yb3+ ion (sensitizer) has large absorption around the commonly used 980 nm excitation source [20, 21]. Yb3+ ions transfer the absorbed radiation to activator ions and the activators show different characteristic emissions from their step-like energy levels. Among various rare earth elements, Er3+ ion is well known due to its characteristic green (2H11/2/ 4S3/2 → 4I15/2) and red (4F9/2 → 4I15/2) emission [22,23,24,25]. These emissions are utilized to prepare phosphors with visible emissions [26, 27].

Herein, we have explored the consequences of varying Er3+ concentrations on UC luminescence properties of Yb, Er codoped BaWO4 synthesized via hydrothermal method. XRD, FTIR, TEM, and XPS studies were done for structural characterization.

2 Experimental

For BaWO4: Yb3+, Er3+ phosphor synthesis, barium nitrate [(Ba(NO3)2) (purity 99.99%, Merck)], sodium tungstate [(Na2WO4.2H2O) (purity 99.99%, Merck)], ytterbium oxide [(Yb2O3) (purity 99.99%, Merck)], and erbium oxide [(Er2O3) (purity 99.99%, Merck)] were taken as precursors. Amounts of the cations were as follows: (1-x–y)Ba2+: xYb3+: yEr3+, where x = 0.05 and y = 0.01, 0.02, 0.03, and 0.04. Firstly, Yb2O3 and Er2O3 were separately taken into small quantity of dilute nitric acid followed by simultaneous heating and stirring. Ba(NO3)2 and Na2WO4·2H2O were added into deionized water in a beaker. This aqueous solution was heated and stirred simultaneously (~ 30 min) until the solution changed into white flocculent liquid. Next, the well dissolved rare earth solutions prepared initially were mixed thoroughly with the white flocculent liquid by 30 min of stirring and pH was raised to 9 by careful addition of ammonium hydroxide solution. The solution was then sealed in Teflon-lined stainless steel autoclave and put at 190 °C for 18 h followed by natural cooling to ambient temperature. Washing of the precipitate was done with ethanol and deionized water for impurity removal and dried for ~ 12 h at 80 °C.

3 Characterization

The crystal structure determination was done using Bruker D8 ADVANCE ECO diffractometer with monochromatic Cu Kα radiation (λ = 1.5406 Å). The tube current and voltage during the experiment were 25 mA and 40 kV, respectively. The microscopic images of nanophosphors were taken by JEOL TEM (FEG-TEM, JEOL-JEM 2100F). FTIR characterization was performed by Shimadzu IRAffinity-1S. XPS measurement was recorded using OMICRON X-ray photoelectron spectrometer (serial number: 0571) with Al Kα X-ray source. Photoluminescence (PL) property was investigated by Edinburgh FLS 980 Photoluminescence Spectrometer with a 980 nm excitation source. Chromaticity coordinates were calculated using the 1931 CIE chromaticity theory.

4 Results and discussion

4.1 Structural analysis

The XRD graphs of BaWO4: Yb, Er is represented in Fig. 1a. The diffraction patterns of both pure and doped samples have shown well-matched peaks with standard JCPDS data card number 85–0588, which belongs to tetragonal scheelite-like BaWO4 with space group I41/a [28, 29]. Due to the small size difference of Yb3+ and Er3+ with the Ba2+ ions, all the dopant elements are expected to reside in the Ba2+ sites in BaWO4 (Fig. 1c) [30]. This is evident from the diffraction patterns, where no extra peak corresponding to dopant ions is present. The diffraction patterns suggest effective incorporation of dopants up to the concentration of 5% Yb3+ and 4% Er3+ without any significant effect on the host crystal structure.

Fig. 1
figure 1

a XRD patterns of BaWO4: Yb, Er. b Williamson–Hall plot of BaWO4: 5%Yb, 3%Er. c Schematic crystal structure of BaWO4

Full size image

The Williamson–Hall (W–H) plot is used to calculate the crystallite size and the micro-strain present in the materials. The W–H equation can be given by

$$\beta \mathrm{cos}\theta =4\varepsilon \mathrm{sin}\theta + \frac{K\lambda }{D}$$
(1)

where \(\beta \) signifies the FWHM measured for different XRD lines corresponding to different planes, ε is the micro-strain inside the material, and D is the crystallite size. The exact value of K is not known for the present materials system, hence K = 1 was used and D value obtained from the intercept are only estimates. Equation 1 represents a straight line between 4sinθ (X-axis) and βcosθ (Y-axis). The slope of the line gives the strain (ε) and the intercept (λ/D) of this line on the Y-axis gives grain size (D) [29]. From the intercept of W–H plot, crystallite size was found to be around 79 nm (Fig. 1b).

Particle morphology and size of synthesized phosphors are illustrated in Fig. 2. Distorted spherical-type particles with size ~ 50–100 nm were found for BaWO4: Yb, Er as indicated in Fig. 2a. The lattice image of the nanophosphor showing different arrays of atoms is presented in Fig. 2b.

Fig. 2
figure 2

a TEM micrographs of BaWO4: 5%Yb, 3%Er. b Lattice image of the nanophosphor

Full size image

Figure 3 shows the FTIR spectra of pure BaWO4 and BaWO4: Yb, Er. The sharp absorption peak observed around 800 cm−1 is attributed to the anti-symmetric stretching of the W–O bonds present in WO42− groups [29]. Small peak near 1500 cm−1 is due to carboxylate groups coming from citrate ions used during synthesis.

Fig. 3
figure 3

FTIR spectra of BaWO4: 5%Yb, 3%Er and pure BaWO4

Full size image

XPS is a useful surface characterization technique to confirm presence of dopant elements in the host material. XPS data taken for BaWO4: 5% Yb, 3% Er sample are presented in Fig. 4. The survey spectrum (Fig. 4a) confirms the presence of Ba, W, O, Yb, and Er in the sample. High-resolution spectrum of Ba 3d (Fig. 4b) reveals two peaks at 778.7 and 794.1 eV corresponding to Ba 3d5/2 and Ba 3d3/2, respectively. For W 4f (Fig. 4c) there are two peaks at 34.2 and 36.4 eV due to W 4f7/2 and W 4f5/2, respectively. Peak fitting for Yb 4d leads to identification of four main peaks in the range of 174–202 eV (Fig. 4d). Fitting for Er 4d shows a peak at 168 eV (Fig. 4e). All the peaks are in well accordance with previously reported data [31].

Fig. 4
figure 4

XPS spectrum of a BaWO4: 5% Yb, 3% Er and high-resolution spectra of b Ba 3d, c W 4f, d Yb 4d, and e Er 4d

Full size image

4.2 Photoluminescence study

As shown in Fig. 5a, emission peaks were observed in the regions of 515–575 nm and 625–685 nm for BaWO4: Yb, Er. For constant Yb3+ concentration of 5 mol %, Er3+ was taken from 1 mol% to 4 mol%. As shown in Fig. 5, the emission intensity gradually increases from 1 mol% up to 3 mol% and decreases at 4 mol% of Er3+. With increasing Er3+ concentration from 1 mol% to 4 mol% and invariable Yb3+ concentration (5 mol%), the variation tendency of UC intensities is presented in Fig. 5a and b. The optimum dopant concentration was found to be 5 mol% Yb3+ and 3 mol% Er3+ for the BaWO4: Yb, Er phosphors.

Fig. 5
figure 5

a PL emission spectra of BaWO4: Yb, Er. b Relative intensities of the 655 nm and 541 nm emissions for different Er3+ concentrations

Full size image

If luminescence quenching is due to the energy transfer among identical lanthanide ions, the determination of critical distance (Rc) can be done by the Blasse equation (Eq. 2) [29]

$$ R_{C} = 2(\frac{{3V}}{{4\pi {\text{X}}_{c}N}})^{{{\raise0.5ex\hbox{$\scriptstyle 1$} \kern-0.1em/\kern-0.15em \lower0.25ex\hbox{$\scriptstyle 3$}}}} $$
(2)

where ‘V’ = unit cell volume, ‘Xc’ = activator ion’s mole fraction beyond which quenching occurs (critical concentration) and ‘N’ = crystallographic sites possessed by Er3+ in unit cell. For optimized 3 mol% Er3+ (Xc = 0.03)-doped phosphors tetragonal system, a = b = 5.613 Å, c = 12.72 Å, Z = 4, V = abc = 400.75 Å3, and N = Z, Rc is calculated to be 18.54 Å. Since Rc is greater than 5 Å, it is concluded that exchange interaction among Er3+ ions is not the reason behind concentration quenching, but rather multipolar interaction is involved in it [29, 32].

Chromaticity coordinates calculated using the 1931 CIE chromaticity theory are shown in Fig. 6. CIE coordinates of different phosphors are tabulated in Table 1.

Fig. 6
figure 6

CIE diagram of BaWO4: 5% Yb, x% Er, x = 1, 2, 3, 4 (in inset 1, 2, 3, 4 are representing the mole % of Er3+)

Full size image
Table 1 CIE coordinates of BaWO4: Yb, Er phosphors
Full size table

4.2.1 Photoluminescence mechanism

The UC photoluminescence mechanism consists of different processes, like ground-state absorption (GSA), excited-state absorption (ESA), energy transfer (ET), and non-radiative transfer (NRT). Upon excitation with 980 nm laser, energy was mostly absorbed by the sensitizer Yb3+ as Yb3+ (~ 11.7 X 10–21 cm2) has almost 7 times higher absorption cross-section area than Er3+ (~ 1.7 X 10–21 cm2) around 980 nm [33]. This GSA process causes the transition from ground level 2F7/2 to the excited level 2F5/2 in Yb3+. While returning to ground level, the Yb3+ ions transfer their energy via different ET processes to the Er3+ ions from where further processes for luminescence take place as presented in Fig. 7. For green emissions, two sequential ET processes populate the 4F7/2 excited state from ground 4I15/2 as presented below:

$$ {}^{{\text{4}}}{\text{I}}_{{15/2}} + {\text{ET}} - 1({\text{GSA)}} \to {}^{{\text{4}}}{\text{I}}_{{11/2}} + {\text{ET}} - 2\;({\text{ESA}}) \to {}^{{\text{4}}}{\text{F}}_{{7/2}} $$
Fig. 7
figure 7

Schematic representation of energy levels in Yb3+ and Er3+ along with proposed mechanism for green and red emissions (Color figure online)

Full size image

Two different non-radiative energy transfers from the excited 4F7/2 level populate the 2H11/2 and 4S3/2 levels. Subsequent relaxation to 4I15/2 level gives rise to two green emission bands, i.e., green1 (527 nm) and green2 (541 nm) due to the transitions 2H11/2 → 4I15/2 and 4S3/2 → 4I15/2, respectively.

$$ {\text{Green1:}}\;{}^{4}{\text{I}}_{{15/2}} + {\text{ET - 1(GSA)}} \to {}^{4}{\text{I}}_{{11/2}} {\text{ + ET - 2}}\;{\text{(ESA)}} \to {}^{4}{\text{F}}_{{7/2}} \to {}^{2}{\text{H}}_{{{\text{11/2}}}} \to {}^{4}{\text{I}}_{{15/2}} $$
$${\text{Green2:}}\;{}^{4}{\text{I}}_{{15/2}} + {\text{ET - 1(GSA)}} \to {}^{4}{\text{I}}_{{11/2}} {\text{ + ET - 2}}\;{\text{(ESA)}} \to {}^{4}{\text{F}}_{{7/2}} \to {}^{2}{\text{S}}_{{{\text{3/2}}}} \to {}^{4}{\text{I}}_{{15/2}}$$

For red emission, electrons from the excited 4I11/2 state populate the 4I13/2 state via an NRT process. An ESA from this 4I13/2 state populate the 4F9/2 state. Relaxation from this excited 4F9/2 level to ground 4I15/2 level is accompanied by red emission between 640 and 690 nm. Whole process is outlined below:

$$ {}^{4}{\text{I}}_{{15/2}} + {\text{ET - 1(GSA)}} \to {}^{4}{\text{I}}_{{11/2}} \to {}^{4}{\text{I}}_{{13/2}} + {\text{ET - 3}}\;{\text{(ESA)}} \to {}^{4}{\text{F}}_{{9/2}} \to {}^{4}{\text{I}}_{{15/2}} $$

Relation between pump power (P) of laser source and UC emission intensity (Iem) is given by: Iem α Pn, where ‘n’ represents the number of photons involved for a particular emission [27]. Value of ‘n’ is found from slope by plotting logarithm of pump power vs logarithm of emitted intensity (lnP vs lnIem graph in Fig. 8 inset). The values of ‘n’ measured by curve fitting for 655 nm and 541 nm are 1.63 and 1.57, respectively. The slopes for both emissions prove two-photon processes occur both in 655 nm and 541 nm UC emissions. The values of n obtained are lower than the theoretical value of 2, which indicates the saturation of the upconversion process [1].

Fig. 8
figure 8

UC emission intensity with varying pump powers for the BaWO4: 5%Yb3+, 3%Er3+. Inset figure shows lnP vs lnIem plot with linear fitting

Full size image

To further explore the mechanism, the decay curves of red 4F9/2 → 4I15/2 (655 nm) and green 4S3/2 → 4I15/2 (541 nm) in BaWO4: 5%Yb3+, 3%Er3+ were recorded under pulsed laser excitation of 980 nm.

Table 2 The experimental values of the parameters τ1, τ2, A1, A2, and τavg
Full size table

Figure 9 shows the decay curves of 655 nm and 541 nm emissions of the nanophosphor that were fitted into a mono-exponential and bi-exponential function, respectively, which is in well accordance with the available literature [23]. The bi-exponential decay indicates that more than one decay channel (both radiative and non-radiative) is involved in the total decay process, whereas mono-exponential decay indicate only one kind (radiative) of luminescent center dominates the PL emission. The bi-exponential decay equation (Eq. 3) and the mono-exponential decay equation (Eq. 4) are given below:

Fig. 9
figure 9

Time decay curve of BaWO4: 5% Yb3+, 3% Er3+ for a 541 nm and b 655 nm

Full size image
$$I(t)= {A}_{1}{\mathrm{exp}}^{\left(-\frac{t}{{\tau }_{1}}\right)}+{A}_{2}{\mathrm{exp}}^{\left(-\frac{t}{{\tau }_{2}}\right)}$$
(3)
$$I\left(t\right)= {I}_{0}+{A}_{1}{\mathrm{exp}}^{\left(-\frac{t}{\tau }\right)}$$
(4)

where I0 is the initial luminescence intensity, τ1 and τ2 represent fast and slow luminescence lifetime, respectively, and A1 and A2 are the constants (Table 2) [23, 34]. The average lifetime for bi-exponential decay curve is calculated using the equation below (Eq. 5).

$$ \tau_{avg} = \frac{{A_{1} \tau_{1}^{2} + A_{2} \tau_{2}^{2} }}{{A_{1} \tau_{1} + A_{2} \tau_{2} }} $$
(5)

The average lifetime for 541 nm emission was calculated using Eq. 5 and observed lifetime of 655 nm (red) emission and 541 nm (green) emissions was 502.95 μs and 388.22 μs, respectively.

5 Conclusion

We have reported BaWO4: Yb, Er nanophosphor prepared via hydrothermal route. The phosphors have shown green and red emissions corresponding to the transitions 2H11/2/4S3/2 → 4I15/2 and 4F9/2 → 4I15/2 of Er3+, respectively. The concentration of Er3+ was optimized to be 3 mol % for the highest luminescence intensity. It was determined that two-photon processes control both red and green emissions. The average decay time was detected to be in the microsecond region. Overall PL mechanism was discussed accordingly. The results signify that the discussed phosphor may be a suitable option for UC-based lighting applications.