Effects of Li⁺, Na⁺, and K⁺ doping on the microstructure, fluorescence thermometry, and thermochromism of Ho³⁺,Yb³⁺:Bi₂WO₆ materials†

Mengliang Jiang ^ab, Linxiang Wang *^ab, Munire Maimaiti ^ab, Xin Feng ^ab, Yan Zhang ^ab and Jiachen Shi ^ab
aSchool of Physics and Electronic Engineering, Xinjiang Normal University, Urumqi, Xinjiang 830054, China. E-mail: wanglinxiang23@126.com; 1793737924@qq.com
^bXinjiang Key Laboratory for Luminescence Minerals and Optical Functional Materials, Xinjiang Normal University, Urumqi, Xinjiang 830054, China

First published on 30th June 2025

---

---

1. Introduction

Near-infrared-responsive rare-earth upconversion luminescent materials have the advantages of good photostability, an adjustable particle size, small background fluorescence interference, and better penetration depth of the emitted light in biological tissues; thus, these materials have great development potential in the fields of bioimaging, marker detection, and temperature sensing and have therefore attracted the attention of many researchers.^1,2 The rare-earth ion Ho³⁺ has abundant energy levels with large energy gaps between adjacent energy levels and strong emission in both the green and red light regions, and strong upconversion fluorescence of Ho³⁺ can be achieved by introducing Yb³⁺, which has a large absorption cross-section for 980 nm laser light, as a sensitizer³. Ho³⁺ and Yb³⁺-codoped luminescent materials and their temperature sensing properties have attracted much attention from researchers. For example, Julijia et al.⁴ prepared Ho³⁺:K₂Gd(PO₄)(WO₄) and Ho³⁺,20% Yb³⁺:K₂Gd(PO₄)(WO₄) samples via the solid-phase method to investigate the effects of the Ho³⁺ concentration and ambient temperature on the luminescence properties of the samples, and the red–green fluorescence intensity ratio of the Ho³⁺,Yb³⁺:K₂Gd(PO₄)(WO₄) sample under 980 nm excitation in the temperature range of 150–500 K was determined. The red–green fluorescence intensity ratio gradually increased with increasing temperature, indicating that the Ho³⁺,Yb³⁺:K₂Gd(PO₄)(WO₄) sample can be used for fluorescence thermometry. Liu et al.³ synthesized two kinds of upconversion phosphors, Ho³⁺,Yb³⁺:BiOF and Ho³⁺,Yb³⁺:BiF₃, and explored the crystal structure and temperature sensing properties of the materials; the results revealed that Ho³⁺,Yb³⁺:BiOF had a relative sensitivity as high as 5.41% K⁻¹ in the temperature range of 298–673 K. Lei et al.⁵ synthesized four different phosphors by doping Tm, Er or Ho as well as all three Tm/Er/Ho into the eulytite-type host Ba₃Yb(PO₄)₃ and investigated the emission spectra, color purity, brightness, energy transfer mechanism, and temperature sensing properties of the phosphors; the results showed that the Ho³⁺:Ba₃Yb(PO₄)₃ samples presented a maximum relative temperature sensing sensitivity of 0.26% K⁻¹ in the temperature range of 303–603 K. Liu et al.⁶ synthesized phase-pure eulytite-type 0.10% Yb³⁺,0.02Ln³⁺:Sr₃Y_0.88(PO₄) (Ln = Ho, Tm, Er) upconversion Our group previously⁷ investigated the upconversion luminescence and fluorescence thermometry performance of Bi₂WO₆ materials doped with Ho³⁺ and Yb³⁺ at different concentrations, and a maximum thermometric sensitivity of 1.85% K⁻¹ was obtained at 523 K for 3% Ho³⁺ and 10% Yb³⁺:Bi₂WO₆. However, to realize practical temperature sensing applications, the luminescence efficiency and temperature sensitivity of synthetic materials still need to be improved.

According to the literature, the doping of the alkali metals Li⁺, Na⁺, and K⁺ improves the luminescence performance of materials.^8–11 LuXiong¹² synthesized a series of novel ternary alkali metal-doped Yb³⁺,Er³⁺:Li_xNa_yK_(1−x−y)YF₄ upconversion luminescent materials via a hydrothermal method and studied the luminescence of the materials. The results showed that the doping of Li, Na and K ions changed the crystal structure of Yb³⁺,Er³⁺:Li_xNa_yK_(1−x−y)YF₄, which affected the spatial structure relationship between the rare earth ions and changed their energy transfer efficiency, thus improving the luminescence intensity of the material. When the molar ratio of doped Li, Na, and K was 10:6:9, the luminescence intensity was 30 times greater than that of the undoped sample. Zaifa Yang¹³ synthesized the red fluorescent powder Mn⁴⁺:La₃Mg₂NbO₉via the solid-phase method and analyzed the effects of adding boric acid and charge compensators (Li⁺, Na⁺, K⁺) on the crystal structure, morphology, quantum efficiency, and luminescence properties of the Mn⁴⁺:La₃Mg₂NbO₉ material. The results showed that the addition of boric acid and a charge compensator could effectively improve the quantum efficiency and luminescence intensity, and the luminescence intensity of the material was enhanced by a factor of 5.9 when Li⁺ was used as the charge compensator. However, few reports have investigated the effects of Li⁺, Na⁺, and K⁺ on the microstructure, upconversion luminescence, and temperature sensing properties of Ho³⁺,Yb³⁺:Bi₂WO₆ materials. Therefore, in this work, the effects of ion doping on the microstructure, upconversion luminescence and temperature sensing properties of Ho³⁺,Yb³⁺:Bi₂WO₆ materials were investigated by doping them with the alkali metal ions Li⁺, Na⁺, and K⁺.

2. Experimental and theoretical calculation methods

2.1. Sample synthesis

A series of 1% M, 3% Ho³⁺, and 10% Yb³⁺:Bi₂WO₆ (M: Li⁺, Na⁺, K⁺) samples were synthesized via the high-temperature solid-phase method. The ion doping concentration in all the samples was the concentration of the corresponding substance. The raw materials used in the experiments were Bi₂O₃ (99.9%), WO₃ (99.99%), Ho₂O₃ (99.99%), Yb₂O₃ (99.99%), Li₂CO₃ (99.99%), Na₂CO₃ (99.99%), and K₂CO₃ (99.99%). An AL104 electronic balance was used to weigh the required raw materials in accordance with the chemical proportions, which were then mixed and ground for 25 min, and the mixture was subsequently placed in a crucible in an SX-12-16 high-temperature box-type muffle furnace for calcination (initially from 50 °C to 350 °C at a rate of 3° min⁻¹ and then from 350 °C to 800 °C at a rate of 5° min⁻¹). The samples were calcined at 800 °C for 3 h, cooled to room temperature, removed, and milled again for 5 min to obtain 1% M, 3% Ho³⁺, 10% Yb³⁺:Bi₂WO₆ (M: Li⁺, Na⁺, K⁺) series samples.

2.2. Performance test equipment

The physical phase and microstructure of the samples were analyzed with a Tsushima XRD-6100 diffractometer using X-rays at a wavelength of 1.54095 Å. The current and voltage of the X-ray tube were 10 mA and 40 kV, respectively, and the scanning range was 10–80°, with a scanning step of 0.02° and a test rate of 2° min⁻¹. A ZEISS SUPRA55 VP scanning electron microscope from Zeiss, Germany, was used to observe the microscopic morphology, size and dispersion of the samples, and a Bruker X-Flash SDD 5010 was used to analyze the elements and their contents on the powder surface. A Hitachi U-3900/3900H UV-vis spectrophotometer was used to obtain the absorption spectra of the materials in the range of 250–800 nm at a rate of 30 nm min⁻¹. Fluorescence detection of the sample was carried out with an FLS920 full-featured steady-state/transient fluorescence spectrometer from Edinburgh, U.K. The excitation source was infrared light from a 980 nm diode laser (MDL-III-980-2 W), and the upconversion emission spectra were measured at different pump powers. The sample was heated by the high-temperature accessory heater of the spectrometer (the temperature control range was from room temperature 298 K to 573 K) to obtain the variable-temperature upconversion spectra. To avoid laser-induced thermal effects, the samples were excited with the appropriate excitation power, and all the equipment was calibrated before testing.

2.3. Theoretical calculation methods

In this work, we used the Vienna ab initio simulation package (VASP)¹⁴ in conjunction with the postprocessing VASPKIT¹⁵ package to perform electronic band calculations via density functional theory (DFT). The projector augmented wave (PAW) pseudopotential was used to describe the interaction between electrons and nuclei. For the exchange–correlation function, the Perdew–Burke–Ernzerhof (PBE) form of the generalized gradient approximation (GGA) was used. The energy cutoff for plane-wave basis expansion was 400 eV. The geometric structure relaxation convergence criterion was −0.02 eV Å⁻¹. The geometry was optimized via the Monkhorst–Pack method with a k-point grid of size 5 × 3 × 9. The energy band calculations were performed in k-space along the following high symmetry points: G (0, 0, 0), X (0.5, 0, 0), S (0.5, 0.5, 0), and Y (0, 0.5, 0). The energy convergence threshold used for the self-consistent calculations was 10⁻⁸ eV.

3. Analysis of results

3.1. Structural calculations

3.2. Structural characterization

The XRD patterns of the 1% M, 3% Ho³⁺, and 10% Yb³⁺:Bi₂WO₆ (M: Li⁺, Na⁺, K⁺) series of samples and the 3% Ho³⁺,10% Yb³⁺:Bi₂WO₆ and Bi₂WO₆ materials are shown in Fig. 3. The positions of the diffraction peaks of all the samples coincide with those of the Bi₂WO₆ standard card #04-001-8551, and there are no other peaks. The doping of Li⁺, Na⁺, K⁺, Ho³⁺, and Yb³⁺ essentially does not alter the orthorhombic crystal structure of the Bi₂WO₆ matrix material, suggesting that the dopant ions partially replace Bi³⁺. As seen from the magnified view of the main peak, Ho³⁺ and Yb³⁺ doping shifts the main peak to a larger angle, and further doping with alkali metal ions (Li⁺, Na⁺, K⁺) shifts it first to a larger angle and then to a smaller angle.

Usually, whether a dopant ion can be successfully doped into a matrix can be assessed on the basis of the relative ionic radius Δr. When the relative ionic radius is less than 30%, the dopant ion can undergo substitution, and the relative ionic radius is calculated as follows:

	(13)

where r₁ and r₂ represent the radius of the substituted ion and the radius of the dopant ion in the matrix, respectively. The calculated results are shown in Table 3. The relative ionic radii are less than 30%, so Li⁺, Na⁺, Ho³⁺ and Yb³⁺ have relatively high probabilities of replacing Bi³⁺, and K⁺ has a certain probability of replacing Bi³⁺.

---

Under the same coordination environment, the trend of the radii of each ion is as follows: (rLi⁺ = 0.076 nm) < (rYb³⁺ = 0.0868 nm) < (rHo³⁺ = 0.0901 nm) < (rNa⁺ = 0.102 nm) < (rBi³⁺ = 0.103 nm) < (rK⁺ = 0.138 nm). According to the Bragg equation 2dsinθ = nλ (d: crystal plane spacing, θ: diffraction angle, n: diffraction order, λ: wavelength of incident X-rays), when n and λ are constant and the crystal plane spacing decreases, sinθ increases, the diffraction angle θ increases, and the diffraction peaks move to a larger angle. When the small-radius ions Ho³⁺ and Yb³⁺ are doped in place of the large-radius ion Bi³⁺, the lattice shrinks, and the crystal plane spacing decreases, so the main diffraction peak shifts to a larger angle. When Li⁺, which has the smallest ionic radius, replaces Bi³⁺, the lattice shrinks, the crystal plane spacing decreases, and the diffraction peak shifts to a larger angle. When Na⁺ enters the lattice, the positions of the diffraction peaks of the Ho³⁺,Yb³⁺:Bi₂WO₆ and Na⁺,Ho³⁺,Yb³⁺:Bi₂WO₆ samples are essentially the same because the difference in the radii of Na⁺ and Bi³⁺ is not large, and Na⁺ doping does not have much effect on the interstitial spacing. The radius of the K⁺ ions is larger than that of Bi³⁺, and K⁺ doping occurs not only in the interstices but also at the Bi³⁺ sites. If a small amount of K⁺ replaces Bi³⁺, the lattice expands, the crystal plane spacing d increases, the diffraction angle θ decreases, and the diffraction peak shifts to a smaller angle. The XRD data were analyzed via Jade software to obtain the cell volume, cell parameters, and crystal plane spacing of the main diffraction peak (113) for all the samples, as shown in Table 4. The results in Table 4 show that the variations in the cell volume and the crystal plane spacing of the main diffraction peak (113) are consistent with the results of the magnified XRD main diffraction peak.

---

Fig. 4 shows the results of scanning electron microscopy (SEM) and energy-dispersive spectroscopy (EDS) of 1% Na⁺,3% Ho³⁺,10% Yb³⁺:Bi₂WO₆ (Fig. 4(a)) and 1% K⁺,3% Ho³⁺,10% Yb³⁺:Bi₂WO₆ (Fig. 4(b)) powder samples after high-temperature solid-state calcination. The samples doped with Li were not tested because the atomic number of Li is too small for EDS to measure. The SEM images in the figure show that the particle sizes of the samples are in the range of 1–3 μm, and the powder is agglomerated. The mapping results show that the elements Na, K, Bi, W, O, Ho, and Yb are uniformly distributed on the powder surface. EDS was used to measure the elemental content on the powder surface, as shown in Fig. 5. Because of the small K⁺ doping content, no potassium is detected in the tested powder area, but all the other elements are detected, and the results indicate that the elements are well distributed in the powder.

Fig. 6(a) shows the UV-vis absorption spectra of the prepared samples, and the broadband absorption peaks at 250–450 nm are the absorption peaks of the matrix material Bi₂WO₆. As shown in the enlarged image in Fig. 6(a), after doping with rare earth ions, the light absorption in the range of 250–450 nm improved, and the characteristic absorption peaks of Ho³⁺ appeared at 450 nm (⁵I₈ → ⁵G₆, ⁵F₁), 540 nm (⁵I₈ → ⁵F₄, ⁵S₂) and 643 nm (⁵I₈ → ⁵F₅)⁴. After further doping of the Ho³⁺,Yb³⁺:Bi₂WO₆ samples with alkali metals, the intensities of all the absorption peaks decreased. The bandgap of the samples was estimated via the Tauc plot method^25,26 and calculated as follows:

(αhv)1/n = A(hv − Eg)	(14)

where α is the absorbance coefficient, h is Planck's constant, ν is the photon frequency, E_g is the forbidden bandwidth, and n is related to the type of semiconductor (direct semiconductor: n = 1/2; indirect semiconductor: n = 2). The bandgap results are shown in Fig. 6(b), and the bandgaps of the Bi₂WO₆, 3% Ho³⁺,10% Yb³⁺:Bi₂WO₆ and alkali-metal Li⁺, Na⁺ or K⁺-doped Ho³⁺,Yb³⁺:Bi₂WO₆ samples are 2.83 eV, 2.75 V, 2.82 eV, 2.80 eV, and 2.79 eV, respectively. After doping with rare earth elements, the bandgap of the material is reduced. The bandgap was reduced, and the material could absorb photons with lower energy, which was beneficial for the absorption of infrared photons. This can improve the light absorption range of the matrix material to some extent. At this time, the energy difference between the minimum conduction band and the lowest excited state increases, and the probability of electrons being excited to the Ho³⁺ excited state energy level increases, which leads to enhanced Ho³⁺ luminescence.

3.3. Luminescence performance analysis

A previous study⁷ reported that the luminescence intensity of Ho³⁺ and Yb³⁺ is the strongest when the ion doping ratio in a Bi₂WO₆ matrix is 3:10. In this work, we doped the alkali metal ions Li⁺, Na⁺ and K⁺ at a molar concentration of 1% to investigate the effects of different alkali metal ions on the luminescence and temperature sensing properties of Ho³⁺, Yb³⁺:Bi₂WO₆ materials. The upconversion emission spectra were obtained under a 980 nm laser with a pump power of 206 mW, as shown in Fig. 7. The four emission peaks of Ho³⁺ at 538 nm (⁵F₄ → ⁵I₈), 546 nm (⁵S₂ → ⁵I₈), 660 nm (⁵F₅ → ⁵I₈), and 756 nm (⁵F₄ → ⁵I₇) all appear in the emission spectra. Doping of Li⁺, Na⁺, and K⁺ ions does not change the position of the emission peaks of Ho³⁺. The Ho³⁺ fluorescence emission peak intensities are enhanced by K⁺ doping, with a 2.1-fold increase in the emission peak intensity at 538 nm, a 2.2-fold increase in that at 546 nm, a 1.9-fold increase in that at 660 nm, and a 2.3-fold increase in that at 756 nm compared with those of the Ho³⁺,Yb³⁺:Bi₂WO₆ samples. Combining the analytical results in Fig. 2 and Fig. 7, the bandgap of the material is significantly reduced after K⁺ doping, which is conducive to the transition of electrons to the excited state and promotes the luminescence of the luminescent center Ho³⁺. The intensity of the Ho³⁺ fluorescence emission peak is weakened after Li⁺ and Na⁺ doping.

The upconversion luminescence process under 980 nm excitation of Yb³⁺-sensitized Ho³⁺ is shown in Fig. 8. First, Yb³⁺ absorbs 980 nm photons, and energy is transferred to the ⁵I₆ energy level of Ho³⁺ (step I). Ho³⁺ again receives energy from Yb³⁺ (ET, energy transfer) or absorbs a 980 nm photon (ESA, excited state absorption, E980 nm for a 980 nm photon energy), and an electron is excited to the ⁵F₄, ⁵S₂ energy level. Three processes are involved in the Ho³⁺ transition to the ⁵F₅ energy level. In the first process, a Ho³⁺ electron at the a⁵I₆ energy level nonradiatively relaxes to the ⁵I₇ energy level and then transitions to the ⁵F₅ energy level by an ET process from Yb³⁺ to Ho³⁺ (step II). In the second process, a Ho³⁺ cross-relaxation process occurs, allowing a Ho³⁺ electron to transition to the ⁵F₅ energy level (step III). The third process is nonradiative relaxation of a Ho³⁺ electron at the ⁵F₄, ⁵S₂ energy level to the ⁵F₅ energy level (step IV, ΔE is the energy released during the nonradiative transition). The transition energy levels for the four emission peaks in Fig. 7 are ⁵F₄ → ⁵I₈ (538 nm), ⁵S₂ → ⁵I₈ (546 nm), ⁵F₅ → ⁵I₈ (660 nm), and ⁵F₄ → ⁵I₇ (756 nm).

	(15)

After Li⁺, Na⁺, or K⁺ codoping, the ions occupy either the lattice sites via Bi³⁺ substitution or the lattice gaps, which will cause lattice distortion such that the local symmetry of the material is reduced, which is conducive to increasing the luminescence intensity of the luminescent centers. However, the local symmetry reduction caused by lattice distortion is also affected by the doping concentration, and if the material is overdoped, the local symmetry of the lattice tends to be symmetric again, thus reducing the luminescence. Carbonates were used as sources of alkali metal ions during sample synthesis. According to the literature,²⁷ Li₂CO₃, Na₂CO₃, and K₂CO₃ can act as fluxes during sample synthesis, and the lower the melting point of the flux is, the greater the flux effect. The melting points of Li₂CO₃, Na₂CO₃, and K₂CO₃ are 723 °C, 851 °C, and 891 °C, respectively, and their decomposition temperatures are 1300 °C, 1744 °C, and 270 °C, respectively. K₂CO₃ decomposes into K₂O and CO₂ above 270 °C, and the melting point of K₂O is 350 °C. Because the samples were synthesized at 800 °C, K₂O, which has the lowest melting point during sample synthesis, had the best flux effect, and Na₂CO₃ had the worst flux effect. The main role of the flux is to reduce the temperature of the reactants and improve the crystallinity of the powder material to improve the luminescence intensity. The relative ionic radius data in Table 2 indicate that Li⁺, Na⁺, and K⁺ all have a probability of replacing Bi³⁺ in the matrix. If alkali metal Li⁺, Na⁺, and K⁺ replacement of Bi³⁺ occurs, then according to the principle of charge conservation, oxygen vacancies will be introduced inside the grains, and the negative electronegativity defects induced by charge compensation will result in the entry of more Ho³⁺ into the lattice, which will promote luminescence. An appropriate amount of doped Li⁺, Na⁺, and K⁺ can combine with the oxygen dangling bonds on the surface of the particles to reduce the fluorescence burst caused by the surface state and improve the luminescence intensity. However, excessive doping introduces more nonluminescent centers, the probability of cross-relaxation increases, and the chance of nonradiative transitions increases, ultimately leading to fluorescence bursts.⁹ At a concentration of 3%Ho³⁺, the Li⁺ and Na⁺ doping concentration is excessive, resulting in a weakening of the fluorescence intensity.

Fig. 9 shows the upconversion emission spectra of the 1% Li⁺, 1% Na⁺, or 1% K⁺ codoped 3% Ho³⁺,10% Yb³⁺:Bi₂WO₆ samples under 980 nm excitation (pump power of 14–206 mW). The intensity of the Ho³⁺ emission bands in the samples gradually increases with increasing excitation pump power. According to the literature,²⁸ the number n of excitation photons absorbed by an upconversion luminescent material to emit one photon can be obtained from I ∝ Pⁿ, where I is the upconversion emission intensity and P is the pump power. The value of n is affected by the competition between the decay rate and the upconversion rate of the intermediate-state ions. When the excitation pump power is increased, the probability of Yb³⁺ being excited increases, the probability of ET from Yb³⁺ to Ho³⁺ increases, and the probability of upconversion radiative emission from Ho³⁺ increases, so the sample luminescence is initially enhanced. However, luminescence attenuation also occurs at the same time, and when the power is sufficiently high, the thermal effect due to the high power increases the chance of interion cross-relaxation and radiation-free relaxation. The rate of attenuation increases, so after a certain excitation power is reached, the calculated value of n is smaller than the actual value. The value of n gradually decreases with increasing power, the “saturation effect” occurs, and the light intensity I and the power p have a nonlinear relationship, resulting in the inability to accurately determine the number of photons absorbed to form the emission peak. Therefore, a high power is not appropriate in the experimental process, and the power range chosen in the experiment was 14–206 mW. Since the pump current of the laser is directly obtained during the experiment, the different pump powers can be replaced by the excitation current under a constant voltage. For the above three samples, the relationships between the light intensities of the four emission peaks at 538, 546, 660 and 756 nm and the pump current are shown in Fig. 11. By fitting the experimental data, the n values corresponding to the four emission peaks of the 1% Li⁺-doped samples are obtained as 1.84, 1.85, 1.75 and 1.88; those corresponding to the four emission peaks of the 1% Na⁺-doped samples are obtained as 1.62, 1.59, 1.43 and 1.34; and those corresponding to the four emission peaks of the 1% K⁺-doped samples are obtained as 1.38, 1.34, 1.81 and 1.64. The calculations show that all four emission peaks of Ho³⁺ in the above three samples arise from two-photon absorption.

3.4. Temperature sensing characteristics

When the relationship between a material and the temperature is tested, a higher laser power affects the experimental results because of the laser thermal effect; however, when a small laser power is chosen, the luminescence of the material will be weaker, which is not conducive to the acquisition of fluorescence signals. Therefore, we chose to carry out a temperature sensing performance experiment with a laser pump power of 206 mW. Fig. 10(a), (c) and (e) show plots of the upconversion emission spectra versus temperature for 980 nm (excitation power of 206 mW) laser excitation conditions in the temperature range of 298–573 K. The positions of all the emission peaks remain unchanged with increasing temperature, but the intensities all decrease with increasing temperature. This behavior occurs because as the temperature increases, the lattice vibrations increase and the number of phonons increases, leading to an increase in the probability of nonradiative transitions, which in turn leads to a decrease in the probability of radiative transitions and ultimately a decrease in the fluorescence emission intensity. Fig. 10(b), (d) and (f) show the variation in the intensity of the four fluorescence emission peaks (538 nm, 546 nm, 660 nm, and 756 nm) of Ho³⁺ with temperature. The small figures show a plot of the intensity ratio of red light (660 nm, 756 nm) to green light (538 nm, 546 nm) as a function of temperature. The red–green intensity ratio increases with temperature.

A polynomial equation was used to fit the Ho³⁺ 756 nm/538 nm fluorescence intensity ratio and temperature in the temperature range of 298–573 K. The fitting formula was as follows:

	(16)

where I_756nm and I_538nm denote the fluorescence intensities of the corresponding 756 nm and 538 nm emission peaks generated by the ⁵F₄ → ⁵I₇ and ⁵F₄ → ⁵I₈ transitions of the nonthermally coupled energy levels of Ho³⁺, respectively.

Fig. 11 shows the results of the fit between the 756 nm/538 nm fluorescence intensity ratio and temperature for the Li⁺, Na⁺- or K⁺-doped Ho³⁺,Yb³⁺:Bi₂WO₆ samples; the following relationships were obtained:

	(17)

	(18)

	(19)

To illustrate the temperature sensing performance of the materials, we used the temperature sensitivity and temperature resolution to evaluate the performance. According to the literature,¹⁸ the absolute temperature measurement sensitivity S_a and relative temperature measurement sensitivity S_r can be expressed as:

	(20)

	(21)

where FIR is the fluorescence intensity ratio and T is the absolute temperature.

According to the literature,²⁹ the temperature resolution can be calculated via the following equation:

	(22)

where δFIR is the fluorescence intensity ratio uncertainty:

	(23)

where FIR is the fluorescence intensity ratio, δI is the uncertainty in the fluorescence intensity, and I₁ and I₂ are the intensities of the two fluorescence peaks used to obtain the fluorescence intensity ratio.

Fig. 12 shows the corresponding thermometric sensitivity results after Li⁺, Na⁺, or K⁺ doping. In the temperature range of 273–573 K, the fluorescence intensity ratio (I_756nm/538nm) of the nonthermally coupled energy levels of Ho³⁺ has good thermometric sensitivity. The maximum relative thermometric sensitivities of 3% Ho³⁺ and 10% Yb³⁺:Bi₂WO₆ doped with Li⁺, Na⁺, or K⁺ are 1.69% K⁻¹ (348 K), 2.54% K⁻¹ (298 K), and 2.65% K⁻¹ (298 K), respectively. Fig. 13 shows the temperature resolution results corresponding to Li⁺, Na⁺, and K⁺ doping. The minimum temperature resolutions corresponding to Li⁺, Na⁺ or K⁺ doping are 0.14 K (323 K), 0.05 K (298 K), and 0.04 K (298 K), respectively.

Table 5 shows the temperature sensing properties of different matrix materials after codoping with Ho³⁺ and Yb³⁺. Under 980 nm laser excitation with a pump power of 206 mW, the present material and other materials can be compared in terms of their temperature characterization on the basis of the fluorescence intensity ratio (I_756nm/538nm) of the Ho³⁺ nonthermally coupled energy levels in the temperature range of 298–573 K. The 1% Li⁺,3% Ho³⁺,10% Yb³⁺:Bi₂WO₆ material prepared in the present work has good relative thermometric sensitivity.

---

The chromaticity diagrams and chromaticity coordinates of the alkali metal Li⁺, Na⁺ or K⁺-doped Ho³⁺,Yb³⁺:Bi₂WO₆ samples are given in Fig. 14 and Table 6. Under 980 nm excitation, the luminescence color of the material moves from the green region to the yellow region and finally to the orange–red region as the temperature increases. As the ambient temperature increases, the green light and red light intensities decrease, but as shown in Fig. 10(b), (d) and (f), the intensity ratio of red light to green light gradually increases. The difference in the rates of the intensity decay of red light and green light with increasing temperature is why the luminescence color of the material changes with temperature. Fig. 14 shows that the luminescence color of the samples at each temperature is close to the edge of the chromaticity diagram, indicating that the color purity of the converted luminescence of the samples is high.

---

From the chromaticity diagram Fig. 14, we observe that the luminescent color of the sample changes with temperature. To further investigate the thermochromic properties of the phosphor samples, four powder samples of 1% M⁺, 3% Ho³⁺, and 10% Yb³⁺:Bi₂WO₆ (M = Li⁺, Na⁺, K⁺) were placed on a heating table and irradiated with a 980 nm laser (2 W), and the luminescence color changes of the powder samples were observed during the warming and cooling processes, as shown in Fig. 15. Yellow–green emission was observed in all four samples after the samples were irradiated by the 980 nm laser in the dark. As the temperature increased (298–573 K), the luminescence of the powder sample sequentially changed from yellow-green to yellow, orange and red. The K⁺-doped sample tends to glow deep red at the slowest rate, whereas the undoped sample tends to glow red at the fastest rate during warming. When the above phosphors were cooled under the same conditions (573–298 K), their luminescence color reversibly changed from red to orange, yellow and finally yellow–green in sequence. These findings suggest that the above synthesized samples also have potential applications in the field of thermochromic optical safety.

Fig. 10(b), (d) and (f) shows that both the green light emission and red light emission of the material tend to decrease with increasing temperature. However, the rates of green and red light weakening are different. As seen from the red–green ratio line graph in the smaller graph, the red–green ratios of the three samples increase as the temperature increases. This results in a shift in the luminescent color of the material from green to red as the temperature increases. The slope of the red–green ratio line graph shows that the red–green ratio of the Li⁺-doped samples increases slowly and then rapidly. The Na⁺- and K⁺-doped samples increased slowly, then rapidly and finally slowly. This suggests that there is a fast and slow transition of the material from green to red as the temperature increases, with the K⁺-doped samples having the slowest conversion rate, which is consistent with thermochromism.

4. Conclusion

Theoretically and experimentally obtained from DFT theoretical calculations and UV-vis absorption spectra, the band gap of the Ho³⁺/Yb³⁺-doped or continued M⁺ (M = Li⁺, Na⁺, K⁺)-doped material is smaller than that of the matrix material Bi₂WO₆, which improves the absorption of infrared photons by the material. Samples of 1% M, 3% Ho³⁺, and 10% Yb³⁺:Bi₂WO₆ (M = Li⁺, Na⁺, K⁺) were prepared via the high-temperature solid-phase method through calcination in air at 800 °C for 3 h. The XRD results show that the doping of Li⁺, Na⁺, K⁺, Ho³⁺, and Yb³⁺ essentially does not change the orthorhombic crystal structure of the Bi₂WO₆ matrix material. SEM and EDS mapping revealed that the sample size ranged from 1–3 μm, and the element distribution on the sample surface was uniform. Excitation at 980 nm of the 1% M, 3% Ho³⁺, and 10% Yb³⁺:Bi₂WO₆ (M = Li⁺, Na⁺, K⁺) samples revealed significantly enhanced luminescence for the 1% K⁺-doped samples. The luminescence intensity–pump power relationships of the 1% M, 3% Ho³⁺, and 10% Yb³⁺:Bi₂WO₆ (M = Li⁺, Na⁺, K⁺) samples under 980 nm excitation at pump powers of 14–206 mW were fitted, and the results revealed that the four Ho³⁺ emission peaks in the samples (538 nm, 546 nm, 660 nm, and 756 nm) originated from two-photon absorption. In this work, the intensity ratios of the 756 nm (⁵F₄ → ⁵I₇) and 538 nm (⁵F₄ → ⁵I₈) fluorescence generated by the nonthermally coupled energy level pair of Ho³⁺ were used to characterize the temperature. The maximum relative thermometric sensitivities after Li⁺, Na⁺, or K⁺ doping of 3% Ho³⁺, 10% Yb³⁺:Bi₂WO₆ in the temperature range of 298–573 K are 1.69% K⁻¹ (348 K), 2.54% K⁻¹ (298 K), and 2.65% K⁻¹ (298 K), and the minimum temperature resolutions are 0.14 K (323 K), 0.05 K (298 K), and 0.04 K (298 K). In the temperature range of 298–573 K, the luminescence of the samples not doped with Li⁺, Na⁺, or K⁺ changes from yellow–green to yellow and then red as the temperature increases, and the luminescence colors of the Li⁺, Na⁺, or K⁺-doped samples all change from green to yellow, orange, and finally red, with the K⁺-doped samples having the slowest rate of change to red. As the temperature decreases, the sample luminescence color is restored, which indicates that the synthesized samples have potential applications in thermochromism and optical anticounterfeiting.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data supporting this article have been included as part of the ESI.† The original model data for the first-principle calculations in this study can be found in the Crystallography Open Database, Original Structural Model COD ID: 2108252. https://www.crystallography.net/cod/result.php. In this work, we used the Vienna ab initio simulation package (VASP) (DOI: https://doi.org/10.1016/0927-0256(96)00008-0) in conjunction with the postprocessing vaspkit (DOI: https://doi.org/10.1016/j.cpc.2021.108033) package to perform electronic band calculations via density functional theory (DFT). This study was performed using PDF#04-001-8551 standard cards from PDF2017 to analyze material XRD. This study used Origin software to process and analyze the experimental data.

Acknowledgements

This project was supported by the Natural Science Foundation of China (No. 12164048) and the Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2023D01A41).

References

1. D. Ruiz, B. del Rosal, M. Acebrón, C. Palencia, C. Sun, J. Cabanillas-González, M. López-Haro, A. B. Hungría, D. Jaque and B. H. Juarez, Adv. Funct. Mater., 2017, 27, 1604629 CrossRef .
2. Y. Zhao, X. Wang, Y. Zhang, Y. Li and X. Yao, J. Alloys Compd., 2020, 817, 152691 CrossRef CAS .
3. R. Liu, Y. Pan, J. Zhang, X. Zhuang, Y. Wang, X. Zhou, S. Du, H. Li and W. Li, J. Lumin., 2023, 258, 119791 CrossRef CAS .
4. J. Grigorjevaite and A. Katelnikovas, Materials, 2024, 16, 917 CrossRef PubMed .
5. Z. Lei, H. Dong, L. Sun, B. Teng, Y. Zou and D. Zhong, J. Mater. Chem. C, 2024, 12, 628–638 RSC .
6. W. Liu, X. Wang, Q. Zhu, X. Li, X. Sun and J.-G. Li, Sci. Technol. Adv. Mater., 2019, 20, 949–963 CrossRef CAS PubMed .
7. M. Jiang, L. Wang, M. Maimaiti, X. Feng and Y. Zhang, Dalton Trans., 2024, 53, 2591–2601 RSC .
8. J. Tuo, Y. Ye, H. Zhao and L. Wang, Chin. Opt., 2019, 12, 1279–1287 CrossRef .
9. A. Xiakeer, L. Wang, M. Maimaiti, X. Feng and M. Jiang, Rev. Adv. Mater. Sci., 2023, 62, 20230104 CAS .
10. Y. Yang, Y. Yan, C. Fan, Y. Zou, J. Wang, C. You and R. Yang, ChemistrySelect, 2021, 6, 13452–13460 CrossRef CAS .
11. W. Liu, Y. Yang, P. Liu, Y. Huang, P. Li, T. Chen, Y. Zhuo, J. Yan and K. Liu, Solid State Ionics, 2020, 356, 115451 CrossRef CAS .
12. L. Xiong, H. Hu, M. Liu and C. Zhang, J. Mater. Sci.: Mater. Electron., 2023, 34, 2260 CrossRef CAS .
13. Z. Yang, R. Wang, S. Yang, H. Bu and J. Zhao, Molecules, 2024, 29, 1402 CrossRef CAS PubMed .
14. G. Kresse and J. Furthmüller, Comput. Mater. Sci., 1996, 6, 15–50 CrossRef CAS .
15. V. Wang, N. Xu, J.-C. Liu, G. Tang and W.-T. Geng, Comput. Phys. Commun., 2021, 267, 108033 CrossRef CAS .
16. Y. Zhou, C. Li and Y. Wang, Adv. Opt. Mater., 2022, 10, 2102246 CrossRef CAS .
17. K. A. Denault, J. Brgoch, S. D. Kloß, M. W. Gaultois, J. Siewenie, K. Page and R. Seshadri, ACS Appl. Mater. Interfaces, 2015, 7, 7264–7272 CrossRef CAS PubMed .
18. K. Li, M. Dai, Z. Fu, Z. Wang, H. Xu and R. Wang, Inorg. Chem. Front., 2024, 11, 172–185 RSC .
19. X. Luo and K. Akimoto, Appl. Surf. Sci., 2013, 273, 257–260 CrossRef CAS .
20. O. V. Solovyev and M. E. Miftakhova, J. Lumin., 2018, 195, 96–101 CrossRef CAS .
21. X. Zhang, W. Gui and Q. Zeng, Ceram. Int., 2017, 43, 3346–3355 CrossRef CAS .
22. A. Phuruangrat, S. Buapoon, T. Bunluesak, P. Suebsom, S. Wannapop, T. Thongtem and S. Thongtem, Solid State Sci., 2022, 128, 106881 CrossRef CAS .
23. H. Okudera, Y. Sakai, K. Yamagata and H. Takeda, Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater., 2018, 74, 295–303 CrossRef CAS PubMed .
24. X. Wang, L. Chen, J. Zhang, B. Xing, B. Xu, G. Mi, B. Liu, Y. Wu and L. Chen, J. Phys. D: Appl. Phys., 2025, 58, 155103 CrossRef CAS .
25. E. A. Davis and N. F. Mott, Philos. Mag., 1970, 22, 0903–0922 CrossRef CAS .
26. J. Tauc, R. Grigorovici and A. Vancu, Phys. Status Solidi B, 1966, 15, 627–637 CrossRef CAS .
27. S. Wang, Q. Sun, B. Devakumar, J. Liang, L. Sun and X. Huang, J. Lumin., 2019, 214, 116525 CrossRef CAS .
28. P. Yuehong, R. Weizhou, Q. Jianbei, H. Jin, Y. Zhengwen and S. Zhiguo, J. Inorg. Mater., 2020, 35, 902 CrossRef .
29. Y. Huang, G. Bai, Y. Zhao, H. Xie, X. Yang and S. Xu, Inorg. Chem., 2020, 59, 14229–14235 CrossRef CAS PubMed .
30. H. Lv, L. Luo, W. Li and P. Du, Inorg. Chem., 2022, 61, 11442–11453 CrossRef CAS PubMed .
31. H. Liu, X. Jian, M. Liu, K. Wang, G. Bai and Y. Zhang, RSC Adv., 2021, 11, 36689–36697 RSC .
32. Y. Pan, H. Lin, R. Hong and D. Zhang, J. Lumin., 2021, 238, 118293 CrossRef CAS .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5tc01422k

This journal is © The Royal Society of Chemistry 2025