Solid-state reaction synthesis and optical property analysis of LaF 3 ­ LaOF:Yb 3 + / Ho 3 + upconversion phosphors

In this study, LaF 3 ­ LaOF:Yb 3 + / Ho 3 + upconversion (UC) phosphors were synthesized via a solid-state reaction and their optical properties were analyzed. The analysis of photoluminescence (PL) characteristics revealed strong emission at a wavelength of 543 nm for molar ratios of La:Yb:Ho = 1: x : y , where x = 0.04 ­¹ 0.06 and y = 0.01 ­¹ 0.03. The pump power dependence of phosphor PL and the ﬂ uorescence lifetime were then measured, and the mechanism of UC PL is discussed.


Introduction
Downconversion, also known as the Stokes shift, is a technique for converting short-wavelength electromagnetic waves, such as UV light, into longer-wavelength electromagnetic waves. Upconversion (UC), or the anti-Stokes shift, on the other hand, is a technique for converting longer-wavelength electromagnetic waves, such as near-IR (NIR), into shorter-wavelength electromagnetic waves. 1) Studies on UC have been conducted since 1966; UC phosphors have been synthesized by introducing rare-earth elements into the host material. 2) In recent years, rare-earth ions have been doped at higher concentrations than before with the aim of avoiding concentration quenching. 3) UC phosphors have attracted considerable attention in bioimaging applications owing to the high tissue penetration depth of NIR excitation and the low autofluorescence background at NIR wavelengths. 4) In addition, UC phosphors are expected to increase the efficiency of solar cells and to be used in noncontact temperature sensors. 5), 6) Fluoride has attracted attention as a host material for UC phosphors because of its low phonon energy; in particular, NaYF 4 and NaGdF 4 have been studied as promising host materials. 7)12) In this study, LaF 3 was used as a host material. 13) 15) In addition to its low phonon energy, LaF 3 has another useful property as a host material: doping with rare-earth ions does not distort the crystal lattice because the ionic radius of La in the +3 oxidation state is close to that of other rare-earth elements in the same oxidation state. 16) Various combinations of doping elements in UC phosphors have been reported, such as YbEr and Yb Tm. 17)20) In this study, we used the YbHo combination; Ho can be used to obtain green ( 5 F 4 , 5 S 2 ¼ 5 I 8 ) and red ( 5 F 5 ¼ 5 I 8 ) emissions. 21), 22) We previously synthesized LaOF:Yb 3+ /Er 3+ UC phosphors and analyzed their optical properties. 23) The photoluminescence (PL) intensity of LaOF was lower than that of LaF 3 , indicating that practical applications for a host material made from LaOF alone would be limited. Also, in general, LaF 3 is considered to have low chemical stability, and it is thought that there will be challenges in practical application if the host material is only LaF 3 . Therefore, the aim of this study is to synthesize a LaF 3 LaOF:Yb 3+ /Ho 3+ UC phosphors with excellent PL intensity and chemical stability and to analyze its optical properties. The phonon energy of LaF 3 is lower than that of LaOF, and LaF 3 LaOF:Yb 3+ /Ho 3+ is considered to emit at a higher intensity than LaOF:Yb 3+ /Ho 3+ . Also, LaF 3 LaOF composites are considered to be more chemically stable than LaF 3 .
In previous studies on LaF 3 :Yb 3+ /Ho 3+ , the samples were synthesized in solution, 24)29) and there have been no reports of solid-phase synthesis. We describe the UC properties of LaF 3 LaOF:Yb 3+ /Ho 3+ synthesized using a solid-state reaction.

Experimental conditions
To synthesize the sample by a solid-state reaction, LaF 3 , Yb 2 O 3 , and Ho 2 O 3 powders, obtained from Kojundo Chemical Lab Co., were mixed to achieve molar ratios of La:Yb:Ho = 1:x:y (x = 0.03, 0.04, 0.05, 0.06, and 0.07 and y = 0.01, 0.02, 0.03, 0.04, and 0.05). Then, the powder mixture was pressurized into pellets, which were placed in an Al 2 O 3 crucible and heated in air at 1100°C for 1 h.
The PL characteristics, crystal structure, pump power dependence of PL intensity, and fluorescence lifetime were analyzed. For the PL characteristics, an NIR laser (wavelength = 980 nm; output = 200 mW) was used as the excitation source. A Flame series spectrometer (Ocean Optics) was used for the analysis. PL characteristics were measured five times for each sample and the average value was calculated. An Ultima IV X-ray diffractometer (XRD) (Rigaku Corporation) was used for the crystal structure analysis. Neutral density (ND) filters were used to modify the excitation light output when measuring the pump power dependence of PL intensity. ND filters with transmittances of 12.5, 25, and 50 % were placed in front of the excitation light source to modify its pump power.
The fluorescence lifetime was measured using the method reported by Zhu. 30) An electric motor (Mabuchi Motor Co. model FA-130RA) was used to rotate the samples. An iPhone 8 (Apple Inc.) digital camera was used to obtain fluorescence images, and the iOS version of Lightroom software (Adobe) was used for image processing. The sample was fixed to the motor shaft with double-sided tape and imaged during and after excitation by NIR while being rotated by the motor at 4800 rpm. After the images were taken, they were analyzed using the free software ImageJ. Images were divided into the three primary colors of light, and the fluorescence lifetime was calculated by measuring the rate of decay of luminance in the green and red spectral bands after NIR excitation had ceased.

Results and discussion
XRD patterns of LaF 3 LaOF:Yb 3+ /Ho 3+ UC phosphors with differing concentrations of Yb 3+ dopant are shown in  Fig. 1 shows the XRD patterns of LaOF (01-089-5168) and LaF 3 (01-082-0684) for reference, which were obtained from the Inorganic Crystal Structure Database (ICSD). The crystal structures of LaOF (01-089-5168) and LaF 3 (01-082-0684) are tetragonal and hexagonal, respectively. Major peaks of LaOF and LaF 3 were observed at 2ª = 26.6°and 2ª = 27.6°in all samples, respectively, indicating that each sample is primarily composed of LaOF and LaF 3 . Figure 2 illustrates the PL characteristics of the samples. The horizontal and vertical axes in Fig. 2(a) are wavelength and emission intensity, respectively, and each spectrum shows the result for a different Yb 3+ doping level. For all samples, peaks were observed at wavelengths of 543 nm ( 5 F 4 , 5 S 2 ¼ 5 I 8 ) and 654 nm ( 5 F 5 ¼ 5 I 8 ). The  highest PL emission intensity was observed at the molar ratio of La:Yb:Ho = 1:0.06:0.01. The horizontal and vertical axes in Fig. 2(b) respectively show the Yb 3+ doping level and the ratio of PL emission intensity (G/R ratio) at wavelengths of 543 nm ( 5 F 4 , 5 S 2 ¼ 5 I 8 ) and 654 nm ( 5 F 5 ¼ 5 I 8 ). In the range of Yb 3+ mole fractions used in the experiment, the green emission intensity at 543 nm was dominant and was more than 10 times stronger than the red emission intensity at 654 nm. The G/R ratio of samples with Yb = 0.040.06 was ³14. XRD patterns of LaF 3 LaOF:Yb 3+ /Ho 3+ UC phosphors with differing concentrations of Ho 3+ dopant are shown in Fig. 3(a). The molar ratios of the samples are La:Yb:Ho = 1:0.06:y, where y = 0.01, 0.02, 0.03, 0.04, and 0.05. The XRD patterns of LaOF (01-089-5168) and LaF 3 (01-082-0684) from ICSD are shown in Fig. 3(a) for reference. Results indicate that each sample primarily comprises LaOF and LaF 3 . The peak value of LaF 3 was greater than that of LaOF in the samples with Ho = 0.01 and 0.02.
The analysis result of the peak positions in LaF 3 LaOF:Yb 3+ /Ho 3+ UC phosphor are shown in Fig. 3(b). The unit cell parameters and volume of LaF 3 LaOF: Yb 3+ /Ho 3+ UC phosphor are shown in Table 1. The molar ratio is La:Yb:Ho = 1:0.06:0.01. The lattice parameters of LaF 3 LaOF:Yb 3+ /Ho 3+ UC phosphor are refined by the least-squares method. For comparison, Table 1 also includes the unit cell parameters and volume of ICSD. Comparing the ICSD with the analysis of this study, it can be seen that the rate of decrease in the volume of LaF 3 is greater than the rate of decrease in the volume of LaOF. The ionic radii of La 3+ (coordination number = 8), Ho 3+ (coordination number = 8), and Yb 3+ (coordination number = 8) are 1.160, 1.015, and 0.985 ¡, respectively. 31) Therefore, it is assumed that the replacement of La 3+ in LaF 3 by Yb 3+ and Ho 3+ caused the decreased the volume of LaF 3 .
The resulting PL characteristics are shown in Fig. 4. Each spectrum in Fig. 4(a) shows the dependence on the Ho 3+ doping level. Peaks at wavelengths of 543 nm ( 5 F 4 , 5 S 2 ¼ 5 I 8 ) and 654 nm ( 5 F 5 ¼ 5 I 8 ) were observed for all samples, with the same spectral distribution as in Fig. 2. Figure 4(b) shows the G/R ratio versus the Ho 3+ doping level. At molar ratios of Ho = 0.010.03, the emission, dominated by the 543 nm peak similarly to in Fig. 2, showed a G/R ratio of ³1; however, the G/R ratio then decreased substantially when the Ho concentration was >0.04. The G/R ratio at Ho = 0.05 was found to be about one-fifth of that at Ho = 0.03.
To examine the UC mechanism, we analyzed the dependence of PL intensity on pump power. The PL intensity I UP of UC depends on the output P and is expressed by where n indicates the number of pumping photons required for excitation from the ground state to the excited state. 22) The pump power dependence of PL intensity is shown in Fig. 5.   wavelengths of 543 and 654 nm were observed to be 1.63 and 1.42, respectively. Both slopes are between 1 and 2, indicating that the PL emission is a two-photon process. 32) The reason that both slope values are <2 is the energy loss due to cross-relaxation (CR) effects and non-radiative (NR) transitions. 32 , it is inferred that the energy loss due to CR effects and NR transitions is smaller than that of the other samples. The slope could not be calculated for the molar ratio of Ho = 0.05, which is therefore absent from Fig. 5(c), because the PL emission intensity was too low. The energy level diagram of Yb 3+ and Ho 3+ in LaF 3 LaOF and UC mechanisms under 980 nm excitation are shown in Fig. 5(d). In the green emission [ 5 F 4 , 5 S 2 (Ho 3+ ) ¼ 5 I 8 (Ho 3+ )], two successive energy transfer (ET) processes [ 5 I 8 (Ho 3+ ) + 2 F 5/2 (Yb 3+ ) ¼ 5 I 6 (Ho 3+ ) + 2 F 7/2 (Yb 3+ )] and [ 5 I 6 (Ho 3+ ) + 2 F 5/2 (Yb 3+ ) ¼ 5 F 4 , 5 S 2 (Ho 3+ ) + 2 F 7/2 (Yb 3+ )] may be involved in the population of the 5 F 4 , 5 S 2 excited states. This green UC emission is a two-photon process, consistent with the results in Figs. 5(a)5(c). On the other hand, in the red emission [ 5 F 5 (Ho 3+ ) ¼ 5 I 8 (Ho 3+ )], there are two pathways to populate in the excited state 5 F 5 level of Ho 3+ . The first is the NR transition from the upper 5 F 4 , 5 S 2 level to the 5 F 5 level. The second is the relaxation from the 5 I 6 level to the 5 I 7 level by the NR process and the excitation to the 5 F 5 level by the ET process [ 5 I 7 (Ho 3+ ) + 2 F 5/2 (Yb 3+ ) ¼ 5 F 5 (Ho 3+ ) + 2 F 7/2 (Yb 3+ )]. In addition, the population of 5 I 6 level via the CR process [ 5 F 4 , 5 S 2 (Ho 3+ ) + 5 I 7 (Ho 3+ ) ¼ 5 F 5 (Ho 3+ ) + 5 I 6 (Ho 3+ )] can't be neglected at relatively high concentrations. 32) The fluorescence lifetime¸was calculated by analyzing the images using where z is PL intensity, t is delay time, and A 0 and A 1 are constants. 25) The fluorescence lifetime results are shown in The decrease in fluorescence lifetime at Yb = 0.07 in Fig. 6(b) may be attributed to the enhanced energy back transfer (EBT) process between Yb 3+ and Ho 3+ ions: 5 F 4 , 5 S 2 (Ho 3+ ) + 2 F 7/2 (Yb 3+ ) ¼ 5 I 6 (Ho 3+ ) + 2 F 5/2 (Yb 3+ ). 33) Furthermore, the decrease in fluorescence lifetime with increasing Ho 3+ concentration in Fig. 6(c) may be caused by CR among Ho 3+ ions. 33)

Conclusion
UC phosphors of LaF 3 LaOF:Yb 3+ /Ho 3+ were synthesized by a solid-state reaction. Several samples were synthesized with differing concentrations of rare-earth dopants. The crystal structure was analyzed via XRD, and LaF 3 and LaOF peaks were detected in all samples. The analysis of PL characteristics revealed strong emission at the wavelength of 543 nm for the molar ratios of La:Yb:Ho = 1:x:y, where x = 0.040.06 and y = 0.01 0.03. The pump power dependence of PL intensity showed that the PL emission of the samples is a two-photon process. Furthermore, samples with the highest G/R ratios had the smallest energy losses. This was due to CR effects and NR transitions. A decrease in fluorescence lifetime was observed at Yb = 0.07, which may be attributed to EBT. Moreover, a decrease in fluorescence lifetime was observed with increasing Ho 3+ concentration owing to the CR of absorbed energy among Ho 3+ ions.