Samarium-Activated La2Hf2O7 Nanoparticles as Multifunctional Phosphors.

Recent developments in the field of designing novel nanostructures with various functionalities have pushed the scientific world to design and develop high-quality nanomaterials with multifunctional applications. Here, we propose a new kind of doped metal oxide pyrochlore nanostructure for solid-state phosphor, X-ray scintillator, and optical thermometry. The developed samarium-activated La2Hf2O7 (LHOS) nanoparticles (NPs) emit a narrow and stable red emission with lower color temperature and adequate critical distance under near-UV and X-ray excitations. When the LHOS NPs are exposed to an energetic X-ray beam, the Sm3+ ions situated at the symmetric environment get excited along with those located at the asymmetric environment, which results in a low asymmetry ratio of Sm3+ under radioluminescence compared to photoluminescence. High activation energy and adequate thermal sensitivity of the LHOS NPs highlight their potential as a thermal sensor. Our results indicate that these Sm3+-activated La2Hf2O7 NPs can serve as a multifunctional UV, X-ray, and thermographic phosphor.

Introduction

In the current era of materials science, new kinds of functional materials have been continuously developed to reduce the burden from increasing human demand in the areas of energy, health, environment, food security, agriculture, and so forth.[1−5] Some materials which can perform multiple functions in a system because of their unique properties are known as multifunctional materials. Such materials can reduce human workload and improve our lifestyle by virtue of their efficiency, reliability, cost-efficiency, and scalability. In the areas of photonics and optoelectronics, a phosphor material which can function for solid-state lighting, scintillation, and luminescence thermometer is desirable for meeting a multitude of demands. Phosphor materials synthesized in a nanodomain offer a high electron–hole overlap integral, large light collection, and ease of coating on films/fibers.[6,7] The host-to-dopant energy-transfer process in nanosized phosphors is highly relevant because of their potential to exhibit higher-resolution displays at lower thermal budget relative to their bulk counterparts.[8] Lanthanide ion-doped nanoparticles (NPs) have displayed outstanding photophysical properties such as large Stoke shift, high color purity, higher excited-state lifetime, and substantial quantum efficiency.[7,9] Such unique properties provide obvious advantages for luminescent materials designed toward UV-excited phosphors, X-ray scintillators, and thermographic phosphors (TGPs). Designing efficient nanosized phosphors which display all these application potentials would help solve a variety of issues in the areas of optical technology. This study is intended to meet these challenges and fill the gap.

UV-based and blue light phosphors are used for solid-state lighting, light-emitting diodes, display panels, sensors, biomedical imaging and diagnostics, catalysts, and so forth.[10,11] Their application in white light-emitting diodes is in high demand because of their desirable power output, longevity, environment benignness, and low cost.[12] However, the present commercial white light lamps through a combination of a yellow phosphor (YAG:Ce3+) and a blue light-emitting diode chip (InGaN) suffer from low color rendering index and high correlated color temperature (CCT) problems because of the lack of suitable red components.[13,14] Therefore, continuous search for better red phosphors with low CCT, high color purity, and high structural and thermal stability is undergoing.

Scintillators are in high demand because of their broad applications for ionizing radiation detectors, X-ray imaging, biological and medical diagnostics, photodynamic therapy, and so forth.[15,16] Their single crystal or bulk form suffers from difficult optical property tuning, long processing time, sophisticated instrumentation, and high fabrication cost.[15] NPs and nanoceramics are expected to meet these challenges because of the dominance of surface atoms with strains and electron–hole overlap integral to fine-tune their optical properties.

Moreover, phosphors have been explored for thermal sensing as TGPs because of their temperature-dependent photoluminescence (PL) characteristics such as emission intensity and excited-state lifetime. As luminescence thermometers, TGPs have obvious advantages over other conventional thermal sensing methods as they do not require direct contact, are operational in harsh conditions where large electromagnetic noises exist, and work well in aqueous systems. They also have fast response and high sensitivity and accuracy. Lanthanide-doped materials, such as silicates, garnets, and niobates, have gained significant momentum for such application because of their high spatial resolution and accuracy of temperature gradients.[17] Most of the relevant work has been performed in the visible region of electromagnetic spectrum to determine the temperature of running jet turbines, gas centrifuges, and chemical reaction vessels.[18] However, current TGPs are restricted to bulk samples with moderate sensitivity and intermediate temperature range.

La2Hf2O7 (LHO) is a desirable host for efficient lanthanide-doped phosphor materials because of its high structural stability, wide band gap, the ability to accommodate dopant ions at both La3+ and Hf4+ sites, and high radiation stability, refractive index, and dielectric constant.[19] Y2Ti2O7 pyrochlore has a phonon frequency <712 cm–1, which is expected to decrease for heavier hafnate-based pyrochlores.[20] As a result, LHO could have low nonradiative relaxations and accordingly enhanced luminescence intensity. It also has high stopping power for X- and γ-rays because of its high density of 7.9 g/cm3 and the presence of hafnium with atomic number 72. Sm3+-doped phosphors exhibit sharp emission bands and long excited-state lifetimes, and the large difference between the lowest luminescent level and the highest nonluminescent level of Sm3+ ions makes it an excellent structural probe.[21] To the best of our knowledge, there is no report on designing multifunctional nanosized phosphors using Sm3+-doped LHO (LHOS) NPs as UV, X-ray, and TGPs.

In this paper, we report the exploration of our LHOS NPs synthesized by a molten salt synthesis (MSS) method as a new type of multifunctional UV-excited phosphor, X-ray scintillator, and TGP. Under near-UV excitation, these NPs demonstrate red emission and 2.5% quenching concentration. Magnetic dipole transition (MDT) at ∼568 nm is absent, which suggests a highly asymmetric environment of Sm3+ ions in the LHO lattice and endows high red color purity. Under X-ray irradiation, these NPs show orange-red emission with the presence of intense MDT peak. The LHOS NPs also display favorable thermal stability and good photo-/radioluminescence efficiencies. The high activation energy and desirable thermal sensitivity for luminescence thermometer are attributed to the low phonon energy, low defect density, high crystal quality, and high structural stability of the LHO host. Overall, our LHOS NPs demonstrate multifunctional application potentials as UV, X-ray, and TGPs.

Results and Discussion

Phase, Crystallite Size, and Morphology of the LHOS NPs

The synthesized LHOS NPs are in the ideal pyrochlore phase with Fd3̅m space group (Figure S2) and are nanospheres with particle size between 20 and 70 nm (Figure S3) as investigated by X-ray diffraction (XRD) and scanning electron microscopy (SEM), respectively.

Structural Analysis: Raman and FTIR Spectroscopy

On the basis of the concepts of group theory, the Raman spectra of ideal pyrochlore A2B2O7 compounds (Fd3̅m space group) consist of six well-defined and sharp vibrational peaks in the range 200–1000 cm–1, which are ascribed to A1g + Eg + 4F2g.[22,23] Radius ratio (rA/rB) plays an important role to decide the exact structure of A2B2O7 compounds. When rA/rB > 1.46, an ordered pyrochlore is the most prevalent structure.[24] LHO prefers to stabilize in an ideal pyrochlore structure because rLa (CN = 8) = 1.16 Å and rHf (CN = 6) = 0.710 Å, giving rLa/rHf = 1.63. Figure a shows the Raman spectra of the LHOS NPs that exhibit peaks located approximately at 306, 320, 397, 498, 519, and 615 cm–1 (see deconvoluted spectra in Figure b), which correspond to F2g, Eg, F2g, A1g, F2g, and F2g modes.[25] Specifically, the peaks at 306, 320, and 397 cm–1 are originated from the vibrations of the metal–oxygen bonds (i.e., La–O and Hf–O bonds). The 519 and 615 cm–1 peaks arise from the Hf–O stretching. The 498 cm–1 peak arising from the A1g mode is attributed to the bending of O–Hf–O bond in HfO6 octahedra.[26] The small peak at around 769 cm–1 is ascribed to the distortion of the HfO6 octahedra.[27] Therefore, based on our Raman data, the LHOS NPs have the ideal pyrochlore structure. Moreover, changing of Sm3+ doping concentration (≤10%) does not distort the basic pyrochlore matrix of the LHO host, as can be seen from the Raman spectra.

Figure 1

(a) Raman spectra and (b) vibrational peak characteristics of ideal pyrochlore structure obtained after Gaussian fitting, (c) FTIR spectra, and (d) expanded FTIR spectra at low wavenumbers of the LHOS NPs.

In the Fourier transform infrared (FTIR) spectra of the LHOS NPs (Figure c), the broad-band peaks located at around 3400 cm–1 are attributed to stretching of surface-adsorbed water molecules on the LHOS NPs.[28] The peaks at around 1330–1500 cm–1 are due to bending vibration of OH bond.[29] The IR bands at around 840–920 cm–1 are attributed to wagging vibration of NH band, which may come from the used NH4OH during the coprecipitation process. The IR peaks at around 510 cm–1 are the absorption bands characteristic to pyrochlore A2Hf2O7 compounds, which is consistent with the Raman data shown above.[30] The expanded FTIR spectra in the low wavelength range clearly depict the characteristic absorption of rare earth hafnates (Figure d). The absorption bands characteristic of A2B2O7 hafnates pyrochlore structure was reported by Klee and Weitz.[31] In fact, Sevastyanov and group[30] have observed absorption from Nd2Hf2O7 and Gd2Hf2O7 pyrochlore at around 530–540 cm–1.

PL Spectroscopy

The PL excitation spectra of the LHOS NPs recorded at 607 nm corresponding to the 4G5/2 → 6H5/2 emission of Sm3+ (Figure a) can be divided into two regions. The first region with a very weak band at around 275–300 nm is attributed to the charge transfer from O2– → Sm3+ ion. The second region from 300 to 500 nm of the excitation spectra includes fine peaks at 322, 345, 362, 377, 405, 420, 446, 461, 467, 472, 480, and 490 nm, which correspond to the transitions of Sm3+ ions: 6H5/2 → 2L15/2, 6H5/2 → 4H9/2, 6H5/2 → 4D3/2, 6H5/2 → 4D1/2, 6H5/2 → 4F7/2, 6H5/2 → 4P5/2, 6H5/2 → 4G9/2, 6H5/2 → 4I9/2, 6H5/2 → 4I11/2, 6H5/2 → 4I13/2, and 6H5/2 → 4I15/2, respectively.[32] More interestingly, the Laporte forbidden f–f excitation bands of the LHOS NPs are found to be more intense than the charge-transfer band. This phenomenon indicates that the LHO host has low phonon energy and the energy levels of the doped Sm3+ ions match well with the band structure of LHO for broad phosphor applications. It is supplemented by lower phonon energy of the LHO host (745 cm–1) together with relaxation of f–f transition in the presence of a crystalline field offered by the LHO host.

Figure 2

(a) PL excitation spectra (λem = 607 nm), (b) PL emission spectra (λex = 405 nm), (c) proposed mechanism of host-to-dopant energy transfer, (d) variation of integrated emission intensity in the range of 575–675 nm as a function of Sm3+ doping concentration, (e) relation between log10(I/x) and log10(x) based on the Van Uitert equation, and (f) luminescence decay profiles of the LHOS NPs with Sm3+ doping level from 0.5 to 10.0%. The inset of Figure a shows the magnified excitation spectra in the wavelength range of 260–315 nm.

The three main PL emission peaks of the LHOS NPs under 405 nm excitation are at 605, 653, and 717 nm (Figure b), corresponding to the 4G5/2 → 6H7/2, 4G5/2 → 6H9/2, and 4G5/2 → 6H11/2 transitions of Sm3+ ions.[33] Intriguingly, the MDT 4G5/2 → 6H5/2 at ∼565 nm is absent, indicating that the majority of Sm3+ ions are located at sites without inversion center. The A2B2O7-type ordered pyrochlore structure of the La2Hf2O7 host with LaO8 and HfO6 coordinations is depicted in Figure S4a.[34] Samarium ions are expected to occupy both La and Hf sites, which is also shown pictorially in Figure S4b. On the basis of ionic size and charge, a large fraction of samarium ions is localized at distorted LaO8 scalenohedra.

In our case, they are localized at highly asymmetric sites in the LHO host. This phenomenon is consistent with (i) the large spectral splitting of the 4G5/2 → 6H7/2, 4G5/2 → 6H9/2, and 4G5/2 → 6H11/2 transitions observed (Figure b) and (ii) the high luminescence intensity of the 653 nm peak corresponding to the electric dipole transition (EDT) 4G5/2 → 6H9/2. Moreover, the characteristic 4G5/2 → 6H7/2 transition of Sm3+ ions has mixed electric and magnetic dipoles, which contribute to the orange-red emission at 605 nm as the strongest peak.[35] Because there is no MDT emission and strong EDT at 653 nm, the asymmetry ratio value of the LHOS NPs is much higher than that of Eu3+-doped La2Hf2O7 NPs (∼3.0).[19,36] It is reported that surface defects and cation vacancies are deleterious to PL efficiency where oxygen vacancies may induce improved emissions by virtue of energy transfer.[37] Previously, we observed violet-blue emission from undoped LHO NPs under UV irradiation, which was attributed to the presence of ionized oxygen vacancies in the band gap of the undoped LHO NPs based on density functional theory calculations.[38] In one of our earlier work, we have estimated the band gap of LHO to be 5.63 eV using hybrid exchange–correlation functional (HSE06)[38] and is in close agreement with the experimental value of 5.6 ± 0.1 eV. LHO with a large band gap value of 5.6 eV can provide enough space to accommodate Sm3+ 4f energy levels and effectively reduce the probability of the overlapping between the lowest 4f state of Sm3+ ions and the conduction band maxima of the LHO host.[39] The energy transfer is mediated through ionized oxygen vacancies which are localized within the band gap of LHO and transfer photon energy to excited states of samarium ions. The energy levels of dopant Sm3+ ions are situated within the band gap of the LHO host. On the basis of our earlier report[38,40] and close similarity between Sm3+ and Eu3+, Sm d- and f-states are present in both valence and conduction bands of LHO as both minority and majority spin components. This makes the interaction between O 2p states and Sm d/f states highly feasible to facilitate efficient host-to-dopant energy transfer. On the basis of the Dexter proposal, an effective energy transfer needs a good spectral overlap between the donor emission and the acceptor excitation.[41] To further validate these results, we have shown experimental results on the spectral overlap between the emission of the LHO host (donor) and the excitation of Sm3+ ions (acceptor) (Figure S5a), which suggests an efficient energy transfer from the host to the dopant ions.

To study the role of oxygen vacancies on the PL spectra, both undoped LHO NPs and the LHOS NPs were annealed under an oxygen-deficient atmosphere and vacuum. Reducing atmosphere annealing was avoided to prevent reduction of Hf4+ ions in the samples.[33] The PL spectral results (Figure S5b) demonstrated that the vacuum annealing enhanced the violet-blue emission intensity for the undoped LHO compared to the initially prepared sample in air.[38] For the LHOS NPs, the red emission intensity increased to 2.5 times after vacuum annealing because of the increase of oxygen vacancy concentration. At the same time, the disappearance of the oxygen vacancy-related band at 400 nm from the LHO NPs to the LHOS NPs confirms the efficient energy transfer between the oxygen vacancy of the LHO host and Sm3+ dopant ions in the LHOS NPs.

Concentration Quenching Study

Dopant concentration plays important roles in the luminescence performance of phosphors. The Sm3+ doping level in the LHOS NPs changes their PL intensity but does not affect the spectral profile because of the strong shielding of 4f electrons of Sm3+ ions by the outer lying 5s2 and 5p6 electrons (Figure b).[42] The initial increase of PL emission intensity is ascribed to the increase of Sm3+ doping percentage; therefore, more luminophores are available for excitation by near-UV light at 405 nm. After 2.5 mol % doping level, the distance between Sm3+ ions decreases to an extent that there is enhanced probability of nonradiative energy transfer between adjacent Sm3+ ions as concentration quenching. The quenching concentration of Sm3+ ions of these LHOS NPs is 2.5 mol % (Figure d), which is comparable with other reported Sm3+-doped inorganic hosts in the literature (Table S2).

To understand the underlying nonradiative energy-transfer mechanism, that is, exchange interaction, radiation reabsorption, or multipole–multipole interaction, of these LHOS NPs, we turn to the empirical relationship proposed by Blasse to determine the critical distance (Rc) at which concentration quenching takes place based on eq S1.[43]

On average, there is one activator ion per V/XcN. The calculated critical distance Rc of the LHOS (2.5 mol %) NPs is 30.9 Å, which is much larger than the typical critical distance responsible for energy transfer via exchange interaction (5 Å). Because radiation reabsorption is only effective when the emission spectrum of the donor and the excitation spectrum of the acceptor overlap broadly, which is not the case for the LHOS NPs, we can rule out the radiation reabsorption mechanism here. Therefore, electric multipole–multipole interaction would be the main mechanism responsible for the energy transfer from one Sm3+ ion to another in the LHOS NPs. It means that the Sm3+ ions localized at two LaO8 scalenohedra at a distance less than 30.9 Å would undergo nonradiative energy transfer and leads to luminescence quenching by orbital overlap.

The calculated critical distance Rc of the LHOS NPs is larger compared to the reported values of other lanthanide-doped pyrochlores, such as Nd2Zr2O7:Eu3+ (13.64 Å), Y2Ti2O7:Eu3+ (6.47 Å), Gd2Zr2O7:Sm3+ (6.28 Å), La2Ti2O7:Er3+ (13.8 Å), La2Zr2O7:Eu3+ (5.0 Å), and CaYTiNbO7:Eu3+ (8.0 Å).[44−49] This can be attributed to the low phonon energy and nanocrystalline nature of the LHO host.

We have used the Van Uitert equation to confirm the multipole–multipole interaction as the exact nonradiative energy-transfer mechanisms for our LHOS NPs.[50] If nonradiative energy transfer involves the same acceptor and donor ions (i.e., Sm3+ in the case here), the type of electric multipolar interaction can be deciphered from the variation in emission intensity per unit concentration using the relation S2.

When θ = 3, nonradiative energy transfer proceeds via exchange interaction, θ = 6 corresponds to dipole–dipole (d–d) interaction, θ = 8 corresponds to dipole–quadrupole (d–p) interaction, and θ = 10 corresponds to quadrupole–quadrupole (q–q) interaction.[51] Applying the condition of β(x)θ/3 ≫ 1, eq S2 is simplified as eq S3.

The slope of this equation (−θ/3) obtained by plotting log(I/x) versus log(x) gives the electric multipolar characteristic value θ. For our LHOS NPs, the slope is −2.19814 (Figure e), so θ is approximately 6, corresponding to the dipole–dipole interaction, as similarly observed from La2Ti2O7:Eu.[52]

Luminescence Lifetime Spectroscopy

Understanding local site occupancy is important to interpret the local symmetry around dopant ions in a host and optimize the luminescence properties of phosphors. Population analysis and number of lifetime components from lifetime spectra enable us to predict the local structure of dopant ions in hosts. The luminescence decay profiles of the LHOS NPs (Figure f) under excitation and emission wavelengths of 405 and 607 nm, respectively, can be fitted with a biexponential function using eq S4 with a time resolution of 1–2 μs.

In ordered pyrochlore LHO, La3+ ions exist in an eightfold coordination in highly distorted scalenohedra, whereas Hf4+ ions exist in a sixfold coordination in ideal octahedra, as confirmed by our XRD and Raman spectra data (Figures S2 and 1).[29] Both the lifetime values are attributed to Sm3+ ions localized at La3+ sites in distorted scalenohedra geometry. It is known that there is high concentration of defects in pyrochlore structure. If LaO8 is surrounded by such defects, the LHOS NPs have more nonradiative channels and therefore short decay lifetime. Otherwise, defects are at long distance with long lifetime of the LHOS NPs.[53,54] Specifically, the long-lived component τ2 is ascribed to Sm3+ ions occupying LaO8 scalenohedra sites, which are at long distance from point or structural defects. On the other hand, the fast decaying component τ1 is due to Sm3+ ions localized at LaO8 scalenohedra but at proximity of defects. In addition, both the lifetime values of the LHOS NPs decrease monotonically with increasing Sm3+ doping concentration (Table S3). This can also be seen from the variations of the short- and long-lived lifetimes with Sm3+ concentration (Figure S6). This may be attributed to enhanced probability of nonradiative energy transfer among Sm3+ ions with increasing doping concentration. In our LHOS NPs, Sm3+ ions occupy La3+ sites, and the matching charge does not invoke the need to compensate defects in the LHO lattice. This is an additional advantage of our LHOS NPs because such defects as nonradiative pathways are deleterious and adversely affect their performance.[55]

Colorimetric Performance of the LHOS NPs

As expected, the Commission Internationale de l’Eclairage (CIE) chromaticity diagram (Figure a) and the calculated CIE color coordinates (Table S4) demonstrate that the orange-red light emits from our LHOS NPs under 405 nm excitation because no PL spectral shift is seen as a function of Sm3+ doping concentration (Figure b). CCT is one of the important parameters to evaluate phosphor performance. It is reported that CCT values below 5000 K are ideal to generate artificial warm white light.[56] The CCT values of our LHOS NPs are calculated using McCamy and Kelly approximation mentioned in eq S5.[57,58]

Figure 3

(a) CIE index diagram (excited at 405 nm) and (b) variation of the CCT with Sm3+ doping concentration of the LHOS NPs. The inset of Figure a shows the magnified color coordinates of the LHOS NPs with different Sm3+ concentrations.

The calculated CCT values of the LHOS NPs are in the range of 3057–3437 K (Table S4), and the variation of the CCT values is shown in Figure b. In that context, our LHOS NPs as red phosphors with CCT of 3266 K can be quite useful in designing a high-quality warm white light, which can be further explored for indoor and office lighting.[59] The CCT values reported for various europium- and samarium-doped inorganic oxides such as GdPO4:Eu, LiEu(WO4)2, CsGd(MoO4)2:Eu, LaPO4:Sm, and GdPO4:Sm are 1704, 2175, 1627, 1984, and 1925 K, respectively.[60−63] In comparison to other lanthanide-doped inorganic oxides in terms of CCT, our LHOS-nanosized phosphors have an optimum value as a red phosphor.

Radioluminescence Performance of the LHOS NPs

Upon X-ray excitation, the LHOS NPs present four emission bands at 568, 611, 656, and 722 nm corresponding to 4G5/2 → 6H5/2, 4G5/2 → 6H7/2, 4G5/2 → 6H9/2, and 4G5/2 → 6H11/2 transitions of Sm3+ ions (Figure a). These radioluminescence (RL) spectra are obviously different from the corresponding PL spectra (Figure b). The RL asymmetry ratio was found to be 1.33 for the LHOS NPs (2.5%). The major difference is that the RL spectra show intense MDT 4G5/2 → 6H5/2 peak at 568 nm, which is absent in the PL spectra. This may be attributed to different mechanisms upon UV and X-ray excitations.

Figure 4

(a) RL emission spectra and (b) variation of RL emission intensity as a function of Sm3+ ion concentration of the LHOS NPs.

It is reported that lanthanide dopants are distributed at both La and Hf sites in multiple site lattices such as La2Hf2O7.[38] LHO has ideal pyrochlore structure with distorted LaO8 scalenohedra and perfect HfO6 octahedra.[19] The EDT peak normally results when Sm3+ ions sitting at asymmetric La site is excited, whereas the MDT peak appears because of excitation of Sm ion sitting at symmetric Hf site. The luminescence processes under UV and X-ray irradiations are distinct because different mechanisms are involved. On X-ray irradiation, e––h+ pairs are created because 4f–4f bands of Sm3+ ions and LHO are coexcited.[64] Under UV excitation, only the 4f–4f bands of Sm3+ ions are excited, but not the LHO host. Hence, in the LHOS NPs, a larger fraction of Sm3+ ions at the La site gets excited under UV excitation, leading to intense EDT and absence of MDT emission. On the other hand, both Sm3+ ions at La and Hf sites get excited under X-ray excitation, leading to both EDT and MDT emission peaks.

The nature of heavy constituents is very important for efficient X-ray scintillation process because X-ray absorption intensity varies proportionally to Zeff4/AE3, where Zeff and A are effective atomic number and atomic mass, respectively, and E is the energy of X-ray.[65] Accordingly, LHO is a desirable host for X-ray scintillators because of its high effective atomic number of Hf (72) and density (7.9 g/cc). In the first step, a scintillating material absorbs X-ray through photoelectric effect (PE), which creates electrons and holes. At the beginning, the X-ray (E < few 100 keV) interacts with Hf atom of the pyrochlore LHOS NPs mostly through PE. A similar interaction mechanism was predominant in CsPbBr3 nanocrystals where X-ray interacts with the Pb atom of the perovskite nanocrystals through PE.[66]

As shown by the variation of integrated RL intensity between 575 and 675 nm corresponding to 4G5/2 → 6H7/2 as a function of Sm3+ concentration (Figure b), the maximum RL output is found to be with 2.5% Sm3+ doping concentration, which is similar to the PL results with quenching at higher concentrations. The emission intensity of phosphors is strongly affected by the surface defects present on nanostructured materials. There are two affecting parameters for luminescence properties: surface defects and agglomeration. Surface defects provide nonradiative pathways, whereas agglomeration scatters incident and emitted light. Hence, increasing these parameters reduces PL and RL intensities. Our SEM images show that there is a gradual decrease of particle size from 70 to 22 nm with increasing Sm3+ doping level from 0.5 to 10.0 mol % (Figure S3). The initial small decrease of the RL intensity with the increase of Sm3+ doping from 0.5 to 1.0% is within experimental error. The increase of RL intensity with the increase of Sm3+ doping from 1.0 to 2.5% is endowed by enhanced dopant concentration, which could overturn the negative effects by the decreased particle size and increased surface defect. When the Sm3+ doping level is higher than 2.5%, concentration quenching governs the observed RL trend.

Thermographic Performance of LHOS as a Phosphor

We have further investigated the thermal stability of the LHOS NPs and explored their potential application as TGPs by in situ PL measurements from 298 to 1073 K (Figure a). The LHOS (2.5%) NPs are used as an example because they possess the highest PL intensity. With increasing measurement temperature, the PL emission intensity initially increases and then decreases rather than monotonously decreases as normally seen from various doped phosphors.[38,40] Meanwhile, the well-resolved PL peaks pertaining to the Sm3+ ion broaden until 773 K (Figure b). The PL peaks are masked by broad envelope in the measurement temperature of 673–773 K, while the PL intensity is very weak between 973 and 1073 K. The broad envelope between 673 K up to 773 K may be attributed to large density of thermal defects (such as cation vacancies and cation interstitials) or oxygen vacancies, which can provide nonradiative pathways.[19,38,39] Another possible reason may be the partial reduction of Sm3+ to Sm2+ at high temperature, which is known to display a broad emission in the green region.[67] Techniques to probe defects in situ such as positron annihilation lifetime spectroscopy and to detect the Sm3+ to Sm2+ reduction in situ such as electron spin resonance spectroscopy would be helpful in this regard. Such an instrumental setup is not available to us, but we would try to seek collaborations in the future.

Figure 5

LHOS (2.5%) NPs: (a) in situ PL spectra measured from 298 to 1073 K. (b) Deconvoluted electronic transition peaks of Sm3+ ions at 773 K. (c) Arrhenius plot of the integrated PL intensity of 4G5/2 → 6H7/2 transition from 298 to 1073 K. (d) Variation of Sa and Sr as a function of temperature.

Of Sm3+-doped phosphors, there are several nearly resonant cross-relaxation pathways which would not demonstrate temperature dependence.[68] The temperature-dependent change of PL emission intensity usually arises because the nonradiative transition probability of some particular energy levels is also temperature-dependent.[69] For our LHOS NPs, with the increase of PL measurement temperature, nonradiative transition probability reduces up to 773 K,[29,37] and therefore, the PL emission intensity increases in the temperature range of 298–773 K. However, at temperature higher than 773 K, there is a sudden thermal quench by ∼ 48% from 773 to 873 K (Figure c). This thermal quenching phenomenon occurs when phosphors are supersaturated with thermal heat and can no longer convert the excitation energy to visible light. Sudden and fast thermal quenching at high temperatures is typical of phosphors such as the LHOS NPs wherein nonresonant cross-relaxation channels (by about few hundred wavenumbers) become resonant at high temperature because of energy activation.[68]

We have extrapolated the activation energy value by Arrhenius fitting of the integrated PL intensity of 4G5/2 → 6H7/2 transition in the temperature range of 298–773 K based on eqs S6 and S7. The relationship of and 1/kT holds a linear fit (Figure S7) with a slope of −0.153, which is proportional to the involved activation energy of the LHOS-2.5 NPs. Therefore, the activation energy is equal to 0.153 eV for the thermal quenching.

The absolute temperature sensitivity (Sa) is a vital parameter to quantitatively evaluate the optical thermometric ability of phosphors. Sa denotes the theoretical variation rate in fluorescence intensity ratio with respect to temperature and can be expressed by eq S8.

The absolute temperature sensitivity of our LHOS NPs increases with the increasing measurement temperature and reaches a maximum value of 0.98 K–1 (Figure d).

Meanwhile, relative temperature sensitivity (Sr) is another critical parameter to quantify the temperature sensing properties of thermometric materials. It is expressed by eq S9 as the relative change of the integrated PL intensity (I) with respect to temperature (T).

There is a gradual decrease of the Sr value with increasing measurement temperature for 4G5/2 → 6H7/2 transition of Sm3+ ions, which is similar to the trend observed from GdVO4:Sm3+(Figure d).[70] The Sr value from the LHOS NPs reaches a maximum of 3.79 % K–1, which is the highest compared with typical temperature sensors (Table ). The high activation energy and relatively good thermal sensitivity of synthesized LHOS NPs could be attributed to (a) low phonon energy of LHO, (b) high structural stability of LHO, and (c) little lattice strain or charge compensating defects as the majority of Sm3+ ions occupy La3+ sites because of similar ionic radii and the same charge.

Table 1

Relative Temperature Sensitivity Sr of Representative Optical Thermometric Materials

materials	maximum Sr (% K–1)	temperature range (K)	refs
Gd2O3:Er3+/Yb3+	0.39	300–900	(71)
NaLuF4:Ho3+/Yb3+	0.12	350–750	(72)
Yb3Al5O12:Er3+	0.48	295–973	(73)
NaLuF4:Gd3+	0.29	298–523	(74)
Y2MgTiO6:Mn4+	0.14	10–513	(75)
GdVO4:Sm3+	1.41	393–603	(70)
La2Hf2O7:Sm3+	3.79	298–873	this work

Thermographic and display panel applications of phosphors require them to possess high structural stability over a wide range of temperatures. Other than in situ PL measurements, we have also taken in situ XRD and Raman measurements on the LHOS (2.5%) NPs from 298 to 1148 and 1173 K, respectively (Figure S8). It is clearly seen that the LHOS NPs are structurally stable. High thermal stability of phosphors is favorable for luminescence thermometer application with high PL and RL efficiencies.

Conclusions

In summary, our LHOS NPs displayed strong orange-red emission with high color purity and low CCT under near-UV excitation at 405 nm, which makes them compatible with commercial NUV light-emitting diodes. The LHOS NPs are also applicable as an X-ray phosphor by converting highly energetic X-ray into orange-red light. Interestingly, their electronic and MDTs are different under UV and X-ray excitations. Meanwhile, the LHOS NPs demonstrate highest thermal sensitivity among typical temperature sensors. Therefore, the superior emission characteristics, efficient RL output, and good thermal sensing capability make the LHOS NPs potential candidates for multifunctional applications as phosphors, scintillators, and luminescence thermometry.

Experimental Section

Synthesis of the LHOS NPs

A set of six samples of LHO/x % Sm3+, where x = 0.5, 1, 2.5, 5, 7.5, and 10, were prepared via a hybrid coprecipitation and MSS. Lanthanum nitrate hexahydrate (La(NO3)3·6H2O, 99.0%), hafnium dichloride oxide octahydrate (HfCl2O·8H2O, 98%), and samarium nitrate (Sm(NO3)3·6H2O, 99.9%) were used as the starting reactants with no further purification. They were mixed and dissolved in distilled water (18.2 MΩ at 25 °C) together at an appropriate stoichiometric ratio and kept stirring for around 30 min. The solution was then titrated with 10% ammonium hydroxide solution (diluted from 28 to 30% NH4OH) over a period of 2 h. The formed precipitate was then washed with deionized water, filtered, dried, and then mixed with potassium nitrate (KNO3, 99.9%) and sodium nitrate (NaNO3, 98%) in a stoichiometric ratio of (1:30:30). The mixture was finely grinded using a mortal and a pestle before being annealed at 650 °C for 6 h. The final resulting powder was washed and dried to obtain the final pure LHOS NPs. The schematic of the synthesis process employed for the LHOS NPs is depicted in Figure S1.

Characterization of the LHOS NPs

The XRD patterns were collected using a Rigaku MiniFlex X-ray diffractometer with a Cu Kα1 radiation (λ = 0.15406 nm, 30 kV, and 15 mA). The scanning mode used was 2θ, with a scanning range from 10° to 90° and a scanning step size of 0.05° with a scanning rate of 2° min–1. FTIR spectra were collected using an FTIR ALPHA II’s Platinum ATR single reflection diamond ATR module. Raman spectra were recorded on a Bruker SENTERRA Raman spectrometer (Bruker Optics SENTERRA R200) using a 785 nm laser with 100 mW power.

In situ Raman measurements were performed using a custom-built micro-Raman setup. The samples were placed in a high-temperature microscope stage (TS-1500, Linkam) and were excited with a continuous wave diode-pumped solid-state laser (Excelsior, Spectra Physics, 532 nm) through an upright microscope using a 50× long-working distance objective with NA (numeric aperture) = 0.5. The typical incident laser power on a sample was maintained at ∼100 μW. The scattered Raman light was analyzed by a spectrometer (SpectraPro 2300i, Acton, f = 0.3 m) that was coupled to the microscope and equipped with a 1800 grooves/mm grating and a CCD camera (Pixis 256BR, Princeton Instruments).

The optical properties of the NPs were evaluated using an Edinburgh Instrument FLS 980 fluorometer system equipped with both a steady-state and a pulsed Xe lamp source. The Xe-pulsed source has a frequency range of 1–100 Hz. RL spectra were acquired with a silver target as the X-ray source with λ = 0.52 Å at a power of 12 W (60 kV and 200 μA). The silver target was adapted to the Edinburgh Instruments FLS 980 fluorimeter system. No filters were used during the process. Lead shielding was used to block the incoming radiation lower than 2 mR/h.