Sensitivity Modulation of Upconverting ... - ACS Publications

1 downloads 0 Views 3MB Size Report
Oct 19, 2016 - School of Sciences, Chongqing University of Posts and ... Nanoscale Biophotonics (CNBP), Macquarie University, North Ryde 2109, Australia.
Research Article www.acsami.org

Sensitivity Modulation of Upconverting Thermometry through Engineering Phonon Energy of a Matrix Hao Suo,† Chongfeng Guo,*,† Jiming Zheng,† Bo Zhou,‡ Chonggeng Ma,§ Xiaoqi Zhao,† Ting Li,† Ping Guo,† and Ewa M. Goldys∥ †

National Key Laboratory of Photoelectric Technology and Functional Materials (Culture Base) in Shaanxi Province, National Photoelectric Technology and Functional Materials & Application of Science and Technology International Cooperation Base, Institute of Photonics & Photon-Technology, Northwest University, Xi’an 710069, China ‡ Institute of Modern Physics, Northwest University, Xi’an 710069, China § School of Sciences, Chongqing University of Posts and Telecommunications, Chongqing 400065, China ∥ ARC Centre of Excellence for Nanoscale Biophotonics (CNBP), Macquarie University, North Ryde 2109, Australia S Supporting Information *

ABSTRACT: Investigation of the unclear influential factors to thermal sensing capability is the only way to achieve highly sensitive thermometry, which is greatly needed to meet the growing demand for potential sensing applications. Here, the effect from the phonon energy of a matrix on the sensitivity of upconversion (UC) microthermometers is elaborately discussed using a controllable method. Uniform truncated octahedral YF3:Er3+/Yb3+ microcrystals were prepared by a hydrothermal approach, and phase transformation from YF3 to YOF and Y2O3 with nearly unchanged morphology and size was successfully realized by controlling the annealing temperature. The phonon energies of blank matrixes were determined by FT-IR spectra and Raman scattering. Upon 980 nm excitation, phonon energy-dependent UC emitting color was finely tuned from green to yellow for three samples, and the mechanisms were proposed. Thermal sensing behaviors based on the TCLs (2H11/2/4S3/2) were evaluated, and the sensitivities gradually grew with the increase in the matrix’s phonon energy. According to chemical bond theory and first-principle calculations, the most intrinsic factors associated with thermometric ability were qualitatively demonstrated through analyzing the inner relation between the phonon energy and bond covalency. The exciting results provide guiding insights into employing appropriate host materials with desired thermometric ability while offering the possibility of highly accurate measurement of temperature. KEYWORDS: upconversion, thermometer, phonon energy, bond covalency, sensitivity field.10 Especially, motivated by smart mini-size technology, FIR-based upconversion (UC) luminescent nanothermometers have been extensively explored in noninvasive subcellular resolution and biologically relevant thermal sensing owing to its deep tissue penetration and minimized background autoluminescence of low-energy near-infrared (NIR) excitation light, low toxicity, and excellent physical and chemical stability.8,11−15 Er3+/Yb3+ codoped systems are typical candidates for optical thermometry, utilizing closely spaced 4S3/2/2H11/2 levels of activator Er3+ as TCLs (ΔE ≈ 800 cm−1) and a broad absorption cross section around 980 nm of Yb3+ ion as a sensitizer to suppress relative detection deviation.16−18 Up to now, most recent publications in the thermometric field are mainly focused on achieving highly accurate sensing or developing multifunctional integrated micro- or nanosized

1. INTRODUCTION As a thermodynamic parameter, temperature determines many properties of matter and plays a dominant role in scientific research, industrial manufacturing processes, and life activities, and its fast and accurate measurement of the objects at different scales is attracting growing interests. Compared with traditional thermal sensing methods, optical thermometry based on the fluorescence intensity ratio (FIR) technique provides fast response, excellent accuracy, high spatial resolution, and tunable size.1−4 Such an optical thermometric method relies on the temperature-dependent intensity ratio between the emissions arising from two adjacent thermal coupled levels (TCLs) with energy separations in the range of 200 cm−1 ≤ ΔE ≤ 2000 cm−1 (for example, Er3+:4S3/2/2H11/2; Tm3+:3F2, 3/3H4; Nd3+:4F7/2/4F3/2, 4F7/2/4F5/2, 4F5/2/4F3/2; Ho3+:5G6/3K8; etc.), which is hardly affected by the fluctuation of excitation power and signal losses.5−9 These advantages endow optical thermometers with extensive and great application prospects in fire detection, oil refinery, power stations, and biomedical © 2016 American Chemical Society

Received: September 25, 2016 Accepted: October 19, 2016 Published: October 19, 2016 30312

DOI: 10.1021/acsami.6b12176 ACS Appl. Mater. Interfaces 2016, 8, 30312−30319

Research Article

ACS Applied Materials & Interfaces

Figure 1. XRD patterns of Er3+/Yb3+ (a) codoped YF3, (b) YOF, and (c) Y2O3 with their standard profiles as well as the corresponding (d) SEM (left column) and TEM (right column) images (scale bars: 1 μm). Er, Yb, Y; 99.99%, Shanghai Yuelong Nonferrous Metals Ltd.) was first dissolved in diluted HNO3 solution (AR); excessive HNO3 was volatilized by evaporation, and deionized water was introduced with vigorous stirring to form 50 mL of a Ln(NO3)3 solution. Subsequently, 0.0175 mol of NH4F was added with constant stirring for 1 h, and then the resulting solution with white colloidal precipitate (fixed at 70 mL) was transferred into a 100 mL Teflon-lined autoclave, which was sealed and maintained at 180 °C for 24 h. After the autoclave cooled to room temperature, the obtained white precipitate was collected by centrifugation, washed with deionized water and ethanol three times, dried at 70 °C in air for 24 h, and finally crushed in an agate mortar to obtain the final sample. The Yb3+ sensitized Er3+ doped YOF and Y2O3 microcrystals were obtained through annealing above YF3:Er3+/Yb3+ samples at 850 and 1100 °C for 2 h. Measurements and Characterization. Powder X-ray diffraction (XRD) data were collected using a Rigaku-Dmax 3C powder diffractometer (Rigaku Corp, Tokyo, Japan) with Cu Kα (λ = 1.54056 Å) radiation. The morphologies of three sample particles were carried out using a field emission scanning electron microscope (FESEM, Hitachi SU-8010) and a transmission electron microscope (TEM, FEI TF-20: accelerating voltage 200 kV). Fourier transform infrared (FT-IR) spectra were performed on an EQUINOX55 spectrometer (Bruker Optics, Germany), and Raman measurements were carried out using an Alpha 500R Confocal Raman Microscopy System (WITec, Germany) equipped with a 633 nm laser and EMCCD (UHTS 300). The UC emission spectra at different temperature were measured on a FLS920 fluorescence spectrophotometer assembled with an external 980 nm semiconductor laser as excitation source and an Oxford OptistatDN2 nitrogen cryogenics temperature controlling system. To reduce error, the duration of stay at the measured temperature was set at 0.5 h. Single-particle luminescent microscopy was performed on a BX43-OLYMPUS microscope equipped with a 976 nm semiconductor laser as the excitation source. The fractional covalence of chemical bond from center ion to ith ligand is symbolized by fc and calculated using the dielectric theory of the chemical bond for complex crystals with known crystal structure and refractive index (n).31,32 The first-principles calculations of total electronic energy and frontier molecular orbital of two typical chemical bonds Y−F and Y−O were performed by utilizing the LDA augmented with gradient correction approximation (GGA) functional in the Perdew−Wang functional form in the Amsterdam density functional (ADF) package.

systems. The temperature sensitivity (S) is the most crucial parameter for evaluating thermometric capability, and realizing high S is a permanent pursuit in the sensing field, but its influencing factor is still an open question for researchers despite tremendous efforts that have been devoted to figuring it out.11,19−23 For the FIR-based UC thermometers, employing hosts with low phonon energy seems to be the most common way to achieve high UC efficiency by inhibiting the electron− phonon coupling.14 However, according to the reported studies carried out within the Er3+/Yb3+ codoped systems, fluoridebased thermometers with relative low phonon energy and high UC efficiency expectedly did not exhibit high sensitivity, for example, NaYF4 (Ehost ∼ 350 cm−1, S ∼ 0.0020 K−1), YF3 (Ehost ∼ 360 cm−1, S ∼ 0.0025 K−1), etc.24−26 On the contrary, outstanding thermometric ability was realized in many oxide matrixes with higher lattice vibration energy, for example, niobates (Ehost ∼ 700 cm−1, S ∼ 0.0073 K−1), vanadates/ tungstates (Ehost ∼ 800 cm−1, S ∼ 0.0116 K−1), double molybdates (Ehost ∼ 880 cm−1, S ∼ 0.0180 K−1), etc.27−30 The above results imply that the sensitivity of optical temperature sensors could be manipulated by engineering the phonon energy of hosts. For the current framework, the phonon energy-dependent temperature sensing capability and UC luminescent property were elaborately discussed in Er3+/Yb3+ ions pair doped with fluoride, oxyfluoride, and oxide using a controllable method. In our design, uniform truncated octahedral YF3:Er3+/Yb3+ microcrystals were synthesized by a hydrothermal procedure, and phase transformation from YF3 to YOF and Y2O3 took place through control of heating temperature without obviously changing the morphology and size. The variation in phonon energy greatly affects the UC luminescent property and thermal sensing ability. Importantly, thorough elucidations about the inner connection between sensitivity and lattice vibration energy were conducted according to the first-principles calculation.

2. EXPERIMENTAL SECTION Sample Preparation. The present YF3:Er3+/Yb3+ phosphors with truncated octahedral morphology were prepared using a hydrothermal approach. In a typical synthesis procedure, 0.0025 mol of Ln2O3 (Ln = 30313

DOI: 10.1021/acsami.6b12176 ACS Appl. Mater. Interfaces 2016, 8, 30312−30319

Research Article

ACS Applied Materials & Interfaces

3. RESULTS AND DISCUSSION The phase-selective synthesis from a single-source precursor after hydrothermal reaction was achieved by controlling heating temperature, and XRD patterns of the as-prepared samples are displayed in Figure 1. For the precursor after hydrothermal treatment (YF3:1% Er3+/5% Yb3+), all the diffraction peaks (Figure 1a) can be indexed well to pure YF3 crystal with an orthorhombic structure (JCPDS 74-0911), which indicates that the introduction of activator Er3+ and sensitizer Yb3+ do not cause any impurity peaks. After the precursor was annealed at 850 and 1100 °C for 2 h in air (Figures 1b and c), two resulting products were identified to be the well-crystallized rhombohedral phase of YOF (JCPDS 71-2100) and cubic-structured Y2O3 (JCPDS 76-0151) without any secondary phases, suggesting that YF3 can be successfully transformed into high purity YOF and Y2O3 through calcination treatment at different temperatures. The corresponding high resolution TEM (HRTEM) images displayed in Figure S1 further prove the formation of YF3, YOF, and Y2O3. To shed light on the morphology and size evolution of the obtained samples, SEM and TEM images of Er3+/Yb3+ codoped YF3, YOF, and Y2O3 were examined and are shown in Figure 1d. The obtained YF3:Er3+/Yb3+ particles exhibit highly uniform and monodispersed truncated octahedral morphology with an average length of approximately 1.5 μm and smooth surfaces (Figure S1). TEM observation provides further insight into a single truncated octahedron in which a rhombic cross-section of about 1.5 μm side lengths could be clearly seen. After an annealing treatment of YF3 precursors, the obtained Er3+/Yb3+ doped YOF and Y2O3 products still maintain well-defined and regular truncated octahedral microstructure with an almost unchanged side length of about 1.5 μm. Although little aggregations and rough surfaces appeared in the annealed samples, phase transformation successfully occurred without obviously changing morphology and particle size. The EDX spectra (Figure S2) further verify the occurrence of conversion process from YF3 to YOF and Y2O3 through calcination treatment in which the atomic ratios of Y:F, Y:O:F, and Y:O in YF3, YOF, and Y2O3 samples are close to theoretical values 1:3, 1:1:1, and 2:3, respectively, with a small quantity of Er3+ and Yb3+. It is generally accepted that phonon energy (lattice vibration energy) exerts great effect on UC luminescence through a phonon-assisted nonradiative transition process. Thus, the FTIR spectra of blank YF3, YOF, and Y2O3 samples were measured in the range from 400 to 4000 cm−1 to determine their phonon energies, as displayed in Figure 2a. It is clearly observed that the strongest phonon cutoff energy bands are centered at 482 and 566 cm−1 for YOF and Y2O3, which demonstrates that their phonon energies are approximately 482 and 566 cm−1, respectively. Because the position of maximum peaks is out of the detection range (beyond 400 cm−1) for YF3, a Raman scattering spectrum was employed to accurately measure its phonon energy, as shown in Figure 2b. The phonon frequency of the strongest vibration peak is around 358 cm−1, corresponding to B2g and Ag vibration modes, which is in accordance with the calculation result based on first-principles (Figure S3), endowing the phonon energy of YF3 at about 358 cm−1.33 The above experimental results and theoretical analysis indicated that the phonon energy gradually increases from fluoride to oxide and agrees well with previous results (YF3 ∼ 350 cm−1; YOF ∼ 400 cm−1; Y2O3 ∼ 600 cm−1).34,35

Figure 2. FT-IR spectra of blank YF3, YOF, and Y2O3 microcrystals (a) along with the Raman spectrum of the YF3 sample (b).

With the aim of illuminating the impact of phase conversion on UC fluorescence, the UC emission spectra at room temperature of the as-synthesized Er3+/Yb3+ doped YF3, YOF, and Y2O3 powder samples with similar morphology and particle size were presented in Figure 3a under the excitation of 980 nm light (P = 20 mW). It can be clearly observed that three UC spectra are mainly composed of two green emission bands and a red emission peak attributed to the 2H11/2, 4S3/2 → 4I15/2, and 4 F9/2 → 4I15/2 intrinsic transitions of the Er3+ ion, but their peak shapes are completely different from each other due to distinct crystal field environment, and the emission bands were split into several incisive peaks, resulting from the effect of spin− orbit coupling and electronic interactions.36 Intriguingly, the integrated emission intensity ratios of red to green strongly depend on the matrixes and increase from 0.35 (YF3) through 1.30 (YOF) to 2.09 (Y2O3), leading to the alteration of output emission color from pure green through yellowish green to yellow, respectively, as displayed in CIE chromaticity diagram (Figure S4a) and in the luminescent digital photographs of samples dispersed in ethanol (insets of Figure 3a) with the 980 nm excitation. Because the emission intensity ratio of samples may change in different solvents in comparison with that of powder,37 single-particle luminescent microscopy equipped with a 976 nm laser was utilized to further illuminate the uniform UC luminescent color of individual truncated octahedral microcrystals. As presented in the bright and dark fields (Figure 3b), Er3+/Yb3+ codoped YF3, YOF, and Y2O3 single particles exhibit similar microstructures with well-defined and regular truncated octahedral morphology (1.5 μm side length) and UC luminescent colors nearly the same as those of the corresponding samples dispersed in ethanol. The crucial mechanism responsible for the modulation of emitting color could be expressed in terms of the phonon energy-dependent nonradiative process. When the energy difference (ΔE) is fixed, the probability of nonradiative transition is directly related to the phonon energy of the matrix (hν) at a given temperature (T) according to the expression of nonradiative decay rate (W): W(T) = W(0)[1 − exp(hν/kT)]−ΔE/hν (k is the Boltzmann constant).38 As for the Er3+ ion (Figure S4b), the 4I11/2 → 4I13/2 (NR1) transition could indirectly promote red emission 30314

DOI: 10.1021/acsami.6b12176 ACS Appl. Mater. Interfaces 2016, 8, 30312−30319

Research Article

ACS Applied Materials & Interfaces

Figure 3. UC emission spectra and ethanol-dispersed luminescent photos of Er3+/Yb3+ codoped YF3, YOF, and Y2O3 (a) and the corresponding single-particle UC luminescent images (scale bars: 5 μm) with 980 nm excitation (b).

Figure 4. Normalized UC emission spectra in green region (a), fitted plot of FIR as a function of temperature (b), and temperature-dependent sensitivity of Er3+/Yb3+ codoped YF3, YOF, and Y2O3 microcrystals (c).

through subsequent energy transfer [ET:Yb3+ (2F5/2) + Er3+ (4I13/2) → Yb3+ (2F7/2) + Er3+ (4F9/2)] or excited state absorption (ESA) process from metastable 4I13/2 state to redemitting 4F9/2 level, while the 2H11/2 → 4F9/2 (NR2) transition could directly facilitate population of 4F9/2 level from greenemitting levels 2H11/2 and 4S3/2. Thus, the emission intensity ratio of red to green is closely related with the probability of the above two nonradiative processes. With the consideration that the phonon energy value increases in the order of YF3 (358 cm−1) < YOF (482 cm−1) < Y2O3 (566 cm−1), the growing probability of NR1 and NR2 transitions results in a marked increase in the red to green ratio and tunable emission color. Irradiated by 980 nm light, thermal evolution behavior of UC emission spectra in the green region from the thermally

coupled levels 2H11/2 and 4S3/2 in Er3+/Yb3+ codoped YF3, YOF, and Y2O3 microcrystals with similar shape and particle size is illustrated in Figure 4 to investigate the effect from the phonon energy of the matrix on thermometric capability. All spectra were normalized at the 2H11/2 → 4I13/2 transition, and the excitation power is low enough (P = 120 mW) to avoid a laserinduced thermal effect. With the increase in temperature from 260 to 490 K (30 K interval), no significant changes were found in the shape and position of green emission bands for the above samples; however, the integrated emission intensity ratios of 2 H11/2 → 4I15/2 to 4S3/2 → 4I15/2 enhanced gradually (Figure 4a and Figure S5). With the consideration that TCLs 2H11/2 and 4 S3/2 of the Er3+ ion are quasithermal balanced and governed by the Boltzmann distribution law, the relative FIR could be 30315

DOI: 10.1021/acsami.6b12176 ACS Appl. Mater. Interfaces 2016, 8, 30312−30319

Research Article

ACS Applied Materials & Interfaces

electrons are protected by the outermost electron shield (5s25p6) from the impact of the crystal field. Therefore, coefficient B exerts the most influence on the thermometric sensitivity, which is described as

mathematically described using following formula (detailed formula derivation is shown in Supporting Information):17 FIR =

ω2A 2 g2 ⎛ ΔE ⎞ I2 N ℏω A ⎟ exp⎜ − = 2 2 2 = ⎝ KT ⎠ I1 N1ℏω1A1 ω1A1g1

⎛ ΔE ⎞ ⎟ = Bexp⎜ − ⎝ KT ⎠

B= (1)

where B=

× A 2 g 2 ω2 A1g1ω1



I1 and I2 denote the integrated emission intensities of 4S3/2 → 4 I15/2 and 2H11/2 → 4I15/2 transitions; B is the proportionality factor, and N, g, ω, and A represent electron occupation number, degeneracy, angular frequency, and spontaneous emission rate of corresponding transitions, respectively. Planck’s constant, absolute temperature, Boltzmann constant, and energy separation between 4S3/2 and 2H11/2 are symbolized by ℏ, T, K, and ΔE. The intuitive monologarithmic plots of ln(FIR) versus inverse absolute temperature (1/T) for Er3+/ Yb3+ codoped YF3, YOF, and Y2O3 microcrystals are presented in Figure S6, in which the experimental points could be well fitted by straight lines with slopes (ΔE/K) of about 991.9, 1034.3, and 1127.3, respectively. The values of the energy gap ΔE between 4S3/2 and 2H11/2 in above samples were calculated to be approximately 685, 714, and 778 cm−1. As seen from the dependence of FIR on temperature (Figure 4b), the FIR values of three samples rise dramatically with the increase in T, and the corresponding coefficient B and regression parameters R obtained from the best fitting curves of experimental results are R1 = 0.9997 and B1 = 4.883 ± 0.0805 (YF3), R2 = 0.9998 and B2 = 11.331 ± 0.1712 (YOF), and R3 = 0.9991 and B3 = 18.080 ± 0.6998 (Y2O3). As a vital parameter of quantitatively evaluating optical thermometric ability, absolute sensitivity (Sa) denotes the theoretical variation rate of FIR along with temperature, which can be expressed as20 S=

⎛ ΔE ⎞ ⎡ ΔE ⎤ d(FIR) ΔE = FIR⎜ 2 ⎟ = B 2 exp⎢ − ⎣ KT ⎥⎦ ⎝ ⎠ dT KT KT

C2 ω24 ω24 × = C1 ω14 ω14 ∑λ =−2,4,6 Ωλ⟨4 I15/2 U (λ) 2 H11/2⟩2 Ω6⟨4 I15/2 U 6 4S3/2 ⟩2 0.7158 × Ω 2 + 04138 × Ω4 0.0927 + 0.2225 × Ω6 0.2225

(3)

ω24/ω14

where is approximately evaluated as the unit one due to the smaller orders of magnitude of the energy separation between the 4S3/2 and 2H11/2 levels (i.e., 102 cm−1) and C represents the line strength of corresponding transitions. Ωλ (λ = 2, 4, or 6) denotes the Judd−Ofelt (J−O) intensity parameters for electric dipole transitions, and the reduced matrix elements ⟨∥ ∥⟩ are assumed to be independent of the studied host.20 As known to all, the 4S3/2 → 4I15/2 transition is insensitive to the environment, while the 2H11/2 → 4I15/2 radiation process belongs to the hypersensitive transition. The Ω2 parameter is closely associated with hypersensitive transition and is more sensitive to the crystal field environments than other J−O parameters which rely on the short-range effects, yet Ω4 and Ω6 are long-range parameters linked with the bulk natures of the matrixes, such as the basicity or the rigidity of the host, revealing that Ω4 and Ω6 hardly exert influence on sensitivity, and the B value is mainly decided by Ω2 in the present system. Moreover, the Ω2 parameter of the Er3+ ion strongly depends on the covalent chemical bonding between lanthanide ions and ligands, according to previous literature.39 Namely, absolute sensitivity of a sensor could be evaluated by analyzing the bond covalency of the materials. According to dielectric theory of the chemical bond for complex crystals with known crystal structure and refractive index (n), the average bond covalency (fc) of Y−A (A = F or O) in YF3, YOF, and Y2O3 hosts were calculated and are listed along with other related parameters in Table 1, in which the values of bond

(2)

The variation of the calculated sensor sensitivities at different temperatures for three samples are plotted in Figure 4c, in which all sensitivities kept rising with the increase in temperature in our experimental range. Impressively, the sensitivity values of Er3+/Yb3+ doped YF3, YOF, and Y2O3 samples gradually increased and achieved maximums of about 0.0027, 0.0060, and 0.0085 K−1 at the 490 K, respectively. Because the morphology and particle size of three products are similar and the calcination-induced variation in UC emission intensity hardly affects sensitivity according to previous work,22 diverse phonon energies of hosts may be responsible for the difference in thermometric ability. According to the theoretical eq 2, thermal sensing sensitivity is directly determined by ΔE and B; thus, the fitting curves of absolute sensitivity (Sa) versus temperature (T) with different values of ΔE and B are displayed in Figure S7 to better comprehend their connection. It can be deduced that the optimal temperature detection range is closely related with energy gap ΔE, while Sa is mainly affected by proportionality factor B. For present system based on the Er3+ ion, the ΔE value between 4S3/2 and 2H11/2 in different hosts could be recognized as a constant within the error range because 4f

Table 1. Calculated Bond Covalency of Er3+/Yb3+ Codoped YF3, YOF, and Y2O3 Microcrystalsa host material

phonon energy (cm−1)

bond

fc ̅

B

S (10−4 K−1)

YF3 YOF Y2O3

358 482 566

Y−F Y−F, Y−O Y−O

0.083 0.131 0.140

4.883 11.331 18.08

27 60 85

a fc:̅ average bond covalency; B: proportionality factor; S: thermometric sensitivity.

covalency gradually increase from fluoride to oxide, which follows the same variation trend with B and absolute sensitivity.40 The above characters suggest a strong dependence of sensitivity on the bond covalency of host materials. Beyond that, the phonon energy is also proportional to the bond covalency, which kept rising from YF3 (358 cm−1) through YOF (482 cm−1) to Y2O3 (566 cm−1). In an attempt to determine the most basic and intrinsic factors associated with thermometric ability, first-principles calculations were carried out to obtain a deeper insight in the inner relationship between phonon energy and bond covalency. 30316

DOI: 10.1021/acsami.6b12176 ACS Appl. Mater. Interfaces 2016, 8, 30312−30319

Research Article

ACS Applied Materials & Interfaces

electronegativity, which provides a convenient strategy to achieve highly sensitive thermometry based on Er3+ by selecting hosts with an appropriate phonon energy.

Generally, the lattice vibration in one unit cell is similar to the others in the same crystal and differs from different unit cells only in phase factor introduced by translation symmetry; thus, it needs to analyze only the vibration in one unit cell in principle. With the consideration that a phonon is still a lattice vibration and YF3, YOF, and Y2O3 were mainly built by Y−F and Y−O, total electronic energy and frontier molecular orbital of two typical chemical bonds Y−F and Y−O were taken as examples in the present system. When the bond length was elongated, the fitting parabolas based on the total electronic energy achieved the lowest energy values (equilibrium point) at bond lengths of 1.9 and 1.8 Å for Y−F and Y−O, respectively (Figure 5). Namely, the most stable states could be realized

4. CONCLUSIONS In summary, the inner qualitative relation between phonon energy and sensing capability of FIR-based thermal sensors in Er3+/Yb3+ codoped UC microcrystals was investigated in detail. Regular truncated octahedral Er3+/Yb3+ codoped YF3 microcrystals were prepared by the hydrothermal method and then transferred to YOF and Y2O 3 with nearly unchanged morphology and size through calculation treatment at 850 and 1100 °C for 2 h in air. The phonon energies of YF3, YOF, and Y2O3 were measured to be 358, 482, and 566 cm−1, respectively. After the samples were irradiated at 980 nm, the output visible color from UC emission spectra and visualized single-particle luminescent images of the three samples were markedly tuned from green to yellow, attributed to the increased possibility of a phonon energy-dependent nonradiative process. Thermal responsive FIR from TCLs (2H11/2/4S3/2) was monitored to evaluate thermal sensing behaviors, and the optimal sensitivity values of the three samples gradually increased and reached maxima of about 0.0027, 0.0060, and 0.0085 K−1 at 490 K, which is closely associated with bond covalency. According to the firstprinciples calculations results when Y−F and Y−O bonds are taken as examples, the electronegativity is bound to lattice vibration energy, which is directly proportional to the S value. The above theoretical and experimental results not only enable a deeper understanding of the relation between intrinsic structure and thermometric ability but also provide a significant step toward highly sensitive thermometry.

Figure 5. Plot of total electronic energy versus bond length of Y−F and Y−O; the inset shows the schematic frontier molecular orbital illustration of corresponding bonds.



ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at 10.1021/acsami.6b12176. Theoretical background of the FIR technique, SEM image of the YF3:Er3+/Yb3 sample, HR-TEM images and EDX spectra of three samples, theoretical Raman scattering spectrum of blank YF3, CIE chromaticity coordinates and simplified schematic energy diagram, temperature-dependent UC emission spectra and monolog plot of FIR versus 1/T for three samples, fitting curves of S versus T with changeable ΔE/K value (B = 10) or B value (ΔE/K = 800 cm−1), calculated vibration frequency value of Y−F and Y−O bonds, and atomic charge of atoms in two bands at an optimized bond length (PDF)

when the bond lengths of Y−F and Y−O were 1.9 and 1.8 Å. Because the chemical bond between two atoms acts as a spring with vibrators on both ends, the natural vibration frequency (ω) is directly proportional to the second derivative value of the potential energy surface curve at its balance point, which is closely linked with lattice vibration energy. As illustrated in Table S1, the calculated vibration frequency of Y−O (40.211) is larger than that of Y−F (29.094), endowing Y−O with a lattice vibration energy larger than that of Y−F. According to the Mulliken population analysis, a larger overlap atomic charge population between atoms represents stronger covalent bonding.41 Compared with that of the Y−O bond, Y−F exhibits a lower overlap atomic charge value at its optimized bond length (Table S2), which indicates an ionicity of Y−F stronger than that of the Y−O bond. As observed from the visualized frontier molecular orbital illustrations (inset of Figure 5), when the bonds are located in the highest occupied molecular orbital (HOMO) and adjacent lowest unoccupied molecular orbital (LUMO-1), the possibility of electrons populating at the middle bond between two atoms of Y−O is higher than that of the Y−F bond, further confirming that Y−O has a bond covalency stronger than that of Y−F. For the present system, the interaction force of the ionic bond is nondirectional, while the overlap of electron clouds in the covalent bond follows a specific direction, resulting in a vibration energy of the covalent bond larger than that of the ionic bond. In light of above theoretical results, as an intrinsic property of host materials, phonon energy is closely bound to



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (Grants 11274251 and 51672215), Research Fund for the Doctoral Program of Higher Education of China (RFDP) (Grant 20136101110017), Natural Science Foundation of Shaanxi Province (Grant 2014JM1004), 30317

DOI: 10.1021/acsami.6b12176 ACS Appl. Mater. Interfaces 2016, 8, 30312−30319

Research Article

ACS Applied Materials & Interfaces

(18) Dong, B.; Cao, B. S.; He, Y. Y.; Liu, Z.; Li, Z. P.; Feng, Z. Q. Temperature Sensing and in Vivo Imaging by Molybdenum Sensitized Visible Upconversion Luminescence of Rare-Earth Oxides. Adv. Mater. 2012, 24, 1987−1993. (19) Debasu, M. L.; Brites, C. D. S.; Balabhadra, S.; Oliveira, H.; Rocha, J.; Carlos, L. D. Nanoplatforms for Plasmon-Induced Heating and Thermometry. ChemNanoMat 2016, 2, 520−527. (20) León-Luis, S. F.; Rodríguez-Mendoza, U. R.; Haro-González, P.; Martín, I. R.; Lavín, V. Role of the Host Matrix on The Thermal Sensitivity of Er3+ Luminescence in Optical Temperature Sensors. Sens. Actuators, B 2012, 174, 176−186. (21) Dong, B.; Hua, R. N.; Cao, B. S.; Li, Z. P.; He, Y. Y.; Zhang, Z. Y.; Wolfbeis, O. S. Size Dependence of the Upconverted Luminescence of Nayf4: Er, Yb Microspheres for Use in Ratiometric Thermometry. Phys. Chem. Chem. Phys. 2014, 16, 20009−20012. (22) He, D.; Guo, C. F.; Zhou, S. S.; Zhang, L. L.; Yang, Z.; Duan, C. K.; Yin, M. Synthesis and Thermometric Properties of Shuttle-like Er3+/Yb3+co-doped NaLa(MoO4)2 Microstructures. CrystEngComm 2015, 17, 7745−7753. (23) Marciniak, L.; Waszniewska, K.; Bednarkiewicz, A.; Hreniak, D.; Strek, W. Sensitivity of a Nanocrystalline Luminescent Thermometer in High and Low Excitation Density Regimes. J. Phys. Chem. C 2016, 120, 8877−8882. (24) Vetrone, F.; Naccache, R.; Zamarron, A.; de la Fuente, A. J.; Sanz-Rodríguez, F.; Maestro, L. M.; Rodriguez, E. M.; Jaque, D.; Solé, J. G.; Capobianco, J. A. Temperature Sensing Using Fluorescent Nanothermometers. ACS Nano 2010, 4, 3254−3258. (25) Aigouy, L.; Saïdi, E.; Lalouat, L.; Labéguerie-Egéa, J.; Mortier, M.; Löw, P.; Bergaud, C. AC Thermal Imaging of a Microwire with a Fluorescent Nanocrystal: Influence of the Near Field on the Thermal Contrast. J. Appl. Phys. 2009, 106, 074301−074310. (26) Chen, D. Q.; Wan, Z. Y.; Zhou, Y.; Zhou, X. Z.; Yu, Y. L.; Zhong, J. S.; Ding, M. Y.; Ji, Z. G. Dual-Phase Glass Ceramic: Structure, Dual-Modal Luminescence, and Temperature Sensing Behaviors. ACS Appl. Mater. Interfaces 2015, 7, 19484−19493. (27) Singh, A. K.; Singh, S. K.; Gupta, B. K.; Prakash, R.; Rai, S. B. Probing a Highly Efficient Dual Mode: Down-Upconversion Luminescence and Temperature Sensing Performance of Rare-Earth Oxide Phosphors. Dalton Trans. 2013, 42, 1065−1072. (28) Mahata, M. K.; Kumar, K.; Rai, V. K. Er3+-Yb3+ doped Vanadate Nanocrystals: A Highly Sensitive Thermographic Phosphor and its Optical Nanoheater Behavior. Sens. Actuators, B 2015, 209, 775−780. (29) Pandey, A.; Rai, V. K.; Kumar, V.; Kumar, V.; Swart, H. C. Upconversion Based Temperature Sensing Ability of Er3+-Yb3+ codoped SrWO4: An Optical Heating Phosphor. Sens. Actuators, B 2015, 209, 352−358. (30) Li, T.; Guo, C. F.; Zhou, S. S.; Duan, C. K.; Yin, M. Highly Sensitive Optical Thermometry of Yb3+-Er3+ codoped AgLa(MoO4)2 Green Upconversion Phosphor. J. Am. Ceram. Soc. 2015, 98, 2812− 2816. (31) Levine, B. F.; Bethea, C. G. Charge Transfer Complexes and Hyperpolarizabilities. J. Chem. Phys. 1977, 66, 1070−1074. (32) Van Vechten, J. A. Quantum Dielectric Theory of Electronegativity in Covalent Systems. I. Electronic Dielectric Constant. Phys. Rev. 1969, 182, 891−905. (33) Wilmarth, W. R.; Begun, G. M.; Nave, S. E.; Peterson, J. R. The Raman Spectra of Polycrystalline Orthorhombic YF3, SmF3, HoF3, YbF3, and Single Crystal TbF3. J. Chem. Phys. 1988, 89, 711−715. (34) Chai, G. Q.; Dong, G. P.; Qiu, J. R.; Zhang, Q. Y.; Yang, Z. M. Phase Transformation and Intense 2.7 mm Emission from Er3+ doped YF3/YOF Submicron-Crystals. Sci. Rep. 2013, 3, 1598−1609. (35) Li, J.; Zhang, J. H.; Hao, Z. D.; Zhang, X.; Zhao, J. H.; Luo, Y. S. Intense Upconversion Luminescence and Origin Study in Tm3+Yb3+ codoped Calcium Scandate. Appl. Phys. Lett. 2012, 101, 121905− 121909. (36) Patra, A.; Ghosh, P.; Chowdhury, P. S.; Alencar, M. A. R. C.; Lozano, W.; Rakov, N.; Maciel, G. S. Red to Blue Tunable Upconversion in Tm3+-doped ZrO2 Nanocrystals. J. Phys. Chem. B 2005, 109, 10142−10146.

Foundation of Shaanxi Province Educational Department (Grants 15JS101 and 15JK1712), and Northwest University (NWU) Graduate Innovation and Creativity Funds (Grant YZZ15030).



REFERENCES

(1) Jaque, D.; Vetrone, F. Luminescence Nanothermometry. Nanoscale 2012, 4, 4301−4326. (2) Wang, X. D.; Wolfbeis, O. S.; Meier, R. J. Luminescent Probes and Sensors for Temperature. Chem. Soc. Rev. 2013, 42, 7834−7869. (3) McLaurin, E. J.; Bradshaw, L. R.; Gamelin, D. R. Dual-emitting Nanoscale Temperature Sensors. Chem. Mater. 2013, 25, 1283−1292. (4) Bai, T. T.; Gu, N. Micro/Nanoscale Thermometry for Cellular Thermal Sensing. Small 2016, 12, 4590−4610. (5) Rodríguez-Sevilla, P.; Zhang, Y. H.; Haro-González, P.; SanzRodríguez, F.; Jaque, F.; Solé, J. G.; Liu, X. G.; Jaque, D. Thermal Scanning at the Cellular Level by an Optically Trapped Upconverting Fluorescent Particle. Adv. Mater. 2016, 28, 2421−2426. (6) Debasu, M. L.; Ananias, D.; Pastoriza-Santos, I.; Liz-Marzán, L. M.; Rocha, J.; Carlos, L. D. All-In-One Optical Heater-Thermometer Nanoplatform Operative from 300 to 2000 K Based on Er3+ Emission and Blackbody Radiation. Adv. Mater. 2013, 25, 4868−4874. (7) Pereira, A. F.; Kumar, K. U.; Silva, W. F.; Santos, W. Q.; Jaque, D.; Jacinto, C. Yb3+/Tm3+ co-doped NaNbO3 Nanocrystals as Threephoton-excited Luminescent Nanothermometers. Sens. Actuators, B 2015, 213, 65−71. (8) Ximendes, E. C.; Santos, W. Q.; Rocha, U.; Kagola, U. K.; SanzRodríguez, F.; Fernández, N.; Gouveia-Neto, A. S.; Bravo, D.; Domingo, A. M.; del Rosal, B.; Brites, C. D. S.; Carlos, L. D.; Jaque, D.; Jacinto, C. Unveiling in Vivo Subcutaneous Thermal Dynamics by Infrared Luminescent Nanothermometers. Nano Lett. 2016, 16, 1695− 1703. (9) Savchuk, O. A.; Carvajal, J. J.; Pujol, M. C.; Barrera, E. W.; Massons, J.; Aguiló, M.; Díaz, F. Ho,Yb: KLu(WO4)2 Nanoparticles: A Versatile Material for Multiple Thermal Sensing Purposes by Luminescent Thermometry. J. Phys. Chem. C 2015, 119, 18546− 18558. (10) Savchuk, O. A.; Carvajal, J. J.; Cascales, C.; Massons, J.; Aguiló, M.; Díaz, F. Thermochromic Upconversion Nanoparticles for Visual Temperature Sensors with High Thermal, Spatial and Temporal Resolution. J. Mater. Chem. C 2016, 4, 6602−6613. (11) Zhu, X. J.; Feng, W.; Chang, J.; Tan, Y. W.; Li, J. C.; Chen, M.; Sun, Y.; Li, F. Y. Temperature-feedback Upconversion Nanocomposite for Accurate Photothermal Therapy at Facile Temperature. Nat. Commun. 2016, 7, 10437−10446. (12) Benayas, A.; del Rosal, B.; Pérez-Delgado, A.; Santacruz-Gómez, K.; Jaque, D.; Hirata, G. A.; Vetrone, F. Nd:YAG Near-Infrared Luminescent Nanothermometers. Adv. Opt. Mater. 2015, 3, 687−694. (13) Cerón, E. N.; Ortgies, D. H.; del Rosal, B.; Ren, F. Q.; Benayas, A.; Vetrone, F.; Ma, D. L.; Sanz-Rodríguez, F.; Solé, J. G.; Jaque, D.; Rodríguez, E. M. Hybrid Nanostructures for High-Sensitivity Luminescence Nanothermometry in the Second Biological Window. Adv. Mater. 2015, 27, 4781−4787. (14) Zhou, J.; Liu, Q.; Feng, W.; Sun, Y. Upconversion Luminescent Materials: Advances and Applications. Chem. Rev. 2015, 115, 395−465. (15) Wang, Z.; Zhang, P.; Yuan, Q. H.; Xu, X.; Lei, P. P.; Liu, X. L.; Su, Y.; Dong, L. L.; Feng, J.; Zhang, H. J. Nd3+-sensitized NaLuF4 luminescent nanoparticles for multimodal imaging and temperature sensing under 808 nm excitation. Nanoscale 2015, 7, 17861−17870. (16) Xu, X.; Wang, Z.; Lei, P. P.; Yu, Y. N.; Yao, S.; Song, S. Y.; Liu, X. L.; Su, Y.; Dong, L. L.; Feng, J.; Zhang, H. J. α-NaYb(Mn)F4:Er3+/ Tm3+@NaYF4 UCNPs as “Band-Shape” Luminescent Nanothermometers over a Wide Temperature Range. ACS Appl. Mater. Interfaces 2015, 7, 20813−20819. (17) Suo, H.; Guo, C. F.; Li, T. Broad-Scope Thermometry Based on Dual-Color Modulation Up-Conversion Phosphor Ba5Gd8Zn4O21:Yb3+/Er3+. J. Phys. Chem. C 2016, 120, 2914−2924. 30318

DOI: 10.1021/acsami.6b12176 ACS Appl. Mater. Interfaces 2016, 8, 30312−30319

Research Article

ACS Applied Materials & Interfaces (37) Savchuk, O. A.; Carvajal, J. J.; Cascales, C.; Aguiló, M.; Díaz, F. Benefits of Silica Core−Shell Structures on the Temperature Sensing Properties of Er, Yb: GdVO4 Up-Conversion Nanoparticles. ACS Appl. Mater. Interfaces 2016, 8, 7266−7273. (38) Chen, G. Y.; Liu, H. C.; Liang, H. J.; Somesfalean, G.; Zhang, Z. G. Upconversion Emission Enhancement in Yb3+/Er3+-codoped Y2O3 Nanocrystals by Tridoping with Li+ Ions. J. Phys. Chem. C 2008, 112, 12030−12036. (39) León-Luis, S. F.; Rodríguez-Mendoza, U. R.; Martín, I. R.; Lalla, E.; Lavín, V. Effects of Er3+ Concentration on Thermal Sensitivity in Optical Temperature Fluorotellurite Glass Sensors. Sens. Actuators, B 2013, 176, 1167−1175. (40) Wu, Z. J.; Zhang, S. Y. Semiempirical Method for the Evaluation of Bond Covalency in Complex Crystals. J. Phys. Chem. A 1999, 103, 4270−4274. (41) Mulliken, R. S. Electronic Population Analysis on LCAO-MO Molecular Wave Functions. J. Chem. Phys. 1955, 23, 1833−1840.

30319

DOI: 10.1021/acsami.6b12176 ACS Appl. Mater. Interfaces 2016, 8, 30312−30319