Yb3+ doped LiYF4 single crystals

Yb3+ doped LiYF4 single crystals

Author’s Accepted Manuscript Examination of Judd-Ofelt calculation and temperature self-reading for Tm3+ and Tm3+/Yb3+ doped LiYF4 single crystals Guo...
Download PDF
1MB Sizes 0 Downloads 18 Views
Report

Author’s Accepted Manuscript Examination of Judd-Ofelt calculation and temperature self-reading for Tm3+ and Tm3+/Yb3+ doped LiYF4 single crystals Guozhu Sui, Baojiu Chen, Jinsu Zhang, Xiangping Li, Sai Xu, Jiashi Sun, Yanqiu Zhang, Lili Tong, Xixian Luo, Haiping Xia

PII: DOI: Reference:

www.elsevier.com/locate/jlumin

S0022-2313(17)31244-9 https://doi.org/10.1016/j.jlumin.2018.02.016 LUMIN15363

To appear in: Journal of Luminescence Received date: 20 July 2017 Revised date: 28 December 2017 Accepted date: 5 February 2018 Cite this article as: Guozhu Sui, Baojiu Chen, Jinsu Zhang, Xiangping Li, Sai Xu, Jiashi Sun, Yanqiu Zhang, Lili Tong, Xixian Luo and Haiping Xia, Examination of Judd-Ofelt calculation and temperature self-reading for Tm3+ and Tm3+/Yb3+ doped LiYF4 single crystals, Journal of Luminescence, https://doi.org/10.1016/j.jlumin.2018.02.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Examination of Judd-Ofelt calculation and temperature self-reading for Tm3+ and Tm3+/Yb3+ doped LiYF4 single crystals Guozhu Suia, Baojiu Chena, * Jinsu Zhanga, Xiangping Lia, Sai Xua, Jiashi Suna, Yanqiu Zhanga, Lili Tonga, Xixian Luoa, Haiping Xiab, * a

Department of Physics, Dalian Maritime University, Dalian, 116026, PR China

b

Kay Laboratory of Photo-electronic Materials, Ningbo University, Ningbo 315211, PR China

ABSTRACT To validate the reliability of Judd-Ofelt results and the influence of involving absorption transition number, the Judd-Ofelt calculations, in which various transitions were adopted, were carried out for Tm3+ doped LiYF4 single crystal. It was found that introducing more transitions into the calculation procedure might get more reliable results. In order to clarify the feasibility of temperature self-reading in Tm3+/Yb3+ doped LiYF4 single crystal during laser operation, the temperature sensing properties of the single crystal were studied. It was found that the fluorescence intensity ratio of 3

F2+3F3→3H6 to 3H4→3H6 can be used for achieving better temperature detection, and

the temperature sensitivity was found much better than that in other materials. Keywords: Judd-Ofelt theory; Temperature sensing; Up-conversion luminescence; LiYF4:Tm3+/Yb3+ single crystal

1. Introduction Nowadays, lasers are widely applied and play important roles in many fields such as medicine, aerospace and even our daily lives [1-3]. The working medium for laser operation is requisite. The media can be divided into solid, liquid and gaseous states. Amongst these media the solid state working media are more preferable than the others due to their easy mechanical treatment, high thermal conductivity and physical/chemical stability. The solid state laser media include single crystals, polycrystals and ceramics. The single crystals including optical nonlinear crystals and activators doped single crystals have attracted much attention, and some successful stories, for example, Y3Al5O12:Nd3+ (YAG:Nd3+), KTiOPO4 (KTP) and Al2O3:Ti3+ are well known [4-6]. Amongst all these laser crystals, we should give much attention to the fluoride based-crystals such as YF3, LiYF4 and NaYF4 because of that they exhibit extremely low phonon energy, which results in low nonradiative transition rate of the laser operation level and high up-conversion efficiency in these crystals doped with rare earth ions [7-11]. Moreover, the fluoride single crystals have high transmissivity in wide wavelength region from vacuum violet to mid infrared. Therefore, the short wavelength lasers operating in ultraviolet and blue region and long wavelength lasers operating in mid infrared are expected to be achieved in fluoride crystals doped with rare earth ions [12]. As an excellent representative of fluoride single crystal media, LiYF4 has been widely investigated since it is relatively easier grown in comparison with others such

as NaYF4 [13]. Down-conversion, up-conversion, scintillation behaviors in LiYF4 doped with rare earth ions were studied for exploring their applications in short wavelength lasers, mercury-free fluorescent lamps, mid infrared lasers, radiation detection and solar cells [14-18]. Though both the spectroscopic properties and laser operation of Tm3+ in LiYF4 are widely reported [19-22], it still leaves us some questions, for example, quality for the results of optical transition calculation and the solution for the heat deposition in the crystal during laser operation. In this work, we examined the Judd-Ofelt calculation reliability for Tm3+ in LiYF4 single crystal by taking various transition numbers into the numerical fitting. Moreover, to detect the medium temperature during the laser operation, the temperature sensing properties were studied. It was found that main/intense transitions played dominant role in the Judd-Ofelt calculation, and the non-contact temperature sensing might be expedient for self-monitoring the crystal temperature of the crystal during the laser operation.

2. Experimental details 2.1 Single crystal growth Tm3+ doped and Tm3+/Yb3+ co-doped LiYF4 single crystals with designed composition 51.5LiF-(48-x)YF3-0.5TmF3-xYbF3 (x=0 or 10, in molar) were grown by a Bridgman method. Here we just mention the main procedure for the sample growth, and the detailed procedure can be found elsewhere in Ref. [23]. First, the highly pure starting materials of LiF, YF3, TmF3 and YbF3 powders (all of them are 99.999% in weight) were fluoridized by using HF gas at at 750-800 ℃ for 8-10 h to

ensure the complete removal of residual moisture and oxides. Then the starting materials were weight according to the designed composition and mixed together. After that, the mixture and a LiYF4 crystal seed were sealed in a platinum crucible which was loaded into the furnace for crystal growth. The temperature gradient along the vertical axis of the crucible was set to be around 50-60 ℃/cm. Later on, the crucible was lowering at a rate of 0.5-0.6 mm/h. Finally, the furnace was cooled to room temperature with a rate of 30-40 ℃/h, following that the single sample was obtained. The samples were cut into small pieces and carefully polished for further measurements. 2.2 Structural and spectroscopic characterization The sample density is needed in the Judd-Ofelt calculation. To measure the sample density, the distilled water was used as immersion liquid, and the density was confirmed based on the Archimedes’ principle to be 3.99 g/cm3. Usually, the actual/true concentration of the doping ion is dependent on the position in the single rod, and the sample used for study is cut off from the rod, therefore, the real concentration of Tm3+ should be experimentally measured. In this sense, the Tm3+ concentration was obtained by using an Optima-3000 (Perkin Elmer Inc.) ICP-AES. The crystal structure of the sample was checked by X-ray diffraction (XRD) using a XD-98X diffractometer (XD-3, Beijing) with a Cu tube and Kα radiation at 0.15406 nm.

The

absorption

spectrum

was

measured

by

Shimadzu

UV-3600

Spectrophotometer at room temperature. The sample temperature was controlled by using a temperature controller DMU-TC 450 assembled in our lab. The emission

spectra measured upon 980 nm at temperatures ranging from 313 K to 713 K by using a Hitachi F-4600 Spectrophotometer.

3. Results and discussion 3.1 Crystal structure The XRD pattern for the sample single doped with Tm3+ was measured by using the powder derived from well grinding the bulk crystals. The pattern in the 2 ranging from 10 to 60 º is shown in Fig. 1. In careful comparison with the standard JCPDS card No. 85-0806, it is found that all the diffraction peaks observed can be well indexed to those for the tetragonal phase with body-centered symmetry. This fact means that the obtained single crystal is the expected target LiYF4, and the Tm3+ doping in the present concentration extent does not obviously affect the crystal structure. <
> 3.2 Optical transition of Tm3+ in LiYF4 single crystal First of all, in this section, we briefly introduce the Judd-Ofelt theory [24, 25] and its application procedure for the studied crystal, based on which we will discuss about the influence of the number of involving optical absorption transitions in the Judd-Ofelt calculation on the result reliability. For an absorption transition, the experimental oscillator strength can be calculated from the absorption cross section by using the following equation [26],

mc2 f ex  2   (v)dv e

(1)

where m is the mass of electron; e represents the charge of electron; c is the light

velocity in the vacuum; v is the wavenumber,  (v) represents the absorption cross section at wavenumber v and can be calculated by the equation,

 (v ) 

ln[ I 0 (v) / I (v)] Nt

(2)

where I 0 (v) and I (v) , respectively, presents the intensities of incident and outgoing light, t stands the thickness of the sample, and its unit is cm ; N represents the number of RE3+ ions per cm 3 in the sample. Fig. 2 shows the ground state absorption cross section spectrum of LiYF4:Tm3+ single crystal. It can be seen that there are 6 absorption bands centered at around 5945, 8265, 12642, 14598, 21186, and 28011 cm-1 corresponding to the transitions from ground state 3H6 to the excited states 3F4, 3H5, 3H4, 3F2, 3, 1G4 and 1D2, respectively. According to Eqs. (1), (2) and the spectral data in Fig. 2, the experimental oscillator strengths of the observed transitions were calculated and are listed in Table 1. <
> <> The theoretical oscillator strength f thed for an electric dipole transition within the 4fN configuration can be written as,

f thed 



8 2 mcv (n(v) 2  2) 2   J U  J   3h(2 J  1) 9n   2, 4, 6



2

(3)

md

The theoretical oscillator strength f th for a magnetic dipole transition can be expressed as

fthmd 

hvn(v) J L  2S J  6mc(2 J  1)

2

(4)

In above Eqs. (3) and (4), m , c and v are the same as in Eqs. (1) and (2); h is

Planck’s constant; J is the angular moment quantum number of the initial state for a transition. In the present case of Tm3+, the absorption transition 3H6→3H5 is a magnetic dipole transition;   (   2，4，6 ) are the Judd-Ofelt parameters; J U   J 

and

J L  2S J 

are the reduced matrix elements for the

electric and magnetic dipole transitions; n is refractive index of the host. According to Eqs. (1), (3) and (4), the Judd-Ofelt parameters can be calculated via a least-square method by using the data in Table 1, and the obtained the Judd-Ofelt parameters are listed in Table 2 as the 2nd raw counting from right side. It should be mentioned that at least 3 transitions are needed for obtaining Judd-Ofelt parameters, and if the transition number involving into the Judd-Ofelt calculation is more, then the calculation results are more reliable. However, the observable optical absorption transitions are usually finite, thus there a question is placed on the desk that how many transitions at least should be adapted in order to achieve relatively reliable results; meanwhile, another question is whether the different transitions adapted in the calculation affect the results. To answer these questions, we choose arbitrary 3 transitions from the total 6 observed transitions in Table 1 as a set for Judd-Ofelt calculation, thus totaled 20 sets of Judd-Ofelt parameters are obtained and listed in Table as the 2nd main raw. In an analogical way, we also choose arbitrary 4 or 5 transitions from the 6 transitions for the Judd-Ofelt calculations, and the results are filled into the 3rd and 4th main raw in Table 2. For the possible cases, the calculation errors  for the oscillator strengths were estimated by using the following equation [26],



N

 i

f

 f thi N 3

i ex



2

(5)

where N is the transition number involved in the Judd-Ofelt calculation, f exi and f thi are, respectively, the experimental and theoretical oscillator strengths of ith transition. The obtained errors are listed as the last column in Table 2. It should be stated that the specific transitions used for calculations are also marked in Table 2 as the 2nd column, and the experimental and theoretical oscillators are listed as 3rd and 4th column. <
>

By inspecting Table 2, it can be found that the results of Judd-Ofelt intensity parameters can be strongly influenced by the transition number involved in the calculation, and following conclusion can be deduced. (I) It is known that the Judd-Ofelt parameters must be positive values. In some calculations, minus value for one or more parameters amongst the three Judd-Ofelt parameters was derived. 5 sets amongst 20 sets (5/20) exhibited minus Judd-Ofelt parameters when 3 transitions were adopted in Judd-Ofelt calculation; 1 sets amongst 15 sets (1/15) displayed minus Judd-Ofelt parameters when 4 transitions were used in Judd-Ofelt calculation; 0 set amongst 6 sets (0/6) showed minus Judd-Ofelt parameters when 5 transitions were used in Judd-Ofelt calculation. The minus values indicated that the calculations were failed. This fact implies that the probability for achieving failed calculation results decreases when the transition number involved into the Judd-Ofelt calculation increases, thus indicating that more transitions used in the Judd-Ofelt calculation is beneficial for getting the correct/reliable results. (II) The Judd-Ofelt calculation for the datum set including 3 transitions were the

processes of solving three-variable linear equation sets, and the calculation error is zero, but it can be seen that the Judd-Ofelt parameters were very large, and sometimes they even reached several hundreds times 10-20 cm2 which is really too far from the f-f transition nature. Moreover, by inspecting the calculation errors in the last column of Table 2, it is not hard to find that the calculation error tend to decrease when the involved transition number in the Judd-Ofelt calculation increases. This again told that more transition used in the Judd-Ofelt calculation is helpful for achieving more accurate results. (III) It can also be found that when 5 or 6 transitions are used in the calculations, the closer results for the Judd-Ofelt parameters are obtained, which indicates that the Judd-Ofelt parameters results confirmed by involving more transitions can improve the reliability of the Judd-Ofelt parameters. Moreover, it is well known that the absorption transitions 3H6→3H4 and 3H6→3F4 are hypersensitive transitions which are strongly dependent on the crystal field environment [27]. However, calculation results were not substantially improved when these transitions were excluded from the calculation process (see the blue raw in Table 2), though the calculation error is less since the 2 is very different from the values derived by using more transitions in the calculations. Finally, what we can conclude is that different results can be found from various transitions involving the Judd-Ofelt calculation, and for achieving more reliable results more than 4 transitions at least should be adapted in the calculation. In the present case, we suggest that the Judd-Ofelt parameters determined by using 6

absorption transitions are more reliable. In order to further check the above obtained Judd-Ofelt parameters, here we compare the radiative transition rate derived by using the Judd-Ofelt parameters with the one derived from a route independent from the Judd-Ofelt parameters for a specific transition. We choose the absorption from ground state to the first excited state, namely, 3H6→3F4. For this transition the radiative transition rate A3F4 can be represented as

A3 F 4



64 4e 2 3n n 2  2  27h2 J  1



2

  

J U  J 

2

(6)

2 , 4 , 6

where all the symbols have the same physical meanings as in Eq. (3). By taking all the already-known values for the physical quantities into the above Eq. (6), A3F4 can be confirmed to be 260.5 s-1. Taking the Einstein relation into account, the radiative transition rate A3F4 for the first excited state 3F4 can be represented by using the integrated absorption cross section



3F 4

(v)dv as follows [28],

8(2 J '1)v 2 n 2c A3F 4    (v)dv 2J  1

(7)

where J and J are the quantum numbers of angular moment for the initial state 3H6 and final state 3F4, the other symbols have the same meaning as in Eq. (3). In Eq. (7) the integral runs over full the absorption band 3H63F4 in Fig. 2. From Eq. (7) the radiative transition rate can be obtained to be 246.4 s-1 which is close to the result derived from Eq. (6), thus implying the Judd-Ofelt parameters we obtained by using 6 absorption is reliable.

It should emphasized that in Judd-Ofelt theory the theoretical oscillator strength for an electric-dipole transition is parameterized and presented by Eq. (3), which is a phenomenological presentation but not the first principle presentation. Therefore the Judd-Ofelt theory is not the final theory [29, 30], natheless it still can explain the optical f-f transition properties in a certain degree. In this sense, the error in the Judd-Ofelt treatment can come from the theory itself, thereby the obtained errors can be different for different sets of absorption transition data though the same number of transitions is involved into the Judd-Ofelt calculations as seen in Table 2. Unfortunately, here is only one set of data including 6 absorption transitions, and the error obtained in the calculation involving these 6 transitions seems larger. If we could observe 7 absorption transitions then some errors from the other Judd-Ofelt calculations involving other different 6 absorption transitions would be much smaller than the present one. This is the trend as we find in Table 2. 3.3 Temperature sensing performance As a laser medium, the LiYF4:Tm3+/Yb3+ usually works with the commercial high power 980 nm semiconductor laser as pump source [7]. It should be mentioned that the LiYF4:Tm3+/Yb3+ single crystal used for laser operation is not ideal/perfect crystal without defects, that is to say, the defects or unintended doping centers are still existent in the crystal. These defects and unintended doping centers usually result in energy deposition of pump energy in the laser crystal. In addition, the input energy of pump source can not fully convert into the laser energy or other radiations due to the nonradiative transitions are existent. Heat will come into being because of the

nonradiative transitions and also remain in the laser crystal. Therefore, the laser medium usually needs cooling to lower its temperature. It would be better that if the temperature of the laser medium can be known, then the cooling process can be controlled. It is known that Tm3+ can be a temperature sensing probe ion based on the fluorescence branching ratio technique [31]. For Tm3+ ion, 3F2, 3F3 and 3H4 states are in a thermal equilibrium, thus following equations can be gotten, I 3 F 2  I10 exp（

 E1 ） kT

(8)

 E2 ） kT

(9)

 E3 ） kT

(10)

I 3 F 3  I 20 exp（

I 3 H 4  I 30 exp（

In above equations, I3F2, I3F3 and I3H4 are the luminescent intensities for 3F2→3H6, 3

F3→3H6, and 3H4→3H6 transitions at temperature T; Ij0 (here j = 1, 2 and 3) present

the luminescent intensities of 3F2, 3F3 and 3H4 levels at temperature approximately closing to 0 K; Ej (j = 1, 2 and 3) stands for the energy of 3F2, 3F3 or 3H4 levels; k represents the Boltzmann’s constant, T is absolute temperature. For a certain Tm3+ doped system, the fluorescence intensity ratio (FIR) Rij can be expressed as Rij 

 Eij Ii  I ij exp（ ） Ij kT

(11)

where Ii and Ij are arbitrary two amongst the fluorescence intensities I3F2, I3F3 and I3H4; Iij is the ratio of Ii0 to Ij0, and Eij = Ei - Ej. Eq. (11) tells that the fluorescence intensity ratio depends only on the temperature of the studied specific system doped

with Tm3+, thus one the fluorescence intensity ratio is known, the temperature of the single crystal can be self-read out, namely, the temperature monitor for the Tm3+ doped crystal is possible when laser is working. To check the temperature sensing performance, the temperature-dependent up-conversion emission spectra of Tm3+/Yb3+ co-doped LiYF4 single crystal were measured under 980 nm excitation in the temperature region of 313-713 K and are shown in Fig. 3. The emission peaks centered at 687, 704 and 801 nm correspond to the electron transitions 3F2→3H6, 3F3→3H6, and 3H4→3H6 of Tm3+, respectively, and their intensities change with the increase of crystal temperature, thus indicating the temperature sensing can be achieved. However, it is found that the emission peaks for 3

F2→3H6 and 3F3→3H6 transitions are overlapped together due to the narrow energy

distance between

3

F2 and 3F3 levels, thus resulting in difficulty for practical

applications since the spectral decomposition for measuring technique is hard and increasing cost. By inspecting the spectra in Fig. 3, it can be found that 3H4→3H6 emission peak is well separated from 3F2,3→3H6 emissions, therefore, we define a fluorescence intensity ratio R  I 3F 2  I 3F 3  I 3H 4 , and then taking Eqs. (8)-(10) into this ratio R, it can be easily derived that R  I 3 F 2  I 3 F 3  I 3 H 4  I13 exp（

 E13  E23 ） I 23 exp（ ） kT kT

(12)

All the symbols in Eq. (12) have their meaning as earlier defined in this paper. Eq. (12) displays that the fluorescence intensity ratio R is also determined by sample temperature, thus it can be used for temperature sensing. <
>

To examine the temperature performance of LiYF4:Tm3+/Yb3+ single crystal, the integrated fluorescence intensities for 3F2,3→3H6 and 3H4→3H6 transitions were individually calculated, and then the fluorescence intensity ratios R at different temperatures were also obtained. The temperature-dependence of fluorescence intensity ratio is shown in Fig. 3 as dispersed squared dots. Eq. (12) was fit to the data in Fig. 3 by taking N13, N23, E13/k and E23/k as free parameters. It can be found that Eq. (12) can fit well with the experimental data, and the free parameters N13, N23, E13/k and E23/k were confirmed to be 1.90.1, 3.80.04, 2949.410.0 K and 2433.310.0 K. By using the values of E13/k and E23/k, the E13 and E23 were confirmed to be 2059 and 1691 cm-1 which are very close to the experimentally measured values of about 2055 and 1693 cm-1 derived from the spectra in Fig. 3. The relative sensitivity is an important factor for characterizing sensors. For optical temperature sensing based on the fluorescence intensity ratio, the relative sensitivity Sr is usually defined as[32], Sr 

1 dR R dT

(13)

Since the R has already been confirmed, the Sr can be obtained via simple mathematical calculation, and the result is marked in the Fig. 4 as circle dots. For comparison purpose, we collected the relative sensitivities of Tm3+ in different hosts, and listed them in Table 3. It is found that the relative sensitivity Sr of our studied sample has a relatively high value at low temperature, which indicates that Tm3+/Yb3+ co-doped LiYF4 crystal may be good candidate for temperature sensor, and the single crystal temperature may can be self-read out when the laser is operating.

<
> <
>

4. Conclusion The optical transition intensity parameters were calculated by adopting different absorption transition numbers, and it was found that the obtained results are very different from each other, and it is confirmed that involving more absorption transitions into the Judd-Ofelt calculation can obtain more reliable results. By carefully inspecting the up-conversion spectra measured at various temperatures, it was proposed that the fluorescence intensity ratio of 3F2+3F3→3H6 to 3H4→3H6 can be used for achieving better temperature detection, and the single temperature might be self-read out during the laser operation.

Acknowledgments This work was partially supported by NSFC [National Natural Science Foundation of China, grant number 11774042], Fundamental Research Funds for the Central Universities [grant number 3132016333], China Postdoctoral Science Foundation [grant number 2016M591420], the Natural Science Foundation of Zhejiang Province [grant number. LZ17E020001], the Open Fund of the State Key Laboratory on Integrated Optoelectronics [IOSKL2015KF27].

References [1] Y.A. Vladimirov, A.N. Osipov, G.I. Klebanov, Photobiological principles of therapeutic applications of laser radiation, Biochemistry (Mosc) 69(2004) 81-90. [2] S.C. Tam, R. Williams, L.J. Yang, S. Jana, L.E.N. Lim, M.W.S. Lau, A review of the laser processing of aircraft components, J. Mater. Process. Technol. 23(1990) 177-194. [3] C.Y. Yeo, S.C. Tam, S. Jana, M.W.S. Lau, A technical review of the laser drilling of aerospace materials, J. Mater. Process. Technol. 42(1994) 15-49. [4] A.A. Kaminkii, K.S. Bagdasarov, L.M. Belyev, Analysis of the optical spectra and stimulated 3+

radiation of Y3Al5O12-Nd crystals, Phys. Status Solidi B 21(1967) K23-K29. [5] P.F. Bordui, J.C. Jacco, G.M Loiacono, R.A. Stolzenberger, J.J. Zola, Growth of large single crystals of KTiOPO4 (KTP) from high-temperature solution using heat pipe based furnace system, J. Cryst. Growth 84(1987) 403-408. [6] R.C. Powell, J.L. Caslavsky, Z. AlShaieb, J.M. Bowen, Growth, characterization, and optical spectroscopy of Al2O3:Ti3+, J. Appl. Phys. 58(1985) 2331. [7] A.M. Tkachuk, I.K. Razumova, M.F. Joubert, R. Moncorge, Up-conversion in YLiF: Yb3+, Tm3+ laser crystals, Proc. SPIE. 3682(1998) 76-82. [8] D.F.D. Sousa, V. Peters, G. Huber, A. Toncelli, D. Parisi, M. Tonelli, Pump modulation frequency resolved excited state absorption spectra in Tm3+ doped YLiF, Appl. Phys. B 77(2003) 817-822. [9] S. Li, H. Xia, Y. Dong, L. Fu, X. Gu, J. Zhang, D. Wang, H. Jiang, B. Chen, Luminescent properties of Tm3+/Ho3+ co-doped LiYF4 crystals, Optoelectron. Lett. 10(2014) 443-446. [10] J.F. Suyver, J. Grimm, M.K.V. Veen, D. Biner, K.W. Kramer, H.U. Gudel, Upconversion spectroscopy and properties of NaYF4 doped with Er3+, Tm3+ and/or Yb3+, J. Lumin. 117(2006) 1-12. [11] Z. Li, L. Zheng, L. Zhang, L. Xiong, Synthesis, characterization and upconversion emission properties of the nanocrystals of Yb3+/Er3+-codoped YF3-YOF-Y2O3 system, J. Lumin. 126(2007) 481-486. [12] K.J. Kim, A. Jouini, A. Yoshikawa, R. Simura, G. Boulon, T. Fukuda, Growth and optical properties of Pr, Yb-codoped KY3F10 fluoride single crystal for up-conversion visible

luminescence, J. Cryst. Growth 299(2007) 171-177. [13] S. Yang, H. Xia, J. Zhang, Y. Jiang, Y. Shi, X. Gu, J. Zhang, Y. Zhang, H. Jiang, B. chen, Growth and spectral properties of Ho3+ doped-NaYF4 single crystal, Opt. Mater. 45(2015) 209-214. [14] T. Jensen, A. Diening, G. Huber, B.H.T. Chai, Investigation of diode-pumped 2.8-μm Er:LiYF4 lasers with various doping levels, Opt. Lett. 21(1996) 585-587. [15] T. Chuang, H.R. Verdun, Energy transfer up-conversion and excited state absorption of laser radiation in Nd:YLF laser crystals, IEEE J. Quantum Electron. 32(1996) 79-91. [16] M. Pollnau, P.J. Hardman, W.A. Clarkson, D.C. Hanna, Upconversion, lifetime quenching, and ground-state bleaching in Nd3+: LiYF4, Opt. Commun. 147(1998) 203-211. [17] N.P. Barnes, W.J. Rodriguez, B.M. Walsh, Ho:Tm:YLF laser amplifiers, J. Opt. Soc. Am. B 13(1996) 2872-2882. [18] C.M. Combes, P. Dorenbos, C.W.E. van Eijk, C. Pedrini, H.W.D. Hartog, J.Y. Gesland, P.A. Rodnyi, Optical and scintillation properties of Ce3+ doped LiYF4 and LiLuF4 crystals, J. Lumin. 71(1997) 65-70. [19] H. Hu, H. Xia, P. Wang, J. Peng, Y. Zhang, H. Jiang, B. Chen, Near and mid-infrared emission characteristics of Er3+/Tm3+/Ho3+-doped LiYF4 single crystals excited by laser diode, J. Nanosci. Nanotechnol. 16(2016) 526-531. [20] S. Yang, H. Xia, J. Zhang, Y. Jiang, X. Gu, J.Z. Zhang, D. Wang, H. Jiang, B. Chen, Energy transfer and 1.8 μm emission in Tm3+/Yb3+ co-doped LiYF4 crystal, J. Nanosci. Nanotechnol. 16(2016) 537-541. [21] L. Fu, H. Xia, Y. Dong, S. Li, X. Gu, J. Zhang, D. Wang, H. Jiang, B. Chen, Efficient near-infrared quantum cutting in Tm3+/Yb3+ codoped LiYF4 single crystals for solar photovoltaic, Chin. J. Chem. Phys. 28(2015) 73-78. [22] G. Zhu, A 60W Tm:YLF Laser with Triple Tm:YLF Rods, Chin. Phys. Lett. 32(2015) 094207. [23] S. Li, H. Xia, L. Fu, Y. Dong, X. Gu, J. Zhang, D. Wang, Y. Zhang, H. Jiang, B. Chen, Optimum fluorescence emission around 1.8 μm for LiYF4 single crystals of various Tm3+-doping concentrations, Chin. Phys. B 23(2014) 7806. [24] B.R. Judd, Optical absorption intensities of rare-earth ions, Phys. Rev. 127(1962) 750.

[25] G.S. Ofelt, Intensities of crystal spectra of rare-earth ions, J. Chem. Phys. 37(1962) 511. [26] Y. Zheng, B Chen, J. Sun, L. Cheng, H. Zhong, Composition-dependent optical transitions and 2.0 m emission properties of Ho3+ in xGeO2-(80-x)TeO2-9.5ZnF2-10NaF-0.5Ho2O3 glasses, J. Rare Earths 29(2011) 924. [27] P.A. Ryabochkina, S.A. Antoshkina, E.V. Bolshakova, M.A. Ivanov, V.V. Kochurihin, A.V. Malov, S.N. Ushakov, N.V. Shchuchkina, K.N. Nishchev, Hypersensitive transitions of Tm3+, Ho3+ and Dy3+ rare-earth ions in garnet crystals, J. Lumin. 132(2012) 1900-1905. [28] B. Wang, L. Cheng, H. Zhong, J. Sun, Y. Tian, X. Zhang, B. Chen, Excited state absorption cross sections of 4I13/2 of Er3+ in ZBLAN, Opt. Mater. 31(2009) 1658-1662. [29] F. Liu, J.H. Zhang, S. Z. Lu, S. X. Liu, S. H. Huang, X. J. Wang, Explicit effects of 4f5d configuration on 4f24f2 electric dipole transitions in Pr3+-doped SrAl12O19, Phys. Rev. B, 74(2006)115112. [30] M. V. V. Kumar, K. R. Gopal, R. R. Reddy, G. V. L. Reddy, N. S. Hussain, B. C. Jamataiah, Application of modified Judd-Ofelt theory and the evaluation of radiative properties of Pr3+-doped lead telluroborate glasses for laser applications, J. Non-Cryst. Solid., 364(2013)20-27. [31] L. Tong, X. Li, R. Hua, L. Cheng, J. Sun, J. Zhang, S. Xu, H. Zheng, Y.Q. Zhang, Optical temperature sensing properties of Yb3+/Tm3+ co-doped, NaLuF4 crystals, Curr. Appl. Phys. 17 (2017) 999-1004. [32] A.F. Pereira, J.F. Silva, A.S. Gouveia-Neto, C. Jacinto, 1.319 μm excited thulium doped nanoparticles for subtissue thermal sensing with deep penetration and high contrast imaging, Sens. Actuators B. 238(2017) 525 -531. [33] A.F. Pereira, K. Upendra Kumar, W.F. Silva, W.Q. Santos, D. Jaque, C. Jacinto, Yb3+/Tm3+ co-doped NaNbO3 nanocrystals as three-photon-excited luminescent nanothermometers. Sens. Actuators B. 213(2015) 65 -71. [34] V. Lojpur, M. Nikolic, L. Mancic, O. Milosevic, M.D. Dramicanin, Y2O3:Yb, Tm and Y2O3: Yb, Ho powders for low-temperature thermometry based on up-conversion fluorescence, Ceram. Int. 39(2013) 1129–1134. [35] X. Wang, J. Zheng, Y. Xuan, X. Yan, Optical temperature sensing of NaYbF4:Tm3+@SiO2 core-shell micro-particles induced by infrared excitation, Opt. Expr. 21(2013) 21596–21606.

Figure captions Fig. 1 XRD pattern for LiYF4:Tm3+ single crystal measured by using the powders derived from grinding the bulk crystal. Fig. 2 Ground state absorption cross section spectrum of LiYF4:Tm3+ single crystal. Fig. 3 Temperature evolution of emission spectra for the Tm3+/Yb3+ co-doped LiYF4 crystal. Fig. 4 Dependence of R on temperature (squares indicates the experimental results; solid curve is the fitting result), and dependence of relative sensitivity on the temperature for Tm3+/Yb3+ co-doped LiYF4 crystal.

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Table arrangements Table 1 Absorption Transitions 3

H6→3F4 3 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→1G4 3 H6→1D2

Experimental Oscillator Strengths (10-8) 344.43 453.32 652.56 858.37 211.21 517.03

Table 1 The experimental oscillator strengths of observed transitions in Tm3+ doped LiYF4 single crystal.

Table 2 Number of Transitions

Absorption Transitions 3

H6→3F4 H6→3H5 3 H6→3H4 3 H6→3F4 3 H6→3H5 3 H6→3F2, 3 3 H6→3F4 3 H6→3H5 3 H6→1G4 3 H6→3F4 3 H6→3H5 3 H6→1D2 3 H6→3F4 3 H6→3H4 3 H6→3F2, 3 3 H6→3F4 3 H6→3H4 3 H6→1G4 3 H6→3F4 3 H6→3H4 3 H6→1D2 3 H6→3F4 3 H6→3F2, 3 3 H6→1G4 3 H6→3F4 3 H6→3F2, 3 3 H6→1D2 3 H6→3F4 3 H6→1G4 3 H6→1D2 3 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→3H5 3 H6→3H4 3 H6→1G4 3 H6→3H5 3 H6→3H4 3 H6→1D2 3 H6→3H5 3 H6→3F2, 3 3 H6→1G4 3 H6→3H5 3 H6→3F2, 3 3 H6→1D2 3 H6→3H5 3 H6→1G4 3 H6→1D2 3 H6→3H4 3 H6→3F2, 3 3 H6→1G4 3

3

Experimental Oscillators Strength (10-8) 344.43 453.32 652.56 344.43 453.32 858.37 344.43 453.32 211.21 344.43 453.32 517.03 344.43 652.56 858.37 344.43 652.56 211.21 344.43 652.56 517.03 344.43 858.37 211.21 344.43 858.37 517.03 344.43 211.21 517.03 453.32 652.56 858.37 453.32 652.56 211.21 453.32 652.56 517.03 453.32 858.37 211.21 453.32 858.37 517.03 453.32 211.21 517.03 652.56 858.37 211.21

Theoretical Oscillators Strength # (10-8) 344.43 412.92(40.4) 652.56 344.43 412.92(40.4) 858.37 344.43 412.92(40.4) 211.21 344.43 412.92(40.4) 517.03 344.43 652.56 858.37 344.43 652.56 211.21 344.43 652.56 517.03 344.43 858.37 211.21 344.43 858.37 517.03 344.43 211.21 517.03 412.92(40.4) 652.56 858.37 412.92(40.4) 652.56 211.21 412.92(40.4) 652.56 517.03 412.92(40.4) 858.37 211.21 412.92(40.4) 858.37 517.03 412.92(40.4) 211.21 517.03 652.56 858.37 211.21

Judd-Ofelt Parameters Ωλ (λ=2, 4, 6) (10-20 cm2) 4.92 1.85 5.77 17.00 -7.55 7.15 -49.62 44.30 -0.43 2.40 3.81 5.49 7.77 0.01 4.98 -124.33 85.44 42.03 2.25 3.51 6.50 -39.53 38.74 -6.16 2.50 4.32 3.74 4.35 14.57 -31.11 9.95 8.33 2.58 8.30 6.21 3.63 6.62 4.04 4.70 11.96 3.81 3.88 11.73 4.33 3.74 11.22 4.29 3.83 9.25 5.65 3.35

Errors

3

H6→3H4 H6→3F2, 3 3 H6→1D2 3 H6→3H4 3 H6→1G4 3 H6→1D2 3 H6→3F2, 3 3 H6→1G4 3 H6→1D2 3 H6→3F4 3 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→3F4 3 H6→3H5 3 H6→3H4 3 H6→1G4 3 H6→3F4 3 H6→3H5 3 H6→3H4 3 H6→1D2 3 H6→3F4 3 H6→3H5 3 H6→3F2, 3 3 H6→1G4 3 H6→3F4 3 H6→3H5 3 H6→3F2, 3 3 H6→1D2 3 H6→3F4 3 H6→3H5 3 H6→1G4 3 H6→1D2 3 H6→3F4 3 H6→3H4 3 H6→3F2, 3 3 H6→1G4 3 H6→3F4 3 H6→3H4 3 H6→3F2, 3 3 H6→1D2 3 H6→3F4 3 H6→3H4 3 H6→1G4 3 H6→1D2 3 H6→3F4 3 H6→3F2, 3 3 H6→1G4 3 H6→1D2 3 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→1G4 3 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→1D2 3

4

652.56 858.37 517.03 652.56 211.21 517.03 858.37 211.21 517.03 344.43 453.32 652.56 858.37 344.43 453.32 652.56 211.21 344.43 453.32 652.56 517.03 344.43 453.32 858.37 211.21 344.43 453.32 858.37 517.03 344.43 453.32 211.21 517.03 344.43 652.56 858.37 211.21 344.43 652.56 858.37 517.03 344.43 652.56 211.21 517.03 344.43 858.37 211.21 517.03 453.32 652.56 858.37 211.21 453.32 652.56 858.37 517.03

652.56 858.37 517.03 652.56 211.21 517.03 858.37 211.21 517.03 347.50 368.07(40.4) 668.82 892.50 347.46 413.58(40.4) 650.17 105.22 347.45 429.51(40.4) 626.55 462.81 348.57 405.85(40.4) 864.35 85.22 353.24 351.62(40.4) 919.04 456.82 347.06 412.09(40.4) 109.91 521.80 347.57 650.36 858.89 102.86 358.83 583.28 902.10 329.22 347.29 651.23 107.13 519.14 346.82 857.59 114.72 522.35 404.17(40.4) 656.15 864.93 231.82 393.22(40.4) 666.19 869.27 534.60

8.90 4.32 3.74 11.02 4.59 2.83 11.20 4.32 3.74 7.78 0.00 5.18

58.74

4.84 1.98 5.76

106.06

3.24 3.12 5.89

62.46

15.46 -6.20 6.80

126.40

3.58 3.52 4.31

105.55

2.41 3.87 4.45

101.45

7.72 0.12 4.95

108.42

5.53 2.09 4.63

205.41

2.41 3.55 6.45

104.15

2.49 4.39 3.71

96.67

9.54 6.69 3.09

23.61

9.21 4.50 3.75

31.64

3

H6→3H5 H6→3H4 3 H6→1G4 3 H6→1D2

453.32 652.56 211.21 517.03

407.38(40.4) 660.78 174.46 535.88

7.72 4.35 4.30

42.48

3

H6→3H5 H6→3F2, 3 3 H6→1G4 3 H6→1D2

453.32 858.37 211.21 517.03

412.13(40.4) 859.19 215.77 516.01

11.63 4.31 3.74

4.81

3

H6→3H4 H6→3F2, 3 3 H6→1G4 3 H6→1D2

652.56 858.37 211.21 517.03

653.90 786.18 200.01 521.27

10.04 4.51 3.26

73.19

3

344.43 453.32 652.56 858.37 211.21 344.43 453.32 652.56 858.37 517.03 344.43 453.32 652.56 211.21 517.03 344.43 453.32 858.37 211.21 517.03 344.43 652.56 858.37 211.21 517.03 453.32 652.56 858.37 211.21 517.03 344.43 453.32 652.56 858.37 211.21 517.03

350.61 368.23(40.4) 666.57 892.89 103.41 361.61 369.52(40.4) 599.91 934.56 331.67 350.19 429.08(40.4) 625.89 108.50 466.28 355.64 351.59(40.4) 918.28 114.56 462.12 361.35 582.92 901.53 112.55 333.69 393.83(40.4) 667.78 867.52 194.28 540.46 364.11 369.63(40.4) 599.51 933.9 112.96 336.11

7.73 0.11 5.15

86.73

5.56 2.06 4.83

150.19

3.27 3.17 5.86

84.06

3.57 3.58 4.30

99.58

5.54 2.14 4.61

158.66

9.30 4.57 3.71

27.53

5.58 2.11 4.81

133.12

3

3

3

H6→3F4 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→1G4 3 H6→3F4 3 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→1D2 3 H6→3F4 3 H6→3H5 3 H6→3H4 3 H6→1G4 3 H6→1D2 3 H6→3F4 3 H6→3H5 3 H6→3F2, 3 3 H6→1G4 3 H6→1D2 3 H6→3F4 3 H6→3H4 3 H6→3F2, 3 3 H6→1G4 3 H6→1D2 3 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→1G4 3 H6→1D2 3 H6→3F4 3 H6→3H5 3 H6→3H4 3 H6→3F2, 3 3 H6→1G4 3 H6→1D2 3

5

6

# in this column the value in a bracket presents the oscillator strength for magnetic transition Table 2 The dependence of experimental oscillators, theoretical oscillators, and Judd-Ofelt parameters on absorption transitions in Tm3+ doped LiYF4 crystal.

Table 3 Ions: host

Srmax (% K−1)

T (K)

Temperature range (K)

References

Tm/Yb: NaNbO3

0.08

-

293-353

[33]

Tm/Yb: Y2O3 (powders)

5.0

300

100-300

[34]

Tm: NaYbF4@SiO2

0.054

430

400-700

[35]

Tm/Yb: LiYF4 crystal

2.98

313

313-713

this work

Table 3 The maximum relative sensitivity Sr of different systems doped with Tm3+, the temperature for maximum relative sensitivity and the temperature range are included.