Mössbauer and Fluorescence Spectroscopic Study on the Local

Jun 27, 1996 - Hikari Shigemura, Masanori Shojiya, Ryoji Kanno, and Yoji Kawamoto , Kohei Kadono , Masahide Takahashi. The Journal of Physical ...
0 downloads 0 Views 294KB Size
J. Phys. Chem. 1996, 100, 11193-11197

11193

Mo1 ssbauer and Fluorescence Spectroscopic Study on the Local Environment around Trivalent Europium Ion in the Chlorofluorozirconate Glass System Masahide Takahashi,† Ryoji Kanno, and Yoji Kawamoto* Photonics Materials Laboratory, Graduate School of Science and Technology, Kobe UniVersity, Nada, Kobe 657, Japan ReceiVed: February 5, 1996; In Final Form: April 18, 1996X

Chemical bonding state and fluorescence properties of Eu3+ ions in the ZrF4-BaF2 glass system in which a part of the F- ions are successively substituted by Cl- ions have been investigated by means of 151Eu-Mo¨ssbauer and Eu3+-fluorescence spectroscopies. Mo¨ssbauer spectra indicated that Eu in the chlorofluorozirconate glasses exists only as the +3 oxidation state. Isomer shift of 151Eu3+ increased with an increasing amount of Cl-. This change corresponded to the intensity of the hypersensitive 5D0 f 7F2 transition in Eu3+ fluorescence. This suggests that the intensity of the hypersensitive 5D0 f 7F2 transition depends on the density of 6s electrons at the Eu nucleus. The intensity of the 5D0 f 7F0 transition increases with increasing Cl- amount. Ligand field strength around Eu3+ and the inhomogeneous line width of Eu3+ fluorescence calculated from fluorescence spectra of the 5D0 f 7F0 transition decreased with increasing amount of Cl-. These results revealed that even a small amount of Cl- substitution greatly changes the local environments around Eu3+.

Introduction Rare-earth-doped fluoride glasses (RDFG) have various applications as active optical functional materials such as high efficiency up-conversion phosphors,1,2 lasers,3,4 and optical fiber amplifiers for 1.3 µm optical telecommunication.5 Such a potentiality originates from the peculiar characteristics of RDFGs: optical transmittance in the range from the middle infrared to the near ultraviolet,6 incorporation of rare-earth elements,7 and low phonon energies.8 It is widely known that the optical properties of rare-earth ions doped in glasses are affected by the structures and compositions of the hosts. A number of studies have already been made on glass structures of RDFGs and local environments around rare-earth ions by means of X-ray or neutron diffraction techniques,9 EXAFS analyses,10,11 fluorescence line narrowing (FLN) techniques,12 and molecular dynamics (MD) simulations.13,14 The fluorescence properties and local environments of Eu3+ ions in several glass systems have been reported by using FLN and MD techniques.15-18 However, studies which treat the chemical bonding states of rare-earth ions in RDFGs are very few.19 Previously we reported a series of studies on the rare-earthdoped ZrF4-BaF2-BaCl2 glass system.20-22 The result of the MD simulation revealed that most of the Cl- ions substituted for F- ions exist around Ba2+ and Eu3+.21 EXAFS analysis clarified that the local structure around Zr4+ is hardly influenced by the Cl- substitution.22 Mo¨ssbauer spectroscopy provides useful information for characterizing the valence state and coordination environments around Mo¨ssbauer-active cations in glasses.23 Europium is one of the appropriate elements with respect to clarifying the electron charge density at the nucleus. The 151Eu resonance has previously been observed for Eu3+ in several oxide and fluoride glasses.24,25 Trivalent europium is also the most suitable element † Present address: Frontier Materials Laboratory, Toyota Technological Institute, 1-12-1, Hisakata, Tempaku, Nagoya 468, Japan (email: masahide@ toyota-ti.ac.jp). * Corresponding author. Tel: +81-78-803 0591. Fax: +81-78-803-0722. Email: [email protected]. X Abstract published in AdVance ACS Abstracts, June 1, 1996.

S0022-3654(96)00349-8 CCC: $12.00

among rare earths to evaluate the changes in fluorescence characteristics because its electronic levels are quite simple.26 In the present study, we have measured the 151Eu-Mo¨ssbauer and Eu3+-fluorescence spectra for the ZrF4-BaF2-BaCl2 glasses doped with Eu3+ to examine changes in the chemical bonding-states and f-f transition characteristics with successive Cl- substitution for F-. Electron densities at the nucleus are discussed by calculating the isomer shifts from the resonance absorption spectra of 151Eu. Ligand field strength, inhomogeneous line width, and intensity of hypersensitive transition of Eu3+ are clarified by analyzing Eu3+-fluorescence spectra. Experimental Procedure (a) Sample Preparation. Glass compositions employed are 58ZrF4‚(39-x)BaF2 ‚xBaCl2‚3EuF3 (x ) 0, 5, 10, 15, 19, 24, and 29). High-purity fluorides were used as the starting materials. Batches were mixed and melted in a Pt crucible at 700 °C for 15 min under an Ar gas atmosphere and quenched into a brass mold preheated to about 100 °C, as described elsewhere.20 The glasses obtained were annealed at the respective glass transition temperatures determined by a differential thermal analysis. The accurate Cl contents of the prepared glasses were determined by an X-ray fluorescence analysis. The molar volumes of the glasses were obtained by density measurement with an Archimedean method. The refractive indexes were measured with an Abbe refractometer. Hereafter, the x ) 0, 5, 10, 15, 19, 24, and 29 glasses are abbreviated as glasses 0Cl, 3Cl, 6Cl, 9Cl, 12Cl, 15Cl, and 18Cl, respectively, in terms of the anion ratio of Cl to (F + Cl) in the batch composition, i.e. 0, 3, 6, 9, 12, 15, and 18%. (b) Spectroscopic Measurements. Room-temperature Mo¨ssbauer measurements were recorded in transmission geometry. A 0.93 GBq 151Sm2O3 source was used. The velocity range of the spectrometer was set between -18 and 18 mm/s. The velocity scale was calibrated with the magnetic hyperfine spectrum of R-Fe using a 57Co source. The zero velocity was taken as the absorption of 151Eu in crystalline EuF3. Measurements were made on the 0Cl, 6Cl, 12Cl, and 15Cl glasses. Emission spectra were measured with a HITACHI F-3010 fluorescence spectrometer. The fluorescence spectra of Eu3+ © 1996 American Chemical Society

11194 J. Phys. Chem., Vol. 100, No. 26, 1996

Takahashi et al. TABLE 2: Isomer Shifts, δ, Line Widths at Half-Maximum, fwhm, and R-Factors in Least Squares Calculations in Mo1 ssbauer Spectra of Glasses 0Cl, 6Cl, 12Cl, and 15Cl and Crystalline EuF3

Figure 1. 151Eu-Mo¨ssbauer spectra of glasses 0Cl, 6Cl, 12Cl, and 15Cl and crystalline EuF3.

TABLE 1: Cl Contents, Cl/(F + Cl), Glass Transition Temperatures, Tg, Refractive Indexes, nD, and Molar Volumes, Vm, of 58ZrF4‚(39-x)BaF2‚xBaCl2‚3EuF3 Glasses glass x

notation

0 5 10 15 19 24 29

0Cl 3Cl 6Cl 9Cl 12Cl 15Cl 18Cl

Cl/(Cl + F) (%) batch

analyzed

nD

Tg (°C)

Vm (cm3/mol)

0 3 6 9 12 15 18

0.00 1.71 3.20 5.09 6.23 7.50 8.95

1.5241 1.5320 1.5435 1.5507 1.5615 1.5665 1.5695

309 287 285 276 271 262 249

36.6 37.4 38.1 39.4 40.1 40.5 41.4

were measured in the wavelength range from 550 to 650 nm under 395 nm excitation. Results and Discussion The analyzed Cl- amounts of the prepared 58ZrF4‚(39-x)BaF2‚xBaCl2‚3EuF3 glasses are given as the anion ratios [Cl-]/ ([F-] + [Cl-]) × 100, which is abbreviated as Cl/(F + Cl), in Table 1, together with the refractive indexes, the glass transition temperatures, and the molar volumes. The analyzed Cl contents were almost in proportion to the Cl contents in batch compositions. The glass transition temperature decreased and the refractive index and molar volume increased in proportion to increasing amount of Cl-. (a) 151Eu-Mo1 ssbauer Spectra. Isomer shift, δ, and quadrupole interaction, QI, determined from 151Eu-Mo¨ssbauer measurements are in proportion to the s-electron density and electric field gradient at the 151Eu nucleus, respectively.23 That is, the δ and QI values represent the covalency and symmetry of the ligand field around Eu. Figure 1 shows the 151Eu-Mo¨ssbauer spectra of glasses 0Cl, 6Cl, 12Cl, and 15Cl and crystalline EuF3. The respective 151Eu-Mo¨ssbauer spectra consist of a broadened resonance line. The positions of the absorption line revealed that Eu in the present chlorofluorozirconate glasses exists only as the +3 oxidation state. No trace of Eu2+ was observed in the measurements. The fine structure of the QI was hardly detected because of the large line width. The peak position and the line width at half-maximum of peak, fwhm, were calculated by employing a least squares fitting program of the Gauss-Lorentz convoluted function. The isomer shift of 151Eu3+ was determined from the center shift of the peak with respect to crystalline EuF3. The calculated isomer shifts and fwhm’s of resonant peaks are given in Table 2. Although

glass

δ × 102 (mm/s)

fwhm (mm/s)

R-factor (%)

0Cl 6Cl 12Cl 15Cl

-5.03 1.14 6.03 7.73

1.55 1.50 1.63 1.55

82.8 93.7 95.0 96.3

EuF3

0

2.71

99.7

vitreous materials generally give broadened line widths compared with crystalline materials because of structural variation of ions from site to site, the fwhm’s of 151Eu resonance absorption in chlorofluorozirconate glasses are smaller than that in crystalline EuF3. This suggests that the quadrupole interactions of 151Eu3+ in these glasses are smaller than that in crystalline EuF3. The crystal structure of EuF3 is of the LaF3 (tysonite) type, which has three kinds of nonequivalent F-. In the case of LaF3, the neighbors of La are comprised of 7F at 2.42-2.48 Å, 2F at 2.64 Å, and 2F at 3.00 Å.27 The large fwhm value in crystalline EuF3 suggests that the asymmetric lattice contributes to the electric field gradient. As shown in Table 2, the δ value increases with increasing Cl/(F+Cl). An increase in δ can be attributed to greater participation of the 6s electrons to covalent bonds. The δ value of crystalline EuCl3 is 0.3 mm/s with respect to EuF3.28 The difference in δ between EuF3 and EuCl3 arises from a difference in bond character. The covalency in the Eu-Cl bond is much greater than that in the Eu-F bond. Therefore, an increase in isomer shifts with increasing amount of Cl- substitution means an increase in the number of covalent Eu-Cl bonds around Eu3+. (b) Hypersensitive 5D0 f 7F2 Transition. The strength of most f-f transitions of trivalent rare-earth ions is little affected by the local environment of the ions. A few transitions, however, are very sensitive to the local environment, so that these transitions are called hypersensitive transitions.29 In the observable emission bands of Eu3+, the emission line due to the 5D0 f 7F2 transition is the hypersensitive transition, being sensitive to the local environment around Eu3+. The emission spectra of 5D0 f 7F2 transition of Eu3+ in chlorofluorozirconate glasses are shown in Figure 2a, together with the magnetic dipole 5D0 f 7F1 transition whose intensity is independent of the local environment of the ion. The 5D0 f 7F1 transition is the parity-allowed magnetic dipole transition, and its strength hardly varies with the glass composition. Therefore, the intensity of the 5D0 f 7F2 transition can be normalized by the intensity of the 5D0 f 7F1 transition. To calculate accurate peak area of the 5D0 f 7F1 emission band, the deconvolution of the 5D f 7F and 5D f 7F emission bands was performed by 1 3 0 1 curve-fitting using the least squares methods as described elsewhere.30 The root mean square values in the least squares fitting were Br- > Cl- > H2O > F-. (3) The hypersensitivity is proportional to the nephelauxetic ratio and to covalency.31 Figure 3 shows the relationship between the normalized intensity of the 5D0 f 7F2 transition, I(0-2)/

Europium Ion in the Chlorofluorozirconate Glass System

J. Phys. Chem., Vol. 100, No. 26, 1996 11195

Figure 4. Emission spectra of the 5D0 f 7F0 transition of Eu3+ in glasses 0Cl, 3Cl, 6Cl, 9Cl, 12Cl, 15Cl, and 18Cl. (The intensities are normalized by the integrated intensities of the 5D0 f 7F1 transition.)

Figure 2. (a) Emission spectra of the 5D0 f 7F2 transition in glasses 0Cl, 3Cl, 6Cl, 9Cl, 12Cl, 15Cl, and 18Cl. (The intensities are normalized by the integrated intensities of the 5D0 f 7F1 transition.) (b) Cl- amount, Cl/(F + Cl), dependence of the integrated intensity of the 5D0 f 7F2 transition normalized by that of the magnetic dipole 5D0 f 7F1 transition, I(0-2)/I(0-1).

Figure 3. Relationship between the normalized intensity of the 5D0 f 7F2 transition, I(0-2)/I(0-1), and isomer shift, δ. The dotted line is a guide for the eye.

I(0-1), and the isomer shift, δ, calculated from Mo¨ssbauer spectra. A linear correlation between both values suggests that the strength of the hypersensitive 5D0 f 7F2 transition depends on the density of the 6s electrons at the Eu nucleus. In other words, the large intensity of the hypersensitive transition is caused by an increase in covalency of the Eu3+ ligand field.

Such a change in chemical bonding is due to the replacement of Cl- for F- at neighbors of Eu3+. Thus, it is concluded that the replacement of Cl- for F- occurs at neighbors of Eu3+ and that the hypersensitive transition appears with large intensity by the Cl- substitution for F-. (c) 5D0 f 7F0 Transition of Eu3+. According to the JuddOfelt theory,32,33 the transitions between the J ) 0 levels are strictly forbidden. However, the 5D0 f 7F0 transition in the fluorescence spectra of Eu3+ in certain matrices can be observed with extremely low intensity.34 Three types of mechanisms are proposed to explain this transition: (1) The transition is induced by the breakdown of the closure approximation in the JuddOfelt theory. (2) The spin-selection rule is relaxed by the spinorbital interaction acting within the high-lying odd-parity states which serve as the intermediate states of the f-f transitions (Wybourne-Downer theory).35,36 (3) The J-mixing effect (the mixing of the 4f6 states into the 5D0 and 7F0 states by the evenparity crystal field perturbation) makes it possible for the 5D0 f 7F0 transition to borrow intensity from the 5D0 f 7FJ (J ) 2, 4, and 6) and 5DJ f 7F0 (J ) 2 and 4) transitions.37 In the present chlorofluorozirconate glasses, the 5D0 f 7F0 transition can be seen in the spectra of Eu3+, as shown in Figure 4. The intensity of the 5D0 f 7F0 transition increases with increasing Cl- amount. Tanaka et al. have reported that the mixture of the 7F2 state into the 7F0 state by the J-mixing effect plays a dominant role in the intensity of the 5D0 f 7F0 fluorescence of Eu3+ in sodium borate and sodium germanate glasses.38 Figure 5 shows the relationship between the integrated intensity ratio of the 5D0 f 7F0 to 5D0 f 7F1 transition, I(0-0)/I(0-1), and that of the 5D0 f 7F2 to 5D0 f 7F1 transition, I(0-2)/I(0-1). There is a linear correlation between I(0-0)/I(0-1) and I(02)/I(0-1). Therefore, the 5D0 f 7F0 transition is shown to be caused by the mixture of the 7F2 state into the 7F0 state by the J-mixing effect in the present glasses. The ligand field strength around Eu3+ and the inhomogeneous line width of Eu3+ fluorescence can be estimated from the spectral profiles of 5D0 f 7F0 transition because both 5D0 and 7F states of transition are nondegenerated J ) 0 levels. The 0 peak position and fwhm of the emission line of the 5D0 f 7F0 transition represent the ligand field strength and inhomogeneous line width, respectively. Changes in the peak positions of the 5D f 7F transition with Cl/(F + Cl) are shown in Figure 6. 0 0 The ligand field strength around Eu3+ decreases with increasing Cl/(F+Cl). The Cl- ions are weak ligands as compared with

11196 J. Phys. Chem., Vol. 100, No. 26, 1996

Takahashi et al. line width of the fluorescence suggests that the present chlorofluorozirconate glass system has a potentiality to the rareearth-doped laser materials because the stimulated emission cross section is inversely proportional to the effective line width of the fluorescence line shape.8 Conclusions

Figure 5. Relationship between the normalized integrated intensity of 5D0 f 7F0, I(0-0)/I(0-1), and that of the 5D0 f 7F2 transition, I(02)/I(0-1). The dotted line is a guide for the eye.

The local environments around Eu3+ in the ZrF4-BaF2BaCl2 glass system doped with Eu3+ have been examined by means of 151Eu-Mo¨ssbauer and Eu3+-fluorescence spectroscopies. The glass composition employed was 58ZrF4‚(39-x)BaF2‚ xBaCl2‚3EuF3 (x ) 0, 5, 10, 15, 19, 24, and 29). It was found from 151Eu-Mo¨ssbauer measurement that Eu in the present chlorofluorozirconate glasses exists only as the +3 oxidation state and that the isomer shifts of 151Eu3+ decrease with increasing Cl- concentration. Eu3+-fluorescence measurements revealed that the ligand field strength around Eu3+ and the inhomogeneous line width of Eu3+ fluorescence decrease, and the intensity of the hypersensitive 5D0 f 7F2 transition increases with increasing Cl- amount. These experimental results conclude that Cl- substituted for F- are present at neighbors of Eu3+. This conclusion supports the previous MD simulation and Zr K-EXFAS results.21,22 Acknowledgment. This work was supported by a Grantin-Aid for Scientific Research (No. 03650626) from the Ministry of Education, Science and Culture. References and Notes

Figure 6. Variation of the peak position of the 5D0 f 7F0 transition, E(0-0), with substituted Cl- amount, Cl/(F + Cl).

Figure 7. Variation of full width at half-maximum of the 5D0 f 7F0 transition, fwhm(0-0), with substituted Cl- amount, Cl/(F + Cl).

F- ions. This also indicates that Cl- exists around Eu3+. Figure 7 shows the fwhm’s of the 5D0 f 7F0 transition as a function of Cl/(F + Cl). It is particularly interesting that the fwhm of the emission peak of 5D0 f 7F0 transition becomes narrower with increasing Cl/(F + Cl). A decrease in the inhomogeneous

(1) Takahashi, M.; Kanno, R.; Kawamoto, Y.; Tanabe, S.; Hirao, K. Mat. Res. Bull. 1993, 168, 557. (2) Chen, Y.; Meichenin, D.; Auzel, F. J. Phys.: Condens. Matter 1995, 7, 3363. (3) Briertly, M. C.; Millar, C. A. Electron. Lett. 1988, 24, 438. (4) Pollack, S. A.; Robinson, M. Electron. Lett. 1988, 24, 320. (5) Ohishi, Y.; Kanamori, T.; Shimizu, M.; Yamada, M.; Terunuma, Y.; Temmyo, J.; Wada, M.; Sudo, S. IECE Trans. Commun. 1994, E77B(4), 421. (6) France, P. W.; Carter S. F.; Day, C. R.; Moore M. W. In Fluoride Glasses; Comyns, A. E., Ed.; John Wiley & Sons: Chichester, 1989; Chapter 5. (7) Poulain, M. In Fluoride Glasses; Comyns, A. E., Ed.; John Wiley & Sons: Chichester, 1989; Chapter 2. (8) Quimby, R. S. In Fluoride Glass Fiber Optics; Aggarwal, I. D., Lu, G., Eds.; Academic Press: San Diego, 1991; Chapter 9. (9) Wasylak, J.; Samek, L. J. Non-Cryst. Solids 1991, 129, 137. (10) Marcus, M. A.; Polman, A. J. Non-Cryst. Solids 1991, 136, 260. (11) Almeida, R. M.; de Barros, M. I.; Gonc¸ alves, M. C. J. Non-Cryst. Solids 1994, 168, 144. (12) Adam, J. L.; Ponc¸ on, V.; Lucas, J.; Boulon, G. J. Non-Cryst. Solids 1987, 91, 191. (13) Brawer, S. A.; Weber, M. J. J. Chem. Phys. 1981, 75, 3522. (14) Brawer, S. A.; Weber, M. J. Phys. ReV. Lett. 1980, 45, 460. (15) Cormeir, G.; Capobianco, J. A.; Monteil, A. J. Non-Cryst. Solids 1994, 168, 115. (16) Capobianco, J. A.; Proux, P. P.; Bettinelli, M.; Negrisolo, F. Phys. ReV. B 1993, 42 (10), 5936. (17) Cormeir, G.; Capobianco, J. A.; Morrison, C. A.; Monteil, A. Phys. ReV. B 1993, 48 (22), 16290. (18) Soga. K.; Uo, M.; Makishima, A.; Inoue, S. J. Am. Ceram. Soc. 1995, 78(1), 129. (19) Tanabe, S.; Takahara, K.; Takahashi, M.; Kawamoto, Y. J. Opt. Soc. Am. B 1995, 22 (5), 786. (20) Kawamoto, Y.; Kanno, R.; Yokota, R.; Takahashi, M.; Tanabe, S.; Hirao, K., J. Solid State Chem. 1993, 103, 334. (21) Takahashi, M.; Yamamoto, R.; Kanno, R.; Kawamoto, Y. J. Phys.: Condens. Matter 1995, 7, 4583. (22) Takahashi, M.; Kanno, R.; Kawamoto, Y. J. Phys.: Condens. Matter 1995, 7, 7797. (23) Long, G. J. In Mo¨ ssbauer Spectroscopy Applied to Inorganic Chemistry; Long, G. J., Ed.; Plenum Press: New York, 1984; Vol. 1, p 7. (24) Taragin, M. F.; Eisenstein, J. C. J. Non-Cryst. Solids 1973, 11, 395.

Europium Ion in the Chlorofluorozirconate Glass System (25) Kawamoto, Y. In Proceedings of Japanese-French Seminar on Inorganic Fluorine Chemistry; JSPS 155th Committee on Fluorine Chemistry, Kyoto, 1995; p 39. (26) Todoroki, S.; Hirao, K.; Soga, N. J. Appl. Phys. 1992, 72 (12), 5853. (27) Elyamani, A.; Poulain, M.; Saggese, S. J.; Sigel, G. H. Jr. J. NonCryst. Solids 1982, 51, 187. (28) Wells, A. F. In Structural Inorganic Chemistry; 5th ed.; Wells, A. F., Ed.; Clarendon Press: Oxford, 1984; p 420. (29) Jørgensen, C. K.; Judd, B. R. Mol. Phys. 1964, 8, 281. (30) Takahashi, M.; Kanno, R.; Kawamoto, Y.; Tanabe, S.; Hirao, K. J. Non-Cryst. Solids 1994, 137. (31) Peacock, R. D. In Structure and Bonding; Dunits, J. D.; Hemmerich, P.; Holm, R. H.; Ibers, J. A.; Jørgensen, C. K.; Neilands, J. B.; Reinen, D.; Williams, R. J. P., Eds.; Springer-Verlag: Berlin, 1975; Vol. 22, p 116.

J. Phys. Chem., Vol. 100, No. 26, 1996 11197 (32) Judd, B. R. Phys. ReV. 1962, 127, 750. (33) Ofelt, G. S. J. Chem. Phys. 1962, 37, 511. (34) Gerth, G.; Kienle, P.; Luchner, K. Phys. Lett. 1968, 27A (8), 557. (35) Downer, M. C.; Burdick, G. W.; Sardar, D. K. J. Chem. Phys. 1988, 89, 1787. (36) Burdick, G. W.; Downer, M. C.; Sardar, D. K. J. Chem. Phys. 1989, 91, 1511. (37) Nishimura, G.; Kushida, T. Phys. ReV. B 1988, 37, 9075. (38) Tanaka, M.; Nishihara, G.; Kushida, T. Phys. ReV B 1994, 49 (24), 16917.

JP9603492