Read doi:10.1016/j.jnoncrysol.2006.03.089 text version

Journal of Non-Crystalline Solids 352 (2006) 3577­3581

Thermal properties of barium titanium borate glasses measured by thermal lens technique

C. Jacinto *, C.A.C. Feitosa, V.R. Mastelaro, T. Catunda

´ Instituto de Fisica de Sao Carlos, Universidade de Sao Paulo ­ USP, CEP 13560-970, Sao Carlos, SP, Brazil ~ ~ ~ Available online 2 August 2006

Abstract In this paper, thermal lens spectrometry was used to determine the thermal properties of barium titanium borate glasses as a function of the amount of TiO2 in the ratios 8/9 and 1/1 of BaO­B2O3. The thermal lens technique was shown to be sensitive to the variation of TiO2 and the ratio of BaO/B2O3, through thermal diffusivity and conductivity, with different behavior for the ratios 1/1 and 8/9. An increase in thermal diffusivity and conductivity as a function of TiO2 was observed, and this behavior was ascribed to the fact that titanium acts as a glassy-former, being more active in the ratio 8/9. Ó 2006 Elsevier B.V. All rights reserved.

PACS: 61.43.Fs; 65.60.+a; 66.30.Xj; 67.80.Gb Keywords: Absorption; Borates; Thermal properties; Thermal conductivity

1. Introduction Glassy systems based on barium titanium borate have several interesting characteristics that make them promissory candidates for several applications. For example, in the crystalline b-phase it presents a large second harmonic generation coefficient, a wide range of transparence, a broad phase-matched region and a high damage threshold. In thin film form, the b-BaB2O4 (b-BBO) phase can be applied in optical devices such as frequency converters, waveguides and switches, and so on [1­6]. These several applications have led us to the study of thermal properties in BaO­B2O3­TiO2 glassy matrices, and thus to the determination of thermal parameters such as thermal diffusivity (D) and conductivity (K). The BaO­B2O3­TiO2 system is of interest for the manufacture of ferroelectric barium titanate glass­ceramic materials [7]. Although the synthesis and properties of the BaO­B2O3­TiO2 system have been widely explored, little

information is found in previous literature concerning the structural role of TiO2 on the BaO­B2O3 system. Under these circumstances, a study of this issue is needed. In this paper, the thermal lens (TL) technique was used to determine the thermal properties of barium titanium borate glasses as a function of the amount of TiO2 in the ratios 8/ 9 and 1/1 of BaO­B2O3. The effect of the change from 1/1 to 8/9 in the BaO/B2O3 ratio is discussed, as well as the effect of the titanium. 2. Experimental procedures 2.1. Thermal lens measurements The thermal lens (TL) effect is caused by the heat generation via non-radiative decay processes after the laser energy has been absorbed by the sample. In this situation, a transverse temperature gradient is established, and owing to the temperature coefficient of optical path length change, ds/dT, a lens-like optical element is created: the so-called TL. The propagation of a probe laser beam through this TL results in a variation of its on-axis intensity, I(t), which can be calculated using diffraction integral theory [8]. In the


Corresponding author. Tel.: +55 16 33738093; fax: +55 16 3371381. E-mail address: [email protected] (C. Jacinto).

0022-3093/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jnoncrysol.2006.03.089


M1 chopper L1 M2 M3 Sample

C. Jacinto et al. / Journal of Non-Crystalline Solids 352 (2006) 3577­3581

Excitation Laser



Laser P1

pinhole Digital Scope signal trigger


Fig. 1. Schematic diagram of the mode-mismatched thermal lens experimental apparatus, where M's are mirrors, P's are photodetectors and L's are convergent lenses. The angle between the excitation and probe beams is indicated by a.

ing lens, L1 (f = 20 cm), and the sample was put at its focal plane. A chopper controlled exposure of the sample to the excitation beam. The probe beam was focused by lens L2 (f = 20 cm) arranged in such a way that the sample was positioned near its confocal position (Z1 $ 1.73 · Zcp). The angle a < 1.5° was used to deviate the probe beam to the TL detection plane (photodiode P2) positioned in the far field. The probe beam was aligned to pass through the TL, induced by the excitation beam, as depicted in Fig. 1, to maximize the signal. In the transient experiment, an oscilloscope was used to record the TL signal buildup, which lasted 240 ms for the sample, and the experimental arrangement of the current work. More details about the experimental configuration can be found elsewhere [9,10]. 2.2. Sample preparation The composition of the glasses under study was xBaO­ yB2O3­zTiO2 with x, y, and z given in mol%. Two sets of samples were prepared: one with x/y ratio = 1/1 (where z = 8, 16, and 33.33) and the other with x/y ratio = 8/9 (with z = 4, 8, and 15). The glasses were prepared by mixing appropriate amounts of reagent grade BaCO3, TiO2 and B2O3, and melting them in a platinum crucible in an electrically heated furnace for 3 h at 1100 °C. The melt was quenched on a steel plate and annealed in a furnace at 450 °C for 1 h. This process yielded 4-mm thick, 20 g glass plates. For our measurements, the samples thickness was $2 mm. The purity of the raw materials was: TiO2 (Merck, 99%), B2O3 (Suprapur, 99.8%), and BaCO3 (Riedel-de Haen, 99%). More details about the manufacture of these samples can be found elsewhere [11]. 3. Results Fig. 2 shows the visible transmission spectra in the range of 300­600 nm for samples with 4 and 15 mol% of TiO2 for

cw excitation regime, an analytical expression can be obtained for the probe beam intensity, I(t) [9,10] ( " #)2 h 2mV IðtÞ ¼ Ið0Þ 1 À a tan ; 2 ½ð1 þ 2mÞ2 þ V 2 tc =2t þ 1 þ 2m þ V 2


where h¼À P abs ds ; Kkp dT m¼ wp we 2 ; V ¼ Z1 Z cp ð2Þ

ðwhen Z cp ( Z 2 Þ:

Transmitance (%)

Here, Zcp is the confocal distance of the probe beam, Z1 is the distance between the probe beam waist and the sample, Z2 is the distance between the sample and the photodiode, wp is the probe beam radius at the sample, we is the excitation laser beam radius at the sample, I(0) = I(t) when the transient time t or h is zero, h is approximately theffi phase pffiffi difference of the probe beam at r = 0 and r ¼ 2we induced by TL, Pabs = Pexc[1 À exp(ÀAL)] % PexcAL (for AL ( 1) is the absorbed pump power (in which Pexc is the pump beam power, L is the sample thickness and A is the optical absorption coefficient of the sample at the excitation beam wavelength), ds/dT is the temperature coefficient of the optical path length, kp is the probe beam wavelength, and K is the thermal conductivity. The temporal evolution of the TL signal depends on the characteristic TL signal time, tc, which is related to the thermal diffusivity (D) by D ¼ w2 =4tc e and this to the thermal conductivity by: K ¼ qcD; ð4Þ where q is the density and c is the specific heat. The TL experiment was performed using the timeresolved mode mismatched configuration [9,10] shown in Fig. 1, in which an Ar+ laser at 488 nm was used as the excitation beam and a He­Ne laser at 632.8 nm was used as the probe beam. The excitation beam was focused by a convergð3Þ





4 mol% - 8/9 8 mol% - 1/1 15 mol% - 8/9 16 mol% - 1/1


0 300 360 420 480 540 600


Fig. 2. Visible transmission spectra in the range of 300­600 nm, for samples of the sets 1/1 (8 and 16 mol% of TiO2) and 8/9 (4 and 15 mol% of TiO2).

C. Jacinto et al. / Journal of Non-Crystalline Solids 352 (2006) 3577­3581


set 8/9 and with 8 and 16 mol% of TiO2 for set 1/1. These measurements were performed with a Hitachi U-2001 spectrophotometer. Fig. 3 shows a typical TL transient signal for the sample with 33 mol% of TiO2 of the set 1/1. The excitation power used in this transient was 1.3 W. The solid curve represents the fit curve obtained from Eq. (1). As an example, by use of the obtained tc value in relation to tc ¼ w2 =4D, jointly e with the we value used, it is possible to determine D = (3.00 ± 0.09) · 10À3 cm2/s. In order to achieve an accurate determination of D, for each sample, tc and h were obtained from several transients at various excitation powers, Pexc. Measurements in several parts of the samples were also performed in order to verify the sample homogeneity. By means of this procedure, the D values are determined with better accuracy. A linear dependence of h versus Pexc was also observed for each sample in all measurements, indicating good determination of these parameters and absence of saturation effects [12]. Fig. 4 presents the thermal diffusivity as a function of the amount TiO2 obtained for the two sets of samples. The error bars were calculated using the error in we measurements and those in the tc determination by means of fitting with Eq. (1). To obtain the thermal conductivity for the two sets of samples, it is necessary to know the densities and the specific heat of the samples. The densities were measured by means of the Archimedes method in CCl4 (±0.01 g cmÀ3). The specific heats were calculated using the same procedure as that adopted by Bruce [13], who analyzed twelve kinds of fluoride glasses and concluded that, at room temperature, the molar specific heat was almost constant and nearly equal to 21.3 J KÀ1 molÀ1 (86% of the Dulong­Petit value). This expression has presented values in good agreement with those obtained experimentally [14]. For example, using this value for ZBLAN glass, qc = 2.7 J/K cm3 was obtained, which was in a good agreement with the experi-



D (10-3 cm2/s )



(BaO-B2O3) - 8/9


(BaO-B2O3) - 1/1

1.2 7 14 21 28 35

TiO2 Concetration (mole %)

Fig. 4. Thermal diffusivity versus TiO2 concentration. The closed and open circles correspond to the BaO­B2O3 compositions of 1/1 and 8/9, respectively.

mental value qc = 2.8 J/K cm3. The measured value for the samples studied in this work was c = 0.7033 J/g K, which is also in good agreement with that found in previous literature for BaOB2O3 [15]. Table 1 reports the calculated values for K and also those for D, q, and h/PexcL. In Fig. 5, the values of K as a function of the TiO2 concentration are depicted. 4. Discussion As can be seen in Fig. 2, the studied samples present a very good transparency in the visible range, and it is slightly diminished when the amount of TiO2 increases as observed in beta barium borate [1,16]. For some purposes, for instance, for the crystallization of large areas in little irradiation time [2,17­19], it is desirable to have a material with low thermal diffusivity and conductivity. In this sense, Gan [18] and Sato et al. [19] presented interesting results concerning the crystallization process induced by laser radiation in amorphous chalcogenide and tellurite glasses, respectively. The samples in this study present D and K smaller than those of crystals such as YAG (D = 48 · 10À3 cm2/s and K = 13 W/m K [20]) and YLF (D = 19 · 10À3 cm2/s and K ¼ 5:8kc and 7.2?c W/m K) [12,21], and comparable to those of glasses such as fluoride (ZBLAN ­ D = 2.7 · 10À3 cm2/s and K = 0.77 W/m K) [22], chalcogenide (GLS ­ D = 2.7 · 10À3 cm2/s and K = 0.59 W/m K) [23], and aluminosilicate (LSCAS ­ D = 5.5 · 10À3 cm2/s and K = 1.5 W/m K) [24,25]. Concerning the increase of D and K with TiO2, Maia et al. [1] have shown that the increase of the TiO2 concentration reduces the band gap, thus probably increasing the conductive nature of thin films. In Refs. [24,25], the authors observed a decrease of D and K in LSCAS glass as a function of the Nd2O3 and





= -(112.8 ± 0.5)x10-3 rad


tc = (1.08 ± 0.01) ms







time (ms)

Fig. 3. Time-resolved experimental data and their best fit curve for the sample with 33 mol% of TiO2. The excitation power was 1.3 W at kexc = 488 nm.


C. Jacinto et al. / Journal of Non-Crystalline Solids 352 (2006) 3577­3581

Table 1 Thermal properties (D = thermal diffusivity, K = thermal conductivity) of the BaO­B2O3 system in the BaO/B2O3 ratios of 1/1 and 8/9 as a function of the amount of TiO2 Sample (Composition in mol%) 4%TiO2:45.18%BaO­50.82%B2O3 (8/9) 8%TiO2:43.3%BaO­48.7%B2O3 (8/9) 15%TiO2:40%BaO­45%B2O3 (8/9) 8%TiO2:46%BaO­46%B2O3 (1/1) 16%TiO2:42%BaO­42%B2O3 (1/1) 33.33%TiO2:33.33%BaO­33.33%B2O3 (1/1) c = 0.7033 J/g K. q (g/cm3) 3.97 ± 0.03 4.01 ± 0.06 3.95 ± 0.05 4.76 ± 0.06 4.73 ± 0.07 4.71 ± 0.08 h/L Æ Pexc (10À2 cmÀ1 WÀ1) À(1.41 ± 0.04) À(2.8 ± 0.1 ) À(8.2 ± 0.4) À(5.7 ± 0.3) À(9.0 ± 0.3) À(45 ± 2) D (10À3 cm2/s) 1.40 ± 0.06 1.74 ± 0.09 2.70 ± 0.09 2.10 ± 0.09 2.22 ± 0.09 3.00 ± 0.09 K (10À3 W/K cm) 3.9 ± 0.2 4.9 ± 0.3 7.5 ± 0.3 7.0 ± 0.3 7.4 ± 0.3 9.9 ± 0.4


ratio of the atoms that compose each of the glassy systems. The closer the structure of the glassy lattice, the easier it will be to conduct heat through the material. 5. Conclusions In summary, in this work it was studied the role of TiO2 on the BaO­B2O3 composition in BaO/B2O3 ratios of 1/1 and 8/9. The results show that titanium acts as glass-former in this system, increasing the thermal diffusivity and conductivity, and that it is more active in the BaO/B2O3 ratio 8/9 than 1/1. These results could be very important to the understanding and the control of the crystallization process in this glassy system. Acknowledgements The authors are thankful to CNPq and FAPESP for the financial support of this research. References

[1] L.J.Q. Maia, M.I.B. Bernardi, C.A.C. Feitosa, V.R. Mastelaro, A.R. Zanatta, A.C. Hernandes, Thin Solid Films 457 (2004) 246. [2] A.F. Maciente, V.R. Matelaro, A.L. Martinez, A.C. Hernandes, C.A.C. Carneiro, J. Non-Cryst. Solids 306 (2002) 309. [3] C.T. Chen, B.C. Wu, A.D. Jiang, G.M. You, Sci. Sinica Ser. B 28 (1985) 235. [4] D. Eimerl, L. Davi, S. Velsko, E.K. Graham, A. Zalkin, J. Appl. Phys. 62 (1987) 1968. [5] P. Becker, Adv. Mater. 10 (1998) 979. [6] M.-Q. Cai, Z. Yin, M.-S. Zhang, Appl. Phys. Lett. 83 (2003) 2805. [7] A. Bhargava, J.E. Shelby, R.L. Snyder, J. Non-Cryst. Solids 102 (1988) 136. [8] S.J. Sheldon, L.V. Knight, J.M. Thorne, Appl. Opt. 21 (1982) 1663. [9] M.L. Baesso, J. Shen, R.D. Snook, J. Appl. Phys. 75 (1994) 3732. [10] S.M. Lima, J.A. Sampaio, T. Catunda, A.C. Bento, L.C.M. Miranda, M.L. Baesso, J. Non-Cryst. Solids 273 (2000) 215. [11] C.A.C. Feitosa, V.R. Mastelaro, A.R. Zanatta, A.C. Hernandes, E.D. Zanotto, Opt. Mat. 28 (2006) 935. ´ [12] C. Jacinto, T. Catunda, D. Jaque, J. Garcia-Sole, Phys. Rev. B 72 (2005) 235111. [13] A.J. Bruce, in: R.M. Almeida (Ed.), NATO on Halide Glasses for Infrared Optica Fiberoptics (Portugal), Martinus Nijhoff, Dordrecht, 1987, pp. 149­162. [14] J. Zarzycki, Glasses and the Vitreous State, Cambridge University Press, New York, 1991, p. 349. [15] H. Bach, N. Neuroth, The Properties of Optical Glass, Springer, Berlin, 1995.

K (10-3W/cm.K)



(BaO-B2O3) - 8/9 (BaO-B2O3) - 1/1








TiO2 Concetration (mole %)

Fig. 5. Thermal conductivity versus TiO2 concentration. The closed and open circles correspond to the BaO­B2O3 compositions of 1/1 and 8/9, respectively.

Er2O3 concentrations. They attributed this behavior to the modifying character of the rare-earth ion. Mansanares et al. [26] have shown for silicate glasses doped with Fe2O3 that there are two distinct situations in which the doping controls the thermal diffusivity. When the Fe3+ ions are network formers, the sample thermal diffusivity becomes larger than that of an undoped sample, while when they are network modifiers, the thermal diffusivity decreases with increasing doping concentration. Thus, the increase of thermal diffusivity with increasing TiO2 concentration showed in Fig. 4 indicates that titanium acts as a glass-former [27,28]. Following this reasoning, the smaller variations of D and K with the amount of TiO2 in the 1/1 ratio compared to the 8/9 ratio, shown in Figs. 4 and 5, must be an indication that TiO2 acts as network former more actively in the 8/9 system than in the 1/ 1. On the other hand, Sampaio in his PhD thesis [29] studied several glassy systems and concluded that the higher value of K in LSCAS glass (aluminosilicate) than in other systems (in order of decreasing K values: aluminate, silicate, borosilicate, phosphate, fluoride, and chalcogenide ­ the borates in this study come between phosphate and chalcogenide) can be attributed to the weight/atomic radius

C. Jacinto et al. / Journal of Non-Crystalline Solids 352 (2006) 3577­3581 [16] A.L. Martinez, R. Lebullenger, C.A.C. Feitosa, A.C. Hernandes, J. Non-Cryst. Solids 351 (2005) 1372. [17] C. Mai, Suppl. Riv. Staz. Sper. Vetro XXIII (1993) 435. [18] F. Gan, J. Non-Cryst. Solids 257 (1999) 176. [19] R. Sato, Y. Benino, T. Fujiwara, T. Komatsu, J. Non-Cryst. Solids 289 (2001) 228. [20] C. Jacinto, A.A. Andrade, T. Catunda, S.M. Lima, M.L. Baesso, Appl. Phys. Lett. 86 (2005) 034104. [21] R.C. Powell, Physics of Solid-State Laser Materials, Springer, New York, 1998. [22] S.M. Lima, J.A. Sampaio, T. Catunda, R. Lebullenger, A.C. Hernandes, M.L. Baesso, A.C. Bento, F.C.G. Gandra, J. Non-Cryst. Solids 256&257 (1999) 337. [23] S.M. Lima, J.A. Sampaio, T. Catunda, A.S.S. de Camargo, L.A.O. Nunes, M.L. Baesso, D.W. Hewak, J. Non-Cryst. Solids 284 (2001) 274.


[24] J.A. Sampaio, S. Gama, M.L. Baesso, T. Catunda, J. Non-Cryst. Solids 351 (2005) 1594. [25] M.L. Baesso, A.C. Bento, A.R. Duarte, A.M. Neto, L.C.M. Miranda, J.A. Sampaio, T. Catunda, S. Gama, F.C.G. Gandra, J. Appl. Phys. 85 (1999) 8112. [26] A.M. Mansanares, M.L. Baesso, E.C. Silva, F.C.G. Gandra, H. Vargas, L.C.M. Miranda, Phys. Rev. B 40 (1989) 7912. [27] P. Pernice, S. Esposito, A. Aronne, Phys. Chem. Glasses 39 (1998) 222. [28] P. Pernice, S. Esposito, A. Aronne, V.N. Siagev, J. Non-Cryst. Solids 258 (1999) 1. [29] J.A. Sampaio, PhD thesis, University of Sao Paulo, Brazil, 2001. ~



5 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Effect of intracellular lipid accumulation in a new model of non-alcoholic fatty liver disease
FluoroLog 3 Spectrofluorometer
12 Experimental Techniques at Synchrotron Light Source Beamlines
Z-Scan Measurements of Organic Nonlinearities