Growth Optimization of Li2B4O7 Crystals and Their Characterization

Oct 26, 2007 - ABSTRACT: Lithium tetraborate (Li2B4O7) crystals were grown by the Czochralski technique. The thermal gradient is the key factor to gro...
0 downloads 0 Views 425KB Size
CRYSTAL GROWTH & DESIGN

Growth Optimization of Li2B4O7 Crystals and Their Characterization S. Kar and K. S. Bartwal*,‡ Laser Materials DeVelopment & DeVices DiVision Raja Ramanna Centre for AdVanced Technology, Indore - 452013, India

2007 VOL. 7, NO. 12 2522–2525

ReceiVed April 10, 2007; ReVised Manuscript ReceiVed September 20, 2007

ABSTRACT: Lithium tetraborate (Li2B4O7) crystals were grown by the Czochralski technique. The thermal gradient is the key factor to grow crack-free crystals. We have optimized the vertical thermal gradient for the growth of crack-free and optical quality crystals. This crystals belongs to the tetragonal system with lattice parameters a ) b ) 9.479 Å, and c ) 10.286 Å. The grown crystals were 20–30 mm in diameter and 20–25 mm in length. Transmission spectra of as-grown crystals show good optical transparency. Powder X-ray diffraction (XRD) was done to confirm the phase of the grown crystals. Crystalline perfection of the as-grown crystals was investigated using the high resolution X-ray diffraction (HRXRD) technique.

1. Introduction The Li2O-B2O3 phase diagram contains eight stoichiometric compounds.1–3 Only five of these are stable at room temperature, namely, LiB3O5, Li2B4O7, LiBO2, Li6B4O9, and Li3BO3. The general formula of these compounds are (Li2O)x(B2O3)1-x. Single crystals of the Li2O-B2O3 system are of considerable interest, not only for practical uses but also for fundamental investigations. Lithium tetraborate (Li2B4O7, LTB) is an important material for generation of 4th and 5th harmonics of Nd: YAG laser and frequency conversion devices for high power UV solid-state lasers.4 LTB crystals belong to the tetragonal system having 4-mm point group symmetry with lattice parameters a ) b ) 9.479 Å, c ) 10.286 Å. High electromechanical coupling coefficient factor k2 and low temperature coefficient of frequency have made it a very attractive material for a surface acoustic wave substrate.5 Microwave devices using surface acoustic waves are in common use for infrared filters for color television and signal processing elements. Transmission range of this crystal is 170–3300 nm. The damage threshold value of ∼40 GW/cm2 for LTB crystal is the largest among the other nonlinear optical (NLO) borate crystals. Lithium triborate and cesium triborate both have vanishing effective nonlinearity in the limit of 90° noncritical phase matching, but effective nonlinearity is highest for the lithium tetraborate crystal at 90° phase matching. An UV solid-state laser system which combines a high power infrared laser with a NLO crystal has been strongly desired for various applications such as photolithography, material processing, and medical applications. In general, UV light is easily absorbed in the crystal because of its large photon energy, and this degrades the crystal like the color center in fused silica. Degradation is more serious when defects and impurities exist in the crystal. Crystals for UV applications therefore require higher optical quality than those for other applications. Li2B4O7 single crystals can be used as bulk acoustic wave-based devices6 and pyroelectric temperature sensors.7 They can also be used as thermoluminescent dosimetry of X-ray, gamma, and neutron radiation.8–10 LTB contains Li and B, which possess a large neutron capture cross-section. The * Corresponding author. Tel.: +91-731-2488656; fax: +91-731-2488650; e-mail: [email protected]. ‡ Present address: Energy Materials Research Centre, Korea Research Institute of Chemical Technology, KRICT, P.O. Box 107, Yuseong, Daejeon-305-600, South Korea.

effective atomic number for lithium tetra borate is 7.23, which agrees closely with the value of 7.22 for soft tissue composition C5H14O18N.11 It is important to point out that the NLO properties of these materials are very sensitive to the cationic substitution. The tetraborate materials doped by rare earths ions may be promising quantum electronic materials.12,13 This paper describes the growth optimization of crack-free transparent lithium tetraborate crystals and their characterizations using powder X-ray diffraction (XRD) and high resolution X-ray diffraction (HRXRD).

2. Experimental Details High purity oxides in the ratio 67.9 mol % of B2O3 and 32.1 mol % Li2CO3 were taken for the preparation of charge. The excess amount of B2O3 over stoichiometric composition was because of its high vapor pressure.14 Special care has to be taken during weighing of charge because the chemicals are hygroscopic in nature. Charge preparation for lithium tetraborate is also a critical process as severe frothing occurs during preparation. To avoid frothing, the mixture was first kept at 350 °C for 24 h in the Pt crucible, and then it was melted directly at 917 °C. The growth was carried out in a Pt crucible of 50 mm diameter and 50 mm height. The [110] oriented seed crystal of dimension 2 × 2 × 10 mm3 was used for seeding. This [110] direction in LBO crystal is most favorable for the generation of 4th and 5th harmonics of the Nd:YAG laser. The seed was properly tied in a ceramic rod with platinum wire and slowly dipped into the melt to avoid thermal shock. The small portion of the seed was then slightly pulled out manually to get an inclusion-free crystal. Initial rotation rate of the seed rod was 8 rpm and was reduced to 6 rpm afterward. A small rotation rate reduces thermal convection and hence reduces temperature fluctuation. The pulling rate for this crystal is very much less than ∼0.3 mm/h. A resistive heating furnace with a suitable axial temperature gradient was used for growth. Eurotherm temperature controller was used to control the temperature of the furnace. Crystals were grown with a different temperature gradient near the melt and the solid interface. The lower the temperature gradient near the interface, the larger the crystal faceting. Initially, the axial temperature gradient was kept 16 °C/cm above the melt level, whereas below the melt level it was 10 °C/cm. In these conditions, we got crack-free crystals with facets and small thickness (Figure 1a). In the second run, the gradient of the furnace was slightly modified, and good quality crack-free crystals without any facet were obtained (Figure 1b,c). In the second run, the crucible was lifted upward to change the gradient above the melt level 18 °C/cm. Elements prepared from these crystals for characterization are shown in Figure 1d. Axial gradient of the furnace is shown in Figure 2. Major difficulties encountered during the growth of lithium tetraborate crystals are cracks, formation of core, defect region consisting largely of opaque inclusions, and twinning.15,16 Crystals crack during post

10.1021/cg070353a CCC: $37.00  2007 American Chemical Society Published on Web 10/26/2007

Growth Optimization of Li2B4O7 Crystals

Figure 1. Photograph of grown crystals and elements. (a) Crystal grown in normal gradient conditions, (b) and (c) crystal grown with modified gradient, (d) cut and polished elements.

Crystal Growth & Design, Vol. 7, No. 12, 2007 2523

Figure 3. XRD pattern of lithium tetraborate.

Figure 4. High resolution X-ray diffraction recorded for (004) planes of Li2B4O7 crystal (error bar 5%). Figure 2. Temperature gradient of the furnace (error bar 2%). growth cooling due to anisotropy of thermal expansion coefficient. Grown crystals were cooled slowly in 24 h to room temperature to avoid the thermal shock. Crystalline powder obtained by crushing a piece of a grown crystal was tested by powder XRD for phase identification. A multicrystal X-ray diffractometer (MCD) was used for HRXRD experiments. A finefocus X-ray tube (Philips, 0.4 × 8 mm, Mo 2 kW) is energized by a well-stabilized Philips X-ray generator (Philips PW1743). The white X-ray beam first passes a long collimator (fitted with a special set of slits at one end) and is then diffracted from two plane (111) Si monochromator crystals of Bonse-Hart type. A well-collimated Mo KR1 beam is isolated and further diffracted from a third plane monochromator crystal of dislocation free silicon in the dispersive symmetrical Bragg geometry (+, -, -) configuration. This geometry gives the rocking curve for that particular plane of the crystal. Therefore, the crystalline perfection can be estimated by the full-width at halfmaximum (fwhm) of the curve. The crystal wafers were cut in the perpendicular direction of the growth axis, and the plates of dimensions 10 × 10 × 1 mm3 were polished with 0.03-µm sized alumina powder and used for transmission study. A crystal plate of dimension 10.8 × 5.5 × 2.16 mm3 was lapped properly for activation energy measurements. Platinum paste was used for electroding purpose. DC conductivity of this sample was measured as a function of temperature by a two-probe method in the temperature range from room temperature to 400 °C.

3. Results and Discussion Crystals of various sizes were grown in different temperature gradients. The maximum size we obtained was 30 mm in

diameter and 25 mm in length. It is difficult to grow LTB crystals at the stoichiometric ratio of Li2CO3/B2O3 (i.e., 1:2). A slight excess of B2O3 is helpful for growth of LTB crystal. The coexistence of BO4 tetrahedral and BO3 triangles in the framework structure of Li2B4O7 makes the crystal prone to cracking.17 To avoid cracking, the proper temperature gradient of the furnace and post growth cooling rate are very essential and critical. Pulling and rotation rates are also very crucial parameters for the growth of this crystal. Slow pulling and low rotation rate are required for successful growth. The powder XRD pattern from the grown crystals matched well with the known single phase for LTB. Figure 3 shows the powder XRD pattern of the grown crystal. It was observed that XRD peaks match well with the known single phase for LTB. Also the sharp peaks in the pattern are indicative of the fact that there is no nonstoichiometry in the grown crystals. Crystalline perfection of the grown crystals was examined by diffraction curves (DCs) or rocking curves. Figure 4 shows the HRXRD for the crystal plate along the 004 plane. A wellcollimated Mo KR1 beam diffracted from a third plane monochromator crystal of Si (111) in the dispersive symmetrical Bragg geometry (+, -, -) configuration is used for this experiment. This arrangement particularly improves the spectral purity of the KR1 beam. The specimen crystal is aligned in the (+, -, -, +) symmetrical Bragg geometry. The rotation given by the main turntable to the specimen around a vertical axis (which changes the glancing angle θ) is produced with the help of a micrometer, which moves a long radial arm of the diffractometer. The micrometer is driven by a stepper motor,

2524 Crystal Growth & Design, Vol. 7, No. 12, 2007

Kar and Bartwal

Figure 6. Variation of current with temperature. Figure 5. Transmission spectra of as grown Li2B4O7 crystal.

which is controlled by a microprocessor control unit (Philips, PW 1710). The specimen can be rotated in steps of 0.5′′. The diffracted X-ray intensity is measured by a scintillation counter. Its output is measured by the counting system incorporated in the microprocessor control unit. The detector is mounted with its axis along a radial arm of the turntable. The measured tilt angle is 66 arc sec. The deconvolution of the DC curve in Figure 4 indicates that the experimental DC consists of two curves (dotted line) with an angular separation of 8–10 arc sec. The fwhm recorded of the main peak is found to be 38 arc sec, while the peak intensity is 130 c/s. For the secondary peak, the fwhm is about 28 arc sec. This indicates the presence of a small angle grain boundary in the crystals with a tilt orientation of about 8–10 arc sec. The overall fwhm can be calculated as 66 arc sec. The values in arc sec are generally accepted as low when we talk about the crystalline perfection. Generally the normal crystal shows this value in several min of arc. Therefore, this value for fwhm suggests that the overall crystalline perfection of the grown crystals is quite good for optical applications such as frequency conversion. Transmission spectra were taken in a UV–vis PC spectrophotometer. The transmission of the crystal was observed at more than 80%. Figure 5 shows the representative transmission spectra of the grown crystal. No variation in transmission spectra was recorded when taken at various points in a crystal plate and also crystal plates taken from different heights of the grown boule. This proves that there is no axial or radial inhomogeneity in the crystal. We have taken the transmission spectra only up to 190 nm, due to limitation of the spectrophotometer used in the present study. However, this crystal is transparent up to about 165 nm.18 The conductivity was calculated as σ ) iL/VA, where i is the current, V is test voltage, L is the thickness, and A is the area of the sample. Below 250 °C, no detectable current was measured, and after that it increases exponentially. Figure 6 shows the current versus temperature plot for LTB crystal plate. The variation of logarithm of conductivity (lσ) with 1000/T K is shown in Figure 7. Activation energy was calculated from the equation σ ) σ0 exp(-Eg/KT). The measured activation energy was found to be 7.6 eV. This is very near the reported band gap energy of the crystal 7.5 eV. It is indicative of the absence of any virtual energy level (trap level) below the conduction band, and hence the crystal is grown with very negligible defects. It is important to point out that the NLO properties of these materials are very sensitive to the defects or

Figure 7. Variation of ln σ vs 1000/T.

cationic substitution. This will open up the renewed interest in these materials for further development of their technology.

4. Conclusions Good quality transparent and crack-free LTB crystals were grown successfully under different temperature gradients. Various growth parameters were optimized for better quality crystals. The tilt angle of the grown crystal was measured by HRXRD. The measured tilt angle was 66 arc sec. Through transmission spectra studies, it was observed that grown crystals are transparent to UV region. Activation energy of as grown crystal was measured by a two-probe method and found nearly same as the energy band gap of the crystal. These studies indicate that the grown crystals are quite good in crystalline perfection and have negligible crystal defects. Acknowledgment. We acknowledge Dr. G Bhagavannarayana, NPL Delhi, for helping with HRXRD experiments on our samples.

References (1) Sastry, B. S. R.; Hummel, F. A. J. Am. Ceram. Soc. 1958, 41, 7–17. (2) Sastry, B. S. R.; Hummel, F. A. J. Am. Ceram. Soc. 1959, 42, 216– 218. (3) Aver’yanov, V. I.; Kalmykov, A. E. Glass Phys. Chem. 1990, 16, 492. (4) Sugawara, T.; Komatsu, R.; Uda, S. J. Crystal Growth 1998, 193, 364–369. (5) Whatmore, R. W.; Shorrocks, N. M.; Hara, C. O.; Anger, F. W.; Young, I. W. Electron. Lett. 1981, 17, 11–13.

Growth Optimization of Li2B4O7 Crystals (6) Shestopalov, K. V.; Nefedov, V. A.; Zadneprovsky, B. I. Proceedings of the 1994 IEEE International Frequency Control Symposium, Boston; Institute of Electrical and Electronics Engineers: Piscataway, NJ, 1994. (7) Ono, M.; Sakai, M.; Fujiwara, Y.; and Wakatsuki, N. Proceedings of the 1997 IEEE Ultrasonics Symposim 2; Institute of Electrical and Electronics Engineers: Piscataway, NJ, 1997; pp 1047–1051. (8) Furetta, C.; Weng, P. S. Operation Thermoluminescent Dosimetry; World Scientific: London, 1998. (9) Mahesh, K.; Weng, P. S. and Furetta, C. Thermoluminescence in Solids and its Applications to New Technology; Ashford: Kent, UK, 1989. (10) Dolzhenkova, E. F.; Baumer, V. N.; Tolmachev, A. V.; Hunda, B. M.; Puga, P. P. Sixth International Conference on Inorganic Scintillators and Their Use in Scientific and Industrial Applications, Chamonix, France, September 16-21, 2001; Book of Abstracts, p 210. (11) Jayachandran, C. A. Phys. Med. Biol. 1971, 16, 617–623.

Crystal Growth & Design, Vol. 7, No. 12, 2007 2525 (12) Adamiv, V. T.; Burak, Ya.V.; Kityk, I. V.; Kasperczyk, J.; Smok, R.; Czerwihki, M. Opt. Mater. 1997, 8, 207–213. (13) Majchrowski, A.; Kityk, I. V.; Mandowska, E.; Mandowski, A.; Ebothe, J.; Lukasiewicz, T. J. Appl. Phys. 2006, 100, 53101–53105. (14) Podgorska, D.; Kaczamarek, S. M.; Drozdowski, W.; Berkowski, M.; Worsztynowicz, A. Acta Phys. Polonica 2005, A107, 507–512. (15) Robertson, D. S.; Young, I. M. J. Mater. Sci. 1982, 17, 1729–1738. (16) Burak, Y. V. J. Cryst. Growth 1998, 186, 302–304. (17) Jiayue, X.; Shiji, F.; Baoliang, L. J. Cryst. Growth 2004, 264, 260– 265. (18) Komatsu, R.; Sugawara, T.; Sassa, K.; Sarukura, N.; Liu, Z.; Izumida, S.; Segwa, Y.; Yamanouchi, K. Appl. Phys. Lett. 1997, 70 (26), 3492–3494.

CG070353A