Study on a multifunctional crystal LiMnBO3

Study on a multifunctional crystal LiMnBO3

Materials Research Bulletin 48 (2013) 277–280 Contents lists available at SciVerse ScienceDirect Materials Research Bulletin journal homepage: www.e...

528KB Sizes 0 Downloads 23 Views

Materials Research Bulletin 48 (2013) 277–280

Contents lists available at SciVerse ScienceDirect

Materials Research Bulletin journal homepage: www.elsevier.com/locate/matresbu

Study on a multifunctional crystal LiMnBO3 Liwei Zhao a,b, R.K. Li a,* a b

Beijing Center for Crystal Research and Development, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, PR China Graduate University of Chinese Academy of Sciences, Beijing 100049, PR China

A R T I C L E I N F O

A B S T R A C T

Article history: Received 21 May 2012 Received in revised form 9 October 2012 Accepted 11 October 2012 Available online 8 November 2012

Single crystal of high temperature phase of LiMnBO3 in the hexagonal (h) lattice has been grown by slowcooling method with LiBO2 as flux. The transmission spectrum reveals that h-LiMnBO3 is highly transparent from 680 to 3000 nm and its absorption bands in the visible and ultraviolet regions are assigned to d–d transitions of Mn2+ ions. Second-harmonic generation (SHG) was measured with a fundamental light at 2090 nm using the Kurtz and Perry technique, which showed that its effective SHG coefficient was about half that of LiB3O5(LBO). AC impedance spectroscopy shows that along the crystallographic b axis the conductivity presumably due to Li ion migration is 1.1  107 S cm1 at 394 8C and the activation energy Ea is 0.599 eV. Together with its chiral magnetic ground state, present study shows that h-LiMnBO3 is an interesting multifunctional material to possess also nonlinear optic and ionic conducting properties. ß 2012 Elsevier Ltd. All rights reserved.

Keywords: A. Oxides B. Crystal growth C. Impedance spectroscopy D. Ionic conductivity D. Optical properties

1. Introduction In recent years, wide arrays of multifunctional materials have been proposed because of their integrated electrical, magnetic, optical and mechanical properties. For instance, the commercial piezoelectric materials, a-quartz (SiO2) and lead titanate (PbTiO3) are the core materials in a great number of devices such as actuators, sensors, and transducers [1,2]. As typical magneticoptical materials, Yttrium iron garnet (Y3Fe5O12, YIG) and dopedYIG have been widely applied in tunable microwave devices, circulators, isolators, phase shifters, nonlinear devices, etc. [3,4]. Lithium niobate (LiNbO3) and lithium tantalite (LiTaO3) have found significant applications that exploit their electro-optic characteristic, such as electro-optic modulation [5,6]. LiMnBO3 was first synthesized by Legagneur et al.; they explored the possibility of using the boron-based compound LiMnBO3 as a candidate electrode material for lithium ion battery [7]. Since then, many efforts have been continuously made to improve the discharge capacity, electrochemical activity, cycling performance of LiMnBO3 [8–10]. However, up to now, as an electrode material LiMnBO3 was synthesized only by solid-state reaction method and there has been no report on the electrochemical properties of bulk LiMnBO3 crystals. As a matter of fact, LiMnBO3 exists in two polymorphs: the low-temperature monoclinic form m-LiMnBO3 which was synthesized below 400 8C in hydrothermal conditions [11], and the high-temperature hexago-

* Corresponding author. Tel.: +86 10 82543711. E-mail address: [email protected] (R.K. Li). 0025-5408/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.materresbull.2012.10.057

nal form h-LiMnBO3 [7] with space group P 6/bar and lattice parameters of a = 8.172 A˚ and c = 3.1473 A˚ which is related to the present work. Magnetic neutron diffraction results reveal that hLiMnBO3 contains antiferromagnetic (AFM) MnO5 chains along c direction and orders in a novel chiral ground state in the ab plane consisting of a mixture of one ferromagnetic (FM) Mn3 triangle and two normal 1208 frustrated AFM Mn3 triangles [12]. In addition to the interesting Li transport and Mn–Mn magnetic interactions, the arrangement of the borate group BO3 in the structure is critical in achieving significant nonlinear optical (NLO) effects. Within the ab plane, all of the three BO3 groups are approximately aligned in the same direction (Fig. 1). According to the anionic group theory which assumes the macroscopic NLO coefficients can be obtained by directional summation of the contributions from the anionic groups within the unit cell of a given crystal [13,14], such an arrangement of BO3 groups in LiMnBO3 can be identified as a favourable factor to produce a large second harmonic generation (SHG) response. It is well established now the borate crystals are the best candidate NLO materials for high power laser frequency conversions since the discovery of BaB2O4 (BBO) and LiB3O5 (LBO) crystals [15,16] due to their large NLO coefficients and high damage threshold. Therefore, efforts were made herein to grow bulk single-crystal LiMnBO3 and investigate its electro-chemical and optical properties. 2. Experimental h-LiMnBO3 single-crystals were prepared from MnO and LiBO2 by spontaneous nucleation using the slow-cooling method according to a two-step process. The starting powders were

278

LW. Zhao, R.K. Li / Materials Research Bulletin 48 (2013) 277–280

Fig. 1. (a) Crystal structure of LiMnBO3 along c-axis. (b) Crystal structure of LiMnBO3 on ac plane.

analytical reagent (AR) grade Li2CO3, MnO and H3BO3. LiBO2 was first prepared from Li2CO3 and H3BO3 with the molar ratios of Li2CO3:H3BO3 = 1:1 by heating at 900 8C for 4 h. In the second step, MnO and LiBO2 with the molar ratios of MnO:LiBO2 = 1:1.33 were placed in a platinum crucible and first heated for 24 h at 950 8C to ensure homogeneity of the melt, and then cooled from 850 to 800 8C at a rate of 2 8C/day, followed by cooling to room temperature at a rate of 10–30 8C per hour. The crystal growth process was kept in a reducing atmosphere of H2(10%)/N2. Wellfaced transparent purple crystals in millimetre-size were obtained at the surface of the content in the Pt crucible. A crystal plate of 4 mm  3 mm  0.5 mm (Fig. 2, inset) in size was chosen for optical and electrochemical properties studies. Room temperature optical transmission spectrum was recorded with a Lambda 900 UV–vis-NIR (Perkin-Elmer) spectrophotometer in the region of 185–3000 nm. The impedance measurements with an Agilent 4294A Precision Impedance Analyzer in the frequency range from 100 Hz to 100 MHz were performed in the temperature range 200–400 8C. The silver-coated single crystal plate was placed in the sample holder about 2 mm next to the thermocouple. Though quite strong SHG signal (green light) from the single crystal was clearly observed, quantitative measurement could not be done with an Nd:YAG laser because LiMnBO3 absorbs part of the SHG signal at 532 nm. The optical SHG response of the crystalline sample of the LiMnBO3 was instead measured by means of the Kurtz–Perry method [17] with an fundamental light at 2090 nm, which was generated with a Q-switched Ho:Tm:Cr:YAG laser. Since SHG efficiencies are known to depend strongly on particle

size, the LiMnBO3 crystallites were ground and sieved into distinct particle sizes: 0–50, 50–74, 74–105, 105–150, 150–200, 200– 300 mm, respectively. For comparison, powdered LiB3O5 (LBO) crystal of similar particle size served as a reference. 3. Results and discussion 3.1. Transmission spectrum The transmission spectrum of the LiMnBO3 crystal, shown in Fig. 2, consists of six absorption bands: a small weak peak at 311 nm, two sharp absorption peaks at 370 and 423 nm, along with a weak shoulder peak at 356 nm, and two broad bands at 513 and 581 nm. These absorption bands can be tentatively associated to the d–d transitions of Mn2+ ions based on Tanabe–Sugano diagram of the d5 configuration [18,19]: the first small, weak peak 311 nm together with the shoulder peak at 356 nm and the two higher and sharp energy peaks at 370 nm, 423 nm are attributable to the crystal-field independent or weak dependent transitions from the ground state 6A1(6S) to the 4T1 (4P); 4E(4D); 4T2(4D); 4A1 and 4E(4G) levels, respectively. The two broad bands at 513 nm and 581 nm are assigned to transitions from the ground state 6A1(6S) to 4T2(4G) and 4T1(4G) with band broadening due to crystal-field effect. The energy levels were calculated using the Racah (B and C), the crystal field splitting (Dq) and the Trees correction parameters (a) [20,21]. In the present study, the same free ion value of a = 76 cm1 was adopted as that for Mg2B2O5:Mn2+ [21]. The energy matrices including Trees correction terms were presented by Tanabe and Sugano and Mehra [20], respectively. The electrostatic parameters of B = 866 cm1, C = 2638 cm1 and Dq = 727 cm1 were evaluated using the following equations [21]:





94a þ

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 49ðT 2  T 1 Þ2  768a2 49

T 1 þ T 2  27B  26a 10

ð10DqÞ2 ¼ ð10B þ 6C þ 12a  T 3 Þ2  ðC  8aÞ2 

We have taken T1 as the energy difference between A1(6S) ! 4E(4G), T2 as 6A1(6S) ! 4E(4D), which are independent of crystal field interactions, and T3 as 6A1(6S) ! 4T1(4G). 6

Fig. 2. The transmission spectrum of the LiMnBO3 crystal.

4ð3B þ 2aÞ2 ð10B þ 5C þ 20a  T 3 Þ 19B þ 7C þ 10a  T 3

LW. Zhao, R.K. Li / Materials Research Bulletin 48 (2013) 277–280

279

Table 1 Optical absorption peak positions of LiMnBO3 crystal. Energy level

l (nm)

nexp (cm1)

ncal (cm1)

4

311 356 370 423 513 581

32,154 28,090 27,027 23,641 19,493 17,212

31,007 28,109 26,408 23,668 21,352 17,212

4

T1( P) E(4D) T2(4D) 4 A1, 4E(4G) 4 T2(4G) 4 T1(4G) 4 4

With above obtained parameters, the rest three energy levels were calculated by diagonalizing Tanabe–Sugano matrices [22]. A satisfactory agreement between the calculated and observed data can be seen from Table 1, justifying the assignments. Although the absorption bands in the ultraviolet (UV) and visible region exist, it is especially noteworthy that LiMnBO3 crystal possesses ultraviolet transmission cut-off wavelength of 285 nm and high transmittance (80%) in the 680–3000 nm optical range, therefore, LiMnBO3 can still be a useful optical crystal in the visible and infrared range.

Fig. 4. The complex impedance taken on LiMnBO3 single crystal along b-axis at various temperatures.

3.3. Impedance spectroscopy 3.2. SHG measurement As aforementioned, since all the three BO3 groups in the unit cell are parallel and point to the same direction, LiMnBO3 should possess relative large SHG (one type of NLOs) coefficients based on the group theory proposed by Chen, Wu and Li [13–16]. The SHG signal intensities as a function of particle sizes from the measurements on ground crystallites of LiMnBO3 and LBO are shown in Fig. 3. The SHG measurement indicates that LiMnBO3 displays SHG signals about a quarter that of LBO. As the output signal is proportional to the square of the effective SHG coefficient deff, we can estimate that deff of LiMnBO3 is about half that of LBO. For comparison, we calculated the SHG coefficients of LiMnBO3 based on the contribution of the three BO3 groups [16], which gave the following values: d22 = d21 = 1.65 pm/v, d12 = d11 = 0.06 pm/v. The results are approximately consistent with the experimental value of deff = 0.5deff (LBO). Furthermore, based on the same level of approximation [23], the refractive indices of LiMnBO3 can also be calculated as: no = 1.7754 and ne = 1.6874 with Dn = 0.088. Considering the relative large birefringence which guarantees phase matchable SHG output together with its large NLO coefficients, we are confident that LiMnBO3 can be a candidate material for application in the visible and IR nonlinear optics.

Fig. 3. SHG measurements of LiMnBO3 ground crystals with LBO as a reference.

Fig. 4 presents the experimental results of the impedances taken on LiMnBO3 single crystal along b-axes at various temperatures. A partial semicircle is obtained at low temperature; with increasing temperature more segments of the semicircle are formed and at 394 8C nearly perfect semicircles are obtained. A typical equivalent RC parallel circuit, for a Cole–Cole plot has been used to fit the data (Fig. 4, inset):

Z RkC ¼

R 1 þ v2 C 2 R2

i

vCR 1 þ v2 C 2 R2

In which v, R and C denote the angular frequency, the AC resistance and the capacitance, respectively. The temperature dependence of the conductivity of LiMnBO3 crystal is plotted as ln(s(T)) versus 1000/T, which can be well fitted by the Arrhenius equation (Fig. 5).

A T

  Ea kT

s ðTÞ ¼ exp

where s(T) is the conductivity, A the pre-exponential constant, Ea the activation energy for ionic or electron migration, k the

Fig. 5. The temperature dependent conductivities along b-axis of LiMnBO3 crystal.

280

LW. Zhao, R.K. Li / Materials Research Bulletin 48 (2013) 277–280

Boltzmann constant, T the absolute temperature. From the slope of the fit, the calculated Ea in the temperature range of 200–400 8C along the b-axis is extracted as 0.599 eV. At this moment, we cannot differentiate whether the measured conductivity is ionic or electronic in origin. Nonetheless, since the obtained conductivities and the activation energy are well in accordance with the other known crystalline Li ion conductors, such as LiFePO4 [24–26], Li5La3Bi2O12, Li6SrLa2Bi2O12 [27], we presumably attributed the measured conductivity to Li ion migration. As can be seen from the structure of h-LiMnBO3 (Fig. 1), it is built up from MnO5 square pyramids, planar BO3 groups, and LiO4 tetrahedra. Similar as the MnO5 chains, LiO4 tetrahedra also form linear chains along the c-axis. It can be expected that lithium hopping along c-axis only involves the neighbouring LiO4 tetrahedra, which should give largest ionic conductivity along caxis. In comparison, hopping along the b-axis must involve cooperative movements of the MnO5 pyramids and planar BO3 groups, which must give a smaller conductivity and higher migration energy. Unfortunately, under present growth conditions, the obtained crystal tends to have a plate habit with the longest dimension along c and shortest dimension along a or b axis. Therefore, growing crystals with large area in the ab plane are needed to fully characterize the ionic conduction performance of LiMnBO3 crystal. 4. Conclusions Hexagonal LiMnBO3 single crystals with sizes up to 4 mm  3 mm  0.5 mm were obtained by slow-cooling method in a reducing atmosphere. Six absorption peaks in the visible and ultraviolet regions in the transmittance spectra, according to the crystal-field theory and Tanabe–Sugano scheme, can be assigned to the d–d transitions of Mn2+ ions as follows: 6A1(6S) to the 4T1 (4P); 4E(4D); 4T2(4D); 4A1, 4E(4G); 4T2(4G); and 4T1(4G), respectively. Second-harmonic generation was performed using the Kurtz and Perry technique and it showed that the effective SHG coefficient deff is about 0.5 times as large as that of LBO. Considering its high transparency in the 680–3000 nm range and relative large NLO coefficients and birefringence, LiMnBO3 may found applications in the visible and IR nonlinear optics. The AC measurements using Ag as electrodes in the symmetric cell Ag/LiMnBO3/Ag shows that Li ion conductivity along b-axis increases with increasing temperature to reach 1.1  107 S cm1 at 394 8C and the activation energy

is about 0.599 eV along b-axis. Further studies on the magnetooptic properties on the title crystal at low temperature may also be interesting when larger crystals can be obtained. Acknowledgments This work is supported by the National Science Foundation of China (Nos. 90922036 and 51032004/E0201). The authors would like to thank Prof. C. Greaves from School of Chemistry, The University of Birmingham for his help in arranging the thermogravimetric experiment on the LiMnBO3 sample and his thoughtful discussion during the writing up of this manuscript. References [1] D.Z. Shen, Q. Kang, Y.H. Xue, L.X. Chen, L.Z. Wang, Sens. Actuators B 50 (1998) 253. [2] T.Y. Chen, S.Y. Chu, R.C. Chang, C.K. Cheng, C.S. Hong, H.H. Nien, Sens. Actuators A 134 (2007) 452. [3] T. Boudiar, B. Payet-Gervy, M.-F. Blanc-Mignon, J.-J. Rousseau, M. Le Berre, H. Joisten, J. Magn. Magn. Mater. 284 (2004) 77. [4] O. Kamada, T. Nakaya, S. Higuchi, Sens. Actuators A 119 (2005) 345. [5] M.J. Geselbracht, A.M. Stacy, A.R. Garcia, B.G. Silbernagel, G.H. Kwei, J. Phys. Chem. 97 (1993) 7102. [6] M. Nyman, T.M. Anderson, P.P. Provencio, Cryst. Growth Des. 9 (2009) 1036. [7] V. Legagneur, Y. An, A. Mosbah, R. Portal, A. Le Gal La Salle, A. Verbaere, D. Guyomard, Y. Piffard, Solid State Ionics 139 (2001) 37. [8] S.M. Wang, X.J. Huang, L.Q. Chen, J. Mater. Chem. 10 (2000) 1465. [9] V. Aravindan, K. Karthikeyan, S. Amaresh, Y.S. Lee, Bull. Korean Chem. Soc. 31 (2010) 1506. [10] L. Chen, Y.M. Zhao, X.N. An, J.M. Liu, Y.Z. Dong, Y.H. Chen, Q. Kuang, J. Alloys Compd. 494 (2010) 415. [11] O. Bondareva, M. Simonov, Y.K. Egorov-Tismenko, N. Belov, Sov. Phys. Crystallogr. 23 (1978) 269. [12] R.K. Li, C.T. Chen, C. Greaves, Phys. Rev. B 66 (2002) 052405. [13] R.K. Li, C.T. Chen, Acta Phys. Sin. 34 (1985) 823. [14] C.T. Chen, Y.C. Wu, R.K. Li, J. Cryst. Growth 99 (1990) 790. [15] C.T. Chen, B.C. Wu, A.D. Jiang, G.M. You, Sci. Sin. Ser. B 28 (1985) 235. [16] C.T. Chen, Y.C. Wu, A.D. Jiang, B.C. Wu, G.M. You, R.K. Li, S.J. Lin, J. Opt. Soc. Am. B 6 (1989) 616. [17] S.K. Kurtz, T.T. Perry, J. Appl. Phys. 39 (1968) 3798. [18] Y. Tanabe, S. Sugano, J. Phys. Soc. Jpn. 9 (1954) 766. [19] B.N. Figgis, M.A. Hitchman, Ligand Field Theory and Its Applications, Wiley-VCH, New York, 2000. [20] A. Mehra, J. Chem. Phys. 48 (1968) 4384. [21] T. Kawano, T. Suehiro, T. Sato, H. Yamane, J. Lumin. 130 (2010) 2161. [22] Y. Tanabe, S. Sugano, J. Phys. Soc. Jpn. 9 (1954) 753. [23] F.L. Qin, R.K. Li, J. Cryst. Growth 318 (2011) 642. [24] J.Y. Li, W.L. Yao, S. Martin, D. Vaknin, Solid State Ionics 179 (2008) 2016. [25] R. Amin, J. Maier, P. Balaya, D.P. Chen, C.T. Lin, Solid State Ionics 179 (2008) 1683. [26] R. Amin, C.T. Lin, J. Maier, Phys. Chem. Chem. Phys. 10 (2008) 3519. [27] R. Murugan, W. Weppner, P. Schmid-Beurmann, V. Thangadurai, Mater. Sci. Eng. B 143 (2007) 14.