- Email: [email protected]
Fabrication of a high lithium ion conducting lithium borosilicate glass
Journal of Non-Crystalline Solids 357 (2011) 2863–2867
Contents lists available at ScienceDirect
Journal of Non-Crystalline Solids j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n o n c r y s o l
Fabrication of a high lithium ion conducting lithium borosilicate glass Chul Eui Kim a, Hyun Chul Hwang a, Mi Young Yoon a, Byung Hyun Choi b, Hae Jin Hwang a,⁎ a b
School of Materials Science and Engineering, Inha University, Incheon, Republic of Korea Korea Institute of Ceramic Engineering and Technology, Seoul, Republic of Korea
a r t i c l e
i n f o
Article history: Received 29 June 2010 Received in revised form 10 March 2011 Available online 15 April 2011 Keywords: Li ion conducting glass; Solid electrolyte; Thin film battery; Lithium borosilicate glass; Impedance spectrum
a b s t r a c t A lithium ion conducting glass, Li2O–B2O3–SiO2, is fabricated by the conventional melt and quenching technique from a mixture of Li2CO3, B2O3 and SiO2 powders. It appears that B2O3 decreases the crystallization tendency of the Li2O–SiO2 binary glass, resulting in an expanded glass forming region in the Li2O–B2O3–SiO2 ternary glass. The maximum conductivity is 2 × 10− 6 S cm− 1 at 25 °C for the 50Li2O–38B2O3–12SiO2 glass sample. The observed high conductivity is due to the mixed former effect. The conductivity strongly depends on the Li2O content, but not on K (SiO2/B2O3) in the Li2O–B2O3–SiO2 ternary glass. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Lithium ion conducting inorganic solid electrolytes have attracted considerable attention, because they are the key materials for all solid state rechargeable thin film batteries, which are potential candidates as a power source for small electronic products, such as semiconductor diagnostic wafers, smart cards, sensors and radio frequency identification tags, and medical products, such as pacemakers and other implantable surgical devices. The lithium ion conducting solid electrolyte should be highly conductive, compatible with electrode materials, such as lithium metal, and chemically stable in an air atmosphere. In addition, it should be easily fabricated at low temperatures via thin film fabrication processes such as sputtering. The materials use for lithium ion conducting solid electrolytes can be divided into two categories, crystalline and amorphous. Lithium lanthanum titanate (LLTO) [1,2], NASICON- [3], NISICON/thio LISICON- [4,5] and garnet- [6] type Li ion conductors are included in the crystalline category. On the other hand, oxide or sulfide glasses and LiPON are amorphous Li ion conductors. Although NASICON- and perovskite-type Li ion conductors show high conductivities of up to 10− 3 to 10− 4 S cm− 1, they are not phase-compatible with Li metal, due to the reduction of the Ti4+ ions. In addition, high temperature heattreatment is required to obtain a dense crystalline body [2]. NISICON and its related materials suffer from low Li ion conductivity and high reactivity with Li metal and CO2 [4]. Amorphous-type Li ion conductors offer several advantages over crystalline conductors, such as their ease of fabrication by thin film techniques, low temperature processing and isotropic conductivity. ⁎ Corresponding author. E-mail address: [email protected] (H.J. Hwang). 0022-3093/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jnoncrysol.2011.03.022
Among them, sulfide-based glasses show a high Li ion conductivity (for example, 10− 3 S cm− 1 for GeS2–LiS2–LiI at 25 °C) and improved thermal stability [7]. However, they are easily devitrified, corrosive and hygroscopic. Lithium phosphorous oxynitride (LiPON) with a typical composition of Li2.88PO3.73N0.14, which was first reported by Bates et al., shows a good Li ion conductivity of the order of 10− 6 S cm− 1 at 25 °C [8]. LiPON is prepared by the sputtering technique using a Li3PO4 target under an N2 atmosphere. The incorporation of nitrogen into lithium phosphate allows LiPON to be chemically stable with Li metal and not to decompose even at 5 V. However, the problems of its poor reproducibility and structural degradation in an air atmosphere remain unsolved [9]. On the other hand, silica-based Li ion conducting glasses such as lithium aluminosilicates [10], lithium borosilicates [11], lithium phosphosilicates [12] and so on, have been systematically studied, due to their excellent chemical and thermal stability. In fact, the Li ion conductivity of lithium silicate glasses is in the range of 10− 7 to 10− 8 S cm− 1 at room temperature. In addition, special techniques, for example, rapid quenching using twin roller are needed to fabricate the glass with high concentration of lithium ion [12]. In this study, lithium borosilicate (Li2O–B2O3–SiO2) glasses with 35 to 50 mol% of Li2O were fabricated by the melt quenching method. The lithium ion conductivities of the glasses were measured at temperatures ranging from room temperature to 200 °C using an ac impedance analyzer and were discussed in terms of the Li2O content and the K value, i.e., the SiO2/B2O3 ratio, in the Li2O–B2O3–SiO2 composition. 2. Experimental procedure Lithium carbonate (Li2CO3), boron oxide (B2O3) and silica (SiO2) were used as the starting materials. Appropriate amounts of the
2864
C.E. Kim et al. / Journal of Non-Crystalline Solids 357 (2011) 2863–2867
Counts (Arb. Units)
starting materials were used so as to fabricate the glass with the composition shown in Table 1, mixed well and heated at 1000– 1200 °C for 3 h in a platinum crucible. A homogeneous free flowing melt was obtained at these temperatures, which was then quenched in a graphite mold. The bar-shaped glass sample (20 × 20 × 100 mm) was then annealed at 400–500 °C for subsequent machining. X-ray diffraction (XRD) patterns were taken using a diffractometer (DMAX-2500, Rigaku, Japan) with Ni filtered Cu–Kα radiation to confirm the crystallization of the sample. Differential thermal analysis (DTA) was carried out to determine glass transition temperature using a TG-DTA instrument (Diamond TG/DTA, Perkin Elmer, USA). Li ion conductivity measurements were performed in the range from room temperature to 200 °C by an ac impedance analyzer (IM6, Zahner Co., Germany), in the frequency range of 1 Hz–1 MHz. A disktype sample (20 × 20 × 0.25 mm) was cut from the bar-shaped glass and polished using a # 2000 SiC paper. Platinum ionic blocking electrodes with an area of 1 cm2 were deposited by sputtering (E-1030, Hitachi Co. Ltd., Japan) on both sides of the sample.
(b) (c) 20
3. Results
30
40
50
60
70
80
2θ (degree)
Li2O
B2O3
SiO2
K (SiO2/B2O3)
35 40 45 35 40 45 47.5 35 40 42.5 45 47.5 50 40 45 47.5 50 52.5 37.5 40 42.5 45 47.5 47.5
21.67 20 18.33 32.5 30 27.5 26.25 43.33 40 38.33 36.67 35 33.33 46.15 42.31 40.38 38.46 36.54 52.08 50 47.92 45.83 43.75 47.73
43.33 40 36.67 32.5 30 27.5 26.25 21.67 20 19.67 18.33 17.5 16.67 13.85 12.69 12.12 11.54 10.96 10.42 10 9.58 9.17 8.75 4.77
2.0 2.0 2.0 1.0 1.0 1.0 1.0 0.5 0.5 0.5 0.5 0.5 0.5 0.3 0.3 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.2 0.1
The composition shown here is nominal.
was partially crystallized. As K decreases from 2.0 to 0.3, the glass forming region is gradually extended to compositions with a high Li2O content, indicating that glasses containing large amounts of Li2O can be obtained in the Li2O–B2O3–SiO2 system. However, the glass forming region shrunk with the further substitution of SiO2 by B2O3. The glass with 47.5 mol% Li2O and K = 0.2 was found to be partially crystallized. Fig. 4(a) shows the impedance spectra, measured at 25 °C, for the Pt/glass electrolyte with 40, 45, 47.5 and 50 mol% Li2O/Pt cells. All of the samples exhibited the typical impedance spectrum observed in the case of cells consisting of a glass electrolyte and blocking electrodes, such as Pt, i.e., one semicircle in the high and medium frequency range, followed by a straight line due to the polarization or diffusion of Li ions at the interface between the glass electrolyte and Pt blocking electrode [13]. The bulk resistance of the glass electrolyte can be obtained from the right intercept of the semicircle with the real axis (x axis) and the conductivity is then calculated from the resistance, thickness of the electrolyte and area of the electrode. Fig. 4(b) presents the log conductivity for the Li2O–B2O3–SiO2 glass
Exothermic
Table 1 Composition for lithium borosilicate glass system, in mol%a.
Fig. 1. XRD patterns of 50Li2O–38B2O3–12SiO2 (K=0.3) (a), 47.5Li2O–35B2O3–17.5SiO2 (K=0.5) (b) and 45Li2O–27.5B2O3–27.5SiO2 (K=1) (c) glasses.
502°C
(a) (b)
Endothermic
Fig. 1 shows the typical XRD patterns of the lithium borosilicate glasses with the compositions of 50Li2O–38B2O3–12SiO2, 47.5Li2O– 35B2O3–17.5SiO2 and 45Li2O–27.5B2O3–27.5SiO2. There was no characteristic peak which corresponds to any crystalline phase and therefore it can be inferred that the obtained samples are amorphous. Fig. 2 shows DTA curve for the 50Li2O–38B2O3–12SiO2, 47.5Li2O– 35B2O3–17.5SiO2 and 45Li2O–27.5B2O3–27.5SiO2 glasses. The small endothermic peak appears between 400 and 500 °C, and followed by large exothermic peak between 500 and 600 °C, respectively. Each peak is attributed to glass transition temperature and crystallization temperature, respectively. Both glass transition and crystallization temperatures decrease with increasing the Li2O content in the glass composition. The glass forming region in the Li2O–B2O3–SiO2 composition diagram is shown in Fig. 3. The gray circles denote the compositions for which glassy and transparent samples were obtained and the closed circles indicate the samples which contain a small amount of crystallites. At K = 2.0, i.e., SiO2/B2O3 = 2.0, the sample containing 42.5 mol% Li2O is glassy, while the sample containing 45 mol% Li2O
a
(a)
545°C
(c)
413°C
430° C
445°C
200
300
400
560°C
500
600
700
Temperature (°C) Fig. 2. DTA curves of 50Li2O–38B2O3–12SiO2 (K = 0.3) (a), 47.5Li2O–35B2O3–17.5SiO2 (K = 0.5) (b) and 45Li2O–27.5B2O3–27.5SiO2 (K = 1) (c) glasses.
C.E. Kim et al. / Journal of Non-Crystalline Solids 357 (2011) 2863–2867
2865
4. Discussion
Fig. 3. Glass forming region of Li2O–B2O3–SiO2 glasses in this study. The gray and closed circles indicate the perfect transparent glass and partially crystallized glass, respectively.
electrolyte with K = 0.3 as a function of the Li2O content. The conductivity increases with increasing Li2O content. Fig. 5 presents the effect of K on the conductivity of the various glass electrolytes with different Li2O contents. The room temperature conductivities of the glass samples are summarized in Table 2. As can be seen in Fig. 5 and Table 2, the conductivity of the glass depended strongly on the Li2O content, i.e., the glass samples containing more Li2O showed higher conductivities. However, the K value has no significant effect on conductivity of the glasses. The Arrhenius plots of the conductivity for the glass samples are shown in Fig. 6 and the activation energies are summarized in Table 3. A weighted least-square method was used to fit a linear model function to data and the correlation coefficients are also displayed in Table 3. The activation energy of the glass samples gradually decreased with increasing Li2O content. On the other hand, the glasses with the same mol% of Li2O and different K value showed similar activation energies.
From Figs. 1 and 2 it was confirmed that the lithium borosilicate glasses with the composition of 50Li2O–38B2O3–12SiO2 is amorphous. This glass has 50 mol% Li2O, which is slightly greater than those reported in other papers. Tatsumisago et al. reported that the limit of glass formation in the Li2O–SiO2 system is 40 mol% Li2O in the case of the normal melt quenching technique [14]. Maia et al. fabricated lithium borosilicate glasses with 40 mol% Li2O by melt quenching [15]. It appears that the ratio of SiO2/B2O3 (=K value) has a large effect on the glass formation in the lithium borosilicate system. As is evident in Fig. 3, the glass formation domain was extended in the composition with a high B2O3 molar ratio. It is considered that this observed phenomenon is partly caused by the lower melting temperature of the Li2O–B2O3 glass, than that of the Li2O–SiO2 glass [14]. In addition, B2O3 can decrease the viscosity and crystallization tendency of Li2O–SiO2 glasses, resulting in a wider glass formation region. Generally, the conductivity of lithium borosilicate glasses increases as a molar ratio of Li2O/(B2O3 + SiO2) increases. This can be explained by two factors; one is an increase in the charge carrier (lithium) concentration and the other is the mobility enhancement of the charge carrier. The increase in conductivity as a function of the Li2O content observed in Fig. 4(b) can be explained by the increase in the Li ion concentration of the glass and the accompanying change in the chemical bonding of the glass structure, i.e., the increased number of non-bridging oxygens (NBOs) [16,17]. In addition, it can be noted that the conductivities of the Li2O– B2O3–SiO2 ternary glasses are much higher than those of the Li2O–B2O3 or Li2O–SiO2 binary glass [18–20], suggesting that the mixed former effect is responsible for the high conductivity observed in the ternary glass. For a lithium borate binary glass, adding SiO2 as the second glass former can allow the borate glass to be more depolymerized, because SiO2 can play the role of a network modifier, resulting in an increased orthoborate or pyroborate content, which can enhance the conductivity in the ternary glass with a small amount of SiO2 [16]. Tatsumisago et al. [14] reported that the conductivity of lithium borosilicate glasses can be improved by mixing two glass formers such as B2O3 and SiO2, due to the so-called mixed former effect. Lee et al. [16] also reported the mixed former effect in the Li2O–SeO2–B2O3 glass system. They also reported that the conductivity decreased with increasing SeO2 content. This means that adding a small amount of SeO2 is effective in improving the conductivity of the Li2O–SiO2 binary
Fig. 4. Impedance spectra (a) and conductivity as a function of Li2O% (b) of Li2O–B2O3–SiO2 glasses with K = 0.3 at 25 °C.
2866
C.E. Kim et al. / Journal of Non-Crystalline Solids 357 (2011) 2863–2867
-5.5 35 Li2O 40 Li2O 45 Li2O
-6.0
47.5 Li 2O
log σ (S/cm)
50 Li2O
-6.5
-7.0
-7.5
-8.0 0.0
0.5
1.0
1.5
2.0
K (SiO2/B2O3) Fig. 6. Conductivity vs. 1/T curves of Li2O–B2O3–SiO2 glasses.
Fig. 5. Conductivity vs. K (SiO2/B2O3) curves of Li2O–B2O3–SiO2 glasses with various Li2O contents.
glass. Salodkar et al. [18] observed two maxima in the conductivity in the Li2O–B2O3–P2O5 glass system and they explained the mixed former effect by using a thermodynamic approach. On the other hand, Maria et al. reported that there is no mixed former effect between B2O3 and SiO2 in the 40Li2O–60(B2O3–Si2O4) system [15]. Although the mixed former effect was observed in the Li2O–B2O3– SiO2 glasses, it appears that K does not affect the conductivity of the ternary glasses, which means that the mixed former effect is effective in those glasses with a small amount of the second glass former. It can be considered that the existence of two conductivity maxima in the Li2O–B2O3–P2O5 glass is closely related to this phenomenon. As can be seen in Fig. 4(b) and Fig. 5, it is clear that the glass with a higher molar ratio of Li2O/(B2O3 + SiO2) shows the higher lithium ion conductivity and this suggests that increasing the Li2O content in the glass composition leads to an increase in the number of non-bridging oxygen, which enhances the Li ion mobility. The decrease in the activation energy of the glass with a higher Li2O content might be associated with the increase in non-bridging oxygen in the glass structure, resulting in the enhanced Li ion mobility. On the other hand, at the same Li2O content, the conductivity and the activation energy remains constant irrespective of the K value. For the glasses with the same Li2O content and different K values, the conductivity is determined by the mobility, i.e., the number of NBOs in the glass
structure. The fact that a similar conductivity and activation energy are observed in the glasses with the same Li2O and different K values shows that their glass structures are similar with respect to the number of NBOs, i.e., the connectivity of the glass former polyhedra. Further studies on the glass structure are needed to clarify the conduction mechanism of the glasses.
5. Conclusion A lithium ion conducting Li2O–B2O3–SiO2 glass was fabricated by the conventional melt and quenching technique. The glass forming region was extended as the SiO2/B2O3 ratio was decreased to 0.3. It appears that B2O3 decreases the viscosity and crystallization tendency of the Li2O–SiO2 glass. However, further increasing the B2O3 content did not have a positive effect on the glass formation range. 50 mol% Li2O was found to be the maximum content at which a glassy sample can be obtained in the Li2O–B2O3–SiO2 glass. The 50Li2O– 38B2O3–12SiO2 glass showed the highest lithium ion conductivity of 2 × 10− 6 S cm− 1 at 25 °C. The conductivity and activation energy increased and decreased with increasing Li2O content in the glass. However, it was found that the K value, i.e. the SiO2/B2O3 ratio, had no significant effect on the conductivity and activation energy in the Li2O–B2O3–SiO2 glass.
Table 2 Room temperature conductivity of Li2O–B2O3–SiO2 glasses with various K values. Li2O
B2O3
SiO2
K (SiO2/B2O3)
σ (S/cm)*
35
22 32.5 43 20 30 40 46 38.3 27.5 37 42 35 40.4 38
43 32.5 22 40 30 20 14 19.2 27.5 18 13 17.5 12.1 12
2.0 1 0.5 2 1.0 0.5 0.3 0.5 1.0 0.5 0.3 0.5 0.3 0.3
2.38 × 10− 8 2.20 × 10− 8 2.00 × 10− 8 1.28 × 10− 7 1.18 × 10− 7 1.10 × 10− 7 1.27 × 10− 7 2.65 × 10− 7 5.40 × 10− 7 5.28 × 10− 7 6.34 × 10− 7 1.17 × 10− 6 1.20 × 10− 6 1.96 × 10− 6
40
42.5 45
47.5 50
*Conductivity measurements have a precision of 3%.
Table 3 Activation energy of the Li2O–B2O3–SiO2 glasses with various K and Li2O contents. Li2O
K (SiO2/B2O3)
Activation energy (eV)
Correlation coefficient
35
2.0 1.0 0.5 2.0 1.0 0.5 0.3 1.0 0.5 0.3 0.5 0.3 0.3
0.640 0.643 0.647 0.599 0.602 0.610 0.596 0.568 0.561 0.562 0.538 0.546 0.523
0.99988 0.99989 0.99993 0.99998 0.99999 0.99997 0.99996 0.99989 0.99989 0.99997 0.99978 0.99992 0.99946
40
45
47.5 50
C.E. Kim et al. / Journal of Non-Crystalline Solids 357 (2011) 2863–2867
Acknowledgements This research was supported by the basic research program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (No. 2008D00186).
[7] [8] [9] [10] [11] [12] [13]
References [1] G.Y. Adachi, N. Imanaka, H. Aono, Adv. Mater. 8 (1996) 127. [2] S. Stramare, V. Thangadurai, W. Weppner, Chem. Mater. 15 (2003) 3974–3990. [3] H. Aono, E. Sugimoto, Y. Sadaoka, N. Imanaka, G. Adachi, J. Electrochem. Soc. 136 (1989) 590. [4] A.D. Robertson, A.W. West, A.G. Ritchie, Solid State Ionics 104 (1997) 1. [5] M. Murayama, R. Kanno, M. Irie, S. Ito, T. Hata, N. Sonoyama, Y. Kawamoto, J. Solid State Chem. 168 (2002) 140. [6] V. Thangadurai, H. Kaack, W.J.F. Weppner, J. Am. Ceram. Soc. 86 (2004) 437–440.
[14] [15] [16] [17] [18] [19] [20]
2867
J. Saienga, S.W. Martin, J. Non-Cryst. Solids 354 (14) (2008) 1475. X. Yu, J.B. Bates, G.E. Jellison, F.X. Hart, J. Electrochem. Soc. 144 (1997) 524. P. Birke, W.F. Chu, W. Weppner, Solid State Ionics 93 (1997) 1–15. J. Rogez, P. Knauth, A. Garnier, H. Ghobarkar, O. Schaf, J. Non-Cryst. Solids 262 (2000) 177–182. P. Kluva'nek, R. Klement, M. Karacon, J. Non-Cryst. Solids 353 (2007) 2004–2007. M. Tatsumisago, K. Yoneo, N. Machida, T. Minami, J. Non-Cryst. Solids 95–96 (1987) 857–864. X. Qian, N. Gu, Z. Cheng, X. Yang, E. Wang, S. Dong, Electrochim. Acta 46 (2001) 1829–1836. M. Tatsumisago, T. Minami, Mater. Chem. Phys. 18 (1987) 1–17. L.F. Maia, A.C.M. Rodrigues, Solid State Ionics 168 (2004) 87–92. C.-H. Lee, K.H. Joo, J.H. Kim, S.G. Woo, H.-J. Sohn, T. Kang, Y. Park, J.Y. Oh, Solid State Ionics 149 (2002) 59–65. S.H. Lee, K.I. Cho, J.B. Choi, D.W. Shin, J. Power Sources 162 (2006) 1341–1345. R.V. Salodkar, V.K. Deshpande, K. Singh, J. Power Sources 25 (1989) 257–263. Y.-I. Lee, J.-H. Lee, S.-H. Hong, Y. Park, Solid State Ionics 175 (2004) 687–690. P. Kluvanek, R. Klement, M. Karacon, J. Non-Cryst. Solids 353 (2007) 2004–2007.
Contents lists available at ScienceDirect
Journal of Non-Crystalline Solids j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n o n c r y s o l
Fabrication of a high lithium ion conducting lithium borosilicate glass Chul Eui Kim a, Hyun Chul Hwang a, Mi Young Yoon a, Byung Hyun Choi b, Hae Jin Hwang a,⁎ a b
School of Materials Science and Engineering, Inha University, Incheon, Republic of Korea Korea Institute of Ceramic Engineering and Technology, Seoul, Republic of Korea
a r t i c l e
i n f o
Article history: Received 29 June 2010 Received in revised form 10 March 2011 Available online 15 April 2011 Keywords: Li ion conducting glass; Solid electrolyte; Thin film battery; Lithium borosilicate glass; Impedance spectrum
a b s t r a c t A lithium ion conducting glass, Li2O–B2O3–SiO2, is fabricated by the conventional melt and quenching technique from a mixture of Li2CO3, B2O3 and SiO2 powders. It appears that B2O3 decreases the crystallization tendency of the Li2O–SiO2 binary glass, resulting in an expanded glass forming region in the Li2O–B2O3–SiO2 ternary glass. The maximum conductivity is 2 × 10− 6 S cm− 1 at 25 °C for the 50Li2O–38B2O3–12SiO2 glass sample. The observed high conductivity is due to the mixed former effect. The conductivity strongly depends on the Li2O content, but not on K (SiO2/B2O3) in the Li2O–B2O3–SiO2 ternary glass. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Lithium ion conducting inorganic solid electrolytes have attracted considerable attention, because they are the key materials for all solid state rechargeable thin film batteries, which are potential candidates as a power source for small electronic products, such as semiconductor diagnostic wafers, smart cards, sensors and radio frequency identification tags, and medical products, such as pacemakers and other implantable surgical devices. The lithium ion conducting solid electrolyte should be highly conductive, compatible with electrode materials, such as lithium metal, and chemically stable in an air atmosphere. In addition, it should be easily fabricated at low temperatures via thin film fabrication processes such as sputtering. The materials use for lithium ion conducting solid electrolytes can be divided into two categories, crystalline and amorphous. Lithium lanthanum titanate (LLTO) [1,2], NASICON- [3], NISICON/thio LISICON- [4,5] and garnet- [6] type Li ion conductors are included in the crystalline category. On the other hand, oxide or sulfide glasses and LiPON are amorphous Li ion conductors. Although NASICON- and perovskite-type Li ion conductors show high conductivities of up to 10− 3 to 10− 4 S cm− 1, they are not phase-compatible with Li metal, due to the reduction of the Ti4+ ions. In addition, high temperature heattreatment is required to obtain a dense crystalline body [2]. NISICON and its related materials suffer from low Li ion conductivity and high reactivity with Li metal and CO2 [4]. Amorphous-type Li ion conductors offer several advantages over crystalline conductors, such as their ease of fabrication by thin film techniques, low temperature processing and isotropic conductivity. ⁎ Corresponding author. E-mail address: [email protected] (H.J. Hwang). 0022-3093/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jnoncrysol.2011.03.022
Among them, sulfide-based glasses show a high Li ion conductivity (for example, 10− 3 S cm− 1 for GeS2–LiS2–LiI at 25 °C) and improved thermal stability [7]. However, they are easily devitrified, corrosive and hygroscopic. Lithium phosphorous oxynitride (LiPON) with a typical composition of Li2.88PO3.73N0.14, which was first reported by Bates et al., shows a good Li ion conductivity of the order of 10− 6 S cm− 1 at 25 °C [8]. LiPON is prepared by the sputtering technique using a Li3PO4 target under an N2 atmosphere. The incorporation of nitrogen into lithium phosphate allows LiPON to be chemically stable with Li metal and not to decompose even at 5 V. However, the problems of its poor reproducibility and structural degradation in an air atmosphere remain unsolved [9]. On the other hand, silica-based Li ion conducting glasses such as lithium aluminosilicates [10], lithium borosilicates [11], lithium phosphosilicates [12] and so on, have been systematically studied, due to their excellent chemical and thermal stability. In fact, the Li ion conductivity of lithium silicate glasses is in the range of 10− 7 to 10− 8 S cm− 1 at room temperature. In addition, special techniques, for example, rapid quenching using twin roller are needed to fabricate the glass with high concentration of lithium ion [12]. In this study, lithium borosilicate (Li2O–B2O3–SiO2) glasses with 35 to 50 mol% of Li2O were fabricated by the melt quenching method. The lithium ion conductivities of the glasses were measured at temperatures ranging from room temperature to 200 °C using an ac impedance analyzer and were discussed in terms of the Li2O content and the K value, i.e., the SiO2/B2O3 ratio, in the Li2O–B2O3–SiO2 composition. 2. Experimental procedure Lithium carbonate (Li2CO3), boron oxide (B2O3) and silica (SiO2) were used as the starting materials. Appropriate amounts of the
2864
C.E. Kim et al. / Journal of Non-Crystalline Solids 357 (2011) 2863–2867
Counts (Arb. Units)
starting materials were used so as to fabricate the glass with the composition shown in Table 1, mixed well and heated at 1000– 1200 °C for 3 h in a platinum crucible. A homogeneous free flowing melt was obtained at these temperatures, which was then quenched in a graphite mold. The bar-shaped glass sample (20 × 20 × 100 mm) was then annealed at 400–500 °C for subsequent machining. X-ray diffraction (XRD) patterns were taken using a diffractometer (DMAX-2500, Rigaku, Japan) with Ni filtered Cu–Kα radiation to confirm the crystallization of the sample. Differential thermal analysis (DTA) was carried out to determine glass transition temperature using a TG-DTA instrument (Diamond TG/DTA, Perkin Elmer, USA). Li ion conductivity measurements were performed in the range from room temperature to 200 °C by an ac impedance analyzer (IM6, Zahner Co., Germany), in the frequency range of 1 Hz–1 MHz. A disktype sample (20 × 20 × 0.25 mm) was cut from the bar-shaped glass and polished using a # 2000 SiC paper. Platinum ionic blocking electrodes with an area of 1 cm2 were deposited by sputtering (E-1030, Hitachi Co. Ltd., Japan) on both sides of the sample.
(b) (c) 20
3. Results
30
40
50
60
70
80
2θ (degree)
Li2O
B2O3
SiO2
K (SiO2/B2O3)
35 40 45 35 40 45 47.5 35 40 42.5 45 47.5 50 40 45 47.5 50 52.5 37.5 40 42.5 45 47.5 47.5
21.67 20 18.33 32.5 30 27.5 26.25 43.33 40 38.33 36.67 35 33.33 46.15 42.31 40.38 38.46 36.54 52.08 50 47.92 45.83 43.75 47.73
43.33 40 36.67 32.5 30 27.5 26.25 21.67 20 19.67 18.33 17.5 16.67 13.85 12.69 12.12 11.54 10.96 10.42 10 9.58 9.17 8.75 4.77
2.0 2.0 2.0 1.0 1.0 1.0 1.0 0.5 0.5 0.5 0.5 0.5 0.5 0.3 0.3 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.2 0.1
The composition shown here is nominal.
was partially crystallized. As K decreases from 2.0 to 0.3, the glass forming region is gradually extended to compositions with a high Li2O content, indicating that glasses containing large amounts of Li2O can be obtained in the Li2O–B2O3–SiO2 system. However, the glass forming region shrunk with the further substitution of SiO2 by B2O3. The glass with 47.5 mol% Li2O and K = 0.2 was found to be partially crystallized. Fig. 4(a) shows the impedance spectra, measured at 25 °C, for the Pt/glass electrolyte with 40, 45, 47.5 and 50 mol% Li2O/Pt cells. All of the samples exhibited the typical impedance spectrum observed in the case of cells consisting of a glass electrolyte and blocking electrodes, such as Pt, i.e., one semicircle in the high and medium frequency range, followed by a straight line due to the polarization or diffusion of Li ions at the interface between the glass electrolyte and Pt blocking electrode [13]. The bulk resistance of the glass electrolyte can be obtained from the right intercept of the semicircle with the real axis (x axis) and the conductivity is then calculated from the resistance, thickness of the electrolyte and area of the electrode. Fig. 4(b) presents the log conductivity for the Li2O–B2O3–SiO2 glass
Exothermic
Table 1 Composition for lithium borosilicate glass system, in mol%a.
Fig. 1. XRD patterns of 50Li2O–38B2O3–12SiO2 (K=0.3) (a), 47.5Li2O–35B2O3–17.5SiO2 (K=0.5) (b) and 45Li2O–27.5B2O3–27.5SiO2 (K=1) (c) glasses.
502°C
(a) (b)
Endothermic
Fig. 1 shows the typical XRD patterns of the lithium borosilicate glasses with the compositions of 50Li2O–38B2O3–12SiO2, 47.5Li2O– 35B2O3–17.5SiO2 and 45Li2O–27.5B2O3–27.5SiO2. There was no characteristic peak which corresponds to any crystalline phase and therefore it can be inferred that the obtained samples are amorphous. Fig. 2 shows DTA curve for the 50Li2O–38B2O3–12SiO2, 47.5Li2O– 35B2O3–17.5SiO2 and 45Li2O–27.5B2O3–27.5SiO2 glasses. The small endothermic peak appears between 400 and 500 °C, and followed by large exothermic peak between 500 and 600 °C, respectively. Each peak is attributed to glass transition temperature and crystallization temperature, respectively. Both glass transition and crystallization temperatures decrease with increasing the Li2O content in the glass composition. The glass forming region in the Li2O–B2O3–SiO2 composition diagram is shown in Fig. 3. The gray circles denote the compositions for which glassy and transparent samples were obtained and the closed circles indicate the samples which contain a small amount of crystallites. At K = 2.0, i.e., SiO2/B2O3 = 2.0, the sample containing 42.5 mol% Li2O is glassy, while the sample containing 45 mol% Li2O
a
(a)
545°C
(c)
413°C
430° C
445°C
200
300
400
560°C
500
600
700
Temperature (°C) Fig. 2. DTA curves of 50Li2O–38B2O3–12SiO2 (K = 0.3) (a), 47.5Li2O–35B2O3–17.5SiO2 (K = 0.5) (b) and 45Li2O–27.5B2O3–27.5SiO2 (K = 1) (c) glasses.
C.E. Kim et al. / Journal of Non-Crystalline Solids 357 (2011) 2863–2867
2865
4. Discussion
Fig. 3. Glass forming region of Li2O–B2O3–SiO2 glasses in this study. The gray and closed circles indicate the perfect transparent glass and partially crystallized glass, respectively.
electrolyte with K = 0.3 as a function of the Li2O content. The conductivity increases with increasing Li2O content. Fig. 5 presents the effect of K on the conductivity of the various glass electrolytes with different Li2O contents. The room temperature conductivities of the glass samples are summarized in Table 2. As can be seen in Fig. 5 and Table 2, the conductivity of the glass depended strongly on the Li2O content, i.e., the glass samples containing more Li2O showed higher conductivities. However, the K value has no significant effect on conductivity of the glasses. The Arrhenius plots of the conductivity for the glass samples are shown in Fig. 6 and the activation energies are summarized in Table 3. A weighted least-square method was used to fit a linear model function to data and the correlation coefficients are also displayed in Table 3. The activation energy of the glass samples gradually decreased with increasing Li2O content. On the other hand, the glasses with the same mol% of Li2O and different K value showed similar activation energies.
From Figs. 1 and 2 it was confirmed that the lithium borosilicate glasses with the composition of 50Li2O–38B2O3–12SiO2 is amorphous. This glass has 50 mol% Li2O, which is slightly greater than those reported in other papers. Tatsumisago et al. reported that the limit of glass formation in the Li2O–SiO2 system is 40 mol% Li2O in the case of the normal melt quenching technique [14]. Maia et al. fabricated lithium borosilicate glasses with 40 mol% Li2O by melt quenching [15]. It appears that the ratio of SiO2/B2O3 (=K value) has a large effect on the glass formation in the lithium borosilicate system. As is evident in Fig. 3, the glass formation domain was extended in the composition with a high B2O3 molar ratio. It is considered that this observed phenomenon is partly caused by the lower melting temperature of the Li2O–B2O3 glass, than that of the Li2O–SiO2 glass [14]. In addition, B2O3 can decrease the viscosity and crystallization tendency of Li2O–SiO2 glasses, resulting in a wider glass formation region. Generally, the conductivity of lithium borosilicate glasses increases as a molar ratio of Li2O/(B2O3 + SiO2) increases. This can be explained by two factors; one is an increase in the charge carrier (lithium) concentration and the other is the mobility enhancement of the charge carrier. The increase in conductivity as a function of the Li2O content observed in Fig. 4(b) can be explained by the increase in the Li ion concentration of the glass and the accompanying change in the chemical bonding of the glass structure, i.e., the increased number of non-bridging oxygens (NBOs) [16,17]. In addition, it can be noted that the conductivities of the Li2O– B2O3–SiO2 ternary glasses are much higher than those of the Li2O–B2O3 or Li2O–SiO2 binary glass [18–20], suggesting that the mixed former effect is responsible for the high conductivity observed in the ternary glass. For a lithium borate binary glass, adding SiO2 as the second glass former can allow the borate glass to be more depolymerized, because SiO2 can play the role of a network modifier, resulting in an increased orthoborate or pyroborate content, which can enhance the conductivity in the ternary glass with a small amount of SiO2 [16]. Tatsumisago et al. [14] reported that the conductivity of lithium borosilicate glasses can be improved by mixing two glass formers such as B2O3 and SiO2, due to the so-called mixed former effect. Lee et al. [16] also reported the mixed former effect in the Li2O–SeO2–B2O3 glass system. They also reported that the conductivity decreased with increasing SeO2 content. This means that adding a small amount of SeO2 is effective in improving the conductivity of the Li2O–SiO2 binary
Fig. 4. Impedance spectra (a) and conductivity as a function of Li2O% (b) of Li2O–B2O3–SiO2 glasses with K = 0.3 at 25 °C.
2866
C.E. Kim et al. / Journal of Non-Crystalline Solids 357 (2011) 2863–2867
-5.5 35 Li2O 40 Li2O 45 Li2O
-6.0
47.5 Li 2O
log σ (S/cm)
50 Li2O
-6.5
-7.0
-7.5
-8.0 0.0
0.5
1.0
1.5
2.0
K (SiO2/B2O3) Fig. 6. Conductivity vs. 1/T curves of Li2O–B2O3–SiO2 glasses.
Fig. 5. Conductivity vs. K (SiO2/B2O3) curves of Li2O–B2O3–SiO2 glasses with various Li2O contents.
glass. Salodkar et al. [18] observed two maxima in the conductivity in the Li2O–B2O3–P2O5 glass system and they explained the mixed former effect by using a thermodynamic approach. On the other hand, Maria et al. reported that there is no mixed former effect between B2O3 and SiO2 in the 40Li2O–60(B2O3–Si2O4) system [15]. Although the mixed former effect was observed in the Li2O–B2O3– SiO2 glasses, it appears that K does not affect the conductivity of the ternary glasses, which means that the mixed former effect is effective in those glasses with a small amount of the second glass former. It can be considered that the existence of two conductivity maxima in the Li2O–B2O3–P2O5 glass is closely related to this phenomenon. As can be seen in Fig. 4(b) and Fig. 5, it is clear that the glass with a higher molar ratio of Li2O/(B2O3 + SiO2) shows the higher lithium ion conductivity and this suggests that increasing the Li2O content in the glass composition leads to an increase in the number of non-bridging oxygen, which enhances the Li ion mobility. The decrease in the activation energy of the glass with a higher Li2O content might be associated with the increase in non-bridging oxygen in the glass structure, resulting in the enhanced Li ion mobility. On the other hand, at the same Li2O content, the conductivity and the activation energy remains constant irrespective of the K value. For the glasses with the same Li2O content and different K values, the conductivity is determined by the mobility, i.e., the number of NBOs in the glass
structure. The fact that a similar conductivity and activation energy are observed in the glasses with the same Li2O and different K values shows that their glass structures are similar with respect to the number of NBOs, i.e., the connectivity of the glass former polyhedra. Further studies on the glass structure are needed to clarify the conduction mechanism of the glasses.
5. Conclusion A lithium ion conducting Li2O–B2O3–SiO2 glass was fabricated by the conventional melt and quenching technique. The glass forming region was extended as the SiO2/B2O3 ratio was decreased to 0.3. It appears that B2O3 decreases the viscosity and crystallization tendency of the Li2O–SiO2 glass. However, further increasing the B2O3 content did not have a positive effect on the glass formation range. 50 mol% Li2O was found to be the maximum content at which a glassy sample can be obtained in the Li2O–B2O3–SiO2 glass. The 50Li2O– 38B2O3–12SiO2 glass showed the highest lithium ion conductivity of 2 × 10− 6 S cm− 1 at 25 °C. The conductivity and activation energy increased and decreased with increasing Li2O content in the glass. However, it was found that the K value, i.e. the SiO2/B2O3 ratio, had no significant effect on the conductivity and activation energy in the Li2O–B2O3–SiO2 glass.
Table 2 Room temperature conductivity of Li2O–B2O3–SiO2 glasses with various K values. Li2O
B2O3
SiO2
K (SiO2/B2O3)
σ (S/cm)*
35
22 32.5 43 20 30 40 46 38.3 27.5 37 42 35 40.4 38
43 32.5 22 40 30 20 14 19.2 27.5 18 13 17.5 12.1 12
2.0 1 0.5 2 1.0 0.5 0.3 0.5 1.0 0.5 0.3 0.5 0.3 0.3
2.38 × 10− 8 2.20 × 10− 8 2.00 × 10− 8 1.28 × 10− 7 1.18 × 10− 7 1.10 × 10− 7 1.27 × 10− 7 2.65 × 10− 7 5.40 × 10− 7 5.28 × 10− 7 6.34 × 10− 7 1.17 × 10− 6 1.20 × 10− 6 1.96 × 10− 6
40
42.5 45
47.5 50
*Conductivity measurements have a precision of 3%.
Table 3 Activation energy of the Li2O–B2O3–SiO2 glasses with various K and Li2O contents. Li2O
K (SiO2/B2O3)
Activation energy (eV)
Correlation coefficient
35
2.0 1.0 0.5 2.0 1.0 0.5 0.3 1.0 0.5 0.3 0.5 0.3 0.3
0.640 0.643 0.647 0.599 0.602 0.610 0.596 0.568 0.561 0.562 0.538 0.546 0.523
0.99988 0.99989 0.99993 0.99998 0.99999 0.99997 0.99996 0.99989 0.99989 0.99997 0.99978 0.99992 0.99946
40
45
47.5 50
C.E. Kim et al. / Journal of Non-Crystalline Solids 357 (2011) 2863–2867
Acknowledgements This research was supported by the basic research program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (No. 2008D00186).
[7] [8] [9] [10] [11] [12] [13]
References [1] G.Y. Adachi, N. Imanaka, H. Aono, Adv. Mater. 8 (1996) 127. [2] S. Stramare, V. Thangadurai, W. Weppner, Chem. Mater. 15 (2003) 3974–3990. [3] H. Aono, E. Sugimoto, Y. Sadaoka, N. Imanaka, G. Adachi, J. Electrochem. Soc. 136 (1989) 590. [4] A.D. Robertson, A.W. West, A.G. Ritchie, Solid State Ionics 104 (1997) 1. [5] M. Murayama, R. Kanno, M. Irie, S. Ito, T. Hata, N. Sonoyama, Y. Kawamoto, J. Solid State Chem. 168 (2002) 140. [6] V. Thangadurai, H. Kaack, W.J.F. Weppner, J. Am. Ceram. Soc. 86 (2004) 437–440.
[14] [15] [16] [17] [18] [19] [20]
2867
J. Saienga, S.W. Martin, J. Non-Cryst. Solids 354 (14) (2008) 1475. X. Yu, J.B. Bates, G.E. Jellison, F.X. Hart, J. Electrochem. Soc. 144 (1997) 524. P. Birke, W.F. Chu, W. Weppner, Solid State Ionics 93 (1997) 1–15. J. Rogez, P. Knauth, A. Garnier, H. Ghobarkar, O. Schaf, J. Non-Cryst. Solids 262 (2000) 177–182. P. Kluva'nek, R. Klement, M. Karacon, J. Non-Cryst. Solids 353 (2007) 2004–2007. M. Tatsumisago, K. Yoneo, N. Machida, T. Minami, J. Non-Cryst. Solids 95–96 (1987) 857–864. X. Qian, N. Gu, Z. Cheng, X. Yang, E. Wang, S. Dong, Electrochim. Acta 46 (2001) 1829–1836. M. Tatsumisago, T. Minami, Mater. Chem. Phys. 18 (1987) 1–17. L.F. Maia, A.C.M. Rodrigues, Solid State Ionics 168 (2004) 87–92. C.-H. Lee, K.H. Joo, J.H. Kim, S.G. Woo, H.-J. Sohn, T. Kang, Y. Park, J.Y. Oh, Solid State Ionics 149 (2002) 59–65. S.H. Lee, K.I. Cho, J.B. Choi, D.W. Shin, J. Power Sources 162 (2006) 1341–1345. R.V. Salodkar, V.K. Deshpande, K. Singh, J. Power Sources 25 (1989) 257–263. Y.-I. Lee, J.-H. Lee, S.-H. Hong, Y. Park, Solid State Ionics 175 (2004) 687–690. P. Kluvanek, R. Klement, M. Karacon, J. Non-Cryst. Solids 353 (2007) 2004–2007.