- Email: [email protected]
Cathode properties of sodium iron phosphate glass for sodium ion batteries
Journal of Non-Crystalline Solids 450 (2016) 109–115
Contents lists available at ScienceDirect
Journal of Non-Crystalline Solids journal homepage: www.elsevier.com/locate/jnoncrysol
Cathode properties of sodium iron phosphate glass for sodium ion batteries Satoshi Nakata, Takuya Togashi, Tsuyoshi Honma ⁎, Takayuki Komatsu Department of Materials Science and Technology, Nagaoka University of Technology, Kamitomioka-cho 1603-1, Nagaoka, Niigata 940-2188, Japan
a r t i c l e
i n f o
Article history: Received 27 May 2016 Received in revised form 21 July 2016 Accepted 3 August 2016 Available online xxxx Keywords: Sodium ion batteries Sodium iron phosphate Electrical conductivity Electrochemical properties
a b s t r a c t In order to clarify the possibility of cathode activity of Na2O-FeO-P2O5 glasses for sodium ion batteries, glass formation tendency, crystallization behavior, electrical conductivity as well as charge and discharge properties are examined. Glass formation in xFeO–(100-x)NaPO3 was confirmed for x ≤ 45 by melt quenching up to 1000 °C in N2 filled electric furnace. By means of electrical conductivity tests, conductivity was increased with increases of x content in xFeO–(100-x)NaPO3 and the activation energy for electrical conductivity was decreased. Charge and discharge profiles for Na anode exhibits high reversible discharge capacity as 115 mAh/g at 0.1 °C rate in 40FeO-60NaPO3 (30Na2O–40FeO–30P2O5) glass. The results on Raman scattering spectra suggests that glass structure is subjected to polymerized Q1 and Q2 phosphate units. And polymerized phosphates in glass cause lower density, open structure and sodium ion diffusive channels as well. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Li-ion batteries (LIB) have been intensively developed after the first launch of the Li-ion batteries in 1990s. LIBs are essential for the development of smart phones, laptop computers, and many other consumer products these days. The next targets of LIBs are considered to be automotive applications and huge energy storage systems. The materials abundance is of the primary importance to design the electrode materials for such large-scale applications. Sodium ion batteries are focused as alternative secondary batteries to save material cost recently [1]. Due to the intensive research, the energy density of the sodium ion battery is increasingly become equal to that of conventional LIBs [2]. In cathode, layer rock salt type oxides as well as poly-anion type materials are considered as active materials in recently. Triclinic P1 Na2FeP2O7 exhibits 3.0 V, 97 mAh/g with good cyclic performance [3–7]. Na2FeP2O7 is also available in aqueous based sodium ion batteries that implies chemical stability is good [8]. The authors group is proposing unique technique to produce Na2FeP2O7 by crystallization of precursor glass so-called ‘glass-ceramic process’, which is also applicable to fabricate LiFePO4, LiMnxFe1 − xPO4 and Li3V2(PO4)3 as well in our previous study [9–11]. By means of transmission electron microscope observation, there are residual amorphous phase on the surface of glass-ceramic grain. Nagakane et al. [11] are proposing that residual amorphous is effective to assist ionic conduction in LiFePO4 cathode, which exists only onedimensional Li+ diffusive channel. On the other hand, to improve ⁎ Corresponding author. E-mail address: [email protected] (T. Honma).
http://dx.doi.org/10.1016/j.jnoncrysol.2016.08.005 0022-3093/© 2016 Elsevier B.V. All rights reserved.
capacity of materials is another important point to achieve high energy density. As shown in Fig. 1 there are five crystals are found in ternary NaO0.5-FeOx-PO2.5 systems. To focus on sodium pre-doped Fe2 + base crystal, there are three candidates, Na2FeP2O7, Na4Fe3(PO4)2P2O7 and maricite type NaFePO4 are remained [12,13]. Comparing with theoretical capacity, unfortunately, NaFePO4, which is stoichiometric composition of olivine type LiFePO4, is inactive for sodium cell cathodes. Isolated PO4 units are exist periodically and they are blocking Na+ ion diffusion in maricite structure. On the other hand, the amorphous material like glass has the random three-dimensional structure. It is known that glass have large free volume and flexible open structure, hence superior alkali ion conduction exhibits in glass-ceramic derived solid state electrolytes for lithium ion batteries and sodium ion batteries as well [14,15]. It is not unique to solid electrolyte, there are possibility to develop for cathode and anode active materials. Okada et al. suggests that mechano-chemical derived amorphous FePO4 have a corner-shared matrix, and they showed similarly good capacity, not only for Li but also for Na anodes [16,17]. SnO-P2O5 glass is also available as anode active materials in Li and Na cell [18–20]. It would be interesting to examine the nature of ionic conduction of glassy state cathode materials such as Na2O-FeOP2O5 system. In this study, glass formation tendency, electrical conductivity and cathode activities of sodium iron phosphate glasses are examined for sodium ion batteries. 2. Experimental procedure The glass composition examined in this study is xFeO–(100-x)NaPO3 (x = 20, 25, 33.3, 40, 45 and 50). Fig. 1 shows glass formation region
110
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
Fig. 1. Glass formation region and examined glass compositions in NaO0.5-FeO1 + δ-PO2.5 systems. We also added typical crystals ever reported.
which is determined by INTERGLAD (glass database) with light blue [21] and illustrated sample composition as red mark. In these glasses there are three stoichiometric compositions of Na2FeP2O7 (x = 33.3), Na3Fe2(PO4)3 (x = 40) and NaFePO4 (x = 50). The fraction of Fe2 + ions in an Na2O-Fe2O3-P2O5 glass varies with the initial batch materials and preparation conditions, initial fraction of raw materials. In generally, the fraction of Fe2+ increase when ferrous raw materials are used such as FeO, Fe3O4 rather than Fe2O3 and when melts are processed under reducing conditions [22,23]. When the melting is process under atmosphere the fraction of Fe2 + is about up to 20%. In this study we are expecting to develop Na+ pre-doped cathode, therefore the fraction of Fe2+ must be high. Glasses were fabricated by a conventional meltquenching method under reducing atmosphere. Starting reagents
Fig. 2. Powder XRD patterns of xFeO–(100-x)NaPO3 (x = 20, 25, 33.3, 40, 45 and 50).
NaH2PO4 (99%, Nakarai tesque Co.), FeO (99.9%, Kojyundo chemicals Co.) were mixed well. A 20 g batch was melted in a gold crucible at 900 °C for 15 min in flowing N2 gas (5 L/min) in muffle furnace. The melts were poured onto an iron plate and pressed to a thickness of 0.5–1 mm by another iron plate. The glass transition and crystallization temperatures were determined by differential thermal analysis (DTA, Rigaku TG-8120). In order to confirm glass formation and to characterize crystallized phase, XRD patterns of all samples were obtained on Rigaku Ultima IV X-ray diffractometer (Rigaku, Japan) with D/tex 1D high-speed detector, which was operated at 40 kV, 40 mA with Cu-Kα radiation (λ = 0.154056 nm). All the measurements were carried out at room temperature under atmospheric air. Raman spectra were recorded using a Nanofinder (Tokyo Instruments, Japan) confocal Raman with Ar laser beam having a wavelength of 488 nm using a CCD detector. Electrical conductivities of glasses were measured by an alternating current (AC) impedance method (HIOKI 3522-50 LCR HiTESTER, Japan) in the temperature range from room temperature to 200 °C. Metal gold was sputtered with 6.5 mmϕ to the glass surface as electrodes (ULVAC QUICK COATER VPS-020, Japan). Pre-pulverized glass flake obtained by automatic mortar was subjected to mechanical milling in a planetary ball mill (Fritsch Premium line Pulversette No. 7). The following milling conditions were used: air atmosphere; milling speed of 600 rpm; ball to powder mass ratio of 10:1; and milling time of 1 h. In order to avoid excessive temperature rising within the grinding chamber, 15 min of ball milling duration was followed by a pause of 5 min. Composite powder was classified by sieving to the size of b 15 μm. Morphologies of glass-ceramics/carbon composites were observed by scanning electron microscope (SEM, Keyence VE-8800). Cathode electrodes were fabricated from a mixture of glass powder, polyvinylidene fluoride (PVDF) and conductive carbon black in a weight ratio of 85:5:10. N-methylpyrrolidone (NMP) was used to make slurry
Fig. 3. DTA profiles of quenched samples of xFeO–(100-x)NaPO3 (x = 20, 25, 33.3, 40, 45 and 50).
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
111
electrolyte. Cells were examined by using a battery testing system (Hokuto-denko Co.) at the charge/discharge current density of 1/100 ° C (0.08 mA cm−2) or 1/10 °C for the theoretical capacity for one-electron reaction between 2.0 and 3.8 V. 3. Results 3.1. Glass formation tendency
Fig. 4. Compositional dependence of bulk density (circle) at room temperature. We also added theoretical density (triangle).
of their mixtures. After homogenization, slurry was coated on a thin aluminum foil and dried at 100 °C for 10 h in a vacuum oven. Electrodes were then pressed and disks were punched out as 16 mmϕ. Electrochemical cells were prepared using coin type cells. Sodium metal foils were used as anode, and glass filter papers (Advantec Co., GA-100) were used as separator. Test cells were assembled in an argon-filled glove box. The dew point of Ar atmosphere in the glove box was kept as − 86 °C. The oxygen content was b0.33 ppm. The solution of 1 MNaPF6 (Tokyo Kasei Co.) in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1, v/v, Kishida Chemicals Co.) was used as
Fig. 5. Temperature dependence of electrical conductivity for vitrified samples. Cole-Cole plot at 336-369 K for 45FeO-55NaPO3 is also shown.
Fig. 2 shows the powder XRD patterns and appearance for meltquenched samples. Amber colored transparent bulk glass was obtained in 20FeO-80NaPO3 and 25FeO-75NaPO3. For x = 33.3, 40 and 45 in xFeO–(100-x)NaPO3, samples exhibit black colored opaque glass was obtained as can be seen in Fig. 2. Devitrification of 50FeO-50NaPO3 which is stoichiometric composition of NaFePO4 was confirmed. Only a halo pattern without any sharp peak was observed for 20 ≤ x ≤ 45, indicating that the melt-quenched samples with the compositions of xFeO–(100-x)NaPO3 (mol%) are amorphous (glasses) state. Meanwhile devitrified glass-ceramics which is composed of maricite type NaFePO4 (ICDD #01-089-0816) was obtained in 50FeO-50NaPO3. The DTA patterns for the as-quenched sample are shown in Fig. 3. The endothermic dips due to the glass transition and exothermic peaks due to the crystallization are clearly observed. The values of the glass transition (Tg) and crystallization peak (Tp) are also indicated in Fig. 3. For 20 ≤ x ≤ 25, Tp was increased. Primary crystallized phase was confirmed as NaPO3 in 20FeO-80NaPO3 by means of XRD. On the other hand, crystallized phase changed from NaPO3 to Na2FeP2O7 when FeO content increase from x = 20 to 25. For the other samples, both values of Tg and Tp was decreased with increases of FeO content in xFeO–(100-x)NaPO3 glasses. In addition, thermal stability criterion for crystallization ΔT = Tp − Tg was drastically attenuated by increases of FeO content. 3.2. Fundamental physical properties Bulk density of glass was determined by Archimedes method at room temperature to compare with that of corresponding crystal and results are shown in Fig. 4. As can be seen in Fig. 4, it is also indicated theoretical density which is calculated from crystal structure with triangular mark [7,24,25]. The density of glasses is monotonically increasing as increases of FeO content. Furthermore, densities of all glass samples are less than those of corresponding crystals. For instance, crystalline
Fig. 6. Compositional dependence of electrical conductivity at room temperature and activation energy for electrical conduction.
112
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
electrical conduction was decreased from 0.8 to 0.64 with increase of FeO content. The increasing of FeO content is effective not only enhance energy density but also electrical conductivity in such glasses. However, the magnitudes of 10−12–10−10 Scm−1 is not enough to offer smooth electrochemical reaction during charge and discharge process. Therefore, planetary ball milling was carried out to grind and mix well with electron conductive agents in cathodes. 3.3. Raman spectra In order to check the structure of phosphate units, micro-Raman scattering spectroscopy was done and results are shown in Fig. 7. According to several literatures, Raman bands are assigned [23]. Main bands are observed from 950 to 1200 cm−1 those are corresponding to symmetric and asymmetric stretching modes of non-bridging oxygen on different P-tetrahedra, with a systematic decrease in peak frequency for P-O bonds on 3-dimensional phosphates (Q2), P2O7 (Q1) dimer and isolated PO4 (Q0). Raman bands at 740 cm−1 which is assigned as P-O-P symmetric stretching modes are also appeared. All samples have Q1 species that means polymerized phosphates are existing in glass matrix. The relative intensity of the peak assign to the Q1 decreased with increasing of FeO content. In addition, Q0 species (950 cm− 1) was appeared drastically at x = 40, and intensity was increase and as FeO content increases. Fig. 7. Micro-Raman spectra for quenched samples.
3.4. Electrochemical properties for sodium anode Na2FeP2O7 have density as 3.22 g·cm−3 at room temperature. On the other hand, corresponding glass (33.3FeO-66.7NaPO3) have 2.994 g·cm−3. In order to clarify the electrical conductive properties, AC impedance spectroscopy was examined in various samples. Fig. 5 shows the temperature dependence and compositional dependence for electrical conductivities those are determined by Cole-Cole plots. For instance, ColeCole plots of 45FeO-55NaPO3 at various temperature are also illustrated in Fig. 5. Most of all samples have only one semi-circle in various temperature. All series represent linear correlation between log σ versus 1/T. As can be seen in Fig. 6, the value of activation energy (Ea) for
As shown in Fig. 8, fine powder b10 μm was obtained after milling process. After milling and classification by sieving, the other sample also have same size distribution. And the results of charge and discharge profiles are shown in Fig. 9. In the sample for 20 ≤ x ≤ 45, both charge and discharge profiles are showing somewhat sloping plateau voltage around 2–3 V. Although an irreversible capacity at initial cycle for 40FeO-60NaPO3 and 45FeO-55NaPO3, but after second cycle reversible reaction progressed continuously. We guess such irreversible capacity is due to disembarrassment of sodium ion during ball milling process or fabrication of electrodes under atmosphere. When we look at
Fig. 8. Scanning electron microscope images of mixture of glass and carbon black after ball milling process.
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
113
Fig. 9. Charge and discharge profiles for x = 20, 40 45 and 50 in xFeO–(100-x)NaPO3 for sodium anode.
discharge capacity, it was increased with increase of FeO content up to x = 40. For the theoretical discharge capacity of glassy state, we assume that all FeO units in glass will participate electrochemical reaction. Fig. 10 shows initial discharge capacity as a function of FeO content in this study. Dashed line as theoretical capacity for one electron reaction is also indicated. Within x = 40, electrochemical reaction follows theoretical line, but actual capacity was dropped drastically in higher FeO content. Maricite NaFePO4 phase, which is inactive for cathode was formed in 50FeO-50NaPO3 instead of homogeneous glassy state. 4. Discussion As we mentioned low density of glassy state, glass network strongly dominated by linking phosphate units in Na2O-FeO-P2O5 glasses [19,23, 26]. These results are implying glass structure, in especially coordination of phosphate units, are different from corresponding crystal structure. According to Raman scattering spectra, most of all glass
structure is dominated by Q1 species in this study. Ma et al. examined structure of Na2O-FeO-Fe2O3-P2O5 glasses by means of Raman spectroscopy and Mössbauer spectroscopy [22]. They suggest that the interaction between phosphate polyanion and iron polyhedral causes systematic changes in the Raman frequencies as well as peak intensities. According to Mössbauer spectroscopy, the degree of Fe-O bond covalency in the Fe2+-O6 octahedral sites increases with increasing Fe/P ratio. The reduction of Fe3 + to Fe2 + will introduce pyrophosphates in the glass network. In this study all glass sample was prepared in strong reduction atmosphere. Therefore, results on Raman are suggesting that ferrous ion cause polymerization of phosphates. Although the primary Raman band is consisted with isolated (PO4)3−, (Q0) at 953 cm−1 as shown in Fig. 7, however the relative in1 tensity of Raman bands is observed due to pyrophosphate (PO2− 3.5 ,Q ) −1 structure at 1023, 1061 and 1128 cm . In theoretically, maricite NaFePO4 have no polymerized P2O7 units and each isolated PO4 units are intercepting sodium ion diffusion. The Raman results for 50FeO-50NaPO3 suppose existing residual glassy phase by quenching, hence Q1 bands are appeared. Therefore,
114
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
Fig. 10. Initial discharge capacity as a function of FeO content.
low density is caused by polymerization of phosphate and they enhance pathway of sodium ion migration in glass matrix like a schematic illustrates as shown in Fig. 11. Otherwise, clustering FeO6 and Na+ should be form at counterpart of phosphates, which is effective for electrochemical reaction as cathode. The addition of divalent Fe2 + into NaPO3, in which all phosphate units are consisted from long chain (Q 2 ) state, will break P-O-P bonds by chain-terminating due to increase of cation field strength. In finally, primary Qn state is shift from Q2, Q1 to completely isolated Q0 state with increases of x in xFeO–(100-x)NaPO3 in this study.
As shown in Fig. 9, charge and discharge reaction performed reversibly. And discharge capacity is almost comparative with theoretical values. To focus on x = 40, the amount of 40 mol% of Na+, which is being at neighbor of iron oxide octahedral site, participate into electrochemical reaction and residual 20 mol% of Na+ may exist as static network modifier near phosphate units in glass matrix. To improve further current density, it is important to activate static Na+ ions near the network modifier in glass. In our previous study, glass-ceramics for lithium/sodium ion battery cathodes exhibit good reversible electrochemical reaction for sodium anode even high current density. At that time, reduction agents was added such as glucose, citric acid and carboxyl methyl cellulose in the process of crystallization of glass matrix. Such carbon sources act as role of reduction agent of transition metal ions and preventing crystal growth in higher temperature [4,5,27]. In this study we just focus discharge capacity of pure glassy state without any carbon coating. It is hard to distinguish effect of crystallization and carbon coating process after crystallization. Crystallization occurs at 300 °C in the case of LiFePO4 glass-ceramics [28]. It is very low temperature to convert conductive carbon at 300 °C from such carbon sources. Therefore, the charge and discharge tests at higher current density were omitted. In order to characterize rate performance of amorphous state, the other coatings must be necessary. On the other hand, if we can control formation of preferable crystalline phase for cathode and hybridization with conducting carbon, the presence of residual glassy phase assumes important role as cathode active materials as well as inorganic oxide based binder to develop densified cathodes and it will be useful for all-solid batteries. 5. Conclusion In this study, glass formation tendency and cathode activity of Na2OFeO-P2O5 glasses are examined for sodium ion batteries. We succeeded to enhance reversible discharge capacity by increasing of FeO content in pure glass. We can say that from this results both of glassy phase and triclinic Na2FeP2O7 is also active for sodium cell. There are several
Fig. 11. Schematic images of local structure and sodium ion conduction pathway in glassy cathode and corresponding crystal.
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
advantages are proposed by using glass and glass-ceramics such as tunable thermal properties as well as discharge capacity with increases of FeO content. In especially, we would like to emphasize that glasses exhibit liquid-phase sintering before crystallization at low temperature. The ability of low-temperature sintering may be useful to construct all-solid battery in the future. Acknowledgement This work was supported by Grant-in-Aid for Scientific Research No. 25288105 from Japan Society for the Promotion of Science (JSPS). References [1] S. Komaba, W. Murata, T. Ishikawa, N. Yabuuchi, T. Ozeki, T. Nakayama, A. Ogata, K. Gotoh, K. Fujiwara, Electrochemical Na insertion and solid electrolyte interphase for hard-carbon electrodes and application to Na-ion batteries, Adv. Funct. Mater. 21 (2011) 3859–3867, http://dx.doi.org/10.1002/adfm.201100854. [2] N. Yabuuchi, K. Kubota, M. Dahbi, S. Komaba, Research development on sodium-ion batteries, Chem. Rev. 114 (2014) 11636–11682, http://dx.doi.org/10.1021/ cr500192f. [3] T. Honma, T. Togashi, N. Ito, T. Komatsu, Fabrication of Na2FeP2O7 glass-ceramics for sodium ion battery, J. Ceram. Soc. Jpn. 120 (2012) 344–346, http://dx.doi.org/10. 2109/jcersj2.120.344. [4] T. Honma, N. Ito, T. Togashi, A. Sato, T. Komatsu, Triclinic Na2 − xFe1 + x / 2P2O7/C glass-ceramics with high current density performance for sodium ion battery, J. Power Sources 227 (2013) 31–34, http://dx.doi.org/10.1016/j.jpowsour.2012.11. 030. [5] T. Honma, A. Sato, N. Ito, T. Togashi, K. Shinozaki, T. Komatsu, Crystallization behavior of sodium iron phosphate glass Na2 − xFe1 + 0.5xP2O7 for sodium ion batteries, J. Non-Cryst. Solids 404 (2014) 26–31, http://dx.doi.org/10.1016/j.jnoncrysol.2014. 07.028. [6] C.-Y. Chen, K. Matsumoto, T. Nohira, R. Hagiwara, Y. Orikasa, Y. Uchimoto, Pyrophosphate Na2FeP2O7 as a low-cost and high-performance positive electrode material for sodium secondary batteries utilizing an inorganic ionic liquid, J. Power Sources 246 (2014) 783–787, http://dx.doi.org/10.1016/j.jpowsour.2013.08.027. [7] P. Barpanda, T. Ye, S. Nishimura, S.-C. Chung, Y. Yamada, M. Okubo, H. Zhou, A. Yamada, Sodium iron pyrophosphate: A novel 3.0 V iron-based cathode for sodium-ion batteries, Electrochem. Commun. 24 (2012) 116–119, http://dx.doi.org/ 10.1016/j.elecom.2012.08.028. [8] Y.H. Jung, C.H. Lim, J.-H. Kim, D.K. Kim, Na2FeP2O7 as a positive electrode material for rechargeable aqueous sodium-ion batteries, RSC Adv. 4 (2014) 9799–9802, http:// dx.doi.org/10.1039/C3RA47560C. [9] T. Honma, K. Nagamine, T. Komatsu, Fabrication of olivine-type LiMnxFe1 − xPO4 crystals via the glass–ceramic route and their lithium ion battery performance, Ceram. Int. 36 (2010) 1137–1141, http://dx.doi.org/10.1016/j.ceramint.2009.10. 003. [10] K. Nagamine, T. Honma, T. Komatsu, A fast synthesis of Li3V2(PO4)3 crystals via glass-ceramic processing and their battery performance, J. Power Sources 196 (2011) 9618–9624, http://dx.doi.org/10.1016/j.jpowsour.2011.06.094. [11] T. Nagakane, H. Yamauchi, K. Yuki, M. Ohji, A. Sakamoto, T. Komatsu, T. Honma, M. Zou, G. Park, T. Sakai, Glass-ceramic LiFePO4 for lithium-ion rechargeable battery, Solid State Ionics 206 (2012) 78–83, http://dx.doi.org/10.1016/j.ssi.2011.10.017.
115
[12] H. Kim, I. Park, D.-H. Seo, S. Lee, S.-W. Kim, W.J. Kwon, Y.-U. Park, C.S. Kim, S. Jeon, K. Kang, New iron-based mixed-polyanion cathodes for lithium and sodium rechargeable batteries: combined first principles calculations and experimental study, J. Am. Chem. Soc. 134 (2012) 10369–10372, http://dx.doi.org/10.1021/ja3038646. [13] Y. Fang, Q. Liu, L. Xiao, X. Ai, H. Yang, Y. Cao, High-performance olivine NaFePO4 microsphere cathode synthesized by aqueous electrochemical displacement method for sodium ion batteries, ACS Appl. Mater. Interfaces 7 (2015) 17977–17984, http://dx.doi.org/10.1021/acsami.5b04691. [14] M. Tatsumisago, M. Nagao, A. Hayashi, Recent development of sulfide solid electrolytes and interfacial modification for all-solid-state rechargeable lithium batteries, J. Asian Ceram. Soc. 1 (2013) 17–25, http://dx.doi.org/10.1016/j.jascer.2013.03.005. [15] A. Hayashi, K. Noi, A. Sakuda, M. Tatsumisago, Superionic glass-ceramic electrolytes for room-temperature rechargeable sodium batteries, Nat. Commun. 3 (2012) 856, http://dx.doi.org/10.1038/ncomms1843. [16] S. Okada, T. Yamamoto, Y. Okazaki, J. Yamaki, M. Tokunaga, T. Nishida, Cathode properties of amorphous and crystalline FePO4, J. Power Sources 146 (2005) 570–574, http://dx.doi.org/10.1016/j.jpowsour.2005.03.200. [17] T. Shiratsuchi, S. Okada, J. Yamaki, T. Nishida, FePO4 cathode properties for Li and Na secondary cells, J. Power Sources 159 (2006) 268–271, http://dx.doi.org/10.1016/j. jpowsour.2006.04.047. [18] H. Yamauchi, G. Park, T. Nagakane, T. Honma, T. Komatsu, T. Sakai, A. Sakamoto, Performance of lithium-ion battery with tin-phosphate glass anode and its characteristics, J. Electrochem. Soc. 160 (2013) A1725–A1730, http://dx.doi.org/10.1149/2. 049310jes. [19] H. Kondo, T. Honma, T. Komatsu, Synthesis and morphology of metal Sn particles in SnO–P2O5 glasses and their battery anode performance, J. Non-Cryst. Solids 402 (2014) 153–159, http://dx.doi.org/10.1016/j.jnoncrysol.2014.05.034. [20] T. Honma, T. Togashi, H. Kondo, T. Komatsu, H. Yamauchi, A. Sakamoto, T. Sakai, Tinphosphate glass anode for sodium ion batteries, APL Mater. 1 (2013) 052101, http://dx.doi.org/10.1063/1.4826938. [21] International Glass Database System, INTERGLAD V.7, New Glass Forum, (n.d.). http://www.interglad.jp/interglad7/. [22] L. Ma, R.K. Brow, L. Ghussn, M.E. Schlesinger, Thermal stability of Na2O–FeO–Fe2O3– P2O5 glasses, J. Non-Cryst. Solids 409 (2015) 131–138, http://dx.doi.org/10.1016/j. jnoncrysol.2014.11.019. [23] L. Ma, R.K. Brow, A. Choudhury, Structural study of Na2O–FeO–Fe2O3–P2O5 glasses by Raman and Mössbauer spectroscopy, J. Non-Crystalline Solids 402 (2014) 64–73, http://dx.doi.org/10.1016/j.jnoncrysol.2014.05.013. [24] H. Kim, I. Park, D. Seo, S. Lee, S. Kim, W.J. Kwon, Y. Park, C.S. Kim, S. Jeon, K. Kang, New Iron-Based Mixed-Polyanion Cathodes for Lithium and Sodium Rechargeable Batteries: Combined First Principles Calculations and Experimental Study, 3, 2012 5–8. [25] C. Masquelier, C. Wurm, J. Rodríguez-Carvajal, J. Gaubicher, L. Nazar, A powder neutron diffraction investigation of the two rhombohedral NASICON analogues: γNa3Fe2(PO4)3 and Li3Fe2(PO4)3, Chem. Mater. 12 (2000) 525–532, http://dx.doi. org/10.1021/cm991138n. [26] L. Zhang, R.K. Brow, A Raman study of iron-phosphate crystalline compounds and glasses, J. Am. Ceram. Soc. 94 (2011) 3123–3130, http://dx.doi.org/10.1111/j. 1551-2916.2011.04486.x. [27] K. Nagamine, T. Honma, T. Komatsu, Fabrication of LiVOPO4/carbon composite via glass-ceramic processing, IOP Conf. Ser. Mater. Sci. Eng.21, 2011, p. 012021, http://dx.doi.org/10.1088/1757-899X/21/1/012021. [28] K. Nagamine, K. Oh-Ishi, T. Honma, T. Komatsu, Formation mechanism of LiFePO4 in crystallization of lithium iron phosphate glass particles, J. Ceram. Soc. Jpn. 120 (2012) 193–198, http://dx.doi.org/10.2109/jcersj2.120.193.
Contents lists available at ScienceDirect
Journal of Non-Crystalline Solids journal homepage: www.elsevier.com/locate/jnoncrysol
Cathode properties of sodium iron phosphate glass for sodium ion batteries Satoshi Nakata, Takuya Togashi, Tsuyoshi Honma ⁎, Takayuki Komatsu Department of Materials Science and Technology, Nagaoka University of Technology, Kamitomioka-cho 1603-1, Nagaoka, Niigata 940-2188, Japan
a r t i c l e
i n f o
Article history: Received 27 May 2016 Received in revised form 21 July 2016 Accepted 3 August 2016 Available online xxxx Keywords: Sodium ion batteries Sodium iron phosphate Electrical conductivity Electrochemical properties
a b s t r a c t In order to clarify the possibility of cathode activity of Na2O-FeO-P2O5 glasses for sodium ion batteries, glass formation tendency, crystallization behavior, electrical conductivity as well as charge and discharge properties are examined. Glass formation in xFeO–(100-x)NaPO3 was confirmed for x ≤ 45 by melt quenching up to 1000 °C in N2 filled electric furnace. By means of electrical conductivity tests, conductivity was increased with increases of x content in xFeO–(100-x)NaPO3 and the activation energy for electrical conductivity was decreased. Charge and discharge profiles for Na anode exhibits high reversible discharge capacity as 115 mAh/g at 0.1 °C rate in 40FeO-60NaPO3 (30Na2O–40FeO–30P2O5) glass. The results on Raman scattering spectra suggests that glass structure is subjected to polymerized Q1 and Q2 phosphate units. And polymerized phosphates in glass cause lower density, open structure and sodium ion diffusive channels as well. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Li-ion batteries (LIB) have been intensively developed after the first launch of the Li-ion batteries in 1990s. LIBs are essential for the development of smart phones, laptop computers, and many other consumer products these days. The next targets of LIBs are considered to be automotive applications and huge energy storage systems. The materials abundance is of the primary importance to design the electrode materials for such large-scale applications. Sodium ion batteries are focused as alternative secondary batteries to save material cost recently [1]. Due to the intensive research, the energy density of the sodium ion battery is increasingly become equal to that of conventional LIBs [2]. In cathode, layer rock salt type oxides as well as poly-anion type materials are considered as active materials in recently. Triclinic P1 Na2FeP2O7 exhibits 3.0 V, 97 mAh/g with good cyclic performance [3–7]. Na2FeP2O7 is also available in aqueous based sodium ion batteries that implies chemical stability is good [8]. The authors group is proposing unique technique to produce Na2FeP2O7 by crystallization of precursor glass so-called ‘glass-ceramic process’, which is also applicable to fabricate LiFePO4, LiMnxFe1 − xPO4 and Li3V2(PO4)3 as well in our previous study [9–11]. By means of transmission electron microscope observation, there are residual amorphous phase on the surface of glass-ceramic grain. Nagakane et al. [11] are proposing that residual amorphous is effective to assist ionic conduction in LiFePO4 cathode, which exists only onedimensional Li+ diffusive channel. On the other hand, to improve ⁎ Corresponding author. E-mail address: [email protected] (T. Honma).
http://dx.doi.org/10.1016/j.jnoncrysol.2016.08.005 0022-3093/© 2016 Elsevier B.V. All rights reserved.
capacity of materials is another important point to achieve high energy density. As shown in Fig. 1 there are five crystals are found in ternary NaO0.5-FeOx-PO2.5 systems. To focus on sodium pre-doped Fe2 + base crystal, there are three candidates, Na2FeP2O7, Na4Fe3(PO4)2P2O7 and maricite type NaFePO4 are remained [12,13]. Comparing with theoretical capacity, unfortunately, NaFePO4, which is stoichiometric composition of olivine type LiFePO4, is inactive for sodium cell cathodes. Isolated PO4 units are exist periodically and they are blocking Na+ ion diffusion in maricite structure. On the other hand, the amorphous material like glass has the random three-dimensional structure. It is known that glass have large free volume and flexible open structure, hence superior alkali ion conduction exhibits in glass-ceramic derived solid state electrolytes for lithium ion batteries and sodium ion batteries as well [14,15]. It is not unique to solid electrolyte, there are possibility to develop for cathode and anode active materials. Okada et al. suggests that mechano-chemical derived amorphous FePO4 have a corner-shared matrix, and they showed similarly good capacity, not only for Li but also for Na anodes [16,17]. SnO-P2O5 glass is also available as anode active materials in Li and Na cell [18–20]. It would be interesting to examine the nature of ionic conduction of glassy state cathode materials such as Na2O-FeOP2O5 system. In this study, glass formation tendency, electrical conductivity and cathode activities of sodium iron phosphate glasses are examined for sodium ion batteries. 2. Experimental procedure The glass composition examined in this study is xFeO–(100-x)NaPO3 (x = 20, 25, 33.3, 40, 45 and 50). Fig. 1 shows glass formation region
110
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
Fig. 1. Glass formation region and examined glass compositions in NaO0.5-FeO1 + δ-PO2.5 systems. We also added typical crystals ever reported.
which is determined by INTERGLAD (glass database) with light blue [21] and illustrated sample composition as red mark. In these glasses there are three stoichiometric compositions of Na2FeP2O7 (x = 33.3), Na3Fe2(PO4)3 (x = 40) and NaFePO4 (x = 50). The fraction of Fe2 + ions in an Na2O-Fe2O3-P2O5 glass varies with the initial batch materials and preparation conditions, initial fraction of raw materials. In generally, the fraction of Fe2+ increase when ferrous raw materials are used such as FeO, Fe3O4 rather than Fe2O3 and when melts are processed under reducing conditions [22,23]. When the melting is process under atmosphere the fraction of Fe2 + is about up to 20%. In this study we are expecting to develop Na+ pre-doped cathode, therefore the fraction of Fe2+ must be high. Glasses were fabricated by a conventional meltquenching method under reducing atmosphere. Starting reagents
Fig. 2. Powder XRD patterns of xFeO–(100-x)NaPO3 (x = 20, 25, 33.3, 40, 45 and 50).
NaH2PO4 (99%, Nakarai tesque Co.), FeO (99.9%, Kojyundo chemicals Co.) were mixed well. A 20 g batch was melted in a gold crucible at 900 °C for 15 min in flowing N2 gas (5 L/min) in muffle furnace. The melts were poured onto an iron plate and pressed to a thickness of 0.5–1 mm by another iron plate. The glass transition and crystallization temperatures were determined by differential thermal analysis (DTA, Rigaku TG-8120). In order to confirm glass formation and to characterize crystallized phase, XRD patterns of all samples were obtained on Rigaku Ultima IV X-ray diffractometer (Rigaku, Japan) with D/tex 1D high-speed detector, which was operated at 40 kV, 40 mA with Cu-Kα radiation (λ = 0.154056 nm). All the measurements were carried out at room temperature under atmospheric air. Raman spectra were recorded using a Nanofinder (Tokyo Instruments, Japan) confocal Raman with Ar laser beam having a wavelength of 488 nm using a CCD detector. Electrical conductivities of glasses were measured by an alternating current (AC) impedance method (HIOKI 3522-50 LCR HiTESTER, Japan) in the temperature range from room temperature to 200 °C. Metal gold was sputtered with 6.5 mmϕ to the glass surface as electrodes (ULVAC QUICK COATER VPS-020, Japan). Pre-pulverized glass flake obtained by automatic mortar was subjected to mechanical milling in a planetary ball mill (Fritsch Premium line Pulversette No. 7). The following milling conditions were used: air atmosphere; milling speed of 600 rpm; ball to powder mass ratio of 10:1; and milling time of 1 h. In order to avoid excessive temperature rising within the grinding chamber, 15 min of ball milling duration was followed by a pause of 5 min. Composite powder was classified by sieving to the size of b 15 μm. Morphologies of glass-ceramics/carbon composites were observed by scanning electron microscope (SEM, Keyence VE-8800). Cathode electrodes were fabricated from a mixture of glass powder, polyvinylidene fluoride (PVDF) and conductive carbon black in a weight ratio of 85:5:10. N-methylpyrrolidone (NMP) was used to make slurry
Fig. 3. DTA profiles of quenched samples of xFeO–(100-x)NaPO3 (x = 20, 25, 33.3, 40, 45 and 50).
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
111
electrolyte. Cells were examined by using a battery testing system (Hokuto-denko Co.) at the charge/discharge current density of 1/100 ° C (0.08 mA cm−2) or 1/10 °C for the theoretical capacity for one-electron reaction between 2.0 and 3.8 V. 3. Results 3.1. Glass formation tendency
Fig. 4. Compositional dependence of bulk density (circle) at room temperature. We also added theoretical density (triangle).
of their mixtures. After homogenization, slurry was coated on a thin aluminum foil and dried at 100 °C for 10 h in a vacuum oven. Electrodes were then pressed and disks were punched out as 16 mmϕ. Electrochemical cells were prepared using coin type cells. Sodium metal foils were used as anode, and glass filter papers (Advantec Co., GA-100) were used as separator. Test cells were assembled in an argon-filled glove box. The dew point of Ar atmosphere in the glove box was kept as − 86 °C. The oxygen content was b0.33 ppm. The solution of 1 MNaPF6 (Tokyo Kasei Co.) in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1, v/v, Kishida Chemicals Co.) was used as
Fig. 5. Temperature dependence of electrical conductivity for vitrified samples. Cole-Cole plot at 336-369 K for 45FeO-55NaPO3 is also shown.
Fig. 2 shows the powder XRD patterns and appearance for meltquenched samples. Amber colored transparent bulk glass was obtained in 20FeO-80NaPO3 and 25FeO-75NaPO3. For x = 33.3, 40 and 45 in xFeO–(100-x)NaPO3, samples exhibit black colored opaque glass was obtained as can be seen in Fig. 2. Devitrification of 50FeO-50NaPO3 which is stoichiometric composition of NaFePO4 was confirmed. Only a halo pattern without any sharp peak was observed for 20 ≤ x ≤ 45, indicating that the melt-quenched samples with the compositions of xFeO–(100-x)NaPO3 (mol%) are amorphous (glasses) state. Meanwhile devitrified glass-ceramics which is composed of maricite type NaFePO4 (ICDD #01-089-0816) was obtained in 50FeO-50NaPO3. The DTA patterns for the as-quenched sample are shown in Fig. 3. The endothermic dips due to the glass transition and exothermic peaks due to the crystallization are clearly observed. The values of the glass transition (Tg) and crystallization peak (Tp) are also indicated in Fig. 3. For 20 ≤ x ≤ 25, Tp was increased. Primary crystallized phase was confirmed as NaPO3 in 20FeO-80NaPO3 by means of XRD. On the other hand, crystallized phase changed from NaPO3 to Na2FeP2O7 when FeO content increase from x = 20 to 25. For the other samples, both values of Tg and Tp was decreased with increases of FeO content in xFeO–(100-x)NaPO3 glasses. In addition, thermal stability criterion for crystallization ΔT = Tp − Tg was drastically attenuated by increases of FeO content. 3.2. Fundamental physical properties Bulk density of glass was determined by Archimedes method at room temperature to compare with that of corresponding crystal and results are shown in Fig. 4. As can be seen in Fig. 4, it is also indicated theoretical density which is calculated from crystal structure with triangular mark [7,24,25]. The density of glasses is monotonically increasing as increases of FeO content. Furthermore, densities of all glass samples are less than those of corresponding crystals. For instance, crystalline
Fig. 6. Compositional dependence of electrical conductivity at room temperature and activation energy for electrical conduction.
112
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
electrical conduction was decreased from 0.8 to 0.64 with increase of FeO content. The increasing of FeO content is effective not only enhance energy density but also electrical conductivity in such glasses. However, the magnitudes of 10−12–10−10 Scm−1 is not enough to offer smooth electrochemical reaction during charge and discharge process. Therefore, planetary ball milling was carried out to grind and mix well with electron conductive agents in cathodes. 3.3. Raman spectra In order to check the structure of phosphate units, micro-Raman scattering spectroscopy was done and results are shown in Fig. 7. According to several literatures, Raman bands are assigned [23]. Main bands are observed from 950 to 1200 cm−1 those are corresponding to symmetric and asymmetric stretching modes of non-bridging oxygen on different P-tetrahedra, with a systematic decrease in peak frequency for P-O bonds on 3-dimensional phosphates (Q2), P2O7 (Q1) dimer and isolated PO4 (Q0). Raman bands at 740 cm−1 which is assigned as P-O-P symmetric stretching modes are also appeared. All samples have Q1 species that means polymerized phosphates are existing in glass matrix. The relative intensity of the peak assign to the Q1 decreased with increasing of FeO content. In addition, Q0 species (950 cm− 1) was appeared drastically at x = 40, and intensity was increase and as FeO content increases. Fig. 7. Micro-Raman spectra for quenched samples.
3.4. Electrochemical properties for sodium anode Na2FeP2O7 have density as 3.22 g·cm−3 at room temperature. On the other hand, corresponding glass (33.3FeO-66.7NaPO3) have 2.994 g·cm−3. In order to clarify the electrical conductive properties, AC impedance spectroscopy was examined in various samples. Fig. 5 shows the temperature dependence and compositional dependence for electrical conductivities those are determined by Cole-Cole plots. For instance, ColeCole plots of 45FeO-55NaPO3 at various temperature are also illustrated in Fig. 5. Most of all samples have only one semi-circle in various temperature. All series represent linear correlation between log σ versus 1/T. As can be seen in Fig. 6, the value of activation energy (Ea) for
As shown in Fig. 8, fine powder b10 μm was obtained after milling process. After milling and classification by sieving, the other sample also have same size distribution. And the results of charge and discharge profiles are shown in Fig. 9. In the sample for 20 ≤ x ≤ 45, both charge and discharge profiles are showing somewhat sloping plateau voltage around 2–3 V. Although an irreversible capacity at initial cycle for 40FeO-60NaPO3 and 45FeO-55NaPO3, but after second cycle reversible reaction progressed continuously. We guess such irreversible capacity is due to disembarrassment of sodium ion during ball milling process or fabrication of electrodes under atmosphere. When we look at
Fig. 8. Scanning electron microscope images of mixture of glass and carbon black after ball milling process.
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
113
Fig. 9. Charge and discharge profiles for x = 20, 40 45 and 50 in xFeO–(100-x)NaPO3 for sodium anode.
discharge capacity, it was increased with increase of FeO content up to x = 40. For the theoretical discharge capacity of glassy state, we assume that all FeO units in glass will participate electrochemical reaction. Fig. 10 shows initial discharge capacity as a function of FeO content in this study. Dashed line as theoretical capacity for one electron reaction is also indicated. Within x = 40, electrochemical reaction follows theoretical line, but actual capacity was dropped drastically in higher FeO content. Maricite NaFePO4 phase, which is inactive for cathode was formed in 50FeO-50NaPO3 instead of homogeneous glassy state. 4. Discussion As we mentioned low density of glassy state, glass network strongly dominated by linking phosphate units in Na2O-FeO-P2O5 glasses [19,23, 26]. These results are implying glass structure, in especially coordination of phosphate units, are different from corresponding crystal structure. According to Raman scattering spectra, most of all glass
structure is dominated by Q1 species in this study. Ma et al. examined structure of Na2O-FeO-Fe2O3-P2O5 glasses by means of Raman spectroscopy and Mössbauer spectroscopy [22]. They suggest that the interaction between phosphate polyanion and iron polyhedral causes systematic changes in the Raman frequencies as well as peak intensities. According to Mössbauer spectroscopy, the degree of Fe-O bond covalency in the Fe2+-O6 octahedral sites increases with increasing Fe/P ratio. The reduction of Fe3 + to Fe2 + will introduce pyrophosphates in the glass network. In this study all glass sample was prepared in strong reduction atmosphere. Therefore, results on Raman are suggesting that ferrous ion cause polymerization of phosphates. Although the primary Raman band is consisted with isolated (PO4)3−, (Q0) at 953 cm−1 as shown in Fig. 7, however the relative in1 tensity of Raman bands is observed due to pyrophosphate (PO2− 3.5 ,Q ) −1 structure at 1023, 1061 and 1128 cm . In theoretically, maricite NaFePO4 have no polymerized P2O7 units and each isolated PO4 units are intercepting sodium ion diffusion. The Raman results for 50FeO-50NaPO3 suppose existing residual glassy phase by quenching, hence Q1 bands are appeared. Therefore,
114
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
Fig. 10. Initial discharge capacity as a function of FeO content.
low density is caused by polymerization of phosphate and they enhance pathway of sodium ion migration in glass matrix like a schematic illustrates as shown in Fig. 11. Otherwise, clustering FeO6 and Na+ should be form at counterpart of phosphates, which is effective for electrochemical reaction as cathode. The addition of divalent Fe2 + into NaPO3, in which all phosphate units are consisted from long chain (Q 2 ) state, will break P-O-P bonds by chain-terminating due to increase of cation field strength. In finally, primary Qn state is shift from Q2, Q1 to completely isolated Q0 state with increases of x in xFeO–(100-x)NaPO3 in this study.
As shown in Fig. 9, charge and discharge reaction performed reversibly. And discharge capacity is almost comparative with theoretical values. To focus on x = 40, the amount of 40 mol% of Na+, which is being at neighbor of iron oxide octahedral site, participate into electrochemical reaction and residual 20 mol% of Na+ may exist as static network modifier near phosphate units in glass matrix. To improve further current density, it is important to activate static Na+ ions near the network modifier in glass. In our previous study, glass-ceramics for lithium/sodium ion battery cathodes exhibit good reversible electrochemical reaction for sodium anode even high current density. At that time, reduction agents was added such as glucose, citric acid and carboxyl methyl cellulose in the process of crystallization of glass matrix. Such carbon sources act as role of reduction agent of transition metal ions and preventing crystal growth in higher temperature [4,5,27]. In this study we just focus discharge capacity of pure glassy state without any carbon coating. It is hard to distinguish effect of crystallization and carbon coating process after crystallization. Crystallization occurs at 300 °C in the case of LiFePO4 glass-ceramics [28]. It is very low temperature to convert conductive carbon at 300 °C from such carbon sources. Therefore, the charge and discharge tests at higher current density were omitted. In order to characterize rate performance of amorphous state, the other coatings must be necessary. On the other hand, if we can control formation of preferable crystalline phase for cathode and hybridization with conducting carbon, the presence of residual glassy phase assumes important role as cathode active materials as well as inorganic oxide based binder to develop densified cathodes and it will be useful for all-solid batteries. 5. Conclusion In this study, glass formation tendency and cathode activity of Na2OFeO-P2O5 glasses are examined for sodium ion batteries. We succeeded to enhance reversible discharge capacity by increasing of FeO content in pure glass. We can say that from this results both of glassy phase and triclinic Na2FeP2O7 is also active for sodium cell. There are several
Fig. 11. Schematic images of local structure and sodium ion conduction pathway in glassy cathode and corresponding crystal.
S. Nakata et al. / Journal of Non-Crystalline Solids 450 (2016) 109–115
advantages are proposed by using glass and glass-ceramics such as tunable thermal properties as well as discharge capacity with increases of FeO content. In especially, we would like to emphasize that glasses exhibit liquid-phase sintering before crystallization at low temperature. The ability of low-temperature sintering may be useful to construct all-solid battery in the future. Acknowledgement This work was supported by Grant-in-Aid for Scientific Research No. 25288105 from Japan Society for the Promotion of Science (JSPS). References [1] S. Komaba, W. Murata, T. Ishikawa, N. Yabuuchi, T. Ozeki, T. Nakayama, A. Ogata, K. Gotoh, K. Fujiwara, Electrochemical Na insertion and solid electrolyte interphase for hard-carbon electrodes and application to Na-ion batteries, Adv. Funct. Mater. 21 (2011) 3859–3867, http://dx.doi.org/10.1002/adfm.201100854. [2] N. Yabuuchi, K. Kubota, M. Dahbi, S. Komaba, Research development on sodium-ion batteries, Chem. Rev. 114 (2014) 11636–11682, http://dx.doi.org/10.1021/ cr500192f. [3] T. Honma, T. Togashi, N. Ito, T. Komatsu, Fabrication of Na2FeP2O7 glass-ceramics for sodium ion battery, J. Ceram. Soc. Jpn. 120 (2012) 344–346, http://dx.doi.org/10. 2109/jcersj2.120.344. [4] T. Honma, N. Ito, T. Togashi, A. Sato, T. Komatsu, Triclinic Na2 − xFe1 + x / 2P2O7/C glass-ceramics with high current density performance for sodium ion battery, J. Power Sources 227 (2013) 31–34, http://dx.doi.org/10.1016/j.jpowsour.2012.11. 030. [5] T. Honma, A. Sato, N. Ito, T. Togashi, K. Shinozaki, T. Komatsu, Crystallization behavior of sodium iron phosphate glass Na2 − xFe1 + 0.5xP2O7 for sodium ion batteries, J. Non-Cryst. Solids 404 (2014) 26–31, http://dx.doi.org/10.1016/j.jnoncrysol.2014. 07.028. [6] C.-Y. Chen, K. Matsumoto, T. Nohira, R. Hagiwara, Y. Orikasa, Y. Uchimoto, Pyrophosphate Na2FeP2O7 as a low-cost and high-performance positive electrode material for sodium secondary batteries utilizing an inorganic ionic liquid, J. Power Sources 246 (2014) 783–787, http://dx.doi.org/10.1016/j.jpowsour.2013.08.027. [7] P. Barpanda, T. Ye, S. Nishimura, S.-C. Chung, Y. Yamada, M. Okubo, H. Zhou, A. Yamada, Sodium iron pyrophosphate: A novel 3.0 V iron-based cathode for sodium-ion batteries, Electrochem. Commun. 24 (2012) 116–119, http://dx.doi.org/ 10.1016/j.elecom.2012.08.028. [8] Y.H. Jung, C.H. Lim, J.-H. Kim, D.K. Kim, Na2FeP2O7 as a positive electrode material for rechargeable aqueous sodium-ion batteries, RSC Adv. 4 (2014) 9799–9802, http:// dx.doi.org/10.1039/C3RA47560C. [9] T. Honma, K. Nagamine, T. Komatsu, Fabrication of olivine-type LiMnxFe1 − xPO4 crystals via the glass–ceramic route and their lithium ion battery performance, Ceram. Int. 36 (2010) 1137–1141, http://dx.doi.org/10.1016/j.ceramint.2009.10. 003. [10] K. Nagamine, T. Honma, T. Komatsu, A fast synthesis of Li3V2(PO4)3 crystals via glass-ceramic processing and their battery performance, J. Power Sources 196 (2011) 9618–9624, http://dx.doi.org/10.1016/j.jpowsour.2011.06.094. [11] T. Nagakane, H. Yamauchi, K. Yuki, M. Ohji, A. Sakamoto, T. Komatsu, T. Honma, M. Zou, G. Park, T. Sakai, Glass-ceramic LiFePO4 for lithium-ion rechargeable battery, Solid State Ionics 206 (2012) 78–83, http://dx.doi.org/10.1016/j.ssi.2011.10.017.
115
[12] H. Kim, I. Park, D.-H. Seo, S. Lee, S.-W. Kim, W.J. Kwon, Y.-U. Park, C.S. Kim, S. Jeon, K. Kang, New iron-based mixed-polyanion cathodes for lithium and sodium rechargeable batteries: combined first principles calculations and experimental study, J. Am. Chem. Soc. 134 (2012) 10369–10372, http://dx.doi.org/10.1021/ja3038646. [13] Y. Fang, Q. Liu, L. Xiao, X. Ai, H. Yang, Y. Cao, High-performance olivine NaFePO4 microsphere cathode synthesized by aqueous electrochemical displacement method for sodium ion batteries, ACS Appl. Mater. Interfaces 7 (2015) 17977–17984, http://dx.doi.org/10.1021/acsami.5b04691. [14] M. Tatsumisago, M. Nagao, A. Hayashi, Recent development of sulfide solid electrolytes and interfacial modification for all-solid-state rechargeable lithium batteries, J. Asian Ceram. Soc. 1 (2013) 17–25, http://dx.doi.org/10.1016/j.jascer.2013.03.005. [15] A. Hayashi, K. Noi, A. Sakuda, M. Tatsumisago, Superionic glass-ceramic electrolytes for room-temperature rechargeable sodium batteries, Nat. Commun. 3 (2012) 856, http://dx.doi.org/10.1038/ncomms1843. [16] S. Okada, T. Yamamoto, Y. Okazaki, J. Yamaki, M. Tokunaga, T. Nishida, Cathode properties of amorphous and crystalline FePO4, J. Power Sources 146 (2005) 570–574, http://dx.doi.org/10.1016/j.jpowsour.2005.03.200. [17] T. Shiratsuchi, S. Okada, J. Yamaki, T. Nishida, FePO4 cathode properties for Li and Na secondary cells, J. Power Sources 159 (2006) 268–271, http://dx.doi.org/10.1016/j. jpowsour.2006.04.047. [18] H. Yamauchi, G. Park, T. Nagakane, T. Honma, T. Komatsu, T. Sakai, A. Sakamoto, Performance of lithium-ion battery with tin-phosphate glass anode and its characteristics, J. Electrochem. Soc. 160 (2013) A1725–A1730, http://dx.doi.org/10.1149/2. 049310jes. [19] H. Kondo, T. Honma, T. Komatsu, Synthesis and morphology of metal Sn particles in SnO–P2O5 glasses and their battery anode performance, J. Non-Cryst. Solids 402 (2014) 153–159, http://dx.doi.org/10.1016/j.jnoncrysol.2014.05.034. [20] T. Honma, T. Togashi, H. Kondo, T. Komatsu, H. Yamauchi, A. Sakamoto, T. Sakai, Tinphosphate glass anode for sodium ion batteries, APL Mater. 1 (2013) 052101, http://dx.doi.org/10.1063/1.4826938. [21] International Glass Database System, INTERGLAD V.7, New Glass Forum, (n.d.). http://www.interglad.jp/interglad7/. [22] L. Ma, R.K. Brow, L. Ghussn, M.E. Schlesinger, Thermal stability of Na2O–FeO–Fe2O3– P2O5 glasses, J. Non-Cryst. Solids 409 (2015) 131–138, http://dx.doi.org/10.1016/j. jnoncrysol.2014.11.019. [23] L. Ma, R.K. Brow, A. Choudhury, Structural study of Na2O–FeO–Fe2O3–P2O5 glasses by Raman and Mössbauer spectroscopy, J. Non-Crystalline Solids 402 (2014) 64–73, http://dx.doi.org/10.1016/j.jnoncrysol.2014.05.013. [24] H. Kim, I. Park, D. Seo, S. Lee, S. Kim, W.J. Kwon, Y. Park, C.S. Kim, S. Jeon, K. Kang, New Iron-Based Mixed-Polyanion Cathodes for Lithium and Sodium Rechargeable Batteries: Combined First Principles Calculations and Experimental Study, 3, 2012 5–8. [25] C. Masquelier, C. Wurm, J. Rodríguez-Carvajal, J. Gaubicher, L. Nazar, A powder neutron diffraction investigation of the two rhombohedral NASICON analogues: γNa3Fe2(PO4)3 and Li3Fe2(PO4)3, Chem. Mater. 12 (2000) 525–532, http://dx.doi. org/10.1021/cm991138n. [26] L. Zhang, R.K. Brow, A Raman study of iron-phosphate crystalline compounds and glasses, J. Am. Ceram. Soc. 94 (2011) 3123–3130, http://dx.doi.org/10.1111/j. 1551-2916.2011.04486.x. [27] K. Nagamine, T. Honma, T. Komatsu, Fabrication of LiVOPO4/carbon composite via glass-ceramic processing, IOP Conf. Ser. Mater. Sci. Eng.21, 2011, p. 012021, http://dx.doi.org/10.1088/1757-899X/21/1/012021. [28] K. Nagamine, K. Oh-Ishi, T. Honma, T. Komatsu, Formation mechanism of LiFePO4 in crystallization of lithium iron phosphate glass particles, J. Ceram. Soc. Jpn. 120 (2012) 193–198, http://dx.doi.org/10.2109/jcersj2.120.193.