Development of Sodium-Ion Conducting Crystalline and Glassy Electrolytes in the Systems Na_2S-M_xS_y (M = Sb, P, B)

概要

For realizing a low-carbon society, clean storages of energy based on various technologies are available [1–5]. Especially, lithium-ion batteries are expanding due to their high energy densities, weight reduction, and miniaturization since the basic concept was developed in 1980s [6–8]. However, safety risks of leakage and explosions in lithium-ion batteries with organic liquid electrolytes are problems to be addressed [9, 10]. There is a need to develop next-generation batteries with higher safety standards.

To solve the issues, all-solid-state batteries which use non-flammable inorganic solid electrolytes are at the center of global attention as safer next-generation batteries [11−15]. Recently, sodium-ion batteries with their associated low costs and abundant sodium reserves have attracted significant attention. Because sodium has a similar standard reduction potential to lithium, sodium batteries are expected to exhibit high performance as batteries using low cost materials [16−18]. Moreover, sodium conducting solid electrolytes will show higher ionic conductivities than lithium-ion conducting solid electrolytes because of low charge density of sodium ions due to larger ionic radius than lithium ions. As one example of sodium-ion conducting electrolytes, β”-alumina (general formula; Na2O·xAl2O3 (5 ≤ x ≤ 7) [19, 20]) is used for sodium–sulfur (NAS) batteries, and they function as a stationary storage system with a high energy density and cost merit due to use sulfur as the electrode [21]. However, they require a high temperature of over 300 °C to operate with molten electrodes to prepare close interfaces of the sintered β”-alumina electrolyte and sodium/sulfur active materials. Because the operating environment at high temperatures causes cost and safety concerns, all-solid-state sodium-ion batteries operating at room temperature are desirable. To solve the issue, superior solid electrolytes with high ionic conductivities at room temperature and high formability are required for achieving large contact areas with electrode active materials.

In general, solid electrolytes are mainly classified into oxide- and sulfide-based electrolytes. The oxide-based electrolytes, such as β-alumina (general formula; Na2O·xAl2O3 (9 ≤ x ≤ 11)) [19, 20, 22−24] and NASICON (Na Super Ionic Conductor, general formula; Na1+xZr2SixP3–xO12) [25−27], have been reported and their ionic conductivities are over 10–3 S cm–1 at 25 °C. These electrolytes have relatively high air stabilities. However, they need a high temperature process at over 1000 °C to obtain sintered bodies and reduce the grain boundary resistance.

Sulfide-based solid electrolytes have favorite formability and their densification is achieved easily just by a cold-pressing process at room temperature [28, 29]. Due to this feature, large contact areas between electrode active materials and solid electrolytes are achieved in all-solid-state batteries. Moreover, sulfide-based solid electrolytes generally exhibit higher ionic conductivities than oxide-based solid electrolytes because of the higher polarizability of sulfur compared to oxygen. One example of recent development is the discovery of the Na3PS4 solid electrolyte with the metastable cubic phase in 2012, which is prepared via a mechanochemical process and consecutive heat treatment [30]. The ionic conductivity of the Na3PS4 solid electrolyte with the cubic phase is 2×10–4 S cm–1, which is higher than those of the crystalline Na3PS4 with the tetragonal phase (1×10–6 S cm–1) [31] or the crystalline Na3PO4 (3×10–11 S cm–1) [32] at 25 °C. In 2014, the conductivity of the cubic Na3PS4 solid electrolyte was enhanced to 4.6×10–4 S cm–1 by using crystalline Na2S as a starting reagent with high purity of over 99% [33]. All-solid-state Na15Sn4 / TiS2 cells with the Na3PS4 solid electrolytes successfully operated as a rechargeable battery at room temperature [30]. Moreover, the Na3PS4 electrolyte experimentally possesses a higher formability than the Li3PS4 one, and the densification and associated sintering behavior are easily promoted in Na3PS4 by pressing at room temperature [34].

To improve the ionic conductivities of these sulfide-based solid electrolytes, aliovalent doping is one of the effective strategies. For Na3PS4-based solid electrolytes with tetrahedral structural units PS 3–, both the compositions of cation-substitution and anion-substitution were reported after developing the cubic Na3PS4 solid electrolyte [35–54]. For example, Na3PS4–xSex [35–37], Na3 P1– x AsxS4 [38, 39], (100–x)Na3PS4·xNa4SiS4 (mol%) [40, 41], Na3–xPS4–xClx [42, 43], Na2–xCaxPS4 [44], Na3+5xP1–xS4 [45], Na10SnP2S12-type [46–53] and Na3–xP1–xWxS4 [54] electrolytes were reported and some of them exhibited high ionic conductivities of over 10–3 S cm–1 at 25 °C as listed in Table 1. In 2016, tetragonal Na3SbS4 solid electrolytes, instead of Na3PS4, were reported to show a high ionic conductivity of over 10–3 S cm–1 [55–57].

In general, sulfide-based solid electrolytes have high formability and ionic conductivities, but most of them generate toxic H2S gas in a humid atmosphere. In contrast, Na3SbS4 generates negligible H2S gas because it forms a hydrate, Na3SbS4·9H2O [56, 57]. This tendency can be explained by the hard and soft acid and base (HSAB) theory, where hard acids prefer binding to hard bases, whereas soft acids prefer binding with soft bases to give covalent complexes. Because Sb–S bonding (a soft acid–soft base combination) is more stable than Sb–O bonding (a soft acid–hard base combination), the SbS43– unit is hardly hydrolyzed by H2O molecules in a humid atmosphere [56, 57]. After the discovery of tetragonal Na3SbS4, there have been a number of efforts at further enhancing the electrochemical properties of Na3SbS4 systems, and element substitution for Na3SbS4 is effective for improving the ionic conductivities [51, 54, 58–70]. Moreover, Tian et al. have reported that the highly reactive interface between Na3SbS4 and Na metal was partially stabilized via electrolyte hydration achieved by exposing the electrolyte pellet to ambient air [71]. The Na3SbS4 system is thus attractive as a solid electrolyte, and the relationship between structure and properties is needed to understand the electrolyte in detail.

To develop sulfide-based solid electrolytes, new approaches from wider perspective are needed. One of the approaches is the development the glassy solid electrolytes. In general, the glassy electrolytes with free volume show higher ionic conductivities and more superior formability than crystalline solid electrolytes. Moreover, glass is useful as a precursor to obtain metastable crystalline phases which are difficult to be synthesized by conventional solid-state reaction. Some of crystalline electrolytes with metastable phases show higher ionic conductivities than those with thermodynamically stable phases. Thus, the formation of metastable phases from glassy electrolytes is also focused on in this study.

The development of the solid electrolytes with the triangular structural units is attractive. Studies on solid electrolytes have been mainly focused on sulfide systems with tetrahedral structure units, such as PS 3–, SnS 3– and SbS 3–. In contrast, solid electrolytes composed of triangular structural units have not been extensively studied. Solid electrolytes containing boron as a center element form triangular structural units in the glass systems of Na2O-B2O3 and Na2S-B2S3. Among oxide-based solid electrolytes, the Na2O-B2O3 glass was first reported in 1936 [72] and the conductivity enhancement was studied [73, 74]. In 2016, the Na3BO3 orthoborate glass with triangular BO33– units was prepared via a mechanochemical process [74]. The ionic conductivity of Na3BO3 glass was over 10–8 S cm–1 at 25 °C. In contrast, Na2S-B2S3 glasses were prepared by conventional melt quenching from 1980s [75–84]. The glass-forming region [75], local structures [76–79], densities [80], and ionic conductivities [81–84] were reported. In general, the Na+ conductivity of glassy electrolytes increases with increasing the Na content. The conductivities of Na2S-B2S3 glasses prepared by conventional melt quenching were only investigated for compositions with less than 67 mol% Na2S [84], while the conductivity of Na3BS3 ortho-thioborate glasses with 75 mol% Na2S has not been reported. Thus, the measurements of ionic conductivities of highly Na+ containing glasses in the system Na2S-B2S3 are highly required. Another challenge for developing the sodium thioborate system is the difficulty in obtaining pure B2S3, which is not commercially available. Although the synthesis procedure of B2S3 was reported [85], B2S3 has low chemical stability with tendency of ease oxidation. Thus, the new synthesis process should be proposed in order to obtain the solid electrolytes with objective compositions. For the aim of finding new candidates of solid electrolytes, the Na2S-B2S3 glasses with high Na content and triangular structural units are studied here.

This thesis focuses on discovery of Na+ conducting sulfide electrolytes with tetrahedral or triangular units. The central metal elements are antimony, phosphorus, and boron in the systems Na2S-M2Sx (M = Sb, P, and B). The mechanochemical process is mainly used to prepare crystalline and glassy electrolytes. It is because that the mechanochemical synthesis does not need high temperatures as the conventional solid-phase synthesis. All of the Na2S-M2Sx solid electrolytes are synthesized and their structural, chemical, and electrochemical properties are analyzed. The formation of metastable phases by heating glassy samples is also investigated. In addition, the relationship between structure and properties of solid electrolytes is examined to achieve a guideline for designing sulfide electrolytes.

For the development of the crystalline solid electrolytes, the Na2S-M2S5 (M = Sb, P) electrolytes with tetrahedral units MS 3– are chosen as the base system. As mentioned above, for Na PS -based solid electrolytes, the addition of excess Na and less Na are both effective for increasing the ionic conductivity of Na3PS4 [39, 41, 54]. In order to enhance the ionic conductivity of Na3SbS4, cation or anion-substitutions are conducted to obtain the solid electrolytes with both excess Na and less Na. The molding pressures and the heating temperatures are also varied to find the superior experimental condition to show high ionic conductivities. The comprehensive evaluation on basic characteristics such as crystal structure and conductivities is performed. One of the merits of Na3SbS4 solid electrolyte is high safety at the atmosphere, and thus the chemical stability to the humid air of the prepared solid electrolytes is also examined. The Na3PS4-based compositions with less Na are investigated because the Na3SbS4 electrolytes with less Na show higher ionic conductivities than Na3SbS4. Moreover, the Na3PS4-Na4SiS4 system with excess Na were also reported to enhance the ionic conductivities [40]. Thus, the Na3PS4-based solid electrolytes with excess Na, where P is partially replaced by Ge or Sn, are also investigated. Many Na3PS4-based solid electrolytes with high ionic conductivities were already reported, but they are mainly prepared by the solid-state method. In this study, Na3PS4-based solid electrolytes are mechanochemically prepared and their structure and properties are examined. In addition, the ionic conductivities and activation energy of prepared solid electrolytes are compared with those of already reported ones. Moreover, the all-solid-state cells with the prepared solid electrolytes are evaluated.

In order to develop glassy solid electrolytes, the solid electrolytes in the system Na2S-B2S3 with the triangular structural units are also researched in this study. As mentioned above, the reported starting material B2S3 is difficult to handle [85]. Therefore, in this study, the Na2S-B2S3 electrolytes are synthesized from Na2S, B, and S as starting materials. A superior route to synthesize the objective solid electrolytes with triangular structural units is investigated. The glasses in the system Na2S-B2S3 were prepared only by the melt quenching process so far. Thus, the glass-forming region in the Na2S-B2S3 system prepared by the mechanochemical method is determined. The glasses are heated to obtain new metastable crystalline phase, because metastable phases precipitated by crystallization of mother glasses tend to show high ionic conductivities. In general, the Na+ conductivity of glassy electrolytes increases with an increase of the Na content. The 75Na2S·25B2S3 (mol%) ortho-thioborate with the highest Na composition is mainly focused in this thesis. A comprehensive evaluation on the properties of conductivity and formability and structure of the prepared electrolytes are also performed. The all-solid-state cells with the prepared solid electrolytes are evaluated.

This doctoral thesis consists of four chapters:

Chapter 1
This chapter describes the background, the objectives and structure of this thesis.

Chapter 2
This chapter focuses on Na3MS4 (75Na2S·25M2S5 (mol%)) (M = Sb, P) based crystalline solid electrolytes with excess Na or less Na by aliovalent doping in order to develop new metastable phases and enhance the ionic conductivity. The solid electrolytes with anion-substituted and the cation-substituted compositions are synthesized via the mechanochemical process and consecutive heat treatment. Changes in ion conducting properties and structures during crystallization are investigated. Moreover, the air stability of the electrolytes is examined. The obtained electrolytes with less Na show higher conductivities than Na3SbS4. In particular, Na2.88Sb0.88W0.12S4 has the highest conductivity of 3.2×10–2 S cm–1 at 25 °C among sulfide Li and Na-ion conductors reported so far and high safety without H2S gas generation. Detailed structural analysis of the solid electrolytes with less Na by the Rietveld refinements with the XRD data is conducted to investigate why the conductivity of the Na3SbS4 solid electrolyte is increased by the partial substitution of elements. Moreover, all-solid-state cells Na-Sn / TiS2 are fabricated using the prepared solid electrolytes. Room-temperature reversible operation of the cells is demonstrated. Because the W-substituted Na3SbS4 electrolytes with less Na show higher ionic conductivities than that of non-substituted Na3SbS4, the Na3–xP1–xWxS4 crystalline electrolytes are also prepared and their structure and ionic conductivities are investigated. Among Na3PS4-based crystalline electrolytes with excess Na, the Na3.33Ge0.33P0.67S4 (Na10GeP2S12) crystalline electrolyte is firstly synthesized. The Na10+xSn1+xP2–xS12 (0 ≤ x ≤ 1.2) crystalline electrolytes are also synthesized. The Na11Sn2PS12 electrolyte shows the highest ionic conductivity among the prepared Na10+xSn1+xP2–xS12 crystalline electrolytes. The composition range of the single phase, the structure and the ionic conductivities are investigated.

Chapter 3
This chapter focuses on the xNa2S·(100–x)B2S3 (33 ≤ x ≤ 80) (mol%) glassy electrolytes with triangular structure units for the development of new solid electrolytes instead of conventional sulfide electrolytes composed of tetrahedral units. The Na2S-B2S3 crystalline electrolytes are synthesized directly from a starting mixture of Na2S, B, and S reagents by the solid-state method. The Na2S-B2S3 glasses are then obtained via mechanochemical process of the prepared crystalline electrolytes. The glass formation region is 33 ≤ x ≤ 75. The Na3BS3 (x = 75) glass, the ortho-thioborate composition with triangular BS33– units, shows the highest ionic conductivity among the Na2S-B2S3 glasses in this study. A metastable crystalline phase is obtained by crystallization of the prepared Na3BS3 glass. The structure, ionic conductivities and formability of the prepared Na3BS3 glassy and crystalline electrolytes are examined. Moreover, the application to all-solid-state sodium cells (Na-Sn / Na3BS3 glass / TiS2) is investigated.

Chapter 4
This chapter summarizes all the conclusions in this thesis.

この論文で使われている画像