Mechanochemical Synthesis of Pyrite Ni_1−xFe_xS₂ Electrode for All-solid-state Sodium Battery

Abstract

All-solid-state sodium secondary batteries are expected to be low-cost, next-generation batteries. NiS₂ and FeS₂ are potential candidates as positive electrode materials owing to their high theoretical capacities. However, it is difficult to achieve sufficient capacity with bulk FeS₂. In this study, pyrite Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) electrodes are prepared by a mechanochemical process. The all-solid-state sodium cells with Ni_1−xFe_xS₂ show higher discharge–charge potentials than those with NiS₂, and higher capacities than those with FeS₂. In addition, Ni_1−xFe_xS₂ exhibits a higher rate performance than those of NiS₂ and FeS₂. The all-solid-state cells using Ni_1−xFe_xS₂ (x = 0.3, 0.5, and 0.7) are discharged and charged with a high capacity of approximately 390 mAh g⁻¹, without significant capacity fading for at least 30 cycles. The solid-solution formation of NiS₂ and FeS₂ results in lower material cost, higher rate performance, higher discharge–charge potential than those of NiS₂, and higher capacity than that of FeS₂. Pyrite Ni_1−xFe_xS₂ is a promising positive electrode material for all-solid-state sodium secondary batteries.

1. Introduction

Conventional lithium-ion batteries (LIBs) are widely used in smartphones and other devices. From a safety standpoint, all-solid-state batteries are considered as next-generation batteries owing to the use of non-flammable inorganic solid electrolytes. Furthermore, all-solid-state sodium secondary batteries are expected to be low-cost batteries due to the abundance of sodium sources.^1,2

It is necessary to develop a positive electrode active material with a high capacity and potential to obtain an all-solid-state sodium secondary battery with a high energy density. Sulfur is a promising candidate as a high-capacity material due to its high theoretical capacity (1672 mAh g⁻¹). However, it is challenging to use sulfur directly as an active material, because it has a low electrical conductivity of 5 × 10⁻³⁰ S cm⁻¹ at 25 °C.³ The preparation of a composite that includes conductive carbon is an effective strategy for increasing electron conduction. In fact, our research group reported that a sulfur-carbon composite showed a high reversible capacity in all-solid-state sodium secondary batteries.⁴

In addition to sulfur-carbon composites, transition metal sulfides are promising electrode materials for high-energy batteries because of their high theoretical capacities and electrical conductivities. Among them, transition-metal polysulfide shows a high capacity derived from the anionic redox reaction involving dissociation and formation of disulfide bonds.⁵ We also reported that amorphous MoS₃ showed a high capacity in all-solid-state sodium secondary batteries.⁶ In addition, metal sulfides with pyrite-type structures that have disulfide bonds (e.g., NiS₂, FeS₂, and CoS₂) are expected to be high-capacity materials. In particular, NiS₂ is expected to be a promising electrode material for all-solid-state sodium secondary batteries because it has been reported that NiS₂ shows a high capacity in sodium-ion batteries using organic electrolytes.^7,8 FeS₂ is also a promising candidate because of its abundance in the Earth’s crust, low material cost, and high theoretical capacity.^9–11 Additionally, FeS₂ showed a higher discharge–charge potential than that of NiS₂, as shown in this study. However, it is difficult to achieve sufficient capacity using bulk FeS₂ due to the slow diffusion of Na and Fe in FeS₂, which decreases the utilization of FeS₂ during the discharge–charge reaction.¹²

In this study, we focused on the solid solutions Ni_1−xFe_xS₂ prepared from NiS₂ and FeS₂. We aimed to lower the cost and improve the discharge–charge potential of NiS₂, and improve the capacity of FeS₂, by combining NiS₂ and FeS₂. Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) were prepared by a mechanochemical (MC) process. MC processes are often used for the synthesis of metal sulfides.^6,13–16 The structures and electrochemical properties of Ni_1−xFe_xS₂ electrodes in all-solid-state sodium secondary batteries were evaluated.

2. Experimental

2.1 Preparation of Ni_1−xFe_xS₂

Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) samples were prepared using an MC process. The starting materials, Ni (99.99 %, Aldrich), Fe (99.9 %, Kojundo Chemical), and S (99.9 %, Kojundo Chemical) were mixed in a molar ratio of 1 − x : x : 2 (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) to obtain a total mass of 0.5 g. The mixtures were placed in a 45 mL zirconia pot with zirconia balls (90 g, diameter 4 mm) in a dry Ar glove box and mechanically milled using a planetary ball mill apparatus (Pulverisette 7, Fritsch). The rotational speed and milling time were set to 510 rpm and 10 h, respectively.

2.2 Material characterization

X-ray diffraction (XRD) measurements were performed using an X-ray diffractometer (SmartLab, Rigaku) equipped with a one-dimensional X-ray detector employing CuKα radiation (λ = 1.54 × 10⁻¹⁰ m). CHNS elemental analysis was performed using an elemental analyzer (Vario EL cube, Elementar). The surface of the Ni_1−xFe_xS₂ particles was analyzed using a field-emission scanning electron microscope (FE-SEM, SU8200; Hitachi High-Technologies Corp.). The electronic structures of sulfur in Ni_1−xFe_xS₂ were analyzed using X-ray photoelectron spectroscopy (XPS) (K-Alpha, Thermo Fisher Scientific) with a monochromatic AlKα source (1486.6 eV). The observed binding energies were calibrated against the adventitious C1s peak at 284.7 eV.

2.3 Electrochemical measurements

The all-solid-state cells for discharge–charge measurements were fabricated as follows: composite positive electrodes were prepared by mixing Ni_1−xFe_xS₂, Na₃PS₄ glass-ceramic solid electrolyte (SE),¹⁷ and acetylene black (AB) in a mass ratio of 40 : 60 : 6 in an agate mortar. SE and AB provided sodium-ion and electron conduction paths to the active material in the composite positive electrodes. In this study, a sufficient amount of SE (60 wt%) was also added to the composite positive electrodes to evaluate the potential performance of Ni_1−xFe_xS₂ as an active material. Tri-layer pellets (10 mm in diameter) consisting of a Ni_1−xFe_xS₂-Na₃PS₄-AB positive electrode layer (6 mg), a Na₃PS₄ glass-ceramic separator layer (80 mg), and a Na₁₅Sn₄-Ketjen black (KB) negative electrode layer (40 mg)¹⁸ were prepared by uniaxial pressing under 360 MPa for 5 min in a polycarbonate cylinder with stainless steel rods acting as current collectors. All preparation steps were conducted in a dry-Ar-filled glovebox. The discharge–charge tests were conducted in a glovebox at 24–27 °C with a cut-off voltage range of 0.92 to 2.72 V vs. Na₁₅Sn₄ using an electrochemical workstation (VMP3, Bio-Logic Co.).

3. Results and Discussion

Figure 1a shows the XRD patterns of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1). Figure 1b and Table 1 show the lattice constants determined from the XRD patterns of Ni_1−xFe_xS₂. All compositions showed similar XRD patterns with broad peaks, and the diffraction patterns of Ni_1−xFe_xS₂ could be indexed by the pyrite structure of the space group Pa-3 (No. 205). On the XRD patterns of Ni_0.3Fe_0.7S₂, Ni_0.1Fe_0.9S₂, and FeS₂, the small peaks derived from the impurity phase Fe₇S₈ were observed. Figure 1c shows the structural model of the pyrite. All sulfur atoms in the pyrite-type structure formed disulfide bonds. With an increase in the Fe content of Ni_1−xFe_xS₂, the diffraction pattern of Ni_1−xFe_xS₂ shifted to higher angles and the lattice constant of Ni_1−xFe_xS₂ decreased, suggesting that solid solutions of Ni_1−xFe_xS₂ were obtained.

Figure 1.

(a) XRD patterns of Ni_1−xFe_xS₂. (b) The lattice constants for x in Ni_1−xFe_xS₂. (c) The structural model of the pyrite.

Table 1. The lattice constants of Ni_1−xFe_xS₂.

	Lattice constants (Å)
NiS2	5.681 (7)
Ni0.7Fe0.3S2	5.602 (7)
Ni0.5Fe0.5S2	5.552 (3)
Ni0.3Fe0.7S2	5.516 (17)
Ni0.1Fe0.9S2	5.456 (3)
FeS2	5.426 (3)

Figures 2a–2f show SEM images of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1). The particle sizes of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) samples ranged from the submicron level to a few microns, and there were no significant differences in their sizes. From the EDS mapping results of Ni_0.3Fe_0.7S₂ (Fig. S1), it was found that the constituent elements were basically uniformly distributed.

Figure 2.

(a–f) The SEM images of Ni_1−xFe_xS₂.

Figure 3a shows the S2p XPS spectra of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1), which were composed of two sets of doublet peaks. Each doublet comprises the S2p_3/2 and S2p_1/2 components. The peak energy separation (1.2 eV) and the area ratio of S2p_3/2/S2p_1/2 = 2/1 for each doublet were set during peak fitting. The S2p spectra of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) showed one S2p_3/2 peak at 162.7–163.0 eV. This is because there is only one type of sulfur in the pyrite-type structure, which forms disulfide bonds. Figures 3b and 3c show the Ni2p and Fe2p XPS spectra of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1), where a Ni2p_3/2 peak was observed at 853.2–853.4 eV and an Fe2p_3/2 peak at 707.0–707.3 eV. The broad peak near 711 eV in Fig. 3c is attributed to the Auger line of Ni KL₁L₁. These results indicate that the electronic states of the Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) samples were similar. In all compositions, the electronic states of Ni and Fe are similar because Ni and Fe exist in the cationic sites in the pyrite structure.

Figure 3.

(a) S2p (b) Ni2p and (c) Fe2p XPS spectra of Ni_1−xFe_xS₂.

Figures 4a–4f show the discharge–charge curves of the all-solid-state cells fabricated using Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) as the positive electrode active material, respectively. The discharge–charge measurements were performed at a current density of 0.013 mA cm⁻². The discharge–charge potentials of Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9) were higher than that of NiS₂ and lower than that of FeS₂. Thus, the discharge–charge profiles of Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9) were intermediate between those of NiS₂ and FeS₂. The plateaus observed during the first discharge of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) were attributed to the structural changes associated with the initial Na ion insertion. With the insertion of Na during the initial discharge, the crystal structure changes and becomes amorphous. The crystal structure does not return to its original state even after the Na is removed during charging. The charging and discharging continue in an amorphous state.¹⁹ The reversible capacities of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) were 492, 485, 411, 390, 418, and 227 mAh g⁻¹, respectively. NiS₂ showed irreversible capacity in the first cycle, and then the capacity decreased slightly over five cycles. In contrast, Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, 0.9, and 1) showed reversible capacities in the first cycle, and no decrease in capacity with each cycle. Furthermore, the potentials of Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9) increased compared to those of NiS₂. This was attributed to the combination of the characteristics of NiS₂ and FeS₂. Here, the reaction equation between Na and MS₂ (M: Ni, Fe) is as follows:¹⁹ MS₂ + 2Na⁺ + 2e⁻ $ \rightleftarrows $ Na₂MS₂. This reaction is an anion redox reaction in which the disulfide bond ((S-S)²⁻) dissociates and changes to S²⁻, unlike the intercalation reaction of the layered compound. The anion redox reaction of Ni_1−xFe_xS₂ is a two-electron reaction with a theoretical capacity of approximately 440 mAh g⁻¹. The capacity of 227 mAh g⁻¹ for the all-solid-state cell using FeS₂ was clearly lower than the theoretical capacity. The electronic conductivity of Ni_1−xFe_xS₂ too high to obtain an accurate value. Therefore, the influence of electronic conductivity on the difference in capacity and overvoltage is almost negligible, and factors such as sodium diffusion during the discharge-charge are considered to be more influential. Additionally, it has been reported that bulk FeS₂ does not exhibit sufficient capacity.^12,20 This is considered to be due to the slow diffusion of Na and Fe in FeS₂; controlling the particle size at the nanoscale is one method for solving this problem.¹² In fact, FeS₂ (99.9 %, Aldrich) with a controlled particle size at the nanoscale showed a higher capacity at 298 mAh g⁻¹ (Fig. S2) than FeS₂ prepared by the MC process. Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9), in which a portion of Fe in FeS₂ was replaced with Ni, showed capacities close to the theoretical capacity.

Figure 4.

(a–f) Discharge-charge curves of Na₁₅Sn₄/Na₃PS₄/Ni_1−xFe_xS₂ in all-solid-state cell.

Figure 5a shows the rate performance of the all-solid-state cells using Ni_1−xFe_xS₂ (x = 0, 0.5, and 1) as the positive electrode active materials. The discharge–charge measurements were performed with current densities of 0.013, 0.038, 0.064, 0.13 and 0.19 mA cm⁻². Both NiS₂ and FeS₂ showed low capacities of approximately 80 mAh g⁻¹ and 25 mAh g⁻¹ at 0.13 mA cm⁻² and 0.19 mA cm⁻², respectively. Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9) showed a high rate performance compared to those of NiS₂ and FeS₂. In particular, Ni_0.5Fe_0.5S₂ showed high capacities of 256 and 162 mAh g⁻¹ at 0.13 and 0.19 mA cm⁻², respectively. It was found that the rate performance of Ni_1−xFe_xS₂ was improved by substituting a portion of Ni in NiS₂ with Fe (or Fe in FeS₂ with Ni) compared to those of NiS₂ and FeS₂. Recently, it was reported that Co substitution into FeS₂ enhances Na diffusion,²¹ and it was considered that a similar phenomenon occurs in Ni_1−xFe_xS₂ in this study. However, the detailed mechanism of Na diffusion enhancement has not been revealed.

Figure 5.

(a) Rate performance of Ni_1−xFe_xS₂ in all-solid-state cell at 0.013–0.19 mA cm⁻² for 0.92–2.72 V vs. Na₁₅Sn₄. (b) Cyclic performance of Ni_1−xFe_xS₂ in all-solid-state cell at 0.038 mA cm⁻² for 0.92–2.72 V vs. Na₁₅Sn₄.

Figure 5b shows the cycle performance of the all-solid-state cells using Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) as the positive electrode active materials. The discharge–charge cycle measurements were performed at a current density of 0.038 mA cm⁻². The capacities at the 30th cycle of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) were 370, 381, 385, 395, 313, and 97 mAh g⁻¹, respectively, while the capacity retention rates of Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1) were 81.9 %, 86.5 %, 89.8 %, 93.6 %, 85.0 %, and 78.1 % between the 2nd and 30th cycles, respectively. NiS₂ and FeS₂ showed a decrease in capacity in the first few cycles, but then showed a stable capacity up to 30 cycles. Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9) had improved capacity degradation results in the first few cycles compared to those observed in NiS₂ and FeS₂; Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9), and did not show significant capacity fading for at least 30 cycles. This indicated that Ni_1−xFe_xS₂ exhibited a high cycle performance for 30 cycles in all-solid-state sodium cells. The Ni_1−xFe_xS₂ (x = 0.3, 0.5, and 0.7) samples, which have higher discharge–charge potentials than that of NiS₂, showed almost the same capacities as that of NiS₂ after 30 cycles. These results show that Fe substitution in NiS₂ can reduce the material cost and improve the energy density compared to those of NiS₂.

4. Conclusion

Pyrite Ni_1−xFe_xS₂ (x = 0, 0.3, 0.5, 0.7, 0.9, and 1), which is a solid solution of NiS₂ and FeS₂, was prepared by the MC process. The XRD results suggested that a solid solution of Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9) was obtained. The all-solid-state sodium cells with Ni_1−xFe_xS₂ (x = 0.3, 0.5, 0.7, and 0.9) positive electrodes showed high reversible capacities comparable to those of the cell with NiS₂. Ni_1−xFe_xS₂ showed a higher discharge–charge potential than that of NiS₂. Ni_1−xFe_xS₂ also showed a higher rate performance than those of NiS₂ and FeS₂. In particular, Ni_0.5Fe_0.5S₂ showed high capacities of 485, 256 and 162 mAh g⁻¹ at 0.013, 0.13, and 0.19 mA cm⁻², respectively. In addition, Ni_1−xFe_xS₂ exhibited an excellent cycle performance for 30 cycles in all-solid-state sodium cells. The solid-solution formation of Ni_1−xFe_xS₂ resulted in lower material cost, higher rate performance, and higher discharge–charge potential than those of NiS₂, and higher capacity than that of FeS₂. Furthermore, it could be determined that the formation of Ni_1−xFe_xS₂ solid solutions is advantageous for Na ion diffusion due to the combination of different discharge–charge potentials. This study suggests that Ni_1−xFe_xS₂ is a promising positive electrode material for all-solid-state sodium secondary batteries.

Acknowledgment

This study was supported by JSPS KAKENHI Grant Numbers JP20K05688 and JP21H04701.

Data Availability Statement

The data that support the findings of this study are openly available under the terms of the designated Creative Commons License in J-STAGE Data at https://doi.org/10.50892/data.electrochemistry.19106876.

CRediT Authorship Contribution Statement

Gaku Shirota: Data curation (Lead), Formal analysis (Lead), Writing – original draft (Lead)

Akira Nasu: Data curation (Supporting), Investigation (Supporting), Writing – review & editing (Supporting)

Atsushi Sakuda: Conceptualization (Lead), Project administration (Lead), Supervision (Equal), Writing – review & editing (Lead)

Minako Deguchi: Data curation (Supporting), Formal analysis (Equal)

Kota Motohashi: Supervision (Supporting), Writing – review & editing (Equal)

Masahiro Tatsumisago: Supervision (Equal), Writing – review & editing (Supporting)

Akitoshi Hayashi: Funding acquisition (Equal), Supervision (Equal), Writing – review & editing (Equal)

Conflict of Interest

The authors declare no conflict of interest in the manuscript.

Funding

Japan Society for the Promotion of Science: JP20K05688

Japan Society for the Promotion of Science: JP21H04701

Footnotes

A. Nasu, A. Sakuda, K. Motohashi, and A. Hayashi: ECSJ Active Members

M. Tatsumisago: ECSJ Fellow

References

• 1)  J. Y. Hwang, S. T. Myung, and Y. K. Sun, Chem. Soc. Rev., 46, 3529 (2017).
• 2)  S. Chou and Y. Yu, Adv. Energy Mater., 7, 1703223 (2017).
• 3)  J. Wang, S. Y. Chew, Z. W. Zhao, S. Ashraf, D. Wexler, J. Chen, S. H. Ng, S. L. Chou, and H. K. Liu, Carbon, 46, 229 (2008).
• 4)  T. Ando, A. Sakuda, M. Tatsumisago, and A. Hayashi, Electrochem. Commun., 116, 106741 (2020).
• 5)  A. Sakuda, K. Ohara, K. Fukuda, K. Nakanishi, T. Kawaguchi, H. Arai, Y. Uchimoto, T. Ohta, E. Matsubara, Z. Ogumi, T. Okumura, H. Kobayashi, H. Kageyama, M. Shikano, H. Sakaebe, and T. Takeuchi, J. Am. Chem. Soc., 139, 8796 (2017).
• 6)  G. Shirota, A. Nasu, M. Deguchi, A. Sakuda, M. Tatsumisago, and A. Hayashi, J. Power Sources Advances, 10, 100061 (2021).
• 7)  K. M. Abraham and J. E. Elliot, J. Electrochem. Soc., 131, 2211 (1984).
• 8)  T. Wang, P. Hu, C. Zhang, H. Du, Z. Zhang, X. Wang, S. Chen, J. Xiong, and G. Cui, ACS Appl. Mater. Interfaces, 8, 7811 (2016).
• 9)  R. Fong, J. R. Dahn, and C. H. W. Jones, J. Electrochem. Soc., 136, 3206 (1989).
• 10)  Z. Hu, Z. Zhu, F. Cheng, K. Zhang, J. Wang, C. Chen, and J. Chen, Energy Environ. Sci., 8, 1309 (2015).
• 11)  X. Li, S. H. Qi, W. C. Zhang, Y. Z. Feng, and J. M. Ma, Rare Met., 39, 1239 (2020).
• 12)  A. Douglas, R. Carter, L. Oakes, K. Share, A. P. Cohn, and C. L. Pint, ACS Nano, 9, 11156 (2015).
• 13)  N. Tanibata, T. Matsuyama, A. Hayashi, and M. Tatsumisago, J. Power Sources, 275, 284 (2015).
• 14)  T. Matsuyama, A. Hayashi, T. Ozaki, S. Mori, and M. Tatsumisago, J. Mater. Chem. A, 3, 14142 (2015).
• 15)  A. Sakuda, N. Taguchi, T. Takeuchi, H. Kobayashi, H. Sakaebe, K. Tatsumi, and Z. Ogumi, ECS Electrochem. Lett., 3, A79 (2014).
• 16)  A. Sakuda, N. Taguchi, T. Takeuchi, H. Kobayashi, H. Sakaebe, K. Tatsumi, and Z. Ogumi, Solid State Ionics, 262, 143 (2014).
• 17)  A. Hayashi, K. Noi, A. Sakuda, and M. Tatsumisago, Nat. Commun., 3, (2012).
• 18)  N. Tanibata, A. Hayashi, and M. Tatsumisago, J. Electrochem. Soc., 162, A793 (2015).
• 19)  A. Kitajou, J. Yamaguchi, S. Hara, and S. Okada, J. Power Sources, 247, 391 (2014).
• 20)  T. B. Kim, W. H. Jung, H. S. Ryu, K. W. Kim, J. H. Ahn, K. K. Cho, G. B. Cho, T. H. Nam, I. S. Ahn, and H. J. Ahn, J. Alloys Compd., 449, 304 (2008).
• 21)  J. Wang, J. Qin, Y. Jiang, X. Wang, and M. Cao, Nanoscale, 12, 13781 (2020).