Effect on Electrochemical Properties of Phases in AB-Type Zr–Ti–Nb–Ni Alloys as Nickel-Metal Hydride Batteries

Abstract

AB-type Zr–Ti–Nb–Ni alloys are good candidates as the negative electrode for Ni-metal hydride (MH) batteries because they have high discharge capacity around 330 mAh g⁻¹ at 303 K. The Zr_0.49Ti_0.5Nb_0.01Ni alloy consists of two phases, the primary-phase with B33-type orthorhombic structure and the secondary-phase with B2-type cubic structure. To compare electrochemical properties of each phase with the mother alloy, we synthesized the primary-phase and secondary phase alloys with compositions of Zr_0.54Ti_0.47Nb_0.01Ni_0.98 and Zr_0.47Ti_0.52Nb_0.01Ni, respectively. The discharge capacity was examined at 25 mA g⁻¹ and 303 K, showing that the primary-phase alloy has the highest value of 362 mAh g⁻¹ than the mother alloy (335 mAh g⁻¹) and the secondary-phase alloy (253 mAh g⁻¹). For cycle performance, all alloys were excellent (≧95%) at 100 mA g⁻¹ and 303 K. For high-rate dischargeability, the secondary-phase alloy was the best, probably because the stability of hydride for the secondary-phase alloy was lower than that for the mother and the primary-phase alloy.

Fig. 6 Discharge curve (1st cycle) for the Zr_0.49Ti_0.5Nb_0.01Ni mother and the primary- and secondary-phase alloy negative electrodes at 25 mA g⁻¹, 303 K.

1. Introduction

Developing high performance electrochemical energy storage devices is important in sustainable society. Nickel-metal hydride (Ni-MH) batteries, which consist of the hydrogen storage alloy negative electrode active material, the Ni(OH)₂ positive electrode active material and alkaline aqueous solutions, were commercialized in 1990s and have been used around the world as power sources for mobile electric devices, hybrid electric vehicles, railways, etc. because of their high safety, good charge-discharge cycle life and high-rate performance.^1–4) However, the gravimetric energy density of Ni-MH batteries is one third of the Lithium-ion batteries (LIBs), therefore the use of LIBs for electric vehicles and electric power storage is on the increase.^5,6) The LIBs, however, have a serious issue in safety because they include flammable organic electrolyte solutions.^7,8) Therefore, it is meaningful to increase the gravimetric energy density of the Ni-MH batteries, in particular, develop new negative electrode active material with high capacity.

AB-type ZrNi hydrogen storage alloy has the theoretical capacity of 536 mAh g⁻¹,⁹⁾ which is higher than that of commercial AB₅- or AB₃-type rare earth-based alloys,^10,11) so the ZrNi alloy is a promising negative electrode active material.^12–14) Recently, we have developed new AB-type Zr–Ti–Nb–Ni quaternary hydrogen storage alloys, and the electrochemical capacities of the Zr_0.58Ti_0.4Nb_0.02Ni negative electrode at 333 and 303 K were 384 and 335 mAh g⁻¹,¹⁵⁾ respectively. The Zr–Ti–Nb–Ni alloys consist of two phases, the primary-phase with B33-type ZrNi orthorhombic structure and the secondary-phase with B2-type Ti_0.6Zr_0.4Ni cubic structure. However, electrochemical properties for each phase have not been elucidated yet. Moreover, it is essential to clarify the role of each phase to improve negative electrode properties.

In this study, first, we synthesized the Zr_0.49Ti_0.5Nb_0.01Ni alloy as the mother alloy, and characterize it by X-ray diffractometry (XRD) with Rietveld analysis, optical microscopy and electron probe microanalysis (EPMA). Second, we separately synthesized the primary- and the secondary-phase alloys, and evaluated the crystal structure, pressure-composition isotherms and electrochemical properties of each phase.

2. Experimental Procedure

2.1 Preparation of Zr–Ti–Nb–Ni alloys

The Zr_0.49Ti_0.5Nb_0.01Ni mother alloy or its primary- or secondary-phase alloys were prepared along the previous method.¹⁵⁾ All reagents used in this study were purchased from the Koujundo Chemical Laboratory. The stoichiometric amounts of sponge Zr (98.0%), Ti powder (99.0%), Nb powder (99.9%) and Ni powder (99.9%) were mixed and alloyed by arc-melting method on a water-cooled copper mold in an Ar atmosphere. The resultant alloy ingot was remelted five times to ensure their homogeneity, and then each ingot was pulverized and sieved between 20 and 40 µm in diameter.

2.2 Characterization of the Zr–Ti–Nb–Ni alloys

The optical micrographs for each alloy were taken with an optical microscope (Nikon ECLIPSE L150). Before the observation with the optical microscope, each alloy ingot with polished cross-section was immersed in a mixed acid solution, which consisted of 1 mL of nitric acid, 0.5 mL of hydrofluoric acid and 98.5 mL deionized water, for 10 s. The chemical composition of the primary- and the secondary-phase alloys was estimated with an Electron Probe Micro Analyzer (EPMA Shimadzu, EPMA-1100).

Crystal structure of the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy or its primary- or secondary-phase alloy powder was analyzed by X-ray diffractometry (XRD, Rigaku, SmartLab) with CuKα radiation (λ = 0.1541 nm, 40 kV, 40 mA). The lattice parameters (a, b, c) and phase abundances of the primary and secondary phases in the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy were refined by the Rietveld analysis using Z-Rietveld software (ver. 1.0.2).^16,17) In the Rietveld refinement, the compositions of the primary and secondary phases determined by EPMA were taken as the initial ones. The atomic coordinates reported by Matar¹⁸⁾ and Cuevas et al.¹⁹⁾ were used as those of the primary and secondary phases, respectively.

2.3 Measurement of pressure-composition isotherms (PCT curves) of the Zr–Ti–Nb–Ni alloys

The pressure-composition isotherms (PCT curves) for the Zr–Ti–Nb–Ni alloy powders were measured between 373 K using a Sieverts-type apparatus (Suzuki Shokan, PCT-6-4SDMDWIN). Before the PCT measurement, the initial activation treatment was performed according to our previous study.¹⁵⁾ Each sample was placed into a stainless steel reactor tube in an Ar-filled glove box, heated in vacuum at 423 K for 1 h and gradually introduced at 0.9 MPa H₂ pressure. After the reactor tube was heated to 623 K, it was evacuated at the same temperature for 2 h. After the initial activation treatment, the PCT curves were measured in the hydrogen pressure between 0.005 and 0.92 MPa at given temperatures.

2.4 Electrochemical measurements of the Zr–Ti–Nb–Ni electrodes

The Zr_0.49Ti_0.5Nb_0.01Ni mother alloy and its primary- and secondary-phase alloy negative electrodes were prepared according to our previous study.¹⁵⁾ Briefly, the alloy powder, Cu powder and 10 mass% polyvinyl alcohol aqueous solution were mixed to make a paste, and the resultant paste was cast on a Ni mesh as a current collector. For the activation treatment, the resultant Zr–Ti–Nb–Ni negative electrodes were immersed in a boiling 6 M KOH alkaline solution for 4 h.¹²⁾ The positive and reference electrodes were NiOOH/Ni(OH)₂ and Hg/HgO electrodes, respectively. The electrolyte solution was a 6 M KOH aqueous solution containing 1 M LiOH.

All Charge/discharge experiments were performed at 303 K. For evaluating the maximum discharge capacity, each negative electrode was charged at a current density of 100 mA g⁻¹ for 5 h and then discharged at a current density of 25 mA g⁻¹ until cut-off potential of −0.5 V vs. Hg/HgO. After charging and discharging, the circuit was opened for 10 min. For evaluating high-rate dischargeability (HRD), each negative electrode was charged at a current density of 100 mA g⁻¹ for 5 h, and then discharged at a current densities of 25–500 mA g⁻¹ until −0.5 V vs. Hg/HgO. Electrochemical impedance spectroscopy was performed in a frequency range between 0.1 Hz and 64 kHz with an amplitude of 5 mV after each electrode was charged at 100 mA g⁻¹ for 5 h, and then kept at an open-circuit potential for 1 h.

3. Results and Discussion

A cross-sectional optical micrograph image of the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy ingot is shown in Fig. 1. The Zr_0.49Ti_0.5Nb_0.01Ni mother alloy consisted of gray and black phases, and their area fractions were about 86 and 14%, respectively. In this study, the gray and black phases were assigned to the primary- and secondary-phase alloys, respectively. Average chemical compositions of the primary- and secondary-phases determined by EPMA were Zr_0.54Ti_0.47Nb_0.01Ni_0.98 and Zr_0.47Ti_0.52Nb_0.01Ni, respectively.

Fig. 1

Cross-sectional optical micrograph image of the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy ingot.

Figure 2 shows the XRD pattern of the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy powder with the Rietveld refinement. In Fig. 2, small upper and lower bars are assigned to Zr_0.54Ti_0.47Nb_0.01Ni_0.98 (B33-type orthorhombic structure) and Zr_0.47Ti_0.52Nb_0.01Ni (B2-type cubic structure), respectively. The mother alloy was composed of two phases, the B33-type orthorhombic primary phase and the B2-type cubic secondary phase, which was consistent with the other Zr–Ti–Nb–Ni alloys.¹⁵⁾ Table 1 shows lattice parameters and abundance of the primary- and secondary-phases estimated by the Rietveld refinement for the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy powders. The residue of the weighted pattern (R_wp) and goodness-of-fit indicator (S) are used to ascertain the quality of the fit between the experimental and calculated patterns. The R_wp and S values are 9.91% and 1.89, respectively, suggesting all the estimated parameters were reasonable. The phase abundance of the primary and secondary phases was 82.1 and 17.9 at%, respectively.

Fig. 2

XRD pattern of the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy powder with the Rietveld refinement.

Table 1 Crystal structure, lattice parameters and abundance for the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy powders by the Rietveld refinement.

Next, we separately synthesized the primary- and secondary-phase alloys with compositions of Zr_0.54Ti_0.47Nb_0.01Ni_0.98 and Zr_0.47Ti_0.52Nb_0.01Ni, respectively.

Figure 3(a) and Fig. 3(b) show the cross-sectional optical micrograph images of the primary- and secondary-phase alloy ingots. In addition, average chemical compositions for both ingots were shown in Table 2. The optical micrograph image of the primary-phase alloy consisted of dark and light gray phases like the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy. The area of the light gray phase was largely occupied in the primary phase alloy. The average chemical composition in the dark gray phase was in close agreement with the theoretical value. On the other hand, for the light gray phase the Ti content was lower than the theoretical value, whereas the Zr content was higher. However, it is suggested that both phases have the B33-type orthorhombic structure. For the secondary-phase alloy, the cross-section was largely occupied by the black phase as shown in Fig. 3(b). As shown in Table 2, the average chemical composition of the black phase was also similar to the theoretical composition. On the contrary, the Ti content in the gray phase, which is slightly observed in Fig. 3(b), was lower than the Zr content, suggesting the gray phase has the B33-type orthorhombic structure like the primary phase.

Fig. 3

Cross-sectional optical micrograph images of (a) the primary- and (b) the secondary-phase alloy ingots.

Table 2 Average chemical compositions for the primary- and secondary-phase alloys by EPMA.

Figure 4 shows the XRD patterns of the primary- and secondary-phase alloy powders with the Rietveld refinement. The primary- and the secondary-phase alloy powders were nearly indexed as the B33-type orthorhombic structure and the cubic structure, respectively. Table 3 show crystal structure, lattice parameters and abundance of the primary- and secondary-phase alloy powders by the Rietveld refinement. Both phase alloys had almost the single phase, which was consistent with the optical micrograph observation.

Fig. 4

XRD patterns of (a) the primary- and (b) secondary-phase alloy powders with the Rietveld refinement.

Table 3 Crystal structure, lattice parameters and abundance for the primary- and secondary-phase alloy powders by the Rietveld refinement.

The lattice parameters for the primary-phase of the mother alloy with the B33-type orthorhombic structure (Table 1) were slightly smaller than those of the primary-phase alloy (Table 3). In our previous study, the lattice constants and lattice volume linearly decreased with the Ti content because of the shrinkage of crystal lattices.²⁰⁾ This suggests that the Ti content for the light gray (primary) phase of the primary-phase alloy is lower than that for the mother alloy, leading to the shrinkage of the B33-type orthorhombic crystal lattices. On the contrary, the lattice parameters for the secondary-phase of the mother alloy with the B2-type cubic structure (Table 1) were almost equal to those for the black (primary) phase of the secondary-phase alloy (Table 3), suggesting that the average chemical composition for the black phase of the secondary-phase alloy were almost to the theoretical one or the chemical composition for the secondary-phase of the mother alloy.

Figure 5 shows PCT curves of the hydrogen absorption and desorption at 373 K for the mother alloy, the primary-phase alloy and the secondary-phase alloy. In the PCT curves for the mother and primary-phase alloys, a distinct plateau was observed regardless of the hydrogen absorption and desorption. Each plateau was assigned to the transformation from monohydride to trihydride. They were characteristic of the typical B33-type orthorhombic structural alloy.^9,20,21) The maximum hydrogen storage capacity, which was estimated at 0.9 MPa, for the mother and the primary-phase alloy at 373 K were 1.8 and 1.9 mass%H₂, respectively. The plateau pressures in hydrogen absorption from ZrNiH to ZrNiH₃ for the mother and primary-phase alloys were 0.025 and 0.015 MPa, whereas those in hydrogen desorption from ZrNiH₃ to ZrNiH were 0.011 and 0.0062 MPa, respectively. The plateau pressures for the mother alloy were higher than those for the primary-phase alloy, because the former had smaller lattice constants than the latter although both alloys had the B33-type orthorhombic structure. This suggests that the stability of hydrides for the mother alloy is lower than that for the primary-phase alloy.

Fig. 5

PCT curves of (a) hydrogen absorption and (b) hydrogen desorption for each phase.

The plateau region of the secondary phase alloy was not clear, which was also observed for the B2-type cubic TiNi alloy.^22,23) The maximum hydrogen storage capacity at 0.9 MPa for the secondary-phase alloy was 1.1 mass%H₂, which was lower than that for the primary-phase alloy, suggesting that the secondary-phase contributes to the decrease in the hydrogen storage capacity of the mother alloy. Because the primary-phase alloy was scarcely had the B2-type TiNi phase, whereas its content in the mother alloy was about 18%. For the secondary-phase alloy, the plateau pressures in hydrogen absorption and desorption defined as the half of hydrogen absorption capacity²⁴⁾ were 0.030 and 0.022 MPa, respectively, which were higher than those for the mother and primary-phase alloys, suggesting that the stability of hydrides in the secondary-phase alloy is the lowest in the three alloys.

It is interesting to discuss whether the hydrogen storage capacity of the mother alloy is represented as a function of that of the primary- and secondary phases or not to clarify the role of each phase. As shown in Table 1, the abundances of the primary- and secondary-phase in the mother alloy (F₁ and F₂) are 0.821 and 0.179, respectively. When the hydrogen storage capacity for the primary- and secondary-phase alloys are named C₁ and C₂, the hydrogen storage capacity of the mother alloy (C_m) at each equilibrium pressure can be represented in the following equation.   

\begin{equation} C_{\text{m}} = C_{1}F_{1} + C_{2}F_{2} \end{equation} (1)

The PCT curves drawn with the C_m values at various equilibrium pressures are shown as broken curves in Fig. 5(a) and Fig. 5(b). Each broken curve was in good agreement with the experimental one for the mother alloy, indicating that both phases contribute to the hydrogen storage capacity of the mother alloy.

Figure 6 shows discharge curves (1st cycle) for the Zr_0.49Ti_0.5Nb_0.01Ni mother and the primary- and secondary-phase alloy negative electrodes at 25 mA g⁻¹ and 303 K. The initial discharge capacity for the mother and the primary- and secondary-phase alloy negative electrodes was 335, 362 and 253 mAh g⁻¹, respectively. Thus the primary-phase alloy negative electrode had the highest discharge capacity among the three electrodes. The discharge capacity of the mother alloy electrode estimated with eq. (1), in which hydrogen storage capacity was replaced with discharge capacity, was 342 mAh g⁻¹, which was close to the experimental discharge capacity. Both phases also contribute to the electrochemical hydrogen storage capacity of the mother alloy.

Fig. 6

Discharge curve (1st cycle) for the Zr_0.49Ti_0.5Nb_0.01Ni mother and the primary- and secondary-phase alloy negative electrodes at 25 mA g⁻¹, 303 K.

The discharge plateau region assigned to the hydrogen desorption from ZrNiH₃ to ZrNiH^14,15) was clearly observed for the mother and primary-phase alloy negative electrodes, which is similar to the PCT curves in Fig. 5. The discharge plateau potentials for the mother and primary- and secondary-phase alloy negative electrodes were −0.842, −0.825 and −0.848 V, respectively, suggesting the stability of hydride in the secondary-phase alloy is lower than that for the mother and primary-phase alloys.

Figure 7 shows discharge capacity and the normalized discharge capacity (NDC) as a function of the number of cycles, respectively. Here, the NDC was defined by NDC(%) = 100C_n/C_max, where C_n is the discharge capacity at n-th cycle and C_max is the maximum discharge capacity. The NDCs after 20 cycles for all alloy negative electrodes were more than 95%. So all the alloy negative electrodes exhibited good cycle performance as shown in Fig. 7(a).

Fig. 7

(a) Discharge capacity and (b) NDC at 303 K as a function of the number of cycles for the Zr_0.49Ti_0.5Nb_0.01Ni mother and the primary- and secondary-phase alloy negative electrodes.

Figure 8 shows high-rate dischargeability (HRD) and normalized one, respectively. HRD was defined by HRD(%) = 100C_i/C₂₅, where C_i and C₂₅ are discharge capacity at i and 25 mA g⁻¹, respectively. The secondary-phase alloy exhibited the best HRD, probably because the stability of hydride for the secondary-phase alloy was lower than that for the mother and the primary-phase alloy.

Fig. 8

(a) High-rate dischargeability and (b) normalized HRD for the Zr_0.49Ti_0.5Nb_0.01Ni mother and the primary- and secondary-phase alloy negative electrodes.

Figure 9 shows AC impedance spectra for the Zr_0.49Ti_0.5Nb_0.01Ni mother and the primary- and secondary-phase alloy negative electrodes after charging at 21th cycle. A distinct semicircle in the low frequency range for all negative electrodes was observed. This was assigned to charge transfer in the negative electrode active material.²⁵⁾ The charge transfer resistance (R_ct) and double layer capacitance (C_dl) of each alloy negative electrode were calculated according to our previous study¹⁵⁾ and are summarized in Table 4. Especially, it is well known that R_ct is determined by the electrocatalytic activity for an alloy. R_ct for the secondary phase alloy negative electrode was much smaller than that for the mother and primary-phase alloy negative electrode, whereas C_dl for the former was much higher than that for latter. As reported previously,^26–28) the B2-type TiNi phase in the V-based V–Ti–Ni bcc (body-centered cubic) alloy or the Ti₂Ni alloy negative electrodes works as an electrocatalyst. In this study, the abundance of the secondary-phase alloy, which has B2-type TiNi cubic structure, also seems to improve the electrical conductivity. Therefore, HRD of the mother alloy electrode was better than the primary-phase alloy electrode, suggesting that the mother alloy has the B2-type secondary-phase network, as shown in Fig. 1.

Fig. 9

AC impedance spectra for the Zr_0.49Ti_0.5Nb_0.01Ni mother and the primary- and secondary-phase alloy negative electrodes after charging at 21th cycle.

Table 4 Charge transfer resistances (R_ct) and double layer capacitance (C_dl) for the Zr_0.49Ti_0.5Nb_0.01Ni mother and the primary- and secondary-phase alloy negative electrodes after charging at the 21th cycle.

Considering these results, it is clearly concluded that the B33-type orthorhombic primary-phase for the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy serves as a hydrogen reservoir, and the B2-type cubic secondary-phase serves as an electrocatalyst in the alloy.

4. Conclusion

We investigated the crystal structure, pressure-composition isotherms and electrochemical properties of each phase with AB-type Zr–Ti–Nb–Ni alloy negative electrodes, and the following findings were obtained.

The Zr_0.49Ti_0.50Nb_0.1Ni mother alloy consisted of gray and black phases assigned to the primary- and the secondary-phase alloys, respectively, and average chemical compositions of the former and latter were identified as Zr_0.54Ti_0.47Nb_0.01Ni_0.98 and Zr_0.47Ti_0.52Nb_0.01Ni, respectively. We separately synthesized the primary- and the secondary-phase alloys. The primary- and secondary-phase alloy powders were nearly indexed as the B33-type orthorhombic structure and the B2-type cubic structure.

The maximum hydrogen storage capacities at 0.9 MPa and 373 K for the mother, primary-phase and secondary phase alloys were 1.8, 1.9 and 1.1 mass%H₂, respectively.

The initial discharge capacity of the mother, primary- and secondary-phase alloy negative electrodes was 335, 362 and 253 mAh g⁻¹, respectively. The primary-phase alloy negative electrode had the highest discharge capacity among the three electrodes in this study. The cycle performance for all alloy negative electrodes was more than 95% even after 20 cycles. So they exhibited good cycle performance. The secondary-phase alloy exhibited the best HRD among the three electrodes.

We clearly concluded that the B33-type orthorhombic primary-phase for the Zr_0.49Ti_0.5Nb_0.01Ni mother alloy serves as a hydrogen reservoir, and the B2-type cubic secondary-phase serves as an electrocatalyst, respectively.

Acknowledgments

This research was funded by Aichi Steel Corporation.

REFERENCES

• 1)   P.  Bäuerlein,  C.  Antonius and  J.  Kümpers: J. Power Sources 176 (2008) 547–554.
• 2)   K.  Nishimura,  T.  Takasaki and  T.  Sakai: J. Alloys Compd. 580 (2013) S353–S358.
• 3)   W.  Li and  Y.L.  Du: Mater. Trans. 49 (2008) 2229–2232.
• 4)   H.  Chen,  J.M.  Wang,  T.  Pan,  J.Q.  Zhang and  C.N.  Cao: Int. J. Hydrogen Energy 27 (2002) 489–496.
• 5)   B.  Scrosati and  J.  Garche: J. Power Sources 195 (2010) 2419–2430.
• 6)   J.B.  Goodenough and  Y.  Kim: J. Power Sources 196 (2011) 6688–6694.
• 7)   F.  Larsson,  P.  Andresson,  A.  Lorën and  B.E.  Mellander: J. Power Sources 271 (2014) 414–420.
• 8)   A.  Lecocq,  G.G.  Eshetu,  S.  Grugeon,  N.  Martin,  S.  Laruelle and  G.  Marlair: J. Power Sources 316 (2016) 197–206.
• 9)   P.  Dantzer,  P.  Millet and  T.B.  Flanagan: Metall. Mater. Trans. A 32 (2001) 29–38.
• 10)   S.  Yasuoka,  Y.  Magari,  T.  Murata,  T.  Tanaka,  J.  Ishida,  H.  Nakamura,  T.  Nohma,  M.  Kihara,  Y.  Baba and  H.  Teraoka: J. Power Sources 156 (2006) 662–666.
• 11)   C.  Khaldi,  M.  Mathlouthi,  J.  Lamloumi and  A.  Percheron-Guégan: Int. J. Hydrogen Energy 29 (2004) 307–311.
• 12)   A.  Matsuyama,  H.  Mizutani,  T.  Kozuka and  H.  Inoue: Int. J. Hydrogen Energy 41 (2016) 9908–9913.
• 13)   A.  Matsuyama,  H.  Mizutani,  T.  Kozuka and  H.  Inoue: Batteries 2 (2016) 30.
• 14)   A.  Matsuyama,  H.  Mizutani,  T.  Kozuka and  H.  Inoue: Int. J. Hydrogen Energy 42 (2017) 22622–22627.
• 15)   A.  Matsuyama,  H.  Takitou,  T.  Kozuka and  H.  Inoue: Metals 8 (2018) 473.
• 16)   R.  Oishi,  M.  Yonemura,  Y.  Nishimaki,  S.  Torii,  A.  Hoshikawa,  T.  Ishigaki,  T.  Morishima,  K.  Mori and  T.  Kamiyama: Nucl. Instrum. Methods Phys. Res. A 600 (2009) 94–96.
• 17)   R.  Oishi-Tomiyasu,  M.  Yonemura,  T.  Morishima,  A.  Hoshikawa,  S.  Torii,  T.  Ishigaki and  T.  Kamiyama: J. Appl. Crystallogr. 45 (2012) 299–308.
• 18)   S.F.  Matar: Chem. Phys. Lett. 473 (2009) 61–65.
• 19)   F.  Cuevas,  M.  Latroche,  F.  Bourée-Vigneron and  A.  Percheron-Guégan: J. Solid State Chem. 179 (2006) 3295–3307.
• 20)   A.  Matsuyama,  H.  Mizutani,  T.  Kozuka and  H.  Inoue: J. Alloys Compd. 714 (2017) 467–475.
• 21)   J.S.  Cantrell,  R.C.  Bowman, Jr.,  L.A.  Wade,  S.  Luo,  J.D.  Clewley and  T.B.  Flanagan: J. Alloys Compd. 231 (1995) 518–523.
• 22)   F.  Cuevas,  M.  Latroche,  P.  Ochin,  A.  Dezellus,  J.F.  Fernández,  C.  Sánchez and  A.  Percheron-Guégan: J. Alloys Compd. 330–332 (2002) 250–255.
• 23)   H.  Emami,  F.  Cuevas and  M.  Latroche: J. Power Sources 265 (2014) 182–191.
• 24)   K.  Young,  T.  Ouchi,  B.  Huang,  B.  Reichman and  R.  Blankenship: J. Alloys Compd. 575 (2013) 65–72.
• 25)   N.  Kuriyama,  T.  Sakai,  H.  Miyamura,  H.  Tanaka,  I.  Uehara,  H.  Ishikawa and  T.  Iwasaki: J. Alloys Compd. 202 (1993) 183–197.
• 26)   M.  Tsukahara,  K.  Takahashi,  T.  Mishima,  T.  Sakai,  H.  Miyamura,  N.  Kurihara and  L.  Uehara: J. Alloys Compd. 226 (1995) 203–207.
• 27)   H.  Inoue,  S.  Koyama and  E.  Higuchi: Electrochim. Acta 59 (2012) 23–31.
• 28)  M.A. Gutjahr, H. Buchner, K.D. Becca and H. Säufferer: Power Sources, 4, (Oriel press, Newcastle upon tyne, 1973) pp. 79–91.