Operando ⁵⁷Fe Mössbauer and XRD investigation of Li_xMn_yFe_1-yPO₄/C composites (y = 0.50; 0.75)

Abstract

LiMn_0.50Fe_0.50PO₄ and LiMn_0.75Fe_0.25PO₄ powders have been synthesized by ceramic route. A comparative investigation of their electrochemical behaviour using operando⁵⁷Fe Mössbauer and X-ray diffraction is reported. The complementarity of operando techniques used in this study allows the monitoring of changes of the local electronic environment and the lattice modifications that are directly linked to the redox reaction mechanisms. During the charge, LiMn_0.50Fe_0.50PO₄ has been found to undergo three well defined and reversible reactions via an intermediate phase containing simultaneously Fe^II and Fe^III cations. LiMn_0.75Fe_0.25PO₄ undergoes a two reaction mechanism. Fe^III Mössbauer signature has been found to be sensitive to the oxidation of Mn^II since this oxidation is accompanied with a significant increase of the iron quadrupole splitting. This study summarizes the different mechanisms observed for compositions with different manganese content.

---

1. Introduction

Since the work on the Fe³⁺/Fe²⁺ redox couple in phospho olivines was first reported in 1997, LiFePO₄ has been intensively studied as a promising cathode material^1–4 for the next generation of lithium-ion batteries because of its low cost, safety and environmental compatibility. Numerous studies have attempted to improve the electrochemical performances of the material and to get deeper insight in the lithium electrochemical mechanism.^5–9 On the other hand, attempts have been made to enhance the potential of olivine cathode materials by using manganese phosphate instead of iron phosphate.^3,10 Despite the recent advances in cycling insulating materials, thanks to nano-coating techniques and new preparation routes^11,12 and even though some authors have reported promising results,^11–13 LiMnPO₄ is not used as active material in lithium battery. This is due to the fact that it shows much lower effective energy density than the lithium iron phosphate owing to the low practical capacity evidenced. In addition, LiMnPO₄ requires very slow charge and discharge rates.^2,14–16 The main reason is the large kinetic barrier at the mismatched interface of MnPO₄/LiMnPO₄.⁴ It is also worth noticing that during the charge, LiMnPO₄ exhibits higher volume change than LiFePO₄.^2,15,17

A good compromise between LiFePO₄ and LiMnPO₄ is the partially substituted phases LiMn_yFe_1−yPO₄,^3,15,18 that would introduce the advantage of combining good electrochemical performances of LiFePO₄ with a higher potential for manganese (∼4.1 V compared with ∼3.4 V vs. Li⁺/Li for iron) in order to enhance the energy density.

It is however not yet clear which composition has to be preferred for practical application. Yamada et al. suggested that LiMn_0.6Fe_0.4PO₄ composition exhibits the best electrochemical performances.¹⁵ LiMn_0.4Fe_0.6PO₄ has also been reported to give good specific capacity after 100 cycles.¹⁹ Chang et al. have reported that LiMn_0.3Fe_0.7PO₄ shows the excellent cycling performances,²⁰ whereas Xu et al. found that LiMn_0.1Fe_0.9PO₄ shows the best cycling performances.²¹ Despite increasing interest in the Mn substituted phases, their electrochemical mechanisms are not yet well understood. It is, for example, not clear why the Fe³⁺/Fe²⁺ and Mn³⁺/Mn²⁺ redox potentials are shifted (vs. Mn content) and why the polarizations related to these two couples show an opposite behaviour.²²

Operando techniques are increasingly employed for the investigation of the batteries electrochemical behaviour. Few analyses of LiMn_yFe_1−yPO₄ have been reported using XRD and X-ray absorption spectroscopy (XAS).^23–25

We have recently reported an operando study of Li_xMn_yFe_1−yPO₄/C composites (y = 0; 0.25).^1,26 In order to establish a global mechanism, we have prepared LiMn_0.50Fe_0.50PO₄/C and LiMn_0.25Fe_0.75PO₄/C using solid-state reaction. Operando XRD and ⁵⁷Fe Mössbauer Spectroscopy analyses have been used to investigate the evolution of iron local environment, oxidation state and lattice structure during the electrochemical redox process.

2. Experimental section

2.1 Synthesis procedure

LiMn_yFe_1−yPO₄/C (y = 0.5; 0.75) composites were obtained by solid-state reaction from Li₂CO₃, FeC₂O₄·2H₂O, Mn(COOCH₃)₂·4H₂O, NH₄H₂PO₄ and a source of carbon (all chemicals of 99% purity from Aldrich) taken in stoichiometric quantities. The precursors were first ball milled during 90 min and then thermally treated in a furnace under argon flow at 600 °C.

2.2 Characterization of materials

After preparation, the compounds were crushed in an agate mortar and powder XRD (PHILIPS X′Pert MPD θ–2θ diffractometer equipped with the X′celerator detector) patterns were recorded using Cu-Kα radiation (λ = 1.5418 Å) and a nickel filter.

The residual carbon content was evaluated by a Flash EA 1112 analyser based on the Dynamic Flash Combustion which produced complete combustion of the sample followed by an accurate and precise determination of the elemental gases produced. ⁵⁷Fe Mössbauer spectra were recorded in the constant acceleration mode and in transmission geometry on a standard Mössbauer spectrometer composed of electronic devices from Ortec and Wissel. A ⁵⁷Co(Rh) source with a nominal activity of 370 MBq was used. The source and the absorber were always kept at RT. All the isomer shifts are given relative to α-Fe at RT.

2.3 Electrochemical behaviour

Electrodes containing 80 wt.% sample and 20 wt.% carbon black were prepared for cycling tests. Standard Swagelok cells Li|1 M LiPF₆ (propylenecarbonate (PC): ethylene carbonate (EC): dimethyl carbonate (DMC) 1 : 1 : 3, v/v)| LiMn_yFe_1−yPO₄ were assembled in an argon-filled glove box and tested using a VMP system.

To identify the electrochemical reaction mechanisms, we carried out operando XRD and ⁵⁷Fe Mössbauer spectroscopy measurements for both LiMn_yFe_1−yPO₄ (y = 0.50; 0.75) electrodes using an electrochemical cell designed especially for this aim.²⁷

To preserve beryllium window from the oxidation due to high working voltage, it was covered by a 2 μm thick foil of pure aluminium (Goodfellow). Here, the cells were charged and discharged using a C/40 rate, at room temperature in the voltage ranges 2.75–4.6 V. Each operando XRD scan was recorded during 1 h in which Li extraction/insertion was performed and the results correspond to a change of composition of Δx = 0.025 Li (where x is in Li_xMn_yFe_1−yPO₄/C (y = 0.50; 0.75)). A 2 h acquisition time was used for the operando Mössbauer measurement. Each spectrum corresponds to Δx = 0.05 extracted/inserted Li. For y = 0.75 the spectra presented are obtained in situ but not operando due to the low amount of iron in the electrode.

3. Results and discussion

XRD patterns of the obtained powders are given in Fig. 1a. As it can be seen, the powders are free of any crystalline impurity.

Fig. 1 Powder X-ray diffraction patterns (a), and evolution of lattice parameters with manganese content of LiMn_yFe_1−yPO₄/C (b).

All the diffraction peaks can be indexed in the orthorhombic Pnma space group. Lebail method²⁸ has been used to estimate the lattice parameters a = 10.368(1) Å, b = 6.050(1) Å, c = 4.710(1) Å for LiMn_0.50Fe_0.50PO₄/C, and a = 10.397(1) Å, b = 6.073(1) Å, c = 4.726(1) Å for LiMn_0.75Fe_0.25PO₄/C. These values are in good agreement with the literature.³ Average crystallite sizes of 120 nm have been obtained using the Scherrer formula.²⁹ The residual carbon content was evaluated at 4.89% and 4.82% for LiMn_yFe_1−yPO₄ (y = 0.50; 0.75), respectively.

Fig. 1b gives the variation of a, b and c for the olivine phases. All the lattice parameters increase linearly with manganese content. The main effect of the substitution of iron-manganese in LiMn_yFe_1−yPO₄ is the expansion of the network that can be explained simply by the difference between the ionic radii of high spin Mn²⁺ (0.97 Å) and Fe²⁺ high spin (0.92 Å).

LiMn_yFe_1−yPO₄/C (y = 0.50; 0.75) spectra (Fig. 2) consist of sharp doublets with an isomer shift of δ = 1.235(2) and a quadrupole splitting of Δ = 2.947(3) mm s⁻¹ for y = 0.5 and an isomer shift of δ = 1.239(1) and a quadrupole splitting of Δ = 2.942(2) mm s⁻¹ for y = 0.75 as discussed previously.¹

Fig. 2 ⁵⁷Fe Mössbauer spectra for LiMn_yFe_1−yPO₄/C, where y = 0.50 and y = 0.75.

The voltage profiles of LiMn_yFe_1−yPO₄ are shown in Fig. 3a. For LiFePO₄, a flat plateau is observed at 3.4 V as previously reported. For the substituted phases, two pseudo plateaus are observed in agreement with the two redox reactions Fe³⁺/Fe²⁺ and Mn³⁺/Mn²⁺. The length of the Mn plateau is proportional to manganese content.

Fig. 3 Electrochemical curves (first cycle) (a) and rate capability (b) of LiMn_yFe_1−yPO₄.

Calculated capacities with the cycle rate for Li_xMn_yFe_1−yPO₄/C (y = 0.25, 0.50, 0.60, 0.75) are summarized in Fig. 3b. This representation allows direct visualization of the reversible capacity of the material depending on the rate of cycling. The capacity is found to be a function of both cycling rate and Mn content. For our synthesis conditions, the material with the best performance is the composition LiMn_0.25Fe_0.75PO₄/C. The loss of capacity between C/10 and C is 12% and 62%, respectively, for the compositions LiMn_0.50Fe_0.50PO₄/C and LiMn_0.75Fe_0.25PO₄/C. The electrochemical performance decreases with the manganese content: for the phase LiMn_0.75Fe_0.25PO₄/C the reversible capacity calculated for a cycle rate of C/10 is 56 mAh.g⁻¹ against 125 mAh.g⁻¹ for the substituted phase with less manganese content LiMn_0.50Fe_0.50PO₄/C.

The redox potentials have been obtained from the derivative curves (Fig. 4). These potentials associated with the redox couples Fe³⁺/Fe²⁺ and Mn³⁺/Mn²⁺ undergo an increase with the manganese content. Kobayashi et al.³⁰ explain this with an expansion of the volume of all compositions. These variations are sufficient to increase the average potential of the couple Fe³⁺/Fe²⁺ from 3.45 V to 3.56 V vs. Li⁺/Li⁰ respectively for y = 0 and y = 0.75. This variation in potential of 0.11 V may appear negligible in the case of a battery with one cell, however, it can be important in the case of batteries with many cells mounted in parallel that can be used, for example, for electrical vehicles.

Fig. 4 Average potential of redox reactions in LiMn_yFe_1−yPO₄.

To better understand the reaction mechanisms, operando analyses have been performed.

For y = 0.50 the Mössbauer spectra corresponding to the first cycle and the parameters associated with the first charge are represented, respectively, in Fig. 5a and 5b. The diffractograms corresponding to the first cycle of phase LiMn_0.50Fe_0.50PO₄ are shown in Fig. 6a. At the beginning of the charge (blue area) (0.70 ≤ x < 0.95) a new phase appears in both XRD patterns and Mössbauer data. The (020) peak of the pristine material vanishes at the expense of a new peak located at 29.3°. In this domain, the Mössbauer spectra (Fig. 7) indicates the transformation of initial Fe²⁺ into two components: the first is associated with Fe³⁺ with the parameters (δ = 0.42 mm s⁻¹ and Δ = 1.12 mm s⁻¹) and a second component associated with a new Fe²⁺ component (δ = 1.26 mm s⁻¹ and Δ = 2.72 mm s⁻¹). The quadrupole splitting of this new Fe²⁺ is lower than that of the initial Fe²⁺ indicating an increase of the distortion of the iron environment³¹ caused by the deintercalation of Li⁺ and iron oxidation. This observation allows us to conclude the formation of a mixed valence phase with the following composition: Li_0.70Mn^II_0.50Fe^II_0.20Fe^III_0.30PO₄. It is worth noting that such phases (called ferrisicklerites Li_x<1(Fe,Mn)PO₄) have been reported by mineralogists and attributed to natural oxidation of triphylite Li(Fe,Mn)PO₄.^32–34 From the study of 45 ferrisicklerite minerals, Fanton et al. have concluded on the absence of any structural order of the Mn and Fe cations in this family of materials.³²

Fig. 5 Evolution of the operando Mössbauer spectra (a), and hyperfine parameters of Fe^II and Fe^III species (b), of Li_xMn_0.5Fe_0.5PO₄/C recorded during the electrochemical process recorded at C/40 rate.

Fig. 6 Operando X-ray diffractograms of LiMn_0.50Fe_0.50PO₄/C (a), and LiMn_0.75Fe_0.25PO₄/C (b).

Fig. 7 Mössbauer spectra of Li_xMn_0.50Fe_0.50PO₄/C obtained at the indicated x values.

From these observations we conclude a biphasic mechanism with the coexistence of LiMn^II_0.50Fe^II_0.50PO₄ and Li_0.70Mn^II_0.50Fe^II,III_0.50PO₄ for the charge and discharge in the range 0.70 ≤ x < 0.95.

The electrochemical reaction in this region can be written:

Li_1.0Mn^II_0.50Fe^II_0.50PO₄ ↔ (α)Li_0.70Mn^II_0.50Fe^II_0.20Fe^III_0.30PO₄ + (1 − α)Li_1.0Mn^II_0.50Fe^II_0.50PO₄ + 0.30 αLi⁺ 0.30αe⁻ (where 0 < α ≤ 1)

In the second region (0.50 ≤ x ≤ 0.70), shown in black in Fig. 6a, the diffraction lines corresponding to the Li_0.70Mn_0.50Fe_0.50PO₄ phase are shifted to upper angles, indicating the existence of a solid solution domain in charge and discharge. This also is supported by gradual increase (∼21%) of the quadrupole splitting of Fe³⁺ with the oxidation of iron. Therefore, in this area the electrochemical reaction can be written as follows:

Li_0.70Mn^II_0.50Fe^II_0.20Fe^III_0.30PO₄ ↔ Li_0.70−αMn^II_0.50Fe^II_0.20−αFe^III_0.30+αPO₄ + αLi⁺ + αe⁻ (where 0 ≤ α ≤ 0.2)

In the third region (∼0.16 ≤ x < 0.5, shown in red in Fig. 6a, the two-phase character is more evident. A new phase appears from x = 0.50 (Li_0.50Mn^III_0.50Fe^III_0.50PO₄) with diffraction peaks located at angles higher than the initial phase:

Li_0.50Mn^II_0.50Fe^III_0.50PO₄ ↔ βMn^III_0.50Fe^III_0.50PO₄ + (1 − β)Li_0.50Mn^II_0.50Fe^III_0.50PO₄ + 0.50βLi⁺ + 0.50βe⁻ (where 0 < β ≤ 1)

All the Fe²⁺ was oxidized into Fe³⁺ at the end of the second region, however, the Mössbauer spectra continue to evolve (Fig. 7), indicating that the environment of iron is affected by the oxidation of Mn²⁺ in Mn³⁺ as already observed for the phase LiMn_0.25Fe_0.75PO₄.¹

From x ≤ 0.50 the spectra (Fig. 7) are refined by adding a new Fe³⁺ doublet shown in magenta. The new doublet corresponding to Fe³⁺ phase Mn^III_0.50Fe^III_0.50PO₄ exhibits a quadrupole splitting of 1.65 mm s⁻¹, which is higher than 1.55 mm s⁻¹ observed for Mn^III_0.25Fe^III_0.75PO₄, and this further confirms that the Fe³⁺ is very sensitive to the oxidation state of Mn since an increase of the amount of Mn³⁺ around the Fe³⁺ site leads to a significant increase of the quadrupole splitting.³¹ Similar spectra have been reported for natural purpurite (Fe,Mn)PO₄.³¹ The electromechanism of the charge and discharge of LiMn_0.50Fe_0.50PO₄/C is done in three steps as the LiMn_0.25Fe_0.75PO₄/C phase. At the beginning, the mechanism is biphasic, with the coexistence of LiMn^II_0.50Fe^II_0.50PO₄ and a mixed composition Li_0.70Mn^II_0.50Fe^II_0.20Fe^III_0.30PO₄ up to x = 0.50. From x = 0.50, the second domain is a solid solution where the phase Li_xMn^II_0.50Fe^II_0.50Fe^III_0.30PO₄ oxidizes up to x = 0.16. The last domain is biphasic between Li_0.50Mn^II_0.50Fe^III_0.50PO₄ and Mn^II_0.50Fe^III_0.50PO₄.

The electrochemical mechanism corresponding to the richest manganese phase LiMn_0.75Fe_0.25PO₄/C was also studied. The diffractograms corresponding to the first cycle with the galvanostatic curves are shown in Fig. 6b. The three ⁵⁷Fe Mössbauer spectra (Fig. 8) do not match the operando mode, indeed the low iron content of this phase would not have sufficient quality for data processing. They were, however, recorded in situ with the battery paused at key points.

Fig. 8 Mössbauer spectra of Li_xMn_0.75Fe_0.25PO₄/C obtained at the indicated x values.

From x = 0.95 the diffraction line corresponding to the plane (020) phase LiMn_0.75Fe_0.25PO₄/C (29.2° (2θ)) shifts to higher angles up to x = 0.75 (black area of Fig. 6b). The reaction corresponding to the oxidation of iron is single phase in this area (0.45 ≤ x < 0.75) for the charge and discharge:

Li_1.0Mn^II_0.75Fe^II_0.25PO₄ ↔ Li_1−αMn^II_0.75Fe^II_0.25+αFe^III_αPO₄ +αLi⁺ + αe⁻(where 0 < α ≤ 0.25)

After x = 0.75, the Mössbauer spectra show clearly that all the Fe²⁺ has been oxidized to Fe³⁺ (Fig. 8).

From x = 0.75, the potential is 4.15 V, an olivine-type phase appeared, the main diffraction peak (2θ ∼ 30.2°) corresponding to this new phase is shown in Fig. 6b. The coexistence of two phases in charge and discharge is established for the oxidation-reduction of Mn³⁺/Mn²⁺ in the area shown in red. The electrochemical reaction in (0.45 ≤ x < 0.75) can be written:

Li_0.75Mn^II_0.75Fe^III_0.25PO₄ ↔ βMn^III_0.75Fe^III_0.25PO₄ + (1 − β) Li_0.75Mn^II_0.75Fe^III_0.25PO₄ + 0.25βLi⁺ + 0.25βe⁻ (where 0 < β ≤ 1)

Mössbauer spectra shown in Fig. 8 (where x = 0.60 and 0.45) correspond to the spectra of Fe³⁺ on the “plateau” of redox reaction Mn³⁺/Mn²⁺. Indeed, it was interesting to see the indirect effect of Mn³⁺ on the environment of Fe³⁺ as in previous phases with less manganese. The spectra cannot be refined with a single component in this area, we must add a second doublet for a correct refinement. Mössbauer parameters for the two Fe³⁺ components (Li_0.75Mn_0.75Fe_0.25PO₄ and Mn_0.75Fe_0.25PO₄) are shown in Table 1. Component Mn_0.75Fe_0.25PO₄ is characterized by a quadrupole splitting of 1.705 mm s⁻¹.

Table 1 Hyperfine parameters of the composition involved during the electrochemical charge process

Composition	δ	Δ
	mm s^−1	mm s^−1
Li0.70Mn^II0.50Fe^II,III0.50PO4	1.32	2.72
	0.43	1.02
Li0.50Mn^II0.50Fe^II,III0.50PO4	0.43	1.25
Mn^III0.50Fe^III0.50PO4	0.43	1.64
Li0.55Mn^II0.75Fe^II,III0.25PO4	1.28	2.71
	0.44	1.08
Li0.25Mn^II0.75Fe^III0.25PO4	0.41	1.30
Mn^III0.75Fe^III0.25PO4	0.41	1.52

---

In conclusion, the electrochemical mechanism of phase Li_xMn_0.75Fe_0.25PO₄/C differs from the three previously studied compositions. The first part corresponding to the domain (0.75 ≤ x < 0.95) is single-phase, the phase LiMn_0.75Fe_0.25PO₄ evolves to the composition Li_0.75Mn_0.75Fe_0.25PO₄. The second area (0.45 ≤ x < 0.75) which is located at the “plateau” of the reaction Mn³⁺/Mn²⁺ is biphasic between the two compositions Li_0.75Mn_0.75Fe_0.25PO₄ and Mn_0.75Fe_0.25PO₄ for the charge and discharge.

To summarize, all the mechanisms observed for these substituted phases LiMn_yFe_1−yPO₄/C in charge are respectively represented in Fig. 9. Other compositions LiMn_yFe_1−yPO₄/C (y = 0.40, 0.60) prepared under the same conditions have also been studied and can complete the graph of Fig. 9. The regions corresponding to different types of mechanism are shown as a function of lithium content:

Fig. 9 Schematic diagram of the different mechanisms observed during the first charge.

-The beginning of the charge corresponding to the reaction Fe²⁺ → Fe³⁺ is biphasic for all compositions, except LiMn_0.75Fe_0.25PO₄/C where the reaction is a single-phase.

-A domain of solid solution is also observed for all compositions, except LiMn_0.75Fe_0.25PO₄/C between the two “plateaus” corresponding to the two redox reactions, this domain decreases with the manganese content.

-The end of the charge where the potential is around 4.2 V corresponding to the Mn³⁺/Mn²⁺ reaction is biphasic for all the compositions.

An inductive effect has already been established in some intercalation compounds polyanionic lithium,³⁵ it can result in changes in potential of more than 1 V. In the case of compounds LiMn_yFe_1−yPO₄/C, the ionicity or covalent bonds (Fe, Mn–O) are controlled by the size of the network (the length of these bonds). It was therefore interesting to compare the values of the isomer shifts of phases LiMn_yFe_1−yPO₄/C with the values of potentials associated with the couple Fe³⁺/Fe²⁺ (Fig. 10). Indeed, there is a correlation between the inductive effect and value of the isomer shift as demonstrated by Menil.³⁶ The representation of values of δ and potentials in Fig. 10 clearly shows a very similar tendency of the two parameters. This interesting result suggests that it is possible to predict the reaction potential from iron Mössbauer data as reported by Naille et al. for tin compounds.³⁷

Fig. 10 Evolution of the isomer shift vs. average potential.

Finally we observed an excellent capacity retention over 100 cycles for all the studies compositions (see Add. Data). Although the substitution of iron with manganese does not improve the capacity (since it decreases with the manganese content), the possibility of tuning the working voltage may be useful for some applications as electric hybrid cars.

4. Conclusion

In this paper we report a comparative operando study for LiMn_yFe_1−yPO₄/C for y = 0.50 and y = 0.75 combining XRD and ⁵⁷Fe Mössbauer spectroscopy. The electrochemical mechanisms have been studied by monitoring the evolution of the local electronic environment (Mössbauer probe) and the lattice changes (XRD). The solid solution phase LiMn_0.50Fe_0.50PO₄ exhibits a complex three steps mechanism during the charge. First, a biphasic reaction in the domain (0.70 ≤ x < 0.95) between LiMn^II_0.50Fe^II_0.50PO₄ and Li_0.70Mn^II_0.50Fe^II_0.20Fe^III_0.30PO₄, the Mössbauer probe helps to determine the composition of this mixed valence phase. The reaction is followed by a solid solution in the domain (0.50 ≤ x ≤ 0.70). The third step is described by a biphasic reaction between Li_0.50Mn^II_0.50Fe^III_0.50PO₄ and Mn^III_0.50Fe^III_0.50PO₄ until x ∼ 0.1. In conclusion, the electrochemical mechanism of phase Li_xMn_0.75Fe_0.25PO₄/C differs from the three previously studied compositions. The first part corresponding to the domain (0.75 ≤ x < 0.95) is single-phase, the phase LiMn_0.75Fe_0.25PO₄ evolves to the composition Li_0.75Mn_0.75Fe_0.25PO₄ followed by a two phase mechanism with LiMn_0.25Fe_0.75PO₄ and LiMn_0.50Fe_0.50PO₄. The second area (0.45 ≤ x < 0.75) which is located at the “plateau” of the reaction Mn³⁺/Mn²⁺ is biphasic between the two compositions Li_0.75Mn_0.75Fe_0.25PO₄ and Mn_0.75Fe_0.25PO₄ for the charge and discharge. An interesting correlation has been established between the isomer shift and the average potential of the Fe³⁺/Fe²⁺ redox reaction, which can be used to predict the potential indirectly from Mössbauer data.

Acknowledgements

The authors would like to thank SAFT Company and CNRS (contract SAFT/CNRS No. 029888) through the Ph.D. grant of Alexis Perea and ANR PHOSPHALION (National Research Agency) Stock-E program (ANR-09-STOCK-E-07) for financial support and Manfred Womes for fruitful discussions. Région Languedoc-Roussillon is gratefully acknowledged for the financial support to X-rays and gamma-rays platform (Contracts No. 2006 Q-086 and 2008 094192).

References

1. A. Perea, M. T. Sougrati, C. M. Ionica-Bousquet, B. Fraisse, C. Tessier, L. Aldon and J.-C. Jumas, Operando ⁵⁷Fe Mossbauer and XRD investigation of Li_xMn_yFe_1-yPO₄/C composites (y = 0; 0.25), RSC Adv., 2012, 2, 2080–2086.
2. C. Delacourt, P. Poizot, M. Morcrette, J. M. Tarascon and C. Masquelier, One-step low-temperature route for the preparation of electrochemically active LiMnPO₄ powders, Chem. Mater., 2004, 16(1), 93–99.
3. A. K. Padhi, K. S. Nanjundaswamy and J. B. Goodenough, Phospho-olivines as positive-electrode materials for rechargeable lithium batteries, J. Electrochem. Soc., 1997, 144(4), 1188–1194.
4. C. Delacourt, L. Laffont, R. Bouchet, C. Wurm, J. B. Leriche, M. Morcrette, J. M. Tarascon and C. Masquelier, Toward understanding of electrical limitations (electronic, ionic) in LiMPO₄ (M = Fe, Mn) electrode materials, J. Electrochem. Soc., 2005, 152, A913–A921.
5. C. Delmas, M. Maccario, L. Croguennec, F. Le Cras and F. Weill, Lithium deintercalation in LiFePO₄ nanoparticles via a domino-cascade model, Nat. Mater., 2008, 7(8), 665–671.
6. R. Dedryvere, M. Maccario, L. Croguennec, F. Le Cras, C. Delmas and D. Gonbeau, X-Ray Photoelectron Spectroscopy Investigations of Carbon-Coated Li_xFePO₄ Materials, Chem. Mater., 2008, 20(22), 7164–7170.
7. C. Delacourt, J. Rodriguez-Carvajal, B. Schmitt, J. M. Tarascon and C. Masquelier, Crystal chemistry of the olivine-type Li_xFePO₄ system (0 ≤ x ≤ 1) between 25 and 370 degrees C, Solid State Sci., 2005, 7(12), 1506–1516.
8. A. Yamada, H. Koizumi, N. Sonoyama and R. Kanno, Phase change in Li_xFePO₄, Electrochem. Solid-State Lett., 2005, 8(8), A409–A413.
9. L. Aldon, A. Perea, M. Womes, C. M. Ionica-Bousquet and J. C. Jumas, Determination of the Lamb-Mossbauer factors of LiFePO₄ and FePO₄ for electrochemical in situ and operando measurements in Li-ion batteries, J. Solid State Chem., 2010, 183(1), 218–222.
10. F. Zhou, K. S. Kang, T. Maxisch, G. Ceder and D. Morgan, The electronic structure and band gap of LiFePO₄ and LiMnPO₄, Solid State Commun., 2004, 132((3–4)), 181–186.
11. G. H. Li, H. Azuma and M. Tohda, LiMnPO₄ as the cathode for lithium batteries, Electrochem. Solid-State Lett., 2002, 5(6), A135–A137.
12. N. H. Kwon, T. Drezen, I. Exnar, I. Teerlinck, M. Isono and M. Graetzel, Enhanced electrochemical performance of mesoparticulate LiMnPO₄ for lithium ion batteries, Electrochem. Solid-State Lett., 2006, 9(6), A277–A280.
13. C. Wang, S. Lu, S. Kan, J. Pang, W. Jin and X. Zhang, Enhanced capacity retention of Co and Li doubly doped LiMn₂O₄, J. Power Sources, 2009, 189(1), 607–610.
14. M. Yonemura, A. Yamada, Y. Takei, N. Sonoyama and R. Kanno, Comparative kinetic study of olivine Li_xMPO₄ (M = Fe, Mn), J. Electrochem. Soc., 2004, 151(9), A1352–A1356.
15. A. Yamada, Y. Kudo and K. Y. Liu, Reaction mechanism of the olivine-type Li_x(Mn_0.6Fe_0.4)PO₄ (0 ≤ x ≤ 1), J. Electrochem. Soc., 2001, 148(7), A747–A754.
16. J. S. Yang and J. J. Xu, Synthesis and characterization of carbon-coated lithium transition metal phosphates LiMPO₄ (M = Fe, Mn, Co, Ni) prepared via a nonaqueous sol–gel route, J. Electrochem. Soc., 2006, 153(4), A716–A723.
17. S. W. Kim, J. Kim, H. Gwon and K. Kang, Phase Stability Study of Li_1-xMnPO₄ (0 ≤ x ≤ 1) Cathode for Li Rechargeable Battery, J. Electrochem. Soc., 2009, 156(8), A635–A638.
18. A. Yamada, M. Hosoya, S. C. Chung, Y. Kudo, K. Hinokuma, K. Y. Liu and Y. Nishi, Olivine-type cathodes achievements and problems, J. Power Sources, 2003, 119, 232–238.
19. Y. J. Shin, J. K. Kim, G. Cheruvally, J. H. Ahn and K. W. Kim, Li(Mn_0.4Fe_0.6)PO₄ cathode active material: Synthesis and electrochemical performance evaluation, J. Phys. Chem. Solids, 2008, 69((5–6)), 1253–1256.
20. J. Xu, G. Chen, H. J. Li and Z. S. Lv, Direct-hydrothermal synthesis of LiFe_1-xMn_xPO₄ cathode materials, J. Appl. Electrochem., 2010, 40(3), 575–580.
21. X. Y. Chang, Z. X. Wang, X. H. Li, L. Zhang, H. J. Guo and W. J. Peng, Synthesis and performance of LiMn_0.7Fe_0.3PO₄ cathode material for lithium ion batteries, Mater. Res. Bull., 2005, 40(9), 1513–1520.
22. G. Kobayashi, A. Yamada, S. Nishimura, R. Kanno, Y. Kobayashi, S. Seki, Y. Ohno and H. Miyashiro, Shift of redox potential and kinetics in Li_x(Mn_yFe_1-y)PO₄, J. Power Sources, 2009, 189(1), 397–401.
23. K. W. Nam, W. S. Yoon, K. Zaghib, K. Y. Chung and X. Q. Yang, The phase transition behaviors of Li_1-xMn_0.5Fe_0.5PO₄ during lithium extraction studied by in situ X-ray absorption and diffraction techniques, Electrochem. Commun., 2009, 11(10), 2023–2026.
24. T. Nedoseykina, M. G. Kim, S. A. Park, H. S. Kim, S. B. Kim, J. Cho and Y. Lee, In situ X-ray absorption spectroscopic study for the electrochemical delithiation of a cathode LiFe_0.4Mn_0.6PO₄ material, Electrochim. Acta, 2010, 55(28), 8876–8882.
25. Y. C. Chen, J. M. Chen, C. H. Hsu, J. F. Lee, J. W. Yeh and H. C. Shih, In situ synchrotron X-ray absorption studies of LiMn_0.25Fe_0.75PO₄ as a cathode material for lithium ion batteries, Solid State Ionics, 2009, 180((20–22)), 1215–1219.
26. L. Aldon and A. Perea, 2D-correlation analysis applied to in situ and operando Mossbauer spectroscopy, J. Power Sources, 196(3), 1342–1348.
27. J. B. Leriche, S. Hamelet, J. Shu, M. Morcrette, C. Masquelier, G. Ouvrard, M. Zerrouki, P. Soudan, S. Belin, E. Elkaim and F. Baudelet, An Electrochemical Cell for Operando Study of Lithium Batteries Using Synchrotron Radiation, J. Electrochem. Soc., 2010, 157(5), A606–A610.
28. J. Rodríguez-Carvajal, Recent advances in magnetic structure determination by neutron powder diffraction, Phys. B, 1993, 192((1–2)), 55–69.
29. J. P. Eberhart, Méthodes physiques d′étude des minéraux et des matériaux solides, DOIN edn, p. 1976.
30. G. Kobayashi, A. Yamada, S. i. Nishimura, R. Kanno, Y. Kobayashi, S. Seki, Y. Ohno and H. Miyashiro, Shift of redox potential and kinetics in Li_x(Mn_yFe_1-y)PO₄, J. Power Sources, 2009, 189(1), 397–401.
31. Z. G. Liu, D. W. Dong, M. L. Liu, Y. Sui, W. H. Su, Z. N. Qian and Z. Li, Quadrupole splitting distributions in purpurite and related minerals, Hyperfine Interact., 2005, 163((1–4)), 13–27.
32. F. Fontan, P. Huvelin, M. Orliac and F. Permingeat, Ferrisicklerite of sidi bou othmane pegmatites (jebilet, morokko) and mineral group with a triphylite structure, Bulletin De La Societe Francaise Mineralogie Et De Cristallographie, 1976, 99(5), 274–286.
33. Z. Liu, D. Dong, M. Liu, Y. Sui, W. Su, Z. Qian and Z. Li, Quadrupole Splitting Distributions in Purpurite and Related Minerals, Hyperfine Interact., 2005, 163(1), 13–27.
34. A. Alberti, The crystal structure of ferrisicklerite, Li_x≤1(Fe³⁺, Mn²⁺)PO₄, Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1976, 32(OCT15), 2761–2764.
35. A. K. Padhi, K. S. Nanjundaswamy, C. Masquelier, S. Okada and J. B. Goodenough, Effect of structure on the Fe³⁺/Fe²⁺ redox couple in iron phosphates, J. Electrochem. Soc., 1997, 144(5), 1609–1613.
36. F. Menil, Systematic trends of the ⁵⁷Fe Mössbauer isomer shifts in (FeO_n) and (FeF_n) polyhedra. Evidence of a new correlation between the isomer shift and the inductive effect of the competing bond T-X (→Fe) (where X is O or F and T any element with a formal positive charge), J. Phys. Chem. Solids, 1985, 46(7), 763–789.
37. S. Naille, J. C. Jumas, P. E. Lippens and J. Olivier-Fourcade, ¹¹⁹Sn Mössbauer parameters as predictive tool for future Sn-based negative electrode materials, J. Power Sources, 2009, 189(1), 814–817.

---

Footnote

† Electronic Supplementary Information (ESI) available. See DOI: 10.1039/c2ra20949g/

---