Probing the Electrochemical Mechanism of Li5FeO4 (LFO) Cathodes in Li-Ion Cells

The push for high capacity cathodes capable of Li storage greater than Li/TM (TM=transition metal) ratio of 1 in Li-ion batteries is extremely challenging.  Considering Li₂MnO₃ (2Li per TM; 459 mAhg^-1 theoretical), we have shown [1], as well as others [2], that two lithiums per Mn can be extracted in an initial charge yielding practical capacities >320 mAhg^-1, but this process occurs only above 4.5 V, and requires a material with high-surface area.  Nevertheless structural changes occur in this cathode on the first charge, and the Li₂MnO₃ phase is lost due to lithium and oxygen removal, wherein the resulting material (i.e. ‘MnO₂’) does not possess good electrochemical reversibility.  

Moving away from capacity limited layered Li(TM)O₂ requires exploring other crystalline structure types.  The antifluorite family of L₂O (Li₈O₄; enlarged unit cell) is an interesting option since transition metals can substitute for tetrahedral Li in the structure.  Note that there are more available tetrahedral interstitials for close-packed oxygen atoms (total of two per O), than for octahedral sites (total of 1 per O) thereby providing a clue about which structure types to pursue.  A representative example is Li₅FeO₄ (LFO) that features replacing a total of three lithium atoms in Li₈O₄ (enlarged unit cell) with one Fe atom (a trivalent charged cation; two vacancies are also created).  This material, if electrochemically active could, in theory, provide 5 Li/Fe or ~ 867 mAhg^-1, if made reversible.  Certainly redox on oxygen could be necessary in order to access all of this lithium and that remains a topic for exploration in hybrid Li-ion Li-oxygen cells [3]. 

In the earlier publication on LFO used as a lithium source for charged cathodes in Li-ion cells, we were able to demonstrate that a total of 4 lithiums can be extracted from the cathode material up to 4.4 V vs. Li metal.  This provided a charge capacity of ~ 690 mAhg^-1 [4].  We revisit this material for Li-ion cells and are now focused on optimizing the synthesis process and to thoroughly characterize the electrochemical-chemical mechanism that occurs on the first charge in hopes of making the cathode reversible to supply 350 mAhg^-1 on discharge above 3 V which can lead to energy densities of ca. >1000 Whkg^-1.

The electrochemical voltage profile of LFO is shown in Figure 1 for conventional (micron-sized Fe₂O₃ precursor) and a nano-Fe₂O₃ precursor.  The over-potential is lower and the extractable capacity is higher for the latter (magenta line).  The subsequent discharge to 1 V is also plotted.  It is evident that the material is irreversibly converted over to another phase that only shows Fe(III/II) redox at lower voltages below 2.5 V.  Reversibility, however, is fairly good over subsequent cycles within a large voltage window between 4 and 1 V yielding ~230 mAhg^-1 (not shown).

 Figure 1.  First charge and discharge voltage profile of Li/Li₅FeO₄cell between 4.7 and 1.0 V; C rate is C/40.

 A voltage window opening experiment was also conducted and better reversibility is seen up to 2 lithium cations suggesting that the following electrochemical reaction may take place:

 Li₅FeO₄  ↔  Li₃FeO₄   +  2 Li⁺  + 2 e^-.

 In this reaction, strictly written formally with Fe(V) product likely does not exactly proceed.  Earlier XANES/XAFS studies doesn’t invoke the Fe(IV) redox state in the oxidation process [5].  Certainly oxygen redox (lattice peroxide formation) must also be ascertained in this electrochemical mechanism [6] and is the subject of our ongoing experimentation.  These results, and are most recent findings to date will be presented in this talk.

Antiflourite Li₅FeO₄ presents an interesting material pathway and design model to high capacity and high energy battery systems and through careful scientific progress is expected to result in considerable advances in the knowledge of advanced Li-ion cathodes utilizing oxygen 2p orbital and metal d-states hybridization for charge storage.

ACKNOWLEDGEMENT

Argonne National Laboratory is operated for the U.S. Department of Energy by UChicago Argonne, LLC, under contract DE-AC02-06CH11357.  This work was supported as part of the Center for Electrochemical Energy Science (CEES), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences.

REFERENCES

1. C. S. Johnson et al. Electrochem. Commun., 6, 1085 (2004)

2. A. Robertson et al, Chem. Mater., 15, 1984 (2003)

3. M. M. Thackeray et al., J. Phys. Chem. Lett., 4, 3607 (2013)

4. C. S. Johnson et al. Chem. Mater., 22, 1263-1270 (2010)

5. T. Okumura et al., J. Mater. Chem. A , 2, 11847 (2014)

6. S. Okuoka et al., Scientific Reports, 4, 5684 (2014)