Introduction

The revolution of electrochemical energy storage took place in the 1970s, when M. Stanley Whittingham demonstrated the use of TiS₂ as a cathode material^[1]. The material was based on a process in which lithium ions diffused in between TiS₂ layers. This diffusion mechanism was later deemed the intercalation process. The next generation cathode material, lithium cobalt oxide (LiCoO₂, LCO), was produced by John Goodenough and Koichi Mizushima in the beginning of the 1980s^[2]. The work lead eventually to Sony’s commercialization of the first lithium-ion battery (LIB) in the early 1990s. In 2019 Whittingham, Mizushima and Goodenough were awarded with a Nobel prize for their efforts in battery research.

The development of battery materials has ever since been in rapid increase and several superior battery chemistries have replaced LCO nearly completely. Though LCO was at its time a revolutionary material, it still suffered from a relatively low specific capacity of only 140 mAh/g (theoretical 274 mAh/g). This was mainly due to irreversible structural changes at higher charging potentials^[3]. Several competing battery chemistries have since emerged to address performance and stability issues. Most notably, lithium nickel manganese cobalt oxide (LNMC or NMC) in varying compositions is one of the most widely used battery materials on today’s market. It possesses an impressive capacity ranging from around 160 mAh/g to 200 mAh/g depending on its composition.

The trend in NMC research has been to decrease the quantity of cobalt due to several factors. Cobalt mining is concentrated in the Democratic republic of Congo, with about 70% of the annual cobalt extraction^[4]. There mining is done in poor conditions, child labor is widely used, the local population is paid poorly for their work and the environment is polluted by the toxic cobalt^[5]. Additionally, the concentration of 70 % of cobalt production in such a small area poses a sever supply risk in case of a crisis. Furthermore, cobalt is an expensive mineral, which drives up the cost of LIBs. Thus, the replacement of cobalt in NMC with another transition metal could decrease cost, supply dependance, and toxicity. Among many elements, iron is an attractive replacement due to its abundance, low cost and relative safety. Additionally, it has a similar size to cobalt, which allows it to replace cobalt in the crystalline lattice with relative ease. A considerable number of papers have been published on lithium nickel manganese iron oxide (LNMF or NMF), which have demonstrated that the material could be a viable option for a Co-free cathode material for high energy density LIBs. The future development of the material is intriguing since it could offer a similar practical capacity to NMC, while other Co-free cathode materials on the market promise capacities similar to LCO. 

Principle of operation

The principle chemical reaction in a LIB is comparatively simple. In the discharged state, the cathode material has all of the lithium stored within its layered structure. As a potential is applied to charge the battery, these ions move out of the structure into the electrolyte. On the anode side, ions begin to intercalate into the anode material, which typically is graphite. The cathodic and anodic reactions are presented below. As about 70 % – 80 % of the lithium has diffused out of the structure, the battery is fully charged. Upon discharge, the process is reversed, and the stored energy can be harvested. The charging and discharging process is illustrated in figure 1. 

The anode and cathode reactions for NMF  are:

Cathode:

Li(Ni_xMn_yFe_z)O₂ ↔ Li_1-a(Ni_xMn_yFe_z)O₂ + a Li⁺ + a e^-

Anode:

Li⁺ + e^- + C₆ ↔ LiC₆

Figure 1. An illustration of the behavior of a LIB during charging and discharging. (Figure by Akseli Rautakorpi)

Synthesis

NMF has been synthesized with several different techniques including solid-state synthesis^[6], sol-gel^{[7][8][9][10][11][12]}, molten salt^[13], mixed hydroxide^[8][14], co-precipitation ^{[8][15][16][17]} and reverse micelle synthesis^[18]. Co-precipitation is considered to be the state-of-the-art and thus it is the one, which will be briefly discussed. The method is a two-step process, where firstly the precursor metal hydroxide is formed through a precipitation technique. The precursor is synthesized by dissolving transition metal salts, typically sulfates, and precipitating them with a suitable base such as LiOH or NaOH. The final material is formed by lithiating the sample between 700 °C and 900 °C with LiOH. The method yields particles with a favorable spherical morphology, which improves the electrochemical performance of the material. Additionally, the particles exhibit a lower surface area compared to nanomaterials, which reduced unwanted surface reactions. The main drawback of the synthesis is its considerable complexity. For repeatable results, variables such as temperature, rotation speed, pH, solution concentrations, addition rates, additives and reactions times need to be carefully controlled ^[15].

Structure

The structure of NMF consists out of repeating transition metal oxide (TMO) layers with lithium intercalated in between them with a stacking sequence of AB CA BC. The material exhibits the structure type of α-NaFeO₂ with a space group of R-3m (166) ^[11][15][17]. The oxygen atoms are packed in a face centered cubic structure and transition metal cations are in the octahedral sites. This structure is often denoted as O3, where “O” refers to octahedral coordination and “3” to the number of lithium layers in a unit cell.

Importantly, the cathode material undergoes several phase transitions as it is being charged i.e., when lithium diffuses out of the structure^[19]. Initially, the structure transitions into a distorted monoclinic O’3 phase. Further deintercalation leads to a hexagonal O1 phase. Additionally, several hybrid phases can be observed at varying levels of delithiation, denoted as H1-H3. These compose of mixed O1 and O3 phases. The structures are illustrated in figure 2.

Figure 2: An illustration of NMF phases present during charging and discharging. Color code as in Figure 1. (Figure adapted from ^[19] by Akseli Rautakorpi)

XRD is among to most important tools for analyzing prepared samples, since it can be used for not only identifying the formation of the correct phase, but also to study more carefully the quality and structure of the material. Figure 3 shows a typical XRD pattern observed for NMF 811. Samples may commonly exhibit cation mixing – cations occupying octahedral sites in lithium layers. This is commonly assessed by examining the intensity ratio between the (003) and (004) reflections (I₀₀₃/I₁₀₄)_. A value above 1 suggests little cation mixing, which enhances the performance of the material ^[8]. Additionally, the separation of the (006)/(012) and (108)/(110) reflection pairs are used to study how well the layered structure has formed with clear separation suggesting a well formed layered structure ^[8]. 

Figure 3. A typical XRD pattern observed between 2-theta values of 15 and 70 for NMF 811 with the relevant reflections marked. Reflections related to another phase (Li₅FeO₄) can be ignored between 2-theta values of 19 and 25. (Figure by Akseli Rautakorpi)

Electrochemistry

NMF has practically an equal theoretical capacity to NMC. Importantly, several papers have reported high reversible specific capacities during the initial cycles reaching values above 200 mAh/g ^{[6] [13][14] [16]}. Thus, NMF can be seen as a promising material for the transition to Co-free battery chemistries.

In NMF Ni, Mn and Fe exist in oxidation states of 2, 4 and 3 respectively ^[15]. This suggests that the Ni²⁺/Ni⁴⁺  redox pair is the most electrochemically active, being responsible for most of the electrochemical storage. Mn remains mostly unchanged during the charging and discharging, which indicates to its role of stabilizing the structure. Similarly, the Fe³⁺/Fe⁴⁺ redox couple is considered mostly electrochemically inactive contributing to the stability of the material during repeated cycles ^[8]. It should be noted that the roles of Mn and Fe are still under research with varying views. The roles of each cation may also be dependent on the stoichiometric ratio of each element, which hinders the formation of a general view on their roles. 

Interestingly, Xi et al. ^[17] studied the capacity of NMF 811 as a function of cell voltage and found that the material stores more energy at higher voltages (> 4.1 V), while storing less energy at lower potentials (< 3.9 V) when compared to NMC. This would suggest that the material reaches its full potential only at higher voltages. Thus, increasing the cut-off potential from the common 4.6 V to 4.8 V or more could yield in a notable capacity increase. Additionally, as higher potentials are used a higher energy density can be reached. Thus, a key area of development is improving the stability of NMF based batteries when charged to higher potentials. This is mainly tackled by either doping the material with other transition metals ^[11], or by surface coating it with a suitable material ^[6]. 

The lithium quantity can also be optimized to increase the capacity. The theoretical capacity of NMF is calculated with the assumption that Li and NMF are precent at a 1:1 ratio. The capacity could increase considerably if even 10% more lithium could be stored in the structure in a stable manner. Several articles highlight such work, owing to higher observed specific capacities ^{[6] [8][10][14][15][16]}.  Nonetheless, NMF requires still considerable research before it could become a competitive material on the market. Mainly its stability under repeated cycles needs to be improved, along with the ability to deliver high amounts of current rapidly. 

References