ZnFe₂O₄, a Green and High-Capacity Anode Material for Lithium-Ion Batteries: A Review

Featured Application

The review, reporting the state of the art on the promising anode material ZnFe₂O₄, can be of interest for the community of researchers working on lithium ion batteries.

Abstract

Ferrites, a broad class of ceramic oxides, possess intriguing physico-chemical properties, mainly due to their unique structural features, that, during these last 50–60 years, made them the materials of choice for many different applications. They are, indeed, applied as inductors, high-frequency materials, for electric field suppression, as catalysts and sensors, in nanomedicine for magneto-fluid hyperthermia and magnetic resonance imaging, and, more recently, in electrochemistry. In particular, ZnFe₂O₄ and its solid solutions are drawing scientists’ attention for the application as anode materials for lithium-ion batteries (LIBs). The main reasons are found in the low cost, abundance, and environmental friendliness of both Zn and Fe precursors, high surface-to-volume ratio, relatively short path for Li-ion diffusion, low working voltage of about 1.5 V for lithium extraction, and the high theoretical specific capacity (1072 mAh g⁻¹). However, some drawbacks are represented by fast capacity fading and poor rate capability, resulting from a low electronic conductivity, severe agglomeration, and large volume change during lithiation/delithiation processes. In this review, the main synthesis methods of spinels will be briefly discussed before presenting the most recent and promising electrochemical results on ZnFe₂O₄ obtained with peculiar morphologies/architectures or as composites, which represent the focus of this review.

1. Introduction

The term ferrite indicates a broad class of ceramic oxide materials, joined by the presence of iron ions and largely applied during the last five decades [1]. They are well-known and -studied materials with an impressively wide range of applications (Figure 1) extending from millimeter wave integrated circuitry to power handling, simple permanent magnets and magnetic recording [2], catalysis [3], sensors [4], energy [5], and in nanomedicine from magnetic hyperthermia to drug delivery [6] and imaging in magnetic resonance [7] up to electrochemistry (for rechargeable ion batteries) [8].

A unique combination of properties makes ferrites so versatile: significant saturation magnetization, high electrical resistivity, low electrical losses, and very good chemical and structural stabilities [1]. Ferrites are commonly broadly divided into three groups based on the crystal structure: cubic spinels with AB₂O₄ stoichiometry, garnets, and hexa-ferrites. The key factor for tailoring their properties for so many different applications is the virtually unlimited number of solid solutions that can be obtained. In addition, the possibility to produce ferrites in the form of nanoparticles has opened a new and exciting research field, with revolutionary applications not only in electronic technology but also in the field of biotechnology. The crystal structures of garnets and hexa-ferrites are relatively complex. The first ones have a cubic structure and derive from the garnet mineral, Mn₃Al₂Si₃O₁₂. The magnetic garnets include Fe³⁺ instead of Al and Si and a rare earth cation (R) substitute Mn ion, to give the general formula R₃Fe₅O₁₂ for ferromagnetic garnets [9]. Hexa-ferrites are a family of synthetic compounds with hexagonal or rhombohedral symmetry, also forming extended solid solutions. For example, barium hexa-ferrite (BaFe₁₂O₁₉) possesses the same crystal structure of the natural mineral magneto-plumbite. The magnetic structure of hexa-ferrites is very complex due to the crystal structure complexity: their most interesting feature is the high coercivity [10].

However, cubic spinels are undoubtedly the most known and exploited ferrites, so that worldwide the term ferrite is synonymous for the AB₂O₄ compounds, that will be the subject of this review.

Firstly, a brief description of the crystal structure of spinels and their main application fields will be provided. Then, we will focus on the zinc ferrite ZnFe₂O₄ (ZFO) that, in these last years, has aroused a renewed interest in very different fields, especially nanomedicine, with an intriguing use in magnetic hyperthermia and electrochemistry, as anode material for lithium-ion batteries (LIBs). In particular, the application in LIBs will be deepened: the most promising synthesis methods for producing pure or composite/hybrid-based ZnFe₂O₄ nanomaterials, with structural and morphological properties particularly suitable for the electrochemical long-term cycling, will be analyzed and thoroughly discussed. The main aim of the review is filling up a lack of inclusive papers on the electrochemical properties of ZnFe₂O₄ by summarizing the most promising electrochemical findings of the state of the art of the last years.

2. Spinel Structure

Spinel ferrites possess the highly stable crystal structure of the natural spinel MgAl₂O₄ [1]. An extremely large variety of oxides can adopt it, fulfilling the conditions of the overall cations/anions ratio of 3/4, a total cation valence of eight, and relatively small cationic radii. A universally known ferrite is magnetite, Fe²⁺Fe³⁺₂O₄ (Fe₃O₄), probably the oldest magnetic solid with practical applications still currently studied due to the fascinating properties associated with the coexistence of ferrous and ferric cations. The oxygen network in the ferrite has an FCC (Face-Centered Cubic) symmetry, in which two types of interstitial sites, 64 tetrahedral and 32 octahedral, can be defined for a unit cell containing eight times the basic formula AB₂O₄. Only one-eighth of tetrahedral and half of octahedral sites are occupied by cations and the resulting space group is $Fd\overline{3}m$.

It is defined as normal spinel the one in which the B³⁺ and A²⁺ cations occupy octahedral and tetrahedral sites, respectively (Figure 2, left), while the inverse spinel has a radically different cation distribution, where half of the trivalent cations occupy the A sites, while the B sites are shared by the divalent and the remaining trivalent cations (Figure 2, right). Complex spinels, in which bivalent and trivalent cations occupy in a random way the tetrahedral and octahedral sites, can be also stabilized.

The described polyhedral arrangement is consistent not only with the high-symmetry space group $Fd\overline{3}m$ but also with several other space groups ranging from P4₁32, P4₁2₁2, I4₁/amd, I4₁/a, and $P\overline{4}m2$ to R3 [11].

The cation distribution in the spinel structure is controlled by many factors: elastic energy, the lattice deformation produced by the differences of cationic radii, electrostatic energy, due to the electrical charge distribution in the crystal and the crystal field stabilization energy, deriving from the geometry of the d orbitals. Magnetic ordering in ferrites produces antiferromagnetic arrangements even if, in most cases, a resulting magnetic moment is maintained due to the different numbers of cations hosted in the magnetic lattice. Some examples of the most studied and applied ferrites are represented by ZnFe₂O₄, CoFe₂O₄, Fe₃O₄, NiFe₂O₄, MgFe₂O_4, and MnFe₂O₄ [1,8,12,13] with very different magnetic properties. For example, ZnFe₂O₄ is a normal spinel with an anti-ferromagnetic character, [12,14] which also exists in nature, known as Franklinite mineral, discovered in 1819 [15]. NiFe₂O₄ is, instead, an inverse spinel with ferromagnetic ordering [16]. Almost all the spinels can form an extremely wide variety of total solid solutions due to the easiness of the spinel structure to host different kinds of cations in tetrahedral and octahedral sites, as well as in usually empty interstitial sites. This feature opens the possibility to strongly modify the chemical composition, maintaining practically unchanged the basic crystalline structure (see also the following paragraphs) [1]. The proper choices of the experimental synthesis conditions, as well as the stoichiometry and the kind of substituents, are the key parameters to produce spinels with a huge variety of functional properties, in particular, optical, magnetic, and electrochemical ones.

3. Application Fields of Ferrite

The spinel ferrites have a broad range of very different applications [17]. They are used as inductors (as low-noise amplifiers, filters, voltage-controlled oscillators, and impedance-matching networks) [18], high-frequency materials (circulators, isolators, phase shifters, and antennas) [9], power supplies (computers, peripherals, TV, and video systems), [19] and Electromagnetic Interference (EMI) Suppression [20] and in biosciences. Many biotechnological applications are based on magnetic nanoparticles. For example, nanoparticles have been used to guide radionuclides to specific tissues as well as in magnetic resonance imaging (MRI). Magnetite superparamagnetic particles are selectively associated with healthy regions of some tissues, liver for instance [6,7,21] (Figure 3).

Thermal energy from hysteresis loss of ferrites can be used in hyperthermia (Figure 3), a technique that makes use of heating of specific tissues or organs to treat cancer: The temperature in tumour tissues rises and more sensitivity to radio- or chemotherapy can be attained [21].

A more recent field of application of spinels is electrochemistry. In these last several years, many spinels have been proposed as electrodes for lithium-ion batteries. We can cite ZnCoO₄, LiMn₂O₄, Fe₃O₄, NiFe₂O₄, Li₄Ti₅O₁₂, and LiV₂O₄ applied as cathodes or anodes in LIBs, but ZnFe₂O₄ stands out from other analogous compounds as the most competitive [22,23]. The main reasons are found in the low cost, abundance, and environmental friendliness of both Zn and Fe precursors, the high surface-to-volume ratio, the relatively short path for Li-ion diffusion, the low working voltage of about 1.5 V for lithium extraction and, most importantly, the high theoretical specific capacity (1072 mAh g⁻¹) [24,25].

The first report about ZnFe₂O₄ as anode is quite recent and dates to 2004 [26]. Up to now, about 200 papers have been published on this topic [27,28,29,30]. However, a large part of works are only academic researches because its practical application is still seriously hampered by many factors: fast capacity fading and poor rate capability, resulting from an inherent low electronic conductivity, severe agglomeration, and large volume change during the lithiation/delithiation processes [12,22,23]. To overcome these limits, different strategies have been proposed, even though the most satisfying and practicable is represented by the coating/embedding of ferrite particles in various architectures with carbon sources (Figure 4) [31,32,33,34,35,36].

The doping route, which is still however scarcely exploited, is also proposed as a successful strategy for improving the electrochemical performances of ferrites, particularly to increase their poor electronic conductivity. The most encouraging dopings of ZnFe₂O₄ were performed with cations, such as Mn, Ni, Co, Mg, and Cu ions [37,38,39,40], to cite the most used. Anionic doping is not common for the ferrite phases, while more diffused is the insertion of N and/or S species on the carbon sources used to obtain composites [41,42,43]. A key role is played by the synergistic effect of N and S elements’ co-doping: [32,43] N ions can bring more defect points into the carbon lattice, while S ion has a lone pair of electrons leading to the polarization of molecules, and, consequently, the chemical activity of the anode is improved thanks to the redistribution of electric and spin density. In addition, the use of a 3D graphene structure can help to accommodate the huge volume fluctuations and improve the electrolyte diffusion and the intercalation of Li⁺ ions into the active phase. More recently, other dopants were also investigated (Ca, Al, and Gd ions) with encouraging electrochemical results [44].

The doping with transition metal ions could offer the advantage, among others, to promote additional redox reactions in the same potential window of Zn and Fe ions, increasing the lithium amount that can be intercalated/deintercalated in the spinel matrix and the capacity of the battery. For example, it has been reported that manganese ions can produce lower discharge potentials and better cycling with respect to the pure ZFO, while nickel ions cause appreciable changes in structure and morphology that, in turn, can positively influence the electrochemical properties [45,46].

The evaluation and comprehension of electrochemical phenomena cannot ignore the knowledge of crystalline phases constituting the sample. To this aim, the most suitable technique is the X-ray powder diffraction (XRPD) that allows determining the sample purity and the eventual presence of unreacted reagents or oxides such as ZnO and hematite Fe₂O₃ [40,47]. The presence of magnetite Fe₃O₄, having the same crystal structure and similar lattice parameter of ZFO, is difficult to determine with XRPD, and, in general, spectroscopic or magnetic measurements are the most suitable. Evenly fundamental is the determination of lattice parameters as well as crystallite sizes: the Rietveld structural refinement on powder diffraction patterns can easily provide this information as well as the weight percentages of the phases and the occupancies of cationic sites, allowing us to determine the possible inversion of the spinel [40,48,49]. For example, the inversion in the ZFO spinel was found in the cases of Ca, Al, Mg, and Ga doping [44,50].

4. Common Synthesis Methods of ZnFe₂O₄

In the following, the most promising methods to produce spinels with structural, morphological, and electronic conductivity properties suitable for the electrochemical application will be discussed. Broad groups of syntheses will be identified, depending on the experimental setup.

4.1. Wet Chemistry Methods

Hydrothermal, solvothermal, sol-gel, and co-precipitation are the most exploited wet synthesis methods [51,52]. They share the use of a liquid solvent and a series of peculiar aspects, such as the easy tuning of the particle size and shape, a high homogeneity of the products, and the easiness to obtain nanoparticles. The main advantage of hydrothermal synthesis, over other conventional wet-chemistry methods, is that it happens under non-standard conditions, so that unconventional crystallization pathways can occur. Together with the numerous advantages, a drawback for the hydrothermal method could be represented by the need that the synthesizing compounds are not sensitive to aqueous atmospheres. In addition, attention should be paid to a proper choice of precipitating agents that can influence the physical and chemical characteristics of spinel; the eventual defects of the obtained spinels can be removed by additional thermal treatments, prolonging the synthesis time. Many factors such as the temperature, reagents’ ratio, solvents, and the kind of salts can be varied to customize the size and shape of the ferrite nanoparticles. It is common to start from zinc and iron nitrates, [38] chlorides, or acetates [53] dissolved in an aqueous medium. In some cases, carbon sources are also added to obtain, at the same time, a coating carbon layer [54,55,56] (Figure 5).

During co-precipitation syntheses, the particle growth is controlled only by kinetic factors, such as pH, temperature, ionic strength, salts’ nature, or Fe(II)/Fe(III) ratio. The sol-gel technique takes advantage of good stoichiometric control and homogeneity, short preparation time, and inexpensive precursors. Generally, to produce multicomponent oxides, alkoxides (or sometimes acetates or nitrates) are mixed in alcohol at a proper pH: Attention should be paid to salts’ concentration and temperature for adequate hydrolysis [57,58]. The obtained peculiar morphology can be commonly determined with microscopies such as SEM and/or TEM. They are mainly applied to evidence the particle shape (round, needle, hollow sphere, etc.) or size (from micron to nanometers) and the eventual porosity or roughness of samples [59,60,61]. SAED images obtained by HR-TEM provide the interplanar distances and a possible phase attribution. In addition, in cases of carbon coating, the identification of the thickness of the amorphous layers is also common. The EDS microanalysis, coupled with microscopies, provides insights into the elemental composition: the visual detection of ions’ distribution can also be obtained in the form of colored maps [58]. The particle sizes (nanoparticles) and surface area, two obviously related parameters, are at the base of Li⁺ diffusion into the ferrite grains.

4.2. Solid-State Syntheses

The conventional solid-state reaction is the oldest method for the synthesis of ceramics, since it is economic, efficient, and easily scalable. In fact, despite some disadvantages, such as particle agglomeration and growth, it is still regularly used to synthesize novel materials for the first time [62,63,64,65]. The introduction of mechanical milling during the preparation could guarantee the formation of the desired compounds without additional thermal treatments or at least at lower temperatures [66]. Some drawbacks are, however, represented by the highly energetic milling for an extended time and a possible contamination of the milled sample by the materials of balls and jars. The tuning of nanoparticles could be obtained by changing the milling parameters: milling container, speed, time, ball-to-powder weight ratio, extent of filling the vial, milling atmosphere, temperature of milling, etc. An interesting example of analysis of the impact of the milling process on the structural and morphological features of ferrites and the clarification of the contamination source during milling (balls or vial) can be found in reference [64]. The microwave-assisted combustion route is equally widespread, thanks to its numerous advantages: it is a green chemical technique, which enables a fast reaction rate, chemical homogeneity, and high reactivity. The microwaves interact with the reactants at the molecular level, leading to a uniform heating, producing ZnFe₂O₄ nanoparticles within few minutes. Drawbacks are represented by the difficulty in varying the spherical form of particles and the control of the valence states of the elements. However, ferrites can be produced at low temperature (under 500 °C) in nanometric sizes [67]. Pure samples can be stabilized with a thermal treatment at 300 °C for a few hours, offering the advantage of lowering the costs of the entire process [50]. By increasing the temperature up to about 1400 °C, the sinterization level increases and pure samples with large particles can be obtained [65]. In a typical synthesis, metal salts (typically nitrates) are mixed with an amount of fuel (citric acid, urea, glycine) in a ratio calculated by the propellant theory. The mixture is then placed in a microwave oven where a proper power (500–800 W) can induce the combustion process producing the ferrite nanoparticles. This method has been applied to synthesize both undoped and doped zinc ferrites (e.g., Sr doped) with excellent results [68].

4.3. Electrospinning

The electrospinning technique is an excellent candidate for fabricating nanomaterials possessing outstanding features such as low cost, flexibility, and simplicity [69,70]. The obtained nanostructured materials offer high surface area, which not only could shorten Li⁺ insertion/extraction pathways but also restrain the volume variations and particle aggregation during cycling, as well as providing good electrical conduction. Peculiar drawbacks include poor reproducibility and inhomogeneity of the mats. The electrospinning technique involves uniaxial stretching of a viscoelastic solution by the application of a strong electric field resulting in non-woven nanofiber layers or mats. Different fibrous arrays with tuneable length and diameter at nano- or meso-levels, high surface area, and controllable textures can be produced by simply varying the precursor viscosity, flow rate, applied electric field, and distance between the syringe and collector. Many examples of the application of this technique to produce 1D ZnFe₂O₄ can be found in the literature [69,70,71,72]. The surface area and the prevalence of macro, meso, or micro porosity, fundamental to hypothesize the electrolyte permeation and the cell functioning, can be determined by using N₂ adsorption measurements, on the base of BET theory [34,73].

In addition, the characterization of carbon, the main component of the resulting fibers, can be performed by Raman spectroscopy [74,75,76,77]. In fact, two characteristic signals in the Raman spectrum at about 1350 and 1580 cm⁻¹ are ascribable to D and G bands of carbon. The D band (A_1g mode) results from a break in the symmetry at the edges or defects in graphene sheets, while the G band, assigned to the E_2g mode, is related to the C–C stretching within the hexagonal lattice. The intensity ratio of the D and G peaks is sensitive to the type and order of the carbon material forming composites with ZFO [75].

In conclusion, we can compare the cited synthesis methods with special attention to the costs and the environmental impact of the entire processes. Equally inexpensive can be considered the wet chemistry and solid-state methods, starting from cheap and easily disposable reagents. In addition, the solid-state synthesis combined with the ball milling can produce pure spinels at low temperature, further reducing the costs of the synthesis process. In general, no toxic wastes are produced and, particularly the solid-state synthesis, can be easily scalable at an industrial level. Therefore, with these premises, they can be fully defined as green methods and, as we will see later, they represent the preferred choice for the ZFO production.

5. Electrochemistry of ZnFe₂O₄

The anode materials for LIBs are conventionally divided into three groups, depending on their reaction mechanism [78,79,80,81]:

(1)

intercalation/de-intercalation (such as carbon-based materials, TiO₂, Li₄Ti₅O₁₂, etc.);

(2)

alloying/de-alloying (such as Si, Zn, Ge, Sn, Al, Bi, Sb, etc.);

(3)

conversion (such as transition metal oxides, sulphides, phosphides, and nitrides).

The materials based on the conversion reaction mechanism are particularly intriguing for their high theoretical capacity and improved discharge potential as compared to graphite and for the ability to avoid lithium dendrites’ formation under fast discharge rates.

ZnFe₂O₄, unlike other oxides, has a lithium insertion mechanism that involves both conversion and alloying reactions. After the conversion reaction of ZFO with lithium ions and the formation of metallic Zn, Fe, and Li₂O, the resulting Zn can further react with lithium to form a Li_xZn alloy, thus contributing additional capacity.

The Li storage mechanism of ZFO is proposed as follows: during discharge

ZnFe₂O₄ + 0.5Li⁺ + 0.5e⁻ → Li_0.5ZnFe₂O₄

(1)

Li_0.5ZnFe₂O₄ + 1.5Li⁺ + 1.5e⁻ → Li₂ZnFe₂O₄

(2)

Li₂ZnFe₂O₄ + 6Li⁺ + 6e⁻ → 4Li₂O + 2Fe + Zn

(3)

Zn + Li⁺ + e⁻ → LiZn (alloy)

(4)

and in the following recharge process, the ferrite is not recovered and the reaction proceeds as follows:

LiZn (alloy) → Li⁺ + Zn + e⁻

(5)

Li₂O + Fe → 6Li⁺ + Fe₂O₃ + 6e⁻

(6)

Li₂O + Zn → ZnO + 2Li⁺ + 2e⁻

(7)

Based on the aforementioned mechanisms, the discharge in the first cycle consists of the destruction of the spinel crystal structure, which yields a mixture of Li₂O and metallic phases. The overall capacity obtained in the first discharge is provided by nine Li⁺ ions’ insertion into the ZFO molecule based on reactions (1)–(4). The formation of Li₂O is partially irreversible and contributes to the irreversible capacity loss in the first cycle, together with the formation of a SEI layer. In the first recharge (5)–(7), only the simple oxides Fe₂O₃ and ZnO can be recovered. In the following cycles, the electrochemically active Li–Zn alloys conveys reversible electrochemical reactivity towards Li.

Unfortunately, as is clear from the reported mechanism, ZFO experiences first-cycle irreversibility and fast decay in capacity with cycling, mainly resulting from poor electrical conductivity and large volumetric changes associated with the conversion reaction.

However, thanks to the downsizing of the particles, the addition of proper carbon sources, and to peculiar morphologies, the long-term cycling and the capacity values of ZnFe₂O₄ are, nowadays, very appealing.

Some of the most encouraging electrochemical results obtained in these last years on the ZFO active material are reported in Table 1. The ink composition, carbon type, and the electrolyte mixture are reported together with the discharge capacity values and the long-cycling performances.

5.1. Components of the Ink

5.1.1. Electrolytes

From the inspection of Table 1, it is obvious that the large part of electrolytes is based on carbonate solvents. The electrolyte acts as a medium to favour the movement of Li⁺ ions between cathode and anode; so, the interface between the electrolyte itself and the electrodes is crucial for the cycling performance of an electrochemical device. The solvents of choice should have high solubility and ionizability towards lithium salts, high stability, high flash point, a wide electrochemical window, and environmental friendliness. A carbonate solvent alone could not possess all the required features; so, mixtures of carbonates are the preferred ones, where EC, DEC, and DMC, usually in equal volumes (or weight amount), are the most used [31,34,47]. The available Li salts for the use in battery applications are numerous: lithium perchlorate (LiClO₄), lithium hexafluoroarsenate (LiAsF₆), lithium hexafluorophosphate (LiPF₆), lithium tetra-fluoroborate (LiBF₄), and lithium bis(tri-fluoromethanesulfonyl)imide (LiTFSI). For the application with ZnFe₂O₄ anodes, the preferred salt is LiPF₆ (see Table 1). LiPF₆, without any single outstanding property, is often commercialized due to the well-balanced combination of features, such as good ionic conductivity and electrochemical stability. In addition, commercial products based on LiPF₆ in carbonates’ mixtures are easily available. However, the classical non-aqueous electrolytes have some important drawbacks such as the oxidative decomposition under high voltage (>4.3 V vs. Li/Li⁺), resulting in the failure of batteries, and the tendency to high flammability causing a major safety issue.

5.1.2. Ink Composition

Even if the electrolytes for ZFO are very similar to one another, the same is not true for the anodic ink composition (Table 1). The ratio between the active material, carbon source, and binder (Ac:Carbon:B) varies between 60:30:10 to 80:10:10, with this last one being the preferred choice [51,62,82,83].

5.1.3. Binders

The electrodes are usually prepared as a slurry, pasted to the current collector thanks to a binder. The most diffuse binder for LIBs is PVdF, [25,36,84,85,86]: however, for its processing, volatile organic compounds such as N-methyl-pyrrolidone (NMP) are required. These solvents are not only expensive, but also environmentally harmful and toxic, thus introducing safety concerns into the manufacturing process. In addition, PVdF is a relatively costly polymer (industrial cost in the multiton scale is around 15–18 EUR kg⁻¹) and not easily disposable at the end of the battery cycle.

More environmentally friendly, cheaper, and green binders are emerging for both LIBs and Sodium Ion Batteries (NIBs). The aim of these binders is to avoid the use of polluting volatile organic compounds in the preparation of slurries. In particular, CarboxyMethylCellulose (CMC) is widely employed for the manufacturing of ZnFe₂O₄ anodes (Table 1) [66,72,75,86,87]. CMC is produced by the insertion of carboxymethyl groups in the natural cellulose, making CMC water soluble. This is certainly the greatest advantage of CMC since it allows the processing of the slurries in aqueous medium. In addition, the CMC industrial price is about 1–2 EUR kg⁻¹, i.e., about one order of magnitude cheaper than PVdF. Another widely employed natural and green binder is Sodium Alginate (SA) that contains carboxylic groups on monomeric units, enabling a great number of hydrogen bonds between itself and the electrode materials [30,41,62]. Its poor swelling ability makes the electrodes less susceptible to a massive electrolyte penetration and, hence, to degradation and passivation processes. It was first studied as a binder for anodic Si-based materials, but nowadays is also used for other electrode anode and cathode materials for LIBs. As can be seen from Table 1, the anodes containing green binders have electrochemical performances even better than those with the classical PVdF.

5.1.4. Carbon

The other fundamental actor for the anode slurry preparation is the conductive carbon, which should provide high electronic conduction, particularly useful for active materials, such as ZFO, with low electronic conductivity. The particle sizes and morphologies of the carbons should be carefully evaluated to better understand the obtained capacities, particularly at low potential. For LIBs, the most diffused carbons are acetylene black, Super P carbon black, Ketjen black, graphene, and Vulcan, with the first two as the most common ones for the preparation of ZnFe₂O₄ anodic ink (see Table 1) [33,47,69,70]. Acetylene black (AB) is a well-known conductive additive for batteries, due to its excellent conductivity, compressibility, liquid-absorbing tendency, and elasticity. Additionally, it is very cheap compared to other carbonaceous materials (such as carbon nanotubes and graphene) and bears a continuous porous structure that can restrain volume variations, stabilizing the electrode during long-term cycling. Its optimum amount is between 10% and 30% by weight (Table 1) [56,61,87,88,89].

Super P carbon black (SP) can be a valid alternative to acetylene black: It is produced from partial oxidation of petrochemical precursors, and it exhibits a large specific surface area and superb electrical conductivity. Lately, it seems to be the carbon source of choice for ZFO anodes [52,74,82,90,91].

5.2. Peculiar Morphologies

The improvement of the electrochemical performances of ferrites starts from the search for particular nanoparticles’ morphologies, as well as the formation of hybrid nanocomposites with peculiar carbon sources. Many different and rummy morphologies, producing interesting electrochemical performances, have been proposed [92,93,94,95,96,97,98,99]. The 3D, cuboid-structured ZnFe₂O₄@C nano-whiskers [85], highly mesoporous carbon nanofibers in free-standing mat form [92], carbon-decorated ZnFe₂O₄ nanowires [86], and 1D ZnFe₂O₄ hollow fibers with integrated tubular mesoporous nanostructures [70] are hybrid materials that combine the advantages of nanostructures with the carbon peculiarities.

Examples of peculiar ZFO morphologies with the corresponding cell performances are reported in Figure 6.

Porous nanorod-like nanostructures (many tiny nanoparticles with smooth surfaces) of the as-prepared ZFO materials guarantee 1458 mAh g⁻¹ after 120 cycles and an excellent rate capability (778 mAh g⁻¹ at 3 Ag⁻¹). More importantly, a discharge capacity of 456 mAh g⁻¹ was maintained at a current density of about 5 C after 200 cycles [95].

ZnFe₂O₄ nanoneedles, uniformly dispersed on the surface of the carbonized polydopamine porous nanofiber, can align perpendicularly to the nanofibers’ surface, whose mesoporous nature may restrict the size of the nucleation sites. The nanoneedle-like morphology can favour the contact between the nanoneedles themselves and the conductive matrix, with the easy diffusion of electrolyte and lithium ions in the nanochannels leading to excellent cyclability and rate capacity. This peculiar structure allowed obtaining normalized capacities between 1000 and 1700 mAh g⁻¹ at 0.1 C and 560 mAh g⁻¹ at 5 C, respectively [92].

A particularly promising morphology is represented by the thin triple-shelled ZFO hollow microspheres with a complex structure formed by the outer shell, void, inner shell, and inner hollow core along the radial direction from the surface (Figure 6). The thin shell offers a highly specific surface area with a lot of electrochemical active sites and short diffusion lengths. The wide voids between the shells lead to an easy mass transfer by storing abundant electrolyte and protecting against the damage of microspheres by providing large spaces, hindering a severe volume change during lithiation/delithiation. As a result, this peculiar architecture offers a high reversible capacity of 932 mAh g⁻¹ at 2 C even at the 200th cycle without obvious decay. Furthermore, it delivers 1235, 1005, 865, 834, and 845 mAh g⁻¹ at current densities of 0.5, 2, 5, 10, and 20 C after 100 cycles (Figure 6) [52].

The use of metal-organic framework MOFs as electrode materials for LIBs is growing, due to their inimitable structure, tunable morphology, well-developed porosity, high specific surface area, and excellent lithium storage capacity. The open crystal frameworks of MOFs can deliver efficient and reversible ionic insertion/extraction [100]. Prussian Blue (a kind of metal-organic framework) can react with Zn acetate under microwave irradiation, forming hydroxides’ and carbonates’ nanosheets on the surface of Prussian Blue itself. After pyrolysis, a composite of ZnO/ZnFe₂O₄ porous nanoparticles was obtained [101], showing a good capacity of 804 mAh g⁻¹ for 500 cycles at a current density of 1 C.

The obvious advantages shown by the nanocomposites with various architectures are, thus, the high capacities at high C rates and prolonged cycle life.

5.3. Doping

As discussed, some critical challenges associated with ZFO anodes in LIBs are represented by the large irreversible capacity loss at the initial discharge–charge cycles and poor cycling performance due to the low electronic conductivity and large volume variation. The doping is well recognized as an easy way to improve the electronic conductivity and/or the crystal structure stability of ferrites [37,38,39,40]. The dopant elements that can satisfy both these requests are obviously the transition metals ions. ZFO nanofibers doped with Mn ions [37] exhibited a good capacity of ∼612 mAh g⁻¹ at the 50th cycle (at 60 mAh g⁻¹) for the Zn_0.3Mn_0.7Fe₂O₄ composition, the best result among the solid-solution series, through the beneficial conversion reaction and alloy−dealloy mechanism (Figure 7). The use of ex situ characterizations such as high-resolution transmission electron microscopy (HRTEM) and electron energy loss spectroscopy (EELS) allowed us to study the involvement of Mn in the battery performance. This paper offers interesting insights on how to flexibly adjust the electrochemical performance by tuning the transition metals’ ratio within the spinel [37].

An appropriate Mn doping and CNTs’ intertwining actively affect the formation of uniform morphology and improve the cycling stability and rate capability [39]. Zn_0.5Mn_0.5Fe₂O₄@CNT composites exhibit excellent electrochemical performances, with enhanced reversible capacity (1375 mAh g⁻¹ after 100 cycles at the current density of 100 mA g⁻¹) and good rate capability (933.5 mAh g⁻¹ at 500 mA g⁻¹, 810 mAh g⁻¹ at 1000 mA g^−1, and 634.2 mAh g⁻¹ at 1500 mA g⁻¹).

The co-doping is another successful strategy to improve the lithium storage properties of ZFO. ZnFe₂O₄ and Mg_xCu_0.2Zn_0.82−xFe_1.98O₄ (x = 0.20, 0.25, 0.30, 0.35, and 0.40) nanoparticles were synthesized by the sol−gel-assisted combustion method. The as-synthesized samples showed large capacity fading, but the addition of a thermal treatment at 800 °C allowed us to obtain better cycling stability [40]. Undoped ZFO exhibited a high reversible capacity of 575 mAh g⁻¹ after 60 cycles at a current density of 100 mA g⁻¹. Mg_0.2Cu_0.2Zn_0.62Fe_1.98O₄ showed a similar capacity after 60 cycles but better capacity retention. Therefore, though the Mg, Cu co-doped compounds exhibited lower theoretical capacity, the co-doping improved the reversible capacity and capacity retention (Figure 7), especially at higher current rates. The nickel ion, with high redox potentials, as a dopant has great potential; in fact, NiFe₂O₄ is deemed as a promising anode material for LIBs. Ni_xZn_1-xFe₂O₄ (x = 0, 0.25, 0.5, 0.75, 1) compounds [38] were prepared by hydrothermal method and subsequent heat treatment. Ni_0.25Zn_0.75Fe₂O₄ showed good electrochemical performance with a discharge capacity of 1488 mAh g⁻¹ in the first cycle and 856 mAh g⁻¹ after 100 cycles. Compared with ZFO, nickel-doped samples, except NiFe₂O₄, increased to some extent the specific discharge capacity. This improvement was attributed to good crystallinity, low resistance, and small particle sizes that could shorten the diffusion length of the lithium ion, accommodate volume expansion, and lead to better electrochemical performance.

5.4. Composites

The strategy to synthesize hybrid nanomaterials with conductive carbonaceous sources (e.g., porous carbon, carbon nanotubes, nanofibers, graphene/graphite, etc.) is particularly promising to overcome the main limitation of ferrites for electrochemical application, i.e., the poor electronic conductivity. Carbonaceous materials possess good electronic conductivity and high-rate capability and can preserve the structural integrity of active materials, thereby giving enhanced cyclability and rate capacity of the resulting hybrid composites. A variety of modifications of ZFO with carbonaceous materials have been exploited: surface carbon coating [31], graphene as substrate with active nanomaterials decorating [102,103], growing, or anchoring on surfaces [25,55], embedding in carbon matrix [34] or graphene nanosheets-wrapped active material particles [32] until the use of sacrificial templates [104,105,106,107,108,109,110] (Figure 8).

Ultrafine nanoparticles with average diameters of about 5 nm homogeneously distributed within a 3D, interconnected, porous carbon framework in the walls of the hollow octahedra work very well (Figure 6). This elaborate nanoarchitecture combines the superior properties of a hollow interior structure, ultrafine building blocks, a conductive elastic buffering framework, and high porosity. High reversible capacities of 1060 mAh g⁻¹ at 0.5 C, 934 mAh g⁻¹ at 1 C, 887 mAh g⁻¹ at 2 C, and 842 mAh g⁻¹ at 5 C could be achieved, demonstrating a magnificent, high-rate performance of the hybrid electrode. Moreover, a capacity of more than 1200 mAh g⁻¹ was recovered at 0.5 C [74].

Mesoporous ZnFe₂O₄ spheres wrapped in graphene sheets provide good performances [104]. These spheres are composed of numerous nanoparticles where the void spaces between the neighboring nanoparticles provide a cushion to alleviate volume variation during the conversion/alloying process as well as large contact areas with the electrolyte to facilitate lithium-ion diffusion and electron transport during cycling. A high specific capacity of 1182 mAh g⁻¹ at 0.1 C and significantly enhanced rate capability and cycling stability after long-term testing were obtained [104]. Cai et al. [31] prepared well-dispersed ZFO nanoparticles anchored on an acetylene black substrate, further covered and interlinked by amorphous carbon layers, resulting in self-assembly into large hierarchical porous granules. The presence of conductive carbon enables better electron transfer and prevents the aggregation of the ZFO nanoparticles, buffering the large volume change of the active material. The reversible discharge capacity of the AB/ZnFe₂O₄-NPs@C decreased from 731 to 229 mAh g⁻¹ as the current density increased from 0.1 to 2.0 C; but it is worth noting that a reversible discharge capacity of 375 mAh g⁻¹ after cycling at 1 C was provided, still higher than the theoretical capacity of graphite. Importantly, when the current density returned to 0.1 A g⁻¹, a discharge capacity of 470 mAh g⁻¹ was recovered, further increasing to 830 mAh g⁻¹ after another 100 cycles [31]. ZnFe₂O₄ nanorods [36], coated with carbon (by self-polymerization of dopamine) on the surfaces followed by carbonization, have better rate capability and cycling stability. When discharged/charged at 0.5 and 2 Ag⁻¹, the ZFO/carbon nanorod anode can deliver reversible capacities of 805 and 504 mAh g⁻¹, respectively. Besides, it shows excellent cycling stability and no obvious capacity fading at the current density of 1Ag⁻¹ for 100 cycles [36].

The obtained unique morphologies can provide materials with high reversible capacity, enhanced rate performance, and superior cycling stability due to the synergetic effect of carbon matrix and carbon coating (see Table 1). The carbon network offers a continuous conductive pathway for electron transport and high surface area, improves the mechanical flexibility of active materials, and, more importantly, can help to maintain the structural integrity of ZnFe₂O₄ during repeated lithiation/delithiation [111,112,113,114,115,116].

Figure 8. (A) Illustration of the synthetic process of ternary metal oxides’ hollow structures via MOF self-sacrificial template [63]; (B) ZFO/GAs composites obtained by two-step method [103], reproduced with permission, Copyright Elsevier 2017; (C) preparation procedure of conventional ZnFe₂O₄/graphene (ZFO–G) and ZnFe₂O₄/graphene by using glucose (ZFO–G–GL) [116]; (D) schematic illustration of Pluronic copolymer-induced self-assembly of TMO nanoparticles into holey 2D TMO nanosheets and STEM images [105].

Figure 8. (A) Illustration of the synthetic process of ternary metal oxides’ hollow structures via MOF self-sacrificial template [63]; (B) ZFO/GAs composites obtained by two-step method [103], reproduced with permission, Copyright Elsevier 2017; (C) preparation procedure of conventional ZnFe₂O₄/graphene (ZFO–G) and ZnFe₂O₄/graphene by using glucose (ZFO–G–GL) [116]; (D) schematic illustration of Pluronic copolymer-induced self-assembly of TMO nanoparticles into holey 2D TMO nanosheets and STEM images [105].

Table 1. Electrochemical performances of ZnFe₂O₄ obtained from different synthesis routes. The data are listed according to groups of methods. When, in the cited papers, different doping levels were used, we chose to report the better performances. 1C was considered 1000 mA/g.

Material	Synthesis	Anode Composition/Electrolyte	First Discharge Capacity (mAh g−1 at mAg−1)	Capacity at the Highest C Rate (mAh g−1/C Rate)	Long Cycling (mAh g−1/C Rate/N Cycles)	Reference
1D ZFO-based hybrid NWs	Electrospinning	Ac:AB:CMC 70:20:10 1M LiPF6 in EC:DMC:EMC (1:1:1 v/v/v)	1108/0.05C	287/10C	881/1C/220	[54]
Composites nanofibers	Electrospinning-annealing	Ac:CB:CMC 80:10:10 1 M LiPF6 in EC:DMC:DEC (1:1:1 w/w/w)	1242/0.1C	450/1C	793/0.1C/200	[72]
ZFO hollow tubular mesoporous nanostructures	Single-spinneret electrospinning post-treatment	Ac:SP:(PVdF+DMF) 70:20:10 1 M LiPF6 in EC:DEC (1:1 w/w)	1306/0.2C	528/1.6C	882/0.2C/200	[70]
ZnO/ZFO sub-microcubes’ composites	Bottom-up self-sacrifice template	Ac:SP:CMC 70:20:10 1 M LiPF 6 in EC:DMC:DEC (1:1:1 v/v/v)	1892/1C	667/2C	837/1C/200	[47]
2D mesoporous ZFO/ZnO nanosheets	MOF strategy	Ac:AB:CMC 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 v/v/v)	1156/0.5C	266/1.5C	537/0.5C/500	[106]
Porous ZnO/ZFO/C octahedra	MOF as precursor and self-sacrificing template	Ac:SP:PAA 70:15:15 1M LiPF6 in EC:EMC:DMC (1:1:1 w/w/w)	1385/0.5C	762/10C	988/2C/100	[74]
ZnO/ZFO porous nanoparticles’ composites	Prussian Blue-Zn acetate with microwave irradiation	Ac:CB:CMC 70:20:10 1M LiPF6 in DMC:EC (1:1 v/v)	1293/0.1C	300/2C	497/2C/1000	[101]
Porous ZnO/ZFO nanostructures	Prussian blue analogue as sacrificial template	Ac:CB:PVdF 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 w/w/w)	998.4/0.2C	---	704/0.2C/200	[107]
3D ball-in-ball ZnO/ZFO@C nanospheres	Carbonization of ZnFe-MOFs	Ac:AB:PVdF 80:10:10 1M LiPF6 in EC:EMC:DMC (1:1:1 v/v/v)	1686/0.1C	155/20C	1085/0.2C/200	[94]
Polydopamine film on surfaces of ZFO	Biomimetic method	Ac:SP:SA 40:40:20 1M LiPF6 in EC:DMC (1:1 v/v)	2079/1C	1215.3/5C	2074/1C/150	[29]
ZFO/C-crafted nano-composites	In situ pyrolysis	Ac:AB:PVdF 80:10:10 1 M LiPF6 in EC:DEC: EMC (1:1:1 v/v/v)	1626/0.1C	521/5C	1100/0.2C/430	[33]
ZFO/graphene composites	Co-precipitation/solid-state reaction	Ac:SP:PVdF 60:30:10 1 M LiPF6 in EC:DMC (1:1 v/v)	1744/0.1	90/20C	331/5C/200	[61]
Carbon nanospheres embedded in ZFO	In situ polymerization-carbonization	Ac:SP:PvdF 80:10:10 1 M LiPF6 in EC:DMC (1:1 v/v)	1356/0.1C	504/20C	392/5C/200	[34]
Shuttle-shaped mesoporous ZFO microrods	Solvothermal post-annealing	Ac:SP:CMC 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 v/v/v)	1520/0.1C	326/1.5C	542/1C/488	[113]
Porous N-doped carbon-coated ZFO	Combustion carbon coating	Ac:SP:SA 40:40:20 1 M LiPF6 in EC:DMC (1:1 v/v)	1477/0.1C	---	705/1C/1000	[41]
Hierarchical porous AB/ZFO@carbon hybrid	Self-assembly-calcination	Ac:AB:PVdF 80:10:10 1 M LiPF6 in EC:DMC (1:1 v/v)	1099/0.1C	229/2C	430/1C/200	[31]
Macroporous MFO	Solution combustion synthesis	Ac:SP:PVdF 70:20:10 1M LiPF6 in EC:DMC (1:1 v/v)	1404.6/0.2C	800/1.5C	794.7/1C/300	[35]
Porous ZFO nanorods	Spray-drying process sintering	Ac:SP:PVdF 80:10:10 1M LiPF6 in EC:DEC (1:1 v/v)	1800/0.1C	778/3C	456/5C/250	[95]
ZFO/C	Hydrothermal carbon coating	Ac:SP:CMC 80:10:10 1M LiPF6 in EC:DEC (3:7 w/w)	1327/0.1C	216/20C	1091/0.1C/190	[5]
Ni-doped ZFO	Hydrothermal heat treatment	Ac:AB:CMC 70:20:10 1M LiPF6 in EC:DEC:EMC (1:1:1 v/v/v)	1488/0.1C	481/0.77C	856/0.1C/100	[38]
ZFO flake/graphite composites	Hydrothermal sintering	Ac:SP:LA132 60:15:25 1M LiPF6 in EC:DEC:EMC (1:1:1 v/v/v)	744/0.1C	---	730/0.1C/100	[102]
Self-assembled 3D mesoporous ZFO spheres in graphene sheets	One-pot hydrothermal	Ac:AB:PVdF 70:20:10 1M LiPF6 in EC:DEC (1:1 v/v)	1978/0.1C	650/2C	770/1C/500	[104]
N/S graphene-supported hollow ZFO nanosphere composites	Two-step hydrothermal method	Ac:AB:PVdF 65:15:20 1M LiPF6 in EC:DEC:DMC (1:1:1 v/v/v)	2478/0.3C	617/1.8C	729/0.3C/100	[43]
Triple-shelled ZFO hollow microspheres	One-pot hydrothermal +calcination	Ac:SP:PAA 70:20:10 1M LiPF6 in EC:DMC (1:1 v/v)	1412/2C	--	932/2C/200	[52]
Multifunctional ZFO yolk–shell hollow architecture	Hydrothermal-SiO2 template	Ac:SP:CMC 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 v/v/v)	1392/1.5C	775/2C	873/1C/500	[24]
Hollow octahedra ZFO@C nanocomposites	Solvothermal	Ac:SP:PVdF 70:20:10 1M LiPF6 in EC:DEC (1:1 v/v)	1752/0.2C	557/15C	1780/1C/400	[20]
ZFO@C hollow microspheres	CVD method from ZFO microspheres via solvothermal method	Ac:SP:SA 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 v/v/v)	1475/0.1C	810/1.6C	1000/0.5C/200	[30]
Hierarchical bead chain ZFO-PEDOT composite	Co-precipitation, thermal decomposition, polymerization	Ac:AB:PTFE 80:10:10 1M LiPF6 in EC:DEC:EMC (1:1:1 v/v/v)	1412/0.1C	827/2C	1005/1C/200	[83]
Mn-doped ZFO @CNT composites	Co-precipitation	Ac:SP:PVdF 80:10:10 1M LiPF6 in EC:DEC:DMC (1:1:1 v/v/v)	1666/0.1C	634/1.5C	1375/0.1C/100	[39]
ZFO/C composites	Co-precipitation	Ac:CB:PVdF 70:20:10 1M LiPF6 in EC:DEC:DC (1:1:1 v/v/v)	1558/2C	321/10C	644/2C/1000	[111]
ZFO nanoneedles on mesoporous carbon nanofibers	Solution route	Freestanding 1M LiPF6 in EC:DMC (1:1 v/v)	1581/0.05C	300/5C	700/1C/200	[92]
3D cuboid-structured ZFO@C nano-whiskers	In situ graft co-polymerization-calcination	Ac:SP:PVdF 80:10:10 1M LiPF6 in DEC:EC:EMC (1:1:1 v/v/v)	1667/0.1C	700/3.2C	1722/0.1C/120	[85]
ZFO nanoparticles embedded in carbon matrix	In situ one-step route	Ac:CB:SA 60:30:10 1M LiPF6 in EC:DMC (1:1 v/v)	1744/1C	---	882/0.5C/1000	[62]
ZFO yolk–shell powders	Spray-drying process	Ac:SP:CMC 70:20:10 1MLiPF6 in EC:DEC (1:5 v/v)+ 5% FEC	1226/0.5C		862/0.5C/200	[91]
ZFO/GN hybrid films	One-step electrophoretic deposition thermal annealing	Ac only 1M LiPF6 in EC:DMC (1:1 v/v)	1249/0.2C	510/3.2C	881/0.2C/200	[49]

Table 1. Electrochemical performances of ZnFe₂O₄ obtained from different synthesis routes. The data are listed according to groups of methods. When, in the cited papers, different doping levels were used, we chose to report the better performances. 1C was considered 1000 mA/g.

Material	Synthesis	Anode Composition/Electrolyte	First Discharge Capacity (mAh g−1 at mAg−1)	Capacity at the Highest C Rate (mAh g−1/C Rate)	Long Cycling (mAh g−1/C Rate/N Cycles)	Reference
1D ZFO-based hybrid NWs	Electrospinning	Ac:AB:CMC 70:20:10 1M LiPF6 in EC:DMC:EMC (1:1:1 v/v/v)	1108/0.05C	287/10C	881/1C/220	[54]
Composites nanofibers	Electrospinning-annealing	Ac:CB:CMC 80:10:10 1 M LiPF6 in EC:DMC:DEC (1:1:1 w/w/w)	1242/0.1C	450/1C	793/0.1C/200	[72]
ZFO hollow tubular mesoporous nanostructures	Single-spinneret electrospinning post-treatment	Ac:SP:(PVdF+DMF) 70:20:10 1 M LiPF6 in EC:DEC (1:1 w/w)	1306/0.2C	528/1.6C	882/0.2C/200	[70]
ZnO/ZFO sub-microcubes’ composites	Bottom-up self-sacrifice template	Ac:SP:CMC 70:20:10 1 M LiPF 6 in EC:DMC:DEC (1:1:1 v/v/v)	1892/1C	667/2C	837/1C/200	[47]
2D mesoporous ZFO/ZnO nanosheets	MOF strategy	Ac:AB:CMC 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 v/v/v)	1156/0.5C	266/1.5C	537/0.5C/500	[106]
Porous ZnO/ZFO/C octahedra	MOF as precursor and self-sacrificing template	Ac:SP:PAA 70:15:15 1M LiPF6 in EC:EMC:DMC (1:1:1 w/w/w)	1385/0.5C	762/10C	988/2C/100	[74]
ZnO/ZFO porous nanoparticles’ composites	Prussian Blue-Zn acetate with microwave irradiation	Ac:CB:CMC 70:20:10 1M LiPF6 in DMC:EC (1:1 v/v)	1293/0.1C	300/2C	497/2C/1000	[101]
Porous ZnO/ZFO nanostructures	Prussian blue analogue as sacrificial template	Ac:CB:PVdF 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 w/w/w)	998.4/0.2C	---	704/0.2C/200	[107]
3D ball-in-ball ZnO/ZFO@C nanospheres	Carbonization of ZnFe-MOFs	Ac:AB:PVdF 80:10:10 1M LiPF6 in EC:EMC:DMC (1:1:1 v/v/v)	1686/0.1C	155/20C	1085/0.2C/200	[94]
Polydopamine film on surfaces of ZFO	Biomimetic method	Ac:SP:SA 40:40:20 1M LiPF6 in EC:DMC (1:1 v/v)	2079/1C	1215.3/5C	2074/1C/150	[29]
ZFO/C-crafted nano-composites	In situ pyrolysis	Ac:AB:PVdF 80:10:10 1 M LiPF6 in EC:DEC: EMC (1:1:1 v/v/v)	1626/0.1C	521/5C	1100/0.2C/430	[33]
ZFO/graphene composites	Co-precipitation/solid-state reaction	Ac:SP:PVdF 60:30:10 1 M LiPF6 in EC:DMC (1:1 v/v)	1744/0.1	90/20C	331/5C/200	[61]
Carbon nanospheres embedded in ZFO	In situ polymerization-carbonization	Ac:SP:PvdF 80:10:10 1 M LiPF6 in EC:DMC (1:1 v/v)	1356/0.1C	504/20C	392/5C/200	[34]
Shuttle-shaped mesoporous ZFO microrods	Solvothermal post-annealing	Ac:SP:CMC 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 v/v/v)	1520/0.1C	326/1.5C	542/1C/488	[113]
Porous N-doped carbon-coated ZFO	Combustion carbon coating	Ac:SP:SA 40:40:20 1 M LiPF6 in EC:DMC (1:1 v/v)	1477/0.1C	---	705/1C/1000	[41]
Hierarchical porous AB/ZFO@carbon hybrid	Self-assembly-calcination	Ac:AB:PVdF 80:10:10 1 M LiPF6 in EC:DMC (1:1 v/v)	1099/0.1C	229/2C	430/1C/200	[31]
Macroporous MFO	Solution combustion synthesis	Ac:SP:PVdF 70:20:10 1M LiPF6 in EC:DMC (1:1 v/v)	1404.6/0.2C	800/1.5C	794.7/1C/300	[35]
Porous ZFO nanorods	Spray-drying process sintering	Ac:SP:PVdF 80:10:10 1M LiPF6 in EC:DEC (1:1 v/v)	1800/0.1C	778/3C	456/5C/250	[95]
ZFO/C	Hydrothermal carbon coating	Ac:SP:CMC 80:10:10 1M LiPF6 in EC:DEC (3:7 w/w)	1327/0.1C	216/20C	1091/0.1C/190	[5]
Ni-doped ZFO	Hydrothermal heat treatment	Ac:AB:CMC 70:20:10 1M LiPF6 in EC:DEC:EMC (1:1:1 v/v/v)	1488/0.1C	481/0.77C	856/0.1C/100	[38]
ZFO flake/graphite composites	Hydrothermal sintering	Ac:SP:LA132 60:15:25 1M LiPF6 in EC:DEC:EMC (1:1:1 v/v/v)	744/0.1C	---	730/0.1C/100	[102]
Self-assembled 3D mesoporous ZFO spheres in graphene sheets	One-pot hydrothermal	Ac:AB:PVdF 70:20:10 1M LiPF6 in EC:DEC (1:1 v/v)	1978/0.1C	650/2C	770/1C/500	[104]
N/S graphene-supported hollow ZFO nanosphere composites	Two-step hydrothermal method	Ac:AB:PVdF 65:15:20 1M LiPF6 in EC:DEC:DMC (1:1:1 v/v/v)	2478/0.3C	617/1.8C	729/0.3C/100	[43]
Triple-shelled ZFO hollow microspheres	One-pot hydrothermal +calcination	Ac:SP:PAA 70:20:10 1M LiPF6 in EC:DMC (1:1 v/v)	1412/2C	--	932/2C/200	[52]
Multifunctional ZFO yolk–shell hollow architecture	Hydrothermal-SiO2 template	Ac:SP:CMC 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 v/v/v)	1392/1.5C	775/2C	873/1C/500	[24]
Hollow octahedra ZFO@C nanocomposites	Solvothermal	Ac:SP:PVdF 70:20:10 1M LiPF6 in EC:DEC (1:1 v/v)	1752/0.2C	557/15C	1780/1C/400	[20]
ZFO@C hollow microspheres	CVD method from ZFO microspheres via solvothermal method	Ac:SP:SA 70:20:10 1M LiPF6 in EC:DMC:DEC (1:1:1 v/v/v)	1475/0.1C	810/1.6C	1000/0.5C/200	[30]
Hierarchical bead chain ZFO-PEDOT composite	Co-precipitation, thermal decomposition, polymerization	Ac:AB:PTFE 80:10:10 1M LiPF6 in EC:DEC:EMC (1:1:1 v/v/v)	1412/0.1C	827/2C	1005/1C/200	[83]
Mn-doped ZFO @CNT composites	Co-precipitation	Ac:SP:PVdF 80:10:10 1M LiPF6 in EC:DEC:DMC (1:1:1 v/v/v)	1666/0.1C	634/1.5C	1375/0.1C/100	[39]
ZFO/C composites	Co-precipitation	Ac:CB:PVdF 70:20:10 1M LiPF6 in EC:DEC:DC (1:1:1 v/v/v)	1558/2C	321/10C	644/2C/1000	[111]
ZFO nanoneedles on mesoporous carbon nanofibers	Solution route	Freestanding 1M LiPF6 in EC:DMC (1:1 v/v)	1581/0.05C	300/5C	700/1C/200	[92]
3D cuboid-structured ZFO@C nano-whiskers	In situ graft co-polymerization-calcination	Ac:SP:PVdF 80:10:10 1M LiPF6 in DEC:EC:EMC (1:1:1 v/v/v)	1667/0.1C	700/3.2C	1722/0.1C/120	[85]
ZFO nanoparticles embedded in carbon matrix	In situ one-step route	Ac:CB:SA 60:30:10 1M LiPF6 in EC:DMC (1:1 v/v)	1744/1C	---	882/0.5C/1000	[62]
ZFO yolk–shell powders	Spray-drying process	Ac:SP:CMC 70:20:10 1MLiPF6 in EC:DEC (1:5 v/v)+ 5% FEC	1226/0.5C		862/0.5C/200	[91]
ZFO/GN hybrid films	One-step electrophoretic deposition thermal annealing	Ac only 1M LiPF6 in EC:DMC (1:1 v/v)	1249/0.2C	510/3.2C	881/0.2C/200	[49]

5.5. Studies on SEI and Mechanism of Conversion/Alloying

The processes involved in lithium intercalation/deintercalation of ZFO are complex and their understanding requires different but combined experimental techniques [117,118,119,120,121,122,123,124]. A crucial part for the cell functioning is related to the solid electrolyte interface (SEI), formed during the first lithium uptake, usually considered as a passivation layer preventing the degradation of the electrodes while not participating in the lithium storage process. The investigation of SEI was fundamental for lithium metal batteries, allowing us to understand that the SEI film is the major reason for the failure of the lithium metal anode because it is unstable against the corrosive electrolyte owing to the formation of unstable SEI; but without SEI, the battery operation would not be feasible [125].

The formation of the SEI layer takes place during the first steps of the charging process, and its thickness reaches about 40 nm at about 1/3 of the full capacity, with a stable total thickness up to 20 working cycles, as demonstrated by XAS studies [117,118]. The structural analysis of the distinct components forming the SEI gave detailed information on the SEI evolution at different charge−discharge stages [119]. The experimental results showed that the electrode reduction process, resulting in the formation of Li₂O in proximity of the electrode active material and the formation of the passivation layer on the electrode core, was accompanied by the formation of a partly reversible layer as the uppermost SEI layer (typical thickness of about 5–7 nm). This mostly consisted of alkyl lithium carbonate (ROCO₂Li), being partly replaced by Li₂CO₃ during delithiation. This layer acts as a lithium reservoir and seems to give a significant contribution to the extra capacity of the electrodes. This result further extends the role of the SEI layer in the functionality of Li-ion batteries.

The use of sophisticated techniques, such as X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) [120], impedance spectroscopy [121], in situ TEM [122], in situ X-ray diffraction data, synchrotron-based powder diffraction, and ex situ extended X-ray absorption fine structure [123,124], allowed us to prove the distinctive feature of the ZFO anode. The lithiation of ZFO nanoparticles was demonstrated to be a multistep process involving the presence of intermediate Li-Zn-Fe-O phases. In fact, ZFO, during the first discharge, formed a mixture of metallic iron, metallic zinc, and Li₂O. Instead of the original ZFO spinel, the metallic iron and zinc particles were re-oxidized to Fe₂O₃ and ZnO during the subsequent de-lithiation (see Equations (1)–(7)). All the cited studies [120,121,122,123,124] agree with this mechanism (Figure 9).

5.6. Full Cell Based on ZFO Anode

The papers above dealt with semi-cell configuration, in which the ZFO anode performances were tested towards Li metal. Clearly, to hypothesize the future commercialization of batteries based on ZFO, it is mandatory to study full cells. Papers about this topic are limited: In these cases, the ZFO anode was coupled with LiFePO₄ cathode, a well-known material, commercially available. A carbon-coated, ZnFe₂O₄ nanoparticle-based spinel was interfaced in a complete cell with a LiFePO₄ multiwalled carbon nanotube-based cathode, both aqueously processed with Na-CMC, obtaining exceptional electrochemical performance [126] (Figure 10). Such a battery showed remarkable rate capabilities, delivering about 50% of its nominal capacity at currents corresponding to about 20 C. The pre-lithiation of the negative electrode, a diffuse strategy offering the possibility of tuning the cell potential, allowed obtaining values of 202 Wh kg⁻¹ and 3.72 Wkg⁻¹ of gravimetric energy and power density, respectively, in addition to granting a lithium reservoir. The high reversibility of the system enabled sustaining more than 10,000 cycles at about 10 C with respect to LiFePO₄ cathode, while retaining up to 85% of its initial capacity.

In another study [127], a ZnFe₂O₄/G nanohybrid was synthesized by a facile in situ route, showing better cycling stability and rate capability than bare ZFO. At a charge current density of 800 mA g^−1, the hybrid still delivered a first-charge capacity of 701 mAh g⁻¹, which was maintained at 464 mAh g⁻¹ after 300 cycles. The excellent high-rate cycling stability was attributed to the three combined effects of graphene: conducting, confining, and dispersing. In addition, the unique 2D sheet-like nanostructure maximized the exposure of the active material to the electrolyte and buffered the volume changes. In the ZnFe₂O₄/G–LiFePO₄/C full cell, ZnFe₂O₄/G can yield a high initial discharge capacity of 805 mAh g⁻¹ at a current density of 100 mA g⁻¹.

In general, the evaluation of the performances of a full cell was based on the analysis of specific gravimetric and/or volumetric capacities, while other key parameters were often neglected. Notwithstanding the tempting specific energy (Whkg⁻¹) of ZFO in full batteries, other parameters such as the Coulombic efficiency (CE), voltage efficiency (VE), and energy efficiency (EE) have to be considered for practical applications in large-sized batteries. In a detailed study [128], ZFO, with its large voltage hysteresis, exhibited lower VE and EE than other anode materials, for instance, graphite, and currently does not seem to be an alarming competitor for the commercialized anodes. However, all these electrochemical parameters particularly depend on electrode characteristics and operating conditions. Therefore, they can be carefully optimized to improve the performances.

6. Summary and Outlook

The current state of the art on ZnFe₂O₄ anode for LIBs demonstrates that they are very appealing materials thanks to their electrochemical features mainly related to the modification/choice of three peculiar aspects: (1) the particles’ morphologies (downsizing the dimensions, obtaining nanowires, nanorods) by using synthetic routes such as co-precipitation, hydrothermal method, electrospinning, and using sacrificial template; (2) the formation of composites with carbonaceous materials (graphene, nanowires, nanotubes) or the proper coating of the particles to obtain exotic hybrid compounds with high porosity, and (3) the proper choice of the binder and electrolyte. The scaling down of the ZFO particle dimensions to nanoscale, the first cited factor, will increase the surface area, allowing access to the maximum number of reaction sites at the electrode–electrolyte interface and reducing the diffusion path length. In addition, the volume expansion during Li insertion is minimized with respect to the bulk form and the specific capacities’ increase. However, the large surface area will lead to unwanted side reactions, which in turn produce the formation of a large amount of the SEI layer that degrades the capacity. The formation of composites with carbonaceous materials, the second factor, can prevent the side reactions with electrode/electrolyte interface and accommodate volume strain. The core shell structure is the simplest strategy to enhance the electrochemical performance. The ZFO core can be covered by a shell consisting of amorphous carbon, which can act as a protection layer to strengthen the performance of core particles, by mainly avoiding the aggregation between the particles. In addition, the carbon shell with a proper thickness provides high electronic conductivity that enhances the lithium-ion diffusion rate. Another widely employed and winning strategy is the production by electrospinning of one-dimensional (1D) nanostructures, developing a well-interconnected conducting network, which can enhance the effective interaction between the active materials and electrolyte, shorten the diffusion length of the lithium ions and electrons, and buffer the strain energy, avoiding cracking and electrode pulverization.

The third in-depth aspect concerns the binders. The widely used PVdF is progressively being substituted by CMC, PAA, and sodium alginate because they are more eco-friendly, less expensive, and able to completely cover the active material and accommodate the swelling of the electrode material. They also restrict the peeling of active material from the current collector, facilitating the tight attachment of the active material on the current collector and enhancing the electronic conductivity.

These three crucial factors explain why, in the recently published papers, great efforts are devoted to synthesizing largely porous nanomaterials with morphologies able to support the huge volume changes during intercalation/deintercalation and to allow the electrolyte permeation. As demonstrated, very good cycling stability was obtained with high-capacity values for hundreds of cycles. The best results for ZFO are represented by some successful examples in which the use of templates, such as a metal organic framework or Prussian blue analogues as precursors to prepare porous and well-dispersed ZFO subunits in carbon matrix, allowed reaching capacities higher than 800 mAh g⁻¹ at 1 C for over 200 cycles. Hollow microspheres, too, obtained from the hydrothermal method guaranteed about 850 mAh g⁻¹ up to 20 C for 100 cycles, while one-dimensional nano-architecture with a polypyrrole shell can provide 880 mAh g⁻¹ at 1 C for 220 cycles. Full cells, even if in a preliminary research state and needing optimization, seem to provide very promising capacity and power density values.

However, some drawbacks are still to be solved. Poor cycling stabilities and low coulombic efficiencies in the first cycles, due to stable SEI film formation and large irreversible capacity loss, can lead to safety problems in the application in LIBs. A possible solution is represented by the use of electrolyte additives such as fluoroethylene carbonate (FEC), allowing the reduction of SEI film thickness and minimizing the irreversible capacity. In this way, the cycle life could be increased with a modification of the surface chemistry of the SEI film.

To this aim, the research on ZnFe₂O₄ is ongoing and many papers are published every year, trying to make the academic research an industrial reality.