Mg-Doped Li₂FeTiO₄ as a High-Performance Cathode Material Enabling Fast and Stable Li-ion Storage

Abstract

As a multi-electron system material, the excellent capacity and environmentally benign properties of Li₂FeTiO₄ cathodes make them attractive for lithium-ion batteries. Nevertheless, their electrochemical performance has been hampered by poor conductivity and limited ion transport. In this work, the synthesis of Mg-doped Li₂Mg_xFe_1−xTiO₄ (LiFT-Mgx, x = 0, 0.01, 0.03, 0.05) cathode materials was successfully achieved. We observed significant gains in interlayer spacing, ionic conductivity, and kinetics. Hence, the sample of the LiFT-Mg0.03 cathode demonstrated charming initial capacity (112.1 mAh g⁻¹, 0.05 C), stability (85.0%, 30 cycles), and rate capability (96.5 mAh g⁻¹, 85.9%). This research provided precious insights into lithium storage with exceptional long-term stability and has the potential to drive the development of next-generation energy storage technologies.

1. Introduction

With the development of society, people’s demand for fast charging has increased significantly [1,2,3,4]. Recently, research on lithium-ion battery cathode materials has shown the advanced nature of keeping pace with the times, and many new materials have slowly entered people’s field of vision [5,6]. Li₂FeTiO₄ is typically characterized by high capacity, low cost, non-toxicity and environmental friendliness [7]. In addition, Li₂FeTiO₄ belongs to a multi-electron system, which can realize the transfer of multiple electrons, thus significantly increasing the specific capacity and the energy density of the battery. Based on the advantages of these products, in this era full of environmental awareness, Li₂FeTiO₄ has good appeal. Especially in the field of lithium-ion batteries, this kind of new green and pollution-free material has attracted more and more attention. However, its lack of electrical conductivity and slow dynamic migration seriously damage its practical application value [8]. Carbon coatings are often constructed to improve the dynamics of Li₂FeTiO₄ [9,10]. However, these strategies have been focused on the realization of the surface angle of the material, neglecting to change the internal electric field angle to optimize intrinsic conductivity and ion transport mechanics. Ion doping introduces impurity ions outside the material, causes internal defects, and thus changes the internal electric field distribution or internal carrier concentration of the material, which will improve its intrinsic conductivity and dynamic mobility [11]. Therefore, the study of bulk phase doping design has important practical value and significance for improving the electrochemical properties of Li₂FeTiO₄ materials.

In this work, MgCl₂ was used as raw material to achieve bulk-phase doping of Mg at the Fe site of Li₂FeTiO₄. Due to the nature of Mg’s large ion radius, the extended lithium-ion channel effect was triggered, which ultimately optimized ion migration dynamics. Innovative design of a novel Li₂Mg_xFe_1−xTiO₄ (LiFT-Mgx, x = 0, 0.01, 0.03, 0.05) cathode material for efficient and stable lithium storage. The mechanism of Mg doping was further discussed. The addition of Mg²⁺ could enlarge the interlayer spacing, promote ion transport, optimize electrode/electrolyte interface conduction, and improve electrochemical kinetics. Electrochemical analysis proved that LiFT-Mg0.03 had a relatively wide ion migration channel and the best charge migration behavior. As a result, the LiFT-Mg0.03 cathode possessed outstanding capacity (112.1 mAh g⁻¹, 0.05 C), stability (85.0%, 30 cycles), and rate performance (96.5 mAh g⁻¹, 85.9%). This research opens up new possibilities for advancing Li₂FeTiO₄ cathode materials aimed at rapid and efficient lithium storage.

2. Results and Discussion

2.1. SEM Analysis of Li₂Mg_xFe_1−xTiO₄ Samples

SEM was employed to explore the microscopic morphology of the LiFT-Mgx samples. There were no significant changes in the structure (Figure 1a–d). They exhibited a clear calcined body structure, and this was formed by the aggregation of numerous tiny and irregular particles with obvious agglomeration phenomena, and the size of the particles forming the agglomerates ranged from 200 nm to 500 nm. It was declared that Mg doping had little effect on the microscopic morphology of the materials. This phenomenon might be due to the fact that the dopant ions were able to effectively occupy the lattice positions without destroying the symmetry and integrity of the original crystals [12]. Therefore, even if the doped samples showed morphological changes, such as the appearance of new agglomerates or a slight increase or decrease in grain size, these changes were usually localised, controllable and not sufficient to change the overall properties of the material. In addition, these elements (Li, Fe, Ti, Mg, and O) were evenly distributed (Figure 1e). The results confirmed the uniformity and consistent particle size of the synthesized samples, which was crucial for the materials’ properties and applications [13].

2.2. XRD Analysis of Li₂Mg_xFe_1−xTiO₄ Samples

In order to further analyse the microstructural morphology as well as the difference in phase species of Li₂Mg_xFe_1−xTiO₄ materials under different doping levels, XRD plots were compared with each other. According to Figure 2a, it could be seen that the main phase of the sample after Mg doping was still a heterogeneous phase of Li₂FeTiO₄ and Fe₃O₄ [14,15]. Meanwhile, when the doping content reached X = 0.05, the characteristic peaks of Li₂FeTiO₄ gradually became shorter and wider. This might be due to the fact that excessive Mg doping amounts could lead to a decrease in the crystallinity of the sample, ultimately leading to a weakening of the characteristic peak strength [16]. In addition, impurity peaks (Fe₃O₄) existed in the X = 0.05 system. The possible explanation was that excessive doping of Mg²⁺ ions might cause lattice distortion, altering the local charge distribution. This change might prompt the redistribution of Fe²⁺ and Fe³⁺ to form a more stable Fe₃O₄ structure. It was worth noting that LiFT-Mg0.03 was relatively significant in both crystallinity and purity, which meant the most significant ion dynamics. Moreover, it could be seen that with the increase of doping amount, the position of the diffraction peaks shifted towards the direction of small angle corresponding to the arrow (Figure 2b). The (200) peaks exhibited a shift towards higher angles, moving from 43.7° to 43.4°. This suggested that the incorporation of Mg elements led to changes in the interlayer spacing, lattice parameters, and cell volume. This could be due to the larger ionic radius of Mg²⁺ [17]. The increased interlayer spacing in LiFT-Mg0.03 was anticipated to enhance the structure for lithium storage, facilitating faster ion movement and boosting kinetic behavior.

2.3. FTIR Analysis of Li₂Mg_xFe_1−xTiO₄ Samples

In order to investigate the structure changes of different samples, FTIR was used, as shown in Figure 3. FTIR was performed on the dry gel, pre-fired material, and cathode materials obtained from LiFT-Mg0.03. Peak regions were observed at 860 cm⁻¹, 1000–1600 cm⁻¹, and 3430 cm⁻¹, corresponding to Ti−O, C−O, C−H, and −OH groups, respectively [18]. It could be found that the samples of the three stages were basically similar in terms of peak position, which indicated that the chemical bond had not changed much. However, the strength of the pre-fired samples and the cathode materials in the 1000–1600 cm⁻¹ region was weakened because of the decomposition of esters caused by high-temperature calcination.

2.4. TG-DSC Analysis of Li₂Mg_xFe_1−xTiO₄ Samples

In order to study the thermal stability of the dry gel of LiFT-Mg0.03 material and its crystallisation process, we decided to carry out TG-DSC analysis (Figure 4). From the figure it can be seen that the whole heating process can be divided into four stages: (1) 25–200 °C; (2) 200–460 °C; (3) 460–680 °C; and (4) 680–1000 °C. In the first stage, about 1.4% of the mass was lost, corresponding to the DSC curve with two absorbing peaks and two exothermic peaks, of which the more obvious absorbing peak at 100 °C corresponded to the loss of water of crystallisation of the compounds; in the second stage, more mass was lost, about 45%, corresponding to the DSC curve with one exothermic peak and two absorbing peaks. One of the absorbing peaks indicated that the thermal decomposition of citric acid occurs to trigger a large amount of weight loss. And the other absorbing peak indicated that the metal salts (e.g., FeCl₂, Ti(OC₄H₉)₄, CH₃COOLi) had begun to decompose thermally. In the third stage, there was no obvious loss of mass, and the corresponding DSC curve had no obvious heat-absorbing and exothermic peaks. This might be the initial generation of LiFT-Mg0.03 crystals. There was an 8% loss of mass in the fourth stage, and the corresponding DSC curve was relatively flat with no obvious heat-absorbing and exothermic peaks. This stage was the further generation of LiFT-Mg0.03 crystals.

2.5. Valence Analysis of Li₂Mg_xFe_1−xTiO₄ Samples

For further analysis of the elemental valence states of LiFT and the LiFT-Mg0.03 samples, XAS was organized. Standard Fe₂O₃ and FeO reference spectra were used to compare with the K-edge of Fe. Both LiFT and the LiFT-Mg0.03 samples displayed signs of Fe existing in a mixed valence state, with valence states of +2.53 for LiFT and +2.61 for LiFT-Mg0.03 (Figure 5a). This phenomenon indicated that the introduction of Mg²⁺ slightly increased the valence state of the Fe layer, which was conducive to the enhancement of the Fe-O effect and facilitated the rapid migration of lithium ions [19,20,21,22,23]. Additionally, the Ti K-edge XANES spectra revealed that Li₂FeTiO₄ exhibited no notable changes, suggesting that the Ti valence remained consistently at +4 (Figure 5b).

2.6. Electrochemical Properties of Li₂Mg_xFe_1−xTiO₄ Samples

A series of experiments were carried out to examine how Mg doping influences the electrochemical performance of LiFT, including LiFT-Mg0.01, LiFT-Mg0.03, and LiFT-Mg0.05. The initial capacities were measured at 114.5, 113.6, 112.1, and 111.4 mAh g⁻¹, respectively (Figure 6a–d). Because magnesium is chemically inert, its capacity tends to decay as the magnesium content increases. In addition, the repeatability of the charge curves of doped cathode materials was better than that of undoped materials. As a result, Mg doping helped improve structural stability and promoted the fast transport of lithium ions.

In addition, the samples underwent further testing to evaluate their cycling performance at 0.05 C. At the 30th cycle, the capacities of LiFT, LiFT-Mg0.01, LiFT-Mg0.03, and LiFT-Mg0.05 were measured at 79.7, 85.8, 95.3 and 93.4 mAh g⁻¹, representing 69.6%, 75.5%, 85.0% and 83.8% of their initial capacities, respectively (Figure 7a). The LiFT-Mg0.03 cathode material showed the greatest stability across the charge/discharge cycles. Finally, rate performance was evaluated across various current densities (0.05–2.0 C) (Figure 7b). It was observed that as the current density increased, the capacity of all the samples declined. At 2.0 C, the capacities were 56.1, 61.7, 68.1 and 65.4 mAh g⁻¹, which accounted for 48.4%, 54.4%, 60.6% and 58.1% of their initial values, respectively. After cycling the samples back to 0.05 C, the reversible capacities were recorded at 83.5, 88.4, 96.5 and 91.6 mAh g⁻¹, representing 72.0%, 78.0%, 85.9% and 82.4% of the original capacities. Clearly, the LiFT-Mg0.03 cathode material demonstrated the best rate performance, which could be attributed to its wider layer spacing and improved ion transmission characteristics. This finding was further confirmed by XRD and EIS. Moreover, compared with LiFT, LiFT-Mg0.03 had better charge–discharge characteristics at different current densities (Figure 7c,d). The coulomb efficiency of the two samples was greatly improved compared with Figure 6. Because of the pre-activation, the growth of the CEI film was conducive to a more stable charge and discharge.

In order to investigate the relationship between impedance and Mg doping, the EIS of different doped materials were compared, as shown in Figure 8. The main features of the curves were a semicircle and a straight line [24,25]. The high-frequency semicircle was said to be closely related to charge transfer resistance (R_ct) and ion migration dynamics [26]. With the increase in doping amount, the impedance of electrode material decreased first and then increased further. The R_ct values for LiFT, LiFT-Mg0.01, LiFT-Mg0.03, and LiFT-Mg0.05 were recorded at 1200.8 Ω, 802.5 Ω, 408.8 Ω and 671.9 Ω (

Table 1. The fitted impedance (R_s and R_ct) for the samples.

	LiFT	LiFT-Mg0.01	LiFT-Mg0.03	LiFT-Mg0.04
Rs (Ω)	8.3	7.4	5.8	6.9
Rct (Ω)	1200.8	802.5	408.8	671.9

). These results convincingly expounded that Mg doping upgraded the sodium ion movement, and further supported the optimization design of LiFT-Mg0.03.

In order to better evaluate the effect of a high current environment on LiFT-Mg0.03, we conducted a high current cycle test again. After the study, it was found that LiFT-Mg0.03 still retained 74.4% of the initial value after 200 cycles at 2.0 C (Figure 9a). Then, the morphology and structure of the recycled materials were tested again. After a long cycle, LiFT-Mg0.03 could still maintain its original morphology and crystal structure while the crystallinity decreased after cycling, possibly due to lattice damage caused by long-term charge and discharge (Figure 9b,c). In general, the sample still had the potential for long-term operation with a certain high current.

In order to study the basic history and thermodynamic characteristics of the electrochemical reaction of LiFT-Mg0.03, CV curves were performed at 1.5–4.8 V. It could be noticed that there were still three sets of oxidation peaks and three sets of reduction peaks in Figure 10. And the locations of the peaks in both cycles were strikingly similar, suggesting its favourable electrochemical reversibility. A notable peak appeared in the potential range around 3.0 V, corresponding to the oxidation of Fe²⁺ to Fe³⁺, a typical reaction observed during cycling [27]. As the voltage increased to 3.8 V, Fe³⁺ was further oxidized to Fe⁴⁺ [28,29]. These indicated that the battery could store energy reliably while maintaining stability and a long lifespan.

3. Materials and Methods

3.1. Material Preparation

The sol–gel technique was employed for creating Li₂Mg_xFe_1−xTiO₄ (LiFT-Mgx, x = 0, 0.01, 0.03, 0.05). Li, Fe, Ti, and Mg came from CH₃COOLi·2H₂O, FeCl₂·4H₂O, tetrabutyl titanate (Ti(OC₄H₉)₄), and MgCl₂·6H₂O, respectively. And citric acid was the chelating agent. The raw materials were initially mixed in ethanol at their stoichiometric ratio. Taking LiFT-Mg0.03 as an example, the molar ratio of CH₃COOLi·2H₂O, Ti(OC₄H₉)₄, FeCl₂·4H₂O, and MgCl₂·6H₂O was 2:1:0.07:0.03. So the amount of CH₃COOLi·2H₂O, Ti(OC₄H₉)₄, FeCl₂·4H₂O, and MgCl₂·6H₂O were 0.2040 g, 0.3403 g, 0.0139 g, and 0.0061 g. And the amount of ethanol was 30 mL. Treating the solution with a water bath (65 °C, 5 h) formed a yellow gel. The gel was then further dried to produce a dry gel (120 °C, 12 h). Next, the high-temperature pretreatment (500 °C, 8 h) in argon was performed on the dry ground gel for the first time. This was the pre-fired sample. Finally, the secondary calcination of the pre-fired sample was performed at 650 °C for 8 h. Among them, the calcination rate was 5 °C/min, and it was cooled with the furnace.

3.2. Characterization

A scanning electron microscope (SEM Hitachi S-4800, Hitachi, Japan) was suitable for microstructure analysis. The crystal structure was determined using X-ray diffractometry (XRD, Cu-Kα, 5° min⁻¹ scan rate, 2θ mode, X/Pert Pro 3040/60), Fourier transform infrared spectroscopy (FTIR, 8400S, Shimadzu, Kyoto, Japan), and Mössbauer spectroscopy (MS, 57Co, Shimadzu). Using X-ray absorption spectroscopy (XAS) at the Beijing Synchrotron Radiation Facility (BSRF)’s beamline 4B9A XRD station, we measured the Ti and Fe K-edges to determine elemental valences.

3.3. Electrochemical Measurements

Synthesized materials, acetylene black, and polyvinylidene fluoride (PVDF) were combined in N-methyl-2-pyrrolidinone (NMP) at an 8:1:1 weight ratio to create the cathode slurry. Subsequently, the cathode slurry was coated on an Al foil, wherein the Al foil was a collector. Following this, it underwent overnight drying at 80 °C. Inside a glovebox, the negative shell, Na sheet, separator, cathode material, gasket, shrapnel and positive shell were assembled in order to obtain the 2032 coin battery. In this group, the electrolyte was lithium iron phosphate (KLD-LFP01), and Celgard 2400 was the separator. Using a LAND system, electrochemical properties were evaluated from 1.5 to 4.8 V.

4. Conclusions

In conclusion, Mg-doped Li₂Mg_xFe_1−xTiO₄ (LiFT-Mgx, x = 0, 0.01, 0.03, 0.05) cathode materials were synthesized using the sol–gel method. The LiFT-Mg0.03 sample exhibited smaller R_ct, indicating superior electrochemical kinetics. As a result, the LiFT-Mg0.03 cathode realized impressive specific capacities (112.1 mAh g⁻¹, 0.05 C) and stability (85.0%, 30 cycles). This could be attributed to the improvement in electrochemical kinetics. This study provided fresh insights into the relationship between structural features and performance, which could open new avenues for the development of advanced Li₂FeTiO₄ cathode materials in future research.