Synthesis of Fe2+ Substituted High-Performance LiMn1−xFexPO4/C (x = 0, 0.1, 0.2, 0.3, 0.4) Cathode Materials for Lithium-Ion Batteries via Sol-Gel Processes

A series of carbon-coated LiMn1−xFexPO4 (x = 0, 0.1, 0.2, 0.3, 0.4) materials are successfully constructed using glucose as carbon sources via sol-gel processes. The morphology of the synthesized material particles are more regular and particle sizes are more homogeneous. The carbon-coated LiMn0.8Fe0.2PO4 material obtains the discharge specific capacity of 152.5 mAh·g−1 at 0.1 C rate and its discharge specific capacity reaches 95.7 mAh·g−1 at 5 C rate. Iron doping offers a viable way to improve the electronic conductivity and lattice defects of materials, as well as improving transmission kinetics, thereby improving the rate performance and cycle performance of materials, which is an effective method to promote the electrical properties.

The reducibility of carbon can inhibit the oxidation of Mn 2+ and the growth and agglomeration of material particles to a certain extent, mainly by reducing the particle size of the cathode material, preventing the oxidation of Fe 2+ [17], increasing the specific surface area, and shortening the diffusion path of Li + to promote the material electrochemical performance. Mixing organic carbon source and inorganic carbon source for coating also helps to improve the performance of materials [5]. At the same time, the carbon coating layer can form a transport network between different particles and improve the conductivity of materials [18]. Note that carbon materials with a mass fraction of 25% should be used for coating modification to achieve the purpose of expanding battery capacity [19].
The sol-gel process is to dissolve the complexing agent and stoichiometric amounts of raw materials in a solvent, react in an inert gas atmosphere to form a gel, and then calcinate to obtain the final product. The sol-gel processes have many advantages, such as the low temperature of the reaction, small particle size, and uniform distribution of composition compared to solid-phase reactions [20][21][22][23]. In addition, one of the characteristics the solgel process has is to control the particle size of powder easily, which is superior to the solid-phase reaction.
The formation of solid solutions between LiFePO 4 and LiMnPO 4 starts a new method in the research and design of electrode materials due to its combination of the high electron transportability of LiFePO 4 and the high-voltage platform of LiMnPO 4 , which was based on olivine-shaped phosphate material with a potential application in HEV [24]. The doping of Fe ions dilutes the concentration of Mn ions, thereby reducing the Jahn-Teller distortion caused by Mn 3+ [18]. The introduction of Mn 2+ in the lattice can widen the lithiumion diffusion channel to some extent, and the presence of Fe 2+ can improve electrode kinetics. The interaction between them [25,26] makes LiFe 1-x Mn x PO 4 /C solid solution exhibit excellent electrochemical performances and kinetic properties.
In this paper, Mn(CH 3 COO) 2 ·4H 2 O, LiOH·H 2 O, FeC 2 O 4 ·2H 2 O, and NH 4 H 2 PO 4 were used as raw materials to prepare Fe 2+ substituted LiMn 1−x Fe x PO 4 /C (0 < x ≤ 0.4) positive electrode materials via sol-gel processes. At the same time, we used the same method to synthesize LiMnPO 4 for comparison with the modified material. LiMn 1-x Fe x PO 4 /C electrode materials' electrochemical properties and their microstructure were investigated by adjusting the ratio of iron and manganese. Thereafter, samples that had remarkable electrochemical performance were selected comprehensively. Figure 1 is the thermogravimetric curve of precursor of LiMnPO 4 /C sample synthesized by sol-gel processes. As can be seen from the figure, the differential thermal gravity (DTG) curve shows two strong endothermic peaks at 250 • C and 400 • C; correspondingly, the TG curve has two obvious weight-loss steps in the range of 25 • C to 800 • C. When the temperature range is between 200 • C and 500 • C, the weight loss of the sample is approximately 50%, which corresponds to the high temperature decomposition process of the precursor organic matter, where the mass loss in the range of 200 • C to 300 • C corresponds mainly to the decomposition of NH 4 H 2 PO 4 and citric acid, and the weight loss in the range of 300 • C to 500 • C corresponds to the decomposition of Mn(CH 3 COO) 2 , accompanied by the generation of LiMnPO 4 . When the temperature is higher than 500 • C, the TG curve tends to be basically flat and enters the relatively constant weight region, indicating that there is no more phase change occurring at temperatures higher than 500 • C, and LiMnPO 4 has crystallized completely, and that higher temperature treatment only increases the crystallinity of LiMnPO 4 . As can be seen from the TG-DTG curve, the decomposition reactions of the precursors and the chemical reactions that take place between them occur mainly before 500 • C, while the weight loss between 30 • C and 200 • C is only 2%, which corresponds to the evaporation process of water contained in the precursors. When the temperature is between 600-800 • C, the weight loss rate is 1%. Therefore, the two-stage sintering method is adopted in the whole sintering process. See the experiment operation section for details. precursors. When the temperature is between 600-800 °C, the weight loss rate is 1%. Therefore, the two-stage sintering method is adopted in the whole sintering process. See the experiment operation section for details.  Figure 2 shows XRD patterns of LiMn1−xFexPO4/C (0 ≤ x ≤ 0.4) synthesized by the sol-gel processes. Compared with the standard card (JCPDS NO.74-0375), the positions of the diffraction peaks of each sample were highly consistent with the standard card, so the materials had an olivine-type LiMnPO4 structure with Pnmb space group. This indicated that Fe-doping did not cause significant changes in LiMnPO4 structure and crystal form. Hence it suggested that Fe 2+ had successfully entered the crystal space. No obvious carbon peak was observed in the composite material spectrum, illustrating the amorphous nature of the residual carbon layer and its low content [27,28]. As shown in Figure 2b, the diffraction peaks of each sample shifted slightly toward the increase of the diffraction angle with the increase of the Fe-doped amount. This is because Fe 2+ partially replaces the Mn 2+ site and the Fe ion radius (0.92 Å) is smaller than the Mn ion radius (0.97 Å); the position of the diffraction peak had slightly shifted [10]. Table S1 (Supplementary Materials) shows the carbon content of LiMn1−xFexPO4/C (0 ≤ x ≤ 0.4) materials. As can be seen from Table S1 (Supplementary Materials), the carbon content of different samples was about 7.2%, and the carbon content of LiMn1−xFexPO4/C materials was similar.  Hence it suggested that Fe 2+ had successfully entered the crystal space. No obvious carbon peak was observed in the composite material spectrum, illustrating the amorphous nature of the residual carbon layer and its low content [27,28]. As shown in Figure 2b, the diffraction peaks of each sample shifted slightly toward the increase of the diffraction angle with the increase of the Fe-doped amount. This is because Fe 2+ partially replaces the Mn 2+ site and the Fe ion radius (0.92 Å) is smaller than the Mn ion radius (0.97 Å); the position of the diffraction peak had slightly shifted [10].  For LiMn1−xFexPO4/C (0 ≤ x ≤ 0.4) cathode materials, their lattice parameters after Rietveld refinement are summarized in Table 1. The calculated lattice parameters decrease linearly along with the shrink of the unit cell with the increase of the Fe-doping amount, and the change is approximately linear, which was consistent with Vegard's For LiMn 1−x Fe x PO 4 /C (0 ≤ x ≤ 0.4) cathode materials, their lattice parameters after Rietveld refinement are summarized in Table 1. The calculated lattice parameters decrease linearly along with the shrink of the unit cell with the increase of the Fe-doping amount, and the change is approximately linear, which was consistent with Vegard's law. This indicated that Fe 2+ doping could cause the lattice shrinkage of LiMnPO 4 , but the crystal structure of LiMnPO 4 was not affected [26,29]. Thus, it can be seen that Fe 2+ was successfully embedded in the olivine structure of LiMnPO 4 .

Micromorphology Analysis
The morphology of LiMn 1−x Fe x PO 4 /C (0 ≤ x ≤ 0.4) materials is presented in Figure 3. It can be seen from Figure 3a-e that the distribution of particle sizes for LiMn 1−x Fe x PO 4 /C materials are very uniform. The primary particle size is about 300 nm. Synthesized materials had a relatively regular morphology, and most of them are spherical. This means that iron doping helps to improve the morphology of the material particles. Figure 3f shows an HRTEM image of the LiMn 0.8 Fe 0.2 PO 4 /C cathode material, from which it can be clearly seen that an amorphous carbon coating with a thickness of approximately 1.5 nm is present on the surface of the material. The particle size distribution statistics shown in Figure S1 (Supplementary Materials) verified that the LiMn 0.8 Fe 0.2 PO 4 /C cathode material possesses a more uniform particle size distribution than the other four samples, with an average particle size of 260 nm. The particle size of materials was decreased first and then increased with the increase of the amount of Fe-doping, and agglomeration of spherical particles intensified. Increasing the particle size of materials will increase the migration distance of lithium ions during charging and discharging. Thus, the utilization rate of active substances is reduced [22], which is not conducive to modify the electrochemical performance of materials.

Electrochemical Analysis
The electrochemical performance of materials was tested under constant current charging and discharging mode. Figure 4 is the charge-discharge capacity diagram of LiMn1−xFexPO4/C (0 ≤ x ≤ 0.4) cathode materials at a rate of 0.1 C. Theoretically, since the

Electrochemical Analysis
The electrochemical performance of materials was tested under constant current charging and discharging mode. Figure 4 is the charge-discharge capacity diagram of LiMn 1−x Fe x PO 4 /C (0 ≤ x ≤ 0.4) cathode materials at a rate of 0.1 C. Theoretically, since the total capacity of each synthetic sample is identical during the charge-discharge process, the lengths of charge/discharge platforms should be equal. However, since the maximum overpotential causes the minimum phase-change activation energy, the length of charge/discharge platforms in the 3.5 V and 4.1 V regions is not symmetrical [30]. It can be seen from Figure 4 that LiMn 1−x Fe x PO 4 /C materials exhibit charge/discharge voltage plateaus at 3.5 V, corresponding to the Fe 2+ /Fe 3+ redox couple, and at 4.1 V, associating with the Mn 2+ /Mn 3+ redox couple, of which the LiMn 0.8 Fe 0.2 PO 4 /C material had the widest charge/discharge platform. The high operating voltage platform is an important parameter for the cathode material to meet the high energy-density requirement of electronic products. The discharge specific capacity of the LiMn 0.8 Fe 0.2 PO 4 /C material also reached a peak of 152.5 mAh·g −1 . Therefore, the LiMn 0.8 Fe 0.2 PO 4 /C material had the smallest degree of polarization during charging and discharging among several prepared materials, and had the highest specific discharge capacity and battery energy density [26]. At the same time, it was confirmed that the Mn site was successfully partially replaced by Fe, and Fe substitution had a positive effect on the two-phase transition between the charge/discharge reactions.  The partial substitution of Fe at Mn sites can significantly enhance the lithium-ion diffusion and reduce the polarization of LiMnPO4, which helps to suppress the Jahn-Teller distortion. With the increase of Mn content and the amount of iron substitution decreases in olivine materials, the discharge specific capacity of materials decreases instead during the charge/discharge cycle. This is due to the loss of the electrochemical activity of the Mn 2+ /Mn 3+ redox center, which has been reported in the previous literature [30,31]. LiMn1−xFexPO4/C materials had better charge-discharge performance, and the cycle performance was enhanced, as shown in Figure 5a. These voltage plateaus agree well with the anode and cathode peaks in the CV curves and are also consistent with the particle size variation trend of LiMn1−xFexPO4/C materials ( Figure S1, Supplementary Materials). The partial substitution of Fe at Mn sites can significantly enhance the lithium-ion diffusion and reduce the polarization of LiMnPO 4 , which helps to suppress the Jahn-Teller distortion. With the increase of Mn content and the amount of iron substitution decreases in olivine materials, the discharge specific capacity of materials decreases instead during the charge/discharge cycle. This is due to the loss of the electrochemical activity of the Mn 2+ /Mn 3+ redox center, which has been reported in the previous literature [30,31]. LiMn 1−x Fe x PO 4 /C materials had better charge-discharge performance, and the cycle performance was enhanced, as shown in Figure 5a. These voltage plateaus agree well with the anode and cathode peaks in the CV curves and are also consistent with the particle size variation trend of LiMn 1−x Fe x PO 4 /C materials ( Figure S1, Supplementary Materials). stead during the charge/discharge cycle. This is due to the loss of the electrochemical activity of the Mn 2+ /Mn 3+ redox center, which has been reported in the previous literature [30,31]. LiMn1−xFexPO4/C materials had better charge-discharge performance, and the cycle performance was enhanced, as shown in Figure 5a. These voltage plateaus agree well with the anode and cathode peaks in the CV curves and are also consistent with the particle size variation trend of LiMn1−xFexPO4/C materials ( Figure S1, Supplementary Materials).    Figure 5 showed the cycle performance of LiMn 1−x Fe x PO 4 /C (0 ≤ x ≤ 0.4) cathode materials and rate performance at current densities of 0.1 C, 0.2 C, 0.5 C, 1 C, 2 C, and 5 C, respectively. The high rate performance at high operating voltage should be attributed to the enhancement of lithium-ion diffusion kinetics and the better contact between the cathode material and the current collector due to the uniform coating of conductive carbon material [32]. As can be seen from Figure 5b, we can see that the discharge specific capacity of LiMn 1-x Fe x PO 4 /C materials decreases with the increase of rate. When the ironmanganese ratio was 2:8, the discharge specific capacities of materials under the condition of 0.1 C to 5 C were higher than that of the other four synthetic materials. The discharge specific capacity at 5 C rate is only 95.7 mAh·g −1 , which was 73.3% of 1 C rate and 56.3% of the theoretical capacity, illustrating that the material had a longer charge/discharge voltage platform and a higher discharge specific capacity. Therefore, LiMn 0.8 Fe 0.2 PO 4 /C exhibited excellent rate performance and cycling stability.
The electrochemical properties of the composite material largely depended on the ratio of Fe to Mn [2]. Increasing the proportion of manganese may cause the discharge specific capacity of the material to decrease. Due to the intense Jahn-Teller distortion caused by Mn 3+ , the lithium-ion extraction process inside the LiMnPO 4 material is hindered, and the LiMnPO 4 material has an olivine structure, and the structural disorder in the crystal lattice may hinder lithium exchange in the [010] channel to a certain extent [33]. The mixed arrangement of Mn 2+ at the Li + position hinders the transmission of lithium ions in one-dimensional channels and reduces the electrochemical activity of the material [29,34], while the addition of iron ions increases the capacity of the electrode material, which means that iron doping is more beneficial to the deintercalation/intercalation process of Li + [1,35]. At the same time, the Jahn-Teller distortion is related to the Mn-Mn distance in the crystal structure [34][35][36]. It is presumed that the ion radius of Fe 2+ (0.92 Å) is smaller than the ion radius of Mn 2+ (0.97 Å), so substitution with Fe 2+ will increase the Mn-Mn distance and reduced the Jahn-Teller distortion. When Fe is embedded in the position of Mn in the crystal lattice of LiMnPO 4 , the smaller Fe atom energy will leave more space for the migration of Li + due to the small size of the Fe atom, and inhibit the dissolution of Mn 3+ in the electrolyte, thus stabilizing the crystal ion configuration of materials. Iron-doping modification can improve the ionic conductivity and lattice defects of materials to a certain extent, and enhance its transmission kinetics, thereby improving the rate performance and cycle performance of materials under charge/discharge conditions, which is an effective improvement method for electrochemical performance. Figure 6 shows the cyclic voltammetry curves of the LiMn 0.8 Fe 0.2 PO 4 /C cathode material and the LiMnPO 4 /C sample. The oxidation peak and reduction peak of the two samples have good symmetry. The CV curves of the LiMn 0.8 Fe 0.2 PO 4 /C cathode material show two pairs of peaks, one at 3.51/3.62 V, corresponding to Fe 2+ /Fe 3+ redox couple, and the other at 4.0/4.23 V, associating with Mn 2+ /Mn 3+ redox couple. The LiMnPO 4 /C material showed only one pair of redox couple associated with Mn 2+ /Mn 3+ , as shown in Figure 6d. The anodic peak current was equal to the corresponding cathodic peak current, and the redox peak shape was not destroyed, which indicated that the prepared sample had good reversibility. Two of the materials had good cycle stability, and the ratio of the peak area corresponded to the stoichiometric ratio of Mn/Fe. The CV curve showed that the potential difference of the LiMn 0.8 Fe 0.2 PO 4 /C cathode material between oxidative and reductive peaks was smaller than the LiMnPO 4 /C sample, and the degree of polarization was lower than the LiMnPO 4 /C sample. The energy density of the modified material is linked to the redox potential of transition metal ions [36]. The results showed that iron doping increases the diffusion coefficient of lithium ions, and especially accelerates the conversion between Mn 2+ ↔Mn 3+ ; the peak in the CV test proved the above statement. The oxidation and reduction peaks of the Mn 2+ /Mn 3+ and Fe 2+ /Fe 3+ redox pairs were sharper, and the peak intensity of Fe 2+ /Fe 3+ increased gradually. This phenomenon can be attributed to the fact that as the iron content in the material increases, the intercalation/deintercalation kinetics of lithium occurring at the electrode/electrolyte interface and the diffusion rate of lithium ions in the crystal lattice also increase. The CV curves showed that the cathodic peak current was equal to the corresponding highest anodic peak, which indicated that the prepared sample had good chemical reversibility. With the change of the scanning rate, the oxidation and reduction peaks shape was not destroyed, and the potential difference between the oxidation The CV curves showed that the cathodic peak current was equal to the corresponding highest anodic peak, which indicated that the prepared sample had good chemical reversibility. With the change of the scanning rate, the oxidation and reduction peaks shape was not destroyed, and the potential difference between the oxidation voltage and the reduction voltage changed little. Nanoparticles can ensure that the material has a high lithium-ion diffusion coefficient. For the reversible reaction, the Randles-Sevcik equation can be used to calculate the chemical diffusion coefficient of lithium ions using the peak current during scanning at different sweep speeds: I p is the peak current of the positive electrode at different scanning speeds, ν is the scan rate (V·s −1 ), A is the area of the electrode pole piece (cm 2 ), C Li is the lithium-ion concentration, D Li is the lithium-ion diffusion coefficient (cm 2 ·s −1 ), and n is the number of electrons in the redox process. The linear response of the cathodic peak current (Ip) as a function of the square root of the scan rate (ν) is shown in Figure 7. The electrochemical impendence spectroscopy spectrum test was carried out to further analyze the electrochemical kinetics of the above samples. The EIS spectrum consists of a high-frequency intercept, a compressed semicircle from high to intermediate frequencies, and a low-frequency oblique line. The small intercept at a high frequency corresponds to the solution resistance of the battery and the recessed semicircle at the high-frequency-intermediate frequency are related to the charge transfer resistance at the electrode/electrolyte interface and the electric double layer capacitance between the electrolyte and the positive electrode. The oblique line is attributed to the Warburg impedance corresponding to lithium-ion diffusion in the electrode. From Figure 8, LiMn1−xFexPO4/C materials showed small interface resistance and faster diffusion of Li + , which could be ascribed to the doping of Fe. A smaller Rct value is conducive to rapid electrochemical reactions and can improve the material's electrochemical performance. This result was also in agreement with the results of Figures 4 and 5. Through comparison, it was found that the EIS impedance value of samples were maintained between 40-100 Ω, and the charge transfer resistance values were reduced by metal ion doping and carbon coating. Among them, the LiMn0.8Fe0.2PO4/C sample had the smallest impedance, indicating that iron doping significantly improved the electronic conductivity of the crystal of the LiMnPO4 sample, and the EIS results were consistent with the electrochemical performance of LiMn1−xFexPO4/C samples. The electrochemical impendence spectroscopy spectrum test was carried out to further analyze the electrochemical kinetics of the above samples. The EIS spectrum consists of a high-frequency intercept, a compressed semicircle from high to intermediate frequencies, and a low-frequency oblique line. The small intercept at a high frequency corresponds to the solution resistance of the battery and the recessed semicircle at the high-frequency-intermediate frequency are related to the charge transfer resistance at the electrode/electrolyte interface and the electric double layer capacitance between the electrolyte and the positive electrode. The oblique line is attributed to the Warburg impedance corresponding to lithium-ion diffusion in the electrode. From Figure 8, LiMn 1−x Fe x PO 4 /C materials showed small interface resistance and faster diffusion of Li + , which could be ascribed to the doping of Fe. A smaller R ct value is conducive to rapid electrochemical reactions and can improve the material's electrochemical performance. This result was also in agreement with the results of Figures 4 and 5. Through comparison, it was found that the EIS impedance value of samples were maintained between 40-100 Ω, and the charge transfer resistance values were reduced by metal ion doping and carbon coating. Among them, the LiMn 0.8 Fe 0.2 PO 4 /C sample had the smallest impedance, indicating that iron doping significantly improved the electronic conductivity of the crystal of the LiMnPO 4 sample, and the EIS results were consistent with the electrochemical performance of LiMn 1−x Fe x PO 4 /C samples.  The sols were formed after 6 h reaction with 80 °C constant-temperature water bath, and then the excess water was removed by vacuum distillation to get gel. To make the gel fully dried, it was transferred to the vacuum drying box for drying at 140 °C for 24 h. The dried gel mixing with anhydrous glucose (25 wt%) was placed in quartz boats. N2, as a protective gas, was introduced into the tube furnace. The mixture was heated from room temperature to 350 °C and kept for 4 h to fully decompose the organic matter; then, the samples were fully crystallized by heating from 350 °C to 600 °C and holding for 10 h. The heating rate of the whole sintering process was set at 3 °С·min -1 . The calcined product was moved out and ground again, whereafter it was transferred to a dry place for storage.

Characterization of Synthesized Materials
Thermal gravimetric analysis (TG) was accomplished via a PerkinElmer TGA4000 instrument (PerkinElmer, Waltham, MA, USA) at a heating rate of 10 °С·min -1 in high purity inert gas to determine the thermal properties of LiMnPO4/C precursors, and the temperature range was 30-800 °C. The crystal structure was researched by X-ray diffraction (XRD, PANalytical Xpert Pro, Almelo, The Netherlands) with the Cu Kα source as the incident beam (λ = 1.5406 Å) at a voltage of 40 kV and a current of 10 mA, and scanning the 2θ range from 5 to 80° at a scan rate of 10°·min −1 . The carbon content of LiMn1−xFexPO4/C materials was tested using elemental analyzer (Vario EL III, Elementar, Berlin, Germany). JADE refined the XRD spectrum to estimate the lattice parameters. The morphology of synthetic materials was observed by scanning electron microscopy (SEM, JSM-7001F, JEOL, Tokyo, Japan), one of the most commonly used surface topography observation methods. The High Resolution Transmission Electron Microscope (HRTEM) image of carbon coating was observed with transmission electron microscopy (TEM, JEM 2100F, JEOL, Tokyo, Japan).  The sols were formed after 6 h reaction with 80 • C constant-temperature water bath, and then the excess water was removed by vacuum distillation to get gel. To make the gel fully dried, it was transferred to the vacuum drying box for drying at 140 • C for 24 h. The dried gel mixing with anhydrous glucose (25 wt%) was placed in quartz boats. N 2 , as a protective gas, was introduced into the tube furnace. The mixture was heated from room temperature to 350 • C and kept for 4 h to fully decompose the organic matter; then, the samples were fully crystallized by heating from 350 • C to 600 • C and holding for 10 h. The heating rate of the whole sintering process was set at 3 • C·min -1 . The calcined product was moved out and ground again, whereafter it was transferred to a dry place for storage.

Characterization of Synthesized Materials
Thermal gravimetric analysis (TG) was accomplished via a PerkinElmer TGA4000 instrument (PerkinElmer, Waltham, MA, USA) at a heating rate of 10 • C·min -1 in high purity inert gas to determine the thermal properties of LiMnPO 4 /C precursors, and the temperature range was 30-800 • C. The crystal structure was researched by X-ray diffraction (XRD, PANalytical Xpert Pro, Almelo, The Netherlands) with the Cu Kα source as the incident beam (λ = 1.5406 Å) at a voltage of 40 kV and a current of 10 mA, and scanning the 2θ range from 5 to 80 • at a scan rate of 10 • ·min −1 . The carbon content of LiMn 1−x Fe x PO 4 /C materials was tested using elemental analyzer (Vario EL III, Elementar, Berlin, Germany). JADE refined the XRD spectrum to estimate the lattice parameters. The morphology of synthetic materials was observed by scanning electron microscopy (SEM, JSM-7001F, JEOL, Tokyo, Japan), one of the most commonly used surface topography observation methods. The High Resolution Transmission Electron Microscope (HRTEM) image of carbon coating was observed with transmission electron microscopy (TEM, JEM 2100F, JEOL, Tokyo, Japan).

Electrode Preparation and Electrochemical Characterization
The tested electrodes of LiMn 1−x Fe x PO 4 /C (0 ≤ x ≤ 0.4) were made by mixing 80 wt% active material, 10 wt% acetylene black, and 10 wt% polyvinylidene fluoride (PVDF) into a slurry substance. After adding N-methyl-2-pyrrolidone (NMP) as a solvent to the mixed materials, the mixed slurry was coated onto an aluminum-foil current collector to prepare electrode materials. The material loading of the cathode film is about 1.6-2.2 mg·cm −2 . The electrodes attached to the aluminum foil were cut into 12 mm diameter discs after a roller pressing operation. CR2032 coin cells, which were used to conduct the electrochemical measurements of synthesized composites, were being assembled in an argon-filled glove box. An amount of 1 M LiPF 6 was selected as the electrolyte, which was dissolved in a mixed solution composed of ethylene carbonate (EC), dimethyl carbonate (DMC), and ethyl methyl carbonate (EMC). The volume ratio of mixed solution is 1:1:1. The rate performance and cycling performance of LiMn 1-x Fe x PO 4 /C (x = 0, 0.1, 0.2, 0.3, 0.4) electrode materials were tested by the Neware battery test system (Neware BTS-5 V-10 mA, Neware, Shenzhen, China) at the potential window between 2.5 and 4.5 V. Cyclic voltammetry (CV) was carried out on the CHI600E electrochemical workstation (CH Instruments, Austin, TX, USA) under the same range of potential window. The electrochemical impendence spectroscopy test (EIS) was conducted using the CHI600E electrochemical workstation with a frequency range of 10 −2 to 10 5 Hz and a range of 5 mV.

Conclusions
In short, we successfully developed a series of LiMn 1−x Fe x PO 4 /C (0 ≤ x ≤ 0.4) materials with uniform particle size distribution and excellent electrochemical performance through the sol-gel process. By modifying the iron-doping ratio of LiMn 1−x Fe x PO 4 /C materials, it was observed by XRD and SEM that when the iron-manganese ratio was 2:8, the sample had an olivine-type LiMnPO 4 structure with Pnmb space group. With the increase of Fe doping amount, the calculated lattice parameters decreased, indicating the successful integration of Fe 2+ . When the discharge rate was 0.1 C, the discharge specific capacity was 152.2 mAh·g −1 , and 130.5 mAh·g −1 at a rate of 1 C. The improved properties of modified materials were attributed to the successful insertion of Fe 2+ , so iron doping was an effective method to improve the electrochemical properties. Therefore, LiMn 1−x Fe x PO 4 /C materials are promising cathode materials for lithium-ion batteries.

Conflicts of Interest:
The authors declare no conflict of interest.