Enhanced electrochemical performance of LiFePO₄/C nanocomposites due to in situ formation of Fe₂P impurities

Abstract

We have studied LiFePO₄/C nanocomposites prepared by sol-gel method using lauric acid as a surfactant and calcined at different temperatures between 600 and 900 °C. In addition to the major LiFePO₄ phase, all the samples show a varying amount of in situ Fe₂P impurity phase characterized by x-ray diffraction, magnetic measurements, and Mössbauer spectroscopy. The amount of Fe₂P impurity phase increases with increasing calcination temperature. Of all the samples studied, the LiFePO₄/C sample calcined at 700 °C which contains ∼15 wt% Fe₂P shows the least charge transfer resistance and a better electrochemical performance with a discharge capacity of 136 mA h g⁻¹ at a rate of 1 C, 121 mA h g⁻¹ at 10 C (∼70 % of the theoretical capacity of LiFePO₄), and excellent cycleability. Although further increase in the amount of Fe₂P reduces the overall capacity, frequency-dependent Warburg impedance analyses show that all samples calcined at temperatures ≥700 °C have an order of magnitude higher Li⁺ diffusion coefficient (∼1.3 × 10⁻¹³ cm² s⁻¹) compared to the one calcined at 600 °C, as well as the values reported in literature. This work suggests that controlling the reduction environment and the temperature during the synthesis process can be used to optimize the amount of conducting Fe₂P for obtaining the best capacity for the high power batteries.

Introduction

Lithium iron phosphate (LiFePO₄) has been used as a cathode material in commercially available rechargeable batteries for nearly two decades. LiFePO₄ is intensely investigated since it was proposed by Padhi et al. [1] as a possible cathode material for Li-ion rechargeable batteries. It is one of the most promising cathode material for the next generation Li-ion batteries for military, electric vehicles, and aerospace applications, due to its high theoretical capacity (170 mA h g⁻¹), availability of inexpensive starting materials, its environmental friendliness, and excellent thermal stability [2–6]. However, the main limitation of LiFePO₄ has been its low electronic conductivity (∼10⁻⁹ S cm⁻¹), which leads to high impedance, low capacity, and low rate capability. Many researchers have been investigating on improving the performance and safety aspects of LiFePO₄. The methods of improving the performance including carbon coating [7–10], particle size reduction [11–13], and metal doping [14–16] have been proposed. Incorporation of conducting metal fluoride phases such as FeF₂ into LiFePO₄ is reported to be a viable method for enhancing the electronic conductivity of this material [14, 17, 18]. The presence of iron phosphide (Fe₂P) has also been found to increase the performance of Li-ion batteries [15].

Fe₂P is known to be a very good electronic conductor [15]. An olivine phosphate, particularly LiFePO₄, has been known to undergo carbothermal reduction at high temperatures to form Fe₂P and other metal phosphides. In presence of a suitable carbon-containing surfactant, at high temperatures, carbon reacts with LiFePO₄ to form CO₂ or CO, which leads to the formation of a two-phase system: Li₃PO₄/Fe₂P [19]. Thus, the conducting Fe₂P phase can form intrinsically in LiFePO₄. Fe₂P-incorporated LiFePO₄ has shown an improvement in electrochemical performance, as it provides a conducting network that enhances the electron transport through the material. It has been known that even a small amount of Fe₂P can enhance the electronic conductivity by 4 orders of magnitude [20]. However, excessive amount of this conducting phase may result in loss of capacity, as it forms at the expense of LiFePO₄, which is responsible for lithium insertion/de-insertion reactions [21]. There are several studies which have reported the effect of addition of Fe₂P on the electrical and electrochemical properties of LiFePO₄ [22–24]. Liu et al. [25] prepared LiFePO₄ by co-precipitation method using polyacrylamide as carbon source that provided a reducing environment for the formation of Fe₂P. They found Fe₂P impurity was formed at 750 °C, and LiFePO₄ with Fe₂P impurity exhibited better electronic conductivity and improved electrochemical performance. Although the amount of Fe₂P was not reported, they found the amount of Fe₂P increased with increasing calcination time. Rahman et al. [26] synthesized LiFePO₄-Fe₂P-C material using solvent-assisted solid-state reaction with various amounts of citric acid as a source of carbon. They found that the use of solvent assists in the formation of Fe₂P in the proximity of LiFePO₄ particles, which provides a percolating network resulting in a very high electronic conductivity. They claimed a very high capacity 136 mA h g⁻¹ at 10 C rate. Kim et al. [27] prepared LiFePO₄/Fe₂P composites by mechanical alloying in order to improve the electrical conductivity. The sample containing 8 % of Fe₂P in LiFePO₄/Fe₂P composite showed a high capacity ∼160 mA h g⁻¹ at rate of C/20, and ∼110 mA h g⁻¹ at a rate of 1 C with good cycleability. Rho et al. [28] studied the surface chemistry of four LiFePO₄ samples, prepared by heat treatment under increasing reducing environment, using Mössbauer and x-ray photoelectron spectroscopy. When the samples were treated for a longer period from 15 to 24 h in 7 % H₂/N₂ atmosphere at 600 °C, the authors found Fe₂P impurity phase between 4 and 18 wt%, along with Li₃PO₄ formed on the surface of the parent LiFePO₄. Their results suggest that an optimum amount, between 5 and 10 wt%, of Fe₂P gives rise to a greatly enhanced electrochemical performance of the composite. Another study reported that the presence of FeP is more favorable over Fe₂P, and indicated that 16 % of Fe₂P greatly lowered the discharge capacity [29].

From the review of the literature, it is clear that the presence of Fe₂P phase influences the performance of LiFePO₄ cathode. However, several questions still remain unanswered. It is important to know (1) how does the calcination temperature affect the amount of Fe₂P phase produced in a reducing environment? (2) what are the effects of the presence of Fe₂P impurity phase on diffusion of Li⁺ ions through LiFePO₄ and the electrochemical performance of LiFePO₄ cathode material? and (3) what is the optimum amount of Fe₂P required to obtain the best performance of LiFePO₄ cathode? To answer these questions, we have performed an extensive temperature-dependent study on LiFePO₄/C prepared by a simple and cost-effective sol-gel method followed by calcination in reducing Ar/H₂. We have synthesized samples at various temperatures between 600 and 900 °C and quantified the Fe₂P content (5–37 wt%) using magnetic and Mössbauer spectroscopy measurements. We found that the sample calcined at 700 °C yields an optimum amount of Fe₂P (15 wt%) phase that shows an enhanced capacity with high rate capability of 136 mA h g⁻¹ at rate of 1 C and 121 mA h g⁻¹ at 10 C. We further demonstrate that although the excessive amount of Fe₂P decreased the overall discharge capacity, all samples calcined at temperatures ≥700 °C have an order of magnitude higher Li⁺ diffusion coefficient (∼1.3 × 10⁻¹³ cm² s⁻¹) compared to the one calcined at 600 °C as well as the values reported in literature.

Experimental details

Synthesis

LiFePO₄/C samples were synthesized by the sol-gel method [29]. CH₃COOLi·2H₂O, 99 % (Alfa Aesar), FeCl₃ (Fisher Scientific), and P₂O₅ (Fisher Scientific) were used as precursors. FeCl₃ and P₂O₅ were dissolved in 200 proof ethanol in two separate beakers, then mixed and stirred for 3 h in a three-neck flask under constant flow of nitrogen. Then lithium acetate dissolved in ethanol was added to the solution and was allowed to stir for additional 3 h. Lauric acid dissolved in ethanol was added as a source of carbon and the solution was kept for 3 more hours of stirring under the same conditions and the resulting solution was dried at 90 °C to form a dry gel, which was subsequently calcined at 600, 700, 800, and 900 °C for 10 h under the flow of Ar/H₂ (90 %/10 %). Carbon content of the samples was measured by CHN elemental analyses, where the sample is combusted in a pure oxygen environment and found to be ∼4 % in the first three samples and ∼2 % in the fourth sample. Hereafter, these samples will be referred as LFP-600, LFP-700, LFP-800, and LFP-900, respectively.

Characterization

X-ray diffraction (XRD) measurements were performed using a Rigaku Minflex-600 diffractometer using Cu Kα (λ = 1.54 Å) x-rays to determine the phase purity of the samples. Electrical conductivity of the composite samples was measured by preparing pressed pellets and attaching electrodes using conducting silver epoxy in a Van der Pauw geometry. Magnetic measurements were done using a Physical Properties Measurement System (PPMS). Temperature-dependent ac magnetization measurements were done at 500 Oe and saturation magnetization (M) was measured by varying the magnetic field (H) from −5 to +5 T. The morphology of the samples was investigated using JSM-6510-LV-LGS SEM and JEOL 2010 TEM.

⁵⁷Fe Mössbauer spectroscopy

⁵⁷Fe Mössbauer spectra were recorded in the transmission geometry using both sides of a (Wissel) transducer coupled to ⁵⁷Co in Rh matrix source of about 50 mCi and 256 channels of a multichannel analyzer. The velocity calibration and the linearity verification were performed using a thin iron foil. For Mössbauer measurements, approximately 70 mg of the sample was uniformly distributed in a Teflon circular cell of 1.7 cm diameter. The isomer shift values are reported with reference to α-Fe foil. The spectra were least square fitted with MossWin program.

Electrochemical measurements

The electrochemical characterization of the samples was performed in a standard coin cell geometry with lithium metal as an active anode. The active cathode materials and Super P as a conducting material were mixed (95:5 ratio) and ground for 20 min (without a binder). The homogenous mixture was put on an aluminum mesh (current collector) and was then hand pressed between two steel cylinders [30]. The prepared cathode was cycled against Li metal electrode as a counter electrode separated by a polymeric separator soaked in a binary electrolyte consisting of ethylene carbonate (EC) and dimethyl carbonate (DMC) (50:50) containing 1 M LiPF₆. The room temperature galvanostatic charge and discharge measurements were carried out at different C-rates in the voltage range 2.2–4.2 V. Electrochemical impedance spectroscopy measurements were carried out using a Gamry electrochemical system in the frequency region of 0.01 Hz–100 kHz with AC amplitude 10 mV.

Results and discussion

Figure 1 shows the XRD patterns of all LiFePO₄ samples calcined at 600–900 °C for 10 h. All the observed Bragg peaks in the LFP-600 sample can be indexed to the olivine LiFPO₄ structure with a space group Pnma, which implies the absence of any detectable impurity phase. Additional peaks, which were indexed to iron phosphide (Fe₂P) and Li₃PO₄, are observed in the XRD patterns of samples annealed at 700 °C and higher, and the amount of these phases steadily increases with calcination temperature. These impurity phases result from carbothermal reduction of LiFePO₄ at higher temperatures in reducing environment provided by the lauric acid (carbon source) and the flowing forming gas. As the formation of Li₃PO₄ requires three Li ions, this phase does not grow as quickly as Fe₂P because dissociation of LiFePO₄ is not fast enough up to 700 °C. Only at calcination temperatures of >700 °C their amount begins to appear considerably according to following carbothermal and/or hydrogen reduction reactions [28]:

Fig. 1

XRD patterns of LiFePO₄ samples annealed for 10 h

$$ 6{\mathrm{Li}\mathrm{FePO}}_4+8\mathrm{C}\to 3{\mathrm{Fe}}_2\mathrm{P} + 2\ {\mathrm{Li}}_3{\mathrm{PO}}_4+\mathrm{P}\uparrow + 8{\mathrm{CO}}_2\uparrow $$

(1)

$$ 6{\mathrm{Li}\mathrm{FePO}}_4+16{\mathrm{H}}_2\to 3{\mathrm{Fe}}_2\mathrm{P} + 2\ {\mathrm{Li}}_3{\mathrm{PO}}_4+\mathrm{P}\uparrow + 8{\mathrm{CO}}_2\uparrow $$

(2)

In order to find the weight fraction of impurity phases, we have performed Rietveld refinement of XRD data of LFP-600, LFP-700, LFP-800, and LFP-900 samples. As an example, we show the Rietveld fitting of XRD data of LFP-700 sample in Fig. 2. The crystallite size of LiFePO₄, obtained from Rietveld fitting, showed an increase from ∼99 nm in LFP-600 to ∼125 nm in LFP-900, as expected. Further, Table 1 lists the weight percentage of LiFePO₄, Li₃PO₄, and Fe₂P in the samples studied in this work. We note that amount of Fe₂P increases with the calcination temperature, but the weight ratio between Fe₂P and Li₃PO₄ is less than the expected ratio of ∼2:1 (molar ratio 3:2) in first three samples, very similar to the observations in ref. [28], implying the presence of amorphous or nano-sized impurity phases. The room temperature electrical conductivity measured on the pressed pellets of the LFP-700, LFP-800, and LFP-900 samples showed 2 orders of magnitude higher conductivity (0.8–2.0 × 10⁻¹ S cm⁻¹) compared to that of LFP-600 sample (∼2 × 10⁻³ S cm⁻¹) which is attributed to the presence of conducting network of Fe₂P impurity phase.

Fig. 2

Rietveld fitting of XRD pattern of LFP-700 sample

Table 1 The weight percentage of LiFePO₄, Fe₂P, and Li₃PO₄ obtained from multiphase Rietveld refinement of XRD data

Figure 3 shows SEM images of the samples. The grain size of LFP-600 and LFP-700 samples appears to be very similar. Both of them contain grains of uniform size, estimated to be ∼100 nm, with some agglomeration. However, the samples calcined at higher temperatures (LFP-800 and LFP-900) show noticeably larger grain size and increased nonuniform size distribution (Fig. 3c, d). The EDXS measurements on the LFP-900 sample showed relatively more Fe-rich regions (see inset Fig. 3d), indicating the presence of more Fe₂P regions compared to LFP-600 sample (see inset Fig. 3a).

Fig. 3

SEM images of LiFePO₄ samples: a LFP-600; b LFP-700; c LFP-800; and d LFP-900. The scale bar is 1 μm. The insets in a and b show EDX spectra as described in the text

We further investigated the particle morphology and size distribution using TEM. Figure 4 shows the TEM images of LFP-700 sample at two different magnifications, revealing particles in the size range of ∼80–100 nm. The particles show a rough morphology due to decomposition of LiFePO₄ into Fe₂P and Li₃PO₄ on the surface. The arrow marks in Fig. 4a show the inter-particle regions consisting of possibly carbon, Fe₂P, and Li₃PO₄ very similar to the observations reported in ref. [26]. Under higher magnification, the particle surface further reveals the formation of sub-nanoregions of 2–4 nm (see Fig. 4b) of the decomposed products, which may not be detected by XRD but could influence the electrochemical properties of the composite material [26].

Fig. 4

TEM images of LFP-700 sample at two different magnifications. The arrow marks in a show inter-particle regions consisting of possibly carbon, Fe₂P, and Li₃PO₄; b magnified image of the circled region showing the sub-nano features on the surface of the particle

Since Fe₂P is magnetic with a first order ferromagnetic transition at ∼220 °C [31, 32], we measured the magnetic properties to confirm its presence in the samples. Figure 5a shows the zero field cooled (ZFC) temperature dependence of magnetization at an applied field of 500 Oe and Fig. 5b shows the measured hysteresis loops of magnetization vs applied field for the samples. Although Fe₂P is not detectable in the XRD (Table 1), its first order ferromagnetic transition at ∼220 °C is seen in all the samples, albeit it is rather weak in LFP-600, whereas it is quite clear in samples calcined at higher temperatures due to the presence of considerable amount of Fe₂P. Thus, the magnetization measurements confirm the presence of Fe₂P in all the samples and demonstrate an increase in the amount of Fe₂P with increasing calcination temperature (see Fig. 5b). Although the magnetization is not saturated at an applied field of ±10 kOe, the maximum magnetization achieved in each sample was compared with the saturation magnetization value, 32.1 emu g⁻¹, observed for nanocrystallites of Fe₂P reported in literature [33]. Considering this value as 100 %, the estimated amount of Fe₂P in LFP-600, LFP-700, LFP-800, and LFP-900 samples is 6, 12, 28, and 56 %, respectively. As a result, the amount of active material for Li insertion/de-insertion is substantially reduced in LFP-800 and LFP-900, which is expected to lower the overall specific capacity.

Fig. 5

a Magnetization vs. temperature of LiFePO₄ samples; b hysteresis loops of LiFePO₄ samples measured at 10 K

To further confirm the amount of Fe₂P, as determined by magnetization measurements, Mössbauer spectra (see Fig. 6) of the samples were recorded. Summary of the Mössbauer parameters is given in Table 2. The ⁵⁷Fe Mössbauer spectra of LiFePO₄ with Fe₂P impurity consists of three quadrupole doublets. First, a symmetric and dominant doublet with an isomer shift ∼1.22 mm/s and a quadrupole splitting of ∼2.97 mm/s arises due to high spin configuration of 3d electrons in Fe²⁺ ion and charge asymmetry around Fe in parent LiFePO₄ [28, 34]. The second doublet with an isomer shift of 0.61 mm/s and a quadrupole splitting of 0.43 mm/s is due to Fe³⁺ occupying 3f site in Fe₂P and the third doublet with an isomer shift of 0.19 mm/s and a quadrupole splitting of 0.1 mm/s is due to Fe³⁺ at the pyramidal 3 g site for Fe₂P structure [28, 35]. In the present study, we did not observe any FeP phase as reported in ref. [28]. It is reasonable to assume that the FeP phase may have transformed into Fe₂P phase due to higher annealing temperatures and reducing environment. A very small signature of one of the two doublets for Fe₂P in LFP-600 indicates an insignificant amount of Fe₂P occupying Fe-3f site only. We have estimated the amount of Fe²⁺ and Fe³⁺ from the relative area under the respective peaks in the Mössbauer spectra. The percentage of Fe²⁺ and Fe³⁺ phases in the samples is given in Table 2, and corresponding mol% and wt% of LiFePO₄, Fe₂P, and Li₃PO₄, calculated using Eq. 1, are listed in Table 3. It is interesting that the total amount of Fe₂P calculated from the Mössbauer spectroscopy (Table 3) and magnetic measurements are close to each other in first three samples, but differ from those estimated by Rietveld refinement of XRD data (Table 1). As explained in a previous section, this is due to the presence of amorphous or sub-nanoregions of Fe₂P and Li₃PO₄ as shown in TEM images (Fig. 4).

Fig. 6

Mössbauer spectra of LiFePO₄ samples measured at room temperature

Table 2 Mössbauer parameters of the LiFePO₄ samples

Table 3 Percentage of LiFePO₄, Fe₂P, and Li₃PO₄in LFP samples deduced from Mössbauer measurements

The galvanostatic charge–discharge voltage profiles of the samples measured at a rate of 1 C are shown in Fig. 7. The observed capacity of 136, 102, and 78 mA h g⁻¹, for LFP-700, LFP-800, and LFP-900 samples, respectively, are very close to the expected capacity calculated using the wt% of LiFeO₄ (Table 2) deduced from Mössbauer measurements. However, LFP-600 sample shows a significantly lower capacity of 120 mA h g⁻¹ compared to the expected capacity of 158 mA h g⁻¹. All the samples show a potential plateau at ∼3.4 V corresponding to Fe²⁺/Fe³⁺ redox couple. The potential plateau of LFP-700 is wider compared to the other samples, which indicates that Li can be inserted and de-inserted more efficiently.

Fig. 7

Charge–discharge profiles of LiFePO₄ samples measured at a rate of 1 C

Figure 8a shows the impedance spectra (Nyquist plots) of the samples. The spectra display typical characteristics of an electrochemical cell consisting of a small intercept in the higher frequency region, a depressed semicircle in the intermediate frequency region, and an inclined line in the low-frequency region. The small intercept corresponds to the ohmic resistance, representing the resistance of the electrolyte. The depressed semicircle is related to the charge transfer resistance and the double layer capacitance between the electrolyte and cathode. The inclined line is related to the frequency-dependent Warburg impedance associated with Li⁺ ion diffusion in the cathode active particles. The values for the charge transfer resistance (R _ct) extracted from fitting the impedance data as described below are listed in Table 4. LFP-700 exhibits the lowest charge transfer resistance and hence highest specific capacity suggesting that the amount of Fe₂P present in this sample may be the optimum amount to provide highly conducting network, which enhances the electron transport in the sample. Although the samples LFP-800 and LFP-900, calcined at 800 and 900 °C, have lower charge transfer resistance compared to the LFP-600 sample, their specific capacity is lower due to the presence of less active material for Li intercalation and deintercalation process.

Fig. 8

a Nyquist plots of LiFePO₄ samples; b a plot of the Z′ vs. ω ^−1/2 in the low-frequency region

Table 4 Electrochemical impedance parameters and the exchange current density of the LiFePO₄ samples

To further understand the above behavior, the diffusion coefficient of lithium ion (D _Li) was determined by the Z′ dependence on ω in the low-frequency region, which is described by [36],

$$ {Z}^{\prime }=\kern0.24em {R}_{\mathrm{s}}+{R}_{\mathrm{ct}}+\sigma {\omega}^{-1/2} $$

(2)

where, σ is the Warburg coefficient and R _s and R _ct are the solution and the charge transfer resistances. σ is related to D _Li by,

$$ {D}_{\mathrm{Li}}=\kern0.24em {R}^2{T}^2/2{A}^2{n}^4{F}^4{C}_{\mathrm{Li}}^2{\sigma}^2s $$

(3)

where, R is the gas constant, T is the absolute temperature, A is the surface area of the cathode (0.28 cm² in our case), n is the number of electrons per molecule during oxidation, F is the Faraday constant, and C _Li is the concentration of lithium ion (0.0228 mol/cm³). As expected, a plot of Z′ versus ω ^−1/2 (see Fig. 7b) shows linear behavior with values of slope σ. The diffusion coefficient and apparent exchange current density (I _o) [36],

$$ {I}_o=RT/n{R}_{\mathrm{ct}}F $$

(4)

along with other relevant parameters for the samples are given in Table 4. The LFP-600 sample has the lowest diffusion coefficient (3.5 × 10⁻¹⁴ cm² s⁻¹) while LFP-700, LFP-800, and LFP-900 samples have similar value, (1.3 ± 0.1) × 10⁻¹³ cm² s⁻¹, which is about an order of magnitude higher than that of LFP-600 and values reported in the literature [37–39]. Of four samples studied in this work, the LFP-700 sample has the least R _ct and the highest exchange current density compared to other carbon-coated LiFePO₄ samples reported in literature [40], perhaps due to the presence of conducting nanoregions of Fe₂P on the surface LiFePO₄ particles. This result corroborates with an earlier study on LiFePO₄/Fe₂P-C composite cathode [26].

Figure 9 shows the discharge capacity y of the LiFePO₄ samples at various rates from 1 to 10 C. Clearly, the LFP-700 sample not only shows the best capacity at 1 C rate but also retains more than 70 % of its capacity when tested at higher rate of 10 C (faster charging and discharging). This sample also retains nearly 100 % of its initial capacity even after 350 cycles at a rate of 5 C (see inset in Fig. 9). We attribute the enhanced electrochemical performance of LFP-700 to faster kinetics of the cell reactions (higher diffusion constant) along with higher catalytic activity (higher exchange current density) compared to other samples. We have also calculated the characteristic diffusion length, ${L}_{\max, \mathrm{L}\mathrm{i}}=\sqrt{\tau\;D{}_{\mathrm{Li}}}$ [41], where τ is the diffusion time constant. It has been shown that at higher C-rate (faster charging/discharging), the characteristic diffusion length decreases with diffusion time constant (taken as the discharging time) [38], and the particle size comparable to the characteristic diffusion length is indicative of better electrochemical performance at higher charge/discharge rates. The estimated characteristic diffusion length for LiFePO₄ samples annealed at higher temperatures, using diffusion time constant equivalent to the discharging time for 10 C rate, to be ≈ 70 nm. This means we can obtain maximum possible capacity from the sample, after accounting for inactive components, at lower charging/discharging rate of 1 C. This result is in agreement with our observations (see Table 3).

Fig. 9

Capacity of LiFePO₄ samples during continuous cycling at different charging rates. The inset shows the capacity retention for LFP-700 sample at 5 C rate

Conclusions

In summary, we have prepared several LiFePO₄/C samples calcined at different temperatures between 600 and 900 °C and studied their structural, magnetic, and electrochemical properties. The impurity phase Fe₂P was analyzed by XRD, magnetic measurements, and Mössbauer spectroscopy. We found that the amount of the impurity phase strongly depends on the calcination temperature. LiFePO₄/C samples calcined at 600 and 700 °C consist of LiFePO₄ particles with size ∼90–100 nm and contain 5 % and 16 wt% Fe₂P phase. The amount of Fe₂P grows considerably in the samples calcined at 800 and 900 °C. Comparing all the properties of the LiFePO₄/C samples, the sample calcined at 700 °C exhibits better electrochemical performance with a capacity of 136 mA h g⁻¹ at 1 C, 121 mA h g⁻¹ at 10 C (∼70 % of theoretical capacity of LiFePO₄), and excellent cycleability. With ∼15 wt% of Fe₂P, it exhibits the least charge transfer resistance of 43 Ω with an electronic conductivity of ∼10⁻¹ s cm⁻¹ and Li⁺ diffusion coefficient of ∼1.3 × 10⁻¹³ cm² s⁻¹. This indicates the importance of controlling both the reduction environment and temperature during the synthesis process to produce the proper amount of Fe₂P phase necessary for enhancing the electrochemical performance of LiFePO₄/C nanocomposites.