Lithium-ion batteries (LIBs) are indispensable energy storage devices widely integrated into modern life. Among the key components of LIBs, the cathode plays a pivotal role in determining the energy density, rate capability, and cycle life of the cells while also representing the largest fraction of the overall cost of the battery [1–6]. Among various cathode materials, lithium iron phosphate (LiFePO₄, LFP) has garnered significant attention in recent years owing to its unique properties and growing applications [7–11]. This interest has been further fueled by the growing popularity of electric vehicles (EVs), where the relatively low-cost LFP cathode is increasingly used as an alternative to the more expensive nickel-cobalt-manganese (NCM) cathodes traditionally used in EVs. A key advantage of LFP lies in its exceptional structural stability, attributed to the strong P–O bonds within the (PO₄)^3– polyanion, which are more robust than the metal–oxygen bonds in conventional NCM cathodes [7–11]. This stability prevents the release of oxygen even under fully charged conditions or high-temperature environments, endowing LFP with exceptional cycling performance and superior safety characteristics compared with NCM cathodes [12–15]. These properties make LFP an attractive choice, particularly for EVs and stationary energy storage systems, where safety and durability are paramount. However, LFP also exhibits notable drawbacks, including a relatively lower energy density and suboptimal rate capability. Electrons in the LiFePO₄ structure are transported through the Fe–O–Fe pathway within the FeO₆ octahedra. However, owing to the corner-sharing arrangement of these FeO₆ octahedra, which are separated by oxygen atoms in the (PO₄)^3– tetrahedra, electron movement occurs in a zigzag pattern, significantly limiting the electronic conductivity of the material [7–11]. Furthermore, lithium-ion transport in LiFePO₄ is confined to one-dimensional channels, inherently restricting Li⁺ mobility compared with cathode materials with two- or three-dimensional conduction pathways [7–11]. The combination of low electronic conductivity and constrained ionic diffusion underpins the poor rate capability of LiFePO₄, posing challenges for its performance in high-power applications.

The surface coating of LFP cathodes with carbon has been widely adopted for enhancing their electrochemical performance [16–20]. The high electronic conductivity of the carbon coating facilitates efficient electron transfer into the LFP particles, significantly improving the rate capability of the material. Various organic materials have been employed as carbon sources to fabricate the coating layer [21–25], and in some cases, carbon is directly applied in its elemental form [26–28]. Despite these advances, achieving an optimal carbon coating with superior properties remains an area of active research and continuous improvement. To address this limitation, this study aims to investigate the combined effects of glucose, polystyrene, and ferrocene on the structure and electrochemical performance of carbon-coated LiFePO₄, with the goal of optimizing carbon coatings derived from mixed organic sources. To characterize the resulting carbon coating layer, Raman spectroscopy and transmission electron microscopy (TEM) were used to analyze its structural and morphological properties. Furthermore, the electrochemical performance of the coated LFP cathodes was systematically evaluated through several performance tests, including impedance analysis and rate capability measurements. This comprehensive approach aims to provide insights into achieving a more effective carbon layer for LFP cathodes.

In this study, LiFePO₄/C composite powders were synthesized via a solid-state reaction using FePO₄, supplied by ECOPRO BM, as the precursor. The precursor was thoroughly mixed with Li₂CO₃ (99.9%, Aldrich), a lithium source. Glucose (99.5%, Aldrich) and polystyrene (Aldrich) served as carbon sources, whereas ferrocene (98%, Aldrich) was introduced as an additive to enhance the quality of the carbon coating. When glucose and polystyrene were simultaneously used as carbon sources, the total weight of the carbon source was held constant, while their ratio was systematically varied, and ferrocene was introduced as a supplementary additive. The precursor and Li₂CO₃ were combined in a molar ratio of Fe:Li=1:1.05 and subjected to high-energy planetary ball milling (Pulverisette 6, Fritsch) at 300 rpm for 3 h. To prevent overheating, the milling process was conducted in 15-min intervals, followed by 10-min rest periods, repeated 12 times. The milled powder was subsequently sieved through a 300-mesh screen and thermally treated in a quartz tube furnace under an argon atmosphere. The heat treatment was performed in two stages: an initial step at 350°C for 2 h, followed by a second stage at 700°C for 6 h, with a controlled heating rate of 2°C·min^–1. The morphology of the resulting carbon coating was characterized using TEM (JEM-2100F, Cs corrector). The structural properties of the carbon were analyzed using Raman spectroscopy (Jobin Yvon, LabRAM HR Evolution), with the intensity ratio of the disorder-induced D-band (1342.8 cm^–1) to the graphitic G-band (1594.0 cm^–1) used to assess the degree of graphitization.

For electrochemical characterization, cathodes were prepared by blending the active material with Super-P, carbon nanotubes, and polyvinylidene fluoride in a precise weight ratio of 80:10:5:5, using anhydrous N-methyl-2-pyrrolidone (Aldrich) as the solvent. The resulting slurry was uniformly coated onto an aluminum foil, which functioned as the current collector, and subsequently dried under vacuum at 90°C for 20 h. The mass loading of the electrode was approximately 5 mg·cm^–2. The electrochemical performance was evaluated using 2032-type coin cells, assembled in an Ar-filled glove box with a lithium metal anode, the prepared cathode, a polypropylene separator (Celgard 2400), and an electrolyte composed of 1M LiPF₆ dissolved in EC/DMC (1:2 vol.%, Enchem Co., Ltd.). Galvanostatic cycling tests were conducted using a WonATech voltammetry system over a voltage range of 2.5–4.2 V. The cells were tested at the current densities of 7.5, 15, 45, 75, and 150 mA·g^–1, corresponding to the approximate C-rates of 0.05, 0.1, 0.3, 0.5, and 1, respectively. All the electrochemical tests were performed at a controlled temperature of 30°C. Electrochemical impedance spectroscopy measurements were conducted after the first discharge cycle at a current density of 150 mA·g^–1, using an AMETEK VersaSTAT3 electrochemical workstation. The impedance spectra were acquired over a frequency range of 0.1 Hz to 100 kHz with an AC voltage amplitude of 10 mV and subsequently fitted using ZSimpWin 3.60 software. The galvanostatic intermittent titration technique (GITT) was employed to further evaluate the Li-ion diffusion characteristics. The GITT measurements were performed by applying a 0.1C pulse current for 10-min intervals, followed by an open-circuit relaxation period. The Li-ion diffusion coefficient (D_Li+) was calculated using the following equation:

where  denotes the pulse duration (600 s), S represents the contact area between the electrolyte and the active material, n_m is the molar amount of the active material, V_m denotes the molar volume of the active material, △E_t represents the transient voltage change, and △E_s corresponds to the steady-state voltage change.

First, we investigated the electrochemical properties of LFP cathodes with a carbon coating formed using glucose as the source material. Fig. 1a illustrates the initial charge–discharge profiles of pristine LFP (without carbon coating) and samples with carbon coatings prepared using 5, 7, 9, and 11 wt.% glucose (denoted as GC 5, 7, 9, and 11 wt.%, respectively) relative to the cathode weight. The pristine LFP exhibited a discharge capacity of approximately 134 mAh·g^–1 at a rate of 0.05C. However, the LFP samples with a glucose-derived carbon coating of 7–11 wt.% demonstrated a significantly enhanced discharge capacity of approximately 155–156 mAh·g^–1 under the same conditions. This enhancement became even more pronounced at higher C rates, as shown in Fig. 1b. For the pristine LFP, the discharge capacity decreased to ~58 mAh·g^–1 at a rate of 1C, with a capacity retention (1C discharge capacity compared with 0.05C discharge capacity) of only 43%. In contrast, as the glucose content used for carbon coating increased, the discharge capacity and capacity retention at 1C improved markedly. Specifically, the sample with 11 wt.% glucose exhibited a discharge capacity of ~132 mAh·g^–1 at 1C, corresponding to a capacity retention of ~85%. The discharge capacities at 0.05C, 0.1C, 0.3C, 0.5C, and 1C rates, as well as the capacity retentions, are summarized in Table 1.

While increasing the glucose content yielded a robust carbon coating and enhanced the electrochemical performance, the objective was to optimize the coating while minimizing the amount of carbon source material required. To achieve this, we introduced polystyrene as a secondary carbon source, mixing it with glucose. By fixing the total carbon source at 7 wt.% and varying the glucose-to-polystyrene ratios to 3:1, 1:1, and 1:3 (denoted as GCPS 3:1, 1:1 and 1:3, respectively), we examined the effect on the electrochemical performance. As shown in Fig. 1c, the discharge capacity at a rate of 0.05C did not vary significantly with the addition of polystyrene. However, under higher C rates (Fig. 1d), a substantial improvement in discharge capacity and retention was observed for the samples containing polystyrene. For example, even with the smallest polystyrene ratio (glucose:polystyrene = 3:1), the capacity retention increased to 87%, surpassing the performance achieved using 11 wt.% glucose alone. These results demonstrate that mixing polystyrene with glucose enables a more efficient formation of the carbon coating, resulting in a superior rate performance with reduced carbon source usage. The discharge capacities and capacity retentions of the mixed-source samples are summarized in Table 2.

To investigate the effects of incorporating polystyrene as a carbon source, we performed a detailed analysis of the pristine LFP, GC 7 wt.%, and GCPS 3:1 samples using TEM, carbon-sulfur (CS) analyses, and Raman spectroscopy. The sample with a higher proportion of polystyrene (e.g., GCPS 1:3) exhibited marginally superior electrochemical performance. However, the GCPS 3:1 sample was selected for further analysis to reduce reliance on the relatively expensive polystyrene in favor of the more cost-effective glucose, thereby achieving a balance between performance and material efficiency. The TEM images in Fig. 2a–c reveal the differences in the surface morphology and thickness of the carbon coatings. The pristine LFP surface appeared clean and devoid of any coating layer, whereas both carbon-coated samples exhibited well-defined surface layers. For GC 7 wt.%, the coating thickness was approximately 2.5 nm, whereas the GCPS 3:1 sample demonstrated a slightly thicker coating layer of ~3.2 nm. The CS analysis results further validated these observations by quantifying the carbon residue on the cathode. For the GC 7 wt.% sample, the carbon content was measured to be ~0.84%, whereas the GCPS 3:1 sample displayed a significantly higher carbon residue of ~1.77%. This suggests that the inclusion of polystyrene enables greater carbon retention during the carbonization process. The reason for this can be attributed to the chemical composition of the precursors: glucose, with the formula C₆H₁₂O₆, has a carbon weight proportion of approximately 40%, whereas polystyrene, with the formula [C₈H₈]_n, consists of ~92% carbon by weight. Consequently, polystyrene provides a greater yield of carbon residue, which directly contributes to the improved electrochemical performance of the GCPS 3:1 sample.

However, the properties of the carbon coating are influenced not only by the amount of carbon residue but also by its crystallinity. Crystalline carbon, such as graphite, exhibits superior electronic conductivity compared with amorphous carbon. Therefore, the degree of graphitization of the carbon coating layer plays a crucial role in determining its electronic conductivity and overall performance. Raman spectroscopy was employed to analyze the crystallinity of the carbon layers in the GC 7 wt.% and GCPS 3:1 samples. The Raman spectra, shown in Fig. 2d, reveal two characteristic peaks: the D band at ~1342.8 cm^–1, associated with disordered carbon, and the G band at ~1594.0 cm^–1, associated with graphitic (ordered) carbon. The intensity ratio of these bands, the I_D/I_G (intensity ratio of D and G bands) ratio , provides a reliable measure of graphitization. For the GC 7 wt.% sample, the I_D/I_G ratio was 0.871, whereas for the GCPS 3:1 sample, the ratio increased to 0.948. This indicates that the incorporation of polystyrene leads to an increase in disordered carbon and a corresponding decrease in the graphitization of the carbon coating. Therefore, although the addition of polystyrene significantly increases the carbon residue, thereby improving the electrochemical properties of LFP, it also reduces the crystallinity of the carbon coating. This reduction in crystallinity results in lower electronic conductivity, diminishing the effectiveness of the carbon per unit amount.

Additives, such as ferrocene, can be effectively used in combination with carbon sources to enhance the crystallinity of the carbon coating layer [29–31]. Fe ions, contained in ferrocene, have been reported to serve as a favorable catalyst for promoting carbon crystallization at low temperatures [32–34]. Fig. 3 illustrates the electrochemical properties of the samples where ferrocene was added as an additive during the carbon coating process. When ferrocene was incorporated into the GC 7 wt.% sample, both the initial discharge capacity and rate capability improved, as shown in Fig. 3a and 3b. Specifically, the addition of 0.5 wt.% ferrocene resulted in the highest discharge capacity of approximately 165 mAh·g^–1 and a capacity retention exceeding 80%. These results suggest that 0.5 wt.% is the optimal amount of ferrocene, as adding less than this amount led to insufficient improvement, whereas adding 0.7 wt.% offered no additional benefits over the 0.5 wt.% sample. Thus, 0.5 wt.% ferrocene was applied to the GCPS 3:1 sample, which had already demonstrated superior electrochemical properties than GC 7 wt.% in earlier experiments. As shown in Fig. 3c and 3d, the addition of ferrocene further enhanced the performance of the GCPS 3:1 sample. The initial discharge capacity at a rate of 0.05C increased to ~166 mAh·g^–1, and the capacity retention reached an impressive 90%, significantly higher than that of the ferrocene-free GCPS 3:1 sample. Table 3 provides a detailed comparison of the discharge capacities and capacity retentions for the samples with and without ferrocene, further illustrating the beneficial impact of the additive.

TEM and Raman spectroscopy were employed to analyze the carbon coating layers to understand the mechanism behind the improved performance. As shown in Fig. 4a and 4b, the TEM images of the GC 7 wt.% and GCPS 3:1 samples prepared using ferrocene (0.5 wt.%) revealed no significant difference in the thickness of the carbon coating layers compared with those of the GC 7 wt.% and GCPS 3:1 samples without ferrocene (shown in Fig. 2b, c). This indicates that ferrocene barely increased the amount of carbon residue. However, a notable difference was observed in the Raman spectra: the intensity of the G band increased in the ferrocene-applied samples compared with those without ferrocene. The I_D/I_G ratios decreased from 0.871 to 0.854 for the GC 7 wt.% sample with ferrocene and from 0.948 to 0.901 for the GCPS 3:1 sample with ferrocene, indicating that ferrocene enhanced the graphitization of the carbon coating layer and reduced the proportion of disordered carbon. This improved graphitization is directly responsible for the enhanced electronic conductivity of the carbon coating, which, in turn, contributed to the superior electrochemical performance of the ferrocene-applied samples.

Impedance measurements were performed on cells containing the pristine LFP, GC 7 wt.%, GC 7 wt.% with ferrocene, GCPS 3:1, and GCPS 3:1 with ferrocene samples to evaluate the combined effects of carbon coating and ferrocene application on improving the electrochemical performance of LFP. Fig. 5 shows the Nyquist plots of these cells after the first cycle, with the semicircle sizes in the plots indicating the impedance resistance. As shown in Fig. 5, the semicircle size of GC 7 wt.% and GCPS 3:1 was significantly smaller than that of pristine LFP, indicating that the carbon coating effectively reduced the impedance resistance. This demonstrates the role of carbon coating in enhancing the charge transfer process by improving the conductivity of the electrode material. Furthermore, the semicircle sizes of the samples with ferrocene (Fig. 5d, e) were somewhat smaller than those of the samples without ferrocene (Fig. 5b, c). This reduction in impedance highlights the positive effect of ferrocene addition, which promotes graphitization during the material preparation process. Graphitization leads to improved electronic conductivity, as evidenced by the reduced impedance values. For a quantitative understanding, the Nyquist plots were fitted using an equivalent circuit model (Fig. 5f), which includes bulk resistance (R_b), solid-electrolyte interface layer resistance (R_SEI), charge transfer resistance (R_CT), and Warburg impedance (W) [35,36]. The fitted impedance values are summarized in Table 4, with the charge transfer resistance (R_CT) showing the most prominent differences among the samples.

The pristine LFP exhibited a high R_CT of ~112 Ω, whereas the carbon-coated GC 7 wt.% showed a significantly reduced value of ~48 Ω. An even lower R_CT of ~40 Ω was observed for the GCPS 3:1 sample, which can be attributed to the addition of polystyrene. The incorporation of polystyrene improves the electronic properties of the electrode, thereby enhancing its electrochemical performance, as demonstrated in Fig. 1c, d. When ferrocene was added, the R_CT values were further reduced, reaching ~38 Ω for GC 7 wt.% with ferrocene and ~30 Ω for GCPS 3:1 with ferrocene. These results confirm the effect of ferrocene on the impedance resistance, which contribute to the improved discharge capacity and rate capability observed in Fig. 3.

To investigate the influence of carbon coating on Li-ion diffusion, the Li-ion diffusion coefficient (D_Li+) was measured using the GITT method after 100 charge–discharge cycles and subsequently compared. As illustrated in Fig. 6 and summarized in Table 5, the application of a carbon coating enhanced the D_Li+ values. The pristine LFP exhibited a D_Li+ range of 9.222×10^–16–8.367×10^–11, whereas the introduction of a 7 wt.% glucose-derived carbon coating (GC 7 wt.%) increased D_Li+ slightly to 3.427×10^–15–3.533×10^–11. A further enhancement of D_Li+ to 2.228×10^–14–1.245×10^–10 was observed with GCPS 3:1. These results confirm that carbon coating facilitates Li-ion diffusion, with polystyrene supplementation proving even more effective. Although the samples incorporating ferrocene exhibited superior D_Li+ values compared with pristine LFP, the direct contribution of ferrocene to the enhancement of Li-ion diffusivity remained ambiguous. This suggests that an increased graphite ratio does not necessarily result in a significant improvement in Li-ion diffusion. Instead, the enhanced electrochemical performance associated with ferrocene addition is likely attributable to the increased electronic conductivity originating from a higher graphite content. Fig. 7 provides a schematic highlighting the advantages of utilizing a mixed carbon source comprising polystyrene and ferrocene. The illustration shows that the combination of polystyrene and ferrocene results in increased residual carbon content and enhanced graphitization, which together contribute to the improved electronic conductivity, ultimately enhancing the electrochemical performance of LFP.

This paper introduced an efficient carbon coating strategy for LFP, a prominent cathode material for secondary batteries. By incorporating polystyrene as part of the carbon source rather than relying solely on glucose, significant improvements in discharge capacity and rate capability were achieved. TEM and carbon-sulfur analyses revealed that the amount of carbon residue was greater when polystyrene was included, which can be attributed to the high carbon content in the chemical structure of polystyrene. However, the use of polystyrene resulted in reduced graphitization, lowering the crystallinity of the carbon layer and potentially diminishing its electronic conductivity. To address this, ferrocene was employed as an additive to the carbon source, effectively enhancing the graphitization of the carbon coating layer and reducing the proportion of disordered carbon. This optimization yielded an impressive discharge capacity of 166 mAh·g^–1, along with a high capacity retention of 90% (calculated as the 1C discharge capacity relative to the 0.05C discharge capacity). Further analysis using impedance spectroscopy demonstrated that ferrocene significantly reduced the interfacial resistance. These findings underscore the potential of this method to optimize the carbon coating of LFP cathodes, achieving enhanced electrochemical performance and advancing the development of high-performance lithium-ion batteries.