Study of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ as an Anode in Li-Ion Battery

Abstract

The application of oxygen-deficient perovskites (ODPs) has attracted interest as anode materials for lithium-ion batteries for their unique properties. One such material, CaSrFe_0.75Co_0.75Mn_0.5O_6-δ, has been studied extensively. The structure of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ was investigated using various techniques, including Rietveld refinements with X-ray diffraction and neutron diffraction. Additionally, iodometric titration and X-ray photoelectron spectroscopy were employed to study the oxygen-deficiency amount and the transition metal’s oxidation states in the material. As an anode material, CaSrFe_0.75Co_0.75Mn_0.5O_6-δ exhibits promising performance. It delivers 393 mAhg⁻¹ of discharge capacity at a current density of 25 mAg⁻¹ after 100 cycles. Notably, this capacity surpasses both the theoretical graphite anode capacity (372 mAhg⁻¹) and that of the calcium analog reported previously. Furthermore, the electrochemical performance of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ remains highly reversible across various current densities ranging from 25 to 500 mAg⁻¹. This suggests the material’s excellent stability and reversibility during charge–discharge cycles, showing its probable application as an anode for lithium-ion batteries. The mechanism of lithium intercalation and deintercalation within CaSrFe_0.75Co_0.75Mn_0.5O_6-δ has also been discussed. Understanding this mechanism is crucial for optimizing the battery’s performance and ensuring long-term stability. Overall, this study highlights the significant potential of oxygen-deficient perovskites, particularly CaSrFe_0.75Co_0.75Mn_0.5O_6-δ, for applications as an anode material for lithium-ion batteries, offering enhanced capacity and stability compared with traditional graphite-based anodes.

1. Introduction

Lithium-ion batteries (LIBs) are integral to modern technologies, powering everything from smartphones to electric vehicles. The performance of LIBs depends largely on the efficiency and stability of their electrodes and electrolytes [1]. The anode plays a crucial role in LIBs, as it plays a role in releasing and storing lithium ions at the time of charge and discharge cycles. Therefore, improving the anode’s properties is essential for enhancing battery performance and longevity [2,3,4,5].

Graphite has been the primary anode material used commercially in LIBs because of its ability to intercalate lithium ions, forming LiC₆ and providing a theoretical specific capacity of 372 mAhg⁻¹. However, practical capacities are typically lower, around 300–320 mAhg⁻¹ [1,4,6,7]. Despite its widespread use, graphite anodes have shortcomings, such as a solid electrolyte interphase (SEI) formation during initial charge and discharge cycles, which reduces efficiency and cycle life [6,8,9].

To overcome these limitations, extensive research has focused on developing alternative anode materials [6,10]. Tin-based materials, including alloys and oxides, refs. [7,10,11,12,13,14] have shown promise because of their high capacity for lithium storage. During charge and discharge cycles, tin-based anodes undergo reversible formation and decomposition of Li-Sn alloys, contributing to their high capacity [7,10,11,12,13,14].

There are studies of transition metal oxides for their potential application as anode materials. Although these materials generally do not create alloys or intercalate lithium, their reversible capacity is attributed to the metal’s oxidation state changing reversibly, along with the formation and breakdown of lithium oxide [10,15]. Various oxide materials, including perovskite and spinel phases, have been investigated for alternative sources of anodes [6,11,12]. Spinel-type materials, such as ZnCo₂O₄, exhibit reversible capacity through lithium intercalation into the spinel structure and subsequent decomposition, followed by alloying with zinc [6,16,17].

One particularly intriguing family of oxides being explored for anode applications is oxygen-deficient perovskites [2,18]. These materials exhibit interesting electrical properties, primarily because of oxygen vacancy presence, which enhances charge transport properties and ionic conductivity [19]. ODPs have been extensively explored in other applications, such as catalytic behavior [20], fuel cells, etc. [21], but their potential as anode materials for LIBs has been relatively understudied until recently. In recent research efforts, oxygen-deficient perovskites such as Ca₂Fe₂O₅ and Ca₂Co₂O₅, with brownmillerite-type structures, and Sr₂Fe₂O_6-δ with a unique structure of tetragonal geometry consisting of alternate octahedra and square pyramids have shown promise as efficient LIB anodes [2,18]. In these materials, the mechanism of electrode operation involves crystal lattice destruction to form nanoparticles of transition metals in an amorphous matrix of lithium and calcium oxides. A reversible reaction between the amorphous matrix and nanoparticles contributes to the observed capacity, while calcium ions aid in stabilizing the matrix.

In the current study, we focus on CaSrFe_0.75Co_0.75Mn_0.5O_6-δ, a multi-doped oxygen-deficient perovskite with a cubic structure, to address critical gaps in perovskite-based LIB anodes. While prior work has explored simpler compositions (e.g., Ca₂Fe₂O₅ and Sr₂Fe₂O_6-δ), the role of co-doping (Fe/Co/Mn) in stabilizing the amorphous Li–oxide matrix and enhancing reversible capacity remains unclear. Here, we aim to (1) Determine how multi-cation doping influences electrochemical performance compared with single-metal perovskites and (2) Evaluate whether this composition offers superior cycling stability over existing ODP anodes. Our findings provide new insights into designing high-performance perovskite anodes for long-life LIBs.

2. Experimental

The double perovskite CaSrFe_0.75Co_0.75Mn_0.5O_6-δ was synthesized via a conventional solid-state reaction process. Highly pure powders—CaCO₃ (99.99% Sigma Aldrich, St. Louis, MO, USA), SrCO₃ (99.99% Sigma Aldrich, USA), Fe₂O₃ (99.99% Sigma Aldrich, USA), Co₃O₄ (99.99% Sigma Aldrich, USA), and Mn₂O₃ (99.99% Sigma Aldrich, USA)—were weighed in stoichiometric proportions and uniformly homogenized using an agate mortar and pestle. The mixed powders were then compacted into pellets under a uniaxial pressure of 19 MPa (equivalent to 3 tons of force). These pellets were subjected to calcination at 1000 °C for 24 h in a muffle furnace. After cooling to ambient temperature, the calcined pellets were reground, repelletized, and sintered at 1200 °C for an additional 24 h. A controlled heating rate of 100 °C/h was maintained during both calcination and sintering to ensure phase purity and homogeneity.

The crystal structure and phase purity of the synthesized material were analyzed using powder X-ray diffraction (PXRD) with Cu Kα₁ and Kα₂ radiation. Rietveld refinement of XRD data was performed using the GSAS (I) software [22] with the EXPEGUI interface [23] for structural modeling and fitting. To examine the surface morphology and microstructure, scanning electron microscopy (SEM) was employed, providing high-resolution micrographs of the sample. The oxidation states of the B site cations are analyzed using X-ray photoelectron spectroscopy (XPS) with Al Kα radiation at 1486.7 eV. Iodometric titrations [24], a widely utilized method, was employed to assess oxygen vacancies in CaSrFe_0.75Co_0.75Mn_0.5O_6-δ. The procedure involved dissolving 50 mg CaSrFe_0.75Co_0.75Mn_0.5O_6-δ and an excess of KI (approximately 2 g) in 100 mL of 1 M HCl. Subsequently, 5 mL of the resulting solution was extracted for titration, where the iodine produced was titrated against 0.025 M Na₂S₂O₃. To indicate the endpoint, 10 drops of starch solution were introduced. The whole process of the experiment was carried out in an argon atmosphere.

Electrochemical Characterization of CaSr_0.75Mn_0.5Co_0.75O_6-δ

The CaSr_0.75Mn_0.5Co_0.75O_6-δ powder was precisely weighed (~2 mg) and combined with a Teflonized acetylene black (TAB-2) binder in a 2:3 (mg) ratio, then homogenized in an agate mortar. The resulting electrode mixture was compressed onto a stainless-steel substrate with an active area of 2 cm² and subsequently dried at 160 °C for 5 h under vacuum. Coin cells (2032-type) were assembled in an argon-filled glovebox, using CaSr_0.75Mn_0.5Co_0.75O_6-δ as the working electrode and lithium foil as the counter/reference electrode. A glass fiber separator (ADVANTEC GB-100R, Hibiya-Kokusai BLDG 5F, 2-2-3, Uchisaiwaicho, Chiyoda-ku, Tokyo, 100-0011, Japan) was placed between the electrodes, and the electrolyte consisted of 1M LiPF₆ in an ethylene carbonate (EC): dimethyl carbonate (DMC) (1:2 v/v) mixture. Electrochemical testing was performed using an Arbin 16-channel battery tester, with galvanostatic discharge–charge cycles conducted between 0.005 and 3.0 V at a current density of 50 mA g⁻¹ (room temperature). Additional rate capability tests were carried out at varying currents (50, 100, 200, 500, and 1000 mA/g). Cyclic voltammetry (CV) was performed using a Biologic SP200 potentiostat, France, scanning between 3.0 and 0.005 V at a rate of 0.1 mV s⁻¹.

3. Results and Discussion

3.1. Crystal Structure

The synthesized compound CaSrFe_0.75Co_0.75Mn_0.5O_6-δ belongs to the family of oxygen-deficient cubic perovskites, crystallizing in the Pm-3m space group. As shown in Figure 1, the X-ray diffraction (XRD) pattern confirms the phase purity, while the Rietveld-refined structural parameters (summarized in

Table 1. Atomic positions of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ, Pm-3m, a = 3.8279(9) Å, χ² = 1.580, Rp = 0.0321 wRp = 0.0423.

Elements	x	y	z	Occupancy	Uiso	Multiplicity
Ca/Sr	0.5	0.5	0.5	0.5/0.5	0.0132 (8)	1
Fe/Co/Mn	0.0	0.0	0.25	0.375/0.375/0.250	0.0295(5)	1
O	0.5	0	0	0.91	0.0487(5)	3

) exhibit excellent agreement with the reported literature values, validating the consistency of our synthesis.

As characteristic of oxygen-deficient perovskites, these materials follow the general formula ABO_3-δ or A₂B₂O_6-δ, where:

A = alkaline earth metals (Ca²⁺, Sr²⁺ in this study)

B = 3d/4d transition metals (Fe, Co, Mn here).

In CaSrFe_0.75Co_0.75Mn_0.5O_6-δ, the A-site is jointly occupied by Ca and Sr, whereas the B-site accommodates a mixed valence of Fe, Co, and Mn, contributing to its unique oxygen-deficient structure.

A detailed analysis of Figure 1b reveals that the Fe/Co/Mn atoms (green spheres) are octahedrally coordinated by six oxygen atoms (red spheres), forming well-defined BO₆ octahedra. For clarity, the octahedral geometry is emphasized by imaginary sky-blue planes connecting the oxygen atoms, highlighting the uniform six-fold coordination of the transition metals throughout the crystal structure. Notably, while this visualization assumes full occupancy for simplicity, potential oxygen-deficient sites—whose exact positions remain uncertain—are omitted from the representation.

In CaSrFe_0.75Co_0.75Mn_0.5O_6-δ, the B-site cations are distributed such that Fe^3+/4+ and Co^3+/4+ ions together account for 75% of the total occupancy, with each metal (Fe and Co) occupying 37.5% of the B-sites. The remaining 25% is filled by Mn^3+/4+ ions, completing the B-site composition. As illustrated in Figure 1b, the A-site cations (Ca²⁺/Sr²⁺), represented by large whitish-grey spheres, display 12-fold coordination with surrounding oxygen atoms in an ideal cubic perovskite arrangement. These A-site positions are equally shared between Ca and Sr, with 50% occupancy for each cation. The A-site coordination geometry forms a cuboctahedral environment, connecting to eight adjacent BO₆ octahedra in the three-dimensional framework. It should be noted that while the visualization assumes complete oxygen occupancy for clarity, actual oxygen-deficient sites—which would reduce the coordination number—are not depicted because of positional uncertainties in the XRD analysis.

The octahedra within the crystal lattice of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ are interconnected through corner-sharing via oxygen atoms, establishing a bonding pattern denoted as B-O-B, where B represents Fe/Co/Mn. The bond angle between B-O-B is consistently measured at 180°.

Additionally, Figure 2 presents an SEM image of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ, revealing a porous structure with diffused grains and a lack of distinct grain boundaries. Notably, iodometric titrations conducted on this compound yielded a value of approximately δ ≈ 0.56, indicative of a considerable degree of oxygen deficiency within the material. Past studies of Fe, Mn, and Co-based oxygen-deficient perovskites have been reported for different oxidation states, such as +3 and +4 for Fe, Mn, and Co. Therefore, we performed an XPS study and iodometric titration to find out the oxidation states of Fe in CaSrFe_0.75Co_0.75Mn_0.5O_6-δ. XPS study was conducted at room temperature using Al Kα radiation (1486.7 eV). The Fe—oxidation state is determined based on its 2P_3/2 peak position in XPS. Figure 3a displays the Fe 2P_3/2 peak at approximately 710 eV, accompanied by a satellite peak located ~8 eV higher in binding energy. This characteristic peak separation confirms the presence of Fe in the trivalent oxidation state (Fe³⁺) [25]. A distinct shoulder appearing at approximately 712 eV on the high-binding-energy side of the Fe 2P_3/2 peak provides evidence for Fe⁴⁺ species in the sample [25].

As shown in Figure 3b, the Co 2P_3/2 core level exhibits a primary peak centered at approximately 780 eV, flanked by two distinct satellite features located ~5 and 10 eV higher in binding. Co²⁺ is expected to show a satellite peak at about 5 eV higher than the 2P_3/2 peak [26]. The satellite peak at ~ 10 eV higher than the 2P_3/2 peak is a signature of Co³⁺ [27]. Thus, the compound contains Co⁺² and Co⁺³ oxidation states. XPS analysis identifies manganese in multiple oxidation states, with the Mn 2P_3/2 spectrum showing signature peaks at 641 eV (attributed to Mn³⁺) and 643 eV (assigned to Mn⁴⁺) [28]. Complementary iodometric titration analysis supports the XPS findings, determining an oxygen vacancy of δ = 0.56 in CaSrFe_0.75Co_0.75Mn_0.5O_6-δ, confirming the oxygen-deficient nature of the material.

3.2. Anode Properties

Figure 4 shows the galvanostatic charge/discharge curves for the CaSrFe_0.75Co_0.75Mn_0.5O_6-δ material at a current density of 50 mA g⁻¹, covering the 1st, 2nd, 10th, 50th, and 100th cycles, with a voltage range from 3.0 to 0.005 V. During the first cycle, the material exhibits an initial discharge capacity of 1100 mAh g⁻¹. This high capacity is likely due to the substantial incorporation of lithium at defects on the crystal surfaces [29]. The first cycle discharge curve features a plateau ranging between 0.9 and 0.75 V, similar to other materials in this category [2,30]. The discharge capacity stabilizes around 380 mAh g⁻¹ after the 2nd cycle onwards, maintaining this value through to the 100th cycle.

The electrochemical testing demonstrates an initial charge capacity of 400 mAh g⁻¹ and a discharge capacity of 1100 mAh g⁻¹, yielding a first-cycle coulombic efficiency of ~36%. This substantial irreversible capacity loss (~700 mAh g⁻¹) is mainly associated with two factors: (1) solid electrolyte interphase (SEI) layer formation and (2) electrolyte reduction side reactions during the first lithiation process [30,31]. This phenomenon is consistent with observations in other oxide materials, where capacity loss is linked to structural degradation and electrolyte reduction [2,6].

The voltage profile of the first discharge cycle is qualitatively similar to that of Ca₂Fe₂O₅ and Ca₂Co₂O₅ [2] and quantitatively comparable to Fe₂O₃ [30], with a plateau near 0.75 V until reaching a capacity of around 280 mAh g⁻¹, after which the voltage decreases gradually to 0.005 V. This indicates an electrochemical process akin to those observed in Ca₂Fe₂O₅, Ca₂Co₂O₅, and Sr₂Fe₂O_6-δ materials [2,18] The plateau voltage position is sensitive to the crystal structure and the nature of the metal ions involved in the electrochemical reaction; a characteristic also noted in other 3d metal oxides [2,30].

After the first cycle, the charge and discharge profiles remain similar, indicating a reversible lithium insertion/extraction process. By the 100th cycle, the average charge and discharge potentials converge around ∼0.15 V, suggesting a low average potential and minimal charge/discharge hysteresis, highlighting the material’s efficiency as an anode. The discharge capacity remains around 430 mAh g⁻¹ at the second cycle and 280 mAh g⁻¹ at the one hundredth cycle, with the corresponding charge capacities stabilizing around 400 mAh g⁻¹. These consistent capacities after the initial cycle demonstrate the electrolyte’s stability and the anode’s superior cycling performance compared with Ca₂Fe₂O₅, Ca₂Co₂O₅, and Sr₂Fe₂O_6-δ [2,18].

As shown in Figure 5, the CaSrFe_0.75Co_0.75Mn_0.5O_6-δ electrode delivers exceptional cycling stability at 50 mA g⁻¹, retaining a reversible capacity of ~380 mAh g⁻¹ over 100 cycles. This performance surpasses the theoretical capacity of graphite (372 mAh g⁻¹), highlighting the material’s potential as a high-capacity anode for lithium-ion batteries [10]. In practical applications, the capacity for graphite anodes ranges between 300 and 320 mAh g⁻¹ [6].

The first cycle of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ exhibits a moderate irreversible capacity loss, primarily attributed to two key factors: (1) the formation of a solid electrolyte interphase (SEI) layer and (2) electrolyte decomposition reactions. These processes are characteristic of initial activation in oxide-based electrode materials [31,32]. As depicted in the figure, the capacity remains consistent up to 100 cycles with an average coulombic efficiency of over 98%, outperforming materials such as Ca₂Fe₂O₅, Ca₂Co₂O₅, Sr₂Fe₂O_6-δ, and other comparable materials.

Figure 6 presents the rate capability test of the CaSrFe_0.75Co_0.75Mn_0.5O_6-δ electrode within a voltage range of 3.0–0.005 V under various current densities (50, 100, 200, 500, and 1000 mA g⁻¹). As the current densities increase from 50 to 1000 mA g⁻¹, the electrode delivers discharge capacities of 400, 365, 300, 240, and 150 mAh g⁻¹, respectively. After the rate performance test, when the current rate is reduced back to 100 and 50 mA g⁻¹, the CaSrFe_0.75Co_0.75Mn_0.5O_6-δ electrode retains discharge capacities of 400 and 365 mAh g⁻¹, respectively. This demonstrates the electrode’s good reversibility and high rate capability. The galvanostatic charge–discharge curves of the CaSrFe_0.75Co_0.75Mn_0.5O_6-δ electrode exhibit slopping profiles without distinct potential plateaus, indicating a solid-solution-like mechanism [33,34]. The cyclic voltammetry measurement of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ electrode at the voltage range of 0.005–3.0 V with a scan speed of 0.1 mV/s shows the first reduction potential peaks at 0.94 and 0.6 V, respectively. During oxidation, a small bump-like peak at 1.99 V was observed. Further cyclic voltammetry measurement shows similar reduction potential peaks with shoulder-like bumps at 1.15 V and 0.6 V, respectively, and anodic oxidation peaks at 0.19 and 2.03 V, respectively, during the repeated cycles.

Figure 7 shows the cyclic voltammetry (CV) curves for the CaSrFe_0.75Co_0.75Mn_0.5O_6-δ anode electrode, scanned over a voltage range of 0.005 to 3.0 V (vs. Li/Li⁺) at a scan rate of 1 mV s⁻¹ to study the electrochemical mechanism. The CV testing begins with the cathodic scan (reduction process) followed by the anodic scan (oxidation process).

In the first cycle’s reduction process, two peaks are observed, whereas only one peak is seen in the subsequent cycles. The voltage profile is similar to previous reports on Fe₂O₃, [30] Ca₂Fe₂O₅, Ca₂Co₂O₅, [2] and Sr₂Fe₂O_6-δ [18] materials showing a long flat plateau at around 0.75 V during first cycle discharge. A peak around 0.95 V in the first cycle’s cathodic scan indicates the non-reversible formation of a solid/electrolyte interphase (SEI) on the electrode surface, leading to irreversible capacity loss.

The first cycle exhibits a cathodic peak near 0.95 V, which is hidden by the CV lines of other cycles. Subsequent cycles (two through five) show a small cathodic peak at around 1.2 V. The overlapping CV curves from the second to the fifth cycle suggest structural stability and good reversibility. The anodic and cathodic peaks may shift slightly between cycles. The peak at approximately 0.55 V in the first cathodic cycle may represent a leftward shift from peaks in later cycles, indicating crystal structure destruction. The other reduction peak at around 0.55 V in the first cycle and around 1.2 V in subsequent cycles is likely because of electrochemical reactions.

During the oxidation process, the CV curves show consistent anodic peak positions across all cycles, indicating the material’s stable anodic characteristics. While the chemical process for Fe⁴⁺ remains unknown, the chemical behavior of Fe³⁺ can be described as reported previously.

The initial discharge process triggers an irreversible structural transformation, wherein the crystalline framework undergoes complete breakdown. This reaction yields metallic nanoparticles uniformly dispersed within an amorphous SrO and Li₂O composite matrix, as confirmed by ex situ XRD analysis [2,18].

Sr₂M₂O_6-δ + 6Li⁺ + 6e⁻→2SrO + 2M⁰ + 3Li₂O

where M represents Fe, Co, or Mn in the compound’s formula.

Upon subsequent charging, these metal particles are oxidized back to their oxides, accompanied by the decomposition of Li₂O: [2,18]

M⁰ + Li₂O↔M²⁺O + 2Li⁺ + 2e⁻

The observed reversible capacity originates from a displacement mechanism involving three distinct phases: (i) the metal oxide framework, (ii) metallic nanoparticles, and (iii) lithium oxide (Li₂O). During cycling, this three-phase system undergoes reversible transformations through concurrent Li₂O formation/decomposition and metal nanoparticle redox processes, as confirmed by in situ XRD analysis.

From the second to the fifth cathodic cycle, the overlapping CV curves indicate the structural stability and reversibility of CaSrFe_0.75Co_0.75Mn_0.5O_6-δ. The specific behavior of M⁴⁺ in this reaction has not been discussed in prior research and remains an area for future investigation.

4. Conclusions

This research suggests that oxygen-deficient perovskites could be potential candidates for use as anodes in lithium-ion batteries. The CaSrFe_0.75Co_0.75Mn_0.5O_6-δ electrode demonstrates a discharge capacity of 393 mAh g⁻¹, surpassing that of the graphite anode. It also exhibits consistent charge–discharge capacities for up to 100 cycles. Rate capability tests of the CaSrFe_0.75Co_0.75Mn_0.5O_6-δ electrode indicate good performance by virtue of cycle, capacity, and reversibility, promising the potential of perovskite oxides for application as anodes in lithium-ion batteries.