Emerging multimetal LMFP-based cathodes for lithium-ion batteries: a review

Abstract

The increasing demand for smart portable electronics and electric vehicles is driving advancements in high-energy-density lithium-ion batteries (LIBs). Among various cathode materials, lithium manganese iron phosphate (LiMn_yFe_1−yPO₄ – LMFP) stands out due to its cost-effectiveness, excellent safety profile, long cycle life, high operating voltage, robust thermal stability, and competitive energy density. However, despite notable progress, LMFP still encounters key challenges such as limited electronic conductivity, sluggish Li-ion diffusion, manganese dissolution affecting cycling stability, and low tap density. To address these issues, significant efforts have been devoted to developing multimetal LMFP and other multimetal olivine-based cathodes with enhanced electrochemical properties. Nevertheless, a comprehensive review of these recent advancements remains lacking. This article aims to bridge this gap by examining emerging strategies and recent developments in multimetal LMFP cathodes for LIBs. It systematically compares various optimization approaches, including (i) doping with single or multiple elements at Li-, M-, or Li & M-sites, (ii) integrating additional primary constituents or structural components, and (iii) increasing the configurational entropy of electrode materials. The advantages and limitations of these strategies are critically assessed. Furthermore, key synthesis methodologies and processing techniques used to enhance LMFP cathode performance are discussed. Finally, the review provides insights into the benefits and challenges associated with these materials, highlighting future perspectives and potential research directions.

---

Josué M. Gonçalves

Josué Martins Gonçalves is a Researcher at the Mackenzie Institute for Research in Graphene and Nanotechnologies (MackGraphe) within the Mackenzie Presbyterian Institute (IPM), Brazil. He graduated in Chemistry at the University Vale do Acaraú (UVA) in 2014 and received his PhD from the Institute of Chemistry at the University of São Paulo (USP) in 2019. He developed cutting-edge research during his postdoctoral internships at USP (2019–2021), the University of Illinois-Chicago (UIC, 2021–2022), and the University of Campinas (Unicamp, 2023–2024). His current research interests include applications of high-entropy materials for energy conversion and storage.

Syra Mubarac

Syra Mubarac is a PhD candidate at the State University of Campinas (UNICAMP) and a member of the Advanced Energy Storage Division of the Center for Innovation on New Energies, Brazil. She obtained her bachelor's degree in chemical engineering from the State University of Amazonas (UEA) and her MSc from the School of Electrical Engineering at the UNICAMP in 2024 with emphasis on Electrical Energy. She has worked on the development of lithium-ion batteries, from synthesis to manufacturing. Her current research interests include the study and synthesis of multi-metallic and high-entropy materials applied to intercalation energy storage systems.

Gustavo T. M. Silva

Gustavo Thalmer M. Silva obtained his undergraduate degree in Chemistry from the State University of Rio Grande do Norte (2012) and a Master’s degree in Chemistry from the Federal University of Rio Grande do Norte (2014). He received his PhD at the Institute of Chemistry of the University of São Paulo (IQ-USP) in 2019 under the supervision of Frank H. Quina and is currently conducting postdoctoral research at the same institution. His research interests include energy sources, colloid and materials chemistry, the chemistry and photochemistry of natural pigments and their synthetic analogues, and quantum chemical methodology for electronically excited states.

Bruno Freitas

Bruno Guilherme Aguiar Freitas is a Postdoctoral Researcher at the Center for Innovation in New Energies, affiliated with the University of Campinas (UNICAMP), Brazil. He holds a B.Sc. (2014) and an M.Sc. (2017) in Chemistry from the Federal Fluminense University (UFF) and a Ph.D. in Electrical Engineering (2023) from UNICAMP. He specializes in energy storage materials, using the Brazilian synchrotron, Sirius, and international facilities like SLAC (USA) and MAX IV (Sweden) for in situ/operando XRD studies of supercapacitors and batteries. His current work focuses on synthesizing and characterizing electrodes for batteries and supercapacitors using electrochemical and synchrotron-based techniques.

Benedito G. Aguiar Neto

Benedito Guimarães Aguiar Neto holds a degree in Electrical Engineering from the Federal University of Paraíba (1977), a Master's degree in Electrical Engineering from the Federal University of Paraíba (1982), Brazil, and a PhD in Electrical Engineering from the Technische Universität Berlin, Germany (1987). Postdoctoral fellow at the University of Washington, USA (2017–2018) and at the University of Minho, Portugal (2021–2022). He was a Professor at the Federal University of Campina Grande, Paraíba, Brazil, until 2011, and since 2011, he has been at the Mackenzie Presbyterian University, São Paulo, Brazil, where he was Rector until 2020. Since December 2022, he has been General Director of the Mackenzie Institute for Research in Graphene and Nanotechnologies (MackGraphe).

Hudson Zanin

Hudson Zanin is a Professor at the School of Electrical and Computer Engineering of the University of Campinas, Brazil, and a Researcher in the Advanced Energy Storage Division of the Center for Innovation on New Energies. His research focuses on developing functional (nano)materials and their application in Energy Storage and Conversion Devices. He has experience in supercapacitors, batteries, and fuel cells' materials and devices. He is the founder of the first South American lab-scale batteries and supercapacitors manufacturing. Dr Zanin has a PhD degree in Electrical Engineering from the University of Campinas (2012).

---

1. Introduction

The growing demand for advanced portable electronics and electric vehicles is driving innovations in lithium-ion batteries (LIBs),^1,2 with enhanced energy density, ushering in a new era in the energy transition. In the pursuit of next-generation energy storage systems, significant efforts have been focused on the development of next-generation cathode materials, with a strong emphasis on designing nickel-rich layered oxides,^3,4 especially super nickel-rich (cobalt-low) compositions. This research particularly centers on compounds such as LiNi_xCo_yMn_zO₂ (NCM), LiNi_xCo_yAl_zO₂ (NCA), and LiNi_xCo_yMn_zAl_{1−x−y−z}O₂ (NCMA), where x ≥ 0.85.⁵ While nickel-rich cathodes hold significant promise, these compositions alone are unlikely to meet future demands, particularly given the growing emphasis on safety in energy storage devices.⁶

Regarded as a strong alternative to nickel-rich cathodes, lithium manganese iron phosphate (LiMn_yFe_1−yPO₄, LMFP) has attracted considerable interest as a potential cathode material for LIBs, owing to its low cost, excellent safety characteristics, prolonged cycle stability, elevated operating voltage, reliable performance under high-temperature conditions, and competitive energy density.⁷ In fact, LMFP is viewed as an advanced version of LiFePO₄ (LFP), offering a slightly higher operating voltage (∼3.5 V vs. Li/Li⁺ for the Fe²⁺/Fe³⁺ redox couple) than single-phase LFP (3.4 V vs. Li/Li⁺). It is compatible with existing LFP production lines, making it a cost-effective choice for industrial-scale production, aiming to accelerate its commercialization.⁸

Despite substantial progress over recent decades, LMFP still faces major challenges, including limited electronic conductivity, sluggish lithium-ion diffusion, manganese dissolution impacting cycling stability, and low tap density. These drawbacks greatly affect its energy storage capacity, rate capability, and cycling stability, limiting its broader adoption in high-performance LIBs.⁷ To overcome these limitations, various modification strategies, such as morphology control, elemental doping, applying conductive surface coating, micronizing primary particle size, and carefully managing defect concentrations, are widely used to boost conductivity, facilitate lithium-ion transport, and optimize both kinetics and electrochemical performance.⁹

More recently, the design of multielemental cathode materials has been an essential strategy to improve stability and overall electrochemical performance, especially in electrode materials for energy storage. Notably, these enhancements can be achieved by (i) doping with one or multiple elements, (ii) adding one or more principal constituents or structural components, and/or (iii) increasing the entropy of the electrode materials. Doping with various elements has been investigated as a strategy to improve both electronic and ionic conductivity. Dopants, which can be either cations or anions, may substitute at Li-sites, Mn/Fe-sites, or anionic (PO₄) sites.¹⁰ On the other hand, when the dopant concentration exceeds 5% by mass or atomic percentage, it should be regarded as a primary constituent or structural component of the electrode material. These primary constituents must be carefully selected, as higher concentrations can improve stability but may lead to reduced specific capacity, particularly when dealing with electrochemically inactive elements. Similar to the addition of primary constituents, when these elements are in near-equimolar proportions, entropy stabilization, and/or synergistic effects may enhance electrochemical properties. For instance, incorporating a third or fourth element can yield a medium-entropy material (MEM), whereas the presence of five or more elements results in a high-entropy material (HEM).¹¹

Recent review articles have spotlighted advancements and innovative strategies in the design of LMFP cathode materials. For instance, Xu et al.⁸ reviewed recent advancements and challenges in optimizing strategies to enhance the electrochemical performance of LMFP. Their work highlights approaches such as fine-tuning the Mn/Fe ratio, compositing with conductive materials, elemental doping, and morphology control. Particular emphasis was placed on regulating the additional discharge plateau, a critical factor for preventing energy density loss while ensuring the consistency and reliability of LMFP-based batteries. Similarly, Zhang and colleagues⁷ reviewed the reaction mechanisms, synthesis approaches, and electrochemical behavior of LMFP, offering insights that provided guidance for better material development. They outlined key challenges, proposing strategies like multi-scale particle tailoring, heteroatom doping, surface coating, and structural morphology engineering to improve kinetics and overall electrochemical properties. Furthermore, Akhmatova and collaborators¹⁰ emphasized recent advancements in multi-element doping of LFP cathodes as an effective strategy to enhance their electrochemical performance. However, to our knowledge, no reviews have comprehensively addressed the promising outcomes, comparisons, and distinctions associated with elemental doping, including principal constituents or structural components, and/or entropy enhancement in LMFP materials. While there has been a noticeable increase in publications on this subject in recent years, showcasing intriguing performances of multimental LMFP cathodes, a detailed analysis remains absent.

In that regard, this review aims to fill this gap, highlighting recent advancements, trends, and emerging strategies in the development of multimental LMFP-based cathodes for LIBs (Scheme 1). It provides a comprehensive comparison of approaches, including (i) doping with single or multiple elements, (ii) incorporating additional primary constituents or structural components, and (iii) enhancing the entropy of electrode materials. The advantages and limitations of these strategies are thoroughly analyzed. Additionally, the review outlines key synthesis protocols and techniques employed to optimize the electrochemical properties of LMFP cathodes. Finally, the benefits and drawbacks of these materials are critically evaluated, and insights into future directions and potential advancements are presented.

Scheme 1 Key optimization strategies and substitution sites employed in the design of multimetal LMFP cathodes. At the center, the primary metal ions utilized in the development of advanced LMFP and other multimetal olivine cathodes. The LMFP structural model shown at the core of the scheme has been adapted with permission from ref. 12. Copyright © 2025 Wiley-VCH GmbH.

2. Principles and challenges in LIBs

Since their commercial introduction by Sony Corporation in 1991, lithium-ion batteries (LIBs) have attracted significant attention as a key technology for energy storage.^5,13–17 Their relatively high energy density, lightweight design, fast recharge capability, long cycle life, and environmental friendliness have established LIBs as the dominant energy source for both portable electronic devices, electric vehicles (EVs), and stationary energy storage devices.^5,13–17 In the context of EVs, there is a growing demand for higher battery capacity, improved performance, and greater stability to effectively replace internal combustion engines and accelerate the global adoption of EVs. Therefore, LIBs have shown great promise in meeting these demands due to their qualities.¹⁸

Fig. 1 shows the composition of a LIB. In LIBs, the charge–discharge process is reversible, requiring both oxidation and reduction to take place at the same electrode; nevertheless, the negative and positive electrodes are conventionally referred to as the anode and cathode, respectively. During the discharging process, positively charged lithium ions move from the anode (where oxidation occurs) to the cathode (where reduction takes place). These ions are transported through the electrolyte, which facilitates ion movement, and pass through the separator, which prevents the free flow of electrons and avoids short circuits between the electrodes.^5,13–19 As the ions reach the electrodes, electrons flow through the external circuit. The current collectors serve as a connection (providing the pathway) for the flow of electrons between the electrodes and the external circuit. On the other hand, during the charging process, the positive lithium ions are released from the cathode to the anode.^5,13–20

Fig. 1 Schematic representation of the composition and discharge/charge cycle for LIBs.

In order to find solutions to mitigate CO₂ emissions, governments have been facilitating the global production of EVs.^21–23 However, the next generation of EVs requires LIBs with enhanced structural stability, higher energy density, and improved performance to meet the demands of long driving ranges between charges.^24,25 Since electrode conductivity plays a decisive role in the electrochemical response of LIBs, advances in anode and cathode design are crucial. Research in this field has fostered the development of alternative active materials intended to reduce production costs and increase safety, while also enhancing stability and key charge–discharge parameters such as voltage, reversibility, and capacity.^26–29 The use of graphite as anode material has been well stablished, with moderate specific capacities of c.a. 372 mAh g⁻¹ and low cost.³⁰ Therefore, the cathode material is a determining factor not only in the cost but also in the energy density and power density of LIBs.

Among conventional cathode materials, LiCoO₂ (LCO), LiNi_1−x−yCo_xMn_yO₂ (NCM), and LiFePO₄ (LFP) are some of the most widely used. Although LCO presents a high theoretical specific capacity (274 mAh g⁻¹) and energy density, the actual specific capacity is only about half of the theoretical specific capacity under a cutoff voltage of 4.2 V.³¹ For a higher cutoff voltage, lattice distortion and deterioration during cycling can occur. In addition, safety problems have occurred due to the dissolution of Co in the electrolyte and the decomposition of electrolyte.⁷ NCM, which tends to replace the determined fraction of Co by Ni and Mn, has been an alternative for improving specific capacity and energy density,⁴ mainly when increasing the Ni-content.⁵ However, problems caused by Li/Ni cation mixing, microcrack formation, and interfacial side reactions have hampered commercial applications.⁵ On the other hand, LFP stands out as a highly attractive option for energy storage. Composed of abundant, low-cost, and environmentally friendly raw materials, it offers a safer, more economical, and less toxic alternative to the LCO and their cobalt-containing derivatives, which face availability constraints.^19,32 In addition, it also exhibits superior thermal stability and longer cycle life.^33,34 However, LFP exhibits unsatisfactory performance in low-temperature environments and has a relatively low energy density³² compared to other LIB cathode materials, which limits its use in applications with high energy demand, such as EVs.³⁴ Besides, the intrinsically sluggish lithium-ion diffusion and limited electronic conductivity significantly hinder its rate capability.³⁵ Fortunately, considering the strengths and limitations of the aforementioned cathode materials, each chemistry can play a complementary role in battery technology. Rather than aiming for full substitution, their optimal value lies in strategic integration (i.e., blended cathode design)³⁶ or application-specific deployment, leveraging the unique advantages of each material for tailored energy storage solutions.

On the other hand, the addition of manganese, replacing partially iron, forms lithium manganese iron phosphate (LiMn_yFe_1−yPO₄, LMFP), an innovative and environmentally friendly cathode with higher energy density and good cycle stability, which combines the safety and stability of LFP with the higher voltage characteristics of lithium manganese phosphate (LMP).³³ LMFP can offer enhanced energy density due to the higher operating voltage of the Mn^2+/3+ redox couple (∼4.0 V) compared to that of Fe^2+/3+ (∼3.4 V), and its theoretical energy density can reach ca. 21% more than that of LFP.³⁷ Indeed, LMFP offers several advantages, including low cost, high safety, long cycling life, elevated voltage, good high-temperature performance, and improved energy density.^7,38,39 However, LMFP also faces several challenges that affect the overall performance and limit its applications.

Despite the numerous advantages of LMFP, there are still structural limitations that compromise its electrochemical performance and hinder its widespread application. Olivine-type structure phosphates exhibit intrinsic limitations, such as low electrical conductivity and slow lithium-ion kinetics.⁴⁰ In fact, some structural challenges in LMFP cathode materials are associated with the Jahn–Teller effect triggered by Mn³⁺ ions. The Jahn–Teller effect of the Mn³⁺ ion causes kinetic limitations of Mn^2+/3+ redox, resulting in lattice strain, poor lithium-ion diffusion, and low electronic conductivity.^33,37,41 This leads to significant polarization under high currents during charge–discharge cycles, affecting its performance and capacity.³³ Additionally, low-temperature environments exacerbate these issues by reducing conductivity and increasing ion migration impedance.^33,41 The oxidation of Mn²⁺ to Mn³⁺ during LMFP delithiation induces significant MnO₆ octahedral distortion, increasing the activation energy for ionic diffusion. This structural deformation manifests through the selective elongation of two equatorial Mn–O bonds (edge-sharing with PO₄ tetrahedra) and the contraction of the remaining Mn–O bonds, differing fundamentally from classical Jahn–Teller distortion.^42–48 Normally, the Jahn–Teller distortion in Mn³⁺O₆ octahedra manifests as axial bond elongation (c/a > 1) with equatorial bond contraction (Fig. 4a).

However, LiMnPO₄ exhibits a distinct distortion pattern due to structural constraints imposed by (PO₄)³⁻ tetrahedra. During delithiation, this pre-existing structural configuration directs the Mn³⁺ Jahn–Teller distortion toward an unusual c/a < 1 geometry, characterized by the shortened axial Mn–O1/Mn–O2 bonds and the elongated equatorial Mn–O3/Mn–O3′ bonds (Fig. 4b). This inverse distortion results from the combined effects of Mn–P Coulomb repulsion and strong (PO₄)³⁻ bonding, which preferentially stabilizes the 3z²–r² orbital configuration while driving Mn³⁺ displacement toward O3–O3 edges (Fig. 4c). The bond length changes confirm this atypical distortion behavior, demonstrating how polyhedral connectivity in olivine phosphates can fundamentally alter conventional Jahn–Teller expectations.⁴³ The structural distortion induced by the formation of Mn³⁺ during the charged state of the MnO₆ octahedra increases the obstruction of Li⁺ migration. This distortion reduces Li⁺ diffusivity between the initial (Mn²⁺-rich) and fully charged (Mn³⁺-rich) states. This kinetic limitation originates from the lattice strain generated by the Mn²⁺/Mn³⁺ redox couple, which distorts the crystal lattice and obstructs Li⁺ transport channels. The strong correlation between Mn³⁺ concentration and decreased Li⁺ mobility confirms that mitigating this Jahn-Teller distortion is critical for developing LMFP cathode materials with enhanced rate capability.^44,45 This deformation significantly increases internal resistance within the LMFP lattice, resulting in an extremely low ion diffusion coefficient, which hampers the efficiency and performance of these materials in electrochemical systems due to the kinetic barrier at the interfacial diffusion in the resulting biphasic interface.⁴⁹

In order to solve the aforementioned challenges and enhance the performance of LMFP, various modification strategies have been implemented. Key approaches include surface coating, elemental doping, morphology control, material nanosizing, and electrolyte system regulation.⁴¹ Among these, elemental doping in the LMFP crystal lattice stands out for its potential to enhance the material's intrinsic conductivity. For instance, cation doping can cause changes in lattice vacancies or atomic bond lengths, facilitating lithium movement within the lattice and improving electrochemical performance. Thus, based on the growing body of literature and several reviews that broadly outline advances in strategies for enhancing LMFP cathodes, this review specifically emphasizes (with greater depth) the promising outcomes, key comparisons, and notable distinctions related to elemental doping, incorporation of principal constituents or structural components, and/or entropy enhancement in LMFP materials. Although a marked increase in publications has been observed in recent years (many reporting impressive performances of multimetal LMFP cathode), a comprehensive and critical analysis of these developments is still lacking, as will be discussed later.

3. LMFP: structural features and electrochemical aspects

Lithium iron and manganese phosphate-based cathodes (LiMn_yFe_1−yPO₄, LMFP) hold great prospects for large-scale production in advanced lithium-ion batteries. In fact, several companies (mainly in China) have announced pilot-scale LMFP cathodes and cell prototypes, aiming to use them in electric vehicles (EVs) and energy storage systems as a mid-point technology between LFP and NMC. The combination of high performance at higher operating voltages provided by lithium manganese phosphate (LiMnPO₄, LMP) with the high safety of lithium iron phosphate (LiFePO₄, LFP) makes this material a strong competitor for the commercial batteries for EVs and high-power energy storage systems, due to its promise of environmental compatibility and high energy density compared to LFP.^7,10 Indeed, partially substituting iron with manganese in the LFP structure can enhance the energy density by approximately 15–20%.⁵⁰

The manganese (Mn²⁺) inserted into the crystal structure of LFP is primarily responsible for giving LMFP the characteristic of operating within a high voltage working window when applied in electrochemical devices. Moreover, LMFP materials exhibit an asymmetric charge–discharge voltage electrochemical signature, primarily due to their intricate phase transition processes.⁵¹ With an increased operational voltage range, the system tends to exhibit a higher upper limit of specific energy density, as this energy density is the result of the product of voltage and specific capacity.⁷ LMFP solid solutions, with varying Fe/Mn ratios,⁵² are isostructural with their LFP and LMP endmembers.⁵³ These mixed-metal phosphates adopt the same olivine-type lattice, maintaining structural consistency across different compositions.^53–55 However, the distinct d-electron configuration of Mn introduces additional complexity in its electronic rearrangement and lattice evolution mechanisms.⁵¹

The olivine structure exhibited by LMFP is orthorhombic (space group Pnma), consisting of PO₄ tetrahedra with both Fe and Mn ions occupying octahedral sites where Fe²⁺ and Mn²⁺ ions share corners (4c sites), phosphorus ions are located in tetrahedral sites, and Li⁺ ions occupy the edge-shared octahedral site (4a sites), with the latter running parallel to the b-axis (Fig. 3a and b).^53,54 In a more detailed study, Kope et al.⁵⁶ described LMFP as a stable solid solution due to slight lattice expansion caused by the small difference in ionic radius between high-spin Fe²⁺ and Mn²⁺ ions. They also demonstrated that the Mn-rich phase (y > 0.8) in Li_xMn_yFe_1−yPO₄ was unsuitable for high-specific-capacity cathode materials due to intrinsic limitations arising from strong electron–lattice interactions (Mn³⁺:3d⁴, Jahn–Teller effect) in the charged state (Mn³⁺/Fe³⁺) that induces lattice distortions.^54,56

Computational studies by Gardiner & Islam⁵³ employing three distinct Li⁺ migration pathways were studied in LiFe_0.5Mn_0.5PO₄: (1) Path α along [010] (b-axis, 2.9–3.0 Å), (2) Path β along (001) (4.6–4.7 Å), and (3) Path γ along (101) (5.6–5.8 Å) (Fig. 3c). The analysis revealed that the lowest energy barrier (E_mig = 0.59 eV) corresponds to one-dimensional migration along the (010) crystallographic direction (Path α), characterized by a 2.9–3.0 Å jump distance between adjacent Li sites (Fig. 3d and e). This finding agrees remarkably well with experimental activation energies (0.63 eV) reported for LiFe_0.45Mn_0.55PO₄. In contrast, alternative pathways along (001) (Path β) and (101) (Path γ) directions exhibited prohibitively high energy barriers (>2.2 eV), effectively restricting Li⁺ mobility to the b-axis channels and confirming the strongly anisotropic nature of ionic diffusion in olivine-type materials. Detailed analysis of the migration trajectory revealed a curved pathway between adjacent lithium sites (Fig. 2d–3e), with a maximum deviation of 0.44 Å from linearity, in agreement with neutron diffraction studies. Therefore, Li⁺ tends to move along the b-axis (010) (Fig. 3c) in one-dimensional channels, and diffusion transport in only one-dimension results in limited Li⁺ kinetics due to potential lattice defects. This severely restricts the rate at which lithium ions can move during electrochemical charge–discharge cycles.⁴⁹

Fig. 2 (a) Classical Jahn–Teller effect. Reproduced with permission from ref. 47. Copyright © 2013 American Chemical Society. (b) The distorted interstitial octahedron in the LiMnPO₄ to MnPO₄ framework. Reproduced with permission from ref. 43. Copyright © 2012 Elsevier B.V. All rights reserved. (c) Effects of (PO₄)³⁻ Coulomb repulsion and strong bonding on framework distortion. Adapted with permission from ref. 44. Copyright © 2016 Elsevier Ltd all rights reserved.

Fig. 3 Olivine-type structure of LMFP illustrating arrangement of FeO₆ and MnO₆ octahedra; (a) regular and (b) segregated planes. Reproduced with permission from ref. 53. Copyright © 2009 American Chemical Society. (c) Illustrates the material LiMPO₄ (M = Mn, Fe) for Li⁺ migration in olivine structure. The dotted circles denote the most probable path for Li⁺ migration in the (d) a and b plane view and (e) a–c plane view. Reproduced with permission from ref. 49. Copyright © 2025 by the Korean Electrochemical Society.

In addition to the aforementioned Li⁺ migration studies, extensive investigations have characterized the phase transition behavior of olivine LMFP cathodes during charge/discharge processes, revealing both “two-phase transition” and “single-phase solid-solution” mechanisms to explain lithiation and delithiation processes.^52,57–62 Moreover, it is widely recognized that the Mn/Fe ratio plays a pivotal role in determining the material's phase transition behavior (see Ling et al.¹² for a detailed review).

The phase transition mechanism of Li⁺ extraction/insertion involves the emergence of a new phase during the cycle and its disappearance after the cycle completion.⁵⁷ This two-phase transition exhibits characteristic reversibility, with a single-phase transition at the beginning and the end of charge (x ≤ 0.42 and x ≥ 0.91, respectively), while a two-phase coexistence region (olivine and heterosite) appears at intermediate states (0.42 < x ≤ 0.91) in Li_(1−x)Mn_0.7Fe_0.3PO₄ (0 ≤ x ≤ 1.0). The resulting phase interface significantly impacts kinetic behavior, where lattice mismatch at the boundary increases electronic transport barriers and reduces conductivity.⁶² In contrast to the two-phase transition model, the solid-solution transformation model proposes continuous and homogeneous phase evolution with lithium content variation. Particularly during Li⁺ insertion, the phase transition from L_xMFP to LMFP occurs through a continuous solid-solution pathway, where the L_xMFP composition progressively evolves until reaching stoichiometric LMFP (Fig. 4). This coherent solid-solution behavior significantly enhances the rate capability during charge/discharge cycles.^63,64 Fortunately, other recent reviews^51,65,66 have summarized and outlined the phase transition mechanisms of LMFP during charge/discharge, describing three reaction regions linked to Fe²⁺/Fe³⁺ and Mn²⁺/Mn³⁺ redox processes, highlighting how factors like Mn/Fe ratio, particle size, and synthesis conditions influence structural evolution and electrochemical performance (see for more details^12,65,66).

Fig. 4 The coherent transformation model shows hysteresis during the first charge (Li extraction) and discharge (Li insertion) cycles as LMFP transforms to L_xMFP. Adapted with permission from ref. 63. Copyright © 2014 American Chemical Society.

Interestingly, doping strategies, whether with single or multiple metals, profoundly influence these phase transition pathways, often shifting the reaction kinetics from a dominant two-phase model towards a more favorable solid-solution behavior, thereby enhancing structural stability and electrochemical performance.^67–69 The beneficial impact of doping is further highlighted by comparing pristine and doped materials. While Zn doping did not radically alter the sequence of phase transitions-both Zn-LMFP/C and undoped LMFP/C underwent a two-phase reaction during the Mn³⁺/Mn²⁺ redox process, followed by a solid-solution reaction during the Fe³⁺/Fe²⁺ redox process-it significantly improved the reaction kinetics.⁶⁹ This suggests that the primary role of certain dopants is not to change the fundamental thermodynamic pathway but to facilitate faster Li⁺ diffusion, thereby reducing the kinetic hindrances that exacerbate phase separation and polarization.

In situ X-ray diffraction (XRD) studies have been pivotal in elucidating how cation doping alters the de/intercalation mechanisms. For instance, Nb⁵⁺ doping was shown to induce a hybrid mechanism. During charging, the extraction of Li⁺ from LiMn_0.6Fe_0.4PO₄ leads to a two-phase transition, evidenced by the disappearance of lithiated-phase peaks and the emergence of new peaks corresponding to the delithiated Mn_0.6Fe_0.4Nb_0.02PO₄ phase. Crucially, at higher voltages (>4.2 V), the diffraction peaks of this delithiated phase shift continuously towards higher angles, indicating a solid-solution process within the Mn-based redox couple.⁶⁷ A similar phenomenon was observed with Cu²⁺ doping, where the reversible transformation between LiMn_0.6Fe_0.39Cu_0.01PO₄ and Mn_0.6Fe_0.039Cu_0.01PO₄ was accompanied by continuous peak shifts, confirming that the doping element stabilizes the structure without disrupting the olivine framework and promotes a more continuous reaction.⁶⁸ Besides, and as highlighted in section 5, the advances in in situ studies have shown that aliovalent metal doping into LMFP can significantly affect the structural evolution of LMFP during lithiation and delithiation, especially when combined with strategies such as carbon coating and Fe-vacancy engineering. These approaches help narrow the two-phase reaction region, leading to more uniform phase transitions, improved reaction stability, and enhanced electrochemical performance.^70,71

The optimization of LMFP cathodes through single or multi-metal doping is profoundly effective due to its ability to modify the fundamental phase transition mechanisms. The overarching trend is a doping-induced shift from a classical, sluggish two-phase reaction towards a faster, more homogeneous solid-solution-type process. This modulation mitigates structural strain and reduces polarization. More importantly, the primary benefit often lies in the dramatic enhancement of Li⁺ diffusion kinetics, which addresses the kinetic origins of voltage decay by ensuring faster transport and minimizing non-equilibrium concentration gradients. Therefore, the ideal doping strategy should aim not only to stabilize the crystal structure but also to create a favorable ionic transport landscape, ensuring that the phase transitions proceed in a more reversible and kinetically unhindered manner.^67–69

In the next sections, the main features related to the synthesis, structure, and commercialization of LMFP are discussed, including potential advantages, performance characteristics, and strategies for optimizing its properties to meet the demands of high-energy applications.

4. General synthesis protocols of multimetal LMFPs

The synthesis process is crucial for optimizing the performance of multimetal LMFP cathodes, with each method offering unique benefits and factors to consider for effective production. Fig. 5 shows the general synthesis methods for multimetal LMFP cathodes. A well-controlled synthesis is essential to enhance ion conductivity, improve structural stability, and achieve better cycling performance and energy density. In addition, factors such as precursor selection, synthesis temperature, and applied methodology significantly influence the crystallinity, morphology, and electrochemical properties of the prepared cathode materials.

Fig. 5 Overview of synthesis routes widely reported in the literature for the multimetal LMFP-based cathodes.

The solid-state method is currently the most widely used for the obtention of multimetal LMFP materials (Fig. 5). This method is favored for its simplicity, high yield, and suitability for large-scale industrial production. However, it requires high temperatures and extended processing times, and challenges remain in achieving uniform particle size, distribution, and morphology.⁷

Fig. 6 shows an illustration that represents a simple process of solid-state reaction. The process begins with the mixing and milling of solid compounds containing lithium, manganese, iron, and phosphorus, i.e., the raw materials (in a solid powder form), in order to obtain atomic-level uniformity. Then, the designed precursor material undergoes two stages of a high-temperature heating treatment in a controlled inert atmosphere. The precursor is pre-sintered at temperatures between 300 and 500 °C to decompose organic matter and remove structural water and then sintered at higher temperatures (600–850 °C) to promote the growth of olivine-type crystals and enhance their crystallinity.⁷ Typically, to facilitate the milling process, the reagents are suspended in a liquid medium, such as ethanol, which is subsequently evaporated.^72,73 In this case, this first step of the process may be considered as a mechanochemical liquid-phase process, in which the mechanical activation is performed in a liquid medium. This method involves the mixture of solid reactants with a small amount of liquid, subjecting the mixture to mechanical forces, such as milling or grinding.⁷⁴ The use of mechanical forces in a liquid medium facilitates interactions and introduces better control of the reaction conditions. In addition, the incorporation of organic carbon additives, such as sucrose, tends to improve the conductivity of LMFP cathodes. Mechanical or high-energy ball milling ensures the effective comminution and uniform dispersion of the carbon precursor, enabling the formation of a continuous, conductive carbon network on particle surfaces during high-temperature carbonization.¹² This carbon coating can boost electrochemical performance by mitigating particle agglomeration during milling, refining microstructural homogeneity, and facilitating more efficient charge transport.⁷⁵ V³⁺-doped LiMn_0.8Fe_0.2PO₄@C were synthesized through a solid-state high-temperature carbothermal reduction reaction ⁷⁶. In the procedure, the raw materials were weighed according to the stoichiometric ratio and then uniformly mixed through mechanical activation, with sucrose as a carbon source. Acetic acid was added during the activation to promote the oxidation of Fe and Mn to their +2 valence states. The mixture was ball-milled with a moderate amount of H₂O in a planetary ball mill using zirconia balls as a mill medium. After milling, the wet mixture was dried, resulting in a precursor, which was then calcined in an argon atmosphere, first at 450 °C for 8 hours, then heated to 650 °C for 8 hours, and finally cooled to room temperature. A resulting composite showed discharge capacities of 115.9 mAh g⁻¹ at 5.0C and high-capacity retention of 93.4% after 100 cycles at 1.0C.⁷⁶ The enhanced rate performance of the composites was attributed to the combined benefits of nanocrystalline particle incorporation and V³⁺-doping. In addition, the mechanical liquid-phase activation approach accelerated the reaction between LiH₂PO₄ and metal powders, while efficiently generating nano-sized, highly reactive precursors.

Fig. 6 Illustration of the solid-state reaction process.

Recently, the ball-milling technique combined with high-temperature solid-state reaction has been successfully employed to synthesize LiMn_0.6Fe_0.4−xNb_xPO₄/C cathode materials.⁷¹ LiH₂PO₄, Li₂CO₃, MnCO₃, FePO₄, Nb₂O₅, and 3 wt% sucrose were dispersed in anhydrous ethanol and subjected to high-speed ball-milling (600 rpm, 8 h) to induce liquid-assisted mechanochemical activation, ensuring intimate mixing, particle size reduction, and enhanced precursor reactivity. The resulting mixture was vacuum-dried and preheated at 350 °C for 3 h in Ar. Subsequently, the precursor was combined with an additional 2 wt% sucrose, ball-milled at 300 rpm for 8 h to achieve a uniform secondary carbon coating, vacuum-dried, and calcined at 650 °C for 10 h in Ar (5 °C min⁻¹), yielding fine, well-dispersed carbon-coated particles. This dual modification strategy, combining Nb⁵⁺ doping with a second-time carbon coating, significantly enhanced conductivity, structural stability, and electrochemical performance.⁷¹

The solvothermal process has emerged as the second most commonly used synthesis technique for LMFPs. In this process, a solution of the raw materials in the determined solvent is put in a stainless-steel autoclave reactor and kept under high pressure and temperature for a period of time. As a result, the morphology, particle size, and purity can be more precisely controlled.⁷⁷ Indeed, this process enables the production of materials with reduced particle sizes, uniform distributions, well-developed crystallinity, and reduced impurities. However, materials obtained through this approach frequently present limited electrochemical performance as a result of atomic disorder within the crystal lattice, which requires high-temperature annealing to restore structural alignment and improve their electrochemical performance. In addition, the solvothermal method faces challenges such as low reproducibility and the high cost of specialized reactors, limiting their scalability and feasibility for large-scale industrial production.⁷

A solvothermal synthesis was employed to produce Nb- and Sb-doped LMFP/C cathode materials,^78,79 where stoichiometric amounts of MnSO₄·4H₂O, FeSO₄·7H₂O, and the respective dopant (C₁₀H₅NbO₂₀ for Nb or C₆H₉O₆Sb for Sb) were dissolved in water, mixed with ethylene glycol, and combined with LiOH·H₂O and H₂PO₄. The mixtures were then heated at 180 °C for 5 hours in a Teflon-lined autoclave, followed by washing, drying, milling with glucose, and calcination at 700 °C in an argon atmosphere to obtain the final Nb- and Sb-doped Li(Fe_0.6Mn_0.4)_1−xPO₄/C cathode materials with varying dopant concentrations (x = 0, 0.01, 0.02, 0.03, 0.04, 0.05).^78,79 The Nb- and Sb-doped samples exhibited enhanced electrochemical performance compared to undoped LMFP/C, with the LMFP/C–Nb_0.02 and LMFP/C–Sb_0.04 samples showing superior cycling stability and rate capability.^78,79 In addition, doping improved lithium-ion diffusion and crystal structure stability, with Nb doping enhancing the metal–oxygen framework and Sb doping mitigating the Jahn–Teller effect.⁷⁹Fig. 7 shows a schematic representation of the synthesis used to obtain the Nb-doped LMFP/C cathode materials, effectively illustrating the key steps of a general solvothermal synthesis.

Fig. 7 Schematic representation of the synthesis of Nb-doped LMFP/C cathode materials. Reproduced with permission from ref. 78. Copyright © 2024 Published by Elsevier Ltd.

The hydrothermal method is similar to the solvothermal method. The basic difference between solvothermal and hydrothermal is due to the type of solvent used: organic solvent for solvothermal and water for hydrothermal. It is evident from the literature that the solvothermal method is preferred over the hydrothermal method for LMFP synthesis. This preference can be attributed to the greater flexibility offered by solvothermal processes, which utilize a range of organic solvents. These solvents allow for better control over reaction conditions, enhancing material properties and improving solubility. In contrast, hydrothermal methods are limited to water, thereby restricting the types of materials that can be synthesized. Furthermore, parameters such as temperature and pressure control, along with pH adjustment, are likely considered in the choice of synthesis method.

LiMn_0.79Fe_0.2Mg_0.01PO₄/C composite materials were synthesized via a hydrothermal method, where metal salts (MnSO₄·H₂O, FeSO₄·7H₂O, and MgSO₄·7H₂O) were dissolved in deionized water and purged with Ar to remove dissolved oxygen, then mixed with a colloidal solution of Li₃PO₄.⁸⁰ This mixture was transferred to a high-pressure hydrothermal reactor, where it was heated at 180 °C for 5 hours, followed by filtration, washing, and drying. The dried precursor was ball-milled with glucose and ascorbic acid in ethanol, then heated at 720 °C in an Ar atmosphere to yield the carbon-coated LiMn_0.79Fe_0.2Mg_0.01PO₄/C material.⁸⁰ The hydrothermal process described above is shown in Fig. 8. In this study, the influence of different sources of Li₃PO₄ on the performance of synthesized Mg trace LMFP cathodes was investigated, demonstrating that the choice of Li₃PO₄ precursor and the synthesis conditions directly dictated the particle morphology, size, and electrochemical performance. Li₃PO₄ with smaller crystal grains dissolves more readily due to its higher surface area, promoting nucleation of Mg-doped LMFP and resulting in smaller, more uniform particles. Conversely, larger Li₃PO₄ grains limit further LMFP formation, leading to residual Li₃PO₄ impurities. Thus, Mg-doped LMFP, prepared from well-dispersed, low-aggregation Li₃PO₄, exhibited the smallest particle size and uniform distribution. During co-sintering with glucose, the LMFP precursor enabled the formation of a uniform carbon coating and a larger specific surface area, which enhanced electron transport and lithium-ion diffusion. Consequently, this material exhibited superior lithium-ion kinetics, high-rate capability, and excellent cycling stability. These results clearly illustrate that controlling the hydrothermal synthesis, particularly the precursor characteristics, is essential for tailoring the structural and electrochemical properties of LMFP cathodes.⁸⁰

Fig. 8 Schematic representation of the hydrothermal synthesis process Reproduced with permission from ref. 80. Copyright © 2024 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

These effects on morphology and size, and consequently on electrochemical properties, are indeed expected, but they are not solely attributable to the intrinsic nature of the precursor properties. Dai et al.⁸¹ explored different synthesis variables, such as the anionic species of the Fe precursor (FeCl₂ (Cl⁻) and FeSO₄ (SO₄²⁻)) and the sequence of precursor addition: P + M + L sequence (LiOH solution dropwise introduced into the mixture of Li₃PO₄ and transition metal salts) or P + L + M sequence (Mn²⁺ and Fe²⁺ mixture slowly added into the mixture of Li₃PO₄ and LiOH). The proposed solvothermal synthesis proved highly effective in controlling the morphology of LMFP nanoparticles by adjusting the anion species and the precursor feeding sequence. The P + M + L sequence favored the formation of spindle-like nanostructures with reduced size and lower agglomeration, particularly when FeSO₄ was used as the iron source, leading to more uniform particles compared to FeCl₂. In contrast, the P + L + M sequence predominantly yielded nanoplates oriented along the ac plane, with lengths below 100 nm and thicknesses of about 20–30 nm. These morphological differences had a direct impact on electrochemical performance, as the particles obtained via the P + M + L sequence exhibited a reversible capacity of 129.7 mAh g⁻¹ at 0.1C and excellent rate, retaining 77% of their capacity at 1C.

Sol–gel and co-precipitation methods have been less commonly used for the synthesis of multielemental LMFPs. The sol–gel process^82,83 begins with the preparation of a colloidal solution (known as a sol) by dissolving metal salts or metal alkoxides in a given solvent, in which phosphoric acid (H₃PO₄) is added as the source of phosphate. The metal precursors in solution undergo hydrolysis, forming hydroxyl groups on the metal centers, and then they interact with the H₃PO₄ to form a metal-phosphate network through a condensation reaction. As the metal ions and phosphate ions interact, they form a gel-like structure, which is essentially a three-dimensional network of metal-phosphate bonds. To aid in this process, chelating agents such as citric acid are often used, which help stabilize the metal ions, provide a carbon source, and control gel formation.⁸⁴

As condensation progresses, the sol transitions into a gel. The gel is then aged to further strengthen its metal-phosphate framework and dried to remove excess solvent. The resultant material is calcined to remove organic residues and enhance the crystallinity and stability of the metal-phosphate structure. This method requires low synthesis temperature and provides excellent control over the reaction process, as well as the composition and structure of the final product.⁸⁵ Some studies^12,38 have highlighted that sol–gel processing allows precise tailoring of LMFP structure and morphology, directly linking synthesis conditions to electrochemical performance. By adjusting parameters such as precursor composition, carbon sources, and gelation conditions, several works have obtained nanoscale particles with uniform morphology, porous structures, and consistent carbon coatings, enhancing lithium-ion transport and electronic conductivity. Additives such as glucose, citric acid, or acetylene black further improve particle dispersion and form conductive networks, resulting in higher discharge capacity, rate performance, and cycling stability. However, according to Zhang et al.,⁷ the sol–gel method remains predominantly limited to laboratory research due to extended drying and processing times, complex procedural steps, high raw material costs, and potential hazards to human health and the environment, which hinder its applicability for large-scale industrial production. Fig. 9 provides a schematic representation of the steps involved in the sol–gel synthesis procedure. Qiao et al.⁸⁶ synthesized Li(Fe_0.65Mn_0.35)_0.98Mg_0.02PO₄/C, Li_0.98Na_0.02Fe_0.65Mn_0.35PO₄/C, and Li_0.98Na_0.02(Fe_0.65Mn_0.35)_1−xMg_xPO₄/C (x = 0.01, 0.02, 0.03, 0.05) composites via a sol–gel process modified by Na⁺ and Mg²⁺ doping. In this approach, stoichiometric amounts of manganese source (Mn(CH₃COO)₂·4H₂O), carbon source (citric acid), phosphorus source (NH₄H₂PO₄), iron source (FeCl₂·4H₂O), and lithium source (CH₃COOLi·2H₂O) were sequentially dissolved in deionized water. The solution was then supplemented with sodium (CH₃COONa·3H₂O) and magnesium (Mg(CH₃COO)₂·4H₂O) sources, followed by gel formation, drying, preheating, and calcination under a nitrogen atmosphere. The study demonstrated that doping with Na⁺ and Mg²⁺ enhanced electrochemical performance by reducing impedance, improving specific capacity, and increasing cycle stability.⁸⁶

Fig. 9 Schematic illustration of sol–gel synthesis.

The co-precipitation method involves preparing a solution of metal salts, to which a precipitating agent is added to initiate the precipitation of the desired compounds. The resulting precipitate is then filtered, washed, dried at moderate temperatures, and calcined at high temperatures to promote the formation of a crystalline material with the desired composition and structure. The co-precipitation method is simple and scalable, allowing precise control over the composition, phase, and morphology of active particles, which is crucial for optimizing electrochemical performance. However, achieving monodisperse particles with regular morphology requires careful control of reaction conditions, including pH, reagent concentrations, temperature, stirring rate, and the use of chelating agents. Furthermore, understanding the nucleation and growth processes is essential to predict and control the synthesis of crystalline precursor particles.⁸⁷ Indeed, the importance of carefully selecting and controlling the synthesis route to optimize particle size, morphology, and porosity, thereby directly enhancing the electrochemical properties of the cathodes. Li et al.⁸⁸ showed by preparing three Ni, Fe co-doped LiMnPO₄/C composite materials (LMP) using co-precipitation and solvothermal methods that the synthesis method critically influenced the morphology, microstructure, and electrochemical performance of the LMP. Co-precipitation produced irregular, large, and agglomerated particles of LMP, whereas solvothermal synthesis yielded more uniform rod-like particles. Glucose-assisted solvothermal synthesis further improved particle uniformity, producing small, spindle-shaped, porous structures, which shortened Li⁺ diffusion paths and enhanced electrolyte contact, resulting in higher discharge capacities (121.4 mAh g⁻¹ at 0.1C), superior rate performance (53.7 mAh g⁻¹ at 2C), and improved cycling stability (91% retention after 100 cycles at 1C).

For instance, LiMn_1/2Fe_1/4Ni_1/4PO₄ precursor was synthesized via a co-precipitation method, with Fe and Ni co-doping to enhance the electrochemical properties.⁸⁹ The schematic representation of the co-precipitation synthesis is shown in Fig. 10. The synthesis of the precursor involved dissolving MnSO₄·H₂O, FeSO₄·7H₂O, and NiSO₄·6H₂O in water according to a 2 : 1 : 1 molar ratio. The Fe²⁺ ions were oxidized to Fe³⁺ with hydrogen peroxide to facilitate phosphate nucleation, while Mn²⁺ and Ni²⁺ remained in their divalent states. The resulting mixture was then added dropwise, together with ammonia, into a combined solution of H₃PO₄ and NaOH, precipitating the phosphate precursor at pH 4. The precipitate was filtered, washed, dried, and then dispersed in a small amount of anhydrous ethanol, together with Li₂CO₃, rock sugar, and NH₄H₂PO₄, to synthesize the cathode material. During the calcination at 650 °C under N₂ with rock sugar, Fe³⁺ was reduced back to Fe²⁺, enabling its incorporation into the structure of LiMn_1/2Fe_1/4Ni_1/4PO₄C. The resulting LiMn_1/2Fe_1/4Ni_1/4PO₄C material exhibited a significantly higher initial charge–discharge capacity and excellent cyclic stability compared to LiMnPO₄/C, demonstrating the effectiveness of Fe–Ni doping in improving the electrochemical performance of the material.⁸⁹

Fig. 10 Co-precipitation method for synthesis of the LiMn_1/2Fe_1/4Ni_1/4PO₄ precursor, followed by milling and calcination for obtention of the LiMn_1/2Fe_1/4Ni_1/4PO₄ cathode material. Reproduced with permission from ref. 89. Copyright © 2021, The Author(s), under exclusive license to Springer-Verlag GmbH Germany, part of Springer Nature.

Techniques based on physical processes, such as freeze-drying and spray-drying, are commonly used to complement the methods described above.^40,90–95 These techniques play a crucial role in material synthesis, particularly in controlling particle size, homogeneity, and processing conditions. Freeze-drying is a method that removes water or other solvents from a sample by sublimation.^96,97 When pressure and temperature are adjusted to the triple point, even a slight increase in temperature will cause the solvent to sublimate directly.^77,96,97 This process avoids the need for high temperatures and ensures that delicate structures remain intact. In addition, it offers several advantages, including the ability to achieve lower calcination temperatures and ensure homogeneous distribution of reactants.^77,96,97 In the spray-drying process, solutions of raw materials (commonly metal salts) are atomized into fine droplets and rapidly dried to obtain dry powder known as a precursor.^98,99 After obtaining the precursor, it is typically subjected to an annealing or calcination step at higher temperatures to form the final product.⁹⁹ As well as freeze-drying, spray-drying provides specific advantages in terms of control over particle size, uniformity, and temperature, making it useful in various applications in material synthesis and processing.^98–100

It is important to emphasize that all of the methods described here use calcination as the final step. This high-temperature process facilitates the development of the desired material properties, including crystallinity, particle size, and phase formation. In addition, the different parameters of the synthesis, such as precursors, sintering temperature, solvent, and time, usually have significant effects on the morphology, chemical composition, and electrochemical performance of the cathode materials. Therefore, the choice of the process and, consequently, its parameters must be carefully considered to produce adequate cathode materials.

5. Multimetal LMFPs-based electrodes for LIBs

As previously described, the design of multimetal LMFP cathodes encompasses different approaches, including (i) doping with single or multiple elements, (ii) incorporating additional primary constituents or structural components, and/or (iii) enhancing the entropy of electrode materials. These approaches are particularly categorized based on the concept of doping, where the dopant concentration is typically limited to 5% by mass or atomic percentage. Beyond this threshold, the added metal ion functions as a structural component rather than a dopant. In fact, many authors in the literature mistakenly use the term doping when referring to the incorporation of a new metal ion at concentrations as high as 10 mol% or mass%. However, under these conditions, the introduced element should be regarded as a structural component of the composition rather than a dopant. Besides, incorporating large amounts of new elements into the LMFP structure is expected to induce major structural disorder. In fact, when incorporating new elements via isomorphic substitution, the presence of ions with different ionic radii can induce lattice distortion and increase structural disorder. Additionally, recent advancements in multimetal materials reveal that the design of medium- and high-entropy materials (MEMs and HEMs) can significantly improve their electrochemical and electrocatalytic properties.^101,102 Indeed, the high-entropy concept has emerged as a promising strategy and a hot topic in the design of novel active materials for lithium batteries.¹⁰³ According to statistical thermodynamics, the link between entropy and the composition of a system can be obtained from Boltzmann's equation. The molar configurational entropy of mixing (ΔS_config) for a crystalline phase containing multiple elements can be determined using eqn (1), where R is the universal gas constant (8.314 J K⁻¹ mol⁻¹) and x represents the molar fraction of each element.¹⁰⁴

Based on calculations obtained by the aforementioned equation, materials are classified as “low-entropy” (LEMs) when their S_config is below 1.0 R, as “medium-entropy” when S_config lies between 1.0 R and 1.5 R, and as “high-entropy” when the value is greater than 1.5 R.^104–106 Accordingly, the library of MEMs and HEMs has increasingly expanded to include olivine-type lithium metal phosphates,¹⁰⁷ a subject that will be examined in the following subsections.

This section explores the design strategies for multimetal LMFP electrode materials, focusing on the incorporation of foreign elements into different sites and the specific types of metal ions introduced. Interestingly, the progress shows that significant experimental advances highlight the use of isovalent ions (M²⁺), more specifically, Mg²⁺ and Ni²⁺. These two cations were reported by approximately 40% of the prepared multimetal LMFPs. On the other hand, among the aliovalent ions, Nb⁵⁺ and V³⁺ were the most studied (Fig. 11). In fact, the incorporation of at least one of these elements represents more than 50% of the compositions reported in the literature.

Fig. 11 Percentage of the main metal ions incorporated in multimetal LMFP cathode materials reported in the literature, with some materials incorporating one or two elements within the same active cathode structure. The LMFP structural model shown at the core of the scheme has been adapted with permission from ref. 12. Copyright © 2025 Wiley-VCH GmbH.

As outlined in the subsequent sections, all multielemetal compositions achieving capacity retention of 90% or higher after 200 or more cycles incorporate some form of carbon-based material (Fig. 12). In the carbon-coating approach, LMFP particles are surrounded by a carbon layer, which improves electrical conductivity, rate capability, and cycling stability, ultimately enhancing their overall electrochemical performance.^9,108 In fact, surface carbon coating is one of the most widely employed modification techniques for LMFP (referred to as LMFP/C), addressing key challenges such as low electronic conductivity, Mn dissolution, structural instability, and effectively creating a conductive network that enhances electron transport efficiency.^9,40 The following highlights the strategies, trends, and key findings, with a focus on the synergistic effects arising from incorporating various isovalent and/or aliovalent metal ions into the structure of carbon-coating-based LMFP cathodes, wherein the overall performance exceeds the sum of the individual contributions.

Fig. 12 Multimetal LMFP cathode materials with capacity retention ≥90% after 200 or more charge–discharge cycles. Maximum capacity discharge at 0.1C and or 0.2C. LiMn_0.59Fe_0.4Na_0.005Co_0.005PO₄/C ref. 90; LiFe_0.491Mn_0.499Nb_0.01PO₄/C ref. 94; LiMn_0.59Fe_0.4Mg_0.005Ni_0.005PO₄/C ref. 40; LiMn_0.6Fe_{0.3850□0.0075}Ti_0.0075PO₄/C ref. 70; LiMn_0.6Fe_0.38Nb_0.02PO₄/C ref. 71; LiMn_0.8Fe_0.155V_0.03□0.015PO₄/C ref. 109; LiMn_0.6Fe_0.395La_0.005PO₄/C ref. 110; Li(Mn_0.5Fe_0.5)_0.98V_0.02PO₄/MXene + C ref. 111; Li_0.98Mn_0.6Fe_0.4Nb_0.02PO₄/C ref. 67; Li_0.98Fe_0.6Mn_0.4Nb_0.02PO₄/C ref. 78; Ti/V–LiMn_0.6Fe_0.4PO₄/C ref. 112; 1Nb3Mg–LiMn_0.5Fe_0.5PO₄/C ref. 92; LiMn_0.6Fe_0.39Mg_0.01PO₄/C ref. 113; Li(Fe_0.6Mn_0.4)_0.96Sb_0.04PO₄/C ref. 79; LiMn_0.79Fe_0.2Mg_0.01PO₄/C ref. 80; LiNa_0.03Mn_0.8Fe_0.2PO₄/C ref. 114; LiMn_0.8Fe_0.19Ni_0.01PO₄/C ref. 115; Li_0.97Mg_0.015Mn_0.8Fe_0.2PO₄/C ref. 116; LiMn_0.6Fe_0.39Cu_0.01PO₄/C ref. 68; Li(Fe_0.5Mn_0.5)_1−3xMo_xPO₄/(x = 0.01)/C ref. 117; LiFe_0.47Mn_0.5Ca_0.03PO₄/C ref. 118.

5.1. Elemental doping strategy

The incorporation of elements by doping has been explored as an effective strategy to improve both electronic and ionic conductivity.¹⁰ In fact, elemental doping is essential for boosting the internal conductivity of LMFP-based electrode materials.⁸ In general, multi-element doping in different sites for LFP or LMFP, given their structural similarities, follows two main groups: single-site doping (Li-site or M-site) and dual-site doping (Li-site and M-site).^40,49,119 Besides, choosing the optimal doping ratio is vital to balance the benefits of Fe and Mn elements while preventing potential deviations or degradation in electrochemical performance, particularly from an excess of Mn.⁸

5.1.1. Cation doping with M-site substitution into LMFP. Fig. 11 highlights materials demonstrating capacity retention of 90% or higher after 200 or more cycles. Analysis of these results indicates that cation doping through M-site substitution in LMFP emerges as a key strategy for developing multi-elemental LMFP with enhanced stability during charge–discharge cycling and improved capacity at high rates. Interestingly, among the primary inactive metal ion dopants, Mg²⁺ and Nb⁵⁺ have been the most widely utilized, yielding significant performance improvements. Besides, Ni²⁺-doping has been explored in various studies, showcasing significant enhancements in performance.

Among divalent ions, the electrochemically inactive isovalent Mg²⁺ emerges as a promising candidate due to its cost-effectiveness, environmental advantages, and its ability to promote solid-solution behavior. By substituting at Fe²⁺/Mn²⁺ sites, Mg²⁺ can mitigate the effects of Jahn–Teller distortion and enhance structural stability in Mn-rich compositions. Notably, among cathode materials exhibiting capacity retention of 90% or higher after 200 or more cycles, 33% are Mg-containing LMFP compositions. The ionic radius of Mg²⁺ incorporated into the LMFP crystal lattice is smaller than that of Mn²⁺ and Fe²⁺, leading to two main structural modifications and reaction mechanisms. Firstly, there is a shortening of the bond length within the olivine MO₆ (M = Mn, Fe, Mg) octahedral structure. Conversely, the Li–O bond length in the LiO₆ octahedron increases. This elongation of the Li–O bond facilitates lithium-ion migration by widening the diffusion channel, thereby enhancing electrochemical reactions with improved rate capability and efficiency in multicomponent olivine structures.^7,113,120

Conversely, the formation of carbon-coated LMFP is a critical approach for achieving high-performance multimetal LMFP-based cathodes. As illustrated in Fig. 12, 100% of the materials are integrated with some form of carbon-based compound.

A typical example of the use of these approaches was employed by Duan et al.,¹²¹ who reported that an eco-efficient mechano-chemical liquid-phase activation method was employed to synthesize Mn-rich Li(Mn_0.9Fe_0.1)_1−xMg_xPO₄/C nanocomposites, optimizing synthesis temperature and Mg²⁺-doping ratio to enhance electrochemical performance. This method successfully produced nano-sized precursors, which were converted into Li(Mn_0.9Fe_0.1)_1−xMg_xPO₄/C composites via one-step calcination. Among the samples, Li(Mn_0.9Fe_0.1)_0.95Mg_0.05PO₄/C, synthesized at 700 °C with ∼3 wt% carbon additive, exhibited the best properties, delivering a specific discharge capacity of 153 mAh g⁻¹ at 0.1C and excellent cycling stability with 100% capacity retention after 100 cycles. Mg-doping reduced the lattice parameters, mitigating structural strain during lithium insertion/extraction and improving conductivity. However, excessive Mg-doping (e.g., 7%) led to unfavorable results, such as reduced electrochemical activity and higher resistance. Thus, 5% Mg²⁺ doping was determined to be optimal, balancing structural stability and electrochemical performance. These findings highlight the potential of Li(Mn_0.9Fe_0.1)_1−xMg_xPO₄/C as a competitive cathode material for advanced LIBs. Similarly, Hu and colleagues¹²² developed an Mn-rich LiMn_0.9Fe_0.09Mg_0.01PO₄/C material that exhibits impressive electrochemical performance, including a high discharge capacity of 160 mAh g⁻¹ at 0.1C and remarkable cycling stability at elevated temperatures. Even under a high discharge rate of 1C at 50 °C, the material achieves a capacity of approximately 135 mAh g⁻¹ with no capacity degradation after 120 cycles. This performance highlights the effective suppression of Mn dissolution in the electrolyte, ensuring enhanced structural and electrochemical stability of the electrode material.

Although the aforementioned studies demonstrate promising results, particularly in terms of high-capacity retention, further investigations into long-term cycling stability are essential. Extensive testing over a greater number of charge–discharge cycles is required to fully validate the advantages of Mg-doping, especially in the context of large-scale prototype applications. From this perspective, other Mg-doped LMFP cathodes have been developed, including Mn-rich compositions such as LiMn_0.79Fe_0.2Mg_0.01PO₄/C⁸⁰ and LiMn_0.6Fe_0.39Mg_0.01PO₄/C.¹¹³ These materials demonstrated impressive cycling stability, achieving capacity retentions of 97.07% at 1C after 200 cycles and 96.2% at 1C after 300 cycles, respectively.

Still considering 1% Mg-doping in Mn-rich LMFP, Kisu et al.¹²³ developed spherical LiMn_0.792Fe_0.198Mg_0.010PO₄ nanocrystals (10–40 nm) embedded within the interstices of supergrowth single-walled carbon nanotubes (SGCNTs) using an innovative in situ processing method called “ultra-centrifuging treatment” followed by brief heat treatment (Fig. 13a and b). Transmission electron microscopy (TEM) images revealed a direct attachment of these single-crystalline nanospheres to the highly conductive SGCNTs, forming a well-dispersed composite (Fig. 13c and d). This approach resulted in a 200% enhancement in Li⁺ diffusivity and a 50% reduction in charge transfer resistance, enabling ultrafast Li⁺ migration. These “nano–nano” LiMn_0.792Fe_0.198Mg_0.010PO₄/SGCNT composites overcame the limitations of one-dimensional diffusion. It delivered exceptional electrochemical performance, achieving a capacity of ∼54 mAh g⁻¹ per composite (equivalent to 77 mAh g⁻¹ per pure LiMn_0.792Fe_0.198Mg_0.010PO₄) at a high rate of 50 C (Fig. 13e). Moreover, the material demonstrated outstanding cycling stability, retaining 84% of its initial capacity after 3000 cycles (Fig. 13f). Interestingly, while the capacity retention is below 90%, the ability to sustain performance over 3000 cycles and operate at high power densities presents a promising opportunity for advancing high-power density devices.

Fig. 13 (a) TEM and (b) HRTEM images of LiMn_0.792Fe_0.198Mg_0.010PO₄/SGCNT composites and (c) their corresponding schematic illustration, with (d) 3D image of a LiMn_0.792Fe_0.198Mg_0.010PO₄ nanocrystal with the direction of Li⁺ diffusion. (e) Discharge capacity as a function of current density for (A, blue) LiMn_0.8Fe_0.2PO₄/SGCNTs and (B, red) LiMn_0.792Fe_0.198Mg_0.010PO₄/SGCNTs. Dashed lines represent the capacity of the composites, while solid lines indicate the capacity of the pure LiMn_0.8(1−z)Fe_0.2(1−z)Mg_zPO₄ (z = 0 or 0.01), excluding the contribution from SGCNTs in the composite. (f) Cycle-life testing of the LiMn_0.792Fe_0.198Mg_0.010PO₄/SGCNT composite over 3000 cycles. The inset highlights the charge–discharge profiles at the 1st, 100th, 500th, 1000th, 2000th, and 3000th cycles. Reproduced with permission from ref. 123. Copyrights © The Royal Society of Chemistry 2014.

Conversely, other isovalent ions have also been reported. Among them, Ni²⁺-doping has been explored in various studies, showcasing significant enhancements in performance. In fact, various doping percentages have been investigated, particularly in Mn-rich LMFP compositions (Table 1). One such study utilized a Ni-doping strategy to optimize the morphology of Mn-rich LiMn_0.8Fe_0.2PO₄ nanoparticles, promoting a preferred growth orientation that enhanced electrical and ionic conductivity as well as electrode kinetics.¹²⁴ In further detail, Ni-D0 (LiMn_0.8Fe_0.2PO₄ pristine) exhibits a mix of flake and ellipsoidal shapes with larger particles (∼53 × 120 nm) (Fig. 14a), while Ni-D5 particles (5 mol% Ni, LiMn_0.8Fe_0.15Ni_0.05PO₄) are ellipsoidal with an average size of 58 nm (Fig. 14c). Ni-D5 features a thin, uniform graphitized carbon coating (∼1 nm, Fig. 14d), whereas Ni-D0 has a thicker, amorphous layer (∼3 nm, Fig. 14b). In more detail, the Ni doping promotes a highly graphitized carbon layer, forming a conductive network that enhances electrical conductivity and Li⁺ transport, thereby improving electrochemical performance (Fig. 14e). In fact, compared to the pristine sample, the LiMn_0.8Fe_0.15Ni_0.05PO₄@C sample shows significant improvements, with electrical conductivity of 9.57 × 10⁻² S cm⁻¹ (Fig. 14f). Additionally, Ni doping induces the grains to grow along the (010) plane of LiMnPO₄, shortening Li⁺ diffusion paths and facilitating efficient ion transport. Furthermore, incorporating 5 mol% Ni into LiMn_0.8Fe_0.2PO₄@C reveals a typical two-phase transition characteristic of the olivine LiMnPO₄ structure with a solid-solution reaction, which ensures exceptional structural stability and reversibility.¹²⁴

Table 1 Electrochemical performance of multimetal LMFP-based electrodes for LIBs

Material	Coating material	Metal doping	Synthesis methods	Capacity retention (%)	First discharge capacity (mAh g^−1)	WVW (V vs. Li/Li^+)	Ref.
a Values were taken from images or graphs. b (SGCNTs) = interstices of supergrowth (single-walled) carbon nanotubes.
LiMn0.9Fe0.09Mg0.01PO4	C	Mg	Solid-state	—	141.0@0.1C	2.0–4.5	130
LiMn0.8Fe0.19Mg0.01PO4	C	Mg	Solid-state	—	138.5@1C	2.0–4.5	131
Li(Mn0.85 Fe0.15)0.92Ti0.08PO4	C	Ti	Solid-state	100% after 50 cycles@1C	144.4@1C	2.0–4.5	132
LiNa0.03Mn0.8Fe0.2PO4	C	Na	Solvothermal	96.65% after 200 cycles@0.5C	138.0@0.1C^a	2.5–4.5	114
Li0.98Na0.02(Fe0.65Mn0.35)0.97Mg0.03PO4	C	Na/Mg	Sol–gel	96.2% after 40 cycles@0.1C	147.7@0.1C	2.5–4.5	86
LiFe0.4Mn0.595Cr0.005PO4	C	Cr	Solid-state	99.2% after 50 cycles@0.1C	164.0@1C	3.5–4.1	133
Li(Mn0.9Fe0.10)0.95Mg0.05PO4	C	Mg	Mechanochemical liquid-phase activation technique	98.1% after 100 cycles@0.1C	153.0@0.1C	2.5–4.1	121
LiMn0.8Fe0.19Ni0.01PO4	C	Ni	Solvothermal	94.1% after 200 cycles@0.5C	163.0@0.1C^a	2.0–4.5	115
LiMn0.792Fe0.198Mg0.01PO4	SGCNT^b	Mg	Sol–gel	84.0% after 3000 cycles@1C	119.0@1C	2.5–4.5	123
LiFe0.95Ni0.02Mn0.03PO4	C	Ni	Solid-state	98.6% after 100 cycles@1C	145.4@1C	2.3–4.2	134
LiMn0.9(FeZn)0.05PO4	C	Zn	Solid-state	96.7% after 100 cycles@1C	153.0@0.1C	2.5–4.5	135
LiMn0.8Fe0.15Ni0.05PO4	C	Ni	Solvothermal	89.45% 100 cycles@0.1C	144.0@0.1C	2.5–4.5	124
Li0.97Mg0.015Mn0.8Fe0.2PO4	C	Mg	Solvothermal	93.5% after 200 cycles@0.5C	156.9@0.1C	2.0–4.5	116
(1−x)Li0.8Mn0.8Fe0.2PO4xLi3V2(PO4)3	LiVPO4 + C	V	Solid-state	93.4% after 100 cycles@1C	155.4@0.1C	2.5–4.5	76
LiMn0.79Fe0.2Mg0.01PO4	C	Mg	Hydrothermal	97.07% after 200 cycles@1C	134.7@0.1C	2.5–4.5	80
Li(Fe0.5Mn0.5)1−3xMoxPO4/(x = 0.01)	C	Mo	Solvothermal	91.2% after 200 cycles@2C	153.2@0.1C	2.0–4.5	117
LiMn0.59Fe0.4Na0.005Co0.005PO4	C	Co/Na	Solvothermal/spray-dryer	97.1% after 1000 cycles@1C	150.0@0.1C	2.5–4.5	90
1Nb3Mg–LiMn0.5Fe0.5PO4	LiNbO3 + C	Mg/Nb	Co-precipitation/spray-dryer	99.0% after 300 cycles@0.5C	∼140@0.1C	2.0–4.4	92
Li0.98Fe0.6Mn0.4Nb0.02PO4	C	Nb	Solvothermal	94.15% after 500 cycles@5C	156.9@0.2C	2.0–4.5	78
LiMnxFey(V)1−x−yPO4 (x = 1/3, y = 1/3)	C	V	Solid-state	98.6% after 50 cycles@0.5C	161@0.05C	2.0–4.5	73
LiMn0.48Fe0.5Mg0.02PO4	C	Mg	Solid-state	97.7% after 100 cycles@1C	160.7@1C	2.0–4.3	136
LiMn0.8Fe0.155V0.03□0.015PO4	C	V	Solid-state	98.9% after 600 cycles@0.2C	155.0@0.1C	2.0–4.5	109
LiMn0.9Fe0.05Co0.05PO4	C	Co	Solid-state	—	145.0@0.05C	2.5–4.8	137
LiMn0.9Fe0.05Mg0.05PO4	C	Mg	Solid-state	99.0% after 30 cycles@0.2C	140.0@0.1C	2.5–4.5	138
LiMn0.9Fe0.09Mg0.01PO4	C	Mg	Solid-state	95.0% after 120 cycles@1C	160.0@0.1C	2.5–4.6	122
LiFe0.47Mn0.5Ca0.03PO4	C	Ca	Solid-state	91.0% after 200 cycles@1C	156.1@0.1C	2.0–4.5	118
LiFe0.4Mn0.595Cr0.005PO4	C	Cr	Solid-state	99.2% after 50 cycles@0.1C	164.0 @0.1C	2.2–4.5	133
LiMn0.8Fe0.1Ni0.1PO4	C	Ni	Solid-state/freeze-drying	∼100% after 100 cycles@C/25	110@C/25	2.5–4.5	93
LiMn0.8Fe0.1Co0.1PO4	C	Co	Solid-state/freeze-drying	∼100% after 100 cycles@C/25	80@C/25	2.5–4.5	93
LiMn0.8Fe0.1Cu0.1PO4	C	Cu	Solid-state/freeze-drying	∼100% after 100 cycles@C/25	57@C/25	2.5–4.5	93
LiFe0.491Mn0.499Nb0.01PO4	C	Nb	Hydrothermal/spray-dryer	95.4% after 1000 cycles@1C	152@0.1C	2.5–4.3	94
Li(Fe0.50Mn0.50)0.88V0.08PO4	—	V	Solid-state	—	∼98.0@0.05C	2.0–4.7	139
LiMn0.9(FeZn)0.05PO4	C	Zn	Solid-state	96.7% after 100 cycles@1C	151.3@ 0.1C	2.0–4.5	135
LiFe0.4Mn0.595Cr0.005PO4	C	Cr	Solid-state	98.84% after 100 cycles@1C	163.6@0.1C	2.2–4.5	140
LiMn0.8Fe0.19Mg0.01PO4	C	Mg	Hydrothermal	—	137.0@0.1C	2.0–4.5	141
LiMn0.6Fe0.38Ni0.02PO4	C	Ni	Solid state/spray-dyer	98.3% after 100 cycles@1C	159.3@0.2C	2.5–4.5	95
LiMn0.6Fe0.39Mg0.01PO4	C	Mg	Solid-state	96.2% after 300 cycles@1C	159.6@0.2C	2.5–4.5	113
LiMn0.59Fe0.4Mg0.005Ni0.005PO4	C	Mg/Ni	Solvothermal/spray-dryer	95.0% after 1000@1C	152.0@0.1C	2.5–4.3	40
LiNa0.02Mn0.6Fe0.4PO4	C	Na	Carbothermal/spray-dryer	96.7% after 100 cycles@1C	153.1@0.2C	2.5–4.5	91
LiK0.03Mn0.05Fe0.95PO4	C	K	Solvothermal	97.0% after 100 cycles@1C	159.2@0.1C	2.5–4.2	127
Li(Fe0.6Mn0.4)0.96Sb0.04PO4	C	Sb	Solvothermal	93.2% after 300 cycles@5C	166.6@0.2C	2.0–4.5	79
LiFe0.48Mn0.48Mg0.04PO4	—	Mg	—	—	152.2@0.1C	2.7–4.4	49
LiMn0.8Fe0.19Mg0.01PO4	C	Mg	Solid-state	96.55% after 50 cycles 0.1C	145.0@0.1C	2.0–4.5	142
LiCo0.33Mn0.33Fe0.33PO4	—	Co	Solid-state	—	140.0@0.05C	2.0–5.0	72
Li(Mn0.5Fe0.5)0.98V0.02PO4/MXene	C	V	Solvothermal	96.19% after 500 cycles@1C	156.62@0.1C	2.5–4.5	111
Li0.995Nb0.005Mn0.85Fe0.15PO4	C	Nb	Solid-state	100% after 50 cycles@1C	166.0@0.12C	2.0–4.5	143
LiMn0.6Fe0.385□0.0075Ti0.0075PO4	C	Ti	Hydrothermal	92.55% after 1000 cycles@1C	139.7@0.2C	2.0–4.5	70
LiMn0.6Fe0.38Nb0.02PO4	C	Nb	High-temperature solid-phase	93.2% after 700 cycles@1C	158.3@0.1C	2.0–4.5	71
Li0.97Na0.03Mn0.8Fe0.2PO4	C	Na	Solvothermal	96.65% after 200 cycles@0.5C	141.7@0.05C	2.5–4.8^a	38
LiMn0.6Fe0.395La0.005PO4	C	La	High-temperature solid-phase	96.45% after 600 cycles@1C	158.61@0.1C	2.0–4.5	110
LiMn0.5Fe0.49Zn0.01PO4	C	Zn	Solvothermal	92.4% after 100 cycles@1C	141.75 @0.2C	2.0–4.5	144
LiFe0.5Mn0.49Y0.01PO4	C	Y	Carbothermal reduction of solid-state	87.74% after 300 cycles@1C	160.0@0.2C	2.0–4.5	145
Ti/V–LiMn0.6Fe0.4PO4	C	V/Ti	Co-precipitation/solid-state	93.81% after 500 cycles@1C	161.9@0.1C	2.0–4.5	112
Li0.98Mn0.6Fe0.4Nb0.02PO4	C	Nb	Co-precipitation/solid-state	95.07% after cycles 500@1C	155.63@0.1C	2.0–4.5	67
LiMn0.84Fe0.15Mg0.01PO4	C	Mg	Solid-state	80.8% after 500 cycles@1C	152.1@0.1C	2.5–4.5	146
LiMn0.6Fe0.39Cu0.01PO4	C	Cu	Solvothermal/solid-state	92.5% after 200 cycles@1C	160.3@0.1C	2.5–4 0.3	68
Li(Mn0.35Fe0.35Co0.1Mg0.1Ca0.1)PO4	—	Co/Mg/Ca	Mechanochemistry	78.0% after 20 cycles 0.1C	104.0@0.1C	2.0–4.9	107
Li(Mn0.2Fe0.2Co0.2Ni0.2Mg0.2)PO4	—	Co/Ni/Mg	Mechanochemistry	—	60.0@0.1C	2.0–4.9	107

---

Fig. 14 SEM and HRTEM images (and FFT patterns) of (a and b) Ni-D0 (pristine LMFP) and (c and d) Ni-D5. (e) Rate capability and (f) electrical conductivity measurements of the samples. Reproduced with permission from ref. 124 Copyrights © 2023 Elsevier B.V. All rights reserved.

Similarly, Tian et al.⁹⁵ reported that lattice distortion induced by an optimal level of Ni-doping effectively reduces the surface energy of LiMn_0.6Fe_0.4PO₄ crystals, limiting crystal growth and ensuring uniform particle size and shape. The resulting spherical LiMn_0.6Fe_0.4PO₄ cathode materials exhibit a smooth surface, consistent sphericity, ideal lattice volume, and high phase purity, contributing to outstanding cycling stability and rate performance. After 100 cycles at 1C, the LiMn_0.6Fe_0.38Ni_0.02PO₄ cathode retains 98.3% of its initial capacity, delivering 147.3 mAh g⁻¹. At higher rates of 10C and 15C, it achieves discharge capacities of 125.1 and 115.4 mAh g⁻¹, respectively.

Despite significant progress in the development of isovalent Mg²⁺- or Ni²⁺-doped LMFP, the dual doping (or co-doping) with Mg/Ni combined with carbon coating emerges as a highly promising strategy for achieving multimetal LMFP materials with excellent capacity retention, even after thousands of charge–discharge cycles. Using this strategy, a Mg/Ni dual-doped and carbon-coated LMFP micro-sized secondary sphere was developed (Fig. 15a), exhibiting significantly enhanced reaction kinetics and exceptional structural stability.⁴⁰ Ni²⁺ contributes to charge compensation during cycling, reducing the average valence state of Mn ions to mitigate Jahn–Teller distortion caused by Mn³⁺ while enhancing electronic conductivity in synergy with the carbon coating (Fig. 15c). Meanwhile, Mg²⁺, with its smaller ionic radius, expands the Li-ion transport channels by elongating the Li–O bonds. As a result, MgNi-LMFP/C achieves an impressive reversible capacity of 152 mAh g⁻¹ at 0.1C, surpassing its counterpart, which delivers 148 mAh g⁻¹ under the same conditions (Fig. 15b). Besides, this optimized LMFP cathode achieves a reversible specific capacity of 115 mAh g⁻¹ at 5C, approximately 2.4 times higher than the pristine LMFP sample (Fig. 15d). Extended cycling tests conducted at a current density of 1C (Fig. 15e) demonstrate that MgNi-LMFP/C retains 95% of its capacity after 1000 cycles, significantly outperforming LMFP/C, which achieves only 88% retention. In a pouch-type full cell, it retains 92% of its initial capacity after 2000 cycles at 1C (Fig. 15f), underscoring its promising potential for high-power and long-life LIB applications.

Fig. 15 (a) TEM image and EDS elemental mapping of the MgNi-LMFP/C sample. (b) Initial charge/discharge profiles at 0.1C. (c) Schematic representation of the crystal structures for MgNi-LMFP/C and LMFP/C in their charged states. (d) The rate capabilities of the four samples. (e) Cycling performance over 1000 cycles at 1C for MgNi-LMFP/C and LMFP/C. (f) Capacity retention after 2000 cycles at 1C for MgNi-LMFP/C and LMFP/C in pouch-type full cells. Reproduced with permission from ref. 40. Copyrights © The Royal Society of Chemistry.

Conversely, as previously emphasized, the Nb⁵⁺ cation stands out as a highly promising aliovalent dopant. Its charge disparity introduces vacancies into the crystal lattice, effectively increasing carrier concentration. This adjustment not only enhances the structural stability of LMFP by mitigating the Jahn–Teller effect but also significantly boosts its intrinsic conductivity.⁸ In fact, Nb-doping not only delivers a high initial specific capacity but also ensures capacity retention exceeding 90% over charge–discharge cycles as high as 500–1000. Similarly, Kong et al.⁷⁸ demonstrated that Nb-doping facilitates the growth [001] crystal orientation of the material, thereby enhancing lithium-ion diffusion, reducing electrode polarization, and lowering charge transfer resistance, all contributing to its superior rate performance (Fig. 16a and b). Thus, a carbon-coated Li(Fe_0.6Mn_0.4)_1−xNb_xPO₄ cathode material with x = 0.02 (named LFMP/C–Nb0.02), exhibited notable improvements in rate performance. The LFMP/C–Nb0.02 cathode achieved a high specific capacity of 156.9 mAh g⁻¹ at 0.2 C and 121.4 mAh g⁻¹ at 5C. Additionally, Nb doping increases the M–O bond energy, stabilizing the LFMP/C metal–oxygen framework and reinforcing the resilience of the crystal structure during lithium-ion intercalation and deintercalation cycles. A capacity retention rate of 94.15%, even after 500 cycles, significantly surpasses the original LFMP/C, which maintained only 61.47% of its capacity after 300 cycles.

Fig. 16 (a) Cycling stability of all samples tested at 5C, and (b) discharge capacity retention comparison between LFMP/C and LFMP/C–Nb0.02 across varying rates. Reproduced with permission from ref. 78. Copyrights © 2024 Published by Elsevier Ltd (c) SEM micrograph, (d) high-magnification view, (e) high-resolution TEM image, and (f) TEM-EDS elemental mapping of LMFP/C–1%Nb. The inset in (e) displays the corresponding FFT pattern. (g) Mn discharge capacity retention across various rates, and (h) cycling performance at 1C for pristine LMFP/C and LMFP/C–Nb samples with varying Nb-doping levels. Reproduced with permission from ref. 94. Copyright © 2023 American Chemical Society.

Analogously, Jin and collaborators⁹⁴ synthesized a 1% Nb-doped carbon-coated LiFe_0.5Mn_0.5PO₄ (LMFP/C–1%Nb) cathode material using a combination of hydrothermal and spray-drying techniques. Microscopy techniques revealed the microstructure of LMFP/C–1%Nb samples, showing secondary microspheres (1–4 μm) composed of tightly packed nanoparticles (50–150 nm) (Fig. 16c and d). High-resolution TEM confirmed favorable crystallinity with lattice fringes matching LMFP planes and a 1.5 nm graphitized C-coating layer (Fig. 16e). Elemental mapping verified the uniform distribution of Mn, Fe, P, O, and Nb, unaffected by Nb-doping levels (Fig. 16f). The mechanism of action for the two strategies in LMFP operates through different pathways. The carbon coating is thought to create a highly conductive network between the primary particles while also preventing secondary reactions by shielding the material from direct electrolyte exposure. The strong Nb–O coordination exhibits higher bond energy compared to Mn–O and Fe–O, thus improving the stability of the crystal structure in LMFP materials.⁹⁴ This helps suppress the Jahn–Teller effect of Mn³⁺, while simultaneously accelerating ion diffusion and electron transport within the nanoparticles. Thus, the dual modification strategy significantly enhanced both the lithium-ion diffusion coefficient and the electronic conductivity of LMFP, resulting in faster reaction kinetics. As a result, the optimized LMFP/C–1%Nb cathode exhibited outstanding electrochemical performance, delivering a lithium storage capacity of 115 mAh g⁻¹ at 5 C and achieving exceptional capacity retention of 95.4% after 1000 cycles at 1 C (Fig. 16g and h).

Recently, advanced studies using in situ techniques have revealed fundamental insights into understanding the contributions of aliovalent ion doping (and/or carbon coating).⁷¹ For instance, in situ XRD analyses were carried out on carbon-coated and Nb-doped LMFP composites (named LMFP/DC-2Nb) and LMFP/C to gain deeper insights into the structural changes of LMFP throughout the lithiation and delithiation processes (Fig. 17a). Notably, LMFP/DC-2Nb exhibits a markedly reduced two-phase region compared to non-doped LMFP/C, with its fraction decreasing from 32% to 27%, indicating a more uniform phase transition and enhanced reaction consistency (Fig. 17b and c). DFT calculations further demonstrate that introducing an optimal amount of Nb⁵⁺ strengthens M–O bond energy, enhances structural stability, and boosts the material's intrinsic conductivity by narrowing its band gap. Still by in situ XRD, Gao et al.⁷⁰ demonstrated a similar behavior in a Ti⁴⁺-doped LMFP (designated 0.75% Ti-LMFP/C) with high-rate performance, where the two-phase reaction region (L_xMFTP → MFTP) becomes narrower through the combined effects of aliovalent ion doping and Fe vacancies (Fig. 17d–f).

Fig. 17 (a) Structural evolution of LMFP/DC-2Nb throughout the charge–discharge cycle. (b) The initial charge–discharge profile at 0.2C over the 2.0–4.5 V range, along with representative in situ XRD patterns. (c) Rate performance at 1C for LMFP/DC-xNb samples (x = 0, 0.5, 1, 2, and 3) compared with LMFP/C. Reproduced with permission from ref. 71. Copyright © 2025 Elsevier Ltd All rights are reserved, including those for text and data mining, AI training, and similar technologies. In situ XRD patterns for (d) LMFP/C and (e) 0.75% Ti-LMFP/C electrodes, and (f) the corresponding phase transition schematic during the charge–discharge process of 0.75% Ti-LMFP/C. Reproduced with permission from ref. 70. Copyright © 2025 Elsevier Ltd All rights are reserved, including those for text and data mining, AI training, and similar technologies.

On the other hand, aliovalent Nb⁵⁺ ions have also been incorporated into coating materials to enhance the stabilization of multimetal LMFP structures. In fact, the use of high-valence ions such as Nb, Ti, V, Zr, and Si has been shown to form oxide-based layers, which can influence the lithium-ion diffusion kinetics and improve stability over cycles by increasing the reactivity of the surface between the cathode and electrolyte.^94,125 In this context, P. Vanaphuti & A. Manthiram⁹² reported that LiNbO₃ predominantly resides on the particle surface, while isovalent Mg²⁺ doping occurs within the bulk structure. The LiNbO₃ surface coating enhances ionic conductivity, leading to a synergistic effect that provides higher specific capacity compared to pristine, Mg-doped, or Nb-doped LMFP. Apart from the higher specific capacity observed in the doubly modified sample, all samples demonstrate outstanding cycling stability over 200 cycles at a C/3 rate, maintaining a capacity retention of ≥98%. This remarkable performance is attributed to improved ionic/electronic transport from the LiNbO₃ coating and accelerated Mn^2+/3+ redox kinetics due to Mg bulk doping. Besides, Mg-containing LMFP with Nb-based coating can promote the formation of Fe₂P, which enhances the material's conductivity and facilitates electronic transport in the cathode. Overall, the study highlights the critical role of synergistic dopants and coating materials in enhancing the capacity and cycling stability of LMFP cathodes.

5.1.2. Cation doping with Li-site substitution into LMFP. As previously mentioned, Li-site doping in LMFP cathodes is another possible strategy, resulting in a multi-elemental compound with generally improved electrochemical properties, especially when the newly introduced metal is present at low doping concentrations. In fact, doping at the Li-site significantly affects the diffusion of Li⁺, increasing the available one-dimensional pathway for its movement.^90,114,126 Interestingly, and in a manner similar to results previously reported for LFP studies, the Li-site can be doped with various metal ions, including alkali metals, alkaline earth metals, and transition metals. For instance, different isovalent and aliovalent metal ions have been used in LMFP Li-sites, such as K⁺, Na⁺, Mg²⁺, V³⁺, Nb⁵⁺, etc. (Table 1).

The isovalent alkaline metal ions (Na⁺ and K⁺) have the same valence state as Li⁺, which does not alter the charge state of the Li-site in the olivine structure.¹⁰ For instance, Xu et al.⁹¹ highlighted the significant impact of Na⁺ doping on the electrochemical performance of Li_1–xNa_xMn_0.6Fe_0.4PO₄/C (x = 0, 0.01, 0.02, 0.03) materials. A combination of experimental analysis and density functional theory (DFT) calculations demonstrated that Na⁺ doping enhances the crystallinity, reduces Li–Fe antisite defects, decreases primary particle size, and homogenizes particle size distribution without altering the olivine structure. The inferior rate and cycling performance of undoped LMFP were attributed to the sluggish Li⁺ diffusion kinetics associated with Mn redox processes. Na⁺-doping effectively increases the Li–O bond length, broadens Li⁺ diffusion channels, and lowers diffusion energy barriers, leading to accelerated Li⁺ diffusion and improved Mn redox kinetics. Additionally, the structural role of Na⁺ as a stabilizing pillar reduces Mn³⁺ content and weakens Mn–Mn interactions, mitigating the Jahn–Teller effect. These combined effects significantly enhance the high-rate capability and cycling stability of the Na⁺-doped materials. Among the tested compositions, Li_0.98Na_0.02Mn_0.6Fe_0.4PO₄/C (x = 0.02) exhibited the best performance, delivering a specific discharge capacity of 125.0 mAh g⁻¹ at 5C and retaining 96.7% of its capacity after 100 cycles at 1C.

Still for Na⁺-doping, Li et al.¹¹⁴ designed Li_1−xNa_xMn_0.8Fe_0.2PO₄/C nanocapsule synthesized by a solvothermal method. Interestingly, it was observed that the doping of Na⁺ can effectively inhibit the dissolution of Mn²⁺ and Fe²⁺ in the electrolyte, a critical issue in maintaining cathode stability. The doping process, confirmed by EDS and XPS analyses, did not induce cation rearrangement in the structure, ensuring the structural integrity of the material. The regular diffraction lattice observed in Li_0.97Na_0.03Mn_0.8Fe_0.2PO₄/C confirms the well-developed crystal structure, while ex situ XRD patterns after 200 cycles demonstrate remarkable structural stability. Importantly, Na⁺ doping does not alter the electrochemical reaction potential, ensuring compatibility with the inherent properties of the LMFP framework. In fact, electrochemical studies reveal that a 3% Na⁺ doping level achieves optimal performance, delivering specific capacities of 141.7, 125.0, and 89.5 mAh g⁻¹ at 0.05C, 1C, and 5C rates, respectively. This improvement is attributed to the enhanced lithium-ion diffusion coefficient and the structural and chemical stability imparted by Na⁺-doping. Additionally, the capacity retention after 200 cycles at 0.5C remains at an impressive 96.65%, indicating excellent cyclic stability. The nanocapsule morphology further facilitates fast lithium-ion diffusion, contributing to the material's superior rate performance. In summary, Na⁺-doping in strategy effectively inhibits the dissolution of Mn²⁺ and Fe²⁺ in the electrolyte while maintaining the structural integrity of the cathode, making it a promising approach for next-generation LIBs.

Analogously to Na⁺-, K⁺-doping provides structural stability by mitigating volume changes during cycling and suppressing the Jahn–Teller effect of Mn, thereby enhancing the long-term durability of cathode material.¹²⁷ This stabilization is evidenced by an increased intensity of the diffraction related to the (111) plane, attributed to lattice growth induced by K⁺. Due to its larger ionic radius, K⁺ induces significant lattice distortion when substituted at Li-sites, widening ion diffusion pathways and acting as a structural pillar. Furthermore, the introduction of point defects at the Li-site lowers the activation barrier for Li⁺ migration, thereby facilitating its transport and improving electrochemical performance. In fact, thanks to the widened ion channels and enhanced structural stability induced by doping, the KLMFP cathode delivers an impressive capacity of 145 mAh g⁻¹ at 5C and maintains 96% of its capacity after 400 cycles. However, excessive K⁺ incorporation can disrupt the lattice structure, degrade cycling performance, and partially occupy Li⁺ sites, negatively impacting electrochemical behavior.¹²⁷ On the other hand, it is important to highlight that the K⁺-containing LMFP reported by Geng et al.¹²⁷ can be considered a dual-doped material, as it also incorporates Mn ions at the M-site with an atomic percentage of 5%.

Although significant advancements have been made through Na⁺- and K⁺-doping, experimental results indicate that only limited amounts of these dopants positively impact electrochemical performance.¹⁰ When their concentration exceeds a critical threshold, both discharge capacity and cycling stability deteriorate. This decline can be attributed to the substantially larger ionic radii of Na⁺ (1.02 Å) and K⁺ (1.31 Å) compared to Li⁺ (0.68 Å), leading to structural distortions that compromise the crystal integrity.^10,90

Beyond isovalent doping, a recent approach involves incorporating aliovalent ions, with supervalent doping emerging as a growing trend in certain cases. In fact, aliovalent doping can serve as a strategic approach to introducing controlled atomic disorder within the ordered olivine structure, ultimately enhancing battery performance.¹²⁸ A noteworthy case that still requires thorough investigation is Li-site doping with Mg²⁺ ions. For instance, while most studies report Mg²⁺-doping in LMFP at the M-site, research has also demonstrated the potential for Mg²⁺ incorporation at the Li-site.¹²⁸ In a study by Hu et al.,¹¹⁶ it was observed that with increasing Mg-doping, the LMFP-based samples exhibited a slight shift toward lower angles (Fig. 18a). This shift occurs because the incorporation of Mg²⁺ at the Li-site extends the Li–O and O–O bond lengths, a consequence of lithium vacancy formation under the charge compensation mechanism, ultimately causing expansion of the olivine unit cells (Fig. 18b). This expansion increases the cell volume and broadens the one-dimensional Li⁺ diffusion channels, thereby enhancing the electrochemical performance of the material. Among the synthesized samples, Li_0.97Mg_0.015Mn_0.8Fe_0.2PO₄ (LMFP-2) exhibits the highest reversible capacity (Fig. 18c) and cycle stability (Fig. 18d), with a discharge capacity of 149.8 mAh g⁻¹ at 0.5C, which remains at 140.1 mAh g⁻¹ after 200 cycles, corresponding to a retention rate of 93.5%. These findings suggest that Mg²⁺ substitution at the Li⁺-site effectively enhances electronic conductivity and Li⁺ mobility (as confirmed by EIS, Fig. 18e), thereby improving overall electrochemical performance. Furthermore, the superior rate capability highlights its potential for high-performance LIBs. The solvothermal synthesis of Mg-doped LMFP thus offers a promising pathway for the sustainable development and optimization of olivine-structured LFP materials.

Fig. 18 (a) The diffractograms of four LMFP-based samples. (b) Crystal structure of the LMFP-0 (top) and LMFP-2 (bottom). (c) Rate capability, (d) cyclability at 1C, and (e) Nyquist plots of pristine LMFP and Mg-doped LMFP cathode materials. Reproduced with permission from ref. 116. Copyright © 2023 Elsevier Ltd all rights reserved.

Interestingly, studies suggest that supervalent dopants smaller than the host Li⁺ and Fe²⁺ tend to occupy the Li-site preferentially rather than replacing ions at the M-site in LFP-based materials.^128,129 In fact, the olivine lattice can accommodate aliovalent cations ranging from 2+ to 5+, with charge compensation occurring through Li vacancies as one of the possible defect mechanisms.¹²⁸ Although progress in the design of supervalent ion doping into the Li-site of LFP has been reported, the progress for doping in LMFP still needs to be studied, demonstrating that there is still plenty of room for further development.

5.1.3. Cation co-doping with Li-site & M-site substitution into LMFP. The investigation of multimetal incorporation in olivine-type materials has gained attention in recent studies due to its potential to improve the electrochemical performance of LMFP cathodes. In fact, the rational design in the selection of dopant ions can lead to synergistic effects, contributing to improvements, particularly in electronic conductivity, which consequently enhances the power density of devices, as well as the stability of LMFP-based cathodes during charge–discharge cycles.

To date, only two studies have reported a dual-site doping approach, indicating that the design of multimetal LMFP by this strategy is still in its infancy. For instance, recently, Wang et al.⁹⁰ employed a dual-site doping design to enhance both ionic and electronic conductivities by simultaneously incorporating Na and Co ions into LMFP-electrode material. As shown in Fig. 19a, the Na and Co ions in LMFP act synergistically within the olivine structure. The large ionic radius of Na⁺ (1.02 Å) enhances the Li⁺ transport dynamics by expanding the Li⁺ channel through its placement at the Li-site. Meanwhile, Co²⁺ ions positioned at the M-site reduce the band gap, improving electronic conductivity, and mitigating the increase in the b-axis parameter, thereby shortening the Li⁺ transport path.⁹⁰ Interestingly, the LMFP-Co/Na material exhibited a decrease in particle size from 91.8 to 75.6 nm compared to the undoped LMFP. This reduction in primary particle size can lead to superior kinetic performance, as the Li⁺ diffusion path is shortened. In Fig. 19b, the spherical shape and the homogeneous dispersion of Fe, Mn, Co, and Na elements in the material can be observed through TEM-EDS analysis. The LMFP modified by double doping exhibits excellent specific capacity values at a rate of 0.1C and a coulombic efficiency of 95.5% (Fig. 19c). In addition to the improved specific capacity at low rates, the material also demonstrates good capacity retention even at higher rates, whereas pure LMFP, when tested at a high rate such as 1C, showed a significant decrease in specific capacity (Fig. 19d). The excellent cycling stability and specific capacity retention of 97.1% after 1000 cycles at a rate of 1C demonstrate the effectiveness of the modification in the LMFP structure in mitigating the challenges of the Li⁺ diffusion path within the crystal lattice of the cathode (Fig. 19e). Thus, the synergistic effect of the Na/Co double doping led to a notable improvement in the electrochemical performance of the olivine cathode. This opens a promising new pathway for the implementation of novel multi-element co-doping combinations at multiple sites in LMFP.⁹⁰

Fig. 19 (a) Schematic diagram of LMFP modified by Na/Co dual-site doping approach; (b) HRTEM images of LMFP-Co/Na and EDS mapping images of LMFP-Co/Na; (c) first charge/discharge curves, (d) rate capability, and (e) ionic conductivity of LMFP, LMFP-Co, LMFP-Na, and LMFP-Co/Na. Reproduced with permission from.⁹⁰ Copyright © 2024, the Author(s), under exclusive license to Springer-Verlag GmbH Germany, part of Springer Nature.

Similarly, Qiao et al.⁸⁶ synthesized Li_0.98Na_0.02(Fe_0.65Mn_0.35)_1−xMg_xPO₄/C (x = 0.01, 0.02, 0.03, 0.05) using the sol–gel method, incorporating Na⁺ and Mg²⁺ through a dual-site doping strategy. Their findings indicate that when x = 0.03, the material exhibits the most favorable electrochemical performance. The initial discharge capacity at 0.1C reaches an impressive 147.7 mAh g⁻¹. Even after 40 cycles, when the current returns to 0.1C, the discharge capacity remains as high as 142.1 mAh g⁻¹. These results suggest that optimized Na⁺ and Mg²⁺ doping enhances electrochemical properties, primarily by improving the Li⁺ diffusion coefficient and reducing charge transfer resistance, facilitating ion transport, and promoting electrochemical reactions.

5.2. Elemental incorporating structural components and/or entropy strategy

As highlighted previously, when the dopant concentration exceeds 5% by mass or atomic percentage, it should be regarded as a primary constituent or structural component of the electrode material. Thus, at high concentrations of the third metal ion, the term “doping” becomes inaccurate and may lead to confusion or misunderstanding. Therefore, this subtopic will mainly discuss advances in the incorporation of other ions at concentrations exceeding 5 mol% into LMFP cathodes. Based on these observations, LMFP cathodes incorporating a third structural component should be classified as ternary electrode materials.

Among the reported ternary electrode materials, metal ions such as Ni²⁺, Co²⁺, Cu²⁺, V³⁺, Ti⁴⁺, and Nb⁵⁺ have been incorporated into the olivine structure of LMFP (Table 1). It is essential to mention that even though such materials are not highlighted in the figure of merit, many of these ternary electrode materials have demonstrated high-capacity retention after 50–100 charge–discharge cycles.

In one of these interesting studies, Iturrondobeitia et al.⁹³ designed three different ternary Mn-rich LiMn_0.8Fe_0.1M_0.1PO₄/C (M = Co²⁺, Ni²⁺, and Cu²⁺, isovalent metal ions) cathode materials via a freeze-drying method. The morphological characterization of ternary LiMn_0.8Fe_0.1M_0.1PO₄/C composites revealed that nanoparticles are embedded within an amorphous carbon matrix. This structural feature plays a crucial role in enhancing the electrochemical performance of these materials. Among the synthesized compositions, LiMn_0.8Fe_0.1Ni_0.1PO₄/C emerged as the most promising cathode, demonstrating superior electrochemical properties (Fig. 20a–c). For instance, the partial substitution of 0.1 mol of Fe²⁺ with Ni²⁺ led to an increase in specific capacity, reaching 110 mAh g⁻¹ (Fig. 20a), while the incorporation of Co²⁺ resulted in a moderate capacity improvement, reaching 80 mAh g⁻¹. These enhancements are attributed to several synergistic factors: the favorable morphological characteristics, the conductive network formed by the in situ generated carbon, and the mitigation of structural stress within the olivine framework due to the presence of Ni²⁺ and Fe²⁺, which contribute to improved stability and charge transport. In contrast, the incorporation of Cu²⁺ led to a marked reduction in specific capacity, reaching only 57 mAh g⁻¹. This decline is likely attributed to the instability of the LiCuPO₄ olivine phase, which impairs lithium-ion diffusion and structural integrity. The rate capability of LiMn_0.8Fe_0.1Ni_0.1PO₄/C was 95% of its initial specific capacity at 1C (Fig. 20c). Similarly, LiMn_0.8Fe_0.1Co_0.1PO₄/C exhibited commendable rate performance, maintaining 93% capacity retention at 1C. Conversely, LiMn_0.8Fe_0.1Cu_0.1PO₄/C displayed the lowest capacity retention of 85% at 1C, corroborating its higher polarization and inferior energy storage capability. These insights underscore the importance of compositional tuning in optimizing olivine-based cathode materials for high-performance LIBs.

Fig. 20 (a) Initial charge–discharge curves and (b) cycling stability of the four evaluated LiMn_0.8Fe_0.1M_0.1PO₄/C (M = Fe, Co, Ni, Cu) composites at a C/25. (c) Corresponding modified Peukert plot. Reproduced with permission from ref. 93. Copyright © 2015 American Chemical Society (d) Initial charge/discharge profiles at 1C, (e) cycling stability at 1C, and (f) rate capability at varying C-rates for pristine and aliovalent cation-doped LMFP. Reproduced with permission from ref. 132. Copyright © 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

In addition to isovalent ions, alliovalent structural components have also been incorporated into the M-site in LMFP cathodes. For instance, the incorporation of Ti⁴⁺ at the M-site significantly enhances the electrochemical performance of LMFP due to the synergistic interaction between Ti⁴⁺ and Fe²⁺ (ref. 132) (Fig. 20d–f). This interaction contributes to improved structural stability of the olivine framework, facilitates lithium-ion diffusion, and enhances electrochemical kinetics. When compared to LiMn_0.85Fe_0.15PO₄/C and LiMn_0.92Ti_0.08PO₄/C, the Li(Mn_0.85Fe_0.15)_0.92Ti_0.08PO₄/C composition demonstrates a significantly higher discharge capacity and superior rate capability. It achieves a capacity of 144.4 mAh g⁻¹ with an exceptional retention rate of 99.9% after 50 cycles at 1C, highlighting its excellent cycling stability and rate performance. These findings confirm that the dual incorporation of an aliovalent cation at the M-site is an effective strategy for enhancing the electrochemical properties of LMFP-based cathodes.¹³²

Even higher proportions of the third metal ion have been incorporated into LMFP cathodes, with reported concentrations reaching close to equimolarity at the M-site. Compounds with these characteristics are common among layered metal oxides for LIBs, such as LiNi_xCo_yMn_zO₂ with x = y = z = 0.33 (NCM111). In fact, the increase in entropy has also extended to other types of materials for cathodes or catalysts, being a recent trend in sodium-ion¹¹ and Li–S batteries.¹⁰² Thus, the design of electrode materials with equimolar proportions of metal ions is an important strategy, especially in the design of electrode materials with higher electrochemical stability, thereby configuring desirable medium- and high-entropy materials (ME- and HEMs).^11,102 For instance, a medium-entropy lithium multi-transition metal phosphate, ternary LiCo_1/3Mn_1/3Fe_1/3PO₄, has been successfully synthesized via a solid-state reaction, exhibiting an olivine-type crystal structure.⁷² This material demonstrates a high specific capacity of 140 mAh g⁻¹ and an operating voltage of 3.72 V at a 0.05C discharge rate, attributed to the redox activity of the Mn²⁺/Mn³⁺ and Co²⁺/Co³⁺ couples. In situ XRD analysis confirms that medium-entropy LiCo_1/3Mn_1/3Fe_1/3PO₄ maintains structural stability even at 5 V during charge/discharge cycles. While all three cations (Co²⁺, Mn²⁺, and Fe²⁺) occupy the same 4c site, the Co²⁺/Co³⁺ redox couple contributes a relatively lower capacity of approximately 20 mAh g⁻¹. This reduced contribution is attributed to the significantly lower lithium-ion diffusivity associated with the Co redox process compared to Fe²⁺/Fe³⁺ and Mn²⁺/Mn³⁺. It is also important to emphasize that, while increased entropy can enhance structural stability, it may come at the cost of reduced specific capacity when compared with low-entropy LFP or LMFP. Therefore, achieving an optimal balance between stability and overall electrochemical performance is crucial.

On the other hand, Fan et al.¹⁰⁷ employed the concept of entropy enhancement to highlight the potential for increasing the capacities of other olivine-type lithium metal phosphates, such as LiMnPO₄ (LMP) and LiNiPO₄ (LNP), despite their inherently insulating nature. They adopted an approach centered on high-entropy cathode engineering, capitalizing on the excellent conductivity imparted by both the maximized entropy and the distinctive antisite disorder in the framework. In more detail, a high-entropy variant of conventional olivine-based cathodes was produced using a mechanochemically assisted annealing method. Owing to the configurational entropy effect, the quinary Li(Mn_0.2Fe_0.2Co_0.2Ni_0.2Mg_0.2)PO₄ delivered higher specific capacities compared with unary LiMnPO₄. Besides, adjusting the cation distribution and introducing Ca as a structural component expanded the crystal lattice in Li(Mn_0.35Fe_0.35Co_0.1Mg_0.1Ca_0.1)PO₄, thereby enhancing Li-ion diffusion. Thus, when employed as a high-voltage cathode, this quinary compound achieved an initial specific capacity of 104 mAh g⁻¹, with three well-defined voltage plateaus attributed to the Fe, Mn, and Co redox processes between 2.0–4.9 V vs. Li⁺/Li, significantly outperforming single-metal LMP analogs (15 mAh g⁻¹).

6. Conclusions and prospects

Lithium manganese iron phosphate (LiMn_yFe_1−yPO₄, LMFP) has garnered significant attention as a cathode material due to its affordability, excellent safety, prolonged cycle life, high operating voltage, superior thermal stability, and remarkable energy density. However, despite notable advancements over the past decades, LMFP continues to encounter significant challenges. Its low electronic conductivity and sluggish Li-ion diffusion significantly restrict high-rate performance, while manganese dissolution compromises long-term cycling stability. Additionally, its low tap density poses further obstacles to practical applications. To facilitate the large-scale commercialization of LMFP in power batteries, continued research and innovative strategies are essential to overcome these technical hurdles.

A key emerging strategy for enhancing the stability and overall electrochemical performance of LMFP cathode materials, particularly in energy storage applications, is the development of multielemental compositions. These improvements can be achieved through various approaches, including (i) doping with one or more elements, (ii) incorporating additional primary constituents or structural components, and (iii) increasing the entropy of the electrode materials. Such design principles have proven effective in optimizing LMFP properties, ultimately advancing the performance and durability of next-generation energy storage systems. Table 1 and Fig. 12 provide a comprehensive overview of all multimetal LMFP cathodes, including key electrochemical performance metrics. Based on the analyzed strategies, several important conclusions can be derived, as highlighted below.

The synthesis process plays a pivotal role in optimizing the performance of multimetal LMFP cathodes since factors such as precursor selection, synthesis temperature, and methodology applied significantly influence the crystallinity, morphology, ion conductivity, structural stability, cycling performance, and energy density of the cathode materials. Among the various synthesis techniques, the solid-state reaction is the most widely used, offering advantages such as simplicity, cost-effectiveness, and the ability to achieve high crystallinity and uniformity through high-temperature treatments. The solvothermal method tends to be advantageous compared to the hydrothermal method, while more expensive and less reproducible, provides reasonable control over morphology, particle size, and crystallinity, making it useful for producing high-quality materials despite its higher cost and scalability limitations. Although mainly confined to laboratory use due to its lengthy procedure and high material costs, the sol–gel process allows for precise control over the composition and structure of the final product, which is valuable for research purposes. The co-precipitation method, with its low cost, short processing times, and simplicity, offers advantages in terms of the availability of raw materials and ease of execution. However, it faces challenges in controlling crystallization and ion behavior. In addition, physical processes such as freeze-drying and spray-drying enhance particle size control, uniformity, and overall processing efficiency. Each synthesis method, when optimized, plays a critical role in tailoring the properties of the LMFP cathodes, thereby impacting their overall electrochemical performance and advancing the development of high-performance battery technologies.

Incorporating multi-metals into LMFP-based electrodes follows distinct approaches, including elemental doping, incorporation of structural modification, and/or entropy-driven strategies, as highlighted previously. While doping typically should be restricted to a maximum of 5% by mass or atomic fraction, higher concentrations lead to structural integration of the introduced metal. Unfortunately, numerous authors in the literature incorrectly use the term doping to describe the incorporation of a new metal ion at concentrations as high as 10 mol% or mass%. However, at such levels, the added element should be considered a structural constituent of the composition rather than a dopant. Medium-entropy LMFP-based materials or high-entropy olivine structures have emerged as promising candidates, offering enhanced electrochemical properties in equimolar metal concentrations. Interestingly, among the investigated strategies, doping with one or more elements remains the most effective in fine-tuning electrochemical performance by optimizing lithium-ion diffusion, enhancing electronic conductivity, and stabilizing structural integrity. The incorporation of additional primary constituents or structural components can further improve electrode performance, particularly by mitigating volume expansion and facilitating charge transfer, but its success heavily depends on the compatibility and synergy of the added element. Increasing the entropy of electrode materials via equimolar multi-component design presents a promising avenue for long-term stability and structural robustness. However, its effectiveness in achieving high specific capacity and rate capability is still limited by sluggish kinetics and the challenge of balancing entropy-driven stabilization with the electrochemical activity of added metal ions. Therefore, a synergistic approach that employs metal or multimetal doping is the most promising path for achieving high-performance electrode materials. In fact, beyond enhancing electrochemical performance, doping also allows the incorporation of costlier elements than Fe and Mn without substantially increasing material costs. This is because only small concentrations of the added element are required (at% ≤3%), unlike the larger amounts needed when the third metal acts as a structural component or in entropy-enhanced electrode materials.

Among multimetal LMFPs, isovalent ions such as Mg²⁺ and Ni²⁺ dominate, whereas aliovalent metals like Nb⁵⁺ and V³⁺ are extensively studied. Transition-metal dopants like divalent Co²⁺ and Ni²⁺ generally enhance electronic conductivity and facilitate faster charge-transfer kinetics by introducing additional redox couples, whereas Mg²⁺ and Zn²⁺ tend to stabilize the olivine lattice and improve cycling stability at high rates. Aliovalent dopants such as Ti⁴⁺, Al³⁺, and particularly Nb⁵⁺ play a crucial role in suppressing antisite defects and reinforcing the crystal framework; Nb⁵⁺ can additionally promote lattice effects that enhance Li⁺ mobility, leading to notable improvements in both rate capability and long-term cycling. Comparative studies indicate that while transition-metal doping often provides higher initial capacity and rate performance, lattice-stabilizing dopants like Nb⁵⁺ yield superior structural integrity and capacity retention over prolonged cycling. Besides, recent in situ XRD studies further reveal that aliovalent ion doping (e.g., Nb⁵⁺ and Ti⁴⁺) markedly reduces the two-phase reaction region compared to non-doped LMFP/C (especially when associated with other strategies such as carbon coating or the presence of Fe vacancies), yielding a more uniform phase transition and enhanced reaction consistency, as well as narrowing the band gap and boosting intrinsic conductivity. Thus, a synergistic doping strategy, combining elements that enhance both conductivity and structural stability, appears to be the most promising route for optimizing LMFP performance for large-scale applications.

Conversely, Nb⁵⁺ aliovalent ions have also been integrated into coating materials to improve the stability of multimetal LMFP structures. Additionally, carbon-based materials, particularly LMFP/C composites, play a crucial role in optimizing electrochemical performance by establishing conductive networks. These advancements highlight the potential of multimetal LMFPs for next-generation lithium-ion batteries, balancing structural stability, conductivity, and electrochemical activity.

Despite considerable research advancements in recent years toward designing multimetal LMFP cathodes, substantial gaps remain unexplored. In this context, the following future perspectives and research directions should be considered for further advancements, as outlined below.

(i) Few research papers report the preparation of medium-entropy olivine structures, demonstrating that there is still much research to be done in developing such systems. In fact, while few medium-entropy cathodes have been explored, the design of high-entropy olivine electrodes remains unreported. In other types of cathode active materials, the development of medium- and high-entropy compounds has not led to significant improvements in specific capacity. However, their stability over charge–discharge cycles has shown promising results, particularly when incorporating electrochemically inactive metal ions. Therefore, future efforts should focus on this area, particularly targeting the development of cathode materials with enhanced energy density and charge retention, ideally under high-power operation. Furthermore, in the context of developing medium- and high-entropy olivine structures, it is essential to consider the total configurational entropy contribution from both the Li-site and M-site, despite the challenges in determining the specific site occupancy of certain metal elements.

(ii) The use of low-, medium-, and high-entropy coordination compounds as precursor materials for cathode-active materials remains largely unexplored. From this perspective, there is still significant research potential in designing multimetal olivine cathodes derived from metal–organic frameworks (MOFs), Prussian blue analogues (PBAs), as well as metal-polyphenol coordination polymers, and other typical coordination compounds. In fact, this strategy has been employed in many other research areas to prepare new materials for energy conversion and storage.¹⁴⁷

(iii) So far, a few studies have explored a dual-site doping strategy, suggesting that the development of multimetal LMFP using this approach remains in its early stages. Future studies could focus on optimizing elemental selection and distribution at both sites to enhance electrochemical performance. Additionally, a deeper understanding of the structural and electronic interactions resulting from dual-site modifications could pave the way for improved cycling stability, rate capability, and overall energy storage efficiency. Expanding this strategy may also contribute to the design of next-generation LMFP cathodes with superior electrochemical properties.

(iv) Similar to the case of the Mg²⁺ ion, while most studies report its incorporation into LMFP at the M-site, some research has also suggested the possibility of its integration at the Li-site. This indicates that Mg²⁺ and other metal ions can potentially occupy both sites. However, accurately determining the exact location of the incorporated metal ion remains challenging. Therefore, it is crucial to develop studies utilizing methodologies capable of unambiguously identifying and quantifying the atomic occupancy of each site. To achieve this, advanced characterization techniques should be employed, particularly those leveraging synchrotron radiation. These include Extended X-ray Absorption Fine Structure (EXAFS), X-ray Absorption Near-Edge Structure (XANES), Near-Edge X-ray Absorption Fine Structure (NEXAFS), and X-ray Photoelectron Spectroscopy (XPS), which provide valuable insights into the structural and electronic environment of the incorporated elements.

(v) Despite the fact that nearly all high-performance multimetal LMFP cathodes incorporate a carbon coating approach, other carbonaceous materials have been scarcely reported in multimetal cathodes. In this regard, the use of alternative carbon-based materials, such as graphene/reduced graphene oxide, carbon nanotubes, and carbon nanofibers, could serve as promising platforms to enhance the power density of energy storage devices.

In summary, recent advancements in developing multimetal LMFP cathodes for LIBs demonstrate promising progress; however, significant challenges remain to fully meet the demands of our energy-intensive society. Given the latest findings, emerging trends, and future research directions, we hope this review contributes to the recognition and establishment of multimetal LMFP cathodes as next-generation electrode materials for LIBs. Furthermore, continued efforts in battery materials research are essential to developing advanced systems that can drive future breakthroughs in energy storage technologies.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this minireview.

Data availability

No primary research results, software, or code have been included, and no new data were generated or analysed as part of this review.

Acknowledgements

The authors wish to convey their profound gratitude for the substantial financial backing rendered by several distinguished Brazilian funding agencies, namely, the São Paulo Research Foundation (FAPESP 2023/17560-0 and 2017/11986-5), the Brazilian Innovation Agency (FINEP), the Foundation for Research Development (FUNDEP), and the National Council for Scientific and Technological Development (CNPq). This indispensable support has played an instrumental role in facilitating the successful execution of our research endeavors. B. G. A. N. and J. M. G acknowledge funding through MACKPESQUISA (Fundo Mackenzie de Pesquisa e Inovação; grant #231020 and #251035). Furthermore, we wish to extend our heartfelt appreciation to Shell for their unwavering commitment and valuable support throughout this study. In addition, we would like to acknowledge the strategic significance of the unwavering assistance provided by the Brazilian National Oil, Natural Gas, and Biofuels Agency (ANP) through the Research and Development (R&D) levy regulation.

References

1. M. M. Hasan, R. Haque, M. I. Jahirul, M. G. Rasul, I. M. R. Fattah, N. M. S. Hassan and M. Mofijur, J. Energy Storage, 2025, 120, 116511.
2. F. M. N. U. Khan, M. G. Rasul, A. S. M. Sayem and N. Mandal, Energy Rep., 2023, 9, 11–21.
3. J. M. Gonçalves, G. T. M. Silva and H. Zanin, J. Mater. Chem. A, 2024, 12, 17756–17770.
4. H. Wang, Q. Shi, J. Dong, M. Wang, Y. Lu, Y. Liu, J. Liu, N. Li, Q. Huang, Y. Su, F. Wu and L. Chen, Adv. Funct. Mater., 2025, 35, 2422806.
5. S. Mubarac, M. N. T. Silva, G. T. M. Silva, B. Freitas, J. M. Gonçalves and H. Zanin, J. Energy Storage, 2024, 96, 112612.
6. J. R. Croy, A. Gutierrez, M. He, B. T. Yonemoto, E. Lee and M. M. Thackeray, J. Power Sources, 2019, 434, 226706.
7. B. Zhang, X. Wang, S. Wang, Y. Li, L. Chen, H. Jiao, Z. Yu and J. Tu, J. Energy Chem., 2025, 100, 1–17.
8. E. Xu, X. Sun, W. Lyv, F. Li, R. Li, W. Cai, H. Wu, K. Wu and Y. Zhang, Ind. Eng. Chem. Res., 2024, 63, 9631–9660.
9. Z. Qiu, Q. Man, Y. Mu, H. Gu, Z. Zou, M. Han and L. Zeng, Chem.: Methods, 2025, 5, 2400065.
10. K. Akhmetova, F. Sultanov, A. Mentbayeva, N. Umirov, Z. Bakenov and B. Tatykayev, J. Power Sources, 2024, 624, 235531.
11. N. G. Garcia, J. M. Gonçalves, C. Real, B. Freitas, J. G. Ruiz-Montoya and H. Zanin, Energy Storage Mater., 2024, 67, 103213.
12. W. Ling, S. Fang, W. Zhou, C. Ye, L. Wang, N. Zhou and X. He, Adv. Energy Mater., 2025, 2502720.
13. D. Bresser, E. Paillard and S. Passerini, in Advances in Batteries for Medium and Large-Scale Energy Storage, Elsevier, 2015, pp. 125–211.
14. M. Li, J. Lu, Z. Chen and K. Amine, Adv. Mater., 2018, 30, 1800561.
15. Y. Liu, R. Zhang, J. Wang and Y. Wang, iScience, 2021, 24, 102332.
16. A. A. Kebede, T. Kalogiannis, J. Van Mierlo and M. Berecibar, Renewable Sustainable Energy Rev., 2022, 159, 112213.
17. Q. Dong, S. Liang, J. Li, H. C. Kim, W. Shen and T. J. Wallington, iScience, 2023, 26, 107195.
18. Z. M. Ali, M. Calasan, F. H. Gandoman, F. Jurado and S. H. E. Abdel Aleem, Ain Shams Eng. J., 2024, 15, 102442.
19. Z. W. B. Iton, S. S. Kim, E. S. Patheria, M. D. Qian, S. D. Ware and K. A. See, in Comprehensive Inorganic Chemistry III, Elsevier, 2023, pp. 308–363.
20. P. Zhu, D. Gastol, J. Marshall, R. Sommerville, V. Goodship and E. Kendrick, J. Power Sources, 2021, 485, 229321.
21. S. Li, X. Zhu, Y. Ma, F. Zhang and H. Zhou, J. Policy Anal. Manage., 2022, 41, 450–485.
22. Y. Li, C. Liang, F. Ye and X. Zhao, Transp. Res. A: Policy Pract., 2023, 167, 103558.
23. S. Abdul Qadir, F. Ahmad, A. Mohsin, A. B. Al-Wahedi, A. Iqbal and A. Ali, Energ. Strateg. Rev., 2024, 53, 101379.
24. S. Yin, W. Deng, J. Chen, X. Gao, G. Zou, H. Hou and X. Ji, Nano Energy, 2021, 83, 105854.
25. G. Zhou, H. Chen and Y. Cui, Nat. Energy, 2022, 7, 312–319.
26. S. Zhao, Z. Guo, K. Yan, S. Wan, F. He, B. Sun and G. Wang, Energy Storage Mater., 2021, 34, 716–734.
27. Y. Zou, G. Liu, K. Zhou, J. Zhang, T. Jiao, X. Zhang, Y. Yang and J. Zheng, ACS Appl. Energy Mater., 2021, 4, 11051–11061.
28. M. Zybert, H. Ronduda, W. Raróg-Pilecka and W. Wieczorek, Front Energy Res., 2023, 11, 1248641.
29. J. Shi, D. Xiao, M. Ge, X. Yu, Y. Chu, X. Huang, X. Zhang, Y. Yin, X. Yang, Y. Guo, L. Gu and L. Wan, Adv. Mater., 2018, 30, 1705575.
30. P. U. Nzereogu, A. D. Omah, F. I. Ezema, E. I. Iwuoha and A. C. Nwanya, Appl. Surf. Sci. Adv., 2022, 9, 100233.
31. J. Qian, L. Liu, J. Yang, S. Li, X. Wang, H. L. Zhuang and Y. Lu, Nat. Commun., 2018, 9, 4918.
32. A. Mauger and C. M. Julien, Batteries, 2018, 4, 39.
33. X. Yao, D. Li, L. Guo, M. Kallel, S. D. Alahmari, J. Ren, I. Seok, G. Roymahapatra and C. Wang, Adv. Compos. Hybrid Mater., 2024, 7, 63.
34. S.-P. Chen, D. Lv, J. Chen, Y.-H. Zhang and F.-N. Shi, Energy Fuels, 2022, 36, 1232–1251.
35. R. Tang, J. Dong, C. Wang, Y. Guan, A. Yin, K. Yan, Y. Lu, N. Li, G. Zhao, B. Li, W. Shen, F. Wu, Y. Su and L. Chen, Adv. Funct. Mater., 2025, 35, 2421284.
36. S. B. Chikkannanavar, D. M. Bernardi and L. Liu, J. Power Sources, 2014, 248, 91–100.
37. P. Vanaphuti, K. Scanlan and A. Manthiram, RSC Sustain., 2024, 2, 1969–1978.
38. N. Du, X. Lei, Q. Yu, C. Wang and S. Luo, Mater. Today Energy, 2025, 52, 101965.
39. G. Bree, J. Zhao, V. Majherova, D. Proprentner, G. J. Paez Fajardo and L. F. J. Piper, Energy Fuels, 2025, 39, 3683–3689.
40. H. Yu, E. Zhang, J. Yu, S. Yu, Y. Fang, L. Chen, H. Jiang and C. Li, J. Mater. Chem. A, 2024, 12, 26076–26082.
41. K. Zhang, Z.-X. Li, X. Li, X.-Y. Chen, H.-Q. Tang, X.-H. Liu, C.-Y. Wang and J.-M. Ma, Rare Met., 2023, 42, 740–750.
42. M. D. Johannes, K. Hoang, J. L. Allen and K. Gaskell, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 85, 115106.
43. Y. Dong, L. Wang, S. Zhang, Y. Zhao, J. Zhou, H. Xie and J. B. Goodenough, J. Power Sources, 2012, 215, 116–121.
44. S. Wi, J. Park, S. Lee, J. Kang, T. Hwang, K. S. Lee, H. K. Lee, S. Nam, C. Kim, Y. E. Sung and B. Park, Nano Energy, 2017, 31, 495–503.
45. S. Wi, J. Park, S. Lee, J. Kim, B. Gil, A. J. Yun, Y. E. Sung, B. Park and C. Kim, Nano Energy, 2017, 39, 371–379.
46. Y. Asari, Y. Suwa and T. Hamada, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 84, 134113.
47. L. F. J. Piper, N. F. Quackenbush, S. Sallis, D. O. Scanlon, G. W. Watson, K. W. Nam, X. Q. Yang, K. E. Smith, F. Omenya, N. A. Chernova and M. S. Whittingham, J. Phys. Chem. C, 2013, 117, 10383–10396.
48. Y. Mishima, T. Hojo, T. Nishio, H. Sadamura, N. Oyama, C. Moriyoshi and Y. Kuroiwa, J. Phys. Chem. C, 2013, 117, 2608–2615.
49. D. Jang, K. Palanisamy, Y. Kim and W.-S. Yoon, J. Electrochem. Sci. Technol., 2013, 4, 102–107.
50. Y. Lu and T. Zhu, MRS Commun., 2024, 14, 888–899.
51. D. Kwon and D. Kim, Adv. Energy Mater., 2025, 15, 2501444.
52. A. K. Padhi, K. S. Nanjundaswamy and J. B. Goodenough, J. Electrochem. Soc., 1997, 144, 1188–1194.
53. G. R. Gardiner and M. S. Islam, Chem. Mater., 2010, 22, 1242–1248.
54. Y. Deng, C. Yang, K. Zou, X. Qin, Z. Zhao and G. Chen, Adv. Energy Mater., 2017, 7, 1601958.
55. S. Li, H. Zhang, Y. Liu, L. Wang and X. He, Adv. Funct. Mater., 2024, 34, 2310057.
56. M. Kope¢, A. Yamada, G. Kobayashi, S. Nishimura, R. Kanno, A. Mauger, F. Gendron and C. M. Julien, J. Power Sources, 2009, 189, 1154–1163.
57. I. Bezza, M. Kaus, R. Heinzmann, M. Yavuz, M. Knapp, S. Mangold, S. Doyle, C. P. Grey, H. Ehrenberg, S. Indris and I. Saadoune, J. Phys. Chem. C, 2015, 119, 9016–9024.
58. A. Yamada, Y. Kudo and K. Y. Liu, J. Electrochem. Soc., 2001, 148, A747–A754.
59. C. C. Yang and W. H. Chen, Mater. Chem. Phys., 2016, 173, 482–490.
60. A. Paolella, G. Bertoni, E. Dilena, S. Marras, A. Ansaldo, L. Manna and C. George, Nano Lett., 2014, 14, 1477–1483.
61. A. Yamada, Y. Kudo and K. Y. Liu, J. Electrochem. Soc., 2001, 148, A1153–A1158.
62. N. D. Trinh, Z. W. Ai, G. Liang and S. B. Schougaard, Solid State Ionics, 2018, 324, 33–39.
63. D. B. Ravnsbæk, K. Xiang, W. Xing, O. J. Borkiewicz, K. M. Wiaderek, P. Gionet, K. W. Chapman, P. J. Chupas and Y. M. Chiang, Nano Lett., 2014, 14, 1484–1491.
64. L. Yang, W. Deng, W. Xu, Y. Tian, A. Wang, B. Wang, G. Zou, H. Hou, W. Deng and X. Ji, J. Mater. Chem. A, 2021, 9, 14214–14232.
65. Y. Li, C. Yu, X. Wang, Y. Zhong, Z. Wu, X. Wang and B. Zhong, Ind. Eng. Chem. Res., 2025, 64, 13920–13949.
66. Z. Peng, X. Xia, Y. Xiao and T. Huang, J. Alloys Compd., 2025, 1016, 178901.
67. S. Xin, H. Zhang, Z. Hu, P. Zhao, C. Zhou, H. He, P. Liu, J. Zhang and J. Zhou, ACS Appl. Mater. Interfaces, 2025, 17, 33783–33795.
68. J. Han, J. Zhu, X. He, M. Yang, C. Yan, D. Ma, L. Zhang and P. Zhang, ACS Appl. Mater. Interfaces, 2025, 17, 32381–32391.
69. J. Huang, X. Tang, Y. Zhou, T. Wang, F. Zhao, W. Wang, Y. Meng, W. Cen and Y. Zhang, J. Mater. Chem. A, 2025, 13, 10550–10560.
70. X. Gao, S. Zheng, Y. Song, C. Chen, X. Long, Z. Qiu, W. Li, P. Liu, X. Li, Y. Cui and W. Xing, J. Energy Storage, 2025, 131, 117670.
71. C. Ye, W. Ling, X. Huang, Y. Wu, N. Zhou and X. Yang, Electrochim. Acta, 2025, 536, 146789.
72. Y.-C. Chen, J.-M. Chen, C.-H. Hsu, J.-J. Lee, T.-C. Lin, J.-W. Yeh and H. C. Shih, J. Power Sources, 2010, 195, 6867–6872.
73. L. Qin, Y. Xia, H. Cao, L. Yang and Z. Liu, Electrochim. Acta, 2016, 222, 1660–1667.
74. E. Boldyreva, Chem. Soc. Rev., 2013, 42, 7719.
75. M. Ratynski, M. Winkowska-Struzik, D. Buchberger, B. Hamankiewicz, M. Krajewski and A. Czerwinski, ACS Appl. Energy Mater., 2025, 8, 11053–11067.
76. Y. Cao, J. He, W. Tu, R. Li, W. Luo, X. Zhang, J. Duan, D. Wang, X. Wang, P. Dong and Y. Zhang, Solid State Ionics, 2024, 411, 116576.
77. M.-J. Uddin, P. K. Alaboina and S.-J. Cho, Mater. Today Energy, 2017, 5, 138–157.
78. D. Kong, H. Chen, F. Wu, R. Zhang, J. Li, Y. Mai, Y. Wei, J. Wang and X. Dai, Electrochim. Acta, 2024, 506, 145060.
79. D. Kong, H. Chen, F. Wu, R. Zhang, J. Li, Y. Mai, Y. Du, J. wang and X. Dai, Ceram. Int., 2024, 50, 47308–47316.
80. Q. Deng, T. Li, J. Wang, S. Zhang, H. Yang, C. Xu and M. Wu, Ceram. Int., 2024, 50, 13702–13710.
81. Z. Dai, L. Wang, X. He, F. Ye, C. Huang, J. Li, J. Gao, J. Wang, G. Tian and M. Ouyang, Electrochim. Acta, 2013, 112, 144–148.
82. D. Bokov, A. Turki Jalil, S. Chupradit, W. Suksatan, M. Javed Ansari, I. H. Shewael, G. H. Valiev and E. Kianfar, Adv. Mater. Sci. Eng., 2021, 2021, 5102014.
83. L. L. Hench and J. K. West, Chem. Rev., 1990, 90, 33–72.
84. A. E. Danks, S. R. Hall and Z. Schnepp, Mater. Horiz., 2016, 3, 91–112.
85. H. Liu, Y. P. Wu, E. Rahm, R. Holze and H. Q. Wu, J. Solid State Electrochem., 2004, 8, 450–466.
86. S. Qiao, L. Zhu, E. Han, L. Li, C. Du and Y. He, Int. J. Electrochem. Sci., 2019, 14, 10616–10629.
87. H. Dong and G. M. Koenig, CrystEngComm, 2020, 22, 1514–1530.
88. J. Li, Z. Liu and J. Ma, Sci. Rep., 2025, 15, 28004.
89. X. Bi, L. Chang, S. Luo, S. Cao, A. Wei, W. Yang, J. Liu and F. Zhang, Ionics, 2022, 28, 577–591.
90. P. Wang, E. Zhang, Y. Fang, Y. Chen, H. Yu, Y. Zhang, Q. Cheng and H. Jiang, J. Solid State Electrochem., 2024, 28, 4303–4310.
91. J. Xu, K. Hou, X. Li, Y. Bian, Y. Wang, L. Wang and G. Liang, ACS Appl. Energy Mater., 2024, 7, 8694–8704.
92. P. Vanaphuti and A. Manthiram, Small, 2024, 20, 2404878.
93. A. Iturrondobeitia, A. Goñi, I. Gil de Muro, L. Lezama, C. Kim, M. Doeff, J. Cabana and T. Rojo, Inorg. Chem., 2015, 54, 2671–2678.
94. H. Jin, J. Zhang, L. Qin, Y. Hu, H. Jiang and C. Li, Ind. Eng. Chem. Res., 2023, 62, 1029–1034.
95. S. Tian, K. Zhang, J. Cao, H. Guo, R. Liu and G. Liang, Ionics, 2021, 27, 2877–2887.
96. S. Mi, Z. Liu, C. Luo, L. Cai, Z. Zhang and L. Li, Drying Technol., 2020, 38, 1544–1564.
97. W. Abdelwahed, G. Degobert, S. Stainmesse and H. Fessi, Adv. Drug Deliv. Rev., 2006, 58, 1688–1713.
98. B. Vertruyen, N. Eshraghi, C. Piffet, J. Bodart, A. Mahmoud and F. Boschini, Materials, 2018, 11, 1076.
99. C. Arpagaus, A. Collenberg and D. Rütti, Drying Technol., 2019, 37, 426–434.
100. A. B. D. Nandiyanto and K. Okuyama, Adv. Powder Technol., 2011, 22, 1–19.
101. R. Palm, A. M. Baena-Moncada and J. M. Gonçalves, J. Mater. Chem. A, 2024, 12, 29402–29431.
102. J. M. Gonçalves, É. A. Santos, P. R. Martins, C. G. Silva and H. Zanin, Energy Storage Mater., 2023, 63, 102999.
103. T. G. Ritter, S. Pappu and R. Shahbazian-Yassar, Batteries, 2024, 10, 96.
104. S. Schweidler, M. Botros, F. Strauss, Q. Wang, Y. Ma, L. Velasco, G. Cadilha Marques, A. Sarkar, C. Kübel, H. Hahn, J. Aghassi-Hagmann, T. Brezesinski and B. Breitung, Nat. Rev. Mater., 2024, 9, 266–281.
105. S. Baúti, J. V. T. Feyh, F. C. Antunes, T. Dias, G. Doubek, J. M. Gonçalves, B. B. N. S. Brandão and H. Zanin, J. Energy Chem., 2025, 111, 513–544.
106. J. M. Gonçalves, É. A. Santos, M. M. Amaral, E. Nossol, R. O. Figueiredo and H. Zanin, ACS Omega, 2025, 10(35), 39309–39333.
107. J. Fan, T. Wang, Y. Yuan, A. Borisevich, C.-L. Do-Thanh, Z. Yang and S. Dai, Appl. Phys. Lett., 2024, 124, 171909.
108. X. Zhang, M. Hou, A. G. Tamirate, H. Zhu, C. Wang and Y. Xia, J. Power Sources, 2020, 448, 227438.
109. T. Wu, J. Liu, L. Sun, L. Cong, H. Xie, A. Abdel-Ghany, A. Mauger and C. M. Julien, J. Power Sources, 2018, 383, 133–143.
110. J. Yu, J. Ma, H. Zhang, J. Hang, W. Tian, X. Tian and H. Yang, Ceram. Int., 2025, 51, 26443–26451.
111. Z. Li, Y. You, Z. Zhu, L. Wang, S. Ou, J. Xu and M. Yuan, J. Energy Storage, 2025, 110, 115111.
112. M. Zhao, Y. Zhou, Y. Chen, X. Liang, J. Zeng, K. Bai, X. Xu, H. Wang, X. Jiang and H. He, Adv. Funct. Mater., 2025, 2509461.
113. K. Zhang, J. Cao, S. Tian, H. Guo, R. Liu, X. Ren, L. Wen and G. Liang, Ionics, 2021, 27, 4629–4637.
114. R. Li, C. Fan, W. Zhang, M. Tan, T. Zeng and S. Han, Ceram. Int., 2019, 45, 10501–10510.
115. Y. Wang, H. Yang, C.-Y. Wu and J.-G. Duh, J. Mater. Chem. A, 2017, 5, 18674–18683.
116. H. Hu, H. Li, Y. Lei, J. Liu, X. Liu, R. Wang, J. Peng and X. Wang, J. Energy Storage, 2023, 73, 109006.
117. D. Kim, S. Lee and W. Choi, J. Alloys Compd., 2024, 989, 174396.
118. W. Liu, X. Liu, R. Hao, Z. Yang, B. Ouyang, M. Zhang, M. Pan and K. Liu, J. Electroanal. Chem., 2023, 929, 117117.
119. J. Ni, Y. Zhao, J. Chen, L. Gao and L. Lu, Electrochem. Commun., 2014, 44, 4–7.
120. G. K. P. Dathar, D. Sheppard, K. J. Stevenson and G. Henkelman, Chem. Mater., 2011, 23, 4032–4037.
121. J. Duan, G. Hu, Y. Cao, K. Du and Z. Peng, Ionics, 2016, 22, 609–619.
122. C. Hu, B. Wang, Huihua, J. Zhang, Y. Hu and J. Li, Int. J. Electrochem. Sci., 2018, 13, 5824–5831.
123. K. Kisu, E. Iwama, W. Onishi, S. Nakashima, W. Naoi and K. Naoi, J. Mater. Chem. A, 2014, 2, 20789–20798.
124. T. Zeng, P. Gao, Z. Zhou, C. Fan, Z. Liu, F. Zhang, J. Liu and J. Liu, Energy Storage Mater., 2024, 65, 103125.
125. L. Li, L. Wu, F. Wu, S. Song, X. Zhang, C. Fu, D. Yuan and Y. Xiang, J. Electrochem. Soc., 2017, 164, A2138–A2150.
126. P. Su, H. Zhang, L. Yang, C. Xing, S. Pan, W. Lu and S. Zhang, Chem. Eng. J., 2022, 433, 133203.
127. J. Geng, Z. Zou, T. Wang, S. Zhang, S. Zhong, W. Ling, X. Peng and X. Hu, J. Electroanal. Chem., 2023, 933, 117275.
128. N. Meethong, Y. Kao, S. A. Speakman and Y. Chiang, Adv. Funct. Mater., 2009, 19, 1060–1070.
129. S.-Y. Chung, J. T. Bloking and Y.-M. Chiang, Nat. Mater., 2002, 1, 123–128.
130. H. Yi, C. Hu, H. Fang, B. Yang, Y. Yao, W. Ma and Y. Dai, Electrochim. Acta, 2011, 56, 4052–4057.
131. S. Liu, H. Fang, E. Dai, B. Yang, Y. Yao, W. Ma and Y. Dai, Electrochim. Acta, 2014, 116, 97–102.
132. Q.-Y. Huang, Z. Wu, J. Su, Y.-F. Long, X.-Y. Lv and Y.-X. Wen, Ceram. Int., 2016, 42, 11348–11354.
133. P. Xiao, Y. Cai, X. Chen, Z. Sheng and C. Chang, RSC Adv., 2017, 7, 31558–31566.
134. H. Shu, X. Wang, Q. Wu, B. Hu, X. Yang, Q. Wei, Q. Liang, Y. Bai, M. Zhou, C. Wu, M. Chen, A. Wang and L. Jiang, J. Power Sources, 2013, 237, 149–155.
135. H. Yi, C. Hu, X. He and H. Xu, Ionics, 2015, 21, 667–671.
136. B.-K. Zou, Y. Shao, Z.-Y. Qiang, J.-Y. Liao, Z.-F. Tang and C.-H. Chen, J. Power Sources, 2016, 336, 231–239.
137. W. Xiang, Y. Zhong, Y. Tang, H. Shen, E. Wang, H. Liu, B. Zhong and X. Guo, J. Alloys Compd., 2015, 635, 180–187.
138. C. Hu, H. Yi, H. Fang, B. Yang, Y. Yao, W. Ma and Y. Dai, Electrochem. Commun., 2010, 12, 1784–1787.
139. J. E. Frąckiewicz and T. K. Pietrzak, Materials, 2021, 14, 6434.
140. Y. Cai, D. Zhang, C. Chang, Z. Sheng and K. Huang, Ionics, 2016, 22, 1011–1019.
141. X. Chu, L. Li, W. Chen and H. Fang, Ionics, 2021, 27, 2927–2935.
142. H. Fang, E. Dai, B. Yang, Y. Yao and W. Ma, J. Power Sources, 2012, 204, 193–196.
143. X.-Y. Lv, Q.-Y. Huang, Z. Wu, J. Su, Y.-F. Long and Y.-X. Wen, J. Solid State Electrochem., 2017, 21, 1499–1507.
144. B. J. Jeong, J. Y. Sung, F. Jiang, S. P. Jung and C. W. Lee, J. Energy Storage, 2024, 96, 112552.
145. J. Zheng, J. Yang, J. Wu, S. Li, M. Wang, B. Huang, Y. Li, S. Xiao and Q. Zhu, J. Alloys Compd., 2023, 960, 170610.
146. Y. Niu, S. Wang, R. Chang, P. Yang, H. Xing and Y. Xu, ACS Appl. Mater. Interfaces, 2025, 17, 18831–18843.
147. J. M. Gonçalves and J. G. Ruiz-Montoya, J. Mater. Chem. A, 2023, 11, 20872–20885.

---

Footnote

† Equal contributors.

---