Water Effect on the Electronic Properties and Lithium-Ion Conduction in a Defect-Engineered LiFePO₄ Electrode

Abstract

Defect-engineering accelerates the conduction of lithium ions in the cathode materials of lithium-ion batteries. However, the effects of defect-engineering on ion conduction and its mechanisms in humid environments remain unclear in the academic discourse. Here, we report on the effect of vacancy defects on the electronic properties of and Li-ion diffusion in a LiFePO₄ material in humid environments. The research findings indicate that vacancy defects reduce the lattice constant and unit cell volume of LiFePO₄. Additionally, the water molecules occupy the Li-ion vacancies, leading to an increase in the lattice constant of LiFePO₄. The computational results of the electronic properties show that the introduction of water molecules induces a transition in LiFePO₄ from a semiconductor to a metallic behavior, with a transfer of 0.38 e of charge from the water molecules to LiFePO₄. Additionally, the migration barrier for Li ions in the H₂O + LiFePO₄ system is found to be 0.50 eV, representing an 11.1% increase compared to the pristine LiFePO₄ migration barrier. Our findings suggest that water molecules impede the migration of Li ions and provide important insights into the effect of defect-engineering on electronic properties and ion conduction under humid conditions.

1. Introduction

The current fossil fuel resources are insufficient to meet the demands of the rapid economic development, and the environmental pollution resulting from fossil fuel combustion is becoming increasingly severe [1,2,3]. Renewable energy contributes to the transformation of the global energy structure, and the consumption of renewable energy is a key constraint to its development [4,5]. Advanced energy materials and devices represented by lithium-ion batteries provide robust technological support for achieving high-efficiency energy storage and energy transition [6,7,8,9]. Lithium-ion batteries possess significant advantages such as high energy density, high power density, long cycle life, and environmental friendliness, which make them extensively utilized in energy storage and consumer electronics [10,11,12]. In recent years, there has been a significant growth trend in the demand for lithium-ion batteries, driven by the rapid development in fields such as electric vehicles and energy storage. LiFePO₄ stands out as a commercially successful cathode material for lithium-ion batteries due to its advantages of low production cost, high safety, and strong cycling stability [13,14,15]. It can effectively store energy, is not constrained by natural environments, and mitigates the adverse impacts of intermittency, thereby improving the utilization efficiency of renewable energy [16,17,18]. The primary factors currently constraining the application of lithium-ion batteries are the cycling stability and safety of the LiFePO₄ material, and among them, humidity control stands out as a significant factor [19,20,21,22].

In the preparation process of LiFePO₄, due to its small particle size and large specific surface area, problems such as powder aggregation and moisture absorption easily occur in humid environments. An uncontrolled moisture content in LiFePO₄ can lead to the decomposition of lithium salts in the electrolyte and hinder the film formation on the cathode material, ultimately resulting in the deterioration of the electrochemical properties of lithium-ion batteries [23,24]. Additionally, when LiFePO₄ is exposed to air for an extended period, moisture continuously infiltrates the material, leading to the formation of amorphous FePO₄, ultimately resulting in aging and failure of LiFePO₄ [25]. Jarolimek et al. [26] research findings indicate that OH groups and H₂O stably adsorb on the (010) surface of LiFePO₄. After immersing LiFePO₄ in distilled water for 96 h, Xu et al. [27] observed that its initial discharge specific capacity and capacity after 20 cycles were 131.8 and 96 mAh/g, respectively. Compared to the values of 140 and 117 mAh/g for non-immersed samples, the cycling performance of LiFePO₄ deteriorated significantly after water immersion. Similarly, Jiang et al. [28] discussed the effect of moisture on LiFePO₄ under humid conditions. The results indicate that at temperatures ranging from 100 to 180 °C, H₂O chemically reacts with LiFePO₄, leading to a reduction in the lithium storage capacity of the material. Ahn et al. [29] investigated the impact of moisture on the electrochemical performance of LiFePO₄ (LFP)/Li₇La₃Zr₂O₁₂ (LLZ) based solid-state batteries. The results revealed that exposure of the samples to humid air led to a decrease in the conductivity rate of Li ions at the grain boundary between LFP and LLZ.

Defect-engineering is regarded as a potent strategy for enhancing the electrochemical performance of batteries due to its capacity to facilitate the diffusion of Li ions within electrode materials. Wang et al. [30] reported the successful creation and introduction of a significant quantity of defects within the crystal lattice of a LiMn₂O₄ material through defect-engineering, aiming to achieve enhanced Li-ion diffusion characteristics and rate performance in lithium-ion batteries. Lee et al. [31] combined first-principles calculations with scanning transmission electron microscopy to elucidate that Li-ion vacancies are confined within one-dimensional channels along the b-axis and are capable of migrating between adjacent Fe-Li sites. Amin et al. [32] discovered that the presence of lithium vacancies in LiFePO₄ crystals leads to a significant increase in Li-ion conductivity by more than three orders of magnitude after prolonged annealing at temperatures ranging from 300 °C to 500 °C. The activation energy for Li-ion diffusion decreased from 0.65 eV to 0.30 eV. In addition, Al doping introduces Li-ion vacancies into LiFePO₄. The presence of Li-ion vacancies significantly enhances the diffusion rate of Li ions, resulting in LiFePO₄ exhibiting superior electrochemical performance at room temperature [33,34]. Aksyonov et al. [35] discovered that OH groups can infiltrate into the Li-ion vacancies of LiFePO₄. Molecular dynamics simulation results indicate that OH groups can stably exist in LiFePO₄. However, relatively little research has been conducted on how vacancy defects affect Li-ion conduction in LiFePO₄ under humid conditions.

Herein, we investigated the effect of defect-engineering on the conduction performance of electrons and Li ions in LiFePO₄ under humid conditions. Water molecule can enter Li-ion vacancies, leading to the expansion of the LiFePO₄ crystal structure. Density functional theory calculations demonstrated that water molecules lose electrons in Li-ion vacancy, causing LiFePO₄ to transition from a semiconductor to a metal. Water molecules enhance the electronic conductivity of LiFePO₄. Furthermore, the introduction of H₂O lowers the Li-ion migration barrier, thereby reducing the rate of ion conduction. Our study provides a physical mechanism for the effects of defect-engineering on Li-ion conduction behavior.

2. Methods

The VASP 5.4 (Vienna ab initio simulation package) software via DFT (density functional theory) was employed [36]. To describe the exchange–correlation effects, the PBE functional under the GGA was utilized [37]. The PAW method was employed to describe electronic wave functions [38,39]. The cut-off energy was 520 eV. The valence electron structures were selected as 1s¹ (H), 2s¹ (Li), 2s²2p⁴ (O), 3s²3p³ (P), and 3d⁷4s¹ (Fe). The convergence criteria for energy and atomic forces were set to 10⁻⁶ eV and 0.005 eV/Å, respectively. To compensate for the underestimated correlation effect of the GGA method in the calculations for the strongly localized d-orbitals of the Fe element, the Hubbard correction (GGA + U) [40,41] was used to relax the ions and the cells, and the U_eff value for Fe atoms was taken as 4.3 eV [42]. The k-point of 5 × 3 × 3 was established. The DFT + D2 functional was employed [43]. The climbing image nudged elastic band (CI-NEB) method was used to calculate the migration paths and barriers of Li ions. The VESTA 4.6 software [44] was used to visualize the LiFePO₄ crystal structure.

3. Results and Discussion

3.1. Intrinsic Structure of LiFePO₄ and H₂O + LiFePO₄

The space group of LiFePO₄ is Pnma, and its configuration is in an orthorhombic olivine structure. In its unit cell, there are 4 Li, 4 Fe, 4 P, and 16 O atoms. Figure 1a depicts the 1 × 2 × 1 supercell of LiFePO₄. The Li ions are situated within the octahedral interstices of the unit cell, forming a six-coordinated structure with the PO₄ groups. Fe²⁺, on the other hand, occupies the tetrahedral interstices of the unit cell, also forming a six-coordinated structure with the PO₄ groups. The PO₄ groups are connected to the Li ions and Fe²⁺ through ionic bonds. The Li element and Fe element occupy the 4a and 4c Wyckoff positions in the octahedral sites, while the P element takes the 4c Wyckoff position in the tetrahedral site. In addition, due to the magnetic properties of iron, four magnetic configurations were considered and calculated for LiFePO₄, i.e., ferromagnetic (FM), A-type antiferromagnetic (A-AFM), C-type antiferromagnetic (C-AFM), and G-type antiferromagnetic (G-AFM). The results indicated that the magnetic ground state of LiFePO₄ was C-AFM. After optimization, the structural parameters of LiFePO₄ were a = 4.727 Å, b = 12.046 Å, c = 10.282 Å, and the volume was 585.47 ų. When a Li vacancy was present in LiFePO₄, its lattice constant and volume were reduced, and the cell volume decreased to 560.78 ų. The data are listed in

Table 1. Lattice constant (Å) and volume (ų) of LiFePO₄, Li_0.875FePO₄, and H₂O + LiFePO₄.

Case	a	b	c	V	Ref.
LiFePO4	4.727	12.046	10.282	585.47	this work
Li0.875FePO4	4.680	11.861	10.103	560.78	this work
H2O + LiFePO4	4.756	12.122	10.346	596.47	this work
LiFePO4	4.692	12.022	10.332	582.80	[45]

. The computational results in this study are consistent with the experimental measurements [45], indicating the reliability of the computational methods employed in this work.

In a humid environment, water molecules infiltrate the LiFePO₄ material. During the discharge process of LiFePO₄, a significant number of Li vacancies are generated; hence, there is a high probability that the water molecules will occupy these Li vacancies. Here, we discuss the structural changes of LiFePO₄ after a water molecule (H₂O) occupies a Li vacancy, as shown in Figure 1b. After structural optimization, the lattice constants were a = 4.756 Å, b = 12.122 Å, c = 10.346 Å, and the unit cell volume was 596.47 ų. In this system, there were 2 H, 7 Li, 8 Fe, 8 P, and 33 O atoms, totaling 58 atoms. In comparison to the LiFePO₄, the addition of H₂O resulted in an increase in both lattice constants and unit cell volume, and the water molecule concentration was 12.5%. This is attributed to the larger radius of the water molecules compared to the Li ions they replaced, which readily induced lattice expansion upon incorporation into the lattice. It is commonly believed that an increase in the unit cell volume of a material can provide more space for the extraction and insertion of Li ions, thereby enhancing the rate of Li-ion conduction within the electrode [46,47,48].

3.2. Electronic Properties of LiFePO₄ and H₂O + LiFePO₄

The total density of states (DOS) is used to describe the number of electronic states within a specified energy range per unit energy or volume. Figure 2 shows the total DOS of LiFePO₄ (gray shadow) and H₂O + LiFePO₄ (red curve). LiFePO₄ is a semiconductor and possesses a band gap of approximately 3.5 eV. Detailed band-gap data are provided in the energy band structure of LiFePO₄ below. As for Fe²⁺, six electrons occupy the spin-up states of the t₂g and e_g orbitals in the 3d subshell, along with one spin-down state of the t₂g orbital, while the spin-down states of the e_g orbitals remain unoccupied. Moreover, due to the C-AFM ground state of LiFePO₄, the spin-up and spin-down DOS are completely symmetric, and the total magnetic moment of the entire system is 0 μ_B. For the H₂O + LiFePO₄ system, from the red curve in Figure 2, it is evident that there are electronic states crossing the Fermi level, indicating that the addition of H₂O resulted in the presence of free electrons in LiFePO₄. The system exhibited an external magnetic moment of 1 μ_B. Therefore, upon the occupation of the Li-ion vacancies by H₂O, the contribution of the free electrons caused LiFePO₄ to undergo a transition from a semiconductor to a metal, thereby enhancing the intrinsic electronic conductivity of LiFePO₄ to a certain extent.

The electronic bands exhibited diverse energy structures due to differences in the orbitals occupied within atoms. A small band gap facilitates electron transitions, leading to better conductivity of the material. Conversely, a large band gap makes electron transitions more difficult, resulting in poorer conductivity. The band gap for pristine LiFePO₄ was calculated to be 3.542 eV, demonstrating strong agreement with theoretical and experimental studies [49,50], as shown in Figure 3a. In the band structure, the blue and red lines represent spin-up and spin-down states, respectively. In addition, the valence band maximum is located at the M-point, and the conduction band minimum is situated at the G-point, thus categorizing LiFePO₄ as an indirect band-gap semiconductor material. For pristine LiFePO₄, due to its ground state magnetic configuration being C-AFM, the spin-up state and spin-down state overlap in the electronic band structure. Furthermore, the C-AFM ground state can induce coupling effects between local electronic states and magnetic states, resulting in changes in the local density of the states, which in turn affects the electronic structure and conductivity.

In the H₂O + LiFePO₄ system, we could clearly observe that the introduction of water molecules caused a splitting of the system’s band structure. The spin-up states and spin-down states did not overlap, as shown in Figure 3b. In this case, the presence of a band crossing the Fermi level directly demonstrated the transition of LiFePO₄ from a semiconductor to a metallic behavior upon the addition of water molecule. Furthermore, the presence of water molecules led to a denser band structure throughout the H₂O + LiFePO₄ system, with an increased number of bands due to the introduction of the water molecule. This facilitated the electron transition process from the valence band to the conduction band, making it easier for electrons to move across this transition. After occupying the Li-ion vacancies, the water molecules increased the electronic conductivity of LiFePO₄, thereby enhancing the electrical performance of the electrode material to some extent in a humid environment.

3.3. Charge Properties of LiFePO₄ and H₂O + LiFePO₄

Bader charge analysis [51] is a quantitative method for computing the charge distribution among individual elements within a material. This method is commonly applied to analyze material structures, bonding characteristics, and interactions between different elements or atoms. According to Bader charge analysis, in pristine LiFePO₄, each Li, Fe, and P atom lost 0.87 e, 1.43 e, and 3.62 e, respectively, and each O atom gained 1.48 e. In the H₂O + LiFePO₄ system, each Li, Fe, and P atom lost 0.86 e, 1.44 e, and 3.60 e, respectively, and each O atom gained 1.46 e. As a whole, H₂O lost 0.38 e. At this point, LiFePO₄ gained 0.38 e from H₂O. The average charge gains and loss data are listed for each element in

Table 2. Average charge gains and losses (e) for each element in LiFePO₄ and H₂O + LiFePO₄, where a negative value indicates a loss of charge, and a positive value a gain of charge.

Case	Li	Fe	P	O	${\displaystyle {\sum }_{{\mathbf{LiFePO}}_4}}$	${\displaystyle {\sum }_{{\mathbf{H}}_2\mathbf{O}}}$
LiFePO4	−0.87	−1.43	−3.62	+1.48	0.00	—
H2O + LiFePO4	−0.86	−1.44	−3.60	+1.46	+0.38	−0.38

.

The charge density difference map is employed to analyze the variations in electron density between two systems, facilitating a detailed investigation of bonding characteristics, inter-system interactions, charge transfer, and other physico-chemical phenomena. Here, we computed the charge density difference between H₂O and LiFePO₄ within the system, as shown in Figure 4, according to the following formula:

$$\Delta \rho (z)={\displaystyle \int {\rho }_{{\mathrm{H}}_2{\mathrm{O}+\mathrm{LiFePO}}_4}(x,y,z)\mathrm{d}x\mathrm{d}y\mathrm{d}z}-{\displaystyle \int {\rho }_{{\mathrm{H}}_2\mathrm{O}}(x,y,z)\mathrm{d}x\mathrm{d}y\mathrm{d}z}-{\displaystyle \int {\rho }_{{\mathrm{LiFePO}}_4}(x,y,z)\mathrm{d}x\mathrm{d}y\mathrm{d}z}$$

where ${\rho }_{{\mathrm{H}}_2{\mathrm{O}+\mathrm{LiFePO}}_4}(x,y,z)$, ${\rho }_{{\mathrm{H}}_2\mathrm{O}}(x,y,z)$, and ${\rho }_{{\mathrm{LiFePO}}_4}(x,y,z)$ are the charge densities of H₂O + LiFePO₄ system, water molecule, and LiFePO₄ at the (x,y,z) point, respectively. In Figure 4, the blue region indicates electron depletion, while the yellow region signifies electron accumulation. We can observe that a water molecule is located in a Li-ion vacancy, surrounded by the blue region, indicating a loss of charge by the water molecule and its transfer into LiFePO₄. The charge density difference not only further confirmed the transfer of charge from water molecules to LiFePO₄, but also provided us with visual evidence of their interaction. This computational result is consistent with the conclusions obtained from Bader charge analysis, demonstrating the occurrence of charge transfer between H₂O and LiFePO₄ in a humid environment.

Furthermore, we conducted a detailed analysis of electron localization function (ELF) maps for pristine LiFePO₄ and the H₂O + LiFePO₄ system, as shown in Figure 5. The ELF method is used to study the electronic structure and bonding states of materials [52,53]. This method is employed to characterize the probability of finding an electron with a specific spin in a particular region, thereby delineating the degree of electron localization around atoms. In Figure 5a, it can be observed that there are red regions clustering around the O atoms, indicating that the electrons are predominantly localized around the O element on the (010) crystal face. After the water molecules occupy the Li-ion vacancies, the color of the red regions around the O atoms becomes lighter, indicating a decrease in the localized charge density around the O element, as shown in Figure 5b. This is consistent with the data from Bader charge analysis. When LiFePO₄ was in a humid environment, the average charge on the O atoms decreased from 1.48 e to 1.46 e, denoting a reduction in the electrons gained by the O element.

3.4. Migration Properties of the Li Ion in LiFePO₄ and H₂O + LiFePO₄

The migration barriers of Li ions within electrode materials have a significantly impact on the rate performance of lithium-ion batteries and serve as a crucial factor determining their overall performance. In LiFePO₄, the migration path of Li ions is constrained within one-dimensional channels along the b-direction [54,55]. Nishimura et al. [56] observed the one-dimensional characteristics of Li-ion diffusion channels within the LiFePO₄ material experimentally. Based on structural symmetry, it is evident that there was only one migration pathway for Li ions, with a migration barrier of ~0.45 eV. In addition, we analyzed the migration properties of Li ions in H₂O + LiFePO₄. The computational results indicated that the migration barrier for the Li ions increased to 0.50 eV, as shown in Figure 6. Compared to the pristine LiFePO₄, the migration barrier for Li ions in H₂O + LiFePO₄ increased by approximately 11.1%. Hence, the addition of H₂O increased the migration barrier for Li ions, hindering their diffusion and consequently reducing their migration rate. After a water molecule occupies a Li-ion vacancy, it can obstruct the one-dimensional diffusion channels of the Li ion, thereby hindering ion migration. Ultimately, this leads to a decline in the rate performance of LiFePO₄. In summary, moisture has a significant impact on the migration of Li ions, and the introduction of H₂O will severely impede the migration of Li ions. Therefore, during the preparation and use of actual electrode materials, it is crucial to strictly control the introduction of moisture.

4. Conclusions

In short, this work systematically investigated the influence of a humid environment on the performance of LiFePO₄ as a lithium-ion battery electrode material, using the first-principles methods. Magnetic studies revealed that the ground state of pristine LiFePO₄ is C-type antiferromagnetism. LiFePO₄ is a semiconductor with a band gap of 3.542 eV. The stable framework structure of FePO₄ allows water molecules to enter and occupy Li vacancies in humid environments. When a water molecule (H₂O) occupies a Li vacancy, the lattice constants and volume of the system increase. In the H₂O + LiFePO₄ system, the charge of 0.38 e lost by H₂O was transferred to LiFePO₄, and the charge density difference and electron localization function both confirmed this result. The total DOS and band structure indicated that the introduction of water molecules enhanced the overall electronic conductivity of the H₂O + LiFePO₄ system. At this point, the H₂O + LiFePO₄ system exhibited an external magnetic moment of 1 μ_B. In addition, the migration barrier for Li ions in pristine LiFePO₄ was 0.45 eV. In the H₂O + LiFePO₄ system, the migration barrier for Li ions increased to 0.50 eV, representing an increase of 11.1%. This severely hindered the migration performance of Li ions, resulting in a decline in the performance of LiFePO₄ during charging and discharge processes. Our study revealed the influence of water molecules on the performance of lithium-ion battery electrode materials in a humid environment. Overall, the above research results revealed the effect of a humid environment on the performance of LiFePO₄ materials, providing a theoretical basis for subsequent experimental investigations into the coupled interaction between water molecules and LiFePO₄.