Lithium Salt Screening for PEO-Based Solid Electrolytes of All Solid-State Li Ion Batteries Using Density Functional Theory

Abstract

As key components in solid-state electrolytes, lithium salts influence the electrochemical window, ionic conductivity, and ultimately the full battery’s performance. To reduce the selection time and costs while providing electric and molecular level insights into the interactions of elements and components in solid polymer electrolytes, this paper proposes a rapid screening method based on Density Functional Theory (DFT). The structure stability, electrochemical stability, and ionic conductivity of eight common inorganic and organic lithium salts were systematically investigated by analyzing five parameters: formation energy, band gap, Li⁺–anion dissociation energy, anion–PEO binding energy, and anion diffusion barriers along PEO chains. Through a comprehensive analysis of these parameters obtained from DFT, LiTFSI has been identified as the most suitable lithium salt. The electrolytes fabricated by LiTFSI exhibited better performance. This approach, characterized by its rapidness, efficiency, and low cost, provides a viable method for screening lithium salts in developing solid-state batteries.

1. Introduction

Lithium ion batteries (LIBs) have been widely used in mobile devices, electric vehicles, renewable energy storage systems, and other fields, due to their high energy density, long life, good cycling performance, and environmental friendliness [1,2]. However, along with the growing demand for higher energy density and safety, the next generation of batteries will inevitably incorporate solid-state electrolytes. The adoption of solid electrolytes can not only enhance a battery’s safety but can also significantly improve energy density by using lithium metal or silicon as its negative electrode [3,4]. Solid electrolytes include three types, namely, inorganic, organic, and composite solid electrolytes, while organic and composite electrolytes require the addition of lithium salt to provide freely moving lithium ions within the solid electrolyte.

Lithium salt greatly influences the battery’s ion transference, energy density, cycle performance, and safety [5,6,7]. High-quality lithium salts have high ionic conductivity, stable low-impedance solid electrolyte interphase (SEI) formation, better thermal and electrochemical stability, low cost, and are non-toxic and pollution-free [8,9]. Therefore, selecting an appropriate lithium salt for the solid electrolyte is necessary to meet optimal electrochemical performance.

Some researchers experimentally compare the impact of lithium salts on battery performance. As reported by S. Lee [10], a gum-like ceramic-polymer composite electrolyte consisting of Li_1.3Ti_1.7Al_0.3(PO₄)₃ ceramic powder, polyethyleneoxide (PEO), and lithium salts had ionic conductivities with different lithium salts at room temperature of 8.39 × 10⁻⁵, 9.88 × 10⁻⁵, and 6.95 × 10⁻⁵ S/cm for LiCl, LiClO₄, and LiClO₄∙3H₂O, respectively. Its initial charge and discharge capacities also vary from 70 to 90 mAh/g. Hybrid polymer-ceramic electrolytes composed of PEO and Li_1.3Ti_1.7Al_0.3(PO₄)₃ with and without LiTFSI exhibit a difference in ionic conductivity by a magnitude [11].

The above experimental comparison requires a significant amount of time to fabricate and test the electrolyte even with only two or three kinds of lithium salts. Furthermore, the results obtained in different articles cannot be directly compared due to the different mass ratios of inorganic particles and the other component ratios or fabrication technologies. Conversely, Density Functional Theory (DFT) provides another way that can significantly reduce the experiment time and resources required. How the solvation [12,13], molecular orbital [13,14,15,16], anion size [17,18], and limiting oxidation potential [19,20,21] of lithium salts affect electrochemical performance have been extensively studied, but research about the dissociation [22], interaction with polymers [23,24,25], and electrochemical stability [23] of lithium salts in solid electrolytes is relatively minimal.

Sakar [23] utilized DFT to investigate several properties of four electrolyte systems: PVDF/IL/LiClO₄, PVDF/PC/LiClO₄, PVDF/IL/NaClO₄, and PVDF/PC/NaClO₄, where PVDF represents polyvinylidene fluoride, IL is an ionic liquid, and PC is propylene carbonate. These properties include cell structures, the density of states, differential charge densities, electrochemical windows, and binding energies among each component, which provides detailed insights into their overall stability. Additionally, molecular dynamics (MD) simulations were employed to investigate the ion diffusion behavior and ionic conductivity. Pan [26] focused on the impact of anions in lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) and lithium difluoro(oxalato)borate (LiDFOB) on battery performance. The DFT results suggest that TFSI⁻ has a higher binding energy with PEO rather than DFOB⁻, which directly proves the important role of the anion in facilitating the transfer of Li⁺ in solid batteries. Given the time-consuming nature of MD simulations and inspired by Pan’s approach in [26], it is possible to investigate the effects of lithium salts on ion transport through the anion–polymer interactions.

These studies mainly concentrate on certain properties of lithium salts, such as stability or diffusivity. By integrating the methods from Sakar’s and Pan’s works and incorporating the lithium salt dissociation energies from Jiang’s research [13], an innovative multi-dimensional DFT screening model is introduced in this paper. The properties of lithium salt and PEO are investigated comprehensively by the screening model from multiple perspectives, including electronic structure, interactions between lithium salt cations and anions, and interactions between lithium salt anions and the PEO segment. It provides a comprehensive assessment of both the stability and ionic conductivity of lithium salts by analyzing five different parameters. This multi-dimensional method can understand the properties of electrolytes more accurately and deeply, providing a more efficient and accurate method for screening lithium salts to optimize electrolyte properties.

2. Materials and Methods

2.1. Lithium Salt Properties and Screening Parameters

Some researchers confirmed that the thermal stability, electrochemical stability, and ionic conductivity of lithium salts are closely related to their intrinsic parameters [23,24,26,27]. This paper employs DFT to investigate these properties of eight lithium salts by formation energy, band gap, dissociation energy, anion–PEO binding energy, and anion diffusion barrier along the PEO chain. The goal is to identify the optimal lithium salt for PEO-based electrolytes.

Table S1 presents the molecular structures and characteristics of common lithium salts in LIBs. LiClO₄ has good comprehensive performance, but it is easy to oxidize with the organic solvent in the electrolyte due to the high valence state of Cl. This can lead to safety issues such as combustion and explosions in LIBs. LiPF₆, as a commercially used lithium salt, has high ionic conductivity and good electrochemical stability. However, its poor thermal stability and decomposition products can lead to a decline in battery performance. On the other hand, fluorine-containing organic lithium salts possess higher oxidative stability, better ionic conductivity, and superior film-forming properties. Due to the differences in phases and components in liquid and solid electrolytes, the performance of these lithium salts in all solid-state batteries needs further investigation.

Based on the characteristics of high-quality lithium salts, specific parameters should be determined for screening. Given the complex formation mechanism of the SEI membrane, this paper focuses on the structure stability, electrochemical stability, and ionic conductivity of lithium salts during the screening process. These properties are essential for determining whether lithium salts can perform their fundamental functions within a lithium battery system, and the relevant parameters affecting these performances are studied. The stability of lithium salt is analyzed from its formation energy and band gap. The influence of different lithium salts on Li-ion transition is analyzed from the Li⁺–anion binding energy, the anion–PEO adsorption energy, and the diffusion barrier of the lithium salt anion along the polymer chains.

2.2. Theoretical Computation

This paper investigates the above five parameters of widely used and environmentally friendly lithium salts: LiClO₄, LiBF₄, LiPF₆, LiPO₂F₂, lithium bis(oxalate) borate(LiBOB), LiDFOB, lithium bis(fluorosulfonyl)imide (LiFSI), and LiTFSI. Their electronic structure, stability, and anion diffusion capability were evaluated by DFT as implemented in the Cambridge Sequential Total Energy Package (CASTEP) [28]. The geometry optimization for electronic structure and stability evaluation was performed using the Generalized Gradient Approximation (GGA) Perdew–Burke–Ernzerh (PBE) [29] functional. An energy cutoff of 520 eV and a k-point grid of 2×2×3 were selected, which are optimized through a testing process. The convergence tolerance of energy, max force, max stress, and max displacement were set to 5.0×10⁻⁶ eV/atom, 0.01 eV Å⁻¹, 0.02 GPa, and 5.0 ×10⁻⁴ Å, respectively. The diffusion property was simulated by transition state search, which adopted the complete linear synchronous transit/quadratic synchronous transit (LST/QST) protocol, and all the transition states were confirmed by phonon spectrum analysis.

The formation energy $E_f$ [30] is calculated using Formula (1) to analyze the structure stability of lithium salt. The value of the formation energy represents the energy released by the compound synthesized by the corresponding element:

$$E_f={\displaystyle \frac{E\left(A_m{B}_n{C}_k\right)-m\ast E\left(A\right)-n\ast E\left(B\right)-k\ast E\left(C\right)}{m+n+k}}$$

(1)

where $E\left(A_m{B}_n{C}_k\right)$ is the energy of the ternary compound $A_m{B}_n{C}_k$; $E\left(A\right)$, $E\left(B\right)$, and $E\left(C\right)$ are the normalized energies of the corresponding elements A, B, and C; and m, n, and k are the number of corresponding atoms in the chemical formula. The formation energy of other multicomponent compounds can be calculated similarly.

Since the Li⁺ of lithium salt can move freely only after dissociation into Li⁺ and anion groups, the energy required to decompose lithium salt into anion and Li⁺ is measured by the binding energy of Li⁺ and anion. The lower the binding energy, the more difficult it is to decompose. For compound AB, which consists of $A^{+}$ and $B^{-}$, its binding energy $E_b$ can be computed as follows [22]:

$$E_b=E\left(AB\right)-E\left(A^{+}\right)-E\left(B^{-}\right)$$

(2)

where $E\left(AB\right)$ represents the energy of the compound, and $E\left(A^{+}\right)$ and $E\left(B^{-}\right)$ represent the energy of cation A and anion B, respectively. For lithium salt dissociation energy calculation, cation A is Li⁺, and anion B is lithium salt anion.

After lithium salt is dissociated, both Li⁺ and anion can move in the electrolyte. By limiting the movement of lithium salt anions, the Li⁺ transference number can be increased. In turn, during the charge and discharge process, concentration polarization was weakened, improving the battery’s power density. Formula (2) can also be used to calculate the binding energy of the anion and polymer [24] and can help to analyze which lithium salt anion is less likely to move in the polymer. In the equation, the total energy of the anion adsorbed by the polymer is represented by $E\left(AB\right)$, while $E\left(A\right)$ and $E\left(B\right)$ represent the energy of the anion and polymer, respectively.

2.3. Structure of Lithium Salts and PEO

This paper investigated eight commonly used lithium salts [8,31]. The initial structures of LiBF₄, LiPF₆, LiBOB, LiDFOB, LiFSI, and LiTFSI were imported from the Materials Project database [32]. Their Materials Project IDs are mp-12403, mp-9143, mp-5656165, mp-1105683, mp-559971, and mp-557395, respectively. The initial structures of LiClO₄ and LiPO₂F₂ were obtained by modifying LiBF₄ and LiPF₆. The optimized lithium salt structures are listed in Figure 1. The cell parameters of each structure are specified in Table S2. The atomic charge and average bond length of each lithium salt anion are specified in Tables S3 and S4.

LiPF₆, LiBF₄, and LiPO₂F₂ belong to the trigonal crystal system, while LiClO₄, LiBOB, LiDFOB, and LiTFSI belong to the orthorhombic crystal system. LiTSI belongs to the hexagonal crystal system. Among the inorganic lithium salts, the lattice parameters of LiPF₆, LiBF₄, and LiPO₂F₂ show little variation due to their similar structure and bond length. However, LiClO₄ has a larger cell size than the others because of the strong repulsive force between oxygen atoms. Among the organic lithium salts, LiDFOB and LiBOB have similar cell sizes since LiDFOB is derived from LiBOB. The distinct molecular structure and charge distribution cause the difference in the cell size of LiTFSI and LiTSI.

The PEO segment structure is built according to the literature [33]. The PEO chain within the periodic boundary is built as shown in Figure S1. The periodic boundary is set to ensure that the distance between the lithium salt anion and its periodic mirror image is greater than 10 angstroms after the PEO chain adsorbs the lithium salt anion; thus, the length of the PEO chain segments within this periodic boundary is set to 10. When PEO interacts with other ions, although the overall structure remains periodic, the local structure around the adsorbed ions undergoes alterations. This is consistent with experimental observations [34,35] where adding lithium salts or inorganic fillers alters the crystallinity of PEO.

2.4. Screening Rules

Based on the above analysis, five parameters of lithium salt were calculated for the screening properties: formation energy (lithium salt stability), band gap (electrochemical stability), dissociation energy (energy required to dissociate into Li⁺ and anion), binding energy (binding energy of anion and PEO), and diffusion barrier (the barrier for anion to diffuse along the PEO chains).

The first three parameters focus on the lithium salt’s intrinsic properties, and the last two assess the interaction between the lithium salt and PEO. The thermal decomposition of lithium salt can be tested by the thermogravimetric method, while the band gap can be calculated by ultraviolet–visible spectroscopy (UV-VIS) diffuse reflection test results. However, the lithium salt dissociation energy, anion–polymer binding energy, and anion diffusion barrier cannot be measured experimentally. These parameters are important for electrolyte ionic conductivity and the lithium ion migration number.

Numerical discrepancies between DFT calculations and experimental results are inevitable due to multiple factors such as computational approximations and experimental conditions; however, the trends predicted by DFT calculations should be consistent with those observed experimentally. Therefore, the above five parameters are compared with the relevant literature to ensure that the calculation results in this paper have theoretical guiding significance for experiments.

Since the above five parameters contain both positive and negative values, they need normalization for effective comparison and screening. The structure stability, dissociation energy, binding energy, and diffusion barrier are normalized to a range of [0,1] following Algorithm 1. The electrochemical values, however, are normalized to the range (−1,1] as depicted in Algorithm 2. This adjustment of overriding negative values is necessary for the lithium salt with a small band gap, which reduces the electrochemical window of solid electrolytes. Algorithm 1 begins by converting all negative values in the input array to their absolute values. An auxiliary element, T[9], is then initialized to 0 and appended to the array. The entire modified array (including T[9]) undergoes normalization, and the resulting normalized values are stored in the output array N[1:8].

Algorithm 2 first identifies the minimum value in the input array. It processes each element as follows: values equal to 3.9 are set to 0, while values below 3.9 are adjusted using a specific formula. After normalization, one element (previously modified during processing) is negated. The final normalized and adjusted array is then stored in N[1:8].

In this context, a value closer to 1 for an indicator signifies a better outcome. In the pseudocode, T[1:8] represents the original values of the parameters that need normalization, while N[1:8] denotes the resulting normalized values. Summarizing all normalized parameters of each lithium salt and ranking them, the lithium salt with the highest score is identified as the optimal one for PEO-based solid electrolytes.

Algorithm 1: Normalize to [0,1]

Input: T[1:8]  Output: N[1:8]  for each T[i] do   if (T[i] < 0) then    T[i] = abs (T[i]);   endif  endfor  T[9] = 0;  Normalize(T[1:9]);  N[1:8] = T[1:8]  Return N[1:8]

Algorithm 2: Normalize to (−1,1]

Input: T[1:8]  Output: N[1:8]  Min = min(T[1:8]);  for each T[i] do    if (T[i] == 3.9) then T[i] = 0;    else if (T[i] < 3.9) then j = i; T[i] = 3.9 + (3.9-T[i]);    end if  endfor  Normalize(T[1:8]); T[j] = -T[j];  N[1:8] = T[1:8]  Return N[1:8]

3. Results

3.1. Electrochemical Stability

As the molecular orbitals of lithium salts have been extensively studied, this paper utilized the electrochemical stability analysis method of inorganic solid electrolytes to calculate the state density and band gap. Since the band gap value is the upper bound of the electrochemical stability window, the partial density of states and the band gap are analyzed, as shown in Figure 2. Inorganic lithium salt has a wider electrochemical window than organic lithium salt except LiPO₂F₂. LiBF₄ is the most electrochemically stable inorganic lithium salt, while LiTFSI is the most stable organic lithium salt.

As can be seen, the lithium salts sequenced by band gap are LiBF₄, LiPF₆, LiClO₄, LiTFSI, LiFSI, LiPO₂F₂, LiDFOB, and LiBOB. The band gap values’ variation trend aligns with the HOMO and LUMO data of Pandian’s [16] and Jiang’s [13] calculation results. By comparing the band gap values of the lithium salts with the electrochemical window of PEO (3.9V), it is possible to determine whether the addition of this lithium salt widens the electrochemical window of the solid-state electrolyte. In summary, in addition to LiBOB and LiDFOB, adding other lithium salts can expand the electrochemical window of the PEO-based solid-state electrolyte. Furthermore, the appropriate cathode material can be identified based on the band gap value. As noted in the literature [31], salts such as LiPF₆, LiTFSI, and LiFSI are suitable for high-voltage lithium metal batteries.

Based on the PDOS of Figure 2, the electronic distribution of inorganic lithium salts is strongly localized, indicating that the interaction of atoms in the anion is tightened. Inorganic lithium salt also has a lower bonding orbital energy compared to organic lithium salt. This indicates that the former requires more energy to decompose, leading to lower formation energy. Similarly, the bonding orbital energy of LiPF₆ is lower than LiPO₂F₂, resulting in a larger formation energy. The valence band bottom of LiBF₄ is contributed by Li elements, indicating that LiBF₄ can withstand higher voltage after decomposing to Li⁺ and BF₄⁻.

To further investigate whether different chemical bonds exist in the lithium salts, its electron localization function (ELF) is calculated. As shown in Figure 3, the electrons of Li are highly delocalized, which is consistent with the characteristics of Li as a metallic element, which allows its outer electrons to be easily delocalized. In contrast, F and O elements are highly electronegative, resulting in a significant localization of electrons around them, indicating that all bonds involving F and O exhibit polar covalent characteristics. The electron cloud between the C and S atoms is relatively evenly distributed between the two atoms and is a non-polar covalent bond. The electron cloud between C (or N) and S atoms is distributed relatively evenly, indicating a non-polar covalent bond. For all lithium salts considered in this paper, the electrons associated with Li are highly delocalized, while other elements are linked through covalent bonds. In this paper, considering all lithium salts, the electrons associated with Li are highly localized, while other elements are linked through covalent bonds. During the dissociation process, it is evident that Li⁺ is released first, while the structure of the anion is preserved due to the strength of the covalent bonds.

3.2. Lithium Salt Stability

Referring to the methods in References [36,37], the stability of lithium salts is analyzed by the formation energy calculated by Equation (1), and the results of each lithium salt are shown in Figure 4. The temperature is fixed at T = 0 K during this calculation, and the dispersion correction is included to ensure the accuracy of energy calculations. Inorganic lithium salts exhibit a formation energy range of −2.46 to −2.93 eV, with LiBF₄ being the most stable structure. Organic lithium salts have a formation energy range of −1.31 to −1.98 eV, with LiDFOB being the most stable one. Based on this, it can be inferred that inorganic lithium salts have better structure stability than their organic counterparts. The variation trends in the calculated formation energy of LiBF₄, LiPF₆, LiBOB, LiDFOB, LiFSI, and LiTFSI in this study align with the formation energy variation trends provided in the Materials Project database.

3.3. Ionic Conductivity

Ionic conductivity is evaluated by three parameters: dissociation energy, anion—PEO binding energy, and anion diffusion barrier along the PEO chain. The dissociation energy of each lithium salt (Figure 5a) is calculated by Equation (2) to analyze its ability to dissociate lithium ions as in Li’s work [12]. The dissociation energy of inorganic lithium salts ranges from −3.19 to −4.60 eV, with LiPO₂F₂ requiring the least amount of energy for dissociation. In comparison, organic lithium salts range from −4.08 to −5.24 eV, with LiTSI requiring the minimum energy for dissociation. The dissociation of organic lithium salts requires more energy due to its more complex structure and the greater variety and number of chemical bonds. The variation trends in dissociation energy for LiDFOB, LiTFSI, and LiFSI calculated in this paper are consistent with the literature [12]. The trends in dissociation energy for LiPF₆, LiClO₄, and LiBF₄ calculated in this paper align with the literature [38].

As mentioned above, the diffusion behavior of lithium salt anion directly affects the ionic conductivity. The anion diffusion property was evaluated by the binding energy between the lithium salt anion and PEO as well as the anion diffusion barriers along PEO. The energy required for the detachment of anion from its current adsorption site depends on the binding energy between the anion of lithium salt and the polymer. This kind of binding energy was used to deeply understand the interaction of the polymer and lithium salt molecules [24]. The more negative the binding energy, the harder it is for the lithium anion to dissociate from the current adsorption site.

Ten different adsorption sites are considered for each lithium salt, due to the restriction of cell parameters and repeatability of the PEO chain. The coordinates of anionic central atoms in the ten adsorption structures of each lithium salt are shown in Table S5. The structures shown in Figure S1 were established by each lithium salt anion and PEO structure. The binding energy of each lithium salt anion and PEO is calculated according to Equation (2), and the results are shown in Figure 5b.

The binding energy of the anion in inorganic lithium salts is higher than that of organic lithium salts due to the smaller size of the inorganic anion. Additionally, the binding energy between lithium salt anions and PEO is positively correlated with the volume of the anion; larger anionic volumes lead to stronger interactions and more negative binding energy. TFSI⁻ has the lowest binding energy, which indicates that its anions bind most strongly to PEO, resulting in poor mobility. Consequently, compared to other lithium salts, TFSI⁻ has the least effect on reducing the ionic conductivity of solid-state electrolytes. Compared to the other anions, BOB⁻ and TFSI⁻ are heavily adsorbed from their current adsorption sites, which could result in less reduction of ionic conductivity and lithium ion mobility.

The mobility of different anions of lithium salts across neighboring adsorption sites of the polymer chain was also assessed. Among the above 10 adsorption configurations, the lowest-energy structure was chosen as the initial state for diffusion barrier calculation, and the adjacent adsorption site of the anion was used as the final state. The diffusion barriers of anions are calculated by the transition state (TS) search method, as shown in Figure 5c. The diffusion barrier of the inorganic lithium anion is lower than that of organic lithium salts. Meanwhile, the organic lithium anion diffusion barrier increases with an increase in the length of the anion chain. Specifically, the diffusion barriers for TFSI⁻ and FSI⁻ are higher, indicating a lower ability to migrate along the PEO chain. According to Serife’s study [18], the ionic conductivity of the composite electrolyte containing LiBOB is higher than LiPF₆. It is suggested that the anion’s diffusion barrier is the primary factor that influences ionic conductivity, rather than its binding energy to PEO.

To further investigate the interaction between the lithium anion groups and polymers, the electrostatic potentials of PEO and lithium salt anion are calculated as displayed in Figure 6. Based on the electrostatic potential results, the negative charge of the PEO chain is predominantly concentrated near the O atom. On the other hand, the positive charge is evenly distributed around the C atom and the H atom. This implies that the freely moving lithium ion within the solid electrolyte can be adsorbed by the O atoms and move along the chain segment. The positive and negative charge centers of the inorganic lithium salt anion largely coincide. The BF4⁻, ClO4⁻, and PF6⁻ groups exhibit a uniform electric field distribution due to their symmetrical structures. The differing charge amounts of F and O in LiPO₂F₂ lead to notable variations in the potential near F and O atoms. Since the charge of the O atom is greater than that of the F atom, it interacts more strongly with the polymer. The binding energy of ClO₄⁻ and PEO is the lowest among the inorganic lithium salts. In the case of the organic lithium salt anion, the positive and negative charge centers do not align. The positive charge is predominantly concentrated near the carbon atom, while the negative charge is concentrated around the O and F atoms. The local electrostatic potential of the LiTFSI and LiDFOB anions is lower than that of LiFSI and LIBOB, which indicates that their corresponding adsorption energies are also lower.

3.4. Screening Results

To screen the optimal lithium salt, radar charts (Figure 7) are plotted based on the relevant parameters of several lithium salts according to previous research findings. Inorganic lithium salts show good structure and electrochemical stability, while organic lithium salt anions have high diffusion barriers. In terms of stability, inorganic lithium salts are prioritized, whereas organic lithium salts are favored for enhancing ionic conductivity. The lithium salts have been evaluated and ranked from highest score to lowest as follows: LiTFSI, LiPF₆, LiBF₄, LiFSI, LiClO₄, LiPO₂F₂, LiBOB, and LiDFOB. Consequently, LiTFSI is predicted as the optimal lithium salt for PEO-based solid-state electrolytes.

The results of this paper are consistent with some experimental results, demonstrating that this method is fast, low cost, and effective. Mauger [39] compared the ion migration characteristics, ion-pair dissociation ability, solubility, and stability of LiBF₄, LiClO₄, LiPF₆, LiAsF₆, LiTf, and LiTFSI, concluding that LiTFSI is the optimal lithium salt in lithium batteries. References [26,40] showed that the LiTFSI/PEO electrolyte outperformed the LiFSI/PEO and LiDFOB/PEO electrolytes in ionic conductivity, capacity retention, and rate performance. DFT in [22] attributed this to the lower LiTFSI-PEO binding energy. Reference [33] confirmed the order of the ionic conductivity of PEO-based electrolytes as follows: LiTFSI > LiClO₄ > LiBF₄. These results validate the theoretical calculated and screening results in this paper.

This rapid screening method can be adapted to other polymer matrices. The first step is to substitute the PEO structure with a relaxed target polymer structure. After this substitution, it is crucial to re-identify the anion adsorption sites in the new polymer matrix. The binding energy and diffusion barrier can then be reassessed using the new structure. By applying the same screening method (criteria 3.9 V in Algorithm 2 need to be replaced according to the properties of another polymer), the optimal lithium salt for another polymer electrolyte can be determined.

4. Conclusions

The lithium salt is an essential additive in polymer and composite solid-state electrolytes. It should have good thermal and electrochemical stability and low anion diffusion properties. This paper proposes a fast screening method of lithium salts for PEO-based solid electrolytes. Five screening parameters are calculated by DFT, which include the formation energy, band gap, dissociation energy, PEO—anion binding energy, and the diffusion barrier of the anion along the PEO chain. From a comprehensive view of the normalized five parameters, it is suggested that LiTFSI is a preferred option for PEO-based solid-state electrolytes. This screening method can be applied to other polymer-based electrolytes and can also be utilized for the modification and innovation of lithium salts by high-throughput computational methods.