Fluorine anion modification of high-voltage Mn-rich phosphate cathodes for high-power and long-lasting sodium-ion batteries

doi:10.21203/rs.3.rs-1847341/v1

Download PDF

Article

Fluorine anion modification of high-voltage Mn-rich phosphate cathodes for high-power and long-lasting sodium-ion batteries

https://doi.org/10.21203/rs.3.rs-1847341/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

High-voltage cathodes with high power and stable cyclability are needed for future high-performance sodium-ion batteries. However, the low electronic conductivity and poor capacity retention impede the development of Mn-rich phosphate cathodes. Here, a light-weight F doping strategy is reported to modify high-voltage Na₄MnCr(PO₄)₃ cathodes. The light-weight F dopant could decrease the energy gap to 0.22 eV from 1.52 eV and strengthen the adjacent chemical bonding around Mn as confirmed by density functional theory calculations, which ensure the boosted electronic conductivity and suppress Mn dissolution. The combination of in-situ and ex-situ techniques determine that the F dopant has no influence on the Na⁺ storage mechanisms. As a result, an outstanding rate performance up to 40 C and an improved cycling stability (1000 cycles at 20 C) are thus achieved (best reported performances until now). This work presents an unusual and widely available light-weight anion doping strategy for high-performance polyanionic cathodes.

Carbon neutrality has been proposed globally to deal with the energy crisis and climate change¹. Nowadays, the development of energy storage systems (ESSs) is a critical project that needs to consider cost-effectiveness, high round-trip efficiency, environmental friendliness, and long lifespan, etc^2–5. The currently widely used lithium-ion batteries (LIBs) might be a candidate for this purpose^6–8. However, they will be excluded when the limited resources and underlying political disputes are taken into consideration^{9, 10}. Therefore, sodium-ion batteries (SIBs) sharing an analogous “rocking-chair” mechanism to LIBs emerge as promising alternatives in large-scale ESSs due to the omnipresent availability of Na resources^{11, 12}. Over the past decade, numerous efforts have been dedicated to developing high-performance cathode materials for SIBs, which mainly include layered oxides and polyanionic compounds^{13, 14}. Layered oxides (Na_xMeO₂, Me = Mg, Fe, Cu, Ni, Mn, etc.) are featured with facile synthesis and high capacities but are still confronted with inferior structural stability, noticeable volume change, and low operating voltage, which hinder their application in ESSs that require long lifespan, high power density, and stable cycling performances^{15, 16}.

On the contrary, polyanionic compounds especially transition metal phosphates are endowed with excellent structural stability (strong covalent bond, e.g., P-O bonds) that can help endure repeating extraction/insertion of Na⁺ and higher output voltage than oxides, which is associated with inductive effects of the polyanion groups (e.g., (PO₄)^3-)^{13, 17–19}. Furthermore, the electrochemical potential tunability is presumed to be a promising characteristic to increase overall voltage and activate multi-electron reaction (higher capacity) by involving different transition metals in their structure and adjusting the operating voltage window of SIBs, which can help compensate the lower output voltage of SIBs compared to LIBs due to the 0.3 V higher inherent electrochemical potential of Na⁺/Na than that of Li⁺/Li^20–22. To illustrate, different transition metal redox couples in the same phosphate framework display various voltages versus Na⁺/Na, such as V²⁺/V³⁺ (~ 1.6 V), Ti³⁺/Ti⁴⁺ (~ 2.1 V), Fe²⁺/Fe³⁺ (~ 2.5V ), V³⁺/V⁴⁺ (~ 3.4 V), Mn²⁺/Mn³⁺ (~ 3.6 V), V⁴⁺/V⁵⁺ (~ 4.0 V) and Mn³⁺/Mn⁴⁺ (~ 4.2 V), etc^{13, 20, 23}. It hence seems that Mn-based phosphates can be good candidates as high-voltage cathodes.

Among the polyanionic compounds, Na superionic conductor (NASICON) type phosphates (Na_xMM^’(PO₄)₃, where x = 1 ~ 4, M/M^’= Cr, Mn, Fe, Ti, V, Zr, Al, etc.) have attracted extensive and increasing attention in recent years due to their ability to foster rapid Na⁺ diffusion, their three-dimensional robust structure as well as their superior air/water endurability^{20, 24–26}. Na₃V₂(PO₄)₃ stands out as the most representative NASICON material with good electrochemical properties and a moderate voltage platform of 3.4 V (V⁴⁺/V³⁺).²⁷ Cation substitution recently is emerging as a valid strategy to reduce the cost, increase the voltage and enhance the electrochemical performances of NASICONs²⁰. For example, Goodenough et al. partially substituted V with low-cost Mn to synthesize Na₄MnV(PO₄)₃²⁸. Apart from the aforementioned 3.4 V plateaus, a higher platform of 3.6 V (Mn³⁺/Mn²⁺) was obtained²⁹. However, a higher potential redox reaction of Mn⁴⁺/Mn³⁺ (~ 4.2 V) is absent from this material; besides, although the V⁴⁺/V⁵⁺ (~ 4.0 V) plateaus could be activated when extending the cut-off voltage to > 4.1V, it rapidly vanished in the second cycle, which results in a low energy density (< 380 Wh kg^-1) and impedes its further application^{30, 31}. Tremendous efforts have been dedicated since 2016 to explain this phenomenon and to completely activate V⁴⁺/V⁵⁺ (~ 4.0 V) as well as Mn⁴⁺/Mn³⁺ (~ 4.2 V) redox reactions in Na₄MnV(PO₄)₃ but limited progress has been made^{30, 32–35}.

In this regard, Ceder et al.³⁶, Zhang et al.³⁷, and Liu et al.³⁸ almost simultaneously reported the Na₄MnCr(PO₄)₃ material that could fully activate the high-potential redox couples of Mn⁴⁺/Mn³⁺ (4.2 V) and Mn³⁺/Mn²⁺ (3.6 V), and partially activate Cr⁴⁺/Cr³⁺ (4.4 V) redox reaction, which gave rise to the highest output voltage among NASICONs. When accompanied by multi-electron reactions, Na₄MnCr(PO₄)₃ could deliver an ultrahigh energy density of ~ 566 Wh kg^-1 (1.4–4.6 V versus Na⁺/Na), which broke the record of the NASICON family. Similar to other Mn-based phosphate materials, nevertheless, it still suffers from low intrinsic electronic conductivity and inferior cycling stability (obvious capacity fading within 20 cycles)³⁶, which may stem from the Mn²⁺ dissolution in the electrolyte and electrolyte decomposition at a high cut-off voltage (> 4.5 V)^{17, 25, 39, 40}. To our knowledge, there is no relevant work to further improve the electrochemical performances of Na₄MnCr(PO₄)₃. Carbon coating and cation doping have been broadly studied in the past years as efficient approaches to modify polyanionic materials but still face tremendous challenges. Carbon coating (usually ~ 10 wt.%) would inevitably increase the inactive mass of the electrode and thus lower the overall energy density. It is also challenging to form a uniform carbon coating layer and the conductive coating layer alone cannot solve all the above issues. The heteroatom doping can be an effective approach to further tune the electronic and microstructures of cathode materials. As for the cation doping, heavy inactive metal ions (e. g., Al³⁺, Mg²⁺, Ca²⁺, and Cu²⁺, etc.) also lower the cathode capacity. Therefore, it is highly desirable to develop a novel doping strategy based on a light-weight element with high electronegativity to increase the intrinsic electronic conductivity and to stabilize the structure with a minimal capacity reduction. The highly electronegative F can be a promising dopant candidate in consideration of its strong interaction with metals and relatively low atomic weight. The underlying mechanisms for doping enhancement of properties are still not well understood.

Herein, we report a novel doping strategy via introducing a well-controlled amount of light-weight F into Na₄MnCr(PO₄)₃ (i.e., Na_3.85MnCr(PO_3.95F_0.05)₃, denoted as NMCPF; NMCP for the pristine sample). The function of the light-weight F dopant is deeply studied by both experimental and theoretical approaches, which was subsequently shown to contribute to an enhanced electronic conductivity and suppress the manganese dissolution. The comprehensive analysis of in-situ and ex-situ measurements confirmed that the F dopant has no impact on structural evolution (i.e., both solid solution and two-phase reactions are involved) and charge compensation mechanisms (Cr⁴⁺/Cr³⁺, Mn⁴⁺/Mn³⁺, and Mn³⁺/Mn²⁺ redox reactions) of the original Na₄MnCr(PO₄)₃. It turned out that an excellent rate performance up to 40 C and an improved cycling stability (1000 cycles at 20 C) were achieved. This work proposes a novel and broadly applicable light-weight F doping strategy for high-performance polyanionic SIB cathodes without side effects.

Materials Characterizations. The designed NASICON Na_4 − 3xMnCr(PO_4 − xF_x)₃ (x = 0, 0.01, 0.02, 0.05, 0.1, 0.15 and 0.2) materials were synthesized by a simple sol-gel method combined with a high-temperature sintering process under an argon atmosphere. The X-ray diffraction (XRD) results (x = 0 ~ 0.05) in Supplementary Fig. 1 present analogous patterns, which determined a pure Na₄MnCr(PO₄)₃ phase within these samples and proved the partial introduction of F had no impact on the structure within this x value range. Of note, NaPO₃ (JCPDS NO. 00-003-0688) and Na₃PO₄ (JCPDS NO. 00-030-1232) impurity phases appear with x value higher than 0.05, in line with the corresponding relatively larger values of R_wp in the XRD Rietveld refinement results (Supplementary Fig. 2). In Supplementary Fig. 3, it can be clearly found that the rate properties improved initially with an increased F ratio (x = 0, 0.01, 0.02, 0.05) and reached a peak value at x = 0.05, accompanied by a descending trend when further increasing x. With the purpose of disclosing the underling mechanism, the bond lengths of Na(2)-O in different samples were calculated (Supplementary Fig. 4). It can be clearly seen that at the initial stage, there was an increasing trend with a maximum Na(2)-O distance of 2.637 Å (x = 0.05) and then the distance decreased with higher fluorine content. According to previous works, longer the bond length of Na(2)-O (weaker bond strength) is associated with faster Na⁺ diffusion, which explains the performance discrepancy of different samples in Supplementary Fig. 3 ^41–43. Therefore, the material with x = 0.05 was selected as the optimal one (NMCPF; the baseline sample (x = 0) is denoted as NMCP).

XRD Rietveld refinements determined that both NMCPF (Fig. 1a) and NMCP (Supplementary Fig. 2, x = 0) were indexed to the R_3c space group with low R_wp values of 1.52% and 1.48%, respectively. The calculated geometric parameters, atomic coordinates and crystallographic information are listed in Supplementary Table 1 and Supplementary Table 4. Figure 1b provides an illustration of their crystalline structure based on the XRD Rietveld refinement results. Both samples crystalized in a similar rhombohedral NASICON structure: Cr and Mn atoms share the same 12c site with 50% fraction for each position; [PO₄] or [PO_3.95F_0.05] tetrahedra are corner-shared with [CrO₆] or [MnO₆] octahedra to construct the lantern-like [MnCr(PO₄)₃] or [MnCr(PO_3.95F_0.05)₃] constituents, and thus two Na⁺ host sites are formed. Na(1) (6b site) is sixfold coordinated while the other Na(2) (18e site) is eightfold coordinated which tends to be more easily extracted/inserted from/to the NASICON structure due to the weaker bonding force of Na(2)-O²⁸. As anticipated, F atoms partially occupy in O1 and O2 sites with the fractions of 1.2% for each position from Supplementary Table 4. According to the valence-induced mechanism, the less negative charge stemming from the substitution of O²⁻ by F⁻ should be compensated by cation deficiency (i.e., Na⁺ deficiency) to form Na_3.85MnCr(PO_3.95F_0.05)₃, which coincides with previous reports^{42, 43}.

Inductively Coupled Plasma- Optical Emission Spectrometry (ICP-OES) was carried out to further determine the chemical formula. As shown in Supplementary Fig. 5, the atomic ratios were measured to be Na(3.84), Mn(1), Cr(1.01) and P(2.95) for NMCPF and Na(3.96), Mn(1), Cr(1.01) and P(2.97) for NMCP, which agree well with previous discussions. These results indicate that some Na vacancies exist in NMCPF as illustrated in Fig. 1b. The Rietveld refinements results of all samples could be found in Supplementary Fig. 2; their detailed structural information was listed in Supplementary Table 1–6. Transmission electron microscopy (TEM) energy dispersive spectrometry (EDS) mapping images of NMCPF (Fig. 1f) and NMCP (Supplementary Fig. 6) present a uniform distribution of all elements throughout these particles, which confirms that F has been successfully introduced into NMCPF. Solid-state nuclear magnetic resonance (NMR) spectroscopy of ¹⁹F was performed to further study the fluorine substitution. The broad resonance of -80 ~ -180 ppm in Supplementary Fig. 7 determined that F atoms partially substituted O atoms in NMCPF⁴². The combination of XRD Rietveld refinements, EDS mapping, and NMR results validates that O atoms were partially occupied by F atoms within NMCPF structure, which provides compelling evidence that this light-weight F doping strategy could be applied to other phosphates or even wider polyanionic materials.

Furthermore, TEM was employed to probe the microstructure and morphology of NMCPF and NMCP. As displayed in Fig. 1c-d and Supplementary Fig. 6a-b, both NMCPF and NMCP particles are homogenously coated with an amorphous carbon layer of 2 ~ 5 nm. Similar irregular morphologies as they displayed, the size of NMCPF (0.5 ~ 1 µm) is smaller than that of NMCP (1 ~ 2 µm), indicating that the F dopant is favorable for suppressing the particle growth to some extent and thereby achieved a shortened diffusion length for Na⁺ in NMCPF⁴⁴. Besides, high-resolution (HR) TEM images (Fig. 1d and Supplementary Fig. 6b) show a lattice fringe of 0.32 nm, which corresponds to the (20_4) lattice plane in both NMCPF and NMCP. The selected area electron diffraction (SAED) in Fig. 1e and Supplementary Fig. 6c determined the well-crystallized NMCPF and NMCP with the R_3c space group.

Thermogravimetric analysis (TGA) was adopted to determine the contents of carbon within the samples. From Supplementary Fig. 8, the amounts of carbon were shown to be 10.85% (NMCPF) and 8.83% (NMCP), respectively. As shown in Supplementary Fig. 9, NMCPF and NMCP possessed identical Fourier transform infrared (FT-IR) spectra: the stretching or bending vibration signals of [CrO₆] or [MnO₆] octahedra and [PO₄] tetrahedra are located at ~ 1000 cm^{− 1 31, 37}. Raman spectra in Supplementary Fig. 10 reveal two representative peaks of 1340 cm^− 1 (D-band) and 1590 cm^− 1 (G-band); the intensity ratios were calculated to be 0.74 (NMCPF) and 0.70 (NMCP), demonstrative of partial graphitization of carbon for these two samples and enhanced electronic conductivity for NMCPF^{31, 41}. X-ray photoelectron spectroscopy (XPS) results (Supplementary Fig. 11) display the existence of Na, Cr, Mn, P, O, C for both NMCPF and NMCP. But it is of note that F only exists in NMCPF and the F1s has a characteristic peak at 683.93 eV, which is in good agreement with previous F-doped works^{42, 44}. On the contrary, there are no distinguishable signals of F in NMCP. In addition, Mn 2p and Cr 2p spectra identified the signals of Mn²⁺ and Cr³⁺ in both NMCPF and NMCP^{45, 46}. Thus, the above results confirmed that both NMCPF and NMCP were successfully prepared.

Electrochemical Performances. The CR2032 type coin cells with metallic Na anode were fabricated to evaluate the effectiveness of our strategy. In Fig. 2a, two redox couples of 3.75/3.44 V and 4.24/4.10 V could easily be found in the CV curves, which correspond to the oxidation/ reduction of Mn²⁺/Mn³⁺ and Mn³⁺/Mn⁴⁺, respectively³⁷. The highly overlapped profiles indicated an excellent reversibility of de-/sodiation processes for both NMCPF and NMCP; the slightly deviated CV curves can be ascribed to the formation of the solid electrolyte interphase (SEI) layer¹⁸. And the narrowed potential gap of NMCPF (0.01 V; 0.03 V for NMCP) showed an improved kinetics after the introduction of fluorine within the NMCP structure. In Fig. 2b, two distinguishable plateaus located at 3.65 V and 4.2 V could be observed as well, thus delivering specific capacities of 110.7 mA h g^− 1 (NMCPF) and 107.1 mA h g^− 1 (NMCP) at 0.1 C accompanied by successive phase transformations of Na_3.85MnCr(PO_3.95F_0.05)₃ ↔ Na_2.85MnCr(PO_3.95F_0.05)₃ ↔ Na_1.85MnCr(PO_3.95F_0.05)₃ and Na₄MnCr(PO₄)₃ ↔ Na₃MnCr(PO₄)₃ ↔ Na₂MnCr(PO₄)₃, respectively, which will be analyzed in detail later.

Rate properties are also of great significance for practical applications. As displayed in Fig. 2c, in general, NMCPF showed much enhanced rate performances as compared to NMCP while they delivered analogous initial capacities at 0.1 C: 110.6 mA h g^− 1 for NMCPF and 107.5 mA h g^− 1 for NMCP. Increasing current densities from 0.1 C to 0.2 C, 0.5 C, 1 C, 3 C, 5 C, 10 C, 20 C and 30 C, accordingly enhanced the capacity gap between NMCPF and NMCP. Notably, even under an ultrahigh current rate of 40 C, NMCPF could still deliver 60.4 mA h g^− 1, which outweighs all of previous reports on Na₄MnCr(PO₄)₃^{36–38, 47}. In comparison, NMCP only revealed a “NEAR ZERO” capacity of 15 mA h g^− 1 under the same test condition. When the current rate went back to 0.5 C, a capacity of 89.1 mA h g^− 1 for NMCPF was still recovered, demonstrative of the remarkable reversibility. Moreover, as depicted in Fig. 2d, the corresponding charge-discharge curves proved a restrained electrochemical polarization of NMCPF¹⁸. The above rate performances again determined a boosted kinetics after F doping in NMCP. Figure 2e also compared the performances of NMCPF with other reported cathode materials hinged in Na half-cells. It turns out that NMCPF possessing higher energy densities surpassed the majority of the proposed materials.

To meet the demands of large-scale ESSs, cycling stabilities under various current densities play pivotal roles as well. In Supplementary Fig. 12, NMCPF showed a capacity of 110.8 mA h g^− 1 for the initial cycle at 0.1 C; it could stably operate for 65 cycles. Nevertheless, NMCP possessed a much poorer cyclability and only a low capacity of 81.5 mA h g^− 1 could be maintained after 65 cycles under the identical condition. Figure 2f (at 1 C), Supplementary Fig. 13 (at 0.5 C) and Supplementary Fig. 14 (at 5 C) showed apparently improved capacity retentions and cycling stabilities for NMCPF in comparison to NMCP. As displayed in Supplementary Fig. 15, we also subjected the batteries to a higher rate of 10 C to investigate their long-term cyclabilities. Markedly, NMCPF could steadily cycle 500 times at 10 C (Supplementary Fig. 15) and work for 1000 cycles at 20 C (Fig. 2g) with boosted Coulombic efficiency. In contrast, capacity retentions of merely 56.7% at 10 C and 51.3% at 20 C were achieved for NMCP.

In addition, the batteries were tested within a wider voltage window of 1.5–4.5 V so as to further determine the merits of our work. In Supplementary Fig. 16a-b, both the CV profiles and charge-discharge curves exhibit three pairs of redox couples and three apparent voltage platforms. Apart from Mn²⁺/Mn³⁺ and Mn³⁺/Mn⁴⁺, a high-potential Cr³⁺/Cr⁴⁺ (4.4 V) was also activated in these two samples thereby 143 mA h g^− 1 for NMCPF and 133.6 mA h g^− 1 for NMCP were obtained, which coincides with previous works^{36, 38, 47}. Furthermore, rate performances (Supplementary Fig. 16c) and the cycling stability (Supplementary Fig. 16d) of NMCPF were better than those of NMCP. Further improvements reply on the development of novel high-voltage electrolyte to endure such a harsh cut-off voltage.

Therefore, the above results confirmed the effectiveness of the F doping strategy on NMCPF. Two typical approaches were carried out to further evaluate the Na⁺ diffusion coefficient (D_Na+) of NMCPF and NMCP. The first one is hinged on galvanostatic intermittent titration technique (GITT) method (Supplementary Fig. 17a) after five activation cycles. As seen in Supplementary Fig. 17b, overall, NMCPF displayed higher D_Na+ (10^− 13~10^− 8 cm² s^− 1) than NMCP (10^− 14~10^− 8 cm² s^− 1), which indicates the enhanced Na⁺ kinetics after F doping in agreement with Fig. 1b and Supplementary Fig. 4. Also, Supplementary Fig. 17c showed suppressed overpotentials for NMCPF. Likewise, the batteries were subjected to various scan rates from 0.1-1.0 mV s^− 1 under CV tests (Supplementary Fig. 19). The linear fitting profiles of Supplementary Fig. 19b and Supplementary Fig. 19d reveal diffusion-controlled processes during Na⁺ extraction/intercalation for both NMCPF and NMCP⁴⁸. Accordingly, NMCPF showed higher D_Na+ values than NMCP in Supplementary Fig. 17d, which were computed based on the Randles-Sevcik equation. This enhancement could be on account of the construction of Na vacancies (Fig. 1b) due to F substitution, which truly facilitated Na⁺ diffusion⁴².

Structural Evolution and Charge Compensation Analysis. Aiming to unravel the structural evolution of NMCPF and NMCP upon de-/sodiation processes, in-situ XRD tests based on a specially designed cell were performed within 1.5–4.5 V (vs. Na⁺/Na) for one cycle (left sides of Fig. 3a, c). Supplementary Fig. 20 depicts the full-scope patterns. Figure 3a and 3c clearly showed the reflections of (104), (2_13), (20_4), (3_11), (2_16) and (3_14) for both NMCPF and NMCP and all of them were highly reversible in general, which can be ascribed to their robust NASICON structures and explains the good reversibility of electrochemical properties as discussed above. All peaks could be indexed to the R_3c group with a typical NASICON rhombohedral phase for both fresh NMCPF and NMCP. At first upon charging to ~ 3.65 V, all reflections underwent a right-shift, which corresponded to a solid-solution reaction. It should be noted that (20_4) reflection was split into two peaks from the end of ~ 3.65 V; both (3_11) and (3_14) vanished during the first voltage platform stage, which confirmed that a second phase appeared with coexistence of the initial phase. At the end of the first voltage plateau, all vanished reflections appeared again. When charging to high voltage (3.72 V ~ 4.5 V), all the peaks gradually shifted to higher 2θ values, corresponding to solid-solution reactions. During discharging, all reflections underwent completely obsequent processes and returned to the initial positions. Therefore, the above results revealed a combination of solid-solution reactions and two-phase reactions during extraction/intercalation of Na⁺ for both NMCPF and NMCP; while solid-solution reactions were in the majority, which ensures the good reversibility and fast kinetics for Na⁺ storge. The residual Na⁺ in the structure at deep charged states are deemed to stabilize the overall framework as the binding pillars. The above results confirmed that F doping has no impact on Na⁺ storage mechanisms for NMCP.

In addition, with the purpose of obtaining a clearer understanding regarding their structural transitions, we collected the lattice parameters at all stages, as presented in Fig. 3a and c. As displayed in Fig. 3b and 3d, two samples underwent analogous transitions for all lattices parameters: to illustrate, at the beginning upon charging, a (= b)-axis slightly decreased but the c-axis kept invariant thus the decreased v (volume) could be obtained, corresponding to the structural shrinkage; when further charging, the decreasing rate was facilitated for a (= b)-axis as well as v although the c-axis was only slightly increased, which is in good agreement with solid-solution reactions; upon discharging, all parameters experienced the reverse processes and finally returned to their initial states, supportive of good reversibility for both NMCPF and NMCP. As a consequence, the volume change of NMCPF (8.9%) was much lower than that of NMCP (10.6%), which strongly confirmed that the F doping strategy is conducive to improve structural stability. This may be due to the high electronegativity of F and the increased F-metal interaction, which will be discussed in detail afterwards. The extent of peak deviation for NMCPF (Fig. 3a) was also smaller than that of NMCP (Fig. 3c), which coincides well with the results in Fig. 3b and 3d. The ex-situ TEM and SAED results of fully charged and discharged NMCPF cathodes are shown in Fig. 3e-i. Some obvious lattice fringes and diffraction spots could be clearly identified, demonstrating that the well-crystallized structure was maintained when even subjected to a deep charge state. Such an enhanced structural stability guaranteed the boosted cyclability of NCMPF after fluorine doping as displayed in Fig. 2.

Additionally, ex-situ XPS was carried out to further determine the charge compensation mechanisms of NMCPF and NMCP (Supplementary Fig. 21). The transformations of Na_3.85MnCr(PO_3.95F_0.05)₃ → Na_2.85MnCr(PO_3.95F_0.05)₃ → Na_1.85MnCr(PO_3.95F_0.05)₃ → Na_0.85MnCr(PO_3.95F_0.05)₃ or Na₄MnCr(PO₄)₃ → Na₃MnCr(PO₄)₃ → Na₂MnCr(PO₄)₃ → Na₁MnCr(PO₄)₃ are anticipated to be accompanied with the changes of valence states of transition metals (i.e., Mn and Cr) during charging³⁸. The initial binding energies of Mn 2p_3/2 (641.68 eV) and Cr 2p_3/2 (577.73 eV) of NMCPF and NMCP are the same as that of the corresponding powders, indicating Mn²⁺ and Cr³⁺ in the fresh electrodes^{45, 46}. When charged to 3.67 V and 4.26 V, the peaks of Mn 2p_3/2 shifted to higher binding energies of 642 eV and 642.2 eV, respectively, indicative of oxidations of manganese from pristine Mn²⁺ to Mn³⁺ and further to Mn^{4+ 45}. While the binding energies of Cr 2p_3/2 were kept constant (577.73 eV) in this process, which demonstrated that this cut-off voltage is not high enough to activate the redox behavior of Cr^{3+ 37, 46}. Whereafter, when the battery was subjected to 4.5 V, there was a left-shift in the Cr 2p_3/2 peak (578.3 eV), a signal of Cr^{4+ 49}. But Mn 2p_3/2 remained at the same position as that for Mn⁴⁺. Therefore, the first two platforms were mainly attributed to the successive oxidations of Mn²⁺ to Mn³⁺ and Mn⁴⁺ while the last high-voltage plateaus coincided with the oxidation of Cr³⁺ to Cr⁴⁺. The calculated magnetic moments (µ_B) on the Mn and Cr sites for both NMCPF and NMCP with different Na contents were collected in Supplementary Table 7 and Supplementary Table 8, respectively. The changes of µ_B were in good agreement with the ex-situ XPS results. Besides, it should be noted that the whole redox reaction processes were identical for both NMCPF and NMCP (Supplementary Fig. 13). Thus, one can conclude that appropriate F modification doesn’t alter the charge compensation mechanism of NMCP.

DFT calculations of the Na ⁺ storage mechanism. To obtain in-depth insights on the physicochemical properties of NMCPF and NMCP from a theoretical perspective, we further employed density functional theory (DFT) calculations. The bond valence (BV) method has been widely adopted to identify the ionic diffusion pathways within structures as a typical empirical approach. From the BV maps in Fig. 4a, it can be clearly found that both NMCPF and NMCP possess 3D well-interconnected pathways for Na⁺ with the lowest energy regions except for some subtle differences, which is the feature of typical NASICONs^{23, 26}. Convex-hull phase diagrams of NMCPF (Fig. 4b) and NMCP (Fig. 4c) were constructed hinged on the formation energy at different de-/sodiation states, which accompanied the structural transformations when cycling. All the metastable phases from Na₄MnCr(PO₄)₃ (or Na_3.85MnCr(PO_3.95F_0.05)₃) to Na₀MnCr(PO₄)₃ (or Na₀MnCr(PO_3.95F_0.05)₃) could be electrochemically achieved on account of the negative values of formation energies, which may be the intrinsic origin for multi-electron redox reactions as displayed in Fig. 2 and Supplementary Fig. 16. Of note, Na₂MnCr(PO₄)₃ or Na_1.85MnCr(PO_3.95F_0.05)₃ emerged as the most stable phase due to the lowest formation energy, which well explained the relatively good performance between 1.5–4.3 V.

The voltage profiles of NMCPF (Fig. 4d) and NMCP (Fig. 4e) were also calculated with regard to Mn²⁺/Mn³⁺, Mn³⁺/Mn⁴⁺ and Cr³⁺/Cr⁴⁺ redox couples, both of which fitted well with the experimental (GITT) results (with slight deviations). In general, the calculated voltage platforms of NMCPF were relatively higher than that of NMCP, which may be ascribed to the high electronegativity of F (similar to Na₃V₂(PO₄)₃ and Na₃V₂(PO₄)₂F₃ cases).⁵⁰ An ultrahigh voltage plateaus at ~ 5 V is anticipated to be activated from the calculation if an optimized electrolyte that endures such a high voltage could be developed, which would greatly increase the overall energy density. In addition, the density of states (DOS) diagrams (Fig. 4f and Fig. 4g) were obtained, from which we could find the hybridized Na 3s, Mn 3d, Cr 3d, P 2p, and O 2p orbitals in both NMCPF and NMCP; one more F 1s orbital existed in NMCPF. It turned out that the forbidden band gap was greatly decreased from 1.52 eV in NMCP to 0.22 eV in NMCPF, which contributed to the enhanced electronic conductivity after fluorine modification. The rest DOS patterns with different Na contents are displayed in Supplementary Fig. 14. The electron density difference in NMCPF (Fig. 4h) and NMCP (Fig. 4i) further supported the modification of electron density through F doping. These results accounted for better electrochemical properties of NMCPF from a kinetics perspective.

To further reveal the origin of the capacity decay of the pristine NMCP, we dissembled the cycled batteries in the glove box to collect the electrolyte and the separator and measure the separate manganese concentration through ICP-OES method. Like other Mn-based materials, NMCP also suffers from severe Mn dissolution with the concentration of 0.24 mg L^− 1 in the electrolyte and 8 ppm in the separator after cycling, which could account for the inferior electrochemical performances in Fig. 2⁵¹. In sharp contrast, the manganese dissolution issue in NMCPF was effectively suppressed with a low concentration of 0.05 mg L^− 1 in the electrolyte and 1 ppm in the separator after cycling. In addition, we employed a theoretical approach to explain this phenomenon. The Crystal Orbital Hamilton Population (COHP) approach has been acknowledged as a reliable tool to visualize the chemical bonding in the battery field in recent years^52–54. The Kohn-Sham states were initially projected into atomic orbitals and subsequently the mutual orbital overlap population was inspected. The COHP and ICOHP (integrated COHP) results are shown in Fig. 5b-f, supplementary Fig. 23, supplementary Fig. 24 and supplementary Table 9–10 based on the adjacent six O/F atoms around a specific Mn atom. As reported, the negative value of COHP of a Mn-O/F pair represents constructive (namely, bonding) interference of atomic orbitals; the modulus of COHP corresponds to the degree of covalency between Mn and O/F and the zero point of COHP suggests non-interacting bonds or a pure ionic bond. In each pattern of Fig. 5c-f, supplementary Fig. 23, and supplementary Fig. 24, the intersection between ICOHP curve and Fermi level is the actual -ICOHP value. From Fig. 5b, NMCPF possesses a higher accumulated -ICOHP value (all six Mn-O/F pairs including spin-up and spin-down direction) of 4.91 eV than NMCP (4.38 eV). This result indicate that F doping strengthens the overall chemical bonding of Mn-O/F bonds in the local scope, which may be associated with the high electronegativity of F. The strengthened chemical bonding between adjacent O/F and Mn helps explain the suppressed manganese dissolution.

In summary, we have proposed an effective modification strategy of well-controlled doping light-weight F into the Na₄MnCr(PO₄)₃ lattice. An enhanced cycling stability (1000 cycles at 20 C) and an excellent rate performance up to 40 C were thus obtained. The good electrochemical performances could be attributed to the reduced energy gap (the boosted electronic conductivity) and strengthened local chemical bonding around Mn as confirmed by DFT calculations. The ICP-OES results revealed suppressed Mn dissolution after the modification of F. The structural transformations (mainly for the solid solution reaction and partially for the two-phase reaction) and charge compensation mechanisms remained unchanged as determined by in-situ XRD, TEM and XPS. This research proves the prospect of light-weight F doping for high-voltage polyanionic cathodes for SIBs technology.

Materials synthesis. A typical sol-gel approach was performed to synthesize Na_4-3xMnCr(PO_4-xF_x)₃, where x = 0, 0.01, 0.02, 0.05, 0.1, 0.15 and 0.2. All chemicals were directly used without further purification. Anhydrous citric acid (99.5%, Alfa Aesar), (CH₃COO)₂Mn (99%, Acros Organics), Cr(NO₃)₃·9H₂O (99%, Aladdin), CH₃COONa (99%, Alfa Aesar) and NH₄H₂PO₄ (99%, Sigma-Aldrich) with the designed ratios were successively dissolved into an aqueous solution (60 mL ultrapure water). Afterwards, the solution was heated at 80°C accompanied by constant magnetic stirring for several hours and then was kept in 120°C overnight. After that, the solid precursor was grinded and further subjected to 650°C for 9 hours under argon atmosphere.

Physicochemical characterizations. The XRD patterns were collected on a PANalytical Empyrean device with Cu K_α radiation and the lattice parameters were refined with a GSAS software by Rietveld method. VESTA software was adopted to depict the structure schematic illustration. A LabRAM HR Evolution instrument (laser wavelength: 532 nm) was carried out to collect the Raman spectra. XPS spectra were obtained on a ThermoFisher ESCALAB 250Xi facility (Al Kα radiation) and were calibrated based on C 1s (284.8 eV). The binding energies of peaks were referred to the National Institute of Standards and Technology (NIST, USA) database. TGA was performed on a STA 2500 Regulus device (heating rate: 10°C min^− 1) under ambient atmosphere. The morphologies of the samples were observed on Titan G260-300. ICP tests were conducted on Spectro Blue SOP. The in-situ XRD patterns were obtained by a well-designed cell (Beijing SciStar technology Co., Ltd) with an X-ray penetrable Be window that is also as the current collector. The cycled cathodes were disassembled in glove box and were thoroughly rinsed by dimethyl carbonate solvent before ex-situ measurements.

Electrochemical characterizations. The electrode consists of NMCPF/NMCP, acetylene black and polyvinylidene fluoride with a typical ratio of 7:2:1 (N-methyl-2-pyrrolidone as the solvent). The slurry was then uniformly pasted onto the surface of aluminum foil by a doctor blade. Afterwards, the wet electrode was further dried at 80°C for 12 hours within a vacuum oven. The batteries (CR2032 coin cell) were assembled in a pure-argon glove box with a low O₂ and H₂O concentration (< 0.1 ppm). The glass fiber (Whatman GF/D) served as the separator. Na metallic disks were counter electrodes. 1.0 M NaClO₄ in propylene carbonate (PC) –ethylene carbonate (EC) (1:1 vol.%) solvent with 5 vol. % fluoroethylene carbonate (FEC) was the electrolyte. Neware battery test system (Shenzhen, China) was adopted to conduct the electrochemical performances and GITT measurements. Five formation cycles were conducted prior to GITT test. CV and EIS test were performed on a Biologic VMP-3 electrochemical workstation.

Computational methods. All calculations were conducted via the projector augmented wave method in the framework of the density functional theory (DFT)⁵⁵, as implemented in the QUANTUM ESPRESSO⁵⁶. The generalized gradient approximation (GGA) and Perdew–Burke–Ernzerhof (PBE) exchange functional⁵⁵ were used. The plane-wave energy cutoff was set to 38 Ry, and the Monkhorst–Pack method⁵⁷ was employed for the Brillouin zone sampling. The convergence criteria of energy and force calculations were set to 10^− 5 Ry/atom and 0.01 Ry/Å, respectively. To consider the strong correlation effects of transition metal in Na_4-xMnCr(PO₄)₃ and Na_3.85-yMnCr(PO_3.95F_0.05)₃, both structural optimizations and electronic structure calculations were carried out by using the spin-dependent GGA plus Hubbard correction U (GGA + U) method⁵⁸, and the effective U_eff parameters are 3.9 and 3.7 eV for Mn-3d and Cr-3d states⁵⁹, respectively. Na₄MnCr(PO₄)₃ is built from Na₄Mn(PO₄)₃ with equal molar of Mn-Cr mixing, that is Mn-Cr cation arrangements in the Mn sites were created within the conventional cell of Na₄Mn(PO₄)₃ using the enumeration method⁶⁰ implemented in the Pymatgen code⁶¹. The Mn-Cr cation arrangement of the most stable Na₄MnCr(PO₄)₃ structure was determined by ranking the Mn-Cr cation arrangements by their DFT energies. Furthermore, determination of the stable structures of the sodium extracted compounds with Na-vacancy mixing, including Na₃MnCr(PO₄)₃, Na₂MnCr(PO₄)₃ and NaMnCr(PO₄)₃, was employed through this same method. Similarly, the most stable F-doped Na_3.85MnCr(PO_3.95F_0.05)₃ structure was constructed by enumerating O anion sites in Na₄MnCr(PO₄)₃ structure and ranking their DFT energies. The bond-valence sum (BVS) mappings were performed by the PyAbstantia code. The crystal orbital Hamilton population (COHP) between neighboring oxygen ions was computed by the Lobster program⁶², in which the negative and positive COHP values indicate bonding and anti-bonding, respectively. The pbeVaspFit2015 basis sets were used in the reconstruction of the PAW wave functions of each element.

Data availability

The data supporting the findings of this study are available from the authors upon reasonable request.

Acknowledgements

W.Z. thanks the funding support from China Scholarship Council/ University College London for the joint Ph.D. scholarship. We thank the Natural Science Foundation of Hunan Province, China (2020JJ1007).

Author contributions

W.Z. prepared the manuscript. W.Z. and Y.W. synthesized the materials and carried out all the electrochemical measurements. Z.X. performed the DFT calculations. J.L. conducted the XRD, XPS and NMR studies. Y.D., W.Z., and R.C. performed the ex-situ XPS, TEM and Raman tests. S.L. conducted the in-situ XRD and the ICP work. Z. Z., G.H., Y.L. and I.P.P supervised the project and co-wrote the manuscript. All authors discussed the results and contributed to writing the manuscript.

Competing interests

The authors declare no competing interests.

Additional information

Supplementary Information The online version contains supplementary material available at

Correspondence and requests for materials should be addressed to Guanjie He.

Tian Y. et al. Promises and Challenges of Next – Generation "Beyond Li – ion" Batteries for Electric Vehicles and Grid Decarbonization. Chem. Rev. 121, 1623–1669 (2021).
Liu Y. et al. Rechargeable aqueous Zn-based energy storage devices. Joule 5, 2845–2903 (2021).
Zong W. et al. Topochemistry-Driven Synthesis of Transition – Metal Selenides with Weakened Van Der Waals Force to Enable 3D – Printed Na – Ion Hybrid Capacitors. Adv. Funct. Mater. 32, 2110016 (2021).
Zheng J. et al. Highly dispersed MoP encapsulated in P – doped porous carbon boosts polysulfide redox kinetics of lithium – sulfur batteries. Mater. Today Energy 18, 100531 (2020).
Dunn B., Kamath H. & Tarascon J. –M. Electrical Energy Storage for the Grid: A Battery of Choices. Science 334, 928–935 (2011).
Dai Y. et al. Nanoribbons and nanoscrolls intertwined three – dimensional vanadium oxide hydrogels for high – rate lithium storage at high mass loading level. Nano Energy 40, 73–81 (2017).
Li J. et al. Air – Stable Li₆CoO₄@Li₅FeO₄ Pre – Lithiation Reagent in Cathode Enabling High Performance Lithium-Ion Batteries. J. Electrochem. Soc. 168, 080510 (2021).
Li S. et al. Constructing stable surface structures enabling fast charging for Li – rich layered oxide cathodes. Chem. Eng. J. 427, 132036 (2022).
Tarascon J. –M. Na – ion versus Li – ion Batteries: Complementarity Rather than Competitiveness. Joule 4, 1616 – 1620 (2020).
Wu F., Maier J. & Yu Y. Guidelines and trends for next – generation rechargeable lithium and lithium – ion batteries. Chem. Soc. Rev. 49, 1569–1614 (2020).
Usiskin R. et al. Fundamentals, status and promise of sodium – based batteries. Nat. Rev. Mater. 6, 1020–1035 (2021).
Wu F. et al. Multi – electron reaction materials for sodium – based batteries. Mater. Today 21, 960–973 (2018).
Li H. et al. Manganese – Based Materials for Rechargeable Batteries beyond Lithium – Ion. Adv. Energy Mater. 11, 2100867 (2021).
Chu S., Guo S. & Zhou H. Advanced cobalt – free cathode materials for sodium – ion batteries. Chem. Soc. Rev. 50, 13189–13235 (2021).
Zhao C. et al. Rational design of layered oxide materials for sodium – ion batteries. Science 370, 708–711 (2020).
Huang Y. et al. Enabling Anionic Redox Stability of P2 – Na_5/6Li_1/4Mn_3/4O₂ by Mg Substitution. Adv. Mater. 34, 2105404 (2022).
Gao H., Li Y., Park K. & Goodenough J. B. Sodium Extraction from NASICON-Structured Na₃MnTi(PO₄)₃ through Mn(III)/Mn(II) and Mn(IV)/Mn(III) Redox Couples. Chem. Mater. 28, 6553–6559 (2016).
Chen M. et al. NASICON – type air – stable and all – climate cathode for sodium – ion batteries with low cost and high – power density. Nat. Commun. 10, 1480 (2019).
Li H. et al. Organic/inorganic anions coupling enabled reversible high-valent redox in vanadium – based polyanionic compound. Energy Storage Mater. 47, 526–533 (2022).
Li H., Xu M., Zhang Z., Lai Y. & Ma J. Engineering of Polyanion Type Cathode Materials for Sodium – Ion Batteries: Toward Higher Energy/Power Density. Adv. Funct. Mater. 28, 2000473 (2020).
Wang T. et al. N – Doped Carbon Nanotubes Decorated Na₃V₂(PO₄)₂F₃ as a Durable Ultrahigh – rate Cathode for Sodium Ion Batteries. ACS Appl. Energy Mater. 3, 3845–3853 (2020).
Wang S. et al. Ultra – High – Rate Na₃V(PO₃)₃N Cathode with Superior Stability for Fast – Charging Sodium – Ion Batteries. ACS Appl. Energy Mater. 4, 10136–10144 (2021).
Zhang W. et al. Rationally Designed Sodium Chromium Vanadium Phosphate Cathodes with Multi – Electron Reaction for Fast – Charging Sodium – Ion Batteries. Adv. Energy Mater., 2201065 (2022).
Li H. et al. Highly efficient, fast and reversible multi – electron reaction of Na₃MnTi(PO₄)₃ cathode for sodium-ion batteries. Energy Storage Mater. 26, 325–333 (2020).
Li H. et al. Pseudocapacitance enhanced by N – defects in Na₃MnTi(PO₄)₃/N – doped carbon composite for symmetric full sodium – ion batteries. Mater. Today Energy 21, 100754 (2021).
Zhang W. et al. All – climate and air – stable NASICON-Na₂TiV(PO₄)₃ cathode with three – electron reaction toward high – performance sodium – ion batteries. Chem. Eng. J. 433, 133542 (2022).
Xiong H. et al. Polymer Stabilized Droplet Templating towards Tunable Hierarchical Porosity in Single Crystalline Na₃V₂(PO₄)₃ for Enhanced Sodium-Ion Storage. Angew. Chem. Int. Ed 60, 10334–10341 (2021).
Zhou W. et al. Na_xMV(PO₄)₃ (M = Mn, Fe, Ni) Structure and Properties for Sodium Extraction. Nano Lett. 16, 7836–7841 (2016).
Zhang W. et al. Engineering 3D Well-Interconnected Na₄MnV(PO₄)₃ Facilitates Ultrafast and Ultrastable Sodium Storage. ACS Appl. Mater. Interfaces 11, 35746–35754 (2019).
Chen F. et al. A NASICON-Type Positive Electrode for Na Batteries with High Energy Density: Na₄MnV(PO₄)₃. Small Methods 3, 1800218 (2018).
Xu C. et al. Mn-Rich Phosphate Cathodes for Na – Ion Batteries with Superior Rate Performance. ACS Energy Lett. 7, 97–107 (2021).
Zakharkin M. V. et al. Enhancing Na⁺ Extraction Limit through High Voltage Activation of the NASICON-Type Na₄MnV(PO₄)₃ Cathode. ACS Appl. Energy Mater. 1, 5842–5846 (2018).
Gao X. et al. Phase Transformation, Charge Transfer, and Ionic Diffusion of Na₄MnV(PO₄)₃ in Sodium – Ion Batteries: A Combined First – principles and Experimental Study. J. Mater. Chem. A 8, 17477–17486 (2020).
Xu C. et al. A Novel NASICON-Typed Na₄VMn_0.5Fe_0.5(PO₄)₃ Cathode for High – Performance Na – Ion Batteries. Adv. Energy Mater. 11, 2100729 (2021).
Buryak N. S. et al. High-voltage structural evolution and its kinetic consequences for the Na₄MnV(PO₄)₃ sodium – ion battery cathode material. J. Power Sources 518, 230769 (2022).
Wang J. et al. A High – Energy NASICON – Type Cathode Material for Na – Ion Batteries. Adv. Energy Mater. 10, 1903968 (2020).
Zhang W. et al. Full Activation of Mn⁴⁺/Mn³⁺ Redox in Na₄MnCr(PO₄)₃ as a High – Voltage and High – Rate Cathode Material for Sodium – Ion Batteries. Small 16, 2001524 (2020).
Zhang J. et al. A Novel NASICON – Type Na₄MnCr(PO₄)₃ Demonstrating the Energy Density Record of Phosphate Cathodes for Sodium – Ion Batteries. Adv. Mater. 32, 1906348 (2020).
Kim H. et al. Anomalous Jahn–Teller behavior in a manganese – based mixed – phosphate cathode for sodium ion batteries. Energy Environ. Sci. 8, 3325–3335 (2015).
Li H. et al. Robust Artificial Interphases Constructed by a Versatile Protein – Based Binder for High – Voltage Na – Ion Battery Cathodes. Adv. Mater., 2202624 (2022).
Xu C. et al. Reversible Activation of V⁴⁺/V⁵⁺ Redox Couples in NASICON Phosphate Cathodes. Adv. Energy Mater., 2200966 (2022).
Hou J. et al. Unlocking fast and reversible sodium intercalation in NASICON Na₄MnV(PO₄)₃ by fluorine substitution. Energy Storage Mater. 42, 307–316 (2021).
Hadouchi M. et al. Fast sodium intercalation in Na_3.41£_0.59FeV(PO₄)₃: A novel sodium – deficient NASICON cathode for sodium – ion batteries. Energy Storage Mater. 35, 192–202 (2021).
Liu J. et al. NASICON – structured Na₃MnTi(PO₄)_2.83F_0.5 cathode with high energy density and rate performance for sodium – ion batteries. Chem. Eng. J. 435, 134839 (2022).
Tan B. J., Klabunde K. J. & Sherwood P. M. A. XPS Studies of Solvated Metal Atom Dispersed Catalysts. Evidence for Layered Cobalt – Manganese Particles on Alumina and Silica. J. Am. Chem. Soc. 113, 855–861 (1991).
T.P. Moffat R. M. L., R.R. Ruf. An X – ray photoelectron spectroscopy study of chromium – metalloid alloys—III. Electrochim. Acta 40, 1723–1734 (1995).
Zhao Y., Gao X., Gao H., Dolocan A. & Goodenough J. B. Elevating Energy Density for Sodium – Ion Batteries through Multielectron Reactions. Nano Lett. 21, 2281–2287 (2021).
Gu Z. Y. et al. An Advanced High – Entropy Fluorophosphate Cathode for Sodium – Ion Batteries with Increased Working Voltage and Energy Density. Adv. Mater. 34, 2110108 (2022).
Allen G. C., Curtis M. T., Hooper A. J. & Tucker P. M. X – Ray photoelectron spectroscopy of chromium–oxygen systems. J. Chem. Soc., Dalton Trans., 1675 – 1683 (1973).
Sharma L., Adiga S. P., Alshareef H. N. & Barpanda P. Fluorophosphates: Next Generation Cathode Materials for Rechargeable Batteries. Adv. Energy Mater. 10, 2001449 (2020).
Asl H. Y. & Manthiram A. Reining in dissolved transition-metal ions. Science 369, 140 – 141 (2020).
Dronskowski R. & Bloechl P. E. Crystal orbital Hamilton populations (COHP): energy-resolved visualization of chemical bonding in solids based on density – functional calculations. J. Phys. Chem. 97, 8617–8624 (1993).
You Y. et al. Insights into the Improved High-Voltage Performance of Li – Incorporated Layered Oxide Cathodes for Sodium – Ion Batteries. Chem 4, 2124 – 2139 (2018).
Wang C. et al. Tuning local chemistry of P2 layered – oxide cathode for high energy and long cycles of sodium-ion battery. Nat. Commun. 12, 2256 (2021).
Kohn W. & Sham L. J. Self – Consistent Equations Including Exchange and Correlation Effects. Phys. Rev. 140, A1133 – A1138 (1965).
Giannozzi P. et al. QUANTUM ESPRESSO: a modular and open-source software project for quantum simulations of materials. J. Phys. Condens. Matter. 21, 395502 (2009).
Monkhorst H. J. & Pack J. D. Special points for Brillouin – zone integrations. Phys. Rev. B 13, 5188–5192 (1976).
Dudarev S. L., Botton G. A., Savrasov S. Y., Humphreys C. J. & Sutton A. P. Electron – energy – loss spectra and the structural stability of nickel oxide: An LSDA + U study. Phys. Rev. B 57, 1505–1509 (1998).
Xiao R., Li H. & Chen L. Density Functional Investigation on Li₂MnO₃. Chem. Mater. 24, 4242–4251 (2012).
Hart G. L. W. & Forcade R. W. Algorithm for generating derivative structures. Phy. Rev. B 77, 224115 (2008).
Ong S. P. et al. Python Materials Genomics (pymatgen): A robust, open – source python library for materials analysis. Comput. Mater. Sci. 68, 314–319 (2013).
Maintz S., Deringer V. L., Tchougreeff A. L. & Dronskowski R. LOBSTER: A tool to extract chemical bonding from plane – wave based DFT. J. Comput. Chem. 37, 1030–1035 (2016).

There is NO Competing Interest.

SupplementaryInformation.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Fluorine anion modification of high-voltage Mn-rich phosphate cathodes for high-power and long-lasting sodium-ion batteries

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Methods

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1