Elevating Operation Voltage and Suppressing Phase Transition for Honeycomb‐Layered Cathodes by a Dual‐Honeycomb Structure Strategy

Honeycomb‐layered oxides are a class of cathode materials for sodium‐ion batteries with great potential due to their high voltage and high capacity. However, the structural instability and voltage fading during cycling limit their practical application. Herein, it is revealed that Te substitution into Na3Ni2SbO6 induces a new dual‐honeycomb structure, which can elevate the average discharge voltage of the cathode materials from 3.2 to 3.8 V with improved cycle stability and alleviated voltage decay. Synchrotron operando X‐ray diffraction demonstrates that Te substitution can suppress the O3−P3−O1‐phase transition during charge and discharge processes effectively, benefited from the strong TeO covalent bonds. The resulted Na2.2Ni2Sb0.2Te0.8O6 cathode exhibits a high capacity retention of 70.9% after 1000 cycles at 1C, with an elevated operating voltage of ≈3.8 V. Theoretical calculations reveal that the introduced TeO bonds break the symmetric distribution of charge in Ni/Sb honeycomb structure and elevate the operation voltage by increased valence band width. Proper Te substitution can promote the rate and cycle capability of the cathode by suppressing phase transition and decreasing the bandgap.

Honeycomb-layered oxides are a class of cathode materials for sodium-ion batteries with great potential due to their high voltage and high capacity.However, the structural instability and voltage fading during cycling limit their practical application.Herein, it is revealed that Te substitution into Na 3 Ni 2 SbO 6 induces a new dual-honeycomb structure, which can elevate the average discharge voltage of the cathode materials from 3.2 to 3.8 V with improved cycle stability and alleviated voltage decay.Synchrotron operando X-ray diffraction demonstrates that Te substitution can suppress the O3ÀP3ÀO1-phase transition during charge and discharge processes effectively, benefited from the strong Te─O covalent bonds.The resulted Na 2.2 Ni 2 Sb 0.2 Te 0.8 O 6 cathode exhibits a high capacity retention of 70.9% after 1000 cycles at 1C, with an elevated operating voltage of ≈3.8 V. Theoretical calculations reveal that the introduced Te─O bonds break the symmetric distribution of charge in Ni/Sb honeycomb structure and elevate the operation voltage by increased valence band width.Proper Te substitution can promote the rate and cycle capability of the cathode by suppressing phase transition and decreasing the bandgap.a maximum of two electrons while most other cations can only provide one electron during Na-ion extraction.In addition, the high redox potential of Ni leads Ni-based honeycomb-layered oxides to promising candidates for high energy density SIBs.In this system, O3-Na 3 Ni 2 SbO 6 and P2-Na 2 Ni 2 TeO 6 are two representative cathodes material and arouse great interest among researchers due to their good electrochemical performance.Yuan et al. first reported the electrochemical behaviors of O3-Na 3 Ni 2 SbO 6 with a decent capacity of 117 mAh g À1 and revealed the three phase transformations of this layered compound (O3ÀP3ÀO1). [25]It is found that the O3ÀP3-phase transition occurs at around 3.35 V and P3ÀO1-phase transition is at around 3.5 V.As we know, complicated phase transformations might result in unsatisfactory battery performance with limited cycle life.[28] The combination of these chemical and structural degradations results in rapid capacity decay. [29,30]To inhibit irreversible phase transition, most researchers usually limit the cutoff voltage to 4 V. [14][15][16]31] In addition, an effective way to improve its performance is cation doping. Thepartial substitution of Ni 2þ by inactive elements such as Mg 2þ and Zn 2þ can significantly suppress the electrochemically unfavorable P3-O1 transition, alleviating the volume variation of the cathode material and contributing to improved cyclability.[26,32] However, the phase transition still exists and the substitution of electrochemically active Ni 2þ element by inert elements inevitably results in a sacrifice of capacity.[33] In addition, the lower electronegativity of these elements might lead to a decreased operation voltage according to the inductive effect, [7,8] which results in low energy density.This inspires us to introduce high valence Te 6þ with high electronegativity to substitute the Sb 5þ site rather than the Ni 2þ site in Na 3 Ni 2 SbO 6 .
Te 6þ ion is an electrochemically inert cation with smaller ionic radius (0.56 Å) than Sb 5þ (0.60 Å).Similar to O3-Na 3 Ni 2 SbO 6 , P2-Na 2 Ni 2 TeO 6 presents the honeycomb structure with Ni 2þ / Te 6þ ordering in the TM layer and its electrochemical performance was first reported by Goodenough's group. [6,9,34]This honeycomb cathode material shows high ionic conductivity and is considered to be one of the highest voltage cathodes for SIBs.The operation voltage of Na 2 Ni 2 TeO 6 is around 3.6 V, which is much higher than other layered cathode materials, such as Na 3 Ni 2 SbO 6 (3.3 V).Compared to traditional layered cathode materials, the high electronegativity elements of Te and Sb in honeycomb-layered cathode materials increase the covalent character of Sb─O and Te─O bond within the layers.Consequently, the energy for Ni oxidation is higher than that in traditional cathode materials, leading to a staggering increase in the voltage of the honeycomb-layered oxides.This process is referred to as "inductive effect". [7,8]The high electronegativity of Te decreases the electronegativity difference with O, but increases the electronegativity difference with Ni, resulting in enhanced covalent character of Te─O bonds and ionic character of Te─Ni bonds.The stronger Te─O and Te─Ni bonds result in higher working voltages of Na 2 Ni 2 TeO 6 in comparison to Na 3 Ni 2 SbO 6 .However, the rate performance of Na 2 Ni 2 TeO 6 is not comparable to O3-Na 3 Ni 2 SbO 6 .When increasing the rate from 0.05C to 1C (1C = 138.3mA g À1 ), a sharp loss in capacity from ≈110 mAh g À1 (0.03C) to ≈30 mAh g À1 (1C) occurs. [3]The poor rate capability severely hampers the application of this material and the intrinsic reason has not been deeply explored until now.Therefore, it is significant to find a new type of cathode material with elevated operation voltage and explore the reaction mechanism by monitoring the structural evolution occurring during the Na þ extraction/insertion process, contributing to developing high-performance honeycomb cathode materials.
Here, we design a new cathode composition of Na 2.2 Ni 2 Sb 0.2 Te 0.8 O 6 with the Sb site substituted by Te for the first time.By tuning the ratio of the Ni 2þ /Sb 5þ order and the Ni 2þ /Te 6þ order, the battery performance reaches the best balance between the high-power density and long cycling capability.The Na 2.2 Ni 2 Sb 0.2 Te 0.8 O 6 electrode exhibits excellent cyclability for 1000 cycles with ≈70.9% capacity retention at 1C and works well even at 50C in Na cells.Moreover, an electrochemical mechanism study combining operando and ex situ X-ray diffraction (XRD) reveals an effective suppression of the O3-P3-O1-phase transition at 3.5-4.0V after the introduction of Te.In addition, the electronic environment of Ni in different Ni 2þ /Sb 5þ (Te 6þ ) order structures was compared through density-functional theory (DFT) calculation.

Crystal Structure
A series of Na 3Àx Ni 2 Sb 1Àx Te x O 6 (x = 0, 0.2, 0.5, 0.8, and 1.0, denoted as TeÀ0, TeÀ0.2, TeÀ0.5, TeÀ0.8, and TeÀ1.0, respectively) were synthesized by solid-state reaction method.The XRD patterns of the whole series of materials are shown in Figure 1a.The result shows that the TeÀ0 end-member adopts a typical O3 structure indexed in the monoclinic C2/m space group (no.12).It can be seen that the O3 structure is well maintained with increasing Te content except for the Te-rich end-member of the tie-line, Na 2 Ni 2 TeO 6 , exhibiting a P2 structure with a hexagonal P6 3 /mcm space group (No. 193).The XRD patterns all show superstructure peaks due to the honeycomb ordering of Ni 2þ /Sb 5þ or Ni 2þ /Te 6þ within the TM layer.The structure of the five samples were refined using Rietveld method, shown in Figure 1b-d and S1a,b, Supporting Information.The detailed atomic occupancy and lattice parameter results derived from Rietveld refinements are shown in Table S1-S5, Supporting Information.The lattice parameters of all the samples are listed in Table S6, Supporting Information.It can be seen that the parameters of TeÀ0 are a = 5.3022(5) Å, b = 9.1912(5) Å, and c = 5.6474 (9) Å.With the increase of Te content, the lattice parameter c increases, implying the increase of interlayer spacing along c axis, beneficial for Na transportation.This can be attributed to the decrease of Na content designed to keep charge balance with the increase of Te ions.As Na content decreases, there will be more vacancies between oxygen in the neighboring TM layers.Thus, the Coulombic repulsion force between oxygen in the neighboring TMO 6 layers increases.In addition, the higher electronegativity of Te than Sb, which brings stronger Te─O and Te─Ni bonds and leads to the contraction of TM layer and expansion of interlayer distance.The schematic diagram of the structural changes with Te substitution is shown in Figure 1e.Viewing along the c-axis direction, it can be seen that each Sb atom is surrounded by six Ni atoms, presenting the honeycomb structure.The layer spacing gradually increases after Te substitution, which corresponds to the increase of lattice parameter c.The coordination environment of Na changes from octahedral to trigonal coordination when all the Sb atoms are substituted by Te atoms.In general, Te substitution and Na decrease lead to the phase change from O3 phase to P2 phase.
The microstructure of TeÀ0.8 was characterized by transmission electron microscope (TEM), as shown in Figure 2. The images show an irregular granular morphology of TeÀ0.8 with particle size of about 0.5-1 μm (Figure 2a,d).Figure 2c is an enlarged view of the box in Figure 2b.It is notable that a clear honeycomb arrangement of the atoms can be seen in the image, which is labeled by the colored balls (the blue atoms represent Ni and the yellow atoms represent Sb or Te).In the high-resolution TEM image shown in Figure 2e, the lattice fringes display three different spaces (2.66, 4.40, and 4.57 Å), corresponding to (111), (011), and (100) planes of the O3 TeÀ0.8 phase, respectively.The selected-area electron diffraction (SAED) pattern along the [01 1] hex zone axis in Figure 2f shows the corresponding planes of the selected area in Figure 2e.This feature is characteristic of stacking faults in honeycomb-ordered [A 1/3 M 2/3 ]O 2 slabs, indicating that the in-plane structure of Te-substituted sample TeÀ0.8 is honeycomb order. [35]The morphologies of TeÀ0, TeÀ0.8, and TeÀ1.0 were examined by scanning electron microscopy (SEM), as shown in Figure S2-S4, Supporting Information.It can be seen that the three samples comprise bulk particles with a size of 0.5-2.0μm, which is in good agreement with the TEM images.[38] It should be noted that TeÀ0.8 shows the relative smaller particle size and its size distribution is more uniform.The inset of energy-dispersive X-ray spectroscopy (EDX) mapping shows that Na, Ni, Sb, Te, and O are distributed homogeneously in the particles.As shown in Table S7-S9, Supporting Information, the stoichiometry of TeÀ0, TeÀ0.8, and TeÀ1.0 was examined by EDX and the ratio is very close to the designed stoichiometry.

Electrochemical Performance
The electrochemical performance of the five samples was tested in Na half-cells.Among the whole series of materials synthesized, TeÀ0.8 delivers the best electrochemical performance in terms of rate performance as well as cycling stability, as shown in Figure 3.The comparison of galvanostatic chargeÀdischarge (GCD) curves for the first cycle at 0.1C between 2.0 and 4.3 V is shown in Figure 3a.The initial discharge capacities are 91.6,87.7, 80.9, 80.3, and 55.3 mAh g À1 for TeÀ0, TeÀ0.2, TeÀ0.5, TeÀ0.8, and TeÀ1.0, respectively.Although the TeÀ0.8 sample shows a slight capacity decay, the average discharge voltage is elevated from ≈3.2 to ≈3.8 V with the increasing amount of Te substitution.The stronger electronegativity of Te than that of Sb leads to stronger Te─O bonds.This increases the covalent character of the Te─O bond within the layer.Consequently, the energy for Ni oxidation is higher than that in Sb-rich materials, leading to a staggering increase in the voltage of the honeycomblayered oxides, which is called the "inductive effect". [7,8]The high operation voltage makes it possible to increase the energy density of SIBs.
The typical GCD curves of TeÀ0.8 at 1C in the 1st, 50th, 100th, 200th, 300th, and 500th cycles are shown in Figure 3b.The initial charge/discharge capacities are 80.9/70.1 mAh g À1 , with a capacity retention of 87% after 300 cycles and 80% after 500 cycles.It is observed that the shape of the voltage curves can be well maintained even after 500 cycles without voltage decay.In contrast, large voltage polarization and poor cycle stability can be observed in the GCD curves of other samples, as shown in Figure S5, Supporting Information.The curves of TeÀ0, TeÀ0.5, and TeÀ1.0 show severe voltage decay after cycling, and the capacity decays to 37.2, 33.1, and 25.5 mAh g À1 , respectively.
Figure 3c compares the rate performance of TeÀ0, TeÀ0.8, and TeÀ1.0.The excellent rate performance of TeÀ0.8 becomes evident as the current density increases to 50C.For TeÀ0.8, the capacities are 79.9, 66.3, 64.3, 61.3, 57.2, 47.6, and 36.7 at the current densities of 0.1C, 1C, 2C, 5C, 10C, 20C, 50C, respectively.In comparison, TeÀ0 and TeÀ1.0 experience severe capacity drops to 4.3 and 5.8 mAh g À1 at 50C, respectively.It should be noted that the rate performance of TeÀ0.8 is the best compared to other honeycomb-layered cathode materials reported so far. [6,10,11,22,33]The outstanding rate capability can be attributed Electrochemical impedance spectroscopy (EIS) measurements were performed to investigate the effect of Te substitution on sodium-ion diffusion.Figure 3e  Galvanostatic intermittent titration technique (GITT) tests of TeÀ0, TeÀ0.2, TeÀ0.5, TeÀ0.8, and TeÀ1.0 under the thermodynamic equilibrium conditions were employed to further compare the diffusion kinetics in both materials during the first charging process.The cells were tested at a current density of 0.2C within the potential window of 2.0-4.3V for 10 min followed by 1 h rest to reach the quasi-equilibrium potential state, as shown in Figure 3f.The Na þ diffusion coefficient (D Naþ ) therefore can be calculated by Fick's second law under ideal thermodynamic equilibrium conditions.Figure S6a-e, Supporting Information, shows the D Naþ as a function of time in the first charging process.The calculated D Naþ values of around 10 À10 cm 2 s À1 for TeÀ0.8 are apparently higher than those for the other materials (about 10 À13 , 10 À12 , 10 À12 , and 10 À11 cm 2 s À1 for TeÀ0, TeÀ0.2, TeÀ0.5, and TeÀ1.0, respectively).The superiorsodium-ion diffusion capability is in good agreement with the result of electrochemical performance, verifying the effective strategy of TM ordering tuning.
Figure 3g shows cycle stability up to 1000 cycles of the five samples at 1C within the voltage range of 2.0À4.3V.It can be seen that TeÀ0.8 performs superior cycling stability with a capacity retention of 70.9% after 1000 cycles, much higher than the other samples.The energy densities of Te-0, Te-0.8, and Te-1.0 based on the mass of the cathode materials are compared in Figure S7, Supporting Information.The high operation voltage and cycle stability of Te-0.8 makes it keep a high energy density over 200 Wh kg À1 in about 600 cycles, while Te-0 and Te-1.0 suffer severe energy density loss during cycling.The improved performance of TeÀ0.8 is attributed to its high structural stability and electrochemical reversibility, which facilitates its possibility for practical applications in energy storage systems.

Structure Evolution
To investigate the effect of Te doping on the structure evolution of the honeycomb cathode during charge and discharge processes, we performed ex situ XRD measurement on TeÀ0, TeÀ0.5, and TeÀ0.8 in the first cycle (Figure 4).The peaks of all the pristine electrodes can be well indexed with an O3 phase and the peaks of (001), (002), (130), (13 1), and (131) are labeled in the figure.Figure 4a shows the phase transition of TeÀ0 during the charge and discharge processes.The enlarged (001) peak and the corresponding GCD curves are plotted on the right side.It shows that the peak intensity of the O3 phase gradually decreases and the peaks of the P3 and O1 phase start to appear upon charging, which are marked with corresponding color blocks in the enlarged figure of (001) peak.The peak at around 16.5°marked by the green block indicates the O3 phase.After charging to 3.5 V, only the peak at around 15.8°(O1 phase) marked by the purple block is observed.The peak at 31.9°marked with a dashed box in the left figure also can be assigned to the O1 phase.It implies that the O1 phase dominates the structure at the voltage of 3.5 V due to the stacking order change, which is consistent with previous reports that the O3ÀP3-phase transition occurs at 3.35 V and the P3ÀO1 phase transition occurs at 3.5 V in Na 3 Ni 2 SbO 6 . [1,25]The small bump on the left of the (001) peak for Te-0 at the pristine state may be attributed to P3-phase impurity in this specific sample.After further charging to 4.0 and 4.3 V, the diffraction peaks are not visible because of the decreased crystallinity of the material.However, it is still possible to see the (001) peak of the O1 phase at 15.8°, which slightly shifted to the left when further charged to 4.3 V, indicating the expansion along c axis.After discharging to 3.6 V, the P3 phase appears with a (001) peak at around 16.2°and two-phase coexistence of O1 and P3 phases can be observed.When further discharging to 3.0 V, the O3 phase appears again, showing the mixed O3 and P3 phases.After discharging to 2.0 V, O3 phase dominates the structure with small amount of P3 phase.For TeÀ0.5 (Figure 4b), the P3 phase appears when charged to 3.5 V, showing a new peak of (001) at around 16.2°.Unlike the TeÀ0 sample, there is no O1 phase at this voltage.When charging to 4.0 V, the (001) peaks of the P3 and O3 phases shift to the left due to the removal of Na þ .As further charging to 4.0 V, the diffraction peaks are not visible because of the decreased crystallinity.The P3 phase disappears and the intensity of the peak of the O3 phase increases during the discharge process.It implies that Te substitution in Sb site can suppress the phase transition of P3ÀO1 in the high-voltage region.This can also be reflected in the smoother charge/discharge curves shown in Figure 3a.When the content of Te substitution increases to 0.8, a different phase-transition behavior can be observed, as shown in Figure 4c.The (001) peak of the O3 phase shifts to a lower angle during the charging process, indicating an expansion along c axis.There is no emergence of P3 and O1 phases in the whole charge process, presenting a solid-solution reaction behavior.During the discharge process, the peak of (001) shifts back to higher angle, indicating a high reversible phase-change process.According to previous reports, the Na 3 Ni 2 SbO 6 suffers from rapid capacity degradation, and the O3ÀP3ÀO1-phase transition during the charging process was responsible for the insufficient cycling stability. [39,40]Furthermore, the XRD results shown in Figure S8, Supporting Information, indicate that Te-0.8 maintains a stable hexagonal O3 structure even after 100 cycles.Apart from the peak corresponding to the aluminum current collector, no other peaks indicative of impurities or phase transitions were observed.It is suggested that the material not only remains phase stable in the initial cycle, but also maintains excellent structural stability after 100 cycles.As mentioned before, the P3ÀO1 irreversible phase transition was suppressed effectively by introducing Mg 2þ , Zn 2þ , Co 2þ , etc., but the O3 À P3-phase transition still lacks substantial suppression.In this work, it is revealed that proper Te 6þ doping can effectively tune the dualhoneycomb structure and inhibit the two-step O3ÀP3ÀO1-phase transition of the cathode material during cycling.
To further gain insight into the detailed structural evolution information for TeÀ0.8 during the charging and discharging processes, we performed synchrotron operando XRD measurement at different current densities of 2C and 5C. Figure 5 shows the phase-transition behavior and its 2D contour plots of the operando XRD patterns during the charging/discharging process in the voltage range of 2.0-4.3V, accompanied by the corresponding voltage profiles.The magnified region of synchrotron XRD patterns for 2C and 5C are shown in Figure S9, Supporting Information.As shown in Figure 5a,b, the (001) and (002) peaks shift to a lower angle during the charging process (current density: 2C), which indicates an increase in the oxygen electrostatic repulsion between the oxygen layers due to the removal of Na þ , leading to an expansion of the c-lattice parameters.Meanwhile, the (130) and (060) peaks move toward higher angles during the charging process, corresponding to the contraction of the lattice parameters a and b due to the oxidation of Ni.After further discharging, the peaks show reversible shift, indicating a reversible solid solution reaction.It should be noted that the (001) peak keeps singlet without splitting even charging to 4.3 V, indicating no phase transition in the entire charge process.However, the (130) peak shows that a slight left shift at the end of the charging (as marked by the dashed box), which suggests a small expansion along the a-b plane, may be caused by inner lattice stress release.During the second charging process, it exhibits a highly reversible phase-change behavior, ensuring the high cycle stability.As the current density increases to 5C (Figure 5c,d), there is no phase transition and structure destruction in TeÀ0.8, similar as the phase-change behavior at 2C, which can explain its superior rate capability.Based on the previous results, it is verified that the TeÀ0.8 with Ni/Sb and Ni/Te dual-honeycomb structure can mitigate the undesired irreversible phase transition, thus ensuring high cycle stability.

Charge Compensation
To further investigate the charge compensation mechanism of TeÀ0.8, the synchrotron-based X-ray absorption spectroscopy (XAS) measurement was performed, shown in Figure 6.The normalized K-edge X-ray absorption near-edge structure (XANES) spectra of Ni in TeÀ0.8 during the first cycle process at the states of the pristine, charged to 3.7 V, charged to 4.3 V, discharged to 2.4 V, and discharged to 2.0 V, are shown in Figure 6a.According to previous research, [41,42] the weak absorption peak A for 1s ! 3 d quadrupole-allowed transition appears at about 8333 eV with Ni 3dÀ4p orbital mixing by the structural distortion of NiO 6 octahedra.Two strong absorption peaks for the 1s !4p dipole-allowed transitions with and without shakedown process of ligand to metal charge transfer (LMCT) are observed at about 8342 eV (peak B) and 8351 eV (peak C), respectively.It can be seen that during the Na-ion extraction process, Ni K-edge moves to higher energy when charging to 4.3 V, indicating the oxidation of Ni 2þ to Ni 3þ referring to NiO and LiNiO 2 .It should be noted that the peak intensity decreases for the LMCT process, which is due to the dominant distribution of Ni 3þ ÀO 2À ion pair from the Ni 2þ ÀO 2À in the pristine. [41]Figure 6b shows the Fouriertransformed (FT) X-ray absorption fine structure (EXAFS) spectra of Ni K edge in TeÀ0.8 electrodes at the corresponding states.The fitting results of the five samples are shown in Figure 6c-e, and Figure S10a,b and Table S11, Supporting Information.The first peak at R = ≈2.1 Å corresponds to the Ni─O coordination shell, and the peak at R = ≈3.0Å is attributed to the Ni─Ni coordination shell.The Ni─O and Ni─Ni distances of the pristine TeÀ0.8 are 2.11 and 3.01 Å, respectively.After charging to 4.3 V, the Ni─O distance decreases to 2.02 Å and Ni─Ni distance decreases to 2.97 Å.In the discharge process, the Ni─O distance increases reversely to 2.11 Å and the Ni─Ni distance increases to 3.02 Å.The average interatomic distances of Ni─O and Ni─Ni bonds change during charging process, which is in good agreement with the XANES study, according to the general relationship between the peak position (E) and interatomic distance (R), (E À E 0 )•R 2 = constant. [41]This is attributed to the change of ionic radius of Ni ions due to the redox reaction.It can be seen that the oxygen shell of Ni─O shows a split (marked by green block) during the charging process, which is consistent with a distorted octahedra due to the strong Jahn-Teller effect of the low-spin Ni 3þ ions. [43]After discharge to 2.0 V, the distortion disappears, indicating the reversibility of the structural change and valence change of Ni.To investigate the surface change of TeÀ0.8 during cycling, X-ray photoelectron spectroscopy (XPS) was also performed for Ni 2p 3/2 and Ni 2p 1/2 in TeÀ0.8 cathode at different states during the first cycle, as shown in Figure S11, Supporting

Theoretical Calculation
To understand the effect of Te 6þ substitution on the electronic structure of Na 3 Ni 2 SbO 6 , DFT calculation was performed and the results are shown in Figure 7. Compared to TeÀ0 and TeÀ1.0 in Figure 7a-c, the bandgap in the d band structure of Ni in TeÀ0.8 disappears, indicating better electronic conductivity after the introducing proper amount of high-valent Te 6þ into the lattice of the pristine cathode material.This can be attributed to the more covalent nature of the Te-O bond in TeÀ0.8. [44]In addition, the density of state (DOS) distribution of Ni 3d band near the Fermi level (E F ) was enhanced.This implies that Te 6þ doping in the cathode material increases the number of electrons near E F , thus improving the electronic conductivity.The DOS distribution of O 2p near E F orbitals increased to maximum, which also explains the best covalent nature of the Te─O bond in TeÀ0.8.However, when all the Sb atoms were replaced by Te atoms in TeÀ1.0, the bandgap of Ni around E F appears and the number of electrons of Ni and O near the E F decreases, indicating a decrease in the electronic conductivity.This is an important factor limiting its rate performance.The electrochemical behavior shown in Figure 3

Conclusion
In summary, we construct a dual-honeycomb-structure-layered cathode material Na 2.2 Ni 2 Sb 0.2 Te 0.8 O 6 by proper amount of Te substitution.It is revealed that Te substitution can not only elevate the average discharge voltage from 3.2 to 3.8 V remarkably, but also improving cycle stability and alleviating voltage decay.Synchrotron operando XRD reveals highly reversible phase-transition behavior of TeÀ0.8 during cycling, and the O3ÀP3ÀO1-phase transition is suppressed effectively by Te substitution.As a result, TeÀ0.8 exhibits promising electrochemical performance with a capacity retention of 70.9% after 1000 cycles at 1C. XPS and XAS results demonstrate that the charge compensation of TeÀ0.8 during charge and discharge is mainly contributed by Ni redox.Combining with DFT calculations, it is revealed that the proper amount of Te substitution in the cathode material can promote the rate and long-cycle capability cathode by suppressing phase transition and decreasing the bandgap.This work provides new opportunities for the future design of high-performance honeycomb-layered cathodes for SIBs.

Experimental Section
Materials Synthesis: A series of Na 3Àx Ni 2 Sb 1Àx Te x O 6 (x = 0, 0.2, 0.5, 0.8, 1.0) were synthesized by the solid-state reaction method.A stoichiometric ratio of Na 2 CO 3 , NiO, Sb 2 O 3 , and TeO 2 were grounded in an agate mortar, and the mixture was pressed into a tablet by a powder pressing machine.A slight excess of 5 mol% Na 2 CO 3 was used to compensate for the loss of sodium volatility.The tablets were calcined at 1000 °C for 12 h under flowing oxygen followed by cooling down to room temperature naturally except that TeÀ1.0 was heated at 900 °C according to the previous report.Afterward, the samples were grounded and transferred to an argon-filled glove box for further handling.
Materials Characterization: The crystal structure of the as-prepared products was investigated by XRD using an X-ray diffractometer (Bruker D8 Advance, Germany) with Cu-Kα radiation (λ = 1.5418Å) in the range of 10°À80°at 40 kV and 40 mA.The morphology of the materials was characterized by SEM (FEI Quanta FEG 250), and the elemental mappings were recorded by EDX on SEM.A TEM (JEOL JEM-200CX, JEOL JEM-2100Plus) was used for analyzing the microstructure of the samples.Synchrotron operando XRD was performed at the Shanghai Synchrotron Radiation Facility (SSRF) on beamline BL14B1 (λ = 0.8857 Å).Hard XAS was performed on beamline BL14W1 in SSRF and BL14B2 at Super Photon Ring-8 in Japan.For FT-EXAFS fitting, the AUTOBK code was used to normalize the absorption coefficient and separate the EXAFS signal χ(k) from the isolated atom absorption background.The extracted EXAFS signal χ(k) was weighted by k 2 to emphasize the high-energy oscillations and then FT in a k range from 3.0 to 11.5 Å À1 using a Hanning window with a window sill Δk of 1.0 Å À1 to obtain magnitude plots of the EXAFS spectra in R-space (Å).The least-squares fits were carried out in R-space.XPS test was performed using the Thermo Kalpha, Thermo ESCALAB 250XI, Axis Ultra DLD Kratos AXIS SUPRA, and PHI-5000 VersaProbe III instruments.The cathode was prepared by mixing 70 wt% active material powder, 20 wt% Super P, and 10 wt% polyvinylidenefluoride dissolved in the N-methyl-2-pyrrolidone.The cathode slurry was coated on aluminum foil and dried at 70 °C under vacuum for 12 h.Afterward, the electrodes were punched into a disk of 12 mm in diameter.The CR2032-coin cells were assembled with Na foil as an anode and the glass fiber as a separator to evaluate the electrochemical performance.And, 1 M NaClO 4 in a nonaqueous solution of ethylene carbonate (EC) and dimethyl carbonate with a volume ratio of 1:1 was used as the electrolyte for the cells.The cells were assembled in an argon-filled glove box (Vacuum Atmospheres Company, H 2 O < 0.5 ppm, O 2 < 0.2 ppm).The charge-discharge performance was measured on a Land CT2001A battery test system at 25 °C.The cells were cycled between 2.0 and 4.3 V versus Na þ /Na.The current density was set as 1C = 100 mA g À1 .GITT experiments were performed by charging/ discharging for 30 min at a current density of 10 mA g À1 and relaxing for 2 h to reach a quasi-equilibrium state.EIS was tested on EC-lab with a frequency range from 10 mHz to 1 MHz.
Theoretical Calculations: DFT calculations were performed to study the differential charge analysis and DOS of TeÀ0, TeÀ0.5, and TeÀ1.0.The theoretical calculations were conducted in Cambridge Sequential Total Energy Package.The structures of TeÀ0, TeÀ0.5, and TeÀ1.0 were built according to the XRD refinement result.A 1 Â 1 Â 1 supercell for TeÀ0, a 5 Â 1 Â 1 supercell for TeÀ0.5, and a 1 Â 1 Â 1 supercell for TeÀ1.0 were built separately.Geometry optimization of each structure was conducted separately and generalized gradient approximation-perdew burke ernzerhof functional was employed.The cutoff energy was 598 eV and ultrasoft pseudopotential was applied.The convergence tolerance was set as follows: 1 Â 10 À5 eV atom À1 for energy, 0.03 eV Å À1 for max force, 0.05 GPa for max stress, and 0.001 Å for max displacement.Self-consistent field tolerance was set to 1 Â 10 À5 eV atom À1 .

Figure 2 .
Figure 2. a-e) TEM image of TeÀ0.8 at different magnification ((c) is an enlarged view of the box in (b); the blue atoms represent Ni and the yellow atoms represent Sb or Te).f ) SAED pattern of TeÀ0.8 along the [01 1] zone axis direction (the selected-area is marked by the yellow box in (e)).
shows the EIS curves recorded at the open-circuit voltage of TeÀ0, TeÀ0.8, and TeÀ1.0, and the corresponding fitting results are listed in Table S10, Supporting Information.The impedance spectra include R s representing electrolyte resistance and a semicircle in the high-mid-frequency zone representing the charge transfer resistance (R ct ).The values of R ct for TeÀ0, TeÀ0.8, and TeÀ1.0 samples are 204.6,108.1, and 171.3 Ω, respectively.It is noted that the R ct value of TeÀ0.8 is the lowest among them, showing fast diffusion of Na þ in the materials.This could be one of the reasons for the high rate capability of TeÀ0.8.The superior-sodium-ion diffusion capability of Te-0.8 aligns with its electrochemical performance, affirming the efficacy of the TM ordering tuning strategy.

Figure 5 .
Figure 5. Structural evolution of TeÀ0.8 and in the first charge, first discharge, and second charge processes cycled at different current densities.a) Operando XRD of TeÀ0.8 cycled at 2C between 2 and 4.3 V and b) its contour plot of the (003) peak evolution.c) Operando XRD of TeÀ0.8 cycled at 5C between 2 and 4.3 V and d) its contour plot of the (003) peak evolution.(λ = 0.8857 Å).
Figure7a-c, the bandgap in the d band structure of Ni in TeÀ0.8 disappears, indicating better electronic conductivity after the introducing proper amount of high-valent Te 6þ into the lattice of the pristine cathode material.This can be attributed to the more covalent nature of the Te-O bond in TeÀ0.8.[44]In addition, the density of state (DOS) distribution of Ni 3d band near the Fermi level (E F ) was enhanced.This implies that Te 6þ doping in the cathode material increases the number of electrons near E F , thus improving the electronic conductivity.The DOS distribution of O 2p near E F orbitals increased to maximum, which also explains the best covalent nature of the Te─O bond in TeÀ0.8.However, when all the Sb atoms were replaced by Te atoms in TeÀ1.0, the bandgap of Ni around E F appears and the number of electrons of Ni and O near the E F decreases, indicating a decrease in the electronic conductivity.This is an important factor limiting its rate performance.The electrochemical behavior shown in Figure3indicates that the increased electronic conductivity in TeÀ0.8 suppresses voltage polarization of the cathode material during cycling, which ultimately improves the cycling performance.It should be noted that the d orbitals of the Te atom in the honeycomb TeÀ0.8 structure help to increase the valence band width, leading to the structure filling the valence band below À6.5 eV, which can explain the high Ni redox voltage in TeÀ0.8 structure, in good agreement with the high charge/discharge plateaus shown in Figure 3a.The schematics of electronic structures in TeÀ0, TeÀ0.8, and TeÀ1.0 are shown in Figure 7d.It suggests that the introduction of

Figure 6 .
Figure 6.Charge compensation mechanism of TeÀ0.8 during cycling.a) Ni K-edge XANES spectra of TeÀ0.8 collected at different charge/discharge states.b) Corresponding FT-EXAFS spectra of Ni K-edge in TeÀ0.8 electrodes during the first charge and discharge processes.c-e) Ni K-edge least-square fits of the calculated FT-EXAFS phase and amplitude functions (solid red lines) to the experimental FT-EXAFS spectra (solid and open circles) for c) pristine and d) fully charged (4.3 V) and (e) fully discharged (2.0 V) TeÀ0.8.