2D Nb2O5@2D Metallic RuO2 Heterostructures as Highly Reversible Anode Materials for Lithium-ion Batteries

The 2D Nb 2 O 5 @RuO 2 heterostructures possess short diffusion length of ions/electrons, easy penetration of liquid electrolyte, and high conductivity transport of electrons. Abstract Constructing two-dimensional (2D) heterostructured materials by stacking different 2D materials could combine the merits of the individual building blocks while getting rid of the associated shortcomings. Orthorhombic Nb 2 O 5 (T-Nb 2 O 5 ) is one of the greatly promising candidates for durable and safety anode for Li-ion batteries (LIBs), but it usually exhibits poor electrochemical performance due to the low electronic conductivity. Herein, we realize excellent lithium storage performance of T-Nb 2 O 5 by designing 2D Nb 2 O 5 @2D metallic RuO 2 heterostructures (Nb 2 O 5 @RuO 2 ). The presence of 2D metallic RuO 2 leads to enhanced electronic conductivity. The 2D Nb 2 O 5 @RuO 2 heterostructures possess very short diffusion length of ions/electrons, easy penetration of liquid electrolyte, and high conductivity transport of electrons through the 2D metallic RuO 2 to 2D Nb 2 O 5 . The Nb 2 O 5 @RuO 2 delivers remarkable rate performance (133 mAh g -1 and 106 mAh g -1 at 50 C and 100 C) and excellent long-life capacity (97 mAh g -1 after 10000 cycles at 50 C). Moreover, Nb 2 O 5 @RuO 2 //LiFePO 4 full batteries also display high rate capability of 140 mAh g -1 and 90 mAh g -1 at 20 C and 50 C, respectively. Theoretical calculation results show that the 2D Nb 2 O 5 @RuO 2 heterostructures possess more large adsorption ability for Li + than that of Nb 2 O 5 , indicating an excellent lithium storage performance.

Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.

Introduction
Nowadays, the energy crises and environmental pollution resulting from the overdependence of fossil fuels have drawn the worldwide attention. There is an increased demand for developing clean and renewable energy. Lithium-ion batteries (LIBs), as chemically storing energy devices, have been widely applied in portable electronic devices and demonstrated as promising energy storage system for hybrid electric vehicles (HEVs), electric vehicles (EVs), and smart grids. [1][2][3] The developments of LIBs with high energy density and high power density rely heavily on the high performance electrode materials. Graphite is the most conventional anode material for LIBs, which possesses a lithiation potential of below 0.2 V (vs. Li + /Li). [4] Such low lithiation potential induce major safety issues due to the electrolytic 1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60 Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.
F o r R e v i e w O n l y decomposition and lithium dendrites formation. Furthermore, it displays unsatisfactory capacity at high current density due to its low Li ion diffusion coefficient and the formation of solid electrolyte interphase (SEI) on its surfaces. [5][6][7][8] Typical insertion-type anode materials, such as Li4Ti5O12, [9][10] TiNb2O7, [11][12] orthorhombic Nb2O5 [13][14][15] are investigated extensively because of their better capacity retention than convention-and alloy-type ones. Among them, orthorhombic Nb2O5 (T-Nb2O5) has shown excellent lithium ions pseudocapacitive in nonaqueous electrolytes since the empty octahedral sites between (001) planes offer natural tunnels for lithium ion transport. [16] When used for LIBs, the T-Nb2O5 shows a suitable voltage range from 1.1 to 2.0 V vs. Li + /Li, corresponding to a comparatively high reversible theoretical capacity of about 200 mAh g -1 to form LixNb2O5 (x = 2). [13,[17][18][19][20] Moreover, it can use aluminum foil as the anode current collector rather than expensive and heavy copper foil, because Li does not alloy with aluminum above 0.3 V vs. Li + /Li. [21][22] However, the intrinsic drawback of poor electrical conductivity (≈ 3 × 10 −6 S cm −1 ) and severe pulverization during charge/discharge prevent its practical applications for high performance LIBs. [23] Considerable attention has been made to enhance the electrochemical performance of T-Nb2O5. Construction of nanosized T-Nb2O5 (i.e. nanobelts, [24] vein-like nanoporous networks, [25] hierarchical microspheres [26] and nanofibers [27] ) is one of the most effective approach. Another strategy for overcoming low electronic conductivity is to mix Nb2O5 with conductive materials such as graphene [28] , carbon nanotubes (CNTs), [29][30] Graphite [31] When used for energy storage system, two-dimensional (2D) materials show unprecedented properties that differ from their bulk systems, because the characteristic time constant τ for ion diffusion is proportional to the square of the diffusion length L (τ ≈ L 2 /D). [32] Single-crystalline mesoporous T-Nb2O5 nanosheets/graphene composites have demonstrated enhanced intercalation/deintercalation characteristics by shortening ions diffusion length and accelerating electron transport. [33] 2D heterostructured materials, prepared by stacking different 2D crystals on top of each other at the nanoscale, could combine the collective advantages of individual building blocks and synergistic properties. In addition, the 2D heterostructured materials have high specific surface area, enabling improved performances for surface dominant reactions. Gogotsi et al reported that MoS2-on-MXene heterostructures show high performance lithium storage performance. [34] One can expected that rational design 2D T-Nb2O5@2D metallic metal oxide heterostructures could significantly improve its energy storage properties. Among 2D metal oxides, RuO2 nanosheets are considered to be the most 1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60 Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.
promising metal oxides because of their metal-like electrical conductivity, which exhibits a highly conductive property (≈2 × 10 4 S cm -1 ) [20,35] and fast Li permeation. [36] In this work, we designed 2D Nb2O5@2D metallic RuO2 heterostructures (denoted as Nb2O5@RuO2) as anode for high performance LIBs. This unique 2D heterostructures combine a variety of advantages: 1) The intimate interfacial interaction between T-Nb2O5 and the metallic RuO2 can not only facilitate the electron transfer, but also offer high electrode/electrolyte contact surface area for quick electrolyte infiltration, reaching the full potential of the active material (Nb2O5).
2) The confined thickness within a nanoscale range of Nb2O5@RuO2 shorten the diffusion length of lithium ions.
3) The Nb2O5 nanosheets possess strong interaction with RuO2 nanosheets, which keeps the structure stability of Nb2O5@RuO2 after repeated Li + intercalation/deintercalation, thus leading to excellent electrochemical performance. When used as anode for LIBs, the Nb2O5@RuO2 can deliver excellent cyclability and rate performance, which demonstrates capacity of 100 mAh g -1 after 10000 cycles at 50 C. Additionally, a full battery has been assembled by using Nb2O5@RuO2 and LiFePO4 as anode and cathode, respectively. The full battery demonstrates enhanced charge and discharge ability (140 mAh g -1 and 90 mAh g -1 at 20 C and 50 C, respectively). Moreover, theoretical calculations reveal the presence of the RuO2 layer indeed improves the stability of lithium ions adsorption and reduces the diffusion barrier of lithium ions in the Nb2O5.

Materials synthesis
As reported previously, [35,37] The KNb3O8 and NaRuO2 were synthesized by calcination of stoichiometric Nb2O5 and K2CO3 at 900 °C for 10 h in air and Na2CO3, Ru and RuO2 at 900 °C for 12h in Ar atmosphere, respectively. Protonation was realized by mixed KNb3O8 and NaRuO2 with acid aqueous for one week. In a typical procedure, 1.0 g of KNb3O8 and NaRuO2 were dispersed in 60 mL nitric acid (6 mol L -1 ) and 500 mL hydrochloric acid (1 mol L -1 ), respectively. The exfoliation of HNb3O8 and HRuO2 were achieved by intercalation of tetrabutylammonium (TBA) cation into the layer. Note that the TBA + /H + = 5. The resultant solution was shaken for two weeks and followed by ultrasonic treatment. Subsequently, the suspensions were centrifuged at 2000 rpm for 30 min to separate the non-exfoliated sediment. The dry TBA-Nb3O8 and TBA-RuO2 were obtained by after centrifugal separation and drying in vacuum freeze

Accepted Article
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.
drier. The TBA-Nb3O8 and TBA-RuO2 were dissolved in water with mass ratio of 6:1 to form uniform solution by stirring and ultrasonic dispersion. Subsequently, the Nb2O5@RuO2 was obtained by freeze drying. Finally, the precursor was annealing at 600 °C for 2h in air. The pure Nb2O5 nanosheets were obtained by directly heating the TBA-Nb3O8 at same condition.

Materials characterization
The morphologies of samples were observed by scanning electron microscopy (CIQTEK SEM3100 and Hitachi, SU8200) and transmission electron microscopy (JEOL, JEM-2100F). The crystalline structure of samples was examined by X-ray diffraction (TTR-III, Rigaku). Bonding states of the atoms were investigated by XPS (Thermo-VG Scientific, ESCALAB 250). Raman spectrum was obtained by a LanRanHR (HORIBA Scientific, Paris, France) spectrometer. Nitrogen adsorption/desorption isotherms were characterized by an ASAP 2020 accelerated surface area and porosimetry system. The thermogravimetric analysis using TGA Q5000 instrument (10 °C/min from 30 to 800 °C in air).

Electrochemical characterization
To prepare the electrodes for 2032 coin cell assembly. The samples were mixed with acetylene black and poly(vinylidene) in a mass ratio of 7:2:1 in an N-methylpyrrolidone solvent. Then the slurry was coated on the copper foil and dried in vacuum oven 60 °C overnight. The cell was assembled in glove box (the water and oxygen concentrations were kept below 0.1 ppm), using Li metal as counter and reference electrode and Celgard 2400 as the separator. The electrolyte was composed of 1 M LiPF6 dissolved in ethylene carbonate (EC) and dimethyl carbonate (DEC) (1:1, w/w). The CV curves and EIS were obtained on electrochemical workstation (CHI 660D). The charge and discharge were evaluated by battery test system (Neware BTS-610). For the fabrication of Nb2O5@RuO2//LiFePO4 full cell, the commercial LiFePO4 electrode was used as cathode and the activated Nb2O5@RuO2 was used as anode. The Nb2O5@RuO2 electrodes used in the full cells were cycled in half cells before they were assembled into full cells. The full cell was fabricated with Nb2O5@RuO2 as the anode and LiFePO4 as the cathode with a mass ratio of 1:3, the capacity ratio of the negative electrode to positive electrode (N/P ratio) is about 0.67.

Accepted Article
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.
The calculations for structure optimization, self-consistent and energy barrier are carried out with the ABACUS package. [38] The exchange-correlation function is treated by the generalized gradient approximation (GGA) in the form of the Perdew-Burke-Ernzerhof (PBE) functional. [39] We use the SG15 optimized Norm-Conserving Vanderbilt (ONCV) multi-projector pseudopotentials to describe the electronic potential field of the core. [40] At the same time, the numerical atomic orbital basis set is expanded using Standard DZP Orbitals. The convergence criterion for electronic selfconsistent calculations is 5.0E-7 eV. Both the atomic positions and the lattice parameters are allowed to change during the structural relaxation until the Hellmann-Feynman forces on each atom are lower than 2.0E-2 eV/Å.We employs numerically tabulated atom-centered orbitals as the basis functions to expand electronic wave functions. Specifically, we choose the basis sets 4s2p2d1f for Ru and Nb elements, 2s2p1d for O element and 4s1p for Li element. Brillouin-Zone (BZ) integration was sampled by Gamma point in the reciprocal space in order to improve computational efficiency. We take the van der Waals DFT-D3 interaction into account in the calculation of the Nb2O5@RuO2 heterojunction system, which makes the simulation results more accurate. [41] The establishment of hypothesis model of T-Nb2O5: Previous experimental fitting proposed a virtual crystal model of T-Nb2O5 (Nb16.8O42), but the fraction occupation of Nb or O made DFT calculation expensive. Therefore, we made the following rational approximations to the T-Nb2O5 structure model: (1) A small amount of the amorphous Nb cannot be considered because the location is not defined and stoichiometry is fractioned; (2) Two nearest Nb atoms (the distance is 0.43 Å, occupations are 0.5) are replaced with their average sites (occupations are 1.0) owing to avoid Nb deviating from the equilibrium positions during self-consistent calculations; (3) Two oxygen atoms at the 2b Wyckoff positions are removed from the model to balance the charge of the entire unit cell, and there is no Raman active modes at 2b Wyckoff positions. [42] Through these approximations, the unit cell of T-Nb2O5 shows higher space symmetry than the original Nb16.8O42 model and space group is Pbam.
The crystalline phase of Nb2O5, RuO2 and Nb2O5@RuO2 were studied by XRD ( Figure S5a). The XRD patterns of Nb2O5 and RuO2 exhibit the standard patterns of Nb2O5 (JCPDS No. 30-0873) [45][46] and RuO2 (JCPDS No. 40-1290), [47] respectively. The XRD pattern of Nb2O5@RuO2 heterostructures confirms that its composition contains both Nb2O5 and RuO2. The construction of the Nb2O5@RuO2 heterostructure reduces the particle size and crystallinity of RuO2, which leads to weaker intensity and larger half width of the diffraction peak. [48][49] Figure S5b displays the Raman spectra of Nb2O5, RuO2 and Nb2O5@RuO2. The RuO2 shows three distinct peaks located at 523, 643 and 713 cm -1 . [50][51] And the Nb2O5 exhibits four clearly peaks located at 126, 239, 687 and 990 cm -1 . [52] In the case of the Nb2O5@RuO2, distinct peaks with those of the RuO2 and Nb2O5 confirm the formation of heterostructures. Figure S6 reveals the X-ray photoelectron spectroscopy (XPS) spectrum of Nb2O5@RuO2, the typical shape peaks of Nb 4s, Nb 4p, Nb 3s, Nb 3p, Ru 3p, Ru 3d and O 1s species can be clearly observed in XPS survey spectrum, which demonstrates that the Nb2O5@RuO2 includes Nb, Ru and O elements. The high-resolution XPS spectrum can analyze the chemical environment of every element in composites. According to the high-resolution Nb 3d XPS spectrum (Figure S7a), two remarkable peaks located at about 209.9 and 207.1 eV, analogous to Nb 3d3/2 and Nb 3d5/2, respectively. [33,53] As for the high-resolution of Ru 3d (Figure S7b), the binding energies of about 284.8 and 280.6 eV match with Ru 3d3/2 and Ru 3d5/2, respectively. [54][55] Compared with the pure Nb2O5 power, the peaks of Nb 3d exhibit slight shift to high bonding energy, demonstrating the enhanced valence state of Nb. Meanwhile, the peaks of Ru 3d display obvious shift to low  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60 Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. During the first Li + insertion process, two reduction peaks located at 1.83 V and 1.5 V were observed, which is attributed to the redox reaction of Nb2O5 (Nb2O5 + x Li + + x e -= LixNb2O5), accompanied by a little decomposition of electrolyte and formation of stable solid electrolyte interphase (SEI) layer. [56][57] The oxidation peaks were observed at around 1.75 V and 1.93 V during charge process, corresponding to the oxidation of Nb from Nb 4+ to Nb 5+ . [57][58] The CV curves almost overlap after the first cycle, revealing good reversibility after the initial capacity loss. The first three galvanostatic charge/discharge curves of Nb2O5@RuO2 are shown in Figure 3b. The Nb2O5@RuO2 delivers a reversible charge capacity about 285 mAh g -1 at 1 C (1 C = 200 mA g -1 ), with an initial Columbic efficiency (ICE) of 61.4%. The initial irreversibility may result from the SEI formation and irreversible Li + insert into Nb2O5 and RuO2. [17] The typical charge/discharge profiles of RuO2 were shown in Figure S10, the RuO2 reacts with Li via conversation mechanism in the voltage window below 1.1 V. Thus, the capacity of Nb2O5@RuO2 can be mainly attributed to the insertion reaction between lithium ion and Nb2O5. The cycling stability of the Nb2O5@RuO2 and the Nb2O5 at 5 C are shown in Figure 3c. Obviously, the Nb2O5@RuO2 shows higher reversible capacity (165 mAh g -1 after 1000 cycles) than that of the Nb2O5 (129 mAh g -1 after 1000 cycles). The capacity decay of the Nb2O5@RuO2 (Figure 3c) was very slow, only 0.03% decay per cycle after the first three cycles at 1 C. The Nb2O5@RuO2 not only displays excellent cyclability at a low

Accepted Article
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.
F o r R e v i e w O n l y current density but also shows stable cycle life under a high current density. The SEM images of Nb2O5@RuO2 for LIBs after 100 cycles at rate of 5 C is shown in Figure   S11, it can still well maintain the 2D morphology. Figure 3d shows the cycling performance of the Nb2O5@RuO2 electrode at 50 C. After the initial capacity loss, the Columbic efficiency reaches ≈ 100%. It delivers extraordinary long-life stability of around 97 mAh g -1 at 50 C after 10000 cycles, which is higher than that of other Nb2O5based anode for LIBs (Table S1).
Where a and b are two parameters of value from 0.5 to 1. It was agreed that b = 0.5 reveals a diffusion-controlled process resulted from Li + intercalation, and b = 1 implies a capacitive process by surface charge storage mechanism. [65][66] In Figure 4b, the b value calculated from cathodic and anodic peaks are 0.88 and 0.96, respectively, exhibiting the pseudocapacitive behavior of Nb2O5@RuO2. It can explain why Nb2O5@RuO2 achieved extraordinary rate performance. The current can be described as a sum of two parts at every potential value, which are capacitive contribution (av) and diffusion-limited contribution (av 1/2 ). Therefore, it can be written as follow: Where k1 and k2 are two parameters, which can be easily work out according to Equation (4). According to the integral area of the CV curve, the contribution of capacitance is about 88% of the total charge storage (Figure 4c) at scan rate of 10 mV s -1 . The contribution ratios of capacitive and diffusion-controlled at different sweep rates have been calculated as shown in Figure 4d. The percentage of capacitive contribution continuously increases with the increases of sweep rate.
To further demonstrate the possibility of practical application, lithium-ion full cells were assembled by using commercial LiFePO4 (LFP) as cathode and activated Nb2O5@RuO2 as anode between 1.0 and 3.0 V, as the schematic illustration in Figure   5a. [67][68] The XRD pattern and SEM image of LFP are shown in Figure S13 and Figure  S14, respectively. The morphology of LFP is near-spherical with size of 1-10 μm. The typical charge/discharge curves of LFP half cell ranging from 2.2-4.0 V indicate a plateau around 3.4 V with a discharge capacity of 150 mAh g -1 at 1 C (Figure S15).
The LiFePO4 electrode delivers the capacities of 158, 154, 150, 142, 125, 103, and 70 mAh g -1 at 0.2, 0.5, 1, 2, 5, 10, and 20 C, respectively ( Figure S16). Figure 5b shows the charge/discharge curves of LPF//Nb2O5@RuO2 full cell at rate of 2 C to 50 C. The full battery delivers a reversible capacity of about 286 mAh g -1 at 2 C with an average voltage plateau at 1.8 V (the specific capacity and current density are based on the weight of anode material).  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60 Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. respectively. Figure 5d shows the cycle performance of full cell at 20 C. Remarkably, the full battery maintains good cycling stability and exhibits high reversible capacity (82 mAh g -1 after 500 cycles) with Coulombic efficiency nearly reach 100%. The excellent electrochemical performance of the full cell confirms the potential of practical application of the Nb2O5@RuO2.   1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60 Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. The density functional theory (DFT) calculations were performed to explain the lithium storage mechanism and ions transport in Nb2O5 and Nb2O5@RuO2. The structure of the Nb2O5 crystal is obtained after structural optimization. The atomic arrangement along the c-axis direction is formed by the stacking of 4h layers and 4g layers in turn, of which 4h layers become the primary Nb-O bonding layers, while the 4g layers are completely occupied by oxygen atoms (Figure 6a). [42,69] The larger spatial gap makes the 4g layer an ideal layer for lithium ions adsorption and bonding. Similarly, the structure-optimized Nb2O5/RuO2 heterojunction system is prepared for subsequent  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60 Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.  (Figure 6b). We found eight possible lithium ions adsorption sites on the Nb2O5 through structure optimization (Figure 6a), and the corresponding adsorption energy values of different adsorption sites (labeled in Figure 6a) range from -3.31 eV to -3.53 eV (Figure 6c and Table S2). Besides, the lithium ions adsorption energy on Nb2O5@RuO2 (labeled in Figure 6b) range from -3.98 to -4.17 eV (Figure  6c and Table S2). The adsorption energy of lithium ions among Nb2O5 and Nb2O5@RuO2 are shown in Figure 6c, revealing the introduction of the RuO2 layer improves the stability of lithium ions adsorption on the T-Nb2O5 4g layer. Then, we simulated the diffusion barrier of lithium ions in the 4g layer along the x-y plane of Nb2O5 and Nb2O5@RuO2 using the Climbing Image Nudged Elastic Band (CINEB) method. [70][71] We simulate the two possible diffusion systems, namely: path 1 is that lithium ions transport from site 4 to another site 4 of the nearest neighbor lattice and path 2 is lithium ions transport path from site 4 to nearest stable site 5 in the same unit cell. The lithium ions diffusivity related migration barriers of Nb2O5 along path 1 and path 2 are summarized and compared with those of Nb2O5@RuO2 in Figure 6d and

Conclusions
In summary, Nb2O5@RuO2 heterostructures have been synthesized via a top-down liquid exfoliation strategy and followed by annealing in air. The intimate interfacial interaction between T-Nb2O5 and the metallic RuO2 combine the collective merits of Nb2O5 NSs and RuO2 NSs and facilitate the synergistic advantages. The metallic RuO2 can enhance the electronic conductivity of Nb2O5 and realize fast Li accessibility. The 2D structure of Nb2O5 with high surface can shorten the diffusion path of ions and electrons. The Nb2O5@RuO2 endows remarkable rate performance (133 mAh g -1 and 106 mAh g -1 at 50 C and 100 C, respectively) and excellent long-life capacity (100 mAh g -1 after 10000 cycles at 50 C). The kinetic analysis reveals the pseudocapacitive storage process of Nb2O5@RuO2, which enhances the electrochemistry performance at rapid charge and discharge process. Furthermore, Nb2O5@RuO2//LiFePO4 full cell also delivers capacities of 140 mAh g -1 and 90 mAh g -1 at 20 C and 50 C, respectively, indicating its excellent electrochemical performance. The computation results  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60 Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.

Conflict of interest
Authors should declare whether or not the submitted work was carried out with a conflict of interest. If yes, please state any personal, professional or financial relationships that could potentially be construed as a conflict of interest. If no, please add "The authors declare no conflict of interest".

Response:
We have added "The authors declare no conflict of interest".

References
Please revise the mode of citation according to References of the attached Manuscript Submission Template.
-Please make sure the format of the authors' name in references is "the first name, the last name".
-Please delete the ending page number and the volume number.
-Please check that the references are up to date.
-Please remove any references that are unpublished and move them to the text, if applicable.

Response:
We have revised the mode of citation according to References of the attached Manuscript Submission Template.