Efficient Surface Modulation of Single- Crystalline Na2Ti3O7 Nanotube Arrays with Ti3+ Self-Doping toward Superior Sodium Storage

Jinlong Liu,†,‡ Zhenyu Wang,‡ Zhouguang Lu,‡ Lei Zhang,† Fangxi Xie,† Anthony Vasileff,† and Shi-Zhang Qiao*,†
†Center for Materials in Energy and Catalysis, School of Chemical Engineering and Advanced Materials, The University of Adelaide, Adelaide, South Australia 5005, Australia
‡Department of Materials Science and Engineering, Southern University of Science and Technology, Shenzhen 518055, P.R. China
*S Supporting Information

ABSTRACT: Although Na₂Ti₃O₇-based anodes have been widely investigated in sodium-ion batteries (SIBs), their Na⁺ storage properties especially high-rate capability and long-term cycling durability are far from practical application, because of their intrinsic low conductivity and unsatisfied Na⁺ diffusion resistance. Here, we report the surface engineering of Na₂Ti₃O₇ nanotube arrays grown in situ on Ti foil through a hydrothermal method and subsequent NH₃-assisted calcination. Benefiting from the effective surface modification, the as-derived free-standing electrode possesses highly crystalline surface with favorable Na⁺ diffusion kinetics and self-incorporation of abundant

Ti³⁺ for improved electronic conductivity. These features enable the electrode to achieve remarkable reversible capacity (237.9 mAh g⁻¹), ultra-high rate capability (88.5 mAh g⁻¹ at 100 C = 17.7 A g⁻¹), and excellent cycling stability (92.32% capacity retention at 50 C after 5000 cycles), which are superior to the counterpart without surface modification, as well as almost all Na₂Ti₃O₇-based anode materials reported so far for SIBs. The outstanding electrochemical performance demonstrates the feasibility of proposed surface modulation in designing more efficient electrode materials for energy storage.
odium-ion batteries (SIBs), as a promising energy storage technology for various applications, such as electric vehicles and smart grids, have aroused extensive research interest because of the easy availability and low cost of sodium resources.1−3 Thanks to the rocking-chair working mechanism of SIBs being essentially same as that of lithium- ion batteries (LIBs), great progress has been achieved by applying successful LIB principles to SIBs, especially in the exploration of cathode materials via replacing lithium ions with sodium ions.1,4 Unfortunately, conventional graphite anodes in LIBs can hardly be applicable to accommodate Na+ because of
the large atomic weight and ionic radius of Na+ with more sluggish diffusion kinetics,5,6 and the advancement of SIBs is severely limited by the development of suitable anode materials that allow for reversible Na+ insertion and extraction.7,8 Therefore, considerable efforts have been

for Na+ storage, considering its unique layered structure with the lowest voltage (0.3 V vs Na+/Na) among different reported oXide insertion SIB electrodes.10
Although Na2Ti3O7 can accommodate ∼3.5 Na+ per formula unit in theory, corresponding a capacity of 310 mAh g−1,11 its intrinsic insulating nature with a large bandgap (3.7 eV) poses
a major barrier to efficient Na+ storage,12 particularly under ultra-high current densities. In this regard, different material modification strategies (e.g., carbon incorporation and surface engineering) have been employed to enhance the electronic conductivity of Na2Ti3O7. Coupling Na2Ti3O7 with carbon materials is the most widely adopted approach to address this issue.13−17 For instance, Xie et al. reported N-doped carbon- coated Na2Ti3O7 hollow spheres as an improved SIB anode with excellent rate performance.13 In spite of the increased conductivity, carbon incorporation strategy especially through

devoted to searching for efficient anode materials for SIBs.

Among the emerged SIB anode candidates, Ti-based materials are very appealing owing to their abundance, cost-effectiveness, and nontoXicity.9 In particular, Na2Ti3O7 holds great potential

Figure 1. Schematic illustration of the synthesis process of N-NTO NTAs and A-NTO NTAs.
carbon coating is likely to exacerbate Na+ diffusion resistance, given the fact that traditional carbon materials provide insufficient interlayer spacing (∼0.34 nm) for Na+ transport.5 By contrast, surface engineering offers another effective way to overcome the poor conductivity of Na2Ti3O7.11,18 For example, Fu and co-workers prepared conductive Na2Ti3O7 nanoarrays through a hydrogenation treatment and demon- strated that the reduction of Ti4+ to Ti3+ on the surface is
responsible for the significantly enhanced conductivity.18 This is likely because Ti3+ features a much higher carrier density than Ti4+ in Na2Ti3O7. Unlike the combination of Na2Ti3O7 and carbon materials that requires more complicated operation and often gives rise to nanocomposites with nonuniformity, surface modulation of Na2Ti3O7 by facile thermal treatment in reducing atmosphere is more advantageous to realize homogeneous Ti3+ doping toward better electronic con- ductivity. As a consequence, it is highly preferable to improve the conductivity of Na2Ti3O7 through more effective surface engineering.
In addition to the low electron conductivity, undesired Na+ diffusion resistance is another critical factor that impairs the Na+ storage properties of Na2Ti3O7. Specifically, Na2Ti3O7 features a zigzag layered structure composed of TiO6 octahedra linked by edges.19 It requires much higher energy for Na+ to travel through layers than that along the layers.12 In general, nanostructure engineering is an effective strategy to facilitate the Na+ transport by reducing the diffusion length. To this end, a series of nanostructured Na2Ti3O7 materials, including
nanoparticles,20,21 nanorods,22 nanowires,23−25 nanofibers,14,26 nanotubes,11,18,27−30 nanoribbons,31 and nanosheets,13,32−37 have been intensively studied in recent years. Of note, Na2Ti3O7 nanotubes can not only assure short diffusion
pathway for the transport of Na+ but also bring other advantages associated with available cavities, such as high surface area accessible to the electrolyte, abundant active sites for surface capacitive processes, and plentiful space as a buffer against the volume changes during Na+ intercalation/ deintercalation.11,18,27−30 In addition, in comparison to polycrystalline structures, single-crystalline nanotubes favor rapid Na+ diffusion along with the ordered interlayer channels.27 Notably, aforementioned hydrogenation driven conductive Na2Ti3O7 nanotube arrays have an amorphous shell wrapping the crystalline core. According to Qian et al.’s
findings based on Li4Ti5O12 in LIBs, amorphous surface layer

could cause a diffusion barrier for ionic migration and might induce side reaction with the electrolyte, while removing the amorphous layers by surface reconstruction enables to achieve better rate capability and cycle stability.38 Accordingly, it is more attractive to modulate conductive and clean Na2Ti3O7 surfaces without incurring additional Na+ diffusion resistance. Yet this remains unexplored in SIBs. More importantly, constructing nanoarrays directly grown on current collectors can dramatically accelerate the charge transfer between the substrate and the active material, thereby improving the kinetics of electrochemical reactions. Meanwhile, self-sup- ported electrode configuration usually contributes a lot to the structural robustness toward durable operation. As a result, more rational surface modulation of single-crystalline Na2Ti3O7 nanotube arrays is expected to obtain efficient anodes for superior Na+ storage, but it is challenging to seek a valid synthetic strategy.
With these criteria in mind, herein, we present the tailored synthesis of single-crystalline Na2Ti3O7 nanotube arrays with high conductivity and clean surface through in situ hydro- thermal growth on Ti foil and thermal treatment in a NH3/Ar atmosphere, which is denoted as N-NTO NTAs. Our purpose is to achieve superior Na+ storage properties of the free- standing nanotube array electrode with special emphasis on surface modulation, that is, modifying the surface structure of Na2Ti3O7 via NH3-assisted calcination can simultaneously enhance the electronic conductivity by self-incorporation of Ti3+ and the ionic diffusion kinetics by eliminating amorphous surface layer. The resultant N-NTO NTAs can be directly employed as an anode for SIBs. Impressively, N-NTO NTAs exhibit remarkably higher reversible capacity, better rate capability, and more durable cycling stability compared with almost all reported Na2Ti3O7 based SIB anodes to date. Its overwhelmingly superior Na+ storage performance can be reasonably attributed to the efficient surface modulation, evidenced by comprehensive comparison with the control group free of NH3-assisted treatment. The proposed strategy in this work can open up new opportunities and avenues to develop higher-performance electrode materials for SIBs and related electrochemical devices.
The fabrication process of N-NTO NTAs is schematically illustrated in Figure 1, which involves two main steps. First, Na2Ti3O7 nanotube arrays (denoted as NTO NTAs) were grown in situ on Ti metal foil through its hydrothermal
Figure 2. (a,b) SEM, (c) TEM, (d) HRTEM, (e) HADDF-STEM, and (f−i) element mapping images of N-NTO NTAs. The insets in panels c and d show the corresponding SAED pattern and FFT pattern, respectively.

reaction with sodium hydroXide aqueous solution at 220 °C for 10 h. Subsequently, the obtained NTO NTAs were further calcinated in NH3/Ar atmosphere at 500 °C for 2 h, producing target N-NTO NTAs. In contrast, the same thermal treatment of NTO NTAs in Ar atmosphere resulted in crystalline Na2Ti3O7 nanotube arrays with amorphous layer on the surface. The sample synthesized in pure Ar is named A-NTO NTAs, serving as a control group to unravel the significance of surface modification.
Figure 2a shows the typical scanning electron microscopy (SEM) image of N-NTO NTAs. The resulting N-NTO NTAs consist of plentiful of 1D nanotubes with length up to tens of micrometers, preserving the morphology of the NTO NTAs precursor (Figure S1). Under low magnification (Figure S2), it can be seen that the 1D nanotubes interweave with each other into small bundles, which further interconnect into huge

meshwork on the whole. Such a hierarchical 3D network is beneficial to promote the transport of electrons and sodium ions. Close observation in Figure 2b reveals the open-ended nanotube morphology of N-NTO NTAs with smooth surface and a diameter of ∼50 nm. The hollow nanotube structure of N-NTO NTAs with open end can be further confirmed by the brighter contrast in the center of the transmission electron microscopy (TEM) image of an individual Na2Ti3O7 nanotube (Figure 2c). Corresponding selected area electron diffraction (SAED) pattern shows well-defined diffraction spots of
Na2Ti3O7 crystal along with the [111] zone axis, suggesting the single-crystalline nature of resultant N-NTO NTAs with uniform orientation. In the high-resolution TEM (HRTEM) image of N-NTO NTAs (Figure 2d), the well-resolved lattice fringes with interlayer spacings of 0.35 and 0.68 nm match well with the (01̅1) and (1̅01) planes of Na2Ti3O7. In addition, the
Figure 3. (a, b) SEM, (c) TEM, (d) HRTEM, (e) HADDF-STEM, and (f−h) element mapping images of A-NTO NTAs. The insets in panels c and d show the corresponding SAED pattern and FFT patterns, respectively.

fast Fourier transform (FFT) pattern of the selected pink area confirms the (01̅1) and (1̅01) faces of Na2Ti3O7 single crystal, in line with the SAED result. Also, it is noticed that as- fabricated N-NTO NTAs display clean surface with clear exposure of (1̅01) face, verifying the successful surface modification suppressing the formation of amorphous layers. On the basis of a previous report,38 compared to Na+ transport through an amorphous layer, a clean crystalline surface is advantageous for improving ionic conductivity. Figure 2e displays the high-angle annular dark field scanning TEM (HADDF-STEM) image of N-NTO NTAs. Corresponding energy dispersive X-ray (EDX) element analyses indicate the presence and homogenous distribution of Na, Ti, and O throughout the N-NTO NTAs, while no obvious N element was detected, implying the absence of titanium nitride
impurities on the as-prepared N-NTO NTAs surface.
As shown in Figure 3, the morphology and microstructure of A-NTO NTAs were also carefully characterized using the same electron microscopy techniques. From the SEM images of A- NTO NTAs (Figures 3a and b and S3), it can be seen that A- NTO NTAs exhibit identical network structure assembled by numerous open-ended nanotubes. This is because both A- NTO NTAs and N-NTO NTAs have the same NTO NTAs

precursor. Likewise, the TEM image of the A-NTO NTAs demonstrates the hollow nanotube structure (Figure 3c). The corresponding SAED pattern of A-NTO NTAs shows similar diffraction spots originating from the [111] zone axis of the Na2Ti3O7 single crystal, but the spots deviate from a perfect circle. Such a phenomenon is likely related to the presence of an amorphous layer on the A-NTO NTAs surface, which can be distinguished by the differing contrast between the nanotube wall and the outmost layer. The core−shell structure of A-NTO NTAs is affirmed by its HRTEM image (Figure
3d), where an amorphous shell with a thickness of about 7 nm covers the inner crystalline core. This is further supported by the FFT patterns of related areas as shown in the insets. To be specific, the FFT pattern of selected red area on the edge displays no distinct spots, validating the amorphousness of the shell layer. In contrast, the FFT pattern of the selected pink area is similar to that of N-NTO NTAs, which agrees with the
crystalline core corresponding to the (01̅1) and (1̅01) planes of Na2Ti3O7. This amorphous shell has also been observed in previous reports.18,38 Considering the same synthesis process
was used for A-NTO NTAs and N-NTO NTAs (apart from the calcination atmosphere), it is likely that, in pure Ar, the amorphous shell originates from the pyrolysis of Na2Ti3O7 on

 

Figure 4. (a) XRD patterns, (b) Raman spectra, and (c) Ti 2p and (d) O 1s high-resolution XPS spectra of N-NTO NTAs and A-NTO NTAs.

the surface. Conversely, the presence of NH3 likely suppresses pyrolysis, resulting in the observed clean and highly crystalline surface. Additionally, the HAADF-STEM and element mapping images of A-NTO NTAs also demonstrate the uniform distribution of Na, Ti, and O along the Na2Ti3O7 nanotubes (Figure 3e−h). Since the production of A-NTO NTAs follows the same synthesis procedures as those for N- NTO NTAs, except for the calcination atmosphere, the absence of NH3 should be responsible for the differing surface structures. At the same time, it is safe to conclude that the
assistance of NH3 is crucial to modulating the surface structure of NTO NTAs.
Furthermore, the crystal structures and phase purities of N- NTO NTAs and A-NTO NTAs were compared systematically using X-ray diffraction (XRD) and Raman spectroscopy. In Figure 4a, N-NTO NTAs and A-NTO NTAs show almost the same XRD patterns, in which all the diffraction peaks can be well indexed to monoclinic Na2Ti3O7 (JCPDS Card No. 31- 1329), apart from the background signal from the underlying Ti substrate (Figure S4, JCPDS Card No. 44-1294). In contrast to the very weak (001) peak in the XRD pattern of NTO NTAs (Figure S5), both N-NTO NTAs and A-NTO NTAs exhibit more obvious (001) peaks, showing the enhanced crystallization after thermal treatment. Nevertheless, the relatively stronger peak intensity of N-NTO NTAs discloses its better crystallinity, consistent with the observa- tions in HRTEM images. As depicted in Figure 4b, the Raman
characteristic peaks around 92, 258, 489, and 746 cm−1 for both N-NTO NTAs and A-NTO NTAs can be assigned to the Na−O−Ti vibration,33 confirming their pure Na2Ti3O7 phase. Again, the improved crystallinity of N-NTO NTAs can be reflected from their intensified Raman bands. Meanwhile, no characteristic G (1500−1750 cm−1) and D bands (1280−1350

cm−1) associated with carbon were detected for either N-NTO NTAs or A-NTO NTAs. This rules out the possibility of a carbon coating resulting from the pyrolysis of organic contaminates.
It is well-known that surface chemistry plays a vital role in regulating the physicochemical properties of Na2Ti3O7. As such, X-ray photoelectron spectroscopy (XPS) analysis was carried out to get more insight into the chemical states of N- NTO NTAs and A-NTO NTAs. The survey XPS spectrum of N-NTO NTAs (Figure S6) shows the coexistence of Na, Ti, and O elements without detectable N signal, and no N 1s peak can be found even in its high-resolution XPS spectrum (Figure S7), in accordance with the EDX element mapping results. The survey XPS analysis of A-NTO NTAs displays the same element signals as those of N-NTO NTAs (Figure S8). Interestingly, the corresponding quantification analysis in- dicates that N-NTO NTAs has relative smaller atomic percentage of O (64.75%) compared with that of A-NTO NTAs (66.44%). The reduced amount of O on the surface of N-NTO NTAs is probably resulted from the formation of oXygen vacancies after calcination in NH3/Ar. Thus, their Ti 2p and O 1s high-resolution XPS spectra were examined. As shown in Figure 4c, the Ti 2p spectrum of A-NTO NTAs shows a Ti 2p3/2 peak at 459.0 eV and a Ti 2p1/2 peak at 464.8 eV. The splitting width of 5.8 eV between the Ti 2p1/2 and Ti 2p3/2 core levels confirms a chemical state of Ti4+ in A-NTO NTAs.39,40 By comparison, the Ti 2p1/2 and Ti 2p3/2 peaks of
N-NTO NTAs exhibit a remarkable negative shift by ∼0.8 eV, indicating considerable reduction of Ti4+ into Ti3+ upon NH3- assisted calcination. Hence, both the Ti 2p1/2 and Ti 2p3/2 peaks of N-NTO NTAs can be deconvoluted into two components, whereby the peaks centred at 458.1 and 463.4
eV can be attributed to Ti3+ species.18,20 It is notable that the

 

Figure 5. (a) Rate capability of N-NTO NTAs and A-NTO NTAs. (b) Comparison of rate capability of N-NTO NTAs with some representative Na2Ti3O7-based anodes reported recently. (c) CV curves of N-NTO NTAs and A-NTO NTAs at a scan rate of 0.1 mV s−1. (d) Capacitive contribution to the total capacity at different scan rates for N-NTO NTAs. (e) Cycling performance of N-NTO NTAs and A-NTO NTAs at a high current density of 8.85 A g−1.

fitted peaks in the Ti 2p spectrum of N-NTO NTAs agree well with the difference between the two Ti 2p spectra. This correlation demonstrates the generation of Ti3+ accompanying with presence of abundant oXygen vacancies in NH3-derived N-NTO NTAs.18 Generally, the formation of abundant oXygen vacancies is favorable for increasing the carrier density and improving the conductivity of Na2Ti3O7.18 In the O 1s region (Figure 4d), N-NTO NTAs also display a shift of about 0.55 eV toward lower binding energy because of the partial reduction of Ti4+. The O 1s spectrum of N-NTO NTAs can be fitted into three peaks located around 529.7, 530.1, and
532.1 eV, which are ascribable to O−Ti3+, O−Ti4+, and
adsorbed water, respectively. Correspondingly, the O 1s region
of A-NTO NTAs can be divided into two peaks taking account of the absence of Ti3+ species. On the basis of the above results, it is reasonable to conclude that the modulation of NTO NTAs by NH3-assisted thermal treatment can not only create a highly crystalline and clean surface but also introduce

ample Ti3+ species and oXygen vacancies. This likely proceeds through the following reaction: Na2Ti3O7 + x/3NH3 → Na2Ti3−x(IV)Tix(III)O7−x/2 + x/6N2 + x/2H2O. Conse- quently, the synthetic process is successful in significantly improving the electronic conductivity of a desired anode electrode for SIBs.
The electrochemical performance of N-NTO NTAs and A- NTO NTAs as anodes for SIBs were evaluated by assembling CR2032 coin cells. Their rate capabilities were first investigated via galvanostatic discharge−charge testing in the
voltage range of 0.01−2.5 V versus Na+/Na under different C rates (1 C = 177 mA g−1, Figure 5a). The galvanostatic discharge−charge profiles of N-NTO NTAs and A-NTO
NTAs at different rates are given in Figure S9. As expected, N-
NTO NTAs manifest much higher specific capacities than those of A-NTO NTAs from 1 to 100 C, attesting ultrahigh rate performance. Specifically, N-NTO NTAs deliver a reversible capacity of 237.9, 215.3, 200.1, 179.6, 156, 129.8,
Figure 6. (a) Nyquist plots (scatter) and fitted curves (line) using the equivalent circuit. (b) Simulated charge transfer resistance and calculated diffusion coefficient of Na+. (c) Schematic illustration of the roles of surface modulation for N-NTO NTAs and A-NTO NTAs.

111.1, and 98.2 mAh g−1 at 1, 2, 5, 10, 20, 40, 60, and 80 C.
Even at the current density of 100 C, a high reversible capacity
of 88.5 mAh g−1 is afforded by N-NTO NTAs, whereas A- NTO NTAs achieve only about half the capacity of N-NTO NTAs (40.1 mAh g−1) under the same condition. Moreover, a specific capacity of 234.4 mAh g−1 can be maintained for N- NTO NTAs, when the current density returned to 1 C. Compared with A-NTO NTAs, the strikingly improved rate
capability of N-NTO NTAs can be ascribed to the conductive Ti3+ species for fast electron transport and highly crystalline surface for efficient Na+ diffusion. To the best of our knowledge, the outstanding rate performance of N-NTO
NTAs outperforms almost all the reported Na2Ti3O7 based SIB anodes,11,13,17,18,23,28,29,33,34,41−46 which is highlighted in
Figure 5b.
To have a better understanding of the superior Na+ storage performance of N-NTO NTAs, the electrochemical behavior of N-NTO NTAs and A-NTO NTAs was further investigated by cyclic voltammetry (CV) technique. Figure 5c gives their
CV curves at a scan rate of 0.1 mV s−1 in the potential window of 0.01−2.5 V versus Na+/Na. The CV curve of N-NTO NTAs shows a couple of sharp redoX peaks at 0.388 and 0.572 V, corresponding to the Na+ insertion and extraction in
Na2Ti3O7, respectively. For A-NTO NTAs, the redoX peaks

where k1v and k2v represent to the current contributions from the pseudocapacitance and the Na+ insertion process, respectively, and the capacitive contribution to the total stored charge can be quantitatively calculated by determining k1 and k2 from CV curves at varied scan rates. The CV curves of N- NTO NTAs at various scan rates are presented in Figure S10, and Figure 5d shows the capacitive contribution ratios. With the scan rate increasing from 0.1 to 10 mV s−1, the capacitive contribution improves from 41.2% to 88.0% for N-NTO NTAs. The great capacitive effect of N-NTO NTAs is favorable to fast Na+ storage under high current densities,
thereby contributing to excellent rate performance. Intrigu- ingly, it is found that A-NTO NTAs also reveal very similar surface capacitive effect, showing 38.3% capacitive contribution at the scan rate of 0.1 mV s−1 and 82.5% at the scan rate of 10
mV s−1 (Figure S11). The pseudocapacitance is normally
proportional to the specific surface area.50 Consequently, one plausible explanation for the similar capacitive contribution is that both N-NTO NTAs and A-NTO NTAs preserved the morphology of NTO NTAs precursor, leading to almost the same specific surface area.
Long-term durability and high Coulombic efficiency are of great significance for practical application. In this aspect, N-
NTO NTAs and A-NTO NTAs were cycled at a high current

shift to 0.249 and 0.592 V. Obviously, N-NTO NTAs exhibit
stronger redoX current with much smaller polarization,

density of 8.85 A g−1 (50 C) for 5000 cycles. As shown in
Figure 5e, N-NTO NTAs also exhibit much better cycling

suggesting their favorable reaction kinetics. The Na+ storage reaction generally involves two separate mechanisms, namely, diffusion-controlled Na+ insertion process and surface capacitive behavior.47,48 Since the current i and the sweep rate v obeys the equation49

performance than that of A-NTO NTAs. In the first cycle, N- NTO NTAs achieve a high specific capacity of 127.6 mAh g−1 with an initial Coulombic efficiency of 64.69%. The relatively low initial Coulombic efficiency is usually attributed to the formation of solid electrolyte interphase (SEI) film.27 Next, the

Coulombic efficiency of N-NTO NTAs increase rapidly to nearly 100% in the following several cycles and keep stable subsequently over the prolonged operation. A-NTO NTAs exhibit similar Coulombic efficiency during the cycling test, but its initial Coulombic efficiency is only 52.35% (Figure S12), possibly due to side reactions between the amorphous shell and the electrolyte. After 5000 cycles, N-NTO NTAs still deliver a high reversible capacity of 117.8 mAh g−1, corresponding to a cyclic retention of 92.32% or an average capacity fading of ∼0.0154‰ per cycle. In contrast, A-NTO
NTAs retain only 56.6 mAh g−1 over 5000 cycles with an
obvious capacity attenuation of 30.21%. Undoubtedly, the excellent stability and Coulombic efficiency under extremely high current density further demonstrate the promising application of N-NTO NTAs as a high-performance anode for SIBs.
Finally, electrochemical impedance spectroscopy (EIS) was performed to probe into more convincing direct evidence of surface modulation for improving electronic conductivity and Na+ diffusion. Figure 6a displays the Nyquist plots of N-NTO NTAs and A-NTO NTAs. Apparently, N-NTO NTAs has much smaller impedance than that of A-NTO NTAs. Moreover, the EIS data was fitted employing the equivalent circuit, and the fitted curves are completely consistent with the measurement results. The values of simulated charge transfer resistance (Rct) are given in Figure 6b, where the markedly decreased Rct of N-NTO NTAs (72.6 Ω) compared with that
of A-NTO NTAs (125.2 Ω) sustains that the surface
engineering has a prominent impact on the enhancement of conductivity. On the other hand, the Na+ diffusion coefficient (DNa+) can also be calculated according to the equations20,51
R2T2 DNa+ 2A2n4F 4C2σ2

clean surface and self-doped Ti3+. The as-developed electrode exhibits remarkably superior Na+ storage properties in comparison to those of the counterpart without surface modification and almost all the reported Na2Ti3O7 anodes, as well as most of transition metal related anodes so far, especially in terms of super-high rate capability and ultralong cycling stability. The fascinating SIB performance can be attributed to the effective surface modulation, endowing the electrode with accelerated Na+ diffusion kinetics and improved electronic conductivity. The results presented in this work can serve as a valuable platform to engineer other electrode materials for diverse energy storage devices beyond SIBs.
ASSOCIATED CONTENT
*S Supporting Information
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsmaterial- slett.9b00213.
EXperimental details, SEM images, XRD patterns, XPS spectra, CV curves, capacity contribution ratios, plots of impedance (Z′) versus inverse square root of angular frequency (ω−1/2), and summary tables of recently reported Na2Ti3O7 based anodes and transition metal
related anodes for SIBs (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected].
ORCID
Zhouguang Lu: 0000-0003-3769-9356
Shi-Zhang Qiao: 0000-0002-4568-8422
Notes

The authors declare no competing financial interest.

where R is the gas constant (8.314 J mol−1 K−1), T is the absolute temperature (298.15 K), A is the surface area of the electrode, n is the number of electrons transferred in the redoX half-reaction, F is the Faraday constant (96485 C mol−1), C is the concentration of Na+ (7.45 × 10−3 mol cm−3), and σ represents the Warburg factor that can be determined from the linear relationship between Z′ and ω−1/2 (the reciprocal root square of frequency) in the low frequency region (Figure S13).
The Na+ diffusion coefficient of N-NTO NTAs is calculated to be 1.71 × 10−13 cm2 s−1. This value is much higher than that for A-NTO NTAs (6.34 × 10−15 cm2 s−1), proving the remarkable increase of Na+ diffusion rate after the surface
modification. In light of these electrochemical results and the aforementioned structural information, the considerably improved Na+ storage performance of N-NTO NTAs is largely related to the elaborate surface modulation. As illustrated in Figure 6c, compared to amorphous Na2Ti3O7 surface, crystalline Na2Ti3−x(IV)Tix(III)O7−x/2 can not only act as highway for electrons to expedite charge transfer but also provide regular channels for rapid Na+ diffusion, whereby N- NTO NTAs attain exceptional rate capability and cycling durability that stick out among both recently reported Na2Ti3O7-based anodes (Table S1) and transition-metal-
related anodes (Table S2) for SIBs.
In summary, a combination of hydrothermal growth and NH3-assisted thermal treatment has been developed to construct highly celastrol crystalline Na2Ti3O7 nanotube arrays with

ACKNOWLEDGMENTS
We would like to acknowledge the financial support from the Australian Research Council (ARC) through the Discovery Project and Linkage Project programs (FL170100154, DP140104062, DP160104866, DP170104464, DE150101234,
and LP160100927), Basic Research Project of the Science and Technology Innovation Commission of Shenzhen (No. JCYJ20170412153139454), and the National Natural Science Foundation of China (Nos. 21875097 and 21671096).
REFERENCES
(1) Yabuuchi, N.; Kubota, K.; Dahbi, M.; Komaba, S. Research Development on Sodium-Ion Batteries. Chem. Rev. 2014, 114, 11636−11682.
(2) Kundu, D.; Talaie, E.; Duffort, V.; Nazar, L. F. The Emerging
Chemistry of Sodium Ion Batteries for Electrochemical Energy Storage. Angew. Chem., Int. Ed. 2015, 54, 3431−3448.
(3) Hwang, J.Y.; Myung, S. T.; Sun, Y. K. Sodium-Ion Batteries: Present and Future. Chem. Soc. Rev. 2017, 46, 3529−3614.
(4) Xiang, X.; Zhang, K.; Chen, J. Recent Advances and Prospects of
Cathode Materials for Sodium-Ion Batteries. Adv. Mater. 2015, 27, 5343−5364.
(5) Cao, Y.; Xiao, L.; Sushko, M. L.; Wang, W.; Schwenzer, B.; Xiao,
J.; Nie, Z.; Saraf, L. V.; Yang, Z.; Liu, J. Sodium Ion Insertion in Hollow Carbon Nanowires for Battery Applications. Nano Lett. 2012, 12, 3783−3787.
(6) Liu, J.; Zhang, Y.; Zhang, L.; Xie, F.; Vasileff, A.; Qiao, S. Z.
Graphitic Carbon Nitride (g-C3N4)-Derived N-Rich Graphene with

Tuneable Interlayer Distance as a High-Rate Anode for Sodium-Ion Batteries. Adv. Mater. 2019, 31, 1901261.
(7) Li, L.; Zheng, Y.; Zhang, S.; Yang, J.; Shao, Z.; Guo, Z. Recent Progress on Sodium Ion Batteries: Potential High-Performance Anodes. Energy Environ. Sci. 2018, 11, 2310−2340.
(8) Zhang, H.; Hasa, I.; Passerini, S. Beyond Insertion for Na-Ion
Batteries: Nanostructured Alloying and Conversion Anode Materials.
Adv. Energy Mater. 2018, 8, 1702582.
(9) Mei, Y.; Huang, Y.; Hu, X. Nanostructured Ti-Based Anode Materials for Na-Ion Batteries. J. Mater. Chem. A 2016, 4, 12001− 12013.
(10) Senguttuvan, P.; Rousse, G.; Seznec, V.; Tarascon, J. M.; Palacín, M. R. Na2Ti3O7: Lowest Voltage Ever Reported OXide Insertion Electrode for Sodium Ion Batteries. Chem. Mater. 2011, 23, 4109−4111.
(11) Ni, J.; Fu, S.; Wu, C.; Zhao, Y.; Maier, J.; Yu, Y.; Li, L. Superior
Sodium Storage in Na2Ti3O7 Nanotube Arrays through Surface Engineering. Adv. Energy Mater. 2016, 6, 1502568.
(12) Pan, H.; Lu, X.; Yu, X.; Hu, Y.-S.; Li, H.; Yang, X. Q.; Chen, L. Sodium Storage and Transport Properties in Layered Na2Ti3O7 for Room-Temperature Sodium-Ion Batteries. Adv. Energy Mater. 2013, 3, 1186−1194.
(13) Xie, F.; Zhang, L.; Su, D.; Jaroniec, M.; Qiao, S. Z. Na2Ti3O7@
N-Doped Carbon Hollow Spheres for Sodium-Ion Batteries with EXcellent Rate Performance. Adv. Mater. 2017, 29, 1700989.
(14) Nie, S.; Liu, L.; Li, M.; Liu, J.; Xia, J.; Zhang, Y.; Wang, X. Na2Ti3O7/C Nanofibers for High-Rate and Ultralong-Life Anodes in Sodium-Ion Batteries. ChemElectroChem 2018, 5, 3498−3505.
(15) Xu, J.; Ma, C.; Balasubramanian, M.; Meng, Y. S. Under-
standing Na2Ti3O7 as an Ultra-Low Voltage Anode Material for a Na- Ion Battery. Chem. Commun. 2014, 50, 12564−12567.
(16) Bi, R.; Zeng, C.; Ma, T.; Etogo, A.; Wang, X.; Zhang, L.
Encapsulated Hollow Na2Ti3O7 Spheres in Reduced Graphene OXide Films for Flexible Sodium-Ion Batteries. Electrochim. Acta 2018, 284, 287−293.
(17) Yan, Z.; Liu, L.; Shu, H.; Yang, X.; Wang, H.; Tan, J.; Zhou, Q.;
Huang, Z.; Wang, X. A Tightly Integrated Sodium Titanate-Carbon Composite as an Anode Material for Rechargeable Sodium Ion Batteries. J. Power Sources 2015, 274, 8−14.
(18) Fu, S.; Ni, J.; Xu, Y.; Zhang, Q.; Li, L. Hydrogenation Driven
Conductive Na2Ti3O7 Nanoarrays as Robust Binder-Free Anodes for Sodium-Ion Batteries. Nano Lett. 2016, 16, 4544−4551.
(19) Rousse, G.; Arroyo-de Dompablo, M. E.; Senguttuvan, P.;
Ponrouch, A.; Tarascon, J. M.; Palacín, M. R. Rationalization of Intercalation Potential and RedoX Mechanism for A2Ti3O7 (A = Li, Na). Chem. Mater. 2013, 25, 4946−4956.
(20) Song, T.; Ye, S.; Liu, H.; Wang, Y. G. Self-Doping of Ti3+ into
Na2Ti3O7 Increases both Ion and Electron Conductivity as a High- Performance Anode Material for Sodium-Ion Batteries. J. Alloys Compd. 2018, 767, 820−828.
(21) Zukalova,́M.; Pitňa Laśkova,́B.; Mocek, K.; Zukal, A.; Bousǎ,
M.; Kavan, L. Electrochemical Performance of Sol-Gel-Made Na2Ti3O7 Anode Material for Na-Ion Batteries. J. Solid State Electrochem. 2018, 22, 2545−2552.
(22) Ho, C. K.; Li, C. Y. V.; Chan, K. Y. Scalable Template-Free
Synthesis of Na2Ti3O7/Na2Ti6O13 Nanorods with Composition Tunable for Synergistic Performance in Sodium-Ion Batteries. Ind. Eng. Chem. Res. 2016, 55, 10065−10072.
(23) Li, Z.; Shen, W.; Wang, C.; Xu, Q.; Liu, H.; Wang, Y.; Xia, Y.
Ultra-Long Na2Ti3O7 Nanowires@Carbon Cloth as a Binder-Free Flexible Electrode with a Large Capacity and Long Lifetime for Sodium-Ion Batteries. J. Mater. Chem. A 2016, 4, 17111−17120.
(24) Zhou, Z.; Xiao, H.; Zhang, F.; Zhang, X.; Tang, Y. Solvothermal
Synthesis of Na2Ti3O7 Nanowires Embedded in 3D Graphene Networks as an Anode for High-Performance Sodium-Ion Batteries. Electrochim. Acta 2016, 211, 430−436.
(25) Li, Z.; Ye, S.; Wang, W.; Xu, Q.; Liu, H.; Wang, Y.; Xia, Y. Free-
Standing Sandwich-Structured Flexible Film Electrode Composed of

Na2Ti3O7 Nanowires@CNT and Reduced Graphene OXide for Advanced Sodium-Ion Batteries. ACS Omega 2017, 2, 5726−5736.
(26) Zou, W.; Fan, C.; Li, J. Sodium Titanate/Carbon (Na2Ti3O7/
C) Nanofibers via Electrospinning Technique as the Anode of Sodium-Ion Batteries. Chin. J. Chem. 2017, 35, 79−85.
(27) Zhang, Y.; Guo, L.; Yang, S. Three-Dimensional Spider-Web
Architecture Assembled from Na2Ti3O7 Nanotubes as a High Performance Anode for a Sodium-Ion Battery. Chem. Commun. 2014, 50, 14029−14032.
(28) Liu, Y.; Fang, X.; Zhang, A.; Shen, C.; Liu, Q.; Enaya, H. A.;
Zhou, C. Layered P2-Na2/3[Ni1/3Mn2/3]O2 as High-Voltage Cathode for Sodium-Ion Batteries: The Capacity Decay Mechanism and Al2O3 Surface Modification. Nano Energy 2016, 27, 27−34.
(29) Li, M.; Xiao, X.; Fan, X.; Huang, X.; Liu, Y.; Chen, L. Carbon
Coated Sodium-Titanate Nanotube as an Advanced Intercalation Anode Material for Sodium-Ion Batteries. J. Alloys Compd. 2017, 712, 365−372.
(30) Li, H.; Peng, L.; Zhu, Y.; Chen, D.; Zhang, X.; Yu, G. An
Advanced High-Energy Sodium Ion Full Battery Based on Nano- structured Na2Ti3O7/VOPO4 Layered Materials. Energy Environ. Sci. 2016, 9, 3399−3405.
(31) Dong, S.; Wu, L.; Wang, J.; Nie, P.; Dou, H.; Zhang, X. Self-
Supported Electrodes of Na2Ti3O7 Nanoribbon Array/Graphene Foam and Graphene Foam for Quasi-Solid-State Na-Ion Capacitors. J. Mater. Chem. A 2017, 5, 5806−5812.
(32) Chen, C.; Zhang, N.; Liu, Y.; Wang, Y.; Chen, J. In-Situ
Preparation of Na2Ti3O7 Nanosheets as High-Performance Anodes for Sodium Ion Batteries. Acta Phys. -Chim. Sin. 2016, 32, 349−355.
(33) Anwer, S.; Huang, Y.; liu, J.; Liu, J.; Xu, M.; Wang, Z.; Chen,
R.; Zhang, J.; Wu, F. Nature-Inspired Na2Ti3O7 Nanosheets-Formed Three-Dimensional Microflowers Architecture as a High-Performance Anode Material for Rechargeable Sodium-Ion Batteries. ACS Appl. Mater. Interfaces 2017, 9, 11669−11677.
(34) Ko, J. S.; Doan-Nguyen, V. V. T.; Kim, H.-S.; Muller, G. A.;
Serino, A. C.; Weiss, P. S.; Dunn, B. S. Na2Ti3O7 Nanoplatelets and Nanosheets Derived from a Modified EXfoliation Process for Use as a High-Capacity Sodium-Ion Negative Electrode. ACS Appl. Mater. Interfaces 2017, 9, 1416−1425.
(35) Chen, S.; Gao, L.; Zhang, L.; Yang, X. Mesoporous Na2Ti3O7
Microspheres with Rigid Framework as Anode Materials for High- Performance Sodium Ion Batteries. Ionics 2019, 25, 2211−2219.
(36) Chen, S.; Pang, Y.; Liang, J.; Ding, S. Red Blood Cell-Like
Hollow Carbon Sphere Anchored Ultrathin Na2Ti3O7 Nanosheets as Long Cycling and High Rate-Performance Anodes for Sodium-Ion Batteries. J. Mater. Chem. A 2018, 6, 13164−13170.
(37) Tsiamtsouri, M. A.; Allan, P. K.; Pell, A. J.; Stratford, J. M.;
Kim, G.; Kerber, R. N.; Magusin, P. C. M. M.; Jefferson, D. A.; Grey,
C. P. EXfoliation of Layered Na-Ion Anode Material Na2Ti3O7 for Enhanced Capacity and Cyclability. Chem. Mater. 2018, 30, 1505− 1516.
(38) Qian, K.; Tang, L.; Wagemaker, M.; He, Y. B.; Liu, D.; Li, H.; Shi, R.; Li, B.; Kang, F. A Facile Surface Reconstruction Mechanism toward Better Electrochemical Performance of Li4Ti5O12 in Lithium- Ion Battery. Adv. Sci. 2017, 4, 1700205.
(39) Luo, Y.; Luo, J.; Jiang, J.; Zhou, W.; Yang, H.; Qi, X.; Zhang, H.; Fan, H. J.; Yu, D. Y. W.; Li, C. M.; Yu, T. Seed-Assisted Synthesis of Highly Ordered TiO2@α-Fe2O3 Core/Shell Arrays on Carbon Textiles for Lithium-Ion Battery Applications. Energy Environ. Sci. 2012, 5, 6559−6566.
(40) Liu, J.; Qian, D.; Feng, H.; Li, J.; Jiang, J.; Peng, S.; Liu, Y.
Designed Synthesis of TiO2-Modified Iron OXides on/among Carbon Nanotubes as a Superior Lithium-Ion Storage Material. J. Mater. Chem. A 2014, 2, 11372−11381.
(41) Xie, M.; Wang, K.; Chen, R.; Li, L.; Wu, F. A Facile Route to
Synthesize Sheet-Like Na2Ti3O7 with Improved Sodium Storage Properties. Chem. Res. Chin. Univ. 2015, 31, 443−446.
(42) Zou, W.; Li, J.; Deng, Q.; Xue, J.; Dai, X.; Zhou, A.; Li, J.
Microspherical Na2Ti3O7 Prepared by Spray-Drying Method as

Anode Material for Sodium-Ion Battery. Solid State Ionics 2014, 262, 192−196.
(43) Chen, J.; Zhou, X.; Mei, C.; Xu, J.; Wong, C.-P. Improving the
Sodiation Performance of Na2Ti3O7 through Nb-Doping. Electrochim. Acta 2017, 224, 446−451.
(44) Yan, X.; Sun, D.; Jiang, J.; Yan, W.; Jin, Y. Self-Assembled
Twine-Like Na2Ti3O7 Nanostructure as Advanced Anode for Sodium- Ion Batteries. J. Alloys Compd. 2017, 697, 208−214.
(45) Dong, S.; Shen, L.; Li, H.; Nie, P.; Zhu, Y.; Sheng, Q.; Zhang,
X. Pseudocapacitive Behaviours of Na2Ti3O7@CNT Coaxial Nano- cables for High-Performance Sodium-Ion Capacitors. J. Mater. Chem. A 2015, 3, 21277−21283.
(46) Dong, S.; Li, Z.; Rodríguez-Peŕez, I. A.; Jiang, H.; Lu, J.; Zhang,
X.; Ji, X. A Novel Coronene//Na2Ti3O7 Dual-Ion Battery. Nano Energy 2017, 40, 233−239.
(47) Li, S.; Qiu, J.; Lai, C.; Ling, M.; Zhao, H.; Zhang, S. Surface
Capacitive Contributions: Towards High Rate Anode Materials for Sodium Ion Batteries. Nano Energy 2015, 12, 224−230.
(48) Liu, J.; Wang, J.; Xu, C.; Jiang, H.; Li, C.; Zhang, L.; Lin, J.;
Shen, Z. X. Advanced Energy Storage Devices: Basic Principles, Analytical Methods, and Rational Materials Design. Adv. Sci. 2018, 5, 1700322.
(49) Liu, T. C.; Pell, W. G.; Conway, B. E.; Roberson, S. L. Behavior of Molybdenum Nitrides as Materials for Electrochemical Capacitors: Comparison with Ruthenium OXide. J. Electrochem. Soc. 1998, 145, 1882−1888.
(50) Frackowiak, E.; Beǵuin, F. Carbon Materials for the
Electrochemical Storage of Energy in Capacitors. Carbon 2001, 39, 937−950.
(51) Zhu, H.; Yang, K.; Lan, H.; Qian, S.; Yu, H.; Yan, L.; Long, N.;
Shui, M.; Shu, J. Electrochemical Kinetics of Na2Ti3O7 as Anode Material for Lithium-Ion Batteries. J. Electroanal. Chem. 2017, 788, 203−209.