Facile synthesis of an Al-doped carbon-coated Li₄Ti₅O₁₂ anode for high-rate lithium-ion batteries

Abstract

In this current work, Al-doped carbon-coated Li₄Ti₅O₁₂ (LTAO/C) composites were firstly synthesized via a facile sol–gel method, and subsequently employed as anode materials for high-rate lithium-ion batteries. According to extensive material characterization including X-ray diffraction spectroscopy, high-resolution transmission microscopy, Raman spectroscopy and X-ray photoelectron spectroscopy, LTAO/C composites were successfully not only included with an Al dopant, but also coated with a uniform carbon layer. On the basis of the synergistic effect of the Al-doping and carbon-coating, the resultant LTAO/C electrode displayed superior rate capability and cycling stability. The LTAO/C electrode demonstrates impressive capacity retention of 85.3% at 20C with respect to the discharge capacity at 1C. Additionally, the LTAO/C electrode still retained 97.9% of its initial capacity even after 100 charge/discharge cycles. These results signify that the LTAO/C electrode can be potentially considered as one of the promising anode materials for high-rate lithium-ion batteries.

---

1. Introduction

Lithium-ion batteries (LIBs) have been extensively regarded as one of the candidates with the most potential for large-scale applications in electric vehicles (EV), hybrid electric vehicles (HEV), and energy-storage devices of smart grids due to their features of high energy density and design flexibility.¹ Nevertheless, the low power density, short cycle life and safety issues of the current LIBs should be further addressed to realize their utilization in the aforementioned applications.² The low lithiation potential for the common graphite-based anode materials (approximately <0.2 V vs. Li/Li⁺) generally causes the dendritic lithium plating and the formation of a solid electrolyte interphase (SEI) layer on the electrode, which would result in the safety issues and irreversible capacity of LIBs.³

More recently, spinel lithium titanium oxide (Li₄Ti₅O₁₂, LTO) has been with growing concerns because it exhibits a stable and relatively high operating voltage of about 1.55 V vs. Li/Li⁺. Although such high operating voltage of LTO could reduce the overall cell voltage, it is beneficial to restrict the dendritic lithium growth and the formation of SEI film.^4–7 LTO is generally regarded as a zero-strain material, since its volume change upon lithiation/delithiation process is negligible, and it can be therefore expected to possess excellent cycling stability. These intrinsic strengths of LTO would improve the safety and stability of LIBs and therefore enable them to achieve the basic requirements for use in HEVs or EVs.⁸ In spite of these promising advantages, LTO still suffers from its poor rate capability due to its inherent insulating properties (poor electronic conductivity of ca. 10⁻¹³ S cm⁻¹ and moderate lithium ionic conductivity of 10⁻⁸ cm² s⁻¹).

To date, several strategies have been adopted to address this issue, including reducing LTO particle size, surface modification with conductive materials (Ag, carbon, and TiN),^9–18 and doping with supervalent cations (such as Mg²⁺, Al³⁺, Zr⁴⁺, V⁵⁺, Mo⁴⁺, Cr³⁺, Nb⁵⁺, Ga³⁺, Sc³⁺, and Fe³⁺).^19–28 Recently, Huang et al.²⁹ found that the reversible capacity and cycling performance of LTO were significantly improved by doping Al elements at its Ti sites. The resultant Li₄Ti_4.85Al_0.15O₁₂ displayed the discharge capacity of 195.6 and 173.6 mA h g⁻¹ at 0.15 mA cm⁻² in the first and second cycle, respectively. Our group reported a sol–gel synthesized Li₄Ti_4.95Al_0.05O₁₂ (designated as LTAO) electrode revealed the much higher discharge capacity of 116 mA h g⁻¹ at 5C than that of the pristine LTO (76 mA h g⁻¹).²⁰ Although these significant improvements are impressive, they are still insufficient for applications in power-oriented devices/systems. To further enhance the rate capability of LTO, the combination of LTO with conducting materials, such as Ag, Cu, and carbonaceous materials, is another efficient strategy.^9–14,30 Wang et al.³¹ prepared Li₄Ti_4.95Al_0.05O₁₂/C by using the convention solid-state reaction. The as-prepared Li₄Ti_4.95Al_0.05O₁₂/C electrode was found to be with the improved reversible capacity and cycling performance. However, their study did not investigate the effect of the carbon coating in the improved electrochemical properties for the composite electrode. More importantly, the displayed charge/discharge performance was only conducted at less than 0.2C, suggesting that such kind of Li₄Ti_4.95Al_0.05O₁₂/C electrode material would be possibly restricted for high-power applications.

In this study, we successfully used a facile sol–gel route to synthesize Al-doped carbon-coated LTO (designated as LTAO/C) electrode. The synergistic effects of the Al doping and carbon coating resulted in the LTO with the oxygen vacancy, highly conductive carbon coating and mixed valence of Ti⁴⁺/Ti³⁺. Moreover, the electrochemical characteristics including the rate performance and cycling performance of LTO, Al-doped LTO (LTAO), carbon-coated LTO (LTO/C), and LTAO/C electrodes were systemically investigated in detail.

2. Experimental

2.1 Materials

All chemicals were basically used as revised without any purification. The chemicals included citric acid (99.5%, Acros), lithium acetate dihydrate (CH₃COOLi·2H₂O) (98%, Acros), aluminium acetate dihydrate (CH₃COOAl·2H₂O) (90%, Acros), tetrabutyl titanate [Ti(OC₄H₉)₄] (99%, Acros), ethanol solution (99 wt%, Shimakyu), metallic Li foil, acetylene black (99.99%, Strem Chemicals Inc.) and polyvinylidene fluoride (PVDF, Kynar 7200, ELF), N-methyl-2-pyrrolidone (Ultra, ISP Technologies Inc.), Celgard 2500 separator (HiporeTM, Asahi Kasei Co.), and a formulated battery electrolyte from Novolyte technologies Co. Ltd (1 M LiPF₆ solution in a 1 : 1 (v/v) mixture of ethylene carbonate (EC)/diethyl carbonate (DEC)).

2.2 Preparation of anode materials

Li₄Ti_4.95Al_0.05O₁₂/C (LTAO/C) anode material was synthesized on the basis of our previous study,²⁰ in which citric acid served as a chelating agent and CH₃COOLi·2H₂O, CH₃COOAl·2H₂O and Ti(OC₄H₉)₄ were employed as starting materials. Firstly, CH₃COOLi·2H₂O, CH₃COOAl·2H₂O and Ti(OC₄H₉)₄ were dissolved in ethanol solutions, which were designated as solution A, B and C, respectively. Then, the solution A and B were separately added to the solution C by drop-to-drop titration approach. To obtain a homogeneous mixture solution, the as-prepared mixture solution was continuously stirred at room temperature for 12 h. Subsequently, the citric acid solution was introduced into the mixture solution. After that, the resultant mixture was stirred at room temperature for a half hour, followed by being continuously stirred and heated at 80 °C for 6 h to obtain a gel precursor. The resultant gel precursor was further placed at an oven at 60 °C for 8 h to remove the remaining ethanol within it. After that, it was calcinated at 350 °C for 4 h, and subsequently at 800 °C for 12 h under nitrogen atmosphere. As a result, the LTAO/C composite was attained. LTO/C composite was synthesized via the similar procedure in the absence of CH₃COOAl·2H₂O. LTO and LTAO were synthesized using the similar procedure for the LTO/C and LTAO/C, respectively, except the calcinations were conducted under air atmosphere. It should be noted that the Al-doped LTO with nominal composition of Li₄Ti_4.95Al_0.05O₁₂ demonstrated the superior electrochemical performance in our previous study,²⁰ and therefore the Al-doping amount of 0.05 was employed to replace partial Ti in the spinel LTO in this study.

2.3 Materials characterization

The crystal structures of the synthesized powders were characterized using a Lab XRD 6000 (Shimadzu Corporation) with Cu-Kα radiation from 10° to 80°. A JEOL JSM-7600 field-emission scanning electron microscope (FESEM) equipped with an energy dispersive spectrometer (EDS) was employed for FESEM images and EDS spectra. The particle size distribution of the as-synthesized powders was analysed by means of a particle size analyzer (PSA, LS 13 320). High-resolution transmission electron microscope (HR-TEM) images were taken using a Philips Tecnai G2 operating at 200 kV. An elemental analyzer (EA, Heraeus Vario EL III) was employed to evaluate the carbon content within as-synthesized powders. Raman spectra were recorded by using a micro-Raman system (Horiba Jobin Yvon), in which the laser is with a wavelength of 532 nm and a spot size of 2 μm. X-ray photoelectron spectroscopy (XPS) was performed on VG Scientific ESCALAB 250 spectrometer using Al K_α radiation. The electrochemical charge/discharge properties of the anode materials were evaluated by a series of galvanostatic tests with CR2032 coin cells. The anodes as working electrodes were fabricated by mixing 83 wt% of the as-synthesized powder, 10 wt% of acetylene black and 7 wt% of PVDF binder in NMP solvent. Subsequently, the slurry was stirred at room temperature for 2 h into a homogeneous slurry and then coated on a copper foil current collector with a thickness of ca. 100 μm. After that, the as-prepared anodes were vacuum-dried at 120 °C for 12 h to remove the residual NMP solvent, and then rolled to make the slurry and copper foil more compact. The mass loading of active materials for the LTO-based anodes was ca. 2.5 mg. The coin cell was assembled in an argon-filled glove box with an as-fabricated working electrode (punched into 10 mm in diameter), a metallic Li foil (11 mm in the diameter) as counter electrode, a Celgard 2500 separator, and electrolyte composed of 1 M LiPF₆ in 1 : 1 v/v mixture of EC/DEC. The coin cells were discharged and charged galvanostatically at room temperature within 1.0–2.5 V voltage window at different current densities. A Zahner ennium potentiostat was used for the electrochemical impedance spectroscopy (EIS) analyses. The EIS spectra were measured at a constant voltage of 1.56 V within the frequency range of 10 kHz to 0.01 Hz with a superimposed 5 mV sinusoidal voltage. The resultant EIS spectra were fitted by the software of ZSimpWin V 3.1.

3. Results and discussion

Fig. 1a shows the XRD patterns of LTO, LTO/C, LTAO, LTAO/C samples. All the diffraction peaks are in well accordance with cubic spinel structure (JCPDS card no. 26-1198). This indicates that the modification of LTO with Al-doping, carbon-coating or both of them does not affect the crystal structure of LTO. Additionally, the EA results summarized in Table 1 reveal that the carbon contents of LTO/C and LTAO/C are ca. 1.72 wt% and 1.76 wt%, respectively. Fig. 1b further displays the peak position variation of (111) plane of all samples. Compared to the pristine LTO, it is observed that the (111) peak is shifted toward lower degree for LTO/C sample. On the contrary, the peak is slightly shifted toward higher degree for the LTAO sample. It is worthy noted that the position of the (111) peak for the LTAO/C sample is located between those of the LTO and LTAO samples. The lattice parameters of all samples can be estimated based on the refined results by FullProf, as listed in Table 1. The a-axis lattice parameter of the LTO/C is estimated to be 8.3557 Å, which is slightly larger than that of the pristine LTO (8.3515 Å). The increased lattice parameter for the LTO/C could be accounted for the reduction of partial Ti⁴⁺ in LTO into Ti³⁺ after being sintered under nitrogen atmosphere,³² because the ionic radius of Ti³⁺ (0.67 Å) is relatively larger than that of Ti⁴⁺ (0.61 Å). Moreover, the lattice parameter at a-axis for the LTAO and LTAO/C are estimated to be 8.3403 Å and 8.3466 Å, respectively. It is interesting that both of them are found to be slightly smaller than that of the pristine LTO (8.3525 Å). This could be ascribed to the smaller ionic radius of Al³⁺ ion (0.53 Å) than that of Ti⁴⁺ ion (0.61 Å),³³ suggesting that Al³⁺ ions were successfully substituted for partial Ti⁴⁺ sites. It should be noted that the lattice parameter of the LTAO/C falls in between those of the LTO/C and LTAO samples. This can be explained by the reason that the partial Ti⁴⁺ ions in the LTO would be either transformed into Ti³⁺ ions or replaced by Al³⁺ ions.

Fig. 1 (a) XRD patterns of the pristine LTO, LTO/C, LTAO and LTAO/C samples. (b) Enlarged views of the LTO (111).

Table 1 Physical properties and EIS parameters of different LTO-based samples

Sample	Carbon content^a (wt%)	Lattice parameter^b (Å)	Particle-size distribution^c (μm)	Rs^d (Ω)	Rct^d (Ω)	Rf^d (Ω)	W^d (Ω)
a Carbon content is obtained by EA.b Lattice parameters are obtained by using FullProf.c Particle-size distribution are obtained by LS 13 320 Laser Diffraction Particle Size Analyzer.d Fitted results from EIS measurements.
LTO	N.A.	8.3515	1.15 ± 0.97	7.7	142.7	11.6	92.49
LTO/C	1.72 wt%	8.3557	0.687 ± 0.29	3.1	74.2	9.1	98.12
LTAO	N.A.	8.3403	1.427 ± 1.23	2.9	91.8	18.1	75.22
LTAO/C	1.77 wt%	8.3466	0.761 ± 0.33	2.8	47.1	10.9	86.28

---

Fig. 2 presents the FESEM images of the LTO, LTO/C, LTAO and LTAO/C samples. It can be observed that the particle size is similar for these four samples. The corresponding enlarged FESEM images in Fig. 2a and c further reveal that the well-crystallized spinel structure and smooth surface morphology can be observed in both LTO and LTAO samples. Nevertheless, the LTO/C and LTAO/C samples as depicted in Fig. 2b and d display the crude and reunion surface morphology. Afterward, the secondary particle-size distribution of all samples were shown and summarized in Fig. 3 and Table 1. It can be found that the mean particle size values of the LTO and LTAO samples have larger average particle size than the those of the LTO/C and LTAO/C samples. This can be ascribed to that partial large particles are included in LTO and LTAO samples. Therefore, these results can be ascribed to the formation of amorphous carbon layer from the carbonization of citric acid form on their surface and therefore prevent the LTO particles from agglomeration.³⁴ Additionally, the EDS results shown in Fig. 4 reveal that the presence of Al elements is only found in LTAO and LTAO/C samples, and the average atomic ratio of Li/Ti/Al for both LTAO and LTAO/C is similar and estimated to be 4.000 : 4.953 : 0.047, which is close to the expected stoichiometric ratio. This suggests that Al³⁺ ions are successfully incorporated within the LTAO and LTAO/C samples. This result is in accordance with the observation from the XRD results.

Fig. 2 FESEM images of (a) pristine LTO, (b) LTO/C, (c) LTAO and (d) LTAO/C samples.

Fig. 3 Particle-size distribution of (a) LTO, (b) LTO/C, (c) LTAO and (d) LTAO/C samples.

Fig. 4 EDS spectra of (a) pristine LTO, (b) LTO/C, (c) LTAO, (d) LTAO/C samples.

To further clarify the effects of Al-doping and carbon-coating on the microstructures of LTOs, HR-TEM observations were conducted. As can be seen in Fig. 5, all HR-TEM results show an interplanar distance of around 0.48 nm, which corresponds to the (111) plane of bulk spinel-type LTO. It can be observed that the lattice distance increases in the order of LTAO (0.476 nm) < LTAO/C (0.479 nm) < LTO (0.481 nm) < LTO/C (0.483 nm), which is well consistent with the tendency of the lattice parameters obtained from the XRD results. More importantly, the particles of the LTO/C and LTAO/C samples are surrounded with amorphous carbon layers (Fig. 5b and d), suggesting that the carbon layer can be successfully coated on their surface via the facile sol–gel method. Fig. 6 displays the Raman spectra of the four samples. Two main characteristic peaks at around 1348 cm⁻¹ and 1595 cm⁻¹ corresponding to the D and G band are only observed for the LTO/C and LTAO/C samples while no characteristic peaks are presented for LTO and LTAO samples. The D and G bands can be regarded as the in-plane bond stretching of pairs of sp²-C atoms, and the D mode is associated with defects or lattice distortion.³⁵ The relative intensity of the D band and the G band (I_D/I_G) can be employed to evaluate the degree of disorder or defects in the carbon network.^36–38 This confirms that only LTO/C and LTAO/C samples are with carbon layers. Moreover, the low ratio of I_D/I_G generally signifies the carbon structure with high degree of graphitization, which possesses high electronic conductivity. The I_D/I_G value for the LTO/C and LTAO/C is 0.69 and 0.68, respectively, suggesting the high graphitization degree of the carbon layer on the surface of LTO and LTAO, which can efficiently improve their electronic conductivity.

Fig. 5 HRTEM images of (a) pristine LTO, (b) LTO/C, (c) LTAO, (d) LTAO/C samples.

Fig. 6 Raman spectra of LTO, LTO/C, LTAO and LTAO/C samples.

To further examine the presence of substitutional Al-doping and the partial Ti⁴⁺/Ti³⁺ transformation, XPS analyses of LTO, LTAO, LTO/C and LTAO/C electrodes were conducted. All XPS Ti 2p, Al 2p and O 1 s spectra were referenced based on an adventitious C 1 s excitation. The Ti 2p spectra of LTO and LTAO samples in Fig. 7a and c could be deconvoluted into two characteristic peaks at binding energies of 458.4 and 463.94 eV, which are typical Ti 2p_3/2 peak and Ti 2p_1/2 peak, respectively, corresponding to Ti⁴⁺.^39,40 No notable change in the Ti 2p spectra is observed, signifying no Ti³⁺ formation within the LTO and LTAO samples. As for the LTO/C and LTAO/C samples (Fig. 7b and d), in additional to the two typical characteristic peaks of Ti⁴⁺, two valleys located at 457.3 eV and 462.84 eV are observed, which can correspond to the formation of Ti³⁺ in the LTO/C and LTAO/C samples. This result confirms that the transformation from Ti⁴⁺ to Ti³⁺ occurs during the calcination of LTO/C and LTAO/C in the inert atmosphere and thus results in the mixing Ti⁴⁺/Ti³⁺ valence within them. Fig. 7e–h further displays the Al 2p spectra of the undoped (LTO and LTO/C) and doped (LTAO and LTAO/C) samples. The Al 2p signal only appears at 74.4 eV for the LTAO and LTAO/C, signifying that the Al³⁺ ions are successfully doped in LTAO and LTAO/C and Al³⁺ ion possesses the ionic bonding characteristic.⁴¹ Fig. 7i–l presents the O 1s spectra of all he undoped (LTO and LTO/C) and doped (LTAO and LTAO/C) samples. The peak located at binding energy of 529.9 eV is observed in all samples, which is associated with O²⁻ species in the lattice (O_L) for Ti–O binding. Nevertheless, it should be worth noting that another peak at binding energy of 531.4 eV can be seen in the LTAO and LTAO/C samples, which can be attributed to the formation of oxygen vacancies or defects (O_V) in their microstructures.⁴² The aforementioned results demonstrate that the Al³⁺ ions are successful doped in the LTAO and LTAO/C samples, and also cause the formation of oxygen vacancy within those samples.

Fig. 7 XPS (a–d) Ti 2p, (e–h) Al 2p, and (i–l) O 1s spectra of the pristine LTO, LTO/C, LTAO, and LTAO/C samples.

The electrochemical performance of the LTO, LTO/C, LTAO and LTAO/C electrodes was evaluated using a series of galvanostatic tests at different current densities. Fig. 8 depicts the charging (delithiation of LTO) and discharging (lithiation of LTO) curves of all electrodes at C-rates ranging from 1 to 20C. The charge/discharge plateaus of all electrodes become shorter and the potential gap between the plateaus increase with increasing the C-rates. This polarization phenomenon would slow the reaction kinetic, therefore causing the decrease in discharge capacities. A comparison of the LTO, LTAO and LTO/C electrodes shows that the pristine LTO electrode reveals much higher degree of polarization than those of the LTAO and LTO/C electrodes at a higher charge/discharge current rates. The pristine LTO electrode presents the discharge capacity of 158, 132, 120 and 82 mA h g⁻¹ at current density of 1, 5, 10 and 20C, respectively. Compared with the pristine LTO electrode, the specific discharge capacities of the LTAO and LTO/C electrodes are substantially improved at all charge/discharge current rates from 1C to 20C. For instance, the LTAO electrode exhibits the discharge capacity of 165, 140, 133, and 92 mA h g⁻¹, while the LTO/C electrode delivers 172, 160, 150 and 139.5 mA h g⁻¹ at current density of 1, 5, 10 and 20C, respectively. More interesting, the discharge capacity of the LTAO/C electrode slightly decreases from 174 to 158 mA h g⁻¹ while the current density increases from 1 to 10C, and it still retains as high as ca. 145 mA h g⁻¹ even at high current density of 20C. These results indicate that the LTAO/C electrode possesses the superior rate capability to the LTAO or LTO/C electrodes. As depicted in Fig. 9, the LTAO/C electrode exhibits exceptional rate capability and cycling response to continuously varied current densities while compared with the other three LTO electrodes (LTO, LTAO and LTO/C). For instance, the capacity retention for the LTAO/C electrode at 20C with respect to the discharge capacity at 1C can be achieved up to 85.3%, which is significantly higher than those of other electrodes (48.5% for LTO, 78.8% for LTO/C, and 54.5% for LTAO). Additionally, the discharge capacities of all electrodes return back to their original values without any significant degradation after total 110 cycles, verifying their excellent electrochemical reversibility. The aforementioned XPS results demonstrate that the substitution of Al³⁺ for Ti⁴⁺ sites in LTAO and LTAO/C can cause the existence of oxygen vacancy. It has been reported that the oxygen vacancy could generate two conduction electrons for facilitating electron transfer in electrode materials,^43,44 and reducing ionic diffusivity.⁴⁵ Additionally, the HRTEM and XPS results reveal that the LTO/C and LTAO/C possess a conductive carbon coating layer and the mixed Ti⁴⁺/Ti³⁺ valence. The electrodes with mixed Ti⁴⁺/Ti³⁺ valence is beneficial for enhancing electronic conductivity due to the high electronic conductivity of Ti³⁺.^46,47 To prove this issue, the electronic conductivities of LTO, LTAO, LTO/C, and LTAO/C electrodes were measured by using a four-point probe. The average electronic conductivity of LTO, LTAO, LTO/C, and LTAO/C electrodes is measured to be 1.33 × 10⁻⁷, 7.54 × 10⁻⁷, 2.11 × 10⁻⁶, and 2.58 × 10⁻⁶, respectively. As a result, the excellent rate capability of the LTAO/C electrode can be ascribed to the improved electronic conductivity due to the synergistic effect of the existence of oxygen vacancy, highly conductive carbon coating and the mixed Ti⁴⁺/Ti³⁺ valence.

Fig. 8 The initial charge/discharge curves of (a) pristine LTO, (b) LTO/C, (c) LTAO, and (d) LTAO/C electrodes at different current rates between 1.0 and 2.5 V vs. Li⁺/Li at the room temperature.

Fig. 9 (a) Rate capability of the pristine LTO, LTO/C, LTAO, and LTAO/C electrodes at different current rates. (b) The relationship between the reversible discharge capacity and the current rates for the electrodes.

To further elaborate the enhanced rate capability of the LTAO/C electrode, AC impedance measurements were carried out in the frequency range from 10 mHz to 100 kHz with an AC voltage signal of ±5 mV. As illustrated in Fig. 10, all Nyquist plots are composed of two depressed semicircles in high-to-medium frequency range and an inclined straight line in the low frequency range. The equivalent circuit model used to fit the above Nyquist plots is also depicted in the inset of Fig. 10. The interception of the real axis at high frequency region represents the solution resistance (R_s), which mainly ascribed to the contact resistance from electrolyte and electrode. The semicircle in the high frequency region is assigned to the charge transfer resistance (R_ct) and the constant phase element (CPE) representing double layer capacitance at the electrode/electrolyte interface. The resistance in the low frequency region corresponds to the Warburg impedance (W) associated with the lithium-diffusion process. R_f is the surface polarization resistance related to the particle-to-particle contact resistance, and the C_f is the corresponding surface capacitance.⁴⁸ The fitted results of the EIS data for all electrodes are listed in Table 1. Compared to the LTO electrode, the R_s values of the LTAO, LTO/C and LTAO/C electrodes are observed to be slightly reduced. It can be possibly explained due to the improved electronic conductivity of the LTAO, LTO/C, and LTAO/C electrodes. Moreover, the R_f value is slightly decreased from 11.6 to 9.1 Ω and 18.1 to 10.9 while the LTO and LTAO electrodes are coated with carbon layer, respectively, suggesting the carbon layer could connect LTO particles and thus reduce the surface polarization between them. As compared to the R_ct value of the pristine LTO electrode (147.2 Ω), those for the LTO/C (74.2 Ω) and LTAO (91.8 Ω) electrodes are significantly lower. This finding once again confirms that the Al doping or carbon coating can preserve the high electronic conductivity of the LTAO or LTO/C electrode, thus significantly promoting the electron transport during the lithiation/delithiation reaction and improving the charge-transfer rate at the electrode/electrolyte interface. By taking both advantages of Al doping and carbon coating, the LTAO/C electrode exhibits the lowest R_ct value (47.1 Ω) among all electrodes. Moreover, the LTAO and LTAO/C electrodes have slightly lower W values than those of the LTO and LTO/C electrode. This fact can be explained by the reason that the existence of oxygen vacancy generated by Al-doping can efficiently reduce ionic diffusivity and facilitate the mass diffusion of Li⁺. Therefore, the remarkably improved rate capability of the LTO/C, LTAO and LTAO/C electrodes could be mainly attributed to their significantly reduced R_ct values.

Fig. 10 Nyquist plots of the LTO, LTO/C, LTAO, and LTAO/C electrodes measured after three consecutive charge/discharge cycles. The inset is the equivalent circuit model used to fit the obtained EIS spectra.

The cycling performance of the LTAO/C electrode was carried out under cycling at various high current densities. Fig. 11 displays the capacity retention of the LTAO/C electrode at current rate of 5, 10 and 20C for consecutive 100 cycles. Its capacity retentions after the consecutive 100 cycles at 5, 10 and 20C are 99.4%, 98.6% and 97.9%, respectively. This confirms the highly electrochemical stability of the LTAO/C electrode, even at ultrahigh charge/discharge C rates. A partial list of reports on the electrochemical performance of various cation-doped carbon-coated LTO anode materials is summarized in Table 2. It is noteworthy that the LTAO/C electrode exhibits the superior electrochemical performance including rate capability and discharged capacity compared to the previously reported cation-doped carbon-coated LTO electrodes. Although Wang et al.³¹ reported Al-doped carbon-coated LTO electrode using the conventional solid-state method, they did not reveal the carbon layer was successfully coated on LTO surface. Moreover, they did not investigate the improved electrochemical performance in detail, and the charge/discharge performance was only conducted at less than 0.2C. More importantly, compared to their proposed solid-state synthesis method, our synthesis approach can enable the large-scale production of LTAO/C electrode materials.

Fig. 11 Cycling performance of LTAO/C electrodes at ultrahigh charge/discharge current rates.

Table 2 Comparison of the electrochemical performance between the various cation-doped carbon-coated LTO electrodes

Electrode material	Synthesis method	Electrochemical performance	Reference
Li4Ti4.95Al0.05O12/C	Solid state	0.2C, ca. 158 mA h g^−1	31
Li4Ti4.95Zr0.05O12/C	Solid state	0.2C, ca. 168 mA h g^−1	49
Li4Ti4.99Ru0.01O12/C	Solid state	5C, 120 mA h g^−1	50
		10C, 110 mA h g^−1
Li4Ti4.9V0.1O12/C	Solid state	10C, 118 mA h g^−1	51
		20C, 79 mA h g^−1
LTAO/C (Li4Ti4.95Al0.05O12/C)	Sol–gel	5C, 166.3 mA h g^−1	This work
		10C, 158.0 mA h g^−1
		20C, 145.2 mA h g^−1

---

4. Conclusions

In summary, spinel Al-doped carbon-coated LTO was successfully synthesized by a facile sol–gel method. The formation of oxygen vacancy was found in the LTAO/C sample due to the Al-doping, and can effectively improve its electronic and ionic conductivity. Moreover, the conductive carbon coating layer on the LTAO/C particles can not only improve the interfacial contact between LTAO particles, but also generate the mixed valance of Ti⁴⁺/Ti³⁺, therefore significantly improving its electronic conductivity. On the basis of the synergistic effects of the Al-doping and carbon coating, the resultant LTAO/C electrode can deliver impressive initial reversible discharge capacity up to 174.1, 166.3, 158 and 145.2 mA h g⁻¹ at current rate of 1C, 5C, 10C and 20C, respectively. The LTAO/C electrode displays an excellent capacity retention of 85.3% at 20C with respect to the discharge capacity at 1C. Most importantly, its discharge capacity even retains up to 97.9% of that of the initial cycle after the consecutive 100 cycles at 20C. Therefore, the LTAO/C composite electrode derived from the facile sol–gel route can be considered as one of the potential anode materials for high-rate LIBs.

Acknowledgements

This work is financially supported by Ministry of Science and Technology Taiwan (NSC102-2632-E-036-001-MY3) and Tatung University (B101-C09-025). The authors are also grateful to Prof. She-huang Wu in Tatung University for his helpful discussion and partial supports in materials and instruments. Ms. Su-Jen Ji of Ministry of Science and Technology for the assistance in FESEM experiments is also appreciated by the authors.

References

1. T. A. Stuart and W. Zhu, J. Power Sources, 2011, 196, 458.
2. J. M. Tarascon and M. Armand, Nature, 2001, 414, 359.
3. M. Winter, J. O. Besenhard, M. E. Spahr and P. Novμk, Adv. Mater., 1998, 10, 725.
4. T. Ohzuku, A. Ueda and N. Yamamoto, J. Electrochem. Soc., 1995, 142, 1431.
5. S. Takai, M. Kamata, S. Fujine, K. Yoneda, K. Kanda and T. Esaka, Solid State Ionics, 1999, 123, 165.
6. S. Bach, J. P. Pereira-Ramos and N. Baffier, J. Power Sources, 1999, 81, 273.
7. G. X. Wang, D. H. Bradhurst, S. X. Dou and H. K. Liu, J. Power Sources, 1999, 83, 156.
8. M. M. Thackeray, W. I. F. David, P. G. Bruce and J. B. Goodenough, Mater. Res. Bull., 1983, 18, 461.
9. N. Zhu, W. Liu, M. Q. Xue, Z. Xie, D. Zhao, M. N. Zhang, J. T. Chen and T. B. Cao, Electrochim. Acta, 2010, 55, 5813.
10. J. Zhu, R. Duan, Y. Zhang and J. Zhu, Ceram. Int., 2016, 42, 334.
11. L. Fan, X. Tan, T. Yu and Z. Shi, RSC Adv., 2016, 6, 26406.
12. Y. B. He, F. Ning, B. H. Li, Q. S. Song, W. Lv, H. D. Du, D. Y. Zhai, F. Y. Su, Q. H. Yang and F. Y. Kang, J. Power Sources, 2012, 202, 253.
13. X. F. Guo, C. Y. Wang, M. M. Chen, J. Z. Wang and J. M. Zheng, J. Power Sources, 2012, 214, 107.
14. Y.-C. Kuo and J.-Y. Lin, Electrochim. Acta, 2014, 142, 43.
15. H. Park, T. Song, H. Han and U. Paik, J. Power Sources, 2013, 244, 726.
16. H. Zhang, Q. Deng, C. Mou, Z. Huang, Y. Wang, A. Zhou and J. Li, J. Power Sources, 2013, 239, 538.
17. L. Wang, H. Zhang, Q. Deng, Z. Huang, A. Zhou and J. Li, Electrochim. Acta, 2014, 142, 202.
18. Y. Wang, W. Zou, X. Dai, L. Feng, H. Zhang, A. Zhou and J. Li, Ionics, 2014, 20, 1377.
19. S. Ji, J. Zhang, W. Wang, Y. Huang, Z. Feng, Z. Zhang and Z. Tang, Mater. Chem. Phys., 2010, 123, 510.
20. J. Y. Lin, C. C. Hsu, H. P. Ho and S. H. Wu, Electrochim. Acta, 2013, 87, 126.
21. H. Zhao, Y. Li, Z. Zhu, J. Lin, Z. Tian and R. Wang, Electrochim. Acta, 2008, 53, 7079.
22. X. Li, M. Qu and Z. Yu, J. Alloys Compd., 2009, 487, L12.
23. T. F. Yi, J. Shu, Y. R. Zhu, X. D. Zhu, C. B. Yue, A. N. Zhou and R. S. Zhu, Electrochim. Acta, 2009, 4, 7464.
24. T. F. Yi, Y. Xie, L. J. Jiang, J. Shu, C. B. Yue, A. N. Zhou and M. F. Ye, RSC Adv., 2012, 2, 3541.
25. T. Ohzuku, K. Tatsumi, N. Matoba and K. Sawai, J. Electrochem. Soc., 2000, 147, 3592.
26. T. F. Yi, Y. Xie, J. Shu, Z. H. Wang, C. B. Yue, R. S. Zhu and H. B. Qiao, J. Electrochem. Soc., 2011, 158, A266.
27. F. Li, M. Zeng, J. Li, X. Tong and H. Xu, RSC Adv., 2016, 6, 26902.
28. A. D. Robertson, L. Trevino, H. Tukamoto and J. T. S. Irvine, J. Power Sources, 1999, 81, 352.
29. S. H. Huang, Z. Y. Wen, X. J. Zh and Z. X. Lin, J. Electrochem. Soc., 2005, 152, A186.
30. N. Li, J. Liang, D. Wei, Y.-C. Zhu and Y.-T. Qian, Electrochim. Acta, 2014, 123, 346.
31. Z. Wang, G. Chen, J. Xu, Z. Lv and W. Yang, J. Phys. Chem. Solids, 2011, 72, 773.
32. Y. H. Yin, S. Y. Li, Z. J. Fan, X. L. Ding and S. T. Yang, Mater. Chem. Phys., 2011, 130, 186.
33. J. Huang and Z. Jiang, Electrochim. Acta, 2008, 53, 7756.
34. L. Wang, Z. Zhang, G. Liang, X. Ou and Y. Xu, Powder Technol., 2012, 215, 79.
35. M. A. Pimenta, G. Dresselhaus, M. S. Dresselhaus, L. G. Cancado, A. Jorio and R. Saito, Phys. Chem. Chem. Phys., 2007, 9, 1276.
36. M. M. Doeff, Y. Hu, F. McLarnon and R. Kostecki, Electrochem. Solid-State Lett., 2003, 6, A207.
37. A. Nugroho, K. Y. Chung and J. Kim, J. Phys. Chem. C, 2014, 118, 183.
38. C. Chen, Y. Huang, H. Zhang, X. Wang, G. Li, Y. Wang, L. Jiao and H. Yuan, J. Power Sources, 2015, 278, 693.
39. F. Werfel and O. Brümmer, Phys. Scr., 1983, 28, 92.
40. Q. Zhang, C. Zhang, B. Li, D. Jiang, S. Kanga, X. Li and Y. Wang, Electrochim. Acta, 2013, 107, 139.
41. J. F. Moulder, W. F. Stickle, P. E. Sobol and K. D. Bomben, Handbook of X-Ray Photoelectron Spectroscopy, Physical Electronics Inc, Eden Prairie, MN, USA, 1995.
42. X. Zhang, J. Qin, Y. Xue, P. Yu, B. Zhang, L. Wang and R. Liu, Sci. Rep., 2014, 4, 4596.
43. A. Chen, K. Zhu, H. Zhong, Q. Shao and G. Ge, Sol. Energy Mater. Sol. Cells, 2014, 120, 157.
44. T. Matsuda, D. Nishimoto, K. Takahashi and M. Kimura, Jpn. J. Appl. Phys., 2014, 53, 03CB03.
45. R. Malik, D. Burch, M. Bazant and G. Ceder, Nano Lett., 2010, 10, 4123.
46. J. Wolfenstine and J. L. Allen, J. Power Sources, 2008, 180, 582.
47. T. F. Yi, J. Shu, Y. R. Zhu, X. D. Zhu, R. S. Zhu and A. N. Zhou, J. Power Sources, 2010, 195, 285.
48. J. Peng, Q. Sun, Z. Zhai, J. Yuan, X. Huang, Z. Jin, K. Li, S. Wang, H. Wang and W. Ma, Nanotechnology, 2013, 24, 484010.
49. F. Gu, G. Chen and Z. Wang, J. Solid State Electrochem., 2012, 16, 375.
50. C. Y. Lin, Y. R. Jhan and J. G. Duh, J. Alloys Compd., 2011, 509, 6965.
51. C. C. Yang, H. C. Hu, S. J. Lin and W. C. Chien, J. Power Sources, 2014, 258, 424.

---