TiO₂ Coated with Carbon via Chemical Vapor Deposition as Li-Ion Batteries Anode

Abstract

With the increasing demand for renewable energy and sustainable technologies, lithium-ion batteries (LIBs) have become crucial energy storage components. Despite the promising properties of the high capacity and stability of TiO₂, its large-scale application as an anode for LIBs is hindered by challenges like poor conductivity and volumetric changes during cycling. Here, a rutile TiO₂ composite material with a thinned carbon coating (TiO₂@TC) was synthesized through chemical vapor deposition (CVD) and a subsequent annealing process, which significantly improved the reversibility, cycling stability, and rate performance of the TiO₂ anode materials. The thickness of the carbon layer on TiO₂ was precisely controlled and thinned from 4.2 nm to 1.9 nm after secondary annealing treatment, leading to a smaller steric hindrance and an improved conductivity while serving as protective coatings by preventing the electrochemical degradation of the TiO₂ surface and hindering volumetric changes during cycling. The resulting TiO₂@TC with the thin carbon layer demonstrated a high specific capacity of 167 mAh g⁻¹ at 0.5 C in Li-based half cells, which could stably run for 200 cycles with nearly 100% capacity retention. The thin carbon layer also contributes to an improved rate performance of 90 mAh g⁻¹ at even 20 C. This work provides an innovational strategy for improving the conductivity and volumetric changes during the cycling of TiO₂ anodes.

1. Introduction

With the continuous growth of the global demand for renewable energy and sustainability, lithium-ion batteries (LIBs) have become key components in electronic devices such as electric vehicles and smartphones, mainly due to their significant high energy density and long cycling life [1,2,3]. However, LIBs are now facing increasingly stringent demands, particularly in terms of energy density, safety, and cost-effectiveness. Consequently, researchers have been actively exploring novel anode materials, especially those that possess higher specific capacity and superior stability [4,5,6]. Among the various potential candidates, titanium dioxide (TiO₂) stands out as a particularly promising LIB anode material thanks to its high theoretical specific capacity and low lithium insertion potential [7,8,9]. Moreover, compared to traditional graphite anodes, TiO₂ is less prone to the formation of lithium dendrites, which significantly reduces the risk of internal short circuits. The excellent chemical and thermal stability of TiO₂ further ensures the safety of its applications in LIBs. In addition, the relatively low cost, non toxicity, and abundant resources of TiO₂ make it a remarkable choice for large-scale commercial applications [10,11].

Nowadays, various TiO₂ materials with morphologies like nanowires [12,13], nanotubes [14,15], nanaosheets [16,17], and nanorods [18,19] have been synthesized and applied as anode materials for LIBs. For instance, Dominic Bresser et al. synthesized oleic acid-capped anatase TiO₂ nanorods (NRs) via a low-temperature colloidal route and prepared a highly porous network composed of NRs and carbon. This composite electrode could provide a reversible capacity of 250 mAh g⁻¹. At 0.5/1/2/5/10 C, the reversible capacities of 210/194/165/130 mAh g⁻¹ could be observed, confirming the excellent high rate performance of the well-dispersed NR [20]. Unfortunately, the large-scale feasibility of TiO₂ anodes is still being challenged by the poor rate capability and fast capacity decay resulting from its low electric conductivity and volumetric change during cycling [21,22,23]. To effectively elevate the conductivity and enhance the electrochemical properties of the TiO₂ anode, researchers have decorated titanium dioxide with carbon materials, which could provide abundant growth sites, restrict the aggregation of TiO₂ nanocrystals, and mitigate the variations of volume, promoting the diffusion rate of electrons/ions and improving the rate capability during cycling [24,25,26].

Previous reports mainly focus on the carbon coating of TiO₂ through hydrothermal or ball milling methods, which suffered from insufficient purity, low crystallinity, and non-uniformity [10,27,28]. Moreover, the hydrothermal method typically takes a long reaction time and specific equipment such as high-pressure reactors are required, making it difficult to scale up and achieve large-scale industrial production [28]. Compared with hydrothermal and ball milling processes, the chemical vapor deposition (CVD) is capable of depositing TiO₂ at high temperatures to produce high-purity and highly crystalline materials, which can be applied to synthesize different TiO₂ with various structures and morphologies, such as thin films, nanowires, and nanotubes. Furthermore, the gas-phase reaction during its deposition process can ensure the uniformity of TiO₂ thin films or particles [29,30,31]. However, there have been few reports that exploit the merit of CVD carbon-coated TiO₂ for anode material to date.

Herein, we report a rutile TiO₂ composite anode material with a thinned carbon coating (TiO₂@TC) synthesized through CVD and a subsequent annealing process, which possesses an improved reversibility, cycling stability, and rate performance. By using secondary annealing treatment after the CVD process, the thickness of the carbon layer on TiO₂ was precisely controlled and thinned from 4.2 nm to 1.9 nm. The resulting thin carbon coating enables the TiO₂ anode to obtain an improved conductivity while protecting the TiO₂ anode from electrochemical degradation and hindering volumetric changes during cycling. Subsequently, the obtained TiO₂@TC delivers a high specific capacity of 167 mAh g⁻¹ at 0.5 C in lithium-ion half cells while maintaining full capacity retention after 200 cycles. At a high current density of 20 C, the TiO₂@TC anode exhibits an improved rate capability of 90 mAh g⁻¹. Significantly, this work provides an innovational strategy for improving the performance of TiO₂ anodes and paves a way towards high performance Li-ion batteries.

2. Experimental Section

2.1. Materials and Characterization Methods

All reagents commercially available were used as received. For example, titanium dioxide powder (TiO₂) was from Shanghai Aladdin Bio-Chem Technology Co., Ltd. (Shanghai, China). The carbon additive of carbon black was provided by Lion Co., Ltd. (Tokyo, Japan). Polyvinylidene difluoride binder (PVDF), lithium hexafluorophosphate (LiPF₆), and ethylene carbonate/diethyl carbonate/dimethyl carbonate (EC/DEC/DMC) were purchased from DoDo-Chem (Shanghai, China). The Li foils (>99%) were purchased from Canrd New Energy Technology Co., Ltd. (Guangzhou, China). The X-ray diffraction (XRD) spectra were performed on X’Pert PRO MPD (Analytical, Breda, The Netherlands). The Raman results were measured by LabRam HR Evolution (Keoda, Chengdu, China). The morphology images of the samples were collected by the field-emission scanning electron microscope (FE-SEM, Hitachi, Tokyo, Japan, S3400N). The transmission electron microscopy (TEM) images were collected by FEI Tecnai F20 (Hillsboro, OR, USA). The cyclic voltammetry (CV) test was measured by a CHI instrument electrochemical workstation (650E, Austin, TX, USA). And the electrochemical impedance spectroscopy (EIS) was also performed on a CHI instrument electrochemical workstation (650E) in the frequency of 0.01 Hz–100 kHz, with 5 mV amplitude. All cells (CR2032) were assembled in the Ar-filled glove box. The galvanostatic charge–discharge curves were collected by a CT2001A cell test instrument (LAND Electronic Co., Wuhan, China) under room temperature. All the half and dual-ion batteries were tested at room temperature.

2.2. Synthesis of TiO₂@C and TiO₂@TC

Initially, TiO₂ powder was evenly spread in a quartz boat and placed inside a CVD furnace. Under the protection of argon gas flowing at 40 mL/min, the temperature was raised to 600 °C. Upon reaching this temperature, acetylene was introduced while adjusting the argon flow rate to 60 mL/min for acetylene and 20 mL/min for argon. The deposition process lasted for 10 min. Afterward, acetylene was shut off, and the sample was maintained at 600 °C for 30 min with an argon flow of 40 mL/min before cooling to room temperature. This resulted in the formation of TiO₂@C nanoparticles, specifically the sample with a thicker carbon layer mentioned in the article.

Subsequently, the obtained TiO₂@C sample was laid flat in a quartz boat and placed in a muffle furnace. Under argon protection, the temperature was raised to 800 °C and maintained for two hours, yielding the TiO₂@TC. This process involved annealing and the graphitization of the carbon layer.

2.3. Half Cells Fabrication

The TiO₂@C and TiO₂@TC anodes were simply fabricated by 80 wt% active materials, 10 wt% carbon black, and 10 wt% polyvinylidene difluoride binder (PVDF). The neat loading mass of active materials on Cu foil was >12 mg cm⁻². The electrolyte for Li-ion half cells was 1M LiPF₆ in ethylene carbonate/diethyl carbonate/dimethyl carbonate (EC/DEC/DMC). The separator was Whatman glass fiber. The anode was Li metals. All half cells (CR2032) were assembled in the Ar-filled glove box. The CV test was tested at a scan rate of 0.2 mV s⁻¹ between 1.0 and 3.0 V (vs. Li⁺/Li) using an electrochemical workstation (CHI 650e). All the capacities were reported based on the mass of active materials in half cells.

3. Results

3.1. TiO₂@TC Design and Characterizations

Figure 1 demonstrates the mechanism of TiO₂, TiO₂@C, and TiO₂@TC during electrochemical reactions. As shown in Figure 1a, the low conductivity of pure TiO₂ leads to the low utilization of the active material. In contrast, after being coated with a graphitic carbon layer (Figure 1b) by the CVD process, the carbon coating of the obtained TiO₂@C uniformly wrapped on the surface of TiO₂ facilitates electron transfer, promotes the insertion of Li-ions during electrochemical reactions, and enhances the electrochemical reversibility and cycling stability of the material. After secondary annealing treatment of TiO₂@C, the carbon layer is significantly thinned and the resulting TiO₂@TC possesses a smaller steric hindrance compared with TiO₂@C, further promoting the insertion of Li-ions and enabling an improved conductivity while hindering volumetric changes during cycling. Therefore, a rutile phase TiO₂ composite material with a carbon coating was synthesized using CVD (TiO₂@C). And TiO₂ with a thin carbon layer (TiO₂@TC) was obtained through secondary annealing. See synthesis details in Supplementary Materials.

Subsequently, the obtained TiO₂@C and TiO₂@TC were collected for structural characterizations. Firstly, the X-ray diffraction (XRD) patterns were carried out. As provided in Figure 2a, both carbon-coated (TiO₂@C) and annealed carbon-coated (TiO₂@TC) TiO₂ samples exhibited complete alignment with the peaks of pure TiO₂, suggesting that the crystal structure of TiO₂ was effectively protected during CVD and the annealing process. Figure 2b demonstrated the fine XRD spectra of the (110) and (101 peaks), indicating no observable changes for TiO₂, TiO₂@C, and TiO₂@TC. This stability ensures long-term efficacy in practical applications, where a stable crystal structure contributes to an enhanced reversibility and cycling stability during electrochemical reactions. Moreover, rutile-phased TiO₂ could be identified and confirmed by indexing all the presenting eminent diffraction peaks to standard rutile TiO₂ (JCPDS No. 21-2176), further validating the phase purity and crystallinity of the samples. Remarkably, there existed no obvious diffraction peaks of the carbon layer (two theta < 20°) (Figure 2a), which could be explained by ascribing the vanished carbon diffraction peaks to the amorphous structure or slight content of the carbon [29]. The following Raman spectra, SEM images, and TEM images also confirmed the existence of a uniform distributed carbon coating on TiO₂@C and TiO₂@TC.

Afterwards, the Raman spectroscopy was performed in order to further investigate the structure of TiO₂@C and TiO₂@TC. The resulting Raman spectra of TiO₂, TiO₂@C, and TiO₂@TC are demonstrated in Figure 3a. The two characteristic peaks located at 1350 cm⁻¹ and 1580 cm⁻¹ in the Raman spectra corresponded to the D and G bands of disordered carbon and graphitized carbon, respectively [32]. On the other hand, the presence of another double maximum at 440.4 cm⁻¹ and 611.8 cm⁻¹ could be observed as features of rutile-phased TiO₂, assigned to the E_g and A_1g bands, respectively [27]. Nonetheless, the intensity of the E_g and A_1g bands significantly decreases compared with the pure TiO₂, which was almost undetectable. And this phenomenon might result from the shielding effect of the outer carbon layer. Therefore, the coexistence of the E_g, A_1g, D, and G bands in the Raman spectra of TiO₂@C and TiO₂@TC solidified the evidence of successful carbon coating on the pristine rutile-phased TiO₂ nanoparticle. Furthermore, the degree of carbon disorder could be measured by the intensity ratio of the D band over G band (I_D/I_G). As depicted in Figure 3b, the I_D/I_G values of TiO₂@C and TiO₂@TC were 1.63 and 1.35, indicating a lower degree of carbon disorder and higher content of graphitized carbon in TiO₂@TC [32,33,34].

In addition, the morphology of TiO₂@C and TiO₂@TC was demonstrated by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Figure 4a–c, the SEM image of TiO₂@C revealed an average particle size of 41.67 nm and a uniform coating of carbon black over TiO₂ nanoparticles. And TiO₂@TC exhibited a similar average particle size of 41.32 nm with a uniform coating of carbon black as well over TiO₂ nanoparticles. All three materials exhibited a similar morphology of grain shape, revealing the maintained crystal structure of TiO₂ during CVD and the annealing process. In the TEM images (Figure 4d,e), TiO₂@C showed the lattice fringes of TiO₂ nanoparticles with a basal distance of 0.24 and 0.35 nm, which is consistent with the (101) and (110) lattice spacing of the rutile TiO₂ phase and in parallel with the above XRD result [32,35]. And the thickness of the carbon layer was 4.2 nm. On the other hand, TiO₂@TC with a smaller thickness of only 1.9 nm also exhibited the lattice fringes of TiO₂ nanoparticles in Figure 4f,g. By comparison, there existed a decrease in carbon coating thickness ranging from 0.15 nm to 0.31 nm, which could be attributed to the annealing process involving carbon volatilization, densification, and interfacial reactions [36], enabling an improved conductivity with a smaller steric hindrance and facilitating the insertion of Li-ions during electrochemical reactions. The lattice spacing in sync with the rutile phase of TiO₂ further supported the above XRD results. Based on the above characterizations, we concluded that the TiO₂@TC and TiO₂@C were successfully synthesized.

3.2. Electrochemical Performance

The electrochemical behaviors of TiO₂@TC were initially evaluated in cyclic voltammetry (CV) for Li-based half cells. The neat loading mass of TiO₂@TC was 80 wt% and >12 mg cm⁻² in the electrode composition on the Cu foil. Notably, TiO₂@TC was used as received and no other technical processing was carried out or hidden. Meanwhile, the electrolytes were selected to be 1 M LiPF₆ in ethylene carbonate/diethyl carbonate/dimethyl carbonate (EC/DEC/DMC), with a volume ratio (v/v) of 1:1:1. The anodes were Li metals. See the half-cell fabrication details in Supplementary Materials. The collected CV results are shown in Figure 5a. It was noticeable that the first cycle was different from the subsequent four cycles, which could be attributed to the manifestation of the side reactions taking place on the electrode surfaces and interfaces, the solid electrolyte interface (SEI) formation, as well as the insertion/intercalation instability in the initial stage. Afterwards, the curves of the 2–5th cycles were almost coincident, where the redox peaks could stabilize at 2.1 V and 1.65 V (vs. Li). The reduction peak at 1.65 V was attributed to the Li-ion insertion/intercalation to TiO₂ and/or the coated carbon layer associated with the reduction of Ti⁴⁺ to Ti³⁺. On the other hand, the oxidation peak at 2.1 V belonged to the Li-ion desertion/de-intercalation from TiO₂ and/or coated carbon layer associated with the oxidation of Ti³⁺ to Ti⁴⁺. Remarkably, the well-overlapped CV curves indicated the highly reversible and highly stable electrochemical reaction of TiO₂@TC in half cells.

Subsequently, the whole anode performances of TiO₂@TC in Li-based half cells were tested. TiO₂@TC exhibited one clear charge–discharge potential platform that corresponded to the above CV results at 0.5 C with the discharge capacity of 167/152/140/125/110/90 mAh g⁻¹ at 0.5/1/2/5/10/20 C, respectively (Figure 5b). And the charge–discharge curve at 0.5 C was chosen to study the charge–discharge plateau’s potential difference. As demonstrated in Figure 5c, the charge–discharge potential platform ascribed to TiO₂@TC displayed a small potential difference of 0.2 V. Meanwhile, the cycle profiles of TiO₂, TiO₂@C, and TiO₂@TC in Li-based half cells are provided in Figure 5d, of which the TiO₂ exhibits a fast capacity decay from 275 mAh g⁻¹ to 82 mAh g⁻¹, with only 30% retentions after only ten cycles. In contrast, the TiO₂ coated with carbon layers exhibited a much slower capacity decay of 77% and 69% for TiO₂@C and TiO₂@TC, respectively. These results indicated that the carbon coatings generated by the CVD approach significantly hinder the volumetric changes of TiO₂ during the first few cycles and improve the electrochemical stability for the TiO₂ anode. After stabilization, each of them exhibited an extraordinary ability in keeping an almost identical reversible capacity even after 200 cycles, where the TiO₂@TC demonstrated the highest specific capacity compared with other two TiO₂-based anodes. Such enhancement could be assigned to how the protective carbon coatings blocked some part of the TiO₂ surface from ion exchange with the electrolyte and protected the whole anode structure from further electrochemical degradation, leading to a lower specific capacity compared with the theoretical specific capacity (~335 mAh g⁻¹) but a much more stable cycle stability for the TiO₂@TC anode. The rate performances of TiO₂, TiO₂@C, and TiO₂@TC in Li-based half cells are performed in Figure 5e. At 0.5 C, 1 C, 2 C, 5 C, 10 C, and 20 C, TiO₂@TC demonstrated the satisfactory rate capacities of 167 mAh g⁻¹, 152 mAh g⁻¹, 140 mAh g⁻¹, 125 mAh g⁻¹, 110 mAh g⁻¹, and 90 mAh g⁻¹, corresponding to the above charge–discharge curves in Figure 5b, respectively. And the capacity nearly fully recovered when the current rate was set back to 0.5 C, revealing the high reversibility of the TiO₂@TC anode. It was noticeable that the initial Coulomb efficiency of all three anode materials was relatively low and a steep decline in capacity could be observed, which might result from the inevitable formation of a solid electrolyte interface (SEI) and electrolyte deposition during the first few cycles.

In addition, to investigate how varying carbon layer thicknesses influence the enhancement of electrical conductivity in TiO₂ anode materials, electrochemical impedance spectroscopy (EIS) measurements were conducted for the TiO₂, TiO₂@C, and TiO₂@TC anodes, which presented significant differences in their impedance characteristics (Figure 5f). The inset displayed the equivalent circuit model. The sloped line indicated the diffusion resistance of Li-ions within the electrolyte, known as the Warburg impedance (Z_w). And the charge–transfer resistance (R_ct) reflected the charge transfer at the electrode/electrolyte boundary and corresponded to the diameter of the semicircle. The ohmic resistance (R_e) represented the Li-ion diffusion resistance in the surface of the anode layer and appeared as the initial point of the semicircle. As shown in Figure 5f and Figure S1, TiO₂@C and TiO₂@TC exhibited a dramatically lower semicircle diameter (R_ct) compared to TiO₂, with a notable gap of approximately 200 versus 7000 Ohms. In terms of electrochemical performance, the TiO₂@C and TiO₂@TC anodes demonstrated a remarkable enhancement over pristine TiO₂, which showed that the incorporation of carbon coating onto the TiO₂ surface significantly reduced the overall impedance and enabled an improved charge transfer efficiency and a faster electron transport within the material. As provided in Table S1, the TiO₂@TC could rival the TiO₂@C anodes by different carbon coat techniques.

4. Conclusions

In summary, the simple chemical vapor deposition method for synthesizing carbon-coated rutile TiO₂ has successfully overcome the challenges posed by poor conductivity and volumetric changes during the cycling of TiO₂-based anode materials, exhibiting exceptional reversibility, cycling stability, and rate capability. The thickness of the carbon layer on TiO₂ was precisely controlled through secondary annealing treatment, and the realized carbon layer protects the TiO₂ anode from further electrochemical degradation while improving conductivity and hindering volumetric changes during cycling. The optimized TiO₂@TC composite with a thin carbon layer achieved a noteworthy specific capacity of 167 mAh g⁻¹ at 0.5 C in Li-based half cells, stably running for 200 cycles with nearly 100% capacity retention. The thin carbon layer enabled TiO₂@TC with an improved rate performance of 90 mAh g⁻¹ at even 20 C. Significantly, this work introduces an innovational strategy for addressing the low conductivity and volumetric changes during the cycling of TiO₂ anodes and opens opportunities towards high performance Li-ion batteries.