A unique dual-shell encapsulated structure design achieves stable and high-rate lithium storage of Si@a-TiO₂@a-C anode

Abstract

Due to high theoretical capacity and low lithium-storage potential, silicon (Si)-based anode materials are considered as one kind of the most promising options for lithium-ion batteries. However, their practical applications are still limited because of significant volume expansion and poor conductivity during cycling. In this study, we prepared a double core‒shell nanostructure through coating commercial Si nanoparticles with both amorphous titanium dioxide (a-TiO₂) and amorphous carbon (a-C) via a facile sol‒gel method combined with chemical vapor deposition. Elastic behaviors of a-TiO₂ shells allowed for the release of strain, maintaining the integrity of Si cores during charge‒discharge processes. Additionally, outer layers of a-C provided numerous pore channels facilitating the transport of both Li⁺ ions and electrons. Using the distribution of relaxation time analysis, we provided a precise kinetic explanation for the observed electrochemical behaviors. Furthermore, the structural evolution of the anode was explored during cycling processes. The Si@a-TiO₂@a-C-6 anode was revealed to exhibit excellent electrochemical properties, achieving a capacity retention rate of 86.7% (877.1 mA·h·g⁻¹ after 500 cycles at a 1 A·g⁻¹). This result offers valuable insights for the design of high-performance and cyclically stable Si-based anode materials.

Graphical abstract

Keywords

lithium-ion battery / Si anode / distribution of relaxation time analysis / dual-shell encapsulated structure / high-rate lithium storage

1 Introduction

The development of advanced lithium-ion batteries (LIBs) with higher energy density and longer lifespan is crucial to meet requirements of next-generation portable devices and electric vehicles [1–3]. However, the specific capacity of commonly used commercial graphite anode materials for LIBs has nearly reached the theoretical value of 372 mA·h·g⁻¹ [4–5]. Silicon (Si) anodes are considered highly promising candidates for LIBs due to their high theoretical capacity (3579 mA·h·g⁻¹ corresponding to Li₁₅Si₄ at room temperature), natural abundance, and relatively low operating potential [6–7]. Nevertheless, significant challenges associated with Si anodes include considerable volume expansion (> 300%) during the lithiation process, continuous side reactions with the electrolyte leading to the partial silicon failure, and inadequate electrical conductivity, which significantly reduce the cycle life of Si-based anodes [8–13].

Reasonable designs of silicon nanostructures with diverse morphologies and modifications of Si using a conductive substrate are commonly employed to enhance the cyclic stability of Si anodes. Various techniques for nanosizing Si nanostructures, such as Si nanocrystals [14], Si nanotubes [15], Si nanowires [16], Si nanosheets [17], and nanoporous Si [18], have been utilized to mitigate the volume expansion effect during cycling of Si anode-based batteries. However, these manufacturing methods often possess certain drawbacks associated with toxic reagents, hazardous etching reactions, high-temperature chemical processes, etc. [19–20]. Another effective approach to suppressing the volume expansion of Si anodes is through constructing core‒shell or yolk‒shell structures. A surface coating material, composed of carbon [21–24], Al₂O₃ [25], Co₃O₄ [26], TiO₂ [27–28], etc., is applied onto the surface of the Si anode, which not only reduces the volume change of the Si anode during cycling, but also prevents the direct contact between the Si anode and the electrolyte, thereby inhibiting the continuous electrolyte decomposition and significantly improving the cycle stability. Among these coatings, the TiO₂ coating exhibits the minimal volume expansion (less than 4%) during the processing of lithium. Meanwhile, the lithiated TiO₂ layer enhances safety due to its thermal stability while suppressing thermal effects between the highly lithiated Si phase and the electrolyte solution [29–31]. Recently, employing flexible amorphous TiO₂ (a-TiO₂) as a shell for rigid Si has been demonstrated an effective method in restraining the volume expansion and preventing the crushing of Si anodes [32–34]. However, modifying the Si anode solely with TiO₂ poses challenges in the enhancement of both ion transport and electron conduction. The carbon coating strategy has been demonstrated effective in improving ion transport and electron conduction characteristics of electrode materials [35–37]. Therefore, to ensure structural integrity and excellent electrical conductivity throughout the cycling process, a dual-coating strategy involving amorphous carbon (a-C) coating and a-TiO₂ coating is necessary for collaborative enhancement.

In this work, a core‒shell nanostructured material was used as the anode electrode, which comprised nanoparticles (NPs) of a-TiO₂ and a-C dual-shell encapsulated commercial silicon (hereafter denoted as Si@a-TiO₂@a-C) synthesized through the sol‒gel method combined with chemical vapor deposition (CVD). The a-TiO₂ inner encapsulation layer exhibited elasticity with strain relief behaviors, ensuring the integrity of the Si core during the charge‒discharge process. The presence of the a-C outer encapsulation layer provided abundant pore channels that facilitated the transfer of both Li⁺ ions and electrons. Furthermore, achieving accurate time-scale modeling of cells is of paramount importance for facilitating kinetic interpretation. Nevertheless, the coupling of electrochemical impedance spectroscopy (EIS) responses presents a challenge in the construction of equivalent circuits. To ensure the precise kinetic interpretation, we employed the distribution of relaxation time (DRT) analysis to investigate the electrochemical process. Additionally, transmission electron microscopy (TEM) was performed to observe the evolution of the dual-shell encapsulated structure during the initial stage of lithiation and that after 500 cycles. We believe that this stable dual-shell encapsulated layer structure makes the Si@a-TiO₂@a-C composite a promising anode material for LIBs.

2 Experimental

2.1 Synthesis of Si@a-TiO₂@a-C composites

A layer of a-TiO₂ was deposited on the surface of commercial silicon nanoparticles (SiNPs) via an improved kinetics-controlled sol‒gel coating method. In a typical process, 0.5 g of commercially available SiNPs with an approximate diameter of 100 nm were added into a mixed solution comprising 500 mL of anhydrous ethanol and 1.5 mL of the ammonia aqueous solution with a weight proportion of 28%. The mixture became a uniformly dispersed solution through ultrasound treatment for 1 h followed by stirring at 40 °C for another hour. After different volumes (4, 6, and 8 mL) of titanium tetraisopropanolate were added dropwise, the solution was stirred gently (200 r·min⁻¹) at 40 °C for 48 h. Subsequently, a-TiO₂-coated silicon (Si@a-TiO₂) core‒shell NPs were obtained through a series of centrifugation and ethanol washing steps followed by drying at 80 °C for 10 h. A layer of a-C was then deposited onto the surface of Si@a-TiO₂ via a CVD coating approach. In a typical process, Si@a-TiO₂ was loaded into a rotary quartz tube reactor and heated to 550 °C in an Ar atmosphere. Finally, the carbon source (C₃H₆) with a flow rate of 0.4 L·min⁻¹ was introduced into the reactor, with the objective of depositing carbon coating layers on Si@a-TiO₂. The product, Si@a-TiO₂@a-C, was thus obtained after a reaction time of 10 min.

2.2 Material characterizations

The morphology and elemental mappings were examined using a Hitachi SU8010 scanning electron microscope equipped with an energy dispersive X-ray spectroscopy (EDS) system. The ultrafine structure and lattice fringes were investigated through TEM on a transmission electron microscope (Tecnai G2 F20). For TEM measurements, the samples were suspended in ethanol and dried on a holey carbon support film located on a Cu grid. X-ray diffraction (XRD) patterns were obtained using a Bruker D8 ADVANCE X-ray diffractometer (Germany). Raman spectra were measured at ambient temperature, and signals were recorded using an Xplora Plus Raman spectrometer with 532 nm laser excitation (Horiba Scientific). The chemical state and elemental composition of the sample surface were analyzed through X-ray photoelectron spectroscopy (XPS) using a Thermo Fisher K-α spectrometer with an Al Kα X-ray source. The specific surface area (SSA) and pore size distribution (PSD) of synthesized samples were determined during the N₂ isothermal adsorption‒desorption process at 77 K using an ASAP 2020 instrument.

2.3 Electrochemical measurements

The anode was obtained through mixing Si@a-TiO₂@a-C, conductive carbon black, and polyantimonic acid (PAA)-Li at a mass ratio of 8:1:1 with H₂O. The electrode films were prepared through pasting the slurry onto the copper foil with the thickness reaching 150 μm followed by drying in a vacuum oven at 80 °C overnight. The working electrodes were then prepared through punching the electrode film into discs 1.2 cm in diameter. To prepare the electrolyte, 1 mol·L⁻¹ LiPF₆ was dissolved in ethylene carbonate (EC) and diethyl carbonate (DEC) at a volume ratio of 1:1 with 5% fluoroethylene carbonate (FEC) additive. The assembled batteries were aged for 24 h before testing. Cyclic voltammetry (CV) was performed using a CHI760E electrochemical workstation from 0.01 to 3.0 V at a sweep rate of 0.01 mV·s⁻¹. EIS was conducted in the frequency range from 100 kHz to 0.01 Hz. DRT data were analyzed via the calculation from impedance data using open-source MATLAB script-based software (DRT Tools).

3 Results and discussion

The synthesis process of Si@a-TiO₂@a-C microspheres is illustrated in Fig.1(a). Initially, a-TiO₂ was deposited onto commercial SiNPs in an alkaline ethanol solution using titanium tetraisopropanolate as a precursor. The kinetically controlled sol‒gel coating method was employed to facilitate the formation of a core‒shell structure. Furthermore, the hydrolysis rate of titanium tetraisopropanolate in ethanol solution was significantly reduced, ensuring the uniform growth of a-TiO₂ [38]. Subsequently, a layer of a-C was deposited on the surface of the a-TiO₂ shell layer via the CVD method using propylene as a carbon source in a uniformly rotating quartz tube reactor. Scanning electron microscopy (SEM) images reveal that the main size of SiNPs is approximately 100 nm with smooth and spherical surfaces, as shown in Fig. S1(a) (included by ESM of Appendix). The prepared intermediate Si@a-TiO₂ NPs retain the structure of SiNPs but exhibit rough surfaces, as shown in Figs. S1(b)‒S1(d) (included by ESM of Appendix), indicating the successful coating with the TiO₂ shell layer. Clearly, the addition of titanium tetraisopropanolate has a significant impact on the morphology of Si@a-TiO₂ NPs. An insufficient addition of titanium tetraisopropanolate results in the incomplete a-TiO₂ shell that fails in effectively suppressing the volume expansion, while an excessive addition causes aggregation of Si@a-TiO₂ NPs, thereby affecting the overall conductivity of the electrode material. Fig.1(b)‒1(d) reveal SEM images of Si@a-TiO₂@a-C-4, Si@a-TiO₂@a-C-6, and Si@a-TiO₂@a-C-8, respectively. The structure and morphology of NPs are found to be minimally affected by the deposition of carbon. Elemental mapping results demonstrate homogeneous distributions of C, O, Si, and Ti elements within the composites (Fig.1(h)). Furthermore, overlapping signals from Si, Ti, and O elements confirm the successful preparation of core‒shell structures. TEM analysis further illustrates microstructures of Si@a-TiO₂@a-C-4, Si@a-TiO₂@a-C-6, and Si@a-TiO₂@a-C-8 (Fig.2(a)‒2(c)), and all the samples exhibit uniform spherical structures. In high-magnification TEM images (Fig.2(d)‒2(f)) of these samples, a thin and transparent layer of a-C wrapped around the outermost layer can be observed. The intermediate layer consists of an elastic a-TiO₂ shell with the thickness positively correlated to the quantity of the added titanium source observed in these samples. Additionally, the high-magnification TEM images indicate that the brief CVD process has not led to the formation of crystalline TiO₂ (c-TiO₂). The SiNP core is well-crystallized and conformally wrapped by the a-TiO₂ layer. Fig.2(g)‒2(i) illustrate that the lattice fringes exhibit an interplanar spacing of 0.31 nm, corresponding to the (1 1 1) crystal planes of silicon [39].

Fig.1 (a) Synthesis process of Si@a-TiO2@a-C composites. (b)(c)(d)(e)(f)(g) SEM images of Si@a-TiO2@a-C-4 (left panels), Si@a-TiO2@a-C-6 (middle panels), Si@a-TiO2@a-C-8 (right panels). (h) EDS elemental mapping results of the Si@a-TiO2@a-C-6 composite.

Full size|PPT slide

Fig.2 Morphological characterizations of Si@a-TiO2@a-C composites: TEM images of (a)(d) Si@a-TiO2@a-C-4, (b)(e) Si@a-TiO2@a-C-6, and (c)(f) Si@a-TiO2@a-C-8. Lattice fringes of (g) Si@a-TiO2@a-C-4, (h) Si@a-TiO2@a-C-6, and (i) Si@a-TiO2@a-C-8.

Full size|PPT slide

Fig.3(a) demonstrates crystal structures of Si@a-TiO₂@a-C-4, Si@a-TiO₂@a-C-6, and Si@a-TiO₂@a-C-8 analyzed via XRD. It is shown that all peaks observed in each sample are consistent with Si (JCPDS no. 27-1402). The absence of any detectable signals assignable to TiO₂ also corroborates the absence of c-TiO₂. Therefore, TiO₂ may exist in the amorphous state. A comparison with Fig. S2 (included by ESM of Appendix) reveals that properties of TiO₂ remain unchanged after undergoing the CVD process, consistent with the TEM results. The SSA and pore structure characteristics of the samples were investigated through the N₂ adsorption–desorption isotherm testing. Fig.3(b) illustrates that the Brunaur–Emmett–Teller (BET) surface areas are 5.8 m²·g⁻¹ for SiNPs, 36.7 m²·g⁻¹ for Si@a-TiO₂-6, and 71.6 m²·g⁻¹ for Si@a-TiO₂@a-C-6. The larger surface area of Si@a-TiO₂@a-C-6 is favorable for the adsorption of Li⁺. As can be seen from Fig. S3 (included by ESM of Appendix), the BET surface areas are 85.9 m²·g⁻¹ for Si@a-TiO₂@a-C-4 while 64.8 m²·g⁻¹ for Si@a-TiO₂@a-C-8. As illustrated in Fig.3(c), the shell of a-TiO₂ provides a substantial number of pores larger than 2 nm in diameter, while the micropores (smaller than 2 nm in diameter) are attributed to the formation of a-C. Specifically, the micropores can supply plentiful active sites for the storage of Li⁺ [40]. Figures S3(a) and S3(b) (included by ESM of Appendix) indicate that the layer thickness of the TiO₂ shell has a minimal impact on the overall surface areas of Si@a-TiO₂@a-C samples. The successful formation of a-C can be inferred from the presence of the D peak and G peak in the Raman spectra, as depicted in Fig.3(d) together with Fig. S4 (included by ESM of Appendix). Additionally, the chemical composition of Si@a-TiO₂@a-C samples was analyzed via XPS. Fig.3(e) illustrates that Si@a-TiO₂@a-C-6 contains four elements, Si, Ti, O, and C. The O 1s spectrum (Fig.3(f)) can be deconvoluted into three peaks corresponding to Ti−O (530.4 eV), Si−O (531.8 eV), and C−O (532.9 eV) [41]. Notably, the formation of C−O bonds is responsible for the robust bond between TiO₂ and a-C, which facilitates the attachment of carbon layers to the TiO₂ shell. The C 1s spectrum (Fig.3(g)) can be deconvoluted into two peaks corresponding to C−C (284.9 eV) and C−O (286.2 eV). Fig.3(h) shows that two distinct fitted peaks are evident at 458.5 and 464.7 eV, which are attributed to Ti 2p_3/2 and Ti 2p_1/2, respectively [42]. Moreover, the Si 2p spectrum in Fig.3(i) illustrates that the peaks located at 98.9 and 102.8 eV belong to Si−Si and Si−O, respectively, attributed to slight oxidation on the surface of SiNPs [43]. As illustrated in Figs. S5 and S6 (included by ESM of Appendix), Si@a-TiO₂@a-C-4 and Si@a-TiO₂@a-C-8 also exhibit similar elemental compositions and properties to those of Si@a-TiO₂@a-C-6, proving the generality of this synthesis method.

Fig.3 (a) XRD patterns of Si@a-TiO2@a-C composites. (b) Nitrogen adsorption–desorption isotherms and (c) PSDs of SiNPs, Si@a-TiO2-6, and Si@a-TiO2@a-C-6. (d) Raman spectra of Si@a-TiO2@a-C composites. (e)(f)(g)(h)(i) XPS survey and O 1s, C 1s, Ti 2p, and Si 2p spectra of Si@a-TiO2@a-C-6.

Full size|PPT slide

The Li⁺ storage performances of Si@a-TiO₂@a-C were evaluated through assembling the 2032-type half-cells. Fig.4(a)‒4(c) exhibit galvanostatic charge–discharge (GCD) curves of Si@a-TiO₂@a-C-4, Si@a-TiO₂@a-C-6, and Si@a-TiO₂@a-C-8 at a current density of 0.2 A·g⁻¹. The initial reversible capacity and initial coulombic efficiency (ICE) of Si@a-TiO₂@a-C-6 are 2741.4 mA·h·g⁻¹ and 87.3%, respectively, while ICE values of Si@a-TiO₂@a-C-4 and Si@a-TiO₂@a-C-8 are 86.5% and 85.1%, respectively. The loss of initial capacity is usually caused by the formation of solid electrolyte interphase (SEI) layers and the occurrence of side reactions. CV curves of the Si@a-TiO₂@a-C-6 electrode at a scan rate of 0.1 mV·s⁻¹ and in a voltage range from 0.005 to 3.0 V are shown in Fig.4(d). The peak at 1.1 V during the first cathodic scan disappears in subsequent cycles, which is typically associated with the formation of SEI layers [44]. In subsequent scans, the peak appearing at 0.18 V is attributed to the formation of Li_xSi, while the peak at 0.6 V is attributed to the dealloying reaction of Si. The gradual intensity enhancement of these peaks suggests an improvement in the kinetics of lithiation and delithiation [45]. Additionally, the two peaks observed at 1.78 and 2.06 V correspond to redox reactions occurring in the TiO₂ shell layer. Meanwhile, Si@a-TiO₂@a-C-4 and Si@a-TiO₂@a-C-8 also display the same alloying–dealloying processes as that of Si@a-TiO₂@a-C-6, which are shown in Figs. S7(a) and S7(b) (included by ESM of Appendix), respectively. To investigate the stability provided by the dual-shell encapsulated structure with the shell of a-TiO₂ and the layer of a-C, the long-term cycling performances of Si@a-TiO₂@a-C-4, Si@a-TiO₂@a-C-6, and Si@a-TiO₂@a-C-8 at a current density of 1 A·g⁻¹ are explored (Fig.4(e)). After 500 cycles, Si@a-TiO₂@a-C-6 retains a high capacity of 877.1 mA·h·g⁻¹ with a capacity retention of 89.8% (following the completion of 25 cycles of activation), superior to those of Si@a-TiO₂@a-C-4 (660.6 mA·h·g⁻¹ with a capacity retention of 82.9%) and Si@a-TiO₂@a-C-8 (567.3 mA·h·g⁻¹ with a capacity retention of 72.5%). As highlighted in Table S1 (included by ESM of Appendix), the excellent long-term cycling performance of Si@a-TiO₂@a-C-6 outperforms many reported Si-based materials used as LIB anodes. In contrast, SiNPs exhibit significant deterioration in cycle performance after 50 cycles, as shown in Fig. S8 (included by ESM of Appendix). Thus, it can be concluded that the integrity and uniformity of the a-TiO₂ shell and the a-C layer have a considerable impact on the cycling performance of the electrode. The rate performances of Si@a-TiO₂@a-C were also explored at different current densities. As shown in Fig.4(f), Si@a-TiO₂@a-C-6 achieves high charge capacities of 2150.5, 1498.6, 1072.1, 760.4, and 418.4 mA·h·g⁻¹ at current densities of 0.2, 0.5, 1.0, 2.0, and 5.0 A·g⁻¹, respectively. Furthermore, the testing at a current density of 0.2 A·g⁻¹ resulted in the specific capacity recovery to 2104.2 mA·h·g⁻¹, showcasing outstanding electrochemical reversibility at a high current density (5.0 A·g⁻¹) for the Si@a-TiO₂@a-C-6 electrode. Interestingly, Fig. S9 (included by ESM of Appendix) illustrates the poor rate performance of Si@a-TiO₂ electrodes without carbon. This phenomenon can be attributed to the presence of a-C promoting both ion mobility and electron transfer.

Fig.4 (a)(b)(c) GCD curves at 0.2 A·g−1 of Si@a-TiO2@a-C composites. (d) CV curves of Si@a-TiO2@a-C-6 at 0.1 mV·s−1. (e) Cycling performance of Si@a-TiO2@a-C composites at 1.0 A·g−1. (f) Rate performance of Si@a-TiO2@a-C composites at different current densities. (g) GITT curves of Si@a-TiO2@a-C composites. (h) Voltage response over time during a single current pulse. (i) Voltage-dependent lithium-ion diffusion coefficients of Si@a-TiO2@a-C composites. (j) Nyquist plots of Si@a-TiO2@a-C composites in their initial cycling.

Full size|PPT slide

The galvanostatic intermittent titration technique (GITT) was applied to explore the Li⁺ diffusion coefficient (D(Li⁺)) [34,46–47]. Fig.4(g) and 4(h) illustrate GITT curves of Si@a-TiO₂@a-C electrodes at 0.05 A·g⁻¹ and the voltage response over time during a single current pulse of Si@a-TiO₂@a-C-6, respectively. In accordance with prior work [48], D(Li⁺) can be calculated based on Eq. (1) as follows:

(1)$D(\text{L}{\text{i}}^{+})=\frac{4}{\text{π}\tau }{\left(\frac{m_{\text{B}}{V}_{\text{M}}}{M_{\text{B}}S}\right)}^2{\left(\frac{\mathrm{\Delta }{E}_{\text{s}}}{\mathrm{\Delta }{E}_{\tau }}\right)}^2$

where, τ is the pulse time, M_B is the molar mass of material, m_B and S are the active mass and the surface area of the tested sample, respectively, V_M is the molar volume, and both ΔE_s and ΔE_τ can be obtained from our test results.

Based on Eq. (1), D(Li⁺) values of Si@a-TiO₂@a-C electrodes can be calculated, as shown in Fig.4(i). The D(Li⁺) value of Si@a-TiO₂@a-C-6 is higher than those of Si@a-TiO₂@a-C-4 and Si@a-TiO₂@a-C-8, demonstrating that the Si@a-TiO₂@a-C-6 electrode possesses the most complete structure and conductive network. As illustrated in Fig.4(j), the EIS curve of each electrode comprises a semicircular component at high frequencies, which is associated with the charge transfer resistance (R_ct) at the electrode‒electrolyte interface (EEI). The sloping line at the low-frequency region is indicative of the impedance associated with the transfer of ions to the anode material. In this work, Si@a-TiO₂@a-C-6 shows a smaller R_ct than those of Si@a-TiO₂@a-C-4 and Si@a-TiO₂@a-C-8, indicating better charge transfer kinetics. Meanwhile, Fig. S10 (included by ESM of Appendix) illustrates a notable reduction in the electrode charge transfer impedance after carbon coating, further demonstrating the promoting effect of the a-C layer on the formation of the overall conductive network in Si@a-TiO₂@a-C-6.

To elucidate the mechanism of electrode alterations during the charging‒discharge process, a variety of kinetic processes with distinctive relaxation characteristics were discerned on different time scales by subjecting the impedance to varying numbers of cycle turns and integrating the results with the DRT analysis [49–50]. As demonstrated in Fig.5(a), the Li⁺ transfer processes occurring within the composites encompass bulk phase diffusion (τ₁), charge transfer reactions (τ₂), and transport processes across the SEI layers (τ₃). Fig.5(b) illustrates the impedance change after different numbers of cycles. With the enhancement of the cycle number, the charge transfer resistance rises gradually, while the diffusion impedance diminishes, resulting in an appropriate increase in the diffusion rate of Li⁺. The overall trend indicates a consistent rise in impedance. The extraction and analysis of the temporal scale information from half-cells will facilitate the advancement of dynamic research. In the DRT spectrum (Fig.5(c)), the x-axis represents the time constant, indicating inherent characteristics of each electrochemical process, and the y-axis represents the impedance intensity of the corresponding process. Generally, values of response time, which correspond to the diffusion of Li⁺ in the bulk phase of the electrode material, the charge transfer reaction, and the transport process of Li⁺ through the SEI layers, are τ₁ > 10 s, 0.01 < τ₂ < 10 s, and 0.001 < τ₃ < 0.01 s, respectively [51–52]. As shown in Fig.5(c) and 5(d), the intensity of the characteristic peak of τ₁ decreases as the number of cycles increases in Si@a-TiO₂@a-C-6, attributed to the fact that the diffusion process becomes more favorable due to the increased wetting degree of the electrolyte on the electrode. The gradual intensity decrease of the τ₂ peak is related to the activation effect of materials and the reduction of side reactions during the cycling process. As cycling proceeds, leading to the thickening of SEI layers, the resistance to the Li⁺ transport in SEI layers increases, and the intensity of the τ₃ peak gradually increases. Fig.5(e)‒5(g) detailedly illustrate variation trends of τ₁, τ₂, and τ₃, respectively, more distinctly describing the variation trend of the Li⁺ transfer process in the electrode. Si@a-TiO₂@a-C-4 and Si@a-TiO₂@a-C-8 also show similar patterns of tendency in Figs. S11 and S12 (included by ESM of Appendix), demonstrating the generalizability of the mechanism of this type of materials.

Fig.5 (a) Schematic diagram of the DRT testing and its processes related to τ1, τ2, and τ3. (b)(c)(d) DRT analysis results obtained from EIS of the Si@a-TiO2@a-C-6 composite. (e)(f)(g) Variation trends of τ1, τ2, and τ3 peak intensities.

Full size|PPT slide

To further explore the structural evolution of the material during initial lithiation and cycling processes, the Si@a-TiO₂@a-C-6 electrode was investigated through ex-situ TEM measurements. Fig.6(a) shows that the a-TiO₂ shell and a-C coatings are robust enough to withstand the large volume expansion of SiNPs without causing particle comminution and structural fracture. The corresponding TEM results (Fig.6(b)) confirm this speculation by showing that the SiNP core undergoes a huge volume expansion during the first discharge process, leading to the distortion of Si spheres. Surprisingly, the distorted and lithiated Si@a-TiO₂@a-C-6 NPs still retain the intact core‒shell structure, suggesting that the a-TiO₂ shell and the a-C layer are robust enough to withstand the volume change. Meanwhile, to further explore the electrode stability, structures of working electrodes before and after 500 cycles were observed, as revealed in Fig.6(c)‒6(h). It is seen that the Si@a-TiO₂@a-C-4, Si@a-TiO₂@a-C-6, and Si@a-TiO₂@a-C-8 electrodes exhibit different volume expansions of 58.1%, 42.1%, and 39.8%, respectively. The result indicates that this amorphous and elastic core‒shell structure possesses excellent stability and is suitable for anode materials applied in LIBs.

Fig.6 (a) Schematic illustration showing the structural evolution of the Si@a-TiO2@a-C electrode during lithium insertion and extraction. (b) Structural evolution of the Si@a-TiO2@a-C-6 electrode with corresponding TEM images. (c)(d)(e)(f)(g)(h) SEM images of Si@a-TiO2@a-C-4 (left panels), Si@a-TiO2@a-C-6 (middle panels), and Si@a-TiO2@a-C-8 (right panels) electrodes before 500 cycles (upper panels) and after 500 cycles (lower panels).

Full size|PPT slide

4 Conclusions

In summary, we utilized a simple sol‒gel approach combined with the CVD method to prepare Si@a-TiO₂@a-C as the anode for LIBs. Si@a-TiO₂@a-C-6 demonstrates excellent electrochemical performance, achieving 877.1 mA·h·g⁻¹ at 1 A·g⁻¹ after 500 cycles, maintaining a capacity of 86.7%. The elastic a-TiO₂ shell significantly limits the volume expansion of the Si core during the lithiation process. Ex-situ TEM measurements clearly reveal the structural evolution process of Si@a-TiO₂@a-C-6 during the initial lithiation stage and cycling process. The shells of a-C provide abundant ion channels and improve the kinetics of alloying reactions. The DRT analysis significantly enhances the accuracy of kinetic interpretation of Si@a-TiO₂@a-C-6 during the lithiation process. These results provide a novel guideline for the rational structural design of high-performance electrode materials for Si-based anodes.

References

[1] Xu J, Dou Y, Wei Z, . Recent progress in graphite intercalation compounds for rechargeable metal (Li, Na, K, Al)-ion batteries.Advanced Science, 2017, 4(10): 1700146

[2] Mei L, Xu J T, Wei Z X, . Chevrel phase Mo₆T₈ (T = S, Se) as electrodes for advanced energy storage.Small, 2017, 13(34): 1701441

[3] Wu X, Lan X X, Hu R Z, . Tin-based anode materials for stable sodium storage: progress and perspective.Advanced Materials, 2022, 34(7): 2106895

[4] Li Y, Liu W, Long Z, . Si@C microsphere composite with multiple buffer structures for high-performance lithium-ion battery anodes.Chemistry, 2018, 24(49): 12912–12919

[5] Lee G, Kim S, Kim S, . SiO₂/TiO₂ composite film for high capacity and excellent cycling stability in lithium-ion battery anodes.Advanced Functional Materials, 2017, 27(39): 1703538

[6] Ogata K, Ko D, Jung C, . Spontaneous pseudo-topological silicon quantization for redesigned Si-based Li-ion batteries.Nano Energy, 2019, 56: 875–883

[7] Zhao C, Wada T, De Andrade V, . Imaging of 3D morphological evolution of nanoporous silicon anode in lithium ion battery by X-ray nano-tomography.Nano Energy, 2018, 52: 381–390

[8] Guo J P, Dong D Q, Wang J, . Silicon-based lithium ion battery systems: state-of-the-art from half and full cell viewpoint.Advanced Functional Materials, 2021, 31(34): 2102546

[9] Zhu G, Chao D, Xu W, . Microscale silicon-based anodes: fundamental understanding and industrial prospects for practical high-energy lithium-ion batteries.ACS Nano, 2021, 15(10): 15567–15593

[10] Park S, King P, Tian R, . High areal capacity battery electrodes enabled by segregated nanotube networks.Nature Energy, 2019, 4(7): 560–567

[11] Chae S, Choi S H, Kim N, . Integration of graphite and silicon anodes for the commercialization of high-energy lithium-ion batteries.Angewandte Chemie International Edition, 2020, 59(1): 110–135

[12] Cheng H, Liu Y, Cheng Z, . Ti₃C₂T_x MXene wrapped, carbon-coated porous Si sheets for improved lithium storage performance.Chinese Chemical Letters, 2024, 35(2): 108923

[13] Chen S, Ma C, Zhu Y, . Interpenetrated tunnel routes in silicon/carbon hollow sphere anodes to boost their lithium storage.Materials Chemistry Frontiers, 2020, 4(9): 2782–2790

[14] Kim H, Seo M, Park M H, . A critical size of silicon nano-anodes for lithium rechargeable batteries.Angewandte Chemie International Edition, 2010, 49(12): 2146–2149

[15] Guo T, Luo G, Shi C, . Thermal burst synthesis of high-performance Si nanotube sheets for lithium-ion batteries.ACS Sustainable Chemistry & Engineering, 2022, 10(12): 4031–4039

[16] Shen C, Fu R, Guo H, . Scalable synthesis of Si nanowires interconnected SiO_x anode for high performance lithium-ion batteries.Journal of Alloys and Compounds, 2019, 783: 128–135

[17] Chen S, Chen Z, Xu X Y, . Scalable 2D mesoporous silicon nanosheets for high-performance lithium-ion battery anode.Small, 2018, 14(12): 1703361

[18] Sun J, Shi J, Ban B, . Porous Si/C anode materials by Al–Si dealloying method with PEA surfactant assisted cross-linked carbon coating for lithium-ion battery applications.Electrochimica Acta, 2019, 327: 134995

[19] Chen S Q, Shen L F, van Aken P A, . Dual-functionalized double carbon shells coated silicon nanoparticles for high performance lithium-ion batteries.Advanced Materials, 2017, 29(21): 1605650

[20] Zhang Z L, Wang Y H, Ren W F, . Scalable synthesis of interconnected porous silicon/carbon composites by the Rochow reaction as high-performance anodes of lithium ion batteries.Angewandte Chemie International Edition, 2014, 53(20): 5165–5169

[21] Shi Q, Zhou J, Ullah S, . A review of recent developments in Si/C composite materials for Li-ion batteries.Energy Storage Materials, 2021, 34: 735–754

[22] Man Q Y, An Y L, Liu C K, . Interfacial design of silicon/carbon anodes for rechargeable batteries: a review.Journal of Energy Chemistry, 2023, 76: 576–600

[23] Xu W, Tang C, Huang N, . Adina rubella-like microsized SiO@N-doped carbon grafted with N-doped carbon nanotubes as anodes for high-performance lithium storage.Small Science, 2022, 2(4): 2100105

[24] Hu L, Luo B, Wu C H, . Yolk–shell Si/C composites with multiple Si nanoparticles encapsulated into double carbon shells as lithium-ion battery anodes.Journal of Energy Chemistry, 2019, 32: 124–130

[25] Li H W, Wang Z Y, Dang L Y, . Precursor induced assembly of Si nanoparticles encapsulated in graphene/carbon matrices and the influence of Al₂O₃ coating on their properties as anode for lithium-ion batteries.Small, 2024, 20(18): 2307722

[26] Zhang L, Huang Q W, Liao X Z, . Scalable and controllable fabrication of CNTs improved yolk-shelled Si anodes with advanced in operando mechanical quantification.Energy & Environmental Science, 2021, 14(6): 3502–3509

[27] Lotfabad E M, Kalisvaart P, Kohandehghan A, . Si nanotubes ALD coated with TiO₂, TiN or Al₂O₃ as high performance lithium ion battery anodes.Journal of Materials Chemistry A: Materials for Energy and Sustainability, 2014, 2(8): 2504–2516

[28] Sun J, Tang C, Li H T, . Anthracite-derived porous carbon@MoS₂ heterostructure for elevated lithium storage regulated by the middle TiO₂ layer.ChemSusChem, 2024, 17: e202401396

[29] Lotfabad E M, Kalisvaart P, Cui K, . ALD TiO₂ coated silicon nanowires for lithium ion battery anodes with enhanced cycling stability and coulombic efficiency.Physical Chemistry Chemical Physics, 2013, 15(32): 13646–13657

[30] Hou L, Xiong S S, Cui R W, . Three-dimensional porous carbon framework confined Si@TiO₂ nanoparticles as anode material for high-capacity lithium-ion batteries.ChemElectroChem, 2022, 9(1): e202101447

[31] Shi J W, Zu L H, Gao H Y, . Silicon-based self-assemblies for high volumetric capacity Li-ion batteries via effective stress management.Advanced Functional Materials, 2020, 30(35): 2002980

[32] Yang J P, Wang Y X, Li W, . Amorphous TiO₂ shells: a vital elastic buffering layer on silicon nanoparticles for high-performance and safe lithium storage.Advanced Materials, 2017, 29(48): 1700523

[33] Vats B N, Gupta R, Gupta A, . Enhancing Li-ion battery performance through the integration of Si@TiO₂ core–shell nanoparticles with natural graphite.ChemistrySelect, 2024, 9(3): e202303545

[34] Pan Q, Zhao J, Xing B, . A hierarchical porous architecture of silicon@TiO₂@carbon composite novel anode materials for high performance Li-ion batteries.New Journal of Chemistry, 2019, 43(38): 15342–15350

[35] Li Z S, Zhao Z Y, Pan S Y, . Covalent coating of micro-sized silicon with dynamically bonded graphene layers toward stably cycled lithium storage.Advanced Energy Materials, 2023, 13(28): 2300874

[36] Xu C, Ma G, Yang W, . Amorphous Sn modified nitrogen-doped porous carbon nanosheets with rapid capacitive mechanism for high-capacity and fast-charging lithium-ion batteries.Frontiers of Materials Science, 2023, 17(3): 230651

[37] Liu H, Yang W, Che S, . Silicon doped graphene as high cycle performance anode for lithium-ion batteries.Carbon, 2022, 196: 633–638

[38] Li W, Yang J, Wu Z, . A versatile kinetics-controlled coating method to construct uniform porous TiO₂ shells for multifunctional core–shell structures.Journal of the American Chemical Society, 2012, 134(29): 11864–11867

[39] Jiao X, Tian Y, Zhang X . Hollow Si nanospheres with amorphous TiO₂ layer used as anode for high-performance Li-ion battery.Applied Surface Science, 2021, 566: 150682

[40] Ma G, Yang W, Xu C, . Nitrogen-doped porous carbon embedded Sn/SnO nanoparticles as high-performance lithium-ion battery anode.Electrochimica Acta, 2022, 428: 140898

[41] Xiang J, Liu H, Na R, . Facile preparation of void-buffered Si@TiO₂/C microspheres for high-capacity lithium ion battery anodes.Electrochimica Acta, 2020, 337: 135841

[42] Jiang J, Liu S, Wang Y, . Auto-adjustment of structure and SnO₂ content of SnO₂/TiO₂ microspheres for lithium-ion batteries.Chemical Engineering Journal, 2019, 359: 746–754

[43] Lu B, Ma B, Deng X, . Dual stabilized architecture of hollow Si@TiO₂@C nanospheres as anode of high-performance Li-ion battery.Chemical Engineering Journal, 2018, 351: 269–279

[44] Shen D, Huang C, Gan L, . Rational design of Si@SiO₂/C composites using sustainable cellulose as a carbon resource for anodes in lithium-ion batteries.ACS Applied Materials & Interfaces, 2018, 10(9): 7946–7954

[45] Guo S, Hu X, Hou Y, . Tunable synthesis of yolk–shell porous silicon@carbon for optimizing Si/C-based anode of lithium-ion batteries.ACS Applied Materials & Interfaces, 2017, 9(48): 42084–42092

[46] Xu Y, Wang C L, Niu P, . Tuning the nitrogen-doping configuration in carbon materials via sulfur doping for ultrastable potassium ion storage.Journal of Materials Chemistry A: Materials for Energy and Sustainability, 2021, 9(29): 16150–16159

[47] Zou Z G, Wang Q, Yan J, . Versatile interfacial self-assembly of Ti₃C₂T_x MXene based composites with enhanced kinetics for superior lithium and sodium storage.ACS Nano, 2021, 15(7): 12140–12150

[48] Li Y M, Hu Y S, Titirici M M, . Hard carbon microtubes made from renewable cotton as high-performance anode material for sodium-ion batteries.Advanced Energy Materials, 2016, 6(18): 1600659

[49] Lu Y, Zhao C, Huang J, . The timescale identification decoupling complicated kinetic processes in lithium batteries.Joule, 2022, 6(6): 1172–1198

[50] Gautam M, Mishra G, Furquan M, . Design of low-stress robust silicon and silicon-carbide anode with high areal capacity and high energy density for next-generation lithium-ion batteries.Chemical Engineering Journal, 2023, 472: 144916

[51] Tian W, Zhang H, Duan X, . Porous carbons: structure-oriented design and versatile applications.Advanced Functional Materials, 2020, 30(17): 1909265

[52] Soni R, Robinson J, Shearing P, . Lithium–sulfur battery diagnostics through distribution of relaxation times analysis.Energy Storage Materials, 2022, 51: 97–107

Authors’ contributions

Guang Ma: writing — review & editing, formal analysis, and writing — original draft; Chong Xu: writing — review & editing, formal analysis, and writing — original draft; Dongyuan Zhang: formal analysis and writing — original draft; Sai Che: writing — review & editing; Yuxin Liu: formal analysis; Gong Cheng: writing — review & editing; Chenlin Wang: writing — original draft; Kexin Wei: writing — review & editing and formal analysis; Yongfeng Li: writing — review & editing and supervision.

Declaration of competing interests

The authors declare no conflict of interests.

Acknowledgements

We gratefully acknowledge the financial support from the National Natural Science Foundation of China (Grant Nos. 22238012, 22178384, 22108301, and 22408398) and the Science Foundation of China University of Petroleum, Beijing (Grant No. ZX20230242).

Online appendix

Electronic supplementary material (ESM) can be found in the online version at https://doi.org/10.1007/s11706-024-0708-6 and https://journal.hep.com.cn/foms/EN/10.1007/s11706-024-0708-6 that includes Figs. S1–S12 and Table S1.

RIGHTS & PERMISSIONS

2024 Higher Education Press