Efficient Lithium Storage of Si‐Based Anode Enabled by a Dual‐Component Protection Strategy

Si, as a highly competitive anode material for lithium‐ion batteries (LIBs), has gained enormous commercial interests due to its high theoretical capacity, low delithiation potential, and natural abundance. However, its poor cycling stability/electron conductivity seriously restrains its practical applications. To address this problem well, herein, a scalable spray‐drying method is explored to construct an ultrahigh stable 3D Si‐based composite (designed as Si@C‐MX) anode, where the Ti3C2Tx MXene nanosheets (NSs) crump the nano‐Si coated uniformly with an ultrathin carbon layer. Synergistically, the coating carbon layer and Ti3C2Tx NSs as the conductive elastomer constrain/buffer the volume expansion of nano‐Si, avoid direct contact with the electrolyte, build a continuous electronic network for rapid electron transport, and meanwhile improve mechanical properties of the electrodes. Thanks to the dual protection (i.e., carbon coating and Ti3C2Tx NSs) strategy, the resultant Si@C‐MX anode exhibits large reversible capacities, superior rate capability, and long‐duration cycle stability. Additionally, the Si@C‐MX‐based full batteries delivered an energy density of 371.8 Wh kg−1 based on the whole device at 123.9 W kg−1 and a desirable capacity retention with cycling, which convincingly highlights its promising application in advanced LIBs.

competitive electrochemical behaviors for LIBs. However, such a complicated yet costly synthesis process is unfavorable for the mass production at all. As is well known, the spray-drying method, as a well-integrated avenue in the manufacturing process, can provide a facile, rapid, and industrially adaptable fabricated process, resulting in lower fabrication cost and shortened production durations. [22,23] While, many researches, up till now, were mainly focused on the influence of process parameters (temperature, pressure, and droplet size) on specific morphologies of final products. [24,25] Considering the comprehensive analysis above, herein, a dualcomponent protection strategy was purposefully explored via a spray-drying method to construct the 3D Si-based composite, in which the Ti 3 C 2 T x nanosheets (NSs) crumped the Si nanoparticles (NPs) coated with the chitosan (CS) derived carbon layer (denoted as Si@C-MX) for LIBs. The uniqueness of the Si@C-MX composite lies in that the moderate carbon layer coating can restrain the volume expansion of Si NPs and avoid direct contact between the electrolyte and Si NPs. On the other hand, the Ti 3 C 2 T x NSs substrate serves as the elastomer to provide hierarchical buffer for volume change of Si NPs. The tight connection between the hydroxylated Si NPs, negatively charged Ti 3 C 2 T x NSs, and positively charged CS under the electrostatic interactions guarantees the successful fabrication of the unique Si@C-MX. It is through the synergistic effects between them that the improved ion/electron transport and mechanical properties are simultaneously achieved in the entire 3D network. Benefit from the aforementioned advantages, the as-obtained Si@C-MX anode provided the ultra-high lithium storage capacity, superb rate performance, and excellent cycle stability. Besides, the Si@C-MX-based full batteries delivered an outstanding discharge capacity of 169.4 mAh g À1 (based on the cathode) and desirable capacity retention of about 100% at 0.05 A g À1 after 70 cycles.

Physicochemical and Structural Characteristics
The preparation process of 3D Si@C-MX is schemetically illustrated in Figure 1a. Initially, the positively charged CS is spontaneously adsorbed on the Si NPs, which are negatively charged due to the -OH terminated groups on their surface after piranha lotion treatment, forming the nanohybrid (i.e., Si@CS NPs). Then, the resultant Si@CS NPs are located on the surface of Ti 3 C 2 T x NSs with a negative charge after the Ti 3 C 2 T x NSs suspension is further dropped. After that, the uniform mixture is delivered from the feed pipe to the atomizer of the spray dryer. During the spray-drying process, the increasing surface capillary force originating from the evaporation of water on the surface of Ti 3 C 2 T x NSs impels their local deformation. And finally, the highly folded flower balls, namely the Si@CS-MX, are compressed into, which are collected after cyclone. Afterwards, the carbonization of organic CS at high temperature forms amorphous carbon in the annealing process, which constructs a closer connection with the nano Si and Ti 3 C 2 T x . Accordingly, the 3D Si@C-MX is successfully prepared.
The microstructures of the synthesized samples were evaluated by field-emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM). After spray drying, the obtained Si@CS-MX precursor exhibits a unique 3D flowerspherical morphology with highly crumpled Ti 3 C 2 T x NSs as the loading matrix ( Figure S1a, Supporting Information). The magnified FESEM image ( Figure S1b, Supporting Information) shows that the Si@CS NPs with a size of around 100 nm are tightly wrapped by the Ti 3 C 2 T x NSs, forming a continuous conductive network. The Si@C-MX specimen, as shown in Figure 1b-d, well inherits the representative morphology of its precursor after being annealed at 600°C, where the formed Si@C NPs are tightly wrapped in the Ti 3 C 2 T x NSs. Further TEM observation (Figure 1e) clearly evidences the existence of the Ti 3 C 2 T x NSs at the edge of the flower spheres. The Si NPs, nano-carbon layer, and Ti 3 C 2 T x NSs couple with each other, as displayed in the enlarged TEM image (Figure 1f ), confirming the co-existence and homogeneously distribution of the three phases. Apparently, the Si@C NPs are evident with a diameter of 100 nm (Figure 1g), which is consistent with the FESEM observations ( Figure S1b, Supporting Information; Figure 1d). The high-resolution TEM (HRTEM) image ( Figure 1h) visualizes a thin yet uniform carbon layer of about 2.5 nm coating on the surface of Si NPs with discerned lattice fringes. The uniform carbon coating can form a functional region to constrain/buffer the severe volume expansion/contraction due to the alloying/dealloying Si during charge and discharge processes. The well-defined fringes in the regions (i and j) (Figure 1h) are estimated as 0.320 ( Figure 1i) and 0.314 nm (Figure 1j), respectively, which correspond to the (111) plane of crystalline Si. The energy-dispersive spectroscopy (EDS) mapping images (Figure 1k) confirm the uniform distributions of C, Ti, and Si elements in the selected region. All above characterizations corroborate the successful spray-drying synthesis of the flower-ball 3D Si@C-MX, in which the Si NPs with carbon coating are uniformly loaded and encapsulated in the 2D Ti 3 C 2 T x NSs, thus building a 3D cross-linked network. For comparison, the other two samples of Si@C and Si-MX are prepared by the same process, which also exhibits the regular flower-ball structures ( Figure S2, Supporting Information). As for the Si@C, the Si NPs are tightly connected by the carbon network ( Figure S2a,b, Supporting Information). And the Si-MX is formed with the Si NPs wrapped with the highly crumped Ti 3 C 2 T x NSs ( Figure S2c,d, Supporting Information). The TEM images ( Figure S3a-d, Supporting Information) further confirm the thin and uniform carbon layer in the Si@C sample and few-layered Ti 3 C 2 T x NSs in the Si-MX, which is well supported by the HRTEM images ( Figure S3c,d, Supporting Information) and detailed EDS mapping images ( Figure S3e,f, Supporting Information). Figure 2a shows the X-Ray diffraction (XRD) patterns of the Ti 3 C 2 T x NSs and Si@C-MX. The main diffraction peak located at 2θ ¼ 7.8°is the typical (002) plane of 2D Ti 3 C 2 T x . As regards the Si@C-MX, the new reflections appearing at 2θ ¼ 28.4, 47.3, 56.1, 69.1, and 76.4°are well fitted with the (111), (220), (311), (400), and (331) planes of pure Si phase (PDF#27-1402), besides other weak peaks originating from the Ti 3 C 2 T x NSs. It is particularly noted that the diffraction peaks in the Si@C-MX are consistent with that of Si@CS-MX ( Figure S4, Supporting Information) and there are no new peaks about TiO x and SiO x appeared, but the (002) peak disappears both in the Si@CS-MX and Si@C-MX. It indicates that the certain aggregation of Ti 3 C 2 T x NSs occurs in the composite and the hightemperature calcination does not destroy the phase structure of Ti 3 C 2 T x and Si. Furthermore, the small broad peaks centering around 21.5°should be ascribed to the contribution from the amorphous carbon. [26] Similar observations are clear as well for the cases of Si@C and Si-MX samples ( Figure S5, Supporting Information).
The X-Ray photoelectron spectroscopy (XPS) was further conducted to shed more light on elemental composition and valence states of the Si@C-MX. The full spectrum (Figure 2b [27,28] The Si 2p spectra of Si@C-MX and Si NPs are comparatively shown in Figure 2d. The fitted peaks at 99.2 eV correspond to the bond of low valence Si, which mainly exists in the pure Si phase, while the peaks located at 103.4 eV are assigned to the high valence Si-O bond, which mainly exists in the SiO x . [5] While, no signals for the SiO x are detected in the XRD pattern ( Figure 2a). It suggests that the amorphous Si-O layer is probably formed on the surface of Si NPs in Si@C-MX sample, which is easily induced by the oxygen-containing functional groups on the surface of Si NPs and Ti 3 C 2 T x NSs over the high-temperature treatment.
The Raman spectra were performed in the different wavelength ranges to investigate the chemical environment of the Si@C-MX sample. The peaks, as displayed in Figure 2e, correspond to the vibration peak of elemental Si. Obviously, compared with pure Si NPs, the peak position exhibits a blue shift (from 511 to 514.7 cm À1 ) in the Si@C-MX, along with www.advancedsciencenews.com www.advenergysustres.com the decreasing strength. The phenomenon is generally attributed to the phonon confinement/masking effect [29,30] due to the thorough carbon coating on the surface of Si NPs, thus affecting the lattice vibration of Si. The presence of the carbon coating was further proved by Raman spectra from 1000 to 1800 cm À1 , as collected in Figure 2f. The characteristic D-band (1368 cm À1 , i.e., the disordered graphitic structure) and G-band (1595 cm À1 , i.e., the E 2g phonon of sp 2 carbon atoms) are apparent coupled with the D 00 -band corresponding to amorphous carbon. [31] Generally, the integrated area ratio of D-band to G-band (I D /I G ) represents the graphitization degree of samples. The I D /I G values of Si-MX, Si@C, and Si@C-MX are estimated as 0.94, 1.73, and 1.88, respectively. Besides this, in the spectra of Si@C and Si@C-MX, the strength of D 00 -band is larger than that of Si-MX, further revealing the amorphous carbon layer both formed in the products of Si@C and Si@C-MX during annealing.

Electrochemical Evaluation
As discussed above, the 3D Si@C-MX composite has been successfully prepared through a facile yet industrially adaptable method, which can provide a dual-component protection by the amorphous carbon layer and flexible Ti 3 C 2 T x NSs. Besides, the tight 3D cross-linked structure renders an efficient conductive network for efficient electrochemical reactions. Thanks to the appealing dual-component protection, the simultaneous achievements in restraining/buffering the volume change of Si NPs and rapid electron transport/ion transfer during cycling are realized in the Si@C-MX flower-spheres. Thus, the Si@C-MX is highly anticipated with appealing electrochemical Li-storage behaviors. For this, electrochemical evaluations were comprehensively conducted. Figure 3a demonstrates cyclic voltammetry (CV) curves of Si@C-MX and Si NPs within the potential range of 1.0-3.0 V (vs Li/Li þ ) at a scanning rate of 0.1 mV s À1 . Evidently, the CV curves of the Si@C-MX anode show a similar feature to those of the pure Si NPs, featuring the same electrochemical lithium storage processes for the two. In the first cathodic process, the broad peak centering at around 0.5-1.0 V can be attributed to the formation of SEI film, [11] owing to the decomposition of electrolyte on the electrode surface and the irreversible lithium ion consumed by the surface functional groups of electrode materials, especially the surface functional groups of Ti 3 C 2 T x NSs, which disappears in the second cycle. The sharp reduction peaks starting from 0.5 V mainly correspond to the lithiation of Si phase, which forms the Li x Si alloy phase (0 < x ≤ 4.4), [17] as described by the following equation The anodic peaks at approximately 0.33 and 0.48 V originate from the two-step dealloying process of Li x Si. [32][33][34] During the subsequent second and third cycles of the Si@C-MX (the lower in Figure 3a), the positions of the second anodic peak increase from 0.48 to 0.52 V, and the integrated areas under electrochemical response currents are enhanced with CV cycle prolonging, indicating the slow activation process of the Si@C-MX anode, which is attributed to the amorphization of the crystalline Si due to the large volume expansion during the initial lithium/ delithiation process. [34] The phenomenon is examined as well for the pure Si NPs (the upper in Figure 3a).   Figure 3b profiles the galvanostatic discharge-charge (GCD) plots of the Si@C-MX at a current density of 0.1 A g À1 . Apparently, the lithiation/delithiation platforms appearing in GCD plots well match with the cathodic/anodic peaks in the CV responses (Figure 3a). The Si@C-MX anode delivers high discharge/charge capacities of 1437.8/1186.4 mAh g À1 , that is, a high initial Coulombic efficiency (ICE) of 82.5%.
To further investigate the dual-component protection effect in the Si@C-MX anode, rate properties of the Si@C-MX, Si@C, and Si-MX electrodes were carried out firstly in a wide current density range from 0.1 to 10.0 A g À1 (Figure 3c; S6a,b, Supporting Information). Remarkably, the Si@C-MX electrode exhibits average discharge capacities of 1036.2, 1039.8, 959.2, 844.5, 747.0, and 593.9 mAh g À1 at the current densities of 0.1, 0.2, 0.5, 1.0, 2.0, and 5.0 A g À1 , respectively. Even at a high current density of 10.0 A g À1 , the anode can still obtain a specific capacity of 451.9 mAh g À1 . With the current density suddenly going back to 0.1 A g À1 , the specific capacity recovers up to 875.4 mAh g À1 (Figure 3c), corroborating the superb rate performance of the Si@C-MX. In addition, the Si@C-MX anode exhibits the smaller voltage drop when discharging due to the small polarization even under the ultra-high current density of 10.0 A g À1 than that of Si@C and Si-MX, which confirms the fast reaction kinetics of the Si@C-MX anode due to its favorable electronic/ionic transmission network ( Figure S7, Supporting Information). In sharp contrast, as regards Si@C ( Figure S6a, Supporting Information) and Si-MX ( Figure S6b, Supporting Information) anodes, the reversible capacities of 2475.6 and 2064.5 mAh g À1 are just provided at 0.1 A g À1 , respectively, and particularly, the capacities are gained as small as 12.8 and 12.5 mAh g À1 at a large current density of 10.0 A g À1 . Additionally, huge gaps from the initial values in capacities of the two electrodes are evidenced when the current density is back to 0.1 A g À1 once again. The high-rate performance of Si@C-MX here should be attributed to the 3D tightly linked conductive network in the composite anode, accelerating the transition of ions and electrons, which can be well supported by the electrochemical impedance spectra (EIS) measurements (Figure 3d,e). After fitted with the circuit diagram ( Figure S8, Supporting Information), the charge transfer resistance (R ct ) and internal resistance (R s ) of the Si@C-MX, Si@C, and Si-MX are collected (Table S1, Supporting Information). The R ct is derived from the www.advancedsciencenews.com www.advenergysustres.com size of semicircle in the medium-frequency region (Figure 3d), and R s from the intercepts with the x-axis (Figure 3e). Notably, the Si@C-MX electrode shows the small R ct (121.0 Ω) and R s (2.0 Ω) due to the highly efficient conductive network from both the uniform coated carbon layer and the Ti 3 C 2 T x NSs matrix, as schematically illustrated in Figure 3f, which is highly conducive to the rapid electrochemical kinetics towards (de) lithiation. The long-term cycling stabilities of the Si@C-MX, Si@C, and Si-MX electrodes are further examined at the current density of 0.5 A g À1 , as illustrated in Figure 3g. Although the Si@C and Si-MX electrodes show considerable discharge capacities of 2704.5 and 2518.0 mAh g À1 , respectively, in the initial discharge process, the capacities decay to 66.4 and 637.4 mAh g À1 just after 20 cycles, respectively. More impressively, after 200 uninterrupted cycles, the capacities of the two degrade down to 20.6 and 26.2 mAh g À1 for Si@C and Si-MX electrodes, respectively. The crucial reason for the rapid capacity decay of Si@C and Si-MX electrodes should be assigned to the repeated and drastic volume change of Si NPs during cycling, which results in the continuous rupture of Si NPs and the formation of new SEI film, causing the capacity decay and eventually the destruction of the electrodes. By contrast, a high capacity of 803.8 mAh g À1 can be obtained by the Si@C-MX after 1000 consecutive cycles. The distinct cycling performance of the Si@C-MX anode can be reasonably attributed to synergistic contributions from both the uniform carbon coating and continuous flexible Ti 3 C 2 T x NSs network, which wrap the Si NPs tightly. It is the "dual insurance" structure, rather than any single one that constrains and buffers the drastic volume change of Si NPs during the charge/discharge process, which can maintain the structural integrity and stabilize the SEI film of the electrode material meanwhile. Therefore, the Si@C-MX anode has obvious advantages compared with other reported Si-based materials (Table S2, Supporting Information).
To gain in-depth insights on the mechanism of dualcomponent protection originating from both carbon coating layer and Ti 3 C 2 T x NSs in the Si@C-MX electrode, the electrode-level changes of the Si@C, Si-MX, and Si@C-MX in thickness over the lithiation process are purposefully explored, as observed from the cross-section SEM images of the electrodes (Figure 4a-f ). The thickness of the active materials loaded on the copper foils of Si@C (Figure 4a Figure 4f ). Furthermore, the material-level morphology/ structure changes after 100 charge-discharge cycles at 0.5 A g À1 are investigated in detail, as shown in Figure 4g-j. Obviously, the single-carbon component or single-Ti 3 C 2 T x component protection strategy, namely the Si@C (Figure 4g,h) and Si-MX (Figure 4i,j), is insufficient for inhibiting the volume variation of Si NPs at all, which renders the structural crack, continuous formation of SEI film on the fresh surface, and eventually the pulverization/agglomerate of the active materials after cycling. While, the Si@C-MX electrode (Figure 4k,l) presents a wellpreserved morphology capability after long-term cycling at high current density. It is therefore easy to draw a conclusion that the dual-component protection strategy we devised here, namely the tight 3D cross-linked network formed by coated carbon layer and Ti 3 C 2 T x NSs can well address the variation issue of Si NPs during cycling, which favors to maintain the structural integrity of the electrode. In general, it is well established that nanocrystallization, carbon coating, and construction of flexible substrate all can alleviate the volume change of Si anodes during cycling to a certain extent. [35] However, as described above, the unique configuration of the Si@C-MX electrode exhibits optimal electrochemical behaviors among the three. This explicitly means that the interaction between the coating carbon layer and Ti 3 C 2 T x NSs plays a crucial role in the enhancement of electrochemical performance, rather than any single one. Therefore, as one of the key characteristics of electrodes, the in-plane elastic modulus, which can represent the mechanical property, was evaluated by density functional theory (DFT) calculations. The relaxed configuration of carbon layer (C), mono-layered Ti 3 C 2 , and interface of C-Ti 3 C 2 are schematically shown in Figure 4m-o. The distance between C and Ti 3 C 2 in C-Ti 3 C 2 is 2.17 Å due to the van der Waals interaction. In view of the hexagonal symmetry of the above structure, there are only three independent in-plane elastic constants, namely C 11 ¼ C 22 , C 12 , and C 66 (Table S3, Supporting Information), which meets the born Huang criteria, [36] indicating that both of them have good mechanical properties. The in-plane Young's modulus (Y 2D ) and Poisson's ratio (v 2D ) can be obtained according to Equation (2) and (3). The Y 2D and v values of C and Ti 3 C 2 are estimated to be 347.9 N m À1 /0.18 and 257.8 N m À1 / 0.16, respectively, which are consistent with the reports before. [37,38] Nevertheless, along with the v of 0.13, the Y 2D is calculated as 597.3 N m À1 for the C-Ti 3 C 2 , even higher than those of C and Ti 3 C 2 . As a result, the significantly improved mechanical properties brought by the smart combination of carbon layer and Ti 3 C 2 T x NSs in the Si@C-MX can greatly enhance the deformation resistance ability of electrodes under high mechanical stress during cycling.
Considering the excellent half-cell electrochemical performance of the Si@C-MX anode itself, the prospective application was further evaluated by the full devices. As schematically depicted in Figure 5a, the full LIBs are assembled with the commercial LiNi 0.8 Co 0.1 Al 0.1 O 2 (NCA) and Si@C-MX as the cathode and anode materials, respectively. As derived from the GCD plots (Figure 5b), the first charge/discharge capacities are www.advancedsciencenews.com www.advenergysustres.com 225.0/181.7 mAh g À1 at 0.2 A g À1 in the voltage range of 3-4.3 V (vs Li/Li þ ), corresponding a high ICE of 80.8%. In addition, the NCA cathode still can achieve a reversible discharge capacity of 183.0 mAh g À1 , along with a high CE value of 99% in the third cycle. As observed from the differential capacity versus voltage profiles of NCA and Si@C-MX (Figure 5c), the combination of them provides the lithium-ion full cells with an averaging voltage of about 3.5 V. Typically, the constructed Si@C-MX//NCA full devices achieve an initial discharge capacity of 165.2 mAh g À1 at 0.05 A g À1 in the voltage range of 2.7-4.2 V, coupled with an ICE value of 80%, and an average working voltage of about 3.8 V can be observed (Figure 5d). Typically, the mass loadings of NCA cathode and Si@C-MX anode in the full battery are separately 2.4 and 0.5 mg, which demonstrates a maximum energy density of 371.8 Wh kg À1 at a power density of 123.9 W kg À1 according to the discharge curve by employing the accepted 40 % penalty factor to account for the weight of the electrolyte and of the auxiliary components. [39] After 70 charge-discharge cycles at 0.05 A g À1 , a discharge capacity of the cell (Figure 5e) can be maintained as 169.4 mAh g À1 (based on the cathode), along with a desirable capacity retention of 100% and CE of 99.0%, which convincingly highlights its promising application in advanced LIBs as a competitive anode material.

Conclusion
In conclusion, in the work, a dual-component protection strategy was smartly developed to construct the advanced Si@C-MX anode by a facile and industrially adaptable spray-drying avenue to address the inherent disadvantages of Si anodes, especially the modest rate performance and cycling stability. The dualprotection strategy here, namely, uniform amorphous carbon coating layer and flexible Ti 3 C 2 T x NSs substrate, which ensured the structural integrity during cycling. Furthermore, the strong interaction between Si NPs, amorphous carbon layer, and Ti 3 C 2 T x NSs established a highly efficient conductive network for efficient lithium storage. As expected, the Si@C-MX anode exhibited high reversible capacities of 1036.2 mAh g À1 at 0.1 A g À1 , and even 451.9 mAh g À1 at a large current density of 10.0 A g À1 and ultra-high stable cycling performance with a retained capacity of 803.8 mAh g À1 after 1000 uninterrupted charge-discharge cycles at 0.5 A g À1 . Additionally, the Si@C-MX//NCA full cells delivered a promising energy density of 371.8 Wh kg À1 based on the whole device at a power density of 123.9 W kg À1 , which convincingly highlights its promising application in advanced LIBs. The design here provided a competitively commercialized anode platform for next-generation LIBs.

Experimental Section
Synthesis of Ti 3 C 2 T x NSs: The Ti 3 C 2 T x NSs was typically prepared by selective etching of Al from Ti 3 AlC 2 (11 technology, co., LTD) with HCl and LiF, according to our previous work. [40] Synthesis of Hydroxylated Si NPs: The commercial Si NPs (around 100 nm, Xinnai metallic materials, co., LTD) was activated by Piranha solution (H 2 SO 4 /H 2 O 2 , 7/3, v/v) at 80°C for 40 min.
Synthesis of Si@C-MX: Firstly, 100 mg of the CS was dissolved in 200 mL of the acetic acid aqueous solution with the concentration of 1% v/v. Then 50 mg of hydroxylated Si NPs were added to the above solution under stirring. After mixing thoroughly, 30 mg of Ti 3 C 2 T x NSs were dispersed in the mixture and stirred for 12 h. Finally, the uniform solution was spray-drying by a B-290 spray-dryer (BUCHI, Switzerland). Typically, the above solution was pumped to the spray-drying atomizer through the feed pipe, then atomized into micron size dispersed droplets. Under high temperature air flow (the inlet temperature is 220°C), the droplets are dried and contracted into the Si@CS-MX powder, which was collected by cyclone separator. After further annealed at 600°C for 2 h under the H 2 /Ar flow at a ramp rate of 1°C min À1 , the Si@CS-MX was completely converted into the Si@C-MX.
Synthesis of Si@C and Si-MX: The synthesis of Si@C and Si-MX was carried out with the same procedure as that for the Si@C-MX, just with the exception of single-phase CS or Ti 3 C 2 T x NSs added in the synthesis.
Material Characterizations: The crystal structures of the samples were characterized by XRD (Cu Kα radiation, Rigaku Ultima IV). Raman spectroscopy was performed by Horiba LabRAM HR Evolution (514.5 nm laser). Morphologies and microstructures of samples were carried out using Field-emmision scanning electron microscopy (FESEM, JEOL-6300 F, 15 kV), TEM/high-resolution TEM (HRTEM), and sanning TEM (STEM) (JEOL JEM 2100 system) with energy dispersive X-Ray spectroscopy (EDS). Specific chemical states of elements in the samples were characterized by X-Ray photoelectron spectroscopy (XPS, Thermo, Escalab 250xi).
Computational Method: The DFT calculations were carried out with the Perdew-Burke-Ernzerhof (PBE) function within the generalized gradient approximation (GGA) using the Vienna ab initio simulation package (VASP). [41][42][43] The projector augmented wave (PAW) method was used to describe the electron-ion interaction. The van der Waals (vdW) interaction was described using the DFT-D3. [44] The relatively small supercell, i.e., ffiffi ffi 3 p Â ffiffi ffi 3 p of Ti 3 C 2 and 2 Â 2 of graphene, were choosen for calculation. In addition, the lattice mismatch between Ti 3 C 2 and graphene was within 6.6 %. A plane wave basis set with a cut-off energy of 700 eV was used. The Brillioun zone was represented by a Monkhorst-Pack mesh of 11 Â 11 Â 1 k-points. [44][45][46] The convergence criterion of Hellmann-Feynman forces and total energy were 0.01 eV Å À1 and 10 À7 eV, respectively. A vacuum region of about 15 Å was adopted to avoid unnecessary interactions.
Electrochemical Measurements: Coin-type cells (CR2032) were fabricated in an argon-filled glove box (MBRAUN, Germany) with oxygen and moisture contents being under 0.5 ppm to evaluate the electrochemical performance of the electrodes. The slurry consisted of the active materials (Si@C-MX, Si@C or Si-MX, 70 wt%), acetylene black (20 wt%), and sodium carboxymethyl cellulose (CMC, 10 wt%) with the water as a solvent. After being coated on the copper foil, the slurry was dried in a vacuum oven at 110°C. The cathode slurry was fabricated by mixing 80 wt% of commercial LiNi 0.8 Co 0.1 Al 0.1 O 2 (NCA), 10 wt% of acetylene black, and 10 wt% of polyvinylidene fluoride with the N-methyl-2 pyrrolidone, and coated on the aluminum foil then dried at 110°C for 11 h under vacuum. The electrolyte was a solution of 1 M LiPF 6 in the mixture of an ethylene carbonate and diethyl carbonatein (1/1, v/v). Lithium foil and Celgard 2400 polyprop membrane were used as the counter electrode and separator, respectively. The GCD were carried out using a battery testing system (Land CT2001A, Wuhan, China). CV and EIS analysis from 100 kHz to 100 mHz were measured by electrochemical workstation (IviumStat.h, The Netherlands).

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.