Surface functionalized 3D carbon fiber boosts the lithium storage behaviour of transition metal oxide nanowires via strong electronic interaction and tunable adsorption energy

Abstract

The Li-ion storage properties of transition metal oxide (TMOs) electrodes such as Li-ion intercalation-based electrodes are usually enhanced by hybridizing with 3D carbon scaffolds. However, understanding of the large variation in performance enhancement is rarely reported. As a proof of concept, intercalation reaction-based TMO (V₂O₅ and TiO₂) nanowires were hybridized with two types of 3D carbon scaffolds (namely pristine carbon fiber cloth, CFC and porous N-doped CFC, PNCFC). Theoretical calculation predicts that the PNCFC@TMO hybrids displayed reasonably lower adsorption energy towards easier Li-ion intercalation than those of CFC@TMOs. Electrochemical properties further disclosed that PNCFC-based hybrids exhibit the best lithium storage performance. Furthermore, in situ Raman, XPS and charge redistribution studies not only decipher that strong electronic interaction exists between PNCFC and TMOs but also consistently affirm that such interaction is ascribed to the shift of the p-adsorption energy, facilitating rapid kinetics and leading to improved Li storage properties.

---

New concepts

Poor kinetics has rendered Li-ion intercalation reaction-based transition metal oxides to be hybridized with highly three-dimensional (3D) carbon-based current collectors. It has been found that the Li-storage performance of these transition metal oxides usually increases with the support of the 3D carbon-based current collectors. The most common reasons for the performance enhancement are habitually ascribed to “all-in-one” integrated advantages arising from the 3D conductive network and high surface area. Less efforts have been devoted to the search for possible mechanisms that attract such enhanced Li-storage performance. In this study, both theoretical and experimental analyses were utilized to address the horizons of enhanced Li-storage properties in 3D carbon scaffold and intercalation reaction-based transition metal oxide hybrids. Li-storage test, in situ Raman, and XPS analyses and density functional theory (DFT) calculations consistently reveal that the enhanced performance is attributed to the alteration in the p-adsorption energy, which helps to facilitate rapid Li-ion diffusion. This present study is beneficial for understanding the Li-ion intercalation reaction storage mechanism towards designing other electrodes for LIBs and beyond.

Introduction

Li-ion intercalation reaction (LiIR)-based electrodes are one of the most promising electrode materials for lithium ion batteries (LIBs) because they offer the advantages of low volume expansion, stable solid electrolyte interface layer formation, little irreversible capacity loss and no electrolyte decomposition.^1–5 The intercalation reaction of TMOs is shown as follows:⁶

MO + xLi⁺ + xe⁻ ⇌ Li_xMO (1)

Transition metal oxides (TMOs)^2,7–9 such as Ti oxide-based^10,11 electrodes especially TiO₂ and V oxide-based^12–15 electrodes and especially V₂O₅ usually exhibit such LiIR mechanism. Meanwhile, these TMOs exhibit either poor electrochemical stability (V₂O₅) or poor electrical conductivity (V₂O₅ and TiO₂) leading to loss in capacity and low rate capability, respectively.^16,17 In order to improve the dual dilemma of both V₂O₅ and TiO₂, one of the common strategies employed is to design self-supporting electrode materials by the direct growth of the active materials on 3D carbon scaffolds (e.g. carbon cloth, graphene, carbon nanotubes and so on).^18–22 Such strategy could automatically prevent the intercalation of Li-ions into the carbon scaffold (Li-ion usually intercalates in carbon-based material at a working potential below 1.0 V) and make it easy to determine the capacity contribution of the active V₂O₅ and TiO₂ materials. Previous literature has shown that the performance enhancement of the carbon scaffolds and TMO hybrids can be attributed to the conductive 3D network and high surface area that allows rapid ion/electron transfer, enhanced conductivity and improved mechanical flexibility.^19,23 However, despite the enhanced storage properties,²⁴ the understanding of the intrinsic properties of the carbon scaffold and active electrodes is rarely reported.^25,26 In addition, less or no attention is given to the theoretical insight into the effect of such hybridization towards enhanced electrochemical activity.²⁷

In this work, density functional theory (DFT) calculations were first studied to probe the intercalation of Li-ions into V₂O₅ on pristine carbon fiber cloth (CFC) and porous N-doped carbon fiber cloth (PNCFC). The PNCFC-based sample requires lower adsorption energy to complete the intercalation process than the CFC counterparts. CFC was selected among the commonly reported carbon scaffolds owing to its good mechanical strength, stability, and wide application in energy storage research.^24,27–29 The electrochemical performance of PNCFC@V₂O₅ presents superior Li storage properties over CFC@V₂O₅. In situ Raman spectra and XPS consistently reveal that there exist strong electronic interactions between PNCFC and V₂O₅, which is valuable towards improving Li storage kinetics. Such strong electronic interaction allows the downshift of the p-band adsorption center, which is favorable for enhancing the Li storage performance. More interestingly, based on the understanding of PNCFC@V₂O₅, theoretical and experimental analyses further verify that a hybrid of TiO₂ and PNCFC can also display a similar phenomenon by modulating the p-orbital adsorption energy.

Results

Theoretical predictions

Previous literature has reported that series of TMOs directly synthesized on 3D carbon scaffolds usually relate the enhanced performance to the 3D network.^19,29,30 However, theoretical studies on the intercalation process of carbon compared to the active materials in their powdery forms are rarely studied. DFT calculations were first employed to investigate the optimum Gibb's free energy change (ΔG_H*) required for Li-ion diffusion through pristine V₂O₅ and CFC@V₂O₅.^31,32 The diffusion barrier during a diffusion process of a Li-ion is very essential for understanding the electrode behaviour and the reasons for enhanced Li-ion storage performance.^33,34 The optimized cluster structures of V₂O₅ and CFC@V₂O₅ for the DFT calculations can be found in the ESI,† Fig. S1 and S2 and the free energy change diagrams of V₂O₅ and CFC@V₂O₅ are shown in Fig. 1a and b, respectively. The free energy change diagrams for these three electrodes are derived based on theoretical intercalation of Li-ions (i.e. lithiated state) into these electrodes. For example, Fig. S3 (ESI†) shows the lithiated state of V₂O₅, while Fig. S4 and S7 (ESI†) also show the lithiated state of both CFC@V₂O₅ and PNCFC@V₂O₅, respectively. Li-ions at the initial state (Fig. S3(I) and S4(I), ESI†) account for ΔG_H* of 0.0 eV in both electrodes (Fig. 1a(i) and b(iii)). At the V₂O₅ surface, CFC@V₂O₅ requires a lower diffusion barrier of 0.14 eV (Fig. S4(II), ESI† and Fig. 1b-ts-3), whereas the ΔG_H* is much higher in V₂O₅ (0.19 eV, Fig. S3(II), ESI† and Fig. 1a-ts-1). Upon complete Li-ion intercalation into the lattice of the electrodes forming Li_xV₂O₅ (Fig. S3(III), ESI†), V₂O₅ will still have to overcome an energy barrier of 1.81 eV (Fig. 1a(ts-2)). The lower diffusion barrier (ΔG_H*) of 0.95 eV (Fig. 1b(ts-4) and Fig. S4(III), ESI†) for CFC@V₂O₅ compared to that of pristine V₂O₅ indicates that rapid Li-ion transportation and enhanced kinetics throughout the intercalation process could result in superior performance of CFC@V₂O₅ over pristine V₂O₅. Finally, the energy barrier of V₂O₅ and CFC@V₂O₅ returned to 0.0 eV (Fig. 1a(ii) and b(iv)) after the entire intercalation process (Fig. S3(IV) and S4(IV), ESI†), respectively.

Fig. 1 Calculated migration ΔG_H* to Li-ion diffusion through (a) V₂O₅, (b) CFC@V₂O₅ and (c) PNCFC@V₂O₅ from DFT calculations.

Previously, we have shown that the surface of CFC can be functionalized with nitrogen (N) atoms, high porosity, and better conductivity, which leads to high surface area and also serves as a promising anode material for LIBs.^27,35 When the porous N-doped functionalized CFC (denoted as PNCFC) was used as a current collector for the growth of conversion reaction-based TMOs such as NiCo₂O₄, experimental and theoretical analyses confirmed that there exists an electronic interaction between PNCFC and NiCo₂O₄, which is responsible for achieving high capacity of the hybrid.²⁷ Additionally, V₂O₅ nanobelt³⁶ and V₂O₅ nanosheet³⁷ cathodes that were synthesized on graphene foam and carbon nanotubes (CNTs) displayed enhanced performance due to the 3D network and harmonious integration.^36,37 Motivated by this concept, we utilized the PNCFC as a current collector for the direct synthesis of V₂O₅ nanowires (denoted as PNCFC@V₂O₅ NWs) and compare its performance with those of CFC@V₂O₅ NWs.

Theoretical predictions on a PNCFC@V₂O₅ hybrid was conducted based on the optimized structure model for PNCFC@V₂O₅ (Fig. S5, ESI†). The enlarged snapshot of covalent bonds between C (a functional group of PNCFC) and O (from V₂O₅), and between O (functional group of PNCFC) and V (from V₂O₅) are depicted in Fig. S6a and b (ESI†), respectively. This bonding shows that the functional groups revealed a close relationship between individual elements in the electrode. Compared to CFC@V₂O₅, the optimum Gibb's free energy change (ΔG_H*) required for Li-ion transportation in PNCFC@V₂O₅ is shown in Fig. 1c. During Li-ion intercalation in PNCFC@V₂O₅ (Fig. S7, ESI†), the ΔG_H* at ts-5 and ts-6 are 0.06 and 0.35 eV, also indicating the complete lithiated state of V₂O₅, and significantly lower than those of CFC@V₂O₅ and pristine V₂O₅.

The result further suggests that the support of a functionalized 3D carbon scaffold with TMOs could not only hasten the Li-ion diffusion pathway and enhance the kinetics but also lower the adsorption energy towards achieving high performance LIBs.

Synthesis and morphological characterization

The scanning electron microscopy (SEM) images of the two current collectors, CFC and PNCFC (highly networked self-supported current collectors) are shown in Fig. S8a and b (ESI†), respectively. The direct growth of V₂O₅ NWs on the CFC and PNCFC self-supported materials was accomplished through hydrothermal and air-annealing treatments as previously reported.¹⁹ Schematic illustrations of the V₂O₅ NW fabrication process can be seen in Scheme S1 (ESI†) (details about the synthetic method can be found in the Experimental section). Firstly, a 24 cm² piece of CFC was converted to PNCFC hydrothermal reaction, thermal reduction and etching treatments (Scheme S1(I), ESI†).³⁸ The SEM image clearly shows the porous and rough nature of the PNCFC surface as seen in Fig. 2a. Transmission electron microscopy (TEM) analysis further reveals an exfoliated and porous surface on the PNCFC (Fig. 2a inset). For the CFC current collector, the approximate thickness of a single fiber is 9.6 μm (Fig. S9a, ESI†), which reduces to 7.7 μm after modification to PNCFC (Fig. S9b, ESI†).

Fig. 2 Morphological characterization. (a) SEM image of PNCFC showing the exfoliated porous surface. Inset is the corresponding TEM image. (b) SEM image of PNCFC@V₂O₅ showing V₂O₅ NWs. (c) SEM image of CFC@V₂O₅ showing the contact between CFC and V₂O₅. (d) SEM image of PNCFC@V₂O₅ showing V₂O₅ NWs strictly adhered to the PNCFC surface. (e) TEM image of PNCFC@V₂O₅ NWs showing the core-double shell-like architecture. Elemental mapping of PNCFC@V₂O₅ showing distribution of the elements from (f) HAADF, (g) C, (h) V, (i) O, and (j) N. (k) TEM image of PNCFC@V₂O₅ NWs and (l) HRTEM image of PNCFC@V₂O₅ NWs displaying the lattice fringes.

Previous reports have demonstrated that lightweight substrates or current collectors with porous morphology could serve as favorable scaffolds for the growth of active materials, while reducing the weight of the entire cell^24,36,39 and also contributed to the flexible properties of the electrodes.^24,35,40,41 Thus, VO_x NWs (Fig. S10, ESI†) were synthesized on both CFC and PNCFC current collectors and annealed at 400 °C in the air to obtain CFC@V₂O₅ (Scheme S1(II), ESI†) and PNCFC@V₂O₅ NWs (Scheme S1(III), ESI†), respectively.

According to the XRD pattern, both samples were characterized with V₂O₅ peaks (JCPDF Card #65-0131) but there's no significant difference in the phases of the two samples (Fig. S11, ESI†). Both the SEM images for CFC@V₂O₅ and PNCFC@V₂O₅ consist of interwoven NWs with diameters in the range of 300–800 nm (Fig. S12, ESI† and Fig. 2b), respectively. Fig. 2c clearly depicts that V₂O₅ NWs were successfully coated on the smooth surface of CFC, while Fig. 2d affirms the growth of the NWs firmly adhered to the PNCFC current collector. Further information regarding the morphology of the adhesiveness of V₂O₅ NWs and the PNCFC current collector interface was accomplished by TEM analyses. Interestingly, the TEM images revealed three-layer (core–double shell) architecture of the sample (Fig. 2e and Fig. S13, ESI†), consisting of an inner-core CFC layer, an exfoliated carbon middle-shell layer and a V₂O₅ NW outer-shell layer. Energy dispersive X-ray spectroscopy (EDS) elemental mapping collected from Fig. 2e with its corresponding HAADF (Fig. 2f) displayed the distribution of C (Fig. 2g-red), V (Fig. 2h-yellow) and O (Fig. 2i-blue) and N (Fig. 2j-orange). Fig. 2g discloses that the intensity of C (red colour) decreases from bulk (top side) to the surface (down side) and the presence of V and O at the outermost shell layer (Fig. 2h) also confirms the formation of V₂O₅. The results obtained from the EDS analyses further certified that the V₂O₅ NWs were successfully coated on a PNCFC current collector demonstrating a core–double shell design. Additionally, the TEM image collected from the PNCFC@V₂O₅ NW surface (Fig. 2k) showed the single crystalline nature (Fig. S14, ESI†) and lattice spacing of 0.34 nm (Fig. 2l), which corresponds to the (110) plane of V₂O₅ and also affirms the successful formation of V₂O₅ on a PNCFC substrate.

Electronic structure characterization

The electronic structure and valence states of both CFC@V₂O₅ and PNCFC@V₂O₅ and more information on the relationship between V₂O₅ and the two types of current collectors were studied using X-ray photoelectron spectroscopy (XPS). The XPS survey spectra of CFC@V₂O₅ and PNCFC@V₂O₅ can be found in Fig. S15a (ESI†). C 1s of the PNCFC@V₂O₅ exhibited a broad signal for the C–C bond and an additional broad peak compared to the CFC@V₂O₅ sample, which suggests some extent of modifications in the modified CFC coated samples (Fig. 3a).^35,42 Moreover, a peak at ∼400 eV in the N 1s spectra of PNCFC@V₂O₅ is totally absent in the CFC@V₂O₅ sample (Fig. 3b). This confirmed that the PNCFC@V₂O₅ sample is N-doped. The N-doping can be obtained during thermal reduction of the Ni(OH)₂·0.75H₂O precursor in a N₂ atmosphere.^27,38 Compared to the V 2p XPS spectra of CFC@V₂O₅, the V 2p XPS spectra of both samples can be deconvoluted into two synthetic peaks (516.3 eV for V⁴⁺ and 517.6 eV for V⁵⁺) (Fig. 3c(I)–(II)) confirming that the V₂O₅ samples are mixed-valence in nature.^43–47 Interestingly, the V⁵⁺ V 2p XPS spectra of CFC@V₂O₅, shifted to a lower binding energy (ca. 0.36 eV) in the PNCFC-based sample, whereas, the V⁴⁺ V 2p XPS spectra shifted to a lower binding energy (ca. 0.36 eV) for the PNCFC-based sample.

Fig. 3 (a) C 1s, (b) N 1s and (c) V 2p_3/2 XPS spectra of CFC@V₂O₅ and PNCFC@V₂O₅.

This phenomenon indicates a change in the electronic bonding environment state, justifies strong electronic interaction between PNCFC and V₂O₅ and suggests rapid transfer of electrons in the PNCFC@V₂O₅ sample.^27,48 Moreover, the higher shift in the binding energy of the V⁴⁺ suggests stronger electronic interaction on the low valence state of PNCFC@V₂O₅, which could be beneficial for enhancing the sample conductivity and kinetics.⁴⁹ Additionally, the percentage ratio of the V⁴⁺/V⁵⁺ in the samples increased from 38% in CFC@V₂O₅ to 47% in the PNCFC@V₂O₅ sample, meaning that a higher composition of V⁴⁺ improved the electronic interactions between V₂O₅ and PNCFC, which could also be an avenue for improving the electrochemical performance of PNCFC@V₂O₅.⁴³

It has been previously proposed that the valence of vanadium oxide composites can be tailored, and could be a promising way to achieve high capacity, good conductivity and good cycling stability.^44,45 For instance, self-doping of V⁴⁺ into V₂O₅ nanoflakes exhibits high structural stability, and long-term and fast reversible Li storage capability resulting from the presence of V⁴⁺.⁴³ Likewise V₂O₅ xerogels showed improved lithium storage performance benefiting from the multi-vanadium oxidation states and electrical conductivity.⁵⁰ Thus, we proposed that the high composition of V⁴⁺ in PNCFC@V₂O₅ arises from the electronic interaction between V₂O₅ and PNCFC, and this could eventually lead to improving the electrochemical performance of PNCFC@V₂O₅ over the CFC@V₂O₅ sample. According to Fig. S15b (ESI†), the O 1s XPS spectra of PNCFC@V₂O₅ are slightly different from that of CFC@V₂O₅, also confirming that characteristic functional groups from the PNCFC current collector and such functional groups are beneficial in improving electrode kinetic properties.^27,51 Based on the ex situ Raman spectroscopy analyses, the entire Raman peak below 1000 cm⁻¹ corresponds with V₂O₅ Raman peaks reported in other literature (Fig. S16a, ESI†).⁵² More importantly, the Raman spectra present a D : G peak intensity ratio of 1.48 for PNCFC@V₂O₅ and also larger than that of the CFC@V₂O₅ (1.32) (Fig. S16b, ESI†). In addition, the G-band of the PNCFC@V₂O₅ sample becomes broader further affirming modification of the treated CFC coated sample. This can be attributed to the porous exfoliated surface of the PNCFC substrate that consists of some surface disorder and the presence of functional groups induced by etching, which is in agreement with the O 1s XPS spectra.^35,53

Li-storage properties

Li storage behaviour and performance for both CFC@V₂O₅ and PNCFC@V₂O₅ positive electrodes was studied in a coin cell. The Li-ion insertion–deinsertion reactions in V₂O₅ can be seen from the 5th cyclic voltammetry (CV) curves for both electrodes (Fig. S17, ESI†) and the corresponding reaction mechanism can be seen as V₂O₅ + xLi⁺ + xe⁻ ⇌ Li_xV₂O₅.³⁶ Regular peaks of V₂O₅ can be found in both electrodes.⁵⁴ The PNCFC@V₂O₅ electrode shows stronger current density peaks than the CFC@V₂O₅ electrode and also shifts to the higher voltage region, which implies that the PNCFC@V₂O₅ electrode has a higher electrochemical reactivity and enhanced conductivity compared to the CFC@V₂O₅.³⁶ Electrochemical impedance spectroscopy (EIS) analysis was carried out to gain information on the kinetics of both CFC@V₂O₅ and PNCFC@V₂O₅ electrodes.

The Nyquist plot is characterized with a semi-circle and diffusion slope representing the charge transfer resistance (R_ct) and Warburg impedance, respectively.¹⁹ The smaller the R_ct, the more the increase there is in the electron transfer rate of V₂O₅ in the PNCFC@V₂O₅ cell. As seen in Fig. 4a, PNCFC@V₂O₅ displays a smaller R_ct value (86 Ω) than that of CFC@V₂O₅ (122 Ω), confirming the superior conductivity of the PNCFC@V₂O₅ electrode over CFC@V₂O₅. As shown in Fig. 4b, the discharge capacity of PNCFC@V₂O₅ is 268 mA h g⁻¹ after the 10th electrochemical cycle, while that of CFC@V₂O₅ is lower (233 mA h g⁻¹). It should be noted that energy efficiency (%) is a vital parameter for the practical development of new electrode materials. The energy efficiency of a battery usually depends on the overpotentials of charging and discharging profiles and can be determined according to the formula below;⁵⁵

The energy density is determined by the area under the charge–discharge curves. According to Fig. 4b, the energy efficiency of CFC@V₂O₅ and PNCFC@V₂O₅ is approximately 95%, which can still be maximized for practical interest.⁵⁵Fig. 4c demonstrates the rate capability performance of PNCFC@V₂O₅ and CFC@V₂O₅ electrodes. With increasing current density from 0.3C to 17C, PNCFC@V₂O₅ electrodes continuously displayed superior performance over the CFC@V₂O₅ electrode. Unlike the CFC@V₂O₅ electrode with capacity retention of 69% at 17C, the PNCFC@V₂O₅ electrode performance delivered a reversible capacity of 233 mA h g⁻¹ (81% capacity retention) at a current rate of 17C. A clear difference can be observed with the PNCFC@V₂O₅ electrode retaining a specific capacity of 243 mA h g⁻¹ (85% capacity retention) with 100% coulombic efficiency, while its counterpart could only deliver 192 mA h g⁻¹ (75% capacity retention) at the same testing current density of 17C for 500 cycles (Fig. 4d). It can be observed that both electrodes show fast capacity decay in the first 20 cycles. Generally, the main problem of bulk V₂O₅ as an electrode material for LIBs is its poor cycling stability and rate capability. Thus, the fast capacity decay in the first 20 cycles for both the electrodes is a common phenomenon to V₂O₅. The SEM images of the PNCFC@V₂O₅ electrode after a stability test confirmed that the V₂O₅ NWs were still strongly adhered to the surface of the PNCFC current collector (Fig. S18, ESI†). These results further verify the exemplary stability of the PNCFC@V₂O₅ electrode.

Fig. 4 Li storage properties. (a) The Nyquist plots, (b) charge–discharge curves at a current density of 2.0C, (c) rate capability and (d) cycling stability at a current density of 2.0C of both CFC@V₂O₅ and PNCFC@V₂O₅. (e) In situ Raman spectra of PNCFC@V₂O₅ at different intercalated voltages. (f) The plot of the I_D/I_Gvs. intercalated voltages of CFC@V₂O₅ and PNCFC@V₂O₅.

In situ Raman analysis

Additional analysis of the intercalation of Li-ions was carried out by in situ Raman spectra for the two electrodes. Raman analysis is an effective strategy to study the effect of defect carbon and graphitic carbon in a sample.²⁷ Detail information on the set-up can be found in Fig. S19 (ESI†). The response of the current collector D and G peaks towards Li-ion intercalation at different voltages can be effective for determining the impact of the different CFC current collectors. The Raman spectra collected for CFC@V₂O₅ showed a non-uniform variation in the carbon D and G bands at different Li-ion voltages (Fig. S19e, ESI†), while that of PNCFC@V₂O₅ showed nearly uniform variations in the D and G bands (Fig. 4e). The I_D/I_G values of the electrodes are plotted against the intercalation at different voltages (Fig. 4f) and the I_D/I_G value of the PNCFC@V₂O₅ cell is nearly the same at different Li-ion voltages, but varies and is non-uniform for CFC@V₂O₅. This result interprets that the V₂O₅ NWs are not strictly attached to the CFC surface leading to instability of the CFC@V₂O₅ electrode. However, the I_D/I_G values for PNCFC@V₂O₅ indicate that there are strong interfacial interactions between the shell layer of the PNCFC and V₂O₅ NWs. These interactions are highly stable regardless of the voltage. Such strict adherence between the PNCFC and V₂O₅ NWs allows easy diffusion of Li-ions and transportation of electrons into the entire electrode cavity. Both the theoretical analyses and experimental results are consistent and also account for the higher capacity retention and better stability of the PNCFC@V₂O₅ electrode over the CFC@V₂O₅.

Discussion

Charge redistribution and adsorption energy capability

The electronic interaction between V₂O₅ and PNCFC demonstrates a crucial role in enhancing the Li-storage properties of PNCFC@V₂O₅. DFT calculations were further employed to investigate the charge distribution relationship between V₂O₅ and carbon scaffolds. The optimized model structures for CFC@V₂O₅ and PNCFC@V₂O₅ are displayed in Fig. S2 and S5 (ESI†), respectively. Fig. 5a distinctly displayed the natural bonding orbital (NBO) charge distribution in the optimized CFC@V₂O₅ and PNCFC@V₂O₅ obtained from DFT calculations.

Fig. 5 (a) NBO charge redistribution of V₂O₅, CFC@V₂O₅ and PNCFC@V₂O₅. (b) DOS plots of CFC and PNCFC and their corresponding p-bands. (c) DOS plots of CFC@V₂O₅ and PNCFC@V₂O₅ and their corresponding p-bands.

The NBO charge distribution of pristine V₂O₅ was also introduced for comparison (optimized model structure in Fig. S1, ESI†). The charge distribution on PNCFC@V₂O₅ shows that the positive charge on V is less electropositive, while the negative charge on O is less electronegative, in the order of PNCFC@V₂O₅ > CFC@V₂O₅ > V₂O₅. However, the positive charge on C of PNCFC@V₂O₅ is more electropositive. These results indicate that the surface functionalization of PNCFC could change the electron density charges on the entire electrode, leading to excellent electronic interactions between PNCFC and V₂O₅ compared to that of CFC@V₂O₅.²⁷ To obtain further understanding on the electronic interaction between the hybrids, the analysis of the density of states (DOS) related to the Fermi level (E_f) is studied and the corresponding s, p, d, f and sum bands DOS are shown in Fig. S20 (ESI†). Firstly, in all the samples, the p-band is more intense than the s, d and f-bands and nearly the same as that of the sum-adsorption energy, indicating that the main Li-ion intercalation process is related to the p-adsorption energy. The p-adsorption energy center (E_p) relative to E_f of CFC is −1.27 eV and that of PNCFC is −6.43 eV (Fig. 5b). The results indicate that E_p of PNCFC is farther away from the E_f, facilitating the energy band-gap to reduce and allowing the p-band to undergo a distinct shift in PNCFC than CFC. The hybridization of PNCFC with V₂O₅ could definitely require lower adsorption energy than that of CFC-based hybrids. Indeed, the E_p in PNCFC@V₂O₅ also depicts a shift towards higher E_f (−10.52 eV) than CFC@V₂O₅ (−10.21 eV) (Fig. 5c) due to the increased bonding strength of the PNCFC-based hybrid, which originates from the easy release and pulling of electrons from PNCFC@V₂O₅. Hence, the energies of the charge density and valence band electrons will increase, enabling rapid electron transfer towards promoting fast Li-ion intercalation kinetics for enhanced Li-ion storage properties. Finally, both the NBO charge distribution and DOS results indicate changes in the bonding or electronic states and also in accordance with the XPS, finally affirming that there exist stronger electronic interactions between PNCFC and V₂O₅ than pristine CFC and V₂O₅.

TiO₂-based anode: electrochemical properties and mechanism studies

To obtain further understanding of the LiIR-based mechanism, we also utilized a TiO₂ anode, which is also well known for its poor electrical conductivity leading to its slow kinetics and poor rate capability. Similar to V₂O₅, theoretical predictions on TiO₂ hybridized with CFC and PNCFC were studied. DFT calculations conducted to determine the ΔG_H* required for Li-ion intercalation through CFC@TiO₂ and PNCFC@TiO₂. The optimized CFC@TiO₂ and PNCFC@TiO₂ cluster structures used for DFT calculations are presented in Fig. S21 and S22 (ESI†), respectively. As a Li-ion passes across the optimized CFC@TiO₂ structure (Fig. S23, ESI†), the ΔG_H* required for the transportation of Li-ions throughout CFC@TiO₂ is 1.11 eV (Fig. 6a). However, the maximum ΔG_H* required is 0.51 eV for PNCFC@TiO₂ to transport the Li-ion through the entire PNCFC@TiO₂ structure (Fig. S24, ESI†), which is also lower than that of CFC@TiO₂. These results also correlate with that of V₂O₅ in Fig. 1. NBO charge distribution studies were also performed for the TiO₂-based electrodes. Pristine TiO₂ was also employed for comparison (optimized structure in Fig. S25, ESI†). The enlarged snapshot of covalent bonds between O (a functional group of ECC) and Ti (from TiO₂) and an enlarged snapshot of covalent bonds between C (a functional group of ECC) and O (from TiO₂) are shown in Fig. S26 (ESI†).

Fig. 6 Properties of CFC@TiO₂ and PNCFC@TiO₂. (a) Calculated migration ΔG_H* to Li-ion diffusion through CFC@TiO₂ and PNCFC@TiO₂. (b) NBO charge redistribution of CFC@TiO₂ and PNCFC@TiO₂. (c) SEM images of PNCFC@TiO₂ NWs. (d) Rate performance and (e) plot of the I_D/I_Gvs. intercalated voltages for CFC@TiO₂ and PNCFC@TiO₂. (f) DOS plots of TiO₂, CFC@TiO₂ and PNCFC@TiO₂ and their corresponding p-bands.

According to the NBO charge redistribution in Fig. 6b, compared to CFC@TiO₂ and pristine TiO₂, the charge value on the positively charged Ti decreases, the positive charge on C increases and O is negatively charged in PNCFC@TiO₂, indicating stronger electronic interactions between PNCFC and TiO₂ of PNCFC@TiO₂. Hence, we hydrothermally coated rutile-type TiO₂ nanowires (TiO₂ NWs) (Fig. S27, ESI†) on CFC (CFC@TiO₂ NWs) and PNCFC (PNCFC@TiO₂ NWs) according to our previous work (Fig. 6c).^11,16 Because TiO₂ is known for its poor electrical conductivity, the rate capability performance was investigated at different current densities and in the voltage range of 1.0–3.0 V (vs. Li/Li⁺), where the PNCFC@TiO₂ exhibited superior storage properties to CFC@TiO₂ (Fig. 6d). In situ Raman spectra also show that the PNCFC@TiO₂ is more stable than CFC@TiO₂ (Fig. 6e). Furthermore, the plot of I_D/I_G values against intercalation voltages collected from the whole results obtained for the TiO₂-based electrodes are in-line with those of the V₂O₅-based electrodes. Moreover, the E_p of PNCFC@TiO₂ is −5.67 eV, which is also farther from the E_f compared to CFC@TiO₂ (Fig. 6f), indicating a shift in the E_p, and could facilitate rapid Li-ion kinetics in PNCFC@TiO₂ than CFC@TiO₂. Hence, these results did not clearly confirm that the Li-ion storage properties of LiIR-based TMO electrodes can be effectively modulated by tuning the p-adsorption energy level but also deciphered understanding of the large variation in performance enhancement in carbon scaffold-TMO electrode hybrids.

Conclusions

In summary, we systematically demonstrated the theoretical and experimental insights into the storage mechanism and large variation in the Li-ion storage performance of different carbon scaffolds as current collectors for LiIR-based TMO electrodes. LiIR-based electrodes, such as V₂O₅ and TiO₂ NWs, were directly synthesized on two different carbon current collectors with different surface properties (CFC and PNCFC). DFT calculations revealed that Li-ion adsorption occurred at a lower adsorption energy (ΔG_H*) level of 7.75 and 4.64 eV for PNCFC@V₂O₅ and PNCFC@TiO₂, respectively compared to those of CFC@V₂O₅ (18.21 eV), V₂O₅ (29.04 eV), CFC@TiO₂ (6.67 eV) and TiO₂ (11.34 eV). Electrochemical performance confirmed that the V₂O₅ and TiO₂ NWs coated on PNCFC exhibited better Li-storage activity than those hybridized on CFC. Additionally, in situ Raman spectra analysis revealed that PNCFC could serve as a better scaffold for the electrodes in enhancing the cycling stability and rate capability of the PNCFC-based electrodes compared to CFC. XPS, NBO charge distribution and DOS analyses showed that the PNCFC-based electrodes are attributed to stronger electronic interaction between the PNCFC and TMOs due to the shift of p-adsorption energy level. This work deciphers insight into the Li-storage mechanism and performance modulation of carbon-based current collectors and LiIR-based electrodes.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21875292 and 51525202), the National Key Research and Development Program of China (2016YFA0202604), the Fundamental Research Funds for the Central Universities (17lgjc36) and Science Starting Foundation of Hunan University (531118010182).

References

1. T. Zhou, Y. Zheng, H. Gao, S. Min, S. Li, H. K. Liu and Z. Guo, Adv. Sci., 2015, 2, 1500027.
2. V. Aravindan, Y.-S. Lee and S. Madhavi, Adv. Energy Mater., 2015, 5, 1402225.
3. S. Cao, Z. Xue, C. Yang, J. Qin, L. Zhang, P. Yu, S. Wang, Y. Zhao, X. Zhang and R. Liu, Nano Energy, 2018, 50, 25–34.
4. S. Joshi, Q. Wang, A. Puntambekar and V. Chakrapani, ACS Energy Lett., 2017, 2, 1257–1262.
5. Z. Yao, X. Xia, D. Xie, Y. Wang, C.-a. Zhou, S. Liu, S. Deng, X. Wang and J. Tu, Adv. Funct. Mater., 2018, 28, 1802756.
6. X. Xia, D. Chao, C. F. Ng, J. Lin, Z. Fan, H. Zhang, Z. X. Shen and H. J. Fan, Mater. Horiz., 2015, 2, 237–244.
7. X. Xia, S. Deng, S. Feng, J. Wu and J. Tu, J. Mater. Chem. A, 2017, 5, 21134–21139.
8. F. Wang, X. Wang, Z. Chang, Y. Zhu, L. Fu, X. Liu and Y. Wu, Nanoscale Horiz., 2016, 1, 272–289.
9. C. Wang, L. Wu, H. Wang, W. Zuo, Y. Li and J. Liu, Adv. Funct. Mater., 2015, 25, 3524–3533.
10. V. Augustyn, E. R. White, J. Ko, G. Grüner, B. C. Regan and B. Dunn, Mater. Horiz., 2014, 1, 219–223.
11. M.-S. Balogun, M. Yu, C. Li, T. Zhai, Y. Liu, X. Lu and Y. Tong, J. Mater. Chem. A, 2014, 2, 10825–10829.
12. Y. Dai, Q. Li, S. Tan, Q. Wei, Y. Pan, X. Tian, K. Zhao, X. Xu, Q. An, L. Mai and Q. Zhang, Nano Energy, 2017, 40, 73–81.
13. Y. Yue and H. Liang, Adv. Energy Mater., 2017, 7, 1602545.
14. J. Yao, Y. Li, R. C. Massé, E. Uchaker and G. Cao, Energy Stor. Mater., 2018, 11, 205–259.
15. S. Wang, K. A. Owusu, L. Mai, Y. Ke, Y. Zhou, P. Hu, S. Magdassi and Y. Long, Appl. Energy, 2018, 211, 200–217.
16. M.-S. Balogun, C. Li, Y. Zeng, M. Yu, Q. Wu, M. Wu, X. Lu and Y. Tong, J. Power Sources, 2014, 272, 946–953.
17. B. Long, M.-S. Balogun, L. Luo, Y. Luo, W. Qiu, S. Song, L. Zhang and Y. Tong, Small, 2017, 13, 1702081.
18. L. Shen, B. Ding, P. Nie, G. Cao and X. Zhang, Adv. Energy Mater., 2013, 3, 1484–1489.
19. M.-S. Balogun, Y. Luo, F. Lyu, F. Wang, H. Yang, H. Li, C. Liang, M. Huang, Y. Huang and Y. Tong, ACS Appl. Mater. Interfaces, 2016, 8, 9733–9744.
20. Y. Shi, L. Wen, G. Zhou, J. Chen, S. Pei, K. Huang, H. M. Cheng and F. Li, 2D Mater., 2015, 2, 024004.
21. S. Yoon, S. Lee, S. Kim, K.-W. Park, D. Cho and Y. Jeong, J. Power Sources, 2015, 279, 495–501.
22. G. Qian, X. Liao, Y. Zhu, F. Pan, X. Chen and Y. Yang, ACS Energy Lett., 2019, 4, 690–701.
23. X. Min, B. Sun, S. Chen, M. Fang, X. Wu, Y. g. Liu, A. Abdelkader, Z. Huang, T. Liu, K. Xi and R. Vasant Kumar, Energy Stor. Mater., 2019, 16, 597–606.
24. S. Liu, Z. Wang, C. Yu, H. B. Wu, G. Wang, Q. Dong, J. Qiu and A. Eychmüller, Adv. Mater., 2013, 25, 3462–3467.
25. Y. Deng, S. Dong, Z. Li, H. Jiang, X. Zhang and X. Ji, Small Methods, 2018, 2, 1700332.
26. X. Huang, G. Diao, S. Li, M.-S. Balogun, N. Li, Y. Huang, Z.-Q. Liu and Y. Tong, RSC Adv., 2018, 8, 17056–17059.
27. M.-S. Balogun, H. Yang, Y. Luo, W. Qiu, Y. Huang, Z.-Q. Liu and Y. Tong, Energy Environ. Sci., 2018, 11, 1859–1869.
28. H.-P. Feng, L. Tang, G.-M. Zeng, J. Tang, Y.-C. Deng, M. Yan, Y.-N. Liu, Y.-Y. Zhou, X.-Y. Ren and S. Chen, J. Mater. Chem. A, 2018, 6, 7310–7337.
29. M.-S. Balogun, M. Yu, Y. Huang, C. Li, P. Fang, Y. Liu, X. Lu and Y. Tong, Nano Energy, 2015, 11, 348–355.
30. G. X. Pan, X. H. Xia, F. Cao, J. Chen and Y. J. Zhang, Electrochim. Acta, 2014, 149, 349–354.
31. Y. Chu, L. Guo, B. Xi, Z. Feng, F. Wu, Y. Lin, J. Liu, D. Sun, J. Feng, Y. Qian and S. Xiong, Adv. Mater., 2018, 30, 1704244.
32. Y. Zheng, T. Zhou, X. Zhao, W. K. Pang, H. Gao, S. Li, Z. Zhou, H. Liu and Z. Guo, Adv. Mater., 2017, 29, 1700396.
33. J. Hu, C. Ouyang, S. A. Yang and H. Y. Yang, Nanoscale Horiz., 2019, 4, 457–463.
34. S. Ullah, P. A. Denis and F. Sato, Appl. Mater. Today, 2017, 9, 333–340.
35. M.-S. Balogun, W. Qiu, F. Lyu, Y. Luo, H. Meng, J. Li, W. Mai, L. Mai and Y. Tong, Nano Energy, 2016, 26, 446–455.
36. D. Chao, X. Xia, J. Liu, Z. Fan, C. F. Ng, J. Lin, H. Zhang, Z. X. Shen and H. J. Fan, Adv. Mater., 2014, 26, 5794–5800.
37. D. Kong, X. Li, Y. Zhang, X. Hai, B. Wang, X. Qiu, Q. Song, Q.-H. Yang and L. Zhi, Energy Environ. Sci., 2016, 9, 906–911.
38. M.-S. Balogun, W. Qiu, H. Yang, W. Fan, Y. Huang, P. Fang, G. Li, H. Ji and Y. Tong, Energy Environ. Sci., 2016, 9, 3411–3416.
39. X. Fang, M. Ge, J. Rong and C. Zhou, ACS Nano, 2014, 8, 4876–4882.
40. K. Wang, S. Luo, Y. Wu, X. He, F. Zhao, J. Wang, K. Jiang and S. Fan, Adv. Funct. Mater., 2013, 23, 846–853.
41. D. Chao, C. Zhu, X. Xia, J. Liu, X. Zhang, J. Wang, P. Liang, J. Lin, H. Zhang and Z. X. Shen, Nano Lett., 2014, 15, 565–573.
42. W. Wang, W. Liu, Y. Zeng, Y. Han, M. Yu, X. Lu and Y. Tong, Adv. Mater., 2015, 27, 3572–3578.
43. H. Song, C. Liu, C. Zhang and G. Cao, Nano Energy, 2016, 22, 1–10.
44. S. D. Perera, B. Patel, N. Nijem, K. Roodenko, O. Seitz, J. P. Ferraris, Y. J. Chabal and K. J. Balkus, Adv. Energy Mater., 2011, 1, 936–945.
45. Y. Song, T.-Y. Liu, B. Yao, T.-Y. Kou, D.-Y. Feng, X.-X. Liu and Y. Li, Small, 2017, 13, 1700067.
46. Y. Huang, Y. Lu, Y. Lin, Y. Mao, G. Ouyang, H. Liu, S. Zhang and Y. Tong, J. Mater. Chem. A, 2018, 6, 24740–24747.
47. Y. Huang, L. Hu, R. Liu, Y. Hu, T. Xiong, W. Qiu, M. S. Balogun, A. Pan and Y. Tong, Appl. Catal., B, 2019, 251, 181–194.
48. Y. Sun, K. Xu, Z. Wei, H. Li, T. Zhang, X. Li, W. Cai, J. Ma, H. J. Fan and Y. Li, Adv. Mater., 2018, 30, 1802121.
49. G. Diao, M.-S. Balogun, S.-Y. Tong, X. Guo, X. Huang, Y. Mao and Y. Tong, J. Mater. Chem. A, 2018, 6, 15274–15283.
50. D. Liu, Y. Liu, A. Pan, K. P. Nagle, G. T. Seidler, Y.-H. Jeong and G. Cao, J. Phys. Chem. C, 2011, 115, 4959–4965.
51. K. Ye, K. Li, Y. Lu, Z. Guo, N. Ni, H. Liu, Y. Huang, H. Ji and P. Wang, TrAC, Trends Anal. Chem., 2019, 116, 102–108.
52. M. Gotić, S. Popović, M. Ivanda and S. Musić, Mater. Lett., 2003, 57, 3186–3192.
53. G. Wang, H. Wang, X. Lu, Y. Ling, M. Yu, T. Zhai, Y. Tong and Y. Li, Adv. Mater., 2014, 26, 2676–2682.
54. P. Zhang, L. Zhao, Q. An, Q. Wei, L. Zhou, X. Wei, J. Sheng and L. Mai, Small, 2016, 12, 1082–1090.
55. A. Eftekhari, Sustainable Energy Fuels, 2017, 1, 2053–2060.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9nh00327d

‡ These authors contributed equally.

---