Self-assembled Ti₃C₂ MXene and N-rich porous carbon hybrids as superior anodes for high-performance potassium-ion batteries

Abstract

Potassium-ion batteries (PIBs) are attracting increased attention because of their low cost and similar energy storage mechanism to lithium-ion batteries. Considering the low structural stability and poor electrochemical redox reaction kinetics resulting from the large size of K⁺ (1.38 Å), we elaborately designed novel PDDA-NPCN/Ti₃C₂ hybrids as PIBs anodes via an electrostatic attraction self-assembly approach, while N-rich porous carbon nanosheets (NPCNs) are derived from metal–hexamine frameworks. The coupled PDDA-NPCN/Ti₃C₂ hybrids with stacked structure and large specific surface area could ensure intimate contact between Ti₃C₂ and the NPCNs to efficiently take advantage of both components and more accessible active sites. The hybrids afford enlarged interlayer spacing and unique 3D interconnected conductive networks to accelerate the ionic/electronic transport rates. Meanwhile, the robust hybrids contribute high chemical stabilities due to favorable tolerance to volume change caused by phase transformations during the fast charge/discharge process. DFT calculations further indicate that the PDDA-NPCN/Ti₃C₂ hybrids efficiently reduce the adsorption energy of K⁺ and accelerate the reaction kinetics. The hybrids possess a remarkable synergetic effect, leading to a high reversible capacity of 358.4 mA h g⁻¹ after 300 cycles at 0.1 A g⁻¹ and long cycling stability of 252.2 mA h g⁻¹ with only 0.03% degradation per cycle within 2000 cycles at 1.0 A g⁻¹. This work paves the way for further self-assembled coupled hybrids in energy storage devices.

---

Broader context

Nowadays, the ever-increasing energy consumption and rapid depletion of fossil fuels have driven extensive research activities on the development of green and cost-effective electrochemical energy storage and conversion technologies. Owing to lithium's rarity, uneven distribution in the Earth's crust and high cost, potassium-ion batteries (PIBs) have attracted tremendous attention recently owing to their low cost, fast ionic conductivity in electrolyte, high operating voltage and similar energy storage mechanism to lithium/sodium-ion batteries (LIBs/SIBs). However, the commercialization of PIBs anodes is greatly hindered by the low structural stability, poor electrochemical redox reaction kinetics and fast capacity fading caused by the large size of K⁺ (1.38 Å) as compared to 0.76 Å/1.02 Å for Li⁺/Na⁺. Herein, we develop an elaborate strategy to synthesize PDDA-NPCN/Ti₃C₂ hybrids as anodes for high-performance PIBs, which exhibit excellent electrochemical performance, showing long cycling stability with only 0.03% degradation per cycle within 2000 cycles at 1.0 A g⁻¹. It is a novel structure for PIBs application, and it seems that these questions for PIBs are well resolved in this work. This ingenious strategy can effectively accelerate the ionic/electronic transport kinetics, paving the way for self-assembled hybrids in energy storage devices.

Introduction

Potassium-ion batteries (PIBs), which possess a similar “rocking chair” mechanism to lithium/sodium-ion batteries (LIBs/SIBs), have attracted tremendous attention in large-scale energy storage applications owing to their low cost, fast ionic conductivity in electrolyte and high operating voltage.^1–5 The redox standard hydrogen potential of K/K⁺ (−2.92 V vs. standard hydrogen electrode (SHE)) is lower than that of Na/Na⁺ (−2.71 V vs. SHE) and close to that of Li/Li⁺ (−3.04 V vs. SHE), which enables PIBs to be operated in a higher potential window and thereby provides a higher energy density. Smaller solvated K⁺ caused by the much weaker Lewis acidity of K⁺ leads to a lower desolvation energy and smaller Stokes’ radius (3.6 Å) compared with Li⁺ (4.8 Å) and Na⁺ (4.6 Å), indicating that it has a better ion mobility and ion conductivity in electrolyte.^6,7 However, the poor ion diffusivity in solid graphitic carbon with confined interspace electrodes leads to sluggish reaction kinetics and large volume variations during cycling due to the larger diameter of K⁺ (1.38 Å vs. 0.76 Å/1.02 Å for Li⁺/Na⁺), which induces structural instability and fast capacity fading.^8–10 Therefore, it is still a great challenge to develop appropriate electrode materials to accommodate huge volume changes and improve the K⁺ diffusivity/reaction kinetics, thus enhancing the rate capability and cycling performance.

MXenes, as an emerging potential 2D material, have gained widespread interest due to their high metallic conductivity, abundant surface functional groups and controllable flexible interlayer spacing, which is beneficial for fast ion diffusion and offering more ion intercalation channels over the whole exposed surface.^11–21 It is revealed that few-layer Ti₃C₂ nanosheets display a high conductivity of 6.76 × 10⁵ S m⁻¹ and a low diffusion barrier of K⁺ of about 0.103 eV, which indicates excellent electron transport and superior K⁺ diffusion kinetics.^11,12 More recently, theoretical simulations and experimental measurements reveal that Ti₃C₂ is electrochemically redoxable, which can act as active sites for efficient potassium storage, delivering reversible capacities of 42–136 mA h g⁻¹, and excellent rate capability.^12,14,19,20 The electrochemical performance of multilayer MXene nanosheets can be further improved by exfoliating them into few-layer nanosheets. Nevertheless, the aggregation and self-restacking of few-layer MXene nanosheets are usually inevitable during the drying and electrode fabrication processes due to the strong van der Waals interactions and hydrogen bonds between adjacent nanosheets, which usually limits the accessibility to electrolyte ions and hinders the full utilization of their functional surfaces.^21,22

In order to overcome this shortcoming, recently, layer-by-layer structures could offer a stable scaffold for volume expansion and a stable SEI layer for side reactions, which have been proved definitely to make full use of both characteristics of the components for superior electrochemical performance compared with other methods.^23–25 Among the various 2D materials,^15,26–31 metal–hexamine framework (MHF) derived N-rich porous carbon nanosheets (NPCNs) with a large amount of open pores will benefit both electron and ion transfer during cycling, as well as fully contact and store sufficient electrolyte for rechargeable batteries. Specifically, N-doping could introduce defects and more electronegative sites in carbon, which further enhances the electronic conductivity of carbon and generates more active sites with stronger attraction for ion storage in carbon.^27,32,33 On the other hand, NPCNs would provide an ideal “buffer” or interlayer spacer for coupling with MXene nanosheets to effectively prevent self-restacking between the MXene adjacent nanosheets, which is desirable to ensure structural stability through accelerating the ionic/electronic transport and overcome the serious volume changes during the fast charge/discharge process. However, it is a great challenge and important to rationally design unique NPCN/Ti₃C₂ hybrids for high-performance PIBs. These robust hybrids are expected to make full utilization of every nanosheet, and enhance interfacial charge separation and transfer between adjacent layers, which is important to further achieve high electrochemical performance.^34,35

In this paper, we elaborately design novel PDDA-NPCN/Ti₃C₂ hybrids via an electrostatic attraction self-assembly approach. The synergistic effect between Ti₃C₂ and NPCNs affords enlarged interlayer spacing and a unique 3D porous interconnected conductive network architecture, which exhibits more accessibility of the electrolyte, and shortens the ion transmission paths to facilitate rapid ionic/electronic transport, thereby contributing to enhanced performance. The robust hybrids may alleviate the serious volume changes caused by the phase transformations during the fast charge/discharge process and assist durable rate capability and cycling stability. DFT calculations further show that the PDDA-NPCN/Ti₃C₂ hybrids improve the interfacial K⁺ adsorption ability and promote the potassiation process. Benefiting from those advantages, the PDDA-NPCN/Ti₃C₂ anode exhibits a high reversible capacity of 358.4 mA h g⁻¹ after 300 cycles at 0.1 A g⁻¹ and long cycling stability of 252.2 mA h g⁻¹ with only 0.03% degradation per cycle within 2000 cycles at 1.0 A g⁻¹. The related mechanisms of NPCNs and Ti₃C₂ in the potassiation/depotassiation process are deeply investigated.

Results and discussion

Microstructure characterization

Fig. 1 schematically illustrates the overall fabrication process of the coupled PDDA-NPCN/Ti₃C₂ hybrids. It follows a static electronic self-assembly process between negatively charged exfoliated Ti₃C₂ nanosheets (ex-Ti₃C₂) and N-rich porous carbon nanosheets modified by positively charged polymer poly(diallyldimethylammonium chloride) (PDDA-NPCNs). The ex-Ti₃C₂ nanosheets would be covered by negative charge groups, such as –OH, –F or –O groups, showing a zeta potential of −33.2 mV.³⁶ The scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of the ex-Ti₃C₂ nanosheets (Fig. 2a and b) clearly present a crinkled and self-restacked sheet-like morphology with an ultrathin thickness. The surface functional groups endow them with a hydrophilic nature and uniform dispersion in water, showing a clear Tyndall effect (Fig. 2a, inset).^15,37 The atomic force microscopy (AFM) image of the ex-Ti₃C₂ nanosheets presents a uniform and flat surface with a few-layer thickness, e.g., ∼3.3 nm (Fig. S1, ESI†) in a bilayer structure.³⁸ The as-prepared NPCNs are modified by PDDA and became positively charged with a zeta potential of +35.3 mV (Fig. S2a–g, ESI†). The TEM image (Fig. 2c) reveals that the PDDA-NPCNs have ample pores that uniformly distribute over the body, and its colloidal suspension (Fig. 2c, inset) is stable and exhibits a typical Tyndall effect.³⁹ The PDDA-NPCNs (Fig. S2h and i, ESI†) show a uniform thickness of ∼3.7 nm after being modified with PDDA, which is in agreement with previous reports.⁴⁰ The PDDA-NPCN/Ti₃C₂ hybrids (Fig. 2d, inset) with a zeta potential of +2.77 mV are obtained through flocculation based on a mass ratio of 2.0.

Fig. 1 Schematic illustration of the fabrication of PDDA-NPCN/Ti₃C₂ hybrids. The water molecules between the layers are not shown here for clarity.

Fig. 2 Mophology and chemical composition characterization. (a and b) SEM and TEM images of ex-Ti₃C₂ nanosheets. (c) TEM image of PDDA-NPCNs. The inset shows the Tyndall effect. (d) SEM, (e) TEM and (f) ED pattern of PDDA-NPCN/Ti₃C₂ flocculation. The inset shows flocculation. T stands for ex-Ti₃C₂ and C is PDDA-NPCNs. (g) EDS spectrum. (h) HAADF-STEM image of PDDA-NPCN/Ti₃C₂ and the corresponding elemental mapping images of Ti, C and N elements, respectively. (i) AFM image and thickness profile of PDDA-NPCN/Ti₃C₂.

The SEM and TEM images (Fig. 2d and e) present the stacking of PDDA-NPCNs and ex-Ti₃C₂ nanosheets, which shows a porous interconnected conductive network structure, tremendously shortening the ion transport pathways and facilitating fast ionic/electronic transport. The clear lattice fringes of 0.34 nm and 0.15 nm under high-resolution transmission electron microscopy (HRTEM) (Fig. S3, ESI†) roughly correspond to the (002) plane of the PDDA-NPCNs and the (110) plane of the ex-Ti₃C₂ nanosheets. The electron diffraction (ED) patterns (Fig. 2f and Fig. S3, inset, ESI†) reveal the in-plane diffraction of the (100) and (110) planes of ex-Ti₃C₂ (T₁₀₀, T₁₁₀) and the (100) and (110) planes of the PDDA-NPCNs (C₁₀₀, C₁₁₀), again proving the existence of both nanosheets. More importantly, a typical energy dispersive X-ray spectroscopy (EDS) spectrum (Fig. 2g) shows that the real mass ratio of ex-Ti₃C₂ to PDDA-NPCNs is estimated to be ∼2.1, which is close to the calculated value. And a typical high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image and the corresponding elemental mapping images (Fig. 2h) indicate that the Ti, C and N elements homogeneously distribute over the PDDA-NPCN/Ti₃C₂ hybrids, revealing the stacking of ex-Ti₃C₂ and the PDDA-NPCNs in the lateral dimension. Typical stacking of both nanosheets is observed in the AFM observation (Fig. 2i), in which the PDDA-NPCNs (∼3.8 nm in thickness) are superimposed on the ex-Ti₃C₂ nanosheets (∼3.3 nm in thickness), in accordance with the TEM results.

The X-ray diffraction (XRD) pattern (Fig. 3a) of ex-Ti₃C₂ shows that the (001), (002) and (003) peaks are initially at 6.8°, 13.6° and 20.5° after the etching, intercalation and exfoliation process, corresponding to an interlayer spacing value of 13.0 Å. The PDDA-NPCNs only exhibit a broad peak at 21.7°, which is indexed to the (002) plane of carbon materials with a low degree of graphitization, and the interlayer spacing could be calculated to be 4.0 Å. The flocculated PDDA-NPCN/Ti₃C₂ material shows coexisting crystallographic phases of ex-Ti₃C₂ with a predominant (001) peak and well maintained PDDA-NPCNs with a weak (002) peak. In particular, the (001) plane downshifts to 4.6°, revealing an interlayer spacing of 19.2 Å. The increased spacing (6.2 Å) might be caused by the intercalation of PDDA-NPCNs into the interlayer of ex-Ti₃C₂ through the strong electrostatic interactions between them as well as the water content in the gallery of the flocculated products, which is in agreement with previous reports.^23,24,37

Fig. 3 Structure and chemical composition characterization. (a) XRD patterns of ex-Ti₃C₂, PDDA-NPCNs and PDDA-NPCN/Ti₃C₂ hybrids, respectively. XPS spectra of ex-Ti₃C₂, PDDA-NPCNs and PDDA-NPCN/Ti₃C₂ hybrids, respectively. (b) Full spectra. (c–e) High-resolution spectra of Ti 2p, C 1s and N 1s. (f) Raman spectra of ex-Ti₃C₂, PDDA-NPCNs and PDDA-NPCN/Ti₃C₂ hybrids.

To profoundly understand the elemental compositions and corresponding chemical bonding states of Ti, C, O, F and N species in the PDDA-NPCN/Ti₃C₂ hybrids, X-ray photoelectron spectroscopy (XPS) (Fig. 3b–e and Fig. S4, ESI†) is performed. The XPS spectra as shown in Fig. 3b indicate that the ex-Ti₃C₂ nanosheets are mainly composed of Ti, C, O and F. After the PDDA-NPCNs are introduced into the ex-Ti₃C₂ nanosheets, a new peak at a binding energy of 399.9 eV in the spectra for the PDDA-NPCN/Ti₃C₂ hybrids is assigned to N 1s. The high-resolution Ti 2p spectra of ex-Ti₃C₂ and the PDDA-NPCN/Ti₃C₂ hybrids (Fig. 3c) could be fitted with three doublets (Ti 2p3/2–Ti 2p1/2).¹⁷ The Ti 2p3/2 components of ex-Ti₃C₂ located at 455.0, 456.1 and 458.8 eV correspond to Ti–C, Ti(II) and Ti–O bonds, respectively. The C 1s core level of the ex-Ti₃C₂ nanosheets (Fig. 3d) could be fitted with four components centered at 281.5, 283.8, 284.6, and 288.5 eV, which could be assigned to Ti–C, C–C, C–O and C=O bonds, respectively.¹⁷ The C 1s spectrum of the PDDA-NPCNs displays four peaks at 289.9, 287.4, 285.5, and 284.7 eV, which could be attributed to π–π*, C=O, C–N, and sp² carbon (C=C/C–C), respectively.⁴⁰ The strongest peak at 284.7 eV indicates that graphitic carbon is the major species for the layered bulk MHF derived porous carbon. Moreover, deconvolution of the C 1s region for the PDDA-NPCN/Ti₃C₂ hybrids shows that the characteristic peaks of Ti–C, C=C/C–C, C–O, C–N, C=O and π–π* are still present. Correspondingly, the N 1s spectrum of the PDDA-NPCNs in Fig. 3e could be fitted with four components centered at 398.4, 400.9, 402.5, and 405.2 eV, which could be assigned to pyridinic N (N-6), pyrrolic N (N-5), graphitic N (N-Q) and N–O, indicating that N atoms are indeed inserted into the carbon lattice of the PDDA-NPCNs.⁴¹ Benefiting from the intrinsic high N content of the HMT ligand (N/C = 66.7%), the PDDA-NPCNs also possess a high N-doping level of 10.35 at% in addition to the novel structure, which can greatly improve the wettability and conductivity of the electrode and increase the active sites for potassium storage. According to previous reports, among the various types of N, pyridinic-N is more active compared with the other types of N because it can efficiently induce sufficient defects and more active sites, and is more energetically favorable towards adsorbing alkali metal ions.^42,43

As shown in the Raman spectra (Fig. 3f), the three I, II and III peaks around 258, 410 and 605 cm⁻¹ correspond to the vibrations from Ti–C bonds for ex-Ti₃C₂, respectively.³⁶ Besides, the two new obvious peaks at 1383 and 1570 cm⁻¹ could be attributed to the D band (disordered/defect-induced) and G band (graphite), respectively, and the ratio of intensity between I_D and I_G of the ex-Ti₃C₂ nanosheets is 0.85, suggesting a high graphitization degree. The PDDA-NPCNs show two typical characteristic bands of the D-band (1346 cm⁻¹) and G-band (1578 cm⁻¹), and the ratio of intensity between I_D and I_G is 1.02, which could be ascribed to the higher defect concentration induced by nitrogen doping. After flocculation with PDDA-NPCNs, the I and III peaks of the PDDA-NPCN/Ti₃C₂ hybrids broaden and downshift. Notably, some shift for the II peak of the PDDA-NPCN/Ti₃C₂ hybrids is possibly due to the slightly-oxidized surface in sonication during the fabrication of the PDDA-NPCN/Ti₃C₂ hybrid process, in accordance with XPS results, which is consistent with a previous report.⁴⁴ And the ratio of 0.97 for the intensity between I_D and I_G of the PDDA-NPCN/Ti₃C₂ hybrids could be ascribed to the combined effects of the higher defect concentration induced by the PDDA-NPCNs and the higher graphitization degree induced by ex-Ti₃C₂. Furthermore, there are no redox reactions occurring between ex-Ti₃C₂ and the PDDA-NPCNs, and the presence of surface groups in the flocculation process is further revealed by FTIR (Fig. S5, ESI†).^{15,36,40,45,46} The designed PDDA-NPCN/Ti₃C₂ hybrids (Fig. S6 and Table S1, ESI†) show a higher specific surface area of 147.87 m² g⁻¹ with a pore volume of 0.809 cm³ g⁻¹ compared with the ex-Ti₃C₂ material, which could provide enough transport pathways for K⁺ insertion and expose more accessible active sites for enhanced performance.

Potassium ion storage performance

The electrochemical behaviors as PIBs anodes are shown in Fig. 4. The galvanostatic charge/discharge profiles of the PDDA-NPCN/Ti₃C₂ anode (Fig. 4a) are measured in the first five cycles at 0.1 A g⁻¹. It delivers initial discharge and charge capacities of 797.3 and 583.6 mA h g⁻¹, respectively (all the capacities are calculated based on the mass of the total electrode), corresponding to a CE of 73.2%. Such a low CE in the first cycle can be attributed to the formation of the solid electrolyte interface (SEI) layer on a large surface area, which leads to large irreversible consumption and decomposition of the electrolyte and electrode material. As the cycle number increases, the capacity gradually becomes stable, with CE up to 95.9% at the fifth cycle, corresponding to high cycling stability of the electrode. The prolonged sloping region and an indistinct plateau below 1.5 V of potassiation for the first cycle could be ascribed to different reactions, including the reactions between K⁺ and surface functional groups, the decomposition of the electrolyte and formation of the SEI layer, and the stepwise insertion of K⁺ into the Ti₃C₂ and NPCNs layers to form a K-intercalated compound, which agrees with the weak peaks observed between 0 and 1.5 V in the first cathodic process (Fig. S7, ESI†).^{12,19,29,32,47} The voltage profile of depotassiation in the first cycle monotonically increasing from 0.01 V to 3 V is ascribed to the combination of multiple charge storage mechanisms with the stepwise extraction of K⁺ from the K-intercalated compound and Ti₃C₂ layers, and the reversible reactions between K⁺ and surface functional groups, which is also revealed by the broad peak between 0.2 and 1.0 V in the CV profiles.

Fig. 4 Potassium ion storage performance. (a) Initial five galvanostatic charge/discharge profiles at 0.1 A g⁻¹ for the PDDA-NPCN/Ti₃C₂ anode. (b) The 5th charge/discharge profiles at 0.1 A g⁻¹, (c) rate performance at different current densities and (d) cycling stability at 0.1 A g⁻¹ for different anodes, respectively. (e) Long cycling performance of the PDDA-NPCN/Ti₃C₂ anode at 1.0 and 2.0 A g⁻¹ for 2000 cycles.

To further reveal the performance of different materials, the galvanostatic charge/discharge profiles of the fifth cycle at 0.1 A g⁻¹ are extracted for comparison as shown in Fig. 4b. The ex-Ti₃C₂, PDDA-NPCNs and PDDA-NPCN/Ti₃C₂ anodes exhibit reversible capacities of 172.3, 387.3 and 526.6 mA h g⁻¹, respectively, which suggests that PDDA-NPCN/Ti₃C₂ is more conducive to making full use of the active materials for efficient energy storage. A high-rate capability and long-term cycle performance are also important for practical applications. As shown in Fig. 4c, among the three anodes, the PDDA-NPCN/Ti₃C₂ anode exhibits excellent rate performance, especially under high current densities. Specifically, the PDDA-NPCN/Ti₃C₂ anode delivers reversible capacities of 499.8, 402.2, 319.4, 254.8 and 191.2 mA h g⁻¹ at 0.1, 0.2, 0.5, 1.0 and 2.0 A g⁻¹, respectively. Noticeably, it could quickly recover a capacity of 363.8 mA h g⁻¹ when the current density is back to 0.1 A g⁻¹. In contrast, both ex-Ti₃C₂ and the PDDA-NPCNs are not able to be comparable with PDDA-NPCN/Ti₃C₂ in capacity and rate performance. For example, they only have reversible capacities of 68.9 (ex-Ti₃C₂) and 152.0 mA h g⁻¹ (PDDA-NPCNs) at 2.0 A g⁻¹ after 50 cycles, obviously worse than that of the PDDA-NPCN/Ti₃C₂ anode. This is because the PDDA-NPCN/Ti₃C₂ anode shows a unique structure with enlarged interlayer spacing and high conductivity, which is favorable for rapid electron transfer and abundant electrolyte infiltration, thereby contributing to enhanced performance compared to the other anodes.

In addition, in order to further investigate the influence of the NPCNs content on the electrochemical performance, PDDA-NPCN/Ti₃C₂ hybrids with three different mass ratios of Ti₃C₂/NPCN are prepared and measured under identical test conditions. As shown in Fig. S8 (ESI†), the PDDA-NPCN/Ti₃C₂-2 anode (a mass ratio of 2.0 for Ti₃C₂/NPCN) shows much better electrochemical performance than the PDDA-NPCN/Ti₃C₂-1 and PDDA-NPCN/Ti₃C₂-3 anodes. The enhanced electrochemical performance of the PDDA-NPCN/Ti₃C₂-2 anode could be attributed to the full utilization of the high theoretical capacity of the NPCNs and the high conductivity of Ti₃C₂, because there is no extra self-stacking of the NPCNs and Ti₃C₂ to reduce self-utilization. Specifically, the initial discharge capacity increases with the NPCNs content, while the initial coulombic efficiency decreases contrarily with the NPCNs content, which is possibly because the NPCNs with a well-developed pore structure and high specific surface area consume a large amount of electrolyte for SEI layer formation. For the PDDA-NPCN/Ti₃C₂-1 anode with higher NPCNs content, although it shows a higher initial discharge capacity, it displays poorer rate performance than the PDDA-NPCN/Ti₃C₂-2 anode. For the PDDA-NPCN/Ti₃C₂-3 anode with a relatively lower content of NPCNs, it shows better rate performance, although it shows lower reversible capacity than the PDDA-NPCN/Ti₃C₂-2 and PDDA-NPCN/Ti₃C₂-1 anodes.

The cycling stabilities of the different anodes are further examined at 0.1 A g⁻¹ as depicted in Fig. 4d. The PDDA-NPCN/Ti₃C₂ anode also exhibits the highest reversible capacity and excellent cycling performance. An excellent reversible capacity of 358.4 mA h g⁻¹ still remains after 300 cycles, which corresponds to 61.4% of the initial reversible capacity. Except for the first few cycles, a stable CE of 99.0% is obtained in the following cycles, revealing excellent cycling stability, which is comparable or superior to those of other previously reported Ti₃C₂-based and carbonaceous anodes (Table S2, ESI†).^{12,19–22,27,29,32,48–56} The ex-Ti₃C₂ and PDDA-NPCNs anodes also display stable reversible capacities of 106.2 and 280.9 mA h g⁻¹, respectively. It should be noticed that the capacity of PDDA-NPCN/Ti₃C₂ is not the combination of Ti₃C₂ and the NPCNs, but much larger than both of them. This means there is a synergetic effect between Ti₃C₂ and NPCNs within such hybrids, which is further verified by DFT calculations as follows. In addition, high-rate and long-term cycling performance (Fig. 4e) is also measured on the PDDA-NPCN/Ti₃C₂ anode. Impressively, it retains reversible capacities of 252.2 mA h g⁻¹ with a capacity decay as low as 0.03% per cycle at 1.0 A g⁻¹ and 151.2 mA h g⁻¹ at 2.0 A g⁻¹ after 2000 cycles, which is comparable or much superior to those of other previously reported anode materials (Table S3, ESI†).^{19,21,22,27,49–58} It evidently reveals that the PDDA-NPCN/Ti₃C₂ hybrid anode intrinsically has a better ability to accommodate and survive large stresses and strains caused by K⁺ insertion/extraction during the potassiation/depotassiation process. It could also provide multiple accessible active sites, further accelerate electron/ion transport and facilitate access of electrolyte ions to the electrode, thus offering high reversible capacity and cycling stability.

Electrochemical kinetic behaviors

Electrochemical impedance spectroscopy (EIS) is further performed to explain the electrochemical performance. As can be seen in Fig. 5a, the Nyquist plots mainly include a single depressed semicircle at the high-medium frequency region associated with the charge transfer resistance (R_ct) and an inclined line in the low-frequency region related to the K⁺ diffusion ability defined as the Warburg impedance (W₀). Specifically, R_ct of the PDDA-NPCN/Ti₃C₂ anode is 769.5 Ω, which is obviously lower than that of ex-Ti₃C₂ (1257.3 Ω) and the PDDA-NPCNs (1050.1 Ω), indicating that this PDDA-NPCN/Ti₃C₂ anode can remarkably accelerate interfacial charge separation and transfer between adjacent layers during repeated K⁺ insertion/extraction processes. The K⁺ diffusion coefficient in the electrode could be obtained by the plots in the Warburg region based on the following equation:⁴¹

|Z| = R_ct + R_e + σω^−1/2 (1)

D_k = R²T²/(2A²n⁴F⁴C_k²σ²) (2)

where D_k represents the K⁺ diffusion coefficient, R stands for the gas constant, T is the absolute temperature, A is the surface area, n is the number of electrons per molecule oxidized, F is Faraday's constant, C is the concentration, σ is the Warburg factor, and ω is the frequency. The relationships between |Z| and ω^−1/2 within the Warburg region for the different anodes are plotted to calculate σ and subsequently determine D_k (Fig. 5b).⁵⁹ Specifically, D_k of the PDDA-NPCNs and PDDA-NPCN/Ti₃C₂ anodes is 2.0 × 10⁻¹⁴ and 1.9 × 10⁻¹⁴ cm² s⁻¹, respectively, which is obviously higher than that of ex-Ti₃C₂ (1.0 × 10⁻¹⁴ cm² s⁻¹), and comparable or much superior to those of other previously reported results (Table S4, ESI†).^60–64 These results prove that the PDDA-NPCN/Ti₃C₂ hybrids can form unobstructed pathways for effective ion diffusion channels, and interfacial charge separation and transfer between adjacent layers, and thus are conducive to offering attractive electrochemical performance.

Fig. 5 Electrochemical kinetic behaviors. (a) Nyquist plots. The inset exhibits the equivalent circuit. (b) Extracted |Z| vs. ω^1/2 plot in the Warburg region for different anodes, respectively. (c) CV curves at various scan rates and (d) relationship between log(i) vs. log(ν) for the PDDA-NPCN/Ti₃C₂ anode. (e) Capacitive-controlled and diffusion-controlled contributions at 1.0 mV s⁻¹ for the PDDA-NPCN/Ti₃C₂ anode. (f) Normalized contribution ratio of capacitive-controlled capacities for the PDDA-NPCN/Ti₃C₂ anode at different scan rates.

To better understand the high-rate capability of the PDDA-NPCN/Ti₃C₂ anode, CV curves (Fig. 5c) are obtained at various scan rates from 0.1 to 1.0 mV s⁻¹ to investigate the K⁺ storage kinetic behaviors. With an increasing scan rate, the measured peak current (i) at a certain potential is not proportional to the square root of the scan rate (ν), indicating that the oxidation and reduction processes are not simple ion diffusion-controlled processes.^17,32 Here, the relationship between i and ν can be described by the following equations:^32,47

i = aν^b, (0.5 ≤ b ≤ 1) (3)

log i = b log ν + log a (4)

where a and b are variable parameters. The K⁺ storage behaviors are reflected by the b value. Particularly, b = 1 stands for a completely surface-controlled (capacitive) process, while b = 0.5 for a completely diffusion-controlled one. As shown in Fig. 5d, the value of b corresponds to the slope of the log i vs. log ν profile. At the oxidation and reduction peak potential of 1.2 V, the b-values are 0.84 and 0.66, respectively, revealing that capacitive behaviors cannot be ignored in the PDDA-NPCN/Ti₃C₂ anode. The same kinetics analysis is performed on the ex-Ti₃C₂ (Fig. S9, ESI†) and PDDA-NPCNs anodes (Fig. S10, ESI†) as well, which demonstrates that the PDDA-NPCN/Ti₃C₂ and PDDA-NPCNs anodes show superior K⁺ diffusion kinetics and better rate capability than the ex-Ti₃C₂ anode. That's the reason why the PDDA-NPCN/Ti₃C₂ anode shows a much larger capacity than the ex-Ti₃C₂ anode at the same current density. In fact, the current response can be further expressed as:^32,47

i(ν) = k₁ν + k₂ν^1/2 (5)

i(ν)/ν^1/2 = k₁ν^1/2 + k₂ (6)

where k₁ and k₂ are adjustable constants. k₁ν and k₂ν^1/2 represent the capacitive-controlled contribution and diffusion-controlled contribution. The diffusive and capacitive component could be quantitatively resolved by plotting i/ν^1/2vs. ν^1/2. As displayed in Fig. 5e, the PDDA-NPCN/Ti₃C₂ anode shows a capacitive contribution ratio of 65.2% at a scan rate of 1.0 mV s⁻¹. Correspondingly, the PDDA-NPCN/Ti₃C₂ anode (Fig. 5f) shows a larger capacitive component at all scan rates, further revealing more efficient K⁺ adsorption and faster reaction kinetics. The excellent electrochemical properties of the PDDA-NPCN/Ti₃C₂ anode are mainly ascribed to the following synergistic effects (Fig. S11, ESI†): (i) the coupled hybrids with a stacked structure and large specific surface area ensure intimate contact between Ti₃C₂ and the NPCNs to efficiently take advantage of both components and more accessible active sites; (ii) the PDDA-NPCN/Ti₃C₂ hybrids afford unique 3D interconnected conductive networks with the expanded interlayer spacing accelerating the ionic/electronic transport rates, remarkably facilitating the K⁺ diffusion length of the electrolyte in the electrodes, and inducing more intrinsic capacitive behavior; and (iii) the robust hybrids are expected to constrain the serious volume changes resulting from phase transformations during the fast charge/discharge process, further enhancing the durability of high reversible capacity and long-term cycling stability.

Electrochemical reaction mechanisms

To better understand the superior cycling and rate performance, the voltage profile and the corresponding ex situ XRD patterns of the PDDA-NPCN/Ti₃C₂ anode in the initial charge/discharge cycle are investigated the structure evolution of the anode shown in Fig. 6a and b. As shown in Fig. 6b, the anode shows a clear XRD pattern, corresponding to that of the diffraction features of the hybrids. Upon potassiation, the peak of the NPCNs gradually decreases and shifts towards a lower angle, which indicates that the interlayer distance of the NPCNs increases due to the K⁺ insertion. Notably, the diffraction peaks of the NPCNs do not vanish until the potential reaches 0.30 V, where new peaks at 22.0 and 29.4° appear, which are attributed to KC₃₆ (Fig. 6c).^48,65 Upon further potassiation, KC₃₆ further transforms to KC₂₄ (Fig. 6d), corresponding to the new peaks at 20.2 and 30.6°. In the full potassiation state, new characteristic peaks at 16.2 and 33.5° are observed, corresponding to the formation of phase-pure KC₈ (Fig. 6e).⁷ The interlayer distance of the fresh NPCNs is 4.08 Å and it increases to 5.37 Å after K⁺ insertion, for which the corresponding volume expansion of the NPCNs is estimated to be 31.6%. The ultimate potassiation product of KC₈ leads to a plateau well above the plating potential of potassium metal, which may relieve the danger of dendrite formation. Meanwhile, the (001) peak of Ti₃C₂ shifts to a lower angle, suggesting a significant expansion of the interlayer distance from 19.2 Å to 24.6 Å, which corresponds to K⁺ insertion into the PDDA-NPCN/Ti₃C₂ anode. During the first depotassiation process (Fig. 6b), the XRD peaks are widened, which suggests that the extraction of K⁺ causes certain damage to the hybrids.⁴⁸ According to the ex situ XRD analysis above, the reactions during the potassiation/depotassiation process could be denoted as the following equations:

8C + xK⁺ + xe⁻ ↔ K_xC₈ (7)

Ti₃C₂T_x + yK⁺ + ye⁻ ↔ K_yTi₃C₂T_x (8)

Fig. 6 Electrochemical mechanisms. (a) Voltage profile and (b) corresponding ex situ XRD patterns for the PDDA-NPCN/Ti₃C₂ anode at different states during the initial potassiation/depotassiation process. Supercells of (c) KC₃₆, (d) KC₂₄ and (e) KC₈. (f) Ex situ Raman spectra and (g and h) ex situ XPS spectra of the PDDA-NPCN/Ti₃C₂ anode at different potassiation/depotassiation states.

To characterize the microstructures, ex situ SEM images of the ex-Ti₃C₂, PDDA-NPCNs and PDDA-NPCN/Ti₃C₂ anodes after 300 cycles are investigated in Fig. S12 (ESI†). Notably, the ex-Ti₃C₂ anode (Fig. S12a, ESI†) agglomerates together seriously after 300 cycles owing to pulverization, which inevitably reduced the K⁺ accessible surface area and impeded electrolyte penetration. However, due to the existence of the PDDA-NPCNs (Fig. S12b, ESI†), the morphology of the PDDA-NPCN/Ti₃C₂ anode (Fig. S12c, ESI†) is well retained compared with ex-Ti₃C₂, exhibiting excellent structure stability. From the ex situ TEM image, a porous interconnected conductive network and the inner void space could be well preserved for the PDDA-NPCN/Ti₃C₂ anode even after 300 cycles as shown by Fig. S13a (ESI†), and a thin SEI coating covers the hybrids, which further demonstrates that the cycling stability is not deteriorated upon repeated large radius K⁺ insertion/extraction, making it possible to build long-life PIBs. Ti, C, O, F, N and K elements are homogeneously distributed among the whole anode (Fig. S13, ESI†), further proving successful potassiation of the PDDA-NPCN/Ti₃C₂ anode. Ex situ Raman and ex situ XPS analysis are further carried out for the PDDA-NPCN/Ti₃C₂ anode to study structure changes in the charge/discharge process. For the fresh PDDA-NPCN/Ti₃C₂ anode (Fig. 6f), the I, II and III peaks located at 273, 409, and 613 cm⁻¹ are assigned to the vibrational modes of Ti–C bonds, respectively.³⁶ The two obvious peaks at 1327 and 1588 cm⁻¹ could be attributed to the D and G bands, respectively.¹⁵ The ratio of intensity between I_D and I_G of the fresh PDDA-NPCN/Ti₃C₂ anode is 0.59, suggesting a high graphitization degree. When discharged to 0.01 V, the I and II peaks shift, while the III peak disappears, corresponding to the partial breaking of Ti–C bonds after formation of the Ti₃C₂T_xK_y phase.¹⁴ Meanwhile, a slight shift of the G band peak occurs accompanied by a decrease in the relative intensity, indicating a more disordered structure as a result of the potassiation process. When charged to 3.0 V, the III peak of the Ti–C bonds appears again but not as strong as that of the fresh anode, which is probably caused by different states of Ti–C bonds between the crystalline Ti₃C₂T_xK_y nanosheets and re-formed Ti₃C₂ nanosheets. Correspondingly, the ratio of intensity between I_D and I_G related to the disorder degree of carbon increases from 0.59 to 0.97 and then decreases to 0.95 for the fresh anode, fully discharged and fully charged, respectively, which illuminates the change of the disorder degree due to the K⁺ insertion/extraction, further confirming the structural sustainability and reversibility of the PDDA-NPCN/Ti₃C₂ anode upon the potassiation/depotassiation process. Furthermore, extra weak peaks appear at 757, 872 and 1037 cm⁻¹, which are possibly associated with the decomposition products of the electrolytes because they are only observed for the cycled samples.⁶⁶

As shown in the ex situ XPS spectra (Fig. S14, ESI†), the atomic percent of O, K and F is greatly enhanced after discharging to 0.01 V compared with that of the fresh PDDA-NPCN/Ti₃C₂ anode. The high-resolution XPS spectrum of K 2p (Fig. 6g) deconvolutes into four peaks: metallic K (293.2 and 295.7 eV), KC₈ (293.6 eV) and –C–O–K/–COOK/NK (292.6 eV), which could be attributed to the insertion of K⁺, the formation of KC₈ and the SEI layer.⁵³ However, when charged to 3.0 V again, the atomic percent of O, K and F falls back. The peak at 293.6 eV of the KC₈ state disappears, meanwhile the decreased percent of K could be readily considered as the reversible storage mechanism of K⁺. The residual K in the PDDA-NPCN/Ti₃C₂ anode could be ascribed to various potassium salts in the SEI layer and the trapped metallic K in the nanopores. However, the formation of the SEI layer occurs vigorously during the discharge process, which looks similar to adsorption. In this regard, the change of F 1s (Fig. 6h) on the PDDA-NPCN/Ti₃C₂ anode surface is distinguished during the charge/discharge process. The major component of the fresh anode is located at 684.3 eV, corresponding to the Ti–F surface functional group.¹⁵ The Ti–F peak shifts to higher binding energy 684.6 eV upon discharging to 0.01 V. Meanwhile, the K–F and P–F bonds at 685.2 and 687.3 eV appear, indicating that the SEI layer starts to form during the potassiation process.⁶⁷ When charged to 3.0 V, the Ti–F peak shifts back to the initial again, demonstrating that K⁺ can be reversibly inserted into/extracted from within the PDDA-NPCN/Ti₃C₂ anode. However, the P–F bonds shift to 687.0 eV, which can explain the irreversible reactions of the SEI layer and side reactions of the surface functional groups during the initial cycling. Based on the discussions above, a schematic illustration (Fig. S15, ESI†) of the possible electrochemical mechanism is proposed. The interlayer distance changes when K⁺ is inserted into/extracted from within the robust PDDA-NPCN/Ti₃C₂ anode through mesoporous tunnels and nanopores. It is noted that the NPCNs could transform into KC₈ at low voltage, and further recover to NPCNs at high voltage during the potassiation/depotassiation process, which provide an ideal “buffer” or interlayer spacer for coupling with MXene nanosheets, further enhancing the durability of high reversible capacity and long-term cycling stability.

Based on first principles, we further perform DFT calculations to investigate the adsorption abilities of K⁺ in detail. We construct four types of computational models, i.e., Ti₃C₂, NPCNs and NPCN/Ti₃C₂ hybrids (Fig. S16a–d, ESI†) and then calculate the corresponding adsorption energies (ΔE_a). As a comparison, the ΔE_a (Fig. S16e, ESI†) of the NPCN/Ti₃C₂ hybrids are −2.26 and −2.42 eV, much lower than those of the bare Ti₃C₂ (−0.43 eV) and NPCNs (−2.11 eV), which indicates NPCN/Ti₃C₂ hybrids with stronger K⁺ adsorption capability than Ti₃C₂ and the NPCNs. It also means that K⁺ would prefer to combine with the Ti₃C₂ and NPCNs anchored on the NPCN/Ti₃C₂ hybrids, which not only greatly decrease the ΔE_a of K⁺, but also facilitate the potassiation process. As presented in Fig. S17 (ESI†), the hybrids modify the electronic structure. In all situations, a net gain of electronic charge between K⁺ and Ti₃C₂, the NPCNs and the NPCN/Ti₃C₂ hybrids can be observed, indicating charge transfer from the adsorbed K to its nearest neighbor atoms. Different from the pure Ti₃C₂ and NPCNs, the surfaces for the NPCN/Ti₃C₂ hybrids also show more electron accumulation, indicating stronger charge transfer than the others. It suggests that the NPCNs evidently change the potassiation state of Ti₃C₂ and provide fast electron transfer pathways, which facilitates the potassiation process. Therefore, the NPCN/Ti₃C₂ hybrids could not only be more favorable for the K⁺ adsorption reactions, but also efficiently decrease the ΔE_a of K⁺ and facilitate the potassiation process, theoretically validating the boosted K⁺ storage kinetics.

Conclusions

In summary, we have employed a facile solution-phase co-feeding and electrostatic attraction self-assembly approach to rationally design novel PDDA-NPCN/Ti₃C₂ hybrids for high-performance potassium-ion batteries. The hybrids show an obvious synergetic effect of both materials, leading to a high reversible capacity of 358.4 mA h g⁻¹ after 300 cycles at 0.1 A g⁻¹, much larger than ex-Ti₃C₂ (106.2 mA h g⁻¹) and the NPCNs (280.9 mA h g⁻¹), and long cycling stability of 252.2 mA h g⁻¹ with only 0.03% degradation per cycle within 2000 cycles at 1.0 A g⁻¹. This effect is attributed to the hybrids with stacked structure ensuring intimate contact between Ti₃C₂ and the NPCNs to efficiently take advantage of both components. The PDDA-NPCN/Ti₃C₂ hybrids afford enlarged interlayer spacing, a unique 3D porous interconnected conductive network and abundant utilization of active sites, which shortens the electrolyte ion transport pathways and facilitates rapid ionic/electronic transport. The robust hybrids show a durable rate capability and cycling stability through alleviating the volumetric changes caused by phase transformations during the fast charge/discharge process. DFT calculations further reveal that the PDDA-NPCN/Ti₃C₂ hybrids help with synergistically ameliorating the electrochemical performance by enhancing interfacial K⁺ adsorption reactions and facilitate the potassiation process. This method opens up the possibility to controllably explore self-assembly of coupled hybrids in energy storage devices.

Methods

Methods and any associated references are available in the online version of the paper.

Data availability

All data used in this study are available from the corresponding authors upon reasonable request.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge support from the project supported by the Sate Key Program of National Natural Science of China (no. 51532005), National Nature Science Foundation of China (no. 51472148, 51272137, 51602181, 51902188), General Financial Grant from the China Postdoctoral Science Foundation (no. 2015M582088), and the Fundamental Research Fund of Shandong University.

Notes and references

1. X. Wu, D. P. Leonard and X. Ji, Chem. Mater., 2017, 29, 5031–5042.
2. J. C. Pramudita, D. Sehrawat, D. Goonetilleke and N. Sharma, Adv. Energy Mater., 2017, 7, 1602911.
3. Y.-S. Xu, S.-Y. Duan, Y.-G. Sun, D.-S. Bin, X.-S. Tao, D. Zhang, Y. Liu, A.-M. Cao and L.-J. Wan, J. Mater. Chem. A, 2019, 7, 4334–4352.
4. J.-Y. Hwang, S.-T. Myung and Y.-K. Sun, Adv. Funct. Mater., 2018, 28, 1802938.
5. X. Zou, P. Xiong, J. Zhao, J. Hu, Z. Liu and Y. Xu, Phys. Chem. Chem. Phys., 2017, 19, 26495–26506.
6. Q. Zhao, J. Wang, Y. Lu, Y. Li, G. Liang and J. Chen, Angew. Chem., Int. Ed., 2016, 55, 12528–12532.
7. S. Komaba, T. Hasegawa, M. Dahbi and K. Kubota, Electrochem. Commun., 2015, 60, 172–175.
8. B. Ji, F. Zhang, X. Song and Y. Tang, Adv. Mater., 2017, 29, 1700519.
9. K. Lei, F. Li, C. Mu, J. Wang, Q. Zhao, C. Chen and J. Chen, Energy Environ. Sci., 2017, 10, 552–557.
10. W. Zhang, J. Mao, S. Li, Z. Chen and Z. Guo, J. Am. Chem. Soc., 2017, 139, 3316–3319.
11. X. Sang, Y. Xie, M. W. Lin, M. Alhabeb, K. L. Van Aken, Y. Gogotsi, P. R. Kent, K. Xiao and R. R. Unocic, ACS Nano, 2016, 10, 9193–9200.
12. Y. Xie, Y. Dall’Agnese, M. Naguib, Y. Gogotsi, M. W. Barsoum, H. L. Zhuang and P. R. Kent, ACS Nano, 2014, 8, 9606–9615.
13. J. Wen, Q. Fu, W. Wu, H. Gao, X. Zhang and B. Wang, ACS Appl. Mater. Interfaces, 2019, 11, 7087–7095.
14. D. Er, J. Li, M. Naguib, Y. Gogotsi and V. B. Shenoy, ACS Appl. Mater. Interfaces, 2014, 6, 11173–11179.
15. R. Zhao, M. Wang, D. Zhao, H. Li, C. Wang and L. Yin, ACS Energy Lett., 2017, 3, 132–140.
16. D. Zhao, R. Zhao, S. Dong, X. Miao, Z. Zhang, C. Wang and L. Yin, Energy Environ. Sci., 2019, 12, 2422–2432.
17. J. Li, X. Yuan, C. Lin, Y. Yang, L. Xu, X. Du, J. Xie, J. Lin and J. Sun, Adv. Energy Mater., 2017, 7, 1602725.
18. X. Hui, R. Zhao, P. Zhang, C. Li, C. Wang and L. Yin, Adv. Energy Mater., 2019, 9, 1901065.
19. P. Lian, Y. Dong, Z.-S. Wu, S. Zheng, X. Wang, W. Sen, C. Sun, J. Qin, X. Shi and X. Bao, Nano Energy, 2017, 40, 1–8.
20. M. Naguib, R. A. Adams, Y. Zhao, D. Zemlyanov, A. Varma, J. Nanda and V. G. Pol, Chem. Commun., 2017, 53, 6883–6886.
21. H. Huang, J. Cui, G. Liu, R. Bi and L. Zhang, ACS Nano, 2019, 13, 3448–3456.
22. Y. Tian, Y. An, S. Xiong, J. Feng and Y. Qian, J. Mater. Chem. A, 2019, 7, 9716–9725.
23. R. Ma, X. Liu, J. Liang, Y. Bando and T. Sasaki, Adv. Mater., 2014, 26, 4173–4178.
24. P. Xiong, R. Ma, N. Sakai, L. Nurdiwijayanto and T. Sasaki, ACS Energy Lett., 2018, 3, 997–1005.
25. R. Zhao, Z. Qian, Z. Liu, D. Zhao, X. Hui, G. Jiang, C. Wang and L. Yin, Nano Energy, 2019, 65, 104037.
26. R. Zhao, L. Zhang, C. Wang and L. Yin, J. Power Sources, 2017, 353, 77–84.
27. D. Li, X. Ren, Q. Ai, Q. Sun, L. Zhu, Y. Liu, Z. Liang, R. Peng, P. Si, J. Lou, J. Feng and L. Ci, Adv. Energy Mater., 2018, 8, 1802386.
28. L. Fang, J. Xu, S. Sun, B. Lin, Q. Guo, D. Luo and H. Xia, Small, 2019, 15, 1804806.
29. W. Luo, J. Wan, B. Ozdemir, W. Bao, Y. Chen, J. Dai, H. Lin, Y. Xu, F. Gu, V. Barone and L. Hu, Nano Lett., 2015, 15, 7671–7677.
30. Y. Dong, Z. S. Wu, S. Zheng, X. Wang, J. Qin, S. Wang, X. Shi and X. Bao, ACS Nano, 2017, 11, 4792–4800.
31. G. Liao, Y. Gong, L. Zhang, H. Gao, G.-J. Yang and B. Fang, Energy Environ. Sci., 2019, 12, 2080–2147.
32. H. He, D. Huang, Y. Tang, Q. Wang, X. Ji, H. Wang and Z. Guo, Nano Energy, 2019, 57, 728–736.
33. K. Share, A. P. Cohn, R. Carter, B. Rogers and C. L. Pint, ACS Nano, 2016, 10, 9738–9744.
34. J. Yan, C. E. Ren, K. Maleski, C. B. Hatter, B. Anasori, P. Urbankowski, A. Sarycheva and Y. Gogotsi, Adv. Funct. Mater., 2017, 27, 1701264.
35. X. Ge, C. Gu, Z. Yin, X. Wang, J. Tu and J. Li, Nano Energy, 2016, 20, 185–193.
36. M. Boota, B. Anasori, C. Voigt, M. Q. Zhao, M. W. Barsoum and Y. Gogotsi, Adv. Mater., 2016, 28, 1517–1522.
37. W. Zhao, J. Peng, W. Wang, B. Jin, T. Chen, S. Liu, Q. Zhao and W. Huang, Small, 2019, 15, 1901351.
38. A. Lipatov, M. Alhabeb, M. R. Lukatskaya, A. Boson, Y. Gogotsi and A. Sinitskii, Adv. Electron. Mater., 2016, 2, 1600255.
39. S. Liu, J. Zhou and H. Song, Chem. Commun., 2018, 54, 9825–9828.
40. S. Liu, J. Zhou and H. Song, Adv. Energy Mater., 2018, 8, 1800569.
41. C. Li, S. Dong, R. Tang, X. Ge, Z. Zhang, C. Wang, Y. Lu and L. Yin, Energy Environ. Sci., 2018, 11, 3201–3211.
42. Y. Xie, Y. Chen, L. Liu, P. Tao, M. Fan, N. Xu, X. Shen and C. Yan, Adv. Mater., 2017, 29, 1702268.
43. J. Gu, Z. Du, C. Zhang and S. Yang, Adv. Energy Mater., 2016, 6, 1600917.
44. Q. Xue, H. Zhang, M. Zhu, Z. Pei, H. Li, Z. Wang, Y. Huang, Y. Huang, Q. Deng, J. Zhou, S. Du, Q. Huang and C. Zhi, Adv. Mater., 2017, 29, 1604847.
45. Y. Huo, Y. Jin, J. Zhu and H. Li, Appl. Catal., B, 2009, 89, 543–550.
46. G. Yang, Z. Jiang, H. Shi, T. Xiao and Z. Yan, J. Mater. Chem., 2010, 20, 5301–5309.
47. W. Bao, L. Liu, C. Wang, S. Choi, D. Wang and G. Wang, Adv. Energy Mater., 2018, 8, 1702485.
48. Z. Jian, W. Luo and X. Ji, J. Am. Chem. Soc., 2015, 137, 11566–11569.
49. G. Ma, K. Huang, J.-S. Ma, Z. Ju, Z. Xing and Q.-C. Zhuang, J. Mater. Chem. A, 2017, 5, 7854–7861.
50. X. Qi, K. Huang, X. Wu, W. Zhao, H. Wang, Q. Zhuang and Z. Ju, Carbon, 2018, 131, 79–85.
51. Z. Ju, P. Li, G. Ma, Z. Xing, Q. Zhuang and Y. Qian, Energy Storage Mater., 2018, 11, 38–46.
52. Y. An, H. Fei, G. Zeng, L. Ci, B. Xi, S. Xiong and J. Feng, J. Power Sources, 2018, 378, 66–72.
53. J. Yang, Z. Ju, Y. Jiang, Z. Xing, B. Xi, J. Feng and S. Xiong, Adv. Mater., 2018, 30, 1700104.
54. Z. Tai, Q. Zhang, Y. Liu, H. Liu and S. Dou, Carbon, 2017, 123, 54–61.
55. Y. Sun, H. Xiao, H. Li, Y. He, Y. Zhang, Y. Hu, Z. Ju, Q. Zhuang and Y. Cui, Chem. – Eur. J., 2019, 25, 7359–7365.
56. Q. Gan, J. Xie, Y. Zhu, F. Zhang, P. Zhang, Z. He and S. Liu, ACS Appl. Mater. Interfaces, 2019, 11, 930–939.
57. A. P. Cohn, N. Muralidharan, R. Carter, K. Share, L. Oakes and C. L. Pint, J. Mater. Chem. A, 2016, 4, 14954–14959.
58. R. A. Adams, J. M. Syu, Y. Zhao, C. T. Lo, A. Varma and V. G. Pol, ACS Appl. Mater. Interfaces, 2017, 9, 17872–17881.
59. L. Kong, J. Zhu, W. Shuang and X.-H. Bu, Adv. Energy Mater., 2018, 8, 1801515.
60. L. Wei, Z. Hou and H. Wei, Electrochim. Acta, 2017, 229, 445–451.
61. G. SandI, R. E. Winans and K. A. Carrado, J. Electrochem. Soc., 1996, 143, L95–L98.
62. H. Gao, L. Song, J. Niu, C. Zhang, T. Kou, Y. Sun, J. Qin, Z. Peng and Z. Zhang, J. Mater. Chem. A, 2019, 7, 13602–13613.
63. Y. Zhang, H. Gao, J. Niu, W. Ma, Y. Shi, M. Song, Z. Peng and Z. Zhang, ACS Nano, 2018, 12, 11678–11688.
64. H. Zhang, K. Xi, K. Jiang, X. Zhang, Z. Liu, S. Guo and H. Zhou, Chem. Commun., 2019, 55, 7910–7913.
65. J. C. Chacón-Torres, L. Wirtz and T. Pichler, Phys. Status Solidi B, 2014, 251, 2337–2355.
66. X. Zhao, P. Xiong, J. Meng, Y. Liang, J. Wang and Y. Xu, J. Mater. Chem. A, 2017, 5, 19237–19244.
67. J. Huang, X. Lin, H. Tan and B. Zhang, Adv. Energy Mater., 2018, 8, 1703496.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9ee03250a

---