The schematically synthetic processes of hierarchical porous Mo-Co3O4-CNTc composites were depicted in Fig. 1a in the oil/water emulsion via a typical and simple sol-gel method, in which the Mo-Co(OH)x-CNTc composites as precursors were obtained and the Mo-Co3O4-CNTc composites in the annealing process corresponding to the CNTc content of 21.8% (Fig. S1). To further confirm the crystal structure of the Mo-Co3O4-CNTc composites to the cubic phase (JCPDS no. 42-1467), the XRD patterns (Fig. 1b) represent the diffraction peaks (111), (220), (311), (222), (400), (422), (511) and (440) planes, respectively. Furthermore, the chemical composition and state were determined via XPS (Fig. S2a). Co 2p spectra (with shake-up satellites (“Sat.”) at 787.6 and 804.2 eV) can be detected for Co3+ and Co2+ (Fig. 1c), thus indicating that the fitting peaks at banding energy of 780.5 and 795.5 eV are assigned to Co3+, and the fitting peaks at 782.0 and 797.0 eV can be ascribed to Co2+ [29, 30]. The Mo 3d spectrum (Fig. 1d) can be assigned to Mo 3d3/2 at 235.3 eV and Mo 3d5/2 at 232.2 eV, thus indicating the existence of Mo6+ with a width of 3.1 eV in the Mo-Co3O4-CNTc composites [31, 32]. Besides, the spectrum of O1s can be resolved as the lattice oxide ions O2– at 530.3 eV, defective oxide ions Ox– at 531.5 eV, and adsorbed surface water at 533.5 eV in Fig. 1e, respectively. The two peaks at 284.8 and 286.2 eV should be respectively divided into C–C/C = C and C–O–C (Fig. S2b). So these can further confirm the successful preparation of the Mo-Co3O4-CNTc composites.
The detailed morphologies of obtained Mo-Co3O4-CNTc composites can be observed from SEM imagines (Fig. 2a−b). Compared with the Mo-Co3O4 electrode materials (Fig. S3), the Mo-Co3O4-CNTc composites are composed of intertwisted and crinkly nanosheets to form the hierarchically porous structures. Meanwhile, the CNTs can be uniformly mingled and inserted into the Mo-Co3O4 nanosheets as express electron transport channels (Fig. S4). Moreover, the detailed morphology of the hierarchically porous structures can be identified by TEM (Fig. 2c−f). The cross-linking and doping CNTs are combined with the Mo-Co3O4 nanosheets as interconnected electric network to facilitate the transfer of electron. Notably, the interlaced ultrathin nanosheets reveal thickness of 2–4 nm and substantial mesoporous scale holes as shown in Fig. 2d. Meanwhile, the hierarchical mesoporous structures of Mo-Co3O4 nanosheets combined with CNTs as electric network are beneficial for the rapid electrolyte ion diffusion and fast electrons transport with low resistance, respectively. Accordingly, the HRTEM image Fig. 2f (inset) delivers the lattice spaces of 0.28, 0.23 and 0.20 nm corresponding to the (220), (222) and (400) planes of Mo-Co3O4, indicating high crystallinity and the polycrystalline nature of the Mo-Co3O4 nanoparticles, respectively. Meanwhile, lattice spaces of 0.34 nm can be detected as the (002) planes of CNTs in the Mo-Co3O4-CNTc composites. Additionally, the EDS pattern (inset Fig. 2d) represents Co, Mo, O, and C elements, thus further indicating the successful preparation of Mo-Co3O4-CNTc composites. The elemental mapping analysis can directly display that these elements are distributed homogeneously on the entire Mo-Co3O4 nanosheets as shown in Fig. 2g−k, which is consistent with the observation from XPS results.
Due to such excellent typical structures, the electrochemical performance of as-prepared Mo-Co3O4-CNTc composites was systematically evaluated in the three electrode configuration. Compared with the pure Mo-Co3O4 electrode materials, the Mo-Co3O4-CNTc composites exhibit superior electrochemical properties based on the CV curves at 50 mV s–1 (Fig. 3a), GCD curves at 0.5 A g–1 (Fig. 3b), and average capacity (four samples) at various current densities (Fig. 3c), respectively. Moreover, the pure Co3O4 and a series of Mo-Co3O4 electrode materials with different Mo-Co molar ratio of 1:10, 5:10, and 10:10 were evaluated for comparison (Fig. 3d), indicating the superior electrochemical properties molar ratio of 1:10. The CV curves of Mo-Co3O4-CNTc cathode materials are clearly exhibited in Fig. 3e with obvious battery-type features at multiple scan rates from 0.5 to 50 mV s–1, in which the oxidative peaks shift toward more positively and reductive peaks shift toward more negative values with the increasing scan rates due to polarization effect and more reversible redox reactions. Furthermore, Fig. 3f shows the log i and log v plots at peak currents, and the b-values can be calculated determined to 0.775 and 0.845 (in the range of 0.5–1.0) by the Dunn method [33] according to the Eq. (1). Consequently, the as-prepared Mo-Co3O4-CNTc cathode materials represent both battery-type and pseudocapacitive-type characteristics.
$$\text{l}\text{o}\text{g}i=b\text{l}\text{o}\text{g}v+\text{l}\text{o}\text{g}a$$
1
Moreover, the capacitive contribution for the total current at 1 mV s− 1 was shown in Fig. 3g. The contribution ratio of the capacitive and diffusion-controlled capacity at various scan rates (Fig. 3h) can be calculated as the following equations [34, 35]:
$$I={I}_{\text{c}\text{a}\text{p}}+ {I}_{diff}={av}^{b}$$
2
Wherein, Icap and Idiff are the surface capacitance-led and diffusion-controlled in the instant current, respectively. The capacitive-controlled process are 46.7%, 49.9%, 53.3%, 55.8%, 59.5%, 65.5%, 72.2%, 79.3% and 91.3% at 0.5, 1, 2, 3, 5, 10, 20, 30 and 50 mV s–1, respectively. Additionally, the typical GCD profiles at various current densities deliver the remarkable specific capacity of 152.9 mAh g–1 at 0.5 A g–1 and 82.7 mAh g–1 at 40 A g–1, reaching 54.1% capacity retention in Fig. 3i. Compared with the pure Co3O4 and a series of Mo-Co3O4 electrode materials, the rate performance displays enhanced specific capacity at the lower current density of 0.5 A g–1 in the initial 5 cycles, and so that in the last 15 cycles corresponding to the good structural stability in Fig. 3j. Furthermore, Fig. 3k delivers excellent cycling performance with 80.3% capacity retention even over 4000 GCD cycles at 25 A g–1 and high columbic efficiency of 99.6%. Additionally, the Nyquist plots (Fig. 3l) of Mo-Co3O4-CNTc deliver lower electrochemical resistance (Rs) around 0.43 Ω and charge transfer resistance (Rct) 0.53 Ω than pure Co3O4 and Mo-Co3O4 electrode materials for comparison due to the typical 3D interpenetrating CNTs conductive networks as the “express channels” (Fig. S5), respectively.
As illustrated in Fig. 4a, the MoCo-Zn batteries were assembled with the hierarchically porous Mo-Co3O4-CNTc composites as advanced cathode materials and zinc anode in the electrolyte of 6 M KOH aqueous with 0.2 M zinc acetate. The CV curves of the MoCo-Zn batteries exhibit similar shapes with obvious redox peaks at the increasing scan rate from 0.5 to 50 mV s–1 (Fig. 4b). The b values of the MoCo-Zn batteries can be calculated as 0.786 and 0.746 by Dunn methods [33] as shown in Fig. 4c, thus revealing the coexistence of battery-type and pseudocapacitive-type characteristics. Moreover, the contribution ratio of capacitive and diffusion-controlled reactions were exhibited in Fig. 4d as 48.0%, 51.5%, 53.8%, 55.3%, 56.6%, 57.6%, 59.4%, 61.1%, 65.3%, 69.0%, 75.1%, 81.6% and 86.8% at various scan rates of 1, 2, 3, 4, 5, 6, 8, 10, 15, 20, 30, 40 and 50 mV s–1, respectively. Furthermore, the GCD curves with an average discharge platform of around 1.68 V represent the voltage window of 1.93 V from 1 to 30 A g–1 as shown in Fig. 4e, and deliver the specific capacity of 195.7 mAh g–1 at 0.5 A g–1 and 97.6 mAh g–1 at 30 A g–1 (with capacity utilization of 49.9%), respectively. The MoCo-Zn batteries display outstanding rate performance and columbic efficiency in Fig. 4f, thus demonstrating the good structural stability. Meanwhile, the energy density and power density can be evaluated in the Ragone plots (Fig. 4g) as 237.6 Wh kg–1 at 1692.4 W kg–1 and 162.7 Wh kg–1 at 50032.0 W kg–1, respectively. Compared with the Mo-Co based supercapacitors and other aqueous rechargeable ZIBs, the as-prpared MoCo-Zn batteries exhibited a superior level in energy density, such as CoMoO4–x//AC 62.3 Wh kg–1 at 800 W kg–1 [36], ZnCo2O4@CoMoO4//AC 29.24 Wh kg–1 at 884.57 W kg–1 [37], CoMoO4@Ni(OH)2//AC 62.5 Wh kg–1 at 776 W kg–1 [38], NiMoO4/CoMoO4//AC 33.1 Wh kg–1 at 199.6 W kg–1 [39], Zn//Co3O4 241 Wh kg–1 at 1487.7 W kg–1 [40], Zn//NiCo 210.1 Wh kg–1 at 11600 W kg–1 [41], Zn//core-shell Co3O4@δ-MnO2/CC 212.8 Wh kg–1 at 313.3 W kg–1 [42], Zn//MnO2 254 Wh kg–1 at 197 W kg–1 [43], Zn//P-MoO3–x@Al2O3 240 Wh kg–1 at 931.3 W kg–1 [44], Zn//LiVPO4F-CNTs@PPy 235.6 Wh kg–1 at 320.8 W kg–1 [45]. More strikingly, the MoCo-Zn batteries exhibited excellent cycling performance with 85.1% capacity retention over 10000 cycles at 25 A g–1 and the capacity also represented any decay at the initial 2000 cycles (Fig. 4h). Meanwhile, the Mo-Co3O4-CNTc composites also possess hierarchical porous structures with opened space as “ion-buffering reservoirs” [46−48], which impressively outperformed most aqueous rechargeable zinc ion batteries. Furthermore, the coulombic efficiency of MoCo-Zn batteries is nearly 100%. The inset displays GCD curves at different cycles from 1st to 10000th, thus indicating the changes of GCD curves in the long cycle life including capacity decay, electrode polarization, stabilization of coulombic efficiency and displacement of discharge platform. Finally, the LEDs (2.2 V, 0.06 W) could be lit up by the series of MoCo-Zn devices for potential applications as demonstrated (Fig. 4h the inset image).
In order to further investigate the MoCo-Zn batteries, we have explored the zinc ions storage mechanism of Mo-Co3O4-CNTc cathode materials via the ex-situ XRD patterns, Raman spectra, and XPS measurements to display the structural evolution at some certain voltages. Figure 5a represents the schematic illustration of charge-discharge process with Zn2+ intercalation/de-intercalation. More concretely, Fig. 5b represents the selecting different labeled states of C0, C1, C2, C3, C4, D2, D1, and D0 (C represents for charge and D for discharge) in the charge-discharge process. As shown in ex-situ XRD patterns (Fig. 5c), the diffraction peaks shift at around 2θ = 20o corresponding to the (111) planes of Co3O4 after the intercalation/deintercalation of Zn2+ during the charge-discharge process. Simultaneously, the new diffraction peaks appear in the range of 11–13o, thus signifying new layered of α-Co(OH)2 on the surface of C3, C4, D2, and D1 states, respectively. Moreover, the ex-situ Raman spectra (Fig. 5d) show that the peaks shift to higher at around 666 cm–1 in the states of C2, C3, C4, D2, and D1 according to the Zn2+ ingress/egress. Additionally, more detailed information of chemical composition and states can be further investigated in the ex-XPS measurements (Fig. S6). Compared with the state of C0 without Zn 2p region, Zn 2p spectrum (Fig. 5e) can be detected as the absorbed Zn2+ Zn 2p3/2 at 1022.0 eV and Zn 2p1/2 at 1045.1 eV, and intercalated Zn2+ at Zn 2p3/2 at 1021.4 eV and Zn 2p1/2 at 1044.5 eV [49], respectively. During the discharge/charge process, it remarkably indicates the Zn2+ intercalation/de-intercalation with increasing peaks of intercalated Zn2+ in the states of C1, C2, C3, C4 and decreasing peaks of intercalated Zn2+ in the states of D2, D1, D0, respectively. Furthermore, the Mo 3d spectrum (Fig. 5f) of Mo-Co3O4-CNTc cathode materials can be deconvoluted as Mo 3d5/2 and Mo 3d3/2, corresponding to Mo6+ at 232.2 eV, Mo4+ at 231.7 eV, Mo6+ at 235.3 eV, and Mo4+ at 234.9 eV, respectively, thus indicating the electrochemical reaction between the redox couples of Mo6+/Mo4+ during the charge-discharge processes. Similarly, Co 2p spectra can be identified as the fitting peaks at 780.5 and 795.5 eV for Co3+, and 782.0 and 797.0 eV for Co2+ [50, 51], respectively. Remarkably, the peaks of C3, C4, and D2 shift to lower binding energy due to more electrochemical oxidation of Co3+ as shown in Fig. 5g. Furthermore, the C4 charge state of Mo-Co3O4-CNTc composites can keep the hierarchically porous structures with nanosheets and the interpenetrating conductive networks of CNTc in the TEM imagines (Fig. 6a−c). Moreover, the EDS elemental pattern of C4 charge state (Fig. 6d) represents the homogeneous distribution of Zn, O, Co, Mo, and C elements, thus further indicating the Zn2+ intercalation/de-intercalation in the whole Mo-Co3O4-CNTc composites and a good agreement with XPS measurements.
To gain deep insight for the interaction between CNT and Mo-Co3O4, the structural and electronic properties of Mo-Co3O4-CNT system were investigated by density-functional first-principles calculations [52–54]. The optimized structures and corresponding plane-averaged electrostatic potentials of Co3O4, Mo-Co3O4, and Mo-Co3O4-CNT have been calculated and displayed in Fig. 7a−c. There are no chemical bonds formed at the interface indicating a typical Van der Waals (vdW) interaction between CNT and Mo-Co3O4. Owing to the potential difference, an internal electric field was formed at the interface, which is beneficial to charge transfer [55, 56]. In addition, the calculated work function of the Mo-Co3O4-CNT (as 4.6 eV) is lower than that of the Co3O4 surface (as 6.1 eV) and Mo-Co3O4 surface (as 5.4 eV). The smaller work function means less loss when electrons escape to the surface for electron emission. It suggests the Mo-Co3O4-CNT composite is more beneficial to realize high electrical conductivity. The charge density difference and plane-averaged charge density difference of Co3O4, Mo-Co3O4, and Mo-Co3O4-CNT are plotted in Fig. 7d−f. The positive (yellow region) and negative (cyan region) values indicate the charge accumulation and depletion, respectively. The Bader charge analysis shows that 0.06 e per supercell has been transferred from CNT to Mo-Co3O4. It suggests that forming Mo-Co3O4-CNT interfaces may improve the electron transport performance of Mo-Co3O4 surface. To further study the interfacial contact properties, the atom-projected density of states (DOS) were analyzed in Fig. 7g−i. It can be seen that the increase in the density of states around the Fermi level that will result in increased conduction at elevated energies. CNT can alter the density of states and therefore alter the conductivity at the interface without damaging the significant characteristics of Mo-Co3O4 surface [57]. It is in good agreement with the EIS measurements.