carbon nanotube hybrid papers as binder-free anodes for potassium-ion batteries

carbon nanotube hybrid papers as binder-free anodes for potassium-ion batteries

Journal of Physics and Chemistry of Solids 138 (2020) 109296 Contents lists available at ScienceDirect Journal of Physics and Chemistry of Solids jo...

2MB Sizes 0 Downloads 15 Views

Journal of Physics and Chemistry of Solids 138 (2020) 109296

Contents lists available at ScienceDirect

Journal of Physics and Chemistry of Solids journal homepage: http://www.elsevier.com/locate/jpcs

Electrochemical performance of reduced graphene oxide/carbon nanotube hybrid papers as binder-free anodes for potassium-ion batteries Shuting Peng a, 1, Lichang Wang b, 1, Ziqiang Zhu b, Kai Han a, * a

Hunan Provincial Key Laboratory of Efficient and Clean Utilization of Manganese Resources, College of Chemistry and Chemical Engineering, Central South University, Changsha, 410083, Hunan, China b School of Geosciences and Info-Physics, Central South University, Changsha, 410083, Hunan, China

A R T I C L E I N F O

A B S T R A C T

Keywords: Potassium-ion batteries Anode Graphene Carbon nanotube Binder-free

We investigated the electrochemical performance of reduced graphene oxide (rGO)/carbon nanotube (CNT) hybrid papers as binder-free anodes for potassium-ion batteries. RGO/CNT hybrid papers were fabricated by vacuum filtration and thermal reduction. The potassium-ion storage ability of rGO paper was significantly enhanced by CNT incorporation and the electrochemical performance of the hybrid papers was related to the weight ratio of CNTs. The optimal rGO/CNT-30% paper can deliver an initial reversible discharge capacity of 351 and 223 mA h g 1 at a current density of 10 and 50 mA g 1, respectively. A capacity of 148 mA h g 1 was achieved after 200 cycles at a current density of 50 mA g 1. The enhanced performance of rGO/CNT hybrid papers is attributed to the formation of a 3D conductive carbon network, in which the graphene layers are bridged by CNTs. Such a structure could effectively facilitate the transport of electrons and potassium ions and provide more storage sites for potassium ions.

1. Introduction Lithium-based secondary batteries have been broadly applied in energy storage areas during the past few decades. However, the high cost, shortage, and uneven global distribution of lithium resources have led to growing concerns, in particular with the rapid development of electric vehicles and renewable energy generation technologies [1–3]. Potassium-ion batteries (KIBs) and sodium-ion batteries (NIBs) have attracted extensive attention owing to their electrochemical principles being similar to those of lithium-ion batteries (LIBs) [4–6]. Moreover, they have been considered as promising alternative secondary battery systems owing to Earth’s rich abundant resources of potassium and so­ dium [7–9]. Since the larger radii of Kþ (1.33 Å) and Naþ (0.97 Å) in comparison with Liþ (0.68 Å) inevitably induce slow kinetics of the electrochemical processes, the main challenge is to explore high-performance electrode materials for both KIBs and NIBs [10–13]. Most recently, Kubota et al. [14]reported that Kþ and Naþ ions show weaker Lewis acidity than Liþ ions, resulting in smaller Stokes radii in various solvents in the order Kþ (3.6 Å) < Naþ (4.6 Å) < Liþ (4.8 Å). This demonstration suggests that Kþ and Naþ could display higher ionic mobility and conductivity than Liþ in an electrolyte [15]. In addition,

KIBs typically possess a lower redox potential (Eþ K /K ¼ 2.9 V) than NIBs (Eþ 2.7 V), theoretically suggesting a higher output energy Na/Na ¼ density of KIBs [16]. These demonstrations further enhance the prom­ ising application of KIBs in the future. Compared with NIBs, which have been increasingly investigated in the past few years, KIBs are still in their infancy regarding their study. Graphite, the most commonly used anode for LIBs, can deliver only a low reversible capacity of about 35 mA h g 1 when applied as an anode for NIBs by forming NaC64 [17–20]. Differently from the sodium ion, the larger-radius potassium ion can reversibly insert itself into graphite and consequently deliver a reversible specific capacity of 273 mA h g 1, which was demonstrated by Jian et al. [21] and Wen et al. [22]. It has been proved that three kinds of potassium-intercalated graphite com­ pounds (KC36, KC24, and KC8) are generated at three stages during the graphite electrochemical potassiation process [21]. Inspired by this, carbon-based materials, especially graphitic carbons, have attracted extensive attention as anodes for KIBs instead of NIBs. Graphene, a single layer of graphite, is a widely used 2D carbon material with large surface area and high conductivity and was first successfully synthesized in 2004 [23–27]. Benefiting from its unique 2D structure, graphene has been viewed as an ideal choice. It can be built

* Corresponding author. E-mail address: [email protected] (K. Han). 1 These two authors contributed equally. https://doi.org/10.1016/j.jpcs.2019.109296 Received 24 June 2019; Received in revised form 1 December 2019; Accepted 2 December 2019 Available online 2 December 2019 0022-3697/© 2019 Elsevier Ltd. All rights reserved.

S. Peng et al.

Journal of Physics and Chemistry of Solids 138 (2020) 109296

Fig. 1. Fabrication process for reduced graphene oxide (rGO) and rGO/carbon nanotube (CNT) hybrid papers and typical digital photographs of the papers. GO, graphene oxide.

into other macroscopic structures such as thin films and free-standing papers with high mechanical stability and flexibility, which can be adopted for use as electrodes for batteries [28–30]. When graphene film or paper is applied as an anode for LIBs, although theoretically it has greater lithium storage ability than graphite, the practical electro­ chemical performance is normally unsatisfactory. The main reason is the barriers for electron and ion diffusion in the cross-plane direction once graphene sheets have been tightly restacked to form the film or paper [8, 16,31–33]. KIBs also face the same problem, and to solve this issue, much work has been done to modify graphene, such N doping, S doping, and B doping [16]. Apart from the aforementioned chemical ways to improve the electrochemical performance of graphene for KIBs, we suggested a physical and hybrid way to increase the interlayer spacing of graphene and achieve excellent performance for KIBs. Carbon nanotubes (CNTs) as a widely used incorporated material make the hybrid method possible [34–38]. Because of their high conductivity and 1D structure, CNTs have been used to bridge graphene layers and thus produce a 3D conductive carbon network, which maintained the high flexibility and mechanical strength of graphene-based carbon papers [39–44]. The CNT/rGO composite materials have been well applied in emerging en­ ergy storage and exhibited good performance [45–50]. In this work, we investigate the electrochemical performance of reduced graphene oxide (rGO)/CNT hybrid papers as high-potential binder-free anodes for KIBs. The rGO/CNT hybrid papers composed of a 3D conductive carbon network were fabricated by vacuum filtration and thermal reduction [39]. We combined rGO and CNT into a 3D network in a facile physical way, which promoted Kþ ion diffusion in the in-plane and cross-plane directions. The incorporation of CNTs in rGO layers is expected to (1) prevent restacking and increase the interlayer spacing of rGO to provide more storage sites for potassium ions and (2) bridge rGO layers to facilitate electron and ion transport in the cross-plane direction. As a result, the potassium-ion storage ability of rGO paper is significantly enhanced by CNT incorporation and the electrochemical performance of the hybrid papers is found to depend on the weight ratio of CNT. The optimal rGO/CNT-30% paper can deliver an initial reversible discharge capacity of 351 and 223 mA h g 1 at a current density of 10 and 50 mA g 1, respectively. A capacity of 148 mA h g 1 was maintained after 200 cycles at a current density of 50 mA g 1. To the best of our knowledge, this is the first time the potential of rGO/CNT hybrid papers as anodes for KIBs has been explored. The re­ sults will extend the application of graphene-based materials beyond LIB systems and provide a material choice for KIBs.

2. Experimental 2.1. Fabrication of rGO and rGO/CNT papers Graphene oxide (GO) was initially synthesized by a modified Hum­ mers’ method to fully oxidize graphite [51]. The process was typically divided into two steps: preoxidation and oxidation as reported in our previous work [52,53]. In the present work, considering the poor dispersion of CNTs in aqueous solution, the purchased CNT powders (multiwalled CNTs, diameter 10–20 nm, length 10–50 μm, purity �95 wt%, Nanjing Xian Feng Nanomaterials) were pretreated by surface acid oxidation with concentrated sulfuric acid and nitric acid at 80 � C for 1 h with magnetic stirring. After centrifugation and washing with deionized water, the CNT aqueous suspension obtained was used for hybrid paper fabrication. To ensure a controllable vacuum filtration process and reproducibility of free-standing papers, 2.7 mL GO aqueous dispersion (5 mg mL 1) and 1.5 mL CNT aqueous suspension (1 mg mL 1) were dropped into a glass vial with 1 mL deionized water and then ultra­ sonicated for 30 min at a sonication power of 150 W. Next, the sus­ pension was vacuum filtered (Fig. 1) with use of an using anodic aluminum oxide membrane (47-mm diameter, 200-nm pore size, Whatman) and dried in air at room temperature for about 3 h. The free-standing GO/CNT paper was then peeled off the anodic aluminum oxide substrate; the mass of every whole paper was about 15 mg. The paper was then cut into circular disks of 10-mm diameter for further cell assembly. Lastly, to reduce GO, the GO/CNT paper disk was annealed in a tube furnace at 700 � C for 1 h in an argon atmosphere with a heating rate of 5 � C min 1 [39]. Finally, we obtained free-standing rGO/CNT papers. The weight ratio of CNT to GO was adjusted by our changing the volume of the CNT suspension for filtration and the corresponding samples were denoted as rGO/CNT-10%, rGO/CNT-20%, and rGO/CNT-30%. Pure rGO paper was also prepared as a control sample by our just filtering the GO dispersion. 2.2. Characterization and electrochemical measurements The morphologies of the free-standing rGO and rGO/CNT papers were examined with a field-emission scanning electron microscope (SEM; FEI Helios Nanolab 600i). X-ray diffraction (XRD) patterns were obtained with a Bruker D8 Advance instrument (Cu Kα, λ ¼ 1.5418 Å). Raman spectra were obtained with a Raman system from Renish. X-ray photoelectron spectroscopy (XPS) was performed with a Thermo Sci­ entific ESCALAB 250Xi instrument with Al Kα radiation (1486.6 eV) as the excitation source. Conductivity characteristics were investigated 2

S. Peng et al.

Journal of Physics and Chemistry of Solids 138 (2020) 109296

Fig. 2. Top-view surface and cross-section scanning electron microscope images of (a, b) graphene oxide, (c, d) reduced graphene oxide (rGO), (e, f) rGO/carbon nanotube (CNT)-10%, (g, h) rGO/CNT-20%, and (i, j) rGO/CNT-30% free-standing papers.

with RTS-8 four-point probes. The electrochemical performances were studied with use of a two-electrode coin-type half cell (CR 2032) with potassium metal as the counter electrode. The free-standing rGO and rGO/CNT papers were directly applied as binder-free anodes for cell assembly without any further treatment. All cells were assembled in a glove box with a full argon atmosphere. The electrolyte was 0.8 M KPF6 dissolved in ethylene carbonate/dimethyl carbonate/ethyl methyl car­ bonate (1:1 by volume) solution. Cyclic voltammograms were obtained with an electrochemical workstation (CHI 660E) in the potential range from 0.01 to 3 V (vs. Kþ/K) at a scan rate of 0.1 mV s 1. Galvanostatic charge/discharge tests were conducted between 0.05 and 2.5 V versus Kþ/K with a LAND battery test system. Electrochemical impedance spectroscopy was performed with an electrochemical workstation (CHI 660E) in the frequency range from 100 kHz to 0.01 Hz under AC stim­ ulus with an amplitude of 10 mV. All current densities and capacities were calculated on the basis of the entire weight of rGO/CNT hybrid papers. To ensure the reproducibility, the electrochemical performance of each paper anode was tested at least four times.

paper thickness is mainly due to the incorporation of CNTs. On the one hand, CNTs can prevent tight restacking of GO nanosheets and increase the interlayer spacing of GO paper to provide enough escape channels for CO2 and H2O gas at high temperature. According to Bragg’s law, the interlayer spacings of GO, GO/CNT-10%, GO/CNT-20%, and GO/CNT30% were 0.449, 0.485, 0.524, and 0.463 nm, respectively, as calculated from the XRD patterns in Fig. S1. On the other hand, the presence of CNTs in close contact with GO nanosheets owing to their strong π-π interaction during the paper fabrication process prevented the interlayer gap expanding as much as in pure rGO paper during thermal treatment [39,44]. From the high-magnification cross-section SEM images in Fig. S2, graphene layers were also seen to be bridged by distributed CNTs to form a sandwich-like 3D structure. Such a 3D carbon structure could potentially exhibit excellent electrical and ion conductivity in both the in-plane direction and the cross-plane direction for the hybrid papers. XRD patterns of the as-prepared papers are shown in Fig. 3a. The characteristic peak at 2θ of 10� was observed for GO paper, which suggested the effective oxidation of graphite to GO [52–54]. The disappearance of such a peak for rGO paper indicated the successful reduction of GO under high-temperature treatment [52,53]. After CNT incorporation, rGO/CNT hybrid papers exhibited a typical graphitic carbon peak at 2θ of about 25� . Raman spectra were obtained to un­ derstand the structure evolution of the rGO/CNT hybrid papers, as shown in Fig. 3b. Two peaks corresponding to the D band (1342 cm 1) and G band (1593 cm 1) for carbon material were detected for all samples. The intensity ratio of the D and G bands (ID/IG) was calculated to explore the defect structure and graphitic carbon structure in the papers. The rGO/CNT-10%, rGO/CNT-20%, and rGO/CNT-30% hybrid papers had ID/IG values of 0.98, 1.05, and 0.97, respectively, which were much higher than those of GO paper (0.77) and rGO paper (0.71). This means that there are more defects in the hybrid papers, potentially supplying more defects for potassium-ion storage. This result could also be interpreted by SEM analysis, as we mentioned that the surface of GO was composed of more oxygen and hydrogen groups than the surface of rGO. After introduction of the preoxidized CNTs, the overall defects of the hybrid papers increased. The reduction of GO was further confirmed by the XPS C 1s peak analysis in Fig. 3c. GO paper showed four char­ acteristic peaks of C 1s at binding energies of 284.8, 286.7, 287.4, and – O, 288.64 eV, which corresponded to tetravalent C–C, C–O/C–O–C, C– – O groups, respectively [42]. For rGO and rGO/CNT-30% and O–C– – O and O–C– – O peaks disappeared and the intensity of the papers, the C– C–C peak was relatively enhanced from the C 1s spectra, suggesting an increase of the carbon-to-oxygen ratio in the rGO-based papers. The survey scan XPS results in Fig. S3 also reflect the reduction of GO. Moreover, as we assumed that the incorporation of CNTs could increase the overall conductivity of rGO paper by facilitating cross-plane electron transport, the conductivity of the papers was measured by the

3. Results and discussion The digital photographs in Fig. 1 show the free-standing papers during sample preparation. The GO and GO/CNT papers both possessed good flexibility, and could be bent back and forth several times without being broken. After thermal reduction, the flexibility of the rGO and rGO/CNT papers became much poorer because of the removal of oxygen-containing groups with the appearance of a metallic black color. However, the free-standing rGO and rGO/CNT papers still had sufficient mechanical strength for coin cell assembly without binder addition. As shown in Fig. 2a and b, the pure GO paper had a relatively flat surface and a typical laminate structure with a thickness of about 20 μm. This resulted from the tight restacking of GO nanosheets during vacuum filtration. Many wrinkles were observed on the surface of rGO paper after thermal treatment (Fig. 2c) and the thickness of the paper was increased to about 200 μm, an increase of nearly ten times (Fig. 2d). Obvious gaps were present between the graphene layers in rGO papers, which would lead to the slow transport of electrons and ions in the crossplane direction. The generation of wrinkles and the increase of the thickness were attributed to the escape of CO2 and H2O gas resulting from the removal of oxygen-containing groups at high temperature. In contrast, as shown in Fig. 2e, g, and i, the entire surface of rGO/CNT papers seemed much flatter than that of pure rGO paper. Meanwhile, more and more CNTs were observed on the surface of rGO/CNT papers as the weight ratio of CNT was increased from 10% to 20%–30%. More importantly, the thickness of rGO/CNT papers gradually decreased from about 35 μm for rGO/CNT-10% (Fig. 2f) to about 28 and 25 μm, respectively, for rGO/CNT-20% (Fig. 2h) and rGO/CNT-30% (Fig. 2j), which were much thinner than the pure rGO paper. The reduction of the 3

S. Peng et al.

Journal of Physics and Chemistry of Solids 138 (2020) 109296

Fig. 3. (a) X-ray diffraction patterns and (b) Raman spectra of graphene oxide (GO), reduced GO (rGO), and rGO/carbon nanotube (CNT) papers. (c) C 1s X-ray photoelectron spectroscopy spectra of GO, rGO, and rGO/CNT-30% papers. (d) Electrical conductivity comparison of GO, rGO, and rGO/CNT hybrid papers.

four-probe method; measured thicknesses of papers are given in Table S1. As shown in Fig. 3d, GO paper showed low conductivity (below 1 � 10 4 S cm 1) because of the presence of the oxygen-containing groups, while the conductivity was increased to about 16 � 10 4 S cm 1 for rGO paper after thermal reduction. Compared with pure rGO paper, the rGO/ CNT hybrid papers exhibited an obvious increase in overall conductiv­ ity, which gradually increased with increase of the CNT ratio. The rGO/ CNT-30% paper showed high conductivity of about 75 � 10 4 S cm 1, nearly five times higher than that of pure rGO paper. The increased conductivity is mainly attributed to the bridging role of CNTs in the 3D structure of rGO/CNT hybrid papers. Such a 3D structure not only provides efficient transport ways for electrons in the cross-plane direc­ tion of graphene layers but also generates much thinner papers during the reduction process to shorten the electron diffusion distance [39,40]. The potassium storage abilities of the as-prepared rGO/CNT hybrid papers were evaluated by our assembling a CR2032 coin cell with po­ tassium metal as the counter electrode. Fig. 4a presents the initial threecycle cyclic voltammograms of the rGO/CNT hybrid paper anode. Dur­ ing the cathodic scan, a broad peak at about 0.8 V appeared in the first cycle and disappeared in the following cycles for the hybrid papers. This indicates the formation of a solid-electrolyte interface, similar to a graphene anode for LIBs, which resulted from the electrolyte decom­ position and irreversible reaction of Kþ on account of the defects at basal planes of rGO/CNT hybrid paper. The disappearance of such a peak in the following cycles suggests the stability of the solid-electrolyte inter­ face in the Kþ electrolyte [44,55]. Moreover, a sharp peak close to 0 V was then observed with the scan, corresponding to the intercalation of potassium ions into the graphitic carbon structure of the rGO/CNT hybrid paper [56]. In the following anodic scan, a broad mountain-like peak at about 0.5 V appeared. This is attributed to the deintercalation of potassium ions from the carbon structure in rGO/CNT hybrid paper [4].

In addition, the peak currents of the rGO/CNT hybrid paper were much higher than those of pure rGO paper (Fig. S4) in both the anodic scan and the cathodic scan, further reflecting the enhanced potassium storage ability of the rGO/CNT hybrid paper. The cyclic voltammetry results demonstrate that the rGO/CNT hybrid paper can reversibly store po­ tassium ions like lithium ions with higher storage ability than pure rGO paper. The rate performance and corresponding charge/discharge profiles of the rGO/CNT hybrid paper anodes for KIBs at current densities from 10 to 100 mA g 1 are presented in Fig. 4b and c. The discharge and charge plateaus were consistent with the cyclic voltammograms, indi­ cating the reversible storage of potassium ions by intercalation and deintercalation in the rGO/CNT hybrid paper in electrochemical con­ ditions. In the first discharge at 10 mA g 1, the rGO/CNT-30% paper anode delivered a high specific capacity of 1209 mA h g 1. However, such high initial capacity is not reversible, resulting in low initial Coulombic efficiency, which is the major challenge for graphene-based materials as anodes for LIBs as well as KIBs found here [57]. In the second cycle at 10 mA g 1, a reversible discharge capacity of 335 mA h g 1 was obtained for the rGO/CNT-30% paper anode. In the following cycles, when the current density increased to 20, 30, 50, and 100 mA g 1, the discharge capacity of rGO/CNT-30% paper gradually decreased to 246, 201, 179, and 110 mA h g 1, respectively. When the current density returned to 50 and 20 mA g 1, the capacity recovered to 176 and 204 mA h g 1, respectively, suggesting high reversibility of the rGO/CNT hybrid paper for potassium storage. Pure rGO paper exhibited low discharge capacities of 105, 92, 76, and 48 mA h g 1 at current densities of 20, 30, 50, and 100 mA g 1, respectively. This directly im­ plies the enhanced potassium storage ability of rGO paper by CNT incorporation. It is worth noting that the weight ratio of CNT has a significant effect on the electrochemical performance of the hybrid paper anode for KIBs. As shown in Fig. 4c, rGO/CNT-10% paper showed 4

S. Peng et al.

Journal of Physics and Chemistry of Solids 138 (2020) 109296

Fig. 4. (a) Cyclic voltammograms and (b) charge/discharge profiles at different current densities of reduced graphene oxide (rGO)/carbon nanotube (CNT)-30% paper. (c) Rate capability and (d) cycling performance of rGO and rGO/CNT hybrid papers. (e) Long cycling performance of rGO/CNT-20% and rGO/CNT-30% hybrid papers. (f) Cycling performance of rGO/CNT-30% paper as an anode for lithium-ion batteries at a current density of 100 mA g 1 and sodium-ion batteries at a current density of 50 mA g 1.

5

S. Peng et al.

Journal of Physics and Chemistry of Solids 138 (2020) 109296

the rGO/CNT hybrid paper as an anode for KIBs. Since the potassium storage capacity contribution of CNTs (<10 mA h g 1) could be ignored, the capacity increase and performance enhancement essentially result from CNT incorporation into graphene layers and the formation of a sandwich-like 3D conductive carbon network of the hybrid papers. Be­ sides as an anode for KIBs, we also studied the performance of rGO/CNT-30% paper for LIBs and NIBs, as shown in Fig. 4f. The rGO/CNT-30% paper displayed excellent lithium storage ability, with a stable discharge capacity of about 260 mA h g 1 after 60 cycles at a current density of 100 mA g 1. The better performance in LIBs than in KIBs is mainly the result of the larger radius of potassium ions than lithium ions. However, only a low capacity of about 40 mA h g 1 was achieved at a current density of 50 mA g 1 when the rGO/CNT-30% paper was used as an anode for NIBs. This further confirms the different electrochemical behavior of intercalated sodium ions and potassium ions in the graphitic carbon structure. To further understand the excellent potassium storage ability of the hybrid papers, electrochemical impedance spectroscopy of the hybrid papers was conducted; the spectra are plotted in Fig. 5. The rGO/CNT30% paper exhibited the smallest semicircle in the high-frequency area among all samples, reflecting the lowest charge transfer resistance for the KIBs, which is consistent with the four-probe conductivity mea­ surement results. An increased slope in the low-frequency region was observed for the rGO/CNT hybrid papers compared with the pure rGO paper, suggesting that the potassium-ion diffusion resistance was decreased after CNT incorporation in the graphene layers to form the 3D interconnected structure. The papers after 200 cycles were also characterized by scanning electron microscopy and Raman spectroscopy to obtain further insight into the potassium-ion intercalation/deintercalation process. Fig. 6a–d presents the surface morphologies of the rGO and rGO/CNT hybrid pa­ pers with inserted digital photographs of rGO/CNT-10% and rGO/CNT30% papers. All the cycled papers maintained the initial appearance from the digital photographs, implying sufficient mechanical and chemical stability of the hybrid papers in KIBs. Compared with the fresh paper shown in Fig. 2, the surface of the cycled rGO papers became rougher and some pores were seen accordingly. This could be caused by the intercalation/deintercalation of potassium. Nevertheless, the cycled hybrid papers showed slight morphology changes (Fig. 6b–d). Some small marginal particles were exposed on the surface of the hybrid pa­ pers, which were possibly remains of potassium salt in the electrolyte. Significantly, the hybrid rGO/CNT papers kept their appearance well. This further suggests that such a 3D conductive carbon network could effectively avoid structure damage during electrochemical cycles [58, 59]. Raman spectra of the cycled papers are shown in Fig. 6e. The D and G bands were both well maintained, suggesting the stability of the

Fig. 5. Electrochemical impedance spectroscopy plots of the fresh reduced graphene oxide (rGO) and rGO/carbon nanotube (CNT) hybrid paper anodes.

a slight capacity increase compared with pure rGO paper, while when the CNT ratio was increased to 20%, the capacity increased substan­ tially. However, with the CNT ratio continuously increased to 30%, the capacities at various current densities were similar to those for rGO/CNT-20% paper. A possible reason is the presence of agglomera­ tions of CNTs, which would destroy the 3D conductive carbon network of rGO/CNT hybrid paper. The long cycling stability of the rGO/CNT hybrid paper anodes for KIBs was studied at a current density of 50 mA g 1. As shown in Fig. 4d, the pure rGO paper showed a stable capacity of about 75 mA h g 1 in the 200 cycles tested. Nevertheless, the potassium storage capacity was obviously increased after CNT incorporation. The rGO/CNT-30% paper delivered a higher initial reversible capacity of 223 mA h g 1 and maintained a stable capacity of 148 mA h g 1 after 200 cycles, which is twice than that of the pure rGO paper. Although the capacity of rGO/CNT-30% paper is similar to that of rGO/CNT-20% paper, rGO/CNT-30% paper has a superior long cycling performance compared with rGO/CNT-20% paper, as shown in Fig. 4e. A possible reason is that the performance of the hybrid papers becomes less dependent on the weight ratio of CNT beyond a certain value of this ratio. To further show the relationship between the adjusted paraments of free-standing papers and electrochemical performance, the number of samples and other adjusted parameters are listed in Table S2. The electrochemical test results demonstrate the excellent performance of

Fig. 6. Top-view surface scanning electron microscope images of the cycled (a) reduced graphene oxide (rGO), (b) rGO/carbon nanotube (CNT)-10%, (c) rGO/CNT20%, and (d) rGO/CNT-30% papers. The insets show digital photographs of the cycled paper. (e) Raman spectra of the cycled paper anodes. 6

Journal of Physics and Chemistry of Solids 138 (2020) 109296

S. Peng et al.

graphitic carbon structure during potassium intercalation and dein­ tercalation. However, the D band corresponding to the defect in the carbon structure was weakened after electrochemical cycles with the decrease of the ID/IG ratio to less than 0.8. This behavior is attributed to the defect spots being occupied irreversibly by potassium, which also led to the capacity decline during long-term cycling [60,61].

[11] R. Wang, S. Wang, Y. Zhang, D. Jin, X. Tao, L. Zhang, Graphene coupled Ti3C2 MXenes-derived TiO2 mesostructure: promising sodium-ion capacitor anode with fast ion storage and long-term cycling, J. Mater. Chem. 6 (2018) 1017–1027. [12] J.W. Choi, D. Aurbach, Promise and reality of post-lithium-ion batteries with high energy densities, Nat. Rev. Mater. 1 (2016) 16013. [13] H. Wang, C. Zhu, D. Chao, Q. Yan, H.J. Fan, Nonaqueous hybrid lithium-ion and sodium-ion capacitors, Adv. Mater. 29 (2017) 1702093. [14] K. Kubota, M. Dahbi, T. Hosaka, S. Kumakura, S. Komaba, Towards K-ion and Naion batteries as “beyond li-ion”, Chem. Rec. 18 (2018) 459–479. [15] F. Sagane, T. Abe, Y. Iriyama, Z. Ogumi, Liþ and Naþ transfer through interfaces between inorganic solid electrolytes and polymer or liquid electrolytes, J. Power Sources 146 (2005) 749–752. [16] X. Wu, Y.L. Chen, Z. Xing, C.W.K. Lam, S.S. Pang, W. Zhang, Z.C. Ju, Advanced carbon-based anodes for potassium-ion batteries, Adv. Energy Mater. (2019) 1900343. [17] N. Yabuuchi, K. Kubota, M. Dahbi, S. Komaba, Research development on sodiumion batteries, Chem. Rev. 114 (2014) 11636–11682. [18] L. Xue, H. Gao, W. Zhou, S. Xin, K. Park, Y. Li, J.B. Goodenough, Liquid K–Na alloy anode enables dendrite-free potassium batteries, Adv. Mater. 28 (2016) 9608–9612. [19] K. Share, A.P. Cohn, R. Carter, B. Rogers, C.L. Pint, Role of nitrogen-doped graphene for improved high-capacity potassium ion battery, Anodes. ACS Nano (2016) 9738–9744. [20] L. Xue, Y. Li, H. Gao, W. Zhou, X. Lu, W. Kaveevivitchai, A. Manthiram, J. B. Goodenough, A low-cost high-energy potassium cathode, J. Am. Chem. Soc. 139 (2017) 2164–2167. [21] Z. Jian, W. Luo, X. Ji, Carbon electrodes for k-ion batteries, J. Am. Chem. Soc. 137 (2015) 11566–11569. [22] Y. Wen, K. He, Y. Zhu, F. Han, Y. Xu, I. Matsuda, Y. Ishii, J. Cumings, C. Wang, Expanded graphite as superior anode for sodium-ion batteries, Nat. Commun. 5 (2015) 4033–4039. [23] A.P. Kauling, A.T. Seefeldt, D.P. Pisoni, R.C. Pradeep, R. Bentini, R.V.B. Oliveira, K. S. Novoselov, A.H. Castro Neto, The worldwide graphene flake production, Adv. Mater. 30 (2018) 1803784. [24] R. Rafee, A. Eskandariyun, Estimating Young’s modulus of graphene/polymer composites using stochastic multi-scale modeling, Composites Part B 173 (2019) 1359–8368. [25] R. Rafee, A. Eskandariyun, Comparative study on predicting Young’s modulus of graphene sheets using nano-scale continuum mechanics approach, Physica E 90 (2017) 42–48. [26] Y. Zhu, Y. Yin, X. Yang, T. Sun, S. Wang, Y. Jiang, J. Yan, X.B. Zhang, Transformation of rusty stainless-steel meshes into stable, low-cost, and binder-free cathodes for high-performance potassium-ion batteries, Angew. Chem. Int. Ed. 56 (2017) 7881–7885. [27] S. Wei, Y. Xie, Y. Xing, L. Wang, H. Ye, X. Xiong, S. Wang, K. Han, Two-dimensional graphene oxide/MXene composite lamellar membranes for efficient solvent permeation and molecular separation, J. Membr. Sci. 582 (2019) 414–422. [28] D. Yu, L. Dai, Self-assembled graphene/carbon nanotube hybrid films for supercapacitors, J. Phys. Chem. Lett. 1 (2010) 467–470. [29] J. Abraham, K.S. Vasu, C.D. Williams, K. Gopinadhan, Y. Su, C.T. Cherian, J. Dix, E. Prestat, S.J. Haigh, I.V. Grigorieva, Tunable sieving of ions using graphene oxide membranes, Nat. Nanotechnol. 12 (2017) 546. [30] L. Bao, T. Li, S. Chen, C. Peng, L. Li, Q. Xu, Y. Chen, E. Ou, W. Xu, 3D graphene frameworks/Co3O4 composites electrode for high-performance supercapacitor and enzymeless glucose detection, Small 13 (2017), 02077. [31] C. Wang, L. Dan, C.O. Too, G.G. Wallace, Electrochemical properties of graphene paper electrodes used in lithium batteries, Chem. Mater. 21 (2009) 2604–2606. [32] H. Gwon, H.S. Kim, K.U. Lee, D.H. Seo, C.P. Yun, Y.S. Lee, B.T. Ahn, K. Kang, Flexible energy storage devices based on graphene paper, Energy Environ. Sci. 4 (2011) 1277–1283. [33] A. Abouimrane, O.C. Compton, K. Amine, S.B.T. Nguyen, Non-annealed graphene paper as a binder-free anode for lithium-ion batteries, J. Phys. Chem. C 114 (2010) 12800–12804. [34] R. Rafee, A. Eskandariyun, Stochastic multi-scale modeling of randomly grown CNTs on carbon fiber, Mech. Mater. 106 (2017) 1–7. [35] R. Rafee, A. Eskandariyun, Predicting mechanical properties of fuzzy fiber reinforced composites: radially grown carbon nanotubes on the carbon fiber, Int. J. Mech. Mater. Des. 14 (2018) 37–50. [36] M. Jamal-Omidi, M. ShayanMehr, S. Shokrollahi, R. Rafiee, A study on nonlinear vibration behavior of CNT-based representative volume element, Aero. Sci. Technol. 55 (2016) 272–281. [37] R. Rafee, M. Mahdavi, Characterizing nanotube–polymer interaction using molecular dynamics simulation, Comput. Mater. Sci. 112 (2016) 356–363. [38] R. Rafee, V. Firouzbakht, Multi-scale modeling of carbon nanotube reinforced polymers using irregular tessellation technique, Mech. Mater. 78 (2014) 74–84. [39] Y. Hu, X. Li, J. Wang, R. Li, X. Sun, Free-standing graphene-carbon nanotube hybrid papers used as current collector and binder free anodes for lithium ion batteries, J. Power Sources 237 (2013) 41–46. [40] Y. Huang, J. Wu, K.C. Hwang, Thickness of graphene and single-wall carbon nanotubes, Phys. Rev. B 74 (2016) 4070–4079. [41] N. Li, Y. Xie, S. Peng, X. Xiong, K. Han, Ultra-lightweight Ti3C2Tx MXene modified separator for Li-S batteries: thickness regulation enabled polysulfide inhibition and lithium ion transportation, J. Energy Chem. 42 (2020) 116–125. [42] Y. Lu, Z. Chao, N. Hu, H. Jing, H. Min, L. Zhang, Flexible graphene/carbon nanotube hybrid papers chemical-reduction-tailored by gallic acid for high-

4. Conclusion In summary, we have successfully fabricated free-standing rGO/CNT hybrid papers with a sandwich-like 3D structure by a facile method combining vacuum filtration and thermal reduction. The incorporation of CNTs could not only prevent restacking and increase the interlayer spacing of graphene to provide more storage sites for potassium ions but could also bridge graphene layers to facilitate electron and ion transport in the cross-plane direction. The overall electrical conductivity of rGO paper was thus significantly increased. As a result, the potassium-ion storage ability of rGO paper was considerably enhanced by CNT incor­ poration and the electrochemical performance of the hybrid papers depended on the weight ratio of CNT. The optimal rGO/CNT-30% paper delivered an initial reversible discharge capacity of 351 and 223 mA h g 1 at a current density of 10 and 50 mA g 1, respectively. A capacity of 148 mA h g 1 was maintained after 200 cycles at a current density of 50 mA g 1. The hybrid papers can be expected to be a material used as a binder-free anode for KIBs. This work should increase the application of graphene-based materials beyond LIBs and suggests that rGO/CNT hybrid papers are potential binder-free anodes for KIBs. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (no. 21706292) and ther Hunan Provincial Natural Science Foundation of China (no. 2017JJ3376). K. Han thanks the Hunan Provincial Science and Technology Plan Project, China (no. 2016TP1007) for support. Materials characterization were supported by the Open Sharing Fund for the Large-scale Instruments and Equipments of Central South University. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.jpcs.2019.109296. References [1] M. Naguib, R.A. Adams, Y. Zhao, D. Zemlyanov, A. Varma, J. Nanda, V.G. Pol, Electrochemical performance of MXenes as K-ion battery anodes, Chem. Commun. 53 (2017) 6883–6886. [2] Z.W. Liu, P. Li, G.Q. Suo, S. Gong, W. Wang, C.Y. Lao, Y.J. Xie, H. Guo, Q.Y. Yu, W. Zhao, K. Han, Q. Wang, M.L. Qin, K. Xi, X.H. Qu, Zero-strain K0.6Mn1F2.7 hollow nanocubes for ultrastable potassium ion storage, Energy Environ. Sci. 11 (2018) 3033–3042. [3] Z. Cang, J. Xu, X. Zhang, Recent progress in electrocatalyst for Li-O2 batteries, Adv. Energy Mater. 6 (2017) 1700875. [4] Y. Dong, Z.-S. Wu, S. Zheng, X. Wang, J. Qin, S. Wang, X. Shi, X. Bao, Ti3C2 MXenederived sodium/potassium titanate nanoribbons for high-performance sodium/ potassium ion batteries with enhanced capacities, ACS Nano 11 (2017) 4792–4800. [5] Y. Mei, Y. Huang, X. Hu, Nanostructured Ti-based anode materials for Na-ion batteries, J. Mater. Chem. 4 (2016) 12001–12013. [6] A. Eftekhari, Z. Jian, X. Ji, Potassium secondary batteries, ACS Appl. Mater. Interfaces 9 (2016) 4404–4419. [7] J. Zhao, X. Zou, Y. Zhu, Y. Xu, C. Wang, Electrochemical intercalation of potassium into graphite, Adv. Funct. Mater. 26 (2016) 8103–8110. [8] W. Luo, J. Wan, B. Ozdemir, W. Bao, Y. Chen, J. Dai, H. Lin, Y. Xu, F. Gu, V. Barone, Potassium ion batteries with graphitic materials, Nano Lett. 15 (2015) 7671–7677. [9] C.L. Liu, S.H. Luo, H.B. Huang, Z.Y. Wang, A.M. Hao, Y.C. Zhai, Z.W. Wang, K0.67Ni0.17Co0.17Mn0.66O2: a cathode material for potassium-ion battery, Electrochem. Commun. 82 (2017) 150–154. [10] P. Lian, Y. Dong, Z.-S. Wu, S. Zheng, X. Wang, S. Wang, C. Sun, J. Qin, X. Shi, X. Bao, Alkalized Ti3C2 MXene nanoribbons with expanded interlayer spacing for high-capacity sodium and potassium ion batteries, Nano Energy 40 (2017) 1–8.

7

S. Peng et al.

[43] [44] [45]

[46] [47]

[48] [49] [50] [51]

Journal of Physics and Chemistry of Solids 138 (2020) 109296

performance electrochemical capacitive energy storages, Appl. Surf. Sci. 435 (2018) 699–707. C. Song, K. Yuan, T. Tian, B.Q. Wei, Flexible sub-micro carbon [email protected] as anodes for potassium-ion batteries, ACS Appl. Mater. Interfaces 11 (2019), 50515021. C. Shen, H. Long, G. Wang, W. Lu, L. Shao, K.Y. Xie, Na3V2(PO4)[email protected] dispersed within carbon nanotube frameworks as a high tap density cathode for highperformance sodium-ion batteries, J. Mater. Chem. 6 (2018) 6007–6014. F. Meng, Z. Chang, J. Xu, X. Zhang, J. Yan, Photoinduced decoration of NiO nanosheets/Ni Foam with Pd nanoparticles towards a carbon-free and self-standing cathode for a lithium–oxygen battery with a low overpotential and long cycle life, Mater. Horiz. 5 (2018) 298–302. Y. Zhu, S. Yuan, D. Bao, Y.B. Yin, H. Zhong, X. Zhang, J. Yan, Q. Jiang, Decorating waste cloth via industrial wastewater for tube type flexible and wearable sodiumion batteries, Adv. Mater. 29 (2017) 1603719. X. Yang, J. Xu, Z. Chang, D. Bao, Y. Yin, T. Liu, J. Yan, D. Liu, Y. Zhang, X. Zhang, Blood-capillary-inspired, free-standing, flexible, and low cost super-hydrophobic [email protected] cathodes for high capacity, high-rate, and stable li-air batteries, Adv. Energy Mater. 8 (2018) 170024. C. Wang, A. Wang, L. Ren, X. Guan, D. Wang, A. Dong, C. Zhang, G. Li, J. Luo, Controlling Li ion flux through materials innovation for dendrite-free lithium metal anodes, Adv. Funct. Mater. 29 (2019) 1905940. A. Wang, X. Zhang, Y. Yang, J. Huang, X. Liu, J. Luo, Horizontal centripetal plating in the patterned voids of Li/graphene composites for stable lithium-metal anodes, Chem 4 (2018) 2192–2200. A. Wang, X. Hu, H. Tang, C. Zhang, S. Liu, Y. Yang, Q. Yang, J. Luo, Processable and moldable sodium-metal anodes, Angew. Chem. Int. Ed. 56 (2017) 11921–11926. W.S. Hummers Jr., R.E. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc. 80 (1958) 1339.

[52] K. Han, Z. Liu, J. Shen, Y. Li, F. Dai, H. Ye, A free-standing and ultralong-life lithium-selenium battery cathode enabled by 3D mesoporous carbon/graphene hierarchical architecture, Adv. Funct. Mater. 25 (2015) 455–463. [53] K. Han, J. Shen, S. Hao, H. Ye, C. Wolverton, M.C. Kung, H.H. Kung, Free-standing nitrogen-doped graphene paper as electrodes for high-performance lithium/ dissolved polysulfide batteries, ChemSusChem 7 (2015) 2545–2553. [54] F. Wu, M.H. Zhang, Y. Bai, C. Wu, Lotus Seedpod-derived hard carbon with hierarchical porous structure as stable anode for sodium-ion batteries, ACS Appl. Mater. Interfaces 11 (2019) 12554–12561. [55] D.G. Yin, Z. Chen, M. Zhang, Sn-interspersed MoS2/C nanosheets with high capacity for Naþ/Kþ storage, J. Phys. Chem. Solids 126 (2019) 72–77. [56] J. Wang, L. Fang, Z.M. Liu, S.H. Chen, Q.F. Zhang, L.L. Wang, H.G. Yang, X.Z. Yu, B.G. Lu, In situ alloying strategy for exceptional potassium ion batteries, ACS Nano 13 (2019) 3703–3713. [57] L.L. Wang, Q.Y. Zhang, J.Y. Zhu, H.G. Yang, B.G. Lu, Nature of extra capacity in MoS2 electrodes: molybdenum atoms accommodate with lithium, Energy Storage Mater. 16 (2019) 37–45. [58] Z. Huang, Z. Chen, S.S. Ding, C.M. Chen, M. Zhang, Multi-protection from nanochannels and graphene of SnSb-graphene-carbon composites ensuring high properties for potassium-ion batteries, Solid State Ion. 324 (2018) 267–275. [59] Y.S. Liao, C.M. Chen, D.G. Yin, M. Zhang, Improved Naþ/Kþ storage properties of ReSe2–carbon nanofibers based on graphene modifications, Nano-Micro Lett. (2019) 11–22. [60] J.M. Ge, L. Fang, J. Wang, X.Z. Yu, B.G. Lu, MoSe2/N-doped carbon as anodes for potassium-ion batteries, Adv. Energy Mater. 8 (2018) 1801477. [61] J.L. Yang, Z.C. Ju, Y. Jiang, Z. Xing, B.J. Xi, J.K. Feng, S.L. Xiong, Enhanced capacity and rate capability of nitrogen/oxygen dual-doped hard carbon in capacitive potassium-ion storage, Adv. Mater. 30 (2017) 1700104.

8