- Email: [email protected]
carbon nanotube composites in alkaline electrolyte
Journal of Power Sources 284 (2015) 38e43
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Enhanced supercapacitive performance of delaminated two-dimensional titanium carbide/carbon nanotube composites in alkaline electrolyte Pengtao Yan a, Ruijun Zhang a, Jin Jia b, Chao Wu a, c, Aiguo Zhou b, Jiang Xu a, *, Xuesha Zhang a a b c
State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao 066004, China School of Materials Science and Engineering, Henan Polytechnic University, Jiaozuo 454000, China Qinggong College, Hebei United University, Tangshan 063009, China
h i g h l i g h t s The introduction of carbon nanotube can impede the stacking of MXene sheets. Introducing carbon nanotube can improve the electrical conductivity of MXenes. The d-Ti3C2/CNT composites exhibit excellent supercapacitive performance.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 January 2015 Received in revised form 2 March 2015 Accepted 4 March 2015 Available online 5 March 2015
MXenes, a new family of two-dimensional materials, are terminated by O, OH and F groups. The existence of the oxygen-containing functional groups indicates a potential application in supercapacitor based on a redox mechanism. However, the irreversible stacking of MXenes will lead to an insufficient utilization of these functional groups and thus a decrease in the supercapacitive performance. To solve the problem, we synthesized a composite material comprised of carbon nanotube (CNT) and Ti3C2 sheets (d-Ti3C2) delaminated from MXenes by ultrasonic stirring. The FTIR result suggests that the ultrasonication has no significant effect on the oxygen-containing functional groups. The resultant composites exhibit significantly higher volumetric capacitance and better capacitance retention (during 5 e100 mv s1) than d-Ti3C2. A highest volumetric capacitance of 393 F cm3 at 5 mv s1 in KOH electrolyte can be obtained when the weight ratio of d-Ti3C2 to CNT is 2:1. In addition, the volumetric capacitance has no significant degradation even after 10000 cycles in cycling stability test, showing an excellent cycling stability compared with metal oxides. These enhanced electrochemical performances can be ascribed to the introduction of CNTs, which impede the stacking of Ti3C2, enlarge the distance between Ti3C2 sheets and improve the electrical conductivity. © 2015 Elsevier B.V. All rights reserved.
Keywords: Two-dimensional material MXene Composite Supercapacitive performance
1. Introduction Supercapacitors, a new energy storage device, bridge the gap between batteries and conventional capacitors and have been attracting considerable interest in the energy storage field [1e4]. Suitable electrode materials are key ingredients enabling the search
* Corresponding author. E-mail addresses: [email protected] (R. Zhang), [email protected] (J. Xu). http://dx.doi.org/10.1016/j.jpowsour.2015.03.017 0378-7753/© 2015 Elsevier B.V. All rights reserved.
for high performance supercapacitors. Up to now, many materials used as electrodes have been investigated, such as activated carbon [5,6], carbide-derived carbon [7e9], metal oxides [10,11], graphene/ metal oxide composite materials [12], conducting polymers [13] and so on. Recently, a new family of two-dimensional (2D) materials (i.e., MXenes) have attracted increasing attention, and shown a potential application in the fields of hydrogen storage [14], lead adsorption [15] and energy storage [16e18]. The MXenes are synthesized by selective removal the A-element from MAX phases (such as Ti3AlC2 [19]), where M is an early transition metal, A is a
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43
III or IV A-group element and X is carbon and/or nitrogen [20]. It is worth noting that the termination of the 2D MXene surfaces mainly contain a lot of oxygen-containing functional groups (such as eOH/¼O) and some fluorine (eF) [18,21], which are introduced after etching with aqueous hydrofluoric acid (HF). The existence of the oxygen-containing functional groups shows a promising application in the supercapacitor based on the redox mechanism. However, the MXene sheets are easy to form irreversible agglomerations due to the strong van der Waals interactions between individual MXene sheets, leading to the insufficient utilization of the MXene surface and oxygen-containing functional groups, and, as a result, a poorer electrochemical performance of pseudocapacitors. One-dimension carbon nanotubes (CNTs) have attracted much attention in energy storage applications. Most researches have proved that the introduction of the CNTs in the graphene structures can effectively impede the aggregations of individual graphene sheets and exhibit enhanced supercapacitive performance [22e24]. Therefore, in this work, through introducing the CNT into a Ti3C2-MXene, which is graphene-like structure [21], we synthesized a delaminated 2D Ti3C2/CNT composite material and investigated the electrochemical performance for the resultant composite. 2. Experimental 2.1. Sample preparation A MXene (2D Ti3C2) used in the present work was produced by immersing Ti3AlC2 in HF [18]. In order to obtain the delaminated Ti3C2 sheets [20], 0.3 g of 2D Ti3C2 were mixed with 6 ml dimethyl sulfoxide (DMSO, Shanghai Lingfeng Chemical Reagent Co. Ltd., China) and then magnetically stirred for 18 h at room temperature. The resultant suspension was centrifuged to separate the powder from the liquid DMSO. The obtained powder was mixed with deionized water and the suspension was under intermittent ultrasonication for designated time (e.g. 2 h and 6 h). After that, the larger particles were removed by centrifuging and the supernatant were collected as the samples for further investigation. The samples obtained after ultrasonication for 2 h and 6 h were denoted as ex-Ti3C2 and d-Ti3C2, respectively. The CNTs with 20e40 nm in diameter were obtained from the Shenzhen Nanotech Co. Ltd. of China. The stable suspension of CNTs (0.3 mg ml1) was produced by ultrasonication in the presence of dispersant. Thereafter, the dTi3C2 (0.3 mg ml1) and CNTs were thoroughly mixed (denoted as d-Ti3C2/CNT) at different weight ratios of d-Ti3C2 to CNT (6:1, 2:1 and 1:1) by ultrasonic stirring. The mixed suspension was filtered using a polytetraflouroethylene (PTFE) filter (0.2 mm in pore size) and dried at 70 C, achieving the d-Ti3C2/CNT composite materials. Besides, the weight ratio of d-Ti3C2/CNT is 2:1 if no particular explanation. 2.2. Characterization The morphology and microstructure of the samples were observed by scanning electron microscopy (SEM, Hitachi S4800, Japan) and transmission electron microscopy (TEM, JEOL JEM-2010, Japan). X-ray diffraction (XRD) patterns between 5 (2q) and 65 (2q) degrees were collected by Rigaku D/MAX-2500 powder diffractometer with Cu-Ka radiation (l ¼ 0.154 nm) operated at 40 kV and 200 mA. Existence of oxygen functional groups was examined by Fourier transform infrared spectroscopy (FTIR, Bruker Optiks E55 þ FRA106, Germany). Electrical conductivity were measured by ST-2722 semiconductor resistivity of the powder tester (Changzhou Sanfeng instrument technology Co. Ltd., China).
39
2.3. Preparation of electrode and electrochemical measurement The electrochemical investigations were carried out in a threeelectrode system using platinum as counter electrode and Hg/ HgO as a reference electrode. The working electrodes were prepared as follows: A slurry consisting of 80 wt.% active materials, 10 wt.% carbon black and 10 wt.% PTFE (60 wt.% suspension in water) binder was smeared into nickel foam and dried in vacuum at 120 C for 10 h. Thereafter, the electrode was pressed at a pressure of 10 MPa. 6 M KOH solution was used as electrolyte. The active area of the electrode is 1 cm2. Before each measurement, the working electrode was impregnated with electrolyte to ensure thoroughly wetted by electrolyte. Cyclic voltammograms (CV) were collected on CHI650D electrochemical workstation in a potential range of 0.1e0.55 V. The scan rates for all CV tests were in the range of 5e100 mV s1. 3. Results and discussion The XRD patterns of 2D Ti3C2, ex-Ti3C2, d-Ti3C2, CNTs and dTi3C2/CNT composite are presented in Fig. 1. Fig. 1a shows the structural evolution from Ti3AlC2 to d-Ti3C2, demonstrating clearly a continuous decrease in crystallinity and the structural order when the pristine Ti3AlC2 is transformed to 2D Ti3C2, then to ex-Ti3C2, and finally to d-Ti3C2. The (002) peak of Ti3AlC2, which is initially at 2q ¼ 9 , broadens and shifts to 2q ¼ 8 (for 2D Ti3C2) after etching with HF. When the ultrasonication on 2D Ti3C2 is carried out for 2 h, the (002) peak further shift to lower angle (2q ¼ 6 for ex-Ti3C2) and the lattice parameter c increases from 21.4 Å (2D Ti3C2) to 31.7 Å (ex-Ti3C2). When the ultrasonication is performed for 6 h, the (002) peak disappears, suggesting the full delamination of 2D Ti3C2 and thus the formation of individual Ti3C2 sheets. Furthermore, with the increasing of ultrasonic time, the gradually disappeared (110) peak at 2q ¼ 61 also provides obvious evidence for the decrease of order and full delamination of 2D Ti3C2 [20]. By comparison to the XRD patterns of d-Ti3C2, CNTs and d-Ti3C2/CNT shown in Fig. 1b, it can be deduced that the d-Ti3C2/CNT composite material, which is comprised of Ti3C2 sheets and CNT, has been successfully synthesized. Fig. 2 exhibits SEM images of these Ti3C2 samples. It is clear that the 2D Ti3C2 behaves a layered structure. Under ultrasonication, the layer number decreases and the surface becomes rougher (Fig. 2b for ex-Ti3C2), and then, the further ultrasonication will result in the formation and the pulverization of individual Ti3C2 sheets (Fig. 2c for d-Ti3C2). However, the pulverized individual Ti3C2 sheets are very easy to agglomerate. The introduction of the CNTs into the dTi3C2 can effectively impede the aggregation of individual Ti3C2 sheets (Fig. 2d). TEM observations further demonstrate (shown in Fig. 3) that the ultrasonication of 2D Ti3C2 can obtain delaminated Ti3C2 sheets and the CNTs and d-Ti3C2 are interweaved together in the d-Ti3C2/CNT composite. The FTIR spectra of CNTs, d-Ti3C2 and d-Ti3C2/CNT are given in Fig. 4. It can be seen that the C]O (~1645 cm1) functional groups exist in all the samples. In addition, the CeO (~1051 cm1), CeF (~1090 cm1) and OeH (~1389 cm1) groups can be observed in the spectra of d-Ti3C2 and d-Ti3C2/CNT. Obviously, almost all the oxygen-containing functional groups in the MXenes can still be retained in the synthesized d-Ti3C2/CNT composite, suggesting that the ultrasonication has no significant effect on the oxygencontaining functional groups. This result also implies that the synthesized d-Ti3C2/CNT composite will exhibit pseudocapacitive performance due to the existence of the oxygen-containing functional groups. The CV curves of 10 mv s1 for d-Ti3C2 and d-Ti3C2/CNT composites at different weight ratios of d-Ti3C2 to CNT (6:1, 2:1 and 1:1)
40
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43
Fig. 1. The XRD patterns of Ti3AlC2, 2D Ti3C2, ex-Ti3C2, d-Ti3C2, CNTs and d-Ti3C2/CNT.
Fig. 2. SEM images of 2D Ti3C2 (a), ex-Ti3C2 (b), d-Ti3C2 (c) and d-Ti3C2/CNT (d).
are shown in Fig. 5a. The curve shapes of these samples are dramatically different from that of porous carbon, which generally show a rectangular-like shape indicating electric double-layer capacitance. The strong peaks can be observed in Fig. 5a, which suggests that the capacitance of these samples mainly results from pseudocapacitive capacitance based on a redox mechanism. This should be ascribed to the oxygen-containing functional groups terminated on the 2D MXene surfaces (Fig. 4). There occurs apparent oxidation peaks (at roughly 0.44 V and 0.49 V) and reduction peak (at roughly 0.32 V) in the CV curves of all the dTi3C2/CNT composites. However, the oxidation peaks (at roughly 0.46 V and 0.49 V) in the CV curve of d-Ti3C2 are significantly weak
and unconspicuous (inset Fig. 5a), indicating the insufficient redox process of oxygen-containing functional groups for the d-Ti3C2 sample compared with those for the d-Ti3C2/CNT composites. This is to say, the introduction of CNTs can greatly facilitate the redox process of oxygen-containing functional groups on the d-Ti3C2. This improved redox process should be related to the effective impedance on the aggregation of individual Ti3C2 sheets due to the introduction of CNTs. In addition, the improved conductivity of the d-Ti3C2/CNT composites (Table 1) also contributes to the enhanced electrochemical performance of these composites. The influence of the CNT content on the volumetric capacitance of d-Ti3C2/CNT composites at 10 mv s1 is also investigated, as
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43
41
Fig. 3. The TEM images of d-Ti3C2 (a) and d-Ti3C2/CNT (b).
shown in Fig. 5b. The volumetric capacitance of d-Ti3C2/CNT composite gradually increases with the increasing CNT content, reaching the maximal value of 384 F cm3 at the 2:1 weight ratio of dTi3C2 to CNT, and thereafter, the volumetric capacitance begins to decrease as the CNT content further increases. In any case, however, the introduction of CNTs at different weight ratios certainly leads to significantly higher volumetric capacitance compared with that of the pure d-Ti3C2. This means that the introduced CNTs, which are interweaved together with d-Ti3C2, impede the stacking of individual Ti3C2 sheets effectively and enlarge the distance between Ti3C2 sheets, thus facilitating the utilization of the surface of samples, the diffusion of electrolyte ion and redox process. The CV curves of d-Ti3C2 and d-Ti3C2/CNT at different scan rates are shown in Fig. 6. Due to the introduction of CNTs and the improvement of conductivity, the d-Ti3C2/CNT shows better electrochemical performance and redox process than d-Ti3C2. Fig. 7 present the relation curves of volumetric capacitance and CV scan rates of d-Ti3C2 and d-Ti3C2/CNT. It can be observed that the volumetric capacitances of these two samples decrease gradually with the increasing of scan rates in a range of 5e100 mv s1. At a same scan rate, however, the d-Ti3C2/CNT exhibits obviously higher volumetric capacitance than the d-Ti3C2. For instance, at a scan rate of 5 mv s1, the volumetric capacitance of d-Ti3C2/CNT is up to 393 F cm3, almost 1.8 times as great as that of the pure d-Ti3C2. As the scan rate increases to 100 mv s1, the capacitance retentions of d-Ti3C2 and d-Ti3C2/CNT are about 52% and 80%, respectively. This
Fig. 4. FTIR spectra of CNTs, d-Ti3C2 and d-Ti3C2/CNT.
capacitance retention of d-Ti3C2/CNT is also greatly higher than that of the d-Ti3C2 reported before [25], implying that the introduction of CNTs can significantly improve the d-Ti3C2/CNT rate performance. The cycling stability is a most important criterion for the supercapacitor, and the long-term cycling stability of d-Ti3C2 and d-
Fig. 5. (a) The CV curves of 10 mv s1 for d-Ti3C2 and d-Ti3C2/CNT at different weight ratios. (b) The volumetric capacitance of d-Ti3C2/CNT as a function of the CNT content at 10 mv s1 in 6 M KOH electrolyte.
42
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43
Table 1 Electrical conductivity of d-Ti3C2 and d-Ti3C2/CNT at different weight ratios. Samples
d-Ti3C2
d-Ti3C2/CNT (6:1)
d-Ti3C2/CNT (2:1)
d-Ti3C2/CNT (1:1)
Conductivity (s m1)
20
113
230
332
Fig. 7. A plot of volumetric capacitance versus the scan rates of d-Ti3C2 and d-Ti3C2/ CNT.
Fig. 8. Cycling stability of d-Ti3C2 and d-Ti3C2/CNT at 10 mv s1.
Fig. 6. The CV curves of d-Ti3C2 (a) and d-Ti3C2/CNT (b) at different scan rates.
Ti3C2/CNT is investigated at a scan rate of 10 mv s1. Fig. 8 shows the volumetric capacitance of these two samples as a function of cycle number. It is worth noting that, when the cycling begins, the volumetric capacitance of both the samples increases, which is quite different from those for other redox materials in most cycling stability tests. The reason may be attributed to the sufficient activation of electrode materials at the early circulation of redox process [26]. Furthermore, the cyclic capacitance of d-Ti3C2/CNT (~384 F cm3) is far larger than that of d-Ti3C2 (~193 F cm3). More interestingly, the d-Ti3C2 and d-Ti3C2/CNT composite exhibit satisfactory cycling stabilities over the entire cycle numbers and almost 100% specific capacitance is retained even after 10000 cycles for these two samples. This excellent cycling stability is greatly better compared with those of the other typical electrode materials of pseudocapacitors, such as various types of metal oxides (MnO2 [11], NiO [12], RuO2 [27], SnO2 [28], and so on). 4. Conclusion In this article, the delaminated Ti3C2 sheets are prepared from MXenes by intercalation and ultrasonication, and the CNT is used to
inhibit the aggregations of delaminated Ti3C2 sheets. At the optimal d-Ti3C2 and CNT ratio of 2:1 in weight, the d-Ti3C2/CNT composite displays high volumetric capacitance (393 F cm3 at 5 mv s1), superior rate capability (80% capacitance retention as the scan rate increases from 5 to 100 mv s1) and excellent cycling stability over metal oxides. The introduction of CNTs can significantly enhance the supercapacitive performance of delaminated Ti3C2 by impeding the stacking of Ti3C2, enlarging the distance between Ti3C2 sheets and improving the electrical conductivity. Therefore, the d-Ti3C2/ CNT composite materials are quite promising pseudocapacitor materials with superior eletrochemical properties. Acknowledgment Financial support of this work by National Science Foundation of China (NSFC) (No. 50975247) and Hebei Natural Science Foundation (No. E2014203204) is acknowledged. References [1] D. Pech, M. Brunet, H. Durou, P. Huang, V. Mochalin, Y. Gogotsi, P.L. Taberna, P. Simon, Nat. Nanotech. 5 (2010) 651e654. [2] P. Simon, Y. Gogotsi, Nat. Mater. 7 (2008) 845e854. [3] B.E. Conway, J. Electrochem. Soc. 138 (1991) 1539e1548. [4] J.R. Miller, A.F. Burke, Electrochem. Soc. Interf. 17 (2008) 53e57. [5] D. Hulicova-Jurcakova, M. Seredych, G.Q. Lu, N.K.A.C. Kodiweera, P.E. Stallworth, S. Greenbaum, T.J. Bandosz, Carbon 47 (2009) 1576e1584. [6] Y. Zhao, W. Ran, J. He, Y. Song, C. Zhang, D. Xiong, F. Gao, J. Wu, Y. Xia, ACS Appl. Mater. Interfaces 7 (2015) 1132e1139. [7] J. Xu, R. Zhang, P. Chen, S. Ge, J. Power Sources 246 (2014) 132e140.
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43 [8] P. Yan, J. Xu, C. Wu, R. Zhang, J. Jin, Mater. Lett. 139 (2015) 340e343. [9] Y. Zhao, W. Wang, D. Xiong, G. Shao, W. Xia, S. Yu, F. Gao, Int. J. Hydrog. Energy 37 (2012) 19395e19400. [10] A.M. Hashem, H.M. Abuzeid, A.E. Abdel-Ghany, A. Mauger, K. Zaghib, C.M. Julien, J. Power Sources 202 (2012) 291e298. [11] X. Li, A. Dhanabalan, C. Wang, J. Power Sources 196 (2011) 9625e9630. [12] Y. Zhao, W. Ran, J. He, Y. Huang, Z. Liu, W. Liu, Y. Tang, L. Zhang, D. Gao, F. Gao, Small (2014), http://dx.doi.org/10.1002/smll.201401922. [13] A. Rudge, J. Davey, I. Raistrick, S. Gottesfeld, J. Power Sources 47 (1994) 89e107. [14] Q. Hu, D. Sun, Q. Wu, H. Wang, L. Wang, B. Liu, A. Zhou, J. He, J. Phys. Chem. A 117 (2013) 14253e14260. [15] Q. Peng, J. Guo, Q. Zhang, J. Xiang, B. Liu, A. Zhou, R. Liu, Y. Tian, J. Am. Chem. Soc. 136 (2014) 4113e4116. [16] M.R. Lukatskaya, O. Mashtalir, C.E. Ren, Y. Dall’Agnese, P. Rozier, P.L. Tabetrna, M. Naguib, P. Simon, M.W. Barsoum, Y. Gogotsi, Science 341 (2013) 1502e1505. [17] Q. Tang, Z. Zhou, P. Shen, J. Am. Chem. Soc. 134 (2012) 16909e16916. [18] D. Sun, M. Wang, Z. Li, G. Fan, L. Fan, A. Zhou, Electrochem. Commun. 47
43
(2014) 80e83. [19] M.W. Barsoum, Prog. Solid State Chem. 28 (2000) 201e281. [20] O. Mashtalir, M. Naguib, V.N. Mochalin, Y. Dall’Agnese, M. Heon, M.W. Barsoum, Y. Gogotsi, Nat. Commun. 4 (2013) 1716. [21] M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi, M.W. Barsoum, Adv. Mater. 23 (2011) 4248e4253. [22] M. Sun, G. Wang, X. Li, C. Li, J. Power Sources 245 (2014) 436e444. [23] F. Zeng, Y. Kuang, N. Zhang, Z. Huang, Y. Pan, Z. Hou, H. Zhou, C. Yan, O.G. Schmidt, J. Power Sources 247 (2014) 396e401. [24] M.Y. Yen, M.C. Hsiao, S.H. Liao, P.I. Liu, H.M. Tsai, C.C.M. Ma, N.W. Pu, M.D. Ger, Carbon 49 (2011) 3597e3606. [25] Y. Dall’Agnese, M.R. Lukatskaya, K.M. Cook, P.L. Taberna, Y. Gogotsi, P. Simon, Electrochem. Commun. 48 (2014) 118e122. [26] T.Y. Wei, C.H. Chen, H.C. Chien, S.Y. Lu, C.C. Hu, Adv. Mater. 22 (2010) 347e351. [27] T.F. Hsieh, C.C. Chuang, W.J. Chen, J.H. Huang, W.T. Chen, C.M. Shu, Carbon 50 (2012) 1740e1747. [28] S.N. Pusawale, P.R. Deshmukh, C.D. Lokhande, Appl. Surf. Sci. 257 (2011) 9498e9502.
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Enhanced supercapacitive performance of delaminated two-dimensional titanium carbide/carbon nanotube composites in alkaline electrolyte Pengtao Yan a, Ruijun Zhang a, Jin Jia b, Chao Wu a, c, Aiguo Zhou b, Jiang Xu a, *, Xuesha Zhang a a b c
State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao 066004, China School of Materials Science and Engineering, Henan Polytechnic University, Jiaozuo 454000, China Qinggong College, Hebei United University, Tangshan 063009, China
h i g h l i g h t s The introduction of carbon nanotube can impede the stacking of MXene sheets. Introducing carbon nanotube can improve the electrical conductivity of MXenes. The d-Ti3C2/CNT composites exhibit excellent supercapacitive performance.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 January 2015 Received in revised form 2 March 2015 Accepted 4 March 2015 Available online 5 March 2015
MXenes, a new family of two-dimensional materials, are terminated by O, OH and F groups. The existence of the oxygen-containing functional groups indicates a potential application in supercapacitor based on a redox mechanism. However, the irreversible stacking of MXenes will lead to an insufficient utilization of these functional groups and thus a decrease in the supercapacitive performance. To solve the problem, we synthesized a composite material comprised of carbon nanotube (CNT) and Ti3C2 sheets (d-Ti3C2) delaminated from MXenes by ultrasonic stirring. The FTIR result suggests that the ultrasonication has no significant effect on the oxygen-containing functional groups. The resultant composites exhibit significantly higher volumetric capacitance and better capacitance retention (during 5 e100 mv s1) than d-Ti3C2. A highest volumetric capacitance of 393 F cm3 at 5 mv s1 in KOH electrolyte can be obtained when the weight ratio of d-Ti3C2 to CNT is 2:1. In addition, the volumetric capacitance has no significant degradation even after 10000 cycles in cycling stability test, showing an excellent cycling stability compared with metal oxides. These enhanced electrochemical performances can be ascribed to the introduction of CNTs, which impede the stacking of Ti3C2, enlarge the distance between Ti3C2 sheets and improve the electrical conductivity. © 2015 Elsevier B.V. All rights reserved.
Keywords: Two-dimensional material MXene Composite Supercapacitive performance
1. Introduction Supercapacitors, a new energy storage device, bridge the gap between batteries and conventional capacitors and have been attracting considerable interest in the energy storage field [1e4]. Suitable electrode materials are key ingredients enabling the search
* Corresponding author. E-mail addresses: [email protected] (R. Zhang), [email protected] (J. Xu). http://dx.doi.org/10.1016/j.jpowsour.2015.03.017 0378-7753/© 2015 Elsevier B.V. All rights reserved.
for high performance supercapacitors. Up to now, many materials used as electrodes have been investigated, such as activated carbon [5,6], carbide-derived carbon [7e9], metal oxides [10,11], graphene/ metal oxide composite materials [12], conducting polymers [13] and so on. Recently, a new family of two-dimensional (2D) materials (i.e., MXenes) have attracted increasing attention, and shown a potential application in the fields of hydrogen storage [14], lead adsorption [15] and energy storage [16e18]. The MXenes are synthesized by selective removal the A-element from MAX phases (such as Ti3AlC2 [19]), where M is an early transition metal, A is a
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43
III or IV A-group element and X is carbon and/or nitrogen [20]. It is worth noting that the termination of the 2D MXene surfaces mainly contain a lot of oxygen-containing functional groups (such as eOH/¼O) and some fluorine (eF) [18,21], which are introduced after etching with aqueous hydrofluoric acid (HF). The existence of the oxygen-containing functional groups shows a promising application in the supercapacitor based on the redox mechanism. However, the MXene sheets are easy to form irreversible agglomerations due to the strong van der Waals interactions between individual MXene sheets, leading to the insufficient utilization of the MXene surface and oxygen-containing functional groups, and, as a result, a poorer electrochemical performance of pseudocapacitors. One-dimension carbon nanotubes (CNTs) have attracted much attention in energy storage applications. Most researches have proved that the introduction of the CNTs in the graphene structures can effectively impede the aggregations of individual graphene sheets and exhibit enhanced supercapacitive performance [22e24]. Therefore, in this work, through introducing the CNT into a Ti3C2-MXene, which is graphene-like structure [21], we synthesized a delaminated 2D Ti3C2/CNT composite material and investigated the electrochemical performance for the resultant composite. 2. Experimental 2.1. Sample preparation A MXene (2D Ti3C2) used in the present work was produced by immersing Ti3AlC2 in HF [18]. In order to obtain the delaminated Ti3C2 sheets [20], 0.3 g of 2D Ti3C2 were mixed with 6 ml dimethyl sulfoxide (DMSO, Shanghai Lingfeng Chemical Reagent Co. Ltd., China) and then magnetically stirred for 18 h at room temperature. The resultant suspension was centrifuged to separate the powder from the liquid DMSO. The obtained powder was mixed with deionized water and the suspension was under intermittent ultrasonication for designated time (e.g. 2 h and 6 h). After that, the larger particles were removed by centrifuging and the supernatant were collected as the samples for further investigation. The samples obtained after ultrasonication for 2 h and 6 h were denoted as ex-Ti3C2 and d-Ti3C2, respectively. The CNTs with 20e40 nm in diameter were obtained from the Shenzhen Nanotech Co. Ltd. of China. The stable suspension of CNTs (0.3 mg ml1) was produced by ultrasonication in the presence of dispersant. Thereafter, the dTi3C2 (0.3 mg ml1) and CNTs were thoroughly mixed (denoted as d-Ti3C2/CNT) at different weight ratios of d-Ti3C2 to CNT (6:1, 2:1 and 1:1) by ultrasonic stirring. The mixed suspension was filtered using a polytetraflouroethylene (PTFE) filter (0.2 mm in pore size) and dried at 70 C, achieving the d-Ti3C2/CNT composite materials. Besides, the weight ratio of d-Ti3C2/CNT is 2:1 if no particular explanation. 2.2. Characterization The morphology and microstructure of the samples were observed by scanning electron microscopy (SEM, Hitachi S4800, Japan) and transmission electron microscopy (TEM, JEOL JEM-2010, Japan). X-ray diffraction (XRD) patterns between 5 (2q) and 65 (2q) degrees were collected by Rigaku D/MAX-2500 powder diffractometer with Cu-Ka radiation (l ¼ 0.154 nm) operated at 40 kV and 200 mA. Existence of oxygen functional groups was examined by Fourier transform infrared spectroscopy (FTIR, Bruker Optiks E55 þ FRA106, Germany). Electrical conductivity were measured by ST-2722 semiconductor resistivity of the powder tester (Changzhou Sanfeng instrument technology Co. Ltd., China).
39
2.3. Preparation of electrode and electrochemical measurement The electrochemical investigations were carried out in a threeelectrode system using platinum as counter electrode and Hg/ HgO as a reference electrode. The working electrodes were prepared as follows: A slurry consisting of 80 wt.% active materials, 10 wt.% carbon black and 10 wt.% PTFE (60 wt.% suspension in water) binder was smeared into nickel foam and dried in vacuum at 120 C for 10 h. Thereafter, the electrode was pressed at a pressure of 10 MPa. 6 M KOH solution was used as electrolyte. The active area of the electrode is 1 cm2. Before each measurement, the working electrode was impregnated with electrolyte to ensure thoroughly wetted by electrolyte. Cyclic voltammograms (CV) were collected on CHI650D electrochemical workstation in a potential range of 0.1e0.55 V. The scan rates for all CV tests were in the range of 5e100 mV s1. 3. Results and discussion The XRD patterns of 2D Ti3C2, ex-Ti3C2, d-Ti3C2, CNTs and dTi3C2/CNT composite are presented in Fig. 1. Fig. 1a shows the structural evolution from Ti3AlC2 to d-Ti3C2, demonstrating clearly a continuous decrease in crystallinity and the structural order when the pristine Ti3AlC2 is transformed to 2D Ti3C2, then to ex-Ti3C2, and finally to d-Ti3C2. The (002) peak of Ti3AlC2, which is initially at 2q ¼ 9 , broadens and shifts to 2q ¼ 8 (for 2D Ti3C2) after etching with HF. When the ultrasonication on 2D Ti3C2 is carried out for 2 h, the (002) peak further shift to lower angle (2q ¼ 6 for ex-Ti3C2) and the lattice parameter c increases from 21.4 Å (2D Ti3C2) to 31.7 Å (ex-Ti3C2). When the ultrasonication is performed for 6 h, the (002) peak disappears, suggesting the full delamination of 2D Ti3C2 and thus the formation of individual Ti3C2 sheets. Furthermore, with the increasing of ultrasonic time, the gradually disappeared (110) peak at 2q ¼ 61 also provides obvious evidence for the decrease of order and full delamination of 2D Ti3C2 [20]. By comparison to the XRD patterns of d-Ti3C2, CNTs and d-Ti3C2/CNT shown in Fig. 1b, it can be deduced that the d-Ti3C2/CNT composite material, which is comprised of Ti3C2 sheets and CNT, has been successfully synthesized. Fig. 2 exhibits SEM images of these Ti3C2 samples. It is clear that the 2D Ti3C2 behaves a layered structure. Under ultrasonication, the layer number decreases and the surface becomes rougher (Fig. 2b for ex-Ti3C2), and then, the further ultrasonication will result in the formation and the pulverization of individual Ti3C2 sheets (Fig. 2c for d-Ti3C2). However, the pulverized individual Ti3C2 sheets are very easy to agglomerate. The introduction of the CNTs into the dTi3C2 can effectively impede the aggregation of individual Ti3C2 sheets (Fig. 2d). TEM observations further demonstrate (shown in Fig. 3) that the ultrasonication of 2D Ti3C2 can obtain delaminated Ti3C2 sheets and the CNTs and d-Ti3C2 are interweaved together in the d-Ti3C2/CNT composite. The FTIR spectra of CNTs, d-Ti3C2 and d-Ti3C2/CNT are given in Fig. 4. It can be seen that the C]O (~1645 cm1) functional groups exist in all the samples. In addition, the CeO (~1051 cm1), CeF (~1090 cm1) and OeH (~1389 cm1) groups can be observed in the spectra of d-Ti3C2 and d-Ti3C2/CNT. Obviously, almost all the oxygen-containing functional groups in the MXenes can still be retained in the synthesized d-Ti3C2/CNT composite, suggesting that the ultrasonication has no significant effect on the oxygencontaining functional groups. This result also implies that the synthesized d-Ti3C2/CNT composite will exhibit pseudocapacitive performance due to the existence of the oxygen-containing functional groups. The CV curves of 10 mv s1 for d-Ti3C2 and d-Ti3C2/CNT composites at different weight ratios of d-Ti3C2 to CNT (6:1, 2:1 and 1:1)
40
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43
Fig. 1. The XRD patterns of Ti3AlC2, 2D Ti3C2, ex-Ti3C2, d-Ti3C2, CNTs and d-Ti3C2/CNT.
Fig. 2. SEM images of 2D Ti3C2 (a), ex-Ti3C2 (b), d-Ti3C2 (c) and d-Ti3C2/CNT (d).
are shown in Fig. 5a. The curve shapes of these samples are dramatically different from that of porous carbon, which generally show a rectangular-like shape indicating electric double-layer capacitance. The strong peaks can be observed in Fig. 5a, which suggests that the capacitance of these samples mainly results from pseudocapacitive capacitance based on a redox mechanism. This should be ascribed to the oxygen-containing functional groups terminated on the 2D MXene surfaces (Fig. 4). There occurs apparent oxidation peaks (at roughly 0.44 V and 0.49 V) and reduction peak (at roughly 0.32 V) in the CV curves of all the dTi3C2/CNT composites. However, the oxidation peaks (at roughly 0.46 V and 0.49 V) in the CV curve of d-Ti3C2 are significantly weak
and unconspicuous (inset Fig. 5a), indicating the insufficient redox process of oxygen-containing functional groups for the d-Ti3C2 sample compared with those for the d-Ti3C2/CNT composites. This is to say, the introduction of CNTs can greatly facilitate the redox process of oxygen-containing functional groups on the d-Ti3C2. This improved redox process should be related to the effective impedance on the aggregation of individual Ti3C2 sheets due to the introduction of CNTs. In addition, the improved conductivity of the d-Ti3C2/CNT composites (Table 1) also contributes to the enhanced electrochemical performance of these composites. The influence of the CNT content on the volumetric capacitance of d-Ti3C2/CNT composites at 10 mv s1 is also investigated, as
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43
41
Fig. 3. The TEM images of d-Ti3C2 (a) and d-Ti3C2/CNT (b).
shown in Fig. 5b. The volumetric capacitance of d-Ti3C2/CNT composite gradually increases with the increasing CNT content, reaching the maximal value of 384 F cm3 at the 2:1 weight ratio of dTi3C2 to CNT, and thereafter, the volumetric capacitance begins to decrease as the CNT content further increases. In any case, however, the introduction of CNTs at different weight ratios certainly leads to significantly higher volumetric capacitance compared with that of the pure d-Ti3C2. This means that the introduced CNTs, which are interweaved together with d-Ti3C2, impede the stacking of individual Ti3C2 sheets effectively and enlarge the distance between Ti3C2 sheets, thus facilitating the utilization of the surface of samples, the diffusion of electrolyte ion and redox process. The CV curves of d-Ti3C2 and d-Ti3C2/CNT at different scan rates are shown in Fig. 6. Due to the introduction of CNTs and the improvement of conductivity, the d-Ti3C2/CNT shows better electrochemical performance and redox process than d-Ti3C2. Fig. 7 present the relation curves of volumetric capacitance and CV scan rates of d-Ti3C2 and d-Ti3C2/CNT. It can be observed that the volumetric capacitances of these two samples decrease gradually with the increasing of scan rates in a range of 5e100 mv s1. At a same scan rate, however, the d-Ti3C2/CNT exhibits obviously higher volumetric capacitance than the d-Ti3C2. For instance, at a scan rate of 5 mv s1, the volumetric capacitance of d-Ti3C2/CNT is up to 393 F cm3, almost 1.8 times as great as that of the pure d-Ti3C2. As the scan rate increases to 100 mv s1, the capacitance retentions of d-Ti3C2 and d-Ti3C2/CNT are about 52% and 80%, respectively. This
Fig. 4. FTIR spectra of CNTs, d-Ti3C2 and d-Ti3C2/CNT.
capacitance retention of d-Ti3C2/CNT is also greatly higher than that of the d-Ti3C2 reported before [25], implying that the introduction of CNTs can significantly improve the d-Ti3C2/CNT rate performance. The cycling stability is a most important criterion for the supercapacitor, and the long-term cycling stability of d-Ti3C2 and d-
Fig. 5. (a) The CV curves of 10 mv s1 for d-Ti3C2 and d-Ti3C2/CNT at different weight ratios. (b) The volumetric capacitance of d-Ti3C2/CNT as a function of the CNT content at 10 mv s1 in 6 M KOH electrolyte.
42
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43
Table 1 Electrical conductivity of d-Ti3C2 and d-Ti3C2/CNT at different weight ratios. Samples
d-Ti3C2
d-Ti3C2/CNT (6:1)
d-Ti3C2/CNT (2:1)
d-Ti3C2/CNT (1:1)
Conductivity (s m1)
20
113
230
332
Fig. 7. A plot of volumetric capacitance versus the scan rates of d-Ti3C2 and d-Ti3C2/ CNT.
Fig. 8. Cycling stability of d-Ti3C2 and d-Ti3C2/CNT at 10 mv s1.
Fig. 6. The CV curves of d-Ti3C2 (a) and d-Ti3C2/CNT (b) at different scan rates.
Ti3C2/CNT is investigated at a scan rate of 10 mv s1. Fig. 8 shows the volumetric capacitance of these two samples as a function of cycle number. It is worth noting that, when the cycling begins, the volumetric capacitance of both the samples increases, which is quite different from those for other redox materials in most cycling stability tests. The reason may be attributed to the sufficient activation of electrode materials at the early circulation of redox process [26]. Furthermore, the cyclic capacitance of d-Ti3C2/CNT (~384 F cm3) is far larger than that of d-Ti3C2 (~193 F cm3). More interestingly, the d-Ti3C2 and d-Ti3C2/CNT composite exhibit satisfactory cycling stabilities over the entire cycle numbers and almost 100% specific capacitance is retained even after 10000 cycles for these two samples. This excellent cycling stability is greatly better compared with those of the other typical electrode materials of pseudocapacitors, such as various types of metal oxides (MnO2 [11], NiO [12], RuO2 [27], SnO2 [28], and so on). 4. Conclusion In this article, the delaminated Ti3C2 sheets are prepared from MXenes by intercalation and ultrasonication, and the CNT is used to
inhibit the aggregations of delaminated Ti3C2 sheets. At the optimal d-Ti3C2 and CNT ratio of 2:1 in weight, the d-Ti3C2/CNT composite displays high volumetric capacitance (393 F cm3 at 5 mv s1), superior rate capability (80% capacitance retention as the scan rate increases from 5 to 100 mv s1) and excellent cycling stability over metal oxides. The introduction of CNTs can significantly enhance the supercapacitive performance of delaminated Ti3C2 by impeding the stacking of Ti3C2, enlarging the distance between Ti3C2 sheets and improving the electrical conductivity. Therefore, the d-Ti3C2/ CNT composite materials are quite promising pseudocapacitor materials with superior eletrochemical properties. Acknowledgment Financial support of this work by National Science Foundation of China (NSFC) (No. 50975247) and Hebei Natural Science Foundation (No. E2014203204) is acknowledged. References [1] D. Pech, M. Brunet, H. Durou, P. Huang, V. Mochalin, Y. Gogotsi, P.L. Taberna, P. Simon, Nat. Nanotech. 5 (2010) 651e654. [2] P. Simon, Y. Gogotsi, Nat. Mater. 7 (2008) 845e854. [3] B.E. Conway, J. Electrochem. Soc. 138 (1991) 1539e1548. [4] J.R. Miller, A.F. Burke, Electrochem. Soc. Interf. 17 (2008) 53e57. [5] D. Hulicova-Jurcakova, M. Seredych, G.Q. Lu, N.K.A.C. Kodiweera, P.E. Stallworth, S. Greenbaum, T.J. Bandosz, Carbon 47 (2009) 1576e1584. [6] Y. Zhao, W. Ran, J. He, Y. Song, C. Zhang, D. Xiong, F. Gao, J. Wu, Y. Xia, ACS Appl. Mater. Interfaces 7 (2015) 1132e1139. [7] J. Xu, R. Zhang, P. Chen, S. Ge, J. Power Sources 246 (2014) 132e140.
P. Yan et al. / Journal of Power Sources 284 (2015) 38e43 [8] P. Yan, J. Xu, C. Wu, R. Zhang, J. Jin, Mater. Lett. 139 (2015) 340e343. [9] Y. Zhao, W. Wang, D. Xiong, G. Shao, W. Xia, S. Yu, F. Gao, Int. J. Hydrog. Energy 37 (2012) 19395e19400. [10] A.M. Hashem, H.M. Abuzeid, A.E. Abdel-Ghany, A. Mauger, K. Zaghib, C.M. Julien, J. Power Sources 202 (2012) 291e298. [11] X. Li, A. Dhanabalan, C. Wang, J. Power Sources 196 (2011) 9625e9630. [12] Y. Zhao, W. Ran, J. He, Y. Huang, Z. Liu, W. Liu, Y. Tang, L. Zhang, D. Gao, F. Gao, Small (2014), http://dx.doi.org/10.1002/smll.201401922. [13] A. Rudge, J. Davey, I. Raistrick, S. Gottesfeld, J. Power Sources 47 (1994) 89e107. [14] Q. Hu, D. Sun, Q. Wu, H. Wang, L. Wang, B. Liu, A. Zhou, J. He, J. Phys. Chem. A 117 (2013) 14253e14260. [15] Q. Peng, J. Guo, Q. Zhang, J. Xiang, B. Liu, A. Zhou, R. Liu, Y. Tian, J. Am. Chem. Soc. 136 (2014) 4113e4116. [16] M.R. Lukatskaya, O. Mashtalir, C.E. Ren, Y. Dall’Agnese, P. Rozier, P.L. Tabetrna, M. Naguib, P. Simon, M.W. Barsoum, Y. Gogotsi, Science 341 (2013) 1502e1505. [17] Q. Tang, Z. Zhou, P. Shen, J. Am. Chem. Soc. 134 (2012) 16909e16916. [18] D. Sun, M. Wang, Z. Li, G. Fan, L. Fan, A. Zhou, Electrochem. Commun. 47
43
(2014) 80e83. [19] M.W. Barsoum, Prog. Solid State Chem. 28 (2000) 201e281. [20] O. Mashtalir, M. Naguib, V.N. Mochalin, Y. Dall’Agnese, M. Heon, M.W. Barsoum, Y. Gogotsi, Nat. Commun. 4 (2013) 1716. [21] M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi, M.W. Barsoum, Adv. Mater. 23 (2011) 4248e4253. [22] M. Sun, G. Wang, X. Li, C. Li, J. Power Sources 245 (2014) 436e444. [23] F. Zeng, Y. Kuang, N. Zhang, Z. Huang, Y. Pan, Z. Hou, H. Zhou, C. Yan, O.G. Schmidt, J. Power Sources 247 (2014) 396e401. [24] M.Y. Yen, M.C. Hsiao, S.H. Liao, P.I. Liu, H.M. Tsai, C.C.M. Ma, N.W. Pu, M.D. Ger, Carbon 49 (2011) 3597e3606. [25] Y. Dall’Agnese, M.R. Lukatskaya, K.M. Cook, P.L. Taberna, Y. Gogotsi, P. Simon, Electrochem. Commun. 48 (2014) 118e122. [26] T.Y. Wei, C.H. Chen, H.C. Chien, S.Y. Lu, C.C. Hu, Adv. Mater. 22 (2010) 347e351. [27] T.F. Hsieh, C.C. Chuang, W.J. Chen, J.H. Huang, W.T. Chen, C.M. Shu, Carbon 50 (2012) 1740e1747. [28] S.N. Pusawale, P.R. Deshmukh, C.D. Lokhande, Appl. Surf. Sci. 257 (2011) 9498e9502.