Chinese Chemical Letters  2019, Vol. 30 Issue (5): 1121-1125   PDF    
Surface modulated hierarchical graphene film via sulfur and phosphorus dual-doping for high performance flexible supercapacitors
Xu Yu, Chengang Pei, Ligang Feng*     
School of Chemistry and Chemical Engineering, Yangzhou University, Yangzhou 225002, China
Abstract: Graphene surface modification by heteroatom incorporation is an attractive strategy to construct flexible electrochemical capacitors. Herein, the steam-assistant heteroatoms of sulfur and phosphorus dualdoped graphene film (s-SPG) is fabricated via an ice-template and thermal-activation approach and they demonstrate an excellent pseudocapacitive behavior in flexible electrochemical capacitors. As probed by various microscopic and spectroscopic analysis, well-maintained porosity structure is formed during the freeze-drying and steam-activation treatment; the increased surface roughness is ascribed to heteroatoms doping by the formation of S- and P-containing functional groups as electrochemical active sites. A flexible device integrated by porous s-SPG film shows high pseudocapacitive behavior with high specific capacitance (169 F/g), rate capability (91.7%) and cyclic stability (92.5%). Even at the bend angle of 120°, no obvious change of specific capacitance is found indicating a good flexibility of s-SPG devices; the current study shows that s-SPG is a very promising electrode to realize the practical applications of all solid flexible supercapacitors.
Keywords: Heteroatom     Flexible devices     Steam-activation     Supercapacitor     Graphene film    

Flexible electrochemical capacitor (EC) is of significant to meet the fast growth of modern electronic devices for consumer discretionary [1-3]. Compared to conventional supercapacitors [4, 5], flexible ECs as a promising candidate have attracted more interest due to their superior mechanical property and electrochemical behaviors [6, 7]. However, the low energy density for flexible ECs has restricted their extensive practical applications, and low capacitance is still a key challenge to enhance the energy density to maintain the power density. To solve this issue, it is necessary to construct the flexible electrodes with a large surface area, high mechanical stability and electrical conductivity.

Pristine carbonaceous nanomaterials (activated carbon [8], carbon nanotube [9] and graphene [10-12]) with electrical double layer capacitive behavior could not provide sufficient specific capacitance. Many efforts have demonstrated that the construction of carbon nanocomposites by hybridization with pseudocapacitive materials is an efficient way to significantly improve the electrochemical performance. Deposition of the common pseudocapacitive materials (transition metal oxides and conducting polymers) [13-16] on the carbon surface shows an obvious enhancement of supercapacitive performance. However, the wide applications are seriously restricted by the low conductivity and poor cyclic stability [17, 18]. Recently, heteroatoms incorporated into graphene lattices have been applied in energy storage systems like lithium-ion battery [19, 20], supercapacitors [21, 22] and fuel cells [23-25]. Due to the different atomic radius and electronegativity, the formation of heteroatom-containing functional groups acting as the electrochemical actives sites is beneficial to increase the capacitance by additional 'pseudocapacitance' accompanying a redox reaction. Excellent electrochemical properties have been observed on the heteroatoms (S, P, B or N) doped graphene films in the flexible all-solid-state supercapacitors [26]. Heteroatoms dualdoped system shows a dramatic enhancement of capacitive behavior due to the synergistic effect of different functional groups [27, 28]. The optimization of the structure feature through ice-template and steam-assistant activation can well-maintain the porous structure for fast electrolyte diffusion, and further efficiently enlarge the surface area of the electrode materials [29]. However, heteroatoms (S and P) dual-doped graphene porous film with additional ice-template and steam-activation treatment has yet to be explored for the flexible electrochemical capacitors.

In this work, heteroatoms dual-doped graphene film via a facial ice-template and steam-assistant activation treatment was proposed for application in the flexible electrochemical capacitor by overcoming the low specific capacitance of graphene. The porous structure is cross-linked by the functionalized graphene nanosheets via ππ interaction and hydrogen bonding; and steamassistant activation is favorable to increase the porosity for fast ion diffusion and expose the efficient active sites. The flexible s-SPG electrochemical capacitors possess a high specific capacitance of 169 F/g with a good rate capability (91.7%) and cyclic stability (92.5%). Moreover, a flexible device is constructed by the above proposed electrode, that exhibits an excellent electrochemical performance at various bend states. The high performance is attributed to the combination of electrical double layer capacitance deriving from the high surface area of graphene and effective pseudocapacitance originating from the heteroatoms doping.

The porous s-SPG electrode was prepared through a selfassembly, freeze-drying and steam-activation as shown in Fig. 1a. In brief, the homogeneous mixture of PA, TGA and graphene oxide was obtained by bath-sonication for 30 min, and then the functionalized graphene film was self-assembled by the ππ interaction of graphene layers and hydrogen bonding between graphene and heteroatoms precursors. By controlling the residual level of water, the functionalized graphene film was frozen by liquid nitrogen and freeze-dried in vacuum condition for 3 days to maintain the porous structure. Finally, the functionalized film was steam-activated at 900 ℃ for 1 h to obtain porous S and P dual doped graphene film. During the thermal activation process, heteroatoms (S and P) were successfully incorporated into graphene lattices at the high annealing temperature.

Download:
Fig. 1. (a) The synthetic procedure of s-SPG film. (b, c) SEM images of s-SPG film. (d) High-resolution TEM, (e) STEM image and corresponding elemental distribution of s-SPG film. (f) Nitrogen adsorption/desorption isotherms of s-SPG and s-G (inset: the surface area, pore volume and pore diameter).

The structure and morphologies of s-SPG film were characterized by electron microscope techniques. Figs. 1b and c exhibit the cross-section scanning electron microscope (SEM) images of s-SPG film. The thickness of s-SPG is about 40~50 mm. The macroporouswall for s-SPG film is constructed by the partial restacking of graphene nanosheets through the weak ππ interaction and strong expended π-stacking. The surface of s-SPG film is wrinkled and crumpled due to the heteroatoms (S and P) doping by the different electronegativity and bond angle. As evaluated by nitrogen adsorption/desorption (Fig. 1f), the surface area ofs-SPG film (256.8 m2/g) is larger than that of s-G film (139.6 m2/g) and the average diameter of mesopores for s-SPG is 10.5 nm. The large surface area of s-SPG film can be attributed to the generation of micro- and mesopores by heteroatoms doping which leads to graphene lattice distortion and formation of defects and vacancies [30]. Furthermore, the crumpled and wrinkled morphology of s-SPG film was confirmed by transmission electron microscopy (TEM) images (Fig. 1d). The wrinkled surface morphology can exposemore active-sites for the fast redox reaction. It can be observed that s-SPG film consists of C, O, S and P elements from the scanning elemental mapping images (Fig. 1e), and all the elements are uniformly distributed implying the successfully doping of heteroatoms S and P into graphene lattices.

The structure of s-SPG was investigated by XRD and Raman spectra. After thermal annealing at 900 ℃, the appearance of (002) pattern at 2θ = 24.2° evidences the reduction of graphene oxides accompanying the disappearance of peak at 2θ = 14.1° from graphene oxide (Fig. 2a). Meanwhile, the typical (002) pattern of s-SPG film at 2θ = 24.1° becomes more weaken and broader than that of s-G film due to the decreased crystallinity degree of graphitic carbon. This feature is induced by the formation of S- and P-containing groups due to the different electronegativity from carbon atoms and different bond angle and bond length from C-C bonds. Raman spectroscopy analysis was further applied to verify the structure change and defect generation of s-SPG film. As shown in Fig. 2b, s-G film shows the typical D band and G band at 1347 cm-1 and 1587 cm-1 corresponding to the distortion of graphitic structure and vibration of graphitic crystal plane, respectively. However, the D band of s-SPG film is downshifted to 1343 cm-1 and G band is shifted to 1581 cm-1 owing to the heteroatom doping into graphene lattices. The value of ID/IG (intensity ratio of D band and G band) is a crucial factor to verify the structural distortion of carbon-based materials [27]. The ID/IG ratio of s-SPG (1.16) is higher than that of s-G film (1.04) owing to the defects formed by dual heteroatom (S and P) doping.

Download:
Fig. 2. (a) XRD patterns and (b) Raman spectra of GO, s-G and s-SPG. High resolution XPS spectra of (c) S 2p and (d) P 2p.

The chemical nature of the s-SPG film was investigated by X-ray photoelectron spectroscopy (XPS). As shown in the full survey of XPS (Fig. S1 in Supporting information), the s-SPG film exhibits four peaks at 133 eV, 164 eV, 284 eV and 532 eV corresponding to P, S, C and O with the atomic ratio of 3.2%, 4.8%, 84.9% and 7.1%, respectively. The bonding configuration and chemical nature of s-SPG film are confirmed by the high-resolution C 1s, S 2p and P 2p peaks (Fig. S2 in Supporting information and Figs. 2c and d). The high magnification of C 1s peak can be fitted to five peaks at 284.1 eV, 284.6 eV, 285.3 eV, 285.8 eV and 287.2 eV corresponding to C–S, C-C, C–P, C–O and C=O configurations. Especially, the appearance of C–S and C–P bonds indicates the successful dual heteroatoms (S and P) doping into graphene nanosheets. The first two deconvoluted peaks at 163.9 eV and 165.2 eV correspond to the spin-orbital of S 2p, and the third peak at 169.3 eV can be assigned to the oxygen-sulfur components. From the high magnification P 2p peak of s-SPG film, the peaks center at 133.2 eV and 134.1 eV owing to the C-P and P-O bonds, respectively. It is proved that graphene modified by sulfur is expected to have a wider band gap due to the electron-withdrawing character of S; and S-doped carbon possesses more discords and better conductivity than pristine carbon [31]. The reversible and pseudocapacitive feature of P-doping graphene is coupled with the variation of electrochemically active and stable C-P=O bonding, and the charge transfer resistance is reduced by the modification of the electronic structure [32]. Thus, the synergistic effect of S- and P-containing groups can be acted as active sites for the improvement of the electrochemical performance.

The flexible solid-state cell with symmetric sandwich-structure was fabricated with two same mass of film electrodes and a PVA/ H2SO4 gel electrolyte; the relative capacitive behavior was initially investigated by cyclic voltammetry (CV) at the potential window of 0–1.0 V with various scan rate from 10 mV/s to 500 mV/s. As shown in Fig. 3a, the CV curves of s-G exhibit a nearly rectangular shape at 0 mV/s and s-SPG shows a distorted rectangular profile with higher specific current density due to the pseudocapacitive effect by heteroatom doping and unique porous structure. Meanwhile, the specific capacitance is calculated by the integration of CV curves at the scan rate of 10 mV/s and the values of s-SPG is approach to 158 F/g, which is superior to SPG (129 F/g), s-G (83 F/g), G (70 F/g). In contrast to SPG, the specific capacitance of s-SPG is increased by 29 F/g and the specific capacitance for s-G and G has a similar trend. This result can be attributed to the structure change by the steam-activation treatment for exposing more electroactive sites. Meanwhile, the specific capacitance for s-SPG is increased by 75 F/g compared with s-G, and the specific capacitance change for SPG and G is in a similar trend. This result indicates that the dominated enhancement of specific capacitance is determined by heteroatom doping effect. As the scan rate increased from 10 mV/s to 500 mV/s (Fig. 3b), s-SPG shows a slight distortion of CV curves. To further confirm the pseudocapacitive properties of s-SPG, the GCD analysis was carried out at the current density of 1 A/g. As shown in Fig. 3c, s-G and G show the symmetric triangle-shape profiles corresponding to the electrical double layer capacitance, which is well matched with the rectangular shape of CV curves. However, s-SPG and SPG show a distorted triangle shape which can be ascribed to the pseudocapacitive effect. Meanwhile, s-SPG shows a smaller IR drop and longer duration time than other devices indicating a good electrical conductivity and reversible pseudocapacitive property. The specific capacitance of s-SPG is calculated to be 169 F/g at the current density of 1 A/g, which is 2.3, 1.9 and 1.2 times higher than that of G, s-G and SPG. The volumetric performance is of great importance for practical applications [33, 34], it is calculated to be 52.48 F/cm3 with the density is 0.31 g/cm3. The significant improvement of the specific capacitance of s-SPG can be attributed to the formation of pseudocapacitive properties by the synergistic effect of dual heteroatoms doping and unique structure morphology. Furthermore, the GCD curves of s-SPG at a current density of 1 A/g was measured with three same solid-state cells in a parallel and series way (Fig. 3d). The discharge time of the three-constructed cells is approached to three times in a parallel way with the potential window of 1.0 V. Meanwhile, the potential of the three-constructed cells reaches up to 3.0 V in a series way and the cell can make the LED light work.

Download:
Fig. 3. (a) CV curves of G, s-G, SPG and s-SPG at the scan rate of 10 mV/s. (b) CV curves of s-SPG at various scan rate. (c) GCD curves of G, s-G, SPG and s-SPG at the current density of 1 A/g. (d) GCD curves of s-SPG in a parallel and series way (inset: the LED lighting by a series way of s-SPG devices). (e) GCD curves of s-SPG at various current density. (f) Rate capability and (g) cyclic stability of G, s-G, SPG and s-SPG. (h) Nyquist plots and (i) enlarged Nyquist plots of G, s-G, SPG and s-SPG.

To further evaluate the pseudocapacitive behavior of the flexible s-SPG devices, the rate capability at the current density from 1 A/g to 30 A/g and cyclic stability with 2000 discharge/ charge cycles for practical application were studied at the current density of 5 A/g. As the current density increased by a factor of 30, the specific capacitance of s-SPG is 155 F/g with the capacitance retention of 91.7%, superior to SPG (85.2%), s-G (65.9%) and G (61.2%) (Figs. 3e and f). Thus, the surface modification by heteroatoms doping plays a dominant role to improve the stability of flexible devices. The Coulombic efficiency of s-SPG at different current density is higher than 100% (Table S1 in Supporting information), which may be attributed to the produced faradic current arising from heteroatom doping during the discharge process. Furthermore, the specific capacitance of s-SPG is maintained with a value of 150 F/g to its initial state after 2000 discharge/charge cycles. And the capacitance retention of s-SPG is 92.5%, which was 8%, 20.3% and 30.7% higher than that of SPG, s-G and G (Fig. 3g). The dynamic behavior and electric resistance of allsolid-state symmetric cell with s-SPG film were evaluated by electrochemical impedance spectroscopy (Figs. 3h and i). The oblique line with a slope of near 90° at low frequency corresponds to an ideal polarizable capacitance, and the existence of semi-circle at the middle frequency is attributed to the charge transfer resistance (Rct) at the electrode/electrolyte interface. At high frequency, the value of resistance can be ascribed to the internal resistance of electrode materials. The Rct of s-SPG is 3.9 Ω, which is smaller than that of SPG (4.7 Ω), s-G (5.7 Ω) and G (7.6 Ω). The feature implies a dramatical acceleration for the charge transportation during the fast faradic reaction.

The mechanical property was further probed by testing the allsolid-state devices with the different bending angle of 60°, 90° and 120° at the scan rate of 10 mV/s. As shown in Fig. 4a, all CV curves of s-SPG at various bending angle maintain almost similarly distorted rectangular shape. The specific capacitance is degraded with a value of 9 F/g, 7 F/g and 4 F/g at bending angle of 60°, 90° and 120°, while the specific capacitance for other devices shows a similar change (Fig. 4b). Furthermore, the electrochemical behavior of the all-solid-state devices is measured by CV after keeping at air condition for 3 days, 7 days and 15 days (Fig. 4c) with a bending angle of 120° at the scan rate of 10 mV/s. The specific capacitance of s-SPG is just 4 F/g, 9 F/g and 16 F/g loss in contrast to the initial state with the bending of 120°, but other devices show a fast degradation trend. These results demonstrate that s-SPG possesses an excellent flexibility and pseudocapacitive properties.

Download:
Fig. 4. (a) CV curve of flexible s-SPG devices with the various bending angle at the scan rate of 10 mV/s. The specific capacitance of flexible s-SPG, SPG, s-G and G devices with different bending angles (b) and conservation time at the bending angle of 120° (c).

The abovementioned electrochemical performance of s-SPG in the flexible supercapacitor can be ascribed to the synergistic effect of hierarchical structure and heteroatom doping. The hierarchical structure is favorable for fast ion diffusion, charge transfer and electro-active sites exposure. Meanwhile, the surface chemistry is uniformly modified by heteroatoms (S and P) doping, and the generation of the S- and P-containing groups can serve as the electrochemical active sites where the fast-faradic reaction occurs. In all, the synergistic effect of porous structure and surface modification by heteroatoms doping for graphene can provide an efficient way to prepare the electrode materials for a flexible electrochemical capacitor with highly stable properties.

Acknowledgments

The work was supported by the Natural Science Foundation of the Jiangsu Higher Education Institutions of China; the National Natural Science Foundation of China (Nos. 21603041, 21805239); and the Priority Academic Program Development of Jiangsu Higher Education Institutions. We also acknowledge Testing Center of Yangzhou University for technical support.

Appendix A. Supplementary data

Supplementary material related to this article can befound, in the online version, at doi:https://doi.org/10.1016/j.cclet.2019.01.009.

References
[1]
J. Masa, W. Xia, I. Sinev, et al., Angew. Chem. Int. Ed. 53 (2014) 8508-8512. DOI:10.1002/anie.v53.32
[2]
Y. Ko, M. Kwon, W.K. Bae, et al., Nat. Commun. 8 (2017) 536. DOI:10.1038/s41467-017-00550-3
[3]
M. Chen, Y. Yang, D. Chen, H. Wang, Chin. Chem. Lett. 29 (2018) 564-570. DOI:10.1016/j.cclet.2017.12.019
[4]
X. Lu, D.Y.C. Leung, H. Wang, M.K.H. Leung, J. Xuan, ChemElectroChem 1 (2014) 836-849. DOI:10.1002/celc.v1.5
[5]
X. Lin, H. Lou, W. Lu, et al., Chin. Chem. Lett. 29 (2018) 633-636. DOI:10.1016/j.cclet.2017.11.024
[6]
J.Y. Shieh, S.Y. Tsai, B.Y. Li, H.H. Yu, Mater. Chem. Phys. 195 (2017) 114-122. DOI:10.1016/j.matchemphys.2017.04.034
[7]
J. Yan, C.E. Ren, K. Maleski, et al., Adv. Funct. Mater. 27 (2017) 1701264. DOI:10.1002/adfm.201701264
[8]
Y. Shao, M.F. El-Kady, L.J. Wang, et al., Chem. Soc. Rev. 44 (2015) 3639-3665. DOI:10.1039/C4CS00316K
[9]
C. Choi, J.A. Lee, A.Y. Choi, et al., Adv. Mater. 26 (2013) 2059-2065.
[10]
Z. Xiong, C. Liao, W. Han, X. Wang, Adv. Mater. 27 (2015) 4469-4475. DOI:10.1002/adma.v27.30
[11]
X. Yu, C.G. Pei, W.S. Chen, L.G. Feng, Electrochim. Acta 272 (2018) 119-126. DOI:10.1016/j.electacta.2018.03.201
[12]
X. Yu, L. Feng, H.S. Park, J. Power Sources 390 (2018) 93-99. DOI:10.1016/j.jpowsour.2018.04.032
[13]
Z. Wang, D.O. Carlsson, P. Tammela, et al., ACS Nano 9 (2015) 7563-7571. DOI:10.1021/acsnano.5b02846
[14]
R.R. Salunkhe, J. Tang, Y. Kamachi, et al., ACS Nano 9 (2015) 6288-6296. DOI:10.1021/acsnano.5b01790
[15]
C. Guan, J. Liu, Y. Wang, et al., ACS Nano 9 (2015) 5198-5207. DOI:10.1021/acsnano.5b00582
[16]
L. Feng, H. Xue, ChemElectroChem 4 (2017) 20-34. DOI:10.1002/celc.201600563
[17]
Z. Zhang, M. Liao, H. Lou, et al., Adv. Mater. 30 (2018) 1704261. DOI:10.1002/adma.v30.13
[18]
H.W. Tang, N. Gao, Z.R. Chang, et al., Chin. Chem. Lett. 25 (2014) 269-272. DOI:10.1016/j.cclet.2013.11.051
[19]
W. Ai, Z. Luo, J. Jiang, et al., Adv. Mater. 26 (2014) 6186-6192. DOI:10.1002/adma.201401427
[20]
Z. Xing, Z. Ju, Y. Zhao, et al., Sci. Rep. 6 (2016) 26146. DOI:10.1038/srep26146
[21]
K. Fujisawa, R. Cruz-Silva, K.S. Yang, et al., J. Mater. Chem. A 2 (2014) 9532-9540. DOI:10.1039/C4TA00936C
[22]
X.W. Wang, G.Z. Sun, P. Routh, et al., Chem. Soc. Rev. 43 (2014) 7067-7098. DOI:10.1039/C4CS00141A
[23]
J. Liang, Y. Jiao, M. Jaroniec, S.Z. Qiao, Angew. Chem. Int. Ed. 51 (2012) 11496-11500. DOI:10.1002/anie.201206720
[24]
Z. Liu, Y. Wang, L. Feng, J. Solid State Electrochem. 22 (2018) 3467-3474. DOI:10.1007/s10008-018-4061-5
[25]
Z. Liu, X. Yu, H. Yu, H. Xue, L. Feng, ChemSusChem 11 (2018) 2703-2709. DOI:10.1002/cssc.v11.16
[26]
A.G. Garcia, S.E. Baltazar, A.H.R. Castro, J.F.P. Robles, A. Rubio, J. Comput. Theor. Nanosci. 5 (2008) 2221-2229. DOI:10.1166/jctn.2008.1123
[27]
X. Yu, Y. Kang, H.S. Park, Carbon 101 (2016) 49-56. DOI:10.1016/j.carbon.2016.01.073
[28]
Z.S. Wu, A. Winter, L. Chen, et al., Adv. Mater. 24 (2012) 5130-5135. DOI:10.1002/adma.201201948
[29]
M. Kota, X. Yu, S.H. Yeon, H.W. Cheong, H.S. Park, J. Power Sources 303 (2016) 372-378. DOI:10.1016/j.jpowsour.2015.11.006
[30]
X. Ma, G. Ning, Y. Kan, et al., Electrochim. Acta 150 (2014) 108-113. DOI:10.1016/j.electacta.2014.10.128
[31]
H.L. Poh, P. Šimek, Z. Sofer, M. Pumera, ACS Nano 7 (2013) 5262-5272. DOI:10.1021/nn401296b
[32]
X. Yu, H.J. Kim, J.Y. Hong, et al., Nano Energy 15 (2015) 576-586. DOI:10.1016/j.nanoen.2015.05.010
[33]
Y. Gogotsi, P. Simon, Science 334 (2011) 917-918. DOI:10.1126/science.1213003
[34]
Q. Wang, J. Yan, Z. Fan, Energy Environ. Sci. 9 (2016) 729-762. DOI:10.1039/C5EE03109E