Journal of Colloid and Interface Science 421 (2014) 160–164
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Heteroatom doped mesoporous carbon/graphene nanosheets as highly efficient electrocatalysts for oxygen reduction Peimin Xu a, Dongqing Wu b, Li Wan a, Pengfei Hu c, Ruili Liu a,⇑ a
Department of Chemical Engineering, School of Environment and Chemical Engineering, Shanghai University, Shangda Road 99, 200444 Shanghai, PR China School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Dongchuan Road 800, 200240 Shanghai, PR China c Lab of Microstructure, Shanghai University, Shangda Road 99, 200444 Shanghai, PR China b
a r t i c l e
i n f o
Article history: Received 24 October 2013 Accepted 2 February 2014 Available online 8 February 2014 Keywords: Heteroatom doped carbon Metal-free catalysts Oxygen reduction Fuel cell
a b s t r a c t The high cost of platinum (Pt) based catalysts for oxygen reduction reaction (ORR) has restricted the widespread commercialization of fuel cells. Heteroatom (N, B, P, S or Se) doped carbon materials have been regarded as the promising metal-free catalysts for replacing Pt based catalysts owing to their high efficiencies, good stability and relative low cost. In this work, we present a cost-effective synthesis approach for heteroatom (N and S) doped mesoporous carbon/graphene (HMCG) nanosheets by using nano-casting technology with mesoporous silica/graphene nanosheets (MSG) as hard templates, and four different amino acids (alanine, serine, arginine and cystine) as heteroatom (N, S) and carbon precursors. The resulting catalysts exhibited excellent electrocatalytic activity for ORR in alkaline media. In particular, HMCGAla with alanine as precursors showed the highest electron transfer numbers and durability. These results indicated the attractive potential of HMCGs as metal-free catalysts in practical fuel cells. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Fuel cell, which directly converts chemical energy into electricity, is regarded as one of the most attractive energy conversion systems because of its high efficiency, low pollution and fuel flexibility [1–3]. Generally, the cell performance of fuel cells is mainly determined by the oxygen-reduction reaction (ORR) happened in their cathodes [4]. Platinum (Pt) based materials have been investigated as the most active ORR catalysts for decades. However, the commercialization of fuel cells is still retarded by the poor durability and high cost of Pt catalysts [5–7]. Therefore, the development of efficient ORR electrocatalysts derived from non-precious metal (e.g. Fe, Co or Ni) [8–12] or heteroatom (e.g. N, B, P, S or Se) doped carbon materials [13–17] has become an urgent research topic for fuel cells. Recently, heteroatom doped carbon materials (e.g. carbon nanotubes [18–22], nanotube cups [23], ordered mesoporous graphitic arrays [24,25] and graphene-based nanosheets [26–30]) have shown excellent electrocatalytic activities as well as low costs, good durability, and environmental friendliness, which open the way to a new class of metal-free catalysts for ORR. Generally, the types of heteroatoms and their weight content ratio decide the intrinsic catalytic behavior of the
⇑ Corresponding author. Fax: +86 21 5633 3063. E-mail address: [email protected] (R. Liu). http://dx.doi.org/10.1016/j.jcis.2014.02.001 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
heteroatom doped carbon materials [31]. On the other hand, the mesostructured properties of these catalysts such as porosity and surface area also hold the key for the catalytic efficiency of them [32]. Based on these considerations, heteroatom doped carbons with porous structures are regarded as excellent candidates for carbon based ORR catalysts [24,26,27]. In the last few years, graphene, a two-dimensional (2D) array of sp2 carbon atoms, has received intensive attentions due to its unique flat morphology, intriguing physical and chemical properties such as high conductivity [33], mechanical strength [34] and thermal stability [35]. It is thus believed that the combination of graphene sheets and porous heteroatom doped carbons will lead to high performance metal-free ORR catalysts. For instance, Yang and the coworkers [26] prepared graphene-based carbon nitride (G-CN) nanosheets by using graphene-based mesoporous silica nanosheets as templates, and ethylenediamine/carbon tetrachloride as precursor. Nevertheless, the toxicity of ethylenediamine and carbon tetrachloride limited the large scale production of the G-CN nanosheets as ORR catalysts. Shi’s group [27] reported sandwichstructure nitrogen-doped mesoporous carbon modified graphene sheets for ORR via a soft template route in which triblock copolymer (Pluronic F127) was used as the template for guiding the growth of nanoporous structures of phenolmelamine–formaldehyde-pre-polymer (PMF) as carbon source on graphene surface. With colloidal silica nanoparticles as hard templates, Qiao’s group obtained sulfur and nitrogen dual-doped mesoporous graphene
P. Xu et al. / Journal of Colloid and Interface Science 421 (2014) 160–164
161
nanosheets (MSG) were applied as the hard template for the growth of heteroatom doped mesoporous carbon layers on graphene sheets. In contrast to other carbon sources, amino acids are abundant, inexpensive, biocompatible and eco-friendly. More importantly, amino acids containing multiple heteroatoms such as nitrogen and sulfur enable the formation of dual-doped carbon with different heteroatoms. As the results, these HMCG catalysts have large specific surface areas (245–281 m2 g1), high onset potential (0.05 V vs. SHE) and excellent long term stability for ORR, which were superior to those observed for the commercially available Pt–C catalyst. A comparison of NMGCs derived from different carbon precursors suggests that both nitrogen (especially graphitelike nitrogen atoms) and sulfur atoms account for the excellent electrochemical performance in the ORR.
2. Experimental Scheme 1. The synthesis procedure of HMCGs: (a) self-assembly of mesoporous TEOS on GO, (b) removal of CTAB and (c) nanocasting approach of each amino acid.
2.1. Preparation of mesoporous silica/graphene (MSG) composites
materials by using melamine and benzyl disulfide (BDS) as N and S precursors [36]. Herein, we report an environmentally friendly approach toward heteroatom (N and S) doped mesoporous carbon/graphene (HMCG) nanosheets in which graphene sheets were sandwiched by mesoporous carbon layers containing heteroatom atoms (Scheme 1). In our method, amid acids were used as the source for carbon and heteroatoms and mesoporous silica/graphene
Graphene oxide (GO) was synthesized from natural graphite powders by a modified Hummers method [37]. MSGs were prepared via a soft template route with the assistance of electrostatic interaction between negatively charged GO and cationic surfactant [38]. In a typical experiment, 30 mg GO was firstly suspended in an aqueous solution containing 1 g CTAB and 40 mg NaOH, and then ultrasonically treated for 3 h. After magnetic stirring for 2 h at 40 °C, 1 mL TEOS was slowly added to the above mixture. After reaction for 12 h, MSGs were obtained by washing the precipitants
Fig. 1. SEM (a and b) and TEM (c and d) images of HMCGAla.
162
P. Xu et al. / Journal of Colloid and Interface Science 421 (2014) 160–164
with warm ethanol, separation, drying and the subsequent pyrolysis at 900 °C for 2 h in nitrogen. 2.2. Preparation of heteroatom doped mesoporous carbon/graphene nanosheets HMCGs were prepared by using MSGs as a hard template via a nanocasting approach. Under magnetic stirring at 40 °C, 0.5 g template was repeatedly impregnated with 50 mL aqueous solution of amino acids. In this work, four kinds of amino acids as alanine (C3H7NO2), arginine (C6H14N4O2), serine (C3H7NO3) and cysteine (C6H12N2O4S2) were utilized as the precursors. The weight of each amino acid was fixed to the pore volume of the MSGs (1.0 cm3 g1). After 6 h, the mixtures of MSGs and corresponding amino acids was placed in a drying oven for 6 h at 160 °C. Later on, the solids were calcined at 900 °C for 2 h in nitrogen. Subsequent etching of the resulting powders in a HF solution led to target HMCGs. According to their precursor, the obtained HMCGs were named as HMCGAla, HMCGArg, HMCGSer, and HMCGCys, respectively.
was polished mechanically with 0.5 lm diamond down to 0.05 lm alumina slurry to obtain a mirror-like surface and then washed with Milli-Q water and acetone. 1 mg HMCGs was dissolved in 1 mL solvent mixture of Nafion (5%) and EtOH (V:V ratio = 1:9) by sonication. For comparison, a commercially available catalyst of 20 wt% Pt supported on black carbon (from Alfa Aesar) was used and 1 mg mL1 Pt–C suspension was also prepared as the same procedure described above. 50 lL HMCGs or Pt–C suspension was pipetted on the glassy carbon electrode surface. The electrode was allowed to dry at RT before measurement. This led to a catalyst (HMCGs or Pt–C) loading 200 lg/cm2. A conventional three-electrode cell was employed incorporating a working glass carbon RRDE (PINE), an Ag/AgCl, KCl (Saturated) electrode as reference electrode, and a Pt electrode as counter electrode. All potentials were measured and reported vs. the Ag/AgCl, KCl (Saturated) reference electrode. The experiments were carried out in O2 saturated 0.1 M KOH solution for the oxygen reduction reaction at the ambient temperature after purging O2 or N2 gas for 15 min. The potential range was cyclically scanned between 1.2 and +0.2 V at a scan rate of 100 mV s1.
2.3. Characterizations 3. Results and discussion SEM micrographs were acquired using JSM-6700F microscope. HRTEM images were conducted on JEM-2010F microscope. Nitrogen physisorption measurements were investigated on ASAP 2010 M+C apparatus. XPS experiments were carried out on AXIS UltraDLD system from Kratos with Al Ka radiation as X-ray source for radiation and the XPS data are analyzed by Thermo Avantage. Elemental analysis measurements were performed on Dalton EA300. 2.4. Electrochemical experiments The procedure of glass carbon rotating ring-disk electrode (disk area = 0.25 cm2, ring area = 0.19 cm2, from PINE) pretreatment and modification were as follows: prior to use, the working electrode
The morphologies of as-prepared HMCGs were investigated by means of scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM). As shown in the SEM images of HMCGAla (Fig. 1a and b), the twisted sheets in size from 200 nm to several micrometers formed flower-like aggregates, possibly because of the shrinkage of the sample during high temperature thermal treatment. It is notable that most sheets maintain the large aspect ratio, which was obviously derived from the MSG templates (Fig. S1a and S1b in Supporting Information). HRTEM images of HMCG (Fig. 1a–d) revealed the existence of the porous carbon layers on the surfaces of graphene sheets. The porous carbon did not exhibit clear lattice fringes, which indicated that they were amorphous.
Fig. 2. Electrocatalytic performance of HMCGs catalysts for ORR. (a) Cyclic voltammetry of different HMCGs. (b) RRDE test of different HMCGs in O2-saturated 0.1 M KOH at a rotation rate of 1600 rpm. (c) Electron transfer numbers of different HMCGs. (d) Current–time (i–t) chronoamperometric response of HMCGAla, HMCGCys and Pt–C electrodes at 0.15 V in O2-saturated 0.1 M KOH at a rotation rate of 1600 rpm.
P. Xu et al. / Journal of Colloid and Interface Science 421 (2014) 160–164
The nitrogen adsorption/desorption isotherms of HMCGs revealed the typical type-IV curves with a H2-type hysteresis loop (Fig. S2a), indicating the existence of a large number of mesopores in HMCGs. The Brunauer–Emmett–Teller (BET) surface area of HMCGs is calculated to be 267, 245, 249, and 281 m2 g1, and the total pore volume was estimated to be 1.01, 0.93, 0.97, and 0.96 cm3 g1 for HMCGAla, HMCGArg, HMCGSer, and HMCGCys, respectively. The pore size distribution (Fig. S2b) derived from the adsorption branch using Barrett–Joyner–Halenda (BJH) model showed the size of mesopores centered at about uniform 7.7 nm, which is in good agreement with the pore size observed from the TEM images (Fig. 1c and d). These data are summarized in Supporting Information (Table S1). The electrocatalytic activity of HMCGs for ORR was assessed by cyclic voltammetry (CV) and rotating ring-disk electrode (RRDE) voltammograms in 0.1 M KOH solutions saturated with O2 at a scan rate of 100 mV s1. Among all the HMCGs, HMCGAla shows well-defined cathodic peak at 0.19 V, which suggests the highest electrochemical activity than other heteroatom doped carbon electrodes (Fig. 2a). Furthermore, the steady-state catalytic current density at HMCGAla is the highest value of 4.7 mA cm2 at a potential range from 0.10 to 0.40 V, but lower than that (4.9 mA cm2) of HMCGCys in the more negative potential, comparable to that (5.1 mA cm2) of Pt–C catalyst (Fig. 2b). In addition, it is noted that the onset potentials of ORR at each HMCG electrodes were measured to be similar value of 0.05 V vs. SHE (Fig. 2a and b), which were comparable with most of the recently reported heteroatom doped carbon materials (N-doped graphene [26], mesoporous graphitic array [24], carbon nanotubes [18,19], N and S dual-doped carbons [36,39]). According to the RRDE voltammograms, the electrons transfer numbers for HMCGs can be analyzed on the basis of the following equation:
163
HMCGs. As calculated from the elemental analysis, the weight content of sulfur in HMCGCys was 0.89%. Chemical surface analysis of HMCGs was performed using X-ray photoelectron spectroscopy (XPS, Fig. S3). From the N1s spectrum of HMCGAla (Fig. 3b), three components of the binding energy were fitted: the peak at 398.4 eV corresponds to pyridine-like nitrogen atoms, the peak at 400.0 eV can be ascribed to pyrrole-like nitrogen atoms, and the peak at 401.3 eV corresponds to protonatedlike nitrogen atoms (Table S1) [24,26]. Since HMCGs obtained from different amide acids have almost identical porosity (Table S1), the significantly enhanced performance including electrocatalytic activity, selectivity and durability of HMCGAla must be attributed to the increased proportions of graphite-like nitrogen atoms (Fig. 3a). This conclusion is further supported by both experimental results [40] and computational studies [41,42]. Remarkably, sulfur atoms in HMCGCys formed thiophene-like structures with neighboring carbon atoms, as shown in the high-resolution S2p spectrum (Fig. 3c). This explained the comparable catalytic performance of HMCGCys to HMCGArg, although the former only
n ¼ 4ID =ðID þ IR =NÞ in which n, ID, IR and N represent electrons transfer numbers, disk current, Pt ring current and collection efficiency (N = 0.37), respectively. The calculated results (Table S1) indicated the electron transfer numbers were 3.67–3.89 for HMCGAla, 3.61–3.87 for HMCGArg, 3.51–3.81 for HMCGSer, and 3.41–3.75 for HMCGCys, in the potential range from 0 to 0.60 V (Fig. 2c), indicating that the catalytic process of HMCGs is dominated by the four electron pathway. Since durability is one of the major concerns in current fuel-cell technology, the stability of HMCGAla was further tested at a constant voltage of 0.15 V for 20,000 s in an 0.1 m KOH solution saturated with O2 at a rotation rate of 1600 rpm. After a fast decrease of 10% within the first 2000 s, the current–time (i–t) chronoamperometric response of HMCGAla showed a very slow attenuation, and a high relative current of 84% still persisted after 20,000 s (Fig. 2d). In contrast, Pt–C was found to be a decrease with a current loss of about 30% during the same test period. And the durability (77%) of HMCGCys is a little lower than that of HMCGAla but superior to that of Pt–C catalyst. On the basis of the facts described above, it is reasonable to conclude that HMCGAla has the best performance for catalyzing ORR among all of these HMCG catalysts. Through elemental analysis, the relative composition (wt%) of C, N, H and S was analyzed (Table S1). It is shown that HMCGs are mainly formulated of carbon and small amounts of nitrogen. And trace amount of sulfur could also be found for HMCGCys. It is clearly revealed that the HMGCs derived from four amide acids have the weight content of nitrogen in the following order: HMCGArg (4.97%) > HMCGAla (4.80%) > HMCGSer (4.46%) > HMCGCys (2.97%), which is in agreement with the composition of the carbon precursors. Moreover, it should be noted that cysteine also contain thiol groups, which enable us to obtain nitrogen and sulfur dual-doped
Fig. 3. The content of three types of nitrogen in HMCGs (a), the X-ray photoelectron spectroscopy (XPS) of N2p of all HMGCs (a) and S2p of HMCGCys (b). The signals in (b) fit into three energy components corresponding to pyridine-like N ( ), pyrrole-like N ( ), and graphite-like N ( ); and the two peaks in (c) originate from the 2p 3/2 ( ) and 2p 1/2 splitting ( ) of the S2p spin orbital.
164
P. Xu et al. / Journal of Colloid and Interface Science 421 (2014) 160–164
have less than 1.5% of nitrogen atoms. The excellent catalytic behavior of HMCGCys could be attributed to the proposed synergistic effect caused by dual heteroatom (N and S) doping [36]. 4. Conclusions In summary, we have developed the fabrication of heteroatom doped mesoporous carbon/graphene (HMCG) nanosheets by using cheap and non-toxic amino acids as precursors. In these 2D materials, graphene was sandwiched in two heteroatom-doped (N and S) mesoporous carbon layers. As metal-free ORR catalysts, these HMCGs exhibited excellent electrocatalytic activity, high selectivity, and outstanding long-term stability. Therefore, it can be concluded that amino acids are ideal candidates for the synthesis of heteroatom-doped carbon atoms. These catalysts are attractive for the applications as a low cost but high efficient catalyst in practical fuel cells. Acknowledgments This work was supported by 973 Program (2013CB328804), the National Natural Science Foundation of China (21343002, 61235007, 21102091 and 21372155), the Program for Professor of Special Appointment (Eastern Scholar), the Scientific Research Foundation for Returned Scholars from Ministry of Education of China. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2014.02.001. References [1] M.Z. Jacobson, W.G. Colella, D.M. Golden, Science 2005 (1901) 308. [2] H.A. Gasteiger, S.S. Kocha, B. Sompalli, F.T. Wagner, Appl. Catal. B 56 (2005) 9. [3] J. Zhang, H. Wang, D.P. Wilkinson, D. Song, J. Shen, Z.-S. Liu, J. Power Sources 147 (2005) 58. [4] B.C.H. Steele, A. Heinzel, Nature 414 (2001) 345. [5] X. Yu, S. Ye, J. Power Sources 172 (2007) 145.
[6] Z. Chen, M. Waje, W. Li, Y. Yan, Angew. Chem. Int. Ed. 46 (2007) 4060. [7] B. Lim, M. Jiang, P.H. Camargo, E.C. Cho, J. Tao, X. Lu, Y. Zhu, Y. Xia, Science 324 (2009) 1302. [8] C.H. Choi, S.H. Park, S.I. Woo, Appl. Catal. B 119–120 (2012) 123. [9] S. Kattel, G. Wang, J. Mater. Chem. A (2013). [10] R. Liu, C. von Malotki, L. Arnold, N. Koshino, H. Higashimura, M. Baumgarten, K. Mullen, J. Am. Chem. Soc. 133 (2011) 10372. [11] M. Lefevre, E. Proietti, F. Jaouen, J.P. Dodelet, Science 324 (2009) 71. [12] R. Bashyam, P. Zelenay, Nature 443 (2006) 63. [13] Z. Yang, Z. Yao, G. Li, G. Fang, H. Nie, Z. Liu, X. Zhou, Xa Chen, S. Huang, ACS Nano 6 (2012) 205. [14] Z. Yao, H. Nie, Z. Yang, X. Zhou, Z. Liu, S. Huang, Chem. Commun. 1027 (2012) 48. [15] L. Yang, S. Jiang, Y. Zhao, L. Zhu, S. Chen, X. Wang, Q. Wu, J. Ma, Y. Ma, Z. Hu, Angew. Chem. Int. Ed. 50 (2011) 7132. [16] Z.W. Liu, F. Peng, H.J. Wang, H. Yu, W.X. Zheng, J. Yang, Angew. Chem. Int. Ed. 50 (2011) 3257. [17] Z. Jin, H. Nie, Z. Yang, J. Zhang, Z. Liu, X. Xu, S. Huang, Nanoscale 4 (2012) 6455. [18] P. Chen, T.Y. Xiao, Y.H. Qian, S.S. Li, S.H. Yu, Adv. Mater. 25 (2013) 3192. [19] S. Wang, D. Yu, L. Dai, J. Am. Chem. Soc. 133 (2011) 5182. [20] K. Gong, F. Du, Z. Xia, M. Durstock, L. Dai, Science 323 (2009) 760. [21] W. Xiong, F. Du, Y. Liu, J. Albert Perez, M. Supp, T.S. Ramakrishnan, L. Dai, L. Jiang, J. Am. Chem. Soc. 132 (2010) 15839. [22] T.C. Nagaiah, S. Kundu, M. Bron, M. Muhler, W. Schuhmann, Electrochem. Commun. 12 (2010) 338. [23] Y. Tang, B.L. Allen, D.R. Kauffman, A. Star, J. Am. Chem. Soc. 131 (2009) 13200. [24] R. Liu, D. Wu, X. Feng, K. Mullen, Angew. Chem. Int. Ed. 49 (2010) 2565. [25] W. Yang, T.-P. Fellinger, M. Antonietti, J. Am. Chem. Soc. 133 (2011) 206. [26] S. Yang, X. Feng, X. Wang, K. Mullen, Angew. Chem. Int. Ed. 50 (2011) 5339. [27] Y. Sun, C. Li, G. Shi, J. Mater. Chem. 22 (2012) 12810. [28] L. Qu, Y. Liu, J.-B. Baek, L. Dai, ACS Nano 4 (2010) 1321. [29] Y. Zhang, K. Fugane, T. Mori, L. Niu, J. Ye, J. Mater. Chem. 22 (2012) 6575. [30] Z.S. Wu, S. Yang, Y. Sun, K. Parvez, X. Feng, K. Müllen, J. Am. Chem. Soc. 134 (2012) 9082. [31] Z. Yang, H. Nie, Xa Chen, X. Chen, S. Huang, J. Power Sources 236 (2013) 238. [32] R. Silva, D. Voiry, M. Chhowalla, T. Asefa, J. Am. Chem. Soc. 135 (2013) 7823. [33] A.K. Geim, Science 324 (2009) 1530. [34] A. Fasolino, J.H. Los, M.I. Katsnelson, Nat. Mater. 6 (2007) 858. [35] A.A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, C.N. Lau, Nano Lett. 8 (2008) 902. [36] J. Liang, Y. Jiao, M. Jaroniec, S.Z. Qiao, Angew. Chem. Int. Ed. 51 (2012) 11496. [37] W.S. Hummers Jr., R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339. [38] S. Yang, X. Feng, L. Wang, K. Tang, J. Maier, K. Müllen, Angew. Chem. Int. Ed. 49 (2010) 4795. [39] C.H. Choi, S.H. Park, S.I. Woo, Green Chem. 13 (2011) 406. [40] X. Wang, J.S. Lee, Q. Zhu, J. Liu, Y. Wang, S. Dai, Chem. Mater. 22 (2010) 2178. [41] T. Ikeda, M. Boero, S.-F. Huang, K. Terakura, M. Oshima, J.-I. Ozaki, J. Phys. Chem. C 112 (2008) 14706. [42] R.A. Sidik, A.B. Anderson, J. Phys. Chem. C 110 (2006) 1787.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Heteroatom doped mesoporous carbon/graphene nanosheets as highly efficient electrocatalysts for oxygen reduction Peimin Xu a, Dongqing Wu b, Li Wan a, Pengfei Hu c, Ruili Liu a,⇑ a
Department of Chemical Engineering, School of Environment and Chemical Engineering, Shanghai University, Shangda Road 99, 200444 Shanghai, PR China School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Dongchuan Road 800, 200240 Shanghai, PR China c Lab of Microstructure, Shanghai University, Shangda Road 99, 200444 Shanghai, PR China b
a r t i c l e
i n f o
Article history: Received 24 October 2013 Accepted 2 February 2014 Available online 8 February 2014 Keywords: Heteroatom doped carbon Metal-free catalysts Oxygen reduction Fuel cell
a b s t r a c t The high cost of platinum (Pt) based catalysts for oxygen reduction reaction (ORR) has restricted the widespread commercialization of fuel cells. Heteroatom (N, B, P, S or Se) doped carbon materials have been regarded as the promising metal-free catalysts for replacing Pt based catalysts owing to their high efficiencies, good stability and relative low cost. In this work, we present a cost-effective synthesis approach for heteroatom (N and S) doped mesoporous carbon/graphene (HMCG) nanosheets by using nano-casting technology with mesoporous silica/graphene nanosheets (MSG) as hard templates, and four different amino acids (alanine, serine, arginine and cystine) as heteroatom (N, S) and carbon precursors. The resulting catalysts exhibited excellent electrocatalytic activity for ORR in alkaline media. In particular, HMCGAla with alanine as precursors showed the highest electron transfer numbers and durability. These results indicated the attractive potential of HMCGs as metal-free catalysts in practical fuel cells. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Fuel cell, which directly converts chemical energy into electricity, is regarded as one of the most attractive energy conversion systems because of its high efficiency, low pollution and fuel flexibility [1–3]. Generally, the cell performance of fuel cells is mainly determined by the oxygen-reduction reaction (ORR) happened in their cathodes [4]. Platinum (Pt) based materials have been investigated as the most active ORR catalysts for decades. However, the commercialization of fuel cells is still retarded by the poor durability and high cost of Pt catalysts [5–7]. Therefore, the development of efficient ORR electrocatalysts derived from non-precious metal (e.g. Fe, Co or Ni) [8–12] or heteroatom (e.g. N, B, P, S or Se) doped carbon materials [13–17] has become an urgent research topic for fuel cells. Recently, heteroatom doped carbon materials (e.g. carbon nanotubes [18–22], nanotube cups [23], ordered mesoporous graphitic arrays [24,25] and graphene-based nanosheets [26–30]) have shown excellent electrocatalytic activities as well as low costs, good durability, and environmental friendliness, which open the way to a new class of metal-free catalysts for ORR. Generally, the types of heteroatoms and their weight content ratio decide the intrinsic catalytic behavior of the
⇑ Corresponding author. Fax: +86 21 5633 3063. E-mail address: [email protected] (R. Liu). http://dx.doi.org/10.1016/j.jcis.2014.02.001 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
heteroatom doped carbon materials [31]. On the other hand, the mesostructured properties of these catalysts such as porosity and surface area also hold the key for the catalytic efficiency of them [32]. Based on these considerations, heteroatom doped carbons with porous structures are regarded as excellent candidates for carbon based ORR catalysts [24,26,27]. In the last few years, graphene, a two-dimensional (2D) array of sp2 carbon atoms, has received intensive attentions due to its unique flat morphology, intriguing physical and chemical properties such as high conductivity [33], mechanical strength [34] and thermal stability [35]. It is thus believed that the combination of graphene sheets and porous heteroatom doped carbons will lead to high performance metal-free ORR catalysts. For instance, Yang and the coworkers [26] prepared graphene-based carbon nitride (G-CN) nanosheets by using graphene-based mesoporous silica nanosheets as templates, and ethylenediamine/carbon tetrachloride as precursor. Nevertheless, the toxicity of ethylenediamine and carbon tetrachloride limited the large scale production of the G-CN nanosheets as ORR catalysts. Shi’s group [27] reported sandwichstructure nitrogen-doped mesoporous carbon modified graphene sheets for ORR via a soft template route in which triblock copolymer (Pluronic F127) was used as the template for guiding the growth of nanoporous structures of phenolmelamine–formaldehyde-pre-polymer (PMF) as carbon source on graphene surface. With colloidal silica nanoparticles as hard templates, Qiao’s group obtained sulfur and nitrogen dual-doped mesoporous graphene
P. Xu et al. / Journal of Colloid and Interface Science 421 (2014) 160–164
161
nanosheets (MSG) were applied as the hard template for the growth of heteroatom doped mesoporous carbon layers on graphene sheets. In contrast to other carbon sources, amino acids are abundant, inexpensive, biocompatible and eco-friendly. More importantly, amino acids containing multiple heteroatoms such as nitrogen and sulfur enable the formation of dual-doped carbon with different heteroatoms. As the results, these HMCG catalysts have large specific surface areas (245–281 m2 g1), high onset potential (0.05 V vs. SHE) and excellent long term stability for ORR, which were superior to those observed for the commercially available Pt–C catalyst. A comparison of NMGCs derived from different carbon precursors suggests that both nitrogen (especially graphitelike nitrogen atoms) and sulfur atoms account for the excellent electrochemical performance in the ORR.
2. Experimental Scheme 1. The synthesis procedure of HMCGs: (a) self-assembly of mesoporous TEOS on GO, (b) removal of CTAB and (c) nanocasting approach of each amino acid.
2.1. Preparation of mesoporous silica/graphene (MSG) composites
materials by using melamine and benzyl disulfide (BDS) as N and S precursors [36]. Herein, we report an environmentally friendly approach toward heteroatom (N and S) doped mesoporous carbon/graphene (HMCG) nanosheets in which graphene sheets were sandwiched by mesoporous carbon layers containing heteroatom atoms (Scheme 1). In our method, amid acids were used as the source for carbon and heteroatoms and mesoporous silica/graphene
Graphene oxide (GO) was synthesized from natural graphite powders by a modified Hummers method [37]. MSGs were prepared via a soft template route with the assistance of electrostatic interaction between negatively charged GO and cationic surfactant [38]. In a typical experiment, 30 mg GO was firstly suspended in an aqueous solution containing 1 g CTAB and 40 mg NaOH, and then ultrasonically treated for 3 h. After magnetic stirring for 2 h at 40 °C, 1 mL TEOS was slowly added to the above mixture. After reaction for 12 h, MSGs were obtained by washing the precipitants
Fig. 1. SEM (a and b) and TEM (c and d) images of HMCGAla.
162
P. Xu et al. / Journal of Colloid and Interface Science 421 (2014) 160–164
with warm ethanol, separation, drying and the subsequent pyrolysis at 900 °C for 2 h in nitrogen. 2.2. Preparation of heteroatom doped mesoporous carbon/graphene nanosheets HMCGs were prepared by using MSGs as a hard template via a nanocasting approach. Under magnetic stirring at 40 °C, 0.5 g template was repeatedly impregnated with 50 mL aqueous solution of amino acids. In this work, four kinds of amino acids as alanine (C3H7NO2), arginine (C6H14N4O2), serine (C3H7NO3) and cysteine (C6H12N2O4S2) were utilized as the precursors. The weight of each amino acid was fixed to the pore volume of the MSGs (1.0 cm3 g1). After 6 h, the mixtures of MSGs and corresponding amino acids was placed in a drying oven for 6 h at 160 °C. Later on, the solids were calcined at 900 °C for 2 h in nitrogen. Subsequent etching of the resulting powders in a HF solution led to target HMCGs. According to their precursor, the obtained HMCGs were named as HMCGAla, HMCGArg, HMCGSer, and HMCGCys, respectively.
was polished mechanically with 0.5 lm diamond down to 0.05 lm alumina slurry to obtain a mirror-like surface and then washed with Milli-Q water and acetone. 1 mg HMCGs was dissolved in 1 mL solvent mixture of Nafion (5%) and EtOH (V:V ratio = 1:9) by sonication. For comparison, a commercially available catalyst of 20 wt% Pt supported on black carbon (from Alfa Aesar) was used and 1 mg mL1 Pt–C suspension was also prepared as the same procedure described above. 50 lL HMCGs or Pt–C suspension was pipetted on the glassy carbon electrode surface. The electrode was allowed to dry at RT before measurement. This led to a catalyst (HMCGs or Pt–C) loading 200 lg/cm2. A conventional three-electrode cell was employed incorporating a working glass carbon RRDE (PINE), an Ag/AgCl, KCl (Saturated) electrode as reference electrode, and a Pt electrode as counter electrode. All potentials were measured and reported vs. the Ag/AgCl, KCl (Saturated) reference electrode. The experiments were carried out in O2 saturated 0.1 M KOH solution for the oxygen reduction reaction at the ambient temperature after purging O2 or N2 gas for 15 min. The potential range was cyclically scanned between 1.2 and +0.2 V at a scan rate of 100 mV s1.
2.3. Characterizations 3. Results and discussion SEM micrographs were acquired using JSM-6700F microscope. HRTEM images were conducted on JEM-2010F microscope. Nitrogen physisorption measurements were investigated on ASAP 2010 M+C apparatus. XPS experiments were carried out on AXIS UltraDLD system from Kratos with Al Ka radiation as X-ray source for radiation and the XPS data are analyzed by Thermo Avantage. Elemental analysis measurements were performed on Dalton EA300. 2.4. Electrochemical experiments The procedure of glass carbon rotating ring-disk electrode (disk area = 0.25 cm2, ring area = 0.19 cm2, from PINE) pretreatment and modification were as follows: prior to use, the working electrode
The morphologies of as-prepared HMCGs were investigated by means of scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM). As shown in the SEM images of HMCGAla (Fig. 1a and b), the twisted sheets in size from 200 nm to several micrometers formed flower-like aggregates, possibly because of the shrinkage of the sample during high temperature thermal treatment. It is notable that most sheets maintain the large aspect ratio, which was obviously derived from the MSG templates (Fig. S1a and S1b in Supporting Information). HRTEM images of HMCG (Fig. 1a–d) revealed the existence of the porous carbon layers on the surfaces of graphene sheets. The porous carbon did not exhibit clear lattice fringes, which indicated that they were amorphous.
Fig. 2. Electrocatalytic performance of HMCGs catalysts for ORR. (a) Cyclic voltammetry of different HMCGs. (b) RRDE test of different HMCGs in O2-saturated 0.1 M KOH at a rotation rate of 1600 rpm. (c) Electron transfer numbers of different HMCGs. (d) Current–time (i–t) chronoamperometric response of HMCGAla, HMCGCys and Pt–C electrodes at 0.15 V in O2-saturated 0.1 M KOH at a rotation rate of 1600 rpm.
P. Xu et al. / Journal of Colloid and Interface Science 421 (2014) 160–164
The nitrogen adsorption/desorption isotherms of HMCGs revealed the typical type-IV curves with a H2-type hysteresis loop (Fig. S2a), indicating the existence of a large number of mesopores in HMCGs. The Brunauer–Emmett–Teller (BET) surface area of HMCGs is calculated to be 267, 245, 249, and 281 m2 g1, and the total pore volume was estimated to be 1.01, 0.93, 0.97, and 0.96 cm3 g1 for HMCGAla, HMCGArg, HMCGSer, and HMCGCys, respectively. The pore size distribution (Fig. S2b) derived from the adsorption branch using Barrett–Joyner–Halenda (BJH) model showed the size of mesopores centered at about uniform 7.7 nm, which is in good agreement with the pore size observed from the TEM images (Fig. 1c and d). These data are summarized in Supporting Information (Table S1). The electrocatalytic activity of HMCGs for ORR was assessed by cyclic voltammetry (CV) and rotating ring-disk electrode (RRDE) voltammograms in 0.1 M KOH solutions saturated with O2 at a scan rate of 100 mV s1. Among all the HMCGs, HMCGAla shows well-defined cathodic peak at 0.19 V, which suggests the highest electrochemical activity than other heteroatom doped carbon electrodes (Fig. 2a). Furthermore, the steady-state catalytic current density at HMCGAla is the highest value of 4.7 mA cm2 at a potential range from 0.10 to 0.40 V, but lower than that (4.9 mA cm2) of HMCGCys in the more negative potential, comparable to that (5.1 mA cm2) of Pt–C catalyst (Fig. 2b). In addition, it is noted that the onset potentials of ORR at each HMCG electrodes were measured to be similar value of 0.05 V vs. SHE (Fig. 2a and b), which were comparable with most of the recently reported heteroatom doped carbon materials (N-doped graphene [26], mesoporous graphitic array [24], carbon nanotubes [18,19], N and S dual-doped carbons [36,39]). According to the RRDE voltammograms, the electrons transfer numbers for HMCGs can be analyzed on the basis of the following equation:
163
HMCGs. As calculated from the elemental analysis, the weight content of sulfur in HMCGCys was 0.89%. Chemical surface analysis of HMCGs was performed using X-ray photoelectron spectroscopy (XPS, Fig. S3). From the N1s spectrum of HMCGAla (Fig. 3b), three components of the binding energy were fitted: the peak at 398.4 eV corresponds to pyridine-like nitrogen atoms, the peak at 400.0 eV can be ascribed to pyrrole-like nitrogen atoms, and the peak at 401.3 eV corresponds to protonatedlike nitrogen atoms (Table S1) [24,26]. Since HMCGs obtained from different amide acids have almost identical porosity (Table S1), the significantly enhanced performance including electrocatalytic activity, selectivity and durability of HMCGAla must be attributed to the increased proportions of graphite-like nitrogen atoms (Fig. 3a). This conclusion is further supported by both experimental results [40] and computational studies [41,42]. Remarkably, sulfur atoms in HMCGCys formed thiophene-like structures with neighboring carbon atoms, as shown in the high-resolution S2p spectrum (Fig. 3c). This explained the comparable catalytic performance of HMCGCys to HMCGArg, although the former only
n ¼ 4ID =ðID þ IR =NÞ in which n, ID, IR and N represent electrons transfer numbers, disk current, Pt ring current and collection efficiency (N = 0.37), respectively. The calculated results (Table S1) indicated the electron transfer numbers were 3.67–3.89 for HMCGAla, 3.61–3.87 for HMCGArg, 3.51–3.81 for HMCGSer, and 3.41–3.75 for HMCGCys, in the potential range from 0 to 0.60 V (Fig. 2c), indicating that the catalytic process of HMCGs is dominated by the four electron pathway. Since durability is one of the major concerns in current fuel-cell technology, the stability of HMCGAla was further tested at a constant voltage of 0.15 V for 20,000 s in an 0.1 m KOH solution saturated with O2 at a rotation rate of 1600 rpm. After a fast decrease of 10% within the first 2000 s, the current–time (i–t) chronoamperometric response of HMCGAla showed a very slow attenuation, and a high relative current of 84% still persisted after 20,000 s (Fig. 2d). In contrast, Pt–C was found to be a decrease with a current loss of about 30% during the same test period. And the durability (77%) of HMCGCys is a little lower than that of HMCGAla but superior to that of Pt–C catalyst. On the basis of the facts described above, it is reasonable to conclude that HMCGAla has the best performance for catalyzing ORR among all of these HMCG catalysts. Through elemental analysis, the relative composition (wt%) of C, N, H and S was analyzed (Table S1). It is shown that HMCGs are mainly formulated of carbon and small amounts of nitrogen. And trace amount of sulfur could also be found for HMCGCys. It is clearly revealed that the HMGCs derived from four amide acids have the weight content of nitrogen in the following order: HMCGArg (4.97%) > HMCGAla (4.80%) > HMCGSer (4.46%) > HMCGCys (2.97%), which is in agreement with the composition of the carbon precursors. Moreover, it should be noted that cysteine also contain thiol groups, which enable us to obtain nitrogen and sulfur dual-doped
Fig. 3. The content of three types of nitrogen in HMCGs (a), the X-ray photoelectron spectroscopy (XPS) of N2p of all HMGCs (a) and S2p of HMCGCys (b). The signals in (b) fit into three energy components corresponding to pyridine-like N ( ), pyrrole-like N ( ), and graphite-like N ( ); and the two peaks in (c) originate from the 2p 3/2 ( ) and 2p 1/2 splitting ( ) of the S2p spin orbital.
164
P. Xu et al. / Journal of Colloid and Interface Science 421 (2014) 160–164
have less than 1.5% of nitrogen atoms. The excellent catalytic behavior of HMCGCys could be attributed to the proposed synergistic effect caused by dual heteroatom (N and S) doping [36]. 4. Conclusions In summary, we have developed the fabrication of heteroatom doped mesoporous carbon/graphene (HMCG) nanosheets by using cheap and non-toxic amino acids as precursors. In these 2D materials, graphene was sandwiched in two heteroatom-doped (N and S) mesoporous carbon layers. As metal-free ORR catalysts, these HMCGs exhibited excellent electrocatalytic activity, high selectivity, and outstanding long-term stability. Therefore, it can be concluded that amino acids are ideal candidates for the synthesis of heteroatom-doped carbon atoms. These catalysts are attractive for the applications as a low cost but high efficient catalyst in practical fuel cells. Acknowledgments This work was supported by 973 Program (2013CB328804), the National Natural Science Foundation of China (21343002, 61235007, 21102091 and 21372155), the Program for Professor of Special Appointment (Eastern Scholar), the Scientific Research Foundation for Returned Scholars from Ministry of Education of China. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2014.02.001. References [1] M.Z. Jacobson, W.G. Colella, D.M. Golden, Science 2005 (1901) 308. [2] H.A. Gasteiger, S.S. Kocha, B. Sompalli, F.T. Wagner, Appl. Catal. B 56 (2005) 9. [3] J. Zhang, H. Wang, D.P. Wilkinson, D. Song, J. Shen, Z.-S. Liu, J. Power Sources 147 (2005) 58. [4] B.C.H. Steele, A. Heinzel, Nature 414 (2001) 345. [5] X. Yu, S. Ye, J. Power Sources 172 (2007) 145.
[6] Z. Chen, M. Waje, W. Li, Y. Yan, Angew. Chem. Int. Ed. 46 (2007) 4060. [7] B. Lim, M. Jiang, P.H. Camargo, E.C. Cho, J. Tao, X. Lu, Y. Zhu, Y. Xia, Science 324 (2009) 1302. [8] C.H. Choi, S.H. Park, S.I. Woo, Appl. Catal. B 119–120 (2012) 123. [9] S. Kattel, G. Wang, J. Mater. Chem. A (2013). [10] R. Liu, C. von Malotki, L. Arnold, N. Koshino, H. Higashimura, M. Baumgarten, K. Mullen, J. Am. Chem. Soc. 133 (2011) 10372. [11] M. Lefevre, E. Proietti, F. Jaouen, J.P. Dodelet, Science 324 (2009) 71. [12] R. Bashyam, P. Zelenay, Nature 443 (2006) 63. [13] Z. Yang, Z. Yao, G. Li, G. Fang, H. Nie, Z. Liu, X. Zhou, Xa Chen, S. Huang, ACS Nano 6 (2012) 205. [14] Z. Yao, H. Nie, Z. Yang, X. Zhou, Z. Liu, S. Huang, Chem. Commun. 1027 (2012) 48. [15] L. Yang, S. Jiang, Y. Zhao, L. Zhu, S. Chen, X. Wang, Q. Wu, J. Ma, Y. Ma, Z. Hu, Angew. Chem. Int. Ed. 50 (2011) 7132. [16] Z.W. Liu, F. Peng, H.J. Wang, H. Yu, W.X. Zheng, J. Yang, Angew. Chem. Int. Ed. 50 (2011) 3257. [17] Z. Jin, H. Nie, Z. Yang, J. Zhang, Z. Liu, X. Xu, S. Huang, Nanoscale 4 (2012) 6455. [18] P. Chen, T.Y. Xiao, Y.H. Qian, S.S. Li, S.H. Yu, Adv. Mater. 25 (2013) 3192. [19] S. Wang, D. Yu, L. Dai, J. Am. Chem. Soc. 133 (2011) 5182. [20] K. Gong, F. Du, Z. Xia, M. Durstock, L. Dai, Science 323 (2009) 760. [21] W. Xiong, F. Du, Y. Liu, J. Albert Perez, M. Supp, T.S. Ramakrishnan, L. Dai, L. Jiang, J. Am. Chem. Soc. 132 (2010) 15839. [22] T.C. Nagaiah, S. Kundu, M. Bron, M. Muhler, W. Schuhmann, Electrochem. Commun. 12 (2010) 338. [23] Y. Tang, B.L. Allen, D.R. Kauffman, A. Star, J. Am. Chem. Soc. 131 (2009) 13200. [24] R. Liu, D. Wu, X. Feng, K. Mullen, Angew. Chem. Int. Ed. 49 (2010) 2565. [25] W. Yang, T.-P. Fellinger, M. Antonietti, J. Am. Chem. Soc. 133 (2011) 206. [26] S. Yang, X. Feng, X. Wang, K. Mullen, Angew. Chem. Int. Ed. 50 (2011) 5339. [27] Y. Sun, C. Li, G. Shi, J. Mater. Chem. 22 (2012) 12810. [28] L. Qu, Y. Liu, J.-B. Baek, L. Dai, ACS Nano 4 (2010) 1321. [29] Y. Zhang, K. Fugane, T. Mori, L. Niu, J. Ye, J. Mater. Chem. 22 (2012) 6575. [30] Z.S. Wu, S. Yang, Y. Sun, K. Parvez, X. Feng, K. Müllen, J. Am. Chem. Soc. 134 (2012) 9082. [31] Z. Yang, H. Nie, Xa Chen, X. Chen, S. Huang, J. Power Sources 236 (2013) 238. [32] R. Silva, D. Voiry, M. Chhowalla, T. Asefa, J. Am. Chem. Soc. 135 (2013) 7823. [33] A.K. Geim, Science 324 (2009) 1530. [34] A. Fasolino, J.H. Los, M.I. Katsnelson, Nat. Mater. 6 (2007) 858. [35] A.A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, C.N. Lau, Nano Lett. 8 (2008) 902. [36] J. Liang, Y. Jiao, M. Jaroniec, S.Z. Qiao, Angew. Chem. Int. Ed. 51 (2012) 11496. [37] W.S. Hummers Jr., R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339. [38] S. Yang, X. Feng, L. Wang, K. Tang, J. Maier, K. Müllen, Angew. Chem. Int. Ed. 49 (2010) 4795. [39] C.H. Choi, S.H. Park, S.I. Woo, Green Chem. 13 (2011) 406. [40] X. Wang, J.S. Lee, Q. Zhu, J. Liu, Y. Wang, S. Dai, Chem. Mater. 22 (2010) 2178. [41] T. Ikeda, M. Boero, S.-F. Huang, K. Terakura, M. Oshima, J.-I. Ozaki, J. Phys. Chem. C 112 (2008) 14706. [42] R.A. Sidik, A.B. Anderson, J. Phys. Chem. C 110 (2006) 1787.
