Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method

ARTICLE IN PRESS

JID: JECHEM

[m5G;April 12, 2017;0:54]

Journal of Energy Chemistry xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Journal of Energy Chemistry journal homepage: www.elsevier.com/locate/jechem

http://www.journals.elsevier.com/ journal-of-energy-chemistry/

Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method Baobing Huang, Lu Peng, Fangfang Yang, Yuchuan Liu, Zailai Xie∗

Q2

College of Chemistry, Fuzhou University, Qishan Campus, Fuzhou 350116, Fujian, China

a r t i c l e

i n f o

Article history: Received 18 January 2017 Accepted 3 March 2017 Available online xxx

Q3

Keywords: Carbon nanotubes Hydrothermal carbon ORR activity

a b s t r a c t Nitrogen doped carbons are an important family of materials with ideal activity for oxygen reduction reaction (ORR). It is always interesting to search functional carbons with high heteroatom contents and desirable structure for ORR. Within this study, the surface modification of carbon nanotubes (CNTs) via hydrothermal carbonization (HTC) technique in the presence of glucose and urea was reported, where the surface of CNTs is successfully coated by nitrogen containing hydrothermal carbon layers. The resulting composite combines both advantages of the outstanding electrical conductivity of CNTs and the effective ORR active sites provided by doped nitrogen in the HTC carbon layers. By controlling the ratio of glucose and urea, the nitrogen contents coated on the surface of CNTs can reach up to 1.7 wt%. The resulting materials show outstanding electrochemical activity towards ORR in alkaline electrolyte, making it one of the valuable metal-free electrode materials and a competent alternative to the state-of-the-art Pt/C catalyst. © 2017 Science Press and Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier B.V. and Science Press. All rights reserved.

1

1. Introduction

2

Oxygen reduction reaction (ORR) is a kind of vital reaction involved in fuel cells, metal-air batteries as well as other energy conversion systems [1–6]. Pt-based catalysts is viewed as the most efficient, however, the extremely high cost, shortage of resource as well as declining activity are detrimental to extensively applications [7–9]. Thus searching of low-cost and highly active nonprecious metal catalysts for ORR is urgently needed [10–12]. Carbon materials, which possess various merits of high conductivity, excellent mechanical and chemical stability and easily surface-functionalization, have been applied to many fields [13– 18]. To be specific, they not only can serve as a kind of supports to homogeneously disperse metal nanoparticles, accordingly enhancing the catalytical activity [19–22]; but also act alone as the active components applied to heterogeneous catalysis [13–18, 23–26]. Since Dai group [18] found nitrogen-doped carbon nanotubes (NCNTs) displayed outstanding performance toward ORR in 2009, heteroatoms-doped carbon materials as metal-free ORR electrocatalysts have been studied extensively, which are regarded as one kind of the most potential alternatives to Pt-based catalysts [27–33]. Regarding of catalytically active sites, it is generally believed that the heteroatoms (especially nitrogen) play the vital role

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

∗

Corresponding author. E-mail address: [email protected] (Z. Xie).

of active sites [34–38]. Theoretical and computational chemistry [39] have also revealed the interaction between oxygen molecule and CNTs were indeed strengthened after doping nitrogen into nanocarbons [39]. Moreover, Yang group [40] used first principles calculation to investigate the cracking process of oxygen molecule adsorbed on CNTs. They found that the doped nitrogen significantly reduced the reaction energy barrier, indicative of incorporating nitrogen into carbon materials not only enhancing the adsorption of oxygen molecule but also facilitating its cracking. In terms of nitrogen effect, both of pyridinic N and graphitic N are crucial for the directly four-electrons pathway of ORR [38]. It was said that pyridinic N is beneficial to reduce the onset potential of ORR, while graphitic N helps to develop larger diffusion-limited current density [35, 41]. Besides, the total content of nitrogen, the specific surface areas as well as conductivity are other key influencing factors for ORR [42–47]. As a result, the introduction of nitrogen within CNTs becomes a popular way to obtain active CNTs electroctalysts, for instance, the in situ growth NCNTs by chemical vapor deposition method using nitrogen containing precursors was developed to produce ORR catalysts. However, this kind of NCNTs possess low nitrogen contents on the surface, because a certain number of N active sites locate in the inner walls of NCNTs which are hardly accessible and consequently contributed to the catalytic property scarcely [48–50]. Therefore, the surface modification of CNTs with nitrogen precursors would be highly desired for an efficient electrocatalysts as all the integrated N atoms present at the surface.

http://dx.doi.org/10.1016/j.jechem.2017.03.016 2095-4956/© 2017 Science Press and Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier B.V. and Science Press. All rights reserved.

Please cite this article as: B. Huang et al., Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.03.016

23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Q4

JID: JECHEM 2

ARTICLE IN PRESS B. Huang et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

71

Hydrothermal carbonization (HTC) technique is a sustainable approach to produce nitrogen-doped carbons [51]. This technique involves the heating of biomass (carbohydrate, starch or lignin) in water in an autoclave at mild temperatures (150–200 °C) and self-generated pressures. Consequently, the soluble biomass can be converted to functional, nanostructured, carbonaceous products [52]. The ease of preparation of the HTC approach provides an elegant addition to the toolbox of carbon material synthesis with the possibility of tuning the material morphology, porosity and surface functionality. Herein, we report a novel and facile route to synthesize surface functional CNTs covered with nitrogen containing hydrothermal carbon (HTC) layer onto the surface of CNTs. The resulting composite combines both advantages of the outstanding electrical conductivity of CNTs with the effective ORR active sites provided by doped nitrogen in the HTC layer. The mesoporosity of the current materials make greatly contribution to mass transfer and diffusion of reactants and products. By comparing distinct ORR performances through selecting different HTC precursors (all are biomass, lowcost and sustainable), OCNT-Glu-Urea-900 with a higher nitrogen content of 1.7 wt% exhibits the most superior ORR activity close to the commercial 20 wt% Pt/C catalyst whether in the onset potential, halfwave potential, or diffusion-limited current density.

72

2. Experimental

73

2.1. Sample synthesis

74

MWCNT was purchased from Shandong Dazhang Nano Materials Co. 15 g of MWCNT was mixed with 500 mL of nitric acid, stirred and holded at 110 °C under reflux for 4 h. The pretreated MWCNT was then washed with deionized water several times until the residual solution was neutral, and then drained, and dried at 100 °C for 24 °C, which was denoted as OCNT. 0.75 g of glucose was firstly dissolved into 15 mL of deionized water, and then 1 g of OCNT was added and dispersed by stirring for 1 h and ultrasonicating for 20 min; finally the mixtures were sealed into a Teflon autoclave and kept at 180 °C for 6 h. After the same washing, draining and drying as above, the obtained hydrothermal carbon was denoted as OCNT-Glu. The OCNT-Glu-Urea was prepared by the similar process, and the only difference was using 1.5 g glucose and 0.75 g urea as nitrogen source instead of single 0.75 g glucose. Similarly, 1.5 g of glucosamine hydrochloride as both carbon and nitrogen sources took the place of 0.75 g glucose and also changing the holding temperature of hydrothermal carbonization into 20 h, obtaining OCNT-GN. All above hydrothermal carbons of OCNT-Glu, OCNT-Glu-Urea and OCNT-GN were subjected to further carbonize at 900 °C in N2 atmosphere for 4 h to get the final products, which were denoted as OCNT-Glu-900, OCNT-Glu-Urea-900 and OCNTGN-900, respectively. For comparison, OCNT-Urea-900 was additionally prepared.

49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70

75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109

[m5G;April 12, 2017;0:54]

2.2. Characterization Thermoanalytic analyses were performed with a NETSCH STA449F3 thermobalance setup in N2 atmosphere with a constant gas flow of 50 mL/min in a temperature range of 40–900 °C with a heating rate of 10 K/min. Elemental analysis of C, H, and N was performed using a Vario EL III CHNOS elemental analyzer. Scanning electron microscopy (SEM) images were acquired on a FEI Nova NanoSEM 230 with Everhard-Thornley secondary electron and inlens detectors. Transmission electron microscopy (TEM) was performed on a FEI Tecnai microscope operated at 200 kV. Nitrogen sorption isotherms were measured at 77 K on a Quadrachrome adsorption instrument (Quantachrome Instruments). Sample was dried at 150 °C for 7 h prior to nitrogen sorption analysis. X-ray

diffraction (XRD) patterns were recorded in reflection mode (Cu Kα radiation) on a Bruker D8 diffractometer between 2° and 80°. Raman spectra were recorded using a Thermo Scientific DXR Raman Microscope with a 50 magnification and a 532 nm laser. X-ray photoelectron spectroscopy (XPS) was conducted on a Thermo Scientific K-alpha photoelectron spectrometer. 2.3. Electrochemical measurements The electrochemical activity of the as-synthesized materials toward ORR was performed by the IviumStat multichannel electrochemical analyzer in a three-electrode cell, wherein a saturated calomel electrode (SCE) was selected as the reference electrode, a Pt wire as the counter electrode as well as rotating disk electrode (RDE) functioning as the working electrode. For the rotating disk electrode (RDE) experiment, a glassy carbon electrode (GCE, d = 4 mm) was pre-polished and rinsed cleanly. 5 mg of the carbon sample was dispersed in the mixture of 0.35 mL deionized water, 0.7 mL ethanol and 0.08 mL Nafion (5 wt%), and then sonicated for 1 h to get a homogeneous catalyst ink. All measurements were conducted in the O2 -saturated solution. For comparison, the commercial 20 wt% Pt/C catalyst was prepared and measured under the same conditions, and the only difference is halving the loading of catalyst (The loading of carbon catalysts is 0.45 mg/cm2 , and the loading is halved for Pt/C catalyst.) With regard to the rotating ring-disk electrode (RRDE) experiment, the ring-disk electrode with a glassy carbon disk (d = 4 mm) and a Pt ring (5 mm inner diameter, 7 mm outer diameter) was served as the working electrode with the same loading of catalyst as the RDE measurement. The ring potential was kept at 0.5 V (vs SCE). The number of electron transfer is evaluated from the following equations:

110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139

ld lr /N + ld ld /N H2 O(% ) = 200 × lr /N + ld n = 4×

wherein Ir and Id are the ring and disk currents, respectively, and together with N (0.44) is the ring collection efficiency.

140

3. Results and discussion

142

Giving that untreated CNTs present inherent smooth graphitic surfaces which are less efficient as catalysts or support to anchor metal nanoparticles, the focus of this article is to active CNT by HTC deposition method. In the present contribution, we show that the HTC method can be successfully applied to modify the surface of CNTs at mild condition. In a typical synthesis of product, 1 g OCNT was dispersed in 20 mL aqueous solution containing a certain amount of glucose. Then, the mixture was sealed in a Teflon lined autoclave (45 mL volume) and heated to 180 °C for 6 h to produce black powder. Thermogravimetric analysis is widely used to determine the surface functional groups by probing the temperature-dependent decomposition behavior. The relative mass loss is related to the type and amount of functional groups (such as aldehydes, ketones and carboxylic groups) on the surface of CNTs. As shown in Fig. 1, all samples show a gradual weight loss process in the entire temperature range. OCNT presents the least mass loss of ca. 4 wt%, suggesting the presence of small amount functional groups introduced by HNO3 treatment. This is due to the decomposition of various oxygen-containing functional groups and subsequent release gases in the form of CO2 or CO at elevated temperature, which was confirmed in earlier reports. It was acknowledged that carbon produced by HTC method normally possesses ca. 30 wt% oxygen. All CNTs after HTC treatment show more pronounced mass

Please cite this article as: B. Huang et al., Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.03.016

141

143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166

JID: JECHEM

ARTICLE IN PRESS

[m5G;April 12, 2017;0:54]

B. Huang et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

Fig. 1. Mass loss of different CNTs composites in dependence of temperature between 40 and 900 °C measured with the heating rate of 10 °C/min in nitrogen atmosphere.

Fig. 2. XRD diffraction patterns of all carbon materials.

167 168 169 170 171 172 173 174 175 Q5 176 177 178 179 180 181 182 183 184 185

loss, indicating the successful deposition of hydrothermal carbon. Among them, OCNT-Glu-Urea and OCNT-GN display similar as well as the most pronounced mass loss with approximate 15 wt% between 50 and 900 °C, mainly coming from the extra contribution of oxygen and nitrogen-containing functional groups existing in the hydrothermal carbon layer. In contrast, OCNT-Glu presents a lower mass loss with ca. 10 wt%. In terms of the ratio of HTC carbons and CNTs, for OCNT-Glu-900 the ratio of HTC induced carbon and pristine CNTs is 16:100, while they are 27:100 for both OCNT-GN-900 and OCNT-Glu-Urea-900 calculated from TG and oven yield (Fig. 2). X-ray diffraction (XRD) was usually performed to investigate the extent of structural order of carbon materials. All carbon materials after 900 °C annealing show a high-intensity (002) diffraction peak characteristic at 2 theta = 25.7°, coupled with clearly observable (100) and (110) reflections located at 2 theta = 42.8° and 53.3°, respectively, indicating still high degree of graphitization despite the partial amorphous surface of CNT appended to different oxygencontaining functional groups or covered by heteroatoms-doped carbon layers. The result suggests that the surface is fully covered by

3

Fig. 3. SEM and TEM images of the sample OCNT-Glu-Urea-900.

the active sites while the inner continuous CNT walls are well preserved. Regarding of morphology, SEM and TEM images show that the morphology of the presented carbon materials is similar. Fig. 3 shows there is no significant change in morphology of HTC modified CNTs as compared to original OCNT. This is intriguing, because hydrothermal carbons appear spherical carbon particles of diameter approximate 50–100 μm. However, the discrete particles are not observed in the final carbon materials. Alternatively, high resolution TEM image indicates that HTC carbons present along the outer surface of CNTs. Furthermore, the typical sample of OCNTGlu-Urea-900 was investigated to understand the surface composition with X-ray elemental mapping technique. X-ray maps indeed confirm that the nitrogen is homogeneously distributed on the surface, and oxygen is also present in the sample (Fig. S1). All electronic microscopy images support that HTC carbons are homogeneously covered on the surface of CNTs. The defect sites and graphitic areas of the CNTs (D/G) are important parameters to determine the material properties, such as catalytic activity. Resonance Raman spectroscopy is often used to analyze the electronic structure of CNTs. In this context, the Raman spectra were investigated and fitted to five peaks with Sadezky fitting method, developed for soot and related carbonaceous materials. The amorphous carbon peak was fitted with Gaussian line shape as described by Sadezky et al. As shown in Fig. 4, the G band of around 1580 cm−1 is induced by the E2g vibration mode of sp2 carbon, standing for the flawless graphitized construction. The D band located at about 1320 cm−1 represents the defects and disorder existing on the surface of carbon material. In addition, the band of 1500 cm−1 corresponds to the amorphous carbon. The ratio of the D band to the G band (ID /IG ) is normally utilized to evaluate the degree of graphitization. The ID /IG of CNTs after HNO2 treatment is ca. 2.3, whereas those values of OCNTGlu-90 0, OCNT-GN-90 0 and OCNT-Glu-Urea-900 increase to 2.56, 2.53 and 2.52, respectively. This indicates that the CNTs surface is adsorbed by a rough layer of the amorphous carbon derived from the further carbonization of hydrothermal carbon. The proportion of amorphous carbon in OCNT, OCNT-Glu-900, OCNT-GN-900 and OCNT-Glu-Urea-900 increases progressively. This is in agreement with the SEM and TEM result, indicative of the successful coating of HTC carbon layers on the surface of CNTs. The elemental composition of the modified CNTs as determined by elemental analysis is summarized in Table 1. Whereas the ox-

Please cite this article as: B. Huang et al., Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.03.016

186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228

ARTICLE IN PRESS

JID: JECHEM 4

[m5G;April 12, 2017;0:54]

B. Huang et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

Fig. 4. Raman spectra of all carbon materials. (a) OCNT; (b) OCNT-Glu-900; (c) OCNT-GN-900; (d) Glu-Urea-900. Q6

Table 1. Textural properties and elemental compositions. Samples

OCNT OCNT-Glu-900 OCNT-GN-900 OCNT-Glu-U-900 a b c

229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247

Porosity data

Elemental analysis, wt%

Raman data

SBET (m2 g−1 )

Vtotal (cm3 g−1 )a

Vmeso (cm3 g−1 )b

Vmicro (cm3 g−1 )b

C

H

N

Oc

ID /IG

Amorphous carbon (%)

230 240 263 284

1.80 1.94 1.47 2.19

1.01 1.20 0.95 1.36

0.016 0.017 0.028 0.024

90.8 97.4 96.2 94.4

0.8 0.9 0.7 0.6

0 0.5 0.6 1.7

8.4 1.2 2.5 3.3

2.31 2.56 2.52 2.53

3.9 4.9 7.0 8.4

From total N2 uptake at P/P0 = 0.95. From DFT method. Calculated.

idation of CNTs in refluxing HNO3 yields an O concentration of around 8.4 wt%, this value is drastically reduced to about 1–3 wt% after HTC and thermal treatment at 900 °C dependent on different precursors. As expected, OCNT and OCNT-Glu-900 are almost nitrogen-free. On the contrary, OCNT-GN-900 as well as OCNT-GluUrea-900 appears pronounced nitrogen contents, of which the content in OCNT-GN-900 and OCNT-Glu-Urea-900 is 0.60 and 1.70 wt%, respectively. These trends are in line with those values obtained from X-ray photoelectron spectroscopy and suggest that nitrogen from GU and urea is successfully incorporated within carbon matrix. The incorporation of nitrogen into the surface of CNT was further studied by X-ray photoelectron spectroscopy (XPS). The survey spectra indicate the presence of C 1s, N 1s and O 1s peaks for the samples of OCNT-GN-900 and OCNT-Glu-Urea-900 (Fig. 5). The high-resolution N 1s XPS spectra are deconvoluted into four typical peaks at binding energies. Peaks at ca. 398 eV (N1) are generally assigned to pyridinic N. The configuration of the corresponding N atoms is attached to two C atoms at the edge of graphene layers.

Bands at ca. 400 eV (N2) are usually attributed to pyrrolic N, where N atoms locate at edge sites and bound to two C atoms and one H atom. The high binding energy region at ca. 401 eV (N3) originates from quaternary N. This type of N species substitutes a C atom in the graphene layer interacting by three σ -bonds with neighboring C atoms. The presence of N–O bonds can be observed in both spectra at ca. 403 eV. As seen from both samples, pyridinic N and graphitic N are dominant species, especially the higher content of pyridinic N in OCNT-Glu-Urea-900, which has been proved to possess more efficiently catalytic performance toward ORR (Fig. 6). The porosity of all samples was determined by nitrogen sorption measurements. All isotherms show a small amount of monolayer sorption at low pressure and a evident hysteresis loop and large capillary condensation at p/p0 > 0.9, suggesting the presence of micropore and large mesopores as well as small macropores, which is confirmed by the pore size distribution (PSDs). The micropore stems from inner channel of CNT opened by hash oxidation. The meso- and macro-porosity results from inter particle voids created by the entanglement of the CNTs bundles, which

Please cite this article as: B. Huang et al., Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.03.016

248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266

JID: JECHEM

ARTICLE IN PRESS

[m5G;April 12, 2017;0:54]

B. Huang et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

5

Fig. 5. X-ray photoelectron spectra of OCNT-GN-900 and OCNT-Glu-Urea-900.

267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290

will be expected to make greatly contribution to mass transfer and diffusion of reactants or products. In spite of a subtle difference in surface areas, both the isotherms and PSDs of three materials are quite similar, demonstrating that there is no significant influence on porous structure of CNT upon surface-functionalized by HTC treatment. The specific surface areas increase gradually from OCNT-Glu-900, OCNT-GN-900 to OCNT-Glu-Urea-900, which could be related to the different HTC carbon nano-layers attached on the surface. For OCNT-Glu-Urea-900, the highest amount of ingredients is expected to generate the thickest HTC layer on the surface of CNT, which can effectively impede the entanglement of the CNTs bundles to some extent, resulting in a higher BET surface area and pore volume. However, the effect will be little if the covering HTC layer is too thin, like the case of OCNT-Glu-900. To further compare electrocatalytic activity of each sample, the electrochemical tests of oxygen reduction reaction (ORR) have been carried out, including cyclic voltammetry (CV), linear sweep voltammetry (LSV) by rotating disk electrode (RDE) as well as electron transfer number determined by rotating ring disk electrode (RRDE). All these measurements operated in 0.1 M KOH oxygensaturated electrolyte. As shown in Fig. 7 and Table S1, all modified CNTs present obvious oxygen reduction cathodic peaks in the range of 0.6 V–0.8 V (vs RHE). In detail, OCNT-Glu-Urea-900 exhibits the most positive peak-potential at round 0.75 (vs RHE) as well as

largest peak current density, next is OCNT-GN-900, and OCNT-Glu900 possess the least activity toward ORR. Furthermore, the RDE technology was performed to further compare the kinetics of ORR. The LSV curves in Fig. 7(b) were recorded at a rotating speed of 1600 rpm with a scan rate of 10 mV/s. It can be observed that OCNT with various oxygencontaining functional groups in the surface of CNT displays typical two-plateau peroxide pathway in the lower overpotential region, indicating poor ORR selectivity, which is also confirmed by the electron transfer number of average 2.2 obtained from RRDE data in the potential range from 0.4 to 0.7 V (vs RHE). Compared to OCNT, OCNT-Glu-900 displays higher average electron transfer number of 2.9 and unobvious two-plateau pathway, presumably due to the contribution of a covered HTC layer which is in existence of more defects and oxygen functional groups. Interestingly, for OCNT-Glu-Urea-900, the most positive onset potential and halfwave potential as well as ultrahigh electron transfer number exceeding 3.55 in the entire potential range are achieved, which are all very close to a commercial 20 wt% Pt/C catalyst. As it was previously discussed, doping of the electron-rich nitrogen atoms into the carbon skeleton actually results in the increased electron density at the Fermi level, which generally provides more catalytic active sites and more outstanding properties, such as better electronic conductivity, proper Lewis base active sites. That is also

Please cite this article as: B. Huang et al., Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.03.016

291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314

JID: JECHEM 6

ARTICLE IN PRESS

[m5G;April 12, 2017;0:54]

B. Huang et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

Fig. 6. Isotherms and pore size distributions determined by nitrogen sorption of HTC modified CNTs.

Fig. 7. (a) Cyclic voltammograms in O2 -saturated 0.1 M KOH, (b) comparison of the RDE polarisation curves in O2 -saturated 0.1 M KOH at 1600 rpm, (c, d) the electron transfer number and Koutecky–Levich plots, respectively. (e) RDE polarisation curves in O2 -saturated 0.1 M KOH at various rpm, (f) i–t response at 0.8 V (vs RHE) in O2 -saturated 0.1 M KOH at 1600 rpm.

Please cite this article as: B. Huang et al., Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.03.016

JID: JECHEM

ARTICLE IN PRESS

[m5G;April 12, 2017;0:54]

B. Huang et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

323

the main reason why OCNT-Glu-Urea-900 appear the best activity among all samples in this context as it possesses the highest total N and pyridinic N content. In fact, urea itself is an effective precursor for N-doping process. We have also prepared urea-functional CNTs for active comparison. The activity of urea-functional CNTs is little lower than that of glucose-urea doping composite presumably due to low carbon yield of urea (Fig. S2). In addition, the stability of OCNT-Glu-Urea-900 displays superior durability after 20,0 0 0 s stability test.

324

4. Conclusions

325

335

In this study, surface functionalization of carbon nanotubes (CNTs) was demonstrated by hydrothermal carbonization (HTC) method in the presence of glucose, urea or glucosamine. HTC process allows the homogeneously deposition of the thin nitrogendoped carbon layers on the surface of CNTs. The nitrogen contents can be easily tuned by varying of the ratio of HTC precursors. The resultant nitrogen-doped carbon/CNT composites combine the advantages of CNTs (high electric conductivity) and nitrogen doped carbons (effective ORR active sites). The resultant composites possess favorable electrochemical activity towards ORR in alkaline electrolyte close to that of commercial Pt/C-catalyst.

336

Acknowledgments

337

340

The Award Program for Fujian Minjiang Scholar Professorship is acknowledged for financial support. We thank financial support from the National Natural Science Foundation of China (NSFC Grant number 21571035).

341

Supplementary materials

342 343

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jechem.2017.03.016.

344

References

345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363

[1] B.C. H.Steele, A. Heinzel, Nature 414 (2001) 345–352. [2] R. Bashyam, P.A Zelenay, Nature 443 (2006) 63–66. [3] A. Manthiram, A.V. Murugan, A. Sarkar, T. Muraliganth, Energy Environ. Sci. 1 (2008) 621–638. [4] Y. Li, M. Gong, Y. Liang, J. Feng, J.E. Kim, H. Wang, G. Hong, B. Zhang, H. Dai, Nat. Commun. 4 (2013) 1805–1811. [5] Q. Liu, Y. Wang, L. Dai, J. Yao, Adv. Mater. 28 (2016) 30 0 0–30 06. [6] T.Y. Ma, J. Ran, S. Dai, M. Jaroniec, S.Z. Qiao, Angew. Chem. Int. Ed. 54 (2015) 4646–4650. [7] V. Stamenkovic, B.S. Mun, K.J.J. Mayrhofer, P.N. Ross, N.M. Markovic, J. Rossmeisl, J. Greeley, J.K. Norskov, Angew. Chem. Int. Ed. 45 (2006) 2897–2901. [8] Y.J. Kang, X.C. Ye, J. Chen, Y. Cai, R.E. Diaz, R.R. Adzic, E.A. Stach, C.B. Murray, J. Am. Chem. Soc. 135 (2013) 42–45. [9] G. Wu, P. Zelenay, Acc. Chem. Res. 46 (2013) 1878–1889. [10] X. Ge, A. Sumboja, D. Wuu, T. An, B. Li, F.W.T. Goh, T.S.A. Hor, Y. Zong, Z. Liu, ACS Catal. 5 (2015) 4643–4667. [11] S. Guo, S. Zhang, S. Sun, Angew. Chem. Int. Ed. 52 (2013) 8526–8544. [12] Z.W. Chen, D. Higgins, A.P. Yu, L. Zhang, J.J. Zhang, Energy Environ. Sci. 4 (2011) 3167–3192.

315 316 317 318 319 320 321 322

326 327 328 329 330 331 332 333 334

338 339

7

[13] J. Gong, H. Lin, M. Antonietti, J. Yuan, J. Mater. Chem. A 4 (2016) 7313–7321. [14] Z. Ling, Z. Wang, M. Zhang, C. Yu, G. Wang, Y. Dong, S. Liu, Y. Wang, J. Qiu, Adv. Funct. Mater. 26 (2016) 111–119. [15] J. Xu, J. Zhu, X. Yang, S. Cao, J. Yu, M. Shalom, M. Antonietti, Adv. Mater. 28 (2016) 6727–6733. [16] Z.L. Yu, G.C. Li, N. Fechler, N. Yang, Z.Y. Ma, X. Wang, M. Antonietti, S.H. Yu, Angew. Chem. Int. Ed. 55 (2016) 14623–14627. [17] Y. Zhao, R. Nakamura, K. Kamiya, S. Nakanishi, K. Hashimoto, Nat. Commun. 4 (2013) 2390–2396. [18] K. Gong, F. Du, Z. Xia, M.F. Durstock, L. Dai, Science 323 (2009) 760–764. [19] J. Cao, Y. Chu, X. Tan, Mater. Chem. Phys. 144 (2014) 17–24. [20] X. Tuaev, J.P. Paraknowitsch, R. Illgen, A. Thomas, P. Strasser, Phys. Chem. Chem. Phys. 14 (2012) 6444–6447. [21] B. Wu, Y. Kuang, Y. Zhang, X. Zhang, J. Chen, J. Mater. Chem. 22 (2012) 13085– 13090. [22] X. Xu, Y. Li, Y. Gong, P. Zhang, H. Li, Y. Wang, J. Am. Chem. Soc 134 (2012) 16987–16990. [23] K. Chizari, A. Deneuve, O. Ersen, I. Florea, Y. Liu, D. Edouard, I. Janowska, D. Begin, P.H. Cuong, ChemSusChem 5 (2012) 102–108. [24] R.M. Freire, A.H.D. Batista, A.G. Souza, J. Mendes, G.D. Saraiva, A.C. Oliveira, Catal. Lett. 131 (2009) 135–145. [25] S. van Dommele, K.P. de Jong, J.H. Bitter, Chem. Commun. 46 (2006) 4859– 4861. [26] B. Zhang, H.-H. Wang, H. Su, L.-B. Lv, T.-J. Zhao, J.-M. Ge, X. Wei, K.-X. Wang, X.-H. Li, J.-S. Chen, Nano Res 9 (2016) 2606–2615. [27] Y.J. Sa, C. Park, H.Y. Jeong, S.H. Park, Z. Lee, K.T. Kim, G.G. Park, S.H. Joo, Angew. Chem. Int. Ed. 53 (2014) 4102–4106. [28] T.N. Ye, L.B. Lv, X.H. Li, M. Xu, J.S. Chen, Angew. Chem. Int. Ed. 53 (2014) 6905– 6909. [29] Y. She, Z. Lu, M. Ni, L. Li, M.K. Leung, ACS Appl. Mater. Inter. 7 (2015) 7214– 7221. [30] K. Elumeeva, N. Fechler, T.P. Fellinger, M. Antonietti, Mater. Horiz. 1 (2014) 588–594. [31] C. Han, J. Wang, Y. Gong, X. Xu, H. Li, Y. Wang, J. Mater. Chem. A 2 (2014) 605–609. [32] L. Wang, W. Jia, X. Liu, J. Li, M.-M. Titirici, J. Energy Chem. 25 (2016) 566–570. [33] K. Preuss, V.K. Kannuchamy, A. Marinovic, M. Isaacs, K. Wilson, I. Abrahams, M.-M. Titirici, J. Energy Chem. 25 (2016) 228–235. [34] W. Ding, Z. Wei, S. Chen, X. Qi, T. Yang, J. Hu, D. Wang, L.J. Wan, S.F. Alvi, L. Li, Angew. Chem. Int. Ed. 52 (2013) 11755–11759. [35] L. Lai, J.R. Potts, D. Zhan, L. Wang, C.K. Poh, C. Tang, H. Gong, Z. Shen, J. Lin, R.S. Ruoff, Energy Environ. Sci. 5 (2012) 7936–7942. [36] T. Xing, Y. Zheng, L.H. Li, B.C.C. Cowie, D. Gunzelmann, S.Z. Qiao, S.M. Huang, Y. Chen, ACS Nano 8 (2014) 6856–6862. [37] Y. Zhao, L. Yang, S. Chen, X. Wang, Y. Ma, Q. Wu, Y. Jiang, W. Qian, Z. Hu, J. Am. Chem. Soc. 135 (2013) 1201–1204. [38] D.-W. Wang, D. Su, Energy Environ. Sci. 7 (2014) 576–591. [39] X. Hu, Y. Wu, H. Li, Z. Zhang, J. Phys. Chem. C 114 (2010) 9603–9607. [40] S. Ni, Z.Y. Li, J.L. Yang, Nanoscale 4 (2012) 1184–1189. [41] D.H. Guo, R. Shibuya, C. Akiba, S. Saji, T. Kondo, J. Nakamura, Science 351 (2016) 361–365. [42] Q. Li, S. Zhang, L. Dai, L.S. Li, J. Am. Chem. Soc. 134 (2012) 18932–18935. [43] M. Qiao, C. Tang, G. He, K. Qiu, R. Binions, I.P. Parkin, Q. Zhang, Z. Guo, M.M. Titirici, J. Mater. Chem. A 4 (2016) 12658–12666. [44] S. Zhang, H.-M. Kwon, Z. Li, A. Ikoma, K. Dokko, M. Watanabe, Chem. Electro. Chem. 2 (2015) 1080–1085. [45] Z. Zhang, G.M. Veith, G.M. Brown, P.F. Fulvio, P.C. Hillesheim, S. Dai, S.H. Overbury, Chem. Commun. 50 (2014) 1469–1471. [46] X. Zhang, X. Cheng, Q. Zhang, J. Energy Chem. 25 (2016) 967–984. [47] C. Tang, Q. Zhang, Adv. Mater. (2017) 1604103, doi:10.1002/adma.201604103. [48] J.Q. Huang, M.Q. Zhao, Q. Zhang, J.Q. Nie, L.D. Yao, D.S. Su, F. Wei, Catal. Today 186 (2012) 83–92. [49] Y.X. Wang, X.Z. Cui, Y.S. Li, L.S. Chen, H.R. Chen, L.X. Zhang, J.L. Shi, Carbon 68 (2014) 232–239. [50] G. Tian, Q. Zhang, B. Zhang, Y. Jin, J. Huang, D. Su, F. Wei, Adv. Funct. Mater. 24 (2014) 5956–5961. [51] C. Falco, N. Baccile, M.-M. Titirici, Green Chem 13 (2011) 3273–3281. [52] B. Hu, K. Wang, L. Wu, S.H. Yu, M. Antonietti, M.M. Titirici, Adv. Mater. 22 (2010) 813–828.

Please cite this article as: B. Huang et al., Improving ORR activity of carbon nanotubes by hydrothermal carbon deposition method, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.03.016

364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432