Electrocatalytic study of a 1,10-phenanthroline–cobalt(II) metal complex catalyst supported on reduced graphene oxide towards oxygen reduction reaction

Abstract

A new class of oxygen reduction reaction (ORR) catalyst was fabricated by loading a 1,10-phenanthroline–cobalt(II) metal-complex onto reduced graphene oxide (rGO) surfaces by π–π interaction. The Co(II)–N₄ was the active center of the metal-complex catalyst and catalyzed the ORR via a highly efficient four-electron reduction pathway. The introduction of the nitro-group substituent in 1,10-phenanthroline highly boosted the catalytic activity of the metal complex in terms of half-wave potential (E_1/2) and kinetic current density (J_K), due to the downshift of the e_g-orbital energy level for central Co(II) resulting from the electron-withdrawing effect of the nitro group. Considering the configuration of the metal complex on rGO surfaces, a single cobalt center-mediated catalytic mechanism was proposed to elucidate the ORR process. Compared with the commercial Pt/C catalyst, the as-prepared metal-complex catalyst exhibited a superior methanol tolerance and catalytic durability for the ORR. Our study provides more information about the relationship between the molecular structure and catalytic activity towards the ORR.

---

Introduction

Oxygen reduction reaction (ORR), usually proceeding under the catalysis of metallic Pt, is a particularly important reaction for clean energy production, such as for metal–air batteries and fuel cells.¹ However, the scarcity and high cost of Pt catalyst seriously limit its commercial application. The drive to replace expensive noble Pt catalysts for ORR has led to a class of catalysts comprising transition metal ions stabilized by nitrogen functional groups on carbonaceous surfaces, including pyrolyzed and non-pyrolyzed nitrogen-containing metal-complexes.² For pyrolyzed nitrogen-containing metal-complexes (denoted as M–N_x–C), they were usually prepared by heat treatment of proper transition metal salts, nitrogen-/carbon-based precursors at high temperature,^2–5 and the embedded M–N_x (x = 2, 4) moieties were regarded as the active sites and responsible for the enhanced ORR catalytic activity.^6,7

Since the discovery of cobalt phthalocyanine towards catalyzing ORR,⁸ the non-pyrolyzed metal-complexes, especially for phthalocyanine-/porphyrin-based macrocycle complexes,^9–11 have also attracted intensive attention for alternative ORR catalysts. According to the molecular orbital (MO) theory, the bonding of O₂ molecule to the central metal ions highly depended on the vacancies of d orbitals in metal center.¹² So the current study mainly focused on Fe(II), Co(II) macrocycle metal-complexes, because Fe(II), Co(II) possessed either an empty or half-filled d orbitals, which could accept electrons from O₂ molecule and facilitate the reduction of O₂.⁹ Being similar with the pyrolyzed metal complexes, the M–N₄ center in non-pyrolyzed macrocycle complex was also the active sites of catalyst, and the reduction of O₂ was reported to involve a single-centered M–O–O or bridged M–O–O–M transition state during ORR process.^13,14 Previous studies also proved that the addition of substituent (including electron-withdrawing and electron-donating group) could modulate the electronic structure of macrocycle ligands and further influenced the property of metal-complexes.^14–16

1,10-Phenanthroline (Phen), containing two pyridinic-type nitrogen species, was one of the most versatile chelating ligands in coordination chemistry.¹⁷ Immobilization and grafting of Phen at electrode surface was highly valuable, because it allowed the subsequent formation of metal-complexes by chelating some metal ions, which could find useful applications in biosensor and electrocatalysis field.^18–20 Besides, the Phen had non-amenable electrochemical character and did not participate in oxidation/reduction reaction under normal conditions.¹⁹ So it was a good choice to construct non-pyrolyzed metal-complexes, which contained similar M–N₄ center with phthalocyanine-/porphyrin-based macrocycle complexes, by using 1,10-phenanthroline ligand and further study the relationship between the molecular structure and catalytic activity towards ORR.

In the present work, we reported a kind of novel 1,10-phenanthroline-cobalt(II) metal-complex catalyst for ORR. The 1,10-phenanthroline ligand was firstly loaded onto rGO surfaces by non-covalent π–π interaction,¹⁹ then Co(II) ion was coordinated by pyridinic-type nitrogen species in 1,10-phenanthroline. So the fabricated metal-complexes also contained M–N₄ center. The present work focused on studying (i) the feasibility of constructing high efficient non-pyrolyzed metal-complex catalyst for ORR by employing non-macrocycle ligands; (ii) the effect of electronic structure modulation of ligand molecule on the catalytic activity for metal-complex catalyst; (iii) the catalytic reduction mechanism during ORR process.

Experimental section

Chemicals and materials

The following reagents were used: 1,10-phenanthroline (Phen, 99%), 1,10-phenanthrolin-5-amine (NH₂–Phen, 97%), 5-nitro-1,10-phenanthroline (NO₂–Phen, 98%), cobalt nitrate (Co(NO₃)₂, 99.0%), absolute ethanol (99.7%). All chemicals were used without further purification. All solutions used in electrochemical tests were well prepared with Millipore-Q water (≥18.2 MΩ). 20 wt% Pt/C was purchased from Shanghai Hesen electrical Co. Ltd.

Synthesis of metal-complex catalysts

GO was firstly prepared following a modified Hummer's method.²¹ In a typical process, 3.0 g of graphite powder, 2.5 g of K₂S₂ O₈, 2.5 g of P₂O₅, and 12 mL of H₂SO₄ (98%) were mixed and kept at 80 °C for 4.5 h. Then the mixture was cooled to room temperature, diluted with 500 mL of water, and left overnight. After removal of the residual acid, the preoxidized graphite was further oxidized following the Hummer's method to prepare GO.²² The rGO suspension (0.5 mg mL⁻¹) was obtained by reducing above GO with sodium borohydride. rGO–Phen suspension was prepared by mixing 1 mL rGO (0.5 mg mL⁻¹) suspension and 67 μL Phen (4.12 mM) aqueous solution, then keeping it at room temperature for 24 h to make sure the full loading of Phen onto rGO surfaces. rGO–NH₂–Phen and rGO–NO₂–Phen suspensions were obtained by the same way except for the Phen being replaced by NH₂–Phen (4.12 mM in ethanol) and NO₂–Phen (4.12 mM in ethanol), respectively. rGO–Co, rGO–Phen–Co, rGO–NH₂–Phen–Co, and rGO–NO₂–Phen–Co suspensions were prepared by adding 18 μL Co²⁺ (50 mM) into above rGO, rGO–Phen, rGO–NH₂–Phen, and rGO–NO₂–Phen suspensions, respectively.

Characterization

Atomic force microscopy (AFM) was performed on a Bruker Multimode 8 atomic force microscope. Transmission electron microscopy (TEM) was accomplished by JEM-2100 transmission electron microscope operated at an accelerating voltage of 200 kV. The X-ray photoelectron spectroscopy (XPS) spectra were recorded on a VGESCA LAB MKII X-ray photoelectron spectrometer.

Electrochemical test

All electrochemical tests were conducted on a CHI832B electrochemical workstation (Shanghai Chenhua Co., China) in a standard three-electrode system with a coiled platinum wire (Φ = 0.5 mm, L = 23 cm) and an Ag/AgCl (3.0 M NaCl) electrode as the counter and reference electrodes, respectively. The working electrode was a glassy carbon rotating disk electrode (RDE, ALS Co., Ltd) and its geometric area was 0.1256 cm². Before each test, the RDE surface was polished with alumina slurries (Al₂O₃, 0.05 μm) until a mirror-finish surface was obtained, and then it was fully rinsed with water. The catalyst ink was prepared by adding 10 μL 5 wt% Nafion into 1 mL prepared catalyst suspensions. 10 μL of catalyst ink was coated on the RDE surface using pipettor and then dried at room temperature. The linear sweep voltammetry (LSV) was recorded by slowly sweeping the potential in range of −1.0–0 V at a scan of 5 mV s⁻¹ in O₂-saturated 0.1 M KOH solution. The current–time (i–t) chromoamperometic response was tested at a constant voltage of −0.35 V in O₂-saturated 0.1 M KOH at a rotation rate of 900 rpm.

Results and discussion

The rGO was selected as support due to the facile loading of 1,10-phenanthroline, 1,10-phenanthroline-5-amine, and 5-nitro-1,10-phenanthroline (Fig. 1) by non-covalent π–π interaction except for its extremely large surface area and good electrical conductivity, which made it particularly attractive for electrochemical applications.^23,24 As shown in Fig. 2a and b, the sizes of obtained rGO reached several micrometers, and the thickness was ∼4 nm on the basis of AFM characterization (Fig. 2c and d).

Fig. 1 Molecular structure of 1,10-phenanthroline, 1,10-phenanthroline-5-amine, and 5-nitro-1,10-phenanthroline.

Fig. 2 (a and b) TEM and (c and d) AFM images of as-prepared rGO support.

X-Ray photoelectron spectroscopy (XPS) spectra was applied to explore the formation of Co(II)–N₄ center. As shown in Fig. 3a, c and e, the high-resolution N 1s XPS spectra for rGO–Phen–Co, rGO–NH₂–Phen–Co and rGO–NO₂–Phen–Co could be well-fitted into three or four peaks. The peak at 398.8 eV corresponded to the free pyridinic-like nitrogen species (denoted as N1).²⁵ The peak at 399.4 ± 0.2 eV was regarded as the cobalt(II)-coordinated pyridinic-like nitrogen species (denoted as N2).²⁶ The peak at 400.1 ± 0.1 eV was ascribed to the nitrogen species (denoted as N3) interacting with rGO support.¹⁵ For rGO–NH₂–Phen–Co, an additional peak at 399.1 eV was the nitrogen species (denoted as N4) of amino group. The Co 2p XPS spectra in Fig. 3b, d and f could be divided into four peaks, including Co(II)2p_3/2 at ∼780.2 eV, Co(II)2p_1/2 at ∼796.0 eV, two Co(II) satellite peaks at ∼786.0 and ∼802.0 eV.¹⁵ Combining N 1s and Co 2p XPS spectra, we could well confirm the successful loading of metal-complex catalysts onto rGO surfaces and the formation of Co(II)–N₄ center.

Fig. 3 High-resolution N 1s XPS spectra for (a) rGO–Phen–Co, (c) rGO–NH₂–Phen–Co and (e) rGO–NO₂–Phen–Co; High-resolution Co 2p XPS spectra for (b) rGO–Phen–Co, (d) rGO–NH₂–Phen–Co and (f) rGO–NO₂–Phen–Co.

To explore the active site of as-prepared metal-complex catalysts, the eletrocatalytic activity for catalyst support of rGO and the supporting ligands of rGO–Phen, rGO–NH₂–Phen, and rGO–NO₂–Phen towards ORR was firstly studied. As shown in Fig. 4a, the LSV curves were recorded between −1.0 and 0 V by using RDE at 1600 rpm in O₂-saturated 0.1 M KOH solution. It was found that the half-wave potentials (E_1/2) followed: rGO (−0.28 V) < rGO–Phen (−0.25 V) = rGO–NH₂–Phen (−0.25 V) ≈ rGO–NO₂–Phen (−0.24 V). We also investigated the LSV curves at different rotating rates for above catalysts. As shown in Fig. S1 (ESI†), the limiting current densities significantly rose following the increase of rotation rates, as high rotation rate leaded to a faster oxygen flux to electrode surface and consequently resulted in larger current.²⁷ To evaluate the ORR performance more quantitatively, the kinetic analysis was carried out by the Koutecky–Levich (K–L) equation:²⁸

where J, J_K, J_L represented the measured, kinetic-, and diffusion-limiting current density, respectively. n was the number of electrons transferred, F Faraday constant, D the diffusion coefficient of O₂, ω the rotation rate, ν the kinetic viscosity of solution, and C* the concentration of dissolved O₂ in solution. The number (n) of electrons transferred in ORR is an important parameter for evaluating the catalytic activity. This feature can be obtained from the slope of the K–L curves by plotting J⁻¹ versus ω^−1/2. Fig. 4b–e show the K–L curves for rGO, rGO–Phen, rGO–NH₂–Phen, and rGO–NO₂–Phen. The data showed good linearity, and the slopes remained approximately constant over the potential range from −0.30 to −0.70 V, suggesting the consistent electron transfer and the first-order reaction kinetics with respect to the O₂ concentration.^29,30 According to above equations, the average n values were calculated to be 2.5, 2.4, 2.4, 2.5 for rGO, rGO–Phen, rGO–NH₂–Phen, and rGO–NO₂–Phen, respectively. It implied that the reduction of molecular oxygen was a typical two-electron reduction pathway, and H₂O₂ was the main reduction product. From the K–L plots, the kinetic current density (J_K) could also be obtained by taking the inverse of the Y-intercept.³¹ It was found that the kinetic current densities (J_K) (at −0.35 V vs. Ag/AgCl) followed: rGO (2.1 mA cm⁻²) < rGO–Phen (2.9 mA cm⁻²) ≈ rGO–NH₂–Phen (3.0 mA cm⁻²) < rGO–NO₂–Phen (3.7 mA cm⁻²). Compared with primary rGO support, the loading of ligand molecules (especially for NO₂–Phen) alone also exhibited an enhanced catalytic activity towards ORR, confirmed by the improvement of half-wave potentials (E_1/2) and kinetic current densities (J_K) to some extent. However, they could not vary the molecular oxygen reduction pathway (two-electron) in ORR process (Fig. 4f).

Fig. 4 (a) Comparison of LSV polarization curves in O₂-saturated 0.1 M KOH solution (scan rate: 5 mV s⁻¹; rotation rate: 1600 rpm). Koutecky–Levich plots (J⁻¹–ω^−1/2) for (b) rGO, (c) rGO–Phen, (d) rGO–NH₂–Phen, and (e) rGO–NO₂–Phen at different potentials (from top to bottom: −0.30, −0.35, −0.40, −0.45, −0.50, −0.55, −0.60, −0.65, −0.70 V). (f) Comparison of the number (n) of electrons transferred for above catalysts towards ORR.

It was notable that the average n values for rGO–Co, rGO–Phen–Co, rGO–NH₂–Phen–Co and rGO–NO₂–Phen–Co could reach 3.4, 3.9, 4.0, and 4.0, respectively when cobalt(II) ion being introduced (Fig. S2, ESI† and Fig. 5a–e). It suggested that the Co(II)–N₄ structure was the active center of metal-complex catalysts and responsible for the high efficient four-electron reduction pathway, corresponding to the complete reduction of molecular oxygen into OH⁻ in alkaline solution. We also investigated the influence of substituent group in aromatic ring of Phen on the catalytic activity. It was found that the half-wave potentials (E_1/2) followed: rGO–Co (−0.29 V) < rGO–Phen–Co (−0.25 V) ≈ rGO–NH₂–Phen–Co (−0.24 V) < rGO–NO₂–Phen–Co (−0.18 V), and the kinetic current densities (J_K) (at −0.35 V vs. Ag/AgCl) followed: rGO–Co (1.7 mA cm⁻²) < rGO–Phen–Co (2.8 mA cm⁻²) < rGO–NH₂–Phen–Co (3.6 mA cm⁻²) < rGO–NO₂–Phen–Co (4.6 mA cm⁻²). So the rGO–NO₂–Phen–Co outperformed other metal-complex catalysts in terms of number (n) of electrons transferred, half-wave potential (E_1/2), and kinetic current density (J_K) (Fig. 5f). On the basis of above results, we could conclude that (i) the Co(II)–N₄ was the active center of metal-complex catalysts for the high efficient four-electron reduction pathway; (ii) the introduction of electron-withdrawing group (NO₂–) in 1,10-phenanthroline could evidently improve the catalytic activity towards ORR.

Fig. 5 Koutecky–Levich plots (J⁻¹–ω^−1/2) for (a) rGO–Co, (b) rGO–Phen–Co, (c) rGO–NH₂–Phen–Co, and (d) rGO–NO₂–Phen–Co at different potentials (from top to bottom: −0.35, −0.40, −0.45, −0.50, −0.55, −0.60, −0.65, −0.70 V). (e) Comparison of the number (n) of electrons transferred for above catalysts towards ORR. (f) Comparison of LSV polarization curves in O₂-saturated 0.1 M KOH solution (scan rate: 5 mV s⁻¹; rotation rate: 1600 rpm).

How to explain the effect of nitro-group on the catalytic activity for metal-complex catalyst? In previous study, Zagal et al. have pointed out that the electrocatalytic performance of metal-complex catalyst towards ORR was governed by the nature of central metal ion and the surrounding ligand molecules.³² According to the first-principle DFT calculation by Kiefer and Shu,^6,7 the O₂ molecule preferred to bind on the top of central metal ion with an end-on configuration. On the basis of MO approach, the end-on M–O₂ interaction primarily involved the σ-type bonding between O₂ molecular orbital and e_g-orbital of the transition metal ion.^12,33 A downshift in energy level of e_g-orbital away from the Fermi level could increase the ORR reduction potential and also optimize the binding strength of ORR intermediates.² Compared with unsupported metal-complex catalysts, the rGO–Phen–Co, rGO–NH₂–Phen–Co, and rGO–NO₂–Phen–Co should have a much lower e_g-orbital energy level for central Co(II) due to the formation of electron donor–acceptor pair between metal-complex and rGO plane via π–π interaction.³⁴ Besides, the introduction of nitro-group, a typical electron-withdrawing group, could further lower the energy level of Co(II) e_g-orbital.³⁵ So it was a rational and convincing conclusion that the rGO–NO₂–Phen–Co metal-complex catalyst exhibited the best catalytic performance towards ORR (Table 1).

Table 1 Comparison of ORR catalytic activity for catalysts in O₂-saturated 0.1 M KOH solution

Catalyst	n	E1/2^a (V)	JK^b mA cm^−2
a The rotation rate was 1600 rpm.b Potential was fixed at −0.35 V.
rGO	2.5	−0.28	2.1
Phen–rGO	2.4	−0.25	2.9
NH2–Phen–rGO	2.4	−0.25	3.0
NO2–Phen–rGO	2.5	−0.24	3.7
Co–rGO	3.4	−0.29	1.7
rGO–Phen–Co	3.9	−0.25	2.8
rGO–NH2–Phen–Co	4.0	−0.24	3.6
rGO–NO2–Phen–Co	4.0	−0.18	4.6

---

Considering the configuration of rGO–NO₂–Phen–Co, we speculated that the ORR in alkaline solution proceeded through a single cobalt center-mediated reduction mechanism.^13,36 And the oxidation state of metal-center Co underwent Co(II)/Co(III) redox transitions during ORR process, which was crucial for the four-electron reduction pathway. According to the Mukerjee's study,³⁷ the oxidation state transition from Co(II) to Co(III) could reduce the number of d-electrons for central Co, which could facilitate the adsorption of O₂ molecules on active sites, and optimize the binding energy of ORR intermediates. Contrast experiment also revealed that the ORR under the catalysis of rGO–NO₂–Phen–Ni(II) was a two-electron reduction pathway due to the difficult transition from Ni(II) to Ni(III) (Fig. S3, ESI†). Based on previous study,² we proposed that the ORR process under the catalysis of rGO–NO₂–Phen–Co proceeded following a series of reactions:

rGO–NO₂–Phen–Co(II) + OH⁻ → rGO–NO₂–Phen–Co(II)–OH⁻ (2a)

rGO–NO₂–Phen–Co(II)–OH⁻ + O₂ → rGO–NO₂–Phen–Co(II)–O–O + OH⁻ (2b)

rGO–NO₂–Phen–Co(II)–O–O → rGO–NO₂–Phen–Co(III)–O–O (2c)

rGO–NO₂–Phen–Co(III)–O–O + H₂O + e⁻ → rGO–NO₂–Phen–Co(III)–O–OH + OH⁻ (2d)

rGO–NO₂–Phen–Co(III)–O–OH + e⁻ → rGO–NO₂–Phen–Co(II)–O–OH⁻ (2e)

rGO–NO₂–Phen–Co(II)–O–OH⁻ + H₂O + e⁻ → rGO–NO₂–Phen–Co(III)–OH + 2OH⁻ (2f)

rGO–NO₂–Phen–Co(III)–OH + e⁻ → rGO–NO₂–Phen–Co(II)–OH⁻ (2g)

rGO–NO₂–Phen–Co(II)–OH⁻ → rGO–NO₂–Phen–Co(II) + OH⁻ (2h)

To well elucidate above reactions, a schematic illustration (Fig. 6) was exhibited to describe the single cobalt center-mediated ORR catalytic mechanism.

Fig. 6 The proposed ORR mechanism under the catalysis of rGO–NO₂–Phen–Co in alkaline medium.

Since catalytic durability was a major concern of ORR catalysts,³⁸ the electrocatalytic stability of rGO–NO₂–Phen–Co and commercial Pt/C catalysts towards ORR was tested at a constant voltage of −0.35 V in O₂-saturated 0.1 M KOH at a rotation rate of 900 rpm. As shown in Fig. 7, the current–time (i–t) chromoamperometic response for rGO–NO₂–Phen–Co exhibited a much slower decrease than the commercial Pt/C catalyst. Besides, the selectivity of catalyst towards ORR was also of great significance, because some fuel molecules, e.g. methanol, may seriously affect the catalytic performance of cathode catalysts. It was found that rGO–NO₂–Phen–Co exhibited a stable amperometric response even after introducing methanol at 1000 s. While the relative current on commercial Pt/C drastically decreased following the addition of methanol. So the as-prepared rGO–NO₂–Phen–Co metal-complex catalyst exhibited a superior methanol tolerance and catalytic durability during electrocatalytic ORR process when comparing with commercial Pt/C catalyst.

Fig. 7 The current–time (i–t) chronoamperometric responses for ORR in O₂-saturated 0.1 M KOH solution at the rGO–NO₂–Phen–Co and Pt/C electrodes followed by adding 3 M methanol at 1000 s (rotation rate: 900 rpm).

Conclusions

In conclusion, a new class metal-complex ORR catalyst was fabricated by facilely loading 1,10-phenanthroline-cobalt(II) onto rGO surfaces by π–π interaction. The Co(II)–N₄ center of metal-complexes was confirmed to be the active sites and responsible for the four-electron reduction pathway in ORR process. The introduction of nitro-group substituent highly enhanced the catalytic activity of metal-complex catalyst due to the electron-withdrawing effect. A single cobalt center-mediated mechanism was proposed to elucidate the ORR process. Compared with commercial Pt/C catalyst, the as-prepared metal-complex catalyst exhibited a superior methanol tolerance and catalytic durability for ORR.

Acknowledgements

The work was financially supported by the National Natural Science Foundation of China (21105041, 21127006, 21205056, 21405070), Natural Science Foundation of Shandong Province (2015GGH317001), and Promotive Research Foundation for Excellent Young and Middle-Aged Scientists of Shandong Province (BS2013CL005).

Notes and references

1. F. Cheng and J. Chen, Chem. Soc. Rev., 2012, 41, 2172–2192.
2. N. Ramaswamy, U. Tylus, Q. Jia and S. Mukerjee, J. Am. Chem. Soc., 2013, 135, 15443–15449.
3. D. Singh, K. Mamtani, C. R. Bruening, J. T. Miller and U. S. Ozkan, ACS Catal., 2014, 4, 3454–3462.
4. A. Kong, X. Zhu, Z. Han, Y. Yu, Y. Zhang, B. Dong and Y. Shan, ACS Catal., 2014, 4, 1793–1800.
5. B. Xia, Y. Yan, N. Li, H. Wu, X. Lou and X. Wang, Nature Energy., 2016, 1, 15006.
6. S. Kattel, P. Atanassov and B. Kiefer, Phys. Chem. Chem. Phys., 2013, 15, 148–153.
7. F. Li, H. Shu, C. Hu, Z. Shi, X. Liu, P. Liang and X. Chen, ACS Appl. Mater. Interfaces, 2015, 7, 27405–27413.
8. R. Jasinski, Nature, 1964, 201, 1212–1213.
9. M. A. Thorseth, C. E. Tornow, E. C. M. Tse and A. A. Gewirth, Chem. Rev., 2013, 257, 130–139.
10. H. J. Choi, N. Ashok Kumar and J. B. Baek, Nanoscale, 2015, 7, 6991–6998.
11. D. Liu and Y. T. Long, ACS Appl. Mater. Interfaces, 2015, 7, 24063–24068.
12. H. Alt, H. Binder and G. Sandstede, J. Catal., 1973, 28, 8–19.
13. W. Zhang, A. U. Shaikh, E. Y. Tsui and T. M. Swager, Chem. Mater., 2009, 21, 3234–3241.
14. E. Song, C. Shi and F. C. Anson, Langmuir, 1998, 14, 4315–4321.
15. Y. Gartia, C. M. Parnell, F. Watanabe, P. Szwedo, A. S. Biris, N. Peddi, Z. A. Nima and A. Ghosh, ACS Sustainable Chem. Eng., 2015, 3, 97–102.
16. B. Sun, Z. Ou, D. Meng, Y. Fang, Y. Song, W. Zhu, P. V. Solntsev, V. N. Nemykin and K. M. Kadish, Inorg. Chem., 2014, 53, 8600–8609.
17. P. G. Sammes and G. Yahioglu, Chem. Soc. Rev., 1994, 23, 327–334.
18. A. D. Pauric, B. J. MacLean and E. B. Easton, J. Electrochem. Soc., 2011, 158, B331–B336.
19. P. Gayathri and A. Senthil Kumar, Langmuir, 2014, 30, 10513–10521.
20. D. Huang, Y. Luo, S. Li, B. Zhang, Y. Shen and M. Wang, Nano Res., 2014, 7, 1054–1064.
21. Y. Xu, H. Bai, G. Lu, C. Li and G. Shi, J. Am. Chem. Soc., 2008, 130, 5856–5857.
22. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
23. L. Dai, Y. Xue, L. Qu, H. J. Choi and J. B. Baek, Chem. Rev., 2015, 115, 4823–4892.
24. B. Xia, Y. Yan, X. Wang and X. Lou, Mater. Horiz., 2014, 1, 379–399.
25. H. Li, W. Kang, L. Wang, Q. Yue, S. Xu, H. Wang and J. Liu, Carbon, 2013, 54, 249–257.
26. R. Zhang, S. He, Y. Lu and W. Chen, J. Mater. Chem. A, 2015, 3, 3559–3567.
27. F. Cheng, Y. Su, J. Liang, Z. Tao and J. Chen, Chem. Mater., 2010, 22, 898–905.
28. R. Liu, D. Wu, X. Feng and K. Müllen, Angew. Chem., Int. Ed., 2010, 49, 2565–2569.
29. S. Shanmugam and T. Osaka, Chem. Commun., 2011, 47, 4463–4465.
30. R. Liu, C. von Malotki, L. Arnold, N. Koshino, H. Higashimura, M. Baumgarten and K. Müllen, J. Am. Chem. Soc., 2011, 133, 10372–10375.
31. Z. Chen, D. Higgins, H. Tao, R. S. Hsu and Z. Chen, J. Phys. Chem. C, 2009, 113, 21008–21013.
32. J. Zagal, M. Páez, J. F. Silva, F. Bedioui and J. P. Dodelet, N4-Macrocyclic Metal Complexes, 2006.
33. H. Nakashima, J. Y. Hasegawa and H. Nakatsuji, J. Comput. Chem., 2006, 27, 426–433.
34. S. i. Yamazaki, Y. Yamada, T. Ioroi, N. Fujiwara, Z. Siroma, K. Yasuda and Y. Miyazaki, J. Electroanal. Chem., 2005, 576, 253–259.
35. H. Yamashige, S. Matsuo, T. Kurisaki, R. C. C. Perera and H. Wakita, Anal. Sci., 2005, 21, 309–314.
36. G. Lu, H. Yang, Y. Zhu, T. Huggins, Z. J. Ren, Z. Liu and W. Zhang, J. Mater. Chem. A, 2015, 3, 4954–4959.
37. Q. Jia, N. Ramaswamy, H. Hafiz, U. Tylus, K. Strickland, G. Wu, B. Barbiellini, A. Bansil, E. F. Holby, P. Zelenay and S. Mukerjee, ACS Nano, 2015, 9, 12496–12505.
38. H. Yu, Y. Li, X. Li, L. Fan and S. Yang, Chem.–Eur. J., 2014, 20, 3457–3462.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra04078k

---