Catalytic conversion of ferrous carbonate to higher hydrocarbons under mild conditions and its application in transformation of CO₂ to liquid fuels

Abstract

The generation of liquid fuels of higher hydrocarbons under mild conditions by consuming CO₂ would bring about many economic and environmental benefits. Here we report a finding that Pt and Ru nanoparticles deposited on FeCO₃ could catalyze the conversion of carbonate ions in FeCO₃ and H₂ to higher hydrocarbons under mild conditions. Coupling this new reaction route with the carbonation of formed surface ferrous species resulted in the conversion of CO₂ and H₂ to higher hydrocarbons (C₂–C₂₆) at low or even ambient temperatures due to the low activation-energy for CH_x coupling in the process. At 40 °C, the selectivity for C₅–C₂₆ hydrocarbons and C₂⁺ compounds reached 49.6% and 77.7%, respectively. This finding sheds a new light on regenerating multi-carbon energy carriers in a sustainable manner.

---

Introduction

Ferrous carbonate is a main component of siderite which is a worldwide distributed iron mineral with large deposits. So far, ferrous carbonate has been utilized as a fertilizer, feed additive, and raw material for steel production.^1,2 Conversion of CO₃²⁻ in ferrous carbonate to organic products such as liquid fuels or feed-stocks in the chemical industry under mild conditions is an attractive new subject. On the other hand, the increasing demand for fossil energy and resources has caused a continuous increase of CO₂ concentration in the atmosphere, resulting in serious environmental problems such as the greenhouse effect and ocean acidification. The catalytic conversion of CO₂ to value-added products has attracted increasing attention.^3–6

Higher hydrocarbons (C_xH_y, x ≥ 2) with carbon numbers of 2–4, 5–12, and 13–22 are used as feed-stocks for manufacture of polymers, gasoline, and diesel, respectively. Currently, these products are predominantly produced by petroleum refining or in coal chemical industry based on Fischer–Tropsch (F–T) synthesis at high temperatures.^7,8

Conversion of ferrous carbonate or CO₂ and hydrogen to higher hydrocarbons at low or ambient temperatures may create many energy and ecological benefits, because it will make it possible to produce high-value products in an environmentally friendly manner using H₂ derived from renewable energy, reduce energy consumption, decrease the dependence on fossil resources, and avoid over emission of CO₂ in the conversion process.

We have been engaging in the development of iron oxide supported platinum group metal nanocluster catalysts for a long time.^9–11 Herein, we report a new finding that Ru and Pt nanoparticles deposited on ferrous carbonate could catalyze the conversion of carbonate ions in FeCO₃ and hydrogen to higher hydrocarbons at low temperatures, and the coupling of this new reaction (CTHC) with carbonation of the formed ferrous species by CO₂ could establish a new route (C-CTHC) to realize the conversion of CO₂ and H₂ to hydrocarbons (C₁–C₂₆) at low or even ambient temperatures (40–130 °C).

Results and discussion

Ferrous carbonate supported Ru and Pt nanoparticles (Ru–Pt/FeCO₃) were prepared by depositing Pt nanoclusters stabilized with ethylene glycol and simple ions¹² and Ru oxide nanoparticles on iron oxide,^13,14 consecutively, followed by treating the obtained solid intermediate with hydrogen and CO₂ in a mixture of cyclohexane and water at 160 °C (see the ESI†). The Pt and Ru contents of Ru–Pt/FeCO₃ were measured using an inductively coupled plasma atomic emission spectrometer (ICP-AES) to be 1.6% and 2.1%, respectively.

The X-ray diffraction (XRD) pattern of Ru–Pt/FeCO₃ exhibited the diffraction signals of crystal planes of FeCO₃, while the diffraction peaks of the supported metal nanoparticles were too weak to analyze due to the strong diffraction signals of the support (Fig. S1, ESI†).

As shown in the transmission electron microscope (TEM) image of Ru–Pt/FeCO₃ (Fig. 1A), metal nanoparticles dispersed on FeCO₃ had an average diameter of 1.8 nm, with a standard deviation of 0.4 nm (Fig. S2, ESI†). Pt and Ru nanoparticles with the inter-fringe distances of 0.227 and 0.206 nm were observed in the high resolution TEM image (Fig. 1B), respectively, indicating that Ru–Pt/FeCO₃ was composed of Ru and Pt nanoparticles deposited on FeCO₃. EDX (energy dispersive X-ray spectroscopy) mapping measurements revealed that many Pt nanoparticles were in contact with Ru nanoparticles in Ru–Pt/FeCO₃ (Fig. 1H).

Fig. 1 (A–C) TEM image, high-resolution TEM image, and HAADF-STEM image of Ru–Pt/FeCO₃, (D) high-resolution TEM image of the selected area in (C), (E–G) EDX elemental mapping of Pt, Fe, and Ru in the same area, and (H) merged image of (E) and (G).

X-ray photoelectron spectroscopy (XPS) measurements on Ru–Pt/FeCO₃ revealed that the electron binding energies of Ru 3d_5/2 and Pt 4f_7/2 levels had values of 280.0 and 71.3 eV, respectively, indicating that Ru and Pt were in metallic states.^15,16 The electron binding energies of O 1s and Fe 2p_3/2 levels were measured to be 531.9 and 710.0 eV, respectively, which were consistent with those in FeCO₃ (Fig. S3, ESI†).¹⁷

We found that heating Ru–Pt/FeCO₃ and 3.75 MPa of H₂ (25 °C) at 130 °C for 24 h in a mixture of cyclohexane and water could produce C₁–C₂₆ hydrocarbons as the main products, with a selectivity for C₂⁺ hydrocarbons of 26.2% and a time-yield of hydrocarbons (based on moles of carbon) of 181.0 mmol mol_metal⁻¹ h⁻¹ at a FeCO₃ conversion of 15.3% (Table 1), indicating that Pt and Ru nanoparticles on FeCO₃ could catalyze a reaction between carbonate ions and H₂ to form higher hydrocarbons (CTHC). The carbon balance, i.e. the ratio of carbon in the product to that consumed in CTHC, was measured to be 90.1% based on the experimental data (see the ESI†).

Table 1 Catalytic properties for CTHC and C-CTHC at different temperatures^a

T (°C)	P CO2 (MPa)	Selectivity^b				Time-yield^b
		CH4	C2^+^c	C5^+^c	ROH^c
a Reaction conditions: Ru–Pt/FeCO3, 0.3 g; water, 15 ml; cyclohexane, 15 ml; H2, 3.75 MPa; time, 24 h. b Selectivity (%) and time-yield (mmol molmetal^−1 h^−1) were calculated based on moles of carbon, where metal denotes platinum group metal. c C2^+ and C5^+ denote hydrocarbons with carbon numbers larger than 2 and 5. ROH denotes C1–C6 alcohols. d Reaction time was 8 hours.
130	—	70.7	26.2	16.0	3.1	186.8
130	1.25	51.8	42.1	26.1	6.1	919.5^d
80	—	58.7	33.7	13.5	7.6	67.2
80	1.25	32.5	55.9	32.3	11.6	114.9

---

To confirm the new chemical conversion pathway, we prepared Ru–Pt/Fe¹³CO₃ using ¹³C labeled carbon dioxide (¹³CO₂) by the same method and performed the reaction of Ru–Pt/Fe¹³CO₃ with H₂ under the same conditions. The produced higher hydrocarbons were collected and analyzed by gas chromatography-mass spectrometry (GC-MS). In the mass spectra (MS) of pentane, a representative of the higher hydrocarbons, the molecular ion peak of [¹³C₅H₁₂]˙⁺ with a m/z value of 77 could be observed clearly, whereas that of [¹²C₅H₁₂]˙⁺ with a m/z value of 72 did not appear (Fig. 2), proving that carbon in the higher hydrocarbons was derived from carbonate ions in FeCO₃.

Fig. 2 Mass spectrum of pentane produced in the reaction of H₂ with Ru–Pt/Fe¹²CO₃ (A) and Ru–Pt/Fe¹³CO₃ (B) at 130 °C, respectively.

Furthermore, the aforementioned CTHC was conducted under an atmosphere of ¹³CO₂ (0.75 MPa, 25 °C) in cyclohexane at 130 °C. In the MS of the produced pentane, the enhancement in the signals for the ¹³C containing species could not be observed (Fig. S4, ESI†), demonstrating that higher hydrocarbons could be produced by the hydrogenation of FeCO₃, without CO₂ participation in the reaction.

After treating Ru–Pt/FeCO₃ with hydrogen in a mixture of water and cyclohexane at 130 °C, the solid residue was separated for XPS measurements. The XPS band of Fe 2p_3/2 in the sample, with a maximum at 710.5 eV and a shoulder band at 710.0 eV, revealed that a part of ferrous carbonate was transformed into iron oxide¹⁸ in the reaction. After the recovered sample was treated with a mixture of CO₂ and H₂ at 130 °C, the binding energy of Fe 2p_3/2 in the collected sample changed back to 710.0 eV of FeCO₃ (Fig. S5, ESI†), revealing the possibility of coupling CTHC with the carbonation reaction of surface ferrous species in the presence of CO₂.

When CTHC was performed under an atmosphere of ¹³CO₂ (0.75 MPa, 25 °C) at 130 °C in a mixture of water and cyclohexane, the relative intensities of signals for [C₂H₅]⁺, [C₃H₇]⁺, [C₄H₉]⁺, and [C₅H₁₂]˙⁺ containing ¹³C, with m/z values of 30–31, 44–46, 58–61, and 73–77, respectively, significantly enhanced in the MS of pentane (Fig. 3A), indicating that the CTHC process could couple with the carbonation reaction of formed ferrous species in the presence of water to form a new catalytic reaction route (C-CTHC) to synthesize higher hydrocarbons from CO₂ and H₂ under mild conditions as shown in Fig. 3B.

Fig. 3 Mass spectrum of pentane produced in C-CTHC using ¹³CO₂ in a mixture of water and cyclohexane (A), and a schematic of the proposed catalytic mechanism for C-CTHC (B).

C-CTHC could produce C₁–C₂₆ hydrocarbons at 130 °C in a mixture of water and cyclohexane with a time-yield of hydrocarbons of 863.4 mmol mol_metal⁻¹ h⁻¹ and a selectivity for C₂⁺ hydrocarbons of 42.1% (Table 1), which were much higher than those of CTHC at the same temperature (Fig. S6, ESI†).

It was found that at 80 °C in the mixture of cyclohexane and water, CTHC could produce C₁–C₁₃ hydrocarbons with a selectivity for C₂⁺ hydrocarbons of 33.7% and a time-yield of hydrocarbons of 62.1 mmol mol_metal⁻¹ h⁻¹, while C-CTHC could produce C₁–C₂₆ hydrocarbons with a selectivity for C₂⁺ hydrocarbons of 55.9% and a time-yield of hydrocarbons of 101.6 mmol mol_metal⁻¹ h⁻¹ (Table 1). These experimental results implied that in the present conversion process of CO₃²⁻ to hydrocarbons, the introduced CO₂ is important for producing hydrocarbons with longer carbon chains and increasing the reaction rate at low temperatures, which may be due to the increase of CH_x species concentration around the Pt nanoparticles (see below).

The product distributions for C-CTHC in a mixture of cyclohexane and water at different temperatures are shown in Fig. 4A. It is seen that at 40 °C, the selectivity for C₅–C₂₆ hydrocarbons and C₂⁺ compounds reached 49.6% and 77.7%, respectively, while that for methane was only 7.3%. With the reaction temperature increasing, the time-yield of hydrocarbons increased obviously, and the selectivity for higher hydrocarbons decreased. At 130 °C, the selectivities for C₅–C₂₆ hydrocarbons and C₂⁺ compounds were 26.1% and 45.8%, respectively, with a conversion of carbon in the substrates (including FeCO₃ and CO₂) of 5.1% in 8 hours. The distribution of the formed hydrocarbon products basically followed ASF (Anderson–Schulz–Flory) statistics with a chain growth factor (α) of 0.77 (Fig. S7, ESI†), which was consistent with the production of hydrocarbons with long carbon chains. CO, which was usually produced in the catalytic hydrogenation of CO₂, was not detected by GC (detection limit: 60 ppm) in the products of the present catalytic system.

Fig. 4 (A) Product distributions of C-CTHC at different temperatures. Reaction conditions: Ru–Pt/FeCO₃, 0.3 g; water, 15 ml; cyclohexane, 15 ml; H₂, 3.75 MPa; CO₂, 1.25 MPa. Reaction time at 130 °C was 8 hours, and that at 100 and 80 °C was 24 hours, while that at 40 °C was 12 days. Alcohols produced at 40 °C were C₁–C₄ alcohols, while those produced at other reaction temperatures were C₁–C₆ alcohols. (B) Apparent activation energy measurements for the formation of methane and higher hydrocarbons (C₂⁺) in C-CTHC, respectively.

It was found that decreasing the pressure of H₂ could increase the selectivity for C₂⁺ hydrocarbons, with a declined time-yield. When H₂ pressure varied from 3.75 MPa (25 °C) to 2.50 MPa (25 °C), the selectivity for C₂⁺ hydrocarbons increased from 42.1% to 55.6%, while the time-yield of C₂⁺ hydrocarbons declined from 387.1 to 292.6 mmol mol_metal⁻¹ h⁻¹ as measured after the reaction was conducted for 8 hours at 130 °C. Upon decreasing H₂ pressure to 1.25 MPa (25 °C), the selectivity for C₂⁺ hydrocarbons reached 68.1%, with a time-yield of 109.2 mmol mol_metal⁻¹ h⁻¹ under the same conditions (Table S1, ESI†).

Based on the time-yield of methane and C₂⁺ hydrocarbons in C-CTHC at different temperatures (313, 373, 403, and 423 K), the apparent activation energy for the formation of methane was calculated to be 79.8 ± 3.2 kJ mol⁻¹, while that for C₂⁺ hydrocarbons was 50.7 ± 3.9 kJ mol⁻¹ (Fig. 4B), implying that the activation energy for hydrogenation of CH_x to form methane is much higher than that for the coupling of CH_x (x = 2, 3) to form higher hydrocarbons in the present process, accounting for the increase in the selectivity for higher hydrocarbons with decreasing the reaction temperature.

It should be mentioned that over the previously reported catalytic systems, the synthesis of hydrocarbons with carbon numbers up to 15 through the CO₂-FT process needed a high reaction temperature (∼290 °C) due to the high activation energy for the carbon–carbon coupling reaction,^19,20 demonstrating the advantage of C-CTHC in the synthesis of multi-carbon compounds at low temperatures over CO₂-FT.

To further study the roles of Ru and Pt nanoparticles in C-CTHC, Ru/FeCO₃ and Pt/FeCO₃ were prepared by the method described in this work in the absence of Pt or Ru, respectively. The diffraction peaks of ferrous carbonate are clearly observed in the XRD patterns of the prepared Ru/FeCO₃ and Pt/FeCO₃ (Fig. S8, ESI†). The average diameters of Ru and Pt nanoparticles in Ru/FeCO₃ and Pt/FeCO₃ were measured to be 1.7 ± 0.4 and 1.9 ± 0.3 nm, respectively, which were similar to those of metal nanoparticles in Ru–Pt/FeCO₃. C-CTHC over Pt/FeCO₃, Ru/FeCO₃, and Ru–Pt/FeCO₃ was conducted in a mixture of CO₂ (1.25 MPa, 25 °C) and H₂ (3.75 MPa, 25 °C) at 130 °C, respectively (Table S2, ESI†). The results revealed that Ru/FeCO₃ could catalyze the reaction of CO₃²⁻ with H₂ to produce hydrocarbons (C₁–C₁₀) with a time-yield of 766 mmol_CH2 mol_metal⁻¹ h⁻¹, and the selectivity for methane was as high as 91.6%, implying that Ru nanoparticles deposited on FeCO₃ could efficiently catalyze the hydrogenation of CO₃²⁻ to produce (* means adsorbed species), and tended to couple with H* to form methane over Ru/FeCO₃. On the other hand, Pt nanoparticles deposited on FeCO₃ could catalyze the hydrogenation conversion of CO₃²⁻ to hydrocarbons (C₁–C₁₄) with a selectivity for methane and C₂⁺ hydrocarbon of 26.1 and 49.6%, respectively, and a time-yield of 175 mmol_CH2 mol_metal⁻¹ h⁻¹, implying that Pt nanoparticles deposited on FeCO₃ were not as efficient as Ru/FeCO₃ in catalyzing the hydrogenation of CO₃²⁻ to form , but the formed tended to couple with each other to form higher hydrocarbons over Pt/FeCO₃. Ru and Pt nanoparticles deposited on FeCO₃ could catalyze the hydrogen conversion of CO₃²⁻ to produce C₁–C₂₆ hydrocarbons with a selectivity for C₂⁺ hydrocarbons of 42.1% and a time-yield of 863 mmol_CH2 mol_metal⁻¹ h⁻¹. Based on these control experimental results, it can be concluded that in Ru–Pt/FeCO₃, Ru nanoparticles were mainly responsible for the reduction of CO₃²⁻ to form species, while Pt nanoparticles played an important role in catalyzing the coupling of produced on themselves, or migrated from the attached or adjacent Ru nanoparticles, to form higher hydrocarbons. The synergistic effect between Pt and Ru nanoparticles supported on FeCO₃ is responsible for the formation of hydrocarbons with very long carbon chains over Ru–Pt/FeCO₃ at low temperatures.

A FeCO₃ supported PtRu alloy nanocluster catalyst (PtRu/FeCO₃, Ru/Pt atomic ratio = 2.5) was prepared with the described method by replacing the Pt or Ru nanoclusters with PtRu alloy nanoclusters²¹ and used as a catalyst for C-CTHC. It was found that the time-yield of hydrocarbons over PtRu/FeCO₃ (80 mmol_CH2 mol_metal⁻¹ h⁻¹) was much lower than that over Ru–Pt/FeCO₃, and the carbon chain of higher hydrocarbons generated over PtRu/FeCO₃ (C₂–C₁₁) was much shorter than that generated over Ru–Pt/FeCO₃ (C₂–C₂₆), implying that the FeCO₃ supported PtRu alloy nanocluster is not as efficient as Ru–Pt/FeCO₃ in catalyzing C-CTHC (Table S2, ESI†).

A control catalyst composed of Ru and Pt nanoclusters deposited on TiO₂ (Ru–Pt/TiO₂) was prepared (see the ESI†) and used for the hydrogenation of CO₂. Ru–Pt/TiO₂ could catalyze the reaction of CO₂ with H₂ in a mixture of water and cyclohexane at 130 °C to produce C₁–C₄ hydrocarbons and C₁–C₂ alcohols, with selectivities of CH₄ and higher hydrocarbons of 97.9% and 1.2%, respectively, demonstrating the important role of FeCO₃ as a functional support in the conversion of CO₂ and H₂ to long-chain hydrocarbons under mild conditions (Table S3, ESI†).

Fe₃O₄ supported Ru and Pt nanoparticles (Ru–Pt/Fe₃O₄) were also prepared for comparison. It was found that Ru–Pt/Fe₃O₄ could catalyze the conversion of CO₂ and H₂ to higher hydrocarbons (C₂–C₂₆) with a selectivity of 11.3%, which supported the proposed mechanism of carbonation reaction of CO₂ with surface iron species in C-CTHC.

Cyclohexane in the reaction system played an important role in generating long-chain hydrocarbons in C-CTHC. When the reaction was conducted in water at 130 °C, hydrocarbons with more than 7 carbon atoms were not detected in the products (Table S4, ESI†). As a hydrophobic solvent, cyclohexane could dissolve the formed long-chain hydrocarbons, promoting its desorption from the catalytic sites. Furthermore, cyclohexane covered on the catalyst might efficiently decrease the probability of carbon-chain termination with OH or H.

The stability of Ru–Pt/FeCO₃ in C-CTHC was tested at 130 °C in a mixture of 1.25 MPa (25 °C) of CO₂ and 3.75 MPa (25 °C) of H₂ (see the ESI†). The conversion of carbon in the substrates over the catalyst in 8 hours maintained at 4.8 ± 0.3% in the three cycles, and the selectivity for C₂⁺ hydrocarbons and C₅⁺ hydrocarbons maintained at 40.0 ± 2.0% and 24.0 ± 2.0%, respectively, implying that the catalyst possessed good stability under the reaction conditions (Fig. S9, ESI†).

Conclusions

This work establishes a new route for the synthesis of higher hydrocarbons (C₂–C₂₆), which is characterized by a reaction of ferrous carbonate and hydrogen to form the hydrocarbons, coupled with carbonation of formed ferrous species by carbon dioxide. The low activation energy for the carbon–carbon coupling reaction in C-CTHC led to the formation of the higher hydrocarbons at low or even ambient temperatures, exhibiting advantages of C-CTHC in synthesis of multi-carbon compounds from CO₂ under mild conditions. The synergistic effect between Pt and Ru nanoparticles on FeCO₃ played an important role in generating higher hydrocarbons.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors express their thanks for the support from the NSFC (grants 21573010 and 21821004) and the Chinese Ministry of Science and Technology (2016YFE0118700).

Notes and references

1. Handbook of Mineralogy, ed. J. W. Anthony, R. A. Bideaux, K. W. Bladh and M. C. Nichols, Mineralogical Society of America, Chantilly, VA 20151-1110, USA, 2003.
2. I. S. Alcala, F. Bellon, M. C. del Campillo, V. Barron and J. Torrent, Sci. Hortic., 2012, 138, 17–23.
3. J. J. Wang, G. N. Li, Z. L. Li, C. Z. Tang, Z. C. Feng, H. Y. An, H. L. Liu, T. F. Liu and C. Li, Sci. Adv., 2017, 3, e1701290.
4. O. Martin, A. J. Martin, C. Mondelli, S. Mitchell, T. F. Segawa, R. Hauert, C. Drouilly, D. Curulla-Ferre and J. Perez-Ramirez, Angew. Chem., Int. Ed., 2016, 55, 6261–6265.
5. Z. H. He, Q. L. Qian, J. Ma, Q. L. Meng, H. C. Zhou, J. L. Song, Z. M. Liu and B. X. Han, Angew. Chem., Int. Ed., 2016, 55, 737–741.
6. Y. Wang, Z. Z. Zhang, L. N. Zhang, Z. B. Luo, J. N. Shen, H. X. Lin, J. L. Long, J. S. Wu, X. Z. Fu, X. X. Wang and C. Li, J. Am. Chem. Soc., 2018, 140, 14595–14598.
7. J. Li, Y. L. He, L. Tan, P. P. Zhang, X. B. Peng, A. Oruganti, G. H. Yang, H. Abe, Y. Wang and N. Tsubaki, Nat. Catal., 2018, 1, 787–793.
8. F. Jiao, J. J. Li, X. L. Pan, J. P. Xiao, H. B. Li, H. Ma, M. M. Wei, Y. Pan, Z. Y. Zhou, M. R. Li, S. Miao, J. Li, Y. F. Zhu, D. Xiao, T. He, J. H. Yang, F. Qi, Q. Fu and X. H. Bao, Science, 2016, 351, 1065–1067.
9. Y. L. Yu, J. Huang and Y. Wang, ChemCatChem, 2018, 10, 4863–4867.
10. J. L. Zhang, Y. Wang, H. Ji, Y. G. Wei, N. Z. Wu, B. J. Zuo and Q. L. Wang, J. Catal., 2005, 229, 114–118.
11. M. H. Liang, X. D. Wang, H. Q. Liu, H. C. Liu and Y. Wang, J. Catal., 2008, 255, 335–342.
12. Y. Wang, J. W. Ren, K. Deng, L. L. Gui and Y. Q. Tang, Chem. Mater., 2000, 12, 1622–1627.
13. C. Lian, Y. L. Yu, K. Zhang, A. Gao and Y. Wang, Catal. Sci. Technol., 2018, 8, 1528–1534.
14. L. S. Zhong, J. S. Hu, H. P. Liang, A. M. Cao, W. G. Song and L. J. Wan, Adv. Mater., 2006, 18, 2426–2431.
15. Z. Y. Sun, Z. M. Liu, B. X. Han, Y. Wang, J. M. Du, Z. L. Xie and G. J. Han, Adv. Mater., 2005, 17, 929–932.
16. C. Lian, K. Zhang and Y. Wang, Acta Phys.-Chim. Sin., 2017, 33, 984–992.
17. J. K. Heuer and J. K. Stubbins, Corros. Sci., 1999, 41, 1231–1243.
18. T. Yamashita and P. Hayes, Appl. Surf. Sci., 2008, 254, 2441–2449.
19. U. Rodemerck, M. Holena, E. Wagner, Q. Smejkal, A. Barkschat and M. Baerns, ChemCatChem, 2013, 5, 1948–1955.
20. M. D. Porosoff, B. H. Yan and J. G. Chen, Energy Environ. Sci., 2016, 9, 62–73.
21. L. W. Zhang, A. Gao, Y. Liu, Y. Wang and J. T. Ma, Electrochim. Acta, 2014, 132, 416–422.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9se00612e

‡ Equally contributed.

---