One-step synthesis of cubic pyrite-type CoSe₂ at low temperature for efficient hydrogen evolution reaction

Abstract

Cubic pyrite-type CoSe₂ was synthesized by a one-step method, via the reaction between drop casted CoCl₂ and Se vapour. The as-grown CoSe₂ nanoparticles exhibited high activity toward hydrogen evolution reaction with a low Tafel slope of ∼40 mV per decade, accompanied by a large current density and excellent durability.

---

Tremendous efforts have been devoted to the efficient generation of hydrogen through electrolysis or photoelectrochemical reaction.^1–9 Platinum is the most famous catalyst for the hydrogen evolution reaction (HER), however, the large scale application of Pt has been limited by its high cost and low abundance, so developing HER catalysts based on cheap and earth abundant elements has been actively pursued.^10–16 Recently, the transition metal dichalcogenides (MX₂, where typically M = Mo, W, Fe, Co, or Ni and X = S or Se) have received considerable attention as cheap and earth abundant catalysts for HER.^17–22 Among these electrocatalysts, cubic pyrite-structure MX₂, without needing any special modification or treatment, has shown the high intrinsic activity and excellent durability toward the HER possibly attributed to the unique crystal structure, where the metal atoms are octahedrally bonded to adjacent chalcogen atoms.^{17,20,23–26} Cubic pyrite-structure CoSe₂, exhibited the best HER activity among that of MX₂ with a Tafel slope of ∼40 mV per decade.^17,20

CoSe₂, as a well-known oxygen reduction reaction catalyst, has been synthesized by hydrothermal synthesis method.^27–30 It is noteworthy that cubic pyrite-type^27,28 and orthorhombic marcasite-type CoSe₂^29,30 under the similar reaction conditions have been both reported, and there is seldom document about the manipulation of the crystal structure of CoSe₂ in the hydrothermal process. In fact, these two phases have the similarity structural and the small lattice mismatch resulting in the easy epitaxial growth between them,³¹ so it is very challenging to prepare the preferred cubic pyrite-type CoSe₂ in the hydrothermal process. Recently, Kong et al.¹⁷ reported that cubic pyrite-type CoSe₂ nanoparticles were synthesized by converting e-beam evaporated Co thin films through a selenization reaction at 500 °C. Recently, Kong et al. further reported a two-step process to synthesize pyrite-type CoSe₂ on carbon fiber paper, with a small fraction of marcasite-structure CoSe₂, through reacting Se with cobalt oxide, which was prepared from the pyrolysis of cobalt(II) nitrate (500 °C).²⁰ Faber et al. synthesized pyrite-type CoS₂, the CoSe₂ analog, through the sulfidation of cobalt hydroxide carbonate hydrate at 500 °C.²⁴ Moreover, as we know,^12,32,33 for large-scale applications in hydrogen energy, HER catalysts should be prepared in scalable processes that are much less energy intensive and costly. Thus, it is desirable to explore easy and scalable approach for the synthesis of cubic pyrite-type CoSe₂.

In this letter, for the first time, we reported one step synthesis of pure cubic pyrite-type CoSe₂, i.e. polycrystalline trogtalite, by reacting drop casted amorphous CoCl₂ coating with Se vapour. We would like to highlight that this method is facile, environmental friendly (use ethanol as solvent), and scalable. The microscopic and structural characterizations were comprehensively conducted. The electrocatalytic activity toward HER of pyrite-type CoSe₂ was in investigated.

Typically, the CoSe₂ nanoparticles on graphite disks (GD) were synthesized through a new reaction, i.e. reacting Se vapour with CoCl₂ at low temperature of 350 °C as shown in Fig. 1A. The experimental methods are discussed in detail in the ESI.† Briefly, CoCl₂·6H₂O was coated on the graphite through the drop casting method using ethanol as solvent as illustrated in Fig. 1A, and a typical loading of CoSe₂ on graphite was 2.8 mg cm⁻². Note that CoCl₂·6H₂O is stable to air and water, moreover, it can be soluble in many solvents. The GD substrate covered with CoCl₂·6H₂O was located in a home-built tube reactor. When the temperature increased to 350 °C, H₂O in CoCl₂·6H₂O would be vaporized as confirmed in Fig. S1A (ESI†), Then the reaction between CoCl₂ and Se vapour was achieved to form CoSe₂ at 350 °C. The reaction is:¹⁸

CoCl₂ (s) + 2Se (g) → CoSe₂ (s) + Cl₂ (g) (1)

Fig. 1 Synthesis of CoSe₂ on GD. (A) Schematic illustration of synthesis of CoSe₂/GD through reacting CoCl₂ with Se vapour. (B) A representative SEM image of the original GD. (C) A typical SEM image of CoSe₂/GD synthesized at 350 °C showing that the surface of GD was covered by CoSe₂ particles.

This reaction is possibly thermodynamically favourable because crystal CoSe₂ is very stable at high temperature, and the reaction generates gaseous products further driving the equilibrium to the right-hand side. Moreover, CoCl₂ was amorphous, as confirmed in Fig. S1B (ESI†), which would decrease the energy of the reaction needed. Thus, the reason of easy and low-temperature formation of cubic pyrite-type CoSe₂ might be attributed to the thermodynamically favourable reaction between amorphous CoCl₂ and Se vapour.

First, we investigated the morphology of CoSe₂ samples grown at 350 °C by SEM as shown in Fig. 1C. It was seen that the particle size of CoSe₂ on GD was about 50–100 nm and uniform. It was interestingly observed that the morphology of CoSe₂ particles was strongly influenced by the synthesis temperature. CoSe₂ particles synthesized at 300 °C were hard to observe because of the aggregation of CoSe₂ particles as shown in Fig. S2A (ESI†). CoSe₂ synthesized at 400 °C were composed of large particles (>500 nm) as shown in Fig. S2B (ESI†).

The phase identity of CoSe₂/GD products was determined using powder X-ray diffraction (XRD) as shown in Fig. 2A. A broad diffraction peak appeared at 2θ ≈ 26.28°, and 2θ ≈ 42.64–44.60° mainly corresponding to graphite 2H (ICDD PDF 75-1621) contributed from the graphite disk support. According to the ICDD PDF 88-1712 (no. 205) card, the XRD pattern showed an excellent agreement with the main peaks of the standard spectrum diagram. The XRD patterns show four peaks located at 2θ ≈ 30.5, 34.3, 37.6 and 51.8°, which correspond to the (2 0 0), (2 1 0), (2 1 1) and (3 1 1) planes of cubic pyrite-type CoSe₂, i.e. trogtalite, respectively. Thus, the peaks of CoSe₂ products can be indexed to pyrite-type CoSe₂ (space group: Pa3, a = 5.87 Å). The marcasite-type CoSe₂ (space group Pmnn, a = 3.60 Å, b = 4.84 Å, c = 5.72 Å) was not detected in the PXRD pattern. In order to further confirm the crystal structure of the as-synthesised CoSe₂, transmission electron microscopy (TEM) was used to characterize it as shown in Fig. 2B. The TEM image shows a clear crystal lattice, which reveals the resolved lattice fringes of CoSe₂ (211) plans with a spacing of 2.39 Å as shown in Fig. 2B. Moreover, as shown in Fig. 2B (inset), the fast Fourier transform (FFT) image further reveals the CoSe₂ herein is the cubic pyrite-type phase. It was noteworthy that the crystal structure of CoSe₂ synthesized at 500 °C was also pyrite-type CoSe₂ as shown in Fig. S5 (ESI†). Raman spectrum of the as-obtained CoSe₂ on GD was shown in Fig. 2C. The sharp peak at 173 cm⁻¹ was assigned to the Se–Se stretching mode of cubic CoSe₂;³⁴ the peak at 187 cm⁻¹ is associated with cubic CoSe₂;³⁵ the broad peak at 250 cm⁻¹ should be attributed to the tiny amorphous Se.^20,36,37 These results further verified the composition and structure of the as-grown CoSe₂ on GD. It was well-documented that the activity of HER catalyst was expected to be sensitive to the valence state and coordination environment of the metal centre. So, we revealed the electron-binding energies of Co and Se in cubic pyrite-type CoSe₂ as shown in Fig. 2D and E. It was seen that the binding energies of Co 2p_3/2 and Co 2p_1/2 were observed corresponding to 778.4 eV and 793.7 eV, respectively.^20,38–41 Se 3d_5/2 and Se 3d_3/2 located at 54.6 and 55.3 eV, indicating the oxidation state of −2 for Se.^42,43 In contrast, the binding energies of Co 2p_3/2 and Co 2p_1/2 of elemental Co are 778.5 eV⁴⁴ and 793.3 eV,⁴⁵ respectively. For elemental Se, the binding energy of Se 3d_5/2 is 55.5 eV.⁴⁶ Thus, these results illustrated the formation of cubic pyrite-type CoSe₂ through the reaction of CoCl₂ and Se vapour.

Fig. 2 Structural characterizations of CoSe₂ in CoSe₂/GD. (A) XRD patterns of CoSe₂/GD synthesized at different temperatures and schematic crystal structure of cubic pyrite-type CoSe₂ (inset). (B) High-resolution TEM image and a typical fast Fourier transform (FFT) image of an as-grown CoSe₂ (inset). (C) Raman spectra of CoSe₂/GD. (D and E) were XPS spectra of Co 2p region and Se 3d region, respectively.

The electrocatalytic activities of various CoSe₂/GD samples were evaluated with a three-electrode electrochemical cell in 0.5 M H₂SO₄ solution purged with H₂ (g) (experimental details in the ESI†). We investigated the effect of the amount of CoSe₂ loadings (0.01–4.0 mg cm⁻²) in CoSe₂/GD on the HER performances as shown in Fig. S6 (ESI†). It was observed that the best CoSe₂ loading on GD toward HER was 2.8 mg cm⁻². We comprehensively investigated the effect of reaction temperature of CoSe₂/GD samples with a CoSe₂ loading of 2.8 mg cm⁻² on their HER performances. Polarization curves after iR corrected showing the geometric current density (J) plotted against the applied potential of CoSe₂/GD samples synthesized at different temperatures were exhibited in Fig. 3A and B. As expected, the bare GD electrode was totally inactive towards HER. It was interesting to observe the high current densities (J_cathodic > 200 mA cm⁻²) under low applied overpotentials on all samples. Moreover, the reaction temperature had a strong influence on the HER performance of CoSe₂/GD, and CoSe₂/GD sample synthesized at 350 °C exhibited the best HER activity. The superior HER activities of CoSe₂ synthesised at 350 °C were further illustrated by comparing their Tafel slopes as shown in Fig. 3C. It was seen that the Tafel slopes of samples grown at 300, 350, and 400 °C were 52.4, 42.2, and 46.3 mV per decade, respectively, showing the same trend as above.

Fig. 3 Electrocatalytic performance toward HER of CoSe₂/GD samples by different temperatures. Polarization curves at (A) higher and (B) lower applied potentials. (C) Tafel analysis of the data presented in (A and B). (D) Nyquist plots showing the electrode kinetics of samples. (E) Cyclic voltammograms recorded for a CoSe₂/GD electrode at various scan rates to determine the double layer capacitance (C_dl). (F) Long-term stability measurement for a representative CoSe₂/GD sample at 350 °C, demonstrating the small change in the overpotential required to maintain a continuous catalytic current density of J_cathodic = 10 mA cm⁻² for 24 h.

Three possible reaction steps have been suggested for the HER in acidic media, commonly named the Volmer (eqn (2)), Heyrovsky (eqn (3)), and Tafel reactions (eqn (4)).⁴⁷

H₃O⁺ + e⁻ → H_ads + H₂O (2)

H_ads + H₃O⁺ + e⁻ → H₂ + H₂O (3)

H_ads + H_ads → H₂ (4)

When the Volmer reaction is the rate-determining step of the HER, a slope of ∼120 mV per decade should be observed. Whereas if Heyrovsky or Tafel reactions is the rate-limiting step, Tafel slopes of ∼30 and ∼40 mV per decade can be obtained, respectively.⁴⁸ In this study, the observed Tafel slope of 42.2 mV per decade indicated that the Volmer–Heyrovsky HER mechanism (eqn (2) and (3)) is operative in the HER catalyzed by CoSe₂/GD.¹³ Five experiments were performed to verify the data accuracy, the Tafel slopes were 43.3, 42.6, 42.7, 42.2, 42.2 mV per decade, respectively, and the corresponding standard deviation was 0.45. It is well known that the Tafel slope is considered as an inherent property of the HER catalyst, so the small Tafel slope of ∼40 mV per decade in this study is expected to be more beneficial for practical applications, because it can provide a remarkably increased HER rate with the moderate increase of applied overpotential.

Additionally, electrochemical impedance spectroscopy (EIS) was used to investigate the kinetics on CoSe₂ under the catalytic HER operating conditions as shown in Fig. 3D. The charge transfer resistance (R_ct) associated with H₂ (g) evolution for each electrode was extracted by modelling the EIS data using a simplified Randles equivalent circuit (ESI†). R_ct for CoSe₂/GD grown at 300, 350 and 400 °C were 3.6, 8.8, 16.2 Ω, respectively. The EIS data confirmed that the superior activity CoSe₂/GD grown at 350 °C was possibly attributed to the facile kinetics toward H₂ (g) evolution. We also measured the capacitive current for CoSe₂/GD samples grown at 300 (Fig. S7A (ESI†)), 350 (Fig. 3E), and 400 °C (Fig. S7B (ESI†)) as a function of scan rate to extract the double layer capacitance (C_dl) of each electrode, which serves as an estimate of the effective electrochemically active surface area of the solid–liquid interface. The C_dl of CoSe₂/GD grown at 300, 350, and 400 °C is 3.4, 5.5 and 4.5 mF cm⁻², respectively. Even though these C_dl values exhibited differences, it was unlikely that such a small variation in C_dl accounted for the obvious differences in observed performance (Fig. 3A and B), although the morphologies of CoSe₂/GD grown at 300 (Fig. S2A (ESI†)), 350 (Fig. 1B), and 400 °C (Fig. S2B (ESI†)) were different. Consequently, the differences of HER performance of CoSe₂/GD grown at 300, 350, and 400 °C were mainly ascribed to the different intrinsic activity, i.e. the different kinetics toward H₂ evolution.

Furthermore, besides the electrocatalytic activity, it is well-established that the stability is also a crucial factor to develop HER catalyst. To evaluate the stability, long term durability of CoSe₂/GD grown at 350 °C was assessed by a 24 h stable current measurement as shown in Fig. 3F. It was seen that over the duration of the 24 h stability measurement, the cathodic overpotential required to maintain J_cathodic = 10 mA cm⁻² was inherently stable, which just increased only about 30 mV. It should be noted that the increasing of cathodic overpotential to maintain J_cathodic = 10 mA cm⁻² mainly occurred at the beginning of the stability test (0–12 h); from 12 h to 24 h, the cathodic overpotential to maintain J_cathodic = 10 mA cm⁻² slightly increased only by 3 mV.

Lastly, we attempt to compare the performance of our CoSe₂/GD with that synthesized by two-step process.²⁰ CoSe₂/GD grown at 350 °C exhibited pretty low onset overpotential of −135 mV. Generally, the onset overpotential depends on several factors, such as catalyst loading amount, intrinsic activity, and purity. Considering the similar intrinsic activity of samples by one-step and two-step (a Tafel slope of ∼40 mV per decade), the lower onset overpotential in this study was possibly attributed to the high-purity crystal phase of pyrite-type CoSe₂ and higher loading of CoSe₂. Moreover, it required a applied overpotential of −193 mV to obtain the significant H₂ (g) evolution (J_cathodic = 10 mA cm⁻²). These values were similar to that of CoSe₂ converted form CoO on carbon paper.²⁰ Thus, it was concluded that our simple method can produce CoSe₂ catalysts with competitive HER performance.

In summary, we reported a one-step method to synthesize pure cubic pyrite-type CoSe₂ electrocatalysts toward HER via a reaction between drop casted CoCl₂ and Se vapour. The structural characterization and electrochemical HER activities of series of CoSe₂/GD samples synthesized at different temperatures were systemically investigated. The CoSe₂ electrocatalyst on GD showed excellent HER activity with a minimum Tafel slope of ∼40 mV per decade. This synthesis approach allows facile preparation of advanced CoSe₂ HER electrocatalysts at large scale. This convenient and scalable synthesis method demonstrates a potentially versatile and low-cost strategy to make other transition metal chalcogenides.

Acknowledgements

This research is supported by NSFC (21276231, 21476201, and U1162128).

Notes and references

1. D. Merki and X. L. Hu, Energy Environ. Sci., 2011, 4, 3878–3888.
2. L. S. Li, Y. H. Yu, F. Meng, Y. Z. Tan, R. J. Hamers and S. Jin, Nano Lett., 2012, 12, 724–731.
3. Y. J. Feng, T. He and N. Alonso-Vante, Fuel Cells, 2010, 10, 77–83.
4. W. F. Chen, J. T. Muckerman and E. Fujita, Chem. Commun., 2013, 49, 8896–8909.
5. A. B. Laursen, P. C. K. Vesborg and I. Chorkendorff, Chem. Commun., 2013, 49, 4965–4967.
6. Z. Y. Lu, H. C. Zhang, W. Zhu, X. Y. Yu, Y. Kuang, Z. Chang, X. D. Lei and X. M. Sun, Chem. Commun., 2013, 49, 7516–7518.
7. Y. Xu, R. Wu, J. F. Zhang, Y. M. Shi and B. Zhang, Chem. Commun., 2013, 49, 6656–6658.
8. Y. Yan, B. Y. Xia, X. Y. Qi, H. B. Wang, R. Xu, J. Y. Wang, H. Zhang and X. Wang, Chem. Commun., 2013, 49, 4884–4886.
9. A. B. Laursen, S. Kegnæs, S. Dahl and I. Chorkendorff, Energy Environ. Sci., 2012, 5, 5577–5591.
10. J. R. McKone, S. C. Marinescu, B. S. Brunschwig, J. R. Winkler and H. B. Gray, Chem. Sci., 2014, 5, 865–878.
11. S. P. Berglund, H. C. He, W. D. Chemelewski, H. Celio, A. Dolocan and C. B. Mullins, J. Am. Chem. Soc., 2014, 136, 1535–1544.
12. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6446–6473.
13. E. J. Popczun, J. R. McKone, C. G. Read, A. J. Biacchi, A. M. Wiltrout, N. S. Lewis and R. E. Schaak, J. Am. Chem. Soc., 2013, 135, 9267–9270.
14. E. J. Popczun, C. G. Read, C. W. Roske, N. S. Lewis and R. E. Schaak, Angew. Chem., Int. Ed., 2014, 53, 5427–5430.
15. Q. Ding, F. Meng, C. R. English, M. Cabán-Acevedo, M. J. Shearer, D. Liang, A. S. Daniel, R. J. Hamers and S. Jin, J. Am. Chem. Soc., 2014, 136, 8504–8507.
16. Q. Liu, J. Tian, W. Cui, P. Jiang, N. Cheng, A. M. Asiri and X. Sun, Angew. Chem., Int. Ed., 2014, 53, 6710–6714.
17. D. S. Kong, J. J. Cha, H. T. Wang, H. R. Lee and Y. Cui, Energy Environ. Sci., 2013, 6, 3553–3558.
18. M. A. Lukowski, A. S. Daniel, F. Meng, A. Forticaux, L. S. Li and S. Jin, J. Am. Chem. Soc., 2013, 135, 10274–10277.
19. J. Bonde, P. G. Moses, T. F. Jaramillo, J. K. Nørskov and I. Chorkendorff, Faraday Discuss., 2008, 140, 219–231.
20. D. Kong, H. Wang, Z. Lu and Y. Cui, J. Am. Chem. Soc., 2014, 136, 4897–4900.
21. M. A. Lukowski, A. S. Daniel, C. R. English, F. Meng, A. Forticaux, R. J. Hamers and S. Jin, Energy Environ. Sci., 2014, 7, 2608–2613.
22. T. F. Jaramillo, K. P. Jorgensen, J. Bonde, J. H. Nielsen, S. Horch and I. Chorkendorff, Science, 2007, 317, 100–102.
23. M. Caban-Acevedo, M. S. Faber, Y. Z. Tan, R. J. Hamers and S. Jin, Nano Lett., 2012, 12, 1977–1982.
24. M. S. Faber, R. Dziedzic, M. A. Lukowski, N. S. Kaiser, Q. Ding and S. Jin, J. Am. Chem. Soc., 2014, 136, 10053–10061.
25. A. Ivanovskaya, N. Singh, R. F. Liu, H. Kreutzer, J. Baltrusaitis, T. V. Nguyen, H. Metiu and E. McFarland, Langmuir, 2013, 29, 480–492.
26. M. S. Faber, K. Park, M. Cabán-Acevedo, P. K. Santra and S. Jin, J. Phys. Chem. Lett., 2013, 4, 1843–1849.
27. M. R. Gao, X. Cao, Q. Gao, Y. F. Xu, Y. R. Zheng, J. Jiang and S. H. Yu, ACS Nano, 2014, 8, 3970–3978.
28. J. L. Yang, X. P. Shen, Z. Y. Ji and G. X. Zhu, J. Mater. Sci., 2013, 48, 7913–7919.
29. Y. J. Feng, T. He and N. Alonso-Vante, Fuel Cells, 2010, 10, 77–83.
30. W. X. Zhang, Z. H. Yang, J. W. Liu, Z. H. Hui, W. C. Yu, Y. T. Qian, G. E. Zhou and L. Yang, Mater. Res. Bull., 2000, 35, 2403–2408.
31. R. Sun, M. K. Y. Chan and G. Ceder, Phys. Rev. B, 2011, 83, 235311–235322.
32. P. P. Edwards, V. L. Kuznetsov and W. I. David, Philos. Trans. R. Soc., A, 2007, 365, 1043–1056.
33. K. Mazloomi and C. Gomes, Renewable Sustainable Energy Rev., 2012, 16, 3024–3033.
34. K. Kodama, R. Jinnouchi, T. Suzuki, T. Hatanaka and Y. Morimoto, Electrochim. Acta, 2012, 78, 592–596.
35. L. Zhu, M. Teo, P. C. Wong, K. C. Wong, I. Narita, F. Ernst, K. A. R. Mitchell and S. A. Campbell, Appl. Catal., A, 2010, 386, 157–165.
36. H. Wang, Z. Lu, S. Xu, D. Kong, J. J. Cha, G. Zheng, P. C. Hsu, K. Yan, D. Bradshaw, F. B. Prinz and Y. Cui, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 19701–19706.
37. V. V. Poborchii, A. V. Kolobov and K. Tanaka, Appl. Phys. Lett., 1998, 72, 1167–1169.
38. W. Huang, Z. Zuo, P. Han, Z. Li and T. Zhao, J. Electron Spectrosc. Relat. Phenom., 2009, 173, 88–95.
39. M.-R. Gao, S. Liu, J. Jiang, C.-H. Cui, W.-T. Yao and S.-H. Yu, J. Mater. Chem., 2010, 20, 9355–9361.
40. M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M. Lau, A. R. Gerson and R. S. C. Smart, Appl. Surf. Sci., 2011, 257, 2717–2730.
41. J. Yang, G.-H. Cheng, J. H. Zeng, S. H. Yu, X. M. Liu and Y.-T. Qian, Chem. Mater., 2001, 13, 848–853.
42. Z. Zhang, S. Pang, H. Xu, Z. Yang, X. Zhang, Z. Liu, X. Wang, X. Zhou, S. Dong, X. Chen, L. Gu and G. Cui, RSC Adv., 2013, 3, 16528–16533.
43. N. D. Boscher, C. J. Carmalt and I. P. Parkin, J. Mater. Chem., 2006, 16, 122–127.
44. A. Mandale, S. Badrinarayanan, S. Date and A. Sinha, J. Electron Spectrosc. Relat. Phenom., 1984, 33, 61–72.
45. J. Bonnelle, J. Grimblot and A. D'huysser, J. Electron Spectrosc. Relat. Phenom., 1975, 7, 151–162.
46. M. Shenasa, S. Sainkar and D. Lichtman, J. Electron Spectrosc. Relat. Phenom., 1986, 40, 329–337.
47. B. E. Conway and B. V. Tilak, Electrochim. Acta, 2002, 47, 3571–3594.
48. Y. G. Li, H. L. Wang, L. M. Xie, Y. Y. Liang, G. S. Hong and H. J. Dai, J. Am. Chem. Soc., 2011, 133, 7296–7299.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed experimental procedures, TGA and SEM images, XRD, EDS, XPS spectra, and other additional graphs. See DOI: 10.1039/c4ra07876d

---