Formation of Ti–Fe mixed sulfide nanoboxes for enhanced electrocatalytic oxygen evolution

Jianwei Nai b, Yan Lu b and Xin-Yao Yu *a
aSchool of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, P. R. China. E-mail:
bSchool of Chemical and Biomedical Engineering, Nanyang Technological University, 62 Nanyang Drive, Singapore 637459, Singapore

Received 13th March 2018 , Accepted 31st May 2018

First published on 8th June 2018

Exploring effective electrocatalysts is always required to boost the efficiency of water splitting. In this work, we report a self-template strategy to synthesize Ti–Fe mixed sulfide nanoboxes for the electrocatalytic oxygen evolution reaction (OER). The mixed sulfide nanoboxes are derived from corresponding Prussian blue analog nanoboxes, which are first obtained by hollowing the nanocube precursor with a solvothermal-engaged etching process. The resulting Ti–Fe mixed sulfide nanoboxes perform as a promising OER electrocatalyst with enhanced activity, affording a current density of 10 mA cm−2 and a high mass activity of 100 A g−1 at a small overpotential of 350 mV, a Tafel slope as small as 55 mV dec−1, and excellent stability in alkaline medium.

Electrochemical water splitting is an environmentally friendly alternative to obtain clean hydrogen fuel from renewable energy sources to address the urgent need for clean and sustainable energy.1 Though the generation of hydrogen is the primary objective of water electrolysis, we cannot neglect the counter-reaction (the oxygen evolution reaction, OER), as it is the sluggish one that affects the faradaic efficiency of the electrolytic cell to a greater extent.2 Exploiting effective OER electrocatalysts is always required to enhance the energy conversion efficiency. The benchmark catalysts for the OER have been Ir and Ru based compounds so far. However, the use of such relatively less abundant precious metals for corrosive and destructive water electrolysis is not advisable as it increases the cost of H2 production and hinders the magnification of productivity on a large scale. One solution to this issue is to reduce the used amount of the noble metal, such as single-atom catalysts.3,4 On the other hand, the rapid development of transition-metal based catalysts that are abundant, active, and stable has promised to be another alternative to reduce costs.5–7 Apart from the well-developed oxides/(oxy)hydroxides,8–11 phosphides,12,13 and selenides,14–16 metal sulfides have recently also provided access to newly advanced electrocatalysts with high activity towards the OER.17 Generally, metal sulfides usually possess good electronic conductivity, and mechanical and thermal stability.18,19 Recently, pyrites (FeS2) and pyrite-type materials are attracting great interest for application in electrocatalysis.20 Unfortunately, the development of such materials is still far from satisfaction. A main obstacle needs to be overcome for the use of pyrite-type materials as electrocatalysts is to exploit novel and facile approaches to make high-quality materials at the nanoscale.20 On the other hand, the ability to explore the electrochemical performance of titanium sulfide nanomaterials is also limited due to the daunting synthetic challenge (e.g., high temperature or pressure),21,22 although they have shown good electronic properties.23,24 Compared to monometal sulfides, mixed metal sulfides could even demonstrate richer redox reactions and higher electronic conductivity, leading to a significant enhancement of the electrochemical performance.25 Therefore, it can be expected that the combination of pyrite and titanium sulfide may not only enhance the materials performance but also hold promise to assess the role of titanium sulfides in the electrocatalytic OER, which has never been investigated so far.

Prussian blue and its analogs (PBAs) are well known as hexacyanometalate compounds or a kind of metal–organic framework due to their three-dimensional framework structures.26 These materials are represented by the general AxMA[MB(CN)6]y·zH2O formula, in which A refers to the cations hosted in the framework channels, while MA and MB are the transition metals linked by the cyano ligands. With these structural features and abundant chemical compositions, PBA materials demonstrate unique stimulus-triggered electronic properties,27–29 and thus have wide applications in biomedicine,30 catalysis,31 sensors,32 and so on. Recently, employing PBA nanomaterials as the versatile precursors to fabricate functional inorganic compounds becomes intriguing.33 This is because the components of the final products are designable by deliberate selection of the chemical composition in the PBA precursors and careful manipulation of the chemical conversion processes. As a result, metal oxides,34–37 hydroxides,38 sulfides,39,40 selenides,16 phosphides,12,41 and carbides42 with well-defined nanostructures can be constructed for electrochemical energy-related applications. Nevertheless, the involved metals are limited to Fe, Co, and Ni in most cases. The synthesis of novel PBA-derived nanostructures with uncommon elements is therefore not only important for a wide range of applications but also significant for developing the modern synthetic methodology.

Herein, we report a facile strategy to synthesize mixed metal sulfide nanoboxes derived from PBA materials for the electrocatalytic OER. The overall scheme for the synthetic strategy can be seen in Fig. 1. Briefly, Ti–Fe PBA nanocubes are first prepared, followed by a facile solvothermal process with dimethylformamide (DMF) as the solvent to obtain the Ti–Fe PBA nanoboxes. After sulfuration of the Ti–Fe PBA nanoboxes by a given S source, the Ti–Fe mixed sulfide nanoboxes can be formed in this self-template approach, showing enhanced electrocatalytic performance towards the OER.

image file: c8ta02334d-f1.tif
Fig. 1 Schematic illustration of the formation of Ti–Fe mixed sulfide nanoboxes.

X-ray diffraction (XRD) and energy-dispersive X-ray spectroscopy (EDX) are employed to characterize the crystal structure and composition, respectively, of the as-prepared Ti–Fe PBA nanocubes. The XRD pattern (Fig. S1a, ESI) displays a typical PBA-type face-centered cubic crystal structure. The EDX spectrum (Fig. S1b, ESI) confirms the Ti–Fe combination in the product. Field-emission scanning electron microscopy (FESEM) images show that the Ti–Fe PBA nanocubes are highly uniform with an average size of 150 nm (Fig. S2a and b, ESI). Transmission electron microscopy (TEM) images of the product further reveal that the nanocubes are solid particles (Fig. S2c and d, ESI).

These Ti–Fe PBA nanocubes are then used in a self-template route to produce nanoboxes. Specifically, the solid nanocubes can be readily converted to hollow nanoboxes by a solvothermal approach with only DMF as the solvent but without any other additives. XRD and EDX analyses indicate that the as-treated product is still Ti–Fe PBA but with a different Ti/Fe atomic ratio from that of nanocubes (Fig. S3, ESI). The FESEM image in Fig. 2a shows that the product consists of uniform and high-quality nanoboxes, inheriting well the morphology and dimension of the nanocube templates. The high-magnification FESEM image in Fig. 2b reveals that the surface of the nanoparticles becomes rough after the solvothermal treatment. The shape and construction of the hollow interior of the nanoboxes are further studied by TEM, as displayed in Fig. 2c–e. Highly uniform nanoboxes can be observed from Fig. 2c and d, which is in agreement with the FESEM findings. The walls of the hollow nanoboxes are as thin as about 20 nm in thickness as revealed in Fig. 2e.

image file: c8ta02334d-f2.tif
Fig. 2 (a and b) FESEM and (c–e) TEM images of the Ti–Fe PBA nanoboxes.

Time-dependent microscopy experiments are carried out to study the structural conversion from nanocubes to nanoboxes. The morphology and structure of the intermediate products obtained at different solvothermal stages are detected by the TEM technique (Fig. S4, ESI). The results show that the hollowing process starts from a void formed at the center of the nanocube with a relatively round shape. With the reaction time proceeding, bigger cavities can be observed and eventually enlarge to an obvious cubic shape with a size of 100 nm. As a result, a hollow box-like nanostructure is formed. This inside-out hollowing mechanism should be attributed to the unstable character of the center of the Ti–Fe PBA nanocubes. The center of the nanocubes is formed at the early stage of the reaction and thus have many defects that make it weak and “soft”, and cause it to gradually dissolve during the solvothermal process.37 Ammonia-containing species released from the decomposition of DMF molecules might also help to etch the unstable center of the PBA nanoparticles.16 Surprisingly, the Ti–Fe PBA nanoboxes retain the single crystal characteristic as that of the nanocube precursor, as revealed by the corresponding selected-area electron diffraction patterns (Fig. S5, ESI).

Ti–Fe based sulfide nanoboxes can then be synthesized by sulfurating the Ti–Fe PBA nanoboxes in hydrogen sulfide decomposed from thiourea under the nitrogen gas flow. The XRD result (Fig. S6a, ESI) demonstrates that the product is a sulfide mixture, mainly composed of Ti3S4 (JCPDS card no. 9-294) and FeS2 (JCPDS card no. 42-1340). The EDX spectrum (Fig. S6b, ESI) reveals that the molar ratio of these two components is around 1[thin space (1/6-em)]:[thin space (1/6-em)]5 (Ti3S4[thin space (1/6-em)]:[thin space (1/6-em)]FeS2), hereafter denoted as c-Ti–Fe–S boxes. The formation of Ti–Fe mixed sulfides is further supported by X-ray photoelectron spectroscopy (XPS). The C 1s spectrum confirms that no carbides generated in the product (Fig. S7, ESI). The carbon element detected in the EDX spectrum (Fig. S6b, ESI) might be attributed to organic carbon compounds and amorphous carbon (Fig. S7, ESI). From the Fe 2p3/2 and Ti 2p3/2 spectra, Fe2+ and Fe3+ species, and Ti2+ and Ti4+ species can be detected, respectively (Fig. S8a and b, ESI). Some peaks attributed to oxidized metal species such as FeOx, TiOx, and FeSOx/TiSOx can also be observed in the XPS spectra (Fig. S8, ESI), indicating a certain degree of oxidation on the surface of the product. It can be seen from Fig. 3a and b that the as-formed Ti–Fe mixed sulfides well retain the box-like architecture as that of their Ti–Fe PBA template, but rougher surfaces are observed for the former ones. These rough surfaces are composed of small nanoparticles, which are generated from the PBA-to-sulfide transformation process. TEM images in Fig. 3c–e show uniform nanoboxes with a wall thickness of around 20 nm and small nanoparticles as the building blocks, consistent with the FESEM observations. As determined by N2 adsorption measurements (Fig. S9, ESI), these c-Ti–Fe–S boxes possess a Brunauer–Emmett–Teller (BET) specific surface area of 20 m2 g−1.

image file: c8ta02334d-f3.tif
Fig. 3 (a and b) FESEM and (c–e) TEM images of the c-Ti–Fe–S nanoboxes.

In a control experiment, we convert the Ti–Fe PBA nanocubes to corresponding Ti–Fe mixed sulphide nanostructures by a similar annealing route. XRD and EDX analyses confirm that the product is also a Ti–Fe sulfide mixture (Fig. S10a and b, ESI). FESEM and TEM images show that these Ti–Fe mixed sulfides are still cube-shaped solid nanoparticles (Fig. S10c and d, ESI), hereafter denoted as c-Ti–Fe–S cubes. In another control experiment, Ti–Fe PBA nanoboxes are used as the precursor, and the sulfur source is replaced from thiourea to S powder. In this case, Ti–Fe mixed sulfides are still the composition of the resultant material. However, only FeS2 is crystalline whereas titanium sulfide should be in an amorphous state, as revealed by the XRD and EDX data (Fig. S11a and b, ESI). FESEM and TEM images display that the nanostructure of the materials is similar to that of c-Ti–Fe–S boxes, hereafter denoted as a-Ti–S/c-Fe–S boxes (Fig. S11c and d, ESI).

The electrocatalytic OER performance of c-Ti–Fe–S boxes is investigated in the alkaline solution (1.0 M KOH) in a standard three-electrode system. A glassy carbon electrode modified with the catalyst, an Ag/AgCl electrode and a Pt wire are used as the working electrode, reference electrode, and counter-electrode, respectively. For comparison, the electrochemical properties of the c-Ti–Fe–S cubes and a-Ti–S/c-Fe–S boxes are also studied. Fig. 4a shows the polarization curves recorded by linear sweep voltammetry (LSV) of all three catalysts. It can be seen that the c-Ti–Fe–S boxes afford an obviously lower onset potential and higher current density for the OER than the other two catalysts. To reach a current density of 10 mA cm−2, which is a common criterion to evaluate the OER activity, the c-Ti–Fe–S boxes need a potential of 1.58 V, corresponding to an overpotential of only 350 mV (Fig. 4a). On the other hand, 10 and 70 mV higher overpotentials are required for the c-Ti–Fe–S cubes and a-Ti–S/c-Fe–S boxes, respectively, to achieve the same current density. It should be noted that this activity of the c-Ti–Fe–S boxes is repeatable (Fig. S12, ESI) and higher than that of PB-derived FeS2 nanoparticles and commercial RuO2 (Fig. S13 and S14, ESI). Moreover, the rise of the current density of the c-Ti–Fe–S boxes is much faster than those of other two catalysts. Specifically, the current density of the c-Ti–Fe–S boxes could increase to 50 mA cm−2 at an overpotential of 440 mV, which is 20 mV lower than that of the c-Ti–Fe–S cubes and a-Ti–S/c-Fe–S boxes, to reach the same current density (Fig. 4a). A rotating ring-disk electrode (RRDE) measurement verifies that the observed current of the catalyst (e.g., c-Ti–Fe–S boxes) in Fig. 4a originates from oxygen evolution rather than side reactions with a high faradaic efficiency of around 95.4% (Fig. S15, ESI). Furthermore, the mass activity is another quantitative active parameter used to define the activity of a catalyst. This value is calculated using the formula: mass activity = j/c, where j is the current density at a given overpotential, and c is the concentration of catalysts modified on the electrode (constant 0.1 mg cm−2 in this work). At a given overpotential of 350 mV (Fig. 4b), the mass activity of the c-Ti–Fe–S boxes (100 A g−1) is much higher than that of the c-Ti–Fe–S cubes (78 A g−1), which is even 20-fold higher than that of the a-Ti–S/c-Fe–S box catalyst (5.3 A g−1). Therefore, the OER activity of the c-Ti–Fe–S boxes is the best among the three catalysts and comparable to those of other reported sulfide, oxide, and carbide electrocatalysts (Table S1, ESI).

image file: c8ta02334d-f4.tif
Fig. 4 (a) LSV curves, (b) mass activities at η = 350 mV, and (c) Tafel slopes of the three catalysts. (d) Chronopotentiometry response of the c-Ti–Fe–S nanobox catalyst.

The Tafel plots of the catalysts are further investigated to get additional insight into their OER performance. The Tafel plots are derived from the polarization curves via the Tafel equation (η = b × log[thin space (1/6-em)]j + a, where η is the overpotential, j is the current density, and b is the Tafel slope). As can be seen in Fig. 4c, the c-Ti–Fe–S boxes obviously exhibit enhanced kinetics for the OER, which is proved by its much lower Tafel slope of 55 mV dec−1 compared with other catalysts, e.g. c-Ti–Fe–S cubes (58 mV dec−1) and a-Ti–S/c-Fe–S boxes (66 mV dec−1). Besides high activity, excellent stability of the electrocatalysts toward the OER is also critical for energy conversion systems. Impressively, the c-Ti–Fe–S boxes could maintain a current density of 5 and 10 mA cm−2 for 12 h with only ∼0.5 and 1.0%, respectively, increases required for the potential, suggesting superior stability of the catalyst (Fig. 4d). After the chronopotentiometry measurements, the structure of the c-Ti–Fe–S boxes is further probed by TEM (Fig. S16, ESI). Remarkably, the box-like nanostructure of the catalyst can still be retained, indicating the good structural robustness of this catalyst. Nevertheless, it should be noted that the surface of the catalyst transfers from sulfides to oxidized metal species after the durability test, as revealed by the XPS measurements (Fig. S17, ESI). These oxidized species may be the real active sites for the OER.

Interestingly, Ti–Fe mixed sulfides present an OER activity trend as follows: c-Ti–Fe–S boxes > c-Ti–Fe–S cubes > a-Ti–S/c-Fe–S boxes. To explore the possible reasons for this activity trend, we first evaluate the electronic conductivity of the catalysts. The electrochemical impedance spectroscopy (EIS) data show that the c-Ti–Fe–S boxes and cubes have much smaller charge-transfer resistance than the a-Ti–S/c-Fe–S boxes (Fig. S18, ESI). Then we estimate the electrochemically active surface area (ECSA) of the catalysts based on their double-layer capacitances (Cdl), since the ECSA is generally proportional to the Cdl of the electrocatalysts.43 Specifically, the scan-rate dependence of cyclic voltammograms is measured in the potential range of 1.177–1.277 V without redox processes, to obtain the capacitive current associated with double-layer charging for the three catalysts (Fig. S19a–c, ESI). Then Cdl can be obtained by plotting the Δj = jajc at 1.24 V vs. RHE against the scan rate (the linear slope is twice the Cdl).44 The results show that the linear slopes are as follows: c-Ti–Fe–S boxes > a-Ti–S/c-Fe–S boxes > c-Ti–Fe–S cubes, implying that the ECSA should follow the same sequence (Fig. S19d, ESI). Combining the EIS and ECSA data with the OER activities data, it can be inferred that: (i) the crystalline titanium sulfide should have better electron conductivity than the amorphous phase of titanium sulfide, endowing either the c-Ti–Fe–S boxes or cubes with better OER activity than the a-Ti–S/c-Fe–S boxes, even though the ECSA of the c-Ti–Fe–S cubes is smaller than that of the a-Ti–S/c-Fe–S boxes. Therefore, compared with the ECSA, the charge transfer resistance should be a more important factor to impact the OER activity in this work. (ii) The OER activity of c-Ti–Fe–S boxes is slightly better than that of cubes, which might be due to the larger ECSA of the former that benefited from a greater electrocatalyst–electrolyte contact interface area provided by the hollow structure.


In summary, we have demonstrated a self-template route to transform the Ti–Fe PBA nanoboxes to corresponding mixed metal sulfide nanoboxes. As the primary structure, the Ti–Fe PBA nanoboxes are found to be formed by an inside-out hollowing process from their nanocube precursor, due to the possible ammonia-evoked etching towards the weak center of the nanocubes during the solvothermal process. The final product of crystalline Ti–Fe mixed sulfide nanoboxes shows enhanced OER activity, affording a current density of 10 mA cm−2 and a high mass activity of 100 A g−1 at a small overpotential of 350 mV, a Tafel slope as small as 55 mV dec−1, and excellent stability in alkaline medium.

Conflicts of interest

There are no conflicts to declare.


The authors sincerely acknowledge the great support from Prof. Xiong Wen (David) Lou of Nanyang Technological University for allowing the use of facility in his lab to conduct this work. This work is supported by the National Research Foundation (NRF) of Singapore via the NRF investigatorship (NRF-NRFI2016-04), the Fundamental Research Funds for the Central Universities (Grant No. 112207*172210171), and the Zhejiang Provincial Natural Science Foundation of China (Grant No. LY18B030002).

Notes and references

  1. Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K. Nørskov and T. F. Jaramillo, Science, 2017, 355, eaad4998 CrossRef PubMed.
  2. Y. Liang, Y. Li, H. Wang and H. Dai, J. Am. Chem. Soc., 2013, 135, 2013 CrossRef PubMed.
  3. X.-F. Yang, A. Wang, B. Qiao, J. Li, J. Liu and T. Zhang, Acc. Chem. Res., 2013, 46, 1740 CrossRef PubMed.
  4. H. Zhang, P. An, W. Zhou, B. Y. Guan, P. Zhang, J. Dong and X. W. Lou, Sci. Adv., 2018, 4, eaao6657 CrossRef PubMed.
  5. I. Roger, M. A. Shipman and M. D. Symes, Nat. Rev. Chem., 2017, 1, 0003 CrossRef.
  6. N.-T. Suen, S.-F. Hung, Q. Quan, N. Zhang, Y.-J. Xu and H. M. Chen, Chem. Soc. Rev., 2017, 46, 337 RSC.
  7. M. W. Kanan and D. G. Nocera, Science, 2008, 321, 1072 CrossRef PubMed.
  8. J. Suntivich, K. J. May, H. A. Gasteiger, J. B. Goodenough and Y. Shao-Horn, Science, 2011, 334, 1383 CrossRef PubMed.
  9. R. Subbaraman, D. Tripkovic, K.-C. Chang, D. Strmcnik, A. P. Paulikas, P. Hirunsit, M. Chan, J. Greeley, V. Stamenkovic and N. M. Markovic, Nat. Mater., 2012, 11, 550 CrossRef PubMed.
  10. H. Wang, H.-W. Lee, Y. Deng, Z. Lu, P.-C. Hsu, Y. Liu, D. Lin and Y. Cui, Nat. Commun., 2015, 6, 7261 CrossRef PubMed.
  11. J. Nai, H. Yin, T. You, L. Zheng, J. Zhang, P. Wang, Z. Jin, Y. Tian, J. Liu, Z. Tang and L. Guo, Adv. Energy Mater., 2015, 5, 1401880 CrossRef.
  12. X.-Y. Yu, Y. Feng, B. Guan, X. W. Lou and U. Paik, Energy Environ. Sci., 2016, 9, 1246 RSC.
  13. P. He, X.-Y. Yu and X. W. Lou, Angew. Chem., Int. Ed., 2017, 56, 3897 CrossRef PubMed.
  14. X. Xu, F. Song and X. Hu, Nat. Commun., 2016, 7, 12324 CrossRef PubMed.
  15. Y. Liu, H. Cheng, M. Lyu, S. Fan, Q. Liu, W. Zhang, Y. Zhi, C. Wang, C. Xiao, S. Wei, B. Ye and Y. Xie, J. Am. Chem. Soc., 2014, 136, 15670 CrossRef PubMed.
  16. J. Nai, Y. Lu, L. Yu, X. Wang and X. W. Lou, Adv. Mater., 2017, 29, 1703870 CrossRef PubMed.
  17. S. Anantharaj, S. R. Ede, K. Sakthikumar, K. Karthick, S. Mishra and S. Kundu, ACS Catal., 2016, 6, 8069 CrossRef.
  18. X.-Y. Yu, L. Yu and X. W. Lou, Adv. Energy Mater., 2016, 6, 1501333 CrossRef.
  19. X.-Y. Yu, L. Yu and X. W. Lou, Small Methods, 2017, 1, 1600020 CrossRef.
  20. M.-R. Gao, Y.-R. Zheng, J. Jiang and S.-H. Yu, Acc. Chem. Res., 2017, 50, 2194 CrossRef PubMed.
  21. J. Chen, S.-L. Li, Z.-L. Tao and F. Gao, Chem. Commun., 2003, 980 RSC.
  22. J. Chen, S.-L. Li, Z.-L. Tao, Y.-T. Shen and C.-X. Cui, J. Am. Chem. Soc., 2003, 125, 5284 CrossRef PubMed.
  23. Z.-L. Tao, L.-N. Xu, X.-L. Gou, J. Chen and H.-T. Yuan, Chem. Commun., 2004, 2080 RSC.
  24. J. Chen, Z.-L. Tao and S.-L. Li, Angew. Chem., Int. Ed., 2003, 42, 2147 CrossRef PubMed.
  25. X.-Y. Yu and X. W. Lou, Adv. Energy Mater., 2018, 8, 1701592 CrossRef.
  26. B. Kong, C. Selomulya, G. Zheng and D. Zhao, Chem. Soc. Rev., 2015, 44, 7997 RSC.
  27. O. Sato, T. Iyoda, A. Fujishima and K. Hashimoto, Science, 1996, 272, 704 CrossRef PubMed.
  28. A. Bleuzen, V. Escax, A. Ferrier, F. Villain, M. Verdaguer, P. Münsch and J.-P. Itié, Angew. Chem., Int. Ed., 2004, 43, 3728 CrossRef PubMed.
  29. E. Coronado, M. C. Giménez-López, G. Levchenko, F. M. Romero, V. García-Baonza, A. Milner and M. Paz-Pasternak, J. Am. Chem. Soc., 2005, 127, 4580 CrossRef PubMed.
  30. X. Cai, W. Gao, M. Ma, M. Wu, L. Zhang, Y. Zheng, H. Chen and J. Shi, Adv. Mater., 2015, 27, 6382 CrossRef PubMed.
  31. S. Pintado, S. Goberna-Ferrón, E. C. Escudero-Adán and J. R. Galán-Mascarós, J. Am. Chem. Soc., 2013, 135, 13270 CrossRef PubMed.
  32. W. Zhang, S. Hu, J.-J. Yin, W. He, W. Lu, M. Ma, N. Gu and Y. Zhang, J. Am. Chem. Soc., 2016, 138, 5860 CrossRef PubMed.
  33. F. Ma, Q. Li, T. Wang, H. Zhang and G. Wu, Sci. Bull., 2017, 62, 358 CrossRef.
  34. L. Zhang, H. B. Wu, S. Madhavi, H. H. Hng and X. W. Lou, J. Am. Chem. Soc., 2012, 134, 17388 CrossRef PubMed.
  35. L. Han, X.-Y. Yu and X. W. Lou, Adv. Mater., 2016, 28, 4601 CrossRef PubMed.
  36. Y. Feng, X.-Y. Yu and U. Paik, Chem. Commun., 2016, 52, 6269 RSC.
  37. J. Nai, B. Y. Guan, L. Yu and X. W. Lou, Sci. Adv., 2017, 3, e1700732 CrossRef PubMed.
  38. L. Zhang, H. B. Wu and X. W. Lou, J. Am. Chem. Soc., 2013, 135, 10664 CrossRef PubMed.
  39. X.-Y. Yu, L. Yu, H. B. Wu and X. W. Lou, Angew. Chem., Int. Ed., 2015, 54, 5331 CrossRef PubMed.
  40. X.-Y. Yu, Y. Feng, Y. Jeon, B. Guan, X. W. Lou and U. Paik, Adv. Mater., 2016, 28, 9006 CrossRef PubMed.
  41. Y. Feng, X.-Y. Yu and U. Paik, Chem. Commun., 2016, 52, 1633 RSC.
  42. M. B. Zakaria, RSC Adv., 2016, 6, 10341 RSC.
  43. C. C. L. McCrory, S. Jung, J. C. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2013, 135, 16977 CrossRef PubMed.
  44. F. Song and X. Hu, J. Am. Chem. Soc., 2014, 136, 16481 CrossRef PubMed.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c8ta02334d

This journal is © The Royal Society of Chemistry 2018