Electric field controlled CO₂ capture and CO₂/N₂ separation on MoS₂ monolayers

Abstract

Developing new materials and technologies for efficient CO₂ capture, particularly for separation of CO₂ post-combustion, will significantly reduce the CO₂ concentration and its impacts on the environment. A challenge for CO₂ capture is to obtain high performance adsorbents with both high selectivity and easy regeneration. Here, CO₂ capture/regeneration on MoS₂ monolayers controlled by turning on/off external electric fields is comprehensively investigated through a density functional theory calculation. The calculated results indicate that CO₂ forms a weak interaction with MoS₂ monolayers in the absence of an electric field, but strongly interacts with MoS₂ monolayers when an electric field of 0.004 a.u. is applied. Moreover, the adsorbed CO₂ can be released from the surface of MoS₂ without any energy barrier once the electric field is turned off. Compared with the adsorption of CO₂, the interactions between N₂ and MoS₂ are not affected significantly by the external electric fields, which indicates that MoS₂ monolayers can be used as a robust absorbent for controllable capture of CO₂ by applying an electric field, especially to separate CO₂ from the post-combustion gas mixture where CO₂ and N₂ are the main components.

---

Introduction

The rapid increase of CO₂ in the atmosphere due to anthropogenic activities is one of the greatest global issues faced by mankind today.^1–3 The development of effective materials and technologies for efficient CO₂ capture and gas separation is a challenging and urgent task. Recently, solid materials have emerged as effective adsorbents for CO₂ capture because of their large surfaces and high selectivity.^4,5 However, high selectivity for CO₂ capture usually means that the regeneration step is difficult and energy consuming. Therefore, searching for high-performance adsorbents with both high selectivity and easy regeneration is crucial for CO₂ capture and reutilization.⁶

It is well-known that external electric fields can affect many chemical reactions, especially the electron transfer reactions. The reaction rates can be boosted to several orders of magnitude by applying external electric fields. There is growing interest in applying external electric fields to modulate chemical reactions. For example, electric fields can dramatically enhance hydrogen storage on boron nitride (BN) nanomaterials.⁷ The recent investigations of CO₂ capture and gas separation on boron containing nanomaterials,^8–14 such as BN and pure boron, indicated that CO₂ capture and release can be controlled by turning on/off the charge states or electric fields applied to these nanomaterials.⁸ With an electric field of 0.030 a.u., BN nanosheets were proven to be an effective adsorbent for CO₂ capture.¹⁵ However, turning on/off the charge states or electric fields are energetic processes, due to the wide band gaps of BN nanostructures (their band gaps vary from 3.6 to 7.1 eV with the preparation methods and conditions), and the strong external electric fields applied to BN monolayers.^8,15 Therefore, the main purpose of this study is to search for advanced materials for efficient CO₂ capture/regeneration.

Recently, molybdenum disulfide (MoS₂) as an emerging two-dimensional material has attracted considerable attention due to its exciting properties and great potential for diverse applications, such as energy conversion, piezotronics, phototransistors, catalysts, drug deliveries, lithium-ion batteries, gas sensors, adsorbents and separation.^16–26 However, MoS₂ cannot be used directly as an adsorbent for CO₂ capture because of the weak interactions between MoS₂ and CO₂.²³ Various technologies, such as introduction of dopants, defects, and strains, have been used to successfully increase the interactions between CO₂ and MoS₂.^27–29 As mentioned previously, however, an ideal adsorbent for CO₂ capture should be featured with high selectivity and easy regeneration. Could MoS₂ monolayers serve as ideal adsorbents for CO₂ capture and separation by engineering their interactions with CO₂ by applying electric fields, and by well-understanding the impacts of electric fields on the mechanisms of CO₂ capture, regeneration and separation?

To answer the above question, we carry out the following investigations: (i) the interactions of CO₂ with MoS₂ monolayers in the absence/presence of an electric field, (ii) the reaction mechanisms of CO₂ adsorbed on MoS₂ monolayers with/without the electric field, (iii) optimization of electric fields for CO₂ capture, and (iv) comparing the adsorption of N₂ on MoS₂ in the presence/absence of electric fields, with that of CO₂ to demonstrate whether CO₂ could be selectively separated from the CO₂/N₂ gas mixture by MoS₂ with the assistance of external electric fields.

Calculation methods

The first-principles density functional theory plus dispersion method implemented in the DMol3 package was used for all the calculations of the study.^30,31 All the structures are fully optimized using the generalized gradient approximation,³² treated by the Perdew–Burke–Ernzerhof exchange–correlation potential (PBE) with long range dispersion correction via Grimme's scheme.³³ The basis set for the system is an all-electron double numerical atomic orbital augmented by d-polarization functions (DNP). The optimized lattice constant of the MoS₂ monolayer is 3.185 Å, and a 4 × 4 supercell of the MoS₂ monolayer with a dimension of 12.74 × 12.74 × 20 ų is used for CO₂ and N₂ adsorption without and with different electric fields. The vacuum between the MoS₂ monolayers is 20 Å which is large enough to avoid interactions between periodic images. Geometry optimizations were performed with convergence thresholds of 0.002 a.u. Å⁻¹ on the gradient, 0.005 Å on the displacement, and 10⁻⁵ a.u. on the energy. The real-space global cut-off radius is set to be 4.10 Å. The self-consistent field (SCF) procedure was used with a convergence threshold of 10⁻⁶ a.u. on energy and electron density. The direct inversion of the iterative subspace technique developed by Pulay is used with a subspace size of 6 to speed up SCF convergence on these systems.³⁴ The Brillouin zones are sampled by 6 × 6 × 1 k-points using the Monkhorst–Pack scheme. Electron distributions and transfer mechanisms are determined by the Mulliken method.³⁵

The adsorption energies of CO₂ and N₂ on MoS₂ monolayers are calculated from eqn (1).

E_ads = (E_MoS2 + E_gas) − E_MoS2-gas (1)

where E_MoS2-gas is the total energy of MoS₂ with adsorbed gas, such as CO₂ and N₂, E_MoS2 is the energy of an isolated MoS₂ monolayer, and E_gas is the energy of a CO₂ or a N₂ molecule in the gas phase.

Results and discussion

CO₂ adsorption on MoS₂ monolayers without an electric field

The adsorption of CO₂ on MoS₂ monolayers without an external electric field at the PBE level with a long range dispersion correction was firstly investigated. The optimized configuration with the minimum energy for CO₂ adsorbed on the surface of MoS₂ monolayers is presented in Fig. 1. The configuration shows that the linear CO₂ molecule is parallel to the MoS₂ monolayer and quite far from the surface. The adsorbed CO₂ molecule has little structural change compared to the free CO₂ molecule, with an O–C–O angle of 179.7°. The S⋯C distance between CO₂ and the MoS₂ sheet in the configuration is long with a value of 3.334 Å, and the adsorption energy is 3.16 kcal mol⁻¹. The charge transfer from the MoS₂ monolayer to the adsorbed CO₂ molecule is negligible with a value of 0.008e⁻. Their interactions are very weak and mainly attributed to the van der Waals forces between them. The above results indicate that CO₂ is physisorbed on the MoS₂ monolayer without an external electric field, which agrees well with the recent report²³ where MoS₂ monolayers can't be directly used as an efficient adsorbent for CO₂ capture.

Fig. 1 Top and side views of physisorbed CO₂ on a single layer MoS₂ surface without an electric field. Atom color code: yellow, sulfur; blue, molybdenum; gray, carbon; red, oxygen.

CO₂ adsorption on MoS₂ monolayers with the electric fields from 0.0020 to 0.0050 a.u.

It has been known that external electric fields can enhance gas adsorption on solid surfaces, especially for the polarizable nanomaterials,^7,24 which suggests that the adsorption of CO₂ on the MoS₂ surface could be improved by applying an electric field. In this part, CO₂ adsorption on MoS₂ monolayers with the applied electric fields in the range of 0.0020–0.0050 a.u. was investigated. The important structural parameters of optimized configurations (such as C–S bond length, charge transfer) and the adsorption energies obtained with the electric fields in the above range are shown in Fig. 2 and 3, and given in Table S1 in the ESI.† When the electric field is increased from 0.0020 a.u. to 0.0031 a.u., the distances between CO₂ and MoS₂ decrease slowly and the C⋯S distances are around 3.0 Å [Fig. 2(a) and Table S1†]. The charge transfer from the MoS₂ monolayer to CO₂ increases from 0.036 to 0.077e⁻ [Fig. 2(b) and Table S1†]. The adsorption energies increase smoothly from 5.29 to 7.30 kcal mol⁻¹ [Fig. 2(c) and Table S1†]. These results demonstrate that the CO₂ molecules are physisorbed on the MoS₂ monolayer when the electric field is below 0.0031 a.u. However, when the strength of electric fields increases to 0.0032 a.u., there is a chemical bond (i.e. C–S bond) formed between CO₂ and MoS₂, and the C–S bond of the optimized configuration is 2.048 Å [Fig. 3(b)].

Fig. 2 The distances of C–S bonds (Å), charge transfer (e⁻) and adsorption energies (kcal mol⁻¹) of CO₂ adsorbed on the surface of MoS₂ monolayers with different electric fields.

Fig. 3 Top and side views of chemisorbed CO₂ on the MoS₂ surface with different electric fields. Atom colors are the same as that in Fig. 1.

The results also indicate the transition of the physisorbed configuration into the chemisorbed configuration, and the charge transfer from the MoS₂ monolayer to CO₂ dramatically increases from 0.077 to 0.527e⁻, despite the fact that the electric field is slightly increased from 0.0031 a.u. to 0.0032 a.u. [Fig. 2(b) and Table S1†]. Further increase of the electric field from 0.0032 a.u. to 0.0050 a.u. leads to further decrease of the C–S bond from 2.048 Å to 1.833 Å and increase of charge transfer from 0.527 to 0.865e⁻. The slope of adsorption energies increases dramatically with the increase of the electric fields [Fig. 2(c)], and the adsorption energies increase from 8.01 to 35.62 kcal mol⁻¹. The ideal adsorption energy of CO₂ on high performance adsorbents should be in the range of 40–80 kJ mol⁻¹.³⁶ According to this criterion, the MoS₂ monolayer should be a good sorbent for CO₂ capture if the applied electric field is in the range of 0.0035–0.0040, because the absorption energy in this range is 11.51–19.03 kcal mol⁻¹ (49.37–79.54 kJ mol⁻¹).

CO₂ adsorption on MoS₂ monolayers in the presence of an electric field with a strength of 0.0040 a.u.

The results in the previous section demonstrate that CO₂ is chemisorbed on MoS₂ monolayers in the presence of electric fields above 0.0031 a.u. To compare the adsorption of CO₂ on BN monolayers, we investigated CO₂ adsorption on the surface of MoS₂ monolayers with an electric field of 0.0040 a.u. The direction of the applied electric field is shown in Fig. 4(a). The optimized structure of CO₂ adsorbed on the MoS₂ surface with a perpendicular external electric field (0.0040 a.u.) is displayed in Fig. 4(b), which illustrates a chemisorbed configuration formed between the MoS₂ monolayer and CO₂. In detail, a C–S bond has a length of 1.904 Å. The adsorbed CO₂ molecule is drastically bent with an O–C–O angle of 138.6°, and the two double C=O bonds are elongated to 1.237 Å. A Mulliken charge population analysis shows that there is 0.771e⁻ charge transferred from the MoS₂ sheet to the CO₂ molecule. The Mulliken population analysis further demonstrates that the charge is mainly transferred from the S atom of the MoS₂ monolayer to the C atom of adsorbed CO₂, and the electron transfer is along the direction of the electric field. The CO₂ molecule strongly interacts with the MoS₂ surface with an adsorption energy of 19.03 kcal mol⁻¹. The adsorption energy is similar to that of CO₂ captured on BN monolayers by applying external electric fields.^8,15 With an external electric field of 0.050 a.u., the adsorption energy of CO₂ adsorbed on the BN monolayer where CO₂ vertically interacts with B atoms is 19.26 kcal mol⁻¹.⁸ In the case of CO₂ strongly interacting with the N atoms of the BN sheet in the presence of an electric field of 0.030 a.u., the adsorption energy is 15.68 kcal mol⁻¹ (0.68 eV).¹⁵ Compared with adsorption of CO₂ on BN monolayers, the strength of the electric field applied to MoS₂ monolayers is only around 10% of that applied to BN monolayers.^8,15 The reason is that the MoS₂ monolayer is more polarizable than BN sheets.

Fig. 4 (a) Representation of the direction of the applied perpendicular electric field and the applied electric field is 0.0040 a.u. (b) Top and side views of adsorbed CO₂ on a single layer MoS₂ surface in the presence of an electric field with a strength of 0.0040 a.u. Atom colors are the same as that in Fig. 1.

To investigate the electronic structure of MoS₂ monolayers for CO₂ capture with the effect of the electric field (0.0040 a.u.), the partial density states (PDOS), including s-, p-, and d-orbitals of atoms from the MoS₂–CO₂ system without and with an electric field are plotted in Fig. 5(a) and (b). Fig. 5(a) indicates that in the absence of an electric field, MoS₂ exhibits the nature of a semiconductor. The corresponding energy band gap of 1.88 eV is slightly smaller than that of bare MoS₂ (1.93 eV). Fig. 5(b) shows that when an electric field of 0.0040 a.u. is applied, the adsorption energy increases to 19.03 kcal mol⁻¹, and the semiconductor properties of MoS₂ are still retained with a similar band gap. The calculated result is consistent with the previous report that the bandgap of the MoS₂ monolayer is insensitive to an external perpendicular field,³⁷ which indicates that the MoS₂ monolayer is electronically stable and would not be destroyed by the applied electric field. The PDOS shown in Fig. 5(b) also demonstrates that the p density of state at −0.25 a.u. contributes to the interactions of CO₂ and MoS₂ in the electric field of 0.0040 a.u. In addition, the applied electric field doesn't change its contribution to the density of state at the Fermi level.

Fig. 5 (a) Density of states of MoS₂–CO₂ without an electric field. (b) Density of states of MoS₂–CO₂ in the presence of an electric field with a strength of 0.0040 a.u. (E = 0.0040 a.u.). The vertical dashed line is for the Fermi level.

The mechanisms of CO₂ capture/release on the MoS₂ monolayer by turning on/off the electric field were also studied. Fig. 6(a) shows the turning of a physisorbed configuration obtained without an electric field into an optimized chemisorbed configuration when an electric field of 0.0040 a.u. is applied to the system. The interactions between CO₂ and MoS₂ drastically increase compared with the case of zero field, and the CO₂ molecule is chemisorbed on the surface of MoS₂ with a C–S distance of 1.904 Å, and the reaction is exothermic by about 11.75 kcal mol⁻¹. The turning of the physisorbed configuration into the chemisorbed configuration is exothermic and without any energy barrier. When the electric field is turned off, the turning of the chemisorbed configuration into the physisorbed configuration is also studied. The results illustrate an increase of C⋯S bond distances from 1.904 Å to 3.334 Å. This transition also has no barrier and it is exothermic by 34.23 kcal mol⁻¹ [Fig. 6(b)], which means that CO₂ is easily released from the surface of MoS₂ monolayers once the electric field is removed. The above results demonstrate that the capture and release of CO₂ are overall controlled by the electric field applied to the system, and the energy costs of these processes are dependent on the applied electric field.

Fig. 6 (a) The energy change due to a change from the physisorbed to the chemisorbed configuration with an electric field of 0.0040 a.u. on the MoS₂ surface. (b) The energy change due to a change from the chemisorbed to the physisorbed configuration without an electric field for CO₂ adsorption on the MoS₂ surface.

CO₂/N₂ separation on MoS₂ monolayers

With population expansion and industrial development, the combustion of fossil fuels (coal, petroleum, and natural gas) increases and accounts for 86% of anthropogenic greenhouse gas emissions.³ Therefore, efficient separation of CO₂ from the post-combustion mixture (the main containments are CO₂ and N₂) can reduce the greenhouse effect and mitigate climate change. In this part, we studied whether MoS₂ monolayers can be used to separate CO₂ from post-combustion gas (CO₂/N₂). Firstly, we investigated N₂ adsorption on MoS₂ monolayer without an electric field. The optimized configuration of N₂ adsorbed on the surface of MoS₂ without an electric field is shown in Fig. 7(a), where the distance between N₂ and MoS₂ is quite far with a value of 3.464 Å, and the adsorption energy is 2.28 kcal mol⁻¹, which indicates that N₂ is physisorbed on MoS₂ and their interactions are weak. Similar weak interactions of CO₂ and N₂ adsorbed on MoS₂ suggest that MoS₂ can't be used to separate CO₂ from the CO₂/N₂ gas mixture without an electric field. Then the adsorption of N₂ on MoS₂ monolayers with the electric fields ranging from 0.0020 a.u. to 0.0050 a.u. was carried out_.

Fig. 7 Top and side views of configurations of N₂ adsorption on the MoS₂ surface (a) without and (b) with an electric field at the strength of 0.0040 a.u. Atom colors are the same as that in Fig. 1.

The optimized configuration of N₂ adsorbed on the surface of the MoS₂ monolayer with an electric field of 0.0040 is displayed in Fig. 7(b). The important structural information, such as bond lengths and bond angles, adsorption energies and electron transfers from MoS₂ to N₂, in the cases of N₂ adsorbed on the MoS₂ surface with different electric fields from 0.0020 a.u. to 0.0050 a.u. are shown in Fig. 8 and given in Table S2 in the ESI.†Fig. 7(b) shows that with the electric field of 0.0040 a.u., the distance between N₂ and MoS₂ is still far with a value of 3.051 Å, the adsorption energy is 7.42 kcal mol⁻¹, and N₂ is physisorbed on the MoS₂ monolayer. This is in contrast to the adsorption of CO₂ on the MoS₂ monolayer under the same electric field (i.e. 0.0040 a.u.), where a chemisorbed configuration is formed. The drastic difference between the adsorptions of CO₂/N₂ on MoS₂ monolayers with the same electric field (0.0040 a.u.) clearly demonstrates that the MoS₂ monolayer can selectively capture CO₂ from the CO₂/N₂ gas mixture with the assistance of an electric field. Table S2† shows that for the adsorption of N₂ on the MoS₂ surface with the electric field fields increasing from 0.0020 a.u. to 0.0050 a.u., the distances between N₂ and MoS₂ are in the range of 2.840–3.329 Å. Therefore, the interactions between N₂ and MoS₂ in all adsorbed configurations are relatively weak. Fig. 8 also clearly shows that the difference in adsorption energies of CO₂ and N₂ adsorbed on MoS₂ monolayers increases with the electric fields. All the above results demonstrate that by applying an electric field (the optimized electric field is in the range of 0.0035–0.0040 a.u.), CO₂ can be efficiently captured with MoS₂ from the CO₂/N₂ gas mixture.

Fig. 8 The adsorption energies (kcal mol⁻¹) of CO₂ and N₂ adsorption on the MoS₂ surface with different electric fields.

In addition, the influences of H₂O, NO, NO₂ and SO₂ on the adsorption of CO₂ on MoS₂ monolayers were also investigated with the effect of an external electric field, as they are the major containments of flue gas. The optimized configurations of H₂O, NO, NO₂ and SO₂ adsorbed on MoS₂ monolayers without an external electric field, and with external electric fields of 0.0020 a.u. and 0.0050 a.u. are listed in Fig. S1 and S2 in the ESI.† The adsorption energies of four gases, CO₂ and N₂ adsorbed on the surface of MoS₂ by applying the electric fields in the range of 0.0020–0.0050 a.u. are shown in Fig. S2 and S4 in the ESI,† which clearly indicate that the adsorption of H₂O, NO, NO₂ and SO₂ on MoS₂ monolayers increases linearly with the increase of electric fields. Moreover, the adsorptions of four gases on MoS₂ monolayers are stronger than that of CO₂ with the same electric field. The above results suggest that with a relatively weak electric field (around 0.0020 a.u.), H₂O, NO, NO₂ and SO₂ can be selectively separated from the gas mixture by MoS₂ monolayers. By increasing the electric field to around 0.0035 a.u., CO₂ can be selectively captured on MoS₂ monolayers from the residual CO₂/N₂ gas mixture. The study also demonstrates selective capture of above gases by controlling the strength of electric fields applied to MoS₂ monolayers.

Conclusions

In summary, the CO₂ capture and regeneration on the MoS₂ surface without and with electric fields were theoretically investigated. The results indicate that the adsorption of CO₂ on MoS₂ can be drastically enhanced by applying electric fields. By turning off the electric field, CO₂ can be released from the MoS₂ surface without any barrier. In contrast, the applied electric fields can only slightly affect the interaction strength between N₂ and MoS₂ compared with that of CO₂ and MoS₂, which means that with the applied electric fields, MoS₂ can be employed to efficiently separate CO₂ from the CO₂/N₂ gas mixture (the main containment of post combustion gas). The study demonstrates a versatile approach to CO₂ capture, regeneration and gas separation by controlling the strength of electric fields applied to the sorbents, which provides interesting and complementary information to experiments in search for advanced materials and technologies for CO₂ capture and gas separation.

Acknowledgements

Q. Sun acknowledges support from Jiangsu province (Grant No. BK20151215). Z. Li greatly appreciates support from the National Nature Science Foundation of China (Grant No. 81471657), the 1000 Plan for Young Talents and Jiangsu Specially-Appointed Professorship. This work is also supported by the National Natural Science Foundation of China (Grant No. 21303009, 21577076, 21303093).

Notes and references

1. D. W. Keith, Science, 2009, 325, 1654.
2. J. Meyer, Nature, 2008, 455, 733.
3. N. Stern, Review on the Economics of Climate Change, Cambridge University Press, Cambridge, 2006.
4. J. Y. Wang, L. Huang, R. Y. Yang, Z. Zhang, J. W. Wu, Y. S. Gao, Q. Wang, D. O'Hare and Z. Y. Zhong, Energy Environ. Sci., 2014, 7, 3478.
5. Y. L. Luo, B. Y. Li, W. Wang, K. B. Wu and B. Tan, Adv. Mater., 2012, 24, 5703.
6. T. M. McDonald, J. A. Mason, X. Q. Kong, E. D. Bloch, D. Gygi, A. Dani, V. Crocella, F. Giordanino, S. O. Odoh, W. S. Drisdell, B. Vlaisavljevich, A. L. Dzubak, R. Poloni, S. K. Schnell, N. Planas, K. Lee, T. Pascal, L. W. F. Wan, D. Prendergast, J. B. Neaton, B. Smit, J. B. Kortright, L. Gagliardi, S. Bordiga, J. A. Reimer and J. R. Long, Nature, 2015, 519, 303.
7. J. Zhou, Q. Wang, Q. Sun, P. Jena and X. S. Chen, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 2801.
8. Q. Sun, Z. Li, D. J. Searles, Y. Chen, G. Lu and A. Du, J. Am. Chem. Soc., 2013, 135, 8246.
9. Q. Sun, C. Sun, A. Du and Z. Li, J. Phys. Chem. C, 2014, 118, 30006.
10. Q. Sun, M. Wang, Z. Li, A. Du and D. J. Searles, J. Phys. Chem. C, 2014, 118, 2170.
11. Q. Sun, M. Wang, Z. Li, A. Du and D. J. Searles, Phys. Chem. Chem. Phys., 2014, 16, 12695.
12. Q. Sun, M. Wang, Z. Li, P. Li, W. Wang, X. Tan and A. Du, Fuel, 2013, 109, 575.
13. Q. Sun, M. Wang, Z. Li, Y. Ma and A. Du, Chem. Phys. Lett., 2013, 575, 59.
14. W. Wang, X. Zhang, P. Li, Q. Sun, Z. Li, C. Ren and C. Guo, J. Phys. Chem. A, 2015, 119, 796.
15. H. Guo, W. Zhang, N. Lu, Z. Zhuo, X. C. Zeng, X. Wu and J. Yang, J. Phys. Chem. C, 2015, 119, 6912.
16. B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti and A. Kis, Nat. Nanotechnol., 2011, 6, 147.
17. W. Wu, L. Wang, Y. Li, F. Zhang, L. Lin, S. Niu, D. Chenet, X. Zhang, Y. Hao, T. F. Heinz, J. Hone and Z. L. Wang, Nature, 2014, 514, 470.
18. G. Fiori, F. Bonaccorso, G. Iannaccone, T. Palacios, D. Neumaier, A. Seabaugh, S. K. Banerjee and L. Colombo, Nat. Nanotechnol., 2014, 9, 768.
19. Y.-C. Lin, D. O. Dumcencon, Y.-S. Huang and K. Suenaga, Nat. Nanotechnol., 2014, 9, 391.
20. Y.-C. Lin, D. O. Dumcenco, H.-P. Komsa, Y. Niimi, A. V. Krasheninnikov, Y.-S. Huang and K. Suenaga, Adv. Mater., 2014, 26, 2857.
21. W. Yin, L. Yan, J. Yu, G. Tian, L. Zhou, X. Zheng, X. Zhang, Y. Yong, J. Li, Z. Gu and Y. Zhao, ACS Nano, 2014, 8, 6922.
22. C. Zhu, X. Mu, P. A. van Aken, Y. Yu and J. Maier, Angew. Chem., Int. Ed., 2014, 53, 2152.
23. S. Zhao, J. Xue and W. Kang, Chem. Phys. Lett., 2014, 595, 35.
24. Q. Yue, Z. Shao, S. Chang and J. Li, Nanoscale Res. Lett., 2013, 8, 425.
25. M. Asadi, B. Kumar, A. Behranginia, B. A. Rosen, A. Baskin, N. Repnin, D. Pisasale, P. Phillips, W. Zhu, R. Haasch, R. F. Klie, P. Kral, J. Abiade and A. Salehi-Khojin, Nat. Commun., 2014, 5, 4470.
26. A. Achari, S. Sahana and M. Eswaramoorthy, Energy Environ. Sci., 2016, 9, 1224.
27. M. P. K. Sahoo, J. Wang, Y. J. Zhang, T. Shimada and T. Kitamura, J. Phys. Chem. C, 2016, 120, 14113.
28. S. J. Ray, Sens. Actuators, B, 2016, 222, 492.
29. H. X. Li, M. Huang and G. Y. Cao, Phys. Chem. Chem. Phys., 2016, 18, 15110.
30. B. Delley, J. Chem. Phys., 1990, 92, 508.
31. B. Delley, J. Chem. Phys., 2000, 113, 7756.
32. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865.
33. S. Grimme, J. Comput. Chem., 2006, 27, 1787.
34. P. Pulay, J. Comput. Chem., 1982, 3, 556.
35. R. S. Mulliken, J. Chem. Phys., 1955, 23, 1833.
36. S. Chu and A. Majumdar, Nature, 2012, 488, 294.
37. Q. Yue, S. Chang, J. Kang, X. Zhang, Z. Shao, S. Qin and J. Li, J. Phys.: Condens. Matter, 2012, 24, 335501.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6nr07001a

---