Mechanical deformation induced charge redistribution to promote the high performance of stretchable magnesium-ion batteries based on two-dimensional C2N anodes

Donghai Wu a, Baocheng Yang a, Houyang Chen *b and Eli Ruckenstein b
aHenan Key Laboratory of Nanocomposites and Applications, Institute of Nanostructured Functional Materials, Huanghe Science and Technology College, Zhengzhou 450006, China
bDepartment of Chemical and Biological Engineering, University at Buffalo, The State University of New York, Buffalo, New York 14260-4200, USA. E-mail: hchen23@buffalo.edu

Received 25th April 2019 , Accepted 27th June 2019

First published on 27th June 2019


Stretchable batteries play a central role in stretchable electronics such as healthcare devices and sensors. However, challenges in stretchable batteries, such as unstable performance during the deformation process and their mechanical resistances, slow down their applications. In this paper, by adopting state-of-the-art first-principles calculations, the high performance of two-dimensional (2D) nitrogenated holey graphene C2N based stretchable magnesium-ion batteries (MgIBs) and its origin are determined. Their maximum capacity and average open circuit voltage are 1175 mA h g−1 and 0.447 eV, respectively, in the strain-free state, which are much more superior to most of the previously reported 2D anode materials. Mechanical activation is a facile and effective approach to induce charge redistribution to promote their high performance, i.e. their capacities are promoted under large compressive and tensile strains, whereas their high capacities are maintained under small compressive and tensile strains. The origin of the high performance of stretchable MgIBs is also proposed based on the evidence of electron localization function and charge density difference. Compressive and tensile strains remodulate the structures of electrode materials at the atomic scale and redistribute the electrons at the electronic scale, resulting in the reactivation of adsorption sites in electrode materials and maintaining the high performance of stretchable MgIBs. Additionally, large compressive strains enhance the stability of the C2N/Mg system, further increasing its capacities. The stretchable battery shows a two-stage diffusion mechanism in strain-free and strained states, i.e. in the first stage, the out-of-plane Mg ions diffuse quickly, and in the second stage, the in-plane Mg ions migrate moderately. Compared with the stretchable battery in the strain-free state, its barrier energies are lowered in the strained states. This study provides new insights and microscale mechanisms for non-stretchable and stretchable batteries with high performance and facilitates the innovation of low cost non-stretchable and stretchable batteries with 2D material anodes.


1. Introduction

Stretchable electronic devices, which possess broad applications from healthcare science1 to soft electronics,2,3 have attracted great attention in both industrial and scientific fields.4,5 As a main component, stretchable batteries play a central role in electronic devices. Compared to non-stretchable batteries, electrode materials for stretchable batteries face many more challenges, e.g., their unstable performance during deformation processes4 and their mechanical resistances. Most of the electrode materials that work well for non-stretchable batteries may not be available for stretchable batteries in stretched states. Besides, the origin of stretchable batteries is rarely investigated.6

On the other hand, the growing demand for sustainable and clean energies requires numerous advanced energy storage technologies. Among them, lithium-ion batteries (LIBs) have achieved great commercial and academic successes over the past few decades. However, the cost and safety issues have hampered their further development.7 Thus, the search for high-performance, low cost and safe next generation LIBs and/or other metal ion batteries has become urgent.8–15 As an alternative, magnesium-ion batteries (MgIBs) could be one of the promising candidates for next generation batteries due to their safety, abundance, suitable specific capacity, and high voltage.16–18

However, a couple of challenges hinder the commercial applications of MgIBs, e.g., the slow migration of Mg2+ in the electrodes16,19–21 and the incompatibility between the electrode materials and the electrolyte.19,22–24 To solve these issues, numerous cathode materials for MgIBs have been studied.16,25–28 However, anode materials for MgIBs have barely been investigated. Currently, most of the anode materials for metal-ion batteries are based on group IV A9,10,15,29 (e.g., graphite and silicon) and group V A materials (e.g., black phosphorene, arsenene, and antimonene).12,30–33 However, anode materials, which are suitable for other metal-ion batteries, may not work well for MgIBs. For example, graphite is a classic anode material for commercial LIBs,34 but it does not work well for MgIBs. Therefore, the search for appropriate anode materials is one of the key issues for high performance MgIBs.

In this paper, a nitrogenated holey graphene C2N monolayer is adopted as the anode material for MgIBs. Our density functional theory (DFT) calculations, which were implemented by the Vienna ab initio simulation package (VASP) (see Section S1 of the ESI for details),35 show that C2N is a high performance anode material for both non-stretchable and stretchable batteries. Its theoretical capacity and average open-circuit voltage (OCV) for MgIBs are 1175 mA h g−1 and 0.447 V, respectively, in the strain-free state. Mechanical activation is an effective and facile method to boost its performance, i.e., large compressive and tensile strains boost its capacities, whereas small compressive and tensile strains retain its high capacities. Based on the theoretical evidence of the electron localization function, Bader charge, density of states, and charge density difference, the origin of the high performance of stretchable MgIBs is also identified. Compressive and tensile strains remodulate the structures of electrode materials at the atomic scale and redistribute the electrons at the electronic scale, reactivating adsorption sites in electrode materials and maintaining the high performance of stretchable MgIBs. The diffusion results indicate that the migration of Mg ions on C2N holds a two-stage diffusion mechanism, i.e., out-of-plane Mg atoms have high migration rates in the first stage, whereas in-plane Mg atoms possess moderate kinetics in the second stage. It should be mentioned that most of the Mg atoms are out-of-plane atoms. The results facilitate the innovation of low cost non-stretchable and stretchable batteries with 2D material anodes and offer a new microscale mechanism for high performance stretchable batteries.

2. Results and discussion

2.1. Tensile and compressive strains: effective ways for charge redistribution to reactivate adsorption sites on electrode materials

As an electrode material, its precondition is that it can adsorb metal ions. First, we examine the adsorption behavior of Mg ions on a strain-free C2N anode. By adopting the single atom adsorption method, only one in-plane site (site A, see Fig. S1) is the stable adsorption site and its adsorption energy Eads is −2.495 eV (Fig. S2a and the details of calculations are provided in Section S2 of ESI). The strong adsorption of Mg atoms at A sites could be ascribed to the delocalized lone pair electrons of N atoms in C2N. The strong Mg–N interaction is also verified by the electron localization function (ELF) (Fig. 1a) and the charge density difference (Fig. S2b). The value of ELF around N atoms is approximately 0.9, whereas the value of ELF around Mg atoms is small. A quantitative result of Bader charge analysis also supports this behavior, i.e., 1.68|e| are transferred from one Mg atom to C2N. These results imply the ionic bond feature between Mg and N.
image file: c9nr03518d-f1.tif
Fig. 1 ELF plots of the most stable configurations for a single Mg atom adsorption on the C2N monolayer without strain (a) and under biaxial strains of −10% (b), −5% (c), 5% (d), and 10% (e).

A stretchable battery works in strained states. With the introduction of strains, the calculated Eads indicates that the Mg atoms still prefer the in-plane A sites of C2N, and their Eads values are −4.725, −3.601, −1.629, and −0.817 eV under −10%, −5%, 5% and 10% biaxial strains, respectively. The values of Eads demonstrate that compressive strain (the value of a strain is smaller than 0) results in a stronger adsorption strength (more negative) between Mg and C2N, while tensile strain (the value of a strain is larger than 0) leads to a weaker interaction (less negative) between Mg and C2N. This indicates that the C2N/Mg system becomes more stable under compressive strains. The calculated Bader charges show that 1.706, 1.697, 1.649, and 1.630|e| are transferred from one Mg atom to MoS2 under −10%, −5%, 5% and 10% biaxial strains, respectively, suggesting that more charge transfer occurs from Mg to C2N with the introduction of compressive strains and less charge transfer occurs with the loading of tensile strains.

The ELF results demonstrate that compressive/tensile strains increase/decrease the ELF values around N atoms (Fig. 1b–e). One can see that the coupling between Mg and N atoms increases, again confirming that compressive strains make the C2N/Mg system more stable. Furthermore, both compressive and tensile strains (Fig. 1) alter the structures of C2N at the atomic scale, and remodulate the electron distribution at the electronic scale, especially the C4N2 rings.

The strong adsorption of Mg in the in-plane A sites of C2N is not beneficial for ion migration in (stretchable) batteries. Thus, we further adopt the atom pair adsorption approach36 to investigate the adsorption behavior of low concentration Mg atoms on the surface of C2N. For a strain-free electrode material, compared with the single Mg atom adsorption on C2N, three additional adsorption sites (out-of-plane sites C, D and E) are determined (Fig. S5), and the Eads values of the second Mg atom adsorption are −0.402, −0.855, and −0.401 eV, respectively.

Using the atom pair adsorption approach on the strained electrode material, with a −10% biaxial compressive strain (Fig. S6), six stable out-of-plane sites for the second Mg adsorption on C2N are determined, and their Eads are −1.594, −1.592, −1.576, −1.532, −1.531, and −1.522 eV for D, F, A, C, E, and B sites, respectively. It should be mentioned that these sites are located at ∼3.0 Å away from C2N. Their adsorption energies are closer, and are much less than that of the first Mg atom (−4.725 eV). With a −5% biaxial compressive strain, three stable out-of-plane sites are determined (Fig. S7), and their Eads are −1.092, −0.960 and −0.951 eV for D, C, and B sites, respectively. These three sites are located at ∼2.8 Å away from C2N. With the tensile strains of 5% and 10%, only out-of-plane D sites (Fig. S8) with Eads of −0.959 (for 5% strain) and −0.952 eV (for 10% strain) are determined. It should be mentioned that, as the tensile strain increases, the holes surrounded by N atoms enlarge, and additional available in-plane adsorption sites in the holes may be activated. For example, a second available adsorption site is determined in the holes with a 10% tensile strain (Fig. S8c). The charge density difference for the most stable configurations of atom pair adsorption on C2N under compressive and tensile strains is also provided (Fig. S9).

In a word, compressive and tensile strains remodulate the structures of electrode materials at the atomic scale and redistribute electrons at the electronic scale, resulting in the reactivation of adsorption sites in electrode materials.

2.2. High performance stretchable Mg-ion batteries and their origins

2.2.1 Specific capacity. A series of Mg ions of increased concentration (Mgx(C2N)6, x = 3–6) are adopted to test the storage capacity of a C2N anode. Their energetically favorable configurations are provided in Fig. 2 and S10, and the computed Eads values for Mg3(C2N)6, Mg4(C2N)6, Mg5(C2N)6 and Mg6(C2N)6 are −0.601, −0.422, −0.094 and 0.034 eV, respectively, implying that the maximum stoichiometric ratio is Mg5(C2N)6 (Fig. 2). Our AIMD result shows that Mg5(C2N)6 is thermally stable (Fig. S11a), and the corresponding simulated maximum specific capacity (the definition is provided in Section S3 of the ESI) of the C2N monolayer as a MgIB anode achieves 1175 mA h g−1, which is much larger than numerous reported MgIBs (e.g., 957 mA h g−1 for TiS2,37 ∼277 mA h g−1 for chloride intercalated layered MoS2,16 570 mA h g−1 for Ti2CO2,8 and 360 mA h g−1 for Mg3Bi2[thin space (1/6-em)]38).
image file: c9nr03518d-f2.tif
Fig. 2 Top (a) and side (b) views of the most stable configuration for five Mg atoms’ adsorption on a C2N monolayer, and (c) is the corresponding ELF projection on the (110) slice. Yellow, orange, and pink balls denote the Mg atoms located in the top, middle, and bottom layers from the side view, respectively.

To further study the performance of the anode for the stretchable MgIBs, the mechanochemistry approach is employed to check their specific capacities under biaxial strains. The calculation results show that it holds maxima of thirteen, five, five, six, and six Mg atoms on both sides of the anode under strains of −10%, −5%, 2%, 5% and 10%, respectively (Fig. S12a–e), and their corresponding specific capacities are 3056, 1175, 1175, 1411, and 1411 mA h g−1. This indicates that small compressive and tensile strains maintain the high capacities of the anode, whereas large strains promote its capacities. The AIMD simulation (Fig. S11b) demonstrates that the C2N sheet in Mg13(C2N)6 bends but does not break, and all Mg atoms are located on both sides of C2N without clustering.

To gain deep insights into the strain-dependent specific capacity, we further explore the structural change and electronic properties during loading the compressive and tensile strains. The tensile test increases the lattice constants of C2N (Fig. S12c–e), enlarging its surface area and lowering the steric hindrance. When the tensile strain exceeds a critical value, different adsorption positions for Mg ions are activated. For example, one more adsorption position in the hole surrounded by the six N atoms is activated with 10% tensile strain (Fig. S8c). Meanwhile, another adsorption position, which is near the F site, is also activated with tensile strains of 5% or 10% (see the red circles in Fig. S12d and e). Distinguishing from the tensile strains, the compressive strains (Fig. S12a and b) result in wrinkles in C2N, forming more adsorption positions. For example, with the application of a −10% strain (compressive strain), the C2N monolayer crinkles, and more available adsorption positions are activated, widening the Mg layers after the structural relaxation (Fig. S12a). Compared to the strain-free C2N, these additional Mg atoms in C2N with −10% strain mainly located on C, D, E and F sites.

Bader charge analysis is also employed to investigate the strain-dependent capacity of stretchable MgIBs at the electronic level. The results of charge transfer (per Mg atom) between C2N and Mg in the ultimate capacity of C2N with and without strains (Table S1) show that compressive and tensile strains increase the electron transfer between C2N and Mg in the ultimate capacity of C2N (Table S1), which may facilitate the loading of Mg atoms. Similar results are also determined from the charge density difference (Fig. S13).

2.2.2 Diffusion behavior. We then consider the ion diffusion behavior, which is one of the key factors for the charge/discharge rate in non-stretchable and stretchable rechargeable batteries. We adopt the single atom diffusion and atom pair diffusion methods36 to investigate the low concentration ion diffusion on a strain-free C2N anode. For the single atom diffusion method, two possible diffusion paths (Fig. 3a), i.e., A → B → A for Path-I and A → C → B → C → A for Path-II, are prepared based on the symmetry of C2N. The diffusion barrier of the first one is 0.003 eV higher than that of the second one (4.247 eV). The large diffusion barriers are ascribed to the difference of adsorption energies of Mg in the in-plane sites (A sites) and in the out-of-plane sites (B and C sites). It should be mentioned that the diffusion barriers of the out-of-plane Mg ions are small (only 0.04 eV), and the percentage of the out-of-plane Mg ions is much more than that of the in-plane Mg ions, demonstrating that most of the Mg ions can diffuse quickly.
image file: c9nr03518d-f3.tif
Fig. 3 Diffusion paths and their corresponding energy barriers of a single (a) and two (b) Mg atoms’ migration on a strain-free C2N monolayer, and of a single (c) and two (d) Mg atoms’ migration on a C2N monolayer under a compressive strain of −10%.

The atom pair diffusion approach36 is also adopted to further study the diffusion behavior of the out-of-plane Mg ions. All possible migration paths between the most stable adsorption sites (D sites) are taken into account. One feasible migration path (Fig. 3b) is prepared (i.e., D → E → C → E → D), and its diffusion barrier is 0.916 eV. This barrier is much smaller than that from the single atom diffusion method, again confirming the fast diffusion of the out-of-plane Mg ions.

Based on the above-mentioned results from the single atom diffusion and atom pair diffusion approaches, one can conclude that the Mg ions around C2N follow a two-stage diffusion mechanism, i.e., in the first stage, the out-of-plane Mg ions diffuse quickly, and in the second stage, the in-plane Mg ions diffuse moderately.

We also explore the diffusion behavior of Mg ions on a strained C2N anode. By employing the single Mg atom diffusion method, all possible migration paths are prepared (Fig. 3c and S14a). The diffusion barriers of paths I and II (both are 3.116 eV) and of paths i and ii (both are 2.579 eV) under strains of −10% and 10% are much smaller than the energy barriers of a strain-free C2N. By adopting the atom pair diffusion method, four paths are designed for the second Mg atom (out-of-plane atom) diffusing on C2N with −10% strain (Fig. 3d), and the largest energy barrier is 0.143 eV, which is much lower than that from the single atom diffusion (3.116 eV) under −10% strain and is much lower than that from the out-of-plane atoms in the strain-free C2N anode (0.916 eV). For the anode under 10% strain, an exclusive path (Fig. S14b) for the second Mg atom diffusion on the C2N monolayer is identified. Unfortunately, the calculated diffusion barrier is 2.788 eV, which is comparable to that from the single Mg atom approach (2.579 eV). The undiminished diffusion barrier from the atomic pair diffusion method under a 10% biaxial tensile strain could be ascribed to the fact that the second atom is still an in-plane atom. This is quite different from the second Mg atom (out-of-plane atoms) diffusion in strain-free states (Fig. 3b) and in −10% compressive strain states (Fig. 3d). When we carefully analyze the anode structure under −10% compressive strain, the coupling between Mg and N increases, decreasing the difference in the energies of out-of-plane adsorption sites and accelerating the ion diffusion of the out-of-plane atoms. Although the coupling between Mg and N increases in C2N under −10% compressive strain, the compressive strain generates tension in C2N, which will reduce the diffusion barriers of the in-plane atom in this situation. For the anode under 10% tensile strain, the vacancies surrounded by six N atoms enlarged, and as a result, the interaction between Mg and N becomes weak. This behavior can also be confirmed by the adsorption behavior.

The above points indicate that both compressive and tensile strains can lower the diffusion barriers of both in-plane and out-of-plane Mg atoms.

Zero point energy (ZPE) correction and the quantum mechanical tunneling (QMT) effect, which are also key factors for ion diffusion,33 are considered with and without strains (the details are provided in Section S4 of the ESI). Calculations (Fig. S15 and Table S2) show that (1) the ZPE corrections of energy barriers are in the range of 0.004–0.032 eV, (2) the QMT effect weakens as the temperature increases, and (3) the diffusion constants are promoted by the QMT effect.

2.2.3. Electronic properties. The charging/discharging rate is also affected by the electronic properties of anodes. To better understand the electronic properties of C2N with Mg ion loading, the density of states (DOS) of the most stable configurations with one, two, and five Mg atoms is computed (Fig. 4). The electrons donated from the Mg atoms to C2N shift down towards the low energy level, resulting in the Fermi level crossing the conduction band with sizable DOS at the Fermi level. Thus, the C2N changes from being a semiconductor to metallic. By further examining the projected density of states (PDOS), we find that there is a distinct overlap between the Mg-s orbitals and the p orbitals of C and N at the Fermi level, forming s–p hybridization. From Fig. 4, as the concentration of Mg atoms increases, the sub-gap below the Fermi level shrinks, and it almost vanishes when five Mg atoms are adsorbed. Correspondingly, the concentration of Mg ions promotes the electronic conductivity of C2N, which is in line with previous reports.36,37 For the maximum capacity of Mg5(C2N)6, the continuous and delocalization DOS curves show the typical metallic characteristic with outstanding electronic conductivity.
image file: c9nr03518d-f4.tif
Fig. 4 Density of states for one (a), two (b), and five (c) Mg atoms’ adsorption on C2N monolayers. The Fermi level was set as 0 eV and labeled with a red vertical dashed line.

The dependence of the electron conductivity of C2N on the strain is also investigated via the total DOS curves. Although the sub-gaps below the Fermi level change under the compressive and tensile strains, C2N with a single Mg atom maintains the metallic or semimetallic property (Fig. S16).

2.2.4 Open circuit voltage. For a strain-free C2N anode, low open circuit voltages (OCVs, see definition in Section S5 of the ESI or ref. 9 and 39) occur, implying the possibility of high net cell voltages (Fig. S17). All OCVs are larger than zero, indicating that the anode can effectively avoid the formation of metal clusters and dendrites. The OCV decreases upon increasing the concentration of Mg ions. The average OCV of 0.447 eV is located in the ideal potential range (0.10–1.00 eV) of a potential anode in the strain-free state. The advanced specific capacities occur in low OCV sections, indicating that C2N is a suitable anode material for MgIBs.39

From Fig. S17, one also can conclude that the average OCVs of C2N anodes in stretchable batteries under compressive and tensile strains remain in the ideal potential range. Compared with the strain-free C2N anode, the compressive strains increase the average OCVs (0.572 and 0.632 V for C2N anodes with −5% and −10% strains, respectively), whereas the tensile strains decrease the average OCVs (0.349 and 0.287 V for C2N anodes with 5% and 10% strains, respectively).

3. Conclusions

By employing the state-of-the-art density functional theory, we have investigated the superior performance of C2N anodes in stretchable MgIBs in strain-free and strained states. A C2N anode in stretchable MgIBs in the strain-free state delivers a maximum capacity of 1175 mA h g−1 and a low average OCV of 0.447 eV. For a C2N anode in the stretchable MgIBs in strained states, mechanical activation is an effective and facile approach to enhance its performance, i.e., large strains boost its capacities, while small strains maintain the high capacities of the anode. It has good cyclability and has small structural deformation during the loading of Mg atoms. Our calculations also showed that the Mg atoms have a two-stage diffusion mechanism, i.e., the out-of-plane Mg atoms migrate quickly in the first stage while the in-plane Mg atoms diffuse moderately in the second stage. Compared to the stretchable battery in the strain-free state, the barrier energies of the stretchable battery are reduced in strained states. It should be mentioned that most of the Mg atoms are out-of-plane atoms. The occurrence of the semiconductor–metal or semiconductor–semimetal transition with the loading of Mg ions promotes the charge/discharge rate of MgIBs in strain-free and strained states. We have attempted to provide fundamental insights into the mechanisms of high performance stretchable metal-ion batteries.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the University at Buffalo and by the National Natural Science Foundation of China (NSFC) (Grant No. 21206049, 51702114, 51872110), the Innovation Scientists and Technicians Team Construction Projects of Henan Province (CXTD2017002), the International Cooperation Project of Zhengzhou Science and Technology Bureau (Optimization and development of electrode for hydrogen production system of biological electrolysis battery, N2014SX0167), the Natural Science Foundation of Henan Province (172102210381), the Natural Science Foundation of Education Department of Henan Province (18A150011), and the Training Program of Youth Backbone Teacher of Henan Province of 2018 (Grant No. 2018GGJS178).

References

  1. I. You, B. Kim, J. Park, K. Koh, S. Shin, S. Jung and U. Jeong, Adv. Mater., 2016, 28, 6359–6364 CrossRef CAS.
  2. M. D. Dickey, Adv. Mater., 2017, 29, 1606425 CrossRef.
  3. D. Rus and M. T. Tolley, Nature, 2015, 521, 467 CrossRef CAS.
  4. K. Liu, Y. Liu, D. Lin, A. Pei and Y. Cui, Sci. Adv., 2018, 4, eaas9820 CrossRef.
  5. W. J. Song, S. Lee, G. Song and S. Park, ACS Energy Lett., 2019, 4, 177–186 CrossRef CAS.
  6. S. Xu, Y. Zhang, J. Cho, J. Lee, X. Huang, L. Jia, J. A. Fan, Y. Su, J. Su, H. Zhang, H. Cheng, B. Lu, C. Yu, C. Chuang, T. Kim, T. Song, K. Shigeta, S. Kang, C. Dagdeviren, I. Petrov, P. V. Braun, Y. Huang, U. Paik and J. A. Rogers, Nat. Commun., 2013, 4, 1543 CrossRef.
  7. J. M. Tarascon, Nat. Chem., 2010, 2, 510–510 CrossRef CAS.
  8. Y. Xie, D. Yohan, M. Naguib, Y. Gogotsi, M. W. Barsoum, H. L. Zhuang and P. R. C. Kent, ACS Nano, 2014, 8, 9606–9615 CrossRef CAS.
  9. B. Xiao, Y. C. Li, X. F. Yu and J. B. Cheng, ACS Appl. Mater. Interfaces, 2016, 8, 35342–35352 CrossRef CAS.
  10. D. Datta, J. W. Li and V. B. Shenoy, ACS Appl. Mater. Interfaces, 2014, 6, 1788–1795 CrossRef CAS.
  11. D. Er, J. W. Li, M. Naguib, Y. Gogotsi and V. B. Shenoy, ACS Appl. Mater. Interfaces, 2014, 6, 11173–11179 CrossRef CAS.
  12. Y. Shao, M. Gu, X. Li, Z. Nie, P. Zuo, G. Li, T. Liu, J. Xiao, Y. Cheng, C. Wang, J. Zhang and J. Liu, Nano Lett., 2013, 14, 255–260 CrossRef.
  13. Y. Wang, R. Chen, T. Chen, H. Lv, G. Zhu, L. Ma, C. Wang, Z. Jin and J. Liu, Energy Storage Mater., 2016, 4, 103–129 CrossRef.
  14. Z. S. Wu, W. Ren, L. Xu, F. Li and H. M. Cheng, ACS Nano, 2011, 5, 5463–5471 CrossRef CAS.
  15. W. J. Lee, T. H. Hwang, J. O. Hwang, H. W. Kim, J. Lim, H. Y. Jeong, J. Shim, T. H. Han, J. Y. Kim, J. W. Choi and S. O. Kim, Energy Environ. Sci., 2014, 7, 621–626 RSC.
  16. Z. Wang and G. Shao, J. Mater. Chem. A, 2018, 6, 6830 RSC.
  17. Y. Zhang, H. Geng, W. Wei, J. Ma, L. Chen and C. Li, Energy Storage Mater., 2019, 20, 118–138 CrossRef.
  18. I. A. Rodríguez-Pérez, Y. Yuan, C. Bommier, X. Wang, L. Ma, D. P. Leonard, M. M. Lerner, R. G. Carter, T. Wu, P. A. Greaney, J. Lu and X. Ji, J. Am. Chem. Soc., 2017, 139, 13031–13037 CrossRef PubMed.
  19. J. Lee, B. Monserrat, I. D. Seymour, Z. Liu, S. E. Dutton and C. P. Grey, J. Mater. Chem. A, 2018, 6, 16983–16991 RSC.
  20. H. Yaghoobnejad Asl, J. Fu, H. Kumar, S. S. Welborn, V. B. Shenoy and E. Detsi, Chem. Mater., 2018, 30, 1815–1824 CrossRef CAS.
  21. F. Wang, X. Fan, T. Gao, W. Sun, Z. Ma, C. Yang, F. Han, K. Xu and C. Wang, ACS Cent. Sci., 2017, 3, 1121–1128 CrossRef CAS.
  22. P. Saha, M. K. Datta, O. I. Velikokhatnyi, A. Manivannan, D. Alman and P. N. Kumta, Prog. Mater. Sci., 2014, 66, 1–86 CrossRef CAS.
  23. K. V. Kravchyk, L. Piveteau, R. Caputo, M. He, N. P. Stadie, M. I. Bodnarchuk, R. T. Lechner and M. V. Kovalenko, ACS Nano, 2018, 12, 8297–8307 CrossRef CAS.
  24. C. Chen, J. Wang, Q. Zhao, Y. Wang and J. Chen, ACS Energy Lett., 2016, 1, 1165–1172 CrossRef CAS.
  25. H. D. Yoo, I. Shterenberg, Y. Gofer, G. Gershinsky, N. Pour and D. Aurbach, Energy Environ. Sci., 2013, 6, 2265–2279 RSC.
  26. J. Muldoon, C. B. Bucur and T. Gregory, Chem. Rev., 2014, 114, 11683–11720 CrossRef CAS.
  27. M. Mao, T. Gao, S. Hou and C. Wang, Chem. Soc. Rev., 2018, 47, 8804–8841 RSC.
  28. C. Liu, Z. G. Neale and G. Cao, Mater. Today, 2016, 19, 109–123 CrossRef CAS.
  29. K. Tang, L. Fu, R. J. White, L. Yu, M. M. Titirici, M. Antonietti and J. Maier, Adv. Energy Mater., 2012, 2, 873–877 CrossRef CAS.
  30. A. Darwiche, C. Marino, M. T. Sougrati, B. Fraisse, L. Stievano and L. Monconduit, J. Am. Chem. Soc., 2012, 134, 20805–20811 CrossRef CAS.
  31. S. Zhang, Z. Yan, Y. Li, Z. Chen and H. Zeng, Angew. Chem., Int. Ed., 2015, 54, 3112–3115 CrossRef CAS.
  32. S. Zhang, S. Guo, Z. Chen, Y. Wang, H. Gao, J. Gómez-Herrero, P. Ares, F. Zamora, Z. Zhu and H. Zeng, Chem. Soc. Rev., 2018, 47, 982–1021 RSC.
  33. S. Guo, X. Hu, W. Zhou, X. Liu, Y. Gao, S. Zhang, K. Zhang, Z. Zhu and H. Zeng, J. Phys. Chem. C, 2018, 122, 29559–29566 CrossRef CAS.
  34. J. M. Tarascon and M. Armand, Nature, 2001, 414, 359 CrossRef CAS.
  35. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169 CrossRef CAS.
  36. D. Wu, B. Yang, H. Chen and E. Ruckenstein, Energy Storage Mater., 2019, 16, 574–580 CrossRef.
  37. A. Samad, A. Shafique and Y. H. Shin, Nanotechnology, 2017, 28, 175401 CrossRef.
  38. Y. H. Tan, W. T. Yao, T. Zhang, T. Ma, L. L. Lu, F. Zhou, H. B. Yao and S. H. Yu, ACS Nano, 2018, 12, 5856–5865 CrossRef CAS PubMed.
  39. D. Wang, Y. Liu, X. Meng, Y. Wei, Y. Zhao, Q. Pang and G. Chen, J. Mater. Chem. A, 2017, 5, 21370–21377 RSC.

Footnote

Electronic supplementary information (ESI) available: Computational methods; adsorption of Mg atoms on C2N (including the adsorption energy); stress–strain curves of C2N; specific capacity; calculations of open circuit voltages and their curves; possible positions of a single Mg atom adsorption on C2N; the most stable configuration and charge density difference of a single Mg atom adsorption on C2N; optimized structures for second Mg atom adsorption at different sites on C2N (with and without strains) in the atom pair adsorption model; the charge density difference for the most stable configurations of two Mg atoms adsorption on C2N with and without strains; the most stable configurations for three and four Mg atoms’ adsorption on C2N; the most stable configurations of five (strain-free) and thirteen (under −10% compressive strain) Mg atoms’ adsorption on C2N at the end of a 5 ps AIMD simulation; the most stable configurations for the maximum Mg atoms’ adsorption on C2N under biaxial strains; charge density differences for the most stable configurations of the maximum Mg atoms’ adsorption on C2N with and without strains; diffusion paths and their corresponding energy barriers of a single and two Mg atoms diffusing on C2N under 10% tensile strain; total density of states for the most stable configuration of a single Mg atom adsorption on C2N under biaxial strains; total energies of C2N with and without strains; Bader charge (|e|) of one and two Mg atoms, and of the maximum capacity per Mg atom for a C2N monolayer with and without external strains. See DOI: 10.1039/c9nr03518d

This journal is © The Royal Society of Chemistry 2019