English

• Metal-CO₂ batteries have recently attracted plenty of attention because they can utilize greenhouse gas CO₂, along with their high specific energy density [1-4]. Moreover, it is promising to adopt metal-CO₂ batteries as power supplies of future exploration on Mars, in which CO₂ accounts for 95% of the atmosphere [5, 6]. Metal Li-CO₂ batteries were firstly investigated and have been widely studied due to both high discharge potential (~2.8 V) and theoretical energy density (1876 Wh/kg) based on the ideal reaction of 4Li+3CO₂↔2Li₂CO₃+C [7-9]. Plenty of catalysts such as RuO₂, Ru, Mo₂C, NiO have been developed to promote the decomposition of thermodynamically stable discharge products (Li₂CO₃) [10-16]. Considering the limited resource of lithium, sodium with similar physical and chemical properties has also been applied as anode to build Na-CO₂ batteries, in which the theoretical energy density can reach 1125 Wh/kg through the reaction of 4Na+3CO₂↔2Na₂CO₃+C (~2.35 V) [8, 17, 18].

  Similar with Li-CO₂ batteries, the discharge products of Na-CO₂ batteries (Na₂CO₃) are also hard to decompose, leading to high charge overpotential and poor cycle stability [17-20]. In 2016, with electrolyte-treated multi-wall carbon nanotube (t-MWCNT) as cathode catalyst, we first enabled the rechargeable room temperature Na-CO₂ battery [5]. Since then, a handful of catalysts such as Co₂MnO_x decorated carbon fiber, ZnCo₂O₄@CNT, ketjen black carbon or CNT supported ruthenium nanoparticles (Ru@KB or Ru@CNT), have been designed to accelerate the kinetics of the battery [7, 18, 21, 22]. Among them, nano-Ru metal composites stand out with high catalytic characteristics towards Na₂CO₃ decomposition, but their high reactivities undergo the concerns of catalyzing the decomposition of electrolyte [23, 24]. As an alternative, RuO₂ catalyst, which is more stable than metal Ru, also exhibits decent electron conductivity. RuO₂@CNTs have been applied as cathode catalysts in Li-CO₂ batteries to facilitate the decomposition of Li₂CO₃ [25, 26], which triggers the interest to study the use of ruthenium oxide-based catalysts in Na-CO₂ batteries.

  Herein, a series of RuO₂ nanoparticles in situ loaded on activated multi-walled carbon nanotubes (RuO₂@a-MWCNTs) have been fabricated through the facile hydrolyzing reaction followed by calcination method [27-29] and acted as cathode catalysts to promote the decomposition of Na₂CO₃ in Na-CO₂ batteries. Through optimizing the temperature of calcination and the content of RuO₂, RuO₂@a-MWCNTs prepared at 150 ℃ with 49.7 wt% RuO₂ demonstrated the best overall performance. The charge/discharge overpotential has been reduced to 1.5 V, in contrast with pure a-MWCNTs catalyst (2.2 V). Further cycling test unveiled the charge voltage of Na-CO₂ batteries with above catalyst were still lower than 4.0 V after 40 cycles at 500 mA/g and after 90 cycles at 100 mA/g under same limited capacity of 500 mAh/g. The results of density functional theory (DFT) calculations showed that RuO₂ was the catalytically active site, and the Ru atoms on RuO₂ and the O atoms on Na₂CO₃ formed covalent bonds, which resulted in the charge transfer on the C=O, elongated the length of C=O bond on the Na₂CO₃, and promoted the decomposition of Na₂CO₃ with low charging overpotential.

  Various RuO₂@a-MWCNTs composites were synthesized through the facile hydrolyzing reaction followed by calcination method. The detailed preparation process was shown in Fig. 1a. The precursor powder from the same batch was divided into three parts and calcined at 150 ℃, 180 ℃, and 200 ℃ for 10 h to obtain the final products, which were respectively defined as RuO₂@a-MWCNTs-150 ℃, RuO₂@a-MWCNTs-180 ℃, and RuO₂@a-MWCNTs-200 ℃. The results from X-ray diffraction (XRD) indicated that the crystallization increased with the increasing of temperature (Fig. 1b). Pure a-MWCNTs displayed two diffraction peaks at 25.9° and 42.8° that could be assigned to the (002) and (100) planes [30], respectively. When the calcination temperature was 180 ℃ or higher, the obtained RuO₂ in the composite exhibited decent crystallinity and the XRD patterns matched well with the standard card (PDF#88–323-RuO₂) (Figs. S1 and S2 in Supporting information). For RuO₂@a-MWCNTs-150 ℃, the crystallization of RuO₂ was highly suppressed with only one broad diffraction peak centered at around 35°. Meanwhile, RuO₂ nanoparticles were ultrafine and evenly dispersed on carbon nanotubes (Fig. 1c). However, RuO₂ nanoparticles became larger and agglomerated together for RuO₂@a-MWCNTs-180 ℃ (Fig. 1d) or 200 ℃ (Fig. 1e).

  Figure 1

  

  Figure 1.  (a) Schematic diagram of preparing RuO₂@a-MWCNTs composite. (b) XRD patterns of a-MWCNTs, RuO₂@a-MWCNTs-150 ℃, RuO₂@a-MWCNTs-180 ℃ and RuO₂@a-MWCNTs-200 ℃. TEM images of (c) RuO₂@a-MWCNTs-150 ℃, (d) RuO₂@a-MWCNTs-180 ℃ and (e) RuO₂@a-MWCNTs-200 ℃.

  DownLoad: Full-Size Img PowerPoint

  Considering that the catalysts with highly dispersed and ultra-fine characteristics can provide more active sites to accelerate the decomposition of Na₂CO₃ [31], we chose the calcination temperature of 150 ℃ to further optimize the loadings of RuO₂ nanoparticles (Fig. S3 in Supporting information). The RuO₂ content in the RuO₂@a-MWCNTs composite was evaluated to be 56.8 wt% (labeled as RuO₂@a-MWCNTs-1), 49.7 wt% (labeled as RuO₂@a-MWCNTs-2), 27.9 wt% (labeled as RuO₂@a-MWCNTs-3) according to the result of thermogravimetric (TG) analysis in air atmosphere (Fig. 2a and Fig. S4 in Supporting information). Based on the N₂- adsorption-desorption measurement, the specific surface area of the composite gradually increased after decreasing the content of RuO₂, which were 147.8, 160.3, 183.8 and 212.8 m²/g for RuO₂@a-MWCNTs-1, RuO₂@a-MWCNTs-2, RuO₂@a-MWCNTs-3 and pristine a-MWCNTs, respectively (Fig. 2b). Nevertheless, the specific surface area for RuO₂@a-MWCNTs composite was still high enough to accommodate the discharge product of Na-CO₂ batteries with maintained porous structure [27].

  Figure 2

  

  Figure 2.  (a) TG curves of pure RuO₂, RuO₂@a-MWCNTs-1, RuO₂@a-MWCNTs-2, RuO₂@ a-MWCNTs-3 and a-MWCNTs. (b) Nitrogen-adsorption-desorption isotherms of RuO₂@a-MWCNTs-1, RuO₂@a-MWCNTs-2, RuO₂@a-MWCNTs-3 and a-MWCNTs. TEM image of (c) RuO₂@a-MWCNTs-1, (d) RuO₂@a-MWCNTs-2 and (e) RuO₂@a-MWCNTs-3. (f) HRTEM image of RuO₂@a-MWCNTs-2 and (g) C, O, Ru EDX mapping images of RuO₂@a-MWCNTs-2. (h) Ru 3p and (i) O 1s high-resolution XPS spectra of RuO₂@a-MWCNTs-2.

  DownLoad: Full-Size Img PowerPoint

  We carried out the Transmission electron microscope (TEM), high resolution transmission electron microscope (HRTEM) and energy dispersive X-ray spectroscopy (EDX) to further investigate the distribution of RuO₂ nanoparticles in a-MWCNTs (Figs. 2c-g and Figs. S5-S7 in Supporting information). High RuO₂ content (RuO₂@a-MWCNTs-1) would result in the aggregation of RuO₂ nanoparticles (Fig. 2c and Fig. S5). Reducing the content of RuO₂ nanoparticles (RuO₂@a-MWCNTs-2) enable the uniform distribution of particles (Fig. 2d and Fig. S6). The RuO₂ was more scattered when the content was further reduced (RuO₂@a-MWCNTs-3) (Fig. 2e and Fig. S7). Through calibrating the lattice fringes and the angles between them, the HRTEM image (Fig. 2f) confirmed that the RuO₂ of RuO₂@a-MWCNTs-2 belonged to the cubic phase (PDF#87–726-RuO₂) with average size of about 2 nm. EDX mapping of RuO₂@a-MWCNTs-2 demonstrated Ru and O elements were uniformly distributed (Fig. 2g). X-ray photoelectron spectroscopy (XPS) was adopted to analyze the element valence and bonding states. The XPS high-resolution spectrum of Ru 3p (Fig. 2h) could be fitted to two peaks at 464.0 eV and 486.7 eV, indicating the dominant +4 oxidation state of Ru [31]. The O 1s XPS spectrum could be fitted to three peaks (Fig. 2i). The peak at binding energy of 532.8 eV and 531.4 eV corresponded to C-O and C=O [32, 33]. Compared with the XPS spectrum of a-MWCNTs (Fig. S8 in Supporting information), there was an additional peak with a binding energy of 530.2 eV in RuO₂@a-MWCNTs-2, which was related to the Ru-O bond of RuO₂ [34].

  The above characterizations suggested that RuO₂@a-MWCNTs-2 took advantages of both high dispersibility and abundant active sites. In order to further study the relationship between the loading ratio of RuO₂ and the catalytic performance, we assembled Na-CO₂ coin cells with RuO₂@a-MWCNTs-1, RuO₂@a-MWCNTs-2, RuO₂@a-MWCNTs-3, or a-MWCNTs as cathode catalysts. The median charging voltage was a straightforward indicator to assess the catalytic capability for Na₂CO₃ decomposition (Figs. 3a-d and Figs. S9-S11 in Supporting information). Obviously, pure a-MWCNTs cathode exhibited the highest potential, exceeding 4 V even on the first lap. RuO₂@a-MWCNTs-3 cathode with lowest content of RuO₂ also demonstrated sharply increased overpotential that exceed 4 V on the 25^th lap. Both RuO₂@a-MWCNTs-1 and RuO₂@a-MWCNTs-2 displayed low median charging voltage, while RuO₂@a-MWCNTs-2 demonstrated higher durability and the charging potential was still lower than 4 V after 90 cycles. We attributed it to better dispersed RuO₂ nanoparticles/higher specific surface area than RuO₂@a-MWCNTs-1 and higher RuO₂ ratio than RuO₂@a-MWCNTs-3. Moreover, long cycle test at a limited capacity of 500 mAh/g demonstrated that Na-CO₂ batteries could afford the operation of 70 cycles at 500 mA/g (Fig. 3e), in which the charge voltage was still lower than 4.0 V after 40 cycles at 500 mA/g. The charging voltage only increased by 0.2 V on the 20^th lap when the current density increased from 100 mA/g to 500 mA/g (Fig. S12 in Supporting information). Fig. 3f and Fig. S13 (Supporting information) compared the performance of Na-CO₂ battery between RuO₂@a-MWCNTs-2 in this work and cathode catalysts reported in the literature (cut-off capacity was 500 mAh/g). Among all the cathode catalysts, RuO₂@a-MWCNTs-2 catalyst displayed decent charge voltage at low current density (100 mA/g) on the first cycle, and the lowest charge voltage at high current density (500 mA/g) with longest cycling life.

  Figure 3

  

  Figure 3.  Performance of Na-CO₂ battery. (a) Median charge voltage over cycles and (b) comparison of the 40^th charge−discharge profile of Na-CO₂ battery with RuO₂@a-MWCNTs-1, RuO₂@a-MWCNTs-2, RuO₂@a-MWCNTs-3 and a-MWCNTs cathode. (c) Cycle performance of Na-CO₂ battery using RuO₂@a-MWCNTs-2 cathode with a cutoff capacity of 500 mAh/g at 100 mA/g. (d) The terminal potentials of cycles with RuO₂@a-MWCNTs-2 cathode at 100 mA/g. (e) Cycle performance of Na-CO₂ batteries with a cutoff capacity of 500 mAh/g at 500 mA/g. (f) Comparison of charge voltage of Na-CO₂ batteries with different cathode catalysts.

  DownLoad: Full-Size Img PowerPoint

  In order to unveil how catalyst contributed to the reaction of Na-CO₂ batteries, ex-situ XRD and scanning electron microscope (SEM) characterization have been conducted at different states of charge and discharge (6 state points were marked in Fig. 4a). Na-CO₂ battery with RuO₂@a-MWCNTs-2 catalyst exhibited outstanding reversibility. From the XRD graph (Fig. 4b), three diffraction peaks of Na₂CO₃ at 34.6°, 37.5° and 41.7° were emerged after discharge, and disappeared when charging process was finished. The reversibility was also confirmed by SEM characterization. Plenty of granular products were generated on the carbon nanotube skeleton during the discharge process (Figs. 4c-e and Fig. S14 in Supporting information). The products gradually disappeared, and a porous structure reappeared during the charging process (Figs. 4f-h and Fig. S15 in Supporting information). Even after 20 cycles with a cutoff capacity of 500 mAh/g at 100 mA/g, the RuO₂@a-MWCNTs-2 cathode still showed intriguing catalytic properties for Na₂CO₃ decomposition, sharply contrasting with a-MWCNTs cathode on which a few by-products were accumulated (Fig. S16 in Supporting information). The intriguing results were due to the synergetic functions of CNTs and RuO₂, in which multi-walled carbon nanotubes could provide a huge specific surface area to hold the discharge products, and the RuO₂ nanoparticles could effectively reduce the energy barrier of electrochemical reaction during the charging process to promote the decomposition of Na₂CO₃. Moreover, the oxygen-containing functional groups carried by the activated carbon tubes (Figs. S17-S19 in Supporting information) enabled the uniform dispersion of RuO₂ nanoparticles [6, 35, 36].

  Figure 4

  

  Figure 4.  (a) 1^st charge and discharge curve of Na-CO₂ battery with marked points for characterizations. (b) Ex situ XRD of RuO₂@a-MWCNTs-2 cathode. Ex situ SEM images of RuO₂@a-MWCNTs-2 cathodes: (c) Pristine, (d) discharged 100 mAh/g, (e) discharged 500 mAh/g and (f) recharged 100 mAh/g, (g) recharged 300 mAh/g, (h) recharged 500 mAh/g.

  DownLoad: Full-Size Img PowerPoint

  To provide atomic understanding between RuO₂ based catalyst and Na₂CO₃, we further applied the density functional theory (DFT) calculations. Figs. 5a-c and Table S1 (in Supporting information) summarized the configurations of Na₂CO₃ on a-MWCNTs, RuO₂ and RuO₂@a-MWCNTs substrates. The results demonstrated that the adsorption energy of Na₂CO₃ on RuO₂ (−3.578 eV) and RuO₂@a-MWCNTs (−2.595 eV) were more negative than that on a-MWCNTs (−1.115 eV), indicating RuO₂ was the preferred sites to grow Na₂CO₃. Besides, compared with the substrate of a-MWCNTs (1.289 Å, 1.288 Å), the C=O bond of Na₂CO₃ on the substrate of RuO₂ (1.316 Å, 1.318 Å) and RuO₂@a-MWCNTs (1.318 Å, 1.314 Å) were lengthened. The lengthened bond meant that Na₂CO₃ was easier to decompose, proving that RuO₂ nanoparticles were beneficial for the charge process of Na-CO₂ batteries.

  Figure 5

  

  Figure 5.  The interactions between Na₂CO₃ and different cathodes. The optimized configurations, adsorption energies, and C=O bond lengths of Na₂CO₃ on (a) a-MWCNTs, (b) RuO₂ and (c) RuO₂@a-MWCNTs substrates. (d-f) The differential charge density of Na₂CO₃ adsorbed on a-MWCNTs, RuO₂ and RuO₂@a-MWCNTs substrates.

  DownLoad: Full-Size Img PowerPoint

  The differential charge density was then used to explain that why the C=O bond length was elongated (Figs. 5d-f). Different from a-MWCNTs system, the electron density around the C=O bond of Na₂CO₃ decreased obviously in the RuO₂ and RuO₂@a-MWCNTs system, which was due to the covalent bond that formed between O atoms on the Na₂CO₃ and the Ru atoms on the RuO₂. The newly formed Ru-O interaction directly weakened the C=O bond and thus favored the decomposition of Na₂CO₃. The DFT calculations unambiguously verified that the RuO₂@a-MWCNTs composite cathode catalyst was very effective for fabricating the high stable Na-CO₂ batteries.

  In summary, ultra-fine and highly dispersed RuO₂ nanoparticles (~2 nm) loaded on a-MWCNTs have been fabricated and for the first time, acted as cathode catalyst to reduce the charge overpotential and enhance the reversibility of Na-CO₂ batteries. a-MWCNTs with O-rich groups facilitated the dispersion of RuO₂ and RuO₂ promoted Na₂CO₃ decomposition by weakening the C=O bond of Na₂CO₃, which together contributed to build the Na-CO₂ batteries with long cycle life (> 120 cycles) and low charge potential (< 4.0 V up to 90 cycles). This work provides a general way to promote the decomposition of Na₂CO₃. With rational designation, more composite catalysts with advanced conductive host such as graphene, porous carbons, MXene can be prepared to further enhance the properties of metal-CO₂ batteries.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (Nos. 52001170, 21835004), the National Key R & D Program of China (Nos. 2017YFA0206700, 2021YFB2500300) and the Natural Science Foundation of Tianjin (No. 20JCQNJC02060).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.04.003.

  
  
• 1. [1]

     J. Sun, Y. Lu, H. Yang, et al., Research 2018 (2018) 6914626.

  2. [2]

     X. Mu, H. Pan, P. He, et al., Adv Mater. 32 (2019) e1903790.

  3. [3]

     Y. Hou, J. Wang, L. Liu, et al., Adv. Funct. Mater. 27 (2017) 1700564. doi: 10.1002/adfm.201700564

  4. [4]

     S. Shen, C. Han, B. Wang, et al., Chin. Chem. Lett. 33 (2022) 3721–3725. doi: 10.1016/j.cclet.2021.10.063

  5. [5]

     X. Hu, J. Sun, Z. Li, et al., Angew. Chem. Int. Ed. 55 (2016) 6482–6486. doi: 10.1002/anie.201602504

  6. [6]

     Y. Lu, Y. Cai, Q. Zhang, et al., Chem. Sci. 10 (2019) 4306–4312. doi: 10.1039/C8SC05178J

  7. [7]

     S. Thoka, C.M. Tsai, Z. Tong, et al., ACS Appl. Mater. Interfaces 13 (2021) 480–490. doi: 10.1021/acsami.0c17373

  8. [8]

     S. Yang, Y. Qiao, P. He, et al., Energy Environ. Sci. 10 (2017) 972–978. doi: 10.1039/C6EE03770D

  9. [9]

     S. Ma, Y. Lu, H. Yao, et al., Chin. Chem. Lett. 33 (2022) 2933–2936. doi: 10.1016/j.cclet.2021.10.089

  10. [10]

      Y. Dong, S. Li, S. Hong, L. Wang, B. Wang, Chin. Chem. Lett. 31 (2020) 635–642. doi: 10.1016/j.cclet.2019.08.021

  11. [11]

      X. Zhang, C. Wang, H. Li, et al., J. Mater. Chem. A 6 (2018) 2792–2796. doi: 10.1039/C7TA11015D

  12. [12]

      Y. Xing, Y. Yang, D. Li, et al., Adv. Mater. 30 (2018) e1803124. doi: 10.1002/adma.201803124

  13. [13]

      C. Yang, K. Guo, D. Yuan, et al., J. Am. Chem. Soc. 142 (2020) 6983–6990. doi: 10.1021/jacs.9b12868

  14. [14]

      Z. Zhang, Q. Zhang, Y. Chen, et al., Angew. Chem. Int. Ed. 54 (2015) 6550–6553. doi: 10.1002/anie.201501214

  15. [15]

      K. Chen, G. Huang, J.L. Ma, et al., Angew. Chem. Int. Ed. 59 (2020) 16661–16667. doi: 10.1002/anie.202006303

  16. [16]

      R. Pipes, A. Bhargav, A. Manthiram, Adv. Energy Mater. 9 (2019) 1900453. doi: 10.1002/aenm.201900453

  17. [17]

      Z. Zheng, C. Wu, Q. Gu, et al., J. Energy Environ. Mater. 4 (2021) 158–177. doi: 10.1002/eem2.12139

  18. [18]

      S. Thoka, Z. Tong, A. Jena, et al., J. Mater. Chem. A 8 (2020) 23974. doi: 10.1039/D0TA09235E

  19. [19]

      Z. Zhang, X.G. Wang, X. Zhang, et al., Adv. Sci. 5 (2018) 1700567. doi: 10.1002/advs.201700567

  20. [20]

      L. Qie, Y. Lin, J.W. Connell, et al., Angew. Chem. Int. Ed. 56 (2017) 6970–6974. doi: 10.1002/anie.201701826

  21. [21]

      L. Guo, B. Li, V. Thirumal, et al., Chem. Commun. 55 (2019) 7946–7949. doi: 10.1039/C9CC02737H

  22. [22]

      C. Fang, J. Luo, C. Jin, et al., ACS Appl. Mater. Interfaces 10 (2018) 17240–17248. doi: 10.1021/acsami.8b04034

  23. [23]

      X. Xing, I. Kimihiko, K. Yoshimi, Electrochim. Acta 261 (2018) 323e329.

  24. [24]

      S. Ma, Y. Wu, J. Wang, et al., Nano Lett. 15 (2015) 8084–8090. doi: 10.1021/acs.nanolett.5b03510

  25. [25]

      Z. Xie, X. Zhang, Z. Zhang, et al., Adv. Mater. 29 (2017) 1605891. doi: 10.1002/adma.201605891

  26. [26]

      S. Bie, M. Du, W. He, et al., ACS Appl. Mater. Interfaces 11 (2019) 5146–5151. doi: 10.1021/acsami.8b20573

  27. [27]

      Y. Eda, Y. Chihiro, Y. Keisuke, et al., Nano Lett. 13 (2013) 4679–4684. doi: 10.1021/nl4020952

  28. [28]

      P. Xua, C. Chen, J. Zhu, et al., J. Electroanal. Chem. 842 (2019) 98–106. doi: 10.1016/j.jelechem.2019.04.055

  29. [29]

      Z. Jian, P. Liu, F. Li, et al., Angew. Chem. Int. Ed. 53 (2014) 442–446. doi: 10.1002/anie.201307976

  30. [30]

      R. Wang, X. Zhang, Y. Cai, et al., Nano Res. 12 (2019) 2543–2548. doi: 10.1007/s12274-019-2482-9

  31. [31]

      S. -. M. Xu, Z. -. C. Ren, X. Liu, et al., Energy Storage Mater. 15 (2018) 291–298. doi: 10.1016/j.ensm.2018.05.015

  32. [32]

      M. Chuai, J. Yang, M. Wang, et al., eScience 1 (2021) 178–185. doi: 10.1016/j.esci.2021.11.002

  33. [33]

      S. Kundu, Y. Wang, W. Xia, et al., J. Phys. Chem. C 112 (2008) 16869–16878. doi: 10.1021/jp804413a

  34. [34]

      X. Gao, J. Chen, X. Sun, et al., ACS Appl. Nano Mater. 3 (2020) 12269–12277. doi: 10.1021/acsanm.0c02739

  35. [35]

      A. Huang, Y. Ma, J. Peng, et al., eScience 1 (2021) 141–162. doi: 10.1016/j.esci.2021.11.006

  36. [36]

      J.G. Zhou, H.T. Fang, Y.F. Hu, et al., J. Phys. Chem. C 113 (2009) 10747–10750. doi: 10.1021/jp902871b

• 

  Figure 1  (a) Schematic diagram of preparing RuO₂@a-MWCNTs composite. (b) XRD patterns of a-MWCNTs, RuO₂@a-MWCNTs-150 ℃, RuO₂@a-MWCNTs-180 ℃ and RuO₂@a-MWCNTs-200 ℃. TEM images of (c) RuO₂@a-MWCNTs-150 ℃, (d) RuO₂@a-MWCNTs-180 ℃ and (e) RuO₂@a-MWCNTs-200 ℃.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) TG curves of pure RuO₂, RuO₂@a-MWCNTs-1, RuO₂@a-MWCNTs-2, RuO₂@ a-MWCNTs-3 and a-MWCNTs. (b) Nitrogen-adsorption-desorption isotherms of RuO₂@a-MWCNTs-1, RuO₂@a-MWCNTs-2, RuO₂@a-MWCNTs-3 and a-MWCNTs. TEM image of (c) RuO₂@a-MWCNTs-1, (d) RuO₂@a-MWCNTs-2 and (e) RuO₂@a-MWCNTs-3. (f) HRTEM image of RuO₂@a-MWCNTs-2 and (g) C, O, Ru EDX mapping images of RuO₂@a-MWCNTs-2. (h) Ru 3p and (i) O 1s high-resolution XPS spectra of RuO₂@a-MWCNTs-2.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Performance of Na-CO₂ battery. (a) Median charge voltage over cycles and (b) comparison of the 40^th charge−discharge profile of Na-CO₂ battery with RuO₂@a-MWCNTs-1, RuO₂@a-MWCNTs-2, RuO₂@a-MWCNTs-3 and a-MWCNTs cathode. (c) Cycle performance of Na-CO₂ battery using RuO₂@a-MWCNTs-2 cathode with a cutoff capacity of 500 mAh/g at 100 mA/g. (d) The terminal potentials of cycles with RuO₂@a-MWCNTs-2 cathode at 100 mA/g. (e) Cycle performance of Na-CO₂ batteries with a cutoff capacity of 500 mAh/g at 500 mA/g. (f) Comparison of charge voltage of Na-CO₂ batteries with different cathode catalysts.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) 1^st charge and discharge curve of Na-CO₂ battery with marked points for characterizations. (b) Ex situ XRD of RuO₂@a-MWCNTs-2 cathode. Ex situ SEM images of RuO₂@a-MWCNTs-2 cathodes: (c) Pristine, (d) discharged 100 mAh/g, (e) discharged 500 mAh/g and (f) recharged 100 mAh/g, (g) recharged 300 mAh/g, (h) recharged 500 mAh/g.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  The interactions between Na₂CO₃ and different cathodes. The optimized configurations, adsorption energies, and C=O bond lengths of Na₂CO₃ on (a) a-MWCNTs, (b) RuO₂ and (c) RuO₂@a-MWCNTs substrates. (d-f) The differential charge density of Na₂CO₃ adsorbed on a-MWCNTs, RuO₂ and RuO₂@a-MWCNTs substrates.

  下载: 全尺寸图片 幻灯片
•