English

• In the past decade, lithium-ion batteries (LIBs) have been widely applied in portable electronics, electrical vehicles and other energy storage devices. To meet the demands with further increasement of low-cost and large-scale energy storage devices, sodium ion batteries (SIBs) or sodium-sulfur (Na-S) batteries are considered as promising candidate because of its abundant reserves and even distribution of sodium resource in earth crust [1–3]. Compared with Cu foil as anode current collectors for LIBs, Al foil that applied in SIBs is cost effectiveness. Although the fundamental mechanism is almost the same with LIBs, the molar mass of Na is heavier that is 23.00 compared to 6.94 (Li) and the atomic diameter is larger that is 1.02 Å regarded to 0.76 Å (Li-ion), which not only sacrifices in gravimetric and energy density of SIBs but also induces different sodium mechanism, such as complex electrochemical reactions and relatively low reaction kinetics [4,5].

  To develop high-energy SIBs, it is significant to design high-capacity and appropriate-redox-potential electrode materials. However, suitable high-performance anode materials for SIBs are still challenging [6,7]. Recently, red phosphorus (rP) as anode has been paid much attention to because of its highest theoretical capacity of 2596 mAh/g (in view of reaction of P with three Na⁺ to form Na₃P) [8,9]. But its low intrinsic electronic conductivity and large volume swelling during cycling are to the disadvantage of application [10]. Introducing metal to phosphors to form metal phosphides (MP_x) is a proven method to mitigate the issues, for example Cu, Sn, and Ge [11–13]. Carbon coating, such as hard carbon, multilayer graphene (mG) and carbon nanotubes were also widely applied to enhance the conductivity and release the mechanical stress during cycling [14–17]. Additionally, the discharge product of MP_x-Na₃P has been investigated by X-ray diffraction (XRD), high resolution transmission electron microscopy (HRTEM) and selected area electron diffraction (SAED) [18–22]. The deficiencies of above conventional diffraction measurements are difficult to obtain information about amorphous phase, thus the electrochemical sodium storage mechanisms during (dis)charging are not fully understood.

  In this work, a copper phosphide and reduced graphene oxide (CuP₂/C) composite material was synthesized by a simple ball milling method, which show a high reversible specific capacity of > 900 mAh/g. Taking this composite anode as a sample, the sodiation and desodiation mechanism and quantitative phase transformation CuP₂/C were explored and revealed by the techniques of synchrotron XRD, ex-situ XPS and solid-state NMR. For the first time, it is found that the native PO_x on the surface of CuP₂/C composite show higher electrochemical reversibility than the bulk CuP₂, due to the existence of Cu catalyst. The PO_x firstly sodiates and forms Na₃PO₄ and Na₄P₂O₇ during the first discharge process, after that the above sodiation products react with Na-P alloys and recover to PO_x during charge process. The reversible phase transformation of PO_x accounts for the high reversible capacity of the CuP₂/C composite anode. The findings also illustrate that the phosphorus transforms into nanocrystalline NaP, Na₃P and Na_xP alloys, among which the size of nanocrystals becomes smaller along cycling and shows crystallization-amorphization evolution process during long-life test. Our work sheds light on the material design of phosphorus-based anode with high capacity and electrochemical reversibility toward high energy density SIBs.

  The synthesis details of CuP₂/C are provided in Experimental Section in Supporting information. For simplicity, CuP₂/C is firstly taken as an example for illustration. Fig. S1a (Supporting information) shows the XRD patterns of CuP₂ and CuP₂/C materials, all peaks can be assigned to CuP₂ phase (JCPDS No. 076–1190) with a monoclinic lattice structure [23]. Compared with CuP₂, the CuP₂/C composite exhibits almost the same diffraction peaks, implying the ball milling process does not change the crystalline nature of CuP₂. To investigate the chemical composition and bonding state of above materials, XPS was carried out. In P 2p spectra in Fig. 1a, there is a broad peak at around 130 eV for both samples, which is matched with the CuP₂ and pristine rP. Additionally, broad peaks appear at 132.9, 133.7 and 135.0 eV due to the surface oxidation of rP, the first two peaks indicate the formation of P−O−P and O−P=O bonds [24], and the last one belongs to P₂O_5. The P₂O₅ was obtained during the ball milling process, which is inevitably incorporating some extent of air. At the same time, the size becomes smaller and the surface area increase, as a result, the reactivity between P and O increases and forms some native P₂O₅ on the surface. The peak at 134.3 eV corresponds to C−P=O bond, in which the C element is probably due to the inevitable C-containing contaminants in CuP₂. Conspicuously, above four peaks all intensified in the CuP₂/C composite. The XPS Cu 2p spectra in Fig. S1b (Supporting information) shows the two characteristic peaks, which are associated with Cu(Ⅰ) 2p_3/2 (932.9 eV) and 2p_1/2 (952.7 eV), respectively [25]. Solid-state ³¹P NMR spectra in Fig. 1b were collected to present a quantitative understanding about the materials. Two intense peaks located at 15 and −91 ppm is observed for both samples, corresponding to two P sites in the crystal structure of CuP₂ as illustrated in Fig. S2 (Supporting information). Another two broad resonances at ~62.7 and −95.2 ppm is ascribed to the unreacted rP. A broad peak located at ~15 ppm attributed to the P−O bonding is distinguished. A weak peak at ~−20 ppm detected in CuP₂/C composite is assigned to the C−P=O bond, which is produced during the ball-milling process by the reaction of reduced graphene oxide and PO_x. The morphologies of CuP₂ and CuP₂/C materials are characterized by SEM and TEM. In Fig. S1c (Supporting information), CuP₂ is composed of irregular shape of particles in a range of 300−600 nm. While for CuP₂/C composite in Fig. S1d (Supporting information), smaller particles with a diameter of 100−300 nm observed, which should be generated during ball milling process. High resolution TEM (HRTEM) image in Fig. S1e (Supporting information) manifests that CuP₂/C composite has a crystal core with amorphous carbon shell. The marked parallel fringes have the space of ~0.289 nm, corresponding to the (112) plane of CuP₂, which is in good agreement with the XRD patterns. Such a carbon layer will enhance the electronic conductivity and accommodate the volume variation during (dis)charge.

  Figure 1

  

  Figure 1.  The physical characterization of as-synthesized CuP₂ and CuP₂/C materials: (a) XPS P 2p spectra and (b) solid-state ³¹P NMR spectra. The asterisks indicate the spinning sidebands. The electrochemical profiles of CuP₂/C material as anode for SIBs: (c) CV curves at a scan rate of 0.05 mV/s and (d) electrochemical profiles during cycling at 50 mA/g. The specific capacity and applied current densities are based on the mass of CuP₂ material exclusive of carbon.

  DownLoad: Full-Size Img PowerPoint

  The electrochemical profiles were firstly evaluated by CV between 0 and 3 V at 0.05 mV/s versus Na⁺/Na and presented in Fig. 1c. During the initial cathodic sweep, there is a broad wave ranging from 1.6 V to 0.7 V, which is indicative of the decomposition of electrolyte and the formation of SEI (solid electrolyte interphase) layer on the fresh electrode. Upon further sodiation to 0 V, the current density increases continuously, stating that the occurrence of electrochemical sodiation on CuP₂/C electrode. In the ensuing anodic sweep, two pronounced peaks centered at 0.70 and 0.89 V correspond to multi-step desodiation process from NaP and Na₃P to P. By contrast, in the second discharge process, a peak located at 0.32 V appears contributed to the sodiation of amorphous rP, and the broad wave (1.6−0.7 V) disappears due to the absence of SEI contribution, which is different from the first cycle. The specific capacity is calculated based on the total mass of CuP₂, PO_x and carbon, the mass loading is about 2 mg/cm² on each of the discs for performance tests and 4 mg/cm² for synchrotron XRD, ex-situ NMR characterization, respectively. Fig. 1d shows the voltage profiles of CuP₂/C electrode at a current density of 50 mA/g, the specific capacity of the initial discharge can reach up to 1200 mAh/g (normalized to the mass of CuP₂ exclusive of carbon), and roughly 900 mAh/g is recovered when the electrode is fully charged. Notably, the reversible capacity of ~900 mAh/g is lower than the theoretical value of 1281 mAh/g according to the formation of Na₃P, which has been reported in the P-based sodium storage materials [26–31]. The rate performance and high rate cycling were also tested. As shown in Fig. S3 (Supporting information), the specific capacities of CuP₂/C at 0.05, 0.2, 0.5, 1, and 2 A/g are 950, 880, 815, 708 and 595 mAh/g, respectively. Even at a high rate of 1 A/g, the CuP₂/C anode could deliver a reversible capacity of 550 mAh/g after 50 cycles (Fig. S4 in Supporting information). With the purpose of elucidating the underlying sodiation/desodiation mechanism of CuP₂/C composite, we disassemble the coin cells at certain (dis)charged states and track the phase evolution by ex-situ studies.

  To investigate the reaction mechanism of CuP₂/C composite as anode for SIBs, the crystal structure variation during (dis)charge process is researched by synchrotron XRD (Fig. S5 in Supporting information). The pattern of each state shows no pronounced peaks ascribed to other crystal phases as (dis)charge proceeds, indicating that amorphous phases present in addition to the residual CuP₂ material.

  The surface oxidation states of the CuP₂/C electrode at selected voltage points during cycling, as shown in Fig. 2a, were investigated by ex-situ XPS. Fig. 2b displays the XPS spectra of Cu 2p at (dis)charged states: the pristine electrode featuring Cu(Ⅰ) 2p_1/2 (953.1 eV) and Cu 2p_3/2 (933.2 eV) [25]. As discharge goes on, the peaks of Cu 2p all move to lower E_b. When fully charged to 3 V (Ⅵ), all of peaks shifted in the higher E_b direction, implying the occurrence of reversible electrochemical reaction. And at 2D0V (discharged to 0 V in the second discharge process, similarly hereinafter) state, it is distinct that no peaks of Cu(Ⅰ) are observed compared with 1D0V state, indicating possibly that the influence of high S/N (signal/noise) ratio. In Fig. 2c, the P 2p spectra at different voltages display a peak centered at ~132 eV, which is attributed to the sodium phosphorus intermediates, Na_xP (x = 1–3) [32]. The peak at ~130 eV is corresponding to the rP and CuP₂, which is consistent with the result in Fig. 1a.

  Figure 2

  

  Figure 2.  (a) The electrochemical profiles of CuP₂/C electrodes for SIBs at 25 mA/g. (b) XPS Cu 2p spectra and (c) XPS P 2p spectra_.

  DownLoad: Full-Size Img PowerPoint

  In addition, Cu-K edge XAS analysis was employed to further resolve the structural ambiguities of above weak XRD patterns and to probe the structural variations around Cu ions during electrochemical process of the CuP₂/C electrode. In the X-ray absorption near-edge structure (XANES) region (Fig. S6a in Supporting information), the pristine electrode exhibited a Cu^I-like characteristic feature with a peak located at ~8986 eV assigned to a single bond of 1s → 4pπ^*, the intensity of which suggests the fourfold coordination geometry of the Cu site. This hypothesis is evidenced by the structure presentation of CuP₂ as illustrated in Fig. S2. For Cu^I, the transition shows highest intensity for the linear two-coordinated complexes and becomes less intense and broaden as the coordination number increases [33]. This peak gradually becomes weak with an increase of the depth of discharge. Upon subsequent charge, the XANES feature returned incompletely to its initial state. The second discharge process is similar with the first one. More interestingly, the radial distribution functions of the Fourier-transformed (FT) k²-weight extended X-ray absorption fine structure (EXAFS) spectra (Fig. S6b in Supporting information) present a distinct geometric structure variation around the central Cu ion. Yet surprisingly, an apparent peak at ~2 Å corresponding to the Cu-O shell is viewed. Upon discharge, this peak shifts to the higher r space region, indicating a distinct elongating of the bond length of the CuP₂/C electrode. During the consecutive charge process, the structure is reversibly recovered. Ex-situ XRD was performed to further confirmed the crystalline phase transformation of CuP₂/C electrode for the first cycle and second discharge processes (Fig. S7 in Supporting information). In Fig. S5a, the diffraction intensity of crystalline CuP₂ decrease during the discharge process, implying the CuP₂ transforms into sodiation products. Even after recharged to 3 V, no obvious diffraction peak was observed, demonstrating the CuP₂ becomes amorphous after initial cycle. Expectedly, no crystalline peak was detected during the second discharge process (Fig. S5b). The ex-situ XRD results further confirms the phase transformation of CuP₂ during charge and discharge. The crystalline phase transformation of CuP₂/C during cycling was revealed while the amorphous phases needed to be further explored.

  To elucidate the structural evolution including crystal and amorphous states of the CuP₂/C composite during cycling, the usage of solid-state ³¹P NMR (Fig. 3) was conducive to circumvent the disadvantage of the traditional diffraction method, and quantify the variation of amorphous products. Accordingly, the peaks at −91 and 15 ppm are ascribed to CuP₂ phase, the areal ratio of which is around 1:1. The peaks at 14.1 and 2.5 ppm are Na₃PO₄ and Na₄P₂O₇, respectively [34,35]. In this case, with the Cu catalyst, the PO_x on the surface tends to be electrochemical active and transforms into Na-P-O at the early stage of the sodiation process, such as 1D0.65 V. With the sodiation process continues, the Na-P-O reacts with more Na atoms and forms Na₃PO₄ and other being Na₄P₂O₇ products, as a result, the peak the transition state of phosphates is divided into two peaks. Red phosphorus shows two relatively broad peaks with FWHM of ~100 ppm, which are located at −95.2 and 62.7 ppm [36]. The presence of two environments of associated ³¹P shifts at −29 and −78 ppm that are correspond to amorphous NaP. While for Na₃P, there are four environments of shifts at −118, −180, −238 and −280 ppm. Thereinto, the peak of −238 ppm with width of 10 ppm is probably originated in the Na_xP species formed in side reactions, minor Na_xP intermediates, or Na near P defects (such as the end of a P chain) [37]. Based on above discussion, we can analysis each ex-situ state during cycling.

  Figure 3

  

  Figure 3.  Tracking the structural evolution of CuP₂/C electrodes for SIBs in the first cycle by solid-state ³¹P NMR spectroscopy: (a) The electrochemical profiles of the CuP₂/C electrode at a current of 25 mA/g. (b) Solid-state ³¹P NMR spectra of the CuP₂/C electrodes at selected states. Tracking the structural evolution of CuP₂/C electrodes for SIBs in the second discharge process by solid-state ³¹P NMR spectroscopy: (c) The electrochemical profiles of the CuP₂/C electrode at a current of 25 mA/g. (d) Solid-state ³¹P NMR spectra of the CuP₂/C electrodes at selected states.

  DownLoad: Full-Size Img PowerPoint

  Solid-state ³¹P NMR spectra were further collected to gain a quantitative understanding on the prepared materials. The content of PO_x on the surface was determined by NMR fitting analysis. Two intense ³¹P NMR peaks located at 15 and −91 ppm, corresponding to two P sites in CuP₂. Two other broad resonances at 62.7 and −95.2 ppm which are due to the unreacted rP as shown in the fitting plots. And a broad peak located at ~15 ppm is distinguished and attributed to the P–O bonding. It is noted that the fitting analysis indicates that the atomic ratio of P in CuP₂, rP and P–O bonds (PO_x) is around 7:1.5:1.5. Also, the contents keep dropping as discharge proceeds on. Additionally, the peak of C−P=O bond shows the shift at −21 ppm with a width of 25 ppm, which is located at the same position of PO_x. During the first discharge, when the voltage is downs to 0.65 V (1D0.65 V, discharged to 0.65 V in the first discharge process, similarly hereafter), there is a peak with FWHM of ~20 ppm at 2.9 ppm accompanied with the vanishment of C−P=O bond, which is attributed to the transition state of phosphates. Then in the state of 1D0.26 V, the peak of mentioned transition state is divided into two peaks. One is the peak at 14.1 ppm with width of 12 ppm of Na₃PO₄, the other is the peak at 2.5 ppm with width of 8 ppm of Na₄P₂O₇. It is worth mentioning that the content of Na₃PO₄ gradually increases but that of Na₄P₂O₇ is almost unchanged. Meanwhile, the phases of Na_xP and amorphous NaP (a-NaP) are observed along with the disappearance of PO_x. The peak of Na_xP centers at 55 ppm with width of 40 ppm. In the next state (1D0.16 V), rP vanishes. When the voltage discharged to 0.12 V (1D0.12 V), the phase of Na₃P generates, three of which could be viewed (shifts at −118, −180, −238 ppm with FWHM of 30, 40, 10 ppm, respectively). Above the contents of a-NaP and Na₃P all increase with discharged depth raising. For 1D0.06 V, fourth peak of Na₃P appears at −280 ppm with width of 50 ppm. In short, the discharge process is linked to a conversion of CuP₂, rP and PO_x to Na-P alloys and phosphates. To demonstrate the role of Cu on the electrochemical reversibility of rP, the NMR data of P/C during charge and discharge was carried out. As shown in Fig. S8 (Supporting informatio), the PO_x peak remains almost the same during initial discharge and charge processes, indiacating that the PO_x is electrochemical inactive, which is totally different with the CuP₂ results. Based on the above results, it is concluded that, the native oxide PO_x on the surface of CuP₂/C becomes electrochemical active and reversible under the existence of Cu catalyst.

  To better understand the sodiation/desodiation mechanism of CuP₂/C, based on the ³¹P NMR spectra during the galvanostatic (dis)charge processes, the quantitative phase transformation diagram is analyzed (detailed in Table S1 in Supporting information) and shown in Fig. 4. Generally speaking, the diagram exhibits the reversibility of CuP₂, rP and PO_x, especially for the high electrochemical reversibility of PO_x. After the first desodiation, only a small ratio of CuP₂ convert to the initial state but PO_x shows high reversibility. It should be noted that the content of PO_x is higher than the pristine CuP₂/C before cycling, which may be caused by atomic structure reconstruction between P and O atoms during cycling. Instead, the Na-O-P and Na-P alloy derived from CuP₂ convert to PO_x, rP and CuP₂, which should be the main reason responsible for the high reversibility of the unique CuP₂/C anode. Besides, after the initial desodiation process, the residual Na-P alloy such as Na_xP, Na₃PO₄ and Na₄P₂O₇ may contribute to the irreversible capacity loss and low Coulombic efficiency. It is suggested that the mechanical stress caused by volume expansion could be regulated by controlling the content of PO_x in the phosphorus-based anodes. In addition, reducing the content of Na_xP, Na₃PO₄ and Na₄P₂O₇ is expected to improve the reversibility and Coulombic efficiency. For the second cycle, the contents of NaP and Na₃P alloy increased compared to the initial cycle, which may be induced by the particles fracture and converting into nanosized P, thus enabling fast sodiation kinetics. During the sodiation process of rP, Na_xP and NaP forms at the same time with the disappearance of PO_x. With the sodiation continues, more Na atoms insert into and forms more NaP and Na_xP. At the end of sodiation process, some extent of Na_xP trnaforms into Na₃P. The NaP and Na₃P tend to be amorphous and degree of disorder increase along repeated cycles.

  Figure 4

  

  Figure 4.  Phase transformation diagram for the CuP₂/C composite with the galvanostatic (dis)charge processes: (a) The contents of produced Na-P alloys and Na-P-O compounds can be directly read in the Y axis, depending on the deconvolution results of ex-situ ³¹P NMR spectra in Fig. 3. (b) The electrochemical profiles of the CuP₂/C electrode cycling at a current of 25 mA/g.

  DownLoad: Full-Size Img PowerPoint

  In summary, combination of synchrotron XRD, ex-situ XPS and NMR, it is revealed that native oxidation components such as PO_x of the CuP₂/C composite show better electrochemical reversibility even than the bulk CuP₂. The surface PO_x firstly sodiate and generate Na₃PO₄ and Na₄P₂O₇ during charge process, and substantially react with Na-P alloys and recover to the PO_x during the desodiation process. This is a critical factor that accounts for the high reversible capacity of 900 mAh/g for the CuP₂/C composite. It is also demonstrated that the formation of amorphous Na₃P and Na_xP along repeated cycles. This work for the first time provides a quantitative understanding on the electrochemical sodiation mechanism of the CuP₂/C composite, which shed light on a new road to design high-capacity P-based anode materials for sodium storage by manipulating the content and bonding nature between P and oxygen.

  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.

  Acknowledgements

  This work was financially supported by National Nature Science Foundation of China (Nos. 21805278, 22272175 and 22209075), the Fujian Science and Technology Planning Projects of China (Nos. 2022T3067 and 2023H0045), the Self-deployment Project Research Programs of Haixi Institutes, Chinese Academy of Sciences (No. CXZX-2022-JQ12), the Self-deployment project of XIREM (No. 2023GG02).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109650.

  
  
• 1. [1]

     N.T. Aristote, K. Zou, A. Di, et al., Chin. Chem. Lett. 33 (2022) 730–742. doi: 10.1016/j.cclet.2021.08.049

  2. [2]

     C. Ma, X. Wang, J. Lan, et al., Adv. Funct. Mater. 33 (2022) 2211821.

  3. [3]

     J. He, A. Bhargav, L. Su, et al., Nat. Commun. 14 (2023) 6568. doi: 10.1038/s41467-023-42383-3

  4. [4]

     Na-Ion batteries K. Kubota, S. Komaba, S. Passerini, D. Bresser, A. Moretti, A. Varzi, Encyclopedia of Electrochemistry Wiley-VCH Verlag GmbH & Co KGaA, Weinheim, 2020, pp. 1–64.

  5. [5]

     N. Yabuuchi, K. Kubota, M. Dahbi, et al., Chem. Rev. 114 (2014) 11636–11682. doi: 10.1021/cr500192f

  6. [6]

     H. Wang, H. Chen, C. Chen, et al., Chin. Chem. Lett. 34 (2023) 107465. doi: 10.1016/j.cclet.2022.04.063

  7. [7]

     B. Qin, M. Wang, S. Wu, et al., Chin. Chem. Lett. 35 (2024) 108921. doi: 10.1016/j.cclet.2023.108921

  8. [8]

     H. Tan, D. Chen, X. Rui, et al., Adv. Funct. Mater. 29 (2019) 1808745. doi: 10.1002/adfm.201808745

  9. [9]

     J. Ge, C. Ma, Y. Wan, et al., Adv. Funct. Mater. 33 (2023) 2305803. doi: 10.1002/adfm.202305803

  10. [10]

      F.H. Yang, H. Gao, J. Chen, et al., Small Methods 1 (2017) 1700216. doi: 10.1002/smtd.201700216

  11. [11]

      S.O. Kim, A. Manthiram, ACS Appl. Mater. Interfaces 9 (2017) 16221–16227. doi: 10.1021/acsami.7b02826

  12. [12]

      Y. Kim, Y. Kim, A. Choi, et al., Adv. Mater. 26 (2014) 4139–4144. doi: 10.1002/adma.201305638

  13. [13]

      W. Li, X. Li, J. Liao, et al., Energy Storage Mater. 20 (2019) 380–387. doi: 10.1016/j.ensm.2019.04.034

  14. [14]

      Y. Kim, Y. Park, A. Choi, et al., Adv. Mater. 25 (2013) 3045–3049. doi: 10.1002/adma.201204877

  15. [15]

      J. Qian, X. Wu, Y. Cao, et al., Angew. Chem. Int. Ed. 52 (2013) 4633–4636. doi: 10.1002/anie.201209689

  16. [16]

      J. Sun, H.W. Lee, M. Pasta, et al., Nat. Nanotechnol. 10 (2015) 980–985. doi: 10.1038/nnano.2015.194

  17. [17]

      W.J. Li, S.L. Chou, J.Z. Wang, et al., Nano Lett. 13 (2013) 5480–5484. doi: 10.1021/nl403053v

  18. [18]

      Q. Liu, Z. Hu, Y. Liang, et al., Angew. Chem. Int. Ed. 59 (2020) 5159–5164. doi: 10.1002/anie.201913683

  19. [19]

      Y. Lu, P. Zhou, K. Lei, et al., Adv. Energy Mater. 7 (2017) 1601973. doi: 10.1002/aenm.201601973

  20. [20]

      L. Ran, B. Luo, I.R. Gentle, et al., ACS Nano 14 (2020) 8826–8837. doi: 10.1021/acsnano.0c03432

  21. [21]

      T. Wang, K. Zhang, M. Park, et al., ACS Nano 14 (2020) 4352–4365. doi: 10.1021/acsnano.9b09869

  22. [22]

      X. Wang, K. Chen, G. Wang, et al., ACS Nano 11 (2017) 11602–11616. doi: 10.1021/acsnano.7b06625

  23. [23]

      C. Zhao, H. Chen, H. Liu, et al., J. Mater. Chem. A 9 (2021) 6274–6283. doi: 10.1039/D0TA12141J

  24. [24]

      P. Nakhanivej, X. Yu, S.K. Park, et al., Nat. Mater. 18 (2019) 156–162. doi: 10.1038/s41563-018-0230-2

  25. [25]

      A.Y. Kim, M.K. Kim, K. Cho, et al., ACS Appl. Mater. Interfaces 8 (2016) 19514–19523. doi: 10.1021/acsami.6b05973

  26. [26]

      Q. Li, Z. Li, Z. Zhang, et al., Adv. Energy Mater. 6 (2016) 1600376. doi: 10.1002/aenm.201600376

  27. [27]

      F.P. Zhao, N. Han, W.J. Huang, et al., J. Mater. Chem. A 3 (2015) 21754–21759. doi: 10.1039/C5TA05781G

  28. [28]

      S. Kaushik, J. Hwang, K. Matsumoto, et al., ChemElectroChem 5 (2018) 1340–1344. doi: 10.1002/celc.201800206

  29. [29]

      S.O. Kim, A. Manthiram, Chem. Commun. 52 (2016) 4337–4340. doi: 10.1039/C5CC10585D

  30. [30]

      S. Chen, F. Wu, L. Shen, et al., ACS Nano 12 (2018) 7018–7027. doi: 10.1021/acsnano.8b02721

  31. [31]

      Y. Zhang, G. Wang, L. Wang, et al., Nano Lett. 19 (2019) 2575–2582. doi: 10.1021/acs.nanolett.9b00342

  32. [32]

      Y. Yan, S. Xia, H. Sun, et al., Chem. Eng. J. 393 (2020) 124788. doi: 10.1016/j.cej.2020.124788

  33. [33]

      I.J. Pickering, G.N. George, C.T. Dameron, et al., J. Am. Chem. Soc. 115 (1993) 9498–9505. doi: 10.1021/ja00074a014

  34. [34]

      S. Hayashi, K. Hayamizu, B. Chem. Soc. Jap. 62 (1989) 3061–3068. doi: 10.1246/bcsj.62.3061

  35. [35]

      S. Hayashi, K. Hayamizu, Chem. Phys. Lett. 161 (1989) 158–162. doi: 10.1016/0009-2614(89)85049-3

  36. [36]

      C. Marino, M. El Kazzi, E.J. Berg, et al., Chem. Mater. 29 (2017) 7151–7158. doi: 10.1021/acs.chemmater.7b01128

  37. [37]

      L.E. Marbella, M.L. Evans, M.F. Groh, et al., J. Am. Chem. Soc. 140 (2018) 7994–8004. doi: 10.1021/jacs.8b04183

• 

  Figure 1  The physical characterization of as-synthesized CuP₂ and CuP₂/C materials: (a) XPS P 2p spectra and (b) solid-state ³¹P NMR spectra. The asterisks indicate the spinning sidebands. The electrochemical profiles of CuP₂/C material as anode for SIBs: (c) CV curves at a scan rate of 0.05 mV/s and (d) electrochemical profiles during cycling at 50 mA/g. The specific capacity and applied current densities are based on the mass of CuP₂ material exclusive of carbon.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) The electrochemical profiles of CuP₂/C electrodes for SIBs at 25 mA/g. (b) XPS Cu 2p spectra and (c) XPS P 2p spectra_.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Tracking the structural evolution of CuP₂/C electrodes for SIBs in the first cycle by solid-state ³¹P NMR spectroscopy: (a) The electrochemical profiles of the CuP₂/C electrode at a current of 25 mA/g. (b) Solid-state ³¹P NMR spectra of the CuP₂/C electrodes at selected states. Tracking the structural evolution of CuP₂/C electrodes for SIBs in the second discharge process by solid-state ³¹P NMR spectroscopy: (c) The electrochemical profiles of the CuP₂/C electrode at a current of 25 mA/g. (d) Solid-state ³¹P NMR spectra of the CuP₂/C electrodes at selected states.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Phase transformation diagram for the CuP₂/C composite with the galvanostatic (dis)charge processes: (a) The contents of produced Na-P alloys and Na-P-O compounds can be directly read in the Y axis, depending on the deconvolution results of ex-situ ³¹P NMR spectra in Fig. 3. (b) The electrochemical profiles of the CuP₂/C electrode cycling at a current of 25 mA/g.

  下载: 全尺寸图片 幻灯片
•