Elsevier

Nano Energy

Volume 34, April 2017, Pages 456-462
Nano Energy

Full paper
Origin of excellent rate and cycle performance of Na+-solvent cointercalated graphite vs. poor performance of Li+-solvent case

https://doi.org/10.1016/j.nanoen.2017.03.015Get rights and content

Highlights

  • Li+- and Na+-solvent cointercalations into graphite were examined.

  • Solvent cointercalation enables Na+ intercalation into graphite thermodynamically.

  • Na+-solvent transport is strikingly faster than Li+-solvent transport.

  • Weak Na+ ion–graphene interaction significantly enhances rate capability.

  • Slow Li+-diglyme conductivity is ascribed to steric hindrance to codiffusion.

Abstract

Despite its high reversibility for Li+ intercalation, graphite is known to be electrochemically inactive for Na+ intercalation. On the contrary, recent studies have demonstrated that graphite is active and shows excellent rate and cycle performance for Na+-solvent cointercalation but it exhibits poor performance for Li+-solvent cointercalation. Herein, we elucidate the mechanism of Li+- and Na+-solvent cointercalation into graphite and the origin of the strikingly different electrochemical performance of Li+- and Na+-solvent cointercalation cells. Na+ intercalation into graphite is thermodynamically unfavorable, but Na+-diglyme cointercalation is very favorable. The diglyme–graphene van der Waals interaction reinforces the interlayer coupling strength and thereby improves the resistance of graphite to exfoliation. The transport of solvated Na ions is so fast that the diffusivity of Na+-diglyme complexes is markedly faster (by five orders of magnitude) than that of Li+-diglyme complexes. The very fast Na+-diglyme conductivity is attributed to facile sliding of flat diglyme molecules, which completely solvate Na ions in the interlayer space of graphite. The slow Li+-diglyme conductivity is ascribed to steric hindrance to codiffusion caused by bent diglyme molecules that incompletely solvate Li ions. The bent and flat diglyme molecules surrounding Li and Na ions, respectively, are highly associated with the strong Li+–graphene and weak Na+–graphene interactions, respectively.

Introduction

Graphite intercalation compounds (GICs) are layered materials with periodically stacked intercalant and graphene layers, and they are formed by inserting guest species, such as atoms, molecules, and ions, into the interlayer space of the host graphite. The control of the type and amount of guest species can lead to the formation of GICs with peculiar features such as superconducting behavior [1] and high transparency [2]. The GICs have a variety of applications as chemical reagents, electrochemical electrodes, highly conductive materials, catalysts, and so on [3]. Since the discovery of monolayer graphene in 2004 [4], GICs have been widely used as starting materials to produce large-area graphene sheets via exfoliation [5]. Many hundreds of GICs have been examined by utilizing various intercalant species such as alkali metals, metal oxides, metal chlorides, bromides, fluorides, oxyhalides, acidic oxides, and strong acids [3]. Particularly, one of the most extensively studied GICs is LixC6 (0<x≤1), namely Li-GICs, used as the standard anode material in lithium-ion batteries (LIBs) [6], [7], [8], [9], [10], [11].

LIBs are currently the most commonly used power sources for portable electronic devices, but are facing a potential challenge in price due to the low abundance of Li resources in the Earth's crust [12]. Sodium-ion batteries (SIBs) have attracted much attention as a promising alternative to LIBs. Unfortunately, however, Na+ intercalation into graphite is electrochemically difficult. The maximum sodiation capacity of graphite is <35 mAh g–1 for NaC64, which is much lower than that for lithiation (372 mAh g–1 for LiC6). The poor Na+ storage capability of graphite has been thought to be due to the small interlayer spacing of graphite, which is not sufficient to accommodate Na ions [13], [14], [15], [16], [17]. However, this prevailing view has been contested in experiments showing that K ions, which are larger than Na ions, can electrochemically intercalate into graphite [18], [19]. The low activity of graphite for sodiation can be ascribed to a weak Na+–graphene cation–π interaction rather than to any mismatch between the graphite interlayer spacing and ion size, considering that, among alkali metals, Na has the weakest binding to graphite [20]. Until recent years, graphite has been considered inappropriate for applications in SIBs unless it is modified by chemical methods such as oxidation, reduction [15], [21], and heteroatom doping [22].

Intriguingly, Jache et al. [23] and Kim et al. [24], [25] recently reported that pristine graphite can be successfully used as the anode material for SIBs by utilizing the Na+-solvent cointercalation mechanism. Na+-solvent cointercalation into graphite occurred with the use of ether-based electrolytes, such as mono-, di-, tri-, and tetraglyme molecules [23], [24], [25], [26], [27], [28], [29], [30], [31], [32]. These molecules are linear and have multiple O atoms that can simultaneously attract one Na+ ion. This structural feature allows a much stronger solvation of Na+ ions by these linear ether solvents than by other solvents (e.g., cyclic ether, linear carbonate, and cyclic carbonate) and leads to the frequent observations of cointercalation in linear ether electrolyte systems [29]. Especially when assisted by diglyme solvent molecules, Na ions are reversibly intercalated into graphite with reversible capacities of 100–150 mAh g–1 for several thousands of cycles at high currents, thereby forming ternary Na-diglyme-GICs [23], [24], [26]. Diglyme therefore shows the best electrochemical performance among the mono-, di-, tri-, and tetraglymes [24], [28], and its performance is also better than that of diglyme derivatives having non-polar side groups attached to linear diglyme [28]. These findings motivated the present investigation of the Na+-(linear diglyme) cointercalation into graphite. The most striking result is that Na-diglyme-GICs exhibit electrochemical performance much better than that of Li-diglyme-GICs [23], [28], [30]. The Na+-diglyme cointercalation cell exhibits a robust cycle stability over one thousand of cycles and retains reversible capacities in the range of 0.1–1.0 C rate (1 C=372 mA g–1), while the Li+-diglyme cointercalation cell shows a rapid capacity decay within a few tens of cycles and gives much reduced capacities at current densities exceeding 0.1 C rate [23]. These poor and excellent performance of graphite in Li+- and Na+-solvent cointercalation cells, respectively, are directly opposite to the success and failure of graphite in Li+ and Na+ intercalation cells, respectively. However, a fundamental understanding of the cointercalation into graphite and the origin of the contrasting performance of Li+- and Na+-solvent cointercalation cells remain elusive.

In this study, using first-principles molecular dynamics simulations, we show how the thermodynamic, mechanical, and kinetic properties of Na+-diglyme cointercalated graphite correlate with its electrochemical performance, which far surpasses that of its Li+-diglyme counterpart. While Na+ intercalated graphite is thermodynamically unfavorable, Na+-diglyme cointercalated graphite is considerably favorable. The diglyme–graphene van der Waals (vdW) interaction leads to the formation of stable ternary Na-diglyme-GICs and reinforces the interlayer coupling strength, improving the mechanical integrity of the graphite. The diglyme molecules that surround the Na ions are nearly flat and move rapidly in the interlayer space of the graphite, without intermolecular interference. The diffusivity of the Na+-diglyme complexes is strikingly faster (by five orders of magnitude) than that of the Li+-diglyme complexes. The slow Li+-diglyme conductivity is associated with steric hindrance arising from the bent diglyme molecules involved with partial Li+ desolvation. The contrasting shapes of the molecules surrounding the Li and Na ions are ascribed to a difference in ion–graphene interaction, i.e., the strong Li+–graphene and weak Na+–graphene interactions, respectively.

Section snippets

Computational details

Density functional theory (DFT) calculations were performed using the Vienna ab initio simulation package (VASP) [33]. We used the projector augmented wave (PAW) method [34] and the van der Waals density functional (vdW-DF) [35], in which the revised Perdew–Burke–Ernzerhof (revPBE) exchange functional [36] is employed. The electronic wave functions were expanded on a plane wave basis set with a kinetic energy of 400 eV. We treated 1s22s1 for Li, 2p63s1 for Na, 2s22p2 for C, 2s22p4 for O, and 1s

Results and discussion

We first examined the solvation of carrier ions by solvent molecules. Ion-solvent cointercalation into an electrode can bypass the slow desolvation step at the electrolyte/electrode interface [44]. The strong binding of ions to solvent molecules is thus necessary for successful cointercalation at the interface without desolvation. Our optimized structure of a free Na+-diglyme complex shows that the solvated Na ion is threefold coordinated with O atoms of diglyme (Fig. 1). We calculated the

Conclusions

In conclusion, our first-principles molecular dynamics study demonstrated that large-size solvated Na ions can successfully intercalate into and rapidly diffuse in graphite without causing mechanical disintegration of graphite. While graphite has no thermodynamic impetus for Na+ intercalation, it exhibits a strong thermodynamic driving force for Na+-diglyme cointercalation. Na+-diglyme cointercalated graphite exhibits a stable mechanical integrity thanks to diglyme–graphene vdW interaction. The

Acknowledgments

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (NRF-2015R1C1A1A01053146 and 2016R1A2B4013374). The authors thank Keon-Joon Lee and Jaeik Yoo for their computational support.

Dr. Sung Chul Jung received his Ph. D in condensed matter physics at Pohang University of Science and Technology (POSTECH), South Korea in 2010. He is currently a research professor at the Department of Energy and Materials Engineering, Dongguk University, South Korea. His research focuses on the fundamental understanding and advanced design of electrode materials in rechargeable batteries using first-principles calculations.

References (47)

  • Y.-X. Wang et al.

    Carbon

    (2013)
  • Z. Zhu et al.

    J. Power Sources

    (2015)
  • I. Hasa et al.

    J. Power Sources

    (2016)
  • Z. Guan et al.

    Electrochim. Acta

    (2016)
  • H.-J. Kim et al.

    Nano Energy

    (2015)
  • Y.-J. Kang et al.

    Nano Energy

    (2016)
  • G. Mills et al.

    Surf. Sci.

    (1995)
  • K.C. Rahnejat et al.

    Nat. Commun.

    (2011)
  • I. Khrapach et al.

    Adv. Mater.

    (2012)
  • M.S. Dresselhaus et al.

    Adv. Phys.

    (2002)
  • K.S. Novoselov et al.

    Science

    (2004)
  • V. Nicolosi et al.

    Science

    (2013)
  • J.-M. Tarascon et al.

    Nature

    (2001)
  • M. Armand et al.

    Nature

    (2008)
  • P.G. Bruce et al.

    Angew. Chem. Int. Ed.

    (2008)
  • A. Ambrosi et al.

    Chem. Rev.

    (2014)
  • M. Ko et al.

    Nat. Energy

    (2016)
  • J. Sun et al.

    Nano Lett.

    (2014)
  • N. Yabuuchi et al.

    Chem. Rev.

    (2014)
  • S. Komaba et al.

    Adv. Funct. Mater.

    (2011)
  • Y. Cao et al.

    Nano Lett.

    (2012)
  • Y. Wen et al.

    Nat. Commun.

    (2014)
  • J. Sun et al.

    Nat. Nanotechnol.

    (2015)
  • Cited by (71)

    • Constructing ultrafine Cu nanoparticles encapsulated by N-doped carbon nanosheets with fast kinetics for high-performance lithium/sodium storage

      2022, Chemical Engineering Journal
      Citation Excerpt :

      The D and G peaks are slightly shifted to low wave numbers during the initial discharging process (Fig. 6g). In the meantime, the ID/IG value of the Cu/NC increased after discharging process, suggesting more disordered structure due to the ether solvent and Na-ion co-insert into the carbon structure [68-70], and then decreased in the subsequent charging process, indicating the structural stability. The (0 0 2) peak shifts to a lower angle during the discharging process and then moves back after full charging, which means that the carbon lattice expansion/contraction caused by intercalation/deintercalation of ether solvent and Na-ion (Fig. 6h).

    • Potassium-ion batteries using KFSI/DME electrolytes: Implications of cation solvation on the K<sup>+</sup>-graphite (co-)intercalation mechanism

      2022, Energy Storage Materials
      Citation Excerpt :

      Both mechanisms, co-intercalation and intercalation, have already been thoroughly discussed in many reviews of LIBs and SIBs [26–32] and are commonly related to the presence of a stable solid electrolyte interphase (SEI) acting as a solvent-blocking layer [33–36]. The co-intercalation of Na+ with diglyme in graphite occurs reversibly with excellent rate and cycling performance [37], either due to negligible SEI formed, both for graphite [38] and hard carbon [39], and/or a specific stabilization of the Na+-diglyme complex (within graphite) [40]. Some recent reports on the stability of the SEI formed by DME-based electrolytes in KIBs show improved coulombic efficiency and cycling retention of graphite [41,42], but its influence on the charge storage mechanism has not been clarified yet.

    View all citing articles on Scopus

    Dr. Sung Chul Jung received his Ph. D in condensed matter physics at Pohang University of Science and Technology (POSTECH), South Korea in 2010. He is currently a research professor at the Department of Energy and Materials Engineering, Dongguk University, South Korea. His research focuses on the fundamental understanding and advanced design of electrode materials in rechargeable batteries using first-principles calculations.

    Dr. Yong-Ju Kang received his Ph. D in condensed matter physics at Korea Advanced Institute of Science and Technology (KAIST), South Korea in 2009. He is currently a research professor at the Department of Energy and Materials Engineering, Dongguk University, South Korea. His research focuses on the fundamental understanding and advanced design of electrode materials in rechargeable batteries using first-principles calculations.

    Dr. Young-Kyu Han received his Ph. D in physical chemistry at Korea Advanced Institute of Science and Technology (KAIST), South Korea in 1999. He is currently a professor at the Department of Energy and Materials Engineering, Dongguk University, South Korea. His research focuses on the fundamental understanding and advanced design of electrode and electrolyte materials in rechargeable batteries and supercapacitors using first-principles calculations.

    View full text