Kinetically Determined Phase Transition from Stage II (LiC12) to Stage

Sep 10, 2018 - An in-depth understanding of this phenomenon will deepen the knowledge of electrode material design and optimize rechargeable Li batter...
0 downloads 0 Views 4MB Size
Download PDF
Letter Cite This: J. Phys. Chem. Lett. 2018, 9, 5567−5573

pubs.acs.org/JPCL

Kinetically Determined Phase Transition from Stage II (LiC12) to Stage I (LiC6) in a Graphite Anode for Li-Ion Batteries Qiang Liu,† Shuai Li,‡ Senhao Wang,‡ Xianggong Zhang,† Sisi Zhou,† Ying Bai,§ Jieyun Zheng,∥ and Xia Lu*,‡,⊥ †

Wuhan Institute of Marine Electric Propulsion, China Shipbuilding Industry Corporation, Wuhan 430064, China Beijing Advanced Innovation Center for Soft Matter Science and Engineering, State Key Laboratory of Organic−Inorganic Composites, Beijing University of Chemical Technology, Beijing 100029, China § School of Physics & Electronics, Henan University, Kaifeng 475004, China ∥ Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China ⊥ School of Materials, Sun Yat-sen University, Guangzhou 510275, China

J. Phys. Chem. Lett. 2018.9:5567-5573. Downloaded from pubs.acs.org by KAOHSIUNG MEDICAL UNIV on 10/01/18. For personal use only.

‡

S Supporting Information *

ABSTRACT: The electrochemical insertion of Li into graphite initiates a series of thermodynamic and kinetic processes. An in-depth understanding of this phenomenon will deepen the knowledge of electrode material design and optimize rechargeable Li batteries. In this context, the phase transition from dense stage II (LiC12) to stage I (LiC6) was comprehensively elucidated in a graphite anode via both experimental characterizations and first-principles calculations. The results indicate that, although the transition from stage II to stage I is thermodynamically allowed, the process is kinetically prohibited because Li ions tend to cluster into stage compounds rather than form a solid solution. Additionally, the phase transitions involve at least three intermediate structures (1T, 2H, and 3R) before reaching the LiC6 (stage I) phase. These findings provide new insights into the electrochemical behavior of graphite and the electrode process kinetics for rechargeable Li batteries.

T

ments of inserted Li are generated among the graphite layers with unclear phase transition kinetics, and (2) asymmetric charging/discharge behavior, which could be attributed to the different Li diffusion kinetics at different Li insertion stages.10−13 Tremendous work has already been performed to determine the related Li transport and storage mechanism in a graphite anode.1,10 In early studies, Safran and Hamann discussed the application of strain considerations as one of the driving forces for staging in GICs according to the Daumas− Hérold domain model.4,14−16 They calculated the straininduced forces between two intercalated atoms in a graphite host material, showing that the intercalated atoms between the same two graphite layers are attracted to form two-dimensional islands that lower its strain energy with the free energy minimized by the formation of a pure stage configuration from a mixed or randomly staged crystal.15 Then, a simple scenario for the intercalation process was suggested by Dresselhaus et al. in which there could be a nucleation of islands near the sample edge, followed by migration toward the center of the sample.1 However, it is still not clear that the above mechanisms can by themselves fully account for the intercalation and staging phenomenon in a graphite anode.

he intercalation phenomena of metal ions (e.g., Li, Na, and K) or small molecules (e.g., HNO3, FeCl3, SbF5, and organic radicals) into graphite lead to the formation of graphite intercalation compounds (GICs).1−3 Upon intercalation, this always demonstrates interesting phenomena, such as ion/ vacancy ordering, selective ion insertion, and spatial ion ordering in the graphite host. In terms of Li intercalation into graphite, it gives rise to numerous intermediate phases whose composition (LixC6) could vary from x = 0 to 1. This richness should be ascribed to the uncontrollable interlayer glide and unpredicted structural distortions induced by van der Waals forces.4 Namely, the interlayer interaction among the graphene sheets is too weak to establish a rigid atomic stacking mode along the z direction ([001]). This results in at least four different stacking structures as the AAA-hexagonal phase,5 the ABAB-bernal phase,6−8 the ABCABC-rhombohedral phase,8,9 and the turbostatic phase.9 Among these, the ABAB-stacking graphite has dominated as electrode materials in Li-ion batteries. However, this does not mean that the abovementioned stacking structures could not all coexist at the beginning or be formed during lithiation/delithiation. Li intercalation into graphite with a composition of LiC6 can deliver a specific capacity of 372 mAh/g when cycling between 0.01 and 2.0 V vs Li+/Li. During cycling, two main electrochemical features have been found, such as (1) Li staging-related storage, where a series of ordering arrange© 2018 American Chemical Society

Received: September 6, 2018 Accepted: September 10, 2018 Published: September 10, 2018 5567

DOI: 10.1021/acs.jpclett.8b02750 J. Phys. Chem. Lett. 2018, 9, 5567−5573

Letter

The Journal of Physical Chemistry Letters

Figure 1. Structure and morphology of graphite materials. (a,b) XRD patterns with magnified insets (42−48°); (c) respective lithiation profiles of the graphite nanosheet and nanoparticle anodes at the second cycle; (d−f) SEM, TEM, and STEM images of the graphite nanosheets; (g) SEM images of the graphite nanoparticles.

contained thousands of graphene layers (layers ≈ 103) in the form of graphite nanoparticles, as shown in Figures 1b,g and S2 (SEM and TEM images in the SI). The structural difference was clearly observed in the XRD results in which more diffraction peaks were present for the graphite nanosheet, as shown in Figure 1a,b, which suggests structural instability and an interlayer relative glide. This kind of structural change can be ascribed to structural relaxation due to the layer reduction. In detail, the XRD pattern in Figure 1a indicates that there are different graphene stacking modes in the graphite nanosheets with respect to the graphite nanoparticles, as shown in Figure 1b. The magnified inset XRD patterns (42−48°) display that, in addition to the common ABAB stacking (denoted as the 2H phase) of graphene layers in the graphite nanoparticles, there is at least one more stacking mode (the ABCABC arrangement (denoted as the 3R phase)) in the thin graphite nanosheets, as shown in Figure 1a,b. From a structural point of view, graphite also has the AAA stacking mode as the 1T phase in which the 1T, 2H, and 3R structures of graphite are illustrated in Figure S3. All three phases could be involved in the phase transition of the graphite anode upon lithiation, as discussed later. Figure 1c shows the second electrochemical lithiation of the graphite nanosheet (black) and nanoparticle (red) at different rates in Swagelok-type cells. The similarity between them indicates that the intrinsic lithiation process is independent of the parameters, such as the morphology, cycling rate, and stacking modes, in graphite. The Li intercalation becomes more vivid when discharging to below 0.2 V with three pronounced plateaus, one located at 0.20 V, one at 0.11 V, and one at 0.09 V. Applying a high current would slightly distort the plateaus, as the red curve shows in Figure 1c. This variation is consistent with the reported results at even higher rates.10 Due to the specific surface area changes, much more of the solid electrolyte interphase (SEI) film formed on the nanosheet, as shown in Figure S4, when compared with that on the nanoparticles (low specific area), as shown in Figure S5a, which is consistent with the reported results.18 During the second lithiation, the nanosheet anode still delivered a capacity of 496 mAh/g at 10 mA/g, as shown in Figure 1c, which is

Recently, a shrinking annuli mechanism was proposed by Heß et al. to interpret the staging transformation in a graphite anode upon lithiation.10 They found that upon lithiation the phase boundaries of the previous stage transitions still progress farther into the center of the particle, whereas a new phase boundary develops at the edge surface of the graphite particles in a similar manner as the annuli of trees. Upon delithiation, the fast liquid-like stage transitions compensate for the slow dense stage transitions, which can explain the long-known asymmetry in the rate capability for the charge and discharge of graphite.10 It is easy to accept that upon lithiation, when the Li diffusion limitation is reached, the increasing overpotential can push the lithiation of graphite into the next step in one particle, probably at the edge region. Therefore, the competition between the Li diffusion kinetics and the applied rate determines the Li staging and solid solution storage behaviors. The stage transformation can be modulated by the Li concentration (adjacent Li−Li interactions), the Li diffusion kinetics (Li-vacancy interactions), and the possible glide of graphene layers (structural strains and rearrangements). Therefore, elucidating the interplay between the Li and graphite necessitates an understanding of the staging storage in the graphite material itself and has a profound influence on the electrochemical process and battery community. In this study, the Li-ion diffusion kinetics and the structural evolution of a graphite anode were comprehensively investigated to determine the stage transformation between stage II (LiC12) and stage I (LiC6) via the electrochemical characterizations and density functional theory (DFT) calculations. The results suggest that the stage transition is a kinetically controlled ionic diffusion process that causes a change in the stacking of graphene sheets from the ABAB to the AAA mode at the end of the lithiation. Two kinds of graphite materials were employed to investigate the electrochemical performance and Li storage and transport mechanism for comparison, as shown in Figure 1. One type of graphite nanosheet was composed of ca. 30 layers of graphene (layers ≈ 101), as shown in Figures 1a,d−f and S1 (Raman spectroscopy17 in the SI). The other sheet 5568

DOI: 10.1021/acs.jpclett.8b02750 J. Phys. Chem. Lett. 2018, 9, 5567−5573

Letter

The Journal of Physical Chemistry Letters

unclear for stages I, II, and III, which have been previously reported to be established with the help of structural distortion, Li 2D diffusion, the Li concentration gradient force, and so on.1,10 To elucidate how these factors interact with the stage transitions of GICs, a theoretical calculation was performed within the DFT framework.11 Meanwhile, the Li storage and transport properties were also investigated to elucidate the structural fundamentals of graphite with dense stages (stage I, II, and III). Inside of a graphite 2H supercell (C48), as shown in Figure 2b, one Li can be accommodated in a position inside of the interlayer, where if the bottom is located at the center of a carbon ring (C6) the upside will be the exact same C site. In this case, there will be a little layer relaxation within the 2H structure. Then, one more Li is placed inside of this supercell (LiC24). After structural relaxation, Figure 2c shows the energy profile that varies with the second Li site, indicating that it will attempt to stay as close as possible to a lower total energy. The calculation also implies that two Li in the same layer will result in the largest change in layer distance and structural distortion within the 2H phase. A typical feature shown in Figure 2c is that the adjacent layer cannot possibly accommodate the second Li. Additionally, the third and fourth adjacent layers seem to be minimally influenced by the second Li insertion. Therefore, at the beginning of Li intercalation, a dilute stage II phase or even higher stages (or disordered structure) is preferred as the result of “selective” Li-ion intercalation in which it minimizes the total energy of the lithiated graphite. Additionally, selective Li insertion could become more recognizable with an increase in Li concentrations and could finally lead to a clear Li staging storage feature, as the electrochemical plateaus shown in Figure 2a. In addition to the crystal change, the electronic structures of GICs also vary upon Li intercalation, which is reflected by the color changes shown in Figure S5c. Figure 3 displays the electronic structures of 2H graphite before and after complete lithiation. The band structure, electronic density of states (DOS), and crystal structure of pristine graphite, respectively, are shown in Figure 3a−c. For the fully lithiated state of LiC6, the calculation was performed in a 3 × 3 × 1 graphite supercell, as shown in Figure 3d−g. Pristine 2H graphite possesses a semimetal property, which implies its superior electronic conductivity. The black graphite electrode (before stage III) first turned gray because of Li intercalation when the stage II compound began to form, and then, a brilliant golden stage I compound formed at the end of the Li-ion intercalation process, as shown in Figure S5c. The brilliant golden graphite electrode turned back to grayish once it was taken out of the liquid electrolyte. This color restoration reflects the strong chemical/electrochemical activity of the LiC6 phase. As shown in Figure 3d,e, the Li ions located at the center of carbon rings form a six-fold coordination with surrounding carbons, where each Li ion is coordinated with six carbons and the adjacent Li ions tend to stay as far as possible from each other to stabilize the structure. The injected electrons seem to be redistributed among the carbon 2p orbitals, as shown in Figure 3g. The Fermi level is lifted, which makes the lithiated graphite electrode metallic, as shown in Figure 3f. The color changes during lithiation reveal improvement on the electronic conductivity of the graphite electrode, which is probably less responsible for the asymmetric charge/discharge behavior and sluggish phase transition kinetics.

more than the theoretical value of 372 mAh/g that shows the continuous growth of the SEI film. Then, for the big graphite nanoparticle anode, only 345 mAh/g could be obtained after the formation cycle at C/5 (∼70 mA/g), as shown in Figures 1c and S5a. Although both graphite samples achieved excellent longterm stability, as shown in Figure S5, they exhibit low initial Coulombic efficiency, especially the graphite nanosheets, as shown in Figure S3. This difference in Coulombic efficiency again is the result of the lower surface area of graphite nanoparticles, alleviating the significant side reactions. Such long-term stability implies a reversible structural transition during lithiation/delithiation, despite a minor decrease in the capacity (e.g., the green curve with respect to the red one in Figure S5a). The structure evolution of graphite could also be referred from the cycling voltammetry tests shown in Figure S6. In Figure S6, three clear redox couples were assigned as a, d, and e. Additionally, two features corresponding to subtle processes were also assigned as b and c. Each redox couple is intimately associated with the structural change in the graphite anode upon Li insertion, i.e., presently called the Li staging storage phenomenon.1,19,20 The stage transition also results in a color change; for example, it changes from black (LiC18) to silver black (LiC12) and to golden (LiC6) at the end of discharge, as shown in the inset of Figure S5c. Such ready variation in the composition of LixC6 compounds further complicates the structural transition pathways and Li-ion storage properties. Let us first consider a simulation of the Li insertion into the graphite anode. Skipping the formation process of the SEI layer and the dilute stages,10 there are three prominent structural changes after 0.2 V that correspond to the structural transitions of the stage from LiC24 to LiC18 (formation of stage III), from LiC18 to LiC12 (formation of stage II), and from LiC12 to LiC6 (formation of stage I), if only thermodynamics is considered, as shown in Figure 2a. The phase transition kinetics are still

Figure 2. Staging phenomenon in a graphite anode upon lithiation. (a) Staging-related Li storage in a graphite anode during electrochemical lithiation. (b) Crystal structure and (c) corresponding energy profile of two Li atoms distributing inside of a 2 × 2 × 6 graphite supercell. 5569

DOI: 10.1021/acs.jpclett.8b02750 J. Phys. Chem. Lett. 2018, 9, 5567−5573

Letter

The Journal of Physical Chemistry Letters

Figure 3. Electronic and crystal structures of graphite without/with lithiation. (a) Band structure, (b) DOS, and (c) crystal structure of pristine 2H graphite. Crystal structure of a 3 × 3 × 2 graphite supercell viewed (d) perpendicular and (e) parallel to the Z axis and the (f) band structure and (g) DOS of graphite that is fully lithiated.

Figure 4. Single Li-ion diffusion inside of the deficient LiC12 phase (Li1−xC12). (a) Structure of a 3 × 3 × 2 graphite supercell, (b) diffusion trajectories, and (c) activation energies of single Li hopping inside of the deficient LiC12 phase.

As shown in Figure 2, the selectivity for Li insertion is facilitated to form a dilute stage phase during the initial lithiation of the graphite electrode. In this study, a 3 × 3 × 1 graphite supercell was built to evaluate the Li diffusion kinetics of stage II during early lithiation. As shown in Figure 4a,b, a single Li atom was first relaxed into the equilibrated site. Then, two different Li ion hopping trajectories were selected, labeled as pathways 1 and 2, where the Li ion diffuses directly along the center of the carbon rings, and pathway 12, where the Li ion migrates from one center of the carbon ring along the C− C bond to another. From the activation energies of each path, as shown in Figure 4c, the Li ion along pathways 1 and 2 with an activation energy of ca. 0.05 eV is much more preferred than pathway 12 because the activation energy of pathway 12 is approximately 7 times higher (ca. 0.37 eV). Thus, Li diffusion and even storage in the dilute stages of GICs can be readily accommodated via facile adaptation in both electronic and

crystal structures. Under this circumstance, the correlation between the Li ion and Li vacancy seems to be significantly weakened and has nearly no influence on the initial Li insertion. The low activation energy probably corresponds to the fast Li-ion diffusion kinetics in the dilute stages of the graphite anode. With an increase in Li concentration, the Li diffusion kinetics changes. The results of the Li-ion diffusion in stage IIrelated phases are shown in Figure 5. In Figure 5a,b,g, Li vacancy is purposely created inside of the LiC12 (stage II) structure as Li2C36 (LiC12-deficient). First, the stage II phase retains its 2H symmetry even with the Li vacancy introduced. Li-ion diffusion in this vacancy-rich configuration encounters an energy barrier of 0.30 eV, which is much higher than that of Li-ion diffusion in dilute stages, as discussed in Figure 4c. Although such an activation energy could still facilitate fast Liion diffusion, the Li and vacancy interplay becomes 5570

DOI: 10.1021/acs.jpclett.8b02750 J. Phys. Chem. Lett. 2018, 9, 5567−5573

Letter

The Journal of Physical Chemistry Letters

Figure 5. Illustrations of the Li diffusion inside of the LiC12-related phases in a 3 × 3 × 1 graphite supercell. (a,b) Structures of the LiC12-deficient phase, (c,d) nominal LiC12 structure with Li diffusion trajectory, (e,f) LiC12 excess structures with Li hopping trajectories, and (g) activation energies for Li diffusion in different structures, as shown above.

Figure 6. Li diffusion inside of the stage I phase. (a) Top-view structure of Li hopping trajectories (along the [001] direction) and (b) activation energies during Li-ion diffusions in the LiC6 structure.

significantly important when the Li content increases in the following Li ion migration. On the basis of the abovementioned LiC12-deficient structure, one more Li is added into the adjacent vacant interlayer (nominal LiC12 phase with a Li disordered distribution, which has only a very small total energy difference from stage II). The Li is precisely located at the top of the vacant Li site, as shown in Figure 5c. The calculation results during Li migration indicate that (1) the original 2H structure is transformed into the new 1T phase, where all of the graphene sheets align, as shown in Figure 5d, and that (2) the Li-ion diffusion kinetics become extraordinarily poor based on the calculated activation energy of 0.51 eV in the vacant layer, as shown in Figure 5g, which is ascribed possibly to the structural/layer distortion, much like the Li-ion diffusion kinetics in LiFePO4.21 Therefore, the formation of stage II could be much more feasible compared with the disordered Li insertion into the graphite at stage I (LiC12). There is a kinetically determined phase transition between stage II and stage I. Upon lithiation, the increase in structural disorder produces a sluggish influence on Li storage and transport in graphite. The least energy is likely consumed for Li

ordering insertion, where less structural distortion and reconstruction occur. Moreover, one more Li is added into the empty interlayer of the stage II phase to constitute a Li1+xC12 phase (LiC12 excess), as shown in Figure 5e−g. This process indicates the trend to form the dense LiC6 structure (stage I). During the initial insertion of Li, pristine 2H graphite displays the strong ability to maintain its original structure. The Li-ion diffusion is fast with an activation energy of approximately 0.11 eV in the LiC12 excess phase, which is two times higher than its diffusion in the dilute stages, as shown in Figures 5e,f and 6. This implies that there is an influence of the structural distortion of dense stages upon Li diffusion. However, after experiencing the sluggish lithiation process in forming stage II, the short fast Li ion insertion period occurs on the way to stage I. In the stage I phase, the graphite anode finishes its lithiation process to accomplish a 372 mAh/g capacity, where each carbon ring (C6) is coordinated with one Li. The Li storage and diffusion feature in stage I are schematically shown in Figure 6. The calculation results show that the Li ion is located at the center of the carbon ring and that the structural transition from the original 2H to the final 1T also succeeded, 5571

DOI: 10.1021/acs.jpclett.8b02750 J. Phys. Chem. Lett. 2018, 9, 5567−5573

Letter

The Journal of Physical Chemistry Letters where the Li ion could align directly along the [001] direction. Of note, if the interlayer distance between the graphene sheets is fixed at 3.3 Å, an intermediate phase for the 3R graphite is observed, which implies that the original 2H graphite should experience an intermediate 3R phase prior to the final 1T lithiated graphite. This confirms the direction of the relative glide of the graphene layer in graphite, which is along the diagonal of the C6 rings. Then, the aforementioned Li-ion diffusion pathways were considered in the framework of the vacancy-assisted diffusion mechanism. The difference is that the Li vacancy is coordinated with six Li ions, as shown in Figure 6a (Li5C36: LiC6-deficient). First, the activation energy along pathways 1 and 2 (i.e., cut through the C−C bonds) is approximately 0.45 eV, compared with that in pathway 12 (C− C bridge) of 0.47 eV, as shown in Figure 6b, which means that the Li-ion diffusion kinetics becomes poor enough with respect to its diffusion in the dilute stages. Such a high barrier also implies the diffusion resistance from adjacent Li ions, which is in general described as the imbalance of the lattice field in lithiated graphite by Li vacancy. In the end, the Li-ion diffusion makes it difficult to finish the LiC6 phase, although a fast process occurs at the beginning of forming stage I, as previously mentioned. Throughout the entire phase transition process, based on the coordination environment change of the intercalated Li ions, the periodic Li-ion diffusion kinetics could be easily predicted for the lithiation process of graphite due to the amount of change and Coulombic repulsion of adjacent Li. In summary, the graphite nanosheet and nanoparticle were employed to investigate the structural transitions among various stage compounds upon lithiation in a Li ion battery. A decrease in graphite layers destabilizes the original 2H graphite phase and leads to a new stacking mode, e.g., a 3R phase. Despite the difference in the pristine structures, the Li storage behavior is independent of the morphology because the same plateaus and transitions appeared on the voltage profile of both electrodes. Then, first-principles calculations demonstrated that the Li could stay in every other layer in graphite to form a dilute stage II phase upon initial lithiation. The Li-ion diffusion at low Li concentration was predicted to be considerably fast on the basis of the derived activation energy of 0.05 eV. However, with an increase in Li concentration, the formation of stage II resulted in Li-ion diffusion kinetics from the initial fast rate to the poor one with an activation barrier of 0.51 eV. As a result, upon lithiation, stage II compounds are kinetically stabilized first because the subsequent transition to stage I compounds is too slow to proceed. Formation of stage I compounds is delayed until the concentration of stage II compounds meets the threshold. At the end of lithiation, stage I compounds transform from the original 2H structure via an intermediate 3R phase and eventually into the 1T phase. The interaction between Li storage properties and such “periodical” structural transitions in graphite involves a series of thermodynamic and kinetic factors that play important roles in understanding the battery chemistry. In-depth knowledge of the transition of intercalated graphite compounds is important for providing guidance and comprehension of electrode material design and improvements for rechargeable batteries.

scan mode and using a Cu Kα X-ray source with a step width of 0.0167°/s. Raman spectroscopy was performed on a laser Raman spectrometer (Renishaw RM-1000) with a 514 nm Arion laser. A scanning electron microscope (SEM) (XL 30 SFEG, FEI Co.) was used to determine the sample morphology. Transmission electron microscopy (TEM) studies were performed on a JEM-2011 electron microscope equipped with an EDS analyzer. Scanning transmission electron microscopy (STEM) was performed on a JEOL 2100F (JEOL, Tokyo, Japan) transmission electron microscope operated at 200 keV. The microscope is equipped with a CEOS (CEOS, Heidelberg, Germany) probe aberration corrector. The attainable spatial resolution of the microscope is 90 pm at an incident semiangle of 20 mrad. Electrochemical Tests. The cycling tests were performed in a Swagelok-type cell between 2.5 and 0.01 V. A polypropylene film (Celgard 2300) was used as the separator. The working electrode was fabricated by casting a mixture of the active material and poly(vinylidene difluoride) (PVDF) at a weight ratio of (graphite/PVDF) 0.90:0.10 onto an aluminum foil. A metallic lithium foil was used as the counter electrode. A 1 M LiPF6 in ethylene carbonate (EC)/dimethyl carbonate (DMC) (1:1 by volume) solvent was employed as the electrolyte during the electrochemical tests. First-Principles Calculations. DFT calculations were performed within the Vienna ab initio simulation package (VASP)22,23 in the framework of the projector augmented wave (PAW) method24,25 using the local density approximation (LDA) functional26,27 with spin polarizations. The configurations of Li 2s and C 2s2p were treated as valence electrons. The cutoff energy for the plane-wave was 600 eV for all of the simulation results. A unit cell of 3 × 3 × 1 and 2 × 2 × 6 supercells was adopted for geometry optimization, electronic structure, and Li-ion diffusion kinetic simulations when necessary. Irreducible Brillouin zone integration was performed using Monkhorst−Pack K-points sampling28 with a Gaussian smearing of 0.1 eV near the Fermi level (EF). Atomic sites and lattice constants were relaxed until the residual forces were less than 0.01 eV/Å in each species. The nudged elastic band calculation with the climbing image method (CINEB) was used to estimate the Li-ion diffusion kinetics inside of the graphite anode. This method duplicates a series of images (here 7) between the starting point and the end point of a diffusing ion to simulate the intermediate states with the positions of the starting point and the end point fixed during the kinetic calculations. Using the extended cell, the mirror effect of the vacant or interstitial atom in a neighboring cell can be effectively avoided.

EXPERIMENTAL AND CALCULATION METHODS Characterizations. X-ray powder diffraction (XRD) patterns were acquired in the range of 10−80° using a Philips X’pert diffractometer with Bragg−Brentano geometry in a continuous

Corresponding Author

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.8b02750. Raman spectroscopy, XRD, SEM, and electrochemical performance of the graphite anode (PDF)

■

■

AUTHOR INFORMATION

*E-mail: [email protected]. ORCID

Ying Bai: 0000-0001-7835-7067 5572

DOI: 10.1021/acs.jpclett.8b02750 J. Phys. Chem. Lett. 2018, 9, 5567−5573

Letter

The Journal of Physical Chemistry Letters

(18) Bai, L. Z.; Zhao, D. L.; Zhang, T. M.; Xie, W. G.; Zhang, J. M.; Shen, Z. M. A comparative study of electrochemical performance of graphene sheets, expanded graphite and natural graphite as anode materials for lithium-ion batteries. Electrochim. Acta 2013, 107, 555− 561. (19) Derosa, P. A.; Balbuena, P. B. A lattice-gas model study of lithium intercalation in graphite. J. Electrochem. Soc. 1999, 146 (10), 3630−3638. (20) Funabiki, A.; Inaba, M.; Abe, T.; Ogumi, Z. Stage transformation of lithium-graphite intercalation compounds caused by electrochemical lithium intercalation. J. Electrochem. Soc. 1999, 146 (7), 2443−2448. (21) Sun, Y.; Lu, X.; Xiao, R.; Li, H.; Huang, X. Kinetically Controlled Lithium-Staging in Delithiated LiFePO4 Driven by the Fe Center Mediated Interlayer Li−Li Interactions. Chem. Mater. 2012, 24 (24), 4693−4703. (22) Kresse, G.; Hafner, J. Abinitio Molecular-Dynamics for LiquidMetals. Phys. Rev. B: Condens. Matter Mater. Phys. 1993, 47 (1), 558− 561. (23) Kresse, G.; Furthmuller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B: Condens. Matter Mater. Phys. 1996, 54 (16), 11169−11186. (24) Kresse, G.; Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59 (3), 1758−1775. (25) Blochl, P. E. Projector Augmented-Wave Method. Phys. Rev. B: Condens. Matter Mater. Phys. 1994, 50 (24), 17953−17979. (26) Perdew, J. P.; Wang, Y. Accurate and simple analytic representation of the electron-gas correlation energy. Phys. Rev. B: Condens. Matter Mater. Phys. 1992, 45 (23), 13244. (27) Perdew, J. P.; Zunger, A. Self-interaction correction to densityfunctional approximations for many-electron systems. Phys. Rev. B: Condens. Matter Mater. Phys. 1981, 23 (10), 5048−5079. (28) Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Phys. Rev. B 1976, 13 (12), 5188−5192.

Xia Lu: 0000-0003-3504-9069 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors acknowledge funding support from the National Natural Science Foundation of China (11704019) and the State Key Laboratory of Organic−Inorganic Composites (oic201701011). The work was carried out at the National Supercomputer Center in Tianjin, and the calculations were performed on TianHe-1(A).

■

REFERENCES

(1) Dresselhaus, M. S.; Dresselhaus, G. Intercalation compounds of graphite. Adv. Phys. 2002, 51 (1), 1−186. (2) Ebert, L. B. Intercalation Compounds of Graphite. Annu. Rev. Mater. Sci. 1976, 6, 181−211. (3) Skipper, N. T.; Walters, J. K.; Lobban, C.; McKewn, J.; Mukerji, R.; Martin, G. J.; de Podesta, M.; Hannon, A. C. Neutron diffraction studies of graphite-potassium-methylamine: Staging transitions and structure of new graphite intercalation compounds. J. Phys. Chem. B 2000, 104 (47), 10969−10972. (4) Safran, S. A.; Hamann, D. R. Long-Range Elastic Interactions and Staging in Graphite-Intercalation Compounds. Phys. Rev. Lett. 1979, 42 (21), 1410−1413. (5) Charlier, J. C.; Michenaud, J. P.; Gonze, X.; Vigneron, J. P. Tight-Binding Model for the Electronic-Properties of Simple Hexagonal Graphite. Phys. Rev. B: Condens. Matter Mater. Phys. 1991, 44 (24), 13237−13249. (6) Charlier, J. C.; Gonze, X.; Michenaud, J. P. 1st-Principles Study of the Electronic-Properties of Graphite. Phys. Rev. B: Condens. Matter Mater. Phys. 1991, 43 (6), 4579−4589. (7) Charlier, J. C.; Michenaud, J. P.; Gonze, X. 1st-Principles Study of the Electronic-Properties of Simple Hexagonal Graphite. Phys. Rev. B: Condens. Matter Mater. Phys. 1992, 46 (8), 4531−4539. (8) Charlier, J. C.; Michenaud, J. P.; Lambin, P. Tight-Binding Density of Electronic States of Pregraphitic Carbon. Phys. Rev. B: Condens. Matter Mater. Phys. 1992, 46 (8), 4540−4543. (9) Charlier, J. C.; Gonze, X.; Michenaud, J. P. 1st-Principles Study of the Stacking Effect on the Electronic-Properties of Graphite(S). Carbon 1994, 32 (2), 289−299. (10) Hess, M.; Novak, P. Shrinking annuli mechanism and stagedependent rate capability of thin-layer graphite electrodes for lithiumion batteries. Electrochim. Acta 2013, 106, 149−158. (11) Persson, K.; Hinuma, Y.; Meng, Y. S.; Van der Ven, A.; Ceder, G. Thermodynamic and kinetic properties of the Li-graphite system from first-principles calculations. Phys. Rev. B: Condens. Matter Mater. Phys. 2010, DOI: 10.1103/PhysRevB.82.125416. (12) Levi, M. D.; Wang, C.; Markevich, E.; Aurbach, D.; Chvoj, Z. Noteworthy electroanalytical features of the stage 4 to stage 3 phase transition in lithiated graphite. J. Solid State Electrochem. 2003, 8 (1), 40−43. (13) Buqa, H.; Goers, D.; Holzapfel, M.; Spahr, M. E.; Novak, P. High rate capability of graphite negative electrodes for lithium-ion batteries. J. Electrochem. Soc. 2005, 152 (2), A474−A481. (14) Safran, S. A.; Hamann, D. R. Long-Range Elastic Interactions and Staging in Graphite Intercalation Compounds. Phys. Rev. Lett. 1979, 42 (21), 1410−1413. (15) Safran, S. A.; Hamann, D. R. Electrostatic interactions and staging in graphite intercalation compounds. Phys. Rev. B: Condens. Matter Mater. Phys. 1980, 22 (2), 606−612. (16) Miyazaki, H. Stochastic-Model of Staging in GraphiteIntercalation Compounds. J. Mater. Res. 1988, 3 (3), 479−490. (17) Dresselhaus, M. S.; Jorio, A.; Saito, R. Characterizing Graphene, Graphite, and Carbon Nanotubes by Raman Spectroscopy. Annu. Rev. Condens. Matter Phys. 2010, 1, 89−108. 5573

DOI: 10.1021/acs.jpclett.8b02750 J. Phys. Chem. Lett. 2018, 9, 5567−5573