Silicon-Based Anodes with Long Cycle Life for Lithium-Ion Batteries

Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces 2019, 11, 2950−2960

Silicon-Based Anodes with Long Cycle Life for Lithium-Ion Batteries Achieved by Significant Suppression of Their Volume Expansion in Ionic-Liquid Electrolyte Yasuhiro Domi,†,§,∥ Hiroyuki Usui,†,§,∥ Kazuki Yamaguchi,†,§,∥ Shuhei Yodoya,‡,§,∥ and Hiroki Sakaguchi*,†,§,∥

Downloaded via UNIV OF LOUISIANA AT LAFAYETTE on January 24, 2019 at 14:42:19 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Department of Chemistry and Biotechnology, Graduate School of Engineering, ‡Department of Engineering, Graduate School of Sustainability Science, and §Center for Research on Green Sustainable Chemistry, Tottori University, 4-101 minami, Koyama-cho, Tottori 680-8552, Japan ∥ Global Research Center for Environment and Energy based on Nanomaterials Science (GREEN), National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan S Supporting Information *

ABSTRACT: Elemental Si has a high theoretical capacity and has attracted attention as an anode material for high energy density lithiumion batteries. Rapid capacity fading is the main problem with Si-based electrodes; this is mainly because of a massive volume change in Si during lithiation−delithiation. Here, we report that combining an ionic-liquid electrolyte with a charge capacity limit of 1000 mA h g−1 significantly suppresses Si volume expansion, improving the cycle life. Phosphorusdoping of Si also enhances the suppression and increases the Li+ diffusion coefficient. In contrast, the Si layer expands significantly in an organic electrolyte even with the charge capacity limit and even in an ionic-liquid electrolyte without the limit. We demonstrated that the homogeneously distributed Si lithiation−delithiation, phase-transition control from the Si to Li-rich Li−Si alloy phases, formation of a surface film with structural and/or mechanical stability, and faster Li+ diffusion contribute to suppressing Si volume expansion. KEYWORDS: lithium-ion batteries, silicon, anode, volume expansion, thickness, ionic liquid, phosphorus doping morphology.18−20 Electrolytes also improve the performance and safety of rechargeable batteries.21−23 An increment in the energy density leads to a probability of burning. Hence, nonflammable electrolytes can improve the safety of rechargeable batteries. As electrolyte solvents, ionic liquids provide superior physicochemical properties, including negligible vapor pressure, nonflammability, and wide electrochemical windows.24−27 We demonstrated that a Si anode offers a greater electrochemical property in some ionic-liquid electrolytes compared to that in usual organic electrolytes.28−30 Tables S1 and S2 summarize the expansion rates reported by other researchers for Si-based composite electrodes in ionicliquid and organic electrolytes, respectively.31−36 The rates are initially lower in ionic-liquid electrolytes, but the trend is unclear for longer cycling. While the rates of composite electrodes have been reported in organic electrolytes,34−39 significant volume expansion follows after extended cycling; thus, Si electrodes are unsuitable for practical use in organic electrolytes. Although these reports addressed the thickness of the composite electrodes, including the active material (Si),

1. INTRODUCTION Development of high-performance lithium-ion batteries (LIBs) with a high energy density and long cycle life is essential for establishing a sustainable society.1−4 Silicon (Si) has drawn attention as an active anode material for next-generation LIBs because of its high theoretical capacity (3580 mA h g−1 for Li15Si4).5−8 However, some of its drawbacks is that Si expands and contracts during alloying (charging) and dealloying (discharging) with Li, respectively. The transition from Si to crystalline Li15Si4 (c-Li15Si4) occurs at an expansion rate of 280%, generating high stress and large strain in the active materials.5,9 The resulting strain leads to pulverization and/or cracking of Si. Subsequently, the Si anode disintegrates and the reversible capacity fades. Additional drawbacks of Si are a low Li+ diffusion coefficient and high electrical resistivity, which prevent the practical use of Si-based electrodes.10,11 Many studies have addressed the above issues, by producing composite electrodes with improved mechanical properties,12,13 increasing the electrical conductivity of Si by coating it with conductive materials,14 constructing nanostructured Si materials that accommodate volume expansion,15−17 and doping Si with impurities, including phosphorus (P) and boron, to lower its electrical resistivity and/or alter its properties, such as its phase transition, crystallinity, and © 2019 American Chemical Society

Received: October 1, 2018 Accepted: January 2, 2019 Published: January 4, 2019 2950

DOI: 10.1021/acsami.8b17123 ACS Appl. Mater. Interfaces 2019, 11, 2950−2960

Research Article

ACS Applied Materials & Interfaces

Galvanostatic charge−discharge testing was performed with an electrochemical measurement system (HJ-1001SD8, Hokuto Denko Co., Ltd.) in the potential range between 0.005 and 2.000 V vs Li+/Li at 30 °C with a charge capacity limit of 1000 mA h g−1 unless stated otherwise. The charge capacity limit was performed by controlling the charging time (ca. 42 min); the discharging time was not limited and the cutoff voltage was set at 2.000 V. The current density was set at 0.36 A g−1 (0.1 C) during the first cycle and 1.44 A g−1 (0.4 C) during subsequent cycles (ca. 1600 cycles). The rate capability was also explored at current rates from 0.1 to 10 C. The peak fitting of the differential capacity (dQ/dV) plot was performed by Origin Pro 8.5.0J (LightStone) software. The Li+ diffusion coefficient in Si or P-doped Si electrode was estimated by a galvanostatic intermittent titration technique (GITT).41 The electrode was charged at a constant current density of 0.18 A g−1 (0.05 C) for 30 min and relaxed at open-circuit voltage (OCV) for 10 h. 2.3. Scanning Electron Microscopy and Soft X-ray Emission Spectroscopy. After galvanostatic charge−discharge testing, the coin cell was disassembled in the glovebox and the electrode was washed with propylene carbonate (PC, Kishida Chemical Co., Ltd.) and diethyl carbonate (Kishida Chemical Co., Ltd.) before drying. An electrode cross section was observed by field emission scanning electron microscopy (FE-SEM; JSM-7800F, JEOL Co., Ltd.) with soft X-ray emission spectroscopy (SXES). The electrode was not exposed to the atmosphere until it was introduced into the chamber of the FESEM using a transfer vessel. The working distance and acceleration voltage were fixed at 10 mm and 5 kV, respectively. The cross section was fabricated using a cross-section polisher (CP; IB09020CP, JEOL Co., Ltd.) or focused ion beam (FIB; JIB-4501, JEOL, Co., Ltd.). For the FIB process, the carbon coating of electrode surface was performed to protect the surface against damage by the Ga-ion beam. The surface morphology of the electrode was observed before and after the charge−discharge test by FE-SEM (JSM-6701F, JEOL Co., Ltd.). The working distance and acceleration voltage fixed at 10 mm and 3 kV, respectively. Confocal laser scanning microscopy (CLSM, VK-9700, Keyence) was also used to estimate the root-mean-square roughness (Sq) of the electrode surface. 2.4. Raman Microscopy. The surface crystallinity of Si or Pdoped Si was explored after charge−discharge testing by Raman microscopy (NanofinderFLEX, Tokyo Instruments, Inc.). The electrode was placed in a sealed cell to avoid exposure to the air and moisture. Raman spectra were excited at 532 nm with a Nd:YAG laser through a 50× long focus objective lens at room temperature. Raman imaging was 7 × 7 μm2, and 400 point spectra (20 × 20 points) were recorded. Surface Raman images were obtained by plotting the position of the band with the maximum intensity in a Raman shift range between 490 and 520 cm−1.

conductive agent, and binder, the Si volume change is not clearly understood. The expansion rates of Si thin films have been reported;40 however, it is essential to understand the expansion rates of microsized Si particles for practical applications. We previously reported that a capacity limit drastically improves the cycle life of a Si electrode since the severe stress accumulation should be suppressed by the moderate reduction in the Si volume expansion.13,28−30 We also demonstrated that a P-doped Si electrode with a minimal concentration of 124 ppm offers better cycling stability in a conventional organic electrolyte than a Si-alone electrode.18 However, the expansion rates of Si and P-doped Si have not been determined in an ionic-liquid electrolyte with a capacity limit during cycling. This study explored thickness changes in a Si or P-doped Si electrode with a capacity limit by employing field-emission scanning electron microscopy (FE-SEM). We confirm differences in the thickness of the Si layer in organic and ionic-liquid electrolytes, considering the Li storage distribution, phase transition, and surface film formation. In addition, we discuss the differences in the thicknesses of the Si and P-doped Si layers in the same ionic-liquid electrolyte based on the Li+ diffusion coefficient.

2. EXPERIMENTAL SECTION 2.1. Electrode Fabrication. Commercial Si powder (FUJIFILM Wako Pure Chemical Corporation, Ltd., 99.9%) and P-doped Si powder (100 ppm, Elkem, Silgrain e-Si) were employed as the active materials. We previously reported that the Si lattice constant decreased as the P concentration increasesd.18 The constant for 100 ppm P-doped Si used in this study is consistent with a previous correlation (Figure S1, Supporting Information). Additionally, the average particle size was approximately 5 μm (size distribution: 0.7− 10 μm), as shown in Figure S2. Si-based electrodes were prepared via a gas-deposition (GD) method with no binder and/or conductive additive to directly determine the Si volume expansion (Figure S3). A Cu current collector (thickness 20 μm, 99.9%, Nilaco Co., Ltd.) was placed 10 mm from the nozzle in a vacuum chamber with a guide tube. The nozzle (0.8 mm diameter) was attached to the end of the guide tube. After the chamber was evacuated to a base pressure of several tens of Pa, an aerosol consisting of the carrier gas (Ar or He, differential pressure 7.0 × 105 Pa) and active material powders (Si or P-doped Si) was generated in the guide tube and immediately sprayed from the nozzle onto the Cu substrate. The Ar and He gases were used to prepare Si and P-doped Si electrodes, respectively. The deposited active material had a thickness and weight of about 1.6 μm and 30 μg, respectively. Because the active material layer with a thickness above 1 μm can be fabricated by the GD method, it is possible to discuss the expansion rate of microsized Si particles. 2.2. Cell Assembly and Charge−Discharge Testing. A 2032type coin cell was constructed, comprising a Si or P-doped Si electrode as the working electrode, a glass fiber filter (Whatman GF/ A) as the separator, and a Li metal sheet (thickness: 1 mm, 99.90%, Rare Metallic Co., Ltd.) as the counter electrode. The ionic-liquid electrolyte was 1 mol dm−3 (M) lithium bis(fluorosulfonyl)amide (LiFSA) dissolved in N-methyl-N-propylpyrrolidinium bis(fluorosulfonyl)amide (Py13-FSA). LiFSA and Py13-FSA were obtained from Kishida Chemical Co., Ltd. and Kanto Chemical Co., Inc., respectively. For comparison, 1 M lithium bis(trifluoromethanesulfonyl)amide dissolved in propylene carbonate (LiTFSA/PC, Kishida Chemical Co., Ltd.) was employed as a conventional organic electrolyte. Cell assembly and electrolyte preparation were conducted in an Ar-filled glovebox (Miwa MFG, DBO-2.5LNKP-TS) with an oxygen content less than 1 ppm and a dew point below −100 °C.

3. RESULTS 3.1. Cycle Life of Si-Based Anodes. Figure 1 shows the cycle life of a Si-alone electrode with a charge capacity limit of 1000 mA h g−1 in organic (1 M LiTFSA/PC) and ionic-liquid electrolytes (1 M LiFSA/Py13-FSA) and its corresponding Coulombic efficiency (CE). The result for a P-doped Si electrode in the ionic-liquid electrolyte is also shown. The discharge capacity did not reach 1000 mA h g−1 during the initial 10 cycles in any of the above cases. Figure S4 shows the corresponding charge−discharge curves of the Si-alone and Pdoped Si electrodes at the 50th cycle in the organic and ionicliquid electrolytes. Potential plateaus were confirmed at around 0.2 and 0.4 V vs Li+/Li on charge and discharge curves regardless of the electrode and electrolyte, respectively.28 These plateaus are attributed to the lithiation and delithiation reactions of Si. The reversible capacity of the Si electrode decayed at around the 100th cycle in the organic electrolyte, and the CE drops around the 50th cycle. It is widely accepted that the surface film is formed through reductive decom2951


Research Article


liquid electrolytes, we investigated the thickness of the lithiated Si-based active material layers by FE-SEM (Figure 2). Table 1 Table 1. Thicknesses of Lithiated Si-Based Electrodes Estimated from Figure 2

Figure 1. Cycle life of Si-alone and P-doped Si electrodes in 1 M LiTFSA/PC or 1 M LiFSA/Py13-FSA with a charge capacity limit of 1000 mA h g−1 at 0.4 C (first cycle: 0.1 C). The corresponding CE is also shown.

position of the electrolyte solution, and thus, the film formation is one of the reasons for a decrease in the CE. A significant change in the Si volume during charge−discharge testing leads to pulverization and cracking of the Si active material layer as well as collapsing of the resulting film. Therefore, the surface film forms again on the newly formed Si surface. Conversely, in the ionic-liquid electrolyte, the Si electrode exhibited a better cycle life with a reversible capacity of 1000 mA h g−1 after the 600th cycle. Additionally, the cycle life of the P-doped Si electrode is greatly superior to that of the Sialone electrode in the same ionic-liquid electrolyte; the Pdoped Si electrode retained a reversible capacity of 1000 mA h g−1 after the 1400th cycle. While the CE of both electrodes was very low (below 5%) for the first cycle in the ionic-liquid electrolyte, this gradually increases and reaches 99% after the 60th cycle without decreasing, indicating that the active material layer does not disintegrate in the ionic-liquid electrolyte. Capacity limit mode also affects the cycle life of the P-doped Si electrode; the electrode exhibits a superior cycle life with a reversible capacity of 1000 mA h g−1 over 6000 cycles with a discharge capacity limit of 1000 mA h g−1 (Figure S5). 3.2. Change in Si Layer Thickness during Charge− Discharge Testing. To determine the difference between the cycle lives of Si and P-doped Si electrodes in organic and ionic-

Si-alone

Si-alone

P-doped Si

cycle number

Organic

Ionic liquid

Ionic liquid

10 50 100 300 600 1400

2.4 ± 1.0 μm 8.1 ± 1.9 μm 24.9 ± 3.2 μm

2.7 2.8 2.7 2.3 13.3

± ± ± ± ±

1.1 0.7 1.0 0.6 1.7

μm μm μm μm μm

1.6 2.7 5.2 12.9

± ± ± ±

1.0 0.8 0.6 1.2

μm μm μm μm

lists the average layer thicknesses and their standard deviations. Regardless of P-doping, the thickness of the nonlithiated Si layer before charge−discharge testing was 1.6 ± 0.3 μm, as reported in our papers.18,28 In the organic electrolyte, the Si layer thickness increases with the cycle number, developing several cracks and becoming porous after the 50th and 100th cycles (Figure 2a−c). The thickness reached 24.9 μm after the 100th cycle when the capacity faded. The amorphous Li1.00Si (a-Li1.00Si) phase may mainly form with a charge capacity limit of 1000 mA h g−1 because the phase has a theoretical capacity of 950 mA h g−1.42−46 The expansion rate of the Si layer was ca. 1460%, whereas the calculated rate of increase in thickness from Si to a-Li1.00Si phases is 17%. In the ionic-liquid electrolyte, the Si electrode retained a thickness of 2.7 μm over 300 cycles, with an expansion rate of about 69%. Unexpectedly, the Si layer did not expand significantly after repeated cycling. There are no reports on the significant suppression of Si volume expansion after longterm cycling, as is evident from Tables S1 and S2. However, the expansion rate of 69% is still above the calculated rate of 17% for the a-Li1.00Si phase. A nonporous Li−Si alloy phase is observed in Figure 2d−g, indicating that Li−Si alloy phases other than a-Li1.00Si form. After the 600th cycle and just before capacity fading, the layer thickness increased to 13.3 μm (expansion rate: 731%), with some void and crack formation inside the layer (Figure 2h).

Figure 2. Cross-sectional SEM images of lithiated Si active material layers with a charge capacity limit of 1000 mA h g−1. Each blue-colored area indicates a Si layer. Each image was observed after (a and d) 10, (b and e) 50, (c, f, and i) 100, (g and f) 300, (h and k) 600, and (l) 1400 cycles. (a−h) Si-alone and (i−l) P-doped Si electrodes were cycled in (a−c) 1 M LiTFSA/PC and (d−l) 1 M LiFSA/Py13-FSA. Cross sections (c), (h), and (l) were processed with a CP. 2952


Research Article


Figure 3. Wide view of the cross-sectional SEM images of the lithiated Si active material layers after the 100th cycle. Each blue-colored area indicates the Si layer. (a, b, d, and e) Si-alone and (c and f) P-doped Si electrodes were cycled (a−c) with a charge capacity limit of 1000 mA h g−1 and (d−f) without a limit in (a and d) 1 M LiTFSA/PC and (b, c, e, and f) 1 M LiFSA/Py13-FSA. Fibers observed in (a and d) are separator glass fibers. Note that the magnification of each part differs. The cross sections, except for parts (e) and (f), were processed by CP, while parts (e) and (f) were processed by FIB.

Figure 4. Time dependency of OCV. OCV data was measured after (a and d) 10, (b and e) 50, (c, f, and i) 100, (g and f) 300, (h and k) 600, and (l) 1400 cycles. (a−h) Si-alone and (i−l) P-doped Si electrodes were cycled in (a−c) 1 M LiTFSA/PC and (d−l) 1 M LiFSA/Py13-FSA with a charge capacity limit of 1000 mA h g−1.

P-doping results in gradual Si lithiation.18 Additionally, the thicknesses of all the Si-based layers exceeded 10 μm at the cycle immediately before capacity fading. The spots shown in Figure 2 are tens of micrometers in size because almost all of these cross sections were processed by FIB. Thus, strong suppression of volume expansion of the lithiated Si or P-doped Si layer in the ionic-liquid electrolyte with a charge capacity limit is doubtful. To eliminate this uncertainty, we processed a cross section of the Si-based electrodes after the 100th cycle using a CP that can process a wide sphere of about 1 mm (Figure 3a−c). Irrespective of the processing method, the thicknesses in a, b, and c of Figure 3 were almost the same as those in c, f, and i, respectively, of Figure 2; volume expansion of the Si or P-doped Si layer is significantly suppressed with a charge capacity limit of 1000 mA h g−1 in the ionic-liquid electrolyte. Thus, the SEM observations dispelled the above uncertainty.

Unbelievably, the P-doped Si layer thickness remained unchanged after 100 cycles, as shown in Figure 2i and Table 1; the P-doped Si electrode maintained a thickness of about 1.6 μm after the 100th cycle. Although it is doubtful that the Pdoped Si layer does not store Li, we verified that an area of the layer increased from 21 to 28 mm2 after the 100th cycle. This indicates that horizontal volume expansion of the layer is favored. It is believed that the P-doped Si layer easily expanded horizontally because it was perpendicularly pressed to the Cu current collector by the separator in the coin cell. We also used electrochemical measurement, Raman spectroscopy, and SXES to demonstrate that the layer stored Li, as discussed later. The thickness of the P-doped Si electrode was ca. 2.7 μm at the 300th cycle (Figure 2j), which was almost the same as that of the Si electrode in the ionic-liquid electrolyte (Figure 2g). The thickness of the P-doped Si layer increased to ca. 5.2 and 12.9 μm at the 600th and 1400th cycles, respectively, indicating that 2953


Research Article


Figure 5. Raman images of (a and b) Si-alone and (c) 100 ppm P-doped Si electrode surface after the 100th cycle in (a) 1 M LiTFSA/PC and (b and c) 1 M LiFSA/Py13-FSA. The electrode was cycled with a charge capacity limit of 1000 mA h g−1 at 0.4 C (first cycle, 0.1 C; after second cycle, 0.4 C). A Raman map was constructed by plotting the position of the maximum intensity in the frequency range from 490 to 520 cm−1. The mapping area is 7 × 7 μm2. Red regions indicate crystalline Si (unreacted with Li).

Figure 6. FE-SEM images of (a and b) Si-alone and (c) 100 ppm P-doped Si electrode surfaces after the 100th cycle in (a) 1 M LiTFSA/PC and (b and c) 1 M LiFSA/Py13-FSA. Part (d) shows the result of Si-alone electrode before the charge−discharge test. The electrodes were cycled with a charge capacity limit of 1000 mA h g−1 at 0.4 C (first cycle, 0.1 C; after second cycle, 0.4 C). Average Sq of the electrode surface estimated by CLSM is shown with a standard deviation at the lower left in each part. The standard deviation of Sq in part (d) was below 0.1 μm.

electrolyte are essential for suppressing volume expansion of the Si-based active material layers. 3.3. Li Storage Distribution. We conducted cell disassembly, CP or FIB process, and FE-SEM observation several days after charge−discharge testing. To confirm the self-discharge progression of the cells, we measured the OCV of Si and P-doped Si electrodes after the final charge− discharge test, as depicted in Figure 4. Each electrode maintains an OCV of about 0.2 V vs Li+/Li for several days after testing. The potential was almost in accordance with the potential plateau for alloying Si with Li.28 Therefore, the cell did not self-discharge until the FE-SEM observation, and Si

As a comparison, we obtained cross-sectional FE-SEM images of the Si and P-doped Si layers after the 100th cycle with no capacity limit (Figure 3d−f). Figure S6 shows the corresponding cycling performance. Si electrode disintegration, that is, exfoliation of the Si active material layer from the copper current collector, takes place in the organic electrolyte (Figure 3d). However, in the ionic-liquid electrolyte, the thicknesses of the Si and P-doped layers reached 32 μm at a maximum and 5.7 ± 1.3 μm, respectively (Figure 3e,f). Even in the ionic-liquid electrolyte, Si volume expansion is not significantly suppressed. Thus, the results in Figures 2 and 3 indicate that both the charge capacity limit and ionic-liquid 2954


Research Article


Figure 7. Li storage distribution. (Left) Cross-sectional FE-SEM and (right) SXE spectra of the lithiated Si electrode after 600 cycles in 1 M LiFSA/Py13-FSA with a constant charge capacity of 1000 mA h g−1. The crosses in the SEM image are the SXES measurement points, and the Roman numerals I−X correspond to the numbering in the SXE spectrum.

Figure 8. dQ/dV plot of the Si-based electrodes for the discharge reaction after (a, d, and i) 10, (b, e, and j) 50, (c, f, and k) 75, (g and l) 300, (h and m) 600, and (n) 1400 cycles. (a−h) Si-alone and (i−n) P-doped electrodes were cycled in (a−c) 1 M LiTFSA/PC and (d−n) 1 M LiFSA/ Py13-FSA with a charge capacity limit of 1000 mA h g−1. Alternative results are shown for the 75th cycle (c, f, and k) because the discharge capacity of the Si-alone electrode faded by the 100th cycle in the organic electrolyte.

stores the amount of Li corresponding to 1000 mA h g−1. The OCV of the Si electrode in the ionic-liquid electrolyte was slightly above 0.2 V vs Li+/Li after the 10th cycle (Figure 4d), indicating that Si lithiation is imperfect because of surface film formation. To confirm the distribution of regions, wherein the Si and Pdoped Si layers alloy with Li, we examined their surface crystallinity. Figure 5 shows Raman images of the Si and Pdoped Si electrode surfaces after the 100th cycle with a charge capacity limit of 1000 mA h g −1 . As we reported previously,28,29 the red regions in the images indicate crystalline Si (c-Si), which is not reacted with Li, while blue regions represent amorphous Si (a-Si) which is fully lithiated. In the regions shown in green and yellow, Si has partially lost

its crystallinity and has become lithiated. In the organic electrolyte (Figure 5a), c-Si is distributed in a patchy fashion on the electrode; local Li storage into the Si electrode occurs. While the Si crystallinity differs slightly between Si and Pdoped Si electrodes in the ionic-liquid electrolyte, the images demonstrate that the electrode surfaces react homogeneously with Li irrespective of P-doping (Figure 5b,c). Thus, the Si and P-doped Si layers alloy with Li. Additionally, the amount of c-Si in Figure 5c is higher than that in Figure 5b, indicating that Pdoping results in suppression of phase transition from c-Si to aSi.18 To investigate the change in surface morphology of the Sibased electrodes before and after the lithiation and delithiation, we observed Si-alone and 100 ppm P-doped Si electrode 2955


Research Article

ACS Applied Materials & Interfaces surfaces after the 100th cycle in organic and ionic-liquid electrolytes using FE-SEM (Figure 6). Average Sq value is also shown with standard deviation in each image. Si-alone electrode surface was markedly uneven in the organic electrolyte and the corresponding Sq was the largest (Figure 6a). The surface of the Si-alone electrode was rougher than that of the P-doped Si electrode in the ionic-liquid electrolyte (Figure 6b,c). A similar tendency was confirmed at the 200th cycle, whereas their Sq values increased slightly (Figure S7). Compared to the FE-SEM image before the charge−discharge test (Figure 6d), the Sq of the P-doped Si electrode was slightly larger at the 100th cycle. We directly verified by SXES that Li is stored in Si or Pdoped Si in the ionic-liquid electrolyte with a charge capacity limit. Figure 7 shows a cross-sectional SEM image of the lithiated Si layer after 600 cycles in the ionic-liquid electrolyte with a charge capacity limit and its SXES results. A peak assigned to Li appeared at around 0.054 keV in all the spectra except II. Although a point without Li is detected, this does not indicate that inhomogeneous Si lithiation occurs because the charge capacity is limited to 1000 mA h g−1. This also demonstrated that the suppression of the Si volume expansion (Figure 2d−g) is not caused by a lack of Li storage in Si. Distinct fluorine and sulfur peaks appear at almost all the measured points, indicating that the electrolyte penetrates the porous Si layer, with decomposition of the bis(fluorosulfonyl)amide (FSA) anion.18,28 It was reported that some inorganic compounds, including LiF and Li2O, form mainly as decomposition products of the FSA anion (N(SO2F)2−) and that these improve the structural and/or mechanical stability of the surface film.31,47−50 Figure S8 shows a cross-sectional SEM image of lithiated P-doped Si after 1400 cycles with the corresponding SXE spectrum. Similar results are obtained regardless of P-doping. 3.4. Phase Transition of Si during Charge−Discharge Testing. Previous experiments revealed that suppressing an increase in the thickness of the lithiated Si and P-doped Si layers improves the cycle life in the ionic-liquid electrolyte with a charge capacity limit. To clarify this mechanism, we examined the phase transition of Si during charge−discharge testing. Figure 8 shows a differential capacity (dQ/dV) plot for the discharge reaction with a constant charge capacity of 1000 mA h g−1 together with the peak-fitting results. Two anodic peaks at about 0.31 and 0.48 V vs Li+/Li correspond to delithiation from the amorphous Li-rich (a-Li-rich, including amorphous Li3.5−3.75Si,) and amorphous Li-poor (including amorphous Li2.0Si) Li−Si alloy phases, respectively.5,29,41,51 In both the organic and ionic-liquid electrolytes, the peak of the a-Li-rich phase, shown as a red line, gradually increases with the cycle number, irrespective of the electrode. Thus, the difference between the measured and calculated thicknesses in Figure 2 is caused by a-Li-rich phase formation. Figure 9 shows the correlation between the peak area of the a-Li-rich Li−Si alloy phase estimated from Figure 8 and the cycle number, together with the relationship between the thickness of the Si and P-doped Si layers and the cycle number. Both the peak area and thickness increase with the cycle number irrespective of the electrode and electrolyte. This demonstrates a correlation between the Si volume expansion and the amount of a-Li-rich phase. While the Si layer thickens exponentially and its peak area increases sharply in the organic electrolyte, the thickness and peak area increase gradually in the ionic-liquid electrolyte. Particularly, the initial peak area of

Figure 9. Correlation between the thickness of the lithiated Si layer or the peak area of the a-Li-rich phase and the cycle number. The peak area was estimated by peak fitting each dQ/dV plot (Figure 7). Blue, black, and red symbols represent the data for the Si-alone electrode in 1 M LiTFSA/PC, the Si-alone electrode in 1 M LiFSA/Py13-FSA, and the P-doped Si electrode in 1 M LiFSA/Py13-FSA, respectively.

the P-doped Si layer is approximately one-third of that of the Si layer in the ionic-liquid electrolyte, indicating that P-doping strongly suppresses formation of the a-Li-rich phase.18 3.5. Li+ Diffusion in Si. To calculate the Li+ diffusion coefficient (DLi+) in the Si and P-doped Si electrodes, we obtained the GITT curve of the electrodes, as shown in Figure 10a. DLi+ is given by

ij ΔES yz jj z jj ΔE zzz (1) k τ{ where mB, Vm, and MB are the mass, molar volume, and molecular weight of the active material (P-doped Si or Si-alone powders), respectively. Herein, we used the following values: Vm = 12.06 cm3 mol−1 and MB = 28.09 g mol−1. S is the area of the active material layer on the Cu current collector, and τ is the time duration of the current pulse. ΔES and ΔEτ are the difference between the steady-state potentials measured at the end of sequential open-circuit relaxation steps and the potential change during charging at the time of the flux current with no IR drop, respectively. Consequently, DLi+ for the Si and P-doped Si electrodes is ca. 3.8 × 10−13 and 2.3 × 10−12 cm2 s−1, respectively (Figure 10b): Li+ diffusion in the Pdoped Si electrode is faster than that in the Si electrode. Figure 11 gives the rate capability of the Si and P-doped Si electrodes in the ionic-liquid electrolyte with a charge capacity limit. The Si electrode retained a reversible capacity of 1000 mA h g−1 at a relatively high current rate of 2.0 C, but decreased to about 750 and 500 mA h g−1 at current rates of 5 and 10 C, respectively. Additionally, the capacity recovers at the initial current rate of 0.1 C; that is, no electrode disintegration occurs. Conversely, the P-doped Si electrode exhibits no capacity fading even at 10 C. The reversible capacity of P-doped Si electrode faded at a high current rate of 20 and 50 C (data not shown). The improved rate performance of the P-doped Si electrode is due to a higher DLi+. D Li+ =

4 ijj mBVm yzz j z πτ jjk MBS zz{

2

2

4. DISCUSSION As described above, the cycle life with a reversible capacity of 1000 mA h g−1 has the following order: P-doped Si electrode in the ionic-liquid electrolyte (ca. 1400 cycles) > Si-alone electrode in the ionic-liquid electrolyte (ca. 600 cycles) > Sialone electrode in the organic electrolyte (ca. 100 cycles). The cycle life differences were attributed to suppression of volume 2956


Research Article


Figure 10. (a) GITT curve of P-doped Si electrode during the first lithiation and (b) diffusion coefficient of Li+ in 100 ppm P-doped Si and Sialone electrodes. Inset in part (a) shows enlarged view between 8 and 21 h.

revealed that the same results are obtained at both the surface and the cross section of the Si layer. While the Raman analysis and SXES measurements were obtained with no charge capacity limit, the same results will be obtained with the limit (we have already partially confirmed this, as shown in Figures 5 and 7). Although the a-Li1.00Si phase should mainly form with a charge capacity limit of 1000 mA h g−1, the a-Li-rich phase exists irrespective of the electrolyte (Figure 8). a-Li-rich phase formation indicates inhomogeneous Si lithiation. The stress and strain in the a-Li-rich phase are greater than those in the aLi-poor phase. In Figure 9, the a-Li-rich phase peak area of the Si-alone electrode reaches about 400 mA h g−1 at the cycle when the discharge capacity fades, irrespective of the electrolyte; Si layer disintegration is expected to occur when the amount of a-Li-rich phase exceeds a threshold level. However, the value of 400 mA h g−1 for the peak area of the aLi-rich phase is inconsistent with a result that the capacity fading of P-doped Si electrode occurs before the a-Li-rich phase peak area reaches 400 mA h g−1. Because the dQ/dV plots merely represent a mathematical process, we need to accurately determine the amount of a-Li-rich phase using nuclear magnetic resonance and/or XRD to determine the threshold level.52 Although the a-Li-rich phase peak area of the Si-alone electrode after 50 cycles in the organic electrolyte is less than that after 300 cycles in the ionic-liquid electrolyte, the Si layer thickness of the former is greater than that of the latter (Figure 9). This indicates that the Si layer thickness is not solely proportional to the amount of a-Li-rich phase. As well as the amount, the distribution of the phase should affect the increase in the Si layer thickness. When the a-Li-rich phase is inhomogeneously distributed, a large local strain accumulates, and a dramatic increase in the Si thickness should occur. Thus, the discharge capacity fades within 100 cycles in the organic electrolyte. Moderate control of the amount and distribution of the a-Li-rich phase is required to suppress the Si volume expansion. A surface film forms on anodes by reductive electrolyte decomposition during the initial charging stage, and the film properties strongly influence the battery performance.53−56 F and S were detected in the Si electrode cross section (Figure 7), indicating that the electrolyte osmosed the porous Si layer and decomposed. Additionally, the surface film arising from FSA anion formed in the ionic-liquid electrolyte because the N-methyl-N-propylpyrrolidinium cation (C8H18N+) does not contain F and S. Some research groups reported that the

Figure 11. Rate performance of Si-alone and 100 ppm P-doped Si electrodes in 1 M LiFSA/Py13-FSA. The electrode was cycled with a charge capacity limitation of 1000 mA h g−1.

expansion of the lithiated active material layer. Irrespective of the capacity limit, the expansion rate of the active material layer differs depending on the electrolyte and electrode. We now discuss differences in the performances of the Si-alone electrode in the organic and ionic-liquid electrolytes, taking into account the Li storage distribution, phase transition, and surface film formation. We then confirm the effect of P-doping in Si based on the Li+ diffusion coefficient. We previously investigated the distributions of reacted regions with Li on a Si electrode surface after charge− discharge test by Raman microspectroscopy.28,29 As a result, it was revealed that local c-Si domains remain on the surface in the PC-based organic electrolyte. That is, Si lithiation− delithiation occurs inhomogeneously. A surface film with an uneven thickness should be formed on the Si surface. Then Li+ is preferentially stored into the electrode through the thinner parts of the film because the thicker parts prevent the Li+ storage. Therefore, local change in the volume of Si occurs, which causes the electrode disintegration and rapid capacity fading. Conversely, the a-Si domains were uniformly distributed on the surface in the FSA-based ionic-liquid electrolyte.28 This indicates that Li+ is homogeneously stored on the entire electrode surface because of the formation of a uniform and thin surface film, preventing the occurrence of local stress. Thus, severe Si electrode disintegration, which causes capacity fading, is suppressed. We also reported the distributions of lithiation and delithiation on cross sections of Si electrodes determined by SXES.28 In the PC-based organic electrolyte, the Li peak appeared at only half of the measured points. Conversely, Li was detected at all the points in the FSA-based ionic-liquid electrolyte, indicating that Li reacts uniformly with the cross section. Raman analysis and SXES 2957


Research Article


ionic-liquid electrolyte. Lithiation of the Si and P-doped Si layers certainly occurred with the capacity limit based on the time dependency of OCV, Raman images, and SXES. The homogeneous distribution of Si lithiation−delithiation, the phase transition control from Si to a-Li-rich phases, and the formation of a structural and/or mechanically stable surface film contribute to suppression of the Si volume expansion in the ionic-liquid electrolyte. Additionally, P-doping exhibits pinning and acceleration effects that suppress the amount of Li storage in Si at an electrode/electrolyte interface and accelerate Li+ diffusion within Si, respectively. These effects also suppress volume expansion and roughening of Si, leading to an excellent electrochemical performance. Si-based electrodes with even longer cycle life can be developed by optimizing the capacity limit, ionic-liquid electrolyte, and limit mode (charge or discharge), and these electrodes will allow a sustainable society to be established.

decomposition products of FSA anions were mainly inorganic compounds including LiF and Li2O.29,31,47,50 These improved the structural and/or mechanical stability of surface films.31,49,57 Although LiTFSA (LiN(SO2CF3)2) in the organic electrolyte contains fluorine, almost no F was detected, as reported previously.28 In addition, carbon and oxygen were detected in the organic electrolyte, indicating that PC mainly decomposes. Therefore, the formation of structurally and/or mechanically stable surface films, including LiF and Li2O, also improves cycling performance of the Si electrode in the ionicliquid electrolyte. We have discussed differences in the cycle life of the Si-alone electrode in the organic and ionic-liquid electrolytes. Because we used the same ionic-liquid electrolyte, the improved cycle life of the P-doped Si electrode cannot be rationalized by the above three points (the Li storage distribution, phase transition, and surface film formation). We also reported that P-doping suppresses phase transition from Si to c-Li15Si4 phases in the PC-based organic electrolyte.18 This is due to the shrinking of Si crystal lattice as a result of the displacement of Si atoms with smaller P atoms, and it is thus difficult to insert Li into Si. We considered this suppression of phase transition to resemble a capacity limit from an external circuit. However, the P-doped Si shows a better cycle life even with a capacity limit by an external circuit, as shown in Figure 1; thus, the cycle life cannot be explained only by simple suppression of the phase transition. The entire P-doped Si layer reacted homogeneously and smoothly with Li due to its higher DLi+, which suppressed a-Lirich phase formation for a long cycle (Figure 8 and Figure 9). Because the a-Li-rich phase was smaller on the P-doped Si electrode, the thickness should be less at the cycle when capacity fades than that of the Si electrode (Figure 2h,l ). Thus, hardly any strain is generated in the active material during repeated cycling and no disintegration of the P-doped Si electrode occurs within the 1400th cycle, resulting in an improved cycle life (Figure 1). Compared to Figure 2h, Pdoping also suppressed crack and pore formation, as shown in Figure 2k,l. It is believed that fast Li+ diffusion in the entire Pdoped Si active material layer accelerates uniform volume expansion and/or phase transition from Si to a-Li-rich Li−Si alloy phases. Although the reason for the higher DLi+ of the Pdoped Si remains unclear, we expected that the diffusion pass of Li+ in the P-doped Si differs from that in the undoped Si. Thus, P-doping has two effects: one is a pinning effect to suppress the amount of Li storage in Si at an electrode/ electrolyte interface and the other is an acceleration effect on Li+ diffusion within Si. To develop Si-based electrodes with even longer cycle life, it is important to moderately suppress the formation and growth of the a-Li-rich phase. While a high capacity cannot be obtained without a-Li-rich phase, an increase in the phase must cause Si-based electrode disintegration. Such electrodes can be developed by optimizing the capacity limit value, ionic-liquid electrolyte, and limit mode (charge or discharge).

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.8b17123.

■

XRD pattern, lattice constant, FE-SEM images, and particle size distribution of P-doped Si, illustration of GD method, charge−discharge curves, cycle life with a discharge capacity limit of 1000 mA h g−1, cycling performance without a capacity limit, Li storage distribution of P-doped Si, and expansion rate (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (H.S.). ORCID

Yasuhiro Domi: 0000-0003-3983-2202 Hiroyuki Usui: 0000-0002-1156-0340 Hiroki Sakaguchi: 0000-0002-4125-7182 Author Contributions

The manuscript was written through contributions of all authors. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was partially supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI (Grant Nos. JP17K17888, JP17H03128, and JP16K05954) and the MEXT Program for Development of Environmental Technology using Nanotechnology. The authors thank Enago (www. enago.jp) for the English language review.

■

REFERENCES

(1) Tarascon, J.-M.; Armand, M. Issues and Challenges Facing Rechargeable Lithium Batteries. Nature 2001, 414, 359−367. (2) Armand, M.; Tarascon, J.-M. Building Better Batteries. Nature 2008, 451, 652−657. (3) Whittingham, M. S. Lithium Batteries and Cathode Materials. Chem. Rev. 2004, 104, 4271−4301. (4) Goodenough, J. B.; Kim, Y. Challenges for Rechargeable Li Batteries. Chem. Mater. 2010, 22, 587−603. (5) Obrovac, M. N.; Krause, L. J. Reversible Cycling of Crystalline Silicon Powder. J. Electrochem. Soc. 2007, 154, A103−A108.

5. CONCLUSION Combining an ionic-liquid electrolyte (1 M LiFSA/Py13-FSA) and a charge capacity limit of 1000 mA h g−1 significantly suppressed the volume expansion of the Si and P-doped Si layers and improved the life cycle. The Si and P-doped Si electrodes have maintained a reversible capacity of 1000 mA h g−1 beyond the 600th and 1400th cycles, respectively, in the 2958


Research Article

ACS Applied Materials & Interfaces (6) Key, B.; Morcrette, M.; Tarascon, J.-M.; Grey, C. P. Pair Distribution Function Analysis and Solid State NMR Studies of Silicon Electrodes for Lithium Ion Batteries: Understanding the (De)lithiation Mechanisms. J. Am. Chem. Soc. 2011, 133, 503−512. (7) Lai, S.-C. Solid Lithium-Silicon Electrode. J. Electrochem. Soc. 1976, 123, 1196−1197. (8) Obrovac, M. N.; Christensen, L. Structural Changes in Silicon Anodes during Lithium Insertion/Extraction. Electrochem. Solid-State Lett. 2004, 7, A93−A96. (9) Liu, X. H.; Zhong, L.; Huang, S.; Mao, S. X.; Zhu, T.; Huang, J. Y. Size-Dependent Fracture of Silicon Nanoparticles during Lithiation. ACS Nano 2012, 6, 1522−1531. (10) Ding, N.; Xu, J.; Yao, Y. X.; Wegner, G.; Fang, X.; Chen, C. H.; Lieberwirth, I. Determination of the Diffusion Coefficient of Lithium Ions in Nano-Si. Solid State Ionics 2009, 180, 222−225. (11) Xie, J.; Imanishi, N.; Zhang, T.; Hirano, A.; Takeda, Y.; Yamamoto, O. Li-Ion Diffusion in Amorphous Si Films Prepared by RF Magnetron Sputtering: A Comparison of Using Liquid and Polymer Electrolytes. Mater. Chem. Phys. 2010, 120, 421−425. (12) Domi, Y.; Usui, H.; Takemoto, Y.; Yamaguchi, K.; Sakaguchi, H. Improved Electrochemical Performance of Lanthanum Silicide/ Silicon Composite Electrode with Nickel Substitution for Lithium-Ion Batteries. J. Phys. Chem. C 2016, 120, 16333−16339. (13) Domi, Y.; Usui, H.; Narita, M.; Fujita, Y.; Yamaguchi, K.; Sakaguchi, H. Advanced Performance of Annealed Ni-P/(Etched Si) Negative Electrodes for Lithium-Ion Batteries. J. Electrochem. Soc. 2017, 164, A3208−A3213. (14) Zhou, X.; Yin, Y.-X.; Cao, A.-M.; Wan, L.-J.; Guo, Y.-G. Efficient 3D Conducting Networks Built by Graphene Sheets and Carbon Nanoparticles for High-Performance Silicon Anode. ACS Appl. Mater. Interfaces 2012, 4, 2824−2828. (15) McDowell, M. T.; Ryu, I.; Lee, S. W.; Wang, C.; Nix, W. D.; Cui, Y. Studying the Kinetics of Crystalline Silicon Nanoparticle Lithiation with In Situ Transmission Electron Microscopy. Adv. Mater. 2012, 24, 6034−6041. (16) Zhou, X.; Wan, L.-J.; Guo, Y.-G. Electrospun Silicon Nanoparticle/Porous Carbon Hybrid Nanofibers for Lithium-Ion Batteries. Small 2013, 9, 2684−2688. (17) Liu, B.; Wang, X.; Chen, H.; Wang, Z.; Chen, D.; Cheng, Y.-B.; Zhou, C.; Shen, G. Hierarchical Silicon Nanowires-Carbon Textiles Matrix as a Binder-Free Anode for High-Performance Advanced Lithium-Ion Batteries. Sci. Rep. 2013, 3, 1622. (18) Domi, Y.; Usui, H.; Shimizu, M.; Kakimoto, Y.; Sakaguchi, H. Effect of Phosphorus-Doping on Electrochemical Performance of Silicon Negative Electrodes in Lithium-Ion Batteries. ACS Appl. Mater. Interfaces 2016, 8, 7125−7132. (19) Long, B. R.; Chan, M. K. Y.; Greeley, J. P.; Gewirth, A. A. Dopant Modulated Li Insertion in Si for Battery Anodes: Theory and Experiment. J. Phys. Chem. C 2011, 115, 18916−18921. (20) Yi, R.; Zai, J.; Dai, F.; Gordin, M. L.; Wang, D. Improved Rate Capability of Si-C Composite Anodes by Boron Doping for LithiumIon Batteries. Electrochem. Commun. 2013, 36, 29−32. (21) Yamada, Y.; Furukawa, K.; Sodeyama, K.; Kikuchi, K.; Yaegashi, M.; Tateyama, Y.; Yamada, A. Unusual Stability of Acetonitrile-Based Superconcentrated Electrolytes for Fast-Charging Lithium-Ion Batteries. J. Am. Chem. Soc. 2014, 136, 5039−5046. (22) Wang, J.; Yamada, Y.; Sodeyama, K.; Chiang, C. H.; Tateyama, Y.; Yamada, A. Superconcentrated Electrolytes for a High-Voltage Lithium-Ion Battery. Nat. Commun. 2016, 7, 12032. (23) Usui, H.; Domi, Y.; Shimizu, M.; Imoto, A.; Yamaguchi, K.; Sakaguchi, H. Niobium-Doped Titanium Oxide Anode and Ionic Liquid Electrolyte for a Safe Sodium-Ion Battery. J. Power Sources 2016, 329, 428−431. (24) Earle, M. J.; Esperança, J. M. S. S.; Gilea, M. A.; Lopes, J. N. C.; Rebelo, L. P. N.; Magee, J. W.; Seddon, K. R.; Widegren, J. A. The Distillation and Volatility of Ionic Liquids. Nature 2006, 439, 831− 834. (25) Hapiot, P.; Lagrost, C. Electrochemical Reactivity in RoomTemperature Ionic Liquids. Chem. Rev. 2008, 108, 2238−2264.

(26) Sippel, P.; Lunkenheimer, P.; Krohns, S.; Thoms, E.; Loidl, A. Importance of Liquid Fragility for Energy Applications of Ionic Liquids. Sci. Rep. 2015, 5, 13922. (27) Ghoufi, A.; Szymczyk, A.; Malfreyt, P. Ultrafast Diffusion of Ionic Liquids Confined in Carbon Nanotubes. Sci. Rep. 2016, 6, 28518. (28) Yamaguchi, K.; Domi, Y.; Usui, H.; Sakaguchi, H. Elucidation of the Reaction Behavior of Silicon Negative Electrodes in a Bis(fluorosulfonyl)amide-Based Ionic Liquid Electrolyte. ChemElectroChem 2017, 4, 3257−3263. (29) Shimizu, M.; Usui, H.; Suzumura, T.; Sakaguchi, H. Analysis of the Deterioration Mechanism of Si Electrode as a Li-Ion Battery Anode Using Raman Microspectroscopy. J. Phys. Chem. C 2015, 119, 2975−2982. (30) Yamaguchi, K.; Domi, Y.; Usui, H.; Shimizu, M.; Matsumoto, K.; Nokami, T.; Itoh, T.; Sakaguchi, H. Influence of the Structure of the Anion in an Ionic Liquid Electrolyte on the Electrochemical Performance of a Silicon Negative Electrode for a Lithium-Ion Battery. J. Power Sources 2017, 338, 103−107. (31) Piper, D. M.; Evans, T.; Leung, K.; Watkins, T.; Olson, J.; Kim, S. C.; Han, S. S.; Bhat, V.; Oh, K. H.; Buttry, D. A.; Lee, S.-H. Stable Silicon-Ionic Liquid Interface for Next-Generation Lithium-Ion Batteries. Nat. Commun. 2015, 6, 6230. (32) Chen, C.-Y.; Sawamura, A.; Tsuda, T.; Uchida, S.; Ishikawa, M.; Kuwabata, S. Visualization of Si Anode Reactions in Coin-Type Cells via Operando Scanning Electron Microscopy. ACS Appl. Mater. Interfaces 2017, 9, 35511−35515. (33) Chen, C.-Y.; Sano, T.; Tsuda, T.; Ui, K.; Oshima, Y.; Yamagata, M.; Ishikawa, M.; Haruta, M.; Doi, T.; Inaba, M.; Kuwabata, S. In situ Scanning Electron Microscopy of Silicon Anode Reactions in Lithium-Ion Batteries during Charge/Discharge Processes. Sci. Rep. 2016, 6, 36153. (34) Inaba, M.; Haruta, M.; Saito, M.; Doi, T. Silicon Nano-Flake Powder as an Anode for the Next Generation Lithium-Ion Batteries: Current Status and Challenges. Electrochemistry 2017, 85, 623−629. (35) Hwang, C.; Joo, S.; Kang, N.-R.; Lee, U.; Kim, T.-H.; Jeon, Y.; Kim, J.; Kim, Y.-J.; Kim, J.-Y.; Kwak, S.-K.; Song, H.-K. Breathing Silicon Anodes for Durable High-Power Operations. Sci. Rep. 2015, 5, 14433. (36) Choi, S.; Kwon, T.-W.; Coskun, A.; Choi, J. W. Highly Elastic Binders Integrating Polyrotaxanes for Silicon Microparticle Anodes in Lithium Ion Batteries. Science 2017, 357, 279−283. (37) Ko, M.; Chae, S.; Jeong, S.; Oh, P.; Cho, J. Elastic a-Silicon Nanoparticle Backboned Graphene Hybrid as a Self-Compacting Anode for High-Rate Lithium Ion Batteries. ACS Nano 2014, 8, 8591−8599. (38) Lu, B.; Ma, B.; Deng, X.; Li, W.; Wu, Z.; Shu, H.; Wang, X. Cornlike Ordered Mesoporous Silicon Particles Modified by Nitrogen-Doped Carbon Layer for the Application of Li-Ion Battery. ACS Appl. Mater. Interfaces 2017, 9, 32829−32839. (39) Yan, L.; Liu, J.; Wang, Q.; Sun, M.; Jiang, Z.; Liang, C.; Pan, F.; Lin, Z. In Situ Wrapping Si Nanoparticles with 2D Carbon Nanosheets as High-Areal-Capacity Anode for Lithium-Ion Batteries. ACS Appl. Mater. Interfaces 2017, 9, 38159−38164. (40) Jiménez, A. R.; Klöpsch, R.; Wagner, R.; Rodehorst, U. C.; Kolek, M.; Nölle, R.; Winter, M.; Placke, T. A Step toward HighEnergy Silicon-Based Thin Film Lithium Ion Batteries. ACS Nano 2017, 11, 4731−4744. (41) Huang, Z.; Wang, Z.; Guo, H.; Li, X. Influence of Mg2+ Doping on the Structure and Electrochemical Performances of Layered LiNi0.6Co0.2‑xMn0.2MgxO2 Cathode Materials. J. Alloys Compd. 2016, 671, 479−485. (42) Ogata, K.; Salager, E.; Kerr, C. J.; Fraser, A. E.; Ducati, C.; Morris, A. J.; Hofmann, S.; Grey, C. P. Revealing Lithium-Silicide Phase Transformations in Nano-Structured Silicon-Based Lithium Ion Batteries via In Situ NMR Spectroscopy. Nat. Commun. 2014, 5, 4217. 2959


Research Article

ACS Applied Materials & Interfaces (43) Li, J.; Dahn, J. R. An In Situ X-Ray Diffraction Study of the Reaction of Li with Crystalline Si. J. Electrochem. Soc. 2007, 154, A156−A161. (44) Misra, S.; Liu, N.; Nelson, J.; Hong, S. S.; Cui, Y.; Toney, M. F. In Situ X-ray Diffraction Studies of (De)lithiation Mechanism in Silicon Nanowire Anodes. ACS Nano 2012, 6, 5465−5473. (45) Wang, C.-M.; Li, X.; Wang, Z.; Xu, W.; Liu, J.; Gao, F.; Kovarik, L.; Zhang, J.-G.; Howe, J.; Burton, D. J.; Liu, Z.; Xiao, X.; Thevuthasan, S.; Baer, D. R. In Situ TEM Investigation of Congruent Phase Transition and Structural Evolution of Nanostructured Silicon/ Carbon Anode for Lithium Ion Batteries. Nano Lett. 2012, 12, 1624− 1632. (46) Wang, J. W.; He, Y.; Fan, F.; Liu, X. H.; Xia, S.; Liu, Y.; Harris, C. T.; Li, H.; Huang, J. Y.; Mao, S. X.; Zhu, T. Two-Phase Electrochemical Lithiation in Amorphous Silicon. Nano Lett. 2013, 13, 709−715. (47) Budi, A.; Basile, A.; Opletal, G.; Hollenkamp, A. F.; Best, A. S.; Rees, R. J.; Bhatt, A. I.; O’Mullane, A. P.; Russo, S. P. Study of the Initial Stage of Solid Electrolyte Interphase Formation upon Chemical Reaction of Lithium Metal and N-Methyl-N-Propyl-PyrrolidiniumBis(Fluorosulfonyl)Imide. J. Phys. Chem. C 2012, 116, 19789−19797. (48) Shkrob, I. A.; Marin, T. W.; Zhu, Y.; Abraham, D. P. Why Bis(fluorosulfonyl)imide Is a “Magic Anion” for Electrochemistry. J. Phys. Chem. C 2014, 118, 19661−19671. (49) Schroder, K.; Alvarado, J.; Yersak, T. A.; Li, J.; Dudney, N.; Webb, L. J.; Meng, Y. S.; Stevenson, K. J. The Effect of Fluoroethylene Carbonate as an Additive on the Solid Electrolyte Interphase on Silicon Lithium-Ion Electrodes. Chem. Mater. 2015, 27, 5531−5542. (50) Philippe, B.; Dedryvère, R.; Gorgoi, M.; Rensmo, H.; Gonbeau, D.; Edström, K. Improved Performances of Nanosilicon Electrodes Using the Salt LiFSI: A Photoelectron Spectroscopy Study. J. Am. Chem. Soc. 2013, 135, 9829−9842. (51) Ogata, K.; Jeon, S.; Ko, D.-S.; Jung, I. S.; Kim, J. H.; Ito, K.; Kubo, Y.; Takei, K.; Saito, S.; Cho, Y.-H.; Park, H.; Jang, J.; Kim, H.G.; Kim, J.-H.; Kim, Y. S.; Choi, W.; Koh, M.; Uosaki, K.; Doo, S. G.; Hwang, Y.; Han, S. Evolving Affinity between Coulombic Reversibility and Hysteretic Phase Transformations in Nanostructured SiliconBased Lithium-Ion Batteries. Nat. Commun. 2018, 9, 479. (52) Key, B.; Bhattacharyya, R.; Morcrette, M.; Seznéc, V.; Tarascon, J.-M.; Grey, C. P. Real-Time NMR Investigations of Structural Changes in Silicon Electrodes for Lithium-Ion Batteries. J. Am. Chem. Soc. 2009, 131, 9239−9249. (53) Tsubouchi, S.; Domi, Y.; Doi, T.; Ochida, M.; Nakagawa, H.; Yamanaka, T.; Abe, T.; Ogumi, Z. Spectroscopic Analysis of Surface Layers in Close Contact with Edge Plane Graphite NegativeElectrodes. J. Electrochem. Soc. 2013, 160, A575−A580. (54) Wu, H.; Yu, G.; Pan, L.; Liu, N.; McDowell, M. T.; Bao, Z.; Cui, Y. Stable Li-Ion Battery Anodes by In-Situ Polymerization of Conducting Hydrogel to Conformally Coat Silicon Nanoparticles. Nat. Commun. 2013, 4, 1943. (55) Domi, Y.; Doi, T.; Tsubouchi, S.; Yamanaka, T.; Abe, T.; Ogumi, Z. Irreversible Morphological Changes of a Graphite Negative-Electrode at High Potentials in LiPF6-Based Electrolyte Solution. Phys. Chem. Chem. Phys. 2016, 18, 22426−22433. (56) Domi, Y.; Doi, T.; Nakagawa, H.; Yamanaka, T.; Abe, T.; Ogumi, Z. In Situ Raman Study on Reversible Structural Changes of Graphite Negative-Electrodes at High Potentials in LiPF6-Based Electrolyte Solution. J. Electrochem. Soc. 2016, 163, A2435−A2440. (57) Kim, G.-T.; Kennedy, T.; Brandon, M.; Geaney, H.; Ryan, K. M.; Passerini, S.; Appetecchi, G. B. Behavior of Germanium and Silicon Nanowire Anodes with Ionic Liquid Electrolytes. ACS Nano 2017, 11, 5933−5943.

2960


Silicon-Based Anodes with Long Cycle Life for Lithium-Ion Batteries

Recommend Documents