Locally Concentrated LiPF6 in a Carbonate-Based ... - ACS Publications

Subscriber access provided by MIDWESTERN UNIVERSITY

Energy, Environmental, and Catalysis Applications

Locally Concentrated LiPF6 in Carbonate-based Electrolyte with Fluoroethylene Carbonate as a Diluent for Anode-Free Lithium Metal Battery Tesfaye Teka Hagos, Balamurugan Thirumalraj, Chen-Jui Huang, Ljalem Hadush Abrha, Teklay Mezgebe Hagos, Gebregziabher Brhane Berhe, Hailemariam Kassa Bezabh, Jim Cherng, Shuo-Feng Chiu, Wei-Nien Su, and Bing-Joe Hwang ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b21052 • Publication Date (Web): 21 Feb 2019 Downloaded from http://pubs.acs.org on February 21, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Locally Concentrated LiPF6 in Carbonate-based Electrolyte with Fluoroethylene Carbonate as a Diluent for Anode-Free Lithium Metal Battery Tesfaye Teka Hagosa, Balamurugan Thirumalrajb, Chen-Jui Huangb, Ljalem Hadush Abrhab, Teklay Mezgebe Hagosb, Gebregziabher Brhane Berhea, Hailemariam Kassa Bezabhb, Jim Cherngc, Shuo-Feng Chiub, Wei-Nien Sua,*, Bing-Joe Hwangb,d,* aNano-Electrochemistry

Laboratory, Graduate Institute of Applied Science and Technology,

National Taiwan University of Science and Technology, Taipei 106, Taiwan. bNano-Electrochemistry

Laboratory, Department of Chemical Engineering, National Taiwan.

University of Science and Technology, Taipei 106, Taiwan. cAmita

Technologies Inc., Taoyuan County 33349, Taiwan.

dNational

Synchrotron Radiation Research Center, Hsin-Chu, Taiwan.

*Corresponding Authors: [email protected]; [email protected]

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Currently, concentrated electrolyte solutions are attracting special attention because of their unique characteristics such as unusually improved oxidative stability on both the cathode and anode side, the absence of free solvent, the presence of more anion content and the improved availability of Li+ ions. Most of the concentrated electrolytes reported are lithium bis(fluorosulfonyl)imide (LiFSI) salt with ether-based solvents due to the high solubility of salts in ether-based solvents. However, their poor anti-oxidation capability hindered their application especially with high potential cathode materials (> 4.0 V). In addition, the salt is very costly, so it is not feasible from cost analysis point of view. Therefore, here we report a locally concentrated electrolyte, 2M LiPF6 in ethylene carbonate (EC)/diethyl carbonate (DEC) (1:1 v/v ratio) diluted with fluoroethylene carbonate (FEC) which is stable within wide potential range (2.5-4.5 V). It shows significant improvement in cycling stability of lithium with an average coulombic efficiency (ACE) of ~98% and small voltage hysteresis (~30 mV) with current density of 0.2 mA/cm2 for over 1066 hr in Li‖Cu cell. Furthermore, we ascertained the compatibility of the electrolyte for anode free Li-metal battery (AFLMB) using Cu‖LiNi1/3Mn1/3Co1/3O2 (NMC, ~2 mAh/cm2) with a current density of 0.2 mA/cm2. It shows stable cyclic performance with ACE of 97.8% and 40% retention capacity at 50th cycle, which is the best result reported for carbonate based solvents with AFLMB. Whereas, the commercial carbonate based electrolyte have < 90% ACE and even cannot proceed more than 15 cycles with retention capacity > 40%. The enhanced cycle life and well retained in capacity of the locally concentrated electrolyte is mainly because of the synergetic effect of FEC as the diluent to increase the ionic conductivity and form stable anion-derived SEI. The locally concentrated electrolyte also shows high robustness to the effect of upper limit cut-off voltage. Keywords: Anion-derived SEI; anode-free Lithium metal battery; carbonate based solvent; copper foil; diluent;

2 ACS Paragon Plus Environment

Page 2 of 22

Page 3 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. INTRODUCTION Consumers are in need of efficient energy storage materials with an affordable cost, better energy density, safe and extended cycle life.1-2 Lithium ion batteries (LIBs) with graphite anode are presently envisioned as promising and highly efficient energy storage devices for different electronics, electric vehicles and smart grid storage systems.3-5 However, the currently existing commercial lithium-ion technology cannot provide adequate energy density in paralleled with the increasing demand for energy density. So Lithium metal has been well thought-out as a significant anode material to replace graphite to further escalate the energy density of LIBs because of its high theoretical specific capacity (3820 mAh/g) which is more than tenfold that of graphite (372 mAh/g). However; the growth of Li dendrite upon Li plating/stripping causes safety issue and short cycle life of Li metal batteries because of challenges embedded from lack of uniform deposition of lithium and unstable SEI layer formation.6-8 So scholars are in searching of other alternative ideas to boost the energy density, safety and minimize cost. One of the new ideas is the use of alternative design such as anode-free system. In principle, if the Li+ ions extracted from the positive electrode can be reversibly plated onto and stripped back from the Cu current collector as Li metal, a rechargeable Li metal battery with Cu/separator/cathode/Al configuration can be assembled. So initially the battery has no active anode thus can be termed as an “anodefree rechargeable lithium metal battery (AFLMB). The absence of Li+ host anode material reduces the cell weight and space required for the anode as a result; energy density of the AFLMB will therefore be significantly larger compared to the conventional LIBs.8-9 The cost, time and energy associated with anode production (including electrode slurry preparation, slurry coating, and drying) are also saved. In addition to the above advantages, the direct use of lithium metal as an anode is also highly problematic because of its high reactivity and thus typically very low plating/stripping efficiency usually less than 80% in most non-aqueous electrolytes.10-14 The formation of lithium metal anode after it starts charging can also be taken as the advantage of this system. Nevertheless, realizing this anode-free battery in practical applications is still handicapped. The major reason for the failure of this system as reported in the above literatures is depletion of the active Li and consumption of electrolyte solvents during repair of the SEI (unstable SEI); as a result low Coulombic efficiency and retention capacity.10, 15-16 Therefore, the key to realize lithium metal battery and also the Li plating/stripping process in the anode-free system is designing better functional electrolyte.17-18 When an 3 ACS Paragon Plus Environment


electrolyte is designed, the Li salt, solvent and concentration should be taken under consideration. Solid polymer or inorganic electrolytes and hybrid ionic liquid electrolytes19 were potentially chosen as they can suppress the growth of the lithium dendrites.20-22 Unfortunately, the kinetic properties of solid-state electrolytes are restricted because of their low conductivity and high interfacial resistance at ambient temperature.23-24 So the traditional LIB with graphite anode uses 1 M or 1.2 M LiPF6 in EC-based liquid electrolytes as optimized electrolyte due to their high conductivity, better wettability, good Al passivation at higher potential, wide electrochemical stability windows, and good compatibility with conventional intercalation electrodes.25 However, the conventional carbonate based electrolyte systems have inferior compatibility with Li metal, inducing Li dendrite growth and low cycling coulombic efficiency. They are also chemically unstable and facilitate transition metal dissolution of the electrode material,26 therefore, unsuitable for anode-free battery.27 For example, Cu‖LiFePO4 and Cu‖LiNi1/3Mn1/3Co1/3O2 (NMC) cell were cycled in an electrolyte consists of 1M LiPF6 and 1.2 M LiPF6 in carbonate based solvents10, 28, their report indicates that from the total amount of plated lithium (first charge), the loss was as high as 75% and 77% respectively in the respective discharge. Therefore, according to their report these carbonate based electrolytes are not suitable to be used in an AFLMB. Nowadays battery researchers are interested on concentrated electrolyte solutions because of their peculiar characteristics, which include unusually improved oxidative stability on both the cathode and anode side, absence of free solvent, presence of more anion content and the improved availability of Li+ ions. These unique characteristics are associated to solution structure change between the low and high concentrated electrolyte solutions.29-32 In principle the competitive Li+-solvent and Li+-anion interactions lead to the formation of various Li salt solvate complexes.33 Increasing salt concentration results in increasing ionic association and decreasing free (un-coordinating) solvent molecules.33 But in concentrated solutions the Li saltsolvent mixture, the anions and solvents competitively coordinate to Li+, and hence the balance between Li+-anion and Li+-solvent interactions dominates the solution structure. Therefore, the viscosity together with ionic neutrality due to the charge balance decreases its ionic conductivity, so that diluents are important to decrease the viscosity and unbalance the charge neutrality. On the other hand most of the concentrated electrolytes reported are with ether-based solvents and LiFSI conductive salt due to the high solubility of salts in ether-based solvents.34-37 However, the poor anti-oxidation capability of ether-based solvents hindered practical application, 4 ACS Paragon Plus Environment

Page 4 of 22

Page 5 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


especially with high potential cathode materials (> 4 V). Recently, Zhang et al. proposed highly concentrated ether-based electrolyte (4 M LiFSI/DME) which shows high anodic stability.25 However, such highly concentrated Li-salt is not viable from cost analysis point of view, because the salt is very costly. Lithium hexaflourophosphate (LiPF6) has some drawbacks like sensitivity toward moisture. If one considers the overall properties required by electrolyte-conducting salt including cost, LiPF6 in EC based organic solvents is still the best option among all other salts for LIBs.36 However, concentrated electrolyte solutions suffer from high viscosity and low conductivity as mentioned above, so their overall performance will be affected. Under such circumstances, diluents are important to reduce the viscosity and to enhance the ionic conductivity without losing the characteristics of the concentrated electrolyte solutions such as high oxidative stability and the solvation structure of the solution.38 Here we report a locally concentrated LiPF6 in EC/DEC (1:1 v/v ratio) diluted by 50% FEC to enhance the electrochemical performance of the AFLMB using NMC as cathode and bare Cu foil as anode current collector. The effect of concentration of LiPF6 in EC/DEC and the use FEC as diluent was systematically investigated by optimizing electrochemical performance tests. Then the optimized electrolyte with relative better performance was subjected to further characterization like X-ray photoelectron spectroscopy (XPS), electrochemical impedance spectroscopy (EIS), scanning electron microscopy (SEM) and Raman spectroscopy. 2. EXPERIMENTAL SECTION 2.1. Preparation of Electrolytes Commercial 1 M LiPF6 electrolyte in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1 v/v%, 99% battery grade from sigma Aldrich) as received, was used as a reference electrolyte and then after termed as E1. Then electrolyte solutions with different concentrations (1.5M (E1.5), 2M (E2), 3M (E3) and 4M (E4)) were prepared by dissolving the calculated amount of powder LiPF6 (99% battery grade from sigma Aldrich) in EC/DEC (1:1 ratio) and it was stirred for 24 h until completely dissolved. The salt was dried at 80 oC in an evacuated oven in glove box and the solvents were dried with activated molecular sieve before used. The concentrated solutions then were diluted by using appropriate amount of fluoroethylene carbonate (FEC). All the solutions were prepared in a glove box filled with argon, where both oxygen and moisture contents were less than 0.1 ppm 2.2. Electrochemical Tests 5 ACS Paragon Plus Environment


Li‖Cu cell and AFLMB with commercial LiNi1/3Mn1/3Co1/3O2 (NMC) coated on Al foil (from Amita technologies) as cathode and Cu-foil as anode current collector (Cu‖NMC) were assembled to test the cells. The electrochemical tests were measured using 2032 type coin cells and assembled in a glove box filled with argon gas (Unilab, Mbraun) where both oxygen and moisture content was less than 0.1ppm. To measure the degree of polarization and stability of Li plating/stripping behavior in anode-free Cu||NMC full cell and Li‖Cu cells with the commercial electrolyte E1 (LiPF6 in EC/DEC 1:1 v/v ratio) as control and E2 (LiPF6 (EC/DEC 1:1 v/v ratio, diluted by 50% FEC), as target (optimized) electrolyte, were assembled. The amount Li to be plated on to the Cu-surface was controlled by restricting the time, based on the Cu||NMC anodefree full cell charge capacity (8:18 h) and the stripping was voltage controlled (0.1 V) and constant current density at 0.2 mA/cm2. The charge-discharge performance tests for the anodefree full cells were performed galvanostatically using a computer controlled 40-Channel battery tester (Arbin, BT-2000) in a potential window between 2.5 V and 4.3 V at 0.2 mA/cm2 current density. The charge /discharge protocol was the common CC-CV mood. To study the effect of upper limit cut-off voltage, it was also tested from 2.5 V - 4.1 V and 2.5 V - 4.5 V. The electrochemical impedance spectroscopy (EIS) was performed at open circuit voltage with in the frequency range between 0.01 Hz and 100 kHz with an applied voltage of 10 mV using impedance analyzer (BioLogic SAS) at room temperature before and after different cycle for the anode free configuration. A symmetric cell using two stainless steel electrodes was assembled to measure the EIS spectra so that the conductivity of each solution can be obtained. 2.3. Characterization The samples (electrode) for surface analyses were collected by disassembling the cells inside the glove box after desired number of cycles. The disassembled electrodes were washed with DEC until all the residuals were removed and then dried well for analysis. The surface analysis of the Cu substrates after 5 cycles (fully discharged) for the cell cycled both using reference electrolyte (E1) and the locally concentrated electrolyte (E2) and the bare 2M LiPF6 in EC:DEC 1:1 v/v ratio (E2B) were done by scanning electron microscopy (SEM, JEOL JSM6500F), X-ray photoelectron spectroscopy (XPS). The XPS was performed at NSRRC using 0.05 eV step and 20 eV pass energy and it was calibrated by gold signal at binding energy of 84.0 eV. A Uni-RAM Raman spectrometer furnished with Electrically mechanized stage sample holder and thermoelectrically cooled charge coupled device (CCD) working at -50 oC and excitation source at 785 nm was used to collect spectra of electrolytes with different concentrations. The 6 ACS Paragon Plus Environment

Page 6 of 22

Page 7 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


working distance of the microscope objective was 50 X in backscattering geometry. The viscosity of different electrolytes was measured using a rolling ball Lovis 2000M/ME micro viscometer at room temperature (25 oC) at Lovis angle of 80o. Cyclic voltammetry (CV) was also performed using two electrode system using VMP3 multi-channel (BioLogic Science Instruments work station) taking the real cathode (NMC) and Li anode in the system. 3. RESULTS AND DISCUSSION 3.1. Li ‖ Cu Cell Performance Li‖Cu half-cell test was performed to understand polarization, cycling performance and cyclic coulombic efficiency. From the cyclic coulombic efficiency (ratio of the amount of Li stripped to the plated on the bare Cu substrate), the rate of Li consumption during the Li plating/stripping process and the degree of electrolyte decomposition can be clearly understood. The coulombic efficiency (CE) of Li‖Cu cells at 0.2 mA/cm2 for both E1 and E2 is shown at Figure 1e. Since the opposite side has surplus Li (since Li foil is used anode), the CE essentially indicates the Li loss on the Cu-current collector due to decomposition of electrolytes. The initial Coulombic efficiency of both assembled cells using E1 and E2 electrolyte was ~94% at 0.2 mA/cm2. This indicated that irrespective of the electrolyte type, almost there was no difference in the first cycle. However, the cell with E1 electrolyte, the coulombic efficiency dropped quickly and fluctuation was observed starting after few cycles. Similarly, high degree of polarization which is 76 mV at 50th cycle as seen from Figure 1c, d and continuous fading in capacity shown in Figure 1f. This showed that, after few cycles of platting/stripping, thick deposition of dead Li and hence fast Li dendrite growth on cupper surface, which resulted in a highly resistive layer SEI intertwined with the dead Li metal was formed when E1 was used.39 This increases the ion and electron diffusion resistance within the anode matrix as a result the polarization was escalated up quickly. This very low and less stable CE, fast capacity fading, large polarization and haphazard voltage fluctuations confirms that conventional carbonate electrolyte like the commercial 1M LiPF6 EC/DEC 1:1 v/v ratio (E1) are incompatible with Li-metal anode and also AFLMB as mentioned above.10 This could be probably because of the side reactions at the anode and continuous degradation of the electrolytes and thick deposition of dead Li that significantly affects the cyclic coulombic efficiency and overall cyclic performance of the battery.12 The remarkably reasonable average coulombic efficiency (~98 %) after cycling for more than 1066 h as shown in Figure 1e and low voltage hysteresis, only ~30 mV at the 50th cycle, as 7 ACS Paragon Plus Environment


seen in Figure 1c, d with stable capacity as shown in Figure 1f, using E2 indicates that, the electrolyte was robust and stable at the negative electrode. The polarization in the first cycle was very high while it decreased after the second cycle which also indicates that formation of stable and conductive SEI derived from the locally electrolyte E2 electrolyte. The robust SEI formed, may probably be because of the synergetic effect of the FEC used as diluent and the salt concentration of the locally concentrated electrolyte in comparison to E1, which attributed to the uniform and dendrite free Li-metal deposition on the current collector surface.

Figure 1. Electrochemical performance of Li‖Cu cells at 0.2 mA/cm2 current density with E1 and E2. (a) voltage profile using E1; (b) voltage profile using E2; (c) comparison of polarization

8 ACS Paragon Plus Environment

Page 8 of 22

Page 9 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


profile of plating/stripping process using E1 and E2; (d) magnified view of figure (c); (e) comparison of coulombic efficiency of E1 and E2; (f) comparison of capacity of E1 and E2. 3.2. Cu‖ LiNi1/3Mn1/3Co1/3O2 (NMC) Anode-Free Cell Performance Anode-free Cu||NMC coin cells were assembled with bare Cu as anode current collector and NMC as cathode with different concentrations of LiPF6 in EC/DEC (1:1 v/v ratio) before and after diluted with FEC electrolytes. The electrolytes with different concentrations like bare LiPF6 in EC: DEC 1:1 v/v%, (1.5 M (E1.5B), 2 M (E2B), 3 M (E3B) and (1.5 M (E1.5), 2 M (E2), 3 M (E3), and 4 M(E5)) each diluted by FEC were optimized using this anode free system as shown in Figure S2 and Figure S3, respectively. The commercial carbonate based electrolyte E1 and the locally concentrated electrolyte E2 (optimized) were used to evaluate the cell performance. Even if the literature10 showed that about 77% lost after the first charge for the carbonate solvents (EC/EMC (3:7 wt%)), our performance test showed that, regardless of the electrolytes type, in the first cycle the loss was almost only ⁓14% in the AFLMB. However, after 15th cycles, the cell cycled using E1 lost its discharge capacity more than 60% of its first discharge capacity as shown in Figure 2c. The Coulombic efficiency of the AFLMB using E1 electrolyte dropped quickly (

Locally Concentrated LiPF6 in a Carbonate-Based ... - ACS Publications

Recommend Documents