Subscriber access provided by BOSTON UNIV
Energy, Environmental, and Catalysis Applications
Iron controllable lithium into lithiotropic carbon fiber fabric: a novel Li metal anode with improved cyclability and dendrite suppression Xi Chen, Yingying Lv, Mingwei Shang, and Junjie Niu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b05364 • Publication Date (Web): 29 May 2019 Downloaded from http://pubs.acs.org on May 30, 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 18 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
Iron controllable lithium into lithiotropic carbon fiber fabric: a novel Li metal anode with improved cyclability and dendrite suppression Xi Chen, Yingying Lv, Mingwei Shang, Junjie Niu* Department of Materials Science and Engineering, CEAS, University of Wisconsin-Milwaukee Milwaukee, WI 53211, USA. *Email:
[email protected] Abstract Lithium metal as anode in lithium ion batteries (LIBs) is attracting more attentions due to the high gravimetric/volumetric energy density and low electrochemical potential. However, the irreversible Li plating/striping can reduce the cycling capability and very possibly introduce dendrite growth, thus leading to a series of issues such as infinite volume change, low Coulombic efficiency (CE), and uncontrollable solid electrolyte interphase (SEI). Here we report a novel, single-side Li-infused carbon fiber fabric (LiCFF) with a controllable, minimized Li loading, which shows a highly reversible plating/stripping with an extremely low overpotential of less than 30 mV (Li foil: >1.0 V over 50 cycles) upon >3000 cycles (6000 and 2000 hours) at 1 and 3 mA/cm2 in symmetric cells, respectively. With a high areal capacity up to 10 mAh/cm2 and a high current density of 10 mA/cm2, the cell still shows a minimum overpotential of 150-175 mV after 250 cycles (500 hours). Full cell batteries using the LiCFF as ‘all-in-one’ anode without additional slurry-making process and nickel-manganese-cobalt oxide (NMC) as cathode exhibit an improved capacity retention when compared with Li foil: 32% at 0.5 C and 119% at 1.0 C capacity improved after 100 cycles. In parallel, the mossy/dendritic Li on the LiCFF was largely suppressed, which was confirmed using in-situ observations of Li plating/striping in a capillary cell. The excellent electronic conductivity of the carbon fabric leads to small contact/transfer resistances of 3.4/3.8 Ω (Li foil: 4.1/44.4 Ω), enabling a drastically lowered energy barrier for Li nucleation/growth. Thus a uniform current distribution results in forming a homogeneous Li layer instead of forming dendrite. The current LiCFF as anode with controllable Li (n/p ratio), improved cycling stability, mitigated dendrite formation, and flexibility displays promising applications in versatile Li-metal batteries such as Li-NMC, Li-S and Li-O2. Keywords: Li metal; Carbon fiber fabric; Low overpotential; Dendrite; Lithium ion battery
1
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
Introduction The rapidly growing demand in large energy-density batteries, e.g., >500 Wh/Kg, for the emerging electrical vehicles (EVs) and portable electronic devices necessitates scientists to explore new electrode materials that can deliver higher capacity.1 Li metal, having a theoretical capacity of 3860 mAh/g and a low electrochemical potential of -3.040 V vs standard hydrogen electrode, is considered as one of the most promising candidates as anode in next-generation LIBs.2 However, the dendrite forming, low CE and unstable SEI pose rather challenges in employing Li metal in LIBs.3 A passivated SEI can stabilize the interface during the cycling. However, the mossy Li can introduce an infinite volume change and penetrate through the SEI, resulting in the generation of cracks/holes. The repeated forming and disappearing of the SEI lead to an inhomogeneous Li plating/stripping and a loss of the isolated Li. In parallel, the consumption of plenty of Li+ on the interface makes the local Li+ concentration close to zero, thus the dendrite starts to nucleate and grow until an internal short circuit happens.4-6 To date several strategies on stabilizing Li metal electrode have been reported: 1). Add electrolyte additives and increase lithium salt concentrations.7-13 2). Use solid electrolytes with high ionic conductivity to mechanically inhibit the dendrite growth.14, 15 3). Develop electrochemically stable artificial SEI.16, 17 4). Apply a stiffness-enhanced membrane with controllable ion transport and dendrite growth orientation.18, 19 5). Pre-store Li in stable hosts.20-26 6). Form a uniform 3D configuration.27, 28 A rational design of Li metal anode should contain measurable pre-stored lithium, which is paired to the cathode with NMCs or Li-poor materials such as S29, 30 and O231. Recently the molten lithium was loaded onto lithiophilic host materials such as carbon fiber, graphene, polymer, metal foam and carbonized wood that showed better mechanical properties and electrochemical stability.20-22, 32, 33 A 3D framework matrix without Li also showed a reduced effective current density, minor volume change, and stable SEI.27, 34 Herein, we develop a mass controllable, single-side Li-infused carbon fiber fabric, which displayed an excellent cyclability with a low overpotential of less than 30 mV upon >3000 cycles at high current densities. The excellent conductivity (9.09x104 S/m) and confinement of the carbon fabric texture result in a uniform current distribution, forming a homogeneous, stable Li layer instead of dendrites. The full cell batteries versus NMC and sulfur demonstrated clearly improved capacity retention in contrast to the batteries with Li foil as anode. Results and Discussion The crosswise-orientated CFF is composed of single carbon fibers (~8 μm in diameter) with 10 μm interval between fibers and 100 μm empty space in joint junctions (Figures 1c and S1a). This weaving texture provides sufficient space to store and accommodate lithium as well as a fast pathway for electrolyte diffusion. Only a small peak of C (002) in the X-ray diffraction (XRD) pattern (Figure S1b) indicates the CFF has a structure with weak crystallinity, which was further confirmed by the high intensity of D band at 1330 cm-1 in the Raman spectrum (Figure S1c). A high carbon purity was confirmed using thermogravimetric analysis (TGA, Figure S1e). Typically lithiophilic hosts such as reduced graphene oxides,21 metal oxide on carbon/polymer framework22 and nanoparticles seeded fiber matrix23 displayed rapid infusion via strong capillary force. The infusion process starts from one side of the lithiophilic hosts, e.g., cross section. Then the whole pellet is quickly filled with Li, which makes the
2
ACS Paragon Plus Environment
Page 2 of 18
Page 3 of 18 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
controllability of quantity not possible. In order to load the molten Li with a controllable manner, a moderate infusing process is needed. The micrometer size of the CFF reduces the capillary force in contrast to nano-structures, which leads to a ‘lithiotropic’ (between lithiphilic and lithiophobic) surface. In addition, the Fourier-transform infrared spectrum (FTIR) curve shows few peaks from the functional groups such as –OH and-COOH those exist in most of other crystal carbon materials. The ironing process slowly starts from the large 2D surface (instead of the small side or cross-section) of the lithiotropic LiCFF, leading to a limited/controllable infusion onto single side. As a result, a homogeneous Li layer with controllable mass is achieved via a slow Li infusion process. In typical experiments, we used a clothes-ironing process to transfer the molten Li to CFF, as shown in Figure 1a and Movie S1. The molten Li (‘steam’) was attached to a flat stainless steel to iron the top surface of the CFF back and forth. During this ironing process, certain amount of liquid Li was well penetrated into the CFF matrix to form a uniform, thin layer on each fiber. In addition, only one side of the CFF was filled with Li, resulting in forming Li-rich and Li-poor sides (Figure 1a-b). The Li-poor side can serve as a buffering area to further decrease the overall Li concentration and contribute more spaces to moderate the volume expansion (Figure 1c-d). As observed from Figure 1e, the Li was tightly covered on the outer surface of fibers, which ensures a high conductive interface for Li+/e- exchange. In general, lithium prefers to cover the surface of the carbon fiber due to its high electronic conductivity. The X-ray photoelectron spectroscopy (XPS) of LiCFF demonstrates the formation of LiCx after ironing process (Figure S2). Compared to conventional C-C bonding at 284.8 eV, the C 1s peak 282.4 eV implies the existence of Li-C bonding (Figure S2a).35 The Li 1s peak at 52.7eV also confirms this Li-C bonding (Figure S2b), which is considered as Kirkendall-type diffusion induced formation of LiCx at the interface.36, 37 According to the existing interface in Figure 1b, the entire thickness of LiCFF is about 300 μm and the Li-rich side is about 120 μm. A variety of the thickness of Li-rich side from 30 to 180 μm was obtained via ironing from 15 to 120 seconds (Figure S3). The n/p ratio from 0.5 to 16 (vs NMC 622 as positive electrode) was obtained by adjusting the ironing time from 15 to 120 seconds (Figure S4 and Table S1), which indicates a tunable mass loading of Li. Morphology evolution of the LiCFF electrode was investigated via a series of lithium plating and stripping cycles at varying current densities in symmetric cells. It is clearly seen that the majority of the Li was coated on the outer surface of fibers to form a thin layer with inter-spaces in between upon plating (Figure 2a-c, g-i). The CFF was recovered with more spaces exposed after stripping (Figure 2d-f, jl). The space between fibers shows sufficient accommodation to Li during charging/discharging. In other words, the volume expansion/shrinkage from Li nucleation/growth even with the infinite change from dendrites can be greatly mitigated through the confinement of this weaving texture. As observed from the cross-sections (Figure 2a-c), the concentration of coated lithium is increased in proportion to the applied current density. This change becomes more apparent from the top-view of the Li-poor surface (Figure 2g-i). When the process was reversed, most of Li was stripped at low current densities of 1 and 3 mA/cm2, leaving the empty CFF, while a small amount of Li was remained at 5 mA/cm2, as shown in Figure 2d-f, j-l. This excellent reversibility was further confirmed after a long cycling (Figure S5), which enables a stable Li plating/stripping performance with a minimum Li loading in batteries. In parallel, no mossy/dendritic Li was found even at 5 mA/cm2 after 100 cycles (Figure S5c,i), indicating a good suppression to the dendrite.
3
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
In order to further study the dendrite suppression, a capillary cell was designed to in situ observe the mossy/dendritic Li nucleation/growth (Figure 3a). Under a plating current of 4.77 mA/cm2, the visible nucleation of mossy Li on Li granules appeared only after 4 min (Figure 3b). This mossy Li continued to grow up to ~2 mm in 14 min. As a comparison, the LiCFF started to show the growth of mossy Li on the surface after 28 min (Figure 3c). A small mossy Li with a length of only ~0.5 mm was observed after 42 min. The real-time dynamic growth of mossy Li on Li granules and LiCFF was also recorded in Movies S23). In general, the excellent electric conductivity of CFF provides a fast channel for e- transport across the whole cross-linked matrix, resulting in a homogenous current distribution with a decreased local current density. Meanwhile, plenty of spaces amongst the fabric suffice the interface between the electrolyte and LiCFF and thus have a rapid diffusion of Li+ to reach the electrode. Hereby a ‘constant’ Li+ concentration near to the interface is maintained which avoids the uneven current density in local areas. The stable, low current across the fiber results in a low energy barrier for Li nucleation, leading to an extended Li dendrite nucleation time and a suppressed Li dendrite growth rate according to the Sand’s law. In other words, the enhanced Li+/e- conductivity of CFF prevents the local depletion of Li+ to reach the ‘sand time’ to suppress the dendrite nucleation, enabling a uniform, reliable formation of Li layer2, 4. In addition, the double-side configuration also helps to further mitigate the mossy/dendritic Li. We found that the Li plating started from underneath the Li-poor side (Movie S3). That means the surface (Li-poor) that faces to the electrolyte can accept newly cycled Li from the base matrix (Li-rich) that contacts to the current collector, to further dilute the overall Li concentration and reduce the opportunity of forming dendrites. In case of an extremely high current is long applied, the mossy Li will first form on the Li-rich side and then pass through the Li-poor side to reach the interface (Figure 3c and Movie S3). This design greatly reduces the safety risk from the large volume expansion of mossy/dendritic Li. Electrochemical performance of the LiCFF electrode was checked using symmetric cells (Figure S6). The voltage variation of Li foil and LiCFF cells was measured upon charging/discharging at different current densities. As shown in Figure 4a, an ultra-low overpotential of 3000 cycles (6000 hours). The high reversibility of the voltage fluctuation upon plating/stripping (Figure 4a inset) demonstrates a uniform, reliable nucleation/growth of Li. However, the Li foil exhibited a high initial potential of ± 100 mV and it was increased to over 1.0 V only after ~130 cycles (black). It clearly elucidates that Li foil has a big nucleation barrier for Li, which introduces an uneven growth of Li and in turns to further increase the overpotential to nucleate/grow new Li. If the cell was cycled under larger current densities, the LiCFF electrode still presented a low, stable overpotential of 25-30 mV at 3 mA/cm2 upon >3000 cycles (2000 hours) (Figure 4b, red). The cells with LiCFF were also tested under the high current densities of 10 and 20 mA/cm2. A small offset of
