Heterolayered, One-Dimensional Nanobuilding Block Mat Batteries

Sep 16, 2014 - The rapidly approaching smart/wearable energy era necessitates advanced rechargeable power sources with reliable electrochemical proper...
1 downloads 8 Views 818KB Size
Download PDF
Letter pubs.acs.org/NanoLett

Heterolayered, One-Dimensional Nanobuilding Block Mat Batteries Keun-Ho Choi,† Sung-Ju Cho,† Sang-Jin Chun,‡ Jong Tae Yoo,§ Chang Kee Lee,§ Woong Kim,⊥ Qinglin Wu,∥ Sang-Bum Park,‡ Don-Ha Choi,‡ Sun-Young Lee,*,‡ and Sang-Young Lee*,† †

Department of Energy Engineering, School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, 689-798, Korea ‡ Department of Forest Products, Korea Forest Research Institute, Seoul, 130-712, Korea § Korea Packaging Center, Korea Institute of Industrial Technology, Bucheon, Gyeonggi-do, 421-742, Korea ⊥ Department of Materials Science and Engineering, Korea University, Seoul, 136-713, Korea ∥ School of Renewable Natural Resources, Louisianan State University Agricultural Center, Baton Rouge, Louisiana 70803, United States S Supporting Information *

ABSTRACT: The rapidly approaching smart/wearable energy era necessitates advanced rechargeable power sources with reliable electrochemical properties and versatile form factors. Here, as a unique and promising energy storage system to address this issue, we demonstrate a new class of heterolayered, one-dimensional (1D) nanobuilding block mat (h-nanomat) battery based on unitized separator/electrode assembly (SEA) architecture. The unitized SEAs consist of wood cellulose nanofibril (CNF) separator membranes and metallic current collector-/polymeric binder-free electrodes comprising solely single-walled carbon nanotube (SWNT)-netted electrode active materials (LiFePO4 (cathode) and Li4Ti5O12 (anode) powders are chosen as model systems to explore the proof of concept for h-nanomat batteries). The nanoporous CNF separator plays a critical role in securing the tightly interlocked electrode−separator interface. The SWNTs in the SEAs exhibit multifunctional roles as electron conductive additives, binders, current collectors and also non-Faradaic active materials. This structural/physicochemical uniqueness of the SEAs allows significant improvements in the mass loading of electrode active materials, electron transport pathways, electrolyte accessibility and misalignment-proof of separator/electrode interface. As a result, the h-nanomat batteries, which are easily fabricated by stacking anode SEA and cathode SEA, provide unprecedented advances in the electrochemical performance, shape flexibility and safety tolerance far beyond those achievable with conventional battery technologies. We anticipate that the h-nanomat batteries will open 1D nanobuilding block-driven new architectural design/opportunity for development of next-generation energy storage systems. KEYWORDS: Nanomat batteries, heterolayer, one-dimensional nanobuilding block, separator/electrode assembly, cellulose nanofibrils, single-walled carbon nanotubes

R

electrode sheets and separator membranes), which are believed to pose a major obstacle impeding dramatic progress of the rechargeable battery systems. Here, as a unique power source to address this challenging issue, we demonstrate a new class of heterolayered, onedimensional (1D) nanobuilding block mat (referred to as “hnanomat”) batteries based on unitized separator/electrode assembly (SEA) architecture. The unitized SEA consists of two different parts: (1) current collector-/binder-free electrode comprising solely single-walled carbon nanotube (SWNT)netted electrode active materials and (2) wood cellulose nanofibril (CNF) separator membranes. The SWNTs in the

echargeable energy storage systems, represented by current state-of-the-art lithium-ion batteries, are incessantly in pursuit of high-energy/high-power density and also aesthetic versatility, in order to extend their applications from portable electronic appliances to rapidly growing industrial fields such as electric vehicles (EVs), grid scale energy storage systems and wearable electronics.1−3 To date, most research and development efforts for advanced batteries have been devoted to the design and synthesis of new electrochemically active materials,4−11 which include high-capacity/high-voltage/ high-rate electrode active materials and high-conductive/ functional electrolytes. Although some meaningful achievements have been reported, most of them still rely on traditional electrode architecture (i.e., a mixture of electrode active materials, polymeric binders and electroconductive additives on current collectors) and also battery configuration (i.e., constructed by winding or stacking two-dimensional (2D) © 2014 American Chemical Society

Received: June 26, 2014 Revised: September 10, 2014 Published: September 16, 2014 5677

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

Letter

Figure 1. Fabrication of 1D nanobuilding block (CNF/SWNT)-based SEAs. (a) A schematic representation of the overall fabrication procedure for unitized SEA (=CNF separator/electrode comprising SWNT-netted electrode active materials). (b) A photograph (left side) of CNF suspension and a TEM image (right side) showing CNFs with nanoscale diameter/length up to a few micrometers. (c) SEM images demonstrating the dispersion state of SWNTs and LiFePO4 (or Li4Ti5O12) powders in the electrode slurries, where the SEM characterization was conducted after eliminating water solvent. (d) Photographs of self-standing, flexible LiFePO4 (cathode) and Li4Ti5O12 (anode) SEAs. (e) Photographs of SEAs with two different faces (upper image, CNF separator; bottom image, LiFePO4 cathode). (f) A conceptual illustration depicting the 1D nanobuilding block (CNF/SWNT)-enabled structural uniqueness of SEAs.

collectors,20−22 mechanical buffers for metallic anode materials23,24 and porous separator membranes.25−28 The h-nanomat batteries presented herein, owing to the structural/physicochemical uniqueness of the SEAs, are expected to bring unprecedented improvements in the electrochemical performance, shape flexibility and internal short-circuit tolerance, which are difficult to reach with conventional battery materials and configurations. First, the removal of current collector and binder allows larger amount of electrode active materials to be loaded in electrodes, thereby remarkably increasing gravimetric capacity of cells. Second, the 1D nanobuilding block (CNF/SWNT)-mediated facilitation of electron/ion transport (i.e., highly networked electron pathways and easy accessibility of liquid electrolytes) boosts electrochemical reaction kinetics of cells even at harsh operating conditions such as fast charge/discharge rates. Third, the unitized configuration of SEAs provides significant

SEAs exhibit multifunctional roles as electron conductive additives, binders, and also current collectors. Intriguingly, the unusual SEA structure allows the SWNTs to activate their nonFaradaic reaction, thus adding extra capacity to intrinsic Faradaic capacities of electrode active materials. Meanwhile, the CNF separator with well-tuned nanoporous structure plays an important role in securing the tightly interlocked electrode− separator interface. The CNF, which consists of close-packed polysaccharide chains, is an eco-friendly, naturally abundant and sustainable mesoscopic material with an anisotropic dimension of micrometer length and nanometer diameter.12−15 These structural features enable the CNF to act as a promising alternative material for battery applications. The representative examples include electrode binders replacing synthetic polymers such as polyvinylidene fluoride (PVdF),16−19 threedimensional (3D) porous layers decorated with electroconductive substances to replace conventional metallic current 5678

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

Letter

Figure 2. Structural/physicochemical characterization of SEAs. (a) SEM image (surface view) depicting the freeze-drying-induced porous structure formation of CNF separator (the inset shows the CNF paper with less-developed porous structure). (b) SEM image (cross-sectional view) of the cathode SEA, where the tightly interlocked separator−electrode interface was established. (c) SEM image (cross-sectional view) of anode SEA. (d) AFM image (surface view) of the cathode SEA, where LiFePO4 powders are reticulated with SWNTs. (e) SEM image (surface view) of LiFePO4 cathode SEA. (f) SEM image (surface view) of Li4Ti5O12 anode SEA. (g) Taping-out test of the cathode SEA using commercial 3M scotch tape (the inset shows the result of a control sample employing a typical filter paper with thick (micrometer scale) fibers). (h) Comparison of electronic conductivity between conventional electrodes and SEAs. (i) Comparison of porosity among conventional electrodes, polyethylene separator and SEAs. (j) Photographs demonstrating physical flexibility and dimensional integrity of SEAs.

SWNTs, sodium dodecylbenzenesulfonate (SDBS)30 was added as a surfactant (concentration = 1.0 wt % in water). The characterization of particle dispersion in the electrode slurries demonstrates that the electrode active materials were well mixed with the SWNTs (Figure 1c), although this result does not directly reflect the dispersion state of the electrode slurries because the scanning electron microscopy (SEM) analysis was conducted after eliminating water solvent. Finally, after vacuum-drying (at 100 °C for 12 h) of the electrode slurry on the CNF paper, the self-standing and highly flexible cathode (or anode) SEAs were obtained (Figure 1d), wherein LiFePO4 (cathode, average particle size ∼500 nm) and Li4Ti5O12 (anode, average particle size ∼300 nm) powders were chosen as model systems to explore the proof of concept for the h-nanomat batteries. Notably, a series of SEAs with diameters ranging from 4 to 30 cm were easily fabricated (Supporting Information Figure S1), demonstrating the

benefits to cell assembly simplification, mechanical/thermal tolerance and misalignment-proof of electrode/separator interface (preventing the occurrence of internal short-circuit). The SEA, a core unit of the h-nanomat battery, was fabricated via a simple vacuum-assisted infiltration process, which is analogous to a traditional paper-making method.16,29 First, the CNF suspension (water as solvent) was poured onto a filter paper positioned inside a Porcelain Buchner funnel and then subjected to vacuum infiltration, leading to the formation of CNF paper (Figure 1a). The CNFs, which are produced from the repeated high-pressure homogenization of wood cellulose powders,27,28 were uniformly dispersed in the suspension and had nanoscale diameter/length up to a few micrometers (Figure 1b). Subsequently, on the surface of formed CNF paper, an electrode slurry consisting of electrode active materials and SWNTs in water was introduced using the same infiltration method. To achieve a good dispersion state of 5679

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

Letter

batteries, significant effort has been made to fabricate metallic current collector-free or binder-free electrodes.20,21,33−36 In the SEAs presented herein, the 1D nanobuilding block (CNF/ SWNT)-driven strong interface between separator and electrode allows the SWNTs (reticulating electrode active materials) to act as current collectors and also binders besides their innate role as electroconductive additives, thus enabling the one-pot removal of both current collectors and binders in electrodes. It should be noted that this unusual multifunctionality of SWNTs in the SEAs was achieved only at the low SWNT content (i.e., LiFePO4/SWNT or Li4Ti5O12/ SWNT = 85/15 (w/w)), which underscores that the SWNTs of the SEAs are different from the previously reported CNTbased electroconductive support layers such as sponge-like carbon sheets37 and bucky papers.38 The electronic resistance of SEAs is dependent on the electrode slurry volume used in the vacuum-assisted infiltration process. The larger volume of electrode slurry could promote the close packing and mutual contact of SWNTs and electrode active powder, thus lowering electronic resistance of electrodes in the SEAs (Supporting Information Figure S6). Another important factor to govern the electronic resistance of SEAs is the SWNT content. At the optimized filtration volume (i.e., 50 mL), the electronic conductivity of cathode SEAs tends to increase with the SWNT content. Notably, the percolation threshold of electronic conduction was found at the SWNT content of 15 wt % (Supporting Information Figure S7). The cathode and anode SEAs, respectively, presented the electronic conductivities of 980 and 960 S m−1 (Figure 2h), where the 15 wt % SWNTs were incorporated in both SEAs. These values are higher than those of conventional LiFePO4 (∼56 S m−1) and Li4Ti5O12 (∼11 S m−1) electrodes (i.e., LiFePO4/PVdF binder/carbon black conductive additives = 80/10/10 (w/w/ w) on an aluminum current collector, Li4Ti5O12/PVdF/carbon black = 88/10/2 (w/w/w) on an aluminum current collector). This higher electronic conductivity of the SEAs is attributed to the well-percolated 3D electronic networks of SWNTs reticulating electrode active materials and also the removal of polymeric binders posing as a barrier to impede electronic/ ionic transport (Figure 2d−f). This result demonstrates the structural superiority of SEAs in tuning the electronic resistance of electrodes. To facilitate electrochemical reaction of a cell, fast/uniform ion migration, along with the aforementioned electron transport, should be secured.39,40 Figure 2i shows that the SEAs present a higher porosity than conventional ones. This improvement in the porosity of the SEAs is ascribed to the welldeveloped interstitial voids formed between the 1D nanobuilding blocks (CNF/SWNT). The measurement of electrolyte-immersion height (Supporting Information Figure S8) exhibits that the SEAs, owing to the presence of polar CNFs and well-developed porous structure facilitating capillary intrusion,27,28 provide a better electrolyte wettability than conventional cell components. This good electrolyte wettability, in combination with the highly porous structure serving as ionic channels, is expected to allow a facile accessibility of liquid electrolyte in the SEAs, thereby enhancing electrochemical kinetics of h-nanomat batteries. Figure 2j shows that the cathode SEA was not mechanically broken even after being fully folded. Also, it was wound several times along a wooden rod (diameter = 5 mm). More notably, it could be knotted without any mechanical rupture. Furthermore, the SEA maintained its physical flexibility and dimensional

excellence in simplicity and scalability of the aforementioned infiltration-based SEA manufacturing process. The SEA shows two different faces (Figure 1e). On top of the CNF separator side, a word of “Wood Nanocellulose” was written using a white ink pen, verifying that the CNF separator is a kind of paper. At the same time, the electrode comprising the SWNT-netted electrode active materials was formed on the opposite side of the CNF separator. This 1D nanobuilding block (CNF/ SWNT)-enabled structural uniqueness of the SEAs is conceptually illustrated in Figure 1f. An important requirement for battery separators is the acquisition of high porosity and well-defined pore size, which enables facile ion transport via the electrolyte-filled separator between electrodes.28,31 The CNFs are known to be densely packed by capillary action due to their hydrogen bonds of β-(1 → 4)-D-glucopyranose repeat units during water evaporation,12−15 resulting in the formation of CNF paper with lessdeveloped porous structure (the inset of Figure 2a). To overcome this limitation, we exploited the solvent (ethanol followed by acetone) exchange-assisted freeze-drying method.32 Figure 2a shows that, after the freeze-drying treatment, the porous structure with a plethora of submicrometer-sized pores was formed in the CNF separator. The pore size distribution of the CNF separator was quantitatively measured by mercury intrusion porosimetry (Supporting Information Figure S2). In addition, from the pore size distribution result, the specific surface area was estimated to be approximately 145 m2 g−1. The cross-sectional SEM images of the SEAs demonstrate the construction of tightly interlocked separator−electrode interface (Figure 2b,c), verifying that the CNF separator (thickness ∼20 μm) enables the strong interfacial adhesion with SWNT-netted electrodes (thickness ∼34 μm for cathode and 40 μm for anode). The atomic force microscopy (AFM) image, together with the SEM results, shows that the electrode active materials were highly reticulated with the SWNTs (Figure 2d−f). Neither conventional polymeric binders nor metallic current collectors were found, underscoring that the SWNTs in the electrode exhibited multifunctional roles as conductive additives, binders and also current collectors. Moreover, the removal of traditional polymeric binders allows the generation of highly developed interstitial voids in the electrodes, thus enabling facile accessibility of liquid electrolytes toward electrode active materials. The strongly interlocked separator−electrode interface is expected to endow the SEAs with dimensional/mechanical robustness, even in the absence of polymeric binders and metallic current collectors in electrodes. Figure 2g shows that neither peel-off nor disintegration of constituents was observed at the cathode SEA during the taping-out test (using commercial 3M scotch tape). In comparison, for a control sample that employed a typical filter paper with excessively thick (micrometer scale) fibers (Supporting Information Figure S3) as a separator, a considerable amount of SWNTs and LiFePO4 powders were detached (the inset of Figure 2g). This result demonstrates that the well-tailored nanoporous CNF separator played an important role in the construction of unitized SEAs. The superior interfacial adhesion in the SEA was further verified by examining SEM images of the tape face and also cathode surface, after the taping-out test (Supporting Information Figure S4). In addition to the cathode SEA, the anode SEA also exhibited a good interfacial adhesion between the anode and CNF separator (Supporting Information Figure S5). Recently, in an attempt to develop high-performance 5680

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

Letter

Figure 3. Electrochemical characterization of SEAs. (a) CV profiles of conventional LiFePO4 cathode, LiFePO4 cathode SEA and SWNTs on CNF separator. A schematic illustration depicting two different electrochemical reaction modes of the SEA is also provided. (b) Comparison of areal mass (i.e., cathode (or anode) + separator, mg cm−2) between conventional electrodes (polyethylene separator is also included) and SEAs. (c) Comparison of gravimetric (based on the combined mass of cathode (or anode) and separator) specific capacity between conventional electrodes and SEAs. (d) A photograph of an h-nanomat full cell and a schematic illustration describing the electron transport pathways and electrolyte accessibility. (e) Charge/discharge profiles of a conventional LiFePO4/Li4Ti5O12 cell and an h-nanomat cell, where cell mass (=cathode + anode + separator)-based gravimetric specific charge/discharge capacities are presented. (f) Cycling performance (=capacity retention as a function of cycle numbers) of a conventional LiFePO4/Li4Ti5O12 cell and an h-nanomat cell, where the cells are cycled at charge/discharge current density = 2.0/2.0 C for the 200 cycles and then further cycled at charge/discharge current density = 10.0/10.0 C up to the additional 300 cycles. Structural variation on electrode surface after the cycling test (200 cycles at 2.0/2.0 C → 300 cycles at 10.0/10.0 C); (g) TOF-SIMS image of the LiF byproduct on cathode SEA (the inset is the result of conventional LiFePO4 cathode); (h) XPS spectra (F 1s) of cathode SEA (the inset shows the result of conventional LiFePO4 cathode). (i) SEM image of cathode SEA (the inset shows the result of anode SEA) after the 500 cycles.

behaved as electrochemically active materials participating in the non-Faradaic reaction, although their contribution to the total capacity was not so large. Different from typical Faradaic reaction of lithium-ion electrode materials,41 CNTs tend to undergo non-Faradaic reaction42 that predominantly occurs at their surface, except for the low voltage region below 1.0 V (vs Li+/Li) in which the intercalation/deintercalation reaction of lithium ions in CNTs may occur.43 A conventional electrochemically inert polymeric binder such as PVdF may shield electrolyte-accessible area of CNTs, thus suppressing the activation of the CNT-driven non-Faradaic reaction. This postulation was verified by examining CV profiles of a control cathode (LiFePO4/SWNT/PVdF = 80/15/5 (w/w/w)). The non-Faradaic reaction-induced capacitance was found to be negligible (Supporting Information Figure S9), despite the presence of a large amount (=15 wt %) of SWNTs in the electrode. This result demonstrates the importance of the effective surface area (i.e., the electrolyte-accessible area) of CNTs required for activating their non-Faradaic reaction and also proves the advantageous effect of binder-free structure in the SEAs, which allows the electrochemical involvement of

integrity in the electrolyte-swollen state. This excellent mechanical/dimensional tolerance of SEAs in the absence of polymeric binders and metallic current collectors is ascribed to the tightly interlocked electrode-separator interface and also the SWNTs reticulating electrode active materials (Figure 2b−g). We performed cyclic voltammetry (CV) analysis to elucidate the electrochemical behavior of SEAs (Figure 3a). In comparison to the conventional LiFePO4 cathode, the cathode SEA shows the unusual CV profiles, which could be divided into two different regions (A and B). Region A is governed by the SWNT-driven non-Faradaic reaction (i.e., adsorption/ desorption of charges on surface of SWNTs). Region B exhibits the mixed non-Faradaic/Faradaic reaction, where the Faradaic reaction (induced by lithium intercalation/deintercalation of LiFePO4 active materials) dominates the overall cell capacity. A schematic illustration on this electrochemical reaction of the SEA is depicted in Figure 3a. The non-Faradaic reaction in the cathode SEA can be further explained by analyzing the CV profile of a control sample (i.e., SWNT sheet without LiFePO4 powders deposited on CNF separator). This result exhibits that the SWNTs of the SEAs 5681

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

Letter

Figure 4. Electrochemical superiority of h-nanomat cell. Schematic illustrations depicting electron/ion transport behavior: (a) a conventional cell composed of powder-piled electrodes (i.e., a mixture of electrode active powders/conductive additive powders/polymeric binders on metallic current collectors) and polyolefin separator; (b) an h-nanomat cell comprising anode SEA and cathode SEA. (c) Discharge rate capability of cells (hnanomat cell vs conventional cell), where cells were charged at a constant current density of 0.5 C and discharged over a wide range of current densities (0.5−50.0 C). (d) Charge rate capability of cells (h-nanomat cell vs conventional cell), where cells were discharged at a constant current density of 0.5 C and charged in the range of current densities (0.5−20.0 C). (e) Cell polarization (h-nanomat cell vs conventional cell) as a function of the applied current density during charge/discharge reaction. (f) GITT profiles of cells (h-nanomat cell vs conventional cell), where the inset shows the variation of internal cell resistance as a function of state of charge and depth of discharge.

was found to be higher than that of a theoretical value (∼170 mAh g(LiFePO4)−1)44 of LiFePO4. This extra increment in the discharge capacity of LiFePO4 active materials of the cathode SEA may be ascribed to the capacity contribution of nonFaradaic reaction of coexisting SWNTs. In addition to the LiFePO4 cathode SEA, the similar electrochemical influence of

non-Faradaic SWNTs as another electrode material. The galvanostatic charge/discharge profiles (Supporting Information Figure S10, charge/discharge current density = 0.5/0.5 C) of cells show that the LiFePO4 cathode SEA presents a higher reversible discharge capacity (∼185 mAh g(LiFePO4)−1) than the conventional LiFePO4 cathode (∼160 mAh g(LiFePO4)−1). Moreover, the discharge capacity of LiFePO4 cathode SEA 5682

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

Letter

water molecules and liquid electrolytes (specifically, lithium salt (LiPF6)) were not observed at the h-nanomat cell, continuous efforts for reducing residual water should be undertaken in future studies. Conventional cells, which are composed of powder-piled electrodes (i.e., a mixture of electrode active powders/ conductive additive powders/polymeric binders on metallic current collectors) and polyolefin separators with relatively poorly developed porous structure,27,28,31 have limitations in attaining the well-established electron networks and ionic channels, which may thus provoke sluggish and nonuniform electronic/ionic flow (Figure 4a). By comparison, the 1D nanobuilding block-based SEAs (Figure 4b) are characterized with highly interconnected, uniform electron pathways and porous structure (enabling facile accessibility of liquid electrolyte), which are expected to provide significant improvements in the electrochemical performance (in particular, rate capability) of h-nanomat cells. Under a constant charge current density of 0.5 C, the h-nanomat cell shows a higher discharge capacity than the conventional LiFePO4/Li4Ti5O12 cell (Figure 4c) over a wide range of discharge current densities (=0.5−50.0 C) at a voltage range of 1.0−2.5 V, which reveals the superior discharge rate capability. Moreover, the advantageous effect of the h-nanomat cell was also observed for the charge rate capability (Figure 4d), where the cells were discharged at a constant current density of 0.5 C and charged in the range of 0.5−20.0 C. This improvement in the fast charging was further confirmed by analyzing the TOF-SIMS images (Supporting Information Figure S15) of cathode surface after being charged at 10.0 C. In the TOF-SIMS images, the yellow color indicates a larger number of lithium ions on the cathode surface. In comparison to the cathode surface of the conventional cell, the homogeneous distribution and smaller number of lithium ions were observed over a wide area of cathode surface for the hnanomat cell, demonstrating the facile and uniform ionic/ electronic transport at the fast-charging condition and also the delithiation of higher number of lithium ions (corresponding to the higher capacity shown in Figure 4d). A comparison of charge/discharge profiles (Supporting Information Figure S16) shows that the h-nanomat cell presents a lower cell polarization than the conventional cell at all charge/discharge current densities examined herein. A quantitative analysis of the polarization as a function of the applied current density, where the polarization was determined by measuring the difference in initial voltage between the current density of 0.5 C and a given current density, is summarized in Figure 4e. This beneficial effect of the hnanomat cell on the cell polarization was further verified through conducting Galvanostatic Intermittent Titration Technique (GITT) experiment47 (Figure 4f), where the current was applied for 6 min at a charge (or discharge) current density of 1 C (=0.32 mA cm−2) and an interruption time of 60 min between each pulse. The h-nanomat cell effectively mitigated the build-up in ohmic polarization of a cell upon the repeated current stimuli during both charge and discharge reaction. The values of internal resistance are also summarized as a function of the state of charge and the depth of discharge (the inset of Figure 4f). This superior charge/ discharge rate performance of the h-nanomat cell is attributed to the improved electrochemical kinetics (specifically, the SWNTs-mediated electron networks and 1D nanobuilding block-induced facile electrolyte accessibility) of the SEAs. To explore the broader applicability of the h-nanomat cells with the

SWNTs was observed at the Li4Ti5O12 anode SEA (Supporting Information Figure S11). In battery components, current collectors are known to provide electron transfer toward electrode active materials and also serve as dimensional support layers for electrodes.33−36 The heavy current collectors (particularly metallic-based ones), however, exert a detrimental influence on the gravimetric capacity of electrodes. In the SEAs, the removal of metallic current collectors as well as polymeric binders allows the larger amount of electrode active materials to be loaded in electrodes, thus enabling a substantial increase in (electrode mass-based) gravimetric specific capacity. Moreover, the lower weight (=0.54 mg cm−2) of CNF separators compared to that (=1.3 mg cm−2) of traditional polyolefin separators also contributes to this capacity improvement. Figure 3b shows that the SEAs allow a drastic reduction in the areal mass (=cathode (or anode) + separator). As a result, the SEAs present approximately 2 times higher gravimetric specific capacity than the conventional electrodes (Figure 3c). The unitized SEA configuration can simplify cell assembly. Alternative stacking of an anode SEA and a cathode SEA leads to the construction of a h-nanomat full cell. A photograph of a h-nanomat full cell and a schematic illustration depicting the electron/ion transport are provided in Figure 3d. Meanwhile, the h-nanomat cell exhibits no appreciable deterioration in the discharge profile after the storage for 100 h at room temperature (Supporting Information Figure S12), demonstrating the long-term storage stability31,35 of the proposed SEA architecture. Figure 3e shows that the h-nanomat cell presents the cell mass (=cathode + anode + separator)-based gravimetric specific discharge capacity of 60 mAh gcell−1 at a constant charge/discharge current density = 2.0/2.0 C under a voltage range of 1.0−2.5 V, which is significantly higher than that of the conventional LiFePO4/Li4Ti5O12 cell (∼27 mAh gcell−1). We evaluated the long-term cycling performance of the h-nanomat cell (Figure 3f). The h-nanomat cell shows the better capacity retention for the 200 cycles (at charge/discharge current density = 2.0/2.0 C) than the conventional full cell. More notably, this superiority in the cycling performance was still maintained up to the additional 300 cycles even at the faster charge/discharge rate of 10.0/10.0 C (Figure 3f, Supporting Information Figure S13). To better understand the excellent cycling performance of the h-nanomat cell, the variation of cathode surface after the cycling test was investigated. The surface analysis using time-offlight secondary ion mass-spectroscopy (TOF-SIMS) and X-ray photoelectron spectroscopy (XPS) shows that the formation of LiF, one of the byproducts generated from unwanted electrochemical decomposition of LiPF6 salts in liquid electrolytes,45,46 was suppressed in the h-nanomat cell compared to the conventional cell (Figure 3g,h). Figure 3i demonstrates that the well-designed porous structure of electrodes featuring the SWNT-netted electrode active material was not impaired after the 500 cycles, verifying the long-term structural/electrochemical robustness of SEAs. Meanwhile, to address possible concerns about residual water that may be trapped in the CNF separators, the water content of the h-nanomat cell was compared with that of the conventional cell (Supporting Information Figure S14). The h-nanomat cell presented slightly larger water content (373 ppm for LiFePO4 cathode SEA and 289 ppm for Li4Ti5O12 anode SEA) than the conventional cell (190 ppm for LiFePO4 cathode and 103 ppm for Li4Ti5O12 anode). Although harmful side reactions between the residual 5683

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

Letter

Figure 5. Shape flexibility and thermal tolerance of h-nanomat cell. (a) A schematic illustration of an h-nanomat cell with the unitized SEA configuration. (b) A photograph showing the electrochemical activity of a h-nanomat cell after the 5 bending/unbending cycles (rod diameter = 5 mm), where the inset shows the corresponding charge/discharge profile at charge/discharge current density = 1.0/1.0 C. Here, to clearly visualize the advantageous effect of an h-nanomat cell, pouch cells (width × length × thickness = 30 × 40 × 0.3 mm/mm/mm) incorporating a polydimethylsiloxane-coated polyethylene film (thickness ∼100 μm) as a transparent packaging substance were fabricated. (c) Variation in the cell voltage of an h-nanomat cell as a function of bending cycle (the inset shows charge/discharge profile (at charge/discharge current density = 1.0/1.0 C) after the 50 bending cycles). (d) A photograph showing the electrochemical activity of an h-nanomat cell after the repeated twisting/untwisting deformation. (e) Voltage fluctuation of cells (h-nanomat cell vs conventional cell) with the repeated bending/unbending cycles (rod diameter = 5 mm) during the charge/discharge reaction at charge/discharge current density = 1.0/1.0 C (the inset presents voltage fluctuation of cells with the repeated bending/unbending cycles as a function of normalized charge/discharge capacity). (f) Thermal tolerance of the h-nanomat cell after exposure to high-temperature shock (= 150 °C/0.5 h), where the inset shows the electrochemical activity loss of the conventional cell. For this measurement, aluminum pouch-type cells (width × length × thickness = 30 × 50 × 2 (mm/mm/mm)) were fabricated.

emerging so-called wearable electronics.48−50 The structural uniqueness of SEAs is expected to bring remarkable improvements in the shape flexibility of a h-nanomat cell. Here, to clearly visualize the advantageous effect of a h-nanomat cell, a polydimethylsiloxane-coated polyethylene film (thickness ∼100 μm) was used as a kind of transparent packaging substance.

unusual electrochemical properties, future studies are devoted to combining the SEA architecture with a wide variety of electrode active materials (particularly focusing on highcapacity or high-voltage applications). Flexible batteries with various form factors have garnered considerable attention as a suitable power source for newly 5684

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

Letter

The h-nanomat cell with the unitized SEA configuration (Figure 5a), as compared to the conventional cell comprising the three independent sheets of anode, cathode and separator, fixates the spatial position between electrode and separator, which thus ensures the misalignment-proof of the separatorelectrode interface. Figure 5b shows that the h-nanomat cell, after the repeated bending/unbending (rod diameter = 5 mm), presented a strong tolerance against the mechanical deformation, without impairing charge/discharge performance (the inset of Figure 5b shows the charge/discharge profile after the 5 bending/unbending cycles). To investigate the long-term stability of a h-nanomat cell under the repeated deformation, the variation in the cell voltage was measured as a function of bending cycle (Figure 5c). The h-nanomat cell shows no appreciable loss in the charge voltage during the repeated bending/unbending and also maintains stable charge/discharge profiles even after the 50 cycles (the inset of Figure 5c). Moreover, the h-nanomat cell could be repeatedly twisted and untwisted without losing its electrochemical activity (Figure 5d). To further elucidate the excellent flexibility of a h-nanomat cell, the voltage fluctuation of cells with the repeated bending/ unbending cycles (time interval between each bending deformation was set at 3 min, rod diameter = 5 mm) was monitored during the charge/discharge reaction. Photographs depicting this experimental scheme are provided in Supporting Information Figure S17. Figure 5e shows that, for the conventional cell, the gap distance between the cathode and anode was periodically changed upon the repeated bending deformation, thereby giving rise to the larger amplitudes of voltage fluctuation. By contrast, the h-nanomat cell featuring the tightly interlocked separator−electrode interface effectively mitigates the bending-induced periodic voltage fluctuation. In addition, the voltage fluctuation of cells with the repeated bending was also plotted as a function of normalized charge/ discharge capacity (the inset of Figure 5e). The advantageous effect of a h-nanomat cell on the thermal stability was elucidated by performing the high-temperature shock test (=150 °C/0.5 h). Under this harsh operating condition, the conventional cell, because of the presence of volatile liquid electrolyte and polyethylene separator having the (area-based) a large thermal shrinkage (ΔA ∼ 91% after exposure to 150 °C/0.5 h, Supporting Information Figure S18a), was easily swollen and lost electrochemical activity (the inset of Figure 5f), revealing the occurrence of internal shortcircuit.51,52 In comparison, the h-nanomat cell exhibited a negligible deterioration in the charge/discharge behavior (Supporting Information Figure S19), although the cell was swollen because of the incorporated volatile liquid electrolyte (Figure 5f). This improvement in the thermal tolerance of a hnanomat cell can be explained by the presence of thermally stable CNF separator27,28 (ΔA ∼ 0%, Supporting Information Figure S18b) and also the architectural robustness of unitized SEAs. The overall dimensional stability of SEAs in the hnanomat cell was not disrupted after the high-temperature shock test (Supporting Information Figure S20). In summary, we have presented a new concept of h-nanomat battery based on unitized SEA architecture as a reliable and sustainable power source to facilitate the advent of smart/ wearable energy era. The unitized SEAs were composed of nanoporous CNF separators and current collector-/binder-free electrode comprising solely SWNT-netted electrode active materials. Herein, the CNF separator played a viable role in achieving tightly interlocked electrode/separator interface and

the SWNTs exhibited the multifunctional roles as electron conductive additives, binders, current collectors and also nonFaradaic active materials. The 1D nanobuilding block (CNF/ SWNT)-driven structural/physicochemical uniqueness of the SEAs brought significant improvements in the mass loading of electrode active materials, electron conduction pathways, electrolyte accessibility and misalignment-proof of separator/ electrode interface (preventing the occurrence of internal shortcircuit). As a consequence, the h-nanomat batteries provided unprecedented advances in the cell performance (particularly at fast charge/discharge rates) and shape flexibility far beyond those accessible with conventional battery materials and configurations. Notably, the tightly interlocked separator/ electrode interface of the SEAs contributed to the good dimensional/thermal tolerance at harsh operating conditions, underlying the superiority of the h-nanomat battery as a flexible/safer power source. We envision that the architectural concept of h-nanomat batteries holds a great deal of promise as 1D nanobuilding block-mediated scalable/versatile platform technology for development of high-performance flexible lithium-ion batteries and can also be easily extended to nextgeneration energy storage systems relying on so-called “beyond lithium-ion chemistry”.

■

ASSOCIATED CONTENT

S Supporting Information *

Details of methods and supplementary results demonstrating additional structural/physicochemical properties and also electrochemical performance of cells. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Author Contributions

K.-H.C. performed the experiments and analyzed the data. S.J.C., J.T.Y. and C.K.L. performed the electrochemical analysis. W.K. participated in discussing the properties of single-walled carbon nanotubes. Q.W., S.-J.C., S.-B.P, D.-H.C and S.-Y.L. prepared and analyzed the cellulose nanofibrils. S.-Y.L. coordinated and supervised the overall project. K.-H.C. and S.-Y.L. wrote the manuscript and all authors discussed the results and participated in manuscript preparation. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS

This work was supported by the Korea Forest Research Institute grant (FP 0400-2007-03). This work was also supported by the National Research Foundation of Korea Grant funded by the Korean Government (MEST) (NRF2009-C1AAA001-2009-0093307). This work was also supported by Energy Efficiency and Resources R&D program (20112010100150) under the Ministry of Knowledge Economy, Republic of Korea. This work was also supported by the BK21 Plus Program (META-material-based Energy Harvest and Storage Technologies, 10Z20130011057) funded by the Ministry of Education (MOE, Korea) and National Research Foundation of Korea (NRF). 5685

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686

Nano Letters

■

Letter

(37) Jia, X.; Chen, Z.; Suwarnasarn, A.; Rice, L.; Wang, X.; Sohn, H.; Zhang, Q.; Wu, B. M.; Wei, F.; Lu, Y. Energy Environ. Sci. 2012, 5, 6845−6849. (38) Wang, K.; Luo, S.; Wu, Y.; He, X.; Zhao, F.; Wang, J.; Jiang, K.; Fan, S. Adv. Funct. Mater. 2013, 23, 846−853. (39) Kim, H.; Lim, H. D.; Kim, S. W.; Hong, J.; Seo, D. H.; Kim, D. C.; Jeon, S.; Park, S.; Kang, K. Sci. Rep. 2013, 3, 1506−1513. (40) Lim, H. D.; Song, H.; Kim, J.; Gwon, H.; Bae, Y.; Park, K. Y.; Hong, J.; Kim, H.; Kim, T.; Kim, Y. H.; Lepro, X.; Ovalle-Robles, R.; Baughman, R. H.; Kang, K. Angew. Chem. 2014, 53, 3926−3931. (41) Tarascon, J. M.; Armand, M. Nature 2001, 414, 359−367. (42) Lee, S. W.; Gallant, B. M.; Lee, Y.; Yoshida, N.; Kim, D. Y.; Yamada, Y.; Noda, S.; Yamada, A.; Shao-Horn, Y. Energy Environ. Sci. 2012, 5, 5437−5444. (43) Landi, B. J.; Ganter, M. J.; Cress, C. D.; DiLeo, R. A.; Raffaelle, R. P. Energy Environ. Sci. 2009, 2, 638−654. (44) Hu, L. H.; Wu, F. Y.; Lin, C. T.; Khlobystov, A. N.; Li, L. J. Nat. Commun. 2013, 4, 1687−1693. (45) Lux, S. F.; Lucas, I. T.; Pollak, E.; Passerini, S.; Winter, M.; Kostecki, R. Electrochem. Commun. 2012, 14, 47−50. (46) Xu, K. Chem. Rev. 2004, 104, 4303−4418. (47) Dees, D. W.; Kawauchi, S.; Abraham, D. P.; Prakash, J. J. Power Sources 2009, 189, 263−268. (48) Kil, E. H.; Choi, K. H.; Ha, H. J.; Xu, S.; Rogers, J. A.; Kim, M. R.; Lee, Y. G.; Kim, K. M.; Cho, K. Y.; Lee, S. Y. Adv. Mater. 2013, 25, 1395−1400. (49) Lee, S.-Y.; Choi, K.-H.; Choi, W.-S.; Kwon, Y. H.; Jung, H.-R.; Shin, H.-C.; Kim, J. Y. Energy Environ. Sci. 2013, 6, 2414−2423. (50) Zhou, G.; Li, F.; Cheng, H.-M. Energy Environ. Sci. 2014, 7, 1307−1338. (51) Wang, Q.; Ping, P.; Zhao, X.; Chu, G.; Sun, J.; Chen, C. J. Power Sources 2012, 208, 210−224. (52) Woo, J.-J.; Zhang, Z.; Amine, K. Adv. Energy Mater. 2014, 4, 1301208.

REFERENCES

(1) Armand, M.; Tarascon, J.-M. Nature 2008, 451, 652−657. (2) Tollefson, J. Nature 2008, 456, 436−440. (3) Dunn, B.; Kamath, H.; Tarascon, J.-M. Science 2011, 334, 928− 935. (4) Bruce, P. G.; Scrosati, B.; Tarascon, J. M. Angew. Chem., Int. Ed. 2008, 47, 2930−2946. (5) Hassoun, J.; Lee, K.-S.; Sun, Y.-K.; Scrosati, B. J. Am. Chem. Soc. 2011, 133, 3139−3143. (6) Lee, M.; Hong, J.; Kim, H.; Lim, H. D.; Cho, S. B.; Kang, K.; Park, C. B. Adv. Mater. 2014, 26, 2558−2565. (7) Chan, C. K.; Peng, H.; Liu, G.; McIlwrath, K.; Zhang, X. F.; Huggins, R. A.; Cui, Y. Nat. Nanotechnol. 2008, 3, 31−35. (8) Jung, D. S.; Ryou, M.-H.; Sung, Y. J.; Park, S. B.; Choi, J. W. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 12229−12234. (9) Kang, B.; Ceder, G. Nature 2009, 458, 190−193. (10) Ji, X.; Lee, K. T.; Nazar, L. F. Nat. Mater. 2009, 8, 500−506. (11) Kamaya, N.; Homma, K.; Yamakawa, Y.; Hirayama, M.; Kanno, R.; Yonemura, M.; Kamiyama, T.; Kato, Y.; Hama, S.; Kawamoto, K.; Mitsui, A. Nat. Mater. 2011, 10, 682−686. (12) Hamedi, M. M.; Hajian, A.; Fall, A. B.; Håkansson, K.; Salajkova, M.; Lundell, F.; Wågberg, L.; Berglund, L. A. ACS Nano 2014, 8, 2467−2476. (13) Klemm, D.; Kramer, F.; Moritz, S.; Lindstrom, T.; Ankerfors, M.; Gray, D.; Dorris, A. Angew. Chem., Int. Ed. 2011, 50, 5438−5466. (14) Moon, R. J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood, J. Chem. Soc. Rev. 2011, 40, 3941−3994. (15) Tobjork, D.; Osterbacka, R. Adv. Mater. 2011, 23, 1935−1961. (16) Jabbour, L.; Destro, M.; Gerbaldi, C.; Chaussy, D.; Penazzi, N.; Beneventi, D. J. Mater. Chem. 2012, 22, 3227−3233. (17) Jeong, S. S.; Böckenfeld, N.; Balducci, A.; Winter, M.; Passerini, S. J. Power Sources 2012, 199, 331−335. (18) Leijonmarck, S.; Cornell, A.; Lindbergh, G.; Wågberg, L. Nano Energy 2013, 2, 794−800. (19) Kang, Y. J.; Chun, S. J.; Lee, S. S.; Kim, B. Y.; Kim, J. H.; Chung, H.; Lee, S. Y.; Kim, W. ACS Nano 2012, 6, 6400−6406. (20) Hu, L.; Choi, J. W.; Yang, Y.; Jeong, S.; La Mantia, F.; Cui, L. F.; Cui, Y. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 21490−21494. (21) Hu, L.; Mantia, F. L.; Wu, H.; Xie, X.; McDonough, J.; Pasta, M.; Cui, Y. Adv. Energy Mater. 2011, 1, 1012−1017. (22) Wang, J.; Li, L.; Wong, C. L.; Madhavi, S. Nanotechnology 2012, 23, 495401. (23) Gómez Cámer, J. L.; Morales, J.; Sánchez, L.; Ruch, P.; Ng, S. H.; Kötz, R.; Novák, P. Electrochim. Acta 2009, 54, 6713−6717. (24) Zhu, H.; Jia, Z.; Chen, Y.; Weadock, N.; Wan, J.; Vaaland, O.; Han, X.; Li, T.; Hu, L. Nano Lett. 2013, 13, 3093−3100. (25) Kuribayashi, I. J. Power Sources 1996, 63, 87−91. (26) Zhang, J.; Liu, Z.; Kong, Q.; Zhang, C.; Pang, S.; Yue, L.; Wang, X.; Yao, J.; Cui, G. ACS Appl. Mater. Interfaces 2013, 5, 128−134. (27) Chun, S.-J.; Choi, E.-S.; Lee, E.-H.; Kim, J. H.; Lee, S.-Y.; Lee, S.-Y. J. Mater. Chem. 2012, 22, 16618−16626. (28) Kim, J.-H.; Kim, J.-H.; Choi, E.-S.; Yu, H. K.; Kim, J. H.; Wu, Q.; Chun, S.-J.; Lee, S.-Y.; Lee, S.-Y. J. Power Sources 2013, 242, 533−540. (29) Leijonmarck, S.; Cornell, A.; Lindbergh, G.; Wågberg, L. J. Mater. Chem. A 2013, 1, 4671−4677. (30) Islam, M.; Rojas, E.; Bergey, D.; Johnson, A.; Yodh, A. Nano Lett. 2003, 3, 269−273. (31) Arora, P.; Zhang, Z. Chem. Rev. 2004, 104, 4419−4462. (32) Sehaqui, H.; Zhou, Q.; Ikkala, O.; Berglund, L. A. Biomacromolecules 2011, 12, 3638−3644. (33) Luo, S.; Wang, K.; Wang, J.; Jiang, K.; Li, Q.; Fan, S. Adv. Mater. 2012, 24, 2294−2298. (34) Zhang, H.-X.; Feng, C.; Zhai, Y.-C.; Jiang, K.-L.; Li, Q.-Q.; Fan, S.-S. Adv. Mater. 2009, 21, 2299−2304. (35) Hu, L.; Wu, H.; La Mantia, F.; Yang, Y.; Cui, Y. ACS Nano 2010, 4, 5843−5848. (36) Gui, Z.; Zhu, H. L.; Gillette, E.; Han, X. G.; Rubloff, G. W.; Hu, L. B.; Lee, S. B. ACS Nano 2013, 7, 6037−6046. 5686

dx.doi.org/10.1021/nl5024029 | Nano Lett. 2014, 14, 5677−5686