Research Article www.acsami.org
Understanding the Effects of Electrode Formulation on the Mechanical Strength of Composite Electrodes for Flexible Batteries Abhinav M. Gaikwad* and Ana Claudia Arias* Electrical Engineering and Computer Sciences Department, University of California Berkeley, Berkeley, California 94720, United States S Supporting Information *
ABSTRACT: Flexible lithium-ion batteries are necessary for powering the next generation of wearable electronic devices. In most designs, the mechanical flexibility of the battery is improved by reducing the thickness of the active layers, which in turn reduces the areal capacity and energy density of the battery. The performance of a battery depends on the electrode composition, and in most flexible batteries, standard electrode formulation is used, which is not suitable for flexing. Even with considerable efforts made toward the development of flexible lithium-ion batteries, the formulation of the electrodes has received very little attention. In this study, we investigate the relation between the electrode formulation and the mechanical strength of the electrodes. Peel and drag tests are used to compare the adhesion and cohesion strength of the electrodes. The strength of an electrode is sensitive to the particle size and the choice of polymeric binder. By optimizing the electrode composition, we were able to fabricate a high areal capacity (∼2 mAh/cm2) flexible lithium-ion battery with conventional metal-based current collectors that shows superior electrochemical and mechanical performance in comparison to that of batteries with standard composition. KEYWORDS: flexible batteries, mechanical testing, peel test, drag test, lithium-ion batteries
■
make flexible batteries that are commercially viable, the battery design should be simple, the fabrication process should be easily scalable, and the energy density of the battery should be comparable to that of conventional nonflexible lithium-ion batteries. In addition to the cell architecture, the electrode parameters such as porosity, thickness, type and quantity of binder, conductive additives, and active particles play an important role toward overall performance of the battery.37−41 In commercial batteries, the electrode parameters are tuned carefully based on the application. For example, batteries in cell phones and laptops have thick and dense electrodes to increase the energy density of the battery, whereas batteries in power tools have thin electrodes to reduce polarization at high Crates.42 Through use of a similar engineering principle, it may be possible to improve the flexibility of the electrodes by optimizing their composition. Presently, the relation between the electrode composition and its mechanical flexibility is not clearly understood. In most reports on flexible batteries, a generic ink composition (eight parts of active particles, one part carbon, and one part binder) is used, which is not suitable for flexible batteries.41 Because flexible batteries are going to play an important role in the future, it is important to have a better
INTRODUCTION In recent years, interest in wearable technologies has fueled research efforts in developing flexible lithium-ion batteries.1−5 One can envision thin bendable batteries integrated within flexible electronic devices such as activity trackers and health sensors.6,7 An ideal flexible battery should meet the power and energy requirements of the device and retain its electrochemical performance when flexed. In a conventional battery, the anode and cathode are printed on thin metal foils and separated with a porous membrane. The whole electrode stack is soaked with an electrolyte solution and encapsulated within metal-laminated pouches.8−11 When flexed, batteries can fail/degrade by delamination of the active layers or by a loss of particle-toparticle contact.1 Researchers have demonstrated numerous innovative designs to improve the flexibility of lithium-ion batteries. For example, the mechanical flexibility of the active layers was improved by embedding them inside porous membranes,12−14 reducing the thickness of active layers,15 and reinforcing them with carbon nanotubes16,17 and graphene foams.18−20 Similarly, the adhesion of the active layers to the current collectors was improved by replacing metal-based current collector foils with conducting fabrics,21−24 conductive ink,12,25−28 and carbon nanotubes.14,29 Even with considerable progress toward the development of flexible lithium-ion batteries, the proposed designs have a poor volumetric energy density, low areal capacity, or use fabrication processes that are not scalable for manufacturing.15,24,29−36 To © 2017 American Chemical Society
Received: November 16, 2016 Accepted: February 2, 2017 Published: February 2, 2017 6390
DOI: 10.1021/acsami.6b14719 ACS Appl. Mater. Interfaces 2017, 9, 6390−6400
Research Article
ACS Applied Materials & Interfaces
Figure 1. (A) Schematic of the peel test setup. The electrode is clamped on one end of the Instron. An adhesive tape is attached to the electrode, and the free end of the tape is clamped on the moveable end of the Instron, which is connected to a 10 N load cell. (B) Schematic of delamination of the electrode at the weakest interface. (C) Photograph of the electrode during the peel test and the typical peel force recorded by the load cell during the test. (D) Optical micrographs of the tape and the electrode after the peel test. weight fractions (x wt %) of the carbon additive and PVDF binder in the electrode were kept constant. For x wt % of carbon additive, the weight fraction of LCO in the electrode was 100 − 2x wt %. The slurry preparation process starts with preparing a dry mix (active particle and conductive additive) and wet mix (binder and solvent). The dry and wet mix was homogenized on a vortex mixer for 3 h. The dry and wet mix were mixed together and further homogenized on a vortex mixer for 12 h. The slurries were printed on the metal foils with a doctor blade at speed of 15 mm/s. The blade height was adjusted to achieve the desired loading. The electrodes were dried in an oven at 100 °C for 2 h. The porosity of the electrodes was adjusted to 25% by a calendering process. Conductivity and Mechanical Testing. The samples for the conductivity measurement were prepared by printing the slurries on a plastic (PET) substrate. The conductivity was measured using a 4point probe setup. Measurements were performed at five locations for each sample. The mechanical integrity of the electrodes was tested with peel and drag test. The peel test was performed on an Instron with a 10 N load cell. During the test, one end of the electrode was fixed to the instrument. Scotch tape (3 M 600) was firmly pressed on the electrode, ensuring that there were no air bubbles between the tape and electrode. The free end of the tape was attached to the moveable end of the instrument. The tape was peeled at an angle of 180° from the film at a speed of 100 μm/s. The load cell measured the force required to peel the tape. The adhesion strength of the electrodes was denoted in terms of the force per unit length of the tape (N/cm). The drag test was performed using a DAGE bondtester. The width was the drag tip was 250 μm. In a typical test, the drag tip was brought in close proximity to the current collector foil and plowed through the electrodes at a speed of 100 μm/s for a distance of 3000 μm. The height of the tip was denoted in terms of its distance to the foil. The instrument measured the horizontal force (g) on the tip during the experiment. For each electrode composition, the test was repeated 3− 4 times. Electrochemical Testing. After the calendering step, the electrodes were cut to a dimension of 1.3 × 1.3 in.2. The loading of the anode and cathode were adjusted such that the ratio of the
understanding of the relation between the electrode composition and its mechanical flexibility. In this paper, we use a peel and modified drag test to measure the mechanical strength of composite battery electrodes. In the peel test, the interface of delamination provides qualitative information about possible failure mechanisms in the electrodes during flexing, and the force on the tip during the drag test denotes the cohesion strength of the electrode. With the goal of fabricating a high areal capacity lithium cobalt oxide/graphite-based flexible lithium-ion battery with conventional metal-based current collectors, we focused our attention on the composition of the electrodes. By optimizing the electrode composition, we were able to improve both the electrochemical performance and mechanical strength of the electrodes. Full cells with an areal capacity of 2 mAh/cm2 were fabricated, and the electrochemical performance and mechanical flexibility of the batteries with standard and modified composition were compared. Batteries with modified composition had a higher capacity retention when flexed repeatedly to a bending radius of 1.27 cm.
■
EXPERIMENTAL SECTION
Electrode Preparation. Lithium cobalt oxide (LCO, MTI Corp.) and synthetic graphite (MTI Corp.) were used as the active materials for the cathode and anode, respectively. Carbon black (C-NERGY C65, IMERYS) and graphite (C-NERGY SFG 6L, IMERYS) were used as the conductive additives. Polyvinylidene fluoride (PVDF, Kureha Corp.) and polystyrene butadiene rubber (PSBR, Targray Technology) were used as the binder for the electrodes. Stainless steel foil (SS, 12.5 μm) and nickel foil (10 μm) were used as the current collectors for the cathode and anode, respectively. The graphite electrode was composed of 90.2 wt % graphite, 4.4 wt % carbon black, and 5.4 wt % binder (PVDF or PSBR). A mixture of carbon black and graphite were used as conductive additives in the LCO electrode. The 6391
DOI: 10.1021/acsami.6b14719 ACS Appl. Mater. Interfaces 2017, 9, 6390−6400
Research Article
ACS Applied Materials & Interfaces
Figure 2. (A) Schematic of the front and top view of the drag test setup. During a typical test, a drag tip with a width of 250 μm is brought in close proximity of the foil and plowed through the electrode at a speed of 100 μm/s. (B) Optical image showing the delamination of the electrode during the drag test. (C) SEM micrograph of the electrode after the drag test showing the delamination of the electrode from the foil. (D) Typical force recorded by the instrument during the drag test. capacities of the anode to cathode was ∼1.05 to 1.10, and the areal capacity of the battery was ∼2 mAh/cm2. The electrodes were heated in a vacuum oven connected to a glovebox at 130 °C for 12 h under vacuum to remove traces of moisture from the electrodes. After the drying process, the electrodes were transferred inside the glovebox. The anode and cathode were stacked together with a 25 μm trilayer polypropylene−polyethylene−polypropylene membrane (Celgard) and soaked with an electrolyte solution of 1 M LiPF6 in a 1:1 mixture of ethylene carbonate and dimethyl carbonate. The electrode stack was encapsulated in an aluminum-laminated pouch. The batteries were allowed to rest for a day before starting the electrochemical testing. The batteries were cycled at C/20 between 4.2 and 3.0 V for 3 cycles to activate the battery. The Coulombic efficiency in the first cycle was ∼85% and increased above 99.5% after the three formation cycles. The batteries were cycled using a battery cycler (MTI Corp.), and rate capabilities of the batteries were tested on a potentiostat (Ivium Technologies). Electrochemical impedance spectroscopy scans were performed in potentiostat mode with fluctuation voltage of 10 mV at frequencies ranging from 105 to 0.1 Hz.
required to remove the tape from the electrode. The electrode delaminates at the weakest interface in the electrode (Figure 1B). Figure 1C shows a photograph of the electrode during the peel test and the typical force recorded by the load cell during the test. The peel strength is represented in terms of the average force required to peel the tape normalized with the width of the tape. Under certain conditions, the peel strength (N/cm) alone may not be sufficient to characterize the mechanical strength of the film, especially for thick composite electrodes, which can delaminate at different interfaces within the electrode. In such conditions, the tape and electrode can be analyzed under a microscope to determine the interface of delamination (Figure 1D). The interface of delamination represents the interface that is likely to fail during flexing. The interface of delamination depends on the relative strength between the adhesion of the electrode to the foil, the interparticle cohesion, and the adhesion of the tape to the electrode. If the adhesion of the electrode to the foil is poor, the tape will remove the electrode completely. If the adhesion strength of the electrode is greater than the cohesion strength, the tape will remove a part of the electrode from the foil. If both the adhesion and cohesion strength of the electrode are greater than the adhesion of the tape to the electrode, then most of the electrode will remain on the foil. The scratch test is another useful test to measure the cohesion and adhesion strength of thin film structures. In a typical setup, a conical tip with a diameter of few micrometers is brought into contact with the film.48,49 The load in the normal direction is gradually increased, and the tip is dragged through the film in the horizontal direction. The instrument records the tangential and normal force on the tip and the distance traversed by the tip. In a conventional test, the tangential force on the tip when the film delaminates denotes the adhesion strength of the film, and the ratio of the tangential to the normal force (coefficient of friction) denotes the cohesion strength of the film. The scratch test is designed for thin film crystalline materials, and recent studies have shown that the scratch test does not provide coherent information when used on thick composite electrodes due to the large mismatch between the width of the tip and the average particle
■
RESULTS AND DISCUSSION A battery electrode consists of randomly oriented particles that are held together with a polymeric binder. The mechanical flexibility of a battery depends on the cohesion strength of the electrode and its adhesion to the current collector. Currently, there are no standard protocols to measure the adhesion and cohesion strength of composite battery electrodes. In this study, we leveraged two mechanical testing methods, the peel test and drag test, as a tool to investigate the mechanical strength of the electrodes. In a conventional tape test, an adhesive tape is pressed on the film and removed with a constant motion. The amount of material removed from the substrate provides qualitative information about the adhesion strength of the film. The peel test is a slight variation to the conventional tape test. In the peel test, the tape is removed with a universal testing machine.43−47 Figure 1A shows a schematic of the peel test setup. The electrode is clamped on one end of the Instron. A tape is firmly pressed on the electrode, and the free end of the tape is clamped to the moveable end of the Instron, which is connected to a 10 N load cell. The moveable end peels the tape at a constant speed, and the load cell records the force 6392
DOI: 10.1021/acsami.6b14719 ACS Appl. Mater. Interfaces 2017, 9, 6390−6400
Research Article
ACS Applied Materials & Interfaces
Figure 3. Conductivity (S/cm) of the LCO electrodes with 1−10 wt % of (A) carbon black and (B) graphite as the conductive additive, respectively. (C) Conductivity of LCO electrodes with a mixture of carbon black and graphite (6 wt % total) as the conductive additives.
size in the composite electrode.49 In a variation of the standard scratch test (drag test), we used a tip with a width of 250 μm, which is much wider than the average particle size in the electrode. Figure 2A shows a schematic of the top and side view of the drag test setup. During the test, the tip is kept at a constant distance from the foil and dragged through the electrode. Figure 2B shows an optical image of the electrode during the test. Figure 2C shows a SEM micrograph of the electrode after the drag test. The drag tip removes the electrode material completely, and the foil underneath the active layer is clearly visible. The instrument measures the horizontal force (drag force) on the tip (Figure 2D). The drag force denotes the amount of force required to break the particle-to-particle bonds and can be correlated to the cohesion strength of the electrode. The mechanical strength and the electrochemical performance of a battery depend on a complex interplay between the electrode thickness, porosity, composition, and particle shape.41,50 To simplify our study, the capacity and porosity of the electrodes were fixed to around 2 mAh/cm2 and 25%, respectively, and the resulting electrode had a thickness in the range of 60−70 μm. Thin stainless steel (12.7 μm) and nickel (10 μm) foils were used as the current collectors for the cathode and anode, respectively. The cathode is a mixture of active particles, conductive additives, and polymer binder to hold them together. LCO was selected as the active material for the cathode. The choice of binder for the cathode was limited to PVDF because the binder should be chemically inert and stable at high voltages. There were a number of options for the conductive additives, such as carbon nanotubes, carbon fiber, carbon black, graphite, and conducting polymers. Due to differences in particle shape, the enhancement in electronic conductivity for a particular loading depends on the type of conductive additive. For example, carbon nanotubes provide a higher improvement in the electronic conductivity in comparison to that of carbon black with the same loading.51 The difference in particle shape is also likely to change the physical interactions between the active particles and the carbon additives. By making a careful selection of the type of carbon additive, we can improve the mechanical flexibility of the electrodes while achieving the required electronic conductivity.
Two types of carbons (carbon black (C65) and graphite (SFG 6L)) were used in this study. C65 and SFG 6L have a large difference in their shape and size and served as a good starting material to study the relation between particle shape and mechanical strength of the electrodes. C65 particles are spherical in shape and have an average diameter of 20 nm, and the SFG 6L particles are plate shaped with an average width of 6 μm and thickness
