a high areal capacity flexible lithium-ion battery with a strain-compliant design

FULL P

APER

© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (1 of 11) 1401389wileyonlinelibrary.com

A High Areal Capacity Flexible Lithium-Ion Battery with a Strain-Compliant Design

Abhinav M. Gaikwad , Brian V. Khau , Greg Davies , Benjamin Hertzberg , Daniel A. Steingart , and Ana Claudia Arias*

Dr. A. M. Gaikwad, B. V. Khau, Prof. A. C. Arias Electrical Engineering and Computer Sciences Department University of California BerkeleyCory Hall Berkeley , CA 94720 , USA E-mail: [email protected] G. Davies, Dr. B. Hertzberg, Prof. D. A. Steingart Department of Mechanical and Aerospace Engineering and the Andlinger Center for Energy and The Environment Princeton University D428 Engineering Quadrangle Princeton , NJ 08544 , USA

DOI: 10.1002/aenm.201401389

to a bending radius of less than 1 cm without reaching its fracture limit (≈1% strain). [ 10,11 ] Electronics fabricated on soft polymeric substrate can conform over cur-vilinear surfaces and form the building blocks for wearable electronics. Powering these devices while maintaining the fl ex-ibility and form-factor of the device is a challenge. Towards this goal there have been numerous attempt to develop thin, fl exible and stretchable supercapaci-tors, [ 12–17 ] energy harvesters, [ 18–22 ] and bat-teries. [ 14,23–52 ] Due to their high energy and power density, fl exible rechargeable batteries are an essential part of a power module. [ 53 ] Commercially available bat-teries are typically rigid due to their pack-aging and thick electrode stack. Flexible batteries require that all of the key com-ponents (current collector, active layer, separator, packaging) be bendable. Such devices are currently fabricated by using a thin conductive layer supported on

a non-conductive substrate as a current collector, on which a thin active layer (20–60 µm) is printed. The areal capacities of fl exible lithium-ion batteries have been comparatively low (0.05–0.20 mAh cm −2 vs 1–2 mAh cm −2 for traditional systems) as the active layers are printed thin to reduce the electrode deg-radation during fl exing and the inactive layers associated with supporting the current collectors increase the thickness of the battery. [ 26,35,41,44,54 ]

Here we demonstrate a fl exible rechargeable lithium-ion battery with an areal capacity of ≈1 mAh cm –2 . We use lithium cobalt oxide (LCO) and lithium titanate oxide (LTO) as the posi-tive (cathode) and negative (anode) electrodes, to form a battery with a nominal potential of ≈2.5 V. [ 55–57 ] The fl exible battery with CNT as the current collector had neglible drop in capacity after electrochemically cycling the battery for 450 cycles at C/2 rate. The areal capacity (mAh cm −2 ) of the battery was increased by printing active layers as thick as 150 µm while improving their mechanical property by embedding the active layers inside a fi brous support. The fi brous membrane binds the active layers while carrying the stress associated with fl exing. The reinforced electrodes have a tensile strength of ≈5.5–7.0 MPa, an order of magnitude higher than conventional non-fl exible electrodes, making the electrode resistant to cracking and mechanical fatigue. The battery maintained capacity during electrochem-ical cycles under fl ex conditions and after undergoing repeated fl exing cycles. ElS was used to analyze the structural changes

Early demonstrations of wearable devices have driven interest in fl exible lithium-ion batteries. Previous demonstrations of fl exible lithium-ion batteries trade off between low areal capacity, poor mechanical fl exibility and/or high thickness of inactive components. Here, a reinforced electrode design is used to support the active layers of the battery and a freestanding carbon nanotube (CNT) layer is used as the current collector. The supported architecture helps to increase the areal capacity (mAh cm −2 ) of the battery and improve the tensile strength and mechanical fl exibility of the electrodes. Batteries based on lithium cobalt oxide and lithium titanate oxide shows excellent electro-chemical and mechanical performance. The battery has an areal capacity of ≈1 mAh cm −2 and a capacity retention of around 94% after cycling the battery for 450 cycles at a C/2 rate. The reinforced electrode has a tensile strength of ≈5.5–7.0 MPa and shows excellent capacity retention after repeat-edly fl exing to a bending radius ranging from 45 to 10 mm. The relationships between mechanical fl exing, electrochemical performance, and mechanical integrity of the battery are studied using electrochemical cycling, electron microscopy, and electrochemical impedance spectroscopy (EIS).

1. Introduction

Flexible batteries and fl exible devices have a common goal but different design challenges. Flexible electronics with a wide variety of functionalities such as displays, [ 1 ] skin-like pres-sure sensors, [ 2 ] health sensors, [ 3,4 ] light-emitting diodes, [ 5 ] solar cells [ 6 ] and fl exible circuits [ 7 ] have been demonstrated. Flexible electronics are typically fabricated by using solution processed organic semiconductors [ 8 ] and conductors or by pat-terning traditional inorganic components in ultrathin form (<100 nm). [ 9 ] This enables the electronic device to be fl exible

Adv. Energy Mater. 2014, 1401389

www.MaterialsViews.comwww.advenergymat.de

http://doi.wiley.com/10.1002/aenm.201401389

FULL

PAPER

© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim1401389 (2 of 11) wileyonlinelibrary.com

within the reinforced electrodes after repeatedly fl exing the battery. The EIS data suggests that bending the reinforced electrodes leads to an slight increase in the porosity of the elec-trodes causing no apparent capacity fade.

2. Results and Discussion

2.1. Flexible Current Collector

Current collector electronic conductivity is important to effi -cient charge transfer in a battery. [ 58 ] Commercial lithium-ion batteries rely on thin metallic foils of aluminum and copper to serve as current collectors for the positive and negative elec-trodes, respectively. Such foils (typically ≈15–20 µm) are stiff, and subject to fatigue failures and therefore cannot be used in fl exible batteries. [ 59 ] In recent reports, nickel-coated fabrics have been used as current collectors in fl exible lithium-ion and alkaline batteries. [ 35,40 ] Nickel was deposited on the fabrics using electroless deposition techniques to produce a conformal coating on the fabric. The low sheet resistance (≈0.35 Ω/�) and high fl exibility of nickel-coated fabrics makes them an excel-lent candidate for current collectors in fl exible batteries. But thicknesses of 100–300 µm lead to a signifi cant increase in the mass of the inactive components in the battery. Previously we used commercially available conductive inks such as silver (≈0.015 Ω/�) and carbon (≈50 Ω/�) to serve as current collec-tors for MnO 2 and zinc electrodes. [ 31,38,39 ] Due to the oxidation and dissolution of silver at potentials above 3 V, [ 30 ] silver cur-rent collectors cannot be used in rechargeable lithium-ion bat-teries, and the low conductivity of carbon leads to a signifi cant Ohmic potential drop during charging and discharging. Hu et al. demonstrated the use of CNT coated on paper to serve as current collectors for lithium-ion batteries. [ 14,28 ] The CNT layer had a sheet resistance of ≈5–10 Ω/� and the batteries did not show any signifi cant Ohmic potential drop during cycling. Recently, a similar approach was used to embed CNT ink inside of laboratory Kimwipes to serve as current collectors for fold-able and origami lithium-ion batteries. [ 44,54 ] Both approaches have relied on the use of a non-conductive substrate to support the CNT layer, which was followed by printing the active ink on top of the CNT layer. It would be benefi cial to remove the non-conductive support after printing the active layer to reduce the overall thickness of the battery. In this work, we used an air-brush to spray aqueous CNT ink on top of a stainless steel (SS) foil to form a thin CNT layer, which can then be transferred on top of a self-standing anodic/cathodic electrode ( Figure 1 ). Spray printing has been used previously to deposit CNT inks on substrates to serve as current collectors for batteries [ 36 ] or elec-trodes for supercapacitors. [ 60 ] Spray printing is a versatile tech-nique and can be easily modifi ed to print inks with wide range of viscosities over large areas. In this work, the inactive sup-port layer is removed during the transfer process, which helps to reduce the thickness of inactive layers within the battery. The aqueous CNT ink was prepared by mixing 2 mg mL −1 of CNT with 20 mg mL −1 of sodium dodecylbenzenesulfonate (SDBS) surfactant. [ 61 ] The fi nal thickness and loading of the CNT layer was ≈3 µm and ≈0.125 mg cm −2 , respectively. The CNT layer prepared with this method shows sheet resistance of ≈10 Ω/�.

2.2. Dip Coated Electrode and CNT Transfer

The electrochemically active layer in a battery is a mixture of active metal oxide, carbon (to improve the conductivity) and binder (to hold the mix together). [ 62 ] The active layer is brittle and can crack during fl exing. [ 26 ] Considerably less attention has been put towards improving the mechanical integrity of the active layer. Previous approaches have relied on using a fl exible current collector as a support for the active layer, and have limited the electrode thickness to 20–60 µm to reduce the fl ex induced strain and cracking. [ 26,28,41,44,54,63 ] The low thick-ness of the active layer reduces the areal capacity and the bat-tery requires a larger footprint to achieve the required capacity. Recent reports have proposed folding the battery multiple times to increase its areal capacity (mAh cm −2 ). But even after 4 folds (16 times thicker), the capacity of the battery increased only to 1.5 mAh cm −2 . [ 44,54 ] Additionally, folding the batteries multiple times would make the battery pack thicker, less fl ex-ible and potentially unsuitable for some applications. It would be preferable to increase the areal capacity of the battery by increasing the thickness of the active layer without degrading its mechanical integrity. In our previous work, we reported on fl exible primary alkaline batteries with an areal capacity of ≈6 mAh/cm 2 using a mesh design. [ 31 ] A similar concept was used to make stretchable alkaline batteries by embedding the active ink within fi bers of a silver coated stretchable fabric. [ 34 ] Here, we use similar principles and embedded the active layers of a lithium ion battery inside a porous membrane composed of non-woven fi bers. The membrane acts as a substrate and absorbs stresses generated during fl exing. We embedded the active material into the membrane by dipping it into an ink bath ( Figure 2 ). The ink is absorbed into the membrane by capillary action. The excess ink (ink not embedded inside the fabric) was removed with a doctor blade. After being coated, the membrane with was heated in an oven at 80 °C for two hours to dry it. The membrane with the LCO or LTO mix had a thick-ness of ≈150 µm. The loading of the LTO and LCO electrode was 13.2 and 11.6 mg cm −2 , respectively. Figure 2 shows optical images of the membranes after embedding them with the LCO and LTO mix. The supported electrodes were quite fl exible and can be fl exed repeatedly without any cracking or delamination of the active layer.



Figure 1. Spray coating carbon nanotube (CNT) ink on top of stainless steel (SS) foil with a thickness of ≈2–3 µm and sheet resistance of ≈10 Ω/�.

FULL P

APER


Before the CNT transfer process, the SS with CNT was cut into smaller squares (2.4 × 2.4 cm 2 ). Due to the poor adhe-sion between the CNT and SS, the CNT layer fl oats in water in ≈1 min ( Figure 3 ). The LCO or LTO electrode was pressed on top of the CNT layer and the LTO/CNT or LCO/CNT electrode was taken out of the water bath. The CNT layer adheres to the LCO and LTO electrode by Van der Waals interactions. Due to the low thickness of the CNT layer (≈3 µm), the CNT forms a conformal coating on top of the LTO/LCO electrode. The LCO/CNT and LTO/CNT electrodes are washed in fresh water 2–3 times to remove traces of surfactants. Figure 3 shows an optical image of the LTO and LCO electrode with the CNT cur-rent collector before sealing them in a pouch.

Figure 4 A shows a scanning electron microscopy (SEM) image of the fi brous non-woven membrane with a thickness of ≈120 µm. The membrane is made of chemically inert polyimide fi ber with diameter of ≈10 µm and is used as a separator in Ni-Cd batteries. Due to its open framework structure it serves as an excellent fl exible support for the active layer. Figure 4 B shows a SEM image of the membrane embedded with LCO mix with CNT current collector on top of the electrode. Due to the low thickness of the CNT layer (≈3 µm), the LCO electrode is visible through the CNT layer. The SEM shows an excellent con-formal coating of CNTs on top of the LCO electrode. The LCO particles have sizes ranging between 5–10 µm (Figure S1A, Supporting Information). Figure 4 C shows top-view of the LTO mix embedded within the membrane. The LTO particles have an average size of ≈0.5 µm (Figure S1B, Supporting Informa-tion). Figure 4 D shows a cross sectional micrograph of the membrane embedded with the LTO mix to give an electrode with thickness of ≈150 µm with CNT layer on top of the elec-trode to serve as the current collector. The SEM shows that the

active mix occupies all the open spaces in the membrane.

2.3. Mechanical Testing

An electrode for fl exible batteries should have high strength and retain its mechanical integ-rity after undergoing repeated fl exing. We study the mechanical stability of the reinforced elec-trodes by repeatedly fl exing them ( Figure 5 A), and observe the region of the electrode under stress using SEM. Figure 5 B,C shows the optical and SEM images of LCO and LTO elec-trodes, respectively, before and after fl exing them 300 times to a bending radius of 1 mm. SEM images of electrodes after fl exing show no cracking or delamination, demonstrating superior mechanical property of the rein-forced electrodes. Unsupported LCO and LTO electrodes (electrodes printed on thin metallic foils) delaminated and cracked within few fl exing cycles.

Breaking stress (maximum forced applied to the electrode before it ruptures) is a useful



Figure 2. Embedding the electrochemically active inks (“LCO” and “LTO” inks) within the sup-porting membrane by dipping inside an ink bath, followed by baking the electrode to remove residual solvent.

Figure 3. Carbon nanotube (CNT) transfer process. The stainless steel (SS) with the CNT layer is dipped in a water bath. The CNT layer delami-nates from the foil due to poor adhesion and fl oats in water after 1 min. The supported LCO and LTO electrodes are mechanically pressed on top of the CNT layer of 1 min and the LCO/CNT and LTO/CNT electrodes are taken out of the water bath and dried.

FULL

PAPER


fi gure of merit while comparing mechanical property of elec-trodes. An electrode with high breaking stress is desirable to reduce electrode rupture or cracking during fl exing. Zheng et al. measured the breaking strength of Li[Ni 1/3 Mn 1/3 Co 1/3 ]O 2 based composite cathodes after pressing them to porosity ranging from 0 to 50%. [ 64 ] The breaking stress of electrodes with 40–20% porosity was ≈0.35–0.80 MPa. We measured the breaking stress of the reinforced electrode using an Instron. The electrode (length = 40 mm, width = 15 mm) was stretched at a speed of 0.197 mm s −1 until the electrode broke. Figure 6 A–C shows the stress-strain curve for the unloaded membrane and the membrane embedded with LCO and LTO mix, respectively. The breaking stress of the unloaded membrane was ≈9.16 MPa (Figure 6 A) and ≈5.62 and ≈6.85 MPa for the reinforced LTO and LCO electrodes, respectively (Figure 6 B,C). The breaking stress of the reinforced electrodes was almost an order of magnitude higher than values reported for unsupported electrodes. [ 64,65 ] This indicates that the membrane supported is benefi cial for increasing the overall mechanical strength/fl exibility of the LCO and LTO composite electrodes.

2.4. Half Cell and Full Cell Testing

The LTO/CNT and LCO/CNT electrodes were tested using a half-cell confi guration with lithium metal as the counter elec-trode in a coin cell. Before carrying out the half-cell and full-cell tests, the cells were cycled for 2 cycles at C/20 rate (referred

as formation-cycle). The columbic effi ciency for the fi rst cycle was around 70–80% due to formation of solid-electrolyte-interphase layer and consumption of lithium due to side reac-tions with impurities in the electrodes. [ 66–68 ] The columbic effi ciency increased to around 97–99% for the 2 nd cycle. The columbic effi -ciency after the initial formation cycle was always above 99%. Figure 7 A,D show the charge and discharge curve for the LTO/CNT and LCO/CNT electrodes vs lithium metal, respectively. The LTO/CNT electrode was cycled between 1 and 2 V at C/4 rate vs Li/Li + and showed a plateau at ≈1.55 V vs Li/Li + . The LCO/CNT electrode was cycled between 3 and 4.2 V at C/4 rate vs Li/Li + . These charge/discharge curves match the results previously obtained for LTO and LCO elec-trodes using standard current collectors. [ 55,57 ] The charge/discharge curve showed no polarization effects, indicating suffi cient con-ductivity of the CNT current collectors. The LTO/CNT electrode demonstrated an initial specifi c capacity of ≈132 mAh g −1 , which decreased to ≈131 mAh g −1 after 80 charge/discharge cycles at C/4 rate, a capacity drop of less than 1% over this period (Figure 7 B). The LCO/CNT electrode demonstrated ini-tial specifi c capacity of ≈134 mAh g −1 , which decreased to 124 mAh g −1 after 80 charge/dis-charge cycles at C/4 rate, showing a capacity

retention of 92.5% (Figure 7 E). The columbic effi ciencies of both the cells were above 99%. The rate capabilities of both the LTO/CNT and LCO/CNT electrodes were studied by cycling the electrode at different C-rate (10 cycles each at C/8, C/4, C/2, C and C/8 rate). The LTO/CNT showed an initial specifi c capacity of ≈132 mAh g −1 at C/8 rate, which decreased to around ≈120 mAh g −1 at 1C rate, demonstrating a capacity reten-tion of 91% at 1C rate (Figure 7 C). The LCO/CNT electrode had relatively poor rate capability showing an initial capacity of ≈134 mAh g −1 , which decreased to ≈110 mAh g −1 and ≈59 mAh g −1 at C/2 and 1C rate, respectively (Figure 7 F). Hence the LCO/CNT electrode limits the rate capability of the full cell.

After studying the electrochemical performances of the LCO/CNT and LTO/CNT electrodes in half-cell confi gura-tion, full cells were made by sealing the electrodes inside an aluminum-laminated pouch ( Figure 8 A). The aluminum bar-rier layer reduces moisture permeation through the pouch. Nickel and aluminum tabs were used to make electric contact to the CNT current collectors. The total thickness of the battery was around 600 µm (including 250 µm for the battery pouch). Figure 8 B shows the charge/discharge capacity for the LTO-LCO full cell cycled over 450 cycles. The active area of the bat-tery was 2.5 × 2.5 cm 2 . The batteries demonstrated excellent capacity retention after 450 charge/discharge cycles. The initial areal capacity of the battery at C/2 rate was ≈0.94 mAh cm −2 and it decreased to ≈0.88 mAh cm −2 after 450 cycles, showing a capacity retention of ≈93% during this period. The excellent electrochemical stability of full cell demonstrates the stability



Figure 4. A) SEM image of the fi brous membrane. B) SEM image of the membrane embedded with LCO mix and CNT (current collector, ≈3 µm) on top of the electrode. C) SEM image of the LTO mix embedded inside the membrane. D) Cross-sectional SEM image of the membrane embedded with LTO mix with CNT (current collector) on top of the electrode. The electrode has a thickness of ≈150 µm and the CNT (current collector) has a thickness of ≈3 µm.

FULL P

APER


of the freestanding CNT based current collectors. The capacity retention and areal capacity (mAh cm −2 ) of the battery was sig-nifi cantly better than previous reports on fl exible lithium-ion battery with similar chemistry and is benefi cial while pow-ering devices with small footprint. [ 28,36,44,54 ] Figure 8C shows the rate capability of the battery. The battery was operated between C/4 to 1C rate for 150 cycles. The battery was able to retain above 75% of capacity at 1C rate. Figure 8D shows

charge–discharge curve of the full cell at various C-rates (C/4, C/2 and 1C). The specifi c capacity of the battery drop to 104 (89%) and 90 (77%) mAh g −1 at C/2 and 1C rates, respectively. The drop in capacity at high C-rates can be related to the low conductivity of the CNT, which leads to an Ohmic potential drop during charging and discharging. The Ohmic potential drop can be reduced by increasing the thickness of the CNT layer or by depositing a thin layer of metal on top of the CNT layer.



Figure 5. A) Mechanical fl exing of the LCO/LTO electrodes over a tube with a radius of 1 mm. SEM images and optical images of B) LCO and C) LTO electrodes after fl exing them 300 times to a bending radius of 1 mm, respectively. The dotted square shows the region of the electrode under stress during the fl exing cycle.

Figure 6. Stress-strain curves of A) the membrane, B) the LCO, and C) the LTO electrode with the membrane support.

FULL

PAPER


2.5. Performance Under Flexing

A battery while powering a fl exing device would be in a state of fl ex and undergo mechanical fatigue during its lifetime.

A fl exible battery in such an application should retain its capacity at different bending radius and have high resistance to fatigue. Currently, fl exible batteries are tested by fl exing them 20–100 times and the capacity of the battery before and after



Figure 7. Galvanostatic charge/discharge curves for A) LTO/CNT anode and D) LCO/CNT cathode cycled between 1.0–2.0 V vs Li/Li + and 3.0–4.2 V vs Li/Li + , respectively, at C/4 rate. Specifi c capacity (mAh g −1 ) and columbic effi ciency (%) vs cycle number for B) LTO/CNT anode and E) LCO/CNT cathode at C/4 rate for 80 cycles. Specifi c capacity (mAh g −1 ) and columbic effi ciency (%) vs cycle number for C) LTO/CNT anode and F) LCO/CNT cathode, respectively, cycled between C/8 and 1C rates.

Figure 8. A) Schematic of the fl exible battery laminated within aluminum-lined pouch. B) Areal capacity (mAh cm −2 ) of the full cell (LTO/LCO) cycled for 450 cycles at C/2 rate. The full cell had a total active area of 2.5 × 2.5 cm 2 and LTO and LCO loading of 13.4 and 12.5 mg cm −2 , respectively. C) Areal capacity (mAh cm −2 ) and columbic effi ciency (%) of the full cell cycled between C/4 and 1C rate. D) Galvanostatic charge/discharge curves for the full cell cycled between C/4 to 1C rate. The capacity at C/4, C/2 and 1C rate was 117, 104 and 90 mAh g −1 , respectively. Columbic effi ciency of all the batteries was above 99.5%.

FULL P

APER


fl exing are compared. [ 28,35,38,52,69 ] In earlier reports, the bat-teries were fl exed only once and the effects of mechanical fatigue on the batteries were not considered. [ 26,45 ] Herein, we study the effect of both fl exing and mechanical fatigue on the battery while it was cycled at C/2 rate for 95 cycles. The bat-tery was mechanically fl exed 100 times to a bending radii of 45, 30, 20 and 10 mm on the 19, 38, 57 and 76 th elec-trochemical cycle. After the battery was fl exed 100 times to a certain bending radius, the battery was kept at that bending radius until the next set of fl exing cycles. Figure 9 A shows the areal capacity of the battery fl exed to different bending radius. Overall, the capacity of the battery did not drop signifi cantly after fl exing. In standard electrochemical cells, fl exing can lead to delamination or cracking/swelling of the

active layers. Delamination of the active layers increase the con-tact resistance between the current collector and the active layer, leading to an Ohmic potential drop during the start of charge/discharge cycles. Cracking/swelling in the active layers can cause a loss in capacity due to loss of electrical contact between the particles and an increase in impedance within the active layers. Delamination/cracking of the active layers will cause a shift in charge/discharge profi les of the battery. Figure 9 B shows the charge/discharge curves of the battery before and after the fl exing 100 times to different bending radius. The comparisons of charge/discharge curves indicate that there is no polarization or loss in capacity relating to delamination or cracking of the active layers. The areal capacity of the battery before the fi rst fl exing cycle (18 th cycle) was 0.936 mAh cm −2



Figure 9. A) Areal capacity (mAh cm −2 ) of the full cell (LTO/LCO) under different state of fl ex ( R = ∞ mm to R = 10 mm). The battery was fl exed 100 times to a bend radii of 45, 30, 20 and 10 mm on the 19, 38, 57 and 76 th electrochemical cycle, respectively. B) Comparison of the Galvanostatic charge/discharge curves of the full cell, before and after fl exing the battery 100 times to a bending radius of 45, 30, 20 and 10 mm on the 19, 37, 57 and 76 th electrochemical cycles, respectively.

FULL

PAPER


and it decreased to 0.880 mAh cm −2 (80 th cycle) after fl exing the battery 400 times to bending radius ranging from 45 to 10 mm, a capacity drop of less than 6% during this period.

We further analyze the effect of mechanical fl exing on the structural changes within the reinforced electrodes using electrochemical impedance spectroscopy (EIS). EIS is a non-destructive technique and is used extensively to study the effect of cycle life, state of charge, electrode composition, and electrode design on the charge transfer/kinetic processes in a battery. [ 70–73 ] The EIS measurements were carried out at 50% depth of discharge at frequencies ranging from 10 6 –0.1 Hz with voltage fl uctuation of 5 mV from the open circuit voltage (OCV). The battery was allowed to equilibrate at OCV for a min-imum of 2–3 h before carrying out the EIS scans. EIS measure-ments were carried out on the cell undergoing fl exing cycles on the 1, 19, 38, 57, 76 and 95 th electrochemical cycle. On the 19, 38, 57 and 76 th electrochemical cycle, the battery was fl exed 100 times to a bending radius of 45, 30, 20 and 10 mm, respectively. EIS measurements were done before and after fl exing the cycles ( Figure 10 A). For comparison, EIS measurements were also carried out on a fl at battery on the 1, 19, 38, 57, 76 and 95 th elec-trochemical cycle (Figure 10 B). The EIS response had two semi-circle loops, one from high frequency to mid-frequency and the second from mid-frequency to low frequency and a small linear tail towards the end of the scan. The battery was mod-eled using equivalent circuit shown in Figure 10 C. [ 74,75 ] A repre-sentative fi t of the model to the experimental data is shown in Figure S2 (Supporting Information). The high frequency inter-cept on the x -axis corresponds to the total Ohmic resistance in the battery and is dominated by the resistance ( R elec ) of the elec-trolyte between the two current collectors. The fi rst depressed semicircle loop is due to multiple charge transfer interfaces

within the active layers, which includes the current collector/active layer interface ( R cc ), [ 76 ] active material/passivation layer interface ( R pas ) [ 75 ] and particle-to-particle interface ( R c ). [ 64 ] Since these processes have similar time constants and it is diffi cult to extract individual resistances. We denote total resistance for processes taking place from high frequency to mid frequency as R A ( R A = R cc + R pas + R c ) and the double layer capacitance as CA. The second depressed semicircle loop from mid to low frequency is related to the charge transfer processes between the active particle and the electrolyte ( R ct , C ct ) and the straight line towards the end of the curve is due to solid state diffusion of lithium ions within the electrode and is represented by the Warburg diffusion element ( W ).

The comparison of EIS scans for the fl at cell and the cell undergoing multiple fl exing cycle show that the resistance of the electrolyte ( R elec ) and the resistance of charge transfer ( R ct ) is almost constant with cycle number and fl exing cycles (Table S1 and S2, Supporting Information). The slight increase in R ct with cycle number is due to ageing of the electrodes. For the fl at cell, the fi rst semicircle loop increases initially and remains contact after certain number of cycles. The initial increase in R A is due to formation of a passivation layer on top of the active particles. The resistance is constant once a stable passivation layer is formed. For cell under going mechanical fl exing cycles, the fi rst semicircle loop increases after the battery undergoes mechanical fl exing cycle on the 19, 38, 57 and 75 th electrochem-ical cycle, which relates to an increase in the contact resistance ( R A ) within the electrode. As mentioned earlier R A includes, R cc , R pas and R c . A delamination or cracking of the passivation layer will lead to a decrease in the columbic effi ciency of the battery after fl exing. From Figure 9 A we observe that there is no drop in columbic effi ciency after fl exing. Hence, the increase



Figure 10. A) EIS scans on the full cell at 50% DOD on the 1, 19, 38, 57, 76 and 95 th electrochemical cycle. The cell was fl exed 100 times to a bending radii of 45, 30, 20 and 10 mm on the 19, 38, 57 and 76 th electrochemical cycle, respectively. EIS scan were carried out before and after fl exing the elec-trode. B) EIS scans on the full cell (LTO/LCO) at 50% depth of discharge (DOD) on the 1, 19, 38, 57, 76 and 95 th electrochemical cycle with the cell in the fl at state ( R = ∞ mm). C) Equivalent circuit to represent the full cell.

FULL P

APER

© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (9 of 11) 1401389wileyonlinelibrary.comAdv. Energy Mater. 2014, 1401389


in R A cannot be related to structural changes in the passiva-tion layer ( R pas ). From the author’s experience, delamination of the current collector from the active layer leads to a signifi cant increase in the Ohmic potential drop during the start of charge and discharge cycle. The comparison of the charge/discharge curves of the battery before and after fl exing (Figure 9 B) shows no Ohmic potential drop. Hence, the increase in R A cannot be related to delamination of the current collector ( R cc ). Zheng et al. showed that the radius of the fi rst semicircle loop increases with the porosity of the electrode. [ 64 ] Electrodes calendared to lower porosity have low contact resistance (R A ) . The increase in R A observed in our cells after fl exing can be related to an increase in the porosity of the electrode (R c ). [ 77 ] The electrochemical data suggests that the increase in porosity is not suffi cient high to cause a drop in capacity. A slight increase in porosity can be benefi cial to improve the transport of electrolyte through the electrode. A detailed analysis of the increase in porosity of the electrode with fl ex extent and micros-copy study of the electrode after fl exing is beyond the scope of this paper and these issues will be addressed in future publica-tions. For current cell design, the electrochemical and EIS data suggest that fl exing leads to a slight increase in the porosity of the electrode. But the increase in porosity is not suffi ciently high to lead to a drop in capacity of the battery.

Figure 11 A–C shows optical images of the fl exible lithium-ion battery connected to a voltmeter. The potential of the bat-tery was constant (2.617 V) after fl exing fi fty times to a bend radius of 4 mm. Figure 11 D and E show optical images of the

fl exible battery connected in series with a 200Ω resistor and a green light-emitting diode (LED). The battery powered the LED continuously after fl exing fi fty times to a bend radius of 4 mm. The brightness of the LED was constant during this period.

3. Conclusion

We have demonstrated a technique to fabricate fl exible recharge-able lithium-ion batteries with high areal capacity by printing thick active layers supported in a porous membrane. The mem-brane supports the active mix and prevents cracking during fl exing. The tensile strength of the electrode was an order of magnitude higher than standard electrodes. A freestanding CNT based current collector minimized the thickness of inactive com-ponents within the battery. The lithium cobalt oxide and lithium titanate oxide based batteries with CNT as current collector have excellent capacity retention and were cycled for more than 450 cycles with a nominal capacity of ≈1 mAh cm −2 and they were able to maintain their capacity after repeated fl exing to a bending radius of 10 mm. The areal capacity (mAh cm −2 ) of the battery with membrane support was 3–4 times higher than other reports on fl exible lithium-ion batteries with LTO and LCO as the active materials. [ 26,33,35,40,44 ] Batteries with high areal capacity and fl exibility are important for powering future generation of fl exible and mobile electronics devices. The technique demon-strated here can be easily used with other batteries chemistry with higher energy density and operating voltages. [ 27,35,41,44,54 ]

Figure 11. Optical images of the fl exible lithium-ion battery, connected to a voltmeter A) when fl at, B) after fl exing once, and C) after fl exing for 50 cycles to a bending radius of 4 mm, respectively. D) Demonstration of the fl exible lithium-ion battery connected in series with a 200 Ω resistor and a green light-emitting diode (LED). The battery was able to power the green-LED continuously even after fl exing 50 times to a bend radius of 4 mm (E).

FULL

PAPER

© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim1401389 (10 of 11) wileyonlinelibrary.com Adv. Energy Mater. 2014, 1401389


4. Experimental Section CNT Ink : Aqueous CNT ink was prepared by mixing 2 mg mL −1

of SWNT (Carbon Solutions Inc.) in D.I. water with 20 mg mL −1 of sodium dodecylbenzenesulfonate (SDBS, Sigma Aldrich) as surfactant to disperse the SWNT in water. The ink was sonicated using a Branson 450 digital sonifi er for 20 min at amplitude of 40% and at 30 W. The CNT ink was spray coated on a stainless steel (SS) foil using commercial available airbrush (Badger Model 150). 4 mL of 0.2 wt% CNT solutions was sprayed over 4.5 × 2.2 inch 2 with an air pressure of 25 PSI to give a CNT layer with thickness of 3 µm and weight of 0.125 mg cm −2 . During spraying, the SS substrate was heated at 140 °C.

LTO/LCO Electrode Fabrication : Lithium cobalt oxide (LCO, MTI Corp.) and lithium titanate oxide (LTO, MTI Corp.) slurries were prepared by mixing 77.5 wt% active material with 12.5 wt% carbon black (Super P, TIMCAL) and 10 wt% polyvinylidene fl uoride (PVDF, Kureha Corp.) binder in N -methyl-2-pyrrolidone (NMP, Sigma Aldrich) as the solvent. The slurries were homogenized by stirring overnight using a vortex mixer. The inks were embedded inside a nonwoven fi brous membrane (Freudenberg, Germany) by dipping the membrane inside an ink bath for 1 min. The electrodes were by dried by heating in an oven at 80 °C for two hours. The solvent fraction in the inks was optimized to help with the embedding process. Every 1 g of LTO and LCO dry mix was mixed with 1.70 and 1.53 g of NMP, respectively.

CNT Transfer : The SS with a spray coated CNT layer was cut into smaller squares (2.4 × 2.4 cm 2 ) and then dipped in a water bath. Due to the poor adhesion of the CNT on SS, the CNT layer delaminates from SS in ≈1 min. The LCO/LTO electrode was mechanically pressed on top of the CNT layer for 1 min before taking the electrode out of the water bath. The LTO/CNT or LCO/CNT electrode was dipped in fresh watch bath 2–3 times to remove traces of surfactant from the CNT layer.

Fabrication and Electrochemical Testing of Half and Full Cell : Half cells for testing LTO/CNT and LCO/CNT electrodes were prepared by making coin cells with lithium metal electrodes as the counter electrode. The lithium metal and the LTO/CNT or LCO/CNT electrode were punched in round shapes with diameters of 0.5 inch. The coin cell parts were purchased from MTI Corp. The full cells were prepared by stacking LTO/CNT and LCO/CNT electrode (2.5 × 2.5 cm 2 ) with polypropylene based separator (25 µm, Celgard) and sealing them inside aluminum-lined battery pouch (Sigma Aldrich). 1 M solution of LiPF 6 in EC/DEC (1:1) was used as the electrolyte. Before assembling the coin cells and full cells, the electrodes were heated in a vacuum oven connected to the glove box at 140 °C under vacuum for 12 h to remove traces of solvent from the electrode. The half and full cells were tested using MTI battery analyzer. SEM microscopy was carried out on TM-1000 (Hitachi) and FEI Quanta. The EIS was carried out using a Gamry Potentiostat, at open circuit potential at frequencies ranging from 10 6 to 0.1 Hz at a amplitude of 5 mV. The mechanical testing of the supported membrane and the electrodes were carried out using an Instron. The electrode (length = 40 mm, width = 15 mm) was stretched at a speed of 0.197 mm s −1 till the electrode broke.

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.

Acknowledgements The authors thank Rich Winslow, Martin Cowell, Alla Zamarayeva, and Yasser Khan for many helpful discussions and Prof. Paul Wright and James Evans for giving access to equipment in their laboratories. This work was partially supported in part by the National Science

Foundation under Cooperative Agreement No. ECCS-1202189 and CBET 1318163.

Received: August 11, 2014 Revised: August 29, 2014

Published online:

[1] J. A. Rogers , Z. Bao , K. Baldwin , A. Dodabalapur , B. Crone , V. R. Raju , V. Kuck , H. Katz , K. Amundson , J. Ewing , P. Drzaic , Proc. Natl. Acad. Sci. USA 2001 , 98 , 4835 .

[2] T. Sekitani , S. Iba , Y. Kato , Y. Noguchi , T. Someya , Appl. Phys. Lett. 2005 , 87 , 3 .

[3] D.-H. Kim , N. Lu , R. Ghaffari , Y.-S. Kim , S. P. Lee , L. Xu , J. Wu , R.-H. Kim , J. Song , Z. Liu , J. Viventi , B. de Graff , B. Elolampi , M. Mansour , M. J. Slepian , S. Hwang , J. D. Moss , S.-M. Won , Y. Huang , B. Litt , J. a Rogers , Nat. Mater. 2011 , 10 , 316 .

[4] D.-H. Kim , N. Lu , R. Ma , Y.-S. Kim , R.-H. Kim , S. Wang , J. Wu , S. M. Won , H. Tao , A. Islam , K. J. Yu , T. Kim , R. Chowdhury , M. Ying , L. Xu , M. Li , H.-J. Chung , H. Keum , M. McCormick , P. Liu , Y.-W. Zhang , F. G. Omenetto , Y. Huang , T. Coleman , J. a Rogers , Science 2011 , 333 , 838 .

[5] M. S. White , M. Kaltenbrunner , E. D. Głowacki , K. Gutnichenko , G. Kettlgruber , I. Graz , S. Aazou , C. Ulbricht , D. a. M. Egbe , M. C. Miron , Z. Major , M. C. Scharber , T. Sekitani , T. Someya , S. Bauer , N. S. Sariciftci , Nat. Photonics 2013 , 7 , 811 .

[6] M. Kaltenbrunner , M. S. White , E. D. Głowacki , T. Sekitani , T. Someya , N. S. Sariciftci , S. Bauer , Nat. Commun. 2012 , 3 , 770 .

[7] K. Takei , T. Takahashi , J. C. Ho , H. Ko , A. G. Gillies , P. W. Leu , R. S. Fearing , A. Javey , Nat. Mater. 2010 , 9 , 821 .

[8] A. C. Arias , J. D. MacKenzie , I. McCulloch , J. Rivnay , A. Salleo , Chem. Rev. 2010 , 110 , 3 .

[9] Y. Sun , J. Rogers , Adv. Mater. 2007 , 19 , 1897 . [10] D.-H. Kim , Y.-S. Kim , J. Wu , Z. Liu , J. Song , H.-S. Kim , Y. Y. Huang ,

K.-C. Hwang , J. A. Rogers , Adv. Mater. 2009 , 21 , 3703 . [11] J. a Rogers , T. Someya , Y. Huang , Science 2010 , 327 , 1603 . [12] V. L. Pushparaj , M. M. Shaijumon , A. Kumar , S. Murugesan , L. Ci ,

R. Vajtai , R. J. Linhardt , O. Nalamasu , P. M. Ajayan , Proc. Natl. Acad. Sci. USA 2007 , 104 , 13574 .

[13] L. Hu , M. Pasta , F. La Mantia , L. Cui , S. Jeong , H. D. Deshazer , J. W. Choi , S. M. Han , Y. Cui , Nano Lett. 2010 , 10 , 708 .

[14] L. Hu , J. W. Choi , Y. Yang , S. Jeong , F. La Mantia , L.-F. Cui , Y. Cui , Proc. Natl. Acad. Sci. USA 2009 , 106 , 21490 .

[15] C. Meng , C. Liu , L. Chen , C. Hu , S. Fan , Nano Lett. 2010 , 10 , 4025 . [16] K. Jost , C. R. Perez , J. K. McDonough , V. Presser , M. Heon , G. Dion ,

Y. Gogotsi , Energy Environ. Sci. 2011 , 4 , 5060 . [17] Z. Niu , H. Dong , B. Zhu , J. Li , H. H. Hng , W. Zhou , X. Chen , S. Xie ,

Adv. Mater. 2013 , 25 , 1058 . [18] S. Roundy , E. S. Leland , J. Baker , E. Carleton , E. Reilly , E. Lai , B. Otis ,

J. M. Rabaey , V. Sundararajan , P. K. Wright , IEEE Pervasive Comput. 2005 , 4 , 28 .

[19] D. Steingart , C. Berkeley , S. Roundy , L. V Sensors , K. Paul , MRS Bull. 2008 , 33 , 408 .

[20] S. Saadon , O. Sidek , Energy Convers. Manag. 2011 , 52 , 500 . [21] Z. L. Wang , Adv. Mater. 2012 , 24 , 4632 . [22] C. Dagdeviren , B. D. Yang , Y. Su , P. L. Tran , P. Joe , E. Anderson ,

J. Xia , V. Doraiswamy , B. Dehdashti , X. Feng , B. Lu , R. Poston , Z. Khalpey , R. Ghaffari , Y. Huang , M. J. Slepian , J. a Rogers , Proc. Natl. Acad. Sci. USA 2014 , 111 , 1927 .

[23] V. L. Pushparaj , M. M. Shaijumon , A. Kumar , S. Murugesan , L. Ci , R. Vajtai , R. J. Linhardt , O. Nalamasu , P. M. Ajayan , Proc. Natl. Acad. Sci. USA 2007 , 104 , 13574 .

[24] H. Gwon , H.-S. Kim , K. U. Lee , D.-H. Seo , Y. C. Park , Y.-S. Lee , B. T. Ahn , K. Kang , Energy Environ. Sci. 2011 , 4 , 1277 .

FULL P

APER

© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (11 of 11) 1401389wileyonlinelibrary.comAdv. Energy Mater. 2014, 1401389


[25] H. Nishide , K. Oyaizu , Science 2008 , 319 , 737 . [26] P. Hiralal , S. Imaizumi , H. E. Unalan , H. Matsumoto , M. Minagawa ,

M. Rouvala , A. Tanioka , G. a J. Amaratunga , ACS Nano 2010 , 4 , 2730 . [27] C. C. Ho , J. W. Evans , P. K. Wright , J. Micromech. Microeng. 2010 ,

20 , 104009 . [28] L. Hu , H. Wu , F. La Mantia , Y. Yang , Y. Cui , ACS Nano 2010 , 4 , 5843 . [29] M. Kaltenbrunner , G. Kettlgruber , C. Siket , R. Schwödiauer , S. Bauer ,

Adv. Mater. 2010 , 22 , 2065 . [30] A. M. Gaikwad , J. W. Gallaway , D. Desai , D. a. Steingart , J. Electro-

chem. Soc. 2011 , 158 , 154 . [31] A. M. Gaikwad , G. L. Whiting , D. a Steingart , A. C. Arias , Adv. Mater.

2011 , 23 , 3251 . [32] L. Hu , F. La Mantia , H. Wu , X. Xie , J. McDonough , M. Pasta , Y. Cui ,

Adv. Energy Mater. 2011 , 1 , 1012 . [33] K. T. Braam , S. K. Volkman , V. Subramanian , J. Power Sources 2012 ,

199 , 367 . [34] A. M. Gaikwad , A. M. Zamarayeva , J. Rousseau , H. Chu , I. Derin ,

D. A. Steingart , Adv. Mater. 2012 , 24 , 5071 . [35] M. Koo , K.-I. Park , S. H. Lee , M. Suh , D. Y. Jeon , J. W. Choi , K. Kang ,

K. J. Lee , Nano Lett. 2012 , 12 , 4810 . [36] N. Singh , C. Galande , A. Miranda , A. Mathkar , W. Gao ,

A. L. M. Reddy , A. Vlad , P. M. Ajayan , Sci. Rep. 2012 , 2 , 481 . [37] C. Huang , Q. Zhang , K. Bhattacharya , Appl. Phys. Lett. 2004 , 4391 . [38] A. M. Gaikwad , H. N. Chu , R. Qeraj , A. M. Zamarayeva ,

D. a. Steingart , Energy Technol. 2013 , 1 , 177 . [39] A. M. Gaikwad , D. a. Steingart , T. Nga Ng , D. E. Schwartz ,

G. L. Whiting , Appl. Phys. Lett. 2013 , 102 , 233302 . [40] G. Kettlgruber , M. Kaltenbrunner , C. M. Siket , R. Moser , I. M. Graz ,

R. Schwödiauer , S. Bauer , J. Mater. Chem. A 2013 , 1 , 5505 . [41] Y.-H. Lee , J.-S. Kim , J. Noh , I. Lee , H. J. Kim , S. Choi , J. Seo , S. Jeon ,

T.-S. Kim , J.-Y. Lee , J. W. Choi , Nano Lett. 2013 , 13 , 5753 . [42] S. Xu , Y. Zhang , J. Cho , J. Lee , X. Huang , L. Jia , J. a Fan , Y. Su , J. Su ,

H. Zhang , H. Cheng , B. Lu , C. Yu , C. Chuang , T.-I. Kim , T. Song , K. Shigeta , S. Kang , C. Dagdeviren , I. Petrov , P. V Braun , Y. Huang , U. Paik , J. a Rogers , Nat. Commun. 2013 , 4 , 1543 .

[43] C. Yan , X. Wang , M. Cui , J. Wang , W. Kang , C. Y. Foo , P. S. Lee , Adv. Energy Mater. 2014 , 4 , 1301396 .

[44] Z. Song , T. Ma , R. Tang , Q. Cheng , X. Wang , D. Krishnaraju , R. Panat , C. K. Chan , H. Yu , H. Jiang , Nat. Commun. 2014 , 5 , 3140 .

[45] Z. Wang , Z. Wu , N. Bramnik , S. Mitra , Adv. Mater. 2014 , 26 , 970 . [46] Z. Wang , A. Chen , R. Winslow , D. Madan , R. C. Juang , M. Nill ,

J. W. Evans , P. K. Wright , J. Micromech. Microeng. 2012 , 22 , 094001 .

[47] E. Madej , M. Espig , R. R. Baumann , W. Schuhmann , F. La Mantia , J. Power Sources 2014 , 261 , 356 .

[48] E.-H. Kil , K.-H. Choi , H.-J. Ha , S. Xu , J. a Rogers , M. R. Kim , Y.-G. Lee , K. M. Kim , K. Y. Cho , S.-Y. Lee , Adv. Mater. 2013 , 25 , 1395 .

[49] Y. H. Kwon , S.-W. Woo , H.-R. Jung , H. K. Yu , K. Kim , B. H. Oh , S. Ahn , S.-Y. Lee , S.-W. Song , J. Cho , H.-C. Shin , J. Y. Kim , Adv. Mater. 2012 , 24 , 5192 .

[50] X. Jia , Z. Chen , A. Suwarnasarn , L. Rice , X. Wang , H. Sohn , Q. Zhang , B. M. Wu , F. Wei , Y. Lu , Energy Environ. Sci. 2012 , 5 , 6845 .

[51] B. Liu , J. Zhang , X. Wang , G. Chen , D. Chen , C. Zhou , G. Shen , Nano Lett. 2012 , 12 , 3005 .

[52] B. Liu , X. Wang , H. Chen , Z. Wang , D. Chen , Y.-B. Cheng , C. Zhou , G. Shen , Sci. Rep. 2013 , 3 , 1622 .

[53] J. M. Tarascon , M. Armand , Nature 2001 , 414 , 359 . [54] Q. Cheng , Z. Song , T. Ma , B. B. Smith , R. Tang , H. Yu , H. Jiang ,

C. K. Chan , Nano Lett. 2013 , 13 , 4969 . [55] S. H. R. J. Gummow , M. M. Thackeray , W. I. F. David , Mater. Res.

Bull. 1992 , 27 , 327 . [56] J. Bates , N. Dudney , B. Neudecker , A. Ueda , C. Evans , Solid State

Ionics 2000 , 135 , 33 . [57] S. Lee , P. Liu , C. Tracy , D. Benson , Electrochem. Solid-State Lett.

1999 , 2 , 425 . [58] S.-T. Myung , Y. Hitoshi , Y.-K. Sun , J. Mater. Chem. 2011 , 21 , 9891 . [59] J.-Y. Choi , D. J. Lee , Y. M. Lee , Y.-G. Lee , K. M. Kim , J.-K. Park ,

K. Y. Cho , Adv. Funct. Mater. 2013 , 23 , 2108 . [60] X. Zhao , B. T. T. Chu , B. Ballesteros , W. Wang , C. Johnston ,

J. M. Sykes , P. S. Grant , Nanotechnology 2009 , 20 , 065605 . [61] M. F. Islam , E. Rojas , D. M. Bergey , a. T. Johnson , a. G. Yodh , Nano

Lett. 2003 , 3 , 269 . [62] T. Reddy , Linden’s Handbook of Batteries , 4 th Edition , McGraw-Hill,

New York 2010 . [63] M. Koo , K.-I. Park , S. H. Lee , M. Suh , D. Y. Jeon , J. W. Choi , K. Kang ,

K. J. Lee , Nano Lett. 2012 , 12 , 4810 . [64] H. Zheng , L. Tan , G. Liu , X. Song , V. S. Battaglia , J. Power Sources

2012 , 208 , 52 . [65] Y. Wu , H. Wu , S. Luo , K. Wang , F. Zhao , Y. Wei , P. Liu , K. Jiang ,

J. Wang , S. Fan , RSC Adv. 2014 , 4 , 20010 . [66] D. Aurbach , M. Levi , E. Levi , J. Electrochem. Soc. 1998 , 145 , 3024 . [67] D. Aurbach , Y. Talyosef , B. Markovsky , E. Markevich , E. Zinigrad ,

L. Asraf , J. S. Gnanaraj , H.-J. Kim , Electrochim. Acta 2004 , 50 , 247 . [68] J. Vetter , P. Novák , M. R. Wagner , C. Veit , K.-C. Möller ,

J. O. Besenhard , M. Winter , M. Wohlfahrt-Mehrens , C. Vogler , a. Hammouche , J. Power Sources 2005 , 147 , 269 .

[69] N. Li , Z. Chen , W. Ren , F. Li , H.-M. Cheng , Proc. Natl. Acad. Sci. USA 2012 , 109 , 17360 .

[70] N. Schweikert , R. Heinzmann , A. Eichhöfer , H. Hahn , S. Indris , Solid State Ionics 2012 , 226 , 15 .

[71] D. Zhang , B. Haran , A. Durairajan , R. White , Y. Podrazhansky , B. Popov , J. Power Sources 2000 , 91 , 122 .

[72] S. Zhang , K. Xu , T. Jow , Electrochim. Acta 2004 , 49 , 1057 . [73] F. Huet , J. Power Sources 1998 , 70 , 59 . [74] S. S. Zhang , K. Xu , T. R. Jow , Electrochim. Acta 2006 , 51 , 1636 . [75] S. S. Zhang , K. Xu , T. R. Jow , Electrochim. Acta 2004 , 49 , 1057 . [76] M. Gaberscek , J. Moskon , B. Erjavec , R. Dominko , J. Jamnik , Elec-

trochem. Solid-State Lett. 2008 , 11 , A170 . [77] A. Van Bommel , R. Divigalpitiya , J. Electrochem. Soc. 2012 , 159 ,

A1791 .

a high areal capacity flexible lithium-ion battery with a strain-compliant design

Documents