Transparent Thin Film Solid-State Lithium Ion Batteries - ACS Applied

Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces 2019, 11, 683−690

Transparent Thin Film Solid-State Lithium Ion Batteries Sami Oukassi,* Loic Baggetto, Christophe Dubarry, Lucie Le Van-Jodin, Séverine Poncet, and Raphaël Salot Univ. Grenoble Alpes, CEA, LETI, MINATEC Campus, 38054 Grenoble, France

ACS Appl. Mater. Interfaces 2019.11:683-690. Downloaded from pubs.acs.org by UNITED ARAB EMIRATES UNIV on 01/10/19. For personal use only.

S Supporting Information *

ABSTRACT: Transparent electrochemical energy storage devices have attracted extensive attention for the power supply of next-generation transparent electronics. In this paper, semitransparent thin film batteries (TFBs) with a grid-structured design have been fabricated on glass substrates using specific photolithography and etching processes to achieve LiCoO2/ LiPON/Si structures below human eye resolution. UV−vis transmittances up to 60% have been measured for the obtained TFBs. A discharge capacity as high as 0.15 mAh has been recorded upon galvanostatic cycling at the C/2 rate within 4.2−3 V voltage range for the highest transmittances. The capacity variation trend exhibits an initial phase of a gradual decrease with an average capacity loss of 0.15% per cycle and thereafter a second phase with almost stable capacity. Particular attention has been given to the effects of architecture parameters on the TFB optical and electrochemical properties. To the best of our knowledge, this work is the first demonstration of transparent, all inorganic, thin film lithium ion batteries. While reported studies are limited to battery structures involving liquid or polymer materials, our devices will contribute to improve form factor freedom, extend operating ranges, enhance long-term stability, and will be relevant to the integration into various optoelectronic devices. KEYWORDS: transparent battery, thin film, solid state, LiPON, microfabrication

1. INTRODUCTION Novel multifunctional electronic devices related to the Internet of Things (IoT) and wearables are becoming increasingly more important in our daily life. These miniaturized electronic devices no longer require a conventional, bulk, energy storage device. Instead, a power source that can be perfectly integrated within the overall architecture and that can ensure an additional structural function has become crucially needed.1,2 Within this context, transparency, among other functions, presents a particular challenge. Lithium ion batteries are considered as one of the most promising candidates to meet a wide range of applications’ demands since they offer high energy density, are lightweight, and can provide a long lifetime. Despite the potential advantages of integrating such energy storage with windows, screens, glasses, and optoelectronic devices, to date only a few studies have been reported on transparent lithium ion batteries. Because most of the active battery materials are optically absorbent, there are mainly two basic approaches to achieve this aim. The first approach is based on reducing the length scale of active materials and minimizing electrode thickness,3−5 in addition to the development of transparent polymer electrolytes.6,7 This approach has proven effective, and examples of transparent batteries have been reported in the literature.8 Electrochemical performance is limited, however, and methods of enhancing energy density are required to further progress in this regard. © 2018 American Chemical Society

The second approach is based on a completely different concept of gridlike electrodes obtained by applying structuring techniques. In this method, batteries appear transparent for a low filling ratio and grid features dimensions below the resolution of the human eye. Electrode structuring has already been applied successfully to the realization of transparent electrochemical storage devices, mainly supercapacitors.9−12 This approach has been so far only modestly explored for the fabrication of transparent batteries. Yang et al. were the first to report on the preparation of a fully transparent lithium ion battery integrating gridlike overlaid electrodes separated by a transparent gel electrolyte.13 Despite achieving a transmittance as high as 60% for a 10 Wh/L full packaged cell, long life cycle performance was not reported, and the microfluidics assisted method in this work presents upscaling and integration difficulties, thus limiting further applications. Herein, we report on the development of transparent all solid inorganic thin film lithium ion batteries (TFBs). Our transparent TFBs are fabricated using advanced microfabrication techniques employed in the microelectronic industry. The obtained experimental results demonstrate that our devices could enhance significantly the transparent Received: September 19, 2018 Accepted: December 10, 2018 Published: December 10, 2018 683

DOI: 10.1021/acsami.8b16364 ACS Appl. Mater. Interfaces 2019, 11, 683−690

Research Article

ACS Applied Materials & Interfaces Table 1. Deposition Conditions for the Transparent TFB Layers gas flow rate (sccm) film

target

Ti Pt LiCoO2 LiPON Si

Ti Pt LiCoO2 Li3PO4 Si

powera (W cm−2) 2.7, 0.5, 2.3, 1.8, 1.3,

dc dc dc rf dc

pressure (Pa)

Ar

1 1.4 2 0.6 0.7

45 70 130

O2

N2

100 100

deposition rate (nm min−1) 33 42 70 11 24

a

The rf and dc power sources are respectively for radio frequency and direct current.

Figure 1. Transparent thin film batteries (TFB): (a) schematic illustration and (b) photograph of a realized 25 × 25 mm2 single device. For simplicity, the Ti adhesion layer beneath Pt is omitted in the schematic. performed using laser ablation process (Coherent GEM 100-CO2 laser with a wavelength of λ = 10.6 μm, an average power of 20 W, and a speed of 2 mm/s) to obtain 25 × 25 mm2 single TFB devices. 2.3. Physical and Electrochemical Characterization. TFB top and cross section were imaged using a high-resolution field emission scanning electron microscope (FE-SEM) coupled with focused ion beam (FIB) (HELIOS 450 S). A layer of Pt was locally deposited at the top of the sample to avoid damages during FIB milling. The TFB stack was chemically characterized with a time-of-flight scanning ion mass spectrometer (ToF-SIMS). The samples were profiled with an ION-TOF ToF-SIMS V instrument, using a 25 keV Bi3+ analysis beam and a 2 keV Cs+ sputtering beam over rastered areas of 200 × 200 μm2. The sample temperature was maintained below −100 °C to avoid Li+ ion displacements. The beams were operated in noninterlaced mode, alternating analysis cycle and sputtering followed by a pause of 0.5 s. During the pause time, an electron flood gun was used for charge compensation. Positive secondary ions were analyzed and allowed characteristic ions to be monitored during depth profiling. The elemental distribution of Li, O, P, N, Si, and Ti within the transparent TFB is visualized by 7LiCs2+, 16 OCs2+, 31PCs2+, 14NCs2+, 28SiCs2+, and 48TiCs2+ secondary ions, normalized by Cs2+ intensity. Electrochemical measurements were performed in an Ar glovebox at 25 °C, using a VMP3 galvanostat

batteries performance and widen the scope of possible form factor and integration schemes.

2. EXPERIMENTAL SECTION 2.1. Materials. Film deposition steps were performed in an ENDURA PVD 200 mm tool (Applied Materials) equipped with Ti, LiCoO2, Li3PO4, and Si sputtering targets and an ALCATEL SCM600 tool equipped with a Pt sputtering target. Both tools were connected to Ar-filled gloveboxes to control the samples transfer and storage atmosphere. Glass wafers (Schott AF32, 200 mm diameter) were used as substrates. The details of all deposition conditions are listed in Table 1. 2.2. Transparent TFB Fabrication Process. Transparent TFBs were prepared by successive deposition and patterning of each layer level using UV photolithography and etching (Figure S1). SPR 220 photoresist was used for all photolithography steps, and UV exposure was performed in an EVG 640 mask aligner. Wet etching of LiCoO2 and LiPON was realized in 1H2SO4:1H2O solution and Pt in 1HCl:3HNO3 solution. Si/Ti top layers were etched in a CHF3/Ar plasma using a Corial IL200 tool. All the active layers were patterned following a gridlike structure except LiPON, which was etched only in dicing streets (TFB edges, Figure S1f). A singulation step was 684


Research Article

ACS Applied Materials & Interfaces potentiostat (Bio-Logic). Cycling was performed between 3 and 4.2 V at a typical rate of C/2. Unless stated otherwise, potentials written versus Li correspond to potentials measured against the Li/Li+ redox couple.

Moreover, the thicknesses match very well with the intended values. Each film appears distinct and continuous, and the interfaces do not exhibit delamination or local blisters, which indicates good adhesion at the various interfaces. 3.2. Functional Properties of the Transparent TFB Devices. 3.2.1. Architecture Optical Properties. The UV−vis transmittance spectra of all TFB designs are depicted in Figure 3a. The absorption observed below 400 nm is related to substrate properties. For higher wavelengths the spectra are relatively flat across the visible region, the observed oscillations in the transmittance are likely due to interference effects within the glass/LiPON open area.14−16 Depending on TFB design (Table 2), average transmittances between 17% and 60% are recorded over the 400−2000 nm wavelength range. Furthermore, the transmittance at 800 nm was extracted and compared to a calculated geometric transmittance Tg for each TFB design obtained following the relation Tg,800 = Tglass/LiPON((X1/(X1 + X2))2, where Tglass/LiPON corresponds to the experimental transmittance measured at 800 nm for a glass/LiPON structure. In our case, Tglass/LiPON is equal to 85% (Figure S2); the detailed study of the LiPON film optical characterization is reported elsewhere.14 Tglass/LiPON value is related to substrate reflection rather than layer absorption, as seen in Figure S2. Furthermore, previous work15,16 reported LiPON refractive index n between 1.7 and 1.9 and an extinction coefficient k close to zero for wavelength above 300 nm, the latter resulting in almost no absorption within this spectral range. As illustrated in Figure 3b, an almost linear correlation exists between measured and calculated transmittances with a correlation coefficient of 0.997. This result strongly suggests that TFB transmittance is controlled by only the architecture design and substantiates that LiPON optical properties were unaltered after microfabrication process steps. Haze, the ratio of diffusive transmittance (the difference between total and specular transmittance) to total transmittance, is another relevant parameter for transparent devices in general and gridlike structures in particular.17−19 In this work, the surface irregularities consisting of the opaque TFB structures with different geometric designs may impact the likelihood of light scattering. Results of haze variation with wavelength for G-type TFB are presented in Figure 4a. The measured haze is around 3% and stable up to 2 μm wavelength. Similar haze levels have been reported for devices based on metal grid structures,17,19 thus indicating a low scattering and a transparent appearance of our TFBs, satisfying requirements for the considered applications. Figure 4b shows the measured specular transmittance of Gtype TFB as a function of incident angle for unpolarized light at different wavelengths. The angular variation of transmittance is determined analytically either using the Fresnel expression for the reflection of unpolarized radiation passing between media with different reflective indices or using simplified expressions such as Duffie and Beckman’s approximate method.20 In this work, TFB transmittance variation with incident angle is almost similar to glass samples, showing a continuous decrease when increasing incident angle for all the considered wavelengths. 3.2.2. Transparent TFB Electrochemical Properties. The storage capacity available for the positive LiCoO2 electrode is higher than the one of the negative Si electrode (respectively 125 and 75 μAh cm−2) to lower Si thickness and avoid fracture during cycling.21,22 Consequently, both Si lithiation/delithia-

3. RESULTS AND DISCUSSION 3.1. Transparent Thin Film Battery Concept. A schematic representation of the transparent TFB is presented in Figure 1. The transparent TFB is built by successive deposition and etching of the active materials, starting from the Ti/Pt bottom current collector, LiCoO2 positive electrode material, LiPON solid-state electrolyte, Si negative electrode, and Ti top current collector (Figure 1a). The architecture is built upon a square gridlike array with tunable line spacing and width (X1 and X2, Figure 1a) to control the optical transmittance through the hollow areas. Hence, the realized demonstrators show partial transparency, as depicted in Figure 1b when the device is placed in front of a landscape visible in the background. An example of the device microstructure is presented in Figure 2a. The grid structure offers large empty spaces in the

Figure 2. (a) Top grid structure and (b) cross-section characterizations of transparent TFBs using SEM (left-hand side) and elemental EDS mapping (right-hand side). Layers of interest and elements are indicated in the figures.

center of the connected nodes for optical transmission purpose (Figure 2a). The elemental analysis shows the presence of the expected elements, of which Co from the cathode and Si from the anode are presented. In the depicted example, the overlay of the bottom LiCoO2 and top Si grid patterns is excellent in one direction and well acceptable in the other direction and matches very well with the TFB layout design. The cross section and the elemental mapping (Figure 2b) demonstrate excellent conformal coverage of the Si top anode and good lateral etching control of the LiCoO2 and LiPON films. 685


Research Article

ACS Applied Materials & Interfaces

Figure 3. (a) Transmittance spectra within UV−vis range and (b) correlation between calculated and measured transmittance at the 600 nm wavelength for different realized TFB architectures.

During discharge (negative dQ/dV), Li stripping takes place at the anode, concomitant with Li ion insertion into Li1−xCoO2, as represented by peaks 3′ and 4′. Upon further discharging, Li ion extraction from Li−Si alloys occurs. The somewhat sharp peak 2′ near 3.45 V is indicative of the conversion of Li15Si4 formed during the charge,22,23 and its conversion into amorphous LiySi products at 3.45 V corresponds to a cathode potential of ∼3.90 V and an anode potential of ∼0.45 V. The formed LiySi products are expected to react around 0.5 V vs Li,22,23 which leads to peak 1′ around 3.4 V in the derivative plot when conjugated with the cathode reaction expected at 3.9 V. On the following charge, LiCoO2 reaction around 3.9 V is concomitant with Li−Si alloying reactions expected around 0.25 and 0.1 V, resulting in peaks 1 and 2 around 3.7 and 3.85 V, respectively. Upon cycling, capacity losses are observed, in particular in the Li plating/stripping region and in the region of higher Li concentration in the Si anode, corresponding to a decrease in intensity of peaks 2 and 2′. As shown by Le Cras et al.,22 the transition from a fully crystallized Li15Si4 to an amorphous LiySi anode encompasses domains of phase coexistence characterized by the weighted sum of both phases contributions. Hence, the Si electrode now displays three delithiation peaks at about 0.25, 0.45, and 0.5 V vs Li, which are measured as peaks 2″, 2′, and 1 near 3.65, 3.45, and 3.4 V, respectively, when measured against LiCoO2 (ca. 3.9 V vs Li). Hence, the decrease in intensity of peak 2′ accompanied by an increase in intensity of peak 2″ represents the formation of amorphous LiySi phase at the expense of Li15Si4. 3.2.3. Correlation between the Optical and Electrochemical Functionalities. The discharge capacity variation with transmittance for all TFB types is depicted in Figure 6. It is clear that the discharge capacity (at cycle 1) exhibits a linear

Table 2. TFB Grid Structure Dimensions and Corresponding Calculated Geometric Transmittance TFB design

X2 (μm)

X1 (μm)

Tga (%)

A B C D E F G

150 130 130 110 170 90 90

130 130 205 130 400 130 380

18.3 21.2 31.8 24.9 41.8 29.6 55.5

a

Tg = Tglass/LiPON(X1/(X1 + X2))2.

tion and Li plating/stripping reactions occur during TFB cycling. The electrochemical potential profiles data and derivative capacity plot measured for a TFB are presented in Figure 5. The potential profile can be divided into three regions during the first charge. The beginning of the charge (positive dQ/dV) is characterized by irreversible losses which can be attributed to the irreversible oxidation of the LiCoO2/LiPON and/or to the irreversible reduction of the Si/LiPON interphasial materials created by the intermixing during sputtering.21,22 These irreversible processes are represented in the derivative capacity plot (Figure 5b) by the pronounced peaks A and B. For this type of design, the TFB exhibits about 70 μAh (charge) and 55 μAh (discharge) capacities during the first cycle, thus an available capacity of 100 μAh cm−2 of active area. Subsequently, Li ion exchange between LiCoO2 and Si takes place, resulting from LiCoO2 delithiation into Li1−xCoO2 and Li−Si alloying. At higher potentials, Li plating occurs at the negative electrode side (peaks 3 and 4), and the potential thereby corresponds to that of LiCoO2 vs Li. 686


Research Article


Figure 4. (a) Haze variation within UV−vis range and (b) specular transmittance as a function of incident angle for G-type TFB architecture.

Figure 5. Typical electrochemical characterization of a transparent type G TFB. (a) Potential profiles during galvanostatic cycling within 3−4.2 V at 25 μA for cycles 1−10 and (b) the corresponding derivative capacity curves.

relationship with the transmittance measured at 600 nm, thus substantiating the correlation between storage capacity and electrodes active area. Furthermore, by integrating thicker positive and negative electrodes within the TFB architecture, the transparency remains constant, whereas the energy stored increases proportionally. For instance, when LiCoO2 electrode thickness is increased from 3.5 to 9.5 μm (Si increased from 0.1 to 0.2 μm), the discharge capacities experience an approximately 21/2-fold increase, ranging between 155 and 650 μAh, respectively, for TFB transmittance between 60 and 18%. It is clear from Figure 6 that for a given design of TFB grid structure, the increase of electrode thickness induces only an increase of the capacity while the transmittance is unaffected since the grid structure design is kept unchanged. Therefore, it is possible to achieve higher specific capacity by increasing electrode thickness without altering the TFB transmittance (Figure S3). 3.3. Understanding the Cycling Behavior of the TFB Devices. 3.3.1. Evolution of Impedance and Voltage Profile upon Cycling. The voltage profile and discharge capacity variation during cycling are presented in Figure 7a. The observed evolution indicates a two-regime pattern. In the first regime (cycles 1−50), the discharge capacity tends to decrease gradually at a moderate rate of 0.5% per cycle. The discharge capacity decay in the second regime (cycles 50−100)

Figure 6. Correlation between the electrochemical storage capacity and the optical transmittance for all TFB types and for two architectures: LiCoO2/Si with thicknesses of 3.5/0.1 μm and 9.5/ 0.2 μm, respectively (1) and (2).

decreases significantly, displaying a very low rate of 0.05% per cycle. Likewise, the two regimes can be observed in the voltage 687


Research Article


cycling. Furthermore, the second contribution clearly depends on the TFB voltage, reaching a 1.5 resistance increase at 4.2 V (end of charge) compared to 3 V at cycle 100. Here again, a two-regime pattern evolution is noticed insofar as resistance variation is larger during the first 50 cycles. At this stage of our study, these observations are consistent with the following hypothesis: TFB capacity fade is mainly the consequence of Li ion losses upon cycling; the primary sources are correlated to both plating/stripping and silicon Li insertion−extraction losses. Previous work reported low capacity retention for TFBs implementing both mechanisms separately. Neudecker et al. observed an inhomogeneous plating of lithium at the LiPON/ Cu interface during cycling of LiCoO2/LiPON/Cu TFBs, which induced a decrease of the Li/electrolyte interfacial contact area and consequently a capacity fade.25 Deficient cycling was observed for TFBs integrating Si electrode and was attributed to Li ion losses due to mechanical instability concomitant with nonreversible phase transformations.21,22,26 Although LiCoO2/LiPON/Si TFB architecture has been studied thoroughly,27−30 to our knowledge there are no reports on cycling behavior similar to the one presented in this work because nearly all of the studies considered configurations of Si excess versus LiCoO2. Gong et al.30 reported evidence of Li accumulation at Si/current collector in an overcharged TFB configuration, but no data were reported for cycling. In this study, it can be assumed that plated Li becomes progressively disconnected and isolated, resulting in a faster initial capacity-fading rate, while Li ion entrapment related to Si activity remains relatively low. On subsequent cycling, the available amount of Li ions becomes insufficient, and only Si activity occurs, thus leading to a slower capacity fading rate. To gain deeper insights into the mechanisms occurring at the negative electrode side, post-mortem ToF-SIMS analyses have been conducted to investigate the origins of the capacity decay. 3.3.2. ToF-SIMS Analysis of Elemental Composition. The SIMS depth profiles of the positive secondary ions for as deposited and after 50 cycles TFBs at the discharged state are depicted in Figure 8. The depth profile covers the TFBs from their surfaces down to their interfaces between Si and LiPON electrolyte.

Figure 7. Evolution of the (a) discharge capacity potential profile and of the impedance spectra as a function of cycling in the discharged (3 V) and charged (4.2 V) states (respectively b and c) during galvanostatic cycling within 3−4.2 V at 25 μA. The black, blue, and gray colored curves are respectively for cycles 1, 50, and 100. The inset in panel a represents the discharge capacity variation with cycling.

profile evolution in Figure 7a. It is shown that during the first regime most of the capacity fade occurs above 4 V with a progressive disappearance of 4.15 and 4.05 V plateaus related to LiCoO2 transformations between the hexagonal and monoclinic phases. Then in the second regime, the voltage profile remains almost unchanged, exhibiting a large initial drop and reactions below 3.9 V, thus related only to LiCoO2/ Li−Si alloy couple without Li plating/stripping contribution. Besides, a high rate capability was observed for the TFBs, with less than 18% of discharge capacity decrease between C/2 and 2C rates (Figure S4). The increase of cell impedance during cycling is provided in Figure 7b,c at the end of charge and discharge steps. Two semicircles can be identified from the Nyquist plots. The first semicircle in the high to medium frequency range can be attributed to LiPON contribution, and the second semicircle in the medium−low frequency range can be attributed to interfacial and charge transfer resistances.24 LiPON contribution is very stable whatever the voltage and cycle number. This suggests that not only the bulk material chemistry but also the contact area of LiPON with the electrodes (TFB electrochemically active area) remains stable. The low-frequency semicircle displays a significant and progressive increase upon

Figure 8. ToF-SIMS depth profiles of the upper part of an asdeposited (a) and after 50 cycles (b) TFBs at discharged state (3 V). 688


Research Article

ACS Applied Materials & Interfaces Figure 8a shows the depth profile of the as-deposited reference TFB. Overall, a homogeneous elemental distribution is detected in the Ti, Si, and LiPON thin films. Moreover, abrupt and sharp interfaces are observed between the successive layers. Oxygen, and to a lesser extent Li and P, are observed at the top of the TFB structure. The surface composition may be considered as a result of Ti oxidation and unremoved Li−P containing contaminants during microfabrication processes. Furthermore, the absence of lithium signal intensity in the Si film indicates a fully delithiated anode at the initial state of an as deposited TFB. The depth profile of TFB after 50 cycles at 3 V (discharged state) is significantly changed, as shown in Figure 8b. The elemental distribution is still homogeneous whereas interfaces are clearly diffuse. In comparison to the reference sample, Ti oxidation and Li−P contaminants are also observed at the top of the structure, but with higher intensities. This fact can be attributed to the TFB surface roughening upon cycling in the presence of Li plating reaction31 (Figure S5). Interestingly, lithium signal exhibits a nearly flat profile with high intensity in the Si region, indicating an entrapment of Li+ ions within the negative electrode at the discharged state upon cycling. Furthermore, apparent phase intermixing is observed at the Si/Ti and Si/LiPON interfaces. Previously, a P−Si interdiffused layer has been observed at the Si/LiPON electrolyte/ negative electrode interface during TFB overcharging.32 Similar intermixing is observed at the discharged state after cycling in our case, which we attribute to the roughening of the top surface after prolonged cycling. In addition we clearly see an increase of oxygen content within the Si film. The oxygen signal exhibits a stepwise profile in the Si film with a low intensity for the first 50 nm, comparable to the noncycled TFB. For the deeper part of the Si film, the oxygen intensity rises to higher level (10-fold increase) to reach a maximum at the Si/ LiPON interface. Therefore, SIMS results show evolutions at the anode side during cycling: the electrode is modified with (i) an increase of the roughness at the top of the structure, (ii) an oxidation at the LiPON side as a consequence of interfacial instabilities, and (iii) and a trapping of Li in the Si layer. This can explain in part the interfacial resistance increase observed previously on EIS data. Moreover, Li entrapment in Si may be related to nonreversible phase transformations or local anode disconnection from Ti current collector after Li plating/ stripping. This observation proves the lithium loss during TFB cycling, which can likely lead to the capacity decrease mentioned earlier.

of electrodes thickness, or exploring the implementation on ultrathin transparent substrates. Based on these superior properties, the transparent TFBs have strong potential to be utilized as integrated energy storage units in various emerging applications, including optoelectronics, smart windows, transparent displays, and wearables.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.8b16364. TFB fabrication process description (Figure S1); transmittance and reflectance variation of substrate and LiPON/substrate within 200−2000 nm wavelength range (Figure S2); variation of calculated transmittance Tg,800 and capacity C for a TFB as a function of geometric design parameters (Figure S3); rate capability behavior of the realized TFB (Figure S4); SEM crosssection images of transparent TFBs illustrating surface roughening upon cycling (Figure S5) (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]. ORCID

Sami Oukassi: 0000-0002-3882-6348 Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Vlad, A.; Singh, N.; Galande, C.; Ajayan, P. M. Design Considerations for Unconventional Electrochemical Energy Storage Architectures. Adv. Energy Mater. 2015, 5, 1402115. (2) Pan, S.; Ren, J.; Fang, X.; Peng, H. Integration: An Effective Strategy to Develop Multifunctional Energy Storage Devices. Adv. Energy Mater. 2016, 6 (4), 1501867. (3) Martín, F.; Navarrete, E.; Morales, J.; Roldán, C.; RamosBarrado, J. R.; Sanchez, L. High-Energy, Efficient and Transparent Electrode for Lithium Batteries. J. Mater. Chem. 2010, 20 (14), 2847− 2852. (4) Gittleson, F. S.; Hwang, D.; Ryu, W. H.; Hashmi, S. M.; Hwang, J.; Goh, T.; Taylor, A. D. Ultrathin Nanotube/Nanowire Electrodes by Spin-Spray Layer-by-Layer Assembly: A Concept for Transparent Energy Storage. ACS Nano 2015, 9 (10), 10005−10017. (5) Hu, L.; Wu, H.; La Mantia, F.; Yang, Y.; Cui, Y. Thin, Flexible Secondary Li-Ion Paper Batteries. ACS Nano 2010, 4 (10), 5843− 5848. (6) Puthirath, A. B.; Patra, S.; Pal, S.; Manoj, M.; Puthirath Balan, A.; Jayalekshmi, S.; Tharangattu, N. N. Transparent Flexible Lithium Ion Conducting Solid Polymer Electrolyte. J. Mater. Chem. A 2017, 5 (22), 11152−11162. (7) Pan, Q.; Chen, Y.; Zhang, Y.; Zeng, D.; Sun, Y.; Cheng, H. A Dense Transparent Polymeric Single Ion Conductor for Lithium Ion Batteries with Remarkable Long-Term Stability. J. Power Sources 2016, 336, 75−82. (8) Nagai, H.; Hara, H.; Enomoto, M.; Mochizuki, C.; Honda, T.; Takano, I.; Sato, M. Synchronous Electrochromism of Lithium Ion Battery With Chemically Fabricated Transparent Thin Films. Funct. Mater. Lett. 2013, 06 (02), 1341001. (9) Sollami Delekta, S.; Smith, A. D.; Li, J.; Ö stling, M. Inkjet Printed Highly Transparent and Flexible Graphene Micro-Supercapacitors. Nanoscale 2017, 9 (21), 6998−7005. (10) Nam, I.; Park, S.; Kim, G.-P.; Park, J.; Yi, J. Transparent and Ultra-Bendable All-Solid-State Supercapacitors without Percolation Problems. Chem. Sci. 2013, 4 (4), 1663.

4. CONCLUSION In conclusion, we have proposed and successfully fabricated a grid-structured all solid inorganic transparent TFBs for the first time. As compared with transparent batteries in the literature, our devices exhibit among the best performances (ranging from discharge capacities of 0.15 to 0.6 mAh at transmittances of 60% and 20%). The fabrication process is viable for industrial large-scale production and allows easy control and fine-tuning of the device optical and electrochemical properties. In addition, the transparent TFBs show very good cycling behavior over 100 cycles, with an average capacity loss of 0.08% per cycle. Furthermore, the versatility of the architecture opens opportunities for further development such as optimizing the pattern design, stacking multiple devices, varying the balancing 689


Research Article

ACS Applied Materials & Interfaces (11) Xu, J. L.; Liu, Y. H.; Gao, X.; Sun, Y.; Shen, S.; Cai, X.; Chen, L.; Wang, S. D. Embedded Ag Grid Electrodes as Current Collector for Ultraflexible Transparent Solid-State Supercapacitor. ACS Appl. Mater. Interfaces 2017, 9 (33), 27649−27656. (12) Lee, K.; Lee, H.; Shin, Y.; Yoon, Y.; Kim, D.; Lee, H. Highly Transparent and Flexible Supercapacitors Using Graphene-Graphene Quantum Dots Chelate. Nano Energy 2016, 26, 746−754. (13) Yang, Y.; Jeong, S.; Hu, L.; Wu, H.; Lee, S. W.; Cui, Y. Transparent Lithium-Ion Batteries. Proc. Natl. Acad. Sci. U. S. A. 2011, 108 (32), 13013−13018. (14) Oukassi, S.; Giroud-Garampon, C.; Dubarry, C.; Ducros, C.; Salot, R. All Inorganic Thin Film Electrochromic Device Using LiPON as the Ion Conductor. Sol. Energy Mater. Sol. Cells 2016, 145, 2−7. (15) Su, Y.; Zhang, J.; Shokhovets, S.; Polity, A.; Meyer, B. K. Spectroscopic Ellipsometry and Optical Transmission Study of LiPON Thin Films Prepared by RF Sputtering. Phys. Status Solidi B 2017, 254 (4), 1600424. (16) Yu, X.; Bates, J. B.; Jellison, G. E.; Hart, F. X. A Stable ThinFilm Lithium Electrolyte : Lithium Phosphorus Oxynitride. J. Electrochem. Soc. 1997, 144 (2), 524−532. (17) Park, J. H.; Lee, D. Y.; Kim, Y.; Kim, J. K.; Lee, J. H.; Park, J. H.; Lee, T.; Cho, J. H. Flexible and Transparent Metallic Grid Electrodes Prepared by Evaporative Assembly. ACS Appl. Mater. Interfaces 2014, 6 (15), 12380−12387. (18) Bley, K.; Semmler, J.; Rey, M.; Zhao, C.; Martic, N.; Taylor, R. N. K.; Stingl, M.; Vogel, N. Hierarchical Design of Metal Micro/ Nanohole Array Films Optimizes Transparency and Haze Factor. Adv. Funct. Mater. 2018, 28 (13), 1706965. (19) Lordan, D.; Burke, M.; Manning, M.; Martin, A.; Amann, A.; O'Connell, D.; Murphy, R.; Lyons, C.; Quinn, A. J. Asymmetric Pentagonal Metal Meshes for Flexible Transparent Electrodes and Heaters. ACS Appl. Mater. Interfaces 2017, 9 (5), 4932−4940. (20) Duffie, J. A.; Beckman, W. A. Solar Engineering of Thermal Processes; Wiley: New York, 1974. (21) Baggetto, L.; Niessen, R. A. H.; Notten, P. H. L. On the Activation and Charge Transfer Kinetics of Evaporated Silicon Electrode/Electrolyte Interfaces. Electrochim. Acta 2009, 54 (24), 5937−5941. (22) Le Cras; Le, F.; Pecquenard, B.; Dubois, V.; Phan, V. P.; GuyBouyssou, D. All-Solid-State Lithium-Ion Microbatteries Using Silicon Nanofilm Anodes: High Performance and Memory Effect. Adv. Energy Mater. 2015, 5 (19), 1501061. (23) Baggetto, L.; Niessen, R. A. H.; Roozeboom, F.; Notten, P. H. L. High Energy Density All-Solid-State Batteries: A Challenging Concept towards 3D Integration. Adv. Funct. Mater. 2008, 18 (7), 1057−1066. (24) Larfaillou, S.; Guy-Bouyssou, D.; Le Cras, F.; Franger, S. Comprehensive Characterization of All-Solid-State Thin Films Commercial Microbatteries by Electrochemical Impedance Spectroscopy. J. Power Sources 2016, 319, 139−146. (25) Neudecker, B. J.; Dudney, N. J.; Bates, J. B. “Lithium-Free” Thin-Film Battery with In Situ Plated Li Anode. J. Electrochem. Soc. 2000, 147 (2), 517−523. (26) Reyes Jiménez, A.; Nölle, R.; Wagner, R.; Hüsker, J.; Kolek, M.; Schmuch, R.; Winter, M.; Placke, T. A Step Towards Understanding the Beneficial Influence of a LIPON-Based Artificial SEI on Silicon Thin Film Anodes in Lithium-Ion Batteries. Nanoscale 2018, 10, 2128−2137. (27) Ruzmetov, D.; Oleshko, V. P.; Haney, P. M.; Lezec, H. J.; Karki, K.; Baloch, K. H.; Agrawal, A. K.; Davydov, A. V.; Krylyuk, S.; Liu, Y.; Huang, J. Y.; Tanase, M.; Cumings, J.; Talin, A. A. Electrolyte Stability Determines Scaling Limits for Solid-State 3D Li Ion Batteries. Nano Lett. 2012, 12 (1), 505−511. (28) Gittleson, F. S.; El Gabaly, F. Non-Faradaic Li+Migration and Chemical Coordination across Solid-State Battery Interfaces. Nano Lett. 2017, 17 (11), 6974−6982. (29) Talin, A. A.; Ruzmetov, D.; Kolmakov, A.; McKelvey, K.; Ware, N.; El Gabaly, F.; Dunn, B.; White, H. S. Fabrication, Testing, and

Simulation of All-Solid-State Three-Dimensional Li-Ion Batteries. ACS Appl. Mater. Interfaces 2016, 8 (47), 32385−32391. (30) Gong, C.; Ruzmetov, D.; Pearse, A.; Ma, D.; Munday, J. N.; Rubloff, G.; Talin, A. A.; Leite, M. S. Surface/Interface Effects on High-Performance Thin-Film All-Solid-State Li-Ion Batteries. ACS Appl. Mater. Interfaces 2015, 7 (47), 26007−26011. (31) Motoyama, M.; Ejiri, M.; Iriyama, Y. Modeling the Nucleation and Growth of Li at Metal Current Collector/LiPON Interfaces. J. Electrochem. Soc. 2015, 162 (13), A7067−A7071. (32) Santhanagopalan, D.; Qian, D.; McGilvray, T.; Wang, Z.; Wang, F.; Camino, F.; Graetz, J.; Dudney, N.; Meng, Y. S. Interface Limited Lithium Transport in Solid-State Batteries. J. Phys. Chem. Lett. 2014, 5 (2), 298−303.

690


Transparent Thin Film Solid-State Lithium Ion Batteries - ACS Applied

Recommend Documents