Characterization of Sputter-Deposited LiCoO2 Thin Film Grown on

Subscriber access provided by University of Colorado Boulder

Article 2

Characterization of sputter-deposited LiCoO thin film grown on NASICON type electrolyte for application in all-solid-state rechargeable lithium battery Hee-Soo Kim, Yoong Oh, Ki Hoon Kang, Ju Hwan Kim, Joosun Kim, and Chong Seung Yoon ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 26 Apr 2017 Downloaded from http://pubs.acs.org on April 27, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Characterization of sputter-deposited LiCoO2 thin film grown on NASICON type electrolyte for application in all-solid-state rechargeable lithium battery Hee-Soo Kim,†, ‡ Yoong Oh,† Ki Hoon Kang,† Ju Hwan Kim,† Joosun Kim,‡ Chong Seung Yoon,*,†

†

Department of Materials Science and Engineering, Hanyang University, 222 Wangsimni-ro,

Seongdong-gu, Seoul, 133-791, Korea ‡

High-Temperature Energy Materials Research Center, Korea Institute of Science and

Technology, 5 Hwarang-ro 14-gil, Seongbuk-gu, Seoul 136-791, Korea.

KEYWORDS all-solid state batteries, lithium-ion batteries, solid electrolytes, interfaces, NASICON

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

ABSTRACT

All-solid state Li-rechargeable batteries using a 500-nm-thick LiCoO2 (LCO) film deposited on two

NASICON-type

solid

electrolyte

substrates:

LICGCTM

(OHARA

Inc.)

and

Li1.3Al0.3Ti1.7(PO4)3 (LATP) are constructed. The post-deposition annealing temperature prior to the cell assembly is critical to produce a stable sharp LCO/electrolyte interface and to develop a strong crystallographic texture in the LCO film, conducive to migration of Li ions. Although the cells deliver a limited discharge capacity, the cells cycled stably for 50 cycles. The analysis of the LCO/electrolyte interfaces after cycling demonstrates that the sharp interface, once formed by proper thermal annealing, will remain stable without any evidence for contamination and with minimal intermixing of the constituent elements during cycling. Hence, although ionic conductivity of the NASICON-type solid electrolyte is lower than that of the sulfide electrolytes, the NACSICON-type electrolytes will maintain a stable interface in contact with a LCO cathode, which should be beneficial to improving the capacity retention as well as the rate capability of the all-solid state cell.


2

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Since its successful commercial introduction of LiCoO2 (LCO) - carbon cell in 1991, Li-ion batteries (LIBs) have easily displaced existing rechargeable batteries (e.g. Ni-Cd, and Ni-MH) mainly because of their high-energy density and the LIBs have become ubiquitous portable power sources for applications ranging from household appliances and electronic devices to electric vehicles. Despite their widespread use, current LIBs use a liquid electrolyte which can lead to potential leakage and has relatively poor and chemical stability. Especially, in a large format energy storage system for smart grids, highly flammable solvents used in the non-aqueous liquid electrolyte can pose potential fire and explosion hazards.1 Compared to a liquid electrolyte, a solid electrolyte based on fast-ion conductors provides a much safer transport medium for Li-ions with minimal electronic conductivity.2 In addition, with a solid electrolyte, the operating voltage can be raised beyond the limit (~4.5 V) imposed by the electrochemical instability of the carbonate-based solvents3,4 to increase the power density of the battery5,6. A solid electrolyte allows the use of a Li metal anode to improve the energy density without the safety issue7. With a solid electrolyte, a liquid cathode can be also used to attain a higher capacity compared to the conventional solid insertion cathodes.8 In spite of the safety improvement and the potential gains in energy and power densities, the performance of a battery using a solid state electrolyte is severely limited in power density compared to the conventional LIBs because of the relatively poor ionic conductivity of the solid electrolyte. Although a number of fast ion-conducting solid electrolytes (e.g. sulfide electrolytes with σ > 10-3 S cm-1)9 have been introduced, the current density of all-solid-state batteries is still significantly inferior (several tens of milliamperes per square centimeter) compared to a LIB using a liquid electrolyte. Because the contact at the solid electrolyte and the electrode is solid-to-solid, one potential


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

problem is the large charge transfer resistance at the solid electrolyte/electrode interface due to poor contact, thermal expansion mismatch and the interface reaction. To reduce the interfacial resistance at the solid electrolyte/electrode interface, various schemes including thermal treatment,10 suppression of a space–charge layer by inserting an ion-conducting buffer film,11,12 coating of the LCO cathode,13 and in-situ electrode formation by high voltage14 have been applied and yet compared to the fairly well understood interface chemistry of organic liquid electrolytes, the interface of the solid electrolyte/electrode is relatively poorly understood. In addition, the microstructure of the LCO cathode contacting the electrolyte can also affect the Li diffusion across the interface as the growth orientation of the epitaxially grown LCO films significantly altered the charge/discharge characteristics as it was shown that improper orientation of the grains in the electrode film with respect to the electrolyte crystals can impede the Li migration.15,16 In this study, a model solid electrolyte/cathode interface was constructed by depositing a thin film of LCO onto a commercially available Li2O-Al2O3-SiO2-P2O5-TiO2-GeO2 (LICGCTM) NASICON-type glass ceramic solid electrolyte substrate (manufactured by OHARA Inc., Japan, σ = 10-4 S cm-1 at room temperature) and the interface was heat treated at different temperatures to observe the chemical and microstructural evolution of the interfacial structure and the concurrent change in the microstructure of the LCO cathode. Moreover, the same experiment was repeated using an in-house synthesized Li1.3Al0.3Ti1.7(PO4)3 (LATP), also a NASICON-type fast ion conductor as a solid electrolyte to demonstrate that the observed physical and chemical changes can be generally applied to other NASICON-type solid electrolytes with different compositions, not specific to the OHARA glass ceramic electrolyte. To study the stability of the


4

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


interface at the LCO/NASICON-type electrolytes during cycling, all-solid-state cells were constructed and detailed structural analysis of the cycled interfaces was carried out. Experimental Procedure A 90-µm-thick Si- and Ge-doped LATP sheet (ionic conductivity = 10-4 S cm-1) was purchased from OHARA Inc., Japan. To prepare the in-house LATP electrolyte, LATP powder was synthesized based on the solid-state reaction method using lithium carbonate (Li2CO3), Aluminum oxide (Al2O3), titanium oxide (TiO2) and Ammonium phosphate ((NH4)2HPO4). The raw materials in stoichiometric amounts were mixed and ground using a ball mill for 6 h. The ground powder was calcined at 900 °C for 2 h and then ball-milled again for 12 h in air. The prepared powder was pressed at 2 metric ton and fully sintered at 1000 °C for 2 h in air.17 A LATP sheet was produced by cutting and polishing the sintered pellet to 300 µm in thickness. Structure and morphology of the synthesized LATP powder were examined by x-ray diffraction (XRD, Rigaku, Rint-2000), scanning electron microscopy (SEM, JEOL, JEM-6330F). To construct a model interface, a 500-nm-thick LCO thin film was deposited on the OHARA or LATP substrate by magnetron sputtering at room temperature with a base pressure of < 5×10-6 Torr. Ar pressure was kept at 5 mtorr during deposition and the film deposition rate was 0.8 Å sec-1. The deposited LCO film was then heat-treated in O2 at different temperatures (400 700 °C) for 1 hour to induce crystallization. Transmission electron microscopy (TEM, JEOL 2100F) was mainly used to characterize the microstructure of the thin film and interface. Rutherford backscattering spectroscopy (RBS) was performed in an analysis chamber connected to 6 MV tandetron accelerator. For the assembly of an all-solid-state battery, a 2032 coin-type half-cell was used. The cell assembly is schematically described in Figure S1 in the electronic supplemental material. In


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

short, 500 nm LCO (cathode)/50 nm Pt(current collector) was first deposited on the solid electrolyte substrate by sputtering. On the reverse side of the substrate, 1-µm-thick film of LIPON was deposited again by sputtering. The electrolyte substrate with the deposited films was then transferred into a glove box in which a 2032 coin cell was assembled using Li foil as an anode. The cell tests were performed between 3.3 V and 4.2 V at a C-rate of 0.01C and 30 °C

Results and Discussion To examine the microstructure of the annealed LCO film, cross-sectional TEM images and corresponding selected area electron diffraction (SAED) patterns taken from a thin section prepared by focused ion beam for samples annealed at different temperatures are shown in Figure 1. The as-deposited film was amorphous as expected for a sputter-deposited metal-oxide film. After annealing at 400 °C, the LCO film was partially crystallized with nano-crystallites embedded in an amorphous matrix (Figure 1a and Figure S2 in the electronic supplemental material). At 500 °C, the amorphous film was fully crystallized and the crystallized LCO film was made up of thin columnar grains which, in turn, were composed of nano-crystallites as can be seen Figure 1b. Raising the annealing temperature above 500 °C considerably increased the crystallite size, also evidenced by the discontinuous and progressively spotty SAED patterns (Figure 1c, d). The LCO film annealed at 600 °C also contained intermittent voids which became exceedingly large at 700 °C. The microstructure of the LCO film annealed at 700 °C was dramatically different compared to the films annealed at lower temperatures, containing 200-nmsized equi-axed grains with an array of large voids. The voids were likely due to the Kirkendall effect which causes formation of voids due to the different interdiffusion rates.18 SAED patterns _

in Figure 1 were compatible with the R3m space group; however, a SAED pattern obtained from


6

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


_

a single LCO grain from the film annealed at 700 °C could no longer be indexed to the R3m structure. Instead the pattern was indexed as the [110] zone of a fcc structure. It appears that annealing at 700 °C produced a large set of voids in the LCO film and also triggered nucleation of a fcc phase (rocksalt or spinel) as LCO is known to transform from the rhombohedral phase to cubic phases at high temperatures. Based on the TEM analysis, 500 °C was the appropriate annealing temperature for a sputter-deposited LCO film on the LICGCTM substrate as the LCO film was fully crystallized without voids. The SAED pattern for the LCO film annealed at 500 °C also displays a strong crystallographic texture developed in the crystallized film as the diffraction pattern consisted of a series of arcs and spots instead of continuous circular rings. The SAED pattern from the film annealed at 500 _

°C was in fact a superposition of three [100] zone (R3 m structure) spot patterns oriented at different angles as shown by different colors in Figure 2a. The orientation of the (003) spots (indicated by blue and red) suggests that a set of layer planes originating from those spots were aligned nearly normal to the substrate, providing relatively fast diffusion paths for Li+ migration. The layer planes belonging to the lone (003) spot (in yellow) were nearly parallel to the substrate. It is interesting that thermal annealing of the amorphous LCO film produced three distinctive crystallographic textures with respect to the substrate considering that the LICGCTM substrate is polycrystalline with randomly oriented grains, providing no preferential growth directions for the LCO grains. Dark field TEM images using the three (003) spots are shown in Figure 2b. Elongated grains consisting of nano-sized crystallites can be clearly observed. The crystallites with (003) planes parallel to the substrate (outlined in yellow) were more or less uniformly distributed throughout the film whereas the crystallites with (003) planes nearly normal to the substrate tended to segregate into individual columnar grains; i.e. grains excited in the dark-field


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

image outlined in red are de-excited in the dark-field image outlined in blue). The dark field images also suggest that there is a preferential growth relationship among the layer planes as two adjacent diffraction spots were at an angle of 80o with respect to the lone spot regardless of the _

regions from which SAED patterns were collected. Upon closely examining the R3m structure, the interplanar spacing of (101) planes (d101 = 0.25 nm) is nearly one half of the spacing of (003) plane (d003 =0.52 nm) and the interplanar angle between (101) and (003) planes is 80o. Based on the crystallographic relation between (101) and (003) planes, a set of (003) planes grew quasiepitaxially on top of (003) planes with the growth direction aligned along [101] direction of the base (003) plane. The angular relationship is similar to the twin configuration observed in a LCO thin film19 such that the structure may be envisioned as a series of twins. The proposed quasiepitaxial growth well explains the angular relationship among the diffraction spots observed from the annealed LCO film. The quasi-epitaxial relationship found in the annealed LCO film is schematically described in Figure 3a. As a further of the quasi-epitaxial relationship, a highresolution TEM image of the two crystallites highlighting the angular relationship is shown in Figure 3b. The Fourier transformed images from the two crystallites also verify the quasiepitaxial relationship between the two crystallites. The [101] texture is important because the texture orients the Li layer planes nearly normal to the electrolyte since the texture expedites the Li+ migration. It was also observed that the quasi-epitaxial growth of the LCO crystallites was not dependent on the substrate as identical crystallographic texture was found in the LCO film grown on a Pt film after annealing at 700 °C (see Figure S3 in the supplemental electronic material). Interfaces of the LCO/LICGCTM substrate annealed at different temperatures were examined by cross-sectional TEM in Figure 4a–c. After annealing at 500 °C, the LCO/LICGCTM interface


8

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


was sharp without any reaction phases as attested by a high-resolution TEM image in Figure S4. However, increasing the annealing temperature above 500 °C developed a porous amorphous interlayer (50 nm for 600 °C and 300 nm for 700 °C) which was impregnated with nanocrystalline phases. The interface layer likely developed as a result of massive interdiffusion between the LCO cathode and the electrolyte. Sharpness of the interface at a microscopic scale after annealing at 500 °C was also verified using energy-dispersive X-ray spectroscopy (EDS) compositional line scan across the LCO/LICGCTM interface as shown in Figure 5a. Based on the line scan, the region in which the constituent atoms from the LCO film and the LICGCTM substrate intermixed were confined to a ~10-nm-thick layer at the interface. Elemental mappings of Co (from LCO) and Ti, Al, Ge, and P (from LICGCTM) shown in Figure 5b present a clean cathode/electrolyte interface. RBS spectra were also obtained from the LCO/LICGCTM interfaces annealed at different temperatures to estimate the extent of the interdiffusion across the interface. RUMP (RBS simulation package) was used to model the LCO(500 nm)/LICGCTM film which were overlaid on the RBS spectrum recorded from the as-deposited film in Figure 6a . Near the interface around channel 550, a peak appeared due to the overlap of signals from Ge in the LICGCTM substrate and from Co in the LCO film in agreement with the simulated spectrum. When the LCO film was annealed at 500 °C, the RBS spectrum hardly changed from the asdeposited film. However, as the annealing temperature was raised to 600 °C, a substantial level of interdiffusion was observed at the interface as the Co and Ge atoms diffused away from the interface. A large drop of Co and Ge concentrations at the interface observed at 700 °C likely originated from the large porous interlayer observed in the TEM image in Figure 4c. Thus, RBS spectra clearly showed that the interdiffusion is minimal after annealing at 500 °C and beyond this temperature, extensive interdiffusion severely damaged the interface. From the TEM and


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

RBS results, it can be concluded that 500 °C is the optimal annealing temperature to produce a sharp LCO/LICGCTM interface. It is also noted that when a LCO film was deposited on top of a Pt film, annealing even at 700 °C produced well-crystallized LCO grains with no distortion at the LCO/Pt interface (see Figure S3). Therefore, the interdiffusion and hence, the quality of the LCO interface were very much dependent on the substrate. In fact, in the case of the pulsed laser deposited LCO film on a (Li0.5La0.5)TiO3 substrate, the discharge capacity of the cell was very much dependent on the deposition temperature with complete loss of the discharge capacity for the LCO films deposited at high temperatures.20 When a LiNi1/3Co1/3Mn1/3O2 film was deposited on a LATP substrate, the interface quality was also dependent on the deposition temperature. When the surface temperature of the substrate was 670 °C during deposition of the LiNi1/3Co1/3Mn1/3O2 film, a large increase in the interfacial resistance was observed due to the deterioration of the interface. It is interesting that the optimal surface temperature of the LiNi1/3Co1/3Mn1/3O2 film deposition on the LATP substrate found in that work was 520 °C.21 A LCO film was also sputter-deposited on the LATP electrolyte which has an ionic conductivity of 1.6 x 10-4 S cm-1 at 30 °C produced in our laboratory. Cross-sectional TEM microstructural analysis and EDS line scan and elemental mapping results confirm that annealing at 500 °C also produced the LCO morphology and texture similar to the case of the LCO deposited on a LICGCTM substrate with a sharp LCO/LATP interface with minimal interdiffusion. Optimal annealing temperature was rather insensitive to the composition (LICGCTM or LATP) or to the microstructure of the substrate as there was a large difference in the grain sizes of the LICGCTM (0.1 µm) and LATP (0.2 µm). In order to characterize the electrochemical properties of the LCO film grown on the LICGCTM and LATP substrates, a 2032 coin-type half-cell was assembled using the film stack shown in


10

Page 11 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7a. A 50-nm-thick Pt was deposited on top of the annealed LCO film as a current collector. In addition, 1-µm-thick LIPON film was deposited on the backside of the LICGCTM or LATP substrates to prevent direct contact of Ti ions in the electrolyte with the Li foil because of the detrimental reactive interaction between NASICON-type electrolytes and Li metal.22 Figure 7b shows a SEM image of the Pt(50 nm)/LCO(500nm)/LICGCTM(150 µm)/LIPON(1 µm)/Li foil film stack with no visible cracks at the interfaces. The charge and discharge curves of the halfcell cycled between 3.3 V and 4.2 V at a C-rate of 0.01C and 30 °C during cycling are shown in Figure 7c. The initial discharge capacity of the LCO cathode was 40 µAm µm-1 cm-2 (circa 80 mAh g-1) which is substantially inferior compared to that of a bulk LCO cathode with a liquid electrolyte (140 mAh g-1)23 and gradually decreased as the cycling proceeded. After the 5th cycle, the discharge capacity stabilized at 31 µAm µm-1 cm-2 (circa 62 mAh g-1). The discharge capacity of

the

LCO

film

does

not

compare

well

with

the

previously

reported

LCO/LICGCTM/Polyethylene oxide (PEO)/Li cell which exhibited a discharge capacity of 180 mAh g-1 in the first cycle.24 The inferior discharge capacity of the LCO/LICGCTM cell was largely from the irreversible capacity loss in the first cycle (Coulombic efficiency of 68%). To identify the reason for the large first-cycle capacity loss, the LCO cathode was tested using wellstudied LIPON solid electrolyte in a thin film battery (Pt/LCO/LIPON/Li). The first chargedischarge profiles and cycling data are shown in Figure S5. Using the LIPON electrolyte, the first cycle Coulombic efficiency of the LCO cathode was still low at 78% with a first-cycle discharge capacity of only 36.1 µAm µm-1 cm-2 although the cell cycled stably after the first cycle. Hence, the inferior discharge capacity and first-cycle Coulombic efficiency in Figure 7c likely stemmed from the structural defects in the annealed LCO film which contains a large fraction of grain boundaries. An LCO thin film (0.5-µm-thick) prepared by sol-gel also exhibited


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

a comparable loss of discharge capacity in the first cycle and the first-cycle loss was attributed to structural defects.25 Although it is not unequivocally identified, the large initial reversible capacity loss compared to the previous work may be also related to the contact problem at the anode side as many trials were needed to optimize the anode portion of the cell. Meanwhile, the discharge capacity plotted during cycling in Figure 7d shows the continuous capacity fading of the LCO/LICGCTM/LIPON/Li cell. In the case of the LCO/LATP cell, the cell also exhibited the similar capacity loss in the first cycle and gradual capacity fading during extended cycling. To examine the microstructure of both the cathode and the LICGCTM electrolyte after cycling, the LCO cathode and the LICGCTM electrolyte were recovered after 10 cycles. A TEM image of the cycled cell in Figure 8a indicates that the electrolyte and the interface remained intact. However, the SAED pattern from the cycled LCO film displayed in Figure 8b together with the SAED pattern before cycling for comparison exhibited a subtle difference as the spot patterns broadened in the circular direction after cycling. The angular spread of the (003) spots prior to cycling was 15° – 20° which increased to 30° – 35° after 10 cycles, suggesting that the (003) texturing was gradually destroyed during cycling. Hence, it is likely that the slow capacity fading observed after the first 5 cycles was manifestation of the loss of the crystallographic texture favoring the Li ion diffusion through the cathode. As for the interface, no secondary phases or an interlayer can be seen from the high-resolution TEM image in Figure 8c and the line scan data in Figure 8d also confirms that there were no extensive interdiffusion of elements as the line scan data well matched that of the pristine interface before cycling. A set of elemental mapping data (in Figure S6) also verified that the interface remained sharp after 10 cycles. The cycling behavior of the LCO/LATP cell was nearly identical to that of LCO/LICGCTM for the first 10 cycles and the cycling was extended up to 50 cycles. Even after 50 cycles, the interface remained


12

Page 13 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


stable without noticeable level of intermixing and formation of any secondary phases (see Figure S7 for the line scan and mapping of the LCO/LATP interface after 50 cycles). The chemical stability of the LCO/LATP interface was also verified using RBS analysis of the cycled cell presented in Figure 9. The RBS spectrum of the LCO/LATP stack in Figure 9a shows a single square peak corresponding to Co in the LCO cathode as peaks from other elements were overlapped and lumped together in the sloping peak below channel 500. After 50 cycles, the Co peak from the cycled cathode hardly changed (Figure 9b), verifying that there was minimal diffusion of Co from the cathode into the LATP electrolyte during cycling. It is suspected that the capacity fading during cycling despite the stable cathode interface is due to the degradation of the anode. To further characterize the interfaces in the cell, electrochemical impedance spectroscopy was used to estimate the surface-film resistance (Rsf) and charge transfer resistance (Rct) prior to cycling and after 10 cycles (see Figure S8). Both Rsf and Rct of the cell increased substantially after 10 cycles. The increase in Rsf and Rct most likely stemmed from the formation of the reaction layer on the anode side, judging by the relatively sharp interface on the LCO/LICGCTM and the degradation of the interface observed at the LICGCTM/LIPON interface (see Figure S9). Previously, it was shown that Co from the LCO cathode will diffuse into the Li2S-P2S5 solid electrolyte without an oxide diffusion barrier after initial charging, which raised the interfacial resistance.13 Hence, although ionic conductivity of the NASICON-type solid electrolyte is lower than that of the sulfide electrolyte, the NACSICON-type electrolytes will maintain a stable interface in contact with the LCO cathode which should be beneficial to improving the capacity retention as well as the rate capability of the cell. Hence, it is important that the anode side needs to be further examined and optimized to improve the cycling stability of the LCO/LICGCTM/LIPON/Li cell.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

Conclusion A 500-nm-thick LCO film was deposited on two different NASICON-type solid electrolytes by sputtering and annealed to observe the microstructural changes in the LCO cathode and LCO/electrolyte interface. During crystallization of the LCO film at 500 oC, a crystallographic texture conducive to migration of Li ions was developed. When tested in a functioning solidstate cell, the LCO/NASICON-type electrolyte interface remained stable during cycling unlike the interfaces between a LCO cathode and sulfide or perovskite-type solid electrolytes. Although capacity retention of the solid state cell did not compare favorably with the well-established LIPON data, the result represents one of the first results to reversibly cycle a solid state cell based on a LATP solid electrolyte for extended cycles. The observed capacity fading was likely due to the anode degradation. Based on the observed degradation of the NASICON-type electrolyte in contact with Li metal,22 it is advisable that the interface at the anode side needs to be optimized to achieve high current density from an all-solid-state battery using a NASICONtype solid electrolyte.


14

Page 15 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


FIGURES.

Figure 1. Cross-sectional TEM images for LCO film deposited on LICGCTM substrate after annealing at (a) 400 °C, (b) 500 °C (c) 600 °C and (d) 700 °C with corresponding electron diffraction patterns.


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

Figure 2. LCO film on LICGCTM substrate after annealing at 500 °C (a) electron diffraction patterns highlighting the superimposed three [010] zone patterns and (b) dark field TEM images using the three (003) spots.


16

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3.

(a) Schematic illustration for quasi-epitaxial crystallographic texture observed in

LCO film annealed at 500 °C, (b) high-resolution TEM image of two crystallites in the annealed LCO film and corresponding Fourier transformed images verifying the angular relationship describe in (a).


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4.

Page 18 of 30

Bright field TEM images for the LCO/LICGCTM interface after annealing at (a)

500 °C, (b) 600 °C and (c) 700 °C.


18

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. (a) Compositional EDS line scan and (b) elemental mapping data for the LCO films on LICGCTM substrate after annealing at 500 °C.


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

Figure 6. RBS spectra for (a) as-deposited LCO film and (b) after annealing at 500 -700 °C on LICGCTM substrate.


20

Page 21 of 30


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30

Figure 7. (a) Schematic diagram of all-solid LCO/LICGCTM/Li cell, (b) SEM image of all-solid LCO/LICGCTM/Li cell, (c) charge and discharge profiles and (d) capacity retention curve of the LCO/LICGCTM/Li cell cycled between 3.3 V and 4.2 V at a C-rate of 0.01C and 30 °C.


22

Page 23 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8. (a) Bright field TEM image of the LCO/LICGCTM cell after 10 cycles, (b) electron diffraction patterns before and after cycling, (c) high-resolution TEM image and (d) compositional EDS line scan of LCO/LICGCTM interface after 10 cycles.


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

Figure 9. RBS spectra from LCO/LATP stack (a) prior to cycling and (b) after 50 cycles.


24

Page 25 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ASSOCIATED CONTENT Supporting Information. Figure S1 - assembly of an all-solid-state half-cell battery cell using a 2032 coin cell; Figure S2 - high-resolution TEM image for the LCO film after annealing at 400 °C showing embedded nano-crystallites; Figure S3 - cross-sectional TEM images for the LCO film on the Pt film after annealing at 700 °C with corresponding indexed electron diffraction pattern of LCO film; Figure S4 - high-resolution TEM images for LCO/LICGCTM interface after annealing at 500 °C; Figure S5 - first charge-discharge profile and capacity retention curve for the Pt/LCO/LIPON/Li thin film cell using the LCO film annealed at 500 oC; Figure S6 - elemental mapping data of LCO/LICGCTM interface after 10 cycles; Figure S7 - compositional EDS line scan and elemental mapping data of LCO/LATP interface after 50 cycles; Figure S8 - Nyquist plots for the Pt/LCO/LICGCTM/LIPON/Li cell prior to cycling and after 10 cycles; Figure S9 - TEM image of the Pt/LCO/LICGCTM/LIPON/Li cell near the anode after 10 cycles.

AUTHOR INFORMATION Corresponding Author *Tel: 82-2-2220-0384; E-mail: [email protected]

Funding Sources Ministry of Trade, Industry & Energy, Republic of Korea (No. 20135020900010)


25


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 30

ACKNOWLEDGMENT This work was supported by the ∗Global Excellent Technology Innovation of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20135020900010).

REFERENCES (1) Suzuki, Y.; Munakata, H.; Kajihara, K.; Kanamura, K.; Sato, Y.; Yamamoto, K.; Yoshida, T. Fabrication of Three-Dimensional Battery Using Ceramic Electrolyte with Honeycomb Structure by Sol-Gel Process. ECS Trans. 2009, 16, 37-43. (2) Takada, K. Progress and Prospective of Solid-State Lithium Batteries. Acta Mater. 2013, 61, 759-770. (3) Tan, S.; Ji, Y. J.; Zhang, Z. R.; Yang, Y. Recent Progress in Research on High-Voltage Electrolytes for Lithium-Ion Batteries. ChemPhysChem 2014, 15, 1956-1969. (4) Hu, M.; Pang, X.; Zhou, Z. J. Recent Progress in High-Voltage Lithium Ion Batteries. J. Power Sources 2013, 237, 229-242. (5) Oh, G.; Hirayama, M.; Kwon, O.; Suzuki, K.; Kanno, R. Bulk-Type All Solid-State Batteries with 5 V Class LiNi0.5Mn1.5O4 Cathode and Li10GeP2S12 Solid Electrolyte. Chem. Mater. 2016, 28, 2634-2640. (6) Yubuchi, S.; Ito, Y.; Matsuyama, T.; Hayashi, A.; Tatsumisago, M. 5 V Class LiNi0.5Mn1.5O4 Positive Electrode Coated with Li3PO4 Thin Film for All-Solid-State Batteries Using Sulfide Solid Electrolyte. Solid State Ionics. 2016, 285, 79-82.


26

Page 27 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(7) Cheng, X.; Zhang, R.; Zhao, C.; Wei, F.; Zhang, J.; Zhang, Q. A Review of Solid Electrolyte Interphases on Lithium Metal Anode. Adv. Sci. 2016, 3, 1500213. (8) Inaguma, Y.; Liquan, C.; Itoh, M.; Nakamura, T. High Ionic Conductivity in Lithium Lanthanum Titanate. Solid State Commun. 1993, 86, 689-693. (9) Aotani, N.; Iwamoto, K.; Takada, K.; Kondo, S. Synthesis and Electrochemical Properties of Lithium Ion Conductive Glass, Li3PO4-Li2S-SiS2. Solid State Ionics. 1994, 68, 35-39. (10) Iriyama, Y.; Kako, T.; Yada, C.; Abe, T.; Ogumi, Z. Charge Transfer Reaction at the Lithium Phosphorus Oxynitride Glass Electrolyte/Lithium Cobalt Oxide Thin Film Interface. Solid State Ionics. 2005, 176, 2371-2376. (11) Ohta, N.; Takada, K.; Zhang, L.; Ma, R.; Osada, M.; Sasaki, T. Enhancement of the HighRate Capability of Solid-State Lithium Batteries by Nanoscale Interfacial Modification. Adv. Mater. 2006, 18, 2226-2229. (12) Yada, C.; Lee, C. E.; Laughman, D.; Hannah, L.; Iba, H.; Hayden, B. E. A HighThroughput Approach Developing Lithium-Niobium-Tantalum Oxides as Electrolyte/Cathode Interlayers for High-Voltage All-Solid-State Lithium Batteries. J. Electrochem. Soc. 2015, 162, A722-A726. (13) Sakuda, A.; Hayashi, A.; Tatsumisago, M. Interfacial Observation Between LiCoO2 Electrode and Li2S-P2S5 Solid Electrolytes of All-Solid-State Lithium Secondary Batteries Using Transmission Electron Microscopy. Chem. Mater. 2010, 22, 949-956.


27


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 30

(14) Iriyama, Y.; Yada, C.; Abe, T.; Ogumi, Z.; Kikuchi, K. A New Kind of All-Solid-State Thin-Film-Type Lithium-Ion Battery Developed by Applying a D.C. High Voltage. Electrochem. Commun. 2006, 8, 1287-1291. (15) Shiraki, S.; Takagi, Y.; Shimizu, R.; Suzuki, T.; Haruta, M.; Sato, Y.; Ikuhara, Y.; Hitosugi, T. Orientation Control of LiCoO2 Epitaxial Thin Films on Metal Substrates. Thin Solid Films 2016, 600, 175-178. (16) Takeuchi, S.; Tan, H.; Bharathi, K. K.; Stafford, G. R.; Shin, J.; Yasui, S.; Takeuchi, I.; Bendersky, L. A. Epitaxial LiCoO2 Films as a Model System for Fundamental Electrochemical Studies of Positive Electrodes. ACS Appl. Mater. Interfaces 2015, 7, 7901-7911. (17) Aono, H.; Sugimoto, E.; Sadaoka, Y.; Imanaka, N.; Adachi, G. Ionic Conductivity of Solid Electrolytes Based on Lithium Titanium Phosphate. J. Electrochem. Soc. 1990, 137, 10231027. (18) He, Y.; Wang, Y.; Yu, X.; Li, H.; Huang, X. Si-Cu Thin Film Electrode with Kirkendall Voids Structure for Lithium-Ion Batteries. J. Electrochem. Soc. 2012, 159, A2076-A2081. (19) Moriwake, H.; Kuwabara, A.; Fisher, C. A. J.; Huang, R.; Hitosugi, T.; Ikuhara, Y. H.; Oki, H.; Ikuhara, Y. First-Principles Calculations of Lithium-Ion Migration at a Coherent Grain Boundary in a Cathode Material, LiCoO2. Adv. Mater. 2013, 25, 618-622. (20) Ahn, J.; Yoon, S. Characteristics of Amorphous Lithium Lanthanum Titanate Electrolyte Thin Films Grown by PLD for Use in Rechargeable Lithium Microbatteries. Electrochem. SolidState Lett. 2005, 8, A75-A78.


28

Page 29 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(21) Kato, T.; Yoshida, R.; Yamamoto, K.; Hirayama, T.; Motoyamab, M.; West, W. C.; Iriyama, Y. Effects of Sintering Temperature on Interfacial Structure and Interfacial Resistance for All-Solid-State Rechargeable Lithium Batteries. J. Power Sources 2013, 325, 584-590. (22) Hartmann, P.; Leichtweiss, T.; Busche, M. R.; Schneider, M.; Reich, M.; Sann, J.; Adelhelm, P.; Janek, J. Degradation of NASICON-Type Materials in Contact with Lithium Metal: Formation of Mixed Conducting Interphases (MCI) on Solid Electrolytes. J. Phys. Chem. C 2013, 117, 21064-21074. (23) Kannan, A. M.; Rabenberg, L.; Manthiram, A. High Capacity Surface-Modified LiCoO2 Cathodes for Lithium-Ion Batteries. Electrochem. Solid-State Lett. 2003, 6, A16-A18. (24) Xie, J.; Imanishi, N.; Zhang, T.; Hirano, A.; Takeda, Y.; Yamamoto, O. Li-Ion Transport in All-Solid-State Lithium Batteries with LiCoO2 Using NASICON-Type Glass Ceramic Electrolytes. J. Power Sources 2009, 189, 365-370. (25) Rho, Y. H.; Kanamura, K.; Umegaki, T. LiCoO2 and LiMn2O4 Thin-Film Electrodes for Rechargeable Lithium Batteries. J. Electrochem. Soc. 2003, 150, A107-A111.


29


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 30

Table of Contents Graphic


30

Characterization of Sputter-Deposited LiCoO2 Thin Film Grown on

Recommend Documents